3D simulations of AGB stellar winds -- II. Ray-tracer implementation and impact of radiation on the outflow morphology
Le résumé fourni par la source
Stars with an initial mass below ~ 8 Msun evolve through the asymptotic giant branch (AGB) phase, during which they develop strong stellar winds. Recent observations have revealed significant morphological complexities in their outflows, most likely caused by a companion. We study the impact of the radiation force on such companion-perturbed AGB outflows. We present the implementation of a ray tracer for radiative transfer in smoothed particle hydrodynamics (SPH) and compared four different descriptions of radiative transfer: the free-wind, the geometrical, the Lucy, and the attenuation approximation. For both low and high mass-loss rates, the velocity profile of the outflow is modified when going from the free-wind to the geometrical approximation, also resulting in a different morphology. In the case of a low mass-loss rate, the effect of the Lucy and attenuation approximation is negligible due to the low densities but morphological differences appear in the high mass-loss rate regime. By comparing the radiative equilibrium temperature and radiation force to full 3D radiative transfer, we show that the Lucy approximation works best. Although, close to the companion, artificial heating occurs and it fails to simulate the shadow cast by the companion. The attenuation approximation produces a lower equilibrium temperature and weaker radiation force, but it produces the shadow cast by the companion. From the predictions of the 3D radiative transfer, we also conclude that a radially directed radiation force is a reasonable assumption. The radiation force thus plays a critical role in dust-driven AGB winds, impacting the velocity profile and morphological structures. For low mass-loss rates, the geometrical approximation suffices, while high mass-loss rates require a more rigorous method, where the Lucy approximation provides the most accurate results although not accounting for all effects.
Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.