Star clusters

Abstract

Star clusters are observed in almost every galaxy. In this thesis we address several fundamental problems concerning the formation, evolution and disruption of star clusters. From observations of (young) star clusters in the interacting galaxy M51, we found that clusters are formed in complexes of stars and star clusters. These complexes share similar properties with giant molecular clouds, from which they are formed. Many (70%) of the young clusters will not survive the fist 10 Myr, due to the removal of left over gas. We study the evolution of clusters that have survived this first 10 Myr, to become bound star clusters that have cleared their primordial gas content. We determined the life time of such star clusters in M51 and the solar neighbourhood and compare these values, including existing values from literature, to the results of N-body simulations. These simulations consider realistic star clusters, with a stellar initial mass function, stellar evolution, accurate treatments of binaries and the tidal field of the host galaxy. We found that the observed disruption times of clusters in the solar neighbourhood and M51 are shorter than predicted by the simulations by a factor of 5 and 10, respectively. We studied the effect of additional perturbations by spiral arm crossings and encounters with giant molecular clouds with N-body simulations. We found that the mass loss due to these external perturbations, combined with the mass loss due to stellar evolution and the galactic tidal field can explain the observed disruption times. The star clusters in the solar neighbourhood have much lower masses than the young clusters observed in merging and interacting galaxies. We show that this can be largely explained by size-of-sample effects, that is, when more star clusters are observed, the chance of finding a more massive one is higher. However, we showed that there can exist a physical maximum to the cluster mass, which should be observable in the cluster luminosity function. We found this observational signature in the luminosity function of clusters in M51. A comparison to a cluster population model, that was developed for this thesis research, suggests that the maximum cluster mass in M51 is 5x10^5 solar masses. In the merging Antennae galaxies a similar luminosity function was observed. However, the maximum mass is four times higher there, suggesting that the maximum mass depends on galactic environment. Finally, we have studied the observability of mass segregation in extra-galactic star clusters. Due to dynamical friction, the massive stars in a star cluster are more centrally concentrated. Recent observational studied claimed to have observed this effect in extra-galactic star clusters. We have modeled star clusters with various degrees of mass segregation and quantified the observability of mass segregation. We showed that mass segregation is in principle observable, however, the observational limitations of determining fundamental cluster parameters, such as the radius, make is very hard to disentangle mass segregation from noise.