New Ni and Co metal-silicate partitioning data and their relevance for an early terrestrial magma ocean

Abstract

It is generally assumed that the abundances of the siderophile (metal-loving) elements in the Earth's mantle are the result of core formation in the early Earth. In the most popular model the low contents of siderophile elements in the Earth's mantle are the result of metal-silicate equilibration followed by metal accumulation at the base of a magma ocean. This model is primarily based on the pressure dependence of Ni and Co metal-silicate partition coefficients first determined by Li and Agee [Li, J., Agee, C.B., 1996. Geochemistry of mantle–core differentiation at high pressures. Nature 381, 686–689]. We have undertaken a further test of this hypothesis by performing a large number of additional partition experiments with the two key elements Ni and Co at a variety of temperatures and a pressure range extending from ambient pressure to 25 GPa (~750 km depth in the Earth). The unexpected result of the new experiments is the recognition of two pressure regimes in the temperature and pressure dependences of Ni–Fe and to a lesser extent of Co–Fe exchange metal-silicate partition coefficients, KDNi–Fe and KDCo–Fe. The strong pressure and temperature dependences of KD⁎Ni–Fe and KD⁎Co–Fe at atmospheric pressure change below about 5 GPa to much weaker dependences at higher pressures. As a consequence the pressure dependences of KDNi–Fe and KDCo–Fe cannot be regressed with a single fit over the entire pressure range. Fits based on partition data obtained in this work and from the literature between 5 GPa and 25 GPa were extrapolated to pressures and temperatures to lower mantle liquidus conditions. The Co–Fe partition coefficients required to satisfy the present mantle composition is reached at about 35 GPa, whereas the proper Ni–Fe ratio requires 45 to 50 GPa. At 55 GPa Ni and Co partition coefficients would be identical but the Ni- and Co-contents imposed on the Earth's mantle by metal-silicate equilibration would be too low. This would exclude single stage magma ocean models for producing approximately chondritic Ni/Co ratio in the Earth's mantle. However, the accuracy of the experimental data obtained here and from the literature is not sufficient to unambiguously exclude single stage equilibration. The results of this and other recent studies exclude that Fe-, Ni-, and Co-contents of the present upper mantle were established by metal-silicate equilibration at upper mantle conditions (<25 GPa).