Full Chapter Resonant Tunneling Quantum Waveguides of Variable Cross Section Asymptotics Numerics and Applications 2Nd Edition Lev Baskin PDF

Resonant Tunneling Quantum
Waveguides of Variable Cross Section

Asymptotics Numerics and
Applications 2nd Edition Lev Baskin
Visit to download the full and correct content document:
https://textbookfull.com/product/resonant-tunneling-quantum-waveguides-of-variable-
cross-section-asymptotics-numerics-and-applications-2nd-edition-lev-baskin/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...
A Cross Section of Psychological Research: Journal

Articles for Discussion and Evaluation 2nd Edition
Andrea K. Milinki
https://textbookfull.com/product/a-cross-section-of-
psychological-research-journal-articles-for-discussion-and-
evaluation-2nd-edition-andrea-k-milinki/
Elasticity: theory, applications, and numerics Fourth

Edition Sadd
https://textbookfull.com/product/elasticity-theory-applications-
and-numerics-fourth-edition-sadd/
Finite Mathematics as the Foundation of Classical

Mathematics and Quantum Theory With Applications to
Gravity and Particle Theory 1st Edition Felix Lev
https://textbookfull.com/product/finite-mathematics-as-the-
foundation-of-classical-mathematics-and-quantum-theory-with-
applications-to-gravity-and-particle-theory-1st-edition-felix-
lev/
Elasticity: Theory, Applications, and Numerics 4th

Edition Martin H. Sadd
https://textbookfull.com/product/elasticity-theory-applications-
and-numerics-4th-edition-martin-h-sadd/
Temporal Quantum Correlations and Hidden Variable
Models 1st Edition Costantino Budroni (Auth.)
https://textbookfull.com/product/temporal-quantum-correlations-
and-hidden-variable-models-1st-edition-costantino-budroni-auth/
Geoengineering, the Anthropocene and the End of Nature

Jeremy Baskin
https://textbookfull.com/product/geoengineering-the-anthropocene-
and-the-end-of-nature-jeremy-baskin/
How to Love Yourself The Art of Loving Myself 2nd

Edition Jonathan Kis-Lev
https://textbookfull.com/product/how-to-love-yourself-the-art-of-
loving-myself-2nd-edition-jonathan-kis-lev/
Health Promotion 2nd Edition Cross
https://textbookfull.com/product/health-promotion-2nd-edition-
cross/
Sustainable Tunneling and Underground Use Mona Badr
https://textbookfull.com/product/sustainable-tunneling-and-
underground-use-mona-badr/
Lecture Notes on Numerical Methods
in Engineering and Sciences
Lev Baskin
Pekka Neittaanmäki
Boris Plamenevskii
Oleg Sarafanov
Resonant Tunneling
Quantum Waveguides of Variable
Cross-Section, Asymptotics, Numerics,
and Applications
Second Edition
Lecture Notes on Numerical Methods
in Engineering and Sciences
Editorial Board
Francisco Chinesta, Ecole Centrale de Nantes, Nantes Cedex 3, France
Series Editor
Eugenio Oñate, Jordi Girona, 1, Edifici C1 - UPC, Universitat Politecnica de
Catalunya, Barcelona, Spain
Editorial Board
Charbel Farhat, Dept. Mechanical Engineering, Stanford University, Stanford, CA,
USA
C.A. Felippa, Department of Aerospace Engineering Scie, Univ of Colorado, Coll
of Engg & App, Boulder, CO, USA
Antonio Huerta, Universitat Politècnica de Cataluny, Barcelona, Spain
Thomas J. R. Hughes, Institute for Computational Enginee, University of Texas at
Austin, Austin, TX, USA
Sergio Idelsohn, CIMNE - UPC, Barcelona, Spain
Pierre Ladevèze, Ecole Normale Supérieure de Cachan, Cachan Cedex, France
Wing Kam Liu, Evanston, IL, USA
Xavier Oliver, Campus Nord UPC, International Center of Numerical M,
Barcelona, Spain
Manolis Papadrakakis, National Technical University of At, Athens, Greece
Jacques Périaux, Barcelona, Spain
Bernhard Schrefler, Mechanical Sciences, CISM - International Centre for, Padua,
Italy
Genki Yagawa, School of Engineering, University of Tokyo, Tokyo, Japan
Mingwu Yuan, Beijing, China
This series publishes text books on topics of general interest in the field of
computational engineering sciences.
The books will focus on subjects in which numerical methods play a fundamental
role for solving problems in engineering and applied sciences. Advances in finite
element, finite volume, finite differences, discrete and particle methods and their
applications to classical single discipline fields and new multidisciplinary domains
are examples of the topics covered by the series.
The main intended audience is the first year graduate student. Some books define
the current state of a field to a highly specialised readership; others are accessible to
final year undergraduates, but essentially the emphasis is on accessibility and
clarity.
The books will be also useful for practising engineers and scientists interested in
state of the art information on the theory and application of numerical methods.
More information about this series at http://www.springer.com/series/8548

Lev Baskin Pekka Neittaanmäki
• •
Boris Plamenevskii Oleg Sarafanov

•
Resonant Tunneling
Quantum Waveguides of Variable
Cross-Section, Asymptotics, Numerics,
and Applications
Second Edition
123
Lev Baskin Pekka Neittaanmäki
Department of Mathematics Mathematical Information Technology
St. Petersburg State University University of Jyväskylä
of Telecommunications Jyväskylä, Finland
St. Petersburg, Russia
Oleg Sarafanov
Boris Plamenevskii Department of Higher Mathematics
Department of Higher Mathematics and Mathematical Physics
and Mathematical Physics St. Petersburg State University
St. Petersburg State University St. Petersburg, Russia
St. Petersburg, Russia
ISSN 1877-7341 ISSN 1877-735X (electronic)

Lecture Notes on Numerical Methods in Engineering and Sciences
ISBN 978-3-030-66455-8 ISBN 978-3-030-66456-5 (eBook)
https://doi.org/10.1007/978-3-030-66456-5
1st edition: © Springer International Publishing Switzerland 2015
2nd edition: © Springer Nature Switzerland AG 2021
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt from
the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, expressed or implied, with respect to the material contained
herein or for any errors or omissions that may have been made. The publisher remains neutral with regard
to jurisdictional claims in published maps and institutional affiliations.
This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface
Devices based on the phenomenon of electron resonant tunneling are widely used in
electronics. Efforts are directed toward refining properties of resonance structures.
There are prospects for building new nanosize electronic elements based on
quantum dot systems. However, the role of resonance structure can also be given to
a quantum wire of variable cross-section. Instead of an “electrode—quantum dot—
electrode” system, one can use a quantum wire with two narrows. A waveguide
narrow is an effective potential barrier for longitudinal electron motion along a
waveguide. The part of the waveguide between two narrows becomes a “resonator,”
where electron resonant tunneling can occur. This phenomenon consists in the fact
that, for an electron with energy E, the probability TðEÞ to pass from one part of the
waveguide to the other part through the resonator has a sharp peak at E ¼ Eres ,
where E res denotes a “resonant” energy. Such quantum resonators can find appli-
cations as elements of nanoelectronic devices and provide some advantages in
regard to operation properties and production technology.
In the book, we study electron resonant tunneling in two- and three-dimensional
quantum waveguides of variable cross-sections in the time-independent approach.
We suggest mathematical models for the resonant tunneling and develop asymp-
totic and numerical approaches for investigating the models. We also present
schemes for several electronic devices based on the phenomenon of resonant tun-
neling. The book is addressed to mathematicians, physicists, and engineers inter-
ested in waveguide theory and its applications in electronics.
v
vi Preface
The second edition of the book contains new subjects: the resonant tunneling in
2D waveguides with several resonators (Chap. 6) and the resonant tunneling of
high-energy electrons in 2D waveguides (Chap. 7). The discussion of a finite
waveguide work function given in the first edition is supplemented by a study of the
impact of a finite work function on transport of high-energy electrons and is now
presented in Chap. 9. Several mistakes (see Rem. 5.5.2) and a number of misprints
are corrected.
St. Petersburg, Russia Lev Baskin

Jyväskylä, Finland Pekka Neittaanmäki
St. Petersburg, Russia Boris Plamenevskii
St. Petersburg, Russia Oleg Sarafanov
Contents
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
2 Waveguides. Radiation Principle. Scattering Matrices . . . . . . . . . . 17
2.1 Boundary Value Problem in a Cylinder . . . . . . . . . . . . . . . . . . 19
2.1.1 Statement of the Problem. Operator Pencil . . . . . . . . . . 19
2.1.2 The Solvability of the Problem in a Cylinder . . . . . . . . 24
2.1.3 Asymptotics of Solutions . . . . . . . . . . . . . . . . . . . . . . 25
2.2 Problem in a Domain G with Cylindrical Ends . . . . . . . . . . . . . 29
2.2.1 Statement and Fredholm Property of the Problem . . . . . 29
2.2.2 Asymptotics of Solutions . . . . . . . . . . . . . . . . . . . . . . 30
2.2.3 Properties of the Index Ind Ab ðlÞ and of the Spaces
kerAb ðlÞ and cokerAb ðlÞ . . . . . . . . . . . . . . . . . . . . . 31
2.2.4 Calculation of the Coefficients in the Asymptotics . . . . 33
2.3 Waves and Scattering Matrices . . . . . . . . . . . . . . . . . . . . . . . . 36
2.3.1 Waves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
2.3.2 Continuous Spectrum Eigenfunctions. The Scattering
Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 38
2.3.3 The Intrinsic Radiation Principle . . . . . . . . . . . . . . . .. 40
3 Properties of Scattering Matrices in a Vicinity of Thresholds . . . . . 43
3.1 Augmented Space of Waves . . . . . . . . . . . . . . . . . . . . . . . . . . 44
3.1.1 Waves in a Cylinder . . . . . . . . . . . . . . . . . . . . . . . . . . 44
3.1.2 Waves in Domain G . . . . . . . . . . . . . . . . . . . . . . . . . . 47
3.2 Continuous Spectrum Eigenfunctions. Scattering Matrices . . . . . 48
3.2.1 Intrinsic and Expanded Radiation Principles . . . . . . . . . 48
3.2.2 Analyticity of Scattering Matrices with Respect
to Spectral Parameter . . . . . . . . . . . . . . . . . . . . . .... 53
3.3 Other Properties of the Scattering Matrices . . . . . . . . . . . . .... 58
3.3.1 The Connection Between SðlÞ and SðlÞ
for ¿ 0 \l\¿ . . . . . . . . . . . . . . . . . . . . . . . . . . . .... 58
vii
viii Contents
3.3.2 The Connection Between SðlÞ and SðlÞ

