Pilot-scale Study on a Multi-stage A/O-MBBR System for Nitrogen Removal at Medium-Low Temperatures
In recent years, China has achieved significant results in water environment management, but still faces issues such as water resource shortages, water environmental pollution, and water ecological environment damage. From the perspective of protecting water resources, preventing water pollution, and restoring water ecology, continuously promoting the improvement of wastewater treatment efficiency and effectiveness is of great significance for increasing water resource utilization rates, improving water environment quality, enhancing national quality of life, accelerating ecological environment construction, and winning the battle for clean water. Currently, based on the existing national "Pollutant Discharge Standard for Urban Wastewater Treatment Plants" (GB18918-2002), local governments have successively proposed new requirements for the effluent quality of urban wastewater treatment plants, with particularly stricter demands on indicators such as organic matter, ammonia nitrogen, and total nitrogen. Traditional water treatment technologies represented by the activated sludge process face bottlenecks like limited biological nitrification at low temperatures. Numerous studies have shown that the nitrification performance of the activated sludge process significantly decreases under low-temperature conditions, accompanied by issues like severe sludge bulking and biological scum. Therefore, breaking through the low-temperature bottleneck and achieving stable and efficient biological nitrogen removal has become an urgent problem to be solved in the field of wastewater treatment. The Moving Bed Biofilm Reactor (MBBR) technology has been applied in hundreds of wastewater treatment plants worldwide. Due to the attached growth state of the biofilm within the reactor and its continuous renewal capability, it not only possesses high biomass but also maintains high activity. Application results in Nordic countries also indicate that it has stronger adaptability to low temperatures compared to the activated sludge process.
For this reason, this study, targeting the characteristics of urban wastewater in China, utilizes the advantages of MBBR and the multi-stage Anoxic/Oxic (A/O) process for biological nitrogen removal to construct a three-stage A/O-MBBR pilot-scale system. The system's removal capacity for organic matter, ammonia nitrogen, and total inorganic nitrogen under medium-low temperature conditions was investigated. The nitrification capacity and morphological changes of the biofilm under static experimental conditions were analyzed, providing technical support for achieving stable and efficient nitrogen removal from urban wastewater under low-temperature conditions and for the construction and regulation of multi-stage A/O-MBBR systems.
1. Materials and Methods
1.1 Pilot-scale System Experimental Setup and Operation Mode
The process flow of the constructed three-stage A/O-MBBR pilot-scale system is shown in Figure 1. The pilot-scale system consists of three stages of anoxic/oxic (A/O), divided into 10 reaction zones in total. The first-stage A/O-MBBR subsystem consists of anoxic reaction zones (A1, A2) and aerobic reaction zones (O3, O4). The second-stage A/O-MBBR subsystem consists of anoxic reaction zones (A5, A6) and aerobic reaction zones (O7, O8). The third-stage A/O-MBBR subsystem consists of an anoxic reaction zone (A9) and an aerobic reaction zone (O10). The effective volume of each aforementioned reaction zone is 1.4 m³ (1m * 1m * 1.4m), with an effective water depth of 1.4 m. Suspended biofilm carriers (media) with a specific surface area of 500 m²/m³ were added to each reaction zone segment, with a carrier filling ratio of 35% for all. Mechanical mixing was used in the anoxic reaction zones to keep the carriers fluidized, while perforated pipe aeration was used in the aerobic reaction zones, controlling the dissolved oxygen concentration at 3-9 mg/L.
The actual inflow rate of the pilot-scale system was (23.6 + 5.4) m³/d, using a two-point influent distribution, with inlet points set at reaction zones A1 and O5, and an influent ratio of 1:1. The pilot-scale system had two sets of nitrified liquid recirculation (from O4 to A1, and from O8 to A5), with a recirculation ratio of 100% to 200% (based on the inflow rate of each stage). To ensure proper post-denitrification, 50-90 mg/L of sodium acetate (calculated as COD) was added as an external carbon source in the A9 reaction zone. The entire experimental study was divided into 2 phases: Phase I - Normal temperature (18-29℃); Phase II - Medium-low temperature (10-16℃).
1.2 Test Water
The pilot test was conducted on-site at an urban wastewater treatment plant in Qingdao City. The test water was taken from the effluent of the primary sedimentation tank of this plant and entered the pilot system after enhanced pretreatment by flotation. The water quality conditions after enhanced flotation pretreatment are shown in Table 1.
