Factors that influence the membrane area of a multistage microfiltration process required to produce a micellar casein concentrate1

The objective of the work reported in this paper was to develop a theoretical model to determine the effect of type of microfiltration (MF)-process feed, number of stages, and flux on the minimization of the MF membrane area required to produce a 95% serum protein-reduced micellar casein concentrate. The MF feed, number of stages, and flux were all factors that had an effect on the MF membrane area and should be taken into consideration when designing a MF system to produce a 95% serum protein-reduced micellar casein concentrate. Feeding the MF process with a diluted ultrafiltration retentate (DUR) diluted to the protein concentration of skim milk, as opposed to skim milk, reduced the required membrane area by 36% for a 5-stage process. When DUR was the MF feed, feed protein concentration, which depended on the number of MF stages, was optimized. The DUR protein concentration that minimized the required MF membrane area was 2.47, 3.85, 4.77, and 5.41% for a 2-, 3-, 4-, or 5-stage MF process, respectively. For a 5-stage process, increasing the protein concentration of the feed from 3.2 to 5.4% decreased the required MF membrane area by 10%. It was also found that as the number of stages increased from 2 to 5, the required MF membrane area decreased by 39%, when the MF feed was DUR at the optimal feed protein concentration. Finally, increasing the flux from 50 to 60 kg/m2 per hour decreased the required MF membrane area by 17% when the MF feed was DUR at the optimal MF feed protein concentration. Overall, using DUR as a feed for MF could reduce the amount of MF membrane area required to make a 95% serum protein-reduced micellar casein concentrate.