- Email: [email protected]
Effects of step-feed on granulation processes and nitrogen removal performances of partial nitrifying granules
Bioresource Technology 123 (2012) 375–381
Contents lists available at SciVerse ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Effects of step-feed on granulation processes and nitrogen removal performances of partial nitrifying granules Xin-Hua Wang, Li-Xiu Jiang, Yi-Jing Shi, Ming-Ming Gao, Sen Yang, Shu-Guang Wang ⇑ Shandong Provincial Key Laboratory of Water Pollution Control and Resource Reuse, School of Environmental Science and Engineering, Shandong University, 27 Shanda Nanlu, Jinan 250100, China
h i g h l i g h t s " PNG was successfully cultivated under step-feed mode. " Step-feed delayed the granulation processes of PNG. " PNG cultivated in both reactors with or without step-feed had similar properties. " Step-feed enhanced denitrification and TN removal, as well as ammonia oxidation.
a r t i c l e
i n f o
Article history: Received 12 June 2012 Received in revised form 18 July 2012 Accepted 22 July 2012 Available online 27 July 2012 Keywords: Aerobic granules Partial nitrification Step-feed Granulation processes Denitrification
a b s t r a c t Two anoxic/oxic sequencing batch reactors (A/O SBRs) were operated to investigate the effects of stepfeed on granulation processes and performances of partial nitrifying granules (PNG). R1 was operated in a traditional single-feed mode, while a two-step-feed strategy was used in R2. Results showed that R1 had a faster granulation process and better performance in maintaining partial nitrification compared with R2, indicating that the step-feed mode had a negative effect of on formation of PNG. However, after full granulation, PNG in both reactors had similar properties in terms of suspended solids (MLSS), sludge volume index (SVI) and granule size. Moreover, mature granules in R2 had a higher nitrite accumulation rate than that in R1. Step-feed strategy was also observed to enhance denitrification and TN removal, as well as ammonia oxidation. It can be concluded that step-feed was unfavorable for cultivating PNG, but it significantly improved the nitrogen removal performance of PNG. Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction Partial nitrifying activated sludge (PNAS) processes have attracted the most attention in recent years, owing to their advantages with respect to conventional nitrate pathway: reduce the energy consumption for aeration and organic carbon requirement for denitrification (Peng and Zhu, 2006; Ruiz et al., 2003). Several factors have been identified to selectively inhibit or washout nitrite oxidizing bacteria (NOB) over ammonia oxidizing bacteria (AOB), mainly including dissolved oxygen (DO), temperature, sludge retention time (SRT), free ammonia (FA) and free nitrous acid (FNA) (Peng and Zhu, 2006; Ruiz et al., 2003). DO is one of the main factors inhibiting NOB, but long-term operation at low DO would result in low nitritation rate, sludge bulking or the increasing N2O production (Chuang et al., 2007; Zeng et al., 2009). In addition, ⇑ Corresponding author. Tel.: +86 531 88362220; fax: +86 531 88364513. E-mail addresses: [email protected] (X.-H. Wang), [email protected] (S.-G. Wang). 0960-8524/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biortech.2012.07.080
the capacity of partial nitrification is limited due to lower biomass concentration in activated sludge (AS) systems without effective biomass retention (Jubany et al., 2009; Mosquera-Corral et al., 2005b). Granules based systems have been proposed as a suitable alternative to obtain stable and robust partial nitrification (Gao et al., 2011; Lopez-Palau et al., 2011; Vázquez-Padín et al., 2010). Partial nitrifying granules (PNG) present high cell density and much fast settling velocities than AS, thus ensuring higher biomass concentration and specific activities and consequent higher maximum load (Bartroli et al., 2010; Gao et al., 2011; Lopez-Palau et al., 2011). Moreover, DO diffusion limitation within granules allow operating reactors reaching stable partial nitrification without strict DO and temperature controls (Shi et al., 2011; Vázquez-Padín et al., 2010). Granules also have special layered microbial structure with more denitrifying bacteria existed inside the granules, giving them a great potential for better denitrification (Mosquera-Corral et al., 2005a; Tay et al., 2002). However, during PNG processes, denitrifi-
376
X.-H. Wang et al. / Bioresource Technology 123 (2012) 375–381
cation is normally restricted by deficient carbon source supply in conventional oxic or anoxic/oxic (A/O) sequencing batch reactors (SBR), especially when treating high strength nitrogen and low C/N ratio wastewater (Shi et al., 2009; Shi et al., 2011). The use of an external carbon source will improve N-removal, but also undermines the overall benefits of the nitrite pathway. Step-feed mode for A/O SBR has been reported to be effective in making good use of carbon source in influent and increasing denitrification rate and further total nitrogen (TN) removal efficiency. In addition, it benefits the growth of AOB and inhibits NOB, accelerates nitrite accumulation and nitritation rate for PNAS systems (Lemaire et al., 2008; Yang et al., 2007). Compared with conventional single-feed SBR, step-feed mode changes some factors which may affect granule formation and/or nitrite accumulation, such as smaller substrate load at the beginning of each react phase and shorter but more frequent feast-famine conditions. However, the feasibility of PNG formation under step-feed mode and the comparison with traditional single-feed mode were still sparse. Thus, this study used two A/O SBRs with and without step-feed to cultivate PNG. The main objective was to find the effects of feeding mode on granulation processes and nitrogen removal performances. It is expected that the results derived from this study will provide useful information or guides for practical applications of PNG in treating high strength nitrogen and low C/N wastewater.
Synthetic wastewater mainly consist of glucose, sodium acetate, ammonium chloride, sodium bicarbonate and other necessary mineral-salts medium. The influent COD and NHþ 4 N were fixed at 800 (50% contribution each of glucose and sodium acetate) and 300 mg/L, respectively. 2.2. Analysis
2. Methods
Suspended solids (MLSS), sludge volume index (SVI), ammonium, nitrate and nitrite were analyzed by standard methods (APHA, 1998). Influent samples were taken directly from the feed tanks, while effluent samples were pretreated by passing through a 0.45 lm filter. Grab samples of mixed liquor were taken from equidistant sampling ports along the height of each reactor and gently mixed prior to the measurement of various parameters and observation of microbial compositions. After mixing, the samples were completely blended for the MLSS and SVI tests, or were filtered through 0.45 lm filters for the measurement of ammonium, nitrate and nitrite. The mean granule size was measured by a laser particle size analysis system (Mastersizer 2000, Malvern Instruments, UK). DO levels were measured with a DO meter (YSI Model 85, USA). The microbial compositions of granules were observed qualitatively with scanning electron microscope (SEM) (HITACHI S-570, Japan). The granule samples were fixed with 3.0% glutaraldehyde in 0.1 M phosphate buffer at pH 7.2. The samples were then dehydrated with ethanol, silver-coated by a sputter and observed in the SEM.
2.1. Reactor set-up and operation
2.3. Calculations
Two identical SBRs (R1 and R2, 10 cm in diameter and 35 cm in height) with working volumes of 2.5 L were used for formation of PNG. The inoculated PNAS exhibited an excellent partial nitrification performance with nitrite accumulation rate higher than 85%. Both reactors were operated at 28 °C and without pH control between 7.5 and 8.5. Sampling ports were built along the height of each reactor at 10 cm intervals. Effluent was discharged from the middle port of the reactor with a volumetric exchange ratio of 50%. The R1 was operated in a traditional A/O cycles, consisting of influent feeding, anoxic phase (mixing), oxic phase (aeration), settling and effluent discharge. While the R2 cycles comprised a two-step-feed strategy, with 50% influent introduced in the first feed and the remaining 50% in the second feed. After each feed a sequence of anoxic and oxic phases were applied. Following the two sequences the cycle ended with settling and discharge phases. Tables 1 and 2 summarizes the operational conditions of SBRs and the different time distribution in the cycles. The time period of each phase and the cycle were modified according to nitrogen removal performance based on analytical data evaluation. The settling time was progressively decreased to improve granulation. In order to facilitate nitrite accumulation, the reactors were initially aerated at a superficial upflow air velocity (SUAV) of 0.2 cm/s, which corresponding to DO concentrations between 2.8 and 4.1 mg/L during oxic phases. At the start of period IV, the aeration rate was raised to 1.0 cm/s in terms of SUAV, giving a higher DO concentrations between 6.0 and 7.5 mg/L.
