- Email: [email protected]
Limited filamentous bulking in order to enhance integrated nutrient removal and effluent quality
w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 4 8 7 7 e4 8 8 4
Available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/watres
Limited filamentous bulking in order to enhance integrated nutrient removal and effluent quality Wen-De Tian a,b, Wei-Guang Li a,b,*, Hui Zhang a,b, Xiao-Rong Kang a,b, Mark C.M. van Loosdrecht c,d a
School of Municipal and Environmental Engineering, Harbin Institute of Technology, Harbin 150090, China State Key Laboratory of Urban Water Resource Environment, Harbin Institute of Technology, Harbin 150090, China c Department of Biotechnology, Delft University of Technology, Julianalaan 67, 2628 BC Delft, The Netherlands d KWR Watercycle Research Institute, Groningenhaven 7, 3422 PE Nieuwegein, The Netherlands b
article info
abstract
Article history:
Limited filamentous bulking has been proposed as a means to enhance floc size and make
Received 23 December 2010
conditions more favorable for simultaneous nitrification/Denitrification (SND). Moreover
Received in revised form
a slightly heightened SVI is supposed to increase the removal of small particulates in the
23 June 2011
clarifier. Integrated nitrogen, phosphorus and COD removal performance under limited
Accepted 25 June 2011
filamentous bulking was investigated using a bench-scale plug-flow enhanced biological
Available online 6 July 2011
phosphorus removal (EBPR) reactor fed with raw domestic wastewater. Limited filamentous bulking in this study was mainly induced by low DO levels, while other influencing
Keywords:
factors associated with filamentous bulking (F/M, nutrients, and wastewater characteris-
Limited filamentous bulking
tics) were not selective for filamentous bacteria. The optimum scenario for integrated
Dissolved oxygen
nitrogen, phosphorus and COD removal was achieved under limited filamentous bulking
Settleability (SVI)
with an SVI level of 170e200 (associated with a DO of 1.0e1.5 mg/L). The removal effi-
Simultaneous nitrification
ciencies of COD, TP and NHþ 4 eN were 90%, 97% and 92%, respectively. Under these
and denitrification
conditions, the solideliquid separation was practically not affected and sludge loss was
EBPR
never observed. A well-clarified effluent with marginal suspended solids was obtained. The results of this study indicated the feasibility of limited filamentous bulking under low DO as a stimulation of simultaneous nitrification/denitrification for enhancing nutrient removal and effluent quality in an EBPR process. ª 2011 Elsevier Ltd. All rights reserved.
1.
Introduction
Enhanced biological phosphorus removal (EBPR) in activated sludge systems characterized by high removal efficiency, economy, environmentally-friendly operation, and potential phosphorus recovery (Barat and van Loosdrecht, 2006; Martı´ et al., 2010), has become a popular and widespread technology in wastewater treatment plants (WWTPs). However,
filamentous sludge bulking has been reported for the EBPR process (Vaiopoulou et al., 2007) leading to sludge loss and poor solideliquid separation, and therefore result in subsequent upsets and deterioration in removal performance. Abundant previous researches were mainly focused on the study of the prevention, control and modeling of filamentous bulking in various activated sludge systems (Cenens et al., 2000; Martins et al., 2004b; Gulez and de los Reyes, 2009).
* Corresponding author. School of Municipal and Environmental Engineering, Harbin Institute of Technology, Harbin 150090, China. Tel./fax: þ86 451 8628 3003. E-mail addresses: [email protected] (W.-D. Tian), [email protected] (W.-G. Li), [email protected] (M.C.M. van Loosdrecht). 0043-1354/$ e see front matter ª 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.watres.2011.06.034
4878
w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 4 8 7 7 e4 8 8 4
Nomenclature WWTP A/O SBR UCT EBPR A2O BCFS EPS F/M A/V SND
wastewater treatment plant anoxic-oxic sequencing batch reactor university of cape town enhanced biological phosphorus removal anaerobic-anoxic-oxic biologisch-chemische-fosfaat-stikstof verwijdering extracellular polymeric substances food/micro-organisms ratio surface-to-volume simultaneous nitrification and denitrification
Specific (Martins et al., 2003a, b; Wanner et al., 2010) and nonspecific (Liao et al., 2004; Martins et al., 2004a) methods as well as various biological selectors (Van Loosdrecht et al., 1998; Vaiopoulou and Aivasidis, 2008) were developed to prevent and control filamentous bulking. Still the costs and the need for chemicals and operational control are the intractable issues, and the general cost effective and easy control solution has not been adopted by the plant operators. Recently, Peng et al. (2008) advocated an energy-saving method of limited filamentous bulking under low DO condition for the first time using anoxic-oxic (A/O) process and domestic wastewater. Thereafter Guo et al. (2010) developed the energy-saving theory and method of limited filamentous bulking, which minimizes energy consumption by taking advantage of higher oxygen transfer rate obtained under low DO level. However, hitherto the utilization of limited filamentous bulking induced by controlling low DO levels for the EBPR process and its influence on overall process performance were marginally documented. The lower DO concentration not only maintains a favorable anoxic environment but also can be favorable for simultaneous nitrification and denitrification (SND). Many researchers have been attracted by the simultaneous nitrification and denitrification (SND) technique because of its simplified process design and smaller anoxic zone, as well as no requirements of external carbon source and alkalinity while minimizing the need for sludge recycles (Ajay et al., 2006; Fu et al., 2009). The biological and physical explanations for SND are the coexistence of denitrifiers and autotrophic nitrifiers and oxygen gradients within activated sludge flocs caused by the limitation of oxygen diffusion (Guo et al., 2005; Chiu et al., 2007). The oxygen gradients in biological floc lead to an interior anoxic microenvironment, which facilitates SND. The larger floc diameter has advantages to limit relative penetration depth of oxygen in the floc and therefore generate oxygen gradients and microbial process stratification, which has been described in the literatures (Chu et al., 2004; Li and Bishop, 2004; Pe´rez et al., 2005). Andreadakis (1993) showed that microenvironments within the floc for SND was better formed with the floc size of 50e100 mm than 10e70 mm. Pochana and Keller (1999) found that the nitrogen removal efficiency via SND was increased by 31% when the average floc size increased from 40 mm to 80 mm. Some researches assumed that a bit larger floc diameter was likely to promote the SND due to diffusional limitation of oxygen in
ORP BOD COD TN SVI MLSS SRT MLVSS TP SS HRT DO FISH
oxidation-reduction potential biochemical oxygen demand, mg L1 chemical oxygen demand, mg L1 total nitrogen, mg L1 sludge volume index, ml g1 mixed liquid suspended solids, g L1 solid retention time, d mixed liquor volatile suspended solids, g L1 total phosphate, mg L1 suspended solids, mg L1 hydraulic retention time, dissolved oxygen, mg L1 fluorescence in situ hybridization
the floc (Zhu et al., 2007; Guo et al., 2009). Dissolved oxygen is important for development of filamentous microorganism (Martins et al., 2003b, 2004a). Having a limited growth of filamentous bacteria will also achieve a better effluent quality due to the irregular filamentous morphology which gives a better filtering out of suspended particles (Wile´n and Balme´r, 1999; Guo et al., 2009, 2010). The main objective of the present study is therefore to demonstrate limited filamentous bulking under low DO as a stimulation of SND to enhance biological nutrient removal and effluent quality. Experiments were carried out in a benchscale EBPR process, designed according to a BCFS process (Van Loosdrecht et al., 1998). The removal performance of nitrogen, phosphorus and chemical oxygen demand (COD) were investigated at different SVI values in aerobic compartment.
2.
Materials and methods
2.1.
Reactor configuration and experimental setup
Long-term experiments were performed in a bench-scale EBPR process, designed according to a BCFS process, as shown in Fig. 1. The reactor was made of plexiglas with a total working volume of 27 L, which was separated in four functional compartments by removable plastic sheets. The working volume of anaerobic, anaerobic selector, anoxic, anoxic/oxic and aerobic compartment is 5.4 L, 1.2 L, 5.4 L, 9 L and 6 L, respectively. A mechanical mixer was used in nonaerated zones to provide well mixed conditions. Aeration was supplied at the bottom of the aerobic compartments by an air compressor. A clarifier with a working volume of 9 L was used for solideliquid separation. The surface loading of the clarifier was 0.36 m3/m2 h. The flow rates of influent, return sludge, and the two internal recycle flows were controlled by four peristaltic pumps (Lange Z1515-100M, China).
