Reducing the Toxicity in Hydrothermal Liquefaction aqueous phase using a New Anaerobic Digestion Bioreactor to maximize resource recovery from sludge

INTRODUCTION Water resource recovery facilities (WRRFs) have recovered energy from waste carbon mostly through Anaerobic Digestion (AD). However, only 10% of the WRRFs in the US have implemented energy recovery resulting in very low sludge-derived biogas utilization nationwide (Jones et al., 2019; Shen et al., 2015). Hydrothermal liquefaction (HTL) can recover more energy from sludge (70% of the energy content from sludge) than AD, remove emerging contaminants and, due to its low footprint (retention time 5 to 60 min), it can be applied at a wide variety of WRRF scales, expanding resource recovery in the US (Basar et al., 2021; Seiple et al., 2020). However, HTL generates a toxic aqueous by-product (HTL-AB) that cannot be safely circulated into the headworks of the WRRF (Watson et al., 2020). Here, the largest WRRF in the US (Great Lakes Water Authority (GLWA)) and a national lab (Pacific Northwest National Lab (PNNL)) created a multidisciplinary partnership to develop a research project to assess the use of a novel Recirculating Anaerobic Dynamic Membrane Bioreactor (RAnDMBR) to transform the HTL-AB into a non-toxic low-strength (<1,000 mg COD/L) effluent to maximize resource recovery at WRRFs. MATERIALS AND METHODS A mixture of primary and secondary sludge (7:1; wet weight) with a 75% moisture content was collected at GLWA WRRF (Detroit, USA) and transported with a refrigerated truck to PNNL. At PNNL the sludge was processed through an HTL system. The system processed the sludge at 12 L/h at 350 oC, 2900 psig and at a liquid hourly space velocity of 4 L/L/h. The HTL-AB produced was sent to GLWA for testing. At GLWA the HTL-AB is being processed through a 5-L (working volume) RAnDMBR. The design of the RAnDMBR involves a tree-like structure (Figure 1) surrounded by a stainless-steel mesh of 25 μm pore size. The tree-like structure provides a very high surface area in the system and the bioreactor is constantly recirculating the partially treated wastewater from the HTL through the mesh, promoting the formation of an enhanced biofilm (dynamic membrane). Biofilms have been shown to be useful when degrading recalcitrant compounds present in HTL-AB due to the promotion of direct interspecies electron transfer and increase in the microbial activity (Smith et al., 2015; Usman et al., 2019). The pore size of the mesh (25 μm) is big enough so the RAnDMBR can be operated at low transmembrane pressures while producing a high-quality final permeate. The dynamic membrane also allows the system to operate under high solid retention times while operating at a lower hydraulic retention time (HRT) (<10 days) than conventional AD (15-40 days), minimizing the overall footprint. The effluent from an anaerobic digester co-digesting food waste and sewage sludge was used as inoculum for the RAnDMBR which has been processing diluted HTL-AB for the last 120 days under mesophilic conditions (37oC). The reactor operating conditions can be seen in Table 1. In this paper, the reactor performance is presented through pH, chemical oxygen demand (COD), and volatile fatty acids (VFA) concentrations measurements (APHA et al., 2012). Biogas production was measured daily, and its composition was determined with a gas analyser (SWR 100, MRU). The presence of different soluble compounds was also measured through GC/MS. RESULTS AND CONCLUSIONS To not overwhelm the microbial community during start-up, a dilution factor of 30 was applied on the HTL-AB with DI water. However, the propionate/acetate ratio was constantly increasing and the pH decreasing during the first 25 days of the experiment (Figure 2). This is a sign of potential inhibition of the syntrophic populations, something common when HTL-AB