Dynamics of fed-batch simultaneous saccharification and co-fermentation described by 2nd order cellulase adsorption kinetics and a two-population segregated cell model
Tuesday, April 29, 2014
Exhibit/Poster Hall, lower level (Hilton Clearwater Beach)
Ruifei Wang, Rakesh Koppram, Lisbeth Olsson and Carl Johan Franzén, Chemical and Biological Engineering - Industrial Biotechnology, Chalmers University of Technology, Gothenburg, Sweden
Economically feasible production of bioethanol from lignocellulosic raw materials requires robustness towards inhibitors and operation at high solid loadings to give high final ethanol titers. In this work, a kinetic model was developed to analyze an enhanced fed-batch simultaneous saccharification and co-fermentation (SSCF) process that addresses both of these issues. The process included feeds of substrate, enzymes as well as yeast cells harvested from a semi-continuous aerobic cultivation. This “multifeed SSCF” was evaluated for ethanol production from steam pretreated birch slurry at up to 20% of WIS.

Enzymatic cellulose hydrolysis was described by pseudo-second order adsorption kinetics and a cellulase-cellulose complex dependent rate equation. The fermentation at low cell growth was modeled by a segregated model with two distinct active cell populations, each of which was described by black box kinetics. The model was fitted to batch hydrolysis and batch SSCF data, and was validated using data from fed-batch experiments. The multifeed SSCF process was modeled as a sequence of batch processes. The approach resulted in better reproducibility, extended life span of dividing yeast cells and improved xylose utilization, compared to batch SSCF. The simulation results indicated a correlation between yeast tolerance towards birch hydrolysate and the level of birch hydrolysate used in the propagation. The second-order adsorption kinetics was an efficient and reliable method to describe enzyme adsorption on lignocellulosic materials. The segrated cell model offered the flexibility required for modeling fermentation at low cell growth. The validated model can be used for e.g. optimization of feed strategies.