This content has been downloaded from IOPscience. Please scroll down to see the full text.

Download details:

IP Address: 146.155.54.32
This content was downloaded on 19/12/2019 at 14:20

Please note that terms and conditions apply.

You may also be interested in:

Linear and nonlinear sloshing in a circular conical tank

I P Gavrilyuk, I A Lukovsky and A N Timokha

Fermion-dyon bound states and fermion number fractionisation

M Ravendranadhan and M Sabir

Electron energy distributions in Townsend discharges in hydrogen

T E Kenny and J D Craggs

Influence of residual achromatic aberration on the isochronicity in the FAIR collector ring
S Litvinov, A Dolinskii, I Koop et al.

Influence of the phase front parameters on the path of a beam of rays

Nail R Sadykov

Mathematical modeling is also physics—interdisciplinary teaching between mathematics and physics in

Danish upper secondary education
Claus Michelsen

Discharge time of a cylindrical leaking bucket

M Blasone, F Dell’Anno, R De Luca et al.

Propagation of singularities along characteristics of Maxwell's equations

Elisabetta Barletta and Sorin Dragomir

Wind energy: an application of Bernoulli's theorem generalized to isentropic flow of ideal gases
R De Luca and P Desideri
Electromagnetism
Problems and solutions
Electromagnetism
Problems and solutions

Carolina C Ilie
State University of New York at Oswego, USA

Zachariah S Schrecengost
State University of New York at Oswego, USA

Morgan & Claypool Publishers

Copyright ª 2016 Morgan & Claypool Publishers

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, photocopying, recording
or otherwise, without the prior permission of the publisher, or as expressly permitted by law or
under terms agreed with the appropriate rights organization. Multiple copying is permitted in
accordance with the terms of licences issued by the Copyright Licensing Agency, the Copyright
Clearance Centre and other reproduction rights organisations.

Rights & Permissions

To obtain permission to re-use copyrighted material from Morgan & Claypool Publishers, please
contact info@morganclaypool.com.

ISBN 978-1-6817-4429-2 (ebook)

ISBN 978-1-6817-4428-5 (print)
ISBN 978-1-6817-4431-5 (mobi)

DOI 10.1088/978-1-6817-4429-2

Version: 20161101

IOP Concise Physics

ISSN 2053-2571 (online)
ISSN 2054-7307 (print)

A Morgan & Claypool publication as part of IOP Concise Physics

Published by Morgan & Claypool Publishers, 40 Oak Drive, San Rafael, CA, 94903 USA

IOP Publishing, Temple Circus, Temple Way, Bristol BS1 6HG, UK

To my family, my mentors, and my students — CCI

To my friends, family, and mentors — ZSS

 Contents

Preface x
Acknowledgements xii
About the authors xiii

1 Mathematical techniques 1-1

1.1 Theory 1-1
1.1.1 Dot and cross product 1-1
1.1.2 Separation vector 1-1
1.1.3 Transformation matrix 1-2
1.1.4 Gradient 1-2
1.1.5 Divergence 1-2
1.1.6 Curl 1-3
1.1.7 Laplacian 1-3
1.1.8 Line integral 1-4
1.1.9 Surface integral 1-4
1.1.10 Volume integral 1-4
1.1.11 Fundamental theorem for gradients 1-4
1.1.12 Fundamental theorem for divergences (Gauss’s theorem, 1-4
Green’s theorem, divergence theorem)
1.1.13 Fundamental theorem for curls (Stoke’s theorem, curl theorem) 1-4
1.1.14 Cylindrical polar coordinates 1-4
1.1.15 Spherical polar coordinates 1-5
1.1.16 One-dimensional Dirac delta function 1-5
1.1.17 Theory of vector fields 1-5
1.2 Problems and solutions 1-5
Bibliography 1-34

2 Electrostatics 2-1
2.1 Theory 2-1
2.1.1 Coulomb’s law 2-1
2.1.2 Electric field 2-1
2.1.3 Gauss’s law 2-2
2.1.4 Curl of E ⃗ 2-2
2.1.5 Energy of a point charge distribution 2-2

vii
 Electromagnetism

2.1.6 Energy of a continuous distribution 2-2

2.1.7 Energy per unit volume 2-2
2.2 Problems and solutions 2-3
Bibliography 2-35

3 Electric potential 3-1

3.1 Theory 3-1
3.1.1 Laplace’s equation 3-1
3.1.2 Solving Laplace’s equation 3-1
3.1.3 General solutions 3-4
3.1.4 Method of images 3-5
3.1.5 Potential due to a dipole 3-6
3.1.6 Multiple expansion 3-6
3.1.7 Monopole moment 3-6
3.2 Problems and solutions 3-6
Bibliography 3-29

4 Magnetostatics 4-1
4.1 Theory 4-1
4.1.1 Magnetic force 4-1
4.1.2 Force on a current carrying wire 4-1
4.1.3 Volume current density 4-1
4.1.4 Continuity equation 4-2
4.1.5 Biot–Savart law 4-2
4.1.6 Divergence of B ⃗ 4-2
4.1.7 Ampère’s law 4-2
4.1.8 Vector potential 4-2
4.1.9 Magnetic dipole moment 4-3
4.1.10 Magnetic field due to dipole moment 4-3
4.2 Problems and solutions 4-3
Bibliography 4-26

5 Electric fields in matter 5-1

5.1 Theory 5-1
5.1.1 Induced dipole moment of an atom in an electric field 5-1
5.1.2 Torque on a dipole due to an electric field 5-1
5.1.3 Force on a dipole 5-1

viii
 Electromagnetism

5.1.4 Energy of a dipole in an electric field 5-2

5.1.5 Surface bound charge due to polarization P ⃗ 5-2
5.1.6 Volume bound charge due to polarization P ⃗ 5-2
5.1.7 Potential due to polarization P ⃗ 5-2
5.1.8 Electric displacement 5-2
5.1.9 Gauss’s law for electric displacement 5-2
5.1.10 Linear dielectrics 5-2
5.1.11 Energy in a dielectric system 5-3
5.2 Problems and solutions 5-3
Bibliography 5-26

6 Magnetic fields in matter 6-1

6.1 Theory 6-1
6.1.1 Torque on a magnetic dipole moment 6-1
6.1.2 Force on a magnetic dipole 6-1
6.1.3 H-field 6-1
6.1.4 Linear media 6-2
6.1.5 Surface bound current due to magnetization M⃗ 6-2
6.1.6 Volume bound current due to magnetization M⃗ 6-2
6.2 Problems and solutions 6-2
Bibliography 6-14

ix
 Preface

We wrote this book of problems and solutions having in mind the undergraduate
student—sophomore, junior, or senior—who may want to work on more problems
and receive immediate feedback while studying. The authors strongly recommend
the textbook by David J Griffiths, Introduction to Electrodynamics, as a first source
manual, since it is recognized as one of the best books on electrodynamics at the
undergraduate level. We consider this book of problems and solutions a companion
volume for the student who would like to work on more electrostatic problems by
herself/himself in order to deepen their understanding and problems solving skills.
We add brief theoretical notes and formulae; for a complete theoretical approach we
suggest Griffiths’ book. Every chapter is organized as follows: brief theoretical notes
followed by the problem text with the solution. Each chapter ends with a brief
bibliography.
We plan to write a second volume on electrodynamics, which will start with
Maxwell’s equations and the conservation laws, and then discuss electromagnetic
(EM) waves, potentials and fields, radiation, and relativistic electrodynamics.
We follow here the notation of Griffiths, and use r ⃗ for the vector from a source
point r ⃗′ to the field point r ⃗ . Please note that rˆ = rr = ∣ rr⃗ ⃗ −− rr⃗ ′⃗ ′ ∣ and, as you see, this
⃗

notation already greatly simplifies complex equations, but you need to be careful
with your notation, in particular if you only use cursive or typed letters. Also, we use
the same notation s for the distance to the z-axis in cylindrical coordinates as is used
in Griffiths’ book.
The chosen units are SI units—the international system. The reader should be
aware that other books may employ either the Gaussian system (CGS) or the
Heaviside–Lorentz (HL) system. The Coulomb force in each of the systems is as
follows,
SI system:
1 q1q2
F⃗ = rˆ
4πε0 r 2
CGS:
q1q2
F⃗ = rˆ
r2
HL:
1 q1q2
F⃗ = rˆ
4π r 2
Some of the problems are typical practice problems with the pedagogical role of
improving understanding and problem solving skills. Several of the problems
presented here appear in a variety of undergraduate textbooks on EM as they are
classic examples; however, we felt it would be incomplete to omit these problems as

x
 Electromagnetism

they are fundamental to the study of EM. We also present problems that are more
general in nature, which may be a bit more challenging. We tried to maintain a
balance between the two types of problems, and we hope that the readers will enjoy
this variation and have as much thrill and excitement as we had while creating and
solving these problems.

xi
 Acknowledgements

We want to thank to Dr Ilie’s students, Nicholas Jira, Vincent DeBiase, Ian Evans,
and Andres Inga, who contributed to the editing (typing) of this book. We are
particularly grateful to our illustrator, Julia D’Rozario, for making all of the figures.
We thank Dr Ildar Sabirianov for providing useful suggestions. We thank the
administration at SUNY Oswego and the office of Research and Individualized
Student Experiences for overall support. We are grateful to Dr Peter Dowben, from
the University of Nebraska at Lincoln, who thought that such a project has a niche.
A thought of appreciation to Dr Charles Ebner, from the Ohio State University for
his perfect Electrodynamics course. Also many thanks to our editors, Joel Claypool,
Publisher at Morgan & Claypool Publishers, Jeanine Burke, Consulting Editor at
the IOP Concise Physics e-book program, and Jacky Mucklow, Production Team
Manager at the Institute of Physics. Lastly, we thank to our families and friends for
their sense of humor, encouragement, and for keeping us sane and happy.

xii
 About the authors

Carolina C Ilie
Carolina C Ilie is an Associate Professor with tenure at the State
University of New York at Oswego. She taught Electromagnetic
Theory for almost ten years and designed various problems for her
students’ exams, group work, and quizzes. Dr Ilie obtained her PhD
in Physics and Astronomy at the University of Nebraska at Lincoln,
an MSc in Physics at Ohio State University and another MSc in
Physics at the University of Bucharest, Romania. She received the
President’s Award for Teaching Excellence in 2016 and the Provost
Award for Mentoring in Scholarly and Creative Activity in 2013. She lives in
Central New York with her spouse, also a physicist, and their two sons.
Photograph courtesy of James Russell/SUNY Oswego Office of Communications
and Marketing.

Zachariah S Schrecengost
Zachariah S Schrecengost is a State University of New York
alumnus. He graduated summa cum laude with a BS degree having
completed majors in Physics, Software Engineering, and Applied
Mathematics. He took the Advanced Electromagnetic Theory
course with Dr Ilie and was thrilled to be involved in creating this
book. He brings to the project both the fresh perspective of the
student taking electrodynamics, as well as the enthusiasm and talent
of an alumnus who is an electrodynamics and upper level
mathematics aficionado. Mr Schrecengost works as a software engineer in Syracuse
and is preparing to begin his graduate school studies in physics.

Julia R D’Rozario
Julia R D’Rozario (illustrator) graduated from the State University
of New York at Oswego in December 2016 where she completed a
BS in Physics and a BA in Cinema and Screen Studies, and
completed a minor in Astronomy by May 2016. She completed the
Advanced Electromagnetic Theory course with Dr Ilie and has
much experience of the arts through her career in film. Ms
D’Rozario contributes her knowledge of electrodynamics and her
talent in drawing using Inkscape software. Her future aim is to
attend graduate school and continue to combine her passions for physics and
cinema.

xiii
 IOP Concise Physics

Electromagnetism
Problems and solutions
Carolina C Ilie and Zachariah S Schrecengost

Chapter 1
Mathematical techniques

There are a variety of mathematical techniques required to solve problems in

electromagnetism. The aim of this chapter is to provide problems that will build
confidence in these techniques. Concepts from vector calculus and curvilinear
coordinate systems are the primary focus.

1.1 Theory
1.1.1 Dot and cross product
Given vectors A⃗ = Ax xˆ + Ay yˆ + Az zˆ and B ⃗ = Bxxˆ + Byyˆ + Bzzˆ

A ⃗ ⋅ B ⃗ = Ax Bx + Ay By + Az Bz = AB cos θ
xˆ yˆ zˆ
⃗ ⃗
A × B = Ax Ay Az with A ⃗ × B ⃗ = AB sin θ
Bx By Bz

where A = ∣A⃗ ∣ = Ax2 + A y2 + Az2 , B = ∣B ∣⃗ = Bx2 + By2 + Bz2 , and θ is the angle
between A⃗ and B ⃗ .

1.1.2 Separation vector

This notation is outlined by David J Griffiths in his book Introduction to
Electrodynamics (1999, 2013). Given a source point r ′⃗ and field point r ⃗ , the
separation vector points from r ′⃗ to r ⃗ and is given by

r ⃗ = r ⃗ − r ⃗′ = (x − x′)xˆ + (y − y′)yˆ + (z − z′)zˆ

doi:10.1088/978-1-6817-4429-2ch1 1-1 ª Morgan & Claypool Publishers 2016

 Electromagnetism

and the unit vector pointing from r ′⃗ to r ⃗ is

r⃗ r ⃗ − r ⃗′ (x − x′)xˆ + (y − y′)yˆ + (z − z′)zˆ
rˆ = = = .
r r ⃗ − r ⃗′ (x − x′)2 + (y − y′)2 + (z − z′)2

As explained by Griffiths, this notation greatly simplifies later equations.

1.1.3 Transformation matrix

Given vector A⃗ = Ax xˆ + Ay yˆ + Az zˆ in coordinate system K, the components of A⃗ in
coordinate system K ′ are determined by rotational matrix R given by
⎛ Rxx Rxy Rxz ⎞
⎜ ⎟
R = ⎜ Ryx Ryy Ryz ⎟
⎜ ⎟
⎝ Rzx Rzy Rzz ⎠
with
⎛ ⎞
⎜ Ax′ ⎟ ⎛ Ax ⎞
⎜ A ′ ⎟ = R ⎜ Ay ⎟ .
⎜ y⎟ ⎜⎜ ⎟⎟
⎜ ′⎟ ⎝ Az ⎠
⎝ Az ⎠

1.1.4 Gradient
Given a scalar function T , the gradients for various coordinate systems are given below.
Cartesian
∂T ∂T ∂T
∇T = xˆ + yˆ + zˆ
∂x ∂y ∂z
Cylindrical
∂T 1 ∂T ˆ ∂T
∇T = sˆ + ϕ+ zˆ
∂s s ∂ϕ ∂z
Spherical
∂T 1 ∂T ˆ 1 ∂T ˆ
∇T = rˆ + θ+ ϕ
∂r r ∂θ r sin θ ∂ϕ

1.1.5 Divergence
Given vector function v ⃗ , the divergences for various coordinate systems are given
below.
Cartesian
∂vx ∂vy ∂v
∇ ⋅ v⃗ = + + z
∂x ∂y ∂z

1-2
 Electromagnetism

Cylindrical
1 ∂ 1 ∂vϕ ∂v
∇ ⋅ v⃗ = ( svs ) + + z
s ∂s s ∂ϕ ∂z
Spherical
1 ∂ 2 1 ∂ 1 ∂vϕ
∇ ⋅ v⃗ = 2
r ∂r
( )
r vr +
r sin θ ∂θ
(
sin θ vθ + )
r sin θ ∂ϕ

1.1.6 Curl
Given vector function v ⃗ , the curls for various coordinate systems are given below.
Cartesian
⎛ ∂v ∂vy ⎞ ⎛ ∂vx ∂v ⎞ ⎛ ∂vy ∂v ⎞
∇ × v⃗ = ⎜ z − ⎟xˆ + ⎜ − z ⎟yˆ + ⎜ − x ⎟zˆ
⎝ ∂y ∂z ⎠ ⎝ ∂z ∂x ⎠ ⎝ ∂x ∂y ⎠

Cylindrical
⎛ 1 ∂vz ∂vϕ ⎞ ⎛ ∂vs ∂v ⎞ 1⎡ ∂ ∂v ⎤
∇ × v⃗ = ⎜ − ⎟sˆ + ⎜ − z ⎟ϕˆ + ⎢ ( svϕ ) − s ⎥zˆ
⎝ s ∂ϕ ∂z ⎠ ⎝ ∂z ∂s ⎠ s ⎣ ∂s ∂ϕ ⎦

Spherical
1 ⎡∂ ∂vθ ⎤ 1 ⎡ 1 ∂vr ∂ ⎤
∇ × v⃗ = ⎢ (sin θ vϕ ) − ⎥rˆ + ⎢ − (rvϕ )⎥θˆ
r sin θ ⎣ ∂θ ∂ϕ ⎦ r ⎣ sin θ ∂ϕ ∂r ⎦
1⎡ ∂ ∂v ⎤
+ ⎢ ( rvθ ) − r ⎥ϕˆ
⎣
r ∂r ∂θ ⎦

1.1.7 Laplacian
Given a scalar function T , the Laplacians for various coordinate systems are given
below.
Cartesian
∂ 2T ∂ 2T ∂ 2T
∇2 T = + +
∂x 2 ∂y 2 ∂z 2
Cylindrical
1 ∂ ⎛ ∂T ⎞ 1 ∂ 2T ∂ 2T
∇2 T = ⎜s ⎟ + +
s ∂s ⎝ ∂s ⎠ s 2 ∂ϕ 2 ∂z 2
Spherical
1 ∂ ⎛ 2 ∂T ⎞ 1 ∂ ⎛ ∂T ⎞ 1 ∂ 2T
∇2 T = ⎜r ⎟ + ⎜sin θ ⎟ +
r 2 ∂r ⎝ ∂r ⎠ r 2 sin θ ∂θ ⎝ ∂θ ⎠ r 2 sin2 θ ∂ϕ 2

1-3
 Electromagnetism

1.1.8 Line integral

Given vector function v ⃗ and path P , a line integral is given by
b⃗
∫a ⃗ P v ⃗ ⋅ dl⃗,
where a ⃗ and b ⃗ are the end points, and dl⃗ is the infinitesimal displacement vector
along P . In Cartesian coordinates dl⃗ = dx xˆ + dy yˆ + dz zˆ .

1.1.9 Surface integral

Given vector function v ⃗ and surface S , a surface integral is given by

∫S v ⃗ ⋅ da ⃗ ,

where da ⃗ is the infinitesimal area vector that has direction normal to the surface.
Note that da ⃗ always depends on the surface involved.

1.1.10 Volume integral

Given scalar function T and volume V , a volume integral is given by

∫V T dτ ,

where dτ is the infinitesimal volume element. In Cartesian coordinates

dτ = dx dy dz .

1.1.11 Fundamental theorem for gradients

b⃗
∫a ⃗ P (∇T ) ⋅ dl⃗ = T ( b ⃗) − T ( a ⃗)
1.1.12 Fundamental theorem for divergences (Gauss’s theorem, Green’s theorem,
divergence theorem)

∫V (∇ ⋅ v ⃗)dτ = ∮S v ⃗ ⋅ da ⃗
1.1.13 Fundamental theorem for curls (Stoke’s theorem, curl theorem)

∫S (∇ × v ⃗) ⋅ da ⃗ = ∮P v ⃗ ⋅ dl⃗
1.1.14 Cylindrical polar coordinates
Here our infinitesimal quantities are

dl⃗ = ds sˆ + s dϕ ϕˆ + dz zˆ

and
dτ = s ds dϕ dz .

1-4
 Electromagnetism

1.1.15 Spherical polar coordinates

Here our infinitesimal quantities are

dl⃗ = dr rˆ + r dθ θˆ + r sin θ dϕ ϕˆ

and
dτ = r 2 sin θ dr dθ dϕ .

1.1.16 One-dimensional Dirac delta function

The one-dimensional Dirac delta function is given by

⎧0 x ≠ a
δ (x − a ) = ⎨
⎩∞ x = a

and has the following properties

∞

∫ δ(x − a )dx = 1
−∞

∫ f (x )δ(x − a )dx = f (a )
−∞

1
δ(kx ) = δ(x ).
k

1.1.17 Theory of vector fields

If the curl of a vector field F ⃗ vanishes everywhere, then F ⃗ can be written as the
gradient of a scalar potential V :

∇ × F ⃗ ↔ F ⃗ = −∇V.
If the divergence of a vector vanishes everywhere, then F ⃗ can be expressed as the curl
of a vector potential A⃗ :

∇ ⋅ F ⃗ = 0 ↔ F ⃗ = ∇ × A.⃗

1.2 Problems and solutions

Problem 1.1. Given vectors A⃗ = 3xˆ + 9yˆ + 5zˆ and B ⃗ = xˆ − 7yˆ + 4zˆ , calculate
A⃗ ⋅ B ⃗ and A⃗ × B ⃗ using vector components and find the angle between A⃗ and B ⃗
using both products.

1-5
 Electromagnetism

Solution
A ⃗ ⋅ B ⃗ = (3xˆ + 9yˆ + 5zˆ ) ⋅ (xˆ − 7yˆ + 4zˆ )
= (3)(1) + (9)( −7) + (5)(4) = 3 − 63 + 20
A ⃗ ⋅ B ⃗ = −40
xˆ yˆ zˆ
A⃗ × B ⃗ = 3 9 5
1 −7 4
= [(9)(4) − ( −7)(5)]xˆ + [(1)(5) − (3)(4)]yˆ + [(3)( −7) − (1)(9)]zˆ
Aˆ × Bˆ = 71xˆ − 7yˆ − 30zˆ
To find the angle θ between A⃗ and B̂ we must first calculate A and B :
A = 32 + 9 2 + 5 2 = 115
B = 12 + ( −7)2 + 42 = 66 .
Using the dot product, we have
⎛ −40 ⎞
A ⃗ ⋅ B ⃗ = AB cos θ → θ = cos−1 ⎜ ⎟
⎝ 115 66 ⎠
θ = 117.3°.
Using the cross product, we have
A ⃗ × B ⃗ = AB sin θ → 712 + ( −7)2 + ( −30)2 = 115 66 sin θ
θ = 62.7°.
Note, however, that we can see that the angle between A⃗ and B ⃗ is greater than 90°.
For any argument γ , −90° ⩽ sin−1(γ ) ⩽ 90°. Since the angle between A⃗ and B ⃗ is
greater than 90°, we must adjust for this by subtracting our angle from 180°.
Therefore, θ = 180° − 62.7° = 117.3° as expected.

1-6
 Electromagnetism

Problem 1.2. The scalar triple product states A⃗ ⋅ (B ⃗ × C ⃗ ) = B ⃗ ⋅ (C ⃗ × A⃗ ). Prove

this by expressing each side in terms of its components.
Solution Starting with the left-hand side, the cross product is

xˆ yˆ zˆ
⃗ ⃗
B × C = Bx By Bz
Cx Cy Cz

(
= (ByCz − BzCy )xˆ + BzCx − BxCz yˆ + (BxCy − ByCx )z.
ˆ )
Now, dotting A⃗ with (B ⃗ × C ⃗ )

( ) ( ) (
A ⃗ ⋅ B ⃗ × C ⃗ = Ax ByCz − BzCy + Ay BzCx − BxCz + Az BxCy − ByCx ) ( )
= Ax ByCz − Ax BzCy + Ay BzCx − Ay BxCz + Az BxCy − Az ByCx

( ) (
= Bx CyAz − CzAy + By CzAx − CxAz + Bz CxAy − CyAx ) ( )
A ⃗ ⋅ B ⃗ × C ⃗ = B ⃗ ⋅ ⎡⎣ CyAz − CzAy xˆ + CzAx − CxAz yˆ + CxAy − CyAx zˆ⎤⎦ .
( ) ( ) ( ) ( )
Note the term in brackets is precisely C ⃗ × A⃗ , therefore

( )
A⃗ ⋅ B ⃗ × C⃗ = B ⃗ ⋅ C⃗ × A⃗ ( )
as desired. This procedure can easily be applied again to prove the final part of the
triple product,

( ) ( )
A⃗ ⋅ B ⃗ × C⃗ = B ⃗ ⋅ C⃗ × A⃗ = C⃗ ⋅ A⃗ × B ⃗ . ( )
Problem 1.3. Given source vector r ′⃗ = r cos θ xˆ + r sin θ yˆ and field vector r ⃗ = zzˆ ,
find the separation vector r ⃗ and the unit vector rˆ .
Solution We have

r ⃗ = r ⃗ − r ⃗′ = zzˆ − r cos θ xˆ + r sin θ yˆ

( )
r ⃗ = −r cos θ xˆ − r sin θ yˆ + zzˆ .

To determine the unit vector rˆ , we must first find the magnitude of r ⃗ ,

r= ( −r cos θ )2 + ( −r sin θ )2 + z 2 = (
r 2 cos2 θ + sin2 θ + z 2 = ) r2 + z2 .

1-7
 Electromagnetism

So

r⃗ −r cos θ xˆ − r sin θ yˆ + zzˆ

rˆ = = .
r r2 + z2

Problem 1.4. Given A⃗ in coordinate system K , find the rotational matrix to give the
components in system K ′.

Solution From the figures, we have

Ax′ = Ay , A y′ = Ax , Az′ = −Az .

We want to find the rotational matrix R that satisfies

⎛ ⎞
⎜ Ax′ ⎟ ⎛ Ax ⎞
⎜ A ′ ⎟ = R⎜ Ay ⎟ .
⎜ y⎟ ⎜⎜ ⎟⎟
⎜ ′⎟ ⎝ Az ⎠
⎝ Az ⎠

From our equations above

⎛ ⎞
⎜ Ax′ ⎟ ⎛ 0 1 0 ⎞ ⎛ Ax ⎞
⎜ A ′ ⎟ = ⎜ 1 0 0 ⎟ ⎜ Ay ⎟ .
⎜ y ⎟ ⎜⎜ ⎟⎟ ⎜⎜ ⎟⎟
⎜ ′⎟ ⎝ 0 0 − 1⎠ ⎝ Az ⎠
⎝ Az ⎠

1-8
 Electromagnetism

Therefore,
⎛0 1 0 ⎞
⎜ ⎟
R = ⎜ 1 0 0 ⎟.
⎜ ⎟
⎝ 0 0 −1⎠

Problem 1.5. Find the gradient of the following functions:

a) T = x 4 + y 2 + z 3
b) T = x 2 ln y z 3
c) T = x 2y + z 3
Solutions
a) T = x 4 + y 2 + z 3
∂T ∂T ∂T
∇T = xˆ + yˆ + zˆ = 4x 3xˆ + 2yyˆ + 3z 2zˆ
∂x ∂y ∂z
b) T = x 2 ln y z 3

∂T ∂T ∂T x 2z 3
∇T = xˆ + yˆ + zˆ = 2xz 3 ln y xˆ + yˆ + 3x 2z 2 ln y zˆ
∂x ∂y ∂z y
c) T = x 2y + z 3
∂T ∂T ∂T
∇T = xˆ + yˆ + zˆ = 2xyxˆ + x 2yˆ + 3z 2zˆ
∂x ∂y ∂z

Problem 1.6. Find the divergence of the following functions:

a) v ⃗ = xyxˆ − 2y 2 zyˆ + z 3zˆ
b) v ⃗ = (x + y )xˆ + (y + z )yˆ + (z + x )zˆ
Solutions
a) v ⃗ = xyxˆ − 2y 2 zyˆ + z 3zˆ
∂vx ∂vy ∂v
∇ ⋅ v⃗ = + + z = y − 4yz + 3z 2
∂x ∂y ∂z
b) v ⃗ = (x + y )xˆ + (y + z )yˆ + (z + x )zˆ
∂vx ∂vy ∂v
∇ ⋅ v⃗ = + + z =1+1+1=3
∂x ∂y ∂z

Problem 1.7. Find the curl of the following functions:

a) v ⃗ = xyxˆ − 2y 2 zyˆ + z 3zˆ
b) v ⃗ = (x + y )xˆ + (y + z )yˆ + (z + x )zˆ
c) v ⃗ = sin x xˆ + cos y yˆ

1-9
 Electromagnetism

Solutions
a) v ⃗ = xyxˆ − 2y 2 zyˆ + z 3zˆ

⎛ ∂v ∂vy ⎞ ⎛ ∂vx ∂v ⎞ ⎛ ∂vy ∂v ⎞

∇ × v⃗ = ⎜ z − ⎟xˆ + ⎜ − z ⎟yˆ + ⎜ − x ⎟zˆ
⎝ ∂y ∂z ⎠ ⎝ ∂z ∂x ⎠ ⎝ ∂x ∂y ⎠

⎡ 3 ∂ −2y 2 z ⎤⎥ ⎡
⎢
=⎢
∂(z )
−
( x
)
ˆ+⎢
∂( xy )
−
( ) ⎤⎥yˆ
∂ z3
∂y ∂z ⎥ ⎢ ∂z ∂x ⎥
⎣ ⎦ ⎣ ⎦
⎡ ∂ − 2y 2 z ⎤
+ ⎢⎢
( −
)
∂( xy ) ⎥
zˆ
∂x ∂y ⎥
⎣ ⎦

( )
= 0 + 2y 2 xˆ + (0 − 0)yˆ + (0 − x )zˆ

∇ × v ⃗ = 2y 2 xˆ − xzˆ

b) v ⃗ = (x + y )xˆ + (y + z )yˆ + (z + x )zˆ

⎛ ∂v ∂vy ⎞ ⎛ ∂vx ∂v ⎞ ⎛ ∂vy ∂v ⎞

∇ × v⃗ = ⎜ z − ⎟xˆ + ⎜ − z ⎟yˆ + ⎜ − x ⎟zˆ
⎝ ∂y ∂z ⎠ ⎝ ∂z ∂x ⎠ ⎝ ∂x ∂y ⎠

⎡ ∂(z + x ) ∂(y + z ) ⎤ ⎡ ∂(x + y ) ∂(z + x ) ⎤

=⎢ − ⎥xˆ + ⎢ − ⎥yˆ
⎣ ∂y ∂z ⎦ ⎣ ∂z ∂x ⎦

⎡ ∂(y + z ) ∂(x + y ) ⎤
+⎢ − ⎥zˆ
⎣ ∂x ∂y ⎦

∇ × v ⃗ = −xˆ − yˆ − zˆ

c) v ⃗ = sin x xˆ + cos y yˆ
⎛ ∂v ∂vy ⎞ ⎛ ∂vx ∂v ⎞ ⎛ ∂vy ∂v ⎞
∇ × v⃗ = ⎜ z − ⎟xˆ + ⎜ − z ⎟yˆ + ⎜ − x ⎟zˆ
⎝ ∂y ∂z ⎠ ⎝ ∂z ∂x ⎠ ⎝ ∂x ∂y ⎠

⎡ ∂(0) ∂(cos y ) ⎤ ⎡ ∂(sin x ) ∂(0) ⎤

=⎢ − ⎥xˆ + ⎢ − ⎥yˆ
⎣ ∂y ∂z ⎦ ⎣ ∂z ∂x ⎦
⎡ ∂(cos y ) ∂(sin x ) ⎤
+⎢ − ⎥zˆ = 0
⎣ ∂x ∂y ⎦

1-10
 Electromagnetism

Problem 1.8. Prove ∇ × (∇T ) = 0.

