<p> The research work presented in this dissertation describes kinetic studies of batch and continuous fermentation of hemicelluloses to lactic acid. Sugar maple wood chips were subjected to hot water to extract hemicelluloses, predominantly as oligomers. Hemicelluloses oligomers were hydrolyzed to fermentable monomeric sugars by dilute acid hydrolysis and concentrated by nanofiltration process. The concentrated wood extract hydrolysate contained 138.7 g/L xylose, 22.2 g/L mannose, 18.7 g/L glucose, 10.7 g/L galactose, 4.6 g/L arabinose, and 5.2 g/L rhamnose. The effect of initial sugar loading was investigated by diluting the concentrated wood extract hydrolysate to obtain desired total sugar concentrations. In the batch fermentation process lower total sugar concentration led to the highest lactic acid yield of 0.83 g/g using <i> Lactobacillus pentosus</i> ATCC 8404 cells. Acetic acid was produced as the byproduct. Adaptation of <i>Lactobacillus pentosus</i> strain to concentrated wood extract hydrolysate led to 10 h reduction in batch fermentation time and 15.5% increase in lactic acid production. </p><p> Adapted <i>Lactobacilus</i> pentosus cells were used to study the kinetics of lactic acid production via batch and continuous fermentation processes. The continuous fermentation process led to higher lactic acid productivity and lower acetic acid to lactic acid ratios ranging between 0.27 and 0.60 as compared to the batch fermentation process which had 0.62 acetic acid to lactic acid ratio. For both batch and continuous fermentation processes all wood hemicellulosic sugars were utilized with glucose being the preferred sugar whereas the rest of sugars were simultaneously utilized. </p><p> A kinetic model for batch lactic acid fermentation from hemicellulosic sugars was developed. Kinetic parameters were determined by ODEXLIMS routine to solve a set of ordinary differential equations for biomass growth rate, product formation rate and substrate utilization rate while minimizing the variance between experimental and predicted values using Microsoft Excel<sup> ®</sup> solver. The model performed satisfactorily for predicting the transient responses of biomass growth, product formation and substrate utilization with squared Pearson correlation coefficient (R<sup>2</sup>) ranging between 0.97 and 0.99 for the initial substrate (total hemicellulosic sugar) concentrations of 40.0 g/L and 55.0 g/L.</p>
| Identifer | oai:union.ndltd.org:PROQUEST/oai:pqdtoai.proquest.com:3614766 |
| Date | 03 May 2014 |
| Creators | Buyondo, John Paul |
| Publisher | State University of New York Col. of Environmental Science & Forestry |
| Source Sets | ProQuest.com |
| Language | English |
| Detected Language | English |
| Type | thesis |
Page generated in 0.0022 seconds