- Email: [email protected]
Effects of carbon sources and operation modes on the performances of aerobic denitrification process and its microbial community shifts
Journal of Environmental Management 239 (2019) 299–305
Contents lists available at ScienceDirect
Journal of Environmental Management journal homepage: www.elsevier.com/locate/jenvman
Research article
Effects of carbon sources and operation modes on the performances of aerobic denitrification process and its microbial community shifts
T
Bo Hua,b,e,∗, Tong Wanga,b,e, Junhong Yec,d,e, Jianqiang Zhaob,c,d,e, Liwei Yanga,b,e, Pei Wua,b,e, Jianlei Duana,b,e, Guiqi Yea,b,e a
School of Civil Engineering, Chang’ an University, Xi'an, 710054, China Key Laboratory of Water Supply & Sewage Engineering, Ministry of Housing and Urban-rural Development, Xi'an, 710054, China c School of Environmental Science and Engineering, Chang’ an University, Xi'an, 710054, China d Key Laboratory of Environmental Protection & Pollution and Remediation of Water and Soil of Shaanxi Province, Xi'an, 710054, China e Chang'an University, The Middle Section of the South 2nd Ring Road, 710064, Xi'an, Shaanxi Province, China b
A R T I C LE I N FO
A B S T R A C T
Keywords: Aerobic denitrification process Operation mode Carbon source Microbial community Shift
Carbon source, operation mode and microbial species have great effects on the synthesis of poly-β-hydroxybutyrate (PHB) which has been identified as the key issue for aerobic denitrification process. In this study, an aerobic denitrification SBR was operated under anoxic/oxic mode and completely oxic mode with the carbon source of CH3COONa and CH3CH2CH2COONa, respectively. Total nitrogen (TN) removal efficiencies, PHB content in activated sludge, production of nitric oxide (NO) and nitrous oxide (N2O) of the process were investigated in great detail. The main results obtained from the trial were: (1) the average TN removal was in the range of 86.11%–90.05%; (2) the maximum TN removal efficiency and the maximum PHB content of the process being achieved when the carbon source of CH3CH2CH2COONa was applied under anoxic/oxic mode; (3) in case of CH3COONa as the carbon source, the concentrations of NO and N2O in the bulk liquid were ∼0.4 mg/L and ∼0.02 mg/L, respectively, while in case of CH3CH2CH2COONa, N2O of ∼0.2 mg/L and NO of ∼2.5 mg/L were recorded and the latter was decreased to ∼1.0 mg/L at the end of the cycle; (4) no obvious dominant genus in case of using CH3COONa, while Plasticicumulans sp. being the major microbial community when using CH3CH2CH2COONa. Overall, the effect of carbon source on microbial community is obvious. Nevertheless, operation mode affects the PHB synthesis, while PHB plays an important role in aerobic denitrification process for achieving a relatively high TN nitrogen removal efficiency. CH3COONa is a better carbon source for aerobic denitrification compared with CH3CH2CH2COONa.
1. Introduction Aerobic denitrification process is proposed in 1980s in which NO3−N or NO2−-N are reduced to dinitrogen under aerobic environment (Watahiki et al., 1983; Robertson and Kuenen, 1984a, 1984b; Robertson et al., 1988). Nitrification and denitrification occur in one reactor, and alkalinity consumed by nitrification process can be supplemented by denitrification concurrently (Zheng et al., 2011). Aerobic denitrifiers are mainly Gram-negative bacteria in the phylum Proteobacteria (Ji et al., 2015), and can be isolated from domestic wastewater treatment plants (Wan et al., 2011; Chen et al., 2015). It has been demonstrated that DO concentration (Huang and Tseng, 2001), temperature (Wang et al., 2015, 2016), C/N ratio (Huang and Tseng, 2001), carbon source (Richardson and Ferguson, 1992) and
∗
operational mode (Alzate Marin et al., 2016) could affect the performance of the aerobic denitrification process, while DO concentration and C/N ratio are the major factors (Huang and Tseng, 2001). It has been noted that the activity of periplasmic nitrate reductases (Nap), which are responsible for aerobic denitrification, is affected by carbon source. Compared with malate and succinate, Nap activity of Thiosphaera pantotropha (which had been revised to Paracoccus pantotrophus) is highest when growing on butyrate and caproate (Richardson and Ferguson, 1992). Ellington et al. (2003) declares that Nap activity of Paracoccus pantotrophus was highest when the growth substrate was butyrate. Poly-β-hydroxybutyrate (PHB) has been identified as the carbon source during aerobic denitrification (Bernat and Wojnowska-Baryła, 2007). The synthesis of PHB can be highly affected by carbon source
Corresponding author. School of Civil Engineering, Chang’ an University, Xi'an, 710054, China. E-mail address: [email protected] (B. Hu).
https://doi.org/10.1016/j.jenvman.2019.03.063 Received 10 October 2018; Received in revised form 10 March 2019; Accepted 13 March 2019 0301-4797/ © 2019 Elsevier Ltd. All rights reserved.
Journal of Environmental Management 239 (2019) 299–305
B. Hu, et al.
and operation mode and microbial species. For carbon sources of acetate, ethanol and glucose, PHB storage is only relevant mechanism for the removal of acetate and ethanol (Majone et al., 2001). Besides, based on the study carried out by Liu et al. (2016), the maximum PHB yield of 64 wt% is achieved in a SBR reactor with anaerobic-aerobic and feast-famine operation mode, however, when the reactor is operated under completely aerobic and feast-famine mode, the maximum PHB yield is reduced to 53 wt%. The synthesis of PHB also depends on the microbial species, for example, Plasticicumulans acidivorans and Thauera selenatis have completely different growth characteristics and PHB producing capacity (Jiang et al., 2011). However, the effects of PHB on the performance of aerobic denitrification are not clear. Meanwhile, the roles of carbon source and operation mode on PHB synthesis in aerobic denitrification process are also vague. The aims of this study are: (1) to study the effects of carbon source and operation mode on the performance of aerobic denitrification process. As sodium acetate (CH3COONa) is commonly used either in full-scale or in lab-scale biological wastewater treatment process (Lee and Welander, 1996). CH3COONa and CH3CH2CH2COONa were used in this study; (2) to reveal the effects of carbon source and operation mode on PHB synthesis; (3) to explore NO and N2O production characteristics of aerobic denitrification process; (4) to examine the microbial community shifts under different operation conditions.
Table 2 Influent qualities of the four experiments.
Anoxic/Oxic Oxic Oxic Anoxic/Oxic
100:10:4
50 mg/L 20 mg/L 500 mg/L 50 mg/L 20 mg/L
In order to study the long-term performances of the process, COD, NH4+eN, NO3−-N, NO2−-N, PO43−-P and TN concentrations in effluent were determined every 1 or 2 days. After cultivation for 40 days, the aerobic denitrification SBR process was successively start-up. When the reactor became pseudo-stable, four experiments were conducted to study the effects of carbon source and operation mode on the aerobic denitrification process. Long-term performance of the aerobic denitrification SBR process was shown in Fig. 2, while the effluent qualities of the process under different conditions were shown in Table 4. In four experiments, COD and TN removal efficiencies were higher than 94% and 86.0%, respectively. By inspecting the data, it showed that when the process was operated under A/O mode with CH3CH2CH2COONa as the carbon source, the TN removal efficiency of 90.05% ± 0.03% (n = 21) was the highest in four experiments. Under complete oxic mode, there were no obvious differences between two carbon sources from the perspective of the effluent qualities. 3.2. Effects of operation mode and carbon source on aerobic denitrification SBR process
Table 1 The carbon sources, operation modes and C:N:P ratios of the four experiments.
CH3COONa CH3COONa CH3CH2CH2COONa CH3CH2CH2COONa
500 mg/L
3.1. Long-term performance of aerobic denitrification SBR process
Four experiments were conducted to study the effects of carbon source and operation mode on the aerobic denitrification process. The carbon sources, operation modes and C: N: P ratio of the four experiments were shown in Table 1. The influent qualities of the four experiments were shown in Table 2. The influent also contained MgSO4·7H2O and CaCl2 of 50.0 mg/L and 20.0 mg/L, respectively, while the trace elements were also dosed with detailed description as Cheong and Hansen (2008). All experiments were operated with a cyclic duration of 12 h. The operation procedures were shown in Fig. 1 and Table 3.
