Efficient removal of spores from skim milk using microfiltration: spore size and surface property considerations

Abstract Text:

Presence of spores in milk can cause numerous quality and shelf life issues for dairy products. Microfiltration (MF) using a 1.4µm pore size can effectively remove vegetative bacterial cells from milk and is used in commercial applications. However, this pore size may be not be equally effective in spore removal. The objective of this study was to determine the effectiveness of MF using 1.4µm and 1.2µm pore size for removing spores of Bacillus licheniformis (BL) and Geobacillus spp. (GEO) from skim milk. Cell size of both spores and vegetative cells was evaluated by scanning electron microscopy (SEM), surface charge by zeta potential analysis and surface hydrophobicity by contact angle measurements, in triplicate. Commercially pasteurized skim milk was inoculated in a sterilized feed tank with a spore suspension, at about 10^6 spores/mL, then treated by MF (in triplicate) using ceramic Isoflux membranes at 6°C, cross-flow velocity of 4.1 m/s, and transmembrane pressure between 69-74 kPa. Total aerobic plate count and spore count of the permeate were conducted. An unpaired t-test was used to determine significant differences between samples at P<0.05 significance level. 

Vegetative cell length ranged between 2.40-3.82µm and the width between 0.39-0.64µm. Spores were shorter and wider, averaging 1.39-1.58µm in length and 0.63-0.88 µm in width, thus having a higher probability to pass through a 1.4µm membrane. Indeed, for BL (1.39µm length x 0.63µm width) an average spore reduction of only 2.17­log was achieved by 1.4µm pore size.  For the 1.2µm membrane, a 4.57log reduction was achieved. For GEO spores, their larger spore size (1.58µm length x 0.81µm width) allowed a practically complete removal using both pore sizes (spore counts in permeate below the detection limit). 

The surface properties of BL and GEO indicated that they may interact differently with the membrane. Both spore species and the ceramic membrane had negative surface charge at the milk pH, indicating slight electrostatic repulsion between them. GEO spores were hydrophilic, while BL spores were slightly hydrophobic; the ceramic membrane surface changes from hydrophilic (in unfouled state) to hydrophobic after adsorption of caseins during MF. Consequently, BL spores may experience slight attractive force to the membrane through hydrophobic interactions, which will facilitate their passage through the membrane. A good understanding of all factors that affect the removal of spores using MF can lead to the production of milk with lower spore count, higher quality, and increased shelf life.

Keywords: microfiltration, skim milk, spore removal