Degree Type


Date of Award


Degree Name

Master of Science


Agricultural and Biosystems Engineering

First Advisor

Robert T. Burns

Second Advisor

Dave R. Raman


The main objective of this thesis was to develop a system for predicting methane production and anaerobic digestion performance of multiple substrates prior to implementation of full-scale application. This thesis is prepared in the journal paper format and includes three papers that were prepared for submission to a journal or conference proceedings.

The objective of the first paper was to analyze multiple substrates using various laboratory techniques so that optimum mixture ratios could be formed. Biochemical methane potentials (BMPs) and anaerobic toxicity assays (ATAs) were used to select and in some cases rule out substrates based on their contribution to methane production. Mixtures were created using constraints arising from the full-scale system. This included the use of all available manure, keeping total solids below 15% to facilitate pumping, maintaining pH between 6.5 and 8.2 for microbial ecology, providing high COD concentrations to maximize methane production, and limiting ammonia levels to avoid toxicity (Speece, 1996). The BMP and ATA results from each mixture were analyzed and compared. The mixture with the best performance was selected for subsequent testing in 100-L sub pilot-scale anaerobic digesters.

The objective of the second paper was to analyze the performance of three 100-L sub pilot-scale anaerobic digesters. These plug flow digesters operated at a 21-d hydraulic retention time (HRT) and were fed the mixture selected in the first paper in a semi-continuous manner twice weekly (6 loadings per HRT). Methane production was measured using submerged tipping buckets. Methane production from the sub-pilot scale reactors was compared to that predicted by the BMP tests. After two hydraulic retention times, the BMP maximum and minimum were observed to be valid boundaries for the sub-pilot scale anaerobic digester methane production, with some of the variability ascribed to seasonal substrate changes.

The objective of third and final paper was to use a series of BMPs and an ATA to predict the methane production in three 100-L sub pilot-scale anaerobic digesters that were subjected to a potential toxicant, glycerin. A group of ATAs were performed with glycerin inclusion rates of 0.5%, 1.0%, 2.0%, 4.0%, 8.0%, 15%, 25%, and 35% by volume. A set of BMPs was performed where a baseline mixture was combined with glycerin such that glycerin was 0.0%, 0.5%, 1.0%, 2.0%, 4.0%, 8.0%, 15%, 25%, and 35% of the combined mixture by volume. In addition, BMPs of 100% glycerin and 50% glycerin/50% DI water by volume were also performed. The three 100-L sub pilot-scale anaerobic digesters were operated at a 21-d hydraulic retention time (HRT) and were each fed in a semi-continuous manner twice weekly (6 loadings per HRT). Each digester was fed a combination of the mixture selected in paper one with a different amount of glycerin (1%, 2%, 4% by volume). The ATAs showed that glycerin was toxic to methane production at all inclusion levels. The BMPs indicated no significant difference between methane production of the 0.0%, 0.5%, 1.0%, 2.0%, and 4.0% mixture combinations; however, at 8.0%, methane production tripled. In contrast, the sub pilot-scale reactors showed signs of toxicity 4.0% glycerin inclusion and little to no effect on methane production for 1.0% and 2.0% glycerin inclusion. Thus, neither the ATA nor the BMP proved to be an adequate predictor for the sub pilot-scale reactors. The most likely cause was lack of mixing within the sub pilot-scale digester to keep glycerin suspended and the mixture well blended. The separation of materials probably lead to short circuiting and prevented adequate microbial activity and methane formation.


Copyright Owner

Steven Thomas Sell



Date Available


File Format


File Size

108 pages