Carbon-Chain Chemistry vs. Complex-Organic-Molecule Chemistry in Envelopes around Three Low-Mass Young Stellar Objects in the Perseus Region
Kotomi Taniguchi, Liton Majumdar, Shigehisa Takakuwa, Masao Saito, Dariusz C. Lis, Paul F. Goldsmith, Eric Herbst
aa r X i v : . [ a s t r o - ph . GA ] F e b Draft version February 23, 2021
Typeset using L A TEX twocolumn style in AASTeX63
Carbon-Chain Chemistry vs. Complex-Organic-Molecule Chemistry in Envelopes around ThreeLow-Mass Young Stellar Objects in the Perseus Region
Kotomi Taniguchi, Liton Majumdar, Shigehisa Takakuwa, Masao Saito,
4, 5
Dariusz C. Lis, Paul F. Goldsmith, and Eric Herbst
7, 8 Department of Physics, Faculty of Science, Gakushuin University, Mejiro, Toshima, Tokyo 171-8588, Japan School of Earth and Planetary Sciences, National Institute of Science Education and Research, HBNI, Jatni 752050, Odisha, India Department of Physics and Astronomy, Graduate School of Science and Engineering, Kagoshima University, Korimoto, Kagoshima,Kagoshima 890-0065, Japan National Astronomical Observatory of Japan (NAOJ), National Institutes of Natural Sciences, Osawa, Mitaka, Tokyo 181-8588, Japan Department of Astronomical Science, School of Physical Science, SOKENDAI (The Graduate University for Advanced Studies), Osawa,Mitaka, Tokyo 181-8588, Japan Jet Propulsion Laboratory, California Institute of Technology, 48010 Oak Grove Drive, Pasadena, CA 91109, USA Department of Astronomy, University of Virginia, Charlottesville, VA 22904, USA Department of Chemistry, University of Virginia, Charlottesville, VA 22903, USA (Received; Revised; Accepted)
Submitted to ApJABSTRACTWe have analyzed ALMA Cycle 5 data in Band 4 toward three low-mass young stellar objects (YSOs),IRAS 03235+3004 (hereafter IRAS 03235), IRAS 03245+3002 (IRAS 03245), and IRAS 03271+3013(IRAS 03271), in the Perseus region. The HC N ( J = 16 − E up /k = 59 . OH lines ( E up /k = 15 . − . N distributions ( ∼ − OH emission in IRAS 03245 is ∼ N in this source. We compare the CH OH/HC N abundance ratios observed in thesesources with predictions of chemical models. We confirm that the observed ratio in IRAS 03245 agreeswith the modeled values at temperatures around 30–35 K, which supports the HC N formation by theWCCC mechanism. In this temperature range, CH OH does not thermally desorb from dust grains.Non-thermal desorption mechanisms or gas-phase formation of CH OH seem to work efficiently aroundIRAS 03245. The fact that IRAS 03245 has the highest bolometric luminosity among the target sourcesseems to support these mechanisms, in particular the non-thermal desorption mechanisms.
Keywords: astrochemistry – ISM:molecules – stars: low-mass INTRODUCTIONStudying the environment and evolution of low-massstar forming regions is important for revealing howour Sun was born. The chemical composition in star-forming regions provides us various information, for ex-ample, evolutionary stages and physical conditions (e.g.,Caselli & Ceccarelli 2012).
