Precipitation of light elements from Earth’s liquid core: Can exsolution power the ancient geodynamo?

Earth’s core currently sustains a geodynamo through chemical convection in the liquid outer core. This power source originates from the growth of the solid inner core, where light elements are partitioned to the liquid at the lower most outer core. The inner core is expected to be ~1 Gyr old, meaning that for most of Earth history, the geodynamo required alternate power sources to produce a magnetic field. The paleomagnetic record shows that the field has been persistent for the last 3.5 Gyrs. Secular cooling is not capable of providing sufficient power for the geodynamo to remain active during this time if conductive heat transport is large. Recent experiments and calculations find that the thermal conductivity of the core is high, suggesting that the power available for geodynamo action would have been exhausted significantly before inner core growth began. Of the alternate power sources available to supplement secular cooling, precipitation of light elements is the most hopeful. We explore the solubility of silicon and other candidate light elements in iron-rich liquids of the core through ab initio calculations of partitioning. We apply these results to a thermodynamic model of partitioning, informed by experimental partitioning. When incorporated into thermal history models of the deep Earth, we find that the geodynamo can be sustained by silicon precipitation, provided that the oxygen concentration of the ancient core is less than 1.1 wt%. These results highlight the importance of the initial composition of the core and interaction between light elements on the available precipitative power in the core.