Ⅰ Ⅱ Ⅲ Ⅳ
CH3COONa NH4HCO3 K2HPO4 KH2PO4 CH3CH2CH2COONa NH4HCO3 K2HPO4 KH2PO4
PO43−-P
3. Results
2.2. Design of experiments
C:N:P ratios
Ⅰ and Ⅱ
NH4+eN
During the experiments, COD, NH4+eN, NO3−-N and NO2−-N of the samples were monitored according to standard methods (APHA, 1998). N2O-N and NO-N were monitored when the process was operated under completely oxic mode. Their concentration profiles in the reactor were continuously monitored using Clark-type microelectrodes (Arhus, Denmark). The microelectrode was calibrated with the twopoint method according to the instruction provided by Unisense (Arhus, Denmark). DO concentration and temperature in the reactor were measured by DO meter (HACH), while pH value was measured by pH meter (Rex, Shanghai). PHB contents in activated sludge were determined based on the gravimetric method which was modified from the method proposed by Hahn et al. (1994) and YüksekdaĞ et al. (2003). The detailed procedures were elucidated in Supplementary Materials. The microbial community structure was analyzed by highthroughput sequencing carried out by Sangon Biotech Co., Ltd. (Shanghai, China). The detailed information was shown in Supplementary Materials.
A lab-scale SBR reactor was made from transparent Plexiglas with a working volume of 10 L. The diameter and the height of the reactor were 200 mm and 400 mm, respectively. The reactor was equipped with a rectangular mixing paddle and an air compressor to supply air through a diffuser placed at the bottom of the reactor. The DO concentration in the reactor was maintained at ∼6.5 mg/L. The temperature of the reactor was controlled at 29 °C ± 1 °C by a water jacket. The sludge retention time was 25 days. Correspondingly, the average sludge concentrations of the reactor were 4207 mg/L and 3468 mg/L when the carbon sources were CH3COONa and CH3CH2CH2COONa, respectively. The drainage ratio was 0.5. Before operation, the reactor was seeded while the seeding sludge was taken from the oxic unit of the 4th wastewater treatment plant of Xi'an city, Shaanxi province, China.
Operational modes
COD
2.3. Analytical methods
2.1. The aerobic denitrification SBR reactor
Carbon source
Compounds
Ⅲ and Ⅳ
2. Material and methods
Experiment
Experiments
After the reactor reached steady state during the four experiments, variations of COD, NH4+eN, NO3−-N, NO2−-N, PO43−-P, TN and PHB contents of 4 typical cycles of each experiment were monitored (Fig. 3). Particularly, as the procedure for PHB determination was quite complex, the activated sludge samples for determining PHB content were less than the wastewater samples. Under A/O mode, TN concentrations in the effluents of the process were 6.71 ± 0.19 mg/L (n = 4) and 4.36 ± 0.2 mg/L (n = 4), 300
Journal of Environmental Management 239 (2019) 299–305
B. Hu, et al.
Fig. 1. The operation procedures of the four experiments.
Under the same operation mode, the TN removal efficiencies of the process and the maximum PHB contents of activated sludge were higher when the carbon source was CH3CH2CH2COONa. Furthermore, there was a distinct relationship between the PHB content and the TN removal efficiency of the process under the same operation mode. Under both the A/O mode and the completely oxic mode, the TN removal efficiency was proportional to the maximum PHB content in the activated sludge, i.e. the high maximum PHB content could lead to the high TN removal. Notably, under completely oxic mode, there were no obvious differences in TN removal efficiencies of the process under two carbon sources.
Table 3 Phases’ duration of the aerobic denitrification SBR process in four experiments. Experiment
Feeding
Mixing
Aeration and Mixing
Settling
Drawing
Resting
Ⅰand Ⅳ Ⅱ and Ⅲ
10 min 10 min
120 min 0 min
540 min 660 min
10 min 10 min
10 min 10 min
30 min 30 min
In experiment Ⅰ and Ⅳ, the duration of feeding, mixing, aeration and mixing, settling, drawing and resting were 10 min, 120 min, 540 min, 10 min, 10 min and 30 min, respectively. In experiment Ⅱ and experiment Ⅲ, mixing was cancelled. The duration of feeding, aeration and mixing, settling, drawing and resting were 10 min, 660 min, 10 min, 10 min and 30 min, respectively.
3.3. N2O and NO productions in aerobic denitrification process
respectively, and the TN removal efficiencies of the process were 86.11 ± 0.0043% (n = 4) and 91.30 ± 0.0036% (n = 4) when carbon sources were CH3COONa and CH3CH2CH2COONa, respectively. Correspondingly, the maximum PHB contents of activated sludge were 19.56% and 22.23% respectively. Under completely oxic mode, TN concentrations in the effluents of the process were 5.41 ± 0.76 mg/L (n = 4) and 5.26 ± 0.86 mg/L (n = 4), respectively, and the TN removal efficiencies of the process were 88.12 ± 0.017% (n = 4) and 89.30 ± 0.018% (n = 4) when carbon sources were CH3COONa and CH3CH2CH2COONa respectively. Correspondingly, the maximum PHB contents were declined to 9.10% and 14.08% respectively. The maximum PHB contents in activated sludge and TN removal efficiencies of the four experiments were listed in Table 5.
In order to examine NO and N2O production of aerobic denitrification process, NO and N2O concentrations in the reactor were continuously monitored when the process was operated under completely oxic mode (Fig. 4). When CH3COONa was the carbon source, the concentrations of NO and N2O in the bulk liquid of the typical cycles were ∼0.4 mg/L and ∼0.02 mg/L, respectively. However, when the carbon source was CH3CH2CH2COONa, the concentrations of N2O in the bulk of the typical cycles were ∼0.2 mg/L. The variation of NO was completely different from that when the carbon source was CH3COONa. The concentrations of NO were stable at the most of the cycle and decreased in the latter of the cycle. In the stable stage, the concentration of NO in the bulk liquid was ∼2.5 mg/L, and at the end of the cycle, it was decreased to
Fig. 2. Long-term performances of the aerobic denitrification SBR process. 301
Journal of Environmental Management 239 (2019) 299–305
B. Hu, et al.
Table 4 The long-term performances of the process under different conditions. Experiment
NH4+eN in effluent (mg/L)
NO2−-N in effluent (mg/L)
NO3−-N in effluent (mg/L)
PO43−-P in effluent (mg/L)
COD in effluent (mg/L)
TN in effluent (mg/L)
TN removal efficiency (%)
No. of Samples
I II III IV
0.31 0.31 0.31 0.27
0.07 0.07 0.69 0.07
5.73 4.98 5.37 4.57
8.94 ± 1.31 9.97 ± 1.26 10.19 ± 2.20 11.14 ± 4.17
24.52 27.20 24.67 24.70
6.11 5.36 5.79 4.91
86.11% 88.02% 88.05% 90.05%
n = 22 n = 21 n = 20 n = 21
± ± ± ±
0.00 0.00 0.00 0.10
± ± ± ±
0.02 0.00 0.00 0.00
± ± ± ±
0.75 0.70 0.82 1.31
∼1.0 mg/L. Generally, when the carbon source CH3CH2CH2COONa, more NO and N2O were produced.
± ± ± ±
2.98 1.84 4.53 2.65
± ± ± ±
0.75 0.70 0.81 1.32
± ± ± ±
1.29% 1.62% 1.93% 0.03%
mode, the abundance sequence of microorganisms with different functions was: COD degradation microorganism > Aerobic denitrifiers > EPS synthesis microorganism > Sulfur oxidation microorganism > PHB production microorganism. When the operation mode was shifted to completely oxic mode, the abundance sequence was: COD degradation microorganism > EPS synthesis microorganism > Aerobic denitrifiers > PHB production microorganism > SMP production microorganism. When carbon source was CH3CH2CH2COONa and operation mode was A/O mode, the abundance sequence of microorganisms with different functions was: COD degradation microorganism > PHB production microorganism > EPS synthesis microorganism > Aerobic denitrifiers > Filamentous bacteria. When the operation mode was shifted to completely oxic mode, the abundance sequence was: PHB production microorganism > Aerobic denitrifiers > COD degradation microorganism > EPS synthesis microorganism > Sulfur oxidation microorganism > Nitrifiers.
was
3.4. Microbial community shifts under different operation conditions The microbial communities of the four experiments and the function of each genus were shown in Table 1–Table 4 of Supplementary Materials. Genera in quite small abundances were classified to Other. Except for Unclassified and Other, the microorganisms in the system can be divided into five classes, which were responsible for degradation of organic matters, PHB production, nitrification, aerobic denitrification and the others respectively. The detailed information of microbial groups was elucidated in Supplementary Materials. In order to understand the microbial community shifts under different conditions, the variations of the ten preponderant phyla of the microbial communities under different conditions were shown in Table 5 of Supplementary Materials. The network diagram based on operational taxonomic units (OTUs) of four experiments was illustrated in Fig. 5, while the diversity indexes of four experiments were shown in Table 6. When carbon source was CH3COONa and operation mode was A/O
Fig. 3. Effects of carbon sources and operation mode on aerobic denitrification process. 302
Journal of Environmental Management 239 (2019) 299–305
B. Hu, et al.
Table 5 The maximum PHB contents in activated sludge during four experiments. Experiments
The maximum PHB content
Average TN removal efficiency
Operation mode
Carbon Source
I II III IV
19.56% 9.10% 14.08% 22.23%
86.11% 88.12% 89.30% 91.30%
A/O mode Completely Oxic mode Completely Oxic mode A/O mode
CH3COONa CH3COONa CH3CH2CH2COONa CH3CH2CH2COONa
(n = 4) (n = 4) (n = 4) (n = 4)
4. Discussion
2016). Under anaerobic condition, external carbon source is mainly used for PHB production (Pereira et al., 1996). In this study, the higher TN removal efficiency is achieved when the carbon source is CH3CH2CH2COONa. According to the published literature, Nap activity in the pure cultivation was increased with the increased extent of the reduced carbons source (Richardson and Ferguson, 1992; Sears et al., 1997; Ellington et al., 2003). Similar to pure cultivation, in mixture cultivation, the more reduced carbon source is beneficial for improving the TN removal efficiencies of the aerobic denitrification process.
