next up previous
Next: About this document Up: Spectroscopic techniques Previous: White Dwarfs and

Wolf-Rayet stars

Obtaining a reliable temperature scale for Wolf-Rayet (WR) stars, the chemically evolved descendents of OB stars or evolved H-deficient CSPNe, is a formidable challenge. Continuum techniques were used to estimate in WR stars (e.g. Underhill 1983) though were shown to be unrealistic by Garmany et al. (1984). Due to extreme mass-loss, their winds are so optically thick that their photospheres are invisible, with the usual plane-parallel, LTE apporximations completely invalid, with radii highly wavelength-dependent (Cherepachchuk et al. 1984). As for O stars, much progress has been made in recent years using extended non-LTE model atmospheres (e.g. Hillier 1987; Schmutz et al. 1989). Unfortunately, an added complication of their dense winds is that the usual radius at =2/3 becomes a poor indicator of excitation in WR stars, with =20 usually defining the stellar temperature, though this is difficult to relate with interior evolutionary models (Hamann 1994). Unfortunately, stellar temperatures of WR stars are also critically dependent on assumed velocity distributions (Hillier 1991), so that observational determinations of WR velocity distributions are keenly sought.

Initially, ionization balance studies restricted to lines of helium from model grids where used to obtain temperatures (e.g. Schmutz et al. 1989). These have subsequently been extended to tailored models including nitrogen in WN stars (e.g. Crowther et al. 1995a) and carbon in WC stars (e.g. Koesterke & Hamann 1995), with heavy element line blanketed models now starting to appear. In general, results from pure helium analyses tend to underestimate stellar temperatures, with the discrepancy greater at earlier spectral type. For HD151932 (WN7), Underhill (1983) obtained a temperature of 25kK from the integrated flux method, increasing to 32kK from non-LTE model grids by Schmutz et al. (1989) and 34kK from line profile fits by Crowther et al. (1995a). However, results obtained for different species in non-blanketed models generally lead to discrepant temperatures: For HD211564 (WN3), Crowther et al. (1995b) obtained =49kK (69kK) from helium (nitrogen) diagnostics! Observations of WR nebulae provide excellent tests of current theoretical flux distributions below the Lyman limit (see Esteban et al. 1993).

To summarise, huge advances have been made over the past decade towards a reliable temperature scale of hot stars. However, significant uncertainties remain, especially for O and WR stars, which may be resolved using nebulae as tests of theoretical energy distributions. Since real stars rotate and pulsate it is unclear what scale of temperature fluctuations are occuring across their stellar disks.

References



next up previous
Next: About this document Up: Spectroscopic techniques Previous: White Dwarfs and



Paul Crowther
Wed Sep 23 13:41:03 BST 1998