Hydrogen chloride in diffuse interstellar clouds along the line of sight to W31C (G10.6-0.4)
R. R. Monje, D. C. Lis, E. Roueff, M. Gerin, M. De Luca, D. A. Neufeld, B. Godard, T. G. Phillips
aa r X i v : . [ a s t r o - ph . GA ] F e b Hydrogen chloride in diffuse interstellar clouds along the line ofsight to W31C (G10.6-0.4)
R. R. Monje, D. C. LisCalifornia Institute of Technology, MC 301–17, 1200 E. California Blvd., Pasadena, CA91125-4700, USA [email protected]
E. RoueffObservatoire de Paris-Meudon, LUTH UMR 8102, 5 Pl. Jules Janssen, F-92195 MeudonCedex, FranceM. Gerin, M. De LucaLERMA, CNRS, Observatoire de Paris and ENS, FranceD. A. NeufeldDepartment of Physics and Astronomy, Johns Hopkins University, 3400 North CharlesStreet, Baltimore, MD 21218, USAB. GodardDepartamento de Astrof´ısica, Centro de Astrobiolog´ıa (CAB), INTA-CSIC, Crta. Torrej´onkm 4, 28850, Torrej´on de Ardoz, Madrid, Spainand T. G. PhillipsCalifornia Institute of Technology, MC 301–17, 1200 E. California Blvd., Pasadena, CA91125-4700, USAReceived ; accepted 2 –
ABSTRACT
We report the detection of hydrogen chloride, HCl, in diffuse molecular cloudson the line of sight towards the star-forming region W31C (G10.6-0.4). The J = 1 − Cl and H Cl, are observedusing the 1b receiver of the Heterodyne Instrument for the Far-Infrared (HIFI)aboard the Herschel Space Observatory. The HCl line is detected in absorption,over a wide range of velocities associated with diffuse clouds along the line ofsight to W31C. The analysis of the absorption strength yields a total HCl columndensity of few 10 cm − , implying that HCl accounts for ∼ . ∼ Cl + and HCl + , for which large column densities have also been reported on thesame line of sight. The source of discrepancy between models and observationsis still unknown; however, the detection of these Cl-bearing molecules, provideskey constraints for the chlorine chemistry in the diffuse gas. Subject headings: astrochemistry — submillimeter: ISM – ISM: molecules–ISM:abundances
1. Introduction
Hydride molecules play an important role in interstellar chemistry, as they are oftenstable end points of chemical reactions, or represent important intermediate stages of thereaction chains theorized to form gas-phase molecules. This makes hydrides a sensitive testof these chemical models, as well as potential tracers of other molecules of interest e.g.molecular hydrogen. Due to their small moment of inertia, hydrides have their fundamentalrotational lines in the submillimeter band. The Heterodyne Instrument for the Far-Infrared(HIFI) on board the Herschel Space Observatory is providing invaluable data on interstellarchemistry in general, and in particular on hydride molecules within the Milky Way and localgalaxies. Halogen atoms fluorine (F) and chlorine (Cl), with estimated solar abundances of3.6 × − and 3.2 × − relative to hydrogen (Asplund et al. 2009), are of special interestbecause they are the major atoms (neutral F and ionic Cl + ) in diffuse environments thatreact exothermically with molecular hydrogen to form hydride molecules (HF and HCl).HF and HCl are strong bound systems, only destroyed by photodissociation, reactions withHe + , H +3 , and C + and, in the case of HCl, also by photoionization. Chlorine chemistry hasbeen determined by extensive theoretical and observational work (Jura 1974; Dalgarno et al.1974; van Dishoeck & Black 1986; Blake et al. 1986; Schilke et al. 1995; Federman et al.1995). In diffuse clouds, Cl atoms can be ionized by UV photons at wavelengths between91.2 and 95.6 nm (Jura 1974). The resulting ion, Cl + , reacts rapidly exothermically (by0.17 eV) with H to form HCl + Cl + + H → HCl + + H , (1)which in turn reacts with H to form chloronium, H Cl + HCl + + H → H Cl + + H . (2) 4 –HCl is then formed through dissociative recombination (DR) of H Cl + H Cl + + e → Cl + H HCl + H . (3)From reactions (1) – (3), HCl is expected to be abundant in regions containing bothH and chlorine-ionizing photons.Observations of HCl have been limited to ground-based observatories at high-altitudesites, under extreme good weather conditions, or space-missions (due to its largerotational