EDP Sciences
Free Access
Issue
A&A
Volume 585, January 2016
Article Number A30
Number of page(s) 17
Section Interstellar and circumstellar matter
DOI https://doi.org/10.1051/0004-6361/201525949
Published online 10 December 2015

© ESO, 2015

1. Introduction

Churchwell et al. (2006, 2007) identified more than 600 infrared (IR) dust bubbles in the Spitzer 8.0 μm-images of the Galactic Legacy Infrared Mid-Plane Survey Extraordinaire (GLIMPSE, Benjamin et al. 2003) between longitudes of 60°to +60°. These IR dust bubbles are bright at 8.0 μm, showing the emission of polycyclic aromatic hydrocarbons (PAHs) typical of photodissociation regions (PDRs), and enclose 24 μm emission, indicating the existence of warm dust inside (Watson et al. 2008). These characteristics indicate excitation sources (O- and/or B-type stars) that dissociate the molecular gas in the border of the PDRs and heat the dust through their UV photons. The dense shells surrounding many IR dust bubbles are believed to favor the formation of new generations of stars through the radiatively driven implosion (RDI) or the collect-and-collapse (CC) processes (Pomarès et al. 2009; Deharveng et al. 2009, 2012; Samal et al. 2014). Indeed, detailed studies of these bubbles have shown young stellar objects (YSOs) in their environs, although in some cases triggered star formation could not be proved (see, for example, Alexander et al. 2013; Dewangan & Ojha 2013).

thumbnail Fig. 1

Composite image of the IR dust bubble S 24 and its environs showing IRAC images: 3.6 μm is in blue, 4.5 μm is in green, and 8.0 μm is in red. The different features in the environment of S 24 are indicated.

Open with DEXTER

We selected the complex of IR dust bubbles S 21-S 24 identified by Churchwell et al. (2006) in the GLIMPSE images to perform a multiwavelength study of the interaction with their environs and search for YSOs in their vicinity. Here we report on the study of S 24. This bubble is centered at RA, Dec(J2000) = (16h52m20s, 44°2810′′) or (l, b) = (341°203, 0°23) and coincides with the IR source IRAS 16487-4423, which has colors typical of ultracompact Hii regions (Wood & Churchwell 1989).

A composite image of S 24 and its environs at 3.6, 4.5, and 8.0 μm is displayed in Fig. 1. At 3.6 μm the emission arises in a faint PAH feature at 3.3 μm and in dispersed stellar light, while at 8.0 μm most of the emission originates in strong features from PAHs. The 4.5 μm band shows emission from Brα and Pfβ, and vibrational H2 lines and ro-vibrational CO lines, typical of shocked gas (see Watson et al. 2008). At 8 μm the bubble presents a full filamentary shell of about 28′′ in radius, brighter toward the north and east, and displays a sharp eastern border. Two small and bright regions of extended emission can be identified at 8 μm in the image, named G341.220-0.213 at RA, Dec(J2000) = (16h52m18.25s, 44°2654′′), of about 10′′ in radius, and G341.217-0.237 at RA, Dec(J2000) = (16h52m23.74s, 44°2753′′), of about 5′′ in radius, placed to the north and east of S 24, respectively. At this wavelength, the shape of G341.220-0.213 resembles a horseshoe opened toward the east. The infrared dark cloud (IRDC) identified by Peretto & Fuller (2009) as SDC341.194-0.221 is evident to the west of the bubble where the emission at 8 μm is lacking. It is centered at RA, Dec(J2000) = (16h52m15.5s, 44°2819.4′′) and extends almost 2 in declination. Two IRDCs were also identified by Peretto & Fuller (2009): one at RA, Dec(J2000) = (16h52m15.4s, 44°306.9′′) (SDC 341.1710.241), to the south of the bubble, and the other one at RA, Dec(J2000) = (16h52m35.7s, 44°2821.8′′) (IRDC SDC 341.232-0.268). Finally, fainter emission can be detected in the whole region, encircling S 24, G341.22-0.213, and G341.217-0.237. An almost circular structure of about 16′′ in radius that we call G341.210-0.252, is also detected in the 8 μm image at RA, Dec(J2000) = (16h52m26.5s, 44°2850′′), to the southeast of S 24. These features are indicated in Fig. 1. The figure also shows that G341.220-0.213, G341.217-0.237, and part of the S 24 bubble coincide with sources at 4.5 μm.

Henning et al. (2001) observed the region at 1.2 mm using the SEST telescope with an angular resolution of 23′′. They detected cold dust emission from G341.220-0.213 and G341.217-0.237 and an extended halo linked to S 24, and estimated that the total mass in the region amounts to almost 5000 M (adopting a gas-to-dust ratio of 100 and a distance of 3.7 kpc).

Previous studies toward the region of S 24 have shown that star formation is active. Two spots of maser emission were detected. The first spot is a methanol and water maser emitter (Bayandina et al. 2012; Urquhart et al. 2013; Walsh et al. 2014) with peak velocities of between ~50 km s-1 and 38 km s-1 (Pestalozzi et al. 2005; Caswell et al. 2010; Voronkov et al. 2014; Walsh et al. 2014), coincident with G341.220-0.213. The second spot is also a methanol maser emitter (Vassilev et al. 2008; Bayandina et al. 2012) with peak velocities of between ~70 km s-1 and 16 km s-1 (Larionov & Val’tts 2008; Voronkov et al. 2014) located at RA, Dec(J2000) (16h52m159, 44°2836.̋7), coincident with the IRDC SDC341.194-0.221. Although a number of EGO candidates have been identified in the S21-S24 complex (Cyganowski et al. 2008), none of them appear projected in the close environs of the IR bubble.

S 24 is detected as a radio continuum source in the Molonglo Galactic Plane Survey at 843 MHz (MGPSJ165220-442802, Murphy et al. 2007; synthesized beam = 65′′× 45′′) with an integrated flux density S0.843 = 0.92 ± 0.04 Jy.

Bronfman et al. (1996) performed CS(21) line observations with an angular resolution of 50′′ using the SEST telescope. They observed the molecular line in one position toward IRAS16487-4423 (RA, Dec(J2000) = (16h52m21.24s, 44°282.̋5)) detecting CS emission with a velocity of 43.4 km s-1, which indicates dense gas toward the IR source. Russeil & Castets (2004) observed the CO(10) and CO(21) lines with an angular resolution of 42′′ and 21′′, respectively, toward a position located 4.́3 to the southeast of the S 24 bubble, detecting molecular gas with velocities in the range 40 km s-1 to 37 km s-1. All velocities in this paper are given with reference to the LSR.

Circular galactic rotation models predict that gas with velocities of 43.4 km s-1 lies at near and far kinematical distances of 3.6 kpc and 1213 kpc (see, for example, Brand & Blitz 1993). Similar distances can be derived from maser velocities. Following Henning et al. (2001), we adopt a distance of 3.7 kpc for the S 24 region and its environment. Taking into account a velocity dispersion of 6 km s-1 for the interstellar gas, we adopt a distance uncertainty of 0.5 kpc.

Here, we present the very first molecular line and dust continuum study toward S 24 and its environs aimed at determining the distribution of the molecular gas and the dust linked to the bubble and the correlation between them, analyzing their physical conditions, masses, and ambient densities, and investigating the status of star formation in the region. The dense gas linked to this region makes it particularly interesting to investigate the physics and kinematics of Hii regions and bubbles in dense media, allowing a better understanding of the evolution and star formation processes around these objects. Preliminary results were published in Cappa et al. (2013).

2. Data sets

2.1. Molecular line observations

The molecular gas characteristics were investigated by performing 13CO(32) line observations (at 330.58797 GHz) of a region of 3.́2 × 3.́2 during August 2009, and 12CO(21), 13CO(21), and C18O(21) line observations of a region of 5× 5 obtained in October 2010, with the APEX telescope using the APEX-1 and APEX-2 receivers, whose system temperatures are 150 K and 300 K, respectively.

The half-power beam-widths of the telescope are 30′′ (for the (21) lines) and 21′′ (for the (32) lines). The data were acquired with a FFT spectrometer, consisting of 4096 channels, with a total bandwidth of 1000 km s-1 and a velocity resolution of 0.33 km s-1. The region was observed in position-switching mode using the OTF technique with a space between dumps in the scanning direction of 9′′ for the (21) transition, and in position-switching mode in the (32) transition. The off-source position free of CO emission was located at RA, Dec(J2000) = (13h33m10.3s, 62°241′′).

