Computer simulation study of surface wave dynamics at the crystal-melt interface

Abstract

We study, by means of computer simulations, the crystal-melt interface of three different systems: hard-spheres, Lennard Jones and the TIP4P/2005 water model. In particular, we focus on the dynamics of surface waves. We observe that the processes involved in the relaxation of surface waves are characterized by distinct time scales: a slow one related to the continuous recrystallization and melting, that is governed by capillary forces; and a fast one which we suggest to be due to a combination of processes that quickly cause small perturbations to the shape of the interface (like e. g. Rayleigh waves, subdiffusion, or attachment/detachment of particles to/from the crystal). The relaxation of surface waves becomes dominated by the slow process as the wavelength increases. Moreover, we see that the slow relaxation is not influenced by the details of the microscopic dynamics. In a time scale characteristic for the diffusion of the liquid phase, the relaxation dynamics of the crystal-melt interface of water is around one order of magnitude slower than that of Lennard Jones or hard spheres, which we ascribe to the presence of orientational degrees of freedom in the water molecule. Finally, we estimate the rate of crystal growth from our analysis of the capillary wave dynamics and compare it with previous simulation studies and with experiments for the case of water.

Capillary fluctuations of surface steps: An atomistic simulation study for the model Cu(111) system

“…From Fig. 5 we also see that the long-wavelength modes (small k n ) are insensitive to the choice of coarse-graining algorithm, as observed before in this type of analysis [42,43]. At intermediate values of k n the cluster algorithm is observed to follow the CFM prediction, Eq.…”

Structure and fluctuations of the premelted liquid film of ice at the triple point

“…When analyzing ice/water systems we follow the same procedure as in Ref. [75]. In this procedure we get rid of fluid-like molecules by making use of the order parameter and the biggest solid cluster is found.…”

A study of the ice–water interface using the TIP4P/2005 water model

“…1, where particles in the ice phase are shown in orange and particles in the liquid phase are shown in blue. Particles are labelled as ice or liquid-like based on local bond order parameters [17,18]. By knowing which particles belong to each phase, one can establish a discretized interface profile along the x direction, h(x n ) (in Ref.…”

“…By knowing which particles belong to each phase, one can establish a discretized interface profile along the x direction, h(x n ) (in Ref. [18] a detailed explanation of the way we establish h(x n ) is given). Then, h(x n ) is Fourier-transformed,…”