IRAS 16547-4247 (G343.12-0.06) is the most luminous young stellar object (YSO) known to harbour a thermal radio jet. At a distance of 2.9 kpc, the YSO has a bolometric luminosity of 6.2 × 10⁴ L_Sun, equivalent to that of a single O8 zero-age main sequence (ZAMS) star. The jet was first detected by Garay et al. (2003) using the Compact Array at 1.4, 2.5, 4.8 and 8.6 GHz and subsequently observed with the VLA at 8.5 and 14.9 GHz (Rodríguez et al. 2005). This is the first reported case of a radio jet associated with a young O-type star. Located at the centre of a massive dense molecular core (approximately 10³ M_Sun), the jet is driving a molecular outflow traced by H₂ 2.12 μm emission that extends over 1.5 pc (Brooks et al., 2003).

Methanol masers are quite common in massive star-forming regions. They fall into two categories first defined by Batrla & Menten (1988) and further considered by Voronkov et al. (2005): class II masers (of which the 6.7 GHz emission is the best known) are closely associated with sites of strong continuum emission, infrared sources and reside in the close environment of existing stars. In contrast, class I masers (emitting e.g. at 95 GHz) are usually found offset (up to a parsec) from continuum sources. Theoretical calculations strongly suggest that these masers are pumped via collisions of molecular hydrogen in contrast to class II masers, which have a radiative pumping mechanism (e.g. Cragg et al., 1992). For more information on classification see Ellingsen (2005), Ellingsen et al. (2004) and Voronkov (2005).

Class I masers are relatively poorly studied. There is direct observational evidence for a relationship between these masers and outflows in a number of sources (Menten, 1996). However, other scenarios have been proposed (Mehringer & Menten, 1996). The outflow scenario is attractive for two reasons: The methanol abundance is significantly increased in the shock-processed regions (e.g. Gibb & Davis, 1998) and the gas is heated and compressed, providing more frequent collisions and, therefore, more efficient pumping. From the theoretical side, Sobolev et al. (2005) have recently identified four different regimes of pumping, each corresponding to a different set of physical parameters such as temperature, density and beaming (elongated geometry). It follows from these calculations that relatively rare masers at 9.9 and 104 GHz as well as bright masers in the J₂-J₁ series near 25 GHz require higher densities and temperatures than the masers at 84 GHz and 95 GHz. One may argue that these rare class I masers exclusively trace the interface regions where an outflow interacts with the ambient material.

Maser line surveys (Val'tts et al., 2000; Voronkov, 2005; Voronkov et al., 2005) have revealed that IRAS 16547-4247 is the only source where masers in all class I transitions falling into the frequency range of the ATCA receivers have been detected together, including the relatively rare 9.9 and 104 GHz masers. No class II maser emission at 6.7 GHz has been detected towards this source (Walsh et al., 1998). To investigate the relations of different maser transitions with the outflow and provide the data for theoretical studies we have undertaken interferometric observations at 9.9 and 25 GHz (we observed 8 transitions of the 25 GHz series, J=2 to J=9) in May and June 2005 and 84, 95 and 104 GHz in August 2005 using the ATCA. Such multi-transitional observations of class I masers are the first of their kind. The ATCA is currently the only interferometer in the world equipped with suitable receivers to carry out this task.

Figure 1: VLT H₂ 2.12 μm image of IRAS 16547-4247 (Brooks et al., 2003). The cross denotes the position of the thermal jet. Click on the image to see the inset, which shows the positions of the 95 GHz maser spots and the 25 GHz continuum sources. The synthesised beam is shown in the bottom-left corner of the inset (open and filled ellipses at 25 GHz and 95 GHz, respectively).

Figure 1 shows a VLT image of H₂ 2.12 μm emission (which traces shocked gas) towards IRAS 16547-4247 (Brooks et al., 2003). There is a complex chain of emission with three major concentrations consistent with the morphological characteristics of Herbig-Haro (HH) objects arising from the interaction of a collimated flow with the ambient medium (Reipurth & Bally, 2001). The outer concentrations are located approximately symmetrically offset from the radio jet detected by Garay et al. (2003), which is marked by a cross in Figure 1. The central concentration has a complex bow-shaped morphology curved away from the jet direction. This could be explained by a number of factors such as irregularities of the ambient medium, a precession of the jet, or an additional outflow from a star less massive than the luminous YSO. The inset in Figure 1 shows the location of all detected maser spots at 95 GHz in the vicinity of the thermal radio jet and the central H₂ 2.12 μm emission concentration (note that the outermost lobes of the H₂ emission are outside the Compact Array field of view at 3 mm). Also shown is the associated 25 GHz continuum emission, which closely resembles the triple-source morphology found at lower frequencies by Garay et al. (2003). The central continuum source is the thermal jet, while the two satellites mark the internal working surfaces of the collimated flow (i.e. where the supersonic flow slams into the slower moving gas within the flow). The satellites are characterised by non-thermal emission (with negative spectral indices).

