Abstract: This study aimed to develop a fermentation process for the production of fruit wine using Chinese wolfberry fruits from Ningxia autonomous region as the starting material. The fermentation conditions temperature, sugar addition, yeast inoculation size and pH were investigated in 4 runs of single factor experiments to preliminarily understand their effects on total flavonoid content in Chinese wolfberry fruit wine. Subsequently, Box-Behnken matrix design followed by regression analysis was used to establish a quadratic regression model for total flavonoid content in Chinese wolfberry fruit wine as a function of the above four fermentation conditions. Finally, response surface analysis was carried out for optimization purpose. The optimum fermentation conditions for improved total flavonoid content in Chinese wolfberry fruit wine were determined as follows: fermentation temperature 21.9 ℃; yeast inoculation size 0.29 g/L; sugar addition 107.0 g/L; and pH 4.3. Under these conditions, the predicted value of total flavonoid content in Chinese wolfberry fruit wine was 0.2404 g/L, and the experimental value was (0.2337 ± 0.0043) g/L. Therefore, the established regression model has good prediction ability.

Key words: Chinese wolberry, fruit wine, fermentation, flavonoids, response surface methodology, optimization