(509cx) Grand Canonical Monte Carlo Simulation Methods for Understanding Coke-Resistance Under Dehydrogenation Conditions

Iron-based catalysts have been shown to be attractive alternatives to platinum-group metals as hydrocarbon dehydrogenation catalysts. However, iron catalysts suffer from cracking side reactions and rapid deactivation through coke formation. Different strategies have been developed to increase coke resistance, such as mixing iron with main group elements or introducing electron-withdrawing groups. In this work, we evaluate the effectiveness of these strategies by developing a simulation protocol that uses grand canonical Monte Carlo (GCMC) with density functional theory (DFT) to sample the large ensemble of coke formation structures present on a catalyst surface. The stability of these structures is then evaluated with ab initio thermodynamics. The results show that coking structures are favored on pristine iron surfaces, which agrees with the fact that pure iron suffers from run-away coking under carbon-rich reaction condition. On all other surfaces shown to have high coke resistance in the literature, we find that the coking structures are not stable under typical dehydrogenation conditions. Thus, we find that the combination of GCMC and ab initio thermodynamics is a suitable tool set for identifying strategies to enhance coke resistance on catalyst surfaces.