Corresponding author: Kotomi Taniguchi, Liton [email protected]; [email protected]
The gas-phase chemical composition around youngstellar objects (YSOs) yields essential information onnot just the current evolutionary stage but also formerenvironments or evolutionary processes in the pre-stellarcore phase (Sakai & Yamamoto 2013; Spezzano et al.2016; Taniguchi et al. 2019a, 2020b). The ice man-tles form in cold and dense starless cores. Al-though saturated complex organic molecules (COMs)were considered to be deficient in the gas phase dur-ing the cold stage, several COMs have now beendetected in starless cores, e.g., the Barnard 5 colddark cloud (Taquet et al. 2017) and the L1544 pre-
Taniguchi et al. stellar core (Jim´enez-Serra et al. 2016). These COMsare considered to be formed in the gas phase (e.g.,Balucani et al. 2015) or formed on dust surfaces followedby non-thermal desorption mechanisms. New dust-surface mechanisms proposed by Jin & Garrod (2020)successfully reproduce observed abundances of COMsin the L1544 pre-stellar core. Furthermore, model-ing studies including radiolysis can also produce highCOM abundances (e.g., Shingledecker & Herbst 2018;Shingledecker et al. 2018; Paulive et al. 2021). Hence,even in cold dense cores, chemical processes formingCOMs are also efficient. After YSOs are born, ice man-tles, including COMs, sublimate when the temperaturerises to ≈
100 K. The chemistry in YSOs then stronglydepends on their density and temperature structures(e.g., Jørgensen et al. 2020, and therein).There are two, well-recognized chemical processes ef-fective around low-mass YSOs. One is hot corino chem-istry, in which saturated COMs are abundant in thehot ( >
100 K) dense ( ≥ cm − ) gas (Ceccarelli2004; Herbst & van Dishoeck 2009). Hot corino chem-istry is initiated by successive hydrogenation reactions ofCO molecules on grain surfaces to form CH OH, whichis a fundamental COM and a precursor of more com-plex COMs ( ¨Oberg & Bergin 2020). The other is warmcarbon chain chemistry (WCCC; Sakai & Yamamoto2013). Carbon-chain molecules are formed from gaseousCH , which is one of the main constituents of ice man-tles ( ¨Oberg & Bergin 2020), in lukewarm envelopes ( ≈ −
35 K, Hassel et al. 2008). Hence, the presence oftwo types of chemistry around low-mass YSOs may in-dicate that some conditions of the prestellar core phaseproduce CO-rich ice and others lead to CH -rich ice.However, the origin of the chemical diversity is still con-troversial. The results of this paper will help to explainthe chemical diversity.Although COMs are generally considered to be abun-dant in hot dense gas around YSOs, a recent survey ob-servation has revealed that CH OH and CH CHO areprevalent. Scibelli & Shirley (2020) conducted surveyobservations of these COMs toward 31 starless cores inthe Taurus region with the Arizona Radio Observatory(ARO) 12 m telescope, and reported that the detectionfractions of CH OH and CH CHO are 100% (31/31)and 70% (22/31), respectively. It was also found thatthe CH OH abundance decreases from young cores toevolved starless cores possibly due to depletion onto dustgrains (Scibelli & Shirley 2020). The detection of COMsin the starless cores suggests that there are some non-thermal desorption mechanisms or efficient gas-phaseformation pathways at work in starless cores. Takakuwa et al. (1998) showed different spatial distri-butions of H CO + and CH OH in the TMC-1C regionand suggested that CH OH cores are younger and far-ther from protostar formation than H CO + cores. Asubsequent study supports the possibility of the differ-ent evolutionary stages of these cores (Takakuwa et al.2003a). Furthermore, Takakuwa et al. (2000) carriedout mapping observations toward the Heiles Cloud 2region in Taurus and found that the CH OH emissionis weak toward the protostars (Class 0 and Class I)and rather stronger toward cores without protostars.Takakuwa et al. (2003b) also conducted mapping obser-vations toward IRAS 04191+1522, a very young Class0 YSO. They found that the CH OH abundance to-ward the YSO is ∼
10 times lower than that towardthe interaction region between the outflow and the sur-rounding dense gas. Similar observational results of thenon-detection or weak emission of CH OH at proto-stellar positions have been reported (e.g., White et al.2006). The weak CH OH emission at protostellar coreshas been reported even in the higher angular-resolutionobservations with ALMA at 3 mm, where the ob-scuring effect by the dust emission is not significant(Maureira et al. 2020). All these observational resultsimply that CH OH is associated with young starlesscores and adsorbed onto dust grains in evolved star-less cores. The weak CH OH emission toward low-massYSOs (Takakuwa et al. 2000, 2003b; White et al. 2006;Maureira et al. 2020) probably indicates that CH OHlines are not necessarily tracers for low-mass protostel-lar cores, and not all the protostellar sources possess hotcore/hot corinos.Cyanoacetylene (HC N), which is the shortest mem-ber of the cyanopolyyne family (HC n +1 N), is known tobe abundant in starless cores, as well as other carbon-chain species (e.g., Suzuki et al. 1992). It is also de-tected around low-mass and high-mass YSOs (Law et al.2018; Taniguchi et al. 2018, 2019b), and protoplanetarydisks ( ¨Oberg et al. 2015; Bergner et al. 2018). Hence,HC N is prevalent in star-forming regions in various evo-lutionary stages. However, it has not been determinedwhether the WCCC mechanism is responsible for for-mation of HC N around low-mass YSOs, because thereis no information about spatial distributions of HC Naround these sources at sufficiently high spatial resolu-tions (Law et al. 2018).In the present paper, we focus on methanol andcyanoacetylene as potential indicators to constrain evo-lutionary stages of low-mass YSOs. For this purpose,high angular-resolution imaging observations of CH OHand HC N in low-mass YSOs are essential. We inves-tigate the spatial distributions of HC N and CH OH arbon Chain vs. COM Chemistry in Low-Mass YSOs in the Perseus Region N and CH OH are shown in Section 3. Theobserved CH OH/HC N abundance ratio is comparedwith the results of the chemical network simulations inSection 4. Finally, we summarize our main conclusionsin Section 5. DATA REDUCTIONIn this paper, we present ALMA Band 4 archival datatoward three low-mass YSOs in the Perseus region takenfrom Cycle 5 data . The properties of our target sourcesare summarized in Table 1. In Table 1, the coordi-nates correspond to the phase reference centers of theseALMA observations carried out with the 7-m Array in2018 July.Table 2 summarizes information concerning each spec-tral window presented in this paper. We used the firstspectral window for continuum data. The second andthird windows contain lines of HC N and CH OH, re-spectively. Observed lines of each molecule are summa-rized in Table 3. The frequency resolution of 121 kHzcorresponds to a velocity resolution of ∼ .
