Skip to main content
NCBI home page
EPA Author Manuscripts logoLink to EPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Jul 15.
Published in final edited form as: Environ Pollut. 2017 Aug 3;229:911–921. doi: 10.1016/j.envpol.2017.07.071

Selenium speciation in phosphate mine soils and evaluation of a sequential extraction procedure using XAFS☆

Jessica E Favorito a, Todd P Luxton b, Matthew J Eick a, Paul R Grossl c
PMCID: PMC7363211  NIHMSID: NIHMS1522615  PMID: 28781183

Abstract

Selenium is a trace element found in western US soils, where ingestion of Se-accumulating plants has resulted in livestock fatalities. Therefore, a reliable understanding of Se speciation and bioavailability is critical for effective mitigation. Sequential extraction procedures (SEP) are often employed to examine Se phases and speciation in contaminated soils but may be limited by experimental conditions. We examined the validity of a SEP using X-ray absorption spectroscopy (XAS) for both whole and a sequence of extracted soils. The sequence included removal of soluble, PO4-extractable, carbonate, amorphous Fe-oxide, crystalline Fe-oxide, organic, and residual Se forms. For whole soils, XANES analyses indicated Se(0) and Se(-II) predominated, with lower amounts of Se(IV) present, related to carbonates and Fe-oxides. Oxidized Se species were more elevated and residual/elemental Se was lower than previous SEP results from ICP-AES suggested. For soils from the SEP sequence, XANES results indicated only partial recovery of carbonate, Fe-oxide and organic Se. This suggests Se was incompletely removed during designated extractions, possibly due to lack of mineral solubilization or reagent specificity. Selenium fractions associated with Fe-oxides were reduced in amount or removed after using hydroxylamine HCl for most soils examined. XANES results indicate partial dissolution of solid-phases may occur during extraction processes. This study demonstrates why precautions should be taken to improve the validity of SEPs. Mineralogical and chemical characterizations should be completed prior to SEP implementation to identify extractable phases or mineral components that may influence extraction effectiveness. Sequential extraction procedures can be appropriately tailored for reliable quantification of speciation in contaminated soils.

Keywords: Selenium, Speciation, Sequential extraction procedure, XANES

Graphical abstract

graphic file with name nihms-1522615-f0004.jpg

1. Introduction

Selenium is a trace element found in elevated concentrations in seleniferous regions. In soils where bioavailable forms are present, toxic levels can accumulate in vegetation. In particular, foraging livestock exhibit Se toxicity following ingestion of Se-hyperaccumulators, such as western aster Symphyotrichum ascendens (Lindl.); (Davis et al., 2012, Fessler et al., 2003).

Seleniferous soils throughout the western US are attributed to mining of the Phosphoria Formation and reclamation practices. Associated waste rock contain reduced Se species, selenide (Se(-II)) and elemental Se (Se(0)), which are not immediately soluble (Desborough et al., 1999). Weathering promotes their oxidation to soluble species, selenate (Se(VI)) and selenite (Se(IV)); (McNeal and Balistrieri, 1989). Selenate is more bioavailable, while Se(IV) is typically strongly sorbed onto mineral surfaces, such as clays, oxides, and carbonates (Duc et al., 2006, Goldberg and Glaubig, 1988, Hayes et al., 1987, Masscheleyn et al., 1990, Rajan, 1979, Su and Suarez, 2000).

Selenium bioavailability is a function of soil pH, redox conditions, and mineralogy (Geering et al., 1968, USEPA, 1996). Often, examining total concentrations is considered inappropriate when evaluating bioavailability (Tessier et al., 1979). Instead, sequential extraction procedures (SEP) are employed to estimate soluble, exchangeable, mineralogical, and organic phases using progressively stronger reagents. These fractions will often differ greatly in their potential bioavailability. Sequential extractions are easily accomplished and employ accessible chemical reagents. However, SEPs are often criticized for being operationally defined. Several limitations authors have noted include: issues with target element and solid-phase chemical properties (Martin et al., 1987), lack of reagent specificity for target phases (Jouanneau et al., 1983), partial dissolution of target elements (Sulkowski and Hirner, 2006), and redistribution of solubilized elements (Qiang et al., 1994, Shan and Chen, 1993).

In contrast to SEP protocols and methods, synchrotron-sourced spectroscopic techniques provide detailed insight into trace element mineralogical and organic phase associations in soils. Several authors have demonstrated the usefulness of X-ray absorption fine structure (XAFS) techniques for directly comparing SEP results and using this to evaluate their validity for target phases. Previous studies have indicated mobile species are better estimated using SEPs, which are often overlooked using XAFS (Scheinost et al., 2002). Scheinost et al. (2002) examined Zn in smelter-contaminated soils using a combination of XAFS and a SEP. The SEP failed to identify Zn associated with several mineral phases due to precipitation and nonspecific dissolution during extractions. While the SEP was not able to accurately target particular Zn phases, it was, however, able to better estimate mobilized species not identified by XAFS (Scheinost et al., 2002). Qin et al. (2014) compared results from a SEP modified from Kulp and Pratt (2004) and XAFS for Se in seleniferous soils. They also determined SEP results provided useful information related to bioavailable Se fractions not observed in XAFS spectra. The authors noted distinct differences in conclusions drawn from the Se XAFS vs. SEP data. They observed overestimation of Se(0) and underestimation of Se(-II) by the SEP. In combination, the techniques are complementary and provide improved information for trace element speciation identification and quantification (Qin et al., 2014).

Previous research on soil Se demonstrated synchrotron-sourced X-ray techniques, especially XAFS, are reliable for probing local structures and electronic environments under a wide range of concentrations for bulk samples (Pickering et al., 1995, Koningsberger and Prins, 1988). The main portion of generated spectra used was X-ray absorption near edge structure (XANES) region, which occurs between −20 and + 50 eV below and above the absorption edge, providing information about elemental oxidation states, associated ligands, and immediate coordination environments (Pickering et al., 2013). XANES is ideal for analyzing speciation in soils because it can be completed in situ with minimal pretreatments (Scheinost et al., 2002).

Seleniferous soils within the western US are dominated by Se(-II), Se(0) and Se(IV); (Oram et al., 2008, Pickering et al., 1995, Ryser et al., 2006, Weres et al., 1989). This includes both Blackfoot River, ID sediments and soils from Conda phosphate mine, located near the sites for this current study (Oram et al., 2008, Ryser et al., 2006). This has been confirmed by several synchrotron-based investigations, which have demonstrated the role of geomorphological setting on primary mineral oxidation and subsequent weathering products. For example, within the Kesterson Reservoir, XAFS results indicated varying Se phases based on geomorphological setting. Selenium species present in a former evaporation pond indicated a predominance of Se(0) with Se(IV) as a minor component (Pickering et al., 1995).

This work aimed to improve understanding of Se speciation in soils systems and to increase reliability of SEPs for estimating bioavailable Se fractions. Demonstrating usefulness and shortcomings of SEPs is important from a research perspective because they are widely used techniques with possible limitations that reduce their validity. The combination of SEP and synchrotron-based techniques can provide more reliable information regarding Se solid-phases. An evaluation of this procedure will also prove useful from a regulatory perspective and for future remediation efforts. The objectives of this work were to:

  1. Examine Se solid-state speciation in bulk phosphate mine soils

  2. Evaluate a SEP for solid-phase specificity formulated for Se in soils using XAFS

  3. Suggest SEP modifications using results gleaned from XAFS data, determined by its accuracy for examining target phases

2. Materials & methodology

2.1. Soil samples

Soils were collected from three previously reclaimed phosphate mines near Soda Springs, ID, USA. Mine reclamation was completed prior to 1996. Soil samples were collected at a depth of 0–20 cm. This was chosen due to a shallow calcic horizon below this depth, indicating the presence of a wetting front above the water table. Soil plots examined at Gay Mine by Booth (1980), located near Sites A and B, received annual precipitation levels between 36 and 41 cm. Water ponding was also observed from snowmelt. There is speculation that reducing conditions are possible near our study locations after snowmelt.

