High-mass Stars are cosmic engines known to dominate the energetics in the Milky Way and other galaxies. However, their formation is still not well understood. Massive, cold, dense clouds, often appearing as Infrared Dark Clouds (IRDCs), are the nurs
eries of massive stars. No measurements of magnetic fields in IRDCs in a state prior to the onset of high-mass star formation (HMSF) have previously been available, and prevailing HMSF theories do not consider strong magnetic fields. Here, we report observations of magnetic fields in two of the most massive IRDCs in the Milky Way. We show that IRDCs G11.11-0.12 and G0.253+0.016 are strongly magnetized and that the strong magnetic field is as important as turbulence and gravity for HMSF. The main dense filament in G11.11-0.12 is perpendicular to the magnetic field, while the lower density filament merging onto the main filament is parallel to the magnetic field. The implied magnetic field is strong enough to suppress fragmentation sufficiently to allow HMSF. Other mechanisms reducing fragmentation, such as the entrapment of heating from young stars via high mass surface densities, are not required to facilitate HMSF.
Whether or not molecular clouds and embedded cloud fragments are stable against collapse is of utmost importance for the study of the star formation process. Only supercritical cloud fragments are able to collapse and form stars. The virial parameter
, alpha=M_vir/M, which compares the virial to the actual mass, provides one way to gauge stability against collapse. Supercritical cloud fragments are characterized by alpha<2, as indicated by a comprehensive stability analysis considering perturbations in pressure and density gradients. Past research has suggested that virial parameters alpha>2 prevail in clouds. This would suggest that collapse towards star formation is a gradual and relatively slow process, and that magnetic fields are not needed to explain the observed cloud structure. Here, we review a range of very recent observational studies that derive virial parameters <<2 and compile a catalogue of 1325 virial parameter estimates. Low values of alpha are in particular observed for regions of high mass star formation (HMSF). These observations may argue for a more rapid and violent evolution during collapse. This would enable competitive accretion in HMSF, constrain some models of monolithic collapse, and might explain the absence of high--mass starless cores. Alternatively, the data could point at the presence of significant magnetic fields ~1 mG at high gas densities. We examine to what extent the derived observational properties might be biased by observational or theoretical uncertainties. For a wide range of reasonable parameters, our conclusions appear to be robust with respect to such biases.
We present the first interferometric molecular line and dust emission maps for the Galactic Center (GC) cloud G0.253+0.016, observed using the Combined Array for Research in Millimeter--wave Astronomy (CARMA) and the Submillimeter Array (SMA). This c
loud is very dense, and concentrates a mass exceeding the Orion Molecular Cloud Complex (2x10^5 M_sun) into a radius of only 3pc, but it is essentially starless. G0.253+0.016 therefore violates star formation laws presently used to explain trends in galactic and extragalactic star formation by a factor ~45. Our observations show a lack of dense cores of significant mass and density, thus explaining the low star formation activity. Instead, cores with low densities and line widths 1km/s---probably the narrowest lines reported for the GC region to date---are found. Evolution over several 10^5 yr is needed before more massive cores, and possibly an Arches--like stellar cluster, could form. Given the disruptive dynamics of the GC region, and the potentially unbound nature of G0.253+0.016, it is not clear that this evolution will happen.
We aim at understanding the massive star formation (MSF) limit $m(r) = 870 M_{odot} (r/pc)^{1.33}$ in the mass-size space of molecular structures recently proposed by Kauffmann & Pillai (2010). As a first step, we build on the hypothesis of a volume
density threshold for overall star formation and the model of Parmentier (2011) to establish the mass-radius relations of molecular clumps containing given masses of star-forming gas. Specifically, we relate the mass $m_{clump}$, radius $r_{clump}$ and density profile slope $-p$ of molecular clumps which contain a mass $m_{th}$ of gas denser than a volume density threshold $rho_{th}$. In a second step, we use the relation between the mass of embedded-clusters and the mass of their most-massive star to estimate the minimum mass of star-forming gas needed to form a $10,M_{odot}$ star. Assuming a star formation efficiency of $SFE simeq 0.30$, this gives $m_{th,crit} simeq 150 M_{odot}$. In a third step, we demonstrate that, for sensible choices of the clump density index ($p simeq 1.7$) and of the cluster formation density threshold ($n_{th} simeq 10^4,cm^{-3}$), the line of constant $m_{th,crit} simeq 150 M_{odot}$ in the mass-radius space of molecular structures equates with the MSF limit for spatial scales larger than 0.3,pc. Hence, the observationally inferred MSF limit of Kauffmann & Pillai is consistent with a threshold in star-forming gas mass beyond which the star-forming gas reservoir is large enough to allow the formation of massive stars. For radii smaller than 0.3,pc, the MSF limit is shown to be consistent with the formation of a $10,M_{odot}$ star out of its individual pre-stellar core of density threshold $n_{th} simeq 10^5,cm^{-3}$. The inferred density thresholds for the formation of star clusters and individual stars within star clusters match those previously suggested in the literature.
AIMS: To study the structure of nearby (< 500 pc) dense starless and star-forming cores with the particular goal to identify and understand evolutionary trends in core properties, and to explore the nature of Very Low Luminosity Objects (< 0.1 L_sun;
VeLLOs). METHODS: Using the MAMBO bolometer array, we create maps unusually sensitive to faint (few mJy per beam) extended (approx. 5 arcmin) thermal dust continuum emission at 1.2 mm wavelength. Complementary information on embedded stars is obtained from Spitzer, IRAS, and 2MASS. RESULTS: Our maps are very rich in structure, and we characterize extended emission features (``subcores) and compact intensity peaks in our data separately to pay attention to this complexity. We derive, e.g., sizes, masses, and aspect ratios for the subcores, as well as column densities and related properties for the peaks. Combination with archival infrared data then enables the derivation of bolometric luminosities and temperatures, as well as envelope masses, for the young embedded stars. CONCLUSIONS: (abridged) Starless and star-forming cores occupy the same parameter space in many core properties; a picture of dense core evolution in which any dense core begins to actively form stars once it exceeds some fixed limit in, e.g., mass, density, or both, is inconsistent with our data. Comparison of various evolutionary indicators for young stellar objects in our sample (e.g., bolometric temperatures) reveals inconsistencies between some of them, possibly suggesting a revision of some of these indicators.
