The crystallisation sequence of Earth's core

Earth's core plays a key role in making the planet active, such as supplying heat to the mantle to partially drive tectonic processes and generating magnetic fields by powering the geodynamo, but the processes are poorly known and understood. Here we will elucidate how Earth's core functions by examining the processes of demixing and crystallisation in an impure liquid core. The core is believed to consist of Fe alloy, but with small amounts of impurities that nevertheless have a strong effect on its properties. This so-called 'light element' content accounts for the core's 4-7 % density deficit relative to pure iron, changing the crystallising temperature of core materials and how core convection is powered. Among proposed candidate light elements, silicon and oxygen are likely present in the ancient core as a consequence of metal-silicate interaction in the magma ocean early in Earth's history, although their respective concentrations are yet to be determined. This project will determine (i) the crystallisation sequence of Fe-Si-O core liquids by constructing a thermodynamic model for their behaviour under high pressure (P) and temperature (T) conditions, and (ii) the chemical evolution of the core and its influences on powering the geodynamo. Constraining cooling process of the core is fundamental for understanding the origin, current state, and evolution of Earth's core, which are linked to surface environment and life through generating magnetic fields that protect the Earth's atmosphere and biosphere from the solar wind and harmful ionising radiation.

This project will take a coupled approach to (i) by combining PI's own novel and very precise laboratory measurements with thermodynamic modelling. We will examine eutectic melting and subsolidus phase relations in the system Fe-Si-O by high-P-T experiment in internally resistive-heated diamond anvil cells (DAC) that we have developed. This technique produces extreme conditions up to P = 200 GPa and T = 4000 K and most notably the temperature precision is much better due to resistive heating (+/-50 K) than in the conventional laser-heating system (ca. +/-200 K at 3000 K). Sample analysis will be made at synchrotron facilities for in-situ X-ray diffraction measurements and University of Edinburgh for chemical and textural observations. We will also employ thermodynamic calculations using the experimentally constrained eutectic melting points to obtain properties of liquids under high-P-T conditions which are not directly constrained by experiment. Use of the internally resistive-heated DAC is key to accurately constraining the thermodynamic properties of liquids, which was not possible until now. From the constructed thermodynamic model, we will calculate the crystallising phase relations and then determine how fractional chemical differentiation occurs upon core cooling either by precipitating SiO2 or demixing.

We will employ geodynamic calculations using obtained physical parameters for crystallisation process to examine (ii), as part of which, we will test the wide range of proposed thermal conductivity of iron and the Si/O ratio of the ancient (i.e., starting) core by comparing the resulting dynamo history and paleomagnetic record of field intensity. Thus, we will report a consistent set of the Si/O ratio in the ancient core, iron conductivity, and recorded paleomagnetic data of field intensity. We will then determine the physicochemical conditions of core formation including the oxygen fugacity of the magma ocean from the Si/O ratio in the ancient core. The constrained conductivity value will also provide a new estimate for the energy flux across the core-mantle boundary and hence its influence on mantle convection. This new insight into the nature of the core drives understanding of fundamental Earth processes through time, and will be pivotal to understanding how the Earth functions, including its surface environment and its ability to sustain life.