Chemical evolution in high-mass star-forming regions

Abstract

Growing evidence shows that most stars in the Milky Way, including our Sun, are born in high-mass star-forming regions, but due to both observational and theoretical challenges, our understanding of their chemical evolution is much less clear than that of their low-mass counterparts. Thanks to the capabilities of new generation telescopes and computers, a growing amount of observational and theoretical results have been recently obtained, which have important implications not only for our understanding of the (still mysterious) formation process of high-mass stars, but also for the chemistry that the primordial Solar System might have inherited from its birth environment. In this review, we summarise the main observational and theoretical results achieved in the last decades in the study of chemistry evolution in high-mass star-forming regions, and in the identification of chemical evolutionary indicators. Emphasis is especially given to observational studies, for which most of the work has been carried out so far. A comparison with the chemical evolution occurring in other astrophysical environments, in particular in low-mass star-forming cores and extragalactic cores, is also briefly presented. Current open questions and future perspectives are also discussed.

1 Introduction - the need for this review

Today, increasing evidence supports the idea that our Sun was born in a crowded stellar cluster that included stars more massive than (Adams, 2010). The traces of the interaction between these stars and the Solar System are recorded in meteoritic material, where anomalous high abundances of daughter species of Short-Lived Radionuclides (SLRs), in particular ²⁶Al, produced by nearby high-mass stars during the primordial evolution of the Solar System, have been measured (e.g., Portegies Zwart et al., 2018; Portegies Zwart, 2019). In fact, it is well-known that most stars are born in rich clusters (e.g., Carpenter, 2000; Lada and Lada, 2003) which likely included high-mass stars (e.g., Rivilla et al., 2014). Therefore, the study of the chemical content of high-mass star-forming regions can give us important information not only about the formation of high-mass stars, but also on the chemical heritage and complexity of both the Solar System and most stars in the Milky Way.

Despite this, for long our knowledge of how chemistry evolves in these regions has remained limited. This is due to both observational and theoretical difficulties. First, observations are challenging because high-mass star-forming cores, i.e., molecular compact structures with masses , which have the potential to form high-mass stars and clusters, are fewer than low-mass cores. Therefore, they are on average located further away from the Sun (at typical distances larger than 1 kpc), thus smaller in angular size and typically surrounded by large amounts of ambient gas and/or nearby star formation activity difficult to disentangle (Motte et al., 2018). Second, theory predicts that massive star-forming cores evolve on timescales of years, typically shorter than those of their low-mass counterparts (McKee and Tan, 2002). These short timescales imply that it is challenging to define a physical evolutionary sequence divided into well separated classes, and although tentative evolutionary classifications have been proposed (Beuther, 2007; Tan et al., 2014; Motte et al., 2018), their phases are debated and not clearly separated as in the low-mass case. Finding a chemical evolutionary sequence would be extremely useful to better understand a physical evolutionary sequence, still uncertain. However, this is not easy because the formation and destruction of molecules may require timescales comparable to or longer than a single phase (i.e., yrs), challenging the identification of reliable indicators of a specific evolutionary phase. Thanks to the improved observational and computational capabilities of new facilities, a huge and growing observational and theoretical effort has been devoted to this topic in recent years, and important results have been obtained in the last decades. This review article summarises the observational and theoretical endeavour devoted to understanding how chemistry evolves in massive star-forming regions and to identify robust chemical evolutionary indicators.

FIGURE 1

In our review, we will refer mostly to this coarse classification, bearing in mind that this scheme is not rigid, because these phases may partly overlap, due to the short evolutionary timescales. Moreover, inside each group there can be a variety of physical parameters (see above), linked to possible chemical diversity. In addition, the way in which high-mass cores can be observationally classified into the three groups can slightly change from one study to another.

2 Observations

In this Section, we will review the main observational works that have studied the chemical content of massive star-forming regions, to find a link between the early and late evolutionary stages. The approach has traditionally been twofold: line surveys towards selected objects, or selected molecular species or lines towards source surveys. We divide this section into two main subsections according to this twofold approach. To help the reader, in Table 1 we list the major acronyms used for instruments, research institutes, and observational surveys mentioned throughout this section.

2.1 Line surveys of selected sources

Line surveys of selected sources, and in particular unbiased spectral line surveys, have historically been the preferred approach to reveal new species in the interstellar medium (ISM). For example, urea detection in Sagittarius B2(N) (hereafter Sgr B2(N)) with the Re-exploring Molecular Complexity with ALMA (ReMoCA) survey (Belloche et al., 2019) or the detection of polycyclic aromatic hydrocarbons (PAHs) in the cold, low-mass starless core TMC-1 with the GBT Observations of TMC-1: Hunting Aromatic Molecules (GOTHAM) survey (McGuire et al., 2021) and the Yebes 40-m telescope Q-band Ultrasensitive Inspection Journey to the Obscure TMC-1 Environment (QUIJOTE) project (Cernicharo et al., 2021). Unbiased surveys used to be very telescope time consuming, especially because the observation of the whole receiver bandwidth required several spectral setups. This translated in many hours of telescope time, distributed in different days or even epochs (i.e., astrophysical times used as references for some varying astronomical quantities) that led to different observing conditions and sometimes spectral noise. This has changed in recent years. Nowadays a new generation of new broadband receivers with much larger instantaneous bandwidth, such as those of the Submillimeter Array (SMA) or of the single-dish telescopes Institute de radioastronomie millimétrique (IRAM) 30-m, Yebes 40-m, or Green Bank Telescope (GBT), allow this kind of observations to be carried out faster and in a more efficient way. This has led to an increase of the detection rate of new species in recent years. According to the Cologne Database for Molecular Spectroscopy (CDMS)¹, there are about 330 molecules detected in the ISM or circumstellar shells, and about 150 are complex organic molecules (COMs), carbon-bearing molecules with at least six atoms. The rate of new detections per year has steadily increased from about 4 detection/year since the end of the 1960s, beginning of the 1970s, when the first COMs were detected, to about 6 detection/year since 2005 (McGuire, 2018). This rate has increased considerably in recent years, thanks to the new broadband receivers, and currently with different projects observing TMC-1, the detection rate is about 20 new species per year.

