Deformation of Elastic Solids AJIT K. MAL Mechanical, Aerospace & Nuclear Engineering Department University of California, Los Angeles SARVA JIT SINGH Department of Mathematics Maharshi Dayanand University, RohtakLibrary of Congress Cataloging in-Publication Data MAL, AK. (AjtK) Deformation of elastic solids / Ajit K. Mal, Sarva Sit Singh Pom Includes biographical references and index, ISBN 0-13-200700-2 1. Elastics, 2. Continuum mechanics. 1. Singh, Sarva J. 1. Title QASSLMES 1991 S31" 382—de20 90-41944 cP Editovaliproduetion supervision: Carlyn Serbreny Interior design: Joan Stone Cover design: Joe DiDamenico Prepress buyer: Linda Behrens “Manufacturing bujer: Dave Dickey (© 1991 by Prentice-Hall, In, {A Division of Simon & Schuster Engleword Cis, New Jersey 07632 ‘All sights reserved. No part of this book may be produced, in any form or by any means, ‘without permission in weiting rom the publisher Printed in the Unit States of America w9876543-24 ISBN O-13-200700-2 nie et iran (OR) Le, anon rentie-Hall of Atala Py. Limited, Sydney Prentice-Hall Canada Tne. Toronto Prentice-Hall Hispmoanericana, S.A., Mexico Prentice-Hall of Inia Private Limited, New Delhi Prentice-Hall of Span, Ine, Tokyo Simeon & Schuster Asia Pte, Lid, Singopore Editor Prentice-Hall do Brasil, Lids Ro de Janeiro Contents PREFACE vii Chapter 1 (A REVIEW OF THE ELEMENTARY THEORY OF ELASTICITY 7 La Strain 1 1.2 Sess 4 1.3 Equations of Equilibrium 6 1.4 Hooke’s Law 10 1.5 Thermal Effects 20 Problems 27 Bibliography 22 Chapter 2 CARTESIAN TENSORS 23 2.1 Algebraic Definition of a Vector; The Transformation Matrix 23 2.2 The Index Notation 28 2.3 General Definition and Algebra of Cartesian Tensors 32 2.4 Some Special Properties of Second-Or''~ Tensors 36 ww Contents 2.5 Principal Axes of a Real Symmetric Second-Order Tensor 40 2.6 Isotropic Tensors 45 2.7 Vector and Tensor Calculus 47 2.8 Physical Interpretation of Some Vector Differential Operators 52 2.9 Dyadics 53 2.10 Some Vector Identities in Cylindrical and Spherical Coordinges 55 Problems 69 Bibliography 63 Chapter 3 KINEMATICS OF DEFORMATION 64 1 Material and Spatial Coordinates 64 2 Displacement, Velocity, and Acceleration Fields 67 3 The Deformation Gradient Tensor 69 4 Deformation of Volume and Surface Elements 72 5 The Strain 2B {6 Homogeneous Deformations 78 7 The Strain-Displacement Equations and the Infinitesimal Strain Tensor 81 3.8 Equations of Compatibility for Infinitesimal Strains 85 3.9 Infinitesimal Strain-Displacement Relations in Cylindrical and Spherical Coordinates 88 Problems 99 Bibliography 93, Chapter 4 BALANCE LAWS AND ANALYSIS OF STRESS 94 4.1 Balance Laws 94 4.2 Properties of the Stress Vector 100 4.3 Cauchy’s Stress Tensor 102 4:4 Cauchy's Equation of Motion 103 4.5 Some Properties of the Cauchy Stress Tensor 107 4.6 Linearization of the Balance Equations 107 4.7 Balance of Mechanical Energy 109 48 Thermal Effects 111 4.9 Equations of Motion in Cylindrical and Spherical Coordinates 115 Problems 118 Bibliography 122 Contents Chapter 5 CONSTITUTIVE EQUATIONS 723 5.1 General Rules of Constitutive Theory 123 5.2 Blasticity 124 '5.3 The Strain Energy Approach: Hyperelasticity 130 5.4 Thermoelasticity 136 5.5 Linear Constitutive Equations 138 5.6 Linear Elastic Constitutive Equations for Fiber-Reinforced Composites 149 5.7 Linear Thermoelasticity 151 Problems 153 Bibliography 155 Chapter 6 ELASTOSTATICS 156 6.1 Exact Solution of Some Nonlinear Elastic Problems 156 6.2 Field Equations of Linear Elasticity 166 6.3 Navier’s Equations 170 6.4 Uniqueness of Solution 171 6.5 Betti-Rayleigh Reciprocity (B-R-R) Relations 173 6.6 Exact Solution of Some Linear Elastic Problems 175 Problems 186 Bibliography 190 Chapter 7 SOLUTION OF LINEAR ELASTOSTATIC PROBLEMS BY SPECIAL TECHNIQUES 797 7.1 Plane Problems 191 7.2 Plane Strain 192 73 Plane Stress 194 7.4 Airy's Stress Function in Cartesian Coordinates 197 7.5 Airy’s Stress Function in Polar Coordinates 204 7.6 Complex Variable Method 224 711 Transform Method 236 7.8 Three-Dimensional Problems 242 Problems 258 Bibliography 264w Contents Chapter 8 LINEAR ELASTODYNAMICS 265 8.1 Wave Motion 265, 8.2 Fickl Equations of Linear Elastodynamics 270 8.3 Plane Waves in Unbounded Media 283 8.4 Formulation of Boundary-Value Problems for Isotropic Solids 297 8.5 Two-Dimensional Problems 292 8.6 Lamb Waves 309 8.7 Corcentrated Force 312 Problems 318 Bibliography 320 Appendix A REVIEW OF MATRICES 327 A.1 Deinitions and Notations 321 ‘A.2 Rules of Matrix Algebra 323 ‘A.3 Eigenvalues and Eigenvectors of a Matrix 324 A4 The Positive-Definite Matrix 328 ‘AS Polar Decomposition of a Matrix 328 Bibliography 329 GENERAL BIBLIOGRAPHY 330 AUTHORINDEX 333 SUBJECT INDEX 335 Preface Most aerospace, mechanical, and civil engineering graduate programs in major USS. universities require a series of courses in the general area of solid me- chanics, The details of these courses vary from school to school and program to program. At onc end of the spectrum, some give preference to courses that ‘emphasize unified and exact treatment of solid and fluid mechanics, often re- ferred to as modern continuum mechanics. At the other end, a more traditional approach is taken, whereby an approximate, linear treatment of engineering problems is emphasized. thas become apparent over the years that neither of the two extremes is appropriate for modern engineering programs. Most engineering students find the continuum mechanics approach too “heavy” for their taste and training, while the classical linear approach denies them the basic knowledge and insight that are essential for facing the challenges of the future. It would certainly be ‘best if both approaches could be taught in two complete sets of courses. How- ever, this can be done only under a Ph.D. degree program in mechanics or un- der 4 major field of specialization in solid mechanics, where the number of other required courses can be adjusted. In recent years, such specialized pro- grams have been abandoned in favor of more traditional, discipline oriented programs—e.g., aerospace engineering, mechanical engineering, and civil en- fineering. The major field of specialization in mechanics of solids is also be- wtvi Preface coming somewhat rare, resulting inthe need for a new approach to the teach ing of solid mechanics, whereby the student is given the opportunity to learn the essential elements of both approaches in atime frame that is reasonable un- der his or her program of study. ‘Asa response to these changes, graduate courses in solid mechanics have been redesigned in many universities in recent years. At the University of California at Los Angeles (UCLA), a sequence of courses has been created. ‘The first course deals with the fundamental principles of continuum mechan- ies; the treatment is exact but restricted to elastic solids only. The approximate linear theory is ako introduced in this course as a special case. The second course is devoted entirely to the solution of relevant linear problems, with emm- phasis on analytical techniques. These are followed by a series of specialized courses on plasticity, wave propagation, fracture mechanics, etc. This book is based on the first two of these courses. Both authors have drawn upon their experience in teaching earlier versions of similar courses for over 15 years. The material has been used by a fairly large number of students and scrutinized by a number of faculty members with interest in solid mechan- ies. We hope thatthe book will be received enthusiastically by students, fac- ulty, and others with a similar interest Chapter 1 contains a review of the theory of elasticity as it is normally taught in undergraduate courses. The discussions are quite terse, since more detailed and advanced treatments of the various topics reviewed here are given in later chapters. The main purpose of including the review is to provide the motivation and setting for the more advanced treatment that follows ‘A full appreciation of the theoretical treatment of solid mechanics re- quires a good knowledge of mathematics. Since the mathematics preparation of the readers of this book is likely to be quite varied, it was deemed necessary