First Results of an ALMA Band 10 Spectral Line Survey of NGC 6334I: Detections of Glycolaldehyde (HC(O)CH 2 OH) and a New Compact Bipolar Outflow in HDO and CS
Brett A. McGuire, Crystal L. Brogan, Todd R. Hunter, Anthony J. Remijan, Geoffrey A. Blake, Andrew M. Burkhardt, P. Brandon Carroll, Ewine F. van Dishoeck, Robin T. Garrod, Harold Linnartz, Christopher N. Shingledecker, Eric R. Willis
DDraft version August 17, 2018
Typeset using L A TEX twocolumn style in AASTeX62
First Results of an ALMA Band 10 Spectral Line Survey of NGC 6334I: Detections of Glycolaldehyde(HC(O)CH OH) and a New Compact Bipolar Outflow in HDO and CS
Brett A. McGuire,
1, 2, ∗ Crystal L. Brogan, Todd R. Hunter, Anthony J. Remijan, Geoffrey A. Blake,
3, 4
Andrew M. Burkhardt, † P. Brandon Carroll, Ewine F. van Dishoeck,
6, 7
Robin T. Garrod,
8, 5
Harold Linnartz, Christopher N. Shingledecker, ‡ and Eric R. Willis National Radio Astronomy Observatory, Charlottesville, VA 22903, USA Harvard-Smithsonian Center for Astrophysics, Cambridge, MA 02138, USA Division of Geological and Planetary Sciences, California Institute of Technology, Pasadena, CA 91125, USA Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, CA 91125, USA Department of Astronomy, University of Virginia, Charlottesville, VA 22904, USA Leiden Observatory, Leiden University, 2300 RA Leiden, The Netherlands Max-Planck Institut f¨ur Extraterrestrische Physik, Giessenbachstr. 1, 85748 Garching, Germany Department of Chemistry, University of Virginia, Charlottesville, VA 22904, USA Sackler Laboratory for Astrophysics, Leiden Observatory, Leiden University, 2300 RA Leiden, The Netherlands
ABSTRACTWe present the first results of a pilot program to conduct an ALMA Band 10 spectral line surveyof the high-mass star-forming region NGC 6334I. The observations were taken in exceptional weatherconditions (0.19 mm precipitable water) with typical system temperatures T sys <
950 K at ∼
890 GHz.A bright, bipolar north-south outflow is seen in HDO and CS emission, driven by the embedded massiveprotostar MM1B. This has allowed, for the first time, a direct comparison of the thermal water in thisoutflow to the location of water maser emission from prior 22 GHz VLA observations. The maserlocations are shown to correspond to the sites along the outflow cavity walls where high velocity gasimpacts the surrounding material. We also compare our new observations to prior
Herschel
HIFIspectral line survey data of this field, detecting an order of magnitude more spectral lines (695 vs 65)in the ALMA data. We focus on the strong detections of the complex organic molecule glycolaldehyde(HC(O)CH OH) in the ALMA data that is not detected in the heavily beam-diluted HIFI spectra.Finally, we stress the need for dedicated THz laboratory spectroscopy to support and exploit futurehigh-frequency molecular line observations with ALMA.
Keywords:
Astrochemistry – ISM: molecules – ISM: individual objects (NGC 6334I) – ISM: jets andoutflows – masers INTRODUCTION
Corresponding author: Brett A. [email protected] ∗ B.A.M. is a Hubble Fellow of the National Radio AstronomyObservatory † Current Address: Submillimeter Array (SMA) PostdoctoralFellow, Harvard-Smithsonian Center for Astrophysics,Cambridge, MA 02138 ‡ Current Address: Alexander von HumboldtFoundation Postdoctoral Research Fellow, Max PlanckInstitute for Extraterrestrial Physics, Garching,Germany; Institute for Theoretical Chemistry,University of Stuttgart, Stuttgart, Germany
Observations with the Atacama Large Millimeter Ar-ray (ALMA) in Bands 3–7 (84–373 GHz) have provento be exceptional tools for the detection of new molec-ular species in the interstellar medium (ISM) and thestudy of their chemical history and interaction withtheir physical environment. As a few examples, Bel-loche et al. (2014) reported the first detection of abranched carbon-chain molecule in the ISM, iso-propylcyanide (C H CN), in Band 3 observations of Sgr B2.Later, McGuire et al. (2017) detected methoxymethanol(CH OCH OH) in surprisingly high abundance towardNGC 6334I in Band 6 and 7 observations, while Fay-olle et al. (2017) identified methyl chloride (CH Cl) forthe first time using Band 7 observations of IRAS 16293-2422. These observations, among many others, demon- a r X i v : . [ a s t r o - ph . GA ] A ug Figure 1.
