Chemistry of the High-Mass Protostellar Molecular Clump IRAS 16562-3959
Andrés E. Guzmán, Viviana V. Guzmán, Guido Garay, Leonardo Bronfman, Federico Hechenleitner
DDraft version May 25, 2018
Typeset using L A TEX preprint style in AASTeX62
CHEMISTRY OF THE HIGH-MASS PROTOSTELLAR MOLECULAR CLUMP IRAS16562 − Andr´es E. Guzm´an,
1, 2
Viviana V. Guzm´an, Guido Garay, Leonardo Bronfman, andFederico Hechenleitner Departamento de Astronom´ıa, Universidad de Chile, Camino El Observatorio 1515, Las Condes, Santiago, Chile National Astronomical Observatory of Japan, National Institutes of Natural Sciences, 2-21-1 Osawa, Mitaka, Tokyo181-8588, Japan Joint ALMA Observatory (JAO), Alonso de C´ordova 3107, Vitacura, Santiago, Chile
ABSTRACTWe present molecular line observations of the high-mass molecular clump IRAS16562 − . (cid:48)(cid:48) .
014 pc spatial resolution). This clump hoststhe actively accreting high-mass young stellar object (HMYSO) G345.4938+01.4677,associated with a hypercompact H ii region. We identify and analyze emission linesfrom 22 molecular species (encompassing 34 isomers) and classify them into twogroups, depending on their spatial distribution within the clump. One of these groupsgathers shock tracers (e.g., SiO, SO, HNCO) and species formed in dust grains likemethanol (CH OH), ethenone or ketene (H CCO), and acetaldehyde (CH CHO). Thesecond group collects species resembling more the dust continuum emission morphologyand are formed mainly in the gas-phase, like hydrocarbons (CCH, c-C H , CH CCH),cyanopolyynes (HC N and HC N) and cyanides (HCN and CH C N). Emission fromcomplex organic molecules (COMs) like CH OH, propanenitrile (CH CH CN), andmethoxymethane (CH OCH ) arise from gas in the vicinity of a hot molecular core( T (cid:38)
100 K) associated with the HMYSO. Other COMs such as propyne (CH CCH),acrylonitrile (CH CHCN), and acetaldehyde seem to better trace warm ( T (cid:46)
80 K)dense gas. In addition, deuterated ammonia (NH D) is detected mostly in the out-skirts of IRAS 16562 − − ii regions — existcoevally. Keywords:
ISM: individual objects (IRAS 16562 − INTRODUCTIONHigh-mass stars and their associated stellar clusters form inside massive ( ≥ M (cid:12) ), dense( n H > cm − ), and compact ( R ≤ . a r X i v : . [ a s t r o - ph . GA ] M a y Guzm´an et al. eventually fragments and develops at least one high-mass young stellar object (HMYSO). High-massclumps in this protostellar phase are among the chemically richest regions in the ISM. Currently,almost 200 molecules have been detected in the interstellar medium (ISM) or circumstellar shells (see also Tielens 2013), with the majority of these molecules being detected toward high-mass starforming regions such as OMC/Orion-KL or Sgr B2/N-LMH. Most high-mass clumps are characterizedby the presence of complex organic molecules (COMs), which are comparatively large ( ≥ AU) reaches temperatures ≥
100 K, releasing to the gas phase all theice mantles from dust grains, and forming what is known as a hot molecular core (HMC). Copiousamounts of UV radiation may arise from the young high-mass star, ionizing the surrounding gas andforming a small ultra- or hyper-compact (HC) H ii region. The ionizing and dissociating radiationhave the potential of greatly affecting the chemistry of the illuminated gas. In addition, the shocksinduced by the energetic outflows associated with the HMYSO introduces turbulence, carve outflowcavities which facilitates radiation to affect farther regions in the clump, and releases elements andmolecules into the gas phase through dust heating and sputtering. All these processes indicate us thatthe chemical composition of high-mass clumps depends critically on their evolutionary state, allowingus to estimate the latter by measuring molecular abundances and ratios. A good understanding ofthe chemistry is key for this purpose.Performing a systematic study of such a complex system entails discerning the different physical andchemical environments which dominates the emission of each molecule. For example, it is importantto distinguish whether the emission comes from regions close to the HMC, from outflows, or frommore quiescent and colder gas located farther from the HMYSO. Single dish observations, associatedwith angular resolutions (cid:38) (cid:48)(cid:48) , allow us to peer into the sub-clump and “core” scales (0 . . d ≈
414 pc) andCepheus A ( d ≈
725 pc). Without spatially resolved observations, in order to establish if and whatspecies trace the same parcel of gas, astronomers resort to differences in the line kinematics (e.g.,Blake et al. 1987), excitation temperature, and sometimes source sizes derived from filling factorestimations (Gibb et al. 2000). While this approach has been very useful, the advantage of directlyobserving the spatial differences between the emission of distinct species is patent. It allows usto distinguish unambiguously the regions traced by the different molecular species and family ofmolecules, facilitating further insight into their formation and chemistry.Sub-millimeter interferometers have typically provided molecular line images with beam sizes be-tween 1 and 10 (cid:48)(cid:48) and noise levels of 10–100 mJy beam − per 1 km s − channel width. Because ofthese sensitivity limitations, most of the studies have focused on clumps hosting one or more HMCand they are usually directed toward the brightest peaks of emission. This implies that the chemistryof the bulk of the clump gas is not generally probed.High spatial resolution studies of high-mass star forming regions show the expected chemical vari-ations on molecular core scales ( < .
05 pc). Among the high resolution studies focused on HMC-containing clumps — other than Orion KL and Sgr B2/N-LMH — are Beuther et al. (2006, on IRAS RAS 16562 − − − − .
014 pc) associated with themolecular emission from IRAS 16562 − ∼ . ≈ M (cid:12) , and it harbors the HMYSO G345.4938+01.4677 (G345.49+1.47hereafter) associated with a hypercompact (HC) H ii region (Guzm´an et al. 2014, hereafter PaperI). This HMYSO is actively accreting as evidenced by the presence of an ionized protostellar jet(Guzm´an et al. 2016), a rotating molecular core/disk (Beltr´an & de Wit 2016), and molecular outflows(Guzm´an et al. 2011). Paper I presents results on the continuum, sulfuretted species, and of thehydrogen recombination lines. Here, we analyze the rest of the rich spectrum and emission mapsassociated with other chemical species detected toward IRAS 16562 − − OBSERVATIONSThe interferometer data used in this work were taken using the Atacama Large Millime-ter/submillimeter Array (ALMA, Wootten & Thompson 2009). Paper I describes in more detailthe observations and its calibration. Briefly, the data were taken during 2012 using the 12m array toward the center of IRAS 16562 − ∼ (cid:48)(cid:48) . Therefore, in the rest of this work we avoid analyzing structures A complete description of the ALMA Cycle 0 capabilities is given in the Technical Handbook:http://almascience.nrao.edu/documents-and-tools/cycle-0.
Guzm´an et al. larger than this limit. In Appendix A we investigate in more detail the effects of this lack of shortbaselines by comparing the CS line data cubes with independent, single dish data. We find thatthe simple approach of adding a constant offset per channel in order to ensure positive intensitiesrecovers more than 80% of the line single dish flux and ∼
60% of the peak. We conclude that thenegatives obtained in the interferometric images are mainly caused by spatial filtering.The data covers four spectral windows (SpW) of 1.875 GHz wide each, centered at 85.4, 87.2, 97.6,and 99.3 GHz with a channel width of 488 kHz ( ≈ . − ), which due to Hanning smoothing isequivalent to an effective spectral resolution of about two times the channel size. The spectral setupchoice was motivated by the main science goal of the observations, which was studying hydrogenrecombination lines. The additional molecular line information was obtained gratuitously. For thisreason, the dataset does not cover lines from some relevant chemical species in the band like CH CNand H CO. Typical synthesized beam is approximately 2 . (cid:48)(cid:48) × . (cid:48)(cid:48) , P.A. = 97 ◦ , with a noise perchannel of 0 . . − . Channels associated with strong emission lines (like masers) areusually noisier, with a dynamic range limit of ∼ − .We re-imaged the continuum subtracted data using the task tclean of the Common AstronomySoftware Applications (CASA) using similar parameters as in Paper I. That is, we performed cleaniterations with no masks until a threshold of 2 σ ≈ . . − (e.g., SO, J, N = 3 , (cid:1) ,
1, CS, J = 2 (cid:1)
1, andthe masering CH OH, (E ) J K a K c = 5 − , (cid:1) , transitions) side lobes and possible aliasing from offbeam emission decrease the quality of the images. In these channels the cleaning iterations reacheda fixed limit number. We determine that a limit of 5000 iterations is sufficient to reach stability ofthe cleaned flux. No further improvement in the quality of the images was detected by performing alarger number of iterations. Fully reduced spectral cubes and images are publicly available (Guzm´anet al. 2018). OBSERVATIONAL RESULTSAnalyzing the observational results obtained toward IRAS 16562 − − Line Identification
Within the observed bands we detect ∼
100 spectral features, which we identify as emission linesassociated with 22 molecular species corresponding to 34 different isomers (besides the hydrogenrecombination lines). Table 1 lists all these spectral features together with the observed frequency,molecular line, and the equivalent temperature of the upper state energy. Table 1 gives the name ofeach species as commonly used in the astronomical literature and alternative names recommendedby the
International Union of Pure and Applied Chemistry . In general, we prefer to use the shortestchemical names avoiding denominations with systematic numbering of carbon atoms.
RAS 16562 − Figure 1.
Continuum subtracted spectrum taken toward G345.49+1.47. Species associated with the mostconspicuous lines are indicated.
Figure 1 shows a one-pixel spectrum observed toward G345.49+1.47 (the richest position in molec-ular lines), where we have merged in two panels the lower- and upper-sideband spectral windows.While this spectrum shows many lines, the moderate line density allows an adequate determinationand subtraction of the continuum level. Despite most of the lines appearing toward the HC H ii regionG345.49+1.47, there is no a single position in the field which displays emission in all of the specieslisted in Table 1.To identify different molecular species we used CASSIS together with the Jet Propulsion Laboratory(JPL, Pickett et al. 1998) and Cologne Database for Molecular Spectroscopy (CDMS, M¨uller et al.2005) spectroscopic databases. For each conspicuous spectral feature, we determine which speciesin the subset of detected ISM species given by CASSIS have V LSR between − −
17 km s − .This V LSR interval is within 5 . − from the ambient V LSR ≈ − . − . We revise otherlines with comparable or lower energy upper energy levels and comparable or higher Einstein spon-taneous emission coefficients in order to discern which is the most likely molecule responsible of theexamined lines. Once a candidate is assigned to a spectral feature, we compared the data with localthermodynamic equilibrium (LTE) models in order to confirm or reject the identification. Becausethe spectrum is not too populated, there is often only a single candidate molecule, and mostly two.Probable candidate species are also determined by previous detection toward other star formationregions (e.g., Blake et al. 1987; Gibb et al. 2000). To identify the line we follow the criteria defined Guzm´an et al. in Herbst & van Dishoeck (2009, § OCHO (methyl formate,sometimes also written as HCOOCH ) and NH CHO (formamide) are presented. We do not includethese in Table 1 because the low signal-to-noise ratio of these lines ( (cid:46)
Main Morphological Features of IRAS 16562 − In this section we analyze the morphology of the emission from different species detected towardIRAS 16562 − immoments . FITS files of the zero moment maps are publicly available in Guzm´anet al. (2018).We note here that the most prominent feature of IRAS 16562 − ii region and HMYSOG345.49+1.47. Figures 2 and 3 show zero moment images toward the inner 23 (cid:48)(cid:48) × (cid:48)(cid:48) of IRAS16562 − ∼ . (cid:48) × . (cid:48) D (1 ,
1) transition, which is not part of neither of the two previous groups and displays a uniquemorphology. Finally, the CS, J = 2 (cid:1) CO + , H CN (Figure 5),and SiO (Figure 4) and they should be interpreted with caution. In this work, we avoid extractingintegrated fluxes from regions larger than 10 (cid:48)(cid:48) . In any case, note that when measuring the flux arisingfrom a compact source — like a molecular core — using the filtered map may be more adequate thanusing the map without short baseline filtering. This is because in the second case the intensities someasured include emission from material more homogeneously distributed within the clump, whichwe would not consider it to be part of the core.Figure 7 shows maps of HC N and CH OH and displays morphological features which are commonto several molecular species. As explained in more detail in Section 3.3, these two molecules are rep-resentative of the two distinct groups we define to separate the species according to their morphology.The maps show emission from relatively low excitation lines, E up = 21 . OH and 4 . N. We also show position velocity diagrams (pv-diagrams) of the data cubes in the lateral andupper panels. The pv-diagrams display, for each position, the maximum intensity measured cutting
RAS 16562 − D e c li n a t i o n ( J ) Right Ascension (J2000) SO SO HNCO3mmC H CN C S CH OCH HC NOCS CH OH (v t =1) CH OH (v t =0)H CN Figure 2.
Zero moment maps of the emission from representative transitions from the central 23 (cid:48)(cid:48) × (cid:48)(cid:48) regionof IRAS 16562 − − .
03 Jy beam − km s − , corresponding to approximately − σ . The dashed contours in the upper leftpanel correspond to those presented in Paper I. through the other dimension of the cube. We find that this way of displaying was more effective inseparating and identifying structures than the integrated intensity of the cubes collapsed in R.A. ordeclination. The main morphological features we identify are: • A central compact core (CC core).
The CC core is the most conspicuous feature in many ofthe transitions detected toward IRAS 16562 − ii region/HMYSO G345.49+1.47, marked as Source 10 (R.A. = 16 h m . s − ◦ (cid:48) . (cid:48)(cid:48)
6) in Figure 2, but displaced from it — as noted in Paper I — between 0 . (cid:48)(cid:48) . (cid:48)(cid:48) CN, J = 12 (cid:1)
11 (methyl cyanide) observations of IRAS 16562 − Guzm´an et al. D e c li n a t i o n ( J ) Right Ascension (J2000)
Peak0 H CCO CH C H CH C N HCS+H CO+ CCH NH D HC NSiO c-C H CH CHO HN C CH CHCN
Figure 3.
Same as Figure 2. Starting from the top left panel to the right and from top to bottom:the central source component becomes less and less prominent. Some transitions do not display a centraldominating component at all, like the ones in the two bottom rows. one associated with the HMYSO/HC H ii region (G345.49+1.47) and another associated witha small ( ∼ . (cid:48)(cid:48) . (cid:48)(cid:48) RAS 16562 − Peak outside 6''radius D e c li n a t i o n ( J ) Right Ascension (J2000)
SO SiO C SHNCOOCSHCS+ H CCOCH OH (2 -1 )CH CHO
D. ridge NW cloudNEC wall Nred -1 km s -1 Figure 4.
Zero moment maps of representative transitions of molecular species related with the Shockgroup (see Section 3.3). The dashed contour level in the panels is − .
03 Jy beam − km s − , correspondingto approximately − σ . The two red crosses mark the position of Sources 8 and 10. To ease comparison, weshow the NW-cloud, the N-red cloud, the NEC-wall, and the position of the Diffuse Ridge together with theDR(a), (b), and (c) positions (black crosses) defined in Section 3.2 and Figure 7. these cores as the CC core because it is centered at V LSR = − . − , which is closerto the central velocity of the hydrogen recombination lines arising from the HC H ii region(Paper I) associated with G345.49+1.47. The CH OH pv-diagrams show that the second core(marked as ‘2nd core’ in Figure 7) is located ≈ . (cid:48)(cid:48) to the northwest (2 (cid:48)(cid:48) from Source 10 in the0 Guzm´an et al. D e c li n a t i o n ( J ) Right Ascension (J2000) CO + HC O + HC N H CN CH CHCN c-C H CH CCH HN C CCH CH C NHC N -1 km s -1 Peak outside 6''radius
Figure 5.
Same as Figure 4, but with species in the Continuum group (see Section 3.3). The top left panelshows the 3 mm continuum. Red crosses indicate the continuum sources 3, 7, 8, 10, and 16.
RAS 16562 − -1 km s -1 Peak outside 6''radius
Right Ascension (J2000) D e c li n a t i o n ( J ) NH D Continuum arc
Figure 6.
Zero moment map of the NH D (1 ,
1) transition. This species shows no evident similarities witheither the Shock or Continuum group. We indicate the structure referred to as the “Continuum Arc” inSection 3.5.
P.A. = − ◦ ) and centered at V LSR = − . − , that is, redshifted respect to G345.49+1.47.Figure 7 shows these two cores in the R.A.-velocity diagram of CH OH. Some emission fromthe secondary redshifted core is detected in HC N, but much fainter than the proper CC corecentered at ≈ − . − . This secondary redshifted core is likewise evident in emissionfrom other species such as H CO + , HC N, and HN C. • Source 8 core (C8).
Many molecules show strong and extended emission near the source markedwith number 8 in Figure 2. We note, however, that the molecular peak is not coincidentwith the continuum peak. Usually, the molecular emission is strongest between 0 . (cid:48)(cid:48) . (cid:48)(cid:48) − ◦ direction from Source 8. This is the case of COMs like HC N, CH C N,CH CCH, and CH CHCN, where the emission usually spreads toward the NE direction formingan elongated cloud. A methanol maser seems to be associated with the outer rim of thismolecular envelope. • A Diffuse Ridge.
It is shown in Figure 7 as the elongated methanol emission feature crossingthe south east part of IRAS 16562 − OH,CH CHO, and in most other tracers shown in Figure 4. The Diffuse Ridge extends for ≈ (cid:48)(cid:48) in the P.A. = 65 ◦ direction from the position ∆R.A. = − (cid:48)(cid:48) , ∆Dec. = − (cid:48)(cid:48) respect to2 Guzm´an et al.
18 17161514 13 12 11109 8 7 654 32 1 18 16141715 1213 11109 658 47 32 1 -4-10-16-4-10-16 -40.054-40.056-40.058-40.06-40.062-40.064-40.066-40.068 -18-14-10 -6 -2 -2-6-10-14-18 V LSR (km s -1 ) -40.054-40.056-40.058-40.06-40.062-40.064-40.066-40.068 -18-14-10 -6 -2-2 -2-6-10-14 V LSR (km s -1 ) Di ff use ridge Di ff use ridge D e c li n a t i o n ( J ) Right Ascension (J2000) Right Ascension (J2000)
NEC-w(a)
NEC-w(b)
NW cloud
NW(a) °° NEC wall N-red (a) (b) (c)
N-red cloud
NW cloud
CC core CH OH HC N Figure 7.
Panels a) and b) show zero moment maps of the CH OH, (E ) 2 , (cid:1) , ( E up = 21 . N, 1 (cid:1) E up = 4 . OH, (E ) 5 − , (cid:1) , maser spots. Blackcrosses mark the DR(a), (b), and (c) positions in the Diffuse Ridge, the NEC-wall (a) and (b) positions, andthe NW-cloud (a) position (Section 4.2). Black-dashed ellipses mark the CC core, the N-red cloud, and theNW cloud. To the left and on top of panel a) we show R.A.-velocity and Declination-velocity pv-diagrams,respectively. In the top R.A.-velocity pv-diagram we mark the N-red cloud with a dashed circle, and twoarrows indicate the sources which compose the blended CC core emission in the CH OH zero moment map.PV-diagrams of the HC N emission are shown to the right and on top of panel b).
G345.49+1.47. Within the Diffuse Ridge there are two distinct emission cores. One is markedwith a ’(b)’ in Figure 7 and is centered at ∆R.A. = +10 (cid:48)(cid:48) , ∆dec. = − (cid:48)(cid:48) from Source 10, with adiameter of ≈ (cid:48)(cid:48) . The other, marked with a ’(c)’, is located at ∆R.A. = − (cid:48)(cid:48) , ∆dec. = − . (cid:48)(cid:48) ◦ with diameters of ≈ (cid:48)(cid:48) × (cid:48)(cid:48) .Position ’(a)’, on the other hand, targets more diffuse gas forming the body of this ridge orfilament. Hereafter, we refer to these three positions as DR(a), DR(b), and DR(c). DR(b)is conspicuous in CH OH and likewise in CH CHO, SiO, SO, HCS + , OCS, CS, HC N, andHNCO. It is less noticeable in HN C, H CN, and C S, but still present. DR(c) is as welldistinguished in CH CHO, SiO, SO, OCS, CS, HNCO, and H CN. Emission from DR(c) isalso detected in H CO + and HC N. In contrast with DR(b), we do not detect HN C norHCS + emission associated with DR(c). • A northeast outflow cavity wall (NEC-wall).
This molecular feature located ≈ (cid:48)(cid:48) south ofSource 16 matches an illuminated section of the outflow cavity wall seen in NIR. The emission isconspicuous in CH OH, CH C N, CCH, HC N, HC N, H CN, HN C and H CO + transitions.Other molecules display emission farther from the cavity wall position, but still associated withit, like c-C H , C S, CH CHO, OCS, and maybe SO. The large scale emission from SiO andHNCO which surrounds the NEC-wall area is not clearly associated with it.
