JEQ Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Related articles in JEQ
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via ISI Web of Science (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zhang, Y.
Right arrow Articles by Frankenberger, W. T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zhang, Y.
Right arrow Articles by Frankenberger, W. T., Jr.
Agricola
Right arrow Articles by Zhang, Y.
Right arrow Articles by Frankenberger, W. T.
Related Collections
Right arrow Water Quality
Right arrow Batch Studies
Right arrow Remediation
Right arrow Heavy Metals
Right arrow Water Pollution
Published in J. Environ. Qual. 34:487-495 (2005).
© ASA, CSSA, SSSA
677 S. Segoe Rd., Madison, WI 53711 USA

TECHNICAL REPORTS

Heavy Metals in the Environment

Removal of Selenate from Water by Zerovalent Iron

Yiqiang Zhang, Juanfang Wang, Chris Amrhein and William T. Frankenberger, Jr.*

Department of Environmental Sciences, University of California, Riverside, CA 92521-0424

* Corresponding author (william.frankenberger{at}ucr.edu)

Received for publication May 25, 2004.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Zerovalent iron (ZVI) has been widely used in the removal of environmental contaminants from water. In this study, ZVI was used to remove selenate [Se(VI)] at a level of 1000 µg L–1 in the presence of varying concentrations of Cl–, SO2–4, NO–3, HCO–3, and PO3–4. Results showed that Se(VI) was rapidly removed during the corrosion of ZVI to iron oxyhydroxides (FeOH). During the 16 h of the experiments, 100 and 56% of the added Se(VI) was removed in 10 mM Cl– and SO2–4 solutions under a closed contained system, respectively. Under an open condition, 100 and 93% of the added Se(VI) were removed in the Cl– and SO2–4 solutions, respectively. Analysis of Se species in ZVI–FeOH revealed that selenite [Se(IV)] and nonextractable Se increased during the first 2 to 4 h of reaction, with a decrease of Se(VI) in the Cl– experiment and no detection of Se(VI) in the SO2–4 experiment. Two mechanisms can be attributed to the rapid removal of Se(VI) from the solutions. One is the reduction of Se(VI) to Se(IV), followed by rapid adsorption of Se(IV) to FeOH. The other is the adsorption of Se(VI) directly to FeOH, followed by its reduction to Se(IV). The results also show that there was little effect on Se(VI) removal in the presence of Cl– (5, 50, and 100 mM), NO–3 (1, 5, and 10 mM), SO2–4 (5 mM), HCO–3 (1 and 5 mM), or PO3–4 (1 mM) and only a slight effect in the presence of SO2–4 (50 and 100 mM), HCO–3 (10 mM), and PO3–4 (5 mM) during a 2-d experiment, whereas 10 mM PO3–4 significantly inhibited Se(VI) removal. This work suggests that ZVI may be an effective agent to remove Se from Se-contaminated agricultural drainage water.

Abbreviations: FeOH, iron oxyhydroxides • Se(0), elemental selenium • Se(IV), selenite • Se(VI), selenate • ZVI, zerovalent iron


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
ELEVATED SELENIUM (Se) concentrations in agricultural drainage water have been found in many sites of California. In the San Joaquin Valley, Se in the drainage water is frequently found at a concentration level of 140 to 1400 µg L–1 (Amweg et al., 2003; Cantafio et al., 1996; Sylvester, 1990, p. 119–124). In the Salton Sea region, elevated Se is in the range of 3 to 300 µg L–1 in the subsurface drainwater (Setmire and Schroeder, 1998). In an effort to minimize environmental impacts of Se, the State of California Water Resources Control Board (1989) has recommended an interim maximum mean monthly Se concentration of 2 to 5 µg L–1 in rivers and wetlands receiving agricultural drainage water. So far, no treatment technology has proven economically feasible for meeting this criterion.

Several biotreatment systems have been used to treat Se-contaminated agricultural drainage water in the San Joaquin Valley. In a pilot-scale Se bioremediation system using a Se(VI) reducer, Thauera selenatis, and acetate in liquid phase as an electron donor, Cantafio et al. (1996) reported that bacterial reduction of Se(VI) to elemental selenium [Se(0)] proceeded rapidly in a series of four columns filled with Jaeger (Houston, TX) Tri-Packs and/or silica sand. About 98% of Se(VI) and Se(IV) in agricultural drainage water was reduced. However, high costs make it less feasible to use acetate as an electron donor and carbon source for bacteria to reduce Se(VI) to Se(0) during full-scale operation in the field.

