The organic chemistry in the innermost, infalling envelope of the Class 0 protostar L483
Steffen K. Jacobsen, Jes K. Jørgensen, James Di Francesco, Neal J. Evans II, Minho Choi, Jeong-Eun Lee
AAstronomy & Astrophysics manuscript no. L483_coms © ESO 2019May 7, 2019
The organic chemistry in the innermost, infalling envelope of theClass 0 protostar L483
Ste ff en K. Jacobsen , Jes K. Jørgensen , James Di Francesco , Neal J. Evans II , , , Minho Choi , and Jeong-Eun Lee Niels Bohr Institute & Centre for Star and Planet Formation, University of Copenhagen, Øster Voldgade 5–7, DK-1350 CopenhagenK., Denmark. NRC Herzberg Astronomy and Astrophysics, 5071 West Saanich Road, Victoria, BC, V9E 2E7, Canada. Department of Astronomy, The University of Texas at Austin, Austin, TX 78712, USA. Humanitas College, Global Campus, Kyung Hee University, Yongin-shi 17104, Korea. Korea Astronomy and Space Science Institute, 776 Daedeokdae-ro, Yuseong-gu, Daejeon 34055, Korea. School of Space Research, Kyung Hee University, 1732, Deogyeong-Daero, Giheung-gu, Yongin-shi, Gyunggi-do 17104, Korea.Received 12 / / / / ABSTRACT
Context.
Observations of the innermost regions of deeply embedded protostellar cores have revealed complicated physical structuresas well as a rich chemistry with the existence of complex organic molecules. The protostellar envelopes, outflow and large-scalechemistry of Class 0 and Class I objects have been well-studied, but while previous works have hinted at or found a few Kepleriandisks at the Class 0 stage, it remains to be seen if their presence in this early stage is the norm. Likewise, while complex organics havebeen detected toward some Class 0 objects, their distribution is unknown as they could reside in the hottest parts of the envelope, inthe emerging disk itself or in other components of the protostellar system, such as shocked regions related to outflows.
Aims.
In this work, we aim to address two related issues regarding protostars: when rotationally supported disks form around deeplyembedded protostars and where complex organic molecules reside in such objects. We wish to observe and constrain the velocityprofile of the gas kinematics near the central protostar and determine whether Keplerian motion or an infalling-rotating collapseunder angular momentum conservation best explains the observations. The distribution of the complex organic molecules are used toinvestigate whether they are associated with the hot inner envelope or a possible Keplerian disk.
Methods.
We observed the deeply embedded protostar, L483, using Atacama Large Millimeter / submillimeter Array (ALMA) Band 7data from Cycles 1 and 3 with a high angular resolution down to ∼ (cid:48)(cid:48) (20 au) scales. We present new HCN J = + J = J = CN J = OH, CH OCHO, C H OH, NH CHO, and other species.
Results.
We find that the kinematics of CS J = CN J = OCHO, is consistent with the infall velocity profile derived from CS J = CN J = ∼
50 au, suggesting thatthe complex organics exists in the hot corino of L483, where the molecules sublimate o ff the dust grain ice-mantles and are injectedinto the gas phase. Conclusions.
We find that L483 does not harbor a Keplerian disk down to at least 15 au in radius. Instead, the innermost regions ofL483 are undergoing a rotating collapse, with the complex organics existing in a hot corino with a radius of ∼ Key words. radiative transfer modeling – stars: formation – stars: protostars – ISM: individual (L483) – astrochemistry – Submil-limeter: ISM
1. Introduction
Low-mass stars like our Sun are formed from the gravitationalcollapse of a dense core within a cold molecular cloud. The in-herent rotation of the cloud necessitates the presence of outflowsand jets in the system to transport angular momentum away andlet the central protostar grow in mass. A protostellar disk willemerge around the growing protostar, due to the conservation ofangular momentum of material not lost from the system (Tere-bey et al. 1984). Another angular momentum loss mechanism,strong magnetic braking, however, can prevent a disk from being
Send o ff print requests to : Ste ff en Kjær Jacobsen, e-mail: [email protected] formed altogether (Garcia 2011). If a disk-like structure is able toform, it will eventually become rotationally supported, resultingin a Keplerian disk. Also, in the innermost, warmest part of theenvelope, a rich chemistry should take place, with the sublima-tion of icy dust grain mantles leading to the presence of complexorganic molecules in the gas phase (e.g., Herbst & van Dishoeck2009). These molecules may end up becoming part of the as-sembling circumstellar disk and thus incorporated into eventualplanetary systems. It is therefore interesting to investigate thelink between the physical and chemical structures of inner en-velopes and emerging circumstellar disks, a topic where the At-acama Large Millimeter / submillimeter Array (ALMA) with itshigh angular resolution and sensitivity is ideally suited to make Article number, page 1 of 19 a r X i v : . [ a s t r o - ph . S R ] M a y & A proofs: manuscript no. L483_coms significant contributions. This paper presents observations downto a radius of 10 au of the Class 0 protostar in the isolated coreLynds 483, with the aim of studying its chemistry and using itskinematical structure to shed light on its physical structure onthese scales.Concerted e ff orts have been made over some time onobserving the innermost regions of deeply embedded protostarsin the Class 0 and Class I stage (e.g., Hogerheijde et al. 1998;Looney et al. 2000; Jørgensen et al. 2004, 2007, 2009; Enochet al. 2011), revealing excess compact dust emission thatcould be early disk-like structures or rotationally supporteddisks. The early evolution and exact formation time of theseearliest Keplerian disks are not well-established, however, dueto the di ffi culty of disentangling cloud and disk emission ininterferometric observations, the low number of known Class 0Keplerian disks and the unknown number of Class 0 objectslacking a rotationally supported disk. With the advent of highangular resolution interferometers such as ALMA, discoveringearly, relatively small disks has become feasible. Kepleriandisks are observed around Class I objects on ∼
100 au scales(Brinch et al. 2007; Jørgensen et al. 2009; Harsono et al. 2014)as well as around some Class 0 objects: for example, theClass 0 / I protostar L1527 is found to have a Keplerian disk witha radius of 50–90 au (Tobin et al. 2012; Ohashi et al. 2014).Also, Murillo et al. (2013) detect a disk around the Class 0protostar VLA1623 that is rotationally supported with a Ke-plerian profile out to at least 150 au, Lindberg et al. (2014)report a 50 au Keplerian disk around the Class 0 / I protostarR CrA-IR7B, and Codella et al. (2014) make a tentative de-tection of a 90 au Keplerian disk around the Class 0 protostarHH212. On the other hand, the Class 0 object B335 is shownto lack an observable Keplerian disk down to a radius of 10 au(Yen et al. 2015) and continuum emission in the innermostregion of B335 is consistent with only a very small disk mass(Evans et al. 2015). Due to the uncertain nature of some ofthese Keplerian disk detections and the small sample size, moredetections of rotationally supported disks in the earliest stages,or equivalently, non-detections and upper limits to the sizesof disks around Class 0 objects, are needed to constrain diskformation theories.Concurrently with the investigation of Class 0 and I disks,hot regions in the innermost parts of envelopes hosting low-mass star formation have been observed. Such regions have beenlinked to the formation of Complex Organic Molecules (COMs).Called a ’hot corino’ in the case of a low-mass star, these re-gions of hot gas, T > ff erent molecules sublimate and the moleculesare released into the gas phase, in which COMs and prebioticmolecules have been discovered (e.g., Bottinelli et al. 2004; Jør-gensen et al. 2005, 2012; Coutens et al. 2015; Taquet et al. 2015).The presence of COMs has also been linked to the transition re-gion between the outer infalling-rotating envelope and the cen-trifugal barrier, i.e., the radius where the kinetic energy of theinfalling material is converted into rotational energy (Sakai et al.2014). Accretion shocks and other heating