Solution

xˆ yˆ zˆ
∂ ∂ ∂
( )
∇ × ∇T = ∂x ∂y ∂z
∂T ∂T ∂T
∂x ∂y ∂z

⎡ ∂ ⎛ ∂T ⎞ ∂ ⎛ ∂T ⎞⎤ ⎡ ∂ ⎛ ∂T ⎞ ∂ ⎛ ∂T ⎞⎤
=⎢ ⎜ ⎟ − ⎜ ⎟⎥xˆ + ⎢ ⎜ ⎟ − ⎜ ⎟⎥yˆ
⎣ ∂y ⎝ ∂z ⎠ ∂z ⎝ ∂y ⎠⎦ ⎣ ∂z ⎝ ∂x ⎠ ∂x ⎝ ∂z ⎠⎦
⎡ ∂ ⎛ ∂T ⎞ ∂ ⎛ ∂T ⎞⎤
+⎢ ⎜ ⎟− ⎜ ⎟⎥zˆ
⎣ ∂x ⎝ ∂y ⎠ ∂y ⎝ ∂x ⎠⎦

∇ × (∇T ) = 0.

Problem 1.9. Find the Laplacian of the following functions:

a) T = x + y 2 + xz + 3
b) T = e x + sin y cos(2z )
c) T = sin x cos y
d) v ⃗ = xyxˆ + z 2yˆ − 2zˆ
Solutions
a) T = x + y 2 + xz + 3

∂ 2T ∂ 2T ∂ 2T
∇2 T = + + =0+2+0=2
∂x 2 ∂y 2 ∂z 2

b) T = e x + sin y cos(2z )

∂ 2T ∂ 2T ∂ 2T
∇2 T = + + = e x − sin y cos(2z ) − 4 sin y cos(2z )
∂x 2 ∂y 2 ∂z 2
= e x − 5 sin y cos(2z )

c) T = sin x cos y

∂ 2T ∂ 2T ∂ 2T
∇2 T = + + = −sin x cos y − sin x cos y = −2 sin x cos y
∂x 2 ∂y 2 ∂z 2

1-11
 Electromagnetism

d) v ⃗ = xyxˆ + z 2yˆ − 2zˆ

⎛ ∂ 2v ∂ 2vx ∂ 2vx ⎞ ⎛ ∂ 2vy ∂ 2vy ∂ 2vy ⎞
∇2 v ⃗ = ⎜ 2x + + ⎟xˆ + ⎜ + + ⎟yˆ
⎝ ∂x ∂y 2 ∂z 2 ⎠ ⎝ ∂x 2 ∂y 2 ∂z 2 ⎠
⎛ ∂ 2v ∂ 2vz ∂ 2vz ⎞
+ ⎜ 2z + + ⎟zˆ
⎝ ∂x ∂y 2
∂z 2 ⎠

∇2 v ⃗ = (0 + 0 + 0)xˆ + (0 + 0 + 2)yˆ+(0 + 0 + 0)zˆ = 2yˆ

Problem 1.10. Test the divergence theorem with v ⃗ = 2xyxˆ + y 2 z 3yˆ + (x 2z − 2y )zˆ
and the volume below.

Solution The divergence theorem states

∫V ∇ ⋅ v ⃗ dτ = ∮S v ⃗ ⋅ da ⃗.
Starting with the left-hand side, we have the divergence

(
∇ ⋅ v ⃗ = 2y + 2yz 3 + x 2 = 2y z 3 + 1 + x 2 . )
We must split the volume into two pieces, (a) 0 ⩽ y ⩽ 1 and (b) 1 ⩽ y ⩽ 2.
(a)
2 2 1
⎡2y z 3 + 1 + x 2⎤dy dx dz = 52
∫ ∇ ⋅ v ⃗ dτ = ∫∫∫ ⎣ ( ) ⎦ 3
0 0 0

1-12
 Electromagnetism

(b)
2 2 4 − 2y
⎡2y z 3 + 1 + x 2⎤dy dx dz = 176
∫ ∇ ⋅ v ⃗ dτ = ∫∫ ∫ ⎣ ( ) ⎦ 15
0 1 0

So,
52 176 436
∫V ∇ ⋅ v ⃗ dτ =
3
+
15
=
15
.

Now we solve ∮ v ⃗ ⋅ da ⃗ , which must be evaluated over the six sides.

S

(i) We must split this region into two sections (a) and (b), and da ⃗ = dy dz zˆ with
x = 2.
In (a), 0 ⩽ y ⩽ 1,
2 1

∫ v ⃗ ⋅ da ⃗ = ∫ ∫ 2(2)y dy dz = 4.
0 0

In (b), 1 ⩽ y ⩽ 2 and 0 ⩽ z ⩽ 4 − 2y
2 4 − 2y
16
∫ v ⃗ ⋅ da ⃗ = ∫ ∫ 2(2)y dz dy =
3
.
1 0

1-13
 Electromagnetism

(ii) Here, da ⃗ = −dy dz xˆ and x = 0, so v ⃗ ⋅ da ⃗ = 2(0)y( −dy dx )xˆ = 0.

(iii) Here, da ⃗ = dx dy zˆ and z = 2
2 1
⎡ x 2(2) − 2y⎤ dy dx = 10 .
∫ v⃗ ⋅ da ⃗ = ∫ ∫ ⎣ ⎦
3
0 0

(iv) Here, da ⃗ = −dx dy zˆ and z = 0

2 2
⎡ x 2(0) − 2y⎤( −dx dy ) = 8.
∫ v ⃗ ⋅ da ⃗ = ∫ ∫ ⎣ ⎦
0 0

(v) Here da ⃗ = dx dz yˆ and y = 0, so v ⃗ ⋅ da ⃗ = 02z 3( −dx dz ) = 0.

(vi) Here, we have da ⃗ = dx dz′ nˆ where nˆ = nn⃗ . We can find n ⃗ by crossing vectors
A⃗ = −yˆ + 2zˆ and B ⃗ = 2xˆ (the edges of the volume):

xˆ yˆ zˆ
n ⃗ = A ⃗ × B ⃗ = 0 −1 2 = 4yˆ + 2zˆ .
2 0 0

So

n= 42 + 22 = 2 5

and

2 5 5
nˆ = yˆ + zˆ .
5 5

We can also obtain dz′ by considering

1-14
 Electromagnetism

5
so dz′ = 2
dz . Now

5 ⎛2 5 5 ⎞ ⎛ 1 ⎞
da ⃗ = dx dz ⎜ yˆ + zˆ⎟ = ⎜ yˆ + zˆ⎟dx dz
2 ⎝ 5 5 ⎠ ⎝ 2 ⎠

and
z
z = 4 − 2y → y = 2 − .
2
So
2 2
⎡ 2 3 1 2 ⎤
∫ v ⃗ ⋅ da ⃗ = ∫ ∫ ⎢⎣y z +
2
( )
x z − 2y ⎥ dx dz
⎦
0 0

2 2
⎧⎛ z ⎞2 1⎡ ⎛ z ⎞⎤⎫ 42
=∫∫ ⎨⎜ 2 − ⎟ z 3 + ⎢x 2z − 2⎜ 2 − ⎟⎥⎬ dx dz = .
⎩⎝ 2⎠ 2⎣ ⎝ 2 ⎠⎦⎭ 5
0 0

Therefore

∮S v ⃗ ⋅ da ⃗ = 4 + 163 +
10
3
+8+
42
5
=
436
15
as expected.

Problem 1.11. Test the curl theorem with v ⃗ = 5xy 2 xˆ + yz 2yˆ + 4x 2zzˆ and the surface
below.

1-15
 Electromagnetism

Solution The curl theorem states

∫S (∇ × v ⃗) ⋅ da ⃗ = ∮P v ⃗ ⋅ dl⃗.
Starting with the left-hand side, the curl is given by

xˆ yˆ zˆ
∂ ∂ ∂
∇ × v⃗ = = −2yzxˆ − 8xzyˆ − 10xyzˆ
∂x ∂y ∂z
5xy 2 yz 2 4x 2z

We also have da ⃗ = dy dz xˆ with 0 ⩽ z ⩽ − (y − 2)2 + 4. So

4 −(y − 2)2 + 4
1024
∫S ( ∇ × v ⃗ ) ⋅ da ⃗ = ∫ ∫ −2yz dz dy = −
15
.
0 0

Now to solve ∮ v ⃗ ⋅ dl⃗ over the two paths (i) and (ii):
P

(i) Here we have x = 0, z = 0, and dl⃗ = dy yˆ . So v ⃗ ⋅ dl⃗ = y(02)dy = 0.

(ii) Here we have dl⃗ = dy yˆ + dz zˆ , x = 0, and z = − (y − 2)2 + 4

0
1024
y⎡⎣ −(y − 2)2 + 4⎤⎦ dy = −
2
∫P v ⃗ ⋅ dl⃗ = ∫P yz 2dy + 4(0 2)z dz = ∫ 15
.
4

1-16
 Electromagnetism

So,

∮P v ⃗ ⋅ dl⃗ = 0 + −1024
15
=
−1024
15
as expected.

Problem 1.12. Test the gradient theorem with T = 3xz 2 − y 2 z and path z = y 2 and
z = y3 from (0, 0, 0) → (0, 1, 1).

Solution The gradient theorem states

b⃗

∫ ()
∇T ⋅ dl⃗ = T b ⃗ − T a ⃗ . ()
a⃗

Starting with the right side, we have

T (0,1,1) − T (0,0,0) = 3(0)(12 ) − (12 )(1) − 0 = − 1.

b⃗

Now to solve ∫ ∇T ⋅ dl⃗ , the gradient of T is given by

a⃗
∂ ∂ ∂
∇T =
∂x
( )
3xz 2 − y 2 z xˆ +
∂y
(
3xz 2 − y 2 z yˆ +
∂z
) ( )
3xz 2 − y 2 z zˆ

(
∇T = 3z 2xˆ − 2yz yˆ + 6xz − y 2 zˆ . )

1-17
 Electromagnetism

Here, dl⃗ = dy yˆ + dz zˆ with x = 0. So

∇T ⋅ dl⃗ = −2yz dy + ⎡⎣6(0)( z ) − y 2 ⎤⎦dz = −2yz dy − y 2 dz .

For path (i), we have z = y 2 → dz = 2y dy . So

( )
∇T ⋅ dl⃗ = −2y y 2 dy − y 2 (2y dy ) = −4y 3 dy

and
b⃗ 1

∫ ∇T ⋅ dl⃗ = ∫ −4y 3dy = −1

a⃗ 0

as expected. For path (ii), we have z = y3 → dz = 3y 2 dy . So

( ) ( )
∇T ⋅ dl⃗ = −2y y 3 dy − y 2 3y 2 dy = −5y 4dy

and
b⃗ 1

∫ ∇T ⋅ dl⃗ = ∫ − 5y 4dy = −1
a⃗ 0

also as expected.

Problem 1.13. Verify the following integration by parts given f = xy 2 z and

A⃗ = z 2xˆ + 4xyyˆ − x 2zzˆ and the surface below,
⎡A ⃗ × ∇f ⎤ ⋅ da ⃗ +
∫S f ( ∇ × A⃗ ) ⋅ da ⃗ = ∫S ⎣ ( )
⎦ ∮P fA⃗ ⋅ dl⃗.

1-18
 Electromagnetism

Solution Starting with the left-hand side

xˆ yˆ zˆ
∂ ∂ ∂
∇ × A⃗ = = [2z − ( −2xz )] yˆ + 4yzˆ = 2z(x + 1)yˆ + 4yzˆ .
∂x ∂y ∂z
z 2 4xy −x 2z

Now

( )
f ∇ × A ⃗ = 2xy 2 z 2(x + 1)yˆ + 4xy 3zzˆ .

2 2
Here we have da ⃗ = dx′ dz nˆ where nˆ = xˆ + yˆ and n = 2 so nˆ = 2
xˆ + 2
yˆ . Also
from

we have

dx′ = 2 dx with y = 1 − x .

Now

⎛ 2 2 ⎞
da ⃗ = 2 dx dz ⎜
⎝ 2
xˆ + yˆ ⎟ = dx dz xˆ + yˆ .
2 ⎠
( )
Therefore,
1 1
⎡ 2xy 2 z 2(x + 1)yˆ + 4xy 3zzˆ⎤ ⋅ (xˆ + yˆ )dx dz
∫S f ( ∇ × A⃗ ) ⋅ da ⃗ = ∫ ∫ ⎣ ⎦
0 0

1 1
7
= ∫∫ 2x(1 − x )2 z 2(x + 1)dx dz =
90
.
0 0

Next, we will solve the ∮ fA⃗ ⋅ dl⃗ term for the four segments.
P

1-19
 Electromagnetism

Segment (i)

z = 0 → f = xy 2 (0) = 0.

Segment (ii)

x = 0 → f = (0)y 2 z = 0.

Segment (iii)

dl⃗ = dx xˆ + dy yˆ , z = 1, and y = 1 − x → dy = − dx .

Segment (iv)

y = 0 → f = x(0 2)z = 0.

So

f A ⃗ ⋅ dl⃗ = xy 2 z 2dx + 4xy dy = x(1 − x )2 ⎡⎣ 1 − 4x(1 − x )⎤⎦dx

( ) ( )
and
1
1
∮P fA ⃗ ⋅ dl⃗ = ∫ x(1 − x )2 ⎡⎣ 1 − 4x(1 − x )⎤⎦dx = .
60
0

Now to solve the ∫S [A × ∇f ] ⋅ da ⃗ term. First, we have

∇f = y 2 zxˆ + 2xyzyˆ + xy 2 zˆ .

1-20
 Electromagnetism

So

xˆ yˆ zˆ
2
A × ( ∇f ) = z 4xy −x 2z
y 2 z 2xyz xy 2

( ) ( ) ( )
= 4x 2y 3 + 2x 3yz 2 xˆ + −x 2y 2 z 2 − xy 2 z 2 yˆ + 2xyz 3 − 4xy 3z zˆ .

As before, da ⃗ = dx dz (xˆ + yˆ ). So
1 1

∫S [A × (∇f )] ⋅ da ⃗ = ∫ ∫ ⎡⎣ 4x 2(1 − x )3 + 2x 3(1 − x )z 2

0 0
−x 2(1 − x )2 z 2 − x(1 − x )2 z 2⎤⎦dx dz

11
∫S [A × (∇f )] ⋅ da ⃗ = 180 .

So
11 1 7
∫S [A × (∇f )] ⋅ da ⃗ + ∮P fA⃗ ⋅ dl⃗ = 180 +
60
=
90

as expected.

Problem 1.14. Find the divergence and curl of the following functions:
a) v ⃗ = r 2rˆ + cos θ sin ϕ θˆ + sin θ cos ϕ ϕˆ
b) v ⃗ = s cos ϕ sˆ + cos ϕ sin ϕ ϕˆ +z sin ϕ zˆ
Solutions
a) v ⃗ = r 2rˆ + cos θ sin ϕ θˆ + sin θ cos ϕ ϕˆ

1 ∂ 2 1 ∂ 1 ∂vϕ
∇ ⋅ v⃗ = 2
r ∂r
( )
r vr +
r sin θ ∂θ
(
sin θ vθ + )
r sin θ ∂ϕ

1 ∂ 4 1 ∂ 1 ∂
= 2
r ∂r
( )
r +
r sin θ ∂θ
(
sin θ cos θ sin ϕ + )
r sin θ ∂ϕ
(sin θ cos ϕ)

1 sin ϕ 1
=
r
( )
2
4r 3 +
r sin θ
( r
)
−sin2 θ + cos2 θ + ( −sin ϕ)

1-21
 Electromagnetism

sin ϕ sin ϕ
= 4r +
r sin θ
(
1 − 2 sin2 θ −
r
)
sin ϕ
∇ ⋅ v ⃗ = 4r +
r
(
csc θ − 2 sin θ − 1 )
1 ⎡∂ ∂v ⎤
∇ × v⃗ = ⎢
r sin θ ⎣ ∂θ
( ∂ϕ ⎦
)
sin θ vϕ − θ ⎥rˆ

1 ⎡ 1 ∂vr ∂ ⎤ 1⎡ ∂ ∂v ⎤
+ ⎢ − ( rvϕ )⎥θˆ + ⎢ ( rvθ ) − r ⎥ϕˆ
r ⎣ sin θ ∂ϕ ∂r ⎦ ⎣
r ∂r ∂θ ⎦

1 ⎡∂ ∂ ⎤
= ⎢
r sin θ ⎣ ∂θ
(
sin2 θ cos ϕ −
∂ϕ
)
(cos θ sin ϕ)⎥rˆ
⎦

1⎡ 1 ∂ 2 ∂ ⎤
+ ⎢
r ⎣ sin θ ∂ϕ
( )
r − (r sin θ cos ϕ)⎥θˆ
∂r ⎦

1⎡ ∂ ∂ 2 ⎤ˆ
+ ⎢ (r cos θ sin ϕ) −
r ⎣ ∂r ∂ϕ
r ⎥ϕ
⎦
( )
1 sin θ cos ϕ ˆ
= (2 sin θ cos θ cos ϕ − cos θ cos ϕ)rˆ − θ
r sin θ r
cos θ sin ϕ ˆ
+ ϕ
r

cos θ cos ϕ sin θ cos ϕ ˆ cos θ sin ϕ ˆ

∇ × v⃗ = (2 − csc θ )rˆ − θ+ ϕ
r r r
b) v ⃗ = s cos ϕ sˆ + cos ϕ sin ϕ ϕˆ + z sin ϕ zˆ

1 ∂ 1 ∂vϕ ∂v
∇ ⋅ v⃗ = ( svs ) + + z
s ∂s s ∂ϕ ∂z

1 ∂ 2 1 ∂ ∂
=
s ∂s
(s cos ϕ +
s ∂ϕ
)
(cos ϕ sin ϕ) +
∂z
(z sin ϕ)

1
= 2 cos ϕ +
s
(
−sin2 ϕ + cos2 ϕ + sin ϕ )
cos2 ϕ − sin2 ϕ
∇ ⋅ v ⃗ = 2 cos ϕ + sin ϕ +
s

1-22
 Electromagnetism

⎛ 1 ∂vz ∂vϕ ⎞ ⎛ ∂vs ∂v ⎞ 1⎡ ∂ ∂v ⎤

∇ × v⃗ = ⎜ − ⎟sˆ + ⎜ − z ⎟ϕˆ + ⎢ ( svϕ ) − s ⎥zˆ
⎝ s ∂ϕ ∂z ⎠ ⎝ ∂z ∂s ⎠ s ⎣ ∂s ∂ϕ ⎦

⎡1 ∂ ∂ ⎤ ⎡∂ ∂ ⎤
=⎢ (z sin ϕ) − (cos ϕ sin ϕ)⎥sˆ + ⎢ (s cos ϕ) − (z sin ϕ)⎥ϕˆ
⎣ s ∂ϕ ∂z ⎦ ⎣ ∂z ∂s ⎦

1⎡ ∂ ∂ ⎤
+ ⎢ (s cos ϕ sin ϕ) − (s cos ϕ)⎥zˆ
s ⎣ ∂s ∂ϕ ⎦

z 1
= cos ϕ sˆ + (cos ϕ sin ϕ + s sin ϕ)zˆ
s s

z sin ϕ
∇ × v⃗ = cos ϕ sˆ + (cos ϕ + s )zˆ
s s

Problem 1.15. Find the gradient and Laplacian of:

a) T = r 2(cos θ sin ϕ + sin θ cos ϕ )
b) T = z 2 sin ϕ − s cos2 ϕ
Solutions
a) T = r 2(cos θ sin ϕ + sin θ cos ϕ )
∂T 1 ∂T ˆ 1 ∂T ˆ
∇T = rˆ + θ+ ϕ
∂r r ∂θ r sin θ ∂ϕ

1 2
= 2r(cos θ sin ϕ + sin θ cos ϕ)rˆ + r ( −sin θ sin ϕ + cos θ cos ϕ)θˆ
r
1
+ r 2(cos θ cos ϕ − sin θ sin ϕ)ϕˆ
r sin θ

= 2r(cos θ sin ϕ + sin θ cos ϕ)rˆ + r(cos θ cos ϕ − sin θ sin ϕ)θˆ
r
+ (cos θ cos ϕ − sin θ sin ϕ)ϕˆ
sin θ

r
∇T = 2r sin(θ + ϕ)rˆ + r cos(θ + ϕ)θˆ + cos(θ + ϕ)ϕˆ .
sin θ

Note we could have written T as T = r 2 sin(θ + ϕ ) and then computed the

gradient.

1 ∂ ⎛ 2 ∂T ⎞ 1 ∂ ⎛ ∂T ⎞ 1 ⎛ ∂ 2T ⎞
∇2 T = ⎜r ⎟ + ⎜ sin θ ⎟ + ⎜ ⎟
r 2 ∂r ⎝ ∂r ⎠ r 2 sin θ ∂θ ⎝ ∂θ ⎠ r 2 sin2 θ ⎝ ∂ϕ 2 ⎠

1-23
 Electromagnetism

1 ∂⎡ 3 ⎤ 1 ∂ ⎡ 2 ⎤
= 2 ⎣ 2r sin(θ + ϕ)⎦ + 2 ⎣ r sin θ cos(θ + ϕ)⎦
r ∂r r sin θ ∂θ
1 ∂ ⎡ 2 ⎤
+ 2 2 ⎣ r cos(θ + ϕ)⎦
r sin θ ∂ϕ

1
= 6 sin(θ + ϕ) + [cos θ cos(θ + ϕ) − sin θ sin(θ + ϕ)]
sin θ
1
+ (
sin2 θ
−sin(θ + ϕ) )
cos θ sin(θ + ϕ)
∇2 T = 5 sin(θ + ϕ) + cos(θ + ϕ) − .
sin θ sin2 θ

b) T = z 2 sin ϕ − s cos2 ϕ
∂T 1 ∂T ˆ ∂T
∇T = sˆ + ϕ+ zˆ
∂s s ∂ϕ ∂z

∂ 2 1 ∂ 2
=
∂s
(
z sin ϕ − s cos2 ϕ sˆ +) s ∂ϕ
(
z sin ϕ − s cos2 ϕ ϕˆ )
∂ 2
+
∂z
(
z sin ϕ − s cos2 ϕ zˆ )
1
= −cos2 ϕ sˆ + ⎡⎣ z 2 cos ϕ − 2s cos ϕ( −sin ϕ)⎤⎦ϕˆ + 2z sin ϕ zˆ
s

cos ϕ 2
∇T = − cos2 ϕ sˆ +
s
( )
z + 2s sin ϕ ϕˆ + 2z sin ϕ zˆ

1 ∂ ⎛ ∂T ⎞ 1 ∂ 2T ∂ 2T
∇2 T = ⎜s ⎟ + +
s ∂s ⎝ ∂s ⎠ s 2 ∂ϕ 2 ∂z 2

∂T ∂T ∂ ⎛ ∂T ⎞
= −cos2 ϕ → s = −s cos2 ϕ → ⎜ s ⎟ = −cos2 ϕ
∂s ∂s ∂s ⎝ ∂s ⎠

∂T ∂ 2T
∂ϕ
= z 2 cos ϕ + 2s cos ϕ sin ϕ →
∂ϕ 2
(
= −z 2 sin ϕ + 2s −sin2 ϕ + cos2 ϕ )
∂T ∂ 2T
= 2z sin ϕ → = 2 sin ϕ
∂z ∂z

cos2 ϕ 2 z2
∇2 T = −
s
+
s
( s
)
−sin2 ϕ + cos2 ϕ − 2 sin ϕ + 2 sin ϕ

1-24
 Electromagnetism

cos2 ϕ 2 ⎛ z2 ⎞
∇2 T = − sin2 ϕ + ⎜ 2 − 2 ⎟sin ϕ
s s ⎝ s ⎠

Problem 1.16. Test the divergence theorem with v ⃗ = r cos ϕ rˆ + r cos θ sin θ θˆ +
r sin ϕ ϕ̂ and the volume below (the upper half of the sphere of radius R with a cone
of radius a = R cut out).
3

Solution The divergence theorem states

∫V ∇ ⋅ v ⃗ dτ = ∮S v ⃗ ⋅ da ⃗.
Starting with the left-hand side, the divergence is
1 ∂ 2 1 ∂ 1 ∂vϕ
∇ ⋅ v⃗ = ( )
r 2 ∂r
r vr +
r sin θ ∂θ
sin θ(vθ + )
r sin θ ∂ϕ

1 ∂ 3 1 ∂ 1 ∂
= 2
r ∂r
(r cos ϕ + )
r sin θ ∂θ
(
r sin2 θ cos θ +
r sin θ ∂ϕ
)(r sin ϕ)

1 cos ϕ
= 3 cos ϕ +
sin θ
(
2 sin θ cos2 θ − sin3 θ +
sin θ
)
cos ϕ
∇ ⋅ v ⃗ = 3 cos ϕ + 2 cos2 θ − sin2 θ + .
sin θ
For the volume,
⎛a⎞ ⎛ 1 ⎞ π π π
0 ⩽ r ⩽ R, tan−1⎜ ⎟ = tan−1⎜ ⎟ = → ⩽θ⩽ , 0 ⩽ ϕ ⩽ 2π .
⎝ R⎠ ⎝ 3⎠ 6 6 2

1-25
 Electromagnetism

So
π
R 2 2π
⎛ cos ϕ ⎞ 2
∫V ∇ ⋅ v ⃗ dτ = ∫ ∫ ∫ ⎜ 3 cos ϕ + 2 cos2 θ − sin2 θ +
⎝ (
⎟ r sin θ dϕ dθ dr
sin θ ⎠
)
0 π 0
6

3 3
∫V ∇ ⋅ v ⃗ dτ = −
12
πR .

Now for the right-hand side, we have three surfaces: the bottom (i), the outer shell
(ii), and the inner part where the cone is cut out (iii). We have

v ⃗ = r cos ϕ rˆ + r sin ϕ ϕˆ + r cos θ sin θ θˆ.

π
For (i), we have da ⃗ = r dr dϕ θˆ and θ = 2 . So
π π
v ⃗ ⋅ da ⃗ = r 2 cos sin dr dϕ = 0
2 2

and

∫ v ⃗ ⋅ da ⃗ = 0.
(i )

For (ii), we have r = R and da ⃗ = r 2 sin θ dθ dϕ rˆ = R2 sin θ dθ dϕ rˆ . So

v ⃗ ⋅ da ⃗ = R3 cos ϕ sin θ dθ dϕ

and
π
2π 2

∫ v ⃗ ⋅ da ⃗ = R3 ∫∫ cos ϕ sin θ dθ dϕ = 0.
(ii ) 0 π
6

For (iii), we have θ = π

6
and da ⃗ = −r sin θ dr dϕ θˆ = − 12 r dr dϕ θˆ . So

1 π π 3 2
v ⃗ ⋅ da ⃗ = − r 2 cos sin = − r
2 6 6 8

and
R 2π
3 3
∫ v ⃗ ⋅ da ⃗ = −
8
∫∫ r 2 dϕ dr = −
12
πR3.
(iii ) 0 0

1-26
 Electromagnetism

Therefore,

∮S v ⃗ ⋅ da ⃗ = − 123 πR3
as expected.

Problem 1.17. Test the curl theorem with v ⃗ = s 2z sˆ + sin ϕ cos ϕ ϕˆ + zs cos ϕ zˆ
and half of a cylindrical shell with radius R and height h.

Solution The curl theorem states

∫S ( ∇ × v ⃗) ⋅ da ⃗ = ∮P v ⃗ ⋅ dl⃗.
Starting with the left-handed side, we have

da ⃗ = s dϕ dz sˆ = R dϕ dz sˆ .

Since we are dotting da ⃗ with ∇ × v ⃗ , we only need the ŝ component of the curl:

⎛ 1 ∂vz ∂vϕ ⎞ ⎡1 ∂ ∂ ⎤
[∇ × v ⃗ ] s = ⎜ − ⎟sˆ = ⎢ (zs cos ϕ) − (sin ϕ cos ϕ)⎥sˆ
⎝ s ∂ϕ ∂z ⎠ ⎣ s ∂ϕ ∂z ⎦
= −z sin ϕ sˆ .

So

( ∇ × v ⃗) ⋅ da ⃗ = −Rz sin ϕ dϕ dz.

