Numerous physical aspects of stellar physics have been presented in Ses- sion 2 and the underlying uncertainties have been tentatively assessed. We try here to highlight some specific points raised after the talks and during the general discus- sion
at the end of the session and eventually at the end of the workshop. A table of model uncertainties is then drawn with the help of the participants in order to give the state of the art in stellar modeling uncertainties as of July 2013.
Massive globular clusters (GCs) contain at least two generations of stars with slightly different ages and clearly distinct light elements abundances. The Na-O anticorrelation is the best studied chemical signature of multiple stellar generations. Lo
w-mass clusters appear instead to be usually chemically homogeneous. We are investigating low-mass GCs to understand what is the lower mass limit where multiple populations can form, mainly using the Na and O abundance distribution. We used VLT/FLAMES spectra of giants in the low-mass, metal-poor GC Terzan 8, belonging to the Sagittarius dwarf galaxy, to determine abundances of Fe, O, Na, alpha-, Fe-peak, and neutron-capture elements in six stars observed with UVES and 14 observed with GIRAFFE. The average metallicity is [Fe/H]=-2.27+/-0.03 (rms=0.08), based on the six high-resolution UVES spectra. Only one star, observed with GIRAFFE, shows an enhanced abundance of Na and we tentatively assign it to the second generation. In this cluster, at variance with what happens in more massive GCs, the second generation seems to represent at most a small minority fraction. We discuss the implications of our findings, comparing Terzan 8 with the other Sgr dSph GCs, to GCs and field stars in the Large Magellanic Cloud, Fornax, and in other dwarfs galaxies.
Two independent studies recently uncovered two distinct populations among giants in the distant, massive globular cluster (GC) NGC 2419. One of these populations has normal magnesium (Mg) and potassium (K) abundances for halo stars: enhanced Mg and r
oughly solar K. The other population has extremely depleted Mg and very enhanced K. To better anchor the peculiar NGC 2419 chemical composition, we have investigated the behavior of K in a few red giant branch stars in NGC 6752, NGC 6121, NGC 1904, and omega Cen. To verify that the high K abundances are intrinsic and not due to some atmospheric features in giants, we also derived K abundances in less evolved turn-off and subgiant stars of clusters 47 Tuc, NGC 6752, NGC 6397, and NGC 7099. We normalized the K abundance as a function of the cluster metallicity using 21 field stars analyzed in a homogeneous manner. For all GCs of our sample, the stars lie in the K-Mg abundance plane on the same locus occupied by the Mg-normal population in NGC 2419 and by field stars. This holds both for giants and less evolved stars. At present, NGC 2419 seems unique among GCs.
It is now commonly accepted that globular clusters (GCs) have undergone a complex formation and that they host at least two stellar generations. This is a recent paradigm and is founded on both photometric and spectroscopic evidence. We concentrate o
n results based on high-resolution spectroscopy and on how we moved from single to multiple stellar populations concept for GCs. We underline that the peculiar chemical composition of GC stars is fundamental in establishing the multiple populations scenario and briefly outline what can be learned from observations. Finally, recent observational results on large samples of stars in different evolutionary phases are discussed.
We present aluminium abundances for a sample of about 100 red giant stars in each of the Galactic globular clusters 47 Tuc (NGC 104) and M 4 (NGC 6121). We have derived homogeneous abundances from intermediate-resolution FLAMES/GIRAFFE spectra. Alumi
nium abundances are from the strong doublet Al I at 8772-8773 A as in previous works done for giants in NGC 6752 and NGC 1851, and nitrogen abundances are extracted from a large number of features of the CN molecules, by assuming a suitable carbon abundance. We added previous homogeneous abundances of O and Na and newly derived abundances of Mg and Si for our samples of 83 stars in M 4 and 116 stars in 47 Tuc to obtain the full set of elements from proton-capture reactions produced by different stellar generations in these clusters. By simultaneously studying the Ne-Na and Mg-Al cycles of H-burning at high temperature our main aims are to understand the nature of the polluters at work in the first generation and to ascertain whether the second generation of cluster stars was formed in one or, rather, several episodes of star formation. Our data confirm that in M 4 only two stellar populations are visible. On the other hand, for 47 Tuc a cluster analysis performed on our full dataset suggests that at least three distinct groups of stars are present on the giant branch. The abundances of O, Na, Mg and Al in the intermediate group can be produced within a pollution scenario; results for N are ambiguous, depending on the C abundance we adopt for the three groups.