By using economical organic carbon sources, an algal–bacterial selenium removal system (ABSRS) has been used as an economic way to remove large amounts of Se from drainage water (Lundquist et al., 1995). However, increase in concentration of the most bioavailable organic Se in the treated water creates greater bioavailability of Se to biota than that in the influent (drainage water) (Amweg et al., 2003), which could propose greater toxicological risk to biota if the treated water flows into nearby wetlands. Therefore, there is a need for alternative techniques to remove Se from drainage water.

Zerovalent iron (ZVI) is an inexpensive and moderately strong reducing agent (Genin et al., 1998). It is used as a catalyst for chemical synthesis in industrial applications (Campbell, 1988) and is capable of removing many common environmental contaminants, such as Cr(VI), U(VI), and NO–3 (Alowitz and Scherer, 2002; Farrell et al., 1999; Huang et al., 1998; Powell et al., 1995; Qiu et al., 2000). The corrosion of ZVI is an electrochemical process during which iron is oxidized to soluble Fe2+. Reaction of Fe2+ with OH– forms Fe(OH)2, which can be further oxidized to green rust I [Fe3(II)Fe(III)(OH)8Cl] ([Fe4(III)Fe2(II)(OH)12][CO3·2H2O]), green rust II ([Fe4(III)Fe3(II)(OH)12][SO4·2H2O]), magnetite ({alpha}-Fe3O4), lepidocrocite ({gamma}-FeOOH), ferrihydrite Fe(OH)3, and goethite ({alpha}-FeOOH) (Furukawa et al., 2002; Genin et al., 1998; Phillips et al., 2003). The green rust can also serve as a reducing agent to abiotically reduce Se(VI) to Se(IV) and Se(0) (Myneni et al., 1997; Refait et al., 2000). Ferrihydrite and goethite are also strong adsorbents that can be used to effectively remove Se(IV) from water (Balistrieri and Chao, 1987; 1990). Therefore, ZVI may be an inexpensive potential agent to remove Se(VI) from Se-contaminated water.

This study was conducted to determine the removal of Se(VI) from water by ZVI in the presence of varying concentrations of Cl–, SO2–4, NO–3, HCO–3, and PO3–4. The removal process was characterized in a series of batch experiments.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Selenium standards used in this study included Se(IV) (selenite reference standard solution) purchased from Fisher Scientific (Hampton, NH) and Se(VI) (Na2SeO4) from Sigma (St. Louis, MO). Elemental Se was obtained by the chemical reaction of Se(IV) with ascorbic acid (Sigma) (Combs et al., 1996). Fifteen milliliters of 1000 mg L–1 of Se(IV) was added to a 50-mL glass beaker, followed by the addition of 0.05 g of ascorbic acid. The beaker was slightly shaken by hand for less than 0.5 min and then red Se(0) was formed. The red Se(0) solution was passed through a 0.1-µm filter, followed by washing six times with deionized water to remove ascorbic acid and unreacted Se(IV). Elemental Se accumulated on the filter membrane was transferred to a 20-mL glass vial with deionized water and was then sonicated for 15 min before use. Other chemicals, such as sodium sulfate (Na2SO4), sodium chloride (NaCl), sodium hydroxide (NaOH), and sodium phosphate (NaH2PO4 and Na2HPO4), were purchased from Fisher Scientific. Zerovalent iron (40–60 mesh) was obtained from Peerless Metal Powers and Abrasive (Detroit, MI) and was used as received. The surface area of ZVI particles was 1.63 m2 g–1.

