A potential new phase of massive star formation: A luminous outflow cavity centred on an infrared quiet core

Abstract

Context. Due to the sparsity and rapid evolution of high-mass stars, a detailed picture of the evolutionary sequence of massive protostellar objects still remains to be drawn. Some of the early phases of their formation are so short that only a select number of objects throughout the Milky Way currently find themselves spending time in those phases. Aims. Star-forming regions going through the shortest stages of massive star formation present different observational characteristics than most regions. By studying the dust continuum and line emission of such unusual clouds, one might be able to set strong constraints on the evolution of massive protostellar objects. Methods. We present a detailed analysis of the G345.88-1.10 hub filament system, which is a newly discovered star-forming cloud that hosts an unusually bright bipolar infrared nebulosity at its centre. We used archival continuum observations from Herschel, WISE, Spitzer, 2MASS, and SUMSS in order to fully characterise the morphology and spectral energy distribution of the region. We further made use of APEX 12CO(2-1), 13CO(2-1), C18O(2-1), and H30α observations to investigate the presence of outflows and map the kinematics of the cloud. Finally, we performed RADMC-3D radiative transfer calculations to constrain the physical origin of the central nebulosity. Results. At a distance of 2.26+0:30 -0:21 kpc, G345.88-1.10 exhibits a network of parsec-long converging filaments. At the junction of these filaments lie four infrared-quiet fragments. The fragment H1 is the densest one (with M = 210 M⊙, Reff = 0:14 pc) and sits right at the centre of a wide (opening angle of ∼90 ± 15°) bipolar nebulosity where the column density reaches local minima. The 12CO(2-1) observations of the region show that these infrared-bright cavities are spatially associated with a powerful molecular outflow that is centred on the H1 fragment. Negligible radio continuum and no H30α emission is detected towards the cavities, seemingly excluding the idea that ionising radiation drives the evolution of the cavities. Radiative transfer calculations of an embedded source surrounded by a disc and/or a dense core are unable to reproduce the observed combination of a low-luminosity (≲500 L⊙) central source and a surrounding high-luminosity (∼4000 L⊙) mid-infrared-bright bipolar cavity. This suggests that radiative heating from a central protostar cannot be responsible for the illumination of the outflow cavities. Conclusions. This is, to our knowledge, the first reported object of this type. The rarity of objects such as G345.88-1.10 is likely related to a very short phase in the massive star and/or cluster formation process that has been unidentified thus far. We discuss whether mechanical energy deposition by one episode or successive episodes of powerful mass accretion in a collapsing hub might explain the observations. While promising in some aspects, a fully coherent scenario that explains the presence of a luminous bipolar cavity centred on an infrared-dark fragment remains elusive at this point.