No Arabic abstract
Understanding the properties of dust produced during the asymptotic giant branch phase of stellar evolution is important for understanding the evolution of stars and galaxies. Recent observations of the carbon AGB star R Scl have shown that observations at far-infrared and submillimetre wavelengths can effectively constrain the grain sizes in the shell, while the total mass depends on the structure of the grains (solid vs. hollow or fluffy). We aim to constrain the properties of the dust observed in the submillimetre in the detached shells around the three carbon AGB stars U Ant, DR Ser, and V644 Sco, and to investigate the constraints on the dust masses and grain sizes provided by far-infrared and submm observations. We observed the carbon AGB stars U Ant, DR Ser, and V644 Sco at 870 micron using LABOCA on APEX. Combined with observations from the optical to far-infrared, we produced dust radiative transfer models of the spectral energy distributions (SEDs) with contributions from the stars, present-day mass-loss and detached shells. We tested the effect of different total dust masses and grain sizes on the SED, and attempted to consistently reproduce the SEDs from the optical to the submm. We derive dust masses in the shells of a few 10e-5 Msun, assuming spherical, solid grains. The best-fit grain radii are comparatively large, and indicate the presence of grains between 0.1 micron-2 micron. The LABOCA observations suffer from contamination from 12CO(3-2), and hence gives fluxes that are higher than the predicted dust emission at submm wavelengths. We investigate the effect on the best-fitting models by assuming different degrees of contamination and show that far-infrared and submillimetre observations are important to constrain the dust mass and grain sizes in the shells.
Detached shells are believed to be created during a thermal pulse, and constrain the time scales and physical properties of one of the main drivers of late stellar evolution. We aim at determining the morphology of the detached dust shells around the carbon AGB stars R Scl and V644 Sco, and compare this to observations of the detached gas shells. We observe the polarised, dust-scattered stellar light around these stars using the PolCor instrument mounted on the ESO 3.6m telescope. Observations were done with a coronographic mask to block out the direct stellar light. The polarised images clearly show the detached shells. Using a dust radiative transfer code to model the dust-scattered polarised light, we constrain the radii and widths of the shells to 19.5 arcsec and 9.4 arcsec for the detached dust shells around R Scl and V644 Sco, respectively. Both shells have an overall spherical symmetry and widths of approx. 2 arcsec. For R Scl we can compare the observed dust emission directly with high spatial-resolution maps of CO(3-2) emission from the shell observed with ALMA. We find that the dust and gas coincide almost exactly, indicating a common evolution. The data presented here for R Scl are the most detailed observations of the entire dusty detached shell to date. For V644 Sco these are the first direct measurements of the detached shell. Also here we find that the dust most likely coincides with the gas shell. The observations are consistent with a scenario where the detached shells are created during a thermal pulse. The determined radii and widths will constrain hydrodynamical models describing the pre-pulse mass loss, the thermal pulse, and post-pulse evolution of the star.
Detached circumstellar dust shells are detected around three carbon variables using Herschel-PACS. Two of them are already known on the basis of their thermal CO emission and two are visible as extensions in IRAS imaging data. By model fits to the new data sets, physical sizes, expansion timescales, dust temperatures, and more are deduced. A comparison with existing molecular CO material shows a high degree of correlation for TT Cyg and U Ant but a few distinct differences with other observables are also found.
The origin of the arc-shaped Sh2-296 nebula is still unclear. Mainly due to its morphology, the nebula has been suggested to be a 0.5 Myr-old supernova remnant (SNR) that could be inducing star formation in the CMa OB1 association. We aim to show, for the first time, that the nebula is part of a large, shell-like structure, which we have designated the ``CMa shell, enclosing a bubble created by successive supernova (SN) explosions. We identified three runaway stars, associated with bow-shock structures, in the direction of the CMa shell and we investigate the possibility that they have originated in the center of the shell. By analyzing images of the CMa OB1 association at several wavelengths, we clearly see that the Sh2-296 nebula is in fact part of a large structure, which can be approximated by a large (with a diameter of ~60 pc) elliptical shell. Using the recent Gaia-DR2 astrometric data, we trace back the path of the three runaway stars, in order to find their original position in the past, with relation to the CMa shell. We also revise the heating and ionization of the Sh2-296 nebula, by comparing the photon budget provided by the O stars in the region with results from radio observations. We find that the runaway stars have likely been ejected from a Trapezium-like progenitor cluster on three successive SN explosions having taken place ~6, ~2 and ~1 Myr ago. We also show that the few late-type O stars in the region cannot explain the ionization of the Sh~2-296 nebula and other mechanisms need to be at work. We argue that, though we now have evidence for several SNe events in the CMa OB1 association, the SNe probably played a minor role in triggering star formation in these clouds. In contrast, the CMa OB1 association, as it is now, likely testifies to the last stages of a star-forming region.
We examine the recent star formation associated with four supergiant shells (SGSs) in the Large Magellanic Cloud (LMC): LMC 1, 4, 5, and 6, which have been shown to have simple expanding-shell structures. H II regions and OB associations are used to infer star formation in the last few Myr, while massive young stellar objects (YSOs) reveal the current ongoing star formation. Distributions of ionized, H I, and molecular components of the interstellar gas are compared with the sites of recent and current star formation to determine whether triggering has taken place. We find that a great majority of the current star formation has occurred in gravitationally unstable regions, and that evidence of triggered star formation is prevalent at both large and local scales.
We present the results of ALMA band 7 observations of dust and CO gas in the disks around 7 objects with spectral types ranging between M5.5 and M7.5 in Upper Scorpius OB1, and one M3 star in Ophiuchus. We detect unresolved continuum emission in all but one source, and the $^{12}$CO J=3-2 line in two sources. We constrain the dust and gas content of these systems using a grid of models calculated with the radiative transfer code MCFOST, and find disk dust masses between 0.1 and 1 M$_oplus$, suggesting that the stellar mass / disk mass correlation can be extrapolated for brown dwarfs with masses as low as 0.05 M$_odot$. The one disk in Upper Sco in which we detect CO emission, 2MASS J15555600, is also the disk with warmest inner disk as traced by its H - [4.5] photometric color. Using our radiative transfer grid, we extend the correlation between stellar luminosity and mass-averaged disk dust temperature originally derived for stellar mass objects to the brown dwarf regime to $langle T_{dust} rangle approx 22 (L_{*} /L_{odot})^{0.16} K$, applicable to spectral types of M5 and later. This is slightly shallower than the relation for earlier spectral type objects and yields warmer low-mass disks. The two prescriptions cross at 0.27 L$_odot$, corresponding to masses between 0.1 and 0.2 M$_odot$ depending on age.