Search

Context. The fragmentation of high-mass star-forming regions depends on a variety of physical parameters, including density, the magnetic field, and turbulent gas properties. Aims. We evaluate the importance of the density and magnetic field structures in relation to the fragmentation properties during high-mass star formation. Methods. Observing the large parsec-scale Stokes I millimeter dust continuum emission with the IRAM 30 m telescope and the intermediate-scale (<0.1 pc) polarized submillimeter dust emission with the Submillimeter Array toward a sample of 20 high-mass star-forming regions allows us to quantify the dependence of the fragmentation behavior of these regions on the density and magnetic field structures. Results. Based on the IRAM 30 m data, we infer density distributions n ∝ r−p of the regions with typical power-law slopes p around ~1.5. There is no obvious correlation between the power-law slopes of the density structures on larger clump scales (~1 pc) and the number of fragments on smaller core scales (<0.1 pc). Comparing the large-scale single-dish density profiles to those derived earlier from interferometric observations at smaller spatial scales, we find that the smaller-scale power-law slopes are steeper, typically around ~2.0. The flattening toward larger scales is consistent with the star-forming regions being embedded in larger cloud structures that do not decrease in density away from a particular core. The magnetic fields of several regions appear to be aligned with filamentary structures that lead toward the densest central cores. Furthermore, we find different polarization structures; some regions exhibit central polarization holes, whereas other regions show polarized emission also toward the central peak positions. Nevertheless, the polarized intensities are inversely related to the Stokes I intensities, following roughly a power-law slope of ∝ SI−0.62. We estimate magnetic field strengths between ~0.2 and ~4.5 mG, and we find no clear correlation between magnetic field strength and the fragmentation level of the regions. A comparison of the turbulent to magnetic energies shows that they are of roughly equal importance in this sample. The mass-to-flux ratios range between ~2 and ~7, consistent with collapsing star-forming regions. Conclusions. Finding no clear correlations between the present-day large-scale density structure, the magnetic field strength, and the smaller-scale fragmentation properties of the regions, indicates that the fragmentation of high-mass star-forming regions may not be affected strongly by the initial density profiles and magnetic field properties. However, considering the limited evolutionary range and spatial scales of the presented CORE analysis, future research directions should include density structure analysis of younger regions that better resemble the initial conditions, as well as connecting the observed intermediate-scale magnetic field structure with the larger-scale magnetic fields of the parental molecular clouds. [ABSTRACT FROM AUTHOR]

Context. The fragmentation mode of high-mass molecular clumps and the accretion processes that form the most massive stars (M ≳ 8 M⊙) are still not well understood. A growing number of case studies have found massive young stellar objects (MYSOs) to harbour disk-like structures, painting a picture that the formation of high-mass stars may proceed through disk accretion, similar to that of lower-mass stars. However, the properties of such structures have yet to be uniformly and systematically characterised. Aims. The aim of this work is to uniformly study the kinematic properties of a large sample of MYSOs and characterise the stability of possible circumstellar disks against gravitational fragmentation. Methods. We have undertaken a large observational programme (CORE) making use of interferometric observations from the Northern Extended Millimetre Array (NOEMA) for a sample of 20 luminous (L > 104L⊙) protostellar objects in the 1.37 mm wavelength regime in both continuum and spectral line emission, reaching 0.4″ resolution (800 au at 2 kpc). Results. We present the gas kinematics of the full sample and detect dense gas emission surrounding 15 regions within the CORE sample. Using the dense gas tracer CH3CN, we find velocity gradients across 13 cores perpendicular to the directions of bipolar molecular outflows, making them excellent disk candidates. The extent of the CH3CN emission tracing the disk candidates varies from 1800 to 8500 au. Analysing the free-fall to rotational timescales, we find that the sources are rotationally supported. The rotation profiles of some disk candidates are well described by differential rotation while for others the profiles are poorly resolved. Fitting the velocity profiles with a Keplerian model, we find protostellar masses in the range of ~ 10–25 M⊙. Modelling the level population of CH3CN (12K–11K) K = 0–6 lines, we present temperature maps and find median temperature in the range 70–210 K with a diversity in distributions. Radial profiles of the specific angular momentum (j) for the best disk candidates span a range of 1–2 orders of magnitude, on average ~10−3 km s−1 pc, and they follow j ∝ r1.7, which is consistent with a poorly resolved rotating and infalling envelope-disk model. Studying the Toomre stability of the disk candidates, we find almost all (11 out of 13) disk candidates to be prone to fragmentation due to gravitational instabilities at the scales probed by our observations, as a result of their high disk to stellar mass ratio. In particular, disks with masses greater than ~ 10–20% of the mass of their host (proto)stars are Toomre unstable, and more luminous YSOs tend to have disks that are more massive compared to their host star and hence more prone to fragmentation. Conclusions. In this work, we show that most disk structures around high-mass YSOs are prone to disk fragmentation early in their formation due to their high disk to stellar mass ratio. This impacts the accretion evolution of high-mass protostars which will have significant implications for the formation of the most massive stars. [ABSTRACT FROM AUTHOR]

