(677b) Optimization of Batch and Continuous Fermenters for Increasing Bioethanol Productivity Using Hybrid Cybernetic Models

A process optimization study has been conducted using cybernetic models to explore ways to increase the productivity of fermenters using recombinant Saccharomyces yeast for converting glucose and xylose to bioethanol. For this purpose, a hybrid cybernetic model has been formulated for the Ho-Purdue strain 1400 (pLNH33) by incorporating elementary mode analysis into the cybernetic modeling framework. The hybrid cybernetic modeling approach models the distribution of uptake fluxes among different elementary modes based on pseudo-steady state hypotheses for the intracellular metabolites. The model shows excellent capability for describing and predicting various metabolic behaviors of the recombinant yeast as validated by several different sets of fermentation data.

The developed model has been used to optimize batch and continuous fermentation systems for maximum bioethanol productivity under the constraint that the unfermented sugar content at the end of fermentation for a batch culture or at the outlet for a continuous culture is below a certain threshold (e.g., 1 g/L). The theoretical maximum limit of ethanol productivity achievable by optimal operation has been estimated as a result. A comparison of two different cultivation methods shows that a batch fermenter outperforms a chemostat with respect to ethanol productivity in this example, i.e., the former shows the ethanol productivity within the range of 1.0 ~ 1.5 g/L/h by the optimization, while the latter 0.2 ~ 0.3 g/L/h. This might be ascribed to the strain characteristic of a strong preference for glucose to xylose as a carbon source. The theoretical implications of the obtained optimal operating strategies are discussed with a link to the internal flux distributions which can be estimated by the hybrid cybernetic model as well.