Structural Geology: The Mechanics of Deforming Metamorphic Rocks

Structural Geology

The Mechanics of Deforming Metamorphic Rocks

Volume I: Principles

Bruce Hobbs

Research Fellow, CSIRO, Adjunct Professor, Centre for Exploration Targeting, The University of Western Australia

Alison Ord

Winthrop Research Professor, Centre for Exploration Targeting, The University of Western Australia.

Table of Contents

Cover image

Title page

Copyright

Dedication

Preface

Acknowledgements

Chapter 1. Introduction

1.1. The Multiscale Nature of Deformation and Metamorphism

1.2. Mechanics, Processes and Mechanisms

1.3. Linear and Nonlinear Processes

1.4. Wavelet Analysis

1.5. Notation, Conventions and Units

Section A. The Mechanics of Deforming Solids: Overview of Section A

Introduction

Chapter 2. Geometry: The Concept of Deformation

2.1. Deformations

2.2. Distortion and Rotation

2.3. Deformation and Strain Tensors and Measures

2.4. Distortion and Volume Change

2.5. Example 1: The Geometry of a Simple Shear deformation

2.6. Pseudo Phase Portraits for Affine Deformations

2.7. Example 2: Non-affine Deformations

2.8. The Deformation Arising from Slip on a Single Plane

2.9. Incremental Strain Measures

2.10. Compatibility of Deformations

Chapter 3. Kinematics – Deformation Histories

3.1. The Movement Picture and the Material Derivative

3.2. Velocity and Velocity (or Flow) Fields

3.3. Other Measures of the Kinematics

3.4. Rate of Change of Deformation Measures

3.5. An Example of a Non-affine Flow

3.6. Kinematic Indicators and Flow Fields

3.7. Some Important Relations and Theorems

Chapter 4. The Balance Laws: Forces Involved in Deformation

4.1. General Statement

4.2. Conservation of Mass

4.3. Balance of Linear Momentum

Chapter 5. Energy Flow – Thermodynamics

5.1. What is Thermodynamics?

5.2. Metamorphic Systems

5.3. Thermodynamic Systems

5.4. Four Different Strands of the Development for Thermodynamics

5.5. Summary of the Non-equilibrium Framework

5.6. State Variables and Internal Variables

5.7. The Laws of Thermodynamics

5.8. The Potentials: Helmholtz and Gibbs Energy for Deforming Solids

5.9. The Dissipation for a Material with Internal Variables

5.10. The Thermo-mechanical Heat Balance Equation

5.11. Entropy Production and System Constraints

5.12. Convex and Non-convex Potentials: Minimisation of Energy

Chapter 6. Constitutive Relations

6.1. What is a Constitutive Relation?

6.2. Why are Constitutive Relations Important?

6.3. The Four Basic Forms of Constitutive Behaviour

6.4. The Rules for Constructing Constitutive Relations

6.5. Elastic Behaviour

6.6. Plastic Behaviour

6.7. Viscous–Plastic–Elastic Behaviour

6.8. Fluids and Solids

Chapter 7. Nonlinear Dynamics

7.1. Introduction

7.2. Patterns in Metamorphic Rocks

7.3. Some Archetype Examples

7.4. Linear Stability Analysis

7.5. Classification of Instabilities

7.6. Bifurcations

7.7. Energy Minimisation and the Growth of Fractal Structures

7.8. Fractals and Multifractals

Section B. Processes Involved in the Development of Geological Structures: Overview of Section B

Introduction

Chapter 8. Brittle Flow

8.1. Introduction: What are the Issues and Problems? What do we mean by Brittle and Ductile?

8.2. The Mechanics of Pattern Formation during Deformation

8.3. The Geometry and Physics of Fracture

8.4. A Kinematic View of Fracture Development

8.5. Accommodation Mechanisms

8.6. Breakage Mechanics

Chapter 9. Visco-Plastic Flow

9.1. Introduction

9.2. The Mechanics of Crystal Plastic Flow

9.3. Deformation Mechanisms in Polycrystalline Aggregates

9.4. The Geometry of Crystal Deformation

9.5. The Formation of Subgrains and Deformation by Grain Boundary Migration

9.6. Coupled Grain Boundary Migration during Deformation and Metamorphism

9.7. Constitutive Equations

Chapter 10. Damage Evolution

10.1. Phenomenology – What is Damage?

10.2. A Thermodynamic Theory of Damage

10.3. Ductile Damage

10.4. Criticality and Phase Transitions

Chapter 11. Transport of Heat

11.1. Fourier's Law of Heat Conduction: Diffusion of Heat

11.2. Temperature and Pressure Dependence of Conductivity

11.3. Equations of Heat Transport: Timescales

11.4. Internal Heat Production

11.5. Tectonic Models of Thermal Evolution

11.6. The Thermodynamics of Heat Conduction, Thermal Expansion and Entropy Production

Chapter 12. Fluid Flow

12.1. Fluid Flow and Deformation/Metamorphism

12.2. Types of Fluid Flow

12.3. Drivers of Fluid Flow

12.4. Focussing of Fluid Flow

12.5. Convective Flow

12.6. The Thermodynamics of Fluid Flow. Entropy Production. Bejan's Postulate

12.7. Overview of Crustal Plumbing Systems

Chapter 13. Microstructural Rearrangements

13.1. Introduction. Why is Microstructure Important?

13.2. Normal Grain Growth. Grain Size Distributions and Grain Shapes in Single Phase Aggregates; No Deformation

13.3. Grain Relationships in Polyphase Aggregates

13.4. Grain Size Growth and Reduction during Deformation

13.5. CPO Development

13.6. Mechanisms of Dynamic Recrystallisation

13.7. Anisotropy and the Influence of Elastic Anisotropy and Flow Stress Anisotropy on Localisation

13.8. Controls on Microstructure

Chapter 14. Mineral Reactions: Equilibrium and Non-Equilibrium Aspects

14.1. Introduction

14.2. Some Preliminaries

14.3. Deforming Metamorphic Systems: A Chemical Equilibrium View

14.4. Mineral Reactions: Systems Not at Equilibrium

14.5. Chemical Dissipation

14.6. Synthesis

Chapter 15. Models for Mineral Phase Nucleation and Growth

15.1. Introduction

15.2. Some Detailed Models of Metamorphic Reactions

15.3. Homogeneous and Heterogeneous Nucleation

15.4. Stress-Assisted Transport

15.5. Stresses Generated by Mineral Growth

15.6. An Example: Porphyroblast Formation

15.7. Synthesis

Epilogue

Appendix A. Commonly Used Symbols

Appendix B. Vectors, Tensors and Matrices

Appendix C. Some Useful Mathematical Concepts and Relations

References

Index

Copyright

Elsevier

Radarweg 29, PO Box 211, 1000 AE Amsterdam, Netherlands

The Boulevard, Langford Lane, Kidlington, Oxford, OX5 1GB, UK

225 Wyman Street, Waltham, MA 02451, USA

Copyright © 2015 Elsevier Inc. All rights reserved.

