Baixe o app para aproveitar ainda mais
Prévia do material em texto
Sanichiro Yoshida Deformation and Fracture of Solid-State Materials Field Theoretical Approach and Engineering Applications Deformation and Fracture of Solid-State Materials Sanichiro Yoshida Deformation and Fracture of Solid-State Materials Field Theoretical Approach and Engineering Applications 123 Sanichiro Yoshida Department of Chemistry and Physics Southeastern Louisiana University Hammond, LA, USA ISBN 978-1-4939-2097-6 ISBN 978-1-4939-2098-3 (eBook) DOI 10.1007/978-1-4939-2098-3 Springer New York Heidelberg Dordrecht London Library of Congress Control Number: 2014949877 © Springer Science+Business Media New York 2015 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. Exempted from this legal reservation are brief excerpts in connection with reviews or scholarly analysis or material supplied specifically for the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work. Duplication of this publication or parts thereof is permitted only under the provisions of the Copyright Law of the Publisher’s location, in its current version, and permission for use must always be obtained from Springer. Permissions for use may be obtained through RightsLink at the Copyright Clearance Center. Violations are liable to prosecution under the respective Copyright Law. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. While the advice and information in this book are believed to be true and accurate at the date of publication, neither the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be made. The publisher makes no warranty, express or implied, with respect to the material contained herein. Printed on acid-free paper Springer is part of Springer Science+Business Media (www.springer.com) Preface This book stems from my first acquaintance with Academician Victor E. Panin of the Soviet Academy of Sciences,1 and my daily research log that I have been keeping since then. At the end of 1990, I had the opportunity to listen to a lecture given by Acad. Panin at a meeting held in Tokyo, Japan. In this meeting, a number of delegates from the Soviet Academy of Sciences gave presentations to Japanese business people. The Acad. Panin’s lecture was on a new theory of plastic deformation that he called physical mesomechanics. Although I understood only 20 %, or probably less, of his lecture, I was greatly fascinated by his enthusiastic presentation and by the physical-mesomechanical view of plastic deformation. (He was among the two or three presenters who gave the talk in English without an interpreter.) In particular, the Maxwell-type field equations that describe plastic deformation dynamics interested me greatly. At that time, my field of research was laser and spectroscopy, and I was using the Maxwell equations of electrodynamics on a daily basis. Although I did not understand the Maxwell-type field equations of physical mesomechanics in any depth necessary to comprehend the deformation dynamics behind them, I was able to understand that the equations described the translational and rotational interaction of the displacement field. With my limited knowledge of continuum mechanics, I was able to sense that material rotation and its interaction with translational displacement is important in the plastic regime, and that the Maxwell-type field equations represent that effect. During the coffee break, I came to Acad. Panin to introduce myself and ask a number of questions about his presentation. He answered each of my questions enthusiastically. Moreover, he kindly invited me to the post-meeting banquet to be held at the USSR embassy later that day. Of course I accepted the invitation and attended the banquet where I was able to discuss with Acad. Panin a wide range of topics in strength physics and material sciences. He gave me a book written in Russian as a gift, and invited me to an international conference being held in the 1Presently the Russian Academy of Sciences. v vi Preface following summer in Tomsk, Siberia. I did not know the language at that time. I was so interested in the book that I took Russian language courses for 2 years. In the summer of 1991, I attended the conference in Tomsk and met a group of scientists working in Acad. Panin’s group. The discussions I had with them were revolutionary to me. They explained the Maxwell-type field equations, the interac- tion between the translational and rotational displacement in plastic deformation, and other gauge theoretical concepts in detail. To be honest, my knowledge about gauge theories at that time was almost none. After returning home, I read a handful of books on gauge theories and got more confused. I kept reading and learned that the electromagnetic field is the gauge field that makes quantum mechanics locally symmetric. This brought me to the turning point. I started to understand the concept of gauge transformation and local symmetry. I analyzed various gauge theoretical concepts in deformation dynamics via analogy to electrodynamics. Interestingly, this exercise deepened my understanding on electrodynamics. I noticed a number of different views on Faraday’s law and Ampere’s law as the interaction between the electric and magnetic field that nature uses as a mechanism to stabilize events, e.g., prevent runaway increase of current. This, in turn, helped me consolidate the basic understanding on the gauge-field nature of plastic deformation dynamics. As I kept deepening my understanding on the physical foundation of physical mesomechanics, I realized that the theory was much more profound than I initially thought. It was an elegant theory capable of describing plastic deformation based on pure physics, unlike most theories of plastic deformation that relies on phe- nomenology or mathematical models. It indicated a number of potential engineering applications as well. However, the work at that time was somewhat inclined toward the mathematical aspect of the theoretical foundation with little experimental proofs. I started to conduct experiments trying to prove various elements of the theory, such as transverse wave characteristics of displacement field in the plastic regime. To measure displacement field, I used an optical interferometric technique known as the ESPI (Electronic-Speckle Pattern Interferometry). I found several interesting phenomena that could be explained by the same physical foundation as physical mesomechanics. Through analysis of these experimental observations, especially with the help of analogy with electrodynamics, I conceived new ideas in the description of deformation dynamics such as the concept of deformation charge and its role of energy dissipation. This helped me advance the theory from the field- theoretical description of plastic deformation dynamics to a comprehensive theory of deformation and fracture based on the same theoretical foundation. To date, I have continued investigating the field theoretical dynamics of deformation using the ESPI. As will be discussed in the following chapters, development of this theory has not been completed. I decided to put together the knowledge and informationI gained so far as a book at this point for several reasons. First, recent experimental observations have convinced me of the validity of the theory. Essentially, the gauge field in deformation dynamics makes the law of linear elasticity locally symmetric. The nonlinear dynamics in the plastic regime is formulated through the potential associated with the gauge field. Second, these experimental observations and their Preface vii field theoretical interpretations demonstrate potential of engineering applications. In particular, the use of ESPI techniques allows us to visualize the deformation field as a full-field image, and along with field theoretical interpretations, it provides us with various information. For example, the use of the field theoretical criteria of plastic deformation and fracture allows us to make diagnosis of the deformation state of a given object. Third, I would like to invite specialists of different disciplines to this research for further development of the theory and applications. On the theoretical side, connections with microscopic theories are very important. At this point the theory incorporates the effect of microscopic defects that causes plastic deformation generally into the field equations via the source terms. If a specific form of the source term is provided by a microscopic theory, it is possible to describe how the microscopic defect can evolve to the final fracture under a given condition. Also, more thermodynamic argument will allow us to discuss the energy dissipation process resulting from irreversible plastic deformation more specifically. For applications, software development for visualization of displacement field in objects under deformation, especially during the transitional stage from one regime to another, e.g., from the elastic to plastic regime, will be not only