Sevgi Demirel1 İbrahim Uyanık2 Adem Yurtsever3 Hakan Çelikten4 Deniz Uçar3
1Faculty of Engineering, Department of Environmental Engineering, Nigde University, Nigde, Turkey
2Faculty of Engineering, Department of Environmental Engineering, Erciyes University, Kayseri, Turkey 3
Faculty of Engineering, Department of Environmental Engineering, Harran University, Sanliurfa, Turkey 4Faculty of Engineering, Department of
Environmental Engineering, Ardahan University, Ardahan, Turkey
Research Article
Simultaneous Bromate and Nitrate Reduction in
Water Using Sulfur-Utilizing Autotrophic and
Mixotrophic Denitri
fication Processes in a Fixed
Bed Column Reactor
Carcinogenic bromate (BrO
3) can be present in drinking water as a result of its
formation from bromide (Br
) during ozonation. A
fixed bed column reactor filled with
elementel sulfur and limestone was operated for about six months under autotrophic
and mixotrophic (autotrophic
þ heterotrophic) conditions at 30°C. The reactor was
operated at the hydraulic retention time (HRT) ranging from 16.5 to 10.1 h at
autotrophic conditions. Under mixotrophic conditions, 45 mg/L NO
3-N was removed
completely at C/N ratio (mg CH
3OH/mg NO
3-N) between 0.55 and 1.66 at HRT of 10.1 h.
The average ef
fluent pH was 7.8 and the sulfate concentration was lower than
the Environmental Protection Agency limits at the mixotrophic stages. Ef
ficient
simultaneous BrO
3and nitrate removal was achieved at feed concentrations of 100
–
500
mg/L BrO
3and 45 mg/L nitrate under autotrophic and mixotrophic conditions.
Ef
fluent Br
measurements indicated that BrO
3was completely reduced without
accumulation of by-products.
Keywords: Carbon requirement; Co-metabolic processes; Drinking water; Organic supplementation; Sulfate production
Received: June 24, 2013; revised: August 14, 2013; accepted: October 13, 2013 DOI: 10.1002/clen.201300475
1 Introduction
Carcinogenic bromate (BrO3) can be present in drinking water as a result of its formation from bromide (Br) during ozonation [1]. Br is found especially in regions with saltwater intrusion, dissolution from sedimentary rock, domestic or industrial effluent. Additionally, Bralso may arise from anthropogenic sources, such as ethylene dibromide, agricultural chemicals. BrO3formation has become an important issue as more utilities are considering ozone to combat Cryptosporidium and Giardia, color, and taste and odor problems [2]. BrO3 at levels ranging from 5 to 100mg/L may be found following ozonation of freshwater containing Br concen-trations of 3.2–58 mg/L [3]. The U.S. Environmental Protection Agency (US EPA) and EU Drinking Water Directive set a maximum BrO3 level within drinking water of 10mg/L [4]. A number of treatment methods and investigations are available for BrO3 removal in drinking water including coagulation–filtration [5], granular activated carbon [6], nanofiltration [7], and ultrafiltra-tion [8]. Although these methods are effective, treating BrO3 containing concentrated waste, they are not cost effective regarding to energy consumption.
Nitrate is a natural anion in the environment and only a pollutant under situations of excess caused by over fertilization in agricultural areas. The US EPA has set the maximum contaminant levels of 10 mg NO3-N/L for drinking water [9, 10]. The ability of biological BrO3 reduction has been demonstrated for some mixed and pure cultures of denitrifying bacteria like Pseudomonas spp. [11]. The biological denitrification, either autotrophic or heterotrophic has been proposed separately for nitrate and/or BrO3removal [12, 13]. The disadvantage of the heterotrophic denitrification is its requirement for additional carbon sources for the activity of microorganisms (Eq. (1)). Consequently, some post-treatment operations for the elimination of the process contaminants are also needed [14, 15].
