Optimization of bacterial cellulose fermentation technology with distiller's grains

Abstract: China is one of the biggest agriculture countries and brewing countries in the world. Distillers' grains are the wastes produced along with brewing, they are full of nutrients but they are thrown away, which brings pollution to the environment. Bacterial cellulose (BC) is a new biomaterial and has special properties such as higher water content, higher mechanical properties. BC has broad applications in food, medicine, environmental industries. But lower productivity with higher price limits the broad application of BC. So this work considered to utilize distillers' grains as fermentation materials of bacterial cellulose to optimize the production of BC from Acetobacter xylinum M12 by using response surface methodology. The ratio of water to distillers' grains was 4:1 (g/g); and filtered by eight-layer gauzes, the vinasse leaching solution was got. The composition of vinasse leaching solution was evaluated by double nitro salicylic acid (DNS) method, bradford method and high performance liquid chromatography (HPLC) method. Eight parameters including glucose, peptone, yeast powder, citric acid, Na2HPO4·12H2O, KH2PO4·3H2O, MgSO4·7H2O and size of inoculum were screened by Plackett-Burman design and 3 factors with significance level of above 95% were optimized by the response surface method using a three-level, three-factor Box-Behnken design. The data were analyzed by the software Design-Expert 8.0, and the R2 for mathematical model for BC fermentation was 0.9734. The regression equation was optimized and evaluated when the optimum conditions and interactive effects were imposed. The microstructure of BC was analyzed by fourier transform-infrared spectrum (FTIR). The results showed that the vinasse leaching solution was full of nutrients, which contained reducing sugar 1 990 mg/L, protein 3 400 mg/L, citric acid 2 060 mg/L and succinic acid 5 638 mg/L. The order of the influence of 3 factors on BC fermentation of distillers' grains from high to low was: glucose, size of inoculum and citric acid, the experimental data were fitted to a second-order polynomial equation using multiple regression analysis. The quadratic regression model was fitted well, as shown by the significance level of 'total regress' (P=0.0001), 'lack of fit' being not significant (P=0.6729) and the high values of determination coefficients before and after adjusting i.e. R2=0.9734 and Radj2=0.9392. The best conditions were as follows: to per 1 L vinasse leaching solution, add 23 g glucose, 25 g peptone, 25 g yeast powder, 4.5 g citric acid, 2 g Na2HPO4·12H2O, 1 g KH2PO4·3H2O and 0.2 g MgSO4·7H2O, with inoculation amount of 8%, fermentation temperature of 30℃ and fermentation time of 7 d. Under these conditions the result of confirmatory experiment (14.44 g/L) was in good agreement with the predicted value (14.42 g/L), and the production increased by 140.6% compared with 6.00 g/L before optimization. The results of analysis con?rmed that the response model was suitable for re?ecting the expected optimization, and the model was accurate and satisfactory. The FTIR analysis of BC verified the product. This study provides the reference for the resourceful use of distillers' grains.