Context. Most massive stars are located in multiple stellar systems. The modeling of disk fragmentation, a mechanism that may plausibly lead to stellar multiplicity, relies on parallel 3D simulation codes whose agreement remains to be evaluated. Aims. Cartesian adaptive-mesh refinement (AMR) and spherical codes have frequently been used in the past decade to study massive star formation. We aim to study how the details of collapse and disk fragmentation depend on these codes. Methods. Using the Cartesian AMR code RAMSES within its self-gravity radiation-hydrodynamical framework, we compared disk fragmentation in a centrally condensed protostellar system to the findings of earlier studies performed on a grid in spherical coordinates using PLUTO. Results. To perform the code comparison, two RAMSES runs were considered, effectively giving qualitatively distinct pictures. On the one hand, when allowing for unlimited sink particle creation with no initial sink, Toomre instability and subsequent gas fragmentation leads to a multiple stellar system whose multiplicity is affected by the grid when triggering fragmentation and via numerically assisted mergers. On the other hand, using a unique, central, fixed-sink particle, a centrally-condensed system forms that is similar to that reported by PLUTO. Hence, the RAMSES-PLUTO comparison was performed with the latter and an agreement between the two codes is found as to the first rotationally supported disk formation, the presence of an accretion shock onto it, and the first fragmentation phase. Gaseous fragments form. The properties of the fragments (i.e., number, mass, and temperature) are dictated by local thermodynamics and are in agreement between the two codes given that the system has entered a highly nonlinear phase. Over the simulations, the stellar accretion rate is made of accretion bursts and continuous accretion on the same order of magnitude. As a minor difference between both codes, the dynamics of the fragments causes the disk structure to be sub-Keplerian in RAMSES, whereas it is found to be Keplerian, thus reaching quiescence, in PLUTO. We attribute this discrepancy to the central star being twice less massive in RAMSES because of the different stellar accretion subgrid models in use - rather than the potential grid effects. Conclusions. In a centrally condensed system, the agreement between RAMSES and PLUTO regarding many of the collapse properties and fragmentation process is good. In contrast, fragmentation occurring in the innermost region and given specific numerical choices (use of sink particles, grid, etc.) have a crucial impact when similar but smooth initial conditions are employed. These aspects prove more crucial than the choice of code, with regard to the system being multiple or centrally condensed. [ABSTRACT FROM AUTHOR]

