Treatment of Swine Wastewater using Sequencing Batch Reactor*

Research Paper EAEF 4(2) : 47-53, 2011 Treatment of Swine Wastewater using Sequencing Batch Reactor* Mohammad N. ISLAM *1 Keum J. PARK *2 Md. J. ALAM *3 Abstract The swine wastewater from Sunchon swine farm was decomposed using a sequencing batch reactor (SBR). The reactor body was fabricated using a plexi glass cylinder and its total volume was 20L with 15L of working volume. Each operating cycle consisted of five phases (fill, react, settle, draw and idle) with a total cycle time of 8 hours, thus resulting in three cycles per day (with 5 days of hydraulic retention time and 41 days of solid retention time). The cycles of the SBR system were controlled by a designed on-site computer and custom software. The results showed removal efficiencies of 85.5%, 80.3% and 87.2% for BOD, COD and TP respectively. It was found however that there were some non-satisfactory results, only attaining removal efficiencies of 61.0%, 31.2% and 54.5% for TN, NH 3 -N and NO - 3 -N respectively. This was possibly due to the lack of enough carbon source and the inadequate aeration rate. It was also observed that removal efficiencies of 61.4%, 62.8%, 77.6% and 73.2% could be obtained for TS, TVS, TSS and TVSS respectively. The study showed that the SBR system could be used to attain good removal efficiencies of BOD, COD and nutrients in swine wastewater treatment if it is supplied with sufficient carbon source for de-nitrification and optimum aeration for nitrification. [Keywords] sequencing batch reactor (SBR), COD removal, BOD removal, swine wastewater, wastewater treatment, removal efficiency I Introduction The sequencing batch reactor (SBR) has gained wide acceptance for the removal of the biochemical oxygen demand (BOD), chemical oxygen demand (COD), and different nutrients from wastewater (Imura et al., 1993, Rusten and Eliassen, 1993). Wastewaters from pig farms are characterized by their high BOD, COD and other nutrient contents. Many pig farms in Korea typically use an activated sludge treatment system for the decomposition of wastewaters. The activated sludge system however has problems with high energy consumption and biomass production, leading to a relatively high operation cost and the disposal of a large amount of sludge. It has been found that biological processes based upon a sequencing batch reactor (SBR) are effective for organic nutrient removal in domestic and industrial wastewater (A Mohseni and Bazari, 2004). In recent years sequencing batch reactors have become of great interest for wastewater treatment due to their simple configuration (all necessary processes take place within a single time sequence in a single basin). SBR can achieve nutrient removal by the nitrification and de-nitrification using alternation of oxic and anoxic periods. Due to its operational flexibility, it is simple to increase SBR efficiency in treating wastewater by changing the timescale of each phase. Several researchers have used SBR to remove nitrogen, phosphorus, COD and BOD from swine wastewater (Bicudo et al., 1999; Kim et al., 2000; Kim et al., 2004; Tilche et al., 1999). Despite this SBR is not widely utilized to treat swine farm wastewater in Korea. This research was conducted to evaluate the performance of an oxic-anoxic SBR system according to a specific time schedule in terms of reduction of COD, BOD and nutrients in treating slurry manure from swine farm. II Materials and Methods 1. SBR construction and set-up The experiment was carried out using a lab scale sequencing batch reactor (SBR), having a total volume of 20L with a working volume 15L. The SBR system was installed at the farm power lab in Sunchon National University, Korea. The SBR body was fabricated using a transparent plexi glass cylinder with an inner diameter of 190mm. * This paper was supported by Sunchon National University Research fund in 2007. *1 Department of Industrial Machinery Engineering, College of Bio-industry Science, Sunchon National University, South Korea *2 KSAM Member, Corresponding author, Professor, Department of Industrial Machinery Engineering, College of Bio-industry Science, Sunchon National University, Jeonnam 540-742, South Korea, E-mail: pkj@sunchon.ac.kr, Tel: +82-617503267 Fax: +82-617503260 *3 Department of Animal Science & Technology, Sunchon National University, South Korea

