Membrane formation via non-solvent induced phase separation using sustainable solvents: A comparative study

Abstract

Non-solvent induced phase separation (NIPS) is a frequently used technique for the production of polymeric membranes. It enables the production of membranes with a broad range of different characteristics. Current solvents used in membrane preparation are often toxic, environmentally unfriendly and prepared from non-sustainable resources. This is why a replacement of solvents like N-methyl-2-pyrrolidone (NMP) and dimethylacetamide (DMAc) is highly desirable. In order to substitute a solvent whilst achieving the same desired membrane properties, it is necessary to understand the formation mechanisms and its influencing factors. One important set of parameters for controlling the membrane features is the polymer solution composition. This is why the aim of this study was to improve the understanding of membrane formation by gaining a holistic picture of the influences of systematic additive variations, focusing on the comparison between conventional and alternative sustainable solvent systems. Thus, 72 different polyethersulfone (PES) membrane prototypes were produced by immersion precipitation from polymer solutions prepared in NMP and DMAc, as well as in the sustainable alternatives 2-pyrrolidone (2P) and dimethyllactamide (DML). In all four solvent systems varying concentrations and molecular weights of the polymeric additives polyvinylpyrrolidone (PVP) and polyethylene glycol (PEG) were applied. The viscosity of the polymer solutions was determined, and thereof formed membranes were analyzed in terms of permeability, protein retention, surface properties, mechanical stability and morphology. The results indicate that both, solvents and additives, significantly impact the membrane properties. It was shown that the influences of the additives on all investigated membrane features were strongly dependent on the applied solvent. The observed effects were similar for the conventional solvents NMP and DMAc, but differed from those found for the alternative solvents 2P and DML, which among themselves also showed comparable outcomes. In conclusion, this study proves that it is possible to obtain desired membrane properties with 2P or DML as long as the solution composition is chosen appropriately.