The ammonia oxidation rate was calculated from the slopes of ammonia consumption from cycle studies. The nitrite accumulation rate (NAR) was calculated using the following equation:
NAR ð%Þ ¼
NO2
NO N N þ NO3 N
ð1Þ
The FA concentration in reactor can be estimated by the expression proposed by (Anthonisen et al., 1976):
FA ðmg=LÞ ¼
17 ½NHþ4 N 10pH 14 exp½6334=ð273 þ TÞ þ 10pH
ð2Þ
where ½NHþ 4 N is the ammonia–nitrogen concentration and T is the temperature in °C. 3. Results and discussion 3.1. Granulation processes 3.1.1. Formation of granules In the two reactors the time needed for the granules development was different. In R1 operated in single-feed A/O SBR mode, initial tiny granules appeared on day 33 (period II). Then granules grew and proliferated quickly in the reactor (Fig. S1). Full granulation achieved on day 56 (period III) when round-shaped granules
Table 1 Operational conditions of SBRs during the experiments. Period
Days
Cycle time (h)
SUAV (cm/s)
DO in oxic phase (mg/L)
Settling time (min)
HRT (h)
I II III IV V
1–28 29–44 45–63 64–90 91–128
6 6 4 4 4
0.2 0.2 0.2 1.0 1.0
2.7–4.2 2.7–4.2 2.7–4.2 6.0–7.5 6.0–7.5
8 4 2 2 1
12 12 8 8 8
377
X.-H. Wang et al. / Bioresource Technology 123 (2012) 375–381 Table 2 Different time distribution in the cycle SBRs.
R1 R2
Cycle time (min)
Feed (min)
Anoxic (min)
Oxic (min)
Feed (min)
Anoxic (min)
Oxic (min)
Settling (min)
Discharge (min)
6 4 6 4
12 12 6 6
84 84 42 42
252/256 138/139 126/128 78/79
– – 6 6
– – 42 42
– – 126/128 60/60
8/4 2/1 8/4 2/1
4 4 4 4
had became the dominant form of biomass. As shown from the SEM images, granules in R1 had a compact and dense structure and a bacilli- and cocci-dominant outer surface (Fig. S2). Initial and full granulation time in the step-feed R2 were both obviously longer (Fig. S1). The biomass in R2 always had a floc-like appearance during period I and II. After the settling time was reduced to 2 min in period III, granules were firstly observed on day 50. At the beginning of period IV, a higher SUAV was supplied in the reactor to enhance granulation. However, a lot of filamentous flocs were still present and coexisted with granules in this period. After the settling time was further reduced to 1 min during period V, filamentous flocs were washed out successively and full granulation was observed on day 109. Granules in R2 had a similar microstructure with those in R1, but a certain amount of filamentous bacteria were found to present (Fig. S2). While the granules were successfully cultivated in R2, the step-feed strategy significantly delayed the sludge granulation process. Aerobic granulation is a gradual process from fluffy seed flocs to dense granules often occurring in SBR. It has been reported that granulation is influenced by a variety of factors, mainly including seed sludge, substrate composition and load, and SBR operation (Lee et al., 2010). Microorganisms in the single-feed R1 are subjected to a periodic feast and famine regime. Because the substrate was equally divided into two streams and introduced two times in R2, the reactor operation in one cycle includes two periodic feast and famine. And also, R2 creates a smaller substrate concentration at the beginning of each feast phase as compared with R1. According to the kinetic selection theory, at low substrate concentration, filamentous organisms would achieve a high substrate removal rate and its outgrowth would be favored (Chudoba, 1985). Because of the proliferation of filamentous bacteria, circumstances are disadvantageous for the formation of granules with regular structures. This may explain why the granulation process in the stepfeed SBR was delayed. McSwain et al. (2004) also investigated the role of feeding strategy for aerobic granulation in three SBRs applied with different feeding times. It was found that pulse feeding which creates a high feast condition was favorable for the formation of aerobic granules, because it affects the selection and growth of filamentous organisms. This is consistent with the results in our study. It should be pointed out that both initial granulation in R1 and R2 were realized under a low shear force in terms of SUAV of 0.2 cm/s. Much research has asserted that high shear forces are necessary for the formation of aerobic granules and the minimum SUAV should not be lower than 1.2 cm/s (Adav et al., 2007; Tay et al., 2001). However, granule formation are increasingly reported under lower shear forces. Zhang et al. (2011) reported that dense and regular-shaped aerobic granules were successfully formed at SUAV of 0.6 cm/s by treating low strength synthetic wastewater in A/O/A SBR during the first 50 days operation. Thus, high shear force do not appear to be a crucial factor for granulation, but may be more important in compacting granule structure and eroding excess cells and attached materials from an already formed granule (Lee et al., 2010). The formation of initial granules in both reactors in this study was partly attributed to the high sludge concentration at the beginning of the operation and A/O mode for the SBRs. On one
hand, as proposed by Su et al. (2012), high sludge concentration causes more frequent collision and stronger friction among microorganisms, resulting in the improvement of microbial autoaggregation. On the other hand, because of degradation of most organic matter by denitrifying bacteria at anoxic period in A/O SBRs, the microorganisms in the reactors would encounter with longer starvation state, which may further improve aerobic granulation. 3.1.2. Properties of biomass The initial seed sludge with a mean floc size of 0.12 mm and a SVI value of 40.5 mL/g showed a fluffy, irregular and loosestructured morphology. According to the observed trend in MLSS, SVI and size of biomass (Fig. 1), both granulation processes in the two reactors could be divided into three typical phases: acclimation phase, multiplication phase followed by maturation phase. In the acclimation phase of R1 from day 1 to day 32, the MLSS and SVI both decreased and the size of biomass increased. The start of the multiplication phase in R1 was marked by the appearance of initial tiny granules on day 33 within the reactor. The MLSS and mean diameter of granules increased substantially to 22.6 g/L and 0.87 mm on day 81, respectively. The appearance of granules was also accompanied by gradual improvement of settleability. The SVI decreased to 12.0 mL/g during the multiplication phase. The following maturation phase lasted from day 82 to the end of R1 operation, which was characterized by a little fluctuation or no change in the values of MLSS (23.3 ± 1.2 g/L), SVI (11.6 ± 1.4 mL/g) and mean diameter (0.91 ± 0.04 mm). There were no much difference between R1 and R2 in terms of MLSS, SVI and size of biomass during period I and II. The acclimation phase in R2 lasted until day 50 when granules were firstly observed. During the multiplication phase lasted from day 51 to day 114, granules increasingly formed within the reactor, resulted in a continual increase of MLSS and mean diameter as well as a gradual improvement of settleability. At the beginning of period V, washout of filamentous flocs caused a sudden decrease in MLSS. During the maturation phase from day 115 to the end of reactor operation, the MLSS and SVI stabilized at 26.8 ± 0.9 g/L and 10.6 ± 1.3 mL/g, while the granule size increased to plateau at 1.10 mm. The step-feed strategy delayed the granulation process, which would inevitably retard the increase of biomass concentration and granule size as well as the improvement of settleability during the acclimation phase and multiplication phase (Fig. 1). However, the step-feed strategy did not impair the physicochemical properties of mature granules. The SVI of the mature granules in both reactors were in the same range of 10–13 mL/g, while the MLSS and mean granule size in the step-feed R2 were slightly higher than that in R1. The SVI values are very close to that reported by Qin and Liu (2006) whom cultivated granules with SVI of 12 mL/g under alternating O/A conditions and nitrogen loading rate of 0.45 kg/m3d, but are much lower than those granules cultivated under absolute aerobic conditions with a typical value of 50– 100 mL/g (Moy et al., 2002; Tay et al., 2001). These seem to indicate that the imposed anoxic denitrification condition plays a positive role on sludge properties, because it would enhance heterotrophic growth/storage deeper in the internal anoxic layer of granules and encourage aggregate densification (Wan et al., 2009).
