Allophane Estimation in the Clay Fraction of Volcanic Soils By Rietveld Refinement.

Volcanic ash-derived soils have characteristic properties that are rarely found in soils derived from other parent materials. Their clay fractions are dominated by non-crystalline components, such as allophane, which confer greater reactivity and surface area. Allophane is a secondary mineral formed from weathering of volcanic materials (ash, pumice, and scoria) forming hollow nano-scale spherules composed by a gibbsite-like sheet in the outer surface and a silica tetrahedral sheet in the inner surface. Because clay fractions of volcanic soils are dominated by this type of X-ray amorphous materials, the classic approach to characterizing their mineralogy has been through selective chemical dissolutions using acid oxalate and sodium pyrophosphate. However, some authors have suggested that these methods underestimate the amorphous minerals in these highly reactive soils. The Rietveld refinement method, originally developed to refine the unit cell and atomic coordinates of a single crystal, has become more popular as a minimizing technique for full powder X-ray diffraction pattern quantification. By spiking samples with an internal standard, this technique has been also used to quantify by difference the amount of amorphous phases in a clay sample. A set of soils from the Pacific Coastal Plain of Guatemala were selected to compare the content of extractable allophane as calculated from acid oxalate and sodium pyrophosphate standard dissolutions with the Rietveld refinement method using TOPAS software (from Bruker-AXS) for spiked samples with Zincite as internal standard. Riedveld refinement estimations were greater than the dissolution estimations, and this technique is sensitive to differences in extractable allophane among samples. A good linear correlation (R² = 0.87) was obtained for the two quantification methods, and a calibration with synthetic mixtures of clays with known composition is proposed to better understand whether standard dissolutions underestimate allophane contents, or Rietveld refinement overestimate amorphous content.