2012A&A...540A.104M

Infrared dark clouds (IRDCs) are promising sites to study the earliest formation stages of stellar clusters and high-mass stars, and the physics of molecular-cloud formation and fragmentation. We attempt to improve our understanding of the physical and chemical properties of the filamentary IRDC G304.74+01.32 (hereafter, G304.74). In particular, we investigate the kinematical and dynamical state of the cloud and clumps within it, and the amount of CO depletion. All of the submillimetre peak positions in the cloud identified from our previous LABOCA 870-µm map were observed in C¹⁷O(2-1) with APEX. These are the first line observations along the whole filament that have been made so far. Selected positions were also observed in the ¹³CO(2-1), SiO(5-4), and CH₃OH(5_k-4_k) transitions at ∼1mm. The C¹⁷O lines were detected towards all target positions at similar radial velocities. CO does not appear to be significantly depleted in the clumps, the largest depletion factors being only about 2. Two to three methanol 5_k-4_k lines near ∼241.8 GHz were detected towards all selected positions, whereas SiO(5-4) was seen in only one of these positions, namely SMM 3. In the band covering SiO(5-4), we also detected the DCN(3-2) line towards SMM 3. The ¹³CO(2-1) lines display blue asymmetric profiles, which are indicative of large-scale infall motions. The clumps show transonic to supersonic non-thermal motions, and a virial-parameter analysis suggests that most of them are gravitationally bound. The external pressure may also play a non-negligible role in the dynamics. Our analysis suggests that the fragmentation of the filament into clumps is caused by a ``sausage''-type instability, in agreement with results from other IRDCs. The uniform C¹⁷O radial velocities along the G304.74 cloud shows that it is a coherent filamentary structure. Although the clumps appear to be gravitationally bound, the ambient turbulent ram pressure may be an important factor in the cloud dynamics. This is qualitatively consistent with our earlier suggestion that the filament was formed by converging supersonic turbulent flows. The poloidal magnetic field could resist the radial cloud collapse, which conforms to the low infall velocites that we derived. The cloud may be unable to form high-mass stars based on the mass-size threshold. The star-formation activity in the cloud, such as outflows, is likely responsible for the release of CO from the icy grain mantles back into the gas phase. Shocks related to outflows may also have injected CH₃OH, SiO, and DCN into the gas-phase in SMM 3.