Free Access
Issue
A&A
Volume 564, April 2014
Article Number A64
Number of page(s) 8
Section Interstellar and circumstellar matter
DOI https://doi.org/10.1051/0004-6361/201323320
Published online 07 April 2014

© ESO, 2014

1. Introduction

Microwave and sub-mm absorption-line spectroscopy have greatly extended the inventory of molecules known to exist in diffuse molecular interstellar clouds, that is, clouds with appreciable H2 content but AV 1 mag. Just in the past few years, sub-mm observations from the PRISMAS project using the HIFI instrument on Herschel, from the APEX telescope in the Atacama and the GREAT instrument on SOFIA have more than doubled the number of known species following observation of the hydrides and hydride ions of oxygen, nitrogen, sulfur, fluorine and chlorine (Gerin et al. 2012; Neufeld et al. 2012). Sub-mm observations of the familiar species CH (Gerin et al. 2010) and CH+ (Falgarone et al. 2010; Godard et al. 2012) have allowed these species, previously seen only in the optical absorption spectra of relatively nearby stars, to be tracked across the disk of the Milky Way.

Even so, the search for new molecules and the overall effort to systematize the diffuse cloud chemistry have been seriously hindered by the narrow bandwidths that were available for high sensitivity observations of heavier polyatomic species in the microwave (cm-wave and mm-wave) domain. This impediment is gradually being overcome by new technology such as the WIDAR correlator at the Karl Guthe Jansky VLA that we used to detect l-C3H2 and survey the abundance of several small hydrocarbons (Liszt et al. 2012). An even larger development is the advent of the very wide-band EMIR receivers at the IRAM 30 m Telescope on Pico de Veleta, which produced the recent 1–3 mm WHISPER surveys of emission from the Horsehead nebula (Guzmán et al. 2012b; Pety et al. 2012; Gratier et al. 2013).

Table 1

Sightlines and continuum targets.

Here we describe a 3 mm band survey of absorption from the galactic diffuse and translucent clouds seen toward the mm-bright blazars BL Lac (EB − V = 0.32 mag) and 3C 111 (EB − V = 1.65 mag). As a result of this work the mm-wave absorption spectrum is now known at a detection level of 1% absorption or better over the 3 mm band and this paper reports the detection of three new species HCO, c-C3H and CF+, toward both objects.

The plan of this work is as follows. The observations are described in Sect. 2. In Sects. 3–5 we discuss the detection and chemistry of HCO, c-C3H and CF+, respectively. Section 6 is the briefest of summaries.

2. Observations and data reduction

2.1. The 3 mm band EMIR spectral sweep at the 30 m

Each of three sources (see Table 1) was observed for some 20 h over the course of five days observing in 2012 August. Only the data from BL Lac (B2200+420) and 3C 111 (B0415+379) will be discussed here because the flux of B0355+508 (NRAO150) was too low to detect the new species that were seen toward the other sources. We used the broadband EMIR receiver with an FTS spectrometer at 195 kHz resolution and channel separation (0.60 km s-1 at 100 GHz) while observing simultaneously over both 8 GHz-wide sidebands, in much the same manner as described previously for the 1–3 mm WHISPER surveys of emission from the edge-on PDR in the Horsehead nebula (Guzmán et al. 2012a). That is, we wobble-switched the secondary mirror symmetrically with an azimuth throw of 60′′ about the positions of the blazar background sources and offset the receiver local oscillator to produce continuous coverage, with some overlap, over very nearly the full receiver band. In principle, observing the 3 mm frequency band requires only two tunings. However, every frequency was observed with two different frequency settings separated by 500 MHz to allow us to remove potential ghost images arising from lines in the image side band, given that the typical rejection of the EMIR sideband separating mixers in only 13 dB (a factor 20).

thumbnail Fig. 1

Properties of the spectral sweep. Top: line profile integrals (equivalent widths) for the spectral sweep plotted against those previously obtained by synthesis at the Plateau de Bure Interferometer in our prior work. Bottom: 8-channel running mean channel-channel rms in the line/continuum ratio at 0-absorption for 3C 111 and BL Lac. The points labelled SO in the upper panel represent the 32–21 line at 99.3 GHz as discussed in Sect 2.1 of the text.

