(226f) Optimizing a Separate Hemicellulosic and Cellulosic Conversion Process Design for Producing Ethanol
Dilute-acid pretreatment is used to convert hemicelluloses present in lignocellulosic feedstocks into fermentable sugars. It also enhances the susceptibility of cellulose for conversion to glucose by enzymatic hydrolysis. Improving the conversion of sugars to ethanol may be possible by separately fermenting the hemicellulosic-sugar-rich stream and the glucose stream derived from enzymatic hydrolysis of cellulose. Fermenting the hemicellulosic sugars separately improves sugar conversion yields because the inhibitory effect of high ethanol concentrations is not present. Also, higher glucose yields from enzymatic cellulose hydrolysis are achieved by reducing inhibition of the enzyme by soluble sugars produced in pretreatment.
However, additional costs are incurred due to the presence of costly solid/liquid separation equipment, the additional capital equipment needed to operate two separate fermentation trains, and the potential additional water needed to dilute the cellulosic solids stream. This process configuration was investigated on pretreated corn stover produced in a 1 ton/d pilot-scale continuous reactor. The pretreated corn stover slurry was separated into a liquid and solid fraction using a custom hydraulic press. The hemicellulosic sugar stream was diluted with water to various concentrations, and then the sugars were fermented to ethanol. The cellulosic stream was re-suspended in either fresh water or spent broth from fermentation of the hemicellulosic sugar stream. The cellulose was converted to ethanol using cellulase enzymes and the yeast Saccharomyces pastorianus in either a simultaneous saccharification and fermentation (SSF) or separate enzymatic hydrolysis and fermentation (SHF) process.
Conversion yields were calculated from initial and final concentrations of sugars and ethanol and used to model process economics as a function of the total solids (TS) loading (based on untreated biomass). For the hemicellulosic sugar stream, we found that neither conditioning nor neutralizing the liquor adversely affected the yields; and we observed a sharp drop after 20% equivalent TS loading. For the cellulosic sugar stream, preliminary data suggests that there is no difference in the yield between the SSF and the SHF process. Further, we saw no significant effect on the yields even at 25% equivalent TS loading. Further results will be presented detailing the impact that the interaction of solids loading, liquor dilution, and recycling the spent broth has on conversion yields and ethanol production.