Environ. Sci. Technol. 2003, 37, 1859-1864
Uptake and Reduction of Cr(VI) to Cr(III) by Mesquite (Prosopis spp.): Chromate-Plant Interaction in Hydroponics and Solid Media Studied Using XAS M. V. ALDRICH, J. L. GARDEA-TORRESDEY,* J. R. PERALTA-VIDEA, AND J. G. PARSONS Department of Chemistry and Environmental Science and Engineering, University of Texas at El Paso, El Paso, Texas 79968
Chromium (Cr) is a well-established carcinogen that is a contaminant at half of the EPA Superfund sites in the United States. Two separate studies were performed to investigate the possibility that mesquite (Prosopis spp.), which is an indigenous desert plant species, can remove Cr from the environment via active transport systems to the aerial portions of the plant. The first study was performed by growing mesquite on solid media (agar) at Cr(VI) concentrations of 75 and 125 ppm. The accumulation found in the leaves under the present conditions indicated that mesquite could be classified as a hyperaccumulator of chromium. The second study was conducted to investigate the differences between the type of Cr ligand involved in Cr uptake with agar and hydroponic cultures. We used X-ray absorption spectroscopy (XAS) to determine the mechanisms involved in the uptake and binding of Cr(VI) in live mesquite tissue. The XAS results for this study showed that some of the supplied Cr(VI) was uptaken by the mesquite roots; however, the data analyses of the plant tissues demonstrated that it was fully reduced to Cr(III) in the leaf tissues. Experiments are currently being performed to evaluate the behavior of the Mesquite plant using lower Cr concentrations.
Introduction Chromium contaminated water and soil are the results of industrial operations such as smelting, tanning, electroplating, and mining activities (1-4). These practices account for nearly half of the chromium wastes at Superfund sites in the United States (5). The health effects derived from chromium exposure range from dermatitis (superficial irritation) and dermatosis (subdermal ulceration) as well as various types of cancer (6-8). An increased risk of public exposure exists due to extensive contaminant translocation since wind erosion is prevalent in arid regions such as the Southwestern United States. Thus, the severity of health effects from chromium exposure in these areas has escalated and has become a public health concern. Hazardous waste clean up technologies such as pump and treat, soil washing, and soil vitrification are expensive * Corresponding author phone: (915)747-5359; fax: (915)747-5748; e-mail:
[email protected]. 10.1021/es0208916 CCC: $25.00 Published on Web 04/02/2003
2003 American Chemical Society
and unsustainable. These technologies create long term liabilities to communities both environmentally and economically because of state and federal requirements to perform continuous hazardous waste monitoring (9). However, a relatively new technology, phytoremediation, which is the use of plants to extract heavy metals, is an especially attractive alternative to conventional technologies because of the low costs for implementation and maintenance of the system (10). Progress in recent years of phytoremediation technological development has led to plant gene manipulation and soil chelation treatments to facilitate plant uptake of heavy metals, which has improved its acceptance as a remediation tool (11-13). Phytoremediation has the potential to become an effective, environmentally friendly, and inexpensive alternative to other remediation methods. It has been well documented that both crop and native plant species are able to uptake metals (14-17). The majority of plants that have been identified for the purpose of phytoremediation grow in fertile, temperate regions. However in the arid Southwestern United States, these species do not grow well. Desert plants have adapted to tolerate and thrive in harsh desert conditions such as high salt content in soils, nutrient-poor soil, and extreme temperatures. Because of their highly specialized physiologies, desert plants are obvious candidates for examination of other stress-induced adaptive mechanisms such as growth in heavy metal contaminated sites. Two studies using the desert plant species creosote bush (Larrea tridentata) have been conducted (18, 19). In these studies it was found that this species is capable of uptaking copper with no apparent physiological damage. With the use of X-ray absorption spectroscopy (XAS), it was possible to determine that copper was in the plant as both Cu(I) and Cu(II) (18). In addition, by using scanning electron microscopy (SEM), crystalline structures of copper within the leaf were revealed. These results could indicate a metal detoxification mechanism of the plant (19). Other XAS studies have shown that silica-immobilized alfalfa (Medicago sativa) is able to effectively bind Fe(II) and Fe(III) at a pH of 5.0 at levels of 2.88 mg/g for Fe(II) and 4.47 mg/g for Fe(III) (20). In the present study, we investigate the ability of the desert plant species mesquite (Prosopis spp.) to actively uptake Cr(VI). Two separate studies were conducted: one to determine the Cr(VI) uptake capacity of mesquite species and the other to determine the oxidation state of the Cr transported throughout the plant. The first study involved growing mesquite for 26 days using as treatments 75 and 125 ppm of Cr(VI) supplied as potassium dichromate to an agar media. The second study was conducted with both agar and hydroponic cultures at 80 ppm to establish if there was a difference in the geometric coordination of Cr(VI) with each media type. X-ray absorption spectroscopy (XAS) was used to determine the coordination chemistry of chromium in the roots, stems, and leaves of the plant. Our results are reported herein.