1-27
 Electromagnetism

We have
0⩽ϕ⩽π and 0⩽z⩽h
so
h π

∫S ( ∇ × v ⃗) ⋅ da ⃗ = ∫ ∫ −Rz sin ϕ dϕ dz = −h 2R .
0 0

For the left-hand side, we have four curves

with
v ⃗ = s 2z sˆ + sin ϕ cos ϕ ϕˆ + z cos ϕ zˆ .

For curve (i), dl⃗ = dϕ ϕˆ , z = 0, and s = R . So

v ⃗ ⋅ dl⃗ = sin ϕ cos ϕ dϕ

and
π

∫ sin ϕ cos ϕ dϕ = 0.
0

For curve (ii), dl⃗ = dz zˆ , ϕ = π , and s = R . So

v ⃗ ⋅ dl⃗ = zs cos ϕ dz = zR cos π dz = −zR dz

and
h
1
∫ −zR dz = − h 2R .
2
0

1-28
 Electromagnetism

For curve (iii), dl⃗ = dϕ ϕˆ , z = h, and s = R . So

v ⃗ ⋅ dl⃗ = sin ϕ cos ϕ dϕ

and
0

∫ sin ϕ cos ϕ dϕ = 0.
π

For curve (iv), dl⃗ = dz zˆ , ϕ = 0, and s = R . So

v ⃗ ⋅ dl⃗ = zs cos ϕ dz = zR cos(0)dz = zR dz

and
0
1
∫ zR dz = − h 2R2 .
2
h

So,

∮ v ⃗ ⋅ dl⃗ = − 12 h2R − 12 h2R = − h2R

P

as expected.

Problem 1.18. Test the gradient theorem using T = sz 2 sin ϕ and the half helix path
(radius R , height h).

1-29
 Electromagnetism

Solution The gradient theorem states

∫P ∇T ⋅ dl⃗ = T ( b ⃗) − T ( a ⃗).
Starting with the right-hand side
⎛ π ⎞ ⎛ π ⎞ π ⎛ π⎞
() ()
T b ⃗ − T a ⃗ = T ⎜ R , , h⎟ − T ⎜ R , − , 0⎟ = Rh 2 sin − R(0)2 sin⎜ − ⎟ = h 2R .
⎝ 2 ⎠ ⎝ 2 ⎠ 2 ⎝ 2⎠
Now, the gradient is
∂T 1 ∂T ˆ ∂T
∇T = sˆ + ϕ+ zˆ = z 2 sin ϕ sˆ + z 2 cos ϕ ϕˆ + 2sz sin ϕ zˆ .
∂s s ∂ϕ ∂z
We also have s = R and l⃗ = s dϕ ϕˆ + dz zˆ = R dϕ ϕˆ + dz zˆ. So
∇T ⋅ dl⃗ = Rz 2 cos ϕ dϕ + 2Rz sin ϕ dz .
We need a way to relate z and ϕ. Note that as ϕ increases, z increases linearly. So,
using the equation of line
z − z0 = γ (ϕ − ϕ0) ,
π
when z = 0 and ϕ = − 2,
⎛ π⎞
z = γ⎜ ϕ + ⎟,
⎝ 2⎠
π
when z = h and ϕ = 2
,
⎛π π⎞ h
h = γ⎜ + ⎟ → γ = ,
⎝2 2 ⎠ π
so
h h
z= ϕ−
π 2
and
h
dz = dϕ .
π
Using our expressions for z and dz , we have
⎡ ⎛h h ⎞2 ⎛h h⎞ ⎛ h ⎞⎤
∇T ⋅ dl⃗ = ⎢R⎜ ϕ + ⎟ cos ϕ + 2R⎜ ϕ + ⎟ sin ϕ ⎜ ⎟⎥dϕ .
⎣ ⎝π 2⎠ ⎝π 2⎠ ⎝ π ⎠⎦

So
π
b⃗ 2 ⎡ ⎛h h ⎞2 ⎛h h⎞ ⎛ h ⎞⎤
∫ ∇ T ⋅ d l⃗ = ∫ ⎢R⎜ ϕ + ⎟ cos ϕ + 2R⎜ ϕ + ⎟ sin ϕ ⎜ ⎟⎥dϕ = h 2R
⎣ ⎝π 2⎠ ⎝π 2⎠ ⎝ π ⎠⎦
a⃗ −π
2

as expected.

1-30
 Electromagnetism

Problem 1.19. Evaluate the following integrals:

3

a) ∫ (2x 2 − x + 4)δ (x − 2)dx

1
1

b) ∫ (x 2 + 4)δ (x − 2)dx
−1
6
3x
c) ∫ sin( 2 )δ (x − π )dx
2
2

d) ∫ (2x 3 + 1)δ (4x )dx

−2
∞

e) ∫ x 2δ (2x + 1)dx
−∞
a

f) ∫ δ (x − b )dx
0

Solutions
a)
3

∫ ( 2x 2 − x + 4)δ(x − 2)dx.
1

Since 2 ∈ (1, 3) and f (x ) = 2x 2 − x + 4, we have

3

∫ ( 2x 2 − x + 4)δ(x − 2)dx = f (2) = 2(2)2 − 2 + 4 = 10.

1

b) 1

∫ ( x 2 + 4)δ(x − 2)dx.
−1

Since 2 ∉ ( −1, 1), we have

1

∫ ( x 2 + 4)δ(x − 2)dx = 0.
−1

c)
6
⎛ 3x ⎞
∫ sin ⎜ ⎟δ(x − π )dx .
⎝ 2 ⎠
2

1-31
 Electromagnetism

3x
Since π ∈ (2, 6) and f (x ) = sin( 2 ), we have
6
⎛ 3x ⎞ ⎛ 3π ⎞
∫ sin ⎜ ⎟δ(x − π )dx = f (π ) = sin ⎜ ⎟ = −1.
⎝ 2 ⎠ ⎝2⎠
2

d)
2

∫ ( 2x3 + 1)δ(4x)dx.
−2

Since 0 ∈ ( −2, 2) and f (x ) = 2x 3 + 1, we have

2
1 1 1
∫ ( 2x3 + 1)δ(4x)dx = 4
f (0) =
4
( 2(0)3 + 1) = .
4
−2

e) ∞

∫ x 2δ(2x + 1)dx .
−∞

This can be rewritten as

∞ ∞ ∞
⎡ ⎛ 1 ⎞⎤ 1 ⎛ 1⎞
∫ 2
x δ(2x + 1)dx = ∫ x δ⎢2⎜x + ⎟⎥dx =
2
∫ x2 δ⎜x + ⎟dx .
⎣ ⎝ 2 ⎠⎦ 2 ⎝ 2⎠
−∞ −∞ −∞
1
Since −2 ∈ ( −∞ , ∞) and f (x ) = x 2 , we have
∞
1 ⎛ 1⎞ 1 ⎛ 1⎞ 1 ⎛ 1 ⎞2 1
∫ x2 δ⎜x + ⎟dx = f ⎜− ⎟ = ⎜− ⎟ = .
2 ⎝ 2⎠ 2 ⎝ 2⎠ 2 ⎝ 2⎠ 8
−∞

f) a

∫ δ(x − b)dx .
0

Here we have
a

∫
0
δ(x − b)dx = { 1 if 0 < b < a .
0 otherwise

Problem 1.20. Suppose we have two vector fields F1⃗ = y 2 zˆ and F2⃗ = xxˆ + yyˆ + zzˆ .
Calculate the divergence and curl of each. Which can be written as the gradient of a
scalar and which can be written as the curl of a vector? Find a scalar and a vector
potential.

1-32
 Electromagnetism

Solution For F1⃗ , we have

⎛∂ ∂ ∂ ⎞ ∂( y 2 )
∇ ⋅ F1⃗ = ⎜ xˆ + yˆ + zˆ⎟ ⋅ ( y 2 zˆ ) = =0
⎝ ∂x ∂y ∂z ⎠ ∂z

and

xˆ yˆ zˆ
∂ ∂ ∂
∇ × F1⃗ = = 2yxˆ .
∂x ∂y ∂z
0 0 y2

For F2⃗ , we have

⎛∂ ∂ ∂ ⎞
∇ ⋅ F2⃗ = ⎜ xˆ + yˆ + zˆ⎟ ⋅ (xxˆ + yyˆ + zzˆ ) = 1 + 1 + 1 = 3
⎝ ∂x ∂y ∂z ⎠

and

xˆ yˆ zˆ
∂ ∂ ∂
∇ × F2⃗ = = (0 − 0)xˆ + (0 − 0)yˆ + (0 − 0)zˆ = 0.
∂x ∂y ∂z
x y z

Since ∇ ⋅ F1⃗ = 0, F1⃗ can be expressed as F1⃗ = ∇ × A⃗ . We can find A⃗ by considering

xˆ yˆ zˆ
∂ ∂ ∂
∇ × A⃗ =
∂x ∂y ∂z
0 0 y2
⎛ ∂A ∂Ay ⎞ ⎛ ∂Ax ∂Az ⎞ ⎛ ∂Ay ∂Ax ⎞
=⎜ z − ⎟xˆ + ⎜ − ⎟yˆ + ⎜ − ⎟zˆ .
⎝ ∂y ∂z ⎠ ⎝ ∂z ∂x ⎠ ⎝ ∂x ∂y ⎠

By inspection:
∂Az ∂Ay ∂Ax ∂Az ∂Ay ∂Ax
− = 0, − = 0, − = y2 .
∂y ∂z ∂z ∂x ∂x ∂y

This is satisfied by

A ⃗ = y 2 xy,
ˆ

which is just one example. Since ∇ × F2⃗ = 0, F2⃗ can be expressed as F2⃗ = −∇V . We
can find V by considering

1-33
 Electromagnetism

⎛ ∂V ∂V ∂V ⎞
F2⃗ = −⎜ xˆ + yˆ + zˆ⎟ .
⎝ ∂x ∂y ∂z ⎠

By inspection:
∂V ∂V ∂V
x=− , y=− , z=− .
∂x ∂y ∂z
This is satisfied by
⎛ x2 y2 z2 ⎞
V = −⎜ + + ⎟
⎝2 2 2⎠
which is again just one example.

Bibliography
Byron F W and Fuller R W 1992 Mathematics of Classical and Quantum Physics (New York:
Dover)
Griffiths D J 1999 Introduction to Electrodynamics 3rd edn (Englewood Cliffs, NJ: Prentice Hall)
Griffiths D J 2013 Introduction to Electrodynamics 4th edn (New York: Pearson)
Halliday D, Resnick R and Walker J 2010 Fundamentals of Physics 9th edn (New York: Wiley)
Halliday D, Resnick R and Walker J 2013 Fundamentals of Physics 10th edn (New York: Wiley)
Purcell E M and Morin D J 2013 Electricity and Magnetism 3rd edn (Cambridge: Cambridge
University Press)
Rogawski J 2011 Calculus: Early Transcendentals 2nd edn (San Francisco, CA: Freeman)

1-34
 IOP Concise Physics

Electromagnetism
Problems and solutions
Carolina C Ilie and Zachariah S Schrecengost

Chapter 2
Electrostatics

Electrostatics is the topic of this chapter. Coulomb’s law, Gauss’s law, and the
energy of various charge distributions are a few ways of understanding the electric
field. The methods employed will make use of the specific degrees of symmetry. The
mathematical skills obtained in chapter 1 will be applied here to analyze different
charge distributions in Cartesian, spherical, or cylindrical coordinates.

2.1 Theory
2.1.1 Coulomb’s law
The force on a point charge q due to a charge Q , separated by a distance r , is given
by
JG 1 Q
F = q rˆ,
4πεo r 2
C2
where ε0(= 8.85 × 10−12 ) is the permittivity of free space.
Nm2

2.1.2 Electric field

In general, for a volume charge density ρ(r ⃗ ), the electric field at r ⃗ is given by
JG 1 ρ(r ⃗′)
E (r ⃗ ) = ∫ rˆ dτ′ .
4πεo V r 2
For a surface charge density σ (r ⃗ ), the electric field is given by
JG 1
E (r ⃗ ) =
4πεo
∫S σr(r2⃗′) rˆ da′.

doi:10.1088/978-1-6817-4429-2ch2 2-1 ª Morgan & Claypool Publishers 2016

 Electromagnetism

For a linear charge density λ(r ⃗ ), the electric field is given by

JG 1 λ(r ⃗′)
E (r ⃗ ) = ∫ rˆ dl′ .
4πεo P r 2

2.1.3 Gauss’s law

JG
For an electric field E and surface S , Gauss’s law states
JG JG q
S
∮
E ⋅ da = enc ,
ε0
where the enclosed charge is
qenc = ∫V ρ dτ.
In differential form
JG ρ
∇⋅ E = ,
εo
where ρ is the volume charge density.
JG
2.1.4 Curl of E
JG JG
∮P E ⋅ dl⃗ = 0 →  ∇ × E = 0,
JG
where E is an electrostatic field.

2.1.5 Energy of a point charge distribution

The energy required to assemble n charges q1, q2, …, qn is given by
⎛ ⎞
1
n ⎜ n 1 qj ⎟ 1 n
W = ∑ qi ⎜ ∑ ⎟ = ∑q V (ri ⃗ ),
2 i = i ⎜⎜ j = i 4πεo rij ⎟⎟ 2 i = 1 i
⎝ j ≠i ⎠
where V (ri ⃗ ) is the potential at charge qi and r ij is the distance between charges qi and qj .

2.1.6 Energy of a continuous distribution

1
W=
2
∫ ρV (r ⃗) dτ =   ε2o ∫ E 2 dτ .
all
space

2.1.7 Energy per unit volume

W ε
= o E2
volume 2

2-2
 Electromagnetism

2.2 Problems and solutions

Problem 2.1. Given the charge distribution below, find the force on charge q1 = q
with q2 = 3q , q3 = −2q , and q4 = q .

Solution The force on q1 from q2 is given by

JG 1 q1q2 1 q1q2 JG
F21 = rˆ = r ,
4πεo r 2
4πεo r 3
where
JG
r = 3xˆ + 4yˆ
and
r = 32 + 42 = 5.
So
JG 1 3q 2 q2 ⎛ 9 12 ⎞
F21 = (3xˆ + 4yˆ ) = ⎜ xˆ + yˆ ⎟ .
4πεo 53
4πεo ⎝ 125 125 ⎠
The force on q1 from q3 is given by
JG 1 q1q3
F31 = rˆ,
4πεo r 2
where
r=4
and
rˆ = ŷ.

2-3
 Electromagnetism

So
JG 1 ⎛ 2q 2 ⎞ q2 1
F31 = ⎜ − 2 ⎟yˆ = − yˆ .
4πεo ⎝ 4 ⎠ 4πεo 8
The force on q1 from q4 is given by
JG 1 q1q4
F41 = rˆ,
4πεo r 2
where
r=2
and
rˆ = x̂.
So
JG 1 q2 q2 1
F41 = 2
rˆ = xˆ .
4πεo 2 4πεo 4
Therefore,
JG JG JG JG q 2 ⎡⎛ 9 1⎞ ⎛ 12 1⎞ ⎤
F1 = F21 + F31 + F41 = ⎢⎜ + ⎟xˆ + ⎜ − ⎟yˆ ⎥
4πεo ⎣⎝ 125 4⎠ ⎝ 125 8⎠ ⎦
JG q 2 ⎛ 161 29 ⎞
F1 = ⎜ xˆ − yˆ ⎟ .
4πεo ⎝ 500 1000 ⎠

Problem 2.2. Given a charged sheet with surface charge density σ = ky (where k is a
constant) and sides of length 2d , find the electric field z above the center of the sheet.

2-4
 Electromagnetism

The electric field is given by

JG 1
E =
4πεo
∫ rσ2 rˆ da.
The horizontal components cancel so we only have the ẑ -component:
rˆ → cos θ zˆ = r zˆ .
z

Also we have da = dx dy and r 2 = x 2 + y 2 + z 2 . Note that the piece of the sheet in

each quadrant of the xy -plane contributes the same amount to the total field.
Therefore,
d d d d
JG JG 4 kyzzˆ kzzˆ y
E = 4Equad = ∫∫ 3/2
dx dy = ∫∫ 3/2 
dx dy
4πεo
0 0 ( x2 + y2 + z2 ) πεo
0 0
(x + y + z 2 )
2 2

d ⎡ ⎤ x= d
d

y⎢ ⎥ dy = kzdzˆ
kzzˆ x y
= ∫ ⎢ y2 + z2 y 2 + z 2 + x 2 ⎥⎦
∫ dy .
πεo
0 ⎣( ) x=0
πεo
0 (
y 2
+ z2 ) y2 + z2 + d 2

Let
u2 = y2 + z2
so
2u du = 2y dy → u du = y dy .
Evaluating u at the endpoints yields
u 2(y = 0) = z 2 → u = z
u 2 (y = d ) = d 2 + z 2 → u = d 2 + z2 .
Now
z 2+ d 2
JG kzdzˆ du
E =
πεo
∫ u u2 + d 2
z

⎛ 2 2
⎞
⎜ d+ d +z ⎟
kzdzˆ ⎡⎢ 1 ⎛ d + d 2 + u 2 ⎞⎤⎥
u=z
kzzˆ ⎜ z ⎟
= ln⎜ ⎟ = ln
πεo ⎢⎣ d ⎝ u ⎠⎥⎦ πεo ⎜⎜ d + 2d 2 + z 2 ⎟
⎟
z 2+d 2
u=
⎜ ⎟
⎝ z2 + d 2 ⎠
⎡ ⎤
2
JG kz ⎢ d + z + d
E = ln
( 2
z2 + d 2 ⎥ )
πεo ⎢⎢ ⎥zˆ .
⎣ z d + 2 d 2
+ z 2
( ⎥⎦ )

2-5
 Electromagnetism

Problem 2.3. Find the electric field d above a cylinder of radius R , height h, and
volume density ρ (ignoring edge effects).

Solution We have
JG 1
E =
4πεo
∫ rρ2 rˆ dτ.
Note our horizontal components cancel, so rˆ → cos θ ẑ with
d+h−z
cos θ = .
r
Also
dτ = s ds dϕ dz
and
r 2 = s 2 + (d + h − z )2 .
Therefore,
2π R h
JG ρ (d + h − z )s zˆ
E =
4πεo
∫ ∫ ∫ ⎡ s 2 + (d + h − z )2 ⎤3/2
dz ds dϕ
0 0 0 ⎣ ⎦
2πρ ⎡ 2
= R + d 2 − R2 + (d + h)2 + h⎤⎦zˆ
4πεo ⎣
JG ρ ⎡ 2
E = R + d 2 − R2 + (d + h)2 + h⎤⎦zˆ .
2εo ⎣

2-6
 Electromagnetism

Note if R ≫ d and R ≫ h, the field reduces to

JG ρh
E = zˆ ,
2εo
which is the field given by an infinite sheet of surface charge σ = ρh.

Problem 2.4. Given the bottom hemisphere of a spherical shell of radius R ,

thickness d , and volume charge density ρ, find the electric field z above the center
(above the open part, ignoring edge effects).

Solution The electric field is given by

JG 1
E =
4πεo
∫ rρ2 rˆ dτ.
We can see from

that
π
R ⩽ r ⩽ R + d, 0 ⩽ ϕ ⩽ 2π , ⩽ θ ⩽ π.
2

2-7
 Electromagnetism

Also, dτ = r 2 sin θ dr dϕ dθ . From the law of cosines,

r 2 = z 2 + r 2 − 2rz cos θ .
Since the horizontal components cancel, rˆ becomes
z+b
rˆ → cos γ zˆ = zˆ ,
r
where b is given by
b
cos(π − θ ) = → b = −r cos θ .
r
So
z − r cos θ
rˆ → zˆ .
r
Therefore,
R+d π 2π
JG ρ r 2(z − r cos θ ) sin θ zˆ
E = ∫ ∫ ∫ 3/2
dϕ dθ dr
4πεo
R π
2
0 (z 2
+ r 2 − 2rz cos θ )
R+d π
ρzˆ r 2(z − r cos θ )sin θ zˆ
= ∫ ∫ 3/2
dθ dr .
2εo
R π
2
(z 2
+ r 2 − 2rz cos θ )
Let u = cos θ and du = −sin θ dθ
⎛ π⎞
u⎜θ = ⎟ = 0
⎝ 2⎠
u (θ = π ) = − 1
R +d 0
JG ρzˆ r 2(z − ru )zˆ
E = ∫ ∫ du dr
2εo
R −1 ( z 2 + r 2 − 2rzu )3/2
ρzˆ
R +d
⎛ r3 ⎞
= ∫ ⎜r 2 − ⎟ dr
2εo z 2
R
⎝ r2 + z2 ⎠

⎡ r 2 + z 2 ( r 2 − 2z 2 ) ⎤
r=R +d
ρzˆ ⎢ r 3 ⎥
= −
2εoz 2 ⎢⎣ 3 3 ⎥⎦
r=R

JG ρzˆ
E =
6εoz 2 { (R + d ) − 3
(R + d )2 + z 2 ⎡⎣ (R + d )2 − 2z 2⎤⎦

− R3 + (R 2
+ z2 )( R 2
− 2z 2 )}.

2-8
 Electromagnetism

JG
Problem 2.5. Given the electric field E = k [2xzxˆ + z 2yˆ + (x 2 + 2yz )zˆ ] (with con-
stant k ) find the following:
a) The charge density ρ.
b) The charge enclosed by a cylinder of height h, radius R , and base on the xy -plane
center at the origin (below).
c) The charge enclosed by an upper hemisphere of radius R centered at the origin.

Solutions
a) The charge density ρ.
Gauss’s law states
JG ρ
∇⋅E = .
εo

So
JG
ρ = εo  ∇ ⋅ E = kεo(2y + 2z ) = 2kεo(y + z ).

b) The charge enclosed by a cylinder of height h, radius R , and base on the xy -plane
center at the origin.
We have

qenc = ∫V ρ dτ,

2-9
 Electromagnetism

with
ρ = 2kεo(y + z ).
We can transform ρ into cylindrical coordinates using x = s cos ϕ, y = s sin ϕ, and
z = z.
So
R h 2π

qenc = 2kεo ∫∫∫ (s sin ϕ + z )s dϕ dz ds = πkεoh 2R2 .

0 0 0

c) The charge enclosed by an upper hemisphere of radius R centered at the origin.

Again
qenc = ∫V ρ dτ
with
ρ = 2kεo(y + z ),
but now y = r sin ϕ sin θ and z = r cos θ . So
π
R 2π 2
kεoπR 4
qenc = 2kεo ∫∫∫ r(sin ϕ sin θ + cos θ )r 2 sin θ dθ dϕ dr =
2
.
0 0 0

Problem 2.6. Given a charge q located in the center of a spherical shell of radius R
and surface charge σ = k sin θ (with constant k ), find the electric field inside and
outside the shell.

Solution We will use Gauss’s law to find the field

JG JG q
S
∮
E ⋅ da = enc ,
εo

2-10
 Electromagnetism

where
JG JG
∮S E ⋅ da = ∮S E da = E ∮S da = E 4πr 2.
For r < R , we have
qenc = q.

So

q JG 1 q
E 4πr 2 = →E = rˆ .
εo 4πεo r 2

For r > R , we have

2π π

qenc = q + ∫ σ da = q + k ∫ ∫ R2 sin2 θ dθ dϕ.

0 0
2 2
qenc = q + kR π .

So

q kR2π 2
E 4πr 2 = +
εo εo

and
JG 1 q + kR2π 2
E = rˆ .
4πεo r2

Problem 2.7. Given a line of charge carrying λ surrounded by a cylindrical shell with
inner radius a , outer radius b, and charge density ρ = ks 2 , find the electric field in the
regions s < a , a < s < b, and b < s .

Solution Gauss’s law states

JG JG q
∮S E ⋅ da = enc ,
εo

2-11
 Electromagnetism

where
JG JG
∮S E ⋅ da = ∮S E da = E ∮S da = E 2πsl .

For s < a , we have

qenc = λl .

So
λl
E 2πsl =
εo

and
JG λ
E = sˆ .
2πεos

For a < s < b, we have

l 2π s
⎡ kπ 4 ⎤
qenc = λl + ∫ ρ dτ = λ l + ∫ ∫ ∫ k (s′)2 s′ ds′ dϕ′ dz′ = l⎢λ +
⎣ 2
( )
s − a 4 ⎥.
⎦
0 0 a

So
⎡ kπ 4 ⎤
JG JG q
l⎢λ +
⎣ 2
(
s − a4 ⎥
⎦ )
∮ S
E ⋅ da = enc
εo
→ E 2πsl =
εo

and

JG 2λ + kπ s 4 − a 4
E =
(
sˆ .
)
4πεos

For b < s , we have

l 2π b
⎡ kπ 4 ⎤
qenc = λl + ∫ ρ dτ = λ l + ∫ ∫ ∫ k (s′)2 s′ ds′ dϕ′ dz′ = l⎢λ +
⎣ 2
( )
b − a 4 ⎥.
⎦
0 0 a

So
⎡ kπ 4 ⎤
JG JG q
l⎢λ +
⎣ 2
(
b − a4 ⎥
⎦ )
∮S E ⋅ da = enc
εo
→ E 2πsl =
εo

2-12
 Electromagnetism

and

(
JG 2λ + kπ b 4 − a 4
E =
)
sˆ .
4πεos

Problem
JG 2.8. Which of the following is a possible electrostatic field?
a) E = k (yzxˆ + xzyˆ + x 2zˆ )
JG
b) E = k (xxˆ + yyˆ + zzˆ )
JG
c) E = k [2xzxˆ + z 2yˆ + (x 2 + 2yz )zˆ ]

where k is a constant with the appropriate units for the given field. For the possible
electric field, find the electric potential
JG using the origin as your reference point.
Check your answer by verifying that E = −∇V .

Solutions
JG
(
a) E = k yzxˆ + xzyˆ + x 2zˆ )
xˆ yˆ zˆ
JG ∂ ∂ ∂
∇×E =
∂x ∂y ∂z

yz xz x 2

⎡∂ ∂ ⎤ ⎡∂ ∂ 2⎤ ⎡∂ ∂ ⎤
=⎢
⎣ ∂y
( )
x2 −
∂z
(xz )⎥xˆ + ⎢ (yz ) −
⎦ ⎣ ∂z ∂x
( x )⎥yˆ + ⎢ (xz ) −
⎦ ⎣ ∂x ∂y
(yz )⎥zˆ
⎦

= (0 − x )xˆ + (y − 2x )yˆ + (z − z )zˆ = −xxˆ + ( −2x + y )yˆ .

JG
Since ∇ × E ≠ 0, this is not a possible electric field.
JG
b) E = k (xxˆ + yyˆ + zzˆ )

xˆ yˆ zˆ
JG ∂ ∂ ∂
∇×E =
∂x ∂y ∂z
x y z

⎡∂ ∂ ⎤ ⎡∂ ∂ ⎤ ⎡∂ ∂ ⎤
= ⎢ (z ) − (y )⎥xˆ + ⎢ (x ) − (z )⎥yˆ + ⎢ (y ) − (x )⎥zˆ = 0.
⎣ ∂y ∂z ⎦ ⎣ ∂z ∂x ⎦ ⎣ ∂x ∂y ⎦
JG
Since ∇ × E = 0, this is a possible electric field. Let us find the electric potential
by integrating along the path given by

2-13
 Electromagnetism

The potential is given by

r⃗
JG
V=− ∫ E ⋅ dl ⃗ .
O

where
JG
E ⋅ dl⃗ = k (xxˆ + yyˆ + zzˆ ) ⋅ (dxxˆ + dyyˆ + dzzˆ )
= k (x dx + y dy + z dz ).

Note along (i) we only have dx , along (ii) we only have dy , and along (iii) we only
have dz . Therefore, taking the origin O=(0, 0, 0) as our reference point, the
potential will be given by

⎛ ⎞
r⃗ ⎜ x y z ⎟
JG ⎜ ⎟
V=− ∫ ⃗
∫
E ⋅ dl = −k ⎜ x′dx′ + y′dy′ + ∫
z′dz′ ⎟ . ∫
O ⎜ 0 ⎟
⎜  
 
 
⎟
0 0
⎝ (i) (ii) (iii) ⎠

So,

k 2
V (r ) = −
2
(x + y2 + z2 . )

2-14
 Electromagnetism

We can check this using

JG
E = − ∇V
⎧∂ ⎡ k ⎤ ∂ ⎡ k 2 ⎤
= − ⎨ ⎢ − ( x 2 + y 2 + z 2 )⎥xˆ + ⎢ − ( x + y 2 + z 2 )⎥yˆ
⎩ ∂x ⎣ 2 ⎦ ∂y ⎣ 2 ⎦

∂⎡ k 2 ⎤ ⎫
⎢⎣ − ( x + y + z )⎥⎦zˆ⎬
2 2
+
∂z 2 ⎭

2kx 2ky 2kz

= xˆ + yˆ + zˆ
2 2 2
JG
= k (xxˆ + yyˆ + zzˆ ) = E .

JG
c) E = k [2xzxˆ + z 2yˆ + (x 2 + 2yz )zˆ ]
xˆ yˆ zˆ
JG ∂ ∂ ∂
∇×E =
∂x ∂y ∂z
2xz z 2 x 2 + 2yz

⎡∂ ∂ 2⎤ ⎡∂ ∂ 2 ⎤
= ⎢ ( x 2 + 2yz ) − ( z )⎥xˆ + ⎢ (2xz ) − ( x + 2yz )⎥yˆ
⎣ ∂y ∂z ⎦ ⎣ ∂z ∂x ⎦

⎡∂ ∂ ⎤
+ ⎢ ( z 2) − (2xz )⎥zˆ
⎣ ∂x ∂y ⎦

= k ⎡⎣ (2z − 2z )xˆ + (2x − 2x )yˆ + (0 − 0)zˆ⎤⎦ = 0.