NO3þ 1:08 CH3OHþ 0:24 H2CO3
! 0:056 C5H7NO2þ 0:47 N2þ 1:68 H2Oþ HCO3 ð1Þ
In contrast, autotrophic denitrification does not require organic supplementation. Additionally, due to low biomass production, the risk of bacterial contamination and the operational cost are reduced in autotrophic denitrification process [14]. Sulfur-based autotrophic denitrification is an effective alternative due to low cost and availability of elemental sulfur [12]. In this process, the elemental sulfur acts as an electron donor, and nitrate serves as an electron acceptor. Hence, when nitrate is reduced to nitrogen gas, sulfur is oxidized to sulfate (Eq. (2)).
NO3þ 1:1 Sþ 0:4 CO2þ 0:76 H2Oþ 0:08 NH4þ
! 0:08 C5H7NO2þ 1:1 SO42þ 0:5 N2þ 1:28 Hþ ð2Þ
Correspondence: Dr. S. Demirel, Faculty of Engineering, Department of Environmental Engineering, Nigde University, 51240, Nigde, Turkey E-mail: sevgi.demirel@gmail.com
Abbreviations: EPA, Environmental Protection Agency; HRT, hydraulic retention time
The ability to reduce BrO3to the Br by nitrate and chlorate reducing microbial cultures can be obtained by cometabolism of nitrate reductase and chlorate reductase enzymes [16]. BrO3 reduction releases less energy compared to nitrate reduction [17, 18]. Drinking waters containing both BrO3and nitrate may necessitate to the development of simultaneous nitrate and BrO3removing technologies.
The objective of this study was to evaluate the robustness of the sulfur-based autotrophic and mixotrophic (autotrophic and heterotro-phic) denitrification process for simultaneous reduction of nitrate and BrO3using a lab-scale column bioreactor. Sulfate production and external carbon requirement should be minimized by stimulating simultaneous sulfur-based autotrophic and heterotrophic denitri fica-tion processes (mixotrophic). This study is thefirst study investigating the simultaneous BrO3and nitrate removal using autotrophic and mixotrophic denitrification processes for drinking water treatment.
2 Materials and methods
2.1 Fixed-bed column reactor
A laboratory-scalefixed-bed column reactor used in this study was made of glass with an empty bed volume of 400 mL (Fig. 1). The
column reactor was filled with sulfur (0.5–1 mm)/limestone (0.5– 1 mm) particles at a volume ratio of 1:3 [19]. The reactor was covered with aluminum foil to prevent the growth of phototrophic bacteria. A denitrifying sludge obtained from the anoxic tank of Bardenpho process located in Harran University Campus (Sanliurfa, Turkey) was used as inoculum. The column reactor was operated in continuous up-flow mode at 28–30°C in a temperature controlled room.
During thefirst seven days, the reactor was fed with tap water containing 45 mg/L NO3-N as KNO3and 50 mg/L K2HPO4solution for adaptation of denitrifiers (data not shown). The feed solution was deoxygenated by passing through the N2gas for 5 min. Then, the feed was kept under anaerobic conditions in a collapsible feed container and stored 4°C in refrigerator during all periods.
After the adaptation period, the fixed bed column reactor was operated at different BrO3concentrations (prepared with 500 mg/L of stock KBrO3 solution) and hydraulic retention times (HRT) at autotrophic conditions (Tab. 1). HRT was calculated considering the empty bed volume. The HRT of the reactor was decreased from 16.5 to 10.1 h periodically (Tab. 1) during autotrophic denitrification. A peristaltic pump (Cole Parmer) was used to deliver the feed solution to the column reactor. In order to stimulate simultaneous autotrophic and heterotrophic denitrification (mixotrophic denitri-fication), methanol was added as an external organic carbon at different concentrations after period 5 (Tab. 1).
