Detailed NLTE calculations have become the tool of choice for the determination of realistic chemical abundances in stars which exhibit marked departures from thermodynamic equilibrium in their atmospheres. In particular, we refer to the hot luminous OB stars in which lines belonging to ionized species are prevalent.
B stars on the main sequence are the progenitors of yellow supergiants, and in particular, classical Cepheids. While evolving off the main sequence, they change the characteristics of their photospheres, and hence, the visual characteristics of their spectra. Therefore, the comparison of abundance analyses of the main sequence B stars and their descendants is complicated by the necessity of analyzing spectral features formed under very different physical conditions. One has to be sure that in both cases the applied methods give reliable results for the elemental abundances, and only then can one be confident that any changes detected in the abundances are real.
This work deals with NLTE calculations for carbon and nitrogen - two elements whose surface abundance is closely related to core nucleosynthetic and dynamical processes which take place during the course of stellar evolution.
There exist a number of problems in the spectroscopy of main sequence B stars that cannot be resolved without accurate NLTE elemental abundance determinations. Among them:
1) Is there evidence of incomplete CNO cycle processed material on the surface of B main sequence stars?
2) If so, does the degree of contamination depend upon rotational velocity?
3) What is the relation between C and N abundances in the progenitor B stars and their descendants - Cepheids and non-variable yellow supergiants?
Some efforts have been made to solve these problems (at least, the first two of them), but here we shall not extensively discuss the work based on fundamentally LTE analyses of carbon-nitrogen abundances in B stars. For example, Gies & Lambert (1992), Cunha & Lambert (1994), Kilian (1992) determined LTE abundances of these elements and attempted to correct for NLTE effects. The method used involves indirect estimates of NLTE corrections based upon corresponding relations between equivalent widths of lines and elemental abundances for different temperature and gravity values calculated with Gold (1984) atmosphere models (Becker & Butler, 1989 and Eber & Butler, 1988). Korotin et al. (1999ab-Paper Iab) performed direct NLTE calculations for lines of C II and N II in the spectrum of Peg based on Kurucz (1992) models. They noted a difference in the NLTE corrections obtained from Kurucz models versus the less blanketed Gold models.
© European Southern Observatory (ESO) 1999
Online publication: October 4, 1999