378
X.-H. Wang et al. / Bioresource Technology 123 (2012) 375–381
Fig. 1. Time profiles of MLSS (a), SVI (b) and mean diameter (c) in R1 (j) and R2 (s).
3.2. Reactor performance 3.2.1. Nitrite accumulation performance R1 exhibited a good performance in maintaining nitrite accumulation during the entire operation period, as shown by the low effluent nitrate concentration (20.2 ± 8.7 mg/L) and high NAR (87 ± 5%) in Figs. 2 and 3. Sudden increase of DO concentration in oxic phase caused by raising of aeration rate since period IV did not deteriorate the partial nitrification performance, which would be attributed to the full granulation in the reactor and the oxygen transfer limitation within granules (Shi et al., 2011; Vázquez-Padín et al., 2010). In contrast, a gradual increase in effluent nitrate concentration was observed in R2 during period I and II, which certainly resulted in a decrease in NAR. To further study the operation status of the reactors, typical cycle measurements were conducted on day 40. As shown in Fig. 4a and c, nitrite oxidation was always lag behind ammonia oxidation in both reactors. But in the cycle of 6 h, when ammonia oxidation was completed, it still had a considerable time for the remaining nitrite to be converted to nitrate by NOB in oxic phases. This phenomenon was especially obvious in the step-feed SBR, probably because of the lower FA inhibition on NOB in the reactor. NOB has been described to be much more sensitive to FA than AOB. Anthonisen et al. (1976) reported that AOB and NOB were inhibited at 10–150 mg/L and 0.1–1.0 mg/L of FA. Vadivelu et al. (2007) found Nitrobacter likely ceased to grow at an FA concentration of 6–9 mg/L. The SBRs in this study were operated at 28 °C temperature and pH was between 7.5 and 8.5 during experimental studies. Under step-feed mode, low concentrations of ammonia were introduced into the reactor before every anoxic phases. Thus, the calculated concentrations of FA were considerable lower in R2 (0.1–6.9 mg/L) than R1 (0.1–19.0 mg/L) (Fig. 4). The maximum FA concentration in R1 was far above the threshold concentration of
Fig. 2. Reactors operation in terms of NHþ 4 N concentration in the feeding (j), and NHþ 4 N (N), NO2 –N (s) and NO3 –N () concentrations in the effluent.
inhibition at which the inhibition of NOB began, which led to a comparative stable nitrite accumulation. When the cycle time was reduced to 4 h since period III (mainly by shortened the duration of oxic phases as shown in Table 2), the
X.-H. Wang et al. / Bioresource Technology 123 (2012) 375–381
379
nitrite accumulation. This conclusion was also demonstrated by previous studies (Lemaire et al., 2008; Yang et al., 2007). As shown from the cycle measurements under steady states in both reactors (day 122) in Fig. 3b and d, nitrite accumulated during oxic phase in R2 was always lower than that in R1 and was almost denitrified in time, which reduced the substrate of nitratation and further prevented the toxicity of nitrite. In addition, a bigger granule size reached in the later reactor operation (Fig. 1c) would enhance the oxygen transfer limitation and further accelerate nitrite accumulation.
Fig. 3. Nitrite accumulation rate in R1 (j) and R2 (s) along the operational period.
nitrite accumulation in R2 was quickly recovered with NAR increased from below 20% to above 85%. Meanwhile, a slight enhancement of nitrite accumulation was also observed in R1. In period IV, a transient deterioration of nitrite accumulation performance occurred due to the sudden increase of DO concentration (6.0–7.5 mg/L) in oxic phase, during when flocs were still dominated in R2. With granules increasingly formed and grew bigger, the NAR increased and finally stabilized at 95 ± 3% with low concentrations of effluent nitrate (6.1 ± 4.6 mg/L) (Figs. 2 and 3). This phenomenon revealed the advantages for granules in maintaining partial nitrification and demonstrated the crucial role of oxygen transfer limitation for reaching stable partial nitrification in granule based reactors (Shi et al., 2011; Vázquez-Padín et al., 2010). As compared with R1, a better nitrite accumulation performance was achieved in R2 after sludge full granulated, which means that the step-feed operational mode was favorable for the
3.2.2. TN removal performance Step-feed strategy for R2 was also observed to enhance denitrification and TN removal. As shown from Fig. 5, the TN removal efficiency in R2 was obviously higher than that in R1 from period III to the end of operation. The TN removal efficiency in R1 was basically fluctuated between 40% and 45%, except for period III when ammonia oxidation was seriously deteriorated. Guo et al. (2008) have reported that TN removal efficiency would increase with increasing of feeding steps. For R2 operated in two-step-feed mode, the value reached above 70% during the maturation phase of PNG. Step-feed strategy for SBR has been intensively regarded as an efficient nitrogen removal system, especially for the wastewater with lower C/N ratios. The step-feed mode greatly improved the availability of influent organics for denitrification in the SBR operated alternating A/O sequences, thereby elevating the TN removal capacity. In period I and II, nitrate was increasingly produced in R2 and this would raise the demand of carbon source for denitrification. The stoichiometric ratio of the heterotrophic denitrification is 1.71 mg COD/mg NO2–N and 2.85 mg COD/mg NO3–N. These values must be corrected because some of the carbon is used for biomass synthesis. With glucose as the carbonaceous substrate,
(a)
(b)
(c)
(d)
Fig. 4. Typical cycle measurement of NHþ 4 N (N), NO2 –N (s), NO3 –N (), TN (j), and FA (w) in R1 on day 40 (a) and 122 (b) and R2 on day 40 (c) and 122 (d).
380
X.-H. Wang et al. / Bioresource Technology 123 (2012) 375–381
removal after granules formed, indicated by the significantly improved TN removal efficiency and ammonia oxidation rate. Acknowledgements This work was supported by the National Natural Science Foundation of China (51108250), the Specialized Research Fund for the Doctoral Program of Higher Education of China (20110131120015), and the Excellent Young Scientist Foundation of Shandong Province (BS2010HZ009). Appendix A. Supplementary data Fig. 5. TN removal efficiency in R1 (j) and R2 (s) along the operational period.