ALMA Band 10 images of the peak (a, b) and integrated intensity (c, d) from − . . − for twotransitions of HDO with 0.35 mm dust continuum contours overlaid. The 0.35 mm dust continuum contour levels are 50, 75,and 150 K on the Planck temperature scale. Magenta × symbols show the location of the spectra extracted for MM1 (Fig. 3).There is substantial absorption of the ground-state transition of HDO toward the MM1 continuum in the integrated intensitymaps. This absorption dominates the integrated intensity map, leaving no net emission signal, whereas the peak intensity mapsshow the locations of this subsumed HDO emission. Primary beam correction has been applied with a cutoff at 0.25 of theFWHM. The synthesized beam of 0 . (cid:48)(cid:48) × . (cid:48)(cid:48)
16 (PA= 39 ◦ ) is shown in the lower right of each panel. strate the power of ALMA for studies of our molecularuniverse in the 1–3 mm wavelength range.Astrochemical observations at higher frequencies, inALMA Bands 9 (602–720 GHz) and 10 (787–950 GHz)offer complementary benefits to the lower-frequencydata, yet few molecular line surveys have been con-ducted at these frequencies. Here, we explore two ad-vantages of high-frequency spectral line observations.First, the fundamental or first few lowest transitions ofmany small molecules of interest fall into this range. Forexample, the HDO 1 , − , fundamental transition oc-curs at 893.6 GHz (Messer & De Lucia 1984), providingone of the best opportunities to obtain ground-basedmeasurements of thermal water (see, e.g., Comito et al.2003).Second, the transitions of most complex organicmolecules (COMs) that fall within this frequency rangeare typically much higher in energy, providing a robustconstraint on excitation conditions within a source. Forexample, the strongest transitions of glycolaldehyde(HC(O)CH OH), the simplest sugar-related molecule,in Band 6 have upper state energies E u ∼ OH in Band 10 have E u = 530-630 K – andcan provide needed confirmatory transitions to securea lower-frequency detection (see, e.g., Jørgensen et al.2012).These higher-energy transitions also provide selectiveaccess to the warmest molecular gas in a source, whichprior studies have shown can have substantially dif-ferent chemistry from the population probed by thelower-energy transitions accessible at lower frequencies(Crockett et al. 2014). There is, however, a relative lackof direct laboratory measurements of molecular spectraabove ∼
600 GHz, meaning that many identifications aremade from extrapolated quantum mechanical fits. Forsome species this is a reasonably accurate process, but,as will be shown later, the richness of the Band 10 spec-tra underscores the need for dedicated high-frequencylaboratory work. Finally, at these frequencies it is rea-sonable to expect that increased dust optical depth ef-fects might “hide” the deepest, most compact regionsof hot molecular cores, and the bright continuum mightdrive many molecular transitions into absorption. To explore the utility of ALMA Band 9/10 observa-tions, we proposed for a full Band 9 survey, and a pilotBand 10 survey, of the high-mass star-forming regionNGC 6334I in Cycle 5. NGC 6334I was chosen as thetarget for three reasons. First, it is an exceptionallymolecular line-rich source (McGuire et al. 2017) with arelatively small heliocentric distance of 1.3 kpc (Reidet al. 2014; Chibueze et al. 2014). Second, it has previ-ously been targeted by single-dish observations in over-lapping frequency ranges by Zernickel et al. (2012) usingthe Heterodyne Instrument for the Far-Infrared (HIFI;de Graauw et al. 2010) on the
Herschel
Space Observa-tory (Pilbratt et al. 2010). Third, it displays a complexspatial structure consisting of a substantial number ofembedded sources and outflows, and several chemicallydistinct regions separated by only ∼ ∼ . (cid:48)(cid:48)
5; Bro-gan et al. 2016; McGuire et al. 2017; Bøgelund et al.2018; Figure 1).Here, we present a first look at ALMA Band 10 obser-vations of any line-rich source, and discuss the resultsin the context of both probing favorable transitions oflight molecules, and in examining the high-excitationlines of complex organic species. The spectra at thesehigh frequencies are as line-rich as those in the millime-ter regime. Contrary to initial expectations, observa-tions of high-mass star-forming regions like NGC 6334Iat ALMA Band 10 do not appear to be substantiallyhampered by dust opacity, and are in fact generallybetter suited than previous single-dish facilities suchas Herschel. These first-look observations demonstratethe power and versatility of high-frequency observationswith ALMA. OBSERVATIONS & DATA REDUCTIONThe Band 10 observation