PDF Micromechanics of Composites Multipole Expansion Approach 2Nd Edition Volodymyr Kushch Ebook Full Chapter

Micromechanics of Composites
Multipole Expansion Approach 2nd

Edition Volodymyr Kushch
Visit to download the full and correct content document:
https://textbookfull.com/product/micromechanics-of-composites-multipole-expansion-
approach-2nd-edition-volodymyr-kushch/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...
Proceedings of the 2nd International Conference on

Building Innovations ICBI 2019 Volodymyr Onyshchenko
https://textbookfull.com/product/proceedings-of-the-2nd-
international-conference-on-building-innovations-
icbi-2019-volodymyr-onyshchenko/
Undergraduate Mathematics Competitions 1995 2016 Taras

Shevchenko National University of Kyiv 2nd Edition
Volodymyr Brayman
https://textbookfull.com/product/undergraduate-mathematics-
competitions-1995-2016-taras-shevchenko-national-university-of-
kyiv-2nd-edition-volodymyr-brayman/
Handbook of Wood Chemistry and Wood Composites 2nd

Edition Roger M. Rowell (Editor)
https://textbookfull.com/product/handbook-of-wood-chemistry-and-
wood-composites-2nd-edition-roger-m-rowell-editor/
Science and Engineering of Short Fibre Reinforced

Polymer Composites 2nd Edition Shao-Yun Fu
https://textbookfull.com/product/science-and-engineering-of-
short-fibre-reinforced-polymer-composites-2nd-edition-shao-yun-
fu/
Electromagnetic Composites Handbook Models Measurement
and Characterization 2nd Edition Rick Moore
https://textbookfull.com/product/electromagnetic-composites-
handbook-models-measurement-and-characterization-2nd-edition-
rick-moore/
Polymer Composites in the Aerospace Industry 2nd

Edition P. E. Irving (Editor)
https://textbookfull.com/product/polymer-composites-in-the-
aerospace-industry-2nd-edition-p-e-irving-editor/
Phenomenological Creep Models of Composites and

Nanomaterials: Phenomenological Creep Models of
Composites and Nanomaterials Leo Razdolsky
https://textbookfull.com/product/phenomenological-creep-models-
of-composites-and-nanomaterials-phenomenological-creep-models-of-
composites-and-nanomaterials-leo-razdolsky/
Modelling of Damage Processes in Biocomposites Fibre

Reinforced Composites and Hybrid Composites 1st Edition
Mohammad Jawaid
https://textbookfull.com/product/modelling-of-damage-processes-
in-biocomposites-fibre-reinforced-composites-and-hybrid-
composites-1st-edition-mohammad-jawaid/
Nanomechanics and Micromechanics-Generalized Models and

Nonclassical Engineering Approaches 1st Edition Satya
Bir Singh (Editor)
https://textbookfull.com/product/nanomechanics-and-
micromechanics-generalized-models-and-nonclassical-engineering-
approaches-1st-edition-satya-bir-singh-editor/
Butterworth-Heinemann is an imprint of Elsevier
The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom
50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States
Copyright © 2020 Elsevier Inc. All rights reserved.
No part of this publication may be reproduced or transmitted in any form or by any means,
electronic or mechanical, including photocopying, recording, or any information storage and
retrieval system, without permission in writing from the publisher. Details on how to seek
permission, further information about the Publisher’s permissions policies and our arrangements
with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency,
can be found at our website: www.elsevier.com/permissions.
This book and the individual contributions contained in it are protected under copyright by the
Publisher (other than as may be noted herein).
Notices
Knowledge and best practice in this field are constantly changing. As new research and experience
broaden our understanding, changes in research methods, professional practices, or medical
treatment may become necessary.
Practitioners and researchers must always rely on their own experience and knowledge in
evaluating and using any information, methods, compounds, or experiments described herein. In
using such information or methods they should be mindful of their own safety and the safety of
others, including parties for whom they have a professional responsibility.
To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors,
assume any liability for any injury and/or damage to persons or property as a matter of products
liability, negligence or otherwise, or from any use or operation of any methods, products,
instructions, or ideas contained in the material herein.
Library of Congress Cataloging-in-Publication Data

A catalog record for this book is available from the Library of Congress
British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library
ISBN: 978-0-12-823253-8
For information on all Butterworth-Heinemann publications

visit our website at https://www.elsevier.com/books-and-journals
Publisher: Matthew Deans

Acquisitions Editor: Dennis McGonagle
Editorial Project Manager: Peter Adamson
Production Project Manager: Kamesh Ramajogi
Designer: Victoria Pearson
Typeset by VTeX
Contents
Preface to the first edition xv

Preface to the second edition xix
1. Multipole expansion approach 1

1.1. Introduction 1
1.2. Structure models 4
1.2.1. Single inhomogeneity 4
1.2.2. Finite arrays of inhomogeneities 4
1.2.3. Composite band and layer 5
1.2.4. Representative unit cell 6
1.3. Bulk and interface field models 8
1.3.1. Conductivity 8
1.3.2. Elasticity 10
1.4. Method of solution 12
1.4.1. Multipole expansion: why and how? 12
1.4.2. Local expansion 14
1.4.3. Superposition principle 15
1.4.4. Summary of the method 16
1.5. Induced dipole moment of inhomogeneity 17
1.5.1. Definition 17
1.5.2. Conservation law 18
1.5.3. Relationship to the property contribution tensor 19
1.6. Macroscopic field parameters 20
1.6.1. Conductivity 20
1.6.2. Elasticity 25
1.7. Homogenization problem 27
1.7.1. Maxwell scheme 28
1.7.2. Rayleigh scheme 30
Part 1. Particulate composites
2. Potential fields of interacting spherical inhomogeneities 35

2.1. Single inhomogeneity 35
2.1.1. Series expansion 35
2.1.2. Resolving equations 36
2.2. Particle coating vs imperfect interface 37
2.3. Finite cluster model 39
2.3.1. FCM boundary-value problem 39
v
vi Contents
2.3.2. Convergence proof 42

2.3.3. Modified Maxwell scheme for effective conductivity 45
2.4. Composite sphere 49
2.4.1. Outer boundary condition 50
2.4.2. Interface conditions 50
2.4.3. RSV and effective conductivity of composite 51
2.5. Half-space FCM 53
2.5.1. Formal solution 53
2.5.2. Half-space boundary condition 54
2.5.3. Interface conditions 54
3. Periodic multipoles and RUC model of a composite 57

3.1. Composite layer 57
3.1.1. 2P fundamental solution of Laplace equation 57
3.1.2. 2P solid harmonics 61
3.1.3. Heat flux through the composite layer 62
3.2. Periodic composite as a sandwich of composite layers 64
3.3. Representative unit cell model 66
3.4. 3P scalar solid harmonics 68
3.4.1. Direct summation 68
3.4.2. Hasimoto’s approach 69
3.4.3. 2P harmonics based approach 71
3.5. Local temperature field 72
3.6. Effective conductivity 73
3.6.1. Rayleigh homogenization scheme 73
3.6.2. Numerical results 74
4. Elastic solid with spherical inhomogeneities 77

4.1. Single inhomogeneity in an unbounded solid 78
4.1.1. Multipole series expansion 78
4.1.2. Induced elastic dipole moment 79
4.1.3. Far field expansion 80
4.1.4. Resolving set of linear equations 81
4.2. Coated spherical inhomogeneity 83
4.3. Application to nanocomposite: Gurtin & Murdoch theory 84
4.3.1. Imperfect interface conditions 85
4.3.3. Numerical examples 87
4.4. Isotropic solid with anisotropic inhomogeneity 89
4.4.2. Resolving set of equations 91
4.5. Finite cluster of inhomogeneities 93
4.5.1. Direct (superposition) sum 93
Contents vii

4.5.3. Infinite system of linear equations 95
4.6. Effective stiffness of composite 99
4.6.1. Modified Maxwell scheme 99
4.6.2. Cubic symmetry 100
4.7. Elastic composite sphere 103
4.7.1. Elastic fields 103
4.7.2. Effective elastic moduli 105
5. Elasticity of composite half-space, layer, and bulk 109

5.1. Finite cluster of spherical inhomogeneities in a half-space 109
5.1.1. Problem statement 109
5.1.2. Homogeneous half-space 110
5.1.3. Heterogeneous half-space 111
5.2. Doubly periodic structures 114
5.2.1. 2P solutions of Lamé equation 115
5.2.2. Composite layer 117
5.2.3. Periodic composite as a sandwich of composite layers 120
5.3. Triply periodic solutions of Lamé equation 120
5.3.1. Scalar 3P biharmonics 120
5.3.2. Periodic solutions of Lamé equation 122
5.4. RUC model 123
5.4.2. Effective stiffness tensor 126
5.5. Numerical study 128
5.5.1. Local stress field 128
6. Conductivity of a solid with spheroidal inhomogeneities 137

6.1.2. Induced dipole moment 139
6.1.3. Resolving equations for perfect interface 140
6.1.4. Limiting cases: spherical, penny-shaped, and needle-like
inhomogeneities 140
6.1.5. LC imperfect interface 145
6.1.6. HC imperfect interface 147
6.2. Finite cluster model 148
6.3. Modified Maxwell scheme 150
6.3.1. Analytical results 150
6.4. Heat conduction in a periodic composite 157
6.4.1. Doubly periodic harmonics 157
viii Contents
6.4.2. Triply periodic harmonics 159

6.4.3. Temperature field in a periodic composite 160
6.4.4. Multiple inhomogeneity RUC model 165
6.5. Rayleigh homogenization scheme 166
6.5.1. Composite with perfect interface 166
6.5.2. Composite with imperfect interface 174
7. Elastic solid with spheroidal inhomogeneities 179

7.1. Single inhomogeneity with perfect interface 179
7.1.1. Displacement field 179
7.1.2. Elastic dipole moment 182
7.1.3. Stress intensity factors for a penny-shaped crack 183
7.2. Single inhomogeneity with imperfect interface 186
7.2.1. Spring type interface 187
7.2.2. Membrane type interface 191
7.3. Finite cluster of spheroidal inhomogeneities 195
7.4. Modified Maxwell scheme 203
7.4.1. Single inhomogeneity model 205
7.4.2. Multiple cavity model 211
7.5. Elastic half-space with spheroidal inhomogeneities 213
7.6. RUC model of an elastic spheroidal particle composite 215
7.6.1. Periodic solutions of Lamé equation 215
7.6.2. Displacement solution 216
7.6.3. Numerical study 217
8. Composites with transversely isotropic constituents 237