Context. There is growing evidence that massive stars grow by disk accretion in a similar way to their low-mass counterparts. Early in evolution, these disks can achieve masses that are comparable to the current stellar mass, and therefore the forming disks are highly susceptible to gravitational fragmentation. Aims. We investigate the formation and early evolution of an accretion disk around a forming massive protostar, focussing on its fragmentation physics. To this end, we follow the collapse of a molecular cloud of gas and dust, the formation of a massive protostar, the formation of its circumstellar disk, and the formation and evolution of the disk fragments. Methods. We used a grid-based, self-gravity radiation hydrodynamics code including a sub-grid module for stellar evolution and dust evolution. We purposely do not use a sub-grid module for fragmentation such as sink particles to allow for all paths of fragment formation and destruction, but instead we keep the spatial grid resolution high enough to properly resolve the physical length scales of the problem, namely the pressure scale height and Jeans length of the disk. Simulations are performed on a grid in spherical coordinates with a logarithmic spacing of the grid cells in the radial direction and a cosine distribution of the grid cells in the polar direction, focusing the spatial resolution on the disk midplane. As a consequence, roughly 25% of the total number of grid cells, corresponding to ~26 million grid cells, are used to model the disk physics. These constitute the highest resolution simulations performed up to now on disk fragmentation around a forming massive star with the physics considered here. For a better understanding of the effects of spatial resolution and to compare our high-resolution results with previous lower resolution studies in the literature, we perform the same simulation at five different resolutions, each run differing in resolution from its predecessor by a factor of two. Results. The cloud collapses and a massive (proto)star is formed in its center surrounded by a fragmenting Keplerian-like accretion disk with spiral arms. The fragments have masses of ~1 M⊙, and their continuous interactions with the disk, spiral arms, and other fragments result in eccentric orbits. Fragments form hydrostatic cores surrounded by secondary disks with spiral arms that also produce new fragments. We identified several mechanisms of fragment formation, interaction, and destruction. Central temperatures of the fragments can reach the hydrogen dissociation limit, form second Larson cores, and evolve into companion stars. Based on this, we study the multiplicity predicted by the simulations and find approximately six companions at different distances from the primary: from possible spectroscopic multiples, to companions at distances between 1000 and 2000 au. [ABSTRACT FROM AUTHOR]

We have studied the dust density, temperature, and velocity distributions of the archetypal massive young stellar object (MYSO) AFGL 2591. Given its high luminosity (⁠|$L=2\times 10^5\, \mbox{L$_{\odot}$}$|⁠) and distance (d  = 3.3 kpc), AFGL 2591 has one of the highest |$\sqrt{L}/d$| ratio, giving better resolved dust emission than any other MYSO. As such, this paper provides a template on how to use resolved multiwavelength data and radiative transfer to obtain a well-constrained 2D axisymmetric analytic rotating infall model. We show for the first time that the resolved dust continuum emission from Herschel 70-  μ m observations is extended along the outflow direction, whose origin is explained in part from warm dust in the outflow cavity walls. However, the model can only explain the kinematic features from CH3CN observations with unrealistically low stellar masses (<15 M⊙), indicating that additional physical processes may be playing a role in slowing down the envelope rotation. As part of our three-step continuum and line fitting, we have identified model parameters that can be further constrained by specific observations. High-resolution mm visibilities were fitted to obtain the disc mass (6 M⊙) and radius (2200 au). A combination of SED and near-infrared observations were used to estimate the luminosity and envelope mass together with the outflow cavity inclination and opening angles. [ABSTRACT FROM AUTHOR]

Context. The formation process of high-mass stars (>8 M⊙) is poorly constrained, particularly the effects of clump fragmentation creating multiple systems and the mechanism of mass accretion onto the cores. Aims. We study the fragmentation of dense gas clumps, and trace the circumstellar rotation and outflows by analyzing observations of the high-mass (~500 M⊙) star-forming region IRAS 23033+5951. Methods. Using the Northern Extended Millimeter Array (NOEMA) in three configurations and the IRAM 30 m single-dish telescope at 220 GHz, we probe the gas and dust emission at an angular resolution of ~0.45′′, corresponding to 1900 au. Results. In the millimeter (mm) continuum emission, we identify a protostellar cluster with at least four mm-sources, where three of them show a significantly higher peak intensity well above a signal-to-noise ratio of 100. Hierarchical fragmentation from large to small spatial scales is discussed. Two fragments are embedded in rotating structures and drive molecular outflows, traced by 13CO (2–1) emission. The velocity profiles across two of the cores are similar to Keplerian but are missing the highest-velocity components close to the center of rotation, which is a common phenomena from observations like these, and other rotation scenarios are not excluded entirely. Position–velocity diagrams suggest protostellar masses of ~6 and 19 M⊙. Rotational temperatures from fitting CH3CN (12K− 11K) spectra are used for estimating the gas temperature and thereby also the disk stability against gravitational fragmentation, utilizing Toomre's Q parameter. Assuming that the candidate disk is in Keplerian rotation about the central stellar object and considering different disk inclination angles, we identify only one candidate disk as being unstable against gravitational instability caused by axisymmetric perturbations. Conclusions. The dominant sources cover different evolutionary stages within the same maternal gas clump. The appearance of rotation and outflows of the cores are similar to those found in low-mass star-forming regions. [ABSTRACT FROM AUTHOR]

