Date of Award


Degree Name

MS in Civil and Environmental Engineering


Civil and Environmental Engineering


Tryg Lundquist



Anaerobic Fermentation of Food Waste and Glycerol to Hydrogen

Eric Krikorian

Hydrogen has several well-known advantages as a fuel and chemical feedstock, but current methods of hydrogen production are costly and energy intensive. A potentially advantageous source of hydrogen is fermentation of organic wastes, especially any abundant, low-cost wastes with a high content of simple sugars. Molar hydrogen yields from fermenters (aka digesters) are affected by pH, organic loading rate (OLR), hydraulic residence time (HRT), and substrate type. A less studied process to increase yield is sparging with low-H2 content gas to strip H2 from the digester liquid. The present study optimized the levels of each of these variables for hydrogen production from glycerol and food waste, building on previous proof-of-concept studies that used glucose as the substrate.

Six bench-scale, semi-continuously fed, stirred, anaerobic digesters were constructed and fed glycerol or food waste as a substrate. In a series of experiments, pH, HRT, OLR, and gas sparging rate were tested over a range of values. pH levels were controlled by use of phosphate buffers. In an envisioned process, low-H2 content from a second-stage methane digester would be used as the sparging gas, allowing subsequent combustion of a high-H2 content biogas with low NOx formation potential. N2 was used as a surrogate for biogas in one set of experiments.

The main conclusions are based on data from periods of steady-state digester performance and daily measurements of pH, alkalinity, biogas production, biogas composition, total and volatile suspended solids, and chemical oxygen demand (COD). COD balances were measured for all experiments and generally showed recoveries of >85%.

With glycerol substrate, the highest molar hydrogen yield (0.071 ± 0.0100 mol H2/mol glycerol) and volumetric hydrogen production (0.281 ± 0.0395 LH2/LReactor-day) were achieved with the following: pH 6.51, OLR 18.8 g COD/L-day, HRT 12 hours, and sparging rate of 3.2 mL/min, and 1-L working volume. Gas type (N2 or biogas) used in sparging did not influence hydrogen production.

The best results with food waste (0.021 ± 0.0013 mol H2/mol COD and 0.478 ± 0.0280 L H2/LReactor-day) were obtained with the following conditions: OLR 33.9 g COD/L-day and nitrogen sparging rate of 1.0 L N2/hour, and 1-L working volume. pH and HRT were not optimized for food waste substrate, but the best values from the glycerol experiments were adopted.

Sparged glycerol and food waste digesters had molar hydrogen yields at least 40% greater than controls. Nonetheless, molar hydrogen yields in the present study were lower than in those reported by other authors, for unknown reasons. Yields from food waste might be improved by optimizing pH and HRT levels. Alkalinity sources need to be identified to replace the non-scalable phosphate buffers of the present research. Lastly, long-term experiments should consider whether attached growth of hydrogen-consuming methanogens develops in hydrogen fermentation reactors.

Available for download on Monday, June 15, 2020