High-throughput GCMC simulation study on helium purification from CH 4 /N 2 /He mixture in pure silica zeolites

Helium, as a strategic resource, is irreplaceable in fields such as healthcare and semiconductors. However, its safe, efficient, and low-cost purification remains a challenge that needs urgent resolution. In recent years, adsorption separation based on porous materials has received extensive attention in the purification of helium. This work first theoretically charactered the structural parameters of 256 pure silica zeolites, including the largest cavity diameter (LCD) of 2.70-15.87 Å, pore limiting diameter (PLD) of 0.84-11.72 Å, accessible surface area (ASA) of 0–2829 m2/g, pore-accessible volume (POAV) of 0.11-0.99 cm3/g, and porosity of 0.01-0.70. Then the adsorption separation properties of CH4/N2/He in 132 zeolite structures selected were studied using the high-throughput grand canonical Monte Carlo (GCMC) simulation. Subsequently, the correlations between structural parameters and adsorption separation properties were studied using machine learning technique. The results indicate that FER zeolite exhibits the highest selectivity of 33.65 for CH4 + H2 over He. Furthermore, the interaction mechanisms between gas molecules and several top-performing structures were further investigated at the molecular level. It is hoped that the results of this study can provide theoretical guidance for experimental research on the adsorption, separation, and purification of helium using pure silica zeolites.