JG
Since ∇ × E = 0, this is a possible electric field. Let us find the electric potential by
integrating along the same path as before. The potential is given by
r⃗
JG
V=− ∫ E ⋅ dl ⃗
O

where
JG
E ⋅ dl⃗ = k ⎡⎣ 2xzxˆ + z 2yˆ + x 2 + 2yz zˆ⎤⎦ ⋅ (dxxˆ + dyyˆ + dzzˆ )
( )
= k ⎡⎣ 2xzdx + z 2dy + x 2 + 2yz dz⎤⎦ .
( )
Note along (i) we only have dx with y = 0 and z = 0, along (ii) we only have dy
with x = 1 and z = 0, and along (iii) we only have dz with x = 1 and y = 1.
Therefore, taking the origin O = (0, 0, 0) as our reference point, the potential will be
given by

2-15
 Electromagnetism

⎛ ⎞
r⃗ ⎜ x y z ⎟
JG ⎜ ⎟
V =− ∫ E ⋅ dl ⃗ = − k ⎜ ∫
2x′z dx′ + ∫
z 2dy′ + ∫
(2yz′ + x 2 )dz′ ⎟
O ⎜ 0 ⎟
⎜   
 0  

0   
⎟
⎝ (i) (ii) (iii) ⎠
⎛ ⎞
⎜ x y z ⎟
⎜ ⎟
= − k⎜ ∫
2x′(0)dx′ + 0 d y′ +
2
∫ ∫
(2(1)z′+1 )dz′ ⎟
2

⎜ 0 ⎟
⎜  
 0  

0   
⎟
⎝ (i) (ii) (iii) ⎠

(
V (r ) = − k yz 2 + x 2z . )
We can check this using

JG
E = −∇V
⎧∂ ⎫
= − ⎨ ⎡⎣ −k yz 2 + x 2z ⎤⎦xˆ +
∂ ⎡ 2 ⎤ˆ ∂⎡ 2 ⎤ ⎬
⎩ ∂x
( )
∂y ⎣
− k yz 2
+ x( z ⎦y + )
∂z ⎣
− k (
yz 2
+ x )
z z
ˆ
⎦ ⎭
JG
= k ⎡⎣ (2xz )xˆ + z 2 yˆ + x 2 + 2yz zˆ⎤⎦ = E .
( ) ( )

Problem 2.9. Find the electric field and the electric potential inside and outside a
thin spherical shell of radius R that carries a uniform surface charge σ . Set the
reference point at infinity.

Solution Let us find the electric field everywhere by using Gauss’s law, given by
JG JG q
S
∮
E ⋅ da = enc ,
ε0

2-16
 Electromagnetism

where
JG JG
∮S E ⋅ da = ∮S E da = E ∮S da = E 4πr 2.
For r < R, we have our Gaussian surface given by

where r is the radius of the Gaussian sphere with radius smaller than R. Note that
qenc = 0.
So,
JG
E 4πr 2 = 0 → E = 0.
For r > R, we have our Gaussian surface given by

Now we have
qenc = σ 4πR2 .
So,
σ 4πR2
E 4πr 2 =
ε0

2-17
 Electromagnetism

and
JG σR2
E = r.
ˆ
ε0r 2
Now let us calculate the electric potential everywhere taking the reference point at
infinity. We will use
r
JG
V=− ∫ E ⋅ dl ⃗ ,
∞

where
dl⃗ = dr rˆ + r dθ θˆ + r sin θ dϕ ϕˆ .
For r > R ,

therefore,
r⃗ r⃗ r
JG σR 2 σR 2
V =− ∫ E ⋅ dl ⃗ = − ∫ ε0r 2
( )
rˆ ⋅ dr rˆ + r dθ θˆ + r sin θ dϕ ϕˆ = − ∫ ε0r′2
dr ′
∞ ∞ ∞
2
σR
V= .
ε0r
For r < R

2-18
 Electromagnetism

r⃗ R r
JG σR 2 σR 2 σR
V=− ∫ E ⋅ dl ⃗ = − ∫ ε0r
dr − ∫ 0 dr ′ =
ε0R
−0=
ε0
= const.
∞ ∞ R

Note that the potential inside the shell is constant, as the electric field is zero.

Problem 2.10. Calculate the electric field and the electric potential inside and outside
a solid sphere of radius R having a uniform charge distribution ρ. Use infinity as
your JG
reference point. Then obtain the gradient of the potential everywhere and check
that E = −∇V . Plot the potential versus distance from the center of the sphere.

Solution Starting with the electric field, we use Gauss’s law, given by
JG JG q
S
∮
E ⋅ da = enc ,
ε0
where
JG JG
∮S E ⋅ da = ∮S E da = E ∮S da = E 4πr 2 .

For r > R, we have our Gaussian surface given by

2-19
 Electromagnetism

Here we simply have

ρ4πR3
qenc = ρVsp = ,
3
where Vsp is the volume of the sphere. So
ρ4πR3
E 4πr 2 =
3ε0
and
JG ρR3
E = rˆ .
3ε0r 2

For r < R , our Gaussian surface becomes

Now,
ρ4πr 3
qenc = ρVenc = ,
3
where Venc is the enclosed volume. So
ρ4πr 3
E 4πr 2 =
3ε0

and
JG ρr
E = rˆ .
3ε0

2-20
 Electromagnetism

The plot of electric field is given by

Now we can calculate the electric potential. This is done using

r⃗
JG
V=− ∫ E ⋅ dl ⃗
∞

with infinity as the reference point. For r > R

with

r⃗ r
JG ρR3 ρR3
V=− ∫ E ⋅ dl ⃗ = − ∫ 3ε0r′2
dr ′ =
3ε0r
.
∞ ∞

2-21
 Electromagnetism

For r < R

with
r⃗ R r
JG ρR2 ρr′ ρR2 ρr 2
V=− ∫ E ⋅ dl ⃗ = − ∫ 3ε0r 2
dr − ∫ 3ε0
dr ′ =
2ε0
−
6ε0
.
∞ ∞ R
JG
We can check using E = −∇V . For r > R ,
JG ∂ ⎛ ρR2 ρr 2 ⎞ ρr
E = −∇V = − ⎜ − ⎟ rˆ = rˆ
∂r ⎝ 2ε0 6ε0 ⎠ 3ε0

and for r < R ,

JG ∂ ⎛ ρR3 ⎞ ρR3
E = −∇V = − ⎜ ⎟ rˆ = rˆ
∂r ⎝ 2ε0r ⎠ 3ε0r 2

both of which are in agreement with what we found from Gauss’s law.

Problem 2.11. Calculate the electric field and the electric potential for a sphere of
radius R that carries a charge density ρ = kr 2 , where k is a constant.

2-22
 Electromagnetism

Solution Starting with the electric field, we use Gauss’s law, given by
JG JG q
∮S E ⋅ da = enc ,
ε0

where
JG JG
∮S E ⋅ da = ∮S E da = E ∮S da = E 4πr 2 .

For r > R, we have our Gaussian surface given by

and our enclosed charge is given by

2π π R
R5
qenc = ∫ ρ dτ = ∫ dϕ ∫ sin θ dθ ∫ kr 2r 2dr = 4πk
5
.
0 0 0

Therefore,

4πkR5
E 4πr 2 =
5ε0

so

JG kR5
E = rˆ .
5ε0r 2

2-23
 Electromagnetism

For r < R , our Gaussian surface becomes

Now,
2π π R
4πkr 5
qenc = ∫ ρ dτ = ∫ dϕ ∫ sin θ dθ ∫ kr′2 r′2 dr′ =
5
0 0 0

so
4πkr 5
E 4πr 2 =
5ε 0
and
JG kr 3
E = rˆ .
5ε0
Now we can calculate the electric potential. This is done using
r⃗
JG
V=− ∫ E ⋅ dl ⃗
∞

with infinity as the reference point. For r > R

2-24
 Electromagnetism

with
r⃗ r
JG kR5 kR5
V=− ∫ E ⋅ dl ⃗ = − ∫ 5ε0r′2
dr ′ =
5ε0r
.
∞ ∞

For r < R

with
r⃗ R r
JG kR5 kr′3 kR5 kr 4 kR 4
V =− ∫ E ⋅ dl ⃗ = − ∫ 5ε0r′2
dr ′ − ∫ 5ε0
dr ′ =
5ε0R
−0−
20ε0
+
20ε0
∞ ∞ R

k
=
20ε0
(
4R 4 − r 4 + R 4 )
kR 4 ⎛ r4 ⎞
V= ⎜5 − 4 ⎟ .
20ε0 ⎝ R ⎠

Problem 2.12 A long cylinder of radius a carries a charge density ρ = ks 2 , where k is

a constant and s is the distance from the axis of the cylinder. Find the electric field
and the electric potential everywhere. Take the reference point at a distance b from
the axis (b > a ).

2-25
 Electromagnetism

Solution Starting with the electric field, we use Gauss’s law, given by
JG JG q
∮ S
E ⋅ da = enc .
ε0
Note the left-hand side is always given by
JG JG
∮S E ⋅ da = ∮S E da = E ∮S da = E 2πsl .

For s > a , we have our Gaussian surface given by

with enclosed charge given by

2π l a
πk la 4
qenc = ∫ ρ dτ = ∫ dϕ ∫ dz ∫ ks 2s ds =
2
.
0 0 0

Therefore,
πk la 4
E 2πsl =
2ε 0
and
JG ka 4
E = sˆ .
4ε0s
For s < a , our Gaussian surface becomes

2-26
 Electromagnetism

Now,
2π l s
πk ls 4
qenc = ∫ ρ dτ = ∫ dϕ ∫ dz ∫ ks′2 s′ ds′ =
2
0 0 0

so
πk ls 4
E 2πsl =
2ε 0

and
JG ks 3
E = sˆ .
4ε0

Now we can calculate the electric potential. This is done using

r⃗
JG
V=− ∫JG E ⋅ dl ⃗
b

with b as the reference point. For s > a ,

with
r⃗ s
JG ka 4 ka 4 ka 4 s
V=− ∫JG E ⋅ dl ⃗ = − ∫ 4ε0s′
ds ′ = −
4ε 0
(ln s − ln b) = −
4ε 0
ln .
b
b b

2-27
 Electromagnetism

For s < a ,

with
a s
JG ka 4 ks′3 ka 4 a k (s 4 − a 4 )
V =− ∫ E ⋅ dl ⃗ = − ∫ 4ε0s
ds − ∫ 4ε 0
ds ′ = −
4ε 0
ln −
b 16ε0
b a

V=
ka 4 b
ln +
(
k a4 − s4
.
)
4ε 0 a 16ε0

Problem 2.13. Verify the electrostatic boundary condition using the charge distri-
bution in problem 2.9.
Solution The electrostatic boundary condition is given by
JG JG σ
Eabove − E below = n.ˆ
ε0
From problem 2.9, our electric fields are
⎧ 0 r<R
JG ⎪
E = ⎨ σR 2 .
⎪ 2 rˆ r > R
⎩ ε0r
At r = R , we have
JG σ
Eabove = rˆ
ε0
and
JG
E below = 0.
Therefore,
JG JG σ σ
Eabove − E below = rˆ − 0 = nˆ ,
ε0 ε0

where n̂ is normal to the sphere which has the direction of r̂ .

2-28
 Electromagnetism

Problem 2.14. Find the work required to assemble the charge distribution below.

Solution We can denote the following q1 = 3q , q2 = −2q, q3 = −q, q4 = q . Starting

with q1, W = 0. Moving in q2 , we have

1 ⎛q ⎞ 6q 2 1 q2 3 2
W2 = q2⎜ 1 ⎟ = − =− .
4πε0 ⎝ r12 ⎠ 4πε0 a 2 4πε0 a

Moving in q3, we have

1 ⎛q q ⎞ q ⎛ 3q 2q ⎞ q 2 ⎛ 2 2 −3 ⎞
W3 = q3⎜ 1 + 2 ⎟ = − ⎜ − ⎟= ⎜ ⎟.
4πε0 ⎝ r13 r 23 ⎠ 4πε0 ⎝ 2a 2 a ⎠ 4πε0 ⎝ 2a ⎠

Moving in q4 , we have

1 ⎛q q q ⎞ q ⎛ 3q 2q q ⎞
W4 = q4⎜ 1 + 2 + 3 ⎟ = ⎜ − − ⎟
4πε0 ⎝ r14 r 24 r34 ⎠ 4πε0 ⎝ a 10 2a a 2⎠

q 2 ⎛ 3 10 − 5 2 − 10 ⎞
= ⎜ ⎟.
4πε0 ⎝ 10a ⎠

Therefore,

q2 ⎛ 3 2 2 2 −3 3 10 − 5 2 − 10 ⎞
W = W2 + W3 + W4 = ⎜− + + ⎟
4πε0 ⎝ a 2a 10a ⎠
⎡ ⎤
W=
(
q 2 ⎢ 3 10 − 25 2 + 1 ⎥ )
4πε0 ⎢ 10a ⎥.
⎣ ⎦

Problem 2.15. Find the energy stored in a spherical shell of inner radius a and outer
radius b with a charge distribution ρ = kr 2 .

2-29
 Electromagnetism

Solution The work is given by

ε0
W= 
2
∫ E 2 dτ
all
space

so we need to find the field in all three regions.

For r < a , we have qenc = 0. So E = 0.
For a < r < b, we have
JG JG q
∮S E ⋅ da = enc
ε0

with
r
4πk 5
qenc = ∫ ρ dτ = 4π ∫ k (r′)2 (r′)2 dr′ =
5
(r − a5)
a

and
JG JG
∮S E ⋅ da = E 4πr 2 .

So

k (r 5 − a 5)
E= .
5ε0r 2

For r > b, we have

b
4πk 5
qenc = ∫ ρ dτ = 4π ∫ k (r′)2 (r′)2 dr′ =
5
(b − a5).
a

2-30
 Electromagnetism

So

E=
(
k b5 − a 5 ).
2
5ε0r
Now the work is given by
⎧ ⎡ k r5 − a5 ⎫
) ⎤⎥ r ⎡ k b5 − a 5
) ⎤⎥ r
2 2
ε0 ⎪
W = ⎨4π
b
⎢ ( ∞
⎢ ( ⎪
dr ⎬
2⎪
∫ ⎢ 5ε0r 2 ⎥
2
dr + 4π ∫ ⎢ 5ε0r 2 ⎥
2
⎪
⎩ a ⎣ ⎦ b ⎣ ⎦ ⎭
k 2π
W=
45ε0
(5a 9 − 9a 5b 4 + 4b9 . )

Problem 2.16. Given a charge density ρ = ke−r , with k a constant, find the radius of
a sphere that maximizes the energy per unit volume.

Solution
The energy per unit volume is given by
W ε
= 0 E 2,
volume 2
where the field is given by
JG JG q
∮S E ⋅ da = enc
ε0
with
r
2
qenc = ∫ ρ dτ = 4π ∫ ( )
k e− r ′ r′ (
dr′ = 4πk e−r 2e r − r 2 − 2r − 2 . )
0

Also,
JG JG
∮S E ⋅ da = E 4πr 2 .

So

E=
(
k e−r 2e r − r 2 − 2r − 2 ).
2
ε0r

The energy per unit volume contained in a sphere of radius r is given by

2
W ε
= 0E2 =
k 2 e−2r 2e r − r 2 − 2r − 2
.
( )
volume 2 2ε0r 4

2-31
 Electromagnetism

To maximize this, we have

d ⎛⎜ W ⎞⎟
=0
dr ⎝ volume ⎠
so

( )(
−k 2 e−2r 2e r − r 2 − 2r − 2 4e r − r 3 − 2r 2 − 4r − 4 ) = 0.
5
ε0r
Since r ≠ 0, k ≠ 0, and e−2r ≠ 0, we have
2e r − r 2 − 2r − 2 = 0 → r = 0
and
4e r − r 3 − 2r 2 − 4r − 4 = 0 → r = 0, r = 1.45123.
But r ≠ 0, so a sphere of radius r = 1.45123 has the maximum energy per unit
volume.

Problem 2.17. A metal sphere of radius R and charge q is surrounded by two

concentric metal shells.
a) Obtain the surface charge density σ at R , a , b, c , and d .
b) Calculate the potential at the center of the sphere by taking infinity as the
reference point.

Solutions
a) Obtain the surface charge density σ at R , a , b, c , and d .
For r = R , the sphere is metallic, therefore all charge q is distributed on the
surface of the sphere. This gives a surface charge density of

2-32
 Electromagnetism

q
σ=
4πR2
By influence and due to the sphere with charge q , the inner shell is redistributing the
electric charge such that the surface with radius a has charge –q and the surface with
radius b has charge +q . Similarly for the outer shell.
Therefore, the surface charge densities are

q q q q
σa = − σb = σc = − σd = .
4πa 2 4πb 2 4πc 2 4πd 2

b) Calculate the potential at the center of the sphere by taking infinity as the
reference point

Taking our reference point at infinity, the electric potential at the center is given by
0
JG
V = − ∫ E ⋅ dl⃗
∞
d c b a R 0
q q q
= ∫ 4πε0r 2
dr − ∫ 0 dr − ∫ 4πε0r 2
dr − ∫ 0 dr − ∫ 4πε0r 2
dr − ∫ 0 dr
∞ d c b a R

q ⎛1 1 1 1 1⎞
V= ⎜ + − + − ⎟.
⎝
4πε0 d b c R a⎠

Problem 2.18. Calculate the capacitance of the spherical shell capacitor of radii a
(inner) and b (outer) shown below.

Solution The electric field is due to the inner charge, so

Q
E= .
4πε0r 2

2-33
 Electromagnetism

The electric potential is then

b a
JG Q Q ⎛1 1⎞ Q (b − a )
V=− ∫ E ⋅ dl ⃗ = ∫ dr = ⎜ − ⎟= .
4πε0r 2 ⎝
4πε0 a b ⎠ 4πε0 ab
a b

We can find capacitance using

Q Q
V= →C= .
C V
So,
Q 4πε0ab
C= = .
Q (b − a ) b−a
4πε0 ab

Problem 2.19. Calculate the capacitance of a cylindrical capacitor of length L with

two metal cylinders of radii a (inner) and b (outer) shown below. Ignore the edge
effects, obtain the capacitance per unit length.

Solution Let us consider that the charge on the inner cylinder is Q (at the radius a ).
The electric field is obtained from Gauss’s law
JG JG q
∮S E ⋅ da = enc ,
ε0

where
JG JG
∮S E ⋅ da = ∮S E da = E ∮S da = E 2πsL

and the enclosed charge is

qenc = Q.

2-34
 Electromagnetism

Therefore,
JG Q
E = sˆ .
2πε0sL
The potential difference between the two cylinders is then
b b
JG Q Q b
V (b ) − V (a ) = − ∫ E ⋅ dl ⃗ = − ∫ 2πε0sL
ds =
2πε0L
ln .
a
a a

Therefore, the capacitance is given by

Q 2πε0L
C= = .
V b
ln
a

Bibliography
Byron F W and Fuller R W 1992 Mathematics of Classical and Quantum Physics (New York:
Dover)
Griffiths D J 1999 Introduction to Electrodynamics 3rd edn (Englewood Cliffs, NJ: Prentice Hall)
Griffiths D J 2013 Introduction to Electrodynamics 4th edn (New York: Pearson)
Halliday D, Resnick R and Walker J 2010 Fundamentals of Physics 9th edn (New York: Wiley)
Halliday D, Resnick R and Walker J 2013 Fundamentals of Physics 10th edn (New York: Wiley)
Jackson J D 1998 Classical Electrodynamics 3rd edn (New York: Wiley)
Rogawski J 2011 Calculus: Early Transcendentals 2nd edn (San Fransisco, CA: Freeman)

2-35
 IOP Concise Physics

Electromagnetism
Problems and solutions
Carolina C Ilie and Zachariah S Schrecengost

Chapter 3
Electric potential

Chapter 3 contains different methods for obtaining the electric potential. We will
focus on calculating the potential as finding the field is a straightforward calculation
once the potential has been determined. Laplace’s equation is solved using different
methods, depending on the type of charge distribution and on the symmetry
of the problem. The method of images, separation of variables, and multipole
(in particular dipole) expansions are discussed using appropriate examples.

3.1 Theory
3.1.1 Laplace’s equation
Cartesian
∂ 2T ∂ 2T ∂ 2T
∇2 T = 2
+ 2
+ =0
∂x ∂y ∂z 2
Cylindrical
1 ∂ ⎛ ∂T ⎞ 1 ∂ 2T ∂ 2T
∇2 T = ⎜s ⎟ + + =0
s ∂s ⎝ ∂s ⎠ s 2 ∂ϕ 2 ∂z 2
Spherical
1 ∂ ⎛ 2 ∂T ⎞ 1 ∂ ⎛ ∂T ⎞ 1 ∂ 2T
∇2 T = ⎜r ⎟ + ⎜sin θ ⎟ + =0
r 2 ∂r ⎝ ∂r ⎠ r 2 sin θ ∂θ ⎝ ∂θ ⎠ r 2 sin2 θ ∂ϕ 2

3.1.2 Solving Laplace’s equation

As an introduction to solving problems using Laplace’s equation, we will outline the
solutions in Cartesian and spherical coordinates. Laplace’s equation can be solved
by the method separation of variables when we know the boundary conditions. The

doi:10.1088/978-1-6817-4429-2ch3 3-1 ª Morgan & Claypool Publishers 2016

 Electromagnetism

general solutions will be outlined below, but seeing how they are derived is
important. We will leave them in a general form and problems in this chapter will
provide examples of using the boundary conditions to solve for the constants.

Two-dimensional Cartesian coordinates

Let us look at a general two-dimensional case where Laplace’s equation is given by
∂ 2V ∂ 2V
∇2 V = + = 0.
∂x 2 ∂y 2
We look for a solution of the type V (x , y ) = X (x )Y (y ) and we replace the desired
solution in Laplace’s equation which becomes
∂ 2X (x ) ∂ 2Y (y )
Y (y ) + X ( x ) =0
∂x 2 ∂y 2
and in a simpler form
∂ 2X ∂ 2Y
Y + X = 0.
∂x 2 ∂y 2
We want to separate the variables, which can easily be done by dividing the equation
by X (x )Y (y ) = V (x , y )
1 ∂ 2X 1 ∂ 2Y
+ = 0.
X ∂x 2 Y ∂y 2
Note that the first term depends only on x and the second term depends only on y .
This means that each of the two terms must be constant, and the two constants must
be equal in magnitude but opposite in sign. So

1 ∂ 2X 1 ∂ 2Y
=F = −F .
X ∂x 2 Y ∂y 2

We choose F positive, and we can rewrite the equations as

1 d2X 1 d2Y
= k2 = −k 2.
X dx 2 Y dy 2

Note that the initial partial differential equation was replaced by two ordinary
differential equations. Rearranging yields

d2X d2Y
= k 2X = −k 2Y .
dx 2 dy 2

The two equations have the following solutions

X (x ) = Ae kx + B e−kx

3-2
 Electromagnetism

and
Y (y ) = C sin(ky ) + D cos(ky ).
Going back to the electric potential, V becomes

V (x , y ) = X (x )Y (y ) = Ae kx + B e−kx ⎡⎣ C sin(ky ) + D cos(ky )⎤⎦ .

( )
The next step is to apply the boundary conditions in order to obtain the constants A,
B , C , and D and to (usually) impose some constraints on k .

Two-dimensional spherical coordinates

Here we will assume azimuthal symmetry (no dependence on ϕ) where Laplace’s
equation is given by
1 ∂ ⎛ 2 ∂V ⎞ 1 ∂ ⎛ ∂V ⎞
∇2 V = ⎜r ⎟ + ⎜sin θ ⎟ = 0.
r ∂r ⎝ ∂r ⎠ r sin θ ∂θ ⎝
2 2
∂θ ⎠

We look for a solution which has a radial component and an angular

component
V (r , θ ) = R(r )Θ(θ ).

Note that here R is the function of r , and not merely the radius of the sphere. We
plug our solution in the previous equation and we obtain
∂ ⎛ 2 ∂R(r ) ⎞ R(r ) ∂ ⎛ ∂Θ(θ ) ⎞
Θ(θ ) ⎜r ⎟+ ⎜sin θ ⎟ = 0.
∂r ⎝ ∂r ⎠ sin θ ∂θ ⎝ ∂θ ⎠

We want to use the method of separation of variables so we will divide the previous
equation by V (r, θ ) = R(r )Θ(θ ),
1 d ⎛ 2 dR(r ) ⎞ 1 d⎛ dΘ(θ ) ⎞
⎜r ⎟+ ⎜sin θ ⎟ = 0.
R(r ) dr ⎝ dr ⎠ Θ(θ )sin θ dθ ⎝ dθ ⎠

Note that each term is only a function of a single variable so we were able to replace
the partial derivates with ordinary derivates. Now we have one term in R(r ) and
another term in Θ(θ ), so we have separated the variables. Therefore, each term must
be constant. For well know reasons (more apparent in quantum mechanics), we
choose the constant as following

1 d ⎛ 2 dR(r ) ⎞
⎜r ⎟ = l (l + 1)
R(r ) dr ⎝ dr ⎠

1 d⎛ dΘ(θ ) ⎞
⎜sin θ ⎟ = − l (l + 1).
Θ(θ )sin θ dθ ⎝ dθ ⎠

Now let us analyze each of the equations and find the solution.

3-3
 Electromagnetism

The radial equation

d ⎛ 2 dR(r ) ⎞
⎜r ⎟ = l (l + 1)R(r )
dr ⎝ dr ⎠
has the general solution, with A and B constants
B
R(r ) = Ar l + .
r l +1
The angular equation
d⎛ dΘ(θ ) ⎞
⎜sin θ ⎟ = −l (l + 1)Θ(θ )sin θ
dθ ⎝ dθ ⎠
is not at all trivial. The solutions constitute Legendre polynomials with the variable
cos θ . Legendre polynomials are a special class of polynomials. So the solution to the
angular equation is

Θ(θ ) = Pl (cos θ ),

where the general form is given by Rodrigues formula

1 ⎛ d ⎞l 2 2
Pl (x ) = l ⎝
⎜ ⎟ x −1 .
2 l! dx ⎠
( )
Therefore, the separable solution of the Laplace equation, considering azimuthal
symmetry, is

⎛ B ⎞
V (r , θ ) = R(r )Θ(θ ) = ⎜Ar l + l +1 ⎟Pl (cos θ )
⎝ r ⎠

and the general solution is the linear combination of the separable solutions
∞
⎛ Bl ⎞
V (r , θ ) = ∑⎜⎝Al r l + ⎟Pl (cos θ ).
r l +1 ⎠
l =0

3.1.3 General solutions

Cartesian
∂ 2V ∂ 2V
+ = 0 → V (x , y ) = (Ae kx + B e−kx)⎡⎣ C sin(ky ) + D cos(ky )⎤⎦
∂x 2 ∂y 2
Spherical
1 ∂ ⎛ 2 ∂V ⎞ 1 ∂ ⎛ ∂V ⎞
⎜r ⎟ + ⎜sin θ ⎟ = 0
r ∂r ⎝ ∂r ⎠ r sin θ ∂θ ⎝
2 2
∂θ ⎠

3-4
 Electromagnetism

∞
⎛ Bl ⎞
V (r , θ ) = ∑ ⎜⎝Al r l + ⎟Pl (cos θ ),
r l +1 ⎠
l =0

where Pl are Legendre polynomials given by the Rodrigues formula

1 ⎛ d ⎞l 2 l
Pl (x ) = l ⎝
⎜ ⎟ x −1 .
2 l! dx ⎠
( )
Note
P0(x ) = 1
P1(x ) = x
3x 2 − 1
P2(x ) =
2
5x 3 − 3x
P3(x ) =
2
35x 4 − 30x 2 + 3
P4(x ) = .
8
Cylindrical
1 ∂ ⎛ ∂V ⎞ 1 ∂ 2V ∂ 2V
⎜s ⎟ + + =0
s ∂s ⎝ ∂s ⎠ s 2 ∂ϕ 2 ∂z 2

∞
V (s , ϕ) = a0 + b0 ln(s ) + ∑ {s k⎡⎣ ak cos(kϕ) + bk sin(kϕ)⎤⎦
k=1

+ s−k⎡⎣ ck cos(kϕ) + dksin(kϕ)⎤⎦ . }

3.1.4 Method of images

The method of images is a very useful technique for calculating the electric field
and the electric potential for problems with symmetry. By using the uniqueness
theorem we know that the electric field is uniquely determined at any point in
space, thus we can replace an apparently difficult problem with another problem
in which we use the initial charge(s) and also the ‘images’ of the charges. The
boundary conditions need to be fulfilled. Typically, this involves the condition for
a zero electric potential for a grounded conductor and the condition for zero
potential very far away from the system of charges. The potential can be
determined in the permitted region, which is in general the region of the real
charge. The region of the image charges is the ‘forbidden’ region; the potential
cannot be calculated there. The best way to learn this method is by solving
problems and checking the examples.

