A Textbook of Quantum Mechanics 2nd Edition P. M. Mathew Download
A Textbook of Quantum Mechanics 2nd Edition P. M. Mathew Download
M. Mathew download
https://ebookname.com/product/a-textbook-of-quantum-
mechanics-2nd-edition-p-m-mathew/
Get the full ebook with Bonus Features for a Better Reading Experience on ebookname.com
Instant digital products (PDF, ePub, MOBI) available
Download now and explore formats that suit you...
https://ebookname.com/product/lectures-on-quantum-
mechanics-59465th-edition-paul-a-m-dirac/
https://ebookname.com/product/quantum-mechanics-6th-edition-
alastair-i-m-rae/
https://ebookname.com/product/quantum-information-theory-and-the-
foundations-of-quantum-mechanics-2nd-edition-christopher-g-
timpson/
https://ebookname.com/product/atlas-of-small-animal-wound-
management-and-reconstructive-surgery-3rd-ed-edition-michael-m-
pavletic/
Reflective practice writing and professional
development 3rd ed Edition Gillie Bolton
https://ebookname.com/product/reflective-practice-writing-and-
professional-development-3rd-ed-edition-gillie-bolton/
https://ebookname.com/product/jstl-practical-guide-for-jsp-
programmers-1st-edition-sue-spielman/
https://ebookname.com/product/oil-gas-and-energy-financing-1st-
edition-mr-howard-palmer/
https://ebookname.com/product/intentional-interviewing-and-
counseling-facilitating-client-development-in-a-multicultural-
society-7th-edition-allen-e-ivey/
https://ebookname.com/product/the-rough-guide-to-
iceland-2013-5th-edition-david-leffman/
The Struts Framework Practical Guide for Java
Programmers 1st Edition Sue Spielman
https://ebookname.com/product/the-struts-framework-practical-
guide-for-java-programmers-1st-edition-sue-spielman/
A Textbook of
Quantum Mechanics
Second Edition
About the Authors
P M Mathews held the position of Professor, Senior Professor and Head of the
Department of Theoretical Physics for nearly three decades, and is now retired. He
has been Visiting Professor and Visiting Scientist several times at numerous distin-
guished institutions (MIT, Cambridge, MA, USA; Harvard-Smithsonian Center for
Astrophysics, Cambridge, MA, USA; Royal Observatory of Belgium at Brussels,
Paris Observatory, Paris, France). He is a fellow of the Indian National Science Acad-
emy, of the Indian Academy of Sciences, and of the American Geophysical Union.
His research interests are in Classical and Quantum Theory of Relativistic Fields,
and Geoastronomy. He has published about 120 papers in international journals.
Dr Mathews is a recipient of the Meghnad Saha Award for Theoretical Sciences.
P M Mathews
Retired Senior Professor and Head of the Department of Theoretical Physics
University of Madras
K Venkatesan
Formerly Associated with
Institute of Mathematical Sciences, Chennai
McGraw-Hill Offices
New Delhi New York St Louis San Francisco Auckland Bogotá Caracas
Kuala Lumpur Lisbon London Madrid Mexico City Milan Montreal
Paris San Juan Santiago Singapore Sydney Tokyo Toronto
Tata McGraw-Hill
Published by the Tata McGraw Hill Education Private Limited,
7 West Patel Nagar, New Delhi 110 008.
A Textbook of Quantum Mechanics
Copyright © 2010, 1976 by Tata McGraw Hill Education Private Limited. No part of this pub-
lication may be reproduced or distributed in any form or by any means, electronic, mechanical,
photocopying, recording, or otherwise or stored in a database or retrieval system without the
prior written permission of the publishers. The program listings (if any) may be entered, stored
and executed in a computer system, but they may not be reproduced for publication.
This edition can be exported from India only by the publishers,
Tata McGraw Hill Education Private Limited
ISBN-13: 978-0-07-0-146174
ISBN-10: 0-07-0-146179
Managing Director: Ajay Shukla
Head—Higher Education Publishing: Vibha Mahajan
Manager—Sponsoring: Shalini Jha
Senior Editorial Researcher: Smruti Snigdha
Development Editor: Renu Upadhyay
Jr Manager—Production: Anjali Razdan
General Manager: Marketing—Higher Education: Michael J Cruz
Dy Marketing Manager—SEM &Tech Ed: Biju Ganesan
Asst Product Manager—SEM &Tech Ed: Amit Paranjpe
General Manager—Production: Rajender P Ghansela
Asst General Manager—Production: B L Dogra
Information contained in this work has been obtained by Tata McGraw-Hill, from sources
believed to be reliable. However, neither Tata McGraw-Hill nor its authors guarantee the
accuracy or completeness of any information published herein, and neither Tata McGraw-
Hill nor its authors shall be responsible for any errors, omissions, or damages arising out of
use of this information. This work is published with the understanding that Tata McGraw-
Hill and its authors are supplying information but are not attempting to render engineering
or other professional services. If such services are required, the assistance of an appropriate
professional should be sought.
Typeset at ACEPRO India Private Limited, Chennai and Printed at Adarsh Printers, C-50-51,
Mohan Park, Naveen Shahdara, Delhi-110032.
Cover Printer: SDR Printers
Contents
Appendices
A. Classical Mechanics 425
B. Relativistic Mechanics 429
xii Contents
The primary motivation for the preparation of a new edition of the book was to
extend its coverage to include a couple of quantum phenomena which are currently
of considerable interest, but were not known at the time of publication of the first edi-
tion: the Integer Quantum Hall Effect (IQHE) and the Aharonov–Bohm effect. The
former effect is the appearance of quantum jumps in the Hall conductivity of a mate-
rial under suitable conditions, as the strength of a uniform magnetic field in which
it is placed crosses certain discrete values. In order to bring out the physics of this
phenomenon, we have gone further, in this edition, with the solutions for the degen-
erate eigenfunctions of a charged particle in a uniform magnetic field in Sec. 4.23,
we have exhibited the quantization of the magnetic flux linked to each such eigen-
function, established the linkage between the strength of the magnetic field and the
number of degenerate states that can exist for each energy level (per unit area perpen-
dicular to the magnetic field), and hence shown, in the new Sec. 4.24, how the IHQE
arises. Similarly, the Path Integral approach to quantum mechanics, which provides
an alternative description of the time evolution of a quantum system and illumines
the process of passing to the classical limit, is presented now as the new Sec. 9.5 in
Chapter 9. The next section, also new, applies the path integral approach to elucidate
the Aharonov–Bohm effect (a phase shift of the wave function of a charged particle
that is caused by a magnetic field in a region which the particle does not even enter).
One other addition to the earlier edition is in Chapter 10, where the exact eigenfunc-
tions of a Dirac electron in a Coloumb potential are derived. Passage to the non-
relativistic limit reveals a singularity in one case that is not shared by the solutions
obtained from the non-relativistic treatment in Chapter 4. Apart from these, a dozen
or so new worked out problems, distributed among the various chapters, have been
added to the body of worked examples already present in the first edition. The new
examples include squeezed states of a harmonic oscillator, and rotation and vibration
spectra of molecules, all in Chapter 4, bound states of and scattering cross-section
for a particle in a delta function potential in 3 dimensions (Chapter 6), and helicity
eigenstates (Chapter 10), among other things. Furthermore, a few passages have been
rearranged and/or rewritten for greater clarity. The problems at the close of the chap-
ters are left untouched, to continue as a challenge to the serious student!
Our concentrated effort has been to make this classic text more up-to-date with a
discussion of the latest developments in the subject, relevant as per latest curricula.
The features of the book are given below:
■ It covers important topics, namely, drawbacks of classical mechanics at
the atomic level, Schrodinger equation, matter–wave dual nature, wave
functions and wave mechanics, eigenvalue problems, scattering the-
ory, Heisenbergs’ uncertainty principle, angular momentum theory, and
relativistic wave equations.
xiv Preface to the Second Edition
In the context of the Heisenberg picture of time evolution (which, along with other
pictures, is discussed in Part C), we give an elementary introduction to the concept
of the quantized electromagnetic field and use it to calculate the rates of emission
and absorption of radiation by atoms. A brief account of the representation of states
of ensembles by density matrices, and of their time evolution, makes up the last part
of Chapter 9. The final chapter is devoted to relativistic quantum mechanics, with
emphasis on the Dirac equation and the natural way in which spin and its manifesta-
tions (magnetic moment, spin-orbit interaction) as well as the concept of the antipar-
ticle emerge from it.
We have included a large number of examples distributed through the text, espe-
cially in the earlier chapters, to facilitate a quick grasp of the principal ideas and
methods, and in some cases, to indicate their extensions or applications. Some sup-
plementary material, as well as background material for ready reference, is given in
the Appendices.
In the preparation of the book for publication, we have received cheerful coop-
eration and assistance from the members of the Department of Theoretical Physics,
University of Madras, for which we are grateful. It is a special pleasure to thank
Dr. M. Seetharaman who has read the manuscript critically and rendered valuable
help in many other ways.
The idea of writing a book of this kind grew out of courses given by the first
author for over ten years at Madras University. The work on the book has been sup-
ported by the University Grants Commission through a Fellowship awarded to the
second author and financial assistance towards preparation of the manuscript. For
this encouragement by the Commission, the authors are deeply grateful.