constant which places the ground state transition, at 625.9187 GHz, near astrong atmospheric water absorption line). The first detections of interstellar chlorine-containing molecules were obtained using the NASA’s Kuiper Airborne Observatory (KAO).Blake et al. (1985) observed the HCl ground-state J = 1 − ′ . Zmuidzinas et al. (1995) detected the line inabsorption towards Sagittarius B2. Following these results, ground-based observations havelead to the detection of the H Cl and H Cl isotopologues, in environments such as evolvedstars and active star-forming regions (Schilke et al. 1995; Salez et al. 1996; Peng et al.2010).Cl chemistry in diffuse clouds on the other hand, has not been fully constrainedbecause of the paucity of observations of Cl-bearing species in the interstellar medium.Prior to Herschel/HIFI, HCl in diffuse clouds was observed mainly by means of ultravioletabsorption studies leading only to a tentative detection towards ζ Oph (Federman et al.1995), with an HCl column density of 2 . × cm − . Given an atomic chlorine columndensity of 3.0 × cm − for this line of sight (Federman et al. 1995), the N (HCl)/ N (Cl)ratio is 9 × − . This result is in good agreement with the theoretical models of the Clchemistry (van Dishoeck & Black 1986 and more recently Neufeld & Wolfire 2009, hereafterNW09). The NW09 models predict column densities in the range of 10 – 10 cm − for 5 –HCl, H Cl + and HCl + , for a diffuse molecular cloud of density n H = 10 . cm − and χ UV in the range of 1 – 10, where χ UV is the UV radiation field normalized with respect to themean interstellar value (Draine 1978; see Figures 6 and 7 in NW09). Model predictions alsoidentified H Cl + and HCl + as potentially detectable Cl-bearing species.The Herschel/HIFI instrument has indeed allowed observations for the first time ofthese two new Cl-bearing species in the interstellar medium, as well as the detection ofHCl in diffuse clouds (Lis et al. 2010; De Luca et al. 2012) and in protostellar shocks(Codella et al. 2012). The molecular ions H Cl + and HCl + have been detected in diffuseclouds on the lines of sight to bright submillimeter continuum sources. H Cl + has beendetected in absorption towards NGC 6334I, Sagittarius B2 (S) (Lis et al. 2010), W31C,Sgr A and in emission towards Orion Bar, Orion South (Neufeld et al. 2012). The H Cl + column densities obtained from these studies, imply that chloronium accounts for ∼ ∼
10 larger thanthat predicted by the chemical models, which predict a H Cl + and HCl + joint contributionof ∼ + towards W31C and W49N (De Luca et al. 2012),where HCl + has been detected in absorption with large column densities suggesting a3 – 5 % contribution of this species to the total gas-phase chlorine content.The discrepancy between the chlorine chemistry models and observations is verypuzzling; diffuse clouds contain only simple molecules, and the number of reactions involvedin describing their abundances is fewer than in dense molecular clouds. Therefore, a test ofthe basic interstellar chemistry should be possible, establishing much higher standards forthe modeling of diffuse clouds compared to those of dense clouds. The diffuse cloud modelsthus should be able to reproduce the measured abundances to a factor of two or better(van Dishoeck 1990). 6 –In this paper, we report the result of Herschel/HIFI observations of HCl J = 1 − ∼ − intheir observations. W31C is an extremely active region of high-mass star formation, andone of the three bright HII regions of the W31 complex, harboring an extremely luminoussubmillimeter and infrared continuum source (L IR ∼ L ⊙ , Wright et al. 1977). Witha kinematic distance of 4.8 +0 . − . kpc (Fish et al. 2003), the W31C line of sight intersectsseveral foreground molecular clouds from the Milky Way spiral arms. The source locationand its strong continuum flux, make W31C one of the best sources towards which to carryout absorption line studies. We compare the Herschel/HIFI results with current chlorinechemistry models and with observations of other Cl-bearing molecules detected along thesame line of sight and toward the galactic center source Sgr B2(S).