The production nitrogen gaseous was measured by liquid displacement methods (Fig. 1) and the gas volume was compared to the theoretical value (mL/d) using the following equation [20]:
Theor: value ðmL=dayÞ ¼ removed NO2 N ðmg=LÞ
22:4 mL 28 mg 273T:15 ððKÞKÞflow rateðL=dayÞ
ð3Þ
2.2 Sampling and analytical techniques
Samples for analysis were collected from the reactor effluent at least three times a week for the measurement of BrO3, Br, NO3, NO2, dissolved organic carbon, sulfate, pH, and alkalinity. The feed solution was sampled once a week for the determination of BrO3, Br, NO3, NO2, dissolved organic carbon, sulfate, pH, and alkalinity. All samples were filtered through a 0.45 mm-pore-size sterile filter and stored at 4°C until analysis. NO3, NO2, and SO42 concentrations were determined by ion chromatography Figure 1. Schematic representation of the lab-scalefixed-bed column
reactor.
Table 1. Operation conditions used in FBCR
Stage Period Days
Bromate feed
(mg/L)a) NO(mg/L)3-N feed HRT(h) Methanol feed(mg/L)b)
Autotropic 1 0–19 100 (0.8) 45 16.5 – 2 19–38 100 (0.8) 45 12.4 – 3 38–58 100 (0.8) 45 10.1 – 4 58–108 500 (3.8) 45 10.1 – Mixotrophic 5 108–119 100 (0.8) 45 10.1 25 (9.8) 6 119–128 100 (0.8) 45 10.1 50 (19.7) 7 128–139 100 (0.8) 45 10.1 75 (29.5) 8 139–151 500 (3.8) 45 10.1 25 (9.8) 9 151–162 500 (3.8) 45 10.1 50 (19.7) 10 162–173 500 (3.8) 45 10.1 75 (29.5) a)
Values in parenthesis represent the bromate feed asmmol/L.
(Schimadzu, Prominence HIC-NS). BrO3 and Br were measured using ion chromotography (DIONEX-ICS 3000 with AS19 column, BrO3 detection limit of 3mg/L). DOC was measured by TOC analyzer (Shimadzu, Japan). Alkalinity was measured according to standard methods.
3 Results and discussion
3.1 Performance of autotrophic denitrification
process
The column reactor wasfirst operated under autotrophic conditions for 108 days (periods 1–4, Tab. 1) at 45 mg/L NO3-N and 100–500 mg/L BrO3(0.8–3.8 mmol/L). During these periods, the HRT of the column reactor was stepwise decreased from 16.5 to 10.1 h and the reduction of BrO3and nitrate monitored (Fig. 2A and B). In thefirst ten days of the operation both nitrate and BrO3 were detectable in the effluent, although each decreased linearly to undetectable levels. Under steady-state conditions, the BrO3concentration was<3 mg/L. Although increasing BrO3 concentration to 500mg/L on day 58 caused effluent BrO3concentration to increase to about 250mg/L, the concentration decreased to undetectable levels within 10 days. The measured Brconcentration at the effluent demonstrated that all the BrO3was stoichiometrically reduced without any by-product formation under steady-state conditions. Decreasing the HRT from 16.5 to 10.1 h at 45 mg/L of influent nitrate concentration did not adversely affect the autotrophic denitrification performance. More-over, BrO3and nitrate were removed simultaneously from water, without detectable nitrite in the effluent.
At the autotrophic denitrification conditions, sulfate formation at the end of the experiments was in good agreement with theoretical sulfate determined based on the consumption of nitrate (Fig. 3A). Although sulfur-based autotrophic denitrification has several advantages, its main disadvantages are sulfate and acid formation. Similarly, Sahinkaya and Dursun [21] reported that mixotrophic
process combining heterotrophic and autotrophic denitrification processes can be used to control the amount of sulfate formation.
3.2 Performance of mixotrophic
(autotrophic
þ heterotrophic) denitrification process
After day 108 (period 5), thefixed-bed column reactor was fed with methanol (25–75 mg/L) in addition to 45 mg/L NO3-N and two concentrations of BrO3(0.8–3.8 mmol/L; Tab. 1).