Stability Test of Selenium Species in a Phosphate Anion Extractant
Stability of Se species in extractants needed to be tested to correctly extract Se adsorbed in ZVI–FeOH described below. In this study, the term "Fe oxyhydroxides" is defined to represent all corrosion products of ZVI in NaCl and Na2SO4 solutions. Phosphate and NaOH, the most commonly used extractants in Se speciation studies (Jackson and Miller, 2000), were selected for extracting Se species in ZVI–FeOH. The extractant solution containing 0.35 M of PO3–4 with a pH range of 7 to 12.6 was used. In the test, 1.25 g of ZVI was added to each 40-mL glass vial, followed by 1 mL of 1 mM Cl– and SO2–4 solution. The vials were opened to the atmosphere and stood overnight allowing the formation of FeOH (yellowish-brown color) on the surface of some ZVI particles before the addition of the extractants. Afterward, 24 mL of the extractant was added to each vial, followed by spiking with standard Se(VI), Se(IV), and Se(0) in different vials to a final concentration of 1000, 1000, and 2600 µg L–1, respectively. The vials were tightly capped and horizontally shaken in a gyrotory shaker for 2 and 5 h at 180 rpm. Then, solution samples were collected for analysis of Se species after centrifugation (Microfuge 11; Beckman, Fullerton, CA) for 14 min at 12000 rpm to precipitate ZVI–FeOH.

Characterization of Selenate Removal
A series of time course experiments were conducted in the laboratory to determine the removal of Se(VI) by ZVI in a 10 mM Cl– and a 10 mM SO2–4 solution. These experiments were designed to examine whether Se(VI) can be removed from solution. Therefore, pH and Eh were not controlled during the experiments and dissolved O2 in solution was not removed before the experiments. In a closed contained experiment, 25 mL of 10 mM Cl– or SO2–4 solution containing 1000 µg L–1 of Se(VI) was placed to 40-mL EPA glass vials, followed by the addition of 1.25 g of ZVI. The vials were tightly capped and placed horizontally in a gyrotory shaker for shaking at a speed of 180 rpm. The vials were removed from the shaker at 0.125, 0.25, 0.5, 1, 2, 4, 7, and 16 h. Each sample was passed through a 0.2-µm membrane filter (Fisher Scientific) into another 40-mL glass vial for Se species, pH, and Eh analysis. The ZVI–FeOH residue in the vials was washed two times with 5 mL of deionized water and refiltrated. The ZVI–FeOH on the filter membrane was replaced into the original vials for extraction of accumulated Se to ZVI–FeOH particles. During the extraction, 25 mL of 0.35 M PO3–4 solution with a pH of 12.6 was added to each vial. The vials were tightly capped, placed horizontally in a gyrotory shaker, and shaken for 2 h at a speed of 180 rpm. Then, solution samples were collected for Se species analysis by centrifugation for 14 min at 12000 rpm to precipitate ZVI–FeOH. All of the experiments were run in triplicate at room temperature (21 ± 1°C).

In an open system experiment, 25 mL of 10 mM Cl– or SO2–4 solution containing 1000 µg L–1 of Se(VI) was placed in 50-mL Pyrex flasks, followed by the addition of 1.25 g of ZVI. The flasks were not capped and shaken in a gyrotory shaker at a speed of 180 rpm. The sample collection and extraction of accumulated Se to ZVI–FeOH were the same as the closed system experiment described above.

Effects of Different Anions on Selenate Removal
Effects of different anions (Cl–, SO2–4, HCO–3, PO3–4, and NO–3) on Se(VI) removal by ZVI were examined in a series of batch experiments under aerobic conditions. In these experiments, 150 mL of 1000 µg L–1 of Se(VI) containing one of the anions was placed in 250-mL Pyrex flasks, followed by the addition of 5 g of ZVI. These anions included a Cl– and SO2–4 level of 5, 50, and 100 mM, and a HCO–3, PO3–4, and NO–3 range of 1, 5, and 10 mM. The flasks were not capped and shaken in a gyrotory shaker at a speed of 180 rpm. We collected 1.5 mL of the water samples in each flask at 0.5, 1, 2, 4, 7, 24, 31, and 48 h for Se species analysis. Before analysis, the samples were centrifuged for 14 min at 12000 rpm to precipitate ZVI–FeOH. The pH and Eh in the samples were not determined. All of the experiments were run in triplicate at room temperature (21 ± 1°C). The rate constant of Se(VI) removal by ZVI in these experiments was calculated using a simple first-order kinetics equation: dSe(VI)/dt = –kSe(VI), where k is the rate constant of Se(VI) removal. Selenium data that were close to zero were not used because these data were not in a linear range in the rate constant calculation.

Analysis
Redox potential and pH in the filtered samples were immediately measured after collection with a 720A pH/ISE meter (Thermo Orion, Beverly, MA). pH was measured using an Accumet pH combination electrode (Fisher Scientific). The redox potential was measured with an Accumet combination platinum electrode (Ag/AgCl). The measured potential (Ehmeasured) was converted to potential in the solution (Ehactual) relative to a standard H electrode as Ehactual = Ehmeasured + 224.4 mV (Jayaweera and Biggar, 1996).

