- Email: [email protected]
Denitrification of groundwater with elemental sulfur
Water Research 36 (2002) 1392–1395
Technical note
Denitrification of groundwater with elemental sulfur M.I.M. Soares* Department of Environmental Hydrology and Microbiology, The J. Blaustein Institute for Desert Research, Ben-Gurion University of the Negev, Sede Boqer Campus, Sede Boqer 84990, Israel Received 1 November 1999; received in revised form 1 April 2001; accepted 29 June 2001
Abstract Autotrophic denitrification was studied in laboratory columns packed with granular elemental sulfur only and operated in an upflow mode. Soluble inorganic carbon, sodium bicarbonate, was supplied as source of carbon for microbial growth. Denitrification rates of up to 0.20 kg N removed m3 d1 were obtained at a hydraulic retention time of 1 h, and a nitrate loading of 0.24 kg N m3 d1. The process is extremely simple, stable and easy to maintain. r 2002 Elsevier Science Ltd. All rights reserved. Keywords: Biological denitrification; Drinking water; Water treatment; Elemental sulfur
1. Introduction Due to its low cost, elemental sulfur is an attractive source of energy for biological denitrification of nitratecontaminated groundwater. Furthermore, elemental sulfur is non-toxic, water insoluble, stable under normal conditions, and readily available as a mined mineral or as a byproduct of fuel processing and SO2 control systems such as power plants burning coal. Among denitrifying microorganisms, only very few species of autotrophic bacteria can carry out Sdependent denitrification, and Thiobacillus denitrificans is the most commonly described. The overall reaction for S-dependent denitrification can be summarized as: 2 þ 5S0 þ6NO 3 þ2H2 O-5SO4 þ3N2 þ4H
ð1Þ
and including the production of biomass þ 55S0 þ50NO 3 þ38H2 O þ 20CO2 þ4NH4 þ -4C5 H7 O2 N þ 55SO2 4 þ25N2 þ64H
ð2Þ
In the process, elemental sulfur is converted into sulfate, and this renders the method unsuitable for the treatment of drinking water containing high levels of *Tel.: +972-8659-6834; fax: +972-8659-6831. E-mail address: [email protected] (M.I.M. Soares).
endogenous sulfate. High concentrations of sulfate can act as laxative, especially in combination with magnesium. Sulfate may also affect the taste of water with the effective threshold being different for different salts. A maximum of 400 mg sulfate l1 is therefore recommended [1]. Early reports on the use of elemental sulfur in water denitrification were confined to wastewater [2–5]. Later, denitrification of groundwater was studied in laboratory columns packed with a granular mixture of elemental sulfur and limestone (Le Cloirec and Martin, 1988 [18]; Blecon and Martin [6]). Limestone served as the source of inorganic carbon for bacterial synthesis and as pH buffering agent. The reactors were operated in an upflow mode with periodic backwashings to remove accumulated biomass. The sulfur-limestone process was further developed by van der [7] in a field system without backwashing, and with a capacity 35 m3 h1. It consisted of four unit operations in series: vacuum-deaeration, nitrate removal in a sulfur-limestone filter, aeration (cascade) and soil infiltration as post-treatment. The upflow filtration rate was between 0.25 and 0.50 m h1 and loadings of 0.29 and 0.55 kg N m3 d1 were applied. The means by which S0 is transformed into SO2 4 are still not known [8]. In view of the low water solubility
0043-1354/02/$ -see front matter r 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 0 4 3 - 1 3 5 4 ( 0 1 ) 0 0 3 2 6 - 8
M.I.M. Soares / Water Research 36 (2002) 1392–1395
and hydrophobicity of elemental sulfur it is likely that surface effects play a role in its biological oxidation, and surface-active agents have been found in cultures of some thiobacilli [9,10]. Close contact between sulfur and bacteria is necessary and sulfur-binding proteins have also been found [11]. Since oxidation requires the adhesion of bacteria to elemental sulfur, it would appear that in a fixed-bed denitrification reactor (1) the sulfur surface area should be maximized, and (2) the role of limestone particles as physical support for the denitrifying bacteria would be secondary. Thus, the feasibility of a denitrification system consisting of a reactor packed with granular sulfur only, and fed with soluble inorganic carbon (sodium bicarbonate) was tested in this laboratory study.
