- Email: [email protected]
Performance change during long-term ozonation aimed at augmenting denitrification and decreasing waste activated sludge
Chemosphere 73 (2008) 1529–1532
Contents lists available at ScienceDirect
Chemosphere journal homepage: www.elsevier.com/locate/chemosphere
Technical Note
Performance change during long-term ozonation aimed at augmenting denitrification and decreasing waste activated sludge Magdalena A. Dytczak, Jan A. Oleszkiewicz * Department of Civil Engineering, University of Manitoba, 15 Gillson Street, Winnipeg, MB, Canada R3T 5V6
a r t i c l e
i n f o
Article history: Received 23 May 2008 Received in revised form 27 August 2008 Accepted 30 August 2008 Available online 10 October 2008 Keywords: Sludge minimization Wastewater Activated sludge Ozone Floc structure
a b s t r a c t Partial ozonation of return activated sludge for waste sludge minimization and soluble COD production was examined. Two nitrifying sequencing batch reactors, one control and one ozonated, were operated under alternating anoxic/aerobic conditions. During the first steady-state period of 95–136 d of ozonation, the amount of wasted solids decreased with the ozone dose up to 25%, generating soluble COD by cell lysis. However, during a subsequent period of 190–232 d of continuous ozonation, the effect of solids destruction and COD production decreased by 50%. The investigations of extracellular polymers content and floc shape analyses showed that, after prolonged daily ozone treatment, sludge floc structure becomes stronger, denser, and more ozone-resistant. The findings suggest that, for prolonged operation of partial sludge ozonation, an increase in ozone doses may be required to continuously maintain the expected solids destruction level. This in turn will increase the operational costs of the treatment. Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction Ozonation of a fraction of the return activated sludge (RAS) followed by recycling of the ozonated portion to the main tank is an attractive technology for sludge production minimization (Yasui and Shibata 1994; Sakai et al., 1997). During the process with applied pre-anoxic zone, additional carbon generated from solubilization of particulate chemical oxygen demand (pCOD) helps improve denitrification (Ahn et al., 2002; Park et al., 2004). In addition, sludge settleability is improved and problems with bulking are reduced due to destruction of the filament network (Weemaes et al., 2000). Ozone production is the key expenditure in this process – optimization is required to achieve the most economical performance. With an increased applied ozone dose, the reduction of solids and the production of soluble chemical oxygen demand sCOD (released from damaged cells) increase. Further dose increase does translate into a further drop in solids concentration but not necessarily to the increase of sCOD as the processes of mineralization will dominate over the solubilization. The key to the method is to establish the optimum ozone dose, where sludge production is satisfactorily reduced, while obtaining the benefits of COD production and denitrification improvement. The issue of nitrification must also be taken into consideration when ozone dose is selected, as the nitrification rate deteriorates with increasing ozone application
* Corresponding author. Tel.: +1 204 474 8722; fax: +1 204 474 7513. E-mail address: [email protected] (J.A. Oleszkiewicz). 0045-6535/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2008.08.039
(Boehler and Siegrist, 2004). Those problems have been addressed in the preceding study (Dytczak et al., 2007). After establishing the optimum ozone dose, one needs also to address long-term process stability. It seems probable that continuous long-term ozone application will influence the evolution of activated sludge properties. The objective of this work was to evaluate solids destruction and sCOD production in two phases of the 276 d experiment to check if there are any changes in the ozonation method’s benefits. Extracellular polymers content and activated sludge floc structure were studied during the long-term ozonation and in reference reactors to elucidate the mechanism of ozone impact on floc shape and process performance. 2. Materials and methods 2.1. Experimental setup Two identical sequencing batch reactors were operated in parallel under alternating anoxic/aerobic conditions as described previously (Dytczak et al., 2007), with a total liquid volume of 3 L each, with a solids residence time (SRT) of 12 d, a hydraulic residence time (HRT) of 36 h, and mixed liquor suspended solids (MLSS) of 1800 mg L1 at 20 ± 1 °C. At the start of each daily cycle, they were fed a synthetic wastewater consisting of beef and yeast extract as a carbon source (resultant COD 740 mg L1), ammonium chloride as 1 and ammonia source (resultant total available 62 mg NHþ 4 —N L 1 10 mL of 40 g L NaNO3 solution at the beginning of the anoxic reaction phase to create a sufficient supply of NO 3 for denitrification. After feeding the biomass was stirred and submitted for
