Atomistic Simulations of Glasses

Jincheng Du, PhD, is Professor of materials science and engineering at the University of North Texas. He is Chair of the TC27 Technical Committee on Atomistic Simulation with the International Commission of Glass and is the Editor of the Journal of the American Ceramic Society.Alastair N. Cormack, PhD, Professor at the New York State College of Ceramics at Alfred University. He is a leading authority in the field of computer modeling of materials, focusing on the atomic-scale physics and chemistry of ceramics and glass.

• PrefacePart I Fundamentals of Atomistic SimulationsChapter 1 Classical simulation methods Abstract1.1 Introduction1.2 Simulation techniques1.2.1 Molecular dynamics (MD)1.2.1.1 Integrating the equations of motion1.2.1.2 Thermostats and barostats1.2.2 Monte Carlo (MC) eimulations1.2.2.1 Kinetic Monte Carlo1.2.2.2 Reverse Monte Carlo1.3 The Born Model1.3.1 Ewald summation1.3.2 Potentials1.3.2.1 Transferability of potential parameters: Self-consistent sets1.3.2.2 Ion polarizability1.3.2.3 Potential models for borates1.3.2.4 Modelling reactivity: electron transfer1.4 Calculation of Observables1.4.1 Atomic structure1.4.2 Hyperdynamics and peridynamics1.5 Glass Formation1.5.1 Bulk structures1.5.2 Surfaces and fibers1.6 Geometry optimization and property calculations1.7 ReferencesChapter 2 Ab initio simulation of amorphous solids Abstract2.1 Introduction2.1.1 Big picture2.1.2 The limits of experiment2.1.3 Synergy between experiment and modeling2.1.4 History of simulations and the need for ab initio methods2.1.5 The difference between ab initio and classical MD2.1.6 Ingredients of DFT2.1.7 What DFT can provide2.1.8 The emerging solution for large systems and long times: Machine Learning2.1.9 A practical aid: Databases2.2 Methods to produce models2.2.1 Simulation Paradigm: Melt Quench2.2.2 Information Paradigm2.2.3 Teaching chemistry to RMC: FEAR2.2.4 Gap Sculpting2.3 Analyzing the models2.3.1 Structure2.3.2 Electronic Structure2.3.3 Vibrational Properties2.4 Conclusion2.5 Acknowledgements2.6 ReferencesChapter 3 Reverse Monte Carlo simulations of non-crystalline solids Abstract3.1 Introduction -- why RMC is needed?3.2 Reverse Monte Carlo modeling3.2.1. Basic RMC algorithm3.2.2. Information deficiency3.2.3. Preparation of reference structures: hard sphere Monte Carlo3.2.4. Other methods for preparing suitable structural models3.3 Topological analyses3.3.1. Ring statistics3.3.2. Cavity analyses3.3.3. Persistent homology analyses3.4 Applications3.4.1 Single component liquid and amorphous materials3.4.1.1 l-Si and a-Si3.4.1.2 l-P under high pressure and high temperature3.4.2 Oxide glasses3.4.2.1 SiO2 glass3.4.2.2 R2O-SiO2 glasses (R=Na, K)3.4.2.3 CaO-Al2O3 glass3.4.3 Chalcogenide glasses3.4.4 Metallic glasses3.5 Summary3.6 Acknowledgments3.7 ReferencesChapter 4 Structure analysis and property calculations abstract4.1 Introduction4.2 Structure Analysis4.2.1 Salient features of glass structures4.2.2 Classification of the range order.4.3 Real Space Correlation functions.Spectroscopic properties: validating the structural models4.3.1 X-ray and Neutron diffraction spectra4.3.2 Vibrational spectra4.3.3 NMR spectra4.4 Transport properties4.4.1 Diffusion coefficient and diffusion activation energy4.4.2 Viscosity4.4.3 Thermal conductivity4.5 Mechanical Properties4.5.1 Elastic constants4.5.2 Stress-strain diagrams and fracture mechanism4.6 Concluding remarks4.7 ReferencesChapter 5 Topological constraint theory of glass: counting constraints by molecular dynamics simulations Abstract5.1 Introduction5.2 Background and topological constraint theory5.2.1 Rigidity of mechanical networks5.2.2 Application to atomic networks5.2.3 Constraint enumeration under mean-field approximation5.2.4 Polytope-based