Effects of Temperature and Hydraulic Residence Time (HRT) on Treatment of Dilute Wastewater in a Carrier Anaerobic Baffled Reactor

www.besjournal.com

BIOMEDICAL AND ENVIRONMENTAL SCIENCES 21, 460-466 (2008)

Effects of Temperature and Hydraulic Residence Time (HRT) on Treatment of Dilute Wastewater in a Carrier Anaerobic Baffled Reactor1 HUA-JUN FENG#, LI-FANG HU#, DAN SHAN+, CHENG-RAN FANG#, AND DONG-SHENG SHEN#,2 #

College of Environmental and Resource Sciences, Zhejiang University, Hangzhou 310029, Zhejiang, China; + Environmental Protection Bureau of Yuhang District of Hangzhou, Hangzhou 311100, Zhejiang, China

Objective To examine the effect of hydraulic residence time (HRT) on the performance and stability, to treat dilute wastewater at different operational temperatures in a carrier anaerobic baffled reactor (CABR), and hence to gain a deeper insight into microbial responses to hydraulic shocks on the base of the relationships among macroscopic performance, catabolic intermediate, and microcosmic alternation. Methods COD, VFAs, and microbial activity were detected with constant feed strength (300 mg/L) at different HRTs (9-18 h) and temperatures (10ć-28ć) in a CABR. Results The removal efficiencies declined with the decreases of HRTs and temperatures. However, the COD removal load was still higher at short HRT than at long HRT. Devastating reactor performance happened at temperature of 10ć and at HRT of 9 h. HRTs had effect on the VFAs in the reactor slightly both at high and low temperatures, but the reasons differed from each other. Microbial activity was sensitive to indicate changes of environmental and operational parameters in the reactor. Conclusion The CABR offers to certain extent an application to treat dilute wastewater under a hydraulic-shock at temperatures from 10ć to 28ć. Key words: Carrier anaerobic baffled reactor; Dilute wastewater; Hydraulic residence time; Temperature

INTRODUCTION

get worse at lower HRT, which increases hydraulic dead space. On the contrary, biological dead space is a major contributor to the overall dead space at a higher HRT, while its contribution decreases at a lower HRT since gas production disrupts channeling within the biomass bed. Therefore, the contradictory effects of hydraulic and biological dead space are not correlated with HRT or overall dead space. Severe channeling, caused by large hydraulic shocks, has been found to be beneficial since most of the biomass is not entrained in the flow, thus resulting in a low washout and a fast recovery in performance[9-10]. Generally speaking, a reactor process is stable with little or no reduction in system performance, irrespective of changes in reactor’s environmental and operational parameters (such as operational temperature, feed strength and flow rate). The duration of most hydraulic-shocks is short during the practical operation of small-scale facilities if the design is perfect. It is important for a purification system to make sure that the effluent quality is always satisfactory with the discharge standard limit, even if the shock occurs. However, the performance cannot exactly reflect characteristics of the reactor

A successful anaerobic treatment requires a microbial balance between the fast-growing acidogens and the slow-growing methanogens. However, it has been reported that anaerobic processes is sensitive to shock loading, which is a serious disadvantage, especially in places where balancing and equalizing facilities are limited[1-3]. Hydraulic-shock loading often results in process souring and failure due to an accumulation of VFAs[2]. A number of previous studies on the effect of HRT/shock loads on anaerobic digestion are available[4-8]. Grobicki and Stuckey have found that “dead space” (hydraulic) of the empty anaerobic baffled reactor (ABR) is around 8% by volume, while the overall dead space (biological and hydraulic) increases to around 18%[9]. The presence of biomass has no significant effect on hydraulic dead space that has been found to be a function of flow rate and number of bases. In addition, channeling of the feed flowing through the biomass bed in each compartment would 1

This research is a key project supported by the Science and Technology Department of Zhejiang Province (2005C13003). Correspondence should be addressed to Dong-Sheng SHEN. Tel: +86-571-86971156. Fax: +86-571-86945370. E-mail: [email protected] 2