48 Engineering in Agriculture, Environment and Food Vol. 4, No. 2 (2011) LabVIEW data logger Air pump Draw pump Mixer Influent pump Mixer Effluent tank Air flow controller Influent tank Surplus sludge Fig. 1 A schematic diagram of the experimental arrangements for the sequencing batch reactor treatment system. The system consisted of the reactor body, two peristaltic pumps (7524-45, Cole-Parmer Instrument Co.) for feeding influent and discharging effluent and 3 probes for the measurement of ph, temperature and the dissolved oxygen content (DO). Air was supplied into the reactor via two porous stone diffusers. Supplemental mixing was provided by turbine stirrers with four blades (radius 6 cm). The SBR system operation and data acquisition were accomplished by an on-site computer using LabVIEW software (National Instrument Corporation). A schematic diagram of the design of the SBR system is shown in Fig. 1. 2. Collection of swine wastewater Wastewater for the experiment was collected from Suncheon swine farm located near Suncheon city. The collected wastewater was sieved to remove coarse materials (particles greater than 600 µm) and then stored at 4 o C if not used immediately. The original wastewater had about 7% of total solids (TS), so it was diluted seven times with fresh water to decrease the quantity of total solids for this experiment. The characteristic compositions of the diluted wastewater are presented in Table 1. Table 1 Basic properties of the influent wastewater fed into the sequencing batch reactor. Parameters Concentration ph 7.47 Biochemical oxygen demand, mg/l 1500 Chemical oxygen demand, mg/l 1972 Total phosphorus (TP), mg/l 1058 Total nitrogen(tn), mg/l 1720 Nitrate-nitrogen(NO - 3 -N), mg/l 377 Ammonia-nitrogen(NH 3 -N), mg/l 948 Total solids(ts), % 1.20 Total volatile solids(tvs), % 0.67 Total suspended solids(tss), % 0.67 Total volatile suspended solids 0.33 (TVSS), % 3. Sampling and analytical method During the experimental period (June 20 July 31; 2009), sampling was carried out every day for the first four days, one time every two days from the 5 th to the 27 th day and once every three or four days after that.

ISLAM, PARK, ALAM : Treatment of Swine Wastewater using Sequencing Batch Reactor 49 FILL REACT SETTLE DRAW IDLE 40 min 360 min 40 min 20 min 20min Anaerobic Oxic (40 min) Anoxic (20 min) Fig. 2 The time frame for one cycle tested on the SBR. All analytical measurements were undertaken according to the standard methods of the American Public Health Association (APHA, 1998). The parameters analyzed include: total solids (TS), total volatile solids (TVS), total suspended solids (TSS), total volatile suspended solids (TVSS), 5-day biochemical oxygen demand (BOD 5 ), chemical oxygen demand (COD), total phosphorus (TP), total nitrogen (TN), ammonia nitrogen (NH 3 -N) and nitrate nitrogen (NO - 3 -N). The COD, TP, TN, NH 3 -N, NO - 3 -N contents were measured using a DR/2800 spectrophotometer ( Hach company, 1993). Suspended solids were measured using a glass micro fiber filter (Cat. No. 1822-047, Whatman). The ph, temperature and the dissolved oxygen content (DO) of the mixed liquor in the SBR tank were measured manually every day using CX-401(Sechang, Korea). 