Principles of

Electromagnetics
2—Dielectric
and Conductive
Materials

Arlon T. Adams
Jay K. Lee
Principles of
Electromagnetics 2—
Dielectric and Conductive
Materials

Arlon T. Adams
Jay K. Lee
Principles of Electromagnetics 2—Dielectric and Conductive Materials

Copyright © Cognella Academic Publishing 2015

www.cognella.com

All rights reserved. No part of this publication may be reproduced,

stored in a retrieval system, or transmitted in any form or by any
means—electronic, mechanical, photocopy, recording, or any other
except for brief quotations, not to exceed 400 words, without the prior
permission of the publisher.

ISBN-13: 978-1-60650-716-2 (e-book)

www.momentumpress.net

Trademark Notice: Product or corporate names may be trademarks

or registered trademarks, and are used only for identification and
explanation without intent to infringe.

A publication in the Momentum Press Electrical Power collection

Cover and interior design by S4Carlisle Publishing Services Private Ltd.,

Chennai, India
 Brief Contents

Preface...................................................................................................ix
Chapter 1 Introduction to Dielectrics.................................................1
Chapter 2 Electric Currents and Conductive Materials.....................45
 Contents

Chapter 1 Introduction to Dielectrics.................................................1

1.1 Introduction.............................................................1
1.2 Polarization...............................................................2
1.3 The Electric Field of a Polarized Dielectric................4
1.4 The Displacement Vector D....................................11
1.4.1 Linear Dielectrics ............................................12
1.4.2 Linear, Homogeneous Dielectrics....................13
1.5 Boundary Conditions .............................................17
1.6 Capacitance ............................................................21
1.7 Benjamin Franklin (1706-1790) and the
Beginnings of Electrical Science�����������������������������28
1.8 The Force on a Conductor in an Electric Field........34
1.9 Energy and Forces in the Electrostatic Field............36
1.9.1 Energy of a Collection of Point Charges..........36
1.9.2 Energy of Continuous Charge Distribution.....37
1.9.3 Forces and Torques in Terms of Energy............40
Chapter 2 Electric Currents and Conductive Materials.....................45
2.1 Introduction...........................................................45
2.2 Electric Current......................................................45
2.2.1 Volume Current Density.................................46
2.2.2 Surface Current Density..................................47
2.2.3 Line Current ...................................................48
2.3 The Equation of Continuity....................................48
2.4 Ohm’s Law and Conductive Materials.....................49
2.5 Georg Simon Ohm (1789-1854) and the
Discovery of Ohm’s Law����������������������������������������52
2.6 Power – Joule’s Law.................................................54
2.7 Relaxation Time......................................................55
2.8 Boundary Conditions for Steady Currents..............56
2.9 A Relationship Between Capacitance and
Resistance��������������������������������������������������������������59
 List of Figures

Figure 1-1. A polarized dielectric........................................................1

Figure 1-2. Dipoles in an electric field................................................3
Figure 1-3. (a) A uniformly polarized block of dielectric.
(b) An equivalent charge distribution................................4
Figure 1-4. A non-uniform polarization Pz(z)......................................5
Figure 1-5. The charge distribution for the step approximation
to the non-uniform polarization Pz(z)...............................6
Figure 1-6. A permanently polarized dielectric cylinder (electret)........9
Figure 1-7. A point charge and a spherical dielectric shell.................14
Figure 1-8. A dielectric-loaded coaxial transmission line...................15
Figure 1-9. Tangential E at a boundary.............................................17
Figure 1-10. (a) Normal D at a boundary (side view)
(b) Full view of surface S................................................18
Figure 1-11. Field lines in a dielectric-loaded parallel-plate region
with air gaps...................................................................20
Figure 1-12. (a) Capacitance...............................................................21
Figure 1-12 (b) A parallel-plate capacitor...........................................23
Figure 1-13. Concentric spherical and cylindrical geometries.
Inner and outer surfaces are perfect conductors..............24
Figure 1-14. (a) A conductor in an electric field..................................34
Figure 1-14 (b) The normal electric field Ez and a self field Esz
at a conductor surface.....................................................34
Figure 1-15. A collection of point charges...........................................36
Figure 1-16. A dielectric slab partially inserted between parallel
plates..............................................................................41
Figure 2-1. Volume current flow through an arbitrary surface...........46
Figure 2-2. Drift velocity..................................................................46
Figure 2-3. Surface current density Js................................................47
Figure 2-4. The equation of continuity.............................................48
Figure 2-5. Uniform current flow in a resistor...................................50
Figure 2-6. Relaxation time..............................................................55
viii THE BASICS OF FOREIGN EXCHANGE MARKETS

Figure 2-7. Continuity of normal J...................................................57

Figure 2-8. Dissimilar dielectrics in a parallel-plate capacitor............57
Figure 2-9. A relationship between capacitance and resistance..........59
Figure 2-10. A coaxial line with inner, outer radii a, b,
respectively and length ℓ.................................................61
Figure 2-11. A truncated conical resistor.............................................63
 Preface
Electromagnetics is not an easy subject for students. The subject
presents a number of challenges, such as: new math, new physics, new
geometry, new insights and difficult problems. As a result, every aspect
needs to be presented to students carefully, with thorough mathematics
and strong physical insights and even alternative ways of viewing and
formulating the subject. The theoretician James Clerk Maxwell and the
experimentalist Michael Faraday, both shown on the cover, had high
respect for physical insights.
This book is written primarily as a text for an undergraduate course
in electromagnetics, taken by junior and senior engineering and phys-
ics students. The book can also serve as a text for beginning graduate
courses by including advanced subjects and problems. The book has been
thoroughly class-tested for many years for a two-semester Electromagnet-
ics course at Syracuse University for electrical engineering and physics
students. It could also be used for a one-semester course, covering up
through Chapter 8 and perhaps skipping Chapter 4 and some other parts.
For a one-semester course with more emphasis on waves, the instructor
could briefly cover basic materials from statics (mainly Chapters 2 and 6)
and then cover Chapters 8 through 12.
The authors have attempted to explain the difficult concepts of elec-
tromagnetic theory in a way that students can readily understand and
follow, without omitting the important details critical to a solid under-
standing of a subject. We have included a large number of examples, sum-
mary tables, alternative formulations, whenever possible, and homework
problems. The examples explain the basic approach, leading the students
step by step, slowly at first, to the conclusion. Then special cases and
limiting cases are examined to draw out analogies, physical insights and
their interpretation. Finally, a very extensive set of problems enables the
instructor to teach the course for several years without repeating problem
assignments. Answers to selected problems at the end allow students to
check if their answers are correct.
x PREFACE

During our years of teaching electromagnetics, we became interested

in its historical aspects and found it useful and instructive to introduce
stories of the basic discoveries into the classroom. We have included short
biographical sketches of some of the leading figures of electromagnetics,
including Josiah Willard Gibbs, Charles Augustin Coulomb, Benjamin
Franklin, Pierre Simon de Laplace, Georg Simon Ohm, Andre Marie
Ampère, Joseph Henry, Michael Faraday, and James Clerk Maxwell.
The text incorporates some unique features that include:

•• Coordinate transformations in 2D (Figures 1-11, 1-12).

•• Summary tables, such as Table 2-1, 4-1, 6-1, 10-1.
•• Repeated use of equivalent forms with R (conceptual) and
|r−r′| (mathematical) for the distance between the source
point and the field point as in Eqs. (2-27), (2-46), (6-18),
(6-19), (12-21).
•• Intuitive derivation of equivalent bound charges from
polarization sources, including piecewise approximation to
non-uniform polarization (Section 3.3).
•• Self-field (Section 3.8).
•• Concept of the equivalent problem in the method of images
(Section 4.3).
•• Intuitive derivation of equivalent bound currents from
magnetization sources, including piecewise approximation to
non-uniform magnetization (Section 7.3).
•• Thorough treatment of Faraday’s law and experiments
(Sections 8.3, 8.4).
•• Uniform plane waves propagating in arbitrary direction
(Section 9.4.1).
•• Treatment of total internal reflection (Section 10.4).
•• Transmission line equations from field theory (Section
11.7.2).
•• Presentation of the retarded potential formulation in Chapter
12.
•• Interpretation of the Hertzian dipole fields (Section 12.3).

Finally, we would like to acknowledge all those who contributed to

the textbook. First of all, we would like to thank all of the undergraduate
 PREFACE xi

and graduate students, too numerous to mention, whose comments and

suggestions have proven invaluable. As well, one million thanks go to Ms.
Brenda Flowers for typing the entire manuscript and making corrections
numerous times. We also wish to express our gratitude to Dr. Eunseok
Park, Professor Tae Hoon Yoo, Dr. Gokhan Aydin, and Mr. Walid M. G.
Dyab for drawing figures and plotting curves, and to Professor Mahmoud
El Sabbagh for reviewing the manuscript. Thanks go to the University of
Poitiers, France and Seoul National University, Korea where an office and
academic facilities were provided to Professor Adams and Professor Lee,
respectively, during their sabbatical years. Thanks especially to Syracuse
University where we taught for a total of over 50 years. Comments and
suggestions from readers would be most welcome.

Arlon T. Adams

Jay Kyoon Lee

[email protected]
June 2012
Syracuse, New York
 CHAPTER 1

Introduction to Dielectrics

1.1 Introduction
We turn our attention now to ideal (perfect) dielectrics. Wood, glass,
chalk, plastics, rubber, paper, quartz, and distilled water are all considered
close to perfect dielectics. Ideal dielectrics do not contain free charges
which can move from molecule to molecule. Instead they possess bound
charges which are tightly bound to the atomic or molecular structure.
Now what happens when an electric field is applied to a dielectric as
in Figure 1-1? The bound charges are not free to move from molecule
to molecule but they can move over very small distances. Positive and
negative charges tend to move in opposite directions. A typical result is
shown in Figure 1-1 with dipoles existing in the body of the dielectric and
surface charges on two surfaces. Dipoles are lined up with their dipole
moments parallel to E. We say that the dielectric is polarized. We see that
in Figure 1-1 bound surface charge density is formed on the right and
left hand surfaces of the dielectric. Bound volume charge density may or
may not exist inside the dielectric, depending on the type of polarization.

+ + + + +
ps
+ + + + + +
+ + + + + +
P
+ + + + + +

E
Figure 1-1. A polarized dielectric
2 PRINCIPLES OF ELECTROMAGNETICS 2

We distinguish bound charges from free charges by adding a subscript

p. Thus bound or polarization charges may exist as bound volume charge
density ρpv or as bound surface charge density ρps. The net bound charge in a
volume is represented as Qp. The definitions for ρpv, ρps are identical to Eqs.
(2-2) except that ΔQ is replaced with ΔQp. Note that ρv, ρs, Q represent free
charges; we add the subscript p for bound or polarization charge densities.

1.2 Polarization
We have seen an example of a polarized dielectric with dipoles existing
within the body of a dielectric. Since dipoles give rise to electric fields and
potentials, we need to characterize the strength of the dipoles within the
dielectric. How many dipoles are there per unit volume and what are the
dipole moments? We define a polarization vector P(x,y,z) to characterize
the polarized state of the dielectric material. To calculate P at a particular
point, we construct a small volume Δv around the point in question, and
add vectorially all the dipole moments pi within the volume Δv.
Then
1
P = lim
∆V → 0
∑ Pi
∆v i
(1-1)

P is called the polarization or the dipole moment per unit volume. By defi-
nition it is automatically zero in vacuum (free space).
In the absence of an applied electric field, most dielectrics are unpo-
larized (P = 0) either because of an orderly arrangement of dipoles whose
moments cancel as in Figures 2-17(a),(b) or because of a random orienta-
tion of dipoles. The individual molecules may be polarized or not. Non-
polar molecules have no net dipole moment. Polar molecules each have a
net dipole moment but a random orientation often tends to produce an
unpolarized dielectric macroscopically in the absence of an electric field.
There are some dielectrics which are permanently polarized, i.e., they
may remain polarized in the absence of an electric field. An example is
barium titanate, which is one of the ferroelectrics. Such materials are
called electrets, the electrical analogue of magnets.
If an electric field is applied, dielectrics which were previously unpo-
larized become polarized (P ≠ 0), as in Figure 1-1, with a polarization P
 Introduction to Dielectrics 3

which tends to be in the direction of the electric field E. There are several
contributors to the polarization. First, consider an unpolarized dielectric
with dipoles whose moments cancel in the absence of an electric field. The
cancellation may be orderly or disorderly. Figure 1-2(a) shows an orderly
cancellation. The application of an electric field leads to forces on the pos-
itive and negative bound charges which make up the dipoles. We assume
that the dipoles can rotate slightly about their centers. The dipoles will
tend to rotate to the new positions shown dotted, each rotation resulting
in a dipole moment with an increased component in the direction of E.
We also assume that the dipole can stretch or contract. Consider a dipole
pair with moment parallel, antiparallel to E (Figure 1-2(b)). The applied
field stretches one dipole and contracts another, in each case increasing
the dipole moment in the direction of E. Figures 3-2(c), (d) show the
effect of an applied field on an individual atom. The nucleus is displaced
with respect to the center of the electron cloud. This produces a net dipole
moment in the direction of E. Thus we see that given an unpolarized di-
electric (P = 0), the addition of an applied electric field tends to polarize
the dielectric with a P in the direction of E.
+ +

+
+

+
+
+ +

E E
(a) Dipole rotation. (b) Stretching or contraction of dipoles.

