Instant Ebook Access, One Click Away – Begin at ebookgate.

com

Lectures on Light Nonlinear and Quantum Optics

using the Density Matrix 2ed. Edition Stephen C.
Rand

https://ebookgate.com/product/lectures-on-light-nonlinear-
and-quantum-optics-using-the-density-matrix-2ed-edition-
stephen-c-rand/

OR CLICK BUTTON

DOWLOAD EBOOK

Get Instant Ebook Downloads – Browse at https://ebookgate.com

Click here to visit ebookgate.com and download ebook now
Instant digital products (PDF, ePub, MOBI) available
Download now and explore formats that suit you...

Quantum optics Raymond Chiao

https://ebookgate.com/product/quantum-optics-raymond-chiao/

ebookgate.com

Introduction to Optics Lectures in Optics Volume 1 George

Asimellis

https://ebookgate.com/product/introduction-to-optics-lectures-in-
optics-volume-1-george-asimellis/

ebookgate.com

Lectures on SL2 C Modules Voldymyr Mazorchuk

https://ebookgate.com/product/lectures-on-sl2-c-modules-voldymyr-
mazorchuk/

ebookgate.com

Applications of Nonlinear Fiber Optics Second Edition

Optics and Photonics Series Govind Agrawal

https://ebookgate.com/product/applications-of-nonlinear-fiber-optics-
second-edition-optics-and-photonics-series-govind-agrawal/

ebookgate.com
Applications of Nonlinear Fiber Optics Optics and
Photonics Series 1st Edition Govind Agrawal

https://ebookgate.com/product/applications-of-nonlinear-fiber-optics-
optics-and-photonics-series-1st-edition-govind-agrawal/

ebookgate.com

Nonlinear Fiber Optics 6th Edition Govind P. Agrawal

https://ebookgate.com/product/nonlinear-fiber-optics-6th-edition-
govind-p-agrawal/

ebookgate.com

Thoughts On Design Paul Rand

https://ebookgate.com/product/thoughts-on-design-paul-rand/

ebookgate.com

Nonlinear Optics Phenomena Materials and Devices 1st

Edition George I. Stegeman

https://ebookgate.com/product/nonlinear-optics-phenomena-materials-
and-devices-1st-edition-george-i-stegeman/

ebookgate.com

Quantum Optics An Introduction 1st Edition Mark Fox

https://ebookgate.com/product/quantum-optics-an-introduction-1st-
edition-mark-fox/

ebookgate.com
LECTURES ON LIGHT
 Lectures on Light
Nonlinear and Quantum Optics using the Density Matrix

Second Edition

Stephen C. Rand
University of Michigan, USA

3
 3
Great Clarendon Street, Oxford, OX2 6DP,
United Kingdom
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide. Oxford is a registered trade mark of
Oxford University Press in the UK and in certain other countries
© Stephen C. Rand 2016
The moral rights of the author have been asserted
First Edition published in 2010
Second Edition published in 2016
Impression: 1
All rights reserved. No part of this publication may be reproduced, stored in
a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by licence or under terms agreed with the appropriate reprographics
rights organization. Enquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above
You must not circulate this work in any other form
and you must impose this same condition on any acquirer
Published in the United States of America by Oxford University Press
198 Madison Avenue, New York, NY 10016, United States of America
British Library Cataloguing in Publication Data
Data available
Library of Congress Control Number: 2016930235
ISBN 978–0–19–875745–0
Printed and bound by
CPI Group (UK) Ltd, Croydon, CR0 4YY
Preface to the First Edition

This book, and the course from which it sprang, attempts to bridge the enormous gap
between introductory quantum mechanics and the research front of modern optics and
other fields that make use of light. This would be an impossibly daunting task, were
it not for the fact that most of us love to hear things again and again that we already
know. Taking that into account, this book uses a single approach repeatedly to tackle
progressively more exciting topics in the science of light, moving systematically and
swiftly from very basic concepts to sophisticated topics.
The reader should be aware from the outset that the approach taken here is uncon-
ventional. It is highly selective instead of encyclopedic, and it teaches the reader only
how to use the density matrix on new problems. Nowadays scientific answers are being
sought to an ever-expanding array of problems across numerous disciplines. The trend
in textbooks on quantum optics is understandably to cover an increasing number of top-
ics comprehensively, and to familiarize students with an ever-widening array of analytic
tools, exhaustively. There are many fine texts that fulfill this function, but this book is
not one of them. Instead, important topics and alternative methods of analysis have been
omitted here to keep it as brief as possible, with a single-minded pedagogical purpose
in mind. The main objective of the book is to provide students and researchers with
one reliable tool, and the confidence that comes from practice, to analyze new optical
phenomena in their chosen field successfully and rigorously.
Thus, the one and only analytic tool developed here for attacking research-level
problems in optical science is the density matrix. A systematic procedure is applied to
representative problems of “critical subjects”—usually only one example each—to show
how virtually any problem can be analyzed with the density matrix. Each successive
example adds one system property at a time, with the result that one qualitatively new
feature appears in the dynamics each time. By using a systematic “building-up” prin-
ciple to approach complicated interactions, students begin to recognize what terms in
the analysis are associated with particular changes in the dynamics. Following two slow-
paced introductory chapters on review material, the text shifts focus to the development
of insights as to when atomic motion, or multi-level structure, or coherence effects dom-
inate the behavior of complicated systems. It ends with fast-paced coverage of selected
topics and applications.
The organization of the book follows the original sequence of lectures on light pre-
pared for applied physics graduates with typical undergraduate physics, chemistry and
engineering preparation, expected to handle interdisciplinary research topics during
their careers. Present day graduate students who use light often face important prob-
lems that no longer fall into the neat categories of unique models from the early days of
quantum mechanics, and require broad perspective and reliable mathematical tools to
vi Preface to the First Edition

handle new cross-disciplinary topics quickly. This course therefore embraces not only
students from traditional subject areas that make use of light (physics, chemistry, elec-
trical engineering, materials science), but also the biophysicist who needs laser tweezers,
the photochemist who wants coherent control, the biomedical engineer who needs to
image through scattering media, the mechanical engineer interested in molecular design
of new materials, and others. The greatest theoretical challenge faced by most students
is to make appropriate connections between standard models and the bewildering land-
scape of new research questions on intersecting boundaries of “hyphenated” subjects
like biophysics, biomedicine, photochemistry, etc. For them, the systematic progression
of Lectures on Light offers an approach to quantum optical analysis that should help
bridge the gaps.
Why choose the density matrix? After all, there are many mathematical tools available
to treat nonlinear and quantum optics. The answer is that the density matrix has features
that make it a natural choice. For example it permits one to ignore parts of a problem
that appear to be irrelevant and to focus mathematically on the dynamics of interest to
the researcher. Also, if desired, it can be reduced to a rate equation treatment—a familiar
approach to analysis that all students of science encounter. In addition, it is particularly
well suited for dealing with coherence in isolated or interactive systems.
This makes it an excellent point of departure for anyone who wishes to use light either
to probe or control systems about which little is known. By focusing on this adaptable
tool, readers can cover a lot of intellectual ground with the minimum investment in
mathematical complexity. What emerges is a reliable analytic framework for use in any
research where light probes or controls or alters a system, regardless of whether the
problems are classical or quantum.
Another hurdle that is addressed explicitly in this book, and that is very important,
is the persistent issue of whether simplified models provide reliable representations of
complicated systems. That is why the last part of Chapter 3 examines in detail the ques-
tion of whether sodium atoms can ever legitimately be viewed as two-level systems given
the complexity of their energy level structure. It turns out the experimentalist is much
more in control of the effective number of levels than one might guess. By the end of
this course, graduate students in interdisciplinary science are able to exercise consider-
able judgment in the creation of useful models for their own frontier research problems
and analyze them by drawing on solid examples and the explicit methodology of the text.
Some familiarity with introductory quantum mechanics is assumed. However, ad-
vanced preparation in optics is not essential to learn and use this material. Over the years,
students from disciplines as diverse as the ones mentioned above—mechanical engi-
neering, materials science, electrical, and biomedical engineering together with physics,
applied physics, and chemistry—have found it to provide the essential insights and
analysis they need for immediate application in their research. Some material that is or-
dinarily omitted from advanced quantum mechanics texts is included to set the stage for
the broadest possible applicability. An example of this is third order perturbation theory.
This topic provides an important bridge to understanding nonlinear behavior of even
the simplest quantum mechanical systems. Since self-saturation effects, cross-saturation,
four-wave mixing processes, and other third order phenomena are encountered much
 Preface to the First Edition vii

more commonly in pump–probe experiments than one might expect, it is important that
readers become familiar with third order effects early in the course of their research. An-
other unusual feature of this book is that a few procedural errors are presented early on
to illustrate what can go wrong when quantum mechanical calculations are formulated
inconsistently. These humbling examples remind us all that in research, and most espe-
cially in cross-disciplinary work, there is no substitute for the use of common sense and
no shortcut to pioneering science.
This book progresses as rapidly as possible from a simple and easy review to challeng-
ing modern applications. One layer of conceptual or computational complexity is added
in each new section. The technical material begins with some uncommon examples of
introductory quantum mechanics that force students from the start to revisit the basic
physical principles of optics and quantum mechanics studied as undergraduates. Poten-
tial pitfalls are also pointed out that arise from the inclusion of relaxation processes in
system dynamics. Once the density matrix approach to dynamics is motivated, formu-
lated and understood, the course progresses at an accelerated rate through important
applications. Hence the material is best studied in sequence.
This lecture material will be most useful to students interested in acquiring rigorous,
broadly-applicable analysis quickly. However, the systematic application of one mathe-
matical tool to many forefront topics in nonlinear and quantum optics will be of interest
to seasoned researchers as well. The heavy reliance in late chapters on the insights
and dynamic effects described in earlier chapters helps to keep the treatment short.
Much of the course relies on the semi-classical description in which only the atoms are
quantized—the light field is not. This is intended to encourage intuitive thinking to as
late a stage as possible. However, in the last two chapters several topics are covered where
both the atoms and the light field are quantized and intuitive notions are sometimes poor
guides. Finally, the selected research topics of Chapter 7 not only illustrate the power of
systematic density matrix analysis but give students confidence that, as they approach
the exciting frontiers of their own research, the combination of density matrix analysis
and common sense perspectives developed throughout this course will facilitate success.
I would like to acknowledge all the help I have received during various stages of prep-
aration of this monograph. First and foremost, I am grateful for the comments and
questions of students who took this course over a period of two decades. Along with
my own graduate students, they helped to make the presentation compact by forcing
me to provide concise answers about confusing notions. I am indebted to Philbrick
Bridgess of Roxbury Latin School for imparting to me his respect for analytic geom-
etry, which ultimately led to the discovery of transverse optical magnetism, covered in
Section 7.4 of this book. On a few topics I have drawn liberally from existing texts,
but most especially from Elements of Quantum Optics by Meystre and Sargent, Quantum
Electronics by Yariv, Quantum Optics by Zhubary and Scully, Laser Physics by Sargent,
Scully, and Lamb, Optical Resonance and Two-Level Atoms by Allen and Eberly, and
Foundations of Laser Spectroscopy by Stenholm. I am thankful for their fine examples
of concise pedagogy. Also I thank my colleagues at the University of Michigan for cre-
ating the intellectual environment that made this book possible. Support was provided
by the Department of Electrical Engineering and Computer Science for typing of a
viii Preface to the First Edition

rough draft by Ruby Sowards, Nick Taylor, and Susan Charnley. The graduate student
course itself was offered through the Department of Physics. I owe a debt of gratitude
to Kevin Rand for preparing many original illustrations and for the adaptation of pub-
lished figures. The cover diagram was furnished by William Fisher. I am particularly
grateful to Boris Stoicheff, Richard Brewer, Art Schawlow, Ted Hansch, and Juan Lam
for their friendship and for sharing what they knew. Their examples were inspirational.
Finally, I am deeply indebted to my family—especially my wife Paula who patiently
endured the taxing process of finalizing the manuscript.
S.C. Rand
February 27, 2010
Preface to the Second Edition

