Discovery of functional oxides through next-generation quantum alchemy (DISCOFOX)

• Applicant: Professor Dr. Guido Falk von Rudorff
• Subject Area: Theoretical Chemistry: Molecules, Materials, Surfaces
• Term: since 2026
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 576690633

Many technologies currently actively pursued to enable more sustainable energy generation and use, mobility, and environmental remediation rely critically on the properties of functional semiconducting oxides. From solid-state batteries to fuel cells to electro- and photo-catalysts, oxides are simultaneously materials providing key functionality and the bottleneck on the way to more performant devices in that there is a strong need to develop oxides with a better balance of properties while being inexpensive, easily sourced, safe, and stable. Theoretical approaches are computationally demanding due to the sheer number of the possible systems to consider. While machine learning models become more popular, they still require large amounts of training data and are interpolative in nature. Therefore, methods which do not require the upfront investment in a large representative database are desirable. In this project, we propose such a method: a perturbative electronic structure computational approach for the design of desirable functional oxides which is rooted in ab initio calculations rather than statistical modelling. Similar approaches have been successful for molecular systems but could not be transferred to materials yet, since careful consideration of the difference in electronic structure calculations and the semiconducting nature of the systems are required. By computationally varying the nuclear charges, we can connect otherwise independent systems via a smooth change of the system Hamiltonian. The corresponding alchemical derivatives (which are physically well-defined) connect those systems and allow for materials design by following the gradients w.r.t. composition for stability, defect formation energies, band gaps, band edges and band curvatures, which may prove useful in materials design.

• DFG Programme: Research Grants