The three sites, labeled A, B and C, were reclaimed using middle waste shale material from the Phosphoria Formation. Y-shaped sampling transects were established with center points in locations where high Se levels were presumed to occur, based on the presence of Se-hyperaccumulating vegetation. Transect lines were extended 30 m in three directions, and soil samples were collected every 6 m (16 samples per site). Samples were dried and sieved to 8-mesh (2 mm). Soil mineralogy was determined by powder X-ray diffraction (XRD) using a PANalytical Xpert Pro MPD (Westborough, MA) with Cu kα radiation and a scan rate of 0.02° Θ, from 5 to 90° 2Θ. XPowder software was employed to conduct peak searches (Martin-Ramos, 2004). Reference d-spacing and relative intensities were obtained from the ICDD library card database.

Samples were chosen for XANES analysis based on total Se data, determined through a soils digestion using a mixture of HNO3, H2O2, and HCl (Hossner, 1996). For the intial twenty soils collected, Site A soils were lowest in Se (22.9–37.1 mg kg−1), Site B soils were highest (117–435 mg kg−1), and Site C soils were moderate (3.73–106 mg kg−1).

Prior to XAFS analyses, a SEP described by Amacher (2010) and modified from Martens and Suarez (1997) was employed (Table 1). Extracts were analyzed for Se using ICP-AES (SPECTRO Analytical Instruments, Inc., Mahwah, NJ). This procedure accounts for 1) water-soluble, 2) PO4-extractable, 3) carbonate-associated, 4) amorphous Fe-oxide-associated, 5) organic, and 6) residual/elemental Se phases. To substantiate results from whole soils analyses, six soil samples (seven subsamples each) from the initial twenty soils were re-analyzed under a stepwise SEP approach using XANES (Table 2A, Table 2B).

Table 1.

Average relative abundance for Se in six fractions of a sequential extraction procedure (SEP).

F1a F2 F3 F4 F5 F6
Min Max Mean Min Max Mean Min Max Mean Min Max Mean Min Max Mean Min Max Mean
Relative Distribution (%)
A 0 4.8 1.4 0 3.3 0.7 0 10.2 4.6 0 11.7 2.4 14.1 19.4 25.8 66.1 81.5 71.6
B 0 2.6 0.9 0 1.1 0.5 0.4 6.7 3.9 0 6.7 4.6 2.8 11.3 25.7 64.5 91.4 76.0
C 0 1.6 0.9 0 0.8 0.2 0 6.7 3.8 0 9.0 2.9 0 14.3 24.8 61.4 100.0 80.5
Open in a new tab
a

F1 = Water-soluble; F2 = PO4-extractable; F3 = Carbonate associated; F4 = Amorphous Iron oxide associated; F5 = Organic; F6 = Residual.

Table 2A.

Sequential extraction procedure (SEP) summary for selenium in calcareous soils.

Fraction Extractants Conditions
Water-Soluble Deionized water Shaken 2 h at 120 rpm, 20 °C
PO4-Extractable 0.01 M KH2PO4/K2HPO4 buffer Shaken 2 h at 120 rpm, 20 °C
Carbonate pH 5, 1.0 M NH4CH3CO2 Shaken 24 h at 120 rpm, 20 °C
Amorphous Fe/Al Hydrous Oxides 0.2 M C2H8N2O4 and H2C2O4 Shaken 2 h at 120 rpm in the dark, 20 °C
Crystalline Fe/Al Oxides 0.04 M Hydroxylamine HCl in 25% (v/v) HAOc Heat to 96 °C for 6 h with occasional agitation
Organic 0.1 M K2S2O8 Heated to 90 °C for 2 h
Residual Concentrated HNO3 30 min in water bath at 95 °C, diluted to 25 mL, and heated again for 1.5 h
Open in a new tab

Table 2B.

Stepwise scheme for sequential extraction procedure (SEP) employed for XANES analysis.

Subsample Extraction
1 F1a
2 F1, F2
3 F1, F2, F3
4 F1, F2, F3, F4
5 F1, F2, F3, F4, F5
6 F1, F2, F3, F4, F5, F6
7 F1, F2, F3, F4, F5, F6, F7
Open in a new tab
a

F1: Water-soluble; F2: PO4-Extractable; F3: Carbonate-Associated; F4: Amorphous Fe-Oxide-Associated; F5 = Crystalline Fe-Oxide-Associated; F6 = Organic Se; F7 = Elemental Se.

Prior to XANES analyses, samples were air-dried and sieved to 325-mesh (<45 μM). Twenty whole soils were analyzed by bulk in-situ XANES. The six soils used in the second analysis were prepared similarly as whole soils.

2.2. XAFS reference compounds

Selenium salt reference compounds, l-selenomethionine, l-selenocysteine, and gray elemental Se were purchased from Alfa Aesar and Acros Organics. Selenium-ferrihydrite reference compounds were prepared through sorbing 0.01 M sodium selenite and selenate at pH 4.5 onto ferrihydrite in batch reactors. Two-line ferrihydrite was synthesized using the procedure described by Schwertmann and Cornell (2000). Surface area was determined from a BET N2 isotherm (Quantachrome Autosorb Automated Gas Sorption System Model Autosorb 1, Quantachrome Instruments, Boynton Beach, FL) and was 258 m2 g−1. Reference compounds for Se(IV) and Se(VI) sorbed onto calcite were prepared in batch reactors by adding 0.01 M sodium selenite or selenate at pH 8.0 with ground calcite (Excalibur Mineral Company, Charlottesville, VA). Reference compounds for Se(IV) and Se(VI) co-precipitated with calcite were prepared using a procedure described by Aurelio et al. (2010). Monoclinic red Se was synthesized from a procedure described by Ebels et al. (2006). Reference compound XANES spectra are shown in Fig. 1A.

Fig. 1.

Fig. 1.

Open in a new tab

A and B. (A) Selenium k-edge XANES spectra for all reference compounds used and (B) 6 subsamples from sites A (SA), B (SB), and C (SC) that were extracted for 6 phases of Se. (1) = water-soluble, (2) = PO4- extractable, (3) = carbonate associated, (4) = amorphous iron oxide associated, (5) = crystalline iron oxide associated, (6) = organic selenide.

2.3. Sequential extraction analysis

Six soils (divided into seven subsamples) were re-evaluated to examine the specificity of extractions for targeted phases. A progressive-approach to the SEP described by Amacher (2010) was employed, accounting for six fractions, with modifications. An extraction described by Tessier et al. (1979) for Se associated with crystalline Fe-oxides using hydroxylamine hydrochloride was supplemented into the procedure (Table 2A). Between extractions, residues were washed with 95% ethanol. After each extraction scheme, subsamples were analyzed using XANES to determine whether particular phases were targeted.

2.4. X-ray absorption spectroscopy

Synchrotron analyses were completed at DND CAT (DuPont-Northwestern-Dow Collaborative Access Team) beamline 5-BM-D, equipped with a double Si (111) monochromator, at the Advanced Photon Source, Argonne National Laboratory (Lemont, IL). Samples were compressed into 13 mm pellets using an IR press and sealed between Kapton® tape before analysis. Spectra were obtained at the Se k-edge energy of 12658 eV, and scans were collected from −200 to 1000 eV above the k-edge. Data collection was completed in fluorescence mode using two solid-state Vortex ME4 Silicon Drift detectors. The synchrotron was operated at 7.0 GeV at a nominal 100 mA top-up fill current. Incident beam energy was calibrated to a Se foil standard.