1.3 Detection Indicators and Methods
1.3.1 Conventional Indicators
Conventional indicators such as SCOD, NH₄⁺-N, NO₂⁻-N, NO₃⁻-N, SS, MLSS, and MLVSS were measured using standard methods from "Water and Wastewater Monitoring and Analysis Methods". Dissolved oxygen, temperature, pH, and ORP were measured using a portable dissolved oxygen meter (HACH HQ40d). Biofilm thickness was measured using an inverted fluorescence microscope (Olympus, IX71).
1.3.2 Nitrification Static Experiment
During system operation, carriers from the aerobic zones were periodically sampled to measure the nitrification capacity of the biofilm under static reaction conditions. Carriers from each aerobic reaction zone were placed into a 5L reactor, with a filling ratio identical to the pilot system at 35%. The test water was artificially configured NH₄Cl solution with a mass concentration of 20-25 mg/L (calculated as N). During the experiment, a small air pump was used for aeration to keep the carriers fluidized while controlling the dissolved oxygen at 7-11 mg/L. The test duration was 2 hours, with sampling intervals of 30 minutes, measuring the change in NH₄⁺-N concentration to calculate the nitrification capacity of the biofilm under static reaction conditions.
2. Results and Analysis
2.1 Operational Performance of the Three-stage A/O-MBBR Pilot System
The operational performance of the three-stage A/O-MBBR pilot system is shown in Figure 2. In the normal temperature phase (Phase I), with a reaction temperature of 18-29℃, treatment flow rate of (23.6+5.4) m³/d, and carbon source dosage of 50 mg/L (calculated as COD, same below) in the anoxic zone of the third-stage A/O-MBBR subsystem, the system's influent SCOD, NH₄⁺-N, and TIN concentrations were (160±31), (35.0±7.2), and (35.8±7.0) mg/L, respectively, and the treated effluent concentrations were (27±8), (0.6±0.5), and (2.7±2.2) mg/L, respectively, with average removal rates reaching 83.1%, 98.3%, and 92.5%. In the medium-low temperature phase (Phase II), with a reaction temperature of 10-16℃, the same treatment flow rate of (23.6+5.4) m³/d, and carbon source dosage of 50-90 mg/L in the anoxic zone of the third-stage A/O-MBBR subsystem, the system's influent SCOD, NH₄⁺-N, and TIN concentrations were (147±30), (38.3±2.1), and (39.6±2.3) mg/L, respectively, and the effluent concentrations were (26±6), (0.4±0.6), and (6.8±3.6) mg/L, respectively, with average removal rates reaching 82.3%, 99.0%, and 82.8%. Furthermore, during days 56-62 of system operation, when the carbon source dosage was 50 mg/L, significant NO₂⁻-N accumulation appeared in the A9 reaction zone. However, after gradually increasing the carbon source dosage to 90 mg/L, the NO₂⁻-N accumulation in the A9 reaction zone gradually disappeared, and the effluent TIN concentration decreased to a reasonable level.
2.2 Changes in Biofilm Nitrification Capacity in Each Aerobic Reaction Zone under Different Reaction Temperatures
To evaluate the changes in the nitrification capacity of the three-stage A/O-MBBR system from an overall perspective, the NH₄⁺-N nitrification contribution rate and the nitrification capacity of the biofilm in each aerobic reaction zone under different reaction temperatures were analyzed, with the results shown in Figures 3 and 4, respectively.
Figure 4 Nitrification removal load and fitting curves in the aerobic zones of the 1st and 2nd stage A/O-MBBR subsystems under different reaction temperatures
Figure 3, it can be seen that within the three-stage A/O-MBBR system, due to the two-point influent, the O3 and O4 reaction zones of the first-stage A/O-MBBR subsystem and the O7 and O8 reaction zones of the second-stage A/O-MBBR subsystem bore the main nitrification load of the system. Under both normal and medium-low temperature conditions, the NH₄⁺-N nitrification contribution rates of these two subsystems were 43.1%, 49.6% and 33.8%, 54.0%, respectively. This shows that under medium-low temperature conditions, the NH₄⁺-N nitrification contribution rate of the second-stage subsystem was 20.2% higher than that of the first-stage subsystem.
