(45b) High Throughput Screening of Alloy Structures for Propane Dehydrogenation Reaction

Non-oxidative propane dehydrogenation (PDH) on intermetallic alloys is a promising technology to convert large amounts of propane available in the shale gas to propylene. However, the high temperatures required for the reaction lead to reduced selectivity towards propylene formation and deactivation of the catalyst surface.[1] Even though Pt and Pd alloys have shown promise in terms of improved selectivity in comparison to pure metals, they ultimately experience loss of reactivity under reaction conditions. This difficulty motivates development of a more fundamental understanding of why Pt and Pd alloys have improved performance compared to pure metals and, further, elucidation of how such insights can be leveraged to find other alloy combinations with simultaneously high selectivity, stability, and activity for PDH.

Through our previous analysis on PdIn alloys, we have demonstrated that, as compared to terraces, the step surfaces of 1:1 alloys are highly active under PDH conditions (5 orders of magnitude higher rates). Furthermore, a comprehensive microkinetic analysis pointed to useful descriptors for activity, selectivity and stability.[2] With these descriptors as a starting point, the current study focuses on high-throughput screening of intermetallic alloy structures, including both steps and terraces, to identify reactivity trends across Pt and Pd alloys for PDH and, ultimately, find improved catalysts for this chemistry. First, we deploy a simple automated framework for adsorption and activation energy calculations on large number of chosen terrace and step alloy structures. The Materials Project is used to identify the most stable bulk structure of the alloy composition, while a surface energy analysis points to the most stable surface terminations, and CatKit is employed to identify all the distinct adsorption sites [3]. Subsequently, an in-house algorithm identifies all unique adsorbate configurations and finds the most stable adsorption site. These data are organized with a python-based databasing tool, leading to identification of BEP-type correlations between bond breaking barriers and the binding energies of various adsorbates. Further the binding energies of key descriptors were correlated with electronic and geometric properties of the alloy surfaces, using Random Forest techniques. This, therefore, allowed for a deeper understanding of the surface features that impacts the binding of descriptors the most.

References:

[1] J.J.H.B. Sattler, J. Ruiz-Martinez, E. Santillan-Jimenez, B.M. Weckhuysen, Catalytic dehydrogenation of light alkanes on metals and metal oxides, Chem. Rev. 114 (2014) 10613–10653. https://doi.org/10.1021/cr5002436.

[2] R.R. Seemakurthi, et al. Structure Sensitivity and Microkinetic Analysis of Propane Dehydrogenation on PdIn alloy and Pd, In Preparation.

[3] J.R. Boes, O. Mamun, K. Winther, T. Bligaard, Graph Theory Approach to High-Throughput Surface Adsorption Structure Generation, J. Phys. Chem. A. 123 (2019) 2281–2285. https://doi.org/10.1021/acs.jpca.9b00311.