Our research focuses mainly on the inaccessible Earth's interior. We are interested in many things, such as chemical compositions, mineral constitutions, and the thermal structure of the Earth's present interior. We also wish to know how the interior structures were initiated at the beginning of the Earth's history, and how they have developed to their present states.

There are several possible methods of obtaining information about the Earth's interior. Using the seismic wave method to determine the elastic structures of the Earth's interior is one such method. The Earth's interior can also be modeled using geoelectric and geomagnetic methods. Mineralogists and petrologists study and describe minerals and rocks from a certain depth in the Earth, and geochemists conduct chemical analyses on these materials to obtain more information about them. For a more comprehensive understanding of the Earth's interior from these observations and analyses, however, we need to know more about the physical and chemical properties of the materials that are thought to compose the Earth's interior.

The Earth's interior is under extremely high pressure and high temperature conditions (henceforth, P-T conditions), and it is difficult to determine material properties under such harsh conditions. We have therefore been making continuous efforts to develop technologies to better detect and determine material properties at high P-T conditions.

We have studied phase relations for more than six years by using the conventional quench method and have clarified the phase relations of the olivine-wadsleyite-ringwoodite transitions in the Mg₂SiO₄-Fe₂SiO₄ system. We have also studied the post-spinel transitions in Fe₂SiO₄ and the Fe-Mg partitionings between (Mg,Fe)SiO₃ perovskite and (Mg,Fe)O periclase by utilizing fluxes.

More recently, we have studied phase relations using high P-T in situ X-ray diffraction at theSPring-8 synchrotron radiation facility. Using this method, we have determined the phase relations of ilmenite-perovskite in MgSiO₃, the post-spinel transition in Mg₂SiO₄, and the olivine-wadsleyite transition in (Mg,Fe)₂SiO₄. Currently, we are studying the binary post-spinel transition in (Mg,Fe)₂SiO₄.


2. Development of high-pressure apparatus

In this decade, the maximum pressure able to be generated by a KAWAI-type high-pressure apparatus has been greatly extended by using sintered diamond (SD) anvils, although it is almost saturated by using tungsten carbide anvils. However, the two drawbacks of SD anvils are their cost and brittleness. Therefore, it is necessary to develop a high-pressure apparatus with a very precise high-pressure vessel. We have designed and constructed a new high-pressure apparatus, called the SPEED-Mk.II, at the SPring-8 synchrotron radiation facility whose cubic compression space is well kept with an increasing press load. Using this apparatus, we have successfully generated a pressure of 54 GPa. One of the serious problems of in situ X-ray diffraction in a multi-anvil press is grain growth. Because the diffraction region is limited in a high-pressure apparatus, the number of grains in the diffraction area rapidly decreases by heating, and this leads to the disappearance of diffraction peaks. The SPEED-Mk.II, however, is equipped with an oscillation system, enabling us to obtain high-quality diffraction patterns even at high temperatures. This is a special function of the SPEED-Mk.II that enables us to determine the phase boundary of the B1-B2 transition in NaCl in addition to the thermal expansion coefficient of Mg₂SiO₄ ringwoodite at high pressures and temperatures.


3. Thermal expansion of mantle minerals and mantle geotherm

Heat is mainly transported by convection in the Earth's interior, and the temperature gradient in the Earth is nearly adiabatic. This adiabatic temperature gradient in the Earth (∂T/∂z)_s can be expressed as:
(∂T/∂z)_s = αgT/C_p
where T is the temperature, z is the depth, g is the gravitational acceleration, and α and C_p are the thermal expansion coefficient and heat capacity at constant pressure of the constituent materials, respectively. Because g and C_p are nearly constant, the thermal expansion coefficients of mantle minerals become important to in the estimation of the mantle geotherm. The thermal expansion coefficient depends on the pressure and temperature; therefore, it must be measured at the realistic P-T conditions in the mantle.

Volume measurement obtained by high P-T in situ X-ray diffraction is a rather practical method to determine the thermal expansion coefficients of minerals at simultaneous high pressures and temperatures. Though it had previously been difficult to obtain high-quality diffraction patterns at high temperatures because of grain growth, our new KAWAI-type apparatus, the SPEED-Mk.II, enables us to obtain high-quality diffraction patterns at high temperatures. Using this apparatus, we have already determined the thermal expansion coefficient of Mg₂SiO₄ ringwoodite at temperatures of 300 to 2000 K and at pressure of 16 to 21 GPa, and of MgSiO₃ perovskite at temperature of 300 to 2300 K and pressure of 17 to 30 GPa. We further plan to measure and determine the thermal expansion coefficients of olivine, wadsleyite, and majorite. We also plan to extend the pressure range of thermal expansion coefficient measurement for MgSiO₃ perovskite.


