Extraction and speciation analysis of roxarsone and its metabolites from soils with different physicochemical properties

Purpose: The application of roxarsone (ROX), an arsenic-containing compound, as a feed additive in the animal production industry results in elevated soil levels of ROX and its metabolites, namely, monomethylarsonic acid (MMA), dimethylarsinic acid (DMA), arsenate (As(V)), and arsenite (As(III)). This study was conducted to study the extraction and speciation analysis of ROX-related arsenicals in soils with different physicochemical properties and the possible effects of soil properties on the extraction of ROX and its metabolites. Materials and methods: Analytical method based on high-performance liquid chromatography (HPLC)-inductively coupled plasma–mass spectrometry (ICP-MS) was employed to determine the concentrations of As(III), DMA, MMA, As(V), and ROX extracted by different extraction solvents from different soils spiked by arsenicals. Validity of the developed method was assessed by the recovery efficiencies of arsenic species in soil-dissolved matter solutions containing 20 μg As · L⁻¹ of each arsenic species. Effects of soil properties on the extraction of ROX and its metabolites were analyzed by Pearson’s correlation. Results and discussion: Arsenic species were separated using gradient elution of water and 20 mmol · L⁻¹ (NH₄)₂HPO₄ + 20 mmol · L⁻¹ NH₄NO₃ + 5 % methanol (v/v) within 27 min. The linear ranges of all arsenicals were 0–200 μg As · L⁻¹ with R² > 0.9996. The developed method provided lower limits of detection for As(III), DMA, MMA, As(V), and ROX (0.80, 0.58, 0.35, 0.24, and 1.52 μg As · L⁻¹, respectively) and excellent recoveries (92.52–102.2 %) for all five species. Arsenic speciation was not altered by 0.1 mol · L⁻¹ NaH₂PO₄ + 0.1 mol · L⁻¹ H₃PO₄ (9:1, v/v), which offered better average extraction efficiencies for As(III), As(V), DMA, MMA, and ROX (32.49, 92.50, 78.24, 77.64, and 84.54 %, respectively). Extraction performance of arsenicals was influenced by soil properties, including pH, cation exchange capacity (CEC), total Fe, and amorphous Fe. Conclusions: ROX and its metabolites from soils could be satisfactorily separated by the developed method for the studied arsenicals. To extract arsenic species from soils, 0.1 mol · L⁻¹ NaH₂PO₄ + 0.1 mol · L⁻¹ H₃PO₄ (9:1, v/v) was recommended. Extraction efficiencies of arsenicals were influenced more by solvent composition than soil physicochemical properties. The present study provides a valuable tool and useful information for determining the concentrations of ROX and its metabolites in contaminated soils.