Massive stars on the main sequence

Abstract

Massive stars have a strong influence on their environment by chemical enrichment of the surrounding interstellar medium due to strong stellar winds. The influence of mass loss and rotation on the main-sequence evolution of massive stars is studied in this thesis with the focus on stars more massive than 100 Msun.<br /> A method is presented to determine age and inclination angle of stars based on their surface nitrogen enhancement (published in Köhler et al. (2012)). The derived equations are based on stellar evolution models of Brott et al. (2011a) for rotating single stars. The method is applied to early B-type main-sequence stars (LMC: NGC 2004 and N 11; SMC: NGC 330 and NGC 346). Age constraines are derived for 17 stars. For 10 of these 17 stars, the method lines out that they can not be explained by rotational mixing in single stars. For stars, were only upper limits of the surface nitrogen abundance could be derived, the method can be used to deduce limits to their inclination angles.<br /> Detailed stellar evolution models of rotating very massive single stars are presented with initial masses of 70–500 Msun and surface rotational velocities up to 550 km/s, assuming LMC metallicity (published in Köhler et al. (2015)). Very massive stars are strongly affected by mass loss. Until the end of the main-sequence phase, they lose more than 50% of their initial mass. For stars initially more massive than 160 Msun, mass loss becomes the main trigger to chemical homogeneous evolution, even for slow rotating stars. Additionally, the strong increase in mass loss at 25 000K leads to the truncation of the red-ward evolution for stars initially more massive than 80 Msun. All presented models reach the Eddington limit in the stellar envelope during hydrogen burning, what leads to envelope inflation.<br /> Synthetic populations of stars are presented based on the presented stellar evolution models and models published in Brott et al. (2011a), using the population synthesis code Starmaker (Brott et al. 2011b). The populations are calculated for two different initial distributions. First, the distributions are based on the observational data of the VLT-FLAMES Tarantula Survey (main-sequence stars; spectral type: O7–O2) assuming star bursts of one, two, three and five Myr old stars. Second, flat distributions in initial rotation rates and initial masses are applied, assuming continuous star formation. Based on the simulation results, the probability to observe evolutionary features related to mass loss and rotational mixing in very massive stars are derived (for example rapidly rotating stars in the Hertzprung-Russell diagram). First comparisons to a sub-sample of the VLT-FLAMES Tarantula Survey are presented (main-sequence stars of spectral type O7–O2).