Calibration was performed using Mars and X-TrA sources. Pointing was made twice during observations using X-TrA, o-Ceti, and VY-CMa. The intensity calibration has an uncertainty of 10%.

The spectra were reduced using the Continuum and Line Analysis Single-dish Software (CLASS) of the Grenoble Image and Line Data Analysis Software (GILDAS) working group1. A linear baseline fitting was applied to the data. The rms noise of the profiles after baseline subtraction and calibration is 0.35 K for the (21) and (32) transitions. The observed line intensities are expressed as main-beam brightness temperatures Tmb, by dividing the antenna temperature TA by the main-beam efficiency ηmb, equal to 0.72 for APEX-1 and 0.82 for APEX-2 (Vassilev et al. 2008). The Astronomical Image Processing System (AIPS) package and CLASS software were used to perform the analysis.

2.2. Dust continuum data

To trace cold dust emission we used far-infrared (FIR) images from the Herschel Space Observatory2 and from the APEX Telescope Large Area Survey of the Galaxy (ATLASGAL, Schuller et al. 2009).

2.2.1. Herschel data

The archival data correspond to the Hi-GAL key program (Hi-GAL:Herschel Infrared GALactic plane survey, Molinari et al. 2010, OBSIDs: 1342204094 and 1342204095).

The data have been taken in parallel mode with the instruments PACS (Poglitsch et al. 2010) at 70 and 160 μm, and SPIRE (Griffin et al. 2010) at 250, 350, and 500 μm. The chosen tile covers a field of 2.2 sq. deg and is approximately centered at (l, b) = (340°, 0°). The angular resolutions for the five photometric bands spans from 8′′ to 35.̋2 for 70 μm to 500 μm. Herschel interactive processing environment (HIPE v123, Ott 2010) was used to reduce the data, with reduction scripts from standard processing. To combine the two obsids (scan and cross-scan) of PACS archive data, we used the HIPE implementation script of the map merger called Scanmorphos. This script starts from Level 1 frames and creates a map from the data. The obtained pixel sizes are 2′′/px to blue and 3′′/px to red.

The zero-level of the PACS and SPIRE archive data is unknown, which may cause problems, for example, to derive dust temperatures. The zero-level offsets of the SPIRE maps were corrected as described in Bernard et al. (2010). A proper zero-level was set bearing in mind that Planck includes the same fraction of the sky as Hi-GAL data and taking into account that the two common photometric channels of Planck and SPIRE (857 GHz ~ 350 μm and 545 GHz ~ 550 μm) are calibrated with respect to FIRAS (Piat et al. 2002). We used the HIPE implementation script to run destriper gain corrections and to recalculate the absolute offset via cross-calibration with Planck-HFI data (HIPE v12, updated version by L. Conversi 2013).

To estimate fluxes, a Gaussian smoothing was applied to convolve all images at a low resolution of 500 μm (a FWHM of 35.̋2).

Assuming that an extended source has a spectral index αs = 4 (2 for the blackbody emission plus 2 for the opacity of the dust), the obtained fluxes of each source were multiplied by 0.98755, 0.98741, and 0.96787 for 250, 350, and 500 μm, respectively, to convert from monochromatic intensities of point sources to monochromatic extended sources.

2.2.2. ATLASGAL data

This survey was performed at 870 μm (345 GHz) with an angular resolution of 19.̋2 (HPBW), using the Large Apex Bolometer Camera (LABOCA; Siringo et al. 2009). The camera operates with 295 pixels at the APEX 12-m submillimeter telescope.

2.3. Complementary data

The millimeter and submillimeter data were complemented with Spitzer images at 3.6, 4.5, and 8.0 μm from the Galactic Legacy Infrared Mid-Plane Survey Extraordinaire (GLIMPSE; Benjamin et al. 2003), and at 24 μm from the MIPS Inner Galactic Plane Survey (MIPSGAL; Carey et al. 2005). We used radio continuum images at 843 MHz from the Molonglo Galactic Plane Survey (MGPS2, Murphy et al. 2007) with a synthesized beam of 65′′× 45′′, and at 1.4 GHz from the Southern Galactic Plane Survey (Haverkorn et al.2006; beam size = 1.́7).

To investigate the existence of YSO candidates projected onto the region, we analyzed infrared point sources from the Midcourse Space Experiment (MSX, Price et al. 2001), the Two Micron All Sky Survey (2MASS, Cutri et al. 2003), the Spitzer (Fazio et al. 2004), and the Wide-field Infrared Survey Explorer (WISE, Wright et al. 2010).

thumbnail Fig. 2

Overlay of the IRAC emissions at 8 μm (in red) and 3.6 μm (in green) and the LABOCA emission at 870 μm (in blue). Positions of identified YSOs are indicated with squares in different colors: Class I (red) and Class II (blue) sources from the WISE database, Class I (yellow), Class II (white), and Class I/II (green) sources from the Spitzer database, 2MASS sources (cyan), and MSX sources (magenta). Dust clumps from Contreras et al. (2013) are indicated.

Open with DEXTER

3. YSOs projected onto the region

Table 1

YSO candidates projected onto S 24 and its environs.

As mentioned in Sect. 1, maser emission linked to star-forming regions has been detected in the region. Bearing this in mind, we decided to search for YSO candidates using the MSX, 2MASS, WISE, and Spitzer point source catalogs. To accomplish this, we analyzed the IR sources in a region of 3 in radius centered at RA, Dec(J2000) = (16h52m20s, 44°2800′′). The results are included in Table 1, and their positions are shown in Fig. 2 with different symbols.

To identify YSO candidates in the MSX catalog, we followed the criteria of Lumsden et al. (2002), which take into account their loci in the (F21/F8, F14/F12) diagram. Two candidates were found, classified as massive YSOs (MYSO), one of them linked to G341.220-0.213 and the other one to the brightest section of the S 24 bubble.

To search for YSO candidates in the 2MASS catalog, we have constructed an [HKS, JH] diagram using only sources with photometric quality A or B. Six candidates were identified. The two upper panels of Fig. 3 show the 2MASS colorcolor (CC) and color-magnitude (CM) diagrams with the YSO candidates represented by red asterisks.

thumbnail Fig. 3

Upper panels: [HKs, JH] and [HKs, KS] diagrams of the 2MASS YSO candidates indicated by red crosses. Green asterisks correspond to main sequence stars projected onto the S 24 bubble (see Sect. 7.2). The dashed lines show the reddening vectors corresponding to M0 III and B2 V stars in the CC-diagram, and to B2 V stars in the CM-diagram. The squares are placed at intervals of ten magnitudes of visual extinction. Middle panel: location of YSO candidates from the Spitzer database. Crosses represent Class I and II sources. Bottom panel: YSO candidates from the WISE database. Class I and II sources are indicated in red and green, respectively.

Open with DEXTER

We have performed a search for candidates in the Spitzer database adopting the criteria by Allen et al. (2004), which allow distinguishing IR sources according to a class scheme: Class I are protostars surrounded by dusty infalling envelopes, and Class II are objects whose emission is dominated by accretion disks. Considering only sources detected in the four Spitzer-IRAC bands, we investigated their evolutionary stage by analyzing their location in a colorcolor diagram (middle panel of Fig. 3) following the color criteria determined empirically by Allen et al. (2004). Class II candidates lie in the region 0 [3.6][4.5] 0.8and 0.4 [5.8][8.0] 1.1, and their IR excess might be produced by accretion disks around the stellar objects. Sources outside the domain of Class II objects are identified as Class I candidate objects, and their IR excess might be due to circumstellar envelopes around the star. We identified 14 Class I candidates, 5 Class I/II candidates, and one Class II. The middle panel of Fig. 3 shows the position of the 20 Spitzer sources in the color-color [3.6][4.5] vs. [5.8][8.0] diagram. In this panel, the area delineated by thin lines corresponds to the locus of Class I objects, while the area within thick dashed lines corresponds to class II objects.

