A reactive force field molecular dynamics simulation of the dynamic properties of hydrogen bonding in supercritical water
A reactive force field molecular dynamics simulation of the dynamic properties of hydrogen bonding in supercritical water
复制标题
超临界水中氢键动态特性的反作用力场分子动力学模拟
DOI：
10.1016/j.molliq.2018.11.133
复制
发表时间：
2019-02-15
影响因子：
6
通讯作者：
Chen, Min
中科院分区：
文献类型：
--
作者：
Ai, Liqiang;Zhou, Yusi;Chen, Min
超临界水中氢键的动力学性质，特别是氢键的寿命和连接性，对水体系的物理化学性质起着至关重要的作用。虽然平均寿命的氢键在超临界水中得到了利用计算机模拟的报道，详细的分布和生存分析尚未进行。在这里，我们进行了经典的分子动力学（MD）模拟与最近发展的反应力场方法，研究详细的动态过程中的氢键形成，开裂，重新键合和交换在超临界水中。我们分析了MD计算的轨迹，采样间隔等于MD时间步长，以获得详细的寿命信息。计算的寿命分布表明，超临界水中的氢键表现出与温度无关的新生不稳定性。超临界水表现出非指数动力学，失效率逐渐降低。我们还研究了氢键的连接性，通过分析水分子参与一定数量的氢键的分数，发现在高密度的分数显着偏离广泛应用的简单二项分布。在假设最大协调数可以是非整数w的基础上，我们提出了一种最大协调数为4.4的非整数二项分布用于增强预测。(C)2018 Elsevier B.V.版权所有。
Dynamic properties especially lifetime and connectivity of hydrogen bonds in supercritical water play essential roles in physical and chemical properties in aqueous systems. Although average lifetime of hydrogen bonds in supercritical water derived using computer simulation has been reported, detailed distribution and survival analyses have yet to be performed. Here, we performed classical molecular dynamics (MD) simulations with a recently developed reactive force field method to investigate the detailed dynamic process of hydrogen bond forming, cracking, re-bonding and exchanging in supercritical water. We analyzed the MD-calculated trajectories with a sample interval equal to the MD time-step to obtain detailed lifetime information. The calculated lifetime distributions indicate that hydrogen bonds in supercritical water exhibit the newborn instability which is independent of temperature. Supercritical water shows non-exponential kinetics with a gradually decreasing failure rate. We also investigated the connectivity of hydrogen bonding by analyzing the fractions of water molecules involved in a certain number of hydrogen bonds and found that the fractions at high densities remarkably deviate from the widely applied simple binomial distribution. On the basis of the assumption that the maximal coordination number can be a non-integer w, we proposed a non-integer binomial distribution with a maximum coordination number of 4.4 for an enhanced prediction. (C) 2018 Elsevier B.V. All rights reserved.