XANES spectra (3–5 scans) were merged, calibrated, and normalized using the software, Athena (Ravel and Newville, 2005). Normalization was completed through fitting a first order polynomial from −150 to 30 eV below the edge and either a second or third order polynomial to spectra from 50 to 150 to 300–500 eV above the edge. Selenium absorption edges (Eo) were determined using the maximum of the first derivative of the signal. Using Athena, linear least-squares combination fitting (LCF) of the XANES region was applied to soil spectra along with a combination of reference compounds. The energy range used for the fit was −20 eV below to +30 eV above the Se edge. The quality of fits of reference compounds to soil spectra were indicated by goodness-of-fit parameters, R-factor and χ-square.

Linear combination fitting provides quantitative information through fitting combinations of reference compounds to spectra. For unprocessed soils, fits were compared to ICP-AES results from a SEP described by Amacher (2010). Using these results, fits were analyzed first with reference compounds consistent with the SEP. Reference compounds with low contributions (<4%) were removed from fits. Goodness-of-fit for soils was evaluated by lowest χ-square and R-factor values with a minimum of reference compounds identified in the ICP-AES SEP results. If fits were considered to be poor, other reference compounds not identified in the SEP were incorporated.

3. Results

3.1. Soil chemical properties and mineralogy

Soil pH was alkaline (6.76–8.26) with moderate proportions of total organic (TOC) (25.2–58.7 g kg−1) and inorganic carbon (TIC) (3.3–10.3 g kg−1); (supplementary data). Soil pH was compared using one-way ANOVA analysis and Tukey’s multiple comparison method in JMP Pro 11.0.0 statistical software (SAS Institute Inc., Cary, NC). Overall, Sites A and B average soil pH values were statistically more alkaline than Site C soils.

Soils generally shared similar mineralogy, comprising of quartz, carbonate-fluorapatite, dolomite, calcite, and muscovite. Smaller peak intensities related to orthorhombic Se(0) were also indicated. Minerals present in few percent amounts or less are typically not detected by XRD (Schultz, 1964). This indicates Se(0) is likely a major Se phase in soils. Carbonate-fluorapatite was the only phosphate mineral present in analyzed soils. Iron oxide minerals were difficult to identify in diffractograms. Iron concentrations in soils from the study sites were lower than average US soil concentrations (Shacklette and Boerngen, 1984).

Soils were derived from middle waste shale, with previous work indicating Se contents ranging from 193 to 216 mg kg−1 (Ryser et al., 2005). Using XANES, unweathered shale samples were determined to contain ferroselite and selenocysteine with some seleniferous pyrite/marcasite. No samples contained Se(0), although Perkins and Foster (2004) and Grauch et al. (2004) identified this using SEM-EDS in middle waste shale (Ryser et al., 2005).

3.2. XANES analysis on whole soils

XANES whole soil spectra indicated Se(0) is the most abundant phase. Both elemental forms (red and gray) would be considered to comprise the SEP residual fraction. XANES results revealed Se(0) comprised between 25 and 80% of total Se in soils (Table 3). Comparably, this is lower but close to percentages indicated by the SEP (Table 1; 61–100%) and is not unexpected. Elemental Se was identified in X-ray diffractograms. For many soils, organic Se(-II) comprised the second largest fraction in both SEP (0–20%) and XANES (6–31%) analyses. Results from XANES indicated many of the soils examined contained Se(IV); (Table 3). This was comparable for both Se(IV) associated with ferrihydrite (5–41%) and co-precipitated with calcite (4–30%). Selenate was not observed in any soils examined. Sequential extraction results suggest this is because relative abundances of water-soluble Se, which is the likely fraction where Se(VI) would be observed, are too low (<4%) to be reliably quantified in the XANES region.

Table 3.

XANES linear least-squares combination fits for 20 samples with compounds used to create reference spectra.

Sample Reference compounds
SeIV-Fha SeIV-Ca SeVI-Ca SeIV-Co-Ca Se-Cys. Se-Me. Se Gray Se Red R-Factor
Relative abundance
Site A (n = 5)
SA NE2 0.29 - - 0.21 - 0.10 0.10 0.30 0.0039
SA NE5 0.29 - - 0.20 - 0.06 0.22 0.22 0.0048
SA NW1 0.41 - - 0.23 0.10 - - 0.25 0.00097
SA NW3 0.39 0.30 - - - - 0.15 0.15 0.0054
SA Center 0.08 0.24 0.13 0.17 - - 0.15 0.24 0.0016
Site B (n=9)
SB NE2 0.05 - - 0.06 - 0.17 0.33 0.40 0.00020
SB NE3 0.05 - - 0.09 0.08 0.13 0.25 0.41 0.00019
SB NE5 0.15 - - 0.07 0.18 - 0.20 0.39 0.000018
SB NW1 0.07 - - 0.11 0.30 - 0.18 0.35 0.000015
SB NW2 0.06 - - 0.08 - 0.21 0.32 0.34 0.00023
SB NW4 0.07 - - 0.08 0.21 - 0.29 0.35 0.00013
SB S2 - - - 0.10 0.31 - 0.23 0.36 0.00026
SB S5 0.20 - - 0.13 0.15 - 0.16 0.35 0.00047
SB Center 0.06 - - 0.08 - 0.18 0.34 0.34 0.00027
Site C (n=6)
SC SW1 0.24 - - 0.13 0.17 - 0.21 0.24 0.00082
SC NE1 0.20 - - - 0.18 - 0.29 0.38 0.00069
SC NE4 - - - 0.30 - 0.11 0.75 0.05 0.099
SC NW2 0.24 - - 0.15 - 0.20 0.13 0.29 0.00069
SC NW5 0.29 - - 0.04 0.09 - 0.31 0.28 0.0020
SC Center 0.24 0.06 - - 0.21 - 0.24 0.26 0.00016
Open in a new tab
a

Fh = Ferrihydrite; Ca = Calcite; Co-Ca = Co-precipitated with calcite; Cys. = Cysteine; Me. = Methionine.

XANES analyses for two soils from Site A and one from C also showed the presence of Se associated with calcite not identified in the SEP (Table 3). XANES results indicated four samples from Site A, one from B, and one from C showed Se associations with ferrihydrite present in XANES data not indicated by the SEP. XANES spectral lines confirmed the presence of sorbed forms of Se(IV) in soils where they were previously unaccounted. For particular soils, goodness-of-fit parameters greatly improved following Se(IV)-ferrihydrite reference incorporation into LCFs. For example, the initial R-factor for Site A point NW1 was 0.0025. With inclusion of Se(IV)-ferrihydrite, this resulted in an average R-factor improvement of 61.2% and a relative standard deviation of 64.7% (R-factor = 0.00097).

Generally, there was improved agreement for Site C between the two techniques than for Sites A and B (Fig. 2A–D). Secondly, there was little agreement for carbonate and Fe-oxide Se phases. XANES results were higher in adsorbed/co-precipitated forms of Se than SEP results indicated, particularly for Sites A and B.

Fig. 2.

Fig. 2.

Open in a new tab

A-E. Shows agreement between sequential extraction procedure (SEP) and XANES relative abundances (%) as a function of total Se for soils from Sites A (SA), B (SB), and C (SC). Each plot corresponds with (A) carbonate, (B) Fe-oxide, (C) organic and (D) residual fractions. (E) Comparison of selenium relative abundances determined by a sequential extraction procedure and XANES in a 1:1 ratio for carbonate, iron oxide, organic, and residual forms of Se.