0895-3988/2008 CN 11-2816/Q Copyright © 2008 by China CDC 460

CABR SHOCK LOAD: TEMPERATURE AND HRT

because the duration of the shock is too short to have these characteristics diluted by the original bulk in them. Theoretically, the time to reach a new steady state is one HRT and three HRTs for the plug flow reactor (PFR) and continuous stirred-tank reactor (CSTR), respectively[11]. Based on the previous studies[9] and our studies, the hydraulic characteristic of CABR is intermediate between PFR and CSTR, and about 95% of original water could be replaced after two HRTs. Therefore, using two HRTs as a duration time is feasible for a short-shock load study. Low temperature during anaerobic treatment is always associated with a poor reactor performance. At a lower temperature, dramatic increase in the saturation coefficient (Ks)[12] results in a higher effluent COD, but it does not necessarily mean that psychrophilic wastewater treatment is unfeasible. Most studies have used mesophilic or thermophilic systems to maintain stability in anaerobic reactor. Psychrophilic reactors are considered to be more stable than thermophilic ones, but biomass loss could be detrimental because of slow bacterial growth at a lower operational temperature[13]. Few studies are available on transient hydraulic shocks tests using dilute sewage wastewater at different temperatures (10 ć -28 ć ) for anaerobic treatment, even though there are a large number of studies on the effect of HRT/shock loads on anaerobic reactor treating high concentrated wastewaters under mesophilic or thermophilic conditions[4-8]. Furthermore, information is scarce on the relationship among macroscopic performances, catabolic intermediates and microcosmic alternation during shock loadings. The carrier anaerobic baffled reactor (CABR) (China patent 200620100157.2) developed by Shen at

461

Zhejiang University is packed with carriers made of hollow-sphere bamboo with a high specific surface area (2100 m2/m3), which combines the advantages of ABR and the characteristics of biofilm reactor. It is also a new high-rate anaerobic reactor for decentralized treatment, and can effectively remove organic pollution. This study aims to determine the effect of HRT (18 h-9 h) and operating temperature (10ć-28ć) on the performance and stability of a carrier anaerobic baffled reactor (CABR) and to gain insight into microbial responses to hydraulic shocks in CABR by investigating the relationships among macroscopic performance (COD), catabolic intermediate (VFAs) and microcosmic alternation (microbial activity). MATERIALS AND METHODS CABR was developed from ABR[14-16]. In brief, the CABR was made of 8 mm thick transparent Perspex (600 mmh140 mmh300 mm, Fig. 1). The reactor is rectangular in shape, comprising 6 chambers with an equal volume, and its active reactor volume is 17 L. The width of the chamber down-comer is 25 mm, while that of up-comer is 60 mm. The sampling ports are located at 200 mm back of the effluent port of each chamber. Six up-comer regions are filled with hollow-sphere carriers made of bamboo (diameter 1.5 cm). The carriers have a high specific surface area, which could reach up to 2100 m2/m3 having a high porosity of 95%. The bamboo carriers enable the biomass retention in attached form and are able to entrap suspended solids (SS) from domestic sewage. The reactor is operated in a thermostat room (28ć±1ć, 18ć±1ć, and 10ć±1ć).

FIG. 1. Schematic of the carrier anaerobic baffled reactor (CABR) having a packing made of hollow-sphere bamboo. 1: Influent inlet, 2: Effluent outlet, 3: Liquid sampling port, 4: Biogas, 5: Hollow-sphere carriers.

The CABR is seeded with screened digested sludge (~3.5l) from Hangzhou Sibao Wastewater Treatment Plant in Zhejiang Province, China. The total suspend solids (TSS) and volatile suspend solids (VSS) of sludge are 23.8 g/L and 9.2 g/L,

respectively. The CABR was initially fed with domestic sewage from Zhejiang University during start-up at 28ć±1ć and at HRT of 48 h, derived mainly from restaurants and dormitories. The reactors were

462

FENG ET AL.

acclimatized for over a period of 21 days at ideal temperature under which digesters were known to perform in an optimal way. The HRT was gradually decreased from 48 h to18 h by increasing the flow rate for 3 months. The total COD and SS removal efficiency was 69% and 82% at HRT of 18 h and 28ć, respectively. In order to avoid problems with varying

compositions and degradability of the feed, a buffered synthetic carbohydrate-protein substrate was supplemented with nutrients and trace metals (Table 1), which could be used to prepare various substrate concentrations by simple dilution with tap water. Fresh synthetic substrate was prepared each day in a 28-liter container. The influent feed was pumped with a peristaltic pump.