No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions.

This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein).

Notice

Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary.

Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility.

To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein.

ISBN: 978-0-12-407820-8

Library of Congress Cataloging-in-Publication Data

Hobbs, Bruce E.

Structural geology : the mechanics of deforming metamorphic rocks / Bruce E. Hobbs, Alison Ord.

pages cm

ISBN 978-0-12-407820-8 (hardback)

1. Metamorphic rocks. 2. Rock deformation. 3. Rock mechanics. I. Ord, Alison, 1955. II. Hobbs, Bruce, 1936. III Title.

QE475.H63 2015

552'.4--dc23

2014033461

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library

For information on all Elsevier publications visit our web site at http://store.elsevier.com/

Dedication

Dedicated to:

Susan Therese Plant (1964–2013)

Evelyn Ethel Ord (1919–2014)

Smile because she lived.

Preface

This volume discusses the processes that operate during the deformation and metamorphism of rocks in the crust of the earth with the goal of understanding how these processes control or influence the structures that we observe in deformed metamorphic rocks. The deformation and metamorphism of rocks involves structural rearrangements of elements of the rock by processes such as mass diffusion, dislocation slip and climb, disclination and disconnection motion, grain-boundary migration and fracturing at the same time as chemical reactions proceed. In some instances infiltrating fluids introduce or remove chemical components and may influence mechanical properties through changes in volume, fluid pressure or temperature. At the same time, heat is added or removed from the system depending on the surrounding tectonic environment and the processes operating within the deforming rock mass. The intent is to consider these processes within a framework set by both the individual mechanisms of deformation and the fundamental mechanics that describe and relate these mechanisms. Such mechanisms do not operate independently of each other but are strongly coupled so that each process has strong feedback influences on the other leading to structures and mineral assemblages that do not develop in the uncoupled environment. This leads to nonlinear behaviour and so a goal of this book is to treat deforming metamorphic systems as nonlinear dynamic systems. The general system we consider in this book is a deforming, chemically reactive system in which fluid and thermal transports play significant roles. The volume concentrates on the principles that govern the development of deformed metamorphic rocks. A second volume deals with applications of these principles.

Over the past 30 years there have been dramatic developments in structural geology, metamorphic petrology, physical metallurgy, nonlinear chemical dynamics, continuum mechanics, nonlinear dynamics and thermodynamics that are relevant to processes that operate within the Earth; but these developments have largely evolved independently of each other and many have not been incorporated into mainstream metamorphic geology. This book attempts to integrate aspects of these developments into a common framework that couples the various processes together.

In this volume we first develop the basic groundwork in Mechanics where thermodynamics plays a fundamental role and then move to apply these principles to rocks where deformation, fluid flow, thermal transport, mineral reactions, damage and microstructural evolution contribute to the evolution of the fabric of the rock mass, and produce the structures we observe.

The processes that operate fall into five inter-related categories: (1) mechanical processes, (2) mineral reactions, (3) fluid flow, (4) transport of heat and (5) microstructural adjustments. A goal of this book is to treat all five of these categories under the same umbrella and integrate these processes within a single framework. The basis of this framework lies in thermodynamics. This is not a book about thermodynamics but it uses thermodynamics as an important tool. What do we mean by thermodynamics? By thermodynamics we mean: the study of the flow of physical and chemical quantities through a system under the influence of thermodynamic forces. The physical and chemical quantities involve momentum (per unit area), fluids, heat and mass; the corresponding thermodynamic forces (an unfortunate but useful term) are gradients in deformation, the inverse of the temperature, hydraulic and chemical potentials.

This volume is arranged into two parts. Both parts begin with an overview of the section. Part A is a discussion of modern mechanics and includes chapters devoted to the geometry, kinematics, nonlinear dynamics and thermodynamics of deforming systems. We follow with a consideration of common constitutive relations including non-steady state evolutionary processes such as hardening and softening and the development of anisotropy. In particular, in Section A we introduce the basic principles that govern the nonlinear behaviour of deforming, chemically reactive systems. This represents a significant departure from current treatments of the mechanics of geological materials.

Part B considers the common processes involved in the development of geological structures including brittle and visco-plastic flow, heat and fluid flow, damage evolution, microstructural rearrangements such as subgrain formation and rotation and the nonlinear kinetics of mineral reactions. We take the opportunity to introduce the concepts of disclinations and disconnections which, along with dislocations, are now recognised as playing important roles in deformation and grain growth/reduction processes during the deformation of polycrystalline materials. Of greater importance is the concept of coupled grain-boundary migration in which grain- (or subgrain) boundary migration is coupled to a shearing deformation parallel to the moving boundary. This process has a profound influence on the development of metamorphic fabrics and the concept has a great unifying influence on understanding their evolution.

Of particular importance is the notion of critical systems of various kinds where the Helmholtz energy passes from being a convex to a nonconvex function as a function of some critical parameter or forcing. In order for a deforming system to minimise such energy functions and remain compatible with the imposed deformation, the system cannot remain homogeneous and forms structure at finer and finer scales. This behaviour is fundamental to rock deformation and is responsible for most geological fabrics including jointing, localised folding and boudinage and rotation recrystallisation.

