Search

Binary stars play a crucial role in our understanding of the formation and evolution of star clusters and their stellar populations. We use Gaia Data Release 3 to homogeneously analyze 78 Galactic open clusters and the unresolved binary systems they host, each composed of two main sequence (MS) stars. We first investigated the structural parameters of these clusters, such as the core radius and the central density, and determined the cluster mass function (MF) and total mass by interpolating the density profile of each cluster. We measured the fraction of binaries with a large mass ratio and the fraction of blue straggler stars (BSSs), and finally investigated possible connections between the populations of binary stars and BSSs with the main parameters of the host cluster. {Remarkably, we find that the MFs of 78 analyzed open clusters follow a similar trend and are well reproduced by two single power-law functions, with a change in slope around masses of 1$M_{\odot}$. The fraction of binary stars ranges from $\sim$15\% to more than $\sim$60\% without significant correlation with the mass and the age of the host cluster. Moreover, we detect hints of a correlation between the total fraction of binary stars and the central density of the host cluster. We compared the fraction of binary stars with that of BSSs, finding that clusters with high and low central density exhibit different trends. The fraction of binaries does not significantly change with the mass of the primary star and the mass ratio. The radial distribution of binary stars depends on cluster age. The binaries of clusters younger than $\sim$800\,Myr typically show a flat radial distribution, with some hints of a double peak. In contrast, the binaries of the remaining clusters are more centrally concentrated than the single stars, which is similar to what is observed in globular clusters., Comment: 27 pages, 27 figures, accepted for publication in Astronomy & Astrophysics

The assembly history of the Milky Way (MW) is a rapidly evolving subject, with numerous small accretion events and at least one major merger proposed in the MW's history. Accreted alongside these dwarf galaxies are globular clusters (GCs), which act as spatially coherent remnants of these past events. Using high precision differential abundance measurements from our recently published study, we investigate the likelihood that the MW clusters NGC 362 and NGC 288 are galactic siblings, accreted as part of the Gaia-Sausage-Enceladus (GSE) merger. To do this, we compare the two GCs at the 0.01 dex level for 20+ elements for the first time. Strong similarities are found, with the two showing chemical similarity on the same order as those seen between the three LMC GCs, NGC 1786, NGC 2210 and NGC 2257. However, when comparing GC abundances directly to GSE stars, marked differences are observed. NGC 362 shows good agreement with GSE stars in the ratio of Eu to Mg and Si, as well as a clear dominance in the r- compared to the s-process, while NGC 288 exhibits only a slight r-process dominance. When fitting the two GC abundances with a GSE-like galactic chemical evolution model, NGC 362 shows agreement with both the model predictions and GSE abundance ratios (considering Si, Ni, Ba and Eu) at the same metallicity. This is not the case for NGC 288. We propose that the two are either not galactic siblings, or GSE was chemically inhomogeneous enough to birth two similar, but not identical clusters with distinct chemistry relative to constituent stars., Comment: Second paper in a series. Accepted for publication by MNRAS, 17 pages, 11 figures. Minor corrections and updates made after helpful feedback from the community, thank you!

The view of globular clusters (GCs) as simple systems continues to unravel, revealing complex objects hosting multiple chemical peculiarities. Using differential abundance analysis, we probe the chemistry of the Type I GC, NGC 288 and the Type II GC, NGC 362 at the 2\% level for the first time. We measure 20 elements and find differential measurement uncertainties on the order 0.01-0.02 dex in both clusters. The smallest uncertainties are measured for Fe I in both clusters, with an average uncertainty of $\sim$0.013 dex. Dispersion in the abundances of Na, Al, Ti I, Ni, Fe I, Y, Zr, Ba and Nd are recovered in NGC 288, none of which can be explained by a spread in He. This is the first time, to our knowledge, a statistically significant spread in $s$-process elements and a potential spread in metallicity has been detected in NGC 288. In NGC 362, we find significant dispersion in the same elements as NGC 288, with the addition of Co, Cu, Zn, Sr, La, Ce, and Eu. Two distinct groups are recovered in NGC 362, separated by 0.3 dex in average differential $s$-process abundances. Given strong correlations between Al and several $s$-process elements, and a significant correlation between Mg and Si, we propose that the $s$-process rich group is younger. This agrees with asymptotic giant branch star (AGB) enrichment between generations, if there is overlap between low- and intermediate-mass AGBs. In our scenario, the older population is dominated by the $r$-process with a $\Delta^{\mathrm{La}}-\Delta^{\mathrm{Eu}}$ ratio of $-0.16\pm0.06$. We propose that the $r$-process dominance and dispersion found in NGC 362 are primordial., Comment: First paper in a series. Accepted for publication in MNRAS

Driven by recent observational findings, we select massive interactive binaries as the most suitable among the existing candidates for producing the chemical patterns typical of multiple populations of Galactic globular clusters. Still, to avoid supernova contamination we are further driven to endorse the notion that above a critical mass stars fail to produce supernova events, but rather eventually sink into black holes without ejecting much energy and heavy metals. This assumption has the attractive implication of suppressing star formation feedback for some 5--10 million years, in practice leading to runaway star formation, analog to {\it overcooling} that in absence of feedback would have turned most baryons into stars in the early Universe. Under such conditions, multiple episodes of stars formation, incorporating binary star ejecta from previous episodes, appear to be unavoidable, thus accounting for the ubiquity of the multiple population phenomenon in globular clusters., 7 pages. Submitted December 13, 2021, in original form, accepter April 2, 2022. To appear on MNRAS

The study of the "chromosome maps" of Galactic Globular Clusters has shown that the stars identified as `first generation' often define an extended sequence in the $m_{F275W}-m_{F814W}$ colour, whose straightforward interpretation, by comparison with synthetic spectra, is that they are inhomogeneous in helium content. The cluster M 3 (NGC 5272) is one of the most prominent example of this phenomenon, since its first generation is distributed on an extended colour range, formally corresponding to a large helium enhancement ($ \sim 0.1$). It is necessary to ask whether the bulk of photometric observations available for this cluster supports or falsifies this interpretation. For this purpose, we examine the horizontal branch morphology, the period and magnitude distributions of the RR Lyrae variables, and the main sequence colour distribution. Simulating the first generation stars with such internal variation of helium content we can not meet all the observational constraints at the same time, concluding that the origin of the first generation colour spread is still without a straightforward explanation., Comment: 13 pages, 9 figures, 2 tables. Accepted for publication in MNRAS

