1998A&A...331.1011V

We present the results of stellar evolutionary computations to study the sensitivity of lithium depletion in models of mass and metallicity close to solar, and its dependence on the micro - macro physical inputs in the models, like thermodynamics, mixing, overshooting and the convective model. We find that even marginal chemical inhomogeneities in stellar formation regions lead to a spread in Li-abundances for such stars. We show that a general update in the physical inputs reverses the previously established framework of the problem: solar models including the most recent opacities and equation of state, a Full Spectrum of Turbulence (FST) convection theory and a diffusive description of mixing deplete lithium very efficiently during the pre-Main Sequence (pre-MS). The present solar abundance could be then compatible with all the initial lithium having been burnt during this phase, with no need to invoke depletion during the MS lifetime. This new standard pre-MS Li-depletion is however not consistent with the observed Li-abundances versus mass in young open clusters (e.g. α Per and the Pleiades). We are then led to interpret it as the maximum possible depletion for non rotating, non magnetic stars, and look for physical mechanisms which can inhibit pre-MS depletion. We then included in our code an approximated modeling of the influence of a magnetic field B on the convective envelope. We show that even relatively low values of B are sufficient to largely inhibit Li-depletion during pre-MS. A dynamo generated magnetic field of intensity related to stellar rotation would then lead to different Li-abundances at the end of the pre-MS phase, according to different rotational histories of stars, in qualitative agreement with the observed spread in Li-abundances in young open clusters stars. We discuss the possibility that the right balance between magnetic field and metallicity (and overshooting, if any) can be the relevant parameter in the evaluation of the amount of ⁷Li which should eventually survive after the pre-MS phase. Models of masses 1.25≤M/M_☉≤0.7 are compared with the open cluster ⁷Li - T_eff observations.