Finally, with the aim of identifying candidate YSOs using photometric data from the Wide-field Infrared Survey Explorer survey, which maps the whole sky in four bands centered at 3.4, 4.6, 12, and 22 μm, we used the criteria of Koenig et al. (2012), using [3.4][4.6] vs. [4.6][12] diagrams as follows: after removing contamination arising from background objects such as galaxies (very red in [4.6][12]), broad-line active galactic nuclei (of similar colors as YSOs, but distinctly fainter), and resolved PAH emission regions (redder than the majority of YSOs), we identified infrared excess sources by requiring that

where [3.4], [4.6], and [12.0] are the WISE bands 1, 2, and 3, respectively, and σ1 = σ( [3.4] − [4.6]) and σ2 = σ( [4.6] − [12.0]) indicate the combined errors of [3.4][4.6] and [4.6][12.0] colors, added in quadrature. Class I sources are a subsample of this defined by

The remaining sources are then Class II objects. Identified Class I and Class II candidates are shown in the bottom panel of Fig. 3. In this panel, the separation between the location of ClassI and Class II sources is indicated by the dashed lines.

As can be seen from Fig. 2, four candidates (one MYSO, one 2MASS source, one Spitzer Class I, and one WISE Class I) appear projected onto G341.220-0.213. This fact along with the presence of maser emission towards the region confirms that this is an active star-forming region. Three candidates coincide in position with the S 24 bubble, while only one WISE Class II candidate is projected onto G341.217-0.237. Most of the candidates appear to the east of the IRDC SDC341.193-0.221 and are projected onto emission at 8 μm. Finally, two Spitzer candidates coincide with the IRDC SDC341.194-0.221, and another three candidates appear projected onto the borders of G341.210-0.252.

It is clear from Table 1 that there is little correspondence among 2MASS, Spitzer, and WISE candidate YSOs (coincidences: #24 with #52; and #2 with #15 and #34). In spite of the similar wavelengths of two of the bands and in the different angular resolution of the Spitzer and WISE catalogs, an analysis of the coincidence of Spitzer and WISE YSOs identified in Table 1 shows that 33% of the WISE and Spitzer sources do not have a counterpart in the respective other catalog. Additionally, 35% of the WISE and Spitzer sources coincide with Spitzer and WISE sources, respectively, which are undetected in at least one band and cannot be properly classified. For the remaining 17%, the Spitzer and WISE sources coincide, but their counterparts are not classified as Class I or II.

A more detailed analysis is performed in connection to the molecular dust and cold dust in the next sections.

4. Distribution of the interstellar dust

The distribution of the interstellar dust was investigated using data at different wavelengths in the IR.

Figure 2 shows a superposition of the IRAC 8 μm (in red) and 3.6 μm (in green) emissions, and the ATLASGAL 870 μm emission (in blue). Emission at 870 μm reveals the presence of cold dust. This emission approximately encircles the S24 bubble. The dust clumps identified at this wavelength by Contreras et al. (2013) are indicated. AGAL341.216-00.236 and AGAL341.217-00.212 are bright, coincide with G341.217-0.237 and G341.220-0.213, respectively, and correlate with the emission at 1.2 mm detected by Henning et al. (2001). Weaker emission at 870 μm (AGAL SDC341.196-00.221) coincides in position with the IRDC SDC341.194-0.221 placed to the west of the bubble following its sharp eastern border, and is evident from the 8 μm emission in Fig. 2. Additionally, two cold dust clumps are located to the south and southeast of S 24, at RA, Dec(J2000) = (16h52m19.6s, 44°3038.̋5) (AGAL SDC341.172-00.239), and RA, Dec(J2000) = (16h52m31.71s, 44°2926.̋1) (AGAL SDC341.21100.271). AGAL SDC341.219-00.259, located at RA, Dec(J2000) = (16h52m32s, 44°2830′′), coincides with a section of the IRDC SDC341.232-0.268 and is being analyzed by Vasquez et al. (in prep.). G341.210-0.252 does not coincide with emission at 870 μm.

thumbnail Fig. 4

Composite image of the IRAC emission at 4.5 (in green) and 8.0 μm (in red), and the Herschel emission at 250 μm (in blue). Contours show the emission at 870 μm and correspond to 0.5 MJy sr-1 to 4.0 MJy sr-1 in steps of 0.5 MJy sr-1.

Open with DEXTER

Figure 4 displays an overlay of the emission at 870 μm (in contours) and the emissions at 4.5 (in green) and 8.0 μm (in red) from Spitzer and 250 μm (in blue) from Herschel. Dust clumps identified at 870 μm have counterparts at 250 μm, although the emission at 250 μm, which also shows the distribution of cold dust, appears more extended. The brightest dust clumps at 250 μm coincide with G341.220-0.213 and G341.217-0.237 and are also detected at 350 and 500 μm (the images are not shown here). The S 24 bubble appears clearly surrounded to the east and north by SPIRE emission. Herschel emission at these wavelengths is also projected onto the IRDC identified to the west of S 24, which can be defined as a Herschel bright IRDC (Peretto et al. 2010). Faint extended emission at 250 μm also coincides with the IRDC SDC341.1710.241, bordering the weak halo of 8 μm emission, and near RA, Dec(J2000) = (16h52m28s, 44°2930′′), in coincidence with a Spitzer point source identified at 4.5 μm. In summary, the correlation between 250 and 870 μm is excellent.

thumbnail Fig. 5

Upper panel: overlay of the PACS emission at 70 μm (contours) and the IRAC 8 μm emission (colorscale). Contours are 0.3 to 0.9 Jy px-1 in steps of 0.2 Jy px-1; 2 to 7 Jy px-1 in steps of 1 Jy px-1; and 9, 11, 13 to 33 Jy px-1 in steps of 4 Jy px-1. Bottom panel: overlay of the PACS emission at 160 μm (contours) and the IRAC 8 μm emission (colorscale). Contours are 2 to 7 Jy px-1 in steps of 1 Jy px-1; 10 to 25 Jy px-1 in steps of 3 Jy px-1; and 30 to 65 Jy px-1 in steps of 5 Jy px-1.

Open with DEXTER

thumbnail Fig. 6

Dust temperature map derived from the Herschel emission at 70 and 160 μm. The grayscale extends from 20 to 50 K. Contour levels correspond to 20 to 32 K in steps of 3 K, 35, 40, 50, and 60 K. The darkest regions indicate the highest dust temperatures.

Open with DEXTER

The PACS emission at 70 and 160 μm differs significantly from the emission at longer wavelengths (Fig. 5). The brightest source at these wavelengths coincides with G341.220-0.213. G341.217-0.237 also emits at 70 and 160 μm, while a bright source is also detected to the east of G341.217-0.237. Faint emission correlates with the S 24 bubble, mainly at 70 μm. This emission delineates the sharp western border of the bubble. The IRDC SDC 341.194-0.221 cannot be identified at either 70 μm or 160 μm. Emission at 70 μm is also projected onto G341.210-0.252, located at 16h52m26.5s, 44°2850′′, but is lacking at 160 μm and higher wavelengths. The presence of emission at 70 μm and its absence at 160 μm suggests the existence of dust at a higher temperature.

Dust temperatures Tdust for the region using the PACS images at 70 μm and 160 μm were estimated by convolving the image at 70 μm down to the angular resolution at 160 μm by applying color corrections, subtracting a background local to the S 24 bubble, and assuming that the emission is optically thin.

The color-temperature map was constructed as the inverse function of the ratio map of Herschel 70 μm and 160 μm color-and-background-corrected maps, that is, Tdust = f(T)-1 (see details in Preibisch et al. 2012; Ohlendorf et al. 2013). In the optically thin thermal dust emission regime f(T) has the following parametric form: (1)where Bν(ν,T) is the blackbody Planck function and βd corresponds to the spectral index of the thermal dust emission. The pixel-to-pixel temperature was calculated assuming βd = 2. This is a typical value adopted for irradiated regions (see Preibisch et al. 2012).

thumbnail Fig. 7

SED for G341.220-0.213 obtained using Herschel and ATLASGAL fluxes.

