Process Understanding of Full-scale Micro-aeration to Improve Biogas Quality and Anaerobic Digestion

High hydrogen sulfide (H2S) concentrations are commonly seen in biogas produced from anaerobic digestion process in municipal water resource recovery facilities (WRRFs). These H2S concentrations cause the corrosion of concrete and steel, increases biogas conditioning costs, compromises the functions of cogeneration units, and leads to sulfur dioxide emissions. In addition, elevated dissolved sulfide levels in the sludge can inhibit the overall digestion process. Traditional methods of removing the sulfide have relied on chemical dosing into the digesters or biogas scrubbing, both increasing the costs of biogas utilization, often maintenance intensive and can present relatively high environmental impacts. This may lead to other issues including unnecessary adverse odours and less efficient operation of power generating assets. Biotechnologies (i.e. aerobic or anoxic biotrickling filtration) have recently emerged as cost-effective and more environmentally friendly alternatives (Munoz et al. 2015). Nevertheless, these biotechnologies are still limited by their high investment costs and their generated byproducts such as sulfuric acid and/or elemental sulfur slurry that need further processing or disposal (Kraakman et al. 2019). This paper explores the full-scale implementation of a relatively new process called 'micro-aeration' (MA). MA is a process integrated technology that employs a very small amount of air introduced into the anaerobic digester to raise the redox potential to partly oxidize the sulfides to elemental sulfur and remove it from the digester together with the digested sludge. Traditionally, exposure of the anaerobic digestion process to oxygen has been avoided due to its perceived negative effects on growth and activity of obligate anaerobes, especially methanogens. However, in recent years numerous studies have reported the potential beneficial effects on anaerobic digestion in terms of digestion process stability and digested sludge dewaterability when a small volume of air or oxygen is injected into the sludge (Jenicek et al., 2014). The MA process has been shown to reduce H2S in the biogas and sulfides in the waste stream by more than 90% (Ngheim et al., 2014, Jenicek et al., 2017). For this reason, MA has the potential to revolutionise sulfide removal in anaerobic digestion. However, the basic mechanisms involved in micro-aerobic sulfide oxidation are not fully understood and control strategies for MA are still being optimised. Complex biological and chemical interactions occur between sulfur, oxygen, iron, and phosphorus throughout the anaerobic digestion processes, which are well characterized in literature (Batstone, 2006). Despite recent developments in whole plant process simulators (Hauduc et al., 2017), and modifications to the anaerobic digestion model (ADM) (Flores-Alsina et al., 2016), the implications of these interactions, including within the MA context, have not been fully explored. This paper presents an increased understanding of the MA process by specifically reviewing the performance of a full-scale MA application by Sydney Water and by proposing a process model capable of representing the performance observed at full-scale MA and considers biological and chemical interactions. A review of other full-scale MA applications is also presented. METHODOLOGY This investigation comprised the following steps: Literature review of the existing MA process configurations and performance. Including recently reported full-scale pilot testing (Kraakman et al, 2019). Critical review of over three years of process data for a Sydney Water installation including MA conditions, solids and biogas balances and analysis of the sulfur balance around the digestion treatment process. Development of a process model for an existing installation using the SUMO platform with integrated sulfur chemistry. This involved developing an anaerobic digester process unit with capabilities to set up aeration conditions and calculate oxygen transfer for uptake by digester biomass and use in oxidative processes such as sulfide oxidation by Sulfur Oxidizing Organisms (SOOs). Figure 1 shows the Sumo process schematic of the modelled installation. RESULTS Analysis of the process data on the MA trials showed that normal H2S concentrations in the biogas of approx. 1,000 ppmv can be reduced by up to 92%. However, there is a clear trend when studying air flow rates and H2S removals where a normal removal of around 50% was achieved at air flows of 177 NL/m3 reactor volume-day or higher, with a maximum observed oxygen consumption of 6.65%/m. Moreover, increasing the oxygen to sulfide ratio did not necessarily increase H2S removal. Results for biogas H2S removal and impacts on methane content due to the dilution mostly by the nitrogen in the MA air injected are summarized in Table 1. These sulfide removals translate in an increased biogas quality for energy recovery in WRRFs with potential to lower costs of biogas conditioning. In terms of digestion performance, no statistical differences were observed in total solids reduction, biogas production and cake solids achieved when the MA was performed compared. General process model calibration was performed following the International Water Association's Good Modeling Practices Unified Protocol and the SUMO Process Simulator (Dynamita, Lyon, France). Calibration of the sulfur chemistry in the raw sludge and the digesters without MA was first achieved by estimating the particulate fraction in the influent sulfur speciation, and by considering the collection system operation in terms of iron addition, discharge of biological sludge from a neighboring treatment plant, and internal sulfur recirculation from a chemical scrubber targeting H2S. To calibrate the process model, model parameters needed to be adjusted to match the observations from the MA trials. Most critical seems the affinity constants and its related rate of sulfide oxidation in the model compared to default microbial and chemical kinetic parameters when applied to full scale digestor systems. Two sulfide oxidation pathways are identified for oxidation of H2S, one mediated by SOOs and another chemically mediated in the presence of a metal. From literature in the subject, it is still unclear which pathway would dominate, and simulations using the developed model in this project showed that both mechanisms have the potential to yield the oxidation rates needed to match results obtained. Given the research conducted by Van der Zee et. al. (2007), Jensen et. al (2009), and Ruan et. al. (2017), the MA model currently considers the biological oxidation as the most significant mechanisms for sulfide reduction during air injection. A better understanding of the residual iron-ion levels in the digester feed sludge and actual mass of the different sulfur species will be useful to refine the model. CONCLUSION This investigation presented a process model for a better understanding of the sulfur chemistry during the digestion process, with biological oxidation likely to be the most significant mechanisms for sulfide removal during the MA process. MA is a promising technology for sulfide control in anaerobic digesters; this investigation showed that up to 92% H2S removal can be achieved, although the performance to date has been variable for various reasons. Lower H2S level in the biogas decreases biogas treatment cost for energy recovery thus contributing to the economical and sustainable performance of the energy recovery process of waste sludge digestion at wastewater treatment facilities.