events in this tran-sition zone are hypothesized to induce a chemical change (e.g.,Sakai et al. 2014; Oya et al. 2016). From an astrochemical pointof view, mapping the molecular inventory and distribution at thisearly stage of the disk, or even before the disk is formed, will setthe stage for subsequent chemical evolution, all the way up tothe more complex, prebiotic molecules.An interesting object for addressing these issues is the densecore, Lynds 483 (L483), constituting the envelope around the Class 0 infrared source IRAS 18148-0440 . Traditionally, L483has been associated with the Aquila Rifts region at a distanceof 200 pc (Dame & Thaddeus 1985). Recently the distance toAquila has been revised upward to 436 ± / Aquila. We therefore adoptthe previous distance estimate of 200 pc in this paper. At thisdistance, the bolometric luminosity of L483 is 10–14 L (cid:12) (Laddet al. 1991; Tafalla et al. 2000) making it one of the more lumi-nous solar-type protostars and a good target for chemical studies.L483 drives a well-collimated bipolar CO outflow (Parker1988; Parker et al. 1991; Fuller et al. 1995; Bontemps et al. 1996;Hatchell et al. 1999). Also, it is associated with a variable H Omaser (Xiang & Turner 1995) and shocked H emission, which issuggested to originate from the head and edges of the jet where itinteracts with ambient molecular gas (Fuller et al. 1995). Chap-man et al. (2013) find the position angle of a suggested magneticpseudodisk to be 36 ◦ based on 4.5 µ m Spitzer imaging, while theoutflow position angle is estimated at 105 ◦ , based on the shockedH emission. Fuller et al. (1995) found the outflow inclination ofL483 to be 40 ◦ based on 2.22 µ m imaging, while analysis by Oyaet al. (2018) of the CS and CCH line emission associated withthe outflows found the outflow inclination angle to be between75 ◦ and 90 ◦ , that is, nearly perpendicular to the line-of-sight.In terms of its spectral energy distribution (SED) and envelopemass (4.4 M (cid:12) ; Jørgensen 2004) L483 appears as a deeply em-bedded Class 0 protostar. However, Tafalla et al. (2000) find thatits bipolar outflow has characteristics seen in both Class 0 andClass I objects, and therefore propose that L483 is in transitionfrom Class 0 to Class I.Shirley et al. (2000) made 450 µ m and 850 µ m contin-uum maps of L483 with SCUBA at the JCMT, revealing itselongated continuum emission in the outflow direction, likelythe outer parts of the envelope being swept up by the out-flowing material. Jørgensen (2004) find that the velocity gradi-ents in HCN, CS, and N H + around the source are perpendicu-lar to its outflows, indicative of a large-scale, infalling-rotatingenvelope, with the velocity vector being consistent with rota-tion around a central object of ∼ (cid:12) . Curiously, the inter-ferometric flux of L483 is consistent with envelope-only emis-sion and does not require a central compact emission source(Jørgensen et al. 2007, 2009; Jørgensen 2004). Oya et al. (2017)use a rotating collapse ballistic model and find that a 0.1–0.2M (cid:12) central protostar with a collapsing-rotating envelope with acentrifugal barrier radius (where the barrier radius is half of thecentrifugal radius) of 30–200 au, assuming an inclination an-gle of 80 ◦ , can roughly explain the observed CS, SO, HNCO,NH CHO, and HCOOCH lines. Oya et al. (2017) suggest thatsome molecular species observed towards L483 may be in aKeplerian disk, very near the protostar, but it remains unclearwhether the COMs are more directly linked to a Keplerian diskor to a hot corino region, which may not contain a Kepleriandisk.In this work, we use high angular resolution from ALMACycles 1 and 3 to image the distribution of COMs as well as toprobe the kinematics of the innermost regions down to a radiusof ∼
10 au. These data enable us to improve our understanding of In the literature both the core and the infrared source are referred toas L483, which we follow for the remainder of the paperArticle number, page 2 of 19te ff en K. Jacobsen et al.: The organic chemistry in the innermost, infalling envelope of the Class 0 protostar L483 disk formation and the early astrochemistry of low-mass proto-stars. This paper is structured as follows: First, the observationsare described in Section 2, while the results are presented in Sec-tion 3. An analysis of the inner region kinematics is presented,first for H CN J = J =
2. Observations
L483 was observed on the nights of 2013 June 1, 2013 June 19,2013 November 2, and 2013 November 3 in ALMA Band 7as part of Cycle 1 observations (PI. N. J. Evans II, projec-tid: 2012 . . . . α = h m s , δ = -04 ◦ (cid:48) (cid:48)(cid:48) , with total integra-tion times of 1.8 hours and 3.6 hours for Cycle 1 and Cycle 3,respectively. The Cycle 1 observations used 44 12-m antennas,with baselines in the range of 20–600 k λ , while the Cycle 3 ob-servations used either 38 or 39 12-m antennas with baselines inthe range of 15–1800 k λ .For Cycle 1, L483 was observed on 2013 June 01 withJ1733–1304 as the phase and flux calibrator and J1924-2914as the bandpass calibrator. For 2013 June 19, J1733-1304 wasthe phase, flux, and bandpass calibrator. For 2013 November2 and 3, J1733–1304 was the phase calibrator and J1924–2914was the flux and bandpass calibrator. For Cycle 3, L483 was ob-served on 2016 August 31 with J1924-2914 as the bandpass andflux calibrator, and J1743-0350 as the phase calibrator. For 2016September 7 and 9, J1751 + CASA v. 4.7. Before combination of the datasets, the Cycle 1data were binned down with 4 channels in each bin to match theCycle 3 channel width, as the Cycle 1 observations had higherspectral resolution than those of Cycle 3. Also, the Cycle 3 datawere trimmed at the spectral window edges, to match the Cycle 1bandwidth (see Table 1 for spectral window details).Phase self-calibration was also performed on the continuumchannels in each dataset before combination. After concatena-tion of the Cycles 1 and 3 data, the continuum was constructedusing line-free channels and subtracted from the line emissioncubes. After primary beam correction, both line emission chan-nel images and the 857 µ m continuum image were created withthe clean algorithm using Briggs weighting with a robust pa-rameter of 0.5, to get a good trade-o ff between sensitivity andangular resolution.An 857 µ m continuum image and line emission cubes werealso constructed using phase self-calibrated Cycle 3 data alone,to investigate the spectrum of the broader bandwidth in the Cy-cle 3 observations, and to investigate the spatial distribution ofCOMs on the smallest scales (Table 1).
3. Results
Fig. 1 shows the combined dust continuum image at 857 µ m, re-vealing concentrated emission with a deconvolved 2D Gaussianfit of 0.36 (cid:48)(cid:48) × (cid:48)(cid:48) at a position angle of 104 ◦ . The dust is elon-gated in the East-West direction, possibly caused by the outflows Fig. 1: Cycles 1 and 3 combined dataset continuum emission at857 µ m in logscale. The coordinates are centered on the con-tinuum center at α = h m s , δ = -04 ◦ (cid:48) (cid:48)(cid:48) .Contours are spaced logarithmically between 5–100 % of thepeak emission, in 12 steps. The box marks the region used forthe continuum analysis in Section 4.2.dragging material with it outwards. We estimate the integrateddeconvolved dust continuum of the combined Cycle 1 + µ m (Fig. 1) to be 68.6 ± . (cid:48)(cid:48) × . (cid:48)(cid:48) ,with a position angle of 156 ◦ and an integrated flux density of53 . ± . . λ ), but lower than a cir-cular fit to the SMA continuum visibilities (integrated flux of 0.2Jy) an indication that the ALMA observations filter out some ofthe larger-scale emission.Fig. 2 shows emission from the four major molecule tran-sitions in the observed spectral windows, H CN J = J = J = + J = CN J = J = J = + J = PVEXTRACTOR , with a path width of0.05 (cid:48)(cid:48) , where the o ff set is defined by the distance to the rota-tion axis (Fig. 4 and Fig. 5). The PV diagrams show that onlyH CN J = J = + J = v = ± − in CS J = J = + J = J = Article number, page 3 of 19 & A proofs: manuscript no. L483_coms
Table 1: Observational spectral windows.ID Frequency range [GHz] rms [mJy] Channelwidth [MHz] Synthesized beam
Combined Cycle 1 and 3 data (cid:48)(cid:48) × . (cid:48)(cid:48) .