Electron
cloud
Electron
cloud
Nucleus

+
Nucleus

+
Center

(c) An atom. (d) An atom with applied field.

Figure 1-2. Dipoles in an electric field

4 PRINCIPLES OF ELECTROMAGNETICS 2

We should emphasize that the movements of the bound charges and

the resultant deformation of the structure, i.e., the resultant rotations and
stretching of dipoles, are very small indeed. The basic reason for this is that
the applied fields produce forces on the bound charges which are very small
compared with the strong internal forces holding the structure together.
The movement shown in our figures is much greater than that usually ob-
tained. The perturbation or deformation of the structure is usually mini-
mal and for this reason the process is often linear (P ~ E). On the other
hand, if extremely large electric fields are applied, the bound charges may
be torn from their molecules and breakdown will occur.

1.3 The Electric Field of a Polarized Dielectric

As we have seen, a dielectric becomes polarized in the presence of an elec-
tric field. An orderly arrangement of dipoles is created with a concomitant
bound surface charge and perhaps, as we will soon see, a bound volume
charge as well. What are the bound electric charge densities associated
with the state of polarization and what is the electric field due to those
charges?
First, let us consider the charge distribution of a uniformly polarized
dielectric. Figure 1-3(a) shows a block of dielectric with uniform polariza-
tion P = az Po.
z
z

+ P + so
so
b
b
d d
so

c c
so

Figure 1-3 (a). A uniformly polarized block of dielectric. (b) An

equivalent charge distribution

We note that an unknown uniform bound surface charge density ± ρso

is present on top and bottom surfaces, respectively. There is no bound
 Introduction to Dielectrics 5

volume charge density within the block since every small volume contains
equal amounts of positive and negative charges, i.e., ρpv = 0. Since there
is no volume charge density, we can represent the charges of the system
as in Figure 1-3(b), which shows uniform surface charge densities ± ρso
separated by a distance d. Now we will solve for ρso by requiring that the
dipole moments of Figures 1-3(a), (b) be equal. We consider a small patch
of surface Δs above and below and the intervening volume d(Δs). The
dipole moment of the intervening volume in Figure 1-3(a) is az (Po d Δs)
because Po is the dipole moment per unit volume. The dipole moment of
the same volume in Figure 1-3(b) is az (q d) = az (ρso Δs d). The moments
must be equal and therefore
ρso = Po (1-2a)
We can also compare the total dipole moments of Figures 1-3(a), (b), i.e.,
az Po (bcd) = az ρso (bcd), with the same result. The bound surface charge
density ρps of Figure 1-3(a) may be characterized as follows:

ρps = ±Po = P ⋅ ( ± a z ) on top and bottom

(1-2b)
= 0 on vertical sides
or, in general,
ρps = P ⋅ a n (1-2c)
where an is the outward unit normal to the dielectric surface.

Pz(z) Step approximation

to P (z)
z

P2 P2(z)

P1

z
z z
Figure 1-4. A non-uniform polarization Pz(z)
6 PRINCIPLES OF ELECTROMAGNETICS 2

It is perhaps surprising that we can throw away all the dipoles in the
body of the dielectric since the individual dipoles do produce electric
fields. This is essentially the same process as replacing two dipoles laid
end-to-end with the two end charges. The dipole moment doubles and
two charges in the middle cancel.
Now let us consider what happens if P is not uniform but varies in one
dimension. Assume that P is z-directed, P = az Pz, and let Pz be a continu-
ous function of z as shown in Figure 1-4.
We approximate the continuous function with a series of steps, each
of length Δz. We then represent each step by a model such as that of
Figure 1-3 with a uniform P in each step and bound surface charges at the
beginning and end of each step. The resulting bound charges are shown
in Figure 1-5.
The first two regions are represented by two slabs of uniformly
polarized material with polarization P1 and P2, respectively. Since the po-
larization of the two slabs differ, there is a net surface charge density P1 − P2
which is equal in magnitude to the step discontinuity at z = Δz. Assume a
surface of cross section area of Δs at z = Δz. Then the total bound surface
charge at z = Δz is given as follows:

∂Pz
∆Q P =ρps ∆s = (P1 − P2 )∆s = − ∆z∆s
∂z
z

+P2 + + + + + + + + + + + + + +

P2
P2 ps =P1 P2
+P1 + + + + + + + + + + + + + +
P1
P1

Figure 1-5. The charge distribution for the step approximation to the
non-uniform polarization Pz(z)

Now let’s spread the excess bound surface charge ΔQp over the region (one
∆z
half subsection (or ) to the right and one half subsection to the left)
2
 Introduction to Dielectrics 7

to obtain a representation of the bound volume charge density ρpv. This is

equivalent to taking smaller and smaller steps in the approximation of the
∂P
continuous function Pz(Z). Then ∆Q P =ρpv ∆s ∆z = − z ∆s ∆z and
∂z

∂Pz
ρpv = −
∂z

If we consider additional components of P, then

∂Px ∂Py ∂Pz
ρpv = − − − = −∇ ⋅ P (1-3)
∂x ∂y ∂z
The result above corresponds to the limit Δz → 0. As Δz → 0, we
obtain an infinite number of steps in the approximation and the surface
charge becomes volume charge density.
Table 1-1 gives the equivalent bound charge densities for polarized
material. These are real bound charges that exist in polarized material.
They are equivalent in the sense that we can obtain the electric field E by
assuming those charges in free space (so that the formulations of Chapter
2 apply). The bound charges are also called the polarization charges.

Table 1-1 Equivalent Bound Charge Densities

Surface Charge [C/m2] Volume Charge [C/m3]

ρps = P · an ρpv = − ∇ · P

Equations (1-2) and (1-3) above specify the bound charge densities which
are associated with the polarization vector P. Note that P in Eq. (1-2)
should be evaluated at the surface and an is a unit vector normal outward
from the dielectric surface. The negative sign in Eq. (1-3) is due to the
previous choice that dipole moment p is directed from negative to posi-
tive charges. We note that for uniform polarization P there is no volume
charge:
ρpv = 0 (if P is constant) (1-4)

Later we will see that uniform P is sufficient but not necessary, in order that
ρpv = 0. We will see that ρpv = 0 if the dielectric is linear and homogeneous.
8 PRINCIPLES OF ELECTROMAGNETICS 2

Given polarization P, the bound volume and surface charge den-

sities are known. We can use this information to obtain the electric po-
tential V and the electric field E merely by substituting ρpv, ρps for ρv, ρs,
respectively, in Eqs. (2-46) and (2-27), to obtain

1 ρpv dv ′ 1 ρpv dv ′
V=
4πε 0 ∫∫∫ r − r ′ or
4πε 0 ∫∫∫ R (1-5a)
for volume charge

1 ρps ds ′ 1 ρps ds ′
4 πε0 ∫∫ r − r ′ 4πε0 ∫∫ R
V =
(1-5b) or

for surface charge

and

1 ρpv (r − r ′ )dv ′
r − r′
E=
4πε 0 ∫∫∫ 3

(1-6a)
1 ρpv Rdv ′
or
4πε 0 ∫∫∫ R3
for volume charge

1 ρps (r − r ′ )ds ′
E=
4πε 0 ∫∫ r − r′
3

(1-6a)
1 ρps Rds ′
4πε 0 ∫∫ R 3
or


for surface ch
harge

Note that Eqs. (2-46) ad (2-27) can now be interpreted in a more general
light. A charge is a charge; bound and free charges contribute identically to V
and E. Thus we can interpret ρv, ρs in Eqs. (2-46) and (2-27) as free charge
densities, bound charge densities, or total (bound plus free) charge densities.
 Introduction to Dielectrics 9

Example 1-1. On-Axis Potential of a Permanently Polarized Cylinder

(Electret)
Consider a uniformly polarized cylinder (Figure 1-6):

 0<ρ<a 
P = a z Po  
 z1 ≤ z ≤ z 2 

Find the electrostatic potential along the z axis.

z
Field point
2
(0,0,z)
1
R2

a
z2

R1

z1

Figure 1-6. A permanently polarized dielectric cylinder (electret)

Solution:
First we find the bound charge densities.

ρps = P ⋅ a n = Po (upper disk, z = z 2 )

= −Po (lower disk, z = z 2 )

We can use directly the results from Example 2-9. Substituting (z − z2) for
z everywhere in Eq. (2-49), we obtain the following contribution to the
potential from the upper disk (ρps = Po):
10 PRINCIPLES OF ELECTROMAGNETICS 2

Po 
( z − z 2 )2 + a 2 − z − z 2 
2ε 0  

Subtracting a similar contribution from the lower disk (ρps = −Po), we

obtain

Po 
V(z) = ( z − z 2 )2 + a 2 − ( z − z1 )2 + a 2 − z − z 2 + z − z1  (1-7)
2ε 0  

which can also be expressed as

Po
V(z) =
2ε 0
[R 2 − R 1 − R 2 | sin θ2 | + R 1 | sin θ1 |] (1-8)

Can you establish the following identity?

− z − z 2 + z − z1 = z 2 − z1 ( z ≥ z 2 )
= z1 − z 2 ( z ≥ z1)
= 2 z − z 2 − z1 ( z1 ≤ z ≤ z 2 )

Equation (1-7) can also be obtained by using Eq. (1-5b) for both upper
and lower surfaces (z = z1, z2). The potential V(z) of Eq. (1-7) is continu-
ous along the z axis. The electric field Ez jumps (is discontinuous) at z =
z1, z2 because of the bound surface charges on the upper, lower surfaces
of the cylinder.
The on-axis electric field E z may be obtained in several ways:
(a) by using E = −∇V, (b) by direct application of Eq. (1-6b), (c)
by using the results of Example 2-7 for both upper and lower discs
at z = z 1, z 2.
 Introduction to Dielectrics 11

1.4 The Displacement Vector D

As we have noted before, free and bound charges are identical in their ef-
fect on V, E. Therefore, in the presence of dielectrics, we replace ρv with
ρv + ρpv in Eq. (2-12) (Gauss’ law) to obtain
ρv + ρpv
∇⋅E = (1-9a)
ε0
or

∇ ⋅ ε 0 E = ρv + ρpv (1-9b)

and note that

∇ ⋅ P = −ρpv (1-3)
Equation (1-9) indicates that E lines begin and end on free or bound
charges, going from positive to negative charges. Equation (1-3) indicates
that P lines begin and end on bound charges, going from negative to posi-
tive charges. Adding Eqs. (1-9b) and (1-3),
∇ ⋅ ( ε 0 E + P ) = ρv (1-10a)

The quantity εo E + P is defined as the displacement vector D:

D = ε0 E + P (1-11)

Then Eq. (1-10a) becomes

∇ ⋅ D = ρv (1-10b)

D is thus a vector whose lines begin and end on free charges. We obtain
the integral form of Eq. (1-10b) by integrating both sides over volume V
and applying the divergence theorem:
' ∫∫ D ⋅ ds = Q f (1-12)
S

where Qf is the free charge enclosed within the surface S. Equation (1-12)
should be recognized as a new form of Gauss’ law which is valid even in
the presence of the dielectric. Equation (1-12) is more useful since we
often know the free charges but not the bound charges. Table 1-2 sum-
marizes the two forms of Gauss’ law for D and E. Note that
12 PRINCIPLES OF ELECTROMAGNETICS 2

Q f = ∫∫∫ ρv dv (total free charge enclosed)

V

Q p = ∫∫∫ ρpv dv (total bound charge enclosed)

V

Table 1-2 Gauss’ law

Point Form Integral Form
∇ · D = ρv
' ∫∫ D ⋅ ds = Q f
S

1 Qf +Qp
∇⋅E = (ρv + ρpv ) ' ∫∫ E ⋅ ds =
ε0 S
ε0

1.4.1 Linear Dielectrics

Many dielectrics are highly linear; thus we can assume that P is proportional to E.
P = ε0 χe E (1-13)

Then
D = ε 0 E + P = ε 0 (1 + χe )E = ε 0 ε r E = εE (1-14)
χe, εr, ε are called electric susceptibility, relative permittivity (or dielec-
tric constant), permittivity, respectively. They represent three different
ways of specifying the same linear relationship. We will usually specify the
permittivity ε. The three parameters (χe, εr, ε) are not functions of E for
linear dielectrics but may be functions of position (x,y,z). For homogeneous
media they are constants. Thus ε(x,y,z) represents a linear, inhomogeneous
medium, and ε = constant represents a linear, homogeneous medium.
For a linear dielectric medium,

D = εE
Linear dielectric medium (1-15)
       
P = ( ε − ε 0 )E   
and both D and P are known once E is known. Equation (1-15) is known
as a constitutive relation. Table 1-3 shows some typical values of dielec-
tric constant εr.
 Introduction to Dielectrics 13

Table 1-3 Dielectric Constants

Material Dielectric Constant

Air 1.0006
Balsa Wood 1.4
Paper 2-4
Teflon 2.1
Polyethylene 2.25
Polystyrene 2.55
Dry Soil 2.6
Glass 4 - 10
Quartz 4.3
Mica 6
Silicon 11.7
Alcohol 28
Water 80
Rutile 100
BaTiO3 500 - 4500

1.4.2 Linear, Homogeneous Dielectrics

If a medium is both linear and homogeneous, then D = ε E and ε is a

constant. Thus

∇ ⋅ D = ∇ ⋅ ( εE ) = ε∇ ⋅ E + ∇
ε ⋅ E = ρv
||
0

(see Eq. (1-43)).

