Complex Organic Molecules in Star-Forming Regions of the Magellanic Clouds

Home , Large Magellanic Cloud, Magellanic Clouds, N44 (emission nebula), Small Magellanic Cloud

Marta Sewiło,∗,†,† Steven B. Charnley,¶ Peter Schilke,§ Vianney Taquet,k Joana M. Oliveira,⊥ Takashi Shimonishi,#,@ Eva Wirström,4 Remy Indebetouw,∇,†† Jacob L. Ward,‡‡ Jacco Th. van Loon,⊥ Jennifer Wiseman,¶¶ Sarolta Zahorecz,§§,kk Toshikazu Onishi,§§ Akiko Kawamura,⊥⊥ C.-H. Rosie Chen,## Yasuo Fukui,@@ and Roya Hamedani Golshan§

†CRESST II and Exoplanets and Stellar Astrophysics Laboratory, NASA Goddard Space Flight Center, Greenbelt, MD 20771, USA ‡Department of Astronomy, University of Maryland, College Park, MD 20742, USA ¶Astrochemistry Laboratory, NASA Goddard Space Flight Center, Greenbelt, MD 20771, USA §I. Physikalisches Institut der Universität zu Köln, Zülpicher Str. 77, 50937, Köln, Germany kINAF, Osservatorio Astroﬁsico di Arcetri, Largo E. Fermi 5, 50125 Firenze, Italy ⊥Lennard-Jones Laboratories, Keele University, ST5 5BG, UK #Frontier Research Institute for Interdisciplinary Sciences, Tohoku University, Aramakiazaaoba 6-3, Aoba-ku, Sendai, Miyagi, 980-8578, Japan @Astronomical Institute, Tohoku University, Aramakiazaaoba 6-3, Aoba-ku, Sendai, Miyagi, 980-8578, Japan 4Department of Space, Earth and Environment, Chalmers University of Technology, Onsala Space Observatory, 43992, Onsala, Sweden ∇Department of Astronomy, University of Virginia, PO Box 400325, Charlottesville, VA 22904, USA ††National Radio Astronomy Observatory, 520 Edgemont Rd, Charlottesville, VA 22903, USA ‡‡Astronomisches Rechen-Institut, Zentrum für Astronomie der Universität Heidelberg, Mönchhofstr. 12-14, 69120 Heidelberg Germany ¶¶NASA Goddard Space Flight Center, 8800 Greenbelt Rd, Greenbelt, MD 20771, USA §§Department of Physical Science, Graduate School of Science, Osaka Prefecture University, 1-1 Gakuen-cho, Naka-ku, Sakai, Osaka 599-8531, Japan arXiv:1909.06843v1 [astro-ph.GA] 15 Sep 2019 kkChile Observatory, National Astronomical Observatory of Japan, National Institutes of Natural Sciences, 2-21-1 Osawa, Mitaka, Tokyo 181-8588, Japan ⊥⊥National Astronomical Observatory of Japan, 2-21-1 Osawa, Mitaka, Tokyo 181-8588, Japan ##Max-Planck-Institut fu?r Radioastronomie, Auf dem Hügel 69 D-53121 Bonn, Germany @@Nagoya University, School of Science, Furo-cho, Chikusa-ku, Nagoya, JP 464-8602

E-mail: marta.m.sewilo@nasa.gov

1 Abstract Introduction

The Large and Small Magellanic Clouds (LMC An important issue for astrochemistry is to un- and SMC), gas-rich dwarf companions of the derstand how the organic chemistry in low- Milky Way, are the nearest laboratories for de- metallicity environments (i.e., with low abun- tailed studies on the formation and survival dances of elements heavier than hydrogen or of complex organic molecules (COMs) under helium), relevant for star formation at ear- metal poor conditions. To date, only methanol, lier epochs of cosmic evolution, differs from methyl formate, and dimethyl ether have been that in the Galaxy. Large aromatic organic detected in these galaxies – all three toward molecules such as polycyclic aromatic hydro- two hot cores in the N113 star-forming region carbons (PAHs) are detected in high-redshift in the LMC, the only extragalactic sources ex- galaxies, but the formation efficiency of smaller hibiting complex hot core chemistry. We de- complex organic molecules has been unknown. scribe a small and diverse sample of the LMC A quantitative determination of the organic and SMC sources associated with COMs or hot chemistry in high-redshift galaxies with differ- core chemistry, and compare the observations ent star formation histories, and lower abun- to theoretical model predictions. Theoretical dances of the important biogenic elements (C, models accounting for the physical conditions O, N, S, P), can shed light on the inventory and metallicity of hot molecular cores in the of complex organic molecules available to plan- Magellanic Clouds have been able to broadly etary systems and their potential for harbor- account for the existing observations, but fail ing life (e.g.1). The nearest laboratories for to reproduce the dimethyl ether abundance by detailed studies of star formation under metal more than an order of magnitude. We discuss poor conditions are the Large and Small Mag- future prospects for research in the field of com- ellanic Clouds (LMC and SMC). plex chemistry in the low-metallicity environ- The LMC and SMC are gas-rich dwarf com- ment. The detection of COMs in the Magellanic panions of the Milky Way located at a distance Clouds has important implications for astrobi- of (50.0 ± 1.1) kpc3 and (62.1 ± 2.0) kpc4, re- ology. The metallicity of the Magellanic Clouds spectively. They are the nearest star-forming is similar to galaxies in the earlier epochs of galaxies with metallicities Z (the mass frac- the Universe, thus the presence of COMs in the tions of all the chemical elements other than LMC and SMC indicates that a similar prebi- hydrogen and helium) lower than that in the so- 5 otic chemistry leading to the emergence of life, lar neighborhood (Z = 0.0134) : ZLMC∼0.3– 6,7 as it happened on Earth, is possible in low- 0.5 Z and ZSMC∼0.2 Z . Apart from the metallicity systems in the earlier Universe. lower elemental abundances of gaseous C, O, and N atoms (e.g.,8), low metallicity leads to less shielding (due to the lower dust abundance; 9,10 LMC e.g., ), greater penetration of UV photons SMC into the interstellar medium, and consequently warmer dust grains (e.g.,11,12). The interstellar ultraviolet radiation field in the LMC and

CH 3OH SMC is 10–100 higher than typical Galactic values (e.g.,13). Gamma-ray observations indicate CH OCH 3 3 that the cosmic-ray density in the LMC and CH 3OCHO Spitzer SMC is, respectively, ∼25% and ∼15% of that 14,15 Accepted for publication in the ACS Earth and measured in the solar neighborhood ( ). All Space Chemistry journal on August 27, 2019. these low metallicity eﬀects may have direct The article is part of the “Complex Organic consequences on the formation eﬃciency and Molecules (COMs) in Star-Forming Regions” survival of COMs, although their relative im- Special Issue. portance remains unclear.

2 LMC -66.0 -68.0 N144

N113

N159 -70.0

250 µm 160 µm 100 µm

(RA, Dec) J2000 90.0 80.0 70.0

SMC N78 -72.0

250 µm 160 µm 100 µm -73.0 -74.0

(RA, Dec) J2000 24.0 20.0 16.0 12.0

Figure 1: Three-color composite mosaics of the LMC (top) and SMC (bottom), combining the Herschel/HERITAGE 250 µm (red), 160 µm (green), and 100 µm (blue) images.2 Star-forming regions harboring sources with the detection of COMs (N 113 and N 159 in the LMC and N 78 in the SMC; see Table 2) are indicated with yellow arrows and labeled. The location of the star- forming region N 144 hosting ST 11, a hot core without COMs, is indicated in white. North is up and east to the left.

3 The range of metallicities observed in the most luminous CO clouds from this survey were Magellanic Clouds is similar to galaxies at observed with 4500 resolution using the Australia redshift z ∼ 1.5 − 2, i.e. at the peak Telescope National Facility (ATNF) Mopra 22- of the star formation in the Universe (be- m Telescope (the MAGMA survey;27). tween ∼2.8 and ∼3.5 billion years after the The molecular gas in the Magellanic Clouds Big Bang; e.g.,16), making them ideal tem- was also investigated using the CO data from plates for studying star formation and com- the ESO SEST (Swedish/ESO Submillimeter plex chemistry in low-metallicity systems in Telescope) Key-Program (CO J = 1 − 0 and an earlier Universe where direct measurements 2 − 1 lines with 5000 and 2500 half-power beam of resolved stellar populations are not possi- width, HPBW;28) toward 92 and 42 positions ble. Only very recently have gaseous com- in the LMC and SMC, respectively, mostly co- plex organic molecules (methanol: CH3OH, incident with IRAS sources; the results were methyl formate: HCOOCH3, dimethyl ether: reported in multiple papers, including detailed 17–19 CH3OCH3) been detected in the LMC and studies of major star-forming giant molecu- in the SMC20, using the Atacama Large Mil- lar clouds (GMCs) and less active cloud com- limeter/submillimeter Array (ALMA). These plexes in the LMC and small molecular cloud observations showed that interstellar complex complexes in the SMC (see29 and references organic molecules (COMs) can form in the low- therein). Many other CO studies of individ- metallicity environments, probably in the hot ual star-forming regions with SEST in both the molecular cores and corinos associated with LMC and SMC followed (e.g., LMC regions massive and low- to intermediate-mass proto- N159W, N113, N44BC, and N214DE30). stars, respectively. COMs had previously been The ﬁrst unbiased spectral line survey of an detected outside the Milky Way, but in galaxies individual star-forming region in the LMC was with metallicities comparable to or higher than performed by31 using SEST and centered on the solar (21 and references therein). N 159W molecular peak in N 159 – the bright- In this article, we summarize the current state est NANTEN GMC with H ii regions. The of knowledge concerning the organic chemistry observations covered a 215–245 GHz frequency in the Magellanic Clouds. range in several bands and selected bright lines at lower frequencies (∼100 GHz; HPBW∼2300 at 220 GHz and 5000 at 100 GHz). No COM Molecular Line Inventories in transitions covered by these observations (e.g., Star-Forming Regions in the propyne: CH3CCH J = 5 − 4 and J = 6 − 5 31 and CH3OCH3 J = 6−5 lines) were detected. Magellanic Clouds found that fractional abundances with respect 13 to H2 for molecules detected in N 159W ( CO, The proximity of the LMC/SMC enables both C18O, CS, SO, HCO+, HCN, HNC, CN, and detailed studies of individual star-forming re- H2CO – formaldehyde) are typically a factor of gions and statistical studies of molecular clouds 10 lower than those observed in the Galaxy. or stellar populations on galactic scales. The The multi-region, multi-line SEST 3 mm (85– systematic studies of molecular clouds in the 115 GHz) survey by32 focused on one SMC Magellanic Clouds started with the 12CO J = ◦ ◦ 22 (LIRS 36 or N 12) and four LMC (N 113, 1 − 0 survey of the LMC (central 6 × 6 ; ) N 44BC, N 159HW, and N 214DE) molecular and the SMC (2◦ × 2◦,23) with a resolu- 0 cloud cores. They detected the C2H, HCN, tion of 8.8 (the Columbia 1.2m Millimeter- HCO+, HNC, CS, 13CO, CN, 12CO transitions, WaveTelescope), followed by a higher resolution but the CH OH J = 2 − 1 line at ∼96.74 GHz (2.06) survey of both galaxies with the NAN- 3 24,25 covered by their observations was not detected TEN 4m telescope (e.g., ). The more sensi- toward any of the regions. The follow-up study tive NANTEN survey that identiﬁed 272 molec- 26 on N 12 in the SMC included the 2, 1.3, and ular clouds in the LMC was published by. The 0.85 mm SEST bands, but also failed to detect

4 a,b Table 1: CH3OH Transitions Detected in the LMC in Single-Dish Observations

Region Species Transition Frequency EU Ref. (GHz) (K) c 17 N 159W CH3OH vt=0 2−1,2–1−1,1 E 96.739362 12.54 + 17 20,2–10,1 A 96.741375 6.97 17 20,2–10,1 E 96.744550 20.09 17 21,1–11,0 E 96.755511 28.01 17 3−1,3–2−1,2 E 145.09747 19.51 + 17 30,3–20,2 A 145.10323 13.93 18 N 113 CH3OH vt=0 2−1,2–1−1,1 E 96.739362 12.54 + 18 20,2–10,1 A 96.741375 6.97 18 3−1,3–2−1,2 E 145.09747 19.51 + 18 30,3–20,2 A 145.10323 13.93 a All the observations were conducted with SEST. The SEST’s half-power beam sizes are 6100–5400 and 4700–2400 in frequency ranges of 85–98 GHz and 109–219 GHz, respectively (from18); b The table uses the Cologne Database for Molecular Spectroscopy (CDMS) notation from the National Radio Astronomy Observatory (NRAO) Spectral Line Catalog (Splatalogue). For CH3OH (NKa Kc p v), a sign value of Ka is used to diﬀerentiate E1 (+) and E2 (−) states (both belong to the same E symmetry species); c The four transitions at ∼96.7 GHz are blended.

33 CH3OH or other COMs. todissociation rates of molecules, both due to The first detection of a COM in the Magel- the lower metallicity in the LMC. lanic Clouds was reported by17 in their SEST An extensive spectral line study on N 113, 0.85, 1.3, 2, and 3 mm survey of the SMC previously observed by32 at 3 mm (see above), 18 LIRS 49 (or N 27) and LMC 30 Dor–10, 30 Dor– provided new detections of CH3OH. Their 27, N 159W, N 159S, and N 160 regions, cover- SEST observations covered the 0.85, 1.3, 2, and ing a range of metallicities and UV radiation 3 mm bands. The rich spectral line inventory strengths. N 159W was the only region where included two J = 2−1 and two J = 3−2 lines of CH3OH was detected: J = 2 − 1 and two 3 − 2 CH3OH (see Table 1). The CH3OH abundance lines at ∼96.7 and ∼145.1 GHz, respectively with respect to H2 was estimated to be ∼5 × (see Table 1). The17’s observations covered the 10−10, similar to that observed in N 159W by17, J = 6 − 5 and J = 8 − 7 lines of CH3CCH, but and an order of magnitude lower than that ob- only marginal (∼2.5σ) detections are reported served in the Galaxy. toward one region – N 159W.17 estimated the N 113 and N 159W remain the best-studied re- CH3OH fractional abundance with respect to gions in the LMC in terms of chemical inven- −10 H2 of 4.3 × 10 in N 159W. The authors in- tory. The initial single-dish surveys were fol- dicated that the derived physical and chemi- lowed by interferometric observations with the cal quantities are averages over spatial scales Australia Telescope Compact Array (ATCA) probed by their observations (10–15 pc in the with 400–600 angular resolutions by34 (N 113 3 mm, and 5–10 pc in the 2/1.3 mm bands). only) and35 (N 113, N 159, N 105, and N 44) Overall, the molecular abundances in N 159W that focused on dense gas tracers such as HCO+ 36 where CH3OH was detected are typically five to and HCN. searched for NH3 (1, 1) and (2, 2) in twenty times lower than those measured in the seven star-forming regions in the LMC (N 113, Galactic molecular clouds, assuming that hy- N 159W, N 159S, 30 Dor 10, N 44 BC, N 105 A, drogen is mainly in molecular form within the and N 83 A) with ATCA (HPBW∼900–1700) and clouds.17 concluded that reduced abundances only detected it toward N 159W with an abun- −10 are likely to be a combined effect of lower abundance of ∼4 × 10 with respect to H2 – 1.5– dances of carbon and oxygen and higher pho- 5 orders of magnitude lower than observed in