the theoretical value obtained from the stoichiometric denitrification equation is 4.9 mg COD/mg NO3–N (Mateˇju˚ et al., 1992). While with acetate as a sole carbon source, the value was reported to be 3.75 mg COD/mg NO3–N (Thauer et al., 1977). Thus, a decrease of TN removal efficiency in period I and II of R2 was probably caused by the deficiency of carbon source for denitrification, because the influent COD/N (800/300) was lower in this study. 3.2.3. Ammonia removal performance Both reactors exhibited an excellent and similar ammonia removal performance in most times, as revealed by the low ammonia–nitrogen concentration in the effluent (below 5.0 mg/L as shown in Fig. 2) and high removal efficiency above 98%. A transitory deterioration in ammonia removal performance in period III was mainly resulted from the shorten of oxic phases duration and the low biomass concentrations in both reactors (Fig. 1). Similarly, a slight increase of effluent ammonia–nitrogen concentration from day 90 in R2 was also caused by the decrease of MLSS concentration. According to the comparison of the ammonia oxidation rate calculated from cycle studies, the step-feed operational mode was considered to accelerate AOB activity. The respective ammonia oxidation rate in R1 and R2 were 7.38 ± 0.42 and 11.40 ± 1.25 mg/g SSh on day 40, and 3.42 ± 0.30 and 4.41 ± 0.52 mg/gSSh on day 122. A decrease in specific biomass activity of both reactors on day 122 are supposed to be related to their bigger biomass size. It is generally accepted that granules with a smaller size are more efficient in terms of mass transfer and substrate conversion, and therefore more effective in degrading substrate (Liu and Tay, 2004). The ammonia oxidation rate in R2 was 54% and 29% higher than that in R1 on day 40 and 122, respectively. Lower ammonia and organic loading in the oxic phase could be responsible for the enhancement of nitritation activity in the granules. Firstly, lower ammonia concentration in R2 conduced to lower FA inhibition on AOB activity, as mentioned previously. Secondly, due to the existence of oxygen and space competition between AOB and heterotrophs within granules, nitritation occurred at a lower organic loading would have a higher rate (Ni et al., 2008; Xavier et al., 2007). 4. Conclusions PNG with similar physicochemical properties could be cultivated in both reactors with or without step-feed. However, the step-feed strategy markedly delayed the granulation processes and caused unstable partial nitrification during the formation of granules, probably owing to the smaller substrate concentration at the beginning of each feast phase. Though the step-feed was unfavorable for cultivating PNG, it had a positive effect on nitrogen
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.biortech.2012. 07.080. References Adav, S.S., Lee, D.J., Lai, J., 2007. Effects of aeration intensity on formation of phenolfed aerobic granules and extracellular polymeric substances. Applied Microbiology and Biotechnology 77, 175–182. Anthonisen, A.C., Loehr, R.C., Prakasam, T.B.S., Srinarh, E.G., 1976. Inhibition of nitrification by ammonia and nitrous acid. Journal of the Water Pollution Control Federation 48, 835–852. APHA, 1998. Standard Methods for the Examination of Water and Wastewater. 20th ed. American Public Health Association, Washington, DC. Bartroli, A., Perez, J., Carrera, J., 2010. Applying ratio control in a continuous granular reactor to achieve full nitritation under stable operating conditions. Environmental Science and Technology 44, 8930–8935. Chuang, H.-P., Ohashi, A., Imachi, H., Tandukar, M., Harada, H., 2007. Effective partial nitrification to nitrite by down-flow hanging sponge reactor under limited oxygen condition. Water Research 41, 295–302. Chudoba, J., 1985. Control of activated sludge filamentous bulking-VI. Formulation of basic principle. Water Research 19, 1017–1022. Gao, D., Yuan, X., Liang, H., Wu, W.-M., 2011. Comparison of biological removal via nitrite with real-time control using aerobic granular sludge and flocculent activated sludge. Applied Microbiology and Biotechnology 89, 1645–1652. Guo, J.H., Peng, Y.Z., Yang, Q., Wang, S.Y., Chen, Y., Zhao, C.H., 2008. Theoretical analysis and enhanced nitrogen removal performance of step-feed SBR. Water Science and Technology 58, 795–802. Jubany, I., Lafuente, J., Baeza, J.A., Carrera, J., 2009. Total and stable washout of nitrite oxidizing bacteria from a nitrifying continuous activated sludge system using automatic control based on Oxygen Uptake Rate measurements. Water Research 43, 2761–2772. Lee, D.J., Chen, Y.Y., Show, K.Y., Whiteley, C.G., Tay, J.H., 2010. Advances in aerobic granule formation and granule stability in the course of storage and reactor operation. Biotechnology Advances 28, 919–934. Lemaire, R., Marcelino, M., Yuan, Z., 2008. Achieving the nitrite pathway using aeration phase length control and step-feed in an SBR removing nutrients from abattoir wastewater. Biotechnology and Bioengineering 100, 1228–1236. Liu, Y., Tay, J.H., 2004. State of the art of biogranulation technology for wastewater treatment. Biotechnology Advances 22, 533–563. Lopez-Palau, S., Dosta, J., Pericas, A., Mata-Alvarez, J., 2011. Partial nitrification of sludge reject water using suspended and granular biomass. Journal of Chemical Technology and Biotechnology 86, 1480–1487. ˇ izˇinská, S., Krejcˇí, J., Janoch, T., 1992. Biological water denitrification – a Mateˇju˚, V., C review. Enzyme and Microbial Technology 14, 170–183. McSwain, B.S., Irvine, R.L., Wilderer, P.A., 2004. Effect of intermittent feeding on aerobic granule structure. Water Science and Technology 49, 19–25. Mosquera-Corral, A., De Kreuk, M.K., Heijnen, J.J., Van Loosdrecht, M.C.M., 2005a. Effects of oxygen concentration on N-removal in an aerobic granular sludge reactor. Water Research 39, 2676–2686. Mosquera-Corral, A., González, F., Campos, J.L., Méndez, R., 2005b. Partial nitrification in a SHARON reactor in the presence of salts and organic carbon compounds. Process Biochemistry 40, 3109–3118. Moy, B.Y.P., Tay, J.H., Toh, S.K., Liu, Y., Tay, S.T.L., 2002. High organic loading influences the physical characteristics of aerobic sludge granules. Letters in Applied Microbiology 34, 407–412. Ni, B.J., Yu, H.Q., Sun, Y.J., 2008. Modeling simultaneous autotrophic and heterotrophic growth in aerobic granules. Water Research 42, 1583–1594. Peng, Y., Zhu, G., 2006. Biological nitrogen removal with nitrification and denitrification via nitrite pathway. Applied Microbiology and Biotechnology 73, 15–26. Qin, L., Liu, Y., 2006. Aerobic granulation for organic carbon and nitrogen removal in alternating aerobic-anaerobic sequencing batch reactor. Chemosphere 63, 926– 933. Ruiz, G., Jeison, D., Chamy, R., 2003. Nitrification with high nitrite accumulation for the treatment of wastewater with high ammonia concentration. Water Research 37, 1371–1377.
X.-H. Wang et al. / Bioresource Technology 123 (2012) 375–381 Shi, X.Y., Yu, H.Q., Sun, Y.J., Huang, X., 2009. Characteristics of aerobic granules rich in autotrophic ammonium-oxidizing bacteria in a sequencing batch reactor. Chemical Engineering Journal 147, 102–109. Shi, Y.J., Wang, X.H., Yu, H.B., Xie, H.J., Teng, S.X., Sun, X.F., Tian, B.H., Wang, S.G., 2011. Aerobic granulation for nitrogen removal via nitrite in a sequencing batch reactor and the emission of nitrous oxide. Bioresource Technology 102, 2536– 2541. Su, B., Cui, X., Zhu, J., 2012. Optimal cultivation and characteristics of aerobic granules with typical domestic sewage in an alternating anaerobic/aerobic sequencing batch reactor. Bioresource Technology 110, 125–129. Tay, J.H., Ivanov, V., Pan, S., Tay, S.T.L., 2002. Specific layers in aerobically grown microbial granules. Letters in Applied Microbiology 34, 254–257. Tay, J.H., Liu, Q.S., Liu, Y., 2001. The effects of shear force on the formation, structure and metabolism of aerobic granules. Applied Microbiology and Biotechnology 57, 227–233. Thauer, R.K., Jungermann, K., Decker, K., 1977. Energy conservation in chemotrophic anaerobic bacteria. Bacteriological Reviews 41, 100–180. Vázquez-Padín, J.R., Figueroa, M., Campos, J.L., Mosquera-Corral, A., Méndez, R., 2010. Nitrifying granular systems: a suitable technology to obtain stable partial nitrification at room temperature. Separation and Purification Technology 74, 178–186.
381
Vadivelu, V.M., Keller, J., Yuan, Z., 2007. Effect of free ammonia on the respiration and growth processes of an enriched Nitrobacter culture. Water Research 41, 826–834. Wan, J.F., Bessiere, Y., Sperandio, M., 2009. Alternating anoxic feast/aerobic famine condition for improving granular sludge formation in sequencing batch airlift reactor at reduced aeration rate. Water Research 43, 5097– 5108. Xavier, J.B., de Kreuk, M.K., Picioreanu, C., van Loosdrecht, M.C.M., 2007. Multiscale individual-based model of microbial and bioconversion dynamics in aerobic granular sludge. Environmental Science and Technology 41, 6410– 6417. Yang, Q., Peng, Y., Liu, X., Zeng, W., Mino, T., Satoh, H., 2007. Nitrogen removal via nitrite from municipal wastewater at low temperatures using real-time control to optimize nitrifying communities. Environmental Science and Technology 41, 8159–8164. Zeng, W., Zhang, Y., Li, L., Peng, Y.-z., Wang, S.-y., 2009. Control and optimization of nitrifying communities for nitritation from domestic wastewater at room temperatures. Enzyme and Microbial Technology 45, 226–232. Zhang, H., Dong, F., Jiang, T., Wei, Y., Wang, T., Yang, F., 2011. Aerobic granulation with low strength wastewater at low aeration rate in A/O/A SBR reactor. Enzyme and Microbial Technology 49, 215–222.