Volume II considers the common geological structures such as foliations, lineations, folds and fracture systems and extends the discussion both to the regional scale and to open flow systems such as hydrothermal systems. In particular, the concept of localised structure development is emphasised in the light of progress in nonlinear mechanics over the past 30 years; this enables the fundamental work of Biot to be seen within a generalised framework. We regard Biot as the father of the application of thermodynamics to deforming, chemically reacting systems.

Both volumes assume a basic knowledge of structural geology and metamorphic petrology and so some familiarity with the books quoted below is necessary. A short list of useful additional references for in-depth reading is given at the end of each chapter. Although there is an emphasis on a mathematical treatment, the mathematics is kept to an elementary level and a basic knowledge of differential calculus and tensor algebra suffices to understand the book. Short appendices on both these subjects are included. The volume is meant to be read by graduate students and research workers in metamorphic geology. The intent also is to supply a vocabulary to enable readers to follow some of the more involved papers on the mechanics of deforming, metamorphic rocks that are appearing with greater frequency.

In an interdisciplinary book such as this, we have found it too demanding to stick to a unique mathematical notation throughout. We have attempted to standardise on commonly used quantities (Appendix A) but for individual examples have mainly used the notation of the original authors of the example considered. Such terms are always defined where they are first used. Equally, for didactic reasons, we have introduced concepts and principles where they are of the greatest use and not where they would be introduced in a standard treatment of the subject. This leads to some repetition and sometimes an ordering of material that is not sequential. Each chapter is largely self-contained. We trust that readers will report errors to one or both of us at bruce.hobbs@csiro.au and alison.ord@uwa.edu.au.

Much of the content is very much in progress and many questions remain open but we hope it will inspire researchers, especially the younger breed, who more commonly already have a grasp of the concepts involved, to question and revisit entrenched wisdom and especially revisit natural examples in the field and at the microscale where most of the fascinating and interesting problems are yet to be solved.

Bruce Hobbs

Alison Ord,     Fremantle, Western Australia, August, 2014.

Further Reading

Fossen H. Structural Geology. Cambridge University Press; 2010.

Hobbs B.E, Means W.D, Williams P.F. An Outline of Structural Geology. John Wiley; 1976.

Jaeger J.C. Elasticity, Fracture and Flow. Methuen; 1969.

Passchier C.W, Trouw R.A.J. Microtectonics. Springer; 1996.

Ramsay J.G. Folding and Fracturing of Rocks. McGraw-Hill; 1967.

Turner F.J, Weiss L.E. Structural Analysis of Metamorphic Tectonites. McGraw-Hill; 1963.

Twiss R.J, Moores E.M. Structural Geology. W. H. Freeman; 2006.

Vernon R.H. A Practical Guide to Rock Microstructure. Cambridge University Press; 2004.

Vernon R.H, Clarke G. Principles of Metamorphic Petrology. Cambridge University Press; 2008.

Yardley B.W.D. An Introduction to Metamorphic Petrology. Longman; 1974.

Acknowledgements

We thank the following for their influence on our evolving thought processes and for their contributions to a nonlinear view of deformation and metamorphism. Some may not recognise that they made such a contribution, but we acknowledge them for sharing with us their approach to mechanics and of deformation during metamorphism. Thanks to: Elias Aifantis, Mike Brown, John Christie, Peter Cundall, David Griggs, Peter Hornby, Giles Hunt, Win Means, Hans Muhlhaus, Gerhard Oertel, Mervyn Paterson, Roger Powell, Klaus Regenauer-Lieb, Ioannis Vardoulakis, Ron Vernon, Vic Wall, Paul Williams, David Yuen and Chongbin Zhao. We would also like to acknowledge the encouragement, patience and help of Elsevier staff John Fedor, Louisa Hutchins and Sharmila Vadivelan.

Chapter 1

Introduction

Abstract

In this introductory chapter we present the basic framework and approach for the remainder of the book. We are concerned with the processes that operate during the deformation and metamorphism of the rocks that comprise the crust and uppermost mantle of the Earth and the mechanical principles that govern the behaviour and resulting geometry of these rocks. The processes operate over a vast range of spatial and temporal scales and are strongly coupled so that each can have a strong feedback influence on the others depending on these scales. Such coupling leads to nonlinear behaviour and the resulting complexity of structure and associated mineral fabrics we observe in the field. The basis for understanding this complexity lies in the thermodynamics of systems far from equilibrium and the nonlinear dynamics of such systems. A convenient technique for analysing this complexity across scales is wavelet analysis.

Keywords

Complexity; Feedback; Mechanics; Metamorphism; Multiscale dynamics; Non-equilibrium; Nonlinear dynamics; Structural geology; Thermodynamics; Wavelet analysis

Outline

1.1 The Multiscale Nature of Deformation and Metamorphism 4

1.2 Mechanics, Processes and Mechanisms 7

1.3 Linear and Nonlinear Processes 10

1.4 Wavelet Analysis 18

1.5 Notation, Conventions and Units 20

Recommended Additional Reading 20

General Statement. The aim of this book is to discuss the processes that operate during the deformation and metamorphism of rocks in the crust and upper lithosphere of the Earth with the goal of understanding how these processes control or influence the structures that we observe in the field. The intent is to consider these processes within a framework set both by the individual mechanisms of deformation and the fundamental mechanics that describe and relate these mechanisms. Such mechanisms do not necessarily operate independently of each other but may be strongly coupled so that each process has strong feedback influences on the others leading to structures and mineral assemblages that do not develop in the uncoupled environment. The general system we consider in this book is a deforming chemically reactive system in which fluid and thermal transport play significant roles.

Over the past 30  years there have been dramatic developments in structural geology, metamorphic petrology, physical metallurgy, nonlinear chemical dynamics, continuum mechanics, nonlinear dynamics and non-equilibrium thermodynamics that are relevant to processes that operate within the Earth but these developments have largely evolved independently of each other and many have not been incorporated into mainstream metamorphic geology. This book attempts to integrate aspects of these developments into a common framework that couples the various processes together.

