New molecules in IRC+10216: confirmation of C 5 S and tentative identification of MgCCH, NCCP, and SiH 3 CN
aa r X i v : . [ a s t r o - ph . GA ] A ug Astronomy&Astrophysicsmanuscript no. c5s˙irc c (cid:13)
ESO 2018January 2, 2018
New molecules in IRC +10216: confirmation of C S and tentativeidentification of MgCCH, NCCP, and SiH CN ⋆ Marcelino Ag´undez , Jos´e Cernicharo , and Michel Gu´elin Instituto de Ciencia de Materiales de Madrid, CSIC, C / Sor Juana In´es de la Cruz 3, 28049 Cantoblanco, Spain Institut de Radioastronomie Millim´etrique, 300 rue de la Piscine, 38406 St. Martin d’H´eres, FranceReceived; accepted
ABSTRACT
The C-star envelope IRC + λ + Sconfirms a previous tentative identification of this molecule by Bell et al. (1993) based on a line at 24.0 GHz. We also report thetentative identification of three molecules not yet observed in space: MgCCH, the first metal acetylide detected in space, and NCCPand SiH CN, the phosphorus and silicon analogs of cyanogen (NCCN) and methyl cyanide (CH CN). We derive the following columndensities: N (C S) = (2-14) × cm − (depending on the rotational temperature adopted), N (MgCCH) = × cm − , N (NCCP) = × cm − , and N (SiH CN) = cm − . The S-bearing carbon chain C S is less abundant than C S, while MgCCH has anabundance in between that of MgNC and those of MgCN and HMgNC. On the other hand, NCCP and SiH CN are the least abundantP- and Si-bearing molecules observed to date in IRC + + Key words. astrochemistry – stars: circumstellar matter – stars: AGB and post-AGB – stars: carbon – stars: individual: IRC +
1. Introduction
The C-star envelope IRC + ∼ × − M ⊙ yr − , and being rel-atively nearby ( ∼
130 pc), it is the prototype of carbon star. Theinner wind is a factory of dust and relatively stable moleculeswhile in its outer layers the photochemistry driven by the pen-etration of interstellar UV photons produces a wealth of exoticmolecules.IRC + + ⋆ Based on observations carried out with the IRAM 30m Telescope.IRAM is supported by INSU / CNRS (France), MPG (Germany) andIGN (Spain). of antenna temperature. Here we report various weak lines ob-served within the λ S, confirming a previous tentative identification of thismolecule based on a line at cm-wavelengths (Bell et al. 1993),while others are tentatively attributed to the molecules MgCCH,NCCP, and SiH CN.
2. Observations
The observations were made with the IRAM 30-m telescope lo-cated at Pico Veleta (Spain) during several sessions, most ofthem between 2002 and 2008 in the context of a λ + ABCD receivers, currently replaced by
EMIR at the 30m tele-scope, were used. A SIS receiver operating at 3 mm was tunedin single sideband mode, with a typical image rejection largerthan 20 dB. Identification of image side band lines was done byshifting the frequency of the local oscillator. Data were takenin the wobbler switching observing mode by nutating the sec-ondary mirror by 3-4 ′ at a rate of 0.5 Hz. Pointing and focuswere checked by observing nearby planets and the quasar OJ287. The back end used was a filterbank with a bandwidth of512 MHz and a spectral resolution of 1.0 MHz. The intensityscale was calibrated using two absorbers at di ff erent tempera-tures and the atmospheric transmission model ATM (Cernicharo1985; Pardo et al. 2001). We express intensities in terms of T ∗ A ,the antenna temperature corrected for atmospheric absorptionand for antenna ohmic and spillover losses. The uncertainty in T ∗ A due to calibration is estimated to be around 10 %. The sys-tem temperature ranged from 100 to 150 K. On source integra- S, MgCCH, NCCP, and SiH CN in IRC + Table 1.
