Journal of Analytical, Bioanalytical and Separation Techniques | Volume 3 | Issue 1 | 2018-02-23

Performance and the Microbial Community Changes in an Intelligent Aeration Control System

Xinsheng Chai

Full text

Biological treatments, including anaerobic fermentation and aerobic phosphorus-uptake process [1] , are the typical processes that are widely used in modern Waste water Treatment Plants (WWTPs), because they can effectively degrade or removal many organic as well as inorganic pollutants (e.g., nitrate and phosphate) from the domestic waste water and industrial effluents through microbial metabolisms in aqueous system [2] . It is well known that the Dissolved Oxygen (DO) in the liquid system has a significant impact to the performance of the pollutants removal in these biological treatments. For instance, the level of DO must be below 0.2 mg/L for a better anaerobic microbial growth and the suitable range is between 2.0 to 3.0 mg/L for the aerobic microbial growth in a traditional (A/O) process (Wu et al., 2015). Therefore, a good environment, e.g., with an ideal DO level for the microbial metabolisms, is crucialtoin the biological treatment processes. Currently, the periodical aeration is a common practice to obtain anaerobic or anaerobic environment in WWTPs. Clearly, the wisely controlled aeration ( i.e., extended aeration and limited aeration) is important to the performance of the biological treatment processes. An effective process means that there are some better microbial communities in the system. Therefore, the identification of the microbial community changes in the effective biological process ( i.e., with a good aeration control) will be valuable in finding and selecting the more effective microbial communities, from which the performance of the pollutants removal can be improved or further optimized. There are many studies available for exploring the good aeration control strategy in the biological treatment processes (Cydzik-Kwiatkowska, 2014; Jin et al., 2015; Capodici et al., 2014).In an example, the performance of a pilot-scale biological treatment process (V = 0.45m 3 ) with a modified cascade controller operated for optimized efficient nitrogen removal was evaluated. They results showed that an 18 % of reduced air flow improved the process treatment (Thunberg et al., 2009). In an another work, a previous on/off control strategy of (MPC) in a same volume process (V = 0.45m 3 ) with a modified cascade controller by a black-box model of the aeration in combination with feed forward of the efficient nitrogen removal was also evaluated. The results showed that 20 % of energy saving and 25 % of process efficiency of nitrogen were achieved [3] . Different from the above nitrogen removal controlled mode, a rule-based feed forward methodology for an A 2 /O process, based on the cumulative frequency distribution of the influent DO set-point, was developed. The results showed that this mode its gained aeration decreasing of 8.5 % and phosphate removal increasing of 11 % as well (Shen et al. 2010). (AOSD) intelligent aeration technology of Advance Logic System Company(Tsukuba, Ibaraki, Japan) proved as a benefit aeration strategy and achieved a 35 % electric energy decrease of aeration device (Xu et al. 2014), which was used in this study as well. The research on effect of traditional aeration, such as and continuous aeration, intermittent aeration and extended aeration, on Ammonium-Oxidizing Bacteria (AOB) and Nitrite-Oxidizing Bacteria (NOB) communities had been wide spread performed [4] . Aeration intensity seemed the primary factor for aeration zone microbial community, based on a lab-scale Sequencing Batch Reaction Kettle (SBRK), to optimize the system performance, which inferred that high aeration pressure played a key role on sewage nutrients rapid removal [5] . In the same way, aeration frequency and volume can positively affected sludge-based soil s urease, phosphatase and catalase activity and soil microbial abundance, the impact of aeration treatment on rhizosphere soil enzyme activity found larger than its impact on non-rhizosphere activity as well [6] . Recently, a model based on intermittently aerated MABRs was developed and predicted strong periodic shifts in the spatial gradients of DO, pH, free ammonia, and free nitrous acid, associated with aerated and non aerated phases, as well as AOB dominated, while NOB were suppressed under the condition of appropriate aeration intermittency and duration were effective control parameters [7] . As much of the research on aeration was performed, however, there is still a lack of knowledge about the structure of the microbial community, except for Ammonium-Oxidizing Bacteria (AOB) and Nitrite-Oxidizing Bacteria (NOB) communities, and the factors affecting the formation of consortiums under the mood of intelligent aeration control [6] . Recently, few studies have