The temperature scale was computed and applied in the GILDAS/MIRA software on 4 GHz chunks at a time. All other reduction steps were done in the GILDAS/CLASS software1. In brief, we fit the continuum and divided the spectra by the continuum baseline to yield spectra as line/continuum ratios. We baselined the spectra with a low order baseline and rejected frequency ranges whose rms variation with time was much larger than the noise level, to remove potential spikes. We then co-added the spectra and improved the baseline solution by subtracting the linear interpolation of median values computed every 50 MHz over 100 MHz.

thumbnail Fig. 2

Spectra of BL Lac and 3C 111 in the region of the HCO quartet. The spectral resolution is 0.195 MHz. The approximate positions of HCO, H13CO+ and SiO J = 2–1 transitions are marked. For spectroscopic data, see Table 2.

Although we are confident that there are no other species remaining to be detected in the survey data, it is not yet entirely free of artifacts, for instance unexplained features, always downward-going, that are too broad to be real absorption in the interstellar medium and are not always present in all tunings. As a check on the validity of the data acquistion path we measured the integrated absorption for previously detected lines toward BL Lac and 3C 111 and a comparison of new and old results is shown at top in Fig. 1. Above an equivalent width of 0.3 MHz (1 km s-1 at 100 GHz) there is a slight tendency for the new results to be smaller, perhaps as the result of under-resolving the stronger lines, hence underestimating their opacity. The SO 3–2 line at 99.3 GHz called out in Fig. 1 falls nearly on the join of two tunings where the system temperature may not have been reliably inferred by the data reduction software and was the only feature so affected.

2.2. Survey properties: rms

Figure 1 at bottom shows the rms line/continuum noise for BL Lac and 3C 111 across the frequency band surveyed. To make these plots, we calculated a running rms over 8-channel intervals in line spectra divided by the continuum and averaged this over absorption-free regions of (approximately) fixed mean rms, determined by visual inspection. When the running mean rms changed beyond twice the expected variance a new data point was created. The rms increases at the extreme of the low end because less time was spent there and more broadly at the high end because of the increased atmospheric opacity and consequent increase of system temperature. Inside the extremes there are patches in the survey with noticeably higher rms. These are caused by some combination of smaller integration time/higher system temperature and artifacts in the IF. Some of these artifacts are narrow and resemble astronomical spectral lines while others are much broader. By careful comparison of tunings and polarizations we concluded that none of the problematic features result from true astronomical signals.

Toward BL Lac 0.0015 < σ < 0.002 over the region 85–110 GHz leading to a single-channel detection sensitivity 3σ/EB − V 0.02/mag and a 3 − σ detection sensitivity 0.023(dV)\hbox{$0.023 \sqrt({\rm d}V)$}/mag when integrated over a velocity interval dV (km s-1) at 100 GHz. The same quantities obtained toward 3C 111 were 0.004 < σ < 0.006 and 3σ/EB − V 0.011/mag for the single-channel sensitivities and a 3 − σ velocity-integrated sensitivity 0.014(dV)/\hbox{$0.014 \sqrt({\rm d}V)/$}mag at 100 GHz.

Thus, although the intrinsic channel-channel rms noise is considerably smaller toward BL Lac owing to its higher flux during the period of observations, the signal/noise of the new detections is better toward 3C 111 owing to the higher column density. Nonetheless, the increased sensitivity toward 3C 111 did not result in the detection of any additional species beyond those also seen toward BL Lac, even tentatively. Future observations that seek to improve significantly on the current results will have to have rms 0.002 at EB − V = 0.32 mag and rms/EB − V 0.003 mag -1 more generally.