2. Observations
The observations were performed with the Herschel/HIFI instrument (de Graauw et al.2010) on March 2 2010, as part of the PRISMAS (Probing Interstellar Molecules withAbsorption lines Studies) guaranteed time key program. Both H Cl and H Cl J = 1 − ′ on either side of the sourceposition ( α J = 18 h m s and δ J = -19 ◦ ′ ′′ ) along an East-West axis.The data have been reduced using the Herschel Interactive Processing Environment 7 –(HIPE) (Ott et al. 2010) with pipeline version 5.2. The resulting Level 2 double sidebandspectra were exported into FITS format for subsequent data reduction and analysis usingthe IRAM GILDAS package . The spectra obtained at different LO setting and for eachpolarization were inspected for possible contamination by emission lines from the imagesideband and averaged to produce the final spectra shown in Figure 1. Beam measurements,reported on November 17 of 2010, towards Mars at 610 GHz give a main beam ( η mb ) of0.744 and 0.764, for the Horizontal (H) and vertical (V) polarization, respectively. The fullwidth at half maximum (FWHM) HIFI beam size at the HCl J = 1 − ∼ ′′ .
3. Results
The spectra obtained for the two stable hydrogen chloride isotopologues, H Cl andH Cl towards W31C are shown in Figure 1, upper and lower panels respectively. The dataquality is excellent with a double sideband continuum antenna temperature of ∼ ∼ Cl and H Cl emission lines from thebackground source and a clear absorption feature in the H Cl spectrum, at velocities ∼
10 – 50 km s − . The emission lines from the source will be discussed in an extendedpaper, which will include the HCl observations towards all PRISMAS sources (Monje et al.in prep.). The H Cl absorption feature is analogous to the one seen in the spectra of othermolecules such as HF, CH and H O (see Figure 2 in Neufeld et al. 2012) and correspondsto the foreground clouds on the line of sight towards W31C. The foreground clouds areknown to harbor low density and cool molecular gas (T kin ∼
50 - 70 K from H and ∼
100 K from H I), where molecular excitation can be highly subthermal, with excitationtemperatures close to the cosmic microwave background radiation ∼ http://iram.fr/IRAMFR/GILDAS/ 8 –2010). Consequently, molecular emission lines are very weak and these clouds are beststudied through absorption spectroscopy.The corresponding H Cl absorption line is contaminated by interfering emission ofdimethyl ether (CH OCH ), methyl cyanide (CH CN, v=0) and sulphur dioxide (SO ). Theemission lines are modeled with the program XCLASS (Schilke et al. 1998; Comito et al.2005), which assumes local thermodynamic equilibrium (LTE) and uses the rest frequenciesprovided by the Cologne Database for Molecular Spectroscopy (CDMS) (M¨uller et al.2001; M¨uller et al 2005) and the Jet Propulsion Laboratory (JPL) molecular spectroscopydatabase (Pickett et al. 1998). The result from the line fits exhibits two relatively strongCH CN and CH OCH lines and a weak SO line within the H Cl absorption spectrum,see Figure 2. To obtain the best LTE fit for CH OCH and SO additional lines withinthe same LO tuning are used, see Figure 2. For the CH CN additional transition linesat 533.123 GHz ( J = 29 −
28) and 606.533 GHz ( J = 33 −
32) are used, see Figure 3.The parameters used in the LTE model, shown in Table 1, assume that the CH CN andSO are tracers of the hot core (Beltr´an et al. 2011) while the CH OCH arises from themore extended envelope. However, the W31C region is known to be a complex source withoutflow activity from embedded high-mass protostars, methanol masers, a dense rotatingtoroid and infalling material from the molecular envelope. The LTE emission line fit modelis thus an approximation and tighter constraints using the combination of key moleculessuch as methanol, formaldehyde, CO, CS and SO and more complex chemical modelingare needed. Due to the uncertainties of the modeling, we only use the contamination-freespectra (grey shaded area in Figure 2) to derive an estimate of the Cl/ Cl isotopic ratio,as described below.The rotational levels of HCl have a hyperfine splitting, caused by the interactionof the quadrupole moment of the chlorine nucleus (I = 3/2) and the electric field. The 9 – J = 0 − h ) of 0, -6.3, and +8.2 km s − forH Cl, and 0, -5 and 6.5 km s − for H Cl. The ∆v h is larger than the velocity differencebetween consecutive velocity components, producing overlaps between different velocitycomponents that results in a complex profile. In order to determine the velocity structureof the foreground absorbing gas, we extract the signal associated with each HFS componentusing the numerical procedure described in Godard et al. (2012). In Figure 4, we presentthe H Cl spectrum, with flux normalized with respect to the single sideband continuum(2T L /T C - 1, where T L /T C is the line-to-continuum ratio) in the velocity range from υ LSR ≈ − . The decomposition of the 625 GHz line into three hyperfine components,the main F = 5 / − / F = 1 / − / F = 3 / − / Cl + (Neufeld et al. 2012) and CH (Gerin et al.2010). For comparison, we also plot in Figure 5 the H Cl + and H Cl + HFS deconvolvedabsorption profiles from Neufeld et al. (2012) and the H I absorption spectrum (Fish et al.2003). The H Cl + and H Cl + profiles are in good agreement with that of H Cl for mostvelocity intervals, as expected from precursor species of HCl. 10 –
