Particle size distribution and removal by a chemical-biological flocculation process

Transcription

1 Journal of Environmental Sciences 19(2007) Particle size distribution and removal by a chemical-biological flocculation process ZHANG Zhi-bin 1,2,, ZHAO Jian-fu 1, XIA Si-qing 1, LIU Chang-qing 1, KANG Xing-sheng 3 1. State Key Laboratory of Pollution Control and Resource Reuse Research, College of Environmental Science and Engineering, Tongji University, Shanghai , China 2. College of Municipal and Environmental Engineering, Shandong Jianzhu University, Jinan , China. zhazhb@163.com 3. Shandong Academy of Environmental Sciences, Jinan , China Received 15 May 2006; revised 7 August 2006; accepted 22 August 2006 Abstract The particle characterization from the influent and effluent of a chemical-biological flocculation (CBF) process was studied with a laser diffraction device. Water samples from a chemically enhanced primary treatment (CEPT) process and a primary sediment tank process were also analyzed for comparison. The results showed that CBF process was not only effective for both the big size particles and small size particles removal, but also the best particle removal process in the three processes of CBF process, CEPT process, and PST process (primary sediment tanks). The results also indicated that CBF process was superior to CEPT process in the heavy metals removal. The high and non-selective removal for heavy metals might be closely related to its strong ability to eliminate small particles. Samples from different locations in CBF reactors showed that small particles were easier to aggregate into big ones and those disrupted flocs could properly flocculate again along CBF reactor because of the biological flocculation. Key words: particle removal; chemical-biological flocculation (CBF); chemically enhanced primary treatment (CEPT) Introduction It was estimated that particle organic matter and phosphorous accounted for more than 70% of the total organic matter and phosphorous in municipal wastewater (Levine et al., 1985; Tiehm et al., 1999). Particles (<10 µm) have the largest specific surface among all particles and play a key role in the dynamics and equilibrium of the physico-chemical surface reactions. Trace elements, such as heavy metals, can be adsorbed on the surface of these small particles (Daniel and Markus, 1997; Kavanaugh et al., 1978). According to a study carried out by Landa and Jimenez (1997), the number of helminth eggs was in close correlation with the suspended solids number. Therefore, the particle removal, especially small particles, is essential and great interest for the municipal wastewater treatment plants. Particle removal in primary sediment tanks (PST) is achieved primarily through gravity in traditional secondary wastewater treatment plants. PST usually has good removal efficiency for the particles (>50 µm), whereas more than 85% of the particles in effluent were smaller than 32 µm (Neis and Tiehm, 1997). Compared with traditional PST, the chemically enhanced primary treatment (CEPT) and the chemical-biological flocculation Project supported by the Hi-Tech Research and Development Program (863) of China (No. 2002AA601320); the Shandong Environment Protection Bureau Program (No , ) and the Ph.D Fund of Shandong Jianzhu University (No , ). *Corresponding author. zhazhb@163.com. (CBF) can improve small particle removal greatly (Blanca et al., 2000; Hallvard, 1998; Poon and Chu, 1999). The basic meaning of CEPT is to use chemical coagulants to induce coagulation or flocculation and allow the finelydivided particles to form large aggregates so that they can be easily separated from water by gravity. Through the chemical addition, CEPT process achieves good treatment efficiency not only in removal of particles, but also in removal of organics and phosphorous. Blanca and his colleagues found that the particle removal efficiency was 82% for the particles in the range of µm at the dosage of 72 mg/l FeSO 4 and 1.2 mg/l P-17, whereas the process was less effective in eliminating particles smaller than 10 µm (Blanca et al., 2000). CEPT process could reduce more than 80% suspended solid (SS) and 70% total phosphorous after the addition of 30 mg/l FeCl 3 and 0.5 mg/l polymer (Poon and Chu, 1999). In Europe, CEPT process is used in some wastewater treatment plants as a useful way to further decrease the level of heavy metals discharged into water body (Charles and Bernard, 2005). CBF process is a new advanced primary treatment process in which particles are precipitated through chemical coagulation and biological flocculation. Compared with CEPT, CBF can improve the removal of small particles because of biological flocculation. Previous studies indicated that CBF was superior to CEPT in the removal of SS (Wu et al., 2003). Since SS only represents the particles >10 µm and smaller particles are a substantial part of solids, it is