1530
M.A. Dytczak, J.A. Oleszkiewicz / Chemosphere 73 (2008) 1529–1532
alternating anoxic/aerobic reaction for 22 h 30 min. Wasting of excess biomass was performed by removing 250 mL of reactor content before the end of the reaction phase. The reactor biomass was then allowed to settle for approximately 1 h which was followed by decanting of 2 L of supernatant (equivalent to effluent). This left 1 L of sludge in the reactor (equivalent to RAS and called RAS in this study). A new cycle was then started with a new feeding. In a pair, the one reactor served as a control and the other went through the ozone treatment. For the ozonated reactor, 20% of the RAS was withdrawn, treated by ozone, and then returned to the activated sludge reactor just before feeding. In the ozonated reactor, the wasting was adjusted daily to maintain TSS at around 1800 mg L1 at the beginning of the reaction period, similar to the control. Ozone was generated and applied as described previously (Dytczak et al., 2007). During the experiments, doses of 7–33 mg O3 consumed per reactor and per day were used, corresponding to doses of 0.015–0.080 mg O3 mg1 TSS of initial excess sludge. Doses were applied from the smallest to the largest, increasing in increments of 0.005 mg O3 mg1 TSS of initial excess sludge. They were applied in two phases: increasingly during 1–140 d of the experiment, and, starting at 190 d until 235 d, again smaller doses to the largest. Even during the periods without sampling, daily applications of the moderate doses of ozone were continued for the duration of the whole experiment, except during the period of 45–77 d.
washed EPS, and bound – washed EPS were analyzed using methods described previously (Dytczak et al., 2006). The TSS destruction and COD production were evaluated in two periods, in full ozone doses coverage, between 95–136 d and 190– 232 d. EPS were measured daily in both the ozonated and control reactors during 239–276 d. Floc structure was analysed during 256–269 d. The map of sampling is shown in Fig. 1. 3. Results 3.1. Decrease of excess sludge produced The RAS that was ozonated experienced a large decrease in biomass proportional to the ozone dose, as reported previously (Dytczak et al., 2007). This in turn consistently reduced the amount of excess solids in the reactor receiving the ozonated RAS compared to the control (up to 25% for the highest doses). After prolonged ozonation, the decrease of solids for the ozonated alternating reactor showed a tendency to decline (Fig. 2a). The destruction in solids in the alternating reactor achieved no more than 15% for the highest ozone doses in the second phase of the experiment. 3.2. Production of additional sCOD Ozonation generated additional soluble organic matter. The increase of sCOD content in the ozonated portion of the RAS is plot-
a
Total and volatile suspended solids (TSS and VSS) were measured in RAS from the ozonated reactor directly before and after ozonation. sCOD was measured in quintuplicate in the portion of ozonated RAS directly before and after ozonation as well as in RAS from control reactors. All analyses were performed according to American Public Health Association (1998). Samples of mixed liquor were observed regularly using Leitz Wetzler Germany microscope (magnification of objective 40 and 100 phase contrast). To capture the pictures and floc structure analysis, Nikon Microscope Eclipse E400 with camera Olympus DP70 and Image ProÒPlus software were used. Samples of RAS in the control and ozonated reactor (before and after ozone treatment) were analyzed over a period of one month, to estimate the direct impact of ozonation on extracellular polymer substances (EPS) as well as to track the long-term influence of ozonation of sludge. Both total (bound and soluble) EPS, called un-
30% y = 1.94x + 0.06 2
25%
R = 0.71 phase 1
Decrease in TSS
2.2. Analysis
20%
phase 2
15%
10%
5%
y = 0.66x + 0.07 2
R = 0.13
0%
b
500% 450% 400%
continuous ozonation (d): 50
100
y = 35.76x + 0.25
150
200
250
TSS and COD data collection:
no ozone
phase 1 (41 d)
phase 2 (42 d)
EPS datacollection:
Increase in sCOD
0
350%
2
R = 0.53
300% 250% 200% 150% 100%
no ozone
(37 d)
Image Pro datacollection:
y = 14.17x + 0.35
50% 0% 0.00
2
R = 0.38
0.02
0.04
0.06
0.08
0.10
-1
no ozone Fig. 1. The schedule of sampling during 276 d of ozonation.