Open with DEXTER

The obtained dust temperature map is shown in Fig. 6. We were able to estimate dust temperatures for G341.220-0.213, G341.217-0.237, G341.210-0.252, and for the S 24 bubble. Dust temperatures using this method were almost impossible to obtain for the IRDCs due to their low emission at 70 μm. Values for G341.220-0.213 are in the range 2630 K, while for G341.217-0.237  temperatures are about 29 K. The image reveals a clear gradient in dust temperatures (from 26 to 60 K) for the S24 bubble: the higher the temperature, the closer to the center of the bubble. This is compatible with ionizing sources that are linked to S 24. Dust temperatures for S 24 are compatible with values derived by Watson et al. (2010) for the interior of IR dust bubbles based on images at 24 and 70 μm from MIPS and by Anderson et al. (2012) for Hii regions using Herschel data. Dust temperatures for G341.217-0.237 are about 28 K.

We also estimated Tdust for G341.220-0.213 from the whole set of Herschel data and the LABOCA image by fitting a Planck function. Fluxes at 70, 160, 250, 350, and 500 μm were calculated with HIPE using rectangular sky aperture photometry and convolving all images down to the angular resolution at 500 μm. For each aperture we measured the average surface brightness with an uncertainty derived from the standard deviation of the surface brightness within each aperture and the flux calibration uncertainty. Derived values are 776.0, 1047.2, 499.5, 213.3, and 69.3 Jy at 70, 160, 250, 350, and 500 μm, respectively. Figure 7 shows the spectral energy distribution (SED), which also includes the flux density at 870 μm obtained by Contreras et al. (2013; 14.63 ± 2.47 Jy), and the best fit (obtained using a graybody), which corresponds to Tdust = 28 ± 1 K. Dust temperature for G341.217-0.237 from an SED is difficult to determine. As a result of the poor angular resolution of the Herschel images at 350 and 500 μm, fluxes estimated from these data include part of S 24, which can give a false impression of the dust temperature.

Dust masses can be estimated from the expression (Hildebrand 1983) (2)where S870 is the flux density at 870 μm, d = 3.7 ± 0.5 kpc, κ870 = 1.0 cm2/gr is the dust opacity per unit mass (Ossenkopf & Henning 1994), and B870(Tdust) is the Planck function for a temperature Tdust.

Two processes may contribute to the emission at this wavelength in addition to the thermal emission from cold dust: molecular emission from the CO(32) line and free-free emission from ionized gas. The continuum emission contribution at 345 GHz due to ionized gas was estimated from the radio continuum image at 1.4 GHz considering that the radio emission is thermal (see Sect. 5) using the expression S345 = (345/1.4)-0.1S1.4. Adopting S1.4 = 1.2 Jy (see Sect. 5), this contribution amounts to about 1% of the emission at 870 μm. To estimate the contribution from CO(32), we took into account the intensity in the 13CO(32) line. This mechanism contributes less than 1%. Thus, the contribution by both mechanisms is within calibration uncertainties.

Adopting Tdust = 28 ± 1 K for G341.220-0.213, and a flux S870 = 14.63 ± 2.47 Jy derived by Contreras et al. (2013) for AGAL341.217-00.212, a dust mass Mdust1 = 12.9 ± 6.3M can be estimated for G341.220-0.213. Assuming the same dust temperature for G341.217-0.237and a flux S870 = 25.55 ± 4.11 Jy (for AGAL341.216-00.236), we estimate Mdust2 = 22.4 ± 11.0M. Finally, for the IRDC to the west of the bubble, we adopt temperatures in the range 1020 K (see Peretto et al. 2010), which, along with a flux S870 = 12.43 ± 2.14 Jy for AGAL341.196-00.221, results in Mdust−irdc = 17−57M. Uncertainties in these values are about 50%. Adopting a gas-to-dust ratio of 100, gas masses are 1300 M for G341.220-0.213, 2300 M for G341.217-0.237, and 17005700 M for the IRDC.

5. Ionized gas

thumbnail Fig. 8

Overlay of the radio continuum emission at 843 MHz (contours), the IRAC emission at 8 μm (in red), the MIPSGAL emission at 24 μm (in blue), and the IRAC emission at 3.6 μm (in green). Contours correspond to 10, 30, and 50 mJy beam-1, and from 100 to 500 mJy beam-1 in steps of 100 mJy beam-1.

Open with DEXTER

Figure 8 displays an overlay of the radio continuum image at 843 MHz (in contours) and the emissions at 8 μm and 24 μm (in colorscale). The image at 843 MHz reveals radio emission coincident with the sources, and in particular with the S 24 bubble. The emission at 1.4 GHz (not shown here) correlates with that at the lower frequency. We derived a flux density S1.4 = 1.2 Jy. Taking into account a flux density S0.843 = 0.92 Jy, a spectral index α = 0.5 (Sννα) can be estimated from the emission at both frequencies. The α-value indicates that at least at 843 MHz, the radio emission is optically thick, and that the source is thermal in nature.

thumbnail Fig. 9

Dec vs. RA map showing 12CO(21) (in red) and 13CO(21) (in black) profiles observed in a 5× 5 region centered in the S 24 bubble. Each profile shows Tmb in the interval from 1 K to 20 K vs. LSR velocity in the range from 80 km s-1 to 0 km s-1. The (0, 0) position corresponds to RA, Dec(J2000) = (16h52m20s, –44°2730′′). Note that RA inceases towards the right.

Open with DEXTER

Figure 8 also shows that S 24, as well as G341.220-0.213, G341.217-0.237, and G341.210-0.252 emit at 24 μm, indicating the presence of warm dust linked to the sources, and consequently, the existence of excitation sources inside them. Emission at this wavelength also coincides with the PDR, as revealed by the emission at 8 μm. In fact, the emission at 24 μm from Hii regions has two components: emission from very small grains out of thermal equilibrium, detected inside the Hii region, and thermal emission from the PDR from grains in thermal equilibrium (Anderson et al. 2012).

Lower limits for the electron density ne and the ionized mass Mi can be derived from the image at 843 MHz. For an estimated flux density of 0.92 Jy, using the classical expressions by (Mezger & Henderson 1967), a typical electron temperature of 104 K, and bearing in mind a He abundance of 10% by particle number and that He is single ionized, we derived ne> 200 cm-3, and Mi> 22M. We note that similar results can be obtained using the emission at 1.4 GHz. Clearly, the fact that the emission at 1.4 GHz is optically thick also can not be ruled out. Bearing in mind that the emission at 8 μm due to PAH in the photodissociation region marks the boundary of the ionized region, we measured a radius of 0.33 pc for the Hii region.

Although these results can not be used to determine either the ionized mass or the electron density, they reveal the existence of an Hii region.

6. Distribution of the molecular gas

Figure 9 displays the CO(21) and 13CO(21) spectra obtained for the region under study. The spatial separation between these profiles is 30′′. Relative coordinates are expressed in arcseconds, referred to RA, Dec(J2000) = (16h52m20s, 44°2730′′). The individual spectra show that the molecular emission appears concentrated between ~55 km s-1 and 20 km s-1, where a number of velocity components are present, and an isolated component near 10 km s-1 (which is not going to be taken into account). The CO(21)/13CO(21) line ratios are close to 11.5 in the central and northeastern parts of the figure, suggesting that the gas is optically thick at 13CO(21) where S 24, G341.220-0.213, and G341.217-0.237 are located.

The CO spectra of Fig. 10 illustrate the general behaviour of the molecular gas towards the entire region. There, we show the CO(21), 13CO(21), and C18O(21) spectra averaged within the observed region. At least three velocity components can be identified in the line of sight within the velocity interval from 55 km s-1 to 20 km s-1. The bulk of the molecular emission is detected in the three molecular lines and peaks at 45 km s-1, while fainter gas components are present at 38 km s-1 and 30 km s-1. The faintest component is present near 10 km s-1 in the 12CO(21) line only. The velocity components peaking at 38 km s-1 and 30 km s-1 can be barely identified in the 13CO(21) spectrum and are absent in the C18O(21) profile. Gas components having the same velocity can be also identified in the averaged J = 3−2 profile (not shown here).

Circular galactic rotation models (e.g. Brand & Blitz 1993) predict near and far kinematical distances in the range 3.23.9 kpc and 1213 kpc, respectively, for the gas component at 45 km s-1, while near and far kinematical distances in the range 2.43.5 and 1214 kpc, respectively, were obtained for gas in the components at 38 km s-1 and 30 km s-1. Particularly for the component at 45 km s-1, near kinematical distances are compatible with previously adopted values for the region of S 24.

thumbnail Fig. 10

CO(21) (in black), 13CO(21) (in red), and C18O(21) (in blue) spectra obtained by averaging the observed spectra in the region of S 24. Intensity is expressed as main-beam brightness-temperature.