In the years following the first edition of this book, some topics have been added in
response to student interest in research advances. These have now been incorporated
in the new edition, making it more comprehensive in its coverage of advanced research.
However by continuing to emphasize compact descriptions it has been possible to keep
the book close to its original length even though the number of problems at the ends
of chapters has doubled. Many students and colleagues have provided suggestions and
corrections to improve the presentation and I gratefully acknowledge their input. I would
especially like to thank Hope Wilson, Alex Fisher, Hamed Razavi, and Austin Tai for
help preparing new figures in Chapters 5, 6, and 7.
S.C. Rand
October 29, 2015
Contents

1 Basic Classical Concepts 1

1.1 Introduction 1
1.2 Electric and Magnetic Interactions 3
1.2.1 Classical Electromagnetism 3
1.2.2 Maxwell’s Equations 3
1.2.3 The Wave Equation 4
1.2.4 Absorption and Dispersion 5
1.2.5 Resonant Response 6
1.2.6 The Vectorial Character of Light 7
Supplementary Reading 9
2 Basic Quantum Mechanics 10
2.1 Particles and Waves 10
2.2 Quantum Observables 12
2.2.1 Calculation of Quantum Observables 12
2.2.2 Time Development 13
2.2.3 Symmetry 15
2.2.4 Examples of Simple Quantum Systems 17
2.3 Dynamics of Two-Level Systems 23
2.4 Representations 26
2.4.1 Representations of Vector States and Operators 26
2.4.2 Equations of Motion in Different Representations 28
2.4.3 Matrix Representations of Operators 33
2.4.4 Changing Representations 35
References 37
Problems 37
3 Atom–Field Interactions 42
3.1 The Interaction Hamiltonian 42
3.2 Perturbation Theory 43
3.3 Exact Analysis 49
3.4 Preliminary Consideration of AC Stark or Rabi Splitting 51
3.5 Transition Rates 53
3.6 The Density Matrix 55
3.6.1 Electric Dipole Transition Moments 56
3.6.2 Pure Case Density Matrix 56
3.6.3 Mixed Case Density Matrix 58
xii Contents

3.7 Decay Phenomena 60

3.8 Bloch Equations 63
3.9 Inhomogeneous Broadening, Polarization, and Signal Fields 67
3.10 Homogeneous Line Broadening through Relaxation 70
3.11 Two-Level Atoms Versus Real Atoms 72
References 76
Problems 76
4 Transient Optical Response 83
4.1 Optical Nutation 83
4.1.1 Optical Nutation without Damping 83
4.1.2 Optical Nutation with Damping 87
4.2 Free Induction Decay 87
4.3 Photon Echoes 92
4.3.1 Algebraic Echo Analysis 93
4.3.2 Rotation Matrix Analysis 98
4.3.3 Density Matrix Operator Analysis 99
References 105
Problems 105
5 Coherent Interactions of Fields with Atoms 111
5.1 Stationary Atoms 111
5.1.1 Stationary Two-Level Atoms in a Traveling Wave 111
5.1.2 Stationary Three-Level Atoms in a Traveling Wave 114
5.1.3 Stationary Two-Level Atoms in a Standing Wave 117
5.2 Moving Atoms 120
5.2.1 Moving Atoms in a Traveling Wave 120
5.2.2 Moving Atoms in a Standing Wave 124
5.3 Tri-Level Coherence 129
5.3.1 Two-Photon Coherence 129
5.3.2 Zeeman Coherence 132
5.4 Coherent Multiple Field Interactions 138
5.4.1 Four-Wave Mixing 138
5.4.2 Pump–Probe Experiments 143
5.4.3 Quantum Interference 149
5.4.4 Higher Order Interactions and Feynman Diagrams 155
References 159
Problems 160
6 Quantized Fields and Coherent States 168
6.1 Quantization of the Electromagnetic Field 168
6.2 Spontaneous Emission 178
6.3 Weisskopf–Wigner Theory 181
 Contents xiii

6.4 Coherent States 185

6.5 Statistics 193
6.5.1 Classical Statistics of Light 193
6.5.2 Quantum Statistics of Light 197
6.6 Quantized Reservoir Theory 201
6.6.1 The Reduced Density Matrix 201
6.6.2 Application of the Reduced Density Matrix 205
6.7 Resonance Fluorescence 209
6.7.1 Fluorescence of Strongly Driven Atoms 210
6.7.2 Coherence of Strongly Driven Two-Level Atoms 218
6.8 Dressed Atom Theory 220
6.8.1 Strong Coupling of Atoms to the Electromagnetic Field 220
6.8.2 Dressed State Population Dynamics 224
References 228
Problems 229
7 Selected Topics and Applications 240
7.1 Mechanical Effects of Light and Laser Cooling 240
7.1.1 Radiation Pressure, Dipole Forces, and Optical Tweezers 240
7.1.2 Laser Cooling via the Doppler Shift 243
7.1.3 Magneto-Optic Trapping 246
7.1.4 Laser Cooling below the Doppler Limit 247
7.2 Dark States and Population Trapping 253
7.2.1 Velocity-Selective Coherent Population Trapping 253
7.2.2 Laser Cooling via VSCPT 257
7.3 Coherent Population Transfer 260
7.3.1 Rapid Adiabatic Passage 260
7.3.2 Laser Cooling of Solids 265
7.4 Coherent Transverse Optical Magnetism 280
7.5 Electromagnetically Induced Transparency 290
7.6 Squeezed Light 294
7.7 Cavity Quantum Electrodynamics 298
7.7.1 Damping of an Optical Field by Two-Level Atoms 298
7.7.2 Weak Coupling Regime 300
7.7.3 Strong Coupling Regime 304
7.8 Quantum Information Processing 307
7.8.1 Introduction 307
7.8.2 Classical Logic Circuits and Computation 308
7.8.3 Quantum Bits and Quantum Logic Gates 309
7.8.4 Realization of Quantum Gates 312
7.8.5 Fidelity of Gate Operations 316
References 316
Problems 320
xiv Contents

Appendices
Appendix A Expectation Values 331
Appendix B The Heisenberg Uncertainty Principle 332
Appendix C The Classical Hamiltonian of Electromagnetic Interactions 335
Appendix D Stationary and Time-Dependent Perturbation Theory 338
Appendix E Second Quantization of Fermions 345
Appendix F Frequency Shifts and Decay Due to Reservoir Coupling 349
Appendix G Solving for Off-Diagonal Density Matrix Elements 353
Appendix H Irreducible Spherical Tensor Operators and the
Wigner–Eckart Theorem 356
Appendix I Derivation of Effective Hamiltonians 368
Appendix J Irreducible Representation of Magnetic Dipole Interactions 370

Index 373
1
Basic Classical Concepts

1.1 Introduction
In classical physics, light interacts with matter as described by a compact set of equations
formulated by James Clerk Maxwell in 1865, namely Maxwell’s equations. Solutions for
the propagation of electromagnetic fields in arbitrary media can generally be built up
from solutions to these equations for individual frequency components, and are particu-
larly easy to find if the fields have slowly varying electric and magnetic field magnitudes.
The medium through which light passes must also be uniform and the timescale of inter-
est must greatly exceed the optical period. Fortunately, these conditions are not terribly
restrictive. They encompass most (though not all) distinctive phenomena in classical as
well as quantum optics. Perhaps more importantly, they form a useful framework for
lectures on light without unnecessarily limiting our horizons.
Nowadays, the optical characteristics of media can often be engineered to enhance
specific interactions deliberately. For example, quasi-phase-matching crystals can be
prepared with periodically inverted domain structure to allow the build-up of new fre-
quency components in the optical field. Similarly, in so-called metamaterials, deliberate
variation in microstructure from point to point within a medium is intended to permit
light fields to evolve in complicated but controllable ways. It has been shown in recent
years that metamaterials can be designed to distort the way electromagnetic waves move
through an occupied region of space, in such a way as to render objects located there
effectively invisible. Yet even the passage of light through these non-uniform media is
again entirely predictable using Maxwell’s equations and modern computational tools.
The relative maturity of the subject of electromagnetism and the power of modern com-
puters have put us in a position to predict in great detail how light moves, and how it is
attenuated, emitted, amplified, or scattered as it progresses through practically any kind
of matter. So why is it that light is still such a vital topic today, and when do we have to
treat problems quantum mechanically? How can it be that so many new marvels have
emerged from the study of electromagnetism and optical science in the last decade or
two? The pace of major discoveries continues unabated. How can light tantalize and sur-
prise experts in the twenty-first century with an ever-expanding landscape of discovery
and applications at a time when Maxwell’s equations are 150 years old?

Lectures on Light. Second Edition. Stephen C. Rand.

© Stephen C. Rand 2016. Published in 2016 by Oxford University Press.
2 Basic Classical Concepts

In addition to being a manual for applications of the density matrix, this book seeks
to answer this question. One of its goals is to teach students how to explore and analyze
the unknown without already knowing everything. For this journey, the density matrix
is a perfect companion, since one finds that with it the wavefunction of complex sys-
tems is no longer needed to calculate most things of interest in optical interactions. The
content of this course has been presented now for two decades as an advanced lecture
course on light for graduate students at the University of Michigan—primarily those
with backgrounds in physics, chemistry, and engineering. It seeks to give students fa-
miliarity with a single analytic tool, the density matrix, by applying it systematically to
a great many forefront problems in modern optics. It presents a concise, broad (though
admittedly incomplete) picture of active research fronts in optical science that students
can absorb in a single semester. It does not pretend to be a comprehensive reference
for research in any specialty area from which one or more examples may have been
drawn. Instead, it shows students how to get started on virtually any research problem
using a standard toolbox, and gives them the requisite perspective on what is physically
possible and essential in the analysis. Students acquire a sense for when classical, semi-
classical, or quantum optical approaches need to be applied, when coherence plays an
important role and when it does not, when an exact solution is required and when it is
not, through an approach that adds one concept at a time systematically with a single
mathematical tool.
Among the more advanced topics that are included in this course are free induction
decay, photon echoes, nutation, spectral and spatial hole-burning, light shifts, two- and
three-level coherence, Zeeman coherence, coherent population transfer, electromagnet-
ically induced transparency, slow light, high-order perturbation theory, laser cooling,
optical magnetism, squeezed light, dressed atoms, quantum computation, and cavity
quantum electrodynamics. Although some of these subjects are typically omitted from
standard textbooks on light, experience has shown they can be handled by intermediate
or beginning graduate students when the progression through these topics is presented
in sufficiently small steps. Moreover, this process succeeds in providing a framework
for understanding optical phenomena in a way that is quite different from that of books
that emphasize a sequence of different mathematical techniques to handle quantum me-
chanics. For this reason, most of the book utilizes a “semi-classical” approach that treats
the system under study as a quantum system but avoids the introduction of operators to
describe the electromagnetic field itself. This has the merit of postponing operator math-
ematics required for the quantization of the electromagnetic field until the final chapters
when they are really needed.
Analysis of many forefront topics in optics with a single approach gives readers con-
fidence that they can proceed into virtually any developing field where light is used to
probe or control dynamics, and confidently formulate an initial attack on their research
problem. Naturally it also has the disadvantage of being just one approach. Many of us
are familiar with the mathematical handicap that can result from a bad choice of co-
ordinate system, or an ill-suited choice of variables. However, the density matrix is a
remarkably complete and forgiving tool that accomplishes the essential things. First, it
incorporates the all-important phases of fields and polarizations responsible for some
 Electric and Magnetic Interactions 3

of the surprising phenomena encountered in optical science. It can describe dephasing,

coherent control, and relaxation processes in a manner consistent with the occupation of
various states of the system. Second, it eliminates the need for detailed wavefunctions in
the prediction of most of the important dynamics in new systems. In beginning courses
in quantum mechanics we are taught that the wavefunction contains all the details of
the quantum system and is essential for understanding or predicting its behavior. Un-
fortunately, solutions are available for wavefunctions of only the simplest systems like
atomic hydrogen. Consequently it is fortunate, to say the least, that the density ma-
trix provides a method for computing all the “interesting” dynamics of new systems
given only limited information on energy levels and symmetries. One can even ignore
parts of multi-component systems through the use of the reduced density matrix. In
this way it becomes possible to analyze complicated systems like macromolecules, for
which analytic wavefunctions are not likely to be available at any time in the foreseeable
future.

