Charged ultrafiltration membranes for whey protein fractionation

Although ultrafiltration offers a number of advantages for the concentration of proteins, its use for protein fractionation is limited to the separation of proteins of widely different molecular weights. Charged ultrafiltration membrane technology is a relatively new concept which promises to widen the use of ultrafiltration for the concentration and fractionation of proteins. The central hypothesis of this research is that charged ultrafiltration membranes can be used to fractionate bioactive whey proteins. This hypothesis is based on the fact that although the different whey proteins do not differ greatly in molecular weight, the isoelectric points are markedly different, and hence membrane charge and protein charge can be utilized to enhance the extent of separation.;The ultrafiltration experiments were performed for the separation of different whey proteins. In particular, the research targeted the acidic whey protein glycomacropeptide. The separation achieved was enhanced by using charged ultrafiltration membranes and by optimization of the physicochemical properties of the solution (pH and conductivity) and operating conditions of the process (flux and mass transfer). Both dead-end and crossflow modes of operations were examined. The data were fitted using the stagnant film model that relates the observed sieving coefficient to flux, mass transfer coefficient, and concentration driving force. The concept of constant concentration polarization was also used to maintain a constant concentration of protein at the membrane surface in order to minimize the loss of protein through the membrane during the protein concentration experiments.;Charged membranes enhanced the selectivity of separation of glycomacropeptide from the other proteins in whey. While using a binary mixture of glycomacropeptide and betalactoglobulin, the selectivity was enhanced by over 600% as compared to an uncharged membrane. Using cheese whey, the increase in selectivity was about 220% for the charged versus uncharged membrane. The stagnant film model was found to fit the data well, and was a good predictor of the observed sieving coefficient versus flux.;This work showed that charged membranes can be used to successfully fractionate similarly sized proteins from whey and also provided a theoretical basis for data analysis and scale up.