Nutrient removal performance of a sequencing batch reactor as a function of the sludge age

Nutrient removal performance of a sequencing batch reactor as a function of the sludge age

Enzyme and Microbial Technology 31 (2002) 842–847 Nutrient removal performance of a sequencing batch reactor as a function of the sludge age Fikret K...

133KB Sizes 1 Downloads 80 Views

Enzyme and Microbial Technology 31 (2002) 842–847

Nutrient removal performance of a sequencing batch reactor as a function of the sludge age Fikret Kargi∗ , Ahmet Uygur Department of Environmental Engineering, Dokuz Eylul Universiy, Buca, Ízmir, Turkey Received 29 November 2001; accepted 12 July 2002

Abstract Nutrient removal from synthetic wastewater by sequencing batch operation was studied at different solids retention times (SRTs). The nutrient removal process was consisted of anaerobic, anoxic I, oxic I, anoxic II, oxic II and settling phases. Hydraulic residence times (HRT) of the aforementioned phases were kept constant at 2/1/4.5/1.5/1.5 h. Settling phase was 0.5 h for all experiments. Solids retention time was varied between 5 and 30 days at six different levels. Effects of SRT (sludge age) on COD, nitrogen (NH4 –N, NO3 –N) and phosphate (PO4 –P) removal were investigated and the optimal sludge age resulting in maximum nutrient removal efficiency was determined. The highest COD (94%), NH4 –N (84%) and PO4 –P (70%) removal efficiencies were obtained at the sludge age of 10 days, although a sludge age of 15 days resulted in only slightly lower values. Sludge ages larger than 15 days resulted in lower nutrient removal efficiencies as compared to those obtained at 10 or 15 days of sludge age. Sludge volume index (SVI) was also minimum (55 ml/g) at sludge age of 10 days. MLSS concentration increased with sludge age resulting in MLSS concentration of 3500 mg/l at SRT of 30 days. On the basis of these results, a sludge age of 10 days was found to be optimal resulting in maximum nutrient removal efficiencies and minimum SVI. © 2002 Elsevier Science Inc. All rights reserved. Keywords: Biological nutrient removal; Sequencing batch reactor (SBR); Sludge age

1. Introduction Sequencing batch reactors were originally used for COD and phosphate removal from wastewaters [1–9]. Recent regulations over nutrient discharges to natural water systems resulted in modifications in sequential batch reactor (SBR) systems to achieve nitrification, denitrification along with COD and phosphate removal. SBR treatment system consists of a sequencing operation including the steps of fill, react, settle, decant, and idle [10]. When biological nutrient removal is desired, the steps in the react cycle are adjusted to provide anaerobic, anoxic and aerobic phases in certain number and sequence. Numerous studies on nutrient removal have been reported [11–23]. Colunga and Martinez [8] studied the effects of aforementioned phases in a biofilm SBR on the removal efficiency of carbonaceous substrates, phosphate and ammonia nitrogen. The highest removal efficiencies of COD and PO4 –P were obtained with a 12 h cycle and phase ratio of 37/63% anaerobic/aerobic. ∗

Corresponding author. Tel.: +90-232-4531143; fax: +90-232-4531153. E-mail address: [email protected] (F. Kargi).

Umble and Ketchum [12] studied a sequencing batch reactor for biological treatment of a municipal wastewater. A 12 h total cycle time resulted in BOD5 , TSS, and NH3 –N removals of 98, 90 and 89%, respectively. Chang and Hao [13] investigated nutrient removal in an SBR to identify process parameters that could be useful for monitoring and real-time control purposes. COD, total nitrogen and phosphate removal efficiencies of 91, 98 and 98% were obtained at the solids retention time (SRT) of 10 days, with a cycle duration of 6 h. Demuynck et al. [19] studied SBR plants for nutrient removal. Addition of supplementary COD was found to be necessary during the anoxic phase to obtain complete nitrogen removal. It was found that a sequence of short aerobic/anoxic phases was better than the usual sequence of aerobic phase followed by anoxic phase. Andreottola et.al. [20] developed an algorithm for optimization of the cycle length and phase distribution in order to minimize effluent nitrogen concentration. Optimization results were carried out with 3.3 h of anoxic and 4.2 h of anaerobic phase durations. Effluent concentrations of nitrate, nitrite and ammonia were 2.9, 0.04 and 0.06 mg/l, respectively.