During mixotrophic operational conditions, the methanol con-centrations in the feed were lower than the heterotrophically required amount (Eq. (1)) as nitrate is removed by both autotrophs and heterotrophs. According to Eqs. (1) and (2), 47 g of methanol is required per g of NO3-N reduction [15]. Therefore, the fraction of nitrate reduced by heterotrophic denitrifiers under mixotrophic conditions was calculated as 22.5, 45.0, and 66.5% (based on Eq. (1)), respectively, for periods 5–8, 6–9, and 7–10 (Tab. 1).
Similar to the autotrophic operating conditions, BrO3 and nitrate removed (Fig. 2). BrO3concentrations were below detection limit (3mg/L) throughout the operation, although again a transient increase in BrO3concentration was observed when the influent concentration was stepwise increased to 500mg/L on day 140. A logarithmic decrease of BrO3simultaneously with a stoichiometric increase in Brwas also noted by Butler et al. [22]. The stoichiometric conversion of BrO3 to Br was almost one (3.86 0.01 mmol/L of BrO3were reduced and 3.94 0.009 mmol/L of Br were formed) similar to Matos et al. [18] (Fig. 2B).
Sulfur utilizing autotrophic denitrifiers produces 1.28 mol of hydrogen ions when 1 mol of nitrate-nitrogen is reduced according to Figure 2. Influent and effluent NO3-N or NO2-N (A) and BrO3or Br
(B) variations.
Figure 3. Influent and effluent sulfate (A), alkalinity (B), and pH (C) variations at different stages.
Eq. (2). On the other hand, heterotrophic denitrifiers produce alkalinity and reduce pH during denitrification. The influent pH was constantly around 8.1. In the reactor, the effluent pH (7.4–8.2) was generally lower than influent pH for both autotrophic and mixotrophic conditions (Fig. 3). Influent and effluent alkalinity concentrations were close to each other, showing that limestone supplied enough alkalinity under studied conditions (Fig. 3B and C). Sulfate is one of the end products of sulfur oxidizing autotrophic denitrification processes. In this study, the SO42 concentration decreased in the mixotrophic denitirification process because a portion of the nitrate was reduced by heterotrophs. The effluent sulfate concentration decreased with increasing feed methanol concentration as expected (Fig. 3A). During thefinal period of the reactor operation with greatest methanol concentration, the effluent sulfate concentration decreased to the US EPA drinking water limit value of 250 mg/L. Lowered SO42concentration demonstrates that autotrophic and heterotrophic denitrification occurred simulta-neously in the column reactor. If the water was treated with autotrophic conditions only, the concentration of SO42in effluent was above the maximum contaminant level (EPA Drinking Water Standards; Fig. 3A).
The C/N ratio is an important factor influencing the efficiency of heterotrophic denitrification process. The effects of type of carbon source on heterotrophic denitrification efficiency were conducted
by Lee et al. [23]. They reported that when methanol was used as a carbon source, denitrification efficiency was 100 and 60% of nitrate was denitrified heterotrophically. In this study, the autotrophic and heterotrophic denitrifications were combined in one column reactor for BrO3 and nitrate removal from water. This study showed that autotrophic and heterotrophic denitrifiers can remove 45 mg NO3-N/L at C/N ratio (mg CH3OH/mg NO3-N) between 0.55 and 1.66, which is lower than the stoichiometric ratio of 2.47 for complete heterotrophic denitrification (Fig. 4A). Additionally, mixotrophic conditions permitted BrO3 concentrations to be reduced to below the environmental limit (10mg/L) when either 100 or 500mg/L (0.8–3.8 mmol/L) was in the feed.
BrO3removal capacity of afixed-biofilm denitrifying bioreactor was investigated by Hijnen et al. [14]. They reported that the application of a denitrifying bioreactor was not feasible due to excessive external organic carbon addition. The residual organic carbon will pollute the treated water when the added external organic carbon is excessive. Additionally, excessive carbon will require a robust post-treatment to avoid microbial problems of the distributed drinking water. The results show that mixotrophic denitrifiers can reduce external organic carbon consumption (Fig. 4A).
The produced gas volume was compared to the theoretical value obtained using Eq. (3). The produced and the theoretical N2 gas production rates were in close agreement at all stages (Fig. 4B).