8.1. Transversely isotropic conductivity 237
8.1.1. Partial solutions 238
8.1.3. Temperature field 239
8.1.4. Effective conductivity tensor 242
8.2. Transversely isotropic elasticity 243
8.2.1. Partial vector solutions 244
8.2.3. Finite array of inhomogeneities 253
8.3. RUC model 255
Contents ix
8.4. Numerical examples 257

8.4.1. Stress concentration 257
8.4.2. Effective stiffness 262
9. Conductivity of an ellipsoidal particle composite 269

9.1. Composite with isotropic constituents 270
9.1.2. Single ellipsoidal inhomogeneity 270
9.1.3. Finite array of ellipsoids 274
9.1.4. Periodic array of ellipsoids 275
9.2. Composite with anisotropic constituents 278
9.2.2. Reduction to isotropic case 279
9.2.3. Single ellipsoidal inhomogeneity 282
9.2.4. Eshelby-type problem 286
9.2.5. Finite cluster model 289
9.2.6. RUC model 291
9.3. Effective conductivity of a composite 293
9.3.1. Maxwell homogenization scheme 294
10. Elasticity of an ellipsoidal particle composite 301

10.1. Single ellipsoidal inhomogeneity 303
10.1.2. Dipole moment 304
10.2. Uniform far filed 305
10.2.1. Ellipsoidal cavity, displacement boundary condition 306
10.2.2. Ellipsoidal cavity: traction boundary condition 309
10.2.3. Ellipsoidal inhomogeneity 310
10.3. Nonuniform far field 311
10.3.1. Displacement boundary condition 312
10.3.2. Traction boundary condition 313
10.3.3. Perfect interface 314
10.3.4. Imperfect interface 315
10.4. Finite cluster of ellipsoidal inhomogeneities 316
10.4.1. Displacement solution 316
10.4.2. Resolving linear system 318
10.5. Orthotropic elastic solid with an arbitrarily oriented inhomogeneity 322
10.5.1. Problem formulation 322
10.5.2. Reduction to isotropic case 323
10.5.3. Multipole expansion solution 325
x Contents

10.6.2. Multiple inhomogeneities 334
Part 2. Fibrous composites: 2D models
11. Circular fiber composite with perfect interfaces 343

11.1. In-plane conductivity 343
11.2. Finite cluster of circular inhomogeneities 345
11.2.1. General solution for a single inhomogeneity 345
11.2.2. Finite array of inhomogeneities in an unbounded plane 347
11.2.3. Half-plane with circular inhomogeneities 350
11.2.4. Modified Maxwell scheme 353
11.3. Infinite arrays of circular inhomogeneities 355
11.3.1. Periodic complex potentials 355
11.3.2. Composite band 357
11.3.3. Composite layer 358
11.4. Representative unit cell 362
11.4.2. 1P approach 364
11.4.3. 2P approach 365
11.4.4. Rayleigh scheme for effective conductivity 368
11.5. In-plane elasticity problem 368
11.5.1. Governing equations in terms of complex potentials 368
11.6. Circular inhomogeneities in a half-plane 374
11.6.2. Determination of p(β) and q(β) 376
11.6.3. Resolving linear system 378
11.7. RUC model of an elastic fibrous composite 381
11.7.2. Transverse effective stiffness 384
12. Fibrous composite with interface cracks 387

12.1. In-plane conductivity 387
12.1.1. Single inhomogeneity problem 387
12.1.2. Temperature field 388
12.1.3. Heat flux intensity factor 391
12.2. Finite cluster of partially debonded fibers 392
12.3. RUC model of a fibrous composite with interface damage 396
12.3.2. Evaluation of lattice sums 397
12.3.3. Effective conductivity tensor 398
Contents xi
12.4. In-plane elasticity 401

12.4.1. General form of displacement solution 401
12.5. Single partially debonded fiber 404
12.5.2. General form of complex potentials 405
12.5.3. Analytical solution 406
12.5.4. Stress intensity factor 409
12.6. Finite cluster of debonded fibers 410
12.6.1. Problem statement and iterative solver 410
12.6.2. Evaluation of integrals 411
12.7. RUC model of a partially debonded fibrous composite 414
13. Solids with elliptic inhomogeneities 423

13.1. Single elliptic inhomogeneity 424
13.1.1. Problem statement and formal solution 424
13.1.2. Interface displacement and traction 426
13.1.3. Resolving system 428
13.2. Finite array of inhomogeneities 432
13.3. Heterogeneous half-plane 436
13.4. Effective stiffness tensor 441
13.4.1. Maxwell homogenization scheme 441
14. Cracked solids 451

14.1. RUC model of a cracked solid 451
14.1.1. Geometry 451
14.1.2. Boundary value problem 452
14.1.3. Out-of-plane shear 454
14.1.4. Plane strain 455
14.3. Stress intensity factors 461
14.4. Numerical examples 462
14.4.1. Solid with nonrandomly oriented cracks 462
14.4.3. SIF statistics 468
15. Elliptic fiber composite with an imperfect interface 473

15.1. Single elliptic inhomogeneity 473
xii Contents
15.1.3. Resolving set of equations 476

15.1.4. Numerical testing 478
15.2. Half-plane with an elliptic nanoinhomogeneity 482
15.3. Multiple elliptic fiber models 483
15.3.2. Representative unit cell 485
15.3.3. Effective longitudinal shear stiffness 487
15.4.1. Effective shear modulus 488
15.4.2. Effective conductivity 493
16. Fibrous composite with anisotropic constituents 499

16.1. Out-of-plane shear 499
16.1.1. Governing equations 499
16.3. Finite array of inhomogeneities 504
16.4. Representative unit cell model 507
16.4.1. Periodic complex potentials 507
16.5. Plane strain 510
16.5.1. General solution 510
16.5.3. Multiple inhomogeneities 513
A. Spherical harmonics and related theory 519

A.1. Scalar spherical harmonics 519
Laplace equation in spherical coordinates 519
Selected properties of solid spherical harmonics 521
Spherical harmonics vs multipole potentials 523
Fourier integral representation 524
A.2. Reexpansion formulas for Yts and yst 525
Equally oriented coordinate systems 525
Arbitrarily oriented coordinate systems 526
A.3. Scalar spherical biharmonics 529
A.4. Vector spherical surface harmonics 531
A.5. Partial solutions of Lamé equation 532
Definition 532
Explicit expressions 533
Normal traction 535
Net force and torque 536
A.6. Partial solutions for a half-space 536
Cartesian vector surface harmonics 536
Contents xiii
Vector solutions of Lamé equation for a half-space 537

Integral transforms and series expansions 538
(i ) (i )
A.7. Reexpansion formulas for Uts and uts 539
Translation 539
Rotation 540
B. Spheroidal harmonics and related theory 543

B.1. Scalar solid harmonics 543
Laplace equation in spheroidal coordinates 543
Spheroidal solid harmonics 545
Relationships between the spherical and spheroidal harmonics 546
Alternative set of spheroidal harmonics 548
Extended set of spheroidal harmonics 549
B.2. Reexpansion formulas 550
Formal series expansion 551
Translation: integral form of the expansion coefficients 552
Translation: rational form of the expansion coefficients 553
Rotation 555
B.3. Double Fourier integral transform of spheroidal harmonics 556
B.4. Vector solutions of Lamé equation 558
Definition 558
Selected properties 559
Integral transforms 564
(i ) (i )
B.5. Reexpansion formulas for Vts and vts 565
General form 565
Rotation 567
C. Ellipsoidal harmonics and related theory 569

C.1. Ellipsoidal harmonics 569
Solid harmonics 569
Surface harmonics 570
C.2. Differentiation and integration 571
Derivatives of solid harmonics 571
Numerical integration 572
C.3. Reexpansion formulas 573
D. Selected properties of functions Rλ and Xλ 577

D.1. Function R(ζ ) 577
D.2. Functions Rλ (ζ ) and Xλ (ζ ) 579
E. Elliptic harmonics and related theory 583

E.1. Elliptic harmonics 583
E.2. Reexpansion formulas 584
xiv Contents
E.3. Integral transforms 588

E.4. Periodic complex potentials 589
E.5. Evaluation of βk coefficients 591
Bibliography 593
Index 609
Preface to the first edition
The subject of this book is micromechanics, that is, the analysis of hetero-
geneous materials on the level of individual constituents. The number of
man-made composites is constantly increasing, in parallel with the rising
need to study the “structure–properties” relationships as this knowledge
enables purposeful tailoring of composite materials with superior proper-
ties by a rational choice of components and composition. This explains the
importance of micromechanics as a science and motivates its rapid devel-
opment in recent decades.
In the past, the main effort in micromechanics was focused on the
macroscopic properties of heterogeneous solids and most of the work in
the area has been done with the aid of the single inclusion (Eshelby) model
which is the theoretical framework of several applied theories. They provide
practically useful bounds and approximations for the effective constants and
so can be grouped under the title of engineering, or applied micromechan-
ics. Most of the up to now published books on micromechanics fall into
this category. Certainly, applied micromechanics is easy and convenient for
use, but, as always, convenience comes at a price. The latter involves low
or uncertain accuracy, inability to account for the microstructure of a com-
posite and, as a consequence, inapplicability to study of the phenomena
(e.g., damage) caused by the local fields.
Now, it is well recognized that a reliable prediction of composite’s
behavior must combine a realistic model of the microstructure with an ad-
equate analysis of the relevant model boundary problem. The need for an
in-depth study has led to the development of computational micromechanics.
Recent dramatic increase in computational power and available commer-
cial FEA software made direct numerical approach accessible (not affordable
for individual researcher, however) and enabled consideration of involved
heterogeneous structures. The drawback of this approach is high computa-
tional effort, especially for 3D models. Another and, probably, even more
substantial problem is extracting meaningful data from a bunch of numbers
generated by FEA code.
A promising alternative to computational micromechanics is the multi-
pole expansion method which also provides an efficient analysis of complex
heterogeneous structures. Being mostly analytical in nature, this method
constitutes a theoretical basis of high-performance computational algo-
rithms and found numerous applications in astronomy, physics, chemistry,
xv
xvi Preface to the first edition
engineering, statistics, etc. Introduced by J.C. Maxwell in 1873 and then

