Astron. Astrophys. 320, 460-468 (1997)

5. Conclusions

This work presents new yields from WR stars of various metallicities and initial masses. These predictions rely on new stellar models incorporating improved physical ingredients (mass loss rates, opacities, nuclear reaction rates). These models account much better than previous ones for a diversity of important observed WR properties, like luminosities, surface compositions, or number statistics.

Mainly as a result of enhanced mass loss rates, we find yields larger by a factor of about two than in previous computations. We also explore quantitatively the impact on those yields of the stellar metallicity and of other uncertainties that still affect the predicted evolution of WR stars. We show that the resulting uncertainties in the yields are of the order of a factor 2 - 3. These uncertainties are significantly smaller than in other potential sources, like supernovae, novae, or asymptotic giant branch stars.

Finally, we calculate the contribution of the WR stars to the galactic amount of under various assumptions concerning the stellar IMF, the presumed galactic metallicity gradient and supernova rate. We find that WR stars can account for about of during the last y. This is about half of the observationally derived amount. This finding suggests that WR stars may be major contributors to the present galactic .

Improvements of our understanding of the galactic 1.8 MeV emission will certainly come from new -ray facilities. The currently tested Liquid Xenon Coded Aperture Telescope will provide a direct image of -ray line or continuum sources with a spatial resolution better than 2 arc minutes (Aprile et al. 1995). This instrument will greatly facilitate the identification of -ray sources with known stellar objects. The INTEGRAL satellite, the launch of which is scheduled for 2001, will also shed new light on the distribution of the sources in the Galaxy, and will provide stringent tests of the models. This satellite will indeed be able to detect 1.8 MeV sources about five times fainter than the faintest ones observed by CGRO. If the predicted yields are correct, both the Vela supernova and the WR star should be detectable by this instrument.

© European Southern Observatory (ESO) 1997

Online publication: June 30, 1998
helpdesk.link@springer.de