0141-0229/02/$ – see front matter © 2002 Elsevier Science Inc. All rights reserved. PII: S 0 1 4 1 - 0 2 2 9 ( 0 2 ) 0 0 2 0 9 - 0

F. Kargi, A. Uygur / Enzyme and Microbial Technology 31 (2002) 842–847

Ho et al. [21] carried out experimental studies in a small-scale SBR system of 30 l to define important parameters affecting the process performance. Varying hydraulic residence times (HRTs) with the BOD concentrations of 100–200 mg/l yielded maximum removal efficiencies for N and P within 1–3–2 h of anaerobic–aerobic–anoxic phases. Zuniga and Martinez [22] investigated the possibility of combined phosphorus and nitrogen removal in a biofilm sequencing batch reactor using an operation strategy with four reaction phases: anaerobic/aerobic/anoxic/aerobic. The system was operated successfully with COD, phosphate and ammonia nitrogen removals of 89±1, 75±15, and 87±10%, respectively. Sang-Ill et al. [23] used fermented swine waste instead of acetate for supplementation of bench-scale sequencing batch reactors (SBRs) to improve nutrient removal. There was essentially no difference in performance of the reactors supplemented with either acetate or fermented swine waste; both achieved a total nitrogen and phosphorus removals of 90 and 89%, respectively. In none of the aforementioned studies the nutrient removal performance of a five-step SBR operation was investigated as a function of SRT (sludge age). Therefore, it is the major objective of this study to systematically investigate the effect of solids retention (SRT) on the performance of a five-step SBR used for nutrient (C, N, P) removal. The cycles consisted of anaerobic–anoxic–oxic–anoxic and oxic (An–Ax–Ox–Ax–Ox) phases with HRTs of 2/1/4.5/1.5/1.5 h which were determined to be optimal in our previous studies. SRTs were varied between 5 and 30 days and the optimal SRT resulting in maximum overall nutrient (COD, N, P) removals were determined. Variation of the effluent nutrient concentrations, biomass concentration and the sludge volume index (SVI) with SRT were also investigated. Nutrient concentration profiles during the whole course of the operation were also determined for the each SRT tested.

2. Materials and methods 2.1. Experimental set-up A schematic diagram of the experimental set up is depicted in Fig. 1. A fermenter (Bioflo IIC, New Brunswick) with 5 l working volume was used as the sequencing batch reactor (SBR). The fermenter was microprocessor controlled for aeration, agitation, pH and dissolved oxygen (DO). Aeration was provided by using an air pump and a sparger. Agitation speed was varied between 25 and 300 rpm. pH and DO of the nutrient medium were continuously monitored by the relevant probes. 2.2. Wastewater composition Synthetic wastewater used throughout the studies was composed of glucose, sodium acetate, NH4 Cl, KH2 PO4 ,

843

Fig. 1. The sequencing batch reactor used in experimental studies.