Methodology Plant Germination and Cultivation. The seeds used in this experiment were supplied by Wild Seeds Company (Tempe, AZ). The seeds were soaked overnight in distilled water, washed, and treated with Captan to eliminate fungal contamination. A modified Hoagland’s nutrient solution (pH 5.3 ( 0.2) was used for these experiments at half strength, which consisted of the following: Ca(NO3)2‚ 4H2O, 3.57 × 10-4 M; H3BO3, 2.31 × 10-5 M; CaCl2‚2H2O, 2.14 × 10-3 M; KH2PO4, 9.68 × 10-4 M; KNO3, 2.55 × 10-4 M; MgClO4, 1.04 VOL. 37, NO. 9, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1859
× 10-3 M; FeCl3, 6.83 × 10-5 M; and MoO3, 1 × 10-5 M, as per Peralta et al. (21). The chromium(VI) was introduced into the nutrient solution as K2Cr2O7. During preparation, an aliquot of 80 mL of a 1000 ppm solution of K2Cr2O7 was added to 1 L of the nutrient solution. The nutrient solution containing the K2Cr2O7 was then autoclaved. The chromium(VI) contaminated nutrient solutions were then separated for either the agar growth treatment or the hydroponic growth treatment. For the agar experiments, Bacto-Agar was added at 0.48% w/v. The seeds were placed directly on the growing medium under a laminar flow hood and set under an 18/6-h light and dark cycle. The temperatures were set to 25 °C during the day and 18 °C at night. In the hydroponics experiment, seeds were sown onto a perlite media and watered with the mineral solution. After 1 week, the seedlings were transplanted to a hydroponic system supplied with Cr(VI). The plants grown in the mineral solution minus the metal served as the controls. These studies were performed in triplicate for quality control purposes. After 2 weeks, the whole plants were carefully extracted from each media type. The samples were washed in 0.01 M HCl, rinsed in deionized water, and separated into roots, stems, and leaf portions.
Inductively Coupled Plasma Optical Emission Spectrometry (ICP-OES) Sample Preparation. The seedlings were separated into roots, stems, and leaf portions and then set in a 70 °C oven for 2 days to dry. The dried samples were prepared for ICP-OES analysis by using 10 mL of trace pure nitric acid. They were then placed in Teflon-lined “bombs” in a Perkin-Elmer Multiwave microwave oven and digested using the preprogrammed EPA method 3051 (22). After digestion, the samples were diluted to a 1:9 (sample:deionized water) ratio for analysis. ICP Analysis. A Perkin-Elmer Optima model 4300DV ICPOES was used to determine the chromium concentration in the samples. The ICP-OES was equipped with a Meinhard nebulizer, and the analyses were performed under the following conditions: sample flow rate of 1.75 mL min-1; gas flow rate of 15 mL h-1; and RF power of 1500 W. A calibration correlation coefficient of 0.99 or better was obtained for all analyses. XAS Sample Preparation. Sample preparation for XAS consisted of sample immersion in liquid nitrogen for 25 min until frozen and then placed in a Labconco Freeze-Dry System (Freezone 4.5) (Labconco Corporation, Kansas City, MO). After lyophilization, the samples were ground and packed into 1.0 mm sample plates with Mylar tape windows for analysis at Stanford Synchrotron Radiation Laboratories (SSRL).
