Astron. Astrophys. 327, 689-698 (1997)

3. Results

We show in Fig. 1 the spectra at the nebula center for all the observed molecular transitions. The adopted coordinates for the central position are: (B1950) 07:39:59.0, -14:35:41. We see that, except for , the line profiles are similar, with a relatively narrow component (between 10-55 km s^-1 LSR) centered at 33 km s^-1 LSR that dominates the emission, and two weaker wings. The central spectral component arises from the central compact condensation while the wings, that reach very high velocities, arise from the nebula lobes. , contrarily, shows a flat line profile between -30 and +80 km s^-1 LSR, the emission from the wings in this velocity range being as intense as the central spectral component. We can see in this spectrum a feature between -80 and -30 km s^-1 LSR also visible at other adjacent positions (see below).

Fig. 1. Central position spectra of the molecular transitions observed in OH 231.8 in units of main-beam temperature and projected velocity (LSR).

In Fig. 2 we can see the spatial distribution of the ¹² CO J =2-1 velocity integrated intensity from OH 231.8. A sketch of the source appearance at optical wavelengths, the beam half-maximum contour and the observed positions are also represented in this figure. The total deconvolved extent along the nebula symmetry axis is about , comparable to that of the optical image. In the right panels we see the spectra of this line in representative points along the axis. Note that the quality of the ¹² CO J =2-1 map of our first observations (Alcolea et al. 1996) has been improved by adding the data of a second observing period.

Fig. 2. Map of the integrated intensity of the 12 CO J =2-1 line and spectra for selected points along the symmetry axis. For the spectra, east and north offsets are indicated within square brackets. Contours are 2, 3, 4, 5 and 10 to 90 by 10% of the maximum (219 K km s-1). The observed points are indicated with small crosses, the encircled ones are the positions for which the spectra are shown. The beam size and a sketch of the source in the visible is also shown in light grey. Central coordinates: (B1950) 07:39:59.0, -14:35:41.

We show in Fig. 3 the displacement of the emitting gas along the axis for different LSR velocity intervals: I1 [-80:-30], I2 [-30:+10], I3 [+10:+55], I4 [+55:+80], I5 [+80:150], and I6 [+150:250] (ranges in km s^-1, see right-top corners in the figure). These velocity intervals are similar to those chosen by Alcolea et al. (1996) and correspond to different components that can be found in both our line profiles and spatial intensity distributions (see Figs. 1 and 4, and discussion below). We confirm that the emission from the top, bottom and center of the nebula takes place at very different velocities, which shows that the features detected in the different map positions correspond to independent emitting gas. It is also remarkable that the clump with the highest (positive) velocity lies on the southern bow-like structure of the nebula in the visible. For cuts along the direction perpendicular to the nebular axis, the deconvolved size of the CO emission at half maximum intensity is always . This is in agreement with the small extent of the nebula in OH maser emission and in the visible, except for the southernmost part: the larger extent of the optical emission in this region has no clear counterpart in the compact high velocity clump observed in CO. No velocity gradient is detected in the direction perpendicular to the nebula axis.

Fig. 3. Maps of the 12 CO J =2-1 integrated intensity in representative velocity ranges. The velocity intervals (km s-1 LSR) are indicated in the upper-right corners. Levels are 5 8.5 13 20 35 65 and 105 K km s-1. In the last panel we also show a sketch of the nebula in the visible and the contour at half maximum of the integrated intensity for the interval [220:250] km s-1 (dashed line). Central coordinates: (B1950) 07:39:59.0, -14:35:41.

In Fig. 4 we show the velocity-position diagrams along the symmetry axis of the nebula for the ten observed transitions. The spectral resolution is 11 km s^-1. The maps are distributed in two columns according to their spatial resolution, for the first column and for the second one. The apparently larger extent of the molecular emission in the second column is then an effect of the lower spatial resolution. We can see, looking at the velocity-position maps, that there is a clear velocity gradient along the symmetry axis.