2. Materials and methods 2.1. Experimental set-up The laboratory set-up is schematically represented in Fig. 1. The denitrification reactor was a glass column packed with sulfur ‘‘lentils’’ (particles with the approximate size and shape of lentils) which were pre-washed to remove any powdered sulfur; a thin layer of glass wool was placed at each end of the bed. The main characteristics of the bed at the time of packing are presented on Table 1. The column was fed in an upflow mode, and a water velocity (v) of 0.1 m h1 corresponded to a feed rate of 2.65 ml min1. The feed solution was gently sparged with N2 and consisted of tap water enriched with nitrate, sodium bicarbonate and phosphate (Table 1). The column was kept in the dark, at the temperature of 2471C1. The reactor was inoculated with slurry obtained by mixing in an electric blender sulfur pellets from a previous denitrification column with a small volume of
Fig. 1. Schematic representation of the experimental apparatus. P, pump.
1393
Table 1 Characteristics of the denitrification reactor Bed
Height (cm) Diameter (cm) Volume (cm3) Void volume (cm3) Porosity (%) Sulfur dry weight (g)
43 4.5 683.7 296 43.3 777
Feed solution
Nitrate (mg l1) Sodium bicarbonate (mg l1) Phosphate (mg l1)
100 600 3
tap water. The original inoculum was an enrichment culture from an oxidation pond sediment incubated in an anaerobic chamber in the medium described by [12] containing (g l1): Na2S2O3.5H2O, 5.0; KH2PO4, 2.0; KNO3, 2.0; Mg SO4.7H2O, 0.6; NaHCO3, 1.5; and 1 ml trace metal solution. Prior to start-up, the biofilm was allowed to develop for four weeks in a recirculation mode, with adequate replenishments of nitrate, bicarbonate and phosphate. 2.2. Analytical methods Nitrate was determined by the colorimetric method of [13]. Nitrite, ammonia and sulfide were assayed according to [14], and sulfate was determined by the turbidimetric method of [15]. Inorganic carbon (IC) and dissolved organic carbon (DOC) were determined by means of a Dohrmann DC-190 (Rosemount Analytical Inc., Santa Clara, CA, USA) high-temperature TOC analyzer.
3. Results and discussion The reactor was operated for five months during which varying water velocities were tested. Complete removal of nitrogen was observed at the start-up v of 0.2 m h1 (Fig. 2). The water velocity was then gradually increased up to 0.19 m h1 (days 20–54), corresponding to a hydraulic retention time of 1 h. At this highest v; breakthrough of nitrate and nitrite occurred, reaching concentrations of 6.97 and 4.47 mg l1 respectively. Lowering v to 0.13 m h1 brought about immediate decline in the concentration of nitrate (Fig. 2, days 55– 72). Breakthrough of nitrite persisted and declined only during the period at v 0.11 m h1 (after day 77). From then on, water velocities between 0.11 and 0.08 m h1 had little effect on the effluent concentrations of nitrate and nitrite. Ammonia was never detected. The highest rate of nitrogen removal was 0.20 kg N m3 d1, obtained at the highest water velocity tested,
1394
M.I.M. Soares / Water Research 36 (2002) 1392–1395
Fig. 2. Influent concentrations of nitrate and effluent concentrations of nitrate and nitrite in a denitrification column operated at various water velocities.
0.19 m h1, and at a nitrate loading of 0.24 kg N m3 d1 (Fig. 3). The overall effect of sulfur-dependent denitrification on the water pH was negligible (Fig. 4). The amount of inorganic carbon used in the process was small and nearly constant throughout the experiment (Fig. 4). The concentrations of DOC in the effluent were up to 3 mg l1 higher than those in the influent (data not shown). Sulfate concentrations in the influent varied from 50 to 80 mg l1, and increased up to 320 mg l1 in the effluent. Thus, the concentration of sulfate in the treated water was always below the recommended level for drinking water [1]. The calculated ratios of sulfate formed to nitrate removed (Fig. 4) varied around the theoretical ratio of 1.7 (Eq. (2)). Sulfide was never detected in the effluent. Van der [7] observed that in sections of the reactor where complete
Fig. 3. Nitrogen loadings and rates of nitrogen removal in the reactor in Fig. 2.