4. Experimental procedure The SBR experiment was performed for 41 days at 23.7±1 o C liquid temperature. At the beginning the SBR was filled with 14 L of influent wastewater and started regular operation according to the time schedule. Previous research indicates that the cycle regime and hydraulic retention time (HRT) are the two critical parameters that affect the performance and economics of an SBR. The HRT was 5 days for this experiment, following that indicated by previous research (Ra et al., 2000). Each operating cycle consisted of five phases, i.e. FILL, REACT, SETTLE, DRAW and IDLE, and lasted for a total of 8 hours, thus resulting in three cycles per day. During the FILL phase, the main reactor received influent wastewater from the storage basin at a feeding rate of 25 ml/min resulting in a fill rate of 1 L of influent per cycle. The REACT phase was composed of an oxic-anoxic process. The liquid in the tank was mixed by a 4 bladed propeller and air was supplied (1-1.5 L/min) during the oxic period in the reaction basin. During the SETTLE phase a thick sludge was formed and this sludge was removed during the IDLE phase to keep the SRT of 41 days. In the DRAW phase, the supernatant was removed by the peristaltic pump with a flow rate of 50 ml/min resulting in 5 days of HRT. Finally, the decanted effluent was collected and analyzed. A detailed operating time frame for one cycle is shown in Fig. 2. III Results and Discussion 1. BOD removal The BOD 5 concentration is used to assess the organic matter removal (Zhu et al., 2006). Because the experiment started after filling 14 L of raw wastewater into the SBR tank, effluent concentrations for all items during the first 5 days had high values and decreased continuously towards constant values. In this present study, the influent had a BOD 5 concentration of 1500 mg/l. This reduced to 190 mg/l by the 9 th day and fluctuated from 110 mg/l to 390 mg/l during the rest of the testing period with a mean of 216.9 mg/l as shown in Fig. 3. The removal rates of the effluents were obtained using the data from the 9 th day to the end of the experiment. Fig. 3 Variation of BOD concentration and removal

50 Engineering in Agriculture, Environment and Food Vol. 4, No. 2 (2011) Removal efficiency was calculated by the difference between the influent and effluent concentration in the wastewater. BOD removal efficiency was 85.5±4.5% during this period with a slight increase towards the end of the experiment. This reduction rate is slightly lower than the BOD removal rate of 93% found in previous research (Lo et al., 1999), possibly due to inadequate aeration. A study for pig wastewater treatment was conducted by Lo et al. (1999) using an aerobic SBR with SRT and HRT of 14 days and 23h, respectively. 2. COD removal The removal efficiency for COD and BOD 5 depends on the reactor cycle working with or without the anoxic phase, at different hydraulic retention times (HRT) in the SBR (Kulikowska et al., 2007). A study of an aerobic bench scale SBR with an influent COD concentration of 2000 mg/l for dairy wastewater treatment was conducted by Mohseni-Bandpi and Bazari (2004). In the experiment one cycle was composed of 8 hours with a 6 hour aeration period. In this study, COD removal efficiency was between 85 and 90%, similar to the presented experiment. In the presented experiment COD concentration decreased from the initial value of 1972 mg/l to 499 mg/l at the 9 th day and varied in the range of 280-499 mg/l with a mean value of 388.1 mg/l from the 9 th day to the end of testing. COD removal efficiency during this period was 80.3±2.6% with a slight increase towards the end of the experimental period as shown in Fig. 4. Fig. 5 Variation of TP concentration and removal Fig. 6 Variation of TN concentration and removal Fig. 4 Variation of COD concentration and removal Fig. 7 Variation of NH 3 -N concentration and removal

ISLAM, PARK, ALAM : Treatment of Swine Wastewater using Sequencing Batch Reactor 51 Fig. 8 Variation of NO - 3 -N concentration and removal 3. Phosphorus removal Figure 5 presents the changes of total phosphorus content in the effluent liquid. The influent concentration of total phosphorus was 1058 mg/l and effluent concentrations decreased to 130 mg/l by the 9 th day and varied from 89 mg/l to 165 mg/l with a mean of 135.4 mg/l after this. During this period the average removal efficiency of total phosphorus was 87.2±2.0%. DO concentration was varied from 1.7 mg/l to 2.1 mg/l with a mean value of 1.9 mg/l during the FILL phase. During the FILL phase, the anaerobic environment forced a group of microbes (phosphorus-accumulating microorganisms) to expend