further developed by Lord Rayleigh in 1892, this is historically the first
method of micromechanics. Substantial progress made since that time (and,
especially, in recent years) in the development of the multipole expansion
method has been reported in numerous papers. However, in the author’s
opinion, the real value of this method for micromechanics is still underesti-
mated and its potential in the area is not fully discovered so far. This book is
the first monograph giving a systematic account of the method, with appli-
cation to the actual problems of micromechanics. The multipole expansion
method uses a classical approach and toolkit of mathematical physics, which
is a compelling reason to consider it as theoretical micromechanics.
The book does not pretend to cover all the aspects and methods of
micromechanics. Its specific aim is to describe the theory and technique of
the method in detail, with application to the selected actual problems of
micromechanics. This is primarily the multipole expansion approach that
sets this monograph apart and is an alternative to what readers would find
in other books. In the author’s opinion, the following features of this work
can be of particular interest for the reader:
• The multipole expansion theory and technique have been de-
scribed and further developed. In this respect, the book is of interest for a
wide readership, including specialists in applied mathematics, mathematical
physics, engineering, and related areas dealing with heterogeneous media.
• A detailed analysis of a variety of micromechanical multiinclusion
models has been performed. The contemporary topics include the com-
posites with imperfect and partially debonded interface, nanostructured
materials, cracked solids, statistics of the local fields, brittle strength, etc.
The obtained complete analytical solutions provide a clear insight into the
physical nature of the problems.
• The book contains a lot of tabulated data and plots for the various
problems. The results of the multipole analysis are commonly considered
as the most reliable and serve as a benchmark for testing applicability of
approximate models and accuracy of numerical solutions.
• The considered mechanical models are readily generalized in many
ways to take the specific features of real-world heterogeneous materials into
account. The Fortran source codes given in the Appendix can be used by
the readers as a starting point in developing their own codes.
• An important feature of the developed approach is high numeri-
cal efficiency. In contrast to computational micromechanics, the multipole
expansion does not require powerful computers and expensive software to
Preface to the first edition xvii
be used and appears, probably, as the most efficient (especially in the fast
multipole version) method of micromechanics.
The book summarizes the work done by the author with colleagues for
more than 20 years in development and application of the multipole ex-
pansion method, and is expected to be of particular interest to researchers
and professionals in applied mathematics, physics, mechanics, materials sci-
ence, engineering, and related areas dealing with the heterogeneous solids.
I am very grateful to all my colleagues and friends, in Ukraine and abroad,
who have contributed to this work, as well as inspired and supported me in
many ways while conducting the research and writing this book.
Kyiv, Ukraine Volodymyr Kushch

October, 2012
Preface to the second edition
During the time since the first edition of the book, substantial progress
has been achieved in the theory and application of the multipole expan-
sion method. The newly obtained results greatly increase the potential of
the method, enable a deeper insight into the microstructure-level phenom-
ena, and offer better understanding the “structure–properties” relationships.
They are incorporated in the new edition for a more complete and system-
atic account of the material.
It is noteworthy that the focus of the second edition is somewhat dif-
ferent from the that of the first. The primary objective of the original book
was to acquaint the reader with the novel approach in micromechanics and
provide its comprehensive theoretical justification. This edition is more fo-
cused on the application of the multipole expansion method to the actual
micromechanical problems. The structure of the book has been adjusted
accordingly by rewriting and rearranging the text, adding new chapters,
and removing less relevant parts. All the background mathematical theory
is reduced to the necessary minimum and moved to the Appendices.
The book comprises several new topics. They cover the recent findings
in the theory and technique of the method, as well as many new theoretical
solutions and a large amount of accurate numerical data. Special emphasis
is made on the heterogeneous solids with imperfect interfaces, including
the advanced models of nanoporous and nanocomposite materials. The
unified, dipole moment-based approach to the homogenization problem
makes its solution an easy and straightforward task. The modified Maxwell
and Rayleigh homogenization schemes formulated in terms of the dipole
moments and taking the microstructure and interaction between the inho-
mogeneities into account are applied systematically throughout the book.
The parts not retained in this edition include the background math (i.e.,
derivations, convergence proofs, etc.), Fortran codes, and other reference
information. The interested reader is encouraged to read the original book.
Kyiv, Ukraine Volodymyr Kushch

October, 2019
xix
CHAPTER ONE
Multipole expansion approach
1.1. Introduction
A vast majority of the natural and man-made materials around us
are heterogeneous, or composite, solids. Typical examples of the natu-
ral composites are timber, stone, human bone, and so on. The com-
monly known engineering composites are concrete, fiberglass, ceramics,
microchips, to mention a few. Even the materials that seem homogeneous
at first glance (e.g., the metals), upon closer examination exhibit polycrys-
talline microstructure. In general terms, composites are the materials that
are inhomogeneous on some (micro) length scale l (above the molecular
scale, however) but which are essentially homogeneous at a larger (macro)
length scale L l (e.g., [1]).
Structural sensitivity is a distinctive inherent feature of heterogeneous
solids. In other words, not only the volume content and properties of the
constituents of composite are important, but also the manner in which these
constituents are assembled into a composite solid. In order to understand
– and, hence, be able to predict – the behavior of heterogeneous materi-
als in various conditions, we need to know how the microstructure of a
material affects its properties at the micro and macro level. What is even
more important, this knowledge enables purposeful tailoring of new, ad-
vanced composite materials by employing a rational choice of components
and composition. The widely known example is the reinforced concrete,
where the successful combination of components, structure, and method
of manufacturing revolutionized the building industry. We are surrounded
by a variety of the man-made composite materials and their number is
permanently growing, in parallel with the growing need to study the
“structure–properties” relationships in order to improve the performance
of these materials.
The rapid development of the industry of composite materials in recent
decades has stimulated the development of a new branch of solid mechan-
ics known now as the micromechanics, or mechanics of materials. To date,
several books have been published on the micromechanics of heteroge-
neous solids; see [2–8,1,9–11], among others. These books also contain
Micromechanics of Composites Copyright © 2020 Elsevier Inc. 1

https://doi.org/10.1016/B978-0-12-823253-8.00007-2 All rights reserved.
2 Micromechanics of Composites
comprehensive reviews of other publications on the subject. Up to now,

the main effort was focused on the evaluation of the macroscopic proper-
ties of heterogeneous solids, and most theoretical work in the area was done
with the aid of the single inhomogeneity models. Among them, the most
successful is the Eshelby [12] model which gave rise to a number of ap-
proximate micromechanical theories (see [13–15,7], among others). These
theories provide O(c ) estimates of effective properties (c being the volume
content of inhomogeneities) owing to the fact that they were developed
from the single inhomogeneity model. All of them are equivalent, in terms
of accuracy, to the famous Maxwell theory [16], and their applicability is
justified for the composites with a low c. After more than 50 years since
its publication, the Eshelby [12] model still is the theoretical basis of many
studies on composites.
In order to get the next c 2 term in the serial expansion of effective
property, the pair interaction effect by means of the two-inhomogeneity
model (e.g., [17–20]) must be taken into account. Further increase in
prediction accuracy requires a geometry model with more interacting in-
homogeneities to be considered. Lord Rayleigh [21] was first to consider
conductivity of a solid containing an infinite, periodic array of inhomo-
geneities. His solution, accurate up to c 8/3 , was further refined several times
before the complete, multipole-type analytical solutions of the conductiv-
ity problem for three cubic arrays of identical spheres have been obtained
(see [22–26], among others). Later on, this approach has been extended
to the elasticity problem [27–30]. In a series of more recent papers (see
[31–35], among others), the problem of a random dispersion was treated
as a triple-periodic problem with random arrangement of the particles in
a cubic unit cell. This geometry, known also as the representative unit cell
(RUC) (e.g., [36,37]), will be discussed below in detail. Here, we only
note two distinctive features of the theories employing the multiple inho-
mogeneity model. In contrast to the single inhomogeneity-based theories,
they take into account the microstructure of a composite and interaction
between the inhomogeneities, which greatly improves the reliability and
accuracy of prediction.
The above discussion relates the macroscopic, or effective, properties
of a composite known in micromechanics as the homogenization problem.
The effective properties provide important information regarding the ma-
terial’s behavior on a macro scale. However, this information is insufficient
for most structural and engineering applications of composites. The lo-
cal damage mode, its onset level (critical load), and accumulation rate as a
Multipole expansion approach 3
function of the load applied to the composite part is crucial information for
the designer. Prediction of composite’s strength is the second (in order, but
not in practical importance) problem of the micromechanics. Strength of
a composite is the highly structure-sensitive property governed mostly by
the peak local stress. Its determination is essentially the micro-level prob-
lem which cannot be solved in terms of the mean fields because the local
stress concentration (SCF) plays a dominant role in the damage origin and
development. In turn, SCF is greatly affected by interaction between the
inhomogeneities and may exceed several times the peak stress observed
on a single inhomogeneity. A reliable model of composite’s strength must
combine the multiple inhomogeneity model of microstructure with an ac-
curate study of the local stress field. The strength theory of composites
is still in its infancy and, in order to make progress in the area, we need
to develop adequate models of composites and efficient methods to study
them.
The pertinent question is: what is an adequate model?
Of course, it is impossible to embrace everything. First, it is very dif-
ficult to ascertain the detailed internal structure of a heterogeneous solid.
Second, even if we assume that such information is obtained in some way
(e.g., by X-ray tomography [38]), its theoretical analysis may require a pro-
hibitively large numerical effort. Fortunately, there is no need in capturing
all the details of the microstructure, since much of them have no appre-
ciable effect on the properties of composites. This allows for a reasonable
idealization of the microstructure, which in turn opens up the possibility of
its theoretical analysis. For example, an approximation of the inhomogene-
ity shape by a quadratic (cylinder, sphere, ellipsoid) surface is justified in
many cases. At the same time, it is important to model the essential features
of an actual microstructure, they are particle arrangement, size and shape
statistics, and so on. Reasonable compromise lies, as always, somewhere
between the “too simple model” and “too difficult solution”. A geo-
metric model applicable equally to all heterogeneous solids is unlikely to
exist. It is possible, however, to specify some typical model geometries,
appropriate for practical use. In this book, we consider the matrix-type
composites comprising the continuous matrix and nonintersecting inho-
mogeneities of various shapes. Below, the ordered in ascending complexity
set of geometry models of heterogeneous solids is considered, and their
application to the study of local fields and macroscopic properties is dis-
cussed.
1.2. Structure models

1.2.1 Single inhomogeneity
We mentioned already the Eshelby model [12] being a single ellipsoidal
inhomogeneity immersed in an unbounded matrix subject to the uniform
far field. A more general problem, namely, that of a single ellipsoidal inho-
mogeneity in the polynomial far field, has been considered in [39]. There
are several works where the spherical (e.g., [40–43]) and spheroidal (e.g.,
[42–44]) inhomogeneities in a nonuniform far field were studied. These
results, on the one hand, can be viewed as the generalized Eshelby model,
expanded for the case of nonuniform far load but still readily implanted in
that or another self-consistent scheme. On the other hand, they provide
a necessary background for solving the multiple inhomogeneity problems
of linear micromechanics by the superposition method. Derivation of the
general solution to the basic, single inhomogeneity problem for the arbitrary
realizable far field constitutes the first step in tailoring the solution to the
advanced, many-particle boundary-value problem.
1.2.2 Finite arrays of inhomogeneities