We calculate the hydrogen chloride column densities for the velocities intervalsassociated with the different Gaussian components. First, we derive optical depths of theHCl lines ( τ = -ln[2T L /T C -1]), assuming that the foreground absorption completely coversthe continuum source and that all HCl molecules are in the ground state. We derive theH Cl column densities for each LSR velocity range using equation 3 of Neufeld et al.(2010), where the absorption optical depth for the H Cl J = 0 − Z τ dυ = A ul g u λ πg l N (H Cl) ⇒ N (H Cl) = 6 . × Z τ dυ (cid:2) cm − (cid:3) where g u = 3 and g l = 1 are the degeneracies of the upper ( J = 1) and lower ( J = 0)state. Since we use the deconvolved spectrum without hyperfine splitting, λ = 478.96 µ mis the transition wavelength, and A ul = 1.17 × − s − is the spontaneous radiative decayrate. Table 2 shows the HCl column densities and the abundances with respect to the totalhydrogen column density. We calculate the Cl/ Cl ratio in the velocity interval free fromemission line contamination (shaded interval in Figure 2), which corresponds to velocitiesfrom 38 to 40.5 km s − . The Cl/ Cl ratio obtained is equal to 2.9, in good agreementwith the solar value of 3.1 (Anders & Grevesse 1989). Thus, the HCl abundances arecalculated with respect to the total molecular and neutral hydrogen assuming a Cl/ Clof 3.1 in all velocity intervals. The H I column densities in the foreground gas are obtainedfrom Fish et al. (2003), while the molecular hydrogen is obtained from CH (Gerin et al.2010), assuming a linear correspondence between CH and H with a scaling factor of 3.5 × − (Sheffer et al. 2008). The average HCl fractional abundance with respect to the totalhydrogen is 6.1 ± × − . Using a chlorine abundance in diffuse clouds, N (Cl)/ N H ,of 1.03 × − measured by Moomey, Federman & Sheffer (2012) towards nearby (a fewhundred pc) stars, we obtain that the average fraction of gas-phase Cl in HCl is ∼ σ uncertainty in the column density is dominated by the error in thelinear baseline fitting. The total HIFI calibration uncertainties for band 1b are ≤ ∼
14 % (Godard et al. 2012). Adding inquadrature all the errors, the total uncertainty in the column densities is about 19 %.
4. Chemical model
We use the Meudon PDR code (Le Petit et al. 2006) to compute the fractionalabundances of the various chlorine bearing species, i.e. the abundance of each speciesnormalized relative to the chlorine elemental abundance, as a function of the total extinctionof the cloud, A v , tot . The Meudon PDR code is a one- or two-sided model, where themolecular cloud is modeled as a stationary plane-parallel slab of gas and dust, exposedto incident radiation field from a bright star or the Interstellar Standard Radiation Field(ISRF). As input parameters, we use the best fit model from Neufeld et al. (2012), whichcorresponds to an initial density of n H = 316 cm − , a radiation field intensity from bothsides of 10 and a primary ionization rate of 1.810 × − s − . Using the standard Galacticgas-to-dust ratio N H / A v = 1.93 × mag − cm − (Whittet 2003), A v , tot probed is ∼ left , the results indicate that the HCl + and H Cl + accountjointly at best for ∼ ∼ Cl + and the corresponding product branching ratios are not yetavailable from laboratory experiments. In their models, NW09 assumed a branching ratioof 90 % for Cl and 10 % for HCl, motivated by the low HCl abundance observed towardsthe diffuse cloud ζ Oph. We have thus investigated the effect of the branching ratio ofH Cl + DR for the physical conditions described above, based on the correlation obtainedbetween the branching ratio of triatomic dihydride ions for 3-body dissociation and thecorresponding energy release obtained by Roueff & Herbst (2011). From that correlation,Roueff & Herbst (2011) obtain a branching ratio of 56 % for the 3-body dissociation ofH Cl + . We estimate the fractional abundances corresponding to the new branching ratioand plot them as dashed lines in Figure 6- right , in comparison with those obtain withprevious assumption (90 % for Cl) in full lines, as a function of A v tot . The results show thatdifferences occur only for the minor species, i.e. HCl, HCl + and H Cl + while the reservoirsCl and Cl + are not affected. The increase in the HCl column density is about a factor of ∼
5, in closer agreement with our observational results. However, the increase for HCl + and H Cl + is much smaller, and then this result cannot explain all the discrepancies.