Figures 4(a) and (c), it can be seen that for the biofilms in the O3 and O7 aerobic reaction zones under normal temperature, they are the main reaction zones in the three-stage A/O-MBBR system for organic matter degradation combined with nitrification function. When the SCOD removal load per carrier surface area (abbreviated as "SCOD removal load", calculated as COD) was less than 2.0 g/(m²·d) and the nitrification load per carrier surface area (abbreviated as "nitrification load", calculated as N) was less than 1.6 g/(m²·d), the relationship between the nitrification removal load per carrier surface area (abbreviated as "nitrification removal load", calculated as N) and the nitrification load followed a first-order linear reaction, with slopes of 0.83 and 0.84, respectively. When the nitrification load increased to 1.6-6.0 g/(m²·d), the relationship between nitrification removal load and nitrification load followed a zero-order reaction, with corresponding average nitrification removal loads of 1.31 and 1.34 g/(m²·d), respectively. When the SCOD removal load was 2.0-4.0 g/(m²·d) and the nitrification load was 1.6-6.0 g/(m²·d), although the zero-order reaction relationship between nitrification removal load and nitrification load remained unchanged, the corresponding average nitrification removal loads decreased to 0.95 and 0.97 g/(m²·d), respectively. For the biofilms in the O3 and O7 aerobic reaction zones under medium-low temperature, when the SCOD removal load was less than 2.0 g/(m²·d) and the nitrification load was less than 1.1 g/(m²·d), the linear slopes of nitrification removal load versus nitrification load decreased to 0.71 and 0.81, respectively. When the nitrification load increased to 1.1-6.0 g/(m²·d), the corresponding average nitrification removal loads decreased to 0.78 and 0.94 g/(m²·d), respectively, representing decreases of 40.4% and 19.4% compared to normal temperature conditions. When the SCOD removal load increased to 2.0-4.0 g/(m²·d), the corresponding average nitrification removal loads decreased to 0.66 and 0.91 g/(m²·d), respectively, representing decreases of 30.5% and 6.2% compared to normal temperature conditions. The nitrification capacity of the biofilm in the O3 reaction zone was consistent with the research results of HEM et al. under corresponding conditions. However, it is noteworthy that under medium-low temperature conditions, compared to the O3 reaction zone biofilm, the O7 reaction zone biofilm exhibited stronger nitrification capacity.
Figures 4(b) and (d), it can be seen that for the biofilms in the O4 and O8 aerobic reaction zones under normal temperature, they are the reaction zones in the three-stage A/O-MBBR system primarily serving a supplementary nitrification function. When the SCOD removal load was less than 1.0 g/(m²·d) and the nitrification load was less than 1.3 g/(m²·d), the relationship between nitrification removal load and nitrification load followed a first-order linear reaction, with slopes of 0.86 and 0.88, respectively. When the nitrification load increased to 1.3-3.0 g/(m²·d), the relationship between nitrification removal load and nitrification load followed a zero-order reaction, with corresponding average nitrification removal loads of 1.11 and 1.13 g/(m²·d), respectively. Under medium-low temperature conditions, when the SCOD removal load was less than 1.0 g/(m²·d) and the nitrification load was less than 1.0 g/(m²·d), the linear slopes of nitrification removal load versus nitrification load decreased to 0.72 and 0.84, respectively. When the nitrification load increased to 1.0-3.0 g/(m²·d), the corresponding average nitrification removal loads were 0.72 and 0.86 g/(m²·d), respectively, representing decreases of 35.1% and 23.9% compared to normal temperature conditions.
From the above analysis, it can be seen that under medium-low temperatures, the inflection points of the relationship between nitrification removal load and nitrification load for the biofilm in each reaction zone occurred earlier compared to normal temperature. This phenomenon is relatively consistent with the research results of SAFWAT. Overall, although the nitrification capacity of the biofilm in each aerobic zone of the system showed a downward trend under medium-low temperatures, the nitrification capacity of the biofilm in the O7 reaction zone of the second-stage A/O-MBBR subsystem increased by 20.5%-37.9% compared to the O3 reaction zone, and the nitrification capacity of the biofilm in the O8 reaction zone increased by about 19.4% compared to the O4 reaction zone. This indicates that the setup of the second-stage reaction zone in the three-stage A/O-MBBR system is beneficial for improving the overall nitrification capacity of the system.
2.3 Changes in Biofilm Denitrification Capacity in Each Anoxic Reaction Zone under Different Reaction Temperatures
To evaluate the changes in the denitrification capacity of the three-stage A/O-MBBR system from an overall perspective, this study analyzed the denitrification capacity of the biofilm in each anoxic reaction zone under different reaction temperatures, with the results shown in Figure 5.