RAS 16562 − • A northwest cloud (NW cloud).
This emission cloud has approximately 5 (cid:48)(cid:48) × (cid:48)(cid:48) of size and it iscentered around the position located approximately 25 (cid:48)(cid:48) in the P.A. = − ◦ direction respectto the CC core. This feature is displayed by most of the molecules shown in Figure 4, includingCH OH, SiO, CH CHO, HNCO, and CS. There is also strong emission associated with HC N.Less prominent emission arises from H CN, HN C, H CO + , CCH, H CCO, HC N, OCS, andSO as well. • A northern redshifted cloud (N-red cloud).
This emission feature of ≈ . (cid:48)(cid:48) radius is speciallyconspicuous in transitions shown in Figure 4. It is one of the emission cores with the largestdifferences in radial velocity respect to the clump. In methanol, its radial velocity is centeredat − . − , that is, redshifted by 4 . − respect to the V LSR of IRAS 16562 − OH) SiO,HNCO, OCS, CS, and SO. Less prominent emission is detected from HC N, CH CHO, andH CCO. In each case, the emission is consistently redshifted respect to the systemic radialvelocity of the clump. The N-red cloud is located ∼ (cid:48)(cid:48) north of the CC core and is not to beconfused with the ‘2nd redshifted core’ we mentioned previously.3.3. Zero Moment Cross Correlations and Grouping
In order to evaluate quantitatively how similar are the zero moment maps of different species, wecalculate the cross correlation between each pair of maps according to ρ = (cid:80) i,j I ,ij I ,ij w ij (cid:16)(cid:80) i,j I ,ij w ij (cid:80) i,j I ,ij w ij (cid:17) / , (1)where the sums are taken over each pixel position, I ,ij and I ,ij are the values measured for eachimage at pixel i, j , and w ij = 0 , ρ = ρ and | ρ | ≤
1. A value of ρ = 1 implies that I = αI , with α a positiveconstant. Therefore, the more similar the zero moment spatial distributions are, the closer their crosscorrelation is to 1.According to the definition of Equation (1), brighter sections of the image weight more into thecalculation of ρ . In order to avoid the correlation being dominated entirely by the central source, wemask the pixels (that is, we set w ij = 0) inside an inner circle of 6 (cid:48)(cid:48) radius centered at G345.49+1.47.We leave outside this specific analysis some molecules which are only detected toward the center ofthe field, e.g., CH CH CN, CH OH, and SO .Figure 8 displays the value of the cross-correlations between the zero moment maps of differentmolecules. We include in this analysis the continuum image at 3 mm presented in Paper I. In orderto gather together molecular emission based on their spatial distributions, we will use the ρ valueas a measure of morphological similarity. Cross correlations have been used to this end (althoughcomparing only with the continuum) by Lu et al. (2017). For each molecule, we determine which oneis the other species (among the set of molecules in this study) with the highest cross correlation. Wecall this species the maximum correlation partner (MCP) of the considered molecule. We stress thatthis relation is not necessarily symmetric: for example, while OCS and CH OH are MCPs of eachother, OCS is the MCP of SO but not vice-versa. The cross correlation values of MCPs are markedin boldface in Figure 8.4
Guzm´an et al. H N C O O C S H CC O C H O H S O C H C H O S i O C S H C S + c - C H H C O + C H CC H mm C H C N C H C H C N CC H H N C N H D H C N H C N H C N H N C O O C S H CC O C H O H S O C H C H O S i O C S H C S + c - C H H C O + C H C H mm C H C N C H C H C N CC HH N C N H D H C N H C N H C N H N C O , → , O C S → H CC O , → , C H O H , → , S O → C H C H O , → , ( -- ) S i O → C S → H C S + → CC H , / , → , / , H N C → H C N → H C N → H C N → S h o c k g r o u p C o n t i nuu m g r o u p N H D ( , ) E x c l u d e d c - C H , → , H C O + → C H CC H → mm C H C N → C H C H C N , → , Figure 8.
Cross correlations between zero moment maps of molecules with significant extended emission,and with the continuum (marked with 3 mm). For each molecule, the maximum correlation is marked inboldface. Each molecule in any of the two enclosed frames has their maximum correlation partner in thesame group they pertain. The top left and bottom right framed groups correspond to the so called Shockand Continuum groups, respectively. Below the cross correlation matrices we list explicitly the membersand the specific transition used. For completeness we include NH D despite this molecule not being part ofeither group.
RAS 16562 − D, which has the lowest cross correlations withits MCP (0.39). Because the zero moment map of this molecule (Figure 6) is so different comparedto the rest, we do not include it in the grouping and we analyze it independently. We estimate theuncertainty of ρ by adding simulated noise to each zero moment map and measure the dispersion ofthe values of the cross correlations thus obtained. In all cases, the uncertainty due to random noiseis < .
01, making no discernible effect in the classification.We denominate “Shock group” the one in which traditionally shock activity tracers such as SiO,HNCO and SO gather. We refer to the second group as the “Continuum group” because it gathersspecies better correlated with the 3 mm continuum. As Figure 8 shows, the MCP of the continuumimage is the zero moment of the H CO + , J = 1 (cid:1) − Kinematics
We analyze the kinematics of IRAS 16562 − OCH , CH CH CN, and SO :first moment maps of the first two are rather uninformative because they display no velocity gradientsand seem to be well characterized by a single V LSR . The SO map, on the other hand, does show avelocity gradient characteristic of rotation which was analyzed in detail in Paper I. We also leave outspecies with only faint extended emission like HC N, CH C N, CH CHCN and H CCO also writtenas CH CO. The first moment maps of these molecules display similar characteristics as those shownin Figure 9, but with lower signal-to-noise ratios.We stress that the kinematic analysis is somewhat hindered by the modest spectral resolution ofthe data. Figure 9 shows that the velocities range between − −
15 km s − for most molecules,with the ambient cloud velocity around −
12 km s − . This is consistent with previous single dishstudies on IRAS 16562 − ∼ −
10 km s − , that is, redshiftedrespect to the ambient cloud) toward the north west section of IRAS 16562 − ∼ −
14 km s − ) toward the south east. This trend is reminiscent of the general orientation ofthe jet and CO outflow detected toward G345.49+1.47 (Guzm´an et al. 2011). This general kinematic6 Guzm´an et al. -24 -22 -19 -17 -14 -12 -9.6 -7.2 -4.8 -2.4 0
SiO SO OCS CH OH (2 -1 ) HNCO CH CHO HCS+ C S HC N HC N HN C CH CCH
CCH c-C H H CO+ NH D Right Ascension (J2000) D e c li n a t i o n ( J ) km s -1 + + +++ + ++++++++++ Figure 9.
Large scale images of the first moment from molecular transitions with extended emission. Tocalculate the first moment, we masked emission below 5 mJy beam − ( ∼ σ ). Black crosses indicate theposition of G345.49+1.47. Black arrows in the SiO panel indicate some of the so-called “fingers.” trend is well illustrated by the H CO + first moment map, but it is also evident in HC N, HC N,HN C, CH CCH, c-C H , and C S. The trend is less evident — but still tantalizing — in SO,CH OH, HC N, CCH, and even NH D . The H CO + map likewise displays conspicuous blueshifted RAS 16562 − (cid:48)(cid:48) south of G345.49+1.47, which is due to gas with V LSR ≈ − . − .This emission is apparent in HC O + and SO, but it is not clearly seen in any other molecule.By far, the transition showing the largest velocity variations across the field is SiO, J = 2 (cid:1) . − , whereas for the restof the molecules the velocity span is (cid:46)
10 km s − . Interestingly enough, the only exception to thistrend is SO, whose velocity span is ∼ . − . This kinematic feature of SO is consistent withthe similar morphology observed between its zero moment map and that of SiO.Another kinematic feature unique to SiO are the conspicuous filaments or “fingers” populatingthe north east region of IRAS 16562 − ∼
10 km s − across a distance of 30 (cid:48)(cid:48) (0.25 pc at 1.7 kpc). Four discernible “fingers” arelocated toward the north east region of the clump. Additional fingers are apparent toward the southeast and western parts of IRAS 16562 − − — and isotopologues — there is no other molecule associatedwith the velocity gradient characteristic of the rotating core around G345.49+1.47 detected in Paper I.3.5. Morphology of the Emission by Molecule
In the following paragraphs we describe the main characteristics of the emission of representativetransitions of each species. This section expands the description of Section 3.2 focusing on thespecific morphology per molecule. Paper I already analyzed in detail the morphology of most of thesulfuretted molecules emission near the central HMYSO, so we refer to their analysis for these species(SO, SO , OCS, CS and their isotopologues). To organize the upcoming discussion we gather themolecules according to their composition and number of atoms.3.5.1. Simple molecules
With simple molecules we refer to molecules with 5 atoms or less.
Nitrogenated.
We include here the HCN isomers, HC N (cyanoacetylene) and its isotopologues, andNH D. Just for the sake of exposition, we choose to include HNCO. The regions where HC N (andthe C isotopologues) emission is most intense are the CC core, the NEC-wall, and C8. The C8emission peaks ∼ . (cid:48)(cid:48) N emitting region is atriangular-like structure between sources 10, 11, and an emission peak located 1 . (cid:48)(cid:48) C isotopologues displays the same morphological features. On aclump scale, emission from HC N is most similar, as evaluated from the cross correlations, to HN Cand H CN. Figure 5 suggests in addition a good match between the HC N zero moment map andthose of H CO + and CH CCH, which is confirmed by the cross-correlations.Other nitrogenated simple molecules are the isotopologues H CN and HC N. Their zero momentmaps are very similar, with a cross correlation of 0.9, which is higher than any correlation betweendifferent species ( ≤ . .
86. Kinematically, the molecules are similar as well. The linewidths range between the 2and 6 km s − , with most of the emission having linewidths between 3 and 4 km s − .8 Guzm´an et al.
An evident difference between the line profiles of H CN and HC N is the hyperfine splitting ofthe J = 1 (cid:1) CN transition produced by the nuclear quadrupole of the N nucleus. We fitGaussians to the three hyperfine components in order to explore further the observed splitting. Theblue and redshifted components are located at velocities of − . . − respect to thecentral component, which is the brightest. Theoretically, the line intensities should be in the 1 : 5 : 3ratio in optically thin conditions. We performed the fitting in every pixel where the peak intensityexceeds 5 σ . The observed intensity ratios of the blue and redshifted components respect to the centralone are 1 ± . ± σ variations. We conclude that our results are consistent — withinthe uncertainties — with the local thermodynamic equilibrium and optically thin predictions.The morphology of the HN C emission resembles that of H CN: they both display strong emissionassociated with the C8 and the NEC-wall. However, there are two important differences, the mostrelevant being the complete absence of HN C emission from the CC core while H CN is brightthere. The other noticeable difference is the HN C emission associated with the arcuate continuumstructure joining sources 7, 3, 2, and 8. Because this arc of emission appears conspicuously in a fewother transitions, we refer to it hereafter as the Continuum Arc.HNCO (isocyanic acid), on the other hand, is a molecule with a strong CC core component. Itszero moment maps are well correlated with those of OCS and CH OH, showing strong emissionassociated with the Diffuse Ridge, the NW cloud, and the N-red cloud. However, there is not muchHNCO emission associated with the C8 or the Source 8 itself, and no emission associated with theNEC-wall. Isocyanic acid is the only nitrogenated species in the Shock group.Finally, NH D is a special case. By and large, the zero moment of NH D (Figure 6) displays littlecorrelation with features conspicuous in other molecules, and we analyze it separately from the Shockor Continuum groups. The peak of the NH D emission is located 36 (cid:48)(cid:48) from G345.49+1.47, in theP.A. = 65 ◦ direction. This NH D core is apparently part of a filament which extends for ≈ (cid:48)(cid:48) .Molecules which have emission related with the location of this core are HN C, CCH, and possiblyHCS + . A secondary peak of NH D emission is associated with a core located 38 (cid:48)(cid:48) south east ofG345.49+1.47, in the P.A. = 122 ◦ direction. This core is not related with any discernible structurein any other molecule, but, as seen below, it seems related with a NIR-dark globule. Less intenseemission is located associated with the NEC-wall and with Source 16, and there is also emissionconnecting both positions. This morphology resembles the continuum (Paper I), which shows thatSource 16 is embedded in an envelope extending to the south until approximately the region weidentify in this work with the NEC-wall. Other features clearly associated with NH D emission areSources 7 and 2 and some diffuse emission apparently tracing the Continuum Arc. As we will seebelow, it does show some faint emission associated with the CC core, barely noticeable in Figure 3.
Sulfuretted.
A rather complete analysis of the morphology of the emission from the sulfurettedmolecules, specially in the central region of IRAS 16562 − , and isotopologues) are well associated with G345.49+1.47, in contrastto other carbon-sulfur species such as OCS and CS. All sulfuretted species whose emission extends onscales comparable with the clump size — that is, all of them except SO — are in the Shock group.From Figures 4 and 8 it appears that SO and OCS correlate more with CH OH and (ignoring the
RAS 16562 − S and HCS + . The zero moment maps ofthese last two molecules are similar between each other, and they display some features observablein some molecules of the Continuum group (like CCH, see below).The only sulfur bearing molecule which was not included in Paper I is HCS + . Figures 3 and4 show the zero moment of the HCS + , J = 2 (cid:1) ∼ . (cid:48)(cid:48) + emission is most similar to that of C S. This similarity is reflected in the velocity distributionof both lines.Two of the most remarkably similar zero moment maps are those of SO and SiO (see Section 3.3),whose similarity is also observed in the first moment maps (see Section 3.4). Several common featuresare recognizable in these two maps (Figure 4), but one equally remarkable difference is the strongemission from SO associated with the CC core, which is absent in SiO. That is, the SO emissionin IRAS 16562 − S. We attribute this to the high optical depth of CS compared to that of the C S. Small carbon chains.
In this category we include CCH and c-C H , the latter being a cyclic molecule.Neither of these two molecules have emission associated with the CC core, with c-C H displayingan strong absorption feature toward the location of G345.49+1.47, centered at − . ± . − with a FWHM of 4 . ± . − . One important feature of the CCH, N = 1 (cid:1) isthat it splits in six hyperfine components (discernible in Figure 1). The relative observed strength ofthese components (whose temperature dependence is negligible) is in good agreement with the ratioexpected for optically thin emission. CCH emission is strongest in the Continuum Arc, specifically,just below Source 3. Its is likewise strong between the CC core and N-red cloud, in the NEC-wall,and in a cloud located ∼ (cid:48)(cid:48) to the south east of G345.49+1.47 (P.A. = 135 ◦ ). This last emissionfeature is also conspicuous in the zero moment map of c-C H and it has no obvious counterpart inany other molecule.Emission from c-C H seems to be less extended compared to that of CCH. It is strongest in theC8 region, extending somewhat toward the Continuum Arc. As pointed out before, there is clearemission associated with the south east cloud mentioned above. Contrary to CCH, there is no strongc-C H emission arising from the NEC-wall. Less prominent c-C H emission is also detected froma cloud located at ∼ (cid:48)(cid:48) to the east of G345.49+1.47 (P.A. = 80 ◦ ). We note that there is evidentCCH and C S emission from this region as well.Additional similarities between CCH and C S include a filament of emission extending for ∼ (cid:48)(cid:48) roughly in the E-W direction, located 30 (cid:48)(cid:48) north east of G345.49+1.47 (P.A. = 45 ◦ ). We note thatthis filament is also visible in HCS + . Oxygenated.
In this category we analyze simple molecules composed of carbon and oxygen, that is,the isotopologues H CO + and HC O + and H CCO (ethenone, also called ketene). The HC O + N represents the pure rotational angular momentum (Gottlieb et al. 1983). Guzm´an et al. map is very similar to that of H CO + (correlation coefficient of 0.79), with the HC O + line beingmore optically thin ( ≈ CO + . Figure 3 shows that the emission from H CCO and H CO + is associated with the CCcore and C8. The H CCO zero moment map displays a more compact distribution around these twolocations compared to that of H CO + or HC O + .On a large scale, both species are different: H CCO and H CO + were classified in the Shock andContinuum groups, respectively. Ethenone displays a good resemblance to the OCS (its MCP) andCH CHO maps, and the cross correlations given in Section 3.3 are practically equal (0.60 and 0.59,respectively). We can identify clearly in the H CCO map, emission coincident with the Diffuse Ridge,the NW cloud, and the N-red cloud. In addition, the H CCO map shows diffuse emission located30 (cid:48)(cid:48) north of G345.49+1.47, which is seen clearly in OCS, CH CHO, HNCO, and CH OH.The H CO + , on the other hand, is the MCP of the 3 mm continuum map. That is, the H CO + zero moment map (Figure 5) is the one which better correlates with the continuum away of thecentral source, dominated by thermal dust (Paper I). H CO + is one of the few molecules withcounterpart emission associated with Source 13 (Figure 3). The zero moment map also shows asource located ∼ (cid:48)(cid:48) south of G345.49+1.47 whose emission is significantly blueshifted respect to theambient material (see Section 3.4). Other regions associated with strong H CO + emission are theNEC-wall and the C8. The C8 emission extends to the Continuum Arc. In addition, there is a southeast diffuse feature which correlates roughly with the position of continuum Sources 12, 15, and 17.Finally, we note that the H CO + emission, as it is the case for HC N and CH CCH, is morecompactly distributed. There is little H CO + emission more than 20 (cid:48)(cid:48) away from G345.49+1.47,which is in stark contrast compared to the molecules of the Shock group.3.5.2. Complex Organic Molecules (COMs)
COMs detected in this work are molecules consisting of carbon and hydrogen atoms plus, exceptCH CCH, either one oxygen or one nitrogen atom. The only molecule detected with an O and aN atom is HNCO, and we do not detect any molecule with more than one N or O atom. In thefollowing, we call a molecule saturated if all its carbon-carbon bonds are single (C C). Conversely,unsaturated COMs have double or triple carbon-carbon bonds (C C or C C). All simple moleculeswith carbon-carbon bonds detected in this work (CCH, c-C H , HC N, and H CCO) are unsaturated,but this is likely due in part to a selection effect: saturated hydrocarbons usually have more atoms.
Propyne (CH CCH). This COM was previously observed toward IRAS 16562 − (cid:48)(cid:48) north-east of G345.49+1.47. Propyne is also brightly associated with the C8:a secondary emission peak is located less than 1 (cid:48)(cid:48) west of Source 8. On a larger scale (Figure 5),CH CCH emission displays a rather smooth distribution, with a very good correlation with H CO + ,although less bright. These two molecules are MCPs of each other. Practically all the bright featuresdistinguished in the H CO + zero moment map have a CH CCH counterpart. Propyne shows a goodcorrelation with HC N as well, excluding the central region. Due to the difficulty in separating thedifferent CH CCH lines due to blending, the zero moment maps shown include the sum of all the5 , K (cid:1) , K transitions. Nitrogenated (CH CH CN, CH C N, CH CHCN, and HC N). Only one of these COMs, propaneni-trile (CH CH CN, hereafter C H CN), is unequivocally linked with the CC core. Figure 2 shows that
RAS 16562 − ∼ . (cid:48)(cid:48) or HNCO. Faint emission of C H CN (not evident in the zero moment map image) is detectedalso toward the C8 (see Section 6). We do not detect this COM toward any other location in theclump.As shown in Figure 3, the other three COMs have less correlation with the CC core, and none of theirpeaks actually correspond with the CC core position. Among these COMs, CH C N (cyanopropyne)is the only one which shows some CC core counterpart. The emission from CH C N has two peaks,associated with the NEC-wall and C8, with practically the same intensity. The C8 CH C N emissionpeaks approximately 1 (cid:48)(cid:48) east of Source 8, and extends somewhat to the Continuum Arc. The NEC-wall emission from this COM apparently joins with the CC core emission and with emission detectedsouth of Source 16. The overall appearance of this structure is similar to what is observed in othermolecules, for example, HCS + (same Figure 3). There is emission correlated with the position ofSource 11, which is a feature seen in a few other molecules (e.g., CCH, CH CCH, and HN C). Figure5 shows the CH C N zero moment map on a larger scale. There is not much more emission comparedwith what is shown in Figure 3. However, we note a small cloud located about 18 (cid:48)(cid:48) south west fromG345.49+1.47 in the P.A. = − ◦ direction. Emission from this location is clearly detectable inCCH, CH CCH, H CN, HN C, H CO + , HC N, HC N, HCS + , and CH CHCN. It is located nearthe south west end of the Diffuse Ridge, thus, we can better discern it in molecules without DiffuseRidge emission like CCH and CH CCH.Emission from acrylonitrile (CH CHCN, hereafter C H CN) is better displayed in Figure 3, becausemost of the emission arises from regions not farther than 15 (cid:48)(cid:48) from G345.49+1.47. The peak ofC H CN emission is clearly associated with C8. Acrylonitrile is one of the faintest molecule we claimdetection in this work. With the exception of the peak emission, the rest of the features shown inFigure 5 are apparently real mostly because their location is consistent with emission seen in othermolecules. There is also emission somewhat consistent with the CC core and with the south westcloud described at the end of the previous paragraph.Finally, Figures 3 and 5 show the zero moment images of the cyanopolyyne HC N (cyanodiacety-lene). This molecule does not show strong emission associated with the CC core. Weak emissionlocated ≈ (cid:48)(cid:48) north of G345.49+1.47 joins to the east with emission arising from the NEC-wall, asseen in many other molecules. As it is likewise the case for CH CCH, CH C N, CCH, and HN C,there is HC N emission associated with the location of Source 11. The HC N peak is clearly locatedin C8. Diffuse emission extends from the C8 to the north following somewhat the Continuum Arc.Relatively intense, diffuse emission, is also associated with Source 7.