focused on characterizing communities of sludge under the intelligent aeration control in multi-species with molecular methods and have provided valuable insights [8] . Additionally, the development of these methods for use in sludge should lead to the development of specific molecular probes for rapid monitoring of the communities present in the reactors based on intelligent aeration control (Carney et al. 2014). The aim of this study was to investigate the performance of A/O process systems by using a black-box model of intelligent aeration control and analyze the bacterial consortiums in the same lab-scale medium compositions. One of the A/O process systems was equipped with intelligent aeration control device in its traditional aeration device in aeration pond. The other one had the same devices, but no additional intelligent aeration control device. Material and methods AOSD construction and operation AOSD was designed as a new type of aeration device which is developed for WWTPs in which an experimental monitoring system could be installed for control research activities. The availability of several measured variables allows the application of both the supervisory control and the data acquisition. Using a laboratory-based approach, an AOSD has been installed in aerobic ponds of a system which based on a continuous aeration A/O process. This system, named as the experimental system (I-A/O), consisted of one anoxic pond, two aerobic ponds and one secondary sedimentation tank, which both built of polymethyl methacrylate and set of the same working volume of 10 L. The control group(C-A/O) and the experimental system (I-A/O) were established as parallel continuous aeration A/O process systems. The front one, conversely, hasn t installed AOSD which given its available functional characteristics by cooperation and coordination of a computing machine, a programmable logic controller (PLC), a DO monitor probe and a temperature sensor. Anoxic pools of two systems, approximately, were controlled theirs rate of inflow in 40 mL min -1 , hydraulic retention time in 12.5 hours, sludge retention time in 20 days, nitrate recirculation ratio in 300 % and aeration flow rate in 2.2 mL min -1 . Besides, the C-A/O system worked in a continuous oxygen supply process. The I-A/O system, however, acquired its measurements of DO and temperature used to periodically update to the computing machine in which calculating and hence gaining its oxygen consumption rate, which provide basis and direction for oxygen supply at the next operating cycle of 120 minutes. Its aerobic/anoxic alternating internal environment and the activated sludge domestication, for the operation condition of on-demand oxygen supply was shown in [Figure. 1]. Figure 1: Control mode of AOSD. Seed sludge and synthetic wastewater The systems were seeded with sludge obtained from Jiangnan waste water treatment plant (Nanning, China) with an aerobic anoxic aerobic (conventional A 2 /O) process. The influent synthetic waste water, was prepared in deionized water containing: glucose, NaAc, NH 4 Cl, KH 2 PO 4 and the micronutrients contained EDTA, ZnSO 4 , CoCl 2 , CuSO 4 , KI, FeCl 3 . Analytical methods During these systems operation periods, sludge samples were taken and analyzed. The Mixed Liquor Suspended Solids (MLSS), Sludge Volume Index (SVI), Settling Velocity (SV), Chemical Oxygen Demand (COD), Ammonium (NH 4 + -N), Nitrite (NO 2 - -N), Nitrates (NO 3 - -N) Total Nitrogen (TN) and turbidity were measured in accordance with standard methods (APHA, 2005). The Dissolved Oxygen (DO) and temperature were measured by using an oxygen electrode (SC1000, HACH, USA). Besides, the Specific Oxygen Uptake Rate (SOUR) was measured, additionally, were protein and polysaccharide concentrations in Extracellular Polymeric Substances (EPS). A protein concentration in EPS (EPSp) was determined by Bradford method with Bovine Serum Albumin (BSA) as standard (Ref), whereas polysaccharides (EPSc) were measured by the phenol sulfuric acid method with glucose as a standard (Ref). Glycogen (Gly) which was determined by phenol sulfuric acid method to represent the carbon source content of sludge. In this method, pellet and supernatant were separated from 5 ml of the mixed liquor sample which had been added 0.5 ml of formaldehyde (20 %) by centrifuging at 4000 rpm for 10 min. After the pellet stored at -20 C for 24 h, an operation of low-temperature vacuum drying was taken at vacuum freeze-drying machine (LGJ-10) at the condition of -54.5 C, 0.1 mbar for 24 h. Then these samples, after 5 ml hydrochloric acid solution had been involved, were removed in digestion tubes and digested at drying oven at 105 C for 6 h. The supernatant of samples, consequently, were obtained by centrifugation which has been measured by the phenol sulfuric acid method with glucose as a standard (Ref). High-throughput sequencing and data