The Putative Liquid-Liquid Transition is a Liquid-Solid Transition in Atomistic Models of Water

TL;DR

This study uses numerical simulations to show that what has been interpreted as a liquid-liquid transition in water models is actually a liquid-solid transition, challenging previous claims of a liquid-liquid critical point.

Contribution

The paper demonstrates through free energy calculations that atomistic water models do not exhibit a true liquid-liquid transition, but rather a liquid-solid transition, refuting earlier hypotheses.

Findings

No evidence of multiple liquid basins in models studied.

Phase coexistence between liquid and crystal confirmed.

Behavior attributed to liquid-liquid transition is due to liquid-solid transition.

Abstract

We use numerical simulation to examine the possibility of a reversible liquid-liquid transition in supercooled water and related systems. In particular, for two atomistic models of water, we have computed free energies as functions of multiple order parameters, where one is density and another distinguishes crystal from liquid. For a range of temperatures and pressures, separate free energy basins for liquid and crystal are found, conditions of phase coexistence between these phases are demonstrated, and time scales for equilibration are determined. We find that at no range of temperatures and pressures is there more than a single liquid basin, even at conditions where amorphous behavior is unstable with respect to the crystal. We find a similar result for a related model of silicon. This result excludes the possibility of the proposed liquid-liquid critical point for the models we have studied. Further, we argue that behaviors others have attributed to a liquid-liquid transition in water and related systems are in fact reflections of transitions between liquid and crystal.