In this section, we will review the results obtained from broad spectral surveys towards selected objects of known evolutionary stage. We will follow the three evolutionary classes described in Sect. 1: (i) high-mass starless core candidates (Secttion 2.1.1, e.g., Pillai et al., 2011; Wang et al., 2014); (ii) high-mass protostellar objects (Secttion 2.1.2), including well-known hot molecular cores (HMCs) like Orion KL (Tercero et al., 2010), Sgr B2 (Belloche et al., 2013), G31.41 + 0.31 (Beltrán et al., 2009), NGC 6334I (Walsh et al., 2010) or proposed hot-core precursors like G328.25510.5321 (Csengeri et al., 2019); and (iii) hyper- and ultra-compact (HC and UC) H_II regions (Section 2.1.3) like Mon R2 (Ginard et al., 2012) and W51 (Watanabe et al., 2017; Rivilla et al., 2016; Rivilla et al., 2017b). We will discuss the most relevant species detected in these objects, with special attention to those with prebiotic potential.

2.1.1 High-mass starless core candidates

The existence of HMSCs, typically embedded in infrared dark clouds (IRDCs), is a fundamental question in astrophysics, whose answer would allow us to discriminate and better constrain different theories of massive star-formation, such as core accretion (McKee and Tan, 2003) and competitive accretion (Bonnell et al., 2001). The reality is that these cores are elusive and despite years of research, many of the proposed candidates have shown evidence of active star-formation, such as outflow activity (e.g., Wang et al., 2011; Tan et al., 2016; Pillai et al., 2019). Therefore, as a result we are left only with few starless core candidates with very cold temperatures ( K) and associated with high deuterium fractionation, , and depletion (e.g., Pillai et al., 2007; Fontani et al., 2011; Tan et al., 2013; Wang et al., 2014; Wang et al., 2016; Pillai et al., 2011; Pillai et al., 2019). Due to the lack of a significant sample of candidates, there are not many line surveys, in particular unbiased ones, toward HMSC candidates, and most of them target specific species, in particular deuterated ones, like the SCUBA Massive Pre/Protocluster core Survey (SCAMPS) carried out with the James Clerk Maxwell Telescope (JCMT) (Thompson et al., 2005; Pillai et al., 2007) or the surveys of Fontani et al. (2011) and Sakai et al. (2012). One of the few studies that has carried out a broadband spectral survey toward massive starless core candidates embedded in IRDC G11.110.12 is that of Wang et al. (2014). The G11.110.12 cloud, also known as the “Snake” nebula, located at 3.6 kpc, is one of the first IRDCs identified by Egan et al. (1998) and one of the best studied. Wang et al. (2014) observed the two more massive clumps embedded in this cloud, named P1 and P6 with total masses of and and bolometric luminosities of and , with the SMA at 1.3 mm, covering the frequency ranges 216.8–220.8 GHz and 228.8–232.8 GHz, and at 850 m, covering the frequency ranges 333.7–337.7 GHz and 345.6–349.6 GHz (see Figure 2). The high-angular resolution observations have resolved the emission of the clumps into several cores, four of them being prestellar core candidates according to Wang et al. (2011): cores P1-SMA3, P6-SMA1, P6-SMA3, and P6-SMA4. Compared with the protostellar cores embedded in P1 and P6, the prestellar cores present a much lower chemical richness, with only very few species detected: all prestellar cores show CO and CO and two of them also show ¹³CO and O. None of them shows deuterated emission. Wang et al. (2011) have detected even fewer species toward the cores embedded in the massive P1 clump of another IRDC, the G28.34 + 0.06 cloud. In fact, these authors have detected only CO among the entire 8 GHz SMA band toward the embedded cores, although in all cases, the CO emission is associated with molecular outflows, indicating a protostellar nature of the cores. Regarding IRDC G11.110.12, the poor chemistry observed in the prestellar cores contrasts with that observed in the protostellar ones. In the latter, in addition to the emission of CO and isotopologues and CO, there is also emission of COMs such as OH and ¹³OH, CN, CHO, and CN, typical of HMCs, emission of OCS, ¹³CS, N, c-HCCCH, and HNCO, and emission of typical shock tracers such as SO and SiO. The richest protostellar core, P1-SMA1, also shows emission of the deuterated species DCN. Based on the different line richness and strength of the emission, even among the protostellar cores, Wang et al. (2014) conclude that chemical differentiation is present in these massive clumps, likely indicating an evolutionary sequence from core to core.