5 Galactic star-forming regions. the LMC/SMC; e.g.,47–52). YSOs detected by The most recent multi-region, multi-line Spitzer include mostly Stage I (protostars with single-dish survey was conducted by37 us- accreting envelopes and disks) and also Stage ing the Mopra 22-m telescope’s 3 mm band II (disk-only) YSOs; the youngest Stage 0/I (HPBW∼3800 at 90 GHz and 3000 at 115 GHz). YSOs (protostars in the main accretion stage) The sample included seven H ii regions: NQC2, were identified with Herschel using the data CO Peak 1, N 79, N 44C, N 11B, N 113, and from the “HERschel Inventory of The Agents N 159W. That work confirmed lower molecular of Galaxy Evolution” (HERITAGE,2; e.g.,53,54) abundances in the LMC star-forming regions. survey. The AKARI Large-area Survey of the Those observations failed to detect CH3OH and Large Magellanic Cloud (LSLMC) also carried other COMs. out near- to mid-infrared photometric survey The results of the single dish molecular line and near-infrared slitless spectroscopic survey surveys of individual H ii regions in the LMC towards the LMC, whose datasets also are used 55–57 and SMC indicate a deficiency of CH3OH in to search for YSOs in the LMC. On a these low-metallicity galaxies. This deficiency galaxy-wide scale, YSO lists can be utilized to is also supported by the low detection rate determine a star formation rate in each galaxy. 38 of CH3OH masers in the Magellanic Clouds. On scales of individual star-forming clouds, conducted a complete systematic survey for YSO catalogs can also be used to study star 6668-MHz CH3OH and 6035-MHz excited-state formation rate, star formation efficiency, the OH masers in both the LMC and SMC using the evolutionary stage of the GMCs, and provide Methanol Multibeam (MMB) survey receiver targets for follow-up detailed observations. on the 64-m Parkes telescope (HPBW∼3.03), Spectroscopic follow-up observations of supplemented by higher sensitivity targeted ob- Spitzer, Herschel, and AKARI YSOs in the servations of known star-forming regions. The LMC and SMC revealed chemical differences survey resulted in a detection of only one in the YSO envelopes between these galaxies maser of each kind, increasing the number of and compared to Galactic YSOs, which can known 6668-MHz CH3OH masers in the LMC be explained as a consequence of differences in to four.39–41 42 later detected a single 12.2 GHz metallicity (see Section ). CH3OH maser in the LMC. To date, no CH3OH masers have been detected in the SMC.38 estimated that CH3OH masers in the LMC are un- Near- to Mid-Infrared Spec- derabundant by a factor of ∼45 (or ∼4–5 after troscopy: More Evidence for correcting for differences in the star formation rates between the galaxies) and interstellar OH Chemical Differences between and H2O masers by a factor of ∼10 compared the LMC, SMC, and the to the Galaxy. The Spitzer Space Telescope (Spitzer;43) and Milky Way Herschel Space Observatory (Herschel;44) enabled studies of young stellar object (YSO) The mid-IR studies on the LMC YSOs with 12,58–60 populations in the LMC and SMC, both galaxy- Spitzer Infrared Spectrograph (IRS) and 61,62 wide and in individual star-forming regions. near-IR studies with the AKARI satellite 12,59 Using the data from the “Spitzer Surveying and ground-based instrumentation found the Agents of Galaxy Evolution” (SAGE;45) differences in ice chemistry between the LMC and “Surveying the Agents of Galaxy Evolu- and the Galaxy. In cold molecular clouds, lay- tion in the Tidally Stripped, Low Metallic- ers of ice form on the surface of dust grains. ity Small Magellanic Cloud” (SAGE-SMC;46), The main constituent of ice in envelopes of thousands of YSOs were identified in the Magel- YSOs is H2O, mixed primarily with CO and 63 −4 lanic Clouds, dramatically increasing the num- CO2. Since the H2O ice abundances (∼10 ; 64,65 ber of previously known YSOs (from 20/1 in e.g., ) exceed by a few orders of magnitude

6 the gas-phase abundances, understanding gas– frared properties consistent with it being in grain processes is crucial to fully understand the the transition from the post-asymptotic gi- chemistry in the YSO envelopes. ant branch (AGB) to the PN stage (e.g.,70 In the LMC and by comparison to Galactic and references therein).70 identified the di- samples, CO2 ice column densities are enhanced acetylene (C4H2), ethylene (C2H4), triacety- 58,60,62 with respect to H2O ice , while rela- lene (C6H2), benzene (C6H6), and possibly 12 12 tive CO-to-CO2 abundances are unchanged. methylacetynele (CH3C2H; commonly known as proposed the scenario where the high CO2/H2O propyne which is found also in Titan’s atmo- ratio is due to the low abundance of H2O in sphere) absorption bands in the Spitzer/IRS the LMC. If there is a difference in the optical spectrum of SMP LMC 11. C6H6 is a simplest extinction AV threshold for H2O and CO2 ices building block of polycyclic aromatic hydrocar- (as the observations suggest), there could be an bons (PAHs), which are abundant and ubiqui- envelope of less shielded material that shows tous in the interstellar medium (e.g.,71). In the H2O ice but no CO2 ice. The strong interstellar subsequent analysis and modeling of the same radiation in the LMC penetrates deeper into Spitzer/IRS spectrum,72 questioned the detec- the YSO envelopes as compared with Galac- tion of C6H2, but confirmed the detection of tic YSOs, possibly destroying H2O ice in less- CH3C2H. SMP LMC 11 shows a peculiar chem- shielded outer layers without affecting CO2 and istry, not typical for PN or post-AGB objects H2O ice mixtures that exist deeper in the enve- in general, e.g., it is one of only two evolved lope. stars in which hydrocarbons up to benzene in In their Very Large Telescope (VLT)’s absorption are detected (e.g.,72–74). Infrared Spectrometer And Array Camera (ISAAC) near-IR spectroscopic observations,69 marginally detected CH3OH ice absorption The Quest for Detecting band toward two embedded massive YSOs in COMs in the Magellanic the LMC with the abundances suggesting that solid CH3OH is less abundant for high-mass Clouds with ALMA YSOs in the LMC than those in the Galaxy. They proposed a model which explains both The detection of CH3OH and more complex the low abundance of CH OH and the higher molecules in the Magellanic Clouds has been 3 accelerated by the advent of ALMA. ALMA abundance of solid CO2 found in previous studies. In this “warm ice chemistry” model, the provides high spatial resolution and sensitivity grain surface reactions at a relatively high tem- which enabled studies in the LMC/SMC that in the pre-ALMA era were only possible in our perature (T & 20 K) are responsible for the observed characteristics of ice chemical compo- Galaxy. The wide frequency coverage of ALMA sition in the LMC. The high dust temperature observations (up to four ∼2 GHz spectral win- in the low-metallicity environment caused by dows in the submm/mm wavelength range) al- the strong interstellar radiation field leads to lows for simultaneous observations of multiple inefficient H atom sticking and CO hydrogena- molecular species and enables serendipitous dis- coveries. tion to CH3OH on grain surfaces. At the same time, the production of CO2 is enhanced as a result of the increased mobility of parent species Hot Core without COMs: LMC (CO and OH; see Section for a discussion on ST 11 theoretical models). 5 The Spitzer/IRS spectroscopy also revealed ST 11 is a high-mass YSO (∼5×10 L , 47,48,54,62,76 several COMs in the circumstellar environment Spitzer YSO 052646.61−684847.2; , of SMP LMC 11 (or LHA 120–N 78) - an object or IRAS 05270–6851) located in the N 144 star- in the LMC classified as a carbon-rich plane- forming region in the LMC (Fig. 1 and ??). In tary nebula (PN) in the optical, but with in- the pre-ALMA era, the spectral properties of

7 ST 11 Ks H J 39.4 47:59.8

-68:50:23.8 8.0 µm

-68:48:50.2 4.5 µm Dec 1.5 pc α RA J2000 H

5:26:48.0 45.6 48.0 5:26:24.0

Figure 2: Left: A three-color mosaic combining the InfraRed Survey Facility (IRSF) Ks (red), H (green), and J (blue;66) images for the LMC source ST 11 – the source with a hot core chemistry, but no COMs detection.67 Right: a three-color mosaic combining the Spitzer/SAGE 8.0 µm (red), 4.5 µm (green;45), and the MCELS Hα (blue;68) images. The ALMA 840 µm continuum contours are overlaid on the image in the left panel; the contour levels are (3, 6, 20, 40) × 0.94 mJy beam−1, the 840 µm image rms noise level. The ALMA beam size is 0.005 × 0.005. The red ‘×’ symbol in the left panel indicates the position of the Spitzer YSO – the bright source in the mosaic shown in the right panel. the source were well studied with mid-infrared outflow in this source. and near-infrared spectroscopy, which have de- CH3OH and larger COMs are not detected to- tected absorption bands due to H2O ice, CO2 ward ST 11, despite the high gas temperature ice, and silicate dust, as well as emission due to that is sufficient for the ice sublimation and de- Polycyclic Aromatic Hydrocarbon (PAH) and spite the detection of H2CO, a small organic hydrogen recombination lines.62,76 molecule. The estimated upper limit on the 00 −10 The high-resolution (∼0. 5 or ∼0.12 pc) sub- CH3OH gas fractional abundance of 8 × 10 millimeter (∼0.86 mm) observations toward is significantly lower than those of Galactic ST 11 with ALMA revealed the presence of a hot cores by a few orders of magnitude (67, hot molecular core associated with the source.67 see Table 3).67 hypothesized that the inhib- According to the rotational diagram analysis of ited formation of CH3OH on the dust surface 32 34 multiple SO2 and SO2 lines, the gas temper- before the hot core stage could be responsi- ature of the hot core region is estimated to be ble for this deficiency.67 also estimated upper ∼100–200 K. The total gas density of the source limits on fractional abundances with respect 6 −3 is estimated to be higher than 2×10 cm to H2 for (CH3OCH3, HCOOCH3,C2H5OH) of based on the analysis of the submillimeter dust (<1.5×10−8, <1.8×10−8, <3.7×10−9). These continuum. Emission lines of CO, C17O, HCO+, upper limits indicate that in ST 11 the frac- 13 + 33 33 H CO ,H2CO, NO, H2CS, SO, SO2, and tional abundance of CH3OCH3 is at least an SiO are also detected from the compact region order of magnitude lower compared to Galactic associated with the source. High-velocity com- hot cores, while the HCOOCH3 and C2H5OH ponents are detected in the line profile of CO fractional abundances are comparable to the (3–2), which suggests the presence of molecular average abundances observed in the Galactic

8 N159W-S Ks H J 04.6

MMS-3 45:53.8

8.0 µm 4.5 µm Hα

Dec 1.5 pc -69:46:47.8 -69:46:15.4 RA J2000

43.2 5:39:40.8 48.0 5:39:36.0

Figure 3: The same as Fig.2, but for N159W–S – the source with a cold methanol detection (PI Peter Schilke, see Sections and ). The ALMA 1.3 mm continuum contours are overlaid on the image in the left panel; the contour levels are (0.1, 0.3, 0.5, 0.7) mJy beam−1, the image rms noise level is 0.027 mJy beam−1.75 The ALMA beam size is 0.0026 × 0.0023. counterparts. dubbed N 159W–North and N 159W–South (or N 159W–N and N 159W–S, respectively). Both Cold Methanol: LMC N159W- N 159W–N and N 159W–S are associated with Spitzer/Herschel YSOs.47–49,51,54 Two Stage 0/I South and SMC IRAS 01042 −7215 YSOs with outflows have masses >30 M and N159W–South are associated with 1.3 mm continuum peaks, one in each N 159W–N (YSO–N or Spitzer The N 159 star-forming region located south 053937.56−694525.4) and N 159W–S (YSO–S from 30 Doradus is one of the best studied or Spitzer 053941.89−694612.0; e.g.,49). Based areas in the LMC hosting H ii regions, young on the kinematics of the molecular gas and the stellar clusters, Spitzer and Herschel massive location of YSO–S at the intersection of two 49 YSOs, and a water maser (e.g., and references 13CO filaments,77 proposed that star-formation 42 therein; ). As described in Section , N 159 was in N 159W–S was triggered by the collision of the target of detailed spectral line observations two filaments ∼105 years ago. with single-dish telescopes. It is the brightest This model was revised by75 based on the 12 region in the LMC in CO at SEST/Mopra res- four times higher resolution observations of CO olution. One of the three giant molecular clouds isotopes (∼0.0025 or 0.06 pc), which resolved in N 159, N 159W, is one of only two regions the filaments in N 159W–S into a complex, with a CH3OH detection in the pre-ALMA era. hub–filament structure. Multiple protostellar 00 The first high-resolution (∼1 or ∼0.25 pc) sources with outflows separated by 0.2–2 pc 13 12 CO and CO (2–1) ALMA observations of were detected along the main massive filament. N159W resolved the molecular gas emission They correspond to the three major 1.3 mm into filaments and detected two sources with continuum peaks dubbed MMS–1, MMS–2, and 77 outflows. It was the first detection of both MMS–3 (see Fig. 5). The75’s observations are filamentary structure and outflows outside the consistent with a scenario that a large-scale Galaxy. These observations identified two main (>100 pc), rather than local (∼10 pc;77), col- regions of high-mass star formation which were

9 N159W-S KMOS 0.5 pc K-band H2 Brγ 08.2 11.8

KMOS K-band -69:46:11.8

ALMA methanol -69:46:13.6 contours: 0.25 pc

ALMA 1.3 mm 1.3 mm Dec RA J2000 42.2 5:39:41.7 41.3 42.2 5:39:42.0 41.7 41.5

Figure 4: Left: Three-color mosaic combining the VLT/KMOS K-band (red), ALMA CH3OH (green), and ALMA 1.3 mm (blue) images of N 159W–S. The contours correspond to the 1.3 mm continuum emission with contour levels the same as in Fig. 3. The yellow circle shows the catalog position of the Spitzer YSO.48 Right: Three-color mosaic combining KMOS images: K-band (red), H2 (green), and Brγ (blue). The 1.3 continuum contours (the same as in the left panel) are overlaid for reference.

N159W-S CH OH -69°46'06.0" 3 0.08 contours: 1.3 mm (Jy beam 08.0" 0.06

10.0" − 0.04 1 km s km

(J2000) MMS-3 Dec

12.0" − MMS-2 0.02 1 MMS-1 )

14.0" 0.00 0.5 pc 42.48s 41.76s 5h39m41.04s RA (J2000)

Figure 5: The integrated intensity image of CH3OH for N 159W–S with the 1.3 mm continuum contours and the positions of the continuum peaks (MMS–1, MMS–2, and MMS–3;75) overlaid. The position of the Spitzer YSO is indicated with the red ﬁlled circle (e.g.,49). The 1.3 mm continuum contour levels and the ALMA beam size are the same as in Fig. 3. The ALMA synthesized beam 00 00 size of the CH3OH observations (shown in the lower left corner) is 0. 3 × 0. 3.

10 Table 2: Positions, Physical Properties, and Fractional Abundances for Sources Described in Section with COMs Detection or Measured Upper Limits with ALMA

m Source R.A. (J2000) Decl. (J2000) Molecule Trot N N/N(H2) Type Ref. (h m s)(◦ 0 00) (K) (cm−2) LMC a g 16 −8 19 N 113 A1 05:13:25.17 −69:22:45.5 CH3OH 134±6 (1.6±0.1)×10 (2.0±0.3)×10 HC h 15 −9 CH3OCH3 ...... (1.8±0.5)×10 (2.2±0.7)×10 h 15 −9 HCOOCH3 ...... (1.1±0.2)×10 (1.4±0.4)×10 a g 15 −9 19 N 113 B3 05:13:17.18 −69:22:21.5 CH3OH 131±15 (6.4±0.8)×10 (9.1±1.7)×10 HC h 15 −9 CH3OCH3 ...... (1.2±0.4)×10 (1.7±0.7)×10 f h 15 −9 HCOOCH3 ...... < 3.4×10 < 0.5×10 b,j k N159W–South 05:39:41 −69:46:06 CH3OH ...... CM c,i l 14 −10 ? 67 ST 11 05:26:46.60 −68:48:47.0 CH3OH 100 < 3.5×10 < 8×10 HC

11 l 15 −9 CH3OCH3 100 < 1.3×10 < 3×10 l 15 −8 HCOOCH3 100 < 7.1×10 < 2×10 l 15 −9 C2H5OH 100 < 2.2×10 < 5×10 SMC d,i +1.1 14 +4.2 −10 20 IRAS 01042−7215 P2 01:05:49.54 −71:59:48.9 CH3OH 18±5 (1.4−0.6) × 10 (5.0−2.3) × 10 CM e,i +1.1 14 +1.0 −9 20 IRAS 01042−7215 P3 01:05:49.50 −71:59:49.4 CH3OH 22±6 (1.6−0.6) × 10 (1.5−0.6) × 10 CM a b The positions of N 113 A1 and B3 correspond to the ∼224.3 GHz continuum peaks; Multiple CH3OH peaks are detected across the N159W–South molecular clump (see Section and Fig. 5); c The position of ST 11 corresponds to the 359 GHz continuum peak; d The 33 e position of IRAS 01042−7215 P2 corresponds to the C S, H2CS, and SiO peak; The position of IRAS 01042−7215 P3 corresponds to the f g CH3OH peak; a tentative detection; Trot and N were determined using the madcubaij software with the initial estimates based on the h 19 i rotational diagram analysis of CH3OH; Trot for CH3OCH3 and HCOOCH3 was assumed to be equal to Trot determined for CH3OH ; Trot 20 and N were determined based on the rotational diagram of CH3OH; the values presented in the table are the revised values from based on new ALMA observations (PI T. Shimonishi); j The analysis of the N159W–South data is preliminary, thus we do not provide physical k l 34 33 parameters for this region; PI P. Schilke; The average rotation temperature of SO2, SO2, and SO2; N and N/N(H2) upper limits are at the 2σ level; m Type: HC – a hot core; HC? – a hot core with no COMs; CM – a source with cold methanol. 8.0 µm IRAS 01042-7215 Ks H J 4.5 µm Hα 42.5 -71:57:35.8