To compare with other biological BrO3removal systems, the main operation parameters for the best removal capacity of each reactor are summarized in Tab. 2. The highest BrO3removal rate obtained in current study was 0.8mg/(L min). This removal rate is directly comparable to the value obtained in previous studies [14]. A higher removal rate of 102.2mg/L min) reported by van Ginkel et al. [24]. The higher influent BrO3concentrations compared with the current study and no nitrate influent were used in van Ginkel et al.’s study [24]. In the current study, BrO3and nitrate were removed simultaneously from water, without detectable nitrite and nitrate in the effluent.
4 Concluding remarks
Over the course of 173 days, the performance of afixed-bed column reactor with combined autotrophic and mixotrophic denitrification was investigated. This study is the first study investigating the simultaneous BrO3 and nitrate removal using autotrophic and mixotrophic denitrification processes in one reactor. The results showed that stimulating simultaneous sulfur-based autotrophic and heterotrophic denitrification processes can decrease sulfate production and external carbon requirement for drinking water treatment. Consequently, BrO3 and nitrate removal can be Figure 4. The variations of influent and effluent DOC concentrations
(A) and observed and theoretical gas production (B).
Table 2. Comparison of bromate removal rates with other biological systems
Reactor Influent bromate (mg/L) Influent nitrate (mg/L) T (°C) HRT Bromate removal
rate (mg/L min) Ref.
Denitrifying bioreactor 25–35 85 12 18 min 0.3–0.8 [14]
Biologically active carbon filter 20 5.1 – 25 min 0.7 [6]
Anaerobic batch reactor 294.4 – 20 48 h 102.2 [24]
Fixed film reactor 1100 30.7 20.1 40 h 0.4 [22]
Denitrifying biofilm reactor 1500 5 – 50 min 30 [25]
Ion exchange membrane bioreactor 200 60 23 8.3 h 0.04 [18]
considered co-metabolic processes. This study can be a contribution for the development of a biological process to remove BrO3and nitrate from drinking water.
Acknowledgments
This research was funded by TUBITAK (Project No.: 111Y165).
The authors have declared no conflict of interest.
References
[1] U. von Gunten, U. Pinkernell, Ozonation of Bromide-Containing Drinking Waters: A Delicate Balance between Disinfection and Bromate Formation, Water Sci. Technol.2000, 41, 53–59.
[2] M. Kirisits, V. Snoeyink, Reduction of Bromate in a BAC Filter, J. Am. Water Works Assoc.1999, 91 (8), 74–84.
[3] S. N. Davis, J. T. Fabryka-Martin, L. E. Wolfsberg, Variations of Bromide in Potable Ground Water in the United States, Ground Water 2004, 42, 902–909.
[4] R. Clark, B. Bouting, Controlling Disinfection By-Products and Microbial Contaminants in Drinking Water, US Environmental Protection Agency, Cincinnati2001.
[5] M. Siddiqui, W. Zhai, G. Amy, C. Mysore, Bromate Ion Removal by Activated Carbon, Water Res.1996, 30, 1651–1660.
[6] M. Kirisits, V. Snoeyink, J. Kruithof, The Reduction of Bromate by Granular Activated Carbon, Water Res.2000, 34, 4250–4260. [7] M. Prados-Ramirez, N. Ciba, M. Bourbigot, Available Techniques
for Reducing Bromate in Drinking Water, Water Supply1995, 13, 61–68.
[8] T. Marhaba, S. Medlar, Treatment/removal of Bromate and Bromide Ions from Drinking Water, in Proceedings of the AWWA Water Quality Technology Conference, Miami, FL1993.
[9] K. C. Lee, B. E. Rittman, Appliying a Novel Autohydrogenotrophic Hollow-Fiber Membrane Biofilm Reactor for Denitrification of Drinking Water, Water Res.2002, 36, 2040–2052.
[10] R. Sierra-Alvarez, R. Beristain-Cardoso, M. Salazar, J. Gómez, E. Razo-Flores, J. A. Field, Chemolithotrophic Denitrification with Elemental Sulfur for Groundwater Treatment, Water Res.2007, 41, 1253–1262.
