5:15 PM - 7:15 PM
[SCG62-P04] Influence of water on creep strength of ringwoodite
Keywords:ringwoodite, water content, mantle transition zone, stagnant slab, deformation experiment, in-situ X-ray observation
Ringwoodite is the major component mineral of the lower part of the Mantle Transition Zone (MTZ). The rheology of ringwoodite is important to understand the behavior of stagnant slabs at ~520-660 km depth because the viscosity contrast between the MTZ and lower mantle play an important role in the stagnation. It is known that ringwoodite can contain up to 2.7 wt.% H2O under water saturated condition (Kohlstedt et al., 1996). Electrical conductivity measurements of the MTZ and hydrous ringwoodite suggest that ringwoodite in the MTZ may contain up to 1 wt.% H2O (e.g. Yoshino et al., 2008; Kelbert et al., 2009). Recently, the dislocation creep strength of ringwoodite, which is thought to be the dominant flow mechanism under the MTZ conditions, was estimated experimentally (Kawazoe et al., 2016). However, there have been no experimental study to determine water dependent on the creep strength of ringwoodite. To reveal the influence of water on creep strength of ringwoodite, we conducted deformation experiments with polycrystalline ringwoodite of various content of water (~100-2200 wt. ppm H2O).
We carried out deformation experiments at ~22 GPa by in-situ X-ray observation method using D-111 type high-pressure deformation apparatus at the synchrotron facilities of PF-AR NE-7A beamline. 2D-XRD patterns and radiography images were taken every ~1-5 min to obtain stress-strain curves. The strain rates were 2.2-23 × 10-5 s-1 during constant-temperature deformation. We used two kinds of polycrystalline ringwoodite with different water contents as starting materials, those are synthesized at ~22 GPa and ~1000-1400°C from San Carlos olivine ((Mg0.91Fe0.09)2SiO4) monocrystal and Mg2SiO4 forsterite powder. When synthesizing from monocrystal, the samples were wrapped in iron foil to keep it dry. In contrast, when synthesizing from powder, the samples was packed on NaCl capsule. Their water contents were estimated to be ~100, 1800-2200 wt. ppm H2O, respectively, by FTIR measurements. For each sample, several deformation experiments were carried out at different temperatures to determine the temperature dependence of flow stress.
Based on the stress dependencies of the strain rates (i.e., n values), Mg2SiO4 ringwoodite deforms by Peierls mechanism or dislocation creep, depending on temperature in our experimental conditions. In both creep regimes, water significantly affects the strength of ringwoodite. Dry ringwoodite (~100 wt. ppm H2O) showed higher strength than the values estimated from the flow law constructed in the previous study (370-1100 wt. ppm H2O, Kawazoe et al., 2016), while hydrous ringwoodite (~1800 wt. ppm H2O) tended to be weaker. The effect of water on the strength of Mg2SiO4 ringwoodite can be described by incorporating a water-content exponent (r~1) and a water-content dependent activation enthalpy H*(CH2O) in the flow law. The latter value decreases with increasing water content. We obtained only preliminary results on the strength of iron-bearing ringwoodite, which showed a similar tendency, at least under dry conditions. We will further conduct deformation experiments, particularly using iron-bearing hydrous ringwoodite, to better constrain water-dependent viscosity in the lower part of the MTZ.
We carried out deformation experiments at ~22 GPa by in-situ X-ray observation method using D-111 type high-pressure deformation apparatus at the synchrotron facilities of PF-AR NE-7A beamline. 2D-XRD patterns and radiography images were taken every ~1-5 min to obtain stress-strain curves. The strain rates were 2.2-23 × 10-5 s-1 during constant-temperature deformation. We used two kinds of polycrystalline ringwoodite with different water contents as starting materials, those are synthesized at ~22 GPa and ~1000-1400°C from San Carlos olivine ((Mg0.91Fe0.09)2SiO4) monocrystal and Mg2SiO4 forsterite powder. When synthesizing from monocrystal, the samples were wrapped in iron foil to keep it dry. In contrast, when synthesizing from powder, the samples was packed on NaCl capsule. Their water contents were estimated to be ~100, 1800-2200 wt. ppm H2O, respectively, by FTIR measurements. For each sample, several deformation experiments were carried out at different temperatures to determine the temperature dependence of flow stress.
Based on the stress dependencies of the strain rates (i.e., n values), Mg2SiO4 ringwoodite deforms by Peierls mechanism or dislocation creep, depending on temperature in our experimental conditions. In both creep regimes, water significantly affects the strength of ringwoodite. Dry ringwoodite (~100 wt. ppm H2O) showed higher strength than the values estimated from the flow law constructed in the previous study (370-1100 wt. ppm H2O, Kawazoe et al., 2016), while hydrous ringwoodite (~1800 wt. ppm H2O) tended to be weaker. The effect of water on the strength of Mg2SiO4 ringwoodite can be described by incorporating a water-content exponent (r~1) and a water-content dependent activation enthalpy H*(CH2O) in the flow law. The latter value decreases with increasing water content. We obtained only preliminary results on the strength of iron-bearing ringwoodite, which showed a similar tendency, at least under dry conditions. We will further conduct deformation experiments, particularly using iron-bearing hydrous ringwoodite, to better constrain water-dependent viscosity in the lower part of the MTZ.