for ¿\l\¿ 00 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
3.3.3 The Limits of SðlÞ as l ! ¿ 0 . . . . . . . . . . . . . . . . 63
4 Method for Computing Scattering Matrices . . . . . . . . . . . . . ..... 69
4.1 A Method for Computing Scattering Matrices Outside
Thresholds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... 69
4.1.1 Statement of the Method . . . . . . . . . . . . . . . . . . ..... 69
4.1.2 The Problem in GR . . . . . . . . . . . . . . . . . . . . . . ..... 71
4.1.3 Justification of the Method for Computing
the Scattering Matrix . . . . . . . . . . . . . . . . . . . . ..... 73
4.2 A Method for Computing Scattering Matrices in Vicinity
of Thresholds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... 77
5 Asymptotic and Numerical Studies of Resonant Tunneling
in 2D-Waveguides for Electrons of Small Energy . . . . . . . . . . . . . . 83
5.1 Statement of the Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
5.2 Limit Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
5.2.1 First Kind Limit Problems . . . . . . . . . . . . . . . . . . . . . 86
5.2.2 Second Kind Limit Problems . . . . . . . . . . . . . . . . . . . 88
5.3 Special Solutions to the First Kind Homogeneous Problems . . . 89
5.4 Asymptotic Formulas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
5.4.1 Asymptotics of the Wave Function . . . . . . . . . . . . . . . 93
5.4.2 Formulas for e S 11 , e
S 12 , and C1j . . . . . . . . . . . . . . . . . . 94
5.4.3 Formulas for Resonant Tunneling Characteristics . . . . . 99
5.5 Justification of the Asymptotics . . . . . . . . . . . . . . . . . . . . . . . . 100
5.6 Comparison of Asymptotic and Numerical Results . . . . . . . . . . 110
5.6.1 Problems and Methods for Numerical Analysis . . . . . . 111
5.6.2 Comparison of Asymptotic and Numerical Results . . . . 114
6 Resonant Tunneling in 2D-Waveguides with Several
Resonators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
6.1 Statement of the Problem and Preliminaries . . . . . . . . . . . . . . . 120
6.1.1 Waveguide and the Radiation Conditions . . . . . . . . . . . 120
6.2 Asymptotics of the Wave Function and the Scattering
Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
6.2.2 Formulas for C0 ; . . .; C2n þ 1 . . . . . . . . . . . . . . . . . . . . 127
6.2.3 Asymptotics of the Scattering Matrix . . . . . . . . . . . . . . 129
6.3 Asymptotics of the Resonant Tunneling . . . . . . . . . . . . . . . . . . 131
6.3.1 Amplitude of the Transmitted Wave . . . . . . . . . . . . . . 131
6.3.2 Poles of the Scattering Matrix . . . . . . . . . . . . . . . . . . . 134
Contents ix
6.3.3 Transmission Coefficient . . . . . . . . . . . . . . . . . . . . . . . 136

6.4 Comparison of Asymptotic and Numerical Results . . . . . . . . . . 140
7 Resonant Tunneling of High-Energy Electrons
in 2D-Waveguides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
7.3 Limit Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
7.3.1 Limit Problems of the First Kind . . . . . . . . . . . . . . . . . 147
7.3.2 Limit Problems of the Second Kind . . . . . . . . . . . . . . . 151
7.4.1 Asymptotics of Wave Functions . . . . . . . . . . . . . . . . . 152
7.4.2 Formulas for Cml . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
7.4.3 Asymptotics of Scattering Matrix . . . . . . . . . . . . . . . . 157
7.4.4 Asymptotics for Poles of Scattering Matrix . . . . . . . . . 158
7.4.5 Asymptotics of Resonant Tunneling . . . . . . . . . . . . . . 160
7.5 Comparison of Asymptotic and Numerical Results
in a Symmetric Waveguide . . . . . . . . . . . . . . . . . . ....... . . 161
7.5.1 Asymptotic Formulas . . . . . . . . . . . . . . . . ....... . . 162
7.5.2 Calculation of the Scattering Matrix . . . . . ....... . . 163
7.5.3 Comparison of Asymptotics and Numerical Results . . . 165
8 Numerical Simulation of High Energy Electron Transport . . . . . . . 169
8.1 Numerical Simulation of Multichannel Resonant Tunneling . . . . 169
8.1.1 Closed Resonator . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
8.1.2 The Method for Computing Scattering Matrix . . . . . . . 170
8.1.3 Discussion of Numerical Results . . . . . . . . . . . . . . . . . 173
8.2 Fano Resonances . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
9 The Impact of a Finite Waveguide Work Function
on Resonant Tunneling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
9.1 Preliminaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
9.2 A Qualitative Analysis of a Finite Work Function Impact
on Electron Transport . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
9.3 Numerical Simulation of Resonant Tunneling with Regard
to the Waveguide Work Function . . . . . . . . . . . . . . . . . . . . . . 193
9.4 Impact of Finite Work Function on High Energy Electron
Transport . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
10 Asymptotics of Resonant Tunneling in 3D Waveguides
for Electrons of Small Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201
10.1 Statement of the Problem and Outline of the Results . . . . . . . . 201
10.2 Limit Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
x Contents
10.3 Tunneling in a Waveguide with One Narrow . . . . . . . . . . . . . . 207

10.3.1 Special Solutions to the First Kind Homogeneous
Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
10.3.2 Passing Through the Narrow . . . . . . . . . . . . . . . . . . . . 209
10.3.3 Formal Asymptotics . . . . . . . . . . . . . . . . . . . . . . . . . . 212
10.3.4 The Estimate of the Remainder . . . . . . . . . . . . . . . . . . 216
10.4 Tunneling in a Waveguide with Two Narrows . . . . . . . . . . . . . 218
10.4.1 Special Solutions to the Problem in the Resonator . . . . 218
11 Resonant Tunneling in 2D Waveguides in Magnetic Field . . . . . . . 235
11.2 The Limit Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238
11.3 Special Solutions to Homogeneous First Kind Limit
Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
S 12 , and C1j . . . . . . . . . . . . . . . . . . 246
11.4.3 Formulas for Resonant Tunneling Characteristics . . . . . 250
11.6 Comparison of Asymptotic and Numerical Approaches . . . . . . . 260
11.6.1 Problems and Methods for Numerical Analysis . . . . . . 260
11.6.2 Comparison of Asymptotic and Numerical Results . . . . 265
11.6.3 Dependence of Resonant Tunneling on the Magnetic
Field Location in the Resonator . . . . . . . . . . . . . . . . . 268
12 Effect of Magnetic Field on Resonant Tunneling in 3D
Waveguides of Variable Cross-Section . . . . . . . . . . . . . . . . . . . . . 269
.
12.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269
.
12.2 Statement of the Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . 270
.
12.3 Limit Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
.
12.3.1 First Kind Limit Problems . . . . . . . . . . . . . . . . . . . . 273
.
12.3.2 Second Kind Limit Problems . . . . . . . . . . . . . . . . . . 275
.
12.4 Special Solutions of Limit Problems . . . . . . . . . . . . . . . . . . . 277
.
12.5 Asymptotic Formulas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279
.
12.5.1 The Asymptotics of a Wave Function . . . . . . . . . . . . 279
.
S 12 , and C11 ; . . .; C14 . . . . . . . . . . . . 281
12.5.3 Asymptotics for Resonant Tunneling Characteristics . . . 284
Contents xi
13 Asymptotic Analysis of Multichannel Resonant Tunneling . . . . . . . 295

13.1 Statement of the Problem and Limit Problems . . . . . . . . . . . . . 295
13.2 Tunneling in a Waveguide with One Narrow . . . . . . . . . . . . . . 298
13.2.1 Special Solutions to the First Kind Homogeneous
Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298
13.2.2 Asymptotic Formulas . . . . . . . . . . . . . . . . . . . . . . . . . 300
13.3 Tunneling in a Waveguide with Two Narrows . . . . . . . . . . . . . 304
14 Electronics Devices Based on Resonant Tunneling . . . . . . . . . . . . . 313
14.1 Magnetic Field Sensors Based on Quantum Waveguides . . . . . . 314
14.2 Transistors Based on Quantum Waveguides . . . . . . . . . . . . . . . 318
14.3 Electron Flow Switch for Quantum Nets . . . . . . . . . . . . . . . . . 321
Bibliographical Sketch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327
Chapter 1
Introduction
Resonance Structures
As a preliminary, we consider a brief example of resonant tunneling. A resonance

structure where resonant tunneling can occur consists of a potential well bordered by
two potential barriers, an electron source, and a drain. A “one-dimensional” model
of such a structure is exemplified by the Schrödinger equation
2
− (x) + U (x)(x) = E(x), −∞ < x < +∞, (1.0.1)
2m
where U (x) = U1 for x ∈ [x1 , x2 ], U (x) = U2 for [x3 , x4 ], U1 and U2 are positive
constants, and x1 < x2 < x3 < x4 ; moreover, U (x) = 0 for the rest x. The parts of
U over [x1 , x2 ] and [x3 , x4 ] are called potential barriers and [x2 , x3 ] is a potential
well; the barriers and the well comprise a resonator. An electron wave function
satisfies equation (1.0.1), where E is the electron energy, m is the electron mass, and
U is the electron energy. Besides, can be chosen to satisfy the equalities (see, e.g.,
[3, 15, 20]) ikx
e + r e−ikx as x < x1 ,
(x) =
teikx as x > x4 ,
where k = (2m E/2 )1/2 . For x < x1 , functions eikx and r e−ikx are considered as an
incoming wave and a reflected wave, respectively, and, for x > x4 , teikx is a transmit-
ted wave. The values T (E) = |t (E)|2 and R(E) = |r (E)|2 are called a transmission
coefficient and a reflection coefficient. It turns out that T (E) + R(E) = 1; the T (E)
(R(E)) is interpreted as a probability for the electron to transmit through the res-
onator (to be reflected from the resonator). Under certain conditions, there exists a
“resonance” Er es , 0 < Er es < min{U1 , U2 }, such that for E = Er es the transmission
coefficient T takes a maximal value (in particular, T (Er es ) = 1 can be the case).
Then with probability close to 1 the electron transmission can take place, under the
barriers, through the resonator. This phenomenon is called resonant tunneling.
© Springer Nature Switzerland AG 2021 1