2.2.
Wastewater composition
The bench-scale EBPR reactor was fed with raw domestic sewage. There were no extra chemicals added. The compositions are described as follows. CODcr: 183.5e367 mg/L; NHþ 4 eN:
w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 4 8 7 7 e4 8 8 4
4879
Fig. 1 e Schematic diagram of bench-scale EBPR reactor. (1) Influent tank; (2) feed pump; (3) mechanical mixer; (4) check valve; (5) diffuser; (6) airflow meter; (7) air compressor; (8) return nitrified liquor pump; (9) return denitrified liquor pump; (10) secondary clarifier; (11) effluent; (12) waste sludge; and (13) return sludge pump.
46.3e62.5 mg/L; TN: 48.6e71.5 mg/L; TP 5.3e10.9 mg/L; SS: 151e200 mg/L; pH: 7.0e7.7. The standard deviations of CODcr, NHþ 4 eN, TN and TP of 90 samples are 18.7, 4.2, 4.5 and 0.8, respectively.
2.3.
Experimental operational conditions and procedures
The reactor was inoculated with activated sludge collected from Wenchang municipal wastewater treatment plant (A/O process) of Harbin, PR China, the initial concentration of sludge in the reactor was set at 4 g/L, and the reactor was fed with diluted raw domestic sewage for seven days to obtain constant colonization and accumulation of microorganism. Afterwards, the reactor was operated in a continuous plugflow mode fed with raw domestic sewage. The reactor gradually stabilized after the acclimatization of 35days at the room temperature 22 3 C, DO of 1.5 mg/L, hydraulic retention time (HRT) of anaerobic compartment (1.8 h), anaerobic selector (0.4 h), anoxic compartment (1.8 h), anoxic/oxic compartment (3 h), aerobic compartment (2 h) and solid retention time (SRT) of 15days. During the steady-state periods, the recycling rate of sludge, nitrified liquor and denitrified liquor was set at 1.0, 2.0 and 1.5 time of total influent flow rate, respectively. Then various investigations were conducted with different DO levels in the aerobic compartment.
2.4.
Analytical methods
Ammonia nitrogen (NHþ 4 eN), nitrate nitrogen(NO3 eN), nitrite eN), total phosphorus (TP) chemical oxygen nitrogen (NO 2 demand (COD), mixed liquor suspended solids (MLSS), mixed liquor volatile suspended solids (MLVSS), alkalinity and sludge volume index (SVI) were measured according to the standard methods for the examination of water and wastewater (APHA, 1998). The total nitrogen (TN) concentration was determined with LiquiTOCII (Elementar, Germany). The DO was measured by SG6-ELK SevenGo Pro (Mettler Toledo, Switzerland). The ORP, pH and temperature were measured with HI-8424 pH meter (HANNA, Italy). Periodically microscopic observations of sludge samples from the aerobic compartment were performed with an Olympus IX51 inverted microscope (Tokyo, Japan) and the microscopic analysis was according to the reference manuals (Eikelboom, 2000; Jenkins et al., 2003).
3.
Results
3.1. Occurrence and control of limited filamentous bulking Extensive experiments have been performing for the optimization of process parameters such as volume ratios of interactive functional compartments, various recycling rate, SRT and DO in the past thirteen months. The occurrence of limited filamentous bulking was observed when the volume ratio of anaerobic, anoxic, anoxic/oxic and aerobic compartment was 1:1:1.7:1.1. Herein, DO concentration of the anoxic/oxic zone was constant in the range of 0.3e1.0 mg/L. Limited filamentous bulking is defined as sludge with an SVI of 140e250. The proliferation of filamentous micro-organisms was found by periodic microscopic observations. However, the removal performance of TN and TP adversely enhanced, and COD removal was stable during the period of limited filamentous bulking, in parallel with a case of good sludge settleability. Particularly the poor solideliquid separation and the loss of sludge were never observed. The phenomenon was consistent with the findings of a previous study (Guo et al., 2010). For the further research, limited filamentous bulking induced by DO level was studied. The relationship between settleability (SVI) and DO level under the prerequisite of suitable nutrients (N, P), food/micro-organisms ratio (F/M), wastewater characteristics, pH and temperature, is presented in Fig. 2. There is an expected relationship between DO and SVI. Experimental results showed that it is easy to control the limited filamentous bulking by adjusting aerobic DO levels in this bench-scale plug-flow EBPR reactor. Subsequently, the effect of limited filamentous bulking on the pollutants removal performance was focused at different aerobic DO levels (1.0e1.5, 1.5e2.0 and 2.0e3.0), which will be discussed further below.
3.2. Overall performance of nutrient removal under limited filamentous bulking 3.2.1.
COD removal
The organic loading rate of the reactor was in the range of 0.20e0.39 kgCOD kgMLSS1 d1, which in general is supposed not to lead to filamentous bulking (Chudoba et al., 1974; Wanner et al., 2010). Fig. 2(a) presented the COD removal efficiency under limited filamentous bulking in the bench-
4880
w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 4 8 7 7 e4 8 8 4
D Fig. 2 e The removal efficiencies of COD, NHD 4 eN, TP at different SVI periods (a: COD removal; b: NH4 eN removal; c: TP removal). The nutrient removal efficiencies were compared under the condition of normal settleability and limited filamentous bulking.
w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 4 8 7 7 e4 8 8 4
scale EBPR reactor. The observed trends of COD removal efficiencies and effluent COD are very similar even though the SVI increased from 100 to 200, which means the COD removal performance is not affected by limited filamentous bulking. Previous researches have showed that COD removal efficiency was slightly interfered under filamentous bulking caused by overpopulation of Haliscomenobacter hydrossis (Kotay et al., 2010). The influent COD concentration had an average value of 260 mg/L ranging from 184 to 367 mg/L, whereas effluent concentrations had an average value of 30 mg/L fluctuating between 17 and 40 mg/L. Total COD removal efficiency with had an average value of 90% ranging from 82% to 95% during the steady-state period.
3.2.2.
Nitrogen removal
The NHþ 4 eN removal efficiency under limited filamentous bulking was evaluated, as illustrated in Fig. 2(b) and Fig. 3. For clear comparison, two vertical dotted lines were used to divide the figure into three sections based on the different SVI values. The influent NHþ 4 eN concentrations with an average value of 54 mg/L fluctuated between 46 and 63 mg/L, while effluent concentrations with an average value of 2.8, 3.7 and 4.7 at respective SVI values. From the Fig. 2(b), it could be readily observed that the trends of NHþ 4 eN removal efficiency slightly declined with the increasing SVI values, the average removal efficiency was 95%, 94% and 92% at respective SVI values. Denitrifying nitrogen removal efficiency remained stable under different DO levels, however, the TN removal efficiency enhanced since the ongoing occurrence of simultaneous nitrification and denitrification (SND) with SVI scale of 170e200 (associated with a DO of 1.0e1.5 mg/L) in aerobic compartment, which could obviously be observed by comparing the section of SND in Fig. 3. A similar result was reported by (Third et al., 2003) who also observed that SND increased during aerobic famine period in an SBR at low DO (<2 mg/L). From Fig. 3 it could be apparently found that SND never occurred at DO of 1.5e2.0 and 2.0e3.0 mg/L, this result is consistent with the record which reported that SND was not
Fig. 3 e Nitrogen mass balance at different SVI periods. The corresponding distribution of total nitrogen under the condition of normal settleability and limited filamentous bulking was displayed respectively.
4881
able to be achieved when DO was above 1.5 mg/L (Guo et al., 2010). Moreover, the higher concentration of nitrate in aerobic compartment at DO of 1.5e2.0 and 2.0e3.0 mg/L led to the upset of foregoing anoxic and anaerobic environment, and hence result in the lower phosphorus removal efficiency. Therefore, the preferable DO concentration of aerobic compartment for SND nitrogen removal was 1.0e1.5 mg/L in this bench-scale EBPR reactor.
3.2.3.
