Cracking The Code of Nitrite Accumulation: Insights into Partial Denitrification Fundamentals

BACKGROUND Recently, every industry has been striving to enhance their carbon management strategies, and the wastewater sector is no different. The pressing need for effective carbon and nutrient management plus compliance with stringent regulatory limits has prompted the need to explore innovative and intensified processes [1]. Anammox is a cost-effective and efficient microbial process that offers a shortcut for nitrogen removal in the nitrogen cycle. When combined with partial nitrification, the PNA process has the potential to achieve complete autotrophic nitrogen removal [2]. PNA has been considered to be an effective and cost saving alternative to conventional biological nitrogen removal, for the treatment of ammonia-rich wastewater [3]. Despite being introduced as an innovative method for nitrogen removal, PNA is not widely used for full-scale mainstream treatment. The primary challenges in applying PNA to mainstream treatment are associated with the unstable selection of nitrite oxidizing bacteria (NOB), mutual inhibition, and competition for the substrates involved in the combined processes [4]. With significant PNA engineering challenges, hindering large-scale mainstream application, the newly developed partial-denitrification (PD) process presents a more promising solution, offering emerging opportunities for a more flexible anammox-based operation [5]. Even with ongoing research, the underlying principles of nitrite accumulation remain incompletely understood and there is still a lack of insight on the mechanisms involved in the nitrite accumulation process. Based on a recent review article by [8], it was determined that nitrite accumulation in denitrification processes can occur through various mechanisms, including ecology shift, carbon type, feast and famine, etc. This study presents a comprehensive approach to assessing nitrite accumulation, taking into account factors such as microbial ecological shifts, alternative carbon sources, feast-famine and carbon internalization, and the effects of COD/N in kinetics. METHODOLOGY The study was carried out using a 10-liter bench-scale Sequential Batch Reactor (SBR) at Toronto Metropolitan University, with operational conditions based on prior batch testing results. The lab-scale SBR aimed to determine the denitrifying kinetics of the biomass, utilizing primary effluent from the Ashbridges Bay Wastewater Treatment Plant in Toronto as the primary carbon source. Batch tests were conducted during the acclimatization period and after full acclimation to assess denitrification rates and kinetics. Additionally, Nitrate/nitrite Utilization Rate (NUR) tests were performed in-situ at low and high F/M ratios. RESULTS The initial carbon used for the partial denitrification assessment was acetate, and the Figure 1-A depicts the peaks of nitrite accumulation over a 57-day period of reactor acclimatization. Batch activity tests were conducted every five to seven days, revealing a noticeable shift in the occurrence of nitrite peaks. During the first 30 days, nitrite peaks were observed 30-60 minutes after test initiation, but after this period, a distinct shift occurred, with most peaks occurring 10-15 minutes after commencement, suggesting potential microbial transformation. Additionally, a correlation was observed in tests conducted after 30 days between the nitrite peak level and the COD/N ratio, although not clearly depicted in the figure. Furthermore, during literature review, studies were found mentioning the significance of specific partial denitrifying microorganisms, which only reduce nitrate to nitrite. The data presented contradicts the theory of the absence of nitrite reductase in some bacterial populations, as nearly all tests resulted in complete nitrite depletion by the end of the test. A similar trend to acetate was observed during the acclimatization with methanol as the second carbon source (Figure 1-B). During the initial methanol-based test, an intriguing observation was made: a nitrite accumulation peak was detected in the system, albeit lower, and it took approximately 60-75 minutes to appear. Given the absence of methylotrophs in the system, efforts are underway to comprehend the reason behind this peak. It is theorized that the available internalized carbon within the system might have been the source of this peak occurrence. One of the key observations made during these tests was the correlation between COD/N and the level of nitrite accumulation, which is considered very important and interesting for practical applications of PD. As depicted in Figure 2, the initial peak within the range of 1.5-2.5 may potentially be attributed to a shortage of electron donors, leading to an imbalance in the system's redox potential. A decline around 4.5-5 was noted, which corresponds well with the stoichiometric demand for complete denitrification. Once again, a peak within the range of 7.5-9 was indicated, suggesting an excessive amount of carbon beyond the half saturation coefficient (ks), creating an imbalance. These observations are considered very important and interesting for practical applications of PD. CONCLUSION This study aims to bridge the existing knowledge gap by elucidating the operational factors that significantly impact the stability and performance of the PD process. The findings will contribute to the development of efficient operational strategies for mainstream PDNA systems.