The very old, very massive cluster NGC 6791 is probably the most famous example of a metal-rich OC. Interestingly, in the first photometric study of this cluster (Kinman, 1965) a possible metal deficiency was reported and only in later studies (in particular spectroscopic ones) was its metal-rich nature ascertained (Origlia et al. 2006 and references therein). Besides NGC 6791, other clusters with higher-than-Solar metallicity have been detected by photometric or spectroscopic means during the last few years. In order to estimate their fraction, we plot in Figure 9.1 the distribution of [Fe/H] values for all clusters for which metallicity is available (from photometry or spectroscopy) and that for OCs with spectroscopic [Fe/H] (from either low- or high-resolution measurements). The information was retrieved from the catalog of Dias et al. (2002) (see also www.astro.iag.usp.br/wilton/). Of the 1756 OCs contained in the catalog (at the time of writing), for only 143 (i.e., less than 10%) is an estimate of metallicity available and for only 68 has there been a determination of [Fe/H] from spectroscopy. In a few cases (like NGC 6791 itself) we updated the [Fe/H] values given in the catalog with more recent and higher-quality determinations.
Considering the whole sample of OCs for which [Fe/H] data are available, about 30% (39 of 143), 12% (17 of 143), and 1.5% (2 of 143) of the clusters have [Fe/H] greater than 0.05, 0.1, and 0.2, respectively. These percentages change to ~23%
Figure 9.1. The metallicity distribution of Galactic open clusters. Solid line: all clusters from the catalog of Dias et al. (2002) for which [Fe/H] determinations are available. Dashed line: clusters with spectroscopic [Fe/H] available.
(16 of 68), 15% (10 of 68), and 4.4% (3 of 68) for clusters with [Fe/H] from spectroscopy. With the caveat that the sample of OCs with a metallicity determination is far from complete and certainly biased toward closer clusters, these percentages suggest that a non-negligible fraction of OCs has a metallicity moderately higher than Solar ([Fe/H] > 0.1), while only a very small fraction of the whole cluster population has a metallicity considerably higher than Solar ([Fe/H] > 0.2).
We choose [Fe/H] = 0.1 (i.e. an iron content ~25% higher than Solar) as a (somewhat arbitrary) threshold between metal-rich and non-metal-rich. Also, the discussion in the following sections will be based on the spectroscopic sample only, under the assumption that [Fe/H] values from spectroscopy are more reliable than estimates from photometry. Determinations of spectroscopic metallicities for a much larger number of clusters are indeed badly needed; we note, however, that there is an overall good agreement between the spectroscopic and photometric metallicity distributions shown in Figure 9.1.
Was this article helpful?