A Comparison between Carbon and Nitrogen Abundances of Bright Giants in the Globular Clusters M13, M10, and NGC 7006

Abstract

Carbon and nitrogen abundances have been derived for red giants in the globular clusters M13, M10, and NGC 7006. These constitute a ''second-parameter'' trio of clusters having very similar metallicities (½Fe=H % À1:55) but different horizontal-branch (HB) morphologies, with NGC 7006 having a much higher fraction of red HB stars than either M13 or M10. The red giants observed have absolute magnitudes in the range À2:5 < M V < 0:7 in the case of M13, À1:8 < M V < 0:4 for M10, and À2:1 < M V < À1:2 for the more distant NGC 7006. Most of the spectra were obtained with the Lick Observatory 3 m Shane reflector, supplemented by observations of M10 with the McDonald Observatory 2.7 m telescope. These data were reduced in a homogeneous manner and compared with synthetic spectra computed from model atmospheres in order to derive ½C=Fe abundances from the k4300 G band and ½N=Fe abundances from the k3883 CN band. The giants in all three clusters appear to exhibit a common relationship between ½C=Fe and M V in which the carbon abundance becomes increasingly diminished with increasing luminosity on the giant branch. Such behavior was previously discovered for M13 by Suntzeff and is known to occur in other low-metallicity globular clusters, such as M3, M92, and M15, as well as among halo field red giants. The phenomenon is widely recognized as indicating a progressive dredge-up of CNO-processed material from regions close to the H-burning shell within Population II giants. The similarities between the ½C=Fe-M V trends in the three second-parameter clusters indicates that the deep-mixing mechanism is relatively uncorrelated with cluster HB type. This result may be inconsistent with scenarios that rely on rotation to both drive deep mixing within metal-poor red giants and to set the eventual zero-age HB position of these stars, unless there is a threshold for rotation-induced mixing that is well below that needed to delay the He-flash. The nitrogen abundance is found to differ by as much as 1.0 dex between giants in the program clusters and to be anticorrelated with the carbon abundance. It is speculated that the behavior of ½N=Fe with respect to M V can be interpreted as a combination of deep mixing and primordial contributions.

On the origin of double main-sequence turn-offs in star clusters of the Magellanic Clouds

“…The origin of the observed star‐to‐star light element abundance inhomogeneities in the Galactic GCs (GGCs) has been extensively discussed both theoretically and observationally (e.g. Sneden et al 1992; Norris & Da Costa 1995; Smith, Briley & Harbeck 2005; see Gratton, Sneden & Carretta 2004 for a recent review). The observed presence of star‐to‐star abundance inhomogeneities in less evolved stars on the main sequence and subgiant branch (e.g.…”

A Moderate Sample Size, Multielement Analysis of the Globular Cluster M12 (NGC 6218)

“…Although C and N abundances for M12 are currently unavailable, the CNO abundances of M3, M13, M4, and NGC 7006 all show evidence of mixing in which, on average, C and O decline, N is enhanced, and the C+N+O sum is roughly constant at all RGB luminosities (Suntzeff 1981;Smith et al 1996Smith et al , 2005Ivans et al 1999;Sneden 1999;Briley et al 2004). This would be expected for stars in which the CNO cycle was operating in the outer hydrogen shell, where convective mixing could bring these products to the surface.…”

Secondary star formation within massive star clusters: origin of multiple stellar populations in globular clusters

“…It has long been discussed both observationally and theoretically why some of the Galactic GCs show star‐to‐star inhomogeneity among the light elements of stars and what physical mechanisms are responsible for the inhomogeneity (e.g. Cottrell & Da Costa 1981; Sneden et al 1992; Norris & Da Costa 1995; Cannon et al 1998; D’Antona & Caloi 2004; Fenner et al 2004; Gratton, Sneden & Carretta 2004; Norris 2004; Lee et al 2005; Smith, Briley & Harbeck 2005; Bekki et al 2007b; Alves‐Brito et al 2008; Catelan 2008; Kayser et al 2008; Piotto 2008; Da Costa et al 2009; Marcolini et al 2009; Yong et al 2009; Carretta et al 2010; D’Ercole et al 2010; Romano et al 2010; van Loon 2010). Although such star‐to‐star inhomogeneity was discovered in 1970s and 1980s (e.g.…”