Online Preconcentration-IC-ICP-MS for Selenium Quantification and

Swiss Federal Institute of Aquatic Science and Technology (Eawag), ... information on precise and accurate Se speciation in deficient environments is ...
0 downloads 0 Views 3MB Size
Article pubs.acs.org/est

Online Preconcentration-IC-ICP-MS for Selenium Quantification and Speciation at Ultratraces Markus Lenz,†,‡,* Geerke H. Floor,§,▽ Lenny H. E. Winkel,∥,⊥ Gabriela Román-Ross,§,○ and Philippe F. X. Corvini†,# †

Institute for Ecopreneurship, University of Applied Sciences and Arts Northwestern Switzerland (FHNW), School of Life Sciences, Gründenstrasse 40, 4132 Muttenz, Switzerland ‡ Sub-Department of Environmental Technology, Wageningen University, 6700 EV Wageningen, The Netherlands § Department of Chemistry, University of Girona, Campus de Montilivi s/n., 17071 Girona, Spain ∥ Institute of Biogeochemistry and Pollutant Dynamics, Department of Environmental Sciences, ETH Zurich, CH-8092 Zurich, Switzerland ⊥ Swiss Federal Institute of Aquatic Science and Technology (Eawag), Ü berlandstrasse 133, Postfach 611, 8600 Dübendorf, Switzerland # School of the Environment, Nanjing University, 22# Hankou Rd., Nanjing, 210093, People’s Republic of China S Supporting Information *

ABSTRACT: Selenium (Se) is of key importance to human health with a very narrow concentration range of optimal dietary intake. Due to the inherent analytical challenge linked with the low natural abundance, information on precise and accurate Se speciation in deficient environments is hardly existent. This study presents a novel approach to determine Se species-specifically at ultratraces, by online coupling of a preconcentration (trap) column to an ion chromatography inductively coupled plasma mass spectrometry (IC-ICP-MS) system. It is demonstrated that with this robust and work/time efficient method, the predominant selenium oxyanions, selenite (SeIV) and selenate (SeVI), can be quantified down to 7.3 and 8.3 picogram total Se, respectively, in an overall analytical time of 420 s, only. The applicability for environmental samples was proven on pristine volcanic ashes collected from seven different volcanoes. The high sensitivity of the novel approach allowed to determine speciation in samples that were strongly depleted in total selenium (99%, Sigma Aldrich, Buchs, Switzerland) using a XP2U ultramicrobalance (0.1 μg readability) (Mettler Toledo, Greifensee, Switzerland). Milli-Q Water (>18.2 mΩ cm) was added by weight to yield stock solutions of 1000 μg L−1 Se, each. From these, IC-ICP-MS standards were prepared freshly on a daily base by dilution with Milli-Q water. Selenosulphate was synthesized by reacting solutions of sodium selenite, sodium sulphite and glutathione like previously described.19 Ten minutes after development of a deep red color characteristic for elemental Se, the reaction solution was centrifuged (21 500g, 10 min) and the supernatant diluted for IC-ICP-MS analysis. Standards for the breakthrough experiment were prepared by serial dilution of a 1 M NaCl solution to concentrations between 10−1 M to 10−6 M NaCl and spiked to a final concentration of 100 ng L−1 SeIV and SeVI, each. 2.3. Origin of Samples and Sample Treatment. Pristine volcanic ashes deposited by the following volcanoes were collected (sample coding given in brackets): Chaiten (Chile, CL1 + 2); Santiaguito (Guatemala, GT1); Fuego (Guatemala,

tion of the analytes. For this, analytes can be either retained on solid phase extraction (SPE) materials,10,11 extracted by (dispersive) liquid−liquid12 and cloud point extraction,13 or coprecipitated using insoluble metal hydroxides.14,15 Regarding sensitivity, some of the latter methods indeed achieve low limits of detection (i.e., ng L−1 range). Still, diverse restrictions such as being labor intensive, time-consuming, costly and/or being species unspecific constrain their applicability for routine analysis needed in environmental studies. This study presents a novel preconcentration method that for the first time unites all advantages of being sensitive for ultratraces (i.e., low ng L−1 range), time and work efficient (by coupling online preconcentration to ion chromatography inductively coupled plasma mass spectrometry, IC-ICP-MS) and truly species specific (by detection of the oxyanions as such). Furthermore, the method presented here does not require sample preparation (except for a filtration step), such as adjusting the sample to the eluent pH, which can lead to precipitation and sorption of Se species,16 allowing to study unaltered selenium speciation. The applicability of the method is demonstrated on freshly deposited volcanic ashes, collected from different locations all over the world (Figure 1). Selenium speciation and bioavailability is of particular interest in soils impacted by volcanic ash deposition, since they are considered to be among the most fertile soils in the world.17 This is illustrated by the fact that 9% of the world’s 1990 population lived within 100 km of a historically active volcano.18 The leachates studied here represent a wide range of pH and major anion/cation composition, which illustrates the versatility of the method for environmental speciation of other oxyanions of interest.

