Letter to the Editor

Li – O anti-correlation in NGC 6752: evidence for Li-enriched polluting gas ^⋆

¹ National Astronomical Observatories, Chinese Academy of Science, 20a Datun road, Chaoyang District, Beijing, PR China
e-mail: zshen@bao.ac.cn
² ESO – European Southern Observatory, Karl-Schwarzschild-Strasse 2, 85748 Garching bei München, Germany
e-mail:lpasquin@eso.org
³ GEPI, Observatoire de Paris, CNRS, Université Paris Diderot, Place Jules Janssen, 92190 Meudon, France
⁴ Istituto Nazionale di Astrofisica, Osservatorio Astronomico di Trieste, via Tiepolo 11, 34143 Trieste, Italy
⁵ Istituto Nazionale di Astrofisica, Osservatorio Astronomico di Padova, Vicolo dell’Osservatorio, Padova, Italy

Received: 10 September 2010
Accepted: 24 October 2010

Abstract

Context. Elemental correlations and anti-correlations are known to be present in globular clusters (GCs) owing to pollution by CNO cycled gas. Because of its fragility Li is destroyed at the temperature at which the CNO cycling occurs, and this makes Li a crucial study for the nature of the contaminating stars.

Aims. We observed 112 un-evolved stars at the Turnoff of the NGC 6752 cluster with FLAMES at the VLT to investigate the presence and the extent of a Li-O correlation. This correlation is expected if there is a simple pollution scenario.

Methods. Li (670.8 nm) and O triplet (771 nm) abundances are derived in NLTE. All stars belong to a very narrow region of the color-magnitude diagram, so they have similar stellar parameters (T_eff, log g).

Results. We find that O and Li correlate, with a high statistical significance that confirms the early results for this cluster. At first glance this is what is expected if a simple pollution of pristine gas with CNO cycled gas (O-poor, Li-poor) occurred. The slope of the relationship, however, is about 0.4, and differs from unity by over 7σ. A slope of one is the value predicted for a pure contamination model.

Conclusions. We confirm an extended Li-O correlation in non evolved stars of NGC 6752. At the same time the characteristic of the correlation shows that a simple pollution scenario is not sufficient to explain the observations. Within this scenario the contaminant gas must have been enriched in Li. This would rule out massive stars as main polluters, and favor the hypothesis that the polluting gas was enriched by intermediate or high-mass AGB stars, unless the former can be shown to be able to produce Li. According to our observations, the fraction of polluting gas contained in the stars observed is a considerable fraction of the stellar mass of the cluster.