Performance of a Full-Scale Biofilm System Retrofitted With an Upflow Multilayer Bioreactor As a Preanoxic Reactor for Advanced Wastewater Treatment

By An, Jin-Young Kwon, Joong-Chun; Ahn, Deog-Won; Shin, Hang-Sik; Won, Sung-Ho; Kim, Byung-Woo

ABSTRACT: To enhance nitrogen removal in an existing microbial contact oxidation (MCO) system with a treatment capacity of 900 m^sup 3^/d, an upflow multilayer bioreactor (UMBR) was chosen as a preanoxic reactor for the removal of organic matter and nitrate. The removal performance of the retrofitted plant was evaluated during the startup phase at a low temperature in winter. The high removal (>80%) of organic matter and suspended solids in the UMBR provided stable nitrification conditions in the MCO system, as a result of the substantial reduction in organic matter and solids loaded onto the MCO system. This treatment system showed a stable nitrogen removal efficiency of 75.3%, even in the low temperature range 7 to 10[degrees]C. Phosphorus was completely removed by chemical precipitation. Production rates of excess sludge, as a function of the loads of influent flowrate and biological oxygen demand (BOD), were 0.022 kg dry solid/m^sup 3^ wastewater and 0.132 kg dry solid/ kg BOD.

Water Environ. Res., 80, 757 (2008).

KEYWORDS: upflow, upflow multilayer bioreactor, wastewater treatment, biological nutrient removal, nitrogen removal.

doi:10.2175/106143008X276723

Introduction

Discharge of untreated or insufficiently treated wastewater containing nitrogen (N) and phosphorus (P) causes severe environmental problems for aquatic life, such as eutrophication (Effler et al., 1990; Luostarinen et al., 2006). The application of biological nutrient removal processes to the wastewater treatment field has been gaining increasing attention in Korea, in accordance with the strengthening environmental regulations and government policies. The Korean regulations concerning wastewater treatment, enforced since 2004, require strict effluent limits of less than 10 mg/L 5-day biochemical oxygen demand (BOD^sub 5^), 10 mg/L suspended solids (SS), 20 mg/L nitrogen, 2 mg/L phosphrus, and 1000 colony- forming units/mL of E. coli. To minimize the health and environmental risks in rural communities and ensure water quality protection at an early stage, the Ministry of Environment of the Republic of Korea has been focusing on the construction and upgrade of small wastewater treatment plants (WWTPs, treatment capacity 50 to 1000 m^sup 3^/d), which can serve small rural communities. A wide array of unit processes, known to be effective in nutrient removal, has been tried. However, these processes lead to an increase in energy consumption and operational complexity (Foess et al., 1999). Among the existing aerobic treatment plants, aerobic biofilm processes, such as contact oxidation systems, rotating biological contactors, aerobic biofilters, trickling filters, and fixed- or moving-bed biofilm reactors, cannot be easily retrofitted and upgraded with anaerobic or anoxic reactors for the removal of nitrogen and phosphorus, because they are not very compatible with continuously stirred tank reactors (CSTR) used as anoxic reactors, as a result of requiring an intermediate clarifier used to separate the denitrifying mixed liquor and to provide return activated sludge to the anoxic tank.

In an attempt to develop a more cost-effective, simple, and stable WWTP for small communities, many studies have focused on a combination of the anaerobic and aerobic processes (Bodik et al., 2003; Luostarinen et al., 2006), for example, the anaerobic-filter- activated sludge system (Kocadagistan et al., 2005), upflow- anaerobic-sludge-blanket-attached aerobic filter (Chernicharo and Machado, 1998; Collivignarelli et al., 1990), and anaerobic-baffled- reactor-activated sludge system (Garuti et al., 1992). These systems have some advantages, such as their low excess sludge production and low energy consumption, while their partial denitriflcation and poor removal of organic matter in the anaerobic reactor at low temperature make it difficult to select an appropriate following aeration process to remove residual organic matter and nitrogen (Bodik et al., 2003). Anaerobic systems for the treatment of low- strength wastewater are not suitable in Korea, with its average atmospheric temperature of less than 0.4[degrees]C in winter.