P M Mathews
K Venkatesan
Towards QuanTum
mechanics 1
A. CONCEPTS OF CLASSICAL MECHANICS
The formulation of quantum mechanics in 1925 was the culmination of the search,
that began around 1900, for a rational basis for the understanding of the submicro-
scopic world of the atom and its constituents. At the end of the nineteenth century,
physicists had every reason to regard the Newtonian laws governing the motion of
material bodies, and Maxwell’s theory of electromagnetism, as fundamental laws of
physics. There was little or no reason to suspect the existence of any limitation on
the validity of these theories which constitute what we now call classical mechanics.
However, the discovery of the phenomenon of radioactivity and of X-rays and the
electron, in the 1890s, set in motion a series of experiments yielding results which
could not be reconciled with classical mechanics. For the resolution of the appar-
ent paradoxes posed by these observations and certain other experimental facts, it
became necessary to introduce new ideas quite foreign to commonsense concepts
regarding the nature of matter and radiation—concepts which were implicit in clas-
sical mechanics and had an essential role in determining its consequences. It was
this revolution in concepts, which led to the mathematical formulation of quantum
mechanics, that had an immediate and spectacular success in the explanation of the
experimental observations. Thus, to appreciate the part played by various discoveries
in bringing about this revolution, it is necessary, from the very outset, to have a clear
idea of the classical concepts. We, therefore, begin our studies with a brief discus-
sion of these concepts. The rest of this chapter reviews the developments during the
first quarter of this century which culminated in the establishment of the authority of
quantum mechanics over the domain of microscopic phenomena.
There are two broad categories of entities which physicists have to deal with: mate-
rial bodies, whose essential attribute is mass, and electromagnetic (and gravitational)
fields which are fundamentally distinct from matter.
2 A Textbook of Quantum Mechanics
The most basic concept in the mechanics of material bodies is that of the particle,
a point object endowed with mass. This concept emerged from Newton’s observation
that while the mass of a body has a central role in determining its motions, there is
a wide variety of circumstances in which the size of the body is quite immaterial
(e.g. in the motion of the planets around the sun). The idealization to point size was a
convenient abstraction under such circumstances. Even when such an idealization of
a body as a whole is obviously impossible, as when considering the internal motions
of an extended object, one could imagine the object to be made up of myriads of
minuscule parts, each little part being then visualized as an idealized particle. In this
manner, the mechanics of any material system could be reduced to the mechanics
of a system of particles. It was taken for granted that the motion of such particles,
however large or small their intrinsic mass and size may be, can be pictured in just
the same way as the motion of projectiles or other macroscopic objects of everyday
experience. An essential part of this picture is the idea that any material object has a
definite position at any instant of time. The particle idealization makes it possible to
specify the position precisely, it being intuitively obvious that the position of a point
object is perfectly well defined. The trajectory or path followed by the particle is then
pictured by a sharply defined line, and the instantaneous position of the particle on
.
the trajectory, its velocity, and its acceleration are represented by vectors x(t), x(t)
..
and x (t) with definite numerical values for their components. The manner in which
these vary with time is governed by Newton’s famous equation of motion. Newton’s
equation for the ith member of a system of N particles, is given by
..
mi x i = Fi (i = 1, 2, … N ) (1.1)
The force Fi acting on the ith particle (of mass mi) is a function of the positions
of all the particles (and possibly also of the velocities). The form of the function is
determined by the nature of the interactions of the particles among themselves and
with external agencies. Once these are specified, Eq. (1.1) which form a coupled
system of second order differential equations can, in principle, be solved to obtain all
the xi as functions of t. Since the general solution of a second order ordinary differ-
ential equation contains two arbitrary constants, the general solution of the system of
N such equations for vectors xi depends on 2N arbitrary constant vectors. These may
be chosen so as to satisfy specified initial conditions. In particular, given the 2N vec-
.
tors {xi (t0)} and {xi (t0)}, i.e. the positions and velocities of all the particles at some
instant t0, as initial conditions, the xi(t) are completely determined as functions of t for
.
all t. The velocities xi(t) are then obtained by differentiation of xi (t), thus determining
completely the state of the system at an arbitrary time t.
The meaning ascribed to the word state is a crucial aspect of the difference between
classical and quantum mechanics. In the classical context, knowledge of the state of
a system of particles means knowing the instantaneous values of all the dynami-
cal variables (like position, momentum, angular momentum, energy, etc.). Since
these are all functions of the position coordinates and velocities (or momenta) of the
constituent particles, complete information about the state is implicit in the knowl-
edge of just these quantities. One may, therefore, say that in classical mechanics, the
Towards Quantum Mechanics 3
and the possibility of characterizing the instantaneous state by precise positions and
velocities. The new concepts which replaced these are necessarily divorced from
such intuitive pictures and therefore appear rather strange initially, but they have their
own beauty (especially in the simplicity and elegance of the associated mathematical
structures) which becomes evident with a little familiarity.
The discovery of the phenomena of interference and polarization of light early in the
nineteenth century provided convincing evidence that light is a wave phenomenon.
The nature of light waves was identified some decades later, following Maxwell’s
formulation of the electromagnetic theory. It was then recognized that light consists
of electromagnetic waves and presents just one manifestation of the general phenom-
ena of electromagnetism, governed by the fundamental equations of Maxwell:
1 ∂E 4π
c ∂t = curl H − c j,
div E = 4πρ, (1.2)
1 ∂H
= − curl E,
c ∂t
div H = 0
Here E ≡ E (x, t) and H ≡ H (x, t) stand for the electric and magnetic fields at
x at time t, and ρ(x,t) and j (x,t) are the electric charge and current densities, respec-
tively. At any time t, the vector fields e(x,t) and h(x,t) are implicitly assumed to
have definite numerical values for their components at any point x. This seemingly
self-evident supposition is the essential feature of classical electromagnetic theory.
Since all properties like the energy density, momentum density, etc., of the electro-
magnetic field are functions of the instantaneous values of e and h, specification of
e and h for all x at t0 amounts to a complete description of the state of the classical
electromagnetic field at that instant. Further, such a specification provides the initial
conditions necessary for identifying a particular solution of the first order differen-
tial Eqs (1.2). Therefore, if the state of the field at some instant t0 is given, the state
at any other time can be determined uniquely (at least in principle), provided of course
that the charge and current densities ρ and j are known as functions of x and t. It is
the set of Eqs (1.2) together with the concept of e and h as ordinary vector fields
amenable to simultaneous specification with arbitrary precision at any given time,
that constitutes what is known as the classical electromagnetic theory.
It is worthwhile at this point to notice the obvious but important fact that the classical
pictures of material particles on the one hand, and of light (or more generally, electro-
magnetic fields) on the other, are mutually exclusive. While the ideal particle is a point
object, a field necessarily exists over a region of space. In particular, the purest form of
light, which is monochromatic, is a simple harmonic wave with a definite wavelength,
existing throughout space with a uniform energy density everywhere. It would be
impossible within the framework of this classical picture to conceive of a particle
Towards Quantum Mechanics 5
One of the very few things for which classical theory had been unable to offer an
explanation till the end of the last century was the nature of the distribution of energy
in the spectrum of radiation from a black body. By definition, a black body is one
which absorbs all the radiation it receives. As is well known, the best practical real-
ization is an isothermal cavity with a small aperture through which radiation from
outside may be admitted. The cavity always contains radiation emitted by the walls,
the spectrum of radiation being characterized by a function u(v) where u(v)dv is
the energy (per unit volume of the cavity) contributed by radiation with frequen-
cies between v and v + dv. It had been deduced from very general thermodynamical
arguments that the form of the function u(v) depends only on the temperature t of
the cavity. Efforts to deduce the actual functional form from classical theory led to
the Rayleigh-Jeans formula, u(v) = const. v2. Except at low frequencies this law was
in violent disagreement with experimental observations which showed u(v) falling
off after reaching a maximum as v was increased. That was how matters stood until
1900, when Planck announced the discovery of a law which reproduced perfectly the
experimental curve for u(v):
8πv 2 . hv
u( v ) = 3 hv/kT
(1.3)
c e −1
This formula contains, besides the Boltzmann constant2 k, a new fundamental con-
stant h, with the value
1
For a fuller account, especially of experimental details, see for example, Max Born, atomic physics, 5th
ed., Blackie and Sons, London, 1952; F. K., Richtmyer, E. H. Kennard and T. Lauritsen, introduction to
Modern physics, McGraw-Hill New York, 1955.
2
A table of fundamental constants and other data of interest is provided at the end of the book. As each
physical problem is considered, the student is urged to familiarize himself with the orders of magnitude
of the numbers involved.
6 A Textbook of Quantum Mechanics
The essential new ingredient in the derivation of the law was the following ad hoc
hypothesis:
The emission and absorption of radiation by matter takes place, not as a continu-
ous process, but in indivisible discrete units or quanta of energy. The magnitude ε of
the quantum is determined solely by the frequency of the radiation concerned, and
is given by
ε = hv (1.5)
where h is Planck’s constant.
This hypothesis was a revolutionary break from classical radiation theory based
on Maxwell’s Eqs (1.2). According to the classical theory, oscillating charges are
responsible for the emission (or absorption) of electromagnetic radiation with fre-
quency equal to that of the charge oscillations. Emission or absorption takes place
continuously at a rate determined by the parameters of the oscillating system. The
success of Planck’s hypothesis was the first indication that one might have to look
beyond classical theories for the understanding of at least some areas of physics.
