4:15 PM - 4:30 PM
[SCG60-15] Experimental study on the hydration rate of peridotites at forearc mantle wedge conditions
Keywords:forearc, mantle, fluid, serpentinite, antigorite, kinetics
Recent ground water studies in southwestern Japan suggest that slab-derived fluid upwells through the forearc mantle wedge without significant reaction with the country rocks (e.g., Kazahaya et al. 2014; Kusuda et al., 2014), which may provide a clue to understanding the hydrological budget in forearc regions. The rate of serpentinization is one of the primary parameter for constraining the flow regime of slab-derived fluid in the stagnant forearc mantle wedge. Hydration experiments for kinetic studies have been vigorously conducted previously at relatively low P-T conditions (up to ca. 350℃ and 0.3 GPa), in which olivine reacts with water to form the low T serpentine variety lizardite (or chrysotile) and brucite. However, antigorite is expected to be the dominant serpentine variety under the higher P-T condition corresponding to the forearc mantle wedge (350 to 650℃ and 1.0 to 2.0 GPa). Moreover, serpentine formation needs a silica source in addition to olivine (e.g., orthopyroxne) at the temperature above 450 to 500℃ due to the instability of phase assemblage serpentine + brucite.
In order to constrain the serpentinization rates of peridotite under the mantle wedge conditions, we conducted piston-cylinder experiments at temperature of 400 ℃ (brucite-present condition) and 500 to 580℃(brucite-absent condition), and pressure of 1.3 and 1.8 GPa. Three types of starting materials were prepared from the crushed powder of a San Carlos lherzolite xenolith: 1) olivine (Ol), 2) orthopyroxene (Opx) + clinopyroxene (Cpx) and 3) Ol + Opx. Hereafter these systems are abbreviated as OL, OPX+CPX and OL+OPX respectively. The starting materials were reacted with 15 wt% distilled water for 4 to 19 days. The hydration reaction proceeded in all the experiments, except for the OL system under the brucite-absent conditions. Based on Raman spectroscopy results and crystal shapes, the synthesized serpentine minerals were identified as lizardite in most of the runs except for antigorite in the OL+OPX system at 1.8 GPa. The Al2O3 in the system possibly stabilized the aluminous lizardite (Caruso and Chernosky, 1979). In the OL+OPX system, the reaction progress followed a diffusion-controlled rate law in the brucite-present condition and an interface-controlled rate law in the brucite-absent conditions. The rate constants were estimated to be 1.5 × 10-16 m2/s and 8.7 × 10-12 to 1.5 × 10-11 m/s in the brucite-present and the brucite-absent condition, respectively.
We applied the experimentally-obtained hydration rates of peridotites to a reactive-transport model for the stagnant mantle wedge hydration. In the case of grain-scale pervasive flow, slab-derived fluid is completely fixed in the mantle wedge peridotite. Otherwise, aqueous fluid possibly penetrate all the way through the mantle wedge via crack-like pathways (we assumed the vertical distance of 10 km) with a spacing > 0.025 to 0.80 m in the brucite-present conditions and > 2.6 to 4600 m in the brucite-absent condition. This indicates that slab-derived fluid may upwell easily through a cold forearc mantle wedge like in Western Shikoku rather than a warm forearc mantle wedge like in Cascadia.
In order to constrain the serpentinization rates of peridotite under the mantle wedge conditions, we conducted piston-cylinder experiments at temperature of 400 ℃ (brucite-present condition) and 500 to 580℃(brucite-absent condition), and pressure of 1.3 and 1.8 GPa. Three types of starting materials were prepared from the crushed powder of a San Carlos lherzolite xenolith: 1) olivine (Ol), 2) orthopyroxene (Opx) + clinopyroxene (Cpx) and 3) Ol + Opx. Hereafter these systems are abbreviated as OL, OPX+CPX and OL+OPX respectively. The starting materials were reacted with 15 wt% distilled water for 4 to 19 days. The hydration reaction proceeded in all the experiments, except for the OL system under the brucite-absent conditions. Based on Raman spectroscopy results and crystal shapes, the synthesized serpentine minerals were identified as lizardite in most of the runs except for antigorite in the OL+OPX system at 1.8 GPa. The Al2O3 in the system possibly stabilized the aluminous lizardite (Caruso and Chernosky, 1979). In the OL+OPX system, the reaction progress followed a diffusion-controlled rate law in the brucite-present condition and an interface-controlled rate law in the brucite-absent conditions. The rate constants were estimated to be 1.5 × 10-16 m2/s and 8.7 × 10-12 to 1.5 × 10-11 m/s in the brucite-present and the brucite-absent condition, respectively.
We applied the experimentally-obtained hydration rates of peridotites to a reactive-transport model for the stagnant mantle wedge hydration. In the case of grain-scale pervasive flow, slab-derived fluid is completely fixed in the mantle wedge peridotite. Otherwise, aqueous fluid possibly penetrate all the way through the mantle wedge via crack-like pathways (we assumed the vertical distance of 10 km) with a spacing > 0.025 to 0.80 m in the brucite-present conditions and > 2.6 to 4600 m in the brucite-absent condition. This indicates that slab-derived fluid may upwell easily through a cold forearc mantle wedge like in Western Shikoku rather than a warm forearc mantle wedge like in Cascadia.