Influence of Processing Temperature on Membrane Performance and Characteristics of Process Streams Generated during Ultrafiltration of Skim Milk

Abstract

The effects of processing temperature on filtration performance and characteristics of retentates and permeates produced during ultrafiltration (UF) of skim milk at 5, 20, and 50 °C were investigated. The results indicate that despite higher flux at 50 °C, UF under these conditions resulted in greater fouling and rapid flux decline in comparison with 5 and 20 °C. The average casein micelle diameter was higher in retentate produced at 5 and 20 °C. The retentate analysed at 5 °C displayed higher viscosity and shear thinning behaviour as compared to retentate analysed at 20 and 50 °C. Greater permeation of calcium and phosphorus was observed at 5 and 20 °C in comparison with 50 °C, which was attributed to the inverse relationship between temperature and solubility of colloidal calcium phosphate. Permeation of α-lactalbumin was observed at all processing temperatures, with permeation of β-lactoglobulin also evident during UF at 50 °C. All UF retentates were shown to have plasmin activity, while lower activity was measured in retentate produced at 5 °C. The findings revealed that UF processing temperature influences the physicochemical, rheological, and biochemical properties of, and thereby govern the resulting quality and functionality of, retentate- and permeate-based dairy ingredients.

Keywords: milk; permeate; plasmin; retentate; temperature; ultrafiltration; viscosity.

Figure 1

Flux excursion plot showing permeate flux as a function of transmembrane pressure (TMP) during ultrafiltration of skim milk in full recirculation mode with 10 kDa membrane at 5 ± 0.5 (■), 20 ± 0.5 (■), and 50 ± 0.5 °C (■). To study temperature as a single test variable, an initial TMP of 0.45–0.5 bar was selected as initial TMP to operate ultrafiltration (UF) experiments near the sub-critical flux region. Cross flow velocity was kept constant at 0.34 m/s. LMH: L/m²/h.

Figure 2

Permeate flux and transmembrane pressure (TMP) as a function of processing time during ultrafiltration of skim milk with 10 kDa membrane at 5 ± 0.5, 20 ± 0.5, and 50 ± 0.5 °C. Line colour represents flux at 5 °C (▪▪▪), 20 °C (▪▪▪), 50 °C (▪▪▪); TMP at 5 °C (▬), at 20 °C (▬), at 50 °C (▬). Values are raw mean ± standard deviation of data from triplicate runs at each test temperature. LMH: L/m²/h.

Figure 3

Normalised water permeability of membrane measured before (■) and after (■) membrane cleaning process. Values are the raw mean ± standard deviation of analysis from triplicate runs at each test temperature. Different superscripted lower-case alphabet letters within the bar graphs indicate significant differences (p < 0.05). LMH: L/m²h.

Figure 4

Log–log plots of viscosity as a function of shear rate for retentates obtained by ultrafiltration of skim milk at 5 ± 0.5 (■), 20 ± 0.5 (■), or 50 ± 0.5 °C (■). Values are raw mean ± standard deviation of duplicate analysis from triplicate runs at each test temperature.

Figure 5

Electrophoretic patterns under reducing conditions of (a) retentate and (b) permeate samples obtained by ultrafiltration of skim milk at 5 ± 0.5, 20 ± 0.5, and 50 ± 0.5 °C. Bands represent samples from two independent runs at each processing temperature. L: ladder; F: feed; R: retentate; P: permeate; p: low-heat skim milk powder; S: α-LA standard; α-CN: α-casein; β-CN: β-casein; κ-CN: κ-casein; β-LG: β-lactoglobulin; α-LA: α-lactalbumin.

Figure 6

Innate plasmin (PL)- and plasminogen (PG)-derived plasmin activity in feed and UF retentate obtained at 5 ± 0.5, 20 ± 0.5, or 50 ± 0.5 °C. Bar colour represents PL at 5 °C (■), PL at 20 °C (■), PL at 50 °C (■), PG at 5 °C (║), PG at 20 °C (║), PG at 50 °C (║). Values are the raw mean ± standard deviation of at least duplicate analysis from triplicate runs at each test temperature. Different lower-case alphabet letters within feed PL and PG and retentate PL and PG indicate significant differences (p < 0.05).