The upflow multilayer bioreactor (UMBR) is a patented unit process (U.S. Patent 6,352,643) able to act as a primary settling tank, anaerobic tank, anoxic tank, and thickener (Kwon et al., 2002). The upflow mode in a UMBR can be allowed to maintain a high biomass concentration, depending on the upflow velocity or hydraulic flux. The UMBR has low operational costs, as a result of mixing with the upflow stream, a rotational distributor, and no mechanical agitator. In a CSTR for denitrification, typical power requirements for mechanical mixing in the anoxic zone range from 8 to 13 kW/1000 m^sup 3^ (Metcalf & Eddy, 2003). However, the UMBR requires just 0.4 to 2.2 kW in the capacity range 30 to 1000 m^sup 3^, because the distributor is attached to the shaft of the cycloreducer, rotating at a low revolution rate (tip speed = 4 to 6 m/min) in the UMBR (Kwon et al., 2003). The UMBR is useful for retrofitting existing aerobic treatment systems for nutrient removal, because it can be used with both single- and dual-sludge systems, according to the operational mode. In the case of a singlesludge system, the activated sludge at the top of the UMBR overflows into an aeration basin. In contrast, for a dual-sludge operation, sludge blankets are kept in the UMBR, and then the supernatant at a concentration of less than 30 mg SS/L is introduced to the aerobic biofilm system, without an intermediate clarifier.

In this study, a full-scale wastewater treatment plant, with a treatment capacity of 900 m^sup 3^/d, was retrofitted with an existing microbial contact oxidation (MCO) system and the newly established UMBR without an intermediate clarifier. The process was evaluated during 4 months of startup operation at low temperatures (7 to 14[degrees]C) in winter. With the main objective to assess the effectiveness of a UMBR as a predenitrification process ahead of an aerobic biofilm process, for the improvement of nitrogen removal, the specific objectives of this study were to investigate the following:

(1) Process stability and performance of the full-scale treatment plant,

(2) Effect of treatment performance of the UMBR on nitrification efficiency in the MCO system, and

(3) Characteristics of nitrogen removal in this dual-sludge system under various BOD-to-nitrogen ratios at a low temperature.

Materials and Methods

Retrofitted Wastewater Treatment Plant. Three sets of UMBRs were newly established ahead of the MCO system covered with topsoil (Figure 1). Each UMBR has a capacity of 87 m^sup 3^, with an internal diameter of 4.2 m and a height of 6.3 m. Influent wastewater and internal recycle are uniformly fed to the bottom of the UMBR by rotating distributors via a feed-well situated at the top of the UMBR. The three distributors on vertical/ horizontal baffles in each UMBR are attached to the shaft of the cycloreducer, which rotates at a tip speed of 4 to 5 m/min, to prevent shortcircuits, nonideal flow, or channeling between solid and water, and to distribute influent uniformly (Figures 1 and 2).

The UMBR is divided into 3 layers on the basis of the upflowing position, as follows:

(1) The dense sludge layer (depth = approximately 0.4 m) below the distributor acts as a thickener, in which excess sludge can be enriched from 20 000 to 40 000 mg/L.

(2) The sludge blanket above the distributor is the anoxic layer, in which nitrate supplied with the internal recycle from the MCO is denitrified to nitrogen gas using the organic matter in the wastewater as an electron donor.

(3) In the supernatant layer above the sludge blanket, the influent passing through the sludge blanket can be clarified to below 30 mg SS/L. When the sludge blanket level in the UMBR, as detected by an optical or ultrasonic sensor, is higher than the designated level (maximum limit = approximately 80% of total height), the excess sludge stored in the thickening layer can be automatically or manually drained by a discharge pump.

The existing MCO system, in which the top layer is covered with soil and grass, consists of a sedimentation basin, a contact oxidation basin #1, a clarifier, an MCO basin #2, and a final clarifier (Figure 1). The MCO basin is filled with porous media (gravel with a diameter of 25 to 45 mm), at a filling ratio of less than 50%. The sedimentation basin and clarifiers are equipped only with airlift pumps for discharging the sediment, excluding any mechanical sludge collector. The capacities of this plant are summarized in Table 1.