3-5
 Electromagnetism

3.1.5 Potential due to a dipole

JG
1 qd cos θ 1 p · rˆ
V (r ⃗ ) ≅ =
4π ε 0 r2 4π ε 0 r 2
3.1.6 Multiple expansion

∞
1 1
V (r ⃗ ) = ∑
4π ε 0 n = 0 r n + 1
∫ (r′)nPn(cos θ′) ρ(r ⃗′)dτ′
1 ⎡1
V (r ⃗ ) = ⎢
4π ε 0 ⎣ r
∫ ρ(r ′⃗ )dτ′ + r12 ∫ r′ cos θ′ ρ(r ⃗′)dτ′
1 ⎛ ⎞ ⎤
+
r3
∫ (r′)2⎜⎝ 32 cos2 θ′ − 12 ⎟⎠ ρ(r ⃗′)dτ′ + ⋯⎥⎦.
3.1.7 Monopole moment
n
Q= ∑qi
i=0

3.2 Problems and solutions

Problem 3.1. Solve the Laplace equation in spherical and cylindrical coordinates for
the cases where V is only dependent on one coordinate at a time.
Solution In spherical coordinates

1 ∂ ⎛ 2 ∂V ⎞ 1 ∂ ⎛ ∂V ⎞ 1 ∂ 2V
∇2 V = ⎜r ⎟ + ⎜sin θ ⎟ + = 0.
r 2 ∂r ⎝ ∂r ⎠ r 2 sin θ ∂θ ⎝ ∂θ ⎠ r 2 sin2 θ ∂ϕ 2

If V only depends on r

1 d ⎛ 2 dV ⎞
⎜r ⎟ = 0,
r 2 dr ⎝ dr ⎠

which means

dV
r2 = C.
dr

So

V=C ∫ r−2 dr → V (r) = C ( −r−1 + A) = k − Cr .

3-6
 Electromagnetism

If V only depends on θ
1 d⎛ dV ⎞
⎜sin θ ⎟ = 0,
r sin θ dθ ⎝
2
dθ ⎠
which mean
dV
sin θ = C.
dθ
So

V =C ∫ sin1 θ dθ = C ∫ csc θ dθ = C ( ln csc θ − cot θ )

+A

V (θ ) = k + C ln csc θ − cot θ .
If V only depends on ϕ
1 d2V
= 0,
r 2 sin2 θ dϕ 2
which means
dV
= C.
dϕ
So
V (ϕ) = k + Cϕ .
In cylindrical coordinates
1 ∂ ⎛ ∂V ⎞ 1 ∂ 2V ∂ 2V
⎜s ⎟ + + = 0.
s ∂s ⎝ ∂s ⎠ s 2 ∂ϕ 2 ∂z 2

If V only depends on s

1 d ⎛ dV ⎞
⎜s ⎟ = 0,
s ds ⎝ ds ⎠

which means
dV
s = C.
ds
So

V =C ∫ s−1 ds = C ( ln s )
+A

V (s ) = k + C ln s .

3-7
 Electromagnetism

If V only depends on ϕ,
1 d2V
= 0,
s 2 dϕ 2
which means
dV
= C,
dϕ
which is the same as ϕ dependence in spherical coordinates. So
V (ϕ) = k + Cϕ .
If V only depends on z
d2V
= 0,
dz 2
which means
dV
= C,
dz
which is the same form as ϕ dependence. So
V (z ) = k + Cz .

Problem 3.2. In two-dimensional Cartesian coordinates, the general solution to the

Laplace equation is

V (x , y ) = Ae kx + B e−kx ⎡⎣ C sin(ky ) + D cos(ky )⎤⎦ .

( )
Verify that this does in fact satisfy the Laplace equation.
Solution Here, the Laplace equation is
∂ 2V ∂ 2V
+ = 0.
∂x 2 ∂y 2
∂ 2V ∂ 2V
So we need to compute and . We have
∂x 2 ∂y 2
∂V
= Ak e kx − Bk e−kx ⎡⎣ C sin(ky ) + D cos(ky )⎤⎦
( )
∂x

and

∂ 2V
= ( Ak e 2 kx
+ Bk 2 e−kx ⎡⎣ C sin(ky ) + D cos(ky )⎤⎦
)
∂x 2
= k 2 Ae kx + B e−kx ⎡⎣ C sin(ky ) + D cos(ky )⎤⎦ .
( )

3-8
 Electromagnetism

Also,
∂V
= Ae kx + B e−kx ⎡⎣ Ck cos(ky ) − Dk sin(ky )⎤⎦
( )
∂y
and
∂ 2V
= Ae kx + B e−kx ⎡⎣ −Ck 2 sin(ky ) − Dk 2 cos(ky )⎤⎦
( )
∂y 2

= − k 2 Ae kx + B e−kx ⎡⎣ C sin(ky ) + D cos(ky )⎤⎦ .

( )
Putting this all together, we have
∂ 2V ∂ 2V
+ = k 2 Ae kx + B e−kx ⎡⎣ C sin(ky ) + D cos(ky )⎤⎦
( )
∂x 2 ∂y 2
− k 2 Ae kx + B e−kx ⎡⎣C sin(ky ) + D cos(ky )⎤⎦
( )
∂ 2V ∂ 2V
+ =0
∂x 2 ∂y 2
as expected.

Problem 3.3. A charge q is placed at a distance d from an infinite grounded

conducting plane. Using the method of images, find the electric potential. Which is
the ‘forbidden’ region, for which we cannot calculate the potential?

Solution We replace the previous problem with a completely different problem:

charge q at (0, 0, d ) and its image, charge −q situated at (0, 0, −d). The grounded
conducting plane disappeared. This new problem is shown below

3-9
 Electromagnetism

The potential needs to satisfy the following boundary conditions:

a) V = 0 for z = 0 (grounded plane in the initial problem).
b) V → 0 for a point far from charge q

x 2 + y 2 + z 2 ≫ d 2.

The electric potential due to both point charges is:

q −q
V (x , y , z ) = + .
4π ε 0 x 2 + y 2 + ( z − d ) 2 4π ε 0 x 2 + y 2 + ( z + d ) 2

Let us check the boundary conditions:

a) For z = 0 it is easy to see that V = 0.
b) For x 2 + y 2 + z 2 ≫ d 2 , x 2 + y 2 + (z − d ) 2 ≅ x 2 + y 2 + (z + d ) 2 , and
V → 0.

It is important to note that the only region for which we are able to obtain the
electric potential is the region in space above the grounded conducting plane, i.e. the
semi-space where charge q is located. For z < 0, we are not able to obtain the electric
potential.

Problem 3.4. A charge q is placed in an opened grounded conducting parallelepiped

at (a, b, c ), where a, b, c are positive. Using the method of images, obtain the
electric potential in the region of the charge q (for which x > 0, y > 0, z > 0).

3-10
 Electromagnetism

Solution We have the one real charge q at (a, b, c ) and seven image charges as
following: −q at ( −a, −b, −c ), ( −a, b, c ), (a, −b, c ), and (a, b, −c ); q at
(a, −b, −c ), ( −a, b, −c ), ( −a, −b, c ). The electric potential is given by

V (x , y , z )
⎛
1 ⎜ q q
= ⎜ +
4π ε 0 ⎝ ( x − a ) + ( y − b ) 2 + ( z − c ) 2
2
(x − a ) + (y + b )2 + (z + c )2
2

q q
+ +
2 2 2
(x + a ) + (y − b ) + (z + c ) (x + a ) + (y + b )2 + (z − c )2
2

−q −q
+ +
(x + a )2 + (y + b )2 + (z + c )2 (x − a )2 + (y + b )2 + (z − c )2

⎞
−q −q ⎟.
+ + ⎟
(x − a )2 + (y − b )2 + (z + c )2 (x + a )2 + (y − b )2 + (z − c )2 ⎠

If we check the limit x = 0, y = 0, z = 0 successively, we obtain a zero potential for

all the three sides of the parallelepiped, where the grounded conductors were in the
equivalent problem. Also, for the points far away from the point (a, b, c ) in the
‘eighth’ part of space for which x ≫ 0, y ≫ 0, z ≫ 0, the potential becomes zero as
well. Again, note that this is the electric potential only for this part of space,
accessible for investigation using the method of images.

Problem 3.5. Let us imagine that we have n charges placed as following: q1 = −q at

(0, 0, d ), q2 = 2q at (0, 0, 2d ), …, qn = ( −1) nnq at (0, 0, nd ) above a grounded,
conducting xy -plane (shown below). Obtain the electric potential using the method
of images.

3-11
 Electromagnetism

Solution The image charges will be equal in magnitude, of different sign, and
situated symmetrically with the xy -plane. In the new problem we eliminate the
grounded, conducting plane, but we use the xy -plane for geometrical purposes.

For each charge qn = ( −1) nnq we have the image charge qn′ = ( −1) n+1nq located at
(0, 0, −nd ). The electric potential is, therefore,

3-12
 Electromagnetism

⎛
q ⎜ −1 1
V (x , y , z ) = ⎜ +
4πε0 ⎝ x + y + (z − d )2
2 2
x + y + (z + d )2
2 2

( −1)2 2 ( −1)2+12
+ + + ⋯
x 2 + y 2 + ( z − 2d ) 2 x 2 + y 2 + ( z + 2d ) 2
( −1)k k ( −1)k+1k
+ +
x 2 + y 2 + (z − kd )2 x 2 + y 2 + (z + kd )2
⎞
( −1)nn ( −1)n+1n ⎟.
+⋯ + + ⎟
x 2 + y 2 + (z − nd )2 x 2 + y 2 + (z + nd )2 ⎠

It is easy to see that V = 0 for z = 0 and also V = 0 for a point very far from charge
x 2 + y 2 + z 2 ≫ (nd )2 .

Problem 3.6. A conducting sphere of radius R , centered at the origin, is grounded.

Find the potential outside the sphere, if a point charge +q is placed at a distance d
from the sphere, d > R . Use the method of images.

Solution We replace our problem with the grounded, conducting sphere of

radius R and the charge +q at distance d > R with a different problem. The
sphere, the charge +q and the image charge q′, situated at (0, 0, a), with a < R ,
is given by

3-13
 Electromagnetism

We need V = 0 everywhere on the sphere of radius R . Note that we can only find
the electric potential outside the sphere. Consider the point P depicted above; here
the electric potential at P is given by
1 ⎛q q′ ⎞ 1 ⎛ q q′ ⎞
V (r ⃗ ) = ⎜ + ⎟= ⎜ JG + JG ⎟ .
4π ε0 ⎝ r1 r 2 ⎠ 4π ε 0 ⎝ r ⃗ − d r⃗ − a ⎠

Using the law of cosines, we have

r 12 = d 2 + r 2 − 2dr cos θ
and
r 22 = a 2 + r 2 − 2ar cos θ .
We can rewrite V as

1 ⎛ q q′ ⎞
V (r ⃗ ) = ⎜ + ⎟.
4π ε0 ⎝ d + r − 2dr cos θ
2 2
a 2 + r 2 − 2ar cos θ ⎠

The potential should be zero for r = R

1 ⎛ q q′ ⎞
V (R ) = ⎜ + ⎟ = 0.
4π ε0 ⎝ d + R − 2dR cos θ
2 2
a 2 + R2 − 2aR cos θ ⎠

So
q − q′
= .
d 2 + R2 − 2Rd cos θ a 2 + R2 − 2aR cos θ
We need to obtain both q′ and a , so we need two equations. We choose two
convenient values for θ , θ = 0 and θ = π . For θ = 0, cos θ = 1, so

3-14
 Electromagnetism

q − q′ q − q′
= → = .
d 2 + R2 − 2Rd a 2 + R2 − 2aR (d − R )2 (R − a )2

Choosing the positive square root,

q − q′
=
d−R R−a

and solving for the image charge, we obtain

R−a
q′ = −q .
d−R

For θ = π , cos θ = −1, so

q − q′
= .
d+R R+a

By substituting q′, we have

q (R − a )q
= .
d+R (d − R )(a + R )

Solving for a , we obtain

R2
a= .
d

Using this, we can find our image charge

⎛ R2 ⎞
⎜R − ⎟q
⎝ d ⎠ R(d − R )q
q′ = − =−
d−R (d − R )d
R
q′ = − q.
d

Now we have the electric potential, since we obtained the image charge and its
position.

Problem 3.7. Given two infinitely long grounded plates at y = 0 and y = a

connected by the metal strip at x = −b with constant potential –V0 and x = b
with constant potential V0. Find the potential inside the pipe.

3-15
 Electromagnetism

Solution This is independent of z so we have

∂ 2V ∂ 2V
2
+ =0
∂x ∂y 2
with boundary conditions
(i) V (y = 0) = 0
(ii) V (y = a ) = 0
(iii) V (x = b ) = V0
(iv) V (x = −b ) = −V0 .

Our general solution is given by

V (x , y ) = Ae kx + B e−kx ⎡⎣ C sin(ky ) + D cos(ky )⎤⎦ .

( )
From boundary condition (i)

( )
V (x , 0) = Ae kx + B e−kx (D ) = 0 ⇒ D = 0.

So our solution becomes

V (x , y ) = Ae kx + B e−kx ⎡⎣ C sin(ky )⎤⎦ .
( )
From boundary condition (ii)

V (x , y ) = Ae kx + B e−kx ⎡⎣ C sin(ka )⎤⎦ = 0

( )
we have
nπ
k= ,
a

3-16
 Electromagnetism

where n is an integer. By symmetry

V (x , y ) = − V ( − x , y )

( Ae kx
) (
+ B e−kx [C sin(ka )] = − Ae−kx + B e kx [C sin(ka )] )
Ae kx + B e−kx = −Ae−kx + B e kx

( )
A e kx + e−kx = −B e kx + e−kx . ( )
So
A = −B.
Absorbing our constants, the solution becomes

( )
V (x , y ) = C e kx + e−kx sin(ky ) = C sinh(kx ) sin(ky )

and in general
∞
⎛ nπx ⎞ ⎛ nπy ⎞
V (x , y ) = ∑Cn sinh⎜⎝ ⎟sin⎜
a ⎠ ⎝ a ⎠
⎟.

n=1

Now to find Cn we take

∞
⎛ nπb ⎞ ⎛ nπy ⎞
V (b , y ) = ∑Cn sinh⎜⎝ ⎟ sin⎜
a ⎠ ⎝ a ⎠
⎟ = V
0
n=1

so
a a
∞
⎛ nπb ⎞ ⎛ nπy ⎞ ⎛ n′πy ⎞ ⎛ n′πy ⎞
∑Cn sinh⎜⎝ ⎟⎠ ∫ sin⎜ ⎟ sin⎜
⎝ a ⎠ ⎝ a ⎠
⎟dy = ∫ V0 sin⎜
⎝ a ⎠
⎟dy .
n=1
a
0 0

Note that when n ≠ n′,

a
⎛ nπy ⎞ ⎛ n′πy ⎞
⎟ sin⎜ ⎟dy = 0
∫ sin⎜
⎝ a ⎠ ⎝ a ⎠
0

and when n = n′,

a
⎛ nπy ⎞ ⎛ n′πy ⎞ a
⎟ sin⎜ ⎟dy = .
∫ sin⎜
⎝ a ⎠ ⎝ a ⎠ 2
0

Therefore,
⎧ 0 n is even
⎛ nπb ⎞ a V0a ⎪
Cn sinh⎜ ⎟ = [1 − cos(nπ )] = ⎨ 2V0a .
⎝ a ⎠2 nπ ⎪ n is odd
⎩ nπ

3-17
 Electromagnetism

So
4V0 1
Cn =
nπ sinh nπb ( )a

for n = 1, 3, 5, … and our potential is given by

4V0
∞
1 sinh ( ) sin⎛ nπy ⎞.
n πx
a
V (x , y ) = ∑ ⎜
⎝ a ⎠
⎟
π n=1,3,5 n sinh ( )
n πb
a

Problem 3.8. Suppose a thin spherical shell of radius R has potential

3
V (θ ) = V0(1 − 2 sin2 θ ) specified at the surface. Find the potential inside and outside
the sphere.
Solution Our general solution is given by
∞
⎛ Bl ⎞
V (r , θ ) = ∑⎜⎝Al r l + ⎟Pl (cos θ ).
r l +1 ⎠
l =0

Inside:
Here, we must have Bl = 0 so the potential does not blow up at the origin. So our
potential becomes
∞
V (r , θ ) = ∑Al r l Pl (cos θ ).
l =0

At the surface, we have

∞
⎛ 3 ⎞
V (R , θ ) = ∑Al Rl Pl (cos θ ) = V0⎜⎝1 − sin2 θ ⎟ .
⎠
l =0
2

Note that
⎛ 3 ⎞ ⎡ 3 ⎤ ⎛ 3 cos2 θ − 1 ⎞
V0⎜1 − sin2 θ ⎟ = V0⎢1 − (1 − cos2 θ )⎥ = V0⎜ ⎟ = V0P2(cos θ ).
⎝ 2 ⎠ ⎣ 2 ⎦ ⎝ 2 ⎠
So
∞
∑Al Rl Pl (cos θ ) = V0P2(cos θ ).
l =0

This means we only have the l = 2 term,

V0
A2 R2P2(cos θ ) = V0P2(cos θ ) → A2 = .
R2
Therefore,
⎛ r ⎞2 ⎛ 3 cos2 θ − 1 ⎞
V (r , θ ) = A2 r 2P2(cos θ ) = V0⎜ ⎟ ⎜ ⎟.
⎝R⎠ ⎝ 2 ⎠

3-18
 Electromagnetism

Outside:
Here, we must have Al = 0 so the potential does not blow up as r → ∞. So our
potential becomes
∞
B
V (r , θ ) = ∑ r l +l 1 Pl (cos θ ).
l =0

At the surface, we have

∞
Bl ⎛ 3 ⎞
V (R , θ ) = ∑ P (cos θ ) = V0⎜1 − sin2 θ ⎟ .
l +1 l
l =0
R ⎝ 2 ⎠

Again, the right-hand side is V0P2(cos θ ), so

B2
P2(cos θ ) = V0P2(cos θ ) → B2 = V0R3.
R2 +1
Therefore,

B2 ⎛ R ⎞3⎛ 3 cos2 θ − 1 ⎞
V (r , θ ) = P (cos θ ) = V ⎜ ⎟ ⎜ ⎟.
r3
2 0
⎝r⎠⎝ 2 ⎠

Problem 3.9. A spherical shell of radius R has surface charge σ0(θ ) = sin θ sin 3θ
smeared on its surface. Find the potential inside and outside the sphere.
Solution Our general solution is given by
∞
⎛ Bl ⎞
V (r , θ ) = ∑⎜⎝Al r l + ⎟Pl (cos θ ).
r l +1 ⎠
l =0

Inside we must have Bl = 0, otherwise V → ∞ as r → 0. So our potential becomes

∞
Vin(r , θ ) = ∑Al r l Pl (cos θ ).
l =0

Outside we must have Al = 0, otherwise V → ∞ as r → ∞. So our potential

becomes
∞
B
Vout(r , θ ) = ∑ r l +l 1 Pl (cos θ ).
l =0

At the surface they must be equal, so

Vin(R , θ ) = Vout(R , θ )
∞ ∞
B
∑Al Rl Pl (cos θ ) = ∑ Rl +l 1 Pl (cos θ )
l=0 l=0

Bl = Al R2l +1.

3-19
 Electromagnetism

We must also have

⎛ ∂Vout ∂Vin ⎞ 1
⎜ − ⎟ = − σ0(θ ),
⎝ ∂r ⎠
∂r r=R ϵ0

where
∞
∂Vout B
∂r
= ∑ − (l + 1) r l +l2 Pl (cos θ )
l=0

and
∞
∂Vin
∂r
= ∑lAl r l −1Pl (cos θ ).
l =0

Thus,
∞
⎛ ∂Vout ∂Vin ⎞ B
⎜
⎝ ∂r
− ⎟
∂r ⎠ r=R
= ∑ − (l + 1) Rl +l 2 Pl (cos θ ) − lAl Rl −1Pl (cos θ ).
l =0

Substitution of Bl yields
∞
⎛ ∂Vout ∂Vin ⎞ 1
⎜
⎝ ∂r
− ⎟
∂r ⎠ r=R
= ∑(2l + 1)Al Rl −1Pl (cos θ ) = σ0(θ ).
l =0
ϵ0

Since Legendre polynomials are orthogonal, when l ≠ l ′ we have

π
2
∫ Pl (cos θ )Pl ′(cos θ )sin θ dθ =
2l + 1
.
0

It follows that
π
1
Al =
2ϵ0Rl −1
∫ σ0(θ )Pl (cos θ )sin θ dθ .
0

Note that σ0(θ ) can be rewritten as

σ0(θ ) = sin θ sin 3θ = sin θ (sin 2θ cos θ + cos 2θ sin θ )

= sin θ ⎡⎣ (2 sin θ cos θ )cos θ + sin θ (2 cos2 θ − 1)⎤⎦

= 2 sin2 θ cos2 θ + sin2 θ (2 cos2 θ − 1)

= 2(1 − cos2 θ )cos2 θ + (1 − cos2 θ )(2 cos2 θ − 1)
= 2 cos2 θ − 2 cos4 θ + 2 cos2 θ − 2 cos4 θ − 1 + cos2 θ
= −4 cos4 θ + 5 cos2 θ − 1.

3-20
 Electromagnetism

We can find α , β , and γ such that

−4 cos4 θ + 5 cos2 θ − 1 = αP4(cos θ ) + βP2(cos θ ) + γP0(cos θ ).

So,
35 cos4 θ − 30 cos2 θ + 3 3 cos2 θ − 1
−4 cos4 θ + 5 cos2 θ − 1 = α +β + γ.
8 2
22
It follows that α = − 32
35
,β= 21
, 2
and γ = − 15 and we can now solve for Al .
π
1 ⎡ 32 22
Al =
2ϵ0Rl −1
∫ ⎢⎣ − P4(cos θ )Pl (cos θ )sin θ +
35 21
P2(cos θ )Pl (cos θ )sin θ
0
2 ⎤
− P0(cos θ )Pl (cos θ )sin θ ⎥dθ .
15 ⎦

If l = 4, we have
π
32 ⎛ 35 cos4 θ − 30 cos2 θ + 3 ⎞
2
1 32
A4 =
2ϵ0R 4−1
∫ − ⎜
35 ⎝ 8
⎟ sin θ dθ = −
⎠ 315ϵ0R3
.
0

If l = 2, we have
π
22 ⎛ 3 cos2 θ − 1 ⎞
2
1 22
A2 =
2ϵ0R2−1
∫ ⎜
21 ⎝ 2
⎟ sin θ dθ =
⎠ 105ϵ0R
.
0

If l = 0, we have
π
1 2 2R
A0 =
2ϵ0R 0−1
∫ −
15
sin θ dθ = −
15ϵ0
.
0

We can now find B4 , B2 , and B0 ,

32R6
B4 = A4 R2(4)+1 = −
315ϵ0
22R 4
B2 = A2 R2(2)+1 =
105ϵ0
2R2
B0 = A0 R2(0)+1 = − .
15ϵ0

Therefore, inside we have

Vin(r , θ ) = A0 + A2 r 2P2(cos θ ) + A4 r 4P4(cos θ ).

3-21
 Electromagnetism

So
R⎡ 2 22 ⎛⎜ r ⎞⎟2 ⎛ 3 cos2 θ − 1 ⎞
Vin(r , θ ) = ⎢− + ⎜ ⎟
ϵ0 ⎣ 15 105ϵ0 ⎝ R ⎠ ⎝ 2 ⎠

32 ⎛⎜ r ⎞⎟4⎛ 35 cos4 θ − 30 cos2 θ + 3 ⎞⎤

− ⎜ ⎟⎥
315ϵ0 ⎝ R ⎠ ⎝ 8 ⎠⎦
and outside we have
B0 B B
Vout(r , θ ) = 1
+ 2+21 P2(cos θ ) + 4+41 P4(cos θ ).
r r r
So
Vout(r , θ )
R2 ⎡ 2 22 ⎛ R ⎞2 ⎛ 3 cos2 θ − 1 ⎞ 32 ⎛ R ⎞4⎛ 35 cos4 θ − 30 cos2 θ + 3 ⎞⎤
= ⎢− + ⎜ ⎟ ⎜ ⎟− ⎜ ⎟ ⎜ ⎟⎥ .
ϵ0r ⎣ 15 105 ⎝ r ⎠ ⎝ 2 ⎠ 315 ⎝ r ⎠ ⎝ 8 ⎠⎦

Problem 3.10. An infinitely long cylindrical shell of radius R is held at a potential

V0(ϕ ) = α cos(4ϕ ). Find the potential inside and outside the shell.
Solution Our general solution in cylindrical coordinates is given by
∞ ⎧
V (s , ϕ) = a 0 + b0 ln(s ) + ∑⎨s k⎡⎣ ak cos(kϕ) + bk sin(kϕ)⎤⎦
⎪

k =1⎩
⎪

⎫
+ s −k⎡⎣ ck cos(kϕ) + dk sin(kϕ)⎤⎦ ⎬ .
⎪

⎭
⎪

Inside:
Here, we must have b0 = ck = dk = 0, otherwise the potential would blow up at the
center. Our potential becomes
∞
V (s , ϕ ) = a 0 + ∑s k⎡⎣ ak cos(kϕ) + bk sin(kϕ)⎤⎦.
k =1

At the surface, we have

∞
V (R , ϕ ) = a 0 + ∑ Rk⎡⎣ ak cos(kϕ) + bk sin(kϕ)⎤⎦ = α cos(4ϕ).
k=1

Note we have a 0 = 0, bk = 0, and ak = 0, except for a 4 . So

α
R 4a 4 cos(4ϕ) = α cos(4ϕ) → a 4 = 4 .
R
Therefore, for s ⩽ R , we have
⎛ s ⎞4
V (s , ϕ) = α⎜ ⎟ cos(4ϕ).
⎝R⎠

3-22
 Electromagnetism

Outside:
Here, we must have b0 = ak = bk = 0, otherwise the potential would blow up
as s → ∞. Also, since we must have V → 0 as s → ∞, a 0 = 0. Our potential
becomes
∞
V (s , ϕ ) = ∑s −k⎡⎣ ck cos(kϕ) + dk sin(kϕ)⎤⎦.
k =1

At the surface, we have

∞
V (R , ϕ ) = ∑R−k⎡⎣ ck cos(kϕ) + dk sin(kϕ)⎤⎦ = α cos(4ϕ).
k =1

Note we have dk = 0 and ck = 0, except for c4 . So

R−4c4 cos(4ϕ) = α cos 4ϕ → c4 = αR 4.
Therefore, for s ⩾ R , we have
⎛ R ⎞4
V (s , ϕ) = α⎜ ⎟ cos(4ϕ).
⎝s⎠

Problem 3.11. Given an infinitely long cylindrical shell of radius R and surface charge
σ0(ϕ ) = α cos(2ϕ ) + β sin(3ϕ ), find the potential inside and outside the cylinder.
Solution Our general solution in cylindrical coordinates is given by
∞ ⎧
∑⎨s k⎡⎣ ak cos(kϕ) + bk sin(kϕ)⎤⎦
⎪

V (s , ϕ) = a 0 + b0 ln(s ) +
k =1⎩
⎪

⎫
+ s −k⎡⎣ ck cos(kϕ) + dk sin(kϕ)⎤⎦ ⎬ .
⎪

⎭
⎪

Inside, we must have b0 = ck = dk = 0, otherwise the potential would blow up at the

center. Our potential becomes
∞
Vin(s , ϕ) = a 0 + ∑s k⎡⎣ ak cos(kϕ) + bk sin(kϕ)⎤⎦.
k =1

Outside we must have b0 = ak = bk = 0, otherwise the potential would blow up as

s → ∞. Also, since we must have V → 0 as s → ∞, a 0 = 0. Our potential becomes
∞
Vout(s , ϕ) = ∑s −k⎡⎣ ck cos(kϕ) + dk sin(kϕ)⎤⎦.
k =1

At the surface, we have

⎛ ∂Vout ∂Vin ⎞ 1
⎜ − ⎟ = − σ0(ϕ),
⎝ ∂s ∂s ⎠ s=R ϵ0

3-23
 Electromagnetism

where
∞
∂Vout
∂s
= ∑ −ks −k−1⎡⎣ ck cos(kϕ) + dk sin(kϕ)⎤⎦
k=1

and
∞
∂Vin
∂s
= ∑ks k−1⎡⎣ ak cos(kϕ) + bk sin(kϕ)⎤⎦.
k =1

Thus,
∞
⎛ ∂Vout ∂Vin ⎞
⎜
⎝ ∂s
− ⎟
∂s ⎠ s=R
= ∑ − kR−k−1⎡⎣ ck cos(kϕ) + dk sin(kϕ)⎤⎦
k =1

− kR k−1⎡⎣ ak cos(kϕ) + bk sin(kϕ)⎤⎦

α cos(2ϕ) + β sin(3ϕ)
=− .
ϵ0
From this, we can see that ck = ak = 0, except when k = 2, and dk = bk = 0, except
when k = 3. This means
α cos(2ϕ) β sin(3ϕ)
( ) (
2 cos(2ϕ) R−3c2 + Ra2 + 3 sin(3ϕ) R−4d3 + R2b3 = ) ϵ0
+
ϵ0
.

Separating out the sine and cosine term, we have

α ⎛ α ⎞
(
2 R−3c2 + Ra2 = ) ϵ0
→ c2 = R3⎜
⎝ 2ϵ 0
− Ra2⎟
⎠

and
β ⎛ β ⎞
3(R−4d3 + R2b3) = → d3 = R 4⎜ − R2b3⎟ .
ϵ0 ⎝ 3ϵ 0 ⎠
Since V is continuous, we have
Vout(R , ϕ) = Vin(R , ϕ)

R−2c2 cos(2ϕ) + R−3d3 sin(3ϕ) = a 0 + R2a2 cos(2ϕ) + R3b3 sin(3ϕ).

We can see that a 0 = 0. Also, considering the sine and cosine terms separately, we
have
R−2c2 = R2a2 .

3-24
 Electromagnetism

Substitution of c2 yields
⎛ α ⎞
R−2R3⎜ − Ra2⎟ = R2a2
⎝ 2ϵ 0 ⎠

so
α
a2 = .
4Rϵ0
Also
R−3d3 = R3b3.

