3.1. Occurrence and distribution of contaminantsThe dry-grind ethanol process involves milling, liquefaction,saccharification, fermentation, and distillation (Kelsall and Lyons,1999; Kwiatkowski et al., 2006). Inspected corn is hammer milledand an aqueous slurry is prepared for jet-cooking and treatmentwith a-amylase (liquefaction). The product is corn mash, whichis cooled and further treated with glucoamylase to convertoligosaccharides to glucose (saccharification). Saccharified cornmash proceeds to a combined liquefaction stream where it maybe mixed with backset (thin stillage from distillation) before transferto the fermentation tank. Yeast is propagated in a separate tankbefore being added to the fermentor. In the current study, eightseparate points were sampled over a period of two years (Fig. 1).It should be noted that the production facility was considered‘‘healthy’’ during the entire sampling period, and that no stuckfermentations occurred.Samples from the slurry tank (1), mash tank (2), and before (3)and after (4) the beer/mash cooler, consistently showed less than100 viable contaminants/mL (the limit of detection in thismethod). The combined liquefaction sample (5) was consistentlycontaminated with >106 viable bacteria/mL (Fig. 2A). Since samplesprior to the combined liquefaction stream showed low levels ofcontamination, it is possible that contaminants originated fromthe backset going into the combined liquefaction sample. It is alsopossible that plumbing associated with the combined liquefactionstream is chronically contaminated, possibly with bacterial bio-films. The yeast propagation tank was also consistently contaminatedover time, with >104 viable bacteria/mL (Fig. 2A). Potentialsources of contamination at this point include the yeast inoculaand plumbing. Contents of the combined liquefaction stream andyeast propagation tank are transferred to the fermentation tank.Not surprisingly then, samples taken early in the fermentation(<8 h) also were consistently contaminated, over a range of about105–109 viable bacteria/mL. Interestingly however, samples takenlate in the fermentation (>50 h) were far more variable in termsof contamination, ranging from >108 to <102 viable bacteria/mL.Low levels of contamination in late fermentation samples mightbe accounted for by competition with the yeast for nutrients, orby inhibition by accumulated ethanol. Although the productionfacility was closed for cleaning five times during the sampling period(Fig. 2A), these events were not consistently correlated withchanges in the levels of contaminants. These results suggestchronic, persistent contamination that is resistant to cleaning,possibly related to recalcitrant bacterial biofilms.
การแปล กรุณารอสักครู่..
