Spectroscopic evolution of very massive stars at Z = 1/2.5 Z⊙

Context. Stars with masses in excess of 100 M_⊙ are observed in the Local Universe, but they remain rare objects. Because of the shape of the mass function, they are expected to be present only in the most massive and youngest clusters. They may thus be formed in number in highly star-forming galaxies.
Aims: Very massive stars (VMSs) experience strong stellar winds that are stronger than those of their less massive OB-type counterparts. These strong winds therefore need to be taken into account in evolutionary models and synthetic spectra to properly predict the appearance of VMS.
Methods: We present evolutionary models computed with the code STAREVOL. They include a recent mass-loss recipe that is relevant for VMSs. We subsequently calculated atmosphere models and synthetic spectra along the resulting tracks with the code CMFGEN. We studied stars with masses between 150 and 400 M_⊙ and focused on a metallicity Z = 1/2.5 Z_⊙. We studied the impact of our VMS spectra on the spectral energy distribution of young starbursts.
Results: We show that the optical and UV range is dominated by He II 4686 and He II 1640 emission for almost the entire main-sequence evolution of VMSs, in contrast to less massive stars. In the UV spectral range, carbon, nitrogen, and iron lines shape the spectra of VMSs, which appear for most of their evolution as WNh objects. The morphology of the synthetic spectra is similar to that of VMSs in the Large Magellanic Cloud. We show that stars with masses higher than 100 M_⊙ emit nearly as much light as all other stars in young starbursts. The integrated UV spectrum of these starbursts is significantly affected by the presence of VMSs.
Conclusions: We conclude that a consistent treatment of the evolution and the atmospheres of VMSs is mandatory to properly study spatially unresolved regions of intense star formation. We make our synthetic spectra and spectral energy distributions available to the community.