Contents lists available at SciVerse ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Effects of step-feed on granulation processes and nitrogen removal performances of partial nitrifying granules Xin-Hua Wang, Li-Xiu Jiang, Yi-Jing Shi, Ming-Ming Gao, Sen Yang, Shu-Guang Wang ⇑ Shandong Provincial Key Laboratory of Water Pollution Control and Resource Reuse, School of Environmental Science and Engineering, Shandong University, 27 Shanda Nanlu, Jinan 250100, China
h i g h l i g h t s " PNG was successfully cultivated under step-feed mode. " Step-feed delayed the granulation processes of PNG. " PNG cultivated in both reactors with or without step-feed had similar properties. " Step-feed enhanced denitrification and TN removal, as well as ammonia oxidation.
a r t i c l e
i n f o
Article history: Received 12 June 2012 Received in revised form 18 July 2012 Accepted 22 July 2012 Available online 27 July 2012 Keywords: Aerobic granules Partial nitrification Step-feed Granulation processes Denitrification
a b s t r a c t Two anoxic/oxic sequencing batch reactors (A/O SBRs) were operated to investigate the effects of stepfeed on granulation processes and performances of partial nitrifying granules (PNG). R1 was operated in a traditional single-feed mode, while a two-step-feed strategy was used in R2. Results showed that R1 had a faster granulation process and better performance in maintaining partial nitrification compared with R2, indicating that the step-feed mode had a negative effect of on formation of PNG. However, after full granulation, PNG in both reactors had similar properties in terms of suspended solids (MLSS), sludge volume index (SVI) and granule size. Moreover, mature granules in R2 had a higher nitrite accumulation rate than that in R1. Step-feed strategy was also observed to enhance denitrification and TN removal, as well as ammonia oxidation. It can be concluded that step-feed was unfavorable for cultivating PNG, but it significantly improved the nitrogen removal performance of PNG. Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction Partial nitrifying activated sludge (PNAS) processes have attracted the most attention in recent years, owing to their advantages with respect to conventional nitrate pathway: reduce the energy consumption for aeration and organic carbon requirement for denitrification (Peng and Zhu, 2006; Ruiz et al., 2003). Several factors have been identified to selectively inhibit or washout nitrite oxidizing bacteria (NOB) over ammonia oxidizing bacteria (AOB), mainly including dissolved oxygen (DO), temperature, sludge retention time (SRT), free ammonia (FA) and free nitrous acid (FNA) (Peng and Zhu, 2006; Ruiz et al., 2003). DO is one of the main factors inhibiting NOB, but long-term operation at low DO would result in low nitritation rate, sludge bulking or the increasing N2O production (Chuang et al., 2007; Zeng et al., 2009). In addition, ⇑ Corresponding author. Tel.: +86 531 88362220; fax: +86 531 88364513. E-mail addresses: [email protected] (X.-H. Wang), [email protected] (S.-G. Wang). 0960-8524/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biortech.2012.07.080
the capacity of partial nitrification is limited due to lower biomass concentration in activated sludge (AS) systems without effective biomass retention (Jubany et al., 2009; Mosquera-Corral et al., 2005b). Granules based systems have been proposed as a suitable alternative to obtain stable and robust partial nitrification (Gao et al., 2011; Lopez-Palau et al., 2011; Vázquez-Padín et al., 2010). Partial nitrifying granules (PNG) present high cell density and much fast settling velocities than AS, thus ensuring higher biomass concentration and specific activities and consequent higher maximum load (Bartroli et al., 2010; Gao et al., 2011; Lopez-Palau et al., 2011). Moreover, DO diffusion limitation within granules allow operating reactors reaching stable partial nitrification without strict DO and temperature controls (Shi et al., 2011; Vázquez-Padín et al., 2010). Granules also have special layered microbial structure with more denitrifying bacteria existed inside the granules, giving them a great potential for better denitrification (Mosquera-Corral et al., 2005a; Tay et al., 2002). However, during PNG processes, denitrifi-
376
X.-H. Wang et al. / Bioresource Technology 123 (2012) 375–381
cation is normally restricted by deficient carbon source supply in conventional oxic or anoxic/oxic (A/O) sequencing batch reactors (SBR), especially when treating high strength nitrogen and low C/N ratio wastewater (Shi et al., 2009; Shi et al., 2011). The use of an external carbon source will improve N-removal, but also undermines the overall benefits of the nitrite pathway. Step-feed mode for A/O SBR has been reported to be effective in making good use of carbon source in influent and increasing denitrification rate and further total nitrogen (TN) removal efficiency. In addition, it benefits the growth of AOB and inhibits NOB, accelerates nitrite accumulation and nitritation rate for PNAS systems (Lemaire et al., 2008; Yang et al., 2007). Compared with conventional single-feed SBR, step-feed mode changes some factors which may affect granule formation and/or nitrite accumulation, such as smaller substrate load at the beginning of each react phase and shorter but more frequent feast-famine conditions. However, the feasibility of PNG formation under step-feed mode and the comparison with traditional single-feed mode were still sparse. Thus, this study used two A/O SBRs with and without step-feed to cultivate PNG. The main objective was to find the effects of feeding mode on granulation processes and nitrogen removal performances. It is expected that the results derived from this study will provide useful information or guides for practical applications of PNG in treating high strength nitrogen and low C/N wastewater.
Synthetic wastewater mainly consist of glucose, sodium acetate, ammonium chloride, sodium bicarbonate and other necessary mineral-salts medium. The influent COD and NHþ 4 N were fixed at 800 (50% contribution each of glucose and sodium acetate) and 300 mg/L, respectively. 2.2. Analysis
2. Methods
Suspended solids (MLSS), sludge volume index (SVI), ammonium, nitrate and nitrite were analyzed by standard methods (APHA, 1998). Influent samples were taken directly from the feed tanks, while effluent samples were pretreated by passing through a 0.45 lm filter. Grab samples of mixed liquor were taken from equidistant sampling ports along the height of each reactor and gently mixed prior to the measurement of various parameters and observation of microbial compositions. After mixing, the samples were completely blended for the MLSS and SVI tests, or were filtered through 0.45 lm filters for the measurement of ammonium, nitrate and nitrite. The mean granule size was measured by a laser particle size analysis system (Mastersizer 2000, Malvern Instruments, UK). DO levels were measured with a DO meter (YSI Model 85, USA). The microbial compositions of granules were observed qualitatively with scanning electron microscope (SEM) (HITACHI S-570, Japan). The granule samples were fixed with 3.0% glutaraldehyde in 0.1 M phosphate buffer at pH 7.2. The samples were then dehydrated with ethanol, silver-coated by a sputter and observed in the SEM.
2.1. Reactor set-up and operation
2.3. Calculations
Two identical SBRs (R1 and R2, 10 cm in diameter and 35 cm in height) with working volumes of 2.5 L were used for formation of PNG. The inoculated PNAS exhibited an excellent partial nitrification performance with nitrite accumulation rate higher than 85%. Both reactors were operated at 28 °C and without pH control between 7.5 and 8.5. Sampling ports were built along the height of each reactor at 10 cm intervals. Effluent was discharged from the middle port of the reactor with a volumetric exchange ratio of 50%. The R1 was operated in a traditional A/O cycles, consisting of influent feeding, anoxic phase (mixing), oxic phase (aeration), settling and effluent discharge. While the R2 cycles comprised a two-step-feed strategy, with 50% influent introduced in the first feed and the remaining 50% in the second feed. After each feed a sequence of anoxic and oxic phases were applied. Following the two sequences the cycle ended with settling and discharge phases. Tables 1 and 2 summarizes the operational conditions of SBRs and the different time distribution in the cycles. The time period of each phase and the cycle were modified according to nitrogen removal performance based on analytical data evaluation. The settling time was progressively decreased to improve granulation. In order to facilitate nitrite accumulation, the reactors were initially aerated at a superficial upflow air velocity (SUAV) of 0.2 cm/s, which corresponding to DO concentrations between 2.8 and 4.1 mg/L during oxic phases. At the start of period IV, the aeration rate was raised to 1.0 cm/s in terms of SUAV, giving a higher DO concentrations between 6.0 and 7.5 mg/L.
