ABSTRACT In this work, the application of the contact-adsorption-regeneration-stabilization activated sludge process for the municipal wastewater treatment was investigated by using continuous experiments coupled with adsorption batch experiments. The process optimization for performance evaluation was studied. The obtained appropriate operational parameters of recycle ratio, hydraulic retention time (HRT) of the regeneration tank, HRT of the adsorption tank and solids retention time (SRT) of the system were 40%, 2 h, 30 min, and 6 d, respectively. Adsorptive kinetics and equilibrium were investigated with batch experiments of adsorption. The results showed that the activated sludge concentration was in a positive proportion to its adsorptive capability, but that temperature was in an inverse proportion to the adsorptive capability. Adsorptive equilibrium was reached within 15 min. In addition, the equilibrium data fitted well to both the Freundlich and Langmuir adsorption models.
Journal of Water and Environment Technology, Vol. 7, No. 2, 2009 - 83 - Contact-Adsorption-Regeneration-Stabilization Process for the Treatment of Municipal Wastewater Shao-Gen LIU* , **, Bing-Jie NI*, Lin WEl***, Yong TANG*,Han-Qing YU* *Department of Chemistry, University of Science & Technology of China, Hefei, 230026, China **Department of Environmental Engineering, Anhui Institute of Architecture & Industry, Hefei 230022, China ***School of Resource and Environmental Engineering, Hefei University of Technology, Hefei 230009, China ABSTRACT In this work, the application of the contact-adsorption-regeneration-stabilization activated sludge process for the municipal wastewater treatment was investigated by using continuous experiments coupled with adsorption batch experiments. The process optimization for performance evaluation was studied. The obtained appropriate operational parameters of recycle ratio, hydraulic retention time (HRT) of the regeneration tank, HRT of the adsorption tank and solids retention time (SRT) of the system were 40%, 2 h, 30 min, and 6 d, respectively. Adsorptive kinetics and equilibrium were investigated with batch experiments of adsorption. The results showed that the activated sludge concentration was in a positive proportion to its adsorptive capability, but that temperature was in an inverse proportion to the adsorptive capability. Adsorptive equilibrium was reached within 15 min. In addition, the equilibrium data fitted well to both the Freundlich and Langmuir adsorption models. Keywords: activated sludge, adsorption, adsorptive equilibrium, adsorption model, municipal wastewater, regeneration. INTRODUCTION With the sustained and rapid economic growth in the last three decades, China is undergoing massive urbanization. Municipal domestic sewage, which results from both the increasing population and the improved lifestyle of the people, is one of the major problems, which affect China’s environmental quality and the sustainable development. Municipal wastewater treatment in China faces serious shortage of financing and land area. Thus, considering the municipal wastewater treatment situations in China and the existence problems, the development of municipal wastewater treatment systems with small land area requirements and low investment is essential. Recently, a cost-effective and land-saving municipal wastewater treatment process, i.e., the contact-adsorption-regeneration-stabilization (CARS) process, was developed by our group. In such a process, activated sludge utilizes its physical, chemical and biological synergistic function to adsorb pollutants (e.g., suspended organic matter, soluble organism and ammonium) in the adsorption tank. Then, the sludge with all these pollutants directly flows into the secondary clarifier for solid-liquid separation. After this, the concentrated sludge is pumped to the regeneration stabilization tank for bio-regeneration. Finally, the renewed activated sludge comes out of the bio-regeneration tank under the hydraulic drive and returns to the adsorption tank for reuse. In this process, the high adsorption capacity of activated sludge is fully utilized and can ensure a better effluent quality (Ulrich and Smith, 1951; Jacobsen