Substitution of d3 yields
⎛ β ⎞
R−3R 4⎜ − R2b3⎟ = R3b3
⎝ 3ϵ0 ⎠
So
β
b3 = .
6R2ϵ0
Therefore,
⎛ α α ⎞ αR3
c2 = R3⎜ −R ⎟=
⎝ 2ϵ 0 4Rϵ0 ⎠ 4ϵ 0

and
⎛ β β ⎞ βR 4
d3 = R 4⎜ − R2 2 ⎟ = .
⎝ 3ϵ 0 6R ϵ 0 ⎠ 6ϵ0

Combining everything, the potential inside is

α β
Vin(s , ϕ) = s 2 cos(2ϕ) + s 3 2 sin(3ϕ)
4Rϵ0 6R ϵ0

R ⎡ α ⎛⎜ s ⎞⎟2 β ⎛ s ⎞3 ⎤
= ⎢ cos(2ϕ) + ⎜ ⎟ sin(3ϕ)⎥
ϵ0 ⎣ 4 ⎝ R ⎠ 6⎝R⎠ ⎦

and outside is

αR3 βR 4
Vout(s , ϕ) = s −2 cos(2ϕ) + s −3 sin(3ϕ)
4ϵ 0 6ϵ0

R ⎡ α ⎛ R ⎞2 β ⎛ R ⎞3 ⎤
= ⎢ ⎜ ⎟ cos(2ϕ) + ⎜ ⎟ sin(3ϕ)⎥ .
ϵ0 ⎣ 4 ⎝ s ⎠ 6⎝ s ⎠ ⎦

3-25
 Electromagnetism

Problem 3.12. The electric potential varies as 1 for a monopole, as 12 for a dipole, as
r r
1
3
for a quadrupole, and as 14 for an octopole. How will the electric potential depend
r r
on r for a mutipole with n charges (n being a k power of 2, n = 2k )?
Solution

Number of charges Potential

Monopole n = 20 = 1; k = 0 1 1
V∼ =
r k +1 r
Dipole n = 21 = 2; k = 1 1 1
V ∼ k +1 =
r r2
Quadrupole n = 22 = 4; k = 2 1 1
V ∼ k +1 =
r r3
Octopole n = 23 = 8; k = 3 1 1
V ∼ k +1 =
r r4
Multipole n = 2k ; k 1
V ∼ k +1
r

Problem 3.13. Let us consider an electric dipole with charges q and –q situated at
distance d from each other, shown below. Calculate the electric potential at a point P
in the far approximation r ≫ d .

Solution The total electric potential is obtained by superposition

1 ⎛q −q ⎞
V (r ⃗ ) = V1 + V2 = ⎜ + ⎟.
4πε0 ⎝ r1 r2 ⎠
From the law of cosines
⎛ d ⎞2 d
r 12 = ⎜ ⎟ + r 2 − 2 r cos θ
⎝2⎠ 2
and
⎛ d ⎞2 d
r 22 = ⎜ ⎟ + r 2 − 2 r cos(π − θ ).
⎝2⎠ 2

3-26
 Electromagnetism

Note that cos(π − θ ) = −cos θ . We can rewrite our potential as

⎛ ⎞
q ⎜ 1 1 ⎟
V= ⎜ − ⎟
4πε0 ⎜
+ r + rd cos θ ⎟⎠
d2 2 d2 2
⎝ 4
+ r − rd cos θ 4

or
⎛ ⎞
q ⎜ 1 1 ⎟
V= ⎜ − ⎟.
4πε0 ⎜
cos θ ⎟⎠
d2 d d2 d
⎝ 1+
r − cos θ r 1+ +
4r 2 r 4r 2 r

d2 d
When r ≫ d, is very small and can be ignored. If we consider x = cos θ ≪ 1, we
4r 2 r
can use the binomial theorem and obtain
1 x
(1 + x )− 2 ≅ 1 −
2
and
1 x
(1 − x )− 2 ≅ 1 + .
2
From this, we have
1 1⎛ d ⎞
= ⎜1 + cos θ ⎟
d r ⎝ 2r ⎠
r 1− r
cos θ

and
1 1⎛ d ⎞
= ⎜1 − cos θ ⎟ .
r 1+
d
cos θ r⎝ 2r ⎠
r

Therefore,
q ⎡1 ⎛ d ⎞ 1⎛ d ⎞⎤
V= ⎢ ⎜1 + cos θ ⎟ − ⎜1 − cos θ ⎟⎥
4πε0 ⎣ r ⎝ 2r ⎠ r⎝ 2r ⎠⎦

q ⎡ d d ⎤
= ⎢⎣1 + cos θ − 1 + cos θ ⎥
⎦
4πε0r 2r 2r

qd
V= cos θ .
4πε0r 2
JG JG
Taking the dipole moment p = qd , we have
JG
p · rˆ
V= .
4πε0r 2

3-27
 Electromagnetism

Problem 3.14. Find the electric field of the dipole in problem 3.13, centered at the
JG
origin with the dipole moment p in the z-direction.
Solution From problem 3.13, the electric potential is given by
JG
qd p · rˆ p cos θ
V (r ⃗ ) = 2
cos θ = 2
= .
4πε0r 4πε0r 4πε0r 2

We can find the field from the potential using

JG
E = −∇V.
We need to use the gradient in spherical coordinates
∂V 2p cos θ
Er = − =
∂r 4πε0r 3

1 ∂V p sin θ
Eθ = − =
r ∂θ 4πε0r 3
1 ∂V
Eϕ = − = 0.
r sin θ ∂ϕ

Therefore, the electric field due to the dipole is

JG p
Edipole(r , θ ) =
4πε0r 3
( )
2 cos θ rˆ + sin θ θˆ .

Problem 3.15. Two point charges +4q and −q are separated by a distance d. The
first charge is placed at (0, 0, d ) and the second one at the origin. Find: (a) the
monopole moment; (b) the dipole moment; (c) the electric potential in spherical
coordinates for r ≫ d . Include only the monopole and dipole contributions.

3-28
 Electromagnetism

Solution
(a) Monopole moment:
Q = 4q − q = 3q .

(b) Dipole moment:

2
JG
p = ∑qiri ⃗ = −q(0, 0, 0) + 4q(0, 0, d ) = 4qdzˆ .
i=1

(c) The electric potential:

1 ⎛1 1 ⎞ 1 ⎛ 3q 1 JG ⎞
V (r ⃗ ) = ⎜⎜ ∑qi + ∑qi ri ′ cos θi′ + ⋯⎟⎟ = ⎜ + 2 p · rˆ⎟
⎠ 4πε0 ⎝ r ⎠
2
4πε0 ⎝ r i r i r

1 ⎛ 3q p cos θ ⎞ 1 ⎛ 3q 4qd cos θ ⎞

= ⎜ + ⎟= ⎜ + ⎟.
4πε0 ⎝ r r 2 ⎠ 4πε0 ⎝ r r2 ⎠

Bibliography
Byron F W and Fuller R W 1992 Mathematics of Classical and Quantum Physics (New York:
Dover)
Griffiths D J 1999 Introduction to Electrodynamics 3rd edn (Englewood Cliffs, NJ: Prentice Hall)
Griffiths D J 2013 Introduction to Electrodynamics 4th edn (New York: Pearson)
Halliday D, Resnick R and Walker J 2010 Fundamentals of Physics 9th edn (New York: Wiley)
Halliday D, Resnick R and Walker J 2013 Fundamentals of Physics 10th edn (New York: Wiley)
Jackson J D 1998 Classical Electrodynamics 3rd edn (New York: Wiley)
Rogawski J 2011 Calculus: Early Transcendentals 2nd edn (San Francisco, CA: Freeman)

3-29
 IOP Concise Physics

Electromagnetism
Problems and solutions
Carolina C Ilie and Zachariah S Schrecengost

Chapter 4
Magnetostatics

This chapter introduces magnetic fields in a vacuum and the methods for calculating
the magnetic field. Magnetic fields are intrinsically determined by electric charges in
motion. We imagine these small currents as magnetic dipoles. From the general
Biot–Savart law, to the more straightforward Ampère’s law applicable to config-
urations with higher degree of symmetry, the suggested problems constitute good
practice in magnetostatics.

4.1 Theory
4.1.1 Magnetic force
A charge q moving with velocity υ⃗ in a magnetic field B ⃗ experiences a force given by

Fm⃗ = qυ ⃗ × B ⃗.

4.1.2 Force on a current carrying wire

The force on a current carrying wire in a magnetic field B ⃗ is

Fm⃗ = ∫ I ( dl⃗ × B ⃗).

4.1.3 Volume current density
The current density of a current I ⃗ is
dI ⃗
J⃗ =
da⊥
and the current density of a charge density ρ moving at velocity υ⃗ is
J ⃗ = ρυ ⃗ .

doi:10.1088/978-1-6817-4429-2ch4 4-1 ª Morgan & Claypool Publishers 2016

 Electromagnetism

4.1.4 Continuity equation

The divergence of the charge density J ⃗ is related to the charge density ρ by
∂ρ
∇ ⋅ J⃗ = − .
∂t
4.1.5 Biot–Savart law
The magnetic field due to current distributions is given by

μ0I dl′⃗ × rˆ
B ⃗( r ⃗ ) =
4π
∫ r2
μ0 K⃗ (r ′⃗ ) × rˆ
B ⃗( r ⃗ ) =
4π
∫ r2
da′

μ0 J ⃗(r ′⃗ ) × rˆ
B ⃗( r ⃗ ) =
4π
∫ r2
dτ ′.

4.1.6 Divergence of B ⃗
Given magnetic field B,⃗ we have

∇ ⋅ B ⃗ = 0.

4.1.7 Ampère’s law

Given magnetic field B,⃗ we have
∇ × B ⃗ = μ 0 J ⃗.
By applying Stoke’s law, we also have

∮S B ⃗ · dl⃗ = μ0Ienc,

where

Ienc = ∫ J ⃗ · da ⃗ .
4.1.8 Vector potential
The vector potential due to current distributions is given by
μI 1 ′⃗
A ⃗ (r ⃗ ) = 0
4π r
dl ∫
μ0 K⃗ (r ′⃗ )
A ⃗ (r ⃗ ) =
4π
∫ r
d a′

μ0 J (⃗ r ′⃗ )
A ⃗ (r ⃗ ) =
4π
∫ r
dτ ′.

4-2
 Electromagnetism

Also,

B ⃗ = ∇ × A⃗

and

∇2 A ⃗ = μ0J ⃗.

4.1.9 Magnetic dipole moment

The magnetic dipole moment due to a current I is

m⃗ = ∫ I da ⃗ .

4.1.10 Magnetic field due to dipole moment

Given magnetic dipole moment m⃗ , the magnetic field is
μ0m
⃗ =
Bdip
4πr 3
(2 cos θ rˆ + sin θ θˆ .)
4.2 Problems and solutions

Problem 4.1. A proton travels through a uniform magnetic and electric field. The
magnetic field is B ⃗ = ayˆ , where a is a positive constant. If at one moment the
velocity of the proton is υ⃗ = bzˆ , where b is a positive constant, what is the force
acting on the proton if the electric field is E ⃗ = −cxˆ ?
Solution

F ⃗ = q E ⃗ + υ ⃗ × B ⃗ = q⎡⎣ −cxˆ + (bzˆ ) × (ayˆ )⎤⎦ = q⎡⎣ − cxˆ + ab( −xˆ ) ⎤⎦ = (cxˆ + abxˆ )
( )
F ⃗ = −q(c + ab)xˆ .

Problem 4.2. A particle of charge q enters a region of uniform magnetic field B ⃗

(out of the page, in the z-direction) with an initial velocity υ⃗ (in the x-direction).
The particle is deflected a distance y above the initial direction. If the region has a
width of x , find the sign of the charge and the deflected distance y as a function of
q , υ, B , and x.

4-3
 Electromagnetism

Solution

F ⃗ = qυ ⃗ × B ⃗
Since the charge is deflected as shown, the charge is negative (determined from the
right-hand rule). In the x -direction we have no force, and therefore no acceleration
x = υt.
In the y -direction
ma = q υB sin 90.
So
q υB
a=
m
and
a yt 2 a yt 2 q υBt 2
y = yo + υoyt + = = .
2 2 2m
By substituting the time
x
t=
υ
we obtain
x2
q υB 2 2
y= υ 2 = q Bx = q Bx ,
2m 2mυ 2p
where p is the momentum of the particle.

Problem 4.3. The current density in a wire of circular cross section of radius R is
dependent on the distance from the axis, given by J ⃗ = ks 2zˆ , where k is a constant.
Find a) the total current in the wire and b) the current density if the current in a) is
uniformly distributed.

4-4
 Electromagnetism

Solution
a) Given current density J ⃗ = ks 2zˆ , the current is
2π R R
s4 πkR 4
I= ∫ J ⋅⃗ da ⃗ = ∫ ( ks zˆ) ⋅ ( s dϕ ds zˆ) = ∫
2
dϕ ∫ ks 3ds = 2πk
4
=
2
.
0 0 0

b) If this current was uniformly distributed, the current density is simply

I 1 πkR 4 kR2
J= = = .
area πR2 2 2

Problem 4.4.
q
a) In the famous experiment of J J Thompson, he measured the charge to mass
m
ratio of the catode rays. Find q when you know B, R , and υ, and that B ⃗ is
m
perpendicular to υ⃗ .
b) He also had the beams going in a region with perpendicular electric field and
magnetic field and ‘tuned’ them such that the electrons left the region with
unchanged direction. If the speed of the electrons is υ and the magnetic field is B,⃗
what should be the value of the electric field?
Solution
a) The magnitude of the magnetic force is given by
Fm⃗ = q υ ⃗ × B ⃗,

where we have υ⃗ ⊥ B ⃗ . Also

Fm⃗ = Fcentripetal
⃗ .
So,
Fm = q υB

and
mυ 2
Fcent = .
R
Therefore from
mυ 2
q υB =
R
we have
q υ
= .
m BR

4-5
 Electromagnetism

b) Setting the magnetic and electric forces as equal, we have

Fm⃗ = Fe⃗ → qυ ⃗ × B ⃗ = qE ⃗ .
Dividing by q and expressing this in terms of magnitudes, we have

υB sin 90 = E
so
E = υB.

Problem 4.5. Find the magnetic field at:

a) The center of a circular wire loop of radius R carrying current I.

b) The center of a wire loop that consists of half a loop of radius R and half a
square loop of side 2R , carrying current I.

Solution
a) The Biot–Savart law states

4-6
 Electromagnetism

∫ dlr×2 r ,
μ0I ⃗ ˆ
B⃗ =
4π
where
dl⃗ = R dϕ ⃗ = R dϕ ϕˆ
and
r̂ = −r̂.
Since r = R , the Biot–Savart law becomes
2π
μ0IR μI
B⃗ = ∫ ϕˆ × rˆ dϕ = 0 zˆ (out of page).
4πR2 2R
0

b) From part a), we can determine the field contribution due to the circular part is
μ0I
Bc⃗ = zˆ ,
4R
which is half that of the full loop. As for the square, we consider the field R above
the wire. We have

So

∫ dl r×2 r ,
μ0I ′⃗ ˆ
B⃗ =
4π

where dl′⃗ × rˆ points in the ẑ -direction (out of the page). Also,

dl′ sin γ = dl′ cos θ

and
R
l′ = R tan θ → dl′ = dθ
cos2 θ

4-7
 Electromagnetism

and
1 cos2 θ
r 2 = l′2 + R2 → = .
r 2
R2
Therefore,
θ2
μI ⎛ cos2 θ ⎞⎛ R ⎞ μ0I
B⃗ = 0
4π
∫ ⎜ ⎟⎜
⎝ R ⎠⎝ cos θ ⎠
2 2
⎟cos θ dθ =
4πR
(sin θ2 − sin θ1).
θ1
π
So for each R-lengthed segment (i) and (ii), we have θ1 = 0 and θ2 = 4 , and for
π π
the 2R -lengthed segment, we have θ1 = − 4 and θ2 = 4 . So

⎛μI π ⎞ μ I ⎡ ⎛π ⎞ ⎛ π ⎞⎤
B ⃗ = Bc⃗ + 2⎜ 0 sin zˆ⎟ + 0 ⎢sin⎜ ⎟ − sin⎜ − ⎟⎥zˆ
⎝ 4πR 4 ⎠ 4πR ⎣ ⎝ 4 ⎠ ⎝ 4 ⎠⎦
μ0I
=
4πR
(
π +2 2 zˆ ) (out of page).

Problem 4.6. Consider a cylindrical shell of radius R and length L , carrying σ and
rotating at ω. Find the magnetic field d from the end of the shell (on the axis).

Solution Here we have

∫ K r×2 r da,
μ0 ⃗ ˆ
B⃗ =
4π
where
da = R dz dϕ

and 0 ⩽ z ⩽ L . The surface charge is given by

K⃗ = συ ⃗ = σωRϕˆ .

4-8
 Electromagnetism

From the figure above, we have l′ = L − z and

r= (l′ + d )2 + R2 = (L − z + d )2 + R 2 .

Note that the field cancels such that the ẑ -component is the only component that
survives. So

2
⎡ K⃗ × rˆ ⎤ = σωR sin θ zˆ = σωR R zˆ = σωR zˆ .
⎣ ⎦z
r r

Putting everything together, we have

2π L
μ0 σ ωR 2 R
B⃗ =
4π
∫ ∫ ⎡ (L − z + d )2 + R2⎤3/2
zˆ dz dϕ
0 0 ⎣ ⎦

L
μ0σωR3zˆ dz
=
2
∫ ⎡ (L − z + d )2 + R2⎤3/2
.
0 ⎣ ⎦

Therefore,

μ0σωR ⎡⎢ d+L d
⎤
⎥zˆ .
B⃗ = –
2 ⎢⎣ R2 + (d + L )2 R2 + d 2 ⎥⎦

Problem 4.7. A hemisphere of radius R and charge density ρ is rotating at ω. Find

the magnetic field d above the center.

4-9
 Electromagnetism

Solution Here we have

∫ J r×2 r dτ,
μ0 ⃗ ˆ
B⃗ =
4π
where
dτ = r 2 sin θ dr dϕ dθ .
From the figure below, we can see that
r 2 = d 2 + r 2 − 2dr cos θ .

Also,
J ⃗ = ρυ ⃗ = ρωr sin θ ϕˆ .

4-10
 Electromagnetism

Note that the field cancels such that the ẑ -component is the only component that
survives. So

⎡ J ⃗ × rˆ ⎤ = ρωr sin θ ⎛⎜ r sin θ ⎞⎟zˆ .

⎣ ⎦z ⎝ r ⎠
Therefore
2π π R
μ ρωr 2 sin2 θ r 2 sin θ zˆ
B⃗ = 0
4π
∫ ∫ ∫ 3
dr dθ dϕ
0 π
2
0 (d 2 2
+ r − 2dr cos θ ) 2

π R
μ ρω r 4 sin3 θ zˆ
= 0
2
∫ ∫ 3
dr dθ
π
2
0 (d 2 2
+ r − 2dr cos θ ) 2

μ0ρω ⎡
B⃗ = R2 + d 2 −2R 4 + d 2R2 − 12d 4 + 2R5 + 5d 3R2 + 12d 5⎤⎦zˆ .
( )
30d 3 ⎣

Problem 4.8. A spherical shell of radius R , carrying σ and rotating at ω, is centered

at the origin. Find the velocity a loop of wire, carrying λ with radius a centered at the
origin, required to cancel the magnetic field at the center.

Solution First, we have

∫ K r×2 r da,
μ0 ⃗ ˆ
B⃗ =
4π
where
r=R

4-11
 Electromagnetism

r̂ = −r̂

K⃗ = συ ⃗ = σωR sin θ ϕˆ
da = R2 sin θ dθ dϕ
and
K⃗ × rˆ = σωR sin2 θ zˆ .
Putting this together, we have
π 2π π
μ0 σωR sin2 θ R2 sin θ zˆ μ σωR zˆ
Bs⃗ =
4π
∫ ∫ R 2
dϕ dθ = 0
2
∫ sin3 θ dθ
0 0 0

2μ0σωR
Bs⃗ = zˆ .
3
Now a line of charge λ rotating at υ⃗ ‘looks’ like a wire carrying current I ⃗ = λυ ⃗=
λωl aϕ̂ . From problem 4.5(a), we know this produces magnetic field
μ0λωl a μ λωl
Bl⃗ = zˆ = 0 zˆ .
2a 2
We want Bl⃗ + Bs⃗ = 0, so
μ0λωl 2μ σωR λωl 2σωR
+ 0 =0→ =− .
2 3 2 3
Therefore,
4ωσR
ωl = − .
3λ

Problem 4.9. A long straight wire carries a steady current I. Obtain the magnetic
field at a distance s from the wire.

4-12
 Electromagnetism

Solution Here, we apply Ampère’s law, with an Amperian loop that is a circle
centered on the wire and in a plane perpendicular to the wire. Ampère’s law is

∮S B ⋅⃗ dl⃗ = μ0Ienc.
Noticing B ⃗ dl⃗ , it follows that B is a constant at a certain distance s from the wire.
So the left-hand side is given by

∮S B ⋅⃗ dl⃗ = ∮S B dl = B ∮S dl = B 2πs.
The enclosed current is just simply given by
Ienc = I
So
μ0I
B= .
2πs
Since the magnetic field is tangent on the circle at every point
μ0I ˆ
B⃗ = ϕ.
2πs

Problem 4.10. An electric current flows through a long cylinder wire of radius a .
Find the magnetic field inside and outside the wire, and plot it, in the following
cases, where k is a constant with the appropriate units:
a) I = constant (steady current).
b) Current density J is proportional to the distance from the axis: J = ks .
c) J = ks 2 .
Solution
a) Here we have constant I . For s > a , our Amperian loop is given by

4-13
 Electromagnetism

Ampère’s law states

∮S B ⋅⃗ dl⃗ = μ0Ienc
In all cases, the left-hand side yields

∮S B ⋅⃗ dl⃗ = ∮S B dl = B ∮S dl = B 2πs.
Here we simply have
Ienc = I
So
μ0I
B 2πs = μ0I → B = .
2πs
Applying the right-hand rule to a current coming out of the page, we have
μ0I ˆ
B⃗ = ϕ.
2πs
For s < a , our Amperian loop is inside the wire, at radius s .

Again we have

∮S B ⋅⃗ dl⃗ = μ0Ienc.
Since I is uniform, the current density is constant,
I I
J= 2
= enc2
πa πs
so
s2
Ienc = I .
a2

4-14
 Electromagnetism

Therefore
μ0sI μ sI
B= 2
→ B ⃗ = 0 2 ϕˆ .
2πa 2πa
The plot of the magnetic field is given below.

b) Here we have J = ks . For s > a , the Amperian loop is the same as part a) for
s > a . We still have the left hand side of Ampère’s law given by

∮S B ⋅⃗ dl⃗ = B 2πs.
Now, since J is not constant, we need to integrate in order to find the enclosed
current, so
a 2π a 2π a
ks 3 2πka 3
Ienc = ∫ J da = ∫ Js ds ∫ dϕ = ∫ ks ds2
∫ d ϕ = 2π
3
=
3
.
0 0 0 0 0

Therefore,
μ02πka 3 μ ka 3
B 2πs = →B= 0
3 3s
with
μ0ka 3 ˆ
B⃗ = ϕ.
3s
For s < a , we again have the same Amperian loop as part a). Here, our enclosed
current is given by
s 2π s
ks′3 2πks 3
Ienc = ∫ J da = ∫ ks′ s′ ds′ ∫ d ϕ = 2π
3
=
3
.
0 0 0

4-15
 Electromagnetism

Therefore,
μ02πks 3 μ ks 2
B 2πs = →B= 0
3 3
with
μ0ks 2 ˆ
B⃗ = ϕ.
3
The plot of the magnetic field is given below.

c) Now we have J = ks 2 . For s > a , the enclosed current is given by

a 2π a a
ks 4 ka 4
Ienc = ∫ J da = ∫ Js ds ∫ d ϕ = 2π ∫ ks 2 s ds = 2π
4
= 2π
4
.
0 0 0 0

Therefore,

ka 4 μ ka 4
B 2πs = μ02π →B= 0
4 4s
with
μ0ka 4 ˆ
B⃗ = ϕ.
4s
For s < a , we have
s 2π s s
2πks′4 2πks 4
Ienc = ∫ J da = ∫ Js′ ds′ ∫ dϕ = 2π ∫ ks′2 s′ ds′′ =
4
=
4
.
0 0 0 0

Therefore,
μ0ks 3
B=
4

4-16
 Electromagnetism

with

μ0ks 3 ˆ
B⃗ = ϕ.
4

The plot of magnetic field is given below.

Problem 4.11. Use Ampère’s law to obtain the magnetic field inside and outside a
N
solenoid of n = L , where N is the number of turns, and L is the length of the
solenoid. The solenoid is carrying the current I .
Solution Let us choose two Amperian loops given by

Starting with the outside loop (loop A), the magnetic field does not have any
radial Br component or Bϕ . Ampère’s law is given by

∮S B ⋅⃗ dl⃗ = μ0Ienc.

4-17
 Electromagnetism

Since we have Ienc = 0, B (c ) = B (d ), but since B → 0 for large distances, B = 0

outside the solenoid. For loop B, the left hand side of Ampère’s law is given
by

∮S B ⋅⃗ dl⃗ = ∮S B dl = BL.
The sides perpendicular on the solenoid yield zero dot product, as the magnetic field
is oriented parallel to the solenoid’s axis in the z -direction (by the right-hand rule).
The enclosed current is given by

Ienc = InL = IN.

Substituting these quantities into Ampère’s law yields

μ0IN
BL = μ0InL → B = μ0In =
L

with

μ0IN
B ⃗ = μ0Inzˆ = z.
ˆ
L

Problem 4.12. A current carrying empty cylinder of inner radius a and outer radius
b has a current density J , which is proportional to the distance from the axis;
J = ks, k constant. Find the magnetic field in all regions.

Solution There are three significant regions: s < a , a < s < b, and s > b. The easiest
to find is the field for s < a ,

4-18
 Electromagnetism

where the enclosed current is zero. Therefore,

∮S B ⋅⃗ dl⃗ = 0,
so B ⃗ = 0. For a < s < b, we have

Using Ampère’s law,

∮S B ⋅⃗ dl⃗ = μ0Ienc.
We have the left-hand side is given by

∮S B ⋅⃗ dl⃗ = B 2πs,
with enclosed current given by
s 2π s
2πks′3
s
(
2πk s 3 − a 3 )
Ienc = ∫ J ⃗ · da ⃗ = ∫ Js′ ds′ ∫ d ϕ = 2π ∫ ks′ s′ ds′ =
3
=
3
a 0 a a

4-19
 Electromagnetism

B=
(
μok s 3 − a 3 )
3s
with

B⃗ =
(
μok s 3 − a 3 ) ϕˆ .
3s
For s > b, we have

Again,

∮S B ⋅⃗ dl⃗ = B 2πs
with
b 2π b
ks 3
b
(
2πk b3 − a 3 ).
Ienc = ∫ J ⃗ ⋅ da ⃗ = ∫ Js ds ∫ dϕ = 2π ∫ ks s ds = 2π
3
=
3
a 0 a a

Therefore,

B=
(
k b3 − a 3 )
3s
with

B⃗ =
(
k b3 − a 3 ) ϕˆ .
3s

Problem 4.13. Find the vector potential d above a spinning disk of radius R, with
angular velocity ω and carrying σ .

4-20
 Electromagnetism

Solution We have
μ0 ⃗
A⃗ =
4π
∫ Kr da,
where
K⃗ = συ ⃗ = σωrϕˆ .
From

we also have
r= r2 + d 2
and
da = 2πr dr .
So
R R
μ σωr 2πrϕˆ μ σωϕˆ r2
A⃗ = 0
4π
∫ r2 + d 2
dr = 0
2
∫ r2 + d 2
dr
0 0
μ σω ⎡ ⎤
= 0 ⎣⎢R R2 + d 2 − d 2 ln R +
4
( )
R2 + d 2 + d 2 ln d ⎦⎥ϕˆ

4-21
 Electromagnetism

μ0σω ⎡ ⎛ d ⎞⎤
A⃗ = ⎢R R2 + d 2 + ln⎜ ⎟⎥ϕˆ .
4 ⎢⎣ ⎝R + R2 + d 2 ⎠⎥⎦

Problem 4.14. What current density produces vector potential A⃗ = sin ϕ zˆ ?

Solution First, check ∇ ⋅ A⃗ = 0:
∂
∇ ⋅ A ⃗ = ∇ ⋅ (sin ϕ zˆ ) = (sin ϕ) = 0.
∂z

Now, B ⃗ = ∇ × A⃗ and ∇ × B ⃗ = μ0 J ,⃗ so

1⎡ ∂ ⎤ cos ϕ
B ⃗ = ∇ × A ⃗ = ∇ × (sin ϕ zˆ ) = ⎢ (sin ϕ)⎥sˆ = sˆ .
s ⎣ ∂ϕ ⎦ s

Also,
⎛ cos ϕ ⎞ 1 ⎡ ∂ ⎛ cos ϕ ⎞⎤ sin ϕ
μ0J ⃗ = ∇ × B ⃗ = ∇ × ⎜ sˆ⎟ = ⎢ − ⎜ ⎟⎥zˆ= zˆ .
⎝ s ⎠ s ⎣ ∂ϕ ⎝ s ⎠⎦ s2

Therefore,
sin ϕ
J⃗ = zˆ .
μ0s 2

This can be checked using the product rule

( ) (
∇ × ∇ × A ⃗ = μ0J ⃗ = ∇ ∇ ⋅ A ⃗ − ∇2 A ⃗ . )
Since ∇ ⋅ A⃗ = 0, ∇2 A⃗ = μ0 J ⃗ where
1 ∂2 1 sin ϕ
−∇2 A ⃗ = − 2
(sin ϕ)zˆ = − 2 ( −sin ϕ)zˆ = zˆ .
s ∂ϕ s s2

So
1 sin ϕ
J⃗ =
μ0
(
−∇2 A ⃗ = )
μ0s 2
zˆ

as expected.