Although there is a tendency to separate structural geology from metamorphic petrology in text books, we attempt to bring these two subjects together in this book under the name metamorphic geology. We are not too concerned with mineral phase equilibria but treat the processes that give rise to mineral assemblages and fabrics from the viewpoint that such processes commonly contribute to the production of the structures and patterns we see in deformed metamorphic rocks. We use the term fabric to mean the internal spatial ordering of geometrical, physical and mineralogical elements in deformed metamorphic rocks (Cf., Turner and Weiss, 1963, p. 19; Vernon, 2004, p. 7). The term structure means the assemblage of folds, boudins, foliations, lineations, grain shapes, mineral phases and so on that make up such rocks. At a microscopic scale the microfabric consists of the grain shapes, grain arrangements and spatial orientations and patterns of grain distributions. We use the term crystallographic preferred orientation (or just CPO) to describe the patterns of lattice orientations of grains. An important additional concept is that of patterns. We will see that the behaviour of nonlinear systems is characterised by the formation of patterns of various kinds (Cross and Greenside, 2009). These patterns are typified by metamorphic differentiation, the distribution of porphyroblasts and mineral lineations defined by mineral segregations to name a few. Just as important as the patterns are defects in the patterns and we will discuss such features in Chapter 7.

At least while these processes are operating, the system dissipates energy and hence is not at equilibrium. Thus, the overarching concepts that unify these various processes are grounded in the thermodynamics of systems not at equilibrium. The traditional approach in metamorphic geology is strongly influenced by the work of Gibbs where, for the most part, equilibrium is assumed and so approaches based on departures from equilibrium have been largely neglected or even dismissed as irrelevant in the Earth Sciences, the argument being that geological processes are so slow that equilibrium can be assumed. There has been an additional important deterrent to progress in that developments in non-equilibrium thermodynamics have been dispersed across a large number of disciplines and languages, with apparently conflicting propositions put forward and sterile discussions devoted to the relative merits of minimum and maximum entropy production principles and the so-called Curie principle (Truesdell, 1966). The result has been difficulty in bringing forward a unified approach to the subject as far as Earth scientists are concerned.

In this book we first develop the basic groundwork in mechanics where thermodynamics plays a fundamental role and then move to apply these principles to rocks where deformation, fluid flow, thermal transport, mineral reactions, damage and microstructural evolution contribute to the evolution of the rock mass and produce the structures we observe. The processes that operate fall into five interrelated categories: (i) mechanical processes such as diffusion, dislocation motion, twinning, grain-boundary sliding and fracturing that contribute directly to the deformation of the material; (ii) chemical reactions that produce changes in mineralogy, volume changes and grain-rearrangements and consume or produce heat; these processes influence the mechanical properties both by the generation of heat and of new weak or strong mineral phases and the ways in which these phases are spatially related to each other; moreover chemical reactions can act as a deformation mechanism in their own right; (iii) fluid flow which not only introduces or removes chemical components such as (OH)−, H+ and CO2 but can also be an agent for dissolution and deposition – processes that also contribute to the deformation; (iv) transport of heat by conduction or by advection in a moving fluid that leads to temperature changes that in turn can influence the rates of mineral reactions and of deformation; (v) structural adjustments such as subgrain formation, recrystallisation, preferred orientation development (in the form of both crystallographic axes and grain-shape), metamorphic differentiation, the development of foliations and lineations, folding, boudinage and localised shear zones.

A goal of this book is to treat all five of these categories under the same umbrella and integrate these processes within a single framework. The basis of this framework lies in thermodynamics. This is not a book about thermodynamics but it uses thermodynamics as an important tool. What do we mean by thermodynamics? By thermodynamics we mean the study of the flow of physical and chemical quantities through a system under the influence of thermodynamic forces. The dominant physical and chemical quantities of interest to metamorphic geologists are momentum (per unit area), heat, fluids and mass; the corresponding thermodynamic forces are gradients in the deformation, the inverse of the temperature, hydraulic potential and chemical potential. If there are no flows the system is at equilibrium.

All of the processes mentioned above produce or consume entropy while they operate. So thermodynamics can also be defined as the study of the production of entropy. We use the concept of entropy in a very simple manner. By entropy we mean the amount of heat produced at an instant at a point in the body divided by the current absolute temperature at that point. Most deformation processes such as plastic flow or fracturing produce entropy, whereas chemical reactions may produce or consume entropy. Even the flow of fluids through fractures or porous media produces entropy. The second law of thermodynamics tells us that for a given system the sum of all the individual entropy production rates is greater than zero (or equal to zero at equilibrium). This means that the processes that operate to constitute a deforming chemically reacting system do not occur independently of each other but are coupled through the second law of thermodynamics to ensure that the total entropy production is greater than zero. Such a simple overarching law forms the basis for the unifying framework we seek for deformed metamorphic rocks.

In this book we introduce the basic principles that govern the nonlinear behaviour of deforming chemically reactive systems. This represents a significant departure from current treatments of the mechanics of geological materials. Of particular importance here is the notion of critical systems where the Helmholtz energy passes from being a convex to a non-convex function as a function of some critical parameter. This behaviour is responsible for a large number of geological structures including localised folding and boudinage and rotation recrystallisation. In particular the concept of localised structure development is emphasised in the light of progress in nonlinear mechanics over the past 30  years; this represents an extension of the fundamental work of Biot who we regard as the father of the application of thermodynamics to deforming chemically reactive systems through publication of two seminal papers: Biot (1955, 1984).

This introductory chapter discusses a number of concepts that may be unfamiliar to some metamorphic geologists but which are fundamental tenets of the approach adopted in this book:

• The multiscale nature of metamorphic geology.

• The coupled nature of most processes that operate during deformation and metamorphism.

• The need to describe and analyse the processes involved and their interactions in terms of non-equilibrium concepts.

• The distinction between linear and nonlinear systems and the implications of these two different modes of behaviour for the development of the structures and metamorphic fabrics observed in deformed metamorphic rocks.

• The use of wavelets as an overarching technique that enables structures and patterns in deformed rocks to be quantified across a range of length scales and characterised in terms of fractal geometrical concepts.

