Storage and Pretreatment of Biomass Feedstocks by Ensiling

Authors: SHINNERS, KEVIN - University Of Wisconsin; DIGMAN, MATTHEW - University Of Wisconsin; Muck, Richard; Weimer, Paul

Submitted to: International Silage Conference
Publication Type: Proceedings
Publication Acceptance Date: 7/13/2009
Publication Date: 7/27/2009
Citation: Shinners, K.J., Digman, M.F., Muck, R.E., Weimer, P.J. 2009. Storage and Pretreatment of Biomass Feedstocks by Ensiling. In: Broderick, G.A., Adesogan, A.T., Bocher, L.W., Bolsen, K.K., Contreras-Govea, F.E., Harrison, J.H., Muck, R.E., editors. XVth International Silage Conference Proceedings, July 27-29, 2009, Madison, Wisconsin. p. 191-192.

Interpretive Summary:

Technical Abstract: Biomass costs can be minimized when it is harvested as chopped feedstock using a single- or two-pass system. At harvest, biomass moisture is too high for stable aerobic storage, requiring anaerobic preservation. Although DM content is often well above that typically acceptable for preserving animal feed, anaerobic storage of biomass was successful. Anaerobic storage also offers the opportunity to add value through on-farm, chemical pretreatment that will lower total pretreatment costs and provide better returns. Our objectives were to quantify losses and chemical composition changes of ensiled and pretreated biomass. Three feedstocks were tested--corn stover (residue), reed canarygrass (RCG) and switchgrass (SWG). Corn stover storage was studied in eight trials over five years using over 100 Mg of DM. Stover was harvested with a single-pass combine that simultaneously harvested grain and stover. DM content ranged from 45% to 73%. Stover was ensiled in 3 m diameter silo bags for periods of 7 to 8 months. RCG and SWG were ensiled in three trials each over three years. DM content ranged from 33% to 75%. In all studies, storage characteristics were quantified by DM losses, changes in chemical composition, and fermentation products. RCG and SWG were subject to pretreatment in lab-scale silos with calcium hydroxide, dilute sulfuric acid or ozone at a loading of 50 g/kg DM. The grasses were stored under ambient conditions for 30 days. Pilot-scale silos of RCG and SWG were also treated with calcium hydroxide and sulfuric acid at a loading of 50 g/kg DM and stored outdoors for 60 or 180 days. Following pretreatment and storage, biomass was fermented to ethanol using Saccharomyces cerevisiae D5A in the presence of commercial cellulase and ß-glucosidase. Stover DM losses ranged from 1.4% to 6.0% and averaged 3.9% over the eight trials. Silo density ranged from 96 to 194 kg DM/m^3 and averaged 136 kg DM/m^3. Fermentation products typically were less than 5% of DM and pH averaged 4.6. RCG and SWG DM losses ranged from 0.3% to 4.6% and averaged 2.1%. Fermentation products typically were less than 3% of DM and pH averaged 5.3. Unlike animal feed, ensiled biomass would be removed quickly in massive amounts, minimizing aerobic spoilage and explaining why we achieved such low losses with biomass ensiled at high DM. It is desirable to maximize DM content for economical transport, but some moisture is needed for fermentation to a pH of less than 5 for good preservation. However, too much fermentation reduces carbohydrate availability for downstream conversion. These goals may be optimized by harvesting and ensiling corn stover between 60 and 70% DM. Sulfuric acid and lime pretreatment look promising considering conversion yields and ease of application. Nearly 54% and 24% of cell wall glucose was converted to ethanol in RCG and SWG, respectively, and up to 23% of hemicellulose was hydrolyzed to xylose. Ozone treatment yielded higher results. At pilot-scale, cellulose conversion to ethanol was higher for 180 than 60 d in acid-pretreated substrate, but lower at the longer duration in lime-pretreated substrates. The latter appeared due to butyric acid from clostridial fermentation. Biomass pretreatment applied at 50 g/kg DM would cost an estimated $4.05 and $5.20/Mg DM for calcium hydroxide and sulfuric acid, respectively.