2. MATERIALS AND METHODS 2.1. Instrumentation. The IC consisted of a Dionex 2100 system, equipped with an online eluent generator (EGC II KOH), a self-regenerating suppressor (ASRS 300), a guard column (AG17-C, 2 × 50 mm) and an analytical column (AS17-C, 2 × 250 mm) (all Dionex, Olten, Switzerland; http:// www.dionex.com). Samples of varying volumes (40 μL up to 1200 μL) and Se concentrations (10 ng L−1 to 500 ng L−1 selenite, SeIV, and selenate, SeVI) were preconcentrated on an 11989

dx.doi.org/10.1021/es302550b | Environ. Sci. Technol. 2012, 46, 11988−11994

Environmental Science & Technology

Article

Using H2 as reaction gas diminished the typically dominant, plasma derived 40Ar2+ interference and allowed to monitor 80 + Se . However, injection of H2 increased the bromide interference on m/z = 80, that is, 1H79Br+ (compare with retention time of 79Br+, Figure 2). The latter was chromatographically separated from Se species, eluting 25 s before SeIV. 3.2. Preconcentration Performance. Response for all three Se isotopes studied was directly linear to the total amount of SeIV and SeVI injected (i.e., volume × concentration) (Figure 3) when standards of different volume and Se concentration were preconcentrated. Limits of Detection (LOD, 3σ of blank baseline) and limits of quantification (LOQ, 10σ of blank baseline) for both SeIV and SeVI were in the low picogram range, whereby 78Se yielded both the lowest LOD and LOQ (2.3 and 7.3 pg for SeIV, 3.0 and 8.3 pg for SeVI, respectively) and the best linear response (r2 = 0.983 and 0.988 for SeIV and SeVI, respectively, Table 1). The anion exchange capacity of the preconcentration column was determined using NaCl solutions of increasing ionic strength spiked with SeIV and SeVI. Breakthrough of the analytes was observed when more than 0.57 mg Cl− (concentration × volume) was injected. External standard addition of both Se oxyanions showed satisfactory recovery in the three representative samples tested (IS1, GT2, MX3), since the trend line intersection point of the spiked samples matched the peak area of the nonspiked sample with deviations between −10% (SeIV, GT2) and +28% (SeVI, GT2), the average being +11% (Figure 4), which was not statistically different at the 95% confidence interval. 3.3. Selenium Mobilized from Volcanic Ash. Selenium speciation, share of total Se mobilized (Figure 5) and major cation/anion composition (SI Table S2) varied considerably within the leachates of freshly deposited volcanic ashes studied here. Only two samples contained a single Se species, SeIV (JP1, MX4, Figure 5), whereas all other samples represented mixtures of either both oxyanions (CL1, GT2, MX3) or both oxyanions plus an additional unknown species. The highest share of mobilized Se was found in the sample collected at Volcan de Colima (35.8% of total, MX2, Figure 5), whereas in the majority of the samples (10 out of 12), less than 10% of the Se present was leached. In none of the samples, more than 0.7 μg L−1 Se (single species) was found (SI Table S3). Two of the Mexican samples studied (MX2 + 3) were collected on the same sample location with ∼15 months in between sampling (SI Table S1). Within this sampling location, a considerate temporal variation in Se mobilized was observed. Whereas little Se was leached from MX3 (total of 3.5%, main species SeIV), MX2 leachates contained factor ∼10 more Se (total of 35.8%), with the main species being the unknown species. Regarding the latter species, no standard could be obtained (i.e., selenosulphate, SeMet, SeCys, SeIV, SeVI) that matched the observed broad peaks between Se IV /Se VI (selenosulphate matched best, eluting 0.3 min before SeVI, SeCys/SeMet were probably suppressed,22 data not shown). Due to the extremely low concentrations of the unknown species (average 0.17 μg L−1) and the lack of sufficiently sensitive complementary techniques (e.g., ESI-MS), it could not be identified. However, preconcentration allowed to quantify the species to concentrations in the low ng L−1 range (e.g., 18 ng L−1, MX1), which is >40 times lower than in conventional ICP-MS operation, that is, without preconcentration and LC coupling (LOQ 0.78 μg L−1 under the tested instrumental settings, data not shown).