Comparable abundances were noted for Se(IV) related to ferrihydrite and co-precipitated with calcite. Abundances were related to SEP results to evaluate similarities and differences between predicted Se speciation as a function of total Se (Fig. 2A–D) and directly (Fig. 2E). Initial evaluations of Fig. 2A–E reveal differences in predicted speciation based on results from the two techniques used. XANES LCFs indicate Se(IV) is associated with several solid-phases, either adsorbed onto Fe-oxides and/or co-precipitated with carbonates (Table 3). However, Fig. 2E highlights inconsistencies between ICP-AES SEP and XANES results, indicating SEPs greatly under-predict carbonate and Fe-oxide-associated Se. Comparisons were also further evaluated using simple linear regression for all soils examined.

Regression analyses were completed for twenty soils examined to identify whether SEP relative abundance results from ICP-AES and XANES were predicative of one another. For Fe-oxide associated phases, results indicated the techniques were not predictive (p-value > 0.10) of one another overall (R2 = 0.13; p-value = 0.121). Additionally, no significant relationships were identified for calcite (R2 = 0.02; p-value = 0.556) or organic Se (R2 = 0.11; p = 0.158) phases as well. A significant relationship between the two techniques was, however, indicated for residual/elemental Se (R2 = 0.37; p-value = 0.005).

Differences in predicted abundance of complexed/co-precipitated phases were not consistent for the three sites. XANES spectral features indicate these species occur mainly as Se(IV) sorbed to ferrihydrite and co-precipitated with calcite. Site B soils showed the most agreement between results from ICP-AES SEP and XANES analyses, although most p-values determined from linear regression were insignificant.

The discrepancy between predicted residual/elemental fraction was consistent over the range of Se levels investigated (Fig. 2D). More agreement between the techniques is noted for Sites A and C and somewhat for B soils. However, according to Fig. 2E, the SEP consistently over-predicted Se in residual/elemental fractions. The second most abundant phase in both the SEP and XANES LCFs were Se(-II) compounds. However, there was no trend identified between the two methods for this phase (Fig. 2C and E).

Agreement between SEP and XANES results appears to be concentration dependent, identified above 300 mg kg−1 for carbonate and Fe-oxide-Se. Regression analyses indicated a strong linear relationship (R2 = 1; p-value = <0.0001) between the two techniques for both Se phases. Relationships were also identified for Se(-II) (R2 = 0.43; p-value = 0.546) and residual/elemental Se (R2 = 0.45; p-value = 0.531), but were considered insignificant.

3.3. XANES analysis of a sequential extraction procedure

Whole soils analyses indicated discrepancies between ICP-AES SEP and XANES results, particularly for sorbed Se. Therefore, soils were extracted using a modified version of the Amacher (2010) procedure and analyzed further using XANES. Following each SEP step, Se phase removal after extraction and speciation were evaluated for two soils from each site. This was accomplished to evaluate the effectiveness of the SEP at removing target mineral and organic phases.

XANES spectra for select soils after each step of the modified extraction scheme are shown in Fig. 1B. For all soils examined, changes in spectral features were not observed following first (water-soluble) and second (PO4-extractable) extractions. This is likely because of their negligible concentrations (≤1% of total Se), as demonstrated by the SEP. Therefore, the spectra for soils extracted for the first two SEP fractions were ignored. Speciation results from Favorito et al. (2017) indicated the two fractions are mainly composed of oxidized Se. Also, a significant relationship was noted for Se(VI) and water-soluble Se (Favorito et al., 2017). Because of low concentrations indicated by the SEP, it is unlikely XANES spectra were sensitive enough for accurate identification. XANES LCFs and spectral lines also indicate Se(IV) is the dominant oxidized/sorbed Se species present in all soils examined.

3.4. Site A

Site A soils were lowest in total Se, with concentrations at points NW1 and NE5 of 37.1 and 35.7 mg kg, respectively. Sequentially extracted and whole soil XANES LCFs indicate the presence of Se associated with carbonates in both Site A soils. X-ray diffraction patterns and total inorganic carbon (TIC) results also confirmed the presence of carbonate minerals. Peak intensities in diffractograms indicate dolomite is the predominant carbonate mineral present with a less intense calcite peak. Little changes were observed in XANES LCF following the third extraction for carbonate-associated Se.

Following the fourth extraction for amorphous Fe-oxide-associated Se, XANES LCFs indicated the relative abundance of Se(IV) sorbed to ferrihydrite was not affected by the extraction for Site A point NW1 soils. Fig. 3A and B demonstrate Se concentrations in this fraction were consistent between whole soils and soils extracted by SEP Steps 3 and 4. After Step 5, Se(IV)-ferrihydrite reference compound was reduced even further and slightly more after Step 6 for point NE5 soils.

Fig. 3.

Fig. 3.

Open in a new tab

A-F. Selenium fractionation in whole soil and sequential extraction procedure (SEP) fractions (F3–6) identified in XANES linear combination fitting (LCF) for soils from Sites (A) A NW1, (B) A NE5, (C) B NE3 (D) B Center Point (E) C NE1 (F) C NW5.

aFh = Ferrihydrite; Co-Ca = Co-precipitated with calcite.

Selenide removal was predicted to occur during Step 6 using potassium persulfate (K2S2O8). Fig. 3A and B indicate no reference compound removal from fits had occurred for both Site A soils during this step, compared to whole soils. When reference compounds were removed from LCFs, goodness-of-fit parameters increased, indicating no fit improvement.

Elemental Se species within residual fractions were consistent for all soils throughout the extraction scheme. X-ray diffractograms also confirmed the presence of gray Se(0). XANES analyses indicate that, following extraction with concentrated HNO3, little Se was present in soils following this step. However, spectral lines show the remaining Se was converted from reduced species to Se(VI), which was likely an artifact caused by oxidation with HNO3 during extraction. Consequently, Se speciation could not be accurately determined.

3.5. Site B

Site B soils were generally highest in total Se. Concentrations at points NE3 and the center were 229 and 350 mg kg−1, respectively. The pH was alkaline, due to the presence of carbonate minerals confirmed using XRD. X-ray diffractograms indicate intense dolomite peaks with calcite also present.

XANES fit results for Site B soils appear to be more comparable to ICP-AES SEP findings. Fig. 3C and D indicates Se associated with carbonates is removed for both soils. XANES results suggest Se(IV) co-precipitated with calcite is present. Removal of carbonate associated Se appropriately occurs for point NE3 soils during Step 3. While most Se is removed from center point soils, XANES results indicate further and complete removal occurs during Step 4.

For Se associated with amorphous and crystalline oxides, complete removal was achieved for both Site B soils (Fig. 3C and D). Fits indicated partial removal of this fraction occurred during Step 4 with complete removal achieved during Step 5. Organic Se(-II) is also consistent throughout the extraction procedure, indicating a lack of removal.

3.6. Site C

Site C soils were intermediate in total Se, compared with previous sites. Concentrations in point NE1 and NW5 were 96.0 and 106 mg kg−1, respectively. Similar to Site A soils, a number of discrepancies were noted for Se phases determined in whole soil analyses and those extracted via the SEP analyzed in the XANES region. Fit results indicated point NE1 soils did not contain carbonate-associated Se, which is in accordance with SEP and XRD results. No obvious diffraction patterns for dolomite or calcite were noted for soils from this point. A prominent dolomite peak was present for point NW5 soils with a smaller calcite peak, demonstrating the presence of sorptive surfaces. The Se(IV) co-precipitated with calcite standard was used in the LCF for extracted point NW5 soils. The relative abundance from the LCF (4%) was near what was indicated by SEP ICP-AES results (5%) for Se associated with carbonates.

