Abstract:Water's unique anomalies are vital in various applications and biological processes, yet the molecular mechanisms behind these anomalies remain debated, particularly in the metastable liquid phase under supercooling and stretching conditions. Experimental challenges in these conditions have led to simulations suggesting a liquid-liquid phase transition between low-density and high-density water phases, culminating in a liquid-liquid critical point (LLCP). However, these simulations are limited by computational expense, small system sizes, and reliability of water models. Using the FS model, we improve accuracy in predicting water's density and response functions across a broad range of temperatures and pressures. The FS model avoid by design first-order phase transitions towards crystalline phases, allowing thorough exploration of the metastable phase diagram. We employ advanced numerical techniques to bypass dynamical slowing down and perform finite-size scaling on systems significantly larger than those used in previous analyses. Our study extrapolates thermodynamic behavior in the infinite-system limit, accurately demonstrating the existence of the LLCP in the 3D Ising universality class at TC = 186 +/- 4 K and PC = 174 +/- 14 MPa, following a liquid-liquid phase separation below 200 MPa. These predictions align with recent experimental data and more sophisticated models, highlighting that hydrogen bond cooperativity governs the LLCP and the origin of water anomalies. Moreover, we observe that the hydrogen bond network exhibits substantial cooperative fluctuations at scales larger than 10 nm, even at temperatures relevant to biopreservation. These findings have significant implications for fields such as nanotechnology and biophysics, offering new insights into water's behavior under varied conditions.

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

Beyond Local Structures In Critical Supercooled Water Through Unsupervised Learning

Do Machine-Learning Atomic Descriptors and Order Parameters Tell the Same Story? The Case of Liquid Water

High Dimensional Fluctuations in Liquid Water: Combining Chemical Intuition with Unsupervised Learning

Automated characterization of spatial and dynamical heterogeneity in supercooled liquids via implementation of Machine Learning

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

Numerical evidence for the existence of three different stable liquid water structures as indicated by local order parameter

Human-machine collaboration: ordering mechanism of rank$-2$ spin liquid on breathing pyrochlore lattice

Pinpointing the Location of the Elusive Liquid-Liquid Critical Point in Water

Machine learning for predictive condensed-phase simulation

Dynamics of supercooled liquids from static averaged quantities using machine learning

Balancing Local Order and Long-Ranged Interactions in the Molecular Theory of Liquid Water

Uncovering novel phase transitions in dense dry polar active fluids using a lattice Boltzmann method

Signatures of a liquid–liquid transition in an ab initio deep neural network model for water

Unsupervised machine learning for supercooled liquids

The dynamics of machine-learned "softness" in supercooled liquids describe dynamical heterogeneity

Phase behavior of metastable water from large-scale simulations of a quantitatively accurate model near ambient conditions: The liquid-liquid critical point

Exploring Challenging Properties of Liquid Metallic Systems through Machine Learning: Liquid La and Li4Pb Systems

Anomalies and Local Structure of Liquid Water from Boiling to the Supercooled Regime as Predicted by the Many-Body MB-pol Model

Phase behavior of metastable water from large-scale simulations of a quantitative accurate model: The liquid-liquid critical point