Operation of the Full-Scale Plant. The wastewater pretreated by the automatic barscreens, with a bar space of 3 to 20 mm, is stored in an equalization tank having a capacity of 412 m^sup 3^. The wastewaters in the equalization tank are automatically fed at a variable flowrate, depending on the hydraulic level, into the distribution tank of the UMBR, to which the nitrified solutions are recirculated from the end of the MCO system at a recycle ratio corresponding to 150 to 200% of the influent flowrate. After denitrification and removal of organic matter and suspended solids in the UMBR, the remaining organic matter and ammonia are treated in the MCO system. The remaining suspended solids are removed by the automatic upflow-type sandfilter, to which a polyaluminum chloride (PAC) solution containing 11% aluminum oxide (Al^sub 2^O^sub 3^) is supplied for the precipitation of the remaining phosphorus. The backwashing water of the sandfilter is continuously returned to the equalization tank at a flowrate of 3 to 6 m^sup 3^/h. The final effluent after sand filtration is discharged to the environment after irradiating it with a UV disinfector. The operation parameters of this plant are summarized in Table 1. Sludge Inoculation. The UMBR was inoculated with the thickened sludge of more than 8000 mg/ L of mixed liquor suspended solids (MLSS) obtained from a gravity thickener in the adjacent Yang-Pyeong wastewater treatment plant (Kyunggi province, Korea), which is designed as a 5-stage biological nutrient removal (BNR) system, with a treatment capacity of 13 000 m^sup 3^/d. Approximately 90 m^sup 3^ of the thickened sludge was inoculated to each UMBR, to maintain a sludge blanket level of more than 3 m.

Analysis. Samples taken regularly from each unit process were analyzed for the purpose of monitoring the BOD^sub 5^, chemical oxygen demand by dichromate method (COD^sub Cr^), suspended solids, total nitrogen (TN), ammonium-nitrogen (NH^sub 4^^sup +^-N), nitrate- nitrogen (NO^sub 3^^sup -^-N), total phosphorus (TP), orthophosphate (PO^sub 4^^sup 3-^-P), alkalinity, MLSS, and mixed liquor volatile suspended solids (MLVSS), according to Standard Methods (APHA et al., 1998). The chemical oxygen demand by permanganate method (COD^sub Mn^) (COD^sup Mn^) was measured with potassium permanganate (KMnO), according to Korean Standard Methods (Ministry of Environment of the Republic of Korea, 2004).

The mixed liquor in the middle of the UMBR (depth = approximately 4m) was collected by a deep-water sampler (JT-1, Lamotte Co., Chestertown, Maryland), to measure the concentration of the sludge blanket. The dissolved oxygen (DO), oxidation-reduction potential (ORP), and pH were measured by a portable pH/DO/ORP meter (D-25, Horiba Co., Kyoto, Japan). The sludge depth in the UMBR was analyzed by an ultrasonic sludge level meter (Sondar 3000, IS Technologies Co., Kyunggi, Korea) with an ultrasonic sensor, which was mounted below the surface of the water on top of the UMBR. The ultrasonic sludge interface analyzer was calibrated regularly by a portable MLSS/interface meter with an optical sensor (SS-5Z, K.R.K. Co., Saitama, Japan) capable of detecting the sludge interface and water depth.

Results and Discussion

Removal Performance of Existing Microbial Contact Oxidation System Before the Retrofit Table 2 shows the removal performance of the existing MCO system over a period of 1 month in the temperature range 18 to 20[degrees]C before the retrofit for nutrient removal. The average removal efficiencies of BOD^sub 5^, suspended solids, total nitrogen, ammonia, and total phosphorus in the influent wastewater were 97.1, 94.9, 13.5, 59.8, and 22.7%, respectively. The MCO system, which simultaneously treated organic matter and nutrients, showed good removal efficiencies of organic matter and suspended solids, but very poor nutrient removal rates. In particular, the nitrification rate was very unstable, varying in the range 47 to 77%, depending on the organic or hydraulic loading rate, and led to a decrease in the removal efficiency of total nitrogen. In attached-growth systems used for nitrification, most of the BOD must be removed before nitrifying organisms can be established. Heterotrophic bacteria have a higher biomass yield and can thus dominate the surface area of the biofilm over nitrifying bacteria in the presence of organic matter (Metcalf & Eddy, 2003). A postdenitrification process, where the anoxic zone follows the aerobic zone for nitrification, requires external carbon sources for denitrification, which lead to increasing the cost and complexity of wastewater treatment (Luostarinen et al., 2006). However, in a predenitrification process, where the anoxic zone is ahead of the aerobic zone for nitrification, the electron donor for denitrification is provided by organic matter in the raw wastewater. For this reason, the UMBR with high removal of BOD and suspended solids was chosen as a predenitrification reactor to improve the availability and performance of the existing MCO system, rather than a postdenitrification biofilter with heterotrophic or autotrophic denitrifiers.

