Evolution of blue supergiants and α Cygni variables: Puzzling CNO surface abundances

A massive star can enter the blue supergiant region either by evolving directly from the main sequence, or by evolving from a previous red supergiant stage. The fractions of the blue supergiants with different histories depend on the internal mixing and mass loss during the red supergiant stage. We study the possibility of using diagnostics based on stellar pulsation to discriminate blue supergiants with different evolution histories. For this purpose, we have studied the pulsation property of massive star models calculated with the Geneva stellar evolution code, for initial masses ranging from 8 to 50M, with a solar metallicity of Z = 0.014. We have found that radial pulsations are excited in the blue supergiant region only in models that had been red supergiants previously. This might provide us with a useful means of diagnosing the history of evolution of each blue supergiant. At a given effective temperature, many more non-radial pulsations are excited in the model after the red supergiant stage than in the model evolving towards the red supergiant. We discuss the properties of radial and non-radial pulsations in blue supergiants, and we compare predicted periods with the period ranges observed in some a Cygni variables in the Galaxy and NGC 300.We have found that blue supergiant models after the red supergiant stage roughly agree with observed period ranges, in most cases. However, we are left with the puzzle that the predicted surface N/C and N/O ratios seem to be too high compared with those of Deneb and Rigel.