Oxygenated (CH OH, CH CHO, and CH OCH ). By far, the CH OH (methanol) transitions displaythe brightest and richest spatial structures of all COMs. Among the eleven detected methanol linesthere is one rotational line from a vibrationally excited state and a class I maser transition ((E ) J K a K c = 5 − , (cid:1) , ). Methanol emission (Figures 2 and 4) is in every line dominated by strongemission detected toward the CC core, except for the maser transition which is dominated by threebright maser “spots.”Figure 10 shows in the left panel the quotient between the zero moment maps of the (E ) J K a K c = 5 − , (cid:1) , and (E ) 2 , (cid:1) , CH OH transitions. The three maser spots marked a , b , and c in the right panel of this figure (which shows the zero moment of the maser transition inlogarithmic color stretch) are associated with line fluxes more than 100 times larger than those of the2 Guzm´an et al.
16 109 8 910 816 abc D e c li n a t i o n ( J ) Right Ascension (J2000) (Jy km s -1 ) ° ° ° Figure 10.
Left panel:
Ratio between the zero moment of the CH OH, (E ) J K a K c = 5 − , (cid:1) , and(E ) 2 , (cid:1) , lines. Right panel:
Zero zero moment of the CH OH, (E ) J K a K c = 5 − , (cid:1) , emission.The three masers (yellow circles) correspond to the three strongest sources (whose coordinates are given inTable 3), conspicuous even in the logarithmic color stretch. Red crosses and numbers show continuum 3mm sources nearby the masers. typically thermal (E ) 2 , (cid:1) , line. Table 3 indicates the parameters of the maser emission. Thestrongest maser is b followed closely by a . Maser a is located 4 (cid:48)(cid:48) in the P.A. = − ◦ direction fromSource 8 and it is associated with bright methanol emission in the rest of the lines. Other moleculeswhich show clearly emission consistent with the location of Maser a are CH CHO and H CCO, andperhaps less clearly HCS + and OCS. Methanol lines associated with low upper energy levels ( < a location than toward Source 8, whereasthe opposite is true for high energy transitions ( >
50 K). Source 8 is also bright in H CCO and OCSbut not in CH CHO. Maser b is located 1 . (cid:48)(cid:48) ◦ ) from G345.49+1.47. Dueto the proximity of the CC core, this region is associated with diffuse emission in several molecules.However, in contrast with Maser a , there is no distinguishable feature in either methanol or any othermolecule coincident with the position of this maser. Maser c , located 1 . (cid:48)(cid:48) OH, (E ) 2 , (cid:1) , to define some of the most noticeable of IRAS 16562 − E up = 21 . E up = 74 . . OH supports the view that this is a HMC. The
RAS 16562 −
18 16 14 13 1110 8 7 4 3 2 18 16 14 13 1110 8 7 4 3 2
Right Ascension (J2000) D e c li n a t i o n ( J ) -1 Peak CH OH 6 -2,5 → -1,7 CH OH (v t =1) 6 → Figure 11.
Left and right panels show, respectively, the zero moment maps of the CH OH, (E ) 6 − , (cid:1) − , and 6 , (cid:1) , ( v t = 1) transitions. Red crosses and numbers show the position of continuum 3mm sources as defined inPaper I. The upper energy levels of these two transitions are 74 . . − , (cid:1) − , ( E up = 74 . ≈ (cid:48)(cid:48) at the R.A. of Source 13.We note that while the low energy methanol (Figure 7) shows extended diffuse envelopes blendingin with the continuum sources, the left panel in Figure 11 shows a much less ambiguous correlationwith Sources 7, 3, 8 and 18. Structures which are not apparent in the high energy CH OH transitionsare the Diffuse Ridge, the NW cloud, and the N-red cloud.Acetaldehyde (CH CHO) is another oxygenated COM we detect toward IRAS 16562 − CHO is not associated with the CC core. In fact, there is no clear emissionassociated unambiguously with any continuum source. On a larger scale (Figure 4) CH CHO emissionshows features characteristic of the Shock group: its MCP is CH OH, with a correlation coefficientbetween them of 0 .
72 (we emphasize that Section 3.3 uses a low energy CH OH transition for theanalysis). Acetaldehyde is bright toward the Diffuse Ridge, the N-red cloud, the NW cloud, andMaser a position. In general, there is CH CHO emission toward bright methanol regions, e.g.: thediffuse CH CHO emission south of the CC core has a similar morphology as CH OH; a small cloudlocated 1 . (cid:48)(cid:48) CHO; and the cloudlocated 30 (cid:48)(cid:48) north of G345.49+1.47 already mentioned in the H CCO description.4
Guzm´an et al.
Finally, CH OCH (methoxymethane) is detected only toward two locations: the CC core andSource 8. The CC core CH OCH emission is evident in Figure 2, but that of Source 8 is very faintand more evident in the data cube of the 4 , (cid:1) , transition instead of the zero moment map. COLUMN DENSITIES AND EXCITATION TEMPERATURESIn this section we model the data and results presented in the previous sections and determinephysical parameters like column densities and temperatures, mainly from LTE models. Section 4.1focuses on the CH CCH emission and LTE modeling. Section 4.2 makes more detailed models of theemission of the species detected toward several sources in IRAS 16562 − OCHO and NH CHO (mentioned in Section 3.1), other moleculescommonly associated with hot-cores (e.g., Gibb et al. 2000) which were observed but not detectedare HCOOH (formic acid) and CH CH OH (ethanol). These non-detection allow us to estimateupper limits on the column densities of the respective species (Section 4.2). On the other hand,neither H S (hydrogen sulfide), CH CN (methyl cyanide), nor their isotopologues were covered byour observations. The spectral setup does not efficiently cover the H CO (formaldehyde) or theHDCO lines either because it only samples transitions predicted to be faint (high E up or very lowEinstein A -coefficients). 4.1. Propyne Temperature and Column Density
Propyne (CH CCH) is a symmetric top molecule, with its dipole moment aligned with the symmetryaxis of the molecule. This implies that radiative rotational transitions do not change the projection K of the angular momentum J onto the symmetry axis (Townes & Schawlow 1975, note that K ≤ J ). Levels with different J and the same K are sometimes refer to as K -ladders. Thus, radiativetransitions only connect J, K (cid:1) ( J − , K states and the relative population between different K -ladders is determined by collisional excitation equilibrium. Therefore, the rotational temperatureof CH CCH between different K -ladders is a good indicator of the kinetic temperature T K of thegas (Bergin et al. 1994). Indeed, propyne has been used to estimate the kinetic temperature ofhigh-mass star forming clumps (e.g., Molinari et al. 2016; Giannetti et al. 2017). Cyanopropyne isanother symmetric top detected in our observation, but it is much rarer than propyne. Our spectralsetup covers the CH CCH, J = 5 (cid:1) CCH by fittingGaussians to the four K components and modeling the rotation diagrams assuming LTE and opticallythin conditions. The latter is justified because the line’s optical depth never exceeds 0 .
1. We performthis fitting on each pixel where we detect at least one CH CCH line over 5 σ , where σ = 1 . − is the rms noise. It is necessary to do this Gaussian fitting in order to calculate each line’sintegrated intensity because the linewidths usually imply that the 5 , K (cid:1) , K K = 0 , , Minuit within the
Perl Data Language .While the temperature characterizing the excitation of different K -levels is close to T K , this isnot necessarily the case for the relative J populations and these may be characterized by non-LTEequilibrium. However, following (Bergin et al. 1994) and estimating the collisional cross section of RAS 16562 − Right Ascension (J2000) D e c li n a t i o n ( J )
10 19 28 37 46 55 64 73 82 91 100 log (N CH3C2H /cm -2 ) (K) NW cloud810NEC wall Nred N CH3CCH T ex Figure 12.
Propyne (CH CCH) physical parameters toward IRAS 16562 − CH CCH using CH CN, we conclude that the critical density for the 5 , K (cid:1) , K transitions is ≈ . × cm − . Using the density profile n ( r ) = 1 . × (0 . /r ) . cm − proposed for IRAS16562 − r being the radius from G345.49+1.47), we determine that thedensity of the clump is above the critical density for r < .
34 pc or for projected angular radii ≤ (cid:48)(cid:48) ,that is, encompasing all detected propyne emission. Thus, to calculate the total column density ofCH CCH we use T K and assume LTE conditions.Figure 12 shows the results of the fitting to the CH CCH lines. Uncertainties are shown in FigureB.1. Both the excitation temperature and column densities are consistent with previous single dishobservations of the 5 , K (cid:1) , K lines by Miettinen et al. (2006). They found a temperature of 35 .
9K and a column density of 1 . × cm − toward IRAS 16562 − CCH is not particularly intense towardthe CC core. Regions north of CC core and C8 are associated with the highest column densities of ≈ × cm − . These regions are associated with temperatures of typically 60–65 ± (cid:38) ±
30 K and they are detected toward low column densities regions. Generally,we detect warmer temperatures to the center of the clump than toward the outskirts. Averaging theCH CCH temperature in annuli of 2 (cid:48)(cid:48) starting from G345.49+1.47 we find that the radial temperatureprofile is well characterized by a decaying power law given by 80 K( r/ .
01 pc) − . . This temperature6 Guzm´an et al. dependance is shallower than the one suggested for the dust temperature by Guzm´an et al. (2010, ∝ r − . ), which may reflect dust and gas requiring densities above 10 cm − to be thermally coupled.Our results are in general agreement with what other studies have found toward high-mass starforming regions. Gibb et al. (2000), using single dish data taken toward the HMC G327.3 − CCH temperatures of 72 K and column densities of 2 . × cm − , respectively.Therefore, they suggest that this hydrocarbon better traces the warm, extended component ratherthan the hot gas. Based on the rotational temperatures derived toward seven HMYSOs, Bisschopet al. (2007) classified propyne as a “cold” molecule. Interferometer studies toward three “organicpoor” HMYSOs — which are expected to be younger and less chemically evolved than HMCs —indicate that CH CCH is as abundant far from the HMYSO as close to it, thus classifying it asan envelope molecule ( ¨Oberg et al. 2013; Fayolle et al. 2015). Propyne is also characterized bytemperatures between 40–60 Kand, in NGC 7538 IRS9, by a temperature profile which roughlyfollows ∝ r − . ..Being an unsaturated hydrocarbon, there are in principle efficient ion-neutral gas formation routesfor CH CCH (Schiff & Bohme 1979). Concordantly, there is no evidence that in IRAS 16562 − CCH has occurred in dust grains. First, the zero momentof all unsaturated molecules (except H CCO) including CH CCH are classified together within theContinuum group, consistent with ion-neutral gas reactions which tend to form unsaturated species.Second, let us assume that a large fraction of CH CCH is formed on dust grains and liberatedafterwards to the gaseous phase. This would imply that CH CCH should be well correlated withother molecules formed in grains, of which one of the best established examples is methanol. However,methanol emission does not correlate well with propyne as shown in Section 3.3. Finally, we notethat the continuum correlates well with propyne in IRAS 16562 − ≥
100 MeV) cosmic ray ionization, whichis homogeneous throughout the clump (Herbst & Klemperer 1973). That is, propyne’s abundanceseems to depends more on the total column density of material rather than other circumstances likethe presence of shocks, a higher temperature, or special illumination. We conclude that the goodcorrelation of CH CCH with the rest of the unsaturated species and the continuum, as well as thelack of correlation with CH OH and with shock tracers like SiO, are consistent with the ion-neutralgas reactions forming a significant fraction of propyne in IRAS 16562 − Column Densities and Excitation Temperatures
To determine excitation conditions and column densities we fit the molecular emission lines usingsimple models. It is possible to constrain the excitation state of species with several observed linessuch as CH OH and CH CCH. Other molecules such as SO, SO , OCS, HNCO, C H CN, c-C H ,HC N, CH C N, C H CN, and NH D also have several transitions which help determining theirexcitation conditions, but these higher excitation lines are only detected toward specific sources. Forthe rest of the molecules we detect either only one transition or the observed lines are unsuitable fordiscerning the excitation conditions of the gas (e.g., CCH and H CN).The sources for which we model the emission spectra correspond to those features identified inSection 3.2. Spectra for each of the sources are obtained by taking the primary beam correctedintensity (in K) versus frequency either toward specific directions or spatially averaging the intensityin the solid angle of the source. Spatial integration can improve the signal-to-noise ratio of the
RAS 16562 − • CC core.
Its spectrum is taken toward the peak methanol position, that is, 0 . (cid:48)(cid:48) − ◦ • C8.
We judge C8 not being a completely coherent structure, with significant differences betweenpositions closer to Source 8 and those closer to the maser a . Hence, we split the emission intwo sources: a 1 . (cid:48)(cid:48) a position. • N-red cloud.
We spatially average the emission in the region marked in Figure 7. • NW-cloud.
We spatially average the ellipse marked in Figure 7. Because this source is ratherlarge, in order not to fade out some weak lines we also take the spectrum toward the positionmarked with NW(a) in Figure 7. NW(a) correspond to the peak position of methanol (E )2 , (cid:1) , in the NW-cloud. • NEC-wall.
Toward this source we consider the spectra in two locations corresponding to themethanol and sulfur monoxide peaks, marked in Figure 7 with NEC-w(a) and NEC-w(b),respectively. • Diffuse Ridge.
The Diffuse Ridge is a much more elongated feature with varying characteristicsalong its extension. The size of the Diffuse Ridge also means it is likely affected by shortbaseline filtering. We select three positions to analyze the Diffuse Ridge, marked from (a) to(c) in Figure 7. Positions DR(b) and DR(c) correspond to the location of two cores in theDiffuse Ridge which have counterparts in several molecules. DR(a) is located on more diffusegas forming the body of this ridge or filament.We model the emission of all molecules except CH OH using a single excitation temperature (SET)model (van der Tak 2011), that is, we assume one excitation temperature per line of sight. Due tothe many detected lines of CH OH and because they are usually affected by non-LTE excitation, wemodel its emission using Radex (van der Tak et al. 2007). We assume, unless explicitly stated, thatthe beam filling factor of the emission is 1. Therefore, derived column densities are beam-averaged.Of course, in the cases of extended sources with spatially averaged spectra (e.g., the N-red cloud)these are source-averaged. For the line profiles, we model them as Gaussians with a single centralvelocity ( V LSR ) and FWHM (∆ V ) for all transitions from a specific molecule. We stress that due tothe Hanning smoothing of the ALMA data, the effective spectral resolution is 976 kHz. That is, theinstrumental broadening amounts to ≈ . − . In the SET model, the excitation temperature( T ex ), column density ( N ), V LSR , and ∆ V are free parameters. For methanol, free parameters are T K , the column density, V LSR , ∆ V , and density of the main collision partner (assumed H ). Forsimplicity, we assume equal abundances of the E- and A-CH OH symmetry states because the kinetictemperature (Section 4.1) is always larger than the 7.9 K energy difference (in k B units) betweenthe ground states of E- and A-CH OH (Friberg et al. 1988). The use of Radex for other moleculesis hindered by the detection of only one or two transitions, which makes the modeling unreliable.We find optimal parameters by minimizing the squared difference between the data and the model,8
Guzm´an et al. weighted by σ − , where σ is the uncertainty of the primary beam corrected data (in K). Typical σ for single pixel spectra (that is, not for spatially integrated) ranges in 0 . .
06 K. To minimize andcalculate formal uncertainties we use
Minuit and follow the prescription in Lampton et al. (1976).Tables 4 and 5 and Figures C.1 to C.13 show the results of the SET and Radex modeling.Table 4 shows the results of the Radex modeling of the CH OH lines. Columns (1) to (5) indicatethe source, T K , the logarithm of the column density in cm − (log ( N )), V LSR , ∆ V , and H density,respectively. Column (6) remarks some noticeable characteristics of the fittings or of the data.For some sources (NEC-wall, C8, and NW-cloud) we exclude from the fitting the CH OH, (E ) J K a K c = 5 − , (cid:1) , maser transition because this strong, non-thermal line would require opacities < −
1. These are characteristic of strong masers and cannot be adequately modeled by Radex (van derTak et al. 2007). In other sources, for example toward the CC core, lines are well modeled assumingLTE conditions, which implies a lower bound on the density. According to Radex, densities (cid:38) cm − are needed to thermalize the masering (E ) J K a K c = 5 − , (cid:1) , line. Densities ≥ cm − are usually enough to thermalize the rest of the observed methanol transitions. For those sourcesin which all methanol lines are thermalized except the (E ) J K a K c = 5 − , (cid:1) , we give a range ofcompatible densities.Comparing the (E ) J K a K c = 5 − , (cid:1) , maser transition with the expected LTE intensities isuseful to confirm the strong non-LTE effects on these lines. The maser spots a , b , and c identifiedin Figure 10 and whose parameters are given in Table 3 are associated with antenna temperaturesbetween 200 and 400 K. These values are larger than the expected LTE emission by factors of 200,1300, and 200, respectively. Assuming the angular size of the maser emitting regions covers less thana third of the beam size we obtain brightness temperatures ranging between 1500 and 3000 K. Inaddition, the linewidths of masers a and c given in Table 3 are also slightly narrower by ≈ . OH lines, which is also a characteristic of masers. The CH OH, (E ) J K a K c = 5 − , (cid:1) , line corresponds to a class I maser, that is, it is collisionally excited followed byspontaneous radiative decay (Cragg et al. 1992). These are the first class I methanol masers detectedtoward IRAS 16562 − OH maser detected previously is the class II (radiativelyexcited) 6.7 GHz maser MMB345.498+1.467 (Caswell 2009) detected toward Source 18. We notethat the CH OH class II maser is not associated with the most luminous source G345.49+1.47,but with the apparently more evolved and less embedded Source 18. In fact, neither class I nor IICH OH masers are associated directly with G345.49+1.47, but they appear scattered throughoutthe clump. This is also the case for the masers observed toward IRAS 16547 − ) J K a K c = 5 − , (cid:1) , transition), another clump believed to be in a similar evolutionary stage asIRAS 16562 − T ex ( T K for CH OH), log ( N ), V LSR , and ∆ V , respectively. In general,the V LSR of different molecules are not the same and it is not rare to find differences of ∼ − or more between different molecules for the same line of sight. For several sources, it is possible togather together sets of molecules which share similar velocities. Column (6) in Table 5 identifies thenumber of the group to which each particular molecule belongs. Considering the limited velocityresolution of the data, two groups are sufficient to account for the V LSR variations in each source.As a criterion for separating the two groups of V LSR , we require the internal standard deviation of
RAS 16562 − V LSR of the source.Molecules with a very different V LSR compared with either group (e.g., SiO or optically thick CS)are not classified. We find that two groups describe adequately the V LSR distribution of the CC core,the N-red cloud, the NEC-wall(b), DR(b), the NW cloud, and of the NW(a). For the rest of thesources, they are either well characterized by a single V LSR or the velocities have a large dispersionwhich cannot be grouped in two well differentiated sets.Table 5 gives formal uncertainties for the best-fit parameters, except for those which have beenassumed or kept fixed during the minimization. Note that most of the T ex are assumed for the SETfitting in Table 5. The procedure to assign the assumed T ex for each molecule start with moleculesfor which it is possible to derive a temperature (generally CH OH, CH CCH, and SO). The CH OHand CH CCH temperatures are estimators of the kinetic temperature of the gas. SO, on the otherhand, is associated with rather low T ex ≤
20 and thus it is likely sub-thermally excited. In mostcases, the assumed temperature for molecules without an independent T ex determination is one ofthese three.In order to assign a sensible T ex to molecules for which an independent temperature estimation isnot possible, we first check its V LSR group. Two molecules having very different V LSR is evidencethat they trace different gas and, therefore, assuming the same T ex is not justified. In addition,we asses the critical densities associated with the lines. Considering that the critical density ofthe detected SO transitions ranges between 1 . . × cm − , we assign the same temperatureas SO to all molecules whose transitions have critical densities above or equal to that of SO, thatis, to HC N, HNCO, SiO, CCH, HCN, CS, and to their isotopologues. An exception to this ruleoccurs when the derived CS peak optical depth — assuming T ex from SO — is above 2. Becausethe CS, J = 2 (cid:1) CO + , NH D, HCS + , and OCS) are closer to LTE, and therefore, we use either the CH OHor CH CCH derived T ex , depending on which one is closer in V LSR . We use CH OH or CH CCHtemperatures for molecules without collisional excitation parameters (like several COMs) and in thecase SO is not detected.Column (7) gives additional information about the fitting and noticeable characteristics of thespectra, for example, whether the line has absorption features or if it is associated with high opticaldepths. In Table 5, the T ex entries which have not been derived from the fit have a superscript ( † , ‡ , or †† ). This marker indicates the origin of the assumed temperature for the SET fitting: the samemarker appears in column (7) of the molecule from which T ex was adopted. In a few cases and lackinga better alternative, we also indicate in this column whether we use T ex from a molecule within adifferent velocity group or from another source.Note that Table 5 includes the CH OH parameters of Sources 3 and 18. Source 3 is a continuumsource with a spectral index characteristic of dust emission (Paper I). The CH OH fitting is of arather discrete quality, but we emphasize that this is one of the few lines of sight with emission inthe (E ) J K a K c = 6 , → , , v t = 1 transition ( E up = 340 . ii region associated with a HMYSO less massive than G345.49+1.47. Toward Source 18 Collision and Einstein coefficients have been obtained from the
Leiden Atomic and Molecular Database (Sch¨oieret al. 2005, http://home.strw.leidenuniv.nl/ ∼ moldata/) Guzm´an et al. æ æ æ æ æ æ æ æ æà à à à à à à à à ì ì ì ì ì ì ì ì ì ò ò ò ò ò ò ò ò òô ô ô ô ô ô ô ô ôç ç ç ç ç ç ç ç çá á á á á á á á áí í í í í í í í í
10 1005020 3015 1507012.012.513.013.514.014.515.015.5 T ex H K L l og H N Σ (cid:144) c m - L ž D v = m s - æ OCS à SO ì SO ò SO ô HCS + ç CS á NH D í SiO æ æ æ æ æ æ æ æ æà à à à à à à à à ì ì ì ì ì ì ì ì ì ò ò ò ò ò ò ò ò òô ô ô ô ô ô ô ô ôç ç ç ç ç ç ç ç çá á á á á á á á áí í í í í í í í í
10 1005020 3015 1507012.012.513.013.514.014.515.015.5 T ex H K L æ CH CCH à H CO + ì c - C H ò CCH ô CH CHO ç CH CH CN á CH C N í HC N æ æ æ æ æ æ æ æ æà à à à à à à à à ì ì ì ì ì ì ì ì ì ò ò ò ò ò ò ò ò òô ô ô ô ô ô ô ô ôç ç ç ç ç ç ç ç çá á á á á á á á áí í í í í í í í í ó ó ó ó ó ó ó ó ó
10 1005020 3015 1507012.012.513.013.514.014.515.015.5 T ex H K L l og H N Σ (cid:144) c m - L ž D v = m s - æ HC N à HC N ì H CN ò HN C ô CH OH ç H CCO á CH OCH í CH CHCN ó HNCO æ æ æ æ æ æ æ æ æà à à à à à à à à ì ì ì ì ì ì ì ì ì ò ò ò ò ò ò ò ò òô ô ô ô ô ô ô ô ô
10 1005020 3015 1507012.012.513.013.514.014.515.015.5 T ex H K L æ t - HCOOH à c - HCOOH ì HCOOCH ò CH CH OH ô NH CHO
Figure 13.
