Crystallization of Ice and Organics

Ice nucleation is at the heart of cloud formation and global precipitation. Ice nucleation from pure water occurs at extremely low temperatures of around -35°C. However, most ice forms through heterogeneous nucleation at much higher temperatures due to the presence of dust, soot, organics, and bacteria. Therefore, predicting the ice-nucleating efficiency induced by biological and organic materials is one of the main challenges in climate science. Molecular simulations and the development of coarse grained water models offer powerful tools to resolve the mechanism of ice crystallization.

The goal of this research is to use molecular dynamic simulations and develop a framework based on Classical Nucleation Theory in order to elucidate the characteristics of surfaces that promote the heterogeneous nucleation of ice. In doing so, one can strategically control and boost ice nucleation efficiency, which would strongly impact the Earth’s energy budget and control climate.

I used Classical Nucleation Theory (CNT) to develop a general framework that links the non-equilibrium ice freezing temperature facilitated by a surface to the equilibrium thermodynamics of ice binding to that surface (see Figure 3). This framework allows us to predict the ice nucleation temperature for any arbitrary surface based on its binding free energy to the ice crystal. It is flexible and can be customized or extended to address various research questions.

I integrated this framework with molecular dynamic simulations to explore the heterogeneous nucleation of ice in several key systems:
1) Ice-nucleating proteins
2) Monolayers of long-chain alcohols and acids
3) Ice-nucleating surfaces with non-planar geometries, including cholesterol monohydrate wedges and pores
4) Antifreeze proteins

Overall, this new framework provides a way to predict and control the non-equilibrium ice freezing temperature using equilibrium surface properties.