Phenomenological Model of Glass Dynamics: Effect of Spatial Heterogeneity in Glass and Supercooled Liquid

Abstract

Energy consumption is exponentially increasing every year in our society. This is caused by the rapid growth of information that needs to be transferred, processed, and stored on a global scale. Therefore, the materials science community must study and develop materials that can serve as building blocks for more energy-efficient and powerful information processing devices. Glasses play a pivotal role in achieving this goal because of their technological relevance, such as silica fibers for information transport across the globe and chalcogenide glasses for electronic and photonic devices for information processing and storage. Designing efficient and effective glasses for such applications requires accurate control of their physical properties. However, predicting the dynamics of glasses is known to be extremely challenging. Thus, developing a model that can predict glass dynamics under various temperature conditions is of great interest to the glass science community and industry. In this dissertation, we propose a new phenomenological model that can predict glass dynamics over a wider range of temperatures and time scales, which exceeds the capability of existing phenomenological models.Various organic and inorganic systems are used to validate the proposed model which correctly predicts the temperature-dependence of non-exponentiality in supercooled liquids, where the non-exponentiality controls the microscopic spatial heterogeneous dynamics of supercooled liquids. The model is also used to investigate the relaxation dynamics of two silicate glasses far below their glass transition temperatures T_g. The proposed model successfully simulate their relaxation behavior after complex thermal histories while existing models such as Tool-Narayanaswamy-Moynihan (TNM) have failed to correctly describe their relaxation dynamics. This study reveals the importance of accounting for the temperature dependence of non-exponentiality, as the temperature dependence of spatial heterogeneity directly affects the glass dynamics far from T_g. Kovacs’ expansion gap paradox is reproduced for the first time in the literature using the proposed phenomenological model. Further, the paradox is investigated over a wide range of temperatures to reveal that the paradox as well as the non-linear relaxation vanishes at high temperatures. It is found that the vanishing of the non-linear relaxation and the expansion gap are caused by the significant narrowing of the distribution of relaxation time at high temperatures, leading to homogenous relaxation regardless of the initial temperature before annealing. Additionally, the model is used to simulate the microscopic fluctuation of density as a function of temperature and time, and the model correctly describes the dynamics of the fluctuation that are observed in experiments. This is the first demonstration of modeling the density fluctuation with a phenomenological model in the literature. Thus, these studies show the importance of accounting for the dynamics of microscopic spatial heterogeneity to successfully predict complex relaxation with a phenomenological model. The proposed model is shown to exceed the capability of existing models and should therefore serve as an important new tool to understand glass dynamics and allows the development of glasses with desired physical properties for a range of applications such as efficient and powerful information processing devices.