Background
The simultaneous and effective conversion of both pentose and hexose in fermentation is a critical and challenging task toward the lignocellulosic economy. This study aims to investigate the feasibility of an innovative co-fermentation process featuring with a cell recycling unit (CF/CR) for mixed sugar utilization. A l-lactic acid-producing strain Enterococcus mundtii QU 25 was applied in the continuous fermentation process, and the mixed sugars were utilized at different productivities after the flowing conditions were changed. A mathematical model was constructed with the experiments to optimize the biological process and clarify the cell metabolism through kinetics analysis. The structured model, kinetic parameters, and achievement of the fermentation strategy shall provide new insights toward whole sugar fermentation via real-time monitoring for process control and optimization.
Results
Significant carbon catabolite repression in co-fermentation using a glucose/xylose mixture was overcome by replacing glucose with cellobiose, and the ratio of consumed pentose to consumed hexose increased significantly from 0.096 to 0.461 by mass. An outstanding product concentration of 65.2 g L−1 and productivity of 13.03 g L−1 h−1 were achieved with 50 g L−1 cellobiose and 30 g L−1 xylose at an optimized dilution rate of 0.2 h−1, and the cell retention time gradually increased. Among the total lactic acid production, xylose contributed to more than 34% of the mixed sugars, which was close to the related contents in agricultural residuals. The model successfully simulated the transition of sugar consumption, cell growth, and lactic acid production among the batch, continuous process, and CF/CR systems.
Conclusion
Cell retention time played a critical role in balancing pentose and hexose consumption, cell decay, and lactic acid production in the CF/CR process. With increasing cell concentration, consumption of mixed sugars increased with the productivity of the final product; hence, the impact of substrate inhibition was reduced. With the validated parameters, the model showed the highest accuracy simulating the CF/CR process, and significantly longer cell retention times compared to hydraulic retention time were tested.
