From CO₂- to H₂O-dominated atmospheres and back

How mixed outgassing changes the volatile distribution in magma oceans around M dwarf stars

¹ Space Research Institute, Austrian Academy of Sciences, Schmiedlstrasse 6, 8042 Graz, Austria
² Centre for Exoplanet Science, University of St Andrews, North Haugh, St Andrews, KY16 9SS, UK
³ Max-Planck-Institut für Astronomie, Königstuhl 17, Heidelberg 69117, Germany
⁴ Department of Astronomy, University of Washington, Seattle, WA 98105, USA
⁵ Institute of Geological Sciences, Freie Universität Berlin, Malteserstr. 74-100, 12249 Berlin, Germany
⁶ University of Bristol, School of Physics, Tyndall Avenue, Bristol, BS8 1TL, UK
⁷ School of Earth & Environmental Sciences, University of St Andrews, Bute Building, Queen’s Terrace, St Andrews, KY16 9TS, UK
⁸ SUPA, School of Physics & Astronomy, University of St Andrews, North Haugh, St Andrews, KY16 9SS, UK
⁹ Stuttgart Center for Simulation Science, University of Stuttgart, Pfaffenwaldring 5a, 70569 Stuttgart, Germany
¹⁰ Department of Physics, University College Cork, Cork, T12 R229, Ireland
¹¹ Fakultät für Mathematik, Physik und Geodäsie, TU Graz, Petersgasse 16, Graz 8010, Austria

^★ Corresponding author; ludmila.carone@oeaw.ac.at

Received: 9 April 2024
Accepted: 9 December 2024

Abstract

Aims. We investigate the impact of CO₂ on the distribution of water on TRAPPIST-1 e, f, and g during the magma ocean stage. These potentially habitable rocky planets are currently the most accessible for astronomical observations. A constraint on the volatile budget during the magma ocean stage is a key link to planet formation and also to judging their habitability.

Methods. We expanded the MagmOc module of the VPLanet environment to perform simulations with 1-100 terrestrial oceans (TOs) of H₂O with and without CO₂ and for albedos 0 and 0.75. The CO₂ mass was scaled with initial H₂O by a constant factor between 0.1 and 1.

Results. The magma ocean state of rocky planets begins with a CO₂-dominated atmosphere but can evolve into a H₂O dominated state, depending on initial conditions. For less than 10 TO initial H₂O, the atmosphere tends to desiccate and the evolution can end with a CO₂ dominated atmosphere. Otherwise, the final state is a thick (>1000 bar) H₂O-CO₂ atmosphere. Complete atmosphere desiccation with less than 10 TO initial H₂O can be significantly delayed for TRAPPIST-1 e and f, when H₂O has to diffuse through a CO₂ atmosphere to reach the upper atmosphere, where photolysis due to extreme ultra violet irradiation occurs. As a consequence of CO₂ diffusion-limited water loss, the time of mantle solidification for TRAPPIST-1 e, f, and g can be significantly extended compared to a pure H₂O evolution by up to 40 Myrs for an albedo of 0.75 and by up to 200 Mys for an albedo of 0. The addition of CO₂ further results in a higher water content in the melt during the magma ocean stage. Thus, more water can be sequestered in the solid mantle. However, only up to 6% of the initial water mass can be stored in the mantle at the end of the magma ocean stage. Our compositional model adjusted for the measured metallicity of TRAPPIST-1 yields for the dry inner planets (b, c, d) an iron fraction of 27 wt%. For TRAPPIST-1 e, this iron fraction would be compatible with a (partially) desiccated evolution scenario and a CO₂ atmosphere with surface pressures of a few 100 bar.

Conclusions. A comparative study between TRAPPIST-1 e and the inner planets may yield the most insights about formation and evolution scenarios by confronting, respectively, a scenario with a desiccated evolution due to volatile-poor formation and a volatile-rich scenario with extended atmospheric erosion.