4.1. TN removal efficiency of the aerobic denitrification SBR process The TN removal efficiencies of the aerobic denitrification process were higher than conventional N removal process, such as SBR, SBBR MBR, etc. (Chen et al., 2015). It has been reported that the C/N ratio required by aerobic denitrification is higher than conventional biological removal process, and the aerobic denitrification process tends to work efficiently when the C/N ratio was in the range of 5–10 (Ji et al., 2015). During the four experiments, the C/N ratio of the influent was 10, indicating that excessive organic carbons were supplied into aerobic denitrifiers. There seems a “bottleneck” in the electron transport chain of the aerobic denitrification process, thus, nitrate, nitrite or thiosulfate can all be used as the electron acceptor (Richardson and Ferguson, 1992). When excessive organic carbons are applied to the system, the number of electrons flowed into the electron transport chain are increased, which might stimulate the aerobic denitrification process and make it achieving a relatively high TN removal efficiency. In addition, an alternate anoxic and oxic operation mode is beneficial for aerobic denitrification process (Alzate Marin et al., 2016). As a result, the highest N removal efficiency is obtained under the A/O mode in this study.
4.3. Effects of carbon source on NO and N2O production in aerobic denitrificaiton process A number of studies have focused on N2O and NO emission of aerobic denitrificaiton process under the pure cultivation of aerobic denitrifiers, such as Pseudomonas stutzeri PCN-1 (Zheng et al., 2014), Chelatococcus daeguensis TAD1 (He et al., 2016), Providencia rettgeri YL (Ye et al., 2016). It has been generally recognized that aerobic denitrification can reduce the emission of N2O and NO (Zheng et al., 2014; He et al., 2016). In this study, NO and N2O productions are in relatively high levels. In addition, the NO emission in this study is higher than the N2O emission. Regarding N2O emission of aerobic denitrification process, it has been reported that N2O emission is related to the bacteria type and their oxygen tolerance of nitrous oxide reductase (Ye et al., 2016). Furthermore, according to Ellington et al. (2002), compared with CH3CH2CH2COONa, CH3COONa was more favorable to Paracoccus pantotrophus which can accelerate Nap during aerobic growth. The use of CH3CH2CH2COONa might lead to the microorganisms with a surfeit of reductant and therefore potentially a surfeit of adenosine triphosphate (ATP). An excessive ATP in microorganisms may result in a low concentration of adenosine diphosphate (ADP) and a constraint of the respiratory rate (Ellington et al., 2002). For microorganisms, this may lead to limitation of some biochemical reactions, and then, some intermediates related to these reactions may accumulate. In aerobic denitrification process, although the process achieves a relatively high TN
4.2. The role of PHB in the aerobic denitrification process PHB is a carbon-energy storage material, and can be served as electron donor for aerobic denitrification (Bernat and WojnowskaBaryła, 2007). Richardson et al. (2001) reported that reduced carbon sources could not be used directly by aerobic denitrifiers but were converted to PHB firstly. Compared with CH3COONa, CH3CH2CH2COONa is more reduced. In four experiments, under the same operation mode, the maximum PHB contents are higher when the carbon source is CH3CH2CH2COONa. Besides, the maximum PHB contents under A/O mode are higher than those under complete oxic mode. Anaerobic/aerobic switches are beneficial for PHB synthesis (Liu et al.,
Fig. 4. N2O and NO concentrations in the bulk of aerobic denitrification SBR process. 303
Journal of Environmental Management 239 (2019) 299–305
B. Hu, et al.
Fig. 5. Network diagram based on OTUs of the four experiments.
and the shear force on flocs is stronger than that of A/O mode. Ohtaekwangia sp. is important for preventing the disintegration of the flocs. When the carbon source is CH3CH2CH2COONa and the operation mode is completely oxic mode, Plasticicumulans sp. which has an amazing ability in PHB synthesis (Jiang et al., 2011) is dominant in activated sludge of the process. As mentioned before, PHB can be served as the carbon source for aerobic denitrification (Bernat and WojnowskaBaryła, 2007), and the Nap is more active when the carbon source is CH3CH2CH2COONa (Richardson and Ferguson, 1992; Ellington et al., 2003). Although the proportion of aerobic denitrifiers is smaller than that when the carbon source is CH3COONa, the TN removal efficiency of the process is not influenced as the high concentration of PHB in the activated sludge. When the carbon source is CH3CH2CH2COONa and the operation mode is converted to A/O mode, the proportion of aerobic denitrifiers is increased from 3.70% to 13.30%. A/O mode is more preferred by aerobic denitrifiers. Meanwhile, the proportion of PHB production microorganism is decreased from 43.68% to 37.83%. As alternative anaerobic and oxic environment is beneficial for the PHB synthesis (Liu et al., 2016), the maximum PHB content of activated sludge is increased from 14.08% to 22.23%. The high PHB content provided a powerful support for the experiment IV achieving the maximum TN removal efficiency. Notably, in four experiments, the proportion of nitrifiers is quite small. Not very long time ago, nitrification was considered to be conducted by autotrophs obligately. However, more and more studies have confirmed the existence of heterotrophic nitrification and aerobic
Table 6 Diversity indexes of microorganisms in the four experiments. Experiment
Shannon Index
Chao Index
Coverage
Simpson Index
Ⅰ Ⅱ Ⅲ Ⅳ
5.6668 4.5619 3.6665 3.9825
39706.7 28627.4 16175.7 20321.3
0.9092 0.9616 0.9580 0.9457
0.0133 0.0314 0.1440 0.1175
nitrogen removal efficiency when the carbon source is CH3CH2CH2COONa, a certain amount of nitrogen is removed via NO and N2O. As the negative effects of NO and N2O on environment, it is fair to say that CH3COONa is a better carbon source for aerobic denitrification, albeit CH3CH2CH2COONa can improve Nap activity and the TN removal efficiency of the process.
4.4. Microbial community shifts under different conditions When the carbon source is CH3COONa and the operation mode is converted from A/O mode to completely oxic mode, the abundance of aerobic denitrifiers and EPS synthesis microorganism (Ohtaekwangia sp.) are increased obviously as well as the aerobic denitrifiers. Ohtaekwangia sp. can synthesize extracellular polymetric substances (EPS) and accumulate phosphorous (Świątczak and CydzikKwiatkowska, 2018). EPS is responsible for the floc formation by agglomerating activated sludge (Sponza, 2003) and stabilizing aerobic granule structure (Świątczak and Cydzik-Kwiatkowska, 2018). In completely oxic mode, DO concentration is maintained at ∼6.5 mg/L 304
Journal of Environmental Management 239 (2019) 299–305
B. Hu, et al.
denitrification, such as Acinetobacter sp. HA2 (Yao et al., 2013), Providencia rettgeri strain YL (Ye et al., 2016), etc. Consequently, the nitrification of the process is not influenced as the existence of heterotrophic nitrifiers in microbial communities. When the carbon source is CH3CH2CH2COONa, the abundance of the phylum Proteobacteria is higher than those when the carbon source is CH3COONa. From the phylum level, the numbers of the microorganisms which abundance are more than 1.0% are 8, 6, 4 and 3 respectively in four experiments. When the carbon source is CH3COONa, the bacteria diversity of the reactor is richer than that when the carbon source is CH3CH2CH2COONa. The effect of carbon source on the microbial community is more obvious and direct.
