New high-pressure and high-temperature metal/silicate partitioning of U and Pb: Implications for the cores of the Earth and Mars

Abstract

In order to quantify possible fractionation of U and Pb into a metallic core, we have performed piston cylinder and multi-anvil press experiments at high pressure (up to 20 GPa) and high temperature (up to 2400 °C) and obtained the distribution coefficient D metal–silicate and the exchange partition coefficient K metal–silicate for these elements between metal and silicates (mineral or liquid). D Pb metal–silicate and D U metal–silicate depend strongly on the S content of the metallic phase, and also on the oxygen fugacity, in agreement with an effective valence state of 4 for U in silicates and 2 for Pb in silicates. K Pb d metal–silicate and K U d metal–silicate show no discernable pressure and temperature trend. U remains lithophile even at high pressure and high temperature but its lithophile nature decreases at very low oxygen fugacity. From our experimental data, it was possible to calculate the U and Pb contents of the cores of Mars and Earth under core-mantle equilibrium conditions at high pressure and high temperature. From the D metal–silicate of the present study, we obtained that: 0.008 ppm < Pb in the core <4.4 ppm, and 0.0003 ppb < U in the core < 0.63 ppb, depending on whether the metal is S-free or S-saturated respectively, and if the mantle was molten or solid during the segregation process of the Earth’s core around ΔIW-2. For Mars, based on a core segregation process around ΔIW-1, we obtained that: 0.005 ppm < Pb in the core < 3 ppm, and 0.00002 ppb < U in the core < 0.05 ppb, depending on the metallic composition: S-free or S-saturated respectively. Our results suggest that the low concentration of Pb in the terrestrial mantle could not be explained by an early Pb sequestration in the Earth’s core even if S is the dominant light element of the core. If we assume a magma ocean scenario, U might produced a maximum value of 1.5% of the total heat budget of the core with a segregation occurring below ΔIW-3. The values found in the present study for U in the Martian core suggest that the magnetic field activity of Mars before ∼0.5 b.y. after its formation would be difficult to ascribe to the decay of U alone.