2.3. Improved results for SiO and other previously-detected species

BL Lac was not included in our earlier survey of SiO J = 2–1 absorption (Lucas & Liszt 2000) so the spectrum in Fig. 2 represents the first detection of SiO in this direction; the SiO spectrum of NRAO150 is superior to that in the earlier work. SiO is one of several species (e.g. N2H+) whose abundances, either detections or upper limits, warrant rediscussion on the basis of the sensitivity achieved in the current work. This work is in progress.

thumbnail Fig. 3

Column density of HCO vs. H13CO+ (left) and C2H (right). For W49 only the “spiral arm” feature at 40 km s-1 is shown at left using data from the present work; at right the C2H data of Godard et al. (2010) is used for the 40 km s-1 and 60 km s-1 features. Data from the Horsehead PDR (TdC) are from Gerin et al. (2009). The dashed lines have power-law slope of unity, with proportions of 16:1 and 1:10 at left and right, respectively. Error bars in this and all other plots are 1σ statistical uncertainties.

2.4. Aperture synthesis of W49 in HCO and H13CO+ at the PdBI

We observed the 86.774 GHz HCO F = 1–1 line at the Plateau de Bure Interferometer for 7 h in September 2007 in the D configuration (baselines 25–100 m) and in 2008 April for 9 h in the C configuration (25–175 m) to achieve a combined spatial resolution of 4.0′′ over a single primary beam pointing. The spectral resolution was 270 kHz or 0.93 km s-1 and the rms noise in the line/continuum ratio at 0 absorption was 0.007 with 4.2 h of usable integration time. The LO was centered slightly too low in velocity for simultaneous coverage of H13CO+ in all of the “spiral arm” features that are seen in absorption in this direction but the strong absorption at 40 km s-1 was captured. All the previously known spiral arm features were seen in HCO.

Table 2

HCO 101–000 transitions observed and column density-optical depth conversionsa.

3. HCO and its detection in diffuse clouds

HCO was studied extensively at H II region-molecular cloud interfaces by Snyder and Schenewerk and their collaborators (Schenewerk et al. 1988), including Ori B and W49 but its presence in diffuse and translucent clouds has not previously been noticed. Table 4 of Schenewerk et al. (1988) has N(HCO)/N(H13CO+) = 1–3 generally and 10 toward NGC 2024 but they did not derive column densities for either species individually.

The spectra of HCO toward BL Lac and 3C 111 are shown in Fig. 2 and toward W49 in Fig. 4. The transitions observed and their transition probabilities are given in Table 2; the observed line depths are as expected in LTE. The HCO column densities quoted here in Table 1 were derived from a simultaneous gaussian fit to all observed transitions assuming that the excitation is in equilibrium with the cosmic microwave background. We find N(HCO)/N(H13CO+) = 16 toward BL Lac, 3C 111 and the 40 km s-1 spiral arm cloud toward W49, much larger than the values 1–3 seen in the denser regions studied by Schenewerk et al. (1988). The abundance of HCO relative to molecular hydrogen from our work is X(HCO) = 8 × 10-10 if N(H13CO+)/N(HCO+) = 1/60 and X(HCO+) = 3 × 10-9. This is about 10 times higher than in the dark clouds TMC-1 or B1, according to recent unpublished results (Marcelino & Bacmann, priv. comm.).

thumbnail Fig. 4

W49 continuum, HCO F = 1–1 and H13CO+ emission and absorption spectra. Left: 86.8 GHz continuum (grayscale; peak =2.5 Jy) and H13CO+ emission (contours of integrated brightness temperature in units of Kelvins km s-1). The 4′′ synthesized HPBW is shown inset at lower left. Right top: spectra of H13CO+ and HCO toward the continuum peak labelled position “A” at left. Right bottom: emission spectra at off-continuum positions “B” (blue in electronic media) and “C” (red). Spectra showing strong emission are H13CO+, those without are spectra of HCO.