The next three models shown in Fig. 1.11 involve a finite array of inhomo-
geneities. The first of them (Fig. 1.1A) is the finite cluster model (FCM)
being an unbounded solid with several inhomogeneities. This is essentially
the model used by Maxwell [16] who derived his famous formula by equat-
ing the potential of an array of inhomogeneities to that of the equivalent
inhomogeneity with unknown effective conductivity while neglecting the
interaction between the inhomogeneities. To obtain an accurate solution to
the FCM boundary-value problem, the above-mentioned general solution
of the single-inhomogeneity problem must be combined with the super-
position principle and the relevant reexpansion formulas. This complete
solution can be viewed as the generalized Maxwell model [45,46], where the
particle-to-particle interactions are taken into account. At the same time,
these results constitute the intermediate, second step in development of the
full-featured model of a composite.
The second model (Fig. 1.1B), where the inhomogeneities are placed
nearby the flat boundary of a half-space, is referred as the half-space FCM.
1 Figs. 1.1 to 1.4 are rather illustrative: for clarity, the two-dimensional (2D) domains
with circular/elliptic inhomogeneities are plotted. However, all said holds true for three-
dimensional (3D) domains and inhomogeneities of arbitrary shape.
Figure 1.1 Finite arrays of inhomogeneities: (A) finite cluster model (FCM); (B) FCM in
half-space; (C) representative spherical volume (RSV).
Its particular purpose is to evaluate the edge effects in the composite parts.
The fact is that the stress field in a vicinity of free/loaded surface differs
from that in the composite bulk. In view of that surface is the place where
the damage typically starts to develop, the considered model is of practical
interest. The third model is a composite sphere (Fig. 1.1C) sometimes re-
ferred to as the representative spherical volume (RSV). This model provides
an alternative way to evaluate the effective properties of a composite.
1.2.3 Composite band and layer

The above models deal with the finite arrays of inhomogeneities and can-
not be extended directly to the case of an infinitely large array. At the
same time, the models involving an infinite number of inhomogeneities are
of particular interest in micromechanics because they enable taking con-
tribution of a whole set of interacting inhomogeneities into account and
make the so-called closure problem a trivial task. The simplest example is
a cubic/square lattice model suggested by Lord Rayleigh [21]. This model
is perfect for the composites having a periodic structure. However, most
real-world heterogeneous solids possess random microstructure. Notewor-
thy, the epithet “random” itself does not specify the microgeometry of
a heterogeneous solid because two random media with the same volume
content of disperse phase but different microstructure statistics may pos-
sess quite dissimilar properties. A good model must take the statistics of
a microstructure into account. At the same time, such a model must be
“realistic” in the above-discussed sense, i.e., it should capture the essential
features of the specific random structure and should not require enormous
numerical effort for solving the model-based micromechanical problem.
A reasonable compromise consists in using the so-called “quasi-

random”, or “generalized periodic” structure model, being a finite array
of randomly placed inhomogeneities replicated periodically in order to fill
an infinite domain. This model is advantageous in that it allows us to sim-
ulate a real microstructure while the model boundary-value problem stays
deterministic and hence can be solved accurately. Two structure models
Figure 1.2 Composite band.
Figure 1.3 Composite layer.
of this kind, namely, the composite band and composite layer are shown
in Figs. 1.2 and 1.3, respectively. The latter is particularly appropriate to
study the composite films and coatings where a free/loaded/joined to sub-
strate boundary substantially affects the local fields and macroresponse of
the composite layer.
1.2.4 Representative unit cell

The representative unit cell (RUC) model is essentially the above consid-
ered FCM, replicated periodically in three orthogonal directions without
overlapping of any two inhomogeneities. In essence, we consider an un-
bounded solid containing N periodic, equally oriented simple cubic (SC)
arrays of inhomogeneities. For a given geometry, an arbitrarily placed cube

with side length equal to the period of the structure and oriented along
the principal periodicity axes can be taken as the elementary unit cell of
RUC; see Fig. 1.4A. Alternatively, one can take the unit cell in the form
Figure 1.4 RUC model of composite bulk: (A) flat boundary; (B) curved boundary pass-
ing through the matrix material.
of a cuboid with curvilinear boundary and parallel opposite faces. For con-
venience, we assume with no loss in generality that the cell boundary S0
entirely belongs to the matrix; see Fig. 1.4B. In fact, the model problem
is formulated and solved for the whole composite space rather than for
the unit cell being simply a means for introducing the model geometry
and averaging the local fields. Due to induced by geometry periodicity of
the fields, the RUC serves as a representative volume element (RVE) of
a composite. This model can be thought as the generalized Rayleigh model.
It is indeed appropriate for studying the local fields and effective proper-
ties of highly-filled and strongly heterogeneous composites where the fiber
arrangement and interactions between them affect the material behavior
quite substantially.
Several methods (see [31,10,47], among others) have been developed to
generate the random structure RUC. This model, as well as the real mi-
crostructure, can be characterized by several structural parameters. They are
packing density, coordination number, radial distribution function, nearest
neighbor distance, etc. Ideally, the model geometry has to be taken appro-
priately to approach (in the statistical sense) the microstructure of an actual
disordered composite – and RUC model is, probably, the most appropri-
ate for this purpose. For more discussions on the subject, see [10,48,11],
among others.
1.3. Bulk and interface field models

An adequate geometry model is an important, but only the first step
in the development of the micromechanical model of a heterogeneous solid.
The subsequent steps comprise the model boundary-value problem formu-
lation, analytical or numerical solution and post-processing, i.e., evaluation
of the local and macro fields and effective constants. In what follows, the
steady state conductivity (scalar potential field) and small strain elasticity
(displacement vector field) problems are considered. Both problems are
assumed linear to enable analytical approach to their analysis. The bulk
potential and displacement fields obey Laplace and Lamé equation, respec-
tively.
The interface contact conditions is yet another important part of the
model boundary-value problem. The real-world interfaces are always im-
perfect due to the atomic lattices mismatch, phonons scattering, poor
mechanical or chemical adherence, surface contamination, oxide and in-
terphase diffusion/reaction layers, debonding, etc. These phenomena may
affect the composite’s behavior quite significantly and so must be taken into
account in the predictive model. Below, the typical bulk and interface field
models are discussed.
1.3.1 Conductivity
In micromechanics, the word “conductivity” is often used as a generic
term for several, mathematically equivalent transfer phenomena [1]. These
are heat conduction, electric conductivity, diffusion flow, dielectric per-
mittivity, magnetic permeability, and so on. To be specific, we use the
terminology of the thermal conductivity problem. The governing equa-
tions for the bulk are
∇ · q = 0, q = − · ∇ T , (1.1)
where q is the heat flux vector, is the second rank thermal conductivity
tensor, T and ∇ T are the temperature and its gradient, respectively. In
the case of isotropic solid where = λI, T obeys the Laplace equation
2 T = 0. Eq. (1.1) applies equally to the matrix solid (T = T (0) , = 0 )
and inhomogeneities (T = T (1) , = 1 ).
At the matrix–inhomogeneity interface S, the imperfect thermal con-

tact between the matrix and inhomogeneities is taken in the form

∂ T (i) ∂ T (i)
1
i α T , (i)
, = 0, α = 1, 2. (1.2)
i=0
∂ xj ∂ xj ∂ xk S
In Eq. (1.2), iα is an arbitrary linear function of temperature, as well as

its first and second derivatives at the matrix (i = 0) and inhomogeneity
(i = 1) side of interface. It is noteworthy that all the known models of
thermal interface, including the perfect bonding, low-conductive [49] and
high-conductive [50–52] imperfect interface models, as well as their mod-
ifications [53–55], fall into this category. The following particular cases of
Eq. (1.2) are of special interest:
• The commonly used perfect interface conditions assume the continuity
of temperature and normal heat flux qn = q · n where n is the unit
normal vector to the matrix–inhomogeneity interface S. This is,

T S
= 0, qn S
= 0, (1.3)

where T S = T (0) − T (1) |S means a jump
of temperature

T across
the interface S. The normal flux jump is qn S = q(n0) − q(n1) |S .
• The low-conductive (LC), or Kapitza, interface model assumes that
the normal heat flux is continuous through the interface whereas the
temperature jump is proportional to the normal flux [49]:

hc T S
+ qn = 0, qn S
= 0, (1.4)

where the coefficient hc measured in Wt/ m2 K is known as the
surface conductivity.
• The opposite case is the high-conductive (HC) interface model that as-
sumes the continuity of temperature and jump of normal heat flux, pro-
portional to the surface Laplacian of temperature s T = divs (∇s T ) =
n · ∇ × (n × ∇ T ) [50–52]:

T S
= 0, qn S
− hs s T (1) = 0. (1.5)

Here, hs is the coefficient measured in Wt/K .
Note that both LC and HC models of imperfect interface can be de-
rived as the limiting cases of a thin interphase layer. For derivation and
discussion on the physical nature of Eq. (1.5), see [52]. Here, we mention
only that the dimensionality analysis [56] predicts dependence of the com-
posite behavior on the size of inhomogeneities in both LC and HC cases.
This size effect vanishes only for the extreme values hc = ∞ in Eq. (1.4) and
hs = 0 in Eq. (1.5) where they reduce to the conventional, perfect interface
bonding condition of Eq. (1.3).
1.3.2 Elasticity
In the linear elasticity, the governing equations for the bulk are
1
∇ · σ = 0, σ =C:ε , ε= ∇ u + (∇ u)T , (1.6)
2
where σ = σij ii ij is the stress tensor, ε = εij ii ij is the small elastic strain tensor,
and u is the displacement vector. Also, C is the fourth rank elastic stiffness
tensor. The elastic behavior of an isotropic solid is governed by a pair of
independent elastic constants, namely, the shear modulus μ and Poisson
ratio ν :
ν
σ = 2μ ε + tr (ε) I ,
1 − 2ν
where I is the second rank unit tensor. The displacement vector (u = u(0)
in the matrix, u = u(1) inside the inhomogeneities) of a composite with
isotropic constituents obeys Lamé equation
1−ν
2 ∇(∇ · u) − ∇ × ∇ × u = 0. (1.7)
1 − 2ν
At the matrix–inhomogeneity interface S, the imperfect elastic contact
between the matrix and inhomogeneities is assumed in the form

