A computational study on molecular transport mechanisms and models for alcohol recovery by polymer pervaporation membranes

Abstract Model development for bio‐alcohol recovery via pervaporation (PV) membranes has been hindered by insufficient molecular‐level transport insights. We conducted molecular dynamics simulations to unravel the transport mechanism of solvent (water and alcohol) molecules in polymer membranes. For membranes with different hydrophobicity, alcohol concentrations at both solution‐membrane and vapor–liquid interfaces were similar and approximated the activity in aqueous solution. We developed a thermodynamic cycle method for calculating interaction parameters and applied Flory‐Huggins theory to quantify the solubility of solvent molecules. In dilute alcohol solutions (≤5 wt%), low solubility promotes molecular‐level dispersion of solvent molecules in membranes, making diffusion coefficients nearly solubility independent. Based on these molecular‐level insights, the factors affecting the separation factor/relative volatility of PV/distillation processes were quantified using proposed coefficients. These coefficients provide an intuitive understanding of why the separation factor of various polymer membranes is higher or lower than the relative volatility.