Thermal Hydrolysis of Biosolids: Reducing Microbial Inhibition in Sidestream Treatment

Introduction
As part of its 'Bio-energy Recovery Project,' the Neuse River Resource Recovery Facility (NRRRF) in Raleigh, NC will implement new processes to be more energy-neutral. In the new system, biosolids will go through a thermal hydrolysis process (THP), anaerobic digestion (AD), and dewatering. While THP will increase biogas production and reduce the overall biosolids production, the liquid from dewatering THP-AD sludge (THP-filtrate) becomes a sidestream that must be further treated due to its high levels of ammonia. The facility will implement ANITA™Mox IFAS sidestream treatment with aerobic and anaerobic ammonia oxidizing bacteria (AOB and AMX, respectively) to perform deammonification, which can theoretically remove greater than 90% of ammonia-N and 75-85% of total N.

However, THP releases recalcitrant organics that are inhibitory to AOB and AMX, posing a major challenge in plants that need to remove nitrogen. We hypothesized that pre-treatment of the sidestream will reduce the inhibition of AOB and AMX compared to no pre-treatment. The selected methods included dilution, coagulation with alum sulfate, adsorption to powder activated carbon, and oxidation with ozone. The impact of these treatment methods on the AOB and AMX maximum activity rates in unacclimated biomass (from an ANITA™Mox MBBR) were quantified through batch activity tests. Reducing the inhibition to AOB and AMX is a major challenge in facilities employing THP, and the results of this study will provide needed guidance on addressing this crucial issue.

Methods
Thermal hydrolysis
Dewatered cake was collected from the NRRRF and processed through a CAMBI THP pilot (Figure 1) three times between June 2022 and May 2023. The dewatered cake (~19% total solids (TS)) consisted of a blend of primary sludge and waste activated sludge, which was diluted to 12% TS before adding it to the pressure vessel. The sludge was thermally hydrolyzed at 165°C for 30 minutes.

Anaerobic digestion and dewatering
A 6-L mesophilic anaerobic digester (Figure 2) was operated as a sequencing batch reactor for 1.5 years at an SRT of 15 days. The digester was fed thermally hydrolyzed sludge following the projected VS loading rate at the NRRF. TS, VS, pH, and methane content were routinely measured. The digestate was dewatered with the same cationic polymer used at the NRRRF at a dose of 30 lb/dt (Figure 3). The polymer dose was selected based on the final specifications of the belt filter presses at the plant. After flocculation, the filtrate was collected via gravity filtration with a belt filter sample, then stored at 4°C.

Activity tests
Activity tests (Figure 4) were conducted in triplicate. Synthetic wastewater was used as positive control. Fresh media was collected the morning of each batch test; 25 media pieces that had a uniform biofilm thickness were selected for each test. A negative control was conducted with only THP-filtrate and the same number of media pieces. The impact of THP-filtrate on AOB activity was first tested by exposing the biomass to increasing levels of filtrate (10%, 20%, 30%, and 50% THP-filtrate on a volume basis). Next, two pre-treatment methods, alum coagulation and PAC adsorption, were tested. Alum was added at two doses (2000 and 4000 mg/L); each bottle was agitated at 300 rpm for 90 seconds, then 60 rpm for 10 minutes. The solids were removed via centrifugation. PAC was also tested at two doses (500 and 1000 mg/L); each bottle was agitated at 80 rpm for 24 hours, and the PAC was removed via centrifugation.

The initial ammonia was between 300 and 350 mg NH3-N/L. Samples were taken at 20-minute intervals and the test lasted 80 minutes, and ammonia, nitrate, and nitrite were measured on 0.45 um filtered samples using Hach kits. The maximum AOB activity was taken as the linear portion of the nitrite generation curve, which was observed in the first 40 minutes.

Results and Discussion
As expected, addition of THP filtrate to synthetic wastewater resulted in a decrease in nitrite generation rates of AOB. This is consistent with the production of recalcitrant nutrient products during TH. These include melanoidins produced during Maillard reactions occurring in thermal hydrolysis. Increasing THP filtrate addition to synthetic wastewater resulted in similar decreases in AOB activity in terms of nitrite generation (Figure 5). Compared to the synthetic control, a 10%, 20%, and 30% THP-filtrate addition decreased the activity by 31%, 30%, and 35%, respectively. For alum and PAC treatment, the results were compared to 50% THP filtrate addition with no treatment (chosen to simulate a filtrate with significant inhibition). Treating with 2000 mg/L alum and 500 mg/L PAC increased the activity by 13% and 25%, respectively, with higher alum and PAC doses not resulting in significantly higher activity increases (Figure 6).

Additional data from AOB and anammox tests will be included in the presentation. The sCOD of different size fractions and 3D-EEM will be shown. The results from this study will provide Raleigh Water with specific insights and guidance on how to manage inhibition in the sidestream treatment unit. The results are also of interest to other enhanced nutrient removal facilities planning on incorporating THP to increase anaerobic biogas production. To our knowledge, this is the first time that alum coagulation and PAC addition have been tested as sidestream pretreatment methods to reduce inhibition.