3.1. Research on Velocity Distribution in MPSR
This test detected the flow velocity distribution of the MPSR main reaction zone when the aeration amounts were 0.04, 0.06, 0.10, 0.15, 0.20, and 0.30 m3
/h, and the results are shown in Figure 3
. It could be seen that as the amount of aeration increased, the average velocity of the water flow also increased. Under the six different aeration amounts, a low-speed circulation area was formed in the central area of the reactor. The formation of this zone provided a prerequisite for the formation of different dissolved oxygen zones in the actual operation. The sizes of the central circulation zones formed by different aeration amounts were different. The larger the aeration amount, the smaller the zone was. The flow velocity around the main reaction zone was high, and the flow velocity gradually decreased toward the center, and a relatively stable circulating flow was formed in the main reaction center zone. Due to the viscous force of the side plate to the water flow and the impact and reflection of the water flow, the four corners of the main reaction zone formed unstable vortex zones. In the aeration zone and the bubble rising zone, the bubbles were dense, and strong kinetic energy and momentum exchange were generated with the contact water flow. The turbulence was obvious, and the flow velocity was the largest. At the junction of the main reaction part and the water level-lifting part, the bubbles moved upwards due to buoyancy and the water they wrap around continued to move forward due to inertia. Here, the gas–liquid separation was intense, and the liquid flowed back strongly, forming a strong turbulence zone. In addition, only when the bubbles in the upper area directly opposite the aeration diffuser head accumulated to a certain amount, did it begin to flow horizontally with water, so that the water flow in the main reaction zone showed a pulse motion state.
Through the study of the flow velocity distribution under different aeration rates, the results showed that the MPSR main reaction zone could be divided into five zones according to the different directions and modes of water flow. The regional distribution diagram is shown in Figure 4
. They were: up-flow zone (I), upper horizontal acceleration zone (II), down-flow zone (III), lower horizontal acceleration zone (IV), and central low-speed circulation zone (V). The up-flow zone was located directly above the aeration diffusion device. Under the action of aeration, the bubbles accelerated upward due to buoyancy, and the water current also accelerated upward under the influence of the bubbles. This zone was the main source of power for the water circulation in the reactor, and it was also the zone with the fastest water flow velocity. The top horizontal accelerated flow zone was located near the top cover of the reactor. When water and air bubbles accelerating in the up-flow zone met the top cover, the flow rate dropped rapidly. Some bubbles started to move horizontally with the inertia of the water flow, but some bubbles would gather near the cover. As the volume of the bubbles increased, the water flow began to accelerate horizontally under the continuous squeeze of the bubbles. The down-flow zone was located on the opposite side of the up-flow zone, and the flow direction was also opposite. This zone would wrap some of the bubbles in the upper horizontal acceleration zone, but due to buoyancy, the number of bubbles that were wrapped was small and the volumes of it were small too. The down-flow zone was also in an accelerated motion state, and its power was mainly derived from the gravity acceleration of the water. The lower horizontal acceleration zone was located in the near wall area at the bottom of the reactor. There were almost no air bubbles in this area. The water flow was accelerated horizontally under the suction effect of the aeration area. Due to the limited scope of aeration, when the amount of aeration was small and the water flow energy on the side near the down-flow zone was insufficient, activated sludge precipitation in this area was prone to occur in actual operation. The central low-speed circulation zone was located in the central area of the main reaction zone. There were no air bubbles in this zone. Due to the viscosity of the water, the kinetic energy of the water flow in the four surrounding zones was transferred to the central area, thereby ensuring the continuity and stability of the cycle. Although there were different acceleration forces in the four surrounding zones, it was difficult to continuously accelerate the water flow due to the blockage of the reactor side wall. At the same time, turbulence existed in the surrounding zones due to aeration disturbance, gas-liquid separation, and side wall reflection of the water flow. The existence of these turbulences promoted the material exchange between the surrounding zones and the central circulation zone, which had a certain promotion effect on the actual sewage treatment of MPSR.