an interesting application but also helps further advancement of the theory. Numerical simulations are also important for further tests of the theory and explore for new applications. Lastly, I would like to share my experience of learning the Maxwell’s formalism and the gauge theories with students. A number of electrodynamic concepts that were unclear to me when I was in the graduate school became crystal clear through this project. I would like to invite students and have them feel the beauty of field theories. In my opinion, this subject is ideal to visualize the concept of local symmetry associated with a gauge field, which is otherwise abstract and difficult to comprehend. I tried my best to portray the complicated concept with plain terms and analogies without going into mathematical details. It is also a unique case in which these concepts, which are usually discussed by scientists specialized in basic physics such as high energy or particle physics, are discussed in connection with real world applications such as nondestructive testing of metal objects. I hope that this book is helpful to people in any of these and related disciplines. I tried my best to cite literature appropriately. If some papers or books do not receive fair credit or are not cited, I apologize. Finally, I would like to express my sincere gratitude to a number of people. First of all, I would like to thank Acad. Victor Panin for introducing to me his beautiful paradigm of deformation dynamics and his friendship ever since. I am grateful to countless colleagues who always supported me during the development of this theory, especially Professor Cesar Sciammarella for his continuous encouragement and precious discussions. Through my learning processes of gauge theories and continuum mechanics, I realized that I owe greatly to all the professors and teachers from whom I received my graduate and undergraduate trainings. Without their excellent instruction, I would have never reached the present level of understanding on the subjects. I also thank all of my friends and students who helped me with the experiments and computations that provided a number of supporting data. Hammond, LA, USA Sanichiro Yoshida Contents 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1 Background. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.2 Aim, Scope, and Organization of This Book . . . . . . . . . . . . . . . . . . . . . . . . . . 5 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 2 Quick Review of Theories of Elastic Deformation . . . . . . . . . . . . . . . . . . . . . . . 9 2.1 Displacement and Deformation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 2.2 Hooke’s Law and Poisson’s Ratio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 2.3 Principal Axis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 2.4 Equation of Motion and Elastic Waves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 2.4.1 One-Dimensional Longitudinal Elastic Waves . . . . . . . . . . . . . . . . 32 2.4.2 Three-Dimensional Compression Waves . . . . . . . . . . . . . . . . . . . . . . 33 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 3 Quick Review of Field Theories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 3.1 The Field . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 3.2 Symmetry in Physics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 3.3 Global and Local Transformation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 3.4 Gauge Field. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 3.5 Lagrangian Formalism and Field Equation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 3.6 Electrodynamics as a Gauge Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 4 Field Theory of Deformation and Fracture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 4.1 Gauge Theories of Deformation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 4.2 The Big Picture of the Present Field Theory. . . . . . . . . . . . . . . . . . . . . . . . . . . 64 4.3 Field Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 4.4 Wave Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 ix x Contents 5 Interpretations of Deformation and Fracture Phenomena from Field Theoretical Viewpoint . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 5.1 Field Equation as an Equation of Motion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 5.2 Comprehensive Description of Deformation . . . . . . . . . . . . . . . . . . . . . . . . . . 93 5.2.1 Elastic Deformation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 5.2.2 Plastic Deformation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . 100 5.2.3 Fracture. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 5.3 Physical Meaning of Potentials. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114 5.3.1 Vector Potential and Scalar Potential from Gauge . . . . . . . . . . . . 114 5.3.2 Scalar and Vector Potential from Viewpoint of Wave Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118 5.3.3 Field Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 5.4 Physical Meaning of Charges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127 5.5 Thermodynamic Consideration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132 6 Optical Interferometry and Application to Material Characterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 6.1 Basics of Light and Optics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136 6.1.1 Light as an Electromagnetic Wave . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136 6.1.2 Interaction with Media and Photon . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142 6.1.3 Laser and Coherence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150 6.1.4 Imaging. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153 6.2 Interference and Interferometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155 6.2.1 Mathematics of Waves. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155 6.2.2 Interference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161 6.2.3 Michelson Interferometer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163 6.2.4 Mach–Zehnder Interferometer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164 6.3 Electronic Speckle-Pattern Interferometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165 6.3.1 Speckles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165 6.3.2 In-Plane Displacement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167 6.3.3 Out-of-Plane Displacement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170 7 Experimental Observations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171 7.1 Plastic Deformation Wave . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171 7.1.1 Decay Characteristics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177 7.2 Vortex-Like Displacement Field. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181 7.3 Observation of Charge-Like Phenomena . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183 7.3.1 Charge-Like Pattern and PLC Band . . . . . . . . . . . . . . . . . . . . . . . . . . . 185 7.3.2 Cyclic Loading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191 7.3.3 Experiment with Notched Specimen. . . . . . . . . . . . . . . . . . . . . . . . . . . 197 7.3.4 Acoustic Emission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199 7.3.5 Temperature Rise Due to Plastic Deformation . . . . . . . . . . . . . . . . 201 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207 Contents xi 8 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209 8.1 Evaluation of Stress Concentration with Charge-Like Patterns . . . . . . . 210 8.1.1 Stabilized and Unstabilized Non-welded Specimens, A5052-S and A5052-N. . . . . . . . . . . . . . . . . . . . . . . . . . . . 