(13 d)
ozone dose (mg O 3 mg TSS initial excess sludge) Fig. 2. Decrease in WAS (a) and increase in soluble COD in ozonated RAS (b) in the reactors after ozone treatment; 1st phase 95–136 d, 2nd phase 190–232 d.
M.A. Dytczak, J.A. Oleszkiewicz / Chemosphere 73 (2008) 1529–1532
ted in Fig. 2b. Production of sCOD correlated with TSS destruction for both phases of the sampling. Consequently, the production of sCOD during ozonation in the alternating reactor decreased for the second phase of the tests (up to 170% only) in comparison to the first one (up to 400% for high ozone doses).
1531
equal in shape. In the control reactor which did not receive ozone, the flocs’ size was more variable, the occurrence of bigger flocs was observed, and flocs tended to have a greater perimeter length in comparison to those from the ozonated reactor. 4. Discussion
3.3. EPS content The alternating control SBR had consistent total EPS (378 ± 48 mg g1 VSS, n = 37) and bound EPS (223 ± 53 mg g1 VSS, n = 37) (Dytczak et al., 2006). In the ozonated reactor after long-term ozonation, both total and bound EPS levels increased significantly and were 488 ± 49 mg g1 VSS and 307 ± 41 mg g1 VSS, respectively (n = 37). Additionally, the ratio of bound to total polymers was higher in the ozonated reactor (62%) in comparison to the control (59%), showing some tendency of strengthening the floc structure. 3.4. Biomass structure Microscopic observation of sludge performed daily from the beginning of the research showed that biomass in the alternating reactors consisted mostly of bacteria with few filaments, inhabiting weak, thin, and open, more dispersed and elongated flocs (Dytczak et al., 2006, 2007). The presence of the anoxic zones in alternating reactors is responsible for partial, non-reversible deflocculation probably due to decreased microbial activity under anoxic conditions. To quantify the difference in floc structure observed microscopically, the floc area and perimeter length were compared. The histograms in Fig. 3 show the frequency of these parameters for control and ozonated alternating reactor. After long-term ozonation (data collected within 256–269 d of ozonation) alternating flocs became smaller, compact, round and more
a
15%
Frequency
control
10%
ozonated
A large deterioration in the effectiveness of solids destruction and sCOD production was observed in the second phase of ozone treatment. It appeared that prolonged ozonation strengthened the structure of flocs and made them more resistant to destruction by further ozone treatment. This phenomenon seems to be caused by a significant amount of additional sCOD delivered to the ozonated system, which in turn stimulates the EPS production. The increase of the fraction of bound EPS in the ozonated reactor shows that combined ozonation-biological treatment favours not only the quantity of EPS production, but also the degree of their binding into the floc structure. We have attributed this phenomenon (Dytczak et al., 2006) to an increase in the food-to-microorganisms (F/M) ratio. Both reactors received the same amount of synthetic wastewater but, because of the destruction of part of the solids and a corresponding additional sCOD release from cells, the F/M ratio is higher in the ozonated reactor which promotes microbial activity and increases the EPS yield. Excess, unutilized carbon in the system with high F/M could have been converted to intracellular storage granules and extracellular polymers that accumulate as EPS (Liao et al., 2001). This might explain why the amounts of both the total and bound EPS levels in ozonated reactors after prolonged ozonation (measured since 239 d of the ozone treatment) are higher than in the control reactor, which represented the initial conditions. Extracellular polymers are essential for floc formation; their increased amount helps in flocculation and overcomes the effect of disintegration by ozone. In this research, the stronger flocs with higher content of bound EPS turned out to be more ozone-resistant. Consequently, the effect of reduction of WAS solids was less pronounced in the reactors subjected to prolonged ozone treatment. Additionally, the benefits regarding the soluble carbon production for denitrification could be diminished in time. On the other hand, smaller and equal in size flocs would improve sludge settling properties, although dewaterability may deteriorate (Yen et al., 2002).
