Regensburg 2022 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

MM: Fachverband Metall- und Materialphysik

MM 34: Data Driven Materials Science: Interatomic Potentials / Reduced Dimensions

MM 34.8: Talk

Thursday, September 8, 2022, 17:45–18:00, H45

How to teach my deep generative model to create new RuO₂ surface structures? — •Patricia König, Hanna Türk, Yonghyuk Lee, Chiara Panosetti, Christoph Scheurer, and Karsten Reuter — Fritz-Haber-Institut der MPG, Germany

Many widely used catalyst systems still hold complicated longstanding structural puzzles that hamper their full atomistic understanding and thus further knowledge based progress. Here, we address the well-known RuO₂ catalyst for the oxidative conversion of CO exhaust gases in combustion processes.

To explore the chemical space of RuO₂ surface structures, we trained a Generative Adversarial Network (GAN) that is capable of cheaply generating diverse structural guesses for novel surface structures. For the training set, 28,903 RuO₂ surface terminations were created with a grand-canonical basin hopping method. The atomic positions of these structures were mapped to Gaussian densities on a three-dimensional grid to generate the GAN input. We demonstrate how two-dimensional images of cuts through RuO₂ structures with inferred lattice lengths and energy conditioning can be created as a first step to realistic three-dimensional surface structures.