Poster for the Cool stars 20.5 virtual conference. Abstract: The amount of mass lost by stars during the red-giant branch (RGB) phase is one of the main parameters to correctly study later stages of stellar evolution. In spite of its importance, a fully-comprehensive understanding of this phenomenon is still missing and we mostly rely on empirical formulations. The Galactic Globular Clusters (GCs) are ideal targets to derive such formulations of mass loss, but the presence of multiple populations has been a major challenge. We combine Hubble Space Telescope photometry with stellar evolutionary models, to analyse the horizontal branch (HB) stars in 53 GCs. We constrained the helium abundance for the first and the “extreme” second generations stars using independent measurements based on the RGB. With these new constrains the parameters degeneracy traditionally associated to these stars has been broken for the first time. Our main results are: the mass loss of first generation stars is tightly correlated with cluster metallicity; the location of helium enriched stars on the HB is reproduced only by adopting a higher RGB mass loss, compared to the first generation; and, finally, the difference in mass loss tightly correlates with helium enhancement and cluster mass. &nbsp

Multiple stellar populations (MPs) are a distinct characteristic of Globular Clusters (GCs). Their general properties have been widely studied among main sequence, red giant branch (RGB) and horizontal branch (HB) stars, but a common framework is still missing at later evolutionary stages. We studied the MP phenomenon along the AGB sequences in 58 GCs, observed with the Hubble Space Telescope in ultraviolet (UV) and optical filters. By using UV-optical color-magnitude diagrams, we selected the AGB members of each cluster and identified the AGB candidates of the metal-enhanced population in type II GCs. We studied the photometric properties of AGB stars and compared them to theoretical models derived from synthetic spectra analysis. We observe the following features: i) the spread of AGB stars in photometric indices sensitive to variations of light-elements and helium is typically larger than that expected from photometric errors; ii) the fraction of metal-enhanced stars in the AGB is lower than in the RGB in most of the type II GCs; iii) the fraction of 1G stars derived from the chromosome map of AGB stars in 15 GCs is larger than that of RGB stars; v) the AGB/HB frequency correlates with the average mass of the most helium-enriched population. These findings represent a clear evidence of the presence of MPs along the AGB of Galactic GCs and indicate that a significant fraction of helium-enriched stars, which have lower mass in the HB, does not evolve to the AGB phase, leaving the HB sequence towards higher effective temperatures, as predicted by the AGB-manqu\'e scenario., Comment: 32 pages, 19 figures, 3 tables

The amount of mass lost by stars during the red-giant branch (RGB) phase is one of the main parameters to understand and correctly model the late stages of stellar evolution. Nevertheless, a fully-comprehensive knowledge of the RGB mass loss is still missing. Galactic Globular Clusters (GCs) are ideal targets to derive empirical formulations of mass loss, but the presence of multiple populations with different chemical compositions has been a major challenge to constrain stellar masses and RGB mass losses. Recent work has disentangled the distinct stellar populations along the RGB and the horizontal branch (HB) of 46 GCs, thus providing the possibility to estimate the RGB mass loss of each stellar population. The mass losses inferred for the stellar populations with pristine chemical composition (called first-generation or 1G stars) tightly correlate with cluster metallicity. This finding allows us to derive an empirical RGB mass-loss law for 1G stars. In this paper we investigate seven GCs with no evidence of multiple populations and derive the RGB mass loss by means of high-precision {\it Hubble-Space Telescope} photometry and accurate synthetic photometry. We find a cluster-to-cluster variation in the mass loss ranging from $\sim$0.1 to $\sim$0.3 $M_{\odot}$. The RGB mass loss of simple-population GCs correlates with the metallicity of the host cluster. The discovery that simple-population GCs and 1G stars of multiple population GCs follow similar mass-loss vs. metallicity relations suggests that the resulting mass-loss law is a standard outcome of stellar evolution., Comment: 10 pages, 7 figures, 3 tables, accepted for publication on MNRAS

Neutron-star mergers were recently confirmed as sites of rapid-neutron-capture (r-process) nucleosynthesis. However, in Galactic chemical evolution models, neutron-star mergers alone cannot reproduce the observed element abundance patterns of extremely metal-poor stars, which indicates the existence of other sites of r-process nucleosynthesis. These sites may be investigated by studying the element abundance patterns of chemically primitive stars in the halo of the Milky Way, because these objects retain the nucleosynthetic signatures of the earliest generation of stars. Here we report the element abundance pattern of the extremely metal-poor star SMSS J200322.54-114203.3. We observe a large enhancement in r-process elements, with very low overall metallicity. The element abundance pattern is well matched by the yields of a single 25-solar-mass magnetorotational hypernova. Such a hypernova could produce not only the r-process elements, but also light elements during stellar evolution, and iron-peak elements during explosive nuclear burning. Hypernovae are often associated with long-duration gamma-ray bursts in the nearby Universe. This connection indicates that similar explosions of fast-spinning strongly magnetized stars occurred during the earliest epochs of star formation in our Galaxy., Comment: Author's version of a Letter published in Nature on July 8th, 2021

We exploit multi-band Hubble Space Telescope photometry to investigate multiple populations (MPs) along the red horizontal branches (HBs) and red clumps of fourteen metal-rich Globular Clusters (GCs), including twelve Milky Way GCs and the Magellanic Cloud GCs NGC 1978 and NGC 416. Based on appropriate two-color diagrams we find that the fraction of 1G stars in Galactic GCs correlates with cluster mass, confirming previous results based on red-giant branch (RGB) stars. Magellanic-Cloud GCs show higher fractions of 1G stars than Galactic GCs with similar masses, thus suggesting that the environment affects the MP phenomenon. We compared and combined our population fractions based on HB with previous estimates from MS and RGB, and we used ground-based UBVI photometry (available for NGC 104, NGC 5927, NGC 6366, NGC 6838) to extend the investigation over a wide field of view. All studied GCs are consistent with flat distributions of 1G and 2G stars within ~1 arcmin from the cluster center except NGC 416, where the 2G is more centrally concentrated. 2G stars of NGC 104 and NGC 5927 are more centrally-concentrated than the 1G, whereas the distribution is flat for NGC 6366 and NGC 6838. We discover that most of the analyzed GCs exhibit extended sequences of 1G stars along the red HB, not consistent with a simple population. The comparison between appropriate synthetic and observed CMDs reveals that these extended distributions are consistent with either star-to-star variation in helium or with an internal metallicity spread, recalling the inhomogeneity of 1G stars along the ChMs., 27 pages, 20 figures, 5 tables, accepted for pubblication in ApJ