Startup Operation of the Retrofitted Plant The inoculation of the sludge was performed without any organic matter or supplemental loading for a period of 1 week before feeding the influent wastewater to the UMBR. When the sludge blanket level in the UMBR was kept at approximately 3 m, influent wastewater was fed to the UMBR, resulting in a very quick startup period within 1 week. Figure 3 shows the removal of nitrogen and phosphorus in each unit process for the 10-day period after startup. The internal recycle ratio (internal recycle flowrate/influent flowrate) was kept between 1.5 and 2.5, to maintain anoxic conditions in the range of ORP from -50 to 0 mV and prevent the emission of odors generated by the anaerobic metabolism. As the BOD and suspended solids concentrations incoming to the MCO were reduced to less than 30 mg/L, without a lag period for acclimation in the UMBR after startup, the ammonium concentration in the MCO decreased dramatically, and then total nitrogen removal was stabilized quickly. Final effluent containing approximately 10 mg-N/L was obtained in 5 days. The inoculated sludge, containing polyphosphorus in the previous 5-stage BNR system, released phosphorus at a concentration up to 2.9 mg-P/L in the UMBR during the first 3 days after feeding the influent wastewater. However, as the backwashing water containing PAC from the sand filter was returned to the UMBR, the concentration of phosphorus in the UMBR effluent decreased to less than 1.0 mg-P/L after 10 days.

Figure 4 shows the effects of the upflow velocity on the MLSS concentration in the UMBR. The MLSS concentration decreased with increasing the upflow velocity, which led to a variation in the height of the sludge blanket. The concentration of the sludge blanket at the depth of 3 m below the surface varied from 5200 to 7840 mg/L, at an upflow velocity ranging from 1.12 to 1.62 m/h. During the entire period of this study, the sludge in the UMBR showed good settleability, with a sludge volume index (SVI) of less than 70 mL/g, and a dense sludge blanket of more than 5000 mg MLSS/ L was maintained.

Removal Performance at Steady-State. Removal of Organic Matter and Suspended Solids in the Upflow Multilayer Bioreactor. The results obtained in this study during the 4 months after stabilization of the plant are summarized in Table 3. This study was performed in winter, from November 2004 to February 2005. The water temperatures decreased continuously from 14.4[degrees]C at startup to 7.3[degrees]C at the end of this study. Figure 5 shows variation in the BOD^sub 5^ and suspended solids at each unit process of this plant. The COD^sub Mn^ was analyzed with potassium permanganate according to Korean legislation. The COD by dichromate (COD^sub Cr^) of raw wastewater was measured irregularly. The ratio of COD^sub Cr^/ COD^sub Mn^ of raw wastewater, which can help presume COD^sub Cr^ values from COD^sub Mn^ values, was 2.6 (2.1 to 3.9), and its standard deviation was 0.74.

The average BOD^sub 5^, COD^sub Mn^, and suspended solids concentration in the influent (169, 97.7, and 93 mg/L) and the UMBR effluent (26.1, 16.4, and 14.1 mg/L) corresponded to removal efficiencies of 83.4, 83.2, and 81.9%, respectively. Considering the concentration of the soluble BODs, which was irregularly measured with the filtrated sample, in the UMBR effluent of approximately 6.6 (4.7 to 8.7) mg/L, it seems that most of the readily biodegradable organics and suspended solids could be removed in the UMBR by the anoxic metabolism.

Anaerobic pretreatment may be beneficial to BNR (Elmitwalli et al., 2001; Kalyuzhnyi et al., 2003; Luostarinen et al., 2006), as a result of efficient removal of organic matter and dissolution of paniculate organic matter, which may enhance nitrification because of less competition for biofilm space between carbon-utilizing heterotrophic microorganisms and autotrophic nitrifiers (Metcalf & Eddy, 2003).

