Astron. Astrophys. 325, 725-744 (1997)

5. A comprehensive model: disk and outflow

The nature of the velocity shifts seen in HCO (Fig. 7) and CH₃ CN (Fig. 10) is puzzling. In Sect. 4.1, we have favoured the interpretation of the HCO velocity gradient as due to a bipolar outflow with axis lying almost in the plane of the sky. However, in principle any kinematical interpretation can be suitable (collapse, expansion, rotation). Here we wish to demonstrate that the hypothesis of a bipolar outflow oriented SE-NW (seen in HCO ) together with a rotating disk (seen in CH₃ CN) is to be preferred, if one takes into account the following points:

• The HCO velocity patterns, the H₂ O maser spots, the K-band continuum, and the H₂ emission are all elongated in the same direction (Figs. 8 and 12).
• The three H₂ O maser spots present the same velocity trend (increasing from SE to NW) as the HCO blue- and red-shifted low velocity emission (Fig. 7, bottom). In fact (see TFTH, their Fig. 14 and Table 2), the H₂ O spot to the SE (C3 in the notation of TFTH) presents a spectrum with features from -6.6 to -2 km s^-1, the central spot (C2) from -14.5 to 3.8 km s^-1, and the one to the NW (C1) from 3.2 to 3.8 km s^-1.
• The H₂ emission is likely to originate from shocks due to the impact of material flowing from the central embedded object with the surrounding molecular gas.
• The K-band emission on the two sides of the core may be reflected light from the embedded star: this would indicate the existence of two low-density - and thus low-extinction - channels along the SE-NW axis, through which NIR photons can escape from the dense core.
• The CH₃ CN emission (Fig. 10, bottom) is elongated perpendicularly to the HCO axis, i.e. in the SW-NE direction.
• The CH₃ CN velocity (Fig. 10, top) increases steadily from NE to SW.

All these features find a natural explanation in the following scenario (see Fig. 17 for a schematic picture of the model). IRAS 20126+4104 is a very young B2.5-B0.5 (proto)star at the centre of a dense rotating disk (seen in CH₃ CN and CH₃ OH) of angular diameter (  pc), temperature  K, mass , H₂ density  cm^-3 and column density  cm^-2. Note that the disk diameter has been derived from the distribution of the points in the bottom panel of Fig. 10 and is consistent with the upper limit in Table 8. The disk is seen almost edge-on, making the (proto)star invisible in the NIR: if the embedded object is a B0.5 star, then this implies a visual extinction  mag, i.e. an H₂ column density  cm^-2, consistent with the value derived above. Along the disk axis the matter is swept away by a molecular outflow (seen in HCO ), which extends up to (  pc) from the centre; the low density along the axis allows the NIR photons to escape and become visible as reflection nebulosities (seen in the K-band continuum). The disk and outflow are embedded in a clump (  pc) large, with  K and mass . The interaction of the flow with the surrounding bulk material gives rise to the H₂ O masers, in the inner part of the flow, and to shock-excited H₂ emission, in the outer part.

Fig. 17. Schematic picture of the main components in the molecular cloud associated with IRAS 20126+4104. The drawing is not to scale. The three components represented are (i) the "clump" (or "ambient" or "bulk" gas), seen in mainly in most tracers; (ii) the "outflow", seen in HCO ; (iii) the "disk" (or "core") seen in CH3 CN and highly excited CH3 OH. We also draw the large scale outflow (dashed curves) seen by WBM in 12 CO and 13 CO

We note that the disk diameter agrees with the estimate =200 K and the measured peak brightness temperature in the synthesised beam of the CH₃ CN lines,  K (see Table 4): in fact, an optically thick disk of size (see Fig. 10 bottom) and  K, observed with the angular resolution in Table 2 gets beam-diluted down to  K, consistent with the observed value.

We wish to stress that the velocity trend shown in Fig. 10, top, is not to be taken as the intrinsic rotation curve of the disk, but only as the convolution with the PdBI beam, which is much larger () than the disk diameter (). Nevertheless, the velocity gradient can be used to give an estimate of the dynamical mass by imposing the equilibrium condition for a rotating disk: this is , where i is the inclination angle of the disk axis with respect to the line of sight. As discussed in Sect. 4.1, very likely i is close to , implying . This must be compared with the sum of the disk mass (see Table 8) and that of the embedded star ( for B0.5 V, see Table 21 of Schmidt-Kaler 1981). Given the uncertainties, the agreement seems very good.

We note that a priori one cannot exclude that is the total mass of a binary (or multiple) system of lower mass stars than B0.5: this would increase the associated stellar mass for the same . However, the detection of continuum emission at 3.6 cm (see Sect. 3.3) and the reasons discussed at the end of Sect. 4.3 (H₂ O maser emission, IRAS colours typical of UC HII regions, etc.) seem to favour the existence of an early type star in the core, thus ruling out the multiple star hypothesis.

5.1. Time scales of outflow and disk

When discussing a disk-outflow system, one must check that their ages are comparable, because disk and outflow should have formed approximately at the same time. This is easily verified: in fact, if the disk is unstable, then its lifetime is of the order of a rotation period, namely  yrs, which is indeed comparable to the age of the outflow, 5800 yrs, as given in Table 6. Moreover, one must remember that we measure projected values of the velocity and linear size: by taking such effect into account, it is easy to demonstrate that the two time scales become equal for an inclination angle . This result is perfectly plausible, as discussed in Sect. 4.1.

Another time scale which matters is the time necessary to destroy the core. In fact, the core may lose mass accreting onto the embedded star and/or feeding the outflow: the fraction of matter lost in either process is unknown. However, ignoring the accretion onto the star and assuming that all the mass lost by the core is turned into the molecular outflow, one obtains an upper limit on the core destruction time scale:  yrs, where the mass input of the flow has been computed in Sect. 4.1.1. Such upper limit is consistent with the age of the flow.

© European Southern Observatory (ESO) 1997

Online publication: April 28, 1998

helpdesk.link@springer.de