to make the book zs self-contained as possible by including most of the rele- vant intermediate-level mathematics. This is done in Chapter 2. The topics in- ‘clude Cartesian tensors and cortain integral theorems. The treatment is rigor- ‘ous but not exhaustive. Chapter 3 deals with kinematics of deformation. A number of exact mea- sures of deformation are introduced, and their geometrical interpretations are given with illustrative examples. Linear approximations leading tothe classical infinitesimal strains are carefully discussed The balance laws of continuum mechanics are presented in Chapter 4 The concept of stress and its various measures are also introduced inthis chap- ter. The equations of motion and the energy equation are derived in their exact forms in the presenee or absence of thermal effects. The linearized forms of these equations arc obtained as special cases. Chapter 5 deals with constitutive equations for elastic solids. The general rules of constitutive theory are stated in simple terms, and the exact forms of these equations for isotropic solids are derived. The linear equations for isotropic as well as anisotropic solids are obtained as special cases. The linear Prefoce constitutive equations for the overall behavior of fiber-reinforced composites are also presented in this chapter. Exact solutions of a number of nonlinear and linear elastic problems is presented in Chapter 6. For the nonlinear case, consideration is restricted to the simplest possible problems that bring out the essential features of nonlin- carity. For the linear ease, some general theorems are discussed fist, followed by the exact solutions of a number of simple but useful problems. Chapter 7 is devoted entirely to solution techniques for linear elastic problems. Typical two- and three-dimensional boundary-value problems are considered in detail A brief exposition of linear elastodynamics is presented in Chapter 8 ‘The main features of elastic waves are discussed through simple illustrative ex- amples ‘A review of matrix algebra is given in the appendix. Results that are needed for the discussions in the main body of the book are presented without proof. ‘We have made a strong effort to present an azcount of the exact theory of deformation in a form that can be easily understood by most frstyear graduate students as well as by others with a baccalaureate degree from a U.S. univer- sity or its equivalent. The treatment is reasonably self-contained in that no significant prior knowledge of solid mechanics or of tensor analysis is needed, although some familiarity with elementary strength of materials can be helpful in appreciating the finer points of the exact theory. Complex variables are used in Chapter 7 to solve a few problems; readers who are not comfortable with this technique can omit this small section without any significant loss, since it is shown that these problems can be solved by other methods. We have selected the material of the book with great care in order to keep its size relatively small without sacrificing the essential features of the subject. The problem sets have been deliberately kept small, with the hope that the students will have sufficient time to work them all thoughtfully. We strongly recommend that they be supplemented by additional problems of in- terest to the instructor and students. ‘There are a number of weaknesses; most of these turned out to be un- avoidable. First, the notations for tensors and other mathematical quantities ‘did not come out to be as uncluttered and consistent as we had originally hoped them to be. This is due primarily to the fact that the theoretical treatment of solid mechanics requires the introduction of an enormous number of variables, and one eventually tends to run out of symbols! There is also the need to con- form to traditions and customs in the use of symbols. In spite of these Uifficulties, the uotations used in the book should be easy to follow and should not cause any confusion, ‘A second weakness is the fact only a few problems of nonlinear elasticity have been included in Chapter 6. Certainly, the inclusion of a number of other relevant problems would have been possible. However, this would require ax Preface fuller exposition of the exact theory than was felt appropriate for this book. As explained earlier, the objective here isto introduce the students to the most es- sential features of the nonlinear theory. Finally, several important topics involving inelastic effects (e.g., vis- coclasticity, plasticity) have not been included. Students interested in these special topiss should be able either to take such courses or to read up on their own after they have mastered the material included in the book ACKNOWLEDGMENTS ‘A major portion of the book was written when the first author was on sabbati- cal leave at the Institut fir Werkstoffmechanik in Freiburg, West Germany, as, 1 Senior Fulbright Scholar. The hospitality ofthe Institute Director, Dr. Erwin Sommer, and the generosity of the Fulbright Foundation are gratefully ac- knowledge Valuadle criticism of an earlier version of the manuscript was provided by Dr. Keita Walton and Dr. Gareth Parry of the University of Bath, England. Professors Russell Westmann of UCLA, Subhendu Datta of the University of Colorado, K. L. Brown of Montana State University, Marijan Dravinski of the University of Southern California, Arthur H. England and W. A. (Tony) Green of the University of Nottingham, and E. Rhian Green of Leicester University read the manuscript and suggested improvements. The input of Dr. M. R. Karim and many graduate students of UCLA, especially Shyh-Shiuh Lih, Ruey-Bin Yang, Ching-Chung Yin, and Russel Lund, was extremely helpful We express our gratitude to these individuals. ‘The subject matter of the book is one of the oldest in the history of sci- ence and engineering. AA large number of excellent books have appeared over the years. Our presentation has been heavily influenced by the books listed in the general bibliography. Finally, the book could not have been completed without the patient un- derstanding and encouragement of Rosita Mal and Harmohinder Singh. Ajit K. Mal Sarva Jit Singh A Review of the Elementary Theory of Elasticity In this chapter we present a brief exposition of the linear theory of elasticity as it can be found in most undergraduate texts. This will set the stage for more advanced treatment, which is the primary objective of this book. 1.1, STRAIN To begin with, let us consider a deformation that is independent of one Carte sian coordinate, say 2, and parallel to the 1y-plane. This type of deformation is known as plane strain. We shall return (othe problem of three-dimensional de- formation after we have completed our study of deformation in a plane Let a point P with coordinates (X, ¥) in the undeformed state be dis- placed to the point P’ with coordinates ‘x, y) due to the deformation of the body (Fig. 1-1). The vector displacement of the point P has Cartesian compo- nents (U,V), where Ues-X, Vey-¥ ay It wil be assumed that the displacement components are continuous and twice differentiable functions of X, Y or, y. Consider a small rectangular ele- ‘ment PORS in the undeformed state, with sides (dX, dY) parallel to the coor-2 ‘AReview ofthe Elementary Theory of Elasticity Chap. 1 Figure 1-1 The displacement vector. dinate axes (Fig. 1-2). Let the points P,Q, R, and S move, respectively, to P', Q', R’, and S’ after deformation. The coordinates of Q relative to P are (dX, 0) and the coordinates of Q” relative to P' are (dx, dy), where de = dx + du = ax +O ax (1.22) ana = Dax (1.26) We thus have Po [(« + Yay + (% ay |" ~ ar Dax where we have negkcted the squares and higher powers of (2U/@X) and (@V/aX). Therefore, the increase in length per unit length of the line PQ, de- noted by én. is given by (1.38) Similarly, the increase in length per unit length of the line PS is a cr (1.36) ey See.11 Strain y s R 2 av Fe 7, ax ex pax aq dx+St ax oO Figure 1-2. Deformation of line elements dX and a ‘The quantities ¢,, and ey, are known as normal strains. If y denotes the angle that P’Q’ makes with the x-axis, we have (from Fig. 