Figure 3 at right shows a comparison of N(HCO) with N(C2H) in the diffuse and spiral arm clouds at 40 and 60 km s-1 toward W49 using the C2H column densities of Godard et al. (2010), and toward the Horsehead. Most of these have abundance ratios N(HCO)/N(C2H) 0.1 with a three times smaller value toward 3C 111, which has a higher than normal ratio N(C2H)/N(HCO+). For the Horsehead the peaks of the spatial distributions of C2H and HCO coincide just inside the illuminated edge of the PDR while the peak of HCO+ lies further inside the neutral region. The Horsehead has a much higher ratio of N(HCO)/N(H13CO+) in Fig. 3 at left because the edge-on viewing geometry allows us to distinguish the HCO and HCO+ peaks. If the Horsehead were observed face-on its HCO/H13CO+ ratio would most likely resemble much more closely that seen in the diffuse and spiral arm clouds because all of the material would be observed along the same sighlines. Whether such a segregation occurs in the diffuse and spiral arm clouds is impossible to discern.

Coincidence of the HCO and hydrocarbon peaks in the Horsehead may be ascribed to the formation mechanism for HCO, O + CH2 HCO + H: Gerin et al. (2009) found that a rapid rate for this reaction was required to explain the observed abundances. In turn, CH2 formation is initiated by the endothermic reaction C+ + H2 but C+ does not survive into dense molecular gas after carbon is converted to CO. Therefore HCO is most properly viewed as existing in interface regions where the density is comparatively high and a substantial H2 fraction is present but carbon remains largely in the form of C+.

thumbnail Fig. 5

Spectra of BL Lac and 3C 111 in the region of the c-C3H quartet. The spectral resolution is 0.195 MHz. For spectroscopic data, see Table 3.

Figure 4 illustrates how HCO is present in the gas within the W49 HII region-molecular cloud complex as a whole but does not persist into the denser neutral regions. The maps at left superpose contours of integrated H13CO+ emission upon the continuum in greyscale, noting three positions A (the continuum), B and C at which line profiles are shown at right. H13CO+ at positions B and C (lower right) shows blue and red-shifted emission features, respectively (with slight overlap), corresponding to the blue-shifted H13CO+ emission and red-shifted absorption seen toward the continuuum at position A. Thus the overall morphology can be explained by a disk geometry whose blue-shifted southeastern part is situated behind the continuum, so representing infalling material. By contrast, HCO at position A toward the continuum shows strong red-shifted absorption but no blue-shifted emission from the denser gas behind the continuum. Likewise, HCO shows no detectable emission off the continuum at positions B or C. HCO exists only in lower density portions of the molecular gas where the abundance of C+ is relatively large but the molecular excitation of HCO is relatively weak.

Table 3

c-C3H 212–111 transitions observed and column density-optical depth conversionsa.

4. c-C3H and its detection in diffuse clouds

The discovery spectra of c-C3H in diffuse and translucent clouds are shown in Fig. 5. The transitions observed and their transition probabilities are given in Table 3; the observed line depths are as expected in LTE. The column densities quoted here in Table 1 were derived from a simultaneous gaussian fit to all observed transitions assuming that all excitation is in equilibrium with the cosmic microwave background.

thumbnail Fig. 6

Column density of c-C3H vs. H13CO+ (at left), and C3H2 and C2H (right). Data for the hydrocarbons toward the Horsehead nebula (TdC) are from Pety et al. (2012). Data for TMC-1 are from Ohishi et al. (1992). The c-C3H survey data of Mangum & Wootten (1990) are shown at right, labelled “MW1990”. The solid line at left has power-law slope 1; at right, both lines have slope 0.8.