215 – 345.450 5 0.244 0.28 (cid:48)(cid:48) × . (cid:48)(cid:48) (cid:48)(cid:48) × . (cid:48)(cid:48) (cid:48)(cid:48) × . (cid:48)(cid:48) Cycle 3 (cid:48)(cid:48) × . (cid:48)(cid:48) (cid:48)(cid:48) × . (cid:48)(cid:48) (cid:48)(cid:48) × . (cid:48)(cid:48) (cid:48)(cid:48) × . (cid:48)(cid:48) Frequency corresponds to v lsr = − . The rms is given as the typical rms in flux beam − in the channels.emission, due to a lack of shorter baselines. While redshifted ab-sorption is seen towards similar Class 0 objects, such as B335(Evans et al. 2015), L483 has an elongated structure, with out-flows at an inclination of 75–90 ◦ and the kinematics of the in-nermost region, where we have extracted our spectrum, are alsodominated by rotational motion, not free-fall (Fig. 2). This com-plex geometry and the kinematics may explain why the inverse PCygni profile with redshifted absorption against the continuum,expected for a spherical collapse, is not present in L483. In addition to the four main species (HCN, H CN, CS, andHCO + ) targeted as part of this program, a multitude of emissionlines are found, belonging to a range of COMs, shown in Fig. 6along with their synthetic line spectra. We investigated COMsin the Cycle 3 spectra only, due to its larger frequency range.Identifying these fainter transitions in the spectra requires care-ful comparison to spectroscopic catalogs, modeling the emissionfrom known species, and comparison to other surveys. To do thiswe calculate synthetic spectra for possible molecules and com-pared those to the data: for a given molecule the spectra are pre-dicted under the assumption of local thermodynamic equilibriumgiven assumptions of the column density, excitation temperature,systemic velocity, line width and source size. Typically besidesthe main (very optically thick) lines the three latter parameterscan be fixed for all lines and species leaving the column densitiesto be constrained.To decide which molecules to assign we compared our L483data directly to those from the ALMA Protostellar Interfero-metric Line Survey (PILS) of the low-mass Class 0 protostel-lar binary IRAS 16293-2422 (Jørgensen et al. 2016). We usedthe spectrum from the B component of the protostellar bi-nary (IRAS16293B in the following). Specifically, in that sur-vey more than 10,000 separate features can be identified to-ward IRAS16293B in a frequency range between 329 GHz and363 GHz. With the large frequency coverage in the PILS datamany species can be well-identified and the column densitiesand excitation temperatures constrained. Given our smaller fre-quency coverage in the observations of L483 the assignmentswould in their own right only be tentative and the inferred col-umn densities mainly a sanity check that the assignments areplausible. The general agreement with the identifications in theIRAS16293B data strengthens this case, however. Table B.2 liststhe assigned transitions, while Table 2 gives the inferred columndensities for L483 and IRAS16293B. For most species, we do not have a su ffi cient number oftransitions to constrain the excitation temperature, except forCH DOH (deuterated methanol) and CH OCHO (methyl for-mate). For the former, an excitation temperature of 100 ±
25 K reproduces the relative line strengths, while the linesof CH OCHO are slightly better fitted with a higher excita-tion temperature of 300 K. This situation is similar to that ofIRAS16293B where a number of species, including methyl for-mate, with binding energies of 5000–7000 K are best fitted witha high excitation temperature of a few hundred K, while otherspecies require a lower excitation temperature of about 100 K.Toward IRAS16293B, optically thin transitions of methanol isalso best fitted with an excitation temperature of 300 K, but to-ward that source a colder component is also present as witnessedby extended emission in a number of lower excited transitions aswell as the low temperatures of highly optically thick transitions.Toward L483, a number of the stronger lines of CH DOH withlow upper energy levels ( ∼
100 K) are marginally optically thickwith τ of 0.1–0.5. Thus, it is plausible that a still higher tem-perature component with a high column density may be presenton even smaller scales, not traced by the lines identified here.Contrary to the methanol isotopologue transitions, most of themethyl formate lines have low opacities of 0.01–0.05, and thusare very likely sampling the most compact, high column densitymaterial. Also, it should be noted that the inferred column den-sities are only weakly dependent on the exact excitation temper-ature, changing by less than 10–20% with temperatures varyingfrom about 100 K to 300 K. For the purpose of this paper, theseuncertainties are less critical. Article number, page 4 of 19te ff en K. Jacobsen et al.: The organic chemistry in the innermost, infalling envelope of the Class 0 protostar L483 Fig. 2: Moment 0 and 1 maps of H CN J = J = J = + J = − km s − for H CN J = J = J = + J = − km s − . All the data shown in this figure are fromthe combined dataset. Article number, page 5 of 19 & A proofs: manuscript no. L483_coms
Fig. 3: PV diagrams of H CN J = J = J = + J = v lsr = . − . Contours arespaced linearly between 5–100 % of the peak emission, in five steps. The PV diagrams are made along the direction of the velocityvector and the o ff set is the distance to the rotation axis (Fig. 4). The emission center, as defined in Section 4, is the intersection ofthe rotation axis and velocity vector. v lsr = . − , derived in Section 4, matches the H CN J = J = J = + J = − (Hatchell et al. 1999).All the data shown in this figure are from the combined dataset. Article number, page 6 of 19te ff en K. Jacobsen et al.: The organic chemistry in the innermost, infalling envelope of the Class 0 protostar L483 Fig. 4: H CN J = imfit datapoints superimposed on the H CN J = σ uncertainty of the peakposition calculated by imfit . Green and black contours show the H CN J = µ m continuum emis-sion, respectively, both spaced logarithmically between 5–100 % of the peak emission, in 10 steps. The inferred rotation axis ofH CN J = imfit routineand also shows the region from which the spectra in Fig. 6 were extracted. All the data shown in this figure are from the Cycle 3dataset.Fig. 5: CS J = imfit datapoints superimposed on the CS J = σ uncertainty of the peak positioncalculated by imfit . Green and black contours show the CS J = µ m continuum emission, respectively,both spaced logarithmically between 5–100 % of the peak emission, in 10 steps. The velocity vector (brown) and rotation axis (grey)was not fitted to the CS J = CN J = imfit routine and alsoshows the region from which the spectra in Fig. 6 were extracted. All the data shown in this figure are from the Cycle 3 dataset. Article number, page 7 of 19 & A proofs: manuscript no. L483_coms(a) Spectral window centered on H CN J = J = J = + J = Fig. 6: Cycle 3 spectral windows centered on H CN J = J = J = + J = v lsr = . − . The line model fits are overlaid in colored lines(red = CH DOH, blue = CH OCHO, purple = C H OH, pink = CH SH, green = all other species), while the observed spectrumis in gray. The top frames show the corresponding IRAS16293B spectrum. The region from which the spectra was extracted can beseen in Fig. 4 and 5. Article number, page 8 of 19te ff en K. Jacobsen et al.: The organic chemistry in the innermost, infalling envelope of the Class 0 protostar L483 We constructed moment 0 and 1 maps of each individualidentified COM transition in Fig. 7, integrating all emissionwithin ± − based on Fig. 8. A range of COM moleculesdemonstrate line emission with clear velocity gradients similarto those seen in the four main species. Deconvolved 2D Gaus-sian fits to the moment 0 maps of the selected COMs reveal thatmost molecules are extended compared to the continuum peakwith deconvolved sizes of approximately 0.2–0.3 (cid:48)(cid:48) (40–60 au). Methanol (CH OH) is the most prominent organic moleculeidentified on small scales towards solar-type protostars with dif-ferent rarer isotopologues typically possible for identification.In our frequency range, the main lines of CH OH are of the two isotopologues, CH OH and CH DOH. CH DOH, in par-ticular, has six transitions in the HCO + spectral window andcover two others in the H CN and CS windows with three sep-arate transitions of CH OH. In the HCO + window, a promi-nent feature is seen at 356.625 GHz. The best option for thisline is a set of relatively highly excited transitions of the mainisotopologue of methanol (CH OH 23 − − − ). For the de-rived column density of CH OH, these transitions should in-deed be present at a temperature of 100 K taking into accountthe standard C: C ratio of 68 (Milam et al. 2005). A simi-lar highly excited CH OH transition at 356.875 GHz is blendedwith a transition of CH OH. Methyl mercaptan (CH SH) at356.627 GHz ( E u =