Open with DEXTER

thumbnail Fig. 11

Channel maps of the 13CO(21) line (in grayscale) for the velocity interval from 49.9 km s-1 to 40.5 km s-1 in steps of 1.05 km s-1 overlaid onto the 8 μm emission (in contours). The grayscale extends from 0.5 K to 17.0 K.

Open with DEXTER

To illustrate the spatial distribution of the molecular gas we show in Fig. 11 the emission of the 13CO(21) in the interval 49.9 km s-1 to 40.5 km s-1, in steps of 1.05 km s-1 (in grayscale) superimpossed to the emission at 8 μm (in contours) for comparison. Most of the molecular emission has velocities in the interval from 46.8 km s-1 to 41.5 km s-1.

thumbnail Fig. 12

Panels A)D): overlay of the CO emission distribution in contours and the emission at 8 μm in grayscale towards G341.220-0.213. Panel A): 13CO(32) line emission in the interval from 43.8 km s-1 to 42.0 km s-1. Contours are from 5 K to 12 K (Tmb) in steps of 1 K. the different symbols mark the position of YSO candidates: stars: MSX sources, triangles: 2MASS sources, crosses: Spitzer sources, diamonds: WISE sources. Panel B): 13CO(21) line emission in the interval from 44.0 km s-1 to 41.9 km s-1. Contours are from 5 K to 11 K (Tmb) in steps of 1 K. Panel C): C18O(21) line emission in the interval from 43.8 km s-1 to 41.9 km s-1. Contours are from 1.5 K to 4 K (Tmb) in steps of 0.5 K. Panel D): 12CO(21) line emission in the interval from 45.7 km s-1 to 44.3 km s-1. Contours are from 9 K to 14 K (Tmb) in steps of 1 K.

Open with DEXTER

thumbnail Fig. 13

Left panel: SED of YSOs from Table 1 in G341.220-0.213. Input data are indicated by filled circles. The triangles indicate upper limits. The black line corresponds to the best fit. The fittings that obey Eq. (1) appear indicated by gray lines. The dashed line shows the emission of the stellar photosphere including foreground interstellar extinction. Right panel: SED of source #43 from Table 1 in G341.217-0.237.

Open with DEXTER

thumbnail Fig. 14

Upper panels: left: 13CO(32) emission distribution at 43.2 km s-1 showing the molecular envelope around the S 24 bubble. Contours are 3.0 K to 6.0 K (Tmb) in steps of 1.0 K, 8.0 K, 10.0 K, and 12.0 K. Right: overlay of the same contours as in the left panel and the IRAC emission at 8 μm. The symbols have the same meaning as in Fig. 12. Middle panels: Left: 13CO(21) emission distribution at 43.2 km s-1. Contours are from 4.0 K to 12.0 K (Tmb) in steps of 1.0 K. Right: overlay of the same contours as in the left panel and the IRAC emission at 8 μm. Bottom panels: left: C13O(21) emission distribution at 43.2 km s-1. Contours are from 1.0 K to 4.0 K (Tmb) in steps of 0.5 K, and 5 K. Right: overlay of the same contours as in the left panel and the IRAC emission at 8 μm.

Open with DEXTER

As regards the S 24 bubble, CO emission encircles the eastern and northern sections at 46.8 km s-1 and 45.7 km s-1. In the images at 44.6 km s-1, 43.6 km s-1, and 42.5 km s-1 this emission extends to the western section of the bubble, where the IRDC SDS 341.194-0.221is present. S 24 appears to be almost completely encircled by molecular gas in the images at 43.6 km s-1 and 42.5 km s-1, while at 41.5 km s-1 and 40.5 km s-1 the CO emission is present to the west of the bubble, coincident with the IRDC.

A close inspection of the emission in the interval from 48.9 km s-1 to 42.5 km s-1 reveals bright molecular emission linked to G341.217-0.237 and the eastern border of the IR bubble.

The CO emission associated with the different IR sources described above is analyzed in more detail in the following sections. Material linked to the IRDC SDC341.232-0.268 and present in the interval from 49.5 km s-1 to 42.1 km s-1 is being analyzed in detail by Vasquez et al. (in prep.).

7. Joining the details

In this section we investigate the gas distribution in S 24 in more detail and study the different features in its environment and its relation to the warm and cold dust and the YSO candidates.

7.1. G341.220-0.213

Panel A of Fig. 12 displays the 13CO(32) emission with velocities in the range 43.8 km s-1 and 42.0 km s-1 superimposed onto the emission at 8 μm for comparison. The emission distribution at these velocities reveals a molecular arc-like structure closely encircling the bright northern and western borders of G341.220-0.213. At 8 μm, G341.220-0.213, which resembles a horseshoe, is open opposite to the molecular emission. The IR source clearly interacts with the surrounding molecular gas and evolves in a medium with a density gradient. The coincidence of G341.220-0.213 with emission at 24 μm (see Sect. 5) suggests excitation sources inside. Bearing in mind the morphology of the source at 8 μm and in molecular lines, a champagne flow (Garay & Lizano 1999) cannot be discarded.

A number of YSO candidates appear inside the horseshoe: one MSX source classified as MYSO (#2 in Table 1); one Spitzer and one WISE source coincident in position, both classified as Class I (#15 and #34). We inspected the SED of sources #2, #15, and #34, which are assumed to be only one source, using the online4 tool developed by Robitaille et al. (2007), which can help to distinguish between evolved stars and reliable candidate YSOs. This tool fits radiation transfer models to observational data according to a χ2 minimization algorithm. Models that accomplished the following condition were selected: (3)where is the minimum χ2 value from all models, and n is the number of input data fluxes.

To perform the fitting, we used the photometric data listed in Table 1 along with fluxes derived from Herschel and ATLASGAL images, visual absorption values in the range 3040 mag, and distances in the range from 3.2 to 4.2 kpc. The result is shown in the left panel of Fig. 13. Fluxes obtained from Herschel and ATLASGAL were considered upper limits. The best fit suggests a massive central source of about 10 M, a disk mass of 5 × 10-4M, an envelope mass of 21 M, a total luminosity of 5300 L, and an age of 1 × 106 yr. With these characteristics, the evolutionary stage of the central source according to Robitaille et al. (2007) would correspond to Stage II (objects with optically thick disks). These results point to a massive central object.

We propose that this source has started modifying its environs by dissociating and then ionizing the nearby material, generating a PDR detected by its emission at 8 μm at the interface between the ionized and molecular material. It may have contributed to the heating of the dust as revealed by the emission at 24 μm. In addition, the detection of maser emission toward G341.220-0.213 (see Sect. 1) indicates that star formation is active.

A low-emission region is located at RA, Dec(J2000) = (16h52m22.5s, 44°2640′′), probably linked to a weak IR arc seen slightly to the north of that position.

Panels B and C of Fig. 12 show the emission distribution in the 13CO(21) and C18O(21) lines for the same velocity interval as panel A, while in panel D the emission corresponds to the CO(21) line in a slightly different velocity interval. The emission distribution reveals a molecular clump coincident with the G341.220-0.213. No signs of the arc-like structure are present in the (21) transitions, probably due to the lower angular resolution of these lines (30′′) in comparison with the one of the 13CO(32) (20′′).

The molecular clump detected in the interval [44.0, 42.0] km s-1 coincides with the emission of cold dust as imaged in the far-IR by LABOCA (AGAL 341.217-00.212) and Herschel (see Fig. 4), indicating that gas and dust are well mixed in the region.

The analysis of the data cubes allows us to identify molecular gas associated with G341.220-0.213 in the velocity interval from 46.5 km s-1 to 40.2 km s-1.

7.2. H II S 24

The upper left panel of Fig. 14 shows the 13CO(32) emission distribution at 43.2 km s-1 in grayscale and contours, while the upper right panel displays an overlay of the same CO contours and the emission at 8 μm. The middle and bottom panels show the emission at the same velocity for the 13CO(21) and C18O(21) lines, respectively.

The emission at this velocity reveals a U-shaped, almost complete molecular shell encircling the S24 bubble. The strongest emission regions of this molecular shell encircle the brightest sections of S 24 as seen at 8 μm. The shell appears open toward RA, Dec(J2000) = (16h52m19s, 44°2830′′). The emission in the western part of the shell coincides with the IRDC SDC341.194-0.221, which borders this section of the bubble. The molecular shell is extends throughout the interval from 46 km s-1 to 41.5 km s-1.

