Development of a mechanistic model for hydrogen production through decentralized anaerobic digestion of high-strength organic wastewater

Abstract

This dissertation describes the investigation of the effects of dissolved H2 recovery and removal from a two-stage anaerobic digestion (AD) system. The work was done in the context of the development of a novel decentralized high-strength wastewater treatment system. The Modular Encapsulated Two-stage Anaerobic Biological (METAB) system is being developed as a solution for both industrial wastewater-producing entities (such as food and beverage manufacturing plants) and municipal wastewater treatment systems. This work hypothesizes that the removal of dissolved H2 from the first-stage of the METAB system will have positive effects on system performance, including an increase in overall H2 production and enhancement of the kinetics of the treatment system. This work can be separated into three parts that is reported in four chapters. The first part consists of the development and validation of a first-principles gas recovery model that can predict the removal rate of dissolved gases using a hollow-fiber membrane contactor. The model was able to achieve an average absolute error of 10-16% against lab-scale experimental data, and 10% against published literature. The second part consists of the development and validation of a modified model to simulate the first-stage of the two-stage AD system. Experimental work on the effects of reducing dissolved H2 concentrations on first-stage reactor performance was done with Adithya Ganapathiraju and detailed in Chapter 3. The model development and validation is reported in Chapter 4. It was shown that reducing dissolved H2 concentration resulted in a 10x increase on H2 production when reactors were fed real brewery wastewater, but had little to no effect when reactors were fed synthetic wastewater. The modified model development to simulate the first-stage system focused on the addition of lactate and ethanol as important intermediates due to their substantial presence in the wastewater feed. Sensitivity analysis and parameter fitting was successful with both baseline and modified models. However, comparing the calibrated models against a validation data set indicated that the modified model had better predictive capability on the effects of reducing dissolved H2 concentration on overall H2 production rates. The final part of the dissertation is a sensitivity analysis and feasibility study on the deployment of the METAB system on a medium-sized brewery. The analysis suggests that the upgrade from a single-stage AD system to a two-stage AD system is likely to be beneficial, but the addition of a H2 recovery system is unlikely to yield a substantial benefit. This work begins to elucidate the effects of H2 recovery specifically on the performance of a staged AD system. Although there is much work left to be done to ensure that the models used can predict the performance of staged systems for other feed compositions and temperatures, this work offers a starting point for future researchers to further investigate how to model and simulate the effects of H2 on staged AD performance