Published in J. Environ. Qual. 33:2026-2029 (2004).
© ASA, CSSA, SSSA
677 S. Segoe Rd., Madison, WI 53711 USA
TECHNICAL REPORTS
Bioremediation and Biodegradation
Biological Reduction of Chlorate in a Gas-Lift Reactor Using Hydrogen as an Energy Source
A. G. M. Kroon and
C. G. van Ginkel*
Akzo Nobel Chemicals, P.O. Box 9300, 6800 SB Arnhem, the Netherlands
* Corresponding author (kees.vanginkel{at}akzonobel-chemicals.com)
Received for publication February 12, 2004.
 |
ABSTRACT
|
|---|
Chlorate release into the environment occurs with its manufacture and use. Biological reduction of chlorate offers an attractive option to decrease this release. A hydrogen gas-lift reactor with microorganisms attached to pumice particles was used for the treatment of wastewater containing high concentrations of chlorate. The microorganisms used chlorate as an electron acceptor and hydrogen gas as a reducing agent. After a start-up period of only a few weeks, chlorate reduction rates of 3.2 mmol L–1 h–1 were achieved during continuous operation. During this period, a hydrogen consumption rate of 14.5 mmol L–1 h–1 was observed. Complete removal of chlorate was maintained at hydraulic retention times of 6 h. This study clearly demonstrates the potential of hydrogen gas-lift bioreactors for the treatment of chlorate-containing waste streams.
Abbreviations: HRT, hydraulic retention time
|
INTRODUCTION
|
|---|
CHLORATE IS PRODUCED on a large scale by the chemical industry and used in a wide variety of applications. The discharge of chlorate into the environment has increased in the last few decades due to chlorine dioxide–based bleaching of pulp and paper, which may lead to environmental problems. The levels of chlorate may be lowered by microbial reduction to chloride.
Aerobic microorganisms capable of using chlorate as an alternative electron acceptor can be found in fresh water, sludge, and soil (van Ginkel et al., 1995; Coates et al., 1999). The biochemical pathway of these microorganisms appears to involve a chlorate reductase that catalyzes the reduction of chlorate to chlorite. Chlorite is subsequently disproportionated into chloride and oxygen by chlorite dismutase (Rikken et al., 1996; van Ginkel et al., 1996). Investigations on the use of biological treatment to remove chlorate have resulted in economically competitive processes (Detaille et al., 1992; Attaway and Smith, 1993; Malmqvist and Gunnarson, 1993; Kim and Logan, 2001). These processes are based on conversions with microorganisms that use chlorate as an electron acceptor and organic compounds as an electron donor and carbon sources (Korenkov et al., 1976; Stepanyuk et al., 1992; Malmqvist and Welander, 1994; Rikken et al., 1996).
Hydrogen gas may be an attractive alternative electron donor (van Ginkel et al., 1995; Miller and Logan, 2000). Hydrogen as an electron donor has proven its value as a reductant in wastewater and drinking water treatment (Kurt et al., 1987; Dries et al., 1988; Gross et al., 1988; van Houten et al., 1994; du Preez and Maree, 1994; Miller and Logan, 2000). The use of hydrogen gas as reductant in biological wastewater treatment offers several important benefits, such as process reliability, low excess sludge production, and no need for removal of residual reductant (Dries et al., 1988; Gross et al., 1988; Kurt et al., 1987; van Houten et al., 1994). As the solubility of hydrogen in water is very low, good mass transfer from gas to liquid is essential. Efficient mass transfer can be achieved in a gas-lift reactor.
In this paper, we describe the reduction of chlorate to chloride by chlorate-reducing microorganisms attached to pumice particles in a gas-lift reactor supplied with hydrogen as the sole reductant and carbon dioxide as the sole source of carbon. Data on reactor performance and reaction stoichiometry are presented.
|
MATERIALS AND METHODS
|
|---|
Materials
Sodium chlorate (>99% purity) was purchased from Fluka (Buchs, Switzerland). Hydrogen gas was purchased from BOC GAS (Zaventem, Belgium). All other chemicals were of the highest purity available (>98%) and purchased from Acros Chemicals (Beerse, Belgium). Pumice particles (Aqua-Volcano, dp = 0.2–0.6 mm, d = 2.4 kg m–3) used as carrier material were obtained from Aqua-Techniek (Rockanje, the Netherlands).