1.2 Electric and Magnetic Interactions

1.2.1 Classical Electromagnetism
The interaction of electromagnetic waves with matter is more complex at optical fre-
quencies than at radio or microwave frequencies. There are innumerable possibilities
for resonance in the optical range, which do not exist at frequencies below 10 GHz, for
example. Such resonances result in large changes to absorption, dispersion, and scatter-
ing when only small changes in frequency are made. Also, relatively large variations of
constitutive parameters, namely permittivity ε and permeability μ, over the frequency
ranges between resonances can be exploited to cause energy exchange between waves of
different frequencies. This emphasizes the importance of finding a general approach to
optical analysis that incorporates resonant, dispersive, and quantum mechanical charac-
ter of atomic and molecular interactions with light. So we shall develop a formalism that
combines Maxwell’s equations and quantum properties of matter from the outset, and
seek perspective by applying it systematically to a variety of problems.

1.2.2 Maxwell’s Equations

The fundamental equations relating time-varying electric and magnetic fields are:

∂ D̄
∇¯ × H̄ = J¯ + , (1.2.1)
∂t
∂ B̄
∇¯ × Ē = – , (1.2.2)
∂t
∇¯ · D̄ = ρv , (1.2.3)

∇¯ · B̄ = 0. (1.2.4)
4 Basic Classical Concepts

Constitutive relations describe the response of charges to applied electric or magnetic

fields in real materials. The displacement field D̄ in Maxwell’s equations is

D̄ = εĒ = ε0 Ē + P̄, (1.2.5)

and the magnetic flux density in Maxwell’s equations is

 
B̄ = μH̄ = μ0 H̄ + M̄ . (1.2.6)

At optical frequencies or in non-magnetic systems, we typically assume M̄ = 0, because

magnetic Lorentz forces are small compared to electric forces at high frequencies. In
(1) (1)
media with linear response the polarization is P̄ = ε0 χe Ē, χe being the linear electric
susceptibility. Thus, the displacement field is

D̄ = ε0 (1 + χe(1) )Ē, (1.2.7)

(1)
and the relative, linear dielectric constant εr = ε/ε0 is given by εr = 1 + χe .
In nonlinear or strongly excited media, there are higher order terms that intro-
(1) (2)
duce field-induced effects. That is, P = P (1) + P (2) + . . . = ε0 (χe E + χe E 2 + . . .).
We deal with nonlinear response when multi-photon transitions of isolated atoms are
considered in later chapters. Nonlinear response can also arise from the finite response
time of bound electrons or charge motion in the case of free carriers. Either mech-
anism can produce a “nonlocal” relationship between the polarization and the field.
That is, P(r̄, t) ≈ ε0 χe (t – t  , r̄ – r̄  )E(r̄  , t ). If we are interested only in slow dynamics in
dielectrics we can often ignore such effects. However, finite response times cannot be
ignored on ultrafast timescales in semiconductors, plasmas, or organic electronic mater-
ials. They also cannot be ignored in photorefractive media, where charges diffuse slowly
from place to place, or in multi-level media where long-lived states lead to time-delayed
response.

1.2.3 The Wave Equation

The key equation describing propagation of classical light is the wave equation, ob-
tained by applying the curl operator to combine Eqs. (1.2.1) and (1.2.2). Consistent
with Panofsky’s expression for ∇¯ × B̄ (see Supplementary Reading list) we find:
   
∂ B̄ ∂ ∂ D̄  
∇¯ × ∇¯ × Ē = –∇¯ × =– μ0 J¯ + + μ0 ∇¯ × M̄
∂t ∂t ∂t

∂ J¯ ∂ 2E ∂ 2 P̄ ∂ M̄
= –μ0 – μ0 ε0 2 – μ0 2 – μ0 ∇¯ × . (1.2.8)
∂t ∂t ∂t ∂t
 Electric and Magnetic Interactions 5

If we restrict ourselves to non-magnetic, insulating materials (M̄ = 0; J¯ = 0), then we

can use a vector identity in Eq. (1.2.8) to find

  ∇¯ 
∇¯ × ∇¯ × Ē = –∇ 2 Ē + ∇¯ ∇¯ · Ē = –∇ 2 E + ∇¯ · D̄ – ∇¯ · P̄ . (1.2.9)
ε

Provided that the local free charge density ρv is zero and there are no spatial variations
of polarization due to charge migration or field gradients, we find ∇¯ · D̄ = ∇¯ · P̄ = 0 and
in free space the wave equation becomes

1 ∂ 2 Ē ∂ 2 P̄
∇ 2 Ē – = μ 0 , (1.2.10)
c2 ∂t2 ∂t2

where c ≡ (μ0 ε0 )–1/2 is the speed of light in vacuum.

1.2.4 Absorption and Dispersion

Solutions of Eq. (1.2.10) are useful in describing many important classical phenomena
in linear dielectrics (P = P (1) ), such as absorption and dispersion. As an example of such
a solution, consider a linearly polarized plane wave E(z, t) propagating along ẑ. If the
wave is polarized along x̂, meaning that the vector E points along the direction x, this
wave can be written in the form

1
Ē(z, t) = E0x (z)x̂ exp[i(kz – ωt)] + c.c., (1.2.11)
2

where c.c. stands for complex conjugate. Other possible polarization states are con-
sidered in Section 1.2.6. Here however we show, by substituting Eq. (1.2.11) into
Eq. (1.2.10), that the wave equation can often be written in a scalar form that ignores
the vector character of light. After substitution, the orientation of the field emerges as a
common factor, which may therefore simply be dropped from the equation. In this way,
we find

∂ 2 E0x (z) ∂E0x (z) ω2 ω2

+ 2ik – k 2
E0x (z) + E0x (z) = – χe E0x (z). (1.2.12)
∂z2 ∂z c2 c2

An important mathematical simplification can be introduced at this point, called the

slowly varying envelope approximation (SVEA), by assuming that the amplitude E0x (z)
varies slowly over distance scales comparable to the wavelength. This permits ne-
glect of small terms like ∂ 2 E0 (z)/∂z2 in Eq. (1.2.12), whereupon the wave equation
reduces to

∂E0x (z) ω2 ω2
2ik – k2 – 2 E0x (z) = – χe E0x (z). (1.2.13)
∂z c c2
6 Basic Classical Concepts

Recognizing that the electric susceptibility χe can be complex (χe = χ  + iχ  ), we can

equate the real parts of Eq. (1.2.13) to find

ω2  
k2 = 2
1 + χ . (1.2.14)
c
This relationship between frequency ω and wavenumber k is known as the linear
dispersion relation. It is a defining relationship for light, usually written as

ω = cn k, (1.2.15)

where cn (ω) = c/n(ω) is the phase velocity of the wave in which c is the speed of light in
vacuum and n is the refractive index given by n2 = 1 + χ  . Equating the imaginary parts
of Eq. (1.2.13), one finds

∂E0x (z) ω2
2k = – 2 χ  E0x (z). (1.2.16)
∂z c
∗
Multiplication of both sides of Eq. (1.2.16) by the conjugate field amplitude E0x (z) yields

∂|E0 |2 ω2 χ 
=– 2 |E0 |2 , (1.2.17)
∂z c k
and thus
∂I
= –αI , (1.2.18)
∂z
where I is the optical intensity and α is the absorption coefficient, defined by

ω2 χ 
α≡ . (1.2.19)
c2 k
In vacuum this result is just α = kχ  . The solution of Eq. (1.2.18) can be written as

I (z) = I0 exp[–αz], (1.2.20)

which is Beer’s law. When α > 0, the electromagnetic wave undergoes exponential ab-
sorption (loss). When α < 0, there is exponential amplification (gain). Beer’s law applies
to optical propagation in many systems. Exceptions include systems where radiation
trapping, saturation, or multiple scattering take place.

1.2.5 Resonant Response

So far, we have characterized the electric response by introducing a polarization P that
depends on the incident field. The proportionality constant is the classical macroscopic
susceptibility χ, and we find that we can describe some well-known propagation effects
 Electric and Magnetic Interactions 7

such as the dependence of the phase velocity cn on refractive index and the exponential
nature of absorption/gain in simple materials. However, to this point we have no model
or fundamental theory for the electric susceptibility χ itself.
For this purpose we turn to a classical harmonic oscillator model based on linear re-
sponse (and Hooke’s law). The electron and proton comprising a fictitious atom are
joined by a mechanical spring with a restoring force constant k0 . It is assumed the pro-
ton does not move significantly in response to the applied field (Born approximation),
whereas the displacement x(t) about equilibrium of the bound electron depends on the
applied field according to F̄ = –eĒ = –k0 x̄. If Ē is harmonic in time, solutions will have
the form x(t) = x0 exp(–iωt). Upon substitution into Newton’s second law

∂ 2 x(t) ∂x(t)
me – me γ + k0 x(t) = –eE0x exp(–iωt), (1.2.21)
∂t2 ∂t
one can solve for the amplitude x(t) of driven charge motion, which is given by

eE0x /me
x0 = , (1.2.22)
ω2 – iωγ – ω02
√
where ω0 ≡ k0 /me is the resonant frequency of the oscillator and γ is an empirical
damping constant. The total polarization is

N
P(t) = – ex(t) = –Nex(t), (1.2.23)
1

for N identically prepared atoms. A comparison with our former expression P = ε0 χ E

determines the frequency-dependent susceptibility components, and through them the
absorption and dispersion curves shown in Figure 1.1.
Although this Lorentz model is strictly empirical, it provides a qualitatively useful
picture of the frequency dependence of system response near electronic resonances.
Regrettably it does not provide a way to derive the spring and damping constants from
first principles or to explain the stability of atoms. We turn to quantum mechanics to
remedy such deficiencies.

1.2.6 The Vectorial Character of Light

The wave equation (Eq. (1.2.10)) is a vector relation. Often, as shown in Section 1.2.4,
it can be reduced to a scalar relation because the vector character of light plays no role
in the particular atom–field interaction of interest. However, this is certainly not always
true, and it is helpful to identify what circumstances require full vector analysis. In later
chapters, a few topics such as Zeeman coherence and transverse optical magnetism are
covered that are strongly governed by vectorial aspects of light. In anticipation of these
subjects, we close this introductory chapter by describing the basic polarization states of
light that are possible and the angular momentum carried by them.
8 Basic Classical Concepts

1.8
1.6

1.4

1.2

1.0
n

0.8

0.6

0.4

0.2

0.0
–10 –5 0 5 10
(ω0 – ω)/Γ

Figure 1.1 Frequency dependence of absorption (solid curve) and dispersion (refractive index; dashed
curve) near a resonance at frequency ω0 in the classical model.

In Section 1.2.4, the polarization of light was introduced as a property of light specify-
ing the direction in which the optical electric field is oriented. Since the field orientation
may vary in time, this definition still admits three fundamentally different states of po-
larization which are called linear, right circular, and left circular. Linear polarization
describes light waves in which the electric field points in a fixed direction. For exam-
ple, x-polarized light propagating along ẑ has an E field oriented along the unit vector x̂.
Hence it may be written in the form Ē(z, t) = Ex (z, t)x̂ or Ē(z, t) = E(z, t)eˆ if we let ê rep-
resent a more general polarization unit vector. Circular polarization describes light with
an electric field vector that executes a circle around the propagation axis. The electric
field rotates once per cycle, either clockwise or counterclockwise, as the wave propagates
forward. The circular basis vectors for these polarizations are
√ 
ε̂± = ∓ 1/ 2 (x̂ ± i ŷ). (1.2.24)

In the case of circular polarization, the vector ê = ε̂ ± is an axial rather than a polar vector.
The 90◦ difference in phase of oscillations along x̂ and ŷ in Eq. (1.2.24) results in fields
of the form Ē(z, t) = Ex (z, t)ε̂± that do not point in a fixed direction. Instead, they
rotate as the wave propagates. Circular and linear states of polarization are illustrated in
Figure 1.2.
Circular polarizations carry spin angular momentum of ±h̄, as indicated in Figure 1.2.
The field direction and the energy density associated with the field twists as it moves,
undergoing a helical rotation with respect to ẑ. Linear polarization carries no angular
momentum, since the electric field vector does not rotate about ẑ at all. This latter state
of polarization is a special combination or superposition state, consisting of the sum of
two equal-amplitude, circularly polarized fields of opposite helicity.
 Supplementary Reading 9

–ћ σ–

Lin
ear

σ+
+ћ

Figure 1.2 Fundamental polarizations of light, illustrating light fields that carry spin angular momenta
of –h̄, 0, and +h̄ (top to bottom, respectively). Angular momentum carried by a light wave affects the way
it interacts with matter.

