Structural water in ferrihydrite and constraints this provides on possible structure models

• Published in: The American mineralogist Vol. 96; no. 4; pp. 513 - 520
• Main Authors: Xu, Wenqian, Hausner, Douglas B, Harrington, Richard, Lee, Peter L, Strongin, Daniel R, Parise, John B
• Format: Journal Article
• Language: English
• Published: Washington Mineralogical Society of America 01.04.2011; De Gruyter; Walter de Gruyter GmbH
• Subjects: calorimetry; coordination; crystal structure; Crystals; Differential scanning calorimetry; Ferrihydrite; goethite; hematite; hydrous ferric oxide; hydroxyl; infrared spectra; iron; metals; Mineralogy; nonsilicates; oxides; pair distribution function; PDF analysis; polyhedra; spectra; Surface water; Temperature; Temperature control; tetrahedra; tetrahedral iron in ferrihydrite; TGA data; Thermogravimetric analysis; Water content; water of crystallization; X-ray spectra
• ISSN: 0003-004X; 1945-3027
• DOI: 10.2138/am.2011.3460

The dry thermal transformation of 2-line ferrihydrite to hematite was investigated using combinations of thermogravimetric (TG) and DSC and pair distribution function (PDF) analysis of X-ray total scattering data and in situ temperature controlled infrared (IR) spectroscopy. TG data show a 25.6±0.1% weight loss below 300°C, ascribed to the removal of surface water since PDF analysis shows no change in the structure of ferrihydrite up to this temperature. The transformation to hematite occurs at around 415±1°C (peak temperature) at a heating rate of 10°C/min, with no obvious weight change during or after the transformation. In situ PDF analysis indicates that the ferrihydrite bulk structure remained intact up to the direct transition to crystalline hematite, with no intermediate phases, crystalline or amorphous, formed. In situ IR data shows the extent of absorption attributable to OH stretching in ferrihydrite at 215°C dropped to 10% of its room-temperature value. These results suggest ferrihydrite contains very little structural OH: the molar ratio of OH/Fe is 0.18±0.01. A recently proposed akdalaite-like ferrihydrite model has an OH/Fe equal to 0.2, consistent with this result. The 3-phase model proposed by Drits et al. (1993) has an average formula close to FeOOH, with an OH/Fe equal to 1.0, far more than suggested by our experiments. Based on the constraints set by the estimated water content and the PDF signatures, we examined possible anion packing types and local structural motifs in ferrihydrite, and demonstrate that ABAC is the only feasible packing type and that a peak at 3.44(2) Å in PDF provides indirect evidence for the presence of tetrahedral Fe.