Then
ρv ρv + ρpv
∇⋅E = =
ε ε0
and
ε − ε0 
ρpv = −  ρ Linar homogeneous dielectric (1-16)
 ε  v
14 PRINCIPLES OF ELECTROMAGNETICS 2

Also, since an ideal (perfect) dielectric has no free charge (ρv = 0),

ρρv = 0 Liner, homogeneous, perfect dielectric (1-17)

Example 1-2. A Point Charge and a Dielectric Shell

Figure 1-7 shows a point charge q surrounded by a dielectric shell of radii
a, b and permittivity ε1. Find the electric field in each region and the
bound surface charges at r = a, b.
Solution:
Because of spherical symmetry, we expect that E and D will be radial.

E = a r E r , D = a r Dr

Apply Gauss’ law for the spherical surface of radius r.

Regardless of where r is chosen,

' ∫∫ D ⋅ ds = 4πr 2 Dr = Q f = q
q
Dr = everywhere (1-18)
4πr 2
(r < a, a < r < b, and r > b)

0
a

q b

Figure 1-7. A point charge and a spherical dielectric shell

Dr q
Er = = (a < r < b) in the dielectric shell
ε1 4πε1r 2
 Introduction to Dielectrics 15

Dr q
Er = = (r < a, r > b)
ε 0 4πε 0 r 2
The polarization (vector) P in the dielectric shell is given by

(ε1 − ε 0 )q
P = (ε1 − ε 0 )E = a r (a < r < b) (1-19a)
4 πε1r 2
Bound charges are:

1 ∂ 2 1 ∂  (ε1 − ε 0 )q 
ρρv = −∇ ⋅ P = −
r ∂r
2 ( r Pr ) = − 2 
r ∂r  4 πε1 
=0

 ε − ε0  q
ρρs = P ⋅ a n = −  1  at r = a [a n = −a r ] (1-19b)
 ε1  4πa
2

 ε − ε0  q
= 1  at r = b [ a n = − a r ]
 ε1  4 πb
2

Note that the electric charge q has drawn up a negative bound charge at r = a and left
a positive bound charge at r = b. This bound charge is related to free charge
(that would be present if there were a conducting shell) by the ratio indicated
in Eq. (1-16). It also weakens the electric field within the dielectric. Note also
that although P is not uniform, ρρv = 0, which confirms Eq. (1-17).

Example 1-3. A Dielectric-Loaded Coaxial Transmission Line

Figure 1-8 shows a coaxial line of inner radius a, outer radius b, which
is loaded with a dielectric of permittivity ε1. A voltage Vo is applied between
inner and outer conductors. The inner conductor is the positive reference.
Find the electrostatic fields and the bound and free charge densities.
Conductor

1
Dielectric
1
a
Conductor 2a 2b
b

Dielectric
1

Conductor

Figure 1-8. A dielectric-loaded coaxial transmission line

16 PRINCIPLES OF ELECTROMAGNETICS 2

Solutions:
First we assume an unknown free charge ρℓ per unit length at ρ = a,
which is drawn up by the application of voltage Vo. Then −ρℓ per unit
length appears at ρ = b. Assuming that the coaxial line is infinitely long,
the fields will be radial (ρ-directed) because of cylindrical symmetry.
Applying Gauss’ law for the cylinder of radius ρ and length ℓ,

' ∫∫ D ⋅ ds = 2 πρ l Dp = Q f = ρl 1
ρl  1  Dρ ρ  1
Dρ =   , Eρ = = l  
2π  ρ  ε1 2 πε1  ρ 

To determine the (linear) relationship between applied voltage Vo and the

resultant free charge ρℓ per unit length:
a a
ρl 1 ρ b
Vo = − ∫ E ⋅ d = − ∫ Eρ dρ = − ∫ dρ = l ln( )
b
2πε1 b ρ 2πε1 a

2πε1 Vo
and ρl =
 b
ln  
 a

The free surface charge density is

ρ ε V
ρs = = 1 o at ρ = a (1-20a)
2πa  b
a ln  
 a

− ρl ε1 Vo
= = at ρ = b (1-20b)
2πb  b
b ln  
 a

The electric field and the polarization are

ρl  1  Vo  1 
E = aρ = aρ (1-21a)
2πε1  ρ   b   ρ 
ln  
 a
 Introduction to Dielectrics 17

( ε1 − ε 0 ) Vo  1 
P = ( ε1 − ε 0 )E = a ρ (1-21b)
 b   ρ 
ln  
 a
The bound surface charge density is
( ε1 − ε 0 ) Vo
ρps = P = a n = − (ρ = a ) (1-22a)
 b
a ln  
 a

( ε1 − ε 0 ) Vo
= (ρ = b) (1-22b)
 b
b ln  
 a
Note that the bound and free surface charge densities at both ρ = a
and ρ = b are related by
 ε − ε0 
ρps = −  1 ρS
 ε1 

as in Eq. (3-16).

1.5 Boundary Conditions

We now turn to problems involving two or more media. To treat these
problems, we need to understand how the vectors E, D behave at an in-
terface. First, we analyze the tangential component of the electric field E.
Consider a particular point P lying on the interface between two arbitrary
media 1, 2 (Figure 1-9).

E1
l l
1
2 P
E2 x

Figure 1-9. Tangential E at a boundary

Let x, y be rectangular coordinates of a point P located in the tangent

plane to the interface between media 1 and 2. We construct a closed
contour C around P as shown in Figure 1-9 and consider the line integral
18 PRINCIPLES OF ELECTROMAGNETICS 2

∫
C
E ⋅ d

The long side ∆ℓ is parallel to the x direction; ∆ℓ is made small so that

E does not vary over the length. α is made arbitrarily small compared to
unity so that the short side does not contribute to the line integral. E1,
E2 are the electric fields at the interface in regions 1, 2, respectively. Then
from the first law, of electrostatics, Eq. (2-11),

∫
C
E ⋅ d = (E1x − E 2x )∆ = 0

and E1x = E2x; similarly, E1y = E2y, therefore

E1t − E 2t (1-24)

i.e., the tangential component of the electric field is continuous across the
boundary.
Next we analyze the normal component of the displacement D. Con-
sider Figure 1-10 which shows a small closed surface S about the point
P on the interface, with top and bottom (horizontal) surfaces of area ∆s
parallel to the tangent plane and vertical side of area α∆s normal to the
tangent plane.
an
D1
an S S S
1 S
2 P
D2
(a) (b)
Figure 1-10. (a) Normal D at a boundary (side view).
(b) Full view of surface S

We consider the integral ∫∫


D ⋅ ds. ∆s is made small so that D does
not vary over S and α is madeS small compared to unity so that the flux
through the vertical side does not contribute to ∫∫

D ⋅ ds . A direction
an normal to the interface is arbitrarily chosen pointing into region 1.
S

Then if there exists a free surface charge ρs at the boundary, Gauss’ law,
Eq. (1-12), leads to
 Introduction to Dielectrics 19

' ∫∫ D ⋅ ds = (D1n − D2n )∆s = Q f = ρs ∆s

S (1-25)
D1n − D2n = ρs

Normal D is discontinuous by an amount equal to the free surface charge

density. The polarities can be checked by noting that a normal vector D,
upon encountering a positive charge, increases. Note that the normal di-
rection (an) was chosen pointing into region 1; Eq. (1-25) will change if
the normal is chosen pointing into region 2.
Because we have used the two basic laws ∫ ∫∫
E ⋅ d and ' D ⋅ ds = Q f ,
which are valid for any electrostatic problem,
C the boundaryS conditions are
also valid for any electrostatic problem. They hold for any media whatsoever,
i.e., dielectrics (perfect and imperfect) and vacuum and any combinations
thereof. There are thus many special cases for regions 1, 2. Table 1-4 outlines
a few of the most important cases.
Note that for arbitrary media, the jump (discontinuity) in normal
D, E, P is proportional to ρs, ρs + ρps, −ρps respectively. At the interface
between two ideal dielectrics, normal D is continuous. The boundary condi-
tions at an air-conductor interface, Eq. (2-67), can be obtained by setting
ε1 = ε0 in the last case of Table 1-4.

Table 1-4 Boundary Conditions

Arbitrary Media

E1t = E2t, D1n − D2n = ρs

ε0 ( E1n − E2n ) = ρs + ρps

P1n − P2n = −ρps

Linear, Ideal Dielectrics (rs = 0)

E1t = E 2t D1n = D2n

ε  ε 
D1t = D2t  1  E1n = E 2n  1 
 ε2   ε2 
 ε − ε0   ε − ε0  ε2
P1t = P2t  1 P1n = P2n  1
 ε 2 − ε 0   ε 2 − ε 0  ε1
 20 PRINCIPLES OF ELECTROMAGNETICS 2

Conductor, Dielectric
(Region #2 is a conductor)

E2 = D2 = P2 = 0; ε0 E1n = (ρs + ρps), E1t = 0

D1n = ρs

P1n = − ρps

Conductor, Linear Dielectric

(Region #2 is a conductor and Region #1 has permittivity ε1)

ρs
E 2 = D2 = P2 = 0; E1n = , E1t = 0
ε1
D1n = ρs
ε -ε 
P1n = ρs  1 0 
 ε1 

Example 1-4. Fields in a Parallel-Plate Region

Figure 1-11 shows the field lines for ε0E and P in a parallel plate region
with a homogeneous dielectric and air gaps. The fringing fields near the edges
are neglected. The electric field E is stronger in the air gap, due to the fact
that some E lines terminate on bound charge. Note that the sum of ε0E +
P equals D and that the D lines begin and end only on free charge. For a
 ε − ε0 
linear dielectric with permittivity ε1 , ρps = −  1  as in Eq. (3-16).
 ε1 
++ ++ +++++++++ s
C

ps

V0+ 0 E P
+ + + + + +

0 E

P
Figure 1-11. Field lines in a dielectric-loaded parallel-plate region
with air gaps
 Introduction to Dielectrics 21

1.6 Capacitance
In this section we study the capacitance of a capacitor. Consider two con-
ductors (1), (2) with charges Q, −Q, and electric field E and voltage V as
shown in Figure 1-12(a).

Q
Q

1 2
Conductor
S1
Conductor S2

Figure 1-12(a). Capacitance

A linear dielectric of permittivity ε exists in the region between the con-

ductors. The charge resides on the surfaces S1, S2 with a surface charge
distribution ρs. Because of the linear relationships between ρs, E, V, there
is a linear relationship between charge and voltage. Doubling one doubles
the other. The ratio of charge to voltage is called the capacitance:

Q
C= (1-26)
V

Note that the voltage V here is the potential difference between the two
conductors. The unit of the capacitance is the farad [F] (named after
Michael Faraday). The capacitance may also be expressed in terms of elec-
tric field E and displacement D:
22 PRINCIPLES OF ELECTROMAGNETICS 2

Q S1
s
S
∫∫ ρ ds ∫∫ D ⋅ ds
C=
= 1 = 11
V
− E ⋅ d − E ⋅ d
(1-27a) ∫2
∫
2

which reduces to the following form for a linear, homogeneous dielectric

ε ∫∫ E ds
S1
C= 1 (1-27b)
− ∫ E ⋅ d
2

(for a linear, homonogeneous dielectric)

The capacitance of a capacitor depends on the geometry of the capacitor
and the dielectric property of the medium between the conductors. The
capacitance C can be calculated by the following steps:

1. Assume charges +Q, −Q on the two conductors.

2.   Find the electric E, using one of the methods we learned so far.
(+)

3.   Calculate the potential difference, V = − ∫ E ⋅ d .

Q (−)
4.   Find C by taking the ratio V .

