Building on results from the Magnetism in Massive Stars (MiMeS) project, this paper shows how a two-parameter classification of massive-star magnetospheres in terms of the magnetic wind confinement (which sets the Alfvén radius R A ) and stellar rotation (which sets the Kepler co-rotation radius R K ) provides a useful organisation of both observational signatures and theoretical predictions. We compile the first comprehensive study of inferred and observed values for relevant stellar and magnetic parameters of 64 confirmed magnetic OB stars with T eff 16 kK. Using these parameters, we locate the stars in the magnetic confinement-rotation diagram, a log-log plot of R K vs. R A . This diagram can be subdivided into regimes of centrifugal magnetospheres (CM), with R A > R K , vs. dynamical magnetospheres (DM), with R K > R A . We show how key observational diagnostics, like the presence and characteristics of Hα emission, depend on a star's position within the diagram, as well as other parameters, especially the expected wind mass-loss rates. In particular, we identify two distinct populations of magnetic stars with Hα emission: namely, slowly rotating O-type stars with narrow emission consistent with a DM, and more rapidly rotating B-type stars with broader emission associated with a CM. For O-type stars, the high mass-loss rates are sufficient to accumulate enough material for line emission even within the relatively short free-fall timescale associated with a DM: this high mass-loss rate also leads to a rapid magnetic spindown of the stellar rotation. For the B-type stars, the longer confinement of a CM is required to accumulate sufficient emitting material from their relatively weak winds, which also lead to much longer spindown timescales. Finally, we discuss how other observational diagnostics, e.g. variability of UV wind lines or X-ray emission, relate to the inferred magnetic properties of these stars, and summarise prospects for future developments in our understanding of massive-star magnetospheres.

Context. The evolution of massive stars is still partly unconstrained. Mass, metallicity, mass loss, and rotation are the main drivers of stellar evolution. Binarity and the magnetic field may also significantly affect the fate of massive stars. Aims. Our goal is to investigate the evolution of single O stars in the Galaxy. Methods. For that, we used a sample of 74 objects comprising all luminosity classes and spectral types from O4 to O9.7. We relied on optical spectroscopy obtained in the context of the MiMeS survey of massive stars. We performed spectral modelling with the code CMFGEN. We determined the surface properties of the sample stars, with special emphasis on abundances of carbon, nitrogen, and oxygen. Results. Most of our sample stars have initial masses in the range of 20 to 50 M . We show that nitrogen is more enriched and carbon and oxygen are more depleted in supergiants than in dwarfs, with giants showing intermediate degrees of mixing. CNO abundances are observed in the range of values predicted by nucleosynthesis through the CNO cycle. More massive stars, within a given luminosity class, appear to be more chemically enriched than lower mass stars. We compare our results with predictions of three types of evolutionary models and show that for two sets of models, 80% of our sample can be explained by stellar evolution including rotation. The effect of magnetism on surface abundances is unconstrained. Conclusions. Our study indicates that in the 20−50 M mass range, the surface chemical abundances of most single O stars in the Galaxy are fairly well accounted for by stellar evolution of rotating stars.

We present the analysis performed on spectropolarimetric data of 97 O-type targets included in the framework of the MiMeS (Magnetism in Massive Stars) Survey. Mean Least-Squares Deconvolved Stokes I and V line profiles were extracted for each observation, from which we measured the radial velocity, rotational and nonrotational broadening velocities, and longitudinal magnetic field B ℓ . The investigation of the Stokes I profiles led to the discovery of 2 new multi-line spectroscopic systems (HD 46106, HD 204827) and confirmed the presence of a suspected companion in HD 37041. We present a modified strategy of the Least-Squares Deconvolution technique aimed at optimising the detection of magnetic signatures while minimising the detection of spurious signatures in Stokes V . Using this analysis, we confirm the detection of a magnetic field in 6 targets previously reported as magnetic by the MiMeS collaboration (HD 108, HD 47129A2, HD 57682, HD 148937, CPD-28 2561, and NGC 1624-2), as well as report the presence of signal in Stokes V in 3 new magnetic candidates (HD 36486, HD 162978, HD 199579). Overall, we find a magnetic incidence rate of 7 ± 3%, for 108 individual O stars (including all O-type components part of multi-line systems), with a median uncertainty of the B ℓ measurements of about 50 G. An inspection of the data reveals no obvious biases affecting the incidence rate or the preference for detecting magnetic signatures in the magnetic stars. Similar to A-and B-type stars, we find no link between the stars' physical properties (e.g. T eff , mass, age) and the presence of a magnetic field. However, the Of?p stars represent a distinct class of magnetic O-type stars.

Magnetic confinement of stellar winds leads to the formation of magnetospheres, which can be sculpted into Centrifugal Magnetospheres (CMs) by rotational support of the corotating plasma. The conditions required for the CMs of magnetic early B-type stars to yield detectable emission in Hα -the principal diagnostic of these structures -are poorly constrained. A key reason is that no detailed study of the magnetic and rotational evolution of this population has yet been performed. Using newly determined rotational periods, modern magnetic measurements, and atmospheric parameters determined via spectroscopic modelling, we have derived fundamental parameters, dipolar oblique rotator models, and magnetospheric parameters for 56 early B-type stars. Comparison to magnetic A-and O-type stars shows that the range of surface magnetic field strength is essentially constant with stellar mass, but that the unsigned surface magnetic flux increases with mass. Both the surface magnetic dipole strength and the total magnetic flux decrease with stellar age, with the rate of flux decay apparently increasing with stellar mass. We find tentative evidence that multipolar magnetic fields may decay more rapidly than dipoles. Rotational periods increase with stellar age, as expected for a magnetic braking scenario. Without exception, all stars with Hα emission originating in a CM are 1) rapid rotators, 2) strongly magnetic, and 3) young, with the latter property consistent with the observation that magnetic fields and rotation both decrease over time.

The groundbreaking detection of gravitational waves produced by the inspiralling and coalescence of the black hole (BH) binary GW150914 confirms the existence of "heavy" stellar-mass BHs with masses > 25 M . Initial modelling of the system by Abbott et al. (2016a) supposes that the formation of black holes with such large masses from the evolution of single massive stars is only feasible if the wind mass-loss rates of the progenitors were greatly reduced relative to the mass-loss rates of massive stars in the Galaxy, concluding that heavy BHs must form in low-metallicity (Z ∼ < 0.25 − 0.5 Z ) environments. However, strong surface magnetic fields also provide a powerful mechanism for modifying mass loss and rotation of massive stars, independent of environmental metallicity (ud-Doula & Owocki 2002;ud-Doula et al. 2008). In this paper we explore the hypothesis that some heavy BHs, with masses > 25 M such as those inferred to compose GW150914, could be the natural end-point of evolution of magnetic massive stars in a solar-metallicity environment. Using the MESA code, we developed a new grid of single, non-rotating, solar metallicity evolutionary models for initial ZAMS masses from 40-80 M that include, for the first time, the quenching of the mass loss due to a realistic dipolar surface magnetic field. The new models predict TAMS masses that are significantly greater than those from equivalent non-magnetic models, reducing the total mass lost by a strongly magnetized 80 M star during its main sequence evolution by 20 M . This corresponds approximately to the mass loss reduction expected from an environment with metallicity Z = 1/30 Z .