Three-Dimensional Simulations of Kelvin-Helmholtz Instability in Settled Dust Layers in Protoplanetary Disks

TL;DR

This study uses 3D simulations to analyze Kelvin-Helmholtz instability in dust layers of protoplanetary disks, revealing how Coriolis force and shear influence stability and potential for planet formation.

Contribution

It provides new insights into the effects of Coriolis force and radial shear on dust layer stability, clarifying conditions for gravitational instability in protoplanetary disks.

Findings

Coriolis force increases instability at high wavenumber for thick layers.

Radial shear's effect depends on perturbation amplitude, with larger perturbations causing disruption.

Stable layers have Richardson numbers between 0.2 and 0.4.

Abstract

As dust settles in a protoplanetary disk, a vertical shear develops because the dust-rich gas in the midplane orbits at a rate closer to true Keplerian than the slower-moving dust-depleted gas above and below. A classical analysis (neglecting the Coriolis force and differential rotation) predicts that Kelvin-Helmholtz instability occurs when the Richardson number of the stratified shear flow is below roughly one-quarter. However, earlier numerical studies showed that the Coriolis force makes layers more unstable, whereas horizontal shear may stabilize the layers. Simulations with a 3D spectral code were used to investigate these opposing influences on the instability in order to resolve whether such layers can ever reach the dense enough conditions for the onset of gravitational instability. I confirm that the Coriolis force, in the absence of radial shear, does indeed make dust layers more unstable, however the instability sets in at high spatial wavenumber for thicker layers. When radial shear is introduced, the onset of instability depends on the amplitude of perturbations: small amplitude perturbations are sheared to high wavenumber where further growth is damped; whereas larger amplitude perturbations grow to magnitudes that disrupt the dust layer. However, this critical amplitude decreases sharply for thinner, more unstable layers. In 3D simulations of unstable layers, turbulence mixes the dust and gas, creating thicker, more stable layers. I find that layers with minimum Richardson numbers in the approximate range 0.2 -- 0.4 are stable in simulations with horizontal shear.