Line parameters observed in IRC + Transition ν calc ν obs V exp R T ∗ A d v(MHz) (MHz) (km / s) (K km / s) J ′ − J ′′ C S44-43 81192.98 81193.4(10) a b N ′ J ′ − N ′′ J ′′ MgCCH9 / -8 / b / -8 / b / -9 / b / -9 / b J ′ − J ′′ NCCP15-14 81131.69 81131.9(10) 14.5 b b J ′ K ′ − J ′′ K ′′ SiH CN c -8 -8 -9 a -9 -10 a b -10 σ uncertainties in units of the last dig-its. a Line blended. b Line width has been fixed. c The K = CN (the strongest ones) are fit to one single line. tion times ranged from 2 to 20 hours, resulting in T ∗ A rms noiselevels per 1 MHz channel ranging from less than 1 mK to 3 mK.
3. Results and discussion S The molecule C S is a relatively heavy linear carbon chain witha Σ + ground electronic state, whose rotational spectrum hasbeen recorded in the laboratory in the 5-18 GHz frequency range(Kasai et al. 1993; Gordon et al. 2001). Its electric dipole mo-ment has been calculated as 4.65 D (Pascoli & Lavendy 1998;P´erez-Juste et al. 2004). In their observations of IRC + J = −
12 rotational transition of C S.To our knowledge, this tentative detection has not been neitherconfirmed nor refuted by subsequent studies.Various rotational transitions of C S with upper level ener-gies E up in the range 90-170 K fall within the frequency cover-age of our 3 mm line survey. We have identified at the lower fre-quency side of the survey three weak emission lines whose fre-quencies coincide with the J = −
43 to J = −
45 transitionsof C S. The next J lines are not detected either because they fallin spectral regions observed with an insu ffi cient sensitivity orbecause they overlap with other strong lines (see Appendix A).The line parameters are given in Table 1 and the lines are shownin Fig. 1. The agreement between calculated and observed fre-quencies is better than 1 MHz, which is good enough taking intoaccount the moderate spectral resolution of the observed spec-tra (1 MHz) and the non-negligible errors of some hundreds ofkHz in the calculated frequencies at 3 mm (extrapolated fromlaboratory measurements at cm-wavelengths). In IRC + J = −
44 and J = −
45 lines of C S appear freeof blending with other lines, although the J = −
43 line over-laps partially with other weak lines arising from HC C CN andCCCC CH.
Fig. 1.
Portions of the 3 mm line survey of IRC + J = − J = −
44, and J = −
45 lines of C S.The J = −
43 line is blended with lines of HC C CN andCCCC CH. The T ∗ A rms noise levels per 1 MHz channel are 0.6,0.5, and 1.0 mK in the bottom, middle, and top panels, respec-tively. Note the coincidence in frequency of a weak shoulder atthe red part of a SiCN line with the Π / J = / − / H (middle panel). Thick gray lines are the calculated lineprofiles under LTE assuming that C S extends between 5 and20 ′′ with an homogeneous fractional abundance and rotationaltemperature of 18 K.Given the limited signal-to-noise ratio of the C S lines it isdi ffi cult to appreciate a distinct line shape, either U-shaped orflat-topped, which could permit us to infer whether the size ofthe emitting region is smaller or larger than the main beam of thetelescope at these frequencies (29-30 ′′ ). Based on the U-shapedline profiles observed at 3 mm for the related molecules C S andC S (Cernicharo et al. 1987) we could also expect an extendedemission for C S. Assuming a source size of radius 20 ′′ , as mostmolecules distributed as a hollow shell in IRC + N (C S) of 1.4 × cm − and a rotational temperature T rot of 18 K. The column density issomewhat lower than that reported by Bell et al. (1993), mainlybecause of the lower source size adopted by these authors. Thederived values of T rot and N (C S) rely to a large extent on the0.8 cm line ( E up = E up = T rot needsto be fixed and N (C S) becomes very sensitive to the adoptedchoice of T rot . For example, for a rotational temperature of 44K (as measured for the related molecule C S; see Table 2) thecolumn density of C S decreases by almost one order of mag-nitude to 2 × cm − . The 3 mm lines involve high energy S, MgCCH, NCCP, and SiH CN in IRC + Table 2.
