Inspecting Acoustic Energy as a Candidate Source of Mass Loss in Red Giant Branch Stars

It has long been known that stars experience mass loss while evolving along the red giant branch (RGB). Nonetheless, the theoretical understanding of the mechanisms driving mass loss is still very poor. The most used prescription for the mass loss rate along the RGB is the Reimers' law (Reimers 1975), which lacks a solid physical foundation. A large set of theories can be grouped under the "family" of thermal winds, where mass loss consists in evaporation of coronal gas, heated by dissipation of the energy associated to magnetohydrodynamic or acoustic waves. The latter are generated by convective motions within the stellar envelope and propagate above the photosphere, where they are damped. Fusi-Pecci and Renzini (1975) (FPR, hereafter) derived an equation where the mass loss rate relates to the total rate of acoustic energy emitted by the envelope. The physical assumptions at the basis of their approach have various limitations, primarily the use of the mixing length theory (MLT). The scope of this project is reviewing the role of acoustic energy in the FPR mass loss prescription, on the basis of the acoustic energy extracted from the 3D atmosphere models of the STAGGER grid (Magic et al. 2013). In particular, the total rate of acoustic energy is corrected on the basis of the information extracted from 3D simulations. Both the standard FPR prescription and the updated one (FPR3D, hereafter) are implemented in the 1D stellar evolutionary code Module for Experiments in Stellar Astrophysics (MESA). A strong concentration of mass loss is found close to the RGB tip, both for the standard FPR and FPR3D models. The dependence of the total mass loss on metallicity is instead very different: for [Fe/H] ≳ −0.5, standard FPR models show a decreasing trend, whereas the FPR3D models are characterized by a monotonically increasing trend. These results are compared with the predictions of Reimers mass loss and recent observational mass-loss estimates.