High-throughput computational screening of Covalent−Organic framework membranes for helium purification

Abstract

Currently, helium purification is carried out by the heat-driven cryogenic distillation process, which is prohibitively energy-intensive and capital-intensive. As an alternative, covalent-organic frameworks (COFs)-based membrane separation technology is considered to be cost-efficient and technically feasible. In this study, a high-throughput computational screening strategy was proposed for the rapid determination of high-performing COF membranes for this particular application. A geometric analysis-based prescreening process was carried out on 688 experimentally synthesized COFs, resulting in 665 COF candidates with pore limited diameter (PLD) no less than 3.8 Å. Subsequently, molecular simulations were performed to estimate the henry coefficients and self-diffusion coefficients of N2, H2, and CH4 in each COF. Based on the results, the membrane selectivity and permeability of all the COFs were calculated, from which 23 high-performing COF membranes were identified. Additionally, the performances of these 23 COF membranes for a ternary mixture of He/N2/CH4 were computationally examined, five of which were highly preferable to permeate He over N2 or CH4. The results indicated that a large number of COF membranes exceeded the performance upper limit of polymer membranes owing to their high He permeability. Additionally, the relation between the PLD of COFs and the corresponding separation performances was investigated. Finally, He permeability, as well as He/CH4 and He/N2 selectivity of mixed matrix membranes were estimated by using the Maxwell model. The results revealed that the addition of COFs into pure polymers significantly improved the permeability of He up to 11689 Barrer.