Turbulence-induced collision velocities and rates between different sized dust grains

Abstract

We study the collision rates and velocities for point-particles of different sizes in turbulent flows. We construct fits for the collision rates at specified velocities (effectively a collisional velocity probability distribution) for particle stopping time ratios up to four; already by that point the collisional partners are very poorly correlated and so the results should be robust for even larger stopping time ratios. Significantly, we find that while particles of very different masses have approximately Maxwellian collisional statistics, as the mass ratio shrinks the distribution changes dramatically. At small stopping time ratios, the collisional partners are highly correlated and we find a population of high number density (clustered), low relative-velocity particle pairs. Unlike in the case of identical stopping time collisional partners, this low relative-velocity clustered population is collisional, but the clustering is barely adequate to trigger bulk effects such as the streaming instability. We conclude our analysis by constructing a master fit to the collisional statistics as a function only of the stopping time ratio. Together with our previous work for identical stopping time particle pairs, this provides a recipe for including collisional velocity probability distributions in dust coagulation models for protoplanetary disks. We also include our recipe for determining particle collisional diagnostics from numerical simulations.