Premelting, fluctuations and coarse-graining of water-ice interfaces

TL;DR

This paper combines statistical field theory and molecular dynamics to analyze premelting phenomena at water-ice interfaces, providing a coarse-grained model that captures interfacial properties and their dependence on temperature and system size.

Contribution

It introduces a coarse-graining procedure linking molecular configurations to a scalar order parameter, enabling analysis of premelting layer properties and their universal features.

Findings

Premelting layer thickness grows logarithmically with system size.

Premelting layer existence is insensitive to crystal face and curvature.

The model accurately predicts interfacial properties like surface tension.

Abstract

Using statistical field theory supplemented with molecular dynamics simulations, we consider premelting on the surface of ice as a generic consequence of broken hydrogen bonds at the boundary between the condensed and gaseous phases. A procedure for coarse-graining molecular configurations onto a continuous scalar order parameter field is discussed, which provides a convenient representation of the interface between locally crystal-like and locally liquid-like regions. A number of interfacial properties are straightforwardly evaluated using this procedure such as the average premelting thickness and surface tension. The temperature and system size dependence of the premelting layer thickness calculated in this way confirms the characteristic logarithmic growth expected for the scalar field theory that the system is mapped onto through coarse-graining, though remains finite due to long-ranged interactions. Finally, from explicit simulations the existence of a premelting layer is shown to be insensitive to bulk lattice geometry, exposed crystal face and curvature.