Highly siderophile elements were stripped from Earth’s mantle by iron sulfide segregation

Final results based on metal-silicate segregation, the Hadean matte and late accretion. The evolution of (A) sulfur and (B, C) HSE concentrations with time are shown with ks = 0.44 (Eq. 2). The accretion history is shown in (A-C). Results are shown for multi-stage FeS segregation (m-s) and late single-stage segregation (s-s) at 118 My. The vertical dashed lines in (A-C) show the time of the final giant impact (GI) at 113 My and the start of late veneer accretion (LV) at 119 My. Error bars, based on the propagation of uncertainties in the partitioning parameters, are shown in (B) and (C) for the final Pd and Ru concentrations; propagated uncertainties for Pt and Ir are ±0.3 and ±0.6 ppb respectively. (D) Final calculated HSE values, normalized to Ir and CI, compared with bulk silicate Earth values.

Final results based on metal-silicate segregation, the Hadean matte and late accretion. The evolution of (A) sulfur and (B, C) HSE concentrations with time are shown with ks = 0.44 (Eq. 2). The accretion history is shown in (A-C). Results are shown for multi-stage FeS segregation (m-s) and late single-stage segregation (s-s) at 118 My. The vertical dashed lines in (A-C) show the time of the final giant impact (GI) at 113 My and the start of late veneer accretion (LV) at 119 My. Error bars, based on the propagation of uncertainties in the partitioning parameters, are shown in (B) and (C) for the final Pd and Ru concentrations; propagated uncertainties for Pt and Ir are ±0.3 and ±0.6 ppb respectively. (D) Final calculated HSE values, normalized to Ir and CI, compared with bulk silicate Earth values.

Our paper describing how highly siderophile elements such as platinum, palladium, rubidium and iridium were removed from Earth's mantle and segregated into the core was published online and will be in the September 9, 2016 issue of Science. Here is a link to the publication in Science.

A one-paragraph summary:

It shows evidence that heavy elements called highly siderophile elements like platinum left the rocky part of Earth into the core as the rocky part of the Earth crystallized from molten magma into solid rock. This is very different than what most experts thought before. The previous hypothesis was that these elements sank to the core with iron during core formation itself, but when we tested this idea with laboratory and numerical experiments, it didn't work. Instead, we had to revive a poorly promoted hypothesis from 1991 called the Hadean Matte. During the Hadean Matte, as the rocks crystallize, there are too many sulfur atoms to fit in the silicate crystal lattices. This sulfur grabs iron atoms to make iron-sulfide and then sink to the core, but as they do so they also grab highly siderophile elements such as platinum, palladium, rubidium, iridium etc. This hypothesis passed all of our tests that the previous model failed.

More details:

We tested both hypotheses via a numerical model that simulates the growth and differentiation of Earth. During core formation events, we determined what quantity of HSEs would be segregated into the core according to partitioning models created from laboratory experiments done at high pressures and temperatures. Similarly, for the iron sulfide segregation model, we conducted appropriate experiments to determine how much HSE segregation occured during those events.