Only total soluble Se and Se(IV) were monitored in the filtered samples and PO3–4 extract because there were no organic materials and organic Se in the solution–ZVI–FeOH system. Selenite in the samples was determined in a 6 M HCl solution. Total soluble Se in the samples was determined after a reduction of Se(VI) to Se(IV) in 6 M HCl (8.9 mL) added with 0.1 mL of 5% K2S2O8 at 90°C for 15 min. Selenium concentrations in all prepared solutions were analyzed by hydride generation atomic absorption spectrometry (HGAAS) (Zhang et al., 1999b). The Se(VI) concentration was calculated as the difference between total soluble Se concentration and Se(IV). Estimation of nonextractable Se was performed as the differences between the added Se and sum of the total soluble Se and total extractable Se in each sample.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Stability of Selenium Species in the Phosphate Anion Solutions with Different pH
Extraction effectiveness of adsorbed Se from soil constituents is commonly related to the composition and concentrations of extractants used (Jackson and Miller, 2000). Selenite adsorption results through its replacement of the surface hydroxyl groups and its adsorption decreases with increasing OH– concentrations (as the pH increases). Phosphate is considered to have a similar affinity with Se(IV) binding on soil constituents (Balistrieri and Chao, 1987; Goldberg, 1985) and it can replace Se(IV) on adsorption sites. In this study, stability of Se species [Se(0), Se(VI), and Se(IV)] in a 0.35 M PO3–4 solution with a pH range of 7 to 12.6 was tested in a ZVI–FeOH system (Fig. 1) . Selenate was stable in the PO3–4 solutions during a 2- and 5-h reaction, indicating that Se(VI) was not removed by ZVI–FeOH. Selenite was also stable in the PO3–4 solutions with a pH range of 9 to 12.6 during a 2-h reaction. About 49 and 94% of the added Se(IV) were recovered in the pH 7 and 8 solutions, respectively. A similar reaction of Se(IV) was observed in the PO3–4 solutions with a pH range of 7 to 11 during a 5-h reaction. In the pH 12 and 12.6 solutions, 6 and 14% of the added Se(IV) were oxidized to Se(VI). Elemental Se was relatively stable in the PO3–4 solutions during a 2-h reaction with less than 3% of the added Se(0) being oxidized to Se(IV). Oxidation of the added Se(0) increased to 1.5 to 8% during a 5-h reaction. Considering stability of the Se species in the PO3–4 solutions and high efficiency of the extraction of adsorbed Se species in a high pH solution (Jackson and Miller, 2000; Zhang et al., 1999a), the PO3–4 solution with the highest pH (12.6) tested and a 2-h extraction period were selected for extracting Se accumulated to ZVI–FeOH on the removal of Se(VI) in the Cl– and SO2–4 solutions.



View larger version (32K):
[in this window]
[in a new window]
  		Fig. 1. Stability of Se species in a 0.35 M PO3–4 solution containing zerovalent iron (ZVI) and iron oxyhydroxides (FeOH) for 2- (left figures) and 5-h reactions (right figures). The pH range of PO3–4 solution was 7 to 12.6. Error bars show one standard deviation (n = 3).

 
Selenate Removal under Closed and Open Systems
Removal of Se(VI) from a 10 mM Cl– and SO2–4 solution under a closed contained system by ZVI is present in the Fig. 2 . Total soluble Se dropped rapidly from 1000 to 32 µg L–1 in the 10 mM Cl– solution during the first hour of the experiment. Selenate was the only dominant form of Se in the solution. pH increased rapidly from 5.9 to 9.5, with a rapid decrease of Eh from 0.62 to 0.26 V. Total soluble Se decreased to zero during rest of the experiment, with a slight change in pH and Eh. Total PO3–4–extractable Se from ZVI–FeOH increased rapidly to 13.8 mg kg–1 in the first 0.5 h of the reaction, and then stabilized at a range of 12.4 to 13.8 mg kg–1. Selenite was the major Se form, increasing from 3.83 to 12.8 mg kg–1 at 0.125 to 1 h, respectively. Selenate increased to 3.16 mg kg–1 at 0.25 h and then decreased to zero at 16 h.



