B.S.

Belikoy

| General
| methods
_ for solving
| physics
_ problems

= be es ee F Papert ‘ -_r

waits rPuDIsShe! ie) Wilerere)

 General Methods
for Solving Physics Problems
Bb. C. Bennkos

PEMIEHHE 3AQZAU 0 ®H3HNKE

OGmue meTroqH

Mocrpa «Bnicwian wrotay 1986

General
methods
for solving
physics
problems

Kade
Ss
oo)

Mir Publishers Moscow

Translated from the Russian by Eugene Yankovsky

First published 1989

TO THE READER

Mir Publishers would be grateful for your com-

ments on the content, translation and design of this
book. We would also be pleased to receive any other
suggestions you may wish to make.
Our address is:
Mir Publishers
2 Pervy Rizhsky Perculok
1-410, GSP, Moscow, 129820

Ha ak2aulicrom aauKe

Printed in the Union of Soviet Socialist Republics

ISBN 5-03-000907-8 © ae «Bricwiaa DIKONa»,

1
© Engen translation, Mir Publishers,
1989
PREFACE

The revolution in science and engineering makes it nec-

essary to seek new ways of mote effectively training spe-
cialists for the national economy. What must be done
so that the theoretical knowledge that a student acquires
in college does not remain passive but is used with maai-
mal efficiency in practical work? Until recently this was
achieved with relative success in the traditional way,
but now with the information explosion this is becoming
ever more difficult. The body of information has become
so large that it cannot be comprehended in the limited
period of education unless organized on anentirely new ba-
sis. This basis could be the extended and systematic use
of generalized methods, general methodological princi-
ples, and very general concepts.
This book attempts to do so in a segment of student
instruction of vital importance, the solution of physics
problems. The approach is based on the application of
the most general concepts of physics to the solution of
any problem. I consider the theoretical aspects underlying
the general approach to problem solution and methods
for solving standard, nonstandard, nonspecific, and
general problems.
Such an approach is not the only one, of course. Many
problems can be solved much faster by applying nonstan-
dard methods or by intuition. Nevertheless, a nonstandard
approach and the use of intuition are possible only when
the student possesses a methodological base for problem
solving and has mastered the general approach. This book
may, therefore, prove helpful for students of diversified
specialities.
Some of the problems discussed here have been taken
from A.G. Chertov and A.A. Vorob’ev, A Problem Book
in Physics (Moscow, 1981, in Russian), and I.E. Irodov,
6 Preface

Problems in General Physics (Moscow, 1979, English

translation: Mir Publishers, Moscow, 1981, 2nd ed. 1983).
The ideas incorporated in others belong to me.
I take this opportunity to thank Assistant Professors
V.B. Zernov and A.F. Menyaev and also the staff
of the Physics Faculty of the Moscow Institute of Elec-
tronic Machinery Industry for their constructive com-
ments to improve this book.
B.S. Belikov
CONTENTS

Introduction

Part 1. The Theoretical Bases of the General Ap-

proach to Solving Any Physics Problem
Chapter 1. The System of Fundamental Concepts of Physics
4. Some General Concepts of Physics
2. Idealization of a Physics Problem
3. Classification of Physics Problems
Chapter 2. Some General Methods for Solving Physics
Problems
4. Pte in Solving a Formulated Problem
5. Method of Analyzing the Physical Content of a
Problem
. General-Particular Methods. The DI Method
. The Simplification and Complication Method. The
Estimate Method
. The Problem Statement Method
oo. Another Classification of Formulated Problems
"MM

Part 2. Solution of Standard Problems

Mechantes
Chapter 3. The Motion of a Particle
10. Particle Kinematics
41. Particle Dynamics
12. Mechanical Oscillations
13. Conservation Laws
Chapter 4. The Motion of a Rigid Body
14. Rigid-Body Dynamics
15. Conservation Laws in Rigid-Body Dynamics
Elements of the Theory of Physical Fields
Chapter 5. The Gravitational Field 114
16. The Basic Problem of Gravitation Theory 144
17. The Gravitational Field Generated by a System of
Particles 113
18. The Gravitational Field Generated by an Arbitrary
Mass Distribution 117
8 Contents

Chapter 6. The Electric Field 427

49. The Electrostatic Ficld in a Vacuum 127
20. The Electrostatic Field in Insulators 140
24. Conductors in an Electrostatic Field 157
22. Direct Current 169
Chapter 7. The Magnetic Field 177
2 3. The Magnetic Field in a Vacuum 177
24. The Magnetic Field in Matler 191
Chapter 8. The Electromagnetic Field 195
25, Electromagnetic Induction and Self-Induction 195
26. Electromagnetic Oscillations
Chapter 9. Electromagnetic Waves 241
27. Interference of Light 211
28. Diffraction of Light 249
Thermodynamics and Kinetic Theory
Chapter 10. Thermodynamics 226
2 9. The First Law of Thermodynamics 226
30. The Second Law of Thermodynamics
oor 11. Kinetic Theory 241
31. The Maxwell- Boleriann Distribution
32. The Boltzmann Distribution 249

Part 3. Solution of Nonstandard, Nonspecified, and

Arbitrary Problems
Chapter 12. Nonstandard and Original Problems 205
33. Nonstandard Problems 259
34. Original Problems 264
Chapter 13. Nonspecified, Research, and Arbitrary Problems 271
35. Nonspecified Problems 271
36. Research Problems 276
37. Arbitrary Problems

Concluston
INTRODUCTION

It often happens that a student has a good knowledge of

the theoretical aspects of the physics course but does not
know how to solve physics problems. Some students ad-
mit that they encounter no difficulties in studying theory
and can memorize and understand formulas, definitions,
etc., but are not able to cope with problems. They do not
even know how to begin, and at times, while solving a prob-
lem, they get bogged down in theory and after writing
down numerous formulas, laws, and equations do not know
whether they have solved the problem or are close to or
far from the solution. Often, after solving a problem cor-
rectly in general form, they make mistakes in calcula-
tions, and an incorrect numerical result means an incor-
rect solution, after all, and cannot be considered valid.
Granted: it is not easy to learn to solve physics prob-
lems. One can know theory well and yet not be able to
solve the simplest preblem. This is no accident. Know-
ing theory is necessary but not sufficient for problem
solving. Besides specific knowledge one must possess
generalized knowledge. And this is usually acquired thro-
ugh experience, in the process of solving physics prob-
lems, usually toward the end of the physics course. But
sometimes generalized knowledge is not acquired at all.
The body of generalized knowledge necessary for prob-
lem solving is discussed mainly in the first two chapters
of this book. The reader will learn what a physics prob-
lem is, from what angle to approach it, what signifies that
a solution has been found, and much more.
The base of generalized knowledge is the fundamental
concepts of physics, which are methodological in character.
These are discussed in the first chapter.
There are relatively few fundamental methodological
concepts of physics. This book uses eight: a physical sys-
10 Introduction

tem, a physical quantity, a physical law. the state of a

physical system, interaction, a physical phenomenon, ideal-
ized objects and processes and a physical model. Being
interrelated, these concepts form a system. Of special
importance is the relation of a physical phenomenon
to all other fundamental concepts.
Using a system of fundamental concepts makes it pos-
sible to formulate the most important definition of a the-
oretical physical problem as a physical phenomenon in
which some of the relationships and quantities are unknown.
To solve a physics problem means to establish the
unknown relationships and determine the sought
physical quantities. -
This definition is extremely important methodologic-
ally. If a physics problem reflects a physical phenome-
non (or collection of phenomena), it is necessary not on-
ly to have specific knowledge about this phenomenon but
also to know how to analyze any physical phenomenon
by applying generalized knowledge. This analysis begins
with the choice and analysis of a physical system and ends
with the construction of a closed system of equations by
the application of the appropriate physical laws. In
view of this the solution process breaks down into three
stages: the physical (which ends with the construction of
the closed system of equations), the mathematical (de-
signed to obtain a solution in general and numerical
forms), and analysis of this solution.
This, naturally, requires building a system of methods
(discussed in Chapter 2) for solving physics problems as a
system of general guidelines for each of the three stages.
It is believed that no one method for solving all phys-
ics problems exists. This may be true. I believe, how-
ever, there does exist a general approach (system of
methods) to the solution of any physics problem.
In Chapters 3 to 13 the general approach is applied to
the solution of problems in practically all the classical
divisions of the general physics course.
Part 1
THE THEORETICAL BASES
OF THE GENERAL APPROACH TO SOLVING
ANY PHYSICS PROBLEM

Chapter 1
THE SYSTEM OF FUNDAMENTAL CONCEPTS
OF PHYSICS

1. Some General Concepts of Physics

The general approach to solving any physics problem is
based on several fundamental physical concepts. These
are well known, and we mention them in this section only
to note some of their specific features. One central con-
cept is that of a physical system.
A physical system constitutes a collection of physi-
cal objects or even a single physical object.
The solution of any physics problem is related to the
study of a particular system. In what follows we will
see that the choice and investigation of a physical sys-
tem constitute the beginning of an analysis of the phys-
ics of a problem.
The physical objects in a system possess certain physi-
cal properties and can participate in various physical
processes. The common way to characterize the properties
of physical objects and processes is to introduce various
Physical quantities.
Another important concept is that of the state of a phys-
ical system. The general concept of the state of any phys-
ical system is fairly complicated. If a physical system
consists of a single particle, its mechanical state is de-
termined by six quantities: three position coordinates
(z, y, 2) and three components of the particle’s momen-
tum (Pz, Py» Pz):
42 Part 1. General Approach to Solving Any Physics Problem

The objects in a physical system are interrelated

both with each other and with external objects.
This interrelationship manifests itself in the inter-
action of physical objects.
Interaction constitutes one of the most important prop-
erties of any physical objects. It is caused by the inner
nature of the objects. Four basic types of interaction be-
tween elementary particles are known in physics: strong,
electromagnetic, weak, and gravitational. Neither strong
nor weak interactions will be touched on in this book.
Interaction may change either the position of a physi-
cal system or its state.
The change in position or state of a physical system
will be called a physical phenomenon.
Within the framework of the general approach to solv-
ing physics problems this concept is the most important.
Nature knows of a great variety of physical phenomena:
the movements of planets and stars, thunder and light-
ning, rain and wind, water evaporation and dew fall,
the vibration of atoms and the movement of molecules,
the emission and absorption of light, and so on. It is
not so simple to understand even the qualitative aspect
of one or another physical phenomenon.-

It has proved expedient to start analyzing a physical

phenomenon by selecting and investigating the ap-
propriate physical system, since the phenomenon
proper always occurs within a system.

In the process of analyzing the physical objects of the

system it is advisable to establish to which ideal objects
they correspond, what properties they possess, with what
objects and in what manner they may interact, and what
the results and consequences of such interaction may be.
A physical phenomenon is characterized by a change
in certain physical quantities. These are interrelated.
As is known, a necessary and stable interrelation
or dependence between physical quantities is ex-
pressed by a physical law.
 Ch. 1. System of Fundamental Concepts of Physics 13

Every physical law has many facets or angles (neces-

sity, objectivity, physical meaning, and so on). In what
follows we will have special need for two specific aspects
of a physical law: the conditions (or limits) of its ap-
plicability and the method of application.
Every physical law is relative in the sense that it is
valid only in certain conditions.
The set of these limitations will be called the condi-
tions (or limits) of applicability of the given law.
If any one of these conditions is violated, the law cannot
be employed, it becomes invalid. For example, Newton’s
second law written in the form F = ma is valid if the
following conditions are maintained: the motion of an
object is considered in relation to an inertial reference
frame, the object is a particle, the particle mass is con-
stant, the velocity of the object (or particle) is much smal-
ler than the speed of light, etc. If any one of these condi-
tions is violated, Newton’s second law in the above form
cannot be employed.
When solving physics problems, it is not enough to
know the right law (i.e. its physical meaning, conditions
of applicability, etc.): one must also know how to apply
the law in specific conditions.
Every physical law has its method (algorithm)
of application.
Say, to write Newton's second law in the form F =
ma correctly, one must perform the following se-
quence of operations. First, check to see whether the ap-
plicability conditions for the law are met (if any one is
violated, the law cannot be employed). Second, select an
inertial reference frame (the given law is valid only in
such reference frames). Third, find all the forces acting
on the object under investigation (the law incorporates
the physical quantity F, which constitutes the vector
sum of all the forces acting on the object of mass m).
Fourth, determine the projection of each force on the coor-
dinate axes (Newton’s second law is a vector law). Fifth,
sum algebraically the projections of the forces on each
coordinate axis (>) F;, >) Fy, > F,). And finally, sixth,
44 Part 1. General Approach to Solving Any Physics Problem

write Newton’s second law in the form of three equations

py F,=ma,, ay Fy=ma,, py F,=ma,,

where ax, ay, and a, are the projections of acceleration a

on the z, y, and z axes. The methods used in applying
other physical laws will be discussed in the appropriate
chapters.

2. Idealization of a Physics Problem

Suppose that a formulated problem contains the neces-
sary data (whose completeness is ensured) and we must
determine certain unknown physical quantities. But this
is not the main thing in a formulated problem. The most
important aspect is that the problem is already idealized.
The author of the problem has introduced many addition-
al conditions simplifying it. By introducing these con-
ditions and limitations, he has artificially severed the
links of the given physical phenomenon with other phe-
nomena. It is also assumed that the influence of other
(additional) phenomena is smal] and can be ignored. Thus,
a formulated physics problem is a problem concerned
with a pure, idealized phenomenon.
Very often in science the object of investigation is not
a real object but its ideal image. The explanation lies
in the fact that real objects and phenomena are so compli-
cated and interrelated that their study and quantitative
investigation with due account for all aspects, interrela-
tions, and interactions would present insurmountable
mathematical difficulties.
Reasonable idealization of concrete physics problems
constitutes the most important feature of physics as
a natural science.
If physicists did not idealize their problems, they could
Dot solve a single concrete problem in full. Very of-
ten the simplifying assumptions and limitations are for-
mulated in the problem itself, but at times they are
present in the problem in latent or nonexplicit form.
 Ch. 1. System of Fundamental Concepts of Physics 415

Example 2.1. A projectile is fired by a gun at an angle

a = 45° to the horizon with an initial speed vg = 600 m/s.
Find the horizontal distance to which the projectile will be
propelled. Ignore air drag.
The problem is formulated. It is idealized. One addi-
tional condition simplifying the problem (ignore air
drag) is formulated directly. However, many other sim-
plifying conditions are tacitly assumed. It is assumed that
(a) the gun is positioned on Earth;
(b) the motion of Earth about the Sun is not taken in-
to account;
(c) the motion of Earth about its axis is not taken in-
to account;
(d) the vector of acceleration of free fall, g, points in
the same direction at all points of the projectile’s tra-
jectory;
(e) the acceleration of free fal] at Earth’s surface, g,
is assumed constant and equal to 9.8 m/s?; and
(f) the projectile is thought of as a particle.
Conditions (b)-(e) simplify the problem considerably.
Say, if we assume Earth to be a sphere, with the result
that the acceleration of free fall points in different direc-
tions at different points of the projectile’s trajectory,
then condition (d) is not met and the problem becomes
more complicated. If all the additional assumptions are
lifted, the problem becomes extremely complex.
The simplifying assumptions vary from problem to
problem, but
a common feature of all idealizations is the ignor-
ing of nonessential, secondary interrelations and
interactions.

Then what should be the criteria? When, in what con-

ditions, can an interrelation or interaction be ignored
and when not? The answer is closely linked with the
method used in analyzing the solution of a problem and
the estimate method, both of which will be discussed in
detail later.
Two ways of idealization are most commonly used in
analyzing physics problems: the introduction of idealized
physical objects and the neglect of nonessential interactions
16 Part 1. General Approach to Solving Any Physics Problem

and processes. The second approach includes the introduc-

tion of idealized physical processes.
It is important to note that in any idealized physical
object only some of its properties are ignored; the object
possesses these properties, but in the specific conditions
of the problem they manifest themselves so weakly that
they can be ignored.
Physics introduces many idealized objects to use in
the idealization of physics problems. Here are some:
(i) Particle. A fundamental and universal physical
object. In this concept the geometric dimensions of an
object are ignored in comparison with the characteristic
distances of the problem. -
(ii) Rigid body. In this idealized object all possible
strains are ignored.
(iii) Elastic body. Here the remnant strain is ignored.
In some problems remnant strain is so small] that it plays
no role. It must be noted that when elastic bodies in-
teract, there is no transformation of mechanical energy
into other types of energy (i.e. the law of energy conser-
vation in mechanics is observed).
(iv) Inelastic body. Such an idealized object is incapa-
ble of sustaining deformation without permanent change
in size or shape. Elastic deformation is ignored.
Examples of other idealized objects will be given later.
The second approach to idealization involves the in-
troduction of idealized physical processes or the neglect
of nonessential physical processes (phenomena) and in-
teractions. Examples of idealized processes are the iso-
choric, isobaric, isothermal, and adiabatic processes.
Very often, the variation of one or another physical
quantity is ignored in solving a concrete problem on the
assumption that this variation is small. For instance, in
the example with the projectile, despite the well-known
fact that the acceleration of free fall depends on altitude,
we assume that g is constant and equal to 9.8 m/s?.
But since the altitude % reached by the projectile in its
flight is small (several kilometers) in comparison with
Earth's radius (R = 6400 km), it is reasonable to assume
that in the given problem the acceleration of free fall,
g, is constant.
 Ch. 1. System of Fundamental Concepts of Physics 17

Thus, as a consequence of idealization and simpli-

fication physicists consider a schematic model of
the phenomenon rather than a real physical pheno-
menon.
Usually the model of a real physical phenomenon re-
flects the essentials, allows only for the most important
interrelations and interactions, and considers idealized
objects rather than real objects. Very often success in
solving a particular physics problem depends on how well
the model is chosen.
The classification of models of physical phenomena,
obviously, coincides with the classification of
phenomena proper.
Hence, depending on the properties of the physical
system and the conditions in which physical phenomena
occur, we distinguish between two general models: clas-
sical physical phenomena (the classical model) and quan-
tum physical phenomena (the quantum model). This book
considers only classical physical phenomena and the
corresponding models.

3. Classification of Physics Problems

As is known, objects can be classified according to any
of their characteristics, but the best classification is ac-
cording to the essential characteristics. Physics problems
possess many characteristics. To arrive at an optimal clas-
sification we must isolate the essential features of a phys-
ics problem. So what is a physics problem, what are its
essential features? It is also expedient to pose other ques-
tions. How and when does a physics problem emerge?
What is meant by solving a physics problem? What
types of physics problems exist? These questions are not
as simple and unimportant as it might seem.
In studying a physical phenomenon some of the physi-
cal quantities characterizing the phenomenon may be
known while others may not. If a person knows everything
about a physical phenomenon (within the framework of
the phenomenon), no questions or problems arise. These
2—0498
48 Part 1. General Approach to Solving Any Physics Problem

arise when in the process of studying a phenomenon the

researcher finds that for some reason some of the physi-
cal quantities characterizing the phenomenon are un-
known. Hence, a problem is posed (stated, formulated)
by a person studying a concrete physical phenomenon
when some of the interrelations, interactions, physical
quantities, etc., in the phenomenon are not known. The
following definition of a physics problem can, therefore,
be suggested:
A physics problem constitutes a physical phenomenon,
more precisely, a verbal model of this phenomenon
(or collection of phenomena), with some known and
some unknown physical quantities. To solve a phys-
ics problem means to find (reconstruct) the unknown
interrelations, physical quantities, etc.

This definition immediately leads us to two classifi-

cations of physics problems. The first is based on the differ-
ence in the methods of finding the unknown quantities,
and the second allows for the content of the physical
phenomenon reflected in each physics problem.
There are two ways of finding the unknown quantities
in a physical phenomenon: the experimental and the
theoretical. In the first the unknown quantities are deter-
mined through measurements performed in experiments.
In the second these unknown quantities are found by a
physical analysis of the given phenomenon via the phys-
ical laws that control the phenomenon. Physical laws
interrelate different physical quantities, some of which
may be known, others not. If by applying the appropriate
physical laws we arrive at a closed system of equations
whose unknowns include the unknown physical quanti-
ties to be found, solving the system of equations can
solve the problem theoretically.
These two ways of finding the unknown quantities
lead us to the first classification of physics problems. Prob-
lems may be experimental and theoretical.

A problem is said to be experimental if its solution

requires using measurements.
 Ch. 1. System of Fundamental Concepts of Physics 19

Experimental problems will not be considered in this

book (they are usually studied in laboratory courses).
A physics problem is said to be theoretical if it corre-
sponds to a physical phenomenon (or collection of
phenomena) with some known and some unknown
physical quantitities characterizing the phenomenon
and can be solved without resorting to measurements.
This book considers only theoretical physical probleins.
In what follows we drop the adjective “theoretical” for
the sake of brevity.
The classification of theoretical problems will be car-
ried out according to the two most important aspects of
a physics problem formulated above (a problem is al-
ways formulated and solved by a specific person and ex-
presses a certain physical phenomenon). The first aspect
divides all physics problems into formulated (or specified)
and nonspecified.
A problem is said to be nonspecified if the data neces-
sary for its solution are not provided in full (except
tabulated values) or if the necessary idealization has
not been carried out or if both requirements are not
fulfilled.
We will consider nonspecified problems later.
In a formulated problem not only is the complete set
of quantities and their values given but also the
idealization process has been carried out.
Hence, in a way a formulated problem is a “prepared”
problem that always has a solution.
We now turn to the classification of formulated prob-
lems according to the second important aspect: a problem
expresses a certain physical phenomenon. The same
criteria used to classify physical phenomena classify
physics problems:
The type of physics problem corresponds to the type
of physical phenomenon that it expresses.
In short, the problem corresponds to the phenomenon-
The general criterion mentioned earlier enables us to di.
Qe
20 Part 1. General Approach to Solving Any Physics Problem

vide all problems into classical and quantum. Next, each

classical problem (and, of course, each quantum prob-
lem) can be referred, via particular criteria, to the cor-
responding types (right down to the elementary criteria).
But it is hardly reasonable to carry out such a detailed
classification of problems, not only because this would
first require elaborating the entire collection of physical
phenomena (i.e. the entire course of general physics)
but also because a student beginning the study of phys-
ics would find this extremely cumbersome and, appa-
rently, of little use in problem solving. Hence, we re-
strict our discussion to the above-mentioned general clas-
sification according to two generalized features (a for-
mulated and nonspecified, classical and quantum). Note
that although an analysis of a physical system enables
us to determine, often even before the final solution has
been obtained, whether a problem is classical or quan-
tum, what idealized objects and processes have been in-
troduced, what interactions take place, and what their
consequences are, it is sometimes only after the problem
has been solved that we can decide whether it belongs to
the formulated type or the nonspecified.
Another important concept that has proved useful is
that of a basic problem. Each physical phenomenon
can be characterized by a set of physical quantities,
which are related by various physical laws. Among the
many physical laws that govern a given physical phe-
nomenon there is always one or several fundamental laws.
Finding the physical quantities that enter into the
fundamental laws constitutes the essence of the
basic problem of a physical phenomenon.
The next step is to determine, by employing secondary
laws, the entire set of physical quantities that character-
ize the given phenomenon. It can be demonstrated that
any basic problem in the general physics course
consists of finding the state of the corresponding
physical system.
 Ch. 2. Methods for Solving Physics Problems 21

Chapter 2
SOME GENERAL METHODS
FOR SOLVING PHYSICS PROBLEMS

4. Stages in Solving a Formulated Problem

In solving a formulated problem it is useful to distin-

guish between three stages: the physical, the mathematical,
and analysis of solution.
The physical stage begins with acquainting oneself
with the terms (or hypothesis) of the problem and finishes
with setting up a closed system of equations whose un-
knowns include the sought quantities. When the closed
system of equations has been set up, the problem is as-
sumed solved from the standpoint of physics.
The mathematical stage begins with solving the closed
system of equations and finishes with obtaining a numer-
ical answer. This stage can be broken down into two
stages:
(a) finding the solution to the problem in general form;
and
(b) finding the numerical answer to the problem.
Once we have solved the system of equations, we have
found the solution in general form. Performing the neces-
sary calculations, we arrive at the numerical answer.
In the mathematical stage there is almost no physics.
It goes without saying that the mathematical stage is less
important than the physical, but it must be stressed that
it is not of secondary importance. Unfortunately, the
role of this stage is sometimes underestimated. It is
sometimes assumed that it can be ignored. It is also wrong
to assume that mistakes made in the mathematical stage
are secondary. If in solving the system of equations or
transferring to new units of measurements or in the cal-
culations proper a mistake is made, the whole solution
becomes wrong.

From the practical standpoint, a problem is solved

correctly only if correct general and numerical an-
swers are obtained.
22 Part 1. General Approach to Soluing Any Physics Problem

Another reason why it is wrong to consider the mathe-

matical stage of secondary importance is that the analy-
sis-of-solution stage must follow. And this stage cannot
be carried out if the general and numerical answers have
not been obtained. Thus, the final solution to a physics
problem relies to an equal degree on both the physical
stage and the mathematical stage.
After the solution has been obtained in general form
and in the form of a numerical answer, the stage of anal-
ysis begins. In this stage we establish how and on what
physical quantities the found quantity depends, in what
conditions this dependence materializes, etc. In conclu-
sion of the analysis we consider the possibility of formu-
lating and solving other problems obtained by varying
and transforming the terms of the given problem. Some-
times the method of dimensionalities is used to check the
correctness of the general solution. Bear in mind that
this method yields only the necessary indication of the
correctness.
Analysis of the numerical answer often involves the
following procedures:
(i) study of the dimensions of the obtained quantity;
(ii) verification that the obtained numerical answer
is physically meaningful and corresponds to the possible
values of the sought quantity; say, if the velocity of an
object is found to be higher than the speed of light in
empty space (c = 3 X 108 m/s), the answer is, obvious-
ly, erroneous; and
(iii) investigation of the correspondence of the answers
to the terms of the problem if a multivalued answer has
been obtained.
To a certain extent analysis of a problem solution is a
creative process. Hence, the method just described must
not be too rigid and may incorporate a number of other
elements, depending on the hypothesis of the problem.
Analysis of the solution is closely related to the problem
statement method, which will be discussed later.
The system of stages in solving a formulated physics
problem is important not in itself. It is necessary to know
more than these stages when solving concrete problems.
But the special feature of the system of stages is that it
 Ch. 2. Methods for Solving Physics Problems 23

is directly related to the system of methods for solving

physics problems. In each stage the person solving the
problem must act independently in accordance with the
particular stage.
It is often said, in fact, that if you want to learn
to solve physics problems, you must solve them in-
dependently.
This is true, of course. But if the person solving the prob-
lem is not taught the general ways (methods) of problem
solving. he will use the torturous trial-and-error method.
This makes it necessary to have a system of general meth-
ods for carrying out all the stages in the solution of an
arbitrary physics problem, a system that serves as a set
of rules for independent work. IIence, the system of gen-
eral methods must have the following properties:
(a) it must be universal, that is, applicable to the so-
lution of any problem in the general physics course; and
(B) it must encompass all the stages of solution.
After analyzing the ways in which each stage in prob-
lem solving needs to be carried out. we suggest the fol-
lowing system of general methods:
(1) analysis of the physical content of a problem;
(2) application of a physical law;
(3) general-particular methods;
(4) simplification and complication method; and esti-
mate;
(5) analysis of the solution; and
(G6) problem statement.
It must be noted that no method taken alone is univer-
sal. Each has meaning and manifests itself fully only
within the system of methods, and this system does not
always automatically guarantee solution of the problem.
Sometimes a problem can be solved without applying
any method, that is by intuition. But solutions will be
obtained much faster and more often if one acts in accor-
dance with these methods. In short,
the system of general methods is not a dogma but a
guide to independent work in solving a physics prob-
lem, not an instruction but a system of intelligent
advice.
24 Part I. General Approach to Solving Any Physics Problem

There are appropriate methods for realizing each stage

in the problem solving process. The following sections
discuss each method in detail.

5. Method of Analyzing the Physical Content

of a Problem
The solution of any physics problem is primarily a men-
tal process. However, we have no wish to go into the psy-
chological aspects of this process. Let us go directly to
the results.
Any physics problem expresses a physical phenom-
enon or group of phenomena. The relationships _be-
tween the sought and known physical processes lie in-
side this phenomenon. To find these relationships, which
must form a closed system of equations, one must not
only know the essence of the given phenomenon, the sys-
tem of its physical parameters, the laws that govern the
system, and the limits of its applicability, but also how
to isolate all these elements in the given problem.
Practically speaking, analysis of the physics of a
problem is reduced to isolating and analyzing the
physical phenomenon.
But how does one begin to analyze the physical con-
tent of a problem?
The introductory part of the method of analyzing the
physical content of a problem is of an auxiliary nature:
a kind of entry into the world of the physical phenomena
inherent in the problem. Analysis of the phenomena takes
place when the student is getting acquainted with the
problem. After reading the problem, it is advisable to
write down its terms and try to understand the data and
sought quantities and the relationship between them. Next
it is necessary to make a sketch, drawing, or diagram and
mark all the data and quantities on it. A graphical image
makes it possible to picture the physical phenomena con-
tained in the problem.
In the main part of the method we make a concrete
analysis of the physical phenomena. As is known, a phys-
ical phenomenon always has two sides, the qualitative
 Ch. 2. Methods for Solving Physics Problems 25

and the quantitative. Hence, we start the analysis by es-

tablishing the qualitative nature of the phenomenon (in
what respects the phenomenon differs from other phe-
nomena, what its essence is, how it occurs, and so on).
Specifically, we, first, select the appropriate physical
system (what physical objects must be included in the
system); second, determine the qualitative characteris-
tics of these objects (the type of idealized object that
each component in the system represents: particle, rig-
id body, etc.); and, third, determine in which physica]
processes the components of the system participate.
After this we establish the various quantitative inter-
relations between the various physical quantities char-
acterizing the given phenomenon. As noted earlier, the
quantitative interrelations between various physical
quantities are reflected in physical laws. Hence, employ-
ing the appropriate physical laws, we arrive at a closed
system of equations. Setting up such a system completes
the physical solution of the problem.
Thus, the method of analyzing the physical content of
a problem indicates how to start solving a problem and
how and what must be done to solve any formulated phys-
ics problem. Understandably, this method can be ap.
plied only at the physical stage of problem solving-

6. General-Particular Methods. The DI Method

The system of general-particular methods is universal in

the sense that it can be applied to solving problems in
practically all parts of the general physics course. Once
one has mastered a fairly smal] number of general-parti-
cular methods, one can successfully solve practically
any formulated problems.
There are relatively few general-particular methods.
Among these we will consider the following: the kine-
matic, the dynamical, the conservation-law, the method
of calculating physical fields, and the method of differentia-
tion and integration. The first four will be studied in sub-
sequent chapters. Here we examine the last method, the
DI method.
Of great importance to the DI method is the principle
26 Part 1. General Approach to Solving Any Physics Problem

of the limits of applicability of physical laws. As is known,

the content of a physical law is not absolute, and the va-
lidity of a law is restricted to the framework of the appli-
cability limits (or conditions).
Often a physical law can be expanded by changing its
form beyond the limits of applicability via the DI meth-
od. Two principles underlie this method: the principle

L(™)

Figure 6.4

that a law can be represented in differential form, and

the superposition principle (if the quantities entering in-
to the law are additive).
The essence of the DI method lies in the following. Sup-
pose that the physical law in question has the form
K = LM, (6.1)
where K, L, and M are physical quantities, with L =
const being the condition of the law’s applicability.
How should one generalize this law to the case where L
is not a constant but a function of M, that is, L =
L (My?
Let us isolate an interval dM so small that the varia-
tion of M over this interval can be ignored (Figure 6.1).
Thus, in the interval dM we can approximately assume
that LZ is constant (ZL = const) and, hence, the law (6.1)
is valid in this interval (approximately). Then
dK = L (M) aM, (6.2)
where dK is the variation of K over dM.
 Ch. 2. Methods for Solving Physics Problems 27

Employing the superposition principle (summing the

quantities (6.2) over all the intervals of variation of M),
we arrive at an expression for K in the form
Ma,

K=\ L(M)aM, (6.3)

M,
where M, and M, are the initial and final values of M.
Thus, the DI method consists of two parts. In the
first we find the differential (6.2) of the sought quantity.
This is done in the majority of cases by partitioning the
object (mentally) into parts so small that they can be con-
sidered as being particles or by partitioning a large time
interval into time intervals dt so small that in the course
of each interval the process being investigated may be
thought of as approximately uniform (or steady-state).
In the second part of the method we sum (or integrate).
The most difficult aspect here is to choose the correct in-
tegration variable and determine the limits of integration.
To determine the integration variable we must analyze
in detail on what quantities the differential of the sought
quantity depends and what variable is the most impor-
tant. This variable is usually chosen as the integration
variable, and the other variables are expressed as func-
tions of the integration variable. As a result the differ-
ential of the sought quantity assumes the form of a func-
tion of the integration variable. The next step is to de-
termine the limits of integration, which are the upper and
lower values of the integration variable. Evaluation of
the definite integral yields the numerical value of the
sought quantity.
Example 6.1. A thin rod of length 1 =1 m has been
uniformly charged with an electric charge Q = 10-™ C.
Determine the potential of the electric field generated by this
charge at a point A on the rod’s azis at a distanced = 1m
from its end (Figure 6.2). The rod ts placed in a vacuum.
Solution. An answer written in the form mg = Q/4neqd,
from which it follows that p = 9 x 10-° V, is erroneous
since this formula is valid only for the potential of an
electric field generated by a point charge. In our case,
28 Part 1. General Approach to Solving Any Physics Problem

however, charge Q is distributed over an object (the rod)

whose dimensions (J = 1 m) cannot be ignored in compa-
rison with the characteristic distance (d = 1 m) con-
sidered in the problem. Hence, charge Q cannot be con-
sidered a point charge.
Let us employ the DI method. We partition the rod in-
to so many small segments that each can be taken as a
particle. Hence, the charge carried by each segment can
be taken as a point charge. We now take one segment of

Figure 6.2

length dz that lies at a distance zxfrom pointA. The charge

of this segment is a point charge and amounts to dQ =
(Q/l) dz. Charge dQ generates an electric field whose
Potential dp at point A can be calculated according to
the following formula:
_ 42
dp = 4ne€gr * (6.4)
Substituting the value dQ = (Q/l) dz, we arrive at a for-
mula that expresses the sought quantity as a function of
a single variable:
__@ dz
dg = Gael Ties * (6.5)

The first part of the method is complete. Now we must

sum the potentials of the fields generated by all the ele-
mentary charges (all are point charges by construction)
into which the initial charge Q has been divided. The
variable of integration z varies from d = 1mtod+l1=
2m. Integrating (6.5) with respect to x within these
limits, we arrive at the following formula for calculating
the sought quantity:

emt £8p—ahrm (144).

d+l

4ne,lz a 4ieol
 Ch. 2. Methods for Solving Physics Problems 29

Substituting the required numerical values, we find that

op ~ 6.3 x 10-3V.
The differentiation and integration method is univer-
sal and is needed both in the study of theory and espe-
cially in solving physics problems. In mechanics the
method is used to calculate the work performed by a var-
iable force and the moments of inertia; in the study of
physical fields it is used to calculate the field strengths
and potentials of fields generated by extended masses,
nonpoint-like charges, macrocurrents, etc.
The differentiation and integration of functions forms
the mathematical basis of this method. Hence, the method
makes it possible to trace the links existing between phy-
sics and higher mathematics (calculus) when studying
these courses.

7. The Simplification and Complication Method.

The Estimate Method
The combination of these two methods is used in solving
complicated problems and also nonspecified and nonstan-
dard problems. It is widely used to analyze the solution
of a physics problem. At this stage the simplification
and complication method makes it possible to develop
practically any problem into a “block” of more compli-
cated or simpler problems. A typical example is Exam-
ple 11.2 (see Section 11).
The component parts of the simplification and compli-
cation method are two interconnected and opposite proc-
esses: the simplification (idealization, estimate and dis-
carding of secondary phenomena, neglect of unessential
elements, etc.) and complication (consideration of previ-
ously discarded objects, phenomena, and elements, and
complication of the physical system, interrelations, etc.).
The estimate method forms the material basis of these two
processes.
This approach is often employed in analyzing a physi-
cal situation by estimating physical quantities or physical
Phenomena.
An estimate of a physical quantity consists, first, in
Dumerically calculating the order of magnitude of the
30 Part 1. General Approach to Solving Any Physics Problem

physical quantity proper (the order-of-magnitude estimate)

and, second, in comparing similar quantities by their
order of magnitude (comparison of the orders of magni-
tude).
When a quantity depending on other quantities is cal-
culated, the numerical value of each of the component
quantities is represented in the scientific notation (in this
notation numbers are expressed as products consisting of
a number between 1 and 10 multiplied by an appropriate
power of 10). Then the order of magnitude of each term
(if the calculated expression is an algebraic sum) is esti-
mated. The terms with the greatest order are isolated,
while the terms whose orders are lower than the highest-
order terms by at least two are discarded. The exact sig-
nificant digit can be found by using a calculator.
Example 7.1. Suppose that the general solution of a
problem gives the following working equation:

VM (pyT— PaT
1)
Am = RTT, ’
where V =9 1 is the volume occupied by a gas, M =
2 x 10-® kg/mol the molecular weight of the gas,
P, = 52 x 10° Pa the initial gas pressure, T; = 296 K
the initial gas temperature, pp = 5 X 104 Pa the final gas
pressure, T, = 283 K the final gas temperature, R =
8.31 J/mol-K the universal gas constant, and Am
the variation of the gas mass. Estimate the order of magni-
tude of Am.
Solution. We transfer the data into the SI system and,
simultaneously, round off their values and represent
these values in the scientific notation. We have: V ~
10-7 m*, M =2 x 10-% kg/mol, p, ~5 x 10° Pa,
T, ~3 x 10? K, pp = 5 x 104 Pa, T, =3 x 10° K,
and R ~ 8 J/mol-K. These data imply, first, that the
approximate values of the initial and final gas tempera-
tures coincide and, hence, instead of the above formula we
arrive at a simpler expression:

Am pees
x ae VM (pi
— Pe)
 Ch. 2. Methods for Solving Physics Problems 31

Second, the order of magnitude of the final pressure

Pp, = 5 X 104 Pa is much lower than that of the initial
pressure p, = 5 X 10° Pa (by two orders of magnitude)
and, hence, the final pressure can be ignored. The
final formula for the estimate of the order of magnitude
of Am is Am = VMp,/RT, whence
_ 10-7? x2 x 10-9 x5 x 10° a ~9

A more exact but more cumbersome calculation yields

the following value of the sought quantity: Am = 3.8 x
10-? kg.
A rough but quick estimate of the order of magnitude
of a sought quantity is extremely important for the sub-
sequent analysis of the solution.
In comparing physical quantities that depend on
other quantities the first step is to find their ratio
in general form and the second to numerically
calculate the order of magnitude of this ratio.
Example 7.2. Compare the force of gravity F, existing
between two protons and the force of repulsion due to the
proton electric charge, F..
Solution. Let us find the ratio of the two forces:

Fg _. Gm? X Aner?
Fey sor? XQ?
where G ~ 6.7 x 10-1! N-m*/kg? is the universal grav-
itational constant, m ~ 1.67 x 10-7? kg the proton
mass, Q = 1.6 x 10-!® C the proton charge, and 4ne, ~
4.1 x 10-7 F/m. Calculation of this ratio yields
F,/F.. = 7 X 10-57 ~ 10-°*, We see that the force of
gravity existing between two protons is weaker than the
force of electric repulsion by 36 orders of magnitude (the
gravitational interaction is fantastically weak if com-
pared with the electromagnetic interaction).

Example 7.3. Which body attracts the Moon with greater

strength, Earth or the Sun?
Solution. On the basis of Newton's law of gravitation
we can find the ratio of the force of gravity exerted on the
32 Part 1. General Approach to Solving Any Physics Problem

Moon by Earth, Fg, to the force of gravity exerted on the

Moon by the Sun, Fs:
Fe Mer§
Fs _ Msri, '

where Mz ~ 6 x 10*%* kg is Earth’s mass, Ms ~2 X

10°° kg the Sun’s mass, rs @ 1.5 X 101! m the mean
distance from the Moon (Earth) to the Sun, and rg =
4x 10® m the mean distance from the Moon to
Earth. The calculation yields F,g/Fs ~ 3/8. Hence, by
order of magnitude the forces of attraction of the Moon to
Earth and the Sun are equal, but, nevertheless, the Sun
attracts the Moon about two and a half times as strongly
as Earth attracts the Moon. There is nothing paradoxical
in this if one recalls that under the Sun’s attraction the
Moon moves around the Sun while under Earth's at-
traction the Moon moves around Earth.
Estimate of a physical phenomenon amounts, first,
to finding the fundamental law governing the phenom-
enon and, second, to numerically calculating the
order of magnitude of the respective physical quan-
tity.

Estimate problems are often nonspecified.

Example 7.4. Estimate the pressure at Earth's center.
Solution. Statement of the problem. We introduce
some simplifying assumptions. We will think of Earth as
a homogeneous solid ball of radius Rg. The field of grav-
ity generated by a homogeneous ball is equivalent to the
field of gravity generated by a particle of the same mass as
the ball’s and placed at the center of the ball. Any object
of mass m placed on Earth’s surface is attracted to Earth
with a force Fy = GmM,/ Rg and, hence, presses on the
surface with a pressure p = F,/A, where A is the surface
area of contact of object and Earth. If many such ob-
jects are distributed over Earth’s surface in the form of a
thin spherical layer, the pressure of this layer of inass
dm on Earth's surface will be
GMgdm
dp =~ —aaRe
 Ch. 2. Methods for Solving Physics Problems 33

The force of attraction to Earth depends on the distance

to Earth’s center. Hence, the thickness of the spheri-
cal layer must be small in
comparison to this distance.
Each spherical layer presses
on the layers lying under it.
It is now clear that to calcu-
late the pressure at Earth’s
center we must employ the
DI method (see Section 6).
We partition the solid spheri-
cal ball representing Earth
into thin spherical layers and
consider a layer of thickness
dr lying at a distance r from Figure 7.4
Earth's center O (Figure 7.1).
This layer is attracted to the part of Earth lying within
it (the outer part does not act on the layer) with a force
dF, = str? dre X 42r3p
G X 4ar 3,

3r? ,

where p is Earth's mean density. This yields the fol-

lowing formula for the pressure produced by the
layer:
dF g __ 4nGp?r dr
dp=
4ar2 3 7
After integration we get
Rp
an
p= \ dp =-> Gp? (RE—r”’),
r
which is the pressure inside Earth at a distance r from
Earth’s center. At r = 0 we have the pressure at Earth's
center:

p= "= GpPRk.
We estimate the order of magnitude of this quantity (as-
suming that p + 5.5 x 10® kg/m’):
p 1.6 x 104 Pa = 2 x 10" Pa.
3-0498
34 Part 1. General Approach to Soluing Any Physics Problem

As is known, the standard atmospheric pressure amo-

unts to about 10° Pa. Hence. the pressure at Earth's cen-
ter exceeds the standard atmospheric pressure by six or-
ders of magnitude.
The estimate method is often used to compare differ-
ent physical phenomena. Here the fundamental
physical quantities characterizing the phenomena are
estimated.
Example 7.5. A flat conducting contour encompassing an
area A = 1 m?* has an electrical resistance R = 1Q. It is
placed in a homogeneous magnetic field whose induction
varies according to the law B = By, — at?/2A, with By =
140 T and a = 107! T-m?/s?. The plane in which the con-
tour lies is perpendicular to the magnetic induction vector
B. Determine the current flowing in the contour at time
t=1 if the inductance of the contour is L and if at
t = 0 the current in the contour is I = 0.
Solution. Depending on the value of L the concrete
physical phenomena will occur in different ways. We con-
sider two limiting cases.
(1) Inductance L is so small that self-induction can
be ignored. But ignored in comparison with what other
physical phenomenon (or quantity)? We incorporate the
contour and the magnetic field into the physical system.
Because of the variation of the magnetic field, electro-
magnetic induction sets in. Since the induction emf €,; =
at is time-dependent, the induction current is also
time-dependent. Flence, the phenomenon of self-induc-
tion sets in, with the self-induction emf€ ._, = —LZ(d/J/dt)
acting as a characteristic of this phenomenon.
Thus, we are considering the case of L-values so small
that the self-induction emf €,_; can be ignored in com-
parison with the induction emf €;. Then Kirchhoff’s
second rule yields €,=J,R. Since €; = at, we get
I, =at/R, I, =1071A. (7.1)
(2) Inductance L is so large that we cannot ignore self-
induction. This means that the self-induction emf @—_,
is comparable to the induction emf @,. Kirchhoff’s
 Ch 2. Methods for Solving Physics Problems 35

second rule then yields

dl I]
€é,-L7=1R8, or at— sp ai.

The solution to this differential equation that satis-

fies the initial condition (J = 0 at t = 0) is given by the
following function:
at aL 5
I= R —-pr (i—e (R/L)t), (7.2)

The second term on the right-hand side accounts for

self-induction. Suppose that inductance L of the contour
isequal to 1H. Substituting the necessary values into
(7.2), we get J = 0.04 A, which differs considerably from
the value obtained earlier. We see that for large values
of the inductance of the contour one cannot ignore self-
induction.
Note that the above conclusion is valid only for short
time intervals. Equation (7.2) clearly shows that the
role of self-induction diminishes with the passage of time.
For example, at = 100 s we have J = 10 A if self-induc-
tion is ignored and J = 9.9 Aifitis not, that is, the cor-
rection introduced by self-induction amounts to only 4
percent. Thus, for time intervals exceeding 100 s we can
ignore self-induction even for such a large value of the
contour’s inductance (L = 1 H).

8. The Problem Statement Method

This method is employed either when the solution is be-
ing analyzed or, more often, when the problem is being
formulated in the case of a nonspecified problem.
The reader will recall that a nonspecified problem has
been defined as a problem that has yet to be idealized or
a problem with an incomplete (nonclosed)* system of
physical quantities and conditions or as a problem in
which both conditions are present. Ilence, a nonspecified

* Tabulated data are not considered as fixed quantities. Thus,

an idealized problem whose system of fixed (or given) quantities is
incomplete only because it lacks tabulated data does not consti-
tute a nonspecrfied problem.
3*
36 Part 1. General Approach to Solving Any Physics Problem

problem differs from a formulated one in that, first, it

is not idealized and, second, its solution is not unique,
with the result that the problem breaks down into sev-
eral formulated problems.
Often a nonspecified problem contains no concrete da-
ta orit isnot known what quantities must be found or
some additional conditions are missing, and so on.
Hence. the first step (the most important but also
the most difficult) in solving a nonspecified problem
is to state (formulate, specify) the problem properly.
Wheb analyzing a physical phenomenon (the problem
statement method starts with this stage), one must estab-
lish what simplifying assumptions can be introduced,
what can be ignored, what additional conditions can be
introduced, and the like. Above, this procedure was called
the idealization process. After reasonable idealization, it
is necessary to establish what data may be known, and
what can be taken from handbooks, tables, and other
sources. Some data may prove superfluous, while other
data may prove lacking, but this can only be established
after the problem has been solved in general form. Appar-
ently, there is no general method (or algorithm) for carry-
ing out the idealization process in a problem. This is
simply a creative process.
Idealization is followed by formulation (or statement)
of the problem: under such andsuch conditions something
is given and something must be found. This finishes the
first stage in the solution and statement of a nonspecilied
problem. The problem has been stated (formulated).
The next stage is already known: the solution of a for-
mulated problem. We must again analyze the physical
phenomenon in question (now this is done much faster),
set up the closed system of equations, and solve the sys-
tem in general form. Before proceeding with the numeri-
cal calculations, we must check to see whether all the
necessary data for this are present. If not, the lacking
data must be added to the given data or taken from ta-
bles, handbooks, and other sources. Only after introducing
the additional data that guarantee the existence of a
u nique solution to the formulated problem can we assume
 Ch. 2. Methods for Solving Physics Problems 37

that the problem has been properly stated (formulated).

Finally, comes the arithmetic to yield the total solution.
Next, lifting one or several additional conditions (say,
by allowing for friction or by assuming that the given
object is not a particle), we can formulate other problems
and solve them in the above manner.
Thus, a single nonspecified problem may be linked
with a whole group, or “block”, of different physics
problems of varying difficulty.
Example 8.1. An object is placed on a wedge (an inclined
plane). Study the motion of the wedge and the object (Fig-
ure 8.1).
Solution. This is a nonspecified problem. It is not
clear what physical quantities are assumed given or what
must be found, and there
are no additional condi-
tions (where must one look
for the data on the object,
what are their properties,
and so on).
In the first stage of ana-
lyzing a possible physical Figure 8.1
phenomenon let us try to
formulate the problem properly. It is advisable to in-
clude the object and the wedge in the physical system and
assume that all other objects are external to the sys-
tem.
Let us now idealize the problem. To this end we intro-
duce additional conditions and limitations which will
ensure that the problem yet to be formulated will have a
solution. We assume that
(1) the given physical system is placed on Earth;
(2) the friction between the wedge and Earth's surface
is so large that the wedge remains fixed in relation to
Earth;
(3) both object and wedge are rigid bodies, that is, all
possible strains are so small that they can be discarded;
however, the elastic forces generated by these strains will
be accounted for (this condition, for one thing, implies
that the surfaces of the wedge can be considered flat‘;
38 Part 1. General Approach to Solving Any Physics Problem

(4) the height of the wedge is so small that over its en-
tire value we can consider g = 9.8 m/s* = const;
(5) the object is a particle;
(6) the friction between the object and the wedge is
small and can be discarded;
(7) the horizontal face of the wedge is so short that the
sphericity of Earth can be ignored (i.e. the acceleration
of free fall g can be assumed to point everywhere in the
same direction).
Now, introducing these conditions and limitations,
we can formulate (state) the first problem:
A particle of mass m = 1 kg moves along a rigid inclined
plane whose height ish = 10m. The initval velocity of the
object is vy = 0. The angle at the base of the inclined plane
is a = 30°. Determine the time it takes the object to slip
down the plane to the base (or the object's acceleration a
or the velocity v or some other parameter) if there is no fric-
tion between object and wedge. Air resistance is ignored.
The problem has been formulated and, as its solution
will shortly show (the solution is quite simple), has been
formulated correctly. Analysis of this solution shows that
the sought time ¢ depends on height 2 and angle a@ in the
following manner:
1 —

Substitution of the numerical values into this formula

yields t ~3 s.
One formulated problem has been solved. Now, by grad-
ually lifting the restrictions and additional conditions,
more complicated problems can be formulated. Say, if we
lift condition (6), we get a problem on the motion of a
particle that allows for friction. It is advisable to compare
the solution of this second problem with the first solution
(8.1). If we lift condition (5), we have a problem on a ri-
gid body (not a particle) moving along an inclined plane.
But here we must introduce the assumption concerning
the shape of the object (a ball, cylinder, or some other
shape). The solution to the third problemcan be compa-
red with the solutions to the first and second and the emerg-
ing differences can be analyzed (say, why ?¢ is different
for the three cases).
 Ch. 2. Methods for Solving Physics Problems 39

Thus, from a single nonspecified problem we can

obtain a whole set (‘cluster’) of diversified prob-
lems.