FIGURE 2

Interferometric observations with different arrays, such as the IRAM Plateau de Bure (PdBI), the Berkeley-Illinois-Maryland Association Array (BIMA), and the Very Large Array (VLA), were also carried out toward the IRDCs G29.96e and G35.20w. Both cores are located near UC H_II regions in the G29.960.02 and G35.201.74 star-forming regions, respectively, as a follow-up of the SCAMPS survey by Pillai et al. (2011). The observations targeted , D, and . The deuterated species D is a high-density tracer of cold gas in very young stellar objects (YSOs). In such a cold and dense environment, carbon-bearing molecules, particularly CO, should be depleted by freeze-out onto dust grains (Tafalla et al., 2002). This should enhance due to the fact that the ion, the progenitor of most deuterated species, is not destroyed by CO (see Sect. 2.2.1). Once the temperature near protostellar YSOs exceeds 20 K, CO should be released from the surface of dust grains, react with , and reduce the deuterium fractionation. D was detected in 11 cores in G29.96e and 13 in G35.20w, while , which is a dense tracer of cold and warm gas, was detected only in five cores in G29.96e and three in G35.20w. The low temperatures estimated ( K) together with the high , D/] in the range 0.06–0.37, several orders of magnitude above the interstellar [D/H] ratio of (Oliveira et al., 2003), suggest that some of these cores could be starless. However, the fact that molecular outflows have been observed toward both IRDCs, indicates that protostellar activity has already started in some of these cores. Unfortunately, the angular resolution of the observations does not allow us to distinguish which cores are indeed protostellar and which are prestellar. A similar study, with the same targeted species (D and ) has been conducted by Busquet et al. (2010) in IRAS 20293 + 3,952 with the PdBI and the VLA. This high-mass star-forming region, with a bolometric luminosity of 6,300 , is associated with YSOs in very different evolutionary stages, from starless core candidates to protostellar YSOs powering molecular outflows to an UC H_II region. The D emission is strongly detected toward the prestellar candidates, with high (0.1–0.8) while it is hardly detected toward the protostellar ones, with low (Busquet et al., 2010).

Deuterium fractionation has also been studied in the starless core candidate C1-N and the early evolutionary phase core C1-S in IRDC G28.37 + 0.07 by Kong et al. (2016). These two cores were originally identified as starless candidates based on their cold temperature ( K) and high (/] 0.38) by Fontani et al. (2011), who carried out a and survey of YSOs in different evolutionary stages with the IRAM 30-m telescope. However, Tan et al. (2016) mapped in CO a narrow, highly collimated bipolar outflow associated with core C1-S, indicating that protostellar activity is already present in this core. Both cores have been detected in , , and , but only C1-S in DCN (Tan et al., 2013; 2016; Kong et al., 2016), and none in o- (Kong et al., 2016). By combining single-dish and interferometric observations, Kong et al. (2016) have estimated very high values (0.2–0.7) toward these two cores.

2.1.2 High-mass protostellar objects

Hot molecular cores, the cradles of OB stars or HMPOs, are dense, n, compact, dusty cores with temperatures in excess of 100 K, and luminosities . These cores are often found in association with typical signposts of massive star formation such as UC H_II regions and maser emission of different species. HMCs exhibit_II the richest chemistry in the ISM as a result of the evaporation of the dust grain mantles by the strong radiation of the deeply embedded early-type star. This chemistry is mainly observable in molecular line emission at millimeter and sub-millimeter wavelengths (e.g., Beuther, 2007) and includes mainly hydrogenated molecules, oxygen-bearing, nitrogen-bearing, sulfur- and silicon-bearing, phosphorus-bearing, and isotopologues and deuterated species. Other species including chlorine, fluorine, sodium, potassium and iron have also been detected toward these massive cores. HMCs are the richest cores in COM emission, with species with atoms recently detected, such as 2-methoxyethanol (OH) in NGC 6334I (Fried et al., 2024), and prebiotic species. The abundance and variety of species detected towards these regions allow us to estimate the physical parameters, such as temperature, density, and mass, and, moreover to study the kinematics of these young massive stellar objects.

Unbiased line surveys have been carried out towards a few selected HMCs, with Sgr B2(N) in the Galactic center and Orion KL in the nearest high-mass star-forming region being the most targeted. The first spectral line unbiased surveys at 1 mm and 3 mm targeted Sgr B2 and OMC-1 (Orion A) in the 1980s (Johansson et al., 1984Sutton et al., 1985; Blake et al., 1986; Cummins et al., 1986; Turner, 1989). The NRAO 3 mm survey (Turner, 1989) was the only one that observed both sources and was the most sensitive one. This led to the detection of more than 700 lines in Sgr B2 and 800 in Orion KL Turner (1991). Some other HMCs, such as G31.41 + 0.31, NGC, 6334I, and NGC, 2264 CMM3, have received increasing attention, especially after the advent of ALMA. Next, we discuss the findings toward these sources in detail.