Such a high removal of BODs and suspended solids in the UMBR can offer a stable nitrification condition in a following aeration system, as a result of the substantial reduction in the amounts of organic matter and solids loaded onto the following aeration system. This suggests that the UMBR is useful as a predenitrification reactor for the retrofit of existing biofilm systems for nitrogen removal.

Nitrogen Removal. The average concentrations of total nitrogen at each stage were 34.7 (20.5 to 52.9) mg-N/L in the influent wastewater, 8.6 (5.1 to 12.4) mg-N/L in the UMBR effluent, and 8.4 (5.1 to 11.9) mg-N/L in the final effluent after sand filtration, which correspond to a removal efficiency of 75.3 (61.2 to 83.5)% (Figure 6a). As shown in Figure 6b, the average concentration of nitrate in the UMBR effluent was approximately 1.5 (0.3 to 4.1) mg- N/L, and the average concentration of NH^sub 4^^sup +^-N in the final effluent after sand filtration was approximately 0.2 (0.02 to 0.4) mg-N/L. Even though there is a slight difference in the removal performance of nitrogen according to the daily operation conditions, most of the nitrate was denitrified in the UMBR, and NH^sub 4^^sup +^-N was completely nitrified in the MCO system.

Figures 6c and 6d show the effects of the BOD-to-nitrogen ratio and temperature on nitrogen removal, respectively, where this treatment system showed stable removal performances of total nitrogen, even in the low temperature range 7 to 10[degrees]C under the BOD-to-nitrogen ratios of 4.9 (1.9 to 9.4). One of the causes of the high removal of total nitrogen at the low temperature may be the long hydraulic retention time (HRT) of this system. However, as reported by Dies and Mavinic (2001), the 4-stage Bardenpho system using single sludge experienced significant inhibitions in both dentrification and nitrification at 1[degrees]0C, and the overall performance could not be improved by changing operation parameters, such as a decreased loading rate. In particular, percentage denitrification decreased to less than 5% of its potential, at the lowest ambient temperature of 10[degrees]C.

Considering serious inhibitions at low temperatures below 10[degrees]C, the dual-sludge operation in this system may be the other cause of its high nitrogen removal. In alternating anoxic/ oxic processes for nitrogen removal, facultative denitrifiers recycled from an aerobic reactor to an anoxic reactor exhibit a lag period of approximately 40 minutes before using nitrate as an electron acceptor. This is caused by the oxygen repression of the nitrate-reduction enzymes (Delwiche and Bryan, 1976). In addition, denitrification is compromised with enhanced phosphorus removal, because there is competition between phosphorus-accumulating organisms and denitrifiers for readily biodegradable substrate in the influent (Patel et al., 2006). However, with this system, using dual sludge grown in the separated basins without alternating, it can be kept to a more favorable condition of denitrification by lack of phosphorus removal (refer to the Phosphorus Removal and Excess Sludge Production section).

High nitrogen removal at a low BOD-to-nitrogen ratio appears to be caused by a high specific endogenous denitrification rate in the dense sludge blanket of the UMBR, where it has a concentration stratification of MLSS in the range 5 to 13 g/L, depending on upflowing position and upflow velocity (An et al., 2007). Beer et al. (1977) suggested that specific endogenous denitrification was not related to MLSS in the range 2.5 to 4.0 g/L. In a laboratory- scale study of a UMBR system using synthetic wastewater (Suh et al., 2006), however, the specific endogenous denitrification rate in the UMBR increased with MLVSS concentration, resulting in a 2.2-fold increase in specific endogenous denitrification rate (from 0.22 to 0.48 mg-N/g MLVSS -h) when the MLVSS was changed from 2 to 10 g/L. This may be influenced by the cryptic growth, that is, microbial growth on lysates released by microbial cell lysis, which can be amplified by prolongation of solids retention time (SRT) and decrease in sludge loading rate, occurred with high MLSS concentration (Liu and Tay, 2001; Low and Chase, 1999; Wei et al., 2003). As reported by Jung et al. (2004), it was possible to denitrify using lysates as an electron donor after fast depletion of substrates. Therefore, a dense anoxic sludge blanket grown only in the anoxic condition of the UMBR may have higher denitrification activity than can be expected in a CSTR with an alternating anoxic/ oxic system with single sludge.

