Computational Screening of Porous Organic Molecules for Xenon/Krypton Separation

We performed a computational screening of previously reported porous molecular materials, including porous organic cages, cucurbiturils, cyclodextrins, and cryptophanes, for Xe/Kr separation. Our approach for rapid screening through analysis of single host molecules, rather than the solid state structure of the materials, is evaluated. We use a set of tools including in-house software for structural evaluations, electronic structure calculations for guest binding energies, and molecular dynamics and metadynamics simulations to study the effect of the hosts' flexibility upon guest diffusion. Our final results confirm that the CC3 cage molecule, previously reported as high performing for Xe/Kr separation, is the most promising of this class' of materials reported to date. The Noria molecule was also found to be promising, and we therefore synthesized two related Noria molecules and tested their performance for Xe/Kr separation in the laboratory.