Example 1-5. Parallel-Plate Capacitor

Consider a parallel-plate capacitor with dielectric material of uni-
form permittivity ε shown in Figure 1-12(b). The area of each plate is
A and plate separation is d. Neglect the fringing fields. Determine the
capacitance.
Solution:
First we assume charges Q, −Q on the plates, resulting in a surface
charge ρs as shown in Figure 1-12(b), where
Q
ρs =
A
We expect that the electric field is uniform as in Example 2-6. Applying
the boundary condition at x = 0,
 Introduction to Dielectrics 23

Q Q
Dn = D x = εE x = ρs = ; Ex =
A εA
and the potential V across the capacitor between plates can now be
determined.
0 d
D x d Qd
V = − ∫ E ⋅ d = ∫ E x dx = =
d 0
ε εA

s ps ps ps

+
+ E
+
+
+
+
+
+
Area +
A
+ E
+
x=0 x=d
Figure 1-12(b). A parallel-plate capacitor

The capacitance is
Q εA
C= = (1-28)
V d
The bound charge ρps at x = 0 is determined from boundary conditions:

( ε1 − ε 0 ) ε − ε0 
ρps = P ⋅ a n = P ⋅ ( −a x ) = − D x = −  1  ρS
ε  ε 

as in Eq. (1-16). Note then an is normal outward from the dielectric.

Consider what happens when we insert the dielectric between parallel
plates. Under constant charge conditions, bound charge is drawn up to
weaken E and V, thereby increasing C. Under constant voltage condi-
tions, the electric field must be maintained. Additional free charge must
be drawn up to maintain E despite the weakening effect of bound charge.
Q and C are therefore increased. Insertion of a dielectric increases the
capacitance. C is proportional to the area of the plate and inversely pro-
portional to the plate separation.
24 PRINCIPLES OF ELECTROMAGNETICS 2

Example 1-6. Spherical Capacitors (see Figure 1-13)

(a)The Isolated Sphere
Consider an isolated conducting sphere of radius a in free space. To
find the capacitance, we assume a charge q uniformly distributed over
the surface of the sphere. Here we assume that the negative terminal is at
infinity. Applying Gauss’ law (see Example 2-4):

q q q
Er = (r > a ); V(r) = ; V(a) = , V(∞ ) = 0
4πε 0 r 2
4πε 0 r 4πε 0 a
Then
q
C= = 4πε 0 a (1-29)
V(a)
The capacitance of the earth is therefore less than a millifarad.

1
a
a

b
b c

(a) A capacitor. (b) Capacitors in series.

3
1 2

a 2
a
1
b

b
c

(c) Capacitors in parallel. (d) Capacitors in a series-parallel combination.

Figure 1-13. Concentric spherical and cylindrical geometries.

Inner and outer surfaces are perfect conductors
 Introduction to Dielectrics 25

(b) Concentric Spheres (see Figure 1-13(a))

Consider concentric spheres (perfect conductors at r = a, b). The
space between spheres is filled with a dielectric of permittivity ε. Assume
a charge q at r = a and −q at r = b. Then

a
q q 1 1 q  b − a
Er = (a < r < b ); Vab = − ∫ E ⋅ d = ( − )=  
4πεr 2
b
4πε a b 4πε  a b 
Then
q 4πεab
= C=
Vab b−a (1-30)

→ 4πεa as b → ∞
which agrees with Eq. (1-29).
(c)Capacitors in Series (Figure 1-13(b))
Figure 1-13(b) shows a spherical capacitor with two dielectrics. Two
conductors are at r = a, c. We assume a charge q at r = a and −q at r = c.
Applying Gauss’ law:
q
Dr = (a < r < c ) (1-31)
4πr 2
Note that normal D is continuous across the boundary of two dielectrics
at r = b as required by the boundary conditions. The electric fields are given by
D q
E= = ar (a < r < b )
ε1 4πε1r 2
D q
= = ar (b < r < c )
ε2 4πε 2 r 2

Vac is then evaluated as follows

q  1 (b − a) 1 (c − b) 
a b a
q q
Vac = − ∫ E ⋅ d = − ∫ dr − ∫ dr = +
c c
4πε 2 r 2
b
4πε1r 2
4π  ε1 ab ε 2 cb 

then −1
q  1 1  C1C 2
C= = + = (1-32)
Vac  C1 C 2  C1 + C 2
where
4πε1ab 4πε 2 bc
C1 = , C2 =
b−a c−b
26 PRINCIPLES OF ELECTROMAGNETICS 2

Note that the result corresponds to capacitances in series.

(d) Capacitors in Parallel (Figure 1-13(c))
Figure 1-13(c) shows a spherical capacitor with two dielectrics in par-
allel. We try a solution of the following form:
K
Er = (a < r < b ) (1-33)
r2
in both dielectrics where K is an unknown constant. Note that tangential
E is continuous at the interface of two dielectrics as required by the bound-
ary conditions. The free charge densities ρs1, ρs2 (at r = a) differ. Using the
boundary condition (1-25), the total charge at r = a is given by

εK ε K
q = 2 πa 2 (ρs1 + ρs2 ) = 2 πa 2  1 2 + 2 2  = 2 πK ( ε1 + ε 2 )
 a a 
a a
K K(b − a)
Vab = − ∫ E ⋅ d = − ∫ 2
dr =
b b
r ab

then
q 2π(ε1 + ε 2 )
C= = = C1 + C 2 (1-34)
Vab b−a
( )
ab
where

2πε1ab 2πε 2 ab
C1 = , C2 =
b−a b−a

The result corresponds to capacitors in parallel.

(e)A Series–Parallel Combination (Figure 1-13(d))
Figure 3-13(d) shows a capacitor with three dielectrics. We are
tempted to call this a series–parallel combination. There is no simple form
for E and normal D which will satisfy continuity of both tangential E and
normal D. In fact, this is a difficult problem involving fringing fields and
requiring numerical computation for precise results. However, for thin
spherical shells, the fringing is less significant and the capacitance may be
approximated by the result given below:
 Introduction to Dielectrics 27

C 2 C3
C ≈ C1 + (if a ? c − a)
C 2 + C3
(1-35)
where
2πε1ac
C1 =
c−a

2πε 2 ab
C2 =
b−a
2πε 3 bc
C3 =
c−b

The result in Eq. (1-35) corresponds to a series-parallel combination.

Example 1-7. A Concentric Cylindrical Capacitor (Figure 1-13(a))

Note that Figure 1-13 applies to both spherical and cylindrical
geometries.
The cross section of a concentric cylindrical capacitor of infinite length
is shown in Figure 1-13(a). Surfaces ρ = a, b are perfect conductors. Find
the capacitance per unit length appropriate to this cylindrical geometry.
Solution:
We assume a charge ρℓ per unit length uniformly distributed
over the surface ρ = a and −ρℓ at ρ = b. Then, applying Gauss’ law (see
Example 1-3)
ρl  1 
Eρ = (a < ρ < b)
2πε  ρ 
a a
ρl  b
Vab = − ∫ E ⋅ d = − ∫ Eρ dρ = ln  
b b
2πε  a 
The capacitance per unit length is given as follows:
C Q/l ρl 2πε  F 
= = = (1-36)
l V Vab  b   m 
ln  
and  a
2πεl
C= (1-37)
 b
ln  
 a
28 PRINCIPLES OF ELECTROMAGNETICS 2

For a concentric cylindrical capacitor of finite length ℓ, fringing at the

ends invalidates the results given in Eq. (1-37), which will, however, be
reasonably accurate if   a,b.

1.7 Benjamin Franklin (1706-1790) and the

Beginnings of Electrical Science
Benjamin Franklin can accurately be described as a Renaissance man: he
did so many things and he did them to perfection. The breadth and depth
of his accomplishments are largely unparalleled. History remembers him
as the inventor of the lightning rod, bifocals, and the Franklin stove. We
recall his role in the American Revolution as Ambassador to France. He
appears in numerous famous paintings: performing his kite experiment,
presenting the Declaration of Independence to John Hancock, and in his
fur cap and spectacles in Paris. As memorable as these well-known im-
ages are, they commemorate only a very small number of his accomplish-
ments. To gain some idea of the scope of his activities, we would have to
read some of the biographies of Franklin; the 1938 biography by Carl Van
Doren is still one of the best.
Benjamin Franklin was born in 1706, the youngest boy of a very large
family. He attended school for about two years and, at age ten, began to
work with his father, a soap and candle maker. At age twelve he was ap-
prenticed to his brother James, a printer. He ran away to Philadelphia at
age seventeen, worked as a printer, and eventually set up his own print-
ing shop at age twenty-two. Within four years, he had started writing
and printing “Poor Richard’s Almanac”, a yearly publication which would
become immensely popular, eventually selling 10,000 copies a year and
continuing for 27 years. By his late twenties, then, Franklin was well-
established as a local printer and businessman, and was well on his way
to wider recognition.
Benjamin Franklin wrote two outstanding books during his lifetime.
The first was a description of his discoveries in electricity, Experiments and
Observations on Electricity, published in 1751. This first publication was
crucial in establishing Franklin’s reputation as a scientist. In it, Franklin
put forward the idea that lightning was electricity, proposed a specific
experiment (the sentry box experiment) for verification of the theory, and
 Introduction to Dielectrics 29

recommended the use of grounded rods to protect buildings from light-

ning strikes. A year later, several French scientists carried out his sentry
box experiment, while Franklin performed his famous kite experiment,
resulting in a thorough confirmation of his basic theory. The book and
these lightning experiments catapulted Franklin into the limelight and es-
tablished him as one of the world’s foremost scientists and a leading figure
in the new science of electricity. Accolades and awards from around the
world followed soon after. Two years later, in 1753, Benjamin Franklin
received the Copley medal, an award that carried prestige equivalent to
today’s Nobel Prize.
Carl Van Doren, Benjamin Franklin, Bramhall House, 1987, Reprint of
Viking Press, NY, 1938.
Benjamin Franklin wrote in a very clear and readable style. His judg-
ment in reaching for a conclusion or drawing an hypothesis is very sure.
His style is engaging and frank, and he readily acknowledges his igno-
rance or uncertainty: “You require the reason. I do not know it. Perhaps
you may discover it and then you will be so good as to communicate it to
me.” His treatment of his own hypotheses was very refreshing. He readily
acknowledged them as only tentative models and he was ready to aban-
don them if they proved inadequate: “I am still at a loss about the manner
in which they (clouds) become charged with electricity; no hypothesis I
have yet formed perfectly satisfying me.”
Benjamin Franklin’s second book is his Autobiography, which remains ex-
tremely popular to this day. Particularly interesting is the description of his
self education and self discipline. It is the story of a brilliant and unusual man
struggling to emerge from ignorance and poverty. He relates many of his mis-
takes and failures and does not bask in his successes and honors.
Throughout his life, Benjamin Franklin was also a prolific writer of
newspaper articles. These were usually written to advance some social or
political purpose. They were entertaining as well as educational, and their
impact was always considerable. The undergraduate of today can read
with edification and delight any of these famous articles, including “Miss
Polly Baker,” “Rattlesnakes for Felons,” “The King of Prussia,” and “Rules
by Which a Great Empire May be Reduced to a Small One.” In the age
of Enlightenment, Franklin’s articles led to personal recognition and to
profound respect for the power of his arguments.
30 PRINCIPLES OF ELECTROMAGNETICS 2