GT2); Eyjafjalla jökull (Iceland, IS1); Etna (Italy, IT1); Sakurajima (Japan, JP1); and Volcan de Colima (Mexico; MX1−4) and stored in air-dried form. Further information regarding the sample origins can be found in the Supporting Information (SI) (Table S1). Total Se content (as nitric acid heat soluble Se) was determined using microwave assisted acid digestion (EPA method 3051) of ground samples followed by ICP-MS analysis (using a hydrogen-pressurized collision-reaction cell with 78Se as monitoring mass and 103Rh as internal standard).20 The method was adapted for mass-limited samples (i.e., using 2.5 mL of HNO3 in quartz vessels). For the certified reference material “volcanic soil JSAC-041” (Japan Society for Analytical Chemistry) a value of 1.28 ± 0.12 mg kg−1 Se was obtained, which fell within the uncertainty of the certified value (1.32 ± 0.27 mg kg−1). For speciation measurements, samples were weighed on the ultramicrobalance (250 mg each; except CL1 50 mg due to limited sample availability), and leached under shaking with Milli-Q Water on a laboratory platform shaker (90 min; solid to water ratio 1:25 w/w) following the standardized methodology for leachates of volcanic ashes.21 Suspensions were then pelleted by centrifugation (10 min, 5000g) and supernatants filtered with a 0.45 μm cellulose nitrate syringe filter (Whatman, Bottmingen, Switzerland). Filtered leachates of volcanic soils were then transferred to polypropylene crimp cap vials (Agilent Technologies). Vials were filled entirely and capped immediately, minimizing sample oxidation by ambient air. All samples were analyzed in duplicates within less than 12h after extraction. For recovery determination, the method of standard addition using 50, 100, and 250 pg SeIV and SeVI as external standards was conducted in three representative samples (injection volume 400 μL) (IS1, GT2, MX3; for full sample description see SI).

3. RESULTS 3.1. Chromatographic Separation of Standards and Potential Interferences. Selenite and selenate were separated within 300 s using the developed eluent gradient, resulting in a total analytical time of 420 s including equilibration of the analytical column for the subsequent injections (Figure 2).

Figure 2. Chromatographic separation of selenite (SeIV) and selenate (SeVI) standards (500 ng/L) from the bromide (Br) resulting in an interference on 80Se (1H79Br+). Monitored isotopes were m/z = 78 (straight line); 79 (dotted line) and 80 (dashed line). 11990

dx.doi.org/10.1021/es302550b | Environ. Sci. Technol. 2012, 46, 11988−11994

Environmental Science & Technology

Article

Figure 3. Calibration for selenite (SeIV) and selenate (SeVI) using different injection volumes (40 − 1200 μL) and SeIV/SeVI concentrations (0−500 ng L−1 each) using online preconcentration IC-ICP-MS, each point corresponding to a single injection. Monitored isotopes m/z=76 (A), 78 (B) and 80 (C); calculated limits of detection (---) and limits of quantification (...).

Table 1. Calibration, Quality of Fit and Related Limits of Detection (LOD) and Limits of Quantification (LOQ) Using Online Preconcentration ICP-MS isotope

species

m

b

r2

LOD [pg]

LOQ [pg]

76

selenite selenate selenite selenate selenite selenate

6.0658 6.1678 15.59 17.004 34.863 39.848

22.514 25.991 48.782 56.266 314.09 242.92

0.976 0.958 0.983 0.988 0.932 0.964

4.6 7.4 2.3 3.0 4.9 6.3

13.6 21.5 7.3 8.3 17.1 20.2

78 80

Figure 5. Selenium speciation in volcanic ashes leachates (for sample coding see SI Table S2). Values above columns represent share of total selenium leached from samples. Error bars based on duplicate measurements.