Rotational temperatures and column densities
Molecule T rot (K) N (cm − ) ReferenceC S 18(1) 5.0(3) × [1]C S 44(2) 1.7(1) × [1]C S 20 a < × b [2]C S 18-44 (2-14) × [2]MgCCH 20 a × [2]NCCP 20 a × [2]SiH CN 39 a × [2]Numbers in parentheses are 1 σ uncertainties in units of the last dig-its. A source size of radius 20 ′′ has been adopted. a T rot has beenfixed. b σ upper limit.References: [1] IRAM 30-m data at λ
3, 2, and 1.3 mm. [2] Thisstudy. levels and have Einstein coe ffi cients for spontaneous emission(and thus critical densities) about 40 times higher than the 0.8cm line, probably implying that they are subthermally excited inthe outer layers of IRC + S is not character-ized by a uniform rotational temperature. For T rot = S has a column density in the range (2-14) × cm − . Fig. 1 shows the line profiles calculated under local ther-modynamic equilibrium (LTE) adopting the parameters of theenvelope from Ag´undez et al. (2012) and assuming that C S ispresent in a hollow shell extending from 5 to 20 ′′ , with an ho-mogeneous fractional abundance and rotational temperature of18 K, which may be adequate for the 3 mm lines, as discussedabove. The agreement between modeled and observed line pro-files is fairly good.The molecule C S is the largest member of the series ofsulfur-containing carbon chains C n S observed in space. Theshorter members C S and C S are observed in various astronom-ical regions, although C S has never been observed. The columndensities derived for these molecules in IRC + S and C S the column densitiesare quite robust as they result from a large number of rotationallines observed across the 3, 2, and 1.3 mm bands. That of C Sis just an upper limit while that of C S has a large uncertainty,which however could be reduced if sensitive observations areperformed at frequencies around 40 GHz, where the brightestC S lines are expected. This spectral region has been surveyedby Kawaguchi et al. (1995) although no line attributable to C Swas found, probably due to an insu ffi cient sensitivity. The car-bon chains C n S seem to show a decline in abundance as thelength of the chain increases (albeit the uncertainty in the valueof C S), as occurs with polyacetylenic and cyanopolyacetilenicchains (see e.g. Ag´undez et al. 2008a). There is no obvious al-ternation in the abundance between chains with n odd ( Σ elec-tronic ground state) and those with an even number of carbonatoms ( Σ electronic ground state), as may be occurring for theseries C n O, of which only C O, but not C O, has been detectedin IRC + S is more than twice that ofC S (see Table 2), which points to di ff erences in the excitationand may be indicative of radiative pumping to vibrationally ex-cited states playing an important role in the case of C S (see e.g.,Ag´undez et al. 2008a).The chemical routes leading to S-bearing carbon chains inIRC + + C n H → C n S + H , n = , , ... (1) Fig. 2.
Parts of the 3 mm line survey of IRC + N = − N = − T ∗ A rms noise levels per1 MHz channel are 0.5 and 0.7 mK in the bottom and top pan-els, respectively. Thick gray lines correspond to the calculatedline profiles under LTE assuming that MgCCH extends between5 and 20 ′′ with an homogeneous fractional abundance and rota-tional temperature of 20 K.for S-containing carbon chains with an even number of carbonatoms, while those with an odd number of carbon atoms wouldform through reactions of the typeCS + C n H → C n + S + H , n = , , ... (2)The model of Millar et al. (2001) yields results in reasonableagreement with our observed column densities of C n S chains,although these are somewhat overestimated by the model prob-ably because of the too high initial