analysis CTAB method was used to isolate the DNA from samples of sludge. Purity and concentration of the extracted DNA were detected by and agarose gel (1 %) electrophoresis as following step. With primer: 5 -ACT CCTACG GGA GGC AGC AG-3 and primer: 5 -GGA CTA CHV GGGTWT CTA AT-3 , 16S rRNA gene libraries of bacteria which target the V3-V4 hyper variable region was amplified by PCR. Illumina HiSeq platform (Illumina, USA) was applied for sequencing the PCR products and its raw data has been deposited into the sequencing reads archive database. After removing low-quality and chimeric sequences, all datasets were proved achieving their high-quality reads. These reads, then, were clustered into Operational Taxonomic Units (OTUs) and the alpha-diversity indices (ACE richness, Chao 1 richness, Simpson diversity, Shannon diversity and Coverage percentage) of each sample were calculated by using UCLUST at 97 % identity level. Functional analysis was performed in PICRUSt using the KEGG pathway database and a p-value cut-off of 0.05. Result Systems performance during the start-up During 22 days of start-up operation, these systems, as shown in Figure.2, have shown their diverse removal efficiency of contaminants, where as both have revealed the efficiency divided into fluctuation period and stabilization period, which found as a series of changes from rapid degeneration to gradually improved and, eventually, reach a stable and highly efficient operation. Mean while, regarded as 16 days was the fluctuation period, when systems both had consistent effluent quality. Figure 2: Removal efficiency of contaminant. On the fluctuation period (1 ~ 16 days), the ammonium (NH 4 + -N) of I-A/O system showed a downward tendency from 20.32 to 4.31 mg/L as the nitrite (NO 4 - -N) decreased likewise from 0.23 to 0.08 mg/L, while the C-A/O system, approximately, stayed at low-density levels of 1.07 and 0.01 mg/L. Above may resulted in a evolution process, from growth-suppressing to promoting, of Ammonia-Oxidizing Bacteria (AOB) and Nitrite-Oxidizing Bacteria (NOB), for Glycogen-Accumulating Organisms (GAOs), which had proved possessing oxygen consumption advantage, had predominated at the preliminary start-up operation of I-A/O system leaded to these growth-inhibitions, yet the yield coefficient (Y H ) and activities of Glycogen-Accumulating Organisms (GAOs) occurred decreasing trend under the low oxygen supply by AOSD and contributed to low Ammonium (NH 4 + -N) and Nitrite (NO 2 - -N) densities of effluent. The effluent COD density of I-A/O system, similarly, had existed a fluctuation of 60 ~ 77 mg/L at 3 ~ 8 days, still it was leveling out at around 30 mg/L after 15 day. The Nitrogen (TN) of effluent removal, what s more, were observed a positive correlation with Ammonium (NH 4 + -N) whose cumulative effect at the early period of start-up might shaped the high density of nitrogen (TN) at I-A/O system. During the start-up period, the Dissolved Oxygen (DO) density at aerobic compartment of C-A/O system generally remained at about 2 ~ 4 mg/L compared with 1.17 ~ 2.61 mg/L at I-A/O system, aeration of which reduced by (71.1 + -4.04) % in a operation cycle (120 min), hence the activated sludge microbial communities may have achieved a low-oxygen-demanding stable. Activated sludge properties during the start-up Figure 3 shows that the activated sludge properties in both two systems during the start-up period with the rapid increase in the Settling Velocity (SV) from approximately 42 % to (91.36 1.42) % in I-A/O system and (86.45 1.37) % in C-A/O system during the first 12 days, and the Sludge Volume Index (SVI), roughly, found as a decreasing trend. Above emerged a circumstance that there was a slightly more intensive activated sludge bulking happened in I-A/O system, still its SVI approximately stayed in 130 ~ 170 ml/g followed with a low turbidity of 4.52 NTU after 21 day. Gradually, after 12 day, the SV and SVI values of both systems seemed staying at a higher level, which followed with down-turn tendency of Mixed Liquor Suspended Solids (MLSS). Figure 3: Evolution characteristic of activated sludge properties. However the Specific Oxygen Uptake Rate (SOUR) proves as a significant parameter which is used to identify the oxygen consumption intensity of activated sludge were not quite the same in both systems with higher values in C-A/O system compared with I-A/O at 4 ~ 12 days, which showed that microbes in C-A/O shows its strong adaptability in a new condition and rapid recovery of metabolic activity in a relatively short period. Microbes in I-A/O system remained an oxygen supply inhibition, which was counted as (13.66 0.62) mg/gh, all just the same it got a relatively high value of SOUR after 12 day. The Extracellular Polymeric Substances (EPS) production of sludge combined with protein, polysaccharide concentrations, humic acid, DNA, etc is a extremely crucial