∂ u(i) ∂ u(i)
1
iα (i)
u , , = 0, α = 1, 2. (1.8)
i=0
∂ xj ∂ xj ∂ xk S
In Eq. (1.8), iα is a linear function of the displacements, as well as their first
and second derivatives at the matrix and inhomogeneity side of interface.
Again, all the known models of imperfect elastic interface, including the
sliding [57], spring- [58], and membrane-type [59] models, as well as their
various modifications [60–62], fall into this category. Below, the following
particular cases of Eq. (1.8) will be studied in detail:
• The commonly used perfect interface conditions assume the continuity
of the displacement vector and normal traction vector Tn = σ · n at the
matrix–inhomogeneity interface S. This is,

[[u]]S = 0, Tn S
= 0, (1.9)

where [[u]]S = u(0) − u(1) |S and Tn S
= Tn (u(0) ) − Tn (u(1) ) |S
mean a jump of displacement and normal traction, respectively, across
the interface S.
• The spring-type (weak) interface [58] assumes the continuity of the
normal traction vector and jump of the displacement vector:

Tn S
= 0, Tn (u(1) ) − D · [[u]]S = 0, (1.10)

where D is the contact stiffness matrix measured in N/m3 . In the
commonly accepted isotropic weak interface model, D = Dn nn +
Dt (I − nn). Also, Dn and Dt are the normal and tangential contact
stiffness, respectively. The dimensionality analysis predicts dependence
of the composite behavior on the size of inhomogeneities. This size
effect vanishes only for the extreme values Dij = ∞ and Dij = 0, corre-
sponding to the perfect contact and porous solid, respectively.
• The membrane-type (stiff) interface model we consider is based on
Gurtin–Murdoch theory of elastic material surface [59]:

[[u]]S = 0, Tn S
+ div σ = 0. (1.11)
Recently, this model has gained wide popularity due to its potential
applicability to nanoporous solids and nanocomposites [63]. First, the
so-called coherent interface condition in Eq. (1.11) imposes continuity
of the displacement vector across the interface. The term div σ in the
right-hand side of the second condition of Eq. (1.11) denotes surface
divergence of the surface stress tensor σ given by the formula [59]
σ = σs P + (λs + σs ) tr(ε )P + 2(μs − σs ) ε + σs ∇ u. (1.12)
In Eq. (1.12), P = I − nn = Pij ii ij is the surface projection tensor, σs is

the surface

tension, μs and λs are the surface elastic constants measured
in N/m . Also, tr(ε ) is a trace of the surface strain tensor ε , and ∇ u
is the surface gradient of the displacement field. The first three terms
in Eq. (1.12) constitute a tensor tangent to the surface, while the last
term includes a component normal to it. The presence of the normal
component in the last term of Eq. (1.12) is due to the existence of the
normal component of the displacement vector u and curvature of the
interface.
1.4. Method of solution

1.4.1 Multipole expansion: why and how?
A solution to the model boundary-value problem can be obtained by vari-
ous methods. The computational power available now for simulation makes
the numerical methods an efficient tool for the analysis of complex hetero-
geneous microstructures (see, e.g., [64,65]). At the same time, a rigorous
analytical solution is advantageous in that it
• Provides a clear insight into the physical essence of the problem;
• Allows an efficient exhaustive parametric study and obtaining the an-
alytical relationships (rather than empirical correlations) between the
parameters under study;
• Serves as a benchmark for testing the applicability of the new/existing
approximate models and accuracy of numerical methods;
• Gives a reliable theoretical basis for derivation of the simplified theories
and engineering formulas convenient for use and, at the same time,
rigorously justified.
Moreover, using an appropriate computational scheme makes an ana-
lytical approach far more efficient than the direct numerical methods. The
primary reason is that any work done analytically reduces the numerical
effort. This fully applies to the multipole expansion method we deal with
in this book. More important is that the efficiency of this method becomes
even more prominent for the large-scale models. A great illustration is the
fast multipole method (see, e.g., [66]) where the numerical effort scales
as O(N ), N being the number of inhomogeneities. This feature enables
consideration of the realistic, large-scale models and makes the multipole
expansion method probably the most efficient tool of micromechanics. The
mentioned arguments in favor of the multipole expansion method are sup-
ported by the author’s personal experience in solving a wide range of the
multiple inhomogeneity problems of the mechanics of heterogeneous solids
and fluids. This method will be discussed in detail in the subsequent chap-
ters. Here, we outline briefly its basic features.
In the scientific literature, the “multipole expansion” term usually refers
to the class of methods designed to study the behavior of large-scale collec-
tions of interacting objects of various nature, from atoms and molecules up
to stars and galaxies. Being mostly analytical in nature, these methods pro-
vide a theoretical basis of very efficient computational algorithms and found
many applications in astronomy, physics, chemistry, engineering, statistics,
etc. This list involves also the mechanics of heterogeneous solids and fluid
suspensions, where a certain progress has been achieved in recent years in

the application of the multipole expansion technique. However, in the au-
thor’s opinion, a true value of this method for the mechanics of composites
is still underestimated and its potential in this area is not fully discovered
so far.
There exists some diversity in the literature in using the terms “mul-
tipole” and “multipole expansion”. Historically, the “multipole” notion is
traced back to Maxwell [16] who defined it as a point source (monopole, or
simple charge, dipole, quadrupole, and so on). Among several definitions
available in the literature, probably the most general one is given by [67]:
“. . . A multipole expansion is a series expansion of the effect produced by
a given system in terms of an expansion parameter which becomes small as
the distance away from the system increases”. This definition is not strictly
related to a specific geometry or class of functions and, in what follows, we
accept it. The basis functions and relevant series coefficients are referred to
as multipole fields and multipole moments/strengths, respectively. One way
or another, we follow the already established tradition and refer to the ap-
proach considered in this book as the multipole expansion method despite
the fact that the multipole expansion (more exactly, the series expansion
over a set of the irregular partial solutions of the equilibrium equation)
constitutes only a part of the technique. The basic idea of the method is
to reduce the primary boundary-value problem formulated on the multi-
ply connected, piecewise-homogeneous domain to an ordinary system of
linear algebraic equations with constant matrix coefficients. In so doing, a
considerable analytical effort involving the theory of special functions and
integral transforms is required. This effort is quite rewarding, in view of the
derived by this way remarkably simple and highly efficient computational
algorithms.
As to the scope and application limits of the method, we note first
that the above mentioned analytical work can be done only for the linear
boundary-value problems. Second, the method we apply is essentially the
series expansion method, where the partial solutions of a differential equa-
tion obtained by separation of variables in an appropriate coordinate system
constitute a countable set of basis functions. For example, Laplace equation
allows separation of variables in 11 coordinate systems and R-separation in
17 coordinate systems (see, e.g., [68]). The specific choice of a coordinate
system is dictated by the inhomogeneity shape and introduced in a way that
the matrix–inhomogeneity interface coincides with the coordinate surface.
The partial solutions obtained by the separation of variables possess the fol-
Another random document with
no related content on Scribd:
that five thousand, belongs to me by rights. Oh, never mind, I’m
going along. Do you hear?” he shouted, his fists raised, “I’m going
along, I tell you. There ain’t a man of you big enough to stop me.
Let’s see you try and stop me going. Let’s see you once, any two of
you.” He filled the barroom with his clamor.
“Lord love you, come along, then,” said the sheriff.
The posse rode out of Keeler that same night. The keeper of the
general merchandise store, from whom Marcus had borrowed a
second pony, had informed them that Cribbens and his partner,
whose description tallied exactly with that given in the notice of
reward, had outfitted at his place with a view to prospecting in the
Panamint hills. The posse trailed them at once to their first camp at
the head of the valley. It was an easy matter. It was only necessary
to inquire of the cowboys and range-riders of the valley if they had
seen and noted the passage of two men, one of whom carried a bird-
cage.
Beyond this first camp the trail was lost, and a week was wasted in
a bootless search around the mine at Gold Gulch, whither it seemed
probable the partners had gone. Then a traveling peddler, who
included Gold Gulch in his route, brought in the news of a wonderful
strike of gold-bearing quartz some ten miles to the south on the
western slope of the range. Two men from Keeler had made a strike,
the peddler had said, and added the curious detail that one of the
men had a canary bird in a cage with him.
The posse made Cribbens’ camp three days after the
unaccountable disappearance of his partner. Their man was gone,
but the narrow hoof-prints of a mule, mixed with those of huge hob-
nailed boots, could be plainly followed in the sand. Here they picked
up the trail and held to it steadily till the point was reached where,
instead of tending southward, it swerved abruptly to the east. The
men could hardly believe their eyes.
“It ain’t reason,” exclaimed the sheriff. “What in thunder is he up
to? This beats me! Cutting out into Death Valley at this time of year!”
“He’s heading for Gold Mountain over in the Armagosa, sure.”
The men decided that this conjecture was true. It was the only
inhabited locality in that direction. A discussion began as to the
further movements of the posse.
“I don’t figure on going into that alkali sink with no eight men and
horses,” declared the sheriff. “One man can’t carry enough water to
take him and his mount across, let alone eight. No, sir. Four couldn’t
do it. No, three couldn’t. We’ve got to make a circuit round the valley
and come up on the other side and head him off at Gold Mountain.
That’s what we got to do, and ride like blazes to do it, too.”
But Marcus protested with all the strength of his lungs against
abandoning the trail now that they had found it. He argued that they
were now but a day and a half behind their man. There was no
possibility of their missing the trail—as distinct in the white alkali as
in snow. They could make a dash into the valley, secure their man,
and return long before their water failed them. He, for one, would not
give up the pursuit, now that they were so close. In the haste of the
departure from Keeler the sheriff had neglected to swear him in. He
was under no orders. He would do as he pleased.
“Go on, then, you darn fool,” answered the sheriff. “We’ll cut on
round the valley, for all that. It’s a gamble he’ll be at Gold Mountain
before you’re half-way across. But if you catch him, here”—he
tossed Marcus a pair of handcuffs—“put ’em on him and bring him
back to Keeler.”
Two days after he had left the posse, and when he was already far
out in the desert, Marcus’s horse gave out. In the fury of his
impatience he had spurred mercilessly forward on the trail, and on
the morning of the third day found that his horse was unable to
move. The joints of his legs seemed locked rigidly. He would go his
own length, stumbling and interfering, then collapse helplessly upon
the ground with a pitiful groan. He was used up.
Marcus believed himself to be close upon McTeague now. The
ashes at his last camp had still been smoldering. Marcus took what
supplies of food and water he could carry, and hurried on. But
McTeague was farther ahead than he had guessed, and by evening
of his third day upon the desert Marcus, raging with thirst, had drunk
his last mouthful of water and had flung away the empty canteen.
“If he ain’t got water with um,” he said to himself, as he pushed on,
“if he ain’t got water with um, I’ll be in a bad way. I will, for a fact.”
At Marcus’s shout McTeague looked up and around him. For the