Fig. 4. Influent and effluent pH and concentrations of inorganic carbon, and variations on the ratio of sulfate formed to nitrate removed in the reactor in Fig. 2.
removal of nitrate and nitrite had been achieved, reduction of elemental sulfur by heterotrophic anaerobes could occur. Although sulfide was readily oxidized into colloidal sulfur upon exposure of the effluent to aerobic conditions, the formation of sulfide should be prevented as the sulfur particles caused clogging of the infiltration pond. After 6 months (one-month recirculation and 5 months operation) the dry weight of the sulfur bed had decreased from the original 777 g to 723 g, representing a loss of 7%. Signs of clogging by entrapped gases, as reported for finer matrixes [16,17] were never detected.
4. Conclusions 1. The denitrification system presented here is extremely simple, stable and easy to maintain. 2. Reactors are packed with the energy source only, which is completely degradable, and no extra volume is required to accommodate a bulky carbon substrate. The surface area offered by limestone would not appear to be important, since oxidation of elemental sulfur is the rate-limiting step in the denitrification treatment and requires the direct contact of bacteria. 3. Dosage of soluble carbon source at the inlet is flexible and can be instantly adjusted according to the needs of the reactor (e.g., buildup of biomass during the start up phase, steady-state operation). Addition of phosphate is routinely carried out in denitrification plants, and carbon can be conveniently supplied at the same time.
M.I.M. Soares / Water Research 36 (2002) 1392–1395
Acknowledgements This research was supported by a grant from the Israeli Water Commission. Sulfur ‘‘lentils’’ were donated by Haifa Chemicals, Israel.
References [1] WHO, Guidelines for drinking water quality. Vol. I, Recommendations. World Health Organization, Geneva, 1984. [2] Batchelor B. Autotrophic denitrification using sulfur electron donors. PhD Thesis, Cornell University, 1976. [3] Batchelor B, Lawrence AW. Autotrophic denitrification using elemental sulfur. J Water Pollut Control Fed 1978;50:1986–2001. [4] Driscoll CT, Bisogni JJ. The use of sulfur and sulfide in packed bed reactors for autotrophic denitrification. J Water Pollut Control Fed 1978;50:569–77. [5] Hashimoto S, Furukawa K, Shioyama M. Autotrophic denitrification using elemental sulfur. J Ferment Technol 1987;65:683–92. [6] Blecon G, Martin G. Conditions de la denitrification autotrophique par Thiobacillus denitrificans sur soufre en traitement d’eau sur lit fixe. Proceedings of the Conference Nitrates in Water, Paris, 22–4, 1985. [7] Hoek JP, van der, Hijnen WAM, van Bennekom CA, Mijnarends BJ. Optimization of the sulphur-limestone filtration process for nitrate removal from groundwater. J Water SRT-Aqua 1992;41:209–18. [8] Kelly DP, Jasvinder KS, Lu W-P, Wood AP. Oxidative metabolism of inorganic sulfur compounds by bacteria. Antonie van Leeuwenhoek 1997;71:95–107.
1395
[9] Cook TM. Growth of Thiobacillus thiooxidans in shaken cultures. J Bacteriol 1964;88:620–3. [10] Agate AD, Korezynski MS, Lundgren DG. Extracellular complex from the culture filtrate of Ferrobacillus ferrooxidans. Can J Microbiol 1969;15:259–64. [11] Ohmura N, Tsugita K, Koizumi J-I, Saiki H. Sulfurbinding protein of flagella of Thiobacillus ferrooxidans. J Bacteriol 1996;178:5776–80. [12] Baldensperger J, Garcia J-L. Reduction of oxidized inorganic compounds by a new strain of Thiobacillus denitrificans. Arch Microbiol 1975;103:31–6. [13] Cataldo DA, Haroon M, Schrader LE, Youngs VL. Rapid colorimetric determination of nitrate in plant tissue by nitration of salicylic acid. Commun Soil Sci Plant Anal 1975;6:71–80. [14] Standard Methods for the Examination of Water and Wastewater. (1992). American Public Health Association/American Water Works Association/Water Environmental Federation, 18th edition, Washington, DC, USA. [15] Sorbo B. Sulfate: turbidimetric and nephelometric methods. Methods Enzmol 1987;143:3–6. [16] Soares MIM, Belkin S, Abeliovich A. Clogging of microbial denitrification sand columns: gas bubbles or biomass accumulation? J Water Wastewater Res 1989; 22:20–4. [17] Soares MIM, Belkin S, Abeliovich A. Denitrification in laboratory sand columns: carbon regime, gas accumulation and hydraulic properties. Water Res 1991;25: 325–32. [18] Le Cloirec P, Martin G. Simulation of the biological denitrification in a sulfur-calcium carbonate reactor. Environ Technol Lett 1988;9:207–18.