L. Baskin et al., Resonant Tunneling, Lecture Notes on Numerical Methods
in Engineering and Sciences, https://doi.org/10.1007/978-3-030-66456-5_1
2 1 Introduction
There is a variety of electronics devices (transistors, key devices, energy

monochromators) based on resonant tunneling. Classic two-barrier resonant devices
use the process of one-dimensional resonant tunneling. Efforts are directed towards
refining production technology and operation properties of resonance structures. At
present, electron tunneling is being studied intensively in the “metallic electrode-
quantum dot-metallic electrode” systems (e.g., see [2, 53]). A quantum dot is a
conductive domain of about 10 nm size and is separated from electrodes by “tunnel”
intervals (vacuum gaps or dielectric layers). Owing to resonant tunneling, the con-
ductivity of such a system can abruptly vary with voltage between the electrodes.
There are prospects for building new nanosize electronics elements that are based
on the aforementioned quantum dot systems and have a frequency-operating range
of around 1012 Hz. However, the properties of such systems heavily depend on
inevitable inhomogeneities of the electrode-vacuum and quantum dot-vacuum inter-
faces. Therefore, the production of the systems must satisfy not easily accessible
accuracy conditions.
The role of resonant structures can be given to quantum wires. Resonant tun-
neling occurs as an electron propagates in a quantum waveguide (wire) of variable
cross-section. Instead of an “electrode-quantum dot-electrode” system, one can use a
quantum wire with two narrows. This can be explained heuristically by the following
reasons. For simplicity, let us consider a waveguide whose cross-section is a disk. If
the waveguide is a cylinder, the full energy of an electron is the sum E = E ⊥ + E ,
E ⊥ being the (quantized) transverse motion energy and E the longitudinal motion
energy; E ⊥ is inversely proportional to the cross-section square. When a waveg-
uide cross-section varies along the axis, the narrows of the waveguide play the role
of effective barriers for the longitudinal motion. Indeed, the full energy E remains
constant. One can consider E = E ⊥ + E as an approximate relation. In a narrow,
E ⊥ is increasing, so E is decreasing. For E ⊥ > E, the electron wave function is
exponentially decaying in the narrow just as it does in electron tunneling under a
potential barrier. The part of the waveguide between two narrows becomes a “res-
onator,” and conditions for electron resonant tunneling can occur. The tunneling
consists of the fact that, for an electron with energy E, the probability T (E) to pass
from one part of the waveguide to the other through the resonator has a sharp peak at
E = Er es , where Er es denotes a “resonant” energy. That resonant tunneling happens
in deformed waveguides was confirmed by numerical experiments in [5, 38].
To analyze the operation of devices based on this phenomenon, it is important
to know Er es , the behavior of T (E) for E close to Er es , the height of the resonant
peak, and its width at the half-height (which is inversely proportional to the so-called
resonator quality factor). Approximate numerical calculations are effective only if the
narrows of a waveguide are “not too narrow” so that the resonant peak is sufficiently
wide. That is why, to obtain a detailed picture of the phenomenon, it is of value to
use both numerical and asymptotics methods which complement each other.
We consider electron propagation in a waveguide with two cylindrical outlets to
infinity and two narrows of small diameters ε1 and ε2 (Fig. 1.1). The boundary of
the waveguide is assumed to be smooth. The electron motion is described by the
Helmholtz equation (or the Pauli system for the electron motion in magnetic field).
1 Introduction 3
Fig. 1.1 The waveguide

with narrows
G(ε , ε )
1 2
ε2
ε
1
In particular, we obtain asymptotic formulas for the aforementioned characteristics

of the resonant tunneling as ε1 and ε2 tend to zero.
Resonant devices based on quantum wires can provide advantages in regard to
both operation properties and production technology. Such a device is homogeneous,
i.e., it is made of one material only. When tunneling, an electron crosses no interfaces
of dielectrics, electrodes, or vacuums. Therefore, the operation of the device is more
stable under small perturbations of its geometry.
Scattering Matrix
The basic characteristics of electron resonant tunneling can be expressed in terms of a

waveguide scattering matrix. Therefore, when studying tunneling, we mainly analyze
the scattering matrix behavior. Chapters 2–4 define scattering matrices, describe their
properties, and present a method for approximate computing of such matrices.
In these chapters, we consider waveguides of somewhat more complicated struc-
ture (with finitely many cylindrical outlets to infinity) than those in the studies of
electron resonant tunneling in the subsequent chapters. In fact, this does not make the
discussion more complicated, rather, it provides possibilities for introducing other
applications (e.g., see the description of an electron flow switch for quantum nets in
Chap. 14).
Chapter 2 presents a radiation principle for the Helmholtz equation in waveguides,
that is the solvability of a boundary value problem with radiation conditions, the
asymptotics of solutions at infinity, and the scattering matrix definition. In essence,
there is given a version (for the Helmholtz equation) of the theory exposed in [43] for
the general self-adjoint elliptic systems in domains with cylindrical ends. (Detailed
references are given in the Bibliographical sketch; as a rule, in the body of the book,
we restrict ourselves to technical references.)
We are now going to define a scattering matrix, and to this end we need to consider
certain issues. Let G be a domain in Rn+1 , n = 1, 2, with smooth boundary ∂G
coinciding, outside a large ball, with the union 1+ ∪ · · · ∪ T+ of finitely many
non-overlapping semi-cylinders
r+ = {(y r , t r ) : y r ∈ r , t r > 0},

4 1 Introduction
Fig. 1.2 The waveguide
where (y r , t r ) are local coordinates in r+ and r is a bounded domain in Rn

(Fig. 1.2). We consider the boundary value problem
− (x) − μ(x) = 0, x ∈ G, (1.0.2)

(x) = 0, x ∈ ∂G,

with = n+1 j=1 ∂ /∂x j . We suppose that, under certain conditions, electron wave
2 2
functions satisfy (1.0.2); moreover, the functions are bounded and do not vanish at
infinity. To describe the wave function behavior at infinity, we will use solutions to
the problem in the cylinder
− ( y,t + μ)u(y, t) = 0, (y, t) ∈ × R = , (1.0.3)

u(y, t) = 0, (y, t) ∈ ∂,
where is a domain in Rn and

n
y,t = y + ∂t2 , y = ∂ 2j , ∂ j = ∂/∂ y j .
j=1
Straightforward calculation shows that the nonzero functions
× R (y, t) → exp (±i(μ − τ )1/2 t)ϕ(y)
satisfy (1.0.3) if and only if
− ( y + τ )ϕ(y) = 0, y ∈ , (1.0.4)
ϕ(y) = 0, y ∈ ∂,
that is, ϕ has to be an eigenfunction of problem (1.0.4), whereas τ is the corre-

sponding eigenvalue. The eigenvalues of problem (1.0.4) form an increasing positive
sequence τ1 < τ2 < . . . that tends to +∞. Let us assume, for the time being, that
τ1 < μ < τ2 (recall that τ1 is a simple eigenvalue). We denote by ϕ1 an eigenfunction
corresponding to τ1 , normalized by the condition
1 Introduction 5

|ϕ1 (y)|2 dy = 1,

and set
u± ∓ −1/2
1 (y, t) = (2|λ1 |) exp(iλ∓
1 t)ϕ1 (y) (1.0.5)
with λ±
1 = ±(μ − τ1 )
1/2
. Functions (1.0.5) are bounded, satisfy (1.0.3), and do not
decay at infinity. We will call the u + −
1 (u 1 ) a wave incoming from +∞ (outgoing to
+∞).
For μ ∈ (τ2 , τ3 ), besides u ±
1 in (1.0.5), we have waves of the form
u ± (y, t) = (2|λ∓
2 |)
−1/2
exp(iλ∓
2 t)ψ(y), (1.0.6)
where λ± 2 = ±(μ − τ2 )
1/2
and ψ is an eigenfunction of problem (1.0.4) correspond-
ing to τ2 . The number of pairs of the form (1.0.6) is equal to the multiplicity κ(τ2 )
of the eigenvalue τ2 ; as an eigenfunction ψ, the elements ψ1 , . . . , ψκ(τ2 ) of a basis
in the eigenspace of problem (1.0.4) have to be chosen, subject to the orthogonality
and normalization conditions

ψ p (y)ψq (y) dy = δ p,q , p, q = 1, . . . , κ(τ2 ).

In general, for μ ∈ (τl , τl+1 ), the number of the wave pairs in the cylinder is equal
to κ(τ1 ) + · · · + κ(τl ).
Let μ be different from the eigenvalues of problems (1.0.4) in 1 , . . . , T . Given
μ, we enumerate all wave pairs in the cylinders 1 , . . . , T by the same index
j = 1, 2, . . . , M. Among electron wave functions in G, there exist 1 , . . . , M that
admit the representations

M
l (x) = u l+ (x) + Sl j u −j (x) + O(exp (−δ|x|)) (1.0.7)
j=1
for |x| → ∞, sufficiently small positive δ, and l = 1, . . . , M. The matrix
S(μ) = Sl j (μ) M

j,l=1 (1.0.8)
is called the scattering matrix.

Let us discuss the definition in more detail. The eigenvalues of problems (1.0.4)
for = 1 , . . . , T are called the thresholds (of the waveguide G). Let τ1 denote
the minimal threshold; τ1 > 0. We have defined the scattering matrix S = S(μ) for
μ > τ1 except the thresholds; later, in Chap. 3, it will be defined at the thresholds as
well. The set [τ1 , +∞) is called the waveguide continuous spectrum. Thus, the S is
a matrix-valued function on the continuous spectrum. The size M = M(μ) of S(μ)
depends on μ, remains constant between two neighboring thresholds, and jumps at
the thresholds increasing to +∞ as μ tends to +∞. It will be shown that, at any
6 1 Introduction
threshold τ , there exist both one-sided limits of S(μ) as μ → τ ± 0 and, moreover,

the S is continuous from the right at the threshold τ .
The scattering matrix S(μ) is unitary for every μ ∈ [τ1 , +∞). Given μ, we con-
sider a wave pair u +j , u −j , j = 1, . . . , M(μ), as a scattering channel. The |Sl j (μ)|2 is
interpreted as the probability of an electron, incoming through the lth channel, to go
out through the jth channel.
Remark 1.0.1 For the one-dimensional resonance structure (1.0.1) there are two
related scattering channels: u + − + − +
1 , u 1 and u 2 , u 2 , where u 1 (x) = e
ikx
(u −
1 (x) = e
−ikx
)
+
is an incoming (outgoing) wave to the left of the resonator, and u 2 (x) = e−ikx
(u −
2 (x) = e
ikx
) is an incoming (outgoing) wave to the right of the resonator. Thus,
the scattering matrix is of size 2 × 2.
Method for Approximate Computation of Scattering

Matrices
Next we are going to state the method employed for numerical simulation of resonant
tunneling. In the Introduction, we restrict ourselves to considering the scattering
matrix on a finite interval of the continuous spectrum containing no thresholds. In
Chap. 4, we modify the method to calculate the scattering matrix also in vicinity of
thresholds and present a justification for the method in both of these situations.
Introduce the notation
r,R r r R N r,R
+ = {(y , t ) ∈ : t > R}, G = G \ ∪r =1 +
r r
for large R. Then ∂G R \ ∂G = R = ∪r r,R , where r,R = {(y r , t r ) ∈ r : t r =