MgSO4 ·7H2 O, NaHCO3 and trace salt minerals of NaCl (100 mg/l), KCl (20 mg/l), CaCl2 ·2H2 O (50 mg/l), FeCl3 ·6H2 0 (50 mg/l). Typical composition of the synthetic wastewater was COD0 = 1, 200 mg/l, NT = 60 mg/l and PT = 18 mg/l, resulting in COD/N/P = 100/5/1.5. MgSO4 and NaHCO3 concentrations in the feed were 0.1 and 0.5 g/l, respectively throughout the studies. Glucose and sodium acetate concentrations were adjusted to yield equal COD of 600 mg/l each. 2.3. Organisms A mixed microbial population composed of heterotrophic organisms capable of oxidizing carbonaceous compounds, denitrifying organisms, autotrophic nitrifying organisms, anaerobic organisms (acid producers) and excess phosphate uptaking organisms (Acinetobacter sp.) were used as inoculum culture. Nitrification organisms (Nitrosomonas and Nitrobacter) were obtained from Clemson University, SC, USA. Heterotrophic culture obtained from Çigli Municipal Wastewater Treatment Plant, Izmir was used as a component of the inoculum culture for carbon and nitrogen removal. Acinetobacter calcoaceticus (NRRL-552) obtained from the USDA, National Research Laboratories, Peoria, IL, USA was used for luxury phosphate uptake. The cultures were cultivated in suitable growth media in the laboratory and used as inoculum in form of mixed culture. 2.4. Experimental procedure Before starting sequencing batch operation, the reactor was filled with the synthetic wastewater, inoculated with a mixed culture of microorganisms and was operated batchwise with aeration and mixing for several days to obtain a dense culture to start with. After sedimentation of the organisms, the clear supernatant on top was removed and

844

F. Kargi, A. Uygur / Enzyme and Microbial Technology 31 (2002) 842–847

the reactor was filled up to 5 l total volume with the defined nutrient media. Then, anaerobic, anoxic, oxic operations were applied in sequence. Nitrogen gas was passed through the media only during anaerobic operation. Agitation speed during anaerobic and anoxic cycles was 25 and 50 rpm, respectively. The media was aerated and agitated (300 rpm) vigorously during oxic (aerobic) operation. Samples were withdrawn from the reactor at the beginning and at the end of each cycle for analysis. At the end of each SBR operation, the organisms were sedimented for 0.5 h and the treated wastewater on top of the sediment was removed. Sedimented organisms were used for the next treatment operation. A fraction of the culture was removed from the reactor daily to adjust the sludge age to the desired level. Temperature and pH were controlled around T = 25 ◦ C and pH = 7–7.5. DO concentration in oxic (aerobic) phase was kept above 2 mg/l, while the DO during anaerobic and anoxic phases was zero. Solids retention times (sludge age) were varied between 5 and 30 days while HRT of each step were kept constant throughout the experimental study. Five-step nutrient removal process consisting of anaerobic/anoxic/oxic/anoxic/ oxic steps with HRTs of 2/1/4.5/1.5/1.5 h was used. Experiments were carried out two times at each SRT tested. The first run was to adapt and develop the culture at the SRT used. The second run was used for collection of data. 2.5. Analytical methods Samples were withdrawn from the liquid media at the beginning and at the end of each treatment period (anaerobic, anoxic, aerobic) and were centrifuged at 6000 rpm for 0.5 h to remove microorganisms from the liquid medium. Clear supernatants were analyzed for COD, ammonium and nitrate nitrogen and phosphate–P contents three times and the average values were reported. Standard kits (Merck-Spectroquant) and spectrometric methods were used for nitrogen and phosphorous analysis. COD, total solids (TS) and total suspended solids (TSS) concentrations were determined by using the Standard Methods [24]. DO and pH measurements were done by using the probes and analyzers associated with the microprocessor controlled fermenter (New Brunswick, Bioflo IIC). Biomass concentrations (MLSS) were determined by filtering the samples through 0.45 micromillipore filter and drying in an oven at 105 ◦ C until constant weight.

3. Results and discussion 3.1. Effect of sludge age on nutrient removal Variation of COD removal efficiency with the SRT is depicted in Fig. 2a. COD removal efficiencies at every SRT tested were above 90% with some minor differences. The highest COD removals were obtained at sludge ages of 10, 15 and 25 days which were 94, 95 and 96%, respectively.

Fig. 2. Variations of nutrient removal efficiencies with sludge age (SRT): (a) COD; (b) NH4 –N; (c) PO4 –P.