5. Discussion
We present the detection of H Cl and H Cl J = 0 − ∼ × cm − ), is comparable with that obtained towards the line of sight to SgrB2 (S) (Lis et al. 2010). Given the HCl + and H Cl + column densities from De Luca et al.(2012) and Neufeld et al. (2012), the HCl + /H Cl + and HCl/H Cl + column density ratios, of ∼ ∼ ±
13 –0.6 × − suggests a HCl fractional abundance with respect to the chlorine elementalabundance ( N (Cl)/ N H = 1.03 × − ) a factor of about 6 larger than those predicted bychemical models of diffuse clouds. Similar discrepancies between models and observationswere also found for other Cl-bearing gas-phase species, HCl + and H Cl + towards W31C,Sgr A (+50 km s − molecular cloud) and W49N (Lis et al. 2010; De Luca et al. 2012;Neufeld et al. 2012).Lis et al. (2010) suggested a geometrical explanation for the discrepancy between theobserved column densities and the models, where enhancement in the absorbing columndensity could result from multiple PDRs present along the line of sight. This scenario couldexplain higher column densities, but the HCl/Cl ratio will remain constant and hence thissolution is not likely to reproduce the W31C line of sight results.Neufeld et al. (2012) explored several possible explanations for the deficiencies in themodel predictions, e.g. the rate coefficient assumed for dissociative recombination (DR)of HCl + and uncertainties in the assumed dipole moment of H Cl + , none of which provedsatisfactory. We have investigated the effect of varying the branching ratio of H Cl + DR.By taking a branching ratio of 56 % for the chlorine (instead of 90 % assumed in NW09models), the predicted fractional abundance of HCl with respect to the gas-phase chlorineabundance increases by about a factor of ∼
5. However, laboratory results are neededto confirm the branching ratio and validate this argument, along with the study of otherunknowns in the chlorine chemistry, such as the DR rates, and the dipole moment of H Cl + .HIFI has been designed and built by a consortium of institutes and universitydepartments from across Europe, Canada and the United States (NASA) underthe leadership of SRON, Netherlands Institute for Space Research, Groningen, TheNetherlands, and with major contributions from Germany, France and the US. Consortiummembers are: Canada: CSA, U. Waterloo; France: CESR, LAB, LERMA, IRAM; Germany: 14 –KOSMA, MPIfR, MPS; Ireland: NUI Maynooth; Italy: ASI, IFSI-INAF, OsservatorioAstrofisico di Arcetri-INAF; Netherlands: SRON, TUD; Poland: CAMK, CBK; Spain:Observatorio Astronomico Nacional (IGN), Centro de Astrobiologia; Sweden: ChalmersUniversity of Technology - MC2, RSS & GARD, Onsala Space Observatory, SwedishNational Space Board, Stockholm University - Stockholm Observatory; Switzerland:ETH Zurich, FHNW; USA: CalTech, JPL, NHSC. Support for this work was providedby the Centre National de Recherche Spatiale (CNES), by the SCHISM project (grantANR-09-BLAN-0231-01), by NASA through an award issued by JPL/Caltech, and by theSpanish MICINN (grants AYA2009-07304 and CSD2009-00038). This research has beensupported in part by the NSF, award AST-0540882 to the CSO. Facilities:
Herschel/HIFI. 15 –
REFERENCES