24 km s − .The field of view (FoV) and largest angular scale (LAS)are 65 . ′′ . ′′
6, respectively.We carried out data reduction and imaging usingthe Common Astronomy Software Application (CASAv 5.4.0; McMullin et al. 2007) on the pipeline-calibratedvisibilities. The data cubes were created with the CASAtclean task. Uniform weighting was applied. The result-ing angular resolution of ∼ ′′ × ′′ corresponds to ∼ × ∼ × . ′′ ×
250 pixels.Continuum ( λ = 2 mm) images were created fromthe data cubes using the IMCONTSUB task. The noiselevels of the continuum images are 4 mJy beam − inIRAS 03235 and IRAS 03245, and 1 mJy beam − inIRAS 03271, respectively. RESULTS AND ANALYSES3.1.
Results: Spatial distributions
Figure 1 shows the continuum images of the three tar-get sources. The green crosses indicate positions of in-frared sources revealed by the Spitzer Core to Disk (c2d) project ID; 2017.1.00955.S, PI; Jennifer Bergner Legacy program (Evans et al. 2003; Enoch et al. 2009).In IRAS 03235 and IRAS 03245, the dust continuumpeaks are consistent with the Spitzer sources. On theother hand, the detection of dust continuum emission isnot significant in IRAS 03271.The HC N ( J = 16 −
15) line has been detected inall of the sources, and panels (a)–(c) of Figure 2 showmoment 0 maps of HC N toward the three sources.Methanol (CH OH) has been detected only in IRAS03245. We used two lines (1 − − and 3 − − ) formaking the CH OH moment 0 map, because these linesare partially blended. Panel (d) of Figure 2 shows itsmoment 0 map toward IRAS 03245.Since all of the target sources have similar enve-lope masses (Table 1), the detection/non-detection ofCH OH does not likely depend on the presence of an en-velope. Graninger et al. (2016) detected CH OH towardall of the three sources with the IRAM 30-m telescopewith a beam size of ∼ ′′ . In their observations, onlytwo lines with low upper-state energies ( E up /k = 6 .
97K and 12.54 K) were detected as weak lines in IRAS03235 and IRAS 03271. The noise levels of the ACA7-m observations could be too high to detect weak linesof CH OH in IRAS 03235 and IRAS 03271.We applied 2D gaussian fitting to the continuum andmoment 0 maps. The fit results are summarized in Table4. In the case that beam-deconvolved sizes could notbe obtained, we indicate the synthesized beam as theirupper limits.In IRAS 03235, the peak position of HC N is consis-tent with that of continuum image as well as the Spitzersource. The size of the HC N emission ( ∼ ′′ , corre-sponding to ∼ N emission ( ∼ N is likely to be efficiently formed in moder-ate temperature regions ( ∼ −
35 K) by the warm car-bon chain chemistry (WCCC; Sakai & Yamamoto 2013)mechanism.In IRAS 03245, the peak positions of dust continuumand CH OH are consistent with the Spitzer point source.The spatial extent of HC N ( ≈ ′′ , ∼ ∼ N. Theemission region of HC N is larger than that of CH OH( ∼ ′′ , corresponding to ∼ N emission is not consistentwith the Spitzer point source, but is located ∼ Taniguchi et al.
Table 1.