No carbonate associated Se was found in point NE1 soils and was inappropriately removed for point NW5 soils. Relative abundances of Se associated with Fe-oxides were higher in LCFs than SEP results indicated for both Site C soils. When compared to whole soils, both extracted soils indicate large removals of this fraction during Step 4 (Fig. 3E and F). Further removal was noted in soils from point NE1 during Step 5, however, it is possible this may not be present because of reduced sensitivity of XANES at low Se concentrations.

Relative abundances of Se(-II) decreased during Step 5, with complete removal during Step 6 for point NW5 soils (Fig. 3F). Soils from point NE1 showed a reduction in Se(-II) abundance following Step 6, but no complete removal (Fig. 3E). Following Step 6 extractions, Se reference compounds were manually removed from the LCF for soils from point NE1. This caused a 24% increase in the R-factor, signifying no fit improvement and that Se(-II) compounds may still be present.

4. Discussion

4.1. XANES analysis on whole soils

XANES results indicate Se(0) predominated in soils. Elemental Se is also the most abundant species derived from phosphatic shale (Stillings and Amacher, 2010). Additionally, soils contained Se(-II), which may occur as organoselenides and/or ferroselite (an iron selenide), which is also present in phosphatic shale (Stillings and Amacher, 2010). Ryser et al. (2006) speculated because these soils are young and reduced species are present, the existence of Se-containing primary minerals in soils are likely. The oxidation of the species, Se(-II) and Se(0), to oxyanions is also a kinetically slow process (Ryser et al., 2006). Bound and co-precipitated forms of Se were also identified in spectra. Sources of organic Se can also be derived by plants through reducing inorganic Se species. XAFS analyses for the plant species, wheat (Triticum aestivum) and Indian mustard (Brassica juncea), indicated high amounts of selenate in leaves (Eiche et al., 2015). Methylselenocysteine was also detected in wheat. Selenate can be reduced to selenite followed by selenocysteine (de Souza et al., 1998, Ng and Anderson, 1978). Selenocysteine can be further incorporated into selenoproteins or transformed to organic and elemental Se (Pilon et al., 2003, Shrift, 1969). Using XANES, El Mehdawi et al. (2015) indicated soils next to hyperaccumulator and nonaccumulator plants were elevated in organic Se. This indicates plant litter and microorganisms may influence soils by providing a source of organic Se. These occurrences may further explain Se composition in the soils examined in this work.

The presence of Se(IV) in XANES LCFs coincides well with μ-XANES work from Ryser et al. (2006) and μ-SXRF work from Oram et al. (2008) who observed a predominance of Se(IV) with lower amounts of Se(VI) in western US soils. Because of oxidizing conditions found in semi-arid regions, Se(VI) was thought to predominate, although this has not been confirmed. Ryser et al. (2006) suggests Se(VI) is rapidly lost in soils through leaching, yet plants throughout this region are accumulating very high concentrations of Se. Based on XANES LCF data, Se(VI) was not observed in any of the soils examined. It was further speculated soil pH within this region may cause Se(IV) to become more available for plant uptake. Above-normal concentrations of phosphate and the presence of dissolved organic carbon (DOC) in soil rhizosphere environments may increase Se solubility through surface site competition with Se(IV); (Dynes and Huang, 1995, Dynes and Huang, 1997, Geelhoed et al., 1998, Grafe et al., 2001, Grafe et al., 2002, Ryser et al., 2006).

Associations with calcite are expected and are in accordance with soil mineralogy and pH. In calcareous soil systems (pH > 7), Se(IV) should be primarily sorbed onto calcite, although the number of sorption sites are less than those on Fe-oxides (Cowan et al., 1990, Goldberg and Glaubig, 1988). Quantities of Se associated with Fe-oxides in XANES were in contrast to results from SEP extractions, which indicated Fe-oxide-sorbed Se made up 10% or less of the total Se present in any of the soils investigated. Similar results for carbonate-Se were also noted.

Because of discrepancies between XANES LCF and SEP results, it was speculated other sources (i.e. crystalline forms) of Fe-oxide–sorbed Se were present and unaccounted for by the Amacher (2010) procedure. Ryser et al. (2006) also indicated reference spectra for Se sorbed to amorphous and crystalline Fe-oxides were indistinguishable in the XANES region.

Soils with lower total Se under-predict elemental and organic Se and over-predict bound or co-precipitated Se(IV) species (in XANES). However, linear regression analyses for all soils identified a significant relationship between SEP Se(0) fractions identified by ICP-AES and XANES. This indicates the two methods are predicative of one another for this phase but may also be related to XANES sensitivity. Because Se(0) is considered the most abundant phase, it can be more easily identified in XANES spectra. Insignificant relationships between carbonate, Fe-oxide, and organic Se phases were also indicated in regression analyses. The lower concentrations of these secondary phases may cause difficulties in identifications using XANES. Weak relationships also suggest XANES may be identifying target Se phases that were incompletely removed using the SEP.

We have established a Se concentration dependence for XANES LCF results. Relationships between XANES and SEP results from ICP-AES were significant above concentrations of 300 mg kg−1. It is important to note that the total Se present in soils will have an impact on the ability to resolve specific Se species by analysis of the XANES data. As Se levels decrease, the subsequent decrease in signal to noise ratios will limit the ability to effectively detect all of the Se present by the XANES LCF. These aforementioned limitations may provide further explanations for discrepancies between XANES and SEP results. Data also suggests SEP limitations are highly likely and are possibly related to incomplete mineral phase dissolution during extraction or lack of reagent specificity for target phases.

4.2. XANES analysis of a sequential extraction procedure

Differences in removal efficiency indicated in XANES spectra were speculated to be due to variations in physicochemical properties (Supplementary Data). In comparison, Site A soils were lowest, Site B soils were highest, and Site C soils were moderate in total Se. Lower Se levels will affect XANES sensitivity, especially for individual fractions.

Little changes in spectra following the third extraction suggests Se(IV) is co-precipitated with calcite, which is identified in XANES LCFs, and is not accounted for or effectively removed by the SEP in Site A soils. In Site C point NW5 soils, this Se fraction was reduced during an inappropriate step, which should occur during Step 3. Unlike previous sites, this phase was appropriately removed in both soils from Site B. We speculate it is also possible Se could be co-precipitated with other minerals or trace elements that decrease its solubility. Selenite can be strongly bound to or co-precipitated with Fe-oxides (Strawn et al., 2002). Merrill et al. (1986) noted significant quantities of Se(IV) can be co-precipitated with Fe-oxyhydroxides. Stillings and Amacher (2004) also found evidence of this in wetland sediments receiving phosphate mine drainage. This may provide additional insight into XANES and SEP discrepancies. If Se(IV) is co-precipitated with Fe-oxides, this may hinder dissolution and recovery of Se during Steps 4 and 5. However, this indicates Se is likely more recalcitrant in soils and is not adequately solubilized during the SEP. During extraction, changes in experimental conditions may also occur. The altered chemistry during extraction may not be sufficient for adequate recovery.