et al., 1996; Address correspondence to Han-Qing YU, Department of Chemistry, University of Science & Technology of China, Hefei, 230026, China. Email: hqyu@ustc.edu.cn Received November 29th, 2008, Accepted February 17th, 2009 Journal of Water and Environment Technology, Vol. 7, No. 2, 2009 - 84 - Grady et al., 1999; Huang et al., 2000). At the same time, the sludge settling characteristics are significantly improved after the bio-flocculation process occurring in the adsorption tank (Bruus et al., 1992; Droppo and Ongley, 1997; Biggs and Lant, 2000; Wilen et al., 2003). As a result, a high concentration of the sludge in regeneration reactor is achieved and the HRT is greatly shortened. Therefore, the land area requirement could be significantly reduced and accordingly construction investment would be cut down. This paper reports the experimental results of the CARS process for the municipal wastewater treatment. The technical feasibility of this process was demonstrated, and the process optimization was performed. In addition, the adsorption characteristics of the activated sludge from the CARS system were evaluated with the adsorptive kinetics and equilibrium experiments. MATERIALS AND METHODS Reactor configuration and wastewater Three lab-scale reactors were used to evaluate the CARS process. The schematic diagram is shown in Fig. 1. The effective volumes of the adsorption tank, the regeneration tank and the secondary clarifier were 5 L, 12 L and 10 L, respectively. The adsorption tank had an inflow of 10 L/h. Domestic wastewater was collected with a central sewerage network for the preparation of the influent wastewater. The composition of the influent was given in Table 1. The average COD concentration in this wastewater was about 250 mg/L. Ef f l uent Wat er t ank Mixer Pump Adsorption tank Regeneration tank Pump Aer at or Flowmeter Sl udge excl udi ng Secondary cl ari f i er Air lift Return acti vated sludge I nf l uent Fig. 1 -The schematic diagram of CARS system Seed sludge The reactors were inoculated with activated sludge taken from the Wangxiaoying Wastewater Treatment Plant, Hefei, China. The seeding sludge had a sludge age of 15 d, a mixed liquor suspended solids (MLSS) concentration of 6.0 g/L and a sludge volume index (SVI) of 90.5 mL/g. Journal of Water and Environment Technology, Vol. 7, No. 2, 2009 - 85 - Table 1 - Composition of the Municipal Wastewater Component Level SS (mg/L) 100~300 COD (mg/L) 100~300 SCOD (mg/L) 30~150 NH 3 -N (mg/L) 15~35 PO 4 3 -P (mg/L) 1.0~2.5 pH 6.5~7.5 Continuous tests Four series of experiments were conducted to investigate the individual effect of the solids retention time (SRT), hydraulic retention time (HRT) of the regeneration tank, recycle ratio (R), and HRT of the adsorption tank. In Series I, the recycle ratio (R) and the SRT were kept at 40% (with a recycling sludge flowrate of 4 L/h) and 6 d, respectively, the HRT of the regeneration tank was decreased stepwise from 3 h to 1 h for the three reactors, while the HRT of the adsorption tank were kept at 30 min. In Series II, R was increased from 20% to 60%, and the SRT, HRT of regeneration tank, and HRT of the adsorption tank were kept at 6 d, 2 h and 30 min, respectively; in Series III, the HRT of the adsorption tank was lowered stepwise from 30 min to 10 min, while the SRT, HRT of the regeneration tank, and R were kept at 6 d, 2 h and 40%, respectively; In Series IV, the SRT was increased stepwise from 3 d to 12 d, the HRT of the regeneration tank, recycle ratio R, and HRT of the adsorption tank were kept at 2 h, 40% and 30 min, respectively. Each series consisted of 3-4 runs. Each run lasted over 3 weeks to ensure the reactors to reach steady-state before changing to the next condition. Effluent compositions were continuously monitored. Only those obtained under steady-state conditions are reported here. Batch tests The sludge samples were taken from the incubation reactors and were then centrifuged and washed using tap water for three times. Before the adsorption tests, the sludge was diluted to the required MLSS level with the same mineral solution, and the initial pH of the mixed liquor was adjusted to 7.0. The adsorption experiments were conducted