5%
5. Conclusions 0% 0
300
600
900
1200
1500
Floc area (mm2)
b
50%
Frequency
40%
control ozonated
30% 20% 10% 0% 0
100
200
300
400
Floc perimeter (μm) Fig. 3. The frequency of floc area (a) and perimeter (b) for flocs for control and ozonated reactor (samples from the RAS, 256–269 d).
The issue of progressive strengthening of continuously ozonated flocs may have large engineering significance for practical implementation of the method, and, to our knowledge, has not previously been reported. This research indicated that, for the prolonged operation of partial sludge ozonation, an increase in ozone doses may be required to continuously maintain the solids destruction and related denitrification improvement effect. The necessity of application of higher ozone doses will increase the operational costs of the treatment related to ozone generation. This negative consequence of process adaptation could of course be different in a full scale facility, where variations in influent composition and temperature cause activated sludge flocs to be more vulnerable to deflocculation. The finding should therefore be verified on a larger scale. Acknowledgements The research was sponsored by Natural Sciences and Engineering Research Council of Canada; Province of Manitoba; City of Winnipeg Water and Waste Department and EarthTech (Canada) Ltd.
1532
M.A. Dytczak, J.A. Oleszkiewicz / Chemosphere 73 (2008) 1529–1532
The excellent technical assistance with floc structure analysis of Jennifer Kroeker, Department of Microbiology, University of Manitoba, is greatly appreciated. References Ahn, K.-H., Yeom, I.T., Park, K.Y., Maeng, S.K., Lee, Y., Song, K.-G., Hwang, J.H., 2002. Reduction of sludge by ozone treatment and production of carbon source for denitrification. Water Sci. Technol. 46 (10), 121–125. American Public Health Association, 1998. Standard Methods for the Examination of Water and Wastewater, 20th ed. Washington, DC, USA. Boehler, M., Siegrist, H., 2004. Partial ozonation of activated sludge to reduce excess sludge, improve denitrification and control scumming and bulking. Water Sci. Technol. 49 (10), 41–49. Dytczak, M.A., Londry, K.L., Siegrist, H., Oleszkiewicz, J.A., 2006. Extracellular polymers in partly ozonated return activated sludge: impact on flocculation and dewaterability. Water Sci Technol. 54 (9), 155–164.
Dytczak, M.A., Londry, K.L., Siegrist, H., Oleszkiewicz, J.A., 2007. Ozonation reduces sludge production and improves denitrification. Water Res. 41, 543–550. Liao, B.Q., Allen, D.G., Droppo, I.G., Leppard, G.G., Liss, S.N., 2001. Surface properties of sludge and their role in bioflocculation and settleability. Water Res. 35, 339– 350. Park, K.Y., Lee, J.W., Ahn, K.H., Maeng, S.K., Hwang, J.H., Song, K.G., 2004. Ozone disintegration of excess biomass and application to nitrogen removal. Water Environ. Res. 76, 162–167. Sakai, Y., Fukase, T., Yasui, H., Shibata, M., 1997. An activated sludge process without excess sludge production. Water Sci. Technol. 36 (11), 163–170. Weemaes, M., Grootaerd, H., Simoens, F., Verstraete, W., 2000. Anaerobic digestion of ozonized biosolids. Water Res. 34, 2330–2336. Yasui, H., Shibata, M., 1994. An innovative approach to reduce excess sludge production in the activated sludge processes. Water Sci. Technol. 30 (9), 11–20. Yen, P.-S., Chen, L.C., Chien, C.Y., Wu, R.M., Lee, D.J., 2002. Network strength and dewaterability of flocculated activated sludge. Water Res. 36, 539–550.