TP removal
The TP removal efficiency under limited filamentous bulking was investigated, as shown in Fig. 2(c), which was divided by vertical dotted lines into three sections according to the respective SVI periods. It could be distinctly observed that the TP removal efficiency steadily enhanced with increasing SVI, the lowest, average and highest removal efficiency was recorded as 74%, 90% and 97% with an SVI of 100e140,140e170 and 170e200, respectively. The influent TP had a mean value of 7.7 mg/L ranged between 5.5 and 11 mg/L, whereas effluent concentrations had a mean value of 2.23, 0.84 and 0.25 mg/L, correspondingly. Constant lower removal efficiency occurred at an SVI level of 100e140 (associated with a DO of 2e3 mg/L), which was mainly attributed to the chain of reactions in the reactor caused by DO level. Since the recycling of nitrified liquor and denitrified liquor were set for denitrification and denitrifying dephosphatation and for the reutilization of BOD, respectively. Therefore, strict anaerobic environment could be disturbed if the concentration of nitrate exceed 0.1 mg N/L in the anaerobic compartment (Van Loosdrecht et al., 1998), and the trespass of a higher DO in anoxic compartment would interfere the denitrifying dephosphatation. Reversely, the higher TP removal efficiency was obtained at higher SVI values of 140e170 and 170e200 (i.e. limited filamentous bulking). Moreover, low DO level can be favorable for maintaining denitrifying dephosphatation and indirectly upholding a comfortable anaerobic environment for efficient phosphorus release and associated substrate uptake. Notwithstanding, to guarantee the solideliquid separation in the secondary clarifier and a satisfactory effluent, DO level is not as lower as better and therefore an optimum settleability (SVI) should be dominated by moderate DO level based on the critical point(1.0e1.5 mg/L in this study). In particularly, phosphorus removal should be integrated with nitrification and denitrification, while DO level is a critical operating parameter of these procedures. Furthermore, the optimal SVI under limited filamentous bulking induced by DO level isn’t an absolute value, which still depends on the configuration of the process and other uncertain factors in practice (Eikelboom, 2000).
4.
Discussion
4.1.
Microscopic observation aspect
The filament index (FI) is a measure of the number of filamentous micro-organisms in activated sludge, which is established by comparing the microscopic image of the sludge with a series of reference photograph of the five FI classes at a low magnification. The predominance of filaments was mainly distinguished according to intrinsic morphological
4882
w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 4 8 7 7 e4 8 8 4
characteristics of the filamentous micro-organisms viz. mobility, branching, filament shape, filament length, attached growth, septa or transverse walls, cell diameter and sheath etc. (Eikelboom, 2000). The observations in this study showed that FI was maintained between 1 and 2 under limited filamentous bulking which means the effect of the filaments on the settling velocity of the sludge is limited (Eikelboom, 2000). Sludge sample under limited filamentous bulking was staining with DAPI fluorochrome and then observed using the fluorescence microscope, the typical epifluorescence micrographs are showed in Fig. 4. The dominant filamentous bacteria was identified as H. hydrossis due to its needle-like appearance in a pin cushion with straight filaments protruding from the flocs, as presented in Fig. 4(a), and which was within the floc structure. In addition, minor S. natans characterized by
straight or smoothly curved filaments with no/tree-like false branching, round-ended and rod shaped cells and clearly visible cell septa with indentations was also recognized as the secondary filamentous bacteria, as shown in Fig. 4(b). S. natans can radiate outward from the floc surface into the bulk solution and results in a high SVI by inter-floc bridging. In this study, the presence of H. hydrossis was mainly caused by low DO, while S. natans was likely to be caused by low DO and long retention time of sludge in secondary clarifier. Nevertheless, H. hydrossis population is usually limited present in domestic treatment plants and it can develop en masse in industrial plants where many easily biodegradable compounds are present in the influent (Eikelboom, 2000). Indeed, low DO tends to cause filamentous bulking by S. natans, type 1701 and H. hydrossis (Jenkins et al., 2003). Moreover, a few amount of Eikelboom Type 0041 with much attached growth was also observed which was beneficial as the backbone structure for the flocs in Fig. 4(c). However, M. parvicella and Type 021N were apparently not observed. The results of this study are in line with the research of Gaval and Pernelle, 2003, in which the dominant filamentous bacteria identifying by morphological criteria and FISH were H. hydrossis and S. natans under respective oxygen deficiency condition, and small Type 021N was also observed. But Guo et al., 2010 demonstrated that Eikelboom Type 0041 was the dominant filamentous bacterium and few Type 021N and M. parvicella were also detected, however, H. hydrossis and S. natans were never observed.
4.2.
Fig. 4 e Epifluorescence micrographs of DAPI stained filament of (a) H. hydrossis, (b) S. natans and (c) Eikelboom Type 0041. The length of the bars corresponds to: a and b 10 mm; c 20 mm. The filamentous bacterium was observed under limited filamentous bulking.
Limited filamentous bulking
It has been hypothesized that filamentous micro-organisms serve as a backbone for the flocs to provide more binding sites for the attachment of free cells or smaller aggregates by extracellular polymeric substances (EPS) (Cenens et al., 2000; Liao et al., 2011). The principle of limited filamentous bulking is to keep a moderate imbalance between floc-forming and filamentous bacteria, which has a slight advantage for filamentous bacteria. This allows a better enmeshing of tiny particles or free flocs, and the larger floc diameter under limited filamentous bulking implies diffusion resistance of oxygen inside the flocs is larger which facilitate SND (Martins et al., 2004a), although filamentous bacteria extend from the flocs make flocs a bit incompact and porous, but denser. Moreover, filamentous micro-organisms have the competitive advantages to access organic substrate based on A/V hypothesis (Jenkins et al., 2003) and diffusion-based selection (Martins et al., 2004a, 2010; Lou and de los Reyes, 2008) since the intrinsic morphological property (preferential growth of one or two directions) facilitates a large contact area and an easy penetration rate. In addition, the kinetic selection hypothesis (Chudoba et al., 1974) can also well explain it because of the lower affinity constant of filamentous bacteria than floc-forming bacteria that low DO concentrations favor the growth of filamentous bacteria. If we can balance the advantages and disadvantages of filamentous bacteria under low DO condition, such as limited filamentous bulking for a good effluent, energy-saving undoubtedly been achieved. The experiments in this study are according to above theory. Although general believe is that the activated sludge with more filamentous bacteria is negative for wastewater
w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 4 8 7 7 e4 8 8 4
operations (Jin et al., 2003), it was found that the solideliquid separation was practically not affected and no sludge loss was observed even with the highest SVI (200) condition under limited filamentous bulking induced by DO level. In the case of bio-P sludge, the density of the flocs might however compensate for the effect of the larger number of filaments (Eikelboom et al., 1998), and less production of sludge attributing to the characteristic of denitrifying dephosphatation in the EBPR reactor (Beun et al., 2000). In addition, a stable wellclarified effluent with marginal suspended solids (<8 mg/L) was obtained, which can be attributed to morphological characteristics of filamentous bacteria (i.e. enmeshment mechanism). The results of this study also show that limited filamentous bulking is repeatable and controllable. Therefore, the technique of limited filamentous bulking could be an alternative solution for enhancing nutrient removal and effluent quality though the control strategy and engineering capital in practice are still in question.
5.
Conclusions
This study investigated the integrated nitrogen, phosphorus and COD removal performance in a bench-scale plug-flow EBPR reactor under limited filamentous bulking. Experimental work lasted for about 200 days to study the removal performance of the EBPR process at different SVI periods. The results show that limited filamentous bulking induced by low DO level could enhance biological nutrient removal and achieve a well-clarified effluent. Moreover, low DO concentration associated to limited filamentous bulking is of importance for energy-saving. The optimum scenario for integrated nitrogen, phosphorus and COD removal was achieved under limited filamentous bulking at an SVI level of 170e200(associated with a DO of 1.0e1.5 mg/L), and the corresponding respective removal efficiencies of COD, TP, NHþ 4 eN were 90%, 97% and 92%. In addition, the solideliquid separation was practically not affected and no sludge loss was observed even at the highest SVI (200) condition, which was likely to attribute to the heavy flocs of bio-P sludge and less production of sludge of denitrifying dephosphatation.