Sgr B2: The Sagittarius B2 molecular cloud, located at less than 100 pc in projected distance from the Galactic center (Molinari et al., 2010) is the most massive star-forming region in our Galaxy. This giant molecular cloud complex hosts two HMCs, Sgr B2(N) also known as the Large Molecule Heimat (LMH), and Sgr B2(M), separated by  pc in projection and with luminosities of and , respectively (Schmiedeke et al., 2016), and masses of a few (Belloche et al., 2013). Both HMCs contain a large number of UC H_II regions, X-ray sources and molecular masers (e.g., Schmiedeke et al., 2016, and references herein) that highlight the extreme environment of this cloud. The Sgr B2 molecular cloud is the most chemically rich region in the Galaxy, and historically it has been the cloud in which to search for new species, being the target of numerous spectral line unbiased surveys, as already mentioned. Focusing on more recent surveys, the two HMCs in the region have been observed with single-dish telescopes, such as with the GBT at centimeter wavelengths, as part of the GBT Legacy Survey of Prebiotic Molecules Toward Sgr B2N (PRIMOS), the IRAM 30-m telescope at millimeter wavelengths, and the Herschel space telescope at infrared wavelengths (e.g., Neill et al. 2012; 2014; Loomis et al., 2013; Zaleski et al., 2013; Belloche et al. 2013; Möller et al. 2021). They were also observed at high-angular resolution with interferometers such as SMA (Qin et al., 2011), the Australia Telescope Compact Array (ATCA) (Corby et al., 2015), BIMA (Friedel et al., 2004), and ALMA (e.g., Sánchez-Monge et al., 2017), often as the target of large line surveys such as Exploring molecular complexity with ALMA (EMoCA; Belloche et al., 2016) and ReMoCA (Belloche et al., 2019). As a result of these observations, most COMs detected in the ISM, some of them heavy COMs with atoms and some of them prebiotic, have been first detected in this cloud. In particular, toward the HMC Sgr B2(N), which is chemically richer than Sgr B2(M) by a factor in terms of detected lines (Belloche et al., 2013; Sánchez-Monge et al., 2017), some important first detections include: formamide (Rubin et al., 1971, CHO,), acetic acid (COOH, Mehringer et al., 1997), the first sugar-like molecule, glycolaldehyde ((OH)CHO, Hollis et al., 2000), propenal (CHCHO) and propanal (HC(O)C, Hollis et al., 2004), the first keto ring molecule detected in a interstellar cloud, acetamide (Hollis et al., 2006a), cyclopropenone (c-O, Hollis et al., 2006b), aminoacetonitrile (Belloche et al., 2008, CN,), ethyl formate (OCHO, Belloche et al., 2009), E-cyanomethanimine (E-HNCHCN, Zaleski et al., 2013), ethanimine (CHNH, Loomis et al., 2013), which could play a role in the formation of the amino acid alanine, or urea (C(O)Belloche et al., 2019), which like formamide contain the peptide-like bond (NCO), key for prebiotic chemistry. Belloche et al. (2016), thanks to the EMoCA ALMA survey at 3 mm, have reported the detection of deuterated methyl cyanide DCN and the tentative detection of a few other deuterated COMs (e.g., DOH, CN, and CHDCN) toward Sgr B2(N), and have estimated a low deuteration fraction of if compared to the values estimated toward other HMCs such as Orion KL (Gerin et al., 1992) and low-mass hot corinos such as 162932,422 (Taquet et al., 2014). According to the authors of this study, the low deuteration fraction could be the result of the higher kinetic temperatures typical of the Galactic center and that could reduce the degree of deuterium fractionation at the end of the prestellar phase as suggested by chemical models (e.g., (Taquet et al., 2014),). Alternatively, it might be due to an overall lower abundance of deuterium in the Galactic center. Besides COMs, Sgr B2 is also the cloud in which some iron-bearing, chlorine-bearing, and fluorine-bearing species have been first (tentatively) detected in the ISM. Walmsley et al. (2002) reported a tentative detection of iron monoxide (FeO) seen in absorption at 1 mm with the IRAM 30-m telescope that later on was confirmed by Furuya et al. (2003) with Nobeyama Millimeter Array (NMA) interferometric observations. Hydrogen fluoride (HF) was detected also in absorption with the Infrared Space Observatory (ISO) by Neufeld et al. (1997), who proposed HF as the dominant reservoir of gas-phase fluorine in this source. Chlorine-bearing species such as chloronium () has also been detected in absorption toward Sgr B2, in this case with the space telescope (Lis et al., 2010). The star formation rate (SFR) of Sgr B2 estimated by Belloche et al. (2013) is 0.04 . This SFR is much higher than that estimated toward other star-forming regions (see, e.g., Beltrán et al., 2013) and is about 2%–3% of the global SFR of the Milky Way (Lee et al., 2012). This indicates that this cloud is a mini-starburst region. This together with its location close to the Galactic center makes the physical conditions in that environment quite extreme, which could have consequences on the chemistry. Therefore, despite the wealth of new detections in this cloud, the chemistry of Sgr B2 may not be representative of that typical of HMCs in the disk of the Galaxy.

