Breaking the barrier of time in computational mineral physics: the example of vacancy diffusion in periclase

Abstract:

Point defects play an essential role in several fields including diffusive phase transformations and mechanical properties in the diffusion creep regime. Vacancies, have generally higher migration energy barriers than other defects which makes their diffusion the controlling factor for these processes. High confining pressures such as those encountered in the mantle tend to slow down diffusion and makes experimental determinations of diffusion coefficients very challenging. This limitation is also met by computational approaches. Atomistic simulations like molecular dynamics can handle very short time durations only. In both cases, the only possibility is to carry experiments and calculations at extremely high temperatures and to extrapolate results to the range of interest. We introduce here a new computational technique which allows to overcome these limitations. The kinetic Activation Relaxation Technique (k-ART), is an off-lattice kinetic Monte Carlo method with on-the-fly catalog building. Using k-ART, we study the diffusion of vacancies in MgO crystal which allows the access time-scales which are unreachable by standard methods. With our approach we cover more than 15 orders of magnitude in time from nano to thousands of seconds leading to the calculation of diffusion coefficients on a very large temperature range (500-2000 K). In addition, the characterisation of the exact kinetic paths of these defects by the calculation of energy barriers sheds new light on the vacancy diffusion mechanism. In particular, we describe how oxygen and magnesium vacancies attract together and aggregate to form a low energy bound configuration (divacancy). We describe the mechanisms which allow this divacancy to move as an aggregated defect.