Acknowledgments The authors gratefully acknowledge the financial support provided by National Water Pollution Control and Management Technology Major Projects of China (No. 2009ZX07317-008).
references
APHA, 1998. Standard Methods for Examination of Water and Wastewater, 20th ed. American Public Health Association, Washington, DC. Ajay, P., Jesse, Z., George, N., 2006. Simultaneous carbon, nitrogen and phosphorous removal from municipal wastewater in
4883
a circulating fluidized bed bioreactor. Chemosphere 65 (7), 1103e1112. Andreadakis, A.D., 1993. Physical and chemical properties lf activated sludge flocs. Water Research 27 (11), 1707e1714. Beun, J.J., Paletta, F., Van Loosdrecht, M.C.M., Heijnen, J.J., 2000. Stoichiometry and kinetics of poly B hydroxybutyrate metabolism under denitrifying conditions in activated sludge cultures. Biotechnology and Bioengineering 67, 379e389. Barat, R., van Loosdrecht, M.C.M., 2006. Potential phosphorus recovery in a WWTP with the BCFS process: interactions with the biological process. Water Research 40, 3507e3516. Chiu, Y.C., Lee, L.L., Chang, C.N., Chao, A.C., 2007. Control of carbon and ammonium ratio for simultaneous nitrification and denitrification in a sequencing batch bioreactor. International Biodeterioration and Biodegradation 59, 1e7. Chu, K.H., van Veldhuizen, H.M., van Loosdrecht, M.C.M., 2004. Respirometric measurement of kinetic parameters: effect of activated sludge floc size. Water Science and Technology 48 (8), 61e68. Chudoba, J., Blaha, J., Madera, V., 1974. Control of activated sludge filamentous bulking eIII. Effect of sludge loading. Water Research 8 (4), 231e237. Cenens, C., Smets, I., Van Impe, J., 2000. Modeling the competition between floc-forming and filamentous bacteria in activated sludge waste water treatment systems. Part II. A prototype mathematical model based on kinetic selection and filamentous backbone theory. Water Research 34, 2535e2541. Eikelboom, D.H., Andreadakis, A., Andreasen, K., 1998. Survey of the filamentous population in nutrient removal plants in four European countries. Water Science and Technology 37 (4/5), 281e290. Eikelboom, D.H., 2000. Process Control of Activated Sludge Plants by Microscopic Investigation. IWA Publishing, London, UK. Fu, Z., Yang, F., An, Y., Xue, Y., 2009. Simultaneous nitrification and denitrification coupled with phosphorus removal in an modified anoxic/oxic-membrane bioreactor (A/O-MBR). Biochemical Engineering Journal 43, 191e196. Gaval, G., Pernelle, J.J., 2003. Impact of the repetition of oxygen deficiencies on the filamentous bacteria proliferation in activated sludge. Water Research 37, 1991e2000. Gulez, G., de los Reyes, F.L., 2009. Multiple approaches to assess filamentous growth in activated sludge under different carbon source conditions. Journal of Applied Microbiology 106, 682e691. Guo, H.Y., Zhou, J.T., Su, J., et al., 2005. Integration of nitrification and denitrification in airlift bioreactor. Biochemical Engineering Journal 23, 57e62. Guo, J.H., Peng, Y.Z., Wang, S.Y., Zheng, Y.N., Huang, H.J., Wang, Z.W., 2009. Long term effect of dissolved oxygen on partial nitrification performance and microbial community structure. Bioresource Technology 100, 2796e2802. Guo, J.H., Peng, Y.Z., Peng, C.Y., Wang, S.Y., Chen, Y., Huang, H.J., Sun, Z.R., 2010. Energy saving achieved by limited filamentous bulking sludge under low dissolved oxygen. Bioresource Technology 101, 1120e1126. Jenkins, D., Richard, M.G., Daigger, G.T., 2003. Manual on the Cause and Control of Activated Sludge Bulking, Foaming, and Other Solids Separation Problems, third ed. Lewis Publishers, NY, USA. Jin, B., Wile´n, B.M., Lant, P., 2003. A comprehensive insight into floc characteristics and their impact on compressibility and settleability of activated sludge. Chemical Engineering Journal 95 (1e3), 221e234. Kotay, S.M., Datta, T., Choi, J., Goel, R., 2010. Biocontrol of biomass bulking caused by Haliscomenobacter hydrossis using a newly isolated lytic bacteriophage. Water Research. doi:10.1016/j. watres.2010.08.038.
4884
w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 4 8 7 7 e4 8 8 4
Li, B., Bishop, L., 2004. Micro-profiles of activated sludge floc determined using microelectrodes. Water Research 38 (5), 1248e1258. Liao, J., Lou, I., delosReyes, F.L., 2004. Relationship of speciesspecific filament levels to filamentous bulking in activated sludge. Applied and Environmental Microbiology 70 (4), 2420e2428. Liao, B.Q., Lin, H.J., Langevin, S.P., Gao, W.J., Leppard, G.G., 2010. Effects of temperature and dissolved oxygen on sludge properties and their role in bioflocculation and settling. Water Research 45 (2), 509e520. Lou, I., de los Reyes, F.L., 2008. Clarifying the roles of kinetics and diffusion in activated sludge filamentous bulking. Biotechnology and Bioengineering 101, 327e336. Martins, A.M.P., Heijnen, J.J., van Loosdrecht, M.C.M., 2003a. Effect of feeding pattern and storage on the sludge settleability under aerobic conditions. Water Research 37 (11), 2555e2570. Martins, A.M.P., Heijnen, J.J., Van Loosdrecht, M.C.M., 2003b. Effect of dissolved oxygen concentration on sludge settleability. Applied Microbiology Biotechnology 62 (5e6), 586e593. Martins, A.M.P., Pagilla, K., Van Loosdrecht, M.C.M., 2004a. Filamentous bulking sludge e a critical review. Water Research 38 (4), 793e817. Martins, A.M.P., Picioreanu, C., Heijnen, J.J., van Loosdrecht, M.C. M., 2004b. Three-dimensional dual-morphotype species modeling of activated sludge flocs. Environmental Science and Technology 38 (21), 5632e5641. Martins, A.M.P., Karahan, O., van Loosdrecht, M.C.M., 2010. Effect of polymeric substrate on sludge settleability. Water Research. doi:10.1016/j.watres.2010.07.055. Martı´, N., Pastor, L., Bouzas, A., Ferrer, J., Seco, A., 2010. Phosphorus recovery by struvite crystallization in WWTPs: influence of the sludge treatment line operation. Water Research 44, 2371e2379.
Peng, Y.Z., Guo, J.H., Wang, S.Y., Chen, Y., 2008. Energy saving achieved by limited filamentous bulking under low dissolved oxygen: derivation, originality and theoretical basis. Journal of Environmental Sciences 29 (12), 3342e3347. Pe´rez, J., Picioreanu, C., van Loosdrecht, M.C.M., 2005. Modeling biofilm and floc diffusion processes based on analytical solution of reaction-diffusion equations. Water Research 39, 1311e1323. Pochana, K., Keller, J., 1999. Study of factors affecting simultaneous nitrification and denitrification (SND). Water Science and Technology 39 (6), 61e68. Third, K.A., Burnett, N., Cord-Ruwisch, R., 2003. Simultaneous nitrification and denitrification using stored substrate (PHB) as the electron donor in an SBR. Biotechnology and Bioengineering 83, 706e720. Van Loosdrecht, M.C.M., Brandse, F.A., de Vries, A.C., 1998. Upgrading of waste water treatment processes for integrated nutrient removal e the BCFS process. Water Science and Technology 37 (9), 209e217. Vaiopoulou, E., Melidis, P., Aivasidis, A., 2007. Growth of filamentous bacteria in an enhanced biological phosphorus removal system. Desalination 213, 288e296. Vaiopoulou, E., Aivasidis, A., 2008. A modified UCT method for biological nutrient removal: configuration and performance. Chemosphere 72, 1062e1068. Wanner, J., Kragelund, C., Nielsen, P., 2010. In: Seviour, R., Nielsen, P.H. (Eds.), Microbial Ecology of Activated Sludge. IWA Publishing, London. Wile´n, B.M., Balme´r, P., 1999. The effect of dissolved oxygen concentration on the structure, size and size distribution of activated sludge flocs. Water Research 33 (2), 391e400. Zhu, G.B., Peng, Y.Z., Wu, S.Y., Wang, S.Y., Xu, S.W., 2007. Simultaneous nitrification and denitrification in step feeding biological nitrogen removal process. Journal of Environmental Science 19, 1043e1048.