occurred 05 April 2018, with40 ALMA antennas in the array in a nominal C43-3 con-figuration with a maximum baseline of 532 m providinga 0.21 (cid:48)(cid:48) × (cid:48)(cid:48) synthesized beam (robust weighting pa-rameter = 0.5). The precipitable water vapor at thetime of observation was 0.19 mm, and the average result-ing system temperature was T sys = 926 K at 880 GHz.Total time on source was 47 minutes, resulting in anRMS noise level of 62 mJy beam − in 0.5 km s − chan-nels. J1517-2422 ( ∼ ∼ ∼ (cid:48)(cid:48) , which is slightly smaller than the total an-gular extent of the emitting regions of interest inNGC 6334I. We therefore targeted two phase centersto cover the entire source while maximizing the UVcoverage and RMS sensitivity in the critical central re-gions. As of publication, only the pointing positiontoward MM1 has been observed; with a phase center of α (J2000) = 17 h m s δ (J2000) = − ◦ (cid:48) (cid:48)(cid:48) (Fig-ure 1). The 0.35 mm continuum was created from (rel-atively) line-free channels, and has an aggregate band-width and rms noise of 2.7 GHz and 50 mJy beam − ,respectively. The method employed for continuum sub-traction is described in detail in Brogan et al. (2018).Self-calibration was performed on the continuum andapplied to the spectral line data after subtracting thecontinuum in the uv-plane. The details of the comple-mentary Band 7 data used in this paper are describedin McGuire et al. (2017), while the Band 4 data aredescribed in Appendix A.For the analysis presented here, spectra were ex-tracted toward MM1 from a single pixel at a position of α (J2000) = 17 h m s δ (J2000) = − ◦ (cid:48) (cid:48)(cid:48) (magenta color cross in Figure 1). The location is ∼
400 au west of the brightest continuum peak, MM1B.This location was chosen for its proximity to the densemolecular gas while being far enough from the brightcontinuum that the majority of the molecular lines arenot driven into absorption due to the high continuumbrightness temperature (McGuire et al. 2017).The NGC 6334I region was previously targeted in anextensive, broadband spectral line survey as part of theCHESS (Chemical HErschel Surveys of Star forming re-gions; Ceccarelli et al. 2010) key program using HIFI(Zernickel et al. 2012). The data used in this manuscriptfor comparison to our ALMA data were obtained fromthe
Herschel
Science Archive, ObsID 1342192328. TheHIFI data are used directly as downloaded from thearchive, with the exception of the subtraction of astatic continuum offset, to place the spectral baselineat T A = 0 K. HDO AND CSMost transitions of H O are blocked from the groundby atmospheric absorption features, and thus observa-tions normally rely either on extreme excitation con-ditions (i.e. maser emission), isotopologues, or space-based observatories to study H O. Spectra towardNGC 6334I, Orion KL, and other star-forming regions inthe rotational and lowest few ground state transitions ofortho-H O recorded by the
Submillimeter Wave Astron-omy Satellite (SWAS) and
Herschel exhibit broad wingcomponents that arise from heated gas in low and highvelocity outflowing gas (Melnick et al. 2000; Neufeldet al. 2000; van der Tak et al. 2013; San Jos´e-Garc´ıa et al.2015). Our Band 10 observations provide access to theHDO fundamental 1 , − , transition at 893.6 GHz( E u = 43 K) and the higher-energy ( E u = 581 K) 6 , − , transition at 895.9 GHz (Messer & De Lu-cia 1984).Figure 1 shows the full velocity extent of these HDOtransitions as both peak intensity and integrated inten-sity images. While no emission signals of this tran-sition were detected in the HIFI data (Emprechtingeret al. 2013), the lower-energy, fundamental HDO tran-sition shows an extended distribution in the high res-olution ALMA data. The higher-energy transition ismore compact and located closer to the primary con-tinuum sources, likely tracing warmer regions withinthe source. This pattern is similar to that measuredin Orion KL for two intermediate energy transitions inALMA Band 6 (Neill et al. 2013). It is notable that thefundamental transition of HDO exhibits significant self-absorption toward the bright continuum sources, andeven the 6 , − , transition is affected by the highcontinuum opacity and brightness temperature at thecontinuum peak locations (see Fig. 1.)Figure 2a shows only the redshifted, high velocitycomponents of the HDO emission, and reveals a struc-ture that traces a north-south bipolar outflow emanat-ing out of the MM1 continuum peak, centered on theMM1B protostellar source. Our data also covered theCS J = 18–17 transition ( E u =402 K) at 880.9 GHz(M¨uller et al. 2005), which shows the same north-southoutflow structure as HDO (Figure 