Lack of appropriate carbonate fraction removal may be caused by extractability hindered due to occlusion via co-precipitation or incomplete phase dissolution. No current work has examined Se sorption onto dolomite. Previous work reveals calcite is a large sink for Se (Aurelio et al., 2010). Because of its dynamic solubility, calcite sorption and co-precipitation are more likely to occur in soils, even with lower amounts of calcium carbonate (Heberling et al., 2014). When soil solutions are supersaturated with respect to calcite, co-precipitation dominates through adsorption and entrapment events. Below calcite saturation, surface exchange and complexation mechanisms will occur (Heberling et al., 2014). Co-precipitated Se phases are more occluded compared to Se adsorbed to surfaces. Because diffractograms identified calcite in soils, which is pedogenic, co-precipitation with Se is a likely mechanism in soils from the three sites. Our LCFs also suggest this reference compound should be utilized in reference spectra for many of the soils examined. The presence of this phase could explain the higher Se abundance in this fraction in XANES fits compared to SEP ICP-AES data. For dolomite-rich soils, Sulkowski and Hirner (2006) observed a rise in pH above what is generally used for carbonate dissolution. Due to subsequent desorption decreases and re-adsorption increases of target elements, the authors recommended monitoring pH and repetition of using acetic acid to achieve constant pH. Complete dissolution of carbonates also promoted a 10-fold higher release of Fe from oxides following hydroxylamine hydrochloride extraction (Sulkowski and Hirner, 2006). Therefore, a lack of carbonate mineral dissolution could explain the lack extractability of Se in our soils and also, the higher abundance of carbonate-associated Se reference compounds in LCFs compared with SEP data. This is imperative to note for future work with SEPs for carbonate-rich soils.

Reductions in amorphous Fe-oxide-Se were observed for Site A point NW1 soils following Step 4 with no further reductions afterwards. Further removal of Se following Step 5 for point NE5 soils signifies associations of Se with crystalline Fe-oxides. Although not indicated in X-ray diffractograms, Fe-oxide surfaces are likely present. Complete and appropriate removal in both soils from Site B and one from Site C at point NE1 was noted. Higher relative abundances for Se associated with Fe-oxides than SEP and removal during Steps 4 and 5 suggests additional forms of Fe-oxides play a part in Se sequestration. Therefore, it is likely this fraction should be incorporated into SEPs formulated for soils in this region.

Complete removal of Se(-II) using K2S2O8 was only noted for Site C point NW5 soils. The reasoning for this is unclear, considering SEP ICP-AES results indicate point NW5 was highest in Se for this fraction compared with the five other soils examined. The Se(-II) phase is likely still present in these soils, perhaps due to lack of reagent specificity, ineffectiveness for the target phase examined, or inadequate target phase dissolution possibly related to pH buffering. Also, according to Wright et al. (2003), K2S2O8 is not specific to organic Se(-II) compounds, but also metal Se(-II), such as iron selenides. It is possible additional standards may be needed to accurately account for these species. The procedure employed by Wright et al. (2003) also unintentionally solubilized Se(0). Our results did not indicate any removal of this occurred during the extraction process. Combined with lack of comparability between ICP-AES SEP data and XANES whole soil LCFs along with lack of Se(-II) extractability, it is likely K2S2O8 was unsuccessful at extracting Se(-II) compounds.

For both Site A and one of the Site C soils, it appears the SEP failed to completely solubilize several components. Of the soils examined, phases identified in XANES were in most agreement between whole and sequentially extracted soils from Site B. XANES results also indicate oxidized Se was appropriately and effectively extracted in these soils. This suggests a possible concentration dependence in removal efficiencies. Further removal of Se associated with Fe-oxides following Step 5 also suggests this fraction is important and should be incorporated into SEPs. An overall lack of specificity of K2S2O8 for Se(-II) extraction was also indicated.

4.3. Implications for SEPs

Disparities between XANES results for whole and extracted soils suggest the occurrence of partial dissolution of solid surfaces. This indicates researchers should exert caution over SEP experimental conditions for Se or similar oxyanions. This may be particularly concerning for oxidized mobile and sorbed species of Se in seleniferous regions. XANES analyses for whole soils indicated Se was associated with additional forms of Fe-oxides not accounted for by the unmodified SEP. A lack of carbonate associated Se removal throughout the extraction scheme for many of the soils also suggested more recalcitrant phases were not accounted for. XANES analyses indicated organic Se species were also not adequately removed from the majority of soils. In combination, XANES-SEP results suggest incomplete mineral dissolution or lack of reagent specificity for target phases. Little could be said for water-soluble and PO4-extractable Se composition because of their low abundance. However, SEP results provide additional insight for these phases not indicated by XANES. Although SEPs are reasonable methods for trace element estimation, the Amacher (2010) procedure fails to reliably quantify several Se phases.

It is important that SEPs be tailored to soil geochemical composition and mineralogy in order to improve reliability. Extracting sorbed or co-precipitated Se in carbonate fractions appears to be problematic, especially for carbonate-rich soils. Modifications for soils with this particular feature should be made and can include extracting target elements under pH monitored conditions, increasing extraction liquid:solid ratio, extraction repetition, or increasing solute concentrations (Sulkowski and Hirner, 2006, Šurija and Branica, 1995, Tack and Verloo, 1997). Calcium and Mg concentrations can also be monitored during extraction to ensure adequate solubilization during carbonate phase extractions. Our results also indicated incomplete Se removal related to Fe-oxides for several soils examined. Previous works have noted the possibility of Fe-oxide-Se co-precipitates in phosphate mine soils that should be further evaluated.

5. Conclusion

This work examines Se distribution in phosphate mine soils using XANES and a SEP formulated for various Se phases. XANES spectra provided the most quantitative information for Se speciation in soils, indicating elemental and organic Se were the largest fractions present. In comparison, lower amounts of sorbed Se(IV) species were present. Higher relative abundances of Se related to Fe-oxides in whole soils suggested an additional Se phase related to crystalline Fe-oxides was present. With this information, a modified SEP was evaluated for its ability to target specific solid-phases for removal. XANES LCFs for extracted soils indicated only partial Se dissolution occurred in the carbonate, Fe-oxide, and organic fractions. XANES results indicate care should be taken when formulating SEPs to evaluate oxyanion contaminants in soils. Sequential extractions should be tailored to prospective soils to be analyzed, with both chemical and mineralogical composition known in order to ascertain the most suitable procedure.

Supplementary Material

Sup1

Acknowledgements

This work was sponsored by the U.S. EPA. Any opinions expressed in this paper are those of the author(s) and do not, necessarily, reflect the official positions and policies of the U.S. EPA. Any mention of products of trade name does not constitute recommendation for you by the U.S. EPA. This research used resources of the Advanced Photon Source, a U.S. Department of Energy (DOE) Office of Science User Facility operated for the DOE Office of Science by Argonne National Laboratory under Contract No. DE-AC02–06CH11357.

Abbreviations

LCF

Linear Least-Squares Combination Fitting

SEP

Sequential Extraction Procedure

XANES

X-ray Absorption Near Edge Structure

Footnotes

☆

This paper has been recommended for acceptance by Prof. W. Wen-Xiong.

