In situ formation of hot Jupiters with companion super-Earths

Observations have confirmed the existence of multiple-planet systems containing a hot Jupiter and smaller planetary companions. Examples include WASP-47, Kepler-730, and TOI-1130. We examine the plausibility of forming such systems in situ using N-body simulations that include a realistic treatment of collisions, an evolving protoplanetary disc, and eccentricity/inclination damping of planetary embryos. Initial conditions are constructed using two different models for the core of the giant planet: a 'seed-model' and an 'equal-mass-model'. The former has a more massive protoplanet placed among multiple small embryos in a compact configuration. The latter consists only of equal-mass embryos. Simulations of the seed-model lead to the formation of systems containing a hot Jupiter and super-Earths. The evolution consistently follows four distinct phases: early giant impacts; runaway gas accretion on to the seed protoplanet; disc damping-dominated evolution of the embryos orbiting exterior to the giant; a late chaotic phase after dispersal of the gas disc. Approximately 1 per cent of the equal-mass simulations form a giant and follow the same four-phase evolution. Synthetic transit observations of the equal-mass simulations provide an occurrence rate of 0.26 per cent for systems containing a hot Jupiter and an inner super-Earth, similar to the 0.2 per cent occurrence rate from actual transit surveys, but simulated hot Jupiters are rarely detected as single transiting planets, in disagreement with observations. A subset of our simulations form two close-in giants, similar to the WASP-148 system. The scenario explored here provides a viable pathway for forming systems with unusual architectures, but does not apply to the majority of hot Jupiters.

Automated summary

Observations have shown the existence of systems with hot Jupiters and smaller planetary companions. N-body simulations were used to examine the formation of such systems, with two different models leading to the formation of systems with hot Jupiters and super-Earths. Approximately 1% of simulations formed giant planets, and synthetic transit observations provided an occurrence rate similar to actual surveys.