Each curve shows the minimum column density necessary to produce a Gaussian line peak of5 σ ≈ . − , in at least one transition in our spectral coverage, versus the excitationtemperature (assuming LTE). we detect CH OH lines including high energy ( E up >
100 K) transitions. Source 18 is one of theuncommon cases of a relatively isolated continuum source with an unambiguous line counterpart andthus a YSO with a reliable V LSR .Figures C.1 to C.13 (appendix C) show the SET and Radex best-fit models and the spectra inprimary beam corrected K versus frequency. Individual panels show usually one transition each, butsome panels show a few closely spaced lines (e.g., for CH CCH) of the same molecule. The name ofthe molecule and the upper energy level of the transition are displayed on top and in the top leftcorner of each panel, respectively. Models and data are shown in red and black, respectively. Thegreen bar indicates the frequency range used to calculate the squared difference between data andmodel. Some panels with faint detections show the ± . σ level, where σ depends on the specificspectrum and it is given in the caption.Finally, we emphasize that the lack of a molecule entry for a specific source in Table 5 is due tonon-detection. The exception are the sulfur oxides for the CC core (SO, SO , and isotopologues),which are not present in the table because their spectra were analyzed in detail in Paper I and theiremission cannot be well fitted by a single temperature model. The general criterion to discard adetection is the absence of any line attributable to the molecule over 2 . σ at a V LSR between − −
20 km s − . However, in some cases detection of the main isotopologue of a species lendscredibility to fainter spectral features located at the same V LSR . The non-detections are judged uponvisual inspection of the spectra in the positions where the strongest lines are expected. Figure 13
RAS 16562 − . σ ≈ .
143 K with∆ V = 4 . − versus temperature, assuming LTE conditions for different molecules. This ∆ V istypical of high-mass star formation regions but, in any case, the diagrams are easily scalable to othervalues because the peak of optically thin lines are proportional to the column density and inverselyproportional to ∆ V . Figure 13 includes the minimum column density of some species typical ofhot-cores which are not (or only dubiously) detected in this work. FRACTIONATION AND ISOMERIZATION IN IRAS 16562 − SiO, and SiO; HC N and H CN; H CO + and HC O + ; SO, SO, and SO; CS, and C S; OCS and O CS; and CH OH and CH OH. In addition to these, we study theH CN/HN C isotopomer fractionation. In principle, we can asses the silicon, sulfur, and carbonfractionations more directly because the isotopic difference between the molecules involve only oneatom. On the other hand, for hydrogen cyanide and formylium the differences between the observedisotopologues involve two isotopes from different elements, complicating somewhat the interpretation.The following sections present the analysis of the isotopic and isomeric fractionation toward IRAS16562 − + , and HCN isomers is extended, hence, Table 2 gives the average fractionation ratiosmeasured toward the clump. The rest of the ratios characterize emission arising from the centralcore.The isotopic proportion of many atoms are found to vary with Galactocentric radius (Wilson1999).We use 6 . − Silicon Fractionation
SiO and SiO have been observed toward IRAS 16562 − . − , Harju et al. 1998). Excitation temperatures reported by Miettinen et al.(2006) are typically 4 . ± . (cid:1) (cid:1) − . ± . . ± . − for SiO and SiO, respectively. Their ratio,15 .
1, is lower than the solar abundance value of 19 . SiO]=17 . ± .
1, who also argue forthe lack of a Galactic gradient in the silicon isotopic distribution. The isotopic ratio varies within aninterquartile range of [9 . , .
2] (Table 2), but with no evident systematic spatial trend. Line ratioslower than the expected isotopic abundance proportion are usually interpreted as evidence of opticallythick emission, which probably characterizes the main isotopologue ( SiO) lines (Penzias 1981).Optical depth introduces an additional complication, because its effects and abundance changes are2
Guzm´an et al. in principle degenerate (the study by Monson et al. 2017 models the opacity broadening of the lineto estimate τ independently).From our ALMA data, we find line ratios between SiO and SiO ranging between 9 . . .
2. We integrated the line emissionin the same velocity range for each line ( −
24 to 0 km s − ) to calculate these values, but we note thatusing the zero moment maps (obtained through moment masking) gives similar results. The ratios weobtain are comparable to the ratios calculated by Monson et al. (2017) toward clouds in the centralmolecular zone of the Galaxy. Monson et al. do not attribute these low values to fractionation, butexplain them as due to optical depths (cid:38)
1. This is likely only part of the explanation for the valuesobtained for IRAS 16562 − SiO line ratio increases to a median valueof 13 . V LSR . Presumably, emission from the line wingsis more likely associated with optically thin column densities. However, the proportion is still lowcompared with the mean Galactic value. Because it is also possible that the main isotopologue linebeing affected by short spacing losses, we refrain from attributing the low [SiO]/[ SiO] line ratios tofractionation.We derive SiO column densities toward the N-red cloud, the S8 maser position, the NW cloud,and position (a) from the NW cloud, and found [SiO]/[ SiO] values of, respectively, 14 . ± . . ± .
0, 18 . ± .
3, and 12 . ± .
7, where the error bars represent formal uncertainties. Thesevalues are somewhat higher that those obtained from the zero moment maps. It may be possible that,because they are associated with emission from rather compact sources, they could be less affectedby short spacing losses. It is apparent that a fair assessment of the SiO fractionation toward IRAS16562 − SiO and SiO emission to have opacities lower than those of the main isotopologue,and also being less affected by short baseline filtering. Indeed, the SiO/ SiO velocity integratedline ratios are distributed within an interquartile range of [1 . , .
6] and a median of 1 .
4. This valueis the same as the median ratio obtained by Monson et al. (2017), and it is very similar to the solarvalue (1 .
5, Asplund et al. 2009). We also derive SiO column densities toward the N-red cloud, theS8 maser position, and the NW cloud, observing little variation: we obtain [ SiO]/[ SiO] values of1 . ± .
1, 1 . ± .
1, and 1 . ± .
26, respectively.Finally, we emphasize that we do not find obvious spatial trend or pattern in the distribution ofany of the SiO isotopologue ratios. This is in line with what is found by Monson et al. (2017).Because the origin of SiO in the gas phase is likely dust sputtering in shocks, it is unlikely thatthis mechanism be sensitive to the rather small relative difference between the molecular masses ofthe SiO isotopologues. Furthermore, any chemical mechanisms affecting the SiO relative isotopicabundance in the clump probably act on timescales larger than the depletion time of SiO onto dustgrains, therefore, not generating detectable fractionation.5.2.
Sulfur Fractionation
In this section we analyze the relative abundances of the sulfur isotopologues S (main), S and S. The only source toward which we claim detection of SO is the CC core. Assuming an excitationtemperature of 64 K, the SO column density toward the CC core is 9 . ± . × cm − . Combiningthis with the SO column density given in Table 5, we obtain [ SO]/[ SO]= 5 . ± .
6. This valueis remarkably close to the solar isotopic proportion [ S]/[ S]= 5 .
64 (Asplund et al. 2009). The
RAS 16562 − SO abundance respect to the main isotopologue ( SO)was consistent with solar abundance. Note that the isotopic proportions of sulfur are predicted tovary little with Galactic radius (Chin et al. 1996): at 6 . S]= 27 ± S] ratio because in the directions where we detect C S, the CS line isoptically thick. 5.3.
Carbon Fractionation
We can compare the abundance of C-bearing molecules with their main isotopologue using linesfrom O CS, CH OH, and HC CCN. We avoid using the lines of H CCCN and HCC CN becausethey are blended with lines of HNCO and C H CN, respectively.O CS and CH OH are only detected toward the CC core. Using the column densities of Table 5,we obtain [OCS]/[O CS]= 25 ± .
4, assuming the same excitation temperature for both molecules.This ratio is below the expected value for [C]/[ C] ∼
60 (Wilson 1999; Milam et al. 2005) and itis similar to the values found by Tercero et al. (2010) toward Orion-KL. The most likely cause forthis difference is that the OCS line is optically thick. LTE fittings with derived column densitiescompatible with the solar isotopic abundances require peak optical depth ( τ p ) of 1 .
65, filling factors (cid:46) .
3, and excitation temperatures higher than 100 K.The isotopologue ratio calculated from the methanol column densities gives [CH OH]/[ CH OH]=26 ±
4. This number is comparable to the one obtained above using OCS, and it is a factor of ∼ OH line ( τ p ≈ ) J K a K c = 5 − , (cid:1) , transition) with a filling factor of 0 .
24 give isotopic ratios ∼
50, closer to the expected [C]/[ C] value.Finally, we can calculate the [HC N]/[HC CCN] quotient toward the CC core, the NEC-wallpositions (a) and (b), the C8 maser position, and the C8 1 . (cid:48)(cid:48) ±
4, 41 ±
4, 37 ±
5, 60 ±
6, and 50 ±
4, respectively. The errorbars give the formal uncertainty, but there is a likely more relevant systematic uncertainty associatedwith the excitation temperature. Note that because we only detect one line for each isotopologue,the assumed excitation temperature is critical. The ratio obtained for the CC core emission is closeto the values obtained for O CS and CH OH. Again, the column density quotient between thetwo isotopologues can accommodate a value closer to 50 if the opacity near the peak of the HC Nline is 1 . (cid:46) .
5. The ratios increase with distance from the center of IRAS16562 − C abundance for the C8 sources. This is consistent with the view that quotients lower than theexpected isotopic abundance ratio are caused by opacity.5.4.
Formylium Cation: Oxygen and Carbon Fractionation
The median value we calculate for the velocity integrated flux ratio between H CO + and HC O + ,within the range −
15 km s − < V LSR < − − , is 6.9, with an interquartile range of [4.8,8.5]. This abundance ratio depend on carbon and oxygen isotopic abundances, which at the clump’slocation are [ C]/[ C]=59 . ±
18 and [ O]/[ O]=442 . ± CO + ]/[HC O + ]=7 . ± .
0. Using instead [ C]/[ C]=63 . Guzm´an et al. [H CO + ]/[HC O + ]=7 .
0, which is within the previous value uncertainty. The median value foundtoward IRAS 16562 − CO + /HC O + line ratios: the central parts of IRAS 16562 − (cid:48)(cid:48) from G345.49+1.47) are associatedwith ratios higher than average, with a mean value of ∼ .
5. Regions farther away from the centerof the clump, like the emission associated with Sources 7 , 3 , and 2 (the Continuum arc) and tosources in south east like 15 and 17; exhibit ratios ∼
5. This behavior of the line ratios is rathersurprising because it is the opposite to what would be expected if the lines from the central parts ofIRAS 16562 − + observations would determine whether the cause of this decrement is fractionation,and of which isotopologue. Finally, we mention a possible path to increase the H CO + abundancenear the center of the clump respect to HC O + : it may merely reflect CO being better (self)shielded than C O from radiation arising from G345.49+1.47, simply because of the lower abun-dance of the latter. We note that the HMYSO has contracted and its high energy radiation hasalready ionized a small region in the center of IRAS 16562 − + is formed mainly bycombining H +3 and CO, an increase of CO respect to C O could produce a corresponding increasein the formylium isotopologue.5.5.
Hydrogen Cyanide: Nitrogen and Carbon Fractionation
Figure 14 shows in the left panel the quotient between the moment zero images of the H CN andHC N, J = 0 (cid:1) C proportion and because HCN transitions are usually opticallythick, the [H CN]/[HC N] ratio is commonly used to study the N vs. N fractionation. Singledish observations of IRAS 16562 − N] behavior with Galacto-centric radius. While Wilson (1999) — based on H CN and HC N observations from Dahmen et al.(1995) and Wannier et al. (1981) — suggests a value of 420 at the distance of IRAS 16562 − C] curves (illustrating theuncertainties introduced by the double isotope corrections) but this is not clear since there are al-ready noticeable differences between their [HN C]/[H NC] and the [H CN]/[HC N] values givenby Dahmen et al. (1995). In view of these uncertainties, we focus the quantitative analysis directlyon the [H CN]/[HC N] ratio.According to the image shown in Figure 14, the mean H CN/HC N line quotient (which is thesame as [H CN]/[HC N], assuming the same excitation temperature for both molecules) rises fromvalues ∼ . . (cid:48)(cid:48) from the center of IRAS 16562 − . . (cid:48)(cid:48) from the clump center.These values are dominated by the higher quotients measured toward the Diffuse ridge and the NW RAS 16562 −
18 1716151413121103:20.030.040.050.0-40:04:00.0
418 17161514131211109 8 7654 32 03:20.030.040.050.0-40:04:00.0
418 17161514131211109 8 7654 3203:20.030.040.050.0-40:04:00.0
Right Ascension (J2000) D e c li n a t i o n ( J ) NW cloudNredNEC wall D. ridge H CN/HC N H CN/HN C Figure 14.
Left panel.
Moment zero quotient between the H CN and HC N, J = 0 (cid:1) Rightpanel.
Moment zero quotient between the H CN and HN C, J = 0 (cid:1) CN moment 0 emission. Cyan contours correspond to NH Demission as in Figure 15. In both panels we indicate the N-red, NW- and Diffuse ridge markers as in Figure7. cloud. The values obtained for [H CN]/[HC N] from the LTE fittings (Table 5) for each source aresimilar to those read from Figure 14. The mean [H CN]/[HC N] is 5 . ±
2, where the uncertaintyrepresents the dispersion observed between different sources. This value is slightly lower, althoughwithin the uncertainties, compared with that given recently by Colzi et al. (2018) at 6.9 kpc.Using single dish data, Dahmen et al. (1995) obtained [H CN]/[HC N]= 6 . ± .
16 for IRAS16562 − − CN line is associated with an optical depth at peak of ∼
4. Such high opacity would affect theobserved relative intensities of the hyperfine components, deteriorating noticeably the quality of thefits: we do not observe in general large hyperfine anomalies which could be evidence of an opticallythick H CN line. It is still not clear how much influence the short baseline filtering has on the linequotients, but at least for the CC core, it is not likely to dominate.Hence, it is possible that there is real fractionation effect toward the densest cores of IRAS16562 − CN]/[HC N] show that this proportion varies dependingon the nature of the source. For example, Wampfler et al. (2014) finds that the quotients associatedwith three low mass protostellar envelopes are 2 .
4, 4 .
2, and 5 .
3. The causes for the different valuesare not entirely clear, but the lowest ratio (highest HC N fractionation) is associated with the hotcorino IRAS 16293 − Guzm´an et al. low ratios ( ≤ .
0) toward proto-planetary disks. Of the two fractionation mechanisms suggestedby Guzm´an et al. (2015b), one of them only acts effectively in cold gas ( ≤
20 K), which is notcharacteristic of IRAS 16562 − N] ratio should increase the more illuminatedwith dissociating radiation and diffuse a cloud is. In this case, the same self-shielding from theHMYSO radiation which may cause the H CO + -HC O + spatial fractionation could also explainthe observed [H CN]/[HC N] ratios.5.6. H CN/HN C Isomer Ratio
We can explore the HCN to HNC isomer ratio in IRAS 16562 − C isotopologues,whose transitions have the advantage of being more likely optically thin compared to those of themain species. The HCN/HNC ratio has been found to depend on kinetic temperature (e.g., Schilkeet al. 1992). Theory proposes that the two molecules are produced in a ratio ∼ + (Herbst 1978; Hirota et al. 1998). Selective destruction ofHNC through the neutral-neutral reaction HNC + H → HCN + H is expected to work at highertemperatures because of the presence of an activation barrier, depleting HNC in favor of HCN. Thequestion of what is the energy of this activation barrier is still not completely resolved (Graningeret al. 2014).The right panel of Figure 14 shows the quotient between the moment zero of the H CN and HN C,1 (cid:1) − CN/HN Cquotient is comparable to unity. Close to the center of IRAS 16562 − CN/HN C quotient increases by one order of magnitude. It isalso apparent that the molecular ratio is higher in the NE side of the Diffuse ridge, being presumablyilluminated and heated more directly by G345.49+1.47. For the CC core, the column density givenin Table 5 compared with the upper limits from Figure 13 suggests a ratio ∼ D in combination with HN C, the H CN/HN C ratios have values close to unity, againconsistent with the low temperatures expected to be traced by deuterated species (Bergin & Tafalla2007).The right panel of Figure 14 shows the quotient between the moment zero of the H CN and HN C,1 (cid:1) CN and the HN C emission, the mostnoticeable in the moment zero map is the lack of a CC core counterpart in HN C, in contrast toH CN. In addition, it is not uncommon these lines having different V LSR , as shown for several sourcesin Table 5.The differences can be understood as due to the depletion of HNC in favor of HCN in gas at tem-peratures >
25 K, which implies that warmer gas along a given line of sight will weight more into theH CN emission than HN C. Arising from different locations, warm and cold gas do not necessarilyshare the same V LSR . This interpretation implies also that the effective temperature of the HCNemission should be higher than that of HNC (Jin et al. 2015). Conversely, low temperatures shouldbe associated with HCN/HNC column density ratios close to unity, the same effective temperaturefor both species, and consistent V LSR . DISCUSSION
RAS 16562 − Right Ascension (J2000) D e c li n a t i o n ( J ) Figure 15.
Both panels show in the background three color NIR images (red, green, and blue for K s , H,and J band filters, respectively) obtained from the VVV survey. Left panel.
White contours show CH OH,2 , (cid:1) , zero moment emission. Levels: 10, 30, 50, 70, and 90% of the peak (0 .
51 Jy beam − km s − ). Right panel.
Cyan contours show deuterated ammonia (
J, K ) = (1 ,
1) zero moment emission. Levels: 10,30, 50, 70, and 90% of the peak (0 .
