The impact of binary interaction on the main-sequence morphology of young star clusters

Abstract

Star clusters play a vital role in our understanding of stellar physics and evolution, because they are believed to contain stars born at the same time, with the same initial conditions. Therefore, the distribution of the main-sequence (MS) stars of a star cluster in the color-magnitude diagram (CMD) should be characterized by one simple isochrone. However, recent high-precision Hubble Space Telescope (HST) observations found complex MS morphologies in young star clusters, indicating the existence of multiple MS components. Despite more than ten years of effort, the origin of these MS components is still unclear. In this thesis, we reinvestigate this problem by using new state-of-the-art stellar models, considering binary interaction and rotation, both of which are widely accepted to have a significant influence on stellar structure and evolution. <br /> We first explore how binary evolution affects the MS morphology of young star clusters, as most massive stars are proposed to be born with close companions. For this aim, we calculate the evolution of more than 50 000 detailed binary models. To work out the main effects of binary interaction, we assume that the two stellar models composing a binary system are initially rotating synchronized with the orbital motion. We find that binary interaction can reproduce the multiple MS components in young star clusters, including slowly-rotating blue stragglers on the upper left side of the cluster turn-off, which may stem from binary mergers, and near-critically-rotating stars in an extended region on the right side of the cluster turn-off, which our models reproduce as mass gainers of the stable mass transfer. <br /> Our detailed binary models indeed demonstrate that binary evolution greatly impacts the MS morphology of young star clusters. However, binary evolution alone cannot explain all observed MS features, in particular, the double MSs, with the red MS containing the majority of stars and the blue MS containing a minority. There is ample evidence that rotation is responsible for the double MSs, with the fast-rotating stars appearing redder than their slowly-rotating counterparts in the CMD. We find that stellar models with natal rotational velocities of 0-35% of their critical values and 50%-65% of their critical values can explain the observed blue and red MSs, respectively. This bi-modal velocity distribution agrees with the observed velocity distribution of the field A and B type stars. <br /> We propose that the slowly-rotating blue MS stars originate from binary mergers, stimulated by recent finding that binary merger products may rotate slowly and appear younger than other cluster members. In order to explain the number of the observed blue MS stars, we require a high merger rate early during the evolution of the clusters, as predicted by recent binary formation models, and by new cluster observations. Based on our findings, we propose a scenario that can explain all the MS components in young star clusters with coeval stars. We assume that all cluster stars are born with nearly the same rotational velocities (i.e. 50%-65% of their critical velocities), no matter whether they are single stars or in binaries, and populate the red MS. The close binaries merge either during their pre-MS evolution or their early MS evolution, and produce slowly-rotating blue MS stars. The mass gainers in mass transfer systems evolve to near-critically-rotating Be stars. We examine this scenario by computing a large grid of new detailed binary models in which the two components have natal velocities of 55% of their critical values. These models indeed reproduce all observed MS components in the CMD. <br /> However, we find that our models fail to explain the large number of Be stars observed in young star clusters, because many of our binary models lead to mergers during mass transfer, according to our adopted merger criterion. When relaxing the merger criterion, we find that this problem can be alleviated. <br /> Our models predict observational signatures for stars in each MS component. Further observations, especially spectroscopic observations that can obtain rotational rate and binary fraction of the stars in young star clusters, are in high demand to thoroughly test our scenario.