abundance of CS adopted (4 × − relative to H ), which may need to be revised down by afactor of 5-6 (Ag´undez et al. 2012). A variety of metal cyanides has been observed in the circumstel-lar gas of IRC + + + Σ + electronic ground state, whose purerotational spectrum has been characterized in the laboratory inthe spectral range 315-525 GHz (Brewster et al. 1999). MgCCHhas a moderate electric dipole moment of 1.68 D, according tothe ab initio calculations of Woon (1996), which indicates thatthe ethynyl-metal bond probably has a fair degree of covalentcharacter.Three rotational transition doublets of MgCCH fall withinour 3 mm line survey. The N = − N = − S, MgCCH, NCCP, and SiH CN in IRC + and 99.3 GHz, respectively, coincide in frequency with two pairsof non-blended weak lines (see Fig. 2) while the N = − ffi cient to detect lines with antenna temperatures ofa few mK (see Appendix A). The agreement between observedand calculated frequencies is reasonably good, less than 1 MHz(see Table 1). It must however be taken into account that the ob-served lines attributed to MgCCH are weak and the number oflines with similar intensities is relatively large in this region ofthe spectrum and at this level of sensitivity. The possibility ofaccidental coincidence cannot be completely ruled out althoughthe assignment to MgCCH is reinforced by the fact that the tran-sitions consist of doublets and both the N = − N = − ′′ ). Magnesium isocyanide has been found to bedistributed in the form of a hollow shell of radius 15-20 ′′ , asmapped with the IRAM Plateau de Bure interferometer at mil-limeter wavelengths (Gu´elin et al. 1993). Based on the U-shapedprofiles and the distribution of the related molecule MgNC wecan assume a source size of radius 20 ′′ for MgCCH. Adopting arotational temperature of 20 K, of the order of other moleculesdistributed in the outer layers of IRC + N (MgCCH) of 2 × cm − ,that is, ∼ ∼ + ffi ciently abundant precursors in the circumstellargas. From these studies it turns out that the most likely forma-tion route of MgCCH probably follows closely that proposed byPetrie (1996) for the synthesis of magnesium cyanides, whichconsists of the radiative association between Mg + and largecyanopolyynes followed by the dissociative recombination ofthe ionic complex with free electrons. In the case of MgCCH,the radiative recombination would take place between Mg + andlarge polyynes, leading to Mg(C n H ) + complexes, whose recom-bination with electrons could yield MgCCH, among other pos-sible fragments. A simple chemical model similar to that de-scribed in Cabezas et al. (2013) adopting the rate constants ofradiative association between Mg + and polyynes calculated byDunbar & Petrie (2002) indicates that this route is capable ofproducing MgCCH with an abundance high enough to accountfor the column density derived from observations. An alternativeroute could be the exchange reaction between MgNC and CCHto yield MgCCH and CN (the precursor MgNC being about 6times more abundant than MgCCH in IRC + + ffi cult to detect if, as found in the case of Mg, they areless abundant than the corresponding (iso)cyanides. For exam-ple, AlNC is detected in IRC + Fig. 3.
Parts of the 3 mm line survey of IRC + J = − J = − J = −
17 rotational lines of NCCP. The J = −
15 line ismarginally detected. The T ∗ A rms noise levels per 1 MHz channelare 0.4, 1.2, and 0.8 mK in the bottom, middle, and top panels,respectively. Line profiles calculated under LTE assuming thatNCCP extends between 5 and 20 ′′ with an homogeneous frac-tional abundance and rotational temperature of 20 K are shownas thick gray lines.AlCCH, with N (AlCCH) / N (AlNC) .
