A common treatment of solid surface heterogeneous catalysts is as an unmoving crystal lattice. However, the nuances of heterogeneous catalyst nanoclusters (NC) are not well understood and will instead be sensitive to subtle variations in operation such as temperature, pressure, support, and reaction species. This results in dynamic structural transformations under operando conditions. Simulating such changes would be computationally burdensome using a traditional computational method, arising a need for a method that can describe such a complex system but can also save time.<br/><br/>We propose using a semiempirical tight-binding-based approach as implemented in DFBT+ using the GFN2-xTB method for this. This software uses a simplified quantum mechanical Hamiltonian and relies heavily on pair-wise interactions. However, these simplifications can lead to inaccurate structural and energetic data. To help combat this we have devised a method tested against DFT results to reparametrize the pair-wise interactions to better represent the structural characters of these nanoclusters.<br/><br/>To understand the accuracy of this method we modeled the absorbance of carbon monoxide (CO) on a series of platinum NC supported on a graphene sheet. To do this algorithmic generation of random clusters having two, three, and four platinum atoms as well as a range of one to eight CO molecules were done. The data resulting from using the modified GFN2 Hamiltonian in DFBT+ will be compared to experimental data as a method to evaluate the validity of this procedure. This method can then be applied to other catalyst systems in the future.

Modeling CO adsorption of platinum nanocluster ensembles on graphene

Where thousands of chemistry professionals meet to share ideas and advance scientific and technical knowledge. Spring 2023 abstracts, posters, and presentations will launch at the end of March.

ACS Spring 2023

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.