The ammonia oxidation rate was calculated from the slopes of ammonia consumption from cycle studies. The nitrite accumulation rate (NAR) was calculated using the following equation:
NAR ð%Þ ¼
NO2
NO N N þ NO3 N
ð1Þ
The FA concentration in reactor can be estimated by the expression proposed by (Anthonisen et al., 1976):
FA ðmg=LÞ ¼
17 ½NHþ4 N 10pH 14 exp½6334=ð273 þ TÞ þ 10pH
ð2Þ
where ½NHþ 4 N is the ammonia–nitrogen concentration and T is the temperature in °C. 3. Results and discussion 3.1. Granulation processes 3.1.1. Formation of granules In the two reactors the time needed for the granules development was different. In R1 operated in single-feed A/O SBR mode, initial tiny granules appeared on day 33 (period II). Then granules grew and proliferated quickly in the reactor (Fig. S1). Full granulation achieved on day 56 (period III) when round-shaped granules
Table 1 Operational conditions of SBRs during the experiments. Period
Days
Cycle time (h)
SUAV (cm/s)
DO in oxic phase (mg/L)
Settling time (min)
HRT (h)
I II III IV V
1–28 29–44 45–63 64–90 91–128
6 6 4 4 4
0.2 0.2 0.2 1.0 1.0
2.7–4.2 2.7–4.2 2.7–4.2 6.0–7.5 6.0–7.5
8 4 2 2 1
12 12 8 8 8
377
X.-H. Wang et al. / Bioresource Technology 123 (2012) 375–381 Table 2 Different time distribution in the cycle SBRs.
R1 R2
Cycle time (min)
Feed (min)
Anoxic (min)
Oxic (min)
Feed (min)
Anoxic (min)
Oxic (min)
Settling (min)
Discharge (min)
6 4 6 4
12 12 6 6
84 84 42 42
252/256 138/139 126/128 78/79
– – 6 6
– – 42 42
– – 126/128 60/60
8/4 2/1 8/4 2/1
4 4 4 4
had became the dominant form of biomass. As shown from the SEM images, granules in R1 had a compact and dense structure and a bacilli- and cocci-dominant outer surface (Fig. S2). Initial and full granulation time in the step-feed R2 were both obviously longer (Fig. S1). The biomass in R2 always had a floc-like appearance during period I and II. After the settling time was reduced to 2 min in period III, granules were firstly observed on day 50. At the beginning of period IV, a higher SUAV was supplied in the reactor to enhance granulation. However, a lot of filamentous flocs were still present and coexisted with granules in this period. After the settling time was further reduced to 1 min during period V, filamentous flocs were washed out successively and full granulation was observed on day 109. Granules in R2 had a similar microstructure with those in R1, but a certain amount of filamentous bacteria were found to present (Fig. S2). While the granules were successfully cultivated in R2, the step-feed strategy significantly delayed the sludge granulation process. Aerobic granulation is a gradual process from fluffy seed flocs to dense granules often occurring in SBR. It has been reported that granulation is influenced by a variety of factors, mainly including seed sludge, substrate composition and load, and SBR operation (Lee et al., 2010). Microorganisms in the single-feed R1 are subjected to a periodic feast and famine regime. Because the substrate was equally divided into two streams and introduced two times in R2, the reactor operation in one cycle includes two periodic feast and famine. And also, R2 creates a smaller substrate concentration at the beginning of each feast phase as compared with R1. According to the kinetic selection theory, at low substrate concentration, filamentous organisms would achieve a high substrate removal rate and its outgrowth would be favored (Chudoba, 1985). Because of the proliferation of filamentous bacteria, circumstances are disadvantageous for the formation of granules with regular structures. This may explain why the granulation process in the stepfeed SBR was delayed. McSwain et al. (2004) also investigated the role of feeding strategy for aerobic granulation in three SBRs applied with different feeding times. It was found that pulse feeding which creates a high feast condition was favorable for the formation of aerobic granules, because it affects the selection and growth of filamentous organisms. This is consistent with the results in our study. It should be pointed out that both initial granulation in R1 and R2 were realized under a low shear force in terms of SUAV of 0.2 cm/s. Much research has asserted that high shear forces are necessary for the formation of aerobic granules and the minimum SUAV should not be lower than 1.2 cm/s (Adav et al., 2007; Tay et al., 2001). However, granule formation are increasingly reported under lower shear forces. Zhang et al. (2011) reported that dense and regular-shaped aerobic granules were successfully formed at SUAV of 0.6 cm/s by treating low strength synthetic wastewater in A/O/A SBR during the first 50 days operation. Thus, high shear force do not appear to be a crucial factor for granulation, but may be more important in compacting granule structure and eroding excess cells and attached materials from an already formed granule (Lee et al., 2010). The formation of initial granules in both reactors in this study was partly attributed to the high sludge concentration at the beginning of the operation and A/O mode for the SBRs. On one
hand, as proposed by Su et al. (2012), high sludge concentration causes more frequent collision and stronger friction among microorganisms, resulting in the improvement of microbial autoaggregation. On the other hand, because of degradation of most organic matter by denitrifying bacteria at anoxic period in A/O SBRs, the microorganisms in the reactors would encounter with longer starvation state, which may further improve aerobic granulation. 3.1.2. Properties of biomass The initial seed sludge with a mean floc size of 0.12 mm and a SVI value of 40.5 mL/g showed a fluffy, irregular and loosestructured morphology. According to the observed trend in MLSS, SVI and size of biomass (Fig. 1), both granulation processes in the two reactors could be divided into three typical phases: acclimation phase, multiplication phase followed by maturation phase. In the acclimation phase of R1 from day 1 to day 32, the MLSS and SVI both decreased and the size of biomass increased. The start of the multiplication phase in R1 was marked by the appearance of initial tiny granules on day 33 within the reactor. The MLSS and mean diameter of granules increased substantially to 22.6 g/L and 0.87 mm on day 81, respectively. The appearance of granules was also accompanied by gradual improvement of settleability. The SVI decreased to 12.0 mL/g during the multiplication phase. The following maturation phase lasted from day 82 to the end of R1 operation, which was characterized by a little fluctuation or no change in the values of MLSS (23.3 ± 1.2 g/L), SVI (11.6 ± 1.4 mL/g) and mean diameter (0.91 ± 0.04 mm). There were no much difference between R1 and R2 in terms of MLSS, SVI and size of biomass during period I and II. The acclimation phase in R2 lasted until day 50 when granules were firstly observed. During the multiplication phase lasted from day 51 to day 114, granules increasingly formed within the reactor, resulted in a continual increase of MLSS and mean diameter as well as a gradual improvement of settleability. At the beginning of period V, washout of filamentous flocs caused a sudden decrease in MLSS. During the maturation phase from day 115 to the end of reactor operation, the MLSS and SVI stabilized at 26.8 ± 0.9 g/L and 10.6 ± 1.3 mL/g, while the granule size increased to plateau at 1.10 mm. The step-feed strategy delayed the granulation process, which would inevitably retard the increase of biomass concentration and granule size as well as the improvement of settleability during the acclimation phase and multiplication phase (Fig. 1). However, the step-feed strategy did not impair the physicochemical properties of mature granules. The SVI of the mature granules in both reactors were in the same range of 10–13 mL/g, while the MLSS and mean granule size in the step-feed R2 were slightly higher than that in R1. The SVI values are very close to that reported by Qin and Liu (2006) whom cultivated granules with SVI of 12 mL/g under alternating O/A conditions and nitrogen loading rate of 0.45 kg/m3d, but are much lower than those granules cultivated under absolute aerobic conditions with a typical value of 50– 100 mL/g (Moy et al., 2002; Tay et al., 2001). These seem to indicate that the imposed anoxic denitrification condition plays a positive role on sludge properties, because it would enhance heterotrophic growth/storage deeper in the internal anoxic layer of granules and encourage aggregate densification (Wan et al., 2009).