We present the highest angular resolution (∼20 × 15 mas–44 × 33 au) Atacama Large Millimeter/sub-millimeter Array (ALMA) observations that are currently possible of the proto-O-star G17.64+0.16 in Band 6. The Cycle 5 observations with baselines out to 16 km probe scales < 50 au and reveal the rotating disc around G17.64+0.16, a massive forming O-type star. The disc has a ring-like enhancement in the dust emission that is especially visible as arc structures to the north and south. The Keplerian kinematics are most prominently seen in the vibrationally excited water line, H2O 55, 0−64, 3 ν2 = 1 (Eu = 3461.9 K). The mass of the central source found by modelling the Keplerian rotation is consistent with 45 ± 10 M⊙. The H30α (231.9 GHz) radio-recombination line and the SiO (5-4) molecular line were detected at up to the ∼10σ level. The estimated disc mass is 0.6 − 2.6 M⊙ under the optically thin assumption. Analysis of the Toomre Q parameter in the optically thin regime indicates that the disc stability is highly dependent on temperature. The disc currently appears stable for temperatures > 150 K; this does not preclude that the substructures formed earlier through disc fragmentation. [ABSTRACT FROM AUTHOR]

Context. This study is part of the CORE project, an IRAM/NOEMA large program consisting of observations of the millimeter continuum and molecular line emission towards 20 selected high-mass star-forming regions. The goal of the program is to search for circumstellar accretion disks, study the fragmentation process of molecular clumps, and investigate the chemical composition of the gas in these regions. Aims. We focus on IRAS 23385+6053, which is believed to be the least-evolved source of the CORE sample. This object is characterized by a compact molecular clump that is IR-dark shortward of 24 μm and is surrounded by a stellar cluster detected in the near-IR. Our aim is to study the structure and velocity field of the clump. Methods. Observations were performed at ~1.4 mm and employed three configurations of NOEMA and additional single-dish maps, merged with the interferometric data to recover the extended emission. Our correlator setup covered a number of lines from well-known hot core tracers and a few outflow tracers. The angular (~0′′.45–0′′.9) and spectral (0.5 km s−1) resolutions were sufficient to resolve the clump in IRAS 23385+6053 and investigate the existence of large-scale motions due to rotation, infall, or expansion. Results. We find that the clump splits into six distinct cores when observed at sub-arcsecond resolution. These are identified through their 1.4 mm continuum and molecular line emission. We produce maps of the velocity, line width, and rotational temperature from the methanol and methyl cyanide lines, which allow us to investigate the cores and reveal a velocity and temperature gradient in the most massive core. We also find evidence of a bipolar outflow, possibly powered by a low-mass star. Conclusions. We present the tentative detection of a circumstellar self-gravitating disk lying in the most massive core and powering a large-scale outflow previously known in the literature. In our scenario, the star powering the flow is responsible for most of the luminosity of IRAS 23385+6053 (~3000 L⊙). The other cores, albeit with masses below the corresponding virial masses, appear to be accreting material from their molecular surroundings and are possibly collapsing or on the verge of collapse. We conclude that we are observing a sample of star-forming cores that is bound to turn into a cluster of massive stars. [ABSTRACT FROM AUTHOR]