using 150-mL Erlenmeyer flasks in duplicate. The flasks were filled with 100-mL activated sludge solution, and spiked with wastewater. They were immediately sealed with rubber plugs and were shaken on a thermostatic rotary shaker at 125 rpm and 20°C. Samples were collected from the flasks at intervals. The mixed liquor was centrifuged at 6000 rpm for 10 min. The supernatant samples were acidified to pH 2.0 by 1 M HCl and stored in a refrigerator at 4°C before they were analyzed. The adsorption isotherms were developed using different sludge and substrate levels. The temperature effect was studied at 20, 30, and 40°C, respectively. Analytical methods Measurement of MLSS, mixed liquor volatile suspended solids (MLVSS), effluent suspended solids (SS), NH 4 + -N, PO 4 3- -P, and COD was performed according to the Standard Methods (APHA, 1995). Journal of Water and Environment Technology, Vol. 7, No. 2, 2009 - 86 - RESULTS AND DISCUSSION Effect of SRT The SRT effect on the CARS performance is shown in Fig. 2. The results revealed that the SRT had a significance effect on the COD, ammonium and phosphorus removals. An increase in SRT from 3 d to 6 d resulted in a significant increase in the COD and NH 4 + -N removals. However, a further increase in SRT from 6 d to 12 d led to a little decrease in the COD and NH 4 + -N removals. In general, the PO 4 3- -P removal decreased slightly with an increase in SRT. The PO 4 3- -P removal in the CARS system was mainly attributed to the sludge discharge from the system, which was governed by the SRT. Thus, the SRT affected the CARS performance significantly. The system had the maximum pollutant removal capacity at an SRT of 6 d. 0246810 0 20 40 60 80 100 0246810 0 10 20 30 40 50 0246810 0 10 20 30 40 50 60 70 COD removal (%) Time (d) 3 d 6 d 12 d NH 4 + -N removal (%) Time (d) PO 4 3- -P removal (%) Time (d) Fig. 2 - Variations of COD, NH 4 + -N, and PO 4 3- -P removals vs. time at different SRTs Effect of HRT of the regeneration tank Fig. 3 illustrates the effect of HRT of the regeneration tank on the CARS system performance. The COD removal increased with an increase in HRT of the regeneration tank from 1 h to 2 h, but decreased from 2 h to 3 h. As shown in Fig. 2, the PO 4 3- -P removal increased as the HRT of the regeneration tank increased. The NH4 + -N removal was not very sensitive to the HRT. It slightly increased as the HRT increased from 1 h to 2 h, but slightly decreased when the HRT increased from 2 h to 3 h. This suggests that the 2 h of HRT of the regeneration tank was appropriate for this CARS process. 02468 0 20 40 60 80 02468 0 10 20 30 40 50 02468 0 10 20 30 40 50 60 COD removal (%) Time (d) 1 h 2 h 3 h NH 4 + -N removal (%) Time (d) PO 4 3- -P removal (%) Time (d) Fig. 3 - Variations of COD, NH 4 + -N, and PO 4 3- -P removals vs. time at different HRTs of the regeneration tank Effect of recycle ratio (R) The recycle ratio of the CARS process played an important role in the substrate removal. They had a significant influence on the output variables (Fig. 4). It is found that ±50% fluctuation in the recycle ratio resulted in ±10% fluctuation in the COD removal, ±30% Journal of Water and Environment Technology, Vol. 7, No. 2, 2009 - 87 - fluctuation in the NH 4 + -N removal and ±15% fluctuation in the PO 4 3- -P removal. An increase in the recycle ratio resulted in an increase in the COD removal and an increase in NH 4 + -N and PO 4 3- -P removals (Fig. 3). Such influences were associated with the direct effect of recycle ratio on the sludge concentrations in the adsorption and regeneration tanks. 