Salix as a biorefinery feedstock : an inquiry into factors affecting conversion performance

Sammanfattning: Lignocellulosic biomass is one of the world’s greatest natural resources, with po-tential for replacing a considerable portion of current fossil resource use. Although lignocellulosic biomass can be readily utilized for generating heat and power via combustion, its conversion into storable fuels such as ethanol and biogas presents a greater challenge. Biomass can be converted into such fuels in biorefineries, although the inherent resistance of lignocellulosic biomass to deconstruction, biomass recalci-trance, incurs financial penalties. The genus Salix contains several woody species known for their high biomass production, low requirements for agricultural inputs, and short harvest rotations. These species could serve as optimal feedstocks for the production of transportation biofuels, if their recalcitrance can be overcome. In this thesis, several aspects related to Salix as a biomass feedstock for use in biorefineries are considered, including breeding, conversion methodology, and cultivation.Biomass recalcitrance and several other traits relevant to the breeding of improved Salix varieties were quantified in a population of natural accessions of S. viminalis, and narrow-sense heritabilities and genetic and phenotypic correlations were calculated. Narrow-sense heritabilities (h2) for biomass recalcitrance were 0.23 to 0.29, indicating a certain level of genetic control. Shoot weights and ratios of syringyl to guaiacyl monomers in lignin were negatively correlated with biomass recalcitrance, serving as possible proxy traits for selection of improved variants in breeding programs.The conversion performance of a subset of clones from the aforementioned pop-ulation were further evaluated using two different conversion systems, enzymatic hydrolysis (EH) and anaerobic digestion (AD). Despite fundamental differences be-tween these systems, measures of conversion performance correlated well, indicating that recalcitrance improvements are not specific to one conversion system. In AD, biomass recalcitrance more strongly affected conversion rates than final yields, and the AD system was overall less sensitive to variations in recalcitrance.Several other aspects related to the use of Salix as an AD feedstock were evaluated. By using a sequential reactor setup, methane yields of non-pretreated Salix were shown to be similar to those reported for steam pretreated material, allowing its use in smaller-scale facilities where pretreatment may not be feasible. Moreover, the choice of clone and rotation length were shown to affect conversion yields under AD, whereas there was no apparent effect of nitrogen fertilization.Making biofuels cost-competitive with their fossil counterparts will likely require a considerable amount of further work in several scientific and engineering disciplines. The contents of this thesis will hopefully contribute toward that goal.