Figure 5 Denitrification removal load in each anoxic zone of the three-stage A/O-MBBR system under different reaction temperatures
Figures 5(a) and (c), it can be seen that for the A1 and A5 anoxic reaction zones, they are the main denitrification zones in the three-stage A/O-MBBR system using raw water carbon sources as substrate. Under both normal and medium-low temperature conditions, when the corresponding anoxic denitrification carbon-to-nitrogen ratio (ΔCBSCOD / CNOx--N) was greater than 5.0 and the denitrification load per carrier surface area (abbreviated as "denitrification load", calculated as NOx--N) was less than 0.95 g/(m²·d), the relationship between the denitrification removal load per carrier surface area (abbreviated as "denitrification removal load", calculated as NOx--N) and the denitrification load followed a first-order linear reaction, with slopes of 0.87, 0.88 and 0.82, 0.84, respectively. When the denitrification load increased above 0.95 g/(m²·d), the relationship between denitrification removal load and denitrification load followed a zero-order reaction, with corresponding average denitrification removal loads of 0.82, 0.82 g/(m²·d) and 0.78, 0.77 g/(m²·d), respectively. As the ΔCBSCOD / CNOx--N decreased, the inflection point of the relationship between denitrification removal load and denitrification load shifted forward, the linear slope under low-load conditions showed a downward trend, and simultaneously, the average denitrification removal load under high-load conditions also showed a downward trend. These results indicate that for the biofilm denitrification in the A1 and A5 reaction zones using raw water carbon sources, the carbon-to-nitrogen ratio is the main factor determining the denitrification function, and under the test water quality conditions, the ideal carbon-to-nitrogen ratio for the A1 and A5 anoxic reaction zones should be greater than 5.
From Figures 5(b) and (d), it can be seen that for the A2 and A6 anoxic reaction zones, because the A1 and A5 anoxic reaction zones removed and consumed the carbon sources in the raw wastewater and most of the nitrate carried by the recirculation flow, the A2 and A6 anoxic reaction zones were long-term substrate-deficient low-load state. Therefore, under both normal and medium-low temperature conditions, when ΔCBSCOD / CNOx--N was between 1.0-2.0 and the denitrification load was less than 0.50 g/(m²·d), the linear slopes of denitrification removal load versus denitrification load were only 0.51, 0.40 and 0.47, 0.37, respectively. Moreover, when the denitrification load increased to 0.50-1.50 g/(m²·d), the corresponding average denitrification removal loads were only 0.25, 0.20 and 0.20, 0.17 g/(m²·d), respectively. However, the static experiment results in this study showed that under conditions of sufficient carbon source and nitrate substrate, the denitrification removal load of the biofilm in the A2 and A6 anoxic reaction zones could reach (0.66±0.14) and (0.68±0.11) g/(m²·d), respectively. This result reflects that the biofilm in the A2 and A6 anoxic reaction zones actually possesses relatively strong denitrification capacity, which is limited by the lack of carbon source and nitrate substrates in this pilot system.
Figure 5(e), it can be seen that for the A9 anoxic reaction zone, it bears the denitrification load for all nitrate flowing out from the first two stages of the three-stage A/O-MBBR system, using externally added sodium acetate as the denitrification carbon source. Under both normal and medium-low temperature conditions, when ΔCBSCOD / CNOx--N was greater than 5 and the denitrification load was less than 2.5 g/(m²·d), the relationship between denitrification removal load and denitrification load followed a first-order linear reaction, with slopes of 0.93 and 0.94, respectively. However, as the ΔCBSCOD / CNOx--N decreased, the linear slope of the relationship between denitrification removal load and denitrification load showed a downward trend. This result also indicates that for the biofilm denitrification in the A9 reaction zone using an external carbon source, the carbon-to-nitrogen ratio is also the main factor determining the denitrification function, with a required denitrification carbon-to-nitrogen ratio greater than 3. Simultaneously, the influence of reaction temperature changes on its denitrification function is relatively small.
2.4 Nitrification Capacity and Morphological Characteristics of Biofilm in Each Aerobic Reaction Zone under Static Experimental Conditions
Figure 6. From Figure 6, it can be seen that under normal temperature, the nitrification capacities of the biofilm in the O3, O4, O7, and O8 aerobic reaction zones were (1.37±0.21), (1.23±0.15), (1.40±0.20), and (1.25±0.13) g/(m²·d), respectively. Under medium-low temperature, the nitrification capacities of the biofilm in the corresponding aerobic reaction zones were (1.07±0.01), (1.00±0.04), (1.08±0.09), and (1.03±0.05) g/(m²·d), respectively, decreasing by 21.9%, 18.7%, 22.9%, and 17.6% compared to normal temperature. These static experiment results are consistent with the trend of measured values in the pilot system. Furthermore, it can be observed that the measured nitrification capacity of the biofilm in each aerobic zone under static experimental conditions was somewhat higher than the actual values in the pilot system. The analysis attributes this to the use of a single ammonium nitrogen substrate and near-saturated high dissolved oxygen conditions during the static experiments, leading to a higher level of biofilm nitrification capacity. Under normal temperature, the actual nitrification capacities in the O3, O4, O7, and O8 reaction zones of the three-stage A/O-MBBR system were 95.6%, 90.6%, 95.7%, and 90.4% of the maximum nitrification capacity under static experiments, respectively. Under medium-low temperature, the actual nitrification capacities in the O3, O4, O7, and O8 reaction zones decreased to 72.9%, 72.0%, 87.0%, and 84.5%, respectively.