Inoculum
The inoculum was obtained from an aeration tank of an activated sludge plant treating predominantly domestic wastewater continuously (Duiven, the Netherlands).
Media
The mineral salts medium used contained the following minerals (per liter deionized water): 0.10 g MgSO4·7H2O, 0.50 g (NH4)2HPO4, 3.10 g K2HPO4, 1.70 g NaH2PO4, 1.7 mg Na2SeO3, 0.13 mg NiSO4·6H2O, 0.5 mg resazurin (redox indicator), and 0.1 mL trace element solution (Vishniac and Santer, 1957). The influent of the gas-lift bioreactor was prepared by adding sodium chlorate as the sole electron acceptor to the mineral salts medium.
Gas-Lift Bioreactor
The working volume of the gas-lift bioreactor made of glass was 1.3 L and the volume of the settler was 1.2 L (Applikon, Schiedam, the Netherlands). The inner diameter was 260 mm and the Hliquid was 480 mm. All tubing used for the experimental setup was made of polytetrafluoroethylene (PTFE). The reactor contained 100-g pumice particles as the carrier material. The temperature was maintained at 30°C with a water jacket. The pH was maintained at 7.5 with a 7.5% solution of phosphoric acid, using a pH electrode connected to a pH controller (ADI 1020; Applikon). Hydrogen was sparged into the reactor through a stainless-steel sintered plate (20 µm). The hydrogen gas feed was initially set at 100 mL min–1. The gas was continuously recycled through an external gas loop with a compressor pump (KNF Neuberger, Freiburg, Germany) at a flow rate of 190 L h–1. Hydrogen gas flows were monitored with mass flow controllers (Type 5850S controller, 5860S meter and Flow Computer 405A; Brooks Instrument, Hatfield, PA) connected to a user interface program (Smart Control Series 0160; Brooks Instrument). Mineral salts medium with chlorate sparged with nitrogen gas and activated sludge used as inoculum were supplied to the reactor using peristaltic pumps (Goffin Meyvis, Bergen op Zoom, the Netherlands). Sodium bicarbonate was pumped into the bioreactor at a load of 8.9 g L–1 d–1, using a 5% aqueous solution. The chlorate loading rate and the hydraulic retention time (HRT) were varied during the course of the experiment.
Operation of Gas-Lift Reactor
During start-up the gas-lift bioreactor particles was continuously inoculated with activated sludge (500 mg L–1 dry weight) at a rate of 1 mL h–1. Samples were collected at different time intervals. The gas-lift reactor was operated at steady-state conditions with an HRT of 6 h. The influent chlorate concentration was 4.7 mmol L–1. The carbon recovery was estimated from the amount of carbon dioxide consumed by the bacteria in the reactor and the amount leaving the reactor as off-gas and as inorganic carbon in the effluent. Chlorate and hydrogen utilized by the microorganisms in the reactor were estimated from the amounts supplied and the amounts leaving the reactor in the effluent and off-gas.
Batch Experiments
Batch experiments were conducted in 110-mL gas-tight glass vessels, which were filled to approximately 10% (v/v) with mineral medium containing chlorate. The gas phase and the medium were flushed with nitrogen before inoculation. Hydrogen gas was supplied by injection of a known amount using a gas-tight syringe (up to 150 kPa) at the start of the experiment. Samples were incubated at 30°C in a shaking water bath. Pumice particles with biomass were withdrawn from the bioreactor and washed twice with phosphate buffer (50 mmol L–1, pH 7.2) before use in the batch experiments. The pH optimum was determined in the batch cultures with either phosphate buffer (140 mmol L–1) or sodium borate buffer (24 mmol L–1) to attain the desired pH values. The effect of temperature on the activity was checked in a temperature-controlled shaking water bath (GFL, Burgwedel, Germany). Various salt concentrations in the batch cultures were obtained by adding sodium chloride. All experiments were performed in triplicate.