Summary of target sourcesSource Name R.A. (J2000) a Decl. (J2000) a T bol (K) b L bol (L ⊙ ) c M env (M ⊙ ) d α b Class e T dust (K) f IRAS 03235+3004 03 h m . s
45 +30 ◦ ′ . ′′ . ± .
11 1.03 0 ...IRAS 03245+3002 03 h m . s
03 +30 ◦ ′ . ′′ . ± .
04 2.46 0 37 ± h m . s
16 +30 ◦ ′ . ′′ . ± .
10 1.58 I 45 ± a Coordinates of the phase reference centers. b Bolometric temperature taken from Dunham et al. (2015). c Bolometric luminosities at an assumed distance of 250 pc taken from Dunham et al. (2015) are scaled to the newly measureddistance of Perseus (293 pc; Ortiz-Le´on et al. 2018). d Envelope masses at an assumed distance of 250 pc taken from Enoch et al. (2009) are scaled to the newly measured distanceof Perseus (293 pc). e IR spectral indexes ( α ) taken from Enoch et al. (2009). f Dust temperatures taken from Emprechtinger et al. (2009). (a) IRAS 03235 (b) IRAS 03245 (c) IRAS 03271
Figure 1.
Continuum images toward the three IRAS sources. The contour levels are 3 σ , 4 σ , 5 σ , and 6 σ of the rms noise levels.In panel (c), only the 3 σ level contour appears. The noise levels are 4 mJy beam − for panels (a) and (b), and 1 mJy beam − for panel (c), respectively. The filled white ellipses indicate the angular resolution of approximately 11 . ′′ × . ′′
6. The crossesindicate source positions taken from the Spitzer c2d program (Enoch et al. 2009).
Table 2.
Summary of spectral windows covered by thecorrelator setupFrequency Frequency Angular Molecularrange (GHz) res. (kHz) resolution Lines147.097–147.224 124 11 . ′′ × . ′′ . ′′ × . ′′ N157.236–157.301 121 10 . ′′ × . ′′ OH clear at the current angular resolution. One possibilityis that HC N emission is associated with the molecular
Table 3.
Summary of target molecular linesMolecule Transition Frequency E up /k log ( A ij )(GHz) (K) (s − )HC N 16–15 145.560946 59.4 -3.61727CH OH 4 − − E − − E − − E − − E Note —Taken from the Jet Propulsion Laboratory (JPL) cat-alog (Pickett et al. 1998). arbon Chain vs. COM Chemistry in Low-Mass YSOs in the Perseus Region (c) IRAS 03271 (HC N) (d) IRAS 03245 (CH OH)(b) IRAS 03245 (HC N)(a) IRAS 03235 (HC N) Figure 2.
Moment 0 maps of (a)–(c) HC N and (d) CH OH. The black contour levels are 5 σ , 6 σ , 7 σ , and 8 σ for panels (a) and(d), 3 σ , 4 σ , 5 σ , and 6 σ for panels (b) and (c), respectively. In panel (c), only the 3 σ and 4 σ level contours appear. The noiselevels are 0.1, 0.2, 0.065, and 0.15 Jy beam − × km s − for panels (a)–(d), respectively. The white contours show the distributionof dust continuum emissions, as in Figure 1. The filled white ellipses represent an angular resolution of approximately 11 . ′′ × . ′′ . ′′ × . ′′ outflow, because the direction of the offset of the HC Nemission peak from the Spitzer source is similar to thedirection of the molecular outflow (Hsieh et al. 2019).3.2.
Spectral analyses
Figures 3 and 4 show spectra of HC N and CH OHat the HC N emission peaks (Table 4) with beam sizesof 14 ′′ , 12 ′′ , and 9 ′′ in IRAS 03235, IRAS 03245, andIRAS 03271. These beam sizes correspond to the sizeof the HC N emission in its moment 0 maps. In theIRAS 03235 and IRAS 03271 panels of Figure 4, bluelines indicate velocity positions of 5.4 km s − and 5.7 km s − , which are the V LSR values of HC N in eachsource.We analyzed spectra using the CASSIS software(Caux et al. 2011). In the analyses, we used the lo-cal thermodynamic equilibrium (LTE) model availablein the CASSIS spectrum analyzer, assuming that thelines are optically thin. Using the Markov chain MonteCarlo (MCMC) method and the LTE model, we derivedthe column densities ( N ) and excitation temperatures( T ex ) of HC N and CH OH, treating N , T ex , line width(FWHM), and V LSR as semi-free parameters within cer-
Taniguchi et al.
Table 4.
