Two bi-stability jumps in theoretical wind models for massive stars and the implications for Luminous Blue Variable supernovae

TL;DR

This study explores two bi-stability jumps in stellar wind mass-loss rates of massive stars, revealing their occurrence at lower temperatures than previously thought and their significant impact near the Eddington limit, influencing supernova progenitors.

Contribution

It confirms the existence of two bi-stability jumps in mass-loss rates at lower Teff and highlights their dramatic increase near the Eddington limit, improving understanding of stellar evolution.

Findings

Bi-stability jumps occur at lower Teff than previously predicted.

Mass-loss rates can increase by up to a factor of 30 near the Eddington limit.

Results help reconcile discrepancies between empirical and theoretical mass-loss models.

Abstract

Luminous Blue Variables have been suggested to be the direct progenitors of supernova types IIb and IIn, with enhanced mass loss prior to explosion. However, the mechanism of this mass loss is not yet known. Here, we investigate the qualitative behaviour of theoretical stellar wind mass-loss as a function of Teff across two bi-stability jumps in blue supergiant regime and also in proximity to the Eddington limit, relevant for LBVs. To investigate the physical ingredients that play a role in the radiative acceleration we calculate blue supergiant wind models with the CMFGEN non-LTE model atmosphere code over an effective temperature range between 30000 and 8800 K. Although our aim is not to provide new mass-loss rates for BA supergiants, we study and confirm the existence of two bi-stability jumps in mass-loss rates predicted by Vink, de Koter, & Lamers (1999). However, they are found to occur at somewhat lower Teff (20000 and 9000 K, respectively) than found previously, which would imply that stars may evolve towards lower Teff before strong mass-loss is induced by the bi-stability jumps. When the combined effects of the second bi-stability jump and the proximity to Eddington limit are accounted for, we find a dramatic increase in the mass-loss rate by up to a factor of 30. Further investigation of both bi-stability jumps is expected to lead to a better understanding of discrepancies between empirical modelling and theoretical mass-loss rates reported in the literature, and to provide key inputs for the evolution of both normal AB supergiants and LBVs, as well as their subsequent supernova type II explosions.