There is no magnetic braking catastrophe: low-mass star cluster and protostellar disc formation with non-ideal magnetohydrodynamics

Abstract

We present results from the first radiation non-ideal magnetohydrodynamics (MHD) simulations of low-mass star cluster formation that resolve the fragmentation process down to the opacity limit. We model 50 M turbulent clouds initially threaded by a uniform magnetic field with strengths of 3, 5 10 and 20 times the critical mass-to-magnetic flux ratio, and at each strength, we model both an ideal and non-ideal (including Ohmic resistivity, ambipolar diffusion and the Hall effect) MHD cloud. Turbulence and magnetic fields shape the large-scale structure of the cloud, and similar structures form regardless of whether ideal or non-ideal MHD is employed. At high densities (10 6 n H 10 11 cm −3 ), all models have a similar magnetic field strength versus density relation, suggesting that the field strength in dense cores is independent of the large-scale environment. Albeit with limited statistics, we find no evidence for the dependence of the initial mass function on the initial magnetic field strength, however, the star formation rate decreases for models with increasing initial field strengths; the exception is the strongest field case where collapse occurs primarily along field lines. Protostellar discs with radii 20 au form in all models, suggesting that disc formation is dependent on the gas turbulence rather than on magnetic field strength. We find no evidence for the magnetic braking catastrophe, and find that magnetic fields do not hinder the formation of protostellar discs.

Can magnetized turbulence set the mass scale of stars?

“…Federrath et al 2010). Overall the good agreement between the present study, Haugbølle et al (2018), , Wurster et al (2019), and arguably Padoan et al (2019) is encouraging, suggesting that these IMF predictions have some robustness to choice of MHD solver and numerical sink particle prescriptions.…”

“…To achieve satisfactory statistics at the high density end we stacked the distribution from 10 snapshots around the target SFE. Despite the different initial conditions all runs saturate to the same B ∝ ρ 1/2 line (corresponding to v A = 2c s ), similar to the results of Wurster et al (2019). The results depart from the power-law above ρ ∼ 3 × 10 −14 g cm −3 , corresponding to the maximum Jeans-resolved density ρ J for these simulations (Equation 10).…”

“…But the effect we see is weakly-dependent on magnetic field strength. Moreover, Wurster et al (2019) investigated the combined effects of non-ideal MHD terms upon the IMF predicted by simulations and found no systematic difference compared to ideal MHD. And even if we imagined the "most extreme non-ideal" limit, where non-ideal terms allowed for either efficient decoupling of magnetic fields from most of the gas (ambipolar diffusion) or efficient magnetic damping (resistivity), this would lead to results more like non-MHD simulations, which as discussed above fare even more poorly at predicting any IMF shape resembling that observed.…”

The properties of clusters, and the orientation of magnetic fields relative to filaments, in magnetohydrodynamic simulations of colliding clouds

“…The first is that the two clouds lie within a low density medium, which is one hundredth of the density of the clouds. This is the same approach as Wurster et al (2019), who modelled isolated clouds within a low density medium. The magnetic field permeates both the clouds and this surrounding medium, preventing the magnetic field becoming unstable at the edge of the cloud, and removing the need for more complex magnetic boundary conditions.…”

“…The extent of the low density medium is ±120 pc in the x dimension, and ±46 pc in the y and z dimensions. Secondly, following the setup of the initial conditions in Wurster et al (2019), the particles are initially allocated on a grid in both the clouds and the low density surrounding medium, rather than randomly.…”