Properties of Free Floating Planets Ejected through Planet-Planet Scattering

TL;DR

This study uses N-body simulations to analyze how planet-planet scattering leads to the ejection of planets, revealing ejection rates, velocities, and dependencies on initial system properties, contributing to understanding free floating planet populations.

Contribution

It provides new insights into the properties and ejection mechanisms of planets through scattering, with detailed simulation results matching observed free floating planet data.

Findings

40-80% of planets are ejected depending on initial system

Ejected planets have velocities of 2-6 km/sec

Bound planets tend to be more massive than ejected ones

Abstract

Multiple studies have shown that planet-planet scattering plays an important role in the dynamical evolution of planetary systems. For instance, it has been shown that planet-planet scattering can reproduce the eccentricity distribution of exoplanets. It can also contribute to the current census of free floating planets. In this work we run an ensemble of N-body simulations of planetary systems, and record the properties of planets which are ejected from the system. In our simulations we sample a wide range of orbital and physical properties of the planets. We find that in general $40-80\%$ of the planets are ejected from the system depending on the number of planets initially in the system. Most of the planets are ejected over a timescale of $\sim 10^8-10^9$ years. The ejected planets have a mean excess velocity in the range of 2-6 km/sec with respect to the host star. The excess velocities of the planets ejected from the system strongly depends on the semi-major axis of the inner most planet. We find that irrespective of their initial location in the planetary system, all planets are equally likely to be ejected from the system. Also, bound and ejected planets have distinct mass distributions, with bound planets being more massive than ejected planets. In addition, increasing the radii of the planets reduces the ejection fraction. The properties of the ejected planets do not strongly depend on the initial spacing between the planets. The timescale over which ejections happen does increase with the initial separation between the planets. We also find that the ejection fraction does not strongly depend on the distance from the host star beyond which the planets are considered unbound. Finally, we compared our results with observed populations of free floating planets. We conclude that on average 5-10 planets should form around each star to reproduce the observations.