Periodic Reporting for period 3 - UltraLVP (Chemistry and transport properties of bridgmanite controlling lower-mantle dynamics)
Période du rapport: 2021-10-01 au 2023-03-31
Seismic tomography indicates that the descent of some slabs is impeded in the region from 660 to 1000 km depth, referred to as the mid-mantle. It is also indicated that the ascent of plumes might be accelerated in the mid-mantle. A recent evaluation of viscosity variation suggested that the viscosity may increase or have a maximum in the mid-mantle. This viscosity increase or maximum may impede the slab subduction and accelerate the plume upwelling. However, there is no first-order phase transition in the mid-mantle, and therefore, the possible viscosity variation should be caused by the rheological change of the dominant lower-mantle constituent mineral, bridgmanite. The absence of seismic anisotropy in the significant part of the lower mantle suggests the dominance of diffusion creep. Element diffusivities and grain size are two essential factors of diffusion creep, and defect chemistry controls element diffusivity. Therefore, the defect chemistry, diffusivity, and grain-growth rate of bridgmanite under mid-mantle conditions are to be investigated in this project.
Although the multianvil press is the most reliable tool to investigate mineral properties under high P-T conditions, the pressure and temperature ranges of the conventional multianvil technology are insufficient for the present purpose. However, we recently developed multianvil technology that can generate mid-mantle conditions. Therefore, we determine the defect chemistry, element diffusivity, and grain growth rate of bridgmanite under mid-mantle conditions using our advanced multianvil technology to understand the origin of the viscosity increase that may cause the slab stagnation and plume acceleration in the mid-mantle.
Until recently, the majority of geophysicists and mineralogists considered that it should be at the 660-km discontinuity where physical properties significantly change because of the first-order phase transition of the dissociation of ringwoodite to bridgmanite plus ferropericlase. As mentioned above, however, the geophysical observations have declined this expectation and shown that the change occurs in the mid-mantle. Therefore, Clarifying the origin of the change in rheology in the mid-mantle is currently an essential task for geophysics and mineralogy.
The current project also has particular importance in high-pressure science. Before our technological development, the pressure limit of multianvil presses with carbide anvils has been kept up to 26 GPa, which has remained unchanged for a few decades. Although much higher pressures (~120 GPa) can be generated using sintered diamond anvils, the use of sintered diamond anvils is impractical due to the required extremely high experimental skills and extremely high prices. Our technology can generate pressures to 40 GPa easily and 50 GPa with some special efforts. Thus we have extended the pressure range by 1.5 to 2 times wider. We also attempt to expand the temperature range of multianvil experiments. Although the practical temperature range of multianvil experiments is 300 to 2000 K, a recent innovation allows us to generate a temperature of 4000 K using a boron-doped diamond heater. We attempt to make this technique more practical for many multianvil workers.
We further aim to expand the pressure range at extremely high temperatures by combining a boron-doped diamond heater and sintered diamond anvils. Our final target is the simultaneous generation of 80 GPa and 2500 K. There are many applications of these technologies not only in Earth science but also in material science. Further development of multianvil technologies through the current project will bring tremendous benefit to both scientific communities.
Although the multianvil press is the most reliable tool to investigate mineral properties under high P-T conditions, the pressure and temperature ranges of the conventional multianvil technology are insufficient for the present purpose. However, we recently developed multianvil technology that can generate mid-mantle conditions. Therefore, we determine the defect chemistry, element diffusivity, and grain growth rate of bridgmanite under mid-mantle conditions using our advanced multianvil technology to understand the origin of the viscosity increase that may cause the slab stagnation and plume acceleration in the mid-mantle.
Until recently, the majority of geophysicists and mineralogists considered that it should be at the 660-km discontinuity where physical properties significantly change because of the first-order phase transition of the dissociation of ringwoodite to bridgmanite plus ferropericlase. As mentioned above, however, the geophysical observations have declined this expectation and shown that the change occurs in the mid-mantle. Therefore, Clarifying the origin of the change in rheology in the mid-mantle is currently an essential task for geophysics and mineralogy.
The current project also has particular importance in high-pressure science. Before our technological development, the pressure limit of multianvil presses with carbide anvils has been kept up to 26 GPa, which has remained unchanged for a few decades. Although much higher pressures (~120 GPa) can be generated using sintered diamond anvils, the use of sintered diamond anvils is impractical due to the required extremely high experimental skills and extremely high prices. Our technology can generate pressures to 40 GPa easily and 50 GPa with some special efforts. Thus we have extended the pressure range by 1.5 to 2 times wider. We also attempt to expand the temperature range of multianvil experiments. Although the practical temperature range of multianvil experiments is 300 to 2000 K, a recent innovation allows us to generate a temperature of 4000 K using a boron-doped diamond heater. We attempt to make this technique more practical for many multianvil workers.
We further aim to expand the pressure range at extremely high temperatures by combining a boron-doped diamond heater and sintered diamond anvils. Our final target is the simultaneous generation of 80 GPa and 2500 K. There are many applications of these technologies not only in Earth science but also in material science. Further development of multianvil technologies through the current project will bring tremendous benefit to both scientific communities.
SP1. We attempted to make it possible to use CVD-synthesized boron-doped diamond as a heating element. We successfully generated a temperature of 2700 K at 15 GPa (Fig. 1). Furthermore, we successfully generated 52 GPa and 3300 K simultaneously using sintered diamond anvils and a boron doped-diamond heater. Since the initial target was 80 GPa and 2500 K, we generated higher temperature than aimed. On the other hand, the target pressure has not been achieved yet.