Further analysis showed that under normal temperature, the specific ammonia oxidation rates (nitrification rate per unit mass MLVSS, calculated as N) of the biofilm in the O3, O4, O7, and O8 aerobic reaction zones were (0.062±0.0095), (0.059±0.0072), (0.060±0.0086), and (0.060±0.0063) g/(g·d), respectively. Under medium-low temperature, the specific ammonia oxidation rates of the biofilm in the O3 and O4 aerobic reaction zones were only (0.046±0.0004) and (0.041±0.0016) g/(g·d), respectively, decreasing by 25.8% and 30.5% compared to normal temperature. In contrast, the specific ammonia oxidation rates of the biofilm in the O7 and O8 aerobic reaction zones were (0.062±0.0051) and (0.060±0.0029) g/(g·d), respectively. Compared to normal temperature conditions, the ammonia oxidation capacity of the O8 reaction zone biofilm remained unchanged, while the ammonia oxidation capacity of the O7 aerobic reaction zone biofilm even increased by 3.3%. This resultwell demonstrates that under medium-low temperature conditions, the biofilm in the second-stage reaction zone of the pilot system has better nitrification capacity and the rationality of the second-stage subsystem's contribution to the overall system nitrification.
The observation results of the biofilm morphology in each aerobic reaction zone of the first and second stage A/O-MBBR subsystems are shown in Figure 7. Under normal temperature, the biofilm thicknesses in the O3, O4, O7, and O8 aerobic reaction zones were (217.6±54.6), (175.7±38.7), (168.1±38.2), and (152.4±37.8) μm, respectively. Under medium-low temperature, the biofilm thicknesses in the O3 and O4 reaction zones were (289.4±59.9) and (285.3±61.9) μm, respectively, representing increases of 33.0% and 62.4% compared to the biofilm thickness under normal temperature. In contrast, the biofilm thicknesses in the O7 and O8 reaction zones were (173.1±40.2) and (178.3±31.2) μm, respectively, increasing by only 3.0% and 17.0% compared to normal temperature. Some studies have shown that thinner biofilms have stronger ammonia oxidation capacity, which is relatively consistent with the experimental results of this study. The analysis attributes this to the fact that nitrifying bacteria in the biofilm are vertically distributed in the layered structure of the biofilm; excessive biofilm thickness leads to reduced substrate mass transfer efficiency and substrate affinity. Moreover, under medium-low temperature conditions, the dissolved oxygen concentration in each aerobic zone of the pilot system was much lower than that in the static experiment reactor (differing by 3.0-5.0 mg/L). Especially for the thicker biofilms in the O3 and O4 reaction zones, the decrease in oxygen mass transfer capacity within the biofilm led to a decrease in their actual nitrification capacity (only about 70% of the maximum nitrification capacity measured under static conditions). Therefore, for a pure biofilm MBBR, it is necessary to enhance biofilm renewal by strengthening shear intensity and reasonably control biofilm thickness to maintain biofilm nitrification capacity.
3. Conclusion
① Under the conditions of a reaction temperature of 10-16℃ (medium-low temperature), a treatment flow rate of (23.6±5.4) m³/d, and a carbon source dosage of 50-90 mg/L (calculated as COD) in the anoxic zone of the third-stage A/O-MBBR subsystem, the effluent SCOD, NH₄⁺-N, and TIN concentrations of the three-stage A/O-MBBR pilot system were (26±6), (0.4±0.6), and (6.8±3.6) mg/L, respectively, with average removal rates reaching 82.3%, 99.0%, and 82.8%.
② Under medium-low temperature conditions, due to differences in the biofilm of the aerobic reaction zones between the first-stage and second-stage A/O-MBBR subsystems, a difference in the nitrification capacity of the biofilm between the two subsystems was formed. Especially for the first-stage A/O-MBBR subsystem, the nitrification capacity decreased due to increased biofilm thickness. To maintain biofilm nitrification capacity, it is necessary to reasonably control the biofilm thickness.
③ In the three-stage A/O-MBBR pilot system, the effect of reaction temperature changes on the denitrification function was relatively small. Under different reaction temperatures, the denitrification carbon-to-nitrogen ratio using raw water as the carbon source needs to be greater than 5, and the denitrification carbon-to-nitrogen ratio using externally added sodium acetate as the carbon source needs to be greater than 3.