R}. We seek the row (Sl1 , . . . , Sl M ) of the scattering matrix S = S(μ). As approxi-
mation to the row, we take the minimizer of a quadratic functional. To construct such
a functional, we consider the problem
− ( + μ)XlR = 0, x ∈ G R ;
XlR = 0, x ∈ ∂G R \ R ;
M
(∂ν + iζ)XlR = (∂ν + iζ)(u l+ + a j u −j ), x ∈ R , (1.0.9)
j=1
where ζ ∈ R \ {0} is an arbitrary fixed number, ν is an outward normal, and

a1 , . . . , a M are complex numbers.
Let us explain the origin of the problem. Being a solution to problem (1.0.2), the
electron wave function l satisfies the first two equations (1.0.9). The asymptotics
(1.0.7) can be differentiated, so
1 Introduction 7

M
(∂ν + iζ)l = (∂ν + iζ)(u l+ + a j u −j ) + O(e−γ R )
j=1
for a j = Sl j . Thus, l satisfies the last equation in (1.0.9) up to an exponentially small

discrepancy. As an approximation for the row (Sl1 , . . . , Sl M ), we take the minimizer
a 0 (R) = (a10 (R), . . . , a 0M (R)) of the functional

M
JlR (a1 , . . . , a M ) = XlR − u l+ − a j u −j ; L 2 ( R )2 , (1.0.10)
j=1
where XlR is a solution to problem (1.0.9). One can expect that a 0j (R, μ) → Sl j (μ)
at exponential rate as R → ∞ and j = 1, . . . , M.
Let [μ , μ ] be an interval of the continuous spectrum without thresholds. In
Chap. 4, we prove, in particular, that for all R ≥ R0 and μ ∈ [μ , μ ] there exists
a unique minimizer a(R, μ) = (a1 (R, μ), . . . , a M (R, μ)) of functional (1.0.10) and
the estimates
|a j (R, μ) − Sl j (μ)| ≤ c()e−R , j = 1, . . . , M, (1.0.11)
hold with some positive constants and c() independent of R and μ.
Asymptotic and Numerical Studies of Resonant Tunneling in

2D Waveguides for Electrons of Small Energy
Chapter 5 begins an asymptotic and numerical study of resonant tunneling. Electrons

propagate in a 2D waveguide that coincides with an infinite strip in a plane having
two identical narrows of the diameter ε and symmetric about the waveguide axis.
Electron wave functions satisfy the Helmholtz equation in the strip and vanish at its
boundary. The electron energy is supposed to be between the first and the second
thresholds, so only one scattering channel relates to each of the waveguide outlets to
infinity. The purpose is to obtain, as ε → 0, the asymptotics for the resonant energy
Er es , the transmission T (E) and reflection R(E) coefficients, and for the resonator
quality factor.
It turns out that such asymptotic formulas depend on the limiting shape of the
narrows. We assume that the limiting waveguide in a neighborhood of each narrow
coincides with two cones intersecting only at their common vertex. We first con-
struct an asymptotics of the corresponding electron wave function by the method of
“compound” asymptotic expansions (the general theory of the method was exposed,
e.g., in [36, 39]). The expansions contain terms of two kinds: the first kind terms
depend on the “slow” variables x and approximate the wave function “far” from
the narrows; the second kind terms depend on the “fast” variables x/ε and serve
8 1 Introduction
Fig. 1.3 Domain
O
ω
as an approximation in a neighborhood of the narrows. The terms are obtained by

solving the so-called first and second kind limit problems, respectively. The analysis
of the obtained expansions enables us to get asymptotic formulas for the mentioned
characteristics of resonant tunneling.
Let us discuss the situation in more detail. To describe the domain G(ε) in R2
occupied by the waveguide, we first introduce two auxiliary domains G and in R2 .
The domain G is the strip
G = R × D = {(x, y) ∈ R2 : x ∈ R = (−∞, +∞); y ∈ D = (−l/2, l/2)}.
We denote by K a double cone with vertex at the origin O that contains the x-axis
and is symmetric about the coordinate axes. The set K ∩ S 1 , where S 1 is a unit circle,
consists of two simple arcs. Assume that contains the cone K and a neighborhood
of its vertex; moreover, outside a large disk (centered at the origin) coincides with
K . The boundary ∂ of is supposed to be smooth (see Fig. 1.3).
Denote by (ε) the domain obtained from by the contraction with center at O
and coefficient ε. In other words, (x, y) ∈ (ε) if and only if (x/ε, y/ε) ∈ . Let
K j and j (ε) stand for K and (ε) shifted by the vector r j = (x 0j , 0), j = 1, 2. We
assume that |x10 − x20 | is sufficiently large so the distance from ∂ K 1 ∩ ∂ K 2 to G is
positive. We put G(ε) = G ∩ 1 (ε) ∩ 2 (ε) (Fig. 1.4).
The wave function of a free electron of energy k 2 satisfies the boundary value
problem
−u(x, y) − k 2 u(x, y) = 0, (x, y) ∈ G(ε),

u(x, y) = 0, (x, y) ∈ ∂G(ε).
Moreover, u is subject to certain radiation conditions at infinity (that correspond, for

example, to an electron wave incoming from −∞).
We set G(0) = G ∩ K 1 ∩ K 2 (Fig. 1.5); thus, G(0) consists of three parts G 0 , G 1 ,
and G 2 , where G 1 and G 2 are infinite domains, while G 0 is a bounded resonator.
1 Introduction 9
Fig. 1.4 Waveguide G(ε)

G(ε)
O O
1 2
Fig. 1.5 The “limit

G1 G0 G2
waveguide” G(0)
O1 O2
The problems
− v(x, y) − k 2 v(x, y) = f, (x, y) ∈ G j ,

v(x, y) = 0, (x, y) ∈ ∂G j , (1.0.12)
where j = 0, 1, 2, are called the first kind limit problems. In the domains j , j =
1, 2, we consider the boundary value problems
w(ξ j , η j )= F(ξ j , η j ), (ξ j , η j ) ∈ j ,
w(ξ j , η j )= 0, (ξ j , η j ) ∈ ∂ j ,
which are called the second kind limit problems; (ξ j , η j ) are Cartesian coordinates
with origin at O j .
We denote by ke2 a simple eigenvalue of problem (1.0.12) in the resonator G 0
and by kr2 (ε) a resonance frequency such that kr2 (ε) → ke2 as ε → 0. For |k 2 − kr2 | =
O(ε2π/ω ) the asymptotic representations hold:
1
T (k, ε) = 1 + O(ε2−τ ) ,
k − kr2 2
2
1 + P2
ε4π/ω

kr2 (ε) = ke2 + Qε2π/ω + O ε4π/ω ,
1
ϒ(ε) = ε4π/ω 1 + O(ε1−τ /2 ) ,
P
where T (k, ε) is the electron transmission coefficient and ϒ(ε) is the width of the
resonant peak at its half-height (which is inversely proportional to the resonator
quality factor), τ being an arbitrarily small positive number; the P and Q are the
products of several constants in the asymptotics of limit problem solutions near
corners or at infinity.
Without numerical values of the constants, the asymptotic formulas provide only a
qualitative picture. To find the constants, one has to solve numerically several bound-
ary value problems. We state the problems and describe a way to solve them. When
the constants are found, the asymptotics can be used as an approximate solution.
10 1 Introduction
However, it remains uncertain for what band of parameters the approximation is reli-
able. On the other hand, one should expect numerical approach to be efficient only if
the waveguide narrows are not too small in diameter and if the resonant peak of the
transmission coefficient is sufficiently wide. Therefore a detailed picture of resonant
tunneling can be achieved when the asymptotic and numerical approaches are com-
bined. Independently of asymptotic approach, an approximation to the waveguide
scattering matrix is calculated. For that purpose, we employ the method from Chap.
4. Then we can compare the asymptotics with calculated constants and the scattering
matrix (the transition and reflection coefficients). It turns out, that there is an interval
for ε, where the asymptotic and numerical results practically coincide. To the right
of the interval, the asymptotics vanishes but the numerical method for calculation of
the scattering matrix is effective; to the left of the interval, the numerical method is
ill-conditioned while the asymptotics is reliable.
Asymptotics of Resonant Tunneling in 2D-Waveguide with

Several Resonators
In this chapter, we consider a two-dimensional waveguide that coincides with a strip

having n + 1 narrows of small diameter ε. All narrows are of the same shape and
are spaced from each other by equal distances (see Fig. 1.6). Parts of the waveguide
between two neighboring narrows play the role of resonators. The wave function
of a free electron satisfies the Dirichlet boundary value problem for the Helmholtz
equation in the waveguide.
Let ke2 be a simple eigenvalue of a closed resonator. Then there are n distinct
2
resonant energies kr,l , l = 1, . . . , n, located in the ε2π/ω -vicinity of ke2 ,
2
kr,l (ε) = ke2 − Q l ε2π/ω + O(ε4π/ω ),
πl
Q l = f + 2g cos .
n+1
The coefficients f and g here do not depend neither on n, nor on ε.

