No Arabic abstract
The most luminous Spitzer point sources in the 30 Doradus triggered second generation are investigated coherently in the 3-8 micron region. Remarkable diversity and complexity in their natures are revealed. Some are also among the brightest JHK sources, while others are not. Several of them are multiple when examined at higher angular resolutions with HST NICMOS and WFPC2/WFC3 as available, or with VISTA/VMC otherwise. One is a dusty compact H II region near the far northwestern edge of the complex, containing a half dozen bright I-band sources. Three others appear closely associated with luminous WN stars and causal connections are suggested. Some are in the heads of dust pillars oriented toward R136, as previously discussed from the NICMOS data. One resides in a compact cluster of much fainter sources, while another appears monolithic at the highest resolutions. Surprisingly, one is the brighter of the two extended mystery spots associated with Knot 2 of Walborn et al. Masses are derived from YSO models for unresolved sources and lie in the 10-30 M_{sun} range. Further analysis of the IR sources in this unique region will advance understanding of triggered massive star formation, perhaps in some unexpected and unprecedented ways.
Determining the efficiency with which gas is converted into stars in galaxies requires an accurate determination of the total reservoir of molecular gas mass. However, despite being the most abundant molecule in the Universe, H$_2$ is challenging to detect through direct observations and indirect methods have to be used to estimate the total molecular gas reservoir. These are often based on scaling relations from tracers such as CO or dust, and are generally calibrated in the Milky Way. Yet, evidence that these scaling relations are environmentally dependent is growing. In particular, the commonly used CO-to-H$_2$ conversion factor (X$_{rm CO}$) is expected to be higher in metal-poor and/or strongly UV-irradiated environments. We use new SOFIA/FIFI-LS observations of far-infrared fine structure lines from the ionised and neutral gas and the Meudon photodissociation region model to constrain the physical properties and the structure of the gas in the massive star-forming region of 30 Doradus in the Large Magellanic Cloud, and determine the spatially resolved distribution of the total reservoir of molecular gas in the proximity of the young massive cluster R136. We compare this value with the molecular gas mass inferred from ground-based CO observations and dust-based estimates to quantify the impact of this extreme environment on commonly used tracers of the molecular gas. We find that the strong radiation field combined with the half-solar metallicity of the surrounding gas are responsible for a large reservoir of CO-dark molecular gas, leaving a large fraction of the total H$_2$ gas (> 75%) undetected when adopting a standard X$_{rm CO}$ factor in this massive star-forming region.
Using observations obtained with the Wide Field Camera 3 (WFC3) on board the Hubble Space Telescope (HST), we have studied the properties of the stellar populations in the central regions of 30 Dor, in the Large Magellanic Cloud. The observations clearly reveal the presence of considerable differential extinction across the field. We characterise and quantify this effect using young massive main sequence stars to derive a statistical reddening correction for most objects in the field. We then search for pre-main sequence (PMS) stars by looking for objects with a strong (> 4 sigma) Halpha excess emission and find about 1150 of them over the entire field. Comparison of their location in the Hertzsprung-Russell diagram with theoretical PMS evolutionary tracks for the appropriate metallicity reveals that about one third of these objects are younger than ~4Myr, compatible with the age of the massive stars in the central ionising cluster R136, whereas the rest have ages up to ~30Myr, with a median age of ~12Myr. This indicates that star formation has proceeded over an extended period of time, although we cannot discriminate between an extended episode and a series of short and frequent bursts that are not resolved in time. While the younger PMS population preferentially occupies the central regions of the cluster, older PMS objects are more uniformly distributed across the field and are remarkably few at the very centre of the cluster. We attribute this latter effect to photoevaporation of the older circumstellar discs caused by the massive ionising members of R136.
X-ray observations provide a potentially powerful tool to study starburst feedback. The analysis and interpretation of such observations remain challenging, however, due to various complications, including the non-isothermality of the diffuse hot plasma and the inhomogeneity of the foreground absorption. We here illustrate such complications and a way to mitigate their effects by presenting an X-ray spectroscopy of the 30 Doradus nebula in the Large Magellanic Clouds, based on a 100 ks Suzaku observation. We measure the thermal and chemical properties of the hot plasma and quantitatively confront them with the feedback expected from embedded massive stars. We find that our spatially resolved measurements can be well reproduced by a global modeling of the nebula with a log-normal temperature distribution of the plasma emission measure and a log-normal foreground absorption distribution. The metal abundances and total mass of the plasma are consistent with the chemically enriched mass ejection expected from the central OB association and a ~55% mass-loading from the ambient medium. The total thermal energy of the plasma is smaller than what is expected from a simple superbubble model, demonstrating that important channels of energy loss are not accounted for. Our analysis indeed shows tentative evidence for a diffuse non-thermal X-ray component, indicating that cosmic-ray acceleration needs to be considered in such a young starburst region. Finally, we suggest that the log-normal modeling may be suitable for the X-ray spectral analysis of other giant HII regions, especially when spatially resolved spectroscopy is not practical.
Based on an analysis of data obtained with the Wide Field Camera 3 (WFC3) on the Hubble Space Telescope (HST) we report the identification of two distinct stellar populations in the core of the giant HII region 30Doradus in the Large Magellanic Cloud. The most compact and richest component coincides with the center of R136 and is ~1 Myr younger than a second more diffuse clump, located ~5.4 pc toward the northeast. We note that published spectral types of massive stars in these two clumps lend support to the proposed age difference. The morphology and age difference between the two sub-clusters suggests that an ongoing merger may occurring within the core of 30Doradus. This finding is consistent with the predictions of models of hierarchical fragmentation of turbulent giant molecular clouds, according to which star clusters would be the final products of merging smaller sub-structures.
We present deep Hubble Space Telescope (HST) NICMOS 2 F160W band observations of the central 56*57 (14pc*14.25pc) region around R136 in the starburst cluster 30 Dor (NGC 2070) located in the Large Magellanic Cloud. Our aim is to derive the stellar Initial Mass Function (IMF) down to ~1 Msun in order to test whether the IMF in a massive metal-poor cluster is similar to that observed in nearby young clusters and the field in our Galaxy. We estimate the mean age of the cluster to be 3 Myr by combining our F160W photometry with previously obtained HST WFPC2 optical F555W and F814W band photometry and comparing the stellar locus in the color-magnitude diagram with main sequence and pre-main sequence isochrones. The color-magnitude diagrams show the presence of differential extinction and possibly an age spread of a few megayears. We convert the magnitudes into masses adopting both a single mean age of 3 Myr isochrone and a constant star formation history from 2 to 4 Myr. We derive the IMF after correcting for incompleteness due to crowding. The faintest stars detected have a mass of 0.5 Msun and the data are more than 50% complete outside a radius of 5 pc down to a mass limit of 1.1 Msun for 3 Myr old objects. We find an IMF of dN/dlog(M) M^(-1.20+-0.2) over the mass range 1.1--20 Msun only slightly shallower than a Salpeter IMF. In particular, we find no strong evidence for a flattening of the IMF down to 1.1 Msun at a distance of 5 pc from the center, in contrast to a flattening at 2 Msun at a radius of 2 pc, reported in a previous optical HST study. We examine several possible reasons for the different results. If the IMF determined here applies to the whole cluster, the cluster would be massive enough to remain bound and evolve into a relatively low-mass globular cluster.