210 8.1.2 AA6063 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213 8.1.3 Welded Specimens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215 8.2 Plastic Deformation and Fracture Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219 8.2.1 Plastic Deformation Criterion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219 8.2.2 Fracture Criterion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221 8.2.3 ESPI Experiment on Plastic Deformation and Fracture Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221 8.2.4 Interpretation of Fringe Patterns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222 8.3 Evaluation of Load Hysteresis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233 Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235 Chapter 1 Introduction 1.1 Background Deformation and fracture of solid-state materials have been studied for centuries. The subject is not only of scientific interest but also of extreme practical importance. A great number of scholars in a wide range of disciplines developed various theories. The history of continuum mechanics can be traced back to the Hellenic period [1]. In 1660, Robert Hooke discovered the law of elasticity, the linear relation between tension and extension in an elastic spring. The scientists in the seventeenth and eighteenth centuries introduced the basic concepts of strain by extending Newton’s law of motion for a point mass to a motion law for a deformable body with a finite volume. By the middle of the nineteenth century, Augustin-Louis Cauchy, a French mathematician, compiled these achievements into the basic framework of three- dimensional continuum mechanics. Cauchy’s contribution is of especial significance because of the mathematical rigorousness of his formulation, which is contrastive to the heuristic approach exploited by earlier scientists. This is evidenced by the fact that his formulations are still used by engineers of the twenty-first century, including numerous software packages for numerical simulations of solid mechanics. In the twentieth century, continuum mechanics developed to a unifying theory combined with the advancement in thermodynamics and rheology. Clifford Truesdell was the major force in this development. A number of textbooks such as the one written by Landau and Lifshitz as a volume of their theoretical physics courses [2] are available. Elasticity as a nonlinear problems is also studied. It was initiated by Poincare and Lyapunov in their study of ordinary differential equations of discrete mechanics at the end of the nineteenth century. Methods for handling nonlinear boundary-value problems were slowly developed by a handful of mathematicians on the first half of the twentiethcentury. Further information about nonlinear elasticity can be found in [3]. © Springer Science+Business Media New York 2015 S. Yoshida, Deformation and Fracture of Solid-State Materials, DOI 10.1007/978-1-4939-2098-3__1 1 2 1 Introduction In the area of plastic deformation, there are several mathematical descriptions [4]. One is nonlinear deformation theory. In this theory, the stress is expressed as a nonlinear function of the strain, as opposed to a linear function of the strain as in the case of the linear elastic theory. This nonlinearity can also be viewed as that the stress–strain relation is locally linear (infinitesimally linear at each value of strain) on the stress–strain curve but the stiffness (the ratio of the stress to the strain) reduces as the strain increases beyond the linear limit. Although this approach is accurate as long as the stiffness is known as a function of strain, it cannot account for irreversibility of the deformation. Flow plasticity theory [5] assumes that the total strain can be decomposed into elastic and plastic parts. The plastic strain is determined from the linear elastic relation to the stress of the material. The plastic part is determined from a flow rule and a hardening model. Another important theory of plasticity is dislocation theory. In 1934, Egon Orowan [6], Michael Polanyi [7], and Geoffrey Ingram Taylor [8], approximately at the same time, published papers to explain plastic deformation in terms of dislocations. Experiments show that plastic deformation results from slip on specific crystallographic planes in response to shear stress along the planes. However, observed shear strength is orders of magnitude lower than theory. This discrepancy indicates that slip is caused by a mechanism where lattice defects referred to as dislocations move along the plane. When the dislocations complete the movement at the end of the plane, slip occurs on the entire slip plane. Fracture mechanics was developed during World War I by an aeronautical engineer, Alan Arnold Griffith [9], to explain the fact that fracture of brittle materials occur under two orders of magnitude lower stress than the theoretically predicted value. He developed a thermodynamic theory that states that the growth of a crack requires the creation of two new surfaces and hence an increase in the surface energy. At the same time, it reduces the elastic energy stored in the material. The difference of them (surface energy–elastic energy) can be interpreted as the free energy. As the crack area increases, the surface energy increases linearly and the elastic energy decreases quadratically, hence the free energy has a maximum. Griffith postulated that the crack would increase spontaneously leading to fracture beyond this maximum point because the free energy would decrease monotonically. Accordingly, he defined the corresponding crack length as the critical crack length. Griffith’s theory explains the fracture behavior of brittle materials well. However, the actual energy necessary to fracture a ductile material is orders of magnitude higher than the corresponding surface energy. During World War II, a group under George Rankine Irwin [10] realized that in ductile materials the plastic zone developing at the crack tip increases in size with the applied load, dissipating the energy as heat. Hence, an energy dissipative term has to be added in the energy balance. With this modification, the theory explains experiment for ductile materials. A number of researchers apply above-mentioned theories to engineering prob- lems as well. Timoshenko [11], for example, describes the use of elasticity for a number of practical applications. 1.1 Background 3 Sensor technology for defect detections has been advanced tremendously. A number of techniques based on various principles such as ultrasonic/optical imag- ing, eddy current, and other nondestructive technology have been developed and used in the fields. Recent advancement allows us to detect micro-cracks in a very early stage. On the material development side, a number of new materials with additional designed properties such as reinforcement and anti-corrosion have been developed. In spite of all these developments, catastrophic accidents still occur. Aircraft parts fail after passing the pre-flight inspection, structures such as bridges and stages collapse totally unexpectedly. In many cases, the cause is unknown. Apparently, the problems at least partly come from the limitation of the theory that the inspection procedure is based on. The problem is not necessarily in incompleteness of individual theories relevant to a given problem. Rather, it is the lack of connection between the regimes of deformation. As discussed above, most of the currently available theories are applicable selectively to a certain regime of deformation. They are accurate in describing the dynamics in the corresponding regime. However, in reality, the mechanical state cannot be characterized by a single regime. Even in a specimen about to fracture, deformation in some parts are still in the elastic regime. Flow plasticity theory may appear to be able to handle elastic and plastic deformation simultaneously via the elastic and plastic parts of the strain. However, this is a parameterized model, and the elasticity and plasticity are differentiated with parameters, not physical laws. The use of these theories does not allow us to model the transition from one regime to another. In the real world, on the other hand, accurate description of the transition is extremely important. The scale level is another issue. Micro and nano-technology is an emergent field in various engineering disciplines. Most theories of solid mechanics were developed for macroscopic objects. Mechanical properties of a material at the micro and nanoscopic levels can be substantially different from those at the macroscopic level. This imposes limitations in the applicability of existing theories to micro and nanoscopic objects. In particular, those theories based on experimentally evaluated parameters may need substantial modifications for applications to micro and nano- scale systems. The issue is not necessarily limited to micro and nano-technologies. Macroscopic fracture begins at the atomic level. The process starts with an atomistic defect, grows to a macroscopic crack, and eventually the fracture of the entire object. It is important that the theory can describe the transitions from one scale level to the next on the same physical basis. If the crack generation can be predicted in an earlier stage, the inspection technology will drastically advance. The above-addressed issue clearly indicates the necessity of a theory capable of describing all stages of deformation on the same physical basis, independent of the scale level. It is obvious that such a theory must be based on a fundamental level of physics. In this regard, the gauge theoretical approach that Panin et al. employed to formulate dynamics of plastic deformation is promising. Details of their approach can be found elsewhere [12–15]. In short, their approach is as follows: they describe deformation with a transformation of GL.3;R/ group (three-dimensional general linear group over real numbers [16]), and request local symmetry [17] in 4 1 Introduction the transformation. In other words, they allow that the transformation matrix is coordinate dependent and request that the dynamics be expressed in the same form before and