It is also worth noting that high removal of organic matter in the UMBR prevents heterotrophic bacteria from dominating the surface area of biofilm over nitrifying bacteria, which affords favorable nitrification conditions in the MCO system. In addition, the nitrification process is estimated to be less temperaturedependent in biofilms than in activated sludge systems, where it is considered that the nitrification rate in a biofilm system decreases by approximately 4.5% per fall of 1[degrees]0C, in comparison with 10% per 1[degrees]C in a conventional activated sludge system (Bodik et al., 2003).

Phosphorus Removal and Excess Sludge Production. In this study, phosphorus was removed simply by chemical precipitation with a PAC solution (11% Al^sub 2^O^sub 3^) supplied to the sand filter at a flowrate of 0.9 to 1.5 L/h. The average total phosphorus concentrations were 3.69 (2.18 to 5.56) mg-P/L in the influent wastewater, 0.61 (0.17 to 1.87) mg-P/L in the UMBR effluent, 0.51 (0.12- to 1.17) mg-P/L in the MCO system, and 0.05 (0.02 to 0.20) mg- P/L in the final effluent after sand filtration (Figure 7). There does not appear to be the release and luxury uptake of phosphorus that occurs in an alternating anaerobic/aerobic process, because this system is extremely biased toward nitrogen removal.

The phosphorus concentration in the UMBR effluent decreased from 2.9 mg-P/L at startup to below 0.2 mg-P/L at the end of this study. This result can be explained by additional chemical precipitation in the UMBR, to which the backwashing water containing substantial alum floes from the sandfilter was returned. Considering the regulations in Korea requiring a concentration of total phosphorus of less than 2.0 mg-P/L, periodic addition of PAC to obtain the effluent below the regulation limit would help to reduce operation costs.

When the height of sludge blanket exceeded 5.0 m, the excess sludge in the thickening layer of the UMBR was manually drained for 0.5 to 1.0 hours at a flowrate of 3.5 m^sup 3^/h, by means of a screw pump, to prevent clogging. The characteristics and production rate of the excess sludge are summarized in Table 4. The average concentrations of the excess sludge were 28 473 mg MLSSA- and 18 962 mg MLVSS/L, with a volatile-solids-to-suspended-solids ratio of approximately 0.67. The production rates of excess sludge, as a function of the loads of the influent flowrate and BOD^sub 5^, were 0.022 kg dry solids/m^sup 3^ wastewater and 0.132 kg dry solids/kg BOD^sub 5^. Because of the low sludge production rate, the UMBR was operated at a longer SRT of 106.4 days. In a previous study, where a pilot-scale UMBR system for wastewater treatment was operated under a shorter HRT of 3.5 hours (2.5 hours based on the sludge blanket) and a suspended solids loading rate 4.8 times higher than this study, the SRT value and sludge production rate were 8.3 days and 0.3 kg dry solids/m^sup 3^, respectively (Kwon et al., 2005). Therefore, it is concluded that prolonged SRT and HRT and lower suspended solids loading rate lead to a substantial reduction in excess sludge production compared with the previous study (Kwon et al., 2005). Even under long SRT operation, the over-release of phosphorus and poor settleability of the sludge in the UMBR did not occur in this study.

Conclusions

In this study, a patented UMBR was successfully applied as a predenitrification reactor to retrofit an existing aerobic biofilm reactor (MCO) to the BNR process, without a primary settling tank and intermediate clarifier. The UMBR showed high removal efficiencies (>80%) of BOD^sub 5^ and suspended solids, resulting in a substantial reduction in the amount of organic matter and solids loaded onto the MCO system. This high removal performance of organic matter and suspended solids in the UMBR led to stable nitrification conditions being obtained in the MCO system. Dualsludge operation with a long SRT improved nitrogen removal in the complete system. An average nitrogen removal rate of 75.3% was obtained in the low temperature range 7 to 14[degrees]C, regardless of the BOD-to- nitrogen ratio. Phosphorus was completely removed by chemical precipitation with PAC. The production rates of excess sludge in the UMBR, as a function of the loads of influent flowrate and BOD^sub 5^, were 0.022 kg dry solids/m^sup 3^ wastewater and 0.132 kg dry solids/kg BOD^sub 5^, which correspond to a long SRT of 106.4 days in the UMBR. Based on the results obtained in this study, it can be concluded that the UMBR, as a predenitrification tank, is very attractive for the retrofit of an existing aerobic biofilm system for the purpose of nitrogen removal.

