Xylitol is a high value polyalcohol being used in pharmaceutical, hygiene, and food products due to its functional properties such as anticariogenic, antibacterial as well as low calorie and low glycemic properties. An alternative route for xylitol production is the biotechnological method in which microorganisms or enzymes are involved as catalysts to convert xylose into xylitol under mild conditions of pressure and temperature. This method is unlike the conventional chemical method that requires high pressure and temperature and results in low product yield. The goal of this research is to employ an integrated process using all fractions of an agro-industrial biomass (oat hull) for xylitol bioproduction, preferably in a repeated batch bioconversion process, with C. guilliermondii as the biocatalyst. Processes including hydrolysis, biomass delignification, hydrolysate detoxification using adsorption process, and finally free- and immobilized-cell bioconversions were employed in this study. The kinetics of acid-catalyzed hydrolysis of hemicellulose was investigated under mild conditions (temperature: 110ºC to 130ºC and catalyst (H2SO4) concentrations from 0.1 to 0.55 N) to determine the kinetic mechanism and generation of monosaccharides (xylose, glucose, and arabinose) as well as the microbial inhibitors consisting of acetic acid, furfural, and hydroxymethylfurfural (HMF) in the hydrolysate. A maximum recovery of 80% was attained for xylose as the main monosaccharide and the substrate for xylitol; its generation in the hydrolysate followed a single-phase 2-step kinetic mechanism similar to that of the HMF. However, a single-phase mechanism with no decomposition could describe the formation of arabinose, acetic acid, and furfural. Glucose generation followed a biphasic mechanism (fast and slow releasing) apparently with no decomposition. In the alkaline delignification of the hydrolysis byproduct (solid fraction) and the intact (crude) biomass, kinetic models based on biphasic mechanism consisting of bulk and terminal phases gave the best results and fit to the experimental data. In the bulk phase, where the temperature ranged from 30ºC to 100ºC, the reaction rate constant varied from 0.15 to 0.19 1/min for the intact biomass and from 0.25 to 0.55 1/min for the hydrolysis byproduct. According to the models, accelerated lignin removal with the increased operating temperature could be due to the shift of the process from the terminal phase to the bulk phase. The values obtained for the activation energies herein ( 33 kJ/mol) were less than the values reported in the literature for other lignocellulosic materials. The removal or reduction of the microbial inhibitors in the medium was carried out by activated carbon (adsorptive detoxification). According to the results using the Langmuir model with the activated carbon as the adsorbent, the maximum monolayer capacities of 341, 211, and 46 mg/g were obtained, respectively, for phenol, furfural, and acetic acid. Thermodynamic analyses indicated that the adsorption of the three abovementioned chemicals by the activated carbon was exothermic (enthalpy: H0), spontaneous (free energy: G0), and based on the affinity of the solute toward the adsorbent (entropy: S0). In the concentrated hydrolysate, the removal of phenols, as the main inhibitor, was very successful such that by activated carbon doses of 1.25%, 2.5%, and 5% (w/v) they could be reduced to 34%, 13%, and 3% of the initial concentration (8.7 g/l), respectively. During xylitol bioproduction process in the repeated batch mode using C. guilliermondii, variables of pH control, medium supplementation, and cell recycling proved to be more important than medium detoxification. Processes involving pH-controlled condition combined with nitrogen supplementation and a mild detoxification performed very well with consistent conversion parameters in the successive batches; values of over 0.8 g/g, 0.55 g/l/h, and 53 g/l were obtained respectively for xylitol yield, volumetric productivity, and final concentration. On the other hand, in a single-batch bioconversion, there was no need for supplementing the medium with the nitrogen source. Kinetic modeling of the process showed that substrate (xylose) as well as co-substrate (glucose) consumption, product (xylitol) formation, and cell regeneration could be predicted by a diauxic model. In the aerated free-cell and immobilized-cell systems, aeration rates of 1.25 vvm and 1.25-1.5 vvm were required for free-cell and immobilized-cell systems, respectively, to reach the maximum bioconversion performance. In the immobilized-cell system, cell support also played an important role in this biotransformation. Application of the support based on the delignified hydrolysis byproduct resulted in high and consistent bioconversion parameters in all batches comparable to the ones in the free-cell system. However, bioconversions using the lignin-rich material (hydrolysis byproduct) resulted in a lower efficiency in the first batch which could be partly improved in the second batch and almost fully increased in the third batch to nearly reach performance parameters comparable to the ones obtained in the free-cell system. Overall, the integrated process employed in this investigation helps fill in the knowledge gaps existing on the lignocellulosic biomass application for xylitol bioproduction and biorefinery industries.

abstract: The application of microalgal biofilms in wastewater treatment has great advantages such as abolishing the need for energy intensive aerators and recovering nutrients as energy, thus reducing the energy requirement of wastewater treatment several-fold. A 162 cm2 algal biofilm reactor with good wastewater treatment performance and a regular harvesting procedure was studied at lab scale to gain an understanding of effectual parameters such as hydraulic retention time (HRT; 2.6 and 1.3 hrs), liquid level (LL; 0.5 and 1.0 cm), and solids retention time (SRT; 3 and 1.5 wks). A revised synthetic wastewater “Syntho 3.7” was used as a surrogate of domestic primary effluent for nutrient concentration consistency in the feed lines. In the base case (2.6 hr HRT, 0.5 cm LL, and 3 wk SRT), percent removals of 69 ± 2 for total nitrogen (TN), 54 ± 21 for total phosphorous (TP), and 60 ± 7 for chemical oxygen demand (COD) were achieved and 4.0 ± 1.6 g/m2/d dry biomass was produced. A diffusion limitation was encountered when increasing the liquid level, while the potential to further decrease the HRT remains. Nonlinear growth kinetics was observed in comparing SRT variations, and promoting autotrophic growth seems possible. Future work will look towards producing a mathematical model and further testing the aptness of this system for large-scale implementation. / Dissertation/Thesis / Masters Thesis Chemical Engineering 2016