We present the orbit and properties of 2MASS J050051.85-093054.9, establishing it as the closest (d ~ 71 pc) extremely low mass white dwarf to the Sun. We find that this star is hydrogen-rich with Teff ~ 10 500 K, log g ~ 5.9, and, following evolutionary models, has a mass of ~ 0.17 solar masses. Independent analysis of radial velocity and TESS photometric time series reveals an orbital period of ~ 9.5 h. Its high velocity amplitude (K ~ 144 km/s) produces a measurable Doppler beaming effect in the TESS light curve with an amplitude of 1 mmag. The unseen companion is most likely a faint white dwarf. J0500-0930 belongs to a class of post-common envelope systems that will most likely merge through unstable mass transfer and in specific circumstances lead to Type Ia supernova explosions., Accepted for publication in MNRAS Letters

In the present work we analyzed seven globular clusters selected from their location in the Galactic bulge and with metallicity values in the range $-1.30\lesssim\rm{[Fe/H]}\lesssim-0.50$. The aim of this work is first to derive cluster ages assuming single stellar populations, and secondly, to identify the stars from first (1G) and second generations (2G) from the main sequence, subgiant and red giant branches, and to derive their age differences. Based on a combination of UV and optical filters used in this project, we apply the Gaussian mixture models to distinguish the multiple stellar populations. Applying statistical isochrone fitting, we derive self-consistent ages, distances, metallicities, and reddening values for the sample clusters. An average of $12.3\pm0.4$ Gyr was obtained both using Dartmouth and BaSTI (accounting atomic diffusion effects) isochrones, without a clear distinction between the moderately metal-poor and the more metal-rich bulge clusters, except for NGC 6717 and the inner halo NGC 6362 with $\sim 13.5$ Gyr. We derived a weighted mean age difference between the multiple populations hosted by each globular cluster of $41\pm170$ Myr adopting canonical He abundances; whereas for higher He in 2G stars, this difference reduces to $17\pm170$ Myr, but with individual uncertainties of $500$ Myr., 25 pages, 16 figures and 7 tables. Accepted for publication in ApJ

The metal-poor stars in the bulge are important relics of the Milky Way's formation history, as simulations predict that they are some of the oldest stars in the Galaxy. In order to determine if they are truly ancient stars, we must understand the origins of this population. Currently, it is unclear if the metal-poor stars in the bulge ([Fe/H] < -1 dex) are merely halo interlopers, a unique accreted population, part of the boxy/peanut-shaped (B/P) bulge or a classical bulge population. In this work, we use spectra from the VLT/FLAMES spectrograph to obtain metallicity estimates using the Ca-II triplet (CaT) of 473 bulge stars (187 of which have [Fe/H] 50\% probability of being bound to the bulge, and half are halo interlopers. We also see that the occurrence rate of halo interlopers increases steadily with decreasing metallicity across the full range of our sample (-3 < [Fe/H] < 0.5). Our examination of the kinematics of the confined compared to the unbound stars indicates the metal-poor bulge ([Fe/H] < -1 dex) comprises at least two populations; those confined to the boxy/peanut bulge and halo stars passing through the inner galaxy. We conclude that an orbital analysis approach, as we have employed, is important to uncover and understand the composite nature of the metal-poor stars in the inner region., Comment: 18 pages, 12 figures, accepted to MNRAS, comments welcome

The series of events, which occurred at high redshift and originated multiple stellar populations in Globular Clusters (GCs) are still poorly understood. Theoretical work suggests that the present-day dynamics of stars in nearby GCs, including the rotation and velocity dispersion, may retain important clues on the formation of multiple populations. So far, the dynamics of multiple populations have been investigated either from radial velocities of a relatively-small sample of stars, or from relative proper motions of stars in the small field of view provided by the Hubble Space Telescope. In this context, Gaia provides the unique opportunity to investigate the dynamics of thousands GC stars over a wide field of view. For the first time, we combine Gaia DR2 proper motions and multi-band photometry to study the internal motions of the two main stellar populations of 47 Tucanae in a wide field of view. We confirm that this cluster exhibits high rotation on the plane of the sky and find that both stellar generations share similar rotation patters. Second-generation stars show stronger anisotropies and smaller tangential-velocity dispersion than the first generation, while there is no significant difference between their radial-velocity dispersion profiles. We discuss the impact of these results in the context of the formation scenarios for multiple stellar populations in GCs., Comment: 7 pages, 5 figures, accepted for publication in MNRAS

As part of a large investigation with Hubble Space Telescope to study the faintest stars within the globular cluster Omega Centauri, in this work we present early results on the multiplicity of its main sequence (MS) stars, based on deep optical and near-infrared observations. By using appropriate color-magnitude diagrams we have identified, for the first time, the two main stellar populations I, and II along the entire MS, from the turn-off towards the hydrogen-burning limit. We have compared the observations with suitable synthetic spectra of MS stars and conclude that the two MSs are consistent with stellar populations with different metallicity, helium, and light-element abundance. Specifically, MS-I corresponds to a metal-poor stellar population ([Fe/H]~-1.7) with Y~ 0.25 and [O/Fe]~0.30. The MS-II hosts helium-rich (Y~0.37-0.40) stars with metallicity ranging from [Fe/H]~-1.7 to -1.4. Below the MS knee (mF160W~19.5, our photometry reveals that each of the two main MSs hosts stellar subpopulations with different oxygen abundances, with very O-poor stars ([O/Fe]~-0.5) populating the MS-II. Such a complexity has never been observed in previous studies of M-dwarfs in globular clusters. A few months before the lunch of the James Webb Space Telescope, these results demonstrate the power of optical and near-infrared photometry in the study of multiple stellar populations in globular clusters., 13 pages, 9 figures, accepted for publication in MNRAS

I present the latest results from our group about the multiple stellar populations in the old Milky Way globular clusters (GCs) and in the young systems both in the Magellanic Clouds and in the Milky Way. For the ancient GCs in our Galaxy I summarize the chemical properties of the stellar populations as observed on the chromosome map. Both Type I and Type II GCs are discussed. For the youngest clusters I will briefly report our latest spectroscopic analysis on the Large Magellanic Cloud cluster NGC1818 and the Galactic open cluster NGC6705 (M11), which supports the co-existence of stellar populations with different rotation rates., 6 pages, 4 figures, to appear on IAU Proceedings Series "Star Clusters: From the Milky Way to the Early Universe"

