MAXIMIZING LIGNIN YIELD USING EXPERIMENTAL DESIGN ANALYZING THE IMPACT OF SOLVENT COMPOSITION AND FEEDSTOCK PARTICLE SIZE ON THE ORGANOSOLV PROCESS IN THE PRESENCE OF FEEDSTOCK CONTAMINATION

The depletion of fossil feedstock and the unfavorable environmental effects accompanying by its exploitation are the driving forces in the process of transitioning to renewable feedstock as the primary resource. Similar to petrorefineries, a new modern biorefinery would use biomass to produce a variety of different chemical products and transportation fuels. Lignin, a potential low-cost, high volume output process stream derived from lignocellulosic biomass is currently being researched to better support the economics of the future biorefinery. In this study, experimental design was used to determine the optimal level for each process factor in an organosolv fractionation process that targets maximum attainable lignin yield, even in the presence of feedstock contaminants. The process factors studied were two different fractionation times (56, 90 min), two different fractionation temperatures (140°C, 160°C), three mixed feedstock loadings containing mixtures of switchgrass (Panicum virgatum) and tulip poplar (Liriodendron tulipifera) in three different weight ratios ([10/90], [50/50], [90/10]), three different poplar chip sizes (coarse, medium, fine), three different solvent compositions containing different ratios of the fractionation solvents methyl isobutylketone (MIBK), ethanol (EtOH) and water (H2O) ([07/30/63], [16/34/50], [62/27/11]), and three different acid concentrations (0.025, 0.05, 0.1 M). Based on the results found it is predicted that, even in the presence of switchgrass contaminants an estimated mean lignin yield of ~ 90 wt % is attainable if the levels of the organosolv process are set to a fractionation time of 90 minutes at a fractionation temperature of 160°C, use of a feedstock mixture containing 10% switchgrass and 90% medium poplar particles, and the use of the 16/34/50 solvent mixture with an added acid concentration of 0.1 M. The practical implications of these results on biorefinery operation will also be discussed