Available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/watres
Limited filamentous bulking in order to enhance integrated nutrient removal and effluent quality Wen-De Tian a,b, Wei-Guang Li a,b,*, Hui Zhang a,b, Xiao-Rong Kang a,b, Mark C.M. van Loosdrecht c,d a
School of Municipal and Environmental Engineering, Harbin Institute of Technology, Harbin 150090, China State Key Laboratory of Urban Water Resource Environment, Harbin Institute of Technology, Harbin 150090, China c Department of Biotechnology, Delft University of Technology, Julianalaan 67, 2628 BC Delft, The Netherlands d KWR Watercycle Research Institute, Groningenhaven 7, 3422 PE Nieuwegein, The Netherlands b
article info
abstract
Article history:
Limited filamentous bulking has been proposed as a means to enhance floc size and make
Received 23 December 2010
conditions more favorable for simultaneous nitrification/Denitrification (SND). Moreover
Received in revised form
a slightly heightened SVI is supposed to increase the removal of small particulates in the
23 June 2011
clarifier. Integrated nitrogen, phosphorus and COD removal performance under limited
Accepted 25 June 2011
filamentous bulking was investigated using a bench-scale plug-flow enhanced biological
Available online 6 July 2011
phosphorus removal (EBPR) reactor fed with raw domestic wastewater. Limited filamentous bulking in this study was mainly induced by low DO levels, while other influencing
Keywords:
factors associated with filamentous bulking (F/M, nutrients, and wastewater characteris-
Limited filamentous bulking
tics) were not selective for filamentous bacteria. The optimum scenario for integrated
Dissolved oxygen
nitrogen, phosphorus and COD removal was achieved under limited filamentous bulking
Settleability (SVI)
with an SVI level of 170e200 (associated with a DO of 1.0e1.5 mg/L). The removal effi-
Simultaneous nitrification
ciencies of COD, TP and NHþ 4 eN were 90%, 97% and 92%, respectively. Under these
and denitrification
conditions, the solideliquid separation was practically not affected and sludge loss was
EBPR
never observed. A well-clarified effluent with marginal suspended solids was obtained. The results of this study indicated the feasibility of limited filamentous bulking under low DO as a stimulation of simultaneous nitrification/denitrification for enhancing nutrient removal and effluent quality in an EBPR process. ª 2011 Elsevier Ltd. All rights reserved.
1.
Introduction
Enhanced biological phosphorus removal (EBPR) in activated sludge systems characterized by high removal efficiency, economy, environmentally-friendly operation, and potential phosphorus recovery (Barat and van Loosdrecht, 2006; Martı´ et al., 2010), has become a popular and widespread technology in wastewater treatment plants (WWTPs). However,
filamentous sludge bulking has been reported for the EBPR process (Vaiopoulou et al., 2007) leading to sludge loss and poor solideliquid separation, and therefore result in subsequent upsets and deterioration in removal performance. Abundant previous researches were mainly focused on the study of the prevention, control and modeling of filamentous bulking in various activated sludge systems (Cenens et al., 2000; Martins et al., 2004b; Gulez and de los Reyes, 2009).
* Corresponding author. School of Municipal and Environmental Engineering, Harbin Institute of Technology, Harbin 150090, China. Tel./fax: þ86 451 8628 3003. E-mail addresses: [email protected] (W.-D. Tian), [email protected] (W.-G. Li), [email protected] (M.C.M. van Loosdrecht). 0043-1354/$ e see front matter ª 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.watres.2011.06.034
4878
w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 4 8 7 7 e4 8 8 4
Nomenclature WWTP A/O SBR UCT EBPR A2O BCFS EPS F/M A/V SND
wastewater treatment plant anoxic-oxic sequencing batch reactor university of cape town enhanced biological phosphorus removal anaerobic-anoxic-oxic biologisch-chemische-fosfaat-stikstof verwijdering extracellular polymeric substances food/micro-organisms ratio surface-to-volume simultaneous nitrification and denitrification
Specific (Martins et al., 2003a, b; Wanner et al., 2010) and nonspecific (Liao et al., 2004; Martins et al., 2004a) methods as well as various biological selectors (Van Loosdrecht et al., 1998; Vaiopoulou and Aivasidis, 2008) were developed to prevent and control filamentous bulking. Still the costs and the need for chemicals and operational control are the intractable issues, and the general cost effective and easy control solution has not been adopted by the plant operators. Recently, Peng et al. (2008) advocated an energy-saving method of limited filamentous bulking under low DO condition for the first time using anoxic-oxic (A/O) process and domestic wastewater. Thereafter Guo et al. (2010) developed the energy-saving theory and method of limited filamentous bulking, which minimizes energy consumption by taking advantage of higher oxygen transfer rate obtained under low DO level. However, hitherto the utilization of limited filamentous bulking induced by controlling low DO levels for the EBPR process and its influence on overall process performance were marginally documented. The lower DO concentration not only maintains a favorable anoxic environment but also can be favorable for simultaneous nitrification and denitrification (SND). Many researchers have been attracted by the simultaneous nitrification and denitrification (SND) technique because of its simplified process design and smaller anoxic zone, as well as no requirements of external carbon source and alkalinity while minimizing the need for sludge recycles (Ajay et al., 2006; Fu et al., 2009). The biological and physical explanations for SND are the coexistence of denitrifiers and autotrophic nitrifiers and oxygen gradients within activated sludge flocs caused by the limitation of oxygen diffusion (Guo et al., 2005; Chiu et al., 2007). The oxygen gradients in biological floc lead to an interior anoxic microenvironment, which facilitates SND. The larger floc diameter has advantages to limit relative penetration depth of oxygen in the floc and therefore generate oxygen gradients and microbial process stratification, which has been described in the literatures (Chu et al., 2004; Li and Bishop, 2004; Pe´rez et al., 2005). Andreadakis (1993) showed that microenvironments within the floc for SND was better formed with the floc size of 50e100 mm than 10e70 mm. Pochana and Keller (1999) found that the nitrogen removal efficiency via SND was increased by 31% when the average floc size increased from 40 mm to 80 mm. Some researches assumed that a bit larger floc diameter was likely to promote the SND due to diffusional limitation of oxygen in
ORP BOD COD TN SVI MLSS SRT MLVSS TP SS HRT DO FISH
oxidation-reduction potential biochemical oxygen demand, mg L1 chemical oxygen demand, mg L1 total nitrogen, mg L1 sludge volume index, ml g1 mixed liquid suspended solids, g L1 solid retention time, d mixed liquor volatile suspended solids, g L1 total phosphate, mg L1 suspended solids, mg L1 hydraulic retention time, dissolved oxygen, mg L1 fluorescence in situ hybridization
the floc (Zhu et al., 2007; Guo et al., 2009). Dissolved oxygen is important for development of filamentous microorganism (Martins et al., 2003b, 2004a). Having a limited growth of filamentous bacteria will also achieve a better effluent quality due to the irregular filamentous morphology which gives a better filtering out of suspended particles (Wile´n and Balme´r, 1999; Guo et al., 2009, 2010). The main objective of the present study is therefore to demonstrate limited filamentous bulking under low DO as a stimulation of SND to enhance biological nutrient removal and effluent quality. Experiments were carried out in a benchscale EBPR process, designed according to a BCFS process (Van Loosdrecht et al., 1998). The removal performance of nitrogen, phosphorus and chemical oxygen demand (COD) were investigated at different SVI values in aerobic compartment.
2.
Materials and methods
2.1.
Reactor configuration and experimental setup
Long-term experiments were performed in a bench-scale EBPR process, designed according to a BCFS process, as shown in Fig. 1. The reactor was made of plexiglas with a total working volume of 27 L, which was separated in four functional compartments by removable plastic sheets. The working volume of anaerobic, anaerobic selector, anoxic, anoxic/oxic and aerobic compartment is 5.4 L, 1.2 L, 5.4 L, 9 L and 6 L, respectively. A mechanical mixer was used in nonaerated zones to provide well mixed conditions. Aeration was supplied at the bottom of the aerobic compartments by an air compressor. A clarifier with a working volume of 9 L was used for solideliquid separation. The surface loading of the clarifier was 0.36 m3/m2 h. The flow rates of influent, return sludge, and the two internal recycle flows were controlled by four peristaltic pumps (Lange Z1515-100M, China).
