Self-diffusion in MgO--a density functional study

Periclase deforms more slowly than bridgmanite under mantle conditions

Transport of heat from the interior of the Earth drives convection in the mantle, which involves the deformation of solid rocks over billions of years. The lower mantle of the Earth is mostly composed of iron-bearing bridgmanite MgSiO₃ and approximately 25% volume periclase MgO (also with some iron). It is commonly accepted that ferropericlase is weaker than bridgmanite¹. Considerable progress has been made in recent years to study assemblages representative of the lower mantle under the relevant pressure and temperature conditions^2,3. However, the natural strain rates are 8 to 10 orders of magnitude lower than in the laboratory, and are still inaccessible to us. Once the deformation mechanisms of rocks and their constituent minerals have been identified, it is possible to overcome this limitation thanks to multiscale numerical modelling, and to determine rheological properties for inaccessible strain rates. In this work we use 2.5-dimensional dislocation dynamics to model the low-stress creep of MgO periclase at lower mantle pressures and temperatures. We show that periclase deforms very slowly under these conditions, in particular, much more slowly than bridgmanite deforming by pure climb creep. This is due to slow diffusion of oxygen in periclase under pressure. In the assemblage, this secondary phase hardly participates in the deformation, so that the rheology of the lower mantle is very well described by that of bridgmanite. Our results show that drastic changes in deformation mechanisms can occur as a function of the strain rate.

Viscosity of hcp iron at Earth's inner core conditions from density functional theory

The inner core, extending to 1,221 km above the Earth’s center at pressures between 329 and 364 GPa, is primarily composed of solid iron. Its rheological properties influence both the Earth’s rotation and deformation of the inner core which is a potential source of the observed seismic anisotropy. However, the rheology of the inner core is poorly understood. We propose a mineral physics approach based on the density functional theory to infer the viscosity of hexagonal close packed (hcp) iron at the inner core pressure (P) and temperature (T). As plastic deformation is rate-limited by atomic diffusion under the extreme conditions of the Earth’s center, we quantify self-diffusion in iron non-empirically. The results are applied to model steady-state creep of hcp iron. Here, we show that dislocation creep is a key mechanism driving deformation of hcp iron at inner core conditions. The associated viscosity agrees well with the estimates from geophysical observations supporting that the inner core is significantly less viscous than the Earth’s mantle. Such low viscosity rules out inner core translation, with melting on one side and solidification on the opposite, but allows for the occurrence of the seismically observed fluctuations in inner core differential rotation.

Hot-Electron-Mediated Ion Diffusion in Semiconductors for Ion-Beam Nanostructuring

Ion-beam-based techniques are widely utilized to synthesize, modify, and characterize materials at the nanoscale, with applications from the semiconductor industry to medicine. Interactions of the beam with the target are fundamentally interesting, as they trigger multilength and time-scale processes that need to be quantitatively understood to achieve nanoscale precision. Here we demonstrate for magnesium oxide, as a testbed semiconductor material, that in a kinetic-energy regime in which electronic effects are usually neglected, a proton beam efficiently excites oxygen-vacancy-related electrons. We quantitatively describe the excited-electron distribution and the emerging ion dynamics using first-principles techniques. Contrary to the common picture of charging the defect, we discover that most of the excited electrons remain locally near the oxygen vacancy. Using these results, we bridge time scales from ultrafast electron dynamics directly after impact to ion diffusion over migration barriers in semiconductors and discover a diffusion mechanism that is mediated by hot electrons. Our quantitative simulations predict that this mechanism strongly depends on the projectile-ion velocity, suggesting the possibility of using it for precise sample manipulation via nanoscale diffusion enhancement in semiconductors with a deep, neutral, intrinsic defect.