View larger version (34K):
[in this window]
[in a new window]
  		Fig. 2. Removal of Se(VI) in 10 mM Cl– (left figures) and SO2–4 (right figures) solutions by zerovalent iron (ZVI) under a closed condition. (A) Changes in pH and Eh during the removal of Se(VI). (B) Removal of Se(VI) in the solutions. (C) Selenium species in the ZVI and iron oxyhydroxides (FeOH) extracted by a 0.35 M PO3–4 solution (pH 12.6). Error bars show one standard deviation (n = 3).

 
Total soluble Se in a 10 mM SO2–4 solution decreased from 1000 to 610 µg L–1, with a rapid increase in pH from 6 to 10.4 and a rapid decrease in Eh from 0.62 to 0.23 V during the first hour of the experiment (Fig. 2). Afterward, total soluble Se slightly decreased to 442 µg L–1 after 16 h, with a slight change in pH and Eh. Selenate was the major form of Se during the 16 h of the experiment, accounting for 97 to 100% of total Se. Total PO3–4–extractable Se increased rapidly to 5.8 mg kg–1 in the first hour of the reaction, and then slightly increased to 7.37 mg kg–1 at 16 h. Selenite also was the dominant form of Se in the extract, accounting for 95 to 100% of the total Se.

Under the open condition (Fig. 3) , total soluble Se in the 10 mM Cl– solution decreased rapidly from 1000 to 46.9 µg L–1, with an increase in pH from 5.7 to 7.6 and a drop of Eh from 0.62 to a range of 0.34 to 0.43 V during the first 2 h of the reaction. Total soluble Se dropped to 4 µg L–1 during rest of the experiment. pH and Eh stabilized at the levels of 6.9 to 7.6 and 0.34 to 0.43 mV, respectively. Selenate was the only Se species in the solution during the 16 h of the experiment. Total PO3–4–extractable Se increased rapidly to 12.8 mg kg–1 during the first 4 h of the experiment, and then decreased to 2.95 mg kg–1 at 16 h. Selenite was the major form of Se in the extract, accounting for 59 to 99%. Selenate was relatively low, having a peak of 2.38 mg kg–1 at 0.5 h.



View larger version (35K):
[in this window]
[in a new window]
  		Fig. 3. Removal of Se(VI) in 10 mM Cl– (left figures) and SO2–4 (right figures) solutions by zerovalent iron (ZVI) under an open condition. (A) Changes in pH and Eh during the removal of Se(VI). (B) Removal of Se(VI) in the solutions. (C) Selenium species in the ZVI and iron oxyhydroxides (FeOH) extracted by a 0.35 M PO3–4 solution (pH 12.6). Error bars show one standard deviation (n = 3).

 
Total soluble Se also dropped from 1000 to 68 µg L–1 in the 10 mM SO2–4 solution during the 16 h of reaction (Fig. 3). Selenate was the only Se species in the solution. pH increased from 5.9 to 9.4 during the first hour of the experiment and then remained at a level of 8.9 to 9.6. Eh decreased from 0.62 to a range of 0.31 to 0.4 V. Total PO3–4–extractable Se increased from 0.81 to 9.18 mg kg–1 at 0.125 h to the end of the experiment. Selenite was the only Se species in the extract.

Effects of Anions on Selenate Removal
Effect of Cl–, SO2–4, HCO–3, PO3–4, and NO–3 on Se(VI) removal by ZVI under aerobic conditions is illustrated in Fig. 4 and 5 . Only Se(VI) (total Se) is presented in Fig. 4 and 5 because no measurable Se(IV) was detected in the samples. Selenate in the all of the solutions decreased during the experiments. More than 99% of the added Se(VI) was removed in the Cl– solutions (5, 50, and 100 mM) (Fig. 4 and Table 1). In contrast, 99, 87.4, and 79% of the added Se(VI) was removed in the 5, 50, and 100 mM SO2–4 solutions, respectively.



View larger version (28K):
[in this window]
[in a new window]
  		Fig. 4. Effect of Cl– and SO2–4 concentrations on the removal of Se(VI) by zerovalent iron (ZVI). Error bars show one standard deviation (n = 3). Inserted small figures show the first-order kinetics of Se(VI) removal.

 


View larger version (24K):
[in this window]
[in a new window]
  		Fig. 5. Effect of HCO–3, PO3–4, and NO–3 concentrations on the removal of Se(VI) by zerovalent iron (ZVI). Error bars show one standard deviation (n = 3). Inserted small figures show the first-order kinetics of Se(VI) removal.