As part of a large Hubble Space Telescope investigation aiming at reaching the faintest stars in the Galactic globular cluster NGC 6752, an ACS/WFC field was the subject of deep optical observations reaching magnitudes as faint as V~30. In this field we report the discovery of Bedin I a dwarf spheroidal galaxy too faint and too close to the core of NGC 6752 for detection in earlier surveys. As it is of broad interest to complete the census of galaxies in the local Universe, in this Letter we provide the position of this new object along with preliminary assessments of its main param eters. Assuming the same reddening as for NGC 6752, we estimate a distance modulus of (m-M)_0 = 29.70+/-0.13 from the observed red giant branch, i.e., 8.7 (+0.5 -0.7) Mpc, and size of ~840x340pc, about 1/5 the size of the LMC. A comparison of the observed colour-magnitude diagram with synthetic counterparts that account for the galaxy distance modulus, reddening and photometric errors, suggests the presence of an old (~13Gyr) and metal poor ([Fe/H]~-1.3) population. This object is most likely a relatively isolated satellite dwarf spheroidal galaxy of the nearby great spiral NGC 6744, or potentially the most distant isolated dwarf spheroidal known with a secure distance., Comment: 5 pages, 1 table, 3 figures (2 in low resolution). Published on MNRAS Letters and freely available https://academic.oup.com/mnrasl/article/484/1/L54/5288002 Related press releases: https://media.stsci.edu/news_release/news/2019-09 and https://www.spacetelescope.org/news/heic1903/ , (and only in italian: https://www.youtube.com/watch?v=VZCop_j3KOo )

We present and discuss the results of a search for extremely metal-poor stars based on photometry from data release DR1.1 of the SkyMapper imaging survey of the southern sky. In particular, we outline our photometric selection procedures and describe the low-resolution ($R$ $\approx$ 3000) spectroscopic follow-up observations that are used to provide estimates of effective temperature, surface gravity and metallicity ([Fe/H]) for the candidates. The selection process is very efficient: of the 2618 candidates with low-resolution spectra that have photometric metallicity estimates less than or equal to -2.0, 41% have [Fe/H] $\leq$ -2.75 and only $\sim$7% have [Fe/H] $>$ -2.0 dex. The most metal-poor candidate in the sample has [Fe/H] $$ 7.3 (predominantly CEMP-$s$) while any bias against stars with A(C)$_{1D, LTE}$ $$ +1 (predominantly CEMP-no) is not readily quantifiable given the uncertainty in the SkyMapper $v$-band DR1.1 photometry. We find that the metallicity distribution function of the observed sample has a power-law slope of $\Delta$(Log N)/$\Delta$[Fe/H] = 1.5 $\pm$ 0.1 dex per dex for -4.0 $\leq$ [Fe/H] $\leq$ -2.75, but appears to drop abruptly at [Fe/H] $\approx$ -4.2, in line with previous studies., Comment: 21 pages, 19 figures, accepted for publication in MNRAS

The internal dynamics of multiple stellar populations in Globular Clusters (GCs) provides unique constraints on the physical processes responsible for their formation. Specifically, the present-day kinematics of cluster stars, such as rotation and velocity dispersion, seems to be related to the initial configuration of the system. In recent work (Milone et al. 2018), we analyzed for the first time the kinematics of the different stellar populations in NGC0104 (47Tucanae) over a large field of view, exploiting the Gaia Data Release 2 proper motions combined with multi-band ground-based photometry. In this paper, based on the work by Cordoni et al. (2019), we extend this analysis to six GCs, namely NGC0288, NGC5904 (M5), NGC6121 (M4), NGC6752, NGC6838 (M71) and further explore NGC0104. Among the analyzed clusters only NGC0104 and NGC5904 show significant rotation on the plane of the sky. Interestingly, multiple stellar populations in NGC5904 exhibit different rotation curves., Comment: 4 pages, 2 figures, proceeding of IAU symposium 351

Studies based on Hubble Space Telescope ($HST$) photometry in F275W, F336W, and F438W have shown that the incidence and complexity of multiple populations (MPs) in Globular Clusters (GCs) depend on cluster mass. This result, however, is based on nearby, low-reddening Galactic GCs, for which we can obtain accurate F275W photometry. In this work we exploit $HST$ photometry in F336W, F438W, and F814W to characterize MPs in 68 Galactic and extragalactic GCs by using the pseudo-color $C_{F336W,F438W,F814W}$, which is more efficient in terms of observation time than the pseudo-color $C_{F275W,F336W,F438W}$ adopted in previous works. We first analyzed the Galactic GCs alone and find that the RGB color width strongly correlates with [Fe/H]. After removing the dependence from metallicity, we obtain a strong correlation with cluster mass, thus confirming previous findings. We then exploited the RGB width to compare MPs in Galactic and extragalactic GCs. Similarly to Galactic GCs, the RGB width of Magellanic Cloud GCs correlates with clusters mass, after removing the dependence from metallicity. This fact demonstrates that cluster mass is the main factor affecting the properties of MPs. Magellanic Cloud clusters exhibit, on average, narrower RGBs than Galactic GCs with similar mass and metallicity. We suggest that this difference is a signature of stellar mass loss in Galactic GCs. As an alternative, the mass-RGB-width relation would depend on the host galaxy. Finally, we used ground-based photometry to investigate Terzan\,7 through the $C_{U,B,I}$ pseudo-color, ground-based analogous of $C_{F336W,F438W,F814W}$, and find that this cluster is consistent with a simple population., 22 pages, 10 figures, accepted for publication on The Astronomical Journal

The Hubble Space Telescope (HST) UV Legacy Survey of Galactic Globular Clusters (GCs) has investigated multiple stellar populations by means of the "chromosome map" (ChM) diagnostic tool that maximises the separation between stars with different chemical composition. One of the most challenging features revealed by ChMs analysis is the apparent inhomogeneity among stars belonging to the first population, a phenomenon largely attributed to He variations. However, this explanation is not supported by the uniformity in p-capture elements of these stars. The HST survey has revealed that the GC NGC 3201 shows an exceptionally wide coverage in the Delta(F275W,F814W) parameter of the ChM. We present a chemical abundance analysis of 24 elements in 18 giants belonging to the first population of this GC, and having a wide range in Delta(F275W,F814W). As far as the p-capture elements are concerned, the chemical abundances are typical of 1G stars, as expected from the location of our targets in the ChM. Based on radial velocities and chemical abundances arguments, we find that the three stars with the lowest Delta(F275W,F814W) values are binary candidates. This suggests that, at least those stars could be explained with binarity. These results are consistent with evidence inferred from multi-band photometry that evolved blue stragglers populate the bluest part of the 1G sequence in the ChM. The remaining 15 spectroscopic targets show a small range in the overall metallicity by ~0.10 dex, with stars at higher Delta(F275W,F814W) values having higher absolute abundances. We suggest that a small variation in metals and binarity govern the color spread of the 1G in the ChM, and that evolved blue stragglers contribute to the bluest tail of the 1G sequence., 24 pages, 12 figures, 8 tables. Accepted for publication in ApJ