Contents lists available at ScienceDirect
Chemosphere journal homepage: www.elsevier.com/locate/chemosphere
Technical Note
Performance change during long-term ozonation aimed at augmenting denitrification and decreasing waste activated sludge Magdalena A. Dytczak, Jan A. Oleszkiewicz * Department of Civil Engineering, University of Manitoba, 15 Gillson Street, Winnipeg, MB, Canada R3T 5V6
a r t i c l e
i n f o
Article history: Received 23 May 2008 Received in revised form 27 August 2008 Accepted 30 August 2008 Available online 10 October 2008 Keywords: Sludge minimization Wastewater Activated sludge Ozone Floc structure
a b s t r a c t Partial ozonation of return activated sludge for waste sludge minimization and soluble COD production was examined. Two nitrifying sequencing batch reactors, one control and one ozonated, were operated under alternating anoxic/aerobic conditions. During the first steady-state period of 95–136 d of ozonation, the amount of wasted solids decreased with the ozone dose up to 25%, generating soluble COD by cell lysis. However, during a subsequent period of 190–232 d of continuous ozonation, the effect of solids destruction and COD production decreased by 50%. The investigations of extracellular polymers content and floc shape analyses showed that, after prolonged daily ozone treatment, sludge floc structure becomes stronger, denser, and more ozone-resistant. The findings suggest that, for prolonged operation of partial sludge ozonation, an increase in ozone doses may be required to continuously maintain the expected solids destruction level. This in turn will increase the operational costs of the treatment. Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction Ozonation of a fraction of the return activated sludge (RAS) followed by recycling of the ozonated portion to the main tank is an attractive technology for sludge production minimization (Yasui and Shibata 1994; Sakai et al., 1997). During the process with applied pre-anoxic zone, additional carbon generated from solubilization of particulate chemical oxygen demand (pCOD) helps improve denitrification (Ahn et al., 2002; Park et al., 2004). In addition, sludge settleability is improved and problems with bulking are reduced due to destruction of the filament network (Weemaes et al., 2000). Ozone production is the key expenditure in this process – optimization is required to achieve the most economical performance. With an increased applied ozone dose, the reduction of solids and the production of soluble chemical oxygen demand sCOD (released from damaged cells) increase. Further dose increase does translate into a further drop in solids concentration but not necessarily to the increase of sCOD as the processes of mineralization will dominate over the solubilization. The key to the method is to establish the optimum ozone dose, where sludge production is satisfactorily reduced, while obtaining the benefits of COD production and denitrification improvement. The issue of nitrification must also be taken into consideration when ozone dose is selected, as the nitrification rate deteriorates with increasing ozone application
* Corresponding author. Tel.: +1 204 474 8722; fax: +1 204 474 7513. E-mail address: [email protected] (J.A. Oleszkiewicz). 0045-6535/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2008.08.039
(Boehler and Siegrist, 2004). Those problems have been addressed in the preceding study (Dytczak et al., 2007). After establishing the optimum ozone dose, one needs also to address long-term process stability. It seems probable that continuous long-term ozone application will influence the evolution of activated sludge properties. The objective of this work was to evaluate solids destruction and sCOD production in two phases of the 276 d experiment to check if there are any changes in the ozonation method’s benefits. Extracellular polymers content and activated sludge floc structure were studied during the long-term ozonation and in reference reactors to elucidate the mechanism of ozone impact on floc shape and process performance. 2. Materials and methods 2.1. Experimental setup Two identical sequencing batch reactors were operated in parallel under alternating anoxic/aerobic conditions as described previously (Dytczak et al., 2007), with a total liquid volume of 3 L each, with a solids residence time (SRT) of 12 d, a hydraulic residence time (HRT) of 36 h, and mixed liquor suspended solids (MLSS) of 1800 mg L1 at 20 ± 1 °C. At the start of each daily cycle, they were fed a synthetic wastewater consisting of beef and yeast extract as a carbon source (resultant COD 740 mg L1), ammonium chloride as 1 and ammonia source (resultant total available 62 mg NHþ 4 —N L 1 10 mL of 40 g L NaNO3 solution at the beginning of the anoxic reaction phase to create a sufficient supply of NO 3 for denitrification. After feeding the biomass was stirred and submitted for