1-2) (av /aX), T+ @U/aX) ‘Assuming that the angle 7 (measured in radians) is small and neglecting small {quantities of the second and higher orders in (@U/@X) and (@V/0X), we find tan y= av na ay (14a) Similarly, the angle y2 of Fig. 1-2 is au nD (1.4b) Let 2, denote the decrease in the angle between the two lines PQ and PS, which ate parallel to the x- and y-axes, respectively, before deformation. Equations (1.4) then yield ww degen t n= Dee (5) ‘The quantity ¢, is the shearing strain. The symbol 7 = 2e, is often used to denote the shearing strain in engineering applications. Relations (1.3) and (1.5) are obtained under the assumption that the dis- placement derivatives are small compared to unity. The theory of elasticity in4 ‘A oviow of the Elementary Theory of Eleticty Chap. 1 which the products and squares of derivatives of the displacement components with respect to the space coordinates are neglected in comparison with the derivatives themselves is known as infinitesimal or linear theory. In this chap- ter we shall confine ourselves to the linear theory only. ‘We have seen that in the two-dimensional problem of plane strain, there are two normal strains, ea and e,,, and one shearing strain, é. In the general case of three-dimensional deformation, there are three normal strains, eu, Es ‘xy and three shearing strains, ey, ex, and é,,. They are related to the displace- ment components U, V, and W through the equations ww ay" az \ =1(% en = 3\ ax (16a) av) _1 a) 3% — (1.6) (1.60) 1.2 STRESS AA solid may be acted upon by two types of extemal forces: body forces and surface forces. Body forces act upon every volume element of the solid. Sur- face forces, in contrast, are forces that act upon every element of the surface of the body. In addition to the extemal forces, there are internal forces, which arise from the mutual interaction between various parts of the body. Let a deformable solid be in its unstrained state with no forces acting on it and let a system of forces be then applied to it. On account of the application of these forces, the solid becomes deformed, and a system of internal forces is set up within it to oppose this deformation. These intemal forces give rise to what is known as stress within the solid. Let us consider a part of the solid occupying a region V enclosed by the surface $ in th: deformed state. The boundary $ is acted upon by surface forces caused by the action of the material exterior to V on that within V. It will be assumed that these surface forces are continuously distributed over S. A sui able measure of such forces is their intensity, ie., the amount of force per unit area of the surface on which they act. ‘To specity the stress acting on a small area 8S at a point P on S. we as- sume that the forces acting across this elementary area, due to the action of the material outside V, can be reduced to a single force 6p (Fig. 1-3) and that the limit lim SP an Sec.12 Stress Ow Az Mf R ° = Figure 13 The sure fore exists. This limit is known as the traction, or the stress vector, on 85 at the point P. In general the traction is inclined to the area 8S on which it acts, and we ‘ean resolve it into to components: a normal stress perpendicular to the area and a shearing siress acting inthe plane ofthe area AS. Alternatively, we can resolve the traction into its Cartesian components. Let the outward drawn unit normal to aS be n and the x-,y-, and 2-components of the traction acting on 55 be denoted by ", 4, and 1°, respectively. ‘Consider a small cubic element with sides parallel to the coordinate axes (Fig. 1-4). The components of the traction acting on the face with normal in the positive s-direction are 4, 2. We use the notations o (1.8) Similarly, the components of the traction acting on the face of the cube with normal in the positive y-direction are oy, Cy, Gj. and the components of the traction acting on the face of the cube with normal in the positive z-direction Are Gx, Oy, Oe. Thus the frst suffix indicates the direction of the normal to the face and the second suffix indicates the direction of the traction component. In all, we have nine components, Gir, Jny,*** » Oy Which are known as the com- ponents of stress. We can display these components in the form of the matrix: M00, A Som On Oy Oe as) Ox Oy Fn6 ‘A Review ofthe Elementary Theory of Elasticity Chap. 1 Figure 1-4 stress components Obviously the components Gx, , 0: fepresent normal stresses and the com- PONENIS Cry, Ty, Tuy Ours Gps Ory Fepresent shearing stresses We use the convention that the normal stress on a surface is positive when it produces tension and negative when it produces compression of the ‘material within the element. The positive direction of a component of the shearing stress on any face of the cubic element is taken in the positive (nega- tive) direction of the coordinate axis if a tensile stress on the same face is in the Positive (negative) cirection of the corresponding axis. This rule is illustrated in Fig. I-4 by indicating the positive dicections of the components 4, dy, and rs tor the two faces of the cubic element with normals in the positive x-direc- tion and the negative x-direction, respectively. 1.3 EQUATIONS OF EQUILIBRIUM Consider the equilibrium of a small rectangular parallelepiped with its center at Plax, y, 2) and edges 8x, 8y, Be parallel to the coordinate axes (Fig. 1-3). The centers ofthe six faces ofthe parallelepiped are atthe points (et Hoy, Oy £482, (eye #452) If the components of stress at P are Ou, Gy * + » On, then the compo- nents of the traction acting on the face with its center at (x + x/2, y, 2) ate | aoe 1 doy 1800 wt TSE bx, Oy +52 br, on + 5 SH aelet 2 rane R Cre Uar Reet sinee the outward normal to this fae isin the postive x-direction. The compo- nents of the traction acting on the face with its center at (x ~ 6/2, y, 2) are (re PH). lor Ha). Sec. 1.3 Equations of Equilibrium 7 bx. Ox! > (c+ $dcz,) v6 Figure 1-5 Equlirium of a volume slement since the outward normal to this face is in the negative x-direction. Similar ex- pressions can be written dawn for the remaining four faces. If (ff) denote the components of the external body forces per unit volume, then summing, forces along the x-axis, we obtain (Brae (o-oo 1 dow 1 do 7 om + £98 5y)828x ~ (ou — 4 Moy) 26x : rot ty tae ( Day”, 1 = 4 isa. oe + (00+ £2226) = (05 ~ 48) fo Dividing by dxdyéz, this yields Mw 5 Wey f= 1.104) a Se (1.100) Similarly, summing forces along the y- and z-directions, we get We 4 ny Way 5 =9 (1.106) ax ay dos, Bo , 30, 4 305 0 1.10¢) Fe + Gay Ses f (1.106) Equations (1.10) are known as the stress equations of equilibrium for the solid. For the equilibrium of the small rectangular parallelepiped under consid- eration, the moments of all the forces about the coordinate axes must vanish separately. Taking moment about a line through P parallel to the x-axis, we find8 ‘A oview ofthe Elementary Theory of Elesticty Chap. 1 1 1 d0% 1 hay ( +4 esy) obs + bby (o 2a Dividing by 8x8y82, we obtain te (oy + 18s) ~ Lae (og 28232) =0 On = Oy (ula) Similarly, taking moments about the lines through P parallel to the y- and z- Ou Ou Gy = Ow (nb) ‘The symmety relations (1.11) imply that there are only six independent com- ponents of stress, ‘We new prove an important result—namely, the traction at a point across ‘an arbitrarily oriented surface element can be expressed as a linear combination of the stress components at that point. To this end, we assume, for simplicity, that =0 (2) Consider the equilibrium of a small element of volume in the form of a right triangular prism (Fig. 1-6). Let the external body force per unit volume be (fs, fn, 0). Because of the conditions (1.12), the two triangular end sections of the prism are traction-ree. Therefore, the material inside the prism is in equi librium under the tactions acting on the three rectangular faces of the prism and the body force (fe fr 0). Noting that the outward drawn normals to the faces in the yz- and xz-planes are, respectively, inthe negative x- and y-direc- tions, the components of tractions on these faces are (Ca, ~diny 0) and (ors, oj 0). Let the unit normal to the slanted face be m and the