P3 -71:59:51.1 1.5 pc -72:01:47.8 Dec RA J2000 50.4 1:05:48.0 06:24.0 1:05:36.0

Figure 6: The same as Fig.2, but for the field around the SMC source IRAS 01042−7215 where CH3OH was detected toward sources P2 and P3 (indicated in the left panel) corresponding to 20 the CH3OH peaks. The ALMA 1.2 mm continuum contours are overlaid. The continuum contour levels are (3, 6, 10, 20) × 21 mJy beam−1, the 1.2 mm image rms noise level. The ALMA synthesized beam size is 0.0035 × 0.0022. lision triggered the formation of both filaments the ALMA continuum contours overlaid for ref- and massive protostars in N 159W–S. erence. The extended H2 emission is associated MMS–1 and MMS–2 continuum sources with outflows from N 159 A7 121 (MMS–1) and are associated with two near-infrared sources 123 (MMS–2) detected by75 with ALMA. The N 159 A7 ‘121’ and ‘123’, respectively, detected Brγ emission is only detected toward N 159 A7 by79 using high-resolution (∼0.002) VLT/NACO 123 (MMS–2). observations. The position of the Spitzer YSO The most recent ALMA observations of 77 (YSO–S in ; see Fig. 3) is offset to the west N 159W–S detected multiple CH3OH peaks hinting at the possibility that a cluster is being throughout the region (see Fig. 5; PI P. Schilke) formed. No infrared source is detected toward with the brightest emission associated with the MMS–3, indicating that it is the youngest of the continuum source MMS–3. The left panel in three protostars. Recent VLT/KMOS observa- Fig. 4 shows a three-color mosaic combining 00 tions (a pixel scale 0. 2, a full width at half the KMOS K-band, ALMA CH3OH and 1.3 00 maximum seeing ∼0. 4; PI Jacob Ward, see mm continuum emission. No CH3OH emission Section for technical details) targeting YSO–S is detected toward MMS–1 and MMS–2 that (see Fig. 3) detected the K-band continuum, have infrared counterparts. Although these the Brackett-gamma emission line of hydrogen observations were designed to detect multi- (Brγ), and the H2 emission line toward N 159 A7 ple COMs (e.g, CH3OH, CH3CN – methyl 121 (MMS–1) and 123 (MMS–2). As the H2 cyanide, HCOOCH3, CH3OCH3), only CH3OH and Brγ lines trace shocks and accretion, re- was found. The detection of multiple peaks spectively, they can shed light on the nature of hints at the possibility that there is an under- the sources and thus possible formation scenar- lying more extended distribution of CH3OH ios of COMs (e.g., non-thermal origin in shocks that is resolved out. traced by H2). The right panel in Fig. 4 shows Preliminary Local Thermodynamic Equilib- a combination of the three KMOS images with rium (LTE) fitting using multiple CH3OH lines

12 IRAS 01042-7215 0.1 pc KMOS: K-band

47.9 H2 Brγ -71:59:49.0

contours: 1.2 mm Dec RA J2000

49.7 1:05:49.4 49.2

Figure 7: Three-color mosaic of the IRAS 01042–7215 ﬁeld combining the VLT/SINFONI K-band 78 (red), H2 (green), and Brγ (blue) images. Red contours correspond to the ALMA 1.2 mm continuum emission with contour levels the same as in Fig. 6.

3.0 -71°59'47.0" IRAS 01042-7215 CH3OH contours: 1.2 mm 2.4

48.0" (K s km P2 1.8

P1 −

49.0" 1.2 1 ) Dec (J2000)Dec

P3 0.6 50.0" 1" 0.3 pc 0.0 49.56s 1h05m49.20s RA (J2000)

Figure 8: The CH3OH integrated intensity image (the average of the CH3OH 5−1,5 − 4−1,4 E and + 50,5 −40,4 A lines) of IRAS 01042−7215 in the SMC. The 1.2 mm continuum contours are overlaid; the contour levels and the size of the ALMA synthesized beam shown in the lower left corner are the same as in Fig. 6.

13 (∼241.7–241.9 GHz; see Table 5) detected to- ALMA observations detected two continuum ward the two brightest peaks indicate that the peaks – ‘P1’ and ‘P2/P3’ toward IRAS 01042– gas is cold (∼14 K). Methanol forms on grain 7215. P2/P3 is further resolved into two sources surfaces and has to be released to the gas phase (P2 and P3) corresponding to the CH3OH emis- by energetic events to be detectable. The des- sion peaks where multiple CH3OH transitions orption mechanisms include sublimation by in- are detected (Fig. 8;20). frared heating by a forming star, photodesorp- The 1.2 mm continuum source P1 is associ- tion by UV photons, sputtering by shocks, and ated with the high-mass YSO corresponding to chemical desorption (see a discussion in Sec- the IRAS source, while no infrared sources are tion ). None of these mechanisms is directly detected at the positions of P2 and P3 (see supported by the observations of N 159W–S: Fig. 7 and20). P2 and P3 correspond to the the gas is cold (desorption by the infrared and two CH3OH emission peaks in the region; no 20 UV heating is less likely) and the line widths CH3OH is detected toward P1 (Fig. 8 and ). 33 13 + are narrow (broad lines are expected in the Besides CH3OH, CS, C S, SO, SO2,H CO , shock sputtering scenario). Moreover, the aver- H13CN, SiO are detected toward P2/P3, but no age cosmic ray flux in the LMC is low, making COMs larger than CH3OH. the cosmic ray induced UV radiation less effec- The rotation diagram analysis of the CH3OH tive than in the Galaxy. Since the analysis of lines for IRAS 01042–7215 P2 and P3 suggests the ALMA CH3OH data is preliminary, we do that the temperature of the CH3OH gas is low not provide physical parameters for N159W–S (∼20 K), well below the sublimation temper- 82 in Table 2. ature of the CH3OH ice (∼80 K; ). This is another example of a source in the Magellanic IRAS 01042−7215 Clouds with the detection of ‘cold methanol’. The CH OH abundances relative to H for P2 4 3 2 IRAS 01042–7215 is a high-mass YSO (∼2×10 and P3 are an order of magnitude lower than 12,80 L ; e.g., ; Spitzer YSO SSTISAGEMA those for hot cores A1 and B3 in N 113 (see Ta- J010549.30–715948.5,52) located in the north- ble 2 and Section ). The origin of CH3OH gas east region of the SMC bar at the outskirts of in P2/P3 is still under debate.20 suggest that the N 78 star-forming region (see Figs. 1 and a possible origin could be sputtering of the ice 6). The source is well studied through near- mantles by shocks triggered by the outflow from infrared and mid-infrared spectroscopy; absorp- nearby YSOs, or chemical desorption of CH3OH tion bands due to H2O ice, CO2 ice, and sili- from dust surfaces. The presence of shocked but cate dust are detected, while prominent PAH cold CH3OH gas has been suggested in Galac- bands and ionized metal lines are not de- tic infrared dark clouds, indicating that CH OH 11,12,59,80 3 tected. No CH3OH ice has been de- may be tracing relatively old shocks (e.g.,83). tected toward IRAS 01042–7215. Recent VLT/SINFONI K-band observations (a pixel scale of 0.001, a full width at half max- Hot Cores with COMs More Com- imum seeing ∼0.006;78) detected the Brγ and plex than Methanol: LMC N 113 78 H2 emission lines towards IRAS 01042−7215. A1 and B3 Figure 7 shows a three-color mosaic combining To date, COMs with more than six atoms have the SINFONI K-band continuum, H2, and Brγ images. The SINFONI image reveals a single only been reliably detected in the Magellanic Clouds toward two sources, both in the LMC compact embedded source (Av = 16 ± 8 mag) with little or no extended line emission. The N 113 star-forming region. The physical and chemical characteristics of N 113 make it one of measured H2 emission line ratios are consistent with expectations for shocked emission.81 the most interesting star-forming regions in the The high-resolution (∼0.0022–0.0037 or ∼0.07– LMC. Like N 159, it was the target of detailed 0.11 pc at the distance of the SMC) 1.2 mm spectral line studies with single-dish telescopes

14 N 113 A1 Ks H J 8.0 µm 4.5 µm Hα 22:11.8

A1 -69:22:42.0 YSO-1

YSO-4 YSO-3 51.4 1.5 pc -69:23:02.2 Dec RA J2000 26.4 5:13:24.0 5:13:30.0 24.0 18.0

N 113 B3 Ks H J 8.0 µm 4.5 µm 21:35.8 Hα

B3 -69:22:19.0 -69:22:11.8 29.8 1.5 pc

47.8 Dec RA J2000 19.2 5:13:16.8 26.4 19.2 5:13:12.0

Figure 9: Left: The same as in Fig. 2, but for N 113 A1 (top) and B3 (bottom) hot cores. The contours in the left panel correspond to the 1.3 mm continuum emission.19 The contour levels are (5, 10, 20, 50) × the image rms noise level of 0.1 mJy beam−1.

(see Section ) that resulted in the detection of II YSO candidates in the LMC.47,48,51,53,54 18 CH3OH. Mid-IR sources, maser emission, and com- N 113 contains: (1) one of the most massive pact H ii regions in the central part of the 5 27 (∼10 M ) and richest GMCs in the LMC ; GMC associated with the extended gas and (2) signatures of both recent (Hα emission) and dust emission, reveal sites of the current star ongoing (e.g., maser and bright IR emission) formation in N 113 (see Fig. 9). The Spitzer star formation34,84; (3) signatures of star for- (3.6–160 µm) and Herschel (160–500 µm) im- mation triggered by winds from massive stars; ages with resolutions ranging from ∼200 to 3800 (4) the largest number of H2O and OH masers show that this region is dominated by three and the brightest H2O maser in the entire bright sources, which were identiﬁed as 30–40 38,42,84–86 LMC ; (5) and it has among the high- M Stage I YSOs (YSO–1, YSO–3, and YSO–4 est concentrations of Spitzer/Herschel Stage 0– in Fig. 9: Spitzer sources 051317.69−692225.0,

15 N 113 A1 CH3OH N 113 B3 CH3OH 0.09 -69°22'43.0" contours: 1.3 mm 0.2 contours: 1.3 mm

(Jy beam -69°22'19.2" CH3OH (Jy beam 44.0" 0.06 0.1 H2O B3

45.0" − −

1 22.8" H O A1 1 2 s km km s km 0.03 46.0" Dec (J2000)Dec Dec (J2000)Dec − OH − 1 1 )

0.0 ) 47.0" 26.4" 2.1" 2.1" 0.00 48.0" 0.5 pc 0.5 pc 25.44s 5h13m24.96s 18.48s 17.40s 5h13m16.32s RA (J2000) RA (J2000) 0.09 -69°22'42.6" N 113 A1 CH3OCH3 N 113 B3 CH3OCH3 contours: 1.3 mm contours: 1.3 mm -69°22'19.2" 0.06 (Jy beam 0.06 (Jy beam

44.4" B3 − 0.03 − 22.8" 1 A1 1 0.03 km s km km s km Dec (J2000)Dec Dec (J2000)Dec 46.2" − − 1 1 ) 0.00 ) 26.4" 0.00 48.0" 0.5 pc 0.5 pc 25.68s 5h13m24.96s 18.48s 17.40s 5h13m16.32s RA (J2000) RA (J2000) 0.08 -69°22'42.6" N 113 A1 HCOOCH3 0.20 N 113 B3 HCOOCH3 contours: 1.3 mm contours: 1.3 mm 0.15 -69°22'19.2" (Jy beam 0.06 (Jy beam

0.10 44.4" 0.04 B3

− 22.8" − A1 1 1 0.05 s km s km 0.02

Dec (J2000)Dec 46.2" (J2000)Dec − − 1 1 0.00 ) 26.4" ) 0.00

48.0" 0.5 pc 0.5 pc 25.68s 5h13m24.96s 18.48s 17.40s 5h13m16.32s RA (J2000) RA (J2000)

Figure 10: Integrated intensity images for CH3OH (top panel; made using the channels corresponding to all CH3OH transitions in the 216.9 GHz spectral window; see Table 5), CH3OCH3 (middle panel; the 130,13–121,12 transition); and HCOOCH3 (bottom panel; integrated over all detected transitions) for the A1 (left column) and B3 (right column) hot cores in the N 113 star-forming region. The H2O and OH masers are indicated. The white contours correspond to the 1.3 mm continuum emission; the contour levels are the same as in Fig. 9. The red contours in the top right panel −1 correspond to the CH3OH emission with contour levels of (3, 5, 7) × 21 mJy beam , the rms noise level of the CH3OH integrated intensity image. The image shows the CH3OH detection in other spots in the region (‘Region B’ in19) where no other COMs are detected. The results presented in this ﬁgure were reported in.19 The ALMA beam size is shown in the lower left corner in each image.

051325.09−692245.1, and 051321.43−692241.5, in prep.). respectively;48) and are characterized by dis- Detailed Spitzer and Herschel studies on the tinct physical conditions (48,87; J. M. Oliveira, YSO population in N 113 were followed by

16 -69°22'12.0" -69°22'12.0" 3.0 3.0 Reg B+D CH3OH (~96.7 GHz) Reg B+D CH3OH (~96.7 GHz) peak peak B1 channel channel 19.2" B5 2.4 19.2" 2.4

B4 B3 (mJy beam B3 (mJy beam B2 B7 1.8 1.8 26.4" 26.4" B6 D4 1.2 1.2 -1 -1 Dec (J2000)Dec (J2000)Dec 33.6" contours: ) 33.6" contours: ) CH OH 1.3 mm D3 0.6 3 0.6 D2 D1 40.8" 40.8" 1 pc 0.0 1 pc 0.0 19.20s 16.80s 5h13m14.40s 19.20s 16.80s 5h13m14.40s RA (J2000) RA (J2000)

1.6 1.6 -69°22'40.8" Reg A CH3OH (~96.7 GHz) -69°22'40.8" Reg A CH3OH (~96.7 GHz) peak peak channel channel 42.6" 1.2 42.6" 1.2 (mJy beam (mJy beam

44.4" 44.4" 0.8 0.8 A1 A1 46.2" 46.2" -1 -1 Dec (J2000)Dec (J2000)Dec

A2 ) ) contours: 0.4 contours: 0.4 48.0" 1.3 mm 48.0" CH3OH

0.0 0.0 49.8" 1 pc 49.8" 1 pc 25.80s 25.20s 5h13m24.60s 25.80s 25.20s 5h13m24.60s RA (J2000) RA (J2000)

Figure 11: The single channel maps corresponding to the peak of the unresolved rotational transition quartet J = 2K − 1K of CH3OH at 96.7 GHz for regions in N 113 around and south of B3 (top) and toward A1 (bottom) hot cores (Sewiło et al., in prep.; ALMA 12m–Array data). The white 1.3 mm continuum contours in the left panel are the same as in Fig. 9. The white CH3OH contours in the right panel correspond to (3, 5, 7) × rms noise level of the CH3OH single-channel map of 0.32 mJy −1 00 00 beam . The ALMA beam size (2. 15 × 1. 66) for the CH3OH observations is shown in the lower left corner in all images.

ALMA Band 6 (∼1.3 mm) observations at a low-metallicity environment. COMs were de- ∼0.18 pc (or ∼0.008) resolution in the molec- tected toward two 1.3 mm continuum sources ular transitions that probe a wide density ‘A1’ and ‘B3’ in the field around Spitzer YSO–3 range (102–107 cm−3) to study both the dense (‘Region A’ in19) and YSO–1 (‘Region B’). Fig- clumps/cores and the lower density gas in the ure 9 shows three-color mosaics covering A1 and inter-clump regions. The ionized gas and out- B3 combining near-infrared bands with 1.3 mm flow tracers were also included in the program. contours overlaid, as well as a larger scale envi- These observations resulted in a serendipi- ronment in N 113 traced by the Spitzer 4.5 and tous discovery of COMs methanol (CH3OH), 8.0 µm, and Hα emission. Figure 10 shows the methyl formate (HCOOCH3), and dimethyl integrated intensity CH3OH, CH3OCH3, and ether (CH3OCH3) toward two locations in the HCOOCH3 images of Regions A and B. region (19; see Figs. 9–10). The detected tran- A1 and B3 are compact (diameter D ∼ 0.17 sitions are listed in Table 5. This was the pc) and are associated with H2O (A1 and B3) first conclusive detection of COMs more com- and OH (A1) masers. H2 column densities of plex than methanol outside the Galaxy and in (8.0±1.2)×1023 cm−2 and (7.0±0.9)×1023 cm−2

17 N 113 A1 SiO N 113 B3 SiO -69°22'42.6" 0.4 contours: 1.3 mm contours: 1.3 mm 0.3 -69°22'19.2"

(Jy beam CH OH (Jy beam 0.3 3 0.2 44.4" 0.2 B3

− 22.8" − H O 1 B4 1 2 s km s km B7 0.1 0.1 Dec (J2000)Dec 46.2" (J2000)Dec OH − − 1 1 ) 26.4" )

2.1" B6 2.1" 48.0" 0.5 pc 0.0 0.5 pc 0.0 25.44s 5h13m24.96s 18.48s 17.40s 5h13m16.32s RA (J2000) RA (J2000)

Figure 12: The SiO (5–4) integrated intensity images of N 113 A1 (left) and B3 (right) hot cores (Sewiło et al., in prep.). The contours are the same as in Fig. 9 (1.3 mm; white) and Fig. 10 (CH3OH; red). The positions of H2O and OH masers, and 1.3 mm continuum sources associated 00 00 with the CH3OH emission are indicated. The ALMA beam size (0. 98 × 0. 61) is shown in the lower left corner in both images.

