evolution of the Earth's mantle. It is widely accepted that the seismic discontinuities at 410 and 660 km depth (Dziewonski and Anderson 1981) are due to the phase transition from olivine to wadsleyite and from ringwoodite to bridgmanite + ferro-periclase in (Mg, Fe) 2 SiO 4 (e.g., Katsura and Ito 1989;Ito and Takahashi 1989). For the natural mantle composition, such as peridotite, the primary minerals are (Mg, Fe) 2 SiO 4 -related phases, and the secondary minerals are pyroxene and garnet in the system (Mg, Fe)SiO 3 -(Mg, Fe) 3 Al 2 Si 3 O 12· MgSiO 3 is the end component in a complicated composition of pyroxene. According to a previous study (Gasparik 1990), wadsleyite and stishovite change to akimotoite (MgSiO 3 ) with increasing pressure along the geotherm. In contrast, Sawamoto (1987) showed that wadsleyite and stishovite change into ringwoodite (Mg 2 SiO 4 ) and stishovite before the formation of akimotoite. This discrepancy is attributed to the estimation of the phase boundary between ringwoodite + stishovite and akimotoite. As this reaction can be observed in the relatively low-temperature region, the previous experiments (Sawamoto 1987;Gasparik 1990) might have involved a significant uncertainty related to the effects of the reaction kinetics.In our study, the use of a multi-anvil high-pressure system combined with a synchrotron radiation source enabled the acquisition of precise data for experimental pressures from samples under high-pressure and high-temperature conditions. To determine more accurate experimental pressures, we measured powder X-ray diffraction (XRD) data of gold, which was used as the pressure standard, in the sample chamber. Herein, we report on the disputed issue of the phase boundary between ringwoodite + stishovite and akimotoite in MgSiO 3 , and a revised pressure-temperature (P-T) phase diagram in MgSiO 3 based on our data will be suggested.
AbstractThe phase boundary between akimotoite and ringwoodite + stishovite in MgSiO 3 was determined using a multi-anvil high-pressure apparatus and synchrotron X-ray radiation. The phase relation was determined by observing the recovered samples using an electron microprobe analyzer. Experimental pressures were monitored by the in situ powdered X-ray diffraction data of gold, which was put in the sample chamber. The reaction boundary between akimotoite and ringwoodite + stishovite was found to occur at P (GPa) = 22.0-0.0012 × T (K). The pressure dependence of the slope of the reaction boundary, dP/dT, determined in our study was smaller than that determined by Gasparik (J Geophys Res 95:15751-15769, 1990). The triple point of ringwoodite + stishovite-wadsleyite + stishovite-akimotoite estimated in our study was at ~20 GPa and ~1700 K.