compo- nents of traction on it be denoted by (¢", «Then, summing forces along the x-axis, we find H°AB ~ 008 ~ 004 + f($0P 4B) where P is the foot of the perpendicular drawn from the origin to the line AB. Noting that OA = AB sin @ and OB = AB cos @ and taking the limit as OP —> 0, keeping n fixed, we find that e 008 8 + oysin 0 (1.13a) Similarly, samming forces slong the y- and z-axes, we obtain 008 8 + oysin 0 (1.136) (1.13¢) “ hea Sec. 1.3 Equations of Equilibrium 9 y Figure 16 Equilibrium ofa prism element Equation (1.13) can be expressed in the matrix form, 1) [ow on O] fm Wh = lay oy Ol)m, (4) my fo o tha, where (n., ym.) = (Cos 8, sin 8, 0) are the components of n. This result can be extended to the general three-dimensional stress state, as will be seen later. Equilibrium of the element shown in Fig. 1-6 leads to another important property of the stress, namely, its transformation due to a rotation of the coor- dinate axes. To derive this property, let the system (Ox'y'z') be obtained from the system (Oxyz) by a rotation about the z-axis through an angle @. The com- ponents of stress relative to the system (Ox’y'2") are denoted by on, 25 tc. (Fig. 1-7). Summing forces along the x’-, y'-, and 2'-directions and taking the limit as OP —> 0, we obtain the relations a, = autos? 0 + oysin? 8 + 2orysin 8 cos 6 o% = (yy — oudsin 8 cos @ + an(cos*@ — sin?) on=00 ‘A Review ofthe Elementary Theory of Elasticty Chap. 1 Figure 1-7 Transformation of sess due to rotation of xy-axes We can similarly show that 0, = asin? + 0,C08*O — 2aysin 8 cos 6 o, = 01, =0 ‘These equations can be expressed in the matrix form [z: 29] [ cos 8 eal le es | [ee sin 4 (ts) oy of] [-sin@ cos 6} Loy oJ Lsin@ cos 6 ‘The transformation property (1.15) can be used to calculate the stress compo- nents on an element oriented in an arbitrary direction. The transformation equation for the three-dimensional stress state wall be obtained later. 1.4 HOOKE’S LAW Up to this point, we have considered the mathematical description of strain, stress, and the equations of equilibrium of a solid without paying any attention ‘6 to the specific nature of the material. In order to complete the mathematical formulation, it is necessary to introduce additional equations that relate the stress and strain components in the solid under consideration. We shall now derive these equations forthe special case of a linear elastic material. By elas- tic we mean thatthe material will return to is original shape and size upon the Sec. 1.4 Hooke's Law " removal of the deforming loads. By linear we mean that the stress-strain rela- tionship is linear. The concept of linear elasticity was first introduced by Robert Hooke in 1678 when he postulated the law wt tensio sic vis, which can be translated as “the extension is proportional to the force.” A typical observed stress-strain Jaw for a material subjected to uniaxial stress is sketched in Fig. 1-8. The law is linear and elastic up to a stress called the proportional limit. The behavior is nonlinear but stil elastic up to a stress called the yieid stress. The material un- dergoes permanent deformation if loaded beyond the yield point; i.e, a certain amount of residual strain remains after the load is removed completely. The terms ultimate stress and rupture stress ate self-explanatory. Thus Hooke’s law holds below the proportional limit. RUPTURE STRESS ULTIMATE STRESS. PROPORTIONAL LIMIT e Figure 1-8 Typical observed stress-strain curve for materials under uniaxial tension, For three-dimensional bodies, we do not have just extension but, instead, require six components of strain to describe the defomation at a point. Simi larly, the complete specification of the internal force operating at a point re- quires the use of six components of stress. Cauchy and others developed the relationship between the components of stress and strain that is known as gen- eralized Hooke's Law. According to this law, each component of stress at any point of a solid is a linear and homogeneous tuncticn of the components of strain at that point. Ifthe elastic properties of a solid at a given point are the same in every direction, it is called elastically isotropic. A medium that is not isotropic is called acolotropic, ot anisotropic. If the material property is the same at all2 ‘A Review of the Elementary Theory of Elasticty Chap. 1 points within the solid, it is called homogeneous. In this chapter, we shall confine our discussions to isotropic and homogeneous solids only. Consider a unit cube of a material with its edges parallel to the coordinate ‘axes and subjected to the action of the normal stress o;; uniformly distributed ‘over opposite faces perpendicular to the x-axis. Then, according to Hooke's Taw the extension of the element is given by in which E is a material constant known as Young's modulus. Clearly, from physical considerations, E > 0 and the extension of the element in the _xdirection is accompanied by lateral contraction resulting in normal strains in the y- an z-directions. It is reasonable to assume that these strains are propor- tional to ef: 0 =e wy = -, eff) = ell! = ~vel? Ou E The material constant is called the Poisson's ratio of the solid. Similarly, corresponding to the normal stress a3,, we have B 2 = Lo, ef P= fom, ‘and corresponding to the normal stress 0, we have eee ee eB —Fom ef ‘Since the stress-strain relationship is linear, the principle of superposition can be applied. Thus, if the element of volume is subjected simultaneously to the action of normal stresses Ou, Jy, and o., uniformly distributed over the faces of the cube, the resultant components of strain can be obtained from the pre- ceding as, = Om = a ~ Voy + 04) (1.168) 67 = PlOy — von + 01) (1.166) flo ~ (ou + oN) (1.166) If shearing stresses oe, uy and dy act on all the sides of the cube, the change in the angle between any two of its intersecting faces depends only on the corresponding shearing-stress component. We have seen in Sec. 1.1 that Sec. 1.4 Hooke's Law 3 2ey is equal to the decrease in the right angle between two line elements that ‘were parallel to the x- and y-axes, respectively, before deformation. We can therefore assume that (1.164) ‘The material constant jis called the modulus of rigidity, oF shear modulus, and from physical considerations, w. > 0. The extensions given by (I.16a, b, ¢) and the distorions given by (1.164) are independent of each other. The general strains, produced by simul- taneous application of the three normal and three shearing components of stress, can be obtained by superposition. Sometimes it is desirable to express the stress components in ferms of the strain components. This is easly achieved through inversion of (1.16), and the resulls ean be expressed inthe form a E Oa = M+ Een (1.17) _ E Oy = M+ Tey (1.17%) on = Ad ace (1.17%) Oy = Ploy a = ews a = Hew (HIT) where tnt byt ee (1.17) and ve 1.184 aI a ‘The three material constants E, jl, nd » are related by the equation E = 1.18 oH (ise) ‘To show this we consider the state of stress represented by the components x = Oy = 0 From (1.16), the strains are given by Lte Lte ea EO ey = y= 0 (119)“ ‘A Review ofthe Ele tary Theory of Elasticity Chap. 1 Let the system of axes (Ox'y’2') be obtained from the system (Oxyz) by a rotation about O: through 45°. Then using transformation relations (1.15), ‘we find that the components of stress in the primed system are Similarly, from (1.19), the components of strain in the primed system are + ye ee = Oy = eh =e =e =O Since the medium under consideration is isotropic, we can use relations (1.16) even in the primed system. Thus, the preceding equations give 2ey = ao Ley ae 20, = 2-0) which leads to the relation (1.186). Hence, of the three constants introduced before, only two are independent, i.e., an isotropic material is characterized by nvo independent material constants. With the help of (1.18b), we may write (1.17a-d) in the form Ou = Ad + Yl Oy = Ad + Ihe, Ty = Ad + Yen (1.202) Hey (1.206) Oy, = phe yey Ox = 2plny Fm = ‘The constants A and x are known as Lamé constants. From (1.18a, b), we can express E, v in terms of A, pas 2x) wGA+2) A aa 2a +h) (28) Inthe case of a