energy to obtain readily biodegradable carbon substrates and store them as poly hydroxyl alkanets (PHAs) causing a release of phosphorus into the liquid (Lee et al., 2001). High phosphorus removal was possibly due to phosphorus release under anoxic condition and the phosphorus uptake during the following oxic period (Zhu and Xiao, 2007). Table 2 Performance of SBR. Parameters Influent Effluent Removal Conc. ppm Range of concentration efficiency, % after 9 th day BOD, mg/l 1500 110-390 85.5 COD, mg/l 1972 280-499 80.3 TP, mg/l 1058 89-165 87.2 TN, mg/l 1720 500-800 61.0 NH 3 -N, mg/l 948 564-758 31.2 NO 3 - -N, mg/l 377 153-191 54.5 Fig. 9 Variation of TS concentration and removal TS, % TVS, % TSS, % 1.20 0.67 0.67 0.30-0.53 0.06-0.36 0.08-0.20 61.4 62.8 77.6 TVSS, % 0.33 0.00-0.2 73.2 Fig. 10 Variations of ph and temperature with time. 4. Nitrogen removal Total nitrogen (TN) is composed of organic nitrogen, ammonia nitrogen (NH 3 -N) including ammonium nitrogen (NH + 4 -N), nitrite nitrogen (NO - 2 -N) and nitrate nitrogen (NO - 3 -N). High mixed liquor suspended solid (MLSS=2.5%) concentration in the reaction tank helps to create anoxic condition as soon as the air supply is cut after the aeration phase to achieve de-nitrification for nitrogen removal. DO concentrations were varied from 6.9 to 5.1 mg/l with a mean value of 6 mg/l during the oxic phase. Nitrogen tied up in high energy compounds such as amino acids and amine is organic nitrogen. One of the intermediate compounds formed during biological metabolism is ammonia nitrogen. Aerobic decomposition

52 Engineering in Agriculture, Environment and Food Vol. 4, No. 2 (2011) changes NH 3 -N into NO - 2 -N and finally into NO - 3 -N. The influent concentration of total nitrogen was 1720 mg/l and effluent concentrations decreased to 780 mg/l by the 9 th day and varied from 500 mg/l to 800 mg/l with a mean of 670.0 mg/l after this as shown in Fig. 6. During this period the average removal efficiency of total nitrogen was 61.0±5.7%. The influent concentration of NH 3 -N was 948 mg/l and effluent concentrations decreased to 664 mg/l by the 9 th day and varied from 564 mg/l to 758 mg/l with a mean of 651.9 mg/l after this (Fig. 7). During this period the average removal efficiency of NH 3 -N was 31.2±6.8%. The low reduction efficiency for ammonia nitrogen was possibly due to the continuous transformation of organic nitrogen to ammonia nitrogen despite the fact that nitrification occurred. The influent concentration of NO - 3 -N was 377 mg/l and effluent concentrations decreased to 191 mg/l by the 9 th day and varied from 153 mg/l to 191 mg/l with a mean of 171.6 mg/l after this as shown in Fig. 8. During this period the average removal efficiency of NO - 3 -N was 54.5±2.8%. 4. Changes of others parameters in this study Total solids (TS) varied from the initial value of 1.2% to 0.48% on the 9 th day of the experiment and the average reduction rate of TS from the 9 th day to the end of testing was 61.4% as shown in Fig. 9. A similar tendency appeared in TVS, with marginally higher reduction being observed in TSS and TVSS as shown in Table 2. Oxic/anoxic SBR experiments for treating swine wastewater (Ra et al.,2000) showed that removal efficiencies for TS, TVS, TSS and TVSS were 75.3%, 84.9%, 95.2% and 96.4% respectively, indicating higher removal efficiencies than those obtained during this study. The solids and hydraulic retention time (SRT and HRT) employed for their SBR were 10 days and 4.7 days respectively. Lo et al. (1991) decomposed pig wastewater using an aerobic SBR with 23 hours of HRT and 14 days of SRT and achieved only about 23% and 31% of TS and TVS reduction respectively, indicating a much lower efficiency than this experiment possibly due to the low HRT and SRT. Another experiment was also conducted using an anaerobic SBR to treat swine wastewater in which the removal efficiency was found to be 59% for TVSS (Ng, 1989) showing marginally lower efficiency than for this experiment. It was found that the effluent appeared to have very low turbidity although this was not measured. The