High-Throughput Computational Screening of Metal-Organic Frameworks for Thiol Capture

We report high-throughput computational screening of 137 953 hypothetical metal-organic frameworks (hMOFs) and 4764 computation-ready experimental MOFs (CoRE-MOFs) for the capture of thiols (methanethiol and ethanethiol) from air. To minimize the competitive adsorption of moisture, 31 816 hydrophobic MOFs are first identified on the basis of a threshold criterion in the Henry constant of water, and then used to assess the adsorption capacity of thiol (N-SH) and the selectivity of thiol over air (S-SH/Air). The highest N-SH and S-SH/Air, are predicted to be 70.86 mg/g and 2.6 x 10(7), respectively. Most of the high-performance MOFs are found to be hMOFs. The structure-property relationships are derived for N-SH and S-SH/Air with MOF descriptors (including the isosteric heat, the largest cavity diameter, surface area, and void fraction). While the relationship with isosteric heat tends to be monotonic, there exist optimal ranges in the other relationships. Principal component analysis is applied to assess the interrelationships among the four descriptors; then multiple linear regression is used to quantitatively determine the respective effects of descriptors on N-SH and S-SH/Air. It is revealed that the isosteric heat is a key descriptor governing thiol capture. Moreover, decision tree modeling is employed to define a clear effective path to screen high-performance MOFs. Finally, the best MOFs are identified. The microscopic insights obtained from our bottom-up approach are useful toward the development of MOFs and other nanoporous materials for thiol capture from air or in a variety of environmental and industrial situations.