uniform hydrostatic pressure of magnitude p, we have Oy = Oe = =P Ou + Oy + On = ~3P Equations (1.16a, b, ¢) and (1.17e) yield (1.210) Sec. 1.4 Hooke’s Law 6 where (1.216) ‘The material constant & is known as bulk modulus, or incompressibilty. Since a hydrostatic pressure leads to a decrease in volu have k > 0. Equa- tions (1.18) and (1.2le), in conjunction with the conditions E > 0, # > 0, ‘and k > 0, imply that 1 < » <4, However, negative values of the Pois- son's ratio » are uncommon in reality. For an ideal fluid, sz = 0 and k(= A) is finie. For an incompressible solid, both k and A are infinite, but jis finite. Tt can be easily seen that for both of these special cases, » = f ‘Any two of the five constants A, 1, E, v, and kmay be used to character- ize a given isotropic material, and the remaining three constants can be ex- pressed in terms of these two. Some of these relations are given in Table 1-1. Equations (1,6), (1.10), and (1.16) or (1.20) form the basis forthe solu- tion of linear elastic problems. There are a total of 15 equations in all, and the unknowns are six components of stress, six components of strain, and three components ofthe displacement. Ifthe stress and stain components are elimi- nated from these equations, three coupled partial differential equations for the three displacement components, called Navier's equations, are obtained. These ‘equations can, in principle, be solved for the displacement components when the boundary conditions are prescribed. The strain end stress components can then be determined by means of equations (1.6) and (1.20), However, it should be noted that the partial derivatives that appear in the strain-displace- ‘ment relations (1.6) are with respect to X, Y, Z, the inital coordinates of a ma- terial particle ofthe solid in the undeformed state. On the other hand, the equa- tions of equilibrium derived in Sec. 1.3, are for an element of the solid in its deformed state, so that the derivatives in (1.10) are with respect to the current coordinates x, y, 2. It will be shown later that in @ consistent linear theory in which all nonlinear effects are ignored, no distinction needs to be made be- tween the coordinates (X, ¥, Z) and (x, ys 2). In many elementary engineering applications, the primary quantities of interest are the stress components produced in a solid under given boundary loads only. In such cases, the body force terms in th: equations of equilibrium (1-10) are absent. I is often possible to construct solitons of the resulting ho- mogeneous system of equations to obtain stress states, which can then be sub- jected to the appropriate boundary conditions. A few simple examples of this approach are presented next. Tn the first example we consider @ beam of arbitrary cross section sub- jected to uniform tensile load a applied to its end faces shown in Fig. 1-9. ‘The objective is to determine the stress produced within the beam.By KE ‘TABLE 1-1 Interrelations Among the Elastic Constants of an Isotropic Solid ws Basi Pair Constant Sec. 1.4 Hooke's Law ” Y v : a oo —> x Ee dd z | ~ 4 w]e Figure 1-9 A beam under uniaxial loud. E In order to construct a solution of the problem, we locate a coordinate system with x-axis parallel to the beam (Fig. 1-9). Let the normal vector to the boundary of the cross section be (0, n,m). Then by (1.14) the components of . the traction or stress vector ata point on the curved boundary of the beam, f ‘yy ty ae related to the stress components atthe point through the equations t= om + ous (4.229) 3k = 20) 2») ME 5 3G. i by = Gyn, + Ome (1.226) « [ale 1 = Gun, + oun, (1.226) Since there is no applied load on the curved boundary, the components of the stress vector must vanish on it, Further, the boundary conditions at both end sections are easily seen to be «| 4| Shy Ou = (1.23a) ‘The boundary conditions as well as the equilibrium equations in the absence of body forces are clearly satisfied by the stress field Ou = Oy Oy = On = Sy = p= Fu =0 (1.230) AU + a — 2») The strain components within the beam can be easily calculated from (1.16) and (1.23) as an Z ate (1.24) Thus, (1,23) and (1.24) give a possible solution forthe stress and strain components within the block. However, not all stress states that satisfy the a equilibrium equations and boundary conditions are physically possible. Some | ‘of these solutions may lead to multiple-valued displacements within the mate- rial, leading to creation of cracks and other features that are absent in the origi- nal problem. The stress state given by (1.23) happens tobe the correct solution ofthe present problem. A detailed discussion of this and related issues will be given later. mw GA + 2)8 A Review ofthe Elementary Theory of Elasticity Chap. 1 Y 4 M. 2 ox igure 1-10 beam subjected to end moment. AAs another example of the elementary solution, assume that the beam of the previous problem is subjected to an end moment of magnitude M, about the z-axis, as shown in Fig. 1-10. The boundary tractions on the curved surface are still given by (1.22). and these must vanish. ‘The conditions at the end sections are specified not in terms of the dis- tributed stress vector, but their resultant force F and moment M are prescribed [ona as f oy dA = F,=0 (1.25b) fo dA=F.=0 (1.25¢) | (you — 20) dA = M, = 0 (1.254) [ (20, — x0.) dA = M, = 0 (1.25e) [ (x0 ~ you) dA = M, (1.25f), Ou = OY, Oy = On = Oy = Oye =e =O (1.26) where ¢ is a constant. Clearly, the stress components satisfy the equilibrium equations (1.10) as well as the boundary conditions on the curved surface of the beam. The end conditions (1.25) are satisfied if the x-axis passes through See.1.4 Hooke's Law 1° the centroid of the cross section of the beam, the y- and z-axes are along the principal axes of inertia of the cross section of the beam and (1.278) where fr dA (1.276) Which isthe area moment of inertia of the cross section of the beam about the Equations (1.26) and (1.27) give the stresses within the beam, and the strains can be obtained from (1.16). It should be noted that the distribution of, the stress vector at the end section must be prescribed as a linear function of y in order that the solution obtained above be exact. Ifthe distribution is differ- ent, then the solution is approximate and is accurate only at some distance ‘way from the ends of the beam. {As the final example of elementary solutions, we consider the stresses de- veloped in uniform circular cylinder of radius a subjected to an axial torque ‘of magnitude M, (Fig. 1-11). If (x, y, 2) denote the coordinates of a point on the curved surface of the cylinder, then the normal vector has the components (0, y/a, 2/a). Thus by (1.22), the boundary conditions on this surface are Wa + 20, = 0 (1.282) Vy + 2. = 0 (1.286) Oy, + 207 = 0 (1.280) + ’ Figure 1-11 Circular cylinder under axial torque.20 ‘ARoview ofthe Elomentary Theory ofElatity Chap. 1 ‘At the end sections, the stress vector is subjected to the conditions (1.25a, b, c) and f (0 ~ 205) dA = My (1.298) | (20m — x0) dA = 0 (1.296) f (20% ~ you) dA = 0 (1.290) Assume that the stress state within the cylinder is of the form Ou Oy = 0n= On =O On = CY, y= cz (1.30) where c is a constant. The equilibrium equations (1.10) without body forces as, well as the boundary conditions (1.28) are identically satisfied by these stresses. The conditions (1.25a, b, c) and (1.29a, b, c) are satisfied if the x- axis passes through the center of the cross section and if cf or +a Thus, n= [ore )dA (3) where /, is the polar area moment of inertia of the cross section and can be eas- ily shown to be equal to 7a*/2. It should again be pointed out that the preced- ing solution is exact if the stress vectors at the end sections are prescribed in accordance with (1.30). Otherwise it is valid at some distance away from the ends. 1.8 THERMAL EFFECTS ‘Temperature changes may be a source of significant stresses in elastic solids. It js reasonable to assume that a linear relationship exists between the tempera- ture difference and the normal strains caused by it and that thermal strains do not involve shearing of the material, Thus, if @s denotes the uniform tempera- ture of the body in its reference