6 −3 and number densities of ∼1.6×10 cm and (see Section ). Table 3 lists CH3OH, CH3OCH3, 6 −3 ∼1.4×10 cm for A1 and B3, respectively, HCOOCH3 fractional abundances for a set of were estimated using the 1.3 mm continuum Galactic hot cores. When scaled by a factor of data. The LTE analysis of six CH3OH tran- 2.5 to account for the lower metallicity in the sitions resulted in rotational temperatures and LMC (the ratio between the solar metallicity total column densities (Trot, Nrot) of (134 ± and the metallicity of the LMC, assuming ZLMC 16 −2 6 K, 1.6±0.1×10 cm ) and (131±15 K, 6.4± = 0.4 Z ), the abundances of COMs detected 0.8 × 1015 cm−2) for A1 and B3, respectively.19 in N 113 are comparable to those found at the These sizes, number and column densities, and lower end of the range in Galactic hot cores. temperatures of A1/B3 are consistent with clas- This was a surprising result because previous sic hot cores observed in the Galaxy. COMs observational and theoretical studies indicated emission and association with masers are also that the abundance of CH3OH in the LMC is among the main characteristics of hot cores. very low.19 concluded that COMs observed in These are the ﬁrst bona ﬁde detections of com- the N113 hot cores could either originate from plex hot core chemistry outside the Galaxy. grain surface chemistry or in post-desorption The (CH3OH, CH3OCH3, HCOOCH3) col- gas chemistry. umn densities are (16 ± 1, 1.8 ± 0.5, 1.1 ± 0.2) × The quartet J = 2K −1K of CH3OH at ∼96.7 1015 cm−2 for A1 and (6.4 ± 0.8, 1.2 ± 0.4, < GHz was later detected serendipitously toward 0.34) × 1015 cm−2 for B3. Column densities for N 113 with ALMA in the project targeting two HCOOCH3 and CH3OCH3 were estimated us- transitions (1–0 and 2–1) of three CO isotopes 12 13 18 ing the same Trot as for CH3OH, assuming that ( CO, CO, and C O) toward three LMC these molecular species are located in the same and three SMC star-forming regions with a res- 00 region as CH3OH in A1 and B3. olution of ∼2 (or ∼0.48 pc). N 113 is the The (CH3OH, CH3OCH3, HCOOCH3) frac- only region in this sample with CH3OH detec- tional abundances with respect to H2 are (20 ± tion. The spectral window with the CH3OH 3, 2.2 ± 0.7, 1.4 ± 0.4) × 10−9 for A1 and (9.1 ± line was dedicated to the continuum observa- 1.7, 1.7 ± 0.7, < 0.5) × 10−9 for B3 (see Ta- tions and has a spectral resolution of ∼50 km −1 ble 2). The CH3OH fractional abundances in s . Such low spectral resolution does not al- A1/B3 are over an order of magnitude larger low for any quantitative analysis, but surpris- than an upper limit estimated for ST 11 by67 ingly the data revealed several additional well-

18 Table 3: Chemical Abundances Relative to H2 for COMs Detected in the LMC/SMC for Selected Galactic Hot Cores

Source N(X)/N(H2) a CH3OH HCOOCH3 CH3OCH3 Ref.

Orion Mol. Cloud: Hot core 1.4 × 10−7 1.4 × 10−8 8.0 × 10−9 1 Compact Ridge 4.0 × 10−7 3.0 × 10−8 1.9 × 10−8 1 G34.26−0.15: NE 8.5 × 10−7 7.3 × 10−8 1.4 × 10−7 1 SE 6.4 × 10−7 6.8 × 10−8 8.5 × 10−8 1 Sgr B2N 2.0 × 10−7 1.0 × 10−9 3.0 × 10−9 1 G327.3−0.6 2.0 × 10−5 2.0 × 10−6 3.4 × 10−7 1 AFGL 2591 7.0 × 10−7 <3.2 × 10−7 <1.0 × 10−7 2 G24.78+0.08 6.5 × 10−7 7.3 × 10−8 3.0 × 10−7 2 G75.78+0.34 9.2 × 10−7 5.9 × 10−8 1.9 × 10−7 2 NGC 6334 IRS 1 4.0 × 10−6 4.6 × 10−7 2.4 × 10−6 2 NGC 7538 IRS 1 5.7 × 10−7 6.7 × 10−8 <7.6 × 10−8 2 −6 −7 −7 W3(H2O) 5.4 × 10 2.9 × 10 8.3 × 10 2 W33A 7.3 × 10−7 9.6 × 10−8 1.0 × 10−7 2 a References: (1)88 and references therein; (2).89 For more information on Galactic hot cores, see also, e.g.,90–92 and references therein.

defined CH3OH peaks and extended emission dance of all deuterated species strongly depends in Region B (see Fig. 11). The CH3OH emis- on temperature: it drops rapidly with increas- sion reported in19 was limited to two compact ing gas temperature as an object evolves (see, sources (hot cores A1 and B3) with some fainter e.g.,93–96). emission detected in other locations in Region B (see Fig. 10); however, it was unclear whether the latter traces physically distinct sources or Theory shocked lobes of the outflows. We can gain some insight as to the origin of In addition to hot cores, the ∼96.7 GHz the COMs detected in the Magellanic Clouds CH OH emission is detected toward continuum 3 by comparing observations with current astro- sources B4, B7, and B6 in Region B and D1– chemical theories of Galactic COM production D4 located toward the south (Fig. 11; Sewiło in star-forming regions. Here we briefly review et al., in prep.). In Region A, the ∼96.7 GHz COM formation mechanisms and how they have CH OH emission peak coincides with hot core 3 been applied to low-metallicity environments. A1; however, the emission extends toward the south. The ∼96.7 GHz CH3OH emission peaks in Complex Molecule Formation in Regions A and B overlap with the SiO (5–4) Star-forming Regions of the Milky peaks (with some faint extended emission visi- Way ble throughout Region B, see Fig. 12;19, Sewiło et al., in prep.), tracing shocks. The same re- In the Galaxy, the largest COMs have until now gions are associated with the DCN (3–2) emis- been found primarily in the hot molecular cores sion indicating their youth and are distributed associated with massive protostars, as well as along the dense gas filament. The observational with the so-called hot corinos found in regions and theoretical studies indicate that the abun- of low-mass star formation (e.g.,91). In cold,

19 dense molecular clouds some of these organics (4) in situ formation in hot core gas following have been known to be present for quite some ice sublimation. time: HNCO, HCOOH, CH2CO, CH3CHO, 97 CH3OH. More recently, the larger ‘hot core’ Cold Gas–Phase Reactions in Dark COMs, including CH3OCH3 and HCOOCH3, Clouds have also been detected in several cold clouds (98–102). These detections have raised several Cosmic-ray ionization of molecular hydrogen challenges for existing theories of COM forma- drives dense cloud chemistry by initiating se- tion. These include the long-standing prob- quences of ion-neutral and neutral–neutral re- lem of identifying the mechanism responsible actions. This chemistry leads to the produc- for returning ice-mantle molecules to the gas tion of CO and many of the simple molecules in cold clouds, and the fact that radical reac- detected in dark interstellar clouds. The com- tions on 10 K grains will not occur. In the case plex molecules produced tend to be long hydro- of the cold methanol detected in the LMC, it carbon chains and associated radicals, such as would appear that reactive desorption (i.e., us- the cyanopolyynes and various carbenes. The ing part of the energy released upon formation) dark cloud TMC-1 in Taurus is recognized as 111 is a plausible explanation.103,104 the best example of this chemistry (e.g., ) and Hot cores are regions of high gas density chemical modeling demonstrates that gas-phase and extinction where elevated dust tempera- chemistry can account quite well for its com- 112 tures have led to the sublimation of molecu- position (e.g., ). It is now appreciated that lar ice mantles that have previously grown on methanol is formed entirely on dust grains (see cold dust grains. The first model of hot core Section ) and so some (non-thermal) desorption chemical evolution was that of 105 . This in- mechanism is required to explain its presence volves a long period of cold (∼10 K) chemistry in dark clouds like TMC-1 since dust temper- during which molecules form by gas–phase re- atures are never sufficiently high for methanol actions and on dust grains, as the gas freezes and water to desorb. However, in dense cores out; this chemistry is similar to that found in where low-mass protostars are forming, it ap- cold dark clouds (e.g.,106). At some point grav- pears that some dust heating (to ∼30–35 K) has itational collapse ensues, a protostar is formed, occurred and this can lead to the sublimation and the resulting large increase in luminosity of more volatile molecules. In this Warm Car- heats the surrounding envelope of gas and dust bon Chain Chemistry (WCCC), sublimation of to temperatures above about 100 K, leading large abundances of methane from the ice man- to the sublimation of ice mantles and the for- tles greatly increases the efficiency of carbon- 113 mation of the hot core/corino. Further refine- chain growth. Although we do not discuss ments to this picture have been (a) the addi- CH4 injection further, the similarly warm dust tion of a slow ‘warm-up’ phase that can allow temperatures found in the Magellanic Clouds UV photolysis and radical–radical reactions to may make this a viable route to widespread car- occur in the ice mantles, prior to formation of bon chain formation there. the hot core proper (107), and (b) gas–phase synthesis in the hot core, driven by the sub- Cold Grain–Surface Reactions 108 limated molecules ( ). Further refinements Ice mantles grow on cold dust grains through have included the spatio-temporal evolution of the sticking of atoms of O, C and N fol- the gas–grain chemistry (109,110). lowed by reactions with H atoms to form H2O, Thus, within this simple picture, the complex CH and NH . Molecules formed in the gas, molecules observed in hot cores can have four 4 3 such as CO and N2, can also accrete and be- chemical origins: (1) formation in cold gas fol- come incorporated in the mantle. Unsaturated lowed by freeze out on dust; (2) formation on molecules can undergo (tunneling) addition re- the surfaces of cold (10 K) dust grains; (3) for- actions with hydrogen atoms to produce molec- mation in UV-photolyzed ice mantles at ∼30 K;

20 ular radicals that can further react and form shown to be a viable mechanism. In these new molecules (e.g.,114). Figure 13 shows that, experiments, the trans–HCOOH conformer of in the case of CO, these processes can lead to formic acid produced the protonated form of 115 a rich organic chemistry (e.g., ) through the trans–HCOOCH3; electron dissociative recom- growth of linear chains by single atom addi- bination, or proton transfer (see below), would tions and their subsequent saturation. This sur- then lead to the neutral ester. However, apart 130 face scheme can explain the presence of CH3OH from one detection of trans–HCOOCH3 , it and many of the COMs known in hot cores is the more stable cis–HCOOCH3 conformer and originally made predictions116 for new sur- that is found in hot cores. It has therefore face molecules (not shown) that were subse- been suggested that interconversion between 117,118 quently detected: glycolaldehyde , ethy- trans–HCOOCH3 and cis–HCOOCH3 struc- 119 131 lene glycol ((CH2OH)2, ), acrolein and propi- tures could occur in the proton loss process. onaldehyde.120 Numerous surface chemistry ex- Chemical models have demonstrated that this periments have validated many of these addi- is a very competitive pathway to HCOOCH3 tion processes.121–123 around protostars132, assuming that interconversion proceeds with high efficiency.133 Of the COM Chemistry Driven by Sublimated COM products shown in Fig. 13, apart from Ice Mantles dimethyl ether and methyl formate, both ethyl formate and ethyl methyl ether have now been 108 demonstrated that differences in mantle detected.134,135 composition could, following sublimation , ac- A problem with this scenario has been that count for the chemical differentiation seen in experiments on electron dissociative recombi- the Orion–KL star-forming region. Rather nation of protonated COMs show that recovery than calculate the mantle composition ab ini- of the neutral molecule occurs with a probabil- tio, they assumed a composition based on ity much lower than previously assumed, typi- observations and showed that the relative cally < 5% as compared to 100% (e.g.,136), and presence, or not, of methanol and ammonia subsequently to COM abundances much lower could drive different chemistries in the hot gas than observed. However, if NH is injected from 124 125 126 3 (see also ; ). first suggested that self- grain mantles at moderate abundances, compa- methylation of methanol could be the source rable with those measured in Galactic ices137, of dimethyl ether in hot cores, although they then proton transfer, e.g. attributed its origin to be protostellar outflows rather than grain-surface chemistry. The self- + + CH3OH2 + NH3 → CH3OCH3 + NH4 (2) methylation reaction is can dominate COM formation and mitigate + + CH3OH2 + CH3OH → CH3OCH4 + H2O (1) against the destructive effects of electron recombinations.132,138 where the neutral CH3OCH3 molecule is pro- As also shown in Fig. 13, gas phase COM for- duced in reactions with either an electron or mation driven by alkyl cation transfer reactions a base, such as ammonia. Based on labora- involving surface-formed HNCO and NH CHO 127 2 tory experiments, showed that, apart from could plausibly be a source of new molecules, CH3OCH3, many more large ethers and es- perhaps explaining the hot core detections of ters could be formed in this manner, in re- methyl isocyanide (CH3NCO) and acetamide actions involving ethanol, propanol and bu- (NH COCH ).139,140 However, the associated 128 2 3 tanol. Figure 13 includes alkylation reactions gas phase reactions have not yet been studied between protonated methanol and ethanol, the experimentally or theoretically and so should neutral alcohols, and formic acid. Methyl for- be regarded as speculative. mate formation by methylation of HCOOH has also been studied in the laboratory129 and

21 + C2H5OH2 CH3OC2H5 (CH3)2O C2H5OH (C2H5)2O

CH OH + + 3 2 2H C2H5OH2

CH3OH CH3CHO

2H 2H

H2CO CH2CO

H H • • CO H HC=O C HC=C=O O N O

+ • CH3OH2 CO2 O=C-OH HNCO CH3NCO

H H + C2H5OH2 • HCOOH NH2C=O C2H5NCO

+ + C2H5OH2 CH3OH2 H

+ CH3OH2 HCOOC2H5 HCOOCH3 NH2CHO NH2COCH3

+ C2H5OH2

NH2COC2H5 Figure 13: Interstellar gas–grain surface chemistry. Molecules involved in surface reactions are in black. Hydrogen atom addition to unsaturated molecules creates reactive radicals and additions of C, O and N atoms, allows a rich organic chemistry seeded by carbon monoxide to develop. Broken arrows indicate reactions with activation energy barriers; where 2H is shown, a barrier penetration reaction followed by an exothermic addition is implicitly indicated. Once these ice mantles have been sublimated into the hot core gas, methanol and ethanol can become protonated and take part in ion–molecule alkyl cation transfer reactions with themselves and with other molecules that were formed in the ices. These processes are denoted in blue and each arrow indicates the initial ion–molecule reaction and the production of the neutral COM, either through electron dissociative recombination or proton transfer to ammonia (adapted from115 and references therein).