2 560 ZHANG Zhi-bin et al. Vol. 19 important to study the removal efficiency of particles <10 µm in CBF process. This paper aimed to study the characters of particles between µm, especially with respect to small particles in the influent and effluent of the CBF process. Samples from a CEPT process and PST process of Anting Wastewater Treatment Plant were also studied for comparison. 1 Materials and methods 1.1 Municipal wastewater Raw municipal wastewater used in this experiment was collected continuously from the Anting Wastewater Treatment Plant, Shanghai, China. After the screens and grit chamber, the wastewater was pumped to the adjusting tank of the pilot-scale treatment process. 1.2 Experimental set-up Pilot-scale experimental set-up The pilot-scale experimental set-up is shown in Fig.1. Fig. 1 Pilot-scale experimental set-up. The CBF process consisted of fast air-stirring mixing tank, CBF reactor, sediment tank, and sludge return device. The mixing tank had an effective volume of 30 m 3, an effective height of 0.55 m, and its hydraulic retention time (HRT) was 60 s. The CBF reactor consisted of three plug-flow channels with declining aeration rate, effective volume of 1.2 m 3, effective height of 0.55 m, and HRT of 35 min. Aeration was supplied to the channels through microhole aeration pipes and the aeration rate could be adjusted separately in each channel. The sediment tank had an effective height of 1.2 m, and HRT of 1.5 h. Sludge was returned to the CBF reactor with pump (NEMO). The CEPT process consisted of mixing pipe, CEPT reactor, and sediment tank. CEPT reactor included three serially-connected chambers and had the same volumes as CBF. Wastewater in the chambers was completely mixed through mechanical force. The capacity of the pilot-scale plants was 100 m 3 /d. The sludge return ratio of CBF was 33% and dissolved oxygen (DO) in three chambers was controlled at mg/l, mg/l, and mg/l, respectively. The liquid polymeric flocculant PAFC (Al 2 O 3 : 10.8%, Fe 2 O 3 : 1.8%, Guoxin, Jiashan, China) and the aiding flocculant PAM (Shanghai Chemical Reagent Corporation, China) were selected after a series of jar tests which were carried out to determine which flocculant was the most suitable in this study. PAFC was added at the CBF mixing tank and the CEPT mixing pipe respectively and the dosage depended on the influent TP concentration. PAM (dosage 0.5 mg/l) was added at the third channels of both CBF and CEPT reactors. The influent temperature was C, and ph was during this experiment Flow chart of Anting Wastewater Treatment Plant The flow chart of Anting Wastewater Treatment Plant is shown in Fig.2. Fig. 2 Flow chart of Anting Wastewater Treatment Plant. The primary sediment tank was of typical radial shape and had an internal diameter of 30 m, and an effective height of 6.5 m. Anting Wastewater Treatment Plant had an average flow of m 3 /d, and the hydraulic retention time of its PST was h. PST effluent was sampled at its overflow weir. 1.3 Removal of COD Cr, SS, NH 4 + -N, TP, and heavy metals The influent, CBF effluent, CEPT effluent and PST effluent were sampled daily for 30 d, and COD Cr, SS, NH 4 + -N, and TP were measured with their respective standard methods (Wei et al., 2002). Influent, CBF effluent and CEPT effluent were sampled at exactly the same point of time (April 15, 2005), of which heavy metals were measured with an IRIS Advantage 1000 Inductively Coupled Plasma Spectrometer (Thermo Jarrell Ash Company). 1.4 Particle size distribution Particle size distribution of the influent, CBF effluent, CEPT effluent and PST effluent was analyzed once a week for a month and the results were comparable to each other. In this article, only one result was presented. Particle size distribution was determined by a Coulter Ls230 which measures particle sizes from 40 nm to 2000 µm by laser diffraction. Three samples (1 L) were taken from each chamber along the CBF reactor and CEPT reactor, and then these samples were allowed to precipitate for 30 min in beakers of 1 L. Top 300 ml supernatant was taken from each sample to analyze its particle size distribution to study the evolution of flocs along the CBF and CEPT process, respectively. 2 Results and discussion 2.1 Removal of COD Cr, SS, NH 4 + -N, and TP The monthly average concentrations of COD Cr, SS, NH 4 + -N, and TP in different samples are presented in