Hahn, S.K., Chang, Y.K., Kim, B.S., Chang, H.N., 1994. Optimization of microbial poly(3hydroxybutyrate) recovery using dispersions of sodium hypochlorite solution and chloroform. Biotechnol. Bioeng. 44, 256–261. He, J.X., Zhou, S.F., Huang, S.B., Zhang, Y.Q., 2016. Pretreated corn hush hydrolysate as the carbon source for aerobic denitrification with low level of N2O emission by thermophilic Chelatococcus daeguensis TAD1. Water, Air, Soil Pollut. 227, 314. Huang, H.K., Tseng, S.K., 2001. Nitrate reduction by Citrobacter diversus under aerobic environment. Appl. Microbiol. Biotechnol. 55, 90–94. Ji, B., Yang, K., Zhu, L., Jiang, Y., Wang, H.Y., Zhou, J., Zhang, H.N., 2015. Aerobic denitrification: a review of important advances of the last 30 years. Biotechnol. Bioproc. Eng. 20, 643–651. Jiang, Y., Marang, L., Kleerebezem, R., Muyzer, G., van Loosdtrecht, M.C.M., 2011. Polyhydroxybutyrate production from lactate using a mixed microbial culture. Biotechnol. Bioeng. 108, 2022–2035. Lee, N.M., Welander, T., 1996. The effect of different carbon sources on respiratory denitrification in biological wastewater treatment. J. Ferment. Bioeng. 82 (3), 277–285. Liu, C.L., Liu, D., Qi, Y.J., Zhang, Y., Liu, X., Zhao, M., 2016. The effect of anaerobic–aerobic and feast–famine cultivation pattern on bacterial diversity during poly-βhydroxybutyrate production from domestic sewage sludge. Environ. Sci. Pollut. Res. 23 (13), 12966–12975. Majone, M., Beccari, M., Dionisi, D., Levantesi, C., Renzi, V., 2001. Role of storage phenomena on removal of different substrates during pre-denitrification. Water Sci. Technol. 43, 151–158. Pereira, H., Lemos, P.C., Reis, M.A.M., Crespo, J.P.S.G., Carrondo, M.J.T., Santos, H., 1996. Model for carbon metabolism in biological phosphorous removal processes based on in vivo 13C-NMR labelling experiments. Water Res. 30, 2128–2138. Richardson, D.J., Ferguson, S.J., 1992. The influence of carbon substrate on the activity of the periplasmic nitrate reductase in aerobically grown Thiosphaera pantotropha. Arch. Microbiol. 157, 535–537. Richardson, D.J., Berks, B.C., Russell, D.A., Spiro, S., Taylor, C.J., 2001. Functional, biochemical and genetic diversity of prokaryotic nitrate reductases. Cell. Mol. Life Sci. 58, 165–178. Robertson, L.A., Kuenen, J.G., 1984a. Aerobic denitrification: a controversy revived. Arch. Microbiol. 139, 351–354. Robertson, L.A., Kuenen, J.G., 1984b. Aerobic denitrification – old wine in new bottles? Antonie Leeuwenhoek 50, 525–544. Robertson, L.A., van Neil, E.W.J., Torremans, R.A.M., Kuenen, J.G., 1988. Simultaneous nitrification and denitrification in aerobic chemostat cultures of Thiosphaera pantotropha. Appl. Environ. Microbiol 54, 2812–2818. Sears, H.J., Little, P.J., Richardson, D.J., Berks, B.C., Spiro, S., Ferguson, S.J., 1997. Identification of an assimilatory nitrate reductase in mutants of Paracoccus denitrificans GB17 deficient in nitrate respiration. Arch. Microbiol. 167, 61–66. Sponza, D.T., 2003. Investigation of extracellular polymer substances (EPS) and physicochemical properties of different activated sludge flocs under steady-state conditions. Enzym. Microb. Technol. 32, 375–385. Świątczak, P., Cydzik-Kwiatkowska, A., 2018. Performance and microbial characteristics of biomass in a full-scale aerobic granular sludge wastewater treatment plant. Environ. Sci. Pollut. Res. 25, 1655–1669. Wan, C.L., Yang, X., Lee, D.J., Du, M.A., Wan, F., Chen, C., 2011. Aerobic denitrification by novel isolated strain using NO2--N as nitrogen source. Bioresour. Technol. 102 (15), 7244–7248. Wang, Y.Y., Zhang, Z.J., Qiu, L., Guo, Y., Wang, X.J., Xiong, X.J., Chen, S.H., 2015. Effect of temperature downshifts on biological nitrogen removal and community structure of a lab-scale aerobic denitrification process. Biochem. Eng. J. 101, 200–208. Wang, H.Y., Zou, Z.C., Chen, D., Yang, K., 2016. Effects of temperature on aerobic denitrification in a bio-ceramsite reactor. Energy Sources, Part A 38, 3236–3241. Watahiki, M., Hata, S., Aida, T., 1983. N2O accumulation and inhibition of N2O reduction by denitrifying Pseudomonas sp. 220A in the presence of oxygen. Agric. Biol. Chem. 47, 1991–1996. Yao, S., Ni, J.R., Ma, T., Li, C., 2013. Heterotrophic nitrification and aerobic denitrification at low temperature by a newly isolated bacterium, Acinetobacter sp. HA2. Bioresour. Technol. 139, 80–86. Ye, J., Zhao, B., An, Q., Huang, Y.S., 2016. Nitrogen removal by Providencia rettgeri strain YL with heterotrophic nitrification and aerobic denitrification. Environ. Technol 37, 2206–2213. YüksekdaĞ, Z.N., Beyatli, Y., Aslim, B., 2003. Determination of poly-β-hydroxybutyrate (PHB) production by some mesophilic and thermophilic lactic acid bacteria. Turk. J. Biol. 27, 37–42. Zheng, H.Y., Liu, Y., Sun, G.D., Gao, X.Y., Zhang, Q.L., Liu, Z.P., 2011. Denitrification characteristics of a marine origin psychrophilic aerobic denitrifying bacterium. J. Environ. Sci. 23, 1888–1893. Zheng, M.S., He, D., Ma, T., Chen, Q., Liu, S.T., Ahamd, M., Gui, M.Y., Ni, J.R., 2014. Reducing NO and N2O emission during aerobic denitrification by newly isolated Pseudomonas stutzeri PCN-1. Bioresour. Technol. 162, 80–88.
5. Conclusions (1) PHB was important for aerobic denitrification process achieving high TN removal efficiency. A/O mode was beneficial for PHB synthesis and aerobic denitrifiers accumulation. (2) For the aerobic denitrification SBR process tested, CH3CH2CH2COONa was helpful for improving the TN removal efficiency of the process, however, CH3COONa was a better carbon source instead of CH3CH2CH2COONa due to the lower production of NO and N2O. (3) Compared with the operation mode, the effects of carbon source on the aerobic denitrification process and microbial community were more obvious. Acknowledgements This work was supported by National Natural Science Foundation of China (Grant No.:51778057); Science and Technology Project of Shaanxi Province, China (Grant No.: 2018JQ5143); National Training Programs of Innovation and Entrepreneurship for Undergraduates, Chang'an University, China (Grant No.: 201810710253). Appendix A. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.jenvman.2019.03.063. References Alzate Marin, J.C., Caravelli, A.H., Zaritzky, N.E., 2016. Nitrification and aerobic denitrification in anoxic-aerobic sequencing batch reactor. Bioresour. Technol. 200, 380–387. APHA, 1998. Standard Methods for the Examination of Water and Wastewater, twentieth ed. American Public Health Association, Washington, DC, USA. Bernat, K., Wojnowska-Baryła, I., 2007. Carbon source in aerobic denitrification. Biochem. Eng. J. 36, 116–122. Chen, Q., Ni, J.R., Ma, T., Liu, T., Zheng, M.S., 2015. Bioaugmentation treatment of municipal wastewater with heterotrophic-aerobic nitrogen removal bacteria in a pilot-scale SBR. Bioresour. Technol. 183, 25–32. Cheong, D.Y., Hansen, C.L., 2008. Effect of feeding strategy on the stability of anaerobic sequencing batch reactor responses to organic loading conditions. Bioresour. Technol. 99, 5058–5068. Ellington, M.J.K., Bhakoo, K.K., Sawers, G., Richardson, D.J., Ferguson, S.J., 2002. Hierarchy of carbon source selection in Paracoccus pantotrophus: strict correlation between reduction state of the carbon substrate and aerobic expression of the nap operon. J. Bacteriol. 184, 4767–4774. Ellington, M.J.K., Sawers, G., Sears, H.J., Spiro, S., Richardson, D.J., Ferguson, S.J., 2003. Characterization of the expression and activity of the periplasmic nitrate reductase of Paracoccus pantotrophus in chemostat cultures. Microbiology 149, 1533–1540.