XAS Studies The chromium laden mesquite samples were taken to Stanford Synchrotron Radiation Laboratories (SSRL) for X-ray absorption spectroscopy, including X-ray absorption near edge spectroscopy (XANES) and extended X-ray absorption fine structure (EXAFS) of the Cr K edge (5.989 keV). These analyses were performed in order to determine the oxidation state, the interatomic distance, and number of nearest neighbor atoms to the chromium atoms in the different parts of the mesquite samples and the agar (23). The beamline operating conditions were a beam current of 60-100 mA and energy of 3 GeV. All samples were run using a helium cryostat at a temperature of about 15 K to reduce DebyeWaller effects that occur from thermal disorder in samples. Fluorescence spectra of the chromium laden mesquite samples were taken using a 30-element Canberra germanium detector. However, transmission measurements were re1860
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 9, 2003
TABLE 1: Uptake data for Mesquite Supplied with Chromium(VI) at 75 and 125 ppm for 26 Days After Sprouting with (95% Confidence Intervala plant portion leaf stem root a
control
75 ppm
125 ppm
0.03 ( 0.0 322.8 ( 69.1 991.5 ( 114.9 0.0 ( 0.0 888.9 ( 137.1 2262.9 ( 81.2 0.02 ( 0.0 7636.3 ( 2463.4 10983.8 ( 1641.9
Values are mg Cr kg-1 dry weight.
corded for the model compounds chromium(III) sulfide, chromium(III) acetate, chromium(III) phosphate, chromium(III) nitrate, and potassium dichromate using helium filled ionization chambers. The model compounds were diluted in boron nitride to give a change of one absorption unit across the edge. The model compounds were diluted and homogenized with the boron nitride using a pestle and mortar prior to analysis. A silicon (220, at φ 90) double crystal monochromator with an entrance slit of 1 mm was used for all measurements. The monochromator was detuned by 50% to reject higher harmonics that may have interfered with the sample analysis. An average of two to three scans of each sample was made for each XANES and EXAFS spectrum in order to improve the signal-to-noise ratio. All samples were packed in 1 mm aluminum holders with X-ray transparent tape and measured as solids (24). All spectra were calibrated against the edge position of an internal standard chromium foil (5.989 keV).
XAS Data Analysis The WinXAS software was used to analyze the data through standard methods (25). The samples were calibrated against the chromium edge energy using first and second degree derivates of the foil edge energy (5.989 keV). The sample spectra were then background corrected using a 1 degree polynomial on the preedge region and a spline of 7 knots on the postedge region. The XANES region was then extracted from the entire spectra, from 5.95 to 6.20 keV. This was then fitted using a LC-XANES (linear combination XANES) fitting of 5000 iterations of the chromium model compounds. The entire XAS spectra were then converted into K space for the extraction of the EXAFS spectra. The conversion into K space is based on the kinetic energy photoelectrons ejected at the edge energy, using first and second derivatives of the edge energy of the sample. The spectra were converted into k space (or wave vector space). The resulting EXAFS spectra were then k weighted to 2, between 2 and 12.2 Å-1 and subsequently Fourier transformed using a modified Hanning window over the first and last 10% of the EXAFS. The fitting of the EXAFS was performed using least-squares fitting of the Fourier filtered spectra, using FEFF V 8.00 an ab initio multiple-scattering code (26). The FEFF input files were created using the ATOMS software from crystallographic data of the model compounds (27). From the ab initio calculations of FEFF, the number of coordinating atoms, the nearest neighboring atoms, and the Debye-Waller factors were calculated.
Results and Discussion Table 1 shows the results of a 26-day Cr(VI) uptake study with mesquite at 75 and 125 ppm in agar. The results for 26 day uptake using dry weights were as follows: root levels were approximately 7636 mg Cr kg-1 for the 75 ppm treatment and approximately 10983 mg Cr kg-1 for the 125 ppm treatment; stem Cr levels were approximately 889 mg Cr kg-1 at 75 ppm treatment and 2263 mg Cr kg-1 for the 125 ppm treatment. The leaves had the lowest amounts of Cr, containing about 323 mg kg-1 at 75 ppm and roughly 992 mg