after the transformation. This requires replacement of usual derivatives with covariant derivatives, or equivalently, introduction of a gauge. They find an appropriate gauge and Lagrangian associated with it. Based on the principle of least action, they derive field equations for each group element. After summation over the group index, the field equations take a form analogous to Maxwell equationsof electrodynamics. The solution to the field equations represents transverse wave characteristics of the displacement field in the plastic regime. The present theory is based on the Panin’s formalism. When summed over the group index (after contraction over the index representing the group elements) the GL.3;R/ transformation matrix becomes the deformation tensor widely used in linear elastic theory. Thus, the Panin’s approach can be interpreted as requesting local symmetry in linear elasticity, which indicates that this formulation should reduce to the conventional continuum mechanics. Subsequent analyses [18–20] indicate that the transverse wave characteristics in the displacement field in the plastic regime are driven by the shear restoring force of the material represented by the shear modulus and that the longitudinal effect in the plastic regime is not elastic force proportional to displacement but rather an energy dissipative effect. Further, it has been found that fracture occurs when the material loses both its shear resiting force mechanism and the longitudinal energy dissipating mechanism, hence it loses all mechanisms to convert the mechanical work done by the external agent to another form of energy. These altogether indicate the possibility of describing all the stages, from the elastic through fracture, based on the same field equations. One quite interesting feature of the field equations is that they are analogous to Maxwell’s equations of electrodynamics. As will be explained in various sections throughout this book, the similarity of the present theory with electrodynamics is not limited to the mathematical resemblance. There are a number of similarities in the physical behaviors between the displacement field and electromagnetic field. In fact, the physical meaning of various behaviors of the displacement field can be interpreted based on the analogy with electrodynamics, and these interpretations have led to further understanding of the deformation dynamics. A number of experimental observations have also been interpreted based on the analogy, and that has facilitated the theoretical development. It should be noted that the present theory does not refer to a cause of deformation, in the same sense as an equation of motion does not refer to the specifics of the external force. The present field equations describe the relation between transla- tional and rotational modes of displacement. The cause of irreversible deformation is incorporated into the field equations through the source terms. It is possible to integrate the present theory with a microscopic theory, e.g., a dislocation theory, to deal with the cause of deformation. It is an important future subject. 1.2 Aim, Scope, and Organization of This Book 5 1.2 Aim, Scope, and Organization of This Book The primary aim of this book is to introduce the field theoretical approach to deformation and fracture. The theoretical foundation is described and supporting experiments are discussed. It should be emphasized that the present theory is still developing. This book is not to present the completed form of the theory; rather it is to invite researchers to consider the viewpoints of the present theory and hopefully apply the approach to their own purpose. The materials that this book tries to cover are quite interdisciplinary. It is likely that most engineers are unfamiliar with the concept of local symmetry and gauge transformation. Continuum mechanics is not a subject that scientists deal with on a regular basis. It is my intention to describe the big picture of the approach, rather than going into details of the content in each discipline. Those who are interested in more detailed information are encouraged to read books or other resource of the subject field. It is also my intention to invite people of various disciplines, engineers, scientists, and technicians. This is because exchanges of opinions among researchers in different disciplines are most important for further development of this approach. For this reason, much efforts have been made to explain the concepts of each discipline in such a way that people in other disciplines can digest them as easily as possible. Special attention is paid so that the reader does not need prior knowledge except for basic physics and engineering. Basic knowledge on electrodynamics, gaseous electronics, solid mechanics, and quantum mechanics will be helpful, but not prerequisite. For this purpose, whenever seems necessary, extra explanations are added. In some occasions, the mathematical derivation may appear lengthy and perhaps redundant. The objectives of each chapter are as follows. It is intended that each chapter is, to some extent, self-contained so that the reader can use them as a reference. For instance, those who are interested in applying an optical interferometric technique to mechanical analysis, Chap. 6 will be useful. In this chapter, basic concepts of light and optics which may not appear directly related to the interferometric techniques are described. The purpose of the provision of these materials is to facilitate the interferometric experiments. Conducting an optical interferometric experiment with and without these knowledges make a huge difference in the experimental efficiency and the quality of the results. The interferometric fringe-contrast will be much better, for example, if we use an optical interferometer with the proper understanding of the coherence in laser light, as opposed to simply follow the procedures to get outputs from the interferometer. Other chapters are written with the same general philosophy. Chapter 2 reviews continuum mechanics. Basic physical concept of elasticity and mathematical description such as the strain tensor, stress tensor, and constitutive equations are discussed. Some of the continuum-mechanical concepts and mathe- matical expressions are used in the present field approach. The goal of this chapter is to facilitate the description of the mathematical procedure used to derive the present 6 1 Introduction field equations in Chap. 5. Those who are familiar with continuum mechanics can skip this chapter. Chapter 3 discusses the gauge transformation and various underlying concepts such as symmetry in physics, covariant derivatives, local symmetry, and gauge potential. These concepts are not easy to digest for those who are not familiar with the field. An effort has been made to discuss the complex concept as easily as possible. The goal of this chapter is to describe the big picture of the concept without going into mathematical details. Those who are familiar with gauge transformation can skip this chapter. Chapters 4 and 5 discuss the present field theory in detail. Chapter 4 focuses on the formalism of the theory. The concept of gauge transformation discussed in Chap. 3 is applied to the displacement field of a solid-state medium under plastic deformation and the resultant field equations are discussed. Chapter 5 discusses the physical meaning of the field equations and various concepts derived from the field equations. One of the field equations is interpreted as the equation of motion that governs the dynamics of a unit volume in the object under deformation. Wave dynamics of the displacement field as solutions to the equation of motion are discussed. The energy dissipative nature of plastic deformation is argued via the concept of deformation charge, which is analogous to the electric charge. The physical meanings of the charge and its interaction with the displacement field are discussed. Through these discussions, the field equations are argued as the governing equations of deformation for all stages; the elastic, plastic, and fracturing stage. The form of the term representing the longitudinal force in the above- mentioned equation of motion differentiates one stage from another. Chapter6 explains optics and optical interferometry that are used in the support- ing experiments discussed in Chaps. 7 and 8. Interestingly, some of the behaviors of light as an electromagnetic wave are analogous to the deformation wave. This is not surprising because the field equations of the deformation field are analogous to Maxwell equations of electrodynamics. Various similarities between the light and deformation fields are discussed. It will help us digest the field theoretical dynamics of deformation discussed in Chaps. 