Orion KL: The Kleinmann-Low cloud, also known as the Orion KL cloud, which is part of the OMC-1 complex, is the closest (415 pc, Menten et al., 2007), most well studied high-mass star-forming region in our Galaxy, and one of the richest molecular reservoirs known. The cloud is very complex and contains several strong IR and millimeter sources, such as IRc2, the Becklin–Neugebauer object (BN), source n, and source I (SrcI). The latter source, with a bolometric luminosity of (Greenhill et al., 2004), is, together with BN, one of the most massive objects in the region (, Ginsburg et al., 2018). SrcI is associated with an UC H_II region and with typical signposts of massive star formation, such as O and SiO masers, which likely trace a massive disk (Matthews et al., 2010; Hirota et al., 2012). The first unbiased line surveys toward Orion KL were conducted in the 1980s with different single-dish telescopes and in various bands, from to 360 GHz, with the Onsala Space Observatory (OSO, GHz) 20m telescope, the Owens Valley Radio Observatory (OVRO, GHz) 10.4m telescope, and the National Radio Astronomy Observatory (NRAO, GHz, GHz, and GHz) 11m telescope (Johansson et al., 1984; Sutton et al., 1985; Blake et al., 1986; Jewell et al., 1989; Turner, 1989). In more recent years, line surveys have been conducted at (sub-)millimeter wavelengths, mainly with single-dish telescopes larger than those used in the ’80s (e.g., Caltech Submillimeter Observatory, James Clerk Maxwell Telescope, GBT, IRAM 30-m, Effelsberg 100-m, Nobeyama 45-m Tianma 65-m), thus probing smaller linear scales in the frequency bands previously observed. The frequency range covered goes from to  GHz (Schilke et al., 2001; White et al., 2003; Comito et al., 2005; Goddi et al., 2009; Tercero et al., 2010; Esplugues et al., 2013; Gong et al., 2015; Rizzo et al., 2017; Suzuki et al., 2018; Liu et al., 2022; 2024), providing lines with a wide range of excitation conditions. Orion KL has also been surveyed with interferometers such as CARMA (Friedel and Looney, 2017), SMA (Beuther et al., 2005b), and it has been selected for ALMA Science Verification observations of Band 6 (214–246 GHz frequency range) and, more recently, of Band 1 (from 35 to 50 GHz). At infrared wavelengths, Orion KL has been observed by the ISO space telescope from 44 to 188 m (Lerate et al., 2006) and by from 480 to 1907 GHz as part of the Herschel Observations of EXtra Ordinary Sources (HEXOS) guaranteed time key program (Bergin et al., 2010; Crockett et al., 2010; Crockett et al., 2010). These line surveys have detected many molecular species, especially COMs, confirming the richness of Orion KL, and have discovered new species in the ISM or in a star-forming region: e.g., COMs, such as methyl acetate () and the gauche conformer of ethyl formate, OCOH (Tercero et al., 2010), salty species, such as NaCl, KCl, and their isotopologues (first time detected in a star-forming region by Ginsburg et al., 2019), and new maser species, such as the vibrationally excited O maser (Hirota et al., 2012). Several studies have indicated that the complex chemistry of the Orion KL region displays chemical differentiation, where the distribution of complex nitrogen-bearing molecules is spatially different from that of oxygen-bearing molecules. In this scenario, nitrogen-bearing molecules would trace the HMC while oxygen-bearing molecules would trace the Orion Compact Ridge (e.g., Suzuki et al., 2018). However, Friedel and Widicus Weaver (2012) suggest that while for some species, such as acetone (), chemical processes could be responsible for the different spatial distribution, for other COMs, such as ethyl cyanide (CN) and methyl formate (OCHO), the spatial distribution will instead be determined by local physical conditions.

G31.41+0.31: This is a well-known HMC located at 3.75 kpc (Immer et al., 2019), with a bolometric luminosity of (Osorio et al., 2009), which does not have a UC H_II region embedded (Cesaroni et al., 2010). Interferometric studies at millimeter wavelengths with the IRAM Plateau de Bure Interferometer, the predecessor of the NOrthern Extended Millimeter Array (NOEMA), SMA, and ALMA have revealed that the core is very chemically rich (Beltrán et al., 2005; Beltrán et al., 2009; Beltrán et al., 2018; Rivilla et al., 2017a, presenting prominent emission in a large number of COMs). In particular, the first detection of glycolaldehyde outside the Galactic center has been obtained toward G31.41 + 0.31 (Beltrán et al., 2009), and heavy COMs such as both conformers of ethylene glycol (10 atoms) have also been observed ((Rivilla et al., 2017a; Mininni et al., 2023). The G31.41 + 0.31 Unbiased ALMA sPectral Observational Survey (GUAPOS) carried out an unbiased line survey of this core of the whole ALMA Band 3 (3 mm; see Figure  3). This survey has identified more than 40 molecules in the region (including isotopologues and lines from vibrationally-excited states), including oxygen-, nitrogen-, sulfur-, silicon-, and phosphorus-bearing species, with some of them being COMs such as formamide, acetamide, and N-methylformamide (NHCHO) (Mininni et al., 2018; Mininni et al., 2023; Colzi et al., 2021; García de la Concepción et al., 2022; Fontani et al., 2024; López-Gallifa et al., 2024). For some COMs, such as the isomers methyl formate (OCHO) and acetic acid (COOH), the abundances estimated toward G31.41 + 0.31 are higher than those detected toward Sgr B2. Regarding phosphorus-bearing species, PN has clearly been detected while PO has only been tentatively detected Fontani et al. (2024). The fact that PN has been detected along the molecular outflows in the region and not toward the position of the HMC, together with its association with SiO and other typical shock tracers, such as SO, confirms that PN is likely a product of shock chemistry, as previously suggested by (Lefloch et al., 2016; Rivilla et al., 2020). The G31.41 + 0.31 HMC is also one of the high-mass targets of the James Webb Space Telescope (JWST) Observations of Young protoStars (JOYS) Guaranteed Time Program, which aims at tracing the physical and chemical properties of about 30 young low- and high-mass protostars with the MIRI medium resolution spectrometer (MRS) instrument and its integral field unit (IFU) between 5 and 28 m (van Dishoeck et al., 2025). One of the goals of this project is to trace COMs, such as ethanol, methyl formate or dimethyl ether, in ices and compare their abundances with those found in gas phase at (sub-)millimeter wavelengths. Initial results on the high-mass protostars IRAS 23385 + 6053 and IRAS 180891732 have identified several COMs in ices. For some of them, such as OH, the absorption feature is clearly and unambiguously identified, in other cases ice mixtures containing absorption features of acetaldehyde (CHO), ethanol (OH), methyl formate (OCHO), dimethyl ether () and acetone () (acetone) have been identified (Rocha et al., 2024; van Dishoeck et al., 2025).