Analytical Methods
Chlorate and chloride were determined by ion chromatography (Model 120; Dionex, Sunnyvale, CA) using conductivity detection with background suppression. The columns used were AG9-HC (guard) and Ionpac AS-9-HC. The eluent consisted of 9 mmol L–1 Na2CO3 at a rate of 1.0 mL min–1. The injection volume was 50 µL.
The gas chromatograph used to determine hydrogen was an Interscience HR GC 8000 model equipped with a thermal conductivity detector (Carlo Erba, Milan, Italy) at a 300°C detector temperature. The injection (100 µL, split ratio 1:25) temperature was 200°C. Nitrogen was used as the carrier gas at a rate of 50 mL min–1. A Chrompack capillary column (Varian, Palo Alto, CA) with molecular sieve (5 Å, df 30 µm) and a length of 25 m was used.
To estimate the biomass carbon content, biomass was liberated from the pumice particles using ultrasonic treatment at 375 W for 10 min (Vibra Cell; Sonics & Materials, Newtown, CT). After removal of the pumice particles, aqueous samples were acidified to enable removal of carbon dioxide by purging with nitrogen before injection in a total organic carbon (TOC) apparatus (Shimadzu, Kyoto, Japan). Carbon dioxide in the gas phase of the reactor was collected in a 1 M NaOH solution followed by injection into the TOC apparatus for analysis of the total inorganic carbon.
|
RESULTS AND DISCUSSION
|
|---|
Reduction of chlorate during start-up at an HRT of 24 h with a stepwise decrease in HRT after 18 d is shown in Fig. 1
. During start-up of the gas-lift reactor activated sludge was continuously added. After 10 d of operation, biological activity became evident as the chlorate concentration dropped and chloride was detected in the effluent, at which time the inoculation procedure was stopped. The period between the apparent onset of degradation and its completion took another 12 d. The start-up period of approximately 2 to 3 wk was short compared with the results obtained with chlorate reduction by chemoheterotrophic microorganisms in a suspended carrier reactor (Malmqvist and Welander, 1994). In suspended carrier reactors, complete removal of chlorate was only shown after 2 mo of operation.
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 1. Reduction of chlorate ( ) during start-up at a hydraulic retention time (HRT) (—) of 24 h. After 18 d of operation the HRT was decreased stepwise down to 6 h.
|
|
Chlorate reduction was also quantified by the detection of chloride in the effluent. The stoichiometric formation of chloride demonstrates that no accumulation of intermediates occurred. Biomass formation on the pumice particles could not be observed visually during the start-up period of 2 wk; however, the carbon attached to the particles increased from 0.1 to 3.1 mg carbon g–1 pumice particles. Biomass on the pumice particle became visible after 2 mo of operation in which the biomass concentration attached to the particles was 32 mg carbon g–1 pumice particles. Biomass growth in the reactor was consistent with an increase in chlorate-reducing activity.
The HRT was decreased on achieving complete chlorate reduction (Fig. 1). Within 2 wk the HRT was reduced from 24 to 3 h. The stepwise decrease in the HRT down to 6 h did not show any decrease in the degradation efficiency. A further decrease in the HRT to 3 h resulted in a reduction of chlorate removal to approximately 30% (data not shown).
After an operation period of 4 wk, steady-state conditions were assumed in the gas-lift bioreactor maintained at an HRT of 6 h. Following this period of 4 wk, the gas-lift bioreactor achieved chlorate reduction rates of 3.2 mmol L–1 h–1. The microorganisms in the gas-lift reactor consumed hydrogen gas at a rate of 14.5 mmol L–1 h–1. Carbon dioxide (bicarbonate), the only carbon source for the microorganisms, was utilized at a rate of 1.6 mmol L–1 h–1 in the reactor during steady-state conditions. Based on these measured rates and a supposed biomass composition of <CH4O>, the following overall reaction can be formulated:
Figure 2
shows the reduction of chlorate by hydrogen gas in a batch culture with washed pumice particles withdrawn from the gas-lift bioreactor. This batch experiment reveals that the oxidation of hydrogen was coupled to the reduction of chlorate according to the following reaction:
 |
This reaction strongly indicates that the washed cell suspension used hydrogen only as an energy source. Comparing both reactions, it is clear that in the gas-lift bioreactor approximately 30% of the hydrogen gas was used for biomass formation.
