Abstract
Multiple stellar populations are observed in almost all globular-clusters, but the origin of this phenomenon is still debated. We investigate the role cool supergiants may have played. To do this, we combine two investigative methods: state-of-the-art massive stellar evolution and calculations of the hydrodynamic structure of the cluster-gas. This approach allows us to study how star-formation in young massive clusters depends on the energy- and mass-input of the first-generation of stars, while predicting the chemical composition of the second-generation. We find that the presence of massive (9-500 M$_{\odot}$) metal-poor supergiants in the young cluster leads to a star-formation episode within the first 4 Myr of the cluster's lifetime, that is, before the first core-collapse supernovae explode or the gas is expelled. The stellar winds accumulate in the cluster center, forming the second-generation there. Its composition is predicted to show variations in O & Na abundances, consistently with observations. The abundance of helium is, similarly to other scenarios involving massive stars, higher than what is referred from observations. Supposing dynamical removal of stars from the outskirts of the cluster, or applying a top-heavy initial-mass-function, we can predict a number ratio of the second-generation as high as 20-80%. The effect of metallicity is shown to be important, as the most luminous supergiants are only predicted at low-metallicity, thus limiting $-$ but not excluding $-$ the extent of a polluted second-generation at high-metallicity. These massive stars becoming black-holes suggests globular-clusters hosting gravitational-wave progenitors. Our scenario predicts a correlation between the mass of the cluster and the extent of the multiple population phenomenon.
Highlights
Young massive clusters (YMCs) are compact star forming regions with a radius of only a few parsecs (Portegies Zwart et al 2010; Longmore et al 2014)
Why do we see multiple stellar populations in practically all GCs (e.g. Yong et al 2003; Gratton et al 2004; Harris 2010; Da Costa et al 2013; Bastian & Lardo 2018), and possibly in other clusters with ages up to 2 Gyr (e.g. Martocchia et al 2018a)? Since one of the main indications that a cluster harbors multiple populations, is the anomalous ratios of light elements—e.g. the observed ratio of sodium and oxygen, which can only be synthesized at temperatures as high as 60−100 MK—it has long been suggested that a first generation of massive or intermediate mass stars is responsible for the formation of an anomalous second generation
We realized a novel synergy between two research fields, massive stellar theory and cluster gas dynamics
Summary
Young massive clusters (YMCs) are compact star forming regions with a radius of only a few parsecs (Portegies Zwart et al 2010; Longmore et al 2014) Since their projected lifetimes are consistent with those of old globular clusters (GCs, Maíz-Apellániz 2002), they have been suggested to become GC-like objects eventually. Old GCs, observed to populate the bulges and halos of many galaxies including our own, are hypothised to start out as massive clusters (Brodie & Strader 2006; Andersen et al 2016). Both YMCs and GCs, as well as their suggested connection, are surrounded by observational puzzles. Responsible in which sense? What are the conditions under which a second star formation episode can happen that feeds on the material ejected from the first generation? Or, to turn the question around, is the amount of material ejected from the first generation stars enough to produce the observed number of second generation stars? This latter puzzle is usually referred to as the ‘mass budget problem’, since most scenarios suggested so far do struggle to answer yes
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