4. DISCUSSION 4.1. Online Preconcentration IC-ICP-MS for Ultratrace Speciation of Se. This study describes for the first time an instrumental approach that allowed species specific Se quantification down to ultratrace concentrations (i.e., a few pg of Se, Table 1) in a work and time efficient manner. The set up can be considered as true online preconcentration approach, since it consisted of a (reusable) trap column for preconcentration of the analytes, software controlled injection of variable sample amounts and integration of the trap column in the

chromatographic train by an automated switching valve. Using this setup, SeIV and SeVI were separated within 5 min only (Figure 2), resulting in an overall time requirement of 7 min per analysis, without a need of sample preparation other than filtration. Using the latter approach, the response for 76Se, 78Se,

Figure 4. External standard addition for selenite (A) and selenate (B) in three representative samples: IS1 (diamond), GT2 (triangle) and MX3 (square). Samples without spiking in open symbols. Error bars are based on duplicate measurements, monitoring isotope is m/z=78. 11991

dx.doi.org/10.1021/es302550b | Environ. Sci. Technol. 2012, 46, 11988−11994

Environmental Science & Technology

Article

and 80Se was directly proportional to the total amount Se injected to the ICP-MS, that is, volume × concentration (Figures 3 and 4) to more than 250 pg total Se. Thus samples that differ in Se concentration can be analyzed in the same batch straightforwardly by changing preconcentration volumes via the chromatographic software, making dilution of higher concentrated samples obsolete. The IC set up can be equipped with commercially available components to inject volumes up to 8 mL fully automated (here 1.2 mL max); in this case minimal quantifiable SeIV/SeVI concentrations can be calculated as 913 pg L−1 and 1038 pg L−1, respectively (using 78Se for quantification, Table 1) (compare e.g., refs 22, 23, 24, and 25). This represents a significant improvement when compared to LOQ in conventional ICP-MS operation, that is, without preconcentration and LC coupling (under the tested instrumental settings 780 ng L−1, data not shown). The LOD and LOQ achieved in the present study is indeed necessary to determine Se speciation in the environment, for instance in agricultural sciences: studies have shown that relevant field concentrations for Se fertilization were 8 g Se ha−1 (for grain).6 By simplified calculation, assuming that added Se is equally distributed in a depth of 20 cm, this results in a total Se concentration of merely 4 μg L−1 soil. Certainly, from this only a fraction is available for transport or plant uptake.26 In order to assess mobility of Se in agricultural soils, it is therefore necessary to have methods at hand to analyze Se speciation to less than μg L−1 level. Furthermore, it is important to note, that the chromatographic conditions applied here allowed separating bromide from both SeIV and SeVI, which means that the method can be applied to study the speciation of 79Se, otherwise hindered by isobaric interference of 79Br.27 79Se is a long-lived fission product of high radioactive waste28 that is of concern even at ultratrace concentrations. Although 80Se is the most abundant natural Se isotope, the LOD and LOQ determined here were more than 2 times higher in contrast to 78Se and approximately equally high for 76 Se (Table 1). This can be either due to the incomplete removal of the 40Ar40Ar+ interference in contrast to the less abundant 38Ar40Ar+ and 36Ar40Ar+, respectively or to some krypton (80Kr+) contamination of the Ar used (i.e., welding grade). Based on the lower LOD/LOQ and higher linear regression coefficients for both SeIV and SeVI observed here, 78 Se should be used for quantification of Se at ultratraces. This study shows that 76Se can provide an additional control to 80Se for quantification: LOD and LOQ were comparable low, yet 76 Se is not prone to bromine based polyatomic interferences.15 However, 1H75As may interfere with 76Se, so that As species need to be chromatographically separated.29 4.2. Advantages of the Instrumental Set up in Contrast to Current Methods for Ultratrace Speciation. The online preconcentration setup of this study has several advantages in comparison to current methods for ultratrace speciation of redox sensitive elements. For instance, SPE requires time-consuming preconditioning sorbent, sorption, elution and transfer of the analyte to analysis.30 For redox sensitive elements such as Se, time and minimal handling of sample is unquestionably crucial,31 especially in samples of strictly anaerobic environments, since a change of speciation can occur in minutes in the presence of ambient air.12 Using automated injection will decrease bias, which usually affects SPE methods, and improve interlaboratory comparability.