Asplund, M., Grevesse, N., Sauval, A. J., & Scott, P. 2009, ARA&A, 47, 481Anders, E., & Grevesse, N. 1989, GeCoA., 53, 197Blake, G. A., Keene, J., & Phillips, T. G. 1985, ApJ, 295, 501Blake, G. A., Anicich, V. G., & Huntress, W. T., Jr. 1986, ApJ, 300, 415Beltr´an, M. T., Cesaroni, R., Neri, R., & Codella, C. 2011, A&A, 525, A151Comito, C., Schilke, P., Phillips, T. G., et al. 2005, ApJS, 156, 127Codella, C., Ceccarelli, C., Bottinelli, S., Salez, M., Viti, S., Lefloch, B., Cabrit, S., Caux,E., Faure, A., Vasta, M., & Wiesenfeld, L. 2012, ApJ, 744, 164Dalgarno, A., de Jong, T., Oppenheimer, M., & Black, J. H. 1974, ApJ, 192, L37Draine, B. T. 1978, ApJS, 36, 595De Graauw, Th., et al., 2010, A&A, 518, L6De Luca, M. et al. 2012, ApJ, 751, L37Federman, S. R., Cardelli, J. A., van Dishoeck, E. F., Lambert, D. L. & Black, J. H. 1995,ApJ, 445, 325Fish, V. L., Reid, M. L., Wilner, D. J. & Churchwell, E. 2003, ApJ, 587, 701Gerin, M. et al. 2010, A&A, 512, L16Godard, B., Falgarone E., Gerin M., Hily-Blant P., & De Luca M., 2010, A&A, 520, 20Godard, B. et al. 2012, A&A, 540, A87 16 –Greaves, J. S., & Williams, P. G. 1994, ApJ, 290, 259Jura, M. 1974, ApJ, 190, L33Le Petit, F., Nehme, C., Le Bourlot, J., & Roueff, E. 2006, ApJS, 164, 506Lis, D. C. et al. 2010, A&A, 521, L9Monje, R. R., Emprechtinger, M., Phillips, T. G., Lis, D. C., Goldsmith, P. F., Bergin, E.A., Bell, T. A., Neufeld, D. A., Sonnentrucker, P. 2011, ApJ, 734, L23Moomey, D., Federman, S. R. & Sheffer, Y. 2012, ApJ, 744, 174M¨uller, H. S. P., Thorwirth, S., Roth, D. A., & Winnewisser, G. 2001, A&A, 370, L49M¨uller, H. S. P., Schl¨oder, F., Stutzki, J., & Winnewisser, G. 2005, J. Molec. Struct., 742,215Neufeld, D. A., & Wolfire, M. G. 2009, ApJ, 706, 1584 (NW09)Neufeld, D. A., et al. 2010, A&A, 518, L108Neufeld, D. A., et al. 2012, ApJ, 748, 37Ott, S. 2010 in Astronomical Data Analysis Software and Systems XIX, ed. Y. Mizuno, K.I. Morita, & M. Ohishi, ASP Conf. Ser., 434, 139Peng, R., Yoshida, H., Chamberlin, R., Phillips, T., Lis, D. C.; Gerin, M. 2010, ApJ, 723,218Pickett, H. M., Poynter, I. R. L., Cohen, E. A., et al. 1998, J. Quant. Spec. Radiat. Transf.,60, 883Pillbratt, G. L., et al. 2010, A&A, 518, L1 17 –Roelfsema, P. R., et al. 2012, A&A 537, A17Roueff, E., & Herbst, E. 2011, J. Ph. Conf Series 300, 2028Salez, M., Frerking, M. A., & Langer, W. D. 1996, ApJ, 467, 708Schilke, P., Phillips, T. G., & Wang, N. 1995, ApJ, 441, 334Schilke, P., Phillips, T. G., & Mehringer, D. M. 1999, in The Physics and Chemistry of theInterstellar Medium, Proceedings of the 3rd Cologne-Zermatt Symposium, held inZermatt, September 22-25, 1998, ed. V. Ossenkopf, J. Stutzki, & G. Winnewisser,GCA-Verlag Herdecke, 330Sheffer, Y., Rogers, M., & Federman, S. R. 2008, ApJ, 687, 1075Sonnentrucker, P., et al., 2010, A&A, 521, L12van Dischoeck, E. F., & Black, J. H. 1986, ApJS, 62, 109van Dischoeck, E. F. 1990, ASPC, 12, 207Whittet, D. 2003, Lunar Planet. Inf. Bull., 94, 11Wright, E. L., Fazio, G. G., & Low, F. J. 1977, ApJ, 217, 724Zmuidzinas, J., Blake, G. A., Carlstrom, J., Kneene, J., & Miller, D. 1995, ApJ, 447, L125This manuscript was prepared with the AAS L A TEX macros v5.2. 18 –Table 1. LTE model parameters.
