Developing A Bench Scale Test To Evaluate Operational Digestion Stability and Capacity

Introduction
Anaerobic digestion is the backbone of the solids stream process for the Rock Creek Water Resources Recovery Facility (WRRF). Despite its importance, operational status is limited to periodic laboratory samples and a small number of online values. The total feed flow and overall gas generation of the five digesters are tracked online while discrete measurements of volatile acids (VA), alkalinity, pH and volatile solids content are performed routinely per digester to satisfy typical operational requirements. The inability to measure individual feed flow rates and track gas production from each digester can slow the identification of failures and the conditions that cause upset events. Figure 1 shows that even though, two digesters are operated under the same loading conditions, instability can occur due to mechanical failures and our ability to recognize them and respond is limited by discrete measurements that might take several days to be completed. To improve knowledge of the digestion process, Clean Water Services (CWS) is working on the development of a bioassay that can help monitor stability which would benefit operations and the Capital Improvement Program, offering guidance to troubleshoot limitations and helping understand maximum loading conditions for co-digestion. Biochemical Methane Potential (BMP) tests are broadly used to determine the anaerobic degradability of substrates and to obtain a qualitative evaluation of process kinetics (Angelidaki et al., 2009). However, limited conclusions are obtained from BMP tests about the stability of continuously operated systems (Koch et al., 2020). The bioassay CWS is developing is a short version of a BMP test and analogous to an assay developed by Conklin et al., in which the maximum aceticlastic methane production rate (Vmax,ac) is measured and used to estimate the Acetate Capacity Number (ACN). The ACN is the ratio of Vmax,ac to a full-scale digester aceticlastic methane production rate, which can be used as an indicator of process upsets caused by VA accumulation and additional capacity of the system to convert organic matter to methane (Conklin et al., 2008). The goal is to develop a bench scale test that is replicable and easily implementable at the CWS facilities to: - Gain insight on digestion performance and health by identifying conditions, such as mechanical failures, that can cause upset events. - Determine capacity limitations and/or requirements for future construction projects. - Evaluate impacts from the co-digestion program. Bioassay Development Initial Test Trial Initially, a bench test was implemented using available resources (shown in Figure 2-a) to measure the additional aceticlastic methanogenic activity. The feasibility of the test and the potential value of the results for operations were evaluated. Each experiment consisted of feeding acetate to one bottle containing freshly collected digestate from an active digester and comparing its gas production rate to that from another bottle containing only digestate. Despite challenges to determine the initial rate of substrate utilization for aceticlastic methanogens, and limited methane measurements, consistent results were obtained during the first trial. The results suggested that the ratio of biogas production from acetate and digestate could correlate to the Volatile Solids Loading Rate (VSLR) of the full-scale digester but that testing conditions, limited data and the number of experiment replicates could mask differences in digester operational conditions.
Refinement of Bioassay Test A commercial system (AMPTS II Light -BPC Instruments) was later purchased to improve the experimental set up and control, to increase the number of replicates, and to facilitate data gathering (Figure 2-b). Initial validation tests are being performed using fresh digestate from the two digesters located at the Durham WRRF which have online measurement of the individual feed and gas production. Replicability and reproducibility of the experimental conditions are being evaluated using the commercial system. Table 1 includes a description of the factors being tested to establish a consistent method. The indicators that are consistently measured are: - Biogas generation per mass of added acetate - Rate of biogas production in an acetate fed bottle (rac) - Rate of biogas production in a bottle containing only digestate (rb) - Ratio of the two measured biogas production rates (rac/rb) The response to different acetate concentrations was evaluated. The results are observed in Figure 3 which help determine the experiment length. A test that can generate indicators in a short time can be valuable for operations to respond to critical conditions. As observed in Figure 3b, full acetate utilization in comparison to the experiment with digestate can be obtained in less than a day when feeding acetate at <25 mM while the rac is measurable within the first hours of the experiment with consistency independent from the acetate concentration. Consistent gas recovery from carbon addition as acetate has been obtained during the tests that are being conducted weekly. As observed in Figure 4, the gas recovered from mass of acetate seems to not be sensitive to the variability or accumulation of acetic acid & VA in the full-scale digester. The background biogas production rate (rb) is used as a surrogate to the current operation of the digester. This is important because individual gas flow measurements are not available at the Rock Creek WRRF. Comparability of the available bioassay results and gas production of the Durham full-scale digester has been obtained (Figure 5). The ratio between the gas production rates (rac/rb) in these early tests has been compared to the digester VSLR. Figure 6 shows a good correlation between these two parameters despite the intrinsic variability of the TVS measurement. The ratio should represent available aceticlastic capacity which would be limited by VS loading conditions or the relative activity of the biomass that may be influenced by other factors apart from loading, such as temperature, detention time, and the variable loading composition. Additional results from the tests described in Table 1 will be provided in the full paper that include the repeatability and reproducibility of the test as well as relationships to detention time and feed characteristics.
Next steps The testing plans for 2022 consist of: - Bench scale tests will be performed at the Rock Creek WRRF and results will be compared to full-scale parameters and to the results obtained from the validation tests performed at the Durham WRRF. - Testing will be continued at the Durham WRRF and extended to Digester 2 to validate current operational benchmarks. - As an additional valuable result from the test, gas recovery from other substrates for co-digestion will be evaluated. The two digesters at Durham WRRF are currently fed FOG and can be used as a model for new substrate evaluation as well as to determine how the recovery obtained in the batch test compares to that obtained in a continuously fed full-scale digester.