Context. The hierarchical process of star formation has so far mostly been studied on scales from thousands of au to parsecs, but the smaller sub-1000 au scales of high-mass star formation are still largely unexplored in the submillimeter regime. Aims. We aim to resolve the dust and gas emission at the highest spatial resolution to study the physical properties of the densest structures during high-mass star formation. Methods. We observed the high-mass hot core region G351.77-0.54 with the Atacama Large Millimeter Array with baselines extending out to more than 16 km. This allowed us to dissect the region at sub-50 au spatial scales. Results. At a spatial resolution of 18/40 au (depending on the distance), we identify twelve sub-structures within the inner few thousand au of the region. The brightness temperatures are high, reaching values greater 1000 K, signposting high optical depth toward the peak positions. Core separations vary between sub-100 au to several 100 and 1000 au. The core separations and approximate masses are largely consistent with thermal Jeans fragmentation of a dense gas core. Due to the high continuum optical depth, most spectral lines are seen in absorption. However, a few exceptional emission lines are found that most likely stem from transitions with excitation conditions above 1000 K. Toward the main continuum source, these emission lines exhibit a velocity gradient across scales of 100–200 au aligned with the molecular outflow and perpendicular to the previously inferred disk orientation. While we cannot exclude that these observational features stem from an inner hot accretion disk, the alignment with the outflow rather suggests that it stems from the inner jet and outflow region. The highest-velocity features are found toward the peak position, and no Hubble-like velocity structure can be identified. Therefore, these data are consistent with steady-state turbulent entrainment of the hot molecular gas via Kelvin–Helmholtz instabilities at the interface between the jet and the outflow. Conclusions. Resolving this high-mass star-forming region at sub-50 au scales indicates that the hierarchical fragmentation process in the framework of thermal Jeans fragmentation can continue down to the smallest accessible spatial scales. Velocity gradients on these small scales have to be treated cautiously and do not necessarily stem from disks, but may be better explained with outflow emission. Studying these small scales is very powerful, but covering all spatial scales and deriving a global picture from large to small scales are the next steps to investigate. [ABSTRACT FROM AUTHOR]

Context. Massive stars live short but intense lives. While less numerous than low-mass stars, they enormously impact their surroundings by several feedback mechanisms. They form in opaque and far-away regions of the galaxy, such that one of these feedback mechanisms also becomes a record of their evolution: their bright large-scale jets and outflows. Aims. In a comprehensive convergence study, we investigate the computational conditions necessary to resolve (pseudo-) disk formation and jet-launching processes, and analyze possible caveats. We explore the magneto-hydrodynamic (MHD) processes of the collapse of massive prestellar cores in detail, including an analysis of the forces involved and their temporal evolution for up to two free-fall times. Methods. We conduct MHD simulations using the state-of-the-art code PLUTO, combining nonideal MHD, self-gravity, and very high resolutions as they have never been achieved before. Our setup includes a 100 M⊙ cloud core that collapses under its own self-gravity to self-consistently form a dense disk structure and launch tightly collimated magneto-centrifugal jets and wide-angle tower flows. Results. We show a comprehensive evolutionary picture of the collapse of a massive prestellar core with a detailed analysis of the physical processes involved and our high-resolution simulations can resolve a magneto-centrifugal jet and a magnetic pressure-driven outflow, separately. The nature of the outflows depends critically on spatial resolution. Only high-resolution simulations are able to differentiate a magneto-centrifugally launched, highly collimated jet from a slow wide-angle magnetic-pressure-driven tower flow. Of these two outflow components, the tower flow dominates angular-momentum transport. The mass outflow rate is dominated by the entrained material from the interaction of the jet with the stellar environment and only part of the ejected medium is directly launched from the accretion disk. A tower flow can only develop to its full extent when much of the original envelope has already dispersed. Taking into account both the mass launched from the surface of the disk and the entrained material from the envelope, we find an ejection-to-accretion efficiency of 10%. Nonideal MHD is required to form centrifugally supported accretion disks and the disk size is strongly dependent on spatial resolution. A converged result for disk and both outflow components requires a spatial resolution of Δx ≤ 0.17 au at 1 au and sink-cell sizes ≤3.1 au. Conclusions. Massive stars not only possess slow wide-angle tower flows, but also produce magneto-centrifugal jets, just as their low-mass counterparts. The actual difference between low-mass and high-mass star formation lies in the "embeddedness" of the high-mass star which implies that the jet and tower flow interact with the infalling large-scale stellar environment, potentially resulting in entrainment. [ABSTRACT FROM AUTHOR]

