The range of meta stability of ice-water melting for two simple models of water

TL;DR

This study investigates the superheating limits of ice phases in water models using Monte Carlo simulations, revealing a consistent superheating range of about 90K beyond coexistence temperatures, with implications for estimating equilibrium melting points.

Contribution

It quantifies the superheating range of ice in two water models and proposes a practical rule of thumb for estimating melting points from mechanical stability limits.

Findings

Ice can be superheated approximately 90K beyond coexistence temperature.

Superheating limit depends weakly on ice phase and water model.

A simple rule of thumb estimates melting points by subtracting 90K from stability limits.

Abstract

A number of crystal structures of water have been `superheated' in Monte Carlo simulations. Two well known models for water were considered; namely the TIP4P model and the SPC/E model. By comparing the fluid-solid coexistence temperature to the temperature at which the solid becomes mechanically unstable and melts it is possible to determine the typical range of temperatures over which is possible to superheat the ice phases in conventional simulation studies. It is found that the ice phases can be superheated to approximately 90K beyond the fluid-solid coexistence temperature. Beyond this limit they spontaneously melt. This limit appears to depend weakly both on the type of ice phase considered and on the chosen model. Obviously only rigorous free energy calculations can determine the equilibrium fluid-solid coexistence of a model. However, a "rule of thumb" is that, by subtracting 90K from the mechanically stability limit of the the ice phase one is provided with a first guess as to the equilibrium fluid-solid coexistence temperature.