High-Throughput Computation Evaluation of Metal-Organic Frameworks for Efficient Perfluorocarbons Recovery

The recovery of perfluorocarbons (PFCs), such as CF4 and C2F6, from exhaust gas can not only reduce the emissions of greenhouse gas but also improve the utilization of PFCs in the semiconductor industry. In this work, a high-throughput computational evaluation for nearly 10 000 MOFs in the CoRE MOF database was performed to evaluate the potential of metal-organic frameworks (MOFs) for the recovery of trace CF4 and C2F6 from N-2-containing gas. Various adsorbent performance metrics, including adsorption selectivity, working capacity, recovery rate, and adsorbent performance score, were calculated to evaluate the top-performing MOFs, and 10 top-performing MOFs for efficient capture of CF4 and C2F6 over N-2 were identified from a computation-ready experimental (CoRE) MOF database. The machine learning model analysis reveals that the LCD as well as the adsorption heat difference between PFCs with N-2 play dominant roles in PFCs recovery. Furthermore, five design and optimization strategies, including adjustment or functionalization of the organic linker, substitution of metal node, regulation of topology net, and optimization of synthesis condition, were provided to guide the development of high-performing MOFs for PFCs recovery.