Besides, SPE cartridges are one-time-use, contributing to the high costs of the latter approach. Liquid−liquid extraction is based on selective complexation of SeIV, whereas SeVI is determined after reduction to the latter and by difference from total Se. Again, sample processing and/ or the extraction is time-consuming. In addition, presence of Se species other than SeIV and SeVI, incomplete complexation and/ or reduction can bias the determination of Se speciation.32 The aforementioned disadvantages also apply to coprecipitation methods, since they rely on the selective precipitation of SeIV with for example, lanthanum or manganese hydroxides (and determination of SeVI after reduction to the latter). The instrumental setup of this study made use of an electrolytic suppressor, which results in two major advantages in contrast to other hyphenated IC-ICP-MS,3334 techniques: Firstly, the suppressor removed cations from the mobile phase, hence minimized potential polyatomic interferences with the analytes (e.g., 38Ar40Ca+ in the case of 78Se+15 or 48Ti14N14N+ on 76Se+ 16). Secondly, hydroxide anions are removed after chromatographic separation. In consequence, one can make use of a gradient that hampers conventional IC-ICP-MS analysis due to matrix effects and plasma instability when eluent composition changes.35 This was illustrated by the stable baseline observed upon increase of the hydroxide concentration at 2.5 min (Figure 2). By applying a gradient, considerably shorter analytical times can be achieved in contrast to isocratic separation. The analytical technique presented in this study does not rely on a specific conversion or extraction (see above) yet makes use of a simple strong anion exchange column for preconcentration and a further anion exchange column for separation. Therefore, it can easily be adapted to other anionic trace analytes such as anions of uranium, tellurium, arsenic or chromium. For the latter elements, however, it may be necessary to ensure separation of chloride due to polyatomic interferences (40Ar35Cl+ on 75As+, 35Cl16O1H+ on 52Cr+, respectively).36 One limitation of the present approach may be its fixed anion exchange capacity (here 0.57 mg Cl− preconcentrated). Salt rich (e.g., marine or brine) samples may require small amounts of analyte to be injected, limiting achievable LOD and LOQ. 4.3. Impact of Selenium Speciation on Mobilization to Soils. Soils are still typically classified as being Se deficient or sufficient based exclusively on total Se concentrations, although it has been pointed out that total Se concentrations are not useful to estimate Se bioavailability.3 One factor that may have favored such a simplifying consideration is that until now it has basically not been possible to carry out speciation studies on large numbers of samples in Se depleted natural environments. The analytical approach presented here enables the latter for the first time. Whereas some volcanic ash leachates may be relatively rich in selenium21 and thus accessible to “conventional” (i.e., IC-ICP-MS without preconcentration) speciation methods,33 only preconcentration allows to study those that are extremely depleted in selenium. Previously it has been shown for soils that the total Se content is not necessarily correlated to its mobility (and eventual bioavailability).9 This study confirms this to be true also for volcanic ashes deposited on them. For example, the sample from Sakura-Jima volcano (JP1) contains the highest total Se concentration (1.1 mg Se kg−1), yet the share of total mobilized Se is rather small (1.0%, Figure 5). In contrast, sample MX2 originating from Volcan de Colima, Mexico, had 11992

dx.doi.org/10.1021/es302550b | Environ. Sci. Technol. 2012, 46, 11988−11994

Environmental Science & Technology

Article

one of the lowest total Se contents (0.07 mg Se kg−1), nonetheless mobilized both the highest relative amount of Se (in sum 35.8% of total), resulting in the highest absolute concentration found in all leachates (sum of 1005 ng L−1, Figure 5). Selenium in the samples studied here can be considered as being mostly immobile, since only 2 out of 12 samples showed a mobilization of more than 10% of the total Se present. It is important to note that little Se was mobilized even though high concentrations of major anions (competing for sorption sites) and large differences in sample pH (influencing the number of charged surface sites available for sorption) were found in the leachates (SI Table S2). This may be explained by the residual Se being present either as inner sphere complexes (described for both SeIV and SeVI,37 and references therein) not available for fast-desorption, or surface precipitated and/or coprecipitated solid species (e.g., in volcanic glass, a common component of volcanic ashes). Although not focus of this study, there are several processes expected to take place in volcanic

to collect data of the appropriate quality, in particular in the extended X-ray absorption fine structure (EXAFS) energies. Sequential extraction schemes applied routinely to assess fate of cationic metal species are not appropriate to quantify elemental Se or metal selenides due to insufficient selectivity and extraction efficiency.42 Accordingly, studying Se speciation in relation to applied leaching conditions remains the only analytical means to conclude on original speciation. In summary, the analytical method presented here enables the study of Se speciation in a high number of chemically different samples, necessary to elucidate Se mobilization in particular in Se depleted environments. Only thorough understanding of element speciation at ultratrace level may finally allow to understand mechanisms leading to Se deficiency affecting millions of people.