Exercise: (a) Show that light which is linearly polarized along x̂ can be written as the
sum of two opposite circularly polarized fields. That is, show from Eq. (1.2.24)
that x-polarized light is a phased, superposition state of two equal amplitude right
and left circularly polarized components:
√ 
x̂ = – 1/ 2 (ε̂+ – ε̂ – ). (1.2.25)

(b) Conversely, show that circular polarization can be expressed as a superposition

of two orthogonal, linearly polarized waves of equal amplitude.

Since momentum must be conserved in optical interactions, just as energy must be con-
served, we can anticipate that any angular momentum carried by light may affect the
way it interacts with materials. This point is addressed in later chapters.

..........................................................................................................

S U P P L E M E N TA RY R E A D I N G

G.R. Fowles, Introduction to Modern Optics. New York: Dover, 1987.

W.K.H. Panofsky and M. Phillips, Classical Electricity and Magnetism, 2nd ed. London: Addison-
Wesley Publishing Co., 1962.
2
Basic Quantum Mechanics

2.1 Particles and Waves

Quantum mechanics postulates a concept known as wave–particle duality in the
description of dynamical systems by associating a de Broglie wavelength λB with each
particle. This wavelength depends on the linear momentum p of the particle, and is
given by
λB = h/p, (2.1.1)
where h is Planck’s constant. The de Broglie wave causes particles to exhibit wave in-
terference effects whenever the wavelength becomes comparable to or larger than the
space occupied by the wave. In fact de Broglie derived the Bohr–Sommerfeld quantiza-
tion rules from Eq. (2.1.1). Remarkably, this is the only essentially new idea of “wave”
mechanics that is missing in “classical” mechanics, and curiously, although matter can
exhibit particle or wave-like properties, its particle and wave characteristics are never
observed together.
It follows specifically from Eq. (2.1.1) that electron waves can form self-consistent
standing wave patterns (“stationary” states, “orbitals,” or “eigenstates”) that have
constant angular momentum and energy. Charge distributions can be simultaneously
localized and stable. The forces from Coulomb attraction between an electron and a
nucleus, together with nuclear repulsion, must balance within atoms at equilibrium.
This balance and the requirement that the standing wave be self-consistent uniquely
determine the ground state of atoms. Excited states must similarly satisfy the boundary
conditions imposed by the de Broglie wavelength, but tend to be short-lived because the
forces on the electron are no longer balanced. Consequently excitation of such a state
is quickly followed by a transition to the ground state, accompanied by emission of the
energy difference between the ground and excited state as electromagnetic energy.
Conversely, light of frequency ν0 has particle-like properties and may cause resonant
transitions of atoms between well-defined initial and final states of an atom with ener-
gies Ei and Ef . To effect a transition between two specific states, light must supply the
energy difference, according to the formula
hν0 = Ef – Ei . (2.1.2)

Lectures on Light. Second Edition. Stephen C. Rand.

© Stephen C. Rand 2016. Published in 2016 by Oxford University Press.
 Particles and Waves 11

where ν0 is called the transition frequency. Individual transitions are rarely observable
because even small particles of bulk matter contain a great many atoms, and light of
even modest intensity I ∝ |E|2 consists of a high density (N/V = I /chν) of energy-bearing
particles (photons) which cause transitions at random times. So individual interactions
generally go unnoticed. Particle-like interactions of light waves are rarely observed un-
less intensities are extremely low. If a light field is pictured as a collection of quantum
particles of energy hν0 , then at low intensity the particles must be widely separated in
time, presuming they are all the same. We then require a description of light in terms
of mathematical operators that produce discrete changes of the field when they interact.
The particle nature of light is therefore an important factor determining the noise and
statistical correlation properties of weak fields. However, a great many important aspects
of light–matter interactions can be described adequately by treating only the atomic vari-
ables quantum mechanically as operators and the light field classically, with an amplitude
and phase that are continuous scalar functions. This approach, called the semi-classical
approach, is the one adopted throughout the first few chapters of this book. Aspects of
light–matter interactions that depend explicitly on the discrete character of the field are
reserved for the final chapters.
Charge motion caused by light is typically a dipolar response such as that determined
in our classical model (Eq. (1.2.21)). However, in systems of bound charges excited near
internal resonant frequencies, the dipole polarization that develops must clearly depend
on the stationary quantum states and the corresponding resonant frequencies. Hence
under these conditions, polarization of the medium becomes a quantum mechanical
observable determined to first order by non-zero values of the first moment of the mi-
croscopic polarization operator p̂. Hence we shall need a procedure for determining the
expected values of the dipole moment and other operators weighted by available states
of the atomic system (specified by the probability amplitude ψ of finding the system in
a given state at a given time and location) to predict the outcome of experimental meas-
urements. This predictor of repeated measurements is called the expectation value, and
it resembles an average weighted by the probability amplitude of the state of the system.
The most important physical observable considered throughout this book is the electric
dipole p̂ = –er̄, which has an expectation value given by

 
p̂ = – ψ|er¯ |ψ ≡ –e ψ ∗ r̄ψdV . (2.1.3)

Here the wavefunction ψ(r, t) accounts for the proportion of eigenstates contributing to
the actual state of the system. (See Section 2.2 and Appendix A for a more complete dis-
cussion of the definition of expectation value.) The bra-ket symbols used in Eq. (2.1.3)
were introduced by Dirac to represent not only the conjugate state functions ψ ↔ |ψ
and ψ ∗ ↔ ψ| respectively but also to simplify spatial integrals where |ψ and ψ| appear
in combination, as in the dipole moment ψ|er̄ |ψ of Eq. (2.1.3). This notation provides
a convenient shorthand and will be used to shorten calculations throughout the book.
The total polarization for N identical atomic dipoles in the quantum mechanical
limit is obtained by summing individual contributions with an expression similar to
12 Basic Quantum Mechanics

Eq. (1.2.21). However, the notation now implicitly includes functions (wavefunctions)
on which the operator of interest must operate for the expression to have meaning:
N 
 
P̄ = – er̄ i . (2.1.4)
i=1

Dipoles may be static or time varying. When P̄ is time varying, it may have observable
effects even if its time average is zero, because the charge distribution oscillates and may
cause a change of state of the atom or lead to radiation. Indeed, the magnitude of the
dipole operator er̄ evaluated between different states of a single atom, molecule, or optical
center will be shown to be the dominant factor determining the rate of optically induced
transitions between quantum states. For a collection of atoms however, we shall find that
the relative phase variation of er¯i from one atom to the next also plays an important
role in determining the magnitude and temporal development of the ensemble-averaged
or macroscopic polarization P̄ in Eq. (2.1.4). The behavior of P̄ due to initial phas-
ing, dephasing, and even rephasing is important in coherent optical processes, and the
density matrix is especially well-suited to keep track of important phase information.

2.2 Quantum Observables

2.2.1 Calculation of Quantum Observables
Operators generate eigenvalues corresponding to specific energy states by their action on
components of a wavefunction ψ. However, all systems, particularly systems undergoing
transition, contain components with a spread in energy and momentum. So for a system
described by a wavefunction ψ(r̄, t) that is in general not an eigenstate, measurement of
an observable O yields a simple average or first moment of O. This moment is weighted
by the states the system is in or passing through. The only quantum mechanical aspect
of this calculation is that O must be replaced by the corresponding operator Ô:

 
Ô = ψ|Ô|ψ = d 3 rψ ∗ Ôψ. (2.2.1)

This expression will then incorporate commutation properties, as discussed in

Appendix A.
Because quantum theory deals with wave-like properties of matter, its predictions re-
flect the delocalization of waves and include a certain amount of “inexactitude.” This
is not a defect of the theory, but merely reflects the difficulty inherent in answering the
question “Where is a wave?” Once a wave-based, probabilistic description of matter is
introduced, there is a minimum uncertainty associated with measurements. To be more
precise, conjugate variables cannot be determined simultaneously with infinite preci-
sion. For example, linear momentum px and position x are conjugate variables that have
uncertainties px and x that are mutually related.
px x ≥ h̄/2 (2.2.2)
 Quantum Observables 13

∆kx = k0∆ θ
k0 ∆x
z

Figure 2.1 Illustration of a typical uncertainty relationship between spatial resolution x and the cor-
responding angular spread that accompanies optical diffraction at an aperture according to the relation
px x = h̄k0 θ x ≥ h̄/2.

The exact form of this Heisenberg uncertainty relation is justified and generalized in
Appendix B for all pairs of conjugate variables. A simple example of the reciprocal re-
lationship between spatial and angular resolution governed by Eq. (2.2.2) is illustrated
by diffraction from an aperture (Figure 2.1), when a plane wave develops a spread of
propagation directions as it passes through a small aperture.
Measurements acquire minimum uncertainty when the product in Eq. (2.2.2) takes
on its minimum value. For the variables in the previous paragraph, the minimum prod-
uct is px x = h̄/2. When systems are prepared in minimum uncertainty or coherent
states they obey this equality, as discussed in Chapter 6. However, it will be shown that
despite the inescapable uncertainty associated with wave mechanics, the Heisenberg limit
expressed by Eq. (2.2.2) can be partly circumvented by preparing systems in special
states called coherent states.

2.2.2 Time Development

The wavefunction ψ(r̄, t) that we use to describe a material system gives the probability
amplitude for finding the particle at (r̄, t), and its square modulus |ψ(r̄, t)|2 yields the
probability density. Typically, experiments measure |ψ(r̄, t)|2 , but we note in passing
that it is possible to measure the eigenstates forming the basis of ψ(r̄, t) directly [2.1].
We expect to find the particle (with unit probability) if we search through all space.
Hence the wavefunction is assumed to be normalized according to

ψ ∗ (r̄, t)ψ(r̄, t)d 3 r = 1. (2.2.3)

Evolution of the wavefunction in time is described by Schrödinger’s wave equation

∂ψ
i h̄ = H ψ. (2.2.4)
∂t
14 Basic Quantum Mechanics

In classical mechanics, the Hamiltonian H describing system energy is expressible

in terms of canonically conjugate variables qi and pi of the motion. The operator
form of the Hamiltonian, designated by Ĥ , is obtained by a procedure that replaces
the Poisson bracket {qi , pj } in Hamilton’s classical equations of motion by a similar
bracket
 that keeps track of operator commutation properties, namely the commutator
q̂i , p̂j ≡ q̂i p̂j – p̂j q̂i = i h̄δij . An analogous procedure is used in a later chapter to quan-
tize the electromagnetic field, and the essential role played by commutation in optical
interactions will be discussed there further.
Solutions to the Schrödinger equation can be constructed from the spatial energy
eigenfunctions Un of the Hamiltonian, together with harmonic functions of time. For
example,

ψn (r, t) = Un (r) exp(–iωn t) (2.2.5)

is a particular solution which satisfies Schrödinger’s equation when the system energy is
constant (i.e., Ĥ ψn = i h̄ ∂t∂ ψn = h̄ωn ψn and h̄ωn must be the energy eigenvalue E of the
spatial eigenstate):

Ĥ Un = h̄ωn Un .