1.1. The Multiscale Nature of Deformation and Metamorphism

This book is concerned with the processes that operate during the deformation and metamorphism of rocks in the crust and upper lithosphere of the Earth. The deformation and metamorphism of rocks involves structural rearrangements of elements of the rock at the microscale by processes such as mass diffusion, dislocation slip and climb, grain-boundary migration and fracturing at the same time as chemical reactions proceed. In some instances infiltrating fluids introduce or remove chemical components and may influence mechanical properties through changes in mineralogical composition, fluid pressure or temperature. At the same time, heat is added to or removed from the system depending on the surrounding tectonic environment. Such mechanisms do not operate independently of each other but are coupled so that each process has feedback influences on the others leading to structures and mineral assemblages that do not develop in the uncoupled environment. Some of the interrelations between these processes are illustrated in Figure 1.1.

Figure 1.1   Coupling and feedback relations for some of the processes that operate during the development of deformed metamorphic rocks.

For instance deformation can induce a pore volume change which in turn produces a change in fluid pressure which alters the effective stress that then influences the deformation. As another example, a chemical reaction can produce fluid and/or a change in temperature both of which induce changes in pore fluid pressure with a consequent change in the deformation. Again, changes in temperature arising from deformation and/or chemical reactions can influence physical properties such as the viscosity or coefficient of friction which in turn influences the deformation. Many feedback relations exist in deforming chemically reacting rock masses and our aim in this book is to discuss the mechanics of these feedback relations and integrate them within a unified framework that describes the formation of the common structures and mineral fabrics we see in deformed metamorphic rocks.

One of the outstanding characteristics of metamorphic processes is the vast range of spatial and temporal scales involved (Figure 1.2). Thus at the atomic scale the forces between atoms need to be modelled with length scales of 10−⁹  m involving timescales for atomic vibrations of 10−¹⁵  s. Such modelling is becoming common for understanding dislocation dynamics at geological strain rates (Cordier et al., 2012) and some progress has been made in understanding metamorphic mineral reactions (Lasaga and Gibbs, 1990; Gale et al., 2010). Towards the other end of the spectrum is the timescale for regional metamorphism and melting which can be ≈100 my (Brown, 2010; Smithies et al., 2011) and involve length scales of 1000  km. The length scales involved in metamorphic geology span ≈15 orders of magnitude and the timescales span ≈32 orders of magnitude.

Such huge spans in length and timescales have advantages and disadvantages. The advantages lie in the opportunity to uncouple processes operating at different scales thus simplifying the mathematical treatment of the processes involved. For instance we will show in Volume II that for a shear zone 1  km thick deforming at a slow shear strain rate of 10−¹³  s−¹ the heat dissipated by deformation does not have time to be conducted out of the system and the shear zone heats up thus influencing the deformation and the rates of mineral reactions within the shear zone. For these conditions the thermal, chemical and deformation processes cannot be considered to be independent. On the other hand if the shear zone is 1  m thick then the heat can diffuse out of the shear zone and the deformation can be considered isothermal. Here the thermal and deformation processes can be uncoupled and considered independently of each other. On the other hand if the deformation is fast, say 10−¹  s−¹ corresponding to a slow aseismic event, then the thin shear zone also heats up and thermal, chemical and deformation processes need to be coupled.

The disadvantages arising from the huge spans in spatial and timescales involved lie in the present limitations of modern computers and the ways we use them. If we want to model, using finite elements, the deformation of the lithosphere say 150  km thick and 1000  km wide in two dimensions with a resolution of 10  m we need ≈1.5  ×  10⁹ finite elements. This is very close to what is possible with a modern computer. If we needed to do the computation in three dimensions the problem is not possible today. However, 10  m is still large compared to the processes such as diffusion and dislocation motion that take place at the atomic to micron scales. The solution adopted is called a multiscale or homogenisation approach (Tadmor and Miller, 2011) wherein the results of modelling particular processes at one scale are used to develop a mathematical description of that process which effectively homogenises the details of the processes operating at that scale. This description is then used to model detailed behaviour at the next scale up and so on. Thus dislocation dynamics modelled at the micron scale can be used to develop a mathematical constitutive relation that links the strain rate to the applied stresses. This relation can then be used to model details of processes at the metre scale and above (Cordier et al., 2012). This upscaling approach is shown in Figure 1.2(a) by the blue arrows. On the other hand conditions established by geological observations at a large length scale can be used as boundary conditions for processes operating at smaller length scales. This downscaling procedure is shown by the red arrows in Figure 1.2(a).

Figure 1.2   Spatial and temporal scales associated with deformation and metamorphism in the upper lithosphere of the Earth. (a) Spatial scale variation leading to a multiscale approach to modelling such systems. Images in order clockwise from lower left are from Zhu et al., (2005), Patrick Cordier, Ron Vernon, Bruce Hobbs, Jean-Pierre Burg, Catherine Spaggiari and Weronika Gorczyk. (b) Timescales associated with some metamorphic processes.

While these various processes are operating, the system dissipates energy (that is, the system produces or consumes heat) and hence is not at equilibrium. This statement is true no matter what the length and timescales are. As long as dissipative chemical and physical processes are operating the system, by definition, is not at equilibrium. We will show in Chapter 5 and Volume II that for some important metamorphic systems equilibrium is never attained no matter how slow the processes are. We will see that this is particularly true if the system is open so that fluids are entering or leaving the system; examples of such situations are H2O or H+ entering a retrograding shear zone, SiO2 bearing fluids leaving a schist undergoing metamorphic differentiation or melt leaving an anatexite. Thus, the overarching concepts that unify these various processes are grounded in the thermodynamics of systems not at equilibrium.

In this book we first develop the basic groundwork in mechanics where thermodynamics plays a fundamental role and then move to apply these principles to rocks where deformation, fluid flow, thermal transport, mineral reactions, damage and microstructural evolution contribute to the evolution of the rock mass and produce the structures we observe. The overarching principle that unites all these process is the second law of thermodynamics which, in a form where the mathematical detail of the processes involved is spelt out, is commonly called the Clausius–Duhem relation (Truesdell, 1966).