10 Jy beam − km s − ). Red crosses mark the continuum sources 2, 3,7, 8, 10, and 12. Green-dashed ellipses show the location of the NW and N-red clouds, and the NEC-wall.Black crosses mark the DR(a), (b), and (c) positions. We can gain some insight on the chemistry and physics of the clump by comparing ours withindependent observations. In Section 6.1 we compare our ALMA with NIR data taken from theliterature. The latter are some of the few data on IRAS 16562 − − Near-Infrared Counterparts of Molecular Features
Figure 15 shows a three color NIR image (JHK S filters) of IRAS 16562 − OH and NH D maps, respectively. Guzm´an et al. (2016) analyzed similar NIRimages toward G345.49+1.47, describing among other features the illuminated blueshifted outflowcavity and some continuum sources with NIR counterparts. In this section we focus on the molecularemission counterparts.There are three NIR conspicuous features which correlate with the CH OH zero moment emission.The most evident is the CC core, which is coincident with a NIR source associated with the HMYSOG345.49+1.47. As already noted before, the CC core emission extends a few arcseconds in the northand north-east directions respect to G345.49+1.47. The second feature is coincident with the northernwall of the blueshifted illuminated cavity, particularly, the strong CH OH emission associated withthe NEC-wall. Additional methanol emission seems to follow the location of the southern wall of thiscavity, although in a less clear way compared to the NEC-wall. The third feature is evident in the left8
Guzm´an et al. panel of Figure 15, which shows that the Diffuse Ridge seen in the CH OH contours coincides witha dark filament seen across the brightly illuminated blueshifted outflow cavity. From the brightnessdecrement produced by the filament in the K s image, which amounts to A K s ≈ . column density of 1 . × cm − (Rieke & Lebofsky 1985; Heiderman et al. 2010). Themedian V LSR of the Diffuse Ridge is − . − (Section 4.2), which is redshifted respect to theCC core. A likely interpretation is that the Diffuse Ridge material is in front of the outflow cavity,obscuring the illuminated cavity, and possibly falling toward the clump.The right panel of Figure 15 shows NH D contours against the NIR background. Also indicated inthis panel are the positions of continuum Sources 1, 2, 3, 7, 15, and 16. We can see that Sources 2,3, and 7 delineate what we called in Section 3 the Continuum Arc, which is also seen in NH D. TheNIR image shows that the Continuum Arc is embedded within a large IR-dark absorption region.The absence of stars observed toward this area suggests large column densities. There is a lessevident NIR counterpart to the NH D peak located in the northeast part of IRAS 16562 − D emission corresponds with an absorption patch where we see only most likely foregroundstars, evidenced by their blueish hue. The deuterated ammonia line peaks at − . − and hasa FWHM of ∼ − . Note that the true linewidth is smaller and closer to 2 km s − due to theinstrumental broadening (see Section 2).The most clear NIR counterpart to the NH D emission is located to the southeast, coincident witha very clear IR-dark cloud ∼ (cid:48)(cid:48) size. This IR-dark cloud was already noted by Guzm´an et al.(2016) because it is located nearby the outer-eastern ionized lobe of the G345.49+1.47 jet. Thespatial coincidence and the kinematics of the jet’s lobe suggest that the jet may be interacting withthis cloud. Its NH D emission peaks at − . − with a FWHM of 4 . − , which impliesthat the dark cloud is indeed part of IRAS 16562 − − D (Bergin & Tafalla 2007) and by all the rest of thedetected spectral features being in absorption. Lines of CS, CH CCH, CCH, H CN, HN C, andCH OH appear in absorption toward both the south and northeast NH D peak positions.6.2.
Chemistry in IRAS 16562 − Studying the chemistry of a single high-mass protostellar molecular clump is a complicated tasksince it encompass many of the mechanisms expected to be at work in the ISM. Until the recent adventof instruments such as ALMA the chemistry of high-mass protostellar clumps had been studied usingsingle-dish telescopes. These mix different environments in their beams, having to rely on velocitydifferences to discern them. With high resolution and sensitive observations we are now able to studyin detail these regions, and spatially separate the emission of each molecule. Such studies are veryuseful to advance our chemical knowledge.Near the end of the prestellar stage, a high-mass molecular clump is characterized by temperatures <
15 K, hydrogen column densities (cid:38) cm − , and densities (cid:38) cm − . These are the characteristicsof the so-called quiescent infrared dark clouds (Guzm´an et al. 2015a; Rathborne et al. 2006). In thisstage, dust grains are covered by (mostly water) icy mantles together with other adsorbed atoms andmolecules like H, CO, N , and possibly CS and H S (Viti et al. 2004). At these low temperatures, onlyH is mobile on the dust’s surface and it reacts with other molecules (like CO) using the dust grain as athird body which helps dissipating the excess energy. Chemical products formed in the dust grain icy
RAS 16562 − CO and CH OH (Vasyunina et al.2014). Meanwhile, the cold gas in the clump also evolves chemically, specially through ion-neutral andbarrier-less neutral-neutral reactions. A permanent, small amount of cosmic-ray ionization facilitatesthe gaseous ion-neutral reactions and formation of simple unsaturated molecules.At these early evolutionary stages, most of the molecules formed on dust grains remain there: thetemperature is too low to sublimate them from the surface. About all of the ices will co-desorb withwater when the temperature reaches ≈
100 K. However, even at low temperature several non-thermalmechanisms can help desorbing some of these products, e.g., shock sputtering, chemical desorption,and photo-desorption. Some of these mechanisms — not being entirely clear which one(s) — areobserved to be efficient at releasing large quantities of CH OH during the prestellar phase (e.g.,Sanhueza et al. 2013; Vastel et al. 2014; Cosentino et al. 2018).As the clump evolves, it contracts, and eventually young stars are born. These (HM)YSOs increasethe clump’s temperature and introduce turbulence through outflows and winds. Temperature risingto 20–30 K reduces the time spent by H in the surface of the grains, hampering further hydrogenationreactions. The temperature rise increases the mobility of heavier radicals in the mantles, allowingfurther reactions and promoting the formation of more complex molecules. This phase of chemicalevolution receives the name of warm-up (Garrod et al. 2008). The radicals may be produced byphotolysis of other molecules induced by secondary cosmic rays UV-photons. Some other species(CO, N , O , CH ) desorb with the increasing temperature (Viti et al. 2004), potentially openingnew gaseous chemical routes.The third phase of chemical evolution comes only near the regions where the temperature increasesto ≥
100 K, that is, very near the young stars and inside the HMCs. Here, the water ices completelyevaporate and liberate all the molecules formed in the dust mantles to the gas phase. The temperaturerise also allows reactions with barriers and endothermic reactions to occur. Characteristic chemicalproducts generated during the three stages described above — cold prestellar, warm-up, and HMC —are called by Herbst & van Dishoeck (2009) zeroth-, first-, and second-generation species, respectively.Note that the dominating chemical mechanism in each stage is mainly determined by the temper-ature of the gas, which ultimately depends on the distance to the young forming stars. Protostellarclumps, like IRAS 16562 − − HMCs and embedded YSOs
HMCs are defined observationally as dense ( > cm − ), warm (100–500 K), and compact ( < . OH andCH CN signpost the presence of a HMC. In general, it is assumed that the heating source of a HMCis a HMYSO located near its center.0
Guzm´an et al.
There are four compact continuum sources within IRAS 16562 − OH lines. Source 3 is a continuum source associated with hot methanol ( T ex ≈
200 K) but noIR counterpart, which prompts us concluding that it correspond to a HMC associated with a veryembedded YSO. Source 18, on the other hand, is conspicuous in IR (2MASS and
Spitzer /IRACbands) and it is associated with the high-to-intermediate mass YSO GLIMPSE G345.4977+01.4668( L bol ≈ L (cid:12) , Benjamin et al. 2003). It is not clear that Source 18 is associated with a HMC because— being bright at NIR — the YSO cannot be very embedded. In addition, the CH OH temperature(64 K) and column density (5 . ± . × cm − ) are significantly lower than those of the rest ofthe HMCs.Source 8 and 10 are associated with two molecular cores, respectively, C8 and the CC core. Thesecores are the richest sources of molecular lines in IRAS 16562 − >
100 K and hydrogen saturated COMs, which are characteristic of HMCs. The heatingsource is the HMYSO G345.49+1.47, which ionizes an associated HC H ii region. It is natural toassume that this ionizing radiation will also dissociate a large fraction of molecules in the immediatesurroundings of the HMYSO, thus, the molecular emission does not need to coincide exactly withthe HMYSO position (e.g., Mookerjea et al. 2007). C8, on the other hand, is embedded in a densemolecular envelope which extends to the northwest. This envelope apparently engulfs the positionof Maser a , which is associated with a previously unreported continuum source of 2 . a pinpoints the location of another embedded YSO. C8 is also characterized byCH OH lines with T ex >
100 K and other COMs with T ex >
70 K revealing the presence of a HMC.In contrast to the CC core, no evidence of free-free emission is observed toward C8.In order to contrast the chemical composition between the CC core and C8, we compare theabundances of several molecules respect to methanol. The main reason to derive methanol-normalizedabundances is that the hydrogen column density toward the CC core is very difficult to estimatedue to most of the 3 mm continuum corresponding to free-free emission arising from the HC H ii region. We focus first on the abundance of O- and N-bearing molecules, which is a rather commonapproach in the literature (e.g., Widicus Weaver & Friedel 2012; Allen et al. 2017). This strategyis also justified because the only hydrocarbon COM (that is, without N or O), CH CCH, seemsto trace the clump material on a large scale rather than gas associated with the molecular cores.Identifying the O- and N-bearing groups was originally motivated by observations of Orion-KL, whichshow that nitrogenated and oxygenated molecules are better associated with two distinct structures,respectively, the HMC and the compact ridge (Feng et al. 2015, and references therein). A thermalsegregation is also observed, with N-bearing molecules tracing hotter gas compared to the O-bearingspecies (Crockett et al. 2015). N-O segregation has been found toward other high-mass star formingregions as well (Su et al. 2005; ¨Oberg et al. 2013; Allen et al. 2017), but it does not seem to beuniversal (Fontani et al. 2007).Figure 16 compares the methanol-normalized abundance of O- and N-bearing molecules with fouratoms or more, detected toward the CC core or C8 (thus excluding CH CHO). Figure 16 also showsabundance ratios associated with well-known HMC sources such as Sgr B2 (N) and (M) (Bellocheet al. 2013), NGC 7538 IRS1 ( ¨Oberg et al. 2014), G24.78, NGC6334 IRS1, and W3(H O) (Bisschopet al. 2007). We note that the relative abundance ratio respect to CH OH for different sources canvary within several orders of magnitude, depending on the molecule. Part of this variation may be
RAS 16562 − Figure 16.
Blue and green bars show column density ratios for O- and N-bearing molecules more complexthan four atoms toward the CC core and C8 (C8 1 . (cid:48)(cid:48) . (cid:48)(cid:48) OH]ratio obtained toward the CC core using the column density of HNCO given in Table 16 associated with the second velocity component. due to most values reported in Figure 16 are obtained from single dish telescope observations, whichcombines emission from several regions in the beam. Therefore, these ratios are useful only as roughestimations. In any case, the abundance ratios toward the CC core and C8 are comparable to thoseof other sources, within the observed variations.The only molecule in Figure 16 more abundant in the CC core than in C8 respect to CH OHis CH OCH . We do not detect NH D toward C8, but its abundance upper limit is above the[NH D]/[CH OH] ratio measured toward the CC core. We measure the same abundances respect tomethanol of C H CN and H CCO for both cores. For the rest of the molecules (HC N, HC N, HNCO,C H CN, and CH C N), we observe larger abundances toward C8 than toward the CC core. We adda note of caution about HNCO. Lines from this molecule toward the CC core are characterized bya T ex and V LSR similar to those of CH CCH, hence they are not directly associated with the HMC.However, there seems to be in addition a hotter component, prompting us to conclude that theobserved HNCO lines blend emission from the HMC and its colder envelope. We indicate in Figure16 both the [CH OH]/[HNCO] lower limit and the ratio using the column density of Table 5.There are two COMs more complex than CH OH detected toward both the CC core and C8:methoxymethane and propanenitrile (C H CN). These two molecules are representative of O-bearingand N-bearing COMs, respectively. The CH OCH and C H CN lines toward C8 peak at the locationof Source 8. Toward the CC core, they peak together with the highest energy CH OH transitionsdespite CH OCH and C H CN lines coming from relatively low upper energy ( E up <
40 K) tran-sitions. Similar to CH OH, a hotter CH OCH component is blended in the CC core emission: theLTE fitting ( T ex = 138 K) underpredicts the 16 , (cid:1) , transition with E up = 136 . Guzm´an et al.
Methoxymethane to methanol ratios of 1 . ± . × − and 4 ± × − are measured towardthe CC core and C8, respectively. These ratios are within the values found toward most high-massstar forming clumps. The [CH OCH ]/[CH OH] ratio does not vary much among different sources,which ¨Oberg et al. (2014) interprets as evidence of CH OCH forming on the ice mantles of dustgrains and co-desorbing with CH OH. Compared with methanol, methoxymethane is slightly lessabundant toward C8 compared to the CC core. In this regard, C8 is somewhat more consistent withcolder ( <
50 K) sources ( ¨Oberg et al. 2014).The [C H CN]/[CH OH] ratio shows a range wider than that of CH OCH between differentsources, from ∼ − in G35.03A (Allen et al. 2017) to 0 . H CN] / [CH OH] ≈ . × − , whichis consistent with other hot cores like G34.26+0.15SE (Mookerjea et al. 2007) and the sample ofBisschop et al. (2007) (see Figure 16). We do not detect clear evidence of thermal or spatial N-OCOM segregation in either core. However, we note that at least in Orion-KL, CH CN shows the N-Osegregation much more clearly than C H CN (Crockett et al. 2015).We can explain some of the abundance differences shown in Figure 16 if C8 is chemically moreevolved than the CC core. There are two facts which support this view: (i) the [C H CN]/[C H CN]abundance is larger in the C8 core than in the CC core — acrylonitrile is not detected in the latter— and (ii) the relative abundance of cyanopolyynes HC N and HC N in both cores.We assume that emission from C H CN and C H CN associated with the CC core and C8 tracesthe hot molecular gas because of their high temperatures, which is also usually the case toward otherHMCs (Fontani et al. 2007). Using HMC chemical models Caselli et al. (1993) find that, followinggrain mantle evaporation, the C H CN abundance increases respect to C H CN because of the activa-tion of several gas destruction routes of C H CN which produce C H CN. The [C H CN]/[C H CN]ratio, therefore, increases monotonically with time. An underlying assumption which justifies thisconclusion and that C8 is chemically more evolved than the CC core is that, right after grain mantleevaporation, the [C H CN]/[C H CN] ratio of the C8 core was similar (or even perhaps lower) thanthat of the CC core.The [HC N]/[HC N] (cyanodiacetylene to cyanoacetylene) ratios of both cores also supports C8being more evolved than the CC core. These ratios are ≈ × − (assuming T ex of cyanoacetyleneranging between 17 and 77 K) and ≤ × − for C8 and the CC core, respectively. This is adifference of more than an order of magnitude in the relative abundances of these two cyanopolyynes.Because toward the CC core the V LSR of HC N and CH OH are very similar we assume that thesetwo molecules trace gas at the same temperature (107 K). We use this temperature to calculate theCC core upper limits on the HC N column density. Toward C8 we are able to derive an excitationtemperature of 77 ±
10 K for HC N, which is very similar to that of C H CN but somewhat lowerthan that of CH OH. Toward C8, HC N and HC N are detected at a consistent V LSR .Cyanopolyynes abundance models in HMCs (Chapman et al. 2009) indicate that important gasformation routes for HC N and HC N start with acetylene (also ethyne, C H ) and CN. Chapmanet al. (2009) find that HC N and HC N are formed in succession. The [HC N]/[HC N] abundanceratio is consistent with HC N not having formed yet in the CC core whereas in C8, being older,HC N and HC N have already formed in large quantities. In addition and in consistency with thispicture, HC N is more abundant respect to CH OH in C8 respect to the CC core. In the literature,cyanopolyynes more complex than HC N are not commonly detected toward HMCs. Figure 16 shows
RAS 16562 − N]/[CH OH] ratios of both cores are comparable to that of Sgr B2 (M). On the otherhand, HC N is not detected toward Sgr B2 (N), while it is very abundant respect to CH OH in SgrB2 (M).We propose that the CC core is younger than C8 despite the former is associated with a HC H ii region whereas C8 is not. This is consistent with the HMYSO associated with the CC core, namelyG345.49+1.47, being more massive than the YSO heating the C8 HMC. The typical timescale forprotostars to contract and evolve to the main sequence is given by the Kelvin-Helmholtz contraction,which is faster for high-mass protostars (Tan et al. 2014). Furthermore, we also expect that pre-stellar contraction — characterized roughly by the free-fall timescale — was faster for the denser CCcore compared to that of C8.Another molecule with a different abundance in both cores is CH C N. It is not clear, however,that cyanopropyne is tracing the HMCs because of its low temperature and V LSR similar to that ofpropyne. More likely, CH C N abundances are representative of the envelope material of the YSOsassociated with the CC core and C8. The cyanopropyne physical parameters are consistent with itforming from propyne through the gas reaction CH CCH + CN → CH C N + H (Balucani et al.2000). Higher amounts of CH C N associated with C8 could reflect an older age for this clumprespect to the CC core. Note that the previous reaction also generates cyanoallene (CH CHCCN)with a theoretical branching ratio of 1. Our data is not sensitive enough to detect a column densityof cyanoallene similar to that of cyanopropyne, but in principle, this would be a way to test thistheoretical chemical formation path.While some differences in the abundance of N- and O-bearing molecules between both cores arenoticeable, the chemical differences are most conspicuous in sulfuretted species, and specially in thesulfur oxides. Figures 2 and 3 show that most of the SO and SO is associated with the centralsource, with no SO and little SO emission detection toward C8. Paper I already shows that sulfuroxides are good tracers of the central HMC and, as the complex COMs, their emission indicate thepresence of more than one thermal component.However, a careful analysis of the sulfur oxides compared with other HMC tracers indicates that theSO and SO emission does not originate from what we have called the CC core. This is evidenced bya kinematic feature which is rather unique to the SO and SO emission: the rotating core morphologyaround G345.49+1.47. Among our detected species (and CH CN, see Cesaroni et al. 2017), thereis no other molecule which displays this feature. Toward the CC core direction — ∼ . (cid:48)(cid:48) and methanol are comparable, that is, their abundance ratiois on the higher end of the range of observed [SO ]/[CH OH] values toward other sources, whichfluctuate between 1 and 10 − (Wright et al. 1996; Mookerjea et al. 2007; Allen et al. 2017). Closer toG345.49+1.47, however, the SO and SO column densities are approximately one order of magnitudelarger than that of CH OH.What special characteristics does the gas in the rotating core have? What chemical processesare responsible for the enhanced abundance of sulfur oxides? Definitive answer to these questionscannot be given, but it is possible that part of it is related with the illumination of the core by UV-photons from G345.49+1.47. Based on the results of Ferrante et al. (2008), Paper I already arguedfor prolonged UV-illumination of ices to explain the dominance of sulfur oxides in the rotating coreversus other sulfuretted molecules like OCS or CS, which are not formed efficiently in the gaseousphase (Charnley 1997). However, explaining the lack of the rest of the molecules is more difficult.4
Guzm´an et al.
Other systems which may bear some resemblance with the rotating core are the circumstellarenvelopes of M-type (O-rich) asymptotic giant branch (AGB) stars. They consist of dense andmostly molecular gas, heavily influenced by the UV radiation from the central degenerate core. Theydo not show signs of methanol or the type of hydrogenated COMs commonly seen toward hot cores(Olofsson 2005). Interestingly enough, SO and SO are common toward O-rich AGB envelopes, withthese molecules sometimes located very near the degenerate core (Danilovich et al. 2016). Formationof sulfur oxides may proceed through gas reactions involving OH such as S + OH → SO + H andSO + OH ↔ SO + H. Consistently, large quantities of OH toward G345.49+1.47 are attested by thepresence of OH masers (Caswell 1998). Detection of the rotating core in molecules beside the sulfuroxides will help supporting or rejecting these speculations. Other sulfuretted molecules like HCS + ,OCS, and C S show no rotation and are not even clearly associated with the CC core because theiremission peak farther in the northwest direction and with a different V LSR .As mentioned in Section 3.2, recent high spatial resolution continuum and methyl cyanide obser-vations of the central source in IRAS 16562 − . (cid:48)(cid:48) CN emission toward the center of IRAS 16562 − OH column densities to normalize theabundances. The first caveat is the use of a filling factor of 1 for the LTE models. While this iscertainly a simplification, we think it is in part justified because CH OH emission is rarely point-like, being commonly associated with extended envelope material which covers the beam. It ispossible, however, that lines with the highest upper energy levels are tracing a more compact andunresolved component. The second caveat is that, at least for the CC core, the SET model does notaccount for the observed intensities of the two lines with upper energy levels E up = 340 . . OH intensities using two LTE models allows us to better fit the lineswith temperatures of 90 and 350 K, and a ∼
50% higher total methanol column density. However,this fit is very uncertain because we do not detect enough lines to adequately constrain all the freeparameters. Hence, we refrain from complicating the model and we use the values given in Table4 on the understanding that these correspond to average values. Note that toward the CC corethe peak position of the CH OH lines varies within 0 . (cid:48)(cid:48)
4, with the lower energy lines peaking fartherfrom G345.49+1.47 compared to the higher energy transitions, consistent with the hottest materialbeing closer to the HMYSO. A similar thermal gradient is observed toward C8, with >
50 K energytransitions peaking closer to Source 8 compared to lower energy lines, which peak 0 . (cid:48)(cid:48) Extended emission in IRAS 16562 − In Section 3.3 we performed a systematic analysis of the extended emission in the clump. Our maininterest was to determine morphological similarities between the emission of different molecules. We
RAS 16562 − The Shock group.