136 K; log A ul = − . − ]) could alsocontribute to this feature. However, unless its excitation wouldbe very peculiar, one would then also expect to see CH SH tran-sitions at 354.372 GHz ( E u =
147 K; log A ul = . − ]) and354.643 GHz ( E u =
146 K; log A ul = − . − ]) in the HCNwindow with approximately the same strengths. CH SH is in-deed detected toward IRAS16293B (Drozdovskaya et al. 2018)in the PILS data with the three transitions at 356.625, 354.372,and 354.643 GHz all clearly seen. As the latter two do not showup toward L483 it appears that CH SH does not contribute at thislevel.A few isolated transitions can be assigned to individualspecies with an assumed excitation temperature of 100 K.These include CH OCH , C H OH, NH CHO, SO , H CS,and HC N. Of these species, the relatively common gas phasemolecules, H CS, SO , and HC N, are found to be relativelymore abundant toward L483 than the complex organics. For theother species, the inferred column densities are in agreementwith those toward IRAS16293B, lending credibility to their as-signments.A few features remain problematic to assign. For example,one feature at 356.52 GHz in the HCO + window could be at-tributed to a few di ff erent species, including ethylene glycol andacetone, but these species would have transitions visible in otherspectral windows. In the PILS data, a feature is also seen at thisfrequency, which is also not easily assigned to any of the tabu-lated species.In the HCN window, the feature at 354.458 GHz is some-what puzzling. By itself, it could be attributed to acetalde-hyde (CH CHO) but this species has a similar transition at354.525 GHz that should be equally strong and in the PILS datathe two transitions in fact show up in this way. This behavior isnoteworthy as acetaldehyde otherwise is considered one of themost easily identifiable of the complex organics, but in our datait can thus only be tentatively identified.In the H CN window, the feature around 345.285 GHz re-mains unassigned. It could be attributed to cyanamide (NH CN)at 345.2869 GHz ( E u =
138 K; log A ul = − . − ]),which would be an excellent fit and was recently identifiedin the PILS data by Coutens et al. (2018). To reproduce theobserved line strength, however, a cyanamide to formamide(NH CN / NH CHO) ratio of 20 would be required, whereas allother interferometric measurements, as well as models, have for-mamide being more abundant than cyanamide by an order ofmagnitude or more. More likely, there remains an issue withspectroscopic predictions for line intensities. For example, twotransitions are seen in PILS data at the same frequencies thatare unassigned: one is likely CH DOH (345.2842 GHz) and theother C H OH (345.2877 GHz). However, the B- and C-typetransitions of deuterated methanol are known to be problematicand previously some issues have been identified for C H OH aswell (Müller et al. 2016). For the feature at 356.546 GHz, the
Article number, page 9 of 19 & A proofs: manuscript no. L483_coms best assignment would be of ethyl cyanide (CH CH CN). Thisspecies does have a relatively bright transition in the HCN win-dow at 354.477 GHz, but due to blending with the HCN transi-tion itself it is not possible to see whether this transition is in-deed present. Another CH CH CN transition is near the HCO + window at 356.960 GHz, but falls just outside of our frequencycoverage. If the transition indeed could be solely attributed toCH CH CN, it would be more abundant by an order of magni-tude relative to C H OH than what is seen toward IRAS16293B(Calcutt et al. 2018). Clearly, more transitions are needed to beobserved of these species for reliable assignments and columndensities.
4. Analysis
To determine whether or not a Keplerian disk is present in L483,we investigated the kinematics of the gas motions in the inner30 au of L483. For this purpose, we fit the position of the peakemission in each spectral cube channel, using the 2D Gaussianfit routine, CASA imfit . H CN J = J = J = + J = J = imfit resultsnear the continuum center in Fig. 5 show a clear velocity pro-file not visibly a ff ected by the outflows in the image maps. Forthis reason we include CS J = imfit to circumvent the large-scale emission seen in the South-East direction (Fig. 2) of H CN J = J = imfit data points are mostlyfound in two “clusters”, with each cluster corresponding to blue-and red-shifted emission, respectively (see Fig. 4 and 5). Wemade a weighted linear regression fit to these clusters of data,which defines the velocity vector, and took the weighted aver-age declination coordinate of each of the two data clusters asinput to the inverse velocity vector function to define the av-erage right ascension coordinate. The midpoint between thesetwo representative points of the data clusters was taken as theemission center, determined to be α = h m ± s , δ = -04 ◦ (cid:48) (cid:48)(cid:48) .595 ± (cid:48)(cid:48) .The rotation axis was then defined as being normal to thevelocity vector, intersecting the emission center, see Fig. 4. Wedefined the velocity vector, rotation axis, and emission centerusing H CN J = J = CN J = − , using the H CN J = J = ff set of the imfit data points, while the data pointdistance from the rotation axis was taken as the o ff set distance.The imfit data points were then converted into (radius, ve-locity) points and used in a reduced χ fit of two di ff erent ve-locity profiles; a Keplerian velocity profile, v = √ GM ∗ / r , where G is the gravitational constant and M ∗ is the central mass, andthe velocity profile of infall with conservation of angular mo-mentum, v φ r = v φ, r , where v φ, and r is the start velocity and starting distance, respectively, of the collapsing material. r canbe interpreted as the starting position of the given material in thecloud, at the time when the collapse started. Such an infall sce-nario will have v φ ∝ r − , a velocity profile previously used toestimate the kinematics of a disk or disk-like structure aroundClass 0 / I objects (e.g., see Lindberg et al. 2014). For the Keple-rian profile, we used a central mass range of 0.01-1.5 M (cid:12) and2 . × uniform steps in the specific angular momentum rangefor the infall velocity profile of 10 – 1.5 × m s − , to findthe best fit. The central mass parameter presents M ∗ sin i , where i is the system inclination, since we only observe radial veloci-ties. Using an inclination between 75–90 ◦ (Oya et al. 2018), wecan in principle extract an estimate of the true stellar mass fromthe best fit central stellar mass. However, the paramount goal isto distinguish between the two velocity profiles, not to obtain aprecise measure of the stellar mass.We combined the H CN J = J = χ fit in the ve-locity regime. We excluded the lowest velocities as these area ff ected by the small extent of the H CN gas line emission( ∼
20 au radius), causing the lower velocities to be dominatedby low-velocity gas toward the emission center, arising from pro-jected gas velocities coming from the edge of the observed gasemission region. These low radial velocities appear at low pro-jected distances, while the true distance to the protostar is un-known. As such, the emission seen at low projected distances inFig. 8 for v < . − , likely arise from larger actual distancesto the protostar. Moreover, emission at these low velocities couldinclude emission from the other disk-half, as thermally broad-ened lines from the other disk-half are convolved with the rel-atively large beam, drawing the imfit results towards a lowero ff set distance. We also excluded higher velocities, where theCS J = imfit results become noisy. The cause of this noiseis unknown, but it could be an e ff ect of high-velocity outflowingmaterial. The di ff erence in the spatial extents of the observedCS and H CN datapoints and the spatial disparity illustrated inFig. 2, are likely related to the lower critical density of CS rela-tive to HCN (Evans 1999).This approach of using the emission peak position in eachchannel to constrain the gas kinematics, is only exactly validwhen each emission component comes from a single position.If this is a poor approximation of the true velocity structurein each velocity component, then systematic uncertainties willbe introduced. However, given the systematic and concentratedpeak emission positions of the gas line emission velocity com-ponents of CS J = CN J = χ fit, as the imfit result precision is a ff ectednegatively by the larger rms and beam in the combined dataset.The Cycle 3 data have more precise imfit data points, due totheir higher angular resolution of ∼ . (cid:48)(cid:48) × . (cid:48)(cid:48) and lowerrms, while missing the shorter baselines of the combined Cy-cle 1 and 3 data. Consequently, spatial filtering is seen in theCycle 3-only continuum emission in Fig. 4 and Fig. 5 when com-pared to the Cycle 1 + ff erent beam-sizes. Fig. 8 shows that,using Cycle 3–only data, the infall profile is strongly favored bythe reduced χ fit, with χ = χ = .