 

View this table:
[in this window]
[in a new window]
  		Table 1. Rate constant (k) of Se(VI) removal by zerovalent iron (ZVI) in the presence of Cl– or SO2–4 (see Fig. 4).

 
In a range of 1, 5, and 10 mM of NO–3, removal of Se(VI) was much faster than that of HCO–3 solutions, which was faster than that in the PO3–4 solutions (Fig. 5 and Table 2). Selenate was almost completely removed in the 1, 5, and 10 mM NO–3, 1 mM HCO–3, and 1 mM PO3–4 solutions during the first 24 h. At the end of the experiment, about 99, 88, 82, and 43% of the added Se(VI) were removed from the 5 and 10 mM HCO–3 and PO3–4 solutions, respectively.


View this table:
[in this window]
[in a new window]
  		Table 2. Rate constant (k) of Se(VI) removal by zerovalent iron (ZVI) in the presence of HCO–3, PO3–4, or NO–3 (see Fig. 5).

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
CONCLUSIONS
 REFERENCES
 
Zerovalent iron as a reducing agent has been used to treat Cr(VI), NO–3, and U(VI)-contaminated waters (Alowitz and Scherer, 2002; Farrell et al., 1999; Huang et al., 1998; Powell et al., 1995; Qiu et al., 2000). This study revealed that ZVI is also capable of removing Se(VI) from 10 mM Cl– and SO2–4 solutions. After a short period of time (16 h), the added Se(VI) was completely removed from the Cl– solution, and 56 and 93% of the added Se(VI) was removed in the SO2–4 solution under a closed contained and open conditions, respectively.

The major mechanism for the rapid removal of Se(VI) by ZVI–FeOH can be attributed to the reduction of Se(VI) to Se(IV) by Fe(II) oxidized from ZVI (Murphy, 1988), followed by rapid adsorption of Se(IV) to FeOH due to lower redox potentials of Fe(III) and Fe(II) than that of Se(VI) and Se(IV) (Genin et al., 1998) and stronger adsorption of Se(IV) to FeOH than Se(VI) (Balistrieri and Chao, 1987, 1990). Zingaro et al. (1997) reported that 99% of added Se(VI) was rapidly reduced after a 7-h reaction with Fe(II) at an initial Fe(II) to Se(VI) molar ratio of 9. Manning and Burau (1995) extracted both Se(VI) and Se(IV) from Fe precipitates after a reaction of Fe(II) with Se(VI) in 0.1 M NaClO4 solution. In this study, little Se(IV) was found in both 10 mM Cl– and SO2–4 solutions during the removal of Se(VI), revealing that Se(IV) formed from the reduction Se(VI) to Se(IV) by ZVI–Fe(II) was rapidly adsorbed to FeOH. Analysis of Se species in the PO3–4 extract shows that Se(IV) increased rapidly in the FeOH, with a rapid decrease of Se(VI) in the Cl– and SO2–4 solutions.

Another mechanism that cannot be ruled out for the removal of Se(VI) might be the direct adsorption of Se(VI) to FeOH, followed by the reduction of adsorbed Se(VI) to Se(IV). Several studies have reported that Se(VI) can be adsorbed to FeOH via surface complexes of outer-sphere layer or inner-sphere layer, and/or both layers (Hayes et al., 1987; Manceau and Charlet, 1994; Su and Suarez, 2000). Refait et al. (2000) reported that SO2–4 in green rust II , one of the Fe corrosion products, can be replaced by SeO2–4 to form GR and Se(VI) can be reduced to Se(IV) after the replacement. In a study on abiotic reduction of Se(VI) by green rust using X-ray absorption near edge structure (XANES) and extended X-ray absorption fine structure (EXAFS) spectroscopy, Myneni et al. (1997) reported that Se(VI) can be incorporated into the interlayers of green rust to form bidentate binuclear and edge-sharing complex with structural Fe(II) and reduced immediately to Se(IV). In this study, we observed greenish FeOH on the filter membranes after filtration at the initial period of the experiments, showing the existence of green rust I in the Cl– solution and green rust II in the SO2–4 solution. Analysis of Se species in the ZVI–FeOH revealed that extracted Se(VI) from ZVI–FeOH accounted for as high as 30 to 40% of total extractable Se during first 0.5 h of the Cl– experiment and then decreased rapidly to zero, providing evidence that Se(VI) was adsorbed to FeOH. These studies indicate that Se(VI) can be reduced rapidly to Se(IV) when it is adsorbed to FeOH (e.g., green rust I and II).