As shown in Fig. 11, in the velocity interval from 46.5 km s-1 to 43.5 km s-1 the molecular gas borders the eastern, northern, and western sections of S 24. Molecular gas at these velocities coincides with cold dust detected in the far IR at λ> 160 μm, as shown in Figs. 2 and 4. Molecular shells have been found around many IR dust bubbles, in particular those identified as Hii regions (e.g., Petriella et al. 2010; Pomarès et al. 2009; Deharveng et al. 2010). The presence of CO in the Hii region interior suggests that the shell corresponds to a three-dimensional structure. The appearance of the shell indicates that it is inhomogeneous, with the molecular gas distributed unisotropically (e.g., Anderson et al. 2015).

The molecular shell might be expanding at 23 km s-1 due to the difference in pressure between the ionized gas inside the bubble and the neutral gas outside, although no signs of expansion are detected in position-velocity diagrams. We note, however, that low expansion velocities such as those expected in this case are difficult to detect. In the case of expansion, the emission detected toward the interior of the region would correspond to the approaching and receding sections of the shell.

Three YSO candidates are projected onto the S 24 bubble: one MYSO (# 1 in Table 1) and two WISE sources identified as Class II (#35 and #36).

A number of YSOs also coincide in position with the molecular shell, suggesting that star formation is active in the environs of the bubble.

Two YSOs appear projected onto the IRDC SDC341.194-0.221. They do not coincide with the spot of maser emission described in Sect. 1 projected onto this IRDC.

In addition to these YSO candidates, three stars could be identified in the 2MASS catalog in the region under study. These stars are indicated with green asterisks in the [HKs, JH] and [HKs, Ks] diagrams of Fig. 3. In particular, 2MASS J17502070-4428012 is projected inside the bubble at RA, Dec(J2000) = (16h52m20.7s, 44°281.̋2). Its locus in the CC and CM diagrams (HKs) = 1.19 mag, (JH) = 2.18 mag, and Ks = 10.94 mag) is indicative of a massive star with an optical absorption AV ≃ 30 mag. Although more studies are necessary, this star might be responsible for dissociating and ionizing the gas, creating the observed PDR at 8 μm.

7.3. Interface between S 24 and G341.217-0.237

As described in Sect. 6, a bright molecular clump appears projected onto the interface between G341.217-0.237 and S 24 at RA, Dec(J2000) = (16h52m23s, 44°28). Figure 15 displays an enlargement of the region of the clump showing the emission in the 13CO(32) line at 47.4 km s-1 (in contours) and the 8 μm image in grayscale. The peak CO emission is coincident with the interface between the two sources. The anticorrelation between the CO and IR emissions strongly suggests that the molecular clump has modeled the eastern border of the PDR in the Hii region.

These facts suggest that S 24 and G341.217-0.237 are interacting with the molecular clump. The molecular clump coincides with emission at 870 μm and in the Herschel bands at 250 and 350 μm.

The YSO candidate # 43, classified as Class I, coincides with G341.217-0.237. The SED corresponding to this source is shown in the right panel of Fig. 13. To perform the fitting we used the photometric data listed in Table 1 and the flux derived from the image at 70 μm. Herschel fluxes at longer wavelengths include a significant part of the S 24 bubble and were not used for the SED. Visual absorption is in the range 3050 mag, and distances are from 3.2 to 4.2 kpc. The best fit corresponds to a central source of about 15 M, a disk mass of 8 × 10-3M, an envelope mass of 1.1 × 10-5M, a total luminosity of 22 000 L, and an age of 1.3 × 106 yr. According to this fitting, the object is probably in Stage III (objects with optically thin disks). However, as pointed out by Deharveng et al. (2012), Offner et al. (2012), and Robitaille (2008), these results should be taken with caution. The presence of emission at 24 μm coincident with G341.217-0.237 suggests that the ionizing photons from the massive source heat the dust and probably dissociate and ionize the molecular gas.

8. Parameters of the molecular gas

The molecular mass linked to G341.220-0.213, the clump at the interface between G341.217-0.237 and S 24 and the S 24 Hii region, and the molecular mass in the whole region were estimated from the emission in the J = 2 → 1 line. The derived parameters are listed in Table 2.

Assuming local thermodinamic equilibrium (LTE) conditions and that the emission in the CO(21) line is optically thick, we derived the excitation temperature Texc from the emission in the CO(21) line using (4)where = hν12/k, ν12 the frequency of the 12CO(21) line, and Tbg = 2.7 K. To obtain the peak main beam temperature in the CO(21) line (Tp−12CO) we took into account the averaged spectrum within regions of about 30′′ in diameter for the cases of G341.220-0.213 and the interface between G341.217-0.237 and S 24, while for the S 24 region and the whole region the spectra were averaged in a region of 2 and 3 in diameter, respectively.

The optical depth τ13 was obtained from the 13CO(21) line by assuming that the excitation temperature is the same for CO(21) and 13CO(21) emission lines using the expression (5)where = hν13/k, ν13 the frequency of the 13CO(21) line. Assuming LTE, the 13CO column density, Ntot(13CO), can be estimated from the 13CO(21) line data following (6)The integral of Eq. (6) can be approximated by (7)This approximation helps to eliminate to some extended optical depth effects and is good within 15% for τ< 2 (Rohlfs & Wilson 2004). Bearing in mind the τ13-values listed in Table 2, it is appropriate for our region. Then, the molecular mass was calculated using (8)where msun is the solar mass (~2 × 1033 g), μ is the mean molecular weight, which is assumed to be equal to 2.76 after allowing for a relative helium abundance of 25% by mass (Yamaguchi et al. 1999), mH is the hydrogen atom mass (~1.67 × 10-24 g), A is the solid angle of the CO emission, d is the adopted distance expressed in cm, and N(H2) is the H2 column density, obtained using an abundance N(H2)/N(13CO)= 5 × 105(Dickman 1978). Uncertainties in molecular masses and ambient densities are about 50% and 70%, respectively, and originate mainly in distance uncertainties and optically thick emission in the 13CO(21) line.

thumbnail Fig. 15

Overlay of the 13CO(32) emission distribution at –47.4 km s-1 (in contours) and the emission at 8 μm (in grayscale) showing the molecular clump at the interface between G341.217-0.237 and the S 24 bubble. Contours are from 3 K to 6 K (Tmb) in steps of 1 K. The symbols are the same as in Fig. 12.

Open with DEXTER

To explore the questions of temperature and density on G341.220-0.213 and G341.217-0.237 in depth, we performed a large velocity gradient (LVG) analysis (Scoville & Solomon 1973; Goldreich & Kwan 1974) for radiative transfer of molecular emission lines. We used the lvg task implemented as part of the MIRIAD5 package of SMA. For a given column density (normalized by the line width), this program estimates the line radiation temperature of a molecular transition as a function of the kinetic temperature (Tk) and the H2 volume density (nH2). We generated 50 × 50 model grids for the 13CO(32), 13CO(21), and CO(21) lines with Tk in the range 1300 K and nH2 with the range 1 × 103 to 1 × 108 cm-3. In all cases we convolved the 13CO(32) data with a Gaussian of FWHM 30′′ to smooth the cubes down to the angular resolution of the 13CO(21) data. We then derived peak main beam temperatures and line widths for the studied components by fitting Gaussian functions to the line profiles. We then calculated two ratios, R(3−2)/(2−1) = Tmb13CO(32)/CO(21) and R12/13 = Tmb13CO(21)/CO(21). The input column densities of CO and 13CO were estimated from the dust-derived column density measured from the 870 μm emission in each region, and assuming relative abundances N(CO)/N(H2) = 1 × 10-4 and N(13CO)/N(H2) = 2 × 10-6 (Dickman 1978). The same procedure was applied to the IRDC SDC341.194-0.221, for which we did not derive molecular mass and ambient density since its molecular emission is included in the S 24 molecular shell.

