Metastable materials through active learning

Metastable materials, phases that are kinetically trapped in an energetically unfavorable state, offer a tremendous playground to discover new compounds with fascinating properties. In the past, I was primarily studying pressure as a thermodynamic tuning knob to synthesize such metastable materials, and the recent discovery of the first quasi-room-temperature superconductor LaH10 has impressively demonstrated the Read More …

High-pressure materials genome – taking materials discovery to the next level

Many known materials are not thermodynamically stable, but exist as metastable states which are kinetically protected. This behavior is analogous to a skier in a mountainous terrain on a downhill trajectory: ultimately, he is bound to end up at the bottom of the hills (ground state, the lowest point in the mountain range), but he Read More …

Higher pressure ⇒ denser packing ⇒ better superconductivity!

Recently, we had reported the discovery of novel Cu-Bi phases at elevated pressures. Both the Cu11Bi7 and CuBi compounds have unexpected low densities, with one and two dimensional channels running through the materials. We showed that these voids are not completely empty, but provide room for the stereochemically active bismuth lone electron pairs. Such low density Read More …

Teaching machines how to find new materials

Machine learning is a technique that was developed in the field of computer science, but is meanwhile rapidly spreading for applications in many other disciplines. Originally intended as a tool for pattern recognition, machine learning algorithms can be taught to extract trends in large data-sets, which can then be used to make accurate predictions beyond the Read More …

Discovery of ultralow thermal conductivity in new Heusler compounds

Semiconducting half and, to a lesser extent, full Heusler compounds are promising thermoelectric materials due to their compelling electronic properties with large power factors. However, intrinsically high thermal conductivity resulting in a limited thermoelectric efficiency has so far impeded their widespread use in practical applications. Here, we report the computational discovery of a class of Read More …