Technical note
Denitrification of groundwater with elemental sulfur M.I.M. Soares* Department of Environmental Hydrology and Microbiology, The J. Blaustein Institute for Desert Research, Ben-Gurion University of the Negev, Sede Boqer Campus, Sede Boqer 84990, Israel Received 1 November 1999; received in revised form 1 April 2001; accepted 29 June 2001
Abstract Autotrophic denitrification was studied in laboratory columns packed with granular elemental sulfur only and operated in an upflow mode. Soluble inorganic carbon, sodium bicarbonate, was supplied as source of carbon for microbial growth. Denitrification rates of up to 0.20 kg N removed m3 d1 were obtained at a hydraulic retention time of 1 h, and a nitrate loading of 0.24 kg N m3 d1. The process is extremely simple, stable and easy to maintain. r 2002 Elsevier Science Ltd. All rights reserved. Keywords: Biological denitrification; Drinking water; Water treatment; Elemental sulfur
1. Introduction Due to its low cost, elemental sulfur is an attractive source of energy for biological denitrification of nitratecontaminated groundwater. Furthermore, elemental sulfur is non-toxic, water insoluble, stable under normal conditions, and readily available as a mined mineral or as a byproduct of fuel processing and SO2 control systems such as power plants burning coal. Among denitrifying microorganisms, only very few species of autotrophic bacteria can carry out Sdependent denitrification, and Thiobacillus denitrificans is the most commonly described. The overall reaction for S-dependent denitrification can be summarized as: 2 þ 5S0 þ6NO 3 þ2H2 O-5SO4 þ3N2 þ4H
ð1Þ
and including the production of biomass þ 55S0 þ50NO 3 þ38H2 O þ 20CO2 þ4NH4 þ -4C5 H7 O2 N þ 55SO2 4 þ25N2 þ64H
ð2Þ
In the process, elemental sulfur is converted into sulfate, and this renders the method unsuitable for the treatment of drinking water containing high levels of *Tel.: +972-8659-6834; fax: +972-8659-6831. E-mail address: [email protected] (M.I.M. Soares).
endogenous sulfate. High concentrations of sulfate can act as laxative, especially in combination with magnesium. Sulfate may also affect the taste of water with the effective threshold being different for different salts. A maximum of 400 mg sulfate l1 is therefore recommended [1]. Early reports on the use of elemental sulfur in water denitrification were confined to wastewater [2–5]. Later, denitrification of groundwater was studied in laboratory columns packed with a granular mixture of elemental sulfur and limestone (Le Cloirec and Martin, 1988 [18]; Blecon and Martin [6]). Limestone served as the source of inorganic carbon for bacterial synthesis and as pH buffering agent. The reactors were operated in an upflow mode with periodic backwashings to remove accumulated biomass. The sulfur-limestone process was further developed by van der [7] in a field system without backwashing, and with a capacity 35 m3 h1. It consisted of four unit operations in series: vacuum-deaeration, nitrate removal in a sulfur-limestone filter, aeration (cascade) and soil infiltration as post-treatment. The upflow filtration rate was between 0.25 and 0.50 m h1 and loadings of 0.29 and 0.55 kg N m3 d1 were applied. The means by which S0 is transformed into SO2 4 are still not known [8]. In view of the low water solubility
0043-1354/02/$ -see front matter r 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 0 4 3 - 1 3 5 4 ( 0 1 ) 0 0 3 2 6 - 8
M.I.M. Soares / Water Research 36 (2002) 1392–1395
and hydrophobicity of elemental sulfur it is likely that surface effects play a role in its biological oxidation, and surface-active agents have been found in cultures of some thiobacilli [9,10]. Close contact between sulfur and bacteria is necessary and sulfur-binding proteins have also been found [11]. Since oxidation requires the adhesion of bacteria to elemental sulfur, it would appear that in a fixed-bed denitrification reactor (1) the sulfur surface area should be maximized, and (2) the role of limestone particles as physical support for the denitrifying bacteria would be secondary. Thus, the feasibility of a denitrification system consisting of a reactor packed with granular sulfur only, and fed with soluble inorganic carbon (sodium bicarbonate) was tested in this laboratory study.
