Disc fragmentation. I. Ejection of Jupiter-mass Free Floating Planets from growing binary systems

TL;DR

This work demonstrates that disc fragmentation in very young binary systems can efficiently eject Jupiter-mass free-floating planets (JFFPs) through chaotic migration and gravitational slingshot interactions with a growing secondary. Using 3D SPH simulations of massive, self-gravitating discs with embedded $1-3\,M_{\rm J}$ planets and a secondary seed ($5-50\,M_{\rm J}$), the authors show ejection fractions rising with secondary mass, typically yielding low ejection velocities ($v_{\rm f}\sim 2$ km s$^{-1}$) and possible retention of compact circum-planetary discs. They contrast this disc-fragmentation channel with core-accretion pathways, noting earlier formation, higher ejection velocities, and distinctive disc signatures for FFPs produced by fragmentation. The results imply that disc fragmentation could account for a substantial fraction of the observed JFFP population, especially in binaries, and provide observable predictions (disc presence around JFFPs, velocity distributions, and age constraints) to distinguish formation routes in young clusters.

Abstract

Over the past 25 years, observations have uncovered a large population of free-floating planets (FFPs), whose origins remain debated. Massive FFPs (several Jupiter masses or more) may form via gravitational collapse of molecular clouds, similar to stars. Lower-mass FFPs likely originate in planetary systems and are later ejected through dynamical interactions. We show that disc fragmentation in very young stellar binaries may be an abundant source of Jupiter-like FFPs (JFFPs), with masses above 0.3 Jupiter masses. In our model, disc fragmentation at tens to 100 au from the primary star produces gas giants, while fragmentation further out forms a more massive object that will eventually evolve into the secondary star. We present 3D simulations of massive, self-gravitating discs with embedded Jupiter-mass planets and a secondary seed ($5-50 M_J$). Chaotic migration leads to frequent planet-secondary interactions, imparting velocity kicks via gravitational slingshot that usually end in planet ejection. The ejection fraction increases steeply with the secondary-to-primary mass ratio, $q_s$, reaching 0.6 for $q_s > 0.05$. Compared to Core Accretion JFFPs, disc fragmentation JFFPs: (i) form earlier, and may be more abundant in young clusters; (ii) are ejected at much lower velocities; (iii) may retain compact circum-planetary discs. To reproduce the observed abundance of JFFPs, disc fragmentation planets must be in the post-collapse configuration. They must also either form more frequently in binary systems than around single stars, or be frequently disrupted in inner disc regions, as previously suggested in certain versions of the theory.

Figures (21)

• Figure 1: The first snapshot of a simulation when the planets are added.
• Figure 2: Same simulation as in Fig. \ref{['fig:initial_conditions']} but 1,400 years later, the massive object has a mass of 50 ${\,{\rm M}_{\rm J}}$ in this simulation.
• Figure 3: Migration tracks for a number of different secondary masses. The simulations all contain only one planet starting at 140 au, which is allowed to migrate for a few thousand years.
• Figure 4: The radial positions with respect to the star for a subset of two planets and the secondary object in a simulation with $\beta$ = 30, $M_2$ = 50 ${\,{\rm M}_{\rm J}}$. The secondary has initial mass of 50 ${\,{\rm M}_{\rm J}}$ and is labeled "9", while planet 7 has an initial mass of 1 ${\,{\rm M}_{\rm J}}$ and planet 5 starts with 3 ${\,{\rm M}_{\rm J}}$ (cf. Table \ref{['tab:inital']}). The vertical dashed line marks the time of ejection for planet 7. Purely by chance, planet 5 early interactions with the secondary send it inwards rather than outwards (cf. Fig. \ref{['fig:close_encounter_59']} below). It then migrates rapidly inwards, avoiding close interactions with the secondary and hence surviving as a result.
• Figure 5: The radial positions of the secondary (planet 9; dark blue, $M_2$ = 50 ${\,{\rm M}_{\rm J}}$) and 5 (green, $M_{\rm p}$ = 3 ${\,{\rm M}_{\rm J}}$). An initial close encounter leads to subsequent encounters that end in planet 5 migrating inwards and escaping ejection.