FIGURE 1. XANES data for the model compounds of Cr(VI) as potassium dichromate (Cr(VI)) and the Cr(III) compounds). kg-1 for the 125-ppm treatment. These data also compare favorably with previous research. Chatterjee and Chatterjee (15) found that cauliflower (Brassica oleracea L.) could accumulate Cr supplied as Cr(VI) in concentrations of 775, 7.2, and 10.6 µg Cr g-1 dry weight in the roots, stem, and leaves, respectively. Other researchers, using biomass from well-established trees (Dalbergia sissoo, Acacia arabica, and Populus euroamericana) grown on Cr-contaminated sites, found an average of 564 mg Cr g-1 dry weight in the roots and 152 mg Cr g-1 dry weight in the leaves (4). Our data indicate that mesquite is a hyperaccumulator of chromium (28). Experiments are currently being performed to evaluate the behavior of Mesquite plant using lower Cr concentrations. The second study that was conducted incorporated the use of XAS to determine differences in Cr coordination in mesquite plants grown in hydroponic and agar media contaminated with Cr(VI) at a concentration of 80 ppm. Similar studies have been conducted to determine the differential uptake between plants supplied with either Cr(III) and Cr(VI). Mishra et al. (29), using sand and soil with Cr(III) and Cr(VI) and maize (Zea mays), showed a greater whole plant accumulation of Cr(III), but in the aerial portion of the plant, there was a higher accumulation of Cr(VI). The result could be explained either by an immobilization of Cr(III) at root level or because Cr(VI) was translocated as is to the aerial part of the plant. However, evidence for active transport of Cr was reported by Skeffington et al. (30) who used metabolic inhibitors to determine if there is difference between Cr(III) and Cr(VI) uptake in barley. That research showed that Cr(VI) translocation was reduced significantly, while Cr(III) translocation was not, which proved that Cr(VI) translocation is via an active transport mechanism. In our study, we determined that nearly 50% of the chromium in the agar remained as Cr(VI) and that the Cr(VI) was reduced to Cr(III) in all portions of the plant. The next sections discuss the chromium ligands formed in the root portion of the plant, which may explain chromium reduction in the root portion. Figure 1 shows the XANES data for the model compounds of Cr(VI) as potassium dichromate as well as Cr(III) compounds of chromium sulfide, nitrate, acetate, and phosphate. The XANES data were used to determine the oxidation state of the chromium. The XANES data for K2Cr2O7 showed the well-defined preedge peak characteristic of Cr(VI) (from 5.958 to 6.000 keV). The lack of this feature in all other model compounds shows chromium in its reduced valence state Cr(III). Concurrently, we show the XANES data for the roots, stems, leaves, and agar media (Figure 2A). The preedge feature in the agar sample is similar to that of the K2Cr2O7 spectra.
FIGURE 2. (A) XANES and LC-XANES fittings data of the mesquite roots, stems, leaves. (B) XANES data and the LC-XANES fittings for hydroponically grown mesquite roots, stems, and leaves. However, the preedge of the agar spectra is approximately half the area of the K2Cr2O7, which indicated that approximately 50% of the Cr(VI) remained as Cr(VI) in the agar. The lack of such a pronounced feature in all of the biomass sample spectra (roots, stems, and leaves) indicates that the Cr is present as Cr(III). This also confirms that the mesquite can actively uptake and reduce Cr(VI) to Cr(III). The mechanisms by which mesquite plants can do this remain unclear. Lytle and co-workers have hypothesized that the reduction of Cr(VI) to Cr(III) occurs in the roots, catalyzed by Cr reductases; however, they were unable to confirm that these enzymes were present (31). Cervantes and co-workers have reported similar findings, concluding that the reduction of Cr(VI) to Cr(III) occurs in the roots despite the lack of enzyme isolation (32). We hypothesize that the initial uptake and sorption of chromium may be explained by root respiration rates and by the high percentage of chromium acetate found in the root, as explained later. It is at the root surface where root respiration can account for as much as 50% of carbon dioxide (33). Since high amounts of chromium acetate were found in the roots (Table 2) and the amount of CO2 at the root surface is high, it is possible that the Cr(VI) was reduced via an oxidation-reduction reaction with an acetate-type ligand. The high negative potential of the redox pair of CO2/ acetate (P of -4.7) (33) could explain the reduction of Cr(VI) at the root/media interface given the high amount of acetate found in the roots. However, the results could vary if mesquite plants are grown in a different environment. The data for the percent of chromium species in the solid media experiment are presented in Table 2. In the agar sample VOL. 37, NO. 9, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1861
