Application of multistage continuous culture to VHG based ethanol fermentations : performance and control of bacteria by PH and pulsed addition of antibiotic
The application of very high gravity (VHG) technology to multistage continuous culture technology proved to be possible. Multistage fermentations were conducted with concentrations of glucose from 152 to 312 g/L in the medium. Steady state concentrations of Saccharomyces cerevisiae, glucose, and ethanol were established in each fermentor at each condition tested. A maximum ethanol concentration of 17% v/v was achieved in the last fermentor in the multistage system when 312 g/L glucose was fermented at a dilution rate of 0.05 h-1 at 28°C. The productivity of ethanol in the multistage system proved to be higher (12.7 g/L/h) than in identical fermentations conducted in batch (2.4 g/L/h) or in a single stage fermentor (1.25 g/L/h). Lactobacillus paracasei was introduced into a multistage fermentation that had been previously brought to steady state with S. cerevisiae. The effects of the contamination were then documented and the fermentation data were compared to the non-contaminated control. Control of L. paracasei in the multistage continuous culture fermentation (MCCF) system was then investigated by both pH and pulsed additions of penicillin G. Both pH and penicillin G addition were effective at reducing (but not eliminating) L. paracasei. The application of penicillin G every six hours (pulsed addition) to give an overall concentration of 2475 U/L in the multistage system was 3.5 times more effective in reducing the viable numbers of L. paracasei than penicillin G maintained continuously at 2475 U/L. Other frequencies of addition and overall average concentrations of penicillin G were tested and documented. Lastly, various unstructured mathematical models were run and assessed for their ability to accurately predict experimental steady state concentrations of biomass (yeast), substrate (glucose), and product (ethanol) concentrations in each fermentor at each condition in the multistage system. The Aiba model (which showed the best fit to experimental data) was then used to predict the course of the fermentation under multistage conditions.