description of glass rigidity5.2.5 Impact of temperature5.2.6 Need for molecular dynamics simulations5.3 Counting constraints from molecular dynamics simulations5.3.1 Constraint enumeration based on the relative motion between atoms5.3.2 Computation of the internal stress5.3.3 Computation of the floppy modes5.3.5 Dynamical matrix analysis5.4 Conclusions5.5 ReferencesPart II Applications of Atomistic Simulations in Glass ResearchChapter 6 History of atomistic simulations of glasses Abstract6.1 Introduction6.2 Simulation techniques6.2.1 Monte Carlo techniques6.2.2 Molecular dynamics6.3 Classical simulations: interatomic potentials6.3.1 Potential models for silica6.3.1.1 Silica: quantum mechanical simulations6.3.2 Modified silicates and aluminosilicates6.3.3 Borate glasses6.3.3.1 Borates: quantum mechanical simulations6.4 Simulation of surfaces6.5 Computer science and engineering6.6.1 Software6.6.2 Hardware6.6 ReferencesChapter 7 Silica and silicate glassesAbstract7.1 Introduction7.2 Atomistic simulations of silicate glasses: ingredients and critical aspects7.3 Characterization and experimental validation of structural and dynamic features of simulated glasses7.3.1 Structural characterizations7.3.2 Dynamic properties of simulated glasses7.3.3 Validation and experimental confirmation of structural and dynamic properties7.3.3.1 Diffraction methods7.3.3.2 Nuclear Magnetic Resonance7.3.3.3 Vibrational spectral characterization7.4 MD simulations of silica glasses7.5 MD simulations of alkali silicate and alkali earth silicate glasses7.5.1 Local environments and distribution of alkali ions7.5.2 The mixed alkali effect7.6 MD simulations of aluminosilicate glasses7.7 MD simulations of nanoporous silica and silicate glasses7.8 AIMD simulations of silica and silicate glasses7.9 Summary and OutlookAcknowledgementsReferencesChapter 8 Borosilicate and boroaluminosilicate glasses 8.1 Abstract8.2 Introduction8.3 Experimental determination and theoretical models of boron N4 values in borosilicate glass8.3.1 Experimental results on boron coordination number8.3.2 Theoretical models in predicting boron N4 value8.4 ab initio versus classical MD simulations of borosilicate glasses8.5 Empirical potentials for borate and borosilicate glasses8.5.1 Recent development of rigid ion potentials for borosilicate glasses8.5.2 Development of polarizable potentials for borate and borosilicate glasses8.6 Evaluation of the potentials8.7 Effects of cooling rate and system size on simulated borosilicate glass structures8.8 Applications of MD simulations of borosilicate glasses8.8.1 Borosilicate glass8.8.2 Boroaluminosilicate glasses8.8.3 Boron oxide-containing multi-component glass8.9 Conclusions8.10 Appendix: Available empirical potentials for boron-containing systems8.10.1 Borosilicate and boroaluminosilicate potentials-Kieu et al and Deng&Du8.10.2 Borosilicate potential- Wang et al8.10.3 Borosilicate potential-Inoue et al8.10.4 Boroaluminosilicate potential-Ha and Garofalini8.10.5 Borosilicate and boron-containing oxide glass potential-Deng and Du8.10.6 Borate, boroaluminate and borosilicate potential-Sundararaman et al8.10.7 Borate and borosilicate polarizable potential-Yu et al8.10 Acknowledgements8.11 ReferencesChapter 9 Nuclear waste glasses 9.1 Preamble9.2 Introduction to French nuclear glass9.2.1 Chemical composition9.2.2 About the long term behavior (irradiation, glass alteration, He accumulation)9.2.3 What can atomistic simulations contribute?9.3 Computational methodology9.3.1 Review of existing classical potentials for borosilicate glasses9.3.2 Preparation of a glass9.3.3 Displacement cascade simulations9.3.4 Short bibliography about simplified nuclear glass structure studies9.4 Simulation of radiation effects in simplified nuclear glasses9.4.1 Accumulation of