Astron. Astrophys. 348, L45-L48 (1999)

4. Discussion

The primary observational features of the ROSAT and EINSTEIN broadband measurements are (a) no correlation between X-ray luminosity and star or wind parameters and (b) evidence for WN stars being 3-4 times more luminous in X-rays than WC stars. Assuming an optically thin hot gas of small filling factor embedded in a dense "cool" Wolf-Rayet wind, we have derived a simple scaling relation for the luminosity of X-ray emission. Given that the WR winds are optically thick with at most X-ray energies, our scaling results qualitatively explain feature (a) if varies from star to star as . Note that for thin winds with constant or for winds with a function of radius, will generally have a dependence on the ratio . In fact, Owocki & Cohen (1999) appeal to a radially varying filling factor to explain the X-ray emission of O stars (however, note that they do not allow for in their analysis).

As regards feature (b), our scaling results predict that from WR winds should depend on relative abundances and ionization. Using typical parameters for the WN and WC classes, we derived an upper limit for the ratio of to that is factors of 4-5 greater than observed, but enhancement factors of a few for (WC) would bring the prediction in line with observations. An important factor leading to the result is the strong influence of metals on the wind attenuation, with for WC stars being times greater than for WN stars. Although of higher emissivity, the larger values for WC stars and consequent lower emission measures result in lower X-ray luminosities than for WN stars. Better knowledge of the hot gas temperature is necessary to determine , thereby allowing a more rigorous test of our scaling results.

It is clear that a drastic improvement of data quality for WR X-ray measurements is desperately needed. There are many interesting questions on the wind driving and structure of WR winds that could be addressed with good S/N X-ray spectral data, especially the influence of multiple scattering on the formation and evolution of wind shocks. Also, since the line emission spectrum and wind absorption is dominated by metals, the X-ray band is especially apt for studying these highly enriched winds, through resolving individual line features and K-shell edges. With the better spectral response and greater collecting area of the latest X-ray telescopes, a much better data set for addressing these issues and advancing our understanding of the WR phenomenon should be forthcoming in the near future.

© European Southern Observatory (ESO) 1999

Online publication: July 26, 1999
helpdesk.link@springer.de