Binary-ejected enrichment for multiple populations in globular clusters

Abstract

Globular clusters are not simple stellar populations. Practically all globular clusters show multiple populations (MPs), where at the same metallicity [Fe/H], approximately half of their stars are enriched by the products of high-temperature hydrogen burning relative to the rest that show field-like abundances. The source of enrichment for forming the enriched population is an unresolved problem. Interacting massive binaries are an underexplored proposed source of enrichment. Many assessments of the theory are based on only one modelled binary. We simulate a suite of metal-intermediate, [Fe/H]=-1.44, interacting binaries with initial primary masses of 10 to 40 solar masses, with mass ratios ranging from 0.15 to 0.9, over periods ranging from about 2 to 700 days using MESA. Our simulations show that binaries at higher masses, higher mass ratios, and near our upper period limit tend to be the most enriching with ejecta showing HeNaCNOAlMg variations consistent with hot-H burning. Some binaries do not eject material, suggesting binary mass loss can contribute to the dilution of enrichment. As a realistic population, binaries within our parameter space eject about ten times as much mass as they would as single stars. Ejection occurs on timescales of about 11 Myr, consistent with observed and theoretical limits on the age spreads for MPs. Our systems are rare, making them more suited to explaining the stochastic nature of MPs but not the large fraction of enriched stars. Spreads in He, N, Na, C, and Al for our ejecta could reasonably explain the observed spreads in clusters. Reduced variation in O and Mg suggests more massive binaries should be investigated. A multi-scale approach to cluster formation with multiple types of enrichment sources is a necessary next step for validating MP formation scenarios.