TABLE 1 Feed Compositions Used in the Study Feed Components NH4Cl

Weight (g)

Trance Nutrients

Weight (g)

40

CoCl2.6H2O

0.0238

Sugar

48

FeCl2.4H2O

0.157

Urea

1.8

MnCl2.4H2O

0.0075

K2HPO4

8.4

Na2MoO4.2H2O

0.0075

KH2PO4

6.6

NiCl2.6H2O

0.009

NaHCO3

60

Peptone

60

Sodium Dodecylbenzenesulfonate

0.8

Humic Acid

0.8

Note. The constituents to make 10 L of feed with concentration of 12 g/L COD.

The shock tests were conducted in three steps as follows. Firstly, the reactor was operated for 1 month with a feed strength of 300 mg/L COD with a HRT of 18 h at 28ć±1ć to provide a steady baseline for this study. After reaching a steady state, HRT was decreased to 12 h at 300 mg/L COD. Before the operational conditions were changed, the reactor was operated for two HRTs (24 h) and the sample was collected. The HRT fell to 9 h at the same feed strength. The reactor was operated for two HRTs (18 h) and the sample was collected. The process temperature was controlled at 28ć±1ć. Secondly, the operational temperature was decreased to 18ć ±1ć by 1ć per day[17]. The reactor was operated for 10 days with a constant feed strength at HRT of 18 h to reach a steady state. Then, the operation was similar to the first step. Finally, the operational temperature was further decreased to 10ć±1ć by 1ć per day. The operating conditions for the third step were similar to the second one except for temperature. COD was performed as previously described[18]. VFAs were determined from membrane-filtered samples by gas chromatography[19]. TTCdehydrogenase activity (DHA) in activated-sludge was measured as previously described[20]. One enzyme unit was defined as the enzyme amount producing 1 ȝg TF per hour[21]. Analysis of variance was performed using

SPSSTM v.13. Paired sample test was also applied. RESULTS Effect of HRT and Temperature on COD CABR performed well at the operational temperature of 28ćin this studyˈand the final effluent COD ranged from 50 mg/L at a HRT of 18 h to 88 mg/L at a HRT of 9 h . At a HRT of 18 h, the removal efficiency dropped from 83.10% at 28ć to 67.19% at 10ć (Fig. 2, Table 2). However, the concentration of effluent COD was less than 100 mg/L, which was below the discharge standard limit (GB8978-1996, China).

FIG. 2. COD profile in the chambers of CABR at different temperatures and HRTs.

CABR SHOCK LOAD: TEMPERATURE AND HRT

463

TABLE 2 Removal Efficiency of Final Effluent in CABR at Different HRTs and Temperatures Temperature (ć)

HRT (h)

Range

Average

67.19%

15.91%

77.73%

77.03%

65.46%

16.79%

74.91%

62.42%

42.63%

27.85%

58.51%

12.62%

20.48%

24.56%

78.61%

74.12%

58.43%

28

18

10

18

83.10%

82.90%

12

82.25%

9

70.48%

Range Average

At a constant feed strength and operational temperature, the removal efficiency declined with the decreasing HRT. The average COD removal efficiency was 77.73% at a HRT of 18 h, 74.91% at a HRT of 12 h, and 58.51% at a HRT of 9 h, respectively. The difference in removal efficiencies was not significant between 18 h and 12 h of the HRT. However, drastic drop in removal efficiency was observed when the HRT decreased to 9 h, indicating that the HRT greatly affected the performance of CABR. However, the COD removal load at a short HRT was still higher than that at a long HRT in this study, which was 0.31 kg/m3gd at a HRT of 18 h, 0.45 kg/m3gd at a HRT of 12 h, and 0.47 kg/m3gd at a HRT of 9 h, respectively. A similar phenomenon also occurred with the decreasing temperature. The performance of the reactor deteriorated extensively when facing a drastic shock at a short HRT. Devastating performance of the reactor was observed as temperature dropped from 28ć to 10ć, and the COD removal efficiency fell from 78.61% to 58.43%. The decrease in HRT and temperature resulted in a weak performance of the whole reactor, but the COD gradually decreased in each succeeding chamber and the removal efficiency could be improved with sufficient domestication except at the

