Evolution of surface CNO abundances in massive stars

The nitrogen to carbon (N/C) and nitrogen to oxygen (N/O) ratios are the most sensitive quantities to mixing in stellar interiors of intermediate and massive stars. We further investigate the theoretical properties of these ratios as well as put in context recent observational results obtained by the VLT-FLAMES Survey of massive stars in the Galaxy and the Magellanic Clouds. We consider analytical relations and numerical models of stellar evolution as well as our own stellar atmosphere models, and we critically re-investigate observed spectra. On the theoretical side, the N/C vs N/O plot shows little dependence on the initial stellar masses, rotation velocities, and nature of the mixing processes up to relative enrichment of N/O by a factor of about four, thus this plot constitutes an ideal quality test for observational results. The comparison between the FLAMES Survey and theoretical values shows overall agreement, despite the observational scatter of the published results. The existence of some mixing of CNO products is clearly confirmed, however the accuracy of the data is not sufficient for allowing a test of the significant differences between different models of rotating stars and the Geneva models. We discuss reasons (for the most part due to observational bias) why part of the observational data points should not be considered for this comparison. When these observational data points are not considered, the scatter is reduced. Finally, the N/C vs N/O plot potentially offers a powerful way for discriminating blue supergiants before the red supergiant stage from those after it. Also, red supergiants of similar low velocities may exhibit different N enrichments, depending on their initial rotation during the main-sequence phase.