22 Radical Reactions in Icy Grain Mantles nations between HCO, CH3O, CH2OH, COOH and other simple radicals (e.g., CH , CH , NH , As well as chemical reactions on grain surfaces 3 2 2 CN, OH) then produce many of the COMs ob- driven by accretion of reactive species, it is also served in hot cores.134,152,153 known that the ices can undergo bulk processing due to radiation damage caused by photons and energetic particles. Cosmic rays in- Models of COM Formation in the teract directly with icy mantles and recently Magellanic Clouds chemical models have been developed to ex- plore their effects on COM formation (see141 Here we summarise and discuss theoretical and references therein). The environment of models of COM formation at reduced metal- young protostars receives large doses of UV licity in the light of recent observations. and EUV radiation.142 Even in dark clouds, a weak ambient flux of internal UV photons Chemical Models of the LMC and SMC 143 will exist as energetic electrons produced 154 considered the chemical evolution of pu- by cosmic-ray ionization of molecular hydrogen tative dark clouds in the LMC and SMC at can subsequently collide with, and excite, other the appropriate elemental depletions. They H2 molecules which undergo de-excitation by considered the purely gas-phase chemistry of emission of Lyman and Werner band photons. fairly simple molecules, with only CH OH and 3 −1 3 The resulting UV flux is ∼ 10 photons s CH CO being the only COMs. These models −2 2 cm which can be significant for chemistry in −8 could produce CH3OH abundances of ∼10 ; dense clouds. Laboratory experiments on UV however, the main formation mechanism was photolysis of interstellar ice analogs show that through a radiative association process that is an extensive organic chemistry can occur, pri- no longer considered viable. marily through reactions involving various radi- More recently,155,156 developed gas–grain als produced from the major (parent) ice species models of the LMC and SMC. Again, cold 144 (e.g. ) . Cold H additions to CO (see Section ) dark clouds were modeled for a wider range have also to be considered since interstellar or- of physical parameters, including appropriately ganic synthesis by photolysis requires that the warmer gas and dust temperatures. In these methanol be present ab initio .144,145 models CH3OH was formed by CO hydrogena- Early theoretical models for organic synthe- tion on dust grains. For the LMC, they found sis from the recombination of radicals on cold that molecular abundances similar to Milky grains appealed to physically unrealistic pro- Way dark clouds could be obtained, whereas cesses to attain the high mobilities needed for for the SMC the abundances were lower for 146–150 heavy particle migration and reaction. most molecules. The model predictions were Detailed modeling has shown that sustained compared to several star-forming regions in the UV photolysis and warming during the early LMC (as no data on Magellanic dark clouds stages of hot core evolution can produce many then existed): N 159W, N 159S, N 160, and 30 107 new organics. In this picture, the ices are Dor–10. When compared to the Magellanic subjected to the Prasad–Tarafdar UV photons source N 159W with a thermal methanol detec- during the initial cold phase of core formation, tion, the calculated abundances were found to producing a population of radicals. This irradi- be ∼10–103 times lower than determined by.17 ation can continue as the core is being grad- The reason for this discrepancy was explained ually (as opposed to instantaneously) heated as being due to the warm dust temperatures, 107,151 – the so called ‘warm-up’ phase. When appropriate for Magellanic dark clouds, that the dust grain temperature reaches, and can were considered in the models. Dust temper- be maintained close to, about 30 K, the asso- ature affects grain-surface chemistry and can ciated increase in radical mobility allows them inhibit methanol formation since, for tempera- to migrate and react. Radical–radical recombi- tures above about 15 K, hydrogen atoms will

23 Table 4: Model Ice Mantle Compositionsa

Molecule LMC SMC −5 −5 H2O 4.5 × 10 2.2 × 10 CO 1.7 × 10−5 7.6 × 10−6 −5 −6 CO2 1.3 × 10 6.0 × 10 −6 −6 CH4 2.1 × 10 1.0 × 10 −6 −7 NH3 1.0 × 10 3.2 × 10 −6 −6 N2 3.2 × 10 1.0 × 10 −6 −7 H2CO 1.1 × 10 5.0 × 10 −6 −6 CH3OH 3.2 × 10 1.4 × 10 HCOOH 6.4 × 10−7 3.2 × 10−7 a Abundances are given relative to total density of hydrogen nuclei.

desorb before they can react with CO. Above produced the observed CH3OH abundances in about 25 K, CO molecules residing in CO-rich both A1 and B3 at times (post- sublimation outer layers of the ice mantle will also sublimate ) when CH3OCH3 and HCOOCH3 had non- , whereas those trapped in the H2O ice matrix trivial abundances. As CH3OH is the proposed 157 require higher dust temperatures. In both parent of CH3OCH3 and HCOOCH3 in both the LMC and SMC models, gaseous NH3 abun- grain-surface reactions (see Section ) and in dances of ∼10−8 were found. For models with ion–molecule reactions (see Fig. 13) the com- dust temperatures more similar to the Milky parison of predicted CH3OCH3/CH3OH and Way (∼10–15 K), the dust ice mantles are pre- HCOOCH3/CH3OH ratios with observations dicted to contain large fractions of CH3OH provides the most stringent tests of their ori- 159 (∼10–40%) and NH3 (∼1–7%). Hence, the gin (see Section ). The relevant model presence of CH3OH in the Magellanic Clouds yields HCOOCH3/CH3OH ratios of ∼0.04 points to formation in a cold pre-hot core phase. in good agreement with that of both N 113 158 studied the formation of ice mantles in cores; whereas the CH3OCH3/CH3OH ratios the Magellanic Clouds but instead employed a are about 5 times greater than observed. correct stochastic calculation of the grain chem- Overall, these models indicate that the hot istry. The relative composition of ices compris- cores in the Magellanic Clouds must have un- ing of H2O, CO, CO2, CH3OH, NH3 and CH4 dergone a period of chemical evolution when were calculated as a function of the interstellar the dust was cold, as in theories of Galactic hot UV radiation field. The CH3OH and NH3 ice core formation. fractions were found to be lower (∼ few%) than those found by155,156 and more compatible with Models for COM Ice–Gas Synthesis those of Galactic ices.137 159 reported the first bona fide hot core mod- Given the importance of NH3 for gaseous COM els of the Magellanic Clouds. Their model formation, the Magellanic Clouds offer a po- was a standard three-phase model (cold qua- tential opportunity to test this scenario in 159 sistatic phase, collapse phase with out without a low-metallicity environment. emphasized ‘warm-up’, and hot core phase; see Section ) the sublimation of COM-containing ice mantles and included the radical chemistry induced into the hot gas, not active formation in situ, as during the ‘warm-up’ phase as well as many the possible source of Magellanic COMs. Post- of the COMs known in Galactic hot cores sublimation gas-phase synthesis (see Section ) (107,152; see Section ). When compared to the has not yet been studied in the context of the N 113 observations, only one of the models (a Magellanic Clouds and here we report relevant 100 K core in which the initial phase of core model calculations as a first approximation. formation proceeded at 10 K, Model 1) re- The calculations were performed with the

24 -4 -4 10 H2O 10 H2O CO CO CH3OH -6 -6 10 HCOOH 10 CH3OH HCOOH

H CH3OCH3 C2H5OH NH3 H C2H5OH NH3

(i)/n -8 (i)/n -8 10 10 CH3OCH3 gas gas n n HCOOCH 3 HCOOCH3

10-10 10-10

10-12 10-12 102 103 104 105 102 103 104 105 Time [years] Time [years]

Figure 14: Chemical evolution in hot core models of the LMC (left panel) and SMC (right panel). Ice mantles with the compositions of Table 4 are instantaneously injected into the gas at t = 0.

-5 10 -6 XNH3 = 10 -7 XNH3 = 10 NH3 -8 -6 XNH3 = 10 10 -9 XNH3 = 10 -10 XNH3 = 10 CH3OCH3 10-7 H

-8

(i)/n 10 gas n

10-9

10-10 HCOOCH3

10-11 102 103 104 105 Time [years]

Figure 15: Evolution of CH3OCH3 and HCOOCH3 abundances in the LMC model as a function of the NH3 abundance sublimated from ices.

25 dynamical-chemical model of132 but modified the level measured by36, is halted as electron according to the simplest version of these mod- dissociative recombinations come to dominate els in which the physical conditions are fixed in the loss of molecular ions. Although models time and the ices are injected instantaneously of ice formation in the Magellanic Clouds do 108 into the hot gas at t = 0 (cf. ), as op- predict NH3/H2O mantle fractions sufficiently posed to computing the physical and chem- large to allow efficient ion–molecule formation ical evolution of the core (e.g., gravitational in the LMC (see Section ), NH3 observations collapse, protostellar luminosity heating, time- of N113 and other COM-containing regions are dependent molecule desorption). Table 4 lists clearly required to validate the efficacy of this the assumed ice mantle compositions. These process. were obtained by scaling the ice abundances The observed abundances of CH3OCH3 and 132 from Table 1 of to the elemental depletions HCOOCH3 in the A1 and B3 hot cores of observed in the SMC and LMC clouds as listed N 113 (see Table 3) can be reproduced in the 155 156 in Table 1 of (fLMC,i) and (fSMC,i), respec- models of Fig. 14 despite the lower elemen- tively. The (C, O, N) scaling factors (1/f) are tal depletions. However, as in the case of (0.41, 0.46, 0.21) for the LMC and (0.20, 0.22, Galactic hot core models these do not oc- 0.064) for the SMC; for each molecule listed in cur at a time when the CH3OCH3/CH3OH Table 4, we used the averaged 1/f of its corre- and HCOOCH3/CH3OH abundance ratios have sponding elements to determine initial ice abun- their observed values (Fig. 16). In the LMC, dances in the LMC/SMC from their Galac- when the methanol abundance is ∼10−8, the ob- 132 tic values used by. Typical hot core physi- served CH3OCH3/CH3OH ratios in A1 (∼0.1) 7 cal conditions were assumed: nH2 = 5 × 10 and B3 (∼0.2) are about an order of magni- cm−3, T =150 K, and a cosmic-ray ionization tude higher than the calculated values (∼0.01); −17 −1 rate ζ = 3×10 s . The chemical network is for HCOOCH3/CH3OH the calculated value of as discussed in.132 Figure 14 shows COM chem- (∼0.003) is about a factor 20 too low. ical evolution in hot core models applicable to 159 claimed a very good agreement with the the LMC and SMC. Generally, one can notice a abundances of19 to within a factor of 5. How- decrease of all gas phase abundances by a factor ever, calculated abundances and the observed of a few depending on the elemental depletion abundances were presented as relative to to- of C, N, and O in the two galaxies. However, tal hydrogen density and molecular hydrogen the stronger N depletion relative to C and O (by density, respectively. Correcting the abundance a factor of 2 and 4 for the LMC and SMC, re- comparison by the factor of 2 means that for spectively) induces a more pronounced decrease CH3OCH3 in A1 and B3, the models over- of NH3 relative to other ice species. predict the abundances by factors of 12–20; 132 As found by , a decrease of the injected only the calculated abundances of CH3OH and ice mantle NH3/CH3OH abundance ratio from HCOOCH3 in A1 agree to within a factor of 5. ∼1 decreases the production efficiency of com- When a detailed treatment of the collapse plex organic molecules; for CH3OCH3 and physics is included in hot core chemistry mod- HCOOCH3 the Magellanic Cloud abundances els, a monotonic increase in the radial gas and are lower by a factor of 3 for the LMC and of dust temperatures occurs as a function of the 5 for the SMC. The NH3 abundance is not ob- increasing (accretion) luminosity of the pro- servationally well-constrained in the LMC and tostar (e.g.,109). In the case of Galactic hot SMC and is unknown in the N 113 region. In cores/corinos,132 showed that molecular recon- the LMC,36 measured an ammonia abundance densation may be just as chemically important of ∼4×10−10 in N 159W and Fig. 15 shows that as molecular depletion onto dust in cold cores, systematically reducing the injected NH3 abun- and sublimation from them in hot cores. This dance from ices dramatically lowers the COM can occur if the accretion onto the protostar abundances. However, any decline in COMs is not monotonic but sporadic.160 Once grains −8 below an NH3 abundance of ∼10 , down to have been raised to a temperature of about

26 10-1

CH3OCH3

10-2 OH) 3 HCOOCH3 X/X(CH 10-3

10-4 10-9 10-8 10-7 10-6 10-5

X(CH3OH)

Figure 16: Evolution of the CH3OCH3/CH3OH and HCOOCH3/CH3OH abundance ratios in the models of Fig. 14 for the LMC (solid curves) and SMC (dashed curves).

120 K or more, most of the ice mantle is re- Discussion moved and, as they cool to 50 K or lower in about 100 years, selective recondensation can Although single-dish observations provided the occur due to the differing binding energies for first detections of methanol in two bright re- physisorption. Production of organic molecules gions in the LMC (N 159 and N 113), it was ensues in the brief post- sublimation period, as ALMA with its unprecedented angular reso- in the standard model, except that the products lution and sensitivity that has started and subsequently condense as ices.132 demonstrated continue revolutionizing the field of complex that the observed large CH3OCH3/CH3OH and organic chemistry in low-metallicity environ- HCOOCH3/CH3OH ratios could be reproduced ments. To date, the sample of sources with due to the relatively higher binding energy of COMs detection is very small and diverse. CH3OH. This may be a fundamental process Only two bona fide hot cores with COMs are for the organic chemistry of material near pro- known in the Magellanic Clouds, both in the tostars as, between sublimation–recondensation N 113 star-forming region in the LMC (N 113 events, there can be a mantle-driven gaseous A1 and B3;19). Another hot core in the LMC chemistry leading to the dust grains being (ST 11) was found by67, but no methanol or coated with the products as they cool. This other COMs were detected. The detection of scenario remains to be evaluated in the context methanol was reported for two other sources, of Magellanic Cloud hot cores. one in the LMC (N 159W–S, Section ) and one in the SMC (IRAS 01042−7215,20), but not in hot cores (‘cold methanol’). Several sources in N 113 (in addition to two hot cores) are associated with methanol, but due to a very low spectral resolution of the observations, a deter-

27 mination of their physical parameters based on likely formed by gas–phase chemistry following the existing data is currently impossible. an event of ice desorption.102 However, the low No firm conclusions about complex organic yield of COMs from ice desorption at cold tem- chemistry under reduced metallicity conditions peratures is at most a few percent of methanol can be drawn based on the current sample of even at Galactic metallicities, thus consistent sources with the detection of COMs. However, with non-detections of COMs towards N 159W– some general trends are noticeable. S, IRAS 01042−7215 P2 and P3. None of the sources with a COMs detection For IRAS 01042−7215 where the CH3OH is associated with an infrared source: COMs emission is detected toward two sources corre- are detected toward mm continuum sources in sponding to the continuum peak,20 discussed the vicinity of bright Spitzer YSOs. ST 11 with three possibilities for the nature of P2/P3 (their no COMs detection but with evidence for hot ‘East core’) where cold methanol was detected core chemistry is the only source associated – they suggested it can be either a massive star- with the infrared source. In fact, this is the less core, an embedded high-mass YSO (or mul- brightest Spitzer YSO in the entire N 144 H ii tiple YSOs) before the emergence of infrared region51; higher resolution near-infrared obser- emission, or a cluster of low-mass embedded vations reveal a cluster with one dominating YSOs. source (see Fig. 2), which likely has the largest Observations confirm that COMs can form in contribution to the emission at longer wave- extragalactic environments. Formation path- lengths. Multiple mm continuum sources are ways have been suggested that involve either detected toward all the other fields. The N 113 grain–surface chemistry or gas–phase chem- B3 field is particularly interesting since it har- istry, or a combination of both. Theoretical bors sources at different evolutionary stages in- models accounting for the physical conditions cluding a hot core and ultra-compact H ii re- and metallicity of hot molecular cores in the gions with younger sources distributed along a Magellanic Clouds have been able to broadly dense gas filament (see Figs. 9 and 12). Several account for the existing observations. mm continuum sources including one associated As cosmic metallicity is increasing with time, with the brightest CH3OH peaks in the region, understanding interstellar chemistry in low are also found along the filament in N 159W–S metallicity environments also gives us insight (see Figs. 4–5 and75). into the chemistry of the past Universe. The de- The detection of cold methanol in N 159W–S tection of COMs in the Magellanic Clouds has and IRAS 01042−7215 P2/P3 is puzzling, how- important implications for astrobiology. The ever, similar results are being obtained in our metallicity of the LMC/SMC is similar to the Galaxy. In Galactic dark clouds, gas–phase mean metallicity of the interstellar medium methanol is found to have a clumpy distribu- during the epoch of peak star formation in the tion, often not following the density distribu- Universe, thus the presence of COMs in these tion traced by continuum and molecular trac- systems indicates that a similar prebiotic chem- ers161, and thus provides evidence of rapid ice istry leading to the emergence of life, as it hap- desorption processes. This methanol tends to pened on Earth, is possible in low-metallicity be cold, at excitation temperatures of 5–20 systems in earlier epochs of the Universe. Ac- K, similar to what is observed in N 159W– cording to one of the theories on the emergence S and IRAS 01042−7215. In the Barnard 5 of life on Earth, interstellar COMs (after incor- cloud in particular, a cold methanol clump with porating into comets) might have been deliv- the CH3OH abundance with respect to H2 of ered to early Earth to provide important ingre- ∼ 4 × 10−8, offset from the IRAS sources, has dients for the origin of life (e.g.,1,94,163). proven to be a signpost of ice desorption – COMs more complex than CH3OH are de- first by the detection of cold gas–phase water tected in the LMC toward hot cores which are at abundances similar to methanol162 and then associated with massive stars and therefore are other COMs like CH3CHO and HCOOCH3, short-lived; a detection of COMs does not have

