[S-CG62_2AM2] Geofluids and dynamics in subduction zones

[SCG62-11] Reaction progress and porosity change in hydrothermal alternation at Olivine/Quartz boundary

*Ryosuke OYANAGI¹, Atsushi OKAMOTO¹, Noriyoshi TSUCHIYA¹ (1.Graduate School of Environmental Studies, Tohoku University)

Keywords:serpentinization, ultramafic rock, Si-metasomatism, Hydrothermal alternation

Serpentinization in oceanic lithosphere is a fundamental process to bring water into deep earth’s interior. It is known that silica activity controls the reaction paths during the hydrothermal alternation of peridotites [e.g. 1,2], however the detailed reaction mechanism induced by silica transport is poorly understood. In this study, we conducted hydrothermal experiments in olivine (Ol)-quartz (Qtz)-H2O system for investigating the mechanism of silica metasomatism at crust/mantle boundary. Composite powders, which was composed of Qtz zone and Ol zone was set in inner tubes, with diameters of 1.7 mm and heights of 50 mm, and then loaded into autoclave with alkaline solution (NaOH,aq, pH = 13.8 at 25 ℃ ). Temperature and pressure are 250℃ and vapor-saturated pressure (= 3.98 MPa), respectively. After the experiments, the inner tube was cut into ten segments to evaluate the reaction progress as a function of the distance from Ol/Qtz boundary (hereafter denoted X), by Thermogravimetry and XRD. In order to evaluate the spatial variation of the reactions, the area of each minerals (olivine and reaction products) and pore was measured from the back-scattered electron (BSE) images of the thin section.After 46 days, the H2O content near the Ol/Qtz boundary is lower (3.9 wt.% H2O) than that in (12 wt.%) at the margin of the reaction tube. The reaction products after olivine changed systematically as away from Ol/Qtz boundary from smectite+serpentine zone to the serpentine+brucite zones. In the smectite+ serpentine zone, the (Mg+Fe)/Si ratio of the products increases from 0.5 to 1.5, indicating that proportion of serpentine with respect to smectite increased away from the boundary. With increasing time, the smectite+ serpentine zone was enlarged, where as the serpentine+brucite zones was retreated. Based on the combined analyzes of BSE images, TG and SEM-EDS, we obtained the reaction progresses of individual elementary reactions between 25 and 46 days as follows:(1) In the smectite+ serpentine zone, smectite was formed via hydration of olivine and dehydration of serpentine by supply of silica. As the result, overall reaction has a variation in the smectite+ serpentine zone; ∆mH2O is negative (hydration) at X=0-4 mm, it is positive (dehydration) at X=4-10 mm. Volume expansion factor (V/V0) is much higher (=1.4) at Ol/Qtz boundary than other zones (=~1.1), mainly due to Si-metasomatic reaction.(2) Far from the Ol/Qtz boundary(X = 20-40 mm), there is no influence of silica supply, indicating that silica was completely consumed in the smectite+ serpentine zone. In these area, serpentinization proceeds as the typical olivine hydration reaction to produce brucite and serpentine with constant Srp/Brc ratio. (3) In the transient zone, serpentine was formed by two ways: hydration of olivine and dehydration of brucite by supply of silica. These two serpentine forming reaction resulted in a large amount of serpentine in this area, and high volume expansion factor (=~1.4).Due to these two volume expansion reactions, low porosity (~5%) area developed locally, never-theless porosity of other area is 30%. The amount of silica (∆mSiO2,aq), which consumed from 25 to 46days, is largest at Ol/Qtz boundary, and monotonically decreases with increasing distance. If excess silica are available, the zones affected by silica will increase gradually with increasing time during hydrothermal alteration around mantle/crust boundary. In contrast, the porosity has a minimum around X = 15 mm in the transition zone, because Ol-hydration and Brc-dehydration reaction proceed with large volume expansion. Such volume expansion reaction and mineral changes causes the mechanical strength of boundary.References: [1] Frost, B. R., & Beard, J. S. (2007). Journal of Petrology, 48(7), 1351?1368. [2] Ogasawara, Y., Okamoto, A., Hirano, N., & Tsuchiya, N. (2013). Geochimica et Cosmochimica Acta, 119, 212?230.