4 and 5 from a different angle. Chapters 7 and 8 present experimental results that support the present field theory. Chapter 7 focuses on various field theoretical concepts such as decaying, transverse displacement-waves in the plastic regime and the deformation charge and its behavior that causes energy dissipation. Chapter 8 discusses engineering applications of the present theory including diagnosis of the current deformation regime (elastic, plastic, or fracturing regime) and evaluation of load hysteresis for a given object. References 7 References 1. Dugas, R.: A History of Mechanics. Editions du Griffon, Neuchatel (1955) 2. Landau, L.D., Lifshitz, E.M.: Theory of Elasticity. Course of Theoretical Physics, vol. 7, 3rd edn. Butterworth-Heinemann, Oxford (1986) 3. Antman, S.S.: Nonlinear Problems of Elasticity. Applied Mathematical Sciences, vol. 107. Springer, New York/Budapest (1995) 4. Hill, R.: The Mathematical Theory of Plasticity. Oxford University Press, Oxford (1998) 5. Lubliner, J.: Plasticity Theory. Courier Dover, New York (2008) 6. Orowan, E.: Z. Phys. 89, 605, 614, 634 (1934) 7. Polanyi, M.: Z. Phys. 89, 660 (1934) 8. Taylor, G.I.: Proc. R. Soc. A145, 362 (1934) 9. Griffith, A.A.: Philos. Trans. A 221, 163–198 (1920) 10. Irwin, G.R.: “Fracture Dynamics,” Fracturing of Metals. American Society for Metals, Cleveland (1948) 11. Timoshenko, S.P., Goodier, J.N.: Theory of Elasticity. McGraw-Hill, New York (1951) 12. Panin, V.E., Grinaev, Yu.V., Egorushkin, V.E., Buchbinder, I.L., Kul’kov, S.N.: Sov. Phys. J. 30, 24–38 (1987) 13. Panin, V.E.: Wave nature of plastic deformation. Sov. Phys. J. 33(2), 99–110 (1990) 14. Danilov, V.I., Zuev, L.B., Panin, V.E.: Wave nature of plastic deformation of solids. In: Panin, V.E. (ed.) Physical Mesomechanics and Computer-aided Design of Materials, vol.1, p. 241. Nauka, Novosibirsk (1995) (Russian) 15. Panin, V.E.: Physical fundamentals of mesomechanics of plastic deformation and fracture of solids. In: Panin, V.E. (ed.) Physical Mesomechanics of Heterogeneous Media and Computer- Aided Design of Materials. Cambridge International Science Publishing, Cambridge (1998) 16. Ibragimov, N.H.: Transformation Groups and Lie Algebras. World Scientific, Singapore (2013) 17. Elliott, J.P., Dawber, P.G.: Symmetry in Physics, vol. 1. Macmillan, London (1984) 18. Yoshida, S., Siahaan, B., Pardede, M.H., Sijabat, N., Simangunsong, H., Simbolon, T., Kusnowo, A.: Phys. Lett. A 251, 54–60 (1999) 19. Yoshida, S.: Phys. Mesomech. 11, 140–146 (2008) 20. Yoshida, S.: Scale-independent approach to deformation and fracture of solid-state materials. J. Strain Anal. 46, 380–388 (2011) Chapter 2 Quick Review of Theories of Elastic Deformation This chapter describes conventional approaches to elastic deformation known as continuum mechanics or theory of elasticity. The goal of this chapter is to discuss mathematical descriptions of kinematic and dynamics that these conventional theories use. Some of them are basis of the present field theory. It is not my intention to cover a wide area of the subject of elastic deformation. Rather, it is to prepare for the mathematical procedures developed in later chapters where we derive various equations of the present theory and interpret their physical meanings via comparison with conventional approaches. For this purpose, some of the concepts are viewed from different angles than conventional theories of elasticity. For complete description of elastic deformation, the reader is encouraged to refer to other books [1–6]. 2.1 Displacement and Deformation Consider point P1 located at a coordinate point .x1; y1; z1/ in an object. We express this point with a position vector ! OP1 as illustrated in Fig. 2.1. Similarly, another point in the same object P2 at .x2; y2; z2/ can be expressed with position vector! OP2. Apparently, the coordinates .x; y; z/ identify position of different points in the same object, and the coordinate origin can be viewed as a reference point affixed to the object. Such coordinates are referred to as Lagrangian coordinates. Now consider in Fig. 2.2 that pointP1 is displaced to another pointP 01. To express the displacement, we introduce a displacement vector �. As a three-dimensional vector, � has three components that can be expressed with a coordinate system, .�x; �y; �z/. While both .x; y; z/ and .�x; �y; �z/ have the dimension of length and are three-dimensional vectors, the meaning of these two vectors are different from each other. The set of components .x; y; z/ identify the position in a given object © Springer Science+Business Media New York 2015 S. Yoshida, Deformation and Fracture of Solid-State Materials, DOI 10.1007/978-1-4939-2098-3__2 9 10 2 Quick Review of Theories of Elastic Deformation Fig. 2.1 Position vector ! OP1 representing a point in an object P1 (x,y,z) OP1 y x z Fig. 2.2 Point P1 and point P2 displaced for different amounts dr dr' x(r) x(r + dr) dx dr' = dr + dx P1 P2 P1' P2' relative to the origin .x; y; z/ D .0; 0; 0/. On the other hand, .�x; �y; �z/ does not identify a point in the object. Instead, it represents the change in the position of a point in the object as a result of some physical event such as exertion of a force by an external agent. Thus, their reference point is not affixed to the object but rather to the frame of analysis. This type of coordinates is referred to as Eulerian coordinates.1 Deformation is defined as the situation where different points in the same object are displaced differently. Note that as will be discussed in detail below, the coordinate point after displacement is the addition of the initial position vector and the displacement vector. As such, the position vector after the displacement is expressed in the Eulerian coordinates. That is why some books [5] state that the point before deformation is in Lagrangian coordinates and that after is in Eulerian coordinates. Figure 2.2 illustrates the situation where points P1 and P2 are displaced differently as the object is deformed. Here the two points are considered to be separated by an infinitesimal distance ! dr . 1Consider you are still at .x; y; z/ on the earth (the object). Since you are not moving on the earth, the value of .x; y; z/ does not change. However, since you are moving with the earth, the displacement vector representing your motion with reference to the sun keeps changing. 2.1 Displacement and Deformation 11 ! drD !OP2 � ! OP1 : (2.1) By the deformation, the two points are displaced by �1 and �2. ! OP 01 D ! OP1 C�1 ! OP 02 D ! OP2 C�2 (2.2) Thus, the distance between the two points after the deformation is expressed as ! dr 0D ! OP 02 � ! OP 01D ! dr C.�2 � �1/: (2.3) The two displacement vectors can be viewed as two values of a common displace- ment function �.x; y; z/ at points .x1; y1; z1/ and .x2; y2; z2/. Further, as the distance between points P1 and P2 is infinitesimally small, we can put �2 � �1 D �.x2; y2; z2/ � �.x1; y1; z1/ D �.x1 C dx; y1 C dy; z1 C d z/ � �.x1; y1; z1/ D ! d� .x; y; z/: (2.4) Therefore, the change in the infinitesimal distance due to the deformation can be put as follows: ! dr 0D ! dr Cd�.x; y; z/: (2.5) Here, for the purpose of generalization, the suffixis dropped in the rightmost term. Since each component of the displacement vector is a function of .x; y; z/. d�i D @�i @x dx C @�i @y dy C @�i @z d z; i D x; y; z: (2.6) In matrix notation, d� can be put d�i D 0 @ d�x d�y d�z 1 A D 0 BBBBB@ @�x @x @�x @y @�x @z @�y @x @�y @y @�y @z @�z @x @�z @y @�z @z 1 CCCCCA 0 @ dx dy d z 1 A (2.7) 12 2 Quick Review of Theories of Elastic Deformation Hence, Eq. (2.5) becomes 0 @ dx0 dy0 d z0 1 A D 0 @ dx dy d z 1 AC 0 BBBBB@ @�x @x dx C @�x @y dy C @�x @z d z @�y @x dx C @�y @y dy C @�y @z d z @�z @x dx C @�z @y dy C @�z @z d z 1 CCCCCA D 0 BBBBB@ 0 @ 1 0 0 0 1 0 0 0 1 1 AC 0 BBBBB@ @�x @x @�x @y @�x @z @�y @x @�y @y @�y @z @�z @x @�z @y @�z @z 1 CCCCCA 1 CCCCCA 0 @ dx dy d z: 1 A (2.8) More concisely, .dxi /0 D � ıij C @�i @xj � dxj � Uij dxj : (2.9) Here ıij is the Kronecker’s delta. The matrices @�i @xj and Uij D ıij C @�i @xj are called the displacement gradient tensor and the deformation gradient tensor, respectively. It is convenient to divide the displacement gradient tensor into the symmetric and asymmetric terms. 0 BBBBBB@ @�x @x @�x @y @�x @z @�y @x @�y @y @�y @z @�z @x @�z @y @�z @z 1 CCCCCCA D 0 BBBBBB@ @�x @x 1 2 � @�x @y C @�y @x � 1 2 � @�z @x C @�x @z � 1 2 � @�x @y C @�y @x � @�y @y 1 2 � @�z @y C @�y @z � 1 2 � @�x @z C @�z @x � 1 2 � @�z @y C @�y @z � @�z @z 1 CCCCCCA C 0 BBBBBB@ 0 �1 2 � @�y @x � @�x @y � 1 2 � @�x @z � @�z @x � 1 2 � @�y @x � @�x @y � 0 �1 2 � @�z @y � @�y @z � �1 2 � @�x @z � @�z @x � 1 2 � @�z @y � @�y @z � 0 1 CCCCCCA (2.10) As Fig. 2.3 illustrates, the symmetric part represents strain and the asymmetric part rotation. The former is referred to as the strain tensor and the latter as the rotation tensor. They can be concisely expressed as follows: 2.1 Displacement and Deformation 13 Fig. 2.3 Strain and rotation part of deformation tensor Shear strain Normal strain + Rotation �ij D @�j @xi C @�i @xj ; (2.11) !ij D @�j @xi � @�i @xj ; (2.12) The strain tensor can further be divided into the normal strain and shear strain terms. �ij D �n C �sh: (2.13) Here, �n D 0 BBBBB@ @�x @x 0 0 0 @�y @y 0 0 0 @�z @z ; 1 CCCCCA (2.14) �sh D 1 2 0 BBBBBB@ 0 � @�x @y C @�y @x � � @�z @x C @�x @z � � @�x @y C @�y @x � 0 � @�z @y C @�y @z � � @�x @z C @�z @x � � @�z @y C @�y @z � 0 1 CCCCCCA (2.15) For simplicity, let’s define the line element vector � for before and �0 for after the deformation. 