Evolution of stars with 60 and 200 M_⊙: predictions for WNh stars in the Milky Way

¹ Nicolaus Copernicus Astronomical Center, Polish Academy of Sciences, Bartycka 18, 00-716 Warsaw, Poland
² Astronomický ústav, Akademie věd Ceské republiky, Fričova 298, 251 65 Ondřejov, Czech Republic
³ Department of Astronomy and Astrophysics, University of California San Diego, La Jolla, California 92093, USA

^⋆ Corresponding author; alex.gormaz@asu.cas.cz

Received: 18 July 2024
Accepted: 3 March 2025

Abstract

Context. Massive stars are characterised by powerful stellar winds driven by radiation; thus, the mass-loss rate is known to play a crucial role in their evolution.

Aims. We study the evolution of two massive stars (a classical massive star and a very massive star) at solar metallicity (Z = 0.014) in detail. We calculate their final masses, radial expansion, and chemical enrichment, at their H-core, He-core, and C-core burning stages, prior to their final collapse.

Methods. We ran evolutionary models for initial masses of 60 and 200 M_⊙ using MESA and the Geneva-evolution-code (GENEC). For the mass loss, we adopted the self-consistent m-CAK prescription for the optically thin winds of OB-type stars, a semi-empirical formula for H-rich optically thick wind of luminous Wolf-Rayet (WR) stars of the nitrogen sequence with hydrogen in their spectra (WNh stars), and a hydrodynamically consistent formula for the H-poor thick wind of classical WR stars. The transition from thin to thick winds was set to Γ_e = 0.5.

Results. The unification of the initial set-up for the stellar structure and wind prescription leads to very similar black hole mass for both GENEC and MESA codes, but both codes predict different tracks across the Hertzsprung-Russell diagram (HRD) For the 60 M_⊙ case, the GENEC model predicts a more efficient rotational mixing and more chemically homogeneous evolution, whereas the MESA model predicts a large radial expansion that reaches the Luminous Blue Variable (LBV) phase. For the 200 M_⊙ case, differences between both evolution codes are less relevant because their evolution is dominated by wind mass loss with a weaker dependence on internal mixing.

Conclusions. The switch of the mass-loss prescription based on the Eddington factor instead of the removal of outer layers, implies the existence of WNh stars with a large mass fraction of hydrogen at the surface (X_surf ≥ 0.3) formed from initial masses of ≳60 M_⊙. These stars are constrained in a T_eff range of the HRD which corresponds to the main sequence band, in agreement with the observations of Galactic WNh stars at Z = 0.014. While our models employ a fixed Γ_e, trans threshold for the switch to thick winds, rather than a continuous thin-to-thick wind model, the good reproduction of observations during the main sequence supports the robustness of the wind model upgrades, allowing its application to studies of late-stage stellar evolution before core collapse.