9. Another Classification of Formulated Problems

Another classification of formulated problems has also

proved useful. It is based on an extremely important as-
pect of the process of problem solution, namely, the means
hecessary and sufficient for solving a physics problem.
According to this, formulated problems can he classified
as elementary, standard, or nonstandard problems. Here
we give definitions of such problems with little attention
paid to rigor, although more rigorous definitions could
also be given.
An elementary problem is one for whose solution it
is necessary and sufficient to recall and use only one
physical law.
A standard problem is a formulated problem for whose
solution it is necessary and sufficient to employ
only a system of “common” knowledge and “stan-
dard” methods.
The widespread collections of physics problems usual-
ly contain standard problems. Below we give examples of
elementary, standard, and nonstandard problems.
Example 9.1. A direct current I = 1 A flows in a con-
ductor made in the form of a circle of radius R = 0.5 m.
Determine the magnetic induction generated by this current
at the center of the circle. The conductor is placed in a vacu-
um.
Solution. The solution is clear-cut. We need only to
write the Biot-Savart law in integral form for a circular
current:

B=pon ge. or B=4nx< 107 T.

Thus, to solve this problem it is necessary and suff-

cient to employ a concrete law, and the method of apply-
ing this law depends on the form in which the law is writ-
40 Part 1. General Approach to Solving Any Physics Problem

ten. Consequently, this problem is elementary. Some-

times elementary problems are called training or plug-in
problems. Problems of this type do justify this classifi-
cation. They can be called training because they train
the student’s memory; they can be called plug-in because
after writing the necessary law the student need only
substitute (or “plug in”) the various values of the quanti-
ties and carry out the calculations; they can also be cal-
led elementary. We will reserve the last name for such

Figure 9.1

problems. Bearing in mind that elementary problems can

be solved without employing the general approach (al-
though some elements of this approach can be used in
solving such problems), we do not consider them in this
book.
Example 9.2. Two blocks of masses m, and my, are
placed on a plane inclined at an angle « to the horizontal
(Figure 9.1). Determine the force of interaction between the
blocks (which are touching each other) during their motion
if the coefficients of friction between the inclined plane and
the blocks are {, and f,, respectively, with f, > fo.
Solution. This fairly simple problem cannot be solved
by simply writing the appropriate physical law (say,
Newton's second law F = ma) if only because we must
know not only the law but also how to apply it.
Let us employ the method of analyzing the physical con-
tent. After writing the hypothesis of the problem, plot-
 Ch. 2. Methods for Solving Physics Problems 41

ting the necessary diagram, and analyzing the data and

the sought quantities, we get down to the main part of
the physical analysis. We incorporate in the physical sys-
tem the blocks m, and m, and assume that al] other ob-
jects are external to this system. The objects of the sys-
tem may be taken to be particles. These particles move
within the system owing to their interaction with each
other and with external objects (Earth and the inclined
plane). We must determine one parameter of such inter-
action, that is, one of the internal forces. This problem
is linked with the basic problem of particle dynamics. Let
us apply Newton's second law to each object (particle).
We associate an inertial reference frame with the incli-
ned plane and point the axes of the coordinates as shown
in Figure 9.1. Clearly, there are four forces acting on each
of the two objects m, and m,, namely, the force of gravi-
ty mg, the force N exerted on a block by the plane, the
friction force F;,, and the sought force of interaction be-
tween the blocks, F. Projecting these forces on the coordi-
nate axes, we get a closed system of two equations in two
unknowns:
mg sin a — fymyg cosa + F = ma,
Mog sin a — fomog cosa — F = maa.
Solving this system, we arrive at an answer in general
form:
Pa ™m™ (fi—f2) cos a
mtm,

We see that to solve this problem it is necessary and

sufficient to use only Newton's second law, the standard
method of analyzing the physical content of a problem,
and the method of applying the necessary physical law.
Hence the solved problem is a standard one.
A nonstandard problem is also a formulated problem,
but the application of only “ordinary” laws and
methods in the process of its solution does not lead
to the goal because the system of equations proves
not closed.
Something remains unaccounted for (which makes the
problem nonstandard), some feature that eludes imme-
42 Part 1. General Approach to Solving Any Physics Problem

diate detection and requires a wild guess. Obviously,

there can be no general or universal practical advice on
how to guess this feature.
Example 9.3. Two particles of masses m, and m, (with
m, >> ms) are connected by a massless and nonexpandable
string as shown in Figure 9.2. The pulleys are also massless.
Find the force of tensile strength of the string that emerges
when the particles move.
Solution. We employ the method of analyzing the
physical content. After writing out the hypothesis of the
problem, plotting the necessary
diagram, and analyzing the data
and sought quantities, we get down
to the second part of the physical
analysis. The objects system incor-
porates the objects m, and m,
and the string. By the hypothesis,
objects m, and m, are particles and
the thread is massless and nonex-
pandable and thus cannot be tho-
ught of as a particle. As a result of
the interaction of the objects in the
system between themselves and
with external objects (such as Earth)
the objects m, and m, movein a
Figure 9.2
straight line with accelerations a,
and ay, respectively. One of the
dynamical parameters of the problem must be found: the
force of tensile strength (or simply tensile strength) of
the string. This problem is related to the basic problem
of particle dynamics. We apply Newton’s second law
to m, and my:

mge—T=ma,, 2T — mg = med,
with 7 the tensile strength.
Now we have a closed system of two equations in three
unknowns (4a,, a), and 7), The concrete laws of dynamics
have been exhausted, so we employ the concrete laws of
kinematics:
$ = a,07/2, ss, = a,t?/2.
 Ch. 2. Methods for Solving Physics Problems 43

We have thus arrived at an open system of four equa-

tions in six unknowns (a,, a2, T, $;, Se, and ¢). Now the
concrete laws of kinematics have been exhausted, but the
problem still remains unsolved. Something in the hypo-
thesis of the problem has not been accounted for, but we
can only guess what this “something” is. We again analyze
the hypothesis. Why are the accelerations a, and a, differ-
ent? The conditions in which the two particles move are
different. But in what way? Well, different forces act on
m, and mz (this is reflected in the dynamics of the prob-
lem). In what other respect is their motion different? In
the kinematics of the motion; concretely, s, differs from
Sp. But why? Because a, differs from a,. But this is what
we started with. So far logic has proved of no help. And
then suddenly, like a bolt out of the blue, we have the
answer: s, = 2s,. Why? Well, the answer is simple. The
guess is indeed correct, and it can be rigorously proved.
After this the solution is indeed obvious.
To conclude this section we examine a particular case
of nonstandard problems, which we will cal] original, or
challenging, problems.
An original problem is a nonstandard problem in the
solution of which the role of the wild guess becomes
dominant in comparison with ordinary knowledge
and methods.
The role of the latter is usually minor in solving such
problems. The definitions of an original and a strictly
nonstandard problem show that the boundary between
the two is hazy. Sometimes in original problems the un-
definable “something,” or the special feature that evades
detection, becomes the seed out of which new nonstan-
dard methods of problem solving grow. Note that an orig-
inal problem can often be solved by standard methods,
but this is so cumbersome, at times involving complicated
transformations and calculations, that it proves best to
look for another, i.e. original, solution.
Example 9.4. Two motorboats simultaneously leave two
ports A and B separated by a distance l. One has a velocity
v, and the other, a velocity v, (Figure 9.3). The direction in
which the first motorboat travels forms an angle a with the
44 Part 1. General Approach to Solving Any Physics Problem

AB line and that in which the second motorboat travels an

angle B with AB. What will be the smallest distance between
the motorboats?
Solution. We start with a standard solution. We employ
the method of analyzing the physical content. In what
follows this method will be called the method of analysis
for the sake of brevity. The physical system is formed by
the two motorboats, which are assumed to be particles.

Figure 9.3

These move uniformly and rectilinearly with respect to

the inertial reference frame in which Earth is at rest. This
motion is considered only formally. We wish to find one
of the parameters of this motion, the minimal distance
between the particles. The problem is associated with the
basic problem of kinematics. We assume the origin to be
at point A. Since the laws of motion of the particles are
known,
r, =v,tcosa X i+ utsina X j,
r, = (1 — vet cos B) i + vet sin B X j,
we can find the distance between the motorboats at any
moment in time:

r =) {l— (v,cosB+ v, cosa) t]?+ [(v, sin B— y, sin a) ¢]?.

(9.4)
There only remains to find the minimum of this expres-
sion. Here, however, we will encounter some difficult
calculations, but these will have to be carried out to the
end. To simplify the calculations (we must find the de-
 Ch. 2. Methods for Solving Physics Problems 45

rivative r’ and, nullifying it, find the value tin, which is

then substituted into (9.1), and the sought value of
rmin is found) we square r and get
r? = [1 —(v, cosh +, cos @) t]?
+ [(v. sin
B— v, sin @)¢]?.
We find the derivatives of both sides of this equation:
arr’ =2{—[l—(v, cosB + v, cos@) é] (v, cos B+v, cosa)
+(v, sin B — v, sin @)?}.
We exclude the trivial case where r = 0 (this would mean
that the motorboats would eventually collide). Then,
equating r’ with zero, we find the time ¢min at which the
distance between the motorboats is minimal:

ft oe (us cosB+ v1cosa) __

min “~ v2 + v? + 2vyve cos (a+) *

Substituting this expression for tmin into (9.1) and per-

forming tedious calculations (we advise the reader to go
through these calculations independently), we arrive at
the following final expression:
1 (ve sin B—v sin @)
Toin=
V vv}
+20,0, cos (a+)
We now give the original solution. We fix the inertial
reference frame with the first motorboat instead of Earth.
But why? In that respect is the new frame better than the
old frame connected with Earth? Perhaps it is better, per-
haps not. We cannot know in advance. But still] let us
select such a reference frame. Then the second motorboat
moves in relation to this reference frame with a relative
velocity
v=v,—-Yyy (9.2)

and its trajectory is the straight line BC (Figure 9.4).

Obviously, the minimal distance between the motor-
boats is equal to the length of the perpendicular AC
dropped from point A onto the straight line BC:

| AC | =lTsing,
46 Part 1. General Approach to Solving Any Physics Problem

where @ is the angle between BA and vector v. What re-

mains to be found is sin g. Projecting v (see Eq. (9.2))

Figure 9.4

on the Y axis, we get

vsin ¢ = v,sinB — vy, sin a.
By the law of cosines,
v= Vv? + v3 + 2v,v, cos (a +f).
Thus,
sin @= vp sin B—v,sina
V v3 + v3 -+ 20,0, cos (2 +B) ©
Hence, finally,
min = |AC|= 1 (v, sin B—v, sin @)
me V v3 03+ 20,0, cos (a+) ”
which coincides with the expression obtained via tedious
calculations by the standard method.
Part 2
SOLUTION OF STANDARD PROBLEMS

MECHANICS

Chapter 3
THE MOTION OF A PARTICLE

10. Particle Kinematics

In kinematics the motion of objects.is considered in a for-

mal manner, without explaining the reasons for the,var-
iations in motion and, hence, without employing the
concepts of force or mass.

Figure 10.1

The simplest physical system consists of a single par-

ticle or a relatively small number. The position of a par-
ticle with respect to a reference frame at an arbitrary mo-
ment in time ¢ is determined by the radius vector r =
r (t) (Figure 10.1). If we introduce the unit vectors i, j,
and k along the X, Y, and Z axes, respectively, the radi-
48 Part 2. Solution of Standard Problems

us vector r can be represented in the following form:

ri =z(ity(t)j+2z(t)k, (10.1)
with z (t), y (t), and z (t) the components of r (¢). Simul-
taneously specifying three functions, z(t), y (t), and
z (t), is equivalent to specifying a single vector function
r (t) of the scalar independent variable ¢. Equation (10.1)
is known as the law of motion of a particle. Thus, the
law of motion (10.1) specifies the position of the particle
at any moment in time.
The velocity vector v = {v; (t), ”y (t), v, (t)} and
the acceleration vector a = {a, (t), (t), a, (t)} are
defined in terms of the Sheu derivatives thus:
dr dz,
v= ae i+st
a ac ’ (10.2)
dvd,
= ar = ae i+a i+a k. (10.3)
The law of motion (10.1) is fundamental to kinematics.
Knowing the law of motion, one can determine other phys-
ical quantities characterizing the motion of a particle,
say, the components of the velocity vector v and the acce-
leration vector a:

a0 a: vy () = i
4 . % (b=; (10.4)
1?
ay (j= t=
ie ’ ay (=r
4 ’ a, (t) = ree . (10.5)

Hence, the law of motion (10.1) is directly related to the

basic problem of kinematics. Formally there are two such
problems: the direct and the inverse.
The direct problem of kinematics consists in finding
any parameter of motion from the known law of
motion.

It is solved by consistently applying the basic laws of

kinematics (10.1)-(40.3).
The inverse problem of kinematics consists in
determining the law of motion from a known para-
meter of motion (the velocity vector v or the ac-
celeration vector a).
 Ch. 3. The Motion of a Particle - 49

The inverse problem is considerably more complicated

than the direct problem. It can be shown that the great
variety of kinematic problems can be reduced to these
two types. Below we consider several examples of the di-
rect and inverse problems of kinematics.
Example 10.1. Determine the absolute value of the veloc-
ity of a particle at time t = 2s if the particle moves ac-
cording to the law r=at*i+ f (sin at)j, with a =
2 m/s? and B =3 m.
Solution. A physical analysis.* The physical system con-
sists of a single idealized object, a particle. The law of
motien is formally specified. Hence, our problem belongs
to the class of direct problems of kinematics (given the
law of motion. find one of the parameters of motion, in
our case the absolute value of the velocity vector). Using
the known law of motion, we find that the components of
the radius vector r (é) are
z(t) = at’, (10.6)
y (t) = B sin xt, (10.7)
z(t) =0. (10.8)
Thus, the particle moves in the XOY plane. Conse-
quently, each of the vectors r, v, and a has two components.
Combining the definition of the velocity vector
with Eqs. (10.2), (10.4), (10.6), and (10.7), we find the
following expressions for the velocity components:
v, = 2at, vy = Ba cos at.

This yields the following expression for the absolute val-

ue of the velocity vector:
lvl = Vuitoui = V 4020? + Bn? cos? at.
Susbstituting the necessary numerical values, we get
v = 12.4 m/s.
* We will not carry out the introductory part of the physical
analysis in full. Hence, the words “a physical analysis” following
the word “solution” mean that only the main part of the method of
analyzing the physical content of a problem is carried out (choice
and analysis of the physical system, the study of the physicul
phenomenon, etc.).
4-—0498
50 Part 2. Solution of Standard Problems

Example 10.2. A particle moves according to the law

r =a (sin 5t) i -+ B (cos? 5t) j, with a =2 m and
p =3 Mm. Find the velocity vector, acceleration vector,
and trajectory of the particle's motion.
Solution. This is also a direct problem of kinematics.
We start with the components of the radius vector:
z (t) = asin 5t, (10.9)
y (t) = B cos? 52, (10.10)
z(t) =0. (40.11)
Thus, the particle moves in the XOY plane. Next we
find the components of the velocity vector:
vx (t) = 5a cos 5t, (10.12)
v (t) = — 5B sin 102. (10.13)
Equations (10.42) and (10.13) yield the expressions for
the components of the acceleration vector:
a, (t) = —25 a sin 5t, (10.14)
a, (t) =— 50 B cos 102. (10.15)
To find the trajectory we exclude ¢ from the system of
equations (10.9)-(10.10). The result is
y = 3— (3/4) 2. (10.16)
Hence, the particle moves along a parabola.
Example 10.3. The velocity of a particle varies according
to the law v = a (2° — B) i — y (sin 2nt/3) j, with a =
1 m/s‘, B = 158°, and y =1 m/s. Find the law of mo-
tion if at the initial moment t = 0 the particle was at the
origin ry = {0, 0, 0}
Solution. A physical analysis.* The physical system con-
sists only of the particle. Formally the law of variation
of the particle’s velocity and the particle's initial posi-
tion are specified. We are looking for the law of motion
of the particle. Hence, the given problem constitutes an
* In what follows we drop the words “a physical analysis” after
the word “solution.”
 Ch. 3. The Motion of a Particle 51

inverse problem of kinematics (given a single parameter

of motion, the velocity, find the law of motion). The law
of motion r = r (¢) and the velocity vector v are related
through the vector differential equation (10.2), which is
equivalent to three differential equations (10.4). In our
case the velocity components v, (t), vy (t), and v, (t) are
known functions of time:
v,=a(2—6), v= —y(sin2nt/3), v,=0.
Substituting these expressions for v,, vy, and v, into
Eqs. (10.4), we obtain a system of three differential equa-
tions in three unknown functions x (t), y (t), and z (t), the
components of the radius vector r:
_ dz . an , dy _ dz
@ (2 —f)=——, —ysin-> t=4-, 0=7.

Separating the variables and integrating, we find that

a=a(— th—ft) +e, (10.17)

y= ohcosa tt ey (10.18)
2= Cy, (10.19)
where ¢,, ¢,, and cy are arbitrary constants that can be de-
termined from the initial conditions. Allowing for the
fact that x = y = z =0 at t =0 and employing the sys-
tem of equations (10.17)-(10.19), we find that c, = 0,
C, = —3y/2n, and cs = 0. We can now write the final
expressions for the components of the radius vector
r:
x(t) =a(— t—fr) : y= st (cos t—1) , 2=0.
Thus, the law of motion has the form

r(t)=a (-t—Ae) i+ St(cost

1—1) j.
Note that if we now solve the direct problem (given
the law of motion, find the velocity), we can obtain the
starting expression for the velocity vector:
v (t) =a (218— B) i—y (sin 27rt/3) j.
ae
52 Part 2. Solution of Standard Problems

Knowing the law of motion we can find any parameter of

motion: velocity v, acceleration a, the trajectory, etc.
Example 10.4. The acceleration of a particle varies ac-
cording to the law a = af*i — Bj, with a = 3 m/s‘ and
Bp = 3 m/s*. Find the distance from the origin to the point
where the particle will be at time t = 1s if vg = 0 and
ro =0 at t =0.
Solution. The hypothesis of the problem shows that
the particle moves in the XOY plane. To find the distan-
ce from the origin to the point where the particle will be
at ¢ = 1s, we must know the particle’s law of motion.
Thus, we have an inverse problem of kinematics: given a
parameter of motion (in our case this is acceleration a),
determine the law of motion r = r (t) and then find the
absolute value of r at tf= 1s.
We start by determining the velocity vector from Eq.
(10.3):
dv diy. , My,
a=’ or a= ita)

This vector differential equation is equivalent to the fol-

lowing two differential equations:
dux_ op tie
(emma irTeo
Separating the variables and integrating, we arrive at
the following expressions for the components of the ve-
locity vector:
¢3
u.= +e, vu,= —ft+e.
Allowing for the initial conditions (v, =v, =0 at
t = 0), we can find the values of the arbitrary constants:
c, = 0 and c, = 0
Next we turn to the system of differential equations
dz__ sat? dy
3S an Ot
and find the components z (t) and y (t) of the radius vec-
tor r (t):
(= tes, yQ=—Pte, (10.20)
 Ch. 3. The Motion of a Partiele 53

where c, and ¢, are arbitrary constants. By allowing for

the initial conditions (c = y = 0 at t = 0) we find from
Eqs. (10.20) that cs = c, = 0. We have now found the
law of motion:
r(y =e j_- 9 j. (10.21)
If we use the formula for the absolute value of a vector, or
the length of a vector, we can determine the sought dis-
tance from the origin to the particle at ¢ = 1s:
Ir} = Vor+y?,
which yields r ~ 1.52 m.
Analysis of the solution. Knowing the law of motion,
we can find any parameter that characterizes the motion
of a particle and, hence, we can formulate and solve many
other kinematic problems concerned with finding these
parameters. Let us formulate, for example, the problem
of finding the trajectory of a given particle: given the ac-
celeration a = at?i — Bj and the same initial conditions
(these could be different as well), find the trajectory des-
cribed by the particle. After the law of motion (10.21) has
been obtained, the trajectory of the particle can be found
by excluding time ¢ from the system of equations
at 1?
rao yap
The set of methods for solving the direct and inverse
problems of kinematics constitutes the essence of the ki-
nematic method mentioned in Section 6
Very often an arbitrary kinematic problem with real
content can be reduced to the schematized direct and in-
verse problems of kinematics just discussed. Let us de-
monstrate this assertion using a concrete example.
Example 10.5. A train is moving rectilinearly with a ve-
locity vg = 180 km/hr. Suddenly an obstacle appears on
the tracks and the engineer brakes. Now the velocity of the
train changes according to the law v = vy — at’, with
a=1 m’s*. What is the braking distance of the train?
How much time must pass from the moment the brakes are
applied before the train stops?
54 Part 2. Solution of Standard Problems

Solution. The physical system consists of only one ob-

ject, the train, and the train can be thought of as a par-
ticle. The motion of the train is studied formally, with-
out clarifying what caused the change in its state of mo-
tion (the principles on which the brakes operate are as-
sumed to be unknown, and knowledge of these principles is
unimportant to the solution of the problem). We know
the law of variation of one of the parameters of motion,
the velocity. We must find other physical quantities
characterizing the train’s motion (the braking distance
and time). Thus, we have an inverse problem of kinemat-
ics, which can be formulated in the following manner:
given the velocity of a particle as a function of time, v =
(vy — at?) i, with a=1 m’s® and v, = 180 km/hr,
find the time it takes the particle to stop and the distance the
particle travels before it stops if at t =O we have r =0
and vy, = voi (the latter condition follows from the law
of variation of velocity v = (v, — at*) i). When the
problem is formulated in this manner, it is unimportant
what real object is moving: a train, car, motorboat, or
submarine (one only has to change the constant paraine-
ters a and vy).
We solve this inverse problem of kinematics by emplo-
ying the kinematic method. Since the motion of the par-
ticle is one-dimensional (along the X axis), to find the
law of motion we need only one differential equation:
dz =uvudt, or dz = (vg — at*) dt.
After integrating the latter equation and allowing for
the initial conditions, we arrive at the law of motion,
r=vyt — al3/3. (10.22)
The time it takes the train to come to a stop can be cal-
culated by nullifying the train's velocity:
0 =r Vo — at?.

This yields ¢ = Wv,/a. Substitution of the necessary

numerical values yields ¢ +7 s. From Eq. (10.22) we
find the braking distance: z ~ 230 m.
Example 10.6. A rocket is launched from a land-based
site vertically with an acceleration a = at?, witha = 1 m/s‘,
 Ch. 3. The Motion of a Particle 595

At an altitude h, = 100 km above Earth’s surface the rock-

et’s boosters fail. How much time will pass (from the mo-
ment the boosters failed) before the rocket crashes? Air drag
is ignored. The initial velocity of the rocket is vp = 0.
Solution. The rocket can be thought of as a particle.
We know the initial conditions and how the acceleration
varies with time. We must find other physical’character-
istics of the rocket’s motion (velocity, time of motion,
and the law of motion). This constitutes an inverse prob-
lem of kinematics. After integrating the equation dv =
adt and allowing for the initial conditions, we find
the law of time variation of the velocity:
v= ai3/3.
After integrating a second time and allowing once more
for the initial conditions, we arrive at the law of motion
of the rocket:
h=at'/12.
These laws are valid only up to the moment when the
boosters failed. Let us determine the velocity of the rocket
at that moment (this velocity constitutes the initial ve-
locity in the rocket’s further motion):
a eee
_«@ 12h, \3
r= $V | a )‘
This yields vg, ~ 12.1 km/s, which exceeds the escape
velocity of objects Jaunched from Earth, roughly
11.2 km/s. Hence, the rocket will never return to Earth.

11. Particle Dynamics

When we study the dynamics of the motion of objects,
we must introduce a new concept that allows for inter-
actions between the objects. This is the concept of
force, F. But how do we find the forces acting on an ob-
ject?
Let us first establish with what other objects a given
object interacts. Then we can find how the given object
interacts with the others and in what ways (type of inter-
action).
58 Part 2. Solution of Standard Problems

As noted earlier, for classical physical systems the fol-

lowing types of interaction play an important role: the
gravitational (Newton’s law of gravitation F =
Gm,m,/r*) and the electromagnetic (its particular ma-
nifestations are the Coulomb force F = Q,Q,/4ne,er’,
the Lorentz force F = Qu x B, the friction force F;, =
{N, and the elastic force F = —kz). Thus, it is only
as the result of the interaction of a given object with an-
other object that several different forces may act on the
given object. It is impertant to understand that these
forces differ qualitatively. The next step is to evaluate
each force quantitatively, that is, determine the order
of magnitude of each force. It may so happen that some
forces are so weak that they can be ignored in the condi-
tions of a specific problem.
Example 11.1. Two objects of masses m, = 1 kg and
m, = 2 kg are connected by a massless string and move in
the horizontal plane (on Earth) under a force F = 10 N

Figure {1.1

directed horizontally and applied to object m, (Figure 11.1).

Determine the forces acting on each object if the coefficient
of friction between m, and m, and the horizontal surface is
{ = 0.5.
Solution. Consider the object m,. Force F is applied to
it, as the hypothesis states. What are the other forces
acting on m,? Object m, interacts with Earth, the string,
and object m,. With Earth m, interacts through New-
ton’s law of gravitation and, hence. the force of gravity
mg acts on object m, downward. Next, object m, inter+
 Ch. 3. The Motion of a Particle 57

acts with Earth elastically, and this fact manifests it-

self in the appearance of an elastic force N, exerted by
the support (the horizontal plane) on object m,. This
force points upward. Furthermore, as a result of the in-
teraction of object m, with Earth there appears a force of
friction Fr, = f{N,. Finally, object m, can interact with
the string only elastically, that is, there is a tensile for-
ce F; pointing to the left and acting on object m, (since
the string is massless, the gravitational interaction be-
tween m, and string is nil). Object m, can interact with
object m, only via Newton’s law of gravitation, but this
interaction is so weak that in the conditions of the pres-
ent problem it can be ignored. Thus, there are five
forces acting on object m,: F, m,g, N,, Fy,, and Ft.
Reasoning in the same manner, we can demonstrate
that there are four forces acting on object m,: the tensile
force Fi, the force of gravity meg, the elastic force Ne
exerted by the support (the horizontal plane), and the
force of friction Fy, = fNz. Only two elastic tensile
forces F; and F; act in the massless and nonexpandable
string. On the basis of Newton’s second law,
F = ma, (14.1)
we conclude that these forces are equal in magnitude,
F, = F{ (since the string is massless, or m = 0, accord-
ing to Newton’s second law we have for the string
Fi — Fi =0 Xa, or Ft = Fi).
Newton’s second law is the fundamental law of parti-
cle dynamics. It is valid only in an inertial reference
frame for a single object (a particle).
-In the particular case where objects move with speeds
considerably lower than the speed of light in empty space
(v<_c), Newton’s second law can be rewritten as
dv
m= F, (11.2)
or
d?r
m Te =F. (a 1.3)

If Newton’s second law is written for a noninertiat

reference frame, the right-hand sides of Eqs. (11.1)-
(11.3) contain forces of inertia.
38 Part 2. Solution of Standard Problems

The content (or physical meaning) of the fundamental

laws (11.1)-(41.3) lies in the fact that a change in momen-
tum mv or velocity v of a particle is due to and deter-
mined by the action of forces. Hence, if we know the forces
and the initial conditions (the position and velocity of
a particle at the initial moment of time), we can find the
variations in the particle's motion. This constitutes the
basic (ideal) problem of dynamics: given the forces and
initial conditions, determine the change in the motion of
the system (the mechanical state of the system).
To find the change in the tnotion of an object, we must
know the law of motion of the object. Determining the
law of motion from a known parameter of the motion (and
the initial conditions) constitutes, as we have just esta-
blished, the essence of an inverse problein of kinematics.
In dynamics a parameter of the motion of a particle can
be found by consistently applying Newton's second law
to describe the motion of each object in the systein. This
law is written in the form
a= Fim (11.4)
(then the acceleration vector a of each object is determined
and, by solving the inverse problem of kinematics, the
law of motion is established) or in the form (11.2) (then
the velocity vector of each object is determined and, by
solving the inverse problem of kinematics, the law of
motion is established) or in the form (11.3) (then the law
of motion is established directly by solving this differen-
tial equation).
To write Newton’s second law correctly in each specific
case, we must know the method of application of this law.
This method has been discussed with sufficient detail in
Section 1.
Example 11.2. A massless pulley is fastenea to the
top of a wedge whose mass is my = 10 kg (Figure 11.2).
A massless and nonexpandable string is flung over the pulley,
and the ends of the string are tied to blocks of masses
m, = 1 kg and m, = 10 kg. The coefficients of friction
of blocks m, and mz, against the faces of the wedge are
f, = 0.2 and f, = 0.1, respectively, and the coefficient of
friction of the wedge against the horizontal plane is f, = 0.3.
 Ch. 3. The Motion of a Particle 59

The angles formed by the faces of the wedge with the hori-
zontal surface are a, = 30° and a, = 60°, respectively.
Find the tensile stress developed by the string.

Figure 11.2 Figure 11.3

Solution. The problem is complicated. Let us try to

simplify it and then lift the simplifying assumptions one
by one.
Let us assume that (a) f, = 0, (b) a, = 0, (c) a, = 90°,
and (d) fy = oo (the wedge is fixed to the horizontal sur-
face). Then we get a relatively simple problem which
can be formulated as follows: two objects (blocks) of mass
m, = 1kgandm, = 10 kg arefastened to the ends of a mass-
less, nonexpandable string flung over a massless pulley
(Figure 11.3); object m, can move along a smooth, fized
horizontal surface; find the tensile stress developed by the
string.
Let us solve this simplified problem. The physical
system will incorporate four objects: block m,, block m.,
the string, and the pulley. Blocks m, and m, may be
thought of as particles. The mechanical motion of the
objects constitutes the physical phenomenon occurring
in the system. The objects constituting the system in-
teract between themselves and with external objects (the
horizontal plane and Earth). Owing to these forces the
objects of the system (except the pulley) move rectilin-
early and with wniform acceleration. Thus, we have
a basic problem of dynamics. For its solution we employ
Newton’s second law in the form (11.4). This law can
 60 Part 2. Solution of Standard Problems

be applied only to particles m, and m, (the string and

the pulley are not particles). Let us take the support as
the inertial reference frame and send the X and Y axes
as shown in Figure 11.3.
Consider particle m,. The following forces act on it:
the force of gravity m,g (which emerges as the result of
the interaction of mdss m, with Earth through Newton's
““Yaw of gravitation), force N exerted by the support (the
elastic force of interaction of m, with the support, which
is the:horizontal plane). and the tensile stress F, devel-
oped by the string (due to the elastic interaction ef particle
m, anjl the string). The other forces. are weak. The above-
mentioned forces have already been projected onto the |
X and 'Y axes. Hence, we can immediately write New
ton’s second law in the form of two equations in terms
of the projections on the X and Y axes:
MyQyx = Fy, (11.5)

m,a,, = mg — N, (11.6)
where a,, and a,, are the projections of the acceleration
vector a, of particle m, on the X and Y axes. Since a,, = 0,
we have N = mg.
Let us now consider particle m.. Two forces act on
this particle, the force of gravity mog and the tensile
stress F, developed by the string. Figure {1.3 shows that
the projection of these forces on the X axis is zero and
that the algebraic sum of the projections of these forces
on the Y axis is mag — F,. Hence. by Newton's second
law for m, we obtain
MoQqy = meg — Fy, (11.7)
where a., is the projection of the acceleration vector a,
of mz. on the Y axis. It can easily be demonstrated that
the projection of a, on the X axis is nil (a., = 0). Since
Q\x = dg, =a, the system of equations (411.5)-(11.7)
acquires the form
ma = Fi, (11.8)
Moa = Mog — Fy. (11.9)

* Thus, we have set up a closed system of two equations

with two unknowns (a and F,). Physically the problem
 Ch. 3. The Motion of a Particle 61

has been solved, that is, the physical stage of the solu-
tion has been completed.
Solving the derived system of equations (11.8), (11.9),
we arrive at an answer in general form:
ome
Ms
(11.10)

F,= g. (44.14)
Substituting the necessary numerical values, we arrive at
a ~ 8.9 m/s? and F, ~ 8.9 N. We have thus completed
the mathematical stage of the solution.
It is now advisable to go through the last stage of the
solution, the analysis. Formula (11.10) shows that the
acceleration of the system depends on both the value of
m, and the value of mz. Let us consider two limiting
cases: (1) m,>>m., and (2) m, < m,. In the first
@ ~~ gm,/m,, that is, acceleration a is low (a small object
m, pulls an extremely large object m,). In the second
case a ~ g, that is, the system moves thanks to the large
object m, with almost the maximally possible (in the
given case) acceleration equal to g. In the same inanner
we can analyze, via (11.11), the dependence of the ten-
sile stress F, on the values of m, and mg.
Now let us lift the simplifying assumptions. (a) Suppose
that the friction coefficient f, is not zero. Then there
appears an additional force acting on object m,, the
friction force F;, =/f,N, pointing in the negative di-
rection of the X axis. The conditions in which object m,.
operates remain unchanged. Applying Newton's second
law to each object, we arrive at a closed system of equa-
tions:
ma = Fy — fymg, maa = mag — Fy.
Solving the system, we get
a= tg, (44.42)
qa
Fy =" Cg, (11.43)
which yield a ~ 8.74 ms? and F, ~ 10.68 N.
62 Part 2. Solution of Standard Problems

If we compare (11.12) with (11.10) and (11.13) with

(11.11), we see that allowing for friction forces diminishes
the acceleration of the system (by what factor and on
what does this reduction depend?) and increases the ten-
sile stress developed by the string.
(b) Suppose that a, #0 and f; #0. The conditions
in which block m, operates remain unchanged. The
forces acting on m, and mg, are shown in Figure 11.4.

Figure 11.4 Figure 11.5

Allowing for the fact that VN, = m,gcosa, and F;, =

1N, =f,myg cos a, we arrive at the following
closed system of equations:
ma = F, — fymyg cos a, — myg sin a,,
Moa = meg — Fy.
Solving the systein, we get
__ Mg— fm, COS @,— Mm, Sin a
= m+ me 8
F,= mym, (1+ /, cos @-+ sin a)

a my+m,
which yield a ~ 8.32 m/s? and F, ~ 14.9 N. Thus, the
acceleration has further diminished and the tensile stress
has increased.
(c) Suppose now that a, #0, a, ~ 90°, /, 0, and
fo #0. The forces acting on objects m, and mz. are de-
picted in Figure 11.5. Newton’s second law as applied
to objects m, and m, yields
ma= F,— Fi,-—mygsina,. (11.14)
moa = Mog sina, —F,—F, (11.15)
 Ch. 3. The Motion of a Particle 63

Allowing for the fact that N, = mgcosa,, N, =

Meg cos @, Fr = f,N, = fymyg cos a,, and
Fir = foN2. = femeg cos @,, we can write the system of
equations (11.14), (11.15) as follows:
ma = F, — fymyg cos a, — myg sin ay,
m,@ = meg sin a. — fpm2g cos @, — Fy.
Solving this system, we get
__ mg Sin a,— fxm, COS a1 — fam, cosa, —m, Sin Gy
ae g, (14.46)
F.= mm, (sina, + fy +TCaaa g, (14.17)

which yield a ~ 6.62 m/s? and Fy = 13.2 N.

We see from (11.16) and (11.17) that acceleration @
has further diminished and the tensile stress developed
by the string has also diminished in comparison with its
value in case (b). The case (d) (the wedge is not fixed to
the horizontal surface) will be considered later.
Equations (11.16) and (41.17) show that the sought
quantities (acceleration a and tensile stress F,) depend in
a very complicated manner on the other parameters of
the physical system: the masses m, and m,, the angles
a, and @., and the coefficients of friction f, and f/,. This
dependence can be studied analytically.
Knowing one of the kinematic quantities of a physical
system (say, acceleration a), we can arrive at the law
of motion by solving the inverse problem of kinematics.
If the initial velocity of the system is zero, the law has
the form z = z, + at?/2, where zy is the initial position
of one of the two blocks. Hence, one can find the velocity
of any block in the system at an arbitrary time ¢t, the
position of each block in space, and other physical quan-
tities characterizing the objects in the system and the
physical phenomena occurring in it. For instance, we can
find the momenta of blocks m, and m, in the system
(Pp; = ,v,, Po = Mev), the values of the respective
kinetic energies Ey, = m,vj/2 and Ey, = mv2/2, and
So on.
64 Part 2, Solution of Standard Problems

Thus, by solving the basic problem of dynamics (this

amounts to finding the law of variation of one kinematic
quantity (acceleration a (t), velocity v (t), or the radius
vector r (t)), we can determine the mechanical state of
a physical system.
The problem we have just solved can be made still
more complicated by assuming, say, that not two but
three or more blocks are connected by strings, that there
are two or more pulleys. that the wedge has three or
more faces (instead of two) along which the blocks can
move, and so on. In short. we can formulate dozens of
similar problems whose principal idea is the same. It
is important to note that all these can be solved by the
same dynamical method. In Example 11.2 we considered
several problems of varying complexity, but in their
essence they were all the same problem and were solved
by applying a single dynamical method.
Note that the problems solved in Example 11.2 had
one very characteristic feature in common, that is, the
forces acting on the objects in the system were all con-
stant. In all problems of this kind the law of motion can
be formulated in advance: if the motion occurs along the
X axis, the law is r = rp + Uoxt + a,t7/2 (similar equa-
tions can be written for the movements along other axes).
Thus, the motion of objects in this case is uniformly ac-
celerated (or uniformly decelerated).
Let un now consider examples of more complicated
problems involving forces that are not constant.
Example 11.3. A skydiver of massm = 100 kg is making
a delayed drop with an initial speed v» = 0. Find the law
by which the skydiver’s speed varies before the parachute is
opened if the air drag is proportional to the skydiver's speed,
F, = — kv, with k = 20 kg’s.
Solution. The physical system in the given case con-
sists of a single object, the skydiver, which can be con-
sidered as being a particle. The physical phenomenon
studied here is the motion of the particle as a result of
the particle’s interaction with external objects (Earth
and air). We must find one of the kinematic parameters
of the particle's motion, the speed as a function of time.
 Ch. 3. The Motion of a Particle 65

This is a basic problem of dynamics. We apply Newton's

second law (the conditions of applicability of this law
are met). For the inertial reference frame we take Earth
(Figure 11.6). We place the origin at the point where the
skydiver begins the jump, point O, and send the X axis
downward. Since the altitude 2 is small compared to
Earth’s radius, the acceleration of free fall may be as-
sumed constant: g ~ 9.8 m/s? = const. Two forces act
on the skydiver, the force of gravity mg and the air drag
Fy = — kv. Newton’s second law enables us to obtain
a differential equation for the unknown function v (t):

- du — _ d(mg/k—v) _ sk
mg/k—v m dt, or me/k—v os dt.

Integration yields
In (mg/k
—v) = — (k/m) t+. (41.18)
We find the arbitrary constant c by employing the ini-
tial conditions (vy =v» =O at ¢ =0), which yield
c = In (mg/k). Substituting this value of constant c
into Eq. (11.18) and performing relatively simple ma-
nipulations, we find the law of variation of the skydiver’s
speed of fall:
v= SA —en(him)t), (11.19)

Equation (11.19) demonstrates that as ¢ tends to in-

finity the speed tends to its maximal value vipa = meg/k,
which amounts to about 50 m/s. Experience has shown
that it takes the skydiver a relatively short time inter-
val to achieve this speed and after that the skydiver
approaches Earth’s surface uniformly at this speed.
Theoretically, the fall of a skydiver is always accelerated
(the speed grows continuously), but starting at a certain
moment in time the change in the skydiver’s speed can
be ignored and the skydiver can be assumed to be fal-
ling uniformly (at a constant speed).
5—0408
66 Part 2. Solution of Standard Problems

Since the law of variation of the skydiver’s speed is

known, we can find the law of motion of the skydiver by
solving an inverse problem of kinematics:
dz=v(t)dt, z()= {v(t) dl,
bbe. " (41.20)
= t— a ae(1 — e- (A/m)t)|

In finding the law of motion (11.20) we employed the

initial condition, z =0 at t =0, to determine the
arbitrary constant.
Thus, the solution of the problem is complete.

Figure 11.6 Figure 11.7

Now let us make the initial conditions of the problem

more complicated: suppose that rz=y =O at t=0
but the initial velocity has a horizontal component,
vo = {0, vo}. In this case the skydiver’s path of fall
is curvilinear (Figure 11.7). As formerly, two forces
act on the skydiver: the force of gravity mg and the air
drag Fy = — kv. But now the air drag Fy, is directed
along the tangent to the path and, hence, we must allow
for the vector nature of Newton's second law. Projecting
the air drag Fy on the X and Y axes and employing
Newton's second law, we obtain
m Ss -. mg — kv,, (11.21)

(11.22)
 Ch. 3. The Motion of a Particle 67

where v, and v, are the unknown components of the ve-

locity vector v.
Separating the variables in Eqs. (41.24) and (411.22)
and integrating with allowance made for the initial
condition (v, = 0 and v, = vp at t = 0), we get

v= ™ (1 etm, (11.23)
Dy = VeeW im)! (11.24)
Let us find the law of motion of the skydiver. Substitut-
ing the v, and vy values from (11.23) and (11.24) into the
relationships dz = v,dt and dy = v,dt, we obtain two
differential equations for determining the two unknown
functions, the z (¢) and y (t) components of the radius
vector r (¢):

dz= “4 (1 —e-th/m) dt, dy =v,e-*/™ de.

After integrating these equations and allowing for the

initial condition (c = y =O at ¢ = 0), we arrive at
the law of motion of the skydiver in parametric form
in the form of two equations:

c=—# t- EU enim), (11.25)

y=2% (4 — e-wvmnt), (14.26)
The law of motion can, of course, also be written in vector
form:

r (9=[ 7% 1 mE (1 — enmty Tics ("2 (1 —e-cnmy)] j.

Now, knowing the law of motion, we can determine any
parameter characterizing the given mechanical phenom-
enon; for one thing, excluding time ¢ from the system
of equations (11.25), (11.26), we arrive at the equation
describing the trajectory of the skydiver:
mg ; k k
r= - 7 [In la y)+—— y|.
Mvp

Thus, this complicated problem too has been fully solved.

5e
68 Part 2. Solution of Standard Problems

The forces acting on a moving object may depend not

only on the object’s velocity but on time ¢, coordinates
z, y, and 2, etc. Let us consider such a problem.
Example 141.4. The thrust of a braking engine is pro-
portional to time, F= — kt, where k = const. Neglecting
friction (and air drag), calculate the time it will take an
object of mass m, with the braking engine mounted on the
object, to come to a halt. The object's speed just before the
engine was turned on was vy. It ts assumed that the mass of
the engine is much smaller than that of the object.
Solution. The physical system consists of a single
object of mass m, which can be assumed to be a particle.

Figure 11.8

The physical phenomenon occurring in this system

amounts to the object moving as a result of the interac-
tion with other objects. We must find one of the parameters
of the system, the duration of motion, ¢,. The initial
conditions are obvious: v = vg and t=O at t=0.
The trajectory of the object’s motion is a straight line
(i.e. the motion is one-dimensional). The final speed of
the object is zero, or v = 0 at ¢ = t,. Thus, the problem
considered here constitutes a basic problem of dynamics.
Let us use Newton’s second law (its applicability con-
ditions are met). We take Earth to be the inertial re-
ference frame. There are three forces acting on the object
(Figure 11.8): the force of gravity mg, the elastic force
N exerted by the support on the object (these two forces
balance each other), and the thrust F = — kti of the
braking engine (the nature of this force is immaterial
to mechanics). Applying Newton’s second law, we ar-
 Ch. 3. The Motion of a Particle 69

rive at a differential equation for one unknown function

of time ¢, the speed v (¢):
dv(t) _
m= —kt.

Separating the variables, integrating, and allowing for

the initial conditions (v = vg at t = 0), we find the
law of variation of the object’s speed:
V =v, — kt?/2m. (11.27)
If in this equation we set the final speed v equal to zero,
we arrive at an equation for finding the duration of mo-
tion, ¢,:
0 =v, — ktt/2m.
This leads us to the formula for ¢,:

ty = V 2mv,/k. (11.28)

After analyzing the solution, we can formulate other

problems, say, finding the braking distance. To determine
the braking distance z, we must know the law of motion.
The law can be found by solving an inverse problem of
kinematics:
I= U,yt — kt8/6m. (11.29)
Substituting into the law of motion (41.29) the value
of the time ¢, taken from (41.28), we find the braking
distauce:
2 amy, io
y= > % “k

Knowing the law of motion of an object, we can deter-

mine any parameters in a given physical phenomenon.
The problem can easily be made more complicated by
allowing, say, for friction.
We have thus considered several different problems in
particle dynamics. In some the forces were constant,
while in others the forces varied, but the approach to
all was the same: the use of the three Newton's laws of
dynamics (especially the second) to determine one para-
meter of motion (velocity v, acceleration a). The next
70 Part 2. Solution of Standard Problems

step in finding the law of motion usually involved solv-

ing an inverse problem of kinematics.
The combined use of the three laws of Newton (es-
pecially the second law) constitutes the essence of
the dynamical method of solving physics problems.
This method can be generalized so as to incorporate
the case of noninertial reference frames. In Example 11.2
we did not consider case (d). Suppose that the wedge is
not fixed to the support, i.e. f, = co. Now in general
the wedge moves with an acceleration (a,) with respect
to the support (Earth) and no inertial reference frame
can be associated with it. If all the conditions of this
problem are taken into account, the problem becomes
extremely complicated (not in principle but technically).
Therefore, to illustrate the essence of applying the dy-
namical method when a noninertial reference frame is
involved, let us simplify the problem as much as possible.
We assume that friction is absent: f, = f, = fy = 0.
Next, we assume that mz, the string, and the pulley are
also absent and that a, = 90° (the last condition is
unessential: angle a, may have any magnitude). Thus,
we are considering the following problem:
Example 11.5. A particle of mass m, = 1 kg ts placed
on the inclined surface of a smooth wedge of mass m,; = 10 kg.
The wedge can move along a smooth horizontal surface.
The angle a, at the base of the wedge is 30°. Determine the
accelerations of wedge and block.
Solution. Two objects will constitute the physical
system: the particle m, and the wedge m,. A wedge can-
not be thought of as a particle, but in the conditions of
the present problem (the wedge moves rectilinearly)
we can assume approximately that, first, all the forces
applied to the wedge are applied at the wedge’s center
of mass and, second, that Newton's second law can be
applied to the wedge.
The physica! phenomenon occurring in this system
is the motion of the two objects, m, and mg, where par-
ticle m, moves with respect to the wedge, and wedge m,
moves with respect to Earth. We must find the kine-
 Ch. 3. The Motion of a Particle 7

matic characteristics of this phenomenon, the accelera-

tion of particle, a,, with respect to the wedge and the
acceleration of wedge, aj, with respect to Earth. This
constitutes a basic problem of dynamics.
Let us start by studying the motion of the wedge with
respect to an inertial reference frame, Earth. The X and
Y axes are directed as shown in Figure 11.9. Three forces

Figure 11.9 Figure 11.10

act on the wedge: the force of gravity msg (due to the

interaction between wedge and Earth via Newton's law
of gravitation), the force N, exerted by the support on
the wedge (due to the elastic interaction between wedge
and Earth), and the force N, (due to the elastic interac-
tion of particle and wedge). By Newton’s second law,
Myg8g = Msg + N, + N3. (41.30)

Projecting Eq. (11.30) on the coordinate axes, we get

M343, = N, sin a,
MgQy, = mg +- N, cos a, — Ng,
where a3, and a3, are the components of the acceleration
vector a, along the X and Y axes, respectively. Since
a3, = 0 and therefore a3, = a3, we get
m;a, = N, sin a,, (41.34)
0 = m3yg + N, cos a, — Nz. (41.32)
We have an open system of two equations in three un-
knowns: a3, N,, and N35.
To find a closed system of equations let us study the
motion of the particle with respect to the wedge. Since
the wedge moves in an accelerated manner, the reference
72 Part 2. Solution of Standard Problems

frame linked with it is noninertial. We direct the coor-

dinate axes as shown in Figure 11.10. Written in a non-
inortial reference frame, Newton's second law assumes
the form
ma=))F+F,, (44.33)
where the summation means the vector sum of “common”
forces acting on an object of mass m moving with an
acceleration a with respect to a noninortial referonce
frame, and F, is the force of inertia, which in our case
(translational motion) is equal to —m,a3. Three forces
act on the particle: the force of gravity m,g (due to the
interaction between particle and Earth via Newton’s law
of gravitation), the force N, exerted by the wedge on
the particle (due to the elastic interaction between par-
ticle and wedgo), and the force of inertia F,;. By Newton’s
second law (11.33),
ma, = mg + N, +- F;.
Projecting this equation on the coordinate axes, we get
ma,, = mg sin a, + mya; cos a, (11.34)
MQ,y = myg cos a, + maz sina, — N,. (41.35)
Since a,, = 0 and, therefore, a, = @,, from (11.34) and
(11.35) we get
mya, = myg sin a, + m,ay cos a, (11.36)
0 = mg cos a, — may sina, — N,. (41.37)
The system of four equations, (11.31), (11.32), (11.36),
and (11.37), is closed (the unknowns are N,, Nz, a,, and
a;). Solving this system of equations, we find the sought
quantities:
____ MyMsf
COS A, = H m, Cos?a;
Me ms+m,sin?a@, ’ N,= m6 (1 a my+m, ina; ) :
= p si me costa,
a, = g Sin Gy + m,sin@,-+-m,/sina, '
= Mg COS Gy
a as m, sin @,+m,/sin @ ?

which yield N,~ 8.2 N, N; = 105 N, a, = 5.25 m/s’,

and a, ~ 0.41 m/s?.
 Ch. 3. The Motion of a Particle 73

12. Mechanical Oscillations

The most common types of oscillations considered in

the general physics course are free continuous oscillations,
free damped oscillations, and forced oscillations. A char-
acteristic feature of oscillatory motion is that such mo-
tion occurs under the action of variable forces. Conse-
quently, after applying Newton’s second’ law, we are
left with a differential equation (usually not in vector
form since one-dimensional problems are considered
in the majority of cases).
Example 12.1. Suppose that a vertical shaft has been
dug through the center of Earth along Earth's diameter. A
small object of mass m is
lowered into the shaft at xX
Earth's surface and released
without initial speed. De-
termine the object's speed at
the center of Earth. Ignore
air drag.
Solution. The physical
system consists of the 0
object, which can be tho-
ught of asa particle. Earth :
is considered being an
external object. Owing to
Earth’s gravity the object
Moves in an _ accelerated Figure 12.4
manner toward the center.
After the object has passed the center, it proceeds toward
Earth’s surface but in a decelerated manner. Since there
is no air drag, the object reaches Earth's surface at the
other end of the shaft and then, reversing its direction
of motion, again begins to move toward Earth's center
in an accelerated manner. Thus, here the physical phenom-
enon consists in the oscillatory motion of the particle.
Let us apply Newton’s second law. We link the inertial
reference frame with Earth, place the origin at Earth’s
center, and direct the X axis as shown in Figure 12.1.
Let us consider an arbitrary point on the X axis where
74 Part 2. Solution of Standard Problems

the particle is at time £; the distance from Earth’s center

to this point is z. When the particle is at this point, there
is a force of gravity acting on it from the ball of radius z:
Rae. z
(12.4)
where M,, is the mass of this ball. Suppose that Earth’s
average density is
= M
= Way
aR *
where M =6 x 1074 kg is Earth’s mass, and R ~
6400 km _ is’ Earth’s _ radius. Then M, =
(4/3) npz®, and the expression (12.1) for the force
of gravity assumes the form
F, = (4/3) nGpmz.
It can be proved that the force of gravity generated by
the remaining spherical layer of thickness (R — z) is
zero.
Applying Newton’s second law, we arrive at the differ-
ential equation describing the oscillations of the particle:
mit=--F,, or 2-+(4/3)nGpr=0,
which coincides with the differential equation describ-
ing free continuous oscillations if we put
ig = (4/3) nGp.
Thus, a particle dropped into the shaft will perform har-
monic oscillations according to the law of motion
r= 2, Sin (Wot + Gp). (12.2)
The amplitude zp and the initial phase a, can be found
by applying the initial conditions (rx = R and v = z=0
at t = 0):
R = zy sin a, O = ZyWo Cos Ap.
Hence, a = n/2 and z = R. The law of motion (12.2)
assumes the form
z= Rsin (wet + 1/2).
 Ch. 3. The Motion of a Particle 75

Knowing the law of motion, we can now determine

any physical quantity characterizing the given phenom-
enon. Let us find the speed of the particle at Earth’s
center:

vera Rody cos (wot + 1/2).

