Dissolved carbon dioxide driven repeated batch fermentation

Abstract

Dissolved carbon dioxide driven repeated batch fermentation has been performed under four glucose concentrations: ~150, ~200, ~250 and ~300 g glucose l-1, with three dissolved carbon dioxide (DCO2) control conditions: without DCO2 control, with DCO2 control at 750 and 1000 mg l-1 levels. No residual glucose was observed under all performed fermentation conditions, and the repeated batch fermentation system could be operated by a computer as self-cycling system. The collected fermentation results presented that, under the same feeding concentration, ethanol concentration in the presence of DCO2 control was significantly lower than that in the absence of DCO2 control; and a higher biomass concentration in the presence of control was observed in this comparison as well. A higher biomass concentration resulted in a shorter fermentation time, which contributed to a higher ethanol production rate. The highest final ethanol concentration was observed as 113.5 g l-1 at 1000 mg DCO2 l-1 control level under ~300 g glucose l-1 condition, where the lowest ethanol production rate of 1.18 g l-1 h-1 was observed. The highest ethanol production rate was 4.57 g l-1 h-1 and its corresponding ethanol concentration was 66.7 g ethanol l-1 at 1000 mg l-1 DCO¬2 control level under ~200 g glucose l-1 condition. For all fermentation conditions, the viabilities of yeast at the end of fermentation were maintained at near 90% where their corresponding final ethanol concentrations were lower than 100 g l-1. As soon as the final ethanol concentration at the end of each cycle was greater than 110 g l-1, its corresponding viability decreased to ~70%. The ethanol conversion efficiency was maintained at ~90% and ~65% in the absence and presence of DCO2 control, respectively. Based on the changing of biomass concentration profiles in the stabilized cycles, two cell growth phases could be identified in the absence of DCO2 control, and only one cell growth phase was noticeable in the presence of DCO2 control cases. Meanwhile, a sudden decline of DCO2 readings at the end of fermentation was constantly observed in both of in the absence and in the presence of DCO2 control cases, which resulted in developing two control algorithms to determine self-cycling time. Comparison of carbon balance analysis between in the absence and in the presence of DCO2 control suggested that the availability of DCO2 control might alter the metabolic flow during fermentation; and the figure of ethanol concentration against fermentation time illustrated that the changing of DCO2 control level did not affect fermentation results, significantly. Moreover, comparisons of ethanol production rate between different processes and different initial glucose concentrations concluded that the ethanol production rate in the presence of DCO2 control was generally higher than that in the absence of DCO2 control under the same glucose concentration; and the ethanol production rate was decreased with the increasing of glucose concentration under the same DCO2 control condition. The experiment results were scaled up to 106 L as a sample analysis in production scale, which suggested that the fermentation with ~200 g glucose l-1 feeding concentration in the absence of DCO2 controlled would provide best profits in the all fermentation conditions.