Historically, multiple populations in Globular Clusters (GCs) have been mostly studied from ultraviolet and optical filters down to stars that are more massive than ~0.6 solar masses. Here we exploit deep near-infrared (NIR) photometry from the Hubble Space Telescope to investigate multiple populations among M-dwarfs in the GC NGC6752. We discovered that the three main populations (A, B and C), previously observed in the brightest part of the color-magnitude diagram, define three distinct sequences that run from the main-sequence (MS) knee towards the bottom of the MS (~0.15 solar masses). These results, together with similar findings on NGC2808, M4, and omega Centauri, demonstrate that multiple sequences of M-dwarfs are common features of the color-magnitude diagrams of GCs. The three sequences of low-mass stars in NGC6752 are consistent with stellar populations with different oxygen abundances. The range of [O/Fe] needed to reproduce the NIR CMD of NGC6752 is similar to the oxygen spread inferred from high-resolution spectroscopy of red-giant branch (RGB) stars. The relative numbers of stars in the three populations of M-dwarfs are similar to those derived among RGB and MS stars more massive than ~0.6 solar masses. As a consequence, the evidence that the properties of multiple populations do not depend on stellar mass is a constraint for the formation scenarios., 7 pages, 5 figures, accepted for publication in MNRAS

Chemistry and kinematic studies can determine the origins of stellar population across the Milky Way. The metallicity distribution function of the bulge indicates that it comprises multiple populations, the more metal-poor end of which is particularly poorly understood. It is currently unknown if metal-poor bulge stars ([Fe/H] $, Comment: 19 pages, 5 tables, accepted to MNRAS

The recent measurements of internal variations of helium in Galactic and extragalactic Globular Clusters (GCs) set binding constraints to the models of formation of Multiple Populations (MPs) in GCs, and gave rise, at the same time, to crucial questions related with the influence of the environment on MP formation as well as with the role played by GCs in the early galactic formation. We present the most recent estimates of helium enrichment in the main populations of a large sample of Galactic and extragalactic GCs., Comment: 4 pages, 4 figures; to appear on International Astronomical Union Proceedings Series "Star Clusters: from the Milky Way to the Early Universe"

Recent work revealed that both the helium variation within globular clusters (GCs) and the relative numbers of first and second-generation stars (1G, 2G) depend on the mass of the host cluster. Precise determination of the internal helium variations and of the fraction of 1G stars are crucial constraints to the formation scenarios of multiple populations (MPs). We exploit multi-band Hubble Space Telescope photometry to investigate MPs in NGC 2419, which is one of the most-massive and distant GCs of the Galaxy, almost isolated from its tidal influence. We find that the 1G hosts the ~37% of the analyzed stars, and identified three populations of 2G stars, namely 2GA, 2GB, and 2GC, which comprise the ~20%, ~31% and ~12% of stars, respectively. We compare the observed colors of these four populations with the colors derived from appropriate synthetic spectra to infer the relative helium abundances. We find that 2GA, 2GB, and 2GC stars are enhanced in helium mass fraction by deltaY ~0.01, 0.06, and 0.19 with respectto 1G stars that have primordial helium (Y=0.246). The high He enrichment of 2GC stars is hardly reconcilable with most of the current scenarios for MPs. Furthermore, the relatively larger fraction of 1G stars (~37%) compared to other massive GCs is noticeable. By exploiting literature results, we find that the fractions of 1G stars of GCs with large perigalactic distance are typically higher than in the other GCs with similar masses. This suggests that NGC 2419, similarly to other distant GCs, lost a lower fraction of 1G stars., Comment: 10 pages, 8 figures, submitted to MNRAS January 22nd

The internal dynamics of multiple stellar populations in Globular Clusters (GCs) provides unique constraints on the physical processes responsible for their formation. Specifically, the present-day kinematics of cluster stars, such as rotation and velocity-dispersion, could be related to the initial configuration of the system. In recent work, we provided the first study of the kinematics of different stellar populations in NGC$\,$0104 over a large field of view in the plane of the sky, exploiting Gaia Data Release 2 (DR2) proper motions combined with multi-band ground-based photometry. In this paper, we combine Gaia DR2 proper motions with Very Large Telescope radial velocities to investigate the kinematics along the line of sight and in the plane of the sky of multiple populations in seven GCs, namely NGC$\,$0104, NGC$\,$0288, NGC$\,$5904, NGC$\,$6121, NGC$\,$6254, NGC$\,$6752 and NGC$\,$6838. Among the analyzed clusters only NGC$\,$0104 and NGC$\,$5904 show significant rotation. Separating our sample into two groups of first- and second-population stars (1P and 2P) we find that overall these two populations exhibit a similar rotation pattern in NGC$\,$0104. However, some hints of different rotation are observed in the external regions of this cluster. Interestingly, 1P and 2P stars in NGC$\,$5904 show different rotation curves, with distinct phases and such difference is significant at the $\sim$2.5-$\sigma$ level. The analysis of the velocity-dispersion profiles of multiple populations confirms that 2P stars of NGC$\,$0104 show stronger anisotropy than the 1P., Comment: Accepted for publication in ApJ, 22 pages, 15 figures, 3 tables

In this paper we present the astro-photometric catalogues of 56 globular clusters and one open cluster. Astrometry and photometry are mainly based on images collected within the "HST Legacy Survey of Galactic Globular Clusters: Shedding UV Light on Their Populations and Formation" (GO-13297, PI:~Piotto), and the "ACS Survey of Galactic Globular Clusters" (GO-10775, PI:~Sarajedini). For each source in the catalogues for which we have reliable proper motion we also publish a membership probability for separation of field and cluster stars. These new catalogues, which we make public in Mikulski Archive for Space Telescopes, replace previous catalogues by Paper VIII of this series., Comment: 14 pages, 5 figures, 4 tables. Accepted for publication in MNRAS on September 11, 2018. Astro-photometric catalogues, stacked and RGB images of the clusters are publicly available at MAST as High Level Science Product at https://archive.stsci.edu/prepds/hugs/