Context. High-mass stars form in clusters, but neither the early fragmentation processes nor the detailed physical processes leading to the most massive stars are well understood. Aims. We aim to understand the fragmentation, as well as the disk formation, outflow generation, and chemical processes during high-mass star formation on spatial scales of individual cores. Methods. Using the IRAM Northern Extended Millimeter Array (NOEMA) in combination with the 30 m telescope, we have observed in the IRAM large program CORE the 1.37 mm continuum and spectral line emission at high angular resolution (~0.4″) for a sample of 20 well-known high-mass star-forming regions with distances below 5.5 kpc and luminosities larger than 104L⊙. Results. We present the overall survey scope, the selected sample, the observational setup, and the main goals of CORE. Scientifically, we concentrated on the mm continuum emission on scales on the order of 1000 AU. We detect strong mm continuum emission from all regions, mostly due to the emission from cold dust. The fragmentation properties of the sample are diverse. We see extremes where some regions are dominated by a single high-mass core whereas others fragment into as many as 20 cores. A minimum-spanning-tree analysis finds fragmentation at scales on the order of the thermal Jeans length or smaller suggesting that turbulent fragmentation is less important than thermal gravitational fragmentation. The diversity of highly fragmented vs. singular regions can be explained by varying initial density structures and/or different initial magnetic field strengths. Conclusions. A large sample of high-mass star-forming regions at high spatial resolution allows us to study the fragmentation properties of young cluster-forming regions. The smallest observed separations between cores are found around the angular resolution limit which indicates that further fragmentation likely takes place on even smaller spatial scales. The CORE project with its numerous spectral line detections will address a diverse set of important physical and chemical questions in the field of high-mass star formation. [ABSTRACT FROM AUTHOR]

Context. Circumstellar discs around massive stars could mediate the accretion onto the star from the infalling envelope, and could minimize the effects of radiation pressure. Despite such a crucial role, only a few convincing candidates have been provided for discs around deeply embedded O-type (proto)stars. Aims. In order to establish whether disc-mediated accretion is the formation mechanism for the most massive stars, we have searched for circumstellar, rotating discs around a limited sample of six luminous (>105 L⊙) young stellar objects. These objects were selected on the basis of their IR and radio properties in order to maximize the likelihood of association with disc+jet systems. Methods. We used ALMA with ~0".2 resolution to observe a large number of molecular lines typical of hot molecular cores. In this paper we limit our analysis to two disc tracers (methyl cyanide, CH3CN, and its isotopologue, 13CH3CN), and an outflow tracer (silicon monoxide, SiO). Results. We reveal many cores, although their number depends dramatically on the target. We focus on the cores that present prominent molecular line emission. In six of these a velocity gradient is seen across the core, three of which show evidence of Keplerian-like rotation. The SiO data reveal clear but poorly collimated bipolar outflow signatures towards two objects only. This can be explained if real jets are rare (perhaps short-lived) in very massive objects and/or if stellar multiplicity significantly affects the outflow structure. For all cores with velocity gradients, the velocity field is analysed through position-velocity plots to establish whether the gas is undergoing rotation with 3rot ∝ R-α, as expected for Keplerian-like discs. Conclusions. Our results suggest that in three objects we are observing rotation in circumstellar discs, with three more tentative cases, and one core where no evidence for rotation is found. In all cases but one, we find that the gas mass is less than the mass of any embedded O-type star, consistent with the (putative) discs undergoing Keplerian-like rotation. With the caveat of low number statistics, we conclude that the disc detection rate could be sensitive to the evolutionary stage of the young stellar object. In young, deeply embedded sources, the evidence for discs could be weak because of confusion with the surrounding envelope, while in the most evolved sources the molecular component of the disc could have already been dispersed. Only in those objects that are at an intermediate stage of the evolution would the molecular disc be sufficiently prominent and relatively less embedded to be detectable by mm/submm observations. [ABSTRACT FROM AUTHOR]

