Extended Main Sequence Turnoffs in Intermediate-Age Star Clusters: A Correlation Between Turnoff Width and Early Escape Velocity

Abstract

We present color-magnitude diagram analysis of deep Hubble Space Telescope imaging of a mass-limited sample of 18 intermediate-age (1 -2 Gyr old) star clusters in the Magellanic Clouds, including 8 clusters for which new data was obtained. We find that all star clusters in our sample feature extended main sequence turnoff (eMSTO) regions that are wider than can be accounted for by a simple stellar population (including unresolved binary stars). FWHM widths of the MSTOs indicate age spreads of 200 -550 Myr. We evaluate dynamical evolution of clusters with and without initial mass segregation. Our main results are: (1) the fraction of red clump (RC) stars in secondary RCs in eMSTO clusters scales with the fraction of MSTO stars having pseudo-ages 1.35 Gyr; (2) the width of the pseudo-age distributions of eMSTO clusters is correlated with their central escape velocity v esc , both currently and at an age of 10 Myr. We find that these two results are unlikely to be reproduced by the effects of interactive binary stars or a range of stellar rotation velocities. We therefore argue that the eMSTO phenomenon is mainly caused by extended star formation within the clusters;(3) we find that v esc ≥ 15 km s −1 out to ages of at least 100 Myr for all clusters featuring eMSTOs, while v esc ≤ 12 km s −1 at all ages for two lower-mass clusters in the same age range that do not show eMSTOs. We argue that eMSTOs only occur for clusters whose early escape velocities are higher than the wind velocities of stars that provide material from which second-generation stars can form. The threshold of 12 -15 km s −1 is consistent with wind velocities of intermediate-mass AGB stars and massive binary stars in the literature.

The morphology of the sub-giant branch and red clump reveal no sign of age spreads in intermediate-age clusters

“…At present, it is unclear how rotation will affect the SGB and RC, which we will leave to a future work. Goudfrooij et al (2014) found a correlation between the fraction of blue/bright stars on the MSTO (i.e., young stars if interpreted as due to age effects) and the fraction of "secondary RC" stars to the main body of RC stars, i.e. stars that are bluer and fainter than the main red clump.…”

“…The authors use this to argue for the presence of age spreads within their cluster sample, although, as shown here, the main body of the RC does not show evidence for age spreads. Goudfrooij et al (2011Goudfrooij et al ( , 2014 have suggested that a correlation exists between the width of the eMSTO and the escape velocity of the cluster. However, this result depends heavily on large "correction factors" to change the current observed escape velocities to the "initial ones".…”

Controversial age spreads from the main sequence turn-off and red clump in intermediate-age clusters in the LMC

“…The data reduction procedure is described in Goudfrooij et al (2014). As with the first data set, we obtained the catalogs from the authors of a previous work, in this case Goudfrooij et al (2014).…”

“…Cabrera-Ziri et al (2014) analyzed the integrated spectrum of NGC 34 cluster 1, a ∼100 Myr, 10 7 M cluster, and did not find evidence for multiple star-forming bursts or an extended SFH. Goudfrooij et al (2011bGoudfrooij et al ( , 2014) put forward a model to explain how a prolonged period of star formation could have happened in intermediate-age LMC clusters. The authors propose a formation of a cluster that has several steps: In the first step a cluster, which is much more massive than observed today, forms within a single burst of star formation.…”

Globular cluster formation with multiple stellar populations from hierarchical star cluster complexes

“…Most of the investigated GCs in the Galaxy are observed to show anti-correlations between light elements (e.g., C, N, and O) of cluster members stars (e.g., Carretta et al 2009; C09) whereas only 8 GCs have been so far confirmed to have star-to-star abundance spreads in heavy elements (e.g., Yong et al 2014;Marino et al 2015). Extended main-sequence turn-offs (eMSTOs) and splits in main-sequence observed in the color magnitude diagrams (CMDs) of some LMC GCs (e.g., Mackey & Broby ⋆ E-mail: kenji.bekki@uwa.edu.au Nielsen 2007; Goudfrooij et al 2014;Milone et al 2016) can be possible evidence for the multiple stellar populations with different ages, though recent observations suggest that internal stellar rotation rather than age spreads could explain the physical properties of LMC clusters with eMSTOs (e.g., Bastian & De Mink 2009;Milone et al 2016;Li et al 2016). These new discoveries stimulated much discussion on the initial stellar mass function of stars in GCs, the formation processes of GCs, and the origin of the observed diversity in chemical and dynamical properties of GCs with multiple stellar populations (e.g., D'Antona& Caloi 2004; Bekki et al 2007;Baumgardt et al 2008;D'Ercole et al 2008, D08;Vesperini et al 2010;Renzini 2015;D'Antona et al 2016).…”