02468 0 20 40 60 80 100 02468 0 10 20 30 40 02468 0 10 20 30 40 50 60 COD removal (%) Time (d) 20% 40% 60% NH 4 + -N removal (%) Time (d) PO 4 3- -P removal (%) Time (d) Fig. 4 - Variations of COD, NH 4 + -N, and PO 4 3- -P removals vs. time at different recycle ratios Effect of HRT of the adsorption tank 02468 0 20 40 60 80 100 02468 0 10 20 30 40 50 02468 0 20 40 60 80 100 COD removal (%) Time (d) 10 min 20 min 30 min NH 4 + -N removal (%) Time (d) PO 4 3- -P removal (%) Time (d) Fig. 5 - Variations of COD, NH 4 + -N, and PO 4 3- -P removals vs. time at different HRTs of the adsorption tank The effect of HRT of the adsorption tank to the CARS process is shown in Fig. 5. The COD removal changed slightly with a change of the HRT of the adsorption tank. The NH 4 + -N removal increased with an increase in the HRT. The HRT had a similar sensitivity to the PO 4 3- -P removal (Fig. 4). An increase in the HRT led to a slight increase in the PO 4 3- -P removal. These results revealed that the selection of the operational parameters of the CARS process had a great influence on the overall system performance. Adsorptive capability of the sludge in CARS system The adsorptive rate and specific adsorptive capacity of the activated sludge are described with the following equations: 0 0 ( ) 100 Adsorptive rate= t CC C −× (1) MLSS CC q t t )( 0 − = (2) where q t is the specific adsorptive capacity of the sludge (g/g), C 0 and C t are the organic Journal of Water and Environment Technology, Vol. 7, No. 2, 2009 - 88 - substance concentrations before and after adsorption at certain time (t) (mg/L), and MLSS is the sludge concentration (mg/L). Different activated sludge concentrations were firstly applied in the adsorption batch tests to reveal the effect of the sludge concentration on the adsorption of organic substances in the municipal wastewater by sludge. Upon approaching the adsorption equilibrium, the adsorptive rate increased with an increase in MLSS concentration, whereas the specific adsorptive capacity decreased with the increasing MLSS concentration, as shown in Fig. 5. The relationship between adsorption time and adsorptive rate and specific adsorptive capacity at equilibrium is also shown in Fig. 6. The adsorptive rate and specific adsorptive capacity both increased with the increasing adsorption time. In addition, the contact time to reach equilibrium extended with the increasing initial organic substance concentrations (data not shown). It took 40 min for the organic substances removal to reach relatively constant at an initial concentration of about 500 mg/L, whereas 30 min for organic substances at an initial concentration of 300 mg/L. 0 10203040 0.03 0.04 0.05 0.06 0.07 0.08 0 100 200 300 400 500 600 20 40 60 80 MLSS (mg/L) Adsorptive rate (%) 0.00 0.05 0.10 0.15 0.20 0.25 q t (g/g) 0 10203040 40 50 60 70 80 Adsorption time (min) Adsorptive rate (%) 0.04 0.05 0.06 0.07 0.08 q t (g/g) q t (g/g) Adsorption time (min) 20 o C 30 o C 40 o C Fig. 6 - Effect of biomass concentration, adsorption time, and temperature on adsorptive rate or specific adsorptive capacity The temperature is an important factor influencing adsorption. Many studies reported that the ability of adsorption increased with the decreasing temperature either for dried sludge (Zhou, 1992; Zhou and Banks, 1993) or for activated carbon (McCreary and Snoeyink, 1980). Therefore, it is hypothesized that there existed common physicochemical principles underlying these adsorption processes. As shown in Fig. 5, the specific adsorptive capacity of the activated sludge increased with the decreasing temperature from 40°C to 20°C. Adsorption isotherm The Langmuir and Freundlich adsorption isotherm equations are the most widely used models for adsorption in an aqueous medium. The Langmuir and Freundlich equations are expressed as: m e me e q C bqq C += 1 (3) Fee KC n q loglog 1 log += (4) where q e is the adsorbed amount of organic substances per gram sludge (mg/g- TSS), Ce Journal of Water and Environment Technology, Vol. 7, No. 2, 2009 - 89 - is the solute equilibrium organic substances concentration (mg/L); q m and b are constant characteristic of the system for the Langmuir model that can be considered as an indicator of adsorptive capacity and appetency, respectively; and K F and n are constant characteristics of the system for the Freundlich model, which can be respectively considered as an indicator of adsorptive capacity and intensity (Esparza-Soto and Westerhoff, 2003). Table 2 - The Freundlich and Langmuir models by fitting the experimental data Model Freundlich model Langmuir model Parameters K F 1/n R 2 b/L.g -1 q m (g.g -1 ) R 2 Values 0.1315 1.3872 0.9886 0.0022 0.7209 0.9534 1.92.02.12.22.32.4 1.8 1.9 2.0 2.1 2.2 2.3 2.4 50 100 150 200 250 0.6 0.8 1.0 1.2 1.4 Experimental data Freundlich model log q e log C e Experimental