2. The observations
We have selected a sample of 11 optically bright carbon stars for which the photospheric 12C/13C-ratios have been determined by Lambert et al. (1986) and Ohnaka & Tsuji (1996). This sample is presented in Table 1, where we have also included S Sct, which was observed by Bergman et al. (1993) and Olofsson el al. (1996), as well as the three J-type stars (i.e., stars with 12C/13C-ratios 3) Y CVn, RY Dra, and T Lyr (see Table 4 for references). These 15 stars were also included in the major survey of circumstellar molecular line emission by Olofsson et al. (1993a,b).
The 13CO observations were carried out in October 1998 using the Swedish-ESO submillimetre telescope (SEST) located on La Silla, Chile, and in March 1999 using the Onsala 20 m telescope (OSO), Sweden. At SEST, a dual channel, heterodyne SIS receiver was used to simultaneously observe at 110 GHz (the J10 line) and 220 GHz (the J21 line). The single sideband temperatures of the receiver are about 110 K and 150 K at 110 GHz and 220 GHz, respectively. All the available acousto-optical spectrometers were used as backends. The two wideband (1 GHz), low resolution spectrometers (LRSs) were used to cover both lines. The third, narrow band (86 MHz), high resolution spectrometer (HRS) was used at 220 GHz, since the J21 line is usually stronger than the lowest rotational line in CSEs. The LRS had 1440 channels separated by 0.7 MHz whereas the HRS used 2000 channels separated by 42 kHz.
At OSO, an SIS receiver, with a single sideband temperature of about 100 K at 110 GHz, was used for the observations. As backends, two filterbanks with bandwidths of 512 MHz (MUL A) and 64 MHz (MUL B) were used. The MUL A used 512 channels separated by 1 MHz and the MUL B filterbank used 256 channels with a separation of 250 kHz.
The observations were made in a dual beamswitch mode, where the source is alternately placed in the signal and the reference beam, using a beam throw of about 12´ at SEST and about 11´ at OSO. This method produces two spectra that are subtracted from each other, which results in very flat baselines, i.e., most of the frequency dependant response of the system is removed. The intensity scales are given in main beam brightness temperature, , where is the antenna temperature corrected for atmospheric attenuation using the chopper wheel method, and is the main beam efficiency. For the SEST is 0.7 at 110 GHz and 0.5 at 220 GHz. At OSO 0.5 at 110 GHz. The beamsizes of the SEST are 45" and 24" at 110 GHz and 220 GHz, respectively. At OSO the beam is 34" at 110 GHz. Regular pointing checks were made on strong SiO masers and the pointing was usually better than for both telescopes. The uncertainty in the intensity scale is estimated to be about 20%. The observational results are summarized in Table 2 and the detections of 13CO line emission are shown in Fig. 1. These observations include first detections of circumstellar 13CO emission towards R Lep, UU Aur, UX Dra, and TX Psc. The line parameters, i.e., the main beam brightness temperature at the line centre (), the line centre velocity (), and half the full line width (), are obtained by fitting the data with an artificial line profile (Olofsson et al. 1993a)
where is a parameter describing the shape of the line. The integrated intensity () is obtained by integrating the emission between . The upper limits are obtained as , where is the peak-to-peak noise obtained from spectra where the velocity resolution has been degraded to twice the expansion velocity. The corresponding 12CO lines were also observed, simultaneously with the 13CO lines, for all the sample stars. Spectra of these lines have been published in Olofsson et al. (1993a).
Table 2. Observational 13CO results. A colon (:) marks an uncertain value due to a low signal to noise ratio (UU Aur) or contamination by interstellar lines (UX Dra).
To complement these new observations we have collected additional 13CO data and, in particular, 12CO data on the sample stars, as well as for an additional seven stars for comparison, from various sources and included them in the modelling (see Table 4).
© European Southern Observatory (ESO) 2000
Online publication: July 7, 2000