Figure1.1 shows the various regions of the electromagnetic spectrum viewed from
both the wave and quantum points of view.
Wavelength (cm)
106 103 1 10– 3 10– 6 10– 9 10– 12
Waves
Frequency (Hz)
105 108 1011 1014 1017 1020
Photons Ergs
10–21 10–18 10–15 10– 12 10– 9 10– 6
(Energy)
eV
10–9 10–6
10–3
1 10 3
106
Let us now examine briefly how Planck’s law (1.3) follows from his quantum
hypothesis. We make use of the fact that the electromagnetic waves, which constitute
the radiation in the cavity, can be analyzed into a superposition of normal modes
characteristic of the cavity. In each normal mode, the fields vary with time in simple
harmonic fashion, in unison throughout the cavity. Thus each normal mode is equiva-
lent to a simple harmonic oscillator, and the radiation field forms an assembly or
ensemble of such oscillators. The absorption (or emission) of radiation by the walls
of the cavity is equivalent to a transfer of energy to (or from) the walls by (or to) the
oscillators. As a result of such energy exchanges, which are continually taking place,
the ensemble of radiation oscillators comes into thermal equilibrium at the tempera-
ture t of the walls of the cavity. Under these conditions, different oscillators having a
given frequency v have different energies at any given time, but their average energy
Towards Quantum Mechanics 7
E (v) has a definite value determined by the temperature t. The energy of radiation
in the frequency range v to v + dv is then simply the number of normal mode oscil-
lators n(v)dv having frequencies within this range, multiplied by the average energy
E (v) per oscillator. Thus if V is the volume of the cavity,
V.u(v)dv = n(v) dv. E (v) (1.6)
The counting of the oscillators is simply a geometrical problem, and it is easily shown
(see end of Sec. 2.5) that
8πv 2V
n( v ) = (1.7)
c3
The determination of E (v) is done by applying the standard results of statistical
mechanics to the ensemble of oscillators. Statistical mechanics tells us that an indi-
vidual member of an ensemble in thermal equilibrium at temperature t has energy
e with probability
e – E/kT
PE = (1.8a)
∑ e – E/kT
E
E = ∑ EPE (l.8b)
E
The summations are to be taken over all values which the energy e (of any member
of the ensemble) may take. The spectrum of permissible values of e is thus of crucial
importance in determining E. It is here that Planck’s hypothesis comes into play. It
implies that normal mode oscillators of the radiation field with the frequency v can
have only the energy values
3
The denominator is a geometric series whose sum is D = (1– e–βhv)–1, where β = (1/kt ) The numerator
is observed to be nothing but − ∂D/∂β = h ve−βhv (1 − e−βhv )−2 .
8 A Textbook of Quantum Mechanics
for the mean energy of a field oscillator. When the expressions (1.10) for E (v) and
(1.7) for n(v) are employed in Eq. (1.6) we obtain the Planck distribution law (1.3).
According to the classical theory, e(v) could have any value from 0 to ∞, and
the same thing would effectively happen if the quantum hv in the above treatment
had a vanishingly small magnitude. Therefore, passage to the limit h → 0 in Eq.
(1.10) should lead to the value kt predicted by the equipartition theorem of classical
statistical mechanics, and indeed it does. With this value [ E (v) = kt] substituted in
Eq. (1.6), one gets the Rayleigh-Jeans law for u(v) which we have already seen to be
incorrect. It appears, therefore, that the very small but nonzero value of the constant
h is a measure of the failure of classical mechanics. This surmise is indeed confirmed
by the mathematical formu1ation of quantum mechanics.
Example 1.1 If quantum effects are to be manifested through a departure of E (v) from
its classical value kt, the frequency should be high enough, so that (hv/kt) becomes compara-
ble to unity. For room temperatures (t ≈ 300° K), (hv/kt) ≈ 1/6 for v = 1012 Hz. It is only when
‘oscillators’ of at least this frequency are involved, that quantum statistical effects become
noticeable at room temperature. n
That the success of Planck’s hypothesis was no mere accident became evident when
precisely the same kind of ideas provided the solution for another puzzling prob-
lem of classical physics. It is well known that atoms in solids execute oscillations
about their mean positions due to thermal agitation. Each atom may be thought of
as a three-dimensional harmonic oscillator and its mean thermal energy should
be three times that of a simple (one-dimensional) harmonic oscillator. As we saw
in the last paragraph, classical theory predicts the latter to be kt. Therefore, the
thermal energy of a solid should be 3kt per atom, or 3rt = 3Nkt per gram-atom
(containing N atoms where N is the Avogadro number, 6·022 × 1023). The atomic
heat (i.e., the rate of increase of thermal energy with temperature, per gram-atom)
then becomes 3r, a universal constant. Many solids do conform to this expecta-
tion (at least approximately) at ordinary temperatures, as observed by Dulong and
Petit empirically. But when the temperature is lowered sufficiently, the specific
heat decreases instead of remaining constant, and indeed goes down to zero as t
approaches 0 K. Einstein4 observed that this behaviour can be simply explained if
it is postulated that the energy of oscillation of any atom in a solid can take only a
discrete set of values—just like the energy of Planck’s field oscillators. More pre-
cisely, it was proposed that the energy associated with each component (in the x,
y, z directions) of the oscillation of an atom be constrained to take only one of the
values nhv (n = 0, 1, 2, ...), where v is now the frequency of oscillation of the atom.
The mean energy per atom then has exactly the form (1.10), except for an extra
4
A. Einstein, ann. d. physik, 22, 180, 1907.
Towards Quantum Mechanics 9
factor 3 coming from the three directions of motion. If it is assumed that all atoms
have the same frequency of oscillation, one immediately obtains the atomic heat as
hv 2
3Nhv
d e hv /kT
C= = 3R .
(1.11)
hv/kT− 1
dT e (e hv/kT −1)2 kT
It is evident that this formula has the desired property of a gradual decrease in c
as the temperature is lowered. It was found in fact that the behaviour of specific heats
of solids is rather well accounted for by this formula, with a suitable choice of v in
each case.
Einstein’s derivation is by no means the last word on the theory of specific heats
of solids. But it suffices for the purpose of displaying one of the early manifestations
of the inadequacy of classical concepts, namely the need for the supposition that
harmonic oscillators—whether radiation oscillators as in Planck’s theory, or material
oscillators as in Einstein’s theory—can take only discrete energy values.
one see why there should be a definite upper limit on the energy so absorbed. Other
properties of the photoelectric emission are also equally difficult to understand on
the classical picture, but are almost self-evident when this phenomenon is viewed
as the instantaneous absorption of light quanta by the electrons with which they
collide. For example, photoelectric emission starts instantly when light falls on the
emitter, however weak the light intensity may be. (Classically, the electron would
need some time to absorb enough energy to escape.) Under irradiation with mono-
chromatic light, the rate of emission of electrons is directly proportional to the light
intensity—which is exactly what would be expected on the quantum picture since
the number of quanta (and hence, of the collisions with electrons) is evidently pro-
portional to the intensity.
We conclude, therefore, without further discussion that in the photoelectric effect,
light behaves as a collection of corpuscles and not as a wave. At the same time, we know
only too well that the phenomena of diffraction, etc, require light to be waves. How are
we to escape the paradox created by the existence of two quite irreconcilable manifes-
tations for one and the same physical entity? One possibility is to suppose that light
propagates in the form of waves and therefore undergoes diffraction, etc. but assumes
corpuscular character (in some unexplained manner) at the instant of absorption (or
emission) by material objects. However, even this supposition, far-fetched as it is, was
made untenable by the discovery of the compton effect6 in the scattering of X-rays.
Example 1.2 It is an experimental fact that if at all there is any delay between the
commencement of irradiation and the emission of photoelectrons, it is less than 10–9 sec.
An electron requires, let us say, 5 × 10–12 ergs (about 3 eV) to escape from the irradiated
metal. If this much energy is to be absorbed classically (in a continuous fashion), the rate of
absorption must be at least 5 × 10–3 ergs/sec. If the light energy is continuously distributed
over the wave front, the electron can only absorb the light incident within a small area near
it, say 10–15 cm2 (i.e. of the order of the square of the interatomic distance). Therefore, the
intensity of illumination required would be at least (5 × 10–3/10–15) = 5 × 1012 ergs/sec/cm2,
that is half a million watts/cm2! Clearly, explanation of the photo-effect in classical terms is
not feasible. n
That X-rays are electromagnetic waves (differing from light only in the considerably
higher values of frequency) had become clear fairly soon after their discovery. Their
wave nature was amply confirmed by the Laue photographs (1913) showing the dif-
fraction of X-rays by crystals. Yet, barely ten years later, Compton had to invoke
the extreme quantum picture to explain the fact that when monochromatic X-rays
are scattered, part of the radiation scattered in any given direction has a definite
wavelength higher than that of the primary beam. Assuming that X-rays of wave-
length λ consist of a stream of corpuscles or quanta of energy E = hv = hc / λ ,
6
A. H. Compton, phys. rev., 21, 483, 1923; 22, 409, 1923.
Towards Quantum Mechanics 11
Compton theorized that when one of these quanta hits any free or loosely bound
electron in the scatterer, the electron (being quite a light particle) would recoil. Its
kinetic energy has to come from the energy of the incident quantum, and the latter
would be left with an energy e9 < e after the collision (in which it gets scattered).