state and © the temperature in the current de- formed state, we may write a = by = bn = OT (1.32) where T(x, ¥,2) = O- O Sec. Problems a and aris the coefficient of thermal expansion, assumed to be a constant, With this concept, itis easy to generalize (1.16) to the form [ou — (Gy + 02) + aT Loy — v(Gu + Ox) + oT (1.33) Co = Lon ~ vou + oy) + a Similarly, the first three relations in (1.20) become Ou = Ad + 2pteu~ 30kT Oy = Ad + 2pteyy ~ 3akT (134) On = Ad + 2pen ~ 3akT Equations (1,33) or (1.34), together with the strin-displacement rela tions (1:6) and the equilibrium equations (1-10), form the basis forthe solution of linear thermoelastic problems for given temperature field and boundary con- ditions. It should again be noted that the independent variables in (1.6) and ((.10) are not the same, and a consistent theory needs to be devetoped before the solution of general problems of thermoclasticity can be attempted PROBLEMS L.A. Given the displacement field uaaKy, Va v4 6xZ, Wm 62) +297 calculate the strain components at the point (1, 0, 2). What is the extension of a line element dX (parallel to the x-axis) at tis point? [Ans. ce = 24, ey = 5, ey = Hy eur = by = ee = O 1.2, Show that equal extension and contraction of two orthogonal linear elements is equivalent to a shearing strain of equal magnitude, which is associated with direc- tions bisecting the angles between the elements. 4.3. The stress components at a point P are given by Cu = 2, Oy = Mn On = Or oy = 0, Byun Uy Oe where ois a constant, Determine the stress st the point P on a plane with normal in the direction (2, 2, 1). Also determine the normal and shearing tresses atthe point P on this pane [Ans.: te = Qos ty = 200s tr = Gow N = 30, 5 = (V3/3)0%}2 ‘A Roview of tho Elemontary Theory of Elasticity Chap. 1 1.4, ‘The state of stess ala point is given as follows: OA Oy =Hl Ona T O20, O4= 0, Oat ‘Show that the normal component of the stress vector on a plane with normal in the direction ', 1, 2) has magnitude 1. Show also that the shear stress has mag- nitude 1/3 and acts in the direction (1, ~1, 0). 1.5. Determine the body forces for which the following stress field describes a state of equilibrium: Oy Det — By? — Oy = 247 oan ax tyt3e-$ 0 y= “3x4 2y 41 Oy = t+ day ~ 6 [Ans.: f= f= f= 01 1.6. Determine whether the following stress feld is admissible in an elastic body when the body forces are negligible: 2ayz 24 dy cutyetd Oy ary, on = 80, ou Sy, [ans.: No] 1.7. ‘The state of stess ata point is given as follows: w= 9? + ex? = y) ay ett c(t 2) aaety? On = 042% oy = flesy) Determine f(x,y) 50 thatthe stess distribution may be in equilibrium in the a sence of body forces. [ans.: fle, 9} = ~2ex9] 1.8. Show that equel tension and compression across two orthogonal planes is equiva Tent to a shear stress of equal magnitude across a plane that bisects the angle be- tween the two given planes. BIBLIOGRAPHY “TiwosHtenKo, S. P. and Goooier, J. N., Theory of Elasticity, MoGraw-Hill, New York, 1951 es Cartesian Tensors 2.1. ALGEBRAIC DEFINITION OF A VECTOR; THE TRANSFORMATION MATRIX Many physical quant .. mass, volume, and temperature of @ solid body-—can be specified by using their numerical values in appropriate units. ‘These are called scalars. Others—e.g., displacement, velocity, and linear ac~ celeration of a moving particle or the force acting on an object—require a mag- nitude as well as a direction for their description. Taese quantities are called vectors. A vector by its very nature can be represented by a directed line sez rent of an arrow in space whose length is proportional to the magnitude of the vector and whose direction and sense coincide with those of the vector. This is the geometrical representation of a vector, and togecher with the geometrical niles of operation between vectors (e.g. the parallelogram law of addition, the cosine law for the dot product, and the sine law for the cross product) form the basis of the vectorial treatment of elementary mechanics. In more advanced treatments it is very convenient to give an algebraic detintion of a vector and to develop algebraic rules of operation between vectors. If we associate the symbol x (in direct notation) with a given directed line segment, then the definition of a vector implies that any other directed line segment of the same length and sense is also x. For example, two forces with 2* Cartesian Tensors Chap.2 tude and direction but acting at two different points of a rigid body are represented by the same vector, although the forces may have differ- ent influerce on the equilibrium or motion of the body. We introduce a right-handed Cartesian coordinate system with axes num- bered as 1,2, and 3 at an arbitrarily chosen origin O (Fig. 2-1) and assume that the dre tine regent OF he veto Xe an ee Po of OP on the coordinate axes numbered 1, 2, and 3, respectively, or ‘Mternatively, the coordinates of P relative to the chosen axes, Then, fora given vecior OP, the numbers x1, 12, x5 can be uniquely determined. Con- versely, ifthe numbers xi, x2. 33 are given, then the vector OP can be uniquely found. In other words, with a given choice of the coordinate system, the or- dered triplet x1, x, a5 8 completely equivalent to the vector OP. The order in which the numbers are given is important, since, in general, the triplet x, 2, 2 is a vector different from OP. The triplet x1, 2, x5 will be denoted by [x] ‘when written as a row matrix and by {x} as a column matrix. Figure 2-1. The coordinate system, Let e,, 2, €s be three vectors of unit magnitude directed along the coord- nate axes (Fig. 2-1). In analogy with the notations introduced in the appendix, the array of these vectors will be denoted by [eas a row and {e} as a column. ‘They define the reference coordinate system for describing all vectors that lie in a three-timensional Euclidean space and are called a vector basis; the unit n ofa Vector; The Transformation Matrix Ea Sec.2.1 Algebraic Defi ‘vectors €1, €2, € are called the base vectors. Then the vector x can be uniquely ‘expressed in terms of the base vectors and its components x1, x2, x3 along the base vectors in the form x= ne, tet t= Se en ‘An obvious consequence of the representation (2.1) is that the sum, dif. ference, and (dot or cross) products between two or mote vectors can be ex- pressed in terms of their components. Thus, if u and v are two vectors with Components ty ws, hs ad Pry Ba, Bs, FEspectvely, in the same vector basis (e}. then 2) 23) XV = (0s — werer + (Wer — Moses + (ties ~ HAO} eee ym os oo 8 where we have used the relations seed, axe=0 (2.58) with i and eX e=-eXe=e exes -eXe=e ) eX eae kee In (2.50), 8) is the Kronecker delta defined via _ ft. itis i - bel ie) es ‘The relations (2.5) can be easily verified from the geometrical definition of the dot and cross products, the orthogonality, and the righthandedness of the vec~ tor basis {e}.28 Cortesian Tensors Chap. 2 2.1.1 The Vector Transformation Rule We first note that if w= ¥ = x in (2.3), we obtain the well-known for- ‘mula for the “length, or magnitude, |x| of the vector x, IxPoat+ ded (2.7) It should be noted that the vector OP itself is not dependent upon the choice of the vector basis and that the triplet {x} is simply the algebraic representation of OP in the chosen vector basis {e}. If we choose another vector basis {e/} at the common origin O (Fig. 2-2), then the same vector OP is represented by different triplet {x’} in the new vector basis. Thus, x Dae! Dae (2.8) Figure 22. Rotation of axes Taking the dot product of (2.8) successively with e}, €3, eS and using the orthogonality property of the vector basis {e!} (ie., ef - €/ = 6;), the follow- ing results are obtained; X= Guay + dan + lox XE = Gay + ban + bows 29) 24 = Gin + bata + boas where Gyreire, f= 12,3 (2.10) Similarly, by taking the dot product of (2.8) successively with e,, e3, @, we obiain S0e.2.1 Algobraie Definition of a Vector; The Transformation Matrix a uxt + lad + nd xt + load + Card uy ” ay = Cpxi + lasxh + atl Equations (2.9) and (2.11) may be written in matrix form as {xl} = L(x} or X" = LX (2.128) and {x} = Lui} or X = LIX" (2.126) where fn ba be L=(Gl=| er & bs 2.13) fr ba bas Equations (2.9) and (2.11)—or, alternatively, the matrix equations (2.12)—give the transformation properties of the components of x in the (wo vector bases {e} and {e/} through the linear transformation matrix L. It can be easily seen from (2.9) and (2.11) that the column vectors of L are the direction cosines of the base vectors e, €2, and e) relative to the vector basis {e!