Problem 4.15. Find the vector potential inside and outside a wire of radius R that is
carrying current density J ⃗ = kszˆ , where k is a constant.

4-22
 Electromagnetism

Solution We can find the field inside by

∮S B ⃗ ⋅ dl⃗ = μ0Ienc,
where

∮S B ⃗ ⋅ dl⃗ = B 2πs
and
s
2πks 3
Ienc = ∫ J da = ∫ ks′2πs′ ds′ =
3
.
0

So,
3
∮S B ⃗ ⋅ dl⃗ = B 2πs = μ0 2π3ks
and
μ0ks 2 ˆ
B⃗ = ϕ.
3
We must have that A⃗ depends only on s and is in the direction of the current. So
A⃗ = A(s )zˆ and ∇ × A⃗ = B ⃗ . Note
∂A ˆ μ ks 2
∇ × A⃗ = − ϕ = B ⃗ = 0 ϕˆ .
∂s 3
Therefore,
μ0k 2
dA = − s
3
and
⎛ μ ks 3 ⎞
A⃗ = ⎜− 0 + C ⎟zˆ .
⎝ 9 ⎠

We will express this as

μ0k 3
A⃗ = −
9
(
s − α 3 zˆ . )
Outside, our total current is
3
Itot = Ienc = ∫ J da = 2πkR
3
.

4-23
 Electromagnetism

From

∮S B ⋅⃗ dl⃗ = μ0Ienc
we have
2πkR3
B 2πs = μ0
3
and
μ0kR3 ˆ
B⃗ = ϕ.
3s
Again,
μ0kR3
dA = − B ds = − ds
3s
so
⎛ μ kR3 ⎞
A⃗ = ⎜− 0 ln s + C ⎟zˆ .
⎝ 3 ⎠

We will express this as

μ0kR3 ⎛ s ⎞
A⃗ = − ⎜ln ⎟zˆ .
3 ⎝ β⎠

Since A⃗ is continuous at R ,

μ0k 3 μ kR3 ⎛ s ⎞
−
9
(
s − α3 = − 0 ) ⎜ln ⎟
3 ⎝ β⎠

we have
R
R3 − α 3 = 3R3 ln
β

with
⎛ R⎞
R3⎜1 − 3 ln ⎟ = α 3
⎝ β⎠

and
R α3
1 − 3 ln = 3.
β R

4-24
 Electromagnetism

If α = β = R,
⎛ R ⎞ R3
1 − 3 ln⎜ ⎟ = 3 → 1 = 1.
⎝R⎠ R
So
⎧ μ0k
⎪ 3 3
(
⎪− 9 s − R zˆ s < R )
A⃗ = ⎨ 3
.
⎪ μ0kR s
⎪ − ln zˆ s > R
⎩ 3 R

Problem 4.16. A disk of radius R is carrying surface charge σ = kr , where k is a

constant, and spinning at angular velocity ω. Find the magnetic dipole moment and
the field it produces.
Solution We have
K⃗ = συ ⃗ = σωrϕˆ = ωkr 2ϕˆ .
So
ωkr 3
dI = ωkr 2dr → I = .
3
Therefore,
R R
ωk 2πωk
m⃗ = ∫ I da ⃗ =
3
∫ 3
r 2πr dr zˆ =
3
∫ r 4 dr
0 0

2πωkR5
m⃗ = zˆ .
15
We have, in spherical coordinates,
μ0m
⃗ =
Bdip
4πr 3
(
2 cos θ rˆ + sin θ θˆ , )
which can be expressed in cylindrical coordinates by considering r = s2 + z2 ,
z s z s s z
cos θ = r , sin θ = r , θ̂ = r sˆ − r zˆ , and rˆ = r sˆ + r zˆ . Therefore,
μ0m ⎡ ⎛ z ⎞⎛ s z ⎞⎟ ⎛⎜ s ⎞⎟⎜⎛ z s ⎞⎟⎤
⃗ =
Bdip ⎢2⎜ ⎟⎜ s ˆ + z
ˆ + s
ˆ − zˆ ⎥
4πr 3 ⎣ ⎝ r ⎠⎝ r r ⎠ ⎝ r ⎠⎝ r r ⎠⎦
μm
(
= 0 5 2zssˆ + 2z 2zˆ + szsˆ − s 2zˆ
4πr
)

4-25
 Electromagnetism

μ0m ⎡
= 3zssˆ + 2z 2 − s 2 zˆ⎤⎦
( )
4πr 5 ⎣

μ0m 1 ⎡ 3zssˆ + 2z 2 − s 2 zˆ⎤ .

⃗ =
Bdip 5/2 ⎣ ( ) ⎦
(
4π s 2 + z 2 )
Substitution of m yields,
μ0ωkR5 1 ⎡ 3zssˆ + 2z 2 − s 2 zˆ⎤ .
⃗ =
Bdip 5/2 ⎣ ( )
⎦
30 (
s2 + z2 )

Bibliography
Griffiths D J 1999 Introduction to Electrodynamics 3rd edn (Englewood Cliffs, NJ: Prentice Hall)
Griffiths D J 2013 Introduction to Electrodynamics 4th edn (New York: Pearson)
Halliday D, Resnick R and Walker J 2010 Fundamentals of Physics 9th edn (New York: Wiley)
Halliday D, Resnick R and Walker J 2013 Fundamentals of Physics 10th edn (New York: Wiley)

4-26
 IOP Concise Physics

Electromagnetism
Problems and solutions
Carolina C Ilie and Zachariah S Schrecengost

Chapter 5
Electric fields in matter

Now we will address problems that deal with electric fields in matter, looking at
problems involving dipole moments, media polarization, and electric displace-
ment. Ideas developed in chapters 2 and 3 will be revisited and expanded upon in
this chapter. Gauss’s law is reformulated for electric displacement and various
ways to calculate the energy of a configuration. Some of the techniques practiced
in chapter 3 will be applied now, including the Laplace equation and Legendre
polynomials.

5.1 Theory
5.1.1 Induced dipole moment of an atom in an electric field
Given an atom with polarizability α in an electric field E ⃗ , the induced dipole
moment is

p ⃗ = αE ⃗ .

5.1.2 Torque on a dipole due to an electric field

Given a dipole moment p ⃗ in an electric field E ⃗ , the torque on the dipole is

N⃗ = p ⃗ × E ⃗ .

5.1.3 Force on a dipole

Given a dipole moment p ⃗ in an electric field E ⃗ , the force on the dipole is

( )
F ⃗ = p ⃗ ⋅ ∇ E ⃗.

doi:10.1088/978-1-6817-4429-2ch5 5-1 ª Morgan & Claypool Publishers 2016

 Electromagnetism

5.1.4 Energy of a dipole in an electric field

Given a dipole moment p ⃗ in an electric field E ⃗ , the energy of the dipole is

U = −p ⃗ ⋅ E ⃗.
5.1.5 Surface bound charge due to polarization P ⃗
Given polarization P ⃗ and normal vector n̂ , the surface bound charge is

σ b = P ⃗ ⋅ n.
ˆ

5.1.6 Volume bound charge due to polarization P ⃗

Given polarization P,⃗ the volume bound charge is

ρb = −∇ ⋅ P.⃗

5.1.7 Potential due to polarization P ⃗

Given a volume V , the potential due to polarization P ⃗(r ⃗ ) is

∫V r
1 ˆ ⋅ P (⃗ r ′⃗ )
V (r ⃗ ) = dτ ′.
4πε0 r2

5.1.8. Electric displacement

Given polarization P ⃗ and electric field E ⃗ , the electric displacement is given by

D⃗ = ε0E ⃗ + P.⃗

5.1.9 Gauss’s law for electric displacement

Considering electric displacement D⃗ and free charge density ρf , Gauss’s law can be
written in differential form as
∇ ⋅ D⃗ = ρf
and in integral form as

∮S D⃗ ⋅ da ⃗ = qf enc
,

where qfenc is the total free charge enclosed in the volume.

5.1.10 Linear dielectrics

Given a medium with electric susceptibility χe , the polarization is given by
P ⃗ = ε0 χe E ⃗ ,

5-2
 Electromagnetism

where E ⃗ is the total electric field. The electric displacement is now

( ) ( )
D⃗ = ε0E ⃗ + P ⃗ = ε0 + ε0 χe E ⃗ = ε0 1 + χe E ⃗ = ε0εrE ⃗ = εE ⃗ ,

where ε is the permittivity of the material and εr is the relative permittivity of the
material. Also, the boundary conditions are now
⊥ ⊥
εaboveEabove − εbelowE below = σf
And
∂Vabove ∂V
εabove − εbelow below = −σf
∂n ∂n
while we still maintain
Vabove = Vbelow.

5.1.11 Energy in a dielectric system

Given electric field E ⃗ and electric displacement D⃗ , the energy in a dielectric system is

W=
ε0
2
∫ εrE 2dτ = 12 ∫ D⃗ ⋅ E ⃗dτ.

5.2 Problems and solutions

Problem 5.1. Given p1⃗ and p2⃗ below, find where to place point charge q such that
there is no net torque on p2⃗ . Assume the center of p1⃗ is the origin and express your
answer in spherical coordinates.

Solution The field at p2⃗ is given by

p1 ˆ p1
⃗ =
Edip θ=− zˆ .
3
4πε0r 4πε0r 3

5-3
 Electromagnetism

Since the field is straight down, we must place the point charge q below it to cancel
the field (thus resulting in zero torque on p2⃗ ). We must place q at a distance d from
the dipole so that
p1 q
Edip + Eq = − + = 0.
4πε0r 3 4πε0d 2
Solving for d 2 yields
qr 3
d2 = .
p1
Therefore we have

with

qr 3 ⎛ qr ⎞
rq2 = + r 2 = r 2⎜ + 1⎟
p1 ⎝ p1 ⎠

so
qr
rq = r + 1.
p1

From
r
cos φ =
rq

we have
⎛ ⎞
⎜ ⎟
−1⎜ 1 ⎟.
φ = cos
⎜ qr ⎟
⎜ p +1⎟
⎝ 1 ⎠

5-4
 Electromagnetism

So the spherical coordinates of q are

⎛ ⎛ ⎞⎞
⎜ ⎜ ⎟⎟
(r , θ ) = ⎜⎜r ⎟⎟ .
qr 1
+ 1 , + cos−1⎜
π
p1 2 ⎜ qr ⎟⎟
⎜ ⎜ p + 1 ⎟⎟
⎝ ⎝ 1 ⎠⎠

Problem 5.2. Consider a neutral atom, with polarizability α , located z above a disk
of radius R carrying surface charge σ . Find the force of attraction between the atom
and the plate.

Solution The field at z is given by

R
1 σ σ 2πrz
E⃗ = ∫ rˆ da = ∫ zˆ dr
4πε0 r 2
4πε0
(z 2
+ r2
3/2
)
0

zσ ⎛ 1 1 ⎞
E⃗ = ⎜ − ⎟zˆ .
2ε0 ⎝ z R2 + z 2 ⎠
This induces a dipole
zσα ⎛ 1 1⎞
p ⃗ = αE ⃗ = ⎜ − ⎟zˆ .
2ε0 ⎝ z R2 + z 2 ⎠
The electric field due to the dipole is given by
⃗ = p
Edip (
2 cos θ rˆ + sin θ θˆ )
4πε0r 3
zσα ⎛ 1 1 ⎞
= ⎜ −
8πε02r 3 ⎝ z
(
⎟ 2 cos θ rˆ + sin θ θˆ .
R2 + z 2 ⎠
)

5-5
 Electromagnetism

The force on a piece of charge dq is given by

dF ⃗ = E ⃗ dq

so

F⃗ = ∫ dF ⃗ = ∫ E ⃗dq = ∫ E ⃗σ dA = ∫ E ⃗σ 2π l dl.
Consider the following

with side view

In our expression fix alignment for E ⃗ , we have the term (2 cos θ rˆ + sin θ θˆ ). Due to
symmetry, we only have a ẑ -component of the force. Therefore,
z
rˆ → −cos γ zˆ = − zˆ
r

5-6
 Electromagnetism

and
l
θˆ = −sin γ zˆ = − zˆ ,
r
where

r= z 2 + l2 .

Also, we have θ = π − γ . So
z
cos θ = cos(π − γ ) = cos(π ) cos( −γ ) + sin(π ) sin( −γ ) = − cos γ = −
r
and
l
sin θ = sin(π − γ ) = sin(π ) cos( −γ ) + cos(π ) sin( −γ ) = −sin( −γ ) = sin γ = .
r
Therefore, we have
⎛ z ⎞⎛ z ⎞ ⎛ l ⎞⎛ l ⎞ 1
2 cos θ rˆ + sin θ θˆ = 2⎜ − ⎟⎜ − ⎟zˆ + ⎜ ⎟⎜ − ⎟zˆ = − 2 −2z 2 + l 2 zˆ .
⎝ r ⎠⎝ r ⎠ ⎝ r ⎠⎝ r ⎠ r
( )
Our force becomes

2πzσ 2α ⎛ 1
⎜ −
1 ⎞
⎟
R
( −2z + l )l zˆ dl
2 2
F⃗ =
8πε02 ⎝ z R2 + z 2 ⎠
∫ − 5/2
0 (z + l ) 2 2

⎡ ⎤
F⃗ =
zσ 2α ⎛ 1
⎜ −
1 ⎞⎢
⎟⎢
R2 ⎥
zˆ =
(
σ 2αR2 R2 + z 2 − z
zˆ .
)
3/2 ⎥
4ε02 ⎝ z R2 + z 2 ⎠⎢⎣ R2 + z 2
( ) ⎥⎦ (
4ε02 R2 + z
2
)
So at z = d , the force of attraction is

F=
σ 2αR2 ( R2 + d 2 − d ).
2
(
4ε02 R2 + d )
We can verify this using

F ⃗ = ( p ⃗ ⋅ ∇) E ⃗

Note

σα ⎛ z ⎞∂
p⃗ ⋅ ∇ = ⎜1 − ⎟
2ε0 ⎝ R2 + z 2 ⎠ ∂z

5-7
 Electromagnetism

and

σα ⎛ z ⎞∂⎡ σ ⎛ z ⎞ ⎤
F ⃗ = ( p ⃗ ⋅ ∇) E ⃗ = ⎜1 − ⎟ ⎢ ⎜1 − ⎟zˆ⎥
2ε0 ⎝ R2 + z 2 ⎠ ∂z ⎢⎣ 2ε0 ⎝ R2 + z 2 ⎠ ⎥⎦
⎡ ⎤
F⃗ = −
zσ 2α ⎛ 1
⎜ −
1 ⎞⎢
⎟⎢
R2 ⎥
zˆ = −
(
σ 2αR2 R2 + z 2 − z )
zˆ ,
3/2 ⎥
4ε02 ⎝ z R2 + z 2 ⎠⎢⎣ R2 + z 2
( ) ⎥⎦ (
4ε02 R2 + z
2
)
which is equal in magnitude and opposite in direction to what was found above.
Why is this? In the first method, we calculated the force on the plate from the atom.
So a positive force is ‘attractive’. In the second method, we are finding the force on
the atom from the plate. A negative force at the atom ‘attracts’ it to the plate.

Problem 5.3. Consider p1⃗ and p2⃗ below. Find the force of p2⃗ on p1⃗ and verify using
the energy of the configuration.

Solution The electric field due to a dipole is

p
⃗ (r , θ ) =
Edip
4πε0r 3
(
2 cos θ rˆ + sin θ θˆ . )
Here, θ = π , so the field at p1⃗ due to p2⃗ is

−2p2 −2p2
⃗ =
Edip rˆ = yˆ .
3
4πε0r 4πε0y 3

The force is given by

(
F ⃗ = p⃗ ⋅ ∇ E⃗ )
where p ⃗ = p1⃗ = −p1 yˆ . So
∂
p ⃗ ⋅ ∇ = − p1 .
∂y

5-8
 Electromagnetism

Therefore,
∂ ⎛ −2p2 ⎞ 2p p ∂ −3
( )
F ⃗ = p ⃗ ⋅ ∇ E ⃗ = −p1 ⎜ yˆ ⎟ = 1 2
∂y ⎝ 4πε0y 3 ⎠ 4πε0 ∂y
( )
y yˆ

3p1 p2
F⃗ = − yˆ .
2πε0y 4
The energy stored in this configuration is given by
⎡ ⎛ −2p2 ⎞⎤ ⎛ −2p2 ⎞ 2p p
U = −p1⃗ ⋅ E ⃗ = −⎢( −p1 yˆ ) ⋅ ⎜ y
ˆ ⎟⎥ = − ( − p1 ) ⎜ ⎟ = − 1 23 .
⎣ ⎝ 4πε0y ⎠⎦
3
⎝ 4πε0y ⎠
3
4πε0y

From this, the force is given by

⎛ −2p1 p2 ⎞ 2p1 p2 ∂ −3
F ⃗ = −∇U = −∇⎜
⎝ 4πε0y ⎠
3
⎟=
4πε0 ∂y
y yˆ ( )
3p1 p2
F⃗ = − yˆ
2πε0y 4
as expected.

Problem 5.4. Consider the two dipoles depicted below. Find the angle γ that
maximizes and minimizes the magnitude of the torque on p2⃗ due to p1⃗ .

Solution The electric field due to a dipole is

p
⃗ (r , θ ) =
Edip
4πε0r 3
(
2 cos θ rˆ + sin θ θˆ .)
Here, θ = π2 , so the field at p2⃗ due to p1⃗ is
p1 ˆ
⃗ =
Edip θ,
4πε0r 3
which points down. The torque is given by
N⃗ = p2⃗ × Edip
⃗ .

5-9
 Electromagnetism

⃗ and p2⃗ by
We can express Edip
p1
⃗ =−
Edip zˆ
4πε0y 3

and

p2⃗ = p2 cos γ yˆ + p2 sin γ zˆ .

So

xˆ yˆ zˆ
0 p2 cos γ p2 sin γ −p1 p2 cos γ
N⃗ = p2⃗ × Edip
⃗ = = xˆ .
p1 4πε0y 3
0 0 −
4πε0y 3

We can see that ∣N⃗ ∣ is maximum when γ = 0 and γ = π

p1 p2
N⃗ =
4πε0y 3
π 3π
and minimum when γ = 2
and γ = 2

N⃗ = 0.

Note the effect of aligning the dipole parallel to the field, ∣p ⃗ × E ∣⃗ = 0, and aligning
the dipole perpendicular to the field ∣ p ⃗ × E ∣⃗ = pE .

Problem 5.5. A sphere of radius R carries polarization P ⃗ = kr rˆ , where k is a

constant, from r = a to r = R . Find the electric field in all regions.

5-10
 Electromagnetism

Solution The bound charges are given by

k k
σ b(a ) = P ⃗ ⋅ nˆ = rˆ ⋅ ( −rˆ ) = −
a a

k k
σ b(R ) = P ⃗ ⋅ nˆ = rˆ ⋅ rˆ =
R R

1 ∂ ⎛ 2k⎞ k
ρb = −∇ ⋅ P ⃗ = − ⎜r ⎟ = − 2 .
2
r ∂r ⎝ r ⎠ r

When r < a , qenc = 0, so E ⃗ = 0. When a ⩽ r ⩽ R , we have

qenc
∮S E ⃗ ⋅ da ⃗ = ε0

with
r
k k
qenc
a
(
= − 4πa 2 + 4π ) ∫ −
(r′) 2
(r′)2 dr′ = −4πkr .
a

So

∮S E ⃗ ⋅ da ⃗ = E 4πr 2 = − 4πεkr
0

and

k 1 k r⃗
E⃗ = − rˆ = − .
ε0 r ε0 r 2

When r > R , we have

R
k k k
qenc = −
a
( )
4πa 2 + 4π ∫ −
(r′) 2
(r′)2 dr′ +
R
(
4πR2 = 0 )
a

so E ⃗ = 0.

Problem 5.6. Consider a very long cylinder of radius R hollowed out to a radius a
and carrying a uniform, radial polarization P ⃗ and charge density ρ = ks , where k
is a constant. Find the electric field in all three regions (P ⃗ from a to R , ρ from 0
to R ).

5-11
 Electromagnetism

Solution For s < a , all we have is charge, so

qenc
∮S E ⃗ ⋅ da ⃗ = ε0
,

where
s
2π lks 3
qenc = 2π l ∫ ks′s′ds′ =
3
.
0

So
3
∮S E ⃗ ⋅ da ⃗ = E 2πsl = 2π3lεks0
and
s 2k
E⃗ = sˆ .
3ε 0

For a ⩽ s ⩽ R , we have bound charge

σ b(a ) = P ⃗ ⋅ nˆ = Psˆ ⋅ ( −sˆ ) = −P

1 ∂ P
ρb = −∇ ⋅ P ⃗ = − (sP ) = − .
s ∂s s
So,
s
2π lks 3 P ⎛ ks 3 ⎞
qenc = −P 2πal+
3
+ 2π l ∫ − s′ ds′ = 2π l⎜
s′ ⎝ 3
− Pa − Ps + Pa⎟
⎠
a

⎛ ks 2 ⎞
qenc = 2π ls⎜ − P⎟.
⎝ 3 ⎠
Therefore,
⎛ 2 ⎞
∮S E ⃗ ⋅ da ⃗ = E 2πsl = 2πε0ls ⎜⎝ ks3 − P⎟
⎠

5-12
 Electromagnetism

So
1
E⃗ = (
3ε 0
ks 2 − 3P sˆ . )
For s > R , we have
σ b(R ) = P ⃗ ⋅ nˆ = Psˆ ⋅ sˆ = P.
So,
R
2π lkR3 P
qenc = −P 2πal +
3
+ 2π l ∫ − s′ ds′ + P 2πRl
s′
a
⎛ kR3 ⎞
= 2π l⎜ − Pa − PR + Pa + PR⎟
⎝ 3 ⎠

2π lkR3
qenc = .
3
Therefore,
3
∮S E ⃗ ⋅ da ⃗ = E 2πsl = 2π3lεkR
0

and
kR3
E⃗ = sˆ .
3sε0

Problem 5.7. Consider a cylinder of radius R and length L , carrying polarization

P ⃗ = Pzˆ . Find the potential d from the cylinder.

Solution The potential is

∫V r r⋅ 2P dτ.
1 ˆ ⃗
V (r ⃗ ) =
4πε0

5-13
 Electromagnetism

We can see from

that

r= (l + d )2 + s 2 .

If z = 0 is the left side of the cylinder, then l goes from L to 0. So

r= (L − z + d )2 + s 2 .

Also,

l+d L−z+d
rˆ = cos θ zˆ = zˆ = zˆ .
r r

Since dτ = s ds dϕ dz , we have
L R 2π
1 P (L − z + d )s
V=
4πε0
∫∫∫ ⎡ (L − z + d )2 + s 2⎤3/2
dϕ ds dz .
0 0 0 ⎣ ⎦

Using u = L − z + d , we have du = −dz . Evaluating u at the endpoints yields

u(z = 0) = L + d

And

u (z = L ) = d .

So

P
L +d R
us P
L +d
⎛ u ⎞
V= ∫ ∫ 3/2
ds du = ∫ ⎜1 − ⎟du .
2ε 0
d 0 ( u2 + s2 ) 2ε 0
d
⎝ R2 + u 2 ⎠

5-14
 Electromagnetism

Using x = R2 + u 2 , we have dx = 2u du and

⎛ R 2 + (L + d ) 2 ⎞
P ⎜ L +d 1 −1 ⎟
V= u∣d − ∫ x 2 dx⎟
2ε0 ⎜⎝ R 2+ d 2
2 ⎠
P ⎡
= L+d−d− R 2 + (L + d )2 + R2 + d 2 ⎤⎦
2ε0 ⎣
P ⎡
V= L− R 2 + (L + d )2 + R2 + d 2 ⎤⎦ .
2ε0 ⎣

Problem 5.8. Consider a sphere of radius R carrying polarization P ⃗(r ⃗ ) = kr nrˆ where
n is an integer and k is a constant. Find the charge density required to cancel the
polarization.
Solution The bound charge is given by
1 ∂ 2 n
ρb = −∇ ⋅ P ⃗ = −
r 2 ∂r
r kr ( )
k ∂ n +2 k (n + 2) n+1
=− 2
r ∂r
r ( )=−
r2
r = −k (n + 2)r n−1.

Therefore, a charge density

ρ = k (n + 2)r n−1

will cancel the bound charge produced by P ⃗(r ⃗ ) = kr nrˆ .

Problem 5.9. A spherical shell of radius R with surface charge density σ is

surrounded up to radius a by an LIH dielectric material of susceptibility χe . Find
the electric displacement and the electric field.

5-15
 Electromagnetism

Solution Gauss’s law for electric displacement is given by

∮S D⃗ ⋅ da ⃗ = qf enc
.

For r < R , we have

qfenc = 0.

Therefore,
D = 0.
For r > R , the left-hand side of Gauss’s law is given by

∮S D⃗ ⋅ da ⃗ = ∮S D da = D ∮S da = D4πr 2.
Now, the enclosed free charge is
qfenc = σ 4πR2 .

So
D 4πr 2 = σ 4πR2
and
σR 2 2
⃗ = σR r.
D= → D ˆ
r2 r2
Let us consider the electric field. For r < R , we have

E⃗ = 0
Since

D ⃗ = εE ⃗ ,
for R < r < a , the electric displacement is
σR 2
D= ,
r2
so the electric field is

D⃗ D⃗ D⃗ σR 2
E⃗ = = = = rˆ .
ε ε0εr (
ε0 1 + χe )
ε0 1 + χe r 2 ( )
Finally, for r > a , the displacement is
σR 2
D=
r2

5-16
 Electromagnetism

so
D⃗ σR 2
E⃗ = = r.
ˆ
ε0 ε0r 2

Problem 5.10. For the previous problem, calculate the electric potential everywhere,
relative to infinity.
Solution The electric field in the three regions is given by

⎧ 0 r<R
⎪ 2
⎪ σR
⎪ R<r<a
E = ⎨ ε0(1 + χe )r 2 .
⎪ 2
⎪ σR
r>a
⎪
⎩ ε0r 2

For r > a ,
r r r
σR 2 σR 2 σR 2
V=− ∫ E ⃗ ⋅ dl⃗ = − ∫ ε0r′2
dr ′ =
ε0r′
=
ε0r
.
∞ ∞ ∞

For R < r < a

a r
σR 2 σR 2
V =− ∫ E ⃗ ⋅ dl⃗ = − ∫ dr − ∫ dr ′
∞
ε0r 2
a (
ε0 1 + χe r′2)
σR 2 σR 2 ⎛ 1 1⎞
= + ⎜ − ⎟.
ε0a ε0(1 + χe ) ⎝ r a⎠

For r < R
a R 0
σR 2 σR 2
V =− ∫ dr − ∫ dr − ∫ 0dr
∞
ε0r 2
a (
ε0 1 + χe r 2) R

σR 2
σR ⎛1 1⎞
2
= + ⎜ − ⎟ = const.
ε0a ε0 1 + χe ⎝ R
( a⎠ )
Problem 5.11. A long cylinder of radius a carries a charge density that is
proportional to the distance from the axis, ρ = ks , k constant. The cylinder
is surrounded by rubber insulation out to a radius R . Find the electric
displacement.

5-17
 Electromagnetism

Solution For s > a , we have

Gauss’s law states

∮S D⃗ ⋅ da ⃗ = qf enc
,

where the left-hand side is given by

∮S D⃗ ⋅ da ⃗ = ∮S D da = D ∮S da = D2πsl
and the enclosed free charge is
a 2π l
2πka 3l
qfenc = ∫V ρ dτ = k ∫ 2
s ds ∫ dϕ ∫ dz =
3
.
0 0 0

So
2πka 3l
D 2πsl =
3
And
ka 3 ka 3
D= → D⃗ = sˆ .
3s 3s
D⃗
From this, E ⃗ = ε0
for s > R ; if we knew P ⃗ , we could find the electric field as well.

5-18
 Electromagnetism

For s < a ,

Now, the enclosed free charge is

s 2π l
2πks 3l
qfenc = ∫V ρ dτ = k ∫ s′s′ds′ ∫ dϕ ∫ dz =
3
.
0 0 0

Therefore,
2πks 3l
D 2πsl =
3
and
ks 2 ks 2
D= → D⃗ = sˆ .
3 3
The electric field can be easily obtained,

D⃗ ks 2
E⃗ = = sˆ .
ε0 3ε 0

Problem 5.12. A sphere of radius R carries a polarization P ⃗ = kr 2rˆ, k constant:

(a) Calculate the bound charges σb and ρ b.
(b) Find the electric field inside and outside the sphere by using the bound charges
and Gauss’s law for E.⃗
(c) Calculate the field by using Gauss’s law for D.⃗
Solutions
(a) Calculate the bound charges σb and ρ b.
The bound charge density is given by

⎡1 ∂ 2 2 ⎤
ρb = −∇ ⋅ P ⃗ = −⎢ 2
⎣ r ∂r (
r kr ⎥ = −4kr
⎦ )
and the surface charge density is given by

σ b = P ⃗ ⋅  nˆ = kr 2rˆ ⋅ nˆ r=R = kR2 = const.

