Modeling the Tidal Downsizing Planetary Formation Hypothesis

Tidal Downsizing Model. III. Planets from sub-Earths to Brown Dwarfs: structure and metallicity preferences

Authors:

Nayakshin et al

Abstract:

We present improved population synthesis calculations in the context of the Tidal Downsizing (TD) hypothesis for planet formation. Our models provide natural explanations and/or quantitative match to exoplanet observations in the following categories: (i) most abundant planets being super-Earths; (ii) cores more massive than ∼5−15M⊕ are enveloped by massive metal-rich atmospheres; (iii) the frequency of occurrence of close-in gas giant planets correlates strongly with metallicity of the host star; (iv) no such correlation is found for sub-Neptune planets; (v) presence of massive cores in giant planets; (vi) the composition of gas giant planets is over-abundant in metals compared to their host stars; (vii) this over-abundance decreases with planet's mass, as observed; (viii) a deep valley in the planet mass function between masses of ∼10−20M⊕ and ∼100M⊕. We provide a number of observational predictions distinguishing the model from Core Accretion: (a) composition of the massive cores is dominated by rocks not ices; (b) the core mass function is smooth with no minimum at ∼3M⊕ and a rollover (rather than rise) below ∼1M⊕; (c) gas giants beyond 10 AU are insensitive to the host star metallicity. Objects more massive than ∼10MJup do not correlate or even anti-correlate with metallicity of the host star, which is consistent with observations showing that brown dwarf/ low mass stellar companions do not correlate/anti-correlate with metallicity of the primary star. One mismatch of the model and exoplanet observations is in the ratio of directly imaged to close-in giant planets, which is a factor ∼10 too high. This however may well be a deficiency of the simple disc model we use. We conclude that TD model is a viable alternative to CA in explaining the observed population of exoplanets