Results of 2D gaussian fittingSource Species Peak Position (J2000) Size Position Angle Peak Intensity a IRAS 03235 continuum 03 h m . s ± . s ◦ ′ . ′′ ± . ′′ < . ′′ × . ′′ . ± . N 03 h m . s ± . s ◦ ′ . ′′ ± . ′′
36 11 . ′′ ± . ′′ × . ′′ ± . ′′ ◦ . ± . h m . s ± . s ◦ ′ . ′′ ± . ′′ < . ′′ ± × . ′′ . ± . N 03 h m . s ± . s ◦ ′ . ′′ ± . ′′
21 9 . ′′ ± . ′′ × . ′′ ± . ′′
53 55 ◦ . ± . OH 03 h m . s ± . s ◦ ′ . ′′ ± . ′′
14 5 . ′′ ± . ′′ × . ′′ ± . ′′
64 49 ◦ . ± . N 03 h m . s ± . s ◦ ′ . ′′ ± . ′′ < . ′′ × . ′′ . ± . a Units for Peak Intensity are “mJy beam − ” and “Jy beam − × km s − ” for continuum and the molecular emissions, respectively. tain ranges. In the MCMC method, the minimum andmaximum values set for parameter in a component de-fine the bounds in which the values are chosen randomly.We consider the following two cases:1. The range of the excitation temperature of bothHC N and CH OH is 15–30 K; hereafter Case 1.2. The excitation temperatures are comparable todust temperatures (Table 1); the excitation tem-perature ranges are set at 35–40 K in IRAS 03235and IRAS 03245, and 35–50 K in IRAS 03271(Case 2).We consider the first case because the WCCC mech-anism (Sakai & Yamamoto 2013) seems to work in ourtarget sources as mentioned in Section 3.1. The assumedtemperature range covers typical excitation tempera-tures of carbon-chain species in L1527 (Yoshida et al.2019). We adopt the second case because the observedlines may come from inner dense cores compared withprevious single-dish observations (Graninger et al. 2016;Bergner et al. 2017; Yoshida et al. 2019), where contri-butions from outer envelopes may be significant due tolow upper-state-energy lines, and the excitation temper-ature may be underestimated in our data.The obtained parameters for each case are summa-rized in Table 5. The modeled spectra using the bestfitted values are overlaid as purple lines in Figures 3and 4. The line widths (FWHM) of HC N in our targetsources ( ∼ − ) are larger than the typical valuein starless cores ( ∼ . − ; Kaifu et al. 2004) butsimilar to that in the L1527 low-mass WCCC source(Yoshida et al. 2019). Thus, the detected HC N lineseems to mainly trace lukewarm regions and not comefrom outer cold regions. In IRAS 03245, the V LSR valueof HC N is consistent with that of CH OH.The obtained excitation temperatures of HC N in allof the three sources in Case 1 are 16 −
18 K, which arecomparable with those in L1527 (Yoshida et al. 2019). The derived column densities of HC N change by onlya factor of ∼ − N in Case 1 are slightly higher than theprevious results obtained with the IRAM 30-m telescope(the telescope beam size is ∼ ′′ ; Bergner et al. 2017)by a factor of ≈ . −
2. Bergner et al. (2017) derivedthe HC N column densities by the rotational diagrammethod using the two lines with almost similar upper-state energies or using only one line with assumed fixedrotational temperatures. The effect of the beam dilutioncould also result in the lower column densities in thesingle-dish observations (Bergner et al. 2017).The CH OH column densities in IRAS 03245 for Cases1 and 2 are higher than that derived by Graninger et al.(2016) by a factor of 1.8–2.8. Graninger et al. (2016) de-rived the CH OH column density to be (1 . ± . × cm − from observations with the IRAM 30-m telescope.Again, this difference in column density could be causedby the beam dilution in the single-dish observations. DISCUSSIONWe now compare the derived CH OH/HC N column-density ratios with those derived by the chemical net-work simulations. We describe our chemical networksimulations in Subsection 4.1, and compare observationsand simulations in Subsection 4.2.4.1.
Chemical network simulations
We used the chemical network code Nautilus(Ruaud et al. 2016) and a hot-core model with a warm-up stage. The physical evolution is the same as thatin Taniguchi et al. (2019a). The initial gas density is n H = 10 cm − and increases to 10 cm − during thefreefall collapse. The initial temperature is 10 K andrises to 200 K during the warm-up period. In this pa-per, we investigated low-mass YSOs, and thus we uti-lized the model with a slow warm-up period (1 × yr).The dust temperature is assumed to be equal to the gastemperature. a r b o n C ha i nv s . C O M C h e m i s t r y i n L o w - M a ss Y S O s i n t h e P e r s e u s R e g i o n Table 5.
