Abstract:Understanding phase behavior of highly confined water, ice, amorphous ice, and clathrate hydrates (or gas hydrates), not only enriches our view of phase transitions and structures of quasi-two-dimensional (Q2D) solids not seen in the bulk phases but also has important implications for diverse phenomena at the intersection between physical chemistry, cell biology, chemical engineering, and nanoscience. Relevant examples include, among others, boundary lubrication in nanofluidic and lab-on-a-chip devices, synthesis of antifreeze proteins for ice-growth inhibition, rapid cooling of biological suspensions or quenching emulsified water under high pressure, and storage of H2 and CO2 in gas hydrates. Classical molecular simulation (MD) is an indispensable tool to explore states and properties of highly confined water and ice. It also has the advantage of precisely monitoring the time and spatial domains in the sub-picosecond and sub-nanometer scales, which are difficult to control in laboratory experiments, and yet allows relatively long simulation at the 10(2) ns time scale that is impractical with ab initio molecular dynamics simulations. In this Account, we present an overview of our MD simulation studies of the structures and phase behaviors of highly confined water, ice, amorphous ice, and clathrate, in slit graphene nanopores. We survey six crystalline phases of monolayer (ML) ice revealed from MD simulations, including one low-density, one mid-density, and four high-density ML ices. We show additional supporting evidence on the structural stabilities of the four high-density ML ices in the vacuum (without the graphene confinement), for the first time, through quantum density-functional theory optimization of their free-standing structures at zero temperature. In addition, we summarize various low-density, high-density, and very-high-density Q2D bilayer (BL) ice and amorphous ice structures revealed from MD simulations. These simulations reinforce the notion that the nanoscale confinement not only can disrupt the hydrogen bonding network in bulk water but also can allow satisfaction of the ice rule for low-density and high-density Q2D crystalline structures. Highly confined water can serve as a generic model system for understanding a variety of Q2D materials science phenomena, for example, liquid-solid, solid-solid, solid-amorphous, and amorphous-amorphous transitions in real time, as well as the Ostwald staging during these transitions. Our simulations also bring new molecular insights into the formation of gas hydrate from a gas and water mixture at low temperature.

Extracting ice phases from liquid water: why a machine-learning water model generalizes so well

Liquid water contains the building blocks of diverse ice phases

Five-Site Water Models for Ice and Liquid Water Generated by a Series-Parallel Machine Learning Strategy

Phase Equilibrium of Water with Hexagonal and Cubic Ice Using the SCAN Functional

Phase Transition Between Crystalline Variants of Ordinary Ice.

Predicting the phase behaviors of superionic water at planetary conditions

Temperature-pressure phase diagram of confined monolayer water/ice at first-principles accuracy with a machine-learning force field

High-dimensional order parameters and neural network classifiers applied to amorphous ices

Unveiling two types of local order in liquid water using machine learning

Efficient parameterization of transferable Atomic Cluster Expansion for water

Machine learning for predictive condensed-phase simulation

Ab initio thermodynamics of liquid and solid water

Quantum-mechanical exploration of the phase diagram of water

Ephemeral ice-like local environments in classical rigid models of liquid water

Machine learning potentials for complex aqueous systems made simple

Highly Confined Water: Two-Dimensional Ice, Amorphous Ice, and Clathrate Hydrates

Ice phase classification made easy with score-based denoising

Liquid-Liquid Transition and Ice Crystallization in a Machine-Learned Coarse grained Water Model

Atomistic View of Homogeneous Nucleation of Water into Polymorphic Ices

IceCoder: Identification of Ice phases in molecular simulation using variational autoencoder