OPTIMAL OPERATION OF A REPEATED FED-BATCH BIOREACTOR FOR BIOMASS PRODUCTION FROM A METHANOL-UTILIZING BACTERIUM
Abstract
For some types of biological growth kinetics, repeated fed-batch fermentor operation provides the opportunity for increased biomass productivity over more widely considered types of fermentors. However, to insure superior performance, optimization of this process is necessary. In this research, growth of a methanol-utilizing bacterium for the purpose of single-cell protein production is considered with the objective of maximizing time averaged biomass generation. Through an approach based on the calculus of variations, a policy for providing a nutrient solution at a programmed rate to achieve this objective is determined. The impact of several important parameters on the behavior of the system is determined through computer simulations and the results are compared to those of simulations of other modes of bioreactor operation. The model used in this computational work is one in which the specific growth rate and biomass yield vary as functions of methanol concentration only. The results indicate that the minimum liquid volume in the fermentor during a repeated fed-batch cycle has a significant impact on the solution. In addition, biomass productivity for optimal operation is many times higher than that of a continuous stirred-tank bioreactor where the residual methanol in the product stream is very low. Experiments were conducted in a highly instrumented computer-coupled fermentor in which the theoretically determined optimal feeding policy was approximated. The resulting behavior was analyzed by evaluating the behavior of key variables. The effect of minimum volume was investigated and the results compared to the predicted behavior of a continuous stirred-tank bioreactor. This showed that for equal biomass productivity, the product stream of the repeated fed-batch fermentor contained less methanol than that of the comparison bioreactor.
Degree
Ph.D.
Subject Area
Chemical engineering|Energy
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.