} and the row vectors of L are the direction cosines of e}, e3, and e5 relative to the vec tor basis {e). For this reason L is also called the direction cosine matrix of the vector basis transformation from {e} to {e/}. Since the unit vectors e:, €2, and ¢ are mutually orthogonal, the sum of the squares of the elements of each column (or row) of L is unity and the sum of the products of the corresponding elements of any (wo columns (or rows) is zero. Thus L is an orthogonal matrix [see the definition in (A. 15)] and, therefore, Lt sir 2.14) in agreement with (2.12) and [epat (2.15) Clearly, ¢11, 2, €1 are the components of the unit vector ein the vector basis {e} and similarly for €1, fz, €:» and €51, €, &y. These results can be ex- pressed in matrix form as fei} = Lie} (2.16) ‘Then, from (2.4), le ee eexes= [lr ba ba br bs bs] and ef (es x es) = [L] 2.17)a Cartesian Tensors Chap. 2 But, since the vector basis {e/} is right-handed, eX eh = et It follows from (2.17) that [L|= +1 (2.18) ‘This result is stronger than (2.15) and is a consequence ofthe fat that both vector bases are right-handed. It can be shown that the system {e/} can be ob- tained by a rigid rotation of {e;} and vice versa. If | | +1, the transforma- tion is called a proper rotation, whereas if | L | = ~1, it is improper, i.e., one of the vector bases isnot ight handed "The preceding diseussons imply tha «vector x in a three-dimensional ucidcar space ean be defined as an ordered array of numbers {sy in} when referred to a (right-handed and orthogonal) vector basis (€1, €2, €3). If a new vector basis (e/, e3, €5) is introduced such that the base vectors of the two thse are linearly related through the equation {ei} = Lie} (2.19) where L is an orthogonal matrix with the property [L]=41 then the same vector x is represented by the array {xi, x, xS}, where {xl} = Lf} or X" = LX (2.20) It should be noted that the preceding definition also holds for @ vector in m di ‘mensions, In that case (x) and {xi} are (n x 1) arrays, each of the vector bases {e} and {e/} has n mutually orthogonal base vectors, and L is a matrix with the stated preperties. We shall be concerned with two- and three-dimensional vec- tors only, ‘Ancther major advantage of the algebraic definition of a vector is that it can be used to define and manipulate other physical entities called tensors, which cannot be represented by directed line segments. Before we introduce the concept of tensors, itis necessary to develop certain notations and conven- tions for ease and compactness in the manipulation of tensors. 2.2 THE INDEX NOTATION In the mathematical formulation of the theory of elasticity, it is often necessary to work with differential or algebraic expressions and equations containing a large number of similar terms. Itis possible to reduce them to more convenient compact forms through the introduction of two simple conventions. The first of these is the range convention, which stipulates that every subscript (or index) ‘Ses.2.2 The Index Notation 2 takes on the values 1, 2, 3 unless otherwise stated. Thus, the symbol x; stands for the collection of the three numbers x1, 42, %» and the symbol ay stands for the collection of the nine numbers ay, azz, - . , sy. Similarly, the equation X= ys represents three separate equations corresponding to i = 1, 2, 3. The indices 7 and j in these examples are called jree indices. The same free indices ‘must appear on both sides of an equation. ication is achieved by means of the summation conven- tion. According to this convention, if an index i appears twice in a given ex- pression, a summation over { must be performed by assigning the values 1, 2, 3 to i successively. Any exception to this rule must be stated explicitly. A re~ peated index is called a dummy index. A dummy index cannot occur more than twice in an expression, It can be seen that many of the equations that appear in the previous sec- tion can be written in compact forms through the use of these conventions. For example, (2.8), (2.9), and (2.11) may be rewritten as x= xfel = xe, = bux |As another example, a homogeneous quadratic function f(x1, x2, x3) may be \written as ayx.xj, where [ay] is a constant matrix. Thus, f(r, Xo, x) has the al- temative expressions Ft, 42,8) = ay3i8) auxt + anad + anxt + (as + an)mxs (2.21) 4 (as. + ausdaias + (aie + ay) mee 2.2.1 The Kronecker Delta and the Alternating Symbol ‘The Kronecker delta, 5, introduced in (2.6) plays an important role in the manipulation of equations written in index notation, A particularly useful property of 6; is illustrated by the results 2b) = aby = Oe (2.22) “Thus, the presence of the dummy index i and 6; in (2.22) results in the replace- rent of { by j in x, and ay. This is known as the substitution property of the Kronecker delta. ‘The orthogonality ofthe vector basis {e) is expressed by the equation e= hy (2.23) and the dot product between any two vectors u and v is given by wey = uty (2.24)0 Carte snTensors Chep.2 On the other hand, the right-handedness of the vector basis {e} is reflected in the cross product rule between the base vectors given in (2.6) and between any ‘wo vectors given in (2.4). In order to express these results in index notation, it is necessary to introduce the alternating (or permutation or Levi-Civita) sym- bol ex, defined via en =H DU WR - 9 Thus, 1 if jk are even permutations of 123, ie., are in cyclic order x= 4-1 if jk are odd permutations of 123, i.e., are in antieyclic order 0 otherwise, i.e., if two or three indices are equal ‘Thus, the only nonzero elements of és are given by €m=en=6n= 1, €m= én zens —1 (2.25) ‘The following results can be easily verified: 0X 6 = ents (2.260) (eX @) oe: = &e (2.26b) WX = EQUiDjes (2.26c) It should be roted that equations written in the index notation as intro- duced here are valid in Cartesian coordinates only, in contrast to those written in the direct vector notation, which are independent of the coordinate system. Example 2.2-1 Show that (2) enue = By Be ~ udp (©) eeyn = 25, (©) nen = 6 Solution. To prove (a), we note that the quantity éwveny- (where a primed index ‘implies that no sum isto be carried out over that index) is nonzero if and only if, j,k” are all different and p, q, &' ate all different. Therefore, either = p and j = q or i= q and j= p. In the first case, ey = eye and, therefore, eeu = 1- In the second case, ey = eye because the interchange of two indices changes the sign of, the value of eye. Consequently, éyréua: = —1. Now, if i,j py and q are fixed and k runs through the values 1, 2, 3, then, as explained before, only one term in the sum is nonzero. Hence, the whole sum is either +1 or —1, This can be written a8 By6j, ~ B45p because this expression equals +1 for i= p, j= q, i * j and ‘equals “1 for = gy! = pyi # jh if neither isthe case, it quale 0, Putting @ = jin (a), we get tue = 5p5y ~ 8yBp = 38y ~ by = 2p Again, putting p = in (b), we find that nen = 25s = 6 SSec,2.2 The Indox Notation 3 Example 2.2-2 Show that be By Br eater = [By By Be 0 By By Bu Solution. Consider the determinant bn Ba ba By Ba Bal = enby5ydu Ba Ba Ba ‘Therefore, Bn 82 a] [Bn 2 Bal ent = [Br b2 Ba] /8s 32 Bo Jou Bz by||84 82 Ba bu Be Ba] [bn By Ba =i: 8 5s[[on ba da] bu Be Bu||89 89 Ba (interchanging the rows and columns in the second determinant) bo By Br bp By Be By Buy bu “The last step follows immediately on noting that, for examgle, the (1, 1) element of the prodvet determinant will be B18p1 + B2dp2 + Bad ,s = Brby = By Putting r = kin i) and expanding the determinant, we get result (a) of Example 2.2-1 Example 2.2-3 Show that (2) detlas} = eindadaaa = eancryass () eadetfay Solution (2) Note that for three vectors u, v, w, (x Yew ean @ In @) let w note that ‘4p, i = ais, proving the first result in (a). Similarly, letting ‘uy gives