Viscosity and Stokes-Einstein relation in deeply supercooled water under pressure

TL;DR

This study measures water's viscosity under high pressure and deep supercooling, revealing non-Arrhenius behavior and pressure effects, and examines the Stokes-Einstein relation, suggesting a possible liquid-liquid critical point in water.

Contribution

First experimental measurement of water viscosity at high pressure and deep supercooling, analyzing its relation to diffusion and implications for water's phase behavior.

Findings

Viscosity decreases non-monotonically with pressure at low temperatures.

Temperature dependence of viscosity is non-Arrhenius across all pressures.

Data supports the existence of a liquid-liquid critical point in water.

Abstract

We report measurements of the shear viscosity $\eta$ in water up to $150\,\mathrm{MPa}$ and down to $229.5\,\mathrm{K}$. This corresponds to more than $30\,\mathrm{K}$ supercooling below the melting line. The temperature dependence is non-Arrhenius at all pressures, but its functional form at $0.1\,\mathrm{MPa}$ is qualitatively different from that at all pressures above $20\,\mathrm{MPa}$. The pressure dependence is non-monotonic, with a pressure-induced decrease of viscosity by more than 50 % at low temperature. Combining our data with literature data on the self-diffusion coefficient $D_\mathrm{s}$ of water, we check the Stokes-Einstein relation which, based on hydrodynamics, predicts constancy of $D_\mathrm{s} \eta/T$, where $T$ is the temperature. The observed temperature and pressure dependence of $D_\mathrm{s} \eta/T$ is analogous to that obtained in simulations of a realistic water model. This analogy suggests that our data are compatible with the existence of a liquid-liquid critical point at positive pressure in water.