14 2 Quick Review of Theories of Elastic Deformation � D 0 @ dx dy d z 1 A (2.16) �0 D 0 @ dx0 dy0 d z0 1 A (2.17) With these expressions, the deformation gradient tensor and the deformation as a transformation become Uij D ıij C �n C �sh C !ij (2.18) �0 D U� (2.19) It is worth exploring the physical meaning of each term of the deformation gradient tensor. Consider two-dimensional deformation in the x-y plane in Fig. 2.4 where the x and y components of an infinitesimal line element vector dx and dy are transformed to dx0 and dy0, respectively, by deformation. Since translational displacement is not of our interest, the tail of the line element vector after the deformation is shifted to that of the before deformation. For simplicity, we consider a two-dimensional case here but the same argument holds for three dimensions. From Fig. 2.4, ! dx0 D ! dx C ! AA0 ! dy0 D ! dy C ! BB 0 (2.20) Fig. 2.4 Physical meaning of each term of deformation gradient tensor dy dy ∂y ∂xy dy ∂y ∂xx dx ∂x ∂xx dx ∂x ∂xy dy' dx dx' B O A B' A' B" A" 2.1 Displacement and Deformation 15 Considering the spatial dependence of the displacement vector � D �x Ox C �y Oy, ! AA0 D � @�x @x dx � Ox C � @�y @x dx � Oy ! BB 0 D � @�x @y dy � Ox C � @�y @y dy � Oy (2.21) So, ! dx0 D dx Ox C � @�x @x dx � Ox C � @�y @x dx � Oy (2.22) ! dy0 D dy Oy C � @�x @y dy � Ox C � @�y @y dy � Oy (2.23) With Eqs. (2.16) and (2.17) substituted into Eq. (2.19), and Fig. 2.4 along with Eqs. (2.22) and (2.23), the meaning of each term of Uij in Eq. (2.18) can be interpreted as follows. The first term, the unit matrix ıij , represents the undeformed part of the line element, i.e., the first terms of Eqs. (2.22) and (2.23). The second term �n represents the normal strain expressed by the second term of Eq. (2.22) and the third term of Eq. (2.23). The geometric meaning of the third term �sh representing the shear strain becomes clear from the following discussion. Consider the scalar product of ! dx0 and ! dy0. ! dx0 � ! dy0 D �� dx C @�x @x dx � Ox C � @�y @x dx � Oy � � �� @�x @y dy � Ox C � dy C @�y @y dy � Oy � � � @�y @x C @�x @y � dxdy (2.24) where the second-order terms of the derivatives of the displacement�� @�x @x �� @�x @y � etc: � are neglected. By definition, this scalar product can be written as follows: ! dx0 � ! dy0 D ˇˇ ˇˇ !dx0 ˇˇ ˇˇ ˇˇ ˇˇ !dy0 ˇˇ ˇˇ cos � 0 D s� 1C @�x @x �2 C � @�y @x �2s� @�x @y �2 C � 1C @�y @y �2 cos � 0dxdy � cos � 0dxdy; (2.25) 16 2 Quick Review of Theories of Elastic Deformation where � 0 is the angle between ! dx0 and ! dy0, and @�x @x ; @�y @y << 1 (small deformation or the length change �x; �y is much smaller than dx; dy) is used. Equating the right-hand side of Eqs. (2.24) and (2.25), we obtain � @�y @x C @�x @y � D cos � 0 D sin �� 2 � � 0 � � � 2 � � 0 (2.26) Here the small angle approximation is used for � 2 �� 0. Equation (2.26) indicates that� @�y @x C @�x @y � is the change in the angle between ! dx and ! dy to ! dx0 and ! dy0 caused by the deformation. The geometrical meaning of the third term !ij in Eq. (2.18) can be understood in a similar fashion. Consider the vector product of ! dx and ! dx0 ! dx � ! dx0D dx Ox � �� dx C @�x @x dx � Ox C � @�y @x dx � Oy � D @�y @x dx2Oz (2.27) By definition ! dx � ! dx0D dx dx0 sin �x Oz � dx2 sin �x Oz (2.28) where �x is the angle between ! dx � ! dx0. From Eqs. (2.27) and (2.28), @�y @x D sin �x � �x: (2.29) Similarly, we can easily find that @�x @y D sin �y � �y: (2.30) From Eqs. (2.29) and (2.30), we find that @�y @x � @�x @y D �x � �y; (2.31) i.e., the rotation represents the difference between the angle of rotation of ! dx and ! dy. 2.2 Hooke’s Law and Poisson’s Ratio 17 2.2 Hooke’s Law and Poisson’s Ratio The argument of the preceding section does not yet refer to dynamics because the concept of force has not been introduced. The underlying force law in elastic deformation is essentially Hooke’s law, which states that elastic media exert resistive force proportional to the displacement from the equilibrium position in response to an external force. Here the equilibrium position is the position that a part of the medium takes when there is no externalforce acting on that part. Different parts of a given elastic object have their own equilibrium positions that are different from one another. The simplest way to visualize this type of force is a series of point masses connected with springs, as illustrated in Fig. 2.5. In this context, the i th mass’ equilibrium is established when neither of the springs directly connected to this mass is stretched or compressed. Note that the force exerted by other stretched springs are not external force on the i th mass. The force proportional to the displacement from the equilibrium position can be evaluated by differentiating the potential energy Up.x/ with respect to the space coordinate. Here x is the distance from the equilibrium position. By Taylor- expanding the potential energy with respect to x, we can express the situation as follows: U.x/ D U.0/C U 0.0/x C 1 2 U 00.0/x2 C � � � (2.32) F.x/ D U 0.0/C U 00.0/x C � � � (2.33) ith ith ith unstretchedstretched stretched Displacement of ith mass Fig. 2.5 Spring mass system to represent elasticity. All springs are unstretched (top). All springs are stretched (middle). Springs connected to the i th mass are unstretched and all other springs are stretched (bottom) 18 2 Quick Review of Theories of Elastic Deformation Fig. 2.6 Normal stress acting on x surface f Dll Ax At the equilibrium x D 0, the force must be zero because the atom is at the bottom of the potential well. Thus, U 0.0/ D 0. From this viewpoint, the spring force is interpreted as the case when we take up to the second-order term of the potential function. By putting U 00.0/ D �k2 where k is the stiffness or the spring constant of the spring, we can rewrite Eq. (2.33) as follows: F.x/ D �kx (2.34) To discuss deformation, the total external force has much less meaning than force per unit area. The force exerted on a unit area is called the stress. The concept of stress is most conveniently explained through consideration of a cube in a medium as shown in Fig. 2.6. In a one-dimensional case where the force is in the positive x-direction, Eq. (2.34) leads to3 �xx D E�xx (2.35) Here �xx is the normal stress, �xx is the normal strain, and E is the Young’s modulus. Here Young’s modulus represents the stiffness, as the spring constant does in Eq. (2.34). Since �xx is acting on a plane of unit area, its dimension is “force per area,” or in the SI unit N=m2. Also, as clear from Fig. 2.6, the force associated with �xx is differential force, or the difference in the normal force acting on the left surface of the cube and the right surface. Each of these forces is the spring force at the corresponding surface. If the normal force acting on the two surfaces is the same, the cube would not be stretched or compressed; it would simply be accelerated as a rigid body. Accordingly, the quantity multiplied to the stiffness on the right-hand side must be “stretch per length,” or strain. The unit of the stiffness E is therefore .N=m2/=.m=m/ D N=m2. �xx D f Ax (2.36) 2A negative sign is used on the right-hand side to emphasize that the force is centripetal, or opposite to the displacement from the equilibrium. 3See Sect. 2.4.1 for further descriptions about the Young’s modulus in the context of dynamics. 