ph of the solution inside the reactor increased over the experiment with fluctuation from the initial value of 7.47 to the maximum value of 8.68 as shown in Fig. 10. IV Conclusions The swine wastewater was decomposed using a sequencing batch reactor (SBR) with 5 days of HRT and 41 days of SRT. The removal efficiencies of BOD, COD and TP were 85.5%, 80.3% and 87.2% respectively, indicating good removal rates. However, the removal efficiencies of TN, NH 3 -N and NO - 3 -N were 61.0%, 31.2% and 54.5%,respectively, exhibiting poor performance, possibly due to the lack of enough carbon sources and an inadequate aeration level for nitrification and de-nitrification during the oxic and anoxic periods. It was also observed that removal efficiencies of TS, TVS, TSS and TVSS were 61.4%, 62.8%, 77.6% and 73.2% respectively. This study showed that an SBR system can be used to get good results for BOD, COD and nutrient removal in swine wastewater treatment if supplied with enough carbon source for de-nitrification and optimum aeration for nitrification. References APHA. 1998. Standard methods for the examination of water and wastewater, 21th Edition, American Public Health Association, Washington D.C. Bicudo, J. R, J. J. Classen, C.D. Goldsmith and T. Smith. 1999. Effects of aeration cycles and hydraulic retention time on the sequencing batch treatment of flushed swine manure. Advances in Environmental Research, 3 (1), 58-73. Hach Inc. 1993. Procedure Manual. Loveland, CO. Imura, M., E. Suzuki., T. Kitao and S. Iwai. 1993. Advanced treatment of domestic wastewater using sequencing batch reactor activated sludge process. Water science and Technology, 28, 267-274. Jun Zhu and W. Xiao. 2007. Full-scale sequencing batch reactor treatment for milking center wastewater. Presentation at the ASABE Annual International Meeting, (American Society of Agriculture and Biological Engineers). Kulikowska D., E. Kliminuk and A. Drzewicki. 2007. BOD5 and COD removal and sludge production in SBR working with or without anoxic phase. Bioresource Technology, 98, 1426-1432. Kim, C. W., M. W. Choi, J. Y. Ha and T. J. Park. 2000. Optimization of operating mode for sequencing batch reactor treating Piggery wastewater. Presented at the 2 nd International symposium on sequencing batch reactor technology, International Water Association. Kim, J. H., M. Chen., N. Kishida and R. Sudo. 2004. Integrated real time control strategy for nitrogen removal in swine wastewater treatment using sequencing batch reactors. Water Research, 38 (14-15), 3340-3348. Lee D. S., C. O. K. Jeon and J. M. Park. 2001. Biological nitrogen removal with enhanced phosphate uptake in a sequencing batch

ISLAM, PARK, ALAM : Treatment of Swine Wastewater using Sequencing Batch Reactor 53 reactor using single sludge system. Water Research, 35 (16), 3968-3976. Lo K. V., P. H. Liao and R. J. Van Kleeck. 1999. A full-scale sequencing batch reactor treatment of dilute swine wastewater. Canadian Agricultural Engineering, 33, 193-194. Mohseni-Bandpi, A. and H. Bazari. 2004. Biological treatment of dairy wastewater by sequencing batch reactors. Iranian J. Env. Health sci Eng, 1(2), 65-69. Ng W-J. 1989. A sequencing batch anaerobic reactor for treating piggery wastewater. Biological Wastes, 28, 39-51. Ra C. S., K. V. Lo., J. S. Shin and B. J. Hong. 2000. Biological nutrient removal with an internal organic carbon source in piggery wastewater treatment. Water Research, 34(3), 965-973. Rusten, B. and H. Eliassen. 1993. Sequencing batch reactors for nutrient removal at small wastewater treatment plants. Water Science and Technology, 28, 233-242. Tilche A., E. Bacilieri., G. Bortone., and L. Stante. 1999. Biological phosphorus and nitrogen removal in a full scale sequencing batch reactor treatment piggery wastewater. Water Science and Technology, 40(1), 199-206. Zhu J., Z. Zhang and C. Miller. 2006. A laboratory scale sequencing batch reactor with the addition of acetate to remove nutrient and organic matter in pig slurry. Biosystems Engineering, 93(4), 437-446. (Received: 5.November. 2009, Accepted: 9.March. 2011)