305
Contents lists available at ScienceDirect
Journal of Environmental Management journal homepage: www.elsevier.com/locate/jenvman
Research article
Effects of carbon sources and operation modes on the performances of aerobic denitrification process and its microbial community shifts
T
Bo Hua,b,e,∗, Tong Wanga,b,e, Junhong Yec,d,e, Jianqiang Zhaob,c,d,e, Liwei Yanga,b,e, Pei Wua,b,e, Jianlei Duana,b,e, Guiqi Yea,b,e a
School of Civil Engineering, Chang’ an University, Xi'an, 710054, China Key Laboratory of Water Supply & Sewage Engineering, Ministry of Housing and Urban-rural Development, Xi'an, 710054, China c School of Environmental Science and Engineering, Chang’ an University, Xi'an, 710054, China d Key Laboratory of Environmental Protection & Pollution and Remediation of Water and Soil of Shaanxi Province, Xi'an, 710054, China e Chang'an University, The Middle Section of the South 2nd Ring Road, 710064, Xi'an, Shaanxi Province, China b
A R T I C LE I N FO
A B S T R A C T
Keywords: Aerobic denitrification process Operation mode Carbon source Microbial community Shift
Carbon source, operation mode and microbial species have great effects on the synthesis of poly-β-hydroxybutyrate (PHB) which has been identified as the key issue for aerobic denitrification process. In this study, an aerobic denitrification SBR was operated under anoxic/oxic mode and completely oxic mode with the carbon source of CH3COONa and CH3CH2CH2COONa, respectively. Total nitrogen (TN) removal efficiencies, PHB content in activated sludge, production of nitric oxide (NO) and nitrous oxide (N2O) of the process were investigated in great detail. The main results obtained from the trial were: (1) the average TN removal was in the range of 86.11%–90.05%; (2) the maximum TN removal efficiency and the maximum PHB content of the process being achieved when the carbon source of CH3CH2CH2COONa was applied under anoxic/oxic mode; (3) in case of CH3COONa as the carbon source, the concentrations of NO and N2O in the bulk liquid were ∼0.4 mg/L and ∼0.02 mg/L, respectively, while in case of CH3CH2CH2COONa, N2O of ∼0.2 mg/L and NO of ∼2.5 mg/L were recorded and the latter was decreased to ∼1.0 mg/L at the end of the cycle; (4) no obvious dominant genus in case of using CH3COONa, while Plasticicumulans sp. being the major microbial community when using CH3CH2CH2COONa. Overall, the effect of carbon source on microbial community is obvious. Nevertheless, operation mode affects the PHB synthesis, while PHB plays an important role in aerobic denitrification process for achieving a relatively high TN nitrogen removal efficiency. CH3COONa is a better carbon source for aerobic denitrification compared with CH3CH2CH2COONa.
1. Introduction Aerobic denitrification process is proposed in 1980s in which NO3−N or NO2−-N are reduced to dinitrogen under aerobic environment (Watahiki et al., 1983; Robertson and Kuenen, 1984a, 1984b; Robertson et al., 1988). Nitrification and denitrification occur in one reactor, and alkalinity consumed by nitrification process can be supplemented by denitrification concurrently (Zheng et al., 2011). Aerobic denitrifiers are mainly Gram-negative bacteria in the phylum Proteobacteria (Ji et al., 2015), and can be isolated from domestic wastewater treatment plants (Wan et al., 2011; Chen et al., 2015). It has been demonstrated that DO concentration (Huang and Tseng, 2001), temperature (Wang et al., 2015, 2016), C/N ratio (Huang and Tseng, 2001), carbon source (Richardson and Ferguson, 1992) and
∗
operational mode (Alzate Marin et al., 2016) could affect the performance of the aerobic denitrification process, while DO concentration and C/N ratio are the major factors (Huang and Tseng, 2001). It has been noted that the activity of periplasmic nitrate reductases (Nap), which are responsible for aerobic denitrification, is affected by carbon source. Compared with malate and succinate, Nap activity of Thiosphaera pantotropha (which had been revised to Paracoccus pantotrophus) is highest when growing on butyrate and caproate (Richardson and Ferguson, 1992). Ellington et al. (2003) declares that Nap activity of Paracoccus pantotrophus was highest when the growth substrate was butyrate. Poly-β-hydroxybutyrate (PHB) has been identified as the carbon source during aerobic denitrification (Bernat and Wojnowska-Baryła, 2007). The synthesis of PHB can be highly affected by carbon source
Corresponding author. School of Civil Engineering, Chang’ an University, Xi'an, 710054, China. E-mail address: [email protected] (B. Hu).
https://doi.org/10.1016/j.jenvman.2019.03.063 Received 10 October 2018; Received in revised form 10 March 2019; Accepted 13 March 2019 0301-4797/ © 2019 Elsevier Ltd. All rights reserved.
Journal of Environmental Management 239 (2019) 299–305
B. Hu, et al.
and operation mode and microbial species. For carbon sources of acetate, ethanol and glucose, PHB storage is only relevant mechanism for the removal of acetate and ethanol (Majone et al., 2001). Besides, based on the study carried out by Liu et al. (2016), the maximum PHB yield of 64 wt% is achieved in a SBR reactor with anaerobic-aerobic and feast-famine operation mode, however, when the reactor is operated under completely aerobic and feast-famine mode, the maximum PHB yield is reduced to 53 wt%. The synthesis of PHB also depends on the microbial species, for example, Plasticicumulans acidivorans and Thauera selenatis have completely different growth characteristics and PHB producing capacity (Jiang et al., 2011). However, the effects of PHB on the performance of aerobic denitrification are not clear. Meanwhile, the roles of carbon source and operation mode on PHB synthesis in aerobic denitrification process are also vague. The aims of this study are: (1) to study the effects of carbon source and operation mode on the performance of aerobic denitrification process. As sodium acetate (CH3COONa) is commonly used either in full-scale or in lab-scale biological wastewater treatment process (Lee and Welander, 1996). CH3COONa and CH3CH2CH2COONa were used in this study; (2) to reveal the effects of carbon source and operation mode on PHB synthesis; (3) to explore NO and N2O production characteristics of aerobic denitrification process; (4) to examine the microbial community shifts under different operation conditions.
Table 2 Influent qualities of the four experiments.
Anoxic/Oxic Oxic Oxic Anoxic/Oxic
100:10:4
50 mg/L 20 mg/L 500 mg/L 50 mg/L 20 mg/L
In order to study the long-term performances of the process, COD, NH4+eN, NO3−-N, NO2−-N, PO43−-P and TN concentrations in effluent were determined every 1 or 2 days. After cultivation for 40 days, the aerobic denitrification SBR process was successively start-up. When the reactor became pseudo-stable, four experiments were conducted to study the effects of carbon source and operation mode on the aerobic denitrification process. Long-term performance of the aerobic denitrification SBR process was shown in Fig. 2, while the effluent qualities of the process under different conditions were shown in Table 4. In four experiments, COD and TN removal efficiencies were higher than 94% and 86.0%, respectively. By inspecting the data, it showed that when the process was operated under A/O mode with CH3CH2CH2COONa as the carbon source, the TN removal efficiency of 90.05% ± 0.03% (n = 21) was the highest in four experiments. Under complete oxic mode, there were no obvious differences between two carbon sources from the perspective of the effluent qualities. 3.2. Effects of operation mode and carbon source on aerobic denitrification SBR process
Table 1 The carbon sources, operation modes and C:N:P ratios of the four experiments.
CH3COONa CH3COONa CH3CH2CH2COONa CH3CH2CH2COONa
500 mg/L
3.1. Long-term performance of aerobic denitrification SBR process
Four experiments were conducted to study the effects of carbon source and operation mode on the aerobic denitrification process. The carbon sources, operation modes and C: N: P ratio of the four experiments were shown in Table 1. The influent qualities of the four experiments were shown in Table 2. The influent also contained MgSO4·7H2O and CaCl2 of 50.0 mg/L and 20.0 mg/L, respectively, while the trace elements were also dosed with detailed description as Cheong and Hansen (2008). All experiments were operated with a cyclic duration of 12 h. The operation procedures were shown in Fig. 1 and Table 3.
