Department of Mechanical and Aerospace Engineering, University of California, Davis, California 95616, USA.
Lawrence Livermore National Laboratory, Livermore, California 94550, USA.
J Chem Phys. 2019 Oct 28;151(16):164501. doi: 10.1063/1.5125948.
Classical nucleation theory (CNT) is a promising way to predictively model the submicrosecond kinetics of phase transitions that occur under dynamic compression, such as the suite of experiments performed over the past two decades on the solidification of liquid water to the high-pressure ice VII phase. Myint et al. [Phys. Rev. Lett. 121, 155701 (2018)] presented the first CNT-based model for these types of rapid phase transitions, but relied on an empirical scaling parameter in their transient induction model to simulate the lag time that occurs prior to the onset of significant formation of ice VII clusters in the system. To build on that study, we model the liquid water-ice VII phase transformation using a numerical discretization scheme to solve the Zel'dovich-Frenkel partial differential equation, which is a fundamental CNT-based kinetic equation that describes the statistical time-dependent behavior of solid cluster formation. The Zel'dovich-Frenkel equation inherently accounts for transience in the nucleation kinetics and eliminates the need for the empirical scaling factor used by Myint et al. One major result of this research is that transience is found to play a relatively small role in the nucleation process for the dynamic-compression time scales of the liquid water-ice VII experiments being simulated. Instead, we show that it is possible to properly model the lag time using steady-state CNT by making small refinements to the interfacial free energy value. We have also developed a new dimensionless parameter that may be applied a priori to predict whether or not transient nucleation will be important in a given dynamic-compression experiment.
经典成核理论（CNT）是一种很有前途的方法，可以对在动态压缩下发生的亚微秒级相变动力学进行预测性建模，例如过去二十年在对液态水向高压冰 VII 相的凝固进行的一系列实验中所进行的实验。Myint 等人[Phys. Rev. Lett. 121, 155701 (2018)]提出了第一个用于此类快速相变的基于 CNT 的模型，但在他们的瞬态感应模型中依赖经验缩放参数来模拟系统中冰 VII 团簇显著形成之前发生的滞后时间。在此研究的基础上，我们使用数值离散化方案对液态水-冰 VII 相变进行建模，以求解 Zel'dovich-Frenkel 偏微分方程，这是一个基本的基于 CNT 的动力学方程，描述了固体团簇形成的统计时变行为。Zel'dovich-Frenkel 方程内在地考虑了成核动力学中的瞬态，并且不需要 Myint 等人使用的经验缩放因子。这项研究的一个主要结果是，在模拟的液态水-冰 VII 实验的动态压缩时间尺度下，瞬态在成核过程中只起到相对较小的作用。相反，我们通过对界面自由能值进行微小改进，证明了使用稳态 CNT 可以正确地对滞后时间进行建模。我们还开发了一个新的无量纲参数，该参数可以在给定的动态压缩实验之前应用于预测瞬态成核是否重要。
