“Real-time” core formation experiments using X-ray tomography at high pressure and temperature

Abstract:

The process of differentiation is a defining moment in a planet’s history. Direct observation of this process at work is impossible in our solar system because it was complete within the first few tens of millions of years. Geochemical and geophysical evidence points to magma ocean scenarios to explain differentiation of large planets such as Earth. Smaller planets and planetesimals likely never achieved the high temperatures necessary for wide scale melting. In these smaller bodies, silicates may have only partially melted, or not melted at all. Furthermore, isotopic signatures in meteorites suggest that some planetesimals differentiated within just a few million years. Achieving efficient core segregation on this rapid timescale is difficult, particularly in a solid or semi-solid silicate matrix. Direct measurements of metallic melt migration velocities have been difficult due to experimental limitations and most previous work has relied on geometric models based on 2-D observations in quenched samples. We have employed a relatively new technique of in-situ, high pressure, high temperature, X-ray micro-tomography coupled with 3-D numerical simulations to evaluate the efficiency of melt percolation in metal/silicate systems. From this, we can place constraints on the timing of core formation in early solar system bodies. Mixtures of olivine and KLB-1 peridotite and up to 12 vol% FeS were pre-synthesized to achieve an initial equilibrium microstructure of silicate and sulfide. The samples were then were then pressed again to ~2GPa, and heated to ~1300°C to collect X-ray tomography images as the partially molten samples were undergoing shear deformation. The reconstructed 3-D images of melt distribution were used as the input for lattice Boltzmann simulations of fluid flow through the melt network and calculations of permeability and melt migration velocity. Our in-situ x-ray tomography results are complemented by traditional 2-D image analysis and high-resolution 3-D imaging of the starting materials and quenched samples using a focused ion beam/SEM cross beam instrument.