A New Model of Roche-lobe Overflow for Short-Period Gaseous Planets and Binary Stars

TL;DR

This paper introduces an improved model for Roche-lobe overflow in short-period gaseous planets and binary stars, accounting for extended atmospheres and arbitrary mass ratios, with implications for planetary mass loss and system stability.

Contribution

The authors develop a revised Roche-lobe overflow model that includes extended atmospheres and arbitrary mass ratios, enhancing the understanding of mass transfer in close-in planetary systems.

Findings

Overflow significant for hot Neptunes out to ~2 days period.

Hot Jupiters may only overflow inward of 0.5 days.

Some planets could lose Earth mass in a few million years.

Abstract

Some close-in gaseous exoplanets are nearly in Roche-lobe contact, and previous studies show tidal decay can drive hot Jupiters into contact during the main sequence of their host stars. Improving upon a previous model, we present a revised model for mass transfer in a semi-detached binary system that incorporates an extended atmosphere around the donor and allows for an arbitrary mass ratio. We apply this new formalism to hypothetical, confirmed, and candidate planetary systems to estimate mass loss rates and compare with models of evaporative mass loss. Overflow may be significant for hot Neptunes out to periods of $\sim$ 2 days, while for hot Jupiters, it may only be important inward of 0.5 days. We find that CoRoT-24 b may be losing mass at a rate of more than an Earth mass in a Gyr. The hot Jupiter WASP-12 b may lose an Earth mass in a Myr, while the putative planet orbiting a T-Tauri star PTFO8-8695 might shed its atmosphere in a few Myrs. We point out that the orbital expansion that can accompany mass transfer may be less effective than previously considered because the gas accreted by the host star removes some of the system's angular momentum from the orbit, but simple scaling arguments suggest that the Roche-lobe overflow might remain stable. Consequently, the recently discovered small planets in ultra-short-periods ($<$ 1 day) may not be the remnants of hot Jupiters/Neptunes. The new model presented here has been incorporated into Modules for Experiments in Stellar Astrophysics (MESA).