Credits

This work was supported by the Postdoctoral Research Program of Sungkyunkwan University (Suwon, Korea).

Submitted for publication July 13, 2007; revised manuscript submitted January 2,2008; accepted for publication February 14, 2008.

The deadline to submit Discussions of this paper is November 15, 2008.

References

American Public Health Association; American Water Works Association; Water Environment Federation (1998) Standard Methods for the Examination of Water and Waste water, 20th ed.; American Public Health Association: New York.

An, J. Y.; Kwon, J. C.; Ann, D. W.; Shin, D. H.; Shin, H. S.; Kim, B. W. (2007) Efficient Nitrogen Removal in a Pilot System Based on Upflow Multi-Layer Bioreactor for Treatment of Strong Nitrogenous Swine Wastewater. Process Biochem., 42, 764-772.

Beer, C.; Bergenthal, J. E; Wang, L. K. (1977) A study of Endogenous Nitrate Respiration of Activated Sludge. Proceedings of the 9th Mid-Atlantic Industrial Waste Conference, Lewisburg, Pennsylvania, Aug. 8-9; Bucknell University, Civil Engineering Department, Lewisburg, Pennsylvania, 207-215.

Bodik, I.; Kratochvfl, K.; Gasparikova, E.; Human, M. (2003) Nitrogen Removal in an Anaerobic Baffled Filter Reactor with Aerobic Post-Treatment. Bioresour. Technol., 86, 79-84.

Chemicharo, C. A.; Machado, R. M. G. (1998) Feasibility of the UASB/AF System for Domestic Sewage Treatment in Developing Countries. Water Sci. Technol, 38 (8-9), 325-332.

Collivignarelli, C.; Urbini, G.; Farneti, A.; Bassetti, A.; Barbaresi, U. (1990) Anaerobic-Aerobic Treatment of Municipal Waste waters with Full-Scale UASB and Attached Biofilm Reactors. Water Sci. Technol., 22 (1/2), 475-482.

Delwiche, C. C.; Bryan, B. A. (1976) Denitrification. Ann. Rev. Microbiol., 30, 241-262.

Effler, S. W.; Brooks, C. M.; Auer, M. T.; Doerr, S. M. (1990) Free Ammonia and Toxicity Criteria in a Polluted Urban Lake. J. Water Pollut. Control Fed., 62, 771-779.

Elmitwalli, T.; Zeeman, G.; Lettinga, G. (2001) Anaerobic Treatment of Domestic Sewage at Low Temperature. Water Sci. Technol., 44 (4), 33-40.

Foess, G. W.; Steinbrecher, P.; Garrett, G. S.; Smith, R. (1999) Cost and Performance Evaluation of Nutrient Removal at Small Wastewater Treatment Plants. Proceedings of the 72nd Annual Water Environment Federation Technical Exposition and Conference, New Orleans, Louisiana, Oct. 9-13; Water Environment Federation: Alexandria, Virginia. Garuti, G.; Dohanyos, M.; Tilche, A. (1992) Anaerobic-Aerobic Combined Process for the Treatment of Sewage with Nutrient Removal: The ANANOX(R) Process. Water Sci. Technol., 25 (7), 383-394.

Ilies, P.; Mavinic, D. S. (2001) The Effect of Decreased Ambient Temperature on the Biological Nitrification and Denitrification of a High Ammonia Landfill Leachate. Water Res., 35 (8), 2065-2072.

Jung, S. Y.; Miyanaga, K.; Tanji, Y.; Unno, H. (2004) Nitrogenous Compounds Transformation by the Sludge Solubilization Under Alternating Aerobic and Anaerobic Conditions. Biochem. Eng. J., 21, 207-212.

