Hydrogen is one of the possible alloying elements in the Earth’s core, but its siderophile (iron-loving) nature is debated. Here we experimentally examined the partitioning of hydrogen between molten iron and silicate melt at 30–60 gigapascals and 3100–4600 kelvin. We find that hydrogen has a metal/silicate partition coefficient DH ≥ 29 and is therefore strongly siderophile at conditions of core formation. Unless water was delivered only in the final stage of accretion, core formation scenarios suggest that 0.3–0.6 wt% H was incorporated into the core, leaving a relatively small residual H2O concentration in silicates. This amount of H explains 30–60% of the density deficit and sound velocity excess of the outer core relative to pure iron. Our results also suggest that hydrogen may be an important constituent in the metallic cores of any terrestrial planet or moon having a mass in excess of ~10% of the Earth.

Melting experiments were performed on the Fe‐C‐H system to 127 GPa in a laser‐heated diamond anvil cell. On the basis of in situ and ex situ sample characterizations, we found that the solubility of carbon in liquid Fe correlates inversely with hydrogen concentration at ~60 GPa and ~3500 K, indicating that liquid Fe preferentially incorporates hydrogen rather than carbon under conditions with abundant C and H. While large amounts of both C and H may have been delivered to the growing Earth, C‐poor/H‐rich metals were likely added to the protocore in the late stages of core formation. We also obtained a melting curve of FeHx (x > 1) far beyond the pressure range in earlier determinations. Its liquidus temperature was found to be 2380 K at 135 GPa, lower than those of Fe alloyed with the other possible core light elements. Relatively low core temperature is thus supported by the presence of hydrogen.

We examined the compression behavior of hexagonal‐close‐packed (hcp) (Fe0.88Si0.12)1H0.61 and (Fe0.88Si0.12)1H0.79 (in atomic ratio) alloys up to 138 GPa in a diamond anvil cell (DAC). While contradicting experimental results were previously reported on the compression curve of double‐hcp (dhcp) FeHx (x ≈ 1), our data show that the compressibility of hcp Fe0.88Si0.12Hx alloys is very similar to those of hcp Fe and Fe0.88Si0.12, indicating that the incorporation of hydrogen into iron does not change its compression behavior remarkably. The present experiments suggest that the inner core may contain up to 0.47 wt % hydrogen (FeH0.26) if temperature is 5000 K. The calculated density profile of Fe0.88Si0.12H0.17 alloy containing 0.32 wt % hydrogen in addition to geochemically required 6.5 wt % silicon matches the seismological observations of the outer core, supporting that hydrogen is an important core light element.

We examined the crystal structure of FeHX (X~1) (FeH hereafter) at high pressure and temperature by X-ray diffraction up to 137 GPa. Results show that FeH adopts a face-centered cubic (fcc) structure at pressures of 43 to 137 GPa and temperatures of ~1000 to 2000 K. Our study revises a phase diagram of stoichiometric FeH in which fcc has a wider-than-expected stability field at high pressure and temperature. Based on our findings, the FeH end-member of the Fe-FeH system is expected to be stable in the fcc structure at the P-T conditions of the Earth's core, rather than in the double-hexagonal close packed (dhcp) structure as previously reported. We compared the experimentally determined unit-cell volumes of FeH with those from ab initio calculations. Additionally, we observed a change in compressibility at ~60 GPa, which could be attributed to a magnetic transition—an interpretation supported by our ab initio computations.

Melting experiments on the Fe‐S‐O ternary system were performed to 208 GPa in a laser‐heated diamond‐anvil cell. Compositions of liquids and coexisting solids in recovered samples were examined using a field‐emission‐type electron microprobe. The results demonstrate that the ternary eutectic point shifts toward the oxygen‐rich, sulfur‐poor side with increasing pressure, in accordance with changes in eutectic liquid compositions in the Fe‐O and Fe‐S binary systems. We also found that solid Fe crystallizing from liquid Fe‐S‐O does not include oxygen, while the partitioning of sulfur into solid Fe is enhanced with increasing pressure. These indicate that oxygen‐rich, sulfur‐poor liquid crystallizes Fe at the inner core boundary; however, it makes a large density difference between the liquid and solid core, which is inconsistent with observations. Alternatively, we found that a range of C‐bearing, S‐poor/O‐rich liquids account for the density and velocity in the outer core and the density in the inner core.