Fig. 4. Intensity as a function of velocity and position along the symmetry axis of the observed lines. For the different transitions, contours in percentage of the maximum (see values in Fig. 1) are, 12 CO 2-1: 2.5, 4, 6.5 and 10 to 100 by 10; 13 CO 2-1: 3, 6.5 and 10 to 100 by 10; SiO: 5, 10, 20, 30, 40 and 50 to 100 by 20; CS: 5 and 10 to 100 by 20; SO2 : 1.5, 3, 4.5 and 10 to 100 by 15; 12 CO 1-0: 3, 5 and 10 to 100 by 10; 13 CO 1-0: 6 and 10 to 100 by 10; : 11 to 100 by 10; HCN: 5 and 10 to 100 by 10; HNC: 15 20 30 to 100 by 20.

By virtue of its large abundance and intense emission, ¹² CO remains the best tracer of the nebular material, showing the largest spatial size and better defining the different components of the molecular envelope. The velocity gradient, determined from the ¹² CO J =2-1 map, is approximately constant ( 6 km s^-1 per arcsec), the highest expansion velocity being associated to the southernmost part of the nebula. Taking into account the inclination of the nebula axis with respect to the plane of the sky, 40 , we deduce that the molecular gas is flowing with velocities (relative to that of the central core) up to 180 km s^-1, in the north lobe, and up to 330 km s^-1, in the southern one. We can also see in Fig. 4 that the velocity-position maps appear fragmented, showing different spectral features located at clearly separated positions along the nebula axis (this structure approximately corresponds to the velocity ranges, I1 to I6, that for simplicity we are considering in some calculations).

Note that the same velocity gradient and nebula fragmentation seem to be present in all the observed transitions. They clearly appear in the ¹³ CO, SiO and maps. We remark that the signal to noise ratio is poor in the the CS, SO₂, HCN and HNC wings. Nevertheless, also in these cases, we find emission from the clumps with velocities between -30: +10 and +55: +80 km s^-1 LSR at the position expected from the axial velocity gradient found for the other molecules. We can better see this in Fig. 5, which shows the variation along the symmetry axis of the integrated intensity in representative (projected) velocity ranges. It is clear that the red/blue wings are shifted toward the south/north lobe for all the molecular transitions. Note that there is probably emission of CS, HCN and HNC also between -80 and -30 km s^-1 LSR (see Fig. 4). These features are located at the positions that correspond to the velocity gradient of the molecular gas, so we think that they are real. We must note that the J =1-0 HCN transition shows hyperfine components separated by 11 km s^-1, which can degrade the effective spectral resolution of our velocity-position map (Figs. 4 and 5). However, this effect could be only important in the central part of the diagram, due to the very large velocity gradient of the nebula.

Fig. 5. Integrated intensity in representative velocity ranges along the symmetry axis. The representation of the LSR velocity intervals is indicated in the upper-left panel.

It is remarkable that the central clump dominates the emission for all the molecular transitions except for . In Fig. 4 we clearly see that a relative minimum of emission appears at the central velocity interval while the other molecules clearly show the most intense emission at this velocity range. This behavior cannot be due to selfabsorption, since in circumstellar envelopes selfabsorption always appears at the negative-velocity edge of the line profile. Indeed, the very weak emission suggests a low optical depth, although we must note that this low intensity can also be due to a strong clumpyness. It seems that the bulk of the emission originates at the nebula lobes, where the gas is flowing faster because of the wind interaction. As we will discuss in Sect. 5, the molecular abundance of in OH 231.8 could be enhanced by shock-induced reactions, explaining the peculiar emission of this molecule. We tentatively detect a similar behavior for the HNC emission. The most intense component appears slightly shifted in velocity and position with respect to the dominant features of other lines, although this shift is compatible with the general velocity gradient. However the low signal to noise ratio in the HNC maps prevents to extract definitive conclusions.

The fragmentation found in the velocity-position maps of the observed molecules shows that the molecular envelope of OH 231.8 is formed by several components. The outer features with large expansion velocities, in the ranges [-80,-30] and [+150,+250], very probably correspond to independent condensations well separated from the rest (inner part) of the nebula. Other components, particularly those corresponding to the ranges [-30,+10] and [+55,+80], stand out from the different intensity distribution of certain molecules, like and SiO. They could perhaps represent regions of peculiar chemistry.

© European Southern Observatory (ESO) 1997

Online publication: April 6, 1998
helpdesk.link@springer.de