Ⅰ Ⅱ Ⅲ Ⅳ
CH3COONa NH4HCO3 K2HPO4 KH2PO4 CH3CH2CH2COONa NH4HCO3 K2HPO4 KH2PO4
PO43−-P
3. Results
2.2. Design of experiments
C:N:P ratios
Ⅰ and Ⅱ
NH4+eN
During the experiments, COD, NH4+eN, NO3−-N and NO2−-N of the samples were monitored according to standard methods (APHA, 1998). N2O-N and NO-N were monitored when the process was operated under completely oxic mode. Their concentration profiles in the reactor were continuously monitored using Clark-type microelectrodes (Arhus, Denmark). The microelectrode was calibrated with the twopoint method according to the instruction provided by Unisense (Arhus, Denmark). DO concentration and temperature in the reactor were measured by DO meter (HACH), while pH value was measured by pH meter (Rex, Shanghai). PHB contents in activated sludge were determined based on the gravimetric method which was modified from the method proposed by Hahn et al. (1994) and YüksekdaĞ et al. (2003). The detailed procedures were elucidated in Supplementary Materials. The microbial community structure was analyzed by highthroughput sequencing carried out by Sangon Biotech Co., Ltd. (Shanghai, China). The detailed information was shown in Supplementary Materials.
A lab-scale SBR reactor was made from transparent Plexiglas with a working volume of 10 L. The diameter and the height of the reactor were 200 mm and 400 mm, respectively. The reactor was equipped with a rectangular mixing paddle and an air compressor to supply air through a diffuser placed at the bottom of the reactor. The DO concentration in the reactor was maintained at ∼6.5 mg/L. The temperature of the reactor was controlled at 29 °C ± 1 °C by a water jacket. The sludge retention time was 25 days. Correspondingly, the average sludge concentrations of the reactor were 4207 mg/L and 3468 mg/L when the carbon sources were CH3COONa and CH3CH2CH2COONa, respectively. The drainage ratio was 0.5. Before operation, the reactor was seeded while the seeding sludge was taken from the oxic unit of the 4th wastewater treatment plant of Xi'an city, Shaanxi province, China.
Operational modes
COD
2.3. Analytical methods
2.1. The aerobic denitrification SBR reactor
Carbon source
Compounds
Ⅲ and Ⅳ
2. Material and methods
Experiment
Experiments
After the reactor reached steady state during the four experiments, variations of COD, NH4+eN, NO3−-N, NO2−-N, PO43−-P, TN and PHB contents of 4 typical cycles of each experiment were monitored (Fig. 3). Particularly, as the procedure for PHB determination was quite complex, the activated sludge samples for determining PHB content were less than the wastewater samples. Under A/O mode, TN concentrations in the effluents of the process were 6.71 ± 0.19 mg/L (n = 4) and 4.36 ± 0.2 mg/L (n = 4), 300
Journal of Environmental Management 239 (2019) 299–305
B. Hu, et al.
Fig. 1. The operation procedures of the four experiments.
Under the same operation mode, the TN removal efficiencies of the process and the maximum PHB contents of activated sludge were higher when the carbon source was CH3CH2CH2COONa. Furthermore, there was a distinct relationship between the PHB content and the TN removal efficiency of the process under the same operation mode. Under both the A/O mode and the completely oxic mode, the TN removal efficiency was proportional to the maximum PHB content in the activated sludge, i.e. the high maximum PHB content could lead to the high TN removal. Notably, under completely oxic mode, there were no obvious differences in TN removal efficiencies of the process under two carbon sources.
Table 3 Phases’ duration of the aerobic denitrification SBR process in four experiments. Experiment
Feeding
Mixing
Aeration and Mixing
Settling
Drawing
Resting
Ⅰand Ⅳ Ⅱ and Ⅲ
10 min 10 min
120 min 0 min
540 min 660 min
10 min 10 min
10 min 10 min
30 min 30 min
In experiment Ⅰ and Ⅳ, the duration of feeding, mixing, aeration and mixing, settling, drawing and resting were 10 min, 120 min, 540 min, 10 min, 10 min and 30 min, respectively. In experiment Ⅱ and experiment Ⅲ, mixing was cancelled. The duration of feeding, aeration and mixing, settling, drawing and resting were 10 min, 660 min, 10 min, 10 min and 30 min, respectively.
3.3. N2O and NO productions in aerobic denitrification process
respectively, and the TN removal efficiencies of the process were 86.11 ± 0.0043% (n = 4) and 91.30 ± 0.0036% (n = 4) when carbon sources were CH3COONa and CH3CH2CH2COONa, respectively. Correspondingly, the maximum PHB contents of activated sludge were 19.56% and 22.23% respectively. Under completely oxic mode, TN concentrations in the effluents of the process were 5.41 ± 0.76 mg/L (n = 4) and 5.26 ± 0.86 mg/L (n = 4), respectively, and the TN removal efficiencies of the process were 88.12 ± 0.017% (n = 4) and 89.30 ± 0.018% (n = 4) when carbon sources were CH3COONa and CH3CH2CH2COONa respectively. Correspondingly, the maximum PHB contents were declined to 9.10% and 14.08% respectively. The maximum PHB contents in activated sludge and TN removal efficiencies of the four experiments were listed in Table 5.
In order to examine NO and N2O production of aerobic denitrification process, NO and N2O concentrations in the reactor were continuously monitored when the process was operated under completely oxic mode (Fig. 4). When CH3COONa was the carbon source, the concentrations of NO and N2O in the bulk liquid of the typical cycles were ∼0.4 mg/L and ∼0.02 mg/L, respectively. However, when the carbon source was CH3CH2CH2COONa, the concentrations of N2O in the bulk of the typical cycles were ∼0.2 mg/L. The variation of NO was completely different from that when the carbon source was CH3COONa. The concentrations of NO were stable at the most of the cycle and decreased in the latter of the cycle. In the stable stage, the concentration of NO in the bulk liquid was ∼2.5 mg/L, and at the end of the cycle, it was decreased to
Fig. 2. Long-term performances of the aerobic denitrification SBR process. 301
Journal of Environmental Management 239 (2019) 299–305
B. Hu, et al.
Table 4 The long-term performances of the process under different conditions. Experiment
NH4+eN in effluent (mg/L)
NO2−-N in effluent (mg/L)
NO3−-N in effluent (mg/L)
PO43−-P in effluent (mg/L)
COD in effluent (mg/L)
TN in effluent (mg/L)
TN removal efficiency (%)
No. of Samples
I II III IV
0.31 0.31 0.31 0.27
0.07 0.07 0.69 0.07
5.73 4.98 5.37 4.57
8.94 ± 1.31 9.97 ± 1.26 10.19 ± 2.20 11.14 ± 4.17
24.52 27.20 24.67 24.70
6.11 5.36 5.79 4.91
86.11% 88.02% 88.05% 90.05%
n = 22 n = 21 n = 20 n = 21
± ± ± ±
0.00 0.00 0.00 0.10
± ± ± ±
0.02 0.00 0.00 0.00
± ± ± ±
0.75 0.70 0.82 1.31
∼1.0 mg/L. Generally, when the carbon source CH3CH2CH2COONa, more NO and N2O were produced.
± ± ± ±
2.98 1.84 4.53 2.65
± ± ± ±
0.75 0.70 0.81 1.32
± ± ± ±
1.29% 1.62% 1.93% 0.03%
mode, the abundance sequence of microorganisms with different functions was: COD degradation microorganism > Aerobic denitrifiers > EPS synthesis microorganism > Sulfur oxidation microorganism > PHB production microorganism. When the operation mode was shifted to completely oxic mode, the abundance sequence was: COD degradation microorganism > EPS synthesis microorganism > Aerobic denitrifiers > PHB production microorganism > SMP production microorganism. When carbon source was CH3CH2CH2COONa and operation mode was A/O mode, the abundance sequence of microorganisms with different functions was: COD degradation microorganism > PHB production microorganism > EPS synthesis microorganism > Aerobic denitrifiers > Filamentous bacteria. When the operation mode was shifted to completely oxic mode, the abundance sequence was: PHB production microorganism > Aerobic denitrifiers > COD degradation microorganism > EPS synthesis microorganism > Sulfur oxidation microorganism > Nitrifiers.
was
3.4. Microbial community shifts under different operation conditions The microbial communities of the four experiments and the function of each genus were shown in Table 1–Table 4 of Supplementary Materials. Genera in quite small abundances were classified to Other. Except for Unclassified and Other, the microorganisms in the system can be divided into five classes, which were responsible for degradation of organic matters, PHB production, nitrification, aerobic denitrification and the others respectively. The detailed information of microbial groups was elucidated in Supplementary Materials. In order to understand the microbial community shifts under different conditions, the variations of the ten preponderant phyla of the microbial communities under different conditions were shown in Table 5 of Supplementary Materials. The network diagram based on operational taxonomic units (OTUs) of four experiments was illustrated in Fig. 5, while the diversity indexes of four experiments were shown in Table 6. When carbon source was CH3COONa and operation mode was A/O
Fig. 3. Effects of carbon sources and operation mode on aerobic denitrification process. 302
Journal of Environmental Management 239 (2019) 299–305
B. Hu, et al.
Table 5 The maximum PHB contents in activated sludge during four experiments. Experiments
The maximum PHB content
Average TN removal efficiency
Operation mode
Carbon Source
I II III IV
19.56% 9.10% 14.08% 22.23%
86.11% 88.12% 89.30% 91.30%
A/O mode Completely Oxic mode Completely Oxic mode A/O mode
CH3COONa CH3COONa CH3CH2CH2COONa CH3CH2CH2COONa
(n = 4) (n = 4) (n = 4) (n = 4)
4. Discussion
2016). Under anaerobic condition, external carbon source is mainly used for PHB production (Pereira et al., 1996). In this study, the higher TN removal efficiency is achieved when the carbon source is CH3CH2CH2COONa. According to the published literature, Nap activity in the pure cultivation was increased with the increased extent of the reduced carbons source (Richardson and Ferguson, 1992; Sears et al., 1997; Ellington et al., 2003). Similar to pure cultivation, in mixture cultivation, the more reduced carbon source is beneficial for improving the TN removal efficiencies of the aerobic denitrification process.