1.2. Mechanics, Processes and Mechanisms

Mechanics is the study of the ways in which physical and chemical systems respond to imposed forces or displacements at the boundaries of the system; it forms the content of section A: The Mechanics of Deforming Solids of this book. The response is strongly influenced by the nature of the material, the chemical reactions taking place and the physical environment defined by the temperature, deformation rate and whether the deforming system is open or closed with respect to transport of mass including fluids. Deformed metamorphic rocks display a vast array of structures such as dislocations, subgrains, recrystallised microstructures, folds, foliations, mineral lineations, boudins, metamorphic layering, porphyroblasts, leucosomes, localised shear zones and fractures. The characteristic of most of these structures is that they formed during the deformation of a chemically reacting polycrystalline solid. In fact the fundamental common characteristic of all of these structures is that the material comprises grains with well-defined crystal structures for most if not all of its evolution. Thus the material resembles a deforming metal rather than a viscous syrup that possesses no structure or a viscous polymer that may develop some of these structures but with no inherent granular structure where the grains have well-defined crystal structures. Thus the response of deforming metamorphic rocks is more akin to that of a solid rather than that of a fluid. From the point of view of mechanics, what do we mean by a solid rather than a fluid?

The common response to this question (Rice, 1993) is that a solid can support a shear stress over finite periods of time, whereas a fluid cannot. A fluid will flow under the influence of shear stresses but a solid may remain undeformed (except for elastic deformations) for periods of time that are long compared to the time involved in the observations or experiments being conducted. We discuss other important differences in Chapter 6. This difference has lead to the development of two different streams of mechanics known as solid and fluid mechanics. These two streams utilise different mathematical tool boxes and have different conceptual approaches. It has been, and still is, common for deforming rocks to be treated as fluids because often the concepts involved and the mathematical treatments are simpler than dealing with solids. Thus the dominant part of the theory of folding in the geological literature is based on fluid mechanics (Johnson and Fletcher, 1994; Ramberg, 1963; Smith, 1977) as are also almost all models of convection within the mantle of the Earth (Davies, 2011). However, the fact that the concepts and mathematics of fluids (especially that of ideal gases or linear fluids) are commonly simpler than dealing with solids is no excuse for proposing that metamorphic rocks actually are fluids or ideal gases. In this book we treat metamorphic rocks as solids and develop the theory of metamorphism within the context of solid mechanics.

The fundamental differences between a fluid and a solid are illustrated in , and so also depends on the mechanical properties of the solid, being different for different solids. For the solid in Figure 1.3(b) deformation is driven locally by the stress difference, (σ1  −  σ2). Elastic solids and fluids always deform elastically even for small boundary forces but elastic–plastic solids only deform inelastically (that is, permanently) once a critical stress (the yield stress) is exceeded. Thus for a fluid the pressure within the fluid is imposed by the external normal forces; for a solid the pressure is defined internally by the mean stress.

Figure 1.3   The difference between a fluid (a) and a solid (b). In both (a) and (b) the external force field is identical with normal forces NF1 and NF2 and shear forces SF1 and SF2. For any orientation of a fluid element in (a) the element feels only the hydrostatic pressure P. For a particular orientation of the solid element in (b) the element feels unequal normal stresses σ1 and σ2 and the pressure within the element is defined as the mean of these normal stresses.

Processes of deformation: As we have seen the processes that operate during deformation and metamorphism fall into five interrelated categories: (i) mechanical processes; (ii) chemical reactions; (iii) fluid flow; (iv) transport of heat by conduction or by advection; and (v) structural adjustments at the subgrain or grain scale.

Mechanisms of deformation: The term mechanism is used here as a subset of the term process and the distinction between the terms is to some extent scale dependent. As an example, a deformation process may be solid state diffusion but the mechanisms involved may be vacancy motion, interstitial motion or some more complicated set of cooperative atomic motions. Similarly the deformation process might be dislocation creep but the mechanisms involved may be dislocation glide, athermal kink drift, climb, interaction of individual dislocations with the forest dislocations or more complicated dislocation, dislocation array, impurity or vacancy interactions. Or the deformation process may be crack propagation where the deformation mechanism is diffusion of impurity atoms to the crack tip and the accumulation of dislocations. The term deformation process is a more general term than deformation mechanism and a number of deformation mechanisms may act together to produce a given deformation process. The deformation process generally operates at larger length scales than the contributing deformation mechanisms. Similarly a chemical reaction may be written as a process, A  +  B  →  C, but may consist of a number of coupled mechanisms such as A  →  X, B  →  2Y, X  +  2Y  →  C. Again, at a length scale large with respect to the pore scale, fluid flow in a porous medium may be described by a process governed by Darcy's law but at the pore scale the mechanism is described by Stokes' law (Coussy, 2010; Phillips, 1990).

For the most part little direct knowledge is available on the mechanisms that operate during the deformation and metamorphism of rocks. Analogues are continuously drawn with what is known in metals and physical chemistry but the necessity for experiments at elevated temperatures and pressures makes experimental observations on mechanisms within mineral systems very difficult. If one adds the additional complexity of making observations at geological strain rates then the problem becomes intractable. We can of course make experimental observations on the processes involved and try to extrapolate this information to geological conditions but in general the direct establishment of mechanisms is very difficult. A way forward that is growing in momentum is the computer simulation of mechanisms and the establishment of the energy necessary for that mechanism to operate under geological conditions. Examples include pioneer work by Lasaga (Xiao and Lasaga, 1996) on the dissolution of quartz, models of the molecular structure of water at quartz–quartz grain boundaries (Adeagbo et al., 2008), models of the dislocation structure of olivine (Castelnau et al., 2010), models of the dissolution of carbonates (Gale et al., 2010), models of the influence of water on the stress solubility of quartz (Zhu et al., 2005) and models of ‘power-law creep’ of MgO at geological strain rates and mantle pressure–temperature (P–T) conditions (Amodeo et al., 2011; Cordier et al., 2012). These studies are shedding new light on the mechanisms of deformation and metamorphism. We present two examples here, namely, that of the stress solubility of quartz in the presence of H2O (Zhu et al., 2005; Figure 1.4) and the development of dislocation structures associated with crack development in a face-centred cubic (FCC) crystal (Abraham, 2003; Figure 1.5)

1.3. Linear and Nonlinear Processes

A stroll through any deformed metamorphic terrain will soon convince an observer that there is a bewildering array of structures developed in such rocks ranging from folds with an endless variety of shapes and sizes to layering and lineations developed solely by metamorphic processes to fractures and veins that pre- and postdate the metamorphism (Figure 1.6).

