Astron. Astrophys. 361, 1079-1094 (2000)

3. Results

3.1. One-point observations

HNCO was detected in 36 SEST sources (from 56 observed) and in 22 OSO sources (from 27). Because of one source belonging to both samples, the total number of detected objects is 57. In many cases transitions were detected too. The gaussian line parameters are presented in Tables 5-14 (Tables 7-14 are available only electronically). It is worth noting that a single-gaussian fit is clearly insufficient in many cases because the lines have broad wings and other non-gaussian features. Therefore, the values in the tables give only a rough representation of the line profiles (the integrated intensities were obtained by integrating over the lines in most cases).

Table 5 summarizes the 220 GHz SEST results for HNCO and C¹⁸O. The Onsala and C¹⁸O results are presented in Table 6. The 220 GHz results for the , 3 ladders are given in Table 7, Table 8. The 110 and 154 GHz SEST data are displayed in Table 9. The Onsala 88 GHz data are summarized in Table 10. The Effelsberg data are presented in Table 11. Table 12-Table 14 contain the HHT data. We fitted the Effelsberg spectra with 3-component gaussians with fixed separations corresponding to the hyperfine structure of the transition.

Table 5. C¹⁸O (2-1) and HNCO () integrated intensities and gaussian line parameters at the indicated positions (cf. Table 1, Table 2) measured at SEST. The numbers in the brackets are the statistical uncertainties in the last digits (standard deviations).

Table 6. C¹⁸O (1-0) and HNCO () integrated intensities and gaussian line parameters at the indicated positions (cf. Table 1, Table 2) measured at Onsala. The numbers in the brackets are the statistical uncertainties in the last digits (standard deviations).

Examples of measured spectra are given in Fig. 2, Fig. 3. Fig. 2 shows spectra of a few sources covering , 2 and 3 transitions at 220 GHz. Fig. 3 presents HNCO spectra in the HNCO transitions at different wavelengths for several sources.

Fig. 2. Examples of 220 GHz HNCO spectra obtained at SEST covering , 2 and 3 transitions

Fig. 3. HNCO lines in four selected sources. For the transition a 3-component gaussian (according for hyperfine structure) is superposed

The HNCO line profile in Orion KL can be decomposed into at least two components which likely correspond to the so-called classical "Hot Core" and "Plateau" outflow components (see, e.g., Harris et al. 1995). The ratio between these components is practically the same for the , and lines: % of the emission originates from the "Plateau" outflow source. The other lines do not allow such decomposition due to their weakness or blending with other spectral features.

An inspection of Table 5 shows that the derived C¹⁸O velocities are systematically lower (more negative) than the HNCO ones. The difference is  km/s on the average. This can be an instrumental effect: the C¹⁸O line was located far away from the center of the spectrometer band and a possible non-linearity in the frequency response could lead to the apparent displacement of the line on the velocity axis. This remark is applicable also for the higher HNCO lines.

3.2. Maps

In order to estimate source sizes and their spatial association with YSO and infrared (IR) sources we mapped 2 southern sources in the HNCO line and Orion KL, W49N and W51M in the , and lines. W51M was mapped also in the line. Three of these maps are presented in Fig. 4.

Fig. 4. The HNCO integrated intensity maps of G301.12-0.20 and G270.26+0.83 and HNCO integrated intensity map of Orion KL. The levels start from 15% of the peak intensities in steps of 7.5%. The peak intensities equal to 4.4, 1.5 and 95.0 Kkm s-1 for G301.12-0.20, G270.26+0.83 and Orion KL, respectively. The beam widths are 24" for the first two objects and 18" for Orion KL. For the first two objects the large crosses indicate IRAS positions, small stars show NIR sources, triangles mark H2O masers, squares correspond to OH masers and diamonds show methanol masers (for references see Lapinov et al. 1998). Open circles mark UC H II regions (Walsh et al. 1998). The IRAS uncertainty ellipses are shown. For Orion KL only the IRc 2 position is indicated

The sources remain spatially unresolved. E.g. for G 301.12-0.20 we obtain a FWHM " in right ascension (from the strip scan across the map) which is very close to the beam size at this frequency (24").

3.3. Detection of the HNCO transition

The highest HNCO transition reported so far was (the line) in Orion (Sutton et al. 1985). This line is located on the shoulder of the strong C¹⁸O line. In Fig. 5 we show parts of our Orion 220 GHz low resolution spectrum and 461 GHz spectra with , 3, 4 and even 5 features (the transition is outside our band). The rest frequencies are assumed to be equal to those given in the JPL catalogue for the strongest components of the corresponding transitions (for at 220 GHz we took the mean of the frequencies of the two strongest components).

