The Ionization in the Winds of O Stars and the Determination of Mass‐Loss Rates from Ultraviolet Lines

Abstract

Empirical ionization fractions of C IV, N V, Si IV, and empiricalionization plus excitation fractions of C III* and N iv^* in the windsof 34 O stars and one B star have been derived. We combine the mass-lossrates derived from radio measurements and H{α} with the line fittingof ultraviolet resonance lines and subordinate lines using the Sobolevplus exact integration (SEI) method. The dependence of the empiricalionization fractions, q, on the stellar effective temperature and on themean wind density is discussed. This sets constraints for the models ofionization in the winds of hot stars. The ionization and excitationfractions can be expressed in terms of an empirical radiationtemperature. This radiation temperature scales with T_eff, and we deriveempirical relations for T_rad as a function of T_eff. The radiationtemperatures are on the order of 0.5-0.9 T_eff, with significantdifferences between the ions. The derived relations between theionization fractions and the stellar parameters have an uncertainty of0.2 dex forSi IV, N V, and C iii^*, and about 0.26 dex for N iv^*. For CIV, we can only derive an expression for the mean ionization fraction inthe wind if the mass-loss rate is small, M{\lt}10^-6 M_solar yr^-1,because the C IV lines are usually saturated for higher mass-loss rates.The resulting expressions for T_rad can be used to derive the mass-lossrates from studies of ultraviolet P Cygni profiles in the range ofstellar parameters studied here: 30,000 K{\lt}~T_eff{\lt}~50,500 K,5.2{\lt}~logL/L_*{\lt}~6.4, and -7.5 M_solar{\lt}~logM{\lt}~-4.6 M_solaryr^-1. An accuracy of about a factor of 2 or better can be reached,depending on the lines that are used and the accuracy of the line fitsand the stellar parameters. The Si IV lines give the most reliablemass-loss rates, because the abundance is about the same for all Ostars, the lines saturate only for high mass-loss rates, the doubletlines only partly overlap, and the mass-loss rate is proportional to thesquare root of the column density. The radiation temperature of N Vshows a surprisingly strict relation with T_eff, with a scatter of onlyDeltaT_rad/T_eff=0.01. The mass-loss rate cannot be derived from the N Vlines, because the column density of the N V ions in the wind isindependent of M. A consistency check and a test of the method for thestars HD 14749 and HD 190429 show that the mass-loss rate derived fromthe UV lines with the ionization fractions of this paper agree very wellwith the mass-loss rate derived from new radio flux measurements.