3 No. 5 Particle size distribution and removal by a chemical-biological flocculation process 561 Table 1. It is shown in Table 1 that the effluent of advanced primary treatment processes (CBF and CEPT) were superior to that of PST, and the best removal efficiencies for COD Cr, SS, NH 4 + -N, and TP was obtained in CBF process. 2.2 Particle size distribution in influent, effluent of CBF, CEPT and PST processes The typical size distributions of particles are shown in Fig.3. Fig. 3 Particle size distributions in influent, CBF effluent, CEPT effluent and PST effluent. The meanings of CBF, CEPT and PST are the same as in Table 1. The particles had the total volume of µm 3 in the influent and were distributed in µm with the mean size at µm. A portion of large particles were removed in PST, which resulted in the decrease of the volume of the particles to µm 3 in PST effluent. PST was less effective in eliminating small particles, particles in its effluent were mainly distributed in µm with the mean size at µm, making no much difference to the influent. The particle volume in CEPT effluent decreased to µm 3 due to the chemical flocculation, and particles >110 µm were completely removed through CEPT. While, particles >3 µm were almost removed completely in CBF effluent, and most of particles <3 µm were also removed, resulting in the total particle volume of µm 3 in CBF effluent. Particles were distributed in µm with the mean size at µm in CBF effluent, showing the highest removal efficiency for both large particles and small particles due to the combination of chemical flocculation and biological flocculation. Stan and Despa (2000) also found that the combination of single particle was the rate-limiting step of flocculation, and the presence of surface proteins over microorganisms promoted the aggregation of small particles. 2.3 Removal of heavy metals in CBF and CEPT The removal results for heavy metals in CEPT and CBF are shown in Table 2. More than 90% of Cr was eliminated whereas the removal rate of Cu, Sn, Ag, Pb, Zn, and Ba was only 29.9%, 43.6%, 78.4%, 30.2%, 57.0%, and 45.5%, respectively in CEPT, suggesting that its removal capacity of heavy metals was selective. CBF appeared to be more effective and non-selective in heavy metals removal compared with CEPT, and the removal rates for Cr, Cu, Cd, Ag, Pb, Zn and Al were all above 75%. The differences in the removal of heavy metals between CBF and CEPT were probably due to their varying capacities in removing small particles. The total surface of particles <10 µm in municipal wastewater had a significant effect on the speed and equilibrium of physico-chemical surface reactions. Trace elements, such as heavy metals, may be adsorbed on the surface of these small particles. CEPT was less effective in small particle removal and its ability to remove heavy metals depended on the chemical coprecipitation. Special coprecipitant was usually selective in nature; therefore CEPT exhibited different removal rates for different heavy metals. CBF achieved high removal rate in small particles, thus heavy metals adsorbed on these particles were eliminated significantly from wastewater, resulting in high heavy metal removal rates with non-selective character. This was confirmed by the research of Kavanaugh et al. (1978), which found that the removal of heavy metals was closely correlated with the removal of small particles in wastewater treatment processes. Table 1 Removal of COD Cr, SS, NH 4 + -N, and TP COD Cr (mg/l) NH 4 + -N (mg/l) TP (mg/l) SS (mg/l) Influent 171.4± ± ± ±121 CBF 55.6± ± ± ±15 CEPT 85.2± ± ± ±20 PST 120.2± ± ± ±66 CBF: Chemical-biological flocculation process; CEPT: chemically enhanced primary treatment process; PST: primary sediment tanks. Table 2 Removal of heavy metals in CBF and CEPT Cr (mg/l) Cu (mg/l) Sn (mg/l) Ag (mg/l) Pb (mg/l) Zn (mg/l) Ba (mg/l) Influent CEPT η CEPT 92.0% 29.9% 43.6% 78.4% 30.2% 57.0% 45.5% CBF η CBF 99.4% 94.2% 84.0% 97.8% 88.5% 95.0% 75.1% η: Removal rate; CBF: chemical-biological flocculation process; CEPT: chemically enhanced primary treatment process.

4 562 ZHANG Zhi-bin et al. Vol Particle size distribution and evolution of flocs along the CEPT process The particle size distribution of samples taken along the CEPT reactor are shown in Fig.4. Fig. 4 Particle size distribution of supernatant sampled along CEPT reactor. Samples 1, 2, and 3 are taken respectively from each chamber along the CEPT reactor. The volume of particles in the supernatant of Sample 1 was µm 3, and these particles were distributed in µm with the mean size at µm, indicating that flocs formed in the first chamber were mainly of small size. In the second chamber, some big flocs which were not present in supernatant were observed. The particle volume of Sample 2 decreased to µm 3 and particles were distributed in µm with the average size at µm. After the PAM addition, more small particles grew into big flocs in the third chamber. The volume of Sample 3 decreased to µm 3 and particles were distributed in µm with the mean size at µm. Along the CEPT reactor, the particle size distributions curve gradually shifted towards small particles, suggesting that the evolution of flocs was slow and gradual. 