Astron. Astrophys. 329, 551-558 (1998)

1. Introduction

Although we have a solid knowledge of the evolution of the deep interior of massive stars since a long time (Weaver et al. 1978, Kippenhahn & Weigert, 1990) it has become more and more clear during the last decades that our understanding of the evolution of their observable features is still rather incomplete. Among others, open problems concern the effective temperature evolution of moderately massive () post main sequence stars (Langer & Maeder 1995) and the evolutionary connections between O stars, Luminous Blue Variables (LBVs) and Wolf-Rayet (WR) stars for higher masses (Schaller et al. 1992, Langer et al. 1994, Stothers & Chin 1996, Pasquali et al. 1997). Two major physical difficulties in the theoretical models of the observable evolutionary stages of massive stars have been identified and made responsible for the persisting lack of reliable models: mass loss and internal mixing processes (Meynet et al. 1994, Langer 1994, Deng et al. 1996). E.g., it has been found that the mass loss of massive main sequence stars should be roughly twice as high as what appears to be observed in order to understand many features of massive post main sequence stars (Meynet et al. 1994, Langer et al. 1994). Additionally, there is growing evidence that stellar rotation may considerably affect the evolution of massive stars (Maeder 1987, Langer 1991a, Fliegner et al. 1996, Maeder & Meynet 1996, Meynet & Maeder 1997, Langer 1997, Langer et al. 1997a,b).

Clearly, massive main sequence stars are rapid rotators (cf. Fukuda 1982, Howarth et al. 1997). Massive stars are also very luminous, which brings about three major problems concerning their theoretical modeling. The first two, which have already been mentioned above, are mass loss - which is driven by their enormous photon fluxes (cf. Puls et al. 1996) - and internal mixing, which is much easier in massive stars due to the large contribution of the radiation pressure - which is independent of the mean molecular weight - to the total pressure in the star (e.g., Maeder 1987). In the present paper, we shall be concerned with a third problem brought along by the high luminosity, i.e. the proximity of the stellar surface to hydrodynamic instability, i.e. to the Eddington-limit. We want to show in the following that in rotating luminous stars the mass loss may be considerably enhanced over the (usually considered) non-rotating case, and that there is a maximum mass loss rate these stars may achieve which is actually controlled by the accompaniing angular momentum loss rate and not by the radiation force.

In the next section, we will spell out and discuss the physical methods and assumptions used for our investigation. Sect. 3 contains the results of exemplary stellar evolution calculations for the case of a stars. In Sect. 4, we shall discuss problems, consequences and further applications of the concept of coupled mass and angular momentum loss. Conclusions are given in Sect. 5.

© European Southern Observatory (ESO) 1998

Online publication: December 8, 1997
helpdesk@link.springer.de