1530
M.A. Dytczak, J.A. Oleszkiewicz / Chemosphere 73 (2008) 1529–1532
alternating anoxic/aerobic reaction for 22 h 30 min. Wasting of excess biomass was performed by removing 250 mL of reactor content before the end of the reaction phase. The reactor biomass was then allowed to settle for approximately 1 h which was followed by decanting of 2 L of supernatant (equivalent to effluent). This left 1 L of sludge in the reactor (equivalent to RAS and called RAS in this study). A new cycle was then started with a new feeding. In a pair, the one reactor served as a control and the other went through the ozone treatment. For the ozonated reactor, 20% of the RAS was withdrawn, treated by ozone, and then returned to the activated sludge reactor just before feeding. In the ozonated reactor, the wasting was adjusted daily to maintain TSS at around 1800 mg L1 at the beginning of the reaction period, similar to the control. Ozone was generated and applied as described previously (Dytczak et al., 2007). During the experiments, doses of 7–33 mg O3 consumed per reactor and per day were used, corresponding to doses of 0.015–0.080 mg O3 mg1 TSS of initial excess sludge. Doses were applied from the smallest to the largest, increasing in increments of 0.005 mg O3 mg1 TSS of initial excess sludge. They were applied in two phases: increasingly during 1–140 d of the experiment, and, starting at 190 d until 235 d, again smaller doses to the largest. Even during the periods without sampling, daily applications of the moderate doses of ozone were continued for the duration of the whole experiment, except during the period of 45–77 d.
washed EPS, and bound – washed EPS were analyzed using methods described previously (Dytczak et al., 2006). The TSS destruction and COD production were evaluated in two periods, in full ozone doses coverage, between 95–136 d and 190– 232 d. EPS were measured daily in both the ozonated and control reactors during 239–276 d. Floc structure was analysed during 256–269 d. The map of sampling is shown in Fig. 1. 3. Results 3.1. Decrease of excess sludge produced The RAS that was ozonated experienced a large decrease in biomass proportional to the ozone dose, as reported previously (Dytczak et al., 2007). This in turn consistently reduced the amount of excess solids in the reactor receiving the ozonated RAS compared to the control (up to 25% for the highest doses). After prolonged ozonation, the decrease of solids for the ozonated alternating reactor showed a tendency to decline (Fig. 2a). The destruction in solids in the alternating reactor achieved no more than 15% for the highest ozone doses in the second phase of the experiment. 3.2. Production of additional sCOD Ozonation generated additional soluble organic matter. The increase of sCOD content in the ozonated portion of the RAS is plot-
a
Total and volatile suspended solids (TSS and VSS) were measured in RAS from the ozonated reactor directly before and after ozonation. sCOD was measured in quintuplicate in the portion of ozonated RAS directly before and after ozonation as well as in RAS from control reactors. All analyses were performed according to American Public Health Association (1998). Samples of mixed liquor were observed regularly using Leitz Wetzler Germany microscope (magnification of objective 40 and 100 phase contrast). To capture the pictures and floc structure analysis, Nikon Microscope Eclipse E400 with camera Olympus DP70 and Image ProÒPlus software were used. Samples of RAS in the control and ozonated reactor (before and after ozone treatment) were analyzed over a period of one month, to estimate the direct impact of ozonation on extracellular polymer substances (EPS) as well as to track the long-term influence of ozonation of sludge. Both total (bound and soluble) EPS, called un-
30% y = 1.94x + 0.06 2
25%
R = 0.71 phase 1
Decrease in TSS
2.2. Analysis
20%
phase 2
15%
10%
5%
y = 0.66x + 0.07 2
R = 0.13
0%
b
500% 450% 400%
continuous ozonation (d): 50
100
y = 35.76x + 0.25
150
200
250
TSS and COD data collection:
no ozone
phase 1 (41 d)
phase 2 (42 d)
EPS datacollection:
Increase in sCOD
0
350%
2
R = 0.53
300% 250% 200% 150% 100%
no ozone
(37 d)
Image Pro datacollection:
y = 14.17x + 0.35
50% 0% 0.00
2
R = 0.38
0.02
0.04
0.06
0.08
0.10
-1
no ozone Fig. 1. The schedule of sampling during 276 d of ozonation.