22 (Cabezas et al. 2012).The low dipole moment of AlCCH ( ∼ ffi cult toimpose a stringent upper limit on its abundance. From a chemical point of view, cyanophosphaethyne (NCCP)can be viewed as a cyanogen molecule (NCCN) in which a ni-trogen atom has been substituted by a phosphorus atom. Froma spectroscopic point of view, NCCP is a linear molecule witha closed-shell singlet electronic ground state. Its pure rota-tional spectrum has been characterized in the laboratory overa broad spectral range, from 25 to 820 GHz (Bizzocchi et al.2001). Having a moderately large dipole moment of 3.44 D,as measured by Cooper et al. (1980), it has been suggestedthat this molecule could be a good candidate for detectionin astronomical sources such as IRC + , and CCP have been already detected (Gu´elin et al. 1990,2000; Ag´undez et al., 2007, 2008b, 2014; Milam et al. 2008;Tenenbaum & Ziurys 2008; Halfen et al. 2008).Seven rotational lines of NCCP from J = −
14 to J = −
20, with upper level energies in the range 30-60 K, fallwithin the 3 mm line survey. A weak line with T ∗ A ∼ S, MgCCH, NCCP, and SiH CN in IRC + ing at 97355.7 ± J = −
17 transition of NCCP (see Table 1 andFig. 3). To confirm whether this assignment is correct we havesearched for other rotational transitions of NCCP within the 3mm band. A weak line at 81131.9 ± J = −
14 transition while the J = −
15 can beidentified with a marginal line at 86539.5 ± ffi cient to detect such weaklines or are a ff ected by severe blendings with other strong lines(see Appendix A). The agreement between calculated and ob-served frequencies is reasonably good and, within the sensitivityreached, there are not missing lines in the 3 mm line survey. Onthe other hand, line confusion and accidental coincidence maybe an issue at this level of sensitivity in the 3 mm spectra ofIRC + ffi cult to distinguish whether the lineprofiles are more U-shaped or flat-topped. Assuming that thismolecule is formed in the outer layers, as seems to be the caseof the related radical CCP (Halfen et al. 2008), we may adoptfor NCCP a distribution size with a radius of 20 ′′ and a rota-tional temperature of 20 K, typical parameters for molecules dis-tributed in the outer shells. The column density of NCCP derivedis then 7 × cm − , about 50-100 times less abundant than CPand slightly less abundant than CCP. The column densities of CPand CCP in IRC + × and 1.2-2.9 × cm − , respectively, according to Gu´elin et al. (1990),Halfen et al. (2008), and an analysis of our IRAM 30-m data. Ascan be seen in Fig 3, the line profiles of NCCP calculated un-der LTE assuming a rotational temperature of 20 K show a goodagreement with the observed ones.The chemistry of phosphorus-containing molecules in cir-cumstellar envelopes such as IRC + ffi cient formation route to NCCN as long asthe reaction is barrierless and both HNC and CN are abundant inIRC + ffi ciently formed in IRC + + CP → NCCP + H , (3)which is slightly exothermic, by ∼
15 kcal mol − adopting theformation enthalpy of NCCP from Pham-Tran et al. (2001), andwhose reactants are abundant enough in IRC + CN Silyl cyanide, SiH CN, is a prolate rotor with a closed electronicshell. Its rotational spectrum has been recorded in the laboratoryat microwave and millimeter wavelengths and its electric dipolemoment has been experimentally determined to be as large as3.44 D (Priem et al. 1998).
Fig. 4.
Portions of the 3 mm line survey of IRC + J = − J = −
10 rotational lines of SiH CN. Each J line is composed of a set of K components whose line profilesoverlap. The K = T ∗ A rms noise levels per 1 MHz channelare 0.5, 0.6, and 1.2 mK in the bottom, middle, and top panels,respectively. The J = − J = − J = − Nin its ν = CN. Thick gray lines correspond to the cal-culated line profiles under LTE assuming that SiH CN extendsbetween 5 and 20 ′′ with an homogeneous fractional abundanceand rotational temperature of 39 K.Three rotational transitions of SiH CN fall within the 3 mmline survey, J = − J = −
10, and the three can beidentified with weak emission features in the observed spectrum.Because lines are relatively broad in IRC + ∼
29 km s − or ∼
10 MHz at 3 mm, each transition of SiH CN is expected toshow a complex profile resulting from the overlap of the di ff er-ent K components, the K = CN lines while Table 1 gives the line pa-rameters. Only the positions of the K = J = − J = − J = −
10 overlap partiallyat the red side with a weak line of C N in its ν = J = − ∼ CN helps to bring some light into theinterpretation of the observed emission features. Assuming thatSiH CN is distributed as a hollow shell of radius between 5 and S, MgCCH, NCCP, and SiH CN in IRC + ′′ with an homogeneous rotational temperature of 39 K, thatderived for the related molecule CH CN according to our IRAM30-m data, a LTE excitation calculation yields the line profilesshown in Fig. 4. It is seen that the line profiles of SiH CN aredominated by the K = K components at lower frequencies. Theobserved emission features cannot be fully explained by the C N ν = CN. However, because twoof the three observed transitions of SiH CN are partially blendedwith other lines and because of the limited signal-to-noise ra-tio achieved, for the moment we consider the identification ofSiH CN in IRC + cm − is derived forSiH CN adopting a rotational temperature of 39 K and a sourcesize of radius 20 ′′ , which makes SiH CN the least abundant Si-containing molecule detected to date in IRC + ∼ × cm − ; Gu´elin et al. 2000, 2004).The synthesis of SiH CN is quite uncertain, although the analo-gous species CH CN has been better studied. A quite direct syn-thetic pathway to CH CN would be the reaction between CH and CN although it has an activation barrier and thus becomestoo slow at low temperatures (Sims et al. 1993). In the absenceof better constraints we could expect that the route to SiH CNfrom SiH and CN is also probably closed at low temperatures.A more e ffi cient route to CH CN in IRC + CNH + , whose dissociative recombination with electronsyields both CH CN and CH CN (Ag´undez et al. 2008a). In thecase of SiH CN, a similar route involving the ion SiH CNH + could also work, although the details on the chemical kinetics ofthe reactions involved are yet to be investigated.