Obtained parameters from the spectral analysesSource Species Case 1 Case 2
N T ex FWHM V LSR
N T ex FWHM V LSR (cm − ) (K) (km s − ) (km s − ) (cm − ) (K) (km s − ) (km s − )IRAS 03235 HC N (4 . ± . × . ± . . ± .
05 5 . ± .
02 (1 . ± . × . ± . . ± .
06 5 . ± . N (7 . ± . × . ± . . ± .
006 4 . ± .
017 (2 . ± . × . ± . . ± .
07 4 . ± . OH (2 . ± . × . ± . . ± .
13 4 . ± .
05 (4 . ± . × . ± . . ± .
17 4 . ± . N (2 . ± . × . ± . . ± .
11 5 . ± .
012 (5 . ± . × . ± . . ± .
16 5 . ± . Note —The errors represent the standard deviation.
Taniguchi et al.
Figure 3.
Spectra of HC N ( J = 16 −
15) toward the positions of peak emission of this molecule. The purple lines are modeledspectra of the best fitted results with CASSIS. arbon Chain vs. COM Chemistry in Low-Mass YSOs in the Perseus Region Figure 4.
Spectra of CH OH (4 − − E , 1 − − E , 3 − − E , and 2 − − E ) at the HC N emission peaks. Thepurple lines are modeled spectra of the best fitted results with CASSIS. In panels of IRAS 03235 and IRAS 03271, the bluelines indicate the positions of the CH OH lines with the same V LSR values as those of HC N, 5.4 km s − and 5.7 km s − forpanels of IRAS 03235 and IRAS 03271, respectively. We changed the cosmic ray ionization rate ( ζ ) andthe initial C/O ratio in order to investigate the de-pendence of the CH OH/HC N ratio on these param-eters. Table 6 summarizes the models used in this pa-per. Models No. 1–No. 3 are the same as those pre-sented in Taniguchi et al. (2019a). The C/O elemen-tal ratio could change the CH OH and HC N abun-dances. We included the high cosmic ray ionizationrates because this rate ( ζ = 4 . × − s − ) can repro-duce the observed abundances of carbon-chain speciesin intermediate-mass protoclusters (Fontani et al. 2017;Favre et al. 2018) and around some massive young stel-lar objects (MYSOs) (Taniguchi et al. 2019a, 2020a).Figure 5 shows the time dependence of theCH OH/HC N ratio during the warm-up period, tak-ing the evolutionary stages of our target sources intoconsideration. As a general trend of these models, theCH OH/HC N ratio decreases after ∼ × yr (23 K < T <
50 K), slightly increases during ∼ × − yr (50 K < T <
70 K), and then drops again. Suchcomplicated features are caused by combinations of the
Table 6.
Summary of modelsModel ζ (s − ) a C/O ratioNo. 1 1 . × − b No. 2 3 . × − b No. 3 4 . × − b No. 4 1 . × − c No. 5 1 . × − d a Cosmic ray ionization rates used inthe model b The C/O ratio of 0.4 correspondsto C + = 7 . × − and O = 1 . × − , and the rest of the elementalabundances remain the same as de-scribed in Taniguchi et al. (2019a). c Corresponding to C + = 1 . × − and O = 3 . × − . d Corresponding to C + = 1 . × − and O = 1 . × − . Taniguchi et al.
Figure 5.
Comparisons of the CH OH/HC N ratios between observations and chemical network simulations (Models No. 1–No. 5, see Table 6). The purple and orange horizontal lines indicate the observed values in IRAS 03245 for Case 1 and Case2, respectively. The black curves indicate gas (and dust) temperature evolution from the model. The symbols of t n and t n ’,( n = 1 −
6) associated with dashed vertical lines represent the ages when the observed CH OH/HC N ratios agree with themodeled values. The details are described in the main text and values are summarized in Table 7. temporal or thermal dependence of the abundance ofCH OH and HC N during the warm-up stage.The detail formation pathways of HC N during thewarm-up stage are discussed in detail in Taniguchi et al.(2019a) and we include only a short description here.Methane (CH ) desorbs from ice mantles at tempera-tures above 25 K, and rapidly starts forming carbon-chain species in the gas phase, via the WCCC mech-anism (Hassel et al. 2008). This explains the first de-crease in the CH OH/HC N ratio, because there is noefficient formation pathway for the gas-phase CH OH atthat time ( T ≈ −
50 K, t ≈ . × − × yr).After HC N is formed in the gas phase, it is soon ad-sorbed onto dust grains and accumulated in ice mantles,leading to a decrease in the gas-phase HC N abundance. Methanol (CH OH) is mainly formed by the electronrecombination reaction of CH OCH +4 ( t ≈ × − yr) in the gas phase, before the temperature reachesits sublimation temperature. The formation of theCH OCH +4 ion becomes enhanced by the reaction be-tween HCO + and CH OCH , for which the formationincreases in rate by the associative gas-phase reactionbetween CH O and CH from t ≈ × yr.After the temperature reaches at 87 K ( t = 1 . × yr), CH OH desorbs from dust grains and its gasphase abundance increases sharply, while HC N startsdesorbing from dust grains at t ≃ . × yr ( T ≃ Comparison of the CH OH/HC N abundanceratio between observations and models arbon Chain vs. COM Chemistry in Low-Mass YSOs in the Perseus Region Table 7.