2b). This outflow,called the MM1B N-S outflow has also been detectedin ALMA CS (6-5) emission by Brogan et al. (2018).These authors find a total linear extent of 6 . (cid:48)(cid:48) (8060 au)for MM1B N-S, and that its orientation is nearly in theplane of the sky, leading to a very young dynamical timeof only 166 years.Extensive observations of H O masers have been madetoward NGC 6334I at cm-wavelengths both with single-dish monitoring (MacLeod et al. 2018) and recent com-plementary Karl G. Jansky Very Large Array (VLA)interferometric observations (Brogan et al. 2016; Bro-gan et al. 2018). By overlaying the locations of cm-wavelength H O masers from the VLA data, it becomesapparent that the masers are predominantly tracing thewalls of this outflow cavity where the high velocity gaslikely impacts the surrounding quiescent gas. These lo-cations are consistent with water maser pumping modelsin which the observed masers arise from velocity coher-ent structures in the hot gas behind shocks propagatingin dense regions (Hollenbach et al. 2013; Melnick et al.1993). The ability to monitor both the thermal water inthe outflow and pinpoint the locations of the H O maseremission to the impact sites of the outflow into the sur-rounding media demonstrates the powerful combinationof ALMA Band 10 data with VLA cm-wave observa-
Figure 2.
ALMA Band 10 images of the integrated intensity of the redshifted high velocity gas ( − − ; thesystemic velocity is − − ) for the (a) ground state of HDO and the (b) CS (18-17) transition showing a highly collimatednorth-south outflow emanating from MM1. Blueshifted high velocity emission (not shown) is co-spatial with the redshiftedemission, but is much weaker. The 0.35 mm continuum is the same as Fig. 1. White circles show the locations of H O masersdetected with the VLA in epoch 2017.8 (Brogan et al. 2018). Primary beam correction has been applied with a cutoff at 0.25the FWHM. The synthesized beam of 0 . (cid:48)(cid:48) × . (cid:48)(cid:48)
16 (PA= 39 ◦ ) is shown in the lower right of each panel. tions. In this case, the detection of a collimated outflowin thermal and maser lines along the same axis as thecompact radio jet from MM1B (Hunter et al. 2018; Bro-gan et al. 2018) enables the recognition of these variousdistinct phenomena as arising from a unified structure.3.1. Spectral Line Survey & Glycolaldehyde(HC(O)CH OH)
The benefits of performing high-frequency spectralline surveys were recognized and exploited by a num-ber of key projects performed with the HIFI instru-ment aboard
Herschel . As mentioned earlier, manysmall molecules have their first few rotational transi-tions at these higher frequencies, and surveys with HIFIresulted in the first reported detections of SH + (Benzet al. 2010), HCl + (De Luca et al. 2012), H O + (Os-senkopf et al. 2010), and H Cl + (Lis et al. 2010). Addi-tionally, full spectral line surveys provided robust con-straints on the excitation conditions and populations oflarger molecules ( > Herschel
HIFI from 500–1900 GHz (see Zer-nickel et al. 2012 and refs. therein). While these sur-veys revealed a line-rich source with a complex molec-ular inventory, the large beam sizes were unable toresolve the underlying structure, and suffered signifi- cantly from beam dilution. The overall angular ex-tent of NGC 6334I is ∼ (cid:48)(cid:48) × (cid:48)(cid:48) (13000 × ≤ (cid:48)(cid:48) (6500 au) (McGuire et al. 2017). In a Herschel beam of ∼ (cid:48)(cid:48) at 880 GHz, this mismatch results in a factor of (cid:38)
25 loss in line brightness due to beam dilution.Indeed, when compared to the spectral line surveyfrom HIFI presented in Zernickel et al. (2012), themost striking feature of our ALMA Band 10 spectra isthe greatly enhanced line density, and the number oflines which are optically thick. Figure 3 shows the full ∼ ∼
10 times that of HIFI (695 vs 65 lines inthe ALMA vs HIFI spectra over the same range). This isexemplified by the emission lines of CH OH ( v t = 1) andC O seen in each spectrum. The complex molecularemission is significantly enhanced in the ALMA spectradue to its more spatially compact distribution. C O,conversely, is not optically thick in the ALMA data, witha substantially decreased intensity relative to the com-plex molecules. This effect is almost certainly due to amore extended distribution which is being partially re-solved out in the ALMA observations, but for which theHIFI observations were well-suited.Because the Band 10 data are so molecular line-rich,they can provide valuable constraints on molecular ex-citation and column density derivations if they can be T A ( K ) Jy / bea m Herschel
HIFI (inverted)ALMA Band 10 C OCH OH v t = 1 Figure 3.
