Kwak et al. studied ways to control the production of ethanol in kefir made
from a defined starter culture and found that a two-stage fermentation was best. They
used a nonlactose fermenting yeast—Saccharomyces cerevisia—initially to ferment
glucose added to milk and then to carry out a lactic acid bacterial fermentation using
a mixture of lactobacilli, lactocococci, leuconostoc, and propionibacteria obtained
commercially. During the yeast fermentation stage, the pH remained stable; it only
dropped when the lactic fermentation started. Adding 0.4 or 0.5% glucose to the
starter milk resulted in ethanol production only during the yeast fermentation stage,
yielding a final product with 0.07 or 0.08% ethanol. Storage experiments showed thatkefir produced with the addition of 0.4% glucose was the most stable. When 1.0%
glucose was added, the production of ethanol continued in the lactic fermentation
and resulted in an ethanol concentration of 0.24% in the final product.