Figure 6 compares the observed column densities of c-C3H with those of H13CO+, ortho-c-C3H2 and C2H. The values quoted for the Horsehead in Fig. 6 are from the singledish IRAM 30 m observations of Pety et al. (2012) at lower spatial resolution than those shown in Fig. 4. The quasar absorption sightlines and TMC-1 have similar N(c-C3H)/N(H13CO+) 3, N(c-C3H)/N(o-c-C3H2)  0.1 and N(c-C3H)/N(H13CO+)  0.01. N(c-C3H) for the Horsehead is high by a factor two-four with regard to H13CO+ and c-C3H2, and perhaps somewhat high with respect to C2H as well.

Mangum & Wootten (1990) noted that the relative abundances of c-C3H and c-C3H2 varied little with environment, whether in dark or giant molecular clouds, HII region-molecular cloud complexes, etc. Our data greatly extend this conclusion downward in column density and across chemical families. At left in Fig. 6 it is seen that N(c-C3H)/N(H13CO+) = 3 toward TMC-1 and the two blazars observed here, over a range of a factor 40 in column density. At right the regression lines have power-law slope 0.80, so that N(c-C3H)/N(C3H2) and N(c-C3H)/N(C2H) vary by a factor ± 2 about a common mean over a factor 100 in N(c-C3H2) and N(C2H).

thumbnail Fig. 7

CF+ spectra toward BL Lac (upper) and 3C 111 (lower). The velocity axis corresponds to the mean LTE intensity-weighted centroid of the two hyperfine components noted in Table 4. Shown for comparison are scaled spectra of the 18.3 GHz transition of c-C3H2 from Liszt et al. (2012). For spectroscopic data, see Table 4.

Table 4

CF+ transitions observed and column density-optical depth conversionsa.

From the lack of variability in their abundance ratio, Mangum & Wootten (1990) inferred that c-C3H and C3H2 shared a common chemical antecedent C3H3+. Although this may be true, Fig. 6 suggests that c-C3H is simply another species, like HCO+, C2H, and C3H2, whose abundance relative to H2 varies little in the ISM at large.

5. CF+ and its detection in diffuse clouds

CF+ has previously been observed only in regions of much higher density, the Orion Bar (Neufeld et al. 2006) and the Horsehead (Guzmán et al. 2012a,b). The discovery spectra of CF+ in diffuse clouds are shown in Fig. 7 and the transitions observed and their transition probabilities are given in Table 4. The low spectral resolution of the sweep data, combined with the 0.34 MHz or 1.01 km s-1 separation of the two hyperfine components prevents a reliable determination of the relative strengths of the kinematic components toward 3C 111. The column densities quoted here in Table 1 were derived by integrating over the observed profiles.

Figure 8 shows that N(CF+) increases with N(H13CO+) and N(C2H), although more slowly than linearly (power law slopes 0.6 and 0.4 respectively). Figure 8 also shows the prediction of Neufeld & Wolfire (2009) that N(CF+) ≈ 4.5 × 1011  cm-2 for AV≳ 1 with most fluorine in HF in the H2-bearing regions. The conversion of carbon to CO and the abrupt depletion of gas phase fluorine at AV> 1 mag both act to confine CF+ at AV< 1 mag and to limit N(CF+) in the models of Neufeld & Wolfire (2009).

CF+ forms from HF via C+ + HF CF+ + H with a rate constant k1 = 7.2 × 10-9(T/300)-0.15  cm3 s-1 and is destroyed by recombination with electrons with a recombination coefficient α1 = 5.3 × 10-8(T/300)-0.8  cm3 s-1. If C+ is the dominant source of free electrons it follows that n(CF+)/n(HF) = 0.136 (T/300)0.65 with most gas-phase fluorine in HF.

thumbnail Fig. 8

Column density of CF+ vs. H13CO+ (at left) and C2H (right). Data for the Horsehead nebula (TdC) are from Guzmán et al. (2012a). The prediction N(CF+) ≈ 4.5 × 1011  cm-2 at AV 1 mag of Neufeld & Wolfire (2009) is shown as a blue dashed horizontal line. The power law slopes of the best-fit lines (0.59, 0.38) are noted.