2.2.
Wastewater composition
The bench-scale EBPR reactor was fed with raw domestic sewage. There were no extra chemicals added. The compositions are described as follows. CODcr: 183.5e367 mg/L; NHþ 4 eN:
w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 4 8 7 7 e4 8 8 4
4879
Fig. 1 e Schematic diagram of bench-scale EBPR reactor. (1) Influent tank; (2) feed pump; (3) mechanical mixer; (4) check valve; (5) diffuser; (6) airflow meter; (7) air compressor; (8) return nitrified liquor pump; (9) return denitrified liquor pump; (10) secondary clarifier; (11) effluent; (12) waste sludge; and (13) return sludge pump.
46.3e62.5 mg/L; TN: 48.6e71.5 mg/L; TP 5.3e10.9 mg/L; SS: 151e200 mg/L; pH: 7.0e7.7. The standard deviations of CODcr, NHþ 4 eN, TN and TP of 90 samples are 18.7, 4.2, 4.5 and 0.8, respectively.
2.3.
Experimental operational conditions and procedures
The reactor was inoculated with activated sludge collected from Wenchang municipal wastewater treatment plant (A/O process) of Harbin, PR China, the initial concentration of sludge in the reactor was set at 4 g/L, and the reactor was fed with diluted raw domestic sewage for seven days to obtain constant colonization and accumulation of microorganism. Afterwards, the reactor was operated in a continuous plugflow mode fed with raw domestic sewage. The reactor gradually stabilized after the acclimatization of 35days at the room temperature 22 3 C, DO of 1.5 mg/L, hydraulic retention time (HRT) of anaerobic compartment (1.8 h), anaerobic selector (0.4 h), anoxic compartment (1.8 h), anoxic/oxic compartment (3 h), aerobic compartment (2 h) and solid retention time (SRT) of 15days. During the steady-state periods, the recycling rate of sludge, nitrified liquor and denitrified liquor was set at 1.0, 2.0 and 1.5 time of total influent flow rate, respectively. Then various investigations were conducted with different DO levels in the aerobic compartment.
2.4.
Analytical methods
Ammonia nitrogen (NHþ 4 eN), nitrate nitrogen(NO3 eN), nitrite eN), total phosphorus (TP) chemical oxygen nitrogen (NO 2 demand (COD), mixed liquor suspended solids (MLSS), mixed liquor volatile suspended solids (MLVSS), alkalinity and sludge volume index (SVI) were measured according to the standard methods for the examination of water and wastewater (APHA, 1998). The total nitrogen (TN) concentration was determined with LiquiTOCII (Elementar, Germany). The DO was measured by SG6-ELK SevenGo Pro (Mettler Toledo, Switzerland). The ORP, pH and temperature were measured with HI-8424 pH meter (HANNA, Italy). Periodically microscopic observations of sludge samples from the aerobic compartment were performed with an Olympus IX51 inverted microscope (Tokyo, Japan) and the microscopic analysis was according to the reference manuals (Eikelboom, 2000; Jenkins et al., 2003).
3.
Results
3.1. Occurrence and control of limited filamentous bulking Extensive experiments have been performing for the optimization of process parameters such as volume ratios of interactive functional compartments, various recycling rate, SRT and DO in the past thirteen months. The occurrence of limited filamentous bulking was observed when the volume ratio of anaerobic, anoxic, anoxic/oxic and aerobic compartment was 1:1:1.7:1.1. Herein, DO concentration of the anoxic/oxic zone was constant in the range of 0.3e1.0 mg/L. Limited filamentous bulking is defined as sludge with an SVI of 140e250. The proliferation of filamentous micro-organisms was found by periodic microscopic observations. However, the removal performance of TN and TP adversely enhanced, and COD removal was stable during the period of limited filamentous bulking, in parallel with a case of good sludge settleability. Particularly the poor solideliquid separation and the loss of sludge were never observed. The phenomenon was consistent with the findings of a previous study (Guo et al., 2010). For the further research, limited filamentous bulking induced by DO level was studied. The relationship between settleability (SVI) and DO level under the prerequisite of suitable nutrients (N, P), food/micro-organisms ratio (F/M), wastewater characteristics, pH and temperature, is presented in Fig. 2. There is an expected relationship between DO and SVI. Experimental results showed that it is easy to control the limited filamentous bulking by adjusting aerobic DO levels in this bench-scale plug-flow EBPR reactor. Subsequently, the effect of limited filamentous bulking on the pollutants removal performance was focused at different aerobic DO levels (1.0e1.5, 1.5e2.0 and 2.0e3.0), which will be discussed further below.
3.2. Overall performance of nutrient removal under limited filamentous bulking 3.2.1.
COD removal
The organic loading rate of the reactor was in the range of 0.20e0.39 kgCOD kgMLSS1 d1, which in general is supposed not to lead to filamentous bulking (Chudoba et al., 1974; Wanner et al., 2010). Fig. 2(a) presented the COD removal efficiency under limited filamentous bulking in the bench-
4880
w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 4 8 7 7 e4 8 8 4
D Fig. 2 e The removal efficiencies of COD, NHD 4 eN, TP at different SVI periods (a: COD removal; b: NH4 eN removal; c: TP removal). The nutrient removal efficiencies were compared under the condition of normal settleability and limited filamentous bulking.
w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 4 8 7 7 e4 8 8 4
scale EBPR reactor. The observed trends of COD removal efficiencies and effluent COD are very similar even though the SVI increased from 100 to 200, which means the COD removal performance is not affected by limited filamentous bulking. Previous researches have showed that COD removal efficiency was slightly interfered under filamentous bulking caused by overpopulation of Haliscomenobacter hydrossis (Kotay et al., 2010). The influent COD concentration had an average value of 260 mg/L ranging from 184 to 367 mg/L, whereas effluent concentrations had an average value of 30 mg/L fluctuating between 17 and 40 mg/L. Total COD removal efficiency with had an average value of 90% ranging from 82% to 95% during the steady-state period.
3.2.2.
Nitrogen removal
The NHþ 4 eN removal efficiency under limited filamentous bulking was evaluated, as illustrated in Fig. 2(b) and Fig. 3. For clear comparison, two vertical dotted lines were used to divide the figure into three sections based on the different SVI values. The influent NHþ 4 eN concentrations with an average value of 54 mg/L fluctuated between 46 and 63 mg/L, while effluent concentrations with an average value of 2.8, 3.7 and 4.7 at respective SVI values. From the Fig. 2(b), it could be readily observed that the trends of NHþ 4 eN removal efficiency slightly declined with the increasing SVI values, the average removal efficiency was 95%, 94% and 92% at respective SVI values. Denitrifying nitrogen removal efficiency remained stable under different DO levels, however, the TN removal efficiency enhanced since the ongoing occurrence of simultaneous nitrification and denitrification (SND) with SVI scale of 170e200 (associated with a DO of 1.0e1.5 mg/L) in aerobic compartment, which could obviously be observed by comparing the section of SND in Fig. 3. A similar result was reported by (Third et al., 2003) who also observed that SND increased during aerobic famine period in an SBR at low DO (<2 mg/L). From Fig. 3 it could be apparently found that SND never occurred at DO of 1.5e2.0 and 2.0e3.0 mg/L, this result is consistent with the record which reported that SND was not
Fig. 3 e Nitrogen mass balance at different SVI periods. The corresponding distribution of total nitrogen under the condition of normal settleability and limited filamentous bulking was displayed respectively.
4881
able to be achieved when DO was above 1.5 mg/L (Guo et al., 2010). Moreover, the higher concentration of nitrate in aerobic compartment at DO of 1.5e2.0 and 2.0e3.0 mg/L led to the upset of foregoing anoxic and anaerobic environment, and hence result in the lower phosphorus removal efficiency. Therefore, the preferable DO concentration of aerobic compartment for SND nitrogen removal was 1.0e1.5 mg/L in this bench-scale EBPR reactor.
3.2.3.