4.1. TN removal efficiency of the aerobic denitrification SBR process The TN removal efficiencies of the aerobic denitrification process were higher than conventional N removal process, such as SBR, SBBR MBR, etc. (Chen et al., 2015). It has been reported that the C/N ratio required by aerobic denitrification is higher than conventional biological removal process, and the aerobic denitrification process tends to work efficiently when the C/N ratio was in the range of 5–10 (Ji et al., 2015). During the four experiments, the C/N ratio of the influent was 10, indicating that excessive organic carbons were supplied into aerobic denitrifiers. There seems a “bottleneck” in the electron transport chain of the aerobic denitrification process, thus, nitrate, nitrite or thiosulfate can all be used as the electron acceptor (Richardson and Ferguson, 1992). When excessive organic carbons are applied to the system, the number of electrons flowed into the electron transport chain are increased, which might stimulate the aerobic denitrification process and make it achieving a relatively high TN removal efficiency. In addition, an alternate anoxic and oxic operation mode is beneficial for aerobic denitrification process (Alzate Marin et al., 2016). As a result, the highest N removal efficiency is obtained under the A/O mode in this study.
4.3. Effects of carbon source on NO and N2O production in aerobic denitrificaiton process A number of studies have focused on N2O and NO emission of aerobic denitrificaiton process under the pure cultivation of aerobic denitrifiers, such as Pseudomonas stutzeri PCN-1 (Zheng et al., 2014), Chelatococcus daeguensis TAD1 (He et al., 2016), Providencia rettgeri YL (Ye et al., 2016). It has been generally recognized that aerobic denitrification can reduce the emission of N2O and NO (Zheng et al., 2014; He et al., 2016). In this study, NO and N2O productions are in relatively high levels. In addition, the NO emission in this study is higher than the N2O emission. Regarding N2O emission of aerobic denitrification process, it has been reported that N2O emission is related to the bacteria type and their oxygen tolerance of nitrous oxide reductase (Ye et al., 2016). Furthermore, according to Ellington et al. (2002), compared with CH3CH2CH2COONa, CH3COONa was more favorable to Paracoccus pantotrophus which can accelerate Nap during aerobic growth. The use of CH3CH2CH2COONa might lead to the microorganisms with a surfeit of reductant and therefore potentially a surfeit of adenosine triphosphate (ATP). An excessive ATP in microorganisms may result in a low concentration of adenosine diphosphate (ADP) and a constraint of the respiratory rate (Ellington et al., 2002). For microorganisms, this may lead to limitation of some biochemical reactions, and then, some intermediates related to these reactions may accumulate. In aerobic denitrification process, although the process achieves a relatively high TN
4.2. The role of PHB in the aerobic denitrification process PHB is a carbon-energy storage material, and can be served as electron donor for aerobic denitrification (Bernat and WojnowskaBaryła, 2007). Richardson et al. (2001) reported that reduced carbon sources could not be used directly by aerobic denitrifiers but were converted to PHB firstly. Compared with CH3COONa, CH3CH2CH2COONa is more reduced. In four experiments, under the same operation mode, the maximum PHB contents are higher when the carbon source is CH3CH2CH2COONa. Besides, the maximum PHB contents under A/O mode are higher than those under complete oxic mode. Anaerobic/aerobic switches are beneficial for PHB synthesis (Liu et al.,
Fig. 4. N2O and NO concentrations in the bulk of aerobic denitrification SBR process. 303
Journal of Environmental Management 239 (2019) 299–305
B. Hu, et al.
Fig. 5. Network diagram based on OTUs of the four experiments.
and the shear force on flocs is stronger than that of A/O mode. Ohtaekwangia sp. is important for preventing the disintegration of the flocs. When the carbon source is CH3CH2CH2COONa and the operation mode is completely oxic mode, Plasticicumulans sp. which has an amazing ability in PHB synthesis (Jiang et al., 2011) is dominant in activated sludge of the process. As mentioned before, PHB can be served as the carbon source for aerobic denitrification (Bernat and WojnowskaBaryła, 2007), and the Nap is more active when the carbon source is CH3CH2CH2COONa (Richardson and Ferguson, 1992; Ellington et al., 2003). Although the proportion of aerobic denitrifiers is smaller than that when the carbon source is CH3COONa, the TN removal efficiency of the process is not influenced as the high concentration of PHB in the activated sludge. When the carbon source is CH3CH2CH2COONa and the operation mode is converted to A/O mode, the proportion of aerobic denitrifiers is increased from 3.70% to 13.30%. A/O mode is more preferred by aerobic denitrifiers. Meanwhile, the proportion of PHB production microorganism is decreased from 43.68% to 37.83%. As alternative anaerobic and oxic environment is beneficial for the PHB synthesis (Liu et al., 2016), the maximum PHB content of activated sludge is increased from 14.08% to 22.23%. The high PHB content provided a powerful support for the experiment IV achieving the maximum TN removal efficiency. Notably, in four experiments, the proportion of nitrifiers is quite small. Not very long time ago, nitrification was considered to be conducted by autotrophs obligately. However, more and more studies have confirmed the existence of heterotrophic nitrification and aerobic
Table 6 Diversity indexes of microorganisms in the four experiments. Experiment
Shannon Index
Chao Index
Coverage
Simpson Index
Ⅰ Ⅱ Ⅲ Ⅳ
5.6668 4.5619 3.6665 3.9825
39706.7 28627.4 16175.7 20321.3
0.9092 0.9616 0.9580 0.9457
0.0133 0.0314 0.1440 0.1175
nitrogen removal efficiency when the carbon source is CH3CH2CH2COONa, a certain amount of nitrogen is removed via NO and N2O. As the negative effects of NO and N2O on environment, it is fair to say that CH3COONa is a better carbon source for aerobic denitrification, albeit CH3CH2CH2COONa can improve Nap activity and the TN removal efficiency of the process.
4.4. Microbial community shifts under different conditions When the carbon source is CH3COONa and the operation mode is converted from A/O mode to completely oxic mode, the abundance of aerobic denitrifiers and EPS synthesis microorganism (Ohtaekwangia sp.) are increased obviously as well as the aerobic denitrifiers. Ohtaekwangia sp. can synthesize extracellular polymetric substances (EPS) and accumulate phosphorous (Świątczak and CydzikKwiatkowska, 2018). EPS is responsible for the floc formation by agglomerating activated sludge (Sponza, 2003) and stabilizing aerobic granule structure (Świątczak and Cydzik-Kwiatkowska, 2018). In completely oxic mode, DO concentration is maintained at ∼6.5 mg/L 304
Journal of Environmental Management 239 (2019) 299–305
B. Hu, et al.
denitrification, such as Acinetobacter sp. HA2 (Yao et al., 2013), Providencia rettgeri strain YL (Ye et al., 2016), etc. Consequently, the nitrification of the process is not influenced as the existence of heterotrophic nitrifiers in microbial communities. When the carbon source is CH3CH2CH2COONa, the abundance of the phylum Proteobacteria is higher than those when the carbon source is CH3COONa. From the phylum level, the numbers of the microorganisms which abundance are more than 1.0% are 8, 6, 4 and 3 respectively in four experiments. When the carbon source is CH3COONa, the bacteria diversity of the reactor is richer than that when the carbon source is CH3CH2CH2COONa. The effect of carbon source on the microbial community is more obvious and direct.