Further reduction of Se(IV) to Se(0) might occur in the FeOH. Calculation of Se mass balance showed that nonextractable Se increased with time. At the end of the experiment, nonextractable Se was 17.3 and 9.75 mg kg–1 in the Cl– and SO2–4 experiments under an open condition, respectively. Under a closed contained system, it was 6.86 and 3.78 mg kg–1 in the Cl– and SO2–4 experiments, respectively. By using XANES and EXAFS spectroscopy, Myneni et al. (1997) detected that Se(0) existed in the green rust after a reduction of Se(VI). Using XANES, Roberson (1999) also reported that Se(IV) and Se(0) was observed in FeOH during a study on the removal of Se(VI) by ZVI. After a reaction of Se(VI) with Fe(II), Zingaro et al. (1997) identified Se(0) in the Fe precipitates using X-ray photoelectron spectroscopy (XPS) and transmission electron microscopy (TEM)–energy dispersive spectroscopy (EDS) analysis. These results may suggest that nonextractable Se in the present study might partly be a form of Se(0), which cannot be extracted by a basic PO3–4 extractant in a short period of the time (2 h). The increase of nonextractable Se may reveal a further reduction of Se(IV) to Se(0). Part of nonextractable Se might be attributed to the formation of Fe(II)–Se(IV) precipitates [Fe2(SeO3)3 and Fe2(OH)4SeO3] in the both Cl– and SO2–4 solutions (Elrashidi et al., 1987).

Experimental conditions affect the removal of Se(VI) by ZVI. Under a closed contained system, ZVI in the vials can be reciprocally moved during shaking due to horizontal placement of the vials in a shaker. Rapid reaction of ZVI–FeOH with Se(VI) resulted in a removal of 97 and 40% of the added Se(VI) during the first hour of the experiment in 10 mM Cl– and SO2–4 solutions, respectively. Removal of Se(VI) was slow in the SO2–4 solution during the rest of the experiment. One possible reason is that ZVI was aerobically oxidized rapidly to Fe(II) (2Fe0 + 2H2O + O2 = 2Fe2+ + 4OH–) at an initial period of time when dissolved O2 and air existed in the closed contained vials. After consuming O2, anaerobic reactions of ZVI took place (Fe0 + 2H2O = Fe2+ + 2OH– + H2), which produced H2. Accumulation of H2 limited continuous corrosion of ZVI under a closed contained system, thus slowing the production of FeOH used for the removal of Se(VI). Under an open condition, ZVI was not mobilized at the bottom of the flasks. Relatively slow corrosion of ZVI resulted in relatively slower removal of Se(VI) at an initial period of the experiment than that under the closed contained condition. However, continuous corrosion of ZVI by dissolved O2 removed all added Se(VI) in the Cl– solution at 7 h, and removed greater amounts of Se(VI) in the SO2–4 solution at end of the experiment than that in the closed contained system. Improvement of Se removal from aqueous solution in the presence of dissolved O2 by ZVI can also be found in Murphy's (1988) study, in which he reported that Se(VI) removal was much greater under open conditions than that under a condition of purging with argon. Recently, Meng et al. (2002) reported that >97% of SeCN– was removed from water by ZVI in the presence of dissolved O2, whereas only 8% of SeCN– was removed in water purged with N2 after 7h.

Relatively slower removal of Se(VI) in the SO2–4 solutions than in the Cl– solutions can be partially attributed to the similarities in the chemical properties of SO2–4 and Se(VI). In this study, the concentrations of SO2–4 were much greater than Se(VI). High SO2–4 concentrations can compete for adsorption sites on FeOH, thus delaying and limiting the removal of Se(VI). Relatively slow removal of Se(VI) in the SO2–4 solutions might also be partially attributed to rapid corrosion of ZVI in the SO2–4 solutions (Hunt, 2000), resulting in a higher pH. This higher pH in the SO2–4 solution can decrease Se adsorption, causing a relatively slow removal of Se(VI).

