The development of transportation fuels with low greenhouse gas emissions is imperative due to growing demand for transportation fuels, decreasing conventional petroleum supplies, and global warming. Ethanol from cellulosic biomass could address all of these challenges but current conversion technologies are not commercially feasible. Expensive pretreatments and enzymes are needed to recover fermentable sugars from cellulose and hemicellulose. Lignin represents an additional barrier to sugar recovery. The pretreatment step, key to the conversion process, hydrolyzes xylan, part of hemicellulose, and disrupts lignin. Cellulose, hemicellulose, and lignin are closely associated in the cell wall but the effect of these associations during pretreatment has not been studied despite suggestions that they are important. Better knowledge of hemicellulose and lignin deconstruction during pretreatment could lead to breakthroughs that would bring cellulosic ethanol one step closer to reality.
Key to this study of biomass deconstruction was the use of a fixed bed flowthrough pretreatment reactor. Pretreatment products are swept out of the reactor quickly thus facilitating the tracking of products as a function of time, limiting side and degradation reactions, and product precipitation at the end of pretreatments. The implementation of a metal 96 well plate for pretreatment resulted in the implementation of indirect steam heating, which raised questions about the adequacy of fluidized sand baths for heating. The results of pretreatment and enzymatic hydrolysis using each system were not significantly different although indirect steam did prove to be the superior heating system. Confident in the performance of the fluidized sand bath, baseline-operating conditions for flowthrough pretreatment were selected based on review of the literature, modeling, and trial and error. Populus trichocarpa, a potential bioenergy crop, was subjected to batch and flowthrough pretreatment along with model substrates: holocellulose, the hemicellulose and cellulose portion of P. trichocarpa, birchwood xylan, and cellulolytic enzyme lignin isolated from P. trichocarpa. The differences in the pretreatment results among these substrates indicate that the associations between cellulose, hemicellulose, and lignin limit the hydrolysis of xylan but increase the extractability of lignin. The findings of this thesis provide new research directions and suggest new plant modification and pretreatment strategies.