Grant: Discovery Grants Program - Individual

Complex Metastable Liquids: Phase Behaviour, Dynamics, and NucleationThe liquid state is an enduring source of challenges for our understanding of matter. Even the simplest classical liquid is a dense, disordered, and strongly-interacting many-body system. Liquids thus present us with rich behavior and formidable complexity, especially at the molecular scale.Some of the most important and perplexing open questions in liquid-state physics occur in supercooled liquids. The nature of amorphous solid formation at the glass transition is a continuing source of debate. Similarly, the nucleation of crystalline solids from the liquid state is also incompletely understood; e.g. our ability to predict crystal nucleation rates with quantitative accuracy remains poor for many important systems. While all these questions are of a fundamental nature, practical applications abound. For example, metallic glass alloys are a growing area of commercial interest, e.g. for aircraft components. Understanding crystal nucleation is vital in a wide swathe of manufacturing, from computer chips to pharmaceuticals.In the context of supercooled liquids, a number of distinct and complex phenomena converge: the thermodynamics of metastable phases; the complex dynamic properties of viscous liquids; and the non-equilibrium kinetics of crystal nucleation. This proposal describes a program of research to investigate supercooled liquids, with a focus on nucleation phenomena in water, because of its complexity, and its pervasive importance across the natural sciences and engineering. At the same time, we will seek to study substances under conditions in which all the complex phenomena listed above become intertwined, in order to seek a broader understanding of metastable systems.We will use molecular dynamics and Monte Carlo computer simulations, from which both molecular-scale details and estimates of bulk properties may be obtained. For example, phase diagrams and free energy barriers for nucleation will be evaluated using umbrella sampling methods, and rare-event simulation techniques such as forward-flux sampling will facilitate calculation of nucleation rates. Our research directions are chosen to provide results of wide scientific interest, and to create training opportunities that prepare students for future work in the quantitative sciences. For example, we will study ice nucleation in deeply supercooled water, close to the proposed liquid-liquid phase transition that has been observed in computer simulations. We will also examine crystal nucleation in systems exhibiting pronounced mobility fluctuations (i.e. dynamical heterogeneities) to explore connections between nucleation and glass formation. We will also study heterogeneous nucleation, especially in water, due to the importance of this process for understanding the contributions of cloud formation to climate modeling.