Technical, Microbial, and Economic Study on Thermophilic Solid-state Anaerobic Digestion of Lignocellulosic Biomass DISSERTATION Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University By Long Lin, M.S. Graduate Program in Environmental Science The Ohio State University 2017 Dissertation Committee: Dr. Zhongtang Yu, Advisor Dr. Harold M. Keener, Co-advisor Dr. Frederick C. Michel Jr. Dr. Ajay Shah Copyrighted by Long Lin 2017 Abstract Yard trimmings (leaves, grass, wood chips, etc.) are abundant biomass wastes, which need to be treated and/or utilized properly. Composting is a conventional approach to divert yard trimmings from the landfill. Alternatively, solid-state anaerobic digestion (SS-AD) can convert yard trimmings to biogas (renewable fuel) and digestate. Thermophilic SS-AD can be more favorable for degradation of lignocellulosic biomass, such as yard trimmings, than mesophilic digestion due to a higher hydrolysis of the recalcitrant lignocellulose biomass. It is promising to develop thermophilic SS-AD that can both produce energy and reduce waste to treat yard trimmings. However, several technical challenges need to be addressed prior to scale up of this technology. One of the challenges is the requirement of nitrogen-rich amendment and effective inoculum. Another challenge is the regional and seasonal variability in the composition of yard trimmings, which can affect the thermophilic SS-AD performance. Moreover, digestate of SS-AD needs to be properly handled or effectively utilized. Optimization of the process is needed to make it both technically and economically feasible. Therefore, interests are raised from technical, microbial and economic aspects to examine the process. This research included five inter-related projects: 1) examination of performance of thermophilic SS-AD and composting of yard trimmings with liquid anaerobic effluent ii as inoculum; 2) evaluation of the effect of yard trimmings on thermophilic SS-AD and prediction of methane yield based on feedstock component ratios; 3) development of a sequential batch thermophilic SS-AD with recirculated digestate as inoculum; 4) investigation of microbial community dynamics in the sequential batch thermophilic SS- AD; and 5) techno-economic comparison of thermophilic SS-AD and composting of yard trimmings. The first project compared the performance of thermophilic SS-AD and composting of yard trimmings with effluent from liquid anaerobic digestion at different total solids (TS) contents and feedstock-to-effluent (F/E) ratios. High total solids content negatively affected both performances. The greatest total carbon loss was observed at 35% TS in composting, which was about 50% higher than that in SS-AD; while, using SS-AD, more than half of the degraded carbon was converted to methane with the highest methane yield of 121 L/kg VS . The preferred F/E ratio for SS-AD was 4–6. feedstock Methane production from SS-AD was low at F/E ratio of 2 and 3, likely due to inhibitory effect of high concentrations of ammonia nitrogen (up to 5.6 g/kg N). Nutrient-rich (N, P, K) end products with non-detectable coliform were generated from both SS-AD and composting. The second project evaluated the effect of three components in yard trimmings, i.e. wood chips, lawn grass, and maple leaves, on thermophilic SS-AD performance. Digestion of mixed feedstocks resulted in sooner and more methane production than digestion of single components. The favorable peaking time (14 days) and methane yield iii (143 L/kg VS) were achieved with equal fractions of the three components, increasing the methane yield by 80–200% compared to digestion of single components. Concentrations of volatile fatty acids and ammonia varied with component ratios and correlated with system performance. While organic components in mixtures were degraded differently, they were mostly degraded within 24 days and agreed with methane yield. A mixture design model was established to predict methane yield and results showed that all interactions were synergistic, with the ternary interaction having the strongest effect. The model was verified using experimental data, which showed good agreement. A sequential batch thermophilic SS-AD of yard trimmings was developed with recirculated digestate as the inoculum in the third project. The SS-AD consisted of 4 consecutive runs (30 days/run), with digestate from the previous run being used as the inoculum of the subsequent run. The substrate-to-inoculum (S/I) ratio of 1 (TS basis) was favored over 2 and 3 due to significantly higher methane yield and volumetric productivity. At an S/I ratio of 1, sequential batch SS-AD gradually reached steady state by 3 runs with increases in both methane yields (up to 11.5%) and cellulose degradation (up to 55%), indicating that recirculated digestate could be a feasible inoculum to establish long term stable SS-AD of lignocellulosic biomass. The initial sharp increases of volatile fatty acids during runs 2–4 indicated faster hydrolysis of organic matter than during run 1, suggesting that microbes were probably more acclimated due to digestate recirculation. At steady state, 51% (w/w) of the digestate was recirculated as the inoculum. iv In the fourth project, the bacterial and archaeal communities (day 0, 4, 8, 12, 20, and 30) in the sequential batch SS-AD were examined using Illumina sequencing of 16S rRNA