6. Using the combined dataset gave the same con-clusion: the infall profile is heavily favored, with χ better bymore than a factor of five. This result suggests that the observed Article number, page 10 of 19te ff en K. Jacobsen et al.: The organic chemistry in the innermost, infalling envelope of the Class 0 protostar L483 Table 2: Inferred column densities and fractional abundances relative to CH OH for the species identified toward L483 and comparedto the one beam o ff set position from IRAS16293B (Jørgensen et al. 2016).Species Column density [cm − ] [X / CH OH]L483 IRAS16293B L483 IRAS16293BMethanol CH OH 1 . × × CH OH 2 . × × × − × − CH DOH 4 . × × × − × − Dimethyl ether CH OCH . × × × − × − Methyl formate CH OCHO 1 . × × × − × − Ethanol C H OH 1 . × × × − × − Acetaldehyde CH CHO 8 . × × × − × − Formamide NH CHO 1 . × × × − × − Cyanopolyyne HC N 5 . × × × − × − Thioformaldehyde H CS 2 . × × × − × − Sulfur-dioxide SO . × × × − × − line emission is not from gas in a rotationally supported disk,but rather from gas in an infalling-rotating structure. While theCS J = J = χ fitwas also made on the combined and Cycle 3 data using onlyH CN J = CN J = imfit data near5 . − , at ∼
15 au, so we performed an independent fit toH CN J = χ fit slightly favored a Keplerian velocity profilewith χ = .
71 vs. an infall velocity profile with χ = .
85. How-ever, we cannot conclude the presence of a Keplerian disk witha radius of 15 au, since the data is too noisy and sparse, andsince an infall velocity profile could fit the data as well. The ab-sence of a Keplerian disk down to at least 15 au is consistent withthe analysis of the submillimeter continuum emission toward thesource (Jørgensen 2004; Jørgensen et al. 2007, 2009) where theinterferometric flux of L483 was consistent with envelope-onlyemission and did not need a central compact emission source.
In order to investigate the dust temperature profile of L483 onsmall scales, we used the density solution to an infalling-rotatingcollapse (Terebey et al. 1984), with an example centrifugal ra-dius of 60 au (consistent with the conclusions of Oya et al. 2017).For an initial total dust mass guess, we performed a 2D Gaussianfit on the 857 µ m emission (Fig. 1), extracted from a box aroundthe elongated emission structure and found a deconvolved size of0 . (cid:48)(cid:48) × . (cid:48)(cid:48) . The box was chosen to avoid the elongated emis-sion in the East-West direction, and instead focus on the inner60–80 au, where the bulk emission is present. We approximatethe deconvolved fit to a circular region, with an extent of 0.28 (cid:48)(cid:48) ,i.e., we approximate the observed dust continuum to a sphericalmodel of 56 au radius. While the 857 µ m dust continuum tracesmaterial swept up in the outflow structure, we do not attempt tomodel the outflow or outflow cavities. Using the available SEDdata (Table B.1), we integrated the SED and estimate the bolo-metric luminosity to be 10.5 L (cid:12) , comparable with previous lumi- nosity estimations of 9 L (cid:12) (Jørgensen et al. 2002) and 13 ± (cid:12) (Shirley et al. 2000).The total dust mass is given by M = S ν d κ ν B ν ( T ) , (1)where S ν is the total source flux, d is the distance, κ ν is thedust opacity, and B ν ( T ) is the spectral radiance. Both κ ν and B ν ( T ) depend on the temperature field as the mean dust opac-ity κ ν will be a mixture of dust with and without ice-mantlesdue to sublimation caused by heating from the central pro-tostar. We used bare-grain and thin ice-mantle opacities fromOssenkopf & Henning (1994), corresponding to coagulated dustgrains in an environment with a gas number density of 10 cm − .We used initial guesses of T av =
100 K, the mean dust tempera-ture of all dust, both with and without ice mantles, within 56 au,and a dust population ratio between icy-dust and bare-grain dustof 0.5, also within 56 au, to get an initial estimate of the totaldust mass, in the innermost region of L483, using Eq. 1.With a dust mass estimate as input, we used
RADMC-3D , a 3DMonte Carlo radiative transfer code (Dullemond et al. 2012), todetermine the dust temperature, which led to a new mass estima-tion, as T av and κ ν change (Eq. 1). The updated mass in turn ledto a di ff erent temperature distribution, which again a ff ects ourestimate of κ ν and T av . After a few iterations, we had a stable es-timate of all parameters, with T av =
125 K from visual inspectionof the temperature distribution within 56 au (Fig. 9). Almost allthe dust within 56 au has temperatures above 90–100 K, leadingus to adopt bare-grain dust opacities exclusively, as the water-icemantle sublimates at these temperatures (Sandford & Allaman-dola 1993). The final temperature profile can be seen in Fig. 9and we estimate the total mass (dust + gas) in the inner region tobe 8 . × − M (cid:12) , using a gas-to-dust mass ratio of 100. Whilethe exact temperature distribution and derived total mass usingEq. 1 is dependent on the used dust density model and its param-eters, we have used a dust density model consistent with bothour observed kinematics (the gas kinematics of the model has a v φ ∝ r − profile) and the earlier research of Oya et al. (2017). Wealso re-performed the continuum analysis and radiative transfermodeling using opacities of bare-grains coagulated with highergas number densities of 10 and 10 cm − , which showed a con-sistent, but minor drop in sublimation radius as the ambient gasdensity of the coagulated bare-grain opacities increased. The 44-52 au sublimation radius of the 10 cm − gas density model isdecreased to a 41-49 au sublimation radius in the 10 cm − gas Article number, page 11 of 19 & A proofs: manuscript no. L483_coms
Fig. 7: Moment 0 and 1 maps of observed COMs (Table B.2). Moment 0 maps are in green contours, overlaid on the moment 1map. The Cycle 3 857 µ m dust continuum image is shown in the lower right frame, in gray contours, and the beamsize of the dustcontinuum observation is shown in the lower left corner of the same frame. Both dust continuum and the moment 0 map contoursare spaced logarithmically between 5–100 % of the peak emission, in 10 steps. The first two panels show the unblended lines ofCH OH and CH OCHO, while the remainder are blended to di ff erent degrees. The beamsizes of the molecule observations areshown in the lower right corner of each frame. The peak flux of the observed transition, F peak , is given in each panel, while the meanrms is 0.016 Jy Beam − km s − . All the data shown in this figure are from Cycle 3.density model. Since changing the model opacity also reducesthe required amount of material in the model (Eq. 1), the totaloptical depth of the model is not significantly changed, and thesublimation radius is therefore not dramatically a ff ected by dif-ferent opacity models.The spatial extents of the COMs (Fig. 7) were fitted with 2DGaussian profiles, from which the deconvolved major and minoraxes were found. The mean major axis is 0.2 (cid:48)(cid:48) with a 2 σ devia-tion of 0.1 (cid:48)(cid:48) . The maximal spatial extents of the COMs emission(0.2–0.3 (cid:48)(cid:48) , i.e., 40–60 au) are consistent with the estimated ice-mantle sublimation front of ∼
50 au (Fig. 9), implying that the COMs reside in the hot corino, which is dominated by rotationalmotion.Figure 7 shows that rotation profiles are also observed forall strong COM lines, with the same North-South velocity gra-dient as seen for CS J = CN J = imfit , but thedata were too noisy and the lines too blended for clear veloc-ity profiles to be extracted, except for CH OCHO. We definedthe emission center and rotation axis of CH OCHO based on its imfit data, as the CH OCHO emission is slightly o ff set com-pared to H CN J = ff set is, however, well within the Article number, page 12 of 19te ff en K. Jacobsen et al.: The organic chemistry in the innermost, infalling envelope of the Class 0 protostar L483 Fig. 8: Distance vs. velocity plot of H CN J = J = imfit datapoints, using Cycle 3 data only, in linear scale(top frame) and logscale (bottom frame). The velocities on the first axis are the velocity o ff set from the employed systemic velocity(6.0 km s − ). The blue- and redshifted imfit datapoints of the CS J = imfit datapoints of H CN J = σ uncertainty. The (a) and (b) regions, with borders marked by vertical dashed lines, show whichdatapoints were used for the fit. Only CS J = CN J = ff ected by the projected velocities from the emission edge. Only H CN J = J = ff ected by outflows in this region. The data diverging from themodel velocity profiles at low velocities, below velocity region (a), arise from the projected rotational gas motions at the edge of theemission region, and are not used in the fit. Some CS J = imfit results are lacking for v < .