The flow velocity distribution test not only revealed the flow velocity distribution of the main reaction zone under each aeration amount, but also obtained the flow pattern distribution of the main reaction zone under each aeration amount. From the experimental results, it could be found that the flow velocity increased in the I, II, III, and IV flow zones distributed around the main reaction zone, and the corresponding acceleration power in each zone helped to maintain and stabilize the circulating flow. Turbulence at the four apex corners of the main reaction zone accelerated the mass exchange. These provided the basic conditions for MPSR to ensure that sewage and microorganisms were in a mixed state and promoted pollutant removal.
For MPSR, aeration has two important functions. One is to ensure the circulating flow of the mixed liquid in the reactor. The other is to provide sufficient dissolved oxygen for biochemical reactions. In order to determine the optimal aeration volume of the reactor under the experimental conditions, the internal dissolved oxygen distribution of the MPSR was studied in a continuous operation mode. Under the test conditions, when the aeration amount was less than 0.1 m3
/h, activated sludge sedimentation existed in the horizontal acceleration zone. Therefore, the test was performed when the aeration amounts were 0.1, 0.2, 0.3, and 0.4 m3
/h, and the test results are shown in Figure 5
. It can be seen that the dissolved oxygen distribution and flow velocity distribution in the main reaction zone were very different. Although the dissolved oxygen also showed a law of lower center and higher edge, this law was not obvious when the aeration amount was relatively low and high. Meanwhile, different aeration levels have a greater impact on the dissolved oxygen distribution in the main reaction zone.
It could be seen from Figure 5
a that when the aeration amount was 0.1 m3
/h, the dissolved oxygen supply was obviously insufficient. The average dissolved oxygen in the entire main reaction zone was below 0.5 mg/L, and there was no obvious aerobic zone. It was not conducive to the removal of organic matter and the occurrence of nitrification. At sampling point 2-1 on the aeration side, the dissolved oxygen value was the largest at 0.58 mg/L. At the sampling point 3-5 near the center of the reactor, the dissolved oxygen was the smallest, which was 0.11 mg/L. Although the edge dissolved oxygen was higher and the center dissolved oxygen was lower, the difference was only 0.47 mg/L. No distinctly different dissolved oxygen partitions were formed. It could be seen from Figure 5
c that when the aeration volume was 0.3 m3
/h, the dissolved oxygen supply was obviously excessive. Although the difference between the maximum dissolved oxygen and the minimum dissolved oxygen was 0.81 mg/L during this aeration, which was higher than that when the aeration was 0.1 m3
/h, the gradient of the dissolved oxygen was still not obvious. The average dissolved oxygen in the entire main reaction zone was 2.85 mg/L, and there was no anaerobic zone and anoxic zone. This would severely interfere with the denitrification process. When the aeration amount was 0.4 m3
/h, it could be found that the dissolved oxygen distribution in the main reaction zone no longer followed the previous rule. From Figure 5
d, in the upper half of the main reaction zone, because the water flow directly contacted the air bubbles, the dissolved oxygen was higher. Due to the greater aeration intensity, the turbulence in the reactor was more intense. The strong material exchange broke the distribution of high oxygen, middle oxygen, and low oxygen in the reactor, and made the dissolved oxygen distribution of the reactor closer to a completely mixed state.
When the aeration amount was 0.2 m3
/h, the characteristics of the DO contour map were similar to those of the velocity profile. As shown in Figure 5
b, the lowest value of dissolved oxygen was 0.24 mg/L at sampling point 3-4, which was close to the center of the reactor. The highest value of dissolved oxygen was 2.02 mg/L at the sampling point 1-8, and the position was on the edge of the reactor. The difference reached 1.8 mg/L. The DO gradient distribution in the main reaction zone was clearly discernible, and the DO concentration decreased from the outer layer to the inner layer. The dissolved oxygen concentration in the outer layer was 1.2 to 1.6 mg/L, the dissolved oxygen concentration in the middle layer was 0.6 to 1.0 mg/L, and the dissolved oxygen concentration in the center layer was 0 to 0.4 mg/L. The aerobic zone, anoxic zone, and anaerobic zone coexisted in the same biochemical aeration tank. In addition, influent was introduced into the central area of the reactor, which could not only provide an electron donor for the denitrifier, but also facilitate the growth of nitrifying bacteria in the outer layer. It was beneficial to the simultaneous removal of nitrogen and phosphorus from the reactor. From the continuous flow MPSR dissolved oxygen distribution, it could be found that the optimal aeration amount was 0.2 m3
/h under the experimental conditions.