TABLE 2. Percent Chromium Found in Mesquite Plants Grown in Agar with 80 ppm for Cr(VI) Supplied as K2Cr2O7 for the Agar, Root, Stem, and Leaf sample
agar
root
stem
leaf
% K2Cr2O7 % chromium(III) acetate % chromium(III) nitrate % chromium(III) phosphate
48 ( 0.6 21 ( 3.5 0.0 31 ( 2.7
0.0 57.8 ( 0.2 33.4 ( 0.5 8.8 ( 1.6
0.0 18 ( 7.3 17 ( 10.1 65 ( 2.3
0.0 18 (6.7 27 ( 5.6 55 ( 2.4
TABLE 3. Percent Chromium Found in Mesquite Grown Hydroponically with 80 ppm Cr(VI) supplied as K2Cr2O7 for Root, Stem, and Leaf sample
root
stem
leaf
% K2Cr2O7 % chromium(III) acetate % chromium(III) nitrate % chromium(III) phosphate
1.2 ( 0.42 43.3 ( 3.8 27.6 ( 1.4 27.9 ( 0.07
6.2 ( 0.5 35.0 ( 1.2 0.0 58.8 ( 1.3
0.0 37.6 ( 0.1 29.1 ( 0.2 33.3 ( 0.2
the best fits is with Cr(VI) as K2Cr2O7 at nearly 50% present as well as a mixture of chromium acetate (∼21%) and chromium phosphate (31%). Since the plant tissues did not contain Cr(VI), one can assume that the Cr(VI) was reduced either before or after entering the plant root, as discussed earlier. The best fits of the chromium compounds found in the roots consisted of almost 60% chromium acetate, 33% chromium nitrate, and a small amount of chromium phosphate (8%). The smaller amounts of chromium nitrate corroborate with the findings of Sharma and co-workers (16), who found a decrease in protein nitrogen with increased amounts of Cr in wheat (Triticum aestivum L.). The best fittings of the stem and leaf data showed high percentages of chromium phosphate, 65% and 55%, respectively, and the same amount of chromium acetate, 18%. Chromium nitrate was higher in the leaves (∼27%) than in the stems (∼17%). Results for the hydroponics experiment were similar for that of the solid media experiment. Table 3 shows the distribution of chromium compounds for the hydroponically cultured mesquite. As with the best fits of the agar experiment, chromium acetate levels are highest in the roots (43%) along with a small amount of Cr(VI) (1%), while the distribution of chromium nitrate and chromium phosphate was the same (∼27%). However, chromium phosphate was higher in the stems (∼59%), but the chromium acetate percentage was almost double than that found in the agar stem sample (35%). Cr(VI) was also present at a low level in the stem (6.2%). The fitting analysis of leaf hydroponics tissue showed that there was a mostly even distribution of chromium(III) acetate, nitrate, and phosphate, ∼38%, ∼29%, and ∼33%, respectively. However, Cr(VI) was not found the leaf portion of the plants. Many structural cells of plant contain nitrogen and phosphate moieties, such as amino acids and phospholyses (34, 35), which may make this distribution data for chromium nitrate and phosphate unremarkable. Two theories exist for metal ligand movement through the plant. One is that the metal ligands go through the xylem and through membranes that are ion specific (36). Salt et al. have theorized that specific carrier cells are the transport mechanisms for these ligands (37). Photosynthetic pathways may be responsible for the subsequent conductance of the ligands through the ATPADP pathways into the leaf portion of the plant, whereupon the ligands are sequestered in the leaf vacuoles (38). These observances are analogous to the studies done for Cu(II) on creosote bush (19). The use of TEM in those experiments showed that Cu(II) was exuded through the leaf stomata and consequently formed a crystalline copper residue on the surface of the leaf (19), which is likely a detoxification pathway for the shrub. 1862
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 9, 2003
FIGURE 3. Fourier transformed EXAFS spectra of model compounds as potassium dichromate and the Cr(III) compounds;
TABLE 4. FEFF Fittings of the Model Compounds Showing the Bond, Coordination Number (N), Atomic Radius R (Å), and Sigma Square (σ2) Values sample chromium(III) acetate chromium(III) sulfide K2Cr2O7 chromium(III) nitrate chromium(III) phosphate
bond
N (( 1)
R
σ2 (()
Cr-O Cr-C Cr-S Cr-Cr Cr-O Cr-O Cr-O Cr-O Cr-O Cr-O Cr-O Cr-P Cr-P
5.9 7.1 6.0 1.6 2.01 1.2 0.9 6.0 7.4 2.5 3.1 3.2 2.9
1.96 3.44 2.38 2.82 1.65 1.67 1.83 1.97 3.34 1.93 2.01 3.22 3.50
0.00081 0.0011 0.00025 0.0035 0.0038 0.0036 0.0075 0.0017 0.0035 0.0011 0.0029 0.0070 0.0061