25 km s − due to heavy, globalabsorption at low velocities. Both frames are overlaid with the best fit infall (black dashed line) and Keplerian velocity profile (blackfull line) to the Cycle 3 data. The red solid line and the red dashed line show the best fit to the H CN J = imfit datapointswith absolute velocities > . − , using a Keplerian and infall velocity profile, respectively.beam. Fig. 10 shows that the velocity profile of CH OCHO isconsistent with the infall profile derived from the CS J = CN J = . (cid:12) vs. 4 M (cid:12) ) and while L483 is more luminous thanIRAS16293B (10 . (cid:12) vs. ∼ (cid:12) ), IRAS16293B is surroundedby massive amounts of disk-like material (Jacobsen et al. 2018),which may explain the comparable amount of material at hightemperatures. The sulfur-species and HC N show higher column densities toward L483 than IRAS16293B, but whether this is achemical e ff ect, or rather reflect di ff erent physical structures onthese scales, is unclear.
5. Discussion
The lack of an observable rotationally supported disk in L483down to ∼
15 au, and the presence of the COMs on scales of40–60 au, have a number of important implications. In terms ofthe system geometry, the position angle of our rotation profilewas 11 ◦ , perpendicular to the outflow in the East-West direc-tion. Our data show that L483 is undergoing a rotating collapse,with a large-scale outflows consistent with earlier works and asmall inner region dominated by rotational motion (Fig. 7). The Article number, page 13 of 19 & A proofs: manuscript no. L483_coms
Fig. 9: Temperature distribution in the density solution to an infalling-rotating collapse model, with r c =
60 au, M ∗ = . (cid:12) , and˙ M = . × − M (cid:12) yr − using only coagulated bare-grain opacities, assuming a gas number density of 10 cm − . The densitycontours in solid lines are logarithmically spaced, increasing towards the midplane (the x-axis), while the dust temperature contoursare in dashed lines.infalling-rotating collapse continues down to at least 15 au in ra-dius, with the outflows of L483 necessarily being launched veryclose to the central protostar, in the absence of a disk >
15 au.L483 shows some similarities with the Class 0 object B335,as both YSOs have a lack of Keplerian disks, >
15 au and >
10 au,respectively, and show very small amounts of dust in the in-nermost region, 8.8 × − M (cid:12) vs. 7.5 × − M (cid:12) , for L483 andB335 (Evans et al. 2015), respectively. Both objects contain ahot corino (Imai et al. 2016), while the L483 hot corino radius of40–60 au is likely larger than that of B335, estimated to be onlya few tens of au (Imai et al. 2016), due to its relatively low lu-minosity of 0.72 L (cid:12) (Evans et al. 2015). B335, however, has lowlevels of rotation in its infall, whereas L483 has a clear infall-rotational signature (Figs. 2 and 7).Oya et al. (2017) invoke a chemical transition at the centrifu-gal barrier of L483, due to the abrupt transition in the physicalenvironment. They also speculate that NH CHO and CH OCHOreside in an unresolved Keplerian disk, explaining the compactemission they observe. Our data of emission from the samemolecules and the absence of a Keplerian disk down to a radiusof at least 15 au, however, illustrate that these species also re-side in the infalling envelope. Also, these COMs, together withthe distributions of the other COMs, can be accounted for by therelease of molecules into the gas phase due to dust ice-mantlesublimation by itself. They invoked a Keplerian disk model in-side the centrifugal barrier to explain the compact, high-velocityemission structure they observe out to ± − in their PVdiagrams, as their rotating-collapse model alone could not ex-plain this emission together with the more spatially extendedemission. However, they do not resolve the hot corino region(0 . (cid:48)(cid:48) –0 . (cid:48)(cid:48) ) in their observations. We find empirically, using thepeak emission position in each channel, that the compact high-velocity CS J = CN J = ± . − , is best matched by an infalling velocity profile, nota Keplerian one.L483, interestingly, also exhibits a di ff erence to some ofthe more evolved protostars with larger disks. For example,Lindberg et al. (2014) find very small levels of CH OH towardthe Class 0 / I protostar R CrA-IRS7B and the presence of a Ke-plerian disk around this source. Based on a detailed line radia-tive transfer analysis, they demonstrate that this lack of CH OHemission may reflect the low column density of the protostellarenvelope at the scales where material is being assembled into thecircumstellar disk. A similar situation is seen toward the Class Iprotostar Oph-IRS67 in Ophiuchus. For this source, Artur de laVillarmois et al. (2018) find a Keplerian disk to be present andalso do not see signs of any CH OH down to low column den-sities. In contrast, acetaldehyde was found towards the Class 0object HH212 in Orion, where a tentative detection of a ∼
90 auKeplerian disk was made (Lee et al. 2014), though the abun-dance could not be determined due to optically thick submil-limeter continuum emission (Codella et al. 2016). More observa-tions are needed to quantify the relationship between the COMcolumn density and abundance, and the presence of Kepleriandisks.For sources with an extended disk, the mass budget is dom-inated by the disk plane in the inner region. If the only sourceof heating at these scales is the radiation by the newly formedprotostar, a significant amount of the material on small scalesmay be relatively cold, causing molecules to freeze-out ontodust grains and thus lowering the column density of COMs.In contrast, for YSOs with smaller disks, the warm envelopewill dominate the mass budget on these scales. As pointed outby Lindberg et al. (2014), in the picture of a simple rotatingcollapse, an early stage hot corino without a sizable disk canonly exist for a limited time: the radius of the region with
Article number, page 14 of 19te ff en K. Jacobsen et al.: The organic chemistry in the innermost, infalling envelope of the Class 0 protostar L483 Fig. 10: Distance vs. velocity plot of CH OCHO imfit datapoints, using Cycle 3 data only, in linear scale (top frame) and logscale(bottom frame). The velocities on the first axis are the velocity o ff set from the employed systemic velocity (6.0 km s − ). The blue-and redshifted imfit datapoints of the CH OCHO emission are shown in blue and red, respectively. The data diverging fromthe model velocity profiles at low velocities, below ∼ − , arise from the projected rotational gas motions at the edge of theemission region. Both frames are overlaid with the best fit infall (dashed line) and Keplerian velocity profile (full line) to the Cycle 3CS J = CN J = R
100 K ∝ M . ∗ , while the centrifugal radiuswithin which rotational support is greater than the gas pressure(Terebey et al. 1984) grows as r c ∝ M ∗ . Thus, while the COMscan be abundant in the infalling envelope in the early stage whereonly a small Keplerian disk is present, the disk will grow morerapidly than the hot corino region and thus suppress its emis-sion, as the COMs are trapped in ices in the disk midplane.The multitude of COMs in L483 and the lack of a Kepleriandisk down to at least 15 au, may therefore indicate that L483is still in a chemical stage dominated by a warm inner enve-lope. In this picture, assuming that we have a protostar lumi-nous enough to create an observable hot region, it is possible thatthese phenomena could be complementary, i.e., the presence ofhot corino chemistry would signify the presence of a small disk.Similarly, sources with extended disks will show relatively smallamounts of methanol and more complex species present in thegas phase. More observations relating the abundances of COMsto disk sizes, are needed to confirm this picture.
