Recently, my Iranian collaborators and I published another paper on mass segregation in outer-halo globular clusters. This time we looked at the effect that primordial mass segregation can have on the size evolution of these clusters (i.e., what happens if heavy stars are preferentially born closer to a star cluster’s center). The problem is the following: if you look at the globular clusters in the outer halo of our Galaxy, you find them to be significantly more extended (i.e., with a radius larger than 5-6 pc) than their counterparts that are closer to the Galactic center (orange points in the Figure above; but see also my previous posts  ).
We are interested in the question: “Have these outer-halo cluster always been extended, or did they form as compact clusters like the majority of all clusters we observe in the Universe?” This is important for our understanding of star formation. Most of our knowledge about the formation of galaxies in the early Universe (i.e., like 10-13 billion years ago) is based on the assumption that star formation followed more or less universal laws. By looking at unusual objects like the outer-halo clusters of the Milky Way, we can test if our understanding of the laws of star formation is good enough to make such assumptions about the past.
Hosein, the lead author of our publication, made extensive N-body simulations of star clusters evolving at different distances from the Galaxy and with different conditions at birth. As shown in the above Figure, we find that star clusters can only reach the large extensions they show today in the outer halo of the Milky Way, when they either start more extended than most of the clusters we observe today in the Universe (blue line), or if they were born with very strong primordial mass segregation (red line). The latter would mean that the most massive stars were all born in the centers of these clusters. When such massive stars explode as super novae within the first few million years of a cluster’s lifetime, they release huge amounts of binding energy, which cause the cluster to expand. If these massive stars explode somewhere else in the cluster, and not in the center, the energy budget of the cluster does not change so dramatically and the cluster expands less.
Together with our previous publications on the outer-halo clusters Palomar 4 and Palomar 14, these results indicate that star formation and the assembly of star clusters in weak tidal fields, like the outer halos of galaxies, may be different from our standard picture of star formation.