The energy of real, non-decaying systems is constant and observable, so the eigenvalues
of such systems are real. It is the adjoint nature of Ĥ that assures us mathematically
that the eigenvalues are real, that the corresponding eigenfunctions Un can be or-
thonormalized, and that they collectively furnish a complete description of the system.
That is,

Un∗ Um d 3 r = δnm , (2.2.6)
 
Un∗ (r̄)Un (r̄  ) = δ r̄ – r̄  . (2.2.7)

Formally, the Un form a complete basis set that can represent an arbitrary state of the sys-
tem. The most general solution for the wavefunction is formed from linear combinations
of solutions like Eq. (2.2.5), such as

ψ(r̄, t) = Cn Un (r̄)e–iωn t . (2.2.8)

n

If the total probability |ψ|2 for the system to occupy the space defined by the basis
functions Un is to be normalized to unity, one can easily show from Eq. (2.2.8) that the
constant coefficients must satisfy

|Cn |2 = 1. (2.2.9)
n
 Quantum Observables 15

In Eq. (2.2.9), |Cn |2 gives the occupation probability associated with a particular eigen-
state labeled by index n. This closure relation is exploited in Section 2.4 as the main tool
for changing between alternative descriptions of a system, expressed in terms of various
basis sets.

2.2.3 Symmetry
When researchers are confronted by a new problem, the greatest simplifications arise
from considering system symmetries. Symmetry can be exploited to reduce complex
problems to simpler ones and to derive general rules for atom–field interactions. The
optimal approach to using symmetry in this way is the domain of group theory, which
is too extensive a topic to be included here. However, we review one symmetry here,
inversion symmetry, because it has important implications for many problems in optical
science. It affects both linear and nonlinear interactions of light with matter. In a later
chapter we shall encounter the Wigner–Eckart theorem (see also Appendix H) which
can be used to implement other symmetry considerations, without group theory. For a
broader discussion of symmetry, see Ref. [2.2].
When a system has inversion symmetry, it is energetically indistinguishable in inverted
and non-inverted states (see Figure 2.2). Then, one can conclude from symmetry alone
that (i) the wavefunctions for different stationary states may be classified into odd or even
“parity,” (ii) the system cannot sustain a permanent dipole moment, and (iii) “parity”
must change when an electromagnetic transition occurs. Hence the presence or absence
of this symmetry often determines whether electromagnetic transitions are allowed or
not. To confirm these results we introduce parity operator Iˆ, defined by its inverting
action on all three spatial coordinates, and allow it to operate on the energy eigenvalue
equation:

Iˆ Ĥ (r̄)ψ(r̄) = IˆEψ(r̄) = Eψ(–r̄). (2.2.10)

Figure 2.2 Example of a three-dimensional object with inversion symmetry. A featureless sphere looks
the same if the coordinates are all reversed.
16 Basic Quantum Mechanics

The left side can be rewritten as


Iˆ Ĥ (r̄)ψ(r̄) = Ĥ (–r̄)ψ(–r̄) = Ĥ (r̄)Iˆψ(r̄), (2.2.11)

since Ĥ (r̄) = Ĥ (–r̄). This last point is the result of the system being physically indistin-
guishable in inverted or non-inverted states. Eq. (2.2.11) shows that the commutator
[Ĥ , Iˆ] = 0. Since Iˆ commutes with Ĥ , the energy eigenfunctions of Ĥ are also eigen-
functions of Iˆ. The eigenvalues I of the parity operator are easily found by applying
inversion twice to return the original wavefunction. Thus
Iˆ2 ψ = I 2 ψ = ψ, (2.2.12)
and wavefunctions of the system separate into two groups distinguished by I = ±1.
In the presence of inversion symmetry we reach three conclusions:

1. Energy eigenfunctions in systems with inversion symmetry have a definite parity.

That is, with respect to inversion of the coordinates they are either even (e) or
odd (o).
ψ (e) = ψ(r) + ψ(–r), for I = +1, (2.2.13)
ψ (o) = ψ(r) – ψ(–r)), for I = –1. (2.2.14)

2. The system has no permanent dipole moment.

ψ(r)|er|ψ(r) = Iˆ ψ(r)|er|ψ(r) = ±ψ|–er| ± ψ = – ψ(r)|er|ψ(r) .
Since the dipole moment must equal its negative, the only solution is

ψ|er|ψ = 0, (2.2.15)
regardless of the state ψ of the system.
3. Transition dipole moments (where the initial and final states are different) only
exist between states of opposite parity.
 (o)   (e)         
ψ (r) er ψ (r) = Iˆ ψ (o) (r) er ψ (e) (r) = –ψ (o) (r) –er ψ (e) (r)
   
= ψ (o) (r) er ψ (e) (r)  = 0, (2.2.16)
               
(o)   (o) (o)   (o) (o)   (o) (o)   (o)
ψ1 er ψ2 = Iˆ ψ1 er ψ2 = –ψ1 –er –ψ2 = – ψ1 er ψ2 = 0,
(2.2.17)
               
(e)   (e) (e)   (e) (e)   (e) (e)   (e)
ψ1 er ψ2 = Iˆ ψ1 er ψ2 = ψ1 –er ψ2 = – ψ1 er ψ2 = 0.
(2.2.18)

These selection rules based on inversion symmetry play an important role in some
of the simple examples of quantum systems that follow.
 Quantum Observables 17

2.2.4 Examples of Simple Quantum Systems

2.2.4.1 Simple Harmonic Oscillator
One of the most rudimentary dynamical systems with inversion symmetry consists of a
simple harmonic oscillator, exemplified by a spring that obeys Hooke’s law of motion on
an immoveable support attached to a mass m. Its allowed energy levels are depicted in
Figure 2.3.
In such a system, the restoring force of the spring is linear in the displacement x from
equilibrium in accord with F̄ = –kx̄ (where k = mω02 ) and the system energy consists of
potential and kinetic energy terms that appear explicitly in the Hamiltonian

Ĥ = p̂ 2 /2m + mω02 x̂ 2 /2. (2.2.19)

By solving Schrödinger’s equation one finds the stationary eigenfunctions

mω0 ½
Un = √ n Hn (ξ ) exp(–ξ 2 /2), (2.2.20)
h̄ π 2 n!

√
where ξ ≡ mω0 /h̄x and Hn (ξ ) is a Hermite polynomial. These will later be referred to
as basis states and written simply as |n ≡ |Un . The eigenvalues are

1
h̄ωn = n + h̄ω0 , n = 0, 1, 2, . . . (2.2.21)
2

Note that the eigenfunctions separate into even and odd sets, reflecting the inversion
symmetry of the motion about the origin.

2
Energy (ћω0 )

n=0

–2 –1 ξ=0 1 2

Figure 2.3 The lowest three eigenfunctions of the simple harmonic oscillator.
18 Basic Quantum Mechanics

2.2.4.2 Particle in a One-Dimensional (Symmetric) Potential Well

The situation of a particle in an infinitely deep potential is illustrated in Figure 2.4:

p̂ 2 0, 0 ≤ z ≤ L
Ĥ = + V̂ , V̂ = . (2.2.22)
2m ∞, z < 0, z > L

The momentum operator is p̂ = –i h̄∂/∂z. Energy eigenfunctions are therefore

√
2/L sin kn z, 0 ≤ z ≤ L
Un (z) = , (2.2.23)
0, z < 0, z > L

where kn = nπ /L and n = 1, 2, 3, . . .
The energy eigenvalues are

h̄2 k2n
h̄ωn = , (2.2.24)
2m
and a general wavefunction for this problem is
1/2
2
ψ= Cn sin (kn z) e–iωn t . (2.2.25)
L n

Notice that the solutions in Eq. (2.2.25) again separate into contributions with n even or
odd, reflecting the inversion symmetry of the potential, as in Section 2.2.3.

2.2.4.3 Particle in a One-Dimensional (Asymmetric) Potential Well

In modulation-doped field-effect transistor (MODFET) structures, energy bands near
interfaces bend in such a way that carriers can be trapped in shallow potential wells that
lack inversion symmetry, forming a two-dimensional (2-D) electron gas, as illustrated in
Figure 2.5.
The potential V (x) is approximately proportional to the distance x from the interface:
V (x) = V0 x. (2.2.26)

ћωn

4
3
2
1
z
0 L

Figure 2.4 The two lowest eigenfunctions of a one-dimensional infinite square well.
 Quantum Observables 19

Strained
GaN
AlGaN

2DEG
EC

Energy AlGaN GaN EV

V(x)

E3

E2

E1

Figure 2.5 (a) Schematic cross-section of a ternary nitride semiconductor structure giving rise to a two-
dimensional electron gas (2DEG) at the AlGaN/GaN interface (in the middle). (b) The linear potential
well and quantized energy levels at the 2DEG interface.

Hence the time-independent Schrödinger equation is

 
–h̄2 d 2
+ (V0 x – E) U (x) = 0. (2.2.27)
2me dx2

By introducing the dimensionless variable

1/3
En 2me V0
ξn ≡ x – , (2.2.28)
V0 h̄2

where n = 1, 2, 3, . . . is an integer index for the various solutions, this equation simpli-
fies to
 
d2
– ξ n Un (ξn ) = 0. (2.2.29)
dξn2
20 Basic Quantum Mechanics

The solution of this equation is an Airy function, related to Bessel functions of fractional
order. A suitable integral form of this function can be written as

∞
1 1 3
n (ξn ) = cos u + uξn du, (2.2.30)
π 3
0

with the explicit eigenfunctions being

Un (ξn ) = An n (–ξn ), (2.2.31)

where An is a constant amplitude for the nth solution. To find energy eigenvalues it is
necessary to apply boundary conditions. For this we can approximate the well height at
the origin as infinite, so that U (ξ ) = 0 at x = 0. That is,
  1/3 
En 2me V0
Un – = Un (–Rn ) = 0. (2.2.32)
V0 h̄2

This shows that the eigenvalues are proportional to roots (Rn ) of the Airy function. The
first few roots are R1 = 2.34, R2 = 4.09, R3 = 5.52, and R4 = 6.78. Hence the eigenvalues,
obtained from Eq. (2.2.32), are
 1/3
V02 h̄2
En = Rn . (2.2.33)
2me

The wavefunctions must be normalized to unity. This requires that

∞
Un∗ (ξ )Un (ξ )dξ = |An |2 , (2.2.34)
–Rn

showing that the undetermined coefficients are given by

⎛ ⎞–1/2
∞
An = ⎝ |Un (ξn )|2 dξn ⎠ . (2.2.35)
–Rn

In this example, no reduction of the wavefunction into sets of even and odd energy
levels takes place. Because of the lack of inversion symmetry, the wavefunctions do not
separate into different parity classes. (See Handbook of Mathematical Functions, NBS,
1964, pp.446–447, Eqs. 10.4.1 and 10.4.32 and Ref. [2.3]).
 Quantum Observables 21

2.2.4.4 Particle in an Adjustable Three-Dimensional Box

F centers in alkali halides are excellent examples of real particles in real cubic boxes
of adjustable size. These color centers consist of an electron in an anion vacancy of
an ionic crystal (a defect of the crystal structure), and give rise to dramatic coloration
of otherwise transparent crystals [2.3]. One does not need to know much in the way
of specifics about them to make quite detailed, verifiable predictions about them. We
picture an electron in a cubic space left by a missing anion, a space somewhat larger
than a hydrogen atom (Figure 2.6), with a positive charge and an infinitely high potential
at the walls. Because it is caged in by the otherwise perfect crystal, the wavefunction
must vanish at the “walls” comprised of neighboring cations. Hence, borrowing from
the one-dimensional symmetric case in Section 2.2.4.3, we would guess that possible
wavefunctions have the form

ψ(x, y, z) = sin(kx x) sin(ky y) sin(kz z) (2.2.36)

where kx = πa nx , ky = πa ny , and kz = π
n.
a z
The energy eigenvalues are
 
  h̄2 k2 h̄2 k2x + k2y + k2z h̄2 π 2  2 
E n x , ny , nz = = = n + n2y + n2z . (2.2.37)
2m 2m 2ma2 x
The ground state energy is

3h2
E0 = , (2.2.38)
8ma2
since nx = ny = nz = 1. The energy of the first excited state is

6h2
E1 = , (2.2.39)
8ma2
since nx = 2 and ny = nz = 1. The transition frequency of the first resonance of the
F center is therefore given by

3n2
hν = E1 – E0 = , (2.2.40)
8ma2
from which one obtains directly the Mollwo–Ivey relation [2.4] for the relationship
between the edge length of the crystal unit cell and the frequency of resonant absorption:

1
ln(a) = ln(ν) + c. (2.2.41)
2
As shown in Figure 2.7, this relation has been confirmed by measurements on various
alkali halide crystals.
22 Basic Quantum Mechanics

(a) z

2
) y
2 ψ(y
ψ (z)
ψ(x) 2

(b) z

2
) y
2 ψ(y
)
ψ(z
ψ(x) 2

Figure 2.6 Pictorial representation of the wavefunctions and charge distributions of an F center in the
(a) ground state and (b) the first excited state. (After Ref. [2.4]).