In addition a large array of different metamorphic mineral fabrics exists including simple foliations, crenulated foliations, porphyroblasts and foliations/lineations defined by melting processes. Relevant questions are: Are there guiding principles behind all this complexity or does each structure form independently of each other? If there are guiding principles what are they and what causes the complexity? The aim of this book is to explore these questions and to propose a systematic basis for the complexity. Part of the answer to these questions lies in the nonlinear response of deforming chemically reacting solids to the deformation conditions and it is important from the outset to understand the implications of such nonlinear behaviour.

A linear system is one that behaves in such a way that the response is proportional to the input. This means that small changes to the input result in small changes to the response. Examples are linear (Hookean) springs that extend in proportion to the load on the spring, linear (Newtonian) fluids that flow at a rate that is proportional to the instantaneous fluid pressure gradient and simple uncoupled isothermal chemical reactions such as A  →  B where the reaction rate is proportional to the affinity or driving force for the reaction. One of the characteristics of such linear systems is stability which means that if we perturb the system by a small amount from its current position it will return to its initial configuration, although it may undergo some oscillatory motion as it returns. Thus if we pull on the weight at the end of a linear spring and then release the weight the system will oscillate but return to its initial position. This initial position is an equilibrium state where no motion occurs. Such systems possess only one equilibrium state and that state is stable.

For linear systems the mathematical principle of superposition holds (Boyce and DiPrima, 2005). The system involved may be a deforming mechanical system or a chemical system undergoing a simple uncoupled reaction. This principle says that if functions f1 and f2 are two possible solutions describing the behaviour of such a system then αf1  +  βf2 is also a solution where α and β are arbitrary constants. Thus if f1  =  sin  (x) and f2  =  sin  (2x) then the principle of superposition says that another solution describing the behaviour is f3  =  sin  (x)  +  2sin  (2x) as shown in Figure 1.7(c); this behaviour is periodic. If the ratio β/α is irrational then the behaviour is quasi-periodic . This means that in a linear system solutions describing the behaviour of the system evolve independently of each other with no mechanical or chemical interference between each other and the final solution can be expressed as a Fourier series of the various independent solutions. One might expect that this could lead to quite complicated geometrical or chemical patterns but for many mechanical and chemical systems of interest it turns out that one solution grows much faster than all others so that a dominant solution appears finally as the only (sinusoidal) solution. This is the process that operates in the classical Biot theory of folding (Biot, 1965) which is rigorously true in two dimensions for a layer composed of any material embedded in a linear matrix. An infinite number of wavelengths start to grow but one grows fastest to become the dominant sinusoidal wavelength.

Figure 1.4   Molecular simulation of the mechanisms involved in the stress solubility of quartz corresponding to a stress σ/σcr  =  0.58 where σcr is the maximum stress at fracture for a pure quartz rod (After Zhu et al. (2005)). (a) A rod of α-quartz in the absence of H2O. Medium-sized spheres are oxygen and large spheres are silicon. An H2O molecule is nearby with hydrogen atoms denoted by the small spheres. Corresponding energy is shown in (e). (b) The H2O molecule approaches the SiO2 rod. Corresponding energy is shown in (e). (c) Saddle point configuration where the energy is a maximum. Corresponding energy is shown in (e). (d) Final chemisorbed state. Corresponding energy is shown in (e). (e) Energy versus the reaction coordinate at high and low stresses measured by the ratio σ/σcr. Atoms are colour coded by charge variation relative to the initial configuration (a).

Figure 1.5   Computer simulated time sequence of the propagation of dislocations from opposing crack tips. From Abraham (2003).

Figure 1.6   Typical structures in deformed rocks. (a) Single layer localised folding in schists from Cap de Creus, Spain. (b) Mineral lineation approximately parallel to bedding/foliation intersection in banded iron formations, Tropicana, Western Australia. (c) Folding in Chewings Range quartzites, Alice Springs, Australia. (d) Quartz veins in deformed turbidites, Bermagui, Australia.

The behaviour of nonlinear systems, however, is quite complicated and at present there is no comprehensive overarching theory that enables general principles to be stated. We will explore many aspects of nonlinear behaviour in the remainder of this book. However, as an example the buckling behaviour of a single layer embedded in nonlinear material is summarised in Figure 1.8. Here we see that the response may be homogeneous, with no buckling behaviour, or a range of buckling behaviours depending sensitively on initial conditions, on the boundary conditions and on the evolution of geometry and mechanical properties of both the layer and the embedding material during buckling. The inhomogeneous responses may be sinusoidal, periodic but nonlocalised, localised and periodic, localised and non-periodic and quasi-periodic or chaotic. The localised packages are also referred to as solitary waves. These various forms of buckling behaviour are discussed in Volume II.

Figure 1.7   The principle of superposition. If f1  =  sin  (x) shown in (a) is a function that describes the evolution of some linear system and so also is f2  =  sin  (2x) shown in (b), then another solution is a linear combination of these two solutions: f3  =  sin  (x)  +  2sin  (2xis shown in (d). A quasi-periodic function is close to periodic but never repeats itself.

Figure 1.8   The various responses to the deformation of a layer embedded in a nonlinear material. The homogeneous and sinusoidal responses are characteristic of linear embedding materials. The periodic unlocalised response is characteristic of linear embedding materials for very small deformations; such structures ultimately evolve to a sinusoidal response. See Volume II for detailed discussions.

The responses illustrated here are characteristic of many nonlinear systems including nonlinear chemical systems, shear zone localisation and the development of more permeable pathways in a reacting porous medium with fluid flow. The development of homogeneous, sinusoidal and nonlocalised periodic behaviour is characteristic of many linear systems. We discuss the details of these behaviours in later sections of this book. The essential differences in behaviour between linear and nonlinear systems are summarised in Table 1.1. It is important to note that the responses for both linear and nonlinear systems are characteristic of situations where there is an initial random distribution of small-amplitude geometrical, physical or chemical heterogeneities. Larger heterogeneities influence the final result for nonlinear systems but are not necessary to produce localised or aperiodic behaviour. The influence of initial imperfections is discussed in Volume II. The localised and chaotic response of nonlinear systems derives solely from nonlinear geometrical, physical and/or chemical interactions.