In the context of the Hubble Space Telescope UV Survey of Galactic Globular Clusters (GCs), we derived high-precision, multi-band photometry to investigate the multiple stellar populations in the massive and metal-poor GC M 15. By creating for red-giant branch (RGB) stars of the cluster a 'chromosome map', which is a pseudo two-colour diagram made with appropriate combination of F275W, F336W, F438W, and F814W magnitudes, we revealed colour spreads around two of the three already known stellar populations. These spreads cannot be produced by photometric errors alone and could hide the existence of (two) additional populations. This discovery increases the complexity of the multiple-population phenomenon in M 15. Our analysis shows that M 15 exhibits a faint sub-giant branch (SGB), which is also detected in colour-magnitude diagrams (CMDs) made with optical magnitudes only. This poorly-populated SGB includes about 5% of the total number of SGB stars and evolves into a red RGB in the m_F336W vs. m_F336W-m_F814W CMD, suggesting that M 15 belongs to the class of Type II GCs. We measured the relative number of stars in each population at various radial distances from the cluster centre, showing that all of these populations share the same radial distribution within statistic uncertainties. These new findings are discussed in the context of the formation and evolution scenarios of the multiple populations., 18 pages, 15 figures (4 at low resolution), 2 tables. Accepted for publication in MNRAS on March 13, 2018

STScI grants [GO-14118, 14662]; Chilean BASAL Centro de Excelencia en Astrofisica y Tecnologias Afines (CATA) grant [PFB-06/2007]; Centre national d'etudes spatiale (CNES) [0101973]; UTINAM Institute of the Universite de Franche-Comte - Region de Franche-Comte and Institut des Sciences de l'Univers (INSU); European Research Council through the ERC-StG project [716082]; NASA [NAS 5-26555]

Recent work, based on data from the Hubble Space Telescope (HST) UV Legacy Survey of Galactic Globular Clusters (GCs), has revealed that all the analyzed clusters host two groups of first- (1G) and second-generation (2G) stars. In most GCs, both 1G and 2G stars host sub-stellar populations with different chemical composition. We compare multi-wavelength HST photometry with synthetic spectra to determine for the first time the average helium difference between the 2G and 1G stars in a large sample of 57 GCs and the maximum helium variation within each of them. We find that in all clusters 2G stars are consistent with being enhanced in helium with respect to 1G. The maximum helium variation ranges from less than 0.01 to more than 0.10 in helium mass fraction and correlates with both the cluster mass and the color extension of the horizontal branch (HB). These findings demonstrate that the internal helium variation is one of the main (second) parameters governing the HB morphology., Comment: 26 pages, 19 figures, accepted for publications in MNRAS

The Hubble Space Telescope photometric survey of Galactic globular clusters (GCs) has revealed a peculiar "chromosome map" for NGC6934. Besides a typical sequence, similar to that observed in Type I GCs, NGC 6934 displays additional stars on the red side, analogous to the anomalous, Type II GCs, as defined in our previous work. We present a chemical abundance analysis of four red giants in this GC. Two stars are located on the chromosome map sequence common to all GCs, and another two on the additional sequence. We find: (i) star-to-star Fe variations, with the two anomalous stars being enriched by ~0.2 dex. Due to our small-size sample, this difference is at the ~2.5 sigma level; (ii) no evidence for variations in the slow neutron-capture abundances over Fe, at odds with what is often observed in anomalous Type II GCs, e.g. M 22 and Omega Centauri; (iii) no large variations in light elements C, O and Na, compatible with the targets location on the lower part of the chromosome map where such variations are not expected. Since the analyzed stars are homogeneous in light elements, the only way to reproduce the photometric splits on the sub-giant (SGB) and the red-giant (RGB) branches is to assume that red-RGB/faint-SGB stars are enhanced in [Fe/H] by ~0.2. This fact corroborates the spectroscopic evidence of a metallicity variation in NGC6934. The observed chemical pattern resembles only partially the other Type II GCs, suggesting that NGC6934 might belong either to a third class of GCs, or be a link between normal Type I and anomalous Type II GCs., Comment: 26 pages, 9 figures, accepted for publication in ApJ

By far, the color-magnitude diagrams (CMDs) of Galactic open clusters have been considered proto-types of single stellar populations. By using photometry in ultraviolet and optical bands we discovered that the nearby young cluster NGC6705 (M11) exhibits an extended main-sequence turn off (eMSTO) and a broadened main-sequence (MS). This is the first evidence of multiple stellar populations in a Galactic open cluster. By using high-resolution VLT spectra we provide direct evidence that the multiple sequences along the CMD correspond to stellar populations with different rotation rates. Specifically, the blue MS is formed of slow-rotating stars, while red-MS stars are fast rotators. Moreover, we exploit photometry from Gaia DR2 to show that three Galactic open clusters, namely NGC2099, NGC2360, and NGC2818, exhibit the eMSTO, thus suggesting that it is a common feature among these objects. Our previous work on the Large Magellanic Cloud star cluster NGC1818 shows that slowly and fastly-rotating stars populate the blue and red MS observed in its CMD. The similarities between M11 and the young clusters of the Magellanic Clouds suggest that rotation is responsible for the appearance of multiple populations in the CMDs of both Milky Way open clusters and Magellanic Clouds young clusters., Comment: 9 pages, 4 figures, Accepted

We present a spectroscopic analysis of main sequence (MS) stars in the young globular cluster NGC1818 (age~40 Myrs) in the Large Magellanic Cloud. Our photometric survey on Magellanic Clouds clusters has revealed that NGC1818, similarly to the other young objects with age 600 Myrs, displays not only an extended MS Turn-Off (eMSTO), as observed in intermediate-age clusters (age~1-2 Gyrs), but also a split MS. The most straightforward interpretation of the double MS is the presence of two stellar populations: a sequence of slowly-rotating stars lying on the blue-MS and a sequence of fast rotators, with rotation close to the breaking speed, defining a red-MS. We report the first direct spectroscopic measurements of projected rotational velocities vsini for the double MS, eMSTO and Be stars of a young cluster. The analysis of line profiles includes non-LTE effects, required for correctly deriving v sini values. Our results suggest that: (i) the mean rotation for blue- and red-MS stars is vsini=71\pm10 km/s (sigma=37 km/s) and vsini=202\pm23 km/s (sigma=91 km/s), respectively; (ii) eMSTO stars have different vsini, which are generally lower than those inferred for red-MS stars, and (iii) as expected, Be stars display the highest vsini values. This analyis supports the idea that distinct rotational velocities play an important role in the appearence of multiple stellar populations in the color-magnitude diagrams of young clusters, and poses new constraints to the current scenarios., Comment: 16 pages, 1 table, 9 figures. Accepted for publication in AJ (11/07/2018)