We know that C+ survives as the dominant form of carbon at and well beyond AV = 1 mag because of the strong fractionation of 13C into 13>CO that is seen toward BL Lac and 3C 111 (Liszt & Lucas 1998); substantial amounts of gas-phase fluorine in the form of HF must also survive beyond AV = 1 mag to provide the behaviour seen in Fig. 8. If the fractional abundances of HCO+ and C2H with respect to H2 are about constant and the kinetic temperature does not vary substantially, the depletion of gas phase fluorine (HF) would vary approximately as AV−1/2. If the temperature were assumed to decline at higher AV, the depletion of fluorine would have to be even weaker in order to compensate for the smaller equilibrium CF+/HF ratio, which varies as T0.65 according to the simple chemical scheme detailed above (Neufeld & Wolfire 2009).

6. Summary

We surveyed the 84–116 GHz 3 mm spectrum in absorption against the compact extragalactic mm-continuum sources BL Lac and 3C 111 as noted in Table 1 at 195 kHz spectral resolution (0.6 km s-1 at 100 GHz), achieving an rms noise level δτ ≈ 0.002 at EB − V = 0.32 mag (AV = 1 mag) toward BL Lac and δτ/EB − V 0.003 mag-1 overall. HCO, c-C3H and CF+ were detected in absorption toward both sources and HCO was also found in the diffuse “spiral-arm” clouds in the galactic plane in front of W49. We discussed observational aspects of their chemistry in Sects. 3–5.

c-C3H is notable for having a nearly fixed abundance with respect to HCO+, C2H and c-C3H2 over a wide range of column density, even across the divide between the diffuse and dark or giant molecular clouds. The increase in N(CF+) beyond AV = 1 mag shows that both fluorine and C+ are abundant in the gas phase at such extinctions but conclusions about the depletion of fluorine are sensitive to assumptions about the temperature profile. The relative abundance ratio N(HCO)/N(H2) is higher in diffuse than dark or dense molecular gas, consistent with its prior interpretation as a species requiring both C+ and H2 for its formation via the reaction O + CH2 HCO + H.

No other new species are present in the spectra toward BL Lac (AV= 1 mag) at a 5-sigma optical depth limit of 0.01 in any single feature, suggesting that the hunt for new species in diffuse clouds in the radio domain will be most profitably conducted in the cm-wave region below 30 GHz where the more heavily-populated lower-lying energy levels of heavier polyatomic molecules are most readily accessible.


1

See http://www.iram.fr/IRAMFR/GILDAS for more information about the GILDAS software (Pety 2005).

Acknowledgments

The National Radio Astronomy Observatory is a facility of the National Science Foundation operated under cooperative agreement by Associated Universities, Inc. IRAM is operated by CNRS (France), the MPG (Germany) and the IGN (Spain). This work was partly funded by grant ANR-09-BLAN-0231-01 from the French Agence Nationale de la Recherche as part of the SCHISM project. We thank the editor, Malcolm Walmsley, and the anonymous referee for their comments.

