Dehydration-hydration mechanisms in synthetic Fe-poor diopside
Sundvall, Rickard; Skogby, Henrik; Stalder, Roland
published: Feb 6, 2009
ArtNo. ESP147052101004, Price: 29.00 €
Small amounts (ppm) of OH in nominally anhydrous minerals (NAMs) can have a dramatic effect on the physical properties of the upper mantle. The pyroxenes of the upper mantle have been shown to incorporate substantial numbers of protons forming hydroxyl ions. Enstatite and diopside are the most important endmembers of the pyroxenes in terms of bulk volume in the upper mantle. To further constrain the behavior of hydroxyl ions in clinopyroxene, the dehydration-hydration mechanisms of synthetic 57Fe-doped diopside were investigated. Dehydration was carried out by stepwise heating in air of crystals synthesized at high pressure under water-saturated conditions. FTIR spectra were obtained after each step. Mössbauer spectra were recorded for three of the crystals when there had been a significant decrease in FTIR absorbance intensity. From the Mössbauer spectra we see an increase in the Fe3+ doublet with successive dehydration, although this increase is less than the decrease in OH in terms of atoms per formula unit. This means that the dehydration only partly follows the redox reaction OH− + Fe2+ = O2− + Fe3+ + ½H2, and that additional reactions occur. Hydration experiments were conducted on one crystal in the same manner as the dehydrations, with the exception that hydrogen gas was used during heating. Hydration experiments resulted in re-hydration of the sample to 73 % of the original amount of OH. The calculated Arrhenius equation derived from the diffusion rates during dehydration along  yields an activation energy (Ea) of −292 ± 50 kJ mol−1, and D0 = 10±1.9 + 2.3 m2 s−1. The result of the rehydration experiment agrees well with the established diffusion law. Diffusion rates determined for synthetic diopside are almost two orders of magnitude slower than for synthetic enstatite with comparable Fe contents. Compared to natural diopside, diffusion rates in these synthetic samples are slower, probably because of the low iron content. Ea is similar to that of dehydration of pure and low-Fe enstatite.