To return to Franklin’s role as a scientist, what were the key scientific ac-
complishments which established his reputation? To begin with, one of his
initial contributions was his analysis of the “Leyden jar,” essentially a glass jar
filled with water and held in one’s hand. We may think of it as a jar which
is coated separately with a conductor on the inside and outside. The Leyden
jar is thus a “warped” parallel plate capacitor. Because of the large surface area
and the thinness of the glass, the capacity is large and considerable electric
energy can be stored in the glass. The Leyden jar was a convenient source of
energy and was therefore used in a wide variety of experiments after it was
discovered by the Dutch physicist Pieter Van Mussenbroek in 1746.
Franklin demonstrated that the conductors inside and outside were
oppositely charged, that the total charge was zero and was therefore un-
changed as the capacitor was discharged. He wrote: “These two states of
electricity, the plus and the minus, are combined and balanced in this
miraculous bottle in a manner that I can by no means comprehend”.
The Leyden jar analysis led to a general theory of positive and negative
charges, the conservation of total charge, and the single fluid theory.
Franklin explained his experiments with the motion of a single type of
charge, namely free electrons. He also concluded that the energy was con-
tained in the glass itself. He “unwrapped” the Leyden jar and formed the
parallel plate capacitor. This was merely a flat glass plate coated with con-
ductors on both sides. He then combined a set of plates in either series or
parallel combinations. The series combination was called a “battery”. The
Leyden jar experiments were crucial because they led directly to Franklin’s
general theory of electricity. His theory swept aside the vague previous
theories of “effluvia” and “affluvia” espoused by the Abbé Nollet, one of
the leading figures in the early study of electricity.
Franklin’s earliest writings on electricity in 1747 described some of
his observations on the effects of pointed objects. His colleague Thomas
Hopkinson had carried out some of the early experiments. By 1750
Franklin had proposed the lightning rod as a protection for houses and
for ships and had added the very important grounding wire for both. He
had also proposed his “sentry box” experiment for determining whether
clouds were electrified, and whether lightning was, in fact, electricity. In
1752 the experiment was performed successfully in France with the sen-
try box and in America with the kite.
 Introduction to Dielectrics 31

The drama of these experiments cannot be overrated. Lightning was

proven to be electrical in origin and identical to the small-scale sparks
produced in the laboratory. The clouds were shown to be electrically
charged and the cloud-earth combination analogous to a giant Leyden
jar. Foremost, a solution to the important practical problem of lightning
hazard was proposed, namely the grounded lightning rod. The potential
usefulness of lightning rods was enormous. It is estimated that in one
thirty-three year period in the eighteenth century 386 church towers were
struck and 120 bell ringers were killed in Germany alone.
Ultimately, Benjamin Franklin’s most important contributions to
electrical science were

(1)   The conservation of electric charge.

(2)   The single-fluid theory.
(3)  The concept of positive and negative charges.
(4)   The analysis and extension of the Leyden Jar.
(5)  The effects of pointed objects.
(6)   The grounded lightning rod.
(7) The electrification of clouds.

Now we begin to see why Franklin, who first organized in a coherent

theory the basic ideas of electricity, was regarded in his time, and for some
time after, as one of the leading scientists of electricity. His experiments
and the book which described them gave order and structure to what had
previously been a disorganized collection of theories. The importance of
his work was immediately recognized by leading European scientists. He
received the highest scientific awards of the day, including the prestigious
Copley medal. He was also elected Fellow of the Royal Society and a
member of the French Academy of Sciences. Long before the American
Revolution, Franklin had established himself as the leading scientist of
electrical phenomena.
So Franklin, in his time and for years afterwards, was widely recog-
nized for his pioneering work in electricity. In recent times, however,
we seem to overlook Franklin’s scientific contributions. It is certainly
possible that you do not think of Benjamin Franklin as a scientist, and
there are several reasons for this oversight. Franklin left no equations
32 PRINCIPLES OF ELECTROMAGNETICS 2

which we recognize as fundamental parts of the theory. We have ab-

sorbed many of his contributions and use them unconsciously. As
well, he was a prolific inventor, and it is easy to think of his inven-
tions as representing his primary technical accomplishments, whereas
they were actually the fruits of his scientific curiosity and scientific
theories.
When Benjamin Franklin came to France as Ambassador, his
scientific reputation preceded him. He was not an unknown Phila-
delphian, but an eminent scientist. His greatest experiment, the
lightning experiment, had first been carried out by Frenchmen. His
book had been translated into French and was in wide circulation.
His greatest antagonist, the French Abbé Nollet, wrote two books at-
tacking the Franklinian theory; he was particularly averse to lightning
rods. Franklin chose not to reply directly to any of the Abbé’s many
attacks, and said, “I have an extreme aversion to public altercation
on philosophic points, and have never yet disputed with anyone who
thought fit to attack my opinions.” He was directly defended by sev-
eral Frenchmen, and, if anything, his reputation was enhanced by this
controversy.
How much of his life did Benjamin Franklin devote to his key
electrical experiments? We are surprised to note that it was a relatively
short period of time, namely, the six years from 1747 thru 1752.
Having achieved financial independence, Benjamin Franklin retired
from the printing business in 1748 in order to devote himself more
completely to his electrical research. His book on experiments was
first published in 1751 and the lightning rod experiments were per-
formed in 1752.
Franklin was the quintessential public relationist. He was very
conscious of his image and strove at an early age to project a positive
one. This is not to say that the image necessarily conflicted with reality
but that Franklin fully realized its importance. Thus, “To show that
I was not above my business, I sometimes brought home the paper
I purchased at the stores through the streets on a wheelbarrow.” In
his scientific writings he projected the image of an open, receptive
person. As a result, other scientists flocked to share their results with
him and to perform experiments for him. When Franklin first came
 Introduction to Dielectrics 33

to France as Ambassador during the American Revolution, he was

already known as an eminent scientist. What image then was appro-
priate for the diplomat Franklin? Could he possibly outdo the other
diplomats in their courtly finery? He decided to do just the opposite
of what was expected and appeared in simple dress with a fur cap,
thereby projecting a unique and lasting image.
Franklin was a very highly skilled diplomat. His efforts on behalf of
the fledgling republic led to diplomatic recognition by France, to mas-
sive military aid, and to huge financial loans. There are many examples
of Franklin’s diplomatic skill but one of the most striking occurred at
the conclusion of the Peace Treaty negotiations. The Americans had
decided to break their agreement with France and their instructions
from Congress and carry out separate negotiations with England. Any
other course of action would have made the negotiations almost im-
possible. An extremely favorable treaty had resulted, including exten-
sions of United States Territory to the Mississippi and to Canada. The
negotiations could not have gone better. The Americans had come to
the table with their most talented statesmen (Franklin, Adams, Jeffer-
son, Jay, Laurens) and had overwhelmed the English negotiators. Now
it was necessary to tell the French about the negotiations. Franklin
reported to his French counterpart, Charles Gravier, the comte de
Vergennes, and sent him a copy of the treaty which had already been
signed. Vergennes sent him back a note of reproach. Franklin now
sent Vergennes a famous letter in which he put forth the American
position in the most favorable light, made a partial apology, and said
“the English, I just now learn, flatter themselves they have already di-
vided us.” Somehow this did the trick! Not only were fences mended
but Vergennes arranged yet another much-needed loan to the United
States. The years of careful diplomacy to establish cordial relationships
with Vergennes combined with just the right touch in the letter, had
preserved America’s unique relationship with France.
Additionally, Franklin had recognized early the injustice of the insti-
tution of slavery and had written many articles on the subject, including
those that highlighted unrecognized talents of African Americans. Two
months before his death he submitted to Congress a petition for the abo-
lition of slavery. It was turned down.
34 PRINCIPLES OF ELECTROMAGNETICS 2

1.8 The Force on a Conductor in an Electric Field

We turn now to the calculation of the force on a conductor in an electric
field. This would seem to be straightforward matter but it’s not. There are
subtleties involved which are concerned with the self field Es. In the course
of the discussion we will learn the very useful fact that in calculating
forces we can always subtract the self force. For the force on a conductor,
this means that we can use (E − Es) in place of E.
To find the force on a conductor we need to consider only the forces
on the electric charges which exist on the surface of the conductor. Figure
3-14(a) shows a conductor with a typical patch of the surface with area ds,
surface charge density ρs, and electric field E. The force dF on the patch
is equal to the charge ρsds times the electric field vector. There is just one
problem and that is, “which electric field?”
E
s
Conductor
+
ds + z
+

Figure 1-14(a). A conductor in an electric field

Figure 1-14(b) shows a plot of the electric field near the surface of the
conductor. We choose z as the direction normal to the patch. Ez is a step
ρs
function which jumps from zero to at the surface of the conductor.
εo
For E should we use the value (zero) just inside the conductor, the value
ρs ρs
just outside, the average , or something else?
εo 2ε o
Ez
s
0
E sz
s
2
Ez 0

s
2 0
E sz

Inside Outside

Figure 1-14(b). The normal electric field Ez and a self field Esz at a
conductor surface
 Introduction to Dielectrics 35

Before going any further, let’s remind ourselves of the behavior of

the electric field E and also the behavior of the self field Es which is the
contribution of the patch itself. We are interested in the behavior of these
fields just inside and just outside the conductor surface.
ρs ρ
E = an = a z s (just outside) (1-38a)
ε0 ε0
= 0 (just inside) (1-38b)

1 ρs
E av = {E(outside) + E(inside)} = a z > (1-39)
2 2ε 0
The self field Es due to the small patch itself is the same as that of an in-
finite plane of charge:
ρs ρ
E s = ±a n = a z s (just outside) (1-40a)
2ε 0 2ε 0

ρs
= −a z (just inside) (1-40b)
2ε 0
Figure 1-14(b) shows a plot of E (solid) and Es (dotted). We note that E
and Es have the same jump (discontinuity) at the surface and therefore
we consider using their difference (E − Es) in calculating forces. This is
legitimate since the conductor exerts no net force on itself. The difference
E − Es is continuous:
ρs
E − Es = az = E av (1-41)
2ε 0
Now we see the culprit. It is the self field Es which is discontinuous at
the surface. Its presence in E leads to an ambiguity; its removal resolves
the ambiguity. Its removal is also equivalent to taking the average of fields
inside and out (since Es is odd). Es is the field at the patch due to the patch
itself; E − Es is the field at the patch due to the rest of the conductor, i.e.,
everything but the patch.
Thus the force F on a conductor in an electric field is given as

ρS2
F = ∫∫ (E − E S )ρS ds = ∫∫ E av ρS ds = ∫∫ a n ds
S S
2ε 0
36 PRINCIPLES OF ELECTROMAGNETICS 2

ρS2
F = ∫∫ a n ds (1-42)
S
2ε 0

1.9 Energy and Forces in the Electrostatic Field

In this section, we calculate the energy of, or the work required to as-
semble, electrostatic systems. In addition, we are interested in the forces
exerted on portions of the system. First we consider the work required to
assemble a collection of point charges.

1.9.1 Energy of a Collection of Point Charges

Figure 1-15 shows a set of N point charges at arbitrary locations. The

distance between q1 and q2 is r12 and the distance between qi and qj is rij.
Let us determine the work required to bring up the charges from infinity.
We start with all the charges an infinite distance from the origin and from
each other. The point charges have been given to us; we do not, for the
moment, inquire into the work required to construct them.
No work is required to bring up the first charge (q1) since there are no
forces exerted on it. Charge q2 is brought up working against the poten-
q1
tial of q1. The potential through which q2 is moved is ; the work
q1q 2 4πε 0 r12
required is thus . Similarly, charge q3 works against the forces of
4πε 0 r12
both q1 and q2. The work required to bring up q3, working against the
forces of both q1 and q2, is
q2 qN
z r12

q1

Figure 1-15. A collection of point charges

 Introduction to Dielectrics 37

 q1 q2 
q3  +
 4πε 0 r13 4πε0 r23 

Finally, the work required to bring up qN is

 q1 q2 q N −1 
qN  + ++
 4πε 0 r1N 4πε 0 r2N 4πε 0 r(N −1)N 

qi q j
Consider an N × N matrix of terms . We have taken into
4πε 0 rij
account all the terms above or below the main diagonal and may thus
express the total electric energy We as follows:
N N qi q j
We = ∑ ∑ (1-43a)
i =1 j =1 4πε 0 rij
j> i

where j > i includes terms above the main diagonal. Because of symmetry
about the main diagonal,

1 N N qi q j
We = ∑∑
2 i =1 j=1 4πε 0 rij
(1-43b)
j≠ i

 
1 N  N qi q j  1 N
= ∑ qi ∑ = ∑ q i Vi (1-44)
2 i =1  j=1 4πε 0 rij  2 i =1
 j≠ i ) 

where Vi is the potential at the location of qi due to all charges except qi

1
itself. The factor of in Eq. (1-44) is related to the linear nature of the
2
problem. The first charge is brought up through zero potential, the last
charge is brought up through the total potential V, and the average charge
1
is brought up through V.
2

1.9.2 Energy of Continuous Charge Distribution

For continuous charge distribution ρv(x,y,z), we divide it into infinitesi-

mally small volumes and consider the charge of an infinitesimally small
38 PRINCIPLES OF ELECTROMAGNETICS 2

volume element to be dQ = ρvdv. We then use the summation of Eq.