HRT of 9 h and 10ćFig. 2 . As the temperature decreased to 10ćand HRT was less than 9 h, the reactor performance was devastating, that is, the later chambers attributed to the COD removal efficiency slightly and excessive loss of solids occurred in the effluent. In addition, the COD removal efficiency failed to improve after operation for 10 d (data not shown). Effect of HRT and Temperature on VFAs The VFA concentration increased with the decreasing operational temperature. A hydraulic shock loading inhibited methanogenic bacteria to some extent, and then caused an accumulation of VFAs, but such an effect was slight at a high temperature (28ć in this study). As the temperature fell to 18ć, the effect was increased (Fig. 3). The concentration of VFAs trended to decrease gradually at 28ć and 18ć, while the HRT decreased from 18 h to 9 h, and the final concentration of effluent VFAs at 28ć and 18ć increased from 8 mg/L to 22 mg/L and from 10 mg/L to 27 mg/L, respectively, suggesting that the effect of HRTs on effluent of VFAs was higher than that of temperatures due to insufficient time of contact between biomass and substrate[22].

FIG. 3. VFA profile in the chambers of CABR at different temperatures and HRTs.

The results at operational temperature of 10ć differed, whereas the concentration of VFAs ascended

gradually irrespective of HRT. Furthermore, the concentration of VFAs at a HRT of 9 h was similar to

464

FENG ET AL.

that at a HRT of 12 h, but the difference was significant in the COD removal efficiency (66% at a HRT of 12 h and 42% at a HRT of 9 h), demonstrating that acidification phase was not fully accomplished at 10ć or at a HRT of 9 h. Detailed results of VFAs in each case were not presented here due to lack of space. A considerably high level of acetate was detected and considered as the main ingredient of VFA in the reactor except for the first two chambers at a HRT of 12 h, and tiny propionate (less than 5 mg/L) was also detected in all the chambers at a HRT of 9 h and 10ć. Effect of HRT and Temperature on DHA The activity level of dehydrogenase (DHA), corresponding to the microbial dehydrogenation capability of organic substance, indirectly indicated the microbial activity[23]. As illustrated in Fig. 4,

temperature and HRT had an effect on the DHA activity, and higher DHA activity was observed at a higher temperature and HRT. The DHA activity decreased gradually in each succeeding chamber at 28ć, but its peak of DHA activity shifted to the later chamber at a lower temperature, that is, the peak presented in the second chamber at 18ć and in the third one at 10ć. Another significant phenomenon was that the activity of the biofilm attached to the carriers was higher than that of floc sludge from bottom reactor (paired samples test, at 1% confidence level). However, the difference in the above sludge was much smaller than that at different temperatures or feed strengths, since the DHA activity at 28ć was much stronger than that at 10ć and in the front than in the rear of the reactor, showing that the microorganism activity in the rear of CABR could not be improved by decreasing the HRT alone.

FIG. 4. DHA profile in the chambers of CABR at different temperatures and HRTs. The upper part indicates the sludge attached to the carriers and the lower part indicates the sludge at the bottom of the reactor.

DISCUSSION Based on the Arrhenius relationship, a decline in temperature results in decreasing reaction rate, and biological reaction rate for a drop at 10ć is expected to decrease by a half value. However, it has been reported that a decline in temperature from 35ć to 25ć does not lead to a significant fluctuation in performance of the reactor, but the mortal performance would occur below 15ć. The saturation coefficient (Ks) increases with the decreasing temperature, resulting in a higher effluent COD. These findings indicate that at a temperature above 18ć there is no difference in CABR performance, which is consistent with the previous reports[12]. The effect of a reduced temperature on system performances could be compensated by the high

