Abstract
We address the turbulent-fragmentation scenario for the origin of the stellar
initial mass function (IMF), using a large set of numerical simulations of
randomly driven supersonic MHD turbulence. While radiation and outflows from
protostars are often invoked, the turbulent-fragmentation model successfully
predicts the main features of the observed stellar IMF assuming an isothermal
equation of state without any stellar feedback. As a test of the model, we
focus on the case of a magnetized isothermal gas, neglecting stellar feedback,
while pursuing a large dynamic range in both space and time scales in order to
generate a large number of stars, covering the full spectrum of stellar masses
from brown dwarfs to massive stars. Our simulations represent a generic 4 pc
region within a typical Galactic molecular cloud (MC), with a mass of 3,000
Msun, an rms velocity ten times the isothermal sound speed and five times the
average Alfven velocity, in agreement with MC observations. We achieve a
maximum resolution of 50 AU and a maximum duration of star formation of 4.0
Myr, forming up to a thousand sink particles whose mass distribution closely
match the observed stellar IMF. A large set of medium-size simulations is used
to test the sink particle scheme, while larger simulations are used to test the
numerical convergence of the IMF and the dependence of the IMF turnover on
physical parameters predicted by the turbulent-fragmentation model. We find
clear evidence of numerical convergence and strong support for the model
predictions, including the initial time evolution of the IMF. We conclude that
the physics of isothermal MHD turbulence is sufficient to explain the origin of
the IMF.
Users
Please
log in to take part in the discussion (add own reviews or comments).