Fig. 5. Left panel: parts of the Orion low-resolution 220 GHz spectrum corresponding to higher HNCO transitions. The profiles are aligned in velocity. No baselines are subtracted but the subspectra are shifted along the y-axis for clarity. The dashed vertical line corresponds to  km s-1. Right panel: the same for the 461 GHz spectrum. Here the dashed vertical line corresponds to  km s-1

There is a weak bump in the redshifted C¹⁸O wing which can be attributed to HNCO . Due to the uncertainty in fitting the C¹⁸O line profile the intensity of the HNCO feature cannot be reliably determined but it is lower than reported by Sutton et al. (1985). Our best estimate for the integrated intensity is  K km/s, but a reliable error cannot be given.

There is also a feature at the frequency in the 220 GHz spectrum. It is located in the wing of a C₂H₃CN line. The integrated intensity is  K km/s. The identification of this feature with HNCO seems to be reliable. The only other candidate is the C₂H₅OH line at 219391.81 MHz. However, there is no sign of other ethanol lines in our spectrum so we reject this alternative. In the 461 GHz spectrum the feature is clearly detected. Its integrated intensity is  K km/s.

3.4. Hyperfine splitting, HN¹³CO and optical depths

The HNCO lines are split into several hyperfine components mainly due to the ¹⁴N spin. This splitting is clearly seen in the transition (Fig. 3) at 22 GHz. Earlier HNCO hyperfine structure in the line was only observed in the dark cloud TMC-1 (Brown 1981) where possible deviations from the optically thin LTE (Local Thermodynamic Equilibrium) intensity ratios (3:5:1) were found. In our spectra the hyperfine ratios are consistent with the optically thin LTE values. Taking into account the measurement uncertainties, an upper limit on the optical depth in this transition for the sources detected in Effelsberg is .

To the best of our knowledge no isotopomer of HNCO except the main one has been detected in space yet. This detection would be important for estimates of HNCO optical depths which are believed to be small (e.g. Jackson et al. 1984; Churchwell et al. 1986). The frequency separations between the HN¹³CO and the main isotopomer lines are rather small corresponding to a few km/s, so in sources with broad lines like Orion A or Sgr A the HN¹³CO lines will be blended. However, there are some strong HNCO sources in our sample with narrower lines which show features attributable to HN¹³CO. The most reliable one is seen in the G 301.12-0.20 spectrum (Fig. 6). A weak feature on the blue shoulder of the main isotope line is very close in frequency to the expected location of the HN¹³CO line.

Fig. 6. The HNCO line in G301.12-0.20 in comparison with the C34S(2-1) line. The expected location of the HN13CO line is shown

For comparison we show in addition to HNCO also the C³⁴S spectrum. It is noteworthy that there is no bump in this spectrum corresponding to the discussed feature in HNCO.

The line we identify with HN¹³CO is shifted by  MHz from the expected HN¹³CO transition frequency. This shift, if it is significant, cannot be explained by instrumental effects like in the case of our C¹⁸O data because the feature is very close to the main isotope line. The shift greatly exceeds the uncertainty of the transition frequency derived from the laboratory data (Winnewisser et al. 1976) which is 25 kHz. This makes the identification questionable. Detection of other HN¹³CO lines would be important in this respect. There is no corresponding feature in the HNCO spectrum (the spectrum is too noisy). This could mean that the optical depth in this transition is significantly lower. Indeed, at sufficiently high temperatures ( K) it can be about 2 times lower than in the transition according to Eq. (2) (see the discussion in Sect. 4.2).

If our identification of the discussed line with HN¹³CO is correct we can estimate the optical depth assuming the same excitation as for the main isotopomer. For G 301.12-0.20 we obtain if we assume the terrestrial ¹²C/¹³C isotope ratio (¹²C/¹³C = 89) and for ¹²C/¹³C = 40. A high optical depth in the HNCO line does not contradict our conclusion of low optical depth in the transition because the line strengths for these transitions are different (see discussion in Sect. 4.2). Therefore, the optical depth in some lines of the main isotopomer might be rather high. This contradicts the usual assumption of low optical depth in all HNCO lines (e.g., Jackson et al. 1984; Churchwell et al. 1986) and could imply serious consequences for the analysis of HNCO excitation and abundances.

© European Southern Observatory (ESO) 2000

Online publication: October 10, 2000
helpdesk.link@springer.de