The Fourth Research Paradigm of Materials Science

Matthias Scheffler’s research focuses on understanding fundamental aspects of physical and chemical properties of surfaces, interfaces, clusters, nanostructures, and bulk materials. Present activities include catalytic reactions at surfaces, thermal conductivity, thermoelectric materials, topological insulators, inorganic / organic hybrid materials, and more. His multi-scale modeling studies link first-principles electronic-structure calculations, ab initio molecular dynamics, and methods from thermo-dynamics and statistical mechanics to enable the understanding of meso- and macroscopic phenomena occurring under realistic (T, p) conditions. Scheffler is also engaged in developing theoretical methods to treat excited states and electron correlations, as well as computer codes to perform large-scale calculations on high-performance computers. In recent years, he developed neural-network and compressed-sensing methods to identify structure and patterns in “big data of materials”. He is coordinating the European Center of Excellence for Novel Materials Discovery (NOMAD) which maintains the largest repository for computational materials science data, a materials encyclopedia, and big-data tools in order to advance materials science and engineering.

New or even novel (new in its kind) materials are interesting for basic science and critical for advancing key technologies for energy, environment, health, mobility, IT, and more. Interestingly, from the 250,000 inorganic materials known today, less than 1% of their functional properties have been determined, so far. And the number of not-yet-known materials is practically infinite. Thus, it is likely that materials exist or can be created that are very different to and far better than those that are used today. My talk will address an innovative approach to this problem – the fourth research paradigm of materials science. I will focusing on recent advancements of artificial-intelligence (AI) methods that help to build “a map of materials” – quite in spirit of the Periodic Table of the Elements. In particular, I will emphasize the importance of knowledge-powered methods and interpretability to enable a trustful description. Applications of these methods will include a high-throughput screening study of the catalytic transformation of “bad” CO2 into fuels and other useful chemicals and topological insulators as a class of quantum materials.

Diskussionsleitung: Prof. Dr. Wolfgang Eberhardt, Wiss. Leiter Magnus-Haus Berlin

Diese Veranstaltung wird gefördert durch die Wilhelm und Else Heraeus-Stiftung.