The lithium-age relationcalibration with open clusters and associations

Abstract

Lithium is a very fragile element that is easily destroyed in stellar interiors. In solar-type and lower mass stars, lithium is slowly being depleted and its surface abundance decreases over time. For this reason, lithium is a very sensitive tracer of stellar evolution, and is especially relevant in the determination of the age of stellar clusters. The lithium abundance (derived from the Li 6707.76 Å line) observed in late-type stars is strongly age-dependent, but also shows a complex pattern depending on several parameters such as rotation, chromospheric activity, metallicity, mixing mechanisms, convection structure, and magnetic activity. The best way to calibrate these effects is to analyse and calibrate coeval groups of stars, such as open clusters and associations. This project has made use of data provided by the Gaia-ESO Survey (GES), a large, public spectroscopic survey that provides an homogeneous overview of the distribution of kinematics, dynamical structure and chemical compositions in the Galaxy. GES has measured data for a great number of stars, as well as near 100 open clusters, from 2011 until 2018. This survey uses the multi-object spectrograph FLAMES on the Very Large Telescope (ESO, Chile) to obtain both high resolution spectra with UVES (Ultraviolet and Visual Echelle Spectrograph) and medium resolution spectra with GIRAFFE. As part of the UCM GES node, in the first stages of this thesis project I performed an extensive analysis of the UVES spectra, manually measuring the equivalent widths (both the Li I line at 6707.76 Å and the adjacent Fe I line at 6707.43 Å) for all UVES spectra, in the context of the homogenization of what then was the fourth data release of GES (iDR4). For all the analysis in the present project, however, we have updated all results using the last GES data release, iDR6, as well as the data provided by Gaia EDR3. The present thesis, entitled The lithium-age relation: Calibration with open clusters and associations, is a large scale work in which we use a series of 42 open clusters and associations observed by GES, ranging in ages from 1 Myr to 5 Gyr, with the aim of studying lithium as an age indicator for pre- and main-sequence FGK late-type stars, and in this way calibrating an empirical Li-age relation. Our first step in this study was to perform a thorough membership analysis in order to obtain lists of candidate members for all 42 clusters in our sample, making use of all available GES parameters, and based on the following criteria: Firstly, we studied the radial velocity distributions of each cluster to obtain probable kinematic members, and we combined this kinematic selection with an analysis of the proper motions and parallaxes provided by Gaia. Gravity indicators such as log g and the γ index enabled us to discard field giant contaminants, and in this way we also obtained, as an additional result of the membership analysis, a series of Li-rich giant stars that we have listed in this work given their interest. We also used photometry from Gaia in colour−magnitude diagrams (CMDs) to confirm the membership of the astrometric selections, as well as [Fe/H] metallicity, which helped rule out further contaminants which might still remain in the list. Finally, we studied lithium as a final criterion by plotting the candidates in EW(Li) vs Teff diagrams, and thus obtained the final lists of candidate stars. All these membership analyses were complemented with detailed bibliographical searches for all clusters in the sample, compiling all previous data regarding Li abundances, age estimations, velocities and metallicities, as well as existing membership studies. Having obtained the member selections for all clusters in the sample, our next step was to conduct a comparative study that allowed us to quantify the observable lithium dispersion in each cluster, and analyse in detail its dependence with several other stellar parameters derived from the GES spectroscopic observations: rotation, accretion indicators, the level of chromospheric activity (Hα), and metallicity. For the study of rotation, we used both rotational velocities provided by GES (vsini), as well as a series of periods obtained from the literature, including CoRoT, Kepler, K2 and TESS measurements. With the aid of several types of figures and diagrams, we studied and confirmed the findings and correlations described in former publications: We found, for example, that members with higher values of Li tended to be faster rotators and often also had higher levels of activity. Lastly, we additionally observed the effects of metallicity in the Li depletion of coeval clusters for those which are metal-rich or metal-poor. Taking all these effects into account, all this information allowed us to finally calibrate a Li-age relation as the final part of this project. In order to do so, we created a series of empirical lithium envelopes for key ages in our cluster sample, from a few Myr to several Gyr. And to be able to obtain the most complete envelopes possible, we have also constrained the lithium depletion boundary (LDB) for those clusters in the 15-500 Myr age range, with the aid of several models. One of the applications of these lithium envelopes when it comes to using lithium as an effective age indicator is to plot field stars whose age is yet unknown in EW(Li) vs Teff diagrams, and so use them as references and a guide to estimate their ages. Regarding future work, our main objective is to use this Li-age relation and the envelopes we have obtained to estimate age ranges for GES field stars whose ages are still unknown, as well as confirm the membership of these field stars to stellar kinematic groups of different ages. An additional point of interest is the study of the unknown Li-rich giant stars observed in the field of these clusters, given their exceptional nature and their usefulness to further understand lithium in stars.