Short-term bonding of freshly added cadmium in Brazilian Oxisols

Full text: Cadmium is one of the most toxic elements for all animals and plants. The chemical speciation of soil Cd should dictate its bioavailability and potential toxicity to plants and animals, and its transfer pathways through the food web to humans. Our research evaluated short-term conversion of freshly added cadmium in Brazilian Oxisols with high Cd sorption capacities. Oxisols found in humid tropical regions, including in extended areas under agriculture in developing countries and Brazil, have pH-dependent charge due to abundant organic matter, oxide minerals, and kaolinite. The objective of this research was to utilize Cd LIII-edge XANES analysis to determine short-term changes in Cd speciation in Oxisols reacted with soluble Cd(II). Six soil samples, comprising three Oxisols (Anionic Acrudox, Rhodic Hapludox and Typic Hapludox), one Mollisol (Mollic Epiaquent), and two Entisols (Typic Argiudolls and Typic Quartzipsamments), were incubated for 0.5 and 6 hours with cadmium chloride solution to obtain a final concentration of 4.45 mmol Cd kg^-1. Complementary incubation for 4 months were submitted to chemical fractionation. Adsorbed Cd standards for XANES analyses were prepared for pure materials of leonardite, kaolinite, montmorillonite, goethite, gibbsite and hematite. Standards of CdS, CdCl₂ and Cd₃(PO₄)₂ were diluted with saccharose to 4.45 mmol Cd kg-1. The samples were dried, homogenized and placed as thin film onto a double-sided carbon tape on a sample holder. Cadmium LIII-edge (3538 eV) XANES data were collected under vacuum at the Soft X-Ray Spectroscopy (D04ASXS) beamline of the Brazilian Synchrotron Light Laboratory – LNLS. For each sample, three to ten spectra were collected in fluorescence mode within energy range between 3500 and 3620 eV. Data were baseline corrected, normalized, and fit using the Athena data analysis software. Scans were merged to improve the signal-to-noise ratio and E0 was set to the maximum in the first derivative spectrum. A unique normalization was needed for samples and some standards due to interference by the K absorption edge of naturally occurring potassium at 3608 eV. Standards of pure (K-free) salts and saccharose normalized using a linear fit to the post edge region served as a guide for normalizing other samples at the point of maximum absorption. Linear Combination Fitting (LCF) was used to fit samples, with CdCl₂ selected as an obligatory standard because it was the source of Cd added to the samples. Cadmiumleonardite, a surrogate for Cd on organic matter, was represented in fits for all soil samples; ranging from 18-19% in the Typic Quartzipsamments, 36% in Typic Argiudolls, and up to 77% in Rhodic Hapludox samples. The Typic Quartzipsamments data could be fit with up to 31% of the CdCl₂ standard, indicating a lack of adsorption and ultimately precipitation of this salt upon drying. The proportions of the standards giving best fits for each soil sample had only minor vartiatins between 0.5 and 6 hours of reaction time, and showed no definite trend across soil classes. In conjunction with longer-term incubation studies involving chemical fractionation, our results suggest that Cd(II) discharged into Oxisols would bind rapidly with soil organic matter or remain in soluble forms with little change in speciation and bioavailability following up to 4 months of residence time in the soils. (author)