Ammonium–N removal efficiencies were also affected by sludge age variations (Fig. 2b). Maximum NH4 –N removal efficiency (84%) was obtained at a sludge age of 10 days, while the efficiency obtained at SRT of 15 days was only slightly lower (83%). NH4 –N removal efficiencies decreased with the sludge age above 15 days of SRT, resulting in 68% removal at 25 days of sludge age. Old population of the nitrification–denitrification organisms at high SRT values may be responsible for low NH4 –N removal performance. Fig. 2c depicts variation of phosphate–P removal efficiency with sludge age. The maximum PO4 –P removal (70%) was obtained at the sludge age of 10 days while the removals at SRT of 15 and 25 days were similar (67 and 68%). When all the results were examined, 10 days of

F. Kargi, A. Uygur / Enzyme and Microbial Technology 31 (2002) 842–847

845

Fig. 4. Variation of biomass concentration (MLSS, mg/l) with the sludge age.

In addition to nutrient removal performances at different sludge ages, variation of biomass concentration (MLSS) and SVI with sludge age (SRT) were also investigated. Variation of biomass concentration with the sludge age is depicted in Fig. 4. As the sludge age was gradually increased from 5 days, biomass concentration increased as a result of sedimentation and transfer of the biomass to the next step. Biomass concentration of 1700 mg/l at SRT of 5 days increased to 3600 mg/l at SRT of 30 days. The increase in the biomass concentration at sludge age values above 20 days was negligible. SVI is an important parameter affecting the performance of the system. Low SVI values (SVI < 100 ml/g) indicate good sedimentation characteristics of the sludge yielding high biomass concentrations in the aeration tank; whereas high SVI values (SVI  100 ml/g) reflect bulking sludge and low biomass concentrations in the aeration tank. Sedimentation characteristics of the sludge are presented in Fig. 5 in form of SVI as a function of the sludge age. SVI was low (52 ml/g) at low sludge ages of 5 and 10 days and increased to 66 ml/g at sludge ages of 25 and 30 days. Apparently, the old cells of high sludge age exhibited lower sedimentation characteristics as compared to the younger cells of Fig. 3. Variations of effluent nutrient concentrations with sludge age (SRT): (a) COD; (b) NH4 –N; (c) PO4 –P.

sludge age was the optimal value resulting in nearly maximum removal efficiencies for COD, NH4 –N and PO4 –P. Variation of the effluent nutrient concentrations with sludge age (SRT) are depicted in Fig. 3. Effluent COD was 40 mg/l at SRT of 15 and 25 days. However, the effluent COD at 10 days of SRT was nearly 70 mg/l. Ammonium–N and phosphate–P concentrations at the end of the operation were 7 and 3.6 mg/l for the sludge age of 10 days. Since, other sludge ages resulted in somewhat higher nitrogen and phosphorous levels, an SRT of 10 days was the optimal value minimizing the effluent nutrient levels.

Fig. 5. Variation of sludge volume index (ml/g MLSS) with the sludge age.

846

F. Kargi, A. Uygur / Enzyme and Microbial Technology 31 (2002) 842–847

Fig. 6. Nutrient concentration profiles for sludge age 10 days in an SBR operation: (a) COD; (b) NH4 –N; (c) NO3 –N; (d) PO4 –P.

