Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 329, 613-623 (1998)

Previous Section Next Section Title Page Table of Contents

2. Microturbulence in early-type stars

It was recognised very early on in the analysis of the spectra of early-type stars, that some additional broadening mechanism, other than thermal or intrinsic, was required to explain the observed widths and strengths of metal lines. This became evident through comparing the relative strengths of lines within a multiplet in dwarfs and supergiants and, using curve of growth arguments, Struve & Elvey (1934) first postulated that this broadening and strengthening of lines might be caused by small-scale turbulent velocities. Subsequently, LTE analyses confirmed these findings, with typical microturbulent velocities of order 5 kms-1 being found for the main sequence B-type stars (Hardorp & Scholz 1970) while even higher values, of order 20 - 30 kms-1, were obtained for supergiants (Lamers 1972, Dufton 1972, Lennon & Dufton 1986). The magnitude of these estimates decreased with the introduction of non-LTE techniques; for example Mihalas (1972) showed that the observed strength of the Mg II 4481Å doublet in early B-type main sequence stars could be reproduced with zero microturbulence. For supergiants however, values close to the speed of sound were still obtained. Lennon et al. (1991) adopted [FORMULA]  = 10 kms-1 in their non-LTE analysis of 3 supergiants, noting that even this value was insufficient to remove the slope in their abundance-equivalent width plots. More recently, Smartt et al. (1997) analysed 4 Galactic supergiants (using methods similar to ours) and found that [FORMULA]  = 30 kms-1 was appropriate for both LTE and non-LTE. Gies & Lambert (1992) analysed a sample of 39 B-type stars (of which 3 were supergiants) in both LTE and non-LTE. They found that the assumption of LTE led to very large microturbulences (typically 25 to 30 kms-1) for their supergiants, whereas a non-LTE analysis yielded smaller values (typically 10 kms-1). Significantly, there are very few microturbulence estimates for O-type stars from photospheric metal lines due to the lack of suitable lines; see for example Peterson & Scholz (1971).

The fact that supergiants have significant winds led naturally to the suggestion that microturbulence might, to a large extent, be the result of an outflow (Kudritzki 1992, Lamers & Achmad 1994). However, it is interesting to note that recent detailed non-LTE investigations of main sequence B-type stars, with mass-loss rates much lower than supergiants, imply that microturbulent velocities as high as 10 kms-1 are applicable (Gies & Lambert 1992, Kilian et al. 1991). Further, it is found that microturbulence is necessary to fit even stellar wind line-profiles in OB-type stars (Groenewegen & Lamers 1989, Haser et al. 1995, McCarthy et al. 1997), although this has been linked to the presence of shocks in the wind. The aim of this paper is to investigate how the inclusion of microturbulence, in the classical sense, might modify the helium abundance results discussed above. It is motivated in part by our involvement in a systematic spectroscopic analysis of a sample of B-type supergiants in our own Galaxy and in the SMC (McErlean et al. 1997, Lennon 1997). Here we present atmospheric parameters ([FORMULA], [FORMULA] and helium fraction, y) for two representative and well studied Galactic objects, [FORMULA] and [FORMULA] Ori, as well as the results of numerical tests performed to investigate the effect of microturbulent velocities on the line profiles of hydrogen and neutral and ionised helium.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1998

Online publication: December 8, 1997