Abstract #T74

# T74
On the use of polymeric microfiltration membranes for the preparation of liquid pre-cheese: Impact on process efficiency.
D. Mercier-Bouchard1, I. Dussault-Chouinard*1, S. Benoit1, A. Doyen1, M. Britten2, Y. Pouliot1, 1STELA Dairy Research Center, Institute of Nutrition and Functional Foods (INAF), Department of Food Science, Université Laval, Québec, QC, Canada, 2Food Research and Development Center (FDRC), Agriculture and Agri-Food Canada, St-Hyacinthe, QC, Canada.

Ultrafiltration (UF) and microfiltration (MF) are widely used for cheesemilk concentration. In a recent study, we observed that filtration performances of the 0.1-µm MF membrane were very close to those of a 10-kDa UF membrane in terms of caseins and serum proteins (SP) rejection. Considering that permeate flux values obtained with a 0.1-µm MF membrane are expected to be higher than those with a 10 kDa UF membrane, it was hypothesized that using an MF membrane would improve the process efficiency. The objectives of this work were to compare 0.1-µm MF and 10 kDa UF membranes in terms of (1) hydraulic and separative performances, (2) energy consumption and fouling behavior and (3) cheesemaking ability of milk retentates. Skim milk concentration (50°C) was carried out in batch mode in triplicate by using 0.1-μm MF and 10-kDa UF membranes, mounted on a Pilot M393 system (Tetra Pak) until a 3.0× concentration factor followed by 2 sequential diafiltration steps with 2 diavolumes. The retentates were standardized with fresh cream to a protein/fat ratio of 0.6 and cheesemaking ability (cheese yield, cheese moisture, fat and protein recovery) were determined. Results showed that the permeate flux values of MF membranes were higher (P < 0.05) than those of UF membranes (0.18 vs 0.09 kg/h·m2·Pa) and the rejection coefficient was slightly lower (0.97 vs 1.00). Energy consumption for the UF system was higher (P < 0.05) than for the MF system (0.024 vs 0.016 kWh/kg of permeate collected). The hydraulic resistance from irreversible fouling was higher for the MF membrane than for the UF membrane (0.11E+13 vs 3.00E+13 m−1). In terms of cheesemaking performances, cheese yield, moisture and fat retention were similar (P > 0.05) but apparent protein losses in whey were lower in cheese made from MF milk due to the removal of SP during concentration. Our results demonstrate that retentates from both processes have similar cheesemaking ability, but using MF leads to better hydraulic performances and uses less energy. The environmental impact of these 2 processes will need to be evaluated through a life-cycle assessment before comparing their efficiency.