28 a direct relation to the origin of life at these lo- the spectral setup covering the most important cations. However, if COMs are detected toward molecules. hot cores, we can expect them to be present in What new COMs await discovery in the Mag- hot corinos, which are associated with low-mass ellanic Clouds? Chemical models of the LMC stars. The evolution of low-mass stars is long have not been successful (to within an order of enough for life to emerge if favorable conditions magnitude) in accounting for the observed pres- exist. ence of dimethyl ether in the LMC (see Sec- tion ) and so may not be reliable indicators for new molecules. At present, taking the in- Future Prospects ventory of well-studied Galactic hot cores may provide the best guidelines for future searches. Systematic studies on YSOs in different envi- Molecules that could be expected to present in- ronments in the LMC and SMC are needed to clude acetaldehyde, ethanol, ethyl cyanide, iso- build a statistically significant sample of sources cyanic acid, and formamide. with COMs detections in the Magellanic Clouds The NASA’s James Webb Space Telescope that would allow us to draw reliable conclu- (JWST; scheduled to launch in 2021) near- sions on the formation and evolution of COMs (0.6–5.3µm; the Near Infrared Spectrograph – in reduced metallicity environments. Both the NIRSpec instrument) to mid-infrared (4.9–28.8 LMC and SMC samples will be important to µm; Mid-Infrared Instrument – MIRI) obser- get as wide a range of metallicities as possi- vations will enable studies of the solid phase ble. In the SMC, the metallicity is lower than material in different Galactic and extragalactic in the LMC (as low as 0.1 Z in the SMC tail environments. The high sensitivity and angular where star formation occurs in pockets; e.g.,164) resolution of JWST will enable detailed studies and the formation and survival of COMs can be of hot cores in the Magellanic Clouds on ∼0.04 tested in an even harsher environment. Enlarg- pc scales in the near-infrared, without contam- ing the sample in both galaxies is important ination from the surrounding regions. It will be to distinguishing between local environmental possible to do an inventory of simple (e.g., H O, conditions in individual clouds, and the more 2 CO , CO) and complex (CH OH and beyond, global effect of metallicity. 2 3 including molecules of prebiotic significance) Reliable determination of Magellanic Clouds ices in the low-metallicity environments in the hot cores’ physical and chemical properties and Magellanic Clouds. For example, in the 5–8 µm a comparison between the two Clouds and the range (MIRI) COMs such as ethylene glycol, Galaxy are of paramount importance to un- glycolaldehyde, methyl formate, and dimethyl derstand the impact of metallicity on complex ether (the latter two already detected in N 113 organic chemistry. High-resolution and sensi- A1/B3 in the gas phase with ALMA), and oth- tivity observations of known hot cores in the ers can be detected.165 The laboratory exper- Magellanic Clouds as their number starts in- iments are underway predicting which COMs creasing are a natural next step in reaching this can be detected at shorter wavelengths covered goal. This direction of research involves chem- by NIRSpec. Such studies will provide infor- ical modeling of hot cores to understand their mation on differences in chemical complexity of chemical evolution. hot cores in different environments. ALMA programs are underway targeting tens of YSOs identified by Spitzer and Herschel, both specifically designed to search for COMs and those with different science goals, but covering frequency ranges where COMs can be detected (serendipitous detections as in N 113). ALMA’s broad spectral windows allow to ob- serve many spectral lines simultaneously with

29 Table 5: COM Transitions Detected in the Magellanic Clouds with ALMAa

b Species Transition Frequency EU Detection (GHz) (K) LMC N 113 A1 B3 CH3OH vt=0 51,4–42,2 E 216.94560 55.87 + + − CH3OH vt=1 61,5–72,6 A 217.299205 373.93 + + CH3OH vt=0 201,19-200,20 E 217.88639 508.38 + − − CH3OH vt=0 102,9-93,6 A 231.28110 165.35 + + + CH3OH vt=0 102,8-93,7 A 232.41859 165.40 + + + CH3OH vt=0 183,16-174,13 A 232.78350 446.53 + + c CH3OH vt=0 2−1,2–1−1,1 E 96.739362 12.54 + + + 20,2–10,1 A 96.741375 6.97 20,2–10,1 E 96.744550 20.09 21,1–11,0 E 96.755511 28.01 HCOOCH3 v=0 182,16–172,15 E 216.83020 105.68 +? − HCOOCH3 v=0 182,16–172,15 A 216.83889 105.67 +? − HCOOCH3 v=0 201,20–191,19 E 216.96476 111.50 + +? HCOOCH3 v=0 201,20–191,19 A 216.96590 111.48 + +? HCOOCH3 v=0 200,20–190,19 E 216.96625 111.50 + +? HCOOCH3 v=0 200,20–190,19 A 216.96742 111.48 + +? CH3OCH3 130,13–121,12 EE 231.98782 80.92 + + LMC N 159W–S MMS–3d CH3OH vt=0 50,5–40,4 E 241.700219 47.93 + CH3OH vt=0 5−1,5–4−1,4 E 241.767224 40.39 + + CH3OH vt=0 50,5–40,4 A 241.791431 34.82 + CH3OH vt=0 51,4–41,3 E 241.879073 55.87 + e CH3OH vt=0 5−2,4–4−2,3 E 241.904152 60.72 + 52,3–42,2 E 241.904645 57.07 SMC IRAS 01042−7215 P2 P3 + CH3OH vt=0 50,5–40,4 A 241.791431 34.82 + + CH3OH vt=0 5−1,5–4−1,4 E 241.767224 40.39 + + CH3OH vt=0 50,5–40,4 E 241.700219 47.93 +? − e CH3OH vt=0 5−2,4–4−2,3 E 241.904152 60.72 + − 52,3–42,2 E 241.904645 57.07 a The table uses the CDMS notation from the Splatalogue. See footnotes to Table 1 for details; b The symbols in these columns indicate a detection (‘+’), a tentative detection (‘+?’), or a non-detection (‘−’) of a given molecular line transition; c The four transitions at ∼96.7 GHz are blended; the spectral resolution of these observations which were dedicated to the continuum detection is ∼50 km s−1. In addition to hot cores A1 and B3, there are several other ∼96.7 GHz d CH3OH peaks in N 113 (see Section ); The two methanol peaks associated with MMS–3 where the largest number of CH3OH lines are detected and the temperature ﬁtting is the most reliable e (see Section ); The CH3OH lines at 241.904152 GHz and 241.904645 GHz are blended.

30 Technical Details on Unpub- VLT/KMOS lished Observations Reported N159W–S was observed with VLT/KMOS in this Paper (Very Large Telescope/K−band Multi-Object Spectrograph) under program 0101.C-0856(A) The ALMA and VLT/KMOS data presented in using the H + K grating with a spectral re- Section for N 159W–S have not been published solving power of 2000. The observations were in literature yet. Below we summarize technical carried out using a standard nod-to-sky pro- details of these observations. cedure with an integration time of 150 s, four DITs and three dither positions, yielding a total on-source integration time of 1800 s. Telluric ALMA absorption correction, response curve correc- N159W–S was observed with ALMA (Atacama tion, and absolute flux calibration was car- Large Millimeter/submillimeter Array; ALMA ried out using observations of telluric standard Partnership et al. 2015) during its Cycle stars using three IFUs. The data were reduced 4 in October 2016 as part of project num- with the standard VLT/KMOS pipeline using ber 2016.1.00308.S. We used the main array the esoreflex data reduction package. The (i.e.,12m antennas) consisting of 40 antennas. K−band continuum image is produced by inte- The phase center was set to RA (J2000) = grating over a third-order polynomial fit to the 05h39m41s, Dec (J2000) = 69◦4601100. The data for every spatial pixel (spaxel) over the ALMA correlator was configured to cover spe- spectral range 2.028–2.290 µm. The Brγ and cific frequency ranges within the Band 6 of H2 line emission images are produced by fitting ALMA. Four broad spectral windows (with a a Gaussian profile to the emission lines at every bandwidth of 1875 MHz each) centered at fre- position in the image. quencies 242.4, 244.8, 257.85 and 259.7 GHz Acknowledgement We thank Dr. Kazuki were tuned at the frequencies of specific molec- Tokuda for providing the ALMA 1.3 mm con- ular transitions of dense gas, shock/hot core tinuum image of N159W–S and Dr. Julia tracers (e.g., CS, SO2, CH3OH and SiO). The Roman-Duval for providing the CH3OH ∼96.7 ALMA data were calibrated using the ALMA GHz image of N 113. We also thank Dr. Ger- calibration pipeline available in CASA version ard Testor for providing the VLT/NACO near- 5.5.1. Flux calibration was obtained through infrared images that were used to improve as- observations of the bright quasar J0519–4546. trometry of the VLT/KMOS images. The work The gains were calibrated by interleaved obser- of M. S. was supported by NASA under award vations of the quasar J0601–7036. The band- number 80GSFC17M0002. The work of S. B. C. pass response was obtained by observing the was supported by the NASA Astrobiology In- bright quasar J0635–7516. After the calibration stitute through the Goddard Center for As- was applied, the line-free channels of the broad trobiology. P. S. acknowledges support from spectral windows were identified and used to the University of Cologne, Collaborative Re- create the continuum images. The CLEANing search Centre 956, sub-project C3, funded by process was done with the task TCLEAN, and the Deutsche Forschungsgemeinschaft (DFG) setting the robust parameter of Briggs equal – project ID 18401886, and Verbundforschung to 0.5 as compromise between resolution and Astrophysics, Project 05A17PK1, funded by sensitivity. The resulting images were restored BMBF (German Ministry of Science and Ed- with a synthesized beam of 0.0022×0.0012. Sensi- ucation). V. T. acknowledges the financial tivity of 1.4 mJy beam−1 was achieved in the support from the European Union’s Horizon data cube covering the CH3OH lines. 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No. 664931. E. S. W. received fund- ing from the Swedish National Space Agency,

31 grants dnr 98/14 and dnr 246/16. The work of Clouds. III - Interpretation of results. J. W. was supported by Sonderforschungsbere- ApJ 1992, 384, 508–522. ich SFB 881 "The Milky Way System" (sub- project B2) of the German Research Founda- (7) Westerlund, B. E. The Magellanic tion (DFG). The work of S. Z. and T. O. were Clouds; by Bengt E. Westerlund, pp. supported by NAOJ ALMA Scientiﬁc Research 292. ISBN 0521480701. Cambridge, UK: Grant Number 2016-03B. The work of T. O. Cambridge University Press; 1997. was also supported by JSPS KAKENHI (Grant (8) Dufour, R. J. The composition of H II re- Nos. 22244014, 26247026, and 18H05440). The gions in the Magellanic Clouds. Structure work of A. K. was supported by JSPS KAK- and Evolution of the Magellanic Clouds, ENHI (Grant No. 23403001). The authors ac- Symposium - International Astronomical knowledge ﬁnancial support from their home in- Union, 108. 1984; pp 353–360. stitutions. (9) Koornneef, J. Gas-to-dust ratios in the Magellanic Clouds. Structure and Evo- References lution of the Magellanic Clouds, Sympo- sium - International Astronomical Union, (1) Ehrenfreund, P.; Charnley, S. B. Organic 108. 1984; pp 333–339. Molecules in the Interstellar Medium, Comets, and Meteorites: A Voyage from (10) Roman-Duval, J. et al. Dust and Gas in Dark Clouds to the Early Earth. ARA&A the Magellanic Clouds from the HER- 2000, 38, 427–483. ITAGE Herschel Key Project. II. Gas-to- dust Ratio Variations across Interstellar (2) Meixner, M. et al. The HERSCHEL In- Medium Phases. ApJ 2014, 797, 86. ventory of The Agents of Galaxy Evolu- tion in the Magellanic Clouds, a Herschel (11) van Loon, J. T.; Oliveira, J. M.; Gor- Open Time Key Program. AJ 2013, 146, don, K. D.; Sloan, G. C.; Engel- 62. bracht, C. W. A Spitzer Space Telescope Far-infrared Spectral Atlas of Compact (3) Pietrzyński, G. et al. An eclipsing-binary Sources in the Magellanic Clouds. II. The distance to the Large Magellanic Cloud Small Magellanic Cloud. AJ 2010, 139, accurate to two per cent. Nature 2013, 1553–1565. 495, 76–79. (12) Oliveira, J. M.; van Loon, J. T.; (4) Graczyk, D.; Pietrzyński, G.; Thomp- Sloan, G. C.; Indebetouw, R.; Kem- son, I. B.; Gieren, W.; Pilecki, B.; per, F.; Tielens, A. G. G. M.; Si- Konorski, P.; Udalski, A.; Soszyński, I.; mon, J. D.; Woods, P. M.; Meixner, M. Villanova, S.; Górski, M.; Suchom- Ice chemistry in massive young stellar ska, K.; Karczmarek, P.; Kudritzki, R.- objects: the role of metallicity. MNRAS P.; Bresolin, F.; Gallenne, A. The Arau- 2011, 411, L36–L40. caria Project. The Distance to the Small Magellanic Cloud from Late-type Eclips- (13) Browning, M. K.; Tumlinson, J.; ing Binaries. ApJ 2014, 780, 59. Shull, J. M. Inferring Physical Condi- tions in Interstellar Clouds of H . ApJ (5) Asplund, M.; Grevesse, N.; Sauval, A. J.; 2 2003, 582, 810–822. Scott, P. The Chemical Composition of the Sun. ARA&A 2009, 47, 481–522. (14) Abdo, A. A. et al. Observations of the Large Magellanic Cloud with Fermi. (6) Russell, S. C.; Dopita, M. A. Abundances A&A 2010, 512, A7. of the heavy elements in the Magellanic

32 (15) Knödlseder, J. GeV Gamma-Ray Emis- (23) Rubio, M.; Garay, G.; Montani, J.; sion from Normal and Starburst Galax- Thaddeus, P. A (C-12)O survey of the ies. Cosmic Rays in Star-Forming Envi- Small Magellanic Cloud. ApJ 1991, 368, ronments, Astrophysics and Space Sci- 173–177. ence Proceedings, 13. 2013; p 169. (24) Fukui, Y. et al. First Results of a CO (16) Pei, Y. C.; Fall, S. M.; Hauser, M. G. Survey of the Large Magellanic Cloud Cosmic Histories of Stars, Gas, Heavy El- with NANTEN; Giant Molecular Clouds ements, and Dust in Galaxies. ApJ 1999, as Formation Sites of Populous Clusters. 522, 604–626. PASJ 1999, 51, 745–749.

(17) Heikkilä, A.; Johansson, L. E. B.; Olofs- (25) Mizuno, N.; Rubio, M.; Mizuno, A.; Ya- son, H. Molecular abundance variations maguchi, R.; Onishi, T.; Fukui, Y. First in the Magellanic Clouds. A&A 1999, Results of a CO Survey of the Small 344, 817–847. Magellanic Cloud with NANTEN. PASJ 2001, 53, L45–L49. (18) Wang, M.; Chin, Y.-N.; Henkel, C.; Whiteoak, J. B.; Cunningham, M. Abun- (26) Fukui, Y.; Kawamura, A.; Minami- dances and Isotope Ratios in the Magel- dani, T.; Mizuno, Y.; Kanai, Y.; lanic Clouds: The Star-Forming Environ- Mizuno, N.; Onishi, T.; Yonekura, Y.; ment of N 113. ApJ 2009, 690, 580–597. Mizuno, A.; Ogawa, H.; Rubio, M. The Second Survey of the Molecular Clouds (19) Sewiło, M.; Indebetouw, R.; Charn- in the Large Magellanic Cloud by NAN- ley, S. B.; Zahorecz, S.; Oliveira, J. M.; TEN. I. Catalog of Molecular Clouds. van Loon, J. T.; Ward, J. L.; Chen, C.- ApJS 2008, 178, 56–70. H. R.; Wiseman, J.; Fukui, Y.; Kawa- mura, A.; Meixner, M.; Onishi, T.; (27) Wong, T. et al. The Magellanic Mopra Schilke, P. The Detection of Hot Cores Assessment (MAGMA). I. The Molecu- and Complex Organic Molecules in the lar Cloud Population of the Large Mag- Large Magellanic Cloud. ApJ Letters ellanic Cloud. ApJS 2011, 197, 16. 2018, 853, L19. (28) Israel, F. P.; Johansson, L. E. B.; (20) Shimonishi, T.; Watanabe, Y.; Lequeux, J.; Booth, R. S.; Nyman, L. A.; Nishimura, Y.; Aikawa, Y.; Ya- Crane, P.; Rubio, M.; de Graauw, T.; mamoto, S.; Onaka, T.; Sakai, N.; Kutner, M. L.; Gredel, R.; Boulanger, F.; Kawamura, A. A Multiline Study of a Garay, G.; Westerlund, B. Results of the High-mass Young Stellar Object in the ESO / SEST Key Programme on Co/ in Small Magellanic Cloud with ALMA: the Magellanic Clouds - Part One - a Sur- The Detection of Methanol Gas at 0.2 vey of Co/ in the Large Magellanic Cloud Solar Metallicity. ApJ 2018, 862, 102. and the Small Magellanic Cloud. A&A 1993, 276, 25. (21) McGuire, B. A. 2018 Census of In- terstellar, Circumstellar, Extragalactic, (29) Israel, F. P.; Johansson, L. E. B.; Ru- Protoplanetary Disk, and Exoplanetary bio, M.; Garay, G.; de Graauw, T.; Molecules. ApJS 2018, 239, 17. Booth, R. S.; Boulanger, F.; Kut- ner, M. L.; Lequeux, J.; Nyman, L.- (22) Cohen, R. S.; Dame, T. M.; Garay, G.; A. Results of the ESO-SEST Key Pro- Montani, J.; Rubio, M.; Thaddeus, P. A gramme on CO in the Magellanic Clouds. complete CO survey of the Large Mag- X. CO emission from star formation re- ellanic Cloud. ApJ Letters 1988, 331, gions in LMC and SMC. A&A 2003, 406, L95–L99. 817–828.