Multi-electron color centers are not only good examples of particle-in-a-box anal-
ysis, but have been widely used in high-gain tunable laser technology [2.5]. Another
important application relies on their mobility under applied fields. Oxygen vacancies
in semiconducting rutile (TiO2 ) are thermally more stable than the alkali F centers
and can be made to migrate by the application of voltage across the crystal, causing
a change in resistance of the medium proportional to the integral of voltage over time.
This understanding led to the first realization of a fundamental circuit element called the
memristor [2.6].
 Dynamics of Two-Level Systems 23

1.0
KI
KCl
0.6 KBr LiCl
NaCl
a (nm)

LiF
NaF
0.4

0.3

3 5 10 15
vmax(1014 Hz)

Figure 2.7 Plot of experimental measurements of the first absorption resonance of F centers in several
alkali halides, compared with the prediction of the Mollwo–Ivey relation (solid curve). (After Ref. [2.4].)

2.3 Dynamics of Two-Level Systems

Knowledge of system symmetries and eigenenergies is often enough to predict dynamics
in systems where the wavefunction and Hamiltonian H0 are unknown. This is fortunate,
because systems encountered in research often have unknown properties and structure
that is complicated enough to prevent determination of the wavefunction. The limited
number of models available for which ψ can be found analytically might even suggest
that these case studies are irrelevant to new problems in which ψ is essentially unknow-
able. However, there are at least three reasons why this is not so. The first is that many
aspects of dynamics are dominated by the interaction of light with only two energy levels
coupled by the light. Other levels in the system often are not strongly coupled by the
electromagnetic field and play a secondary role. So, two-level models capture dominant
aspects of the dynamics even in complex systems. The second reason is that complicated
systems can sometimes be converted into truly two-level systems by a judicious choice
of experimental conditions. Thirdly, as subsequently illustrated, the development that
a system undergoes as time progresses depends on its resonant frequencies and state
symmetries, not on any particular mathematical representation of H0 or ψ.
Consider a solid known to have two levels separated by an energy in the range of an
available light source. When light propagates along the z-axis of the crystal, the structure
absorbs strongly regardless of polarization. What, if anything, can be said about such
a system, its dynamics and transitions? Presuming the absorption process involves a
single electron as in the vast majority of cases, and recalling that the classical energy of
an electric dipole formed by the electron in a field Ē is given by the projection of the
moment on the external field, the interaction energy should be

Ĥ I = –μ̄ · Ē(z, t). (2.3.1)

24 Basic Quantum Mechanics

Since the light field is under the experimentalist’s control, assume it is a plane wave
traveling along z with no variation of its amplitude in x or y. The dipole moment could
be oriented in an arbitrary, fixed direction described by

μ̄ = μx x̂ + μy ŷ + μz ẑ. (2.3.2)

Perpendicular to z, only the x and y components of the moment are relevant for interac-
tion with the wave. In free space the orientation of Ē is perpendicular to the direction of
propagation, so μz ẑ · Ē(z, t) = 0. Since we are told the absorption is isotropic, we may
also assume μx = μy = μ0 , and without loss of generality reduce Eq. (2.3.1) to

Ĥ int = –μ0 E0 , (2.3.3)

where E0 is the amplitude of the electromagnetic wave. The charge oscillations excited
by the field are restricted to the x–y plane. So the interaction spans a two-dimensional
Hilbert space. Equation (2.3.3) therefore describes a 2-D interaction operator, and can
be represented using 2×2 matrices—the Pauli spin matrices which (together with the
unit matrix σI ) form a complete basis set:

     
0 1 0 –i 1 0
σ̂x = ; σ̂y = ; σ̂z = . (2.3.4)
1 0 i 0 0 –1

This set of matrices is called complete because any complex two-component vector can
be written as a linear combination of σx , σy , and σz together with the unit matrix σI using
appropriate coefficients. Also, the product of any two Pauli matrices is a Pauli matrix,
showing that they form a complete group. Since light can at most transform the state
of the atom among its available states in Hilbert space, the interaction Hamiltonian Ĥ I
must be an operator that can convert a 2-vector into another 2-vector in the same Hilbert
space. So its matrix representation must also be 2×2 and we choose the Pauli matrices  
0
as basis operators. The operator representation of Eq. (2.3.3) must change state
1
 
1
into and vice versa. Hence it is given by
0

Ĥ int = –μ0 E0 σ̂x . (2.3.5)

Exercise: Confirm that the form of Ĥ I in Eq. (2.3.5) is correct by applying it to ei-
ther eigenstate of the two-level atom and showing that it causes a transition to the
other.
 Dynamics of Two-Level Systems 25

Now σ̂x can be written in terms of raising and lowering operators for atomic states to
simplify interpretation of dynamics. Consider the following two matrices.
     
1  0 1 0 1
σ̂ + ≡ σ̂x + i σ̂y = → σ̂ + = , (2.3.6)
2 0 0 1 0
     
1  0 0 1 0
σ̂ – ≡ σ̂x – i σ̂y = → σ̂ – = . (2.3.7)
2 1 0 0 1

Writing the interaction Hamiltonian in terms of these operators, which raise or lower the
state of the atom, Eq. (2.3.5) becomes
 
0 1  
Ĥ int = –μ0 E0 = –μ0 E0 σ̂ + + σ̂ – . (2.3.8)
1 0

It may now be deduced that a continuous field–atom interaction causes the atom to
make periodic transitions up and down between states 1 and 2 by stimulated absorption
and emission, respectively, at a frequency related to the absorption coefficient (since it
depends on the transition moment μ0 ).
As shown by the exercise, discrete applications of the interaction Hamiltonian show
the individual transitions caused by optical irradiation:
   
0 1
Ĥ int = –μ0 E0 (2.3.9)
1 0

and
   
1 0
Ĥ int = –μ0 E0 . (2.3.10)
0 1

A more detailed picture of system dynamics can only emerge by solving explicitly for
temporal evolution of the state of the system. In the present problem there is no infor-
mation on the stationary states of the system. So we do not know the wavefunction itself,
but we do know there are at least two states in the system. There is a ground state which
might be represented as ψ – and an excited state ψ + . The challenge of predicting what
the system will do in response to irradiation requires the determination of the most gen-
eral wavefunction we can write based on this information. This is ψ(t) = c+ ψ + + c– ψ – ,
where c+ and c– are time-dependent coefficients. So the time-dependent Schrödinger
equation yields

d +    
c (t)ψ + + c– (t)ψ – = iΩ σ̂ + + σ̂ – c+ (t)ψ + + c– (t)ψ –
dt
 
= i c+ (t)ψ – + c– (t)ψ + , (2.3.11)
26 Basic Quantum Mechanics

where  ≡ μ0 E0 /h̄. Hence

 
∂c+ (t) + ∂c– (t) – 
ψ + ψ = i c+ (t)ψ – + c– (t)ψ + , (2.3.12)
∂t ∂t

and because ψ + and ψ – are independent vector states, Eq. (2.3.12) may be decomposed
into the two equations

∂c+
= ic– , (2.3.13)
∂t
∂c–
= ic+ . (2.3.14)
∂t

These two coupled equations are readily solved by differentiation and cross substitution.

∂ 2 c±
= –2 c± (2.3.15)
∂t 2
 2
By taking into account the normalization condition c+  + |c– |2 = 1 and initial condition
c– (0) = 1, the overall wavefunction can therefore be written

ψ(t) = c+ ψ + + c– ψ – = ψ + sin t + ψ – cos t. (2.3.16)

The system oscillates between the ground state ψ – and the excited state ψ + at a fre-
quency  called the resonant Rabi frequency. This oscillatory behavior is called Rabi
flopping. What is interesting here is that we have predicted some system dynamics—an
oscillation at a very specific rate in a poorly characterized system—without knowing the
functional form of the wavefunction ψ or even the unperturbed system Hamiltonian H0 .
It was sufficient to know a little about system symmetry, the energy eigenvalues, and
to assume the optical interaction was dominated by the electric dipole Hamiltonian. It is
particularly important to realize that one can calculate optical properties and dynamic re-
sponse to light without knowing the wavefunction, since the wavefunction itself is rarely
known in systems of interest.

2.4 Representations
2.4.1 Representations of Vector States and Operators
The Hamiltonian describing the interaction of matter with a perturbing influence is often
written as the sum of a static portion Ĥ 0 and an interaction term V̂ (the electric dipole
interaction between light and matter is considered in detail in Appendix C):

Ĥ = Ĥ 0 + V̂ (2.4.1)
 Representations 27

The Schrödinger equation then becomes

d i 
|ψ(t) = – Ĥ 0 + V̂ |ψ(t) . (2.4.2)
dt h̄

If the entire Hamiltonian Ĥ itself is time independent, direct integration of Eq. (2.4.2)
yields

i
|ψ(t) = exp – Ĥ t |ψ(0) , (2.4.3)
h̄

showing that the system wavefunction can still evolve by changing the admixture of its
eigenstates. Hence an observable O acquires a time dependence which, if not too rapid,
is reflected in the measured matrix element
 
Ô(t) = ψ|Ô |ψ . (2.4.4)

There are three common ways to partition the time dependence in Eq. (2.4.4) be-
tween the state vector |ψ and the operator Ô. The most common divisions are the
following:

Schrödinger picture:
   S  S  S 
Ô = ψ (t)Ô (0)ψ (t) , (2.4.5)
   
where ψ S (t) ≡ exp(–i Ĥt/h̄) ψ S (0) .
Interaction picture:
   I  I  I 
Ô = ψ (t)Ô (t)ψ (t) , (2.4.6)
   
where ψ I (t) ≡ exp(i Ĥ 0 t/h̄) ψ S (t) and ÔI (t) ≡ exp(i Ĥ 0 t/h̄)ÔS (0) exp(–i Ĥ0 t/h̄).
Heisenberg picture:
   H H  H
Ô = ψ  Ô (t) ψ , (2.4.7)
   
where ÔH (t) ≡ exp(i Ĥt/h̄)ÔS (0) exp(–i Ĥt/h̄) and ψ H ≡ ψ S (0) .

These options for representing problems are explored further in the next section. One
of the three pictures is generally preferable to another for a given problem. The choice
of representation is determined by what aspect of system dynamics is to be emphasized
in calculations.
28 Basic Quantum Mechanics

2.4.2 Equations of Motion in Different Representations

The basic equation of motion is Schrödinger’s equation, given by Eq. (2.4.2). However,
it is a matter of convenience as to whether one views the physical effects of system dy-
namics as being associated primarily with the atomic wavefunction, or the perturbing
field operator, or as being divided between the two. For easy analysis, the choice of rep-
resentation should reflect whether it is the state of the atom, or the state of the field, or
the interaction between them (i.e., coherent atom–field coupling) that is of most interest
respectively. Why is this a concern? This matters because we generally have in mind
experiments of a specific type with which we plan to probe a system. It may be just the
atomic state that we wish to monitor. Or, alterations in the spectrum and intensity of the
perturbing field itself may be the chief interest. Or, the chief concern might be measure-
ment of the amplitude of a transient coherent polarization that simultaneously reflects
the harmonic oscillation of a superposition state of the atom and its phase-dependent
coupling with the applied field. For ease of interpretation, our calculations should reflect
the way we choose to view the problem.