FIGURE 3

NGC 6334I: Another HMC that has received a lot of attention from a chemical point of view is NGC 6334I. This is one of the sources embedded in the NGC 6334 giant molecular cloud located at a distance of 1.7 kpc (Neckel, 1978), and is the strongest one from millimeter to far-infrared wavelengths, with a bolometric luminosity of (Sandell, 2000). High-angular resolution dust continuum emission observations have resolved the emission of this HMC in several millimeter sources, suggesting that this HMC hosts a protocluster (Hunter et al., 2006). NGC 6334I is very chemically rich, and for the density of lines, it has been compared to Sgr B2(N) and Orion KL (Thorwirth et al., 2003). Unbiased spectral surveys at millimeter and submillimter have been conducted with single-dish telescopes (e.g., Swedish-ESO Submillimetre Telescope (SEST), the Atacama Pathfinder EXperiment (APEX), JCMT, Mopra) in various bands, from to 810 GHz (McCutcheon et al., 2000; Thorwirth et al., 2003; Schilke et al., 2006). The source has also been observed with from 480 to 1907 GHz with the Heterodyne Instrument for the Far-Infrared (HIFI) as part of the Chemical HErschel Surveys of star-forming regions (CHESS) guaranteed time key program (Zernickel et al., 2012). High-angular resolution narrowband line observations with SMA or ATCA or ALMA are available (e.g., Beuther et al., 2005a; Zernickel et al., 2012; Fried et al., 2024; McGuire et al., 2018) have carried out a pilot project with ALMA Band 10 by observing the source in the frequency range 874 to 881 GHz. Like the previous HMCs analysed, NGC 6334I is rich in emission of COMs, including prebiotic species like glycolaldehyde (e.g., McGuire et al., 2018). In a recent work, Fried et al. (2024) have detected with ALMA, for the first time in the ISM, one of the heaviest COMs (13 atoms), 2-methoxyethanol (OH). To secure the detection of such a heavy COM, these authors have used ALMA Band 4 observations from 130 to  GHz.

There are many more HMCs that have been observed with broadband or unbiased spectral surveys. As an example, NGC 2264 CMM3 at the center of the protocluter, NGC 2264 C, was observed with the Nobeyama 45-m telescope and the Atacama Submillimeter Telescope Experiment (ASTE) 10-m telescope in the 4 mm, 3 mm, and 0.8 mm bands and this led to the identification of 36 molecular species and 30 isotopologues, including many emission lines of carbon-chain molecules (e.g., N, H, and CCS), deuterated species (with a deuteration fractionation ranging 1%–4%), and COMs (Watanabe et al., 2015). On the other hand, van der Walt et al. (2021) observed CygX-N30 also known as W75N (B), a star-forming region containing several millimeter cores in the Cygnus X molecular cloud, with the SMA interferometer. They observed from 329 to 361 GHz as part of the Protostellar Interferometric Line Survey of the Cygnus X region (PILS-Cygnus), and identified 29 different molecular species and their isotopologues, including many COMs and deuterated water (HDO). The authors claim that chemical differentiation is detected in this region, with oxygen-bearing molecular species peaking toward the inner cores and nitrogen- and sulfur-bearing species peaking toward the outer ones. van der Walt et al. (2021) interpreted this result as being due to an evolutionary effect: The outer cores could be more evolved than the inner ones, where the gas-phase chemistry had less time to form N- and S-bearing species, and formed predominantly O-bearing ones.

Line surveys have also been conducted toward proposed HMC precursors such as G328.25510.5321 (Csengeri et al., 2018; Bouscasse et al., 2022). This high-mass clump, with weak or no emission in the 21–24 m range and a bolometric luminosity of , is embedded in the MSXDC G328.2500.51 dark cloud, which is located at a distance 2.5 kpc (Csengeri et al., 2017). The source was observed with ALMA from 333 to 349 GHz as part of the Search for High-mass Protostars with ALMA up to kilo-parsec scales (SPARKS) project by Csengeri et al. (2018) and with an unbiased line 2 mm, 1.2 mm, and 0.8 mm with APEX by Bouscasse et al. (2022). The SPARKS survey has detected emission from 10 different COMs toward this core, such as the oxygen-bearing species ethanol, acetone, and ethylene glycol, and the nitrogen-bearing ones vinyl cyanide, ethyl cyanide, and formamide, which is a prebiotic species. The spatial distribution of oxygen- and nitrogen-bearing COMs in this region is different, suggesting chemical differentiation in this core: nitrogen-bearing species peak toward the protostar and oxygen-bearing COMs being associated with two accretion shock spots located in the inner envelope (Csengeri et al., 2019). Based on the association with the accretion spots, these authors suggest that the COM emission in this core is produced by the sublimation of these species, but not as a result of radiative heating of dust grains, as is usually assumed for HMCs. Deuterated water has also been observed in this core with a distribution similar to that of the oxygen-bearing COMs methyl formate and methanol. On the other hand, Bouscasse et al. (2022) have detected 39 species plus 26 isotopologues with their unbiased millimeter survey, including a sulfur-bearing COM, CSH, and a high abundance of the sulfur-bearing molecular ions and .

2.1.3 Hyper- and ultra-compact H_II regions

OB (proto)stars deeply embedded in molecular clouds start heating and ionizing the surrounding environment as they reach the Zero Age Main Sequence (ZAMS). As the heated region grows, the ultraviolet (UV) radiation from the young embedded star will break through the core to first become a compact and dense bubble of ionised gas, known as hyper-compact H_II (HC H_II) region (Kurtz, 2005), with a size 0.05 pc and a density , that eventually will start expanding, becoming first an ultra-compact H_II (UC H_II) region of size 0.1 pc and a density , and later on, an extended or giant H_II region, with sizes of the order of 100 pc and densities as low a few 10s of (Kurtz, 2002). The spectra of HC and UC H_II regions are characterised by a chemistry similar to that of photo-dissociated regions (PDRs) with forbidden atomic lines of neutral oxygen and neutral and ionised carbon and nitrogen, and isotopologues, such as [OI] at 63 and 145 m, [CII] and CII], at 158 m, [NII] at 205 m, and [CI] at 370 and 609 m, as well as J-ladder emission of CO and isotopologues (e.g., Huang et al., 1999; Rodón et al., 2010; Ossenkopf et al., 2013; Anderson et al., 2019; Kirsanova et al., 2020), and the presence of recombination lines (RRLs) from centimeter to ultraviolet wavelengths with very broad widths, of the order of 30–40 km , or even wider (Kurtz, 2005).