Hahn, S.K., Chang, Y.K., Kim, B.S., Chang, H.N., 1994. Optimization of microbial poly(3hydroxybutyrate) recovery using dispersions of sodium hypochlorite solution and chloroform. Biotechnol. Bioeng. 44, 256–261. He, J.X., Zhou, S.F., Huang, S.B., Zhang, Y.Q., 2016. Pretreated corn hush hydrolysate as the carbon source for aerobic denitrification with low level of N2O emission by thermophilic Chelatococcus daeguensis TAD1. Water, Air, Soil Pollut. 227, 314. Huang, H.K., Tseng, S.K., 2001. Nitrate reduction by Citrobacter diversus under aerobic environment. Appl. Microbiol. Biotechnol. 55, 90–94. Ji, B., Yang, K., Zhu, L., Jiang, Y., Wang, H.Y., Zhou, J., Zhang, H.N., 2015. Aerobic denitrification: a review of important advances of the last 30 years. Biotechnol. Bioproc. Eng. 20, 643–651. Jiang, Y., Marang, L., Kleerebezem, R., Muyzer, G., van Loosdtrecht, M.C.M., 2011. Polyhydroxybutyrate production from lactate using a mixed microbial culture. Biotechnol. Bioeng. 108, 2022–2035. Lee, N.M., Welander, T., 1996. The effect of different carbon sources on respiratory denitrification in biological wastewater treatment. J. Ferment. Bioeng. 82 (3), 277–285. Liu, C.L., Liu, D., Qi, Y.J., Zhang, Y., Liu, X., Zhao, M., 2016. The effect of anaerobic–aerobic and feast–famine cultivation pattern on bacterial diversity during poly-βhydroxybutyrate production from domestic sewage sludge. Environ. Sci. Pollut. Res. 23 (13), 12966–12975. Majone, M., Beccari, M., Dionisi, D., Levantesi, C., Renzi, V., 2001. Role of storage phenomena on removal of different substrates during pre-denitrification. Water Sci. Technol. 43, 151–158. Pereira, H., Lemos, P.C., Reis, M.A.M., Crespo, J.P.S.G., Carrondo, M.J.T., Santos, H., 1996. Model for carbon metabolism in biological phosphorous removal processes based on in vivo 13C-NMR labelling experiments. Water Res. 30, 2128–2138. Richardson, D.J., Ferguson, S.J., 1992. The influence of carbon substrate on the activity of the periplasmic nitrate reductase in aerobically grown Thiosphaera pantotropha. Arch. Microbiol. 157, 535–537. Richardson, D.J., Berks, B.C., Russell, D.A., Spiro, S., Taylor, C.J., 2001. Functional, biochemical and genetic diversity of prokaryotic nitrate reductases. Cell. Mol. Life Sci. 58, 165–178. Robertson, L.A., Kuenen, J.G., 1984a. Aerobic denitrification: a controversy revived. Arch. Microbiol. 139, 351–354. Robertson, L.A., Kuenen, J.G., 1984b. Aerobic denitrification – old wine in new bottles? Antonie Leeuwenhoek 50, 525–544. Robertson, L.A., van Neil, E.W.J., Torremans, R.A.M., Kuenen, J.G., 1988. Simultaneous nitrification and denitrification in aerobic chemostat cultures of Thiosphaera pantotropha. Appl. Environ. Microbiol 54, 2812–2818. Sears, H.J., Little, P.J., Richardson, D.J., Berks, B.C., Spiro, S., Ferguson, S.J., 1997. Identification of an assimilatory nitrate reductase in mutants of Paracoccus denitrificans GB17 deficient in nitrate respiration. Arch. Microbiol. 167, 61–66. Sponza, D.T., 2003. Investigation of extracellular polymer substances (EPS) and physicochemical properties of different activated sludge flocs under steady-state conditions. Enzym. Microb. Technol. 32, 375–385. Świątczak, P., Cydzik-Kwiatkowska, A., 2018. Performance and microbial characteristics of biomass in a full-scale aerobic granular sludge wastewater treatment plant. Environ. Sci. Pollut. Res. 25, 1655–1669. Wan, C.L., Yang, X., Lee, D.J., Du, M.A., Wan, F., Chen, C., 2011. Aerobic denitrification by novel isolated strain using NO2--N as nitrogen source. Bioresour. Technol. 102 (15), 7244–7248. Wang, Y.Y., Zhang, Z.J., Qiu, L., Guo, Y., Wang, X.J., Xiong, X.J., Chen, S.H., 2015. Effect of temperature downshifts on biological nitrogen removal and community structure of a lab-scale aerobic denitrification process. Biochem. Eng. J. 101, 200–208. Wang, H.Y., Zou, Z.C., Chen, D., Yang, K., 2016. Effects of temperature on aerobic denitrification in a bio-ceramsite reactor. Energy Sources, Part A 38, 3236–3241. Watahiki, M., Hata, S., Aida, T., 1983. N2O accumulation and inhibition of N2O reduction by denitrifying Pseudomonas sp. 220A in the presence of oxygen. Agric. Biol. Chem. 47, 1991–1996. Yao, S., Ni, J.R., Ma, T., Li, C., 2013. Heterotrophic nitrification and aerobic denitrification at low temperature by a newly isolated bacterium, Acinetobacter sp. HA2. Bioresour. Technol. 139, 80–86. Ye, J., Zhao, B., An, Q., Huang, Y.S., 2016. Nitrogen removal by Providencia rettgeri strain YL with heterotrophic nitrification and aerobic denitrification. Environ. Technol 37, 2206–2213. YüksekdaĞ, Z.N., Beyatli, Y., Aslim, B., 2003. Determination of poly-β-hydroxybutyrate (PHB) production by some mesophilic and thermophilic lactic acid bacteria. Turk. J. Biol. 27, 37–42. Zheng, H.Y., Liu, Y., Sun, G.D., Gao, X.Y., Zhang, Q.L., Liu, Z.P., 2011. Denitrification characteristics of a marine origin psychrophilic aerobic denitrifying bacterium. J. Environ. Sci. 23, 1888–1893. Zheng, M.S., He, D., Ma, T., Chen, Q., Liu, S.T., Ahamd, M., Gui, M.Y., Ni, J.R., 2014. Reducing NO and N2O emission during aerobic denitrification by newly isolated Pseudomonas stutzeri PCN-1. Bioresour. Technol. 162, 80–88.
5. Conclusions (1) PHB was important for aerobic denitrification process achieving high TN removal efficiency. A/O mode was beneficial for PHB synthesis and aerobic denitrifiers accumulation. (2) For the aerobic denitrification SBR process tested, CH3CH2CH2COONa was helpful for improving the TN removal efficiency of the process, however, CH3COONa was a better carbon source instead of CH3CH2CH2COONa due to the lower production of NO and N2O. (3) Compared with the operation mode, the effects of carbon source on the aerobic denitrification process and microbial community were more obvious. Acknowledgements This work was supported by National Natural Science Foundation of China (Grant No.:51778057); Science and Technology Project of Shaanxi Province, China (Grant No.: 2018JQ5143); National Training Programs of Innovation and Entrepreneurship for Undergraduates, Chang'an University, China (Grant No.: 201810710253). Appendix A. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.jenvman.2019.03.063. References Alzate Marin, J.C., Caravelli, A.H., Zaritzky, N.E., 2016. Nitrification and aerobic denitrification in anoxic-aerobic sequencing batch reactor. Bioresour. Technol. 200, 380–387. APHA, 1998. Standard Methods for the Examination of Water and Wastewater, twentieth ed. American Public Health Association, Washington, DC, USA. Bernat, K., Wojnowska-Baryła, I., 2007. Carbon source in aerobic denitrification. Biochem. Eng. J. 36, 116–122. Chen, Q., Ni, J.R., Ma, T., Liu, T., Zheng, M.S., 2015. Bioaugmentation treatment of municipal wastewater with heterotrophic-aerobic nitrogen removal bacteria in a pilot-scale SBR. Bioresour. Technol. 183, 25–32. Cheong, D.Y., Hansen, C.L., 2008. Effect of feeding strategy on the stability of anaerobic sequencing batch reactor responses to organic loading conditions. Bioresour. Technol. 99, 5058–5068. Ellington, M.J.K., Bhakoo, K.K., Sawers, G., Richardson, D.J., Ferguson, S.J., 2002. Hierarchy of carbon source selection in Paracoccus pantotrophus: strict correlation between reduction state of the carbon substrate and aerobic expression of the nap operon. J. Bacteriol. 184, 4767–4774. Ellington, M.J.K., Sawers, G., Sears, H.J., Spiro, S., Richardson, D.J., Ferguson, S.J., 2003. Characterization of the expression and activity of the periplasmic nitrate reductase of Paracoccus pantotrophus in chemostat cultures. Microbiology 149, 1533–1540.
305