Optimization of polyvinylidene fluoride (PVDF) membrane fabrication for protein binding using statistical experimental design.

Statistical experimental design was employed to optimize the preparation conditions of polyvinylidenefluoride (PVDF) membranes. Three variables considered were polymer concentration, dissolving temperature, and casting thickness, whereby the response variable was membrane-protein binding. The optimum preparation for the PVDF membrane was a polymer concentration of 16.55 wt%, a dissolving temperature of 27.5°C, and a casting thickness of 450 µm. The statistical model exhibits a deviation between the predicted and actual responses of less than 5%. Further characterization of the formed PVDF membrane showed that the morphology of the membrane was in line with the membrane-protein binding performance.