Multiple stellar populations in globular clusters

(A. F. Marino)

Being among the oldest objects in the Milky Way, globular clusters (GCs) provide a privileged observation window into the infancy of the Universe. These stellar fossil records are among the objects used in Galactic archeology to uncover the history of the Primordial Universe by observing the Local Universe (e.g. Frebel & Norris 2015, ARA&A, 53, 631).

For many years GCs have served as laboratories for testing the predictions of stellar evolution models since their stars were considered the best example of Simple Stellar Population, a population of monometallic and coeval stars on which estimates of distances and ages are based (Renzini & Buzzoni 1986, 1986, Spectral Evolution of Galaxies, 195). The presence of more than one stellar population in GCs is one of the most fascinating and intriguing discoveries in the field of stellar populations in the recent years.

Nowadays, GCs are known to host more than one stellar population, and the term Multiple Stellar Populations is now used to refer to this phenomenon, replacing the Simple Stellar Population definition. The most powerful tool currently available for the description and characterisation of the multiple population phenomenon is a pseudo-colour photometric diagram dubbed Chromosome Map (ChM) (Milone et al. 2015, ApJ, 808, 51), which efficiently maximises the separation between stars with even slightly different chemical abundances (see Figure 1 which represents the universal chromosome map introduced in Marino et al. 2019, MNRAS, 487, 3815).

Figure 1. By combining all the available GC chromosome maps in a metal-free plane, we get a Universal Map of multiple stellar populations. The universal chromosome map introduced in Marino et al. (2019, MNRAS, 487, 3815) is displayed as a Hess diagram. Given to their different chemical abundances, first (1P) and second population (2P) stars in GCs occupy distinct location on this photometric plane.

As of now, multiple stellar populations in GCs are basically not understood. One of the most surprising features of the multiple populations is the variety. GCs host different number of populations, with the extension and morphology of the chromosome map changing from one cluster to another. In general, we have discovered that GCs can be divided into two main classes: Type I, including ~83% of the GCs in the Milky Way, and Type II clusters. Both classes host the two main groups of 1P and 2P stars, but Type II clusters host more-complex ChMs and internal variations in neutron-capture elements (e.g. La, see Figure 2), Fe, not seen in typical Type I GCs. The existence of a significant fraction of Type II GCs in the Galaxy poses serious doubts on the origin of these objects, that could be relics of much more massive systems, e.g. nuclei of disrupted dwarf galaxies (e.g. Bekki & Norris 2006, ApJ, 637L, 109).

Figure 2. Comparison of the spectra of La lines (top panels) and Eu lines (bottom panels) in two stars with similar atmospheric parameters in the Type II GC M22. The two stars have substantially different derived La abundances, but similar Eu (see Marino et al. 2009, A&A, 505, 1099). The spectrum in red is that of the s-rich star III-3, and the one in blue is that of the s-poor star IV-102.

By using high-resolution spectra, combined with high-precision photometry from the Hubble Space Telescope and Gaia data, we investigate the origin of the enigmatic phenomenon of multiple stellar populations in GCs. These data will allow us to infer chemical abundances, the main properties of all the stellar populations in each GC from the photometric diagrams (e.g. their He content), and their kinematics. All these observational ingredients will be fundamental constraints to answer fundamental questions like:

- Are the observed stellar populations the result of multiple bursts of star formation?

- Could there be alternative phenomena that can account for the stellar populations without invoking different generations of stars?

- Is there a link between GCs with metallicity variations and the dwarf satellite galaxies?