Ever since the discovery of the edge-on circumstellar disk around beta Pictoris, a standing question has been why the gas observed against the star in absorption is not rapidly expelled by the strong radiation pressure from the star. A solution to th
e puzzle has been suggested to be that the neutral elements that experience the radiation force also are rapidly ionized, and so are only able to accelerate to an average limiting velocity v_ion. Once ionized, the elements are rapidly braked by C II, which is observed to be at least 20x overabundant in the disk with respect to other species. A prediction from this scenario is that different neutral elements should reach different v_ion, depending on the ionization thresholds and strengths of driving line transitions. In particular, neutral Fe and Na are predicted to reach the radial velocities 0.5 and 3.3 km/s, respectively, before being ionized. In this paper we study the absorption profiles of Fe and Na from the circumstellar gas disk around beta Pic, as obtained by HARPS at the ESO 3.6m telescope. We find that the Fe and Na velocity profiles are indeed shifted with respect to each other, confirming the model. The absence of an extended blue wing in the profile of Na, however, indicates that there must be some additional braking on the neutrals. We explore the possibility that the ion gas (dominated by C II) can brake the neutrals, and conclude that about 2-5x more C than previously estimated is needed for the predicted line profile to be consistent with the observed one.
We estimate cluster ages from lithium depletion in five pre-main-sequence groups found within 100 pc of the Sun: TW Hydrae Association, Eta Chamaeleontis Cluster, Beta Pictoris Moving Group, Tucanae-Horologium Association and AB Doradus Moving Group.
We determine surface gravities, effective temperatures and lithium abundances for over 900 spectra through least squares fitting to model-atmosphere spectra. For each group, we compare the dependence of lithium abundance on temperature with isochrones from pre-main-sequence evolutionary tracks to obtain model dependent ages. We find that the Eta Chamaelontis Cluster and the TW Hydrae Association are the youngest, with ages of 12+/-6 Myr and 12+/-8 Myr, respectively, followed by the Beta Pictoris Moving Group at 21+/-9 Myr, the Tucanae-Horologium Association at 27+/-11 Myr, and the AB Doradus Moving Group at an age of at least 45 Myr (where we can only set a lower limit since the models -- unlike real stars -- do not show much lithium depletion beyond this age). Here, the ordering is robust, but the precise ages depend on our choice of both atmospheric and evolutionary models. As a result, while our ages are consistent with estimates based on Hertzsprung-Russell isochrone fitting and dynamical expansion, they are not yet more precise. Our observations do show that with improved models, much stronger constraints should be feasible: the intrinsic uncertainties, as measured from the scatter between measurements from different spectra of the same star, are very low: around 10 K in effective temperature, 0.05 dex in surface gravity, and 0.03 dex in lithium abundance.