Comparison of this ALMA Band 10 dataset (top; extracted toward MM1) to observations retrieved from the
Herschel archive and taken as part of the CHESS key program using the HIFI instrument covering the same data range (bottom; Zernickelet al. 2012). The
Herschel beam ( ∼ (cid:48)(cid:48) at 880 GHz) covered the entirety of the NGC 6334I region. Note that the ALMA spectraare in Jy/beam while the Herschel data are in antenna temperature, and that the
Herschel spectra have been inverted forcomparison. A constant offset has been subtracted from the
Herschel data to remove the continuum for presentation purposes.Transitions of CH OH v t = 1 and C O are labeled. robustly analyzed alongside lower-frequency observa-tions. To test this, we extracted spectra toward theMM1 spectral analysis position from data obtainedin previous ALMA observations toward NGC 6334Iin Band 4 (ADS/JAO.ALMA (cid:48)(cid:48) × (cid:48)(cid:48) . A full molec-ular analysis of the Band 10 survey is beyond the scopeof this Letter. Once the full Band 9/10 spectral survey iscomplete, we plan to provide a fully-reduced line surveyto the community. A preliminary inventory of molecules,however, includes CO, C O, H CO, HNCO, CH OH, CH OH, CH
OH, CH OH v t = 1, CH CN, CH NH,CH NH , NH CHO, CH CH OH, HC(O)CH OH, andCH OCHO.Here, we focus only on HC(O)CH OH, the simplestsugar-related molecule. This complex organic moleculewas successfully detected by ALMA toward the solar-mass protostar IRAS 16293-2422 by Jørgensen et al.(2012), but has not been reported in the HIFI data atBand 10 frequencies.While collisional cross sections for molecules morecomplex than methanol (CH OH) are generally notavailable, especially at these frequencies, the high densi-ties in the region (source averaged n H > cm − ; Rus-seil et al. 2010) suggest that molecules should be well-described by a single T ex . We therefore used the for-malism described in Turner (1991) to simulate the spec-trum of HC(O)CH OH across the entire ∼
750 GHz spanin observational coverage, accounting for differences inbackground continuum temperature at each frequency.The simulated spectra were converted to Jy/beam from Kelvin using the Planck scale, as the Rayleigh-Jeansapproximation introduces significant errors at Band 10frequencies. The simulated HC(O)CH OH spectra areenabled by the high-frequency laboratory work of Car-roll et al. (2010) that extended the measured frequenciesfrom 354 GHz to 1.2 THz.Figure 4 shows the resulting simulated HC(O)CH OHemission overlaid on the observational data in Bands4, 7, and 10. We find that assuming a singleexcitation temperature ( T ex = 135 K), linewidth(∆ V = 3.2 km s − ), and column density ( N T = 1 . × cm − ) across the bands well reproduces the ob-served emission to zeroth order. The Band 10 data pro-vide a high-energy anchor to the excitation conditions;the six lines shown in Figure 4 have upper state energiesbetween E u = 530–631 K, compared to E u = 63–171 Kfor the Band 4 lines shown. A full fit of the spectrum,including line contamination and blending from otherspecies, is beyond the scope of this first-look paper.Preliminary work indicates the availability of these highand low-energy anchors provides definite constraintson the excitation temperature of glycolaldehyde to be ∼
135 K.The peak and integrated intensities of the HC(O)CH OHtransition at 892.12 GHz ( J K a ,K c = 28 ,x − ,x ; E u = 546 K) is shown in Fig. 5. Toward the MM1cluster, these high-energy transitions seem to trace asimilar, although not identical, distribution to the high-energy HDO lines shown in Figure 1, but has a markedlydifferent distribution toward MM2. Recent laboratorystudies have shown that HC(O)CH OH is readily formedon icy dust grains through hydrogenation of solid CO(Fedoseev et al. 2015). The observed distribution may Jy / B ea m Jy / B ea m Jy / B ea m Jy / B ea m BA ND BA ND BA ND ObservationsGlycolaldehydeSimulation
Figure 4.