References

  1. Blake, G. A., Sastry, K. V. L. N., & de Lucia, F. C. 1984, J. Phys. Chem., 80, 95 [NASA ADS] [CrossRef] [Google Scholar]
  2. Chandra, S., Shinde, S. V., Kegel, W. H., & Sedlmayr, E. 2007, A&A, 467, 371 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  3. Falgarone, E., Godard, B., Cernicharo, J., et al. 2010, A&A, 521, L15 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  4. Gerin, M., Goicoechea, J. R., Pety, J., & Hily-Blant, P. 2009, A&A, 494, 977 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  5. Gerin, M., de Luca, M., Goicoechea, J. R., et al. 2010, A&A, 521, L16 [Google Scholar]
  6. Gerin, M., Levrier, F., Falgarone, E., et al. 2012, Roy. Soc. London Philos. Trans. Ser. A, 370, 5174 [Google Scholar]
  7. Godard, B., Falgarone, E., Gerin, M., Hily-Blant, P., & de Luca, M. 2010, A&A, 520, A20 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  8. Godard, B., Falgarone, E., Gerin, M., et al. 2012, A&A, 540, A87 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  9. Gratier, P., Pety, J., Guzmán, V., et al. 2013, A&A, 557, A101 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  10. Guzmán, V., Pety, J., Gratier, P., et al. 2012a, A&A, 543, L1 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  11. Guzmán, V., Roueff, E., Gauss, J., et al. 2012b, A&A, 548, A94 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  12. Liszt, H. S., & Lucas, R. 1998, A&A, 339, 561 [NASA ADS] [Google Scholar]
  13. Liszt, H., Lucas, R., & Pety, J. 2006, A&A, 448, 253 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  14. Liszt, H., Sonnentrucker, P., Cordiner, M., & Gerin, M. 2012, ApJ, 753, L28 [NASA ADS] [CrossRef] [Google Scholar]
  15. Lucas, R., & Liszt, H. S. 2000, A&A, 355, 327 [NASA ADS] [Google Scholar]
  16. Mangum, J. G., & Wootten, A. 1990, A&A, 239, 319 [NASA ADS] [Google Scholar]
  17. Neufeld, D. A., & Wolfire, M. G. 2009, ApJ, 706, 1594 [NASA ADS] [CrossRef] [Google Scholar]
  18. Neufeld, D. A., Schilke, P., Menten, K. M., et al. 2006, Astrophys., 454, L37 [Google Scholar]
  19. Neufeld, D. A., Falgarone, E., Gerin, M., et al. 2012, A&A, 542, L6 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  20. Ohishi, M., Irvine, W., & Kaifu, N. 1992, in Astrochemistry of cosmic phenomena: Proc. of the 150th Symp. IAU, held at Campos do Jordao, Sao Paulo, Brazil, August 5–9, 1991, ed. P. D. Singh (Dordrecht: Kluwer), 171 [Google Scholar]
  21. Pety, J. 2005, in SF2A-2005: Semaine de l’Astrophysique Francaise, Conf. Ser., eds. F. Casoli, T. Contini, J. M. Hameury, & L. Pagani (EDP Sciences), 721 [Google Scholar]
  22. Pety, J., Gratier, P., Guzmán, V., et al. 2012, A&A, 548, A68 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  23. Schenewerk, M. S., Jewell, P. R., Snyder, L. E., Hollis, J. M., & Ziurys, L. M. 1988, ApJ, 328, 785 [NASA ADS] [CrossRef] [Google Scholar]
  24. Schlegel, D. J., Finkbeiner, D. P., & Davis, M. 1998, ApJ, 500, 525 [NASA ADS] [CrossRef] [Google Scholar]

Appendix A: Upper limits on undetected species

In most cases we have examined, the upper limits we can derive for undetected species are not sufficiently sensitive to be interesting. For instance, we derive 3σ upper limits N(l-C3H+) <2.3 × 1011  cm-2 and N(l-C3H+) < 6.0 × 1011  cm-2 toward BL Lac and 3C 111, respectively, but these are at most barely below the value N(l-C3H+) = 4.8 × 1011  cm-2 for the Horsehead (Pety et al. 2012) whose column densities are generally 3–10 times larger. For the neutral l-C3H the upper limits that may be derived from this work are not competitive with those accessible at lower frequency using the VLA and likewise with C4H (Liszt et al. 2012).

One species that is better-constrained is H2CS whose J = 3–2 transitions appear at 101.5 and 104.6 GHz. We find 3σ limits N(H2CO)/N(H2CS) > 11 ± 2.5 and N(H2CO)/N(H2CS) >32 ± 8 toward BL Lac and 3C 111 respectively using the prior results of Liszt et al. (2006), compared to N(H2CO)/N(H2CS) = 7 toward TMC-1 (Ohishi et al. 1992).

