Antimony mobility in sulfidic systems: Coupling with sulfide-induced iron oxide transformations

Iron (Fe) oxides are important host phases for antimony (Sb), a toxic metalloid of environmental concern. In wetland soils and sediments, poorly ordered Fe oxides such as ferrihydrite may undergo reductive dissolution and mineralogical transformation upon reaction with dissolved sulfide (S(-II)). The consequences of these processes for the mobility of associated Sb have not been investigated to date. Here, we allowed Sb(V)-bearing ferrihydrite (molar ratio of Fe:Sb = 400) to react with varying S(-II) concentrations (Fe(III):S(-II) = 0.2, 0.5, and 1) at pH 6 and 8 over 32 days. Changes in speciation and concentration of Fe, S and Sb in the aqueous, colloidal and solid phase were examined through a combination of aqueous-phase analyses, X-ray diffraction and synchrotron X-ray absorption spectroscopy. Addition of S(-II) caused rapid reduction of Fe(III), thereby producing elemental S and Fe(II). X-ray diffractometry and Fe K-edge extended X-ray absorption fine structure (EXAFS) spectroscopy revealed that S(-II) addition resulted in the precipitation of Fe(II) sulfides (mackinawite (FeS) and pyrite (FeS2)) and formation of secondary Fe(III) oxides (goethite (FeOOH) and hematite (Fe2O3)). The formation of mackinawite and pyrite was further confirmed by S K-edge X-ray absorption near-edge structure (XANES) spectroscopy, and was found to occur to the greatest extent in the high-sulfide treatments. The initial reductive dissolution of ferrihydrite was paralleled by a fast increase in dissolved Sb concentrations, with ∼25% of total Sb being released in the high-sulfide treatments. The Sb release was followed by Sb immobilization within ∼1–7 days. Since ion-chromatography ICP-MS revealed antimonate (Sb(OH)6-) as the primary Sb aqueous phase species throughout the experiment, with only negligible concentrations of antimonite (Sb(OH)3) and only very minor amounts (<4% of total Sb) of tri- and tetrathioantimonate (HxSbS3Ox−3/HxSbS4x−3), the decrease in Sb concentrations was attributed to surface-based sorption and structural incorporation of primarily Sb(V) by the secondary Fe oxides. In accordance with the dominance of aqueous Sb(V), Sb K-edge XANES spectroscopy showed that Sb(V) was also the dominant Sb species in the solid phase, comprising up to 90% of solid-phase Sb in the low-sulfide treatments. However, higher S(-II) addition and lower pH favored production of Sb(III) and resulted in up to 40% and 20% of solid-phase Sb(V) being reduced to Sb(III) at pH 6 and 8, respectively, with this Sb(III) comprising a mixture of O- and S-coordinated species. Around 15% of < 0.45-µm Sb occurred in the colloidal (>3 kDa) size fraction at pH 8 under medium and high S(-II) conditions, while no colloidal Sb was found in other treatments. Together, these results show that Fe oxide sulfidization can have opposing effects on Sb mobility. On the one hand, the initial sulfide-promoted Fe oxide dissolution triggers Sb release into the aqueous phase. On the other hand, Sb can subsequently be immobilized via sorption to secondary Fe oxides and newly-formed Fe sulfides during the later stages of sulfidization. Sulfidization reactions, and the complex opposing impacts on Sb mobility, should therefore be considered for the risk assessment and derivation of adequate management strategies at Sb-impacted sites which experience sulfidic conditions.