Species Source size T ex υ LSR
FWHM N (”) (K) (km s − ) (km s − ) (cm − )CH OCH
33 60 -2.5 5.5 2 × CH CN 4 150 -1 5.5 7.4 × SO ×
19 –Table 2. HCl column densities and abundances on the line of sighttowards W31C. v LSR N (H I) a N (H ) b N (H Cl) N (HCl) c N (HCl)/ N Hd km s − × cm − × cm − × cm − × cm − × − [15 ,
19] 1.3 ± ± ± ± ± ,
24] 1.6 ± ± ± ± ± ,
29] 1.6 ± ± ± ± ± ,
36] 2.1 ± ± ± ± ± ,
40] 1.7 ± ± ± ± ± ,
40] 8.4 ± ± ± ± ± a The H I column densities are obtained from Fish et al. (2003) absorption data with N (H I) = 1.84 × (T spin /100K) R τ HI d υ [cm − /km s − ], assuming a T spin ∼
100 K b The H column densities are derived from CH observations (Gerin et al. 2010), assuming N (CH) = 3.5 × − N (H ) (Sheffer et al. 2008). c Assuming a solar Cl/ Cl ratio of 3.1 d N H =2 N (H )+ N (H I)
20 –Fig. 1.— Spectra of the J = 1 − Cl ( upper panel ) and H Cl ( lower panel )towards W31C. The velocity scale in the upper axis is with respect to the frequency of themain ( F = 5 / − /
2) hyperfine component of both isotopologues. The blue lines indicatethe position of the HFS components. 21 –Fig. 2.— Upper panel, XCLASS LTE models of the dimethyl ether (red), methyl cyanide(blue) and sulfur dioxide (green) emission lines blended into the H Cl absorption line spec-trum, using the parameters given in Table 1. Additional lines (within the observationsbandwidth) used in the model are shown in small windows at the top of the plot. Thehorizontal axis is in frequency units, with the corresponding velocity scale plotted on thetop horizontal axis for consistency with Figure 1. Lower panel, for illustration purposesonly shows the resulting H Cl spectrum (black) after subtracting the contamination fromemission lines shown in red. The grey area highlights the emission line contamination-freearea within the H Cl absorption line spectrum. 22 –Fig. 3.— LTE model of CH CN (red), using CH CN transition lines J = 29 −
28 (upperpanel) at 533.123 GHz, J = 33 −
32 (middle panel) at 606.533 GHz and J = 34 −
33 (lowerpanel) at 624.878 GHz. The fit parameters are shown in Table 1. 23 –Fig. 4.— Spectra associated with each HFS components of the H Cl line: F = 3 / − / F = 2 − green ), F = 5 / − / F = 3 − blue ); F = 1 / − / F = 1 − red )(vertical offset introduced for clarity), the sum of the three components ( magenta ) and theobserved spectrum ( black ) 24 –Fig. 5.— Top panel, multiple Gaussian fit (red) to the hyperfine deconvolved spectrum ofH Cl J = 0 − Cl + (blue), H Cl + (red) from Neufeld et al. (2012) and H I (magenta) from Fish et al. (2003). 25 –Fig. 6.— Left, abundance of the Cl-bearing molecules, chlorine atom or ion (X), normalizedrelative to the gas-phase chlorine abundance, f (X) = N (X)/ N (Cl), obtained with theMeudon PDR code, Cl + (magenta), Cl (cyan), HCl (blue), HCl + (red), and H Cl + (green)together with the H fractions (black), f (X) = 2 N (H )/(2 N (H )+ N (H I)), averaged alongthe line of sight, as a function of total visual extinction A v , tot . The initial density is n H = 316 cm − , and the radiation field intensity is 10. Right, comparison of the predictedfraction of gas-phase chlorine present in HCl, HCl + , H Cl + , Cl and Cl ++