The outcomes from this joint experimental-theoretical project represent important contributions to our understanding of the crystallization of the metallic core of the Earth and other terrestrial bodies, with a focus on the effect of core cooling on the magnetic history.

First, electrical experiments on core analogs in the Fe-S-O(+/-Mg,Si) system and extrapolation of the results up to 40 GPa and 3100 K indicated that electrical resistivity is highly sensitive to the nature and amount of light elements. For each composition, thermal conductivity-temperature equations were estimated using the experimental electrical results and a modified Wiedemann-Franz law. These equations were implemented to a thermochemical core cooling model to study the evolution of the dynamo. Modeling results suggest that bulk chemistry affects significantly the entropy available to power dynamo action during core cooling. In the case of Mars, we showed that the presence of oxygen would delay the dynamo cessation by up to 1 billion years compared to an O-free, Fe-S core. Models with 3 wt.% O can be reconciled with the inferred cessation time of the Martian dynamo if the core-mantle boundary heat flow falls from >2 TW to ~0.1 TW in the first 0.5 Gyrs following core formation.

Second, we conducted an experimental investigation of Ni-bearing alloys at 4.5 and 8 GPa to understand the effect of nickel on the electrical and thermal properties of metallic cores. Nickel is abundant in planetary cores (about 10 wt.% Ni is thought to be present in the Earth’s core) but its effect on transport properties is not well-known and requires experimental investigation. We showed that at defined temperature, Fe-Ni(-S) alloys are more resistive than Fe by a factor of ~ 3, and Fe-Ni alloys containing 5 and 10 wt.% Ni present comparable electrical resistivity values. Based on these electrical results and experimentally-determined thermal conductivity values from the literature, lower and upper bounds of thermal conductivity were calculated. For all Ni-bearing alloys, thermal conductivity estimates range between ~12 and 20 W/m.K over the considered pressure and temperature ranges, which is relevant to model the core of small planets and moons.

Third, we investigated the partitioning of silicon between the Earth’s mantle and core. Silicon is potentially present in significant amount in the Earth’s core and its incorporation into the core in the early history of the planet remains unclear. The objective of the study on Si partitioning is to understand how different chemical reactions (dissociation, dissolution, exchange) between a magma ocean and a metallic core affect the distribution of Si at depth and potentially provide significant power for generating the ancient geomagnetic field. Results from 30 silicate-metal partitioning studies have been compiled, spanning a 1-100 GPa pressure range and a 1883-5700 K temperature range. A new thermodynamic partitioning model has been developed to describe the partitioning (K_D) of Si and O. Our work shows a significant increase in K_D of Si with increasing temperature for the three reactions considered (dissociation, dissolution, exchange). Our model places bounds on the Si content of the Earth’s core through time and reveals the available power to the geodynamo from Si precipitation before the formation of a solid inner core.

Fourth, significant efforts were dedicated to the successful integration of a silicate mantle and a thermally stable layer in the core into our existing cooling model. 1-D parameterized models of the thermal evolution of a metallic core were developed, employing the latest constraints from laboratory (first activity mentioned above) and computational studies. Applied to a Mars-sized body, we observed that 94% of the successful models (i.e., models with a dynamo cessation after ~500 Myr) produce a core that is fully conductive with no convective region. We also showed that a low thermal conductivity makes it difficult to reproduce the dynamo cessation constraint. We recently modified this model to apply it to an example of a planet with a large core (Mercury).

Fifth, from a technical standpoint, PI Pommier has developed an electrical cell assembly for multi-anvil experiments at high pressure (>10 GPa) (10/5 assembly), allowing to achieve higher pressures than with the 14/8 assembly.

Sixth, the PI reviewed the physical properties of iron alloys to define the composition of terrestrial cores and dynamics that satisfy the currently available observational constraints. Possible “candidates” for core chemistry of several terrestrial bodies are discussed together with the input from geochemistry, especially metal-silicate partitioning experiments. The interpretation of field measurements (in particular, density and seismic velocity of cores) with laboratory experiments on iron alloys also aids in planning future investigations.

Finally, this NSF-NERC award has resulted in six peer-reviewed publications. Results from this project have also been shared widely through international scientific conferences and presentations at several institutions. This project resulted in professional development opportunities for 2 undergraduate students, 1 graduate student, and 3 postdoctoral researchers.

Last Modified: 05/07/2025
Modified by: Anne Pommier