Modelling surface mineral diversity of atmosphere-free rocky exoplanets for spectroscopic characterisation

Observations of several short-period rocky exoplanets (e.g., LHS 3844 b, TRAPPIST-1 b, GJ 367 b) suggest that they do no host substantial secondary atmospheres, which makes their surfaces directly accessible to spectral characterisation. Various minerals and rock types have potentially distinguishable surface reflectance spectra, allowing for observational characterisation of surface geology for such atmosphere-less exoplanets. While extensive surface spectra for Solar System lithologies are available, they may not capture the full range of surface diversity, as rocky exoplanets display a bulk compositional diversity far exceeding that seen in the Solar System. To address this gap, we explore potential surface mineralogies of volatile-free rocky exoplanets, with compositional diversity informed by stellar abundances.
 
We model magma compositions formed from bulk mantle melting in the NCFMASCr system with a Gibbs free energy minimization algorithm, Perple_X. Bulk mantle compositions are systematically varied in terms of relative abundances of Mg, Si, Ca, Al, and Na, informed by stellar abundances, while keeping Fe and Cr constant and equal to the Earth bulk mantle. We then use the same modelling set-up to derive crustal mineralogy for bulk crust compositions based on these magmas. 
 
Surface mineralogy primarily varies with the bulk mantle Mg/Si ratio: Si-rich mantles produce quartz- and plagioclase-dominated crusts, intermediate planets produce pyroxene- and plagioclase-dominated crusts, and Mg-rich planets produce crusts consisting of olivine, spinel, and nepheline. Increasing the abundances of Ca, Al, and Na mainly results in a widening of the clinopyroxene, spinel, and nepheline stability fields. The crusts of Mg-rich planets are experimentally under-explored, while we predict a significant fraction of all rocky exoplanets to form such crusts. Thus, additional surface reflectance spectra measurements are required to fully cover the diversity of potential rocky exoplanet surfaces and to enable accurate interpretation of future observations of their surface geology.

We further show with geodynamical simulations that the high-pressure density contrast between crustal and mantle rocks plays a first-order role in thermal and dynamical evolution of rocky exoplanet interiors. Planets with a greater density contrast tend to stabilize a layered mantle structure, where subducted crust accumulates at the bottom of the mantle, overlain by a cold, depleted, and typically ultramafic upper mantle. Calculating the density contrast between crust and mantle rocks for our sample of exoplanet compositions at a pressure of 140 GPa, we find that most Mg-rich planets form crusts that are significantly denser than the residual mantle, forming such a double-layered mantle structure. Meanwhile, the most Si-rich mantles produce granite-like crusts, which we predict to be too buoyant to subduct. Only planets with intermediate Mg/Si, which includes the solar system planets, have crustal buoyancy that allows for subduction and mixing of subducted crust with the mantle on geological timescales. Thus, constraining rocky exoplanet crust mineralogy and density is essential for understanding their long-term evolution and for interpreting spectroscopic observations of such planets, which is possible with JWST.