TP removal
The TP removal efficiency under limited filamentous bulking was investigated, as shown in Fig. 2(c), which was divided by vertical dotted lines into three sections according to the respective SVI periods. It could be distinctly observed that the TP removal efficiency steadily enhanced with increasing SVI, the lowest, average and highest removal efficiency was recorded as 74%, 90% and 97% with an SVI of 100e140,140e170 and 170e200, respectively. The influent TP had a mean value of 7.7 mg/L ranged between 5.5 and 11 mg/L, whereas effluent concentrations had a mean value of 2.23, 0.84 and 0.25 mg/L, correspondingly. Constant lower removal efficiency occurred at an SVI level of 100e140 (associated with a DO of 2e3 mg/L), which was mainly attributed to the chain of reactions in the reactor caused by DO level. Since the recycling of nitrified liquor and denitrified liquor were set for denitrification and denitrifying dephosphatation and for the reutilization of BOD, respectively. Therefore, strict anaerobic environment could be disturbed if the concentration of nitrate exceed 0.1 mg N/L in the anaerobic compartment (Van Loosdrecht et al., 1998), and the trespass of a higher DO in anoxic compartment would interfere the denitrifying dephosphatation. Reversely, the higher TP removal efficiency was obtained at higher SVI values of 140e170 and 170e200 (i.e. limited filamentous bulking). Moreover, low DO level can be favorable for maintaining denitrifying dephosphatation and indirectly upholding a comfortable anaerobic environment for efficient phosphorus release and associated substrate uptake. Notwithstanding, to guarantee the solideliquid separation in the secondary clarifier and a satisfactory effluent, DO level is not as lower as better and therefore an optimum settleability (SVI) should be dominated by moderate DO level based on the critical point(1.0e1.5 mg/L in this study). In particularly, phosphorus removal should be integrated with nitrification and denitrification, while DO level is a critical operating parameter of these procedures. Furthermore, the optimal SVI under limited filamentous bulking induced by DO level isn’t an absolute value, which still depends on the configuration of the process and other uncertain factors in practice (Eikelboom, 2000).
4.
Discussion
4.1.
Microscopic observation aspect
The filament index (FI) is a measure of the number of filamentous micro-organisms in activated sludge, which is established by comparing the microscopic image of the sludge with a series of reference photograph of the five FI classes at a low magnification. The predominance of filaments was mainly distinguished according to intrinsic morphological
4882
w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 4 8 7 7 e4 8 8 4
characteristics of the filamentous micro-organisms viz. mobility, branching, filament shape, filament length, attached growth, septa or transverse walls, cell diameter and sheath etc. (Eikelboom, 2000). The observations in this study showed that FI was maintained between 1 and 2 under limited filamentous bulking which means the effect of the filaments on the settling velocity of the sludge is limited (Eikelboom, 2000). Sludge sample under limited filamentous bulking was staining with DAPI fluorochrome and then observed using the fluorescence microscope, the typical epifluorescence micrographs are showed in Fig. 4. The dominant filamentous bacteria was identified as H. hydrossis due to its needle-like appearance in a pin cushion with straight filaments protruding from the flocs, as presented in Fig. 4(a), and which was within the floc structure. In addition, minor S. natans characterized by
straight or smoothly curved filaments with no/tree-like false branching, round-ended and rod shaped cells and clearly visible cell septa with indentations was also recognized as the secondary filamentous bacteria, as shown in Fig. 4(b). S. natans can radiate outward from the floc surface into the bulk solution and results in a high SVI by inter-floc bridging. In this study, the presence of H. hydrossis was mainly caused by low DO, while S. natans was likely to be caused by low DO and long retention time of sludge in secondary clarifier. Nevertheless, H. hydrossis population is usually limited present in domestic treatment plants and it can develop en masse in industrial plants where many easily biodegradable compounds are present in the influent (Eikelboom, 2000). Indeed, low DO tends to cause filamentous bulking by S. natans, type 1701 and H. hydrossis (Jenkins et al., 2003). Moreover, a few amount of Eikelboom Type 0041 with much attached growth was also observed which was beneficial as the backbone structure for the flocs in Fig. 4(c). However, M. parvicella and Type 021N were apparently not observed. The results of this study are in line with the research of Gaval and Pernelle, 2003, in which the dominant filamentous bacteria identifying by morphological criteria and FISH were H. hydrossis and S. natans under respective oxygen deficiency condition, and small Type 021N was also observed. But Guo et al., 2010 demonstrated that Eikelboom Type 0041 was the dominant filamentous bacterium and few Type 021N and M. parvicella were also detected, however, H. hydrossis and S. natans were never observed.
4.2.
Fig. 4 e Epifluorescence micrographs of DAPI stained filament of (a) H. hydrossis, (b) S. natans and (c) Eikelboom Type 0041. The length of the bars corresponds to: a and b 10 mm; c 20 mm. The filamentous bacterium was observed under limited filamentous bulking.
Limited filamentous bulking
It has been hypothesized that filamentous micro-organisms serve as a backbone for the flocs to provide more binding sites for the attachment of free cells or smaller aggregates by extracellular polymeric substances (EPS) (Cenens et al., 2000; Liao et al., 2011). The principle of limited filamentous bulking is to keep a moderate imbalance between floc-forming and filamentous bacteria, which has a slight advantage for filamentous bacteria. This allows a better enmeshing of tiny particles or free flocs, and the larger floc diameter under limited filamentous bulking implies diffusion resistance of oxygen inside the flocs is larger which facilitate SND (Martins et al., 2004a), although filamentous bacteria extend from the flocs make flocs a bit incompact and porous, but denser. Moreover, filamentous micro-organisms have the competitive advantages to access organic substrate based on A/V hypothesis (Jenkins et al., 2003) and diffusion-based selection (Martins et al., 2004a, 2010; Lou and de los Reyes, 2008) since the intrinsic morphological property (preferential growth of one or two directions) facilitates a large contact area and an easy penetration rate. In addition, the kinetic selection hypothesis (Chudoba et al., 1974) can also well explain it because of the lower affinity constant of filamentous bacteria than floc-forming bacteria that low DO concentrations favor the growth of filamentous bacteria. If we can balance the advantages and disadvantages of filamentous bacteria under low DO condition, such as limited filamentous bulking for a good effluent, energy-saving undoubtedly been achieved. The experiments in this study are according to above theory. Although general believe is that the activated sludge with more filamentous bacteria is negative for wastewater
w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 4 8 7 7 e4 8 8 4
operations (Jin et al., 2003), it was found that the solideliquid separation was practically not affected and no sludge loss was observed even with the highest SVI (200) condition under limited filamentous bulking induced by DO level. In the case of bio-P sludge, the density of the flocs might however compensate for the effect of the larger number of filaments (Eikelboom et al., 1998), and less production of sludge attributing to the characteristic of denitrifying dephosphatation in the EBPR reactor (Beun et al., 2000). In addition, a stable wellclarified effluent with marginal suspended solids (<8 mg/L) was obtained, which can be attributed to morphological characteristics of filamentous bacteria (i.e. enmeshment mechanism). The results of this study also show that limited filamentous bulking is repeatable and controllable. Therefore, the technique of limited filamentous bulking could be an alternative solution for enhancing nutrient removal and effluent quality though the control strategy and engineering capital in practice are still in question.
5.
Conclusions
This study investigated the integrated nitrogen, phosphorus and COD removal performance in a bench-scale plug-flow EBPR reactor under limited filamentous bulking. Experimental work lasted for about 200 days to study the removal performance of the EBPR process at different SVI periods. The results show that limited filamentous bulking induced by low DO level could enhance biological nutrient removal and achieve a well-clarified effluent. Moreover, low DO concentration associated to limited filamentous bulking is of importance for energy-saving. The optimum scenario for integrated nitrogen, phosphorus and COD removal was achieved under limited filamentous bulking at an SVI level of 170e200(associated with a DO of 1.0e1.5 mg/L), and the corresponding respective removal efficiencies of COD, TP, NHþ 4 eN were 90%, 97% and 92%. In addition, the solideliquid separation was practically not affected and no sludge loss was observed even at the highest SVI (200) condition, which was likely to attribute to the heavy flocs of bio-P sludge and less production of sludge of denitrifying dephosphatation.