378
X.-H. Wang et al. / Bioresource Technology 123 (2012) 375–381
Fig. 1. Time profiles of MLSS (a), SVI (b) and mean diameter (c) in R1 (j) and R2 (s).
3.2. Reactor performance 3.2.1. Nitrite accumulation performance R1 exhibited a good performance in maintaining nitrite accumulation during the entire operation period, as shown by the low effluent nitrate concentration (20.2 ± 8.7 mg/L) and high NAR (87 ± 5%) in Figs. 2 and 3. Sudden increase of DO concentration in oxic phase caused by raising of aeration rate since period IV did not deteriorate the partial nitrification performance, which would be attributed to the full granulation in the reactor and the oxygen transfer limitation within granules (Shi et al., 2011; Vázquez-Padín et al., 2010). In contrast, a gradual increase in effluent nitrate concentration was observed in R2 during period I and II, which certainly resulted in a decrease in NAR. To further study the operation status of the reactors, typical cycle measurements were conducted on day 40. As shown in Fig. 4a and c, nitrite oxidation was always lag behind ammonia oxidation in both reactors. But in the cycle of 6 h, when ammonia oxidation was completed, it still had a considerable time for the remaining nitrite to be converted to nitrate by NOB in oxic phases. This phenomenon was especially obvious in the step-feed SBR, probably because of the lower FA inhibition on NOB in the reactor. NOB has been described to be much more sensitive to FA than AOB. Anthonisen et al. (1976) reported that AOB and NOB were inhibited at 10–150 mg/L and 0.1–1.0 mg/L of FA. Vadivelu et al. (2007) found Nitrobacter likely ceased to grow at an FA concentration of 6–9 mg/L. The SBRs in this study were operated at 28 °C temperature and pH was between 7.5 and 8.5 during experimental studies. Under step-feed mode, low concentrations of ammonia were introduced into the reactor before every anoxic phases. Thus, the calculated concentrations of FA were considerable lower in R2 (0.1–6.9 mg/L) than R1 (0.1–19.0 mg/L) (Fig. 4). The maximum FA concentration in R1 was far above the threshold concentration of
Fig. 2. Reactors operation in terms of NHþ 4 N concentration in the feeding (j), and NHþ 4 N (N), NO2 –N (s) and NO3 –N () concentrations in the effluent.
inhibition at which the inhibition of NOB began, which led to a comparative stable nitrite accumulation. When the cycle time was reduced to 4 h since period III (mainly by shortened the duration of oxic phases as shown in Table 2), the
X.-H. Wang et al. / Bioresource Technology 123 (2012) 375–381
379
nitrite accumulation. This conclusion was also demonstrated by previous studies (Lemaire et al., 2008; Yang et al., 2007). As shown from the cycle measurements under steady states in both reactors (day 122) in Fig. 3b and d, nitrite accumulated during oxic phase in R2 was always lower than that in R1 and was almost denitrified in time, which reduced the substrate of nitratation and further prevented the toxicity of nitrite. In addition, a bigger granule size reached in the later reactor operation (Fig. 1c) would enhance the oxygen transfer limitation and further accelerate nitrite accumulation.
Fig. 3. Nitrite accumulation rate in R1 (j) and R2 (s) along the operational period.
nitrite accumulation in R2 was quickly recovered with NAR increased from below 20% to above 85%. Meanwhile, a slight enhancement of nitrite accumulation was also observed in R1. In period IV, a transient deterioration of nitrite accumulation performance occurred due to the sudden increase of DO concentration (6.0–7.5 mg/L) in oxic phase, during when flocs were still dominated in R2. With granules increasingly formed and grew bigger, the NAR increased and finally stabilized at 95 ± 3% with low concentrations of effluent nitrate (6.1 ± 4.6 mg/L) (Figs. 2 and 3). This phenomenon revealed the advantages for granules in maintaining partial nitrification and demonstrated the crucial role of oxygen transfer limitation for reaching stable partial nitrification in granule based reactors (Shi et al., 2011; Vázquez-Padín et al., 2010). As compared with R1, a better nitrite accumulation performance was achieved in R2 after sludge full granulated, which means that the step-feed operational mode was favorable for the
3.2.2. TN removal performance Step-feed strategy for R2 was also observed to enhance denitrification and TN removal. As shown from Fig. 5, the TN removal efficiency in R2 was obviously higher than that in R1 from period III to the end of operation. The TN removal efficiency in R1 was basically fluctuated between 40% and 45%, except for period III when ammonia oxidation was seriously deteriorated. Guo et al. (2008) have reported that TN removal efficiency would increase with increasing of feeding steps. For R2 operated in two-step-feed mode, the value reached above 70% during the maturation phase of PNG. Step-feed strategy for SBR has been intensively regarded as an efficient nitrogen removal system, especially for the wastewater with lower C/N ratios. The step-feed mode greatly improved the availability of influent organics for denitrification in the SBR operated alternating A/O sequences, thereby elevating the TN removal capacity. In period I and II, nitrate was increasingly produced in R2 and this would raise the demand of carbon source for denitrification. The stoichiometric ratio of the heterotrophic denitrification is 1.71 mg COD/mg NO2–N and 2.85 mg COD/mg NO3–N. These values must be corrected because some of the carbon is used for biomass synthesis. With glucose as the carbonaceous substrate,
(a)
(b)
(c)
(d)
Fig. 4. Typical cycle measurement of NHþ 4 N (N), NO2 –N (s), NO3 –N (), TN (j), and FA (w) in R1 on day 40 (a) and 122 (b) and R2 on day 40 (c) and 122 (d).
380
X.-H. Wang et al. / Bioresource Technology 123 (2012) 375–381
removal after granules formed, indicated by the significantly improved TN removal efficiency and ammonia oxidation rate. Acknowledgements This work was supported by the National Natural Science Foundation of China (51108250), the Specialized Research Fund for the Doctoral Program of Higher Education of China (20110131120015), and the Excellent Young Scientist Foundation of Shandong Province (BS2010HZ009). Appendix A. Supplementary data Fig. 5. TN removal efficiency in R1 (j) and R2 (s) along the operational period.