ASSOCIATED CONTENT

S Supporting Information *

Tables S1, S2, and S3. This material is available free of charge via the Internet at http://pubs.acs.org.



AUTHOR INFORMATION

Corresponding Author

*Fax.: +41 614 674 290; e-mail: [email protected]. Present Addresses ▽

Institute for Reference Materials and Measurements, Joint Research Centre - European Commission, Retieseweg 111, 2440 Geel, Belgium. ○ Amphos 21 Consulting S. L., Passeig de Garcia i Faria 49, 08019 Barcelona, Spain. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS The work was supported by the Swiss National Science Foundation (SNF 200021_126899, 200020_138210 and PP00P2_133619) and the European Commission Sixth Framework Programme Research Training Network AquaTRAIN (Contract No. MRTN-CT-2006-035420). We thank Nick Varley, Gustavo Chigna, Tesuo Kabayashi, Walter D́ Alessandro, Olgeir Sigmarsson and Astrid Bengtsson for providing the volcanic ash samples.

Figure 6. Proposed processes that may transform primary selenium species in volcanic gases (selenium dioxide, SeO2; elemental selenium, Se; hydrogen selenide, H2Se) to solid species (i.e., elemental selenium, Se; metal selenides, MeSe) deposited in volcanic ash.

environments, making the presence of a precipitated solid (i.e., elemental or metal selenide) Se species plausible (Figure 6). Thermodynamic modeling predicts that Se in high temperature environments such as volcanic plumes is present as either Se dioxide, elemental Se or hydrogen selenide, depending on temperature and oxidizing/reducing conditions.38−40 When present as such, hydrogen selenide is highly reactive toward metal cations and readily forms insoluble metal selenides.41 Upon cooling and in contact with water, hydrogen selenide rapidly decomposes to elemental Se, whereas Se dioxide dissolves to form the SeIV oxyanion.40 Selenite again can be reduced by sulfur dioxide (present in volcanic plumes in high concentrations) to from elemental Se and sulfate.40 In consequence, it may well be that insoluble Se species, i.e. elemental Se or metal selenides, contributed at least to parts of the nonmobilized Se found in the samples of this study. Certainly, in the future this hypothesis has to be supported on a large set of samples by alternative, direct speciation methods, above all X-ray absorption fine structure spectroscopy (XAFS). However, these studies will be limited to Se rich volcanic ashes. At concentrations of as little as 0.05 mg Se kg−1 with potential solid species representing only a share of this, even using the most advanced synchrotron beamlines, it is virtually impossible



REFERENCES

(1) Levander, O., Burk, R. Update of human dietary standards for selenium. In SeleniumIts Molecular Biology and Role in Human Health; Hatfield, D. L., Berry, M. J., Gladyshev, V. N., Eds.; Springer: New York, 2007; pp 399−410. (2) Lenz, M.; Lens, P. N. L. The essential toxin: The changing perception of selenium in environmental sciences. Sci. Total Environ. 2009, 407 (12), 3620−3633. (3) Winkel, L. H. E.; Johnson, C. A.; Lenz, M.; Grundl, T.; Leupin, O. X.; Amini, M.; Charlet, L. Environmental selenium research: From microscopic processes to global understanding. Environ. Sci. Technol. 2012, 46 (2), 571−579. (4) Fordyce, F. Selenium geochemistry and health. Ambio 2007, 36 (1), 94−97. (5) Presser, T. S. The Kesterson effect. Environ. Manage. 1994, 18 (3), 437−454. (6) Hamilton, S. J. Review of selenium toxicity in the aquatic food chain. Sci. Total Environ. 2004, 326 (1−3), 1−31. (7) Haug, A.; Graham, R. D.; Christophersen, O. A.; Lyons, G. H. How to use the world’s scarce selenium resources efficiently to 11993