Acknowledgments The authors gratefully acknowledge the financial support provided by National Water Pollution Control and Management Technology Major Projects of China (No. 2009ZX07317-008).
references
APHA, 1998. Standard Methods for Examination of Water and Wastewater, 20th ed. American Public Health Association, Washington, DC. Ajay, P., Jesse, Z., George, N., 2006. Simultaneous carbon, nitrogen and phosphorous removal from municipal wastewater in
4883
a circulating fluidized bed bioreactor. Chemosphere 65 (7), 1103e1112. Andreadakis, A.D., 1993. Physical and chemical properties lf activated sludge flocs. Water Research 27 (11), 1707e1714. Beun, J.J., Paletta, F., Van Loosdrecht, M.C.M., Heijnen, J.J., 2000. Stoichiometry and kinetics of poly B hydroxybutyrate metabolism under denitrifying conditions in activated sludge cultures. Biotechnology and Bioengineering 67, 379e389. Barat, R., van Loosdrecht, M.C.M., 2006. Potential phosphorus recovery in a WWTP with the BCFS process: interactions with the biological process. Water Research 40, 3507e3516. Chiu, Y.C., Lee, L.L., Chang, C.N., Chao, A.C., 2007. Control of carbon and ammonium ratio for simultaneous nitrification and denitrification in a sequencing batch bioreactor. International Biodeterioration and Biodegradation 59, 1e7. Chu, K.H., van Veldhuizen, H.M., van Loosdrecht, M.C.M., 2004. Respirometric measurement of kinetic parameters: effect of activated sludge floc size. Water Science and Technology 48 (8), 61e68. Chudoba, J., Blaha, J., Madera, V., 1974. Control of activated sludge filamentous bulking eIII. Effect of sludge loading. Water Research 8 (4), 231e237. Cenens, C., Smets, I., Van Impe, J., 2000. Modeling the competition between floc-forming and filamentous bacteria in activated sludge waste water treatment systems. Part II. A prototype mathematical model based on kinetic selection and filamentous backbone theory. Water Research 34, 2535e2541. Eikelboom, D.H., Andreadakis, A., Andreasen, K., 1998. Survey of the filamentous population in nutrient removal plants in four European countries. Water Science and Technology 37 (4/5), 281e290. Eikelboom, D.H., 2000. Process Control of Activated Sludge Plants by Microscopic Investigation. IWA Publishing, London, UK. Fu, Z., Yang, F., An, Y., Xue, Y., 2009. Simultaneous nitrification and denitrification coupled with phosphorus removal in an modified anoxic/oxic-membrane bioreactor (A/O-MBR). Biochemical Engineering Journal 43, 191e196. Gaval, G., Pernelle, J.J., 2003. Impact of the repetition of oxygen deficiencies on the filamentous bacteria proliferation in activated sludge. Water Research 37, 1991e2000. Gulez, G., de los Reyes, F.L., 2009. Multiple approaches to assess filamentous growth in activated sludge under different carbon source conditions. Journal of Applied Microbiology 106, 682e691. Guo, H.Y., Zhou, J.T., Su, J., et al., 2005. Integration of nitrification and denitrification in airlift bioreactor. Biochemical Engineering Journal 23, 57e62. Guo, J.H., Peng, Y.Z., Wang, S.Y., Zheng, Y.N., Huang, H.J., Wang, Z.W., 2009. Long term effect of dissolved oxygen on partial nitrification performance and microbial community structure. Bioresource Technology 100, 2796e2802. Guo, J.H., Peng, Y.Z., Peng, C.Y., Wang, S.Y., Chen, Y., Huang, H.J., Sun, Z.R., 2010. Energy saving achieved by limited filamentous bulking sludge under low dissolved oxygen. Bioresource Technology 101, 1120e1126. Jenkins, D., Richard, M.G., Daigger, G.T., 2003. Manual on the Cause and Control of Activated Sludge Bulking, Foaming, and Other Solids Separation Problems, third ed. Lewis Publishers, NY, USA. Jin, B., Wile´n, B.M., Lant, P., 2003. A comprehensive insight into floc characteristics and their impact on compressibility and settleability of activated sludge. Chemical Engineering Journal 95 (1e3), 221e234. Kotay, S.M., Datta, T., Choi, J., Goel, R., 2010. Biocontrol of biomass bulking caused by Haliscomenobacter hydrossis using a newly isolated lytic bacteriophage. Water Research. doi:10.1016/j. watres.2010.08.038.
4884
w a t e r r e s e a r c h 4 5 ( 2 0 1 1 ) 4 8 7 7 e4 8 8 4
Li, B., Bishop, L., 2004. Micro-profiles of activated sludge floc determined using microelectrodes. Water Research 38 (5), 1248e1258. Liao, J., Lou, I., delosReyes, F.L., 2004. Relationship of speciesspecific filament levels to filamentous bulking in activated sludge. Applied and Environmental Microbiology 70 (4), 2420e2428. Liao, B.Q., Lin, H.J., Langevin, S.P., Gao, W.J., Leppard, G.G., 2010. Effects of temperature and dissolved oxygen on sludge properties and their role in bioflocculation and settling. Water Research 45 (2), 509e520. Lou, I., de los Reyes, F.L., 2008. Clarifying the roles of kinetics and diffusion in activated sludge filamentous bulking. Biotechnology and Bioengineering 101, 327e336. Martins, A.M.P., Heijnen, J.J., van Loosdrecht, M.C.M., 2003a. Effect of feeding pattern and storage on the sludge settleability under aerobic conditions. Water Research 37 (11), 2555e2570. Martins, A.M.P., Heijnen, J.J., Van Loosdrecht, M.C.M., 2003b. Effect of dissolved oxygen concentration on sludge settleability. Applied Microbiology Biotechnology 62 (5e6), 586e593. Martins, A.M.P., Pagilla, K., Van Loosdrecht, M.C.M., 2004a. Filamentous bulking sludge e a critical review. Water Research 38 (4), 793e817. Martins, A.M.P., Picioreanu, C., Heijnen, J.J., van Loosdrecht, M.C. M., 2004b. Three-dimensional dual-morphotype species modeling of activated sludge flocs. Environmental Science and Technology 38 (21), 5632e5641. Martins, A.M.P., Karahan, O., van Loosdrecht, M.C.M., 2010. Effect of polymeric substrate on sludge settleability. Water Research. doi:10.1016/j.watres.2010.07.055. Martı´, N., Pastor, L., Bouzas, A., Ferrer, J., Seco, A., 2010. Phosphorus recovery by struvite crystallization in WWTPs: influence of the sludge treatment line operation. Water Research 44, 2371e2379.
Peng, Y.Z., Guo, J.H., Wang, S.Y., Chen, Y., 2008. Energy saving achieved by limited filamentous bulking under low dissolved oxygen: derivation, originality and theoretical basis. Journal of Environmental Sciences 29 (12), 3342e3347. Pe´rez, J., Picioreanu, C., van Loosdrecht, M.C.M., 2005. Modeling biofilm and floc diffusion processes based on analytical solution of reaction-diffusion equations. Water Research 39, 1311e1323. Pochana, K., Keller, J., 1999. Study of factors affecting simultaneous nitrification and denitrification (SND). Water Science and Technology 39 (6), 61e68. Third, K.A., Burnett, N., Cord-Ruwisch, R., 2003. Simultaneous nitrification and denitrification using stored substrate (PHB) as the electron donor in an SBR. Biotechnology and Bioengineering 83, 706e720. Van Loosdrecht, M.C.M., Brandse, F.A., de Vries, A.C., 1998. Upgrading of waste water treatment processes for integrated nutrient removal e the BCFS process. Water Science and Technology 37 (9), 209e217. Vaiopoulou, E., Melidis, P., Aivasidis, A., 2007. Growth of filamentous bacteria in an enhanced biological phosphorus removal system. Desalination 213, 288e296. Vaiopoulou, E., Aivasidis, A., 2008. A modified UCT method for biological nutrient removal: configuration and performance. Chemosphere 72, 1062e1068. Wanner, J., Kragelund, C., Nielsen, P., 2010. In: Seviour, R., Nielsen, P.H. (Eds.), Microbial Ecology of Activated Sludge. IWA Publishing, London. Wile´n, B.M., Balme´r, P., 1999. The effect of dissolved oxygen concentration on the structure, size and size distribution of activated sludge flocs. Water Research 33 (2), 391e400. Zhu, G.B., Peng, Y.Z., Wu, S.Y., Wang, S.Y., Xu, S.W., 2007. Simultaneous nitrification and denitrification in step feeding biological nitrogen removal process. Journal of Environmental Science 19, 1043e1048.