2.2 Hooke’s Law and Poisson’s Ratio 19 Note that the first subscript in �xx denotes the plane defined by the direction of the outward unit vector normal to the plane and the second subscript denotes the direction of the force acting on the plane. It is useful to consider the relation between the spring constant and the Young’s modulus. Consider an elastic material (spring) of length l and its cross-sectional area Ax being stretched by force f . Denoting the stretch of this material with �l , we can relate f and �l with the spring constant k.4 f D k ��l (2.37) Here the direction of the stretch is x. Then the normal strain �xx is �xx D �l l (2.38) Substituting Eqs. (2.36) and (2.38) into Eq. (2.37), we obtain �xxA D .kl/�xx (2.39) Comparing Eqs. (2.35) and (2.39), we find k D EA l (2.40) Notice that the spring constant k depends on the size of the object, whereas the stiffness (Young’s modulus) E is a material constant. Below, we will find that the phase velocity of one-dimensional longitudinal wave is p E=�, where � is the density of the material. This indicates that the longitudinal wave’s velocity, e.g, sound velocity, is uniquely determined by the material. This is contrastive to the angular resonant frequency of an object of mass m and spring constant k is p k=m; it depends on the object’s size (mass). In the above argument, the differential force is normal to the surface. What if the differential force is parallel to the surface; for example, the force acting on the top surface of the cube in Fig. 2.6 parallel to the surface is different from the force acting on the bottom surface parallel to the other force on the top surface? The cube will experience shear strain as shown in Fig. 2.3. The stress corresponding to the transversely differential force is referred to as the shear stress and the constant of proportionality in its relation to the shear strain is referred to as the shear modulusG. �yx D G�yx (2.41) 4To avoid complexity, only the absolute value of the force is considered here. 20 2 Quick Review of Theories of Elastic Deformation Fig. 2.7 Stress vectors x y z o C B A sy sx s n sz Here, as is the case of the normal stress �xx , the first subscript y indicates the plane that the force is acting on and the second subscript x indicates the direction of the force. So far, the direction of the differential force has been one-dimensional. To extend this to three dimensions, we need to treat the stress as a three-dimensional vector. Consider a stress on a given plane ABC in Fig. 2.7. From the force equilibrium on the infinitesimally small tetrahedron OABC .5 � ndSn D � xdSx C � ydSy C � zdSz (2.42) Here � k; k D n; x; y; z is the stress vector acting on plane k and dSk is the area of plane k. Considering that the ratio dsi=dsn; i D x; y; z is the direction cosine of a normal vector of triangular area ABC with axis i , Eq. (2.42) can be put in the following form: � n D nx� x C ny� y C nz� z (2.43) where ni ; i D x; y; z is the direction cosine of the normal vector to the axis xi . Now the stress vector acting on each of the x, y, z-plane can be expressed in terms of the unit vector Oxi . � x D �xx Ox C �xy Oy C �xzOz (2.44) � y D �yx Ox C �yy Oy C �yzOz (2.45) � z D �zx Ox C �zy Oy C �zzOz (2.46) The coefficients of the unit vectors can be put in the form of a tensor referred to as the stress tensor as follows: 5The effect of body force such as gravity is omitted as it does not have a substantial effect in the argument in this chapter. 2.2 Hooke’s Law and Poisson’s Ratio 21 Œ�ij D 0 @ �xx �xy �xz �yx �yy �yz �zx �zy �zz 1 A (2.47) From Eq. (2.43) and Eqs. (2.44)–(2.46), � n D � nx�xx C ny�yx C nz�zx Ox C � nx�xy C ny�yy C nz�zy Oy C � nx�xz C ny�yz C nz�zz Oz � . ni�ix/ Ox C . ni�iy/ Oy C . ni�iz/Oz (2.48) Here row i represents the stress vector � i introduced in Eq. (2.43), and column j represents the j -component of each stress vector � i . In the last line of Eq. (2.48), the summation over index i is omitted for simplicity. So, for example, �yz is the z component of stress vector �y , which is acting on plane y. This convention is consistent with that for Eq. (2.36). Now consider relations among stress tensor components. Considering force equilibrium for a given volume enclosed by surface S , we obtain the following equation. Z Z S � n dS D Z Z S � nx�xi C ny�yi C nz�zi dS D 0 (2.49) where n is the unit vector normal to the surface S (� n is the componentof � normal to the surface dS) and Eq. (2.48) is used in going through the second equal sign. Using Green theorem, Eq. (2.49) can be converted to an integration over the volume as Z Z Z V � @�xi @x C @�yi @y C @�zi @z � dV D 0 (2.50) In order for Eq. (2.50) to hold in any volume, it follows that @�xi @x C @�yi @y C @�zi @z D 0 (2.51) or for each component @�xx @x C @�yx @y C @�zx @z D 0 @�xy @x C @�yy @y C @�zy @z D 0 @�xz @x C @�yz @y C @�zz @z D 0 22 2 Quick Review of Theories of Elastic Deformation Next considering rotational equilibrium, we obtain Z Z S r � � n dS D 0 (2.52) Here r D x OxCy OyC zOz. With the use of Eq. (2.48) and Green theorem, this leads to Ox Z Z Z V � y � @�xz @x C @�yz @y C @�zz @z � � z � @�xy @x C @�yy @y C @�zy @z � C ��yz � �zy � C Oy Œ� � � C Oz Œ� � � (2.53) From Eq. (2.52), the first part of the angle bracket Œ is zero. This leads to the following relation between �ij and its diagonal counterpart �ji as �xy D �yx; �yz D �zy; �zx D �xz (2.54) This indicates that the stress tensor Eq. (2.47) is symmetric. Œ�ij D 0 @ �xx �xy �zx �xy �yy �yz �zx �yz �zz 1 A (2.55) With the stress tensor (2.55), the one-dimensional constitutive equation (2.35) can be extended to three-dimensions as follows: �ij D Cklij �kl (2.56) Here �ij is the .i; j / component of strain tensor defined by Eq. (2.11). In accordance with the above convention, �ij denotes the j component of the stress vector acting on plane i . Similarly, �kl denotes the .k; l/ component of the strain tensor. Each combination of k and l represents the degree of freedom in deformation; e.g., .k; l/ D .x; y/ represents the shear deformation in a plane parallel to the x-y plane as defined by Eq. (2.57). The coefficient Cklij represents the response of the material to the external force represented by �ij for each degree of freedom. �xy D 1 2 � @�y @x C @�x @y � (2.57) With the use of symmetry of �ij and �ij , eq. (2.56) can be expressed in a matrix form as follows: 2.2 Hooke’s Law and Poisson’s Ratio 23 Fig. 2.8 Meanings of C12 and C21 x y z l l sxx eyy Dl exx syy Dl 0 BBBBBBB@ �xx �yy �zz �xy �yz �zx 1 CCCCCCCA D 0 BBBBBBB@ C11 C12 C13 C14 C15 C16 C21 C22 C23 � � � � � � C26 C31 � � � � � � � � � � � � � � � C41 � � � � � � � � � � � � � � � C51 � � � � � � � � � � � � � � � C61 � � � � � � � � � � � � C66 1 CCCCCCCA 0 BBBBBBB@ �xx �yy �zz �xy �yz �zx 1 CCCCCCCA (2.58) The matrix Ckl is called the stiffness tensor. Consider the meanings of C12 and C21. If we write only the relevant term in �xx and �yy in Eq. (2.58), we obtain the following expressions. �xx D C12�yy (2.59) �yy D C21�xx (2.60) As Fig. 2.8 illustrates, Eq. (2.59) describes how much the volume expands normally along y-axis when the volume is subject to the normal stress applied on the x-plane. Similarly, Eq. (2.60) describes how much the volume expands normally along x-axis when the volume is subject to the normal stress applied on the y-plane. Apparently, the volume does not know which axis is x or y, meaning that if we switch the x-axis with the y-axis, the physical situation is unchanged. It is obvious that Ckl D Clk , or the stiffness matrix is symmetric. We can repeat the same argument for C13 and C31 � � � to find there are C.6; 2/ D 15 (the number of 2-combinations from a set of 6-elements) redundancies. This reduces the number of independent Ckl from the total number of 6 � 6 D 36 to 21. In the case that the material’s response is symmetric, e.g., symmetric about the x-y plane, for example, we can further reduce the number of the elements in the stiffness matrix. Consider a new coordinate system x0y0z0 that is symmetric with xyz system about the x-y plane; x D x0, y D y0 and z D �z0. With this coordinate transformation, the sign of the z-component of a given vector is flipped, but the signs of the x and y components are unchanged. Therefore, �xz D ��x0z0 , �yz D ��y0z0 and the sign of the other stress tensor components is unchanged. Similarly, �yz D ��y0z0 , �zx D ��z0x0 and the other strain tensor components are unchanged. Now consider the normal stress on the x D x0 plane with the two coordinate systems. 24 2 Quick Review of Theories of Elastic Deformation Table 2.1 Coordinate transformation x y z . Ox/ . Oy/ .Oz/ x0 x0x x0y x0z . Ox0/ . Ox0 � Ox/ . Ox0 � Oy/ . Ox0 � Oz/ y0 y0x y0y y0z . Oy0/ . Oy0 � Ox/ . Oy0 � Oy/ . Oy0 � Oz/ z0 z0x z0y z0z . Oz0/ . Oz0 � Ox/ . Oz0 � Oy/ . Oz0 � Oz/ �x0x0 D C11�x0x0 C C12�y0y0 C C13�z0z0 C C14�x0y0 C C15�y0z0 C C16�z0x0 D C11�x0x0 C C12�y0y0 C C13�z0z0 C C14�x0y0 � C15�y0z0 � C16�z0x0 (2.61) �xx D C11�xx C C12�yy C C13�zz C C14�xy C C15�yz C C16�zx (2.62) As mentioned above, this coordinate transformation does not change the normal stress on the x-plane; �x0x0 D �xx . It follows that C15 D C16 D 0. Repeating the same argument for the symmetry about the yz and zx-plane, we can simplify the stiffness matrix as follows: ŒCkl D 0 BBBBBBB@ C11 C12 C13 0 0 0 C12 C22 C23 0 0 0 C13 C23 C33 0 0 0 0 0 0 C44 0 0 0 0 0 0 C55 0 0 0 0 0 0 C66 1 CCCCCCCA (2.63) Next, consider that the material is isotropic. In this case, the constitutive relation should remain the same if we rotate the material around an axis. First examine how the stress vector is transformed by coordinate transformation from O � xyz to O � x0y0z0. Referring to Eq. (2.43), � x 0 D x0x� x C x0y� y C x0z� z (2.64) where x0x etc are the direction cosine (see Table 2.1). Substituting Eqs. (2.44)– (2.46) into the right-hand side of Eq. (2.64), we obtain � x 0 D � x0x�xx C x0y�yx C x0z�zx Ox C � x0x�xy C x0y�yy C x0z�zy Oy C � x0x�xz C x0y�yz C x0z�zz Oz � x0i �ix Ox C x0i �iy Oy C x0i �izOz (2.65) Now by definition, �x0x0 is the Ox0 component of � x0 . �x0x0 D � x0 � � x0x Ox C x0y Oy C x0zOz 2.2 Hooke’s Law and Poisson’s Ratio 25 Table 2.2 Coordinate transformation x y z x0 cos � sin � 0 y0 � sin � cos � 0 z0 0 0 1 D x0x x0i �ix C x0y x0i �iy C x0z x0i �iz � x0j x0i �ij (2.66) Applying the same procedure as Eq. (2.66) to the other components, we find the general expression to relate the stress tensor components before and after the coordinate transformation as follows.6 �i 0j 0 D i 0i j 0j �ij (2.67) Now transform the coordinate system around the z-axis and make analysis on the constitutive relation. Referring to Table 2.2, �x0x0 D x0x x0x�xx C x0y x0x�yx C x0x x0y�xy C x0y x0y�yy D �xx cos2 � C �yx sin � cos � C �xy cos � sin � C �yy sin2 � D �xx cos2 � C �yy sin2 � C 2�xy cos � sin � �y0y0 D �xx sin2 � C �yy cos2 � � 2�xy cos � sin � �z0z0 D �zz �x0y0 D � �yy � �xx cos � sin � C �xy � cos2 � � sin2 � �y0z0 D �yz cos � � �zx sin � �z0x0 D �yz sin � C �zx cos � (2.68) Similarly, the strain tensor components are expressed with the new coordinate system as follows: �x0x0 D x0x x0x�xx C x0y x0y�yy C x0y x0x�yx C x0x x0y�xy D �xx cos2 � C �yy sin2 � C 2�xy cos � sin � �y0y0 D �xx sin2 � C �yy cos2 � � 2�xy cos � sin � �z0z0 D �zz �x0y0 D � �yy � �xx cos � sin � C �xy � cos2 � � sin2 � 6In Eq. (2.67) the order of the coefficients x0j x0i in Eq. (2.66) is switched for better visibility. 