Summary of ages at which models agree with the ob-served CH OH/HC N ratio and corresponding temperaturesLabel Age Temperature pair of(yr) (K) Model and observationt . ×
32 No. 1, 3, 5 & Case 1t (9.2–9.3) ×
55 No. 1, 3 & Case 1t ’ (7.3–7.5) × ’ (9.6–9.7) × . ×
34 No. 4 & Case 1t . ×
60 No. 4 & Case 1t ’ (7.7–7.8) × ’ 9 . ×
63 No. 4 & Case 2t (9.8–10) × . ×
26 No. 5 & Case 2
The observed CH OH/HC N ratios in IRAS 03245 arederived to be 3 . ± . +14 − for Case 1 and Case2, respectively, using the results summarized in Table5. The upper panel of Figure 5 shows the time de-pendence of the CH OH/HC N ratio for three modelswith the different cosmic ray ionization rates (ModelsNo. 1–No. 3), while the lower panel shows three mod-els with the different C/O ratios (Models No. 1, No. 4,and No. 5). The observed CH OH/HC N ratios are indi-cated as horizontal lines including error bars; purple andorange lines indicate the results of Case 1 and Case 2,respectively. The temperature scale arises from only themodel. We now focus on comparison with observations,starting with ages after the temperature reaches 25 K,at which the WCCC mechanism starts ( t ≥ . × yr).We first compare the observed ratio in IRAS 03245 toModel No. 1. The observed ratio for Case 1 (3 . ± . t ≈ . × yr (labeledas t ), at which the temperature is ≃
32 K, as shownin the upper panel of Figure 5. This dust temperatureagrees with the dust temperature derived in IRAS 03245(37 ± t ≈ (9 . − . × yr (t , T ≈ +14 − )is consistent with Model No. 1 around (7 . − . × yr(t ’, T ≃ −
28 K). The dust temperature of 26–28 K isslightly lower than the observed dust temperature in thissource. Since we assumed that the higher temperatureconditions in Case 2 compared with Case 1 and the ex- citation temperatures of HC N and CH OH are derivedto be around 37 K in Case 2 (Table 5), such a lowertemperature (26–28 K) compared to Case 1 ( ≃
32 K)is contradictory. Besides, the CH OH/HC N ratio withCase 2 agrees with Model No. 1 around (9 . − . × yr (t ’, 62–65 K). However, the temperature at this ageis higher than the observed dust temperature in thissource.In the similar manner, we investigate the ages andtemperatures when the observed CH OH/HC N ratiosmatch those derived by other models (No. 2 – No. 5),and the results are summarized in Table 7. In ModelNo. 2, the modeled CH OH/HC N ratio is lower thanthe observed ratios in IRAS 03245 for Cases 1 and 2 after t ≥ . × yr, so that Model No. 2 cannot reproducethe observed ratio in this source.Taking all of the results summarized in Table 7into consideration, the observed CH OH/HC N column-density ratio and dust temperature in IRAS 03245 canbe best reproduced simultaneously by our models at t ≈ (7.7–7.9) × yr ( T ≈ −
34 K). This temperaturerange agrees with the WCCC mechanism (Hassel et al.2008). The other temperature ranges ( ∼ −
28 K and ∼ −
70 K) are not consistent with the observed dusttemperatures, and hence these ages and temperaturesare unlikely.In IRAS 03235 and IRAS 03271, we could not detectCH OH, and the CH OH/HC N ratio should be lowerthan that in IRAS 03245 ( < . OH/HC N ratios toward these two low-mass YSOsagree with Model No. 1 around t (7 . × yr) < t < t (9 . × yr), corresponding to 32 K < T < ± ± OH/HC N ratios become lower than theobserved ratios in IRAS 03245 after ≈ yr. However,the dust temperature becomes higher than 70 K after ≈ yr. These dust temperatures above 70 K are notconsistent with the observed dust temperatures in ourtarget sources. Thus, we omit a possibility of ages after ≈ yr.Since the dust temperatures in the observed sourcesdo not reach the sublimation temperature of CH OH, weconclude that the non-thermal desorption mechanismsor gas-phase formation reactions leading to CH OH arethe primary contributors to gas-phase CH OH forma-tion around low-mass Class 0 and Class I YSOs at the2