[11] A. Assunçao, M. Martins, G. Silva, H. Lucas, M. R. Coelho, M. C. Costa, Bromate Removal by Anaerobic Bacterial Community: Mechanism and Phylogenetic Characterization, J. Hazard. Mater.2011, 197, 237– 243.
[12] E. Sahinkaya, N. Dursun, A. Kilic, S. Demirel, S. Uyanik, O. Cinar, Simultaneous Heterotrophic and Sulfur-Oxidizing Autotrophic Denitrification Process for Drinking Water Treatment: Control of Sulfate Production, Water Res.2011, 45, 6661–6667.
[13] E. J. McAdam, S. J. Judd, Denitrification from Drinking Water Using a Membrane Bioreactor: Chemical and Biochemical Feasibility, Water Res.2007, 41, 4242–4250.
[14] W. A. M. Hijnen, R. Jong, V. D. Kooij, Bromate Removal in a Denitrifying Bioreactor Used in Water Treatment, Water Res.1999, 33, 1049–1053.
[15] S. E. Oh, Y. B. Yoo, J. C. Young, I. S. Kim, Effect of Organics on Sulfur-Utilizing Autotrophic Denitrification under Mixotrophic Condi-tions, J. Biotechnol.2001, 92, 1–8.
[16] R. Butler, A. Godley, L. Lytton, E. Cartmell, Bromate Environmental Contamination: Review of Impact and Possible Treatment, Crit. Rev. Environ. Sci. Technol.2005, 5, 193–217.
[17] W. A. M. Hijnen, R. Voogt, H. Veenendaal, V. D. Jaght, V. D. Kooij, Bromate Reduction by Denitrifying Bacteria, Appl. Environ. Microbiol. 1995, 61, 239–244.
[18] C. T. Matos, S. Velizarov, M. A. M. Reis, J. G. Crespo, Removal of Bromate from Drinking Water Using Ion Exchange Membrane Bioreactor Concept, Environ. Sci. Technol.2008, 42, 7702–7708. [19] E. Sahinkaya, Ö. Çinar, S. Uyanik, S. Demirel, T. E. Ergüder, Performance
Evaluation of Sulfur-Based Autotrophic and Mixotrophic Denitrification Process for the Removal of Nitrate from Ground Water-Effect of Arsenic on Process Performance, TUBITAK Project Code: 110Y256, TUBITAK, Ankara2012. [20] H. S. Moon, D. Y. Shin, K. Nam, J. Y. Kim, A Long Term Performance
Test on an Autotrophic Denitrification Column for Application as a Permeable Reactive Barrier, Chemosphere2008, 73, 723–728. [21] E. Sahinkaya, N. Dursun, Sulfur-Oxidizing Autotrophic and
Mixotrophic Denitrification Processes for Drinking Water Treat-ment: Elimination of Excess Sulfate Production and Alkalinity Requirement, Chemosphere2012, 89, 144–149.
[22] R. Butler, S. Ehrenberg, A. R. Godley, R. Lake, L. Lytton, E. Cartmell, Remediation of Bromate-Contaminated Groundwater in an ex Situ Fixed-Film Biorector, Sci. Total Environ.2006, 366, 12–20.
[23] D. U. Lee, I. S. Lee, Y. D. Chai, J. H. Bae, Effects of External Carbon Source and Empty Bed Contact Time on Simultaneous Heterotrophic and Sulfur-Utilizing Autotrophic Denitrification, Process Biochem. 2001, 36, 1215–1224.
[24] C. G. van Ginkel, A. M. van Haperen, B. van der Togt, Reduction of Bromate to Bromide Coupled to Acetate Oxidation by Anaerobic Mixed Microbial Cultures, Water Res.2005, 39, 59–64.
[25] L. S. Downing, R. Nerenberg, Kinetics of Microbial Bromate Reduction in a Hydrogen-Oxidizing, Denitrifying Biofilm Reactor, Biotechnol. Bioeng.2007, 98 (3), 543–550.
