Changes in the Distribution of Selenium Oxidation States with Sample Storage

Significant changes in the distribution of selenium (Se) oxidation states of fresh soil were noted with oven-drying (90°C) or with outdoor storage (>5 yr) in compromised plastic containers compared with small changes in Se oxidation state distribution in air-dried samples. Compromised outdoor storage resulted in a 25% increase in the selenate (Se6+) concentrations with a concomitant 18.4 and 5.7% decrease in elemental Se and selenite (Se4+) concentrations, respectively, on a total Se basis. Substantial increases in phosphate-soluble selenide from 15.7% of total Se for fresh seleniferous alfalfa (Medicago sativa L.) to 32.6% phosphate-soluble Se were noted with air-dried alfalfa. The results indicate that air-drying will not substantially alter the distribution of soil Se oxidation states, but will result in a significant increase in the concentration of phosphate-soluble Se in seleniferous alfalfa residue. S has become one of the major elemental contaminant problems in irrigation drainage waters in the western portion of the USA due to the toxicological effects of Se on fish and wildlife. An overview of the Se problem in California evaporation ponds including the Kesterson evaporation ponds is presented by Ohlendorf and Santolo (1994). In environmental and biological samples, Se can exist in inorganic forms as elemental Se (Se°), metal selenides (Se2-), and selenite (Se4+) and selenate (Se6÷), and as organic species with direct Se-C bonds such as methylated compounds, seleno-amino acids, and selenoproteins. Research has identified that the toxic dose of Se is very much dependent on its chemical form, with different toxicity for organic and inorganic Se compounds (Heinz et al., 1987, 1988, 1989). Thus, identification of each Se oxidation state in soil, plant, and water samples is more important than the determination of a total Se concentration (Martens and Suarez, 1997). Accurate determination of the distribution of Se oxidation states, especially in water samples, has been shown to be dependent on sample preparation and storage (Wang, 1994; May and Kane, 1984; Robberecht and USDA-ARS, U.S. Salinity Lab., 450 W. Big Springs Rd., Riverside CA 92507. Trade names and company names are included for the benefit of the reader and do not imply any endorsement or preferential treatment of the product listed by the USDA. Received 13 Feb. 1997. *Corresponding author ([email protected]). Published in J. Environ. Qual. 26:1711-1714 (1997). Van Greiken, 1982; Cheam and Agemian, 1980). Possible loss of Se from fresh plant tissue with high concentrations of Se has been noted upon drying for sample preparation (Grant, 1981), but currently little information available to document the changes in the distribution of Se oxidation states in organic material with different sample preparation or storage methods. Soil Sampling and Storage The first sampling of soil from the research plots of W.T. Frankenberger in Kesterson evaporation pond 4 was conducted in 1987. At the time of sampling, a subsample was immediately air-dried and stored in the laboratory in a closed wide-mouth polyethylene bottle. The remaining soil was stored outside between two glasshouse enclosures in plastic trash containers. This soil was soon exposed to the elements due to the loss of the original lids and the development of cracks in the sides of the trash containers. A second sampling of the Kesterson evaporation pond 4 soil near the site of the initial sample collection was collected May 1996, 2 d after a 3-cm rain event. The field-moist soil was transported via an ice chest back to the laboratory and immediately sieved to pass a 2-mm opening in the field-moist state to remove large pieces of organic material. The fresh soil sample was immediately analyzed for Se oxidation state distribution, then stored in a field-moist state at 4°C. A subsample was then air-dried (ambient temperatures) for 48 h in the laboratory and stored in a widemouth polyethylene plastic bottle. The Kesterson soil (Turlock