Figure 2B shows the XANES data for the roots, stems, and leaves grown in hydroponics. The presence of Cr(VI) can be seen in the roots and stems from the small preedge feature. The low amounts of Cr(VI) found in the roots (1.2%) and the stems (6.2%) suggest that the plants have the ability to reduce Cr(VI) to Cr(III). Similar to the agar-grown plants, a Cr(III) acetate structure makes up a major portion of the components of the sample spectrum. In a study using irrigation water, Mishra and co-workers (39) reported a significant difference between Cr(III) and Cr(VI) uptake in paddy (Oryza sativa). This research showed a higher amount of Cr(III) translocated to all portions of the plant compared to the amount of Cr(VI). However, their study did not confirm the valence state of Cr within the plant. Another study conducted by Skeffington et al., who used CrO4-2, found that a small percentage of Cr supplied to barley seedlings (Hordeum vulgare L.) was translocated from the roots to the stems (30). Figure 3 and Table 4 show the Fourier Transformed EXAFS K weighted to 2 and the FEFF fittings of the EXAFS for the model compounds, respectively. The EXAFS of the model compound chromium(III) acetate and chromium(III) sulfide are the model compounds that show a coordination number of 6 and interatomic bonding distances of 1.96 Å, and 2.38 Å, respectively, close to the absorbing chromium atom. This indicates that chromium(III) is in a geometrical arrangement of an octahedron or a distorted octahedron (40). Chromium(III) nitrate shows a first shell coordination of 6, at an interatomic bonding distance of 1.97 Å; however, slightly different EXAFS features are seen in the EXAFS data. The chromium(III) phosphate model compound shows a two shell coordination of two oxygen atoms at 1.93 and 2.01 Å and two
TABLE 5. Results of the FEFF Fittings of the Chromium Plant Samples Showing the Bond, Coordination Number (N), Atomic Radius R (Å), and Sigma Squared (σ2 (Å)) Values sample Cr(VI) agar 80 ppm
bond N (( 1) R (Å) σ2 (Å) (()
Cr-O Cr-O Cr-O Cr(VI) agar 80 ppm mesquite root Cr-O Cr-C Cr(VI) agar 80 ppm mesquite stem Cr-O Cr-C Cr(VI) agar 80 ppm mesquite leaf Cr-O Cr-C Cr(VI) hydroponics 80 ppm Cr-O mesquite root Cr-C Cr(VI) hydroponics 80 ppm Cr-O mesquite stem Cr-C Cr(VI) hydroponics 80 ppm Cr-O mesquite leaf Cr-C
2.7 7.3 5.8 5.8 6.1 5.9 6.8 5.9 5.7 6.0 5.5 5.9 5.9 5.9 6.2
1.62 2.00 3.40 1.98 3.58 1.99 3.45 1.98 3.46 1.98 3.43 1.99 3.56 1.99 3.41
0.0071 0.0046 0.0089 0.00081 0.0033 0.0053 0.0063 0.0032 0.0052 0.0016 0.0041 0.0024 0.0031 0.0024 0.0072
phosphorus shells containing two atoms (each) at approximately 3.22 and 3.50 Å. The chromium(VI) model compound and the supplied chromium(VI) for the experiment (K2Cr2O7) show a coordination number of approximately 4 and approximately 3, respectively, closely bound to the central chromium(VI), indicating a tetrahedral arrangement of oxygen atoms around the chromium atom (41). Figures 4A and Table 5 show Fourier transformed EXAFS K weighted to 2 and the FEFF fittings of the K2 EXAFS for the mesquite hydroponics experiment, respectively. The FEFF fittings of the first shell of the plant tissue sample EXAFS show coordination numbers approximately at 6 indicating that the chromium(III) in the plant tissues is in an octahedral arrangement around the central chromium atom (40). Also, the bonding distances of the chromium(III) oxygen atoms are at approximately the same lengths (1.96 Å for the chromium(III) acetate and 1.98 Å for the chromium(III) laden plant samples). Furthermore, the second coordination shell of the chromium(III) acetate and the chromium(III) laden plant samples are very similar in coordination number and bond lengths. The coordination numbers of the second shell for the chromium(III) acetate is 7 (( 1), and for the chromium(III) laden plant samples they range from approximately 6 (( 1) to 7 (( 1). In addition, the second shell bond lengths of the chromium(III) plant tissue samples and the chromium(III) acetate are approximately at the same length (3.44-3.56 Å). The data from the EXAFS fittings indicate that chromium(III) acetate is the only model compound (used in this study) that is similar in coordination and bond in both shells to the chromium(III) in the plant samples. This further indicates that the chromium(III) is bound to organic acids or sugars within the plant structure. The other model compounds are eliminated due to either the coordination numbers or the interatomic distances between the chromium absorber and the neighboring atoms. In addition, compound stability also limits the possibility of a chromium(III) sulfide, which is highly unstable, from being a good candidate for the compound within the plant. The results from both the EXAFS fittings and the LC-XANES fittings coincide with each other and strengthen the argument that the chromium(III) within the plants is coordinated to sugars and carboxylic acids. The agar sample presents a difficult problem in fitting the EXAFS data as the chromium in the sample exists in two different oxidation states. The agar sample is shown in Figure 4B and the EXAFS and FEFF Fittings are shown in Table 5. In the EXAFS fitting the presence of the chromium(VI) is shown by the relatively short chromium bond length of 1.62 Å with a coordination number of 2.7 (( 1). However the chromium(III) presence is also seen with a chromiumoxygen bond distance of 2.00 Å and a coordination number