displacement cascades and the thermal quench model9.4.2 Preparation of disordered and depolymerized glasses9.4.3 Origin of the hardness change under irradiation9.4.4 Origin of the fracture toughness change under irradiation9.5 Simulation of glass alteration by water9.5.1 Contribution from ab initio calculations9.5.2 Contribution from Monte Carlo simulations9.6 Gas incorporation: radiation effects on He solubility9.6.1 Solubility model9.6.2 Interstitial sites in SiO2-B2O3-Na2O glasses9.6.3 Discussion about He solubility in relation to the radiation effects9.7 Conclusions9.8 Acknowledgements9.9 ReferencesChapter 10 Phosphate glasses Abstract10.1 Introduction to phosphate glasses10.1.1 Applications of phosphate glasses10.1.2 Synthesis of phosphate glasses10.1.3 The modified random network model applied to phosphate glasses10.1.4 The tetrahedral phosphate glass network10.1.5 Modifier cations in phosphate glasses10.2 Modelling methods for phosphate glasses10.2.1 Configurations of atomic coordinates10.2.2 Molecular modelling versus reverse Monte Carlo modelling10.2.3 Classical vs. ab initio molecular modelling10.2.4 Evaluating the simulation of interatomic interactions10.2.5 Evaluating models of glasses by comparison with experimental data10.3 Modelling pure vitreous P2O510.3.1 Modelling of crystalline P2O510.3.2 Modelling of vitreous P2O510.3.3 Cluster models of vitreous P2O510.4 Modelling phosphate glasses with monovalent cations10.4.1 Modelling lithium phosphate glasses10.4.2 Modelling sodium phosphate glasses10.4.3 Modelling phosphate glasses with other monovalent cations10.4.4 Modelling phosphate glasses with monovalent cations and addition of halides10.4.5 Cluster models of alkali phosphate glasses10.5 Modelling phosphate glasses with divalent cations10.5.1 Modelling zinc phosphate glasses10.5.2 Modelling zinc phosphate glasses with additional cations10.5.3 Modelling alkaline earth phosphate glasses10.5.4 Modelling lead phosphate glasses10.6 Modelling phosphate based glasses for biomaterials applications10.6.1 Modelling Na2O-CaO-P2O5 glasses with 45 mol% P2O510.6.2 Modelling Na2O-CaO-P2O5 glasses with 50 mol% P2O510.6.3 Modelling Na2O-CaO-P2O5 glasses with additional cations10.7 Modelling phosphate glasses with trivalent cations10.7.1 Modelling iron phosphate glasses10.7.2 Cluster models of iron phosphate glasses10.7.3 Modelling trivalent rare earth phosphate glasses10.7.4 Modelling aluminophosphate glasses10.8 Modelling phosphate glasses with tetravalent and pentavalent cations10.9 Modelling phosphate glasses with mixed network formers10.9.1 Modelling borophosphate glasses10.9.2 Modelling phosphosilicate glasses10.10 Modelling bioglass 45S and related glasses10.10.1 Modelling bioglass 45S and related glasses from the same system10.10.2 Modelling bioglass 45S and related glasses with additional components10.11 Summary10.12 ReferencesChapter 11 Bioactive glasses Abstract11.1 Introduction11.2 Methodology11.3 Development of interatomic potentials11.4 Structure of 45S5 Bioglass11.5 Inclusion of ions into bioactive glass and the effect on structure and bioactivity11.6 Glass nanoparticles and surfaces11.7 Discussion and future workBibliographyChapter 12 Rare earth and transition metal containing glasses Abstract12.1 Introduction12.1.1 Transition metal and rare earth oxides in glasses: importance and potential applications12.1.2 Effects of local structures and clustering behaviors of RE and TM ions on properties12.1.3 Redox reaction and multioxidation states of TM and RE ions12.1.4 Effect of composition on multioxidation states in glasses containing TM12.1.5 The role of MD in investigating TM and RE containing glasses12.2 Simulation methodologies12.2.1 Interatomic potentials and glass simulations12.2.2 Cation environment and clustering