(13 d)
ozone dose (mg O 3 mg TSS initial excess sludge) Fig. 2. Decrease in WAS (a) and increase in soluble COD in ozonated RAS (b) in the reactors after ozone treatment; 1st phase 95–136 d, 2nd phase 190–232 d.
M.A. Dytczak, J.A. Oleszkiewicz / Chemosphere 73 (2008) 1529–1532
ted in Fig. 2b. Production of sCOD correlated with TSS destruction for both phases of the sampling. Consequently, the production of sCOD during ozonation in the alternating reactor decreased for the second phase of the tests (up to 170% only) in comparison to the first one (up to 400% for high ozone doses).
1531
equal in shape. In the control reactor which did not receive ozone, the flocs’ size was more variable, the occurrence of bigger flocs was observed, and flocs tended to have a greater perimeter length in comparison to those from the ozonated reactor. 4. Discussion
3.3. EPS content The alternating control SBR had consistent total EPS (378 ± 48 mg g1 VSS, n = 37) and bound EPS (223 ± 53 mg g1 VSS, n = 37) (Dytczak et al., 2006). In the ozonated reactor after long-term ozonation, both total and bound EPS levels increased significantly and were 488 ± 49 mg g1 VSS and 307 ± 41 mg g1 VSS, respectively (n = 37). Additionally, the ratio of bound to total polymers was higher in the ozonated reactor (62%) in comparison to the control (59%), showing some tendency of strengthening the floc structure. 3.4. Biomass structure Microscopic observation of sludge performed daily from the beginning of the research showed that biomass in the alternating reactors consisted mostly of bacteria with few filaments, inhabiting weak, thin, and open, more dispersed and elongated flocs (Dytczak et al., 2006, 2007). The presence of the anoxic zones in alternating reactors is responsible for partial, non-reversible deflocculation probably due to decreased microbial activity under anoxic conditions. To quantify the difference in floc structure observed microscopically, the floc area and perimeter length were compared. The histograms in Fig. 3 show the frequency of these parameters for control and ozonated alternating reactor. After long-term ozonation (data collected within 256–269 d of ozonation) alternating flocs became smaller, compact, round and more
a
15%
Frequency
control
10%
ozonated
A large deterioration in the effectiveness of solids destruction and sCOD production was observed in the second phase of ozone treatment. It appeared that prolonged ozonation strengthened the structure of flocs and made them more resistant to destruction by further ozone treatment. This phenomenon seems to be caused by a significant amount of additional sCOD delivered to the ozonated system, which in turn stimulates the EPS production. The increase of the fraction of bound EPS in the ozonated reactor shows that combined ozonation-biological treatment favours not only the quantity of EPS production, but also the degree of their binding into the floc structure. We have attributed this phenomenon (Dytczak et al., 2006) to an increase in the food-to-microorganisms (F/M) ratio. Both reactors received the same amount of synthetic wastewater but, because of the destruction of part of the solids and a corresponding additional sCOD release from cells, the F/M ratio is higher in the ozonated reactor which promotes microbial activity and increases the EPS yield. Excess, unutilized carbon in the system with high F/M could have been converted to intracellular storage granules and extracellular polymers that accumulate as EPS (Liao et al., 2001). This might explain why the amounts of both the total and bound EPS levels in ozonated reactors after prolonged ozonation (measured since 239 d of the ozone treatment) are higher than in the control reactor, which represented the initial conditions. Extracellular polymers are essential for floc formation; their increased amount helps in flocculation and overcomes the effect of disintegration by ozone. In this research, the stronger flocs with higher content of bound EPS turned out to be more ozone-resistant. Consequently, the effect of reduction of WAS solids was less pronounced in the reactors subjected to prolonged ozone treatment. Additionally, the benefits regarding the soluble carbon production for denitrification could be diminished in time. On the other hand, smaller and equal in size flocs would improve sludge settling properties, although dewaterability may deteriorate (Yen et al., 2002).