All the detected masers are located within a dense molecular core (with a deconvolved FWHM of 27 arcsec, Garay et al., 2003), with velocities close to that of the ambient molecular material in the core (of -30.0 km s^-1). The behaviour of the detected maser spots also depends on the transition. All 95 GHz maser spots, except those at -30.3 km s^-1 and -30.5 km s^-1, show 84 GHz maser emission. However, only the -31.6 km s^-1 spot is active at 9.9 GHz, 25 GHz and 104 GHz.

The southern maser spots clearly trace the bow-shaped shocked gas delineated by the 2.12 μm H₂ emission. This correlation points to an association between the class I methanol masers and the interface region where the outflow interacts with the dense molecular core. The maser at -31.6 km s^-1 is located near the edge of the brightest H₂ knot. This may be an indication that indeed more energetic conditions are required to form rare 9.9 GHz, 104 GHz and bright 25 GHz masers.

The northern maser spots show no association with the H₂ emission. However, this may be attributed to higher extinction in the northern region which diminishes the strength of H₂ 2.12 μm emission. For instance, the northern satellite radio continuum source has no associated detected H₂ 2.12 μm emission, unlike its southern companion. There is little doubt, however, that the northern satellite radio continuum source represents the working surface of a collimated flow. Moreover, it is evident from the large-scale H₂ 2.12 μm emission that an outflow lobe extending northwards is present. The -30.5 km s^-1 spot resides close to the northern non-thermal continuum lobe and we therefore assume that all the northern maser spots are associated with the outflow as well. More data are required to confirm this.

Further observations of masers can help to elucidate the connection between outflows and class I methanol masers in detail. For example, a proper motion study will be able to reveal the nature of the shock associated with the masers and test whether it is related to another YSO. The similar radial velocities of the maser spots suggest that the shock is moving in the tangential plane on the sky. Therefore, proper motions are expected to be detectable with ATCA observations on timescales of approximately 3-10 years, depending on the outflow velocity (typically 10³ km s^-1 for jets associated with luminous objects). The association with outflows encourages us to extend searches for class I methanol masers to sites of intermediate- and low-mass star formation. Outflows are common in these sites (Reipurth & Bally, 2001) and, indeed, a few class I masers have recently been found (Kalenskii et al., 2005). Further observations of H₂ 2.12 μm as well as other shock tracers towards known class I methanol masers, particularly rare masers at 9.9 GHz and 104 GHz and bright 25 GHz masers, are required to understand whether or not class I masers are always associated with shocked gas.

References

Batrla W., Menten K.M., 1988, ApJ, 329, 117

Brooks K.J., Garay G., Mardones D., Bronfman L., 2003, ApJ, 594, L131

Cragg D.M., Johns K.P., Godfreq P.D., Brown R.D., 1992, MNRAS, 259, 203

Garay G., Brooks K.J., Mardones D., Norris R.P., 2003, ApJ, 587, 739

Gibb A.G., Davis C.J., 1998, MNRAS, 298, 644

Ellingsen S.P., Caswell J.L., Voronkov M.A., Cimó G., 2004, ATNF Newsletter, 54, 18

Ellingsen S.P., 2005, MNRAS, 359, 1498

Kalenskii S.V., Promyslov V.G., Slysh V.I., Bergman P., Winnberg A., Astron. Reports., 2005, in press (astro-ph/0505225)

Menten K.M., 1996, proceedings of IAU Symposium 178 (eds. E.F. van Dishoeck), 163

Mehringer D.M., Menten K.M., 1996, ApJ, 474, 346

Reipurth B., Bally J., 2001, ARA&A, 39, 403

Rodríguez L.F., Garay G., Brooks K.J., Mardones D., 2005, ApJ, 626, 953

Sobolev A.M., Ostrovskii A.B., Kirsanova M.S., Shelemei O.V., Voronkov M.A., Malyshev A.V., 2005, proceedings of IAU Symposium 227 (eds. E. Churchwell, P. Conti and M. Felli), in press

Val'tts I.E., Ellingsen S.P., Slysh V.I., Kalenskii S.V., Otrupcek R., Larionov G.M., 2000, MNRAS, 317, 315

Voronkov M.A., Sobolev A.M., Ellingsen S.P., Ostrovskii A.B., Alakoz A.V., 2005, Ap&SS, 295, 217

Voronkov M.A., 2005, ATNF Newsletter 55, 10

Walsh A.J., Burton M.G., Hyland A.R., Robinson G., 1998, MNRAS, 301, 640

Maxim Voronkov (ATNF), Kate Brooks (ATNF), Simon Ellingsen (University of Tasmania), Andrej Sobolev (Ural State University), Jim Caswell (ATNF), Andrei Ostrovskii (Ural State University), Guido Garay (Universidad de Chile) and Diego Mardones (Universidad de Chile)
(Maxim.Voronkov@csiro.au)