Simulated spectra of HC(O)CH OH plotted in red over ALMA observations of NGC 6334I MM1 plotted in black .The simulated spectra assume T ex = 135 K, ∆ V = 3.2 km s − , and N T = 1 . × cm − , with a v lsr = -7 km s − . The ALMAobservations have been convolved to a uniform synthesized beam of 0.26 (cid:48)(cid:48) × (cid:48)(cid:48) . In the lower six panels, smaller regions of thefrequency coverage in Band 10 have been selected to show detail. Figure 5.
ALMA Band 10 images of the peak (a) and integrated intensity (b) of the HC(O)CH OH transition at 892.12 GHz(28,23 - 27,22; Eu=546 K) shown in Fig. 4. The 0.35 mm continuum is the same as Fig. 1 and the synthesized beam is shownin the lower right corner. indicate that after being formed in the ice, some of thiscondense-phased HC(O)CH OH is being driven into thegas-phase by a non-thermal desorption mechanism suchas shock-induced sputtering, as previously hypothesizedin observations of complex molecules in galactic centerregions (Requena-Torres et al. 2006). Thermal desorp-tion mechanisms are likely also relevant in many warmerregions of the source.The richness of the Band 10 spectra underscores aneed for accurate, high-resolution gas phase spectra ofcomplex molecules from laboratory studies in these fre-quency ranges. While laboratory data in the millimeter-wave range below ∼
500 GHz are increasingly common,the technical difficulties in measuring and analyzing sub-millimeter and THz spectra, combined with a generallack of observational applications, has limited the num-ber of groups working in this regime. While a dedicatedeffort was undertaken to provide HC(O)CH OH spec-tra through 1.2 THz, laboratory spectra for many othercomplex molecules are missing, and transitions in thisrange must be extrapolated from lower frequencies, lead-ing to increasingly large uncertainties. CONCLUSIONSWe have presented a first look at ALMA Band 10 spec-tral line survey toward a line-rich source – the high-massstar-forming region NGC 6334I – obtained in exceptionalweather conditions. The resulting map shows a bright,bi-polar north-south outflow from the central massiveprotostar MM1b as traced by both HDO and CS emis-sion. A comparison to archival
Herschel
HIFI data ofthe source shows the power of spatially resolving under-lying substructure with a beam size well-matched to the source, resulting in the unambiguous identification ofCH(O)CH OH. A wealth of additional transitions sug-gest the presence of additional complex molecules thatcan be identified once high resolution laboratory dataare available.The authors thank the anonymous referee for theircareful evaluation which improved the quality of thismanuscript. This paper makes use of the follow-ing ALMA data: ADS/JAO.ALMA
Astropy Collaboration, Robitaille, T. P., Tollerud, E. J.,et al. 2013, A&A, 558, A33Belloche, A., Garrod, R. T., M¨uller, H. S. P., & Menten,K. M. 2014, Science, 345, 1584Benz, A. O., Bruderer, S., van Dishoeck, E. F., et al. 2010,Astronomy & Astrophysics, 521, L35Bøgelund, E. G., McGuire, B. A., Ligterink, N. F. W.,et al. 2018, A&A, 615, A88Brogan, C. L., Hunter, T. R., Cyganowski, C. J., et al.2016, ApJ, 832, 1Brogan, C. L., Hunter, T. R., Cyganowski, C. J., et al.2018, ApJ, submittedCarroll, P. B., Drouin, B. J., & Widicus Weaver, S. L. 2010,The Astrophysical