The Atmosphere-Interior Connection: Rocky Planets as Linked Chemical Systems

Abstract

The goal of near-term exoplanet observations is to identify and characterize the atmospheres of habitable rocky planets. Solar System studies have taught us that rocky planets undergo differentiation into three different reservoirs: the atmosphere and ocean, the mantle, and the core. The chemistry of the protoplanetary disk from which they formed, the physics and chemistry of the accretion process, and subsequent loss and exchange between the three reservoirs will all affect the present-day, measurable atmospheric compositions of rocky exoplanets. Understanding these reservoirs and the processes that shaped them is essential to help predict and understand the atmospheres of the new types of rocky planets being discovered. In this thesis I explore each of these reservoirs and some of their interactions through a range of modeling techniques. In the first study, I model the deep water cycle on Earth- and super-Earth-sized planets through the use of parameterized convection models originally developed to study the Earth. I incorporate the effect of pressure and water abundance on the viscosity of the mantle and investigate how these parameters alter the evolution of oceans on the surface of a rocky planet. I find that ocean mass and persistence depends on the mantle convection style and that planets larger than the Earth may have delayed onset of ocean formation. In the second study, I model atmosphere-interior exchange for a planet hot enough to experience a global magma ocean due to greenhouse warming by a thick steam atmosphere. This model is applied to the exoplanet GJ 1132 b to make predictions for its atmospheric composition in a simple two-component (H2O and O2) model. Hydrogen is lost from the upper atmosphere, taking some oxygen with it. A small amount of oxygen, at most 10 %, is absorbed by the magma ocean by oxidizing Fe2+ to Fe3+ in the melt. The most common outcome of the models across a wide range of initial water and Fe abundances is that the planet has a very tenuous O2 atmosphere. Thick atmospheres of steam and O2 are possible only if the planet has very large initial water abundances. I also model the process of core formation on planets larger than the Earth using metal-silicate fractionation. Core formation can remove major and minor elements from a planet's mantle and controls the internal structure of a planet. Core formation can also alter the mantle's oxidation state, which controls the composition of dissolved volatiles and erupted volcanic gases and therefore has implications for atmospheric composition. I utilize experimental data from high-pressure shock experiments done at the Sandia National Laboratory to calculate Si and Ni partition coefficients up to 276 GPa. I apply this model to planets up to 10 Earth masses in size, and find that above a critical pressure, little to no light elements may enter the planet's core. This is in contrast to the Earth, which is known to have up to 10 wt % of lighter elements in the core. In the final study of this thesis, I determine the oxygen fugacities and gas compositions for primitive materials from the solar nebula and their mixtures. Exoplanet host stars have been shown to be approximately solar in their major elemental ratios, so although we cannot measure the elemental compositions of their solid materials, we can use the primitive materials of our own Solar System to constrain the range of compositions and constituents that could exist. These calculations can be used in future models of atmospheric formation and evolution as well as in models of volatile transport and trace element partitioning on planetesimals and meteorite parent bodies.