Technologie Hero Istock-1462646898 Wmaster890

Atomistic Corrosion Informatics

We are focusing on approaches that put individual atoms under the spotlight and combine them with machine learning methods. This is how we want to find out how materials react to their environment.

Developing methods that enable such atomistic electrochemical simulations is one of the scientific challenges. This also includes force fields generated by machine learning. With these simulations, we can explain local electrochemical phenomena and decipher fundamental corrosion mechanisms. With our highly precise simulations, we can also provide important and difficult-to-access physical parameters for multiscale modelling. This is an important step for the development of complex, large-scale and temporally extended electrochemical models, which are being developed here at the institute. In addition, we develop predictive methods, for example to forecast the corrosion influence of small organic molecules based on the atomic structure of the molecules. This provides a link to the high-throughput experiments at the institute and helps to advance the development of new materials.

Our goal is to use artificial intelligence to gain a fundamental understanding of surface processes. In doing so, we want to pave the way for the development of advanced technologies for sustainable electrochemical applications.