5%
5. Conclusions 0% 0
300
600
900
1200
1500
Floc area (mm2)
b
50%
Frequency
40%
control ozonated
30% 20% 10% 0% 0
100
200
300
400
Floc perimeter (μm) Fig. 3. The frequency of floc area (a) and perimeter (b) for flocs for control and ozonated reactor (samples from the RAS, 256–269 d).
The issue of progressive strengthening of continuously ozonated flocs may have large engineering significance for practical implementation of the method, and, to our knowledge, has not previously been reported. This research indicated that, for the prolonged operation of partial sludge ozonation, an increase in ozone doses may be required to continuously maintain the solids destruction and related denitrification improvement effect. The necessity of application of higher ozone doses will increase the operational costs of the treatment related to ozone generation. This negative consequence of process adaptation could of course be different in a full scale facility, where variations in influent composition and temperature cause activated sludge flocs to be more vulnerable to deflocculation. The finding should therefore be verified on a larger scale. Acknowledgements The research was sponsored by Natural Sciences and Engineering Research Council of Canada; Province of Manitoba; City of Winnipeg Water and Waste Department and EarthTech (Canada) Ltd.
1532
M.A. Dytczak, J.A. Oleszkiewicz / Chemosphere 73 (2008) 1529–1532
The excellent technical assistance with floc structure analysis of Jennifer Kroeker, Department of Microbiology, University of Manitoba, is greatly appreciated. References Ahn, K.-H., Yeom, I.T., Park, K.Y., Maeng, S.K., Lee, Y., Song, K.-G., Hwang, J.H., 2002. Reduction of sludge by ozone treatment and production of carbon source for denitrification. Water Sci. Technol. 46 (10), 121–125. American Public Health Association, 1998. Standard Methods for the Examination of Water and Wastewater, 20th ed. Washington, DC, USA. Boehler, M., Siegrist, H., 2004. Partial ozonation of activated sludge to reduce excess sludge, improve denitrification and control scumming and bulking. Water Sci. Technol. 49 (10), 41–49. Dytczak, M.A., Londry, K.L., Siegrist, H., Oleszkiewicz, J.A., 2006. Extracellular polymers in partly ozonated return activated sludge: impact on flocculation and dewaterability. Water Sci Technol. 54 (9), 155–164.
Dytczak, M.A., Londry, K.L., Siegrist, H., Oleszkiewicz, J.A., 2007. Ozonation reduces sludge production and improves denitrification. Water Res. 41, 543–550. Liao, B.Q., Allen, D.G., Droppo, I.G., Leppard, G.G., Liss, S.N., 2001. Surface properties of sludge and their role in bioflocculation and settleability. Water Res. 35, 339– 350. Park, K.Y., Lee, J.W., Ahn, K.H., Maeng, S.K., Hwang, J.H., Song, K.G., 2004. Ozone disintegration of excess biomass and application to nitrogen removal. Water Environ. Res. 76, 162–167. Sakai, Y., Fukase, T., Yasui, H., Shibata, M., 1997. An activated sludge process without excess sludge production. Water Sci. Technol. 36 (11), 163–170. Weemaes, M., Grootaerd, H., Simoens, F., Verstraete, W., 2000. Anaerobic digestion of ozonized biosolids. Water Res. 34, 2330–2336. Yasui, H., Shibata, M., 1994. An innovative approach to reduce excess sludge production in the activated sludge processes. Water Sci. Technol. 30 (9), 11–20. Yen, P.-S., Chen, L.C., Chien, C.Y., Wu, R.M., Lee, D.J., 2002. Network strength and dewaterability of flocculated activated sludge. Water Res. 36, 539–550.