4. Discussion and conclusions
The CS(2-1) and CS(3-2) peak line temperatures (0.5 K) observed in the hotspot are much lower than those found with similar observations in a sample of dense molecular clumps where massive stars are just being formed, in which values as high as K have been observed (Cesaroni et al., in preparation). Considering that these lines require densities cm-3 to be thermalized, this suggests that in later evolutionary stages, as seems to be the case for the hotspot , the molecular densities are lower. This confirms that the high brightness temperature observed in 12CO by MWW (48 K) is due to high temperature ( K), lower density molecular gas located close to a heating source.
The ratio of CS(3-2) to 13CO antenna temperature (using the 13CO value of MWW) is 0.08, about a factor 5 lower than the mean value found in the Cesaroni et al. sample. This may suggest that CS is depleted close to the ionization front, as also suggested by the LVG model.
Our higher resolution observations of the CS(2-1), (3-2) and (5-4) rotational transitions of the hotspot have confirmed that the blister-type HII region has evacuated a cavity and formed a sharp dense front at the edge of an extended molecular cloud, particularly intense in the S-SW part. The morphology confirms that A-NIR is the stellar source responsible for the heating source of the dust emitting in the far IR and, in turn, of the hotspot , as discussed by MWW. To estimate hydrogen densities we position-averaged the spectra in different regions, to mitigate the effects of undersampling at the higher frequency CS transitions, and compared the resulting spectra with a set of LVG models at fixed temperature. We thus found densities of order of cm-3, and total CS column density cm-2.
The mean electron density in source A is cm-3 (TOHTFG). Assuming the standard temperature in an HII region of 104 K and using the molecular gas temperature of 60 K, the pressure equilibrium density in the molecular front is cm-3. The molecular density in the larger scale molecular cloud are much smaller, of order of 103 cm-3 (MWW). The higher value at the front is close to the values derived from the LVG analysis. The density increase must be produced by shock compression ahead of the ionization front.
At the southern edge of source A, both the integrated line intensity (i.e., column density) and the line-width of CS(2-1) and (3-2) are enhanced. This effect is more pronunced at blue-shifted velocities, where the integrated intensity shows a sharp edge, almost bordering source A. Line profiles show some degree of asymmetry, consistent in some cases with a two-components Gaussian fit. We interpret these effects as an indication of shocked molecular gas in a shell around the blister being accelerated away and towards us.
The line profiles associated with the NW part of our maps are consistent with a single component, and emission is observed at less blueshifted velocities. This molecular gas is thus likely to be associated with the remnant of the dense clump from which A-NIR originally formed. The coincidence of the NW clump with source B still leaves open the query if this non-thermal source is associated with it or not.
© European Southern Observatory (ESO) 1998
Online publication: October 22, 1998