Our results indicate that species in the Shock group giving rise to the extendedemission can be related with shocks in two ways: (i) as directly sputtered from the grain mantle orthe refractory core, or (ii) as being formed in recently shocked gas.Perhaps the most common and specific shock tracer molecule is SiO, which is assumed to beproduced by sputtering of Si/SiO from dust grains due to shocks (Jim´enez-Serra et al. 2008) usuallyattributed to YSO outflows. Significant amounts of SiO exists in IRAS 16562 − − ◦ from the approximate location of Source6. However, in general it was not possible to identify collimated molecular outflows unambiguously.Part of the reason is likely confusion: less massive embedded clusters are observed to be associatedwith a plethora of outflows (e.g. NGC 1333, see Plunkett et al. 2013) and we expect this to be thecase in high-mass protostellar clumps as well.The remarkable correlation between the SO extended emission (that is, away from G345.49+1.47)and SiO also indicates the association of SO with shocks. This association has been observed previ-ously (e.g., Jim´enez-Serra et al. 2005; Podio et al. 2015). The SO zero moment map illustrates clearlythe double origin of SO: on the one hand, associated with hot molecular gas near the HMYSO andno SiO emission, and on the other, extended emission tracing the shocked gas. The origin of sul-furetted species can be explained (at least in part) by gas reactions facilitated by the special physicalconditions associated with shocked gas. In particular, sequential reactions of S with OH and O willenhance the formation of sulfur oxides. In IRAS 16562 − whatsoever. This is consistent with models of low velocity shocks which predict largeamounts of SO and little SO (Pineau des Forets et al. 1993). Shocks and sputtering are ways ofreleasing molecules from the dust grains to the gaseous phase. While other mechanisms like photo-and chemical desorption are likely to be at work in IRAS 16562 − (cid:38)
15 km s − (Suutarinen et al. 2014).The production of secondary UV photons associated with the shocks can also increase the amountof S + and accelerate the rate of ion neutral reactions like S + + CH → HCS + + H (Yamamoto 2017).HCS + may later form CS through dissociative recombination. The good correlation between C S andHCS + indicates that this route is maybe important in the formation of these two molecules, althoughthere are also several other relevant formation routes for CS, for example, from SO + C → CS + O(Pineau des Forets et al. 1993). The latter reaction path is consistent with the observed correlationbetween SO and CS. Additionally, ion-neutral formation routes for HCS + and CS are also efficient,which helps explaining the relatively good correlation of the former with CCH. In fact, HCS + issomewhat in a middle ground between both the Shock and the Continuum group.6 Guzm´an et al.
Molecules like CH OH, OCS, CH CHO, H CCO, and HNCO show very similar spatial features.Because methanol is the archetypal dust mantle molecule, we interpret the morphological similaritiesas these species being formed in grain mantles and subsequently removed from the dust, probablyby ice sputtering. CH OH, CH CHO, and H CCO are part of two hydrogenation paths which wereactive when the clump’s temperature was lower, as it is now too high for H to remain long in the dustgrain surface. One of these paths starts with CO, leading to H CO and finally CH OH. The otherincludes an additional carbon atom addition and progresses through H CCO, CH CHO, and finallyCH CH OH (ethanol) (Charnley 1999). The latter path is probably less common than the first one,judging from the usually lower amounts of ethanol compared to methanol. In IRAS 16562 − > × cm − at 100 K).Because hydrogenation in the dust mantles drops once the grain reaches T d (cid:38)
20 K due to fastH evaporation, some of the intermediate unsaturated products like H CCO and CH CHO remainin the ices and they are prone to be removed afterwards by shocks (or other forms of ice erosion).This interpretation for the origin of CH OH, CH CHO, and H CCO is supported by, for example,the detection and correlation of these species toward embedded low-mass protostars (Bergner et al.2017) and in shocks associated with low-mass star formation outflows in L1157-B1 (Lefloch et al.2017).Despite the overall good correlation of H CCO and CH CHO in the diffuse gas, contrary to H CCO,CH CHO is not detected toward either the CC core or C8. Consistently, ¨Oberg et al. (2014) reportsacetaldehyde depletion with increasing temperature in HMCs. A possible explanation for the lack ofCH CHO in the HMCs of IRAS 16562 − CHO,possibly back into H CCO. H CCO and CH CHO have both been detected toward relatively coldgas ( <
50 K, Bisschop et al. 2007) and hot HMCs at temperatures above ≥
100 K (Belloche et al.2013).The good correlation observed between the OCS and methanol zero moment maps can be explainedby most of the OCS coming from dust mantles as well. This picture is confirmed by OCS being oneof the few molecules which has been detected directly in solid state in the ISM (Gibb et al. 2004).Note that this proposed formation mechanisms opposes to that of other sulfuretted species, such asthe sulfur oxides, CS, and HCS + , at least in the diffuse clump gas. There is no difference between theCC core and C8 in their [OCS]/[CH OH] abundance ratios, consistent with most of the OCS actuallycodesorbing (thermally and non-thermally) with CH OH to the gas phase. Isocyanic acid has alsobeen identified as a (slow) shock tracer at galactic scales (Watanabe et al. 2016; Ueda et al. 2017;Kelly et al. 2017) and in combination with CH OH. The HNCO abundance is seemingly enhancedin shocks due to ice erosion and direct formation in the heated post-shock gas (Rodr´ıguez-Fern´andezet al. 2010).
The Continuum group.
Emission from molecules in this group display good correlation with the 3mm continuum map. Extended emission from this map is thought to be dominated by thermal dustemission, with free-free being relevant toward specific sources like 10 and 18 (Paper I). While theamounts of all species are expected to increase with increasing material, some molecules (like thoseof the Shock group) increment their abundances in a higher proportion due to specific physical phe-nomena. The abundance of molecules in the Continuum group, in contrast, seems to be more stable,
RAS 16562 − + isotopologues.Gas phase formation routes dominated by ion-neutral and barrierless reactions characterize theformation of hydrocarbons such as CCH, c-C H , and CH CCH. The three molecules display a goodcorrelation between each other as expected for a common formation pathway. However, reactionsinvolving these molecules are only a small part of the complex chemical networks which character-ize hydrocarbon formation. This complicates reaching solid conclusions about their chemistry. Ingeneral, there are three main formation paths for hydrocarbons and carbon chains: (i) ion-neutralreactions, which dominate at high densities and are expected to correlate with the column density,(ii) warm carbon chain chemistry (WCCC), which could dominate at high temperatures, and (iii)top-down chemistry, which is efficient in UV-illuminated, diffuse regions.From the correlations observed involving CCH, c-C H , and CH CCH, we can conclude thattheir production has not been likely affected by WCCC processes. The defining characteristic ofWCCC is the expansion of the hydrocarbon’s chemical networks allowed by the injection of methane(CH ) into the gas phase from dust mantles (Sakai & Yamamoto 2013). In IRAS 16562 − − − CN, HN C, andHC N), and the hydrocarbons indicate that these species likely share common formation (or de-struction) paths. The most natural molecule acting as the bridge between all these species is CN,which can react with several hydrocarbons to form larger N-bearing molecules. We already describedformation paths for HC N and HC N involving the combination of, respectively, acetylene and di-acetylene with CN. These neutral-neutral reactions are barrier-less, therefore, they are also importantin the clump away from heating sources. As remarked in the previous subsection, CH C N can beformed from CH CCH through the addition of CN. Consistently, CH CCH is the MCP of CH C N.Formation of HCN, on the other hand, naturally starts with CN which reacts with H +3 (ion-neutralreaction, presumably close to the Langevin rate) to create HCNH + , which will form HCN by dis-sociative recombination (e.g., Prasad & Huntress 1980). Of course, H +3 is not a hydrocarbon butit plays a crucial role in starting the hydrocarbon reaction chains. In the same vein, C H CN canbe formed in the gas phase through a CN + ’hydrocarbon’ reaction (specifically, ethylene Herbst &8
Guzm´an et al.
Leung 1990). We caution, however, that C H CN is only well detected near C8 (see Figure 5), hence,it is unclear how well is C H CN related with the rest of the molecules in the Continuum group.Summarizing, molecules in the Continuum group are hydrocarbons, cyanides, cyanopolyynes, andHCO + . All these species have efficient gas phase formation routes which are chemically connectedthrough key molecules like H +3 and CN. Hydrocarbons have relevant formation routes starting from Cand H +3 , the latter being important as well in the creation of HCO + and HCN. In addition, a fraction ofthe hydrocarbons combines with CN producing the cyanopolyynes, CH C N, and possibly C H CN.Gas phase chemistry is efficacious in producing all these species even at low temperatures, whichhelps explaining why their abundances are relatively unaffected by special circumstances like shocksor photo illumination. In fact, emission from species in this group correlates well with the total masscolumn density as traced by the optically thin dust continuum.6.2.3.
Cold, starless gas
One remarkable feature found in IRAS 16562 − ii region — is the presence of cold and dense gas clearly differentiated spatially andchemically from the rest of the molecular emission. There are two starless clouds seen only in NH Dlocated in the northeast and southeast regions of the clump. We refer to these two clouds as the NEand SE clouds, respectively. As noted in Section 6.1, the SE cloud is clearly evident as an IR-darkfeature against the brightly illuminated outflow cavity associated with G345.49+1.47.Emission from NH D is most readily explained by a high deuterated ammonia fractionation. Deu-terium fractionation is expected in cold and dense molecular environments for all species whoseformation path involves H +3 (Bergin & Tafalla 2007). Gas phase reactions which produce ammonia inthis way also include N H + (N D + for NH D) in the formation route (Yamamoto 2017). Therefore,because CO readily destroys N H + , this chemical path is most effective combined with strong COdepletion (Roueff et al. 2005). Confirming large fractions of NH D and of N D + respect to theirmain isotopologues, and also measuring CO depletion factors, will help solidifying these theoreticalexpectations. Alternatively, deuterium enrichment may occur directly in the dust grains (Fedoseevet al. 2015) but this requires an efficient non-thermal desorption mechanism to release deuteratedammonia into the gas phase.Determining whether these starless clouds — namely, without evidence of embedded YSOs in anyof our observations — could collapse and form stars in the future entails deriving their masses. Wederive a NH D mean column density of 3 . . × cm − toward the SE and NE clouds,respectively, assuming an excitation temperature of 15 K. Assuming a mean deuterated ammoniafractionation of 0.02 (Lackington et al. 2016) characteristic of IRDCs, an ammonia abundance respectto H ranging between 5–30 × − (Wienen et al. 2012), and a size given by the 30% of the peakcontour of the NH D zero moment map in Figure 15, we derive mass ranges for the SE and NEclouds of 0.25–1.5 and 0.4–2.3 M (cid:12) , respectively. Column densities of H range between 10 cm − .Whereas these column densities are large, the cloud masses are too low to virialize these clouds ofradius ∼ .
02 pc and ∆ V ≥ − . We conclude that the clouds are the remnants of the coldprestellar clump which gave birth to IRAS 16562 − SUMMARY
RAS 16562 − − σ level withconsistent V LSR .2. We derive physical parameters — column densities and excitation temperatures — based onthe SET modeling of the spectra associated with the most prominent features in the clump.We use the Radex non-LTE model to fit the methanol emission. Typical temperatures for theprotostellar cores are 70–120 K. Most of the extended emission is characterized by temperatures20–40 K, depending on the distance to the central, dominating HMYSO G345.49+1.47.3. The morphological characteristics of the extended emission allow us to collect the molecules intwo groups. The Shock group gathers molecules whose emission is more similar to that of SiO, J = 2 (cid:1)
1. It collects molecules formed in shocked gas (like SO) and in dust grains (like CH OH,H CCO, and CH CHO) later released into the gas phase. The Continuum group collectsmolecules whose emission morphology is more similar to that of the 3 mm dust continuum andseem to depend first on the column density. This group includes hydrocarbons, cyanopolyynes,and other cyanides (HCN and CH C N) which have effective gas phase formation routes. Thereis no clear evidence of efficient WCCC processes in forming hydrocarbons.4. The HMYSO G345.49+1.47 is associated with two different structures: a HMC conspicuousin hydrogen saturated COMs and a previously detected rotating core, conspicuous in sulfuroxides. This rotating core is not detected in any other species.5. Source 8 is associated with a second HMC within IRAS 16562 − CHCN/CH CH CN andcyanopolyyne abundances — suggest that the core associated with Source 8 is more evolvedthan the one associated with G345.49+1.47.6. We detect NH D emission arising from cold, IR-dark, starless clouds of ∼ M (cid:12) . These cloudsare chemically and spatially differentiated from the rest of the gas in the clump. They areremnants of the prestellar stage of the clump, and they will not likely sustain further starformation activity.7. We observe a strong spatial segregation in IRAS 16562 − D, Shock, andContinuum group molecules. This segregation illustrates the need to separate the physicalorigin of different (groups of) molecules when modeling the chemistry of high-mass clumpsusing unresolved observations from, for example, single dish telescopes.A.E.G. thanks support from Fondecyt 3150570. V.V.G. acknowledges support from the NationalAeronautics and Space Administration under grant No. 15XRP15 20140 issued through the Exo-planets Research Program. G.G. and L.B. acknowledge support from CONICYT project PFB-06.This paper makes use of the following ALMA data: ADS/JAO.ALMA
Guzm´an et al. with NRC (Canada), NSC and ASIAA (Taiwan), and KASI (Republic of Korea), in cooperationwith the Republic of Chile. The Joint ALMA Observatory is operated by ESO, AUI/NRAO andNAOJ. Portions of the analysis presented here made use of the Perl Data Language (PDL) devel-oped by K. Glazebrook, J. Brinchmann, J. Cerney, C. DeForest, D. Hunt, T. Jenness, T. Lukka,R. Schwebel, and C. Soeller. Analysis of this paper made use of CASSIS, developed by IRAP-UPS/CNRS (http://cassis.irap.omp.eu).
Facility:
Atacama Large Millimeter/submillimeter Array (ALMA).
Software:
CASA v4.7.2 (Petry & CASA Development Team 2012). Radex (van der Tak et al. 2007).CASSIS (Caux et al. 2011). Perl Data Language (http://pdl.perl.org).
Table 1 . Observed frequencies and upper energies from identified lines E up Frequency (GHz) (K)Carbon monosulfide1 CS J = 2 → S J = 2 → J = 7 → J = 8 → CS J = 7 → J = 8 → + J = 2 → J, N = 2 , → , J, N = 3 , → , J, N = 4 , → , SO J, N = 4 , → , Table 1 continued on next page
RAS 16562 − Table 1 (continued) E up Frequency (GHz) (K)
J, N = 3 , → , SO J, N, F = 3 , , → , , J K a K c = 8 , → , J K a K c = 20 , → , J K a K c = 7 , → , J K a K c = 28 , → , J K a K c = 29 , → , D J K a K c = 1 , , s → , , a J = 2 → SiO J = 2 → SiO J = 2 → b CCH
J, K = 5 , → , J, K = 5 , → , J, K = 5 , → , b J, K = 5 , → , b CO + J = 1 → b O + J = 1 → Table 1 continued on next page Guzm´an et al.
Table 1 (continued) E up Frequency (GHz) (K)16 c-C H J K a K c = 2 , → , J K a K c = 4 , → , J, N, F = , , → , , J, N, F = , , → , , J, N, F = , , → , , J, N, F = , , → , , J, N, F = , , → , , J, N, F = , , → , , CHO (E) J K a K c = 5 , → , J K a K c = 5 − , → − , CH CN J K a K c = 10 , → , J K a K c = 11 , → , J K a K c = 11 , → , J K a K c = 11 , → , b J K a K c = 11 , → , b J K a K c = 11 , → , b J K a K c = 11 , → , b J K a K c = 11 , → , b J K a K c = 11 , → , b J K a K c = 11 , → , b J K a K c = 11 , → , b J K a K c = 11 , → , b J K a K c = 11 , → , b Table 1 continued on next page
RAS 16562 − Table 1 (continued) E up Frequency (GHz) (K) J K a K c = 11 , → , b J K a K c = 11 , → , b J K a K c = 11 , → , J K a K c = 11 , → , J K a K c = 11 , → , C N J, K = 21 , → , b J, K = 24 , → , J, K = 24 , → , J, K = 24 , → , b J, K = 24 , → , b N J = 32 →
31 85.2057 67.5 J = 33 →
32 87.8675 71.7 J = 39 →
38 98.5172 89.8Cyanoacetylene (propynenitrile)22 HC N J = 11 →
10 100.0815 28.823 HCC CN J = 11 →
10 99.6665 b CCN J = 11 →
10 99.6568 28.725 H CCCN J = 11 →
10 96.9873 b N J = 1 → CN J = 1 → C J = 1 → Table 1 continued on next page Guzm´an et al.
Table 1 (continued) E up Frequency (GHz) (K)29 HNCO J K a K c = 4 , → , J K a K c = 4 , → , b J K a K c = 4 , → , b J K a K c = 4 , → , OH (E ) J K a K c = 5 − , → , , v t = 0 84.5246 32.5(E ) J K a K c = 19 , → , , v t = 0 84.7486 b ) J K a K c = 6 − , → − , , v t = 0 85.5725 66.8(E ) J K a K c = 2 − , → − , , v t = 0 96.7442 4.6(E ) J K a K c = 2 , → , , v t = 0 96.7495 12.2(E ) J K a K c = 2 , → , , v t = 0 96.7604 28.0(E ) J K a K c = 6 , → , , v t = 1 99.7363 340.1CH OH (A) (A − ) J K a K c P = 7 − , → − , , v t = 0 86.6201 102.7(A + ) J K a K c P = 7 +2 , → +3 , , v t = 0 86.9074 102.7(A + ) J K a K c P = 2 +0 , → +0 , , v t = 0 96.7463 7.0(A − ) J K a K c P = 2 − , → − , , v t = 0 97.5878 21.631 CH OH (A + ) J K a K c P = 8 +0 , → +1 , , v t = 0 84.9745 81.5Ethenone (ketene)32 H CCO J K a K c = 5 , → , OCH J K a K c = 5 , → , (EE) 96.8554 19.3 J K a K c = 16 , → , (EE) 97.9983 136.6 J K a K c = 4 , → , † CHCN J K a K c = 9 , → , Table 1 continued on next page
RAS 16562 − Table 2.
Isotopic and Isomeric Ratios
Species Observed Solar ISM at R Gal = 6 . / [ SiO] 15 . ± . . . ± . SiO] / [ SiO] 1 . ± . . . ± . / [ SO] 22 . a . . ± . ] / [ SO ] 22 . a . . ± . SO] / [ SO] 5 . ± . . . ± / [O CS] 25 ± . . ± OH] / [ CH OH] 26 ± . ± N] / [HC CCN] 32 ± . ± CO + ] / [HC O + ] 6 . ± . . . ± . CN] / [HC N] 5 . ± . . ± . CN] / [HN C] 1 . ± . · · · References —Solar abundances from Asplund et al. (2009)Silicon fractionation from Monson et al. (2017). Sulfur andcarbon’s from Wilson (1999). Nitrogen fractionation fromAdande & Ziurys (2012). a Assumed in Paper I and found consistent with the data.
Table 1 (continued) E up Frequency (GHz) (K) J K a K c = 9 , → , J K a K c = 9 , → , b J K a K c = 9 , → , b J K a K c = 9 , → , J K a K c = 9 , → , J K a K c = 9 , → , J K a K c = 10 , → , b J K a K c = 3 , → , b bn Blending between the lines marked with the same bn , where n is a number. † Blending of the four (AA, EE, EA, and AE) transitions. b † Possible blending with CH OCHO. Guzm´an et al.
Table 3.
Parameters of CH OH, 5 − , (cid:1) , Maser Spots
Maser R.A. Decl. Flux density a V LSR
FWHM(J2000) (J2000) (Jy) (km s − ) (km s − )16:59:. . . − a . ± .
001 36 . ± .
003 8 . ± . − . ± .
04 3 . ± . b . ± .
001 45 . ± .
005 8 . ± . − . ± .
04 2 . ± . c . ± .
006 39 . ± .
05 4 . ± . − . ± .