All Tables

Table 1

Sightlines and continuum targets.

Table 2

HCO 101–000 transitions observed and column density-optical depth conversionsa.

Table 3

c-C3H 212–111 transitions observed and column density-optical depth conversionsa.

Table 4

CF+ transitions observed and column density-optical depth conversionsa.

All Figures

thumbnail Fig. 1

Properties of the spectral sweep. Top: line profile integrals (equivalent widths) for the spectral sweep plotted against those previously obtained by synthesis at the Plateau de Bure Interferometer in our prior work. Bottom: 8-channel running mean channel-channel rms in the line/continuum ratio at 0-absorption for 3C 111 and BL Lac. The points labelled SO in the upper panel represent the 32–21 line at 99.3 GHz as discussed in Sect 2.1 of the text.

In the text
thumbnail Fig. 2

Spectra of BL Lac and 3C 111 in the region of the HCO quartet. The spectral resolution is 0.195 MHz. The approximate positions of HCO, H13CO+ and SiO J = 2–1 transitions are marked. For spectroscopic data, see Table 2.

In the text
thumbnail Fig. 3

Column density of HCO vs. H13CO+ (left) and C2H (right). For W49 only the “spiral arm” feature at 40 km s-1 is shown at left using data from the present work; at right the C2H data of Godard et al. (2010) is used for the 40 km s-1 and 60 km s-1 features. Data from the Horsehead PDR (TdC) are from Gerin et al. (2009). The dashed lines have power-law slope of unity, with proportions of 16:1 and 1:10 at left and right, respectively. Error bars in this and all other plots are 1σ statistical uncertainties.

In the text
thumbnail Fig. 4

W49 continuum, HCO F = 1–1 and H13CO+ emission and absorption spectra. Left: 86.8 GHz continuum (grayscale; peak =2.5 Jy) and H13CO+ emission (contours of integrated brightness temperature in units of Kelvins km s-1). The 4′′ synthesized HPBW is shown inset at lower left. Right top: spectra of H13CO+ and HCO toward the continuum peak labelled position “A” at left. Right bottom: emission spectra at off-continuum positions “B” (blue in electronic media) and “C” (red). Spectra showing strong emission are H13CO+, those without are spectra of HCO.

In the text
thumbnail Fig. 5

Spectra of BL Lac and 3C 111 in the region of the c-C3H quartet. The spectral resolution is 0.195 MHz. For spectroscopic data, see Table 3.

In the text
thumbnail Fig. 6

Column density of c-C3H vs. H13CO+ (at left), and C3H2 and C2H (right). Data for the hydrocarbons toward the Horsehead nebula (TdC) are from Pety et al. (2012). Data for TMC-1 are from Ohishi et al. (1992). The c-C3H survey data of Mangum & Wootten (1990) are shown at right, labelled “MW1990”. The solid line at left has power-law slope 1; at right, both lines have slope 0.8.

In the text
thumbnail Fig. 7

CF+ spectra toward BL Lac (upper) and 3C 111 (lower). The velocity axis corresponds to the mean LTE intensity-weighted centroid of the two hyperfine components noted in Table 4. Shown for comparison are scaled spectra of the 18.3 GHz transition of c-C3H2 from Liszt et al. (2012). For spectroscopic data, see Table 4.

In the text
thumbnail Fig. 8

Column density of CF+ vs. H13CO+ (at left) and C2H (right). Data for the Horsehead nebula (TdC) are from Guzmán et al. (2012a). The prediction N(CF+) ≈ 4.5 × 1011  cm-2 at AV 1 mag of Neufeld & Wolfire (2009) is shown as a blue dashed horizontal line. The power law slopes of the best-fit lines (0.59, 0.38) are noted.

In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.