Since at Earth’s center (the origin) z = 0 and, hence,
sin (Wot + 2/2) = 0, we have cos (wot + 1/2) = 1. The
sought speed is
v=Ro,=VGM/R, v=7.8 km/s.
We see that the speed is equal to the circular-orbit speed
for Earth. The period of oscillations is

T= =n ae, T, ~ 90 min,
which is equal to the period of revolution of an artificial
satellite around Earth in an orbit whose radius is equal
to Earth’s radius.
This problem can be made more complicated if we al-
low for air drag. Let us assume that air drag is propor-
tional to the speed of the particle: Fy = — rz. Then
Newton’s second law yields a differential equation for
the resulting damped oscillations:
‘E+ 262
+ (4/3) nGpz =0,
whose solution is
r= 2,e~* sin (wt + ap),
with
4 r? r
w= VYznGp—
Fy, b= aR.
The initial amplitude z, and the initial phase a) can be
found from the initial conditions (c = R and vg = z=0
at ¢ = 0):
R=z,.sina,, 0= — Ssinay + w cos a.
Hence,
Ig = RV 1+ (6/w)?, oa = tan“! (w/S).
76 Part 2. Solution of Standard Problems

Consequently, here too the dynamical method has led us

to the complete form of the law of motion:
z=RYT-+ (6/0)? 0-*% sin [wt + tan-! (w/d)).
Example 12.2. A chunk of ice in the form of a parallele-
piped of base area A = 1 m? and height H = 0.5 m is
floating in water. The chunk is submerged to a small depth
Zo = 5 cm-and then released. Find the period of oscilla-
tions. Ignore the force of resistance of water.

Solution. The physical system consists of one object,

the chunk of ice. Earth and water are the external objects.
The physical phenomenon here consists in the fact that

Figure 12.2 Figure 12.3

first the chunk was at rest and then it started to oscillate.

In the conditions of the problem we cannot consider the
chunk to be a particle, but it is easy to see that all of
its points behave in the same manner. Hence, to solve
the problem we need only describe the motion of one
point of the chunk, say its center of mass. Let us apply
Newton's second law. For the inertial reference frame
we take the water (it is assumed that it remains static
and that any change in its level due to the submergence
of the ice can be ignored). We place the origin at the
surface of the water and direct the X axis in the manner
shown in Figure 12.2.
Let us consider the state of the ice parallelepiped
prior to submergence. The ice is in a state of equilibrium.
 Ch. 3. The Motion of a Particle V7

Two forces act on it: the force of gravity mg=pycg Vg =

Pice AHg (where Pico = 900 kg/m® is the dens-
ity of the ice) and the buoyancy force F, = pyAhg
(where py, = 10° kg/m® is the density of the water and
h is the depth of submergence in the state of equilib-
rium). Newton's second law yields
PicAHg — pwAhg = 0, (12.3)
whence h = (pice/py) A.
We now wish to know what happens when the ice is
submerged to an additional depth x (where z is arbitrary
but small; see Figure 12.3). As a result of the additional
submergence there appears an additional buoyancy force
F = p,Azg = pyAgz. Combining (12.3) with Newton's
second law, we arrive at the differential equation of free
continuous oscillations:

mz= —pwAgr, or «+ ix = 0 (12.4)

where 7 7
=2_ SCH”
Pw& (12.5)

Equation (12.5) can be used to determine the sought

quantity, the period of oscillations:

Ty = oe aon V/ Pcl, (12.6)

@ Pw8

This formula yields 7, ~ 1.3 s.

Equation (12.6) shows that the period of oscillations
is independent of the base area A of the chunk of ice, which
implies that this quantity is superfluous. The densities
Pice and py must be taken from tables.
From (12.4) and the initial conditions (2 = x») and
z=Oatt = 0) we can find the law of motion:
X= Ly sin (Mot + 1/2).
Thus, the ice oscillates harmonically. In real conditions
the oscillations are, of course, damped. Let us then make
the problem more complicated and allow for the resis-
tance of water. We will also change the initial conditions.
78 Part 2, Solution of Standard Problems

Example 12.3. The wce from the previous example is giv-

en a push downward and thus a speed vy is imparted to
it at the initial moment of time. Determine the speed at an
arbitrary moment of time if the force of resistance of the
water is proportional to the speed of the chunk, F, = — rv,
where r is the proportionality factor.
Solution. Obviously, the ice will perform damped os-
cillations. Applying Newton’s second law, we arrive at
a differential equation describing these oscillations:
mr= —r2z— PwAgz,
or

1+ 262+ ws =0, (12.7)

where 6 =r/2m is the damping factor, and Ww, =
V Pw8/Piceff is the natural frequency.
As is known, the solution to Eq. (12.7) is given by the
function
z= 20-8! sin (wt +p),
where » is the frequency of damped oscillations:
wo = Vo? — &.
The initial amplitude z, and the initial phase a, can
be found from the initial conditions (x = 0 and x = vu,
at t = 0):
= 2%, SING, Vg = — 2,5sin a > zw cosa,
whence a, = 0 and x) = v,/w.
Thus, we have the following law of motion of the
chunk of ice:
1} .
z= —e-5sin wt.
@

From this we can easily proceed to the formula for the

speed of the ice at any arbitrary moment in time:

U=Z =U, (cos wt — > sin wt )eof,

We can easily see that in the conditions of Example

42.3 the base area A is no longer superfluous and plays
an important role in calculations.
 Ch. 3. The Motion of a Particle 79

Example 12.4. The plates of a plane air capacitor are

positioned vertically. A smooth dielectric shaft connects
the plates horizontally, and along this there can slip a
small hollow cylinder of mass
m = 10-° kg attached to a spring U=Upsincot
whose constant isk = 10-' N/m
(Figure 12.4). The cylinder car-
ries an electric charge Q = 10-8C.
An alternating voltage U=
U, sin wt with U, = 10" V is ap-
plied to the plates. Determine at
what frequency » the amplitude
of the oscillations of the cylinder Figure 12.4
will be x, = 1 cm. The distance
between the plates is d =10 cm. Air drag can be ignored.
Solution. The physical system consists of a single object,
the cylinder carrying an electric charge. The variable
electric field existing between the plates drives the cy-
linder, which performs forced oscillations. Applying
Newton's second law, we arrive at a differential equa-
tion for these oscillations:

mz= -- kat © U,sinwt, or r+ ofr = £20 sin ot.

The solution to this equation is given by the function

r= 2 sin at, (12.8)

Qu
I= Taio) ar . (12.9)

The Jaw of motion (12.8) implies that the cylinder oscil-

lates harmonically. The sought frequency can be found
by solving Eq. (12.9):

om ff eeebe
“mary
m 3

which yields o ~ 9.5 rad/s.

We have studied several examples involving mechan-
ical oscillations. All have been solved by the same dy-
80 Part 2, Solution of Standard Problems

namical method. Thus, mechanical oscillation problems

constitute ‘a particular case of the basic problems of dy-
namics.

13. Conservation Laws

In addition to the kinematic and dynamical methods for
solving physics problems there is one more method, pos-
sibly a more important one, the method of applying con-
servation laws. This method is more universal than the
first two. While the use of the kinematic and dynamical
methods is restricted to classical physical systems, the
conservation-law method can be applied to both classical
and quantum systems.
It must be noted, however, that when applied to clas-
sical physical systems the kinematic and dynamical me-
thods are more general than the conservation-law method.
This is especially true of mechanical systems. In prin-
ciple any formulated mechanical problem can be phys-
ically solved via the kinematic and dynamical methods,
or simply the kinematic-dynamical method. The same
cannot be said of the conservation-law method: far from
all mechanical systems can be solved by applying the
conservation-law method. But in the more complicated
systems the conservation-law method sometimes leads
to a result faster than the kinematic-dynamical method.
We have already noted that there is no one universal
method for solving physics problems. What is important
is a system of methods. Therefore, it is meaningless to
contrast one method with another: each has its strong and
weak points. Nature is so diversified in its properties and
manifestations that to reveal the various relationships
in physical phenomena we need an intelligent combina-
tion of various methods. Hence, in solving physics prob-
leins it is also advisable to use a system of methods, in-
cluding the kinematic-dynamical method and the conser-
vation-law method.
The method now to be ccnsiilered is based on a group
of conservation laws. There are quite a number of such
laws in physics. The following four are used for classical
systems: the law of momentum conservation, the law of
 Ch. 3. The Motion of a Particle 81

energy conservation (for mechanical systems its particular

case is employed, the law of energy conservation in me-
chanics), the law of angular momentum conservation, and
the law of electric charge conservation. What is common
to all four laws is the statement that a certain physical
quantity is conserved in certain conditions. If the con-
served physical quantity is denoted by A and the set of
conditions in which the particular law holds true by B,
conservation laws can be formulated in a generalized
manner thus:
if conditions B are met, then A = const, or in
equivalent form, if conditions B are met, then
AA = 0,_where AA is a variation_of A.
In the majority of cases conservation laws are applied
if the objects in the system interact. Three stages must
be distinguished in such interaction: the first is the state
of the objects prior to interaction, the second is the
interaction itself, and the third is the state of the objects
after interaction. The interaction process is unimportant
to conservation laws. What is important is that the
value of the corresponding physical quantity does not
change as a result of such interaction (i.e. its values be-
fore and after the interaction must be equal). Therefore,
the conservation-law method consists of the following
steps:
(1) establish what objects are included,in the physical
system;
(2) check to see whether conditions B are met;
(3) select an inertial reference ‘rame (with respect to
which we will subsequently find the values of A);
(4) find the value of A prior to interaction (we denote
this by A;);
(5) determine the value of A after interaction (we
denote this by Ag);
(6) write the conservation law in the form A, = A,
or in the form AA = 0 (A, — A, = 0); and
(7) if the law is a vector one, it is usually “projected”
onto the coordinate axes; the result is three equations,
A\, = Aes, Ay, = Agy, and A,, = Ag,, which are
equivalent to the vector equation.
6—0498
82 Part 2. Solution of Standard Problems

In this section we consider only two laws: the law of

Momentum conservation and the law of energy conser-
vation in mechanics. The other laws will be discussed later.
Example 13.1. The inelastic collision. Two objects with
masses m, = 2 kg and m, = 3 kg that have been moving
with velocities v. = (3i + 4j) and v, = (—2i -}- 3j) with
respect to a certain inertial reference frame collide inelastic-
ally. Find their velocity v after collision. The effect of other
objects can be wgnored.
Solution. The physical system incorporates two objects,
m, and m,. Since the terms of the problem allow us to
neglect the effect of external objects, the system is closed.
Note that the laws of motion cannot be established (if we
use the kinematic method) unless we know the initial
conditions (i.e. the position of the objects at ¢ = 0). The
physical phenomenon here is the inelastic interaction of
the two objects constituting the closed system. Given the
masses and the velocities of the objects prior to interac-
tion, find the velocity of the objects after interaction.
We apply the momentum conservation law. The pos-
sibility of applying this law has been verified. The iner-
tial reference frame has been selected in the hypothesis.
Let us determine the momentum of each object prior to
interaction and find the vector sum of the momenta:
Py = my,V, -+ mov. Next we find the momentuin of the
system after interaction (as a result of the inelastic col-
lision the objects stick together and move with a common
velocity v): p. = (m, -+ m,) v. By the law of momen-
tum conservation,
mv, + mev. = (m, 4+ m,) v,
whence
v=
my
m, +m, Yi aa
pm Vo:
Projecting this vector equation on the coordinate axes,
we find the components of thedee velocity vector:

vy = —T1_
m, +m, Yer mim
ae Vo,
2x7 VD.
x 0;=U;

my gett
v= Dv. —— m/s.
yo mtm, My t npg
rae ayy "5 /
 Ch. 3. The Motion of a Particle 83

Thus, the two objects will move along the Y axis with
a speed v, = 17/5 m/s.
Sometimes the chosen physical system as a whole is
not closed and, hence, the momentum conservation law
cannot be applied. But the system may prove to be closed
along a certain direction (say, along the X axis); in other
words, the algebraic sum of the projection of external
forces on this direction is zero. Then we can write the
Momentum conservation law (only for this direction)
in the form
Pix = Pex:
Example 13.2. A cart with sand whose combined mass
M is 100 kg is moving in a straight line and uniformly
along a horizontal surface with a speed vo = 3 m/s (Figu-
re 13.1), A ball of mass m = 20 kg falls onto the cart from

Figure 13.1

a height h = 10 m reckoned from the surface of the sand

(the initial speed of the ball is zero). Determine the speed
of the cart-sand-ball system after the interaction. Friction
can be neglected.
Solution. The physical system consists of the cart
with sand (considered as a single object) and the ball.
It is not closed, since prior to interaction Earth’s grav-
ity acted on the ball and this force was not balanced by
any other external force. So, generally speaking, the
Momentum conservation law cannot be applied. But in
6*
84 Part 2. Solution of Standard Problems

the direction in which the cart is moving no external

forces act. Hence, in this direction the momentum con-
servation law can be applied. We link the inertial re-
ference frame with Earth and direct the coordinate axes
as shown in Figure 13.1. The component of the momen-
tum vector p, of the system along the X axis was pyx =
Mv, prior to interaction; the same component after
interaction was po; = (M+ ™m)v, with v the sought
speed. By momentum conservation,
Mv, = (M + m) v,
whence
_ _Mx%

which yields, after we substitute numerical values,

v = 2.5 m/s. Equation (13.1) shows that the sought speed
does not depend on k and, hence, the data on hk is super-
fluous.
We could include Earth as an object comprising the
physical system. Then the system of the three objects will
be closed. Since Earth now belongs to the physical sys-
tem and owing to the force of gravity generated by the
ball will move in an accelerated manner with respect to
an inertial reference frame, we cannot link an inertial
reference frame with Earth, strictly speaking. But it can
easily be demonstrated that Earth’s speed and accelera-
tion (in conditions of this and similar problems, where
the masses of objects are extremely small if compared
to Earth’s mass) at any moment in time are so small that
they can be ignored and Earth can be considered as a
fixed object.
Let us find, for instance, the speed that Earth will
gain as a result of interaction with the ball (the maximal
value of the speed which Earth could gain in the condi-
tions of the problem). Very often in physics an inertial
reference frame is selected as linked to the center of mass
or center of inertia of the system. In what follows we
denote this reference frame as the center-of-mass (CM)
reference frame, or the center-of-inertia (CI) reference
frame. By the center of mass of a system we mean the
point whose radius vector rcy is determined from the
 Ch. 3. The Motion of a Particle 85

equation

Piha (13.2)
Tom =
Sm
It can be demonstrated that the CM of a system moves like
a particle whose mass is equal to the total mass of the system,
while the acting force is equal to the geometric sum of all the
external forces acting on the system (the theorem on the
motion of center of mass). Let us write the equation of
motion of the center of mass:
dV,
m Ti = > F,,

where m = >)m, is the total mass of the system, Vou

the velocity vector of the CM, and >)F, the vector sum
of the external forces.
If the physical system is closed, then >yF; = 0 and
Vom = const, that is, the center of mass of a closed sys-
tem moves uniformly and in a straight line. Hence, the
reference frame linked with the center of mass of such
a system is inertial. Since in the CM reference frame the
origin coincides with the center of mass, rom = 0, and
(13.2) yields
> mr, =0. (13.3)
Differentiating (13.3) with respect to time ¢, we get
> m,v,=0, (13.4)
that is, the momentum of a closed system with respect
to the CM reference frame is zero at any moment in time.
Let us apply this result to calculate the increase in
Earth's speed as the result of interaction between Earth
and the ball (Figure 13.2; the origin of the CM reference
frame, O, is shifted slightly to the right). Equation (13.4)
yields
Mog — MVpal = 0, (13.5)

where M is Earth's mass, vg Earth’s speed (more exactly,

the increment of the speed due to the interaction between
Earth and the ball), m the mass of the ball, and vyan
86 Part 2. Solution of Standard Problems

the speed of the ball. Equation (13.5) can be used to

find the sought speed:
vg = a Vpal] = 7 V 28h, VEY 5 x 40-23 m/s.

This speed is fantastically small. Moving at such a speed,

it would take Earth 6 x 10!? years to shift by a distance
of 1 cm. In what follows, when studying the motion of

Yat x

Figure 13.2

objects whose mass is smal] compared with that of Earth,

we ignore the interaction between Earth and such objects
and assume Earth to be stationary.
The law of energy conservation in mechanics is linked
to the notions of kinetic, Ey, and potential, Ey, energies.
Another extremely important concept is that of work,
W. As is known, a force F performs an amount of work
over an elementary displacement dr equal to
dW = F-dr. (13.6)
The work performed by force F over a path J is expressed
by the integral
W = | F.dr, (13.7)
t
where the integral is evaluated along curve J.
There are cases where we have to know the work per-
formed in rectilinear motion. Since dr = idz + jdy +
kdz, we can represent (13.6) in the form
dW = F.i dz + F-j dy + F-k dz
= F dx cos a, + F dy cos a, + F dz cosa,
 Ch. 3. The Motion of a Particle 87

where @,, @, and @, are the angles that the force (vector)
F forms, respectively, with the unit vectors i, j, and k
pointing along the X, Y, and Z axes. If the motion is
along a straight line (say, the X axis),
dW = F drcos a.
The work performed by force F along the segment from
x, to 2, in this case is determined by the formula
Xs

W-= (F cosadz.
%1

If the force is constant, we have no difficulty in cal-

culating the work. Often the DI method is used when a
calculation is done of the work performed by a variable
force (see Section 6). Let us restrict our discussion to the
rectilinear case and assume that | cosa |= 1. A force
may depend on the z coordinate (in the general case, on
the y and z coordinates also), on the components of the
velocity v, = v (in the general case on the other compo-
nents of v as well), and on time ¢.
Let us start with the case where the force depends on
position, or F = F (x). The elementary work done by
such a force is
dW = F (x) dz.
The work perforined on the segment from x, to zy is
As

W = \ F(z)
dz.
7]

Example 13.3. Furst an object is lifted from the bottom

of a shaft of depth h, == R/2 (Ris Earth’s radius) to Earth's
surface and then it is lifted still higher, to an altitude h, =
h, = R/2 above Earth’s surface. In which zase is the
work done greater?
Solution. We can easily see that this problem involves
an estimate. To answer the question posed by the prob-
lem, let us find the ratio W,/W,, where W, is the work
done in the first case and W, in the second. In both the
work is done against the force of gravity, but the laws
describing these forces are different. Exampe 12.1 showed
88 Part 2. Solution of Standard Problems

that the force of gravity in the first case is

F, = (4/3) nGpmz,
while in the second it is
F,=GmM/z*.
The variation of these forces is illustrated in Figure 13.3.
Thus, the forces are variable, and to calculate W, and

Figure 13.3

W,, we must apply the DI method. The elementary work

done over dz are
dW, = F,(z)dz and dW, = F, (z) dz.
Upon integrating within the appropriate limits we get
R
W,= ( + nGomadz = 3 So F
R/2

W,= \ oeM da= GmM

baa i.
Se m

R
and, hence,
W,/W, = 9/8, i e. W 4 > W,.

A force may depend on a component of the velocity,

v, = v. When calculating the work done by such a force
(or against it), we must find the law of variation of v
with time #, that is, solve a basic problem of dynamics by
employing Newton's second law. In this case the ele-
mentary work done by the force is
dW = F (v) dz = F (v) v de. (13.8)
 Ch. 8. The Motion of a Particle 89

By Newton’s second law,

mS? =F(vy)+ DF, (13.9)

where 2 F, stands for the algebraic sum of the projections
on the direction of motion of other forces acting on the
given object. Solving Eq. (13.9) and allowing for the
initial conditions, we find the law of speed variation
with time: v = v (t). Substitution of this law into (13.8)
and integration yield a formula for work:
t;
w=\ Fv(yae. (13.10)
t

Example 13.4. Calculate the air drag on the skydiver of

Example 11.3 in the first three seconds and in the first thirty
seconds.
Solution. Since air drag depends on speed (Fy = — kv)
and the law (41.19) of speed variation has been found,
formula (13.10) gives us the amount of work done against
air drag:
t a
w= |wiv
(tydt= 3 ((1—e- mye ar
0 0
= mg 2m,
[e+e chimytmi,
—1)—3 (@-2him=t— 4),
k
(13.41)
Substituting the values t, = 3s and t, = 30s, we get
W,210'J and W, x 1.1 x 10°J,
Next we find the ratio of the work:
W/W, = 110.
Hence, the work done against air drag increased by a factor
of 110 when time increased only 10-fold. This can be ex-
plained by the growth in air drag and the speed of fall.
In conclusion we investigate the case where the force
is time-dependent: F = F (t). Here, too. to find the law
of variation of speed on time, v = v (t), we must solve
90 Part 2. Solution of Standard Problems

a basic problem of dynamics. The elementary work done

by the force is
dW = F(t) dz = F(t) v (2) dt.
After finding the law of variation of the speed, we can
write the formula for the work in the following form:
te

W= |F()v(t) at. (13.42)

t

Example 13.5. Determine the work done by the thrust

of the braking engine inExample 11.4 in the first second.
Solution. Since the thrust of the braking engine depends
on time (F = kt), the law (411.27) of speed variation
has been found, and duration of braking is known, we
apply formula (13.12) and get

w= (ke (»— ke?

5) dt= kugt®
ee2
t

Ww2.5 x 1083. (13.13)

Sometimes the work can be calculated by using the
theorem on the change of the kinetic energy of a physical
system consisting of particles. By this theorem the work
of all the forces acting on such a system is equal to the
variation in the kinetic energy of the system:
W = AEy. (13.14)
In Example 11.4, two of the three forces acting on the
object balance each other. The remaining (unbalanced)
force is the thrust produced by the braking engine whose
work we have set out to find. Hence, W in (13.14) is
the work done by the thrust produced by the braking
engine, and AE, = mv3/2 — mv*/2. Applying formula
(13.14) and allo ing for the law (11.27) of speed varia-
tion, we get
w= mz
5 m ( ate
kt? )’= kt? = ts
= = (% =
2m 2 8m?

W = 2.5 x 10°J,
 Ch. 3. The Motion of a Particle 91

which coincides with the result obtained earlier via the

DI method (see (13.13)).
By the law of energy conservation in mechanics, the
total mechanical energy of a closed system in which only
conservative forces act is a constant:
E = Ey + E, =const, or A (Ex + E,) = 0. (13.45)
If there are nonconservative forces in a closed system, the
system’s total mechanical energy is not conserved and
its variation is equalto the work done by the noncon-
servative forces:
AE = Was, (13.16)
where W,,, is the work done by the dissipative (or non-
conservative) forces.
Example 13.6. Determine the velocity that a meteorite
of mass m has at a distance r = 1.5 X 10" m from the Sun
(mass M) if at infinity it had a zero velocity and is moving
toward the Sun. The effect of all other objects can be neg-
lected.
Solution. The physical system consists of two bodies,
the meteorite and the Sun. The meteorite can be thought
of as being a particle, while the Sun is assumed to be a
solid ball of radius R = 7 x 10° km. The physical phe-
nomenon associated with this system consists in the
meteorite moving toward the Sun under the Sun’s gravity.
Given the initial state of the physical system, we must
determine one of the parameters of the meteorite’s motion
(the velocity v) in the finite state. This constitutes a basic
problem of dynamics, which could be solved via the dy-
namical method by applying Newton’s second law. But
here there is no need to find the v vs. t dependence, since
what we are looking for is the value of v in the final state;
in other words, there is no need to describe the entire pro-
cess of the meteorite’s motion. Hence, it is advisable to
employ the law of energy conservation in mechanics.
The chosen system of bodies is closed (by hypothesis
the effect of other bodies can be ignored). Only conser-
vative gravitational forces act in the system. We select
the inertial reference frame as the one linked with the
92 Part 2. Solution of Standard Problems

Sun (we assume the Sun to be fixed; see Example 13.2).

The total mechanical energy £, of the system at the be-
ginning of the interaction of the bodies is zero (their ki-
netic energies are zero, and assuming the initial position of
the system to be the zero position we can set the initial
potential energy at zero, too). Let us determine the total
M wes

dr

Figure 13.4

mechanical energy of the system at the end of the inter-

action, E,, when the meteorite is at a distance
r= 1.5 x 10" m from the Sun (Figure 13.4). It consists
of the meteorite’s kinetic energy E, = mv*/2 and its po-
tential energy. The latter is determined by the work
performed by the force of gravity in the course of the me-
teorite’s movement from the final to the initial position.
Since the force of gravity depends on distance r, that
is, it constitutes a variable force, we can employ the DI
method to calculate the work performed by this force.
Let us divide the entire path of the meteorite into inter-
vals so small that in each such interval of length dr we
can ignore the variation in force of gravity, assuming it
to be constant. Then the elementary work on such an in-
terval is
mM mM
dW =G r
—,- cosadr= ~G— r
dr.

Summing the elementary work done on each interval, we

get the total work W, which gives the value of the mutual
potential energy of the system E,:

W=E,=— \G ne dr= —GmM (—-) ———

r
 Ch. 3. The Motion of a Particle 93

Thus, the total mechanical energy of the system in

the initial state, Z,, is zero, and in the final state E, =
mv’/2 — GmM/r. By the law of energy conservation
in mechanics,
= mv? GmM
2 r
Solving this equation for v, we find the sought velocity:
v=V2GMi/r, vx 42.2kmis,
where the values of the gravitational constant G and the
Sun’s mass M were taken from tables.
Example 13.7. A small steel cube of mass M =1 kg
is at rest on a horizontal surface. A small steel ball of mass

iA

my yO iss
( \——> <=s7 2

9 Y x

Figure 13.5

m = 10 g is flying at the cube with a velocity v, = 10° m/s,

hits it, and bounces off elastically in the opposite direction
(see Figure 13.5). Find the distance that the cube will travel
before it stops if the coefficient of friction between the hori-
zontal surface and the cube is k = 0.2.
Solution. Two bodies constitute the physical system:
the cube and the ball. These may be considered as parti-
cles. Earth is taken as an external object. The physical
phenomenon consists in the elastic interaction of ball and
cube (the interaction with the external object, Earth, can
be ignored in their subsequent motion. The initial state
ef the system (prior to interaction) is known. What is
sought is one of the parameters of the cube’s motion
(the distance the cube travels after impact).
Since the forces that emerge as the result of interac-
tion of cube and ball are unknown, there is no way in
which we can employ the dynamical method to describe
the process. Let us apply the laws of momentum and
54 Part 2. Solution of Standard Problems

energy conservation in mechanics. On the whole the

system is not closed, but in the direction in which the
ball moves it can be considered closed. We select the
inertial reference frame linked with Earth and point
the X axis as shown in Figure 13.5. Prior to the interac-
tion the momentum of the system is p, = mv,. The mo-
mentum of the system after interaction is p, = mv, +
Mu,,, where v, and u,, are the velocity vectors of
the cube and the ball, respectively, after interaction.
By momentum conservation,
mv, = mv, + Mu).
Projecting this vector equation on the X axis, we get
mv, = — mv, + Muy).
By energy conservation in mechanics,

mv}3/2 = mv3/2+ Mu},/2.

Allowing for the fact that m< M and solving the above
system of equations, we find that
se ___2m m~ _2m
bo Vy, Ya Wm ~~ 1

Let us now study the motion of the cube after impact.

The statement of the new problem is obvious: a cube of
mass M = 1 kg having an initial velocity u,, skids along
the horizontal surface (the friction coefficient f is equal to
2 X 107") and finally stops; find the distance it has traveled.
Two bodies constitute the physical system, the cube
and Earth. The physical phenomenon consists in the
cube moving in a decelerated manner as a result of its
interaction with Earth. The initial configuration of the
system is known. We must find one of the parameters of
the cube’s motion (distance of travel). This constitute
a basic problem of dynamics. Since the forces of interac-
tion of the cube with Earth are known, the problem can
be solved either by the dynamical method or by the con-
servation-law method. Applying Newton’s second law,
Ma = {Mg,
 Ch. 3. The Motion of a Parttele 95
2

we find the acceleration. Solving an inverse problem of

kinematics, we find the distance J, the cube traveled
before it stopped:
I, = 2 2m 1, ~ 100 m. (43.17)
1" 2fg fgM??
Let_us take the alternative approach. We wish to solve
the problem via the conservation-law method (in me-
chanics). The selected system is closed, but the law of
energy conservation in the form (13.15) cannot be ap-
plied (there is a nonconservative friction force
F,, = {Mg). Assuming that the internal nonconservative
friction force is an external one, we find that Eq. (13.16)
yields
eet ={Megl,.

This leads us to a result that coincides with (13.17),

which was arrived at via the dynamical method:
ui, = 2m*vi a
= Sig = jeu? Ll, ~ 100 m.

The solved problem could have also been formulated

in, say, the following manner (a situation problem):
as a result of what interaction of the ball and the cube will
the length of the path traveled by the cube be maximal?
Let us change the terms of Example 13.7, namely,
we assume the cube to be an wnelastic body, with all other
conditions remaining unchanged. Find the distance that
the cube travels before it stops.
The interaction process (inelastic collision) is des-
cribed by the law of momentum conservation:
mv, =(m + M)ue.
This gives us the initial velocity of the cube (we allow
for the fact that m<M):
Ugo = mv,/M.
Solving the dynamical problem of the cube’s motion
after impact (by either of the two methods), we find the
distance of travel:
ly = Use mvt
96 Part 2. Solution of Standard Problems

Equations (13.17) and (13.18) show that in the case

of an elastic collision the distance traveled by the cube
after impact is four times the distance in the case of an
inelastic collision.
Let us change the terms of Example 13.7: suppose that
as the result of collision with the cube the ball pierces the
cube and continues its motion in the same direction with
a velocity v, = 500 m/s, while all other conditions remain
the same. Find the distance the cube travels before stopping.
We again apply the momentum conservation law and
get
mv, = mv, s Murs,

which yields the initial velocity of the cube at

Uog =m (v, — v,)/M.

The distance the cube travels before stopping is

l=
— Gs _ m? (vy —v9)?
3e = —djeM?
oe
Ll, ey 6.25 m.

Thus, in the case of an elastic collision (with m< M)

both the velocity u., and the distance /, traveled by the
cube before stopping are the maximum possible values.
The dependence of the initial velocity and the distance
traveled by the cube before stopping on the m/M ratio
can also be studied. For instance, for any mass ratio
m/M (but m < M) we have
Up4/Ug9
==2
and, hence,
L/l,=4

Chapter 4
THE MOTION OF A RIGID BODY

14, Rigid-Body Dynamics

The acceleration acy of the center of mass of a rigid body
is determined by the theorem on the motion of the center
of mass:
macm = >; F, (14.1)
 Ch. 4. The Motion of a Rigid Body 97

where m is the mass, and >)F stands for the vector sum
of all the external forces acting on the body.
The form of Eq. (14.1) coincides with Newton’s sec-
ond law for a particle, (11.4), and, hence, the method
of applying this law consists of the same operations. The
vector equation (14.1) is equivalent to the following three
equations:

Macmx = py Fy, Macmy = py Fi, magn:= » F,. (14.2)

For a particle and, hence, for a rigid body we have the

following equation of motion:

L=> M, (14.3)
where L = dL/dt (vector L is defined below), and =M is
the vector sum of the moments of the external forces
Ghee moments are also called torques) about a fixed point

If point O is taken as the origin of a Cartesian coordi-

nate system, then, as usual the vector equation (14.3)
is equivalent to the following three equations:
oe
aL = > ,, =
dL MM, ZH
dL,
> ™., (14.4)

where L,, L,, and L, are the projections of the angular

Momentum vector Lon the coordinate axes. They are
known as the angular momenta of a rigid body about the
fized X, Y, and Z azes, respectively. It can be shown that
for a particle and a rigid body the following equations
hold true:
L,=J,0,, Ly=J,o,, L,= J,0,, (14.5)

where J,, Jy, and J, are the moments of inertia of a

particle or rigid body about the X, Y, and Z axes, re-
spectively, and ,, @,, and w, are the projections of the
angular velocity o on the same axes.
If we allow for (14.5), we can rewrite Eqs. (14.4) as

S00)
A(Jx0x)
= M,, Sy)
d(Jy@y)
—y,, 102°) cay,
d(J20,)__
(14.0)
7~-0498
98 Part 2. Solution of Standard Problems

If the moments of inertia J,, J,, and J, are constant, the

equations of motion assume the form
J.$.=,, J,by =, JA,= M,, (14.7)

where Px = do,/dt, Bp, = dw,/dt, and Bp, = do,/dt are

the projections of the angular momentum vector B on
the,coordinate axes. These equations are known as the
equations of motion with respect to the fized X, Y, and Z
axes, respectively.
A rigid body has six degrees of freedom; hence, we
need six independent equations to describe its motion.
These are either the two
vector equations (14.1) and
(14.3) or the equivalent
system of six equations
(14.2) and (14.6). The meth-
od of applying the laws
(14.2) differ in no way
from that of applying New-
ton’s second law. The meth-
od of applying the laws
(14.6) also closely resem-
Figure 14.4 bles the method of apply-
ing Newton’s second law
if two additional operations are introduced: finding the
moments of inertia of bodies and the moment (or torque)
of the external forces about the appropriate axes. Thus,
for a rigid body the dynamical method remains practical-
ly the same as for a particle.
Example 14.1. Let us consider the simplified version
of Example 11.2 and assume that the pulley is a solid cy-
linder of radius R = 10 cm with a mass m, = 8 kg. Deter-
mine the system's acceleration and the tensile stress de-
veloped by the string.
Solution. The same four bodies constitute the physical
system: the two blocks with masses m, and mz, the
string, and the pulley (Figure 14.1). But now the pulley
is not only a body whose mass must be taken into account.
We must allow for its dimensions, i.e. we cannot as-
sume that it is a particle any more. One assumption is
 Ch. 4. The Motion of a Rigid Body 99

that the pulley is a rigid body. Its center of mass is

fixed, and the pulley rotates about a fixed axis, the Z
axis, which passes through the pulley’s center of mass.
We apply Eq. (14.7) with respect to the Z axis to the
pulley. An inertial reference frame has been selected. Two
unequal tensile stresses act on the pulley: Fi 3 Ft.
All the other forces acting on the pulley compensate
each other. The moments of forces Fi and F¢ about the
Z axis are M; = — FiR and M; = FiR. The moment
of inertia of the pulley (a solid cylinder) about the same
axis is J, = (4/2) m,R?. In what follows the subscripts
standing for the axes at the moments of forces, angular
momenta, and other quantities will be dropped. Em-
ploying the equation of motion (14.7), we get
(1/2) m, RB = (Fi — Fi) R.
Applying Newton’s second jaw to particles m, and me,
we find that
ma=Fi,, ma=ma,g— Fi.
The equation that links the linear acceleration a with
the angular acceleration 6,
p = a/R,
completes the system of equations. Solving the system,
we get
= mM, ~ 2°
m,+m,-+(1/2)m, ®? CSRs aue
po ;
i Ls SMEBIg = F, ~ 6.5N;
a= m,-+m,+(1/2)m, °°

+-(4/2) Mm, )meg, Fi x 33N;

yar .

Fi= (4
oa oa
mm, +m
—__—_*2_

B= wpm FU © ne
= m,/R ae

We see that the acceleration a of the system has di-

minished considerably. It is also interesting to note how
different the tensile stresses developed by the string
now are: the tensile stress F{ must be considerably higher
than F, since the moments of these forces have different
signs.
Te
100 Part 2. Solution of Standard Problems

Example 14.2. One end of a hanging string is tied to

a support at point O (Figure 14.2), while the other is wound
around a solid narrow cylinder (disk) of mass m = 10 kg
and radius R = 10 cm. Determine the acceleration of the
disk’s center of mass and the tensile stress developed by the
string, which is massless and noneztendable.

Figure 14.2 Figure 14.3

Solution. Two bodies, cylinder and string, constitute

the physical system. The cylinder cannot be thought of
as being a particle. We assume it to be a rigid body. Its
center of mass (point C) moves downward, and it itself
rotates about a moving axis passing through the center
of mass. Let us employ the theorem on the motion of
the center of mass, (14.1), and the equation of motion,
(14.7). We link the inertial reference frame with Earth
and point the coordinate axes as shown in Figure 14.2.
Two forces act on the cylinder, the force of gravity mg
and the tensile stress F,. By the theorem on the motion
of the center of mass,
macm = mg + F;.
Projecting this vector equation on the X axis, we get
macy = mg — F;.
The cylinder moves with respect to the moving axis,
but this axis is also in translational motion; in this case
 Ch. 4. The Motion of a Rigid Body 101

the equation of motion (14.7) remains valid:

1
= MRB=F,R,

where 6 = dcy/R is the angular acceleration. Solving

the system of equations, we get
2 5
acm = > 8 hs mg.

IIence, acon ~ 6.6 m/s? and F, = 163 N.

Let us make the problem more complicated: suppose
that a block (particle) of mass m, = 1 kg is attached to
the upper end of the string, the block can move (skid) without
friction along a horizontal surface, and the string is swung
over a massless pulley (Figure 14.3).
We denote the mass of the cylinder by m,. Applying
the dynamical method, we set up a closed system of equa-
tions for the translational motion of block and cylinder,
respectively,

Md, = Fi, M2dcm = mag — Fi,

and for the rotational motion of the cylinder,
+ m, RB = FR.
The acceleration acy of the cylinder’s center of mass,
the acceleration a, of the block (particle), and the angular
acceleration f, are linked through the following relation-
ship:
acm = a, + BR.
Solving the obtained system of equations, we get
= 2+ mg/m, __m3/m,
20M = m/m,
3 8 a= 3Em,)m, g,

_ 7 _ __2+ms/m,
P= 35mm 8 = P= REF mimy &
Substituting the numerical values, we get
acm + 9.1 m/s?, a, 27.5 m/s?, F, = 7.5N,
Bw 15.1 rad/s?.
102 Part 2. Solution of Standard Problems

The problem can be made still more complicated by

allowing for friction between block m, and the horizontal
surface, by taking into account the pulley’s mass, by
assuming that the pulley is a rigid body, and the like.
All such problems can be solved by the same dynamical
method.
Let us drastically change the terms of Example 14.2
by introducing a friction force, static friction. This
force is inevitable in the problem about to be discussed
and cannot be ignored.
Example 14.3. A massless and nonextendable string is
wound around a solid cylinder (disk) of mass m = 10 kg
and radius R =10 om. The
cylinder can move without
slippage along a horizontal
surface. A constant horizon-
tal force F = 30N is ap-
plied to the free end of the
string (Figure 14.4). Find
the acceleration of the center
of mass.
Solution. The physical
system consists of a single
rigid body, the cylinder,
Figure 14.4 whose center of mass moves
rectilinearly. The cylinder
also rotates about a moving axis whose direction of
motion does not vary. We are looking for the accele-
ration of the center of mass. This constitutes the
basic problem of rigid-body dynamics. Let us apply
the dynamical method. The inertial reference frame is
linked with Earth, the X axis points to the right, and the
rotation axis is parallel to the Z axis. Four forces act on
the cylinder: the tensile stress developed by the string
(it is equal to the given force F), the force of gravity mg,
the force N exerted by the support on the cylinder (these
two forces compensate each other), and the static-friction
force F;,.
The static-friction force can assume any value within
certain limits: 0< Fy,< fN. In the given case the fric-
 Ch. 4. The Motion of a Rigid Body 103

tion force has a value that prevents slippage (pure rolling

friction).
By the theorem on the motion of the center of mass,

macy = F + Fr.
The equation of motion about the axis passing through
the center of mass yields
+ mR2B = (F — Fy,) R.
Allowing for the fact that
p= acm/R

and solving the obtained system of equations, we find

that
4F i
CM Sm? Fra F.
Substituting the numerical values, we get a¢q = 4 m/s*
and F,, = 10 N.
The condition of absence of slippage assumes the form
+ F<fmg,
which yields
f>F/3mg~ 0.4.
If the friction coefficient is lower, slippage sets in.
Let us make the terms of the above problem more com-
plicated. Suppose that the string is swung over a massless
pulley and a load of mass m, = 20 kg (a particle) is tied
to the free end of the string; all other conditions remain
unchanged with the exception of force F, which is now the
tensile stress F, developed by the string. Find the acceleration
Qcm Of the center of mass, the acceleration a of the load,
and the tensile stress (Figure 14.5).
We denote the cylinder’s mass by m,. Applying the
dynamical method, we arrive at the following equations:
maa = mg — F, (for the translational motion of the
load),
m,acm = F, + F,, (by the theorem on the center-
of-mass motion),
+ mR'B = (F, — F;,,) R (the equation of motion).
104 Part 2, Solution of Standard Problems

The load’s acceleration and the acceleration of the

center of mass are connected by the relationship
a=dcn + BR. Since acm = BR, we have a = 2am.
m

Figure 14.5

Solving this system of equations, we get

3m, \-! Bim1)"
a=(1+ =
8m, | & aom=—y (1+ 2

Peo (ita) & Mott ge) &

_ my 3m, \~! = a ( o, -1

Substituting the numerical values vials

a~8.4m/s*, acm + 4.2m/s?, Fy, + 10.5N,
F, = 31.5N.
The terms of the problem can be made still more com-
plicated by allowing for the mass of the pulley, by as-
suming that the cylinder is a rigid body and moves along
an inclined plane rather than a horizontal surface, by
supposing that the load (particle) moves along an in-
clined plane, and the like. Clearly all these problems can
be solved; by applying the same dynamical method.

15. Conservation Laws in Rigid-Body Dynamics

The elementary work resulting from the rotation of a
rigid body through an angle dq is defined by the follow-
ing formula:
dW = M dg, (15.1)
 Ch. 4. The Motion of a Rigid Body 105

where M is the moment of force about the rotation axis.

The total work is obtained by integrating Eq. (195.1):
Pe

w=| M dg. (15.2)

zy

The kinetic energy of a rigid body in arbitrary motion

breaks down into the kinetic energy of translational
motion and the kinetic energy of rotational motion:

Ey = mvgy/2 + Jw?/2, (15.3)

where Uvcm is the velocity of translational motion of
the center of mass, and J the moment of inertia of the
rigid body about the rotation axis.
In addition to momentum and energy conservation
laws in mechanics, rigid-body dynamics also employs the
law of angular-momentum conservation. This law fol-
lows from the equation of motion (14.3) with respect to
a point: if the vector sum of the moments of external forces
about a fixed point O is zero, the angular momentum about
this point is constant:
L = const. (15.4)
More often the law of angular-momentum conservation
is used in a form that follows from the equation of motion
(14.4) with respect to a fixed axis: if the algebraic sum of
the moments of external forces about a fixed azis is zero, the
angular momentum of the system about this azis is a con-
stant:

L= Jw=const, (15.5)
where the summation sign stands for the algebraic sum
of the angular momenta of all the bodies in the system.
Application of conservation laws in rigid-body dy-
namics is carried out along the same lines as in particle
dynamics.
Example 15.1. A wooden rod of mass M = 6 kg and
length | = 2 m can rotate in the vertical plane about a
horizontal azis passing through point O (Figure 15.1).
A bullet of mass m, = 10g, flying with a velocity
106 Part 2. Solution of Standard Problems

Vo = 10° m/s at right angles to the rod, hits the lower end
and buries itself in the rod. Determine the kinetic energy
of the rod after impact.
Solution. Two bodies form the physical system: the
rod and the bullet. The bullet can be thought of as a par-
ticle, while the rod is assumed to be a rigid body. The
physical phenomenon consists
of the bullet interacting with
the rod (an inelastic collision).
The state of the system prior
to collision is known. We must
determine a parameter of the
system, the kinetic energy, after
the interaction has ceased.
The nature of the forces acting
in the interaction is assumed to
be unknown. Therefore, the
dynamical method cannot be
applied. Let us apply the con-
servation-law method. Prior to
collision, the bullet was moving
Figure 15.1 rectilinearly, but after collision
it is in rotational motion togeth-
er with the rod. It is, therefore, advisable to employ
the law of angular momentum conservation about the
fixed rotation axis, since the conditions of applica-
bility of this law are met.
As usual, the inertial reference frame is linked with
Earth, the origin is placed at point O, the X axis is di-
rected along the axis of rotation. The angular momentum
of the bullet about the rotation axis prior to collision
iS MgVol and that of the rod is zero. After collision the
angular momentum of the rod together with the bullet
is Jo, where J is the moment of inertia of rod and bullet
about the X axis, and » the angular velocity of their
rotation after collision. Since the moment of inertia of the
bullet is much smaller than that of the rod, ml? <
(1/3) Ml?, we can assume that J ~ (4/3) MI’. By the
law of conservation of angular momentum,
MyVol a Jo.
 Ch. 4. The Motion of a Rigid Body 107

The kinetic energy of the rod is

Ey=
= = TE Seal E,=255. (15.6)
J 2 2 2p 2

Note that initially the kinetic energy of the bullet

(prior to impact) was Exo = mov3/2 that is, Eyg = 5 X
10° J, which is considerably higher than the kinetic
energy of the rod after impact. As a result of the inelastic
collision the greater part of the initial mechanical energy
was transferred into nonmechanical! forms of energy. As
a result of the interaction there emerged very strong
nonconservative forces, which dissipated the mechanical
energy of the system. Therefore, it would be wrong to
use the law of conservation of energy in mechanics di-
rectly in the form Jw?/2 = myv?/2, and it would also
be wrong to use the law of momentum conservation, since
after collision the rod together with the bullet is in ro-
tational motion. If we did use the law of momentum con-
servation, we would have mgvy = (M -- Mo)u, where
= ol, whence, neglecting the bullet mass m, in compari-
son to the rod mass M, we would find that © = mgv,/ ML.
The kinetic energy of the rod after impact would be
Jo?/2 = miv3/6M, which is roughly 2.7 J and lower than
the result (15.6) almost by a factor of ten.
Let us find the maximal angle a by which the rod is
deflected from the vertical after impact. There are no
nonconservative forces in the system after impact and,
hence, we can apply the energy conservation law to the
motion of rod and bullet after collision. By this law,
3
said = Mgh,

where kh is the height to which the rod’s center of mass

is raised (in relation to the point prior to impact, A) as
a result of the impact (Figure 15.2). Here we have allowed
for the fact that my>< M. From triangle OBC it follows
that
y2—h
cos a@ = 12
108 Part 2. Solution of Standard Problems

Solving the system of equations, we get

3m2v2
@ == Cos a |
(1=, Meioro
), aw
~My
54°.
°

We can also consider many variants of this problem,

say, by replacing the bullet with a steel ball and the
wooden rod with a steel rod, by assuming an clastic

Figure 15.2

collision instead of an inelastic, by studying the case of

a glancing collision, and the like. All these variants can
be solved by the conservation-law method.
In conclusion we consider a problem whose solution
will be found by all four methods: the kinematic, dy-
namical, conservation-law, and DI.
Example 15.2. A solid homogeneous disk of radius
R=10cm which initially had an angular velocity
@>o = 50 rad/s (about an axis perpendicular to the disk’s
plane and passing through the disk’s center of mass) is placed
on its base on a horizontal surface. How many rotations
will the disk make before it stops if the coefficient of friction
between the base and the horizontal surface, {, is 10-1 and
does not depend on the disk’s angular velocity?
Solution. The physical system consists of a single body,
the disk, which cannot be considered a particle (we assume
 Ch. 4. The Motion of a Rigid Body 109

it to be a rigid body). The physical phenomenon con-

sists of the disk being decelerated in its rotation about a
fixed rotation axis under the forces of friction (all other
forces are balanced). The initial and final states of the
disk are known. We are looking for one of the parameters
of this motion (the number of rotations N the disk will
make before it stops). This constitutes a basic problem
of rigid-body dynamics.
Let us apply the dynamical method. The disk’s center
of mass is at rest as the disk rotates. The equation of
motion (14.7) yields

+-mRp=M, (45.7) av /
where m = nRhp is the disk’s
mass, h its height (thickness),
p the density of its material,
B the angular acceleration,
and M the total moment of
the forces of friction about
the axis.
The force of friction is ap-
plied to each section of the
disk, and since these sections Figure 15.3
lie at different distances from
the axis, the moments of the forces of friction differ
from section to section. To find M we apply the DI me-
thod. We partition the disk into thin rings (Figure 15.3).
Each ring is also partitioned into smal] elements by neigh-
boring radii that form a small angle dg. In Figure 15.3
one such element is hatched. The force of friction acting
on the element is
dF,, = f dg r dr hpg.

The moment of this force of friction is

dM =rdF,, = foghr’dr dg.