(1-44) which becomes an integral. Integrating over all volume elements
yields

1
ρv ( x,y,z)V(x,y,z ) dv
2 ∫∫∫
We = (1-45)

Eq. (1-45) is the energy expression for volume charge distribution. For
1
surface charge distribution it reduces to
1 2 ∫∫ ρs Vds if the surface is a
conductor it reduces to where Q is the total charge on the conductor
2
and V is the potential of the conductor, since the potential V is constant
on the conductor. The energy stored in the capacitor, which consists of
two conductors carrying charges
+Q, −Q (see Figure 1-12(a)), can be expressed as follows:

1 1 1 1 1
We = QV1 + ( − Q)V1 = Q(V1 − V2 ) = QV= CV 2
2 2 2 2 2

where V1 and V2 are the potentials of the two conductors and V is the
potential difference or voltage.

Energy in Terms of Fields

A second general form may be obtained by extending the volume of Eq.

(1-45) to infinity. This is permitted since ρv = 0 outside the charge distri-
bution. Then
1 1
We = ∫∫∫
2 all space
ρv Vdv = ∫∫∫ (∇ ⋅ D)Vdv
2 all space
1
{∇ ⋅ ( VD) − D ⋅ ∇V } dv
2 all∫∫∫
=
space

where we have used the vector identity

∇ ⋅ ( VD) = V(∇ ⋅ D) + (∇V ) ⋅ D

 Introduction to Dielectrics 39

We now use E = −∇V and apply the divergence theorem to the first term
of the integral above to obtain

1  
We =  ∫∫∫ E ⋅ D dv + ' ∫∫ V D) ⋅ ds
2 all space sphere at ∞ 
1 1
For a charge distribution of finite extent,
V ~ , |D|~ 2 , ds ~ r 2 , as r → ∞
r r
,
and the surface integral above vanishes. Thus
1
2 all∫∫∫
We = E ⋅ D dv (1-46)
space

In Eq. (1-45), the energy is viewed as being stored with the charge source,
whereas Eq. (1-46) shows the energy being stored with the fields. Both
equations yield the same result for an electrostatic system, but for time-
varying sources and fields, Eq. (1-46) makes more sense because the fields
move in space, carrying the energy,

Example 1-8. A Spherical Cloud of Charge

Consider a spherical cloud of uniform volume charge density ρv:

ρv = ρo (r < a)
Find the work required to assemble the cloud of charge.
Solution:
Applying Gauss’ Law:

4 3
' ∫∫∫ D ⋅ ds = 4πr 2 Dr = ρo πr (r ≤ a)
S
3
4
= ρo πa 3 (r ≥ a)
3

ρo r
Dr = ε o E r = (r ≤ a )
3
ρ a3
= o 2 (r ≥ a )
3r
40 PRINCIPLES OF ELECTROMAGNETICS 2

The electric potential is given by

r r
ρo a 3 ρo a 3
V = − ∫ E ⋅ d = − ∫ E r
dr = r ≥ a dr = − ∫
(r ≥ a)
3ε 0 r

∞ ∞
2
3ε 0 r

a r
ρo a 2 ρo r 2

∞
∫ a
∫
= − E r r ≥ a dr − E r r ≤ a dr =
2ε 0
−
6ε 0
(r ≤ a)
(1-47)
1
2 ∫∫∫
We = ` v V dv

2  a r2  2
2 a 2
1
= ∫ ∫∫`o  −  r sin ‚ dr d‚ dφ
2 0 0 0 2 0 6 0 
4 ` 2o a 5 3Q 2
= =
15 0 20 0 a

4
(where Q = πa ρo ). We can also derive the same result, using the en-
3

3
ergy in terms of fields, Eq. (1-46) [see Problem 3-26]. Note that
We → 0 as a → 0 (ρo constant)
We → ∞ as a → 0 (Q constant)
This implies that (a) a small volume of uniform finite ρo has negligible
energy and (b) a point charge Q has infinite energy.

1.9.3 Forces and Torques in Terms of Energy

Let us consider an electrostatic system consisting of conductors and di-

electrics. Allow one of the parts of the system to move under the influence
of the force F acting upon it. We allow a small (virtual) displacement
Δx of the movable part. Consider two situations: (a) constant charge for
which conductors are isolated from each other and the total charge on
each conductor remains constant during the displacement, (b) constant
potentials for which potentials between conductors are constant during
the displacement. Figure 1-16 shows parallel plates with a dielectric block
partially inserted between plates. In the constant charge case the plates are
unconnected and charge Q remains constant during a small displacement
Δx. x is defined as the distance between a point on the fixed plates and the
movable block of dielectric. In the constant voltage case an ideal voltage
source is connected between plates. In this case, additional charge ΔQ is
deposited on the upper plate during the displacement Δx.
 Introduction to Dielectrics 41

L1 L1
Q

x d x d

Q
x x
(a) Constant charge case. (b) Constant voltage case.

Figure 1-16. A dielectric slab partially inserted between parallel

plates

Constant Charge

In the constant charge case, charge is fixed but potentials and fields may vary.
Therefore the electrostatic energy We may vary with displacement Δx. As
the dielectric block moves through displacement Δx, under the influence of
force Fx, mechanical work FxΔx is done. The mechanical energy is thereby
changed. It could be stored (as in a spring) or it could show up as kinetic
1 2
energy ( mv ) . In any case, both the mechanical and electrostatic energy
2
may change. Conservation of energy may be expressed as follows:
Fx = ∆x+We = 0
.

∆We ∂We (Q,x )

Fx = − →− (as ∆x → 0) (1-48a)
∆x ∂x

Constant Voltage

In the constant voltage case, the voltage is fixed but the charge may vary.
If the charge is increased by ΔQ then the electrostatic energy, which is
1 1 1
equal to VQ , increases by V∆Q to V(Q + ∆Q) . The mechan-
2 2 2
ical work done is FxΔx as before. The battery is a third contributor to
energy storage. The battery does work (supplies energy) V(ΔQ) in moving
charge ΔQ through voltage V. Conservation of energy may be expressed
as follows:
1
∆We + Fx ∆x = V∆Q = 2 ∆We (since ∆We = V∆Q).
2
(1-48b)
∆We ∂We ( V,x )
Fx = →− as ∆x → 0
∆x ∂x
If the dielectric block is hinged at one end and free to rotate within
the parallel plates, the mechanical work done is Tθ Δθ in each case, where
42 PRINCIPLES OF ELECTROMAGNETICS 2

θ is an angle between the fixed plates and the rotatable block of dielectric.
Tθ is the torque. The corresponding results are

∂We (Q,θ)
Tθ = − (constant charge) (1-49a)
∂θ

∂We ( V,θ)
= (constant voltage) (1-49b)
∂θ
Forces and displacement in directions other than x may also be consid-
ered. The results may be generalized as follows:

FQ = −∇We (1-50a)

FV = +∇We (1-50b)
where the subscripts Q and V represent constant charge and voltage,
respectively.

Example 1-9. Force on a Dielectric Block between Parallel Plates

Consider the configuration of Figure 1-16. The plates are rectangular
with plate area equal to L1L2. Assuming that L1, L2  d, we neglect
fringing fields. In other words, we assume that the electric field is uniform
in the region between the plates and zero outside. The electrostatic energy
of the system and the forces are given by

1 1 1 Q2
We = QV = CV 2 = (1-51)
2 2 2 C

∂We ( V,x ) 1 2 dC(x)

Fx = = V (1-52a)
∂x 2 dx

∂We (Q,x ) 1 d  1  1 Q 2 dC(x)

=− = − Q2 = (1-52b)
∂x 2 dx  C(x)  2 C 2 dx

Note that the constant charge and constant voltage cases yield identi-
cal expressions, as they must, for Fx. Considering the configuration as a
 Introduction to Dielectrics 43

parallel connection of two parallel-plate capacitors and using the result of

Eq. (1-28),

εL 2 x ε 0 L 2 (L1 − x)
C(x)= +
d d
dC(x) L 2 (ε − ε 0 )
=
dx d
Then
V 2 L 2 (ε − ε 0 )
Fx = (1-53)
2d

Since ε > εo, Fx is always positive and the dielectric slab is pulled in be-
tween the plates. Note that we have assumed a uniform electric field in
finding the force. However, it is not this uniform field but a nonuniform
fringing field that pulls the dielectric into the capacitor. Without knowing
the fringing field, which is difficult to calculate, we were able to find the
force on the dielectric from the change in the electrostatic energy of the
system.
44 PRINCIPLES OF ELECTROMAGNETICS 2
 CHAPTER 2

Electric Currents and

Conductive Materials

2.1 Introduction
Now we turn to problems which involve moving charges and resultant
current flow. We are particularly interested in steady currents such as
those which arise when steady voltages are applied across conductive or
resistive materials. In Chapter 1, perfect dielectrics with zero conductivity
were studied. Now we consider dielectrics with finite conductivity.
First we consider definitions of current. Then we establish some gen-
eral relationships, including the equation of continuity, Ohm’s law, Joule’s
law, and relaxation time. These relationships may be applied to the gen-
eral time-varying case, as well as to the special case, of steady currents. Fi-
nally, we consider a new boundary condition which is required for steady
currents, namely, the continuity of normal volume current density.

2.2 Electric Current

Let’s consider charges which are moving within a conductor. Consider
any surface S within the conductor. Let Q(t) be the charge which has
passed through the surface. Then the current I(t) through the surface is
defined as
dQ(t)
I(t) = (2-1)
dt
The unit of current is the ampere [A], or coulomb/second, named after
A. M. Ampère. The current is thus a measure of the rate at which charges
are crossing the surface S.
46 PRINCIPLES OF ELECTROMAGNETICS 2

2.2.1 Volume Current Density

Consider charges which are moving through a volume. To define the vol-
ume current density J at a point, construct a small area ds perpendicular
to current flow. dI is the current passing through the surface. Then the
magnitude of vector J may be represented as follows:
dI
J= [ A/m 2 ] (2-2)
ds
Vector J has magnitude J and a direction identical to that of positive
charge movement or current flow. The current through a differential
surface ds not necessarily perpendicular to current flow (Figure 2-1)
is given by

dI = J ⋅ ds

J ds
S

Figure 2-1. Volume current flow through an arbitrary surface

and the total current I passing through surface S is

I = ∫∫ J ⋅ ds (2-3)
S
Next consider charges which are moving at a steady drift velocity v
(Figure 2-2). Construct a rectangular solid of volume (ds v dt) as shown
with surface ds perpendicular to current flow. In time dt all charges in the
box will pass through the surface ds. Let there be N charges per unit vol-
ume, each of charge q. The current passing through the surface ds is then

dQ(t) Nq(ds vdt)

dI = = = Nq dsv
dt dt
vdt

ds an

+
v
+
+

Figure 2-2. Drift velocity

 Electric Currents and Conductive Materials 47

and dI
J= = Nq v where v =| v |
ds
J = N q v = ρv v

J = ρv v (2-4)
In the most general case where charges are of different types,

J = ∑ N i q i v i = ∑ ρvi v i
i i

where Ni, qi, vi, ρvi are number of charges per unit volume, charge per
carrier, velocity and volume charge density, respectively, for each type of
charge carrier.

2.2.2 Surface Current Density

Surface current flows in a zero-thickness layer on a surface (Figure 2-3).

To define the surface current density Js at a point we construct a line dℓ⊥.
The subscript ⊥ denotes that the line is perpendicular to the direction of
current flow. Then

d
Js

Figure 2-3. Surface current density Js

dI
Js = [ A/m] (2-5)
dl ⊥

and the vector Js has magnitude Js and a direction identical to that of

current flow. Note that surface current density has the unit of amps/m
whereas volume current density has the unit of amps/m2. The total cur-
rent passing through the strip (Figure 2-3) is given by

I = ∫ J s d ⊥ (2-6)
48 PRINCIPLES OF ELECTROMAGNETICS 2

2.2.3 Line Current

For line current, that is, current which flows in a filament of zero cross
section, we merely count the charges passing through a point to deter-
mine the current I there. Then

dQ
I=
dt
I is the current vector, which has magnitude I and whose direction is that
of current flow.

2.3 The Equation of Continuity

Current is produced by the motion of charges. Current and charge quan-
tities are thus related. The fundamental relationship may be obtained by
considering an arbitrary volume V bounded by surface S (Figure 2-4).
Current I(t) is flowing inward through the surface and charge Q(t) is ac-
cumulating within.