biomass population[17]. However, the reactor faces a drastic shock at the shortest HRT, and the performance deteriorates extensively. The disturbance for sludge is mainly caused by flow mixture, rather than by biogas due to a small quantity of biogas production with a lower feed strength of 300 mg/L COD (most biogas dissolves in the water body and discharges). The disturbance increases with the decreasing HRT, leading to an increased mass transfer rate between the substrate and biomass, which improves the reaction rate of the system[24]. That is why the COD removal load is still higher at a short HRT than at a long HRT. Some investigations indicate that the saturation coefficient (Ks) decreases with thee increasing temperature[12], resulting in lower effluent VFA concentration. In our study, although the HRT decreased from 18 h to 9 h at 28ć, the substrate

CABR SHOCK LOAD: TEMPERATURE AND HRT

could be completely consumed by microorganism. However, the lower the temperature was, the worse the buffer capacity. As the temperature decreased to 10ć, no significant difference was observed in the concentration of VFAs at three different HRTs, due to the incomplete acidification by severe inhibition of microbial activity. That is why a certain quantity of propionate is produced during fermentation production. Propionate is produced in the periods of startup, shock-load or over-load[25]. Propionate production and accumulation could demonstrate whether a well-balanced anaerobic system is operated. The propionate consumers are inhibited by a low temperature, resulting in a higher accumulation of propionate. Based on the standard Gibbs free energy data[26], it is the most difficult to translate propionate into methane, which results in acidification. However, such acidification fails to occur in our reactor, since the total concentration of VFAs (less than 66 mg/L in this study) is limited with a low feed strength. Microbial activity is the most sensitive indicator for the changes in reactor’s environment and operational parameters. In the present study, when the temperature dropped from 28ć to 18ć at a HRT of 18 h, the COD removal efficiency obviously remains unchanged , but the DHA removal efficiency does change. In other words, microbial activity is also a buffer capacity of microorganism. Of course, the high biomass population could compensate for the low microbial activity. Theoretically, microorganism in the rear of CABR could obtain more nutrients at a shorter HRT (at a constant feed strength), but it does not necessarily mean that its activity should be enhanced consequently because the new operational environment is not optimal due to a short acclimation time (two HRTs). The difference in microbial activity between biofilm and floc sludge indicates that saturation constant Ks of biofilm, the affinity of the bacteria to the substrate, is lower than that of the floc sludge at the bottom of the reactor. In conclusion, CABR reactor performs well at a higher temperature (18ć and 28ć), and could resist various hydraulic load shocks. However, such a resistance capacity may decline with the deceasing temperature. HRT has a slight effect on the VFAs in the reactor at high and low temperature. Microbial activity is the most sensitive indicator for the changes in reactor’s environment and operational parameters. CABR can be used for the treatment of low-strength wastewater and is promising for the decentralized treatment of dilute domestic sewage in rural areas. ACKNOWLEDGEMENTS This research was supported by the Science and