2.5 Particle size distribution and evolution of flocs along CBF process The particle size distributions of samples taken along the CBF reactor are shown in Fig.5. Fig. 5 Particle size distribution of supernatant sampled along CBF reactor. Samples 1 and 4 were taken respectively from the beginning of the first and the third channel; Samples 2, 3 and 5 were from the end of each channel. The concentration of particles in CBF reactor was high due to sludge recycle, so these particles collided more frequently with each other than in CEPT. On the other hand, biological flocculation promoted small particles to be adsorbed on active large flocs. The volume of particles in Samples 1, 2, 3, 4, and 5 was µm 3, µm 3, µm 3, µm 3, and µm 3, and particles were distributed in µm, µm, µm, µm, and µm respectively. It was inferred from Fig.5 that along the CBF reactor, although flocs were disrupted for many times, they could properly reaggraded. These phenomena could not be explained appropriately by traditional chemical flocculation theory which holds that disrupted flocs hardly agglomerate again, so typical chemical flocculation plants always consisted of serially connected tanks mixed by mechanical force with deckling rate. Reaggration took place in CBF because biological flocculation helped to bridge disrupted flocs together. It is in agreement with the research by Wilen and Kristian (2004), which found that aerobic biological activity was of the greatest significance for the observed reflocculation of broken flocs. Previous research on microbial community in CBF also showed that there was stable and special biological community after 30 d of running duration (Xia et al., 2005). Although at the end of the CBF reactor, the volume of particles in Sample 5 increased to µm 3 due to flocs disruption, there was certain height of sludge bed in sediment tank which could catch small particles due to the biological adsorption. Therefore CBF achieved good particle removal rate, and the volume of particles in CBF effluent was only µm 3 in Fig.3. 3 Conclusions Particles in the influent, CBF effluent, CEPT effluent, and PST effluent were distributed in µm, µm, µm, and µm, respectively. The best particle removal efficiency was achieved in CBF and little small particles was left in CBF effluent compared with those in CEPT effluent and PST effluent, which indicated that CBF were effective in both big size and small size particles removals. The removal efficiencies in CEPT process for Cr, Cu, Sn, Ag, Pb, Zn, and Ba were 92.0%, 29.9%, 43.6%, 78.4%, 30.2%, 57.0%, and 45.5% respectively. While, those in CBF for Cr, Cu, Sn, Ag, Pb, Zn, and Ba were 99.4%, 94.2%, 84.0%, 97.8%, 88.5%, 95.0%, and 75.1% respectively. CBF appeared to be superior to CEPT in the removal of heavy metals, and the capability of high and non-selective removal of heavy metals might be closely related to small particle removal. Samples from different locations in flocculation reactors showed that small particles were easier to aggraded into bigger ones and those disrupted flocs could properly flocculate again in CBF because of the biological flocculation.

5 No. 5 Particle size distribution and removal by a chemical-biological flocculation process 563 References Blanca J, Alma C, Alberto L et al., Sand and synthetic medium filtration of advanced primary treatment effluent from Mexico City[J]. Wat Res, 34(2): Charles G, Bernard R, Contribution of municipal effluents to metal fluxes in the ST. Lawrence River[J]. Environ Sci Technol, 39: Daniel K, Markus B, Particle removal in different filteration systems for tertiary wastewater treatment a comparison[j]. Wat Sci Tech, 36(4): Hallvard D, Optimized particle separation in the primary step of wastewater treatment[j]. Wat Sci Tech, 37(10): Kavanaugh M, Toregas G, Moon C, Particulates and trace pollutant removal by depth filtration[j]. Prog Wat Tech, 10(5/6): Landa H, Jimenez B, Particle size distribution in an effluent from an advanced primary treatment and removal during filtration[j]. Wat Sci Tech, 36(4): Levine A D, Tchobanoglous G, Asano T, Characterization of the size distribution of contaminants in wastewater treatment and reuse implications[j]. J Wat Pollut Control Fed, 57: Neis U, Tiehm A, Particle size analysis in primary and secondary wastewater effluents[j]. Wat Sci Tech, 36(4): Poon C S, Chu C W, The use of ferric chloride and anionic polymer in the chemically assisted primary sedimentation process[j]. Chemosphere, 39(10): Stan S, Despa F, On the doublet formation in the flocculation process of the yeast cells[j]. Biosystems, 57: Tiehm A, Herwig V, Neis U, Particle size analysis for improved sedimentation and filtration in waste water treatment[j]. Wat Sci Tech, 39(8): Wei F S, Xu X B, Yan J C et al., Water and wastewater analysis manual[m]. 3th ed. Beijing: China Environment Science Publisher. Wilen B M, Kristian K, Flocculation of activated sludge flocs by stimulation of the aerobic biological activity[j]. Wat Res, 38: Wu Z P, Xia S Q, Yang D H et al., Comparative research of chemical and biological flocculation process to municipal wastewater[j]. Chongqing Environmental Sciences, 25(7): Xia S Q, Wang F, Fu Y G et al., Biodiversity analysis of microbial community in the chem-bioflocculation treatment process[j]. Biotechnology Bioengineering, 89(6):