instant he saw no one. The white glare of alkali was still unbroken.
Then his swiftly rolling eyes lighted upon a head and shoulder that
protruded above the low crest of the break directly in front of him. A
man was there, lying at full length upon the ground, covering him
with a revolver. For a few seconds McTeague looked at the man
stupidly, bewildered, confused, as yet without definite thought. Then
he noticed that the man was singularly like Marcus Schouler. It was
Marcus Schouler. How in the world did Marcus Schouler happen to
be in that desert? What did he mean by pointing a pistol at him that
way? He’d best look out or the pistol would go off. Then his thoughts
readjusted themselves with a swiftness born of a vivid sense of
danger. Here was the enemy at last, the tracker he had felt upon his
footsteps. Now at length he had “come on” and shown himself, after
all those days of skulking. McTeague was glad of it. He’d show him
now. They two would have it out right then and there. His rifle! He
had thrown it away long since. He was helpless. Marcus had ordered
him to put up his hands. If he did not, Marcus would kill him. He had
the drop on him. McTeague stared, scowling fiercely at the leveled
pistol. He did not move.
“Hands up!” shouted Marcus a second time. “I’ll give you three to
do it in. One, two—” Instinctively McTeague put his hands above his
head.
Marcus rose and came towards him over the break.
“Keep ’em up,” he cried. “If you move ’em once I’ll kill you, sure.”
He came up to McTeague and searched him, going through his
pockets; but McTeague had no revolver; not even a hunting knife.
“What did you do with that money, with that five thousand dollars?”
“It’s on the mule,” answered McTeague, sullenly.
Marcus grunted, and cast a glance at the mule, who was standing
some distance away, snorting nervously, and from time to time
flattening his long ears.
“Is that it there on the horn of the saddle, there in that canvas
sack?” Marcus demanded.
“Yes, that’s it.”
A gleam of satisfaction came into Marcus’s eyes, and under his
breath he muttered: “Got it at last.”
He was singularly puzzled to know what next to do. He had got
McTeague. There he stood at length, with his big hands over his
head, scowling at him sullenly. Marcus had caught his enemy, had
run down the man for whom every officer in the state had been
looking. What should he do with him now? He couldn’t keep him
standing there forever with his hands over his head.
“Got any water?” he demanded.
“There’s a canteen of water on the mule.”
Marcus moved toward the mule and made as if to reach the bridle-
rein. The mule squealed, threw up his head, and galloped to a little
distance, rolling his eyes and flattening his ears.
Marcus swore wrathfully.
“He acted that way once before,” explained McTeague, his hands
still in the air. “He ate some loco-weed back in the hills before I
started.”
For a moment Marcus hesitated. While he was catching the mule
McTeague might get away. But where to, in heaven’s name? A rat
could not hide on the surface of that glistening alkali, and besides, all
McTeague’s store of provisions and his priceless supply of water
were on the mule. Marcus ran after the mule, revolver in hand,
shouting and cursing. But the mule would not be caught. He acted as
if possessed, squealing, lashing out, and galloping in wide circles,
his head high in the air.
“Come on,” shouted Marcus, furious, turning back to McTeague.
“Come on, help me catch him. We got to catch him. All the water we
got is on the saddle.”
McTeague came up.
“He’s eatun some loco-weed,” he repeated. “He went kinda crazy
before.”
“If he should take it into his head to bolt and keep on running—”
Marcus did not finish. A sudden great fear seemed to widen
around and inclose the two men. Once their water gone, the end
would not be long.
“We can catch him all right,” said the dentist. “I caught him once
before.”
“Oh, I guess we can catch him,” answered Marcus, reassuringly.
Already the sense of enmity between the two had weakened in the
face of a common peril. Marcus let down the hammer of his revolver
and slid it back into the holster.
The mule was trotting on ahead, snorting and throwing up great
clouds of alkali dust. At every step the canvas sack jingled, and
McTeague’s bird-cage, still wrapped in the flour bags, bumped
against the saddle-pads. By and by the mule stopped, blowing out
his nostrils excitedly.
“He’s clean crazy,” fumed Marcus, panting and swearing.
“We ought to come up on him quiet,” observed McTeague.
“I’ll try and sneak up,” said Marcus; “two of us would scare him
again. You stay here.”
Marcus went forward a step at a time. He was almost within arm’s
length of the bridle when the mule shied from him abruptly and
galloped away.
Marcus danced with rage, shaking his fists, and swearing horribly.
Some hundred yards away the mule paused and began blowing and
snuffing in the alkali as though in search of food. Then, for no
reason, he shied again, and started off on a jog trot toward the east.
“We’ve got to follow him,” exclaimed Marcus, as McTeague came
up. “There’s no water within seventy miles of here.”
Then began an interminable pursuit. Mile after mile, under the
terrible heat of the desert sun, the two men followed the mule,
racked with a thirst that grew fiercer every hour. A dozen times they
could almost touch the canteen of water, and as often the distraught
animal shied away and fled before them. At length Marcus cried:
“It’s no use, we can’t catch him, and we’re killing ourselves with
thirst. We got to take our chances.” He drew his revolver from its
holster, cocked it, and crept forward.
“Steady now,” said McTeague; “it won’t do to shoot through the
canteen.”
Within twenty yards Marcus paused, made a rest of his left
forearm and fired.
“You got him,” cried McTeague. “No, he’s up again. Shoot him
again. He’s going to bolt.”
Marcus ran on, firing as he went. The mule, one foreleg trailing,
scrambled along, squealing and snorting. Marcus fired his last shot.
The mule pitched forward upon his head, then, rolling sideways, fell
upon the canteen, bursting it open and spilling its entire contents into
the sand.
Marcus and McTeague ran up, and Marcus snatched the battered
canteen from under the reeking, bloody hide. There was no water
left. Marcus flung the canteen from him and stood up, facing
McTeague. There was a pause.
“We’re dead men,” said Marcus.
McTeague looked from him out over the desert. Chaotic desolation
stretched from them on either hand, flaming and glaring with the
afternoon heat. There was the brazen sky and the leagues upon
leagues of alkali, leper white. There was nothing more. They were in
the heart of Death Valley.
“Not a drop of water,” muttered McTeague; “not a drop of water.”
“We can drink the mule’s blood,” said Marcus. “It’s been done
before. But—but—” he looked down at the quivering, gory body “—
but I ain’t thirsty enough for that yet.”
“Where’s the nearest water?”
“Well, it’s about a hundred miles or more back of us in the
Panamint hills,” returned Marcus, doggedly. “We’d be crazy long
before we reached it. I tell you we’re done for. We ain’t ever going to
get outa here.”
“Done for?” murmured the other, looking about stupidly. “Done for,
that’s the word. Done for? Yes, I guess we’re done for.”
“What are we going to do now?” exclaimed Marcus, sharply, after
a while.
“Well, let’s be moving along—somewhere.”
“Where, I’d like to know? What’s the good of moving on?”
“Wat’s the good of stopping here?”
There was a silence.
“Lord, it’s hot,” said the dentist, finally, wiping his forehead with the
back of his hand. Marcus ground his teeth.
“Done for,” he muttered; “done for.”
“I never was so thirsty,” continued McTeague. “I’m that dry I can
hear my tongue rubbing against the roof of my mouth.”
“Well, we can’t stop here,” said Marcus, finally; “we got to go
somewhere. We’ll try and get back, but it ain’t no manner of use.
Anything we want to take along with us from the mule? We can—”
Suddenly he paused. In an instant the eyes of the two doomed
men had met as the same thought simultaneously rose in their
minds. The canvas sack with its five thousand dollars was still tied to
the horn of the saddle.
Marcus had emptied his revolver at the mule, and though he still
wore his cartridge belt, he was for the moment as unarmed as
McTeague.
“I guess,” began McTeague, coming forward a step, “I guess even
if we are done for, I’ll take—some of my truck along.”
“Hold on,” exclaimed Marcus, with rising aggressiveness. “Let’s
talk about that. I ain’t so sure about who that—who that money
belongs to.”
“Well, I am, you see,” growled the dentist.
The old enmity between the two men, their ancient hate, was
flaming up again.
“Don’t try an’ load that gun either,” cried McTeague, fixing Marcus
with his little eyes.
“Then don’t lay your finger on that sack,” shouted the other.
“You’re my prisoner, do you understand? You’ll do as I say.” Marcus
had drawn the handcuffs from his pocket, and stood ready with his
revolver held as a club. “You soldiered me out of that money once,
and played me for a sucker, an’ it’s my turn now. Don’t you lay your
finger on that sack.”
Marcus barred McTeague’s way, white with passion. McTeague did
not answer. His eyes drew to two fine, twinkling points, and his
enormous hands knotted themselves into fists, hard as wooden
mallets. He moved a step nearer to Marcus, then another.
Suddenly the men grappled, and in another instant were rolling
and struggling upon the hot, white ground. McTeague thrust Marcus
backward until he tripped and fell over the body of the dead mule.
The little bird-cage broke from the saddle with the violence of their
fall, and rolled out upon the ground, the flour-bags slipping from it.
McTeague tore the revolver from Marcus’s grip and struck out with it
blindly. Clouds of alkali dust, fine and pungent, enveloped the two
fighting men, all but strangling them.
McTeague did not know how he killed his enemy, but all at once
Marcus grew still beneath his blows. Then there was a sudden last
return of energy. McTeague’s right wrist was caught, something
clicked upon it, then the struggling body fell limp and motionless with
a long breath.
As McTeague rose to his feet, he felt a pull at his right wrist;
something held it fast. Looking down, he saw that Marcus in that last
struggle had found strength to handcuff their right wrists together.
Marcus was dead now; McTeague was locked to the body. All about
him, vast, interminable, stretched the measureless leagues of Death
Valley.
McTeague remained stupidly looking around him, now at the
distant horizon, now at the ground, now at the half-dead canary
chittering feebly in its little gilt prison.—Arranged from “McTeague.”
Copyright by Doubleday, Page & Co., New York, and used by kind
permission.
MICHAEL STROGOFF, COURIER OF THE CZAR