part of activated sludge, in which protein (EPSp) and polysaccharide (EPSc) are two dominant compositions and serve as cell protection, adhesion and flocculation even providing carbon resources in endogenous phase. What s more can be seen in Figure 4 that both systems appeared curative decreases of EPSc at the beginning of start-up period, all the same they showed reverse trends of EPSp. The EPSc in I-A/O system, respectively, increased from 6.2 mg/g to 12.62 mg/g at 4 ~ 12 days and reached a stable state after 12 days, yet this value in C-A/O system enhanced from 7.69 mg/g to 11.36 mg/g at 4 ~ 12 days, and then, gradually, decreased to 9.28 mg/g at 22 day. Since a high content of EPSp, more specifically, has been associated with improvements in flocculation properties, which is more involved than the EPSc in electrostatic bonds with multivalent cations, a key factor in stabilizing the aggregate structure. The ratio of protein and polysaccharide (EPSp/ EPSc ratio) derived at different settling conditions were significant inconsistent, too. In I-A/O system, the EPSp/ EPSc ratio has been consistent growth from 5.36 to 22.64 at 4 ~ 22 days, whereas the index has been increased from 8.72 to 19.16 in C-A/O system at the same time. There was an obvious correlativity between the ratio of protein and polysaccharide (EPSp/EPSc ratio) and the change of chemical functional group on sludge surface, for the reason that EPSp and EPSc have diverse functions in the treatment process. Also appeared was a strong correlation between the EPSp/EPSc ratio in I-A/O system and the density of floc COD (r = -0.985, p = 0.002 < 0.05), which might contribute to underutilized soluble COD (SCOD) transferred into the gel-like properties of sludge and stored as polysaccharide (EPSc) at the relatively low DO condition. Microbial community composition of the systems To investigate the microbial community composition of the I-A/O system and the C-A/O system at the end of start-up period, 16S rRNA gene based high-throughput sequencing technology was applied to analyze the sludge samples. In the present high-throughput sequencing study, 159,946 raw sequences were obtained. After quality trimming and chimera checking was conducted, 114,507 high-qualities reads remained, accounting for 71.6 % of the valid reads with an average of 57,254 reads per bar-coded sample recovered for downstream analysis. The summary information and the detailed characteristics of each sample are shown in Table 1. In total, 1,220 unique sequences were obtained from the two groups, representing all of the phylotype present in the I-A/O system and the C-A/O system at the end of start-up period. The Good s coverage values were high for all of the sequences in the two groups, indicating that the sequencing depth was sufficient for the investigation of the microbiota of the two systems. Indicators of the alpha diversity, such as the ACE and the Chao 1 richness estimators, demonstrated that the quantity of species of the microbiota of the two groups were roughly similar, yet I-A/O group s was slightly higher than C-A/O group s as well. The other indicators of the alpha diversity, such as Shannon and the Simpson diversity indices, illuminated that the level of diversity of the microbiota, which is determined by its spices richness and evenness, of I-A/O group and C-A/O group remained low, hence the low spices abundance of both I-A/O group and C-A/O group were signified and clarified. Table 1: Comparison of phylotype coverage and diversity estimation of the 16S rRNA gene libraries at 97 % similarity from the high-throughput sequencing analysis. (1) The operational taxonomic units (OTUs) were defined at the 97 % similarity level. (2) The coverage percentage (Good s), richness estimators (ACE and Chao1) and diversity indices (Shannon and Simpson) were calculated using good s method and the mothur program, respectively. OTU ACE Chao 1 Simpson Shannon Coverage I-A/O 1063 1096.0277 1130.8814 0.0451 4.9187 0.9982 C-A/O 1026 1054.7856 1070.507 0.0242 5.0117 0.9984 Not only did the rarefaction analysis estimates show that the species richness of the sludge microbiota of I-A/O and C-A/O were significantly similar, but also their OTU number tend to be flat when the number of sequencing reads was beyond 20000, approximately (Figure 4a). In addition, a Venn diagram displaying the overlapping OTU data for the two groups was developed to better understand their shared richness. The analysis showed at Figure. 