is processed through AD (Watson et al., 2020). To avoid problems due to inhibition of the microbial community, the HRT was increased from 6.4 ± 2.3 days to 11.5 ± 3.5 days at day 25 and the recirculation rate was decreased on day 36 from 3.3 ± 1.0 L/(LR*h) to 0.8 ± 0.3 L/(LR*h). Following this change, the bioreactor performance improved, and the pH increased up to 7.3 ± 0.1 and the propionate/actetate ratio decreased to 0.03. The system has been stable since the change, even after changing the dilution factor of the HTL-AB from 30 to 15 on day 66 (pH=7.3 ± 0.1 and VFA<35 mg COD/L. The COD removal has been constant (68 ± 3%; last 30 days average) (Figure 3). In most of the studies where AD configurations were used for HTL-AB, COD removals were between 14-67% with feedstock concentrations between 0.75-20 g COD/L. Therefore, the presence of an enhanced biofilm in the RAnDMBR is likely promoting high COD removal. The GC-MS data show that most of the components in HTL-AB are being removed (Figure 4). However, nitrogen organic compounds (NOC) like pyrrolidine or pyrazine are still present in the effluent of the RAnDMBR. While we still don't know the concentrations of these compounds, we can assume that NOCs would be removed during the activated sludge process when the effluent of the RAnDMBR is recirculated to the headworks (Van Der Zee and Villaverde, 2005). Moreover, the methane yield during the experiment was 0.26 ± 0.03 LCH4/gCODfed (Figure 3), also quite high when compared to the values obtained by systems using AD for HTL-AB treatment (Watson et al., 2020). CONCLUSIONS AND ONGOING WORK RAnDMBR is a promising technology for HTL-AB treatment and the combination of HTL-RAnDMBR could be a leading-edge application to promote resource recovery at WRRFs. This work is still ongoing with the goal of treating undiluted HTL-AB. During the ongoing work we are also studying the microbial populations present in the biofilm and the effect that the recirculation rate has on COD removal and on the removal of NOCs. REFERENCES APHA, AWWA, WEF, 2012. Standard Methods for the Examination of Water and Wastewater, 22nd ed. Washington DC. Basar, I.A., Liu, H., Carrere, H., Trably, E., Eskicioglu, C., 2021. A Review on Key Design and Operational Parameters to Optimize and Develop Hydrothermal Liquefaction of Biomass for Biorefinery Applications. Green Chem. 1404&1446. Jones, C.A., Coker, C., Kirk, K., Reynolds, L., 2019. Food Waste Co-Digestion at Water Resource Recovery Facilities Business Case Analysis. Seiple, T.E., Skaggs, R.L., Fillmore, L., Coleman, A.M., 2020. Municipal wastewater sludge as a renewable, cost-effective feedstock for transportation biofuels using hydrothermal liquefaction. J. Environ. Manage. 270. Shen, Y., Linville, J.L., Urgun-Demirtas, M., Mintz, M.M., Snyder, S.W., 2015. An overview of biogas production and utilization at full-scale wastewater treatment plants (WWTPs) in the United States: Challenges and opportunities towards energy-neutral WWTPs. Renew. Sustain. Energy Rev. 50, 346&362. Smith, a. L., Skerlos, S.J., Raskin, L., 2015. Anaerobic membrane bioreactor treatment of domestic wastewater at psychrophilic temperatures ranging from 15 °C to 3 °C. Environ. Sci. Water Res. Technol. 1, 56&64. Usman, M., Chen, H., Chen, K., Ren, S., Clark, J.H., Fan, J., Luo, G., Zhang, S., 2019. Characterization and utilization of aqueous products from hydrothermal conversion of biomass for bio-oil and hydro-char production: A review. Green Chem. 21, 1553&1572. Van Der Zee, F.P., Villaverde, S., 2005. Combined anaerobic-aerobic treatment of azo dyes - A short review of bioreactor studies. Water Res. 39, 1425&1440. Watson, J., Wang, T., Si, B., Chen, W.T., Aierzhati, A., Zhang, Y., 2020. Valorization of hydrothermal liquefaction aqueous phase: pathways towards commercial viability. Prog. Energy Combust. Sci. 77, 100819.