Formation of the Janus-Epimetheus system through collisions

Abstract

Context. Co-orbital systems are bodies that share the same mean orbit. They can be divided into different families according to the relative mass of the co-orbital partners and the particularities of their movement. Janus and Epimetheus are unique in that they are the only known co-orbital pair of comparable masses and thus the only known system in mutual horseshoe orbit. Aims. We aim to establish whether the Janus-Epimetheus system might have formed by disruption of an object in the current orbit of Epimetheus. Methods. We assumed that four large main fragments were formed and neglected smaller fragments. We used numerical integration of the full N-body problem to study the evolution of different fragment arrangements. Collisions were assumed to result in perfectly inelastic merging of bodies. We statistically analysed the outcome of these simulations to infer whether co-orbital systems might have formed from the chosen initial conditions. Results. Depending on the range of initial conditions, up to 9% of the simulations evolve into co-orbital systems. Initial velocities around the escape velocity of Janus yield the highest formation probability. Analysis of the evolution shows that all co-orbital systems are produced via secondary collisions. The velocity of these collisions needs to be low enough that the fragments can merge and not be destroyed. Generally, collisions are found to be faster than an approximate cut-off velocity threshold. However, given a sufficiently low initial velocity, up to 15% of collisions is expected to result in merging. Hence, the results of this study show that the considered formation scenario is viable.