4b that only 869 of the 1,220 OTUs accounting for the total richness were common to all of the samples, whereas approximately 200 of the OTUs identified in the I-A/O samples were not detected in C-A/O samples. These data demonstrated that their similarities of microbial community stayed a comparatively high level. Figure 4: Comparison of the structures of the microbiota of the I-A/O and C-A/O systems. Rarefaction curves were used to estimate the richness (at a 97 % similarity level) of the sludge microbiota of the two groups (a). Venn diagram illustrating the overlap of the OTUs identified in the microbiota of the two groups (b). Both consistencies and inconsistencies of I-A/O and C-A/O systems in sludge microbiota species distribution were found, which, likewise, has been shown in Figure 5 based on relative abundance of phyla, classes, families and genera. At the phylum level, the sludge microbiota of two systems fell into basic tenphylums, including the proportions of Proteobacteria, Bacteroidetes, Candidate division TM7, Planctomycetes, Chloroflexi, Chlorobi, Nitrospirae, Acidobacteria, Actinobacteria etc. The phyla of Proteobacteria and Bacteroidetes , particularly, seemed in dominant position in two systems, whereas they both more abundant in I-A/O system than in C-A/O system. At the class level, what s more, the members of -proteobacteria, -proteobacteria, -proteobacteria, &del-proteobacteria, Cytophagia and Sphingobacteria were prevalent in both two systems. Proteobacteria were not diverse in two systems where -proteobacteria identified as the most abundant Proteobacteria , which found as its relative abundance of 28.22 % in I-A/O system contrasted with 31.03 % in C-A/O system. In addition to the differences in the dominant species, the relative abundance of the microbiota differed substantially between two systems in which the proportion of other members, except for -proteobacteria , showed their relative abundances were lower in I-A/O system than theirs in C-A/O system. At the lower taxonomic level, differences in the distribution of species in two systems were also found, still the most abundant species in two systems both were Cytophagaceae and Chitinophagaceae . There were not identified to species level; however, there were genetic diversity within the genus. The members of Candidatus Accumulibacter was the most dominant genus in I-A/O system and its relative abundance was 9.77 % and the members of Dechloromonas was in dominant place in C-A/O system and its relative abundance was 10.14 %. As shown in Figure 6, although using Welch s t-test, there were only statistically differences between I-A/O system and C-A/O system in the functional relative abundance. Figure 5: Species distribution. Relative abundance of dominant bacterial phyla (a), classes (b), families (c) and genera (d) in sludge samples of I-A/O system and C-A/O system. Figure 6 : Comparison of relative abundance of KEGG pathway for the metagenomes of I-A/O system and C-A/O system. Welch s t-test was used as the statistical test. The confidence intervals were calculated using Welch s inverted method. Corrected p-valued were calculated using the Benjamini-Hochberg FDR approach. References 1. Narayanasamy, S., Muller, E.E., Sheik, A. R., et al. Integrated omics for the identification of key functionalities in biological wastewater treatment microbial communities. (2015) Microb Biotechnol 8(3): 363-368. Pubmed| Crossref| Others 2. Pire-Sierra, M.C., Cegarra-Badell, D.D., Carrasquero, S., et al. Nitrogen and cod removal from tannery wastewater using biological and physicochemical treatments. (2016) Revista Facultad De Ingenieri?a 2016(80): 63-73. Pubmed | Crossref| Others 3. Jimenez, J., Regmi, P., Wise, G. Mainstream nitrite-shunt with biological phosphorus removal using low dissolved oxygen aeration control strategy. (2015) Proceedings of the Water Environment Federation 1-6(6). Pubmed | Crossref| Others 4. Effect of Aeration Intensity on n2o Emission During Simultaneous Nitrification and Denitrification (denitrification). (2011) Chinese Journal of Environmental Engineering 12 (12): 2658 -2662. Pubmed | Crossref | Others 5. Xin, X., He, J., Wang, Y., et al. Role of aeration intensity on performance and microbial community profiles in a sequencing batch reaction kettle (sbrk) for wastewater nutrients rapid removal. (2016) Bioresour Technol 201: 140-147. Pubmed| Crossref| Others 6. Li, Y., Niu, W., Wang, J., et al. Effects of artificial soil aeration volume and frequency on soil enzyme activity and microbial abundance when cultivating greenhouse tomato. (2016) Soil Science Society of America Journal 80(5): 1208-1221. Pubmed | Crossref| Others 7. Ma, Y., Domingof lez, C., Pl sz, B. G., et al. Intermittent aeration suppresses nitrite-oxidizing bacteria in membrane-aerated biofilms: a model-based explanation. (2017) Environ Sci Technol 51(11): 6146-6155. Pubmed| Crossref| Others 8. Bilodeau, K., Desilva, V. Ammonia-based aeration control on-line probes to save energy cost and meet compliance. (2015) Proceedings of the Water Environment Federation 435-442(8). Pubmed | Crossref| Others 9. Flores-Tlacuahuac, A., Pedraza-Segura, L. Optimal model-based aeration control policies in a sequencing batch reactor. (2016) Computers Chemical Engineering 85: 124-135. Pubmed | Crossref | Others