Study Of Anammox Process Performance And Microbial Characteristic

The anammox (anaerobic ammonium oxidation) process is a novel and promising alternative to biological treatment of ammonium, in which ammonium is oxidized to nitrogen gas using nitrite as the electron acceptor autotrophically. This process can be considered as an environmental friendly and sustainable process with several advantages, such as no need for addition of external carbon sources, low power consumption, low operational costs and biomass yields. However, this process has strong drawbacks, including slowly bacteria growing rate, limited availability of biomass and being sensitive to oxygen, organic matters and substrate, which limit the application of this process critically. In order to solve these difficulties, performance of anammox process and anammox microbial characteristics were studied in this dissertation. Main contents and conclusions obtained from this research are concluded as follows:Ⅰ. An innovative reactor configuration for anammox enrichment by connecting a non-woven membrane module with an anaerobic reactor was developed in this study. The anammox non-woven membrane reactor (ANMR) exhibited high biomass retention ability through the formation of aggregates in the reactor and biofilm on the interior surface of the non-woven membrane. No fouling problems occurred on the membrane after the development of mature biofilms. At steady state, the average ammonium and nitrite removal efficiencies were 90.9% and 95.0%, respectively. The enrichment of anammox bacteria was quantified by real-time polymerase chain reaction (PCR) analysis as 97.7%.Ⅱ. Different mathematical models were used to study the process kinetics of the nitrogen removal in the ANMR. The kinetics of nitrogen gas production of anammox process was first evaluated in this paper. Both model kinetics study and modeling test showed that the Grau second-order model and the Van der Meer and Heertjes model seemed to be the best models to describe the nitrogen removal and nitrogen gas production in the ANMR, respectively.Ⅲ. In previous granular anammox process, the washout of other species and enrichment of anammox biomass were very slow because of the competition of the coexisting bacteria. In this study, inactive methanogenic granules were proved to be suitable for rapid anammox granulation under high nitrogen concentrations by investigating their interaction with anammox bacteria. The start-up nitrite concentration was significantly higher than the published toxic level for anammox bacteria and other lab-scale studies. The nitrogen loading rate increased from 141 to 480 mg/L/d in 120 days operation with a total nitrogen removal efficiency of 96.0±0.6%. Anammox granules with high purity were observed over the course of three months. The accommodations and proliferations of anammox bacteria in the inactive methanogenic granules might be the main reason for the high anammox purity in a short period.IV. The application of anammox process is partly limited by the availability of anammox biomass. The possibility to introduce the exotic anammox sludge to seed the reactor and fast start-up of an anammox granular reactor for treating high nitrogen concentration wastewater were confirmed in this study. High nitrogen removal was achieved for a long period. During the stable period, the NH4+-N and NO2--N removal efficiencies varied from 95%to 99.2% and from 97.7% to 100%, respectively. The Stover-Kincannon model was first applied in granular anammox process with high correlation coefficient, indicating that the granular anammox reactor in this study possessed high nitrogen removal potential of 27.8 kg/m3/d. The anammox granules in the reactor were characterized via microscope observation and fluorescence in-situ hybridization technique. Moreover, the microbial community of the granules was quantified to be composed of 91.4-92.4% anammox bacteria by real-time polymerase chain reaction.Ⅴ. The performance, inhibition and recovery processes of different types of anammox sludge, including flocculent sludge, granular sludge, and cultured inactive methanogenic granules were evaluated. Both process performance and kinetics study indicated that the reactor seeded with cultured inactive methanogenic granules possessed the highest nitrogen removal potential, followed by the granular anammox reactor and the flocculent anammox reactor. The study suggested that the concentration that as high as 988.3 mg NH4+-N/L and 484.4 mg NO2--N/L could totally inhibit granular anammox bacteria and result in a inhibition of 50% flocculent anammox activity. In addition, reactors seeded with flocculent sludge and anammox granules could be fully recovered by decreasing their influent substrate concentrations. However, the decrease of influent substrate concentration for the reactor with cultured inactive methanogenic granules could only restore about 75% of its bacterial activity. In this study, anammox bacteria purity was the major factor to evaluate the recovery ability in comparison with sludge type. Free ammonia was a more appropriate indicator for the anammox recovery process compared to free nitric acid.VI. The presence of organic matter (OM) would affect the anammox process adversely. In practice, wastewaters containing ammonium are not free from OM. In this paper, the performance of anammox granular sludge in presence of OM was first evaluated under different COD to N ratios. Low OM concentration did not affect ammonium and nitrite removal significantly but improved the total nitrogen removal via denitrifiers. High OM could suppress anammox activity, resulting in a lower ammonium removal. Nitrite removal was not affected by the existence of OM in the presence of denitrifiers. PCR test revealed that there was a reduction in the number of anammox bacteria and denitrifiers quantity increased when 400mg COD/L influent was applied. A COD threshold concentration for anammox inhibition, which was, defined when ammonium removal dropped to 80%, was 308 mg COD/L. Anammox granular sludge had higher tolerance to OM than flocculent sludge.