Back to Search Start Over

Rates of dehydration in hydrous, high-Fo, magmatic olivines.

Authors :
Barth, Anna
Plank, Terry
Towbin, Henry
Source :
Geochimica et Cosmochimica Acta. Feb2023, Vol. 342, p62-73. 12p.
Publication Year :
2023

Abstract

Natural olivine from the Etna Fall Stratified eruption (3930 BP) was dehydrated to determine hydrogen diffusivity. Dehydration was carried out in several time-steps, from 0.5 to 40 cumulative hours, in a 1 atm. furnace at 810 and 1000 °C. This starting material has high H 2 O concentration (∼50 ppm), high forsterite (Fo ∼ 90 %), and high fO 2 (∼NNO + 1.2), reflecting characteristics of its host magma. H 2 O concentration was measured using Fourier transform infrared spectroscopy after each heating step. Spectral peaks were fit with Gaussian curves to determine the contribution from each H+ defect. Zonation profiles along each crystallographic direction were modelled using the 'whole-block' method to determine bulk and defect-specific H+ diffusivity. Our modelling reveals a common apparent diffusivity for all defects, which is faster along ([1 0 0]) than b ([0 1 0]) or c ([0 0 1]) by more than an order of magnitude. A common defect-specific diffusivity has not been found in experiments on pure forsterite, so it appears that the presence of Fe enables H+ from all defects to diffuse via the proton-polaron mechanism (supported by the observed anisotropy). However, the rates that we observe are slower than the proton-polaron mechanism, which supports the recent proposal by Ferriss et al. (2018) that the apparent diffusivity is rate-limited by reactions between H+ bound in defect sites and Fe2+. A coupled reaction and diffusion process may explain the observed decrease in diffusivity along a over time, as well as an apparent non-zero H 2 O concentration at the crystal edge. Both of these effects need to be considered when modelling H+ diffusion profiles to determine decompression rate – fixing the edge concentration to zero will underestimate the diffusion timescale. Arrhenius laws describing diffusion parallel to a (at the start and end of the experiments), b , and c are developed from the diffusivity at 810 and 1000 °C: Da start = 10 - 2.94 e x p - 182000 RT , Da end = 10 - 4.63 e x p - 154000 RT , Db = 10 - 2.35 e x p - 243000 RT , Dc = 10 - 4.01 e x p - 187000 RT , where R is the gas constant 8.314 (J/mol K); T is the temperature (K), the units of the diffusivities (D) are in m2/s. A comparison of experimental dehydration studies on natural olivines reveals that the diffusivity during dehydration likely depends on how the olivine was experimentally hydrated. To avoid this experimental effect and understand the diffusivity operating during ascent in a volcanic conduit, crystals which are naturally hydrous (such as the Etna olivines in this study) are particularly useful. Despite some ambiguity in the literature owing to different experimental approaches, our results suggest that for the studied range of pressure and temperature, forsterite content is the major control on H+ diffusivity, while H 2 O concentration and the fO 2 at which the crystal last equilibrated have a smaller effect within the ranges studied for magmatic olivines. [ABSTRACT FROM AUTHOR]

Details

Language :
English
ISSN :
00167037
Volume :
342
Database :
Academic Search Index
Journal :
Geochimica et Cosmochimica Acta
Publication Type :
Academic Journal
Accession number :
161362511
Full Text :
https://doi.org/10.1016/j.gca.2022.11.009