Geochemical tools such as the chemical and isotopic compositions of the samples with a wide range of ages have revealed the chemical evolution of the Earth. A model of the very early history of the earth with accumulation of the planetesimal, the giant impact, core segregation and late veneer has been proposed based on siderophile element abundance and isotopic composition of the old mantle-derived rocks. Especially, late veneer following core segregation hypothesis is supported by evidence that gradual increase with time in platinum group elements abundance of komatiite originated from deep mantle and the chondritic Os isotopic evolution of the earth's mantle. 182W (decay product of extinct 182Hf) anomaly relative to present-day mantle value of komatiite and modern ocean island basalt has indicated core-mantle interaction. These models have been proposed based on geochemical studies, but they still have many unclear problems. In line of this, it is very important to precisely constrain the behaviors and partition coefficients of the elements under deep-earth conditions. Here we mention these problems and propose importance of constraints by high-pressure and high-temperature experiments to provide data on behavior of elements, especially siderophile elements.
We conducted isotopic model calculations and high-pressure melting experiments in order to estimate the major element composition of the “missing reservoir”, which is a supposed component that should compensate the difference in 142Nd/144Nd ratio between the bulk silicate Earth and carbonaceous chondrite, from which the Earth is assumed to have formed. Our estimation demonstrated that the missing reservoir should have picritic to komatiitic composition, and that it was likely to have been lost from the Earth's surface by a giant impact event at the last stage of the Earth formation.
Noble gas isotopes have been a strong argument for the presence of primordial reservoir in the mantle, which has been isolated from the mantle convection throughout the Earth's history. However, recent understanding of mantle dynamics has revealed difficulty with the isolation within the mantle. Alternatively the core is a promising candidate, validity of which depends on partition coefficient of noble gases between liquid iron-rich metal and silicate melt at condition of core segregation in the magma ocean. A few experiments showed that the core could hold primordial noble gases though there are several uncertainties possibly due to extreme noble gas concentrations. For a better understanding of the primordial noble gas reservoir, we are conducting new experiments to determine noble gas partition coefficients between molten metal and silicate.
Carbon, the fourth most abundant element in the solar system, is believed to be an important light element constituent in the Earth's core. The high carbon content of carbonaceous chondrites (3.2 wt.%) compared to bulk earth estimates, the presence of graphite/diamond and metal carbides in Iron meteorites, the high solubility of carbon into iron melts in the Fe-C system, all suggests the plausible presence of carbon in the Earth's core. However, the distribution of carbon isotopes in the core and deep mantle remains elusive. Newly reported experimental data and theoretical estimates on equilibrium carbon isotope fractionation between graphite/diamond and carbide phases suggests that iron carbide melt will preferentially gather 12C than 13C. These results are consistent with the carbon isotope distribution between graphite and cohenite (Fe3C) observed in iron meteorites. The temperature dependent fractionation of carbon isotopes between carbide phases and elemental carbon can be an effective mechanism that might have created a 12C-enriched core. If the Earth's core is a large reservoir of 12C-enriched carbon, then it can result in large perturbations in surface carbon cycle caused by the flux of isotopically lighter carbon from the core-mantle boundary.