A Helium Spread among the Main-Sequence Stars in NGC 2808

We have made a detailed study of the color distribution of the main sequence of the globular cluster (GC) NGC 2808, based on new deep HST WFPC2 photometry of a field in the uncrowded outskirts of the cluster. The observed color distribution of main-sequence stars is not Gaussian and is wider than expected for a single stellar population, given our (carefully determined) measurement errors. About 20% of the sample stars are much bluer than expected and are most plausibly explained as a population having a much larger helium abundance than the bulk of the main sequence. Using synthetic color-magnitude diagrams based on new stellar models, we estimate that the helium mass fraction of these stars is Y ~ 0.4. The newly found anomaly on the main sequence gives credence to the idea that GCs like NGC 2808 have undergone self-enrichment and that different stellar populations were born from the ejecta of the intermediate-mass asymptotic giant branch (AGB) stars of the first generation. The enhancement and spread of helium among the stars in NGC 2808 have recently been suggested as a simple way to explain the very peculiar morphology of its horizontal branch. We find that if in addition to the Y = 0.40 stars, roughly 30% of the stars have Y distributed between 0.26-0.29, while 50% have primordial Y, this leads to a horizontal-branch morphology similar to that observed. In this framework, three main stages of star formation are identified, the first with primordial helium content Y ≃ 0.24, the second born from the winds of the most massive AGBs of the first stellar generation (~6-7 M☉), with Y ~ 0.4, and a third born from the matter ejected from less massive AGBs (~3.5-4.5 M☉), with Y ~ 0.26-0.29. There could have been a long hiatus (several times 107 yr), between the second and third generation in which no star formed in the protocluster. We suggest that during this period, star formation has been inhibited by the explosion of late Type II supernovae deriving from binary evolution.