Q(t)
V
S

J
I(t)
Figure 2-4. The equation of continuity

The total current flowing into the volume V is

I(t) = ' ∫∫ J ⋅ ds = − ∫∫∫ ∇ ⋅ J dv

S V

(by the divergence theorem)

dQ(t) d ∂ρ
= = ∫∫∫ ρv dv = ∫∫∫ v dv
dt dt V V
∂t

(by Leibnitz’s rule for the derivative of an integral)

 Electric Currents and Conductive Materials 49

The above relationship is equivalent to conservation of charge within

volume V.

∂ρ v
∫∫∫ ∇ ⋅ J dv = ∫∫∫
V V
∂t
dv

The volume V of integration is arbitrary and thus the integrands must be

equal:
∂ρ v
∇⋅ J = − (2-7)
∂t

Equation (2-7) is called the equation of continuity. It expresses the fun-

damental relationship between current and charge densities. It plays an
important role for time-varying fields in Chapter 8.
In the steady-state (charge and current do not vary in time), Eq. (2-7)
reduces to
∇⋅ J = 0 for steady currents (2-8)

2.4 Ohm’s Law and Conductive Materials

The electromagnetic form of Ohm’s law gives the relationship between
volume current density J and the electric field E in a conducting material:
J = σE (2-9)
where σ is the conductivity of the material. The conductivity of various
materials is given in Table 2-2. The table includes both good conductors
and good insulators. Silver, copper, and gold are the three best conduc-
tors. Silver is best but it is expensive and tarnishes readily which reduces
the conductivity. Often materials are silver-plated to improve their con-
ductivity. Copper is much less expensive than silver or gold and so is the
conductor of choice where large quantities are needed. Gold is often used
in microelectronics. It is the most corrosion-free of the three. Finally, we note
that rubber and fused quartz are two of the best insulators.
50 PRINCIPLES OF ELECTROMAGNETICS 2

Table 2-1 Conductivities

Material Conductivity (S/m)

Silver 6.2 × 107
Copper 5.8 × 107
Gold 4.1 × 107
Aluminum 3.5 × 107
Brass 2.6 × 107
Iron 107
Sea Water 4
Silicon 4.4 × 10−4
Distilled Water 10−4
Dry Earth 10−5
Glass 10−12
Rubber 10−15
Fused Quartz 10−17

To see that Eq. (2-9) is equivalent to the usual form of Ohm’s law
in circuit theory, consider a rectangular conductor with uniform J, E, σ
throughout (Figure 2-5). Multiply both sides of Eq. (2-9) by S, the cross
sectional area, and equate components to obtain

σS
JS = σES = (E )

or

σS
I= V (since I =JS, V =El).
l
or

l
V= I
σS

S
J

l
z

Figure 2-5. Uniform current flow in a resistor

 Electric Currents and Conductive Materials 51

We have, then, a linear relationship between V and I. The term


is called the resistance R. The familiar circuit relation
σS
V = IR (2-10)
is thus obtained. The resistance,
V l
R= = (2-11)
I σS
applies to any homogeneous resistor of length ℓ and uniform cross section
 l
S.   , the inverse of conductivity, is called resistivity.
 √
“Why should J be proportional to E? Why should J not increase in-
definitely as electrons are accelerated by the electric field E?” To answer
this question, we need to understand the processes of current flow in a
conductor, which can be explained as follows. In the absence of an ap-
plied electric field, at room temperature, the conduction electrons are
moving at about 106 m/s through the “lattice” structure of the material.
The movement is in random directions so that the average vector velocity
is zero and there is no net current (J = 0). We then apply a voltage across
the conductor and an electric field E is established within the conduc-
tor. The conduction electrons are accelerated by the force qE = − eE and
produce a current J in the direction of E. However, the electrons are
accelerated for only a very short time about (about 10−14 s). They travel
only a very short distance (about 10−8m) before colliding with the lattice
structure and being scattered in arbitrary directions. Before collision, the
electron motion contributes to current J in the direction of E. Immedi-
ately after collision, the direction is arbitrary and so the electron motion
does not, on average, contribute to J in the direction of E. The contribu-
tion to current is “lost”, so to speak. Then the process starts all over again.
The electron is accelerated, contributes to J between collisions and loses
its contribution upon collision. There is however an average “drift” veloc-
ity contributing to J. This is merely the average velocity acquired during
the brief acceleration period between collisions. The drift velocity is very
slow, on the order of 10−4 m/s.
Thus we have electrons which are moving very rapidly (106 m/s) in
random directions. Superimposed on this very rapid motion is a very slow
drift velocity (10−4 m/s) which contributes to J in the direction of E. If we
52 PRINCIPLES OF ELECTROMAGNETICS 2

double the voltage, the time between collisions (the acceleration period)
is unchanged since it is determined primarily by the much larger random
velocity. The acceleration time is constant, the acceleration force qE is
doubled, and so the average “drift” velocity is doubled. Thus we have a
linear system and J is proportional to E as in Eq. (2-9). The constant σ is
called conductivity.
It should be pointed out that a conductor is usually electrically neutral.
The moving negative charges travel through a stationary lattice structure
which is oppositely charged. For linear homogeneous media characterized
by a constant conductivity σ, the conductor or resistor is uncharged (ρv =
0) within the volume. This is analogous to the case for the linear homo-
geneous dielectric (ρpv = 0). Surface charge will in general accumulate at
interfaces. Volume charge density may exist (ρv ≠ 0) for inhomogeneous
conductors.

2.5 Georg Simon Ohm (1789-1854) and the

Discovery of Ohm’s Law
1827 should have been a very good year for Georg Simon Ohm. He had
just published his comprehensive book on resistive circuits. He had in-
vestigated the subject very thoroughly and had confirmed his theory with
very carefully performed experiments. He was now almost 40 years old.
Surely he would receive the recognition which he deserved and with it,
perhaps, the university appointment which he had long sought.
In his book, The Galvanic Circuit Mathematically Investigated, Ohm
had explored the subject as completely as was possible at the time. He
had shown that in a simple loop circuit with a battery of voltage V and
uniform conductor size, the current in the wire is uniform throughout
the length. Thus the circuit can be characterized with a single current,
I. He also demonstrated that the potential falls uniformly with distance
along the wire. Both of these essential facts, which are taken for granted
by modern students, were unknown at the time.
Then, he had established the linear relationship between voltage and
current
V = IR (2-10)
 Electric Currents and Conductive Materials 53

where the resistance R of the circuit is proportional to wire length ℓ and

inversely proportional to wire cross section S:
kl l
R= = (2-11)
S √S
where k is a constant, the resistivity, the inverse of conductivity, which
depends on the material of the wire. Ohm also determined the relative
conductivity of various metals, the best conductors being silver, copper,
gold, in that order, as shown in Table 2-2. Ohm’s experiments were very
sophisticated and reliable. He used the Seebeck thermoelectric battery, in-
vented in 1821, because of its predictable internal resistance. To measure
the current and its magnetic field, he used the torsion balance compass
needle galvanometer invented by Coulomb in his measurements of the
inverse square law for magnetic poles.
There is a direct analogy between resistive circuits and heat flow. The
quantity of heat flowing in a wire is proportional to the temperature dif-
ference between the ends of the wire and inversely proportional to ther-
mal resistance, which, in turn, is proportional to the length of the wire
and inversely proportional to its cross-sectional area. Ohm had studied
Fourier’s book, The Analytical Study of Heat, and had, with great insight,
seen the analogy to electrical current flow. Voltage, current, and electrical
resistance are analogous to temperature difference, heat flow, and thermal
resistance, respectively.
Note that our familiar rules for combinations of resistances in series
and parallel follow directly from Ohm’s law. For instance, placing two
identical sections of wire in series doubles the resistance since total length
ℓ is doubled; placing them in parallel halves the resistance since cross-
section area S is doubled.
Georg Ohm was born in Erlangen, Germany, in 1789, the son of a
locksmith. He and his younger brother Martin were well-educated and
Georg received a Ph.D. from the University of Erlangen in 1812. He then
held several teaching positions in secondary schools, finally settling in
Cologne, where he taught for about nine years. It was in Cologne that he
carried out most of his basic research. It was a long and difficult process.
He had to educate himself in the mathematics and science of the times
and become an expert in the current experimental techniques of electri-
cal science, all the while carrying the heavy teaching load of a secondary
54 PRINCIPLES OF ELECTROMAGNETICS 2

school. We are told that he was a very serious, effective and innovative
teacher.
The early 19th century was a time of great advances in electrical sci-
ence. In 1800 Alessandro Volta invented the voltaic cell or battery, which
supplied a steady source of current. This was much preferable to the Ley-
den jar or capacitor as a source. The voltaic cell served as a great stimulus
to research. Oersted’s dramatic discovery in 1820 of the magnetic field of
an electric current is described in Volume 4. Within a year, this discovery
led to a method of measuring current by determining its magnetic field.
This was done by noting the deflection of a compass needle. In this way,
Volta and Oersted provided a steady current and the means to measure it,
making it possible for Ohm to carry out his experiments.
Unfortunately, Ohm’s book was not well received in Germany. In
fact, his theory was widely rejected, condemned as too complicated by
some and too simple by others. In general, reaction to his work was very
negative and the backward educational establishment criticized him in
response. Ohm felt that he must resign his post in Cologne and went to
teach at a military school in Berlin.
However, scientists in other countries slowly began to recognize his
work and to give it the respect which it deserved. He received the Copley
medal in 1841 and finally, in 1849, twenty-two years after the publica-
tion of his book, he was appointed Professor of Physics at the University
of Munich, where he continued to teach, honored and revered, until his
death at 65 in 1854.
Henry Cavendish and other researchers had preceded Ohm in his
discovery of Ohm’s law but the thoroughness and completeness of Ohm’s
work guaranteed that he would receive primary credit for the discovery.

2.6 Power – Joule’s Law

The power dissipated in a conductive or resistive material may be calcu-
lated from the electric field and the current density as follows:

P = ∫∫∫ E ⋅ J dv [W] (2-12)

V
where V is the volume of the resistor. E · J is thus a volume density of power
dissipated with units of watts/m3. Equation (2-12) is known as Joule’s law.
 Electric Currents and Conductive Materials 55

To see the correspondence with circuit theory, consider the resis-

tor of Figure 2-5 with uniform E, J, σ. Making use of Eq. (2-9), we have
1 2 J2
P = ∫∫∫ σE dv = ∫∫∫ J dv = σE (Sl ) = (Sl)
2 2

V V
σ σ
Sσ V 2
= (El )2
=
l R
l
= ( JS)2 = I2 R
Sσ
= ∫ E dl∫∫ J ds = (El )( JS)=VI

2.7 Relaxation Time

In Volume 1, we noted that charges placed within a conductor move al-
most immediately by mutual repulsion toward the surface of the conduc-
tor. In other words, volume charge density disappears very rapidly giving
rise to surface charge density. Now we are in a position to examine the
phenomenon quantitatively and ask the question “how fast?”. Consider a
linear homogeneous medium of conductivity σ, permittivity ε, volume V
and surface S (Figure 2-6).

V v(r,t)
,

Figure 2-6. Relaxation time

We start with the equation of continuity:

∂ρ v σ
∇⋅ J = − = ∇ ⋅ (σE ) = ∇ ⋅  D
∂t ε 
σ σ
= ∇ ⋅ D = ρv [Eq.(3-10b) is used.]
ε ε

∂ρv σ
=  −  ρv
∂t  ε 

56 PRINCIPLES OF ELECTROMAGNETICS 2

The solution to the differential equation above is

σ
− t
ρv ( x, y, z, t ) = ρvo ( x, y, z, t ) e ε
(2-13)
or

− t/τ
ρv ( x, y, z, t ) = ρvo ( x, y, z, t ) e
ε
where τ = is called the relaxation time. We may interpret Eq. (2-13)
σ
in the following way.
Suppose we start with an initial volume charge density ρvo (x,y,z) within
volume V at time t = 0. Then the volume charge density decays expo-
nentially at every point inside the volume. The net charge will eventually
reside on the surface S since charge is conserved. So the surface charge
density is increasing as the volume charge density decreases. τ is the time
required for the volume charge density to decay to 1/e or about 37% of
its value at any point in V. Table 2-2 shows relaxation times for some
conductors and insulators. Relaxation time is so short, on the order of
10−19 seconds, for good conductors, that we may consider the process to
be practically instantaneous for good conductors.