2.4.2.1 Time-Independent Ĥ
Direct integration of the Schrödinger equation when Ĥ is time independent gives

|ψ(t) = exp –i Ĥt/h̄ |ψ(0) , (2.4.8)

where
  1 2
exp –i Ĥt/h̄ ≡ 1 – i Ĥt/h̄ – Ĥt/h̄ + . . . (2.4.9)
2

This result can then be used to develop slightly different equations of motion for the
various pictures of system dynamics.

(i) Schrödinger picture:

d  S   
i h̄ ψ (t) = Ĥ ψ S (t) , (2.4.10)
dt
     
Ô(t) = ψ S (t)Ô(0)ψ S (t) , (2.4.11)

where
 S  
ψ (t) = exp –i Ĥt/h̄ |ψ(0) . (2.4.12)

In Eq. (2.4.11) the expectation value of the operator has been written in terms
of its initial value Ô(0) to emphasize that it does not change with time. All time
dependence is associated with the wavefunctions in the Schrödinger picture.
 Representations 29

(ii) Interaction picture:

To extract the purely sinusoidal dynamics associated with the static Hamilto-
nian H0 in our description of the time development of the state vector, the
contribution of H0 can be removed with the transformation
 S   
ψ (t) = exp –i Ĥ 0 t/h̄ ψ I (t) . (2.4.13)

Then

d     
i h̄ exp –i Ĥ 0 t/h̄ ψ I (t) = Ĥ S0 + V̂ S exp –i Ĥ 0 t/h̄ ψ I (t)
dt
d     
i h̄ ψ I (t) = exp i Ĥ 0 t/h̄ V̂ S exp –i Ĥ 0 t/h̄ ψ I (t) .
dt

This result may be rewritten in the form

d  I   
i h̄ ψ (t) = V̂ I (t) ψ I (t) , (2.4.14)
dt

where
 
V̂ I (t) ≡ exp i Ĥ 0 t/h̄ V̂ S exp –i Ĥ 0 t/h̄ (2.4.15)
 
is an effective Hamiltonian for the transformed state vector ψ I (t) . This rep-
resentation is called the interaction
 picture, because according to Eq. (2.4.14)
the development of ψ I (t) is determined by the interaction V̂ I (t) alone, rather
than Ĥ 0 or Ĥ .
On the other hand the time development of operators in this picture is deter-
mined by Ĥ 0 . To make a comparison with the Schrödinger picture we note from
Eq. (2.4.15) that an arbitrary operator in the interaction picture relates to one in
the Schrödinger picture according to
 
ÔI = exp i Ĥ 0 t/h̄ ÔS (0) exp –i Ĥ 0 t/h̄ . (2.4.16)

Consequently its expectation value may be evaluated by making use of

Eq. (2.4.13):
 I   I    
Ô (t) = ψ (t) exp i Ĥ 0 t/h̄ ÔS (0) exp –i Ĥ 0 t/h̄ ψ I (t)
     
 
= ψ S (t)exp –i Ĥ 0 t/h̄ exp i Ĥ 0 t/h̄ ÔS (0) exp –i Ĥ 0 t/h̄ exp i Ĥ 0 t/h̄ ψ S (t)
 
= ÔS (t) . (2.4.17)
30 Basic Quantum Mechanics

This shows that expectation values of operators are independent of representation.

However, in the interaction picture operators are assigned part of the overall time
dependence, unlike the Schrödinger picture. Hence we shall need an equation of
motion for the operator ÔI (t). Direct differentiation of Eq. (2.4.16) yields:

d I i  
Ô (t) = Ĥ 0 , ÔI . (2.4.18)
dt h̄

Note that the expansion coefficients in the Schrödinger and interaction pictures
are different. The complete time dependence is described by Schrödinger coef-
ficients cn (t) in the expansion |ψ = cn (t)|n , whereas the time dependence
n
due only to the interaction is described by interaction picture coefficients Cn (t)
in |ψ = Cn (t) exp(–iωn t)|n . For each eigenfrequency ωn , the coefficients are
n
therefore related by

cn (t) = Cn (t) exp(–iωn t). (2.4.19)

(iii) Heisenberg picture:

     
Ô(t) = ψ S (0) exp(i Ĥ t/h̄)Ô exp(–i Ĥ t/h̄) ψ S (0)
   
= ψ H ÔH (t) ψ H , (2.4.20)

where
 
ÔH (t) ≡ exp i Ĥt/h̄ Ô exp –i Ĥt/h̄ (2.4.21)

and
 H  S 
ψ = ψ (0) . (2.4.22)

By direct differentiation of Eq. (2.4.21) we obtain

d H d
ψ = ψ(0) = 0, (2.4.23)
dt dt
d H i  H H
Ô = Ĥ , Ô . (2.4.24)
dt h̄

Great care must be exercised when manipulating functions of operators, for

the simple reason that operators do not generally commute with one another.
Some decompositions of the time dependence which might suggest themselves
intuitively are invalid.
Exploring the Variety of Random
Documents with Different Content
The Project Gutenberg eBook of An ethical
philosophy of life presented in its main
outlines
 This ebook is for the use of anyone anywhere in the United
States and most other parts of the world at no cost and with
almost no restrictions whatsoever. You may copy it, give it away
or re-use it under the terms of the Project Gutenberg License
included with this ebook or online at www.gutenberg.org. If you
are not located in the United States, you will have to check the
laws of the country where you are located before using this
eBook.

Title: An ethical philosophy of life presented in its main outlines

Author: Felix Adler

Release date: August 6, 2019 [eBook #60068]

Most recently updated: October 17, 2024

Language: English

Credits: Produced by Turgut Dincer, Les Galloway and the Online

Distributed Proofreading Team at http://www.pgdp.net
(This
book was produced from images made available by the
HathiTrust Digital Library.)

*** START OF THE PROJECT GUTENBERG EBOOK AN ETHICAL

PHILOSOPHY OF LIFE PRESENTED IN ITS MAIN OUTLINES ***
 Transcriber’s Notes

Obvious typographical errors have been silently corrected. Variations in hyphenation

and accents have been standardised but all other spelling and punctuation remains
unchanged.
The cover was prepared by the transcriber and is placed in the public domain.
AN ETHICAL PHILOSOPHY OF LIFE