genes to assess the effect of recirculated digestate as inoculum on the microbial community dynamics. Microbes shifted substantially toward a stable state with increased diversity in the first 3 runs, which was consistent with the findings from the third project. The proportions of Firmicutes that contained cellulolytic bacteria increased from 40% to 80% from run 1 to run 3, which might explain the concomitantly increased cellulose degradation and volatile fatty acids (VFAs). The VFA accumulation likely induced dynamic shifts of methanogens. Proportions of archaea rose from 1% to 5% from run 1 to run 4 when methane peaked. Particularly, hydrogenotrophic Methanothermobacter was enriched at volatile fatty acid (VFA) levels of 6–14 g/kg, implying that non-acetoclastic oxidative pathway dominated during the steady-state thermophilic SS-AD. Results suggested that recirculating SS-AD digestate might be an effective way for inoculation. Finally, a techno-economic analysis was conducted to compare the economic feasibility of commercial-scale SS-AD and composting for a 20,000 MT/year capacity of yard trimmings and L-AD effluent. Composting (NPV $2 million) was shown to be more economically competitive than SS-AD (NPV $0.2 million) without financial incentives. SS-AD had higher capital cost but lower labor associated operational cost than composting. The addition of digestate drying improved the economics of SS-AD, but it was also the most energy intensive step relying on heat recovery to reduce cost ($18.8/MT). Operating costs were estimated be to $44/MT capacity for SS-AD and $31/MT for composting. The revenues were comparable at $48/MT for both systems. v With financial incentives, SS-AD was shown to be slightly more profitable than composting. However, RINs and RECs had minor effects on SS-AD economics compared to investment tax credits and grants. Both systems were most sensitives to plant size and tipping fees. The results of this research indicate that thermophilic SS-AD of yard trimmings is technically feasible and its economics could be comparable with composting with some financial incentives. The knowledge obtained from these studies could be used to assist in the optimization and scale-up of SS-AD and enhance the renewable energy from lignocellulosic biomass. vi Dedication This document is dedicated to my family. vii Acknowledgments I want to address my utmost gratitude to my original advisor, Dr. Yebo Li, for his professional guidance and endless encouragement throughout my Ph.D. study. This dissertation would not have been possible without him. I really appreciated him offering me this opportunity four years ago to study and do research in the Bioproducts and Bioenergy Research Laboratory. This opportunity introduced me to the professional training that will benefit me throughout the years. The time working with Dr. Li will continue being one of the most valuable experiences in my life. My sincere thanks go to my dissertation committee members including Dr. Zhongtang Yu, Dr. Harold M. Keener, Dr. Frederick C. Michel Jr., and Dr. Ajay Shah for their advice and collaboration on the work of my dissertation. I especially thank Dr. Zhangtang Yu and Dr. Keener for their mentorship during the last year of my Ph.D. work. I also thank Dr. Lingying Zhao for her service in my candidacy exam committee. Other members of the Department of Food, Agricultural, and Biological Engineering and Environmental Science Graduate Program are important, including Mrs. Mary Wicks, Ms. Candy McBride, Mrs. Peggy Chrisman, Mr. Michael Klingman, Mr. Scott Wolfe, Dr. Gil Bohrer, Ms. Sarah Straley, and Ms. Maura Eze for their academic, administrative, and engineering support. I especially thank Mrs. Mary Wicks for reviewing my publications as well as offering continuous encouragement throughout the study. viii Many thanks go to the talented members of the Bioproducts and Bioenergy Research Laboratory that I have worked with, including Dr. Liangcheng Yang, Dr. Xumeng Ge, Dr. Fuqing Xu, Dr. Stephen Park, Ms. Zhe Liu, Mr. John Sheets, Mr. Adam Khalaf, Mr. Lu Zhang, Ms. Kathryn Lawson, Ms Danping Jiang, Ms. Juliana Vasco, Mr. Josh Borgemenke, Ms. Lo Niee Liew, and several others. I especially want to thank Dr. Liangcheng Yang and Dr. Xumeng Ge for their mentorship and generous help. I am thankful for the funding supports to this research, including the United States Department of Agriculture (USDA) National Institute of Food and Agriculture (NIFA) Biomass Research and Development Initiative Program, the USDA NIFA Hatch Program, the Ohio Agricultural Research and Development Center (OARDC) Environmental Fellowship Program, the OARDC SEEDS program, and The Ohio State University Environmental Science Graduate Teaching Assistantship program. Thanks to the industry collaborators including quasar energy group and KB Bioenergy Inc. for willingness to provide materials for this research. I am extremely grateful to my parents, parents-in-law, siblings/siblings-in-law, and extended family for their endless support. Most of all, I want to express my deepest thanks to my loving husband Fenglong You and our precious daughter Anna Mu-Chen You. You provide me major motivation throughout this journey. I especially thank my mother Mrs. Xuetan Lan for taking care of Anna while I was busy with my study. Thank you for always being there to support me. ix