Figure 16 shows the solutions for G341.220-0.213, G341.217-0.237, and the IRDC. They indicate that kinetic temperatures in the IRDC are about 20 K, not different than derived in other IRDCs (Egan et al. 1998), and not much higher than typical temperatures of molecular gas without an extra heat source other than interstellar radiation field. The volume density of the IRDC is about 104 cm-3. The temperatures for G341.220-0.213, G341.217-0.237 are slightly higher (~3040 K), as expected, because they are active star-forming regions (see Sect. 7). Their volume densities are less well constrained, but values higher than 4 × 104 cm-3 are inferred from the plots.

Table 2 shows that molecular gas densities are in the range 104−105 for particular regions including the shell around the S 24 Hii region. Masses derived from molecular data are compatible, within errors, with those obtained from the emission at 870 μm, assuming a gas-to-dust ratio equal to 100. Additionally, molecular ambient densities are compatible with values derived from LVG analysis. The large difference for G341.217-0.237 arises because the mass estimate from molecular emission only corresponds to the emission at the interface between G341.217-0.237 and the S 24 Hii region. The ambient density obtained by distributing the total molecular mass in the whole observed region is high (5900 cm-3), indicating that G341.220-0.213, G341.217-0.237  and S 24 are evolving in a high-density ambient medium.

The high opacity estimated for G341.220-0.213 is expected considering the high column density of this source.

thumbnail Fig. 16

LVG plots for G341.220-0.213, G341.217-0.237, and the IRDC SDC341.194-0.221. The green lines show the solutions for the ratios 13CO(32)/13CO(21), red lines those for 13CO(21)/CO(21). The central lines correspond to the observed ratios. The outer lines indicate ~10% calibration uncertainties.

Open with DEXTER

Table 2

Molecular gas parameters.

9. Triggered star formation scenario in S 24?

We have found evidence for ongoing star formation in the dense molecular envelope encircling the S 24 Hii region. Were these YSOs triggered by the collect-and-collapse (CC) mechanism? To answer this question, we can apply the analytical model by Whitworth et al. (1994). For the Hii regions, the model predicts the age of the Hii region at which the fragmentation occurs (the fragmentation timescale), tfrag, the size of the Hii region at that moment, Rfrag, the mass of the fragments, Mfrag, and their separation along the compressed layer, rfrag. The parameters required to derive these quantities are the UV photon flux of the exciting star, NLy, the ambient density of the surrounding medium into which the Hii region is evolving, n0, and the isothermal sound speed in the shocked gas, as.

Since the exciting star is unknown, we took into account a wide range of spectral types, that is, from O5V to O9V stars, with UV fluxes in the range NLy = (180.8) × 1048 s-1 (Martins et al. 2002). Using the mean H2 ambient density nH2 = 5900 cm-3 (see Table 2), and as = 0.2−0.6 km s-1, we obtained tfrag = (6.68.8) × 105 yr, Rfrag = 1.8−2.3 pc, Mfrag = 22−29M, and, rfrag = 0.5−0.4 pc.

The dynamical age of the Hii region can be estimated using the equation (Dyson & Williams 1997) (9)where RS is the original Strömgren radius, equal to 0.090.25 pc for the adopted spectral types, and cs is the sound velocity in the ionized gas. Derived dynamical ages span the range (943) × 103 yr. We find that the dynamical age is younger than the fragmentation timescale tfrag for the adopted ambient density. The comparison of these two quantities does not support a collect-and-collapse process, which is a relatively slow process, for the triggering of star formation in the envelope. An RDI scenario could be investigated, but evidence of this process (such as pillars) appear to be absent. Very probably, the compact Hii region is too young for triggering to have begun.

10. Conclusions

Based on the molecular emission in the CO(21), 13CO(21), C18O(21), and 13CO(32) lines that were obtained with the APEX telescope and images in the near-, mid-, and far-IR from IRAC-Glimpse, Herschel, and ATLASGAL, we performed a multiwavelength analysis of the infrared dust bubble S 24 and two extended IR sources (G341.220-0.213 and G341.217-0.237) located in its environs. The region coincides with the IRAS point source 16487-4423, which is classified as UCHII. We also investigated the presence of YSO candidates in the region using the MSX, 2MASS, Spitzer, and WISE catalogs.

The molecular emission distribution shows that gas linked to the S 24 bubble, G341.220-0.213, and G341.217-0.237 has velocities between 48.0 km s-1 and 40.0 km s-1, compatible with the kinematical distance of 3.7 kpc previously adopted for the region.

The gas distribution reveals a shell-like molecular structure of ~0.8 pc in radius detected in the velocity interval of 46 km s-1 to 41.5 km s-1, encircling the S 24 bubble. A cold dust counterpart of this shell is detected at wavelengths longer than 160 μm, that is, in the LABOCA and SPIRE images. In contrast, PACS emission at 70 μm and MIPS emission at 24 μm from warm dust appears to be projected inside the bubble. Dust temperatures derived from the emission at 70 and 160 μm are in the range 26 to 60 K, with the higher values closer to the center of the bubble. The detection of radio continuum emission and the warm dust indicate exciting sources and that the bubble is a compact Hii region. We estimated a molecular gas and H2 ambient density in the shell of 3100 M and 2.5 × 104 cm-3. A search for excitation sources allowed identifying a massive star candidate projected onto the Hii region, which would provide the ultraviolet photons to heat the gas and dissociate and ionize the gas. A number of YSO candidates appear to be projected onto the molecular shell.

Part of the molecular gas bordering the S 24 bubble coincides with the extended IRDC SDC341.194-0.221, for which we derived a kinetic temperature of about 20 K and a volume density of 104 cm-3, based on LVG analysis.

For the source G341.220-0.213, which resembles a horseshoe open toward the east, the 13CO(32) line emission allowed detecting an arc-like structure with velocities in the range 43.8 km s-1 to 42.0 km s-1, encircling the brightest sections of the IR source. The morphology indicates that the molecular gas interacts with the IR source. A dust counterpart is detected in the Herschel (SPIRE and PACS) and LABOCA images. The spectral energy distribution constructed using Herschel and ATLASGAL fluxes allowed deriving a dust temperature of 28 ± 1 K. A molecular mass of 1550 M is linked to this source, compatible with the mass derived from ATLASGAL. The H2 ambient density amounts to 6.6 × 104 cm-3, in agreement with the ambient density value estimated from the LVG analysis. A number of IR point sources with characteristics of YSOs coincide with this region, which along with the maser emission indicates that star formation is active. The SED for the derived fluxes suggests a massive central source of 10 M, which would be responsible for heating the dust and dissociating the molecular gas.

The source G341.217-0.237 also coincides with molecular gas and cold dust. The line emission allowed detecting a molecular clump at the interface between the S 24 Hii region and G341.217-0.237, shaping the eastern border of the IR bubble. A high ambient density (4.2 × 105 cm-3) was estimated for the clump.

The total mass of molecular gas in the region and the H2 ambient density amount to 10 300 M and 5900 cm-3, indicating that G341.220-0.213, G341.217-0.237, and the S 24 Hii region are evolving into a high-density region.

Finally, we investigated whether a collect-and-collapse scenario can explain the YSOs in the dense envelope around the S 24 Hii region. We conclude that neither this mechanism nor the RDI process seem to occur in this region. Very probably, the Hii region is too young for triggering to have begun.


2

Herschel is an ESA space observatory with science instruments provided by European-led Principal Investigator consortia and with important participation from NASA.

3

HIPE is a joint development by the Herschel Science Ground Segment Consortium, consisting of ESA, the NASA Herschel Science Center, and the HIFI, PACS and SPIRE consortia members, see http://herschel.esac.esa.int/HerschelPeople.shtml

Acknowledgments

We acknowledge the anonymous referee for their comments. The ATLASGAL project is a collaboration between the Max-Planck-Gesellschaft, the European Southern Observatory (ESO) and the Universidad de Chile. C.E.C. acknowledges the kind hospitality of M. Rubio and her family during her stay in Chile, and C. Kobulnicky for having suggested studying this object. L.C. acknowledges support from DIULS. V.F. acknowledges support from ESO-Chile Joint Committee and DIULS and would like to thank Ivan Valtchanov, Bruno Altieri, and Luca Conversi for their support and valuable assistance in Herschel data processing. V.F. also acknowledges support from the Faculty of the European Space Astronomy Centre (ESAC). This project was partially financed by CONICET of Argentina under project PIP 0356, UNLP under project 11/G120, and CONICyT of Chile through FONDECYT grant No. 1140839. This work is based [in part] on observations made with the Spitzer Space Telescope, which was operated by the Jet Propulsion Laboratory, California Institute of Technology under a contract with NASA. This publication makes use of data products from the Two Micron All Sky Survey, which is a joint project of the University of Massachusetts and the Infrared Processing and Analysis Center/California Institute of Technology, funded by the National Aeronautics and Space Administration and the National Science Foundation. The MSX mission was sponsored by the Ballistic Missile Defense Organization (BMDO).