4. Electrical conductivity of the Earth's materials

Electrical conductivity profiles in the mantle have been estimated by geomagnetic and geoelectrical observations. We hope to obtain important information about the structures of the mantle by interpreting the electrical conductivity profiles in the mantle using laboratory data. For this reason, we have tried to measure the electrical conductivity of mantle minerals at high pressures and temperatures under various physical and chemical conditions. We constructed a system for high P-T impedance measurements that consists of function generators, digital multimeters, reference resistance, and a personal computer. This system can withstand enormously noisy environments and enables us to measure the resistance of mineral samples at high pressures and temperatures. We used this system to measure the electrical conductivity of (Mg,Fe)SiO₃ perovskite at the physical conditions in the top of the lower mantle. We will subsequently try to measure the electrical conductivity of (Mg,Fe)SiO₃ perovskite at higher pressures using sintered diamond anvils. 　This system can also be used for studies in different areas. For example, we have used it to simulate the core formation processes at the early stage of the Earth by estimating the connectivity of metal-sulfide in silicate matrix by means of electric conductivity measurements. We also have used it to measure the electrical conductivity of rocks from the lower crust of Japan and to estimate the thermal structure of the island arc of Japan. Finally, we have used it to study the electrical conductivity of brucite in order to better understand the behavior of protons in hydrous minerals.


5. Elasticity of minerals

Seismology is a powerful method used to observe the Earth's interior. We can obtain more information on the composition and thermal structure of Earth's interior by comparing the seismological observations with the elastic properties of the Earth's constituents. Therefore, the determination of elastic properties is one of the most important issues of this field of research. Although we do not measure the elastic properties of minerals by ourselves; our contribution to this area of research is the synthesizing of high-quality samples for elasticity measurements, as follows.

The resonance technique is a method used to precisely determine the elastic moduli of materials. Because of the high precision of this technique, we are able to determine the temperature dependence of elastic moduli by elevating the temperatures only by several tens of kelvins. High pressure minerals are unstable at ambient pressures and can easily decompose or become amorphous at high temperatures. Hence, it is a very useful technique to determine the temperature derivatives of elastic properties particulary for high pressure minerals. We synthesized high-quality sintered aggregates of wadsleyite and ringwoodite with (Mg_0.9Fe_0.1)₂SiO₄ composition and perovskite with MgSiO₃ composition, and discovered the temperature dependence of adiabatic bulk and shear moduli of these minerals. Our next goal is to conduct measurement for single crystal samples of high-pressure mantle minerals, as elastic properties measured for sintered aggregates are always problematic due to the presence of a grain boundary. We are now trying to synthesize of large single crystals of mantle minerals 1-mm size.

Another important method used to determine the elastic moduli of high pressure minerals is Brillouin scattering. The Brillouin scattering method does not require large samples. To date, we have synthesized single crystals of (Mg_0.9Fe_0.1)₂SiO₄ wadsleyite and ringwoodite, and Professor Jay Bass, Dr. Stas Sinogeikin, and their colleagues at the Univesity of Illinois have determined their elastic moduli and pressure and their temperature derivatives. We are now trying to synthesize large single crystals of high-pressure mantle minerals also for Brillouin scattering.

The synthesis of large single crystals is useful for many kinds of measurement of the physical properties of minerals, e.g., IR and Raman spectroscopies, high-resolution X-ray inelastic scattering, the measurement of diffusion coefficients, and so on. We hope that the synthesized large single crystals can be utilized in a variety of studies.

6. Heat transfer properties of mantle minerals

In most parts of the mantle, the transfer of heat is done by convection. However, in some regions, such as in the subducted slabs, heat is transferred by conduction. In order to reveal the thermal structure of the subducted slabs, we must measure the thermal diffusivity of minerals at high pressures and high temperatures. We developed a technique to measure the thermal diffusivity of minerals at pressures up to 10 GPa and temperatures ranging from 350 K to 1700 K using the Angstrom method in cylindrical geometry. We used this method to measure the thermal diffusivity of silica glass, olivine, and periclase.

There are two major kinds of heat transfer mechanisms: conduction and radiation. In a higher temperature heat transfer, the contribution of radiation could be larger than that of conduction. The method that we have adopted cannot correctly detect the amount of heat transfer by radiation in samples of limited sizes. Therefore, we have recently not been able to make significant progress in the study of heat transfer. However, we are planning to search for radically better methods of determining heat transfer properties by both conduction and radiation.