| Nanofiltration (NF) is a membrane technique between ultrafiltration (UF) and reverse osmosis (RO), and its application has gained popularity in water treatment, food, pharmaceutical and chemical industries. Solvent resistant nanofiltration (SRNF) may be considered as an extension to the NF process applied in organic solvent system. With the development of stable and high performance SRNF membranes, research on SRNF membrane transport mechanisms has gained much interest in recent years. Compared to traditional NF processes, SRNF is less predictable due to the complicated solvent-solute-membrane interactions and the limited knowledge on the membrane transport mechanism, i.e. whether transport (both the solvent and solute) through SRNF membrane in organic solvents is dominated by viscous flow or diffusion. Present transport models have various limitations and a generalized model is not available, which hamper the practical applications of SRNF. Therefore, investigation into the transport mechanisms of NF in organic solvents is a priority research area in the world.Solvent transport in SRNF membranes was studied using common solvents (methanol, ethanol, toluene, acetone, etc.). Considering the dependence of solvent flux both on solvent properties and on membrane-solvent interactions, a correlation analysis was performed between the fluxes of 23 solvents and their properties, membrane-solvent interaction parameters. Significant correlation coefficients were found between solvent flux and four parameters, i.e. solvent viscosity, solvent dielectric constant, membrane-solvent solubility parameter difference and membrane-solvent surface tension difference. A semi-empirical solvent transport model was developed based upon the solution-diffusion with imperfection (SDI) model, and its applicability was demonstrated by describing solvent transport behavior in a wide range of solvents.This work systematically investigated the feasibility of separating binary solvent mixtures with SRNF membranes. A series of binary solvent mixtures with different physical and chemical properties (dielectric constant, surface tension, Hansen solubility parameter and viscosity) was chosen, and single factor analysis was performed to tap the potential of solvent separation using different commercial SRNF membranes (MPFTM series, StarMemTM series, DuraMemTM Series). It was found that the DuraMemTM 150 allowed the separation of ethanol/hexane mixtures with a separation factor of around 6, which is the highest for solvent separation in SRNF so far. The presence of preferential permeation was found to be mainly dependent on the membrane type used and differences in the Hansen solubility parameter and polarity between the solvents and membrane, and some hypothesis of great importance that could make the solvent separation available in SRNF were put forward for the first time. This work developed an improved model derived from the classical solution-diffusion model, specifically for solvent separation process in SRNF. Important assumptions for this process were adopted and only two parameters, i.e. the Hansen solubility parameter and dielectric constant, were integrated in this model. Separation factors of DuraMemTM 150,200,300 and MPF-44 fitted well with this new model and that of DuraMemTM 500 and DuraMemTM 900 had a poor correlation, which was consistent with the author's hypotheses. This model provides a new approach to understand solvent separation processes in SRNF, and offers a good breakthrough point for further work in this field. |
PDF Full Text Request