The frequency v⬘ = E⬘/ h of the X-rays so scattered would therefore be less than ν
and the corresponding wavelength λ9 > λ. On this picture, quantitative calculation
of ∆λ = λ⬘ − λ can be made from considerations of energy and momentum con-
servation in the collision. Since ‘the energy and momentum transported by electro-
magnetic radiation are known to be related by a factor c, the momenta of the X-ray
quantum before and after scattering are given by
E hv h E⬘ hv⬘ h
p= = = , and p⬘ = = = (1.13)
c c λ c c λ⬘
The electron recoiling from the collision may have a velocity comparable to c, and
therefore the relation between its energy W and its momentum p has to be taken as
the relativistic one (see Appendix B):
W = ( m2 c 4 + c 2 P 2 )1/2 (1.14)
where m is the rest mass of the electron. The energy of the electron before colli-
sion may be taken to be the rest energy W0 = mc 2 since the initial kinetic energy
is relatively very small. The configuration of the scattering event is shown in
Fig. 1.2 where, following current practice, the quantum of radiation is depicted by
a wavy line and the electron by a straight line. The equations of conservation of the
E′
p ′,
p, E θ
P,
W
These equations are to be solved for p9 which is related to λ9. We can elimi-
nate ϕ first by squaring Eqs (l.15a) and (1.15b) and adding. Then we get
P 2 = p2 + p2 ⬘ − 2 pp⬘ cos θ . On substituting this expression for p2 in Eq. (1.l5c)
and eliminating the square root, we obtain
[c( p − p⬘) + mc 2 ]2 = m2 c 4 + c 2 ( p2 + p2⬘ − 2 pp⬘ cos θ)
The first outlines of the structure of atoms, of which all matter is constituted, began
to be discernible soon after Thomson’s discovery of the electron (1897). It became
known then that an atom consists of a number of negatively charged electrons plus
a positive residue which carries almost the entire mass of the atom. But it was only
in 1911, with Rutherford’s7 analysis of the data on scattering of alpha particles by
thin foils, that the picture of the atom became clearly defined. Rutherford came to
the conclusion that the observed high proportion of alpha particles suffering large-
angle scattering required that the heavy positive part of the atom be concentrated
in a nucleus, whose size is extremely small compared to the dimension of the atom
itself. This immediately suggested a structure for the atom resembling that of the
solar system, with the electrons revolving in orbits around the nucleus (like planets
around the sun). The Coulomb (electrostatic) attraction between each electron and
the oppositely charged nucleus provides the force which holds the atom together.
It was realized immediately that this picture of the atom encounters serious difficul-
ties of principle in the context of classical theory. In fact, such a structure should not
be stable at all, for the orbital motion of the electrons (which are charged particles)
should cause them to emit radiation continuously. The consequent loss of energy
should make the paths go spiralling inwards until the electrons ‘collapsed’ into the
nucleus. During this process the frequency of the emitted radiation, which coincides
with that of orbital motion, should be continually increasing. Obviously none of these
things happens. The collapse of the kind envisaged does not take place; in fact, atoms
have tremendous stability. Nor does the light actually emitted by atoms have the con-
tinuum character demanded by the above picture. As is well known the most impor-
tant feature of atomic spectra is the presence of very sharp, discrete, spectral lines
which are characteristic of the emitting atom. In brief, acceptance of Rutherford’s
nuclear model of the atom meant also recognition of a complete breakdown of the
classical mechanism of radiation in the case of the atom.
This situation was tackled by Niels Bohr8 who adopted the Rutherford model of the
atom, overcoming its unacceptable consequences by postulating that the classical
theory of radiation does not apply to the atom. He enunciated the following further
postulates concerning the dynamical behaviour of atoms:
(i) The system of electrons and nucleus which constitute the atom cannot exist in
any arbitrary state of motion allowed by the classical mechanics. The system
7
E: Rutherford, phil. Mag., 21, 669, 1911.
8
N. Bohr, phil. Mag., 26, 1, 1913.
14 A Textbook of Quantum Mechanics
Ei − E f
v= (1.18)
h
The first postulate extends the idea of discreteness of energy values, which originated
with Planck, to the individual atom. The energy values associated with stationary states
are called the energy levels of the atom. The second ·postulate incorporates the idea that
emission of radiation takes place in discrete quanta, and adopts the Einstein relation
between the energy of a quantum and the frequency of the associated radiation (first
employed in the explanation of the photoelectric effect). It is an immediate consequence
of these postulates that atomic spectra should consist of discrete lines, as observed.
Equation (1.18) states that the frequencies of spectral lines of any atom are differences
between ‘spectral terms’ (stationary state energies, divided by h) which are character-
istic of the atom. This general property of atomic spectra had been already observed
empirically and was known as the Rydberg-Ritz combination principle (1905). Thus
it seemed certain that Bohr’s ideas were essentially sound. In fact, the supposition that
the atom can have only discrete energy levels was directly verified from experiments
by Franck and Hertz9 on the scattering of monoenergetic electrons by atoms. They
found that as long as the electron energy was below a certain minimum value, the scat-
tering was purely elastic, indicating that the atom is incapable of accepting energies
less· than this amount. This behaviour is exactly what is demanded by the Bohr picture:
an atom in its lowest energy level e0 must get an amount of energy (e1-e0) in order
to go to the next permissible level e1. If energy exceeding this minimum is supplied to
the atom, it can still take up only the exact amount (e1-e0), or (e2-e0) etc. The results
of the Franck-Hertz experiment were indeed in accordance with this expectation. When
the energy of the incident electrons was increased sufficiently, inelastic scattering with
the absorption of discrete amounts of energy was found to take place.
While the postulates stated above provide a frame-work for the understanding
of the stability of atoms and the general features of atomic spectra, they do not
indicate how the stationary states are to be identified or how the energy levels ei are
to be determined for a particular system. But by supplementing these postulates by
a quantum condition, Bohr was able to calculate the energy levels of the hydrogen
atom and thus determine its spectral frequencies. His results were in agreement with
the empirically deduced Balmer formula,
v 1 1
= R 2 − 2 , n, m = 1, 2, … ; m > n (1.19)
c n m
9
J. Franck and G. Hertz, Verhandl. deut. phys. ges., 16, 457, 512, 1914.
Towards Quantum Mechanics 15
where r is the so-called Rydberg constant. This was a spectacular triumph for the
Bohr theory and inspired much of the later work which provided the guidance towards
a more fundamental theory. Let us therefore consider it briefly before discussing the
implications and limitations of Bohr’s theory in general.
The hydrogen atom consists of a proton (which is its nucleus) and a single
electron. According to the Rutherford model, the electron moves in an orbit around
the nucleus; the latter, being relatively very heavy, remains practically at rest. Since
the force of attraction between the two is electrostatic and therefore obeys the inverse
square law (just like the gravitational force in the problem of planetary motion)
the possible orbits are circular or elliptical. Suppose the electron is moving in a cir-
cular orbit of radius a with speed u (which is constant). In this orbit, the electrostatic
attractive force e2/a2 is balanced by the centrifugal force mu2/a, e being the charge of
the electron. This fact gives us the relation
e2
mυ 2 = (1.20)
a
which can be used to eliminate u from the expression for the energy e:
1 e2 1 e2
E= mυ 2 − =− (1.21)
2 a 2 a
Thus the total energy is half the potential energy. Classically, the orbital radius a can
take any positive value, and therefore e can be anything from − ∞ to 0. However,
Bohr’s first postulate asserts that only a special set from among these orbits is avail-
able to the electron. Bohr proposed that these special orbits which characterize the
stationary states are those in which the angular momentum l of the electron about
the centre of the orbit (i.e. the position of the nucleus) is an integral multiple of
� ≡ h/2π. The angular momentum in a circular orbit is of course just the product of
the linear momentum mu and the orbital radius a. Thus Bohr’s quantum condition is
given by
mυ a = n� (1.22)
The integer n whose values identify the various stationary states is called a quantum
number. The radii of the allowed (‘quantized’) orbits are now obtained by eliminating
u between the Eqs (1.20) and (1.22). For the nth orbit one has
n2 � 2
a= (1.23)
me 2
Substitution of this expression for a into Eq. (1.21) gives us the quantized energy
levels of the hydrogen atom as:
me 4
En = − , n = 1, 2, … (1.24)
2 n2 � 2
16 A Textbook of Quantum Mechanics
The radius a = (h2/me2) of the first Bohr Orbit (n = 1), which belongs to the ground
state, is known as the Bohr radius.” The state of the lowest energy (the ground state)
corresponds to n = 1. The energies of the other ‘excited’ states increase with n, tend-
ing to 0 as n → ∞ .
Knowing the energy levels, we can immediately obtain the frequencies of the hydro-
gen spectrum, using Eq. (1.18). If the atom jumps from an initial state with the quantum
numbers n = ni to another with n = nf < ni we find by substitution of the corresponding
energies from Eq. (1.24) into (1.18), that the radiation emitted has the frequency
me 4 1 1
v= 2 − 2
3
(1.25)
4π� n f ni
As already noted, this result agrees with the Balmer formula (1.19) and provides a
theoretical value R = ( me 4 /4π c �3 ) for the Rydberg constant.10 We will see later
that exactly the same formula follows from the quantum mechanical theory also,
though the meaning of the quantum number n there is quite different.