FIGURE 4. (A) Fourier transformed EXAFS spectra of the hydroponically grown roots, stems, and leaves with 80 ppm chromium. (B) Fourier transformed EXAFS spectra of the roots, stems, and leaves for the agar experiment with 80 ppm chromium. of 7 (( 1) and a chromium-carbon bond distance of 3.40 and a coordination number of approximately 6 (( 1). This reaffirms the results of the LC-XANES fitting, the presence of chromium(III) compound and a chromium(VI)-oxygen compound. From the results of the fittings, the chromium in the agar sample appears to be present as K2Cr2O7, while the chromium(III) in the plant tissues seems to be bound to an organic ligand such as a sugar or a carboxylic acid. The uptake data indicate that mesquite is a hyperaccumulator of Cr. The XANES data reveal two valuable insights: (1) that mesquite reduces most of the highly toxic Cr(VI) to Cr(III) at the root/media interface by ligand formation with an acetate-type structure and (2) that mesquite plant is able to transport a small percentage of Cr(VI) to the stem from the root, which had not been confirmed previously. In addition, the EXAFS data confirm the LC-XANES fitting data that the chromium coordination in the agar media consists of both chromium(III) and chromium(VI). In addition, the XAS experiment showed that the chromium(III) in the samples exists in either the structure of a sugar or a carboxylic acid complex.
Acknowledgments The authors acknowledge financial support from the National Institutes of Health (NIH) (Grant S06GM8012-30). The authors also thank the financial support from the University of Texas at El Paso (UTEP), Center for Environmental Resource Management (CERM) through funding from the Office of Exploratory Research of the EPA (Cooperative Agreement CR-819849-01-4). Portions of this research were carried out VOL. 37, NO. 9, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1863
at Stanford Synchrotron Radiation Laboratory (SSRL), a national user facility operated by Stanford University. In addition, the authors acknowledge the DOE funded Gateway Program and the Materials Corridor Initiative. The authors also acknowledge financial support from the Southwest Center for Environmental Research and Policy (SCERP) program.
Literature Cited (1) Ajmal, M.; Rafaqat; A. K. R.; Siddiqui, B. A. Water Res. 1996, 30, 1478. (2) Achterberg, E. P.; Braungardt, C.; Moley, N. H.; Elbaz-Poulichet, F.; Leblanc, M. Water Res. 1999, 33, 3387. (3) Ernst, W. H. O.; Nelissen, H. J. M. Environ. Pollut. 2000, 107, 329. (4) Khan, A. G. Envrion. Int. 2001, 26, 417. (5) EPA website: http//es.epa.gov/ncer_abstracts/sbir/91/phase1/ topicf13.html (6) Pausterbach, D. J.; Sheenan, P. J.; Paull, J. M.; Wisser, L. M.; Finley, B. L. J. Toxicol. Environ. Health 1992, 37, 177. (7) Ringenberg, Q. S.; Doll, D. C.; Patterson, W. P.; Perry, M. C.; Yarbro, J. W. South. Med. J. 1988, 81, 1132. (8) Hartwig, A.; Schwerdtle, T. Toxicol. Lett. 2002, 127, 47. (9) Allen, A. Eng. Geol. 2001, 60, 3. (10) Chaney, R. L. In Land Treatment of Hazardous Wastes; Parr, J. F., Marsh, P. D., Kla, J. M., Eds.; Noyes Data Corp.: Park Ridge, NJ, 1983; p 50. (11) Ow, D. W. Res. Conserv. Recycl. 1996, 18, 135. (12) Cobbett, C. S. Curr. Opin. Plant Biol. 2000, 3, 211. (13) Nedelkoska, T. V.; Doran, P. M. Miner. Eng. 2000, 13, 549. (14) Fargasova, A.; Beinrohr, E. Chemosphere 1998, 36, 1305. (15) Chatterjee, J.; Chatterjee, C. Environ. Pollut. 2000, 109, 69. (16) Sharma, D. C.; Chatterjee, C.; Sharma, C. P. Plant Sci. 1999, 111, 145. (17) Peralta-Vida, J. R.; Gardea-Torresdey, J. L.; Gomez, E.; Tiemann, K. J.; Parsons, J. G.; De La Rosa, G.; Carrillo, G. Bull. Environ. Contam. Toxicol. 2002, 69, 74. (18) Gardea-Torresdey, J. L.; Arteaga, S.; Tiemann, K. J.; Chianelli, R.; Pigitore, N.; Mackay, W. Environ. Toxicol. Chem. 2001, 20, 2572. (19) Polette, L. A.; Gardea-Torresdey, J. L.; Chianellli, R.; George, G. N.; Pickering, I. J.; Arenas, J. Microchem. J. 2000, 65, 227. (20) Tiemann, K. J.; Gardea-Torresdey, J. L.; Gamez, G.; Dokken, K.; Cano-Aguilera, I.; Renner, M.; Furenlid, L. R. Environ. Sci. Technol. 2000, 34, 693.