The influence of temperature on the chlorate conversion rates was also investigated in batch cultures with washed pumice particles. The optimum temperature curve was found at 36°C, and almost complete inactivation was assessed at 50°C. The optimum process temperature for chlorate reduction was in the same range as found for dissimilatory nitrate reduction (Kurt et al., 1987) and sulfate reduction (Maree and Strydom, 1987) by hydrogenotrophic organisms. High chlorate reduction rates were observed in a pH range of 7 to 8.5, with a maximum value at pH 7.7. No activity was measured below pH 6 and above pH 9. The pH optimum at slightly alkaline pH is common among lithoautotrophic organisms (Focht and Verstraete, 1977).
The effect of salinity in the reactor was also investigated in batch experiments because industrial effluents may contain high salt concentrations. The activity of the microorganisms attached to pumice particles decreased only slightly with increasing salinity. An activity of more than 50% of the maximum was still observed at a concentration of 40 g L–1 sodium chloride. Logan et al. (2001) observed growth of perchlorate-reducing bacteria at sodium chloride concentrations ranging from 1 to 15%. Yang et al. (1995) found that denitrification was almost completely inhibited at a concentration of 30 g L–1 sodium chloride. The limited inhibitory effect of chloride on (per)chlorate-reducing microorganisms may result from acclimatization to chloride, the end product of chlorate reduction.
Finally, the colonized pumice particles were tested for their ability to reduce chlorate after chlorate starvation for several weeks at 4 and 20°C under anaerobic conditions (Fig. 3)
. At the first measurement time of 4 wk, pumice particles kept at 20°C had already lost the majority of their chlorate reduction activity. After 18 wk the activity had decreased below the detection limit of the analytical method. The activity of the pumice particles decreased only slowly when stored at 4°C for a time period of 10 wk. The total activity after 10 wk of starvation had decreased only 35%, and after 18 wk approximately 40% of the activity was still present. Chlorate-reducing microorganisms therefore only survive long periods of chlorate starvation periods at low temperatures.
|
CONCLUSIONS
|
|---|
A microbial population capable of reducing chlorate with hydrogen gas was enriched in a gas-lift reactor. The gas-lift bioreactor readily demonstrated a capacity for chlorate removal with consistent concomitant chloride formation. Complete removal of chlorate present at a concentration of 4.7 mmol L–1 in artificial wastewater can be maintained at hydraulic retention times of 6 h. The chlorate conversion rate at an HRT of 6 h was 3.2 mmol L–1 h–1, and the hydrogen consumption rate derived from this was 14.5 mmol L–1 h–1. Batch experiments showed that biological chlorate reduction is possible within a pH range of 7 to 8.5 and temperatures up to 35°C. The effects of increased salinity are very minor. In conclusion, the ability of microorganisms to reduce chlorate with hydrogen gas demonstrates that chlorate-contaminated water can be treated successfully in gas-lift bioreactors.
|
REFERENCES
|
|---|
- Attaway, H., and M. Smith. 1993. Reduction of perchlorate by an anaerobic enrichment culture. J. Ind. Microbiol. 12:408–412.
- Coates, J.D., U. Michaelidou, S. O'Connor, R.A. Bruce, J.N. Crespi, and L.A. Achenbach. 1999. The ubiquity and diversity of dissimilatory (per)chlorate-reducing bacteria. Appl. Environ. Microbiol. 65:5234–5241.[Abstract/Free Full Text]
- Detaille, R., S. Deswaef, M. Schlitz, and M. Crine. 1992. Microbial removal of chlorate in an upflow fixed bed reactor. Meded. Fac. Landbouwwet. Univ. Gent 57(4A):1701–1703.
- Dries, D., J. Liessens, W. Verstraete, R. Stevens, P. de Vos, and J. de Ley. 1988. Nitrate removal from drinking water by means of hydrogenotrophic denitrifiers in a polyurethane carrier reactor. Water Supply 6:181–192.