Higher PO3–4 levels significantly affected the removal of Se(VI) by ZVI–FeOH. In comparison with Se(VI) removal in the 5 mM Cl– solution, 1 mM PO3–4 only had a slight effect on the removal of Se(VI). Increases in the concentrations of PO3–4 to 5 and 10 mM significantly reduced Se(VI) removal, with low k values of 0.037 and 0.011 h–1, respectively. This decrease of Se(VI) removal might be attributed to the reaction of Fe(II) with PO3–4. The formation of Fe-PO3–4 minerals [e.g., Fe3(PO4)2(H2O)8, vivianite] consumes Fe(II) used for reducing Se(VI), resulting in a lower removal of Se(VI). The presence of PO3–4 can also compete for adsorption sites of FeOH, thus reducing adsorption of Se (Balistrieri and Chao, 1987; Goldberg, 1985).

Higher levels of HCO–3 can also reduce the removal of Se(VI) in HCO–3–CO2–3–ZVI–FeOH. Reaction of Fe(II) with CO2–3 can form FeCO3 (siderite) that is often found in ZVI barriers used for removing environmental contaminants from ground water (Phillips et al., 2003), and reduce the amounts of Fe(II) available for reducing Se(VI). In this study, increasing the concentrations of HCO–3 from 1 to 10 mM led to a 18% decrease of Se(VI) removal in the ZVI–FeOH system.

Nitrate is commonly considered as a competitive electron acceptor in bacterial reduction of Se(VI) to Se(0) because the redox potential of NO–3 to N2 in an aquatic system is very similar to that of Se(VI) to Se(IV) and much higher than Se(IV) to Se(0) (Masscheleyn and Patrick, 1993). However, this study reveals that NO–3 in a range of 1 to 10 mM had little effect on the removal of Se(VI) by ZVI–FeOH in the Se–NO–3–ZVI system. Like the 5 mM Cl– solution, about 100% of the added Se(VI) was removed from solutions containing 1, 5, and 10 mM of NO–3 during the first 24 h of reaction. The most likely reason is that the major reduction of Se(VI) to Se(IV) was caused by Fe(II) so that the removal of Se(VI) is controlled by oxidation of ZVI to Fe(II) by dissolved O2 under an open condition. The presence of NO–3 did not significantly affect oxidation of ZVI to Fe(II) by dissolved O2, thus resulting in little effect of NO–3 on Se(VI) removal by ZVI.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
REFERENCES
 
Results from this study reveal that Se(VI) can be removed from aqueous solutions with varying concentrations of Cl–, SO2–4, NO–3, HCO–3, and PO3–4. During the 16 h of the experiment, 100 and 56% of the added Se(VI) were removed in 10 mM Cl– and SO2–4 solutions under a closed contained system, respectively. Under an open condition, 100 and 93% of the added Se(VI) were removed in the Cl– and SO2–4 solutions, respectively. The experiments also showed that there was little effect of Cl– (5, 50, and 100 mM), SO2–4 (5 and 50 mM), NO–3 (1, 5, and 10 mM), HCO–3 (1 and 5 mM), and PO3–4 (1 mM) and a slight effect of SO2–4 (100 mM), HCO–3 (10 mM), and PO3–4 (5 mM) on Se(VI) removal. Only higher PO3–4 (10 mM) levels significantly inhibited Se(VI) removal by ZVI. In California, Se-contaminated drainage water contains an average concentration of 987 (Cl–), 2282 , 97 , 214 , and 0.12 mg L–1 (Oswald et al., 1989), which are much lower levels than those tested in our study. Our work suggests that ZVI may be an inexpensive agent that can be used to treat Se-contaminated drainage water.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 


Related articles in JEQ:

This Issue in Journal of Environmental Quality

JEQ 2005 34: 403-407. [Full Text]  




This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Related articles in JEQ
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via ISI Web of Science (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zhang, Y.
Right arrow Articles by Frankenberger, W. T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zhang, Y.
Right arrow Articles by Frankenberger, W. T., Jr.
Agricola
Right arrow Articles by Zhang, Y.
Right arrow Articles by Frankenberger, W. T.
Related Collections
Right arrow Water Quality
Right arrow Batch Studies
Right arrow Remediation
Right arrow Heavy Metals
Right arrow Water Pollution

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Vadose Zone Journal Journal of Plant Registrations
Journal of Natural Resources
and Life Sciences Education		 Soil Science Society of America Journal