5-19
 Electromagnetism

(b) Find the electric field inside and outside the sphere by using the bound charges
and Gauss’s law for E ⃗ .
What is the total bound charge (what do you expect it to be)?
R 2π π

qb = σ b(area) + ∫V ρb dτ = kR 4πR 2 2
+ ∫ ( −4kr)r 2
dr ∫ dϕ ∫ sin θ dθ
0 0 0
R
r4
= 4πkR 4 − 4π 4k = 4πkR 4 − 4πkR 4 = 0.
4 0

For r < R , we have Gauss’s law

qenc
∮S E ⃗ ⋅ da ⃗ = ε0

with the left-hand side given by

∮S E ⃗ ⋅ da ⃗ = E 4πr 2
and the enclosed charge
r 2π π

qenc = ∫ ρb dτ =
V
∫( )
−4kr′ r′2dr′ ∫ dϕ ∫ sin θ dθ
0 0 0

4 r
r′
= −4π 4k = −4πkr 4.
4 0

Therefore,

−4πkr 4
E 4πr 2 =
ε0

with

kr 2 kr 2
E= → E⃗ = r.
ˆ
ε0 ε0

For r > R , we found qenc = 0 so E ⃗ = 0.

(c) Calculate the field by using Gauss’s law for D⃗ .
To find the electric displacement, we consider Gauss’s law for dielectrics

∮S D⃗ ⋅ da ⃗ = qf enc
.

5-20
 Electromagnetism

However, we have no free charge, qfenc = 0, and D⃗ = 0. Therefore, from

D⃗ = ε0E ⃗ + P ⃗
we have

P⃗
E⃗ = − .
ε0

So
kr 2rˆ
E⃗ = − (r < R )
ε0
and

E ⃗ = 0 (r > R )

as expected.

Problem 5.13. A spherical conductor of radius R carries a surface charge density σ .

The sphere is surrounded by alinear homogenous dielectric of susceptibility χe .
Calculate the energy of this configuration.
Solution The energy is given by

W=
ε0
2
∫ εrE 2dτ = 12 ∫ D⃗ ⋅ E ⃗dτ.
It is very easy to obtain D⃗ and E ⃗ . For r < R , we have Gauss’s law
qenc
∮S E ⃗ ⋅ da ⃗ = ε0

with qenc = 0 so E ⃗ = 0. Also,

∮S D⃗ ⋅ da ⃗ = qf enc
=0

So
D⃗ = 0.
For r > R , the total enclosed charge is simply
qfenc = σ 4πR2

and

∮S D⃗ ⋅ da ⃗ = D4πr 2.

5-21
 Electromagnetism

So
D 4πr 2 = σ 4πR2
With
σR 2
D= .
r2
For R < r < a , the polarization is given by

P ⃗ = ε0 χe E ⃗

so the electric displacement is

D⃗ = ε0E ⃗ + P ⃗ = εE ⃗ .

Solving for the electric field, we have

D⃗ σR 2 σR 2
E⃗ = = = .
ε εr 2 (
ε0 1 + χe r 2 )
For r > a , the field is just
σR 2
E= .
ε0r 2

Therefore, our electric displacements are

⎧ 0 r<R
⎪
D = ⎨ σR 2
⃗
⎪ 2 r>R
⎩ r

and our electric fields are

⎧ 0 r<R
⎪ 2
⎪ σR
⎪ R<r<a
E = ⎨ 0( + χe )r 2
⃗ ε 1 .
⎪ 2
⎪ σR
r>a
⎪
⎩ ε0r 2

Returning to the energy, we have

a ∞
1 4π σ 2R 4 2 4π σ 2R 4 2
W=
2
∫ D ⃗ ⋅ E ⃗ dτ =
2
∫ εr 4
r dr +
2
∫ ε0r 4
r dr ,
R a

5-22
 Electromagnetism

where we used

∫ dτ = 4π ∫ r 2dr.
Therefore,
2πσ 2R 4 ⎡ 1 ⎛ 1 1 ⎞ 1 ⎤ 2πσ 2R 4 ⎛ 1 χ⎞ 2πσ 2R 4 ⎛ 1 χ⎞
W= ⎢ ⎜ − ⎟+ ⎥= ⎜ + e⎟ = ⎜ + e ⎟.
ε0 ⎣ εr ⎝ R a⎠ a⎦ ε0εr ⎝ R a ⎠ ε0 1 + χe ⎝ R a⎠ ( )
Problem 5.14. A sphere of radius R, made of linear homogeneous dielectric
material, is brought into a uniform electric field of magnitude E0⃗ . Using the
Laplace equation and Legendre polynomials, find the electric field inside the sphere.
Solution In spherical coordinates, Laplace’s equation is given by
1 ∂ ⎛ 2 ∂V ⎞ 1 ∂ ⎛ ∂V ⎞ 1 ∂ 2V
⎜r ⎟ + ⎜sin θ ⎟ + = 0.
r 2 ∂r ⎝ ∂r ⎠ r 2 sin θ ∂θ ⎝ ∂θ ⎠ r 2(sin θ )2 ∂ϕ 2

We have azimuthal symmetry, therefore the potential is ϕ independent, so

1 ∂ ⎛ 2 ∂V ⎞ 1 ∂ ⎛ ∂V ⎞
⎜r ⎟ + ⎜sin θ ⎟ = 0.
r ∂r ⎝ ∂r ⎠ r sin θ ∂θ ⎝
2 2
∂θ ⎠

We have outlined the solutions to this in chapter 3, and found the general solution to
be given by
∞
⎛ Bl ⎞
V (r , θ ) = ∑⎜⎝Al r l + ⎟Pl (cos θ ).
r l +1 ⎠
l =0

Now we can look at the boundary conditions for this particular problem. We need
the electric potential to satisfy:
1) Vin = Vout at r = R .
⊥ ⊥
2) εaboveEabove − ε belowE below = σfree .
3) At large distance from the sphere: r ≫ R , the potential must be
Vout = −E0r cos θ .

Since the free surface charge density is zero, and by using the relationship between
the electric field and the electric potential, the second condition becomes
∂Vin ∂V
ε = ε0 out
∂r ∂r
at r = R .
Now with clear boundary conditions and the general solution for the potential,
we can write the potential inside the sphere and the potential outside the sphere.
Looking at our general solution, we require Bl = 0 for r < R ; otherwise V → ∞ as
r → 0. Similarly, we require Al = 0 for r > R ; otherwise V → ∞ as r → ∞. For the

5-23
 Electromagnetism

potential outside the sphere we want to make sure we cover the third boundary
condition, and this is why we will have two terms.
Inside the sphere, we have
∞
Vin(r , θ ) = ∑Al r l Pl (cos θ )
l=0

and outside the sphere, we have

∞
B
Vout(r , θ ) = −E 0r cos θ + ∑ r l +l 1 Pl (cos θ ).
l =0

From the first boundary condition, at r = R ,

Vin = Vout

So
∞ ∞
B
∑Al Rl Pl (cos θ ) = −E 0R cos θ + ∑ Rl +l 1 Pl (cos θ ).
l =0 l =0

For l = 1, P1(cos θ ) = cos θ . So

B1
A1R cos θ = −E 0R cos θ + cos θ
R2
And
B1
A1R = −E 0R + .
R2
For l ≠ 1,
Bl
Al Rl = .
Rl +1
From the second boundary condition
∂Vin ∂V
ε = ε0 out
∂r ∂r
we have
∞ ∞
(l + 1)Bl
εr∑lAl Rl −1Pl (cos θ ) = −E 0 cos θ − ∑ Pl (cos θ ).
l =0 l =0
Rl +2

For l ≠ 1,
(l + 1)Bl
εrlAl Rl −1 = −
Rl +2

5-24
 Electromagnetism

so we must have
Al = Bl = 0.
For l = 1
2B1
εrA1 = −E 0 − .
R3
Consider our two equations relating A1 and B1,
B1
A1R = −E 0R +
R2
and
2B1
εrA1 = −E 0 − .
R3
From the first,
B1
A1 = −E 0 − ,
R3
and substitution into the second yields
⎛ B ⎞ 2B ⎛ε 2 ⎞
R ⎠ R ⎝R R ⎠
(
εr⎜ −E 0 − 13 ⎟ = −E 0 − 31 → B1⎜ r3 + 3 ⎟ = E 0 εr − 1
⎝ )
So
εr − 1 3
B1 = R E0
εr + 2

and
3E 0
A1 = − .
εr + 2
Therefore the potential is
3E 0
Vin(r , θ ) = − r cos θ .
εr + 2
Noting that z = r cos θ ,
3E 0
Vin = − z.
εr + 2
The field inside the sphere is uniform and in the same direction as E0⃗ :
3
E⃗ = E 0⃗ .
εr + 2

5-25
 Electromagnetism

Bibliography
Byron F W and Fuller R W 1992 Mathematics of Classical and Quantum Physics (New York:
Dover)
Griffiths D J 1999 Introduction to Electrodynamics 3rd edn (Englewood Cliffs, NJ: Prentice Hall)
Griffiths D J 2013 Introduction to Electrodynamics 4th edn (New York: Pearson)
Halliday D, Resnick R and Walker J 2010 Fundamentals of Physics 9th edn (New York: Wiley)
Halliday D, Resnick R and Walker J 2013 Fundamentals of Physics 10th edn (New York: Wiley)
Jackson J D 1998 Classical Electrodynamics 3rd edn (New York: Wiley)
Purcell E M and Morin D J 2013 Electricity and Magnetism 3rd edn (Cambridge: Cambridge
University Press)
Rogawski J 2011 Calculus: Early Transcendentals 2nd edn (San Francisco, CA: Freeman)

5-26
 IOP Concise Physics

Electromagnetism
Problems and solutions
Carolina C Ilie and Zachariah S Schrecengost

Chapter 6
Magnetic fields in matter

Similarly to the electric field in matter and the electric dipoles, when magnetic dipoles
are subjected to a magnetic field, they may align and the medium becomes magnetized.
Depending on the magnetization, we define different magnetic materials: para-
magnets (the magnetization M⃗ is parallel to the magnetic field B ⃗ ), diamagnets (the
magnetization M⃗ is opposite to the magnetic field B ⃗ ), and the special class of materials,
ferromagnets, which remain magnetized even after the magnetic field becomes zero.
We would like to mention here that different sources may have different names for B ⃗
and H⃗ . Here B ⃗ is the magnetic field and H⃗ is simply the H-field. Griffiths refers to H⃗ as
the auxiliary field but we have chosen the H-field to eliminate any confusion. In other
books, you may find that B ⃗ is the magnetic flux density, while H⃗ is the magnetic field.

6.1 Theory
6.1.1 Torque on a magnetic dipole moment
The torque on a magnetic dipole moment m⃗ in a magnetic field B ⃗ is
N⃗ = m⃗ × B ⃗.

6.1.2 Force on a magnetic dipole

The force on a magnetic dipole moment m⃗ in a magnetic field B ⃗ is
F ⃗ = ∇(m⃗ ⋅ B ⃗ ) .
6.1.3 H-field
Given magnetic field B ⃗ and magnetization M⃗ , the H-field is
1
H⃗ = B ⃗ − M⃗
μ0

doi:10.1088/978-1-6817-4429-2ch6 6-1 ª Morgan & Claypool Publishers 2016

 Electromagnetism

with
∇ × H⃗ = Jf⃗ ,
where Jf⃗ is the free current density. From Stoke’s law,

∮S H⃗ ⋅ dl⃗ = If enc
,

where Ifenc is the free current.

6.1.4 Linear media

Given H-field H⃗ , the magnetization is given by
M⃗ = χm H⃗ ,
where χm is the magnetic susceptibility. The magnetic field is given by

( ) (
B ⃗ = μ0 H⃗ + M⃗ = μ0 1 + χm H⃗ = μH⃗ , )
where μ is the magnetic permeability of the material, and μ0 is the permeability of
the vacuum.

6.1.5 Surface bound current due to magnetization M⃗

Given magnetization M⃗ and normal vector n̂ , the surface bound current is
K⃗b = M⃗ × n.
ˆ

6.1.6 Volume bound current due to magnetization M⃗

Given magnetization M⃗ and normal vector n̂ , the volume bound current is
Jb⃗ = ∇ × M.⃗

6.2 Problems and solutions

Problem 6.1. Find the force between the two magnetic dipoles below.

6-2
 Electromagnetism

Solution The magnetic field due to dipole m⃗1 at m⃗ 2 is given by

μm
(
⃗ m1 = 0 1 2 cos θ rˆ + sin θ θˆ ,
Bdip,
4πr 3
)
π
where θ = 2
. So
μ0m1 ˆ μ m1
⃗ m1 =
Bdip, 3
θ = − 0 3 zˆ .
4πr 4πr
⃗ m1 is
The force on m⃗ 2 due to Bdip,

(
F ⃗ = ∇ m⃗ 2 ⋅ Bdip,
⃗ m1 , )
where
⎛ μm ⎞ μ mm
⃗ m1 = m2zˆ ⋅ ⎜ − 0 1 zˆ⎟ = − 0 1 2 .
m⃗ 2 ⋅ Bdip,
⎝ 4πr ⎠
3
4πr 3
So
⎛ μ m1m2 ⎞ μ m1m2 3μ m1m2
F ⃗ = ∇⎜ − 0 3 ⎟ = − 0
⎝ 4πr ⎠ 4π
(
−3r −4 rˆ = 0 4 rˆ .
4πr
)

Problem 6.2. Find the force on a dipole located on the axis of an infinitely long
cylinder of radius R , rotating at ω and carrying surface charge σ .

Solution We can think of the rotating cylinder as a solenoid with nI → K . So

K = σv = σωR
and
B ⃗ = μ0nIzˆ = μ0Kz,
ˆ
which is the magnetic field inside a solenoid. The force is given by
( )
F ⃗ = ∇ m⃗ ⋅ B ⃗ = ∇(mzˆ ⋅ μ0Kzˆ ) = ∇(μ0mK ) = 0.

This is an example that shows the force on a dipole in a uniform field is zero. Since
we can think of a dipole as a current loop, this is equivalent to saying a current loop
in a uniform field experiences zero net force.

6-3
 Electromagnetism

Problem 6.3. Consider two current loops of radius R whose orientation is depicted
below. Find the torque between them and the angle γ that minimizes this torque.

Solution Looking at this from the side, we have

Considering m⃗1, the field it produces is

μm
(
⃗ m1 = 0 1 2 cos θ rˆ + sin θ θˆ ,
Bdip,
4πr 3
)
π
where θ = γ + 2 , r = y, rˆ = yˆ , and θ̂ = −ẑ . Also, we have
m1 = πR2I
⎛ π⎞
cos θ = cos⎜γ + ⎟ = −sin γ
⎝ 2⎠
and
⎛ π⎞
sin θ = sin⎜γ + ⎟ = cos γ .
⎝ 2⎠
Using these values,
μ0πR2I μ 0 R 2I
⃗ m1 =
Bdip,
4πy 3
(
− 2 sin γ y
ˆ − cos γ z
ˆ = − )4y 3
( )
2 sin γ yˆ + cos γ zˆ .

6-4
 Electromagnetism

We have m⃗ 2 given by
m⃗ 2 = πR2I (sin γ yˆ + cos γ zˆ ).
Now the torque is
μ 0 R 2I
N⃗ = m⃗ 2 × Bdip,
⃗ m1 = − ( πR2I ⎡⎣ (sin γ yˆ + cos γ zˆ ) × (2 sin γ yˆ + cos γ zˆ )⎤⎦ .
)
4y 3
Looking at just the cross product term, we have
xˆ yˆ zˆ
(sin γ yˆ + cos γ zˆ ) × (2 sin γ yˆ + cos γ zˆ ) = 0 sin γ cos γ = −sin γ cos γ xˆ .
0 2 sin γ cos γ

Therefore,
μ0πR 4I 2 μ0πR 4I 2 sin(2γ )
N⃗ = m⃗ 2 × Bdip,
⃗ m1 = − ( − sin γ cos γ x
ˆ ) = xˆ .
4y 3 8y 3
To find the γ that minimizes this, we must differentiate the torque with respect to γ ,
∂N μ πR 4I 2 ∂ μ πR 4I 2
= 0 3 (sin(2γ )) = 0 3 cos(2γ ).
∂γ 8y ∂γ 4y
We can find the extreme values by setting this equal to zero. Note we have
∂N
=0
∂γ
when
cos(2γ ) = 0,
which is zero when
⎛ 2n − 1 ⎞
2γ = ⎜ ⎟π
⎝ 2 ⎠
for a positive integer n. Solving for γ we have
⎛ 2n − 1 ⎞
γ =⎜ ⎟π .
⎝ 4 ⎠
To find the minimum, we must find the second derivative of the torque. So
∂ 2N μ0πR 4I 2 ∂ μ0πR 4I 2 sin(2γ )
= [cos(2γ )] = − ,
∂γ 2 4y 3 ∂γ 2y 3
Substitution of γ yields
μ0πR 4I 2 ⎡ ⎛ 2n − 1 ⎞ ⎤
β=− sin⎢2⎜ ⎟π ⎥ ,
2y 3 ⎣ ⎝ 4 ⎠ ⎦

6-5
 Electromagnetism

where a β > 0 indicates a minimum. Dropping all but the sign and the sine, we have
⎛ π⎞ ⎡ ⎛ π⎞ ⎛ π ⎞⎤
β = −sin⎜nπ − ⎟ = −⎢sin(nπ ) cos⎜ − ⎟ + cos(nπ ) sin⎜ − ⎟⎥ = cos(nπ ).
⎝ 2⎠ ⎣ ⎝ 2⎠ ⎝ 2 ⎠⎦

We have β is positive for n = 2, 4, 6, … so the torque is minimized for

⎛ 2n − 1 ⎞
γ=⎜ ⎟π ,
⎝ 4 ⎠
when n = 2, 4, 6, … or for any positive integer n ,
⎛ 2(2n ) − 1 ⎞ ⎛ 1⎞ π
γ=⎜ ⎟π = ⎜n − ⎟π = nπ − .
⎝ 4 ⎠ ⎝ 4 ⎠ 4
Since each dipole moment is at an angle γ , the angle between them is
π
2γ = 2nπ −
2
Note the multiple of 2nπ is just the addition of another complete circle. The result
that minimizes the torque is a γ that causes the dipoles to be perpendicular to each
other.

Problem 6.4. Consider the rotating cylindrical shell in problem 4.6, where the z-axis
starts at the left side of the cylinder. Suppose we place a dipole m⃗ = mzˆ at a distance
d from the right-hand side of the cylinder, as depicted below. Find the force on the
dipole.

Solution From problem 4.6, the field is given by

μ0σωR ⎡⎢ d+L d
⎤
⎥zˆ ,
B⃗ = –
2 ⎢⎣ R2 + (d + L )2 R2 + d 2 ⎥⎦

where d was the distance from our point to the right-hand side of the cylinder. We
can rewrite this considering d = z − L . So

6-6
 Electromagnetism

μ0σωR ⎡⎢ z z−L
⎤
⎥zˆ .
B⃗ = –
2 ⎢⎣ R2 + z 2 R2 + (z − L )2 ⎥⎦

The force is given by

(
F ⃗ = ∇ m⃗ ⋅ B ⃗ )
with

μ0σωRm ⎡⎢ z z−L
⎤
⎥.
m⃗ ⋅ B ⃗ = –
2 ⎢⎣ R + z 2
2
R2 + (z − L )2 ⎥⎦

So

μ0σωRm ⎡⎢ z z−L
⎤
⎥
F⃗ = ∇ –
2 ⎢⎣ R2 + z 2 2 ⎥
R + (z − L ) ⎦
2

and
⎡ ⎤
μ0σωR3m ⎢ 1 1 ⎥
F⃗ = ⎢ 2 − zˆ .
3/2 ⎡ R2 + (z − L )2 ⎤3/2 ⎥
2 (
⎢⎣ R + z 2 ) ⎣ ⎦ ⎥⎦

At a distance d from the right-hand side of the cylinder, z = d + L . Therefore

⎡ ⎤
μ0σωR3m ⎢ 1 1 ⎥
F⃗ = ⎢ − 3/2 ⎥
zˆ .
⎡ ⎤3/2
2 ⎢⎣ ⎣ R2 + (d + L )2 ⎦ (
R2 + d 2 ) ⎥⎦

Problem 6.5. An infinitely long cylinder has a constant magnetization M⃗ parallel to

the axis of the cylinder. Find the magnetic field due to M⃗ everywhere.

6-7
 Electromagnetism

Solution The magnetization is given by

M⃗ = Mz.
ˆ
The bound volume current Jb⃗ is
Jb⃗ = ∇ × M⃗ = 0
since M⃗ = constant. The bound surface current Kb⃗ is
Kb⃗ = M⃗ × nˆ = Mzˆ × sˆ = Mϕˆ .
For s < R
B ⃗ = μ0M⃗ = μ0Mzˆ
and for s > R
B ⃗ = 0.
Outside, the field is zero (B ⃗ = 0 outside a solenoid).

Problem 6.6. A long circular cylinder of radius R has a magnetization M⃗ = ksϕˆ ,

where k is a constant, s the distance from the axis of the cylinder, and ϕ̂ the
azimuthal unit vector. Find the magnetic field due to M⃗ for s < R and s > R .
Solution Let us first find the bound volume current Jb⃗ and the bound surface current
Kb⃗ . The bound volume current is given by
1 ∂ 1 ∂ 1
Jb⃗ = ∇ × M⃗ =
s ∂s
(sMϕ )zˆ =
s ∂s
( )
ks 2 zˆ = 2kszˆ = 2kzˆ = const,
s
and the bound surface current is given by

(
Kb⃗ = M⃗ × nˆ = ks ϕˆ × sˆ ) s=R
= −kRz.
ˆ

So the bound current flows up the cylinder, and returns down the surface. Let us
check that the total current is zero. The total current due to the bound volume
current is given by
R
4πkR2
Itot,Jb = ∫ Jb⃗ ⋅ da ⃗ = ∫ Jb da = ∫ (2k )(2πs ds ) =
2
= 2πkR2
0

and the total current due to the bound surface current is

Itot,Kb = ∫ Kb dl = (−kR)2πR = −2πkR2.
Since they are equal and opposite, the total current is zero. Now we can find the
magnetic field using Ampère’s law. For s < R ,

∮S B ⃗ ⋅ dl⃗ = μ0Ienc.

6-8
 Electromagnetism

The left-hand side is given by

∮S B ⃗ ⋅ dl⃗ = B 2πs

and enclosed current is given by

s s

Ienc = ∫ Jb da = ∫ 2k 2πs′ds′ = 2kπs 2 .

0 0

Therefore,
B ⃗ = μ0ksϕˆ = μ0M.⃗
For s > R , our enclosed current is zero, so
B ⃗ = 0.

Problem 6.7. A long cylinder of radius R carries a magnetization M⃗ = ks 3ϕˆ , where k

is a constant. Find the magnetic field due to M⃗ everywhere.
Solution Let us start by finding the bound currents. The volume bound current is
given by
1 ∂ 1 ∂ 1 ∂ 1
Jb⃗ = ∇ × M⃗ =
s ∂s
(sMϕ )zˆ =
s ∂s
sks 3 zˆ = (
s ∂s
) ( )
ks 4 zˆ = 4ks 3zˆ = 4ks 2zˆ
s
and the surface bound current is
Kb⃗ = M⃗ × nˆ = ks 3 ϕˆ × sˆ ( ) s=R
= −kR3z.
ˆ

We can check that the total bound current is zero. From the bound volume current,
we have
R
4 R
Itot,Jb = ∫ Jb⃗ ⋅ da ⃗ = ∫ ( 4ks 2)(2πs ds) = 8π4ks = 2πkR 4
0 0

and from the bound surface current, we have

Itot,Kb = ∫ Kb dl = (−kR3)2πR = −2πkR 4.
Therefore, the total bound current is zero, Ib = 0. Now we can find the field by using
Ampère’s law. For s < R ,

∮S B ⃗ ⋅ dl⃗ = μ0Ienc.

The left-hand side is

∮S B ⃗ ⋅ dl⃗ = B 2πs

6-9
 Electromagnetism

and the enclosed current is

s s s

Ienc = ∫ Jb da = ∫ 2
4ks′ 2πs′ds′ = ∫ 8πk (s′)3ds′ = 2πks 4.
0 0 0

Therefore,
B 2πs = μ02πks 4 = μ0ks 3

And
B ⃗ = μ0ks 3ϕˆ = μ0M.⃗

For s > R , we have zero enclosed current. So,

B ⃗ = 0.

Problem 6.8. A sphere of radius R carries magnetization M⃗ = krϕˆ . Find the

magnetic field inside and outside.
Solution Since there is no free current, H⃗ = 0. Inside, we have magnetization, so
1
H⃗ = Bin⃗ − M⃗ .
μ0

So Bin⃗ is given by
Bin⃗ = μ0M⃗ = μ0krϕˆ .

Outside, we have no magnetization either, so

⃗ = 0.
Bout

Problem 6.9. An infinitely long wire carries current I and is surrounded by material,
out to radius R , with magnetization M⃗ = kϕˆ . Find the magnetic field for s < R and
s > R.
Solution For s < R, we have

∮S H⃗ ⋅ dl⃗ = Ifenc

with

∮S H⃗ ⋅ dl⃗ = H 2πs

and
Ifenc = I.

6-10
 Electromagnetism

Therefore,
I ˆ
H= ϕ.
2πs
Using
1
H⃗ = Bin⃗ − M⃗
μ0

the magnetic field inside is given by

⎛ I ⎞
(
Bin⃗ = μ0 H⃗ + M⃗ = μ0⎜)
⎝ 2πs
+ k ⎟ϕˆ .
⎠
I ˆ
For s > R , we still have H = 2πs
ϕ, but we do not have any magnetization. So
μ0I ˆ
⃗ = μ0H⃗ =
Bout ϕ,
2πs
which is exactly what we would expect from a wire carrying current I .

Problem 6.10. An infinitely long wire, of radius R carries magnetization M⃗ = ks 2zˆ .

At s = R , there is a surface current K⃗ = K 0ϕˆ . Find the field for s < R and
s > R.
Solution For s < R , both M⃗ and K⃗ contribute to the field. We can see the
contribution due to K⃗ is that of a solenoid,
BK⃗ = μ0K 0z.
ˆ

We also have H⃗ = 0, so
BM⃗ = μ0ks 2z.
ˆ

Therefore,

(
Bin⃗ = BK⃗ + BM⃗ = μ0 K 0 + ks 2 z.
ˆ )
For s > R , we have zero magnetization, so
BM⃗ = 0.
Also, there is zero field outside of a solenoid, so
BK⃗ = 0.
Therefore,
⃗ = 0.
Bout

6-11
 Electromagnetism

Problem 6.11. Find the H-field produced from a current density Jf⃗ = J0szˆ in two
ways.
Solution First, we will use
∇ × H⃗ = Jf⃗ .
Note, we must have H⃗ = H (s )ϕˆ . So
1 ∂
∇ × H⃗ = [sH (s )]zˆ = J0 szˆ
s ∂s
and
∂
[s H (s )] = J0 s 2 .
∂s
From this, we have
1 2
H (s ) = J0 s + C .
3
Since there is zero current at s = 0, we have H (0) = 0 → C = 0. Therefore,
J s2
H⃗ = 0 ϕˆ .
3
Now we will use

∮S H⃗ ⋅ dl⃗ = Ifenc,

where
s
2πJ0 s 3
Ifenc = ∫ J da = ∫ 2πs′J0 s′ds′ =
3
0

and

∮S H⃗ ⋅ dl⃗ = H 2πs .

Therefore,
J s2
H⃗ = 0 ϕˆ .
3
As expected from the first method. Note each equation we used is simply Stoke’s
theorem applied to the other.

Problem 6.12. This problem was inspired by a different problem presented in the
Electrodynamics graduate course by Dr Charles Ebner at the Ohio State University
in 2002. A sphere of radius R is uniformly polarized with a polarization P ⃗ . Within

6-12
 Electromagnetism

⃗
such a sphere, one can show that D⃗ = 23 P ⃗ and E ⃗ = − 3Pε . By using the similarity of
0
the equations of electrostatics and magnetostatics, find B ⃗ and H⃗ within a uniformly
magnetized sphere having magnetism M⃗ .
Solution The equivalent equations for electrostatics and magnetostatics are the
following
∇ × E⃗ = 0
∇ ⋅ D⃗ = 0
D⃗ P⃗
E⃗ = − .
ε0 ε0
For a system with no free current,
∇ × H⃗ = 0
∇ ⋅ B⃗ = 0
B⃗
H⃗ = − M⃗ .
μ0
D⃗
Comparing the equations, we note that E ⃗ is equivalent to H⃗ , E ⃗ ⇔ H⃗ ; is
ε0
B ⃗ D⃗ B⃗ P⃗ P⃗
equivalent to , ⇔ ; and is equivalent to M⃗ , ⇔ M⃗ .
μ0 ε 0 μ0 ε0 ε0
Starting with
2 ⃗
D⃗ = P
3
we can divide both sides by ε0
D⃗ 2 ⃗
= P.
ε0 3ε0
P⃗
Since ε0
⇔ M,⃗ we have
2 ⃗ 2
P ⇔ M⃗ .
3ε0 3
D⃗ B⃗
Using ε0
⇔ μ0
, we have
B⃗ 2
= M⃗ .
μ0 3
Therefore,
2μ 0
B⃗ = M⃗ .
3

6-13
 Electromagnetism

Now we will look at the electric field,

H⃗ ⇔ E ⃗ .
P⃗
Using ε0
⇔ M,⃗
P⃗ M⃗
E⃗ = − =− .
3ε0 3
Therefore,
M⃗
H⃗ = − .
3

Bibliography
Griffiths D J 1999 Introduction to Electrodynamics 3rd edn (Englewood Cliffs, NJ: Prentice Hall)
Griffiths D J 2013 Introduction to Electrodynamics 4th edn (New York: Pearson)
Halliday D, Resnick R and Walker J 2010 Fundamentals of Physics 9th edn (New York: Wiley)
Halliday D, Resnick R and Walker J 2013 Fundamentals of Physics 10th edn (New York: Wiley)
Jackson J D 1998 Classical Electrodynamics 3rd edn (New York: Wiley)
Purcell E M and Morin D J 2013 Electricity and Magnetism 3rd edn (Cambridge: Cambridge
University Press)

6-14