Abstract
Anaerobic digestion (AD) is extensively adopted in wastewater treatment plants (WWTPs), aids in solid dissolution, sludge volume reduction, and bioenergy recovery through biogas production. However, less than 30 % of organic carbon from AD feedstock is converted to CH4, with about 8 % emitted as CO2. The CO2 content significantly lowers biogas calorific value, often requiring further removal to produce biomethane or renewable natural gas suitable for pipelines or sensitive equipment. Currently, fewer than 10 % of WWTPs in the United States utilize biogas effectively, resulting in energy waste. Enhancing energy recovery from AD would support the transition to sustainable energy systems. One approach to improving biogas utilization is to enrich hydrogenotrophic methanogens to convert CO2 and H2 into CH4. While promising, challenges such as pH shifts due to CO2 removal, H2 partial pressure impacts, and low gas-liquid mass transfer rates limit efficiency. Developing an efficient biological upgrading system to achieve high CO2 translation rate with less external H2 input is in need. To address these challenges, an inverse design modeling framework integrating machine learning with multi-objective optimization was developed to identify optimal operational configurations. Predictive models were constructed using literature derived data to capture the relationships between operating conditions and upgrading performance and were embedded within an optimization algorithm to determine optimal operating regimes. Model predictions were further evaluated through experimental validation under optimized conditions, demonstrating good agreement between predicted and experimental results. Another strategy involves halting the AD process at the acidogenesis stage to accumulate volatile fatty acids (VFAs), converting 50-70 % of organic material into energy. VFAs have great potential values in biomanufacturing. Our Meta-analysis revealed sustained VFA accumulation requires long-term suppression of methanogens, yet effective in situ strategies remain limited. Methanogens’ sensitivity to reactive oxygen species (ROS), particularly H₂O₂, offers a controllable approach for selective inhibition. While there is no further study exploration of direct H2O2 addition to the treatment of actual sewage sludge, which was more difficult to degrade, and evaluation of the long-term performance. Here, I designed an integrated strategy combining on-site H2O2 generation (∼4.2 g L-1) with controlled dosing (up to 80 mg L-1) to suppress methanogens, achieving an average VFAs concentration of 10.6 g COD L-1, while accumulated VFAs were recovered via electrodialysis with a maximum of 26.7 g COD L-1. The effects of light on H2O2 mediated inhibition were further investigated, showing illumination significantly enhanced H2O2 mediated inhibition, allowing complete methane suppression at lower H2O2 doses (from 380 to 80 mg L-1). Microbial community analysis revealed Firmicutes dominance under oxidative stress and functional adaptations to mitigate ROS. The system was further evaluated under tetracycline exposure to simulate real wastewater conditions. Traditional AD exhibited partial methanogenesis inhibition and increased cell damage, whereas the H2O2/Light AD system maintained stable VFAs production and enhanced tetracycline removal via oxidative degradation. Enhanced exopolysaccharide production was observed, likely serving as a microbial defense mechanism to reduce direct contact with tetracycline or oxidative products. This highlights the robustness of the H2O2/Light AD system under antibiotic stress and its potential for treating contaminated sludge. Overall, this dissertation demonstrates integrated strategies for enhanced energy and resource recovery from complex sludge systems. Key innovations include data driven CO2 to CH4 upgrading, long-term methanogen suppression for VFAs accumulation, light assisted H2O2 control, and robust performance under antibiotic stress. Microbial analyses provide insights into selective inhibition, functional adaptation, and protective responses. The dissertation concludes with identifying future research opportunities include optimizing methanogen suppression for maximal VFAs yield, integrating VFAs recovery with nutrient recycling, improving electrodialysis efficiency, and scaling up H2O2 assisted AD systems for practical wastewater treatment.
Committee Chair
Zhen He
Committee Members
Arpita Bose; Fuzhong Zhang; George Wells; Yinjie Tang
Degree
Doctor of Philosophy (PhD)
Author's Department
Energy, Environmental & Chemical Engineering
Document Type
Dissertation
Date of Award
3-23-2026
Language
English (en)
DOI
https://doi.org/10.7936/p8a2-4x25
Recommended Citation
Sun, Jiasi, "Optimization and Mechanistic Understanding for Value-added Products Production through Anaerobic Digestion" (2026). McKelvey School of Engineering Graduate Student Theses & Dissertations. 1362.
The definitive version is available at https://doi.org/10.7936/p8a2-4x25