Supercritical water at near-experimental accuracy with MB-pol(2023)

Area of Science:

• Physical Chemistry
• Computational Materials Science
• Fluid Dynamics

Background:

• Supercritical water (SCW) exhibits complex thermophysical properties challenging for computational modeling.
• Disruption of hydrogen bonds, many-body interactions, and density fluctuations complicate SCW simulations.

Purpose of the Study:

• To evaluate the data-driven many-body MB-pol(2023) potential for accurately simulating SCW.
• To assess the model's ability to reproduce the liquid-to-supercritical crossover and anomalous behaviors of water.

Main Methods:

• Molecular dynamics simulations using the MB-pol(2023) potential.
• Analysis of thermophysical properties (density, compressibility, expansivity, heat capacity).
• Structural and dynamical analysis of hydrogen-bond networks and molecular mobility.

Main Results:

• MB-pol(2023) quantitatively reproduces SCW properties with near-experimental accuracy (400-900 K, 25-100 MPa).
• Accurate prediction of anomalous behaviors near the critical region, including a pseudo-critical Widom-line temperature of ~660 K.
• Demonstrated continuous collapse of hydrogen-bond network and density-driven fluid behavior.

Conclusions:

• MB-pol(2023) is a reliable framework for describing water across the liquid-to-supercritical continuum.
• The model accurately captures SCW's density-driven nature and molecular dynamics.
• Confirms the non-reactive MB-pol(2023) model's validity under extreme conditions.