fine loamy, mixed, thermic Albic Natraqualf) had the following chemical and physical properties: pH, 7.63; organic C, 26.7 g C kg-1 soil; inorganic C, 3.3 g C kg-1 soil; total N, 4.96 g N kg-1 soil; sand content, 580 g kg-1 soil; clay content, 150 g kg-1 soil; and total Se, 47.2 mg Se kg-~ soil (1987 sampling; Frankenberger and Karlson, 1994) and 31.4 mg Se kg-~ soil (1996 sampling). The soil chemical and physical parameters were determined as described by Martens and Suarez (1997). Seleniferous alfalfa was produced in sand culture tanks irrigated every 5 h with a 0.5 M Hoagland’s soluAbbreviations: HGAAS, hydride generation atomic absorption spectrophotometry; MANOVA, one-way multivariate analysis of variance; LSD, least significant difference. 1712 J. ENVIRON. QUAL., VOL. 26, NOVEMBER-DECEMBER 1997 tion (Hoagland and Arnon, 1950) or with a 0.5 M Hoagland’s solution containing 2 mg Se L-1 as sodium selenate. The alfalfa was harvested at the 0.1 bloom stage for Se analyses. Determination of Selenium Oxidation States by Sequential Extraction Hydride generation atomic absorption measurements (HGAAS) were made with a Perkin Elmer 3030B spectrophotometer (Perkin-Elmer Corp., Norwalk, CT) equipped with a Varian Model VGA-76 (Mulgrave, Victoria, Australia) vapor generation apparatus for Se analysis. Operating conditions and quality assurance program for HGAAS were followed as detailed by Martens and Suarez (1997). Alfalfa was digested for total content by the HNO3 digestion method described by Martens and Suarez (1997). To determine the distribution of Se oxidation states present in the seleniferous alfalfa and soils before and after drying, 5 g soil (ovendry basis) or 200 mg alfalfa was placed into a 40-mL PTFE (Teflon) centrifuge tube and exposed to a duplicated sequential extraction scheme composed of a water extraction, a buffered phosphate (P-buffer) extraction, a sodium hydroxide (NaOH) or sodium persulfate extraction, and a nitric acid extraction (HNO3) (Martens and Suarez, 1997). Distribution of Se oxidation states Se4+, Se6+, and Se2solubilized by the water and Pbuffer extractions was determined by selective HGAAS determination of the Se4+ oxidation state (Martens and Suarez, 1997). The Se oxidation states Sea+ and Se 2were determined in the NaOH extraction and the elemental Se (SeO) concentration was measured as 6+ in the HNO3 oxidation by HGAAS procedures. All allotropes of SeO, insoluble S-Se associations, and possible S-amino acid-Se complexes are grouped together in the SeO fraction (Weres et al., 1989). Storage conditions evaluated for the first set of collected Kesterson 4 soil were (i) air-dried for 9 yr and (ii) stored outside for yr. Storage conditions evaluated for the second set of Kesterson 4 soil were (i) field-fresh soil, (ii) air-dried soil (1 d), (iii) air-dried soil (4 mo), and (iv) oven-dried (18 h; 90°C). A second plant extraction methodology, a methanol/ chloroform/water (M/Ch/W) extraction was performed to fractionate the soluble, polar Se compounds present in the alfalfa tissue from the nonpolar, insoluble components. Before and after air-drying, alfalfa tissue (200 mg) was added to 40 mL centrifuge tubes and extracted with 15 mL M/Ch/W (15:5:3, v/v/v) overnight with gentle shaking (60 oscillations rain-l). Alfalfa remaining was separated by centrifugation, dried, and digested for total Se. To isolate the polar compounds (M/W phase), from the pigments and lipids (Ch plase), 6-mL DI water was added and the isolated lipid phase digested for total Se. Distribution of Se oxidation states in M/W phase was as described for the water and P-buffer extractions. The phase separation and Se analysis of the aqueous, Ch and insoluble phases are described by Martens and Suarez (1997, unpublished data). Treatment differences were evaluated by a one-way Table 1. Distribution of Se oxidation states in Kesterson pond 4 (1987 sampling) upon storage in an air-dry state or with compromised outside storage for 9 yr as determined by sequential extractions. Selenium oxidation state Treatment Se ~+ Se 4+ Se ° Se 2Total