Giants are bullies: How their growth influences systems of inner sub-Neptunes and super-Earths

Abstract

Observational evidence points to an unexpected correlation between outer giant planets and inner sub-Neptunes, which has remained unexplained by simulations so far. We utilize N-body simulations including pebble and gas accretion as well as planetary migration to investigate how the gas accretion rates, which depend on the envelope opacity and the core mass, influence the formation of systems of inner sub-Neptunes and outer gas giants as well as the eccentricity distribution of the outer giant planets. We find that less efficient envelope contraction rates allow for a more efficient formation of systems with inner sub-Neptunes and outer gas giants. This is caused by the fact that the cores that formed in the inner disk are too small to effectively accrete large envelopes and only cores growing in the outer disk, where the cores are more massive due to the larger pebble isolation mass, can become giants. As a result, instabilities between the outer giant planets do not necessarily destroy the inner systems of sub-Neptunes unlike simulations with more efficient envelope contraction where giant planets can form closer in. Our simulations show that up to 50% of the systems of cold Jupiters could have inner sub-Neptunes, in agreement with observations. At the same time, our simulations show a good agreement with the eccentricity distribution of giant exoplanets, even though we find a slight mismatch to the mass and semi-major axes’ distributions. Synthetic transit observations of the inner systems (r < 0.7 AU) that formed in our simulations reveal an excellent match to the Kepler observations, where our simulations can especially match the period ratios of adjacent planet pairs. As a consequence, the breaking the chains model for super-Earth and sub-Neptune formation remains consistent with observations even when outer giant planets are present. However, simulations with outer giant planets produce more systems with mostly only one inner planet and with larger eccentricities, in contrast to simulations without outer giants. We thus predict that systems with truly single close-in planets are more likely to host outer gas giants. We consequently suggest radial velocity follow-up observations of systems of close-in transiting sub-Neptunes to understand if these inner sub-Neptunes are truly alone in the inner systems or not.