Core-powered mass-loss and the radius distribution of small exoplanets

Abstract

Recent observations identify a valley in the radius distribution of small exoplanets, with planets in the range 1.5−2.0 R ⊕ significantly less common than somewhat smaller or larger planets. This valley may suggest a bimodal population of rocky planets that are either engulfed by massive gas envelopes that significantly enlarge their radius, or do not have detectable atmospheres at all. One explanation of such a bimodal distribution is atmospheric erosion by high-energy stellar photons. We investigate an alternative mechanism: the luminosity of the cooling rocky core, which can completely erode light envelopes while preserving heavy ones, produces a deficit of intermediate sized planets. We evolve planetary populations that are derived from observations using a simple analytical prescription, accounting self-consistently for envelope accretion, cooling and mass loss, and demonstrate that core-powered mass loss naturally reproduces the observed radius distribution, regardless of the high-energy incident flux. Observations of planets around different stellar types may distinguish between photoevaporation, which is powered by the high-energy tail of the stellar radiation, and core-powered mass loss, which depends on the bolometric flux through the planet's equilibrium temperature that sets both its cooling and mass-loss rates.

Occurrence Rates of Planets Orbiting FGK Stars: Combining Kepler DR25, Gaia DR2, and Bayesian Inference

“…While photoevaporation (e.g. Owen & Wu 2017; Lopez & Rice 2018) is often cited as a potential mechanism for the planet radius valley, other physical mechanisms have been proposed including core-powered mass loss (Ginzburg et al 2018;Gupta & Schlichting 2019a,b), impact erosion via planetesimals (Shuvalov 2009;Schlichting et al 2015;Wyatt et al 2020), and the formation of two distant exoplanet populations due to a gas-poor environment for the rocky planets , which provides an occurrence rate normalized to the width of the bin and therefore is not dependent on choice of grid density. The uncertainties shown are the differences between the median and the 15.87th or 84.13th percentile.…”

Signatures of the core-powered mass-loss mechanism in the exoplanet population: dependence on stellar properties and observational predictions

“…Recently, Ginzburg et al (2018) and Gupta & Schlichting (2019) showed that the core-powered mass-loss mechanism is also able to reproduce the observed radius valley, even in the absence of photoevaporation. In this mechanism, a planet's internal luminosity drives the loss of its atmosphere (Ginzburg et al 2016(Ginzburg et al , 2018Gupta & Schlichting 2019). The source of this luminosity is a planet's primordial energy from formation, which can be of the order of, or even larger than, its atmosphere's gravitational binding energy.…”

“…In this work, we extended previous work on the corepowered mass-loss mechanism (Ginzburg et al 2018;Gupta & Schlichting 2019), and investigated how stellar mass, metallicity and age impact the resulting planet size distribution. We model our host star population on the stars in the CKS survey and investigate, collectively and in isolation, the impact of stellar mass, metallicity and age on the resulting exoplanet size distribution.…”

Testing exoplanet evaporation with multitransiting systems

“…There is a degeneracy between the derived core-composition and photoevaporative massloss rates, with lower mass-loss rates favouring lower core densities (Wu 2019;Owen & Adams 2019). Furthermore, alternative hypothesis have been suggested for the origin of the observed gap in the radius distribution, including corepowered mass-loss (Ginzburg et al 2018;Gupta & Schlichting 2019a,b), or two distinct formation pathways for the two sub-groupings. In the latter scenario, the two sub-groupings are water-worlds and rocky, terrestrial planets (Zeng et al 2019).…”