6. Conclusion
We have presented ALMA Cycles 1 and 3 Band 7 high angularresolution ( ∼ . (cid:48)(cid:48) ) observations of HCN J = CN J = J = + J = – We fitted combined ALMA Cycles 1 and 3 observations, aswell as Cycle 3-only observations, of H CN J = J = ∼
15 au radius. – A range of complex organic molecules was observed with thesame rotational signature as H CN J = J = OCHO was extracted that follow the same infall profileas H CN J = J = – The emission of the observed complex organic molecules ex-tends to ∼ ∼
50 au, where the molecules sublimates
Article number, page 15 of 19 & A proofs: manuscript no. L483_coms into the gas phase o ff the dust grains, suggesting that thecomplex organic molecules exists in the hot corino of theL483 envelope. – The lack of a Keplerian disk down to at least a 15 au ra-dius, and the presence of complex organic molecules in theenvelope at ∼ Acknowledgements.
This paper makes use of the following ALMA data:2012.1.00346.S and 2015.1.00377.S. ALMA is a partnership of ESO (repre-senting its member states), NSF (USA) and NINS (Japan), together with NRC(Canada) and NSC and ASIAA (Taiwan) and KASI (Republic of Korea), in co-operation with the Republic of Chile. The Joint ALMA Observatory is oper-ated by ESO, AUI / NRAO and NAOJ. S.K.J. and J.K.J. acknowledges supportfrom the European Research Council (ERC) under the European Union’s Hori-zon 2020 research and innovation programme (grant agreement No. 646908)through ERC Consolidator Grant “S4F”. Research at the Centre for Star andPlanet Formation is funded by the Danish National Research Foundation. Thisresearch has made use of NASA’s Astrophysics Data System. This researchmade use of Astropy, a community-developed core Python package for Astron-omy Astropy Collaboration et al. (2013). This work uses
PVEXTRACTOR , see https://github.com/radio-astro-tools/pvextractor
References
Artur de la Villarmois, E., Kristensen, L. E., Jørgensen, J. K., et al. 2018, A&Ain press. (arXiv:1802.09286)Astropy Collaboration, Robitaille, T. P., Tollerud, E. J., et al. 2013, A&A, 558,A33Bontemps, S., Andre, P., Terebey, S., & Cabrit, S. 1996, A&A, 311, 858Bottinelli, S., Ceccarelli, C., Neri, R., et al. 2004, ApJ, 617, L69Brinch, C., Crapsi, A., Jørgensen, J. K., Hogerheijde, M. R., & Hill, T. 2007,A&A, 475, 915Calcutt, H., Jørgensen, J. K., Müller, H. S. P., et al. 2018, A&A, 616, A90Chapman, N. L., Davidson, J. A., Goldsmith, P. F., et al. 2013, ApJ, 770, 151Codella, C., Cabrit, S., Gueth, F., et al. 2014, A&A, 568, L5Codella, C., Ceccarelli, C., Cabrit, S., et al. 2016, A&A, 586, L3Coutens, A., Persson, M. V., Jørgensen, J. K., Wampfler, S. F., & Lykke, J. M.2015, A&A, 576, A5Coutens, A., Viti, S., Rawlings, J. M. C., et al. 2018, MNRAS, 475, 2016Dame, T. M. & Thaddeus, P. 1985, ApJ, 297, 751Dotson, J. L., Vaillancourt, J. E., Kirby, L., et al. 2010, ApJS, 186, 406Drozdovskaya, M. N., van Dishoeck, E. F., Jørgensen, J. K., et al. 2018, MNRAS,476, 4949Dullemond, C. P., Juhasz, A., Pohl, A., et al. 2012, RADMC-3D: A multi-purpose radiative transfer tool, Astrophysics Source Code LibraryEnoch, M. L., Corder, S., Duchêne, G., et al. 2011, ApJS, 195, 21Evans, II, N. J. 1999, ARA&A, 37, 311Evans, II, N. J., Di Francesco, J., Lee, J.-E., et al. 2015, ApJ, 814, 22Fuller, G. A., Lada, E. A., Masson, C. R., & Myers, P. C. 1995, ApJ, 453, 754Garcia, P. J. V. 2011, Physical Processes in Circumstellar Disks around YoungStarsHarsono, D., Jørgensen, J. K., van Dishoeck, E. F., et al. 2014, A&A, 562, A77Hatchell, J., Fuller, G. A., & Ladd, E. F. 1999, A&A, 344, 687Helou, G. & Walker, D. W., eds. 1988, Infrared astronomical satellite (IRAS)catalogs and atlases. Volume 7: The small scale structure catalog, Vol. 7, 1–265Herbst, E. & van Dishoeck, E. F. 2009, ARA&A, 47, 427 Hogerheijde, M. R., van Dishoeck, E. F., Blake, G. A., & van Langevelde, H. J.1998, ApJ, 502, 315Imai, M., Sakai, N., Oya, Y., et al. 2016, ApJ, 830, L37Jacobsen, S. K., Jørgensen, J. K., van der Wiel, M. H. D., et al. 2018, A&A, 612,A72Jørgensen, J. K. 2004, A&A, 424, 589Jørgensen, J. K., Bourke, T. L., Myers, P. C., et al. 2007, ApJ, 659, 479Jørgensen, J. K., Bourke, T. L., Myers, P. C., et al. 2005, ApJ, 632, 973Jørgensen, J. K., Favre, C., Bisschop, S. E., et al. 2012, ApJ, 757, L4Jørgensen, J. K., Schöier, F. L., & van Dishoeck, E. F. 2002, A&A, 389, 908Jørgensen, J. K., Schöier, F. L., & van Dishoeck, E. F. 2004, A&A, 416, 603Jørgensen, J. K., van der Wiel, M. H. D., Coutens, A., et al. 2016, A&A, 595,A117Jørgensen, J. K., van Dishoeck, E. F., Visser, R., et al. 2009, A&A, 507, 861Ladd, E. F., Adams, F. C., Casey, S., et al. 1991, ApJ, 366, 203Lee, C.-F., Hirano, N., Zhang, Q., et al. 2014, ApJ, 786, 114Lindberg, J. E., Jørgensen, J. K., Brinch, C., et al. 2014, A&A, 566, A74Looney, L. W., Mundy, L. G., & Welch, W. J. 2000, ApJ, 529, 477Milam, S. N., Savage, C., Brewster, M. A., Ziurys, L. M., & Wycko ff , S. 2005,ApJ, 634, 1126Müller, H. S. P., Belloche, A., Xu, L.-H., et al. 2016, A&A, 587, A92Murillo, N. M., Lai, S.