∫� p dq = nh (1.26)
which had been employed already by Planck in connection with his theory of black
body radiation.11 Here q is some generalized coordinate and p, the corresponding
canonically conjugate momentum. The condition is applicable in the case of periodic
motion only, and the integral is to be taken over one period, treating p as a function
of the position q of the particle on the actual trajectory. Eq. (1.22) corresponds to
choosing q to be the angular position ϕ (varying from 0 to 2π for one period) and p
as its conjugate, the angular momentum (which is independent of ϕ, being a constant
of the motion). Bohr postulated that the quantum condition (1.26) is applicable to
any system with one degree of freedom. This quantum rule was further generalized
by Sommerfeld to multiply-periodic systems, with many degrees of freedom, i.e.
10
This value is strictly correct only for an infinitely heavy nucleus which remains perfectly static. To
indicate this fact explicitly, the notation r∞ is often used for the constant. To take into account the finite-
ness of the mass (mN) of the nucleus (and the consequent motion of the nucleus) we have to replace the
electronic mass m in the expression for r by the reduced mass m mN/(m + mN). The Rydberg constant
for a finite nucleus is thus rN = r∞(1 + m/mN)-1.
11
Planck assumed that the emission and absorption of radiation in quanta hv was done by hypothetical har-
monic oscillators capable of having only discrete energy values, nhv (n = 0, 1, 2, ... ). He showed that this
quantum condition on the energy levels of the oscillator was equivalent to quantizing the ‘action integral’
as in Eq. (1.26). For this reason, the name ‘quantum of action’ has been applied to Planck’s constant.
Note that the value n = 0 does not make any sense in Bohr’s quantum condition, and had to be dropped.
Towards Quantum Mechanics 17
∫� pr dqr = nr h, r = 1, 2, … N (1.27)
where the integration in the case of each pair of conjugate variables is to be taken over
one period of that particular pair. The quantum numbers nr take integral values.
Bohr’s general postulates together with the quantum rule (1.27) constitute what is now
known as the old Quantum theory..13 Details of the applications of the theory are of no
particular interest now, since the theory itself has been superseded by the new quantum
theory or Quantum Mechanics. But the quantization of the elliptical orbits of the hydro-
gen atom deserves mention because it gave the first example of two general properties
which persist in the quantum mechanical theory. One is the property of degeneracy of
energy levels of systems possessing symmetries. Sommerfeld observed that the quan-
tized elliptical orbits in a given plane are identified by two quantum numbers n9 and k,
characterizing the radial and angular parts of the motion in the orbit. He found however
that the energies associated with such quantum states depend only on the sum n = n9 + k
of these quantum numbers, and are given by the Bohr formula (1.24). Thus for a given
value of the total or ‘principal’ quantum number n, there are n different states (elliptical
orbits of various eccentricities, corresponding to14 k = 1, 2, ... n) all of which have the
same energy en. We say that the energy level en is n-fold degenerate (when the quantum
orbits in one plane alone are considered). It is now recognized that the degeneracy in
the case of the hydrogen atom is due to the special nature or ‘symmetry’ of the distance-
dependence of the electrostatic potential, and does not occur for other potentials (unless
they have other symmetries, of course). The second general property exemplified in the
Sommerfeld treatment is the removal of degeneracy (i.e., departure from equality of the
energy values of the previously degenerate quantum states) when the symmetry is ‘bro-
ken’. In the case of the hydrogen atom there is indeed a slight departure from symmetry.
It is caused by the fact that the variation of the speed of the electron as it moves along an
elliptical orbit induces corresponding changes in its mass as given by the theory of rela-
tivity. Sommerfeld showed that this mass variation gives rise to a slow precession of the
orbit in its own plane,15 and that because of this, the energy associated with each orbit is
12
W. Wilson, phil. Mag., 29,795, 1915; A. Sommerfeld, ann. d. physik, 51, 1, 1916.
13
Discussion of the old quantum theory and many of its applications may be found in L. Pauling and
E. B. Wilson, introduction to Quantum Mechanics, McGraw-Hill, New York, 1935; A. Sommerfeld,
atomic structure and spectral lines, 3rd ed., Methuen and Co., London, 1934.
14
In the Bohr-Sommerfeld theory the angular momentum in a stationary state was identified as kh but
quantum mechanics shows that it is given by lh, where l takes the same values as (k-1) for given n,
namely l = 0, 1, ... , n–1. In the discussion of space quantization below, we use this quantum number l
in preference to k.
15
More precisely, the orbit no longer closes on itself, but remains very nearly elliptical for each revolution,
with the direction of the major axis changing slightly (in the plane of the ellipse) from one revolution to
the next. This change of orientation of the ellipse at a steady rate is called precession.
18 A Textbook of Quantum Mechanics
changed by a very small amount depending on k. Consequently, the nth Bohr level gets
split into n closely-spaced levels—there is no more degeneracy. The spectral lines also
then get split, developing what is called a fine structure. Sommerfeld’s calculation gave
complete agreement with the observed fine structure.
In giving the degree of degeneracy of the nth Bohr level (i.e. the number of quan-
tum states belonging to this level) as n, and in asserting that the relativistic mass
variation removes this degeneracy, we have taken account of orbits in any one plane
only. To put this in another way, only orbits with a specific direction for the angular
momentum vector (which is normal to the plane of the orbit) were considered. Actu-
ally there is a further degeneracy associated with the possibility of various orienta-
tions for the angular momentum vector with respect to any fixed axis. Such an axis
may be defined, for instance, by the direction of some externally applied field. From
the consideration of the quantum condition (1.27) in axially symmetric situations
it was inferred that the direction of angular momentum should be quantized. To be
more specific, the component of angular momentum parallel to the axis has to be mh
with the quantum number m taking integer values only. This is called quantization
of direction or space quantization. When the quantum number characterizing the
magnitude of the angular momentum has a value l, the values which m can take are
limited to m = l, l - 1, … , - l + 1, - l.
The existence of space quantization was experimentally demonstrated in a very
direct and beautiful fashion by Stern and Gerlach.l6 They exploited the fact that an
atom with nonzero angular momentum has a magnetic moment µ in the same direc-
tion as the angular momentum vector L; for, the orbital motion of the electron (with
which the angular momentum is associated in the Rutherford-Bohr picture) also pro-
duces a magnetic moment since the electron is a charged particle. If such an atom
is placed in a magnetic field h, the field exerts a torque µ × h tending to turn the
direction of µ and hence that of L too into alignment with the field h. Now, it is
well known that any torque acting on an angular momentum vector has a gyroscopic
effect. Therefore L precesses around the direction of h, keeping a constant angle θ
to h all the time. This is all that happens if h is a uniform field. If it is not uniform,
there is also a net force ( µ .∇) h on the atom. Thus, if a coordinate system is chosen
with z-axis in the direction (Fig. 1.3) of h and if the field strength h increases in
the z-direction, we have a situation where µz and lz are constants for the atom, and
there is also a force µ z (∂H/∂ z ) on the atom, acting in the z-direction. If a beam of
atoms is shot through the field in a direction perpendicular to the field, say along
the x-direction, the above force causes the individual atoms to be deflected up or
down (i.e., in the positive or negative z-directions) by amounts proportional to their
respective values of µz . Therefore, if the values of µz form a continuous range (in
accordance with classical concepts) the beam would widen out into an expanding strip in
the x-z plane. On the other hand, if there is space quantization, so that lz (and hence µz )
can take only a discrete set of values, the beam of atoms would split into a number
16
O. Stern and W. Gerlach, Z. physik, 8, 110, 1922.
Towards Quantum Mechanics 19
EM
Stream of Atoms
z
EM
x S
0
Fig. 1.3 Schematic diagram of the Stern-Gerlach experiment
showing splitting of the beam of atoms between the
pole-pieces (EM) of electromagnet.
of distinct diverging beams, each of which is characterized by a specific value of µz .
Impinging on a plane perpendicular to the x-axis, these beams would leave spots spaced
out along the z-direction (instead of a continuous line which would appear if there were
no space quantization). It is the appearance of such distinct spots in the Stern-Gerlach
experiment which gave direct confirmation of the idea of space quantization. In this
experiment a fine, well-collimated beam of silver atoms was passed through an inho-
mogeneous field created by an electromagnet with specially shaped pole pieces. One
of these had a ridge along the middle, and facing this was a hollowed-out channel in
the other pole piece. The net result was a concentration of field lines (high intensity)
near the former and dispersal (low intensity) near the latter. It was found that when the
atomic beam was made to pass between the pole pieces, traversing their whole length,
the beam split into two—one part deflected upwards, and the other downwards.
While this confirmation of space quantization was a success for the old Quantum
theory, it must be mentioned that the appearance of just two values for µz was cor-
rectly explained only after the discovery of the spin of the electron a couple of years
later. Unlike l, which can have only integral values, the quantum number s charac-
terizing the angular momentum associated with the spinning motion has the value
1 , and the component of spin in any direction can have just the two values + 1 �
2 2
or − 12 �. Deferring further discussion of this subject, we return now to our main
theme: the progression of concepts from the classical to the quantum mechanical.