The Hubble Space Telescope UV Legacy Survey of Galactic Globular Clusters is providing a major breakthrough in our knowledge of Globular Clusters (GCs) and their stellar populations. Among the main results, we discovered that all the studied GCs host two main discrete groups consisting of first generation (1G) and second generation (2G) stars. We exploit the multiwavelength photometry from this project to investigate, for the first time, the Red Giant Branch Bump (RGBB) of the two generations in a large sample of GCs. We identified, with high statistical significance, the RGBB of 1G and 2G stars in 26 GCs and found that their magnitude separation as a function of the filter wavelength follows comparable trends. The comparison of observations to synthetic spectra reveals that the RGBB luminosity depends on the stellar chemical composition and that the 2G RGBB is consistent with stars enhanced in He and N and depleted in C and O with respect to 1G stars. For metal-poor GCs the 1G and 2G RGBB relative luminosity in optical bands mostly depends on helium content, Y. We used the RGBB observations in F606W and F814W bands to infer the relative helium abundance of 1G and 2G stars in 18 GCs, finding an average helium enhancement $\Delta$Y=0.011$\pm$0.002 of 2G stars with respect to 1G stars. This is the first determination of the average difference in helium abundance of multiple populations in a large number of clusters and provides a lower limit to the maximum internal variation of helium in GCs., Comment: 16 pages, 11 figures, 2 tables. Accepted for publication in MNRAS

NASA [NAS 5-26555]; STScI [GO-14118]; European Research Council through ERC-StG [716082]; Space Telescope Science Institute under U.S. Government [NAG W-2166]; U.S. Government grant [NAG W-2166]

The split main sequences (MSs) and extended MS turnoffs (eMSTOs) detected in a few young clusters have demonstrated that these stellar systems host multiple populations differing in a number of properties such as rotation and, possibly, age.We analyze Hubble Space Telescope photometry for thirteen clusters with ages between ~40 and ~1000 Myrs and of different masses. Our goal is to investigate for the first time the occurrence of multiple populations in a large sample of young clusters. We find that all the clusters exhibit the eMSTO phenomenon and that MS stars more massive than ~1.6 solar masses define a blue and red MS, with the latter hosting the majority of MS stars. The comparison between the observations and isochrones suggests that the blue MSs are made of slow-rotating stars, while the red MSs host stars with rotational velocities close to the breakup value. About half of the bright MS stars in the youngest clusters are H-alpha emitters. These Be stars populate the red MS and the reddest part of the eMSTO thus supporting the idea that the red MS is made of fast rotators. We conclude that the split MS and the eMSTO are a common feature of young clusters in both Magellanic Clouds. The phenomena of a split MS and an eMSTO occur for stars that are more massive than a specific threshold which is independent of the host-cluster mass. As a by-product, we report the serendipitous discovery of a young SMC cluster, GALFOR1., Comment: 22 pages, 20 figures, accepted for publication in MNRAS

Multiple stellar populations in the Milky Way globular clusters manifest themselves with a large variety. Although chemical abundance variations in light elements, including He, are ubiquitous, the amount of these variations is different in different globulars. Stellar populations with distinct Fe, C+N+O and slow-neutron capture elements have been now detected in some globular clusters, whose number will likely increase. All these chemical features correspond to specific photometric patterns. I review the chemical+photometric features of the multiple stellar populations in globular clusters and discuss how the interpretation of data is being more and more challenging. Very excitingly, the origin and evolution of globular clusters is being a complex puzzle to compose.

We present a high resolution spectroscopic analysis of 62 red giants in the Milky Way globular cluster NGC5286. We have determined abundances of representative light proton-capture, alpha, Fe-peak and neutron-capture element groups, and combined them with photometry of multiple sequences observed along the colour-magnitude diagram. Our principal results are: (i) a broad, bimodal distribution in s-process element abundance ratios, with two main groups, the s-poor and s-rich groups; (ii) substantial star-to-star Fe variations, with the s-rich stars having higher Fe, e.g. _s-rich - _s-poor ~ 0.2~dex; and (iii) the presence of O-Na-Al (anti-)correlations in both stellar groups. We have defined a new photometric index, c_{BVI}=(B-V)-(V-I), to maximise the separation in the colour-magnitude diagram between the two stellar groups with different Fe and s-element content, and this index is not significantly affected by variations in light elements (such as the O-Na anticorrelation). The variations in the overall metallicity present in NGC5286 add this object to the class of "anomalous" GCs. Furthermore, the chemical abundance pattern of NGC5286 resembles that observed in some of the anomalous GCs, e.g. M22, NGC1851, M2, and the more extreme Omega Centauri, that also show internal variations in s-elements, and in light elements within stars with different Fe and s-elements content. In view of the common variations in s-elements, we propose the term s-Fe-anomalous GCs to describe this sub-class of objects. The similarities in chemical abundance ratios between these objects strongly suggest similar formation and evolution histories, possibly associated with an origin in tidally disrupted dwarf satellites., Comment: 28 pages, 21 figures, accepted for publication in MNRAS

We present a multi-year survey aimed at collecting (1) high-precision (~5 milli-mag), (2) fast-cadence (~3 min), and (3) relatively long duration (~10 days) multi-band photometric series. The goal of the survey is to discover and characterize efficiently variable objects and exoplanetary transits in four fields containing five nearby open clusters spanning a broad range of ages. More in detail, our project will (1) constitute a preparatory survey for HARPS-N@TNG, which will be used for spectroscopic follow-up of any target of interest that this survey discovers or characterizes, (2) measure rotational periods and estimate the activity level of targets we are already monitoring with HARPS and HARPS-N for exoplanet transit search, and (3) long term characterization of selected targets of interest in open clusters within the planned K2 fields. In this first paper we give an overview of the project, and report on the variability of objects within the first of our selected fields, which contains two open clusters: M 35 and NGC 2158. We detect 519 variable objects, 273 of which are new discoveries, while the periods of most of the previously known variables are considerably improved., 12 pages, 11 figures (4 at low resolution), 2 tables. Accepted for publication in MNRAS on December 17, 2014. Electronic materials available at the url http://groups.dfa.unipd.it/ESPG/aphn.html , and later on the Journal and at the CDS