-P., Bruderer, S., Harsono, D., & van Dishoeck, E. F. 2013,A&A, 560, A103Ohashi, N., Saigo, K., Aso, Y., et al. 2014, ApJ, 796, 131Ossenkopf, V. & Henning, T. 1994, A&A, 291, 943Oya, Y., Sakai, N., López-Sepulcre, A., et al. 2016, ApJ, 824, 88Oya, Y., Sakai, N., Watanabe, Y., et al. 2017, ApJ, 837, 174Oya, Y., Sakai, N., Watanabe, Y., et al. 2018, ApJ, 863, 72Parker, N. D. 1988, MNRAS, 235, 139Parker, N. D., Padman, R., & Scott, P. F. 1991, MNRAS, 252, 442Sakai, N., Sakai, T., Hirota, T., et al. 2014, Nature, 507, 78Sandford, S. A. & Allamandola, L. J. 1993, ApJ, 417, 815Shirley, Y. L., Evans, II, N. J., Rawlings, J. M. C., & Gregersen, E. M. 2000,ApJS, 131, 249Skrutskie, M. F., Cutri, R. M., Stiening, R., et al. 2006, AJ, 131, 1163Strauss, M. A., Davis, M., Yahil, A., & Huchra, J. P. 1990, ApJ, 361, 49Tafalla, M., Myers, P. C., Mardones, D., & Bachiller, R. 2000, A&A, 359, 967Taquet, V., López-Sepulcre, A., Ceccarelli, C., et al. 2015, ApJ, 804, 81Terebey, S., Shu, F. H., & Cassen, P. 1984, ApJ, 286, 529Tobin, J. J., Hartmann, L., Chiang, H.-F., et al. 2012, Nature, 492, 83Tóth, L. V., Marton, G., Zahorecz, S., et al. 2014, PASJ, 66, 17Xiang, D. & Turner, B. E. 1995, ApJS, 99, 121Yamamura, I., Makiuti, S., Ikeda, N., et al. 2010, VizieR Online Data Catalog,2298Yen, H.-W., Takakuwa, S., Koch, P. M., et al. 2015, ApJ, 812, 129Gaia Collaboration, Brown, A. G. A., Vallenari, A., et al. 2018, A&A, 616, A1Hogerheijde, M. R., van Dishoeck, E. F., Salverda, J. M., & Blake, G. A. 1999,ApJ, 513, 350Kirk, H., Myers, P. C., Bourke, T. L., et al. 2013, ApJ, 766, 115Lindegren, L., Hernández, J., Bombrun, A., et al. 2018, A&A, 616, A2Ortiz-León, G. N., Loinard, L., Dzib, S. A., Kounkel, M., Galli, P. A. B. Tobin,J. J., Evans, N. J. II., Hartmann, L., Rodríguez, L. F., Bricen õ, C., Torres,R. M. & Mioduszewski, A. J., 2018, ApJ, submitted Article number, page 16 of 19te ff en K. Jacobsen et al.: The organic chemistry in the innermost, infalling envelope of the Class 0 protostar L483 Fig. A.1: Extinction vs. distance for stars with reliable paral-laxes from the Gaia-DR2 catalog within 1 ◦ of L483. The verticaldashed lines indicate distances of 200, 225, and 250 pc, wherethe extinction jumps significantly. Appendix A: Distance of L483 from Gaia-DR2measurements
As noted in the introduction, L483 has traditionally been associ-ated with the Aquila Rift / Serpens region toward which it appearsin projection. Recent estimates of the distance toward the larger-scale Serpens and Aquila environment based on VLBA and GaiaData Release 2 (DR2) parallax measurements (Ortiz-Leon et al.2018), place those clouds at a distance of 436 ± − . This suggests that L483 itself is atthis nearer distance of 200–250 pc and thus not associated withthe larger scale Serpens / Aquila cloud material. Also, the LSRvelocities of protostars in those larger-scale regions are foundto be 7.5–8.5 km s − (e.g., Hogerheijde et al. 1999; Kirk et al.2013), another indication that L483 and these clouds are notphysically associated. Appendix B: Figures and tables
Article number, page 17 of 19 & A proofs: manuscript no. L483_coms
Table B.1: SED of L483. a b Tóth et al. (2014), c Yamamura et al. (2010).Wavelength [ µ m] Flux [Jy] Flux uncertainty [Jy] Reference1.25 5.39 × − × − a × − × − a × − × − a × − × − WISE b × − × − WISE b × − × − WISE b × − Strauss et al. (1990)18.39 1.05 0.06 Akari c b b c
90 87.4 4.0 AKARI c
102 166.0 20.0 Helou & Walker (1988)140 150.0 7.0 AKARI c
160 156.0 7.0 AKARI c
350 31.0 6.2 Dotson et al. (2010)450 15.0 2.0 Shirley et al. (2000)800 1.98 0.02 Shirley et al. (2000)1100.0 0.64 0.02 Fuller et al. (1995)
Article number, page 18 of 19te ff en K. Jacobsen et al.: The organic chemistry in the innermost, infalling envelope of the Class 0 protostar L483 Table B.2: List of transitions of identified species predicted to be significant in modeling in Fig. 6.
Species Transition Freq [GHz] log A ul [s − ] E u [K] Main targeted lines
HCN 4–3 354.5055 − + − − CN 4–3 345.3398 − Other assigned lines CH OH 23 − , − − , ∗ − − , − − , ∗ − CH OH 4 , − − , ∗ − , − , ∗ − , − , ∗ − DOH 17 , − , − , − , − , / − , / a − , / − , / ∗ − , − , † − , / − , / † − , − , − , / − , / ∗ − OCH , − , (AE / EE) ) 356.5753 † − , − , (AA) 356.5829 † − , − , (EE / EA) ) 356.5868 † − , − , (EE) 356.7130 ∗ − , − , (EE) 356.7237 † − , − , (AA) 356.7245 † − OCHO 28 , − , (A) ) 345.1480 ∗ − , − , (E) ) 345.2482 − , − , (E) ) 345.4610 † − , − , (A) ) 345.4670 ∗ , † − , − , (E) ) 345.4732 † − , − , (E) ) 345.4866 ∗ − , − , (A) ) 345.5100 ? − , − , (E) ) 354.3487 − / , − / , (E / A) 354.6078 ∗ − , − , (E) 354.6287 − , − , (A) 356.5398 ∗ − , − , (E) 356.5559 − , − , (E) 356.5663 b − , − , (A) 356.6869 ? − , / − , / (A) 356.7118 c − , − , (E) 356.7239 d − , − , (E) 356.7384 a − , − , (E) 356.7770 ? − , − , (E) 356.8582 ? − , / − , / (A) 356.9287 − , − , (E) 356.9363 − , − , (E) 356.9545 − H OH 7 , / − , / , vt = − e − , − , , vt = − − , − , , vt = − − , − , , vt = − f − , − , , vt = − − , − , , vt = − ? − CHO 18 , − , ∗ , g − , − , − CHO 17 , − , ∗ − , − , − , − , c − CS 10 , − , − N 39 −
38 354.6975 ∗ − , − , f − , − , − , − , − Notes: † Blended with nearby transitions of same species (blending taken into account in synthetic spectrum). ∗ Transition shown in Fig. 7. ? Faint emission seen at frequency of transitionbut blend with other unidentified transitions possible. a Blended with HCO + b Blended with set of stronger CH OCH transitions at 356.575–356.587 GHz. c Blended withstronger CH OCH transition at 356.7130 GHz. d Blended with set of stronger CH OCH transitions at 356.723–356.724 GHz. e Blended with stronger NH CHO transition at 345.1813GHz. f Blended with H CN J = g Synthetic spectrum cannot account for full observed line flux. Blending with unassigned line possible (see text). h Blended with HCN J =4–3transition.