Abundances in Globular Clusters

Stars in each GC, with the exception of $\omega$ Cen and possibly of M 22, generally have a very homogeneous composition as far as Iron-peak and $\alpha$ -elements are considered, while abundances of lighter elements (from C to Al) show a complex, not yet fully explained, pattern (i.e., CN-CH band strength anti-correlation, Na-O anti-correlation, etc.) not seen in field stars. Proposed explanations have varied from an in situ mechanism (e.g., very deep mixing of nuclear-processed material) to an external source of material (either primordial proto-cluster gas or processed material from a polluting companion). Both explanations could work for RGB stars, while main sequence stars require the latter.

Abundances along the RGB
People involved at OAB: Bragaglia, Bellazzini, Cacciari, Monaco, Pancino.

In the framework of a large programme in collaboration with researchers in Padova and Roma, we aim at deriving accurate abundances for a representative number of giants in several globular clusters by using the multiplex FLAMES facility at the VLT (UVES + GIRAFFE). Our first target has been M 22, since there are reasons to believe that M 22 could have a significant dispersion in metallicity. Even though the effect is going to be smaller than in $\omega$ Cen, the detection of such a spread in metallicity would indicate that at least another globular cluster in our Galaxy has experienced the same kind of chemical enrichment history as $\omega$ Cen. Such a discovery would be of invaluable help in understanding the mechanism of formation of $\omega$ Cen and would shed light on the processes that are at the basis of the halo formation and evolution. Accurate wide field photometry of M22 has already been published (Monaco et al. 2004), showing that the metallicity spread should be $\leq0.1$ dex. Using this catalog, we have selected $\sim250$ stars that have been observed with UVES and FLAMES and the data analysis is under way.

Another project on the RGB of a quite peculiar GC (NGC 2808) has been conducted, based on Science Verification observations for FLAMES@VLT, proposed by Cacciari and Bragaglia. A sample of 137 RGB stars (the largest ever in a given cluster) have been observed with UVES/FLAMES and GIRAFFE, monitoring about 3 mag from the RGB tip, and taking spectra in regions containing mass loss and chromospheric activity diagnostics: Ca II H+K, Na I D, and H $\alpha$ lines (see Fig. 2). We have found indication of outward mass motions, and confirmed the presence of chromospheres (Cacciari et al. 2004).

**Figure 2:** Region centered on H $\alpha$ of 13 stars on the RGB in NGC 2808 observed with UVES. From all spectra we have subtracted a template profile (star 53390, which did not show any evidence of irregularities) to enhance asymmetries; results are plotted as solid gray lines. The dotted lines show instead the differences with the corresponding theoretical profiles.
$\begin{figure}\centerline{\psfig{file=figura5_SV.ps,width=0.75 \hsize}}\end{figure}$

The same dataset has been used to study the Na abundance along the RGB. From 82 stars we determined a large star-to-star scatter in Na, at all magnitude levels. Comparison with other clusters has led us to suggest a primordial origin for the different Na abundances (Carretta et al. 2003).

Similar work has recently begun on science commissioning data obtained on 47 Tuc.

Abundances of Main Sequence Turn-Off stars
People involved at OAB: Bragaglia, Clementini

Only recently stars near the MSTO have become observable at the necessary S/N and resolution. As part of an ESO Large Programme (PI Gratton), we have acquired with UVES@VLT in 2000, 2001 and 2002 (30 nights in total) high resolution spectra (R $\geq$ 40000) of a large number of stars at the MSTO and at the base of the sub-giant branch in NGC 6752, NGC 6397, 47 Tuc, NGC 6809 and NGC 7099. Abundance analysis for Fe, Li, and other elements has been completed for the first three clusters (e.g., Gratton et al. 2001, Bonifacio et al. 2002, James et al. 2004, Carretta et al. 2004).

The [Fe/H] abundances determined for NGC 6752, NGC 6397 and 47 Tuc (identical for MSTO and subgiant stars; Gratton et al. 2001, 2003) have removed one of the possible major sources of uncertainty claimed to affect the MSF distances (see Section 1.2.2). Our analysis has also put a strong constraint on sedimentation.

Furthermore, we have found presence of O-Na and Mg-Al anticorrelations in MSTO stars in NGC 6752 (Gratton et al. 2001) and 47 Tuc (Carretta et al. 2004), hence ruling out internal mixing as the cause of such anomalies.

This work is in collaboration with Gratton and Carretta (INAF-Padova Obs.), Bonifacio (INAF-Trieste Obs.), Pasquini (ESO), and many more (see list of publications).

The galactic bulge: globular clusters and field populations
People involved at OAB: Origlia, Valenti.

Bulge GCs are a fundamental stellar population of our Galaxy and it is most interesting to compare their detailed compositions with the Galactic bulge field population (McWilliam & Rich 1994). For many of these bulge clusters, foreground extinction is so large as to preclude any photometric and spectroscopic optical study. In the last few years we have undertaken a long-term project devoted to study a representative sample of Bulge globular clusters in the infrared.

Using the SOFI medium-resolution imager/spectrograph at the ESO NTT telescope and the NIRSPEC high-resolution echelle spectrograph at Keck II, we are performing a systematic survey to measure spectra of several tens of M-giant stars in bulge globular clusters, in Baade's window and near the galactic center. The spectral analysis is in progress.

We measure several single roto-vibrational OH lines and CO bandheads to derive accurate Oxygen and Carbon abundances. Other metal abundances can be derived from the atomic lines of Fe I, Mg I, Si I, Ti I, Ca I and Al I. Abundance analysis is performed by using full spectral synthesis techniques and equivalent width measurements of representative lines.

We find an overall [ $\alpha$ /Fe] enhancement up to about solar metallicities, both in the cluster and field populations, that is consistent with a scenario in which the bulge formed early, with rapid enrichment (Origlia, Rich & Castro 2002, Origlia & Rich 2004).

This reasearch is in collaboration with F. Ferraro (Astron. Dept., Bologna University) and R.M. Rich (UCLA, USA).