In a neighborhood of the resonances, the transmission coefficient T (k, ε) admits
the following representation
Fig. 1.6 Waveguide G(ε)

1 Introduction 11
1
T (k, ε) = 1 + O(ε4π/3ω ) ,
n 2 2
l=1 k − kr,l
2
1+ P2
ε2(n+1)π/ω
where P is a constant coefficient. The height of each of the resonant peaks is 1 +

O(ε4π/3ω ), while the width ϒl of the peak located at k 2 = kr,l
2
is
4 πl 4π/ω
ϒl = sin2 ε + O(ε6π/ω ).
(n + 1)g n−1 P n+1
Asymptotics of the Resonant Tunneling of High-Energy

Electrons in 2 D-waveguides
We consider a waveguide in R2 having two identical narrows of small diameter ε as

in Chap. 5. The energy of electrons (spectral parameter) may be rather high, i.e. any
(fixed) number of waves can propagate in the strip far from the narrows. The spectral
parameter varies in the vicinity of a degenerate eigenvalue of the resonator.
In the case of a degenerate eigenvalue, the leading term of the transmission coef-
ficient has two peaks. The asymptotics of the transmission coefficient has the form
Pε8π/ω (k 2 − ke2 )2
T (k, ε) = + O(ε4π/ω ),
[(k 2 − 2 2
kr,1 ) + Q 1 ε8π/ω ] [(k 2 − kr,2
2 2
) + Q 2 ε8π/ω ]
where k 2 is the spectral parameter, ke2 is an eigenvalue of the resonator,
kr,2 j = ke2 − f j ε2π/ω + O(ε4π/ω ), j = 1, 2;
the constants ω, P, Q j , and f j are positive, they depend only on the waveguide
geometry; τ is a positive number. Note that the leading term of the asymptotics
vanishes at k = ke , i.e. the resonant reflection occurs, which was not observed in the
case of a simple eigenvalue of the resonator.
Numerical Simulation of High Energy Electron Resonant

Tunneling, the Fano Resonances
Chapter 8 is devoted to the numerical simulation of high energy electron scatter-

ing. We consider multi-channel resonant tunneling. An electron wave of energy E
incident on a resonator with transverse quantum number n passes through the res-
onator and arises with transverse number k; shortly, the wave passes from state n
12 1 Introduction
to state k. We denote by Tnk (E) the transmission coefficient of the wave, calculate
the dependence E → Tnk (E) by computing the scattering matrix S(E), and obtain
Tnk (E) = |Snk (E)|2 , where Snk (E) is the entry of S(E). The curve E → Tnk (E)
can be sufficiently complicated and not always easily interpreted. To explain the
curve, we consider Snk (E) as a probability amplitude and represent it in the form
Snk (E) = s Ansk (E), where Ansk (E) is the probability amplitude of the transmis-
sion from n to k through an intermediate state s; the summation is over all intermediate
states (cf. [21]).
As before, we denote by G(ε) a waveguide with two narrows and let G 0 be
the closed resonator, that is, the bounded part of the limit waveguide G(0)(see
Fig. 1.5); generally, the resonator form may be arbitrary. We denote by k12 ≤ k22 ≤ . . .
the eigenvalues of problem (1.0.12) with j = 0 numbered according to their mul-
tiplicities. Then the resonant energies of the waveguide G(ε) form the sequence
ReE 1 , ReE 2 , . . . , where E 1 , E 2 , . . . can be viewed as the “perturbed” k12 , k22 , . . .
and ImE j < 0 for all j = 1, 2, . . . . The amplitude Ansk admits the representation
(s)
(s) Rnk (E)
Ansk (E) = Hnk (E) +
E − Es
(s) (s)
with continuous functions E → Hnk (E) and E → Rnk (E). In a small neighborhood
of ReEr ,
Rnk (Er ) 2
|Snk (E)|2 = | Ansk (E)|2 ≈ |Hnk (Er ) + | ≡ Tnk (E),
s
E − Er
where Hnk (Er ) and Rnk (Er ) are constant. We take the function Tnk (E) as an approx-
imation to the calculated |Snk (E)|2 and find the constants Hnk (Er ), Rnk (Er ), and Er
by the method of least squares.
The Impact of a Finite Waveguide Work Function on

Resonant Tunneling
This subject is discussed in Chap. 9. When considering electron transport in a waveg-

uide, we assume that the electron wave functions vanish at the waveguide boundary.
This means that an electron can not get out of the waveguide because of the infinite
potential barrier at the boundary. In reality, the assumption has never been fulfilled:
the surface potential barrier is always of a finite height and some electrons can pen-
etrate through the waveguide boundary and go away some distance from the waveg-
uide. In other words, in reality we deal with a waveguide of a finite work function.
Due to this phenomenon, the effective widths of a waveguide and waveguide narrows
are greater than their geometric widths. Therefore, to draw a conclusion about the
1 Introduction 13
adequacy of the boundary condition used in the mathematical model, we have to

clarify the impact of a finite waveguide work function on the resonant tunneling.
To this end, we present some physics preliminaries concerning work functions,
introduce a boundary value problem with regard to a finite work function, and ana-
lyze the problem numerically. The results show the need, when employing resonant
tunneling in a waveguide with narrows, to restrict somewhat the range of narrow
parameters and that of electron energy. In particular, by decreasing the narrow diam-
eter at a resonator, one can not diminish the effective narrow diameter beyond a
certain critical value. This restricts the possibility to improve the resonator quality
factor by diminishing the narrow diameter. The angle of a wedge-like narrow should
not be too small. However, increasing the angle causes an increase in the effective
width of the potential barrier and a decrease in the width of the resonant peak. This
increases the resonant tunneling time and affects the frequency properties of the
system. Optimal angles for wedge-like narrows range between 20◦ and 35◦ .
Electrons of high energies penetrate the waveguide walls more effectively, hence,
the model with infinite work function (i.e. with the homogeneous Dirichlet bound-
ary condition) becomes inadequate if the energy of electrons exceeds several first
thresholds.
Asymptotic Study of Resonant Tunneling in 3D Waveguides

for Electrons of Small Energy
In Chap. 10, we consider 3D waveguide with two non-overlapping cylindrical outlets

C1 and C2 to infinity; the axes of the outlets may be of any directions. There are two
waveguide narrows, one narrow in C1 and the other one in C2 . Generally, the narrow
diameters ε1 and ε2 are different. The resonator (that is, the waveguide part between
the narrows) can be of arbitrary form. The boundary of the waveguide is supposed
to be smooth. We denote the waveguide by G(ε1 , ε2 ). The limit set G(0, 0) consists
of unbounded parts G 1 , G 2 , and a bounded resonator G 0 . In a neighborhood of the
point O j = G 0 ∩ G j , the set G(0, 0) coincides with a double cone K j , j = 1, 2.
A wave function of a free electron of energy E = 2 k 2 /2m satisfies the boundary
value problem
−u − k 2 u = 0 in G(ε1 , ε2 ), u = 0 on ∂G(ε1 , ε2 ),
and certain radiation conditions at infinity. We consider the scattering of a wave

coming from C1 and seek the resonant values kr = kr (ε1 , ε2 ) of the parameter k,
where the transition coefficient T = T (k, ε1 , ε2 ) takes maximal values.
Let ke2 be a simple eigenvalue (between the first and second thresholds) of the
boundary value problem in the resonator,
−v(x) − k 2 v(x) = f, x ∈ G 0 ; v(x) = 0, x ∈ ∂G 0 .

14 1 Introduction
Near such an eigenvalue there is a resonant value kr (ε1 , ε2 ) satisfying

kr2 (ε1 , ε2 ) = ke2 + D1 εν11 + D2 εν22 + O εν11 +τ1 + εν22 +τ2
as ε1 , ε2 → 0. The coefficients D1 and D2 are constant, ν j and τ j are some positive

2μ +1+τ1 2μ +1+τ2
numbers, j = 1, 2. Under the condition |k 2 − kr2 | = O ε1 11 + ε2 21 , the
transition coefficient T1 (k, ε1 , ε2 ) admits the asymptotics
2 −1
1 1 2 k2 − k2
T1 (k, ε1 , ε2 ) = z+ + P 2 2μ11 +1 2μr21 +1 1 + O(ετ11 + ετ22 ) ,
4 z ε1 ε2
2μ +1 2μ +1
where τ j are the same as in (10.1.6), z = Qε1 11 /ε2 21 , while P and Q are
constant. The width of the resonant peak at its half-height (calculated for the principal
part in the asymptotics of T ) is

1 1 2μ11 +1 2μ21 +1
ϒ(ε1 , ε2 ) = z+ ε ε2 (1 + O(ετ11 + ετ22 )).
P z 1
Electron Resonant Tunneling in the Presence of Magnetic

Fields
The presence of a magnetic field can essentially affect the basic characteristics of
the resonant tunneling and bring new possibilities for applications in electronics. In
particular, in the presence of a magnetic field, the tunneling phenomenon is feasible
for producing spin-polarized electron flows consisting of electrons with spins of the
same direction. In Chaps. 11 and 12 we consider the same 2D and 3D waveguides with
narrows as in Chaps. 5 and 10, respectively. A part of the resonator is occupied by a
homogeneous magnetic field. An electron wave function satisfies the Pauli equation
in a waveguide and vanishes on its boundary. An electron energy is in between
the first and the second thresholds. The asymptotics of basic resonant tunneling
characteristics are presented as the narrow diameters tend to zero. Moreover, in
Chap. 11, the asymptotic results for 2D waveguides are compared with numerical
ones obtained by approximate computing the scattering matrix; there is an interval
of ε (the narrow diameter) where the asymptotic and numerical results practically
coincide. Using the approximate scattering matrix, we also observe the dependence
of the tunneling characteristics on a magnetic field position in the resonator.
1 Introduction 15
Asymptotic Analysis of Multichannel Resonant Tunneling
In Chap. 13, for electrons of high energy, we generalize the asymptotic theory exposed
in Chap. 10. We present and justify the asymptotics of tunneling characteristics as
the narrow diameters tend to zero.
Electronics Devices Based on Resonant Tunneling in

Waveguides of Variable Cross-Sections
Chapter 14 presents electronics devices based on a quantum waveguide with narrows:

transistors controlled by external electric field and magnetic field sensors controlled
by external magnetic field. Besides, we describe an electron flow switch for quantum
nets. The switch is not related to resonant tunneling, however, the description is based
on analyzing the corresponding scattering matrix calculated by the method of Chap. 4.
We present the switch to demonstrate the method.
Chapter 2
Waveguides. Radiation Principle.
Scattering Matrices
First, we briefly outline the chapter content. Section 2.1 is devoted to the boundary
value problem
(− − μ)u(y, t) = f (y, t), (y, t) ∈ ,

u(y, t) = 0, (y, t) ∈ ∂, (2.0.1)
in the cylinder = {(y, t) : y = (y1 , . . . , yn ) ∈ , t ∈ R}, where is a bounded

domain in Rn with smooth boundary and μ ∈ R. The Fourier transform
+∞
v (λ) = (2π)−1/2
exp (−iλt)v(t) dt (2.0.2)
−∞
reduces the problem to the family of problems depending on the parameter λ:
u (y, λ) =
(− y + λ2 − μ) f (y, λ), y ∈ ,

u (y, λ) = 0, y ∈ ∂. (2.0.3)
If the inverse operator A(λ, μ)−1 of problem (2.0.3) exists for all λ ∈ R, the μ being
fixed, we obtain a solution u to problem (2.0.1) of the form
+∞
u(·, t) = (2π) −1/2
exp (iλt)A(λ, μ)−1
f (·, λ) dλ. (2.0.4)
−∞
However, the spectrum of the pencil λ → A(λ, μ), that is, the set of numbers λ such
that the operator A(λ, μ) is not invertible, consists of an imaginary number sequence
accumulating at infinity and, for sufficiently large μ, additionally contains finitely
many real numbers. Therefore, formula (2.0.4) can fail and we will use the complex
Fourier transform
© Springer Nature Switzerland AG 2021 17

L. Baskin et al., Resonant Tunneling, Lecture Notes on Numerical Methods
in Engineering and Sciences, https://doi.org/10.1007/978-3-030-66456-5_2
18 2 Waveguides. Radiation Principle. Scattering Matrices

v (λ) = (2π)−1/2
exp (−iλt)v(t) dt, λ ∈ R + iβ,
R
where R + iβ = {λ ∈ C : Imλ = β}; there are the inversion formula

v(t) = (2π)−1/2 exp (iλt)v (λ)dλ
R+iβ
and the Parseval equality

exp (2βt)|v(t)|2 dt = |
v (λ)|2 dλ.
R R+iβ
Let us assume that the line R + iβ is free from the spectrum of A(·, μ) and the f in
(2.0.1) satisfies the condition

exp (2βt)| f (y, t)| dydt =
2
|
f (y, λ)|2 dydλ < ∞.
R+iβ
Then, according to Theorem 2.1.4, there exists a unique solution u to problem (2.0.1)
such that
u(·, t) = (2π)−1/2
exp (iλt)A(λ, μ)−1 f (·, λ) dλ (2.0.5)
R+iβ
and the inequality

exp (2βt)|∂tk ∂ yα u(y, t)|2 dydt ≤ C exp (2βt)| f (y, t)|2 dydt
|α|+k≤2
holds with a constant C independent of f .

