Professional Documents
Culture Documents
PDF Micromechanics of Composites Multipole Expansion Approach 2Nd Edition Volodymyr Kushch Ebook Full Chapter
PDF Micromechanics of Composites Multipole Expansion Approach 2Nd Edition Volodymyr Kushch Ebook Full Chapter
PDF Micromechanics of Composites Multipole Expansion Approach 2Nd Edition Volodymyr Kushch Ebook Full Chapter
https://textbookfull.com/product/proceedings-of-the-2nd-
international-conference-on-building-innovations-
icbi-2019-volodymyr-onyshchenko/
https://textbookfull.com/product/undergraduate-mathematics-
competitions-1995-2016-taras-shevchenko-national-university-of-
kyiv-2nd-edition-volodymyr-brayman/
https://textbookfull.com/product/handbook-of-wood-chemistry-and-
wood-composites-2nd-edition-roger-m-rowell-editor/
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/
https://textbookfull.com/product/polymer-composites-in-the-
aerospace-industry-2nd-edition-p-e-irving-editor/
https://textbookfull.com/product/phenomenological-creep-models-
of-composites-and-nanomaterials-phenomenological-creep-models-of-
composites-and-nanomaterials-leo-razdolsky/
https://textbookfull.com/product/modelling-of-damage-processes-
in-biocomposites-fibre-reinforced-composites-and-hybrid-
composites-1st-edition-mohammad-jawaid/
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.
ISBN: 978-0-12-823253-8
v
vi Contents
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
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.
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.
xix
CHAPTER ONE
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
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.
4 Micromechanics of Composites
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.
Multipole expansion approach 5
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.
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.
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-
8 Micromechanics of Composites
ate for this purpose. For more discussions on the subject, see [10,48,11],
among others.
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 ).
Multipole expansion approach 9
∂ T (i) ∂ T (i)
1
i α T , (i)
, = 0, α = 1, 2. (1.2)
i=0
∂ xj ∂ xj ∂ xk S
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
Multipole expansion approach 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]