SpringerLink
Forum Springer Astron. Astrophys.
Forum Whats New Search Orders


Astron. Astrophys. 350, 476-484 (1999)

Previous Section Next Section Title Page Table of Contents

1. Introduction

The phenomenon of chemical inhomogeneity within galactic globular clusters is still not clearly understood (see reviews by Suntzeff 1993 and Kraft 1994). While there is only [FORMULA] Centauri and perhaps also M22 which show variations in their iron abundances, many globular clusters have variations in elements like C, N and O (e.g. Hesser 1976). However, other clusters seem to be chemically very homogeneous.

In searching for explanations of abundance variations within globular clusters there are basically two possibilities: primordial variations and inhomogeneities caused by stellar evolution during the giant branch (GB) or later phases. Studies of the abundances of CNO elements in globular clusters have not led to clear results yet. On the one hand, most of the cyanogen (CN) variations can be successfully explained by processes during the CNO cycle (Kraft 1994). On the other hand, CN variations in globular clusters also have been found in main-sequence stars (Suntzeff 1989, Briley et al. 1991), which points to primordial inhomogeneities. Therefore it is very interesting to make a large census of CN band strenghts in both giants and main-sequence stars.

Since heavy elements like iron are not synthesized in present globular-cluster stars, variations of the iron abundance among cluster stars are believed to be of primordial origin. However, in the case of the extraordinarily massive globular cluster [FORMULA] Centauri two other explanations are under discussion. First, there might have been secondary star formation within the cluster from enriched gas that was not blown out of the cluster during its formation (e.g. Norris et al. 1996). Second, [FORMULA] Centauri might be the result of the merging of two clusters with different metallicities (Searle 1977, Icke & Alcaino 1988, Norris et al. 1997).

The inhomogeneity in [FORMULA] Centauri is visible in the colour-magnitude diagram (CMD) by way of a significant colour dispersion among the giant-branch stars (e.g. Persson et al. 1980). Since a broad colour dispersion is visible on the giant branch of the smaller cluster M22 as well (as mentioned already by Arp & Melbourne 1959), this cluster has also been regarded as a candidate for a primordial abundance spread. However, all the iron-rich stars of the sample (DDO survey) by Hesser et al. (1977) have been identified as non-members in later studies (Lloyd Evans 1978, Peterson & Cudworth 1994). Moreover, in more recent spectroscopic investigations (Lehnert et al. 1991, Brown & Wallerstein 1992) measurable variations in iron are not confirmed, whereas variations in CN are clearly detectable (Norris & Freeman 1982, 1983; Brown et al. 1990). Lehnert et al. (1991) give an upper limit for a possible metallicity spread of [FORMULA][Fe/H][FORMULA] dex. The origin of the colour dispersion in M22 might rather be explained by differential foreground reddening. A spectroscopic analysis by Crocker (1988) and polarisation measurements by Minniti et al. (1990, 1992) constrain the reddening variations to be less than 0.08 mag. Bates et al. (1992), investigating the region of M22 with IRAS data, give a value of [FORMULA] 0.05 for the cluster field.

Several studies have shown that the Strömgren [FORMULA] index ([FORMULA] is the difference between the colour indices [FORMULA] and [FORMULA]) ist not only a good indicator for the mean metallicity of late type stars, but is also sensitive for CN abundances (e.g. Bell & Gustafsson 1978). CCD Strömgren photometry offers the possibility to measure metallicities and cyanogen strengths of many giants in globular clusters simultaneously. It is therefore a appropriate tool for adressing the question of metallicity/CN band distribution with a much larger sample of stars than spectroscopy.

Anthony-Twarog et al. (1995) already applied the Strömgren system to M22. They found that the m1 index is indeed closely correlated with CN-band strengths and that CN-variations are found all over the giant branch.

The intention of the present paper is to make a differential comparison between M55 and M22 and to investigate the influence of the differential reddening in M22. We therefore re-investigated the anomal abundance anomalies of M22 with vby photometry and combine our results with spectroscopic measurements of Norris & Freeman (1982, 1983; hereafter refered to as NF). Moreover, we discuss photometric results for M55, for which spectroscopic measured CN abundances are also available (Smith & Norris 1982; Briley et al. 1993). M55 is known to be rather monometallic (Zinn & West 1984) and it is therefore an excellent comparison object in order to investigate CN abundances of two galactic globular clusters with different chemical properties.

Sect. 2 will give an overview of the observations and the data reduction, Sect. 3 and Sect. 4 will present the results and their discussion for the two clusters. In Sect. 5 we summarize our results and give options for future works.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1999

Online publication: October 4, 1999
helpdesk.link@springer.de