Machine learning (ML) is increasingly becoming a helpful tool in the search for novel functional compounds. Here we use classification via random forests to predict the stability of half-Heusler (HH) compounds, using only experimentally reported compounds as a training set. Cross-validation yields an excellent agreement between the fraction of compounds classified as stable and the actual fraction of truly stable compounds in the ICSD. The ML model is then employed to screen 71,178 different 1:1:1 compositions, yielding 481 likely stable candidates. The predicted stability of HH compounds from three previous high throughput ab initio studies is critically analyzed from the perspective of the alternative ML approach. The incomplete consistency among the three separate ab initio studies and between them and the ML predictions suggests that additional factors beyond those considered by ab initio phase stability calculations might be determinant to the stability of the compounds. Such factors can include configurational entropies and quasihar-monic contributions.

So far most kinetic Monte Carlo (kMC) simulations of heterogeneously catalyzed gas phase reactions were limited to flat crystal surfaces. The newly developed program MoCKA (Monte Carlo Karlsruhe) combines graph-theoretical and lattice-based principles to be able to efficiently handle multiple lattices with a large number of sites, which account for different facets of the catalytic nanoparticle and the support material, and pursues a general approach, which is not restricted to a specific surface or reaction. The implementation uses the efficient variable step size method and applies a fast update algorithm for its process list. It is shown that the analysis of communication between facets and of (reverse) spillover effects is possible by rewinding the kMC simulation. Hence, this approach offers a wide range of new applications for kMC simulations in heterogeneous catalysis.

No abstract

Review: 188 refs.

Understanding the domain structure of ferroelectric ceramics is very important to develop sound knowledge of the influence of the microstructure on the macroscopic properties. To proceed in this direction, experimental tools are necessary in order to quantify the domain patterns in ferroelectrics. This study on BaTiO 3 single crystals exemplifies how vector piezoresponse force microscopy can be used to obtain statistical information about domain directions.