2. Materials and methods 2.1. Experimental set-up The laboratory set-up is schematically represented in Fig. 1. The denitrification reactor was a glass column packed with sulfur ‘‘lentils’’ (particles with the approximate size and shape of lentils) which were pre-washed to remove any powdered sulfur; a thin layer of glass wool was placed at each end of the bed. The main characteristics of the bed at the time of packing are presented on Table 1. The column was fed in an upflow mode, and a water velocity (v) of 0.1 m h1 corresponded to a feed rate of 2.65 ml min1. The feed solution was gently sparged with N2 and consisted of tap water enriched with nitrate, sodium bicarbonate and phosphate (Table 1). The column was kept in the dark, at the temperature of 2471C1. The reactor was inoculated with slurry obtained by mixing in an electric blender sulfur pellets from a previous denitrification column with a small volume of
Fig. 1. Schematic representation of the experimental apparatus. P, pump.
1393
Table 1 Characteristics of the denitrification reactor Bed
Height (cm) Diameter (cm) Volume (cm3) Void volume (cm3) Porosity (%) Sulfur dry weight (g)
43 4.5 683.7 296 43.3 777
Feed solution
Nitrate (mg l1) Sodium bicarbonate (mg l1) Phosphate (mg l1)
100 600 3
tap water. The original inoculum was an enrichment culture from an oxidation pond sediment incubated in an anaerobic chamber in the medium described by [12] containing (g l1): Na2S2O3.5H2O, 5.0; KH2PO4, 2.0; KNO3, 2.0; Mg SO4.7H2O, 0.6; NaHCO3, 1.5; and 1 ml trace metal solution. Prior to start-up, the biofilm was allowed to develop for four weeks in a recirculation mode, with adequate replenishments of nitrate, bicarbonate and phosphate. 2.2. Analytical methods Nitrate was determined by the colorimetric method of [13]. Nitrite, ammonia and sulfide were assayed according to [14], and sulfate was determined by the turbidimetric method of [15]. Inorganic carbon (IC) and dissolved organic carbon (DOC) were determined by means of a Dohrmann DC-190 (Rosemount Analytical Inc., Santa Clara, CA, USA) high-temperature TOC analyzer.
3. Results and discussion The reactor was operated for five months during which varying water velocities were tested. Complete removal of nitrogen was observed at the start-up v of 0.2 m h1 (Fig. 2). The water velocity was then gradually increased up to 0.19 m h1 (days 20–54), corresponding to a hydraulic retention time of 1 h. At this highest v; breakthrough of nitrate and nitrite occurred, reaching concentrations of 6.97 and 4.47 mg l1 respectively. Lowering v to 0.13 m h1 brought about immediate decline in the concentration of nitrate (Fig. 2, days 55– 72). Breakthrough of nitrite persisted and declined only during the period at v 0.11 m h1 (after day 77). From then on, water velocities between 0.11 and 0.08 m h1 had little effect on the effluent concentrations of nitrate and nitrite. Ammonia was never detected. The highest rate of nitrogen removal was 0.20 kg N m3 d1, obtained at the highest water velocity tested,
1394
M.I.M. Soares / Water Research 36 (2002) 1392–1395
Fig. 2. Influent concentrations of nitrate and effluent concentrations of nitrate and nitrite in a denitrification column operated at various water velocities.
0.19 m h1, and at a nitrate loading of 0.24 kg N m3 d1 (Fig. 3). The overall effect of sulfur-dependent denitrification on the water pH was negligible (Fig. 4). The amount of inorganic carbon used in the process was small and nearly constant throughout the experiment (Fig. 4). The concentrations of DOC in the effluent were up to 3 mg l1 higher than those in the influent (data not shown). Sulfate concentrations in the influent varied from 50 to 80 mg l1, and increased up to 320 mg l1 in the effluent. Thus, the concentration of sulfate in the treated water was always below the recommended level for drinking water [1]. The calculated ratios of sulfate formed to nitrate removed (Fig. 4) varied around the theoretical ratio of 1.7 (Eq. (2)). Sulfide was never detected in the effluent. Van der [7] observed that in sections of the reactor where complete
Fig. 3. Nitrogen loadings and rates of nitrogen removal in the reactor in Fig. 2.
Fig. 4. Influent and effluent pH and concentrations of inorganic carbon, and variations on the ratio of sulfate formed to nitrate removed in the reactor in Fig. 2.
removal of nitrate and nitrite had been achieved, reduction of elemental sulfur by heterotrophic anaerobes could occur. Although sulfide was readily oxidized into colloidal sulfur upon exposure of the effluent to aerobic conditions, the formation of sulfide should be prevented as the sulfur particles caused clogging of the infiltration pond. After 6 months (one-month recirculation and 5 months operation) the dry weight of the sulfur bed had decreased from the original 777 g to 723 g, representing a loss of 7%. Signs of clogging by entrapped gases, as reported for finer matrixes [16,17] were never detected.