The nonlinear behaviour may be material or geometrical in nature and in this introductory discussion we concentrate on geometrical nonlinearity and give two examples to emphasise that even the simplest of mechanical behaviour coupled with simple geometrical nonlinearity can lead to localised and nonintuitive behaviour. A more extensive discussion is presented in Chapter 7. As a first example, slight complications in the geometrical arrangements of systems of linear springs can soon lead to complexity. Consider the system shown in Figure 1.9 where two linear springs are arranged initially as shown in Figure 1.9(a). A displacement under the influence of a vertical force leads to the configuration shown in Figure 1.9(b). In doing so the force increases as shown in Figure 1.9(d) from O through D to A. However, this configuration is unstable; a further small displacement results in a jump to the configuration shown in Figure 1.9(c). In doing so the displacement jumps from A to B in Figure 1.9(d). This behaviour is referred to as snap-through instability. Continued downward displacement continues on the path indicated by the black arrow from B. In reverse, indicated by the blue arrows in Figure 1.9(d) the system unloads to C and then jumps to D to continue unloading back to the initial configuration.

Table 1.1

The Different Behaviours of Linear and Nonlinear Systems

In the force–displacement diagram in Figure 1.9(d) the branches OA and BC represent stable configurations such that a small increase or decrease in force results in a small change in displacement. The branch AC is unstable in that a small change in force can result in a large displacement. The system behaves differently under loading and unloading conditions resulting in what is called hysteresis. This behaviour results in an otherwise linear system becoming unstable when a critical geometrical configuration is reached. Hysteresis is typical of many systems, particularly mechanical systems where some mechanical parameter softens due to thermal or chemical feedback, of many chemical reactions such as a simple exothermic reaction, A  →  B, and of many chemically reacting systems with open fluid flow. One should appreciate that the above scenario is true for force boundary conditions. If the system is loaded at constant velocity the complete force–displacement curve is followed including the unstable branch with no snap-through instability.

Figure 1.9   A simple system of linear springs that displays nonlinear behaviour arising from changes in geometry under force-controlled loading. (a) Initial unloaded condition. This corresponds to the point O in (d). (b) An unstable configuration corresponding to point A in (d). (c) A new stable configuration that the system jumps to after (b). This corresponds to point B in (d). (d) Force–displacement diagram showing a stable loading path from O to A, an unstable path which is never accessed from A to C and a stable path corresponding to a loading path past B and an unloading path BC. On loading the system follows the path OAB and beyond represented by the black arrows and on unloading follows the path BCDO represented by the blue arrows. The fact that the loading and unloading paths are different is referred to as hysteresis. If the same system is subjected to constant displacement rate loading conditions the system follows the complete force–displacement curve. That is, it follows the unstable branch as well as the stable branches.

As a second example consider a number of balls arranged in layers and connected by linear springs as shown in Figure 1.10. The packing of the balls is initially close. The nonlinearity in this system is geometrical and results solely from the fact that, during deformation, balls need to ride over each other, producing a dilation, especially as the rotation of an individual chain of balls increases. In the two panels shown in Figure 1.10 no energy is associated with the dilation in the left-hand panel, whereas an energy penalty is associated with dilation in the right-hand panel. By penalty we mean that some work needs to be done for the material to dilate. The initial length of the system before deformation in both cases is 100  units.

Figure 1.10   A three-layer system consisting of initially close packed balls connected by linear springs. The nonlinearity in the system arises from large rotations of groups of balls. In the left-hand panel no work is done against dilation as the balls slide over each other. In the right-hand panel work is done against dilation and hence dilation is minimised and the initial close packing tends to be preserved. Dilation sites are represented by small white circles. (a) One of the two initial solutions which evolves to become (e). (b) The second kind of initial solution which evolves to become (c) and (d). From Hunt and Hammond (2012).

For the left-hand panel of Figure 1.10 even the initial response is complex in that at least two initial solutions are possible. The first shown in (a) is sinusoidal but progresses to strong aperiodic localisation shown in (e). A second solution is shown in (b) which occurs at a slightly lower initial stress than (a), and consists of weak localisation which intensifies by (c) and progressively becomes aperiodic, (d), as deformation proceeds. The lack of any penalty for the formation of voids means that local departures from the initial close packing are common. This enables a rounded waveform to develop so that the fold style could be labelled concentric. Differences in the pattern of dilation are responsible for the two different initial responses shown in (a) and (b) in the right and left panels.

In the right-hand panel of Figure 1.10, the response is localised from the start of deformation and the fold style is kink or chevron like. Again there are two initial solutions (a) and (b). Solution (a) evolves to become (e), whereas (b) evolves through the stages (c) and (d). The penalty imposed by introducing the work due to dilation forces the packing to remain close packed even for large rotations and any necessary dilation is localised. The result is kink or chevron styles of folds. There is an added element of complexity here in that the width of kink bands can evolve from even numbers of balls to odd numbers of balls or vice versa in quite a complex manner (Hunt and Hammond, 2012). Such behaviour is a common feature of nonlinear systems and will be discussed in greater detail in Chapter 7.

1.4. Wavelet Analysis

Since many deformed rocks have structures that are developed over a range of length scales and (we will show) fractal geometries are ubiquitous, it is natural to use a technology that takes such characteristics into account when attempting to quantify the fabrics of deformed rocks and the associated patterns (and defects in the patterns) that are characteristic of these fabrics. In this book and Volume II we use wavelet transforms as a convenient tool to achieve quantification of all deformation/metamorphic fabrics and patterns including the distribution of mineral phases, the CPOs developed during deformation, the shapes of folds at all scales and the geometry of fracture and vein networks. A wavelet transform consists of two parts. The so-called mother wavelet is a mathematical function, usually of a wave shape, that is commonly localised in space and designed in order to emphasise some aspect of the pattern being quantified. The second part of the transform consists of scanning the pattern with this function as it is stretched or compressed to become daughter wavelets. The amount of stretch or compression is the scaling factor. At each point in the fabric the degree of match between some measure of the fabric with the stretched or compressed wavelet is recorded in the form of a series of coefficients.

The wavelet transform can be viewed as a microscope that enables one to probe a pattern