References

  1. Amacher M. Sequential Extraction of Selenium, USDA-Forest Service Open File Report (2010) pp. 155–159 [Google Scholar]
  2. Aurelio G, Fernandez-Martinez A, Cuello GJ, Roman-Ross G, Alliot I, Charlet L. Structural study of selenium(IV) substitutions in calcite. Chem. Geol., 270 (2010), pp. 249–256 [Google Scholar]
  3. Booth DT. Emergence of bitterbrush seedlings on land disturbed by phosphate mining. J. Range Manag., 33 (1980), pp. 439–441 [Google Scholar]
  4. Cowan CE, Zachara JM, Resch CT. Solution ion effects on the surface exchange of selenite on calcite. Geochim. Cosmochim. Acta, 54 (1990), pp. 2223–2234 [Google Scholar]
  5. Davis TZ, Stegelmeier BL, Panter KE, Cook D, Gardner DR, Hall JO. Toxicokinetics and pathology of plant-associated acute selenium toxicosis in steers. J. Veterinary Diagnostic Investig., 24 (2012), pp. 319–327 [DOI] [PubMed] [Google Scholar]
  6. de Souza MP, Pilon-Smits EAH, Lytle CM, Hwang S, Tai J, Honma TSU, Yeh L, Terry N. Rate-limiting steps in selenium assimilation and volatilization by Indian mustard. Plant Physiol., 117 (1998), pp. 1487–1494 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Desborough G, DeWitt E, Jones J, Meier A, Meeker G. Preliminary mineralogical and chemical studies related to the potential mobility of selenium and associated elements in Phosphoria Formation strata, Southeastern Idaho. USGS Open File Rep. (1999), pp. 99–120 [Google Scholar]
  8. Duc M, Lefevre G, Fedoroff M. Sorption of selenite ions on hematite. J. Colloid Interface Sci., 289 (2006), pp. 556–563 [DOI] [PubMed] [Google Scholar]
  9. Dynes JJ, Huang PM. Influence of citrate on selenite sorption-desorption on short-range ordered aluminum hydroxides Huang PM(Ed.), Environmental Impact of Soil Component Interactions - Metals, Other Inorganics, and Microbial Activities, CRC/Lewis Publishers, Boca Raton, FL: (1995), pp. 47–61 [Google Scholar]
  10. Dynes JJ, Huang PM. Influence of organic acids on selenite sorption by poorly ordered aluminum hydroxides. Soil Sci. Soc. Am. J., 61 (1997), pp. 772–783 [Google Scholar]
  11. Ebels J, Sprik S, Pietschnig R, 2006. A Facile Lab Scale Synthesis of Red Selenium, 10th International Electronic Conference on Synthetic Organic Chemistry 10. [Google Scholar]
  12. Eiche E, Bardelli F, Nothstein AK, Charlet L, Gottlicher J, Steinberger R, Dhillon KS, Sadana US. Selenium distribution and speciation in plant parts of wheat (Triticum aestivum) and Indian mustard (Brassica juncea) from a seleniferous area of Punjab, India. Sci. Total Environ., 505 (2015), pp. 952–961 [DOI] [PubMed] [Google Scholar]
  13. El Mehdawi AF, Lindblom SD, Cappa JJ, Pilon-Smits EAH. Do selenium hyperaccumulators affect selenium speciation in neighboring plants and soil? An X-ray microprobe analysis. Int. J. Phytorem., 17 (2015), pp. 753–765 [DOI] [PubMed] [Google Scholar]
  14. Favorito JE, Eick MJ, Grossl PR, Davis TZ. Selenium geochemistry in calcareous phosphate mine soils and its relationship with plant bioavailability. Plant Soil (2017), pp. 1–15 [Google Scholar]
  15. Fessler AJ, Moller G, Talcott PA, Exon JH. Selenium toxicity in sheep grazing reclaimed phosphate mining sites. Vet. Hum. Toxicol., 45 (2003), pp. 294–298 [PubMed] [Google Scholar]
  16. Geelhoed JS, Hiemstra T, Van Riemsdijk WH. Competitive interaction between phosphate and citrate on goethite. Environ. Sci. Technol., 32 (1998), pp. 2119–2123 [Google Scholar]
  17. Geering HR, Cary EE, Jones LHP, Allaway WH. Solubility and redox criteria for the possible forms of selenium in soils. Soil Sci. Soc. Am. J., 32 (1968), pp. 35–40 [Google Scholar]
  18. Goldberg S, Glaubig RA. Anion sorption on a calcareous, montmorillonitic soil-selenium. Soil Sci. Soc. Am. J., 52 (1988), pp. 954–958 [Google Scholar]
  19. Grafe M, Eick MJ, Grossl PR. Adsorption of arsenate (V) and arsenite (III) on goethite in the presence and absence of dissolved organic carbon. Soil Sci. Soc. Am. J., 65 (2001), pp. 1680–1687 [Google Scholar]
  20. Grafe M, Eick MJ, Grossl PR, Saunders AM. Adsorption of arsenate and arsenite on ferrihydrite in the presence and absence of dissolved organic carbon. J. Env. Qual., 31 (2002), pp. 1115–1123 [DOI] [PubMed] [Google Scholar]
  21. Grauch RI, Desborough GA, Meeker GP, Foster AL, Tysdal RG, Herring JR, Lowers HA, Ball BA, Zielinski RA, Johnson EA. Chapter 8. Petrogenesis and Mineralogic Residence of Selected Elements in the Meade Peak Phosphatic Shale Member of the Permian Phosphoria Formation, Southeast Idaho. Elsevier, Amsterdam: (2004) [Google Scholar]
  22. Hayes KF, Roe AL, Brown GE Jr., Hodgson KO, Leckie JO, Parks GA. In situ X-ray absorption study of surface complexes: selenium oxyanions on α-FeOOH Sci. (New York, N.Y.), 238 (1987), pp. 783–786 [DOI] [PubMed] [Google Scholar]
  23. Heberling F, Vinograd VL, Polly R, Gale JD, Heck S, Rothe J, Bosbach D, Geckeis H, Winkler B. A thermodynamic adsorption/entrapment model for selenium(IV) coprecipitation with calcite. Geochim. Cosmochim. Acta, 134 (2014), pp. 16–38 [Google Scholar]
  24. Hossner LR. Dissolution for Total Elemental Analysis. Soil Science Society of America, Inc. and American Society of Agronomy, Inc, Madison, WI: (1996) [Google Scholar]
  25. Jouanneau JM, Latouche C, Pautrizel F. Analyse critique des extractions sequentielles a travers l’etude de quelques constituants des residus d’attaque critical analysis of sequential extractions through the study of several attack constituent residues. Environ. Technol. Lett., 4 (1983), pp. 509–514 [Google Scholar]
  26. Koningsberger DC, Prins R. X-ray Absorption: Principles, Applications, Techniques of EXAFS, SEXAFS and MES. John Wiley and Sons, New York: (1988) [Google Scholar]
  27. Kulp TR, Pratt LM. Speciation and weathering of selenium in Upper Cretaceous chalk and shale from South Dakota and Wyoming, USA. Geochim. Cosmochim. Acta, 68 (2004), pp. 3687–3701 [Google Scholar]
  28. Martens DA, Suarez DL. Selenium speciation of soil/sediment determined with sequential extractions and hydride generation atomic absorption spectrophotometry. Environ. Sci. Technol., 31 (1997), pp. 133–139 [Google Scholar]
  29. Martin JM, Nirel P, Thomas AJ. Sequential extraction techniques: promises and problems. Mar. Chem., 22 (1987), pp. 313–341 [Google Scholar]
  30. Martin-Ramos D. XPowder: Departamento de Mineralogia y Petrologia. Universidad de Granada (2004) [Google Scholar]
  31. Masscheleyn PH, Delaune RD, Patrick WH. Transformations of selenium as affected by sediment oxidation-reduction potential and pH. Environ. Sci. Technol., 24 (1990), pp. 91–96 [Google Scholar]
  32. McNeal JM, Balistrieri LS. Geochemistry and occurrence of selenium: an overview Jacobs LW (Ed.), Selenium in Agriculture and the Environment, Soil Science Society of America and American Society of Agronomy, Madison, WI: (1989) [Google Scholar]
