1Forrest Gilfoy,1Jie Li
Earth and Planetary Science Letters 541, 116285 Link to Article [https://doi.org/10.1016/j.epsl.2020.116285]
1Department of Earth and Environmental Sciences, University of Michigan, Ann Arbor, MI, 48109, USA
A series of multi-anvil experiments have been conducted to define the iron-rich liquidus of the iron-nickel-sulfur (FeNi-S) system at 20 GPa, the estimated pressure of the martian core-mantle boundary (CMB). The liquidus curve of FeNi-S containing about 9 wt.% Ni has a concave up shape, and is as much as 400 K lower than the liquidi previously applied to the martian core with sparse experimental constraints. Unlike existing liquidi of Fe-S and FeNi-S at 23 GPa, which predict a fully molten core for a narrow range of sulfur content between 14 and 15 wt.% S, our results are consistent with a molten state for all proposed core compositions, and establishes a new minimum CMB temperature of 1500 K for 10 wt.% S and 1250 K for 16 wt.% S. Extrapolating our FeNi-S liquidus to high pressures and comparing it to calculated areotherms, we find that three core crystallization regimes are possible. For a martian core with moderate sulfur content (10 to 13 wt.%) or lower, crystallization takes the form of iron snow near the CMB, while for cores with higher sulfur content (15-16 wt.%), solidification occurs near the center of the planet in the form of solid Fe3S. At an intermediate sulfur content of 14 wt.%, Fe3S would precipitate over a broad depth range and may appear fully molten to surface observations.