465

Technology Department of Zhejiang Province (2005C13003). The authors thank J. L. LIU for his providing technical assistance, and acknowledge the reviewers of this article. REFERENCES 1. Borja R, Banks C J (1995). Response of an anaerobic fluidized bed reactor treating ice-cream wastewater to organic, hydraulic, temperature and pH shocks. Journal of Biotechnology 39, 251-259. 2. Chua H, Hu W F, Yu P H F, et al. (1997). Responses of an anaerobic fixed-film reactor to hydraulic shock loadings. Bioresource Technology 61(1), 79-83. 3. Boardman G D, Tisinger J L, Gallagher D L (1995). Treatment of clam processing wastewaters by means of upflow anaerobic sludge blanket technology. Water Research 29(6), 1483-1490. 4. Denac M, Dunn I J (1988). Packed- and fluidized-bed biofilm reactor performance for anaerobic wastewater treatment. Biotechnology and Bioengineering 32, 159-173. 5. Guiot S R, Podruzny M F, Mclean D D (1989). Assessment of macroenergetic parameters for an anaerobic upflow biomass and filter (UBF) reactor. Biotechnology and Bioengineering 34, 1277-1288. 6. Borja R, Banik G C, Wang Z (1994). Stability and performance of an anaerobic downflow filter treating slaughterhouse wastewater under transient changes in process parameters. Biotechnology and Applied Biochemistry 20, 371-383. 7. Xing J, Tilche A (1992). The effect of hydraulic retention time on the hybrid anaerobic baffled reactor performance at constant loading. Biomass Bioenergy 3(1), 25-29. 8. Converti A, Borghi M D, Ferraiolo G (1993). Influence of organic loading rate on the anaerobic treatment of high strength semisynthetic waste waters in a biological fluidized bed. Chemical Engineering Journal 52, B21-B28. 9. Grobicki A, Stuckey D C (1992). Hydrodynamic characteristics of the anaerobic baffled reactor. Water Research 26(3), 371-378. 10. Nachaiyasit S, Stuckey D C (1997). The effect of shock loads on the performance of an anaerobic baffled reactor (ABR). 2. Step and transient hydraulic shocks at constant feed strength. Water Research 31(11), 2747-2754. 11. Flora J R V, Suidan M T, Biswas P, et al. (1995). A modeling study of anaerobic biofilm system. I. detailed biofilm modeling. Biotechnology and Bioengineering 46, 43-53. 12. Nachaiyasit S, Stuckey D C (1997). Effect of low temperatures on the performance of an anaerobic baffled reactor (ABR). Journal of Chemical Technology and Biotechnology 69(2), 276-284. 13. Masse L, Masse D I (2005). Effect of soluble organic, particulate organic, and hydraulic shock loads on anaerobic sequencing batch reactors treating slaughterhouse wastewater at 20ć. Process Biochemistry 40(3-4), 1225-1232. 14. Barber W P, Stuckey D C (1999). The use of the anaerobic baffled reactor (ABR) for wastewater treatment: A review. Water Research 33(7), 1559-1578. 15. Li G, OuYang F, Yang L Z (2001). Study on the performance of ABR reactor: review and summary. China Biogas 19(3), 9-14. 16. Wang J L, Huang Y H, Zhao X (2004). Performance and characteristics of an anaerobic baffled reactor. Bioresource Technology 93(2), 205-208. 17. Ndon U J, Dague R R (1997). Effects of temperature and hydraulic retention time on anaerobic sequencing batch reactor treatment of low-strength wastewater. Water Research 31(10), 2455-2466. 18. APHA (1992). Standard Methods for the Examination of water and wastewater, 18th ed. America Public Health Association,

466

FENG ET AL.

American Water Works Association and Water Environment Federation. Washington DC, USA. 19. Elmitwalli T A, Soellner J, De Keizer A, et al. (2001). Biodegradability and change of physical characteristics of particles during anaerobic digestion of domestic sewage. Water Research 35(5), 1311-1317. 20. Zhu N W, Min H, Chen M C, et al. (1996). The study of determination on TTC-Dehydrogenase activity. China Biogas 14(2), 3-5. 21. Klapwuk A, Drent J, Steenvoorden J H A M (1974). A modified procedure for the TTC-dehydrogenase test in activated-sludge. Water Research 8(2), 121-125. 22. Nachaiyasit S, Stuckey D C (1997). The effect of shock loads on the performance of an anaerobic baffled reactor (ABR). 1. Step changes in feed concentration at constant retention time. Water Research 31(11), 2737-2746.

23. Song T H, Zhao Y S, Cui Y B, et al. (2006). Study on Acidate Bacteria Activity in Anaerobic Baffled Reactor. Journal of Jmn University (Earth Science Edition) 36(3), 429-432. 24. Angenent L T, Banik G C, Sung S W (2001). Anaerobic migrating blanket reactor treatment of low-strength wastewater at low temperatures. Water Environment Research 73(5), 567-574. 25. Cohen A, Breure A M, van Andel J G, et al. (1980). Influence of phase separation on the anaerobic digestion of glucose--I maximum COD-turnover rate during continuous operation. Water Research 14(10), 1439-1448. 26. Rittman B E, McCarty P L (2001). Environmental Biotechnology: Principles and Applications, 2nd ed. McGraw-Hill College. (Received August 8, 2007

Accepted July 28, 2008)