the theoretical value obtained from the stoichiometric denitrification equation is 4.9 mg COD/mg NO3–N (Mateˇju˚ et al., 1992). While with acetate as a sole carbon source, the value was reported to be 3.75 mg COD/mg NO3–N (Thauer et al., 1977). Thus, a decrease of TN removal efficiency in period I and II of R2 was probably caused by the deficiency of carbon source for denitrification, because the influent COD/N (800/300) was lower in this study. 3.2.3. Ammonia removal performance Both reactors exhibited an excellent and similar ammonia removal performance in most times, as revealed by the low ammonia–nitrogen concentration in the effluent (below 5.0 mg/L as shown in Fig. 2) and high removal efficiency above 98%. A transitory deterioration in ammonia removal performance in period III was mainly resulted from the shorten of oxic phases duration and the low biomass concentrations in both reactors (Fig. 1). Similarly, a slight increase of effluent ammonia–nitrogen concentration from day 90 in R2 was also caused by the decrease of MLSS concentration. According to the comparison of the ammonia oxidation rate calculated from cycle studies, the step-feed operational mode was considered to accelerate AOB activity. The respective ammonia oxidation rate in R1 and R2 were 7.38 ± 0.42 and 11.40 ± 1.25 mg/g SSh on day 40, and 3.42 ± 0.30 and 4.41 ± 0.52 mg/gSSh on day 122. A decrease in specific biomass activity of both reactors on day 122 are supposed to be related to their bigger biomass size. It is generally accepted that granules with a smaller size are more efficient in terms of mass transfer and substrate conversion, and therefore more effective in degrading substrate (Liu and Tay, 2004). The ammonia oxidation rate in R2 was 54% and 29% higher than that in R1 on day 40 and 122, respectively. Lower ammonia and organic loading in the oxic phase could be responsible for the enhancement of nitritation activity in the granules. Firstly, lower ammonia concentration in R2 conduced to lower FA inhibition on AOB activity, as mentioned previously. Secondly, due to the existence of oxygen and space competition between AOB and heterotrophs within granules, nitritation occurred at a lower organic loading would have a higher rate (Ni et al., 2008; Xavier et al., 2007). 4. Conclusions PNG with similar physicochemical properties could be cultivated in both reactors with or without step-feed. However, the step-feed strategy markedly delayed the granulation processes and caused unstable partial nitrification during the formation of granules, probably owing to the smaller substrate concentration at the beginning of each feast phase. Though the step-feed was unfavorable for cultivating PNG, it had a positive effect on nitrogen
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.biortech.2012. 07.080. References Adav, S.S., Lee, D.J., Lai, J., 2007. Effects of aeration intensity on formation of phenolfed aerobic granules and extracellular polymeric substances. Applied Microbiology and Biotechnology 77, 175–182. Anthonisen, A.C., Loehr, R.C., Prakasam, T.B.S., Srinarh, E.G., 1976. Inhibition of nitrification by ammonia and nitrous acid. Journal of the Water Pollution Control Federation 48, 835–852. APHA, 1998. Standard Methods for the Examination of Water and Wastewater. 20th ed. American Public Health Association, Washington, DC. Bartroli, A., Perez, J., Carrera, J., 2010. Applying ratio control in a continuous granular reactor to achieve full nitritation under stable operating conditions. Environmental Science and Technology 44, 8930–8935. Chuang, H.-P., Ohashi, A., Imachi, H., Tandukar, M., Harada, H., 2007. Effective partial nitrification to nitrite by down-flow hanging sponge reactor under limited oxygen condition. Water Research 41, 295–302. Chudoba, J., 1985. Control of activated sludge filamentous bulking-VI. Formulation of basic principle. Water Research 19, 1017–1022. Gao, D., Yuan, X., Liang, H., Wu, W.-M., 2011. Comparison of biological removal via nitrite with real-time control using aerobic granular sludge and flocculent activated sludge. Applied Microbiology and Biotechnology 89, 1645–1652. Guo, J.H., Peng, Y.Z., Yang, Q., Wang, S.Y., Chen, Y., Zhao, C.H., 2008. Theoretical analysis and enhanced nitrogen removal performance of step-feed SBR. Water Science and Technology 58, 795–802. Jubany, I., Lafuente, J., Baeza, J.A., Carrera, J., 2009. Total and stable washout of nitrite oxidizing bacteria from a nitrifying continuous activated sludge system using automatic control based on Oxygen Uptake Rate measurements. Water Research 43, 2761–2772. Lee, D.J., Chen, Y.Y., Show, K.Y., Whiteley, C.G., Tay, J.H., 2010. Advances in aerobic granule formation and granule stability in the course of storage and reactor operation. Biotechnology Advances 28, 919–934. Lemaire, R., Marcelino, M., Yuan, Z., 2008. Achieving the nitrite pathway using aeration phase length control and step-feed in an SBR removing nutrients from abattoir wastewater. Biotechnology and Bioengineering 100, 1228–1236. Liu, Y., Tay, J.H., 2004. State of the art of biogranulation technology for wastewater treatment. Biotechnology Advances 22, 533–563. Lopez-Palau, S., Dosta, J., Pericas, A., Mata-Alvarez, J., 2011. Partial nitrification of sludge reject water using suspended and granular biomass. Journal of Chemical Technology and Biotechnology 86, 1480–1487. ˇ izˇinská, S., Krejcˇí, J., Janoch, T., 1992. Biological water denitrification – a Mateˇju˚, V., C review. Enzyme and Microbial Technology 14, 170–183. McSwain, B.S., Irvine, R.L., Wilderer, P.A., 2004. Effect of intermittent feeding on aerobic granule structure. Water Science and Technology 49, 19–25. Mosquera-Corral, A., De Kreuk, M.K., Heijnen, J.J., Van Loosdrecht, M.C.M., 2005a. Effects of oxygen concentration on N-removal in an aerobic granular sludge reactor. Water Research 39, 2676–2686. Mosquera-Corral, A., González, F., Campos, J.L., Méndez, R., 2005b. Partial nitrification in a SHARON reactor in the presence of salts and organic carbon compounds. Process Biochemistry 40, 3109–3118. Moy, B.Y.P., Tay, J.H., Toh, S.K., Liu, Y., Tay, S.T.L., 2002. High organic loading influences the physical characteristics of aerobic sludge granules. Letters in Applied Microbiology 34, 407–412. Ni, B.J., Yu, H.Q., Sun, Y.J., 2008. Modeling simultaneous autotrophic and heterotrophic growth in aerobic granules. Water Research 42, 1583–1594. Peng, Y., Zhu, G., 2006. Biological nitrogen removal with nitrification and denitrification via nitrite pathway. Applied Microbiology and Biotechnology 73, 15–26. Qin, L., Liu, Y., 2006. Aerobic granulation for organic carbon and nitrogen removal in alternating aerobic-anaerobic sequencing batch reactor. Chemosphere 63, 926– 933. Ruiz, G., Jeison, D., Chamy, R., 2003. Nitrification with high nitrite accumulation for the treatment of wastewater with high ammonia concentration. Water Research 37, 1371–1377.
X.-H. Wang et al. / Bioresource Technology 123 (2012) 375–381 Shi, X.Y., Yu, H.Q., Sun, Y.J., Huang, X., 2009. Characteristics of aerobic granules rich in autotrophic ammonium-oxidizing bacteria in a sequencing batch reactor. Chemical Engineering Journal 147, 102–109. Shi, Y.J., Wang, X.H., Yu, H.B., Xie, H.J., Teng, S.X., Sun, X.F., Tian, B.H., Wang, S.G., 2011. Aerobic granulation for nitrogen removal via nitrite in a sequencing batch reactor and the emission of nitrous oxide. Bioresource Technology 102, 2536– 2541. Su, B., Cui, X., Zhu, J., 2012. Optimal cultivation and characteristics of aerobic granules with typical domestic sewage in an alternating anaerobic/aerobic sequencing batch reactor. Bioresource Technology 110, 125–129. Tay, J.H., Ivanov, V., Pan, S., Tay, S.T.L., 2002. Specific layers in aerobically grown microbial granules. Letters in Applied Microbiology 34, 254–257. Tay, J.H., Liu, Q.S., Liu, Y., 2001. The effects of shear force on the formation, structure and metabolism of aerobic granules. Applied Microbiology and Biotechnology 57, 227–233. Thauer, R.K., Jungermann, K., Decker, K., 1977. Energy conservation in chemotrophic anaerobic bacteria. Bacteriological Reviews 41, 100–180. Vázquez-Padín, J.R., Figueroa, M., Campos, J.L., Mosquera-Corral, A., Méndez, R., 2010. Nitrifying granular systems: a suitable technology to obtain stable partial nitrification at room temperature. Separation and Purification Technology 74, 178–186.
381
Vadivelu, V.M., Keller, J., Yuan, Z., 2007. Effect of free ammonia on the respiration and growth processes of an enriched Nitrobacter culture. Water Research 41, 826–834. Wan, J.F., Bessiere, Y., Sperandio, M., 2009. Alternating anoxic feast/aerobic famine condition for improving granular sludge formation in sequencing batch airlift reactor at reduced aeration rate. Water Research 43, 5097– 5108. Xavier, J.B., de Kreuk, M.K., Picioreanu, C., van Loosdrecht, M.C.M., 2007. Multiscale individual-based model of microbial and bioconversion dynamics in aerobic granular sludge. Environmental Science and Technology 41, 6410– 6417. Yang, Q., Peng, Y., Liu, X., Zeng, W., Mino, T., Satoh, H., 2007. Nitrogen removal via nitrite from municipal wastewater at low temperatures using real-time control to optimize nitrifying communities. Environmental Science and Technology 41, 8159–8164. Zeng, W., Zhang, Y., Li, L., Peng, Y.-z., Wang, S.-y., 2009. Control and optimization of nitrifying communities for nitritation from domestic wastewater at room temperatures. Enzyme and Microbial Technology 45, 226–232. Zhang, H., Dong, F., Jiang, T., Wei, Y., Wang, T., Yang, F., 2011. Aerobic granulation with low strength wastewater at low aeration rate in A/O/A SBR reactor. Enzyme and Microbial Technology 49, 215–222.