Aims. We study the fragmentation and collapse properties of the dense gas during the onset of high-mass star formation. Methods. We observed the massive (~800 M⊙) starless gas clump IRDC18310-4 with the Plateau de Bure Interferometer (PdBI) at subarcsecond resolution in the 1.07 mm continuum and N2H+(3-2) line emission. Results. Zooming from a single-dish low-resolution map to previous 3 mm PdBI data, and now the new 1.07 mm continuum observations, the substructures hierarchically fragment on the increasingly smaller spatial scales. While the fragment separations may still be roughly consistent with pure thermal Jeans fragmentation, the derived core masses are almost two orders of magnitude larger than the typical Jeans mass at the given densities and temperatures. However, the data can be reconciled with models using non-homogeneous initial density structures, turbulence, and/or magnetic fields. While most subcores remain (far-)infrared dark even at 70 µm, we identify weak 70 µm emission toward one core with a comparably low luminosity of ~16 L⊙, supporting the notion of the general youth of the region. The spectral line data always exhibit multiple spectral components toward each core with comparably small line widths for the individual components (in the 0.3 to 1.0 km s-1 regime). Based on single-dish C18O(2-1) data we estimate a low virial-to-gas-mass ratio ≤0.25. We propose that the likely origin of these spectral properties may be the global collapse of the original gas clump that results in multiple spectral components along each line of sight. Even within this dynamic picture the individual collapsing gas cores appear to have very low levels of internal turbulence. [ABSTRACT FROM AUTHOR]

Context. When massive stars exert a radiation pressure onto their environment that is higher than their gravitational attraction (super- Eddington condition), they launch a radiation-pressure-driven outflow, which creates cleared cavities. These cavities should prevent any further accretion onto the star from the direction of the bubble, although it has been claimed that a radiative Rayleigh-Taylor instability should lead to the collapse of the outflow cavity and foster the growth of massive stars. Aims. We investigate the stability of idealized radiation-pressure-dominated cavities, focusing on its dependence on the radiation transport approach used in numerical simulations for the stellar radiation feedback. Methods. We compare two different methods for stellar radiation feedback: gray flux-limited diffusion (FLD) and ray-tracing (RT). Both methods are implemented in our self-gravity radiation hydrodynamics simulations for various initial density structures of the collapsing clouds, eventually forming massive stars. We also derive simple analytical models to support our findings. Results. Both methods lead to the launch of a radiation-pressure-dominated outflow cavity. However, only the FLD cases lead to prominent instability in the cavity shell. The RT cases do not show such instability; once the outflow has started, it precedes continuously. The FLD cases display extended epochs of marginal Eddington equilibrium in the cavity shell, making them prone to the radiative Rayleigh-Taylor instability. In the RT cases, the radiation pressure exceeds gravity by 1-2 orders of magnitude. The radiative Rayleigh-Taylor instability is then consequently suppressed. It is a fundamental property of the gray FLD method to neglect the stellar radiation temperature at the location of absorption and thus to underestimate the opacity at the location of the cavity shell. Conclusions. Treating the stellar irradiation in the gray FLD approximation underestimates the radiative forces acting on the cavity shell. This can lead artificially to situations that are affected by the radiative Rayleigh-Taylor instability. The proper treatment of direct stellar irradiation by massive stars is crucial for the stability of radiation-pressure-dominated cavities [ABSTRACT FROM AUTHOR]