dx.doi.org/10.1021/es302550b | Environ. Sci. Technol. 2012, 46, 11988−11994

Environmental Science & Technology

Article

increase the selenium concentration in food. Microb. Ecol. Health Dis. 2007, 19 (4), 209−228. (8) Fairweather-Tait, Susan J.; Bao, Yongping; Broadley, Martin R.; Collings, Rachel; Ford, Dianne; Hesketh, John E.; Hurst, Rachel Selenium in human health and disease. Antioxid. Redox Signaling. 2011, 14 (7), 1337−1383. (9) Johnson, C.; Ge, X.; Green, K.; Liu, X. Selenium distribution in the local environment of selected villages of the Keshan Disease belt, Zhangjiakou District, Hebei Province, People’s Republic of China. Appl. Geochem. 2000, 15 (3), 385−401. (10) Wake, B. D.; Bowie, A. R.; Butler, E. C. V.; Haddad, P. R. Modern preconcentration methods for the determination of selenium species in environmental water samples. TrAC, Trends Anal. Chem. 2004, 23 (7), 491−500. (11) Gómez-Ariza, J. L.; Giráldez, I.; Morales, E.; Pozas, J. A. Use of solid phase extraction for speciation of selenium compounds in aqueous environmental samples. Analyst 1999, 124, 75−78. (12) Bidari, A.; Zeini Jahromi, E.; Assadi, Y.; Milani Hosseini, M. R. Monitoring of selenium in water samples using dispersive liquid− liquid microextraction followed by iridium-modified tube graphite furnace atomic absorption spectrometry. Microchem. J. 2007, 87 (1), 6−12. (13) Güler, N.; Maden, M.; Bakırdere, S.; Yavuz Ataman, O.; Volkan, M. Speciation of selenium in vitamin tablets using spectrofluorometry following cloud point extraction. Food Chem. 2011, 129 (4), 1793− 1799. (14) Chen, M.; Yang, T.; Wang, J. Precipitate coating on cellulose fibre as sorption medium for selenium preconcentration and speciation with hydride generation atomic fluorescence spectrometry. Anal. Chim. Acta 2009, 631 (1), 74−79. (15) Tuzen, M.; Saygi, K. O.; Soylak, M. Separation and speciation of selenium in food and water samples by the combination of magnesium hydroxide coprecipitation-graphite furnace atomic absorption spectrometric determination. Talanta 2007, 71 (1), 424−429. (16) Floor, G. H.; Iglesías, M.; Román-Ross, G.; Corvini, P. F. X.; Lenz, M. Selenium speciation in acidic environmental samples: Application to acid rain-soil interaction at Mount Etna volcano. Chemosphere 2011, 84 (11), 1664−1670. (17) Satti, P.; Mazzarino, M. J.; Roselli, L.; Crego, P. Factors affecting soil P dynamics in temperate volcanic soils of southern Argentina. Geoderma 2007, 139 (1−2), 229−240. (18) Small, C.; Naumann, T. The global distribution of human population and recent volcanism. Environ. Hazards 2001, 3, 93−109. (19) Zhang, J.; Lu, H.; Wang, X. MS07116 sodium selenosulfate synthesis and demonstration of its in vitro cytotoxic activity against HepG2, Caco2, and three kinds of leukemia cells. Biol. Trace Elem. Res. 2008, 125 (1), 13−21. (20) Floor, G. H.; Iglesias, M.; Roman-Ross, G. Selenium determination in volcanic soils by ICP-QMS: Influence of reaction cell pressurization and methanol addition on the occurrence of spectral interferences. J. Anal. At. Spectrom. 2009, 24 (7), 944−948. (21) Witham, C. S.; Oppenheimer, C.; Horwell, C. J. Volcanic ashleachates: A review and recommendations for sampling methods. J. Volcanol. Geotherm. Res. 2005, 141 (3−4), 299−326. (22) Wallschläger, D.; Roehl, R. Determination of inorganic selenium speciation in waters by ion chromatography-inductively coupled plasma-mass spectrometry using eluant elimination with a membrane suppressor. J. Anal. At. Spectrom. 2001, 16, 922−925. (23) Darrouzès, J.; Bueno, M.; Lespès, G.; Holeman, M.; PotinGautier, M. Optimisation of ICPMS collision/reaction cell conditions for the simultaneous removal of argon based interferences of arsenic and selenium in water samples. Talanta 2007, 71 (5), 2080−2084. (24) Jackson, B. P.; Miller, W. P. Arsenic and selenium speciation in coal fly ash extracts by ion chromatography-inductively coupled plasma mass spectrometry. J. Anal. At. Spectrom. 1998, 13, 1107−1112. (25) Sloth, J. J.; Larsen, E. H. The application of inductively coupled plasma dynamic reaction cell mass spectrometry for measurement of selenium isotopes, isotope ratios and chromatographic detection of selenoamino acids. J. Anal. At. Spectrom. 2000, 15, 669−672.