Integrating with respect to angle @ from zero to 2n and

with respect to r from zero to R, we get the total mo-
410 Part 2. Solution of Standard Problems

ment of friction forces:

on R

M = \ \foghr? dr dp = fogh2nR3/3 =2fRgm/3. (15.8)

0 6
Substituting this value of M into the equation of mo-
tion (15.7), we find the angular acceleration of the disk:
p= 4fg/3R.
Solving the inverse problem of kinematics (the kine-
matic method), we determine the law of variation of
the angular velocity,
© = @ — fe, (15.9)
and the respective law of motion,
P = Wot— Bt?/2. (15.10)
Taking into account the fact that the final angular ve-
locity of the disk is zero, © = 0, and employing Eq. (15.9),
we can find the time of the rotation:

t= B= yg? t = 3.75s.
— %o __ 3Ra ey

Substituting this value of ¢ into Eq. (15.10) and bearing

in mind that @ = 2nN, we get
3Rw}
N= 16xfg?
N = 15. (15.14)

Now let us solve this problem using the conservation-

law method. The physical system consists of two bodies:
the disk and Earth. The system is closed, and the law of
conservation of energy in mechanics could be employed
if there were no nonconservative forces acting in the sys-
tem. Assuming that these forces are external, we obtain
from Eq. (13.16) the following:
Ju?/2=W, (15.12)
where J = (1/2) mR? is the disk’s moment of inertia, and
W is the work performed by the nonconservative forces
(friction). Since we already know the moment of these
 Ch. 5. The Gravitational Field 411

forces (see (15.8)), we can use (15.2) to find

Wa Mdm | 2REE gg — Rene

9 9
= _ ( 2/Rgm _ 2fRgemep

0 0
Substituting this value of W into Eq. (15.12) and bearing
in mind that gp = 2nN, we get
N= aeTexfe * Nz aero 15,

which coincides with formula (15.11) found by using the

dynamical method.
Concluding the study of the mechanical model, we see
that any standard formulated problem in this depart-
ment can be solved by applying a fairly small number
of universal methods (aside from the method of analyz-
ing the physical content of a problem): the kinematic,
dynamical, conservation-law, and DI.

ELEMENTS OF THE THEORY OF PHYSICAL FIELDS

Chapter 5
THE GRAVITATIONAL FIELD

16. The Basic Problem of Gravitation Theory

The basic law of gravitation theory is Newton’s law of
gravitation,
F=G 7rm? (16.1)
where G ~ 6.67 x 107"! N.m/kg? is the universal grav-
itational constant. In the form (16.1) the law is valid
only for particles and spherical bodies. It can be written
in vector form thus:
Fy.= —G Pr, (46.2)
where F,, is the vector of the gravitational force acting
on body me, and r the radius vector pointing from body
m, to body m, (Figure 16.1).
112 Part 2. Solution of Standard Problems

The main characteristic of each point in a gravitational

field is the field strength E, a vector quantity defined thus:
E-=F/m,, (16.3)
where F is the gravitational force acting on a particle of
mass my placed at the given point.
The strength and potential of the gravitational field
generated by a particle of mass m at a point positioned

m Fi2 Mo
——<_—_+—___
r

Figure 16.1

at a distance r from the mass are expressed by the follow-

ing formulas:
E=Gml/r’, (16.4)
= —Gmir. (16.5)
The field strength E at a point and the potential
at the same point in a gravitational field are linked by
the formula
= —grad g. (16.6)
The state of the physical system considered (the gravi-
tational field) is determined by the value and direction
of vector E at any point in the field. The field strength E
of a gravitational field is the fundamental characteristic
in the sense that, knowing E, we can not only determine
any parameter characterizing the field but describe the
behavior of physical systems in this field. Indeed,
Eq. (16.6) can be used to find potential @, and Eq. (16.3)
makes it possible to determine the force with which the
field acts on a body placed in it. If the initial conditions
for this body are known, by applying the dynamical meth-
od we can determine the law of the body’s motion. And
knowing this law makes it possible to find all other char-
acteristics and parameters determining the motion. This
leads to the following formulation of the basic problem
i erouleeHion theory, the problem of calculating the
ield.
 Ch. 5. The Gravitational Field 143

Calculating a gravitational field means determining

at each point in the field the field strength vector
E or the potential q.

17. The Gravitational Field Generated by a System

of Particles
A fundamental physical principle lies at the base of the
method for calculating physical fields, the superposition
principle. If the field is generated by a system of par-
ticles, we must first determine the field generated by
each particle separately (i.e. the corresponding vector E,).
Then in accordance with the superposition principle we
find the resultant field vector E as the vector sum of the
field strength vectors:
Bebe
uk te “dj
The gravitational field generated by a single particle
was calculated in Section 16. Description of the motion of
even one body in the gravitational field generated by one
particle presents certain mathematical difficulties. It is
fairly easy to solve such a problem physically, that is,
set up a closed system of equations, by employing either
the dynamical method or the conservation-law method.
Difficulties for first-course students appear at the mathe-
matical level, when it is necessary to solve the system of
equations (often differential equations).
First it is advisable to solve a number of elementary
problems that involves estimation, say calculate the field
strength and potential of the gravitational field on the
surface of the Moon, the Sun, or Mars (note that g =
GM/R? ~ 9.8 m/s* is the gravitational field strength
at Earth’s surface), determine (estimate) the orbital and
escape velocities for Earth, the Moon, and Mars, and
the like.
Then the first problem concerning the motion of a par-
ticle in a known gravitational field can be formulated.
It is even advisable to present it in the form of a non-
specified problem.
Example 17.1. A rocket is launched vertically with an
initial velocity vy at the North Pole (it is assumed that the
8—0408
444 Part 2. Solution of Standard Problems

boosters impart a velocity of vy to the rocket instantaneously

and are then switched off). Describe the motion of the rocket.
Solution. The problem is nonspecified. The first simpli-
fying assumption is obvious: the air drag is ignored. The
rocket can be thought of as being a particle. Its motion
can be described if we find its law of motion. The law of
motion depends essentially on the value of the initial
velocity v». Let us assume that vy is so low that at the
point of the rocket’s greatest altitude the acceleration of
free fall g, (which constitutes the strength of Earth's
gravitational field) differs very little (say, not more than
by one percent) from the acceleration of free fall on
Earth’s surface, gy. It is useful to estimate this altitude
h, and the initial velocity vp, that the rocket must have
to reach this altitude. Since, by assumption, (gy) — g,)/
g = 10-? and
GM GM
So=—Ra >» 81= Rpm?
we haveh, ~ R (1/V0.99 — 1) and vp, = V 2gh,, that
is, hy ~ 32 km and v,, ~ 800 m/s. .
Thus, if v9 <u 9,, then the rocket’s acceleration is
approximately constant, and we have a trivial high-
school problem concerning the uniformly decelerated
upward motion of a particle with a constant ‘acceleration
Zo. The law of motion in this case can be written in the
form
Z = vot — g,t?/2,
which provides all the parameters of motion.
We will not consider the case where the initial velocity
Vo is greater than or equal to v, ~ 11.2 kin/s, the escape
velocity for Earth. Thus, the first problem can be for-
mulated as follows:
Example 17.2. A rocket is launched vertically at Earth's
North Pole with an initial velocity vg satisfying the condi-
tions Vg, << Vg << vg. Find the law of its motion. Ignore air
drag and the effect of other planets, the Sun, and the Moon
on the motion of the rocket.
Solution. Two bodies constitute the physical system,
the rocket and Earth. The rocket can be thought of as
 Ch. 5. The Gravitational Pteld 415

being a particle. The gravitational field generated by

Earth (assumed to be a spherical body) is known. The
physical phenomenon consists of the particle (rocket)
moving in a nonuniform gravitational field. We wish
to determine the law of motion of the rocket. This consti-
tutes the basic problem of particle dynamics.
We apply Newton's second law. The inertial reference
frame is that linked with Earth (since Earth’s mass is
considerably greater than the rocket’s mass, we assume
Earth to be fixed), the X axis is directed upward, and
the origin is placed at Earth’s center. There is only one
force acting on the rocket, the gravitational. It is impor-
tant to note that this force is variable. Then, by New-
ton’s second law,
mz= —GmM/z? for x>R. (417.2)
Physically the problem has been solved because we
have obtained a single differential equation in one un-
known function z (t), the coordinate of the rocket in
space. Solving the equation we get the law of motion.
However, first-year students have great difficulty in solv-
ing this equation. Two points must be stressed in this
connection. First, Eq. (17.2) is solvable in principle, and
in the final analysis we can obtain the law of the motion
of the rocket. Second, even at this level students can be
told that solving physics problems may lead to equations
that have no exact (analytical) solutions. Then a com-
puter must be brought into the picture to obtain numeri-
cal, approximate solutions.
Let us simplify the formulation of the problem by
using the conservation-law method instead of the dynam-
ical. We apply the law of energy conservation to the
rocket-Barth system:
myy mGM _ my? mGM
5 =a 5 a (17.3)
where vu is the velocity of the rocket at a point with coor-
dinate z. This gives us the maximum value of z (the po-
sition of the rocket at v = 0):

Tmax = 36M
— ER
2GMR
(17.4)
146 Part 2. Solution of Standard Problems

If the initial velocity vp is equal, say, to the orbital

velocity for Earth, vy, = V GM/R, the maximum value
of x is tmgy = 2R, and the maximum altitude reached
by the rocket ishpay = R ~ 6400 km. Equation (17.3)
can be used to find the dependence of the rocket's velo-
city on coordinate z:

v= 26m (4-4), (17.5)

The diagram of this dependence is shown in Figure 17.1.
Now we can formulate the second (simpler) problem.
Example 17.3. A rocket is launched vertically at Earth's
North Pole with an initial velocity vg satisfying the con-
ditions vg, << Vp <<. Vg. Determine the maximum altitude

0 R

Figure 17.4

reached by the rocket and the rocket’s velocity at an arbitrary

point of its path. Ignore air drag and the effect of other
Planets, the Sun, and the Moon on the motion of the rocket.
Solution of this problem was obtained earlier (see
Eqs. (17.4) and (17.5)).
Note that in the above problems we must appraise in
more detail the upper bound on the initial velocity, since
at velocities close to v, the altitude reached by the rocket
is so great that the effect of the Moon, Sun, and other
bodies can no more be ignored. It is advisable for more
advanced students to make the necessary estimates.
To conclude this section let us consider one more
problem.
 Ch. 5. The Gravitational Field 417

Example 17.4. A rocket is circling Earth along an

orbit that almost coincides with the Moon's orbit. When
a retroengine is fired, the rocket rapidly loses speed and
begins to fall toward Earth (Figure 17.2). Determine the
time it will take the rocket
to reach Earth. Ignore air \
drag and the effect of other
bodies on the rocket’s motion.
Solution. The physical
system consists of the rock-
et and Earth. The physi-
cal phenomenon consists
of the rocket moving in
Earth's gravitational field.
The solutions of the pre- Figure 17.2
vious problems in_ this
section show that the
dynamical method leads to a complicated differential
equation, while the conservation-law method makes it
possible to find the rocket’s velocity at each point of
the path but not the sought time of fall. So standard
methods have led us nowhere. But suppose that we con-
sider the motion of the rocket as the motion of a satellite
of Earth along an extremely prolate ellipse whose major
axis is as long as the radius of the Moon's orbit, Ry ~
4 x 10° km and whose eccentricity e = 1. This means
that we can employ Kepler’s third law,
(+) = (“4*)
’

where ¢ is the time of fall, and 7 = 27.3 days is the pe-

riod of revolution of the Moon about Earth. Calculations
yield ¢ = T/4V/2, which amounts to about 4.85 days.

18. The Gravitational Field Generated by

an Arbitrary Mass Distribution
The common approach in calculating the gravitational
field generated by an arbitrary mass distribution is to
employ the superposition principle and the DI method
(see Section 6). In calculating the field strength by this
118 Part 2. Solution of Standard Problems

method it is extremely important to take into account

the vector nature of this quantity. After the elementary
field strength vector dE has been found, its projections
dE., dE,, and dE, on the respective coordinate axes are
determined, with subsequent integration (summation) car-
ried out for each projection separately.
If the field strength is known, the problem on the mo-
tion of bodies in such fields is solved by either the dy-
namical method or the conservation-law method.
Example 18.1. Describe the motion of a particle in the
gravitational field generated by a long thin, homogeneous
rod of mass M and length |. The effect of other bodies can
be ignored.
Solution. Let us restrict our discussion to the one-di-
mensional case. We assume that initially the particle

Figure 18.4

was positioned on the rod at a distance z, = | from one

of the rod’s ends (point B in Figure 18.1) and had a zero
velocity (vo = 0). The physical system consists of two
bodies: the rod and the particle (whose mass we denote
by m). The physical phenomenon consists of the particle
moving in the gravitational field generated by the rod.
The gravitational force acting on the particle is un-
known (it is not F = GmM/z? since the rod is not a par-
ticle). To use the dynamical method we must calculate
the gravitational field generated by the rod on the rod’s
axis, that is, we must find the field strength vector E and
potential @. Let us apply the DI method. We assume
that m < M. We select the inertial reference frame as
the one linked with the rod, place the origin O at the
left end of the rod, and direct the X axis to the right.
We partition the rod into segments so smal] that each
 Ch. 5. The Gravitational Field 119

can be thought of as being a particle. We take one seg-

ment of length dz positioned at a distance z from an ar-
bitrary point B on the axis. The mass of the segment is
dm = pA dz, where A is the cross-sectional area of the
rod, and p the density of the rod’s material. Since the
selected segment is a particle, the characteristics of the
gravitational field generated by it (the field strength dE
and potential dg) are known:
Gdm GpAdz d Gdm _ GpA dz
dE= z zm ? x z .

Note that in our case all the elementary field strength

vectors dE point in the same direction. Integration yields
the overall characteristics of the field generated by all
the clementary segmonts (i.e. the field generated by the
rod as a whole):
lx,
E= | GpAdr GM
a ze tq (I-29) ?
Xo

_ _"O Gpddz ot l
ga f Se =--F in (t+ 2).
The force acting on a particle placed at a distance z
from the origin is
GMm
F-= Seay &

Newton's second law,

ns d2z ___GM
de? z(z+l ?

results in a differential equation whose solution will

enable us to establish the appropriate law of motion of
the particle.
Applying the law of energy conservation in mechanics,

—~Sn (1+—)= On (1+ +)+ a ,

120 Part 2. Solutton of Standard Problems

we can determine the velocity of the particle positioned

at a distance z from the right end of the rod:
26M In i+Uz
v=
T i+1/z, °
Let us consider examples of more complicated fields.
Example 18.2. Determine the field strength of the gravi-
tational field generated by a thin ring of radius R and

Figure 18.2 Figure 18.3

mass M at a point A lying on the ring’s azis at a distance

z from the ring’s plane (Figure 18.2).
Solution. The physical system consists of the ring and
the ring’s gravitational field. We wish to solve a basic
problem of gravitational theory, namely, find the field
generated by the ring. A ring, however, cannot be thought
of as a particle and, hence, formula (16.4) is invalid.
Let us employ the DI method. We partition the ring
into segments so small that each segment can he regarded
as a particle. We take one such segment of length di =
Rdg (see Figure 18.2). This generates a gravitational
field whose’ field strength vector at point A is dE of a
magnitude
GdM
 Ch. 5. The Gravitational Field 421

where dM = (M/2nR) dl is the mass of element di. Vec-

tor dE forms an angle a@ with the X axis and an angle »
with the Z axis. The projection of dE on the X axis is
GdM GMzd
OE. = pape O84= Toppan — (18.2)
Integration of (18.2) with respect to m yields the projec-
tion of the sought field vector on the X axis:
on
GMz dg GMz es GMx
a |axcrEaA = apap |=a 0

(18.3)
We can clearly see that in view of the symmetry of
the problem the sum of the projections of the elementary
field strengths on the Y axis is zero: E, = 0; so is the
sum of the projections on the Z axis: FE, = 0. Hence, the
sought field vector is directed along the X axis and its
magnitude is given by (18.3). After calculating the field
generated by a ring we can formulate a number of prob-
lems on the motion of bodies in such a field.
Example 18.3, Describe the motion of a particle of mass
m that initially was at rest at point O, on the axis of a thin
ring of mass M and radius R; point O, lies at a distance
Zo < R from the ring’s plane. Assume that m < M (Figu-
re 18.3).
Solution. To describe the motion of a particle in a
known gravitational field means to find the law of motion
of this particle. This constitute a basic problem of dy-
namics.
Using the dynamical method, from Newton’s second
oe we obtain the differential equation of harmonic oscil-
ations:
ee GM.

mz= — ae a (18.4)

Here we have allowed for the fact that for z< AR the
strength of the gravitational field generated by the ring
is given by (18.3), or E =~ (GM/R°)x. Thus, the par-
ticle oscillates harmonically according to the law of mg-
122 Part 2. Solution of Standard Problems

tion + = 1) sin (Wot + @), the period of these oscilla

tions being T = 2nRV RIGM.
If the condition z < R is not met, the same dynami-
cal method leads to a more complicated differential equa-
tion:
ee GMm
miz= — (R27 z. (48.5)

Solution of the equation will result in finding the law of

motion z = z (t), and using the law of energy conser-
vation, we can find the dependence of the velocity v of
the particle on the position (coordinate zx) of the particle,
but first we must determine the potential of the gravita-
tional field generated by the ring by applying cither the
DI method or formula (16.6).
Let us make the terms of Example 18.2 more complicat-
ed. We wish to calculate the strength of the gravitational
field generated by a semiring at the same point. Clearly,
the projection of the resulting field strength vector on
the X axis diminishes by a factor of two in this case:
GMz
E, = 2 (R223) ’ (18.6)

But the most important fact is that because the symmetry

of the problem breaks down, there emerges a nonzero
projection of the field strength vector on the Z axis, E,.
Since
_ GMsinacosgdl —§ GMRcosad@
dE.= aR (Rp at) once pana + (18.7)
we have

+e? CMR cos od GMR

E.= |meee
= cosgap

—aap”
~n/
(188)
Note that in view of the remaining, ymmetry E, = 0.
The terms of Example 18.2 could be made still more
complicated, say, if the aim were to calculate the gravita-
tional field generated by a quarter of a ring, an arc with a
central angle @ < n/2, and so on. All these problems can
be solved by the same method.
 Ch. 5. The Gravitational Field 123

Let us employ the above results to calculate the field

generated by a hemisphere.
Example 18.4. A sphere of mass M and radius R is se-
parated into two hemispheres by a plane passing through
its center. Determine the potential of the gravitational field
generated by each hemisphere at a point O on a straight line

=
=
L_}
=a
L__}
L_}
v,

Figure 18.4

that is perpendicular to the separating plane and passes

through the sphere’s center, the distance from point O to
the center of the sphere isx > R (Figure 18.4).
Solution. To calculate the gravitational field generated
by each hemisphere (not a particle) we employ the DI
method. We partition each hemisphere into narrow rings
of width R d® and radius R sin 8. Now we consider one
ring. Its surface area is
dA=2xR sin 0 R dO = 2nF? sin
8d0 (18.9)
and its mass is
dA M sin 0d0

Since all the elements of this ring are positioned at the

same distance | from point O, the elementary potential
of the gravitational field generated by the ring at point
O is

eat
fattaoe
GM sin®8d6
=a (18.11)
124 Part 2. Solution of Standard Problems

Take triangle CBO. By the law of cosines,

P= R24 7?—2RzrcosO. (18.12)
Differentiation of both sides of (18.12) yields
2l dl = 2Rz sin 8 dé,
which implies that
sin 8d0=1dl/Rz.
Substituting this into (18.11), we arrive at the final ex-
pression for the elementary potential du as a function ofa
single variable, 1:
GdM dl
du= ——5-— (18.13)

The elementary potential could also be expressed as a

function of a single variable 6 by excluding/ from (18.11)
via (18.12). But the subsequent integration in this case
would be more complicated than integration of (18.13).
Denoting the potential generated by the right hemisphere
by U, and that generated by the left by U, and inte-
grating (18.13), we get

Mae GM GM Se
= \ (- 2Rz ) di= — + [V Re+2
zx-R

—(r—R)], (18.14)
x+R

U,= | (--Sae-) l= --ape We +A)

Viigs
—-V R42. (18.15)
Comparing (18.14) with (18.15), we arrive at an expres-
sion for the potential of the gravitational field generated
by a sphere at a point lying outside the sphere:
U=U,+U,= —GMI/z. (18.16)
The same method can be used to consider the case where
z< R. We can also calculate the gravitational field
generated outside and inside a homogeneous ball, which
was partially done in the problem of oscillations of an
 Ch. 5. The Gravitational Fiteld 425

object in a shaft (e.g. see Example 12.1)), etc. To conclude

this section let us consider the following problem.
Example 18.5. A spacecraft has posvitioned itself at a
point on the azis of a planetary nebula. The point is at a
distance ro = 5d from the nebula’s centeer of mass, and the
rocket engines of the spacecraft have beens switched off. How
much time will it take the spacecraft to» reach the nebula,
moving only because of gravitational attreaction? The nebula
is assumed to be a disk of diameter d = ‘10-* parsec, thick-
ness (depth) h = 10-* parsec, and homaogeneous distribu-
tion of matter with a density p =10-"" Ikg/m*. The initial
velocity of the spacecraft with respect to the nebula is as-
sumed zero, Vy = 0, and the spacecraft meass is m = 10° kg
(1 parsec = 3.08 x 1075 km = 3.08 x 101° m).
Solution. The physical system consissts of two physical
objects, the nebula and the spacecraft. The latter can be
considered a particle. Since the deptha of the nebula is
small compared to hoth theldistance ro, and the diameter
d of the nebula, we will assume that the nebula is a thin
disk. The physical phenomenon cons:ists of a particle
(spacecraft) moving in the gravitatiomal field generated
by the nebula (the disk is not a partcicle). We wish to
find the time of flight of the spacecraft. The law of motion
can be determined by the dynamical method if we know
the strength of the gravitational field generated by the
nebula.
Thus, the solution procedure is clesar: we must first
calculate the gravitational field geneerated by a thin
disk (nebula) of mass
M,, = nd?hol4, (18.47)
then via the dynamical method determine the law of mo-
tion of the spacecraft, and then find the time of flight
from the law of motion.
Let us calculate the strength of the gravitational field
generated by the disk. We employ th.e DI method. We
partition the disk into thin rings of widith dr and consider
one ring of radius r (Figure 18.5). Its mmass is
dM = 2mrho dr. (18.18)
426 Part 2. Solutton of Standard Problems

Using formula (18.3), we can find the elementary strength

of the gravitational field generated by the ring:
dE, = 2x d
(18.49)
Integration of (18.19) yields the strength of the gravita-

Figure 18.5

tional field generated by the entire nebula:

=f 2aGrphrdr_ ongon(4———_4____
d/2

= apr = 2nGeh (1 —aS

(18.20)
Next, applying the dynamical method, we find that
Newton's second law leads us to the following differential
equation:

mzpees
= —2nGphm (1 eee SS). (18.24)
EXCS a
Solution of this equation would enable us to find the law
of motion of the spacecraft, z = z(t), and, hence, cal-
culate the time of flight t). But it seems that there was
no need to set up the differential equation (18.21) and
even less need to solve the equation. The sought time of
flight t, can be found approximately in a simpler way,
namely, by the estimate method. We start by estimating
the mass of the nebula. Formula (18.17) yields
M,,=nd2hpl2, M,~ 2x 10% kg.
 Ch. 6. Phe Electric Pield _ 127

Thus, the nebula’s mass is very small: it is less than the

solar mass Mg, = 2 x 10° kg by a factor of 10°. Allowing
for the fact that dimensions of the nebula are great (the
depth of the nebula kh= 3 x 10!5 m is several times gre-
ater than the diameter of the solar system d, ~ 1.2 X
40'3 m), we can easily predict that the gravitational field
generated by the nebula will be very weak even at the
nebula’s boundaries.
Let us estimate the order of magnitude of E, for two
values of ry: (1) ro = 5d, and (2) rp ~ O (at the boundary
of the nebula). From (18.20) we find that E, ~ 10-?® m/s?
and E;, ~ 10-'% m/s?, respectively. These values are
extremely small. Even if the spacecraft is moving with
maximum acceleration a = Ej, it will take it
t= V 2s/Er, t,~+4.5x 10° sx 52 days,
to cover the distance s = 1m, and
to= V 5d/E; ’

which amounts to about 1.7 x 10's, or 5 x 10° years,

to cover the distance ry = 5d.
Thus, in such a weak gravitational field the spacecraft
is practically at rest. This is the result of applying the
estimate method for solving the problem. The result is
instructive and demonstrates that sometimes, before
applying the laws of physics and setting up (differential)
equations, it is advisable to make a rough estimate of
several quantities and analyze (compare) the results
obtained in the process.

Chapter 6
TIE ELECTRIC FIELD

19. The Electrostatic Field in a Vacuum

The basic law of electrostatic-field theory is Coulomb’s
law,
Fa —210 (19.4)
 128 Part 2. Solutton of Standard Problems

The law is valid for point electric charges that are at rest.
It closely resembles Newton's law of gravitation. Hence,
everything said in Chapter 5 concerning a gravitational
field can be said of an electrostatic field.
The main characteristics of an electrostatic field are the
field strength E and the potential o. For a field generated
by a point charge we have
@
E=aor (49.2)
== ; (49.3)
The field strength E and potential @ of an electrostatic
field are linked by formula (16.6).
The state of an electrostatic field as a physical system is
determined entirely by the direction and magnitude of
the field strength vector at every point in the field. Hence,
the basic problem of electrostatics consists of calculating
the electric field. Here it is advisable to distinguish three
cases:
(1) the field is generated by a system of point charges;
(2) the field is generated by a system of point charges
and charges carried by bodies of regular shape; and
(3) the field is generated by an arbitrary distribution
of electric charge.
Although the first case was considered earlier in connec-
tion with the gravitational field, it is highly advisable
to calculate the fields generated by a dipole (not only at
points lying on the dipole’s axis but at arbitrary points),
a quadrupole, and other point-like systems. In the second
case we first use Gauss’ law of flux to calculate the fields
generated by charges distributed over regularly shaped
objects and then, using the superposition principle, de-
termine the total field. For an arbitrary distribution of
charge we employ the DI method (see Section 6).
If the characteristics of the field have been calculated,
problems on the motion of electrically charged particles
in a known field can be solved by either the dynamical
method or the conservation-law method.
Example 19.1. Calculate the strength of an electric field
generated by a straight infinitely long string uniformly
 Ch. 6. The Electrie Field 129

charged with a linear density y, at a point O that is ro

distant from the string.
Solution. Since the charge cannot be considered point-
like, we cannot use formula (19.2). Let us apply Gauss’ law
of flux. In view of the symmetry of the field, the field

Figure 19.1 Figure 19.2

strength vector al each point is directed along the normal

to the cylindrical surface on which the point lies, with
the symmetry axis of this cylindrical surface coinciding
with the string. For this reason, for the closed surface
we take a cylinder of length 7 whose symmetry axis coin-
cides with the string and on whose lateral surface point O
lies (Figure 19.1). The flux of vector E passing out of the
lateral surface of the cylinder is PD, = 2ar,lE, and the
electric charge inside the cylinder is Q@ = yl. By Gauss’
law,
2nrlE = yl/ey.
Hence, the sought field strength is
baits eS
E=>.. (19.4)
Now let us approach this problem from the DI angle.
We partition the string into segments so small that the
P— 0498
130 Part 2. Solution of Standard Problems

charge carried by each can be considered point-like. We

select one such segment of length di carrying a charge
dQ = y dl (Figure 19.2). At point O the elementary field
generated by this charge has a strength of

dE = men =i... (19.5)

From triangle ADO we get
r=r,/cosa@.
Since |AC| =rda =r, da/cos a, we find that triangle
ABC yields
dl =|AC |/cosa= ry da/cos? a.
Substituting the values of r and dl into Eq. (19.5), we get
dE dy= ed
tne (19.6)

The projections of vector dE on the X and Y axes are

vos
@ da
dz, = ae ; (19.7)
sina da
dE, = (19.8)
Integration yields
+n/2 +7/2
E.= ( yoosada “ss a \ ysinada _
xo 4Negrgo = TMgTy * yO 4N@ry
7/2 ~Aj2

Thus, the final result is

een ae
E= 2Neoro *
which coincides with formula (19.4) obtained via Gauss’
law.
At first glance the DI method seems to have proved more
involved than the Gauss’-law approach. In the given
example this is indeed the case. But the DI method is
universal and can be applied in cases where the Gauss’-
law approach proves useless.
Example 19.2. Determine the strength of the electric
field generated by a straight piece of string carrying an elec-
 Ch. 6. The Electric Field 131

tric charge with a linear density , at a point O that is ro

distant from the string. The angles a, and a, are specified
(Figure 19.3).
Solution. Clearly, the symmetry of the field generated
by an infinite string is broken: the field is not symmetric.
It is extremely difficult to enclose the piece of string

Figure 19.3 Figure 19.4

with a surface using which it would be fairly easy to

calculate, via Gauss’ law, the flux of vector E.
Let us apply the DI method. The projections of the
elementary field vector dE on the X and Y axes were
obtained in Example 19.1. Integrating (19.7) and (19.8),
we find the projections (or components) of the sought
vector E on the X and Y axes:
+a@2
d
B,= | mere = ag (sina,+sina@,), (19.9)
-@)

a2
= ysinada _
E, ~ \ 4Ne gro = on
(cosa@,—cos@,). (19.10)
-ay

Clearly, the field generated by a charged infinitely

straight string, (19.4), constitutes a particular case of the
field generated by a piece of charged straight string. In-
ge
132 Part 2. Solution of Standard Problems

deed, for a, = —n/2 and a, = +n/2, Eqs. (19.9) and

(19.10) yield E, = y/2negro and E, = 0, which coincide
with (19.4).
Now that we have formulas for expressing the strength
of fields generated by a segment and an infinitely long
charged string, we can formulate dozens of problems in-
volving the calculation of fields generated by various
combinations of uniformly charged segments, and infi-
nite and semi-infinite strings (triangles, squares, angles,
etc.).
Example 19.3. An infinitely long string uniformly char-
ged with a linear density », = 4-3 x 10-7 C/m and a seg-
ment of length 1 = 20 cm uniformly charged with a linear
density V2 = 4-2 X 10-7 C/m lie in a plane at right angles
to each other and separated by a distance ry) = 10 cm (Fig-
ure 19.4.) Determine the force with which these two bodies
interact.

Solution. Two objects constitute the physical system,

the infinitely long string and the segment. Neither of the
two can be considered a particle. The physical phenomenon
consists of the effect that the field of the string has on the
charge of the segment. We wish to find the force of this
interaction. The charge Q, = yl carried by the segment
is positioned in the electric field of the string, which is
known (see (19.4)).
It would seem that to find the force acting on the charge
we need only use the formula F = Q,F, where E = y,/
2n€gry. This is not correct, however, since the formula is
valid either in the case of a homogeneous electric field
(the electric field generated by the string is nonhomoge-
neous: E =< const) or in the case of a point charge (Q, is
distributed over the segment). On different sections (of
equal length) of the segment of length / different forces
are acting. Therefore, to calculate the force with which
the nonhomogeneous field generated by the string acts on
the distributed charge Q, we apply the DI method. We
partition segment J into sections of length dz so sinall
that the charge dQ = y, dz of each section can be consid-
ered a point charge. The charge dQ is in the electric
field of the string. Since this is a point charge, the force
 Ch. 6. The Electric Field 133

acting on It ts
dF = E dQ = the dz, (49.44)
where z is the distance from charge dQ to the string.
We now have the differential of the sought quantity.
The force acting on cach section of the segment depends
on the distance z from the segment to the string, and so
we select z as the variable of integration (it varies from
Zt} =o to x =r, - Ll. Integrating Eq. (19.41) with
respect to z, we get
rot d '
Vi¥2 GF _ Vive
E= \ Qneg = 2, In (1 +) :
ro

Substitution of numerical values yields the result F ~

1.2 x 10-3 N.
The terms of the above problem can be changed by plac-
ing the segment parallel to the string, at an angle to the
string, in a plane perpendicular to the string, and so on.
All these variants can be solved by the same DI method.
Let us consider fields generated by curved charged
lines, surfaces, etc.
Example 19.4. A puece of string ts laid out in the form of
a semicircle of radius R = 2 m and is charged uniformly
with an electric charge Q = 10-® C. Find the strength of
the electric field generated by the string at the point that ts
the geometrical center of the semicircle.
Solution. The physical system consists of two objects,
the semicircle uniformly charged with Q and the electric
field generated by the charge. The field strength is un-
known. The charge carried by the semicircle cannot be a
point charge since the size of the semicircle, nR, is com-
parable to the distance R considered in the problem, so
that the solution
B= ae
mesh? ’ E~2.2 Vim, (19.12)
is wrong. Gauss’ law involves extremely complicated cal-
culations. We, therefore, turn to the DI method.
Let us select the inertial reference frame as the one
linked to the semicircle and direct the X axis as shown in
134 Part 2. Solutton of Standard Problems

Figure 19.5. We partition the semicircle into arcs of

length dl so small that the charge d@ = Q dl/xR carried
by an arc can be considered a point charge. We take one
such point charge. It generates an electric field whose
strength vector dE, at a point A makes an angle a with
the X axis. Obviously, to each elementary charge in the
upper half-plane there corresponds a symmetrically placed
charge in the lower half-plane. The vector sum of dE,

Figure 19.5

and dE, is a vector directed along the X axis. Hence, in

the summation process we need only take into account
the projections of the elementary field strength vectors
on the X axis:
d dl
dE, = dE, cos a= zee cos a = SERS (19.13)

The first stage (finding the differential of the sought quan-

tity) has been completed. Let us go on to the second (in-
tegration, summation). We must select the variable of
integration. The position of a point charge on the semicir-
cle is defined by angle a, so it is natural to select this
angle as the integration variable. By definition angle a
is defined as the ratio of the length of the arc J to the ra-
dius R of the circle: @ = l/R. Since dl = R da, we have

dE,= 208% dar,

 Ch. 6. The Flectric Field 135

Integrating this equation with respect to angle @ yields

+VY2
E= \ aaatar os ada= she, E~
1.4 V/m.
-n/2
(19.14)
We sec that the correct result (19.14) differs considerably
from the incorrect one (19.12). If we transform formula
(19.14) by introducing the linear density of the electric
charge carried by the semicircle, y = Q/nR, we find that
the strength of the electric field at the center of a uni-
formly charged arc in the form of a semicircle,
7 v
E=—eR
is given by the same formula as the strength of the elec-
ctric field generated by an infinitely long and uniformly
charged string (19.4).
The same method can be used to calculate the field
generated by a uniformly charged ring (half-ring, etc.)
at any point lying on its axis at a distance z from the
ring’s plane. Having calculated the field of a ring, we
can formulate the problem of the motion of a charged par-
ticle along the ring’s axis; for 7 < HR this motion consti-
tutes harmonic oscillations. After this is done we can
formulate the problem of calculating the electric field gen-
erated by a uniformly charged hemisphere, a part of a
sphere, etc.
Example 19.5. At the center of a hemisphere uniformly
charged with electricity with a surface charge density o there
ts placed a freely oriented point dipole with an electric
moment p. Determine the potential energy of the dipole and
the period of the dipole’s small oscillations about an azxis
perpendicular to the symmetry axis of the hemisphere. The
moment of inertia of the dipole about the rotation azis is J.
Solution. The physical system consists of the uniformly
charged hemisphere and the dipole. Neither can be
thought of as a particle. A dipole is said to be a point dipole
if its length is so small that in any nonhomogeneous field
(the field generated by a hemisphere is nonhomogeneous)
the torque acting on the dipole can be calculated by the
136 Part 2. Solution of Standard Problems

formula
M=pxE, (19.15)
where p is the electric dipole moment. As is known, in a
homogeneous field this formula holds true for any dipole.
To solve the problem, we must first calculate the field
of the hemisphere (its electric field vector E) at the hemi-
sphere’s center. We apply
the DI method. We partition
the hemisphere into narrow
rings and consider one rin
(Figure 19.6). The charge o
the ring is dQ = 2nR*o x
sin 6 d6, where # is the ra-
dius of the sphere.
The projection of the
elementary strength vector
Figure 19.6 dE of the field generated
by the ring on the X axis
(the symmetry axis of the hemisphere) at point 0 is
dE, = qo
d
008 0 = SENT
in Ocos 6 dé
(19.16)
Integrating this equation with respect to 6 from 6, = 0
(the farthest ring) to 6, = 2/2 (the closest ring), we find
that
n/2
a osinOcos@d@ oa c
oa Be tsC (19.17)

Since the torque (19.15) acting on the dipole is hnown,

we can use the equation of motion (14.7) to oblain the
differential equation of the small oscillations of the di-
pole:
Jo=—pEg, or 9+ oy =0,
where w; = pE/J. The sought period is
T=4n V 2,J/po.
To calculate symmetric fields (the electric field strength
of an infinitely long cylindrical surface, of an infinite
 Ch. 6. The Electric Field 137

cylinder, of a sphere, elc.) it has proved expedient to

employ Gauss’ law of flua.
Example 19.6, A straight infinitely long cylinder of ra-
dius Ry = 10 cm ts uniformly charged with electricity with
a surface charge density 6 = 74-10-!* C/m?. The cylinder
serves as a source of electrons, with the velocity vector of
the emitted electrons perpendicular to tts surface. What
must the electron velocity be to ensure that the electrons can
move away from the azis of the cylinder to a distance greater
than r = 108 m?
Solution. The physical system consists of two objects:
the positively charged cylinder and an electron. The phys-
ical phenomenon consists of the electron moving in a
decelerated manner in the electric field of the cylinder.
We wish to find one of the parameters of motion, the
electron velocity.
To describe the motion of the electron we must first
calculate the electric field of the cylinder. The charge on
the cylinder cannot be considered a point charge. We
apply Gauss’ law. For this we surround the cylinder with
a cylindrical surface (coaxial with the cylinder) of arbi-
trary radius r > R, (Figure 19.7). In view of the sym-
metry of the problem, the electric vector E of the field
of the cylinder is perpendicular at all points to the con-
structed cylindrical surface. Hence, the flux of E out of
the cylindrical surface of length L is
, = 2nrLeE.

By Gauss’ theorem,
2urLE = 2nkyLa/ey.
whence
Ea fet
or
(19.18)
Now, by applying the dynamical method we find that
Newton’s second law yields
dr Ro
Ne r=
We dt? S €or’

where m, is the electron mass, and e the electron charge.

From the standpoint of physics the problem is solved,
{38 Part 2. Solution of Standard Problems

It would be solved completely if we were to solve the

above differential equation and obtain the law of motion

cp J
of the electron r =r (t). Knowing this law, we could

: Ma
|

Figure 19.7 Figure 19.8

find the law of variation of the electron’s velocity with

time, v = r (t), and so on. But instead let us apply the
law of energy conservation. By this law,
Meve
— ep)-=— eg, (19.19)
where (oq is the potential of the cylinder, and @ the poten-
tial of the field of the cylinder at a point r distant from
the cylinder’s axis. Employing the relationship E =
—dq dr that exists between the field strength E and po-
tential ¢ and allowing for (19.18), we arrive at the fol-
lowing differential equation:
Roo _ dep
er r*
Integrating, we find that

p= ~ In r+e, (19.20)
with c an arbitrary constant. Hence,
yee In R, +e. (19.21)
The system of equations (19.19), (19.20), and (19.21)
yields the following value for the sought initial velocity
 Ch. 6. The Electric Field 139

of the electron:

a YV Beate Vp © 3.7 x 105 m/s.

e

Concluding this section, we consider the following prob-

em.
Example 19.7. A solid ball made of an insulator (2 ~ 1)
has been drilled along the diameter and air has been removed
from the cavity. An electron is placed in the cavity.
What is the magnitude of the positive charge that should be
imparted to the ball if we want the ball to perform harmonic
oscillations in the cavity with a given frequency vo (the
charge is assumed to be evenly distributed over the ball’s
volume)? Assume that the cross-sectional area A of the cavity
is considerably smaller than nR*, with R the radius of the
ball.
Solution. The problem is similar to Example 12.1 con-
cerned with oscillations of an object in a shaft dug along
Earth’s diameter. To solve it we must calculate the clec-
tric field strength inside the ball. Let us apply Gauss’
law. Suppose that the volume density of the charge, p, is
equal to 3Q/4nR%. We take an arbitrary point z distant
from the center of the ball and draw a sphere of radius r
centered at the ball’s center O and passing through thal
point (Figure 19.8). The flux of vector E out of the sphere
is, in view of the symmetry of the field,

O,=E x 4nz’.
By Gauss’ law,
123
Ex 4a77 = & p ;
£9
whence

E=/-c.
3eq

Thus, the force acting on the electron is

pe
F. 3e, Q.
140 Part 2. Solution of Standard Problems

From Newton's second law we get the differential equa-

tion of the electron’s harmonic oscillations:
mrt = peg
325
Consequently, the angular frequency wo» is equal to
V pe/3em,. Since w, = 2nv,, we can find the sought
volume charge density,
p= 12n2e,vim,/e,
and the charge on the ball,

_4 77k 3Pp.
=

For v, = 10° I[z = 1 MIIz and R = 107! m we have

p ~ 6 X 10-® C/m® and Q = 2.4 x 107" C.

20. The Electrostatic Field in Insulators

When considering the electrostatic field in insulators (di-

electrics), one uses Gauss’ law of flux,

hD-dA= 5)Q,, (20.1)

A
where
D=eE+P (20 2)

is the electric displacement vector, >)Q, the sum of the

free charges lying within the closed surface A, and P
the polarization vector.
The strength E of an electric field in an insulator and
the polarization P are interconnected through the fol-
lowing relation:
P=.e, (e—1)E. (20.3)
Thus,
D=e,eE. (20.4)
By the superposition principle, the electric field E in
an insulator is the vector sum of the electric field generat-
ed by the free charges, E,, and by the bound charges, E’:
E=E,+E’. (20.5)
 Ch. 6. The Electric Field 144

The surface density of the bound charges is

o’ = P, = & (e—1) E,, (20.6)
with P, and E, the normal components of the polariza-
tion and the field strength.
In calculating the field it is advisable to use one of the
following two methods.
The first is based on the superposition principle (20.5);
for the sake of brevity we will call it the superposition
method. First the field generated by the free charges
(sometimes called extraneous charges), Ep, is calculated.
Then the field E’ generated by the bound charges is cal-
culated. Finally, (20.5) is used to find the electric field
vector in the insulator. This enables us to find the expres-
sion for the potential @ of the field in the insulator. Not
all is as simple in this method as it might seem at first
glance. Often the DI method must be used (see Section 6),
and difficulties emerge in determining the density of the
bound charges o’ (according to (20.6) this depends on E,,
which is also unknown) and the field E’ generated by
these charges, and other subtleties, which will be dis-
cussed later.
The second method uses Gauss’ law (20.1) to find the
electric displacement vector D. Then (20.4) is employed
to determine the electric field vector E in the insulator.
If necessary, potential @ is calculated via (16.6). Gauss’
method (for the sake of brevity we use this name to desig-
nate the second method) often leads to results faster and
More simply than the superposition method, but some-
times Gauss’ method proves inapplicable, while the super-
position method can be applied even in such cases.
Note that in the majority of problems in this section
the following conditions are assumed met: the insulators
are homogeneous and isotropic and their boundaries coin-
cide with equipotential surfaces.
Example 20.1. A charge Q = 10-® C is imparted to one
of the plates of a plane-parallel capacitor with a surface
area A = 0.2 in? (the other plate is grounded). The distance
between the plates isd = 2mm. A glass plate and a por-
celain plate are inserted between the plates of the capacitor
142 Part 2. Solution of Standard Problems

(parallel to them), with the thickness of the first being d,=

0.5 mm and that of the second d, = 1.5 mm, so that there
ts no air gap between them. Determine the electric field

Figure 20.1

strength in each plate and the surface densities 0’ and a” of

the bound charges carried by these plates (Figure 20.1).
Solution. The physical system consists of the capacitor,
whose plates carry free electric charges with a density
o = Q/A, and the two insulators, on which bound elec-
tric charges with densities o’ and o” appear. We wish
to determine the electric field strengths FE, and E, in the
insulators and the densities o’ and o” of the bound char-
ges. This constitutes a basic problem of field theory. We
apply both methods.
The superposition method. In each insulator the field is
generated by the free charges carried by the two parallel
capacitor plates and, respectively, by the bound charges
o’ and o” also carried by two planes. Note that the bound
charges generate a field that is nonzero only inside “its
own” insulator. Obviously, the electric fields generated
by all types of charges are
o_— Q ,_ oO" ,_ o"
Pag
ag 8 BT age ne He
Since according to (20.6)
o’=e,(e,—1)£, and o”’=e, (e,—1) £,,
and according to (20.5)
E,=E,—E; and £,=E,—E,,
 Ch. 6. The Electric Field 143

we solve the system of equations and obtain

1
~9e814 ’ 2
pi 9 eoe2A ”
gr a= 0
8A (20.7)
eo’ = (€2—1)Q

eA

Gauss’ method. Using Gauss’ law, we can find the elec-

tric displacement vector in both insulators:
DAA=cAA, D=0=QI/A.
Next, using (20.4), we can find the electric field strengths
E, and E, in the insulators,
=e E =
i €9€,A : 2 &o€oA :

and by employing (20.6) we can calculate the densities

o’ and o” of the bound charges,
1_
al aerry gaat(1-1) » =_ V2 (&2—1)Q
o/s
£24 .

which coincides with the results (20.7) obtained by the

superposition method.
Example 20.2. Two infinitely long thin-walled coazial
cylinders of radii Ry =5cm
and R, = 10cm have been
uniformly charged electrical-
ly with surface densities
o, =10 nC/m? and o,=
—3 nC/m?. The space be-
tween the cylinders is filled
with paraffin (e = 2). Find
the strength E of the field at
points that lie at distances
r; =2cm, re =6cm, and
rg = 15 cm from the azes
of the cylinders.
Solution. (1) The superpo-
sition method. The total field
is generated hy four charges: Figure 20.0
the free charges with densi-
ties o, and o, and the bound charges with densities —o,
and +o; (Figure 20.2). The bound charges generate a field
144 Part 2. Solution of Standard Problems

that is nonzero only inside the insulator (paraffin). Clear-

ly, the field at point A (positioned at a distancer, = 2cm
from the axes) is zero (this can be proved by applying
Gauss’ law for a vacuum).
Let us now consider point B (positioned at a distance
r, = 6cm from the axes). At this point the field is gener-
ated by the charges with densities o, and —o; (the field
generated by the charges +o; and oy at this point is
zero). By Gauss’ law (for a vacuum), the field generated
by 9, is
2 Ryo, A
E,= at (20.8)
In the same manner we can find the field generated by oj:
»_ Ryo, 1
E\ = at 7 (20.9)
According to (20.6),
0; = &y (e — 1) E (Aj). (20.10)
It is important to note that E (R,) in (20.10) is the
strength of the total electric field in the insulator at a
point that is R, distant from the axes. This quantity is
unknown, but let us relate it with Z (r,), the total elec-
tric field in the insulator at point B. Since both E, and
FE; are inversely proportional to the distance r, from
point 2 to the axes, the total field strengths E (R,) and
E (r.) must obey the same condition:
E(Ri) _ re
Fry Re (20.11)
ITence,
E\ = (e—1)
E (ry).
Since according to (20.5)
E (r.) = E,— E},

we get

E (r.) eee are

cdyt 20.12
(20.12)
This yields E (r,) ~ 4.7 x 10? V/m.
At point C lying at a distance rs = 15 cm from the
axes, the field is generated only by the free charges with
 Ch. 6. The Electric Field 145

densities o, and o,:

This yields E (rz) ~ 1.5 xX 10? Vim.

