Chapter 2 Anion Sorption Topology on Hematite: Comparison of Arsenate and Silicate

Abstract

Arsenate and silicate are tetrahedral anions that strongly sorb to positive Fe oxide surfaces over the pH range 2–7. Both are important agents for modification of Fe oxide surface reactivity, and notably passivate against other sorption reactions. Arsenate is a significant health hazard as a sorbed pollutant associated with acid mine drainage, while silicate is a common anion in natural solutions. Our aim is to understand the types of sorption complexes that form with these anions on different crystal faces, and whether polymerization occurs with the silicate units. Silicate polymerization could dramatically alter Fe oxide surface reactivity. The structural characterization is conducted using both grazing incidence extended X-ray absorption fine structure (GIEXAFS) at Stanford Synchrotron Radiation Laboratory (SSRL) and the National Synchrotron Light Source (NSLS), and surface diffraction (using crystal truncation rod (CTR) analysis) at the Advanced Photon Source (APS). GIEXAFS yields interatomic distances from arsenic and silicon to their oxygen first neighbor shell and second Fe or other next neighbor shell, and thus allows identification of the local geometry of sorption. Polarized X-ray fine structure spectroscopy further allows determination of the orientation and density of the complexes on the various Fe surface planes. However, this information is incomplete as any response of the surface to sorption is not revealed, and hydrogen bonding and water molecule arrangement at the surface can be changed due to the sorption process. To access these we use CTR experiments and compare the results with samples without sorbed anions. GIEXAFS results for both hematite (0 0 0 1) and ( 1 1 ¯ 02 ) planes show arsenate sorbed in two ways: bidentate binuclear and bidentate mononuclear. Most of the latter type of sorption geometry appears to be present on surface step edges on the ( 1 1 ¯ 02 ) surface, while there is little or no such attachment to the ( 1 1 ¯ 02 ) surface terraces. These results appear consistent with preliminary structural models obtained by CTR work for the clean and sorbed wet hematite surfaces, and further analysis is in progress. In the case of silicate, strong changes in the CTR are observed indicating considerable surface interactions. CTR measurements of the silicate on the ( 1 1 ¯ 02 ) surface of hematite suggest that a monodentate sorption geometry is more dominant than other possible sorption geometries. However GIEXAFS analysis done on analogous samples at the Si K-edge shows second shell contributions that are similar to amorphous silica, and are largely independent of sorption density. These results suggest that there are at least two kinds of silicate on the hematite surface: a chemically bound fraction present as a silicate–Fe 3+ octahedral complex, and an incoherent portion which is not detectable by the CTR measurements but appears to dominate the GIEXAFS results. The complexed silicate on the hematite ( 1 1 ¯ 02 ) surface is linked by a single oxygen to surface Fe, i.e., a monodentate connection, with an interatomic Si-Fe distance close to those observed in the nontronite and acmite structures. This is the first evidence to our knowledge that identifies silicate as a well-defined sorption complex rather than only as an amorphous surface precipitate. More significantly, the monodentate complex is apparently favored over a more strongly bound bidentate complexation geometry, suggesting that polymerization at the interface may originate synchronously with sorption and be mitigated by the position of surface sites consistent with a polymerized network. These findings have important significance for the nature of passivation processes on Fe oxides and other important reactive natural surfaces.