These considerations motivate the statement of the boundary value problem in the
domain G with cylindrical ends
− u(x) − μu(x) = f (x), x ∈ G,

u(x) = 0, x ∈ ∂G, (2.0.6)
in function spaces with weighted norms (see Fig. 1.2 and the definition of G just after
the figure). For integer l ≥ 0, we denote by H l (G) the Sobolev space with norm
l
1/2
v; H (G) =
l
|Dxα v(x)|2 d x .
j=0 G |α|= j
For real β, we denote by ρβ a smooth positive function on G given by the equality

ρβ (x) = exp(β|x|) for large |x|. We also introduce the space Hβl (G) with norm
Another random document with
no related content on Scribd:
CHAPTER XVI.
CAUSES OF THE AFRICANDERS’ SECOND WAR
OF INDEPENDENCE.—CONCLUDED.
The “later dispatch” promised by the British cabinet was never sent.
The answer to it of the Transvaal government was, therefore,
delayed for several days, awaiting the new proposals that were to
come as the result of further deliberations on the part of Her
Majesty’s government. At last, on the eve of the outbreak of war, Mr.
Chamberlain gave out, on the 10th of October, the text of the
republic’s rejoinder to the British cabinet’s note of the 25th of
September. It was transmitted by cable, through Sir Alfred Milner,
and read thus:
“Dear Sir: The government of the South African Republic feels itself
compelled to refer the government of Her Majesty, Queen of Great
Britain and Ireland, once more to the convention of London, 1884,
concluded between this republic and the United Kingdom, and
which, in Article [242]XIV., secures certain specific rights to the white
population of this republic—namely: That all persons other than
natives, on conforming themselves to the laws of the South African
Republic—
“A—Will have full liberty, with their families, to enter, travel or reside
in any part of the South African Republic.
“B—They will be entitled to hire or possess houses, manufactories,

warehouses, shops and other premises.
“C—They may carry on their commerce either in person or by any

agents whom they may think fit to employ.
“D—They shall not be subject, in respect of their premises or
property or in respect of their commerce and industry, to any taxes
other than those which are or may be imposed upon the citizens of
the said republic.
“This government wishes further to observe that these are the only
rights which Her Majesty’s government has reserved in the above
convention with regard to the outlander population of this republic,
and that a violation only of those rights could give that government a
right to diplomatic representations or intervention; while, moreover,
the regulation of all other questions [243]affecting the position of the
rights of the outlander population under the above-mentioned
convention is handed over to the government and representatives of
the people of the South African Republic.
“Among the questions the regulation of which falls exclusively within

the competence of this government and of the Volksraad are
included those of the franchise and the representation of the people
in this republic; and, although this exclusive right of this government
and of the Volksraad for the regulation of the franchise and the
representation of the people is indisputable, yet this government has
found occasion to discuss, in friendly fashion, the franchise and
representation of the people with Her Majesty’s government—
without, however, recognizing any right thereto on the part of Her
Majesty’s government.
“This government has also, by the formulation of the now existing

franchise law and by a resolution with regard to the representation,
constantly held these friendly discussions before its eyes. On the
part of Her Majesty’s government, however, the friendly nature of
these discussions has assumed more and more a threatening tone,
and the minds of the people of this republic and [244]the whole of
South Africa have been excited and a condition of extreme tension
has been created, owing to the fact that Her Majesty’s government
could no longer agree to the legislation respecting the franchise and
the resolution respecting representation in this republic, and, finally,
by your note of Sept. 25, 1899, which broke off all friendly
correspondence on the subject and intimated that Her Majesty’s
government must now proceed to formulate its own proposals for the
final settlement.
“This government can only see in the above intimation from Her
Majesty’s government a new violation of the convention of London,
1884, which does not reserve to Her Majesty’s government the right
to a unilateral settlement of a question which is exclusively a
domestic one for this government, and which has been already
regulated by this government.
“On account of the strained situation and the consequent serious

loss in and interruption of trade in general, which the
correspondence respecting franchise and the representation of the
people of this republic has carried in its train, Her Majesty’s
government has recently pressed for an early settlement, and finally
pressed, by your intervention, for an answer within forty-eight
[245]hours, a demand subsequently somewhat modified, to your note
of September 12, replied to by the note of this government of
September 15, and to your note of September 25, 1899, and
thereafter further friendly negotiations were broken off, this
government receiving an intimation that a proposal for a final
settlement would shortly be made.
“Although this promise was once more repeated, the proposal, up to

now, has not reached this government.
“Even while this friendly correspondence was still going on the

increase of troops on a large scale was introduced by Her Majesty’s
government, the troops being stationed in the neighborhood of the
borders of this republic.
“Having regard to occurrences in the history of this republic, which it

is unnecessary here to call to mind, this republic felt obliged to
regard this military force in the neighborhood of its borders as a
threat against the independence of the South African Republic, since
it was aware of no circumstances which could justify the presence of
such a military force in South Africa and in the neighborhood of its
borders.
“In answer to an inquiry with respect thereto, addressed to His

Excellency, the High Commissioner, [246]this government received, to
its great astonishment, in answer a veiled insinuation that from the
side of the republic an attack was being made on Her Majesty’s
colonies, and, at the same time, a mysterious reference to
possibilities whereby this government was strengthened in its
suspicion that the independence of this republic was being
threatened.
“As a defensive measure this government was, therefore, obliged to

send a portion of the burghers of this republic in order to offer
requisite resistance to similar possibilities.”
It will be seen from this correspondence that the British government

had failed to send the formulation of “its own proposals for a final
settlement” promised in the note of September 25, and that active
preparations for war, even to the mobilization of troops, had been
going on—on both sides—for some weeks.
On the 7th of August, forty-nine days before the British cabinet

engaged to prolong friendly diplomatic correspondence on the
subjects at issue by promising a later dispatch containing its own
proposals for a final settlement, Mr. Chamberlain delivered a speech
in the House of Commons which has become historic—a speech
which signified past all possibility of mistake that at [247]that early
date war was a foregone conclusion. After deprecating the use of the
word “war” unless it were absolutely necessary, he went on to say:
“The government had stated that they recognized the grievances

under which their subjects in Africa were laboring. They had stated
that they found those grievances not only in themselves a serious
cause for interposition, but a source of danger to the whole of South
Africa.
“They (the government) said that their predominance, which both

sides of the House had constantly asserted, was menaced by the
action of the Transvaal government in refusing the redress of
grievances, and in refusing any consideration of the requests
hitherto put in the most moderate language of the suzerain power.
They said that that was a state of things which could not be long
tolerated. They had said: ‘We have put our hands to the plow and we
will not turn back,’ and with that statement I propose to rest content.”
Language could not be plainer. It was the British government’s

demand that the South African Republic must accept British control
of her internal affairs—of affairs so purely domestic as the franchise
and the representation of her citizens[248]—or fight. It is not a little
remarkable in this connection that Germany, France, the United
States of America and other powerful nations whose subjects were
mingled with the English in that vast foreign population in the
Transvaal, heard of no grievances inflicted on their subjects by the
South African Republic sufficient to call forth even a friendly
diplomatic representation and request for redress.
On the morning of August the 8th, the day after Mr. Chamberlain’s
warlike speech, the London papers announced that the Liverpool
and Manchester regiments, then at the Cape, had been ordered to
Natal; that the Fifteenth Hussars were to embark on the 23d of
August, and that troops were to be massed along the Transvaal
frontier. On the 11th of August it was announced that 12,000 British
troops were to be dispatched from India to South Africa, and on the
same day a large consignment of war stores, including medical
requisites, was given out from the royal arsenal, Woolwich, for
shipment to Natal, and the sum of $2,000,000 in gold was sent to
South Africa for the War Office account. British troops began to
arrive in South Africa from India and from England in the first week of
October. By the 10th some 15,000 had landed. [249]These were
hurried to the frontiers of the Orange Free State—both west and east
—most of them being concentrated along the northern boundary of
Natal, convenient to the southern frontier of the Transvaal.
The government of the South African Republic made no mistake as

to the meaning of Chamberlain’s belligerent speech in the House of
Commons. On the 8th of August orders were given for the purchase
of 1,000 trek oxen, to be used in the operations of the commissary
department. On the 11th the German steamer Reichstag arrived at
Lorenzo Marquez with 401 cases of ammunition. On the 12th it was
decided to proceed at once with the construction of fortified camps at
Laing’s Nek and Majuba Hill, and orders were issued for the
preparation of armored trains. The mobilization of artillery was begun
on the 13th, and the next day that force went into camps of
instruction to learn the handling of guns of the latest pattern. On the
14th of August the Field Cornets were ordered to distribute Mauser
rifles to the burghers, and the government began the purchase of
mules, provisions and general war supplies. Large quantities of arms
and ammunition were dispatched on the 15th of August to
Oudtshoorn, Aliwal Bethany, [250]and other points in Cape Colony
and the Orange Free State for the use of any Africanders who
should rise against Great Britain when hostilities began. On the 19th
of August another German steamer, the Kœnig, arrived in Delagoa
Bay with 2,000 cases of cartridges for the Transvaal government.
The same day fifty cases of ammunition each were dispatched to
Kimberley, Jagersfontein and Aliwal North for the arming of
sympathizers in those districts of Cape Colony. On the same day
300 Transvaal artillerists, with guns, ammunition and camp
equipage, left Johannesburg for Komati Pass, in the Libombo
Mountains.
And so it went on during the “friendly diplomatic correspondence,”

which terminated on the 25th of September—awaiting the “later
dispatch” from the British cabinet, which never came; both sides
arming and maneuvering for strategic advantages in preparation for
the struggle that was seen to be inevitable.
Perceiving that all the days spent in waiting for that “later dispatch”
were being used by Great Britain in massing her gigantic powers of
war in South Africa and along the Transvaal frontier, and believing
that no such dispatch would now come until the points of war were
all [251]secured by his great antagonist, President Kruger at last
astonished the world—and, most of all, Great Britain—by issuing an
ultimatum sufficiently bold and defiant to have come from any of the
first-rate powers of the earth.
The document was dated 5 o’clock, p. m., on Monday, October the