(26) Tolu, J.; Le Hécho, I.; Bueno, M.; Thiry, Y.; Potin-Gautier, M. Selenium speciation analysis at trace level in soils. Anal. Chim. Acta 2011, 684 (1−2), 126−133. (27) Comte, J.; Bienvenu, P.; Brochard, E.; Fernandez, J.-M.; Andreoletti, G. Determination of selenium-79 in solutions of fission products after pre-treatment by ion exchange chromatography and ETV-ICP-MS. J. Anal. At. Spectrom. 2003, 18 (7), 702−707. (28) Koch-Steindl, H.; Pröhl, G. Considerations on the behaviour of long-lived radionuclides in the soil. Radiat. Environ. Biophys. 2001, 40 (2), 93−104. (29) Pan, F.; Tyson, J. F.; Uden, P. C. Simultaneous speciation of arsenic and selenium in human urine by high-performance liquid chromatography inductively coupled plasma mass spectrometry. J. Anal. At. Spectrom. 2007, 22, 931. (30) Urbánková, K.; Moos, M.; Machát, J.; Sommer, L. Simultaneous determination of inorganic arsenic, antimony, selenium and tellurium by ICP-MS in environmental waters using SPE preconcentration on modified silica. Int. J. Environ. Anal. Chem. 2011, 91 (11), 1077−1087. (31) Gómez Ariza, J.; Morales, E.; Sánchez-Rodas, D.; Giráldez, I. Stability of chemical species in environmental matrices. TrAC, Trends Anal. Chem. 2000, 19 (2−3), 200−209. (32) Lenz, M.; Gmerek, A.; Lens, P. N. L. Selenium speciation in anaerobic granular sludge. Int. J. Environ. Anal. Chem. 2006, 86 (9), 615−627. (33) Uden, P. C. Modern trends in the speciation of selenium by hyphenated techniques. Anal. Bioanal. Chem. 2002, 373 (6), 422−431. (34) B’Hymer, C.; Caruso, J. A. Selenium speciation analysis using inductively coupled plasma-mass spectrometry. J. Chromatogr. A 2006, 1114 (1), 1−20. (35) Ferrarello, C. N.; Ruiz Encinar, J.; Centineo, G.; Garcia Alonso, J. I.; Fernandez de la Campa, M. R.; Sanz-Medel, A. Comparison of three different ICP-MS instruments in the study of cadmium speciation in rabbit liver metallothionein-1 using reversed-phase HPLC and post-column isotope dilution analysis. J. Anal. At. Spectrom. 2002, 17 (9), 1024−1029. (36) May, T. W.; Wiedmeyer, R. H. A table of polyatomic interferences in ICP-MS. At. Spectrosc. 1998, 19, 150−155. (37) Fernández-Martínez, A.; Charlet, L. Selenium environmental cycling and bioavailability: A structural chemist point of view. Rev. Environ. Sci. Biotech. 2009, 8 (1), 81−110. (38) Floor, G. H.; Román-Ross, G. Selenium in volcanic environments: A review. Appl. Geochem. 2011, 27 (3), 517−531. (39) Yan, R.; Gauthier, D.; Flamant, G.; Wang, Y. Behavior of selenium in the combustion of coal or coke spiked with Se. Combust. Flame 2004, 138 (1−2), 20−29. (40) Holleman, A. F.; Wiberg, E. Inorganic Chemistry; De Gryter: New York, USA, 2001. (41) Séby, F.; Potin-Gautier, M.; Giffaut, E.; Borge, G.; Donard, O. F. X. A critical review of thermodynamic data for selenium species at 25°C. Chem. Geol. 2001, 171 (3−4), 173−194. (42) Lenz, M.; Van Hullebusch, E. D.; Farges, F.; Nikitenko, S.; Borca, C. N.; Grolimund, D.; Lens, P. N. L. Selenium speciation assessed by X-ray absorption spectroscopy of sequentially extracted anaerobic biofilms. Environ. Sci. Technol. 2008, 42 (20), 7587−7593.

11994

dx.doi.org/10.1021/es302550b | Environ. Sci. Technol. 2012, 46, 11988−11994