01 3 . ± . a Flux density values are primary beam corrected.
Table 4.
Methanol (non-)LTE Model ParametersSource T Column V LSR ∆ V Density Comment(K) log (cid:0) N cm − (cid:1) (km s − ) (km s − ) log (cid:0) n cm − (cid:1) CC core 107 ±
10 16 . ± . − . ± .
05 4 . ± . ≥
10 +Hotter component.N-red cloud 28 ± . ± . − . ± . . ± . . ± . ±
10 15 . ± . − . ± .
04 3 . ± . . ± . ) J K a K c = 5 − , (cid:1) , .NEC-wall (b) 56 ± . ± . − . ± .
06 3 . ± . . . ) J K a K c = 5 − , (cid:1) , .Diffuse Ridge (a) 33 ± . ± . − . ± . . ± . . ± . ± . ± . − . . . ± . ±
15 15 . ± . − . ± . . ± . . ± . ) J K a K c = 5 − , (cid:1) , .C8 maser 20 ± . ± . − . ± .
05 2 . ± . . . ) J K a K c = 5 − , (cid:1) , .C8 1 . (cid:48)(cid:48) ±
30 15 . ± . − . ± . . ± . . . E up ≤
40 K lines.NW cloud 20 ±
10 15 . ± . − . ± . . ± . . ± . ) J K a K c = 5 − , (cid:1) , .NW cloud (a) 50 ±
10 15 . ± . − . ± . . ± . . ± . ±
30 16 . ± . − . ± . . ± . . . ) J K a K c = 5 − , (cid:1) , .Source 18 64 ± . ± . − . ± .
06 4 . ± . ≥
10 Absorption in E up ≤
40 K lines.
RAS 16562 − Table 5 . Physical Parameters of SET Emission Models (cid:63)
Molecule T a Column V LSR ∆ V Vel. CommentDensity Cmp.(K) log (cid:0) N cm − (cid:1) (km s − ) (km s − ) CC core NH D 108 ±
10 14 . ± . − . ± . . ± . OCH ±
20 15 . ± . − . ± . . ± . CCO 107 † . ± . − . ± . . ± N 107 † . ± . − . ± .
07 4 . ± . CCN 107 † . ± . − . ± . . ± . OH 107 ±
10 16 . ± . − . ± .
05 4 . ± . † , +hotter component. CH OH 107 † . ± . − .
33 4 .
65 1 V LSR and ∆ V fromCH OH.CH CH CN 123 ± . ± . − . ± .
04 4 . ± .
09 1H CN 64 ‡ . ± . − . ± . . ± . . ± . ± . − . ± .
09 6 . ± . N 64 ‡ . ± . − . ± .
06 5 . ± . ‡ . ± . − . ± .
05 5 . ± . CS 64 ‡ . ± . − . ± . . ± . CCH 64 . ± . ± . − . ± .
06 4 . ± . ‡ SO 64 ‡ . ± . − . ± . . CO + ‡ . ± . − . ± . . ± . CCH T ex from group 2.HC O + ‡ . ± . − . ± . . ± CCH T ex from group 2.C S 64 ‡ . ± . − . ± . . ± CCH T ex from group 2.HCS + ‡ . ± . − . ± . . ± CCH T ex from group 2.CCH 64 ‡ . ± . − . ± . . ± . CCH T ex from group 2. N-red cloud CH OH 28 ± . ± . − . ± . . ± . † , blueshift absorption.OCS 28 † . ± . − . ± . . ± . . ± . ± . − . ± .
07 5 . ± . ‡ , subthermal. SO 8 . ‡ . ± . − . ± . . ± . . ‡ . ± . − . ± . . ± . SiO 8 . ‡ . ± . − . ± . . ± . Table 5 continued on next page Guzm´an et al.
Table 5 (continued)
Molecule T a Column V LSR ∆ V Vel. CommentDensity Cmp.(K) log (cid:0) N cm − (cid:1) (km s − ) (km s − ) SiO 8 . ‡ . ± . − . ± . . ± . CHO 28 † . ± . − . ± . . ± . N 8 . ‡ . ± . − . ± . . ± . . ‡ . ± . − . ± . . ± . CCO 28 † . ± . − . ± . . ± † . ± . − . ± . . ± . S 28 † . ± . − .
93 5 . N 8 . ‡ . ± . − . ± . . ± . NEC-wall
NEC-wall (a): Methanol Peak b CS 57 † . ± . − .
13 3 . † . ± . − .
13 3 . OH 57 ±
10 15 . ± . − . ± .
04 3 . ± . † , exclude (E ) J K a K c = 5 − , (cid:1) , .CH CHO 57 † . ± . − . ± . . ± . CH OH 57 † . ± . − . ± . . ± . † . ± . − . ± .
02 3 . ± . N 57 † . ± . − . ± . . ± . CN 57 † . ± . − . ± .
06 5 . ± . CCN 42 ‡ . ± . − . ± . . ± . C 42 ‡ . ± . − . ± .
08 4 . ± . CO + ‡ . ± . − . ± .
07 4 . ± . O + ‡ . ± . − . ± . . ± . CCH 42 . ± . ± . − . ± .
07 5 . ± . ‡ HC N 42 ‡ . ± . − . ± .
08 5 . ± . D 42 ‡ . ± . − . ± . . ± OH 56 ± . ± . − . ± .
06 3 . ± . ‡ , exclude (E ) J K a K c = 5 − , (cid:1) , . Table 5 continued on next page
RAS 16562 − Table 5 (continued)
Molecule T a Column V LSR ∆ V Vel. CommentDensity Cmp.(K) log (cid:0) N cm − (cid:1) (km s − ) (km s − )CS 56 ‡ . ± . − . ± . . ± . S 56 ‡ . ± . − . ± . . ± . ‡ . ± . − . ± .
08 3 . ± . + ‡ . ± . − . ± . . ± . ‡ . ± . − . ± .
06 5 . ± . CO + † . ± . − . ± .
07 4 . ± . CCH 43 . ± . ± . − . ± .
07 4 . ± . ±
40 15 . ± . − . ± .
08 5 . ± . † , poorly constrainedtemperature. SO 80 † . ± . − . ± . . ± . N 32 . ± . ± . − . ± . . ± . N 80 † . ± . − . ± .
04 4 . ± . CCN 80 † . ± . − . ± . . ± . N 80 † . ± . − . ± .
07 4 . ± . CN 80 † . ± . − . ± .
06 4 . ± . C 80 † . ± . − . ± . . ± . † . ± . − . ± . . ± . Diffuse Ridge
DR(a) b CS 33 † . ± . − . ± .
04 2 . ± .
08 Optically thick. Radiativetrapping.HC N 6 ‡ . ± . − . ± . . ± . S 6 ‡ . ± . − . ± . . ± . OH 33 ± . ± . − . ± . . ± . † HNCO 6 ‡ . ± . − . ± . . ± . † . ± . − . ± . . ± . CN 6 ‡ . ± . − . ± . . ± . CHO 33 † . ± . − . ± . . ± . CCO 6 ‡ . ± . − . ± . . ± . CO + ‡ . ± . − . ± .
08 2 . ± . Table 5 continued on next page Guzm´an et al.
Table 5 (continued)
Molecule T a Column V LSR ∆ V Vel. CommentDensity Cmp.(K) log (cid:0) N cm − (cid:1) (km s − ) (km s − )HN C 6 ‡ . ± . − . ± .
09 2 . ± . ‡ . ± . − . ± . . ± . O + ‡ . ± . − . ± . . ± . . ± . . ± . − . ± .
09 2 . ± . ‡ HCS + . †† . ± . − . ± . . ± . CCH 36 . ± . ± . − . ± .
08 3 . ± . †† CCH 6 ‡ . ± . − . ± . . ± . OH 38 ± . ± . − . . CCH 37 . ± . ± . − . ± . . ± . † H CO + † . ± . − . ± .
05 3 . ± . O + † . ± . − . ± . . ± . ‡ . ± . − . ± .
05 3 . ± . T ex from group 2.CH CHO 38 † . ± . − . ± . . ± . N 11 ‡ . ± . − . ± .
03 2 . ± .
07 1 SO T ex from group 2.HN C 11 ‡ . ± . − . ± . . ± . T ex from group 2.HC N 11 ‡ . ± . − . ± . . ± . CN 11 ‡ . ± . − . ± . . ± . ‡ . ± . − . ± . . ± . . ± . ± . − . ± . . ± . ‡ SO 11 ‡ . ± . − . ± . . ± † . ± . − . ± . . ± . S 11 ‡ . ± . − . ± . . ± . + † . ± . − . ± . . ± . CCH T ex from group1.SiO 11 ‡ . ± . − . ± . . ± . T ex from group 2.DR(c) b CS 7 † . ± . − . ± . . ± . S 7 † . ± . − .
88 4 . ± CN 7 † . ± . − . ± . . ± . Table 5 continued on next page
RAS 16562 − Table 5 (continued)
Molecule T a Column V LSR ∆ V Vel. CommentDensity Cmp.(K) log (cid:0) N cm − (cid:1) (km s − ) (km s − )SiO 7 † . ± . − . ± . . ± . N 7 † . ± . − . ± . . ± . . ± . . ± . − . ± .
08 5 . ± . † OCS 19 . ± . ± . − . ± . . ± . OH 25 ±
15 15 . ± . − . ± . . ± . ‡ , exclude (E ) J K a K c = 5 − , (cid:1) , .HNCO 7 † . ± . − . ± . . ± . N 7 † . ± . − . ± . . ± . CHO 25 ‡ . ± . − . ± . . ± . C 7 † . ± . − . ± . . ± . C8 C8 maserCS 17 † . ± . − . ± . . ± † . ± . − . ± . . ± SiO 17 † . ± . − . ± . . ± . SiO 17 † . ± . − . ± . . ± . S 17 † . ± . − . ± . . ± . OH 20 ± . ± . − . ± .
05 2 . ± . †† , exclude (E ) J K a K c = 5 − , (cid:1) , .c-C H . ± . ± . − . ± .
07 3 . ± . CN 17 † . ± . − . ± . . ± . C 17 † . ± . − . ± .
07 3 . ± . + †† . ± . − . ± . . ± . † . ± . − . ± .
03 3 . ± . . ± . ± . − . ± . . ± . † CH CHO 17 † . ± . − . ± . . ± . O + ‡ . ± . − . ± . . ± . N 28 . ± . ± . − . ± . . ± . N 17 † . ± . − . ± . . ± . ‡ . ± . − . ± . . ± . Table 5 continued on next page Guzm´an et al.
Table 5 (continued)
Molecule T a Column V LSR ∆ V Vel. CommentDensity Cmp.(K) log (cid:0) N cm − (cid:1) (km s − ) (km s − )H CO + ‡ . ± . − . ± .
06 3 . ± . CCN 17 † . ± . − . ± . . ± . CCH 40 . ± . ± . − . ± .
03 3 . ± . ‡ HNCO 17 † . ± . − . ± . . ± . N 17 † . ± . − . ± .
06 2 . ± . . (cid:48)(cid:48) b SiO 17 † . ± . − . ± .
08 5 . ± . . ± . ± . − . ± .
07 3 . ± . † HN C 17 † . ± . − . ± .
06 3 . ± . ±
70 14 . ± . − . ± . . ± † . ± . − . ± .
03 3 . ± . OH 114 . ±
30 15 . ± . − . ± . . ± . E up transitions.c-C H . ± . ± . − . ± .
03 4 . ± . CO + ‡ . ± . − . ± .
04 4 . ± . C N 48 . ± . ± . − . ± . . ± . ‡ . ± . − . ± .
07 4 . ± . CCH 59 . ± . ± . − . ± .
02 3 . ± . ‡ HC O + ‡ . ± . − . ± . . ± . CH CN 74 . ±
20 13 . ± . − . ± . . ± . N 17 † . ± . − . ± .
09 4 . ± . N 77 . ±
10 13 . ± . − . ± .
05 3 . ± . S 17 † . ± . − . ± . . ± . CN 17 † . ± . − . ± .
07 4 . ± . CHCN 81 . ±
20 13 . ± . − . ± . . ± . CCCN blend at 99986MHz.HC CCN 17 † . ± . − . ± .
09 3 . ± . OCH † . ± . − . ± . . ± † . ± . − . ± .
04 2 . ± . CCO 17 † . ± . − . ± .
08 3 . ± . N 17 † . ± . − . ± .
05 3 . ± . Table 5 continued on next page
RAS 16562 − Table 5 (continued)
Molecule T a Column V LSR ∆ V Vel. CommentDensity Cmp.(K) log (cid:0) N cm − (cid:1) (km s − ) (km s − ) NW cloud H CCO 20 †† . ± . − . ± . . ± . OH T ex from group 1.HC N 16 † . ± . − . ± .
08 5 . ± . T ex from group 1.CH OH 20 ±
10 15 . ± . − . ± . . ± . †† CH CCH 44 . ±
10 14 . ± . − . ± . . ± . ‡ SO 16 † . ± . − . ± . . ± . SO 16 † . ± . − . ± . . ± . . ± . ± . − . ± .
07 4 . ± . † , possibly subthermalH CO + ‡ . ± . − . ± .
04 2 . ± .
09 1HC O + ‡ . ± . − . ± . . ± . † . ± . − . ± .
05 2 . ± . CHO 16 † . ± . − . ± . . ± . N 16 † . ± . − . ± . . ± . CN 16 † . ± . − . ± . . ± . C 16 † . ± . − . ± .
04 3 . ± .
09 1HNCO 16 † . ± . − . ± .
06 4 . ± . † . ± . − . ± .
08 6 . ± . T ex from group 1.SiO 16 † . ± . − . ± . . ± . T ex from group 1,blueshift absorption. SiO 16 † . ± . − . ± . . ± . T ex from group 1,blueshift absorption. SiO 16 † . ± . − . ± . . ± T ex from group 1.NW cloud (a)CH OH 50 ±
10 15 . ± . − . ± . . ± . † OCS 16 14 . ± . − . ± . . ± . T ex from NW cloud.CH CHO 50 † . ± . − . ± . . ± . . ± . − . ± . . ± . T ex from NW cloud.SiO 16 13 . ± . − . ± . . ± . T ex from NW cloud. SiO 16 12 . ± . − . .
96 2 T ex from NW cloud. Faint. Source 3
Table 5 continued on next page Guzm´an et al.
Table 5 (continued)
Molecule T a Column V LSR ∆ V Vel. CommentDensity Cmp.(K) log (cid:0) N cm − (cid:1) (km s − ) (km s − )CH OH 221 ±
30 16 . ± . − . ± . . ± . ) J K a K c = 5 − , (cid:1) , profile. Source 18 CH OH 64 ± . ± . − . ± .
06 4 . ± . E up transitions. (cid:63) Except for the CH OH non-LTE modeling described by the parameter in Table 4.aFor each source, temperatures marked with † , ‡ , and †† are taken from the molecule with the same symbolin column (7).bThis source has a continuous distribution of velocities. No discrete V LSR components are identified.
RAS 16562 − A. SHORT SPATIAL FREQUENCIES FILTERINGThe data presented in this work was taken using the ALMA with only the 12 m antennas withbaselines between 453 and 21 m. This coverage does not guarantee good recovery of structures withangular scales larger than ∼ (cid:48)(cid:48) . The most noticeable effect produced by this lack of short spacingon the images are regions of negative emission. The filtering of emission on large scales has twoeffect: first, it filters out diffuse, extended components associated with compact sources, and second,it leaves local decrements in intensity below the zero level of the synthesized images. Thus, negativeemission is not necessarily consequence of negative sidelobes or spurious data: they could be merelythe reflection of diminished intensity compared with nearby positions, either because of absorption,or just because of the specific morphology of the source.In order to evaluate how the observations presented in this work recover the short spacing, wecompare the flux of the CS, J = 2 (cid:1) Swedish-ESO Submillimetre Telescope (SEST). This 15 m single dish antenna has a beam onlyslightly smaller than ALMA’s primary beam. The emission from CS toward IRAS 16562 − − σ . Thisconstant is the negative of the minimum intensity of each channel, ensuring that each channel cor-rected has only positive emission. This procedure can be thought as a crude way to correct the lackof zero spacing from the data cube. We integrate spatially this corrected CS cube using the SESTbeam. Figure A.1 show the SEST data and the corrected CS spectrum in blue and black lines, re-spectively. We regard the agreement as reasonable, considering the systematic uncertainties involvedin the single dish and interferometer flux calibrations of these two instruments. Our rough correctionprocedure appears to overestimate the emission in the channels with velocities < −
15 km s − . Thisis expected if, for example, the zero-space emission does not cover the entire field of view. On theother hand, at the peak of the line, our zero space correction accounts for about 60% of the singledish flux.Therefore, we conclude that most of the negative emission seen in the maps presented in this studyare produced by zero-space filtering. In a way, the interferometer has recovered spatial informationindicating us that in these negative regions there is a decrement of the emission compared to thesurrounding clump. Figure A.1 also shows in green and red the ALMA CS spectra of the uncorrecteddata cube and of the positive part of the emission, respectively. The uncorrected and positive partspectra reach only 8 and ∼
1% of the line peak flux. http://almascience.nrao.edu/about-alma/alma-basics Guzm´an et al.
Right Ascension (J2000) D e c li n a t i o n ( J ) Jy beam -1 km s -1 CS -0.86 -0.46 -0.061 0.34 0.74 1.1 1.5 1.9 2.4 2.8 3.2 Figure A.1.
Left panel: zero moment of the CS map. Dashed contours indicate the − .
03 Jy beam − km s − level. Right panel: in blue, we show the SEST CS, J = 2 (cid:1) − RAS 16562 − B. UNCERTAINTIES OF THE PROPYNE COLUMN DENSITY AND TEMPERATUREFITTING log (N CH3C2H /cm -2 ) (K) D e c li n a t i o n ( J ) Right Ascension (J2000) σ (N CH3CCH ) σ (T ex ) Figure B.1.
Left and right panels show the formal uncertainties in the CH CCH column density andexcitation temperature fitting, respectively (Section 4.2). The best-fit parameters are shown in Figure 12. Guzm´an et al. C. SPECTRA FROM CONSPICUOUS SOURCES (a)(b) (c) (d)
Figure C.1.
Emission lines from several molecules toward the CC core, in K (primary beam correctedantenna temperatures) vs. frequency (MHz). Black and red continuous lines show the data and the modelwhose parameters are given in Table 5. In each panel we mark the V LSR of the model with a vertical blackline, and the upper energy level of each transition. Thick green lines indicate the section of the spectraused to get the best-fit values. In panels where the peak intensity does not reach 5 σ , σ = 0 .
06 K we showthe ± . σ levels using dotted black lines. Panels (a) to (d) show CH OH, SO, CH OH, and HC N,respectively. The CH OH spectrum suggests the presence of a additional and hotter component.
RAS 16562 − (e) (f) (g) (h) Figure C.1. (cont.)
Same as the previous plot. Panels (e) to (h) show HC CCN, H CCO, HC N, andH CN, respectively. SO (i) Figure C.1. (cont.)
Same as the previous plot. Panels (i) show the C H CN lines. Strong lines from otherspecies within the displayed frequency window are marked in blue Guzm´an et al. (j) (k)(l) (m)(n)
Figure C.1. (cont.)
Same as the previous plot. Panels (j) to (n) show, respectively, CH CCH, OCS,O CS, NH D, and HNCO.
RAS 16562 − CH OH CSC H CN (o) Figure C.1. (cont.)
Same as the previous plot. Panel (o) shows the CH OCH spectra of the CC core.Strong lines from other species within the displayed frequency windows are marked in blue. (p) (q) (r) c-C H (s)(t) Figure C.1. (cont.)
Same as the previous plot. Panels (q) to (r) show, respectively, the H CO + , HC O + ,C S, HCS + , and CCH spectra of the CC core. A strong absorption feature due to c-C H within thedisplayed frequency window of the HCS + (1-0) line is marked in blue. Guzm´an et al. (a)(b) (c) (d)(e) (f)
Figure C.2.
Emission lines from several molecules toward the N-red cloud. Line types and colors asdescribed in Figure C.1, with σ = 0 .
02 K. Panels (a) to (f) show lines of CH OH, CS, C S, OCS, SO, and SO, respectively.
RAS 16562 − (g) (h) (i) (j)(k) (l) (m)(n) Figure C.2. (cont.)
Same as in Figure C.2. Panels (g) to (n) show lines of SiO, SiO, SiO, HC N,CH CHO, HC N, HNCO, and H CCO, respectively. Guzm´an et al. (a)(b) (c) (d)(e) (f) (g) (h)
Figure C.3.
Emission lines from several molecules toward the methanol peak of the NEC-wall. Line typesand colors as described in Figure C.1, with σ = 0 .
06 K. Panels (a) to (h) show lines of CH OH, CS, SO,NH D, CH CCH, H CO + , HC O + , and HC N, respectively. We note in (a) that the strong non-LTECH OH, (E ) J K a K c = 5 − , (cid:1) , emission cannot be reproduced with the models used in this work. RAS 16562 − (i)(j) (k) (l)(m) (n) (o) Figure C.3. (cont.)
