... stars[*]
The whole Tables 3-6, 8, 9 are only available in electronic form at http://www.edpsciences.org 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
...3,2,[*]
Visiting Astronomer at the LATT - OMP, Toulouse, France. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
...(Li)[*]
A(Li) = 12 + log N(Li)/N(H). 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... birth[*]
Actually the primordial Li abundance could even be lower than the plateau value because of production in the early Galaxy (Ryan et al. 1999; Suzuki et al. 2000). 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... tomographer[*]
Since lithium is destroyed at quite low temperatures (for stellar interiors) of the order of $2.5 \times 10^6~$K, it is a powerful tool to identify the mechanisms active in stellar interiors and responsible for convective and/or radiative transports, mixing, diffusion, presence of gravitational waves. Together with beryllium and boron, that burn at $3.5 \times 10^6~$K and $5.0 \times 10^6~$K respectively, lithium abundances allow us to make a stellar tomography of the external atmospheric layers where these three light nuclides are "nuclearly'' preserved (since the epoch of formation when looking at unevolved objects). 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
...[*]
Available in its entirety on-line. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
...[*]
Only available in its entirety on-line. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
...[*]
Available in its entirety on-line. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... Sect. 13[*]
For the metallicity and effective temperature range of the plateau, only three stars have Li upper limits, namely HIP 72561, 81276 and 100682. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
...[*]
Available in their entirety on-line. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... mechanisms[*]
Nuclear burning at the basis of the convective envelope on the pre- and early-main sequence, diffusion by gravitational settling on the main sequence and at the turnoff, convective dredge-up on the early post-main sequence, and dilution on the post-main sequence. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... samples[*]
The ubvy and $\beta $ samples do not contain additional objects. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
...$\leq $ 1.287)[*]
We note that the error on the Hipparcos parallax is relatively high for HIP 36430, 92775 and 83320, which turns into a significant error bar on the determined luminosities. The status of subgiant can however be attributed relatively safely to these three objects (see also Fig. 16). 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... stars[*]
Additional ULDs which can be found in the literature do not have Hipparcos parallaxes and are thus excluded from our sample and from the present discussion. See references in Ryan et al. (2002). 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... ULDs[*]
Two of these objects do not appear in our study, one of them (CD-31 $\hbox{$^\circ$ }$19466) because it does not have an Hipparcos parallax, the other because of its relatively high metallicity (BD+51 $\hbox{$^\circ$ }$1817, [Fe/H] = -0.88). 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... R02[*]
This followed Fuhrmann & Bernkopf (1999) who suggested such a relation in the case of thick-disk, binary blue stragglers. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... values[*]
Standard models of Pop II stars predict only a slight depletion of 6Li (which occurs mainly during the pre-main sequence phase) for stars which are now at the turnoff. The depletion factor increases for lower mass dwarf stars (see e.g. Deliyannis & Malaney 1995; and Cayrel et al. 1999b). This explains why 6Li has been found so far only in turnoff stars relatively metal-poor. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... models[*]
This refers to the modeling of non-rotating, non-magnetic stars in which convection is the only transport process considered. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... model[*]
It is now widely accepted that the standard stellar models are those in which the effects of atomic diffusion are taken into account and not counterbalanced by any macroscopic process. They are calculated from first principles without any arbitrary parameter except for the mixing length. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... Ri02 [*]
In a Tx.y model, the turbulent diffusion coefficient is 400 times larger than the He atomic diffusion at log T= x.y and varies as $\rho^{-3}$. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... rotation[*]
The mass dependance of the gravity waves efficiency leads to a natural explanation of the fast horizontal branch rotators. It also provides a solution to the enigma of the so-called Li dip observed in Pop I stars in terms of rotational mixing, forming a coherent picture of mixing in main sequence stars of all masses. Also, gravity waves are able to shape the Sun's flat rotation profile deduced from helioseismology. See Talon & Charbonnel (2003, 2004) and Charbonnel & Talon (2005) for more details. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Copyright ESO 2005