Table 2-2. Relaxation Times

Material Relaxation Time

Copper 2.5 × 10−19 (s)
Sea Water 2 × 10−10 (s)
Distilled Water 10−6 (s)
Fused Quartz 105 (s) or about 11 days

2.8 Boundary Conditions for Steady Currents

In the presence of steady currents it is necessary to add one boundary
condition, namely, the continuity of normal J, to the basic boundary
conditions already derived in Chapter 1. Figure 2-7 shows a boundary
between two arbitrary media (1), (2) with characteristics (σ1, ε1) and (σ2,
ε2), respectively.
 Electric Currents and Conductive Materials 57

an J1n ,
1 1 1
J2n ,
2 2 2

Figure 2-7. Continuity of normal J

Consider the normal components J1n, J2n of volume current density at the
interface. J2n specifies the charge flowing into the interface and J1n deter-
mines charge flowing out from the interface. J1n, J2n must be equal or else
charge would accumulate indefinitely at the interface. Therefore

J1n = J2n for steady currents (2-14)

In summary, the three basic boundary conditions at an interface are:
E1t = E2t Tangential E is continuous
D1n − D2n = ρs Normal D is discontinuous by the amount of free
surface charge
J1n = J2n Normal J is continuous

Example 2-2. Dissimilar Dielectrics in a Parallel Plate Capacitor

Figure 2-8 shows parallel plates of area A and separation d1 + d2. The
parallel plate region is filled with two different materials of permittivity,
conductivity, ε1, σ1, and ε2, σ2, respectively. The switch is closed and the
steady state is eventually reached. Find the current, the total resistance,
and the free surface charge at the interface between dissimilar materials.
Perfect conductor - Area A

d2
2, 2
V s z
d1 1, 1

Perfect conductor - Area A

Figure 2-8. Dissimilar dielectrics in a parallel-plate capacitor

Solution:
In such problems, we concentrate on the current. Because of the conti-
nuity of normal J in the steady state, we assume a single volume current
density for both regions:

J = −a z Jo
58 PRINCIPLES OF ELECTROMAGNETICS 2

We find the voltage by applying Ohm’s law [Eq. (5-9)]:

Jo d1 Jo d 2 d d 
V = − ∫ E ⋅ d = E1d1 + E 2 d 2 = + = Jo  1 + 2 
σ1 σ2  σ1 σ 2 
and
V
Jo =
 d1 d 2 
 σ + σ 
1 2

then
VA
I = ∫∫ J ⋅ ds = Jo A =
 d1 d 2 
 σ + σ 
and 1 2

V d d
R= = 1 + 2 (2-15)
I σ1 A σ 2 A
Note that the potential at the interface (z = d1) is

d1
d1 σ1
E1d1 = Jo or V =
σ1 d1 d 2
+
σ1 σ 2

with respect to the potential at z = 0. It is determined solely by the con-

ductivities σ1, σ2.
To find the surface charge density at the interface we use the
boundary condition for the normal component of D,
Jo J
D1 = −a z ε1 , D1 = −a z o ε 2
σ1 σ2
ε ε 
ρS = D2z − D1z = Jo  1 − 2 
 σ1 σ 2 
ε ε
ρS ⇒ 0 if 1 = 2
σ1 σ 2
(2-16)

 Electric Currents and Conductive Materials 59

Where did the charge ρs come from? Normal J is continuous in the

steady state and leads to no charge accumulation. We may consider that
the charge ρs is deposited on the interface during the transient period
when normal J is not necessarily continuous. A net charge is deposited
only if relaxation times are unequal.

2.9 A Relationship Between Capacitance and

Resistance
Figure 2-9 shows two conductors in an infinite homogeneous medium of
permittivity ε or conductivity σ or both.

+Q
—Q
1 2
Conductor
Conductor

or

V+

Figure 2-9. A relationship between capacitance and resistance

Let’s consider first the ideal dielectric with permittivity ε. A voltage V is

applied between (perfect) conductors, an electric field E is established in
the dielectric and charges Q, −Q reside on the conductors. The capaci-
tance is
60 PRINCIPLES OF ELECTROMAGNETICS 2

Q
∫∫ D ⋅ ds ε ∫∫ E ⋅ ds
C= = S
= S (2-17)
V V V

Next consider the resistor with conductivity σ. Voltage V is applied and

a steady current I flows in the circuit and through the resistor (J = σ E in
the resistor). The resistance is

1 V V
R= = = (2-18)
G I σ ∫∫ E ⋅ ds
S

since
I = ∫∫ J ⋅ ds = σ∫∫ E ⋅ ds
S S

G, the inverse of resistance, is called the conductance. We assume

that the voltage V is identical in the two cases. The electric field E then
is also identical in the two cases because of uniqueness and, multiplying
Eqs. (2-17) and (2-18), we obtain

C ε
RC = = (2-19)
G σ

What happens as we change from a perfect to an imperfect dielec-

tric in Figure 5-9? The only thing that changes is that current flows and
increases with increasing conductivity σ. V, E, Q, C remain unchanged
as σ increases. Only J and R vary. Thus the resistance and capacitance
problems are decoupled.

Example 2-2. Shunt Resistance and Capacitance of a Coaxial Line

Figure 2-10(a) shows the cross section of a coaxial line with inner,
outer radii a, b, respectively and length ℓ. The coaxial line is completely
filled with a material of permittivity ε and conductivity σ. A voltage V is
applied between inner and outer conductors and a current I then flows
radially outward from inner to outer conductor. Find the shunt resistance
and capacitance.
 Electric Currents and Conductive Materials 61

,
,

V a

b d

(a) A coaxial line. (b) A thin cylindrical shell.

=b

J
,
=a

(c) A truncated wedge.

Figure 2-10. A coaxial line with inner, outer radii a, b, respectively

and length ℓ

Solution:
 l
The volume current density is independent of φ and varies as   so
 ρ
that the total current is independent of ρ:
I
J = aρ (2-20)
2πρ
An integration over a cylindrical surface of radius ρ and length ℓ shows
that the total current is I:

l 2
I
∫∫ J ⋅ ds = ∫ ∫ 0 0
2πρl
ρ dφdz = I

Then
J I
E= = aρ
σ 2πρσl

a
I I  b
V = − ∫ E ⋅ d = − ∫ dρ = ln  
b
2πσlρ 2πσl  a 
2πσl V V  1
I= ; E = aρ
ln(b/a) ln(b/a)  ρ 
62 PRINCIPLES OF ELECTROMAGNETICS 2

and
V ln(b/a)
R= =I= (2-21)
I 2πσl

2πσ G 2πσ
G= ; = (2-22)
ln(b/a)  ln(b/a)

The charge Q in length ℓ (on the inner conductor) is determined as follows:

V l 2πεl
Q = (area )ρS = (area )Dρ = (2πσl )ε = V
ln(b/a) a ln(b/a)
Then
Q 2πσ C 2πε
C= = ; = (2-23)
V ln(b/a)  ln(b/a)
C ε
Note that R in Eq. (2-21) and C in Eq. (2-23) satisy RC = = as
G σ
in Eq. (2-19).
The resistance R may also be determined by considering a thin cylin-
drical shell of radius ρ, length ℓ (Figure 2-10(b)).

dl dρ
dR = =
σA 2πρσl
b b
dρ ln(b/a)
R = ∫ dR = ∫ =
a a
2πσlρ 2πσl

which agrees with Eq. (2-21).

The conductance G may be determined by considering a truncated
wedge (Figure 2-10(c)) with radial current flow:
I
J=a
ρ ; ∫∫ J ⋅ ds = I
αρ
I I ln(b/a)
E = aρ

αρlσ ∫
; V = − E ⋅ d =
αlσ
(2-24)

V I ln(b/a) αlσ
R= = ; G= (wedge)
I αlσ I ln(b/a)
 Electric Currents and Conductive Materials 63

For a thin wedge of angle dφ,

dφσ
dG =
ln(b/a)
and for the whole coaxial line,

2π
lσ d φ 2πσl
G = ∫ dG = ∫ ln(b/a) = ln(b/a)
0

which agrees with Eq. (2-22).

The power losses in the coaxial line are determined as follows:

P = ∫∫∫ E ⋅ J dv = ∫∫∫ σEρ2 dv

l 2¿ b
σV 2 ρdρdφdz V 2 (2πlσ )
=∫∫∫ =
0 0 0
ln 2 (b/a)ρ2 ln(b/a)
V2
= = I2 R
R
Example 2-3. A Truncated Conical Resistor
Figure 2-11 shows a truncated conical resistor bounded by surfaces
r = a, r = b, and θ = 15°. The material of the resistor is inhomogeneous:
r
σ = σo
a
A voltage V is applied between surfaces r = a, b. Find the electric field and
the resistance between the surfaces r = a, b.

r=b

=15˚

J
J
V

r=a

˚
30
+

Figure 2-11. A truncated conical resistor

64 PRINCIPLES OF ELECTROMAGNETICS 2

Solution:
First we assume a radial, inverse square current density which makes
the total current independent of r:
C1
J = ar
r2

2 /12
C1r 2 sin θ dθdφ
I = ∫∫ J ⋅ ds = ∫ ∫
0 0
r2
= 2 πC1 (1 − cos15 )
J C 1 Ca
E = = a r 21 = ar 1 3
σ r σ r σo r
o
a
a
Ca 1 1
V = − ∫ E ⋅ d = − ∫ E r dr = 1  2 − 2 
b
2σ o  a b 

and

 b2 a  2b2 a 2  1 
C1 = 2σ o  2 V; E = a V
 b − a 2  r
b2 − a 2  r 3 
(2-25)
V b2 − a 2
R= =
I 4πσ o ab2 (1 − cos15 )
 INDEX
Boundary conditions, 17–20 Joule’s law, 54–55
for steady currents, 56–59
Linear dielectrics, 12
Capacitance, 21–28 Linear, homogeneous dielectrics,
and resistance, relationship 13–14
between, 59–64 Line current, 48
Cavendish, Henry, 54
Collection of point charges, energy of, Nollet, Abbé, 32
36–37
Concentric cylindrical capacitor, Oersted, Hans, 54
27–28 Ohm, Georg Simon, 52–54
Conductance, 60 Ohm’s law, 49–52
Conductive materials, 49, 50
Continuous charge distribution, Parallel plate capacitor, 22–23
energy of, 37–40 dissimilar dielectrics in, 57–59
Permittivity, 12
Dielectric constant, 12, 13 Polarization, 2–4
Displacement vector, 11–17 charges, 7
Drift velocity, 46 vector, 2
Polarized dielectric, 1
Electrets, 2 electric field of, 4–10
Electrical science, 28–33
Electric current, 45–48 Relaxation time, 55–56
line current, 48 Resistance, 51
surface current density, 47 and capacitance, relationship
volume current density, 46–47 between, 59–64
Electric susceptibility, 12 Resistivity, 51
Equation of continuity, 48–49
Spherical capacitors, 24–27
Force on conductor in electric field, Steady currents, boundary conditions
34–36 for, 56–59
Forces, in terms of energy, 40–43 Surface current density, 47
Franklin, Benjamin, 28–33
Torques, in terms of energy, 40–43
Gauss’ law, 12
Gravier, Charles, 33 Volta, Alessandro, 54
Volume current density, 46–47
Hopkinson, Thomas, 30
EBOOKS Principles of Electromagnetics
FOR THE 2—Dielectric and Conductive
ENGINEERING Materials
LIBRARY
Arlon T. Adams
Create your own Jay K. Lee
Customized Content
Bundle—the more
books you buy,
the greater your
discount!
Arlon T. Adams (PhD, University of Michigan) is a Professor
THE CONTENT Emeritus of Electrical and Computer Engineering at Syracuse
• Manufacturing University, where he taught and conducted research in elec-
Engineering tromagnetics for many years, focusing on antennas and micro-
• Mechanical waves. He served as electronics officer in the U. S. Navy and
& Chemical worked as an engineer for the Sperry Gyroscope Company.
Engineering
• Materials Science
& Engineering
• Civil & Jay Kyoon Lee (Ph.D., Massachusetts Institute of Technology)
Environmental is a Professor of Electrical Engineering and Computer Science
Engineering at Syracuse University, where he teaches Electromagnetics,
• Advanced Energy among other courses. His current research interests are elec-
Technologies tromagnetic theory, microwave remote sensing, waves in aniso-
tropic media, antennas and propagation. He was a Research
THE TERMS Fellow at Naval Air Development Center, Rome Air Develop-
ment ­Center and Naval Research Laboratory and was an Invited
• Perpetual access for
Visiting Professor at Seoul National University in Seoul, Korea.
a one time fee
He has received the Eta Kappa Nu Outstanding Undergraduate
• No subscriptions or
Teacher Award (1999), the IEEE Third Millennium Medal (2000),
access fees
and the College Educator of the Year Award from the Technol-
• Unlimited
ogy Alliance of Central New York (2002).
concurrent usage
• Downloadable PDFs
• Free MARC records

For further information,

a free trial, or to order,
contact:
[email protected]