1864
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 9, 2003
(21) Peralta, J. R.; Gardea-Torresdey, J. L.; Tiemann, K. J.; Gomez, E.; Arteaga, S.; Rascon, E.; Parsons, J. G. Bull. Environ. Contam. Toxicol. 2001, 66, 727. (22) ACS Professional Reference Book Series; Kingston, H. M., Jassie, L. B., Eds; American Chemical Society: Washington, DC, 1988. (23) Penner-Haln, J. E. Coord. Chem. Rev. 1999, 190-192, 1101. (24) Gamez, G.; Gardea-Torresdey, J. L.; Tiemann, K. J.; Parsons, J. G.; Herrera, I.; Jose-Yacaman, M. Microchem. J. 2002, 71, 193. (25) Ressler, T. J. Synchrotron Radiat. 1998, 5, 118. (26) Ankudinov, A. L.; Ravel, B.; Rehr, J. J.; Conradson, S. D. Phys. Rev. B 1998, 58, 7565. (27) Ravel, B. J. Synchrotron Radiat. 2001, 8, 314. (28) Reeves, R. D.; Baker, A. J. M. In Phytoremediation of Toxic Metals, Using Plants to Clean Up the Environment; Raskin, I., Ensley, B. D., Eds.; John Wiley & Sons: 2000; p 193. (29) Mishra, S.; Singh, V.; Srivastava, S.; Srivastava, R.; Srivastava, M. M.; Dass, S.; Satsangi, G. P.; Prakash, S. Food Chem. Toxicol. 1995, 33, 393. (30) Skeffington, R. A.; Shewry, P. R.; Peterseon, P. J. Planta 1976, 132, 209. (31) Lytle, C. M.; Lytle, F. W.; Yang, N.; Qian, J.-H.; Hansen, D.; Zayed, A.; Terry, N. Environ. Sci. Technol. 1998, 32, 3087. (32) Cervantes, C.; Campos-Garcia, J.; Devars, S.; Gutierrez-Corona, F.; Loza-Tavera, H.; Torres-Guzman, J. C.; Moreno-Sanchez, R. FEMS Microbiol. Rev. 2001, 25, 335. (33) Paul, E. A.; Clark, F. E. In Soil Microbiology and Biochemistry; Academic Press: NY, 1989; p 1. (34) Mejare, M.; Bulow, L. Trends Biotechnol. 2001, 19, 67. (35) Satofuka, H.; Fukui, T.; Takagi, M.; Atomi, H.; Imanake, T. J. Inorg. Biochem. 2001, 86, 595. (36) Zenk, M. H. GENE 1996, 179, 21. (37) Salt, D. E.; Wagner, G. J. J. Biol. Chem. 1993, 268, 12297. (38) Williams, L. E.; Pittman, J. K.; Hall, J. L. Biochim. Biophys. Acta 2000, 1465, 104. (39) Mishra, S.; Shanker, K.; Srivastava, M. M.; Srivastava, S.; Srivastava, R.; Dass, S.; and Prakash, S. Agric. Ecosyst. Environ. 1997, 62, 53. (40) Engemann, C.; Hormes, J.; Longen, A.; Do¨tz, K. H. Chem. Phys. 1998, 237, 471. (41) Gili, P.; Lorenzo-Luis, P. A. Coord. Chem. Rev. 1999, 193-195, 747.
Received for review August 22, 2002. Revised manuscript received February 24, 2003. Accepted February 25, 2003. ES0208916
