Prediction of MOF Performance in Vacuum Swing Adsorption Systems for Postcombustion CO₂ Capture Based on Integrated Molecular Simulations, Process Optimizations, and Machine Learning Models

. Machine learning models were developed using common adsorption metrics to predict a material's ability to meet the 95/90-PRT with an overall prediction accuracy of 91%. It was found that accurate parasitic energy and productivity estimates of a VSA process require full process simulations.