Formation of the Asymmetric Accretion Disk from Stellar Wind Accretion in an S-type Symbiotic Star

TL;DR

This study uses 3D hydrodynamical simulations to explore how accretion disks form around white dwarfs in S-type symbiotic stars, revealing asymmetric, flared disks with spiral features and sub-Keplerian flow, challenging previous assumptions.

Contribution

It demonstrates that accretion disks always form in these systems regardless of wind speed and uncovers their asymmetric, spiral structure and flow dynamics through detailed simulations.

Findings

Accretion disks form around white dwarfs regardless of stellar wind velocity.

Disk size and accretion rate vary with wind speed, exceeding BHL predictions in slower winds.

The disk exhibits asymmetry, spiral features, and sub-Keplerian flow speeds.

Abstract

The accretion process in a typical S-type symbiotic star, targeting~AG Draconis, is investigated through 3D hydrodynamical simulations using the FLASH code. Regardless of the wind velocity of the giant star, an accretion disk surrounding the white dwarf is always formed. In the wind models faster than the orbital velocity of the white dwarf, the disk size and accretion rate are consistent with the predictions under the Bondi-Hoyle-Lyttleton (BHL) condition. In slower wind models, unlike the BHL predictions, the disk size does not grow and the accretion rate increases to a considerably higher level, up to $>20\%$ of the mass-loss rate of the giant star. The accretion disk in our fiducial model is characterized by a flared disk with a radius of 0.16~au and a scale height of 0.03 au. The disk mass of $\sim 5 \times 10^{-8} M_\odot$ is asymmetrically distributed with the density peak toward the giant star, being about $50\%$ higher than the density minimum in the disk. Two inflowing spiral features are clearly identified and their relevance to the azimuthal asymmetry of disk is pointed out. The flow in the accretion disk is found to be sub-Keplerian with about $90\%$ of the Keplerian speed, which indicates the caveat of overestimating the O VI emission region from spectroscopy of Raman-scattered O VI features at 6825 {\AA} and 7082 {\AA}.