  33. Merrill DT, Manzione MA, Peterson JJ, Parker DS, Chow W, Hobbs AO. Field evaluation of arsenic and selenium removal by iron coprecipitation. J. Water Pollut. Control Fed., 58 (1986) [Google Scholar]
  34. Ng BH, Anderson JW. Synthesis of selenocysteine by cysteine synthase from selenium accumulator and non-accumulator plants. Phytochemistry, 17 (1978), pp. 2069–2074 [Google Scholar]
  35. Oram LL, Strawn DG, Marcus MA, Fakra SC, Möller G. Macro- and microscale investigation of selenium speciation in Blackfoot River, Idaho sediments. Environ. Sci. Technol., 42 (2008), pp. 6830–6836 [DOI] [PubMed] [Google Scholar]
  36. Perkins RB, Foster AL. Chapter 10. Mineral affinities and distribution of selenium and other trace elements in black shale and phosphorite of the Phosphoria formation Hein JR (Ed.), Handbook of Exploration and Environmental Geochemistry, Elsevier, Amsterdam: (2004), p. 251 [Google Scholar]
  37. Pickering IJ, Brown GE, Tokunaga TK. Quantitative speciation of selenium in soils using X-ray absorption spectroscopy. Environ. Sci. Technol., 29 (1995), pp. 2456–2459 [DOI] [PubMed] [Google Scholar]
  38. Pickering IJ, Ponomarenko O, George GN, La Porte PF, Strait K, Gailer J, Leslie EM, Spallholz J. Synchrotron studies of selenium interactions with arsenic. Banuelos GS, Lin Z, Yin X (Eds.), The 3rd International Conference on Selenium in the Environment and Human Health, CRC Press, Hefei, China (2013) [Google Scholar]
  39. Pilon M, Owen JD, Garifullina GF, Kurihara T, Mihara H, Esaki N, Pilon-Smits EAH. Enhanced selenium tolerance and accumulation in transgenic arabidopsis expressing a mouse selenocysteine lyase. Plant Physiol., 131 (2003), pp. 1250–1257 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Qiang T, Xiao-Quan S, Jin Q, Zhe-Ming N. Trace metal redistribution during extraction of model soils by acetic acid/sodium acetate. Anal. Chem., 66 (1994), pp. 3562–3568 [Google Scholar]
  41. Qin HB, Zhu JM, Takahashi Y, Zheng LR. Selenium Speciation in Soils from Selenosis Area: Comparison between a Sequential Extraction Procedure and XAFS. Taylor & Francis Group, London: (2014) [Google Scholar]
  42. Rajan SSS. Adsorption of selenite, phosphate and sulphate on hydrous alumina. J. Soil Sci., 30 (1979), pp. 709–718 [Google Scholar]
  43. Ravel B, Newville M. ATHENA, ARTEMIS, HEPHAESTUS: data analysis for X-ray absorption spectroscopy using IFEFFIT. J. Synchrotron Radiat., 12 (2005), pp. 537–541 [DOI] [PubMed] [Google Scholar]
  44. Ryser AL, Strawn DG, Marcus MA, Fakra S, Johnson-Maynard JL, Möller G. Microscopically focused synchrotron X-ray investigation of selenium speciation in soils developing on reclaimed mine lands. Environ. Sci. Technol., 40 (2006), pp. 462–467 [DOI] [PubMed] [Google Scholar]
  45. Ryser AL, Strawn DG, Marcus MA, Johnson-Maynard JL, Gunter ME, Möller G. Micro-spectroscopic investigation of selenium-bearing minerals from the Western US phosphate resource area. Geochem. Trans., 6 (2005), pp. 1–11 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Scheinost AC, Kretzschmar R, Pfister S, Roberts DR. Combining selective sequential extractions, X-ray absorption spectroscopy, and principal component analysis for quantitative zinc speciation in soil. Environ. Sci. Technol., 36 (2002), pp. 5021–5028 [DOI] [PubMed] [Google Scholar]
  47. Schultz LG. Quantitative Interpretation of Mineralogical Composition from X-ray and Chemical Data for the Pierre Shale, Analytical Methods in Geochemical Investigations of the Pierre Shale (1964) US Geological Survey Professional Paper 391-C, Washington [Google Scholar]
  48. Schwertmann U, Cornell RM. Iron Oxides in the Laboratory: Preparation and Characterization. (second ed.), Wiley-VHC, Weinheim: (2000) [Google Scholar]
  49. Shacklette HT, Boerngen JG. Element Concentrations in Soils and Other Surficial Materials of the Conterminous United States. U.S. Geological Survey Professional Paper 1270 United States Government Print Office, Washington: (1984) [Google Scholar]
  50. Shan X, Chen B. Evaluation of sequential extraction for speciation of trace metals in model soil containing natural minerals and humic acid. Anal. Chem., 65 (1993), pp. 802–807 [Google Scholar]
  51. Shrift A. Aspects of selenium metabolism in higher plants. Annu. Rev. Plant Physiol., 20 (1969), pp. 475–494 [Google Scholar]
  52. Stillings L, Amacher M. Selenium Attenuation in a Wetland Formed from Mine Drainage in the Phosphoria Formation, Southeast Idaho. Life Cycle of the Phosphoria Formation: from Deposition to the Post-Mining Environment. Elsevier B.V., Amsterdam: (2004) [Google Scholar]
  53. Stillings LL, Amacher MC. Kinetics of selenium release in mine waste from the meade peak phosphatic shale, Phosphoria Formation, Wooley Valley, Idaho, USA. Chem. Geol., 269 (2010), pp. 113–123 [Google Scholar]
  54. Strawn DG, Doner H, Mavrik Z, McHugo S. Microscale investigation into the geochemistry of arsenic, selenium, and iron in soil developed in pyritic shale materials. Geoderma, 108 (2002), pp. 237–257 [Google Scholar]
  55. Su C, Suarez DL. Selenate and selenite sorption on iron oxides: an infrared and electrophoretic study. Soil Sci. Soc. Am. J., 64 (2000), pp. 101–111 [Google Scholar]
  56. Sulkowski M, Hirner AV. Element fractionation by sequential extraction in a soil with high carbonate content. Appl. Geochem, 21 (2006), pp. 16–28 [Google Scholar]
  57. Šurija B, Branica M. Distribution of Cd, Pb, Cu and Zn in carbonate sediments from the Krka river estuary obtained by sequential extraction. Sci. Total Environ., 170 (1995), pp. 101–118 [Google Scholar]
  58. Tack FM, Verloo MG. Determination of the acid extractable metal fraction in a sediment with a high carbonate content: a critical evaluation. Prost R (Ed.), Contaminated Soils: Third International Conference on the Biogeochemistry of Trace Elements, INRA Editions, Paris, Paris (1997), pp. 167–173 [Google Scholar]
  59. Tessier A, Campbell PGC, Bisson M. Sequential extraction procedure for the speciation of particulate trace metals. Anal. Chem., 51 (1979), pp. 844–851 [Google Scholar]
  60. USEPA. EPA Environmental Assessment Sourcebook. Ann Arbor Press, Inc, Chelsea, Michigan: (1996) [Google Scholar]
  61. Weres O, Jaouni A, Tsao L. The distribution, speciation and geochemical cycling of selenium in a sedimentary environment, Kesterson Reservoir, California, U.S.A. Appl. Geochem, 4 (1989), pp. 543–563 [Google Scholar]
  62. Wright MT, Parker DR, Amrhein C. Critical evaluation of the ability of sequential extraction procedures to quantify discrete forms of selenium in sediments and soils. Environ. Sci. Technol., 37 (2003), pp. 4709–4716 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Sup1
NIHMS1522615-supplement-Sup1.xlsx (50.8KB, xlsx)

RESOURCES