1999A&A...348..487R

We present a lithium survey for a sample of 91 Pop. I stars. JHKL photometry was also obtained for 61 stars in the sample. Besides Li abundances, [Fe/H] values were derived. Thanks to Hipparcos parallaxes, we could infer absolute V magnitudes for our sample stars and were able to place them on the color-magnitude diagram, which allowed us to constrain their evolutionary status. Masses and ages were derived for most of the stars by comparison with evolutionary tracks. The sample was originally selected so to include class IV stars later than spectral-type F0, but, based on the location on the color-magnitude diagram, we found a posteriori that a fraction of the stars (about 20%) are either main sequence stars or evolved giants. As it is the case for dwarfs and giants, a large spread in lithium abundance is present among the subgiants in our sample. As expected, the average lithium decreases as the stars evolve along the subgiant branch; however, there is not a one-to-one relationship between the position on the color-magnitude diagram and lithium abundance, and the observed dispersion is only partially explainable as due to a dispersion in mass, metallicity, and age. In particular, a dispersion in lithium is seen among slightly evolved subgiants with masses close to solar but in the same evolutionary stage as the G2 IV star β Hyi. The comparison of the β Hyi-like sample with a sample of non evolved solar-like stars indeed suggests that β Hyi has most likely evolved from a main sequence Li-rich star, rather than from a Li-poor star (like the Sun) that has dredged-up previously stored lithium. Our sample includes several stars that have completed the first-dredge up lithium dilution, but that have not yet evolved to the evolutionary point where extra-mixing in the giant phase is thought to occur. A large number of them have Li abundances considerably below the theoretical predictions of first dredge-up dilution. We confirm that this is due to the fact that the progenitors of these stars are most likely stars that have depleted lithium while on the main sequence; the fraction of post-dredge up Li rich/poor stars, in fact, is consistent with the observed distribution of Li abundances among stars that have just left the main sequence. The signature of the second mixing (or RGB extra-mixing) episode is evident in the log n(Li) vs. B-V and log n(Li) vs. M_bol distributions of the stars in the sample; it seems however that the extra-mixing occurs at luminosities lower than predicted by the models of Charbonnel (1994). Finally, a few evolved giants are found that should have passed the second mixing episode, but that do not show signs of it. At least half of them are spectroscopic binaries.