We present high-precision multi-band photometry for the globular cluster (GC) M2. We combine the analysis of the photometric data obtained from the Hubble Space Telescope UV Legacy Survey of Galactic GCs GO-13297, with chemical abundances by Yong et al.(2014), and compare the photometry with models in order to analyze the multiple stellar sequences we identified in the color-magnitude diagram (CMD). We find three main stellar components, composed of metal-poor, metal-intermediate, and metal-rich stars (hereafter referred to as population A, B, and C, respectively). The components A and B include stars with different $s$-process element abundances. They host six sub-populations with different light-element abundances, and exhibit an internal variation in helium up to Delta Y~0.07 dex. In contrast with M22, another cluster characterized by the presence of populations with different metallicities, M2 contains a third stellar component, C, which shows neither evidence for sub-populations nor an internal spread in light-elements. Population C does not exhibit the typical photometric signatures that are associated with abundance variations of light elements produced by hydrogen burning at hot temperatures. We compare M2 with other GCs with intrinsic heavy-element variations and conclude that M2 resembles M22, but it includes an additional stellar component that makes it more similar to the central region of the Sagittarius galaxy, which hosts a GC (M54) and the nucleus of the Sagittarius galaxy itself., 12 pages, 7 figures, accepted for publication in MNRAS

We use high-precision photometry of red-giant-branch (RGB) stars in 57 Galactic globular clusters (GCs), mostly from the `Hubble Space Telescope (HST) UV Legacy Survey of Galactic globular clusters', to identify and characterize their multiple stellar populations. For each cluster the pseudo two-color diagram (or `chromosome map') is presented, built with a suitable combination of stellar magnitudes in the F275W, F336W, F438W and F814W filters that maximizes the separation between multiple populations. In the chromosome map of most GCs (Type I clusters), stars separate in two distinct groups that we identify with the first (1G) and the second generation (2G). This identification is further supported by noticing that 1G stars have primordial (oxygen-rich, sodium-poor) chemical composition, whereas 2G stars are enhanced in sodium and depleted in oxygen. This 1G-2G separation is not possible for a few GCs where the two sequences have apparently merged into an extended, continuous sequence. In some GCs (Type II clusters) the 1G and/or the 2G sequences appear to be split, hence displaying more complex chromosome maps. These clusters exhibit multiple SGBs also in purely optical color-magnitude diagrams, with the fainter SGB joining into a red RGB which is populated by stars with enhanced heavy-element abundance. We measure the RGB width by using appropriate colors and pseudo-colors. When the metallicity dependence is removed, the RGB width correlates with the cluster mass. The fraction of 1G stars ranges from ~8% to ~67% and anticorrelates with the cluster mass, indicating that incidence and complexity of the multiple population phenomenon both increase with cluster mass., Comment: 21 pages, 24 figures, accepted for publication in MNRAS

We take advantage of the exquisite quality of the Hubble Space Telescope 26-filter astro-photometric catalog of the core of Omega Cen presented in the first paper of this series and the empirical differential-reddening correction presented in the second paper in order to distill the main sequence into its constituent populations. To this end, we restrict ourselves to the five most useful filters: the magic "trio" of F275W, F336W, and F438W, along with F606W and F814W. We develop a strategy for identifying color systems where different populations stand out most distinctly, then we isolate those populations and examine them in other filters where their sub-populations also come to light. In this way, we have identified at least 15 sub-populations, each of which has a distinctive fiducial curve through our 5-dimensional photometric space. We confirm the MSa to be split into two subcomponents, and find that both the bMS and the rMS are split into three subcomponents. Moreover, we have discovered two additional MS groups: the MSd (which has three subcomponents) shares similar properties with the bMS, and the MSe (which has four subcomponents), has properties more similar to those of the rMS. We examine the fiducial curves together and use synthetic spectra to infer relative heavy-element, light-element, and Helium abundances for the populations. Our findings show that the stellar populations and star formation history of Omega Cen are even more complex than inferred previously. Finally, we provide as a supplement to the original catalog a list that identifies for each star which population it most likely is associated with., Comment: 22 pages, 17 figures (most in lower res), 2 tables, accepted for publication in ApJ

High-resolution spectroscopic observations were taken of 29 extended main sequence turn-off (eMSTO) stars in the young ($\sim$200 Myr) LMC cluster, NGC 1866 using the Michigan/Magellan Fiber System and MSpec spectrograph on the Magellan-Clay 6.5-m telescope. These spectra reveal the first direct detection of rapidly rotating stars whose presence has only been inferred from photometric studies. The eMSTO stars exhibit H-alpha emission (indicative of Be-star decretion disks), others have shallow broad H-alpha absorption (consistent with rotation $\gtrsim $150 km s$^{-1}$), or deep H-alpha core absorption signaling lower rotation velocities ($ \lesssim $150 km s$^{-1}$ ). The spectra appear consistent with two populations of stars - one rapidly rotating, and the other, younger and slowly rotating., Comment: 9 pages, 4 figures, Accepted for publication in ApJ Letters

We present CN and CH indices and CaII triplet metallicities for 34 giant stars and chemical abundances for 33 elements in 14 giants in the globular cluster M2. Assuming the program stars are cluster members, our analysis reveals (i) an extreme variation in CN and CH line strengths, (ii) a metallicity dispersion with a dominant peak at [Fe/H] = -1.7 and smaller peaks at -1.5 and -1.0, (iii) star-to-star abundance variations and correlations for the light elements O, Na, Al and Si and (iv) a large (and possibly bimodal) distribution in the abundances of all elements produced mainly via the s-process in solar system material. Following Roederer et al. (2011), we define two groups of stars, "r+s" and "r-only", and subtract the average abundances of the latter from the former group to obtain a "s-process residual". This s-process residual is remarkably similar to that found in M22 and in M4 despite the range in metallicity covered by these three systems. With recent studies identifying a double subgiant branch in M2 and a dispersion in Sr and Ba abundances, our spectroscopic analysis confirms that this globular cluster has experienced a complex formation history with similarities to M22, NGC 1851 and omega Centauri., Comment: Accepted for publication in MNRAS