Journal, 723, 845 Ceccarelli, C., Bacmann, A., Boogert, A., et al. 2010, A&A,521, L22Chibueze, J. O., Omodaka, T., Handa, T., et al. 2014, ApJ,784, 114Comito, C., Schilke, P., Gerin, M., et al. 2003, A&A, 402,635Crockett, N. R., Bergin, E. A., Neill, J. L., et al. 2014, ApJ,787, 112de Graauw, T., Helmich, F. P., Phillips, T. G., et al. 2010,Astronomy & Astrophysics, 518, L6De Luca, M., Gupta, H., Neufeld, D., et al. 2012,Astrophysical Journal Letters, 751, L37Emprechtinger, M., Lis, D. C., Rolffs, R., et al. 2013, ApJ,765, 61 Fayolle, E. C., ¨Oberg, K. I., Jørgensen, J. K., et al. 2017,Nature Astronomy, 1Fedoseev, G., Cuppen, H. M., Ioppolo, S., Lamberts, T., &Linnartz, H. 2015, Monthly Notices of the RoyalAstronomical Society, 448, 1288Hollenbach, D., Elitzur, M., & McKee, C. F. 2013, ApJ,773, 70Hunter, T. R., Brogan, C. L., MacLeod, G. C., et al. 2018,ApJ, 854, 170Jørgensen, J. K., Favre, C., Bisschop, S. E., et al. 2012, TheAstrophysical Journal Letters, 757, L4Lis, D. C., Pearson, J. C., Neufeld, D. A., et al. 2010, A&A,521, L9MacLeod, G. C., Smits, D. P., Goedhart, S., et al. 2018,Monthly Notices of the Royal Astronomical Society, 478,1077McGuire, B. A., Shingledecker, C. N., Willis, E. R., et al.2017, The Astrophysical Journal Letters, 851, L46Melnick, G. J., Menten, K. M., Phillips, T. G., & Hunter,T. 1993, ApJL, 416, L37Melnick, G. J., Ashby, M. L. N., Plume, R., et al. 2000,ApJL, 539, L87Messer, J. K., & De Lucia, F. C. 1984, J Mol Spectrosc,105, 139 M¨uller, H. S. P., Schl¨oder, F., Stutzki, J., & Winnewisser,G. 2005, Journal of Molecular Structure, 742, 215Neill, J. L., Wang, S., Bergin, E. A., et al. 2013, ApJ, 770,142Neill, J. L., Bergin, E. A., Lis, D. C., et al. 2014, TheAstrophysical Journal, 789, 8Neufeld, D. A., Snell, R. L., Ashby, M. L. N., et al. 2000,ApJL, 539, L107Ossenkopf, V., M¨uller, H. S. P., Lis, D. C., et al. 2010,A&A, 518, L111Pilbratt, G. L., Riedinger, J. R., Passvogel, T., et al. 2010,Astronomy & Astrophysics, 518, L1Reid, M. J., Menten, K. M., Brunthaler, A., et al. 2014,The Astrophysical Journal, 783, 130Requena-Torres, M. A., Martin-Pintado, J.,Rodr´ıguez-Franco, A., et al. 2006, A&A, 455, 971Russeil, D., Zavagno, A., Motte, F., et al. 2010, Astronomy& Astrophysics, 515, A55San Jos´e-Garc´ıa, I., Mottram, J. C., van Dishoeck, E. F.,et al. 2015, Astronomy & Astrophysics, 585, A103Turner, B. E. 1991, ApJS, 76, 617van der Tak, F. F. S., Chavarr´ıa, L., Herpin, F., et al. 2013,Astronomy & Astrophysics, 554, A83Zernickel, A., Schilke, P., Schmiedeke, A., et al. 2012,Astronomy & Astrophysics, 546, A87 A. BAND 4 OBSERVATIONSThe salient parameters for the Cycle 5 Band 4 obser-vations are given in Table A1 and will be discussed indetail in a forthcoming paper by Brogan et al.
Table A1.
Observing parameters for Band 4 ALMA data
Parameter Band 4 (2.1 mm)Project Code ALMA 2017.1.00661.SObservation date(s) 2017 Dec 3, 7; 2018, Jan 4Configuration C43-6Time on Source (minutes) 170FWHM Primary Beam 0 . (cid:48) Polarization products dual linearGain calibrator J1713-3418Bandpass calibrator J1617-5848Flux calibrator J1617-5848Spectral window center freqs. (GHz) 130.5, 131.5, 144.5, 145.4Spectral window Bandwidth (MHz) 4 × . − ) 1.1 km s − Robust parameter 0.5Ang. Res. ( (cid:48)(cid:48) × (cid:48)(cid:48) (P.A. ◦ )) 0 . × . − . − * km s −1