2024-01-05 Hydrides in Sub-Neptune Exoplanets

Last week I noticed a paper about the "Outcomes of Collisions between sub-Neptunes", noting that sub-Neptunes (not defined in that paper, but around 5~10×M_⊕ and < 3×R_⊕) have rather different collisional properties to terrestrial properties, due to their thick, gaseous atmospheres.

This paper ("Stability of Hydrides in Sub-Neptune Exoplanets with Thick Hydrogen-Rich Atmospheres") reveals another aspect of these planets, which have no analogues in the Solar system.

The question might be asked, why are these called "sub-Neptunes", not "sub-Uranuses"? After all, Uranus is less massive than Neptune (86.8 vs 102 ×10²⁴kg). But Neptune is smaller than Uranus (R_≆=49528 km vs R_≅=51118 km). There is clearly an interior-compression process going on (described as a "radius cliff"), and Neptune is well inside the effect, while Uranus may or may not be. (Another "radius cliff" is thought to occur in planets larger than Jupiter, going up into the "brown dwarf" star range, though the "turn-over" radius is somewhat uncertain.)

This paper describes another effect which probably does not happen in the Solar system. At least, not outside diamond-anvil high pressure machines on Earth. The bottom of the hot hydrogen-helium atmosphere can interact with silicate minerals in the planetary core, potentially reducing some of the silicates (and their component metal ions) to release metals. Iron (present as Fe²⁺ in fayalite (iron-olivine) can be directly reduced to Fe⁰_metal by interaction with the hydrogen. It is less clear if the magnesium in forsterite (magnesium olivine) can be reduced to magnesium metal, but it seems that a magnesium-iron hydride Mg₂FeH₆ is possible. Water (H₂O) is a minor byproduct, which might accumulate at the top of the rock/ metal core, or might diffuse up to the atmosphere to freeze out at a high level.

The work is experimentally difficult, because at these temperatures (3000 K +) and pressures, the hydrogen makes the diamond cells brittle, while also making it interfere with X-ray diffraction to identify the products before they decompose on decompression. Raman spectroscopy does support the presence of Mg-H bonding though.

The new "hydride" phases, and possibly water-derived compounds in the hdyrgen could lead to a more gradual radial change in properties in these planets, which further work on Neptune might detect, in the same way that Juno's recent close orbiting of Jupiter has suggested it has a "diffuse" core-atmosphere boundary.

Sub-Neptunes are an interesting type of planet - sufficient justification in themselves to go out into the galaxy.