Rapid Core Formation in Terrestrial Planets by Percolative Flow: In-Situ Imaging of Metallic Melt Migration Under High Pressure/Temperature Conditions

Abstract

Core formation has left a lasting geochemical signature on the Earth. In order to constrain the composition of the Earth we must fully understand the processes by which newly formed Earth, and the bodies which accreted to it, differentiated. Percolation of iron-rich melt through solid silicate has been invoked as a mechanism for differentiation and core formation in terrestrial bodies in the early solar system. However, to date the contribution of percolation to core formation cannot be assessed due to the absence of data on Fe-rich melt migration velocities. Here we use a novel experimental design to investigate textural changes in an analog system, Au melt in polycrystalline h-BN, at 3 GPa, relevant to core formation in the early solar system. Using a combination of high resolution, in-situ X-ray tomography and fast 2-D radiographic imaging, we obtain the first direct data on melt migration velocities at high PT. Melt migration is highly variable and episodic, driven by variations in differential pressure during melt migration and matrix compaction. Smaller scale melt processes, representing migration of melt along pre-existing melt networks, give comparatively fast velocities of 0.6-60 µms −1. Ex-situ experiments are used to compare melt networks in analog systems to Fe-rich melt in silicates. Two competing processes for melt migration are percolation of melt along grain boundaries, and hydraulic fracturing induced by melt injection. Typically, both processes are noted in experimental and natural systems, although the relative importance of each mechanism is variable. Using a simple model for melt flow through a porous media, migration velocities determined here account for full differentiation of Earth-sized bodies within 10 1-10 3 Myr, for submicron diameter melt bands, or within a few Myr or micron-sized melt bands. This is consistent with rapid timescales inferred from geochemistry for core formation in planetesimals, implying that percolation may have had an important contribution to core differentiation in the Earth.

Reconciling fast and slow cooling during planetary formation as recorded in the main group pallasites

“…While previous models have suggested that the metal may have an internal source (eg., the molten core of the planetesimal, suggested by a ferrovolcanism origin; Johnson et al, 2020), recent isotopic studies show a statistically significant disequilibrium between the metal and silicate phases in pallasites, strengthening the argument for an external source delivered via impact (Windmill et al, 2022). Studies of core formation via percolative flow Berg et al, 2018) and intrusion propagation and emplacement alongside microstructural evidence from pallasite samples can be utilised to better understand this. Our model can aid in this research, as it provides a range of mantle temperatures over which pallasite-like textures can be produced.…”

A partially equilibrated initial mantle and core indicated by stress-induced percolative core formation through a bridgmanite matrix

“…where d is grain size and C is a constant that depends on the melt network geometry. Many studies have estimated the V m of coreforming melts in silicate (15,27,41,42,(44)(45)(46)(47)(48), and n ranging from 1 to 3.8 was proposed (42,47). Accordingly, the estimated V m varies by orders of magnitude.…”

Deformation of two-phase aggregates with in situ X-ray tomography in rotating Paris–Edinburgh cell at GPa pressures and high temperature

“…The common artefacts due to motions consisted of ring artefacts (e.g. Koeberl et al, 2002;Wang et al, 2011;Gharbi & Blunt, 2012;Mirone et al, 2014;Berg et al, 2018;Kastner & Heinzl, 2018), image 'blurriness', shading and 'triple-point' (or the so-called 'Mercedes') structures [Fig. 4(b)].…”

“…Philippe et al (2016) andA ´lvarez-Murga et al (2017) suggested its potential to image samples in situ at simple shear strain of > 2 or > 3. It has been employed for studying metallic phase transitions and melt percolation (Berg et al, 2018) at high pressure and temperature, but without the use of controlled deformation.…”