Taniguchi et al. time/temperature regime studied here. These mech-anisms seem to be especially efficient in IRAS 03245,among the observed three sources. In the chemical net-work simulation, gas-phase ion-molecule reactions con-tribute to the formation of gas-phase CH OH, via theelectron recombination reaction of CH OCH +4 . Thesehypotheses for gas-phase CH OH formation, in par-ticular those involving non-thermal desorption such asphoto-desorption, are supported by the fact that thebolometric luminosity in IRAS 03245 is higher than theother two sources. CONCLUSIONSWe have analyzed ALMA Cycle 5 data in Band 4 to-ward three low-mass YSOs: IRAS 03235+3004 (IRAS03235), IRAS 03245+3002 (IRAS 03245), and IRAS03271+3013 (IRAS 03271), in the Perseus region. TheHC N ( J = 16 −
15) line was detected toward all sources,while the CH OH lines were detected only toward IRAS03245. The detection/non-detection of CH OH is inde-pendent of the mass of envelopes. These results mayimply the chemical diversity around low-mass YSOs.In IRAS 03235 and IRAS 03245, both of which areClass 0 sources, the continuum peaks correspond to theSpitzer sources. The spatial distributions of HC N areextended relative to those of the continuum emission inthese two sources. On the other hand, the spatial dis-tribution of CH OH matches the continuum emission inIRAS 03245. In IRAS 03271, the spatial distribution ofthe continuum emission shows an elongated feature andits peak does not correspond to the Spitzer source. Thespatial distribution of HC N differs from that of con-tinuum emission and its peak does not match with theSpitzer source. Although we could not understand thediscrepancy between the spatial distributions of HC Nand that of continuum with the current angular resolu-tion, HC N may be originated from the molecular out-flow.We derived the column densities and excitation tem-peratures of HC N and CH OH with the CASSIS soft-ware. The CH OH/HC N ratio in IRAS 03245 was de-rived to be 3 . ± . +14 − on the assumptions thatthe excitation temperature is close to that of L1527,and the excitation temperature is close to the dusttemperatures, respectively. We compared the observedCH OH/HC N ratios to those derived by the chemical network simulations. The observed CH OH/HC N ra-tios in IRAS 03245 are reproduced at ages when thedust temperature lie around ≈ −
35 K, which agreeswith the dust temperature in IRAS 03245. Regardingthe other two sources, where CH OH has not been de-tected, the CH OH/HC N ratios, which should be lowerthan that in IRAS 03245, can be reproduced when thetemperatures lie between 32 K < T <
55 K, which agreeswith the dust temperatures in these low-mass YSOs.At the times when the observed CH OH/HC N ra-tios and the dust temperatures are reproduced, HC Nis efficiently formed by the warm carbon chain chem-istry (WCCC) mechanism. The non-thermal desorptionor sublimation or the gas-phase reactions contribute tothe formation of gas-phase CH OH during the times.The higher bolometric luminosity in IRAS 03245 seemsto support these conclusions, especially the importanceof non-thermal desorption mechanisms.ACKNOWLEDGMENTSThis paper makes use of the following ALMA data:ADS/JAO.ALMA
Facilities:
Atacama Large Millimeter/submillimeterArray (ALMA)
Software:
Common Astronomy Software Applica-tions package (CASA; McMullin et al. 2007), CASSIS(Caux et al. 2011), Nautilus (Ruaud et al. 2016)REFERENCES
Acharyya, K. & Herbst, E. 2017, ApJ, 850, 105.doi:10.3847/1538-4357/aa937e Balucani, N., Ceccarelli, C., & Taquet, V. 2015, MNRAS,449, L16. doi:10.1093/mnrasl/slv009 arbon Chain vs. COM Chemistry in Low-Mass YSOs in the Perseus Region13