9th, and read as follows:
“Her Majesty’s unlawful intervention in the internal affairs of this

republic, in conflict with the London convention of 1884, and by the
extraordinary strengthening of her troops in the neighborhood of the
borders of this republic, has caused an intolerable condition of things
to arise, to which this government feels itself obliged, in the interest
not only of this republic, but also of all South Africa, to make an end
as soon as possible.
“This government feels itself called upon and obliged to press
earnestly and with emphasis for an immediate termination of this
state of things, and to request Her Majesty’s government to give
assurances upon the following four demands:
“First—That all points of mutual difference be regulated by friendly

recourse to arbitration or by whatever amicable way may be agreed
upon by this government and Her Majesty’s government. [252]
“Second—That all troops on the borders of this republic shall be

instantly withdrawn.
“Third—That all re-enforcements of troops which have arrived in

South Africa since June 1, 1899, shall be removed from South Africa
within a reasonable time, to be agreed upon with this government,
and with the mutual assurance and guaranty on the part of this
government that no attack upon or hostilities against any portion of
the possessions of the British government shall be made by this
republic during the further negotiations within a period of time to be
subsequently agreed upon between the governments; and this
government will, on compliance therewith, be prepared to withdraw
the burghers of this republic from the borders.
“Fourth—That Her Majesty’s troops which are now on the high seas
shall not be landed in any part of South Africa.
“This government presses for an immediate and an affirmative

answer to these four questions and earnestly requests Her Majesty’s
government to return an answer before or upon Wednesday,
October 11, 1899, not later than 5 o’clock p. m.
“It desires further to add that in the unexpected event of an answer

not satisfactory being [253]received by it within the interval, it will with
great regret be compelled to regard the action of Her Majesty’s
government as a formal declaration of war and will not hold itself
responsible for the consequences thereof, and that, in the event of
any further movement of troops occurring within the above-
mentioned time in a nearer direction to our borders, this government
will be compelled to regard that also as a formal declaration of war.”
This document was signed by F. W. Reitz, State Secretary, and

handed by him to Mr. Conyngham Greene, Her Majesty’s agent at
Pretoria. On Wednesday afternoon, October the 11th, at 3 o’clock,
Mr. Greene delivered the reply of his government, which read thus:
“Her Majesty’s government declines even to consider the peremptory

demands of the Transvaal government.”
Within an hour the telegraphic wires had flashed through all the
South African Republic the ominous word “Oorlog”—war!
Mr. Conyngham Greene at once asked for his passport, and on the
next day, October the 12th, with his family, he was sent, attended by
a guard of honor, to the border of the Orange Free State, [254]where
a similar guard received and conducted him to British territory in
Cape Colony.
Thursday, the 12th of October, was a busy and exciting day in both
the Transvaal and the Orange Free State—for the two republics
stood as one in the struggle. That night—twenty-four hours after war
had been declared—30,000 burghers were on the borders ready to
do battle. Of these 20,000 invaded Natal under General Joubert, and
the vanguard under General Kock occupied Newcastle on the 13th
of October. The other 10,000, under General Piet Cronje, crossed
the western border into British Bechuanaland and marched on
Mafeking.
Thus, and for the causes set forth, began the Africanders’ Second
War of Independence. It was not in the proposed scope of this book
to treat of its fortunes. The prospect is that it will be a long and
sanguinary war. The story of it will afford abundant and interesting
matter for a later volume.
It only remains to show that in all the matters in dispute between the
government of the Transvaal and that of Great Britain, and in the war
which resulted therefrom, the two Africander republics acted in
solidarity. Early in November, 1899, the President of the Orange
Free State [255]announced this to his people and to the world in the
following proclamation:
“Burghers of the Orange Free State: The time which we had so

much desired to avoid—the moment when we as a nation are
compelled with arms to oppose injustice and shameless violence—is
at hand. Our sister republic to the north of the Vaal river is about to
be attacked by an unscrupulous enemy, who for many years has
prepared herself and sought pretexts for the violence of which he is
now guilty, whose purpose is to destroy the existence of the
Africander race.
“With our sister republic we are not only bound by ties of blood, of
sympathy and of common interests, but also by formal treaty which
has been necessitated by circumstances. This treaty demands of us
that we assist her if she should be unjustly attacked, which we
unfortunately for a long time have had too much reason to expect.
We therefore cannot passively look on while injustice is done her,
and while also our own dearly bought freedom is endangered, but
are called as men to resist, trusting the Almighty, firmly believing that
He will never permit injustice and unrighteousness to triumph.
“Now that we thus resist a powerful enemy, with whom it has always
been our highest desire [256]to live in friendship, notwithstanding
injustice and wrong done by him to us in the past, we solemnly
declare in the presence of the Almighty God that we are compelled
thereto by the injustice done to our kinsmen and by the
consciousness that the end of their independence will make our
existence as an independent state of no significance, and that their
fate, should they be obliged to bend under an overwhelming power,
will also soon after be our own fate.
“Solemn treaties have not protected our sister republic against

annexation, against conspiracy, against the claim of an abolished
suzerainty, against continuous oppression and interference, and now
against a renewed attack which aims only at her downfall.
“Our own unfortunate experiences in the past have also made it

sufficiently clear to us that we cannot rely on the most solemn
promises and agreements of Great Britain, when she has at her
helm a government prepared to trample on treaties, to look for
feigned pretexts for every violation of good faith by her committed.
This is proved among other things by the unjust and unlawful British
intervention, after we had overcome an armed and barbarous black
tribe on our eastern frontier, as also by the forcible appropriation
[257]of the dominion over part of our territory where the discovery of
diamonds had caused the desire for this appropriation, although
contrary to existing treaties. The desire and intention to trample on
our rights as an independent and sovereign nation, notwithstanding
a solemn convention existing between this state and Great Britain,
have also been more than once and are now again shown by the
present government, by giving expressions in public documents to
an unfounded claim of paramountcy over the whole of South Africa,
and therefore also over this state.
“With regard to the South African Republic, Great Britain has

moreover refused until the present to allow her to regain her original
position in respect to foreign affairs, a position which she had lost in
no sense by her own faults. The original intention of the conventions
to which the republic had consented under pressure and
circumstances has been perverted and continually been used by the
present British administration as a means for the practice of tyranny
and of injustice, and, among other things, for the support of a
revolutionary propaganda within the republic in favor of Great Britain.
“And while no redress has been offered, as justice demands, for

injustice done to the South [258]African Republic on the part of the
British government; and while no gratitude is exhibited for the
magnanimity shown at the request of the British government to
British subjects who had forfeited under the laws of the republic their
lives and property, yet no feeling of shame has prevented the British
government, now that the gold mines of immense value have been
discovered in the country, to make claims of the republic, the
consequence of which, if allowed, will be that those who—or whose
forefathers—have saved the country from barbarism and have won it
for civilization with their blood and their tears, will lose their control
over the interests of the country to which they are justly entitled
according to divine and human laws. The consequence of these
claims would be, moreover, that the greater part of the power will be
placed in the hands of those who, foreigners by birth, enjoy the
privilege of depriving the country of its chief treasure, while they
have never shown any loyalty to a foreign government. Besides, the
inevitable consequence of the acceptance of these claims would be
that the independence of the country as a self-governing,
independent sovereign republic would be irreparably lost. For years
past British troops in great numbers have been placed on the
[259]frontiers of our sister republic in order to compel her by fear to
accede to the demands which would be pressed upon her, and in
order to encourage revolutionary disturbances and the cunning plans
of those whose greed for gold is the cause of their shameless
undertakings.
“Those plans have now reached their climax in the open violence to
which the present British government now resorts. While we readily
acknowledge the honorable character of thousands of Englishmen
who loathe such deeds of robbery and wrong, we cannot but abhor
the shameless breaking of treaties, the feigned pretexts for the
transgression of law, the violation of international law and of justice
and the numerous right-rending deeds of the British statesmen, who
will now force a war upon the South African Republic. On their heads
be the guilt of blood, and may a just Providence reward all as they
deserve.
“Burghers of the Orange Free State, rise as one man against the
oppressor and the violator of right!
“In the strife to which we are now driven have care to commit no
deed unworthy of a Christian and of a burgher of the Orange Free
State. Let us look forward with confidence to a fortunate end of this
conflict, trusting to the [260]Higher Power without whose help human
weapons are of no avail.
“May He bless our arms. Under His banner we advance to battle for
liberty and for fatherland.
M. T. Steyn, State President.” [261]

[Contents]
CHAPTER XVII.
THE COUNTRY OF THE AFRICANDERS.
Some knowledge of the physical structure of South Africa is necessary to an

understanding of its resources, economic conditions and the longstanding
political problems which, to all appearance, are now nearing a final solution.
Nature has divided that part of Africa lying south of the Zambesi River into three
distinct and well-defined regions. A strip of lowland skirts the coast of the Indian
Ocean all the way from Cape Town around to Natal, Delagoa Bay and still
northeast to the mouths of the Zambesi. Between Durban, the principal port of
Natal, and Cape Town this strip is very narrow in places—the hills coming down
almost to the margin of the sea. Beyond Durban, to the northeast, the low plain
grows wider. This belt of lowland is more or less swampy, and from Durban
northward is exceedingly malarious and unhealthful. This feature is a prime
factor in the physical [262]structure of the country and has had much to do with
shaping its history.
The second region is composed of the elevated and much broken surface
presented by the Drakensburg or Quathlamba range of mountains, reaching
from Cape Town to the Zambesi Valley—a distance of sixteen hundred miles. In
traveling inland, after leaving the level belt, at from thirty to sixty miles from the
sea the hills rise higher and higher—from three thousand to six thousand feet.
These hills are only the spurs of the principal range, some of whose peaks rise
to an elevation of eleven thousand feet.
Beyond the Quathlamba Mountains, to the west and north, is the third natural
division of South Africa—a vast tableland or plateau, varying from three
thousand to five thousand feet above the sea level. This region occupies about
seven-eighths of the area of South Africa.
To a bird’s-eye view of the country the physical scheme is exceedingly simple—

