Connecting the Observed Properties of Exoplanet Populations to Their Formation

Abstract

The wealth of exoplanetary data as displayed on the mass-semimajor axis and mass-radius distributions reveal a tremendous amount of information constraining our understanding of their formation. We link the variety of outcomes shown in exoplanet populations to the observed ranges of protoplanetary disk properties using the core accretion model of planet formation. For this purpose, we consider a population synthesis framework that samples disk properties' observationally-constrained distributions as inputs to thousands of planet formation models. Planet traps are a key feature of our approach in that they are barriers to rapid type-I migration, and are sites of early stages of the core accretion process in our models. We show that a low setting of forming planets' atmospheric envelope opacities $\kappa_{\rm{env}}\simeq10^{-3}$ cm$^2$ g$^{-1}$ is necessary to achieve a range of gas giants' orbital radii that agrees with the data. At this low setting of $\kappa_{\rm{env}}$, X-ray ionization and its related dead zone results in a clear separation between hot Jupiters and warm gas giants near 1 AU. When radial dust drift is included in our models, the rapid migration of solids into the ice line makes it a crucial trap for the formation of super Earths and warm gas giants. The ratio of the formation frequency of these two planet types has an interesting dependence on the initial disk radius $R_0$, with intermediate $R_0\simeq$50 AU producing the largest super Earth population, and both larger and smaller disk sizes forming more gas giants. When including disk chemistry, the range of disk radii over which planet formation in traps occurs leads to a wide range of solid compositions, from ice rich planets (up to 50\% ice by mass) to dry, Earth-like compositions. We show planet compositions and post-disk atmospheric photoevaporation to be two key factors affecting the mass-radius distribution of our populations.