26 2 Quick Review of Theories of Elastic Deformation �y0z0 D �yz cos � � �zx sin � �z0x0 D �yz sin � C �zx cos � (2.69) Consider expressing �x0x0 in two different ways. First, express the stress tensor component on the right-hand side using the strain-tensor components �ij and the stiffness tensor components in Eq. (2.63). �x0x0 D �xx cos2 � C �yy sin2 � C2�xy cos � sin � D �C11�xx C C12�yy C C13�zz cos2 � C �C12�xx C C22�yy C C23�zz sin2 � C2C44�xy cos � sin � D �xx � C11 cos 2 � C C12 sin2 � C�yy � C12 cos 2 � C C22 sin2 � C�zz � C13 cos 2 � C C23 sin2 � C�xy .2C44/ cos � sin � (2.70) Next, express �x0x0 using �i 0j 0 and the stiffness tensor components in Eq. (2.63), and convert the strain component expression after the coordinate transformation with those of before the transformation, �ij . �x0x0 D C11�x0x0 C C12�y0y0 C C13�z0z0 D C11 � �xx cos 2 � C �yy sin2 � C 2�xy cos � sin � CC12 � �xx sin 2 � C �yy cos2 � � 2�xy cos � sin � CC13�zz D �xx � C11 cos 2 � C C12 sin2 � C�yy � C11 sin 2 � C C12 cos2 � C�zzC13 C�xy2 .C11 � C12/ cos � sin � (2.71) Compare Eqs. (2.70) and (2.71) for the coefficients of the same strain tensor matrix, �ij . In order for �x0x0 expressed in these two ways to be the same for any angle � , the following equalities are necessary for the coefficients of �yy , �zz, and �xy . From �yy C12 cos 2 � C C22 sin2 � D C12 cos2 � C C11 sin2 � (2.72) 2.2 Hooke’s Law and Poisson’s Ratio 27 From �zz C13 cos 2 � C C23 sin2 � D C13 (2.73) And from �xy 2C44 cos � sin � D 2 .C11 � C12/ cos � sin � (2.74) From Eqs. (2.73)–(2.74), it follows that C11 D C22 (2.75) C13 D C23 (2.76) C44 D C11 � C12 (2.77) Repeating the same procedure for coordinate rotations about the x and y axes, we find the following conditions for the case of isotropic materials. C11 D C22 D C33; C12 D C13 D C23; C44 D C55 D C66 D C11 � C12 (2.78) Since C11 D C22 D C33 are related to C44 D C55 D C66 in the last expression, Eq. (2.78) indicates that the stiffness tensor of isotropic materials has two independent components as follows: C12 D C13 D C23 � � (2.79) C44 D C55 D C66 � 2� (2.80) � and � are called as Lamé’s first and second constants, respectively. All these arguments simplify the stiffness tensor for isotropic materials into the form with the use of the two Lamé’s constants, and allows us to express the constitutive relation as follows: 0 BBBBBBB@ �xx �yy �zz �xy �yz �zx 1 CCCCCCCA D 0 BBBBBBB@ �C 2� � � 0 0 0 � �C 2� � 0 0 0 � � �C 2� 0 0 0 0 0 0 2� 0 0 0 0 0 0 2� 0 0 0 0 0 0 2� 1 CCCCCCCA 0 BBBBBBB@ �xx �yy �zz �xy �yz �zx 1 CCCCCCCA (2.81) Here the physical meaning of C44 is the stiffness that connect the shear stress �xy and shear strain �xy [Eq. (2.57)] as �xy D C44�xy D C44 2 � @�y @x C @�x @y � (2.82) 28 2 Quick Review of Theories of Elastic Deformation Fig. 2.9 Schematic illustration of Poisson ratio ezz eyy exx From Eq. (2.41), the shear stress can be related to the shear strain with the shear modulus (G). �xy D G � @�y @x C @�x @y � (2.83) From Eqs. (2.82) and (2.83), we find C44 D 2G. Next relate the Lamé’s constants to Young’s modulus and Poisson ratio. Consider a volume is subject to one-dimensional force in Fig. 2.9. As the volume is stretched along the x-axis, the material is usually compressed in the orthogonal direction. For the isotropic case, the compressions in the y and z direction are the same. The ratio is referred to as the Poisson’s ratio. D ˇˇ ˇˇ �t �n ˇˇ ˇˇ (2.84) Here �n is the normal strain in the direction of the applied, one-dimensional stress, and �t is the transverse strain. In the case of Fig. 2.9, �n D �xx and �t D �yy D �zz. Putting �xx D �n, �yy D �zz D �xy D �yz D �zx D 0, �xx D �n, �yy D �zz D �t , �xy D �yz D �zx D 0 in Eq. (2.81), we obtain �n D .�C 2�/�n C 2��t (2.85) 0 D ��n C 2.�C �/�t (2.86) From Eq. (2.86), D ˇˇ ˇˇ �t �n ˇˇ ˇˇ D � 2.�C �/ (2.87) Eliminating �t from Eqs. (2.85) and (2.86), �n D �.3�C 2�/ .�C �/ �n (2.88) Therefore, 2.3 Principal Axis 29 Table 2.3 Various moduli .�; �/ .E;G/ .E; / .G; / E �.3�C 2�/ .�C � E E 2.1C /G G � G E 2.1C / G � 2.�C �/ E � 2G 2G � � G.E � 2G/ 3G �E E .1C /.1� 2 / 2 G .1� 2 / E D �n �n D �.3�C 2�/ .�C �/ (2.89) Table 2.3 shows relation among different moduli. 2.3 Principal Axis Strain tensor Eq. (2.11) is symmetric and can be diagonalized by a rotational transformation of the coordinate system [7]. Once diagonalized, all the shear components of the resultant strain tensor become zero; �ij D 0; i ¤ j . This means that we can always find a coordinate system with which a given strain can be expressed in terms of normal strains only. This is understandable because when an infinitesimally small volume at a point is deformed, the cor- responding displacement vector has three-translational degrees of freedom, and stretch/compression is the differential displacement along the axes of these degrees of freedom. By aligning the coordinate axes to these three direction, we can always express the total strain as a combination of the three normal strains. Similarly, stress tensor Eq. (2.55) can be diagonalized, and this can be understood as follows. Whatever coordinate system you may chose, the stress tensor vector has three components. In other words, the stress tensor can be expressed as the summation of the three components. The three axes of the coordinate system corresponding to the diagonal tensor are called the principal axes. If we apply a diagonalized strain tensor to the constitutive equation (2.81), it is clear that the stress tensor corresponding to the left-hand side is also diagonal because all shear stress terms �ij ; i ¤ j become zero. This is not surprising because Hooke’s law states that stretch or compression is parallel to the external force causing it. Non-diagonal terms in the stiffness tensor such as C12 that connects a normal stress �xx and normal strain �yy in an orthogonal direction is not a direct consequence of Hooke’s law. It is not associated with a elastic modulus but instead due to the Poisson’s effect represented by the Poisson ratio . In other words, linear elastic deformation is an orientation preserving mapping [2]. 30 2 Quick Review of Theories of Elastic Deformation The procedure to make a diagonal tensor from a square tensor is known as diagonalization. Take a moment to review the procedure and discuss the physical meaning. Consider diagonalization of a strain tensor expressed in the symmetric form as follows: 0 @ �xx �xy �zx �xy �yy �yz �zx �yz �zz 1 A 0 @ dx1 dx2 dx3 dy1 dy2 dy3 d z1 d z2 d z3 1 A D 0 @ �1 0 0 0 �2 0 0 0 �3 1 A 0 @ dx1 dx2 dx3 dy1 dy2 dy3 d z1 d z2 d z3 1 A (2.90) Rewrite Eq. (2.90) in the following form 2 4 0 @ �xx �xy �zx �xy �yy �yz �zx �yz �zz 1 A � 0 @ �1 0 0 0 �2 0 0 0 �3 1 A 3 5 0 @ dx1 dx2 dx3 dy1 dy2 dy3 d z1 d z2 d z3 1 A D 0 (2.91) and consider the first column of Eq. (2.91). 0 @ �xx � �1 �xy �zx �xy �yy � �2 �yz �zx �yz �zz � �3 1 A 0 @ dx1 dy1 d z1 1 A D 0 (2.92) The physical meaning of Eq. (2.92) is as follows. If we multiply a strain matrix to the infinitesimal line element vector expressed in the x1y1z1 coordinate system, the resultant differential displacement vector associated with the strain has only a normal strain component; d�xx ¤ 0, d�xy D d�zx D 0. In order for this condition to be true for a nonzero line element vector, it is necessary that Eq. (2.92) is not solvable as a set of linear equations. This condition is expressed as the determinant of the matrix is zero. That is, det 0 @ �xx � �1 �xy �zx �xy �yy � �2 �yz �zx �yz �zz � �3 1 A D 0 (2.93) This leads to the following characteristic equation, �i ; i D 1; 2; 3 is one of the three roots. �3 � I1�2 C I2� � I3 D 0 (2.94) Here I1, I2, and I3 are known as the invariants defined as follows: I1 D �xx C �yy C �zz (2.95) I2 D �xx�yy C �yy�zz C �zz�xx � �2xy
Compartilhar