low sludge age. Therefore, 10 days of sludge age was also optimal resulting in minimum SVI and the best sedimentation characteristics. 3.2. Nutrient concentration profiles at the optimal SRT Fig. 6 shows variations of nutrient (COD, NH4 –N, NO3 –N, PO4 –P) concentrations with time when the system was operated at the optimal SRT of 10 days. COD concentration dropped steadily reaching a level of below 70 mg/l at the end of 10.5 h of operation. The major fraction of COD was removed mainly during the first oxic step by effective oxidation of the carbonaceous compounds. NH4 –N concentration increased slightly during the anaerobic step and than decreased in anoxic step. Most of the NH4 –N was removed during the first oxic (aerobic) step by assimilation and nitrification. The last two steps did not contribute significantly to the removal of NH4 –N resulting in nearly 7 mg/l NH4 –N concentration at the end of the operation. Initial nitrate–N concentration of 1.1 mg/l was nearly constant for the first two steps (3 h of operation) and increased to 4.3 mg/l at the end of the first oxic step because of nitrification. As a result of denitrification in the second anoxic step NO3 –N concentration decreased to 0.9 mg/l. The final value of NO3 –N at the end of the operation was 1.3 mg/l.

Initial phosphate–P concentration of 12.2 mg/l increased to 13.6 mg/l at the end of the first anoxic step because of phosphate release by the luxury phosphate uptaking organisms. As a result of phosphate uptake in the first oxic step, PO4 –P concentration decreased to 3.6 mg/l. The last two steps did not change PO4 –P concentration. DO concentration was also monitored and controlled at the desired levels during the operation. DO concentrations were less than 0.1 mg/l during the first two steps and increased to 4.65 mg/l by vigorous aeration during the oxic step. DO values were 0.2 and 3.1 mg/l during the second anoxic and oxic steps. Therefore, by a five-step SBR operation with a total cycle time of 10.5 h, nearly 94% of COD, 84% of NH4 –N and 70% of PO4 –P were removed at the optimal SRT of 10 days, resulting in final COD, NH4 –N, NO3 –N and PO4 –P of 70, 7, 1.3 and 3.6 mg/l, respectively. These results compare well with the literature values.

4. Conclusions A five-step SBR operation consisting of anaerobic/anoxic/ oxic/anoxic/oxic steps was used for nutrient removal from a synthetic wastewater. HRTs were kept constant as

F. Kargi, A. Uygur / Enzyme and Microbial Technology 31 (2002) 842–847

2/1/4.5/1.5/1.5 h for the aforementioned steps while the sludge age (SRT) was varied between 5 and 30 days. The optimal sludge age resulting in maximum COD (94%), NH4 –N (84%) and PO4 –P (70%) removal efficiencies was 10 days although 15 days of SRT resulted in only slightly lower values. COD concentration decreased from 1120 mg/l to nearly 70 mg/l; NH4 –N and PO4 –P concentrations decreased from initial values of 46 mg/l and 12 mg/l to nearly 7 and 3.6 mg/l, respectively, at the end of 10.5 h of SBR operation. Biomass concentration (MLSS) increased with the sludge age starting from 1700 mg/l at SRT of 5 days and reaching to nearly 3600 mg/l at SRT of 30 days. SVI which is a measure of settling ability of the sludge varied between 52 and 66 ml/g depending on the sludge age. Minimum SVI (52 ml/g) was obtained at SRT of 5 days. However, SVI value (53 ml/g) obtained at SRT of 10 days was comparable with that of the 5 days. On the basis of the experimental results, the optimal SRT value was found to be 10 days, resulting in maximum nutrient removal and minimum SVI. Acknowledgments This study was supported by the research funds of Uludag University, Bursa, Turkey. References [1] Tasli R, Artan N, Orhon D. The influence of different substrates on enhanced biological phosphorus removal in a sequencing batch reactor. Water Sci Technol 1997;35:75–80. [2] Baozhen W, Jun L, Lin W, Meisheng N, Ji L. Mechanism of phosphorus removal by SBR submerged biofilm system. Water Res 1998;32:2633–8. [3] Ramirez CN, Martinez SG. Phosphorus uptake kinetics in a biofilm sequencing batch reactor. Bioprocess Eng 2000;23:143–7. [4] Belia E, Smith PG. The bioaugmentation of sequencing batch reactor sludges for biological phosphorus removal. Water Sci Technol 1997;35:19–26. [5] Shin HS, Jun HB. Developments of excess phosphorus removal characteristics in a sequencing batch reactor. Water Sci Technol 1992;25:440–3. [6] Danesh S, Oleszkiewicz JA. Use of a new anaerobic–aerobic sequencing batch reactor system to enhance biological phosphorus removal. Water Sci Technol 1997;35:137–44. [7] Carucci A, Majone M, Ramadori R, Rosetti S. Biological phosphorus removal with different organic substrates in an anaerobic/aerobic sequencing batch reactor. Water Sci Technol 1997;35:161–87.