33 (30) Heikkila, A.; Johansson, L. E. B.; Olof- and maser abundance estimates. MN- sson, H. The Cˆ18O/Cˆ17O ratio in RAS 2008, 385, 948–956. the Large Magellanic Cloud. A&A 1998, 332, 493–502. (39) Sinclair, M. W.; Carrad, G. J.; Caswell, J. L.; Norris, R. P.; (31) Johansson, L. E. B.; Olofsson, H.; Hjal- Whiteoak, J. B. A methanol maser marson, A.; Gredel, R.; Black, J. H. In- in the Large Magellanic Cloud. MNRAS terstellar molecules in the Large Magel- 1992, 256, 33P. lanic Cloud. A&A 1994, 291, 89–105. (40) Ellingsen, S. P.; Whiteoak, J. B.; Nor- (32) Chin, Y.-N.; Henkel, C.; Whiteoak, J. B.; ris, R. P.; Caswell, J. L.; Vaile, R. A. A Millar, T. J.; Hunt, M. R.; Lemme, C. Search for Methanol Masers in the Mag- Molecular abundances in the Magellanic ellanic Clouds. MNRAS 1994, 269, 1019. Clouds. I. A multiline study of ﬁve cloud cores. A&A 1997, 317, 548–562. (41) Beasley, A. J.; Ellingsen, S. P.; Claussen, M. J.; Wilcots, E. A Methanol (33) Chin, Y.-N.; Henkel, C.; Millar, T. J.; Maser Survey of IRAS-selected Regions Whiteoak, J. B.; Marx-Zimmer, M. in the Magellanic Clouds. ApJ 1996, Molecular abundances in the Magellanic 459, 600. Clouds. III. LIRS36, a star-forming region in the Small Magellanic Cloud. A&A (42) Ellingsen, S. P.; Breen, S. L.; 1998, 330, 901–909. Caswell, J. L.; Quinn, L. J.; Fuller, G. A. Masers associated with high-mass star (34) Wong, T.; Whiteoak, J. B.; Ott, J.; formation regions in the Large Mag- Chin, Y.-n.; Cunningham, M. R. Synthe- ellanic Cloud. MNRAS 2010, 404, sis Imaging of Dense Molecular Gas in 779–791. the N113 H II Region of the Large Mag- ellanic Cloud. ApJ 2006, 649, 224–234. (43) Werner, M. W. et al. The Spitzer Space Telescope Mission. ApJS 2004, 154, 1–9. (35) Seale, J. P.; Looney, L. W.; Wong, T.; Ott, J.; Klein, U.; Pineda, J. L. The Life (44) Pilbratt, G. L.; Riedinger, J. R.; Passvo- and Death of Dense Molecular Clumps in gel, T.; Crone, G.; Doyle, D.; Gageur, U.; the Large Magellanic Cloud. ApJ 2012, Heras, A. M.; Jewell, C.; Metcalfe, L.; 751, 42. Ott, S.; Schmidt, M. Herschel Space Observatory. An ESA facility for far- (36) Ott, J.; Henkel, C.; Staveley-Smith, L.; infrared and submillimetre astronomy. Weiß, A. First Detection of Ammonia in A&A 2010, 518, L1. the Large Magellanic Cloud: The Kinetic Temperature of Dense Molecular Cores (45) Meixner, M. et al. Spitzer Survey of in N 159 W. ApJ 2010, 710, 105–111. the Large Magellanic Cloud: Survey- ing the Agents of a Galaxy’s Evolution (37) Nishimura, Y.; Shimonishi, T.; Watan- (SAGE). I. Overview and Initial Results. abe, Y.; Sakai, N.; Aikawa, Y.; Kawa- AJ 2006, 132, 2268–2288. mura, A.; Yamamoto, S. Spectral Line Survey toward Molecular Clouds in the (46) Gordon, K. D. et al. Surveying the Large Magellanic Cloud. ApJ 2016, 818, Agents of Galaxy Evolution in the 161. Tidally Stripped, Low Metallicity Small Magellanic Cloud (SAGE-SMC). I. (38) Green, J. A. et al. Multibeam maser Overview. AJ 2011, 142, 102. survey of methanol and excited OH in the Magellanic Clouds: new detections (47) Whitney, B. A. et al. Spitzer Sage Sur- vey of the Large Magellanic Cloud. III.

34 Star Formation and ˜1000 New Candi- (56) Kato, D.; Ita, Y.; Onaka, T.; Tanabé, T.; date Young Stellar Objects. AJ 2008, Shimonishi, T.; Sakon, I.; Kaneda, H.; 136, 18–43. Kawamura, A.; Wada, T.; Usui, F.; Koo, B.-C.; Matsuura, M.; Takahashi, H. (48) Gruendl, R. A.; Chu, Y. High- and AKARI Infrared Camera Survey of the Intermediate-Mass Young Stellar Ob- Large Magellanic Cloud. I. Point-source jects in the Large Magellanic Cloud. Catalog. AJ 2012, 144, 179. ApJS 2009, 184, 172–197. (57) Shimonishi, T.; Onaka, T.; Kato, D.; (49) Chen, C.-H. R.; Indebetouw, R.; Chu, Y.- Sakon, I.; Ita, Y.; Kawamura, A.; H.; Gruendl, R. A.; Testor, G.; Kaneda, H. AKARI Infrared Camera Heitsch, F.; Seale, J. P.; Meixner, M.; Survey of the Large Magellanic Cloud. Sewilo, M. Spitzer View of Young Mas- II. The Near-infrared Spectroscopic Cat- sive Stars in the Large Magellanic Cloud alog. AJ 2013, 145, 32. H II Complexes. II. N 159. ApJ 2010, 721, 1206–1232. (58) Oliveira, J. M.; van Loon, J. T.; Chen, C.-H. R.; Tielens, A. G. G. M.; (50) Romita, K. A.; Carlson, L. R.; Sloan, G. C.; Woods, P. M.; Kem- Meixner, M.; Sewiło, M.; Whitney, B.; per, F.; Indebetouw, R.; Gordon, K. D.; Babler, B.; Indebetouw, R.; Hora, J. L.; Boyer, M. L.; Shiao, B.; Madden, S.; Meade, M.; Shiao, B. Young Stellar Speck, A. K.; Meixner, M.; Marengo, M. Objects in the Large Magellanic Cloud Ice Chemistry in Embedded Young Stel- Star-forming Region N206. ApJ 2010, lar Objects in the Large Magellanic 721, 357–368. Cloud. ApJ 2009, 707, 1269–1295.

(51) Carlson, L. R.; Sewiło, M.; Meixner, M.; (59) Oliveira, J. M.; van Loon, J. T.; Romita, K. A.; Lawton, B. Identifying Sloan, G. C.; Sewiło, M.; Krae- young stellar objects in nine Large Mag- mer, K. E.; Wood, P. R.; Indebe- ellanic Cloud star-forming regions. A&A touw, R.; Filipović, M. D.; Craw- 2012, 542, A66. ford, E. J.; Wong, G. F.; Hora, J. L.; (52) Sewiło, M. et al. Surveying the Agents Meixner, M.; Robitaille, T. P.; Shiao, B.; of Galaxy Evolution in the Tidally Simon, J. D. Early-stage young stellar Stripped, Low Metallicity Small Mag- objects in the Small Magellanic Cloud. ellanic Cloud (SAGE-SMC). III. Young MNRAS 2013, 428, 3001–3033. Stellar Objects. ApJ 2013, 778, 15. (60) Seale, J. P.; Looney, L. W.; Chen, C.- (53) Sewiło, M. et al. The youngest mas- H. R.; Chu, Y.-H.; Gruendl, R. A. The sive protostars in the Large Magellanic Evolution of Massive Young Stellar Ob- Cloud. A&A 2010, 518, L73. jects in the Large Magellanic Cloud. II. Thermal Processing of Circumstellar (54) Seale, J. P. et al. Herschel Key Program Ices. ApJ 2011, 727, 36. Heritage: a Far-Infrared Source Catalog for the Magellanic Clouds. AJ 2014, 148, (61) Shimonishi, T.; Onaka, T.; Kato, D.; 124. Sakon, I.; Ita, Y.; Kawamura, A.; Kaneda, H. AKARI Near-Infrared Spec- (55) Ita, Y. et al. AKARI IRC Survey of the troscopy: Detection of H2O and CO2 Large Magellanic Cloud: Outline of the Ices toward Young Stellar Objects in Survey and Initial Results. PASJ 2008, the Large Magellanic Cloud. ApJ Letters 60, S435–S451. 2008, 686, L99.

35 (62) Shimonishi, T.; Onaka, T.; Kato, D.; (71) Tielens, A. G. G. M. Interstellar poly- Sakon, I.; Ita, Y.; Kawamura, A.; cyclic aromatic hydrocarbon molecules. Kaneda, H. Spectroscopic observations of ARA&A 2008, 46, 289–337. ices around embedded young stellar objects in the Large Magellanic Cloud with (72) Malek, S. E.; Cami, J.; Bernard-Salas, J. AKARI. A&A 2010, 514, A12. The Rich Circumstellar Chemistry of SMP LMC 11. ApJ 2012, 744, 16. (63) van Dishoeck, E. F. Astrochemistry of dust, ice and gas: introduction and (73) Cernicharo, J.; Heras, A. M.; Tie- overview. Faraday Discussions 2014, lens, A. G. G. M.; Pardo, J. R.; Her- 168, 9. pin, F.; Guélin, M.; Waters, L. B. F. M. Infrared Space Observatory’s Discovery (64) Whittet, D. C. B.; Bode, M. F.; of C4H2,C6H2, and Benzene in CRL 618. Longmore, A. J.; Adamson, A. J.; ApJ Letters 2001, 546, L123–L126. McFadzean, A. D.; Aitken, D. K.; (74) Jones, O. C. et al. The SAGE-Spec Roche, P. F. Infrared spectroscopy of Spitzer Legacy program: the life-cycle dust in the Taurus dark clouds - Ice and of dust and gas in the Large Magellanic silicates. MNRAS 1988, 233, 321–336. Cloud. Point source classification - III. (65) Pontoppidan, K. M.; Salyk, C.; MNRAS 2017, 470, 3250–3282. Bergin, E. A.; Brittain, S.; Marty, B.; (75) Tokuda, K. et al. An ALMA view of Mousis, O.; Öberg, K. I. Volatiles in molecular filaments in the Large Magel- Protoplanetary Disks. Protostars and lanic Cloud II: An early stage of high- Planets VI 2014, 363–385. mass star formation embedded at col- (66) Kato, D. et al. The IRSF Magellanic liding clouds in N159W-South. arXiv e- Clouds Point Source Catalog. PASJ prints 2018, arXiv:1811.04400 . 2007, 59, 615–641. (76) Seale, J. P.; Looney, L. W.; Chu, Y.-H.; (67) Shimonishi, T.; Onaka, T.; Kawa- Gruendl, R. A.; Brandl, B.; Chen, C.- mura, A.; Aikawa, Y. The Detection of a H. R.; Brandner, W.; Blake, G. A. The Hot Molecular Core in the Large Magel- Evolution Of Massive Young Stellar Ob- lanic Cloud with ALMA. ApJ 2016, 827, jects in the Large Magellanic Cloud. I. 72. Identification and Spectral Classification. ApJ 2009, 699, 150–167. (68) Smith, R. C.; MCELS Team, The UM/CTIO Magellanic Cloud emission- (77) Fukui, Y. et al. High-mass Star Forma- line survey. PASA 1998, 15, 163–64. tion Triggered by Collision between CO Filaments in N159 West in the Large (69) Shimonishi, T.; Dartois, E.; Onaka, T.; Magellanic Cloud. ApJ Letters 2015, Boulanger, F. VLT/ISAAC infrared 807, L4. spectroscopy of embedded high-mass YSOs in the Large Magellanic Cloud: (78) Ward, J. L.; Oliveira, J. M.; van Methanol and the 3.47 µm band. A&A Loon, J. T.; Sewiło, M. K- band inte- 2016, 585, A107. gral field spectroscopy and optical spectroscopy of massive young stellar objects (70) Bernard-Salas, J.; Peeters, E.; in the Small Magellanic Cloud. MNRAS Sloan, G. C.; Cami, J.; Guiles, S.; 2017, 464, 1512–1552. Houck, J. R. The Spitzer IRS spectrum of SMP LMC 11. ApJ Letters 2006, (79) Testor, G.; Lemaire, J. L.; Field, D.; Di- 652, L29–L32. ana, S. VLT/NACO near-infrared imaging and spectroscopy of N159A in the

36 LMC HII complex N159. A&A 2006, (88) Mookerjea, B.; Casper, E.; Mundy, L. G.; 453, 517–524. Looney, L. W. Kinematics and Chem- istry of the Hot Molecular Core in (80) van Loon, J. T.; Cohen, M.; G34.26+0.15 at High Resolution. ApJ Oliveira, J. M.; Matsuura, M.; Mc- 2007, 659, 447–458. Donald, I.; Sloan, G. C.; Wood, P. R.; Zijlstra, A. A. Molecules and dust (89) Bisschop, S. E.; Jørgensen, J. K.; van production in the Magellanic Clouds. Dishoeck, E. F.; de Wachter, E. B. M. A&A 2008, 487, 1055–1073. Testing grain-surface chemistry in massive hot-core regions. A&A 2007, 465, (81) Shull, J. M.; Hollenbach, D. J. H2 cool- 913–929. ing, dissociation, and infrared emission in shocked molecular clouds. ApJ 1978, (90) Kurtz, S.; Cesaroni, R.; Churchwell, E.; 220, 525–537. Hofner, P.; Walmsley, C. M. Hot Molec- ular Cores and the Earliest Phases of (82) Tielens, A. G. G. M. The Physics and High-Mass Star Formation. Protostars Chemistry of the Interstellar Medium, and Planets IV 2000, 299–326. by A. G. G. M. Tielens, pp. . ISBN 0521826349. Cambridge, UK: Cambridge (91) Herbst, E.; van Dishoeck, E. F. Complex University Press, 2005.; 2005. Organic Interstellar Molecules. ARA&A 2009, 47, 427–480. (83) Sakai, T.; Sakai, N.; Hirota, T.; Ya- mamoto, S. A Survey of Molecular Lines (92) Widicus Weaver, S. L.; Laas, J. C.; Toward Massive Clumps in Early Evolu- Zou, L.; Kroll, J. A.; Rad, M. L.; tionary Stages of High-mass Star Forma- Hays, B. M.; Sanders, J. L.; tion. ApJ 2010, 714, 1658–1671. Lis, D. C.; Cross, T. N.; Wehres, N.; McGuire, B. A.; Sumner, M. C. Deep, (84) Oliveira, J. M.; van Loon, J. T.; Sta- Broadband Spectral Line Surveys of nimirović, S.; Zijlstra, A. A. Massive Molecule-rich Interstellar Clouds. ApJS young stellar objects in the Large Magel- 2017, 232, 3. lanic Cloud: water masers and ESO-VLT 3-4 µm spectroscopy. MNRAS 2006, (93) Fontani, F.; Palau, A.; Caselli, P.; 372, 1509–1524. Sánchez-Monge, Á.; Butler, M. J.; Tan, J. C.; Jiménez-Serra, I.; Bus- (85) Whiteoak, J. B.; Gardner, F. F. Observa- quet, G.; Leurini, S.; Audard, M. tions of H2O masers in nearby galaxies. Deuteration as an evolutionary tracer in MNRAS 1986, 222, 513–523. massive-star formation. A&A 2011, 529, L7. (86) Lazendic, J. S.; Whiteoak, J. B.; Klamer, I.; Harbison, P. D.; (94) Caselli, P.; Ceccarelli, C. Our astrochem- Kuiper, T. B. H. Accurate positions ical heritage. A&AR 2012, 20, 56. of H2 O masers in the Large Magellanic Cloud. MNRAS 2002, 331, 969–974. (95) Albertsson, T.; Semenov, D. A.; Vasyunin, A. I.; Henning, T.; Herbst, E. (87) Ward, J. L.; Oliveira, J. M.; van New Extended Deuterium Fractionation Loon, J. T.; Sewiło, M. Integral ﬁeld Model: Assessment at Dense ISM Con- spectroscopy of massive young stellar ob- ditions and Sensitivity Analysis. ApJS jects in the N113 H II region in the Large 2013, 207, 27. Magellanic Cloud. MNRAS 2016, 455, 2345–2362. (96) Gerner, T.; Shirley, Y. L.; Beuther, H.; Semenov, D.; Linz, H.; Albertsson, T.; Henning, T. Chemical evolution in the

37 early phases of massive star formation. stellar Core Conditions. ApJ 2018, 853, II. Deuteration. A&A 2015, 579, A80. 102.