4. Summary
We have detected various weak lines in the course of a IRAM30-m λ + S, confirming a pre-vious tentative detection of this molecule by Bell et al. (1993)based on a line at 0.8 cm, while some others are tentatively as-signed to rotational transitions of three new molecules not yetobserved in space. These three molecules are MgCCH, the firstmetal acetylide detected in space, and NCCP and SiH CN, twoexotic molecules which are the phosphorus and silicon analogsof cyanogen (NCCN) and methyl cyanide (CH CN).The S-bearing carbon chain C S is likely to have a non uni-form rotational temperature, with values in the range 18-44 K,which translates to column densities in the range (2-14) × cm − , implying that C S is less abundant than C S in the en-velope of IRC + CN wederive column densities of the order of 10 cm − , which im-plies that MgCCH has an abundance in between that of MgNCand those of MgCN and HMgNC, while NCCP and SiH CN be-come the least abundant P- and Si-bearing molecules observedto date in IRC + + S probably occurs in asimilar fashion to other sulfur-containing carbon chains such asC S, while that of MgCCH probably takes place through chem-ical routes similar to those proposed for the formation of metalcyanides. The formation of NCCP and SiH CN is more uncer-tain and is suggested to occur through analogous pathways tothose leading to NCCN and CH CN. The identification of these four molecules through such weaklines suggests that pushing observations towards very low noiselevels will probably bring a good number of molecular discover-ies in chemically rich environments such as IRC + Acknowledgements.
M.A. and J.C. thank Spanish MINECO for funding sup-port through grants CSD2009-00038, AYA2009-07304, and AYA2012-32032.We thank the anonymous referee for a critical and constructive report.
References
Ag´undez, M., Cernicharo, J., & Gu´elin, M. 2007, ApJ, 662, L91Ag´undez, M., Fonfr´ıa, J. P., Cernicharo, J., et al. 2008a, A&A, 479, 493Ag´undez, M., Cernicharo, J., Pardo, J. R., et al. 2008b, A&A, 485, L33Ag´undez, M., Cernicharo, J., Gu´elin, M., et al. 2010, A&A, 517, L2Ag´undez, M., Fonfr´ıa, J. P., Cernicharo, J., et al. 2012, A&A, 543, A48Ag´undez, M., Cernicharo, J., Decin, L., et al. 2014, ApJ, 790, L27Bell, M. B., Avery, L. W., & Feldman, P. A. 1993, ApJ, 417, L37Bizzocchi, L., Thornwirth, S., M¨uller, H. S. P. et al. 2001, J. Mol. Spec., 205,110Brewster, M. A., Apponi, A. J., Xin, J. & Ziurys, L. M. 1999, Chem. Phys. Lett.,310, 411Cabezas, C., Mata, S., Daly, A. M., et al. 2012, J. Mol. Spec., 278, 31Cabezas, C., Cernicharo, J., Alonso, J. L., et al. 2013, ApJ, 775, 133Cernicharo, J. 1985, IRAM Internal Report 52Cernicharo, J., Gu´elin, M., Hein, H., & Kahane, C. 1987, A&A, 181, L9Cernicharo, J. & Gu´elin, M. 1996, A&A, 309, L27Cernicharo, J., Gu´elin, M., & Kahane, C. 2000, A&AS, 142, 181Cernicharo, J., Gu´elin, M., Ag´undez, M., et al. 2007, A&A, 467, L37Cernicharo, J., Ag´undez, M., & Gu´elin, M. 2011, in The Molecular Universe,ed. J. Cernicharo & R. Bachiller (Cambridge Univ. Press), IAU