Long-Time-Scale Simulations of H₂O Admolecule Diffusion on Ice Ih(0001) Surfaces

Abstract

Long-timescale simulations of the diffusion of a H 2 O admolecule on the (0001) basal plane of ice Ih were carried out over a temperature range of 100 to 200 K using the adaptive kinetic Monte Carlo method and TIP4P/2005f interaction potential function. The arrangement of dangling H atoms was varied from the proton-disordered surface to the perfectly ordered Fletcher surface. A large variety of sites was found leading to a broad distribution in adsorption energy at both types of surfaces. Up to 4% of the sites on the proton-disordered surface have an adsorption energy exceeding the cohesive energy of ice Ih. The mean squared displacement of a simulated trajectory at 175 K for the proton-disordered surface gave a diffusion constant of 6·10 −10 cm 2 /s, consistent with an upper bound previously reported from experimental measurements. During the simulation, dangling H atoms were found to rearrange so as to reduce clustering, thereby approaching a linear Fletcher type arrangement. Diffusion on the perfectly ordered Fletcher surface was estimated to be significantly faster, especially in the direction along the rows of dangling hydrogen atoms. From simulations over the range in temperature, an effective activation energy of diffusion was estimated to be 0.16 eV and 0.22 eV for diffusion parallel and perpendicular to the rows, respectively. Even a slight disruption of the rows of the Fletcher surface made the diffusion isotropic.

Atomic scale simulations of heterogeneous electrocatalysis: recent advances

“…Since sampling of water configurations can be difficult because of the relatively strong hydrogen bonds, long-timescale simulation approaches, for example, the adaptive kinetic Monte Carlo approach, which has been used in studies of i.e. surfaces [64][65][66] and diffusion in complex systems [67], could prove useful in electrochemical simulations. Another approach is to develop a functional that implicitly includes configurational averaging [68].…”

The effect of temperature and external field on transitions in elements of kagome spin ice

“…In this way, the KMC method can be made more rigorous and consistent with basic physical properties of the system. Such approaches have been developed for systems undergoing atomic rearrangements [13][14][15], for example dynamics at proton disordered ice surfaces [16,17], but have yet to be developed for magnetic systems.…”

A Practical Guide to Surface Kinetic Monte Carlo Simulations

“…However, they are generally also much more computationally demanding, as they require many orders of magnitude more energy and force evaluations to determine all processes and their barriers. Applications have therefore been limited to rather simple systems or systems where the energy and force evaluations can be done with classical force fields instead of DFT (Xu and Henkelman, 2008; Konwar et al, 2011; Pedersen et al, 2015). At least for the time being, first-principles KMC simulations in the application areas covered here are instead in practice only tractable within the rigid setup, which is why we concentrate on it from now on in this practical guide.…”