We have seen above how the Old Quantum Theory has been able to provide the expla-
nation of the spectrum of the hydrogen atom, including its fine structure. Recognition
of the quantum character of the magnitude and direction of angular momentum remains
as one of its finest achievements. However, despite these and other very considerable
successes of the Old Quantum Theory, it is quite obvious that it is not really a funda-
mental theory and is, in any case, only of limited applicability. The scope of the Bohr-
Sommerfeld quantum rules is restricted to periodic or multiply-periodic motions; they
have nothing to say about situations where other kinds of motion are involved. Even in
the Franck-Hertz experiment which gave direct support to Bohr’s concept, the behav-
iour of the electrons scattered by the atoms is outside the purview of the Old Quantum
Theory. The limitations of the theory were greatly mitigated by skilful exploitation of
20 A Textbook of Quantum Mechanics
the idea that the results of quantum theory should tend to those of the classical theory
under circumstances where the quantum discontinuities are negligibly small. This idea,
which places powerful constraints on the quantum theory, played a considerable role
in the developments of the decade preceding the birth of quantum mechanics. A for-
mal enunciation of the idea, under the name correspondence principle, was given by
Bohr.17 In considering the quantum mechanics in later chapters we will have occasion
to discuss its correspondence with classical mechanics in certain aspects of their math-
ematical structures as well as in the sense of a passage to the limit h → 0.
Looking back on the essentials of the Old Quantum Theory, we see that its fundamen-
tal shortcoming is that it is a peculiar hybrid of quantum concepts grafted on to classical
mechanics. The existence of discrete stationary states is experimentally well substanti-
ated. But as long as the classical picture of well-defined particle orbits is retained, it
remains incomprehensible why certain orbits should be completely stable and others not
allowed to exist at all. This perplexing question was responsible in part for the ultimate
realization that particle states at the microscopic level are not describable in terms of
well-defined orbits, but must be pictured in terms of some kind of waves.
Example 1.3 Show that the frequency (en+1 - en)/h of the line emitted by the quantum
transition from level (n + 1) to n in the Bohr model of the hydrogen atom is, in the limit of
large n, the same as if the radiation was being emitted classically by an electron moving in the
Bohr orbit associated with either of these levels.
The velocity u of motion of the electron in a Bohr orbit may be obtained by eliminating a
between Eqs. (1.20) and (1.22). One finds that u = e2/(nh̄). The orbital frequency of the elec-
tron is evidently vcl = u/(2πa). Since a = n2h̄2/(me2) from Eq. (1.23), we have
υ me 4
vcl = = . (E1.1)
2πa 2πn3� 3
If this electron were emitting radiation classically, the radiation would have the same fre-
quency ν.
According to quantum concepts, radiation is associated with transitions between quantum
states. For a transition between adjacent Bohr orbits, the frequency of the radiation emitted is
given by hv = (en+1 - en). Thus
1 me 4 1 1 me 4
v= −
≈ (E1.2)
h 2�2 n2 ( n + 1)2 2π�3n3
for n >> 1. The equality of this expression to the frequency νcl for classical emission provides
an illustration of Bohr’s correspondence principle. n
The suggestion that matter may have wave-like properties was first put forward in
1924–25 by Louis de Broglie.18 He argued that if light (which consists of waves
according to the classical picture) can sometimes behave like particles, then it should
17
N. Bohr, Nature, 121, 580, 1923.
18
L. de Broglie, phil. Mag., 47, 446, 1924; annales de physique, 3, 22, 1925.
Towards Quantum Mechanics 21
E = hv = �ω (1.28a)
where ω = 2πv is the angular frequency. He noted then that according to the theory of
relativity, the energy e and the momentum p of a particle form the components of a
single four-vector, and so do the angular frequency ω and the propagation vector k of
wave. Since the components of both the four-vectors have to transform in an identi-
cal manner (according to the Lorentz transformation) for a given change of reference
frame, any proportionality relation like (1.28a) between e and ω must necessarily
hold also between p and k. It was concluded, therefore, that
p = hk (1.28b)
In terms of the magnitudes of the vectors, this gives
h h
p = �k = , or λ = (1.28c)
λ p
19
C. Davisson and L. H. Germer, Nature, 119, 558, 1927; phys. rev., 30, 705, 1927.
20
Kikuchi, Japan. J. phys., 5, 83, 1928.
21
G. P. Thomson, Nature, 120, 802, 1927; proc. roy. soc., london, A117, 600, 1928.
22
Appropriate momenta are those for which the de Broglie wavelengths as given by Eq. (1.28c) are in the
range of X-ray wavelengths used in diffraction studies—from a few angstroms down to a tenth of an
angstrom or less.
22 A Textbook of Quantum Mechanics
is then obtained from the relativistic relation p2 = 2m ekin + (e2kin /c2), which is
equation (B.16) of Appendix B with e identified as mc2 + ekin. Thus
h h −1/2
λ= = 1 + Ekin (1.29)
p 2mEkin 2mc 2
When ekin is much less than the rest energy mc2, the last factor is negligible; and
then, on substituting ekin = eV/300, we get
150 12.25 8
λ=h = × 10 cm
meV V
where V is to be expressed in volts.
Example 1.4 For an electron accelerated through 100 volts, the wavelength is (12.25 ×
10–8/10) cm = 1.225 Å. Thermal neutrons (with mean kinetic energy Ekin = 23
kT ≈ 6.2 × 10−14 ergs at 300°K) also have wavelengths of the same order. Substituting the
neutron mass and the above value of ekin in Eq. 1.29 we get λ 1.5 Å. In contrast the wavelength
of a one gram mass with the same thermal energy, has the fantastically small value of about
1·2 × 10-20 cm. There is no wonder then that macroscopic objects show no noticeable diffrac-
tion effects or other wave-like behaviour. n
The Davisson-Germer experiment involved diffraction by a single crystal; Kiku-
chi reproduced the Laue type of diffraction pattern by transmission through very
thin mica crystals, and Thomson obtained the analogue of the Debye-Scherrer rings
by passage of electrons through thin (poly-crystalline) metal foils. In each case, the
wavelength λ of the electron waves, as determined from the diffraction pattern, was
found to be equal to (h/p) in agreement with de Broglie’s theoretical prediction. Thus
the dual nature of matter, like that of radiation, was firmly established.
Actually the theoretical exploitation of de Broglie’s idea did not await its experi-
mental verification. Immediately after the publication of the hypothesis, Erwin
Schrödinger23 proposed that the behaviour of matter waves associated with material
particles (whether free or subject to forces) is governed by a certain differential equa-
tion for the wave function ψ (i.e., the function which represents the matter wave). He
showed that this wave equation, together with physically-motivated conditions on the
wave function, leads in a very natural way to stationary states characterized by discrete
energy values. Thus Schrodinger was able to explain the basic fact of quantization as
a consequence of the wave nature of matter, and to replace the ad hoc quantization
rules of the Old Quantum Theory by mathematical conditions of a very general nature
on the wave functions. The Schrödinger theory, called wave mechanics, is one of the
alternative formulations of the general theory of quantum mechanics which, as far
as we know at present, gives the correct fundamental theory of the physical world at
the microscopic level. Much of this book will deal with the development of quantum
mechanics following Schrödinger’s approach. We defer the introduction of the wave
equation to the next chapter, and devote the rest of this chapter to a discussion of the
conceptual problems raised by the developments described in the preceding sections.
23
E. Schrödinger, ann. d. physik, 79, 361, 489, 1926; 80, 437, 1926; 81, 109, 1926.
Towards Quantum Mechanics 23
The most serious problem raised by the discovery of the wave nature of matter con-
cerns the very definition of the word ‘particle’. We have already noted that classically,
the mental picture of a particle was that of a point object endowed with a precise
position and momentum at every instant of time. The discovery of the quantum prop-
erties (discreteness of energy values, etc.) of the atom made it clear that such a pic-
ture cannot hold, at least in its entirety, in the atomic domain. However, the success
of the de Broglie hypothesis, that a particle of momentum p is to be associated with
a wave of definite wavelength λ = h/p, leaves one in a very uncomfortable situation.
For, a pure harmonic wave necessarily extends over all space, and this fact makes it
impossible to get any idea of the position of a particle described by such a wave. On
the other hand, if we have a wave packet, i.e., a wave which is confined to a small
region of space, it would seem reasonable to suppose that the particle is within the
region of the packet. The position of the particle would then be approximately deter-
mined by that of the wave packet, there being an uncertainty in the position which is
of the order of the dimension or size of the packet. If this ‘fuzzy’ wave packet picture
of a particle is to be taken seriously, it is necessary that such a wave packet should
move like a classical particle, to a good approximation. Let us now test whether this
is the case. For simplicity we consider first a wave packet in one-dimensional space.
