摘要
Water confined in nanopores is ubiquitous in geological, planetary and biological environments and in nanofluidic settings. Understanding the phase behaviour and proton dynamics of nanoconfined water under high-pressure conditions is therefore important from both fundamental and applied points of view. Here we report a machine learning potential and present evidence from large-scale path-integral molecular dynamics simulations of the proton dynamics and phase behaviours of monolayer and bilayer ice under nanoconfinement and high pressures. We find that the symmetry breaking of the underlying hydrogen-bonding network of the two-dimensional (2D) ices together with strong nuclear quantum effects are responsible for the rich proton dynamics, such as the ultrafast one-dimensional proton-hopping within 2D ices. We also predict ten 2D ice phases. Notably, a 2D dynamic partially ionic phase and a superionic phase can be produced in the laboratory at pressures one order of magnitude lower than those measured for the bulk superionic phase or predicted for the partially ionic phase. We also identify a 2D solid-melting behaviour, namely consecutive double or triple continuous phase transitions from bilayer molecular ice to plastic ice and then to hexatic ice and the superionic fluid. @@@ The phase diagram of confined ice is different from that of bulk ice. Simulations now reveal several 2D ice phases and show how strong nuclear quantum effects result in rich proton dynamics in 2D confined ices.
-
单位中国科学院; 中山大学; 国家纳米科学中心