Symp. 280,237Cooper, T. A., Kroto, H. W., Nixon, J. F., & Ohashi, O. 1980, J. Chem. Soc.Chem. Commun., 8, 333Dunbar, R. C. & Petrie, S. 2002, ApJ, 564, 792Gordon, V. D., McCarthy, M. C., Apponi, A. J., & Thaddeus, P. 2001, ApJS, 134,311Gu´elin, M., G´omez-Gonz´alez, J., Cernicharo, J., & Kahane, C. 1986, A&A, 157,L17Gu´elin, M., Cernicharo, J., Paubert, G., & Turner, B. E. 1990, A&A, 230, L9Gu´elin, M., Lucas, R., & Cernicharo, J. 1993, A&A, 280, L19Gu´elin, M., Neininger, N., & Cernicharo, J. 1998, A&A, 335, L1Gu´elin, M., Lucas, R., & Neri, R. 1997, in CO: Twenty-Five Years of MillimeterWave Spectroscopy, ed. W. B. Latter et al. (Dordrecht: Kluwer), IAU Symp.170, 359Gu´elin, M., Muller, S., Cernicharo, J., et al. 2000, A&A, 363, L9Gu´elin, M., Muller, S., Cernicharo, J., et al. 2004, A&A, 426, L49Halfen, D. T., Clouthier, D. J., & Ziurys, L. M. 2008, ApJ, 677, L101Kasai, Y., Obi, K., Ohshima, Y., et al. 1993, ApJ, 410, L45Kawaguchi, K., Kagi, E., Hirano, T., et al. 1993, ApJ, 406, L39Kawaguchi, K., Kasai, Y., Ishikawa, S.-I., & Kaifu, N. 1995, PASJ, 47, 853MacKay, D. D. S. & Charnley, S. B. 2001, MNRAS, 325, 545McCarthy, M. C., Gottlieb, C. A., Gupta, H., & Thaddeus, P. 2006, ApJ, 652,L141Milam, S. N., Halfen, D. T., Tenenbaum, E. D., et al. 2008, ApJ, 684, 618Millar, T. J., Flores, J. R., & Markwick, A. J. 2001, MNRAS, 327, 1173Millar, T. J. 2008, Ap&SS, 313, 223Ohishi, M., Kaifu, N., Kawaguchi, K., et al. 1989, ApJ, 435, L83Pardo, J. R., Cernicharo, J., & Serabyn, E. 2001, IEEE Trans. Antennas Propag.,49, 1683Pascoli, G. & Lavendy, H. 1998, Int. J. Mass Spectr., 181, 11P´erez-Juste, I., Gra˜na, A. M., Carballeira, L., & Mosquera, R. A. 2004,J. Chem. Phys., 121, 10447Petrie, S. 1996, MNRAS, 282, 807Petrie, S., Millar, T. J., & Markwick, A. J. 2003, MNRAS, 341, 609Pham-Tran, N.-N., Hajgat´o, B., Veszpr´emi, T., & Nguyen, M. T. 2001, Phys.Chem. Chem. Phys., 3, 1588Priem, D., Cosl´eou, J., Demaison, J., et al. 1998, J. Mol. Spec., 191, 183Sims, I. R., Que ff elec, J.-L., Travers, D., et al. 1993, Chem. Phys. Lett., 211, 461Tenenbaum, E. D., Apponi, A. J., Ziurys, L. M., et al. 2006, ApJ, 649, L17Tenenbaum, E. D. & Ziurys, L. M. 2008, ApJ, 680, L121Woon, D. E. 1996, ApJ, 456, 602Yang, D. L., Yu, T., Lin, M. C., & Melius, C. F. 1992, J. Chem. Phys., 97, 222Ziurys, L. M., Apponi, A. J., Gu´elin, M., & Cernicharo, J. 1995, ApJ, 445, L47Ziurys, L. M., Savage, C., Highberger, J. L., et al. 2002, ApJ, 564, L45 S, MgCCH, NCCP, and SiH CN in IRC + Appendix A: Spectra covering detected and nondetected lines in the 3 mm band
In the case of silyl cyanide the three rotational transitions cov-ered within the 3 mm band are observed, but for the moleculesC S, MgCCH, and NCCP not all rotational transitions lying inthe 3 mm band (80-116 GHz) are detected. The lines which arenot detected either fall in spectral regions where the sensitivityachieved is not good enough or are blended with stronger linesof other molecules.In Fig. A.1 we show the spectra covering the J = − J = −
