Cu I resonance lines in turn-off stars of NGC 6752 and NGC 6397. Effects of granulation from CO5BOLD models

TL;DR

This study assesses the reliability of Cu I resonance lines in turn-off stars for copper abundance measurements, highlighting the importance of 3D models and NLTE effects, and recommends using giant star lines for more accurate results.

Contribution

It provides a comparative analysis of copper abundance indicators in globular cluster stars using 3D hydrodynamical models and discusses the impact of NLTE effects on these measurements.

Findings

Cu I resonance lines in TO stars are unreliable for abundance measurements.

Large 3D LTE corrections are needed for TO stars, small for giants.

NLTE effects may significantly alter copper abundance estimates.

Abstract

Context. Copper is an element whose interesting evolution with metallicity is not fully understood. Observations of copper abundances rely on a very limited number of lines, the strongest are the Cu I lines of Mult. 1 at 324.7 nm and 327.3 nm which can be measured even at extremely low metallicities. Aims. We investigate the quality of these lines as abundance indicators. Method. We measure these lines in two turn-off (TO) stars in the Globular Cluster NGC 6752 and two TO stars in the Globular Cluster NGC 6397 and derive abundances with 3D hydrodynamical model atmospheres computed with the CO5BOLD code. These abundances are compared to the Cu abundances measured in giant stars of the same clusters, using the lines of Mult. 2 at 510.5 nm and 578.2 nm. Results. The abundances derived from the lines of Mult. 1 in TO stars differ from the abundances of giants of the same clusters. This is true both using CO5BOLD models and using traditional 1D model atmospheres. The LTE 3D corrections for TO stars are large, while they are small for giant stars. Conclusions. The Cu I resonance lines of Mult. 1 are not reliable abundance indicators. It is likely that departures from LTE should be taken into account to properly describe these lines, although it is not clear if these alone can account for the observations. An investigation of these departures is indeed encouraged for both dwarfs and giants. Our recommendation to those interested in the study of the evolution of copper abundances is to rely on the measurements in giants, based on the lines of Mult. 2. We caution, however, that NLTE studies may imply a revision in all the Cu abundances, both in dwarfs and giants.