other querymodes : Identifierquery Coordinatequery Criteriaquery Referencequery Basicquery Scriptsubmission TAP Outputoptions Help

2003ApJ...586..996M - Astrophys. J., 586, 996-1018 (2003/April-1)

Constraints the ionization balance of hot-star winds from FUSE observations of O stars in the Large Magellanic Cloud.

MASSA D., FULLERTON A.W., SONNEBORN G. and HUTCHINGS J.B.

Abstract (from CDS):

We present Far Ultraviolet Spectroscopic Explorer (FUSE) spectra for 25 O stars in the Large Magellanic Cloud. We analyze wind profiles for the resonance lines from C III, N III, S IV, P V, S VI, and O VI in the FUSE range using a Sobolev with exact integration'' method. In addition, the available data from either IUE or the Hubble Space Telescope for the resonance lines of Si IV, C IV, and N V are also modeled. Because several of the FUSE wind lines are unsaturated, the analysis provides meaningful optical depths (or equivalently, mass-loss rate times ionization fractions, M{dot}q) as a function of normalized velocity, w=v/v. Ratios of M{dot}q (which are independent of M{dot}) determine the behavior of the relative ionization as a function of w. The results demonstrate that, with the exception of O VI in all stars and S VI in the later stars, the ionization in the winds shifts toward lower ionization stages at higher w (contrary to the expectations of the nebular approximation). This result implies that the dominant production mechanism for O VI and S VI in the late O stars differs from the other ions. Using the Vink et al. relationship between stellar parameters and mass-loss rate, we convert the measurements into mean ionization fractions for each ion, <qi>. Because the derived ion fractions never exceed unity, we conclude that the derived values of M{dot} are not too small. However, q(P V), which is expected to be the dominant stage of ionization in some of these winds, is never greater than 0.20. This implies that either the calculated values of M{dot} are too large, the assumed abundance of phosphorus is too large, or the winds are strongly clumped. The implications of each possibility are discussed. Correlations between the mean ion fractions and physical parameters such as Teff, v, and the mean wind density, <ρ>, are examined. Two clear relationships emerge. First, as expected, the mean ionization fraction of the lower ions (C III, N III, Si IV, S IV) decreases with increasing Teff. Second, the mean ion fraction of S VI in the latest stars and O VI in all stars increases with increasing v. This reaffirms the notion, first introduced by Cassinelli & Olson, that O VI is produced nonradiatively. Finally, we discuss specific characteristics of three stars, BI 272, BI 208, and Sk -67°166. For BI 272, the ionic species present in its wind suggest it is much hotter than its available (uncertain) spectral type of O7: II-III:. In the case of BI 208, our inability to fit its observed profiles suggests that its wind is not spherically symmetric. For Sk -67°166, quantitative measurements of its line strengths confirm the suggestion by Walborn et al. that it is a nitrogen-rich O star.