Rob Spaargaren
With more observations of terrestrial exoplanets becoming available, the importance of geodynamical studies focusing on exoplanets is increasing. We know from observations that stellar chemical abundances vary in the Solar neighbourhood, and this is likely to result in terrestrial exoplanets with a chemical diversity. Bulk planet composition affects many properties of the interior directly (core size, mantle viscosity) or indirectly (thermal evolution, layering). This may extend to atmospheric properties, since terrestrial planet atmospheres form and evolve under continuous interaction with the interior. In order to better understand the variability of interior properties among terrestrial exoplanets, we here attempt to constrain the range of bulk compositions of terrestrial exoplanets in the Solar neighbourhood. We approximate exoplanet compositions by adjusting host-star compositions from a stellar catalogue according to the condensation temperature, which leads to a depletion of volatile elements (devolatilization). We consider core-mantle differentiation by distributing oxygen among elements according to their tendency to stabilize oxides over metals, and partitioning metals in the core and oxides in the mantle. We include partitioning of light elements into the core. Thus, we obtain a proxy for both bulk silicate and core compositions. Using this methodology, we identify a small number of end-member bulk planet compositions, which we recommend for use in modelling of terrestrial exoplanet interiors. We also present mineralogical mantle profiles based on these end-member compositions. Finally, we explore the effect of this variability in bulk composition on long-term evolution of the planetary interior using a 2D parametrized convection model.
We convert stellar to planet compositions, by considering the compositional trend between the Sun and the Earth. This trend, called the devolatilization trend, shows a growing depletion of elements in Earth (compared to the Sun) with increasing volatility, or how quickly that element can evaporate during planet formation. This is a general process, so we can apply this to exoplanets. This trend can change for exoplanets, if their formation history is very different from Earth, but for the most common elements on Earth, the differences should be small. We consider O, Si, Mg, Fe, Al, Ca, Ni, Na, K, and S.
After applying the devolatilization trend to stellar compositions, we obtain compositions of Earth-like exoplanets. However, one additional step is required, as these planets are split into two major compositional reservoirs: the metallic, iron-rich core, and the rocky, silicate part, which consists of the mantle and crust. For simplicity, we combine the mantle and crust here, referring to both as the mantle. We consider cores similar to that of Earth, that consist mostly of Fe and Ni, with 2.5 wt% O and 6 wt% Si, alongside all S present in the planet. We find that most exoplanets have core sizes between 20 and 35 wt% of the planet. This means that the Earth core, at 32.5 wt%, is fairly large.
Subtracting core compositions from the bulk composition, leads to the mantle compositions of these planets. We present compositions for two important quantities: the mantle iron content, and the mantle Mg/Si-ratio. The mantle iron content controls melting behaviour of the mantle, and is mostly between 5 and 10 wt% for exoplanets (Earth has 6.32 wt% Fe in the mantle). The higher the mantle iron content, the easier the mantle melts. The mantle Mg/Si-ratio is a quantity that affects how easy the mantle deforms, and is able to transport hot material from the hot core-mantle boundary to the cold surface. The higher the Mg/Si-ratio, the easier it is for the mantle to deform, and transport material to the surface. Planets with high Mg/Si-ratios therefore cool down faster. We find molar Mg/Si-ratios mostly between 1 and 1.7 (Earth has Mg/Si of ~1.2).
The compositions we find here, can now be used in models of planetary interiors, to study the effects of composition on planetary evolution. Combining these compositions with data from experiments on different rock types and compositions allows us to study how the mantle behaves as a function of its composition. Eventually, we would like to couple these effects to the transport of water and CO2 between the mantle and the surface, and therefore whether interior composition can affect atmospheric evolution.