the second result in (a). To prove the third result, we ay,a Cartesian Tenors Chep.2 e12€imisdan i + €21€imirAindn = Emddandse ~ Edandye = 2 deta] from the second result in (a). Writing two more relations of this type and adding all three, we get the desired result (b) There sre three possibilities: f, m, n a cyclic permutation of 1 23, l,m, man anti- cyclic permutation of I 2.3, and atleast two of f,m, m equal. In the first case, caAQudin dy = Edn dntin = detlag) = Comdetlay] the first ste follows from the property that any eyclic interchange of the columns of a determinant leaves its value unchanged and the second step from (a). Similarly, in the second case, EAQ4dindin = Eanes ~det{ay] = eredetlag] In te third case, delay] since if any two columns of a determinant are identical, its value is zero and ee = Oi two of fm, m ate equal. This proves the result. e442 2.3 GENERAL DEFINITION AND ALGEBRA (OF CARTESIAN TENSORS Cartesian teisors are defined through a generalization of the algebraic definition ofa vector discussed in Sec. 2.1. Let a physical entity be defined in ‘a three-dimensional Euclidean space such that for a given choice of a vector basis {e} it is possible to specify the entity completely in terms of an ordered set of 3" numbers denoted by t,.. containing a total of indices. The collec- tion of these numbers will be denoted by (the boldface capital) T or by ¢ if there is no confusion with the direct notation for a vector, and the numbers yx... will be called the components of T. Let a new vector basis {e/} be intro- diced through a proper rotation of {e} described by means of the direction cosine matrix L = [€,]. Ifthe components of Tin the new vector basis become fin, Where i = blir foe (2.27) then T is called a Cartesian tensor of order ‘An important Implication of tie definition is that the components of a tensor need be specified only in any one conveniently chosen vector basis, since its components in any other vector basis can be obtained from the trans- formation law (2.27). Furthermore, an ordered set of numbers does not by it- self constitute a tensor unless the ordering is related to a vector basis and a proper rotation of the vector basis results in the transformation of the numbers according to (2.27). Thus the array of the three single-subscripted numbers a; Sec,2.3 General Definition and Algebra of Canesian Tensors 3 ‘and the array ofthe nine double-subscripteé numbers ay form matrices, and the rules of matrix algebra apply to these arrays. But they do not necessarily repre- sent tensors. In general, subscripted quantities that appear in the mathematical description of physical laws are tensors, i.., they also obey the transformation law (2.27). Since a vector can be specified by an array of three single-subseripted numbers in any vector basis and since these numbers transform according to 2.27) with n = 1, a vector is also a tensor of order one, Thus the symbols v in the direct notation or V in the tensor notation can be used interchangeably to denote a vector. A scalar is a tensor of orcer zero. A collection of 3° zeros is called the null tensor of order n and will be denoted by 0. Examples of mathe- matically constructed higher-order tensors will be given in subsequent discus- sions. Two important second-order tensors, called strain and stress tensors, arise in the mathematical description of the elastic behavior of solids. ‘These are discussed in the next two chapters. ‘We now discuss some useful rules for the algebraic manipulation of ten- sors. In what follows the symbols S and T will denote tensors of order n (= 1) with components sy... and fy... respectively, in the vector basis {e)} and sj and 1}, in the basis {e/} introduced earlier. Then, in accordance with (2.27), Ce ee OE) th = Calin ta It can be easily seen that the multiplication of a tensor T by a scalar a produces another tensor aT of the same order with components ary... since axth,.. = Cabin atin ‘The sum (or difference) of $ and TT is another tensor denoted by $ + T (or S — T) with components 6)... + f.. (OFS... ~ fy.)s Since from (2.28) Sho Eth = abn (in E th) ‘The two tensors $ and T are called equal if their difference is the null tensor, e., if $i. 0 2.3.1 Inner and Outer Products ‘The multiplication between two (or more) tensors can be classified into two categories called outer and inner products. In order to illustrate the con- struction and the properties of these produc:s, we assume that S and Tare ten- sors of second order and U and V are vectors (or tensors of first order), Then, the components of U and V in the two vector bases {e} and (e/} are related by the equations ul = ata, fav, (2.29)Fn Cartesian Tensors Chap. 2 whereas those of § and T are related by Cabsion — Uy = Ex Eatin (2.30) Consider the set of nine numbers a, defined by the product 10, 31) ‘We would like to examine whether the collection of these numbers is a tensor. Since wand v; are the components of U and V in the vector basis {e}, the mu- ‘merical values as well as the ordering of the elements of the array {aj} are clearly dependent cn the choice of the vector basis. Furthermore, by changing the free index in the second equation of (2.29) from i to jand the dummy index from k to m, it can be writen in the form 0) = bint so that Cab tts (2.32) Hence, if we use the symbol aj for the transformed quantities u/vj, then (2.32) becomes 4h = €abindon which is the transformation law of a second-order tensor A. Thus, the product tu, defines a second-order tensor. It can be shown in a similar manner that the set of 27 numbers dy defined by the product bye = tye (2.33) represents a third-order tensor, and the set of 81 numbers defined by the product Ci = Shin (2.34) represents a tensor of the fourth order. The products defined in (2.31), (2.33), and (2.34) are called ower products between the tensors involved, and they can be extended to higher-order tensors. An outer product between two tensors always generates a tensor whose order is the sum of the orders of the con~ sfituent tensors. Inner producis between two tensors are of the following types: Oy Py da = Sly (2.35) ‘These products are characterized by the presence of one or more dummy suffixes, It can be easily shown that e is a fensor of order U or a scalar, p, is a component of a tensor of order | or a vector, and ga is a component of a tensor of order 2, Since the total number of indices in the left-hand sides of (2.35) is reduced from that of the right-hand sides, the order of the resulting tensors is always less in inner products than in the outer products. Sec,2.3 General Definition and Algebra of Cartesian Tensors 8 It can be seen that the inner products defined in (2.35) are related to the ‘outer products (2.31), (2.33), and (2.34) through the equations in Pi bw da cy 2.36) ‘The quantities in the right-hand sides of (2.36) are obtained by putting two of the indices in the components of the corresponding tensors equal and then sum- ‘ming over this index according to the summation convention. This operation is called a contraction, and it results in the reduction of the order of the tensor. Thus the scalar a is obtained by contracting the second-order tensor A, the vector P is obtained by contracting the third-order tensor B, and the second- ‘order tensor Q is obtained by contracting the fourth-order tensor C. In general, the contraction of a tensor of order n(= 2) results in a tensor of order n ~ 2. 2.3.2 The Quotiont Rule [As another application of the inner product, consider an array (ay). Let U be an arbitrary vector with components w in the vector basis {e} and let ut) @.37) fv), b2, bs are the components of another vector V in the vector basis (e}, then ay ae the components of a second-order tensor A. To prove ths result we note that in the vector basis {e/}, the elements of the array [ay] become [aj], where iu (2.38) Furthermore, since U and V are vectors, their components transform according to ot = fave (2.39) 4 = bat (2.40) ‘Also, (2.37) may be rewritten as m= ayy 241) Then, by (2.39.41), Of = Cxaybastn = Fudan Ct) and by (2.38) (ai, = Cabanas = 0 (2.42) But, since U is an arbitrary vector, (2.42) must holé for This gives