2017MNRAS.465.2654W

In order to constrain the evolutionary state of the red supergiant Betelgeuse (α Orionis), we have produced a suite of models with zero-age main sequence masses from 15 to 25 M in intervals of 1 M including the effects of rotation. The models were computed with the stellar evolutionary code MESA. For non-rotating models, we find results that are similar to other work. It is somewhat difficult to find models that agree within 1σ of the observed values of R, T_eff and L, but modestly easy within 3σ uncertainty. Incorporating the nominal observed rotational velocity, ∼15 km/s, yields significantly different and challenging constraints. This velocity constraint is only matched when the models first approach the base of the red supergiant branch (RSB), having crossed the Hertzsprung gap, but not yet having ascended the RSB and most violate even generous error bars on R, T_eff and L. Models at the tip of the RSB typically rotate at only ∼0.1 km/s, independent of any reasonable choice of initial rotation. We discuss the possible uncertainties in our modelling and the observations, including the distance to Betelgeuse, the rotation velocity and model parameters. We summarize various options to account for the rotational velocity and suggest that one possibility is that Betelgeuse merged with a companion star of about 1 M as it ascended the RSB, in the process producing the ring structure observed at about 7 arcmin away. A past coalescence would complicate attempts to understand the evolutionary history and future of Betelgeuse.