847

[8] Colunga AM, Martinez SG. Effects of population displacements on biological phosphorus removal in a biofilm SBR. Water Sci Technol 1996;34:303–13. [9] Colmenarejo MF, Bustos A, Garcia MG, Borja R, Banks CJ. An analysis of the factors that influence biological phosphorus removal (BPR) in a sequencing batch anaerobic/aerobic reactor. Bioprocess Eng 1998;19:171–4. [10] Tchobanoglous G, Burton FL (Metcalf & Eddy). Wastewater engineering: treatment, disposal, reuse. 3rd ed. NY, USA: McGraw Hill, 1991. [11] Arora ML, Barth EF, Umphres MB. Technology evaluation of sequencing batch reactors. J Water Poll Cont Fed 1985;57:867–75. [12] Umble AK, Ketchum AL. A strategy for coupling municipal wastewater treatment using the sequencing batch reactor with effluent nutrient recovery through aquaculture. Water Sci Technol 1997;35:177–84. [13] Chang CH, Hao OJ. Sequencing batch reactor system for nutrient removal: ORP and pH profiles. J Chem Technol Biotechnol 1996;67:27–38. [14] Pastorelli G, Canziani R, Pedrazzi L, Rozzi A. Phosphorus and nitrogen removal in moving-bed sequencing batch biofilm reactor. Water Sci Technol 1999;40:169–76. [15] Furumai H, Kazmi AA, Fujita M, Furuya Y, Sasaki K. Modeling long term nutrient removal in a sequencing batch reactor. Water Res 1999;33:2708–14. [16] Keller J, Subramaniam K, Gösswein J, Greenfield PF. Nutrient removal from industrial wastewater using single tank sequencing batch reactors. Water Sci Technol 1997;35:137–44. [17] Subramaniam K, Greenfield PF, Ho KM, Johns MR, Keller J. Efficient biological nutrient removal in high strength wastewater using combined anaerobic-sequencing batch reactor treatment. Water Sci Technol 1994;30:315–21. [18] Demoulin G, Goronszy IC, Wutscher K, Forsthuber E. Co-current nitrification/denitrification and biological P-removal in cyclic activated sludge plants by redox controlled cycle operation. Water Sci Technol 1997;35:215–24. [19] Demuynck C, Vanrolleghem P, Mingneau C, Liessens J, Verstraete W. NDBEPR process optimization in SBRs: reduction of external carbon-source and oxygen supply. Water Sci Technol 1994;30: 169–79. [20] Andreottola G, Bortone G, Tilche A. Experimental validation of a simulation and design model for nitrogen removal in sequencing batch reactors. Water Sci Technol 1997;35:113–20. [21] Ho NC, Ra KM, Byung GP, Seong-Jin L, Dong WC, Woo GL. Simulation of sequential batch reactor (SBR) operation for simultaneous removal of nitrogen and phosphorus. Bioprocess Eng 2000;23:513–21. [22] Zuniga MAG, Martinez SG. Biological phosphate and nitrogen removal in a biofilm sequencing batch reactor. Water Sci Technol 1996;34:293–301. [23] Sang-Ill L, Jong-Ho P, Kwang-Baik K, Ben K. Effect of fermented swine wastes on biological nutrient removal in sequencing batch reactors. Water Res 1997;31:1807–12. [24] American Public Health Association (APHA) Standard methods for the examination of water and wastewater. 17th ed. Washington DC, USA, 1998.