PRESENTED IN ITS MAIN OUTLINES

BY
FELIX ADLER

D. APPLETON AND COMPANY

NEW YORK LONDON
1920

Copyright, 1918, by
D. APPLETON AND COMPANY

Printed in the United States of America

 PREFACE
This book records a philosophy of life growing out of the
experience of a lifetime. The convictions put in it are not dogmatic,
for dogma is the conviction of one man imposed authoritatively upon
others. The convictions herein expounded are submitted to those
who search, as the writer has searched, for light on the problems of
life, in order that they may compare their experience with his, and
their interpretations of their experience with his interpretation.1
It is a great hope that some of the readers of this book may find
the general world-view expounded congenial, and for them also real
and true. It is believed that others may find the practical suggestions
as to the conduct of life in which the theory issues helpful in part, if
not in whole, as many of us accept from the teachings of the Stoics,
or of other thinkers, practical precepts, without on that account
adopting the philosophy from which these precepts are derived.
The book is divided into four parts: the first an autobiographical
introduction describing the various stations on the road by which the
author arrived at his present position, and offering incidental
appreciations and appraisements of the Hebrew religion, of
Emerson, of the ethics of the Gospels, of Socialism and of other
social reform movements.
The second part expounds the philosophical theory.
The third part contains the applications of the theory to the more
strictly personal life, under the captions of the Three Shadows of
Sickness, Sorrow and Sin, and also to the principal so-called Rights
to Life, Property, Reputation.
The fourth part applies the theory to the social institutions, to the
Family, the Vocation, the State, the International Society, and the
Church, these institutions being considered as an expanding series
through which the individual is to pass on his pilgrimage in the
direction of the supreme spiritual end.
The principal problems considered are:
1. How to establish the fundamental ethical dictum that every
human being ought to count, and is intrinsically worth while. This
dictum has been denied by many of the greatest thinkers, who
assert the intrinsic inferiority of some men, the intrinsic superiority of
others. The practice of the world also runs most distinctly contrary to
it. How then is it to be validated?
2. The problem of how to attach a precise meaning to the term
“spiritual,” thereby divesting it of the flavor of sentimentality and
vagueness that attaches to it.
3. How to link up the world’s activities in science, art, politics,
business, to the supreme ethical end.
4. How to lay foundations whereon to erect the conviction that
there verily is a supersensible reality.
For the repetitions that occur throughout the volume indulgence is
requested. In presenting an unfamiliar system of thought they may
sometimes assist the reader in retaining the thread.
The work is conceived as a whole, and should be read through
before any part of it is more minutely examined. The theory of Part
II especially should be read in the light of the applications submitted
in Parts III and IV.
 CONTENTS
BOOK I
AUTOBIOGRAPHICAL INTRODUCTION
CHAPTER PAGE
I.Prelude 3
II.The Hebrew Religion 14
III.Emerson 27
IV.The Teachings of Jesus 30
V.Social Reform 43
VI.The Influence of My Vocation on Inner Development 58
BOOK II
PHILOSOPHICAL THEORY
I.Introductory Remarks: Critique of Kant 73
II.Critique of Kant (Continued) 82
III.Preliminary Remarks on Worth, and on the Reasons Why the Method
Employed by Ethics Must Be the Opposite of That Employed by the
Physical Sciences 91
IV.The Ideal of the Whole 100
V.The Ideal of the Whole and the Ethical Manifold 114
VI.The Ideal of the Spiritual Universe and the God-Ideal 125
BOOK III
APPLICATIONS: THE THREE SHADOWS, SICKNESS, SORROW AND SIN, AND THE
RIGHT TO LIFE, PROPERTY AND REPUTATION
I.Introduction 147
II.The Three Shadows: Sickness, Sorrow, Sin 154
III.Bereavement 162
IV.The Shadow of Sin 171
V.The Spiritual Attitude to be Observed towards Fellow-Men in General,
Irrespective of the Special Relations Which Connect Us More Closely
with Some than Others 179
VI.The Meaning of Forgiveness 202
VII.The Supreme Ethical Rule: Act so as to Elicit the Best in Others and
Thereby in Thyself 208
VIII.The Supreme Ethical Rule (Continued) 220
 IX.How to Learn to SeeSpiritual Numen in Others
the 223
BOOK IV
APPLICATIONS: THE ETHICS OF THE FAMILY, THE STATE, THE INTERNATIONAL
RELATIONS, ETC.
I.The Collective Task of Mankind and the Three-fold Reverence 241
II.The Family 249
III.The Vocations 260
IV.The Practical Vocations 270
V.The Vocation of the Artist: Outline of a Theory of the Relation of Art to
Ethics 277
VI.Educational Vocations, or Vocations Connected with the State 289
VII.The State 305
VIII.The National Character Spiritually Transformed: the International Society,
or the Organization of Mankind 324
IX.Religious Fellowship as the Culminating Social Institution 341
X.The Last Outlook on Life 354
APPENDIX
Appendix I: Spiritual Self-Discipline 365
Appendix II: The Exercise of Force in the Interest of Freedom 369
INDEX 375
 BOOK I
AUTOBIOGRAPHICAL
INTRODUCTION
CHAPTER I
PRELUDE
What this book offers is a system of thought and of points of view
as to conduct, as these have jointly grown out of personal
experience. It will be useful to introduce them with an
autobiographical statement. The ideas which follow are such as have
been found by me, the author, to be fruitful. Certainly I claim for
them objectivity; but I do so because of what I have found them to
mean in my own life. He who has been scorched by lightning knows
that the effects of the lightning will be felt by all who are exposed to
the same experience. I narrate my experience; let others compare
with it theirs.
There is, however, a serious, and most embarrassing difficulty in
the way of discussing the phases and vicissitudes of one’s ethical
development. Self-appraisement is necessarily involved in the
narration. The outstanding subject of ethics is the self and its
relations. The physicist, the chemist, the biologist, however the
methods they use may differ in other respects, agree in the
endeavor to eliminate the personal equation. The psychologist
likewise does his best to see the procession that moves across the
inner stage like an interested but detached spectator. In the case of
ethics, however, the personal factor cannot be eliminated, because
the personal factor is just the Alpha and the Omega of the whole
matter; and if this be left out of account, the very object to be
studied disappears.
 Ethical standards are exacting, separated often from performance
by the widest interval. To set up a standard, therefore, is to reflect
upon oneself, to expose oneself to the backstroke of one’s own
deliverances, to be plunged perhaps into deep pits of self-
humiliation. How shall anyone have the courage to face so searching
a test, or the hardihood to discuss with a lofty air, and to
recommend to others ideals of conduct against which he knows that
he daily offends? How can anyone teach ethics or write about it?
The words of the Sermon on the Mount, “Judge not that ye be not
judged,” seem to apply very closely. Do not judge others, do not lay
down the law for others, because in so doing you will be judged in
the inner forum, becoming a repulsive object in your own eyes, or
standing forth a whited sepulcher. In brief, to touch the subject of
ethics is to handle a knife that cuts both ways, to cast a weapon
which returns upon him who sends it.
The difficulty then which confronts the ethical writer is that the
attitude of detachment possible in other branches of investigation is
found to be impossible when one attempts to sound the profundities
of that kind of inner experience which is called ethical. The self
obtrudes itself at every point, and it instinctively refuses to be
humbled. What may be denominated the struggle for self-esteem
has indeed played a leading rôle both in the outer and inner history
of mankind. This struggle, whose immense importance is often
overlooked, accounts for even more interesting facts than the
biological struggle for existence. The desire to exercise power over
others, often ruthless in the means adopted, is frequently nothing
more than a miserable attempt to save self-esteem by covering up
the inner sense of the weakness of the self. But the same struggle
penetrates also into the realm of theoretical ethics with which we are
concerned. Here it tampers with the standards which mortify self-
esteem, by inventing such ethical theories as seem to make the
problems of personality easy of solution, and by blinking the tragic
facts of guilt, remorse, etc. Various ethical systems that are in vogue
at the present time are, at least in part, exemplars of this process—
the theory for instance that ethics is nothing more than a calculus of
self-interest, or a matter of sympathetic feeling, or a balancing of
the more refined against the grosser pleasures. The instinct of self-
preservation, in the shape of the preservation of self-esteem, is quite
incorrigible, and against its insidious suggestion we have reason to
be particularly on our guard in the discussion which we are entering.
Are we then to refrain, out of sheer regard for decency, from
touching on this subject at all? Is everyone who writes on ethics, or
attempts to teach it, either a pedant or a hypocrite? But we cannot
avoid discussing it, nor resist the impulse to teach and write about it,
for it is the subject on which more than any other we and others
sorely need help and enlightenment. And we shall get help in the
endeavor to afford it to others. This, then, is my position: I do not
presume to lay down the law for anyone. I find that I can set forth
the better standards which in the course of trial and error I have
come to recognize. I would not shamelessly expose mere private
failures and failings after the manner of Rousseau in the
“Confessions”; for there is a tract of the inner life which ought to be
kept from publicity and prying intrusion. I shall then deal with
deflections only in so far as they can be traced to false standards or
principles, and as they tend to illustrate the flaw in those standards
and principles.
What I state as certain is certain for me. It has approved itself as
such in my experience. Let others consult their experience, and see
how far it tallies with that which is here set forth. A distinction,
however, I wish to call attention to between the theory as
expounded in the second part of this volume, and the practical
applications to be found in the third and fourth parts. Persons who
are not trained in metaphysical thinking or interested in it, may do
well to omit the reading of the second part. To those who are
competent in philosophical thinking, and who disagree with the
positions there taken, I may perhaps be permitted to suggest that
one can dissent from a philosophy and yet find help in the
applications to which it leads. And, after all, it is the practice that
counts.
 With these preliminaries, I now proceed to delineate briefly the
stages of inner development which have led me slowly and with
much labor to the system of thought described in the following
pages.
One of the leading principles to which I early gave assent, and to
which I have ever since adhered as a correct fundamental insight, is
expressed in the statement that every human being is an end per se,
worth while on his own account.2
Every human personality is to be safe against infringement and is,
in this sense, sacred. There is a certain precinct which may not be
invaded. The experience which served me especially as the matrix of
this idea was the adolescent experience of sex-life,—the necessity
felt of inhibiting, out of reverence for the personality of women, the
powerful instincts then awakened.3
The fact that I had lived abroad for three years in frequent contact
with young men, especially students, who derided my scruples, and
in the impure atmosphere of three capital cities of Europe, Berlin,
Paris and Vienna, where the “primrose path” is easy, tended to make
the retention of my point of view more difficult, and at the same
time to give it greater fixity, also to drive me into a kind of inward
solitude. I felt myself in opposition to my surroundings, and acquired
a confidence, perhaps exaggerated, to persevere along my own lines
against prevailing tendencies.
I ought next to mention the decay of theism which took place in
my mind in consequence of philosophic reading. Already at an early
age I had stumbled over the doctrine of Creation. I remember asking
my Sunday School teacher—How is creation possible? How can
something originate out of nothing? The answer I received was
evasive, and left me uneasy and unsatisfied. On another occasion I
ventured to suggest to the same authority—a revered and beloved
authority—that the conception of God seemed to me too much like
that of a man, too much fashioned on the human model; and he
amazed me beyond words by replying that he himself sympathized
more or less with the ideas of Spinoza. This chance remark set me
thinking, and seemed to open wide spaces in which my mind felt
free to travel—though I never tended in the direction of Spinoza.4
My thoughts were driven still further by reaction against the
narrow theology of the lectures on Christian Evidences as taught at
that time in Columbia College, where I was a student. And all these
influences came to a head in the atmosphere of the German
university at Berlin. There I heard Zeller, Duhring, Steinthal, Bonitz.
Above all I came into contact with Herman Cohen, subsequently and
for many years professor of philosophy at the University of Marburg,
and undertook to grapple in grim earnest with the philosophy of
Immanuel Kant. The net outcome was not atheism in the moral
sense,—I have never been what is called an atheist,—but the
definite and permanent disappearance of the individualistic
conception of Deity. I was attracted by the rigor, the sublimity, of
Kant’s system, and especially by his transcendental derivation of the
moral law. The individualistic basis of his ethics, which is quite
uncongenial to me, I ignored, and for a time simply accounted
myself a follower of Kant. Very often since then I have discovered
that men, unbeknown to themselves, are apt to sail under false
flags, ranking themselves Kantians, Socialists, or what not, because
the system to which they give their adherence attracts them at some
one outstanding point, the point namely, where it sharply conflicts
with views which they themselves strongly reprobate; and they are
thus led to overlook other features no less important in which the
system is really uncongenial to them. Thus a person who recognizes
the evils of the present wage system may label himself a Socialist,
simply because Socialism is most in evidence as an adversary of the
wage system, while he may by no means agree with the positive
principles that underlie Socialism, when he comes to examine them
dispassionately.
I thought at that time of the Moral Law as that which answers to
or should replace the individualistic God-idea. I believed in an
unknown principle or power in things of which the Moral Law is the
manifestation, and I found the evidence of the moral law in man’s
consciousness. Matthew Arnold’s “the power that makes for
righteousness” is a phrase which at that time would have suited me,
—though perhaps not entirely even at that time. I have since come
to see that “making for righteousness” is a conception inapplicable
to the ultimate reality, and is properly applied only to human effort;
since purpose implies that the end sought has not as yet been
realized, and non-realization and ultimate reality are contradictory
ideas. The power that only makes for righteousness cannot be the
ultimate truth in things. The utmost we can say is that the ultimate
reality expresses itself in the human world as the power that inspires
in men moral purpose.
To return to my personal experiences, there fell into my hands,
while still a student abroad, a book by Friedrich Albert Lange entitled
Die Arbeiterfrage (The Labor Question), which proved epoch-making
in my life. Bacon says in his essay Of Studies: “Some books are to be
tasted, others to be swallowed, and some few to be chewed and
digested.” He might have added that there are books that make a
man over, changing the current of his existence, or at least opening
channels which previously had been blocked.5
Die Arbeiterfrage is not a great book. In the literature of the
subject it has long since been superseded. Yet it opened for me a
wide and tragic prospect, an outlook of which I had been until then
in great measure oblivious, an outlook on all the moral as well as
economic issues involved in what is called the Labor Question. My
teacher in philosophy, Cohen, once said to me sharply, that if there
is to be anything like religion in the world hereafter, Socialism must
be the expression of it. I did not agree with his statement that
Socialism spells religion, and have not seen my way to this day
toward identifying the two. But I realized that there was a measure
of truth in what he said,—and that I must square myself with the
issues that Socialism raises. Lange helped me to do this.
 He aided me in other respects as well. His History of Materialism
dispelled some of the fictitious glamor that still hung about the
materialistic hypothesis at that time,—though the last chapter on the
ultimate philosophy of life, in which he identifies religion with poetry,
is distinctly weak. I read his book on the Labor Question with
burning cheeks; no work of fiction ever excited me as did this little
treatise. It was ethical in spirit, if not in its ruling ideas. It favored
productive co-operation, and seemed to point a way to immediate
action, as Socialism did not.
The upshot of it was that I now possessed a second object,
namely, the laborer, to whom I could apply my non-violation ethics. I
had always felt an instinctive, idealizing reverence for women, and
this had its influence in the first practical outcome of the philosophy
of life with which I started on my career. I would go out as the
minister of a new religious evangel. Instead of preaching the
individual God, I was to stir men up to enact the Moral Law; and to
enact the Moral Law meant at that time primarily to influence the
young men with whom I came into contact to reverence
womanhood, and to keep inviolate the sacred thing, woman’s honor.
And now I had a second arrow in my quiver. I was to go out to help
to arouse the conscience of the wealthy, the advantaged, the
educated classes, to a sense of their guilt in violating the human
personality of the laborer. My mother had often sent me as a child
on errands of charity, and had always impressed upon me the duty
of respecting the dignity of the poor while ministering
sympathetically to their needs. I was prepared by this youthful
training to resent the indignity offered to the personality of the
laborer, as well as the suffering endured by him in consequence of
existing conditions.
Accordingly, on returning from abroad, my first action consisted in
founding among men of my own or nearly my own age a little
society which we ambitiously called a Union for the Higher Life,
based on three tacit assumptions: sex purity, the principle of
devoting the surplus of one’s income beyond that required for one’s
own genuine needs to the elevation of the working class, and thirdly,
continued intellectual development. A second practical enterprise
attempted was the establishment of a co-operative printing shop.
This having failed because of the selfishness actuating the members,
the Workingman’s School was founded, with the avowed object of
creating a truly co-operative spirit among workingmen.
I must, however, pause at this point to explain how the
development described led me to separation from the Hebrew
religion, the religion in which I was born, and to the service of which
as a Jewish minister it was expected that I should devote my life.
Welcome to Our Bookstore - The Ultimate Destination for Book Lovers
Are you passionate about books and eager to explore new worlds of
knowledge? At our website, we offer a vast collection of books that
cater to every interest and age group. From classic literature to
specialized publications, self-help books, and children’s stories, we
have it all! Each book is a gateway to new adventures, helping you
expand your knowledge and nourish your soul
Experience Convenient and Enjoyable Book Shopping Our website is more
than just an online bookstore—it’s a bridge connecting readers to the
timeless values of culture and wisdom. With a sleek and user-friendly
interface and a smart search system, you can find your favorite books
quickly and easily. Enjoy special promotions, fast home delivery, and
a seamless shopping experience that saves you time and enhances your
love for reading.
Let us accompany you on the journey of exploring knowledge and
personal growth!

ebookgate.com