(2) Gauss’ method. Let us first calculate the field at
point B. By Gauss’ law of flux,
D X 2ar,l = 2aR,o,l,
which yields D = Rjo,/r,. Now, using (20.4), we can find
the sought field:

E (r,)=eS Ene T; ?

which coincides with the expression (20.12) obtained by

the superposition method. Hence, E (r.) ~ 4.7 x 10? V/m.
If the field (i.e. the electric vector E) is known, other
quence such as the potential and the energy, can be
ound.
Example 20.3. Two concentric metal spheres of radii
R, = 4 cm and R, = 10 cm carry charges Q, = —2 nC
and Q, = 3 nC, respectively. The space between the spheres
is filled with ebonite (e = 3). Determine the potential
of the electric field at distances r, = 2 cm, rz = 6 cm,
and rz; = 20 cm from the common center of the spheres.
Solution. (4) The superposition method. The total field is
generated by the free charges Q, and Q, and the bound
charges Q; and Q’ (Figure 20.3). To find Q; and Q7 we must
know the field strength & (r) in the insulator. This quan-
tity can be found by applying Gauss’ law:

E(r)= Qi (20.13)
4sle,er® *

Suppose that o, and o7 are the surface densities of the

’
bound charges Q; and On respectively. Then from (20.6)
and (20.13) it follows that
’ —-1
0, = 69 (@— 1) Em = pe

7- = & (¢ ~1) Ens = Tene

—1

10—0498
146 Part 2. Solution of Standard Problems

Thus,
Q, =0,4nRi = eaOer
QQ = o%4nR? =: LAN
=) Qi
(20.14)
It is known that for a uniformly charged sphere of radi-
us R the potential of the field (in a vacuum) inside the

Figure 20.3

sphere and on its surface is

oar: (20.45)
while at points outside the sphere the potential is
ico

where Q is the charge on the sphere, and r the distance

from the center of the sphere, O, to the point in question.
Since the charges (Q,, Qs, Q;, and Q’) generating the field
are distributed over spherical surfaces, by allowing for
the superposition principle and Eqs. (20.14), (20.15), and
(20.16) we can determine potential @,, at the point A
positioned at a distance r, from the center:

Por = Pi + Pot Pa t+
— 91 Q' —9% Qz
~~ Ane gR, lp 4negR, dp 418)R, nie4ne,R,
 Ch. 6. The Flectric Field 147

a Q1 eres (e—1) Qi 4. Q2
~ ~ “Ane sR 4neeR, AnegeRs 4neghs
pe Os OO, QQ. 7
7 4ne eR, 47e eRe + 4negR, ’ (20.17)

the potential @gg at the point B positioned at a distance

r, from the center:
Q1 . (e—1)Q1 ae Ce Qe
Lo2 = — 41ers 4neere 4ne eR, 4ne,Ro
= Qi (&—1)Qi Qe.
~~ ~ 4rtegere 4neeRe r 4nesR, ’ (20.18)
and the potential qo, at the point C positioned at a
distance r, from the center:
Q1 glee es _ (e—1)Q1 Qs
Go = —
AN org 4negers 4ne gery Angry
= Q Q
a cre ce ane : (20.19)
(2) Gauss’ method. Using Gauss’ law, we find the field
strengths Eo,, Eo, and Eo, at points A, B. and C, re-
spectively:
Qi
Eu= Freed
—_ 92—-Q1
Eu = 0, Eqs = “Gegry (20.20)
Since in our case the potentialj@ is continuous, employing
the relationship (16.6) between field strength and poten-
tial we can find the value of the potential at any point if
we know the potential at least at one point. This poten-
tial is @o3 given by (20.19) because it is generated by the
free charges Q, and Q, or the potential at the surface of
the second sphere,
9 (R,) = Ge, (20.21)
Integrating (20.13) with respect to r from r, to Rz, we
obtain
+ (Ry) — 9 (m2) = A
Ane erg —
4nejeR,*
This yields 9 (r.) = qo2:
Qi (@ ~1)01 sis 1Qe
Po2= — ANeoere 47e eRe 4ne,R, *
10*
148 Part 2, Solution of Standard Problems

which coincides with the expression (20.18) obtained by

the superposition method.
Integrating (20.13) from R, to r,, we find that

@ (To) — ¢ (Ry) =es Teh,

| eee eee
Tnerer, :
This yields ¢ (Ry) = Qo:
G1 (0 1) Ar
Poi= ““GneeR, "Geshe
47eeR, 0 ane
arhee

which coincides with the expression (20.17) obtained by

the superposition method.
This problem can be made more complicated if we add
one or several charged concentric spheres and place
different insulators between
the spheres and inside the
first sphere. Clearly, all
these problems can be sol-
ved either by the superpo-
fe,0 sition method or Gauss’
g method.
Nae Let us now consider a
problem that, if solved by
the superposition method,
requires special care, with
the precise analysis and
calculation of the fields
Figure 20.4 generated by different
charges.
Example 20.4. A thick-walled hollow ball of glass (e =
7) is uniformly charged over its volume with a density
e = 1.5 pC/m3. The inner radius of the ball is R, =2cm
and the outer R, = 6 cm. Find the distribution of the po-
tential in the glass and calculate the potential at the outer
and inner surfaces of the ball and at its center.
Solution. (1) The superposition method. Let us first find
the distribution of the potential in the glass of the ball,
that is, the potential at an arbitrary point A that is r
distant from the ball’s center with Ry << r< R, (Fig-
ure 20.4). What charges generate the field in the insulator?
 Ch. 6. The Electric Field 149

First, this is the bound space charge of the insulator

Q=54(Ri—R})p (20.22)
and, second, the bound charge Q’ on the outer surface of
the ball, which according to (20.14) is

O's fae, (20.23)

Let us find at point A the potential of the field generat-
ed by the bound space charge Q. Since this is not a
point charge, we must employ the DI method. But if we
apply this method formally, we may arrive at a wrong
result. In view of the symmetric distribution of charge
Q, we partition the hollow ball into thin concentric spher-
ical layers whose thickness dz is so smal] that the ele-
mentary charge
dQ = 4nz* dzp (20.24)
(where z is the radius of the layer) of each layer can be
thought of as distributed over a sphere. Then, according
to the superposition principle, the total potential at
point A is equal to the sum of the potentials of the fields
generated by the elementary charges on these spheres.
But the elementary potential generated at point A by a
sphere must be calculated by different formulas, (20.15)
or (20.16), depending on whether point A is an interior
or exterior point in relation to the particular sphere.
To allow for this complication, we draw through point
A a sphere of radius r centered at point O (see Fig-
ure 20.4). This sphere divides the hollow ball into two
sublayers: a sublayer with radii R, and r and a sublayer
with radii r and R.. In relation to the first sublayer,
point A is always an exterior point with respect to the
elementary spheres of the sublayer, while in relation to
the second, point A is always an interior point. Consider
an elementary spherical layer of thickness dz in the first
sublayer (Figure 20.5). Its elementary potential at
point A is, according to (20.16),
dQ
dg = “Freer . (20.25)
150 Part 2. Solution of Standard Problems

where we have allowed for the fact that the charge dQ

is located in an insulator.
The validity of (20.25) in the superposition method is
justified in the following manner. The potential at point
A is generated by the free charge dQ and the bound charge
—dQ’ = — [(e — 1)'e) dQ distributed over the sphere of

Figure 20.5 Figure 20.6

radius z. The total potential generated by these charges

at point A is
dQ dQ’ dQ (e—1)d@__ dQ
4Ne€or 4neyr ~~ 4megr 4neer = Astejer

which coincides with (20.25).

Integrating (20.25) with respect to z from R, to r, we
find the potential @, generated at point A by the charge
of the first sublayer:
r

Be ial =Lee
gp 4ne,er 2 Fit)°
9
(20.26)
See (r r
v1 = \
R
We now consider the second sublayer and take an ele-
mentary spherical layer of thickness dz in this sublayer
(Figure 20.6). The elementary potential generated at point
A by the charge on this spherical layer is, according
to (20.15),
ee ~~ Gmteger
(20.27)
 Ch. 6. The Electric Field 151

Allowing for (20.24) and integrating (20.27) with respect

to z from r to Re, we find the potential @, generated at
point A by the charge of the second sublayer:
Re
ed ee
4negez 286 ee “2e9e-
r

Finally, the potential @, generated at point A by the

bound charge Q’ (see (20.23)) is, according to (20.15),
—1)(R3—R?
ge SESSet ADD.
ve (20.29)
Thus, the total potential generated at point A by all
charges is
G=Ht 24+99= — Zee (F+AL)+C, (20.30)
2 R?

with
_ PRI ta
C= (e— 1) (R3—Ri)p (20.31)
a constant. Let us transform C to a form needed later:
C= A(R}—Ri)e
eoSER; , oar
9(R34+2R)
77), Pea
Liang (20.32)

Knowing the potential distribution (20.30), we can

find the potential at the outer ca of the ball,
= e_ (_R
(y= =a. (=
2 +t L)+¢,

and at the inner surface,

o(Ry= 2ee
—S+¢
The potential at the center of the ball, go, is equal to the
potential at the inner surface, g (2).
(2) Gauss’ method. Using Gauss’ law, we find the mag-
nitude of the electric displacement vector at point A
(see Figure 20.6):

Dx 4ar?= = a(r>— Ry p,

D=£(r-#t).
3 r?
(20.33)
152 Part 2. Solution of Standard Problems

Next we find the magnitude of the electric field strength

at the same point:
ual Ri
E= Bene (rt). (20.34)
Upon integrating the relationship E = —dq/dr we get
the potential distribution in the ball:
= p R3 = p jr? , Ri
o= — | Beck (r—-—+
+ )dr= = sag (F4+—+)4+¢,.
(20.35)
The integration constant C, will be found from the requi-
rement that the potential be continuous and the fact that
the potential at the outer surface of the ball is determined
solely by the value of the free charge Q (20.22):
_ Q_ __ (R}—Ri)p
¢ (Fa) = 47e Ry = ae,
Substituting this into (20.35),
(R§—Ri)p _ e Ri | R
TBighg gE (> +3) +61,
we arrive at the following expression for the integration
constant:
(R3—Ri)e | (R}+2R%) "
ae ao a Gees.
aT ee r (20.36)
Thus, the final result is

RE), (RI—RYp , (RI+2Re

Gag (at ])
— aan;
3eoR, Gah
aGe eRe ee

(20.37)
which coincides with the expression (20.30) obtained by
the superposition method if we allow for the value of C
(20.32).
The problem can be made more complicated if, say,
we put in the cavity a metal ball or a dielectric ball (with
a different dielectric constant e,) uncharged or charged
over its volume with another charge density 0,, etc. For
example, from (20.37) we can obtain an expression for
the potential distribution inside a uniformly chaiged
 Ch. 6. The Electric Field 153

solid ball of radius R,

ae =
9= >a Ttoe (14+37).

and at the center of the ball,

PRE (1+).
9 (=F (yp A
All these and similar problems can be solved by apply-
ing either the superposition method or Gauss’ method.
The reader must have noticed that in all the problems
solved in this section Gauss’ method proved to be simpler
and led to a result faster than the superposition method.
However, let us not hurry with conclusions but rather
consider a few more problems.
Example 20.5. A sufficiently long, round cylinder made
from a homogeneous and isotropic insulator with a known
dielectric constant © is placed in a homogeneous electric
field E, in such a manner that the cylinder’s axis coincides

(oe)
f ,
-@ +Q

———————

Eo
Figure 20.7

with the direction of Ey (Figure 20.7). Determine the elec-

tric field strength near the cylinder (inside and outside).
Solution. Clearly, Gauss’ method is useless here. Ap-
plying Gauss’ law, we arrive at the trivial identity D, =
D, expressing the continuity of the normal components
of the electric displacement vector. Let us apply the su-
perposition method. By FE, we denote the electric field
strength inside the cylinder and by E£, the electric field
Strength outside. Owing to the polarization of the insula-
tor, bound charges —Q’ and -—-Q’ gather on the bases of
the cylinder with a density o’. The resulting electric
fields E, and E, are the vector sums of E, and the elec-
tric felds generated by the bound charges —Q’ and +-Q’.
154 Part 2. Solution of Standard Problems

Let us now discuss the meaning of the words “sufficient-

ly long cylinder’. The cylinder considered here is so long
that the field generated, say, by charge +@Q’ is weak in
the vicinity of charge —Q’ and can be neglected in com-
parison to the field generated by —Q’ in that vicinity.
The same is true of the field generated by —Q’ in the vi-
cinity of charge +-Q’. Thus,
E, = Ey = E’", E, = Ey 4- E’,

where E’ is the electric field strength generated by —Q’

(or +-Q’). Let us find E’.
E’ is the field of a uniformly charged disk. Applying the
DI method, we find (see Figure 20.7, Example (18.5),
and formula (48.20)) the projection of the elementary
electric field vector on the disk’s axis generated by a
thin ring (the X axis is directed along the axis of the
disk):
dE. = rdQ = 3aro’x dr
x AN (r? 4- ?)9/2 4ne, (r®4-27)3/3 °

Integration with respect to r from zero to R (the radius of

the disk) yields the electric field strength generated by
the disk (or the field of the bound charge —Q’):
R
, ° o'zrdr _ 90 tice 2
E =E.= \ 2€q (r® + 3*)8/2 = 28 [1 - V 2?4- R? i

(20.38)
i—]

From this it follows, for one, that E’ is roughly zero

when z is very large. This completes the justification for
using the term “sufficiently long cylinder”.
Near the base of the cylinder z ~ 0 and
E' = 0'/(2€,). (20.39)
Allowing for (20.6), we obtain
Ey a Eo = £y — Ey

and, hence,

| pane Oe (20.40)
 Ch. 6. The Electrie Field 155

Note that the £, in (20.6) is the electric field strength in-

side the insulator. In our case E, = E,. Then (20.6)
yields
e—1
o = 285 r= 0 Eo. (20.4 1)

From (20.39) we obtain

r] e—i

~ e+! E,
Hence,
2e ‘ :
E,= cess E,- (20 42)
Example 20.6. An infinite homogeneous, isotropic insu-
lator in which a specified homogeneous electric field Ey has
been created contains a spherical cavity of radius R (Fig-
ure 20.8). A point dipole with an electric moment pis placed
at the center of the cavity.
Determine the period of
the dipole’s small oscilla-
tions if the moment of
inertia of the dipole about
the rotation axis is J.
Solution. The problem
is similar to Example
19.5. We can easily find
the period of the dipole’s
small oscillations if we
know the field in the Figure 20.8
cavity. Clearly, Gauss’
method is useless here. Let us apply the superposition
method. Owing to the polarization of the insulator.
bound charges -}-Q’ and —Q’ gather on the two hemisphe-
res, and the densities o’ of these charges are not constants.
To calculate the field generated by a charged hemisphere
with a variable surface charge density o’ let us employ
the DI method. It can easily be demonstrated that on an
elementary ring the surface charge density is
o’ = 0, cos 8, (20.43)
where
Oo = £5 (e — 1) Ey (20.44)
156 Part 2. Solution of Standard Problems

is the maximum surface charge density (at point A in

Figure 20.8). Then the projection of the elementary elec-
tric field vector dE generated by
the ring at point O on the X axis
(the symmetry axis of the hemi-
sphere) is, according to (19.16),
R 0, sin 8 cos? 6 d6
dE. => ~ a, .

Integration yields the electric

field generated by one charged
Figure 20.9 hemisphere:
n/2
E ; oosinOcos?0d0 (e—1)E,
ae \ oe ee

The field generated by the two hemispheres is

—A)E,
en, eee ~LL
$ .

Thus, the sought field at the center of the spherical cavity

is
E=E,+£,=—t* &,. (20.45)
Since the field is now known, the subsequent solution of
the problem on the oscillations of the dipole is obvious
(see Example 19.5).
To conclude this section we examine another problem
involving estimation.
Example 20.7. An ebonite ball of radius R is uniformly
charged with electricity with a volume density p. What is
the radius R, of the sphere that divides the ball into two
parts whose energies are equal.
Solution. Let us draw a sphere of radius R, (Figure 20.9).
We must determine the energy W, of a ball of radius R,
and the energy JV. of a spherical layer with radii R,
and A. For this we must know the field in the ball. This
can easily be found by employing Gauss’ method. By
 Ch. 6. The Electric Field 157

Gauss’ law of flux,

D x 4ar? == > mrp.

Thus, the field inside the ball is

er
See °

Applying the DI method, we find the energy dW of the

field existing inside a thin spherical layer of thickness dr:
dW = wd = 282 dart
dr= 22
Ee
rtdr,
where w is the energy density of the electric field. Inte-
grating, we obtain
_._2mp?
RE _ arp? (A> — Rt)
i 45e,e ” We 45298 ¥
Since W, = W,, we find that
R
R= v5 ~~ 0.87R.
The numerical answer is somewhat unexpected: the
outer spherical layer whose thickness is only (approxima-
tely) one-tenth of the radius contains half of the energy
of the entire ball.

21. Conductors in an Electrostatic Field

The surface of a conductor constitutes an equipotential

surface. The method of images is based on this property.
This method makes it possible to calculate various elec-
trostatic fields, determine the capacitance of a system of
conductors, etc.
The method of images relies on the following state-
ment: if in an arbitrary electrostatic field we replace an
equipotential surface with a metal surface of the same
shape and create the same potential on the metal surface,
the electrostatic field does not change.
Let us consider the electric field that exists in the space
between a point charge +@Q and an infinite metal plane
whose potential is zero. In view of the above principle this
field is equivalent to an electric field generated by the
158 Part 2. Solution of Standard Problems

given point charge +-Q and a point charge —Q that is

the mirror image of the given charge --Q in the metal
plane (Figure 21.1).
Example 29.1. A point charge Q = --2 x 10-8 C is
placed at a distance 1= 1 m from an infinite metal plane
that is grounded (Figure 21.1). Determine the interaction
between the charge and the plane.
Solution. The metal plane is in the electrostatic field of
the point charge. Owing to electrostatic induction, on the
side of the plane closest to the charge there appear induced

Pp

Figure 21.4 Figure 21.2

electric charges of the opposite sign. Hence, a force of in-

teraction develops between the given point charge and
the charges induced on the metal surface. By the hypothe-
sis, the potential of the metal plane is zero (Earth's po-
tenlial is assumed to be zero). Consequently, according
to the method of images, the electric field existing in
the space between the point charge and the metal surface
is equivalent to the field generated by the given point
charge and its mirror image in the metal plane. According
asCoulomb’s law, the sought interaction is given by the
orce
Q?

F F2x9x 107 N.
= "Gme, (21)? ?
Example 21.2. A point dipole with an electric moment p
is placed at a distance | from an infinite conducting plane.
 Ch. 6. The Electrie Field 159

Determine the magnitude ofthe force acting on the dipole tf

vector p is perpendicular to the plane.
Solution. According to the method of images, the field
generated by the given dipole and the charges induced on
the plane is equivalent to the field generated by two di-
poles, namely, the given dipole and its mirror image in
the plane (Figure 21.2). The dipoles are separated by a
distance of 2l. Thus, the sought force F is that with which
the image-dipole acts on the given dipole. It can easily
be proved that the field strength E at a point on the di-
pole’s axis at a distance r > l, (the dipole’s length) is given
by the following formula
—_2P
E= 4megr3 *
Hence, the force acting on the given dipole is
F = F, — F,,
where
_2pQ__
=QE,= “Gre gr?
is the force acting on ae negative charge of the given di-
pole, and
= = 2pQ
ie
the force acting on the positive charge of the given di-
pole. Allowing for the fact that r = 2l and p = Ql, 21 > 1,
and performing simple manipulations, we obtain
3p?
ek: 271egl* *

Example 21.3. A thin, infinitely long string is uniformly

charged with electricity with a linear density t and is placed
parallel to an infinite conducting plane at a distance |
from the latter (Figure 21.3). Find (a) the magnitude of the
force acting on a section of the string of unit length, and (b)
the distribution of the surface charge density o (x) in the
plane, where x is the distance from the plane that is perpen-
dicular to the conducting plane and passes through the
string.
160 Part 2. Solution of Standard Problems

Solution. Clearly, to find the force acting on a segment

of the string of unit length we must calculate the field
generated by the mirror image of the string (see Figure
21.3). The field generated by the string can easily be

Figure 21.3

found, say by employing Gauss’ law of flux (see (19.4)):

Tt
E = Qaeor .

In our case r = 21. Thus, the force acting on the segment

of the string of unit length is
12

F= Et= Gnegl ‘

To carry out the second task, we determine the field gen-

erated by the string and the string’s image at point A
(see Figure 21.3):
| ee 2tl _ tl
1° One, (z?+1%) ~ eg (z? +P) *
The induced charges generate a field £, near the conduct-
ing plane (outside it), which we find by employing
Gauss’ law:
E, =oley. (24.4)
 Ch. 6. The Electrie Field 161

These induced charges distribute themselves in such a

manner that their field inside the plane neutralizes the
effect of the external field E, (i.e. the field inside a con-
ductor placed in an electrostatic field is nil):
E, + E, = 0. (24.2)
Consequently,
a tl
= aie +y °
Thus, in the method of images we are most often con-
fronted with the problem of calculating the field charac-
teristics of specified charges and their mirror images,
that is, solving the basic problem of field theory.
Example 21.4. A very long straight string has been uni-
formly charged with electricity with a linear density +t and
is placed at right angles to an infinite conducting plane in
such a way that its lower end ws 1 distant from the plane
(Figure 21.4). Point O is the trace of the string on the plane.
Determine the surface density of the charge induced on
the plane (a) at point O, and (b) at a point A that isz
distant from point O (in the plane).
Solution. Using the method of images, we first calculate
the field generated by the string and its image at point O.
To determine the field generated by the string alone at
point O we employ the DI method. The point charge
dQ = tdr carried by an element of the string of length
dr generates at an arbitrary point on the string’s axis r
distant from the element (see Figure 21.4) an electric
field
dE= dQ __—_tdr
4neor? ~— 4egr? °
Integration yields

ee G
4neyr? ~ 4neyr *

In our case r = 1. Thus, the field strength E, generated at

point O by the string and its image is
Ey=Eaz.
T

11—0498
162 Part 2. Solution of Standard Problems

Allowing for (21.1) and (21.2), we find the density 0 of

the charges at point O in the conducting plane:
tT
Og = “onl ry

Let us now find the density o of the charges induced at

point A in the plane (Figure 21.5). To do this we must
again calculate the field of the string and its image, but
this timo the field génerated at point A. Applying the

Figure 21.4 Figure 21.5

DI method, we find the magnitude of the vector dE re-

presenting the elementary field generated at point A by
the point charge dQ = t dl of the element dl (r distant
from A) of the string alone:
dE dQ __itdl
~ Gnegr? ~—— 4egr® *
Since dl = rda/cos a and r = z/cos a, we have
tda
dE= tee
Figure 21.5 shows that the resultant electric field gen-
erated by the string and its image at point A is directed
along the Y axis and, hence, E,, = 0. Therefore, we will
only find the projection dE, of dE:
de u = ak sing a eee.
4n@yz
 Ch. 6. The Electric Field 483

Integration yields the projection of the electric field vector

of the string on the Y axis:
mj2
E =| tsinada tcosa,
vo 4nepzri(‘é‘éA Mg:
a4
Thus, the electric field of the string and its image gener-
ated at point A:
= _ tcosa, _ t
E, =2E, = Qnegt ome, (19-23) *
Allowing for (21.1) and (21.2), we can determine the
surface density of the charges induced on the plane:

o= Fairy air
Tt

Clearly, for x = 0 we have o = 4,4.

Example 21.5. A thin ring of radius R has been uniform-
ly charged with an amount of electricity Q and placed in
relation to a _ conducting
Sphere in such away that the
center of the sphere, O, lies
on the ring’s axis at a dis-
tance of | from the plane of
the ring (Figure 21.6). De-
termine the potential of the
sphere.
Solution. The conducting ;
sphere is situated in the Figure 21.6
field of the ring. We wish
to calculate the potential of the conductor. This con-
stitutes a basic problem of field theory. Since the field
is not symmetric, it is doubtful that Gauss’ law of flux
will lead to meaningful results. Let us employ the super-
position method.
The field of the ring induces charges of magnitude —Q’
and +Q” on the conducting sphere. The resultant field is
generated by three charges: Q, —Q’, and +@Q”. Hence,
according to the superposition principle, the potential of
the conducting sphere isp = g, -!- @, 4- Gy, where @, Go,
and gs; are the potentials of the fields generated by the
i1e
164 Part 2. Solution of Standard Problems

charges Q, —Q’, and -++Q”, respectively. But at what

point of the sphere? The answer is: at any point, since
the potential of a conductor placed in an electrostatic
field is the same for all points of the conductor.
In our case, the entire volume bound by the conducting
sphere is equipotential. Thus, we need only calculate
the potential at the most convenient point, the center of
the sphere. Indeed, notwithstanding the fact that we
know neither the values of the induced charges —Q’
and +Q” nor the distributions of the respective charge
densities —o’ and -}-o” over the sphere, we can state that
the total potential of the field of these charges at the
special point (the center of the sphere) is zero: @, +
, = O (the induced charges —Q’ and -+Q” lie at equal
distances from the center of the sphere, are equal in
magnitude, | —Q’ |= |-+Q* |, and are opposite in
sign). Hence, we need only calculate the potential g,
of the ring’s field at O (see Figure 21.6):
= Q (21.3)
41 = Gre? FR
This constitutes the sought potential of the sphere,
— = fi.
The terms of the problem can be made more complicat-
ed. Suppose that the charge Q on the ring is not distri-
buted evenly. Clearly, solution (21.3) remains unchanged.
Now suppose that in addition to the charge Q of the ring
there are other charges contributing to the field: a point
charge Q,, and charge Q, carried by a thin segment, etc.
It is easy to show that in all] these cases the solution is
reduced to calculating the field (the potential) of the new
charges Q,, Qo, etc. at the center of the sphere. The prob-
lem has proved to be nonstandard: we are supposed to
guess that the most “preferable” point is the sphere’s cen-
ter.
The capacitance
C = Q/g (21.4)
of a single conductor specifies the electric field that appears
outside the conductor and on its surface (potential @)
when a charge Q is imparted to the conductor. This im-
 Ch. 6. The Electric Field 165

plies that finding the capacitances of conductors is reduced

to calculating the potential of this field (i.e. constitutes a
basic problem of field theory).
Example 21.6. Determine the capacitance of a single
spherical conductor of radius R, surrounded by an adjacent
concentric layer of a homogen-
eous insulator with a dielectric
constant e and an outer radius R,
(Figure 21.7).
Solution. Let us impart a char-
ge Q to the spherical conductor.
Then there appears an electric
field at the surface of the con-
ductor and outside of it. If we
calculate the potential of the
conductor at its surface, (R,),
we can use (21.4) to find the Figure 21.7
capacitance C. Calculation of
the potential of the field (the field is symmetric) can be
done by Gauss’ method. By Gauss’ law of flux,
D X 4nr? =Q,
where R, <r < R,. Hence, the electric field in the insu-
lator

E
=F 4ne,er? *

After integrating the relationship E = —dg/dr, we arrive

at the following distribution of potential in the insulator:
-_ Qdr _ @Q
ai { 4neer? ~— Gneger +e.

The constant of seas blag can be found from the

condition @ (R.) = Q/4ne,R,:
ea Ce)
4neyeR, ~

Thus, the final formula for the distribution of potential

in the insulator is

ee
oa (ana no
166 Part 2. Solution of Standard Problems

Employing the condition that the potential is distributed

continuously, we can find the potential of the spherical
conductor,
4 e—1
p (Ry) = a (4-+ Rz )’

and its capacitance (21.4),

_~ @ _ 4ne eR,
@(Ri) 1+ (€—1) Ri/R, *
Example 21.7. The gap between the plates of a plane-
parallel capacitor is filled with an isotropic insulator whose
dielectric constant varies in the direction perpendicular to
the plates according to the linear law from e, to €, with
&) > &,. The area of each plate is A, and the distance be-
tween the plates is d (Figure 21.8). Determine the capacitance
of the capacitor.
Solution. We point the X axis upward and place the
origin on the lower plate (see Figure 21.8). Since € va-
ries according to the linear law, we can write
e=a-+ bz, (21.5)
where the constants a and b are determined from the
boundary conditions (e = e, at x=0 and e=e, at x=
d):
a=e,, b= (e,—8,)/d. (21.6)
Thus,
e=e+-—4—
a 4
Zz.

We impart a charge Q Lo the ee plate of the capacitor

and use Gauss’ method to calculate the strength of the
field generated by this charge:
ie oe SO
eeA &) (a-+bz)A °
After integrating the relationship E = dg/dz, or dg =
Q dz/e,A (a + bz), we can determine the potential differ-
ence Aq between the plates:
d

bom |careyyarn(t+ F)-

 Ch. 6. The Electric Field 167

Hence, the sought capacitance is given by the formula

c. @ pAb __— BoA (&2 ~)
= “Ap~ In(i-4-0d/a) dn (es/ex) *
This problem can be generalized by assuming that the
dielectric constant varies according to an arbitrary law
e =f (x), with f(z) an arbitrary function of z, say
{ (z) = x". It is easy to demonstrate that such problems
can be solved by the same method.
Example 21.8. Determine the capacitance of a spherical
capacitor with plate radii R, and R, and R, > R,; the

Figure 21.8 Figure 21.9

space between the plates is filled with an isotropic insulator

whose dielectric constant varies according to the law e =
alr, where a is a constant, and r the distance from the ca-
pacitor’s center (Figure 21.9).
Solution. We impart a charge Q to the inner plate of
the capacitor and use Gauss’ method to calculate the field
strength inside the insulator,

E ag 4ne,er? = 4nega ’

and the potential difference between the plates,

Re
Ap={ 4nena
2% - a 72—(R,—
4nea Ry)
Ry
168 Part 2. Solution of Standard Problems

Hence, the capacitance of such a spherical capacitor is

given by the formula
_ _4Neoa
cS R,—R, °*
To conclude this section we consider one more problem.
Example 21.9. Determine the capacitance of a section of
unit length of a two-wire line.
Solution. The formulation of the problem is incomplete.
Let us idealize the problem. We assume that the linear
charge density (charge per unt Iength) on one wire is

Figure 21.10

—t and on the other, +t. We also assume that al] other

bodies are so far from the line that their effect on the
electric field in the space between the wires can be ig-
nored. Finally, we assume that the wires have the same ra-
dius and that r < l, where 1 is the distance between the
wires. Thus, the physical system consists of three objects:
two infinitely long thin, straight parallel wires uniformly
charged with linear charge densities —t and --t and the
electric field generated by these charges. We wish to
find the capacitance of a segment of unit length of such
a system.
This problem is linked to the basic problem of field
theory. Let us calculate the field strength between the
wires at an arbitrary point A that is positioned at a dis-
tance x from the left wire (Figure 21.10). Employing the
superposition principle and the formula for the strength
of the field generated by an infinitely long straight, uni-
formly charged string, we get
t t
ee oneyt + 2ney (l—z) *
 Ch. 6. The Electric Field 169

Allowing for the relationship between field strength and

potential, we get
r
g= - \ Edx= — >t (Inz—In(1—2)]
+e,
where c is an arbitrary constant. This gives us the poten-
tials of the left and right wires:

i= — Frey Un —In (lr) +e,

= — Tre [ln(l—r)—In(r)J +e.
Next we find the potential difference between the wires:
Mo = 91 -- 92= Geo In—.
l—r

Since r <1 by hypothesis, we have

\e= Fe In+
Employing relationship (21.4), we can determine the ca-
pacitance of a section of unit length of a two-wire line:
Se
as ow
C= ag In@/r)’

22. Direct Current

The basic problem of direct-current theory deals with the

calculation (or design) of an electric circuit. In general
this problem can be formulated as follows: given an arbit-
rary electric circuit and some of its parameters (emf's, re-
sistances, etc.), find some other (unknown) quantities (cur-
rents, work, power, quantity of heat, etc.). Note that one
should assume current J to be the most important, or
fundamental, quantity in direct-current phenomena.
Knowing (or finding) this quantity, we can find practical-
ly any other quantities (work, power, amount of heat,
energy, parameters of the magnetic field, etc.) characteriz-
ing a particular phenomenon. Therefore, the basic prob-
lem of direct-current theory consists of finding the cur-
rents. Such a formulation of the problem is too general,
and we, therefore, break it down into more concrete and
harrow types.
{70 Part 2. Solution of Siandard Problems

(1) An electric circuit contains only one current source.

(2) An electric circuit contains several identical cur-
rent sources.
(3) An electric circuit contains several different cur-
rent sources.
Problems of the first type are solved by consistently
applying Ohm’s law for a closed circuit, Ohm’s law for a
uniform circuit element, and sometimes applying Kirch-
hoff’s first law. If the problem is formulated correctly,
the system of equations that arises from these laws is
closed and, hence, the problem can be considered physi-
cally solved.
Problems of the second type can easily be reduced to
those of the first if by using the rules of connecting iden-
tical current sources into batteries one finds the result-
ing emf of the circuit, €), and by applying the rules for
connecting resistances one finds the resulting internal
resistance of the battery, ro.
Problems of the first and second types are solved pri-
marily in secondary school and will not be considered
here.
Problems of the third type are the most general and
cannot be reduced to problems of the first and second
types. They are solved by applying laws that differ essen-
tially from Ohm’s laws for a uniform circuit element and
a closed circuit. The latter cannot be applied because in
the majority of such problems we cannot determine the
resulting emf &p.
Several methods exist for solving problems of the
third type. The most widespread is based on the applica-
tion of Kirchhoff’s laws. We will examine the essence
of this method by considering a concrete example.
Example 22.1. Determine the current flowing piroueh a
cell with an emf €,if 6) = 1V, @€, =2V,
ry = 19, r, = 0.59, r, = 1/3Q, R, = 1Q, and Rs == 1/39
(Figure 22.1).
Solution. The physical system consists of an electric
circuit containing several different current sources. It is
impossible to find the resulting emf and, therefore, we
cannot apply Ohm's law for a closed circuit. In this case
 Ch. 6. The Electric Field 174

the electric circuit can be calculated by applying Kirch-

hoff's laws.
First we must select (arbitrarily) the directions of cur-
rents in the branches. We select these directions as shown
in Figure 22.1. If a direction is chosen incorrectly, the
respective current will emerge as negative in the final

—_ ers

Figure 22.4

solution, and if it has been chosen correctly, the current

will emerge as positive.
Let us apply Kirchhoff’s first law. This law (or rule) is
valid for the nodes in a circuit. The given circuit has two
nodes, points A and C. For node A Kirchhoff’s first law
yields
+l, +/; = 0. (22.1)
For node C Kirchhoff’s first law gives nothing new-
Now let us apply Kirchhoff’s second law, which is va-
lid only for closed Joops. There are three loops in the
given circuit: ABCA, ACDA, and ABCDA. We consider
loop ABCA. It contains two emf’s (€, and €,), three
resistors (r,, r2, and Ay), and two currents (J, and J,). To
apply Kirchhoff’s second law we must select (arbitrarily)
a sense of traversal of a loop that is positive by conven-
tion. This is necessary to determine the signs of the emf’s
and currents. If the directions of the emf’s and currents
coincide with the sense of traversal of a particular loop,
the emf’s and currents are considered positive; and if the
directions do not coincide, they are considered negative.
172 Part 2. Solution of Standard Problems

We select the counterclockwise sense of traversal of

loop ABCA as positive. The emf €, is directed counter-
clockwise, so it is positive; the emf €, is directed clock-
wise (i.e. opposite to that of traversal of the loop), and
so it will enter into the equation expressing Kirchhoff’s
second law with a “minus”. Current J, passes through re-
sistors rr, and #2, and its direction coincides with the sense
of traversal of this loop. Current J, passes through resistor
rg and is directed opposite to the sense of traversal. Hence,
current J, is positive and current J, negative. By
Kirchhoff’s second law as applied to loop ABCA,
€,— @,=1,
(ry + Ry) —- Lore. (22.2)
lf we were to take the clockwise sense of traversal of the
same loop as positive, Kirchhoff’s second law would yield
~€;4+6€.= —Iy (ry 4+ Ry) + Tire,
which is simply Eq. (22.2) multiplied by —1. Obviously,
the two equations are equivalent to each other. Thus, the
essence of Kirchhoff’s second law does not depend on the
arbitrarily chosen sense of traversal of a loop.
Let us now turn to loop ACDA. The counterclockwise
senso of traversal of this loop will be chosen as positive.
Applying Kirchhoff’s second law, we obtain
€2— €3 == Jere — 13 (r3 + Rs). (22.3)

Equations (22.1)-(22.3) constitute a closed system.

Ifence, physically the problem can be considered solved.
Solving the system of equation, we find that

= -ZA, h=-FA, Laz.

We see that currents J, and J, are negative. This means
that by accident the directions of the currents J, and J,
were chosen wrongly. Current J, is positive, which means
that its direction was chosen correctly.
Example 22.2. A cylindrical air capacitor with an inner
radius R, and an outer radius R, has been charged to a po-
tential difference Ag, (Figure 22.2). The space between
the plates is filled with a low-conducting material with q
 Ch. 6. The Electric Field 173

resistivity p. Find the leakage current tf the height (or length)

of the capacitor is l.
Solution. The physical system consists of the section
of the electric circuit in which the reason for the direction-
al motion of the free charges
in the low-conducting medium
is the electrostatic field exist-
ing in the space between the
plates. We consider the po-
tential difference Ag, of this
field as constant. Since the
circuit element is uniform (i.e.
there are no emf’s in it), the
current can be found by ap-
plying Ohm’s law for a uni-
form circuit element,
I= Ag,/R, (22.4)
if we hnow the lotal resistance
R of this element. This
quantity, R, can he found
via the DI method. The ele- :
mentary resistance of a thin- Figure}22.2
walled cylindrical layer of
thickness (length) dr and radius r (see Figure 22.2) is
given by the formula
dr
dR=p
2mrl °
Integration yields the value of the total resistance of the
given element:
Re

ae
aa
sar sersp ne.
dr _ R
(22.5)
1

Hence
2nlAgy
o= (Re/®
Sin * (22.6)

Is solution (22.6) always valid? We can also pose a

number of other questions. We have said that after de-
termining the basic quantity, current, it is easy to deter-
144 bart 2. Solutton of Standard Problems

mine all other parameters of a circuit and the processes

that occur during the passage of current. For one, from
Joule’s law,
Q = FRt, (22.7)
we can find the amount of heat Q liberated by a circuit
element during time ¢; or the law
q=lt (22.8)
can be used to find the amount of electricity that has
passed through a circuit element during time ¢.
Let us find the time interval ¢, during which the charge
initially on the capacitor, gg = C Ago, will pass through
the circuit element, with
__ ___ 2me gel
C= (Re/RD 22-8)
the capacitor’s capacitance. The solution

= __ 97 =_ salle,
CAgop In (Ro/Ri) _
eee (22.10)

!s formal and essentially incorrect. Indeed, solution

(22.10) is valid only if the current is constant. From (22.4)
and (22.6) it follows that condition J = const is met only
if R =const and Ag, = const. But condition Ag, =
const means that at C = const the charge on the plates
of the capacitor must remain constant, which contradicts
the terms of the problem. The charge on the plates decreases,
which means that the potential difference Ag also
decreases. This implies that, strictly speaking, the leak-
age current does not remain constant, as it should accor-
ding to (22.6). Solution (22.6) is valid if Ag = const,
which, strictly speaking, is never the case.
Obviously, the way out of this situation is as follows:
we consider the potential difference Ap as changing so
slowly in the given conditions that these changes can be
ignored and it can be assumed that Ag = const. This
condition corresponds to the notion of a “low-conducting
medium” in Example 22.2. Now we make the problem
more complicated by assuming that in reality the poten-
tial difference Ag is not constant.
 Ch. 6. Phe Electric Field 475

Example 22.3. Using the terms of Example 22.2, find

the law of variation of the leakage current with time.
Solution. Suppose that the current changes so slowly
that at every fixed moment in time Ohm's law (22.4)
holds. Then
d A
~f=4 ‘ (22.41)

where { = —dg/dt and Ag = q/C are the instantaneous

values of current and potential difference. Thus, we have
arrived at the differential equation
_ T=
dg _ tr
9 (22.12)

for the unknown function g = gq (t), which is the charge g

on the plates of the capacitor at any moment ¢ in time.
Separating variables and integrating, we get
1
Ing= — Ro tte:

Allowing for the initial conditions (¢ = qa =C Ag, at

t = 0), we can find the constant: c, = Inq). Thus,

G— qge7 RE. (22.13)

We will call the constant t = RC the relaxation time, or

the relaxation constant. It is easy to see that t is the time
it takes the initial charge g, to diminish by a factor of
e ~ 2.78 .... In our case,

_ pln(R2/R3) 2meel
t= “Sal “In (Re/Ra) = E&P, (22.14)

which coincides with (22.10). Thus, ft, is not the time

that it takes the entire charge g, to flow from one plate
to the other, but only the relaxation time. From (22.13)
we see that it takes an infinitely long time for the charge
to flow from one plate to the other (q = 0 at t = oo).
From (22.13) we can also arrive at the laws of variation
of other quantities, the potential difference and the cur-
176 Part 2. Solution of Standard Problems

Ag = 4 = 2 e-M/RC— Ago HRC, (22.15)

I —= — 99
a _ pe
9% ORO
9-1/Rc APo 9-t/RC_
yt er J 9-t/RC
RO = Te" URC. (22.16)

Thus, formula (22.6) gives only the initial value of the

leakage current.
Now let us clarify the meaning of the concept of a low-
conducting medium. This is a medium in relation to
which we can ignore the variations in the instantaneous
values of current (22.16), potential difference (22.15),
charge (22.13), and other quantities. These variations can
be ignored if the relaxation time t is relatively large.
Hence, a low-conducting medium is one in which the re-
laxation time 7 is relatively large.
Let us estimate the relaxation times of some of the ma-
terials in our case. We start with paraffin (e = 2, p =
3 x 107® Q-m). Then (22.14) yields
T= e9ep, T+ 5.3 X 10° s = 6.1 days.
Let us assume, for the sake of convenience, that the dura
tion of observation is t, ~ 1s. Thus, paraffin can be con-
sidered a low-conducting medium with a very close ap-
proximation.
If we use formula (22.14) to estimate the relaxation
times of two types of quartz with e, = 4.4, p, =3 x
104 Q-m and e, = 4.7, p. = 1 X 10!2 Q-m and of mar-
ble with e, = 8.3, p3 = 1 X 108 Q-m, we obtain 3.25 hr,
42 s, and 7 x 10-° s, respectively. Ilence, while both
types of quartz can be considered low-conducting mate-
rials, marble cannot. Of course, if the characteristic du-
ration of observation is t) = 10-®s, we can even consider
marble a low-conducting medium.
 Ch. 7. The Magnetic Field 177

Chapter 7
TIE MAGNETIC FIELD

23. The Magnetic Field in a Vacuum

When studying magnetic fields, we must include in the
physical system the sources of magnetic fields and their
fields.
The basic problem of the theory of magnetic field (as
well as the theories of gravitation and electric field)
consists of calculating the characteristics of the
magnetic field of an arbitrary system of currents and
moving electric charges,
which is equivalent to determining the magnetic induc-
tion B at an arbitrary point
in the field. This problem is
solved by applying the Biot-
Savart law in differential
form
dp=-bol Xr, (23.1)
the superposition principle,
and the Df method (see Sec-
tion 6). A theorem often em- Figure 23.4
ployed in this connection is
concerned with the circulation of vector B,
§ B-dl=p, 5\/, (23.2)
especially in cases where (23.1) is invalid.
It is advisable first to solve several elementary problems
involving two widespread sources of magnetic field (the
results will be given without calculations): the magnetic
induction of a circular current J of radius R at a point A
on the axis at a distance z from the plane of the current
(Figure 23.1),
= folR®
B= Tape ee)
and the magnetic induction of a section of straight wire
carrying a current J at a point A lying at a distance ry
12—0408
178 Part 2, Solution of Standard Problems

from the wire (Figure 23.2),

B= Ee (cos @, — Cos &,). (23.4)

Then we can formulate and solve literally dozens of

problems involving the calculation of magnetic fields gen-
erated by various combinations of the above sources:
squares, triangles, rectangles, trapezoids, figures formed
by combining circles, half-lines, segments, etc. All] these
problems can be solved by the superposition method.
What is most essential is to allow for the vector nature
of the superposition principle.
Example 23.1. Find the magnitude of the magnetic
induction B of a magnetic field generated by a system of

| \%

Figure 23.2 Figure 23.3

thin conductors (Figure 23.3) along which a current I is

flowing at a point A {0, R, 0} that is the center of a circular
conductor of radius R.
Solution. The magnetic field is generated by three sour-
ces: the XO half-line (the positive half of the X axis), the
circular conductor of radius R centered at point A {0, R, 0}
and lying in the ZOY plane, and the half-line OZ (the
positive half of the Z axis). The current in all] three
conductors is the same, J. The magnetic induction B, of
 Ch. 7. The Magnetic Field 179

the magnetic field generated by the current flowing in

the XO conductor lies in the ZOY plane and points in
the direction opposite to that of the Z axis, the magnetic
induction B, of the magnetic field generated by the
current flowing in the circular conductor lies in the XOY
plane and points in the direction opposite to that of the X-
axis, and the magnetic induction Bg of the magnetic
field generated by the current flowing in the OZ conduc-
tor lies in the same XOY plane but points in the direction
opposite to that of B, (Figure 23.3). We use (23.4) to
find the magnitudes of B, and B,,
_ p — bol
By = By= aR
and (23.3) to find the magnitude of Ba,
Hol
B,= 32R °
According to the superposition pore
B=V B+ (B,— B,?=+TER! V2 xt —n+1).
Example 23.2. A current of density j flows along an
infinitely long solid cylindrical conductor of radius R.
Calculate the magnetic induction inside and outside the
conductor.
Solution. Since the conductor is not thin, the Biot-Sa-
vart law (23.1) and its corollary (23.4) cannot be employed.
Let us use the circulation theorem (23.2). We consider
a point A, lying at a distance r, from the conductor's axis
(Figure 23.4). We draw a circle of radius r, centered at
point O on the conductor's axis. In view of the symmetry
of the problem, the magnitude of B, is the same at each
point of the circle. The sum of the currents 2 J encompas-
sed by this loop (the circle) is equal to jnr}. Thus, by the
circulation theorem (23.2),
By, X 2nr,= pojmr'.
This yields the following formula for the magnitude of the
magnetic induction at point A,:
i. :
By= > Poll (23.5)
128
180 Part 2. Solution of Standard Problems

Let us take a point A, lying at a distance r, > A from

the conductor’s axis (see Figure 23.4). Applying the
circulation theorem, we get

By X 2mr, = pojnR2.
Hence, the magnitude of the magnetic induction outside
the field is given by the formula
Bec Bole ars
(23.6)
The diagram that demonstrates the behavior of the magnet-
ic induction of a solid cylindrical conductor is given
in Figure 23.5.

Figure 23.4 Figure 23.5

Example 23.3. A thin band of width | has been wound

into the shape of a pipe of radius R (Figure 23.6). A current
I flows along the pipe and is evenly distributed throughout
its width (the direction of the current isshown in Figure 23.6).
Determine the magnitude of the magnetic induction at an
arbitrary point on the pipe’s azis.
Solution. The conductor can be considered to be neither
thin nor a current element. Hence, we cannot employ
the Biot-Savart law (23.1) or its corollary (23.3). It is
also difficult to employ the circulation theorem (23.2)
since the magnetic field possesses no symmetry. Let us
employ the DI method.
We partition the pipe into rings so narrow thal each
can be considered to be a thin circular conductor. We
 Ch. 7. The Magnetic Field 181

consider one such narrow ring of width dz positioned at

a distance z from an arbitrary point A, (Figure 23.6).

Figure 23.6

The elementary current in this narrow ring,

d= (23.7)
generates, according to A,, a magnetic field whose ele-
mentary magnetic induction is
__ bof
RP dz
dB = SiR
aap
It is convenient to take the angle @ at which the radius of
each thin ring is seen from point A, as the integration
variable. Since
Rda 2 Bias A
z=Reota, dr= —- —,— » R+p= sint'a ’
sin*@

we have
_ Hof sine da
dB = => :

Integration yields
ae .
B= | Hof sinada i Hol
7 ay (COS @ — COs @). (23.8)
1]
182 Part 2. Solution of Standard Problems

If we introduce the concept of a current per unit Jength

of pipe,
io Th, (23.9)
then (23.8) assumes the form
B= Here (cos @, — CoS @»). (23.10)
This formula is also true for a solenoid if we allow for
the obvious formula J, = nJ,, where n is the number of
turns of wire per unit length of solenoid, and /, the current
in the solenoid. Hence, for a finite solenoid we have
B= Hott (cos a,— cos a). (23.11)
The above formulas (23.8), (23.10), and (23.11) also
hold true for a point A, positioned on the axis outside the
solenoid (see Figure 23.6).
Note that for point A, the
angle @, is always obtuse,
while for point A, it is always
acute (excluding the points
at the end faces). It is useful
to study various particular
cases; point A, is at the
middle of the pipe or at its
ends, the pipe is infinitely
long, or the solenoid is inf-
nitely long (lJ + oo).
Figure 23.7 Example 23.4. A current I
flows in a long straight con-
ductor whose cross section has the shape of a thin arc of length
l and radius R (Figure 23.7), Determine the magnetic induc-
tion x of the magnetic field induced !by the current at
point O.
Solution. 1t is easy to seo that the conductor can be
considered to be neither a thin straight conductor nor
a current element. Hence, we cannot employ the Biot-
Savart law (23.1)or its corollary (23.4). Since the magnet-
ic field possesses no symmetry, it is doubtful that the
circulation theorem (23.2) will yield any results. Let
us apply the D1 method.
 Ch. 7. The Magnetic Field 183

We partition the conductor into long, straight conduc-

tors so narrow that each can be thought of as a long,
thin, straight conductor. From (23.4) it follows that the
magnetic induction produced by the current in an infini-
tely long, thin, straight conductor can be calculated via
the formula
Mol ‘
ie z (23.12)

Let us consider one such conductor whose width is dl

(Figure 23.7). The elementary current in such a conduc-
tor is
dl
dl =a ’

and at point O it generates a maguetic field whose elemen-

tary magnetic induction is (see (23.12))
_ Modl —_ pol dl
(Bo, Oni
It is casy to see that the resullant vector B is directed
along the Y axis (i.e. B, = 0). Ihe projection of vector
dB on the Y axis is

For the integration variable we take angle a. Since

dil = Ada, we have
dB, = Hol cos ada
yo onl
Integration yields

B,== Tee
\ bolnol cosa
Mol cos
eeecd = pol
da
teI sin,
.. to
(23.13)
Q4¢

-Ao/2

where a) = J/R is the central angle of the arc J. Ifa, =

x, (23.13) yields
Hol
yo” nm?R *

Example 23.5. A current I flows along an infinitely long-

thin, straight band of width |. Calculate the magnetic induc-
184 Part 2. Solution of Standard Problems

tion of the magnetic field generated by this current at an

arbitrary point O (Figure 23.8).
Solution. With point O we link a coordinate system
whose axes are directed as
shown in Figure 23.8. To
calculate the magnetic field
we employ the DI method (as
in the two previous examples,
neither the Biot-Savart law
nor its corollary can be ap-
plied).
Let us partition the infinite-
ly long band into infinitely
long, narrow, straight sec-
tions each of which can be
thought of as an infinitely
long, thin, straight conductor.
Figure 23.8 Let us consider one such sec-
tion of width dl (see Figu-
re 23.8). The elementary current flowing in this section is

df ET,
and it generates at point O a magnetic field whose inducti-
on is given by the formula (see (23.12))
_ bod! _ pofdl
dB = 2nr ~ 2xir °

Suppose that point O is rp distant from the plane of the

band. Then
To rda@ ro da
=—, dl = = 7°
cos a cosas cos? a
Thus,
__ Hel da
dB = 2x1 cos @ °

Let us find the projections of vector dB on the X and Y

axes:
= a Hof sina da 2 ol dot
dB, = dB sina = : dB, = dBcosa=—S—.
 Ch. 7. The Magnetic Field 185

Integration yields
+@
B,= \ Mol sinada pol In £98.% (23.14)
Qnlcosa aml COS By”
-@

—_ "Pol Mol dae

da __ Mol
iol(S217 (cyte)
G1
B, a ( ~~ 2nl ont (23.45)
-@,

Introducing the concept of a current per unit width of

band, J, = — we find that

In the case where sae O is positioned symmetrically

(i.e. @ = @)) we have B, = 0 and B, = pola,/n. Eo
a band of infinite width (i.e. a plane) »we have B,
and B, = p,l,/2 (i.e. the field generated by a ois
with an evenly distributed current J, is homogeneous).
If the magnetic induction is known (or has been calcu-
lated by the described methods), the solution of most
problems is reduced to solving the respective problems of
mechanics (often by applying the DI method). The
most widespread problems are those related to the behav-
ior of a flat current-carrying loop in a magnetic field.
Often one has to calculate the forces and torques acting on
the loop, determine the work done in the process of
moving the loop in a magnetic field, etc.
Example 23.6. A thin conductor in the shape of a semicir-
cle of radius R carries a current I in the direction shown in
Figure 23.9. The conductor is placed in a homogeneous
magnetic field with a magnetic induction B = {0, Bg, 0}.
Determine the force acting on the conductor.
Solution. It would be a mistake to use Ampére's law
in the form F = JIB,, where 1= xR is the length of the
conductor, since each element of the conductor is posi-
tioned differently in relation to the magnetic field. Let us
apply the DI method.
e partition the conductor into sections so small that
each can be considered to be a current element. We
take one such section whose length is d/. The magnitude
186 Part 2. Solution of Siandard Problems

of the elementary force dF acting on this section is, by

Ampéere’s law,
dF = J dl Bysing. (23.16)
It is easy Lo see that all the elementary vectors dF, are
directed along the Z axis. Hence, vector summation is

Figure 23.9

reduced to arithmetic. Since dl = Rda, integration

of (23.16) with respect to angle @ yields
a

F= \IRB, sin ada =2IRB,.

0

Many variants of this problem can be considered: the

magnetic induction may be directed along the X axis or
along the Z axis or at various angles to the coordinate
axes. All these problems can be solved by the DI method.
Example 23.7. A square current-carrying loop made of
thin wire and having a mass m = 10 g can rotate without
friction with respect to the vertical axis OO, passing thro-
ugh the center of the loop at right angles to two opposite
sides of the loop (Figure 23.10). The loop is placed in a ho-
mogeneous magnetic field with an induction B = 10-! T
directed at right angles to the plane of the drawing. A current
I =2 A is flowing in the loop. Find the period of small
oscillations that the loop performs about its position of
stable equilibrium.
 Ch. 7. The Magnetic Field 187

Solution. The physical system consists of a known

(homogeneous) magnetic field, the current-carrying square
loop, and the free char-
ges moving in the material 0
of the loop (current J). The l im
physical phenomenon con- |
sists of the loop performing |
small oscillations under ! |
the forces exerted by the | QBs
magnetic field on each cur- |
rent element.
. : Since
: thei 10,
magnetic induction is
known, we can find these
forces and their resulting Figure 23.10
torque.
When the loop turns through a sinall angle @ away
from its position of equilibrium, a moment of Ampére
forces act on it:
M = p,B sin a, (23.17)
where
Pm = 1A = Ia? (23.18)
is the magnetic moment of the loop, and a the length of
the loop’s side. Applying the equation of motion (14.7),
we obtain
JB = M, (23.19)
where J is the moment of inertia of the loop about the OO,
axis, and B = @ the angular acceleration of the loop.
The loop’s moment of inertia is
g m a? 1 m 1
J=2 XPXp+eAxwBxz x @=-— me. (23.20)

Substituting into Eq. (23.19) the value of the moment of

Ampére forces, (23.17), and the value of the moment of
inertia of the loop, (23.20), we obtain the following
equation:

a+ 58 sina =0.
188 Part 2. Solution of Standard Problems

Bearing in mind that sin @ ~ @ for small a's, we obtain

a differential equation for the harmonic oscillations of
the loop:
a+ o=0. (23.21)
If we compare this equation with the general equation
for harmonic oscillations, we arrive at the following for-
mula for the angular frequency of the loop’s oscillations,
@) = V 6IB/m,
and the period of oscillations,
T,—2n V mIGIB, T, 0.57 s.
As is known, when a flat current-carrying loop is moved
in a magnetic field, the following amount of work is
performed:
W = IAQ, (23.22)
where A® is the variation of the magnetic flux passing
through the area limited by the loop. If a point magnetic
dipole (a flat loop carrying a current J and having suffi-
ciently smal] dimensions) whose magnetic-moment vector is
Pm =JAn (23.23)
is parallel to the magnetic induction B, calculating the
work amounts to calculating the magnetic field:

W=1AD=1 (®, — ,) = (B,— B,) A= pm (B, — B,).

(23.24)
Example 23.8. Assuming that the terms of Example 23.3
remain valid, we place a point magnetic dipole with magnetic
moment pm at point A, in the middle of the pipe’s axis
(Figure 23.11). The dipole is then moved from point A,
to point A, along the axis in such a manner that vector pm
remains parallel to vector B. Find the work done in moving
the dipole.
Solution. Equation (23.24) shows that to solve the prob-
lem it is sufficient to calculate the induction B, of the
magnetic field at point A, and induction B, at point Ag.
 Ch. 7. The Magnetic Field 189

According to (23.8), we have

es ry B,= =.
2 Y R+12/4 2 2 VRP

Substituting these values into (23.24), we find that

Wi = BebeI [ 1 1
2 ( FRR Tae)

If the magnetic dipole that is moved in a magnetic

field is not point-like but an ordinary flat loop carrying

Figure 23.41

a current J, the DI method is often used to calculate

the magnetic flux.
Example 23.9. An infinitely long, straight wire carrying
a current I, = 5 A lies in the plane of a rectangular loop
carrying a current I, = 3 A. The two are positioned in
such a manner that one side of the rectangular loop whose
length is | = 1 m is parallel to the straight wire and isr =
0.4 b distant from it, with b the length of the other side
of the rectangular loop (Figure 23.12). Determine the work
that must be done so that the loop isturned through an angle
a = 90° about the OO, azis, which is parallel to the straight
wire and passes through the middle of the opposite sides of
the loop whose length is b.
Solution. It is easy to see that when the loop is in its
final position, the magnetic flux passing through it is zero:
@®, = 0. Thus, we need only to calculate the magnetic flux
®, passing through the loop in the initial position. Since
the field generated by a current J, flowing along an
190 Part 2. Solution of Standard Problems

infinitely long, straight wire,

I
B,= Hot ‘
(23.25)
(see (23.4)), is nonhomogeneous, the solution 0, = B,A,
with A = lb the area of the loop, is incorrect. We, there-
fore, apply the DI method.
We partition the area encompassed by the loop into
strips so narrow that within each strip the magnetic

Figure 23.12

field can be assumed to be homogeneous. We take one

such strip of width dz (see Figure 23.12) that is z distant
from the straight wire carrying current J,. The elementary
magnetic flux passing through this strip is
d® = BAA = #4 Jaz. (23.26)
After integrating with respect to z, we find the sought
magnetic flux:
O.1b+4b

O,= | Hebb ge = Hot in tt.

0.18
Thus,
TT gl
W = 1,A0=
1,0, = * In 11.
Substitution of numerical values yields W ~7.1 x 10-°J.
 Ch. 7. The Magnetic Field 191

24. The Magnetic Field in Matter

When considering the maguetic field in a magnetic sub-
stance, we introduce two physical quantities in addition
to the magnetic induction B: the magnetization J (the
magnetic moment per unil volume of the substance) and
the magnetic field strength
B
=——
a J J. 4.
(24.1)

For a homogeneous and isotropic magnetic substance,

B= pop. (24.2)
The relative permeability of a ferromagnetic substance,
pt, is a nonlinear function of BT
the magnetic field strength H. +80
Hence, when solving prob-
lems involving ferromag-
netic substances one often i,
uses empirical B vs. H
curves. Figure 24.1 shows
such curves for iron, steel, 9.
and pig iron.
Finding the magnetic
induction vector B consti-
tutes the basic problem of 0 i 2H, kA/m
the theory of magnetic sub- Figure 24.4
stances. A common proce-
dure is to employ the magnetic circulation theorem,

$H-dl= 1, (24.3)
the B vs. H curves of the type shown in Figure 24.1, and
the fact that at the boundary between two different mag-
netic substances the normal component of vector B
varices continuously:
B,n = Ben. (24.4)
Example 24.1. A closed toroid with an iron core has
N = 400 turns of thin wire in a single layer. The mean
diameter of the toroid isd = 25 cm. Determine the magnetic
192 Part 2. Solution of Standard Problems

field strength and the magnetic induction inside the toroid,

the permeability p of the iron, and the magnetization J for
two values of the current flowing in the wire: 1, = 0.5 A
and J, = 5 A.
Solution. Applying the theorem on the circulation
of H (24.3) along the circumference of diameter d (the

Figure 24.2 Figure 24.3

median line of the toroid; Figure 24.2),

H x nd =IN, ,
we find the magnetic field strength inside the toroid:
H=IN/ad.
Substituting the values of the current, we get
H,=255 A/m, H,= 2550 A/m.
Next, using the curve for iron in Figure 24.1, we deter-
mine the respective values of magnetic induction:
B, =0.9 T, B, =1.45 T.
We can now use Eq. (24.2) to find the values of the relative
permeability » = B/y,H of the iron core:
wy, = 2.8 x 1085, pp. 4.5 x 107.
Finally, using Eq. (24.1), we can find the magnetization
J = Bly, — H:
J, 274 x 10° A/m, J, ~ 1.4 x 10° A/m.
 Ch. 7. The Magnetic Field 19

Analysis of the above data shows that only the magnetic

field strength inside the ferromagnetic substance (iron) is
directly proportional to the current flowing in the wire,
while the other quantities, namely, induction B, perme-
ability p, and magnetization J, are nonlinear functions
of H and, hence, of the current J.
Example 24.2. The winding of a toroid with an tron core
containing a vacuum gap that is 1, = 3 mm wide has n =
1000 turns per meter. The mean diameter of the toroid is
= 30cm. What must the current I flowing in the winding
be if the magnetic induction B, in the gap is to be 1 T
(Figure 24.3)?
Solution. Employing the theorem on the circulation of
vector H (24.3), we find that
Hnd + Hyl, = Innd, (24.5)
where H is the magnetic field strength in the core, and H,
the magnetic field strength in the gap. Since the relative
permeability p of a vaccuum is unity, we can use (24.2)
and find the magnetic field strength in the gap:
Hy ==,
B
Hy= 7a
10?
Alm. (24.6)
Because the vacuum gap is narrow, we assume that the
radial components of the magnetic induction in gap and in
core are zero. Then, allowing for (24.4), we conclude
that induction B in the core is equal in absolute value
to By. Using the appropriate curve in Figure 24.1, we can
find the magnetic field strength in the core: H =7 xX
10? A/m. Thus, Eq. (24.5) yields
TatH +h,
Bol
Te BRA. (24.7)
Example 24.3. Let us change the terms of Example 24.2.
Suppose that the current flowing in the winding of the toroid
is [ = 3.2 A. Find the magnetic induction Bg, in the gap.
All other eonditions remain unaltered.
Solution. At first glance we seem to have formulated
a problem that is the reverse of the Problem 24.2 and the
solution can be obtained by employing formulas (24.5)
13—0498
194 Part 2. Solution of Standard Problems

and (24.6). This system of equations contains three un-

knowns, H, Hy, and By, and although these are connected
through the appropriate curve in Figure 24.1, we cannot
employ the curve directly. Nevertheless, the problem

Figure 24.4

is solved via a diagram. Since B = Bo, from (24.7) we can

find the relationship between B, H, and 7:
B= tiene_ waht 7, (24.8)
0 0

For a given J, Eq. (24.8) specifies a linear B vs. H

dependence for different cores. For a given core (made of
steel), the values of B and H in (24.8) must satisfy the
relationship specified by the appropriate curve in Fig-
ure 24.4. Hence, the sought values of B and H are the
parameters of the point M = {B, H} where the curve in
Figure 24.1 corresponding to steel intersects the straight
line reflecting Eq. (24.8) (Figure 24.4). In Figure 24.4 the
straight line (24.8) intersects the coordinate axes at
points A = {0, B,} and C = {H,, 0}, with
By = | Bye 1.26 T,
0

and
A, = Ih, A, = 3.2 x 408 A/m.

The reader can easily see that the coordinates of point M

are B = B, = 1T and H = 700 A/m, which corresponds
to the data of Example 24.2.
 Ch. 8. The Electromagnetic Field 195

Chapter 8
THE ELECTROMAGNETIC FIELD

25. Electromagnetic Induction and Self-Induction

The basic law governing the electromagnetic induction
phenomenon is Faraday’s law
d@® :

Hence,
the problem of finding the induction emf &, consti-
tutes the basic problem in the electromagnetic in-
duction theory.
When performing the stage of physical analysis one
must thoroughly investigate the causes of the changes in
the magnetic flux @ and how this quantity actually varies.
Then one must determine the magnetic flux passing
through the surface encompassed by the loop as a function
of time ¢, that is, D = @ (t). Now Faraday’s law (25.1)
can be used to find the induction emf.
Example 25.1. A flat square loop with a side a = 20 cm
is placed in a magnetic field whose induction B = (a +
t?)i, where a =10-'! T, B = 10-? T/s*, and i is the
unit vector pointing along the X azis; the plane of the loop is
at right angles to B. Find the induction emf generated in
the loop at time t = 5s.
Solution. The physical system consists of the magnetic
field varying with time, the loop placed in this field,
and the induction current generated by this field in the
loop. We want to know the induction emf. The reason
why the magnetic flux passing through the loop varies is
the variation with time of the magnetic induction vector.
Let ns find the magnetic flux.
Since the magnetic field is homogeneous and the loop
is flat, we have
® = B-A = (a4 + ft?) a’. (25.2)
This gives the following formula for the induction emf:
81! =|— SF |= 26%, 1:1=4x 10°V. (25.3)
ise
 196 Part 2. Solution of Standard Problems

As we have repeatedly done in the past, we can now sim-

plify the problem or make it more complicated, thus
formulating dozens of new similar problems. But what
does “similar” mean in this context? Here we consider
ways of constructing clusters of problems and introduce
the concept of the generalized problem of a cluster. The
word “similar” means that in all problems of a specific
type the underlying phenomenon is the same (in our case
it is electromagnetic induction), the magnetic field is
homogeneous and varies in time, and the loop is flat and
is positioned at right angles to B. To solve all such prob-
lems (the cluster) we can employ Eqs. (25.2) and (25.3).
We can now formulate the generalized problem of the
first “cluster”: a flat loop encompassing an area A is placed
in a magnetic field whose induction varies according to the
law B = f (t) i, where f (t) is an arbitrary (differentiable)
function of time t, in such a manner that the plane of the loop
is at right angles to B; we wish to determine the induction
emf in the loop at an arbitrary moment in time.
The generalized problem of the first cluster can be
solved by the same equations (25. 2) and (25.3) but in
generalized form:
@=B-A=f (t)i-A, (25.4)
il=|—-Sr l=" OA (25.5)
Now the solution of any specific problem belonging to
this cluster can be solved directly by employing Eqs. (25.4)
and (25.5). Thus, every specific problem belonging to a clus-
ter becomes elementary after we have formulated and
solved the appropriate generalized problem.
Consider, for instance, the following specific problem:
a flat loop in the form of an equilateral triangle with a side
of length a is placed in a magnetic field whose induction
varies according to the law B = B, sin wt xX i, where By
and w are constants, at right angles to B; find the induction
emf in the loop at time t.
This elementary problem can easily be solved via
Eqs. (25.4) and (25.5):
® = B,sinwt x A, | 8, | = B,Aw cos ot,
where A = a?// 3/2 is the area subtended by the loop.
 Ch. 8. The Electromagnetic Field 197

We are now ready to formulate and solve problems be-

longing to the second cluster, which includes considering
other physical phenomena related to phenomena discussed
in problems of the first cluster. Let us consider, for
example, the various phenomena related to the induction
current in Example 25.1 (thermal, magnetic, etc.).
Suppose that we are required to determine the amount of
heat liberated in the loop during the first five seconds if the
loop resistance R is 0.5 Q.
Ignoring the inductance and capacitance of the loop
and allowing for Ohm’s law, we find the induction current
flowing in the loop:
I=6(/R = 2Bat/R. (25.6)
Since this quantity varies in time, we must apply the DI
method to find the sought amount of heat:
; : AB%ate? dpes 4Btate> 15
Q={ rrae= | R aR (25.7)
0 0

ore of numerical values yields Q + 5.3 x

~~ J.
It is now easy to formulate and solve the following gen-
eral problem of the second cluster: a flat contour encompas-
sing an area A and having an ohmic resistance R is placed in
a magnetic field whose induction vector varies according to
the law B = f (t) i, with f (t) an arbitrary (differentiable)
function of time t, in such a manner that the plane of the loop
is at right angles to B; find the amount of heat liberated by
the loop during an arbitrary time interval t.
Allowing for (25.4)-(25.7), we arrive at the solution to
the generalized problem of the second cluster:
t

Q=F |i wrde.
At (og
(25.8)
0
Now any specific problem of the second cluster becomes
elementary and can be solved by applying formula (25.8).
We are now ready to formulate and solve specific prob-
lems and the generalized problem of the third “cluster”
498 Part 2. Solution of Standard Problems

(and the fourth, fifth, and so on) by changing still further

the conditions in which the physical phenomena of the
first cluster proceed. Suppose that in Ezample 25.1 the
magnetic field is no more homogeneous. Then Eq. (25.4)
becomes invalid and we must employ the DI method to
calculate the magnetic flux.
Example 25.2. An infinitely long, straight conductor lies
in the plane of a square loop with an ohmic resistance
R=7Q and a side length
a = 20 cm ata distance ry =
20cm parallel to one of the
loop’s sides (Figure 25.1). The
current flowing in the conduc-
tor varies according to the law
IT=at’, with a =2 A/s3,
Find the current in the loop
at time t = 10s.
Solution. Owing to the
change in the current flowing”
in the straight conductor, the
magnetic flux that passes
through the loop varies and
Figure 25.1 an induction current is gener-
ated in the loop. The loop,
therefore, finds itself in a nonhomogeneous magnetic
field. Hence, we are forced to use the DI method to
calculate the magnetic flux (see Example 23.9).
Let us partition the area subtended by the loop into
strips so narrow that within each strip the magnetic
field can be thought of as homogeneous (see Figure 25.1).
The elementary magnetic flux passing through the narrow
strip is
23 __Wo Ja dz
d® = Badr=— >.

Integrating this equation with respect to z from ro to

ro + a, we find that
Tota
o= ( Hofadr p,aa in (1 4-a/ro) 2
= e aaxr 2n 7
Toa
 Ch. 8. The Electromagnetic Field 199

Using Faraday’s law (25.1), we can find the induction emf,

_ 3ppeaIn (1-Fa/ro) 49
a= re e,
and the current,

_ 81 __3ptgae
In (1 + @/r9) ae =
l= =“ aR
onR 42, [~%2.4x 108A.

We could, of course, formulate and solve the generalized

problem of the third cluster. But let us consider a specific
problem belonging to the fourth.
Example 25.3. The loop of Example 25.2 is moving away
from the infinitely long, straight wire with a velocity v =
100 m/s in the direction perpendicular to the conductor.
A direct current | = 10 A ts flowing in the conductor. Find
the induction emf generated in the loop ten minutes after
the loop started its motion if initially the loop was at a dis-
tance ry = 20cm from the conductor.
Solution. The current in the conductor remains constant,
which means that the magnetic field generated by this
current is time independent. However, the magnetic
flux passing through the loop does not remain constant
because the position of the loop in relation to the conduc-
tor changes. Let us find the flux as a function of time ¢.
Applying the DI method, we get

>=!9 In (1++), (25.9)

with z = vt -+ rg the separation between conductor and
loop at time ft. If we now employ Faraday’s law, in other
words, find the time derivative of ®, we have the induc-
tion emf generated in the loop:
f= Pola?v
2n (a+ vt-+ry) (vt-+r9)°

Carrying out the necessary calculations (it is important

to note that vé>>r,, and vt>>a for t > 410-7! s, which
means that we can ignore r, and @ inside the parentheses),
we obtain
£,=8 x 10-13 V.
200 Part 2. Solution of Standard Problems

The numerical value of the induction emf is negligible

because the loop’s velocity is so high that, first, at time
t = 10s the loop is at a distance z = 1 km from the con-
ductor and in this region the magnetic field is weak, and,
second, the variation of the magnetic flux passing through
the loop is low, too. Now let us change somewhat the terms
of Example 25.3.
Example 25.4. Suppose that in Example 25.3 only the
side of the loop farthest from the conductor rather than the
entire loop is moving away
from the conductor with a veloc-
ity v (Figure 25.2). The length
of the mobile side is a, the resis-
tance of the loop is considered
known, and the resistance of
the leads and mobile side is
assumed to be zero. Find the
current generated in the loop e

at an arbitrary moment of
time t.
Figure 25.2 Solution. We denote the
current in the infinitely long,
straight conductor by J/;, which is constant (by hy-
pothesis). The variation in the magnetic flux passing
through the loop is caused by the motion of the moving
side of the loop. Applying the DI method, we find the
magnetic flux passing through the loop:
vt

= [ HoI dz— HhI n(2t), (25.10)

which yields the following formulas for the induction emf
and current:
— Hottie — Hope
a= ont’ is 2nRt °
We can now make the problem more complicated by
assuming, for example, that the current in the conductor
varies with time, J, = f (t). Then, according to (25.10),
__ Moftaf (4) v
o= in(> t)
 Ch. 8. The Electromagnetic Field 201

and, hence,

— Holes! (t) v Hoptaf (t)

g&=— on In(t)+ “Int :
_ Moltaf’ (4), fv Holtaf (t)
P= oR MAG, +e
Example 25.5. Along two smooth copper buses positioned
at an angle a to the horizontal line there slides, owing to
the force of gravity, a copper bar of mass m (Figure 25.8).
The two upper ends of
the buses are connected by
a capacitor of capacitan-
ce C. The distance be-
tween the buses is l. The
entire system is placed in
a homogeneous magnetic
field, with induction B at
right angles to the plane
in which the bar moves.
The resistance of the buses,
bar, and sliding con-
tacts and the self-induc- Figure 25.3
tance of the loop are
assumed negligible. Find the acceleration of the bar.
Solution. As in the previous example, the variation of
the magnetic flux passing through the loop is caused by
the movement of the bar. According to Ohm's law for
a nonuniform section of a circuit, the induction emf §, at
any moment of time is equal to the potential difference
Ag across the capacitor’s plates:
&; = Ag.

But Ag = Q/C. Hence, the induction current in the

loop is given by the following formula:
_ 4Q _ pd (Ag)_ dg;
dt °

Since the magnetic field is homogeneous, we can write

dA dz
6, =B 7 = Bl7- = Bly,
202 Part 2. Solution of Standard Problems

where A is the area enclosed by the loop. Thus,

dv
I = CBI =CBla,
with @ the sought acceleration of the bar.
Two forces act on the bar, the force of gravity mg and
Ampere’s force JIB = CB*l’a. By Newton’s second law,
ma = mg sina — CB?la.
Uence,
__ mgsina
= n+ CB
If, in addition, there is friction that acts on the bar, we
can easily show that
mg sina — fmg cosa
a=
m+ CB?13 *
where f is the coefficient of friction.
Let us now assume that there is no external magnetic
field but the current J in the loop varies with time ¢.
Then the magnetic flux generated by this current and
passing through the loop,
® = LI, (25.11)
varies and there appears a self-induction emf
Sa=—L. (25.12)
The self-induction emf generates a self-induction current.
When an electric circuit is broken or closed there appears
a break induced current
T= I,e-(R/b)t (25.13)
or a make induced current
T=], (1—e-(R/D)), (25.14)

respectively, with Jy = €,/R the steady-state value of

the current in the circuit and 6, the emf of the power
source.
Example 25.6. A solenoid with inductance L = 10-1 H
and resistance R = 2 x 10-2 Q is connected to a source of
emf €, = 2 V whose internal resistance 1s negligible.
 Ch. 8. The Electromagnetic Field 203

What amount of electricity will pass through the solenoid

during the first five seconds after closure of the circuit?

Solution. When the circuit is closed, there appears

a (variable) make induced current (25.14). To find the
amount of electricity passing through the solenoid we
employ the DI method.
Let us partition the time interval ¢ into segments dt so
small that within each the current may be considered to
be approximately constant. Then the elementary amount
of electricity dQ that passes through the solenoid during
dt is given by the following formula:
dQ=Idt= 5o (1-- e- (B/D) dt.

Integration with respect to ¢ yields

5
Q= \3 6 (1-e 2 ILM) du = 652(44 —Z
Br:
0 rt) 5
’
0
0
Qx181C. (25.15)
If we had assumed, erroueously, that the current instan-
taneously reaches its steady-state value J, = 6,/R (which
is, indeed, possible if L is small), we would have found
that Q = I,t = (6,/R)t, @=500 C, which differs
considerably from the correct answer given by (25.15).
The answer Q = 500 C would be correct if the terms of
the problem allowed us to neglect self-induction. Direct
calculation shows that at L = 10-* If we would have
Q ~ 495 C. Thus, at L = 107° H self-induction can be
ignored in the present problem.

26. Electromagnetic Oscillations

When studying electromagnetic oscillations, the common
approach is to include in the physical system the electro-
magnetic field and the objects (which are of secondary
importance) in which this field is localized (conductors,
induction coils, capacitors, and the like).
The basic problem of the theory of electromagnetic
oscillations is to find the law of variation in time
204 Part 2. Solution of Standard Problems

of an electric or magnetic physical quantity charac-

terizing the process.
Using the equations that link this quantity with other
quantities, we can find the values of these quantities, too.
Example 26.1. Find the magnetic induction that exists
within an ideal LC circuit at time t = 10-4 n/6 8 if at
t = 0 the charge on the capacitor was Q, = 107° C and the
current I, was zero. The inductance of the coil is L = 10-5 H,
the number of turns per unit of coil length isn = 10° m-},
the capacitance of the capacitor is C = 10-5 F, and the
contour ts placed in a vacuum.
Solution. The physical system consists of the conduc-
tors forming the induction coil, the capacitor, and the
electromagnetic field varying with time. We are seeking
one of the parameters of this field (the magnetic induction)
at a certain moment in time. This constitutes a basic”
problem of the theory of electromagnetic oscillations.
Let us find the law of variation of anelectric or magnet-
ic quantity. The physical process consists of free undam-
ped (or natural) electromagnetic oscillations occurring in
the circuit. As is known, the differential equation of
such osciliations has the form

0+ 0i0=0, (26.4)
whose solution is given by the equation of harmonic
oscillations
Q = Qy sin (Wot + Ap). (26.2)
Note that equations similar to (26.1) and (26.2) can be
written for other quantities (current, voltage, etc.). There
are three unknown parameters in Eq. (26.2): the angular
frequency @o, the amplitude Q), and the initial phase a.
The angular frequency can be found from the equation
w, = 1/LC, (26.3)
while the amplitude Q, and the initial phase a, can be
found from the initial conditions (Q = Q, and J, =
— dQ/dt = 0 at t = 0),
Q, = Qo sin a, O = — Qos Cos a.
 Ch. 8. The Electromagnetic Field 205

Whence, a, = x/2 and Q, = Q,. Thus, the equation of

harmonic electromagnetic oscillations in the circuit has
the form
Q=Q sin (Gattt>). (26.4)
We have, therefore, found the law of time variation of
an electric or magnetic quantity (in our case, electric
charge Q).
We can now calculate the value of the current flowing
in the circuit at an arbitrary time ¢,
I=-f=
= dQ
TE
Q1
° os (Tatty ), I1=~5x
re
107A,
=

and the magnetic induction,

B=p,prl = — ete os(—— t+ 4),

Bx 6.3 x 105 T.
If we now use equations that link these quantities with
other quantities, we can find any physical quantity
characterizing the phenomenon. For instance, the poten-
tial difference between the plates of the capacitor is
do = sin (Te t+ +)»

the electric field strength in the capacitor (assuming it is

a plane-parallel capacitor with a plate area A) is
Oa Oi Oe
Te, gd GA sin(Tae tty ).
the electric-field energy density inside the capacitor is
egeE? __ eG}. n
wo=
2 2ABe, put (=
Vic t+a)

the magnetic-field energy density inside the coil is

OF aBP SLe
= Qin®
Hap cos? ( "a i++),
273

etc. It is clear that if we abstract ourselves from the con-

crete numerical values in this example, we practically
206 Part 2. Solution of Standard Problems

have a ready solution to the generalized problem on free

undamped electromagnetic oscillations in an LC circuit.
Example 26.2. The ohmic resistance of an LC circuit is
R = 10" Q, the inductance L = 10-* H, and the capacitance
C =10-°F. Find the value of the current in the circuit
at time t = 5-10-® s if at t = 0 the charge on the capacitor
was Qy, = 10-® C and the current was zero.
Solution. Electromagnetic oscillations occur in the LC
circuit. To solve the basic problem of the theory of elec-
tromagnetic oscillations, we must find all the parameters
(w, 5, Qo, and ap) of the equation of damped oscilla-
tions:
Q= Q,e-* sin (wt + a%). (26.5)
The damping factor 6 and the angular frequency w can
be found from the terms of the problem:
6= a » oF IC &.

This yields 6 = 5 x 10° rad/s and w = 8.7 X 10° rad/s.

The initial phase a) and amplitude Q, can be found from
the initial conditions. Allowing for the fact that att = 0
the charge on the capacitor Q is Qo,, we get the first equa-
tion for finding a, and Q,:
Qoit = Qo Sin a. (26.6)
The fact that at ¢ = 0 the current
_ __ 4Q
~~ dt
= —Q,[— de-* sin (wt + a) + we-5! cos (wt + a>)| (26.7)
is zero provides the second equation:
— Ssin a -+ o cosa, = 0. (26.8)
Solving the system of equations (26.6) and (26.8), we find
a) and Q,
Gy =tan-!(w/$), a 21/3; Qy=2Q,/V 3.
Thus, we have the law of variation of charge Q with time
(see Eq. (26.5)) in complete form:
Q= 2Q01
200.
ve 4-(n/2L)t sin (V ole)
o- )ot+) ‘
 Ch. 8. The Electromagnetic Field 207

We are now ready to determine any physical quantity

that characterizes this specific physical phenomenon
(damped electromagnetic oscillations). The sought value of
the current can be found by solving Eq. (26.7):
IT =Q, [5 sin (wt + &) —w cos (wt + a&)] e- 5,
14.6 x 10° A.
As with problems on free electromagnetic oscillations,
when dealing with problems on steady forced oscillations
we must first determine the law of variation of an electric

A ~ D

yo € \= R t
a

(¢) I Cc

Figure 26.1 Figure 26.2

or magnetic quantity and then, using relationships that

link different physical quantities, find the laws of vari-
ation of other sought quantities.
Often the phasor description is employed to solve
problems on forced electromagnetic oscillations. In this
approach a harmonic oscillation Ag = Agg sin (Q¢ | a)
is represented by a vector Ag called a phasor; the length
of the vector is equal to Ag, and the angle formed by the
vector with a certain horizontal axis (the axis of currents J
or the axis of voltages Ag) is equal initially to the initial
phase a (Figure 26.1). Phasor A@ rotates counterclock-
wise with an angular velocity Q. Let us examine several
examples illustrating the use of the phasor method.
Example 26.3. An electric circuit consists of a source of
emf varying harmonically, an ohmic resistance R, a capaci-
tance C, and an inductance L, all connected in series
(Figure 26.2). Find the law of voltage variation on section
ARCLD as a function of time t.
Solution. Let us employ the phasor method (Figure 26.3).
Suppose that the law of variation of current is given in
208 Part 2. Solution of Standard Problems

the form
I = ZT, sin Qt, (26.9)
where Q is the angular frequency (or rate) of variation of
the external emf. We direct the current axis horizontally.
Then the voltage variation
on resistance R is depicted
by a vector A@or directed
along the current axis, the
voltage variation on induc-
tance L by a vector Aqozr
directed at right angles to
the current axis, and the
voltage variation on capac-
itance C by a vector A@oc
also directed at right angles
Figure 26.3 to the current axis but in
the direction opposite to
that of A@o,. The length of each vector is, respectively,
Agen = J oR, Ago, =MQh, Aqgc = M/C.
The resultant voltage is depicted by the vector Aq, =
A@or -+ AGor + A@oc. The sum of voltages on in-
ductance and capacitance,
Ager I, (QL—gz),
is known as the reactive component of the voltage. Thus, the
net voltage varies according to the law
Ag = Aggy sin (Q¢ + a), (26.10)
where the amplitude ©
Ay, =I, V w+(a1-2)
oL—)’ (26.44)
and the initial phase
QL —1/AC )
a= tan! ( (26.42)
R

are found from the vector triangle OAB (see Figure 26.3).
Let us analyze Eq. (26.11). Only the amplitudes of the
voltage and current, Ag, and J, enter into this equation
and not the instantaneous values Ag and J. Equa-
 Ch. 8. The Electromagnetic Field 209

tion (26.11) shows that J, depends on the frequency Q of the

external emf. As Q grows from zero to the value
Qies = = 1/V LC, (26.13)
the amplitude of current, J), increases, since the impedance

Za r+ (QL—1)
(at—a.)" (26.14)
of the circuit drops. At a value Q = Q,,, the amplitude of
the current reaches its maximum value, the reactive com-
ponent of the voltage vanishes, and the circuit behaves
like a purely resistive circuit. This phenomenon is known
as the resonance of voltages. From Eq. (26.12) it follows
that in the event of a resonance of voltages the phase
difference @ between current oscillations and voltage
oscillations vanishes. If Q is further increased (Q > @,),
the amplitude of the current, J), decreases, asymptotical-
ly tending to zero.
Example 26.4. A resistance R = 10 Q and an inductance
L=0.1 Hf are connected in series. What capacitance
should be inserted in series into the circuit so that the phase
shift between the emf and the current decreases by Aa = 27°?
The driving frequency of the external emf is v = 50 IIz.
Solution. Let usemploy the phasor approach. Figure 26.4
shows that
tan a, =
IQL _ QL
7
This yields a, = tan“! (QL/R), or a, + 72°. Hence
GQ, = a, — Aa, or a, = 45°. By formula (26.12),
taney
=SEC
This gives us the value of the sought capacitance (bearing
in mind that Q = 2zv):
4
C 1.5 x 10? pF.
C= ie (2nvL—R) °
Example 26.5. A section of a circuit consists of a capaci-
tance C = 200 pF and a resistance R = 10* Q connected in
parallel. Find the impedance of the section if the driving
frequency of the harmonic emf is v = 50 Hz.
14-0498
210 Part 2. Solution of Standard Problems

Solution. In the phasor method as applied to calcula-

tions of parallel circuits, the horizontal axis is the voltage
axis (Figure 26.5). Then the current in the ohmic resis-
tance, Jy, coincides in phase with the voltage, and the

4Pou= To Sel

Tog=4H * $C

Figure 26.4 Figure 26.5

current flowing through the capacitance leads the voltage

in phase by an angle ac = 90°. The net-current ampli-
tude J, can be found from triangle OAB:

rm AQ \?_ Ap
I, _ V (dq2C)-+ (+) ——R [i envCR) ‘

This yields the impedance of the section:

R
= \/7 LIP Z ~ 15.6 Q.
V 1+ (2xvCR)
 Ch. 9. Electromagnetic Waves 244

Chapter 9
ELECTROMAGNETIC WAVES

27. Interference of Light

The basic problem in studying the interference of
light is to calculate the interference pattern.
‘This means finding the distiibution of intensity J of
elecliomagnetic waves in space. Since intensity is pro-
portional to the square of the amplitude £, of the electric
field strength in the clectromaguetic wave, the basic
problem of the theory of interference is reduced to finding
the amplitude Fy of the resultant oscillation at an arbi-
trary point in the medium.
Most often, when calculating an interference pattern, it
is necessary to determine the position of an arbitrary kth
order maximum (or minimum) and the separation between
two adjacent maxima (or minima). The method of solv-
ing the majority of problems on light interference can
be reduced to two basic stages: finding the optical path
difference 6 and employing the maximum condition
6 =k, (27.1)
or the minimum condition

§=(2k41) 2, (27.2)
Example 27.1. Calculate the interference pattern pro-
duced by two coherent sources1 and 1) (Figure 27.1) posi-
tioned d == 5 mm apart at a distance L = 6 m from the
screen. The wavelength of the light generated by the sources
in a vacuum is 4, = 5 X 10-7 m. Also find the position
of the fifth maximum on the screen and the distance between
the adjacent maxima. The medium ts a vacuum.
Solution. Before meeting at an arbitrary point F on the
screcn (see Figure 27.1), at which point the result of
interference is evaluated, each of the waves travels its own
geometrical path, x, and z,. For the sake of simplicity
we will assume that the initial phases are equal to zero
and the amplitudes are equal to each other. Then we can
14*
212 Part 2. Solution of Standard Problems

write the equations for the waves generated by the

sources as follows:

E, = Ey, sin (2ave— a ) :

E,= E,,sin (2nve — “322 );

According to the superposition principle, the resultant

Figure 27.1

oscillation at point F,
E=E,+E,

= 2E,, cos lz (2,— 2) |sin [2ave— = (a, — 7) |'

is harmonic oscillation with the same frequency v but

with an amplitude
E,=2E,, cos [= (z,— 1) | (27.3)
that depends on a parameter, (1/A,) (rz, — re) = (11/Aq) 6.
Squaring (27.3), we obtain the distribution of the light
intensity on the screen:
6 2nd
I= 4], cos? (4—) = 2% [1 + cos( Io ye (27.4)

Let us link the path difference 6 with the coordinate z of

point F on the screen. From the similarity of triangles
ABC and DFO (assuming that 6 ~ | BC | and | FO | =
 Ch. 9. Electromagnetic Waves 213

z), we find that

6/d =2/I,. (27.5)
Hence.
6 = (dL) x. (27.6)
Thus, the intensity distribution is given by the fol-
lowing formula:
T=2ly
‘
[1 +cos (; 2nd
aE r)]. (27.7)
The graph representing the function (27.7) is depicted
in Figure 27.2. Allowing for the maximum conditions (27.1)
I

“tor.

/
4

Figure 27.2

and (27.5), we can find the position of the kth maximum,

Lp=Lbd=kLa/d, 2,=3x10%m, (27.8)
and the distance between adjacent maxima:
Ar = 2y4, — Zp = Ldg/d, Az =6 xX 107m. (27.9)
Two real sources of light are not coherent. Hence, the
above problem on the calculation of interference pattern
produced by two coherent sources is ideal. However, the
results and the method of solution are often employed in
calculations of real interference devices. In the majority
of such devices the bean of light is split into two coherent
parts. After the parts of the initial beam have traveled
different optical paths. they form an interference pattern.
Example 27.2. A point-like source S of light with a wave-
length }, = 5 x 107? m is placed at a distance r = 10 cm
from the line of intersection of two flat mirrors, with the
angle a between the mirrors equal to 20’ (Fresnel mirrors).
Determine the number of bright lines of the interference pat-
214 Part 2. Solution of Standard Problems

tern formed on a screen that ts | = 190 cm distant from the

line of intersection of the mirrors (Figure 27.3).

Figure 27.3

Solution. The interference pattern is formed by the

light from two coherent sources I and II that are posi-
tioned at pointsA and B and are the virtual images of the
source S§ of light in the two flat mirrors. This ideal prob-
lem has been solved in Example 27.1. Thus, to calculate
the interference pattern we must determine the distance
| AB | =d between the sources. The distance from the
sources to the screen is L ~ 1 | r, In triangle AOC the
angle AOC is equal toa. Hence,d =2|AC|=2|AO|x
sin a@ = 2ra, since sin « ~ @ for small @’s. Using formu-
la (27.9), we can find the distance between adjacent bright
lines:
Ag = 4 — Ao (l-rr)
2ra

The number of bright lines can be found if we determine

the width of the interference pattern, and this is determi-
ned by the region where the waves generated by saurces |
and II overlap. Figure 27.3 shows that the width of the
interference pattern is depicted by the interval | DE | =
z=2|0,D|=2l tana ~2le. Dividing the width
zx of the interference pattern by the width Az of a bright
 Ch. 9. Electromagnetic Waves 215

line, we find the number N of bright lines:

_ 2 4a%rl 1 OR

Example 27.3. What must be the admissible width dy of

the slits in Young’s experiment so that a distinct interfer-
ence pattern is formed on a screen S positioned at a distance
L = 2m from the slits (Figure 27.4)? The distance between
the slits isd = 5 mm, and the wavelength of the light is
Ag = 95 X 10-7 m.
Solution. In Young’s experiment two slits—points A
and B in Figure 27.4—are the coherent sources of light
that form the interference pattern on screen S. Let us
assume that these sources are point-like. Then the inter-
ference pattern can be calculated via formulas (27.8)
and (27.9). If we shift the sources upward by a distance
d,, the interference pattern also is shifted by d,. Let
us now consider the interference pattern that is formed by
the light of four point sources positioned at points A, A’,
B, and B’. This will consist of two interference patterns
shifted in relation to each other by dp. If this distance is
smaller than the distance between the adjacent dark and
bright lines, which according to (27.9) is equal to A,L/2d,
the total interference pattern is distinct.
Now suppose that we have two nonpoint-likhe coherent
sources (slits of width AA’ = BB’ = d,). According Lo
what has just been said, the total interference pattern
is distinct if

dx, that is, d)<0.1 mm.

Example 27.4. In the device for obtaining Newton's rings

the space between the lens (index of refraction ny = 1.55)
and the transparent flat plate (index of refraction nz = 1.50)
is filled with a liquid with an index of refraction ng = 1.60
(Figure 27.5). The device is illuminated with monochromatic
light (Ag = 6 X 10-7? m) incident at right angles on the
flat surface of the lens, Find the radius Rt of curvature of the
lens if the radius of the fourth (k = 4) bright ring wn trans-
mitted light isp, = 1 mm.
216 Part 2. Solution of Standard Problems

Solution. Interference occurs in the thin liquid wedge

(the index of refraction of the liquid, mo, is greater than n,
and ny). In this thin liquid film of nonuniform thickness
each ray of light splits into coherent rays. In transmitted
light the kth maximum emerges because of the interfer-
ence of ray I, which enters the plate at point A, and

Figure 27.4 Figure 27.5

part II of the same ray ABC, which is reflected at points

A and B and entering the plate at point C (see Fig-
ure 27.5). Since nz > ns and n, > ny, there is no loss of
a half-wave in the course of reflection at points A and B.
ITence, the optical path difference acquired by rays
{ and II is
6 = 2dnz,

where d is the thickness of the liquid film at point A.

Allowing for the fact that
d=p}/2R
and the maximum condition (27.1), we find that

Pang _
2

2 oR = “ho,
which yields the following formula for the radius of
curvature of the lens:

PRs
R==- Rw 66 cm
 Ch. 9. Electromagnetic Waves 217

Example 27.5. Light of wavelength } = 6 xX 10-7 m and

a degree of monochromaticity AX = 5 x 10-'° m is incident
on a plane-parallel glass plate whose index of refraction is
n = 1.5. The angle of incidence is i = 45°. What is the
mazimum thickness of the plate at which the interference pat-
tern in reflected light is still distinct?
Solution. As is known, in the event of interference of
Monochromatic light (A) in a thin film of thickness # and
index of refraction n, the maximum condition has the
form (in reflected light)
2h V a sinti= (k++). (27.10)
If the light is nonmonochromatic, the angular width
of the Ath interference maximum A’ is found from
Eq. (27.10) by differentiating the left- and right-hand
sides of this equation al & = const:
a (2h V n?— sin? i) Ai = (x ++) Ai,

(44)
whence

Ai= q —_—_——.
— (2h | n*—sin? 2)
di

The angular separation &¢ of adjacent maxima for mo-

nochromatic light can also be found from Eq. (27.10)
by differentiating the left- and right-hand side of this
equation at A = const:
4 (2h |’ n? — sin? i) 6i = 06k,

whence, at 6k = 1 (adjacent maxima),

A
6i= So
a (2h Yn?
—sin? t)

An interference pattern is distinct if | Ai | < | 5é |, or

(k++)<q. (27.44)
218 Part 2. Solution of Standard Problems

Substituting the expression for k + 1/2 that follows from

Eq. (27.10) into (27.11), we find the maximum thickness
of the film, hm,., at which an interference pattern can
still be observed:
a2
hmaxS Rinax ~ 0.27 mm.
2A0 Vn2—sin?t’
The degree of monochromaticity of laser light may be
as high as AX = 4 x 10-!? m. Hence, to observe inter-
ference patterns in laser
light we can take a plate
of enormous _ thickness:
hmax © 3.3 cm. The deg-
ree of monochromaticity
of white (visible) light is
AX = 3.6 x 10-7 m= and,
hence, in this case Rma, ©
3.7 X 10-7 m, that is,
to observe interference pat-
terns in white light we must
take an extremely thin film,
with a thickness of about
Figure 27.6 a few tenths of a microme-
ter. A film of such thick-
ness can be obtained in liquid or solid form.
How can we calculate the interference pattern produced
by light from many coherent sources of light rather than
two? A common approach is to employ the vector dia-
gram method. Let us take the simple case of equal ampli-
tudes. We also assume that the phase difference of any
two adjacent sources differs by the same value, Ag =
const.
Figure 27 6 depicts a vector diagram corresponding to
the addition of NM=6 oscillations with the same ampli-
tudes,
|AB| = |BC| =|CD| = |DE| = |EF| = |FG| = Ey.
The amplitude of the resultant oscillation is depicted by
segment AG = E,. Let us find this quantity. Obviously,
points A, B, C, D, E, F, and G lie on the circumference of
a circle of radius R = |]OA}]=J|OB|]=.... Lel us
 Ch. 9. Electromagnetic Waves 219

drop perpendiculars OK and OL from the center of the

circle, O, onto segments AB and BC. Then xKOL =
Ag and, hence, xKOB = Ag/2. From triangle KOB,
the radius of the circle is given by the following formula:
_ _1|KB _ Eon
~ sin(Ag/2)~ 2 sin (AP) : (27.12)

Since | AH | = | HG | (OH__AG by construction), the

resultant amplitude is
E,= |AG| =2|AH| (27.13)
Angle AQH is equal to (4/2) (2n — NAg) = n—-(1/2)N Aq,
and from triangle AOH we find that

|AH| = Rein (x —4°?) = Rsin (OP).

Substituting this value of | AH | into Eq. (27.13) and
allowing for (27.12), we get
sin (NAgQ/2)
Ey= Ey, sin (Ag/2) °
The energy of the oscillations (and the intensity /) is
proportional to the square of the amplitude. Consequent-
ly, the intensity of the resultant oscillations is
= sin? (NAqg/2)
T=Ip, “sin? (Aq/2) (27.14)

where /, is the intensity of the light from one source.

For a small phase difference (Ag +0), Eq. (27.14)
assumes the form
[= Iy,N?.
Thus, the intensity of the principal maximum tn the tnter-
ference pattern produced by the light from N sources ts
proportional to the square of the number of sources.

28. Diffraction of Light

The basic problem in studying diffraction is to
calculate the diffraction pattern, that is, find the
distribution of light inlensity J.
220 Part 2. Solution of Standard Problems

A more restricted problem is to find the position of the

maxima and minima in the diffraction spectruin. Often
the Fresnel zone method and the DI method (see Section 6)
are employed in calculating diffraction patterns.
Example 28.1. A plane monochromatic wave with a wave-
length i is incident at right angles on an infinitely long
rectangular slit of width a (Figure 28.1). Find the distribu-
tion of the light intensity I in the diffraction pattern on

ty aap
dt 4

Figure 28.1

a screen S. Solve the same problem for a system of N parallel

slits separated by opaque sections of width b (a diffraction
grating).
Solution. Employing the Fresnel zone method, we can
easily find the minimum condition for diffraction on
a single slit,
asing = ka, (28.1)
and the maximum condition,

asing = (2k+1)—+. (28.2)

This, however, does not provide us with the sought dis-
tribution of the light intensity / in the diffraction pat-
tern. Let us employ the DI method. The zone of width dz
(see Figure 28.1) that is x distant from the edge C of the
slit sends a wave in the direction specified by an angle q,
 Ch. 9. Electromagnetic Waves 221

with the equation of this wave being

dE =dE, cos (ot— = zsing) . (28.3)

where
dE, —=c'a, ¢ = const. (28.4)
In Eq. (28.3) we have allowed for the fact that for
waves propagating in the direction CD the distances are
reckoned on this straight line. Hence, | CD |= z sin g,
and the phase of the wave emitted by zone dz is wt —
(2n/d) x sin q.
Integrating Eq. (28.4) over the entire slit for point O,
we obtain the value of the arbilrary constant:

Ye
E,= \qlee.

Substituting the value of dE. from (28.4) into Eq. (28.3),

we find that
-
dE -= fo
: cos(wot aA
an
1 zsin 9) dz. (28.5)

Integrating Eq. (28.5) over the entire slit, we get

a

E= ( Fo cos (or — sin ¢) dz

0
az! sin [(1/A) asin q] 1 :
— [fee |cos (wt ae on ¢) °

Hence, the amplitude of oscillations at point A is

_ p sin [(4/A) asin 9}

Since light intensity is proportional to the square of

the amplitude, Eq. (28.6) yields the expression for the
distribution of light intensity on a screen in the event of
diffraction on a single slit:
= sin? [(1/A) a sin g)
To, = 10 —Tayasin gh? een)
222 Part 2. Solution of Standard Problems

The reader can easily see that the minimum condition

(28.1) follows from Eqs. (28.6) and (28.7).
A diffraction grating consists ef N parallel slits, each of
width a, separated by opaque segments, each of width b
(Figure 28.2), The sum a -} 6 is known as the grating space.
To calculate a diffraction pattern obtained via a diffrac-
lion grating, let us employ the Fresnel zone method. We

Figure 28.2

divide the wavefront of a plane monochromatic wave that

is incident vertically on a diffraction grating (see Fig-
ure 28.2) in each slit into Fresnel zones by parallel planes,
just as in the case of diffraction on a single slit. The
distance between adjacent planes is 4/2. If into each slit
there fils an even number of zones, then in the given
direction (point A) a diffraction minimum forms. But
if into each slit there fits an odd number of zones, each
slit contains a nonextinguished zone. Suppose that these
zones are positioned at the left edges of the slits (points B,
C, D, etc.). The path difference between adjacent sources
(nonextinguished zones) is constant:
§6=|BE|=|CF| =... =(a+) sing. (28.8)
To this path difference 6 there corresponds a constant
phase difference
Ag = 28 . ner
ot sine
)sin (28.9)
 Ch. 9. Electromagnetic Waves 223

Hence, the problem of calculating a diffraction pattern

produced by a diffraction grating has been reduced to
a problem of calculaling the interference pattern pro-
duced by the light from many coherent sources wilh a con-
stant phase difference (28.9). This has been solved in
Section 27 by the vector diagram method. Allowing for
(27.14) and (28.9), we oblain the distribution for the
light intensity in a diffraction spectrum produced by
a diffraction grating:
sin? [Nx (a 10 sin |

LonON = 163.wi ——
sin? To
[ serene | (28.40)

where J, is the intensity generated by a single slit (see

formula (28.7)).
Equation (28.10) provides the principal maximum con-
dition:
(a + b) sin pg = ka. (28.11)
Example 28.2. A plane monochromatic wave of wavelength
Mo = 5 X 10-7 m is incident on a slit of width a=
10-? mm at right angles to the slit plane. Find the angu-
lar position of the first maximum in the diffraction pattern.
The medium is a vacuum.
Solution. The angular position of the first maximum can
be found from the maximum condition (28.2). Hence,

g=sint (3), pw 448’. (28.12)

The angular position may be determined more precisely
with the help of formula (28.7). Let us determine the
extremum of the function J5, = J,, (g) by finding the
first derivative with respect to m and nullifying it:

dl, 21, = acos sin (= asing)

0= L =
dg (= asing)

x {cos ( asin 9) = asin g—sin (Be

72 sing)}.
224 Part 2. Solution of Standard Problems

This yields the following transcendental equation for

finding the extremum values of 9g:
t an (>
x asi
in y)=z
j= asing,
asin

which after introduction of the notation

m =(ad)sing (28.13)
assumes the form
tan mum = am. (28.14)
The roots of this transcendental equation are the fol-
lowing numbers:
m, = 1.43, mz, = 2.46, my = 3.47, .. .. (28.15)
Combining (28.15) with Eq. (28.13), we can find the angu-
lar position of the first diffraction maximum:
ga sin (+) gw 46". (28.16)
From (28.16) and (28.12) it follows that the more exact
solution (28.16) differs considerably from the approximate
solution (28.12). The error in the approximate solution
can easily be calculated:
e AP 100%, = 12%
100% W 504,
(q) 252’

It should be noted that formula (28.7) allows us not on-

ly to find the exact angular position of the maxima in
the diffraction pattern produced by a single slit but also
to determine the intensity of these maxima.

Example 28.3. Find the maximum order of a diffraction

spectrum produced by adiffraction grating with a grating
space a |- 6 = 0.005 mm on which a plane monochromatic
wave with wavelength hy =6 x 10-7 m is incident in
a vacuum at right angles to the grating plane.
Solution. The maximum order of the diffraction spect-
rum can be found from the maximum condition (28.41).
Since the absolute value of sin ¢ cannot exceed unity,
we have
a+b=hyag, Ao.
 Ch. 9. Electromagnetic Waves 225

Hence,
Kmax=(@+6)/Ag, Kmax © 8.

This equation is inexact, however. It was obtained on

the assumption that in (28.10) the light intensity Jo,
created by a single slit is constant and independent of
angle ¢. But Eq. (28.7) shows that J, depends on angle
q@ and can assuine zero values for certain values of 9.
The solution found determines only the maximum possible
order of the spectrum. However, not all the principal
maxima defined in (28.11) realize themselves: those whose
position coincides with a minimum in the diffraction pat-
tern from a single slit, (28.1), disappear; the only prin-
cipal maxima that materialize are those that fall onto the
central maximum of the diffraction pattern from a single
slit. Hence, the maximum order of the realized principal
maxima is determined by the combination of (28.1)
with (28.11).
If in Eq. (28.1) we put k = 4, we find the angular half-
width of the central maximum of the diffraction pattern
from a single slit:
a SiD min = A. (28.17)
Using Eq. (28.11), we can find the maximum order of the
realized principal maxima:
. (a + 0) Sin @min = KmazA-
Allowing for condition (28.17), we get
kmax = (a-+b)/a.
For the final solution of the problem we need to specify
the width of a slit, a. For a, = 10-* mm and a, = b =
2.5 x 10-3 mm we obtain, via the last relationship,
the following results:
kmmax =5, knax = 2.

Note that in the last solution we have neglected the re-

alized principal maxima that fall onto the region of the
maxima that follow the central maximum of the diffrac-
tion pattern from a single slit. It can be demonstrated
that the intensity of these subsequent maxima is low.
15—0498
226 Part 2. Solution of Standard Problems

Example 28.4. The intensity of the central marimum in

the diffraction pattern from a single slit is Ig. Find the
ratios of the intensities of the three subsequent maxima to the
intensity of the central mazimum, I).
Solution. From the maximum condition (28.43) for
diffraction on a single slit,
(a/A) sing = m,
with m = 1.43, 2.46, 3.47 . . . (see (28.15)), and formu-
la (28.7) we find the sought ratios:

(“eyey ay
I I

(ey (RYT
== mm,
Substitution of numerical values yields
I
I Tm,

(1q,/Io)’ = 0.047, (Ip,/1o)"0.017, (1q,/1o)” © 0.008.

THERMODYNAMICS AND KINETIC THEORY

Chapter 10
THERMODYNAMICS

29. The First Law of Thermodynamics

Experience shows that all macro-objects consist of micro-
objects such as molecules, ions, atoms. Micro-objects are
in a state of chaotic (thermal) motion. Since molecules,
atoms and the like are extremely small, a medium-sized
macro-object contains an enormous number of micro-
objects. For example, in one cubic centimeter of ideal
gas in standard conditions there are 2.7 x 10! molecules.
Hence, the physical systems that must be considered when
solving problems of this section consist of a large num-
ber of objects. It can easily be shown that a dynamical
(mechanistic) description of such systems is not only
practically impossible but even meaningless. Therefore,
 Ch. 10. Thermodynamics 227

there are two methods of studying physical systems in

molecular physics, methods mutually complementary,
namely, the thermodynamic and the statistical. The
statistical method is considered in Chapter 11.
The basis of the thermodynamic method is several fun-
damental laws derived from experience. First, the equa-
tion of state
f(p, V, T) =90, (29.1)
with p the pressure, V the volume, and T the thermody-
namic temperature of the system. In this and the follow-
ing chapter we consider only one physical system, the
ideal, or perfect, gas. For an ideal gas the equation of
state (29.1) is transformed into the ideal gas law
pV = + RT, (29.2)
where m is the mass of the gas, M its molecular mass, and
R = 8.3144 J-mol-? K-" the molar gas constant.
Equations of state (29.1) and (29.2) remain valid only
for physical systems in a state of thermodynamic equilibri-
um. In such a state a physical system is characterized
at each point of volume V by well-specified values of p
and T that are the same for all points. Hence, a thermo-
dynamically stable state of a physical system consisting
of a large number of molecules is characterized by asmal]
number of parameters (pressure p, volume V, tempera-
ture 7, and a few other). These are known as macro- para-
meters, and thé state of such a system is a macro-state.
The concept of a thermodynamically stable (or equili-
brium) state of a system is an idealized one. In any real
case, the pressure p or temperature T at a point in vol-
ume V occupied by the system varies, but these variations
(for an equilibrium state) must be so small that they can
be ignored.
Another basic component of the thermodynamic method
is the first and second laws of thermodynamics. By the
first law of thermodynamics,
8Q = dU + 4A, (29.3)
where
§Q == CaT (29.4)
15*
228 Part 2. Solution of Standard Problems

is the elementary amount of heat received by the system,

C the molar heat capacity of the system, dU the variation
of the internal energy of the physical system, and
8W = pdV (29.5)
is the elementary work done by the system. For an ideal
gas,
dU=" ar, (29.6)
where z is the number of degrees of freedom of the mo-
lecules of the ideal gas. '
The first law of thermodynamics in the form (29.3) is
valid for elementary quasi-static processes. As a result of
a quasi-static process thesystem passes through a sequence
of equilibrium states. Since an equilibrium state can be
depicted by a point in a certain system of coordinates
(usually the p-V, or pressure-volume, coordinates),
a quasi-static process is depicted in the same system of
coordinates by a curve. Such diagramatic representation of
various processes is very often used in solving problems
by the thermodynamic method.
The following processes are assumed to be quasi-static:
the isochoric (V = const, m = const), the isobaric (p =
const, m =const), and the isothermal (T = const,
m = const). Other processes, say, the adiabatic (6Q = 0),
can also be quasi-static if they proceed so slowly that
the system passes through a sequence of equilibrium
States.
The amount of heat 6Q and the work 6W are characteris-
tics of heat transfer and the capacity for work. These two
processes are distinct: the first occurs on the micro-
level as a result of the interaction of micro-objects (mole-
cules, atoms, and the like), while the second occurs on the
macro-level as a result of the interaction of macro-objects.
Heat transfer is said to be elementary if the temperature
variation dT is so small that the molar heat capacity C
may be assumed constant. Then the amount of heat can
be calculated by formula (29.4). A common way to
calculate the amount of heat in the case of a nonelementa-
ry process of heat transfer is to employ the DI method
 Ch. 10. Thermodynamics 229

(see Section 6):

Ts
Q@=\5/Car. (29.7)
T;
To evaluate this integral we must know the dependence
of C on other parameters.
If C is constant (C = const), the process is said to be
polytropic, with :
=-4-C(T,—T,) (29.8)
for such processes.
The process of performing work is said to be elementary
if the variation of volume,
dV, is so small that the ? 2
pressure p can be assumed
to be constant. Of course,
the pressure changes in an
elementary process, but 1
this variation dp must be
so smal] that it can be
ignored and the pressurecan 0 Vj Vo V
be considered approximate-
ly constant. Then work Figure 29.4
can be calculated by for-
mula (29.5). For a nonelementary process work is calcu-
lated by employing the DI method:
Vs
W= (pav. (29.9)
Vs
In the p-V system of coordinates, work is numerical-
ly equal to the magnitude of the hatched area in Fig-
ure 29.1 (curve 1-2 depicts the corresponding process).
Thus, an elementary process for which the equation
expressing the first law of thermodynamics in the form
(29.3) must satisfy the two conditions formulated above.
For a nonelementary process the first law of thermo-
dynamics is written in the form
Ts Vs

| 4 CaT=AU+ \pay, (29.10)

Th Vi
230 Part 2. Solution of Standard Problems

or, if we allow for (29.7) and (29.9),

Q@=AU+W, (29.41)
where AU = U, — U, is the variation of the internal
energy in the process (this variation does not depend on
the type of process but only on the initial and final states
of the physical system).
The basic problem of the thermodynamics of equi-
librium processes is to find all the macro-states
of a physical system.
If the initial and final states of a system are known, we
can find the variation of the system’s internal energy. If,
in addition, we know
the intermediate states
(i.e. the process), we can
find the work performed
by the system, calculate
the amount of heat re-
ceived or liberated, and so
on.
Example 29.1. An
Yo kV Y amount of hydrogen con-
: tained in one cubic metre
Figure 29.2 under standard conditions
is first isochorically trans-
ferred to a state with a pressure that is n times higher
than the initial pressure and then isobarically transferred
to a state with a volume that isk times greater than the initial
volume. Determine the variation of the internal energy of
the gas, the work done by the gas, and the amount of heat
received in the process.
Solution. The physical system consists of a certain
mass m (which can easily be calculated) of an ideal gas
whose molecular mass M is known. The initial macro-
state of the system (point J in Figure 29.2) is known (the
standard pressure py ~ 10° Pa, the standard temperature
T, = 273 K, and the volume V = 1 mi? specified in the
terms of the problem). The states and processes in which
the system takes part are depicted on a p-V diagram
 Ch. 10. Thermodynamics 231

(see Figure 29.2). Let us find the parameters of the

second (point 2) and third (point 3) macro-states of the
system. For this we use the ideal gas law (29.2) and the
definitions of isoprocesses:
MpoV2 _ MnpoVo
P2=NPy, V,=Vy, T= aR RR (29.12)

Pa= P2=NPp, Ve=kVy, T,=Absts

— Mubple |(99.13)
Since the processes in which the system participates are
quasi-static and polytropic, the sought quantities can be
found by using the formulas found earlier. The variation
in internal energy is
AU = Hr- T,) = “pam ( Atetale _ Mro¥o )

=> PoV, (nk — 1). (29.14)

In an isochoric process dV = 0 and no work is done.
The work done in the isobaric process is
W = pz (Vs — Ve) = npo (KVo — Vo) = Po Von (k — 1).
(29.15)
The amount of heat
Q=Q+ Qe = Cy (T2 — To) + Gp Cp (Ts — To)
= Pole fi (nk — 1) + 2n (k—1)], (29.16)
where we have allowed for Mayer's formula
Cp=Cy +R. (29.17)
The amount of heat can be calculated by using the
formula (29.11), which expresses the first law of thermo-
dynamics:
Q=AU
+ W = pV, (nk—1) + PoVon (k—1)
= Pale [i (nk — 1) + 2n(k - 4)];
this coincides with the earlier result (29.16). To carry out
a numerical calculation we must select reasonable values
for n and k. Equations (29.13) show that the value of n
232 Part 2. Solution of Standard Problems

determines the maximum value of pressure, py = nDo,

while the product nk determines the maximum temperatu-
re, 75 = nkT,. The value of n cannot exceed nm,x = 100,
since at pressures equal to (or greater than) 100 p, the gas
ceases to be ideal. The product nk cannot exceed the value
(nk)max, * 10 since at temperature 7, ~ 107, ~ 3 X
10° K (and higher) the walls of the vessel containing
the hydrogen could melt (they must be cooled) and molec-
ular hydrogen transforms into atomic hydrogen; at still
higher temperatures the atomic hydrogen transforms into
hydrogen plasma. If we put n = 5 and k = 2, we find
that Eqs. (29.14), (29.15), and (29.16) yield
AU = 2.2 xX 10°J, W=5 x 10°J, Q = 2.7 x 109J.
It is advisable to study the following problem: what
must be the relationship between n and k (at nk = const)
if we want the ratio W/Q to be maximal? But we will
leave this problem for the reader to solve. Instead we pose
the following problem: how do the sought quantities
change if the system proceeds from the initial state to
the final state quasi-statically along the dashed straight
line in Figure 29.2? The answer can easily be found if we
study formula (29.11) for the first law of thermodynamics.
The variation of internal energy AU, does not depend
on the type of process but on the initial and final states
of the system. Hence, the variation in internal energy
does not change: AU = 2.2 x 10° J. The amount of
work done by the system will decrease (the area of the
trapezoid ABDE is smaller than the area of the rectangle
ACDE). Hence, according to the first law of thermody-
namics (29.11), the system wil] receive less heat.
Example 29.2. Two moles of nitrogen, Nz, are under stan-
dard conditions. They are then transformed isothermically
into a certain state and then quasi-statically and adiabatical-
ly into a finite state with a volume that is four times the
initial one. Find the work performed by the gas if Q =
11 300 J of heat was transmitted to the gas in the isother-
mal process.
Solution. Let us determine the intermediate and final
macro-states of the system. The ideal gas law results in
 Ch. 10. Thermodynamics 233

an indeterminate system of equations. Let us employ the

first law of thermodynamics in the form (29.11). For the
isothermal process,
Ve Vs

Q=W,=( pav=| rom2 rt, dv

= m
RT In a
Vv

Vo Vo

where W, is the work done by the gas in the isothermal

process. This enables us to find the volume that the gas
must occupy in the intermediate state:
V.=VpeQM/mR,
Next, using the equation governing an adiabatic process,
-1 -1
TV} =P; (4V)” ’
with py = C,/Cy the molar heat capacity ratio (commonly
known as the specific heat ratio), we can find the final
temperature:
eQM(y-1)/mRT
T; = T> 4v-t

Formula (29.6) yields the following expression for the

internal-energy variation:
m iR __ m GIRTy Lf e@MV—
1/mRT
AU = 37-7 (Ts
T) =F “3 |ar — —!]-
This gives the following result: W ~ 4500 J. Note that
after finding the parameters of the macro-states we could
have calculated the sought amount of work directly by
employing formula (29.5).
Example 29.3. For an ideal gas find the equation of
a process in which the heat capacity of the gas varies with
temperature according to the law C=a/T, with a=
const.
Solution. The process is not polytropic. Hence, we can
apply the first law of thermodynamics in the form (29.3)
for one mole of the gas:
+ dT =Cy d+ pav.
234 Part 2. Solution of Standard Problems :

Using the ideal gas law (29.2), we can rewrite this

equation:
racy.
dT ear
Dividing the left- and right-hand sides by RT and inte-
grating, we get

payeetes= In 7+1n V + const.

RT
This gives us the sought equation of the process:
V7'/-Yea/RT — const.

30. The Second Law of Thermodynamics

As a result of some process a system may return to its
initial state. Such a process is known as cyclic. Using the
first law of thermodynamics, we can prove that the ther-
mal efficiency of an arbitrary cycle is
N= (Q,— Q2)/Q1, (30.1)
where Q, is the amount of heat received by the system
from the heater, and Q, the heat rejected by the system
to the cooler. For the Carnot cycle (two isotherms and
two adiabatic curves) we have
n=(T,—T,)/T,, (30.2)
where 7, is the temperature of the heater, and T, the tem-
perature of the cooler.
The ratio 6Q T is known as the reduced heat of an ele-
mentary process. According to Clausius’ theorem,
the sum of reduced heats for an arbitrary cycle is
a negative quantity, and for a reversible cycle it
is zero:
§2 <0. (30.3)
Consequently, it follows that
2
the sum of reduced heats, \T-'5Q, for any reversible
1
process does not depend on the type of process but is
 Ch. 10. Thermodynamics 235

determined solely by the initial (7) and final (2)

states of the system.
Next we introduce the notion of entropy S of a system
as a state function whose variation depends only on the
initial and final states of the system in the following
moanner:
2
BeeG \a (30.4)
1

where integration is carried out over any reversible pro-

cess as a result of which the system goes from state J to
state 2.
Example 30.1. The cycle depicted in Figure 30.1 consists
of two isotherms (T, = 600 K and T, = 300 K) and two
isobars (p, = 4p,). Determine the thermal efficiency of the
cycle if the working substance
is an ideal gas whose mole-
cules have five degrees of freedom
i = 5).
Solution. The physical sys-
tem consists of one mole of
an ideal gas. A cyclic process
consisting of two ‘isotherms
and two isobars (see Figure V, Vay Ve V
30.1) occurs in the system.
To find the efficiency of the Figure 30.1
cycle using formula (30.1), we
must determine Q, and Q,. The system receives an
amount of heat Q, in the isobaric transition from state /
with parameters p,, V,, T, to state 2 with parameters p,,
Vi, 7, and in the isothermal expansion from state 2 to
state 3 with parameters p,, V2, 7:
Q,=C, (T,—T,.) + RT, In (V./V)). (30.5)
The system rejects an amount of heat Q, in the isobaric
transition from state 3 to state 4 with parameters p., V;,
T, and in the isothermal compression from state ¢ to
state 1:
Q.=C>p (T, —T.) + RT, In (V5/V)). (30.6)
238 Part 2. Solution of Standard Problems

From Boyle’s law for the isotherms 7, and To,

PiVi=P2V. and p,V,= pV),
it follows that
Veset
Vi VY Ps * .
Substituting these volume ratios into (30.5) and (30.6)
and allowing for (30.1), we obtain
y= —_R(Ti=Ts)\n (pil Pe)
Cp (T1—T2)
+ RT, In (Pi/po) *
Using the well-known relationship Cp = (1/2) (i + 2) R,
with i the number of degrees of freedom, we finally get
T,—T,
(t-+2)(7T1—Ta) *
Pit——3in (Py/pa)
Calculation yields n ~ 22.5%.
The thermal efficiency of a Carnot cycle with the same
temperatures 7, of the heater and 7, of the cooler is
n=(T,—T,)/T,, 4 = 50%.
If the degree of compression is increased (say by putting
Pi/P2 = 20) and the number of degrees of freedom of the
gas molecules is decreased
(say, to i = 3), the thermal
efficiency of a cycle consis-
ting of two isotherms and
two isobars can be increased
by up to yn ~ 35%. But in
all cases it remains lower
than the efficiency of the
respective Carnot cycle.
Figure 30.2 Example 30.2. A cycle
consists of an isotherm (T, =
600 K), an isobar, and an isochor (Figure 30.2). The
volume ratio V2/Vg is equal to two. The working substance
is an ideal gas with i = 5. Determine the efficiency of the
cycle as a function of the maximum (T;) and minimum
temperatures of the working substance.
 Ch. 10. Thermodynamics 237

Solution. Let us find the minimum temperature. In the

isobaric process the gas cools off and in the isochoric pro-
cess it heats up. Hence, the minimum temperature is the
one which the substance has in state 3, or Ty. From the
ideal gas laws for states 2 and 3,
pV. = RT, and pV; = RTs,
we can find the minimum temperature:
T; = T,V5/V. = 0.57,, T; = 300 K.

Since all the processes in the cycle are polytropic, we

can use the first law of thermodynamics for the isothermal
process and formula (29.8) for the isobaric and isochoric
processes to find the amount of heat absorbed by the
working substance (or rejected by it) in these processes.
For the isothermal process,
Qie = RT, In (V,/V,).
Since V, = Vs and V./V, = 7,/T;, we have
Qre = RI, In (T/T).
For the isobaric process,
Qes
=Cp (Ty— T) = +42 RT,-1)).

Finally, for the isochoric process,

Qs =Cyv (71-7)
= (7, —T))-
R

Combining these results according to formula (30.1), we

find the efficiency of the cycle:
n= Q1r2 + 0s1 —Qaa = (1/2) (i+-2) (T1—T)
Qi2z + Qa1 Ty In (74/73) +(1/2) t(7;—T3) ?
n ~ 10%.
Example 30.3. Find the variation of the entropy on one
mole of an ideal gas in the isobaric, isochoric, and isother-
mal processes.
Solution. The physical system, one mole of an ideal gas,
Participates in three isoprocesses. The processes are
quasi-static and reversible. Hence, the entropy variation
238 Part 2. Solution of Standard Problems

can be found directly by employing formula (30.4). For

the isobaric process,
Te Te
CpaT T Vv
As,=| 2=( pet =C,inz*=C,in>*. (30.7)
T, T1 :

For the isochoric process,

® 50?
Asy ==| =|. cyar
2=( T
errpo =Cy Ingt=Cy
a In,
Pa
T1 T1
(30.8)
Finally, for the isothermal process,

ssn | Pmf nl ah Ve
_(
=| Rta
AR _=Rin V3
tt. (30.9)
V1
Let us now apply the above results to the cycle of
Example 30.2. The cycle consists of an isobar, isochore,
and isotherm. All the processes are reversible, and so is
the cycle. According to Clausius’ law (30.3), the variation
in entropy in a reversible cycle is zero:
AS, + ASy + AS; = 0.
Hence, allowing for (30.7)-(30.9) and the notation used
in Example 30.2, we find that
T T V
Cplnae +Cy In=- +R In T= ,

Since V, = Vg and V./V, = T,/T, (see Example 30.2),

we have
—C, In
qT: + Cy nar
T, +R>Ti...=0,

which leads to the well-known Mayer relation (29.17):

Cp = Cy + R.

We depict the entropy variation in the cycle using the

S-ln T coordinates (Figure 30.3). On the segment J-2 of
 Ch, 10. Thermodynamics 239

isothermal expansion (In 7 = const) the entropy in-

creased by AS; = 0.7R; on the segment 2-3 of isobaric
cooling the entropy decreased by AS, = 3.5 x 0.7R;
finally, on the segment 3-1 of isochoric heating the entro-
py increased by ASy = 2.5 x 0.7R. The total entropy
variation in the cycle is zero: AS, + ASy + AS7 = 0.
The increase in entropy on segments 3-7 and 1-2 seems
natural (this agrees with the general law of entropy
increase), but at first glance it seems that the entropy on
segment 2-3 should increase too (according to the same

INS
SSNSPowe \
078
48,%3.5*
]
eal
WY) feed
Int

Figure 30.3 Figure 30.4

law). However, there is no contradiction here. The law is

valid only for an adiabatically isolated system, and the
system considered in Example 30.2 is not adiabatically
isolated: during isothermal expansion J-2 and isochoric
heating 3-1 it receives heat, while during isobaric cooling
2-3 it rejects heat to external objects.
Example 30.4. An adiabatically isolated vessel is divided
into two equal parts by a rigid partition that does not conduct
heat (Figure 30.4). Each half of the vessel is filled with one
mole of the same ideal triatomic gas: at T, = 600 K in the
left half and at T, = 300 K in the right. Then thepartition
is removed from the vessel. Determine the entropy variation
in the gas after an equilibrium state sets in.
Solution. Let us consider three physical systems. Sys-
tem I consists of one mole of the gas in the left half of
240 Part 2. Solution of Standard Problems

the vessel at a temperature 7,. Prior to removal of the

partition the gas in the left half is adiabatically isolated
and in an equilibrium state. System II consists of one
mole of the same gas in the right half of the vessel at
a temperature 7, < 7,. Prior to removal of the parti-
tion the gas in the right half is also adiabatically isolated
and in an equilibrium state. System III (the final one)
appears because the partition between systems I and II
has been removed. Prior to removal and after removal
in systems I and II and in system III, respectively, there
occur quasi-static and irreversible processes as a result
of which equilibrium states set in in all three.
Since the processes are irreversible, we cannot use for-
mula (30.4) to find the entropy variation in system III
directly. What is needed are processes that will take
systems I and II from the initial state to the final state
reversibly. For this the initial adiabatic isolation of
systems I and II must be violated. Suppose that instead
of the partition that cannot conduct heat we take a par-
tition that is massless and conducts heat ideally. Then
systems I and II are in thermal! contact, that is, they
are no more adiabatically isolated. In each there occurs
a (reversible) isochoric process: isochoric cooling in the
left half of the vessel and isochoric heating in the right
half to a common temperature 6. The final equilibrium
temperature can easily be found:

8 = (T, + T,)/2.

The next step is to remove the modified partition. Since

both subsystems, I and II, are in a state of thermodynamic
equilibrium at a temperature 6, the overall system III
is also in a state of thermodynamic equilibrium. Note
that while the isochoric processes occurring in systems
I and II can be considered reversible, the process of heat
transfer in system III cannot be thought of as reversible.
Let us denote the entropy variations in systems I and II
by AS, and AS,, respectively. Then the variation of entro-
py in system III is

AS = AS, + AS».
 Ch. 11, Kinetic Theory 241

According to (30.8), for system I we have

8
\ar 8 1 T
AS, = =Cy In--=Cy Ine (1+ 4%),

and for a II we have

0

AS, = \
. Cy dT
= =Cy nz(3) =Cy In=1 (1 +44).
Te

It is easy to see that AS, is negative and AS, positive,

that is, the entropy of system I decreases and that of sys-
tem II increases (the reader will recall that these sys-
tems ceased to be adiabatically isolated after a heat-
conducting partition was inserted instead of the initial
one; hence, the entropy of cach can increase or decrease).
The overall system III remains adiabatically isolated
and its entropy must increase as a result of an irrevers-
ible process. Indeed, | AS, |< | AS, | and
AS =AS, +AS,
=Cy[In + (1 +7t)+h +(1 +#)] > 0.
Since Cy = iR/2, we have AS ~ 3 J-mol-!K~-). Thus,
if the amount of heat Q, rejected by system I is equal to the
amount of heat Q, received by system II (| Q, | = | Qe |),
the absolute value of the entropy variation AS, in sys-
tem I is not equal to the absolute value of the entropy
variation AS, in system II in the same heat transfer
process (| AS, | # | AS, |).

Chapter 11
KINETIC THEORY

31. The Maxwell-Boltzmann Distribution

In the statistical method, in contrast to the thermodynam-
ic, an essential assumption concerns the “granular”
structure of macro-objects. This method employs the
following propositions (corroborated by numerous ex-
periments):
16-0498
242 Part 2, Solution of Standard Problems

all macro-objects consist of micro-objects, which

participate in chaotic motion and interact with
each other.
In classical statistical physics it is assumed that
no two similar macro-objects are identical.
The behavior of a single micro-object (a particle) is
studied in the six-dimensional phase space (p-space) of
three position coordinates (z, y, z) and three projections
of momentum (pz, py, Pz) or three projections of velocity
(vz, Uy, Vz). The state of a single micro-object is spec-
ified by a point in this space. If the micro-object moves
chaolically, its occurrence inside a volume element
dt = dzdy dz dp, dp, dp, in this space constitutes a ran-
dom event whose probability is
dw =f (x, y, 2, Px» Py» Pz) OT, (34.1)
with f the distribution function (the probability density),
The function / satisfies the normalization condition
(fdat=1. (31.2)
In (31.2) integration is carried out over the entire phase
space. Using the concept of a distribution f, we can define
the mean value of a function @ (z, y, Z, Px: Py, Pz) thuss
(@) = \fo dt. (34.3)
The Masxwell-Boltzmann distribution of molecules in
the p-space hasthe form
dw (z, Y, 2, Per Pys P:)

p2t+pz+p? 7|
_ Lee ee, y. a | RT
dzdydzdp,dp,dp,, (31.4)
where U (x, y, z) is the potential energy of a molecule,
and m the molecule’s mass.
The Maxwell-Boltzmann distribution can be thought
of as two independent distributions in a three-dimension-
al momentum space (the Maxwell distribution),
~(ph+P}+P2)/2mkT
dw (Py, Pys Pz)
=Ae Pe TP dp,dpydp,, (31.5)
 Ch. 11. Kinetic Theory 243

and in a three-dimensional coordinate space (the Boltz-

mann distribution),
dw(z, y, 2)=Be-Ue.u 2W/2kT dzdydz, (31.6)
where A and 3 are constants that can be found from the
normalization condition (31.2). Allowing for the nor-
malization condition (31.2), we arrive at the Maxwell
distribution in the following form:
dw(p,, Py» Pz) = 3 es
(p2, t pe I pg)/2mat
= (2mk7')3/2 ¢ dp,dpydp,, (84.7)
which makes it possible to find the distribution in velocity
components (vz, V,, U2),
dw (vx, Vy, Vz)
m(v2 +07 +02)

= (sar e 2aT dv, dv, dv,, (34.8)

the distribution in speed (v),
moz

dw (v) = 4n (oaiF \ vie *T dy, (31.9)

the distribution in kinetic energy (£,),
ole.
dw (Ey)
=2n (5 rete *T dey, (34.10)
and other distributions.
When a system consisting of N particles is in thermo-
dynamic equilibrium, its macro-state is characterized
by a relatively small number of macro-parameters (phys-
ical quantities that can be found from measurements in
experiments) having definite,- time-independent values.
Owing to the chaotic motion of the particles their position
and velocity change constantly. This means that while
the macro-parameters remain unchanged, the micro-
parameters vary. Thus, a single macro-state has corre-
sponding to it a multitude of micro-states, which implies
that any macroscopic quantity depends on the microscopic
parameters. In statistical physics it is assumed that
physical quantities observable in experiments
(macro-parameters) can be found as mean values
calculated over the set of admissible micro-states
244 Part 2. Solution of Standard Problems

(see (31.3)). Consequently, one of the main problems

solved by the statistical method is finding mean values
of various physical quantities and determining the mean
number of particles dN (belonging to a collection of
N particles) that possess a certain property.
Example 31.1. Nitrogen ts in a vessel at a pressure p =
1 atm and a temperature T = 300 K. Find the fractional
number of nitrogen molecules whose speed lies within the
interval ranging from (v) to (v) -- dv, with dv = 1 m/s.
No external forces are present.
Solution. At 1 atm and 300 K nitrogen may be assumed
to be an ideal gas. In the absence of external forces,

fa (V)

0 v V

Figure 31.1

the molecules of an ideal gas obey the Maxwell distri-

bution. The concrete form of this distribution is deter-
mined by the terms of the problem, in short, we must
use the Maxwell distribution in the absolute value of the
velocity, (31.9):
= N4n (575 91 y2g-mo%/24T
dN m
dy, (34.44)
where dW is the number of molecules (of the given N mol-
ecules) whose speed lies within the interval from v
to v + dv, and m is the mass of a nitrogen molecule. As
is known, formula (31.11) is valid if dv is so small that
any variation of the distribution function

fur (v) = = An (ae)? e-merznT 2 (34.42)

 Ch. 11, Kinetic Theory 245

in this interval can be ignored, that is, the distribution

function is assumed to be constant (within the interval).
In our case the interval dv = 1 m/s is small (compared
to the mean speed (v) = V 8kT/nm ~ 475 m/s). In
addition, the distribution function fy(v) varies very
weakly in the neighborhood of (v) (Figure 31.1). Hence,
formula (31.11) practically solves the problem. Sub-
stituting the value of the mean speed (vy) = VY8kT/nm
into (31.11), we arrive at the solution to the problem in
general form:
dN 8V2 ee -4/n

(ar) oW NFdv.

Carrying out the necessary calculations, we get

dN/N =1.9-10°3=0.19%,

where we have used the tabulated values of the function

f(x) =e*.
Example 31.2. Find the fractional number of molecules
whose speed exceeds the absolute value of the mean velocity.
Solution. In this case dv is infinite (ranging from (v)
to oo) and formula (31.11) cannot be used directly. How-
ever, if we integrate (31.11) within the above-noted
limits, we find the sought fractional number of molecules:

N 3/2 _ mez
ne m 2kT 0
= an (oF) e v2 dv
(0)

— va
44 93/2 | e-av2
Ne v2y2(|
dv, t
(31.13)

where @ = m/2kT, and N, is the number of molecules

(with the total number being V) whose speed exceeds the
mean speed,
246 Part 2. Solution of Standard Problems

Let us represent the last integral in (31.13) in the

following form:

<1 ast ( e203 p2 dy

Va (0)
oD (0)
4
= 4 3/2 (@- 2? y2 dy— —— ai/? ( e- 20? y2 du.
Vn Va

The first integral on the right-hand side is equal to unity

according to the normalization condition (31.2):

4 gs ( e-2?? y2dv=1.
Pi4 .

To calculate the second integral, we change the variable

in it by setting ¢ = v Ya. Then
(v) Var)
ron e~ 29? 2 v= ( e-Pdt
0 0
1.43
=—_ ( Pedi, (31.14)
Vn y
since Ya (v) =V 4/n = 1.13. We integrate (31.14) by
parts and reduce the last integral to the error function
z

@ (=Te (e-* dr,

la) ae
0

whose values are listed in special tables. We obtain

4.13 113
4 4 eH & 1.43 {
7] fe-Mdt=—_[ - t | | > 3 dr]
Va Va 2 {0 2
= — 0.39-+4+
®(4.13).
From error-function tables we find that ® (1.13) = 0.89.
Hence the fractional number of molecules whose speed
 Ch. 11. Kinetic Theory 247

is greater than the mean is

N,/N == 0.50.
Thus, half of the molecules have a speed higher than
the mean and half lower than the mean.
Example 31.3. Find the number of hydrogen molecules
that every second cross an area of one square centimeter
positioned at right angles to the X azis of a coordinate
system (the hydrogen is kept in standard conditions).
Solution. We give two solutions of this problem.
The first model. Since there is no preferential direction
in the motion of the gas molecules, we must assume that
one-third of all the molecules fly along the X axis, one-
third along the Y axis, and one-third along the Z axis.
Hence, one-sixth of the molecules fly in the positive di-
rection of the X axis. We also assume that all the mole-
cules have the same speed equal to (v.) Then the sought
number of molecules is
n, = (1/6) ng (v) AA At, (31.15)
where n, is the number of molecules per unit volume,
AA = 1 cm? the area through which the molecules fly,
and At =1 s a time interval.
The second model. In the first model all the molecules
were assumed to be moving with the same speed (v).
However, as we know, molecules are distributed in the
components of velocity according to the Maxwell distri-
bution, which in the case of one-dimensional motion
can easily be obtained from distribution (31.8):
dn(v.)—n¢ (se) ' 172
eT MATy
- 4 ’
——(31 16)
¥

Hence, the number of molecules flying through the area

AA = 1 cm*in'the course of At = 1s can be found from
the relationship

n,=AA At
248 Part 2. Solution of Standard Problems

— MoVain
2a
y 4 At |e°* 3d (av?)

=-AA Ay tel ole

AA At
=AAA = —e! =o
. 2Va [elo 2Van
with a = m/2kT. Since (v)= V 8kT/nm, we have
Ng = (1/4) ng (v) AA At. (31.17)
We see that the expressions (341.15) and (31.17) differ
considerably. Carrying out the necessary calculations

f(Ex)

I
|
|
|
I

0 e ‘k. mp

Figure 34.2

and allowing for the fact that mg = po/kT,, where po

and 7, are the standard pressure and temperature, we get
ny + 7.4 K 1073, ng 11.1 x 1078.
Example 31.4. A vessel of volume V = 30 | contains
m = 100 g of ozygen under a pressure p = 3 X 10° Pa.
Determine the most probable value of the kinetic energy of
the oxygen molecules.
Solution. One can easily show that oxygen in the spec-
ified conditions constitutes an ideal gas. The most pro-
bable value of the kinetic energy of the oxygen molecules
corresponds to the maximum in the Maxwell distribution
(31.10) in kinetic energy (Figure 31.2) and can he found
 Ch. 11. Kinetic Theory 249

from the appropriate distribution function,

4 \3/2 e
fu (Ex)
=2n(5 yen Eee" FRAT, (34.18)
IIence, the problem is reduced to finding the extremum
of function (31.18). Finding the first derivative of fi(Ex)
and nullifying it, we get
2.
fm (Ey)
= 20 (<r)” 7 EAT (-7) EI?9

+2n (=)
1
=
3/2 1
Ee
,~1/2 -Ey/kT _
=0.
Hence, the most probable kinetic energy of the molecules
is
Ry mp =AT/2. (31.19)
The temperature can be found from the ideal gas law
T = pVM/mR.
Using (31.18) and (31.3), we can find the mean value
of the kinetic energy of the molecules (in translational
motion):
(Ey) => kT.
Thus, the mean kinetic energy of molecules of an ideal
gas is three times the most probable value of the kinetic
energy:
{Ey )/Ey mp= 3.
Nole that the ratio of the average speed of molecules
and the most probable speed is lower than three:

Ap) VERTIS
Ump V 2k7/m
22449
+

32. The Boltzmann Distribution

The Boltzmann distribution (31.6) for the one-dimensional

case assumes the form
dw (x) = 24) = Beever dz, (32.1)
250 Part 2. Solution of Standard Problems

where dN (z) is the number of particles contained in

a layer of thickness dz in the neighborhood of coordinate x
(N is the total number of particles). Let us apply distri-
bution (32.1) to the atmosphere of Earth. We assume
that the temperature of air in Earth’s atmosphere is
constant (J = const; the case of an isothermal atmo-
sphere) and that the altitude h to which the atmosphere
extends is much smaller than Earth’s radius R (h < R),
which means that the acceleration of free fall in the
atmospheric layer is constant (g = 9.8 m/s? = const).
Then the potential energy of a molecule of mass m at
an altitude x above Earth’s surface is U (xr) = mgr.
Using the normalization condition (31.2), we can find
the value of constant B,:

1=( Bye-mevAT dx, By = mg/KT.

0
Thus, the number of molecules dN (z) in a layer of air
of thickness dz at an altitude x above the surface of
Earth is
dN (2) =T7E o-mex/Ar de. (32.2)
Suppose that dA is the elementary area perpendicular
to the X axis. Then dx dA = dV is the volume clement,
and
Nmg _
“ia > Po

is the pressure of the atmosphere at the surface of Earth.

Hence, the number of molecules contained in a volume dV
at an altitude z is
AN (2) <= Po_
P& 9-msvs/hT AV,
kT

Since dN (z)/dV =n is the number of molecules per

unit volume at altitude x and po/kT =n, is the same
quantity in the surface of Earth, we have
n= Nne~™Ex/AT, (32.3)
 Ch. 11. Kinetic Theory 254

This yields the following barometric height formula

Dp = poo MSz/AT,

Standard problems involving the Boltzmann distri-

bution, like those involving the Maxwell distribution,
are reduced to finding the means of physical quantities
and the number of particles possessing specific properties.
Example 32.1. Find the mean potential energy of air
molecules in Earth's gravitational field. At what altitude
above the surface of Earth is the energy of the molecules
equal to the mean potential energy? The temperature of
the air is assumed constant and equal to 0 °C.
Solution. The gas (air) is in Earth’s gravitational field.
Hence, its molecules are distributed in energy according
to the Boltzmann distribution:

fp (J) = Be-U/r,
where U = mgh is the potential energy of a molecule.
If we know the distribution function f for the mole-
cules in a specific physical parameter / (speed v, moment-
um p, energy £, etc.), the mean value of a specific physical
quantity that is a function of this parameter, 9 = g (I),
is determined from the expression

Somrma
(@ (1)) ==.
(rat
0

In our case g (lt) = U and f = fg. Thus, the mean

value of the potential energy of air molecules in Earth’s
gravitational field is

[BueU"Tay =fue“ VAT ay

0 0

f Be“ VRP ay feVAT ay

0 0
252 Part 2. Solution of Standard Problems

Ilere the denominator

\ e-YATAU=kT
i)
and the numerator

\ Ue-YaTdU = 27? \ te-! dt = KeT?.

0 0
Hence, (U) = kT.
Now let us find the altitude A at which the potential
energy of the air molecules is equal to the mean potential
energy: (U) = mgh, or kT = mgh. Hence,

h== AT
me RT
My? h a= 8x1033 m.

Example 32.2. Determine the mass of air contained by

a cylinder with base area AA = 1 m? and altitude h =
{ km. Assume that the air is in standard conditions.
Solution. Here we cannot apply the ideal gas law since
the physical system, the ideal gas (air), is in Earth’s
gravitational field. We cannot directly employ formula
(32.2) either because the layer thickness dz = h = 1 kin
is large. Integrating (32.2) with respect to x from 0
to h, we can find the total number of air molecules in the
cylinder:
ho __mgx -—2
mgh

Ni=7Re\e dr=n(t—-e * }

_ mA (4 9)
mg

Multiplying (32.5) by the mass m of one molecule, we

arrive at the expression for the sought mass:
__mgh _ Meh

M,=mN,= pea (4. jee eee RT )

where M = 29 kg/mol is the molar mass of air,
 Ch. 11. Kinetic Theory 293

If the air did not experience a gravitational pull,

then the ideal gas law would yield
My = pp AAhM/RT.
Let us find the M, to M, ratio:
M, Mgh

hy ~ RT (Le MEnIRT) *
A nuerical calculation for altitudes kh, = 100 m, h, =
1 km, and hj = 10 km yields the following values for
this ratio: a, = 1.008, a, = 1.08, and a, = 1.8
We see that for air contained in a volume with an
altitude of several hundred meters we can apply the
ideal gas law and ignore the Boltzmann distribution.
For air contained in a volume with an altitude of one
kilometer and more the use of the ideal gas law leads
to considerable errors and in such cases we must allow
for Earth's gravitational field.
It would be interesting to study the dependence of a
on parameters (the temperature 7, the molar mass M,
and the acceleration of free fall g).
Example 32.3. The atmosphere contains dust particles
with a particle mass m = 8 xX 10-*? kg and a particle
volume V = 5 xX 10-** m°. Find the decrease in the con-
centration of these particles at altitudes h, =3 m and
h, = 30 m. The air is in standard conditions.
Solution. The physical system consists of dust particles
and air molecules in the Earth’s gravitational field.
Hence, both the dust particles and the air molecules
obey the Boltzmann distribution (32.3). For air mole-
cules this distribution can be applied directly, while for
dust particles doing so may lead to considerable errors.
The fact is that in addition to being subjected to the
force of gravity mg, the dust particles are acted upon by
a buoyancy force F, since they are submerged in air.
A simple calculation shows that F,, is comparable in
order of magnitude with the force of gravity mg. Indeed,
the density of a dust particle is pg = m/V, or pg = 8 X
10-7/5 x 10-?? ~ 1.6 kg/m’, which differs little from
the air density p,;, 1.3 kg/m®, This means that the
254 Part 2. Solution of Standard Problems

buoyancy force differs little from the force of gravity.

For this reason we first find the effective mass of a dust
particle, Meg = g — Fy/g, or Mme = m — pasrV, where
Pair is the air density. The air density can be found from
the ideal gas law:
Pair = PAUIRT.
Hence,
pMV
Met — Mm --
RT °

Employing the Boltzmann density (32.3), we find the

variation of the concentration of dust particles with
altitude:
Mesh _ (m-pMV/RT)gh

p=—=e ‘TF =e ma

A numerical calculation for kh, = 3 m and h, = 30 m

yields B, = 0.29 and B, = 3 x 10-°, respectively. Thus,
while at the altitude h, = 3m (the height of a one-story
building) the dust concentration is still approximately
one-third of the dust concentration at the surface of
Earth, at an altitude kh, = 30 m (the top of a ten-story
building) there is practically no dust. This conclusion
is valid if there is no updrift.
Part 3
SOLUTION OF NONSTANDARD, NONSPECIFIED,
AND ARBITRARY PROBLEMS

Chapter 12
NONSTANDARD AND ORIGINAL PROBLEMS

33. Nonstandard Problems

Earlier we noted that concrete and generalized knowledge

is not sufficient to solve nonstandard problems. Asa rule
when we apply such knowledge to problems of this type,
we end up with an open system of equations. Then we
have to look for an unspecified “something” that will
enable us to close the system of equations. In nonstandard
problems this unspecified “something” is so diversified
that any attempt to classify such problems proves futile.
In the few examples considered below we will specify
the characteristic features of this “something” and ways
of establishing them.
Example 33.1. An object of mass m is lying on a hori-
zontal surface, with a coefficient f of friction between surface
and object. At time t = 0 a horizontal force varying accord-
ing to the law F = ta, with a a constant vector, is applied
to the object. Find the length of the path traveled by the
object after the first t seconds have elapsed.
Solution. We carry out an ordinary analysis of the
problem by applying the method of analyzing its phys-
ical content. Only the object of mass m constitutes the
physical system, and we consider this object a particle.
Several forces influence the motion of the body, and one
depends on time. We must find one of the parameters
of this motion, the path traveled by the object. This
constitutes a basic problem of particle mechanics.
256 Part 3. Nonstandard, Nonspecified, Arbitrary Problems

Let us apply Newton's second law. We link an inertial

reference frame with the horizontal surface and direct
the X axis along vector a. Four forces act on the object
during the motion: the given force F = at = iat, where
iis the unit vector pointing in the positive direction of
the X axis, the force of friction F,;, = —fmgi, the force
N by the support (the horizontal surface) on the object,
and the force of gravity mg. The last two forces compen-
sate each other. By Newton's second law,

mY —ati ~- {mgi,

or in terms of projection on the X axis,

doy
= at — fmg.
ee
Integrating this equation and allowing for the initial
conditions, we arrive at the law of variation of velocity:
t2
Ve= > fet. &

Integrating the equation dz/dt = v,, weobtain the law

of motion:
at fgt®
=a (33.1)

The last expression provides the answer to the prob-

lem. However, the solution is invalid. The physical
analysis has been carried out formally, that is, we did
not allow for static friction and, especially, for the fact
that this force, just as the given force F = at, varies
(grows) with time. The object starts moving only at
time t = fmg/a, when the static friction reaches its
maximum value. Prior to this moment the object was
at rest. Substituting into Eq. (33.1) ¢ — t) for ¢, we arrive
at the correct equation:
a(t—te)> fg (¢—t)8
i
Saar7 nena
6m
eer, eee

where t > fy.

Thus, in the given nonstandard problem the “some-
thing” was a thorough analysis of the force of friction
 Ch. 12. Nonstandard and Original Problems 257

and the fact that this force varies with time, specifically,
grows from zero to its maximum value fmg as the given
force F = at grows.
A condition very often encountered in nonstandard
problems is the separation condition; when the interaction
of objects ceases, the elastic force exerted by the support
vanishes, or N = 0.
Example 33.2. A time-dependent force F = at, where
a is constant, starts to act at time t= 0 on a small object
of mass m lying on a smooth horizontal surface. The forc-
always forms an angle a with the horizontal surface. Find
the moment in time when the object is separated from the
surface and the velocity of the object before and after sepa-
ration (lift-off).
Solution. The physical system consists of only one
body, object m. All other bodies are considered external.
The given object may be considered a particle. As a result
of the interaction with external bodies the given object
moves. Note that one of the external forces depends on
time t. We wish to find the moment when a certain event
takes place (lift-off) and the velocity of the object before
and after the event. Since the motion of the object is not
considered formally (the force is specified), the given
example is related to a basic problem of particle ine-
chanics.
Let us employ Newton's second law. We link an inertial
reference frame with the horizontal plane, direct the X
axis along the plane, and direct the Y axis vertically
upward. By Newton's second law,
m we =at cosa, (33.2)

m oy =N-4atsina— mg. (33.3)

It is easy to guess (!) that prior to lift-off v, = const = 0
and, hence, the system of equations (33.2), (33.3)
assumes the form
dvx
m = at cos a, (33.4)
dt
0=N-+atsina—mg. (33.5)
258 Part 3. Nonstandard, Nonspecified, Arbitrary Problems

We have arrived at a system of two equations in three

unknowns (v,, N, and ¢), but if we guess (!) that at the
moment of lift-off the force exerted by the support on the
object vanishes (V = 0), then Eq. (33.5) immediately
gives us that moment:
t= —s~—.
asing
(33.6)
Next, integrating Eq. (33.4) and allowing for the
initial conditions, we obtain the law of variation of
velocity prior to lift-off:
y= ee (33.7)
2

After lift-off (NV = 0) the system of equations (33.2),

(33.3) assumes the form

m —= =atcosa, (33.8)
dvx

moe =a (t— ty) sina — mg. (33.9)

In Eq. (33.9) we have allowed (1) for the’ fact that the
motion along the Y axis starts at time of lift-off 4, =
mg/a sin a. Integrating Eqs. (33.8) and (33.9) and allow-
ing for the initial conditions, we arrive at the law of
variation of the velocity after lift-off (¢ > ?,):
v = SE
at®
gos i+
.
[Speers
a (t—?ty)? sin
— g(r—4)] j,: (33.10)
where i and j are unit vectors directed along the X and Y
axes.
At first it may seem that the guesses made in the pro-
cess of solving the problem (prior to lift-off v, = 0, at
the moment of lift-off N vanishes, and after lift-off the
time it takes the object to move along the Y axis is
t — t)) are minute, inessential. Indeed, they are, but
without making them we cannot solve the problem. This
illustrates the important role played by the little “some-
thing.” After solving such problems our experience and
physical intuition become richer. Gradually these details,
or guesses, do indeed become obvious. But everything
that has been mastered seems simple and obvious, while
 Ch. 12. Nonstandard and Original Problems 259

the unknown and unsolved seems complicated and in-

comprehensible.
Another widely used tool in solving nonstandard prob-
lems is the choice of a simple reference frame (RF). Here
are three examples. In the first the choice of the reference
frame makes no difference, in the second it is so important
that one choice makes the problem standard while another
makes it nonstandard, and in the third a successful choice
of RF becomes a decisive factor (a common standard
problem becomes an original one).
Example 33.3. A chain of mass m is laid out tn a circle
of radius R and fitted onto a smooth circular cone whose
semi-vertez angle is @ (Figure 33.1). Find the tensile stress

4m 0

Figure 33.1 Figure 33.2

developed by the chain if the cone (with the chain) rotates

with a constant angular velocity w about the vertical azis,
which coincides with the cones symmetry azis.
Solution. The physical system consists of a single
object, the chain. The chain cannot be thought of as
being a particle. The sought tensile stress acts on each
element of the chain. Hence, we partition the chain into
similar elements so small that each may be assumed to
be a particle. We consider one such element (or particle)
of mass Am. It moves along a circle of known radius R
with a known angular velocity . and certain forces
17¢
260 Part 3. Nonstandard, Nonspecified, Arbitrary Problems

act on it. We must find one of these forces. This constitutes

one of the basic problems of the dynamics of rotational
motion of a particle.
Let us employ Newton’s second law. It can easily be
shown that the choice of a reference frame (inertial or
noninertial) is irrelevant; we will write the system of
equations for this problem simultaneously in two refer-
ence frames. Let us select the noninertial reference frame
(NIRF) as the one linked to element Am and the inertial
reference frame (IRF) as the one linked to any (external)
fixed object. Four forces act on Am in IRF: the force of
gravity Amg (Figure 33.1), the elastic force N exerted
by the support (cone) on the element, and two equal
(why?) tensile stresses T (Figure 33.2), each of which
is directed along the tangent to the circle at the appro-
priate point. In the NIRF there is an additional fifth
force, the centrifugal force of inertia Amw?R. Projecting
the forces on the coordinate axes, we can write Newton's
second law in the following form for the IRF:
N sin @— 27 sin\(a/2)=0 (for the Z axi8), (33.11)
2T sin (@/2) —N cos8=Amw?R (for the X axis). (33.12)
Correspondingly, for the NIRF:
N sin 0 — 27 sin (@/2) =0, (33. 13)
2T sin (a/2) — N cos 8 — Amw?R = 0. (33.14)
Obviously, the system of equations (33.11), (33.12)
is equivalent to the system (33.13), (33.14), but in either
case it constitutes an open system of two equations in
three unknowns (7, a, and N). Hence, we must find
the “something” that will close the system of equations.
Looking at Figure 33.2, we guess that angle @ is in some
way connected with the element Am. It is easy to see
(why?) that this connection has the form
Am/m = @/2nx.

Finally, if we allow for another “something,” the fact

that angle @ is small (and, therefore, sina ~ a), we
arrive at a simple, closed system of equations (in the
 Ch. 12. Nonstandard and Original Problems 264

IRF)
N sin 86 — Amg = 0,
Ta — N cos 8 = Amw?R,
Am/m
= a/2n.

After solving this system we arrive at the answer in

general form:
T=m (w*R-+ g cot 6)
=e

Note that although concrete and generalized knowledge

plays an essential part in the solution of this problem,
the role of the insignificant “something” has increased.
Example 33.4. A massless pulley is attached to the ceiling
of an elevator, then a massless string is swung over the pulley,
and the ends of the string are
tied to two loads of mass m, and m, 0
Figure (33.3). The elevator is
lifted with an acceleration a.
Ignoring friction, find the force
with which the pulley acts on the
ceiling of the elevator.
Solution. We will solve the
problem in two reference frames,
one linked with the elevator
cabin (NIRF) and the other
linked with Earth (IRF).
Solution in NIRF. The phys-
ical system incorporates the two
loads m, and m, (which can be Figure 33.3
thought of as particles), the
massless pulley, and the massless string. The objects of
the system are in accelerated motion due to the action
of certain forces. We must find the force with which
the pulley acts on the ceiling of the elevator cabin. This
constitutes a basic problem of dynamics.
Let us apply Newton’s second law to m, and m, in
the selected NIRF. We send the X axis downward (see
Figure 33.3). Three forces act on object m,: the force of
gravity m,g, the tensile stress T developed by the string,
262 Part 3. Nonstandard, Nonspecified, Arbitrary Problems

and the force of inertia —m,a. Projecting these forces

on the X axis. we obtain the expression for Newton's
second law as applied to m,:
mg + ma — T = my,b,
where b is the projection of the acceleration of m, in the
selected NIRF. Here we have assumed that m, > mg,
so that b is directed downward. Correspondingly, for
m, we have

mag + ma — T = —m,b.
We have a closed system of two equations 1n two un-
knowns (T and 6). The sought force F can be found by
wTiting Newton's second law for the pulley’s center of
mass (which is fixed in the NIRF): 27 — F = 0. Hence,

0__ 4imymeene
(g 4-4) : 29 46
(33.15)

Thus, in the selected NIRF the problem has proved to be

standard.
Solution in IRF. Newton’s second law as applied to m,
and m, yields
mg — T = may, (33.16)
mog — T = —Mmeag, (33.17)
where a, and a, are absolute values of the projection of the
accelerations a, and a, of objects m, and m2, respectively,
in the selected IRF. We have an open system of two equ-
ations in three unknowns (7, a,, and a,). Hence, we
must find the “something” that will close the system.
We guess that a, and a, are related. But how? Obvi-
ously (!),
@, =a, 2a. (33.18)
Indeed, since a, = b + a and a, = b — a, we have
a, — a, = 2a.
Solving the closed systems of equations (33.16)-(33.18),
we find that
T= 2m me (g+4)
mim,
 Ch. 12. Nonstandard and Original Problems 263

Hence,
os _4mym , (g- 4)
Peers mytm, °

which coincides with (33.15). Thus, the same problem

in the specified IRF has proved to be nonstandard.
Example 33.5. The projectile of an antiaircraft gun is
fired upward with muzzle velocity vy and explodes at the
highest point of its trajectory into n equal parts. The frag-
ments have equal velocities, Uo, directed at different polar
(0) and azimuthal (p) angles. Determine the position of an
arbitrary fragment at any moment in time.
Solution. We select an inertial reference frame, the one
linked with Earth. It is easy to see that with such a choice

Figure 33.4

the problem is standard. The physical system consists

of only one fragment, which can be thought of as being
a particle. The particle’s initial altitude hy = vj/2g,
the initial velocity u,, and the angle a = 90° — @ that
uy makes with the horizon (Figure 33.4) are known. We
must find the position of the fragment at any moment of
time t. This constitutes a basic problem of particle kine-
matics.
Let us select the plane in which an arbitrary fragment
moves (along a parabola) as the coordinate plane OXY
(Figure 33.4). Then the law of motion of the fragment in
264 Part 3. Nonstandard, Nonspecified, Arbitrary Problems

parametric form has the form

v2 ‘ gt?
r=u,cosat, y= — gg cE Mo sin at + Tz:

Obviously, this law of motion makes it possible to find

the position of the fragment at any moment in time.
Now we take the reference frame linked with the center
of mass of the projectile as the NIRF. After the explosion
the center of mass (the origin of the NIRF) moves down-
ward with an acceleration g. In this reference frame the
motion of any fragment is uniform (with a velocity wp)
and, therefore, at every moment all fragments form a
sphere of radius R = uot with the center at the origin.
We have solved the problem, and the solution is not only
simple and elegant but very graphic.
Thus, the same problem considered in a specified NIRF
has proved to be an original problem.

34. Original Problems :

It is better to speak not of standard, nonstandard, or
original problems (since we have just seen that the same
problem can be considered a standard, nonstandard, or
even original depending, for one thing, on the choice of
reference frame) but of ways of solving problems (stan-
dard, nonstandard, and original).
Trying to classify original problems is, obviously, just
as meaningless as trying to classify nonstandard problems
in general. It can only be noted that original problems
often allow for a standard, nonstandard, or original
solution. In the first case it is sufficient to employ only
concrete and generalized knowledge, in the second one
often resorts to guesses (though this plays no essential
role in the solution process), and, finally, in the third
the problem can be solved only by intuition and a wild
guess. It is problems of the third group that we can call
original. Here are some examples.
Example 34.1. Out of a uniform solid circle of radius R
we cut a circle of radius r <Q R/2 centered at a distance
a<(R—vr) from the center of the larger circle (Fig-
 Ch. 12. Nonstandard and Original Problems 265

ure 34.1a). Find the position of the center of mass of the

resulting figure.
Solution. The problem allows for a standard solution,
following from the definition of the center of mass (see
Eq. (13.2)). But the standard solution involves rather
cumbersome computations. Let us try a nonstandard, or
maybe even an original, solution. What special feature
characterizes the given physical system? As is known,

(a) (b)
Figure 34.1

circles are symmetric figures with an infinitude of sym-

metry axes (the diameters of the circles). The center of
mass of the larger circle (we denote this figure by the
Roman numeral II in Figure 34.16) lies (before the smaller
circle was cut out) at the center of the circle (point O).
The figure left after we cut out the smaller circle (we
denote this figure by the Roman numeral [) is still sym-
metric, but possesses only one axis of symmetry, O0,.
But what if we were to cut out two small circles instead
of one in such a way that the two are positioned symmetric-
ally in relation to the larger circle (Figure 34.2a). The
center of mass of the resulting figure (we denote this
figure by III) will be positioned at point O (this follows
from symmetry considerations). Now we return the
second small circle (the left one) to its place (the resulting
figure is denoted by IV in Figure 34.26). Then the pro-
blem of finding the center of mass of figure I is reduced
to finding the center of mass of the system of figures ITI
266 Part 3. Nonstandard, Nonspecified, Arbitrary Problems

and IV, whose centers of mass are known. Since the center
of mass of two objects lies on the straight line connecting
their centers of mass at a point A (Figure 34.2a) that

(b)
Figure 34.2 :

divides the distance a) between them in a ratio inversely

proportional to their masses, we have
Pa TL:
ao—rom,’ (34.4)
where zr = | AO |, m, = anr’ is the mass of figure IV,
and m, = a (nR? — 2nr?) the mass of figure III, with
a = const. From (34.1) we find the sought position of the
center of mass of figure I:

t=
r?

Note that the most important point in the second (ori-

ginal) solution of this problem was the guess about cut-
ting out an additional small circle. This guess constitutes
an element of experience, of physical intuition.
Let us make the terms of the problem more compli-
cated. Suppose that we are dealing with an asymmetric
figure, say a triangle, out of which we have cut out a
circle.
Example 34.2. Out of a triangular plate with sides a, b,
and c we cut a circle of radius r centered on a median AD
 Ch, 12. Nonstandard and Original Problems 267

of the triangle at a distance a, = | MN | from the pount M

where the medians intersect (Figure 34.3). Find the positron
of the center of muss of the resulting figure.
Solution. It is easy to see that this problem allows for
a standard solution via formula (43.2) involving cumber-
some calculations. Let us try another approach. The
method of cutting out a second circle so as to obtain
a symmetric figure does not work here because the initial

sf A

fe) c

A
6) D . B x

Figure 34.3

figure, the triangle, is not symmetric. But the idea of

combining various figures can still be used. Let us return
the small circle to its proper place (we denote this small
circle by I). Then we have a triangle OAB (we denote
it by II]) whose center of mass lies at the point Mf where
all three medians intersect. But the center of mass of
figure II (the triangular plate with the circle cut out)
lies at point Z on median AD at an unknown distance z
from point M, and we guess that figure II] may be con-
sidered as the combination of figures I and IJ. Since the
mass of each figure is known (mm, = anr?, m. = my — m,,
and m, = a [) p (p — a) (p — b) (p — cc) — ar*l, where
p = (a + 6 + c)/2), Eq. (84.1), which gives the center
of mass of figures I and II, assumes the form
xla, = m,/m,.

In general form the solution is

nr°a,
LE
V p(p—4) (p—b) (p—ce) —2r?
———————————————_
268 Part 3. Nonstandard, Nonspecified, Arbitrary Problems

Example 34.3. A uniformly charged ball has a spherical

cavity whose center lies at a distance a from the ball’s center
(Figure 34.4). Find the electric field strength at an arbitrary
point inside the cavity if the charge density is p.
Solution. The physical system consists of a uniformly
charged ball with a cavity. We need to calculate the
electric field inside the
cavity. This constitutes a
basic problem of the theory
of an electrostatic field.
Since the charge of the
ball cannot be considered
point-like, the DI method
could be employed (see Sec-
tion 6), but this method
involves time-consuming
evaluation of integrals. Let
us instead employ the com-
bination ideas discussed
above. We denote the radius
of the cavity by r and that
Figure 34.4 of the ball by R. We con-
sider a collection of three
objects uniformly charged with electricity with a charge
density p: a small ball of radiusr (denoted by I),
a large ball of radius R (denoted by III), and the ball
with the cavity (denoted by II). According to the super-
position principle, the electric field strength E; at any
point inside the big ball is equal to the vector sum of the
electric field strengths of the small ball, E,, and the ball
with the cavity, E,:
E; = E, + E,.

Hence, the sought field strength is

E, rd EK; = E,.

Suppose that an arbitrary point A inside the cavity is

positioned at a distance y from the center of the cavity
and at a distance z from the center of the ball (Figure 34.4).
Then the formula for the yield strength inside a charged
 Ch. 12. Nonstandard and Original Problems 269

ball yields
E, = py/3e,, (34.2)
E3= pz/3e,. (34.3)
Let us consider triangles AOO, and ABC. Allowing for
(34.2) and (34.3), we find that
[AB| Ey, —y_ {| AOy4
|AC| Ey | AO |
Hence, these triangles are similar, whereby
{BC| _ | AC| or ft
a £8
[O0,| ~~ | AO|’ a6’
Thus,
E, = ap/3e,.
Since E, is parallel to OO, (this follows from the similar-
ity of triangles AOO, and BAC), we finally conclude that
the electric field inside the
cavity is homogeneous.
Example 34.4. A direct
constant current of densityj
is flowing in an infinitely
long cylindrical conductor.
The conductor contains an
infinitely long cylindrical
cavity whose azis is parallel
to that of the conductor and
is a distant from it (Figure
34.5). Determine the magnet-
ic field strength at an arbi-
trary point inside the cavity.
Solution. Let us employ Figure 34.5
the method developed above.
We consider an arbitrary point inside the cavity, denot-
ed by A, that is z distant from the axis of the big cylinder
and y distant from that of the cylindrical cavity (Fig-
ure 34.5). It is expedient to consider three systems:
the current and its magnetic field inside a solid (i.e.
270 =6Part 3. Nonstandard. Nonspecified, Arbitrary Problems

having no cavity) infinitely long cylinder (we denote it

by III), the current and its magnetic field inside the cy-
linder with the cavity (we denote it by I1), and the current
and its magnetic field inside a small (but infinitely long)
cylinder whose radius is equal to that of the cavity (we
denote it by 1). Since by the superposition principle
H, = H, -+ H,,
we find that the sought magnetic field strength is

H, = H, — H,.
From the theorem on the circulation of vector H,

Ilence,

Since the angles BAC and OAO, are equal, the triangles
ABC and OAQ, are similar. Thus,

HJa==H3/x, or H,=ja/2.
It can easily be shown that H, is perpendicular to 00,;
hence, the magnetic field inside the cavity is homogene-
ous.
Note that solution of the last problem proved to be
standard because prior to the problem we solved three
almost similar problems and our experience and physical
intuition grew with each one. We are now able to for-
mulate dozens of similar problems, and they will be
standard rather than original because in the process of
solving the first three problems we found a special method
for their solution. Thus, the concepts of standard, non-
standard and original problems are very arbitrary and
relative and depend on the experience and physical intui-
tion of the person solving the particular problem. Nev-
ertheless, it is useful to classify problems as standard,
nonstandard, or original.
 Ch. 13. Nonspecified, Research, and Arbitrary Problems 274

Chapter 13
NONSPECIFIED, RESEARCH, AND ARBITRARY PROBLEMS

35. Nonspecified Problems

In Section 8 we mentioned nonspecified problems in con-
nection with the problem statement method. Here we
discuss such problems in detail.
The reader will recall that we defined a nonspecified
problem as one with an incomplete system of data for its
solution or a nonidealized one or a problem with both
features present. The solution of a nonspecified problem
begins with specifying the problem:
before solving a problem, formulate it.
However, presenting the general solution of a nonspe-
cified problem as two consecutive steps (statement and
solution) reflects only the external aspect of solving
a nonspecified problem. For a general solution one must
see not only the interrelation and permeation of these
steps but also their order of succession. The very process
of formulating the first problem (usually the simplest)
prepares its solution. The subsequent process of analyzing
the first problem prepares the conditions for making the
necessary step in solving a more complicated problem,
and so on. Here we have the inner dialectics of the process
of finding a general solution for a nonspecified problem.
Nevertheless, the most important, decisive step is to
formulate the problem. There is good reason to say that
once a problem has been properly formulated, half of
the work of solving it has been done.
Formulation of a problem (as well as solution of a spe-
cified problem) begins with the choice of a physical sys-
tem. One must establish which objects are included in
a given system and which will be considered external. Then
the physical system is analyzed, that is, first, one estab-
lishes what properties the objects in the system possess
(classical or quantum, elastic or inelastic, rigid or other-
wise, etc.) and, second, in what conditions the objects
in the system operate. As is known, the properties of the
272 Part 3. Nonstandard, Nonspecified, Arbitrary Problems

objects of a system and the conditions in which they

operate determine the physical phenomena occurring
in the system. A physics problem always emerges inside
a physical phenomenon and reflects the phenomenon.
The process of analyzing a physical system begins with
idealization of the problem: we introduce ideal objects
(particles, massless objects, point charges, etc.) and
estimate interactions and interrelations (determine what
interactions can be ignored, etc.). Idealization of a prob-
lem is carried out practically to the very end of the
solution of a nonspecified problem, and here it is import-
ant to distinguish two interconnected and interwoven
processes: simplification of the terms of the problem and
their complication. At the beginning the first process
is predominant. Starting the solution, one must introduce
as many simplifying assumptions as possible, ignore some
properties of the objects, disregard various conditions
and the like. However, as the problem gets more com-
plicated (in other words, as we ignore fewer and fewer
conditions), the importance of the second process, com-
plication, grows, although here too one may introduce
some simplifying assumptions and limitations.
After the physical system has been chosen and analyzed,
we begin to analyze the physical phenomena that may
occur in the system under certain conditions. At this
stage, too, idealization of the problem continues. Various
ideal processes are introduced and studied, further simpli-
fication of conditions occurs, possible limitations are
studied, and so on. It is at this stage, that is, during the
analysis of a physical phenomenon, that the first problem
emerges: certain data and the sought quantities are se-
lected and the terms of the problem are formulated. Now
the problem has been specified. Whether the formulation
is meaningful or not will be apparent only when the
problem has been solved in general form. Only then will
it be clear whether all the data for obtaining a numerical
answer are present. If some quantities prove to be un-
known, their values must be added to the original con-
ditions. After the solution of the first problem has been
analyzed, the terms are made more complicated, and
he second problem is formulated. This process continues.
 Ch. 13, Nonspecified, Resetrch, and Arbitrary Problems 273

Thus, as noted earlier, to each nonspecified problem there

can be assigned a whole cluster of problems of varying
difficulty.
Nonspecified problems are so diverse that it is not always
possible to apply the above scheme to their solution.
Indeed, there simply cannot be any rigid, unified scheine
of this sort because solving a nonspecified problem is a
creative act. But no matter what nonspecified problem
we may be solving, it is impossible to bypass the for-
mulating process and, hence, the idealization of the
problem. Below we give an example of a nonspecified
problem that at first glance seems to be of a very general
nature.

Example 35.1. Study the motion of two electrically

charged objects.

Solution. The terms of the problem do not state what

is known and what must be found, in short, the problem
is nonspecified. Let us start with the first step, the state-
ment of the problem. The physical system consists of the
two given objects and Earth (we assume that all physical
phenomena involving the objects occur on Earth). The
effect of all other external objects will be ignored. We
know that at this stage we must idealize the problem,
and, hence, must introduce and allow for various sim-
plifying assumptions and conditions. A number of these
conditions have already been stated. We continue the
process of simplification. For the sake of simplicity we
will assume the following:
(1) Both objects are particles with masses m, and mg.
Hence, the charges Q, and Q, are point-like.
(2) The charges Q, and Q, have the same sign.
(3) The effect produced by Earth’s electric field can be
ignored.
(4) The object of mass m, carrying charges Q, is fixed
to Earth’s surface, and the object of mass m, carrying
charge Q, is positioned right above object m, at an alti-
tude h (above the surface of Earth).
(5) Altitude h is small compared with Earth’s radius R.
This means that we can ignore all variations of the accel-
1/,_ 18-0498
274 Part 3. Nonstandard, Nonspecified, Arbitrary Problems

eration of free fall g within this altitude (i.e. g =

9.8 m/s* = const).
(6) The initial velocity of m, is zero (v9, = 0).
(7) Air drag is negligible.
Thus, we are ready to formulate the terms of the first
and simplest problem.
Problem 1. A particle of mass m, carrying a charge Qs,
is fized on the surface of Earth. Another particle, mass m,
and charge Q,, is positioned right above the first one at an
altitude h < R, with R the radius of Earth. Charges Q,
and Q, have the same sign. Determine the velocity of m,
at a distance h, from the surface of Earth if the initial
velocity of m, was zero. Air drag and Earth’s electric field
are ignored.
Solution of Problem 1. Applying the law of energy con-
servation to the closed system consisting of Earth and
particles m, and m,, in which only conservative forces
(the force of gravity and the Coulomb force) operate, we
obtain

m,gh + 210s
oeMeh + a 35 eee :
This yields the sought velocity:
2 h—h =
os V 2 eh) sean rae : (35.1)
Analysis and statement of other problems. Let us assume
that kh, <h. Then formula (35.1) assumes the form

pe V 20h— ee
2219 (35.2)
z

Analyzing this formula, we ‘can state, for example, the

following problem.
Problem 2. Remaining within the terms of Problem 1,
find the magnitude of charge Q. at which the velocity of m,
at altitude h, is zero.
Solution of Problem 2. Formula (35.2) immediately
yields the solution to Problem 2:
__ 4meggmhyh
Qa Q1 :
 Ch. 13, Nonspecified, Research, and Arbitrary Problems 275

Performing the necessary calculations (assuming that

m, = 10 kg, kh; = 10cm. h = 10 m, and Q, = 10-8 C),
we obtain Q, = 10-* C. The electric field strength gener-
ated by such a charge at altitudes k, andh can be found
from the following formulas:
»_ _ Qe ~_ _ Qe
Ea = Greg #2 amegh*
Substituting the numerical values yields FE, ~ 9 x 10°V m
and E} ~ 9 x 10* V/m, which are, indeed, much higher
than the electric field strength of Earth, roughly 130 Vm.
We can now formulate the following problem.
Problem 3. What will happen to object m, if its velocity
vanishes at altitude h,? At what altitude h, will object m,
be in equilibrium and what will be the
Q
nature of the object's oscillations if it m
is disturbed from equilibrium? l
We can then formulate hundreds of
problems by lifting or varying the l
conditions just stated. All these prob-
lems, however, are only particular |
cases of the following |
Generalized problem. An object of |
mass m carrying a charge Q is moving R
un an arbitrary electric field and an
arbitrary gravitational field. Determine Figure 35.1
the nature of its motion.
It is important to note that theoretically this gener-
alized problem can be solved by both the dynamical
method and the conservation-law method. This means
that all the particular problems as well can be solved by
these methods. Suppose that we have the following
Particular problem. £lectrically charged drops of mercury
fall from an altitude h into a spherical metal vessel of
radius R in the upper part of which there is a small opening.
The mass of each drop is m and the charge on the drop is Q
(Figure 35.1). What will be the number n of the last drop
that can still enter the sphere?
Solution of the particular problem. One can easily see
that the given problem is a particular case of Problems 2
and 3. Each drop of mercury reaching the vessel increases
18°
276 =©Part 3. Nonstandard, Nonspecified, Arbitrary Problems

the charge of the vessel by Q. This charge is evenly dis-

tributed over the outer surface of the sphere and generates
an electric field. As is known, the electric field of a uni-
formly charged sphere is equivalent to the field created
by a point charge of the same magnitude but positioned
at the center of the sphere. Thus, Q. = nQ, and Q, is
the charge of the (n -+ 1)st drop, which is in a state of
eguilibrium at an altitude 2 above the surface of the
Earth. The fact that the sum of the force of gravity mg
on the (n + 1)st drop and the Coulomb force with which
charge Q. acts on the charge Q of the (n -- 1)st drop is
zero implies the validity of the following equation for n:
nQ? ="
Gxe,(h— Rp 8:
Here, in accordance with item (3) in the simplifying
assumptions applied to the terms of Problem 1, we ignored
the electric field of Earth. Hence, a meaningful formu-
lation of the mercury-drop problem requires such values
of R, h, m, and Q that satisfy the given assumption (and
other assumptions as well).

36. Research Problems

Although the problems discussed above belong to the
category of nonspecified and nonidealized problems, the
“crux” of the problems is not clearly evident. Let us
consider a problem (we will call it Example 36.1) that
not only is nonspecified and nonidealized but contains
the idea of the problem. In analyzing this problem we will
see that it is a particular case of the nonspecilied problem
considered in Example 35.1 (the motion of a charged
object in an electric field and a gravitational field), but,
in contrast to the generalized problem in Example 35.1,
we will formulate a concrete problem.
An approach to the statement of the problem. It is
well-known that a gas belonging to the atmosphere of
a planet escapes the planet’s gravitational pull. Over
a certain period of time the planet may lose its entire
atmosphere (as was the case with the Moon, for exainple).
This phenomenon can be explained by the fact that the
 Ch. 13. Nonspecified, Research, and Arbitrary Problems 277

molecules in the atmosphere are distributed in velocity

according to the Maxwell law (precisely, the Maxwell-
Boltzmann law). The result is that in the atmosphere
there are always molecules whose velocity exceeds the
escape velocity for the planet. Such molecules may over-
come the gravitational pull of the planet and leave the
planet's atmosphere for ever. In place of the molecules
that have escaped from the gravitational pull there come
other molecules whose velocity exceeds the escape velocity,
and they too will Jeave the atmosphere. If the gravitation-
al pull is not very great, the process will continue until
the planet loses its entire atmosphere.
How easily molecules acquire the escape velocity. The
line of reasoning went like this: the escape velocity,
escape from gravitational pull, the force of gravity. But
other fields, too, may be used to overcome the gravi-
tational field, say the electric field. It is well-known that
Earth has an electric field whose strength at Earth’s
surface is about 130 V/m. Can this field be used to impart
to a charged object the escape velocity?
It can be calculated that if an electron passes a poten-
tial difference of only one volt, its velocity will become
approximately 600 km/s, which considerably exceeds the
escape velocity for Earth (roughly 11.2 km/s). A proton
after passing a potential difference of 100 V acquires
a velocity of 14 km/s, which also exceeds the escape
velocity. Thus, even protons accelerated in this manner
can easily leave Earth.
Statement of problem. Study the possibility of using
an electric field for launching a spaceship.
Approach to the statement of a research problem. For
the planet we take a hypothetical planet with Earth’s
parameters. We assume that Earth is a ball of radius
R ~ 6400 km and mass M = 6 x 10° kg. Let us make
some estimates, that is, numerical calculations of the
order of magnitude of certain quantities. We start with
the electric charge Q and potential » of Earth:
Q= E4ne,R?, Q 5.9 x 105 C,;
8
¢=—TneR’ es 8.3 x ‘
yp ~ 108 V.
278 Part 3. Nonstandard, Nonspectfied, Arbitrary Problems

We know (this can be verified by calculations) that

a proton, with a charge Q@, = 1.6 x 10-!®C and a mass
mp ~ 1.67 x 10-*7 kg, moving in the electric field gener-
ated by Earth’s charge, may easily acquire a velocity
exceeding the escape velocity for Earth and fly off into
outer space. We pose the following question: what must
the maximal mass m be of an object carrying an electric
charge equal to that of the proton and moving in the
electric field of Earth so that the object may escape Earth's
gravitational pull and fly off into outer space? Let us
assume that the object, a particle, is launched from
Earth's surface with an initial velocity v, = 0. Then,
according to the law of energy conservation,

—G BE 00n_ 0,
4ne)R
(36.4)
with G = 6.67 x 10-"' N-m?/kg? the universal gravi-
tational constant. This yields the following formula for
the maximal mass of the object:

m= QQp m = 2.1 x 10-8 kg,

4ne,GM ’

which is approximately the mass of a dust particle. Thus,

a dust particle with a mass 2.4 x 10-'8 kg carrying
a charge Q, = 1.6 x 10-'® C may escape from Earth’s
gravitational pull. But a dust particle may carry a charge
much greater than the proton charge Q». Two questions
emerge in this connection: what can the maximal charge
Qmax carried by a dust particle (an object) be and how
can such a charge be imparted to the object (the dust
particle)? The answer to both questions may be obtained
if we imagine the object (the dust particle) to be a small
metal ball of radius r that acquires its charge from Earth,
with which it is in direct contact. The electric charge
flows to the ball from Earth until the potentials of both
object and Earth become equal, and then the flow stops.
Since the capacitance of the ball is C = 4negr and C =
Qmax'P, We obtain
4neorQ
Qmiax = 4€ yr = Te,R = + 7
 Ch. 13. Nonspecified, Research, and Arbitrary Problems 279

where @ is the potential of Earth. Let o be the density of

the metal ball. Then its mass is

mar atmr3p. (36.2)

Substituting the values of the maximal charge of the

ball, Qmax» and mass of the ball, m, into (36.1), we arrive
at an equation for determining the radius r of a ball (or
density o) that can escape from Earth’s gravitational
pull and fly off into outer space:
M (4/3) ar99 |_QrQ___
—G "3.- Ro oe qneyk?
Bake 0.
Whence
_@Vae
r= 4> V aur (36.3)

We assume that the density p of the ball is 10° kg/in®

(the ball may be hollow). Carrying out the calculations
in the SI system of units, we obtainr + 1.8 x 10-2? m =
1.8 cm. According to formula (36.2), the mass of such
a ball is about seventeen grams.
Statement of a research problem. Thus, the electric
field of Earth may be used to launch a miniature space-
ship of radius r ~ 1.8 cm and mass of about 17 g with a
zero launching velocity. For real spaceships to be launched
we must change (increase) the electric field of Earth.
Suppose that now the radius of the spaceship is r, ~
180 cm = 1.8 m. We can now formulate the terms of the
first problem.
Problem 1. What must the electric charge Q, of Earth
be so that a spherical spaceship of radius r, = 1.8 m and
density 9 = 10° kg/m® can be launched with zero launch
velocity if it acquires a maximal charge Qma, from Earth?
Air drag is ignored.

Solution. The solution is easily obtained from for-

mula (36.3):

Q,=4nr, YiMR 9, 5.9.x 10°C.

280 3 =Part 3. Nonstandard, Nonspecitied, Arbitrary Problems

The surface electric field strength of Earth will then be

E, ~ 13 000 V/m and the potential will be @,; ~ 8.2 x
10° V. Formula (36.2) then yields the mass of the space-
ship: m, ~ 17 x 10* kg. A similar problem can also be
stated for the Moon.
Let us now consider another example (we will call it
Example 36.2).
Approach to the statement of a problem. As is known,
in the state of weightlessness in outer space many phys-
ical phenomena proceed in a manner different from that

Figure 36.4 Figure 36.2

on Earth. Let us take the oscillations of a simple pen-

dulum.
If on Earth the pendulum is placed in a vacuum, it
oscillates under the force of gravity mg and the tensile
stress developed by the string, F, (Figure 36.1). The
period of these oscillations is

T,=2n V Ug, (36.4)

where / is the length of the pendulum.
If the pendulum is placed in a nonviscous medium (an
ideal fluid), an additional force appears, the buoyancy
force F;, acting against the force of gravity mg (Fig-
 Ch. 18. Nonspecified, Research, and Arbitrary Problems 281

ure 36.2). The period of oscillations in this case is

Tata Ve
l

and it becomes infinitely large at F, = mg. If this

happens, the tensile stress developed by the string van-

Figure 36.3

ishes (the force of gravity is compensated for by the

buoyancy force).
Suppose that an electric charge Q is imparted to the
pendulum, which is then placed in a homogeneous elec-
tric field E, as shown in Figure 36.3. The pendulum
one ete and its period is given by the following for-
mula:
i T
T= 20 V TOE
In the state of weightlessness (the pendulum is placed
inside an elevator cabin that falls freely with an accel-
eration g) a force of inertia acts on the pendulum. This
force F, is equal in magnitude to the force of gravity mg
but is opposite in direction (Figure 36.4). The stress F,
developed by the string vanishes and the pendulum does
not oscillate (the force of gravity is compensated for by
19-0498
282 Part 3. Nonstandard, Nonspectfied, Arbitrary Problems

the force of inertia). If a charge is imparted to the pen-

dulum and the pendulum is placed in a homogeneous
electric field, oscillations
S can occur even in weight-
lessness.
Statement of the problem.
Study the possibility of
using an electric field as
F.=-mg the “medium” in which a
pendulum clock can op-
erate in weightlessness.
Statement of a research
problem. The physical sys-
mg a-g tem consists of a simple
pendulum of mass m and
. length lJ carrying a charge Q
Figure 36.4 and two fields” the gravi-
tational field and a homo-
geneous electric field E. A force of inertia F; = —m
acts on the pendulum. To compensate this force the
electric field strength must be such that
mg = QE. (36.5)
Then the period of oscillations can be calculated by for-
mula (36.4), that is, the pendulum will oscillate with
the same frequency as on Earth under ordinary conditions.
Let us imagine the bob of the pendulum to be a small
ball of radius r and density p. The charge carried by the
bob must be so smal] that its field can be ignored in com-
parison with the external electric field E. The latter can
be generated inside a parallel-plate capacitor with the
distance between the plates designated by d. Suppose
that the ball oscillates at a distance d/2 from each plate.
Then the condition that the field of the charge Q on the
ball be weak can be written as follows:
9; = 10-*9, (36.6)
where g, = Q/4ne,r is the potential of the ball, and
gq = Ed/2 the potential of the external field at the point
occupied by the ball. Condition (36.6) then assumes the
 Ch. 13. Nonspecified, Research, and Arbitrary Problems 283

form

Grey OES
4 d

or

Q = Steud 419-2, (36.7)

Substituting this into (36.5) and allowing for (36.2), we
obtain

E=10r 2pg *
Seed (36.8)

Calculating the value of E for r = 10-3 m, p = 10®kg/m!,

and d = 2 x 10-! m, we obtain E =6 x 10° V/m.
This leads us to the following value of the potential
difference across the plates: Ag = Ed = 1.2 x 10° V.
Apparently, it is difficult to generate such a field inside
a spaceship. Therefore, we will calculate the field strength
for a miniature pendulum: r = 10-5 m, p = 10° kg/m’,
and d = 2 x 10-? m. Then formula (36.8) yields E ~
2 x 10' V’m. The potential difference Ag ~ 400 V.
Making the radius r of the ball still smaller, we can obtain
practically realizable values of the electric field strength
E and the potential difference Ag.
Thus, a pendulum clock operating in an electric “me-
dium” in weightlessness must necessarily be minute. Let
us formulate the terms of
Problem 1. Jn a jreely falling elevator cabin (on Earth)
there is a parallel plate capacitor the distance between
whose plates is d = 2 X 10-* m. A simple pendulum of
length d/2 is attached to the upper plate of the capacitor.
The bob of the pendulum has the shape of a ball of radius
r = 10-° m and density p = 10° kg/m®. What potential
difference Aq must be applied across the capacitor plates
and what charge Q must be imparted to the bob if we want
the pendulum to oscillate with the same frequency as it
would in a motionless elevator on Earth? The electric field
generated by charge Q carried by the bob must be weak com-
pared to the electric field between the plates. Air drag on the
pendulum is to be ignored.
19*
284 Part 3. Nonstandard, Nonspecified, Arbitrary Problems

We already know the solution to this problem: it is

given by formulas (36.8) and (36.7).
We can formulate other problems, say, that of con-
sidering the oscillations of a charged physical pendulum
in a homogeneous electric field.

37. Arbitrary Problems

So far we have solved problems that dealt with a known
topic. We usually started with the theory of the topic,
then considered the basic problem, employed certain
methods for its solution, suggested other methods for its
solution and for solution of similar problems belonging
to the same topic. Note that the solution of these problems
was also carried out on the basis of a general approach
developed in Part 1. In this way we moved from topic
to topic, gradually acquiring experience in problem solv-
ing. This is a necessary path, but it is not at all suffi-
cient. In life one is confronted more often with problems
belonging not to one (known) topic but involving several
topics, or what we call arbitrary problems.
The theory, methods, and tricks of solving physics
problems already discussed are aimed at teaching the
reader to solve problems whose topic is not known in
advance. Now we will apply this approach to such pro-
blems.
Example 37.1. In isothermal-expansion of one mole of
oxygen with a temperature T = 300 K the gas absorbed
an amount of heat Q = 2kJ. By what factor did the volume
occupied by the gas increase?
Solution. The problem is not stated quite right because
the initial pressure of the gas is not given. Depending
on the pressure, the gas may be considered either ideal
or real. Two solutions are possible, therefore: one for
a physical system consisting of an ideal gas, the other
for a system consisting of a real gas.
Suppose that we are dealing with an ideal gas. Then,
as a result of an isothermal process the amount of heat Q
is absorbed and the amount of work
 Ch. 13. Nonspecified, Research, and Arbitrary Problems 285

is performed. We wish to find the ratio of two values of

a macro-parameter of the system, its volume (the volume
ratio V,/V,). This constitutes a basic problem of thermo-
dynamics. So we employ the thermodynamic method.
Applying the first law of thermodynamics, we find that
Q= RT |n (V,/V;),
where we have allowed for the fact that the variation of
the internal energy of an ideal gas is zero, AU = 0.
Hence,

pe
Vv
=eORT, VaIV,, 2 23.
Now let us assume that the physical system consists
of a real gas that obeys the van der Waals equation

(ptr) (¥-%)=
where a and b are the van der Waals constants. The inter-
nal energy of one mole of a real gas,
U=cyT —alV,

depends not only on temperature T but also on volume V.

By the first law of thermodynamics,
G=—-aS
a +AT Int==
Since for oxygen b =008) ce, the last equation
can be approximately written as
Q=a aon
FAT Ing,
with z = V,/V,. To te sits transcendental equation,
we must specify the final volume of the system, V5.

Example 37.2. A particle ts moving in the positive

direction along the X azis in such a manner that its velocity
varies according to the law v =a) x, with a a positive
constant. Assuming that at time t = 0 the particle was at
point x = 0, find (a) the time dependence of the particle's
velocity and acceleration, and (b) the average velocity of the
particle over the time interval that it takes the particle to
travél the distance from point x = 0 to point x.
286 Part 3. Nonstandard, Nonspecified, Arbitrary Problems

Solution. The physical system consists of a single object,

the particle. which moves in a straight line (the one-
dimensional case) along the X axis, and this motion is
considered formally. An interrelationship between some
of the parameters of motion (the velocity v and the po-
sition coordinate z) is fixed. We must find certain other
parameters of motion as functions of time and the average
velocity. This constitutes an inverse problem of kine-
matics.
Let us find the law of motion of the particle. Since v, =
dz/dt, we can write the interrelationship between velocity
and the position ee v, =a Vz, in the form

A ae
= Ve
Zz.
Separating variables, integrating, and allowing for the
initial conditions, we arrive at the law of motion:
z=a"t?/4. (37.1)
This leads us to the following laws of time-variation of
velocity,
dz a’t
Ysa 3
time-variation of acceleration

dv a*
nr ra
and time-variation of the average velocity,
t t
1 1 a*t a*t
w=
Dy
| v,dt= aa 5 dt= zl (37.2)
i) 0

Substituting the time of motion ¢ = 2 Yz/a, which can

be found from (37.1), into (37.2), we find the final expres-
sion for the average velocity:

v=5Vz.
Analysis of the solution shows that the problem can be
made more complicated by using instead of the equation
 Ch. 13, Nonspecified, Research, and Arbitrary Problems 287

v; =a Vx a more general equation v, = f (x), where

function f(z) must satisfy several conditions.
Example 37.3. A piston of mass M can move without
friction inside a vertical cylindrical pipe whose lower end
is closed. The volume under the piston is filled with a gas
whose mass is considerably smaller than that of the piston.
In equilibrium the distance between the lower end of the
piston and the bottom of the pipe is l, (Figure 37.1). Find
the period of the small oscillations
that the piston will make if the
equilibrium is disturbed, assuming
that oscillations occur isothermally
and the gas is ideal. The cross-sec-
tional area of the pipe is A and the
standard atmospheric pressure is Po.
Consider the limiting case when
Po =9
Solution. Three objects will be
included in the physical system:
the piston (a rigid body of mass M), :
the gas under the piston, and the Figure 37.1
air above the piston (an ideal gas).
How will the objects of the system behave when the
piston is moved downward by a distance z from the
position of equilibrium (see Figure 37.1)? The state of
the gas above the piston does not change (its pressure Do
and temperature 7, remain unchanged), while the state
of the gas under the piston changes (the temperature 7)
remains unchanged, the volume decreases by AV = 2A,
and the pressure increases by Ap). Hence, the piston
is under an additional force F = ApA directed upward.
This force pushes the piston upward. In equilibrium
the force F vanishes. But since the velocity of the piston
at this point is nonzero, the piston moves past the equi-
librium position. The distance it travels upward from
the equilibrium position is also z since there is no friction.
The pressure of the gas under the piston decreases by Ap.
Under the force ApA, now directed downward. the piston
moves downward. In this way the piston oscillates about
the equilibrium position.
 288 Part 3. Nonstandard, Nonspecified, Arbiirary Problems

Let us employ the dynamical method. First we find

the additional force F = ApA acting on the piston. By
the Boyle law for an isothermal process involving the gas
under the piston,

PylyA = (py + Ap) (Aly — Az), (37.3)

where p, = (Mg + p,A)/A is the pressure of the gas
under the piston in the state of equilibrium. Assuming
that z < 1,, we find from (37.3) the variation of the gas
pressure:
Ap= Pytlly.
Hence, the additional force
F=— Mg+ pA zr
ly
is proportional to the displacement z of the piston from
the equilibrium position and is directed toward the equi-
librium position.
Under this force the piston oscillates harmonically. By
Newton’s second law we can find the differential equation
describing these oscillations:
Mr + teh Pt 2=0.
By comparing this equation with the general differential
equation for free undamped oscillations we can find the
period of the piston’s oscillations:
Ml,
T,=2n “Metcae
pAA‘
Hence, in the limiting case when py = 0 we have 7) =
2x V1,/g, which is the period of oscillations of a simple
pendulum of length Jp.
Example 37.4. A bullet passing through a wooden board
of thickness h changes velocity from vy to v,. Find the time
that the bullet spent in passing through the board if the
resistance force is proportional to the square of velocity.
Solution. A particle (the bullet) changes its state of
motion because of a known force. The initial conditions
are known: vo = {vo, 0, 0} and ry = {0, 0, O} att = 0.
 Ch. 13. Nonspecified, Research, and Arbitrary Problems 289

We wish to find one parameter of motion, time ¢. This

constitutes a basic problem of particle dynamics.
Let us employ the dynamical method. By Newton's
second law,
dv
m— = —av’, (37.4)
“dt
where m is the bullet’s mass, and a is a proportionality
factor. Note that both parameters (m and a) are unknown.
Integrating Eq. (37.4) and allowing for the initial con-
dition, we arrive at the law of time-variation of the
bullet’s velocity:
v= ——"o_
1+ avot/m *

Putting v = v, in this equation, we obtain the sought time

interval:
fics ESavyail (37.5)
To determine the unknown ratio m/a we find the law of
motion of the bullet by solving the inverse problem of
kinematics:
z= In (1+2),
or
h=2 In (1+ wot), (37.6)
Expressing the ratio m/a by employing Eq. (37.5),
Ts, ae Uyty

a 1—v4/v9 ,

and substituting it into (37.6), we arrive at the final

expression for the sought time interval:
h= h(vg—v))
; Ygv1 IN (v9/v) *
Example 37.5. Two square plates with sides a = 300 mm
are fixed at a distance d = 2.00 mm from each other and
form a parallel-plate capacitor. They are connected to a de
source of voltage Ag = 250 V. The plates are positioned
290 Part 3. Nonstandard, Nonspecified, Arbitrary Problems

vertically and lowered into kerosene with a rate v =

5.00 mm/s. Find the current I that flows in the leads during
immersion.
Solution. The physical system consists of the parallel-
plate capacitor connected to the dc voltage source. Prior
to immersion, the charge on one plate was
Q = CAg,

where C = ¢€,a7/d is the capacitance of the capacitor.

When the plates are being immersed, an electric current
flows in the circuit. Why? Kerosene is an insulator (the
dielectric constant ¢ = 2). The appearance of kerosene
between the plates changes the electric field in the ca-
pacitor. This leads to a redistribution of charges on the
plates of the capacitor and current begins to flow in the
leads. The build-up of charge on the plates at a constant
voltage (Ag = const) is caused by the growth in the
<apacitor’s capacitance C
Let us find C at an arbitrary moment in time. By this
time the plates will be immersed to a depth h = vt.
The capacitor may be imagined as conSisting of two par-
allel-plate capacitors connected in-parallel, one with
an insulator between the plates and the other with air
({e =~ 1). The capacitance of this system is
e,evta & (a—vt)a &9a
C= te ile — 1) vt +a.
The charge Q on a capacitor plate changes in time ¢
according to the law
Q = SAP [(e—1) vt + a].
This gives us the following formula for the sought current:
fa Be = Be AO Ay:
Substituting of numerical values yields J = 1.7 X 10-°A.
Example 37.6. A spool of thread lies on a horizontal plane.
With what acceleration a will the azis of the spool move
if the free end of the thread is pulled with a force F (Fig-
ure 37.2). How should the thread be pulled so that the spool
 Ch. 13. Nonspecified, Research, and Arbitrary Problems 291

will move in the direction of the tense thread? The spool

moves along the surface of the horizontal plane without
slippage. Find the force of friction between spool and hori-
zontal plane.
Solution. The physical system consists of a single
object, the spool, which we consider a rigid body. The

Figure 37.2

forces acting on the spool are known (they can be deter-

mined). We wish to find the acceleration of the spvol.
This constitutes a basic problem of rigid-body dynamics.
The following forces act on the spool: the given tensile
stress F developed by the thread, the force of gravity mg,
the force of friction F;,, and the force N exerted by the
support on the spool. We link the inertial reference frame
with Earth and direct the coordinate axes as shown in
Figure 37.2. By the theorem on the motion of the center
of mass,
ma=Fcosa—F;,, 0=N+Fsina—mg.
From the equation of motion of a rigid body about an
axis passing through the center of mass we find that
JB =F,,R—Fr,
292 Part 3. Nonstandard, Nonspecified, Arbitrary Problems

where Bp = a/R is the angular acceleration of the spool,

and J the spool’s moment of inertia about the rotation
axis. Solving the obtained system of equations, we find
the sought acceleration,
ae RF (Ros a—r)
. J+mR? ’
(37.7)
and the force of friction,
F,, = F cosa — ma.
From Eq. (37.7) it follows that condition a > 0 is met if
cos @ >r/R. To solve the problem numerically we must
know the mass m and moment of inertia J of the spool.
Example 37.7. A planoconvez glass lens touches a glass
plate with its convex surface (Figure 37.3). The radius of

Figure 37.3

curvature of the conver surface of the lens is R, and the

wavelength of the light falling on the lens is X. Find the
width Ar of a Newton ring as a function of the ring’s radius
in the region where Ar <r.
Solution. The following objects may be included in the
physical system: the glass plate, the lens, and the thin
air wedge between the lens and the plate (see Figure 37.3).
As a result of reflection of waves from the upper and lower
surfaces of the air wedge an interference pattern is formed
and Newton rings appear. We wish to find the width of
a ring Ar or, which is the same, find the distance between
 Ch. 13. Nonspectfied, Research, and Arbitrary Problems 293

the centers of the neighboring dark and bright rings.

This constitutes a basic problem in the theory of inter-
ference of waves.
First we must find the optical path difference and then
employ the maxima and minima conditions. The optical
path difference of the rays reflected from the upper and
lower surfaces of the air wedge is
§=2d+2/2, (37.8)
where d is the thickness of the air wedge at the point
where the rays are reflected. From geometrical considera-
tions it follows that
dx 2R=r}, (37.9)
In this equation r, is the radius of the Ath dark or bright
ring. Substituting the value of d from Eq. (37.9) into
Eq. (37.8) and employing the maxima condition (27.1)
and the minima condition (27.2), we find the radii of the
kth dark ring,
Tra = V kAR ’

and the kth bright ring,

Trp = V (x —+) R.

These expressions yield

Tad —Thb = (Tra — Tab) (Tra + Tab) = RA/2.
Putting rig + Tap © 2r, we find the width of a ring:
Ri
Ar=Trpa—Try © FZ -
Example 37.8. An ebonite ball of radius R = 50 mm
is charged electrically by friction with a uniformly distri-
buted surface charge of density o = 10.0 wC/m?. The ball
is rotated about its axis with a velocity v = 600 rotations
per minute. Find the magnetic induction B generated
at the center of the ball.
Solution. The surface charges move in circles as the
ball rotates. This generates circular currents, and around
each current there is a magnetic field. We must find the
294 Part 3. Nonstandard, Nonspecified, Arbitrary Problems

total magnetic induction of these fields at the center of

the ball. This constitutes a basic problem in the theory
of magnetic fields.
Let us employ the superposition principle and DI
method (see Section 6). In view of the symmetry of the
problem, we select a spherical system of coordinates and
place its origin at the center of the ball. With planes

Figure 37.4

perpendicular to the rotation axis we partition the ball’s

surface into spherical layers so narrow that the magnetic
field of the current generated by the motion of the charge
carried by this layer can be calculated using the Biot-
Savart law (23.3). Let us take an infinitely small surface
element belonging to such a layer (in Figure 37.4 this
element is hatched). The area of this surface element is
dA = R*sin 6 d6 dq, and the electric charge carried
by it is d@=odA =oR?*sin@d6dqg. Since dg =
w dé, the circular current generated by the motion of the
charges on this layer is

1 = 82 — wR? sin 640. (37.10)

As is known (this result can easily be obtained), a cir-
cular current J of radius r generates a magnetic field whose
induction B at a point lying on the axis of this current
 Ch. 13, Nonspecified, Research, and Arbitrary Problems 295

at a distance d from the current’s plane can be calculated

by the formula
{r?
B= Wot apa -
Thus, the circular current (37.10) generates a magnetic
field whose induction at the ball’s center is
dB = pop RS O08 (37.44)
Integrating (37.11) with respect to 6 from 0 to x, we find
that

B= ooR sin? 6 dé 2h ,nhooR

= Hot 3 _—— = =3
|
ee

Allowing for the fact that wm = 2nv, we arrive at the

final expression for the sought induction
B = 4nyp,.povR/3.
Substituting the numerical values, we get B ~ 2.6 xX
10-1! T.
CONCLUSION

It is time to summarize. We have used many examples to

illustrate the simple fact that the general approach to
the solution of any problem in the course of college phys-
ics amounts basically to the ability to analyze an arbi-
trary physical phenomenon or a collection of phenomena.
The fundamental concept of a physical phenomenon is
linked to the majority of generalized concepts of physics:
a physical system, a physical quantity, a physical law
and its main aspects (the physical meaning, conditions
in which the law is valid, the method of application),
an interaction, the state of a physical system, a basic
problem, idealized physical objects and processes, a phys-
ical model, etc. It is important not only to know all
these concepts but to know how to manipulate and use
them as elements of the various methods. In this system
the two most general methods are the most important, the
method of analyzing the physical content of a problem
and the problem statement method. The first makes it
possible to solve any formulated problem in the college
physics course, while the second helps not only to find
an approach to the solution of a nonspecified problem and
to formulate and solve the “first” problem but, also, via
simplification and complication, to formulate and solve
dozens of problems of varying degrees of difficulty, that
is, to consider a “cluster” of problems.
Above we noted that formulated and nonspecified prob-
lems exhaust the entire spectrum of problems. So it
would seem that we have the answer to the question as
to how to learn to solve physics problems. Basically one
must master these two methods. How simple! But to
do so requires a lot of hard work. In study as in science
there is no easy way to achieve something.