It was an important breakthrough to realize the coexistence of anaerobic, anoxic, and aerobic in a single biological aeration tank. Although the presence of simultaneous nitrification and denitrification confirmed the existence of an anaerobic microenvironment in the SBR aeration tank, the actual anaerobic zone was more important for practical operation. As shown in Figure 3
e and Figure 5
b, the sewage could enter from the MPSR center at points 4-5, which were located in the central circulation area and the anaerobic area. Due to the low flow velocity of the central circulation and the anaerobic environment, the organic matters in the influent could be fully used for anaerobic phosphorus release and denitrification. With the circulation of the mixed solution, the influent gradually diffused into the anoxic zone. The anoxic zone was in direct contact with the aerobic zone, and nitrate nitrogen produced by nitrification in the aerobic zone was easily exchanged to the anoxic zone, where the denitrification mainly occurred. Finally, the influent diffused into the aerobic zone. Due to the consumption of organic matters, aerobic phosphorus absorption and nitrification occurred mainly in the aerobic zone. Therefore, MPSR could achieve simultaneous nitrogen and phosphorus removal.
3.2. The Results of MPSR Sewage Treatment
During the operation of MPSR in continuous operation mode and sequential batch mode, the change of the mixed liquor suspended solids (MLSS) concentration with the running time is shown in Figure 6
a. MLSS was tested every two days, in which continuous operation mode ran for 42 days and sequential batch mode ran for 48 days. The change of MLSS concentration had an important impact on the distribution of dissolved oxygen and the removal of pollutants in MPSR. Therefore, maintaining stable MLSS concentration was a prerequisite for MPSR to run well. From Figure 6
a, the MLSS changed little during operation, and its concentration was basically maintained between 4000~4500 mg/L.
b showed the change law of DO at the three sampling points of the main reaction zone 1-1, 4-5, and 7-9 during a single aeration period when MPSR was in sequential batch mode. During the operation of the reactor in the batch mode, the aeration time of a single cycle was 600 min (10 h). Unlike the dissolved oxygen distribution in the continuous operation mode (Figure 5
a), in the sequential batch mode, all zones of the MPSR were in a low-oxygen environment during an aeration time of 0~360 min. With the increase of the reaction time, DO in each zone of MPSR increased. Although the DO distribution in the MPSR at the initial stage of aeration was similar to that of ordinary SBR aeration tanks, the MPSR was characterized by different DO rise times in various zones in the later stage of the aeration. From Figure 6
b, during the aeration from 0 to 360 min, the DO had been slowly rising at the sample point 1-1, and the average DO at this stage was 0.5 mg/L. After 360 min, DO began to rise sharply. At the end of the aeration period, the DO at point 1-1 was stable at about 3.5 mg/L. From Figure 2
a, the sampling point 1-1 was located directly above the aeration diffusion device. This zone was also the place where the bubbles were easy to gather. Therefore, the DO at point 1-1 was higher than that at point 4-5 and 7-9. Its sharp increase in DO was also the earliest. During the 0~360 min aeration, the DO at sampling points 4-5 and 7-9 were basically unchanged and maintained at a low level. During this period, the average value of DO at both points was below 0.1 mg/L. After 360 min of aeration, there was a difference in DO change between the two points. After aeration for 360 min at sampling points 7-9, the DO showed a slight increase. At this time, the DO was maintained at about 0.35 mg/L. The reason for the analysis was that the DO in the up-flow zone and the upper horizontal acceleration zone started to rise, and with the circulation of the water flow, the DO at points 7-9 increased. Points 4-5 were located in the area of the main reaction center. Due to the smaller impact, its DO continued to remain below 0.1 mg/L. As the biochemical reaction continued, after 450 min of aeration, the DO at point 7-9 began to rise sharply, until the end of the aeration period, the DO increased to a maximum of 3.81 mg/L. At point 4-5, DO began to increase gradually after 550 min of aeration. Until the end of the aeration period, the highest DO value was 0.77 mg/L. This change rule of DO in MPSR had prolonged the existence time of the anoxic/anaerobic zone in the reactor to the greatest extent. Compared with an ordinary SBR aeration tank, MPSR will continue to perform denitrification in the later stage of aeration to further improve TN removal efficiency, thereby avoiding the contradiction between the order of denitrification and nitrification to a certain extent.
