A sulfur-poor terrestrial core inferred from metal–silicate partitioning experiments

• Published in: Earth and planetary science letters 2017-07, Vol.469, p.84-97
• Main Authors: Suer, Terry-Ann, Siebert, Julien, Remusat, Laurent, Menguy, Nicolas, Fiquet, Guillaume
• Format: Article
• Language: English
• Subjects: Condensed Matter; core formation; diamond anvil cell; Earth Sciences; Materials Science; metal–silicate partitioning; Mineralogy; Physics; quantitative NanoSIMS analyses; Sciences of the Universe; sulfur
• ISSN: 0012-821X; 1385-013X
• DOI: 10.1016/j.epsl.2017.04.016

As a siderophile and a volatile element, sulfur's partitioning behavior allows constraints to be placed on processes in the primitive Earth. Sulfur's core–mantle distribution during Earth's accretion has consequences for core content and implications for volatile accretion. In this study, metal–silicate partitioning experiments of sulfur were conducted in a diamond anvil cell at pressures from 46 to 91 GPa and temperatures between 3100 and 4100 K, conditions that are directly relevant to core segregation in a deep magma ocean. The sulfur partition coefficients measured from these experiments are an order of magnitude less than those obtained from extrapolation of previous results to core formation conditions (e.g., Rose-Weston et al., 2009; Boujibar et al., 2014). These measurements challenge the idea that sulfur becomes a highly siderophile element at high pressures and temperatures. A relationship was derived that describes sulfur's partitioning behavior at the pressure–temperature range of core formation. This relationship combined with an accretion model was used to explore the effects of varying impactor sizes and volatile compositions on the sulfur contents of the Earth's core and mantle. The results show that homogeneous delivery of sulfur throughout accretion would overenrich the mantle in sulfur relative to the present day observations of 200 ± 80 ppm (Lorand et al., 2013) unless the bulk Earth sulfur content is lower than its cosmochemical estimate of ∼6400 ppm (e.g., McDonough, 2003). On the other hand, the mantle's sulfur content is matched if sulfur is delivered with large bodies (3 to 10% Earth mass) during the last 20% of Earth's accretion, combined with a chondritic late veneer of 0.5% Earth mass. These results are conditional on the lowered equilibration efficiency of large impactor cores in a terrestrial magma ocean. In each accretion scenario, the core sulfur content remains below ∼2 wt.% in close agreement with cosmochemical estimates and is a further indication that sulfur is not a dominant light element in the core. •Metal–silicate partitioning of sulfur obtained at core formation conditions.•NanoSIMS used for analysis of diamond anvil cell experiments.•Sulfur is only a minor component of the terrestrial core.•Sulfur and other volatiles were likely delivered during late stage of accretion and core formation with large impactors.