Computational mineral physics and the physical properties of perovskite

Merits and Demerits of Machine Learning of Ferroelectric, Flexoelectric, and Electrolytic Properties of Ceramic Materials

In the present review, the merits and demerits of machine learning (ML) in materials science are discussed, compared with first principles calculations (PDE (partial differential equations) model) and physical or phenomenological ODE (ordinary differential equations) model calculations. ML is basically a fitting procedure of pre-existing (experimental) data as a function of various factors called descriptors. If excellent descriptors can be selected and the training data contain negligible error, the predictive power of a ML model is relatively high. However, it is currently very difficult for a ML model to predict experimental results beyond the parameter space of the training experimental data. For example, it is pointed out that all-dislocation-ceramics, which could be a new type of solid electrolyte filled with appropriate dislocations for high ionic conductivity without dendrite formation, could not be predicted by ML. The merits and demerits of first principles calculations and physical or phenomenological ODE model calculations are also discussed with some examples of the flexoelectric effect, dielectric constant, and ionic conductivity in solid electrolytes.

The physical and chemical composition of the lower mantle

This article reviews some of the recent advances made within the field of mineral physics. In order to link the observed seismic and density structures of the lower mantle with a particular mineral composition, knowledge of the thermodynamic properties of the candidate materials is required. Determining which compositional model best matches the observed data is difficult because of the wide variety of possible mineral structures and compositions. State-of-the-art experimental and analytical techniques have pushed forward our knowledge of mineral physics, yet certain properties, such as the elastic properties of lower mantle minerals at high pressures and temperatures, are difficult to determine experimentally and remain elusive. Fortunately, computational techniques are now sufficiently advanced to enable the prediction of these properties in a self-consistent manner, but more results are required.A fundamental question is whether or not the upper and lower mantles are mixing. Traditional models that involve chemically separate upper and lower mantles cannot yet be ruled out despite recent conflicting seismological evidence showing that subducting slabs penetrate deep into the lower mantle and that chemically distinct layers are, therefore, unlikely.Recent seismic tomography studies giving three-dimensional models of the seismic wave velocities in the Earth also base their interpretations on the thermodynamic properties of minerals. These studies reveal heterogeneous velocity and density anomalies in the lower mantle, which are difficult to reconcile with mineral physics data.

The elasticity of the MgSiO3 post-perovskite phase in the Earth's lowermost mantle

MgSiO3 perovskite has been assumed to be the dominant component of the Earth's lower mantle, although this phase alone cannot explain the discontinuity in seismic velocities observed 200-300 km above the core-mantle boundary (the D" discontinuity) or the polarization anisotropy observed in the lowermost mantle. Experimental and theoretical studies that have attempted to attribute these phenomena to a phase transition in the perovskite phase have tended to simply confirm the stability of the perovskite phase. However, recent in situ X-ray diffraction measurements have revealed a transition to a 'post-perovskite' phase above 125 GPa and 2,500 K--conditions close to those at the D" discontinuity. Here we show the results of first-principles calculations of the structure, stability and elasticity of both phases at zero temperature. We find that the post-perovskite phase becomes the stable phase above 98 GPa, and may be responsible for the observed seismic discontinuity and anisotropy in the lowermost mantle. Although our ground-state calculations of the unit cell do not include the effects of temperature and minor elements, they do provide a consistent explanation for a number of properties of the D" layer.