analysis12.2.3 Diffusion and dynamic property calculations12.2.4 Electronic structure calculations12.3 Case studies of MD simulations of RE and TM containing glasses12.3.1 Rare earth doped silicate and aluminophosphate glasses for optical applications12.3.1.1 Erbium doped silica and silicate glasses: from melt-quench to ion implantation12.3.1.2 Europium and praseodymium doped silicate glasses12.3.1.3 Cerium doped aluminophosphate glasses: atomic structure and charge trapping12.3.2 Alkali vanadophosphate glasses as a mixed conductor12.3.2.1 General features of vanadophosphate glasses12.3.2.2 Sodium vanadophosphate glass12.3.2.3 Lithium vanadophosphate glass12.3.3 Zirconia containing aluminosilicate and borosilicate glasses for nuclear waste disposal12.4 ConclusionsAcknowledgementReferencesChapter 13 Halide and oxyhalide glasses Abstract13.1 Introduction13.2 General Structure Features of Fluoride and Oxyfluoride Glasses13.2.1 Structure Features of Fluoride Glasses13.2.2 Structure Features of Oxyfluoride Glasses13.2.3 Phase Separation in Fluoride and Oxyfluoride Glasses13.3 Structures and Properties of Fluoride Glasses from MD Simulations13.3.1 General Structures from MD simulations13.3.2 Cation Coordination and Structural Roles13.3.3 Fluorine Environments13.4 MD Simulations of Fluoroaluminosilicate Oxyfluoride Glasses13.4.1 Oxide and Fluoride Glass Phase Separation Observed from MD Simulations13.4.2 Oxide-Fluoride Interfacial Structure Features from MD simulations13.4.3 Correlation of Structural Features between MD and Crystallization13.5 ab initio MD simulations of oxyfluoride glasses13.6 ConclusionsAcknowledgementsReferencesChapter 14 Glass surface simulations abstract14.1 Introduction14.2 Classical molecular dynamics surface simulations14.2.1 amorphous silica surfaces14.2.2 Multicomponent oxide glass surfaces14.2.2.1 Bioactive glasses14.2.3 Wet glass surfaces14.2.3.1 Reactive potentials14.3 First Principles Surface Simulations14.3.1 Silica glass surfaces14.3.2 Multicomponent glass surfaces14.3.3 Wet glass surfaces14.4 SummaryAcknowledgementsReferencesChapter 15 Simulations of glass - water interactions Abstract15.1 Introduction15.1.1 Glass Dissolution Process and Experimental Characterizations15.1.2 Types of Atomistic Simulation Methods for Studying Glass-Water Interactions15.2 First-Principles Simulations of Glass-Water Interactions15.2.1 Brief Introduction to Methods15.2.2 Energy Barriers for Si-O-Si Bond Breakage15.2.3 Reaction Mechanism for Si-O-Si Bond Breakage15.2.4 Strained Si-O-Si linkages15.2.5 Reaction Energies for Multicomponent Linkages15.2.6 Effect of pH on Si-O-Si Hydrolysis Reactions15.2.7 Nanoconfinement of water in porous materials15.2.8 Oniom or QM/MM simulations15.2.9 Areas for improvement/additional research15.3 Classical Molecular Dynamics Simulations of water-glass interactions15.3.1 Brief Introduction and History15.3.2 Non-Reactive Potentials15.3.3 Reactive Potentials15.3.4 Silica Glass-Water Interactions15.3.5 Silicate Glass – Water Interactions15.3.6 Other glasses – water interactions15.3.7 Areas for Improvement15.4 Challenges and Outlook15.4.1 Extending the Length and Time Scales of Atomistic Simulation15.4.2 Reactive Potential Development15.5 Conclusion Remarks15.6 Acknowledgements15.7 References