Kalyuzhunyi, S.; Gladchenko, M.; Epov, A. (2003) Combined AnaerobicAerobic Treatment of Landfill Leachate Under Mesophilic, Submesophilic and Psychrophilic Condtions. Water Sci. Technol, 48 (6), 311-318.

Kocadagistan, B.; Kocadasgistan, E.; Topcu, N.; Demircioglu, N. (2005) Wastewater Treatment with Combined Upflow Anaerobic Fixed- Bed and Suspended Aerobic Reactor Equipped with a Membrane Unit. Proc. Biochem., 40, 177-182.

Kwon, J. C.; An, J. Y.; Shim, K. B.; Shin, H. S.; Jun, H. B.; Park, H. S. (2003) Nitrogen and Phosphorus Removal of Pilot Scale Plant Using UMBR (Upflow Multi-Layer Bioreactor) as Anaerobic/ Anoxic Reactor. Proceedings of the 76th Annual Water Environment Federation Technical Exposition and Conference, Los Angeles, California, Oct. 11-15; Water Environment Federation: Alexandria, Virginia.

Kwon, J. C.; Park, H. S.; An, J. Y.; Shim, K. B.; Kirn, Y. H.; Shin, H. S. (2005) Biological Nutrient Removal in Simple Dual Sludge System with an UMBR (Upflow Multi-Layer Bioreactor) and Aerobic Biofilm Reactor. Water Sci. Technol., 52 (10-11), 443-451.

Kwon, J. C.; Shin, H. S.; Bae, B. W.; Yoo, K. S. (2002) Wastewater Treatment Plant Comprising Upflow Anaerobic Reactor and Wastewater Treatment Method Using Thereof. U.S. Patent 6, 352,643.

Liu, Y.; Tay, J. H. (2001) Strategy for Minimization of Excess Sludge Production from the Activated Sludge Process. Biotechnol. Adv., 19, 97-107.

Low, E. W.; Chase, H. A. (1999) Reducing Production of Excess Biomass During Wastewater Treatment. Water Res., 33 (5), 1119-1132.

Luostarinen, S.; Luste, S.; Valentin, L.; Rintala, J. (2006) Nitrogen Removal from On-Site Treated Anaerobic Effluents Using Intermittently Aerated Moving Bed Biofilm Reactors at Low Temperatures. Water Res., 40, 1607-1615.

Metcalf & Eddy (2003) Wastewater Engineering: Treatment and Reuse, 4th ed., Tchobanoglous, G., Burton, F. L., Stensel, H. D. (Eds.); Metcalf & Eddy: New York, 749-798.

Ministry of Environment of the Republic of Korea (2004) Korean Standard Methods for Water Quality Contamination, No. 2004-188; Ministry of Environment of the Republic of Korea: Gwacheon, Kyunggi- do (in Korean).

Patel, A.; Zhu, J.; Nakhla, G. (2006) Simultaneous Carbon, Nitrogen and Phosphorous Removal from Municipal Wastewater in a Circulating Fluidized Bed Bioreactor. Chemosphere, 65, 1103-1112.

Suh, C. W.; Lee, S. H.; Jeong, H. S.; Kwon, J. C.; Shin, H. S. (2006) Effects of Influent COD/N Ratio and Internal Recycle Ratio on Nitrogen Removal Efficiency in the KNR Process. Water Sci. Technol, 53 (9), 265-270.

Wei, Y.; Van Houten, R. T.; Borger, A. R.; Eikelboom, D. H.; Fan, Y. (2003) Minimization of Excess Sludge Production for Biological Wastewater Treatment. Water Res., 37, 4453-4467.

Jin-Young An1, Joong-Chun Kwon2, Deog-Won Ahn2, Hang-Sik Shin3, Sung-Ho Won1, Byung-Woo Kim1*

1 Department of Chemical Engineering, Sungkyunkwan University, Suwon, Korea.

2 R&D Center, Ecodigm Co., Ltd., Daejeon, Korea.

3 Department of Civil and Environmental Engineering, Korea Advanced Institute of Science and Technology, Daejon, Korea.

* Department of Chemical Engineering, Sungkyunkwan University, Suwon 440-746, Korea; e-mail: bwkim@skku.edu.

Copyright Water Environment Federation Aug 2008

(c) 2008 Water Environment Research. Provided by ProQuest LLC. All rights Reserved.