data Langmuir model C e /q e C e (mg/L) Fig. 7 - The linearized Freundlich and Langmuir adsorption isotherms Although previous studies demonstrated that the biosorption of heavy metallic ions by activated sludge followed the Langmuir mode (Wu et al., 2004), the present study showed that the Freundlich isotherm equation (Fig. 7) also matched the experimental results well. The obtained values of the constant characteristics for the two models are listed in Table 2. CONCLUSIONS In this study, technical feasibility of this process was demonstrated. The effects of recycle ratio, HRT of the regeneration tank, HRT of the adsorption tank and SRT of the system were evaluated. The removal efficiency improved with the increased HRT of the adsorption tank. From an engineering application point of view, the appropriate operational parameters of the recycle ratio, HRT of the regeneration tank, HRT of the adsorption tank and SRT of the system were determined to be 40%, 2 h, 30 min, and 6 d, respectively. In this CARS system, biosorption instead of biodegradation, was mainly responsible for the organic substance removal from municipal wastewater. Both the Freundlich and Langmuir isotherm models matched the biosorption of organic substances by activated sludge well. These results demonstrated that the CARS system was a cost-effective and land-saving municipal wastewater treatment process. Journal of Water and Environment Technology, Vol. 7, No. 2, 2009 - 90 - ACKNOWLEDGEMENTS The authors wish to thank the National Key Project for Water Pollution Control (2008ZX07103-001 and 2008ZX07316-002), and the Anhui R&D Key Project (07010301022 and 08010302109) for the partial support of this study. REFERENCES Standard Methods for the Examination of Water and Wastewater (1995). 19th edn, American Public Health Association/American Water Works Association/Water Environment Federation, Washington DC, USA. Biggs C. A. and Lant P. A. (2000). Activated sludge flocculation: on-line determination of floc size and the effect of shear. Water Res., 34, 2542-2550. Bruus J. H., Nielsen P. H. and Keiding K. (1992). On the stability of activated sludge flocs with implications to dewatering. Water Res., 26, 1597-1604. Droppo I. G. and Ongley E. D. (1997). Flocculation of suspended sediment in rivers of southeastern Canada. Water Res., 26, 65-72. Esparza-Soto M. and Westerhoff P. (2003). Biosorption of humic and fulvic acids to live activated sludge biomass. Water Res., 37, 2301–2310. Grady Jr. C. P. L., Glen T. D. and Henry, C. L. (1999). Biological wastewater treatment: Second edition, Marcel Dekker, Inc. Huang J. C. and Li L. (2000). An innovative approach to maximize primary treatment performance. Water Sci. and Tech., 42(12), 209-222. Jacobsen B. N., Aruin E. and Reinders M. (1996). Factors affecting sorption of pentachlorophenol to suspended microbial biomass. Water Res., 30, 13-20. Kennedy K. J. and Pham T. P. (1996). Effect of anaerobic sludge source and condition on biosorption of halogenated aromatics. Water Res., 29, 13-20. McCreary J. J. and Snoeyink V. L. (1980). Characterization and activated carbon adsorption of several humic substances. Water Res., 14, 151–160. Ulrich A.H. and Smith M.W. (1951). The biosorption process of sewage and waste treatment. Sewage and Industrial Waste, 23, 1248-1253. Wilen B. M. and Jin B. (2003). The influence of key chemical constituents in activated sludge on surface and flocculation properties. Water Res., 37, 2127-2139. Wu H. S., Zhang A. Q. and Wang L. S. (2004). Immobilization study of biosorption of heavy metal ions onto activated sludge, J. of Environ. Sci., 16, 640–645. Zhou J. L. (1992). Biosorption and desorption of humic acid by microbial biomass. Chemosphere, 24, 1573-1589. Zhou J. L. and Banks C. J. (1993). Mechanism of humic acid colour removal from natural waters by fungal biomass biosorption. Chemosphere, 27, 607–620. . Vol. 7, No. 2, 2009 - 90 - ACKNOWLEDGEMENTS The authors wish to thank the National Key Project for Water Pollution Control (2008ZX 071 03-001 and 2008ZX 073 16-002),. Res., 26, 15 97- 1604. Droppo I. G. and Ongley E. D. (19 97) . Flocculation of suspended sediment in rivers of southeastern Canada. Water Res., 26, 65 -72 . Esparza-Soto