References

All Tables

Table 1

YSO candidates projected onto S 24 and its environs.

Table 2

Molecular gas parameters.

All Figures

thumbnail Fig. 1

Composite image of the IR dust bubble S 24 and its environs showing IRAC images: 3.6 μm is in blue, 4.5 μm is in green, and 8.0 μm is in red. The different features in the environment of S 24 are indicated.

Open with DEXTER
In the text
thumbnail Fig. 2

Overlay of the IRAC emissions at 8 μm (in red) and 3.6 μm (in green) and the LABOCA emission at 870 μm (in blue). Positions of identified YSOs are indicated with squares in different colors: Class I (red) and Class II (blue) sources from the WISE database, Class I (yellow), Class II (white), and Class I/II (green) sources from the Spitzer database, 2MASS sources (cyan), and MSX sources (magenta). Dust clumps from Contreras et al. (2013) are indicated.

Open with DEXTER
In the text
thumbnail Fig. 3

Upper panels: [HKs, JH] and [HKs, KS] diagrams of the 2MASS YSO candidates indicated by red crosses. Green asterisks correspond to main sequence stars projected onto the S 24 bubble (see Sect. 7.2). The dashed lines show the reddening vectors corresponding to M0 III and B2 V stars in the CC-diagram, and to B2 V stars in the CM-diagram. The squares are placed at intervals of ten magnitudes of visual extinction. Middle panel: location of YSO candidates from the Spitzer database. Crosses represent Class I and II sources. Bottom panel: YSO candidates from the WISE database. Class I and II sources are indicated in red and green, respectively.

Open with DEXTER
In the text
thumbnail Fig. 4

Composite image of the IRAC emission at 4.5 (in green) and 8.0 μm (in red), and the Herschel emission at 250 μm (in blue). Contours show the emission at 870 μm and correspond to 0.5 MJy sr-1 to 4.0 MJy sr-1 in steps of 0.5 MJy sr-1.

Open with DEXTER
In the text
thumbnail Fig. 5

Upper panel: overlay of the PACS emission at 70 μm (contours) and the IRAC 8 μm emission (colorscale). Contours are 0.3 to 0.9 Jy px-1 in steps of 0.2 Jy px-1; 2 to 7 Jy px-1 in steps of 1 Jy px-1; and 9, 11, 13 to 33 Jy px-1 in steps of 4 Jy px-1. Bottom panel: overlay of the PACS emission at 160 μm (contours) and the IRAC 8 μm emission (colorscale). Contours are 2 to 7 Jy px-1 in steps of 1 Jy px-1; 10 to 25 Jy px-1 in steps of 3 Jy px-1; and 30 to 65 Jy px-1 in steps of 5 Jy px-1.

Open with DEXTER
In the text
thumbnail Fig. 6

Dust temperature map derived from the Herschel emission at 70 and 160 μm. The grayscale extends from 20 to 50 K. Contour levels correspond to 20 to 32 K in steps of 3 K, 35, 40, 50, and 60 K. The darkest regions indicate the highest dust temperatures.

Open with DEXTER
In the text
thumbnail Fig. 7

SED for G341.220-0.213 obtained using Herschel and ATLASGAL fluxes.

Open with DEXTER
In the text
thumbnail Fig. 8

Overlay of the radio continuum emission at 843 MHz (contours), the IRAC emission at 8 μm (in red), the MIPSGAL emission at 24 μm (in blue), and the IRAC emission at 3.6 μm (in green). Contours correspond to 10, 30, and 50 mJy beam-1, and from 100 to 500 mJy beam-1 in steps of 100 mJy beam-1.

Open with DEXTER
In the text
thumbnail Fig. 9

Dec vs. RA map showing 12CO(21) (in red) and 13CO(21) (in black) profiles observed in a 5× 5 region centered in the S 24 bubble. Each profile shows Tmb in the interval from 1 K to 20 K vs. LSR velocity in the range from 80 km s-1 to 0 km s-1. The (0, 0) position corresponds to RA, Dec(J2000) = (16h52m20s, –44°2730′′). Note that RA inceases towards the right.

Open with DEXTER
In the text
thumbnail Fig. 10

CO(21) (in black), 13CO(21) (in red), and C18O(21) (in blue) spectra obtained by averaging the observed spectra in the region of S 24. Intensity is expressed as main-beam brightness-temperature.

Open with DEXTER
In the text
thumbnail Fig. 11

Channel maps of the 13CO(21) line (in grayscale) for the velocity interval from 49.9 km s-1 to 40.5 km s-1 in steps of 1.05 km s-1 overlaid onto the 8 μm emission (in contours). The grayscale extends from 0.5 K to 17.0 K.

Open with DEXTER
In the text
thumbnail Fig. 12

Panels A)D): overlay of the CO emission distribution in contours and the emission at 8 μm in grayscale towards G341.220-0.213. Panel A): 13CO(32) line emission in the interval from 43.8 km s-1 to 42.0 km s-1. Contours are from 5 K to 12 K (Tmb) in steps of 1 K. the different symbols mark the position of YSO candidates: stars: MSX sources, triangles: 2MASS sources, crosses: Spitzer sources, diamonds: WISE sources. Panel B): 13CO(21) line emission in the interval from 44.0 km s-1 to 41.9 km s-1. Contours are from 5 K to 11 K (Tmb) in steps of 1 K. Panel C): C18O(21) line emission in the interval from 43.8 km s-1 to 41.9 km s-1. Contours are from 1.5 K to 4 K (Tmb) in steps of 0.5 K. Panel D): 12CO(21) line emission in the interval from 45.7 km s-1 to 44.3 km s-1. Contours are from 9 K to 14 K (Tmb) in steps of 1 K.

Open with DEXTER
In the text
thumbnail Fig. 13

Left panel: SED of YSOs from Table 1 in G341.220-0.213. Input data are indicated by filled circles. The triangles indicate upper limits. The black line corresponds to the best fit. The fittings that obey Eq. (1) appear indicated by gray lines. The dashed line shows the emission of the stellar photosphere including foreground interstellar extinction. Right panel: SED of source #43 from Table 1 in G341.217-0.237.

Open with DEXTER
In the text
thumbnail Fig. 14

Upper panels: left: 13CO(32) emission distribution at 43.2 km s-1 showing the molecular envelope around the S 24 bubble. Contours are 3.0 K to 6.0 K (Tmb) in steps of 1.0 K, 8.0 K, 10.0 K, and 12.0 K. Right: overlay of the same contours as in the left panel and the IRAC emission at 8 μm. The symbols have the same meaning as in Fig. 12. Middle panels: Left: 13CO(21) emission distribution at 43.2 km s-1. Contours are from 4.0 K to 12.0 K (Tmb) in steps of 1.0 K. Right: overlay of the same contours as in the left panel and the IRAC emission at 8 μm. Bottom panels: left: C13O(21) emission distribution at 43.2 km s-1. Contours are from 1.0 K to 4.0 K (Tmb) in steps of 0.5 K, and 5 K. Right: overlay of the same contours as in the left panel and the IRAC emission at 8 μm.

Open with DEXTER
In the text
thumbnail Fig. 15

Overlay of the 13CO(32) emission distribution at –47.4 km s-1 (in contours) and the emission at 8 μm (in grayscale) showing the molecular clump at the interface between G341.217-0.237 and the S 24 bubble. Contours are from 3 K to 6 K (Tmb) in steps of 1 K. The symbols are the same as in Fig. 12.

Open with DEXTER
In the text
thumbnail Fig. 16

LVG plots for G341.220-0.213, G341.217-0.237, and the IRDC SDC341.194-0.221. The green lines show the solutions for the ratios 13CO(32)/13CO(21), red lines those for 13CO(21)/CO(21). The central lines correspond to the observed ratios. The outer lines indicate ~10% calibration uncertainties.

Open with DEXTER
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.