ABSTRACT Effects of Filtration Temperature and Heat Treatment on Composition and Rheological Properties of Whole Milk Ultrafiltration Retentates John William Montella For the first part of my thesis, the effects of filtration temperature and heat treatment on the compositional properties of whole milk Ultrafiltration retentate (UF) were studied. Ultrafiltration is primarily run at temperatures in the range of 50-55°C but more and more plants are starting to filter at refrigeration temperatures. In the ultrafiltration of milk, filtration temperature can affect the composition of the retentate by affecting the chemistry of milk components. The application of a pasteurization step can also affect the chemistry of milk components. There were two filtration temperatures used: 10°C and 50°C. The effect of stage in the filtration process in which the pasteurization step is applied (before UF vs. after UF) is also studied. The heat treatment used was a batch pasteurization treatment of 63°C for 30 minutes. The milk was concentrated to a Volume Concentration Ratio (VCR) of 3X through a 10,000 Molecular Weight Cut Off polysulfone membrane. Compositional analysis was performed on permeate and retentate. According to my results, there were significant treatment effects on the retention of true protein (both casein and whey protein nitrogen), total protein, non-casein nitrogen, minerals (including Ca) and pH of the retentate. The chemistry of the milk components were considered as possible reasons for these differences. The week of processing did not affect the results. For the second part, the effect of composition of the retentate on their viscosity and flow properties was observed. Rheological properties are very important in process design and for consumer acceptability. Flow and viscosity data was collected using a dynamic stress rheometer. Three analytical temperatures were used during the rheological measurements: 10°C, 40°C, and 70°C. A shear rate of 500 s-1 was used for viscosity analysis. Flow properties were also observed using the same three temperatures. According to the results, all the retentate displayed shear thinning behavior and this behavior became more pronounced as the testing temperature increased. As the shear rate increased, there was a shear thickening effect that became more pronounced as temperature increased. There was a significant effect of treatment on the viscosities of the retentate. Compositional differences in the retentate are possible contributors to observed results. The week of processing had no effect on the results. For the final part, the effect of filtration temperature and heat treatment on rennet coagulation time of retentate was observed. A 22μl aliquot of chymosin was added to 100 ml of retentate heated to 30°C prior to analysis. Rennet coagulation time was monitored using a dynamic stress rheometer. The rennet coagulation time was recorded as the time at which the G’ value reached 1 Pa. There was a significant effect of filtration temperature and heat treatment on the rennet coagulation time of the retentate. Compositional differences are all possible contributors to these differences. From the observations from all three studies, the following conclusions can be made: (1) There were significant differences observed with respect to filtration temperature and heat treatments on chemical composition of the retentate; (2) The retentate displayed a shear thinning behavior and the chemical composition of the retentate could be a contributing factor as well as the sample testing temperature. There was also a significant treatment effect on the viscosity of the retentate; and (3) Significant differences in rennet coagulation times were observed, possibly due to compositional differences of retentate. Processing week did not have a significant effect on my results.

Reverse osmosis (RO) and ultrafiltration (UF) retentates were ultra-high temperature (UHT) processed and compared for storage life at room temperature. Viscosity studies indicated that UHT-treated, RO retentate delayed age gelation longer than UF retentate at the same total solids level (26% TS). When compared at 6.4% protein level (2x RO vs 2.7x UF where x=ratio of the feed volume to concentrate volume), the storage life for both RO and UF retentates was about 6 to 8 months. Sodium hexametaphosphate (SHMP) and disodium phosphate (DSP) at 1, 3, 5, 10, and 20 mM concentrations were incorporated prior to UHT processing of each sample to improve the shelf life. SHMP at 1 and 3 mM concentrations was effective in delaying age gelation, whereas all levels of DSP accelerated gelation. However, SHMP accelerated age gelation at concentrations of 10 and 20 mM. SHMP at 1 mM in RO retentate was more effective in delaying age gelation than the same SHMP level in two UF samples (22 and 26% TS). Analysis showed that RO/UHT-treated samples had higher soluble calcium and ionic calcium than did UF/UHT-treated samples. The coefficient of determination (R2) was .80 between soluble calcium-to-protein ratio and shelf life.