Towards a population synthesis model of self-gravitating disc fragmentation and tidal downsizing II: the effect of fragment–fragment interactions

Abstract

It is likely that most protostellar systems undergo a brief phase where the protostellar disc is self-gravitating. If these discs are prone to fragmentation, then they are able to rapidly form objects that are initially of several Jupiter masses and larger. The fate of these disc fragments (and the fate of planetary bodies formed afterwards via core accretion) depends sensitively not only on the fragment's interaction with the disc, but with its neighbouring fragments.We return to and revise our population synthesis model of self-gravitating disc fragmentation and tidal downsizing. Amongst other improvements, the model now directly incorporates fragment-fragment interactions while the disc is still present. We find that fragment-fragment scattering dominates the orbital evolution, even when we enforce rapid migration and inefficient gap formation.Compared to our previous model, we see a small increase in the number of terrestrial-type objects being formed, although their survival under tidal evolution is at best unclear. We also see evidence for disrupted fragments with evolved grain populations -this is circumstantial evidence for the formation of planetesimal belts, a phenomenon not seen in runs where fragment-fragment interactions are ignored.In spite of intense dynamical evolution, our population is dominated by massive giant planets and brown dwarfs at large semimajor axis, which direct imaging surveys should, but only rarely, detect. Finally, disc fragmentation is shown to be an efficient manufacturer of free floating planetary mass objects, and the typical multiplicity of systems formed via gravitational instability will be low.

Isochronal age-mass discrepancy of young stars: SCExAO/CHARIS integral field spectroscopy of the HIP 79124 triple system

“…As HIP 79124 should be coeval with the three objects forming in timescales < 1 Myr (Bate 2012), this result strongly points towards the fact that the models do not reproduce well enough the PMS phase of low-mass stars. -Adopting the age of the primary star for the entire system, we find a mass of B of ∼ 100 M Jup , and ∼ 330 M Jup for C. Given their masses and small orbital separation, there is the possibility that these objects formed via disk instability (e.g., Forgan et al 2018). -This effect can alter the mass of the directly-imaged companions to low-mass stars, if the age of the system is derived from isochronal fits to the photometric data of the host star.…”

Giant planets and brown dwarfs on wide orbits: a code comparison project

“…as used by Bate et al (2003). We note that Forgan et al (2018) have recently presented an updated population synthesis model. We do not include this study in our code comparison here because its migration module is similar to the Müller et al (2018) treatment, which is discussed below.…”

“…We do not include this study in our code comparison here because its migration module is similar to the Müller et al (2018) treatment, which is discussed below. Furthermore, Forgan et al (2018) also consider multiple gas clumps and model their N-body interactions. These effects can be very important in modifying the outcome of disc fragmentation (Hall et al 2017) but is beyond the scope of our one-clump study.…”

Fragmentation favoured in discs around higher mass stars

“…Models of the Jeans mass in spiral arms of self-gravitating discs predict that fragmentation will primarily form objects with initial masses greater than ∼3 Jupiter masses (Kratter et al 2010;Forgan & Rice 2011, 2013a. It is therefore likely that planet formation through the gravitational instability (hereafter GI) favours the for-mation of wide-orbit gas giants and brown dwarfs (Stamatellos & Whitworth 2009;Forgan & Rice 2013b;Vigan et al 2017;Hall et al 2017;Forgan et al 2018).…”