We present aluminium, magnesium, and silicon abundances in the metal-poor globular cluster NGC 6752 for a sample of more than 130 red giants with homogeneous oxygen and sodium abundances. We find that [Al/Fe] shows a spread of about 1.4 dex among gia
nts in NGC 6752 and is anticorrelated with [Mg/Fe] and [O/Fe] and correlated with [Na/Fe] and [Si/Fe]. These relations are not continuous in nature, but the distribution of stars is clearly clustered around three distinct Al values, low, intermediate, and high. These three groups nicely correspond to the three distinct sequences previously detected using Stromgren photometry along the red giant branch. These two independent findings strongly indicate the existence of three distinct stellar populations in NGC 6752. Comparing the abundances of O and Mg, we find that the population with intermediate chemical abundances cannot originate from material with the same composition of the most O- and Mg-poor population, diluted by material with that of the most O- and Mg-rich one. This calls for different polluters.
Recent progress in studies of globular clusters has shown that they are not simple stellar populations, being rather made of multiple generations. Evidence stems both from photometry and spectroscopy. A new paradigm is then arising for the formation
of massive star clusters, which includes several episodes of star formation. While this provides an explanation for several features of globular clusters, including the second parameter problem, it also opens new perspectives about the relation between globular clusters and the halo of our Galaxy, and by extension of all populations with a high specific frequency of globular clusters, such as, e.g., giant elliptical galaxies. We review progress in this area, focusing on the most recent studies. Several points remain to be properly understood, in particular those concerning the nature of the polluters producing the abundance pattern in the clusters and the typical timescale, the range of cluster masses where this phenomenon is active, and the relation between globular clusters and other satellites of our Galaxy.
To study the crucial range of Galactocentric distances between 12 and 16 kpc, where little information is available, we have obtained VI CCD imaging of Berkeley 20 and BVI CCD imaging of Berkeley 66 and Tombaugh 2, three distant, old open clusters. U
sing the synthetic colour magnitude diagram (CMD) technique with three types of evolutionary tracks of different metallicities, we have determined age, distance, reddening and indicative metallicity of these systems. The CMD of Be 20 is best reproduced by stellar models with a metallicity about half of solar (Z=0.008 or 0.01), in perfect agreement with high resolution spectroscopic estimates. Its age is between 5 and 6 Gyr from stellar models with overshooting and between 4.3 and 4.5 Gyr from models without it. The distance modulus from the best fitting models is always (m-M)0=14.7 (corresponding to a Galactocentric radius of about 16 kpc), and the reddening E(B-V) ranges between 0.13 and 0.16. A slightly lower metallicity (Z ~ 0.006) appears to be more appropriate for Be 66. This cluster is younger, (age of 3 Gyr), and closer, (m-M)0=13.3 (i.e., at 12 kpc from the Galactic centre), than Be 20, and suffers from high extinction, 1.2 < E(B-V) < 1.3, variable at the 2-3 per cent level. Finally, the results for To 2 indicate that it is an intermediate age cluster, with an age of about 1.4 Gyr or 1.6-1.8 Gyr for models without and with overshooting, respectively. The metallicity is about half of solar (Z=0.006 to 0.01), in agreement with spectroscopic determinations. The distance modulus is (m-M)0=14.5, implying a distance of about 14 kpc from the Galactic centre; the reddening E(B-V) is 0.31-0.4, depending on the model and metallicity, with a preferred value around 0.34.
We study the radial distribution of stars of different stellar generations in the globular cluster NGC 3201. From recently published multicolour photometry, a radial dependence of the location of stars on the giant branch was found. We coupled the ph
otometric information to our sample of 100 red giants with Na, O abundances and known classification as first or second-generation stars. We find that giants stars of the second generation in NGC 3201 show a tendency to be more centrally concentrated than stars of the first generation, supporting less robust results from our spectroscopic analysis.
(abridged) Recent spectroscopic and photometric observations show the existence of various generations of stars in GCs, differing in the abundances of products of H-burning at high temperatures (the main final product being He). It is important to st
udy the connections between stars properties and He content. We consider here the about 1400 stars on the Red Giant Branch (RGB) observed with FLAMES@VLT in 19 Galactic GCs, part of out Na-O anticorrelation projet. Stars with different He are expected to have different temperatures (i.e. colours), slightly different [Fe/H], and different luminosity levels of the RGB bump. All these differences are small, but our study has the necessary precision, good statistics, and homogeneity to detect them. We also computed suitable sets of stellar models (BaSTI) for various assumptions about the initial helium content. Differences in observable quantities that can be attributed to variations in He content are generally detectable between stars of the Primordial (P, first-generation) and Extreme (E, second-generation) populations, but not between the Primordial and Intermediate ones (I). The only exception (differences are significant also between P and I populations) is NGC2808, where three populations are clearly separated also on the Main Sequence and the Horizontal Branch. The average enhancement in the He mass fraction Y between P and E stars is about 0.05-0.11, depending on the assumptions. The differences in Y, for NGC2808 alone, are about 0.11-0.14 between P and I stars, and about 0.15-0.19 between P and E stars, again depending on the assumptions. The RGB bump luminosity of first and second-generation stars has different levels; the implied Y difference is more difficult to quantify, but is in agreement with the other determinations.
