CO2 capture in cation-exchanged metal–organic frameworks: Holistic modeling from molecular simulation to process optimization

• CO2 capture from dry flue gas is investigated by holistic multi-scale modeling.
• Pure-component adsorption and diffusion data are provided by molecular simulation.
• A four-step vacuum swing adsorption process is simulated and optimized.
• Energy penalty-productivity Paretos are presented for CO2 capture in rho-ZMOFs.

While CO2 capture has been extensively investigated in different metal–organic frameworks (MOFs), their performance under practical process conditions is scarcely examined. In this study, a multi-scale modeling study is reported to examine CO2 capture from flue gas by vacuum swing adsorption (VSA) process using cation-exchanged rho zeolite-like MOFs (rho-ZMOFs) as adsorbents. Three rho-ZMOFs namely Na-, Mg- and Al-rho-ZMOFs are considered and compared with 13X Zeolite, which is the current benchmark for post-combustion capture from dry flue gas. First, Monte Carlo (MC) simulations are conducted to estimate the adsorption isotherms of pure CO2 and N2 in the range from 0 to 100 kPa and their mixtures of varying composition at a total pressure of 100 kPa. The pure gas isotherms are then fitted by the dual-site Langmuir model. Subsequently, the extended dual-site Langmuir model is used to satisfactorily predict the equilibrium behavior of CO2/N2 binary mixtures over a wide range of composition. In addition, micropore diffusivities are calculated from molecular dynamics (MD) simulations and compared with the estimated macropore diffusivities in order to determine the controlling mechanism of mass transfer. Macropore is expected when these adsorbents are pelletized before using in a separation process. For each rho-ZMOF adsorbent, a four-step VSA process with light product pressurization for CO2 capture and concentration (CCC) from dry flue gas containing 15% CO2 in balance N2 is then simulated using a non-isothermal, non-isobaric model. The process is optimized using multi-objective optimization based on a genetic algorithm for minimizing the energy penalty and maximizing the process productivity, subject to the purity and recovery constraints of 95 and 90%, respectively. The operating spaces of the three rho-ZMOFs are similar to that of 13X zeolite. However, the minimum energy penalty in Al-rho-ZMOF (156 kW h/t CO2) is lower than in 13X (165 kW h/t CO2). While identifying the cation-exchanged rho-ZMOFs as interesting candidates for CO2 capture, this study also demonstrates that the multi-scale modeling approach adopted here is an effective methodology to screen and design novel MOFs for CCC and other separation applications.

Figure optionsDownload high-quality image (262 K)Download as PowerPoint slide