4. Conclusions 1. The denitrification system presented here is extremely simple, stable and easy to maintain. 2. Reactors are packed with the energy source only, which is completely degradable, and no extra volume is required to accommodate a bulky carbon substrate. The surface area offered by limestone would not appear to be important, since oxidation of elemental sulfur is the rate-limiting step in the denitrification treatment and requires the direct contact of bacteria. 3. Dosage of soluble carbon source at the inlet is flexible and can be instantly adjusted according to the needs of the reactor (e.g., buildup of biomass during the start up phase, steady-state operation). Addition of phosphate is routinely carried out in denitrification plants, and carbon can be conveniently supplied at the same time.
M.I.M. Soares / Water Research 36 (2002) 1392–1395
Acknowledgements This research was supported by a grant from the Israeli Water Commission. Sulfur ‘‘lentils’’ were donated by Haifa Chemicals, Israel.
References [1] WHO, Guidelines for drinking water quality. Vol. I, Recommendations. World Health Organization, Geneva, 1984. [2] Batchelor B. Autotrophic denitrification using sulfur electron donors. PhD Thesis, Cornell University, 1976. [3] Batchelor B, Lawrence AW. Autotrophic denitrification using elemental sulfur. J Water Pollut Control Fed 1978;50:1986–2001. [4] Driscoll CT, Bisogni JJ. The use of sulfur and sulfide in packed bed reactors for autotrophic denitrification. J Water Pollut Control Fed 1978;50:569–77. [5] Hashimoto S, Furukawa K, Shioyama M. Autotrophic denitrification using elemental sulfur. J Ferment Technol 1987;65:683–92. [6] Blecon G, Martin G. Conditions de la denitrification autotrophique par Thiobacillus denitrificans sur soufre en traitement d’eau sur lit fixe. Proceedings of the Conference Nitrates in Water, Paris, 22–4, 1985. [7] Hoek JP, van der, Hijnen WAM, van Bennekom CA, Mijnarends BJ. Optimization of the sulphur-limestone filtration process for nitrate removal from groundwater. J Water SRT-Aqua 1992;41:209–18. [8] Kelly DP, Jasvinder KS, Lu W-P, Wood AP. Oxidative metabolism of inorganic sulfur compounds by bacteria. Antonie van Leeuwenhoek 1997;71:95–107.
1395
[9] Cook TM. Growth of Thiobacillus thiooxidans in shaken cultures. J Bacteriol 1964;88:620–3. [10] Agate AD, Korezynski MS, Lundgren DG. Extracellular complex from the culture filtrate of Ferrobacillus ferrooxidans. Can J Microbiol 1969;15:259–64. [11] Ohmura N, Tsugita K, Koizumi J-I, Saiki H. Sulfurbinding protein of flagella of Thiobacillus ferrooxidans. J Bacteriol 1996;178:5776–80. [12] Baldensperger J, Garcia J-L. Reduction of oxidized inorganic compounds by a new strain of Thiobacillus denitrificans. Arch Microbiol 1975;103:31–6. [13] Cataldo DA, Haroon M, Schrader LE, Youngs VL. Rapid colorimetric determination of nitrate in plant tissue by nitration of salicylic acid. Commun Soil Sci Plant Anal 1975;6:71–80. [14] Standard Methods for the Examination of Water and Wastewater. (1992). American Public Health Association/American Water Works Association/Water Environmental Federation, 18th edition, Washington, DC, USA. [15] Sorbo B. Sulfate: turbidimetric and nephelometric methods. Methods Enzmol 1987;143:3–6. [16] Soares MIM, Belkin S, Abeliovich A. Clogging of microbial denitrification sand columns: gas bubbles or biomass accumulation? J Water Wastewater Res 1989; 22:20–4. [17] Soares MIM, Belkin S, Abeliovich A. Denitrification in laboratory sand columns: carbon regime, gas accumulation and hydraulic properties. Water Res 1991;25: 325–32. [18] Le Cloirec P, Martin G. Simulation of the biological denitrification in a sulfur-calcium carbonate reactor. Environ Technol Lett 1988;9:207–18.