In April 2024, the Federal Ministry of Education and Research (BMBF) approved the project: "Reduction of nitrous oxide emissions and precious metal consumption in the industrial oxidation of ammonia through AI-based development of new catalysts"

With the help of machine learning methods, new metal alloys for ammonia oxidation are to be developed that produce as little nitrous oxide (a greenhouse gas with a strong impact on the climate) as possible and do not corrode within a few months (previously approx. 0.5g Pt loss / tonne HNO3). The process was developed over a hundred years ago by Wilhelm Ostwald in Leipzig (Ostwald process) and despite decades of searching for new catalyser materials, platinum is still used today. The project aims to use descriptor-based machine learning methods, combined with the elucidation of the corrosion mechanism of platinum-based catalysts, to find new, more selective and more stable catalysts for this process that are just as active as platinum but contain as few or no precious metals as possible.

The project partners are: Reacnostics (consortium leader, CFD simulations, software development), TUHH (elucidation of the corrosion mechanism by laboratory and synchrotron experiments), Fritz Haber Institute Berlin, Theory Department (development of machine learning algorithms and descriptors), TU Munich (operando electron microscopy), YARA Norway (user & industrial partner).

The project volume amounts 2.4 Mio. Euro, of which 1.147 Mio. go to the Technical University of Hamburg.