Modeling and simulation are crucial for understanding structure-property relationships in glass-forming systems and for accelerating the design of next-generation glassy materials. Atomistic Simulations of Glasses is a comprehensive volume dedicated to the topic of atomic-scale modeling of glassy materials, with particular emphasis on silicate glasses of practical industrial interest. As such, this book fills a critical gap in the literature, offering an excellent introduction for newcomers to atomistic modeling, as well as a comprehensive and state-of-the-art reference for practitioners in the field.Atomistic Simulations of Glasses, published by ACerS-Wiley, consists of 15 chapters written by experts from around the world. It is edited by two leading authorities in computational glass science: Jincheng Du (University of North Texas) and Alastair N. Cormack (Alfred University). The book itself is gorgeous, printed in full color on high-quality paper. It is designed in a reader-friendly format, including a comprehensive index, an extensive list of references at the end of each chapter, and a helpful table to decode every acronym used throughout the book. Each chapter is well written and has been carefully polished. The text also flows smoothly across chapters, which is sometimes a problem in edited volumes.The first five chapters are devoted to fundamentals of atomistic modeling techniques for glassy systems, including classical simulation methods (Chapter 1), quantum mechanical techniques (Chapter 2), reverse Monte Carlo (Chapter 3), structural analysis methods (Chapter 4), and topological constraint theory (Chapter 5). Each of these chapters does a great job at providing both foundational knowledge and discussing the state-of-the-art in methods and tools. The chapter on topological constraint theory is especially interesting because this is a family of techniques developed specifically for glassy materials.The latter 10 chapters of the book focus on application of these techniques for simulating various glass families of interest. These chapters cover a wide range of silicate, aluminosilicate, and borosilicate glasses, as well as phosphate, fluoride, and oxyfluoride systems. The coverage of transition metal and rare-earth-containing glasses is also a nice touch. There is a particular emphasis on bioactive glasses and glasses for nuclear waste immobilization. As a whole, the 10 application-focused chapters do an excellent job demonstrating the utility and versatility of atomistic simulation approaches for addressing problems of practical concern in the glass science and engineering community. These chapters also provide good perspective on specific needs for future developments in the field.There are a few missing topics that would have been valuable to include in the book. While reactive force fields are mentioned briefly, an entire chapter devoted to the principles and applications of reactive force fields such as ReaxFF would have been a nice addition, especially because reactive force fields are becoming increasingly important in the glass science community. Also, given the importance of thermal history in governing the structure and properties of glasses, it would have been worthwhile to include a chapter on accessing long time scales, e.g., using kinetic Monte Carlo, meta-dynamics, or the activation-relaxation technique, all of which have been applied to noncrystalline systems in the literature and can enable simulations to access experimental time scales. It also would have been helpful to expand the chapter on reverse Monte Carlo to include other Monte Carlo techniques more broadly; for example, Metropolis Monte Carlo is a computationally efficient alternative to molecular dynamics for calculating glass structure and static properties. Finally, given the large amount of research activity in modeling of metallic glasses, a chapter on atomistic simulations of metallic glasses would be a nice addition.Overall, Atomistic Simulations of Glasses is a very welcome addition to the literature and highly recommended for both students and professionals in the field of computational glass science.—John C. Mauro is a Dorothy Pate Enright Professor in the Department of Materials Science and Engineering at The Pennsylvania State University