By Jules Verne
The door of the imperial cabinet was opened and General Kissoff
was announced.
“The courier?” inquired the Czar eagerly.
“He is here, sire,” replied General Kissoff.
“Let him come in,” said the Czar.
In a few moments Michael Strogoff, the courier, entered. The Czar
fixed a penetrating look upon him without uttering a word. Then in an
abrupt tone—
“Thy name?”
“Michael Strogoff, sire.”
“Thy rank?”
“Captain in the corps of Couriers to the Czar.”
“Thou dost know Siberia?”
“I am a Siberian.”
“A native of—?”
“Omsk, sire.”
“Hast thou relations there?”
“Yes, sire, my aged mother.”
The Czar suspended his questions for a moment; then pointed to
a letter which he held in his hand:
“Here is a letter which I charge thee, Michael Strogoff, to deliver
into the hands of the Grand Duke, and to no one but him.”
“I will deliver it, sire.”
“The Grand Duke is at Irkutsk. Thou wilt have to traverse a
rebellious country, invaded by Tartars, whose interest it will be to
intercept this letter.”
“I will traverse it.”
“Above all, beware of the traitor, Ivan Ogareff, who will perhaps
meet thee on the way.”
“I will beware of him.”
“Wilt thou pass through Omsk?”
“Sire, that is my route.”
“If thou dost see thy mother, there will be the risk of being
recognized. Thou must not see her!”
Michael Strogoff hesitated a moment, and then said:
“I will not see her.”
“Swear to me that nothing will make thee acknowledge who thou
art, nor whither thou art going.”
“I swear it.”
“Michael Strogoff, take this letter. On it depends the safety of all
Siberia, and perhaps the life of my brother, the Grand Duke.”
“This letter shall be delivered to His Highness, the Grand Duke.”
“Go, thou, for God, for the Czar, and for your native land.”
The courier saluted his sovereign and that very night set out to
fulfill his perilous mission. All went well until he reached Omsk.
Compelled to stop here for food and a change of horses, he was
about to leave the posting house to continue his journey when
suddenly a cry made him tremble—a cry which penetrated to the
depths of his soul—and these two words rushed into his ear: “My
son!”
His mother, the old woman, Marfa, was before him! Trembling, she
smiled upon him and stretched forth her arms to him. Michael
Strogoff stepped forward; he was about to throw himself—when the
thought of duty, the serious danger to himself, and his mother, in this
unfortunate meeting, stopped him, and so great was his self-
command that not a muscle of his face moved. There were twenty
people in the public room, and among them perhaps spies, and was
it not known that the son of Marfa Strogoff belonged to the corps of
Couriers to the Czar? Michael Strogoff did not move.
“Michael!” cried his mother.
“Who are you, my good woman?”
“Who am I? Dost thou no longer know thy mother?”
“You are mistaken; a resemblance deceives you.”
Marfa went up to him, and looking straight into his eyes, said: “Art
thou not the son of Peter and Marfa Strogoff?”
Michael would have given his life to have locked his mother in his
arms. But if he yielded now it was all over with him, with her, with his
mission, with his oath. Completely master of himself, he closed his
eyes that he might not see the inexpressible anguish of his mother.
“I do not know, in truth, what it is you say, my good woman.”
“Michael!”
“My name is not Michael. I never was your son! I am Nicholas
Kopanoff, a merchant of Irkutsk.”
And suddenly he left the room, while for the last time the words
echoed in his ears,—
“My son! My son!”
Michael Strogoff by a desperate effort had gone. He did not heed
his old mother, who had fallen back almost inanimate on a bench.
But when the postmaster hastened to assist her, the aged woman
raised herself. Suddenly the thought occurred to her: She denied by
her own son. It was impossible! As for being deceived, it was equally
impossible. It was certainly her son whom she had just seen; and if
he had not recognized her it was because he had some strong
reason for acting thus. And then, her mother-feelings arising within
her, she had only one thought: Can I unwittingly have ruined him?
“I am mad,” she said to her interrogators. “This young man was
not my son; he had not his voice. Let us think no more of it. If we do,
I shall end in finding him everywhere.”
This scene, however, was immediately reported to Ivan Ogareff,
who was stationed in the town. He at once arrested Michael Strogoff,
and then had Marfa brought before him. Marfa, standing before Ivan
Ogareff, drew herself up, crossed her arms on her breast, and
waited. “You are Marfa Strogoff?” asked Ogareff.
“Yes.”
“Do you retract what you said a few hours ago?”
“No.”
“Then you do not know that your son, Michael Strogoff, Courier to
the Czar, has passed through Omsk?”
“I do not know it.”
“And the man whom you thought you recognized as your son was
not your son?”
“He was not my son.”
“And since then, you have seen him among the prisoners?”
“No.”
“If he were pointed out to you, would you recognize him?”
“No.”
“Listen! Your son is here, and you shall immediately point him out
to me.”
“No.”
“All these men will file before you, and if you do not show me
Michael Strogoff, you shall receive as many blows from the knout as
men shall have passed before you.”
On an order from Ogareff, the prisoners filed one by one past
Marfa, who was immovable as a statue, and whose face expressed
only perfect indifference. Michael was to all appearances unmoved,
but the palms of his hands bled under the nails which were pressed
into the flesh.
Marfa, seized by two soldiers, was forced on her knees on the
ground. Her dress torn off left her back bare. A saber was placed
before her breast at a few inches’ distance. If she bent beneath her
sufferings, her breast would be pierced by the sharp steel. The Tartar
drew himself up and waited.
“Begin,” said Ogareff.
The whip whistled through the air, but, before it fell, a powerful
hand stopped the Tartar’s arm. Ivan Ogareff had succeeded.
“Michael Strogoff!” cried he.
“Himself!” said Michael, and raising the knout, he struck Ogareff a
blow across the face.
“Blow for blow!” said he.
Twenty soldiers threw themselves on Michael and in another
instant he would have been slain, but Ogareff stopped them.
“This man is reserved for the Emir’s judgment. Search him.”
The letter bearing the imperial arms was found in Michael’s
bosom; he had not time to destroy it. It was handed to Ogareff.
Michael was then led before the Emir.
“Your forehead to the ground!” exclaimed Ogareff.
“No!”
Two soldiers tried to make him bend, but were themselves laid on
the ground by a blow from Michael’s fist.
“Who is this prisoner?” asked the Emir.
“A Russian spy,” answered Ogareff.
In asserting that Michael was a spy, he knew that the sentence
would be terrible. The Emir made a sign, at which all bowed low their
heads. Then he pointed to the Koran, which was brought to him. He
opened the sacred book, and placing his finger on one of its pages,
read in a loud voice a verse ending in these words: “And he shall no
more see the things of this earth.”
“Russian spy, you have come to see what is going on in the Tartar
camp; then look while you may! You have seen for the last time. In
an instant your eyes will be for ever shut to the light of day.”
Michael’s fate was to be not death, but blindness. He was going to
be blinded in the Tartar fashion, with a hot saber-blade passed
before his eyes.
The Emir’s orders executed, Ivan Ogareff approached Michael,
drew from his pocket the Imperial letter, opened it and held it up
before the face of the Czar’s courier, saying with supreme irony:
“Read, now, Michael Strogoff, read, and go and repeat at Irkutsk
what you have read. The true Courier of the Czar is henceforth Ivan
Ogareff.”
The Emir retired with his train. Ivan followed after, and sightless
Michael was left alone to his fate. One thought possessed him. He
must somehow arrive at Irkutsk before the traitor and warn the
Grand Duke of the intended deception.
Some months later Michael Strogoff had reached his journey’s
end! He was in Irkutsk. Hastening to the governor’s palace to see
the Grand Duke, he meets in a waiting-room Ivan Ogareff, the traitor.
The latter must act quickly. Ogareff arose, and thinking he had an
immeasureable advantage over the blind man threw himself upon
him. But with one hand Michael grasps the arm of his enemy and
hurls him to the ground. Ogareff gathers himself together like a tiger
about to spring, and utters not a word. The noise of his footsteps, his
very breathing, he tries to conceal from the blind man. At last, with a
spring, he drives his sword full blast at Michael’s breast. An
imperceptible movement of the blind man’s knife turns aside the
blow. Michael is not touched, and coolly waits a second attack. Cold
drops stand on Ogareff’s brow; he draws back a step and again
leaps forward. But like the first, this attempt fails. Michael’s knife has
parried the blow from the traitor’s useless sword. Mad with rage and
terror, he gazes into the wide-open eyes of the blind man. Those
eyes which seem to pierce to the bottom of his soul, and which do
not, cannot, see, exercise a sort of dreadful fascination over him.
Suddenly Ogareff utters a cry: “He sees! He sees!”
“Yes, I see. Thinking of my mother, the tears which sprang to my
eyes saved my sight. I see the mark of the knout which I gave you,
traitor and coward! I see the place where I am about to strike you!
Defend your life! It is a duel I offer you! My knife against your sword!”
Ogareff now feels that he is lost, but, mustering up all his courage,
he springs forward. The two blades cross, but at a touch from
Michael’s knife the sword flies in splinters, and the wretch, stabbed
to the heart, falls lifeless to the ground.
At the same moment the door is thrown open, and the Grand
Duke, accompanied by some of his officers, enters. The Grand Duke
advances. In the body lying on the ground he recognizes the man
whom he believes to be the Czar’s Courier. Then in threatening
voice:
“Who killed this man?”
“I,” answered Michael.
“Thy name? Who dares kill the servant of my brother, the Czar’s
Courier?”
“That man, your highness, is not a Courier of the Czar! He is Ivan
Ogareff!”
“Ivan Ogareff!”
“Yes, Ivan the traitor.”
“But who are you, then?”
“Michael Strogoff.”
THE TIGER’S CAVE