Same as Figure C.3. Panels (i) to (o) show lines of CCH, CH CHO, HC CCN,HC N, H CN, and CH OH, respectively. Guzm´an et al. (a)(b) (c) (d)(e) (f)(g) (h) (i) (j)
Figure C.4.
Emission lines from several molecules toward the sulfur monoxide peak of the NEC-wall. Linetypes and colors as described in Figure C.1 with σ = 0 .
05 K. Panels (a) to (j) show lines of CH OH, CS,C S, OCS, SO, CH CCH, SiO, H CO + , HC N, and HC CCN, respectively. The strong non-LTE CH OH,(E ) J K a K c = 5 − , (cid:1) , emission shown in (a) cannot be reproduced by the models used in this work. RAS 16562 − (k)(l) (m)(n) (o) Figure C.4. (cont.)
Same as Figure C.4. Panels (k) to (n) show lines of CCH, HC N, HC N, H CN, andHN C, respectively. Guzm´an et al. (a) (b) (c)
Figure C.5.
Emission lines from several molecules toward the point (a) in the Diffuse ridge (DR (a)). Linetypes and colors as described in Figure C.1 with σ = 0 .
06 K. Panels (a) to (c) show lines of CS, C S, andOCS, respectively.
RAS 16562 − (d)(e) (f) (g)(h) (i) (j) (k)(l) (m) (n)(o) Figure C.5. (cont.)
Same as Figure C.5. Panels (d) to (o) show lines of CH OH, CH CHO, H CN,HC N, HNCO, H CCO, H CO + , SiO, HC O + , SO, HCS + , and CH CCH, respectively. Guzm´an et al. (p)
Figure C.5. (cont.)
Same as Figure C.5. Panel (p) shows the CCH lines toward DR (a). (a)(b) (c) (d)(e) (f) (g)
Figure C.6.
Emission lines from several molecules toward the point (b) in the Diffuse ridge (DR (b)).Line types and colors as described in Figure C.1 with σ = 0 .
06 K. Panels (a) to (g) show lines of CH OH,CH CCH, H CO + , HC O + , CH CHO, HC N, and HN C, respectively.
RAS 16562 − (h)(i) (j) (k) (l)(m) (n) (o)(p) (q) Figure C.6. (cont.)
Same as Figure C.6. Panels (h) to (q) show lines of CCH, HC N, H CN, HNCO, SO, SO, CS, C S, HCS + , and SiO, respectively. Guzm´an et al. (a) (b) (c) (d)(e) (f) (g)(h) (i) (j)(k)(l)
Figure C.7.
Emission lines from several molecules toward the point (c) in the Diffuse ridge (DR (c)). Linetypes and colors as described in Figure C.1 with σ = 0 .
05 K. Panels (a) to (l) show lines of CS, C S,H CN, SiO, HC N, SO, HNCO, OCS, HC N, HN C, CH OH, and CH CHO, respectively. We note thatthe strong non-LTE CH OH, (E ) J K a K c = 5 − , (cid:1) , emission shown in (k) is not well reproduced withthe models used in this work. RAS 16562 − (a) (b) (c) (d)(e) (f)(g)(h) (i) (j) (k) Figure C.8.
Emission lines from several molecules toward the point S8E maser. Line types and colors asdescribed in Figure C.1 with σ = 0 .
02 K. Panels (a) to (k) show lines of CS, SiO, SiO, SiO, C S, c-C H ,CH OH, H CN, HN C, HCS + , and HC O + , respectively. The CH OH, (E ) J K a K c = 5 − , (cid:1) , masertransition shown in (g) cannot be reproduced with the models used in this work. Guzm´an et al. (l)(m) (n)(o) (p)(q) (r)
Figure C.8. (cont.)
Same as Figure C.8. Panels (l) to (r) show lines of CCH, SO, HC N, CH OH, OCS,HC N, and H CO + , respectively. RAS 16562 − (s) (t)(u) (v) Figure C.8. (cont.)
Same as Figure C.8. Panels (s) to (v) show lines of HC CCN, CH CCH, HNCO, andHC N, respectively. (a) (b) (c)
Figure C.9.
Emission lines from several molecules toward the point S8E 1 . (cid:48)(cid:48) σ = 0 .
02 K. Panels (a) to (c) show lines of SiO, SO, and HN C, respectively. Guzm´an et al. (d)(e)
Figure C.9. (cont.)
Same as Figure C.9. Panels (d) and (e) show the CCH and CH OH lines, respectively.Note that the model does not reproduce well the strong absorption features associated with the lowest energyCH OH transitions and the CH OH, (E ) J K a K c = 5 − , (cid:1) , line. The latter is usually affected by strongnon-LTE effects. RAS 16562 − (f)(g) H CO + (h) (i) Figure C.9. (cont.)
Same as Figure C.9. Panels (f) and (i) show the lines of c-C H , H CO + , CH C N,and HC O + , respectively. A strong feature due to H CO + within one of the displayed frequency windowsof CH C N is marked in blue. (j) (k) (l)(m) (n)
Figure C.9. (cont.)
Same as Figure C.9. Panels (j) to (n) show the OCS, HC N, C S, CH CCH, andH CN lines, respectively. Guzm´an et al. CH OCHO ? HC N (o) CH CCHCH CCHCCH CH CCH (p)
Figure C.9. (cont.)
Same as Figure C.9. Panels (o) and (p) show the C H CN and CH CHCN lines,respectively. Panel (o) shows also the location of two possible lines of CH OCHO. Strong lines from otherspecies within the displayed frequency windows are marked in blue.
RAS 16562 − (q) (r) (s) (t)(u) Figure C.9. (cont.)
Same as Figure C.9. Panels (q) to (u) show the lines of HC CCN, CS, H CCO,HC N, and CH OCH , respectively. (a) (b) (c) (d)(e) Figure C.10.
Emission lines from several molecules toward the integrated NW cloud. Line types and colorsas described in Figure C.1 with σ = 0 .
009 K. Panels (a) to (e) show lines of H CCO, HC N, SO, H CO + ,and CH OH, respectively. The strong non-LTE CH OH, (E ) J K a K c = 5 − , (cid:1) , emission shown in panel(e) cannot be reproduced with the models used in this work. Guzm´an et al. (f) (g)(h) (i)(j) (k) (l)(m)
Figure C.10. (cont.)
Same as Figure C.10. Panels (f) to (m) show lines of CH CCH, HC O + , SO, HC N,OCS, H CN, HN C, and CCH, respectively.
RAS 16562 − (n)(o) (p) (q)(r) (s) Figure C.10. (cont.)
Same as Figure C.10. Panels (n) to (s) show lines of CH CHO, HNCO, CS, SiO, SiO, and SiO, respectively. (a)
Figure C.11. CH OH lines from several molecules toward the point NW cloud (a), the methanol peak.Line types and colors as described in Figure C.1 with σ = 0 .
08 K. Guzm´an et al. (b) (c)(d) (e) (f)
Figure C.11. (cont.)
Same as Figure C.11. Panels (b) to (f) show lines of OCS, CH CHO, HNCO, SiO,and SiO, respectively.
Figure C.12.
Emission lines from CH OH toward Source 3. Line types and colors as described in FigureC.1 with σ = 0 .
05 K.
RAS 16562 − Figure C.13.
Emission lines from CH OH toward Source 18. Line types and colors as described in FigureC.1 with σ = 0 .
05 K. The model used in this work does not reproduce well the strong absorption featuresassociated with the lowest energy CH OH transitions and the CH OH, (E ) J K a K c = 5 − , (cid:1) , line. Guzm´an et al.
REFERENCES
Adande, G. R., & Ziurys, L. M. 2012, ApJ, 744,194, doi: 10.1088/0004-637X/744/2/194Allen, V., van der Tak, F. F. S., S´anchez-Monge,´A., Cesaroni, R., & Beltr´an, M. T. 2017, A&A,603, A133, doi: 10.1051/0004-6361/201629118Asplund, M., Grevesse, N., Sauval, A. J., & Scott,P. 2009, ARA&A, 47, 481,doi: 10.1146/annurev.astro.46.060407.145222Bally, J., Ginsburg, A., Arce, H., et al. 2017, ApJ,837, 60, doi: 10.3847/1538-4357/aa5c8bBalucani, N., Asvany, O., Huang, L. C. L., et al.2000, ApJ, 545, 892, doi: 10.1086/317848Belloche, A., M¨uller, H. S. P., Menten, K. M.,Schilke, P., & Comito, C. 2013, A&A, 559, A47,doi: 10.1051/0004-6361/201321096Beltr´an, M. T., & de Wit, W. J. 2016, A&A Rv,24, 6, doi: 10.1007/s00159-015-0089-zBenjamin, R. A., Churchwell, E., Babler, B. L.,et al. 2003, PASP, 115, 953, doi: 10.1086/376696Bergin, E. A., Goldsmith, P. F., Snell, R. L., &Ungerechts, H. 1994, ApJ, 431, 674,doi: 10.1086/174518Bergin, E. A., & Tafalla, M. 2007, ARA&A, 45,339,doi: 10.1146/annurev.astro.45.071206.100404Bergner, J. B., ¨Oberg, K. I., Garrod, R. T., &Graninger, D. M. 2017, ApJ, 841, 120,doi: 10.3847/1538-4357/aa72f6Beuther, H., Zhang, Q., Bergin, E. A., &Sridharan, T. K. 2009, AJ, 137, 406,doi: 10.1088/0004-6256/137/1/406Beuther, H., Zhang, Q., Sridharan, T. K., Lee,C.-F., & Zapata, L. A. 2006, A&A, 454, 221,doi: 10.1051/0004-6361:20064887Beuther, H., Zhang, Q., Greenhill, L. J., et al.2005, ApJ, 632, 355, doi: 10.1086/432974Bisschop, S. E., Jørgensen, J. K., van Dishoeck,E. F., & de Wachter, E. B. M. 2007, A&A, 465,913, doi: 10.1051/0004-6361:20065963Blake, G. A., Mundy, L. G., Carlstrom, J. E.,et al. 1996, ApJL, 472, L49, doi: 10.1086/310347Blake, G. A., Sutton, E. C., Masson, C. R., &Phillips, T. G. 1987, ApJ, 315, 621,doi: 10.1086/165165Bronfman, L., Nyman, L.-A., & May, J. 1996,A&AS, 115, 81Caselli, P., Hasegawa, T. I., & Herbst, E. 1993,ApJ, 408, 548, doi: 10.1086/172612 Caswell, J. L. 1998, MNRAS, 297, 215,doi: 10.1046/j.1365-8711.1998.01468.x—. 2009, PASA, 26, 454, doi: 10.1071/AS09013Caux, E., Bottinelli, S., Vastel, C., & Glorian,J. M. 2011, in IAU Symposium, Vol. 280, TheMolecular Universe, ed. J. Cernicharo &R. Bachiller, 120Cesaroni, R., S´anchez-Monge, ´A., Beltr´an, M. T.,et al. 2017, A&A, 602, A59,doi: 10.1051/0004-6361/201630184Chapman, J. F., Millar, T. J., Wardle, M.,Burton, M. G., & Walsh, A. J. 2009, MNRAS,394, 221, doi: 10.1111/j.1365-2966.2008.14144.xCharnley, S. B. 1997, ApJ, 481, 396,doi: 10.1086/304011Charnley, S. B. 1999, in NATO Advanced ScienceInstitutes (ASI) Series C, Vol. 523, Formationand Evolution of Solids in Space, ed. J. M.Greenberg & A. Li (Springer Science+BusinessMedia Dordrecht), 131Chin, Y.-N., Henkel, C., Whiteoak, J. B., Langer,N., & Churchwell, E. B. 1996, A&A, 305, 960Colzi, L., Fontani, F., Rivilla, V. M.,S´anchez-Monge, A., Testi, L., Beltr´an, M. T.,Caselli, P. 2018, MNRAS,doi: 10.1093/mnras/sty1027Cosentino, G., Jim´enez-Serra, I., Henshaw, J. D.,et al. 2018, MNRAS, 474, 3760,doi: 10.1093/mnras/stx3013Cragg, D. M., Johns, K. P., Godfrey, P. D., &Brown, R. D. 1992, MNRAS, 259, 203,doi: 10.1093/mnras/259.1.203Crockett, N. R., Bergin, E. A., Neill, J. L., et al.2015, ApJ, 806, 239,doi: 10.1088/0004-637X/806/2/239Dahmen, G., Wilson, T. L., & Matteucci, F. 1995,A&A, 295, 194Dame, T. M. 2011, ArXiv e-prints.https://arxiv.org/abs/1101.1499Danilovich, T., De Beck, E., Black, J. H.,Olofsson, H., & Justtanont, K. 2016, A&A, 588,A119, doi: 10.1051/0004-6361/201527943Fayolle, E. C., ¨Oberg, K. I., Garrod, R. T., vanDishoeck, E. F., & Bisschop, S. E. 2015, A&A,576, A45, doi: 10.1051/0004-6361/201323114Fedoseev, G., Ioppolo, S., & Linnartz, H. 2015,MNRAS, 446, 449, doi: 10.1093/mnras/stu1852
RAS 16562 − Feng, S., Beuther, H., Henning, T., et al. 2015,A&A, 581, A71,doi: 10.1051/0004-6361/201322725Ferrante, R. F., Moore, M. H., Spiliotis, M. M., &Hudson, R. L. 2008, ApJ, 684, 1210,doi: 10.1086/590362Fontani, F., Pascucci, I., Caselli, P., et al. 2007,A&A, 470, 639,doi: 10.1051/0004-6361:20077485Friberg, P., Hjalmarson, A., Madden, S. C., &Irvine, W. M. 1988, A&A, 195, 281Friedel, D. N., & Widicus Weaver, S. L. 2012,ApJS, 201, 17,doi: 10.1088/0067-0049/201/2/17Garrod, R. T., Widicus Weaver, S. L., & Herbst,E. 2008, ApJ, 682, 283, doi: 10.1086/588035Giannetti, A., Leurini, S., Wyrowski, F., et al.2017, A&A, 603, A33,doi: 10.1051/0004-6361/201630048Gibb, E., Nummelin, A., Irvine, W. M., Whittet,D. C. B., & Bergman, P. 2000, ApJ, 545, 309,doi: 10.1086/317805Gibb, E. L., Whittet, D. C. B., Boogert, A. C. A.,& Tielens, A. G. G. M. 2004, ApJS, 151, 35,doi: 10.1086/381182Gong, Y., Henkel, C., Thorwirth, S., et al. 2015,A&A, 581, A48,doi: 10.1051/0004-6361/201526275Gottlieb, C. A., Gottlieb, E. W., & Thaddeus, P.1983, ApJ, 264, 740, doi: 10.1086/160647Graninger, D. M., Herbst, E., ¨Oberg, K. I., &Vasyunin, A. I. 2014, ApJ, 787, 74,doi: 10.1088/0004-637X/787/1/74Guzm´an, A. E., Garay, G., & Brooks, K. J. 2010,ApJ, 725, 734,doi: 10.1088/0004-637X/725/1/734Guzm´an, A. E., Garay, G., Brooks, K. J.,Rathborne, J., & G¨usten, R. 2011, ApJ, 736,150, doi: 10.1088/0004-637X/736/2/150Guzm´an, A. E., Garay, G., Rodr´ıguez, L. F., et al.2016, ApJ, 826, 208,doi: 10.3847/0004-637X/826/2/208Guzm´an, A. E., Guzm´an, V. V., Garay, G.,Bronfman, L., & Hechenleitner, F. 2018,Replication data for: Chemistry of theHigh-Mass Protostellar Molecular Clump IRAS16562 − Guzm´an et al.
Lampton, M., Margon, B., & Bowyer, S. 1976,ApJ, 208, 177, doi: 10.1086/154592Lefloch, B., Ceccarelli, C., Codella, C., et al. 2017,MNRAS, 469, L73, doi: 10.1093/mnrasl/slx050L´opez-Calder´on, C., Bronfman, L., Nyman, L.-˚A.,et al. 2016, A&A, 595, A88,doi: 10.1051/0004-6361/201321696Lu, X., Zhang, Q., Kauffmann, J., et al. 2017,ApJ, 839, 1, doi: 10.3847/1538-4357/aa67f7Miettinen, O., Harju, J., Haikala, L. K., &Pomr´en, C. 2006, A&A, 460, 721,doi: 10.1051/0004-6361:20064815Milam, S. N., Savage, C., Brewster, M. A., Ziurys,L. M., & Wyckoff, S. 2005, ApJ, 634, 1126,doi: 10.1086/497123Minniti, D., Lucas, P. W., Emerson, J. P., et al.2010, NewA, 15, 433,doi: 10.1016/j.newast.2009.12.002Molinari, S., Merello, M., Elia, D., et al. 2016,ApJL, 826, L8,doi: 10.3847/2041-8205/826/1/L8Monson, N. N., Morris, M. R., & Young, E. D.2017, ApJ, 839, 123,doi: 10.3847/1538-4357/aa67e6Mookerjea, B., Casper, E., Mundy, L. G., &Looney, L. W. 2007, ApJ, 659, 447,doi: 10.1086/512095M¨uller, H. S. P., Schl¨oder, F., Stutzki, J., &Winnewisser, G. 2005, Journal of MolecularStructure, 742, 215,doi: 10.1016/j.molstruc.2005.01.027¨Oberg, K. I., Boamah, M. D., Fayolle, E. C., et al.2013, ApJ, 771, 95,doi: 10.1088/0004-637X/771/2/95¨Oberg, K. I., Fayolle, E. C., Reiter, J. B., &Cyganowski, C. 2014, Faraday Discussions, 168,81, doi: 10.1039/C3FD00146FOlofsson, H. 2005, in IAU Symposium, Vol. 231,Astrochemistry: Recent Successes and CurrentChallenges, ed. D. C. Lis, G. A. Blake, &E. Herbst, 499–508Palau, A., Walsh, C., S´anchez-Monge, ´A., et al.2017, MNRAS, 467, 2723,doi: 10.1093/mnras/stx004Penzias, A. A. 1981, ApJ, 249, 513,doi: 10.1086/159310Persson, C. M., Olofsson, A. O. H., Koning, N.,et al. 2007, A&A, 476, 807,doi: 10.1051/0004-6361:20077225 Petry, D., & CASA Development Team. 2012, inAstronomical Society of the Pacific ConferenceSeries, Vol. 461, Astronomical Data AnalysisSoftware and Systems XXI, ed. P. Ballester,D. Egret, & N. P. F. Lorente, 849Pickett, H. M., Poynter, R. L., Cohen, E. A., et al.1998, JQSRT, 60, 883,doi: 10.1016/S0022-4073(98)00091-0Pineau des Forets, G., Roueff, E., Schilke, P., &Flower, D. R. 1993, MNRAS, 262, 915,doi: 10.1093/mnras/262.4.915Plunkett, A. L., Arce, H. G., Corder, S. A., et al.2013, ApJ, 774, 22,doi: 10.1088/0004-637X/774/1/22Podio, L., Codella, C., Gueth, F., et al. 2015,A&A, 581, A85,doi: 10.1051/0004-6361/201525778Prasad, S. S., & Huntress, Jr., W. T. 1980, ApJS,43, 1, doi: 10.1086/190665Qin, S.-L., Wu, Y., Huang, M., et al. 2010, ApJ,711, 399, doi: 10.1088/0004-637X/711/1/399Rathborne, J. M., Jackson, J. M., & Simon, R.2006, ApJ, 641, 389, doi: 10.1086/500423Rieke, G. H., & Lebofsky, M. J. 1985, ApJ, 288,618, doi: 10.1086/162827Rodr´ıguez-Fern´andez, N. J., Tafalla, M., Gueth,F., & Bachiller, R. 2010, A&A, 516, A98,doi: 10.1051/0004-6361/201013997Roueff, E., Lis, D. C., van der Tak, F. F. S.,Gerin, M., & Goldsmith, P. F. 2005, A&A, 438,585, doi: 10.1051/0004-6361:20052724Sakai, N., & Yamamoto, S. 2013, ChemicalReviews, 113, 8981, doi: 10.1021/cr4001308Sakai, T., Sakai, N., Foster, J. B., et al. 2013,ApJL, 775, L31,doi: 10.1088/2041-8205/775/1/L31Sanhueza, P., Jackson, J. M., Foster, J. B., et al.2013, ApJ, 773, 123,doi: 10.1088/0004-637X/773/2/123Schiff, H. I., & Bohme, D. K. 1979, ApJ, 232, 740,doi: 10.1086/157334Schilke, P., Walmsley, C. M., Pineau Des Forets,G., et al. 1992, A&A, 256, 595Sch¨oier, F. L., van der Tak, F. F. S., vanDishoeck, E. F., & Black, J. H. 2005, A&A, 432,369, doi: 10.1051/0004-6361:20041729Shirley, Y. L. 2015, PASP, 127, 299,doi: 10.1086/680342
RAS 16562 −97