Determination and Chemical Speciation of Selenium in Farmlands from Southeastern Spain: Relation to Levels Found in Sugar Cane Juana P. Diaz-Alarco ´n, † Miguel Navarro-Alarco ´n,* ,† Herminia Lo ´pez-Garcı ´a de la Serrana, † Carlos Asensio-Drima, ‡ and Maria C. Lo ´pez-Martı ´nez † Department of Nutrition and Bromatology, Faculty of Pharmacy, University of Granada, E-18071 Granada, Spain, and Department of Edaphology, Faculty of Pharmacy, University of Granada, E-18071 Granada, Spain The concentration and the oxidation state of selenium in farmland soils, sewage sludges, and sands was determined by using hydride generation atomic absorption spectrometry. Total Se concentra- tions ranged from 0.071 to 0.390 μg/g, whereas Se IV concentrations varied from not detectable to 0.050 μg/g. Selenium is predominantly present as Se VI ; a minor fraction was found as Se IV . We also studied the influence of pH, its values varied between 7.03 and 8.33, on the oxidation state and the content in Se and its possible correlation with the selenium uptake and accumulation in the sugar cane plants. Significant differences were observed between the levels of total Se (p < 0.01) and Se IV (p < 0.05) in farmland soils, whether they are located in the industrialized or nonindustrialized areas. Total Se is found in higher concentrations and mainly as Se IV (selenite) in industrialized area. The content of total Se in the samples of sugar cane taken from the zone is independant of the concentration of total Se and Se IV in farmland soils of the zone (p < 0.05). Keywords: Se speciation; pH; farmlands and sewage sludges; uptake by sugar cane; influence of human and industrial activities INTRODUCTION Selenium is found in the earth’s crust in concentra- tions that vary from 0.5 to 0.09 μg/g; it is frequently related with minerals that contain sulfur (National Research Council, 1983). The existence of Se in soil is related to several factors, such as geographic location, kind of rock, oxidation-reduction potential, pH, nature of drainage waters, and type of plant grown (Kabata- Pendias and Pendias, 1984; Elrashidi et al., 1987; Alexander et al., 1988; Banuelos and Meek, 1990; Simonoff and Simonoff, 1991; Paya-Pe ´rez et al., 1993; Zhao et al., 1993; Wahl et al.,1994). Normally, there is a direct correlation between soil Se content and the plants grown on such soil (Varo et al., 1988; Golubkina et al., 1992; Levander and Burk, 1994) so that high levels of Se may concentrate in some species of accumulator plants. However, the fact of finding high levels of Se in soils does not always imply that Se can be found in plants producing toxic effects as the Se uptake depends on different factors (Burau et al., 1988) such as the species of plant, the way in which Se appears in soils, the presence of other ions in the solution of the soil, SO 4 2- (Bisbjerg and Gissel- Nielsen, 1969; Mikkelsen et al., 1988), the management of the crop, pruning, and harvest time. The soil pH also plays an important role as, in those soils with an acid character, Se is found mainly in the form of selenite, not very soluble and assimilable, while in alkaline soils, Se (SeO 3 2- ) becomes oxidated and then selenate is produced; selenate is much more soluble and easily assimilable. Moreover, it can originate several selenium-containing organic compounds (Hartfiel and Bahners, 1988; Banuelos and Meek, 1990; Gondi et al., 1992; Zhao et al., 1993). This study describes techniques to widen Se specia- tion methods to assess total Se and Se IV levels in farmland soils, sewage sludges, and sands and its later direct determination by hydride generation atomic absorption spectrometry. Furthermore, pH values of the samples have been determined for their correlation with the oxidation state prevailing in Se in these samples. This way, we could have better knowledge of the factors that influence both bioavailability and, as a consequence, Se levels in the plants grown in this zone, which values were formerly determined (Diaz-Alarco ´n et al., 1994a). Apart from this, we have also found it interesting to study the influence of human and industrial activities on Se levels in farmland soils of the zone and the influence on concentrations of this element in sugar cane plants (the most important crop of the zone) grown in these soils (Diaz-Alarco ´n et al., 1994a). MATERIALS AND METHODS Apparatus. The samples were mineralized in an Invester sand bath. All atomic measurements were made with a Perkin-Elmer Model 1100B atomic absorption spectrometer equipped with a Perkin-Elmer MHS-10 hydride generator. A selenium hollow cathode lamp (Perkin-Elmer Corp.) was operated under the conditions recommended by the manufac- turer. A spectral slit width of 2.0 nm was selected to isolate the 196.0-nm line. All analyses were performed in peak height mode to calculate absorbance values. The measurements of pH were carried out with a Radiometer Model 26 pHmeter. Reagents. All solutions were prepared with ultrapure water with a specific resistivity of 18 MΩ cm obtained by filtering double-distilled water through a Milli-Q purifier (Millipore) immediately before use. A commercially available 1000 μg/mL selenium standard solution (prepared from SeO2 in 0.5 mmol/mL HNO3) was used (Tritisol, Merck). The alternate standards for calibration (10 mg/L and 100 μg/L) were prepared by serial dilutions with 1.9% HCl solution. HNO3 (65%), HCl (37%), HClO4 (65%) (Carlo Elba), NaBH4, and NaOH (Merck analytical grade) were also used. Samples. Farmland soil samples were obtained from the fields of the study zone, according to the random model † Department of Nutrition and Bromatology. ‡ Department of Edaphology. 2423 J. Agric. Food Chem. 1996, 44, 2423-2427 S0021-8561(95)00433-X CCC: $12.00 © 1996 American Chemical Society