a great plateau filling the interior, a belt of lowland bordering the Indian Ocean
and one principal mountain range between the two.
Geologically considered, the oldest formation is found in the northern part of the
tableland and toward the northeastern end of the Quathlamba [263]Mountains.
The principal formations in this region are granite and gneiss, believed to be of
great antiquity—probably of the same age as the Laurentian formations in
America. The rocks of the Karoo district are not so ancient. There are no traces
anywhere in South Africa of late volcanic action, nor has any active volcano
been discovered there; but eruptive rocks of ancient date—porphyries and
greenstones—are found overlying the sedimentary deposits in the Karoo district
and in the mountain systems of Basutoland and the Orange Free State.
The African coast is notably poor in harbors. There is no haven between Cape
Town and Durban. From Durban to the Zambesi there are but two good ports—
that of Delagoa Bay and Beira. With the exception of Saldanha Bay, twenty
miles north of Cape Town, the western coast, for a thousand miles, has no
harbor.
The temperature in Southern Africa is much lower than the latitude would lead
one to expect. This is accounted for by the fact that there is a vast
preponderance of water in the southern hemisphere, which has the effect of
giving a cooler temperature than prevails in a corresponding northern latitude.
The difference in both heat and cold represents over two degrees of difference
[264]in latitude. Thus, Cape Town, 34° S., has a lower temperature in both
summer and winter than Gibraltar and Aleppo, in 36° N. Nevertheless, the
thermometer registers high in some parts of South Africa. Even at Durban, in
latitude 30° S., the heat is often severe, and the northern part of the Transvaal
and the British territories to the north of it lie within the Tropic of Capricorn. The
mean temperature in South Africa proper is 70° Fahrenheit in January and 80° in
July.
Over most of the country the climate is exceptionally dry. In the region of Cape
Colony there are well-defined summer and winter; but in the rest of South Africa
for about two-thirds of the year there is only a dry season, when the weather is
cooler, and a wet season of four or five months, when the sun is the highest and
the heat is most intense. The rainy season is not so continuous, nor is there so
great a precipitation, as in some other hot countries. In the parts where the
rainfall is heaviest, averaging over thirty inches in the year, the moisture soon
disappears by evaporation and absorption, and the surface remains parched till
the next wet season. As a consequence of this the air is generally dry, clear and
stimulating.
CATTLE ON THE VAAL RIVER.
[265]
It is interesting to note the effect upon climate of the physical structure described
above. The prevailing and rain-bringing winds are from the east and the
southeast. They bring sufficient moisture to the low plain along the sea coast,
and passing inland the rain-bearing clouds water the foothills of the Quathlamba
Mountains and precipitate snow on the loftier peaks beyond them. A portion of
the moisture is carried still farther to the west and falls in showers on the eastern
part of the plateau—the Transvaal, the Orange Free State, the eastern border of
Bechuanaland and the region northward toward the Zambesi. Sections farther to
the north and west receive but little of the annual rainfall, ranging from five to ten
inches in the year. That little is soon dissipated, the surface becomes dry and
hard, and such vegetation as springs up under the brief showers soon dies.
Much of this region is a desert, and so must remain until more and more
continuous moisture is supplied, either by artificial irrigation or by some
favorable change in natural conditions.
From these permanent physical features—the lowlands along the coast, the
elevated plateau in the interior, the mountain range running between them, a
burning sun and a dry atmosphere[266]—have developed many of the other
natural phenomena of South Africa.
The rivers of that country—laid down in great numbers on the maps—are not
rivers during much of the year. In the dry season they are either without water
altogether or consist of a succession of little pools scarcely sufficient to supply
the cattle on their banks with drink. And when they are rivers they are, most of
the time, such as can neither be forded nor navigated; the violent rains—
continuing for hours and sometimes for many days—have converted them into
roaring torrents.
Now, if that country could have been entered by waterways, as were North and
South America, it would not have remained an unknown land so long. But there
was no other means of penetrating it than the lumbering ox-wagon, making at
best a dozen miles a day, with frequent long halts in the neighborhood of good
grass in order to rest and recuperate the cattle. It is this lack of navigable rivers
that now compels the people to depend exclusively on railways for internal
transportation and travel. With the exception of tidal streams there is no internal
water communication of any value.
Another peculiarity of the east coast rivers [267]arises out of the nearness of the
Quathlamba Mountains to the sea. Such rivers as take their rise in the
mountains have very short courses, and the few that come from beyond, finding
channels through the mountain passes, are so obstructed by rapids and
cataracts at the point of descent from the higher levels that no boat can ascend
them.
South Africa presents to the foreigner from cooler climates no serious danger as
to health. The sun-heat would be trying were it not for the dryness of the
atmosphere and the invariable coolness of the nights, which have the effect of a
refreshing tonic. With due care in providing sufficient wraps for the occasional
cold day in the dry season, and the means of comfortable sleep during the cool
nights, there is nothing to fear.
The much-dreaded malarial fever has its habitat in the lowlands of both the east
and the west coast. Persons who are not immune to it can choose their place of
residence on the higher lands, or take refuge in quinine.
The dryness and purity of the air in many parts of South Africa—notably Ceres,
Kimberley, Beauport West and other places in the interior plateau—make it
peculiarly suitable for [268]persons suffering from any form of chest disease—
always excepting tuberculosis, for which the sure remedy has not yet been

Full Chapter Resonant Tunneling Quantum Waveguides of Variable Cross Section Asymptotics Numerics and Applications 2Nd Edition Lev Baskin PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Full Chapter Resonant Tunneling Quantum Waveguides of Variable Cross Section Asymptotics Numerics and Applications 2Nd Edition Lev Baskin PDF

Uploaded by

Copyright:

Available Formats

Resonant Tunneling Quantum

Waveguides of Variable Cross Section

A Cross Section of Psychological Research: Journal

Elasticity: theory, applications, and numerics Fourth

Finite Mathematics as the Foundation of Classical

Elasticity: Theory, Applications, and Numerics 4th

Geoengineering, the Anthropocene and the End of Nature

How to Love Yourself The Art of Loving Myself 2nd

Health Promotion 2nd Edition Cross

Sustainable Tunneling and Underground Use Mona Badr

More information about this series at http://www.springer.com/series/8548

Boris Plamenevskii Oleg Sarafanov

ISSN 1877-7341 ISSN 1877-735X (electronic)

St. Petersburg, Russia Lev Baskin

3.3.2 The Connection Between SðlÞ and SðlÞ

6.3.3 Transmission Coefﬁcient . . . . . . . . . . . . . . . . . . . . . . . 136

10.3 Tunneling in a Waveguide with One Narrow . . . . . . . . . . . . . . 207

13 Asymptotic Analysis of Multichannel Resonant Tunneling . . . . . . . 295

Bibliographical Sketch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325

As a preliminary, we consider a brief example of resonant tunneling. A resonance

© Springer Nature Switzerland AG 2021 1

There is a variety of electronics devices (transistors, key devices, energy

Fig. 1.1 The waveguide

In particular, we obtain asymptotic formulas for the aforementioned characteristics

The basic characteristics of electron resonant tunneling can be expressed in terms of a

r+ = {(y r , t r ) : y r ∈ r , t r > 0},

Fig. 1.2 The waveguide

where (y r , t r ) are local coordinates in r+ and r is a bounded domain in Rn

− (x) − μ(x) = 0, x ∈ G, (1.0.2)

− ( y,t + μ)u(y, t) = 0, (y, t) ∈ × R = , (1.0.3)

where is a domain in Rn and

Straightforward calculation shows that the nonzero functions

× R  (y, t) → exp (±i(μ − τ )1/2 t)ϕ(y)

satisfy (1.0.3) if and only if

that is, ϕ has to be an eigenfunction of problem (1.0.4), whereas τ is the corre-

for |x| → ∞, sufficiently small positive δ, and l = 1, . . . , M. The matrix

S(μ) = Sl j (μ) M

is called the scattering matrix.

threshold τ , there exist both one-sided limits of S(μ) as μ → τ ± 0 and, moreover,

Method for Approximate Computation of Scattering

for large R. Then ∂G R \ ∂G =  R = ∪r r,R , where r,R = {(y r , t r ) ∈ r : t r =

where ζ ∈ R \ {0} is an arbitrary fixed number, ν is an outward normal, and

for a j = Sl j . Thus, l satisfies the last equation in (1.0.9) up to an exponentially small

|a j (R, μ) − Sl j (μ)| ≤ c()e−R , j = 1, . . . , M, (1.0.11)

hold with some positive constants  and c() independent of R and μ.

Asymptotic and Numerical Studies of Resonant Tunneling in

Chapter 5 begins an asymptotic and numerical study of resonant tunneling. Electrons

Fig. 1.3 Domain

as an approximation in a neighborhood of the narrows. The terms are obtained by

G = R × D = {(x, y) ∈ R2 : x ∈ R = (−∞, +∞); y ∈ D = (−l/2, l/2)}.

−u(x, y) − k 2 u(x, y) = 0, (x, y) ∈ G(ε),

Moreover, u is subject to certain radiation conditions at infinity (that correspond, for

Fig. 1.4 Waveguide G(ε)

Fig. 1.5 The “limit

− v(x, y) − k 2 v(x, y) = f, (x, y) ∈ G j ,

Asymptotics of Resonant Tunneling in 2D-Waveguide with

In this chapter, we consider a two-dimensional waveguide that coincides with a strip

The coefficients f and g here do not depend neither on n, nor on ε.

Fig. 1.6 Waveguide G(ε)

where P is a constant coefficient. The height of each of the resonant peaks is 1 +

Asymptotics of the Resonant Tunneling of High-Energy

We consider a waveguide in R2 having two identical narrows of small diameter ε as

where k 2 is the spectral parameter, ke2 is an eigenvalue of the resonator,

− (x) − μ(x) = 0, x ∈ G, (1.0.2)

− ( y,t + μ)u(y, t) = 0, (y, t) ∈ × R = , (1.0.3)

× R (y, t) → exp (±i(μ − τ )1/2 t)ϕ(y)

S(μ) = Sl j (μ) M

for large R. Then ∂G R \ ∂G = R = ∪r r,R , where r,R = {(y r , t r ) ∈ r : t r =

|a j (R, μ) − Sl j (μ)| ≤ c()e−R , j = 1, . . . , M, (1.0.11)

hold with some positive constants and c() independent of R and μ.

−u(x, y) − k 2 u(x, y) = 0, (x, y) ∈ G(ε),

− v(x, y) − k 2 v(x, y) = f, (x, y) ∈ G j ,

−u − k 2 u = 0 in G(ε1 , ε2 ), u = 0 on ∂G(ε1 , ε2 ),

−v(x) − k 2 v(x) = f, x ∈ G 0 ; v(x) = 0, x ∈ ∂G 0 .