49 rotational transitions of C S, lying in the lowfrequency side (80-93 GHz) of the 3 mm band. Only the threerotational transitions lying at the lowest frequencies ( J = − J = −
44, and J = −
45) are detected (lines also shown inFig. 1 and already discussed in section 3.1). The spectral regionswhere the next higher J lines of C S are expected are shown inthe upper panels of Fig. A.1. None of these transitions can beidentified with clear emission features in the observed spectra.The spectral region around the J = −
46 and J = −
48 tran-sitions are relatively noisy. The antenna temperature rms noiselevel per 1 MHz channel is 1.5 mK around the J = −
46 tran-sition, while the expected C S line intensity is similar, makingvery di ffi cult to ascertain whether or not the weak C S line ispresent. The situation is even worse for the J = −
48 transi-tion, with an expected antenna temperature of just ∼ J = − J = −
49 transitions of C S are severely blended withother lines much stronger. At higher frequencies the chances ofobserving a C S line become lower because the expected C Sline intensities decrease as J increases (C S is a quite heavy ro-tor and levels with J >
50 become poorly populated at rota-tional temperatures around 20 K) and also because in our 3 mmline survey of IRC + N = − N J = − rotational transition of CN,while the high frequency component cannot be appreciated dueto an insu ffi cient sensitivity (the T ∗ A rms noise level is 1.8 mKper 1 MHz in this spectral region and the MgCCH component isexpected with an intensity of just ∼ J = − J = −
15, and J = −
17) are identifiedwith weak emission features (see also Fig. 3 and section 3.3).The J = −
16 transition falls in a spectral region crowdedby stronger lines arising from CH CN, and unidentified line at91953 MHz, and CH CN, making very di ffi cult to infer its pres-ence. The J = −
18 transition is partially blended with astronger line corresponding to C N − , and in any case the limitedsensitivity of the spectrum (rms of 1.5 mK per 1 MHz) makesit very di ffi cult to distinguish the NCCP line from the noise. Aneven worse situation occurs in the case of the J = −
19 tran-sition, which falls in a spectral region where it partially overlapswith a line of C P and the data is quite noisy (rms of 2.1 mK per1 MHz). The last transition of NCCP within the 3 mm band is the J = −
20, which falls in a spectral region where the noise level
Fig. A.1.
Spectra covering the C S lines in the 80-93 GHz fre-quency range. Detected lines are indicated by red arrows and nondetected lines by red dashed lines. LTE calculated line profilesare shown in red. S, MgCCH, NCCP, and SiH CN in IRC + Fig. A.2.
Spectra covering the MgCCH lines in the 3 mm band.Detected lines are indicated by red arrows and non detected linesby red dashed lines. LTE calculated line profiles are shown inred.is moderately low (rms of 1.1 mK per 1 MHz if we do not con-sider the much noiser region at frequencies higher than 113600MHz), although it is not low enough to allow for a clear detec-tion of lines with antenna temperatures of ∼ Fig. A.3.