AN ADVENTURE AMONG THE MOUNTAINS OF QUITO
On leaving the Indian village, we continued to wind round
Chimborazo’s wide base; but its snow-crowned head no longer
shone above us in clear brilliancy, for a dense fog was gathering
gradually around it. Our guides looked anxiously towards it, and
announced their apprehensions of a violent storm. We soon found
that their fears were well founded. The fog rapidly covered and
obscured the whole of the mountain; the atmosphere was
suffocating, and yet so humid that the steelwork of our watches was
covered with rust, and the watches stopped. The river beside which
we were traveling rushed down with still greater impetuosity; and
from the clefts of the rocks which lay on the left of our path were
suddenly precipitated small rivulets that bore the roots of trees and
innumerable serpents along with them. These rivulets often came
down so suddenly and violently that we had great difficulty in
preserving our footing. The thunder at length began to roll, and
resounded through the mountainous passes with the most terrific
grandeur. Then came the vivid lightning, flash following flash—
above, around, beneath—everywhere a sea of fire. We sought a
momentary shelter in a cleft of the rocks, while one of our guides
hastened forward to seek a more secure asylum. In a short time he
returned, and informed us that he had discovered a spacious cavern,
which would afford us sufficient protection from the elements. We
proceeded thither immediately, and with great difficulty, and not a
little danger, at last got into it.
The noise and raging of the storm continued with so much
violence that we could not hear the sound of our voices. I had placed
myself near the entrance of the cave, and could observe, through the
opening, which was straight and narrow, the singular scene without.
The highest cedar-trees were struck down, or bent like reeds;
monkeys and parrots lay strewed upon the ground, killed by the
falling branches; the water had collected in the path we had just
passed, and hurried along it like a mountain stream. From everything
I saw I thought it extremely probable that we should be obliged to
pass some days in this cavern. When the storm, however, had
somewhat abated, our guides ventured out in order to ascertain if it
were possible to continue our journey. The cave in which we had
taken refuge was extremely dark, so that if we moved a few paces
from the entrance we could see no more than an inch before us; and
we were debating as to the propriety of leaving it, even before the
Indians came back, when we suddenly heard a singular rumbling or
growling at the farther end of the cavern, which instantly fixed all our
attention. Wharton and myself listened anxiously, but our daring and
inconsiderate young friend Lincoln, together with my huntsman, crept
about upon their hands and knees, and endeavored to discover, by
groping, from whence the sound proceeded. They had not advanced
far into the cavern before we heard them utter an exclamation of
surprise; and they returned to us, each carrying in his arms an
animal singularly marked, and about the size of a cat, seemingly of
great strength and power, and furnished with immense fangs. The
eyes were of a green color; strong claws were upon their feet; and a
blood-red tongue hung out of their mouths. Wharton had scarcely
glanced at them, when he exclaimed, in consternation, “Good God!
we have come into the den of a—.” He was interrupted by a fearful
cry of dismay from our guides, who came rushing precipitately
towards us, calling out, “A tiger! a tiger!” and at the same time, with
extraordinary rapidity, they climbed up a cedar-tree, which stood at
the entrance of the cave, and hid themselves among the branches.
After the first sensation of horror and surprise, which rendered me
motionless for a moment, had subsided, I grasped my firearms.
Wharton had already regained his composure and self-possession;
and he called to us to assist him instantly in blocking up the mouth of
the cave with an immense stone, which fortunately lay near it. The
sense of approaching danger augmented our strength, for we now
distinctly heard the growl of the ferocious animal, and we were lost
beyond redemption if it reached the entrance before we could get it
closed. Ere this was done, we could distinctly see the tiger bounding
towards the spot and stooping in order to creep into his den by the
narrow opening. At this fearful moment our exertions were
successful, and the great stone kept the wild beast at bay. There
was a small open space, however, left between the top of the
entrance and the stone, through which we could see the head of the
animal, illuminated by its glowing eyes, which rolled, glaring with
fury, upon us. Its frightful roaring, too, penetrated to the depths of the
cavern, and was answered by the hoarse growling of the cubs, which
Lincoln and Frank had now tossed from them. Our ferocious enemy
attempted first to remove the stone with his powerful claws, and then
to push it with his head from its place; and these efforts, proving
abortive, served only to increase his wrath. He uttered a
tremendous, heart-piercing howl, and his flaming eyes darted light
into the darkness of our retreat.
“Now is the time to fire at him,” said Wharton, with his usual
calmness; “aim at his eyes; the ball will go through his brain, and we
shall then have a chance to get rid of him.”
Frank seized his double-barreled gun, and Lincoln his pistols; the
former placed the muzzle within a few inches of the tiger, and Lincoln
did the same. At Wharton’s command, they both drew the triggers at
the same moment, but no shot followed. The tiger, who seemed
aware that the flash indicated an attack upon him, sprang growling
from the entrance, but, feeling himself unhurt, immediately turned
back again, and stationed himself in his former place. The powder in
both pieces was wet. Frank and Lincoln, therefore, proceeded to
draw the useless charges while Wharton and myself hastened to
seek our powder-flask. It was so extremely dark that we were
obliged to grope about the cave; and, at last, coming in contact with
the cubs, we heard a rustling noise, as if they were playing with
some metal substance, which we soon discovered was the cannister
we were looking for. Most unfortunately, however, the animals had
pushed off the lid with their claws, and the powder had been strewed
over the damp earth and rendered entirely useless. This fearful
discovery excited the greatest consternation.
“All is now over,” said Wharton. “We have only now to choose
whether we shall die of hunger, together with these animals who are
shut up along with us, or open the entrance to the blood-thirsty
monster without, and so make a quicker end of the matter.”
So saying, he placed himself close beside the stone, which, for the
moment, defended us, and looked undauntedly upon the lightning
eyes of the tiger.
Lincoln raved and swore; and Frank took a piece of strong cord
from his pocket and hastened to the farther end of the cave—I knew
not with what design. We soon, however, heard a low, stifled
groaning; and the tiger, who had heard it also, became more restless
and disturbed than ever. He went backwards and forwards before
the entrance of the cave, in the most wild, impetuous manner, then
stood still, and stretching out his neck in the direction of the forest,
broke forth into a deafening howl. Our two Indian guides took
advantage of this opportunity to discharge several arrows from the
tree. The animal was struck more than once, but the light weapons
bounded back harmless from his thick skin. At length, however, one
of them struck him near the eye, and the arrow remained sticking in
the wound. He now broke anew into the wildest fury, and sprang at
the tree, and tore it with his claws, as if he would have dragged it to
the ground. But, having at length succeeded in getting rid of the
arrow, became more calm, and laid himself down as before in front
of the cave.
Frank now returned from the lower end of the den, and a glance
showed us what he had been doing. In each hand, and dangling
from the end of a string, were the two cubs. He had strangled them,
and before we were aware what he intended, he threw them through
the opening to the tiger. No sooner did the animal perceive them,
than he gazed earnestly upon them, and began to examine them
closely, turning them cautiously from side to side. As soon as he
became aware that they were dead, he uttered so piercing a howl of
sorrow that we were obliged to put our hands to our ears. When I
upbraided my huntsman for the cruel action he had so rashly
committed, I perceived by his blunt and abrupt answers that he also
had lost all hope of rescue from our impending fate, and, that, under
these circumstances, the ties between master and servant were
dissolved. For my own part, without knowing why, I could not help
believing that some unexpected assistance would yet rescue us from
so horrible a fate. Alas! I little anticipated the sacrifice that my rescue
was to cost.
The thunder had now ceased, and the storm had sunk to a gentle
gale; the songs of the birds were again heard in the neighboring
forest, and the sunbeams sparkled in the drops that hung from the
leaves. We saw through the aperture that all nature was reviving
after the wild war of elements which had so recently taken place; but
the contrast only made our situation the more terrible. We were in a
grave from which there was no deliverance; and a monster, worse
than the fabled Cerberus, kept watch over us. The tiger had laid
himself down beside his whelps. He was a beautiful animal, of great
size and strength, and his limbs being stretched out at their full
length, displayed his immense power of muscle. A double row of
great teeth stood far enough apart to show his large red tongue, from
which the white foam fell in large drops. All at once, another roar was
heard at a distance, and the tiger immediately rose and answered it
with a mournful howl. At the same instant, our Indians uttered a
shriek, which announced that some new danger threatened us. A
few moments confirmed our worst fears, for another tiger, not quite
so large as the former, came rapidly towards the spot where we
were.
“This enemy will prove more cruel than the other,” said Wharton;
“for this is the female, and she knows no pity for those who deprive
her of her young.”
The howls which the tigress gave when she had examined the
bodies of her cubs, surpassed everything of the horrible that we had
yet heard; and the tiger mingled his mournful cries with hers.
Suddenly her roaring was lowered to a hoarse growling, and we saw
her anxiously stretch out her head, extend her wide and smoking
nostrils, and look as if she were determined to discover immediately
the murderers of her young. Her eyes quickly fell upon us, and she
made a spring forward with the intention of penetrating to our place
of refuge. Perhaps she might have been enabled by her immense

PDF Micromechanics of Composites Multipole Expansion Approach 2Nd Edition Volodymyr Kushch Ebook Full Chapter

Uploaded by

Copyright:

Available Formats

You might also like

PDF Micromechanics of Composites Multipole Expansion Approach 2Nd Edition Volodymyr Kushch Ebook Full Chapter

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

PDF Micromechanics of Composites Multipole Expansion Approach 2Nd Edition Volodymyr Kushch Ebook Full Chapter

Uploaded by

Copyright:

Available Formats

Micromechanics of Composites

Multipole Expansion Approach 2nd

Proceedings of the 2nd International Conference on

Undergraduate Mathematics Competitions 1995 2016 Taras

Handbook of Wood Chemistry and Wood Composites 2nd

Science and Engineering of Short Fibre Reinforced

Polymer Composites in the Aerospace Industry 2nd

Phenomenological Creep Models of Composites and

Modelling of Damage Processes in Biocomposites Fibre

Nanomechanics and Micromechanics-Generalized Models and

Library of Congress Cataloging-in-Publication Data

British Library Cataloguing-in-Publication Data

For information on all Butterworth-Heinemann publications

Publisher: Matthew Deans

Preface to the ﬁrst edition xv

1. Multipole expansion approach 1

Part 1. Particulate composites

2. Potential ﬁelds of interacting spherical inhomogeneities 35

2.3.2. Convergence proof 42

3. Periodic multipoles and RUC model of a composite 57

4. Elastic solid with spherical inhomogeneities 77

4.5.2. Local expansion 94

5. Elasticity of composite half-space, layer, and bulk 109

6. Conductivity of a solid with spheroidal inhomogeneities 137

6.4.2. Triply periodic harmonics 159

7. Elastic solid with spheroidal inhomogeneities 179

8. Composites with transversely isotropic constituents 237

8.4. Numerical examples 257

9. Conductivity of an ellipsoidal particle composite 269

10. Elasticity of an ellipsoidal particle composite 301

10.6. Numerical study 328

Part 2. Fibrous composites: 2D models

11. Circular ﬁber composite with perfect interfaces 343

12. Fibrous composite with interface cracks 387

12.4. In-plane elasticity 401

13. Solids with elliptic inhomogeneities 423

14. Cracked solids 451

15. Elliptic ﬁber composite with an imperfect interface 473

15.1.3. Resolving set of equations 476

16. Fibrous composite with anisotropic constituents 499

A. Spherical harmonics and related theory 519

Vector solutions of Lamé equation for a half-space 537

B. Spheroidal harmonics and related theory 543

C. Ellipsoidal harmonics and related theory 569

D. Selected properties of functions Rλ and Xλ 577

E. Elliptic harmonics and related theory 583

E.3. Integral transforms 588

engineering, statistics, etc. Introduced by J.C. Maxwell in 1873 and then

Kyiv, Ukraine Volodymyr Kushch

Kyiv, Ukraine Volodymyr Kushch

Multipole expansion approach

Micromechanics of Composites Copyright © 2020 Elsevier Inc. 1

comprehensive reviews of other publications on the subject. Up to now,

1.2. Structure models

1.2.2 Finite arrays of inhomogeneities

1.2.3 Composite band and layer

A reasonable compromise consists in using the so-called “quasi-

Figure 1.2 Composite band.