Unbiased spectral line surveys of selected HC and UC H_II have been carried out mainly at centimeter and millimeter wavelengths with single-dish telescopes (e.g., Bell et al., 1993; Ginard et al., 2012; Li et al., 2017; Watanabe et al., 2017; Liu et al., 2022) and at infrared wavelengths with (e.g., Rodón et al., 2010). Bell et al. (1993) carried out a survey at centimeter wavelengths (between 17.6 and 22 GHz) of part of the W51 H_II complex, including the well-known HC and UC H_II regions W51 IRS1, W51e1, and W51e2 regions with the NRAO 43-m telescope. This survey detected 94 hydrogen (H) or helium (He) broad RRLs, as well as carbon chain molecules such as N and N, , and its isotopologue and the sulfur-bearing COM SH. Similar species, including CN, H, CCH, different sulfur-bearing molecules such as CS and isotopologues, OCS, and SO, and recombination lines have been observed more recently by Watanabe et al. (2017) thanks to their spectral line survey at 3 mm (between 85.1 and 101.1 GHz and between 107.0 and 114.9 GHz) carried out with the Mopra 22-m telescope. It is also worth mentioning that Rivilla et al. (2016) report the first detection of the prebiotic important species phosphorus monoxide, PO, toward W51e1 and W51e2.

One of the most observed UC H_II regions is Mon R2, which has been the target of spectral line surveys at 1, 2, and 3 mm carried out with the IRAM 30-m telescope (e.g., Ginard et al., 2012; Treviño-Morales et al., 2014). Ginard et al. (2012) detected more than 30 different species (including isotopologues) at 1 and 3 mm. Some of the species, such as CN, HCN, HCO, or the hydrocarbons H, and c- are typical of PDRs, although some can also be detected in HMCs (e.g., HCO and CN in G31.41 + 0.31; Beltrán, private communication), while species such as and H confirm that UV radiation plays an important role in the chemistry of this region. Other species important for PDR chemistry, such as the reactive ions C and HO, have been detected in Mon R2 and also in G29.96 + 0.02 by Rizzo et al. (2003), who reported the first detection of these species in UC H_II regions. The survey of Ginard et al. (2012) has also revealed sulfur-bearing species, such as CS, SO, and , and the deuterated species DCN and D, with high values of 0.03 and 0.05, respectively, for this kind of regions. On the other hand, Treviño-Morales et al. (2014) detected, in addition to DCN and D, many more deuterated species (DNC, , HDCO, CO, D, and at 1, 2, and 3 mm, and estimated values of 0.01 for HNC, HCN, H and CO, and for , , and . Ginard et al. (2012) also report the surprising detection of emission from light COMs such as CN, OH, and H that is usually not detected in such environments. The origin of this emission is puzzling and it could be associated with dense and well-shielded cores within the molecular cloud instead of with the UC H_II region or PDR-like region. Obviously, RRLs have also been detected toward this region and, in particular; Jiménez-Serra et al. (2013) report the detection of the H30 at 231.9 GHz (1.3 mm) and H26 at 353.6 GHz (0.85 mm) recombination line masers toward Mon R2, where the emission is maser amplified.

Li et al. (2017) carried out a line survey with the IRAM 30m telescope from 84.5 to 92.3 GHz at 3 mm and from 143.7 to 147.7 GHz at 2 mm of four massive star-forming regions associated with H_II regions at different stages of evolution: Cepheus A (Cep A), hosting HC H_II regions, DR21S, associated with cometary-shaped H_II regions, S76E, hosting UC H_II regions but also with more evolved cometary-shaped and extended ring-like H_II regions, and G34.26 + 0.15 (G34), containing an extended H_II region (see Figure 4). This survey has detected many different species towards these regions, including long carbon chain molecules such as N and istopologues (with ), H, c-, and H, deuterated molecules such as DCN, , and D, and shock tracers such as SiO, SO, and HNCO toward all sources. Species associated with PDRs and hydrogen RRLs (H42 only in DR21S), were detected toward all except Cepheus A. In fact, the study concluded that the chemical richness of the sources increased with their evolution, as indicated by the fact that while 78 different molecular species plus 3 RRLs were detected in G34.26 + 0.15, “only” 35 different species were identified in Cepheus A in the molecular envelope surrounding the HC H_II region.

FIGURE 4

In their single-dish Q-band line survey toward Orion KL, covering the 34.8–50 GHz frequency range, Liu et al. (2022) also observed the well-known H_II region M42. A total of 126 hydrogen RRLs and 40 He RRLs, with a maximum of 16 and 7, respectively (H135 and He99), were detected toward this H_II region. The survey has also tentatively detected 11 carbon RRLs with a maximum of 3 (C81), but as Liu et al. (2022) discuss, these lines should originate from the PDRs located between M42 and the Orion Bar.