SP2. We determined bridgmanite chemistry in MgO-Al2O3-SiO2, MgO-Fe2O3-SiO2, and MgO-Al2O3-Fe2O3-SiO2 systems as a function of temperature at a constant pressure of 27 GPa and as a function of pressure at a constant temperature of 2000 K. What we mainly found is as follows. 1) In both MgO-Al2O3-SiO2 and MgO-Fe2O3-SiO2 systems, the oxygen vacancy components are contained under MgO-rich conditions, whereas they are zero under MgO-poor conditions (Fig. 2). (2) The oxygen vacancy contents in MgO-Fe2O3-SiO2 are low (Fig. 3), and lower than MgO-Al2O3-SiO2. They are independent of the temperature and decrease with increasing pressure. (3) The FeAlO3 component dominates in bridgmanite chemistry. Bridgmanite can contain even larger amounts of FeAlO3 than the main component of MgSiO3. (4) The FeAlO3 contents are almost independent of temperature and pressure. (5) The oxygen vacancy contents are essentially zero when the FeAlO3 component dominates.
SP3. The Si lattice diffusivity in bridgmanite is found too small to measure because the diffusion lengths are comparable to the spatial resolution of the current method. The Si lattice diffusivity in bridgmanite previously reported seems an experimental artifact.
SP4. We found that the grain size exponent of the grain growth kinetics of bridgmanite + ferropericlase is much smaller than previously thought (Fig. 4). Therefore, the grain size of bridgmanite in the lower mantle should be larger than previously considered. We also found that the grain-growth rate decreases by two orders of magnitude by increasing the fraction of coexisting ferropericlase from zero to 20 vol.%. This effect can produce viscosity variation in the lower mantle. Furthermore, the growth rate significantly decreases with pressure. These experimental data lead to a large activation volume of 8~9 cm3/mol.
SP5. No results have been obtained.
SP2. We determined bridgmanite chemistry in MgO-Al2O3-SiO2, MgO-Fe2O3-SiO2, and MgO-Al2O3-Fe2O3-SiO2 systems as a function of temperature at a constant pressure of 27 GPa and as a function of pressure at a constant temperature of 2000 K. What we mainly found is as follows. 1) In both MgO-Al2O3-SiO2 and MgO-Fe2O3-SiO2 systems, the oxygen vacancy components are contained under MgO-rich conditions, whereas they are zero under MgO-poor conditions (Fig. 2). (2) The oxygen vacancy contents in MgO-Fe2O3-SiO2 are low (Fig. 3), and lower than MgO-Al2O3-SiO2. They are independent of the temperature and decrease with increasing pressure. (3) The FeAlO3 component dominates in bridgmanite chemistry. Bridgmanite can contain even larger amounts of FeAlO3 than the main component of MgSiO3. (4) The FeAlO3 contents are almost independent of temperature and pressure. (5) The oxygen vacancy contents are essentially zero when the FeAlO3 component dominates.
SP3. The Si lattice diffusivity in bridgmanite is found too small to measure because the diffusion lengths are comparable to the spatial resolution of the current method. The Si lattice diffusivity in bridgmanite previously reported seems an experimental artifact.
SP4. We found that the grain size exponent of the grain growth kinetics of bridgmanite + ferropericlase is much smaller than previously thought (Fig. 4). Therefore, the grain size of bridgmanite in the lower mantle should be larger than previously considered. We also found that the grain-growth rate decreases by two orders of magnitude by increasing the fraction of coexisting ferropericlase from zero to 20 vol.%. This effect can produce viscosity variation in the lower mantle. Furthermore, the growth rate significantly decreases with pressure. These experimental data lead to a large activation volume of 8~9 cm3/mol.
SP5. No results have been obtained.
We have made the following technical development beyond the project description.
(1) We succeeded in generating a temperature of 3300 K in sintered diamond anvil assembly, although the target temperature was 2500 K in the project description.
(2) We increased the high-temperature pressure limit using carbide anvils from 48 GPa to 52 GPa, which was not planned initially.
(3) We have developed a novel method to determine a phase boundary precisely using in situ X-ray diffraction, which was not planned in the project description.
We have been applying our advanced multianvil technology to many projects proposed by other researchers by collaboration with them. The following are examples that have been obtained.
(1) Determination of the binary loop thickness in the dissociation of ringwoodite to bridgmanite + ferropericlase (Publ. #5).
(2) Determination of the boundary of the post-garnet transition in Mg3Al2Si3O12 (in revision).
(3) Determination of the transition boundary between akimotoite and bridgmanite (in revision).
(4) Discovery of superhydrous silica phase (submitted).
(5) Synthesis of a novel hard material of rhenium pernitride ReN2 (Publ. #10).
(6) Nuclear magnetic resonance (NMR) spectroscopy on aluminous bridgmanite (Publ. #4).
(7) Synthesis of a new carbon phase (under review).
(1) We succeeded in generating a temperature of 3300 K in sintered diamond anvil assembly, although the target temperature was 2500 K in the project description.
(2) We increased the high-temperature pressure limit using carbide anvils from 48 GPa to 52 GPa, which was not planned initially.
(3) We have developed a novel method to determine a phase boundary precisely using in situ X-ray diffraction, which was not planned in the project description.
We have been applying our advanced multianvil technology to many projects proposed by other researchers by collaboration with them. The following are examples that have been obtained.
(1) Determination of the binary loop thickness in the dissociation of ringwoodite to bridgmanite + ferropericlase (Publ. #5).
(2) Determination of the boundary of the post-garnet transition in Mg3Al2Si3O12 (in revision).
(3) Determination of the transition boundary between akimotoite and bridgmanite (in revision).
(4) Discovery of superhydrous silica phase (submitted).
(5) Synthesis of a novel hard material of rhenium pernitride ReN2 (Publ. #10).
(6) Nuclear magnetic resonance (NMR) spectroscopy on aluminous bridgmanite (Publ. #4).
(7) Synthesis of a new carbon phase (under review).