Consider a wave packet represented by a wave function ψ( x, t ) which, at the instant t,
has a maximum at the point X(t). We will suppose for convenience that the function ψ
falls off monotonically as in Fig. 1.4 as x moves to either side from the maximum
position. If the position of the wave packet changes with time, the rate at which the
maximum point moves is clearly a good measure of the velocity ug of the packet:
dX (t )
υg = (1.30)
dt
ψ (x,t)
X (t) x
The subscript g on u stands for group velocity24 and is a reminder that the wave packet
is composed of a group of harmonic waves having a certain range of wavelengths:
∞
ψ ( x, t ) = ∫ a( k ) ei ( kx – ωt ) dk (1.31)
–∞
where a(k) is the amplitude of the harmonic component with propagation constant k
(wavelength λ = 2π/k) and angular frequency ω, in the wave function ψ. Equation (1.31)
simply expresses the well-known mathematical fact that any reasonable function can
be analyzed into a Fourier integral. We will assume that the harmonic waves present
in ψ (x,t) have values of k lying within a small range centred about some value k . In
other words, a(k) in Eq. (1.31) is taken to be non-vanishing only when k is very close
to k . This assumption is made in order to ensure that the momentum of the particle
described by the wave packet is reasonably well defined: When k is restricted to a
narrow range, the momentum associated with it according to the de Broglie relation
is also restricted likewise. We note further that the frequencies ω of these harmonic
waves will also, for the same reason, remain very close to ω = ω( k ) . Therefore, if
ω(k) is expanded as a Taylor series in powers of ( k – k ) we can, to a good approxi-
mation, neglect terms of higher order than the first, and write ω(k) as
ω( k ) = ω + ω⬘. ( k – k ) (1.32)
where
d ω dω
ω = ω( k ) and ω⬘ = =
dk dk
k
Let us now turn to the determination of the group velocity ug for a packet with
the above properties. First we note the elementary fact that the position X(t) of the
maximum of ψ is the point at which (∂ψ/∂x ) vanishes. Thus
∂ψ
0 = = ∫ a( k ) ⋅ ikei [ kX ( t ) – ωt ] dk
∂x (1.33)
x = X (t )
(∆ p ) (∆ x ) ⲏ � (1.37)
This inequality is the essential content of the heisenberg uncertainty principle,
which will be derived later (Sec. 3.11) in a more precise form from quantum mechan-
ics. It is easy to convince oneself that in the case of a three-dimensional wave packet25
25
To save, writing, we shall use only a single integral sign even in the case of multiple integrals. Wher-
ever the number of integrals involved is relevant, it will be indicated by the notation for the ‘volume
element’ such as d 3k (meaning dkx dky dkz) or d 3 x ≡ dx dy dz . The integration is always from − ∞
to + ∞ for each variable, unless otherwise specified.
26 A Textbook of Quantum Mechanics
(We omit details of evaluation of the integral above. Its value may be obtained from the last equation
under Example 3.11 (p. 98), by making the replacements σ 2 → σ 2 + i �t /m, x → x − i σ 2 k ) .
Observe that the absolute value of the wave function, | ψ( x, t ) | , has a Gaussian form. Its
peak is at x = t � k /m i.e. it moves with the velocity ( �k /m). This is just what we expect of
a particle of momentum p = �k and is in conformity with Eq. (1.35). However, it is to be
noted that the size or width of the wave packet at time t is σ 2 + �2t 2 /m2σ 2 1/2 , and it increases
indefinitely with the passage of time. On account of this spreading, any possibility of identify-
ing a wave packet directly as a particle is ruled out. n
Here λ is the wavelength of light used for illuminating the object, and α is the half-
angle subtended by the objective lens of the microscope at the position of the object
being examined. The higher the accuracy needed, the smaller the wavelength λ has
to be. But light consists of photons of momentum p = h/λ and the x-component of
the momentum of a photon scattered by the object into the microscope is uncertain
by an amount ±( h/λ ) sin α . In scattering the light, the object itself recoils, and the
x-component of the recoil momentum is also evidently uncertain to the same extent:
h h
∆ px ∼ sin α = .
λ ∆x
It follows, therefore, that in the very process of trying to determine position pre-
cisely, the momentum is made uncertain, the extent of the uncertainty being what is
demanded by the Uncertainty Principle.
It is worth observing that in this discussion, the wave characteristics of the object
whose position-momentum uncertainty is under consideration have played no role.
(In this sense the uncertainty we are talking of here is quite a different thing from
what was considered in the last section). It is the quantum properties of the agent
(light) used in the measuring process which brought about the uncertainty. As long as
all possible agents (matter and light) have quantum properties, no measurement can,
even in principle, lead to absolutely precise determinations of positions and momenta
(even if there were no theoretical objection to the object itself having a precise posi-
tion and momentum).
Let us now return to Sec. 1.14 and summarize the main conclusions. The wave nature
of matter has the consequence that particles in the strict classical sense (with pre-
cise positions and momenta) are unrealizable in nature. The closest one can get to a
classical particle is through a wave packet with ∆x and ∆p made as small as pos-
sible subject to the constraint (1.37). Such a wave packet does move approximately
like a classical particle. Our verification of this last statement is actually incomplete.
To justify it fully we must show that the correspondence (1.28) between the corpuscu-
lar and wave aspects holds good also in the case of particles moving under the influ-
ence of forces. Such a proof can be given on the basis of the perfect analogy which
28 A Textbook of Quantum Mechanics
where ds is the length of an infinitesimal segment of the path. Fermat’s principle states
that the actual path taken by the light ray is one which minimizes the integral. Mau-
pertuis principle concerns the motion of a particle in a varying force field. It states
that the actual path taken by the particle is one which minimizes the action integral
x2
∫ p ds (1.41)
x1
A comparison of the two principles shows that the path of a ray associated with a
wave of variable wavelength λ would be identical with that of a particle in a force
field if p ∝(1/λ). But this is just what we have if we assume the de Broglie relation
to be applicable also to particles subject to forces. We note that, if a particle mov-
ing nonrelativistically in a force field has potential energy V(x) at the position x, its
momentum (for a given total energy e ) is given by
p = [2m ( E − V )]1/2
It varies with the position of the particle. Correspondingly, the wavelength λ attrib-
uted to the particle is also a function of x.
h h
λ= = (1.42)
p [2m( E − V ( x))]1/2
Now, it makes sense to talk of a position-dependent wave-length only if λ(x) has an
appreciably constant value over many waves in the neighbourhood of x, i.e., from
one wave to the next, the fractional change in wavelength should be very small.
In symbols,
| ∇λ | Ⰶ 1 (1.43)
This means in turn that the potential energy V(x) should be a slowly-varying function
of x. Thus, the motion of a matter-wave packet in a force field will coincide approxi-
mately with that of a classical particle provided the variation of the field is slow.
Our efforts in the last section have been to convince ourselves that despite the wave
nature of matter, classical mechanics holds in an approximate sense under suitable
Towards Quantum Mechanics 29
circumstances, and that the approximation is in fact perfectly adequate for the descrip-
tion of the macroscopic world. Under such circumstances, the wave aspect of mat-
ter remains unobtrusive and does not create any serious problems of visualization.
Difficulties of interpretation arising from the wave-particle dualism appear in an
acute form when phenomena (e.g., diffraction) in which the wave aspect plays an
essential role are considered. The fundamental problem then is how to reconcile the
discrete nature of material entities with the wave-like behaviour exhibited by them.
The same problem occurs in reverse in the case of radiation, whenever it exhibits the
discrete photon character in an essential way.
Let us now analyse a diffraction experiment to see how the mutual association
of the two (seemingly incompatible) aspects, whether of matter or of radiation, is to
be understood. To be specific, let us think of the experiments of G. P. Thomson or
Kikuchi. We recall that electron diffraction photographs identical with the Laue and
Debye-Scherrer X-ray diffraction patterns were obtained in these experiments. More
specifically, in electron as well as X-ray diffraction, the observed pattern is precisely
what one would expect on the basis of the wave theory, assuming the incident beam
to be a harmonic wave and the intensity at any point on the pattern to be propor-
tional to the absolute square of the amplitude of the wave at that point. It is also an
empirical fact that the nature of the diffraction pattern is quite independent of the
intensity of the incident beam, i.e. on taking a long-exposure diffraction photograph
with a weak incident beam one gets exactly the same pattern as with a beam many
times more intense and a correspondingly shorter exposure. This is again just what
one would expect on the wave theory, but it acquires profound significance when we
recall that the beam actually consists of discrete particles (electrons or photons)26.
For it implies that the diffraction process is independent of the number of particles
simultaneously present in the beam, and hence that the wave property manifested
through diffraction is not the result of some conspiracy among the particles present.
Instead, the wave nature has to be an inherent property of each particle. This infer-
ence is confirmed by experiments in which the beam intensity is made so low that
there is effectively only one particle at a time going through the apparatus; the
diffraction pattern still emerges if recording is made for a sufficiently long period
of time. We have to conclude, therefore that associated with each particle there
is a wave which undergoes diffraction in the crystal. If the wave is represented
by a (possibly time dependent) function ψ( x, t ) , diffraction manifests itself in
2
a variation of | ψ | with x at points along the surface of the photographic film.
2
The form of | ψ | as a function of x is the same for all the particles, and it is the
accumulated effect of all of them which makes up the actually observed diffraction
pattern. The intensity distribution i(x) in the diffraction photograph is then
proportional to | ψ (x) |2 .
26
One needs no persuasion to accept that an electron beam consists of electrons as discrete particles. In
the case of X-ray diffraction, one could keep a Geiger counter-in the path of the incident or diffracted
X-rays and actually count the individual photons. The counting rate is proportional to the X-ray intensity
(for monochromatic X-rays). In fact, intensity measurements are often made using Geiger counters.