At infrared wavelengths, Rodón et al. (2010) observed the optically visible UC H_II region Sh2-104 with the PACS and SPIRE instruments of at 100, 160, 250, 350, and 500 m. They detected the -ladders of ¹²CO and ¹³CO up to the = 13–12 and = 9–8 transitions, respectively, as well as proxies for ionizing flux such as ionised nitrogen [NII] and neutral carbon [C I]. Some other authors, like Kirsanova et al. (2020), have observed PDRs around H_II regions, and have revealed forbidden atomic lines of ionised carbon and isotopologue ([CII], CII]) and neutral oxygen ([OI]), as expected for the chemistry of such regions.

2.1.4 Summary, open questions, and future prospects

We have analysed the main results of line surveys of selected sources in different evolutionary stages. Starting with HMSCs, as discussed above, the main problem that one encounters is the real existence of such starless cores. Many studies have revealed that previously thought starless cores embedded in IRDCs are indeed protostellar objects as indicated by the fact that they are powering molecular outflows (e.g., Wang et al., 2011; Tan et al., 2013). However, despite the fact that these cores show already protostellar activity, their evolutionary stage is earlier than that of the well-known HMPOs discussed in Sect. 2.1.2) and their chemistry is also different. The most promising prestellar core candidates exhibit a poorly rich and very simple chemistry, with emission of simple species such as CO and isotopologues, CO, and and emission of deuterated species such as DCN, D, and . In fact, the main characteristic of HMSCs is the high , which can be orders of magnitude above the interstellar one.

The chemistry of slightly more evolved YSOs, i.e., protostellar cores embedded in IRDCs, is also characterised by the presence of deuterated species. However, the chemical richness of these protostellar cores is greater, and in fact, there spectra show that the emission of COMs, such as CN, CHO, or CN, and of shock tracers, such as SiO and SO, associated with their driven molecular outflows, is important.

The richest chemistry in the ISM, as revealed by unbiased spectral line surveys, is that of HMPOs associated with HMCs. Such chemistry is dominated by the emission of COMs, which can be very heavy or large, i.e., they can consist of many atoms (e.g., 2-methoxyethanol, OH, Fried et al., 2024), and can be prebiotic, such as formamide or glycolaldehyde. The most common species observed in HMCs are oxygen-bearing, nitrogen-bearing, sulfur-bearing, and silicon-bearing species, although species containing phosphorus, iron, chlorine, fluorine, and potassium, and salty species, such as NaCl and KCl and their isotopologues, have also been detected. In some HMCs, the spatial distribution of oxygen- and nitrogen-bearing species, particularly COMs, is different and this has been explained in some cases as the result of chemical differentiation in the core. However, it cannot be ruled out that local chemical processes or physical conditions are responsible for the different spatial distribution. The chemistry of HMPOs is also rich in emission from vibrationally excited states of (e.g., N , CN , and OH ), and maser emission, mainly of O, OH, and OH. On the other hand, despite deuterated species of, e.g., CN, OH, and O, have been detected toward HMCs, the deuteration fractions are low.

The chemistry of the most evolved massive YSOs, i.e., HC and UC H_II regions, is that of highly UV-irradiated environments, and therefore, it resembles that of PDRs. Their IR spectra is rich in forbidden atomic lines of neutral oxygen and neutral and ionised carbon and nitrogen, J-ladder emission of CO, and isotopologues. Another characteristic feature of their spectra from radio to ultraviolet wavelengths is the presence of very broad recombination lines (RRLs).

The future of unbiased spectral line surveys is bright thanks to the new Atacama Large Millimeter/submillimeter Array (ALMA) Wideband Sensitivity Upgrade (WSU). This upgrade, which will start in 2025–26 with the new Band two receiver operating from 67 to 116 GHz, will increase the instantaneous bandwidth of the ALMA receivers and the speed of spectral line surveys by a factor of 2–4, as fewer tunings will be required to cover the full ALMA bands. The WSU will also increase the spectral line sensitivity by improving the receiver temperatures. In summary, the WSU will advance astrochemistry studies by enabling the detection of fainter species and rapidly increasing the number of observed sources. This will improve statistics and will allow to more accurately determine the chemical complexity of massive YSOs at all evolutionary stages.

2.2 Source surveys of selected lines

Several molecular abundances, or abundance ratios, were proposed as chemical clocks by investigating their variation in source surveys of cores divided in different evolutionary stages. In this section, we review these studies focussing on the following type of molecules: simple molecules, less abundant isotopologues and isotopic ratios, COMs, and shock tracers. We conclude this section with a quick overview of the studies that proposed maser emission lines as indicators of evolution.

2.2.1 Simple molecules

Several chemical evolution studies took N₂H⁺as a reference species. N₂H⁺ is one of the most abundant molecular ions, and it is clearly detected in all stages of the high-mass star-formation process with a moderate variation from the HMSC to the UC H_II phase (Hoq et al., 2013; Gerner et al., 2015). Moreover, its progenitor species, that is , appears to be less depleted than CO, the progenitor of another very abundant ion, . Wang et al. (2023) found that the N/N₂H⁺ ratio increases with evolution when comparing IRAM 30m observations of 61 UC H_II regions with those towards HMSCs and HMPOs previously obtained by Taniguchi et al. (2019). The increase, mostly due to the increase of the N abundance, was attributed to the enhanced amount in the gas of , the main progenitor of N through the neutral-neutral reaction (Equation 1, Takano et al., 1998):In contrast, N₂H⁺ produced in the gas by the ion-neutral reaction , is also formed efficiently in the early cold phases. Taniguchi et al. (2018) have also measured an average increase of the N column density with evolution through observations of 17 HMSCs and 35 HMPOs. However, its fractional abundance with respect to is found to decrease with