- Du Preez, L.A., and J.P. Maree. 1994. Pilot-scale biological sulphate and nitrate removal utilizing producer gas as energy source. Water Sci. Technol. 30:275–285.
- Focht, D.D., and W. Verstraete. 1977. Biochemical ecology of nitrification and denitrification. Adv. Microb. Ecol. 1:135–214.
- Gross, H., G. Schnoor, and P. Rutten. 1988. Biological denitrification process with hydrogen-oxidizing bacteria for drinking water treatment. Water Supply 6:193–198.
- Kim, K., and B.E. Logan. 2001. Microbial reduction of perchlorate in pure and mixed culture packed bed bioreactors. Water Res. 13:2071–2076.
- Korenkov, V.N., V.I. Romanenko, S.I. Kuznetsov, and J.V. Voronov. 1976. Process for purification of industrial wastewaters from perchlorates and chlorates. U.S. Patent 3 943 055. Date issued: 9 March.
- Kurt, M., I.J. Dunn, and J.R. Bourne. 1987. Biological denitrification of drinking water using autotrophic organisms with hydrogen in a fluidized-bed biofilm reactor. Biotechnol. Bioeng. 29:493–501.
- Logan, B.E., J. Wu, and R.F. Unz. 2001. Biological perchlorate reduction in high-salinity solutions. Water Res. 35:3034–3038.[Medline]
- Malmqvist, A., and L. Gunnarson. 1993. Biological removal of chlorate from kraft bleach plant effluent. Nord. Pulp Pap. Res. J. 8:302–306.
- Malmqvist, A., and T. Welander. 1994. Biological removal of chlorate from bleach plant effluent. Water Sci. Technol. 29:365–372.
- Maree, J.P., and W.F. Strydom. 1987. Biological sulphate removal from industrial effluent in an upflow packed bed reactor. Water Res. 21:141–146.
- Miller, J.P., and B.E. Logan. 2000. Sustained perchlorate degradation in an autotrophic, gas phase, packed bed bioreactor. Environ. Sci. Technol. 34:3018–3022.
- Rikken, G.B., A.G.M. Kroon, and C.G. van Ginkel. 1996. Transformation of (per)chlorate into chloride by a newly isolated bacterium: Reduction and dismutation. Appl. Microbiol. Biotechnol. 45:420–426.
- Stepanyuk, V.V., G.F. Smirnova, T.M. Klyushnikova, N.I. Kanyuk, L.P. Panchenko, T.M. Nogina, and V.I. Prima. 1992. New species of the Acinetobacter genus Acinetobacter thermotoleranticus sp. nov. Mikrobiologiya 61:490–500.
- Van Ginkel, C.G., C.M. Plugge, and C.A. Stroo. 1995. Reduction of chlorate with various energy substrates and inocula under anaerobic conditions. Chemosphere 31:4057–4066.
- Van Ginkel, C.G., G.B. Rikken, A.G.M. Kroon, and S.W.M. Kengen. 1996. Purification and characterization of chlorite dismutase: A novel oxygen-generating enzyme. Arch. Microbiol. 166:321–326.[Medline]
- Van Houten, R.T., L.W. Hulshoff Pol, and G.L. Lettinga. 1994. Biological sulphate reduction using gas-lift reactors fed with hydrogen and carbon dioxide as energy and carbon source. Biotechnol. Bioeng. 44:586–594.
- Vishniac, W., and M. Santer. 1957. The thiobacilli. Bacteriol. Rev. 21:195–213.[Free Full Text]
- Yang, P.Y., S. Nitisoravut, and J.Y.S. Wu. 1995. Nitrate removal using a mixed-culture entrapped microbial cell immobilization process under high salt conditions. Water Res. 29:1525–1532.
Related articles in JEQ:
- This Issue in Journal of Environmental Quality
JEQ 2004 33: 1947-1953.
[Full Text]
TOP
ABSTRACT
INTRODUCTION
) and the stoichiometric formation of chloride (
) as the sole energy source. Each point represents the mean (±SD) of three replicates.