2014A&A...572L...1S

Massive stars are extremely important for the evolution of the galaxies; there are large gaps in our understanding of their properties and formation, however, mainly because they evolve rapidly, are rare, and distant. Recent findings suggest that most O-stars belong to multiple systems. It may well be that almost all massive stars are born as triples or higher multiples, but their large distances require very high angular resolution to directly detect the companions at milliarcsecond scales. Herschel36 is a young massive system located at 1.3kpc. It has a combined smallest predicted mass of 45M_☉. Multi-epoch spectroscopic data suggest the existence of at least three gravitationally bound components. Two of them, system Ab, are tightly bound in a spectroscopic binary, and the third one, component Aa, orbits in a wider orbit. Our aim was to image and obtain astrometric and photometric measurements of components Aa and Ab using, for the first time, long-baseline optical interferometry to further constrain its nature. We observed Herschel36 with the near-infrared instrument AMBER attached to the ESO VLT Interferometer, which provides an angular resolution of ∼2mas. We used the code BSMEM to perform the interferometric image reconstruction. We fitted the interferometric observables using proprietary IDL routines and the code LitPro. We imaged the Aa+Ab components of Herschel36 in H and K filters. Component Ab is located at a projected distance of 1.81 mas, at a position angle of ∼222° east of north, the flux ratio between components Aa and Ab is close to one. These findings agree with previous predictions about the properties of Herschel 36. The small measured angular separation indicates that system Ab and Ab may be approaching the periastron of their orbits. These results, only achievable with long-baseline near-infrared interferometry, constitute the first step toward a thorough understanding of this massive triple system.