(97) Ohishi, M.; Irvine, W. M.; Kaifu, N. (105) Brown, P. D.; Charnley, S. B.; Mil- Molecular Abundance Variations among lar, T. J. A model of the chemistry in and Within Cold, Dark Molecular hot molecular cores. MNRAS 1988, 231, Clouds(rp). Astrochemistry of Cosmic 409–417. Phenomena. 1992; p 171. (106) Bergin, E. A.; Tafalla, M. Cold Dark (98) Bacmann, A.; Taquet, V.; Faure, A.; Clouds: The Initial Conditions for Star Kahane, C.; Ceccarelli, C. Detection of Formation. ARA&A 2007, 45, 339–396. complex organic molecules in a prestellar core: a new challenge for astrochemical (107) Garrod, R. T.; Herbst, E. Formation of models. A&A 2012, 541, L12. methyl formate and other organic species in the warm-up phase of hot molecular (99) Cernicharo, J.; Marcelino, N.; Roueff, E.; cores. A&A 2006, 457, 927–936. Gerin, M.; Jiménez-Escobar, A.; Muñoz Caro, G. M. Discovery of the Methoxy (108) Charnley, S. B.; Tielens, A. G. G. M.; Radical, CH3O, toward B1: Dust Grain Millar, T. J. On the molecular complex- and Gas-phase Chemistry in Cold Dark ity of the hot cores in Orion A - Grain Clouds. ApJ Letters 2012, 759, L43. surface chemistry as ’The last refuge of the scoundrel’. ApJ Letters 1992, 399, (100) Vastel, C.; Ceccarelli, C.; Lefloch, B.; L71–L74. Bachiller, R. The Origin of Complex Or- ganic Molecules in Prestellar Cores. ApJ (109) Rodgers, S. D.; Charnley, S. B. Chemi- Letters 2014, 795, L2. cal Evolution in Protostellar Envelopes: Cocoon Chemistry. ApJ 2003, 585, 355– (101) Jiménez-Serra, I.; Vasyunin, A. I.; 371. Caselli, P.; Marcelino, N.; Billot, N.; Viti, S.; Testi, L.; Vastel, C.; Lefloch, B.; (110) Aikawa, Y.; Wakelam, V.; Garrod, R. T.; Bachiller, R. The Spatial Distribution of Herbst, E. Molecular Evolution and Star Complex Organic Molecules in the L1544 Formation: From Prestellar Cores to Pre-stellar Core. ApJ Letters 2016, 830, Protostellar Cores. ApJ 2008, 674, 984– L6. 996. (102) Taquet, V.; Wirström, E. S.; Charn- (111) Gratier, P.; Majumdar, L.; Ohishi, M.; ley, S. B.; Faure, A.; López-Sepulcre, A.; Roueff, E.; Loison, J. C.; Hickson, K. M.; Persson, C. M. Chemical complexity in- Wakelam, V. A New Reference Chemi- duced by efficient ice evaporation in the cal Composition for TMC-1. ApJS 2016, Barnard 5 molecular cloud. A&A 2017, 225, 25. 607, A20. (112) Maffucci, D. M.; Wenger, T. V.; Le (103) Minissale, M.; Moudens, A.; Baouche, S.; Gal, R.; Herbst, E. Astrochemical Ki- Chaabouni, H.; Dulieu, F. Hydrogena- netic Grid Models of Groups of Observed tion of CO-bearing species on grains: Molecular Abundances: Taurus Molecu- unexpected chemical desorption of CO. lar Cloud 1 (TMC-1). ApJ 2018, 868, MNRAS 2016, 458, 2953–2961. 41. (104) Chuang, K. J.; Fedoseev, G.; Qasim, D.; (113) Sakai, N.; Yamamoto, S. Warm Carbon- Ioppolo, S.; van Dishoeck, E. F.; Lin- Chain Chemistry. Chem. Rev. 2013, 113, nartz, H. Reactive Desorption of CO Hy- 8981–9015. drogenation Products under Cold Pre-

38 (114) Tielens, A. G. G. M.; Hagen, W. Model (123) Linnartz, H.; Ioppolo, S.; Fedoseev, G. calculations of the molecular composition Atom addition reactions in interstellar of interstellar grain mantles. A&A 1982, ice analogues. International Reviews in 114, 245–260. Physical Chemistry 2015, 34, 205–237.

(115) Charnley, S. B.; Rodgers, S. D. Interstel- (124) Caselli, P.; Hasegawa, T. I.; Herbst, E. lar Reservoirs of Cometary Matter. Space Chemical differentiation between star- Science Reviews 2008, 138, 59–73. forming regions - The Orion Hot Core and Compact Ridge. ApJ 1993, 408, (116) Charnley, S. Interstellar Organic Chem- 548–558. istry. The Bridge Between the Big Bang and Biology: Stars, Planetary Systems, (125) Rodgers, S. D.; Charnley, S. B. Chemi- Atmospheres, Volcanoes: Their Link to cal Differentiation in Regions of Massive Life, International workshop, Stromboli, Star Formation. ApJ 2001, 546, 324– Italy, September 13-17, 1999 Rome: Con- 329. siglio Nazionale delle Ricerche, Edited by (126) Blake, G. A.; Sutton, E. C.; Mas- Franco Giovanelli. 2001; p 139. son, C. R.; Phillips, T. G. Molecu- (117) Hollis, J. M.; Lovas, F. J.; Jewell, P. R. lar abundances in OMC-1 - The chem- Interstellar Glycolaldehyde: The First ical composition of interstellar molecular Sugar. ApJ Letters 2000, 540, L107– clouds and the influence of massive star L110. formation. ApJ 1987, 315, 621–645.

(118) Hollis, J. M.; Jewell, P. R.; Lovas, F. J.; (127) Charnley, S. B.; Kress, M. E.; Tie- Remijan, A. Green Bank Telescope Ob- lens, A. G. G. M.; Millar, T. J. Inter- servations of Interstellar Glycolaldehyde: stellar Alcohols. ApJ 1995, 448, 232. Low-Temperature Sugar. ApJ Letters (128) Karpas, Z.; Mautner, M. Alkyl-transfer 2004, 613, L45–L48. reactions between protonated alcohols (119) Hollis, J. M.; Lovas, F. J.; Jewell, P. R.; and ethers: gas-phase alkylation of Coudert, L. H. Interstellar Antifreeze: formaldehyde. J. Phys. Chem. 1989, 93, Ethylene Glycol. ApJ Letters 2002, 571, 1859–63. L59–L62. (129) Cole, C. A.; Wehres, N.; Yang, Z.; Thom- (120) Hollis, J. M.; Jewell, P. R.; Lovas, F. J.; sen, D. L.; Snow, T. P.; Bierbaum, V. M. Remijan, A.; Møllendal, H. Green Bank A Gas-phase Formation Route to Inter- Telescope Detection of New Interstellar stellar Trans-methyl Formate. ApJ Let- Aldehydes: Propenal and Propanal. ApJ ters 2012, 754, L5. Letters 2004, 610, L21–L24. (130) Neill, J. L.; Steber, A. L.; Muckle, M. T.; (121) Watanabe, N.; Kouchi, A. Ice surface re- Zaleski, D. P.; Lattanzi, V.; Spezzano, S.; actions: A key to chemical evolution in McCarthy, M. C.; Remijan, A. J.; space. Progress in Surface Science 2008, Friedel, D. N.; Widicus Weaver, S. L.; 83, 439–489. Pate, B. H. Spatial Distributions and In- terstellar Reaction Processes. J. Phys. (122) Ioppolo, S.; Cuppen, H. M.; Linnartz, H. Chem. 2011, 115, 6472–6480. Surface formation routes of interstellar molecules: hydrogenation reactions in (131) Neill, J. L.; Muckle, M. T.; Zaleski, D. P.; simple ices. Rend. Fis. Acc. Lincei 2011, Steber, A. L.; Pate, B. H.; Lattanzi, V.; 22, 211. Spezzano, S.; McCarthy, M. C.; Remi- jan, A. J. Laboratory and Tentative Interstellar Detection of Trans-Methyl

39 Formate Using the Publicly Available quantum chemistry and kinetics study. Green Bank Telescope Primos Survey. MNRAS 2019, 482, 3567–3575. ApJ 2012, 755, 153. (139) Cernicharo, J.; Kisiel, Z.; Tercero, B.; (132) Taquet, V.; Wirström, E. S.; Charn- Kolesniková, L.; Medvedev, I. R.; ley, S. B. Formation and Recondensa- López, A.; Fortman, S.; Win- tion of Complex Organic Molecules dur- newisser, M.; de Lucia, F. C.; ing Protostellar Luminosity Outbursts. Alonso, J. L.; Guillemin, J. C. A rigorous ApJ 2016, 821, 46. detection of interstellar CH3NCO: An important missing species in astro- (133) Laas, J. C.; Garrod, R. T.; Herbst, E.; chemical networks. A&A 2016, 587, Widicus Weaver, S. L. Contributions L4. from Grain Surface and Gas Phase Chemistry to the Formation of Methyl (140) Hollis, J. M.; Lovas, F. J.; Remijan, A. J.; Formate and Its Structural Isomers. ApJ Jewell, P. R.; Ilyushin, V. V.; Kleiner, I. 2011, 728, 71. Detection of Acetamide (CH3CONH2): The Largest Interstellar Molecule with a (134) Belloche, A.; Garrod, R. T.; Peptide Bond. ApJ Letters 2006, 643, Müller, H. S. P.; Menten, K. M.; L25–L28. Comito, C.; Schilke, P. Increased complexity in interstellar chemistry: (141) Shingledecker, C. N.; Tennis, J.; Le detection and chemical modeling of Gal, R.; Herbst, E. On Cosmic-Ray- ethyl formate and n-propyl cyanide in driven Grain Chemistry in Cold Core Sagittarius B2(N). A&A 2009, 499, Models. ApJ 2018, 861, 20. 215–232. (142) Benz, A. O. et al. Hydrides in young (135) Tercero, B.; Cernicharo, J.; López, A.; stellar objects: Radiation tracers in Brouillet, N.; Kolesniková, L.; a protostar-disk-outﬂow system. A&A Motiyenko, R. A.; Margulès, L.; 2010, 521, L35. Alonso, J. L.; Guillemin, J.-C. Searching for trans ethyl methyl ether in Orion (143) Prasad, S. S.; Tarafdar, S. P. UV radia- KL. A&A 2015, 582, L1. tion ﬁeld inside dense clouds - Its possible existence and chemical implications. ApJ (136) Hamberg, M.; Österdahl, F.; 1983, 267, 603–609. Thomas, R. D.; Zhaunerchyk, V.; Vigren, E.; Kaminska, M.; Af Ug- (144) Bernstein, M. P.; Sandford, S. A.; Al- glas, M.; Källberg, A.; Simonsson, A.; lamandola, L. J.; Chang, S.; Schar- Paál, A.; Larsson, M.; Geppert, W. D. berg, M. A. Organic Compounds Pro- Experimental studies of the dissociative duced by Photolysis of Realistic Inter- recombination processes for the dimethyl stellar and Cometary Ice Analogs Con- + + taining Methanol. ApJ 1995, 454, 327. ether ions CD3OCD2 and (CD3)2OD . A&A 2010, 514, A83. (145) Öberg, K. I.; Garrod, R. T.; van (137) Boogert, A. C. A.; Gerakines, P. A.; Dishoeck, E. F.; Linnartz, H. Formation Whittet, D. C. B. Observations of the icy rates of complex organics in UV irradi- universe. ARA&A 2015, 53, 541–581. ated CH_3OH-rich ices. I. Experiments. A&A 2009, 504, 891–913. (138) Skouteris, D.; Balucani, N.; Cecca- relli, C.; Faginas Lago, N.; Codella, C.; (146) Allen, M.; Robinson, G. W. The molec- Falcinelli, S.; Rosi, M. Interstellar ular composition of dense interstellar dimethyl ether gas-phase formation: a clouds. ApJ 1977, 212, 396–415.

40 (147) D’Hendecourt, L. B.; Allamandola, L. J.; (157) Collings, M. P.; Anderson, M. A.; Grim, R. J. A.; Greenberg, J. M. Time- Chen, R.; Dever, J. W.; Viti, S.; dependent chemistry in dense molecular Williams, D. A.; McCoustra, M. R. S. clouds. II - Ultraviolet photoprocessing A laboratory survey of the thermal and infrared spectroscopy of grain man- desorption of astrophysically relevant tles. A&A 1986, 158, 119–134. molecules. MNRAS 2004, 354, 1133– 1140. (148) Brown, P. D. The grain-surface formation of complex molecules. MNRAS (158) Pauly, T.; Garrod, R. T. Modeling CO, 1990, 243, 65–71. CO2, and H2O Ice Abundances in the En- velopes of Young Stellar Objects in the (149) Hollis, J. M.; Vogel, S. N.; Snyder, L. E.; Magellanic Clouds. ApJ 2018, 854, 13. Jewell, P. R.; Lovas, F. J. The Spatial Scale of Glycolaldehyde in the Galactic (159) Acharyya, K.; Herbst, E. Hot Cores in Center. ApJ Letters 2001, 554, L81–L85. Magellanic Clouds. ApJ 2018, 859, 51.

(150) Sorrell, W. H. Origin of Amino Acids and (160) Audard, M.; Ábrahám, P.; Dun- Organic Sugars in Interstellar Clouds. ham, M. M.; Green, J. D.; Grosso, N.; ApJ Letters 2001, 555, L129–L132. Hamaguchi, K.; Kastner, J. H.; Kóspál, Á.; Lodato, G.; Ro- (151) Viti, S.; Williams, D. A. Time-dependent manova, M. M.; Skinner, S. L.; evaporation of icy mantles in hot cores. Vorobyov, E. I.; Zhu, Z. Episodic MNRAS 1999, 305, 755–762. Accretion in Young Stars. Protostars (152) Garrod, R. T.; Widicus Weaver, S. L.; and Planets VI 2014, 387–410. Herbst, E. Complex Chemistry in Star- (161) Buckle, J. V.; Rodgers, S. D.; forming Regions: An Expanded Gas- Wirstrom, E. S.; Charnley, S. B.; Grain Warm-up Chemical Model. ApJ Markwick-Kemper, A. J.; Butner, H. M.; 2008, 682, 283–302. Takakuwa, S. Observations of chemical (153) Ligterink, N. F. W.; Coutens, A.; diﬀerentiation in clumpy molecular Kofman, V.; Müller, H. S. P.; Gar- clouds. Faraday Discussions 2006, 133, rod, R. T.; Calcutt, H.; Wampﬂer, S. F.; 63–82. Jørgensen, J. K.; Linnartz, H.; van (162) Wirström, E. S.; Charnley, S. B.; Dishoeck, E. F. The ALMA-PILS sur- Persson, C. M.; Buckle, J. V.; vey: detection of CH3NCO towards the Cordiner, M. A.; Takakuwa, S. Cold low-mass protostar IRAS 16293-2422 and Water Vapor in the Barnard 5 Molecular laboratory constraints on its formation. Cloud. ApJ Letters 2014, 788, L32. MNRAS 2017, 469, 2219–2229. (163) Mumma, M. J.; Charnley, S. B. The (154) Millar, T. J.; Herbst, E. Chemical mod- Chemical Composition of Comets - elling of dark clouds in the LMC and Emerging Taxonomies and Natal Her- SMC. MNRAS 1990, 242, 92–97. itage. ARA&A 2011, 49, 471–524.

(155) Acharyya, K.; Herbst, E. Molecular (164) Oliveira, J. M. The star formation pro- Development in the Large Magellanic cess in the Magellanic Clouds. The Mag- Cloud. ApJ 2015, 812, 142. ellanic System: Stars, Gas, and Galaxies, (156) Acharyya, K.; Herbst, E. Simulations of Symposium - International Astronomical the Chemistry in the Small Magellanic Union, 256, Edited by J. T. van Loon and Cloud. ApJ 2016, 822, 105. J. M. Oliveira. 2009; pp 191–202.

41 (165) Fedoseev, G.; Chuang, K.-J.; Ioppolo, S.; Qasim, D.; van Dishoeck, E. F.; Lin- nartz, H. Formation of Glycerol through Hydrogenation of CO Ice under Prestel- lar Core Conditions. ApJ 2017, 842, 52.