shows the removal effect of COD, TN, NH4+
-N, and TP in MPSR under two operating modes. In continuous operation mode and sequential batch operation mode, the average removal rates of the four pollutants were 91.12%, 70.95%, 98.65%, 92.77%, and 94.46%, 74.23%, 96.47%, 95.54%. The average effluent concentrations of the four pollutants were 31.90, 10.7, 0.30, 0.31 mg/L, and 21.09, 9.68, 1.23, 0.18 mg/L. The effluent concentration of the four pollutants had reached the grade A standard of Chinese urban sewage discharge standards, which fully met the current requirements of Chinese rural pollution control [17
3.3. Microbial Community Analysis
When MPSR was in continuous flow operation mode, activated sludge samples were collected from five sampling points, 1-5, 4-5, 7-5, 4-3, and 4-7, and their microbial community structure was detected. The test results are shown in Figure 8
. From Figure 8
a, there were obvious differences in the richness and diversity of microorganisms in the activated sludge mixture at different points in the reactor. In each sample lane, there was a stable number of major species, and new major species propagated under different dissolved oxygen environments. For example, the bacteria represented by bands a
, and l
were always present at five different sampling points, and the strength was basically unchanged. It was shown that this type of bacteria was very suitable for the internal environment of the reactor and had an important effect on the removal of pollutants, belonging to the dominant species. The strains represented by bands i
, and n
had also always existed, but the brightness of the bands of this type of bacteria varied greatly at different points, indicating that this type of bacteria will only play a role in some specific zones. The bacteria represented by bands c
, and m
were not found at 1-5, and the bands were also weak in the other four samples, indicating that this type of flora was poorly adaptable to the internal environment of the reactor, and the high dissolved oxygen environment at points 1-5 might affect its survival. The bacteria represented by bands o
were not found in the lanes 1-5 and 4-7, but they were evident in the lanes of samples 4-5, 7-5, and 4-3, indicating that this kind of bacteria was not suitable for zones with high water flow velocity. The species represented by band g
only appeared in samples 4-5, indicating that this group of bacteria may only exist in a hypoxic environment.
The UPGMA (Unweighted pair group method using arithmetic average) algorithm was used to perform cluster analysis of the similarity degree of the bacterial communities on the five sample lane maps. The results are shown in Figure 8
b. Samples 1-5 and 4-7 had the highest similarity, while sample 4-5 had the lowest similarity to other samples. From the figure, the similarity between the sample flora was closely related to the external environment where the sample sampling point was located. Because the sewage in the reactor was in a cyclic push flow movement, the closer the sampling layer to the stratosphere, the higher the similarity of bacteria between the samples. At the sampling point 4-5 in the center of the reactor, because the water entered the reactor from there, and it is also the lowest DO of the reactor, the community distribution was more complicated, and the difference with other samples was also large. The above analysis showed that during the operation of the reactor, the structure of the microbial population had undergone complex succession changes and there were large differences in space.
DNA bands from the DGGE profiles were cloned and sequenced, and the sequences were compared with the data in the Gene Bank for homology analysis. As shown in Table 3
, most bacteria in the reactor were Uncultured bacterium
. Among the 16 DNA bands, 8 bands had the highest similarity with Proteobacteria
, 3 bands had the highest similarity with Bacteroidetes
, 3 bands had the highest similarity with Firmicutes
, and 1 band had the highest similarity with Acidobacteria
. Numerically, Acidobacteriumsp
, and Flavobacteriales
bacterium were dominant and played an important role in the organic carbon and nitrogen removal process in the reactor.