High-throughput screening of hypothetical aluminosilicate zeolites for CO2 capture from flue gas

To accelerate the discovery of promising adsorbents for CO2 capture from flue gas, many efforts have been made toward the high-throughput computational prediction of new high-efficiency porous materials, such as aluminosilicate zeolites. To reduce the computational cost, most of the previous efforts evaluate the CO2 capture capability according to the CO2 adsorption selectivity estimated by empirical single-component simulations. In this study, we perform high-throughput grand canonical Monte Carlo (GCMC) simulations to predict the CO2 capture capability of 2625 aluminosilicate zeolite structures, which are selected from 262,500 hypothetical ABC-6 models with Si/Al ratios ranging from 1.0 to 10. Different from previous efforts, our GCMC simulations are conducted on CO2/N-2 binary mixtures, which are much closer to the realistic compositions of flue gas than conventional single-component simulations. Moreover, in addition to conventional CO2 adsorption selectivity, we introduce working capacity, adsorbent performance score, and regenerability to get a comprehensive overview of the performance of these candidate structures under temperature swing adsorption (TSA), vacuum swing adsorption (VSA), and pressure swing adsorption (PSA) processes, respectively. We identify hundreds of hypothetical aluminosilicate zeolite structures that exhibit promising CO2 capture capability in TSA and VSA processes even superior to zeolite Na-X, the most efficient zeolite for CO2 capture known so far. Our results will provide important guidance toward experimental discovery of high-performance zeolites for practical CO2 capture