30 A Textbook of Quantum Mechanics
This conceptual picture of the diffraction phenomenon still leaves us with the fol-
lowing question. Does the electron (or photon) smear itself out over the photographic
plate in a manner determined by its own wave function (or rather, by | ψ |2 )? The
answer is no. It remains discrete and may be detected as a discrete particle at some
point of the diffraction pattern. In that case, how can the photographic plate possibly
know of the existence of peaks and zeros of | ψ |2 ? The only possible answer seems
to be a statistical one. Each individual particle is recorded at some point or other of
the photographic plate, at random, with a probability proportional to the value of
| ψ |2 at that point. When a large number of particles get recorded in this fashion,
the average number arriving in the neighbourhood of a particular point x of the pho-
tographic plate is proportional to | ψ (x) |2 ; on the other hand, the intensity I(x) of
the diffraction pattern (represented for instance by the degree of blackening on the
photograph) is also proportional to this average number. Hence I (x) ∝ | ψ (x)|2 , and
thus the formation of the diffraction pattern through the diffraction of individual
particles is explained.
We will see later that the above statistical interpretation is a fundamental fea-
ture of wave mechanics. Every particle (or system of particles) has a wave function
associated with it. The wave function determines the corpuscular characteristics like
position, momentum, angular momentum, etc. in a statistical sense. For instance a
particle with a wave function ψ ( x, t ) has a probability proportional to | ψ ( x, t ) |2 for
being found in the neighbourhood of the point x, and there are other formulae which
give probability distributions for momentum, etc., in terms of ψ . Nature does not
permit any more precise specification of the state of a particle than what is provided
by the wave function and the probability distributions obtainable from it.
1.18 COMPLEMENTArITy
There is one final question we would like to consider. To make its import clear, we
pose it in the context of a simple situation like Young’s two-slit experiment show-
ing the interference of light, or a hypothetical analogue involving electrons instead
of light. Can we say that the photon (or electron) which reaches some point of the
interference pattern has followed a definite trajectory passing through a particular
one of the two slits? The probability interpretation, which we were forced to adopt
above, makes it pretty clear that we cannot. For, the particle has a definite probability
of appearing at any point where the wave function is non-vanishing, and the wave
function has to have non-vanishing values in a region which includes the locations of
both the slits if interference is to take place. Thus the particle has a chance of being
found at either of the slits and we cannot say that it has passed through a particular
one of them. Another slightly different viewpoint which also leads to the same con-
clusion, is the following: Let us assume that the particle has a trajectory. We have
seen already that sharp trajectories in the classical sense are not possible; so let us
suppose the trajectory is a moving wave packet, with a lateral extension less than
half the distance a between the two slits. By this condition we make the trajectory
sufficiently well defined, so that we can be sure that it can go through only one of
Towards Quantum Mechanics 31
the two slits, if at all. Suppose all particle trajectories are defined with this degree of
precision (though the individual trajectories may be shifted sideways with respect to
each other). Can we have an interference pattern under these conditions? The answer
is clearly no, from the wave point of view, since any wave which does not cover
both the slits cannot cause/interference, as already noted. But it is instructive to note
that the same result can be deduced from the wave-particle dualism. Since we have
assumed that the extension of the wave packet laterally is less than 12 a, the uncer-
tainty in any component of the position vector transverse to the direction of motion
of the particle is ∆y < ( 12 a ) . Therefore, according to the uncertainty principle
[Eq. (1.39)], the corresponding (transverse) momentum component has an uncer-
tainty ∆p y of at least ( h/∆y ) = ( 2h/a) . The direction of motion of the particle is
therefore uncertain by an angle of at least
θ ≈ tan θ = (∆p y / p) = 2h/ pa = 2λ/a (1.44)
where λ is the de Broglie wavelength of the particle. Now the angular separation
between successive maxima of the interference pattern due to two slits separated by
a distance a is (λ/a). If the (uncontrollable) uncertainty in the direction of motion of
the particle exceeds this value, as it does according to Eq. (1.44), the interference
pattern will obviously get washed out. It may be noted that for the purposes of the
above argument it does not matter whether the ‘particle’ we refer to is a material
particle or a photon.
The. principal result of the above discussion may be stated in a general form as
follows: in any experimental situation in which a physical entity (matter or radia-
tion) exhibits its wave properties, it is impossible to attribute corpuscular character-
istics to it. The entity does behave like a particle if its wave packet is made compact
enough, as seen in Sec. 1.14, but when this is done, it loses the ability to display any
wave properties. In other words, the particle and wave aspects of a physical entity are
complementary and cannot be exhibited at the same time. This is the complementarity
principle of Bohr.
The classification of the relation between the particle and wave aspects of physi-
cal entities, in the manner presented above, came about only sometime after quan-
tum mechanics had been formulated as a mathematical theory. As we have already
observed in Sec. 1.13, Schrödinger’s formulation (the so-called wave mechanics
which is based on the de Broglie hypothesis of matter waves), preceded even the
direct experimental verification of the existence of matter waves. Schrödinger (1926)
proposed that the wave function ψ describing the matter waves satisfies a partial
differential equation, and gave a prescription for writing down the equation for any
particular system of particles.27 He showed that there exist solutions of the equation
27
The Schrödinger equation is applicable only to non-relativistic particles. A new wave equation, which
meets also the requirements of relativity theory, was formulated by Dirac in 1928. We will take up the
Dirac equation towards the end of the book.
32 A Textbook of Quantum Mechanics
which correspond to a stationary wave pattern, rather like the familiar standing waves
or normal modes of a string or membrane. And just as boundary conditions on the
latter restrict the possible normal mode frequencies to a discrete set of values, the
natural boundary conditions on matter waves restrict the energies associated with
the stationary patterns to discrete values (in the case of particles which are ‘bound’,
i.e. confined to a finite region of space by the action of a potential). In this manner,
a natural explanation was found for the existence of stationary states, originally pos-
tulated ad hoc by Bohr. This, and the correct quantitative prediction of the stationary
state energy levels of some important systems, constituted the first spectacular suc-
cess of Schrödinger’s wave mechanics. We will not pursue here any further the suc-
cesses of this theory since these form the subject matter of the next several chapters.
At just about the same time as the announcement of Schrödinger’s theory came that
of another theory (now called Matrix Mechanics), due to Heisenberg28. Despite the
vastly different appearance of the two theories it was very soon recognized that they
are completely equivalent. While the Schrödinger formulation is based on explicit
use of a mental picture of matter as waves (replacing the classical point-particle pic-
ture), Matrix Mechanics was born out of a deliberate effort to exclude from the theory
any reference to conceptual pictures which are not amenable to direct experimental
verification. Working against the background of the Old Quantum Theory of atomic
spectra, Heisenberg noted that the inevitable disturbance of the motion of a particle
which is caused by any attempt to observe it (cf. Sec. 1.15) is, in the case of elec-
trons in atoms, of sufficient magnitude to make their orbits unobservable. He set out
then to reformulate the quantum theory in terms of observable quantities alone (like
intensities and frequencies of spectral lines). The resulting formulation was found to
be equivalent to replacing the position, momentum and other, dynamical variables of
classical mechanics by matrices, while keeping the form of the equations of motion
superficially the same as in the classical mechanics. As is well known, the product
of two matrices depends, in general, on the order in which they are multiplied, i.e.
matrix multiplication is not commutative. It is through the non-commutativity of the
product of the position and momentum variables (contrasted with their perfect com-
mutativity, as ordinary numbers, in classical mechanics) that the quantum nature of
the Heisenberg theory is manifested.
Of the two alternative forms of quantum mechanics—wave mechanics and matrix
mechanics—it is the former which lends itself more easily to the ·solution of a wide
variety of practical problems. For this reason, as well as because of the insight pro-
vided by the underlying wave picture, we will be dealing, for the most part, with
the Schrödinger version of quantum mechanics. However, the Heisenberg form has
its advantages, especially in considering formal questions. And in the quantum the-
ory of electromagnetic and other fields, the representation of dynamical variables
[e.g., the field quantities E(x,t) and, H(x,t)] by matrix operators as in Heisenberg’s
theory is almost an inescapable necessity.
28
W. Heisenberg, Z. phys., 33, 879, 1925. M. Born, W. Heisenberg and P. Jordan, Z. phys., 35, 557, 1925.
The equivalence of wave and matrix mechanics was shown by E. Schrödinger, ann. d. physik, 79, 734,
1926.
Exploring the Variety of Random
Documents with Different Content
sombra de tu augusto velo | Las artes viven en concierto amigo. »
Quint. Poes. A la paz (R. 19. 91). ' 3. adj.n) Que gusta mucho de
alguna cosa. Con de. « Tanto del oro te mostraste amiga, | Que
echaste á las espaldas mis pasiones, s Cerv. Gal. 3 (R. 1. 37'). « Era
de complexión recia, seco de carnes, enjuto de rostro, gran
madrugador y amigo de la caza. » Id. Quij. 1. 1 (II. 1.2571). « Era el
cura tan buen cristiano y tan amigo de la verdad, que no diría otra
cosa por todas las del mundo. » Id. ib. 1. (i 1 1!. 1. 2001). «
Siempre, hermano, fui amiga de la igualdad. » Id. ib. 2. 5 (B. 1.
414a). « Yo te perdono con que te enmiendes, y con que no te
muestres de aquí adelante tan amigo de tu interés. » Id. ib. 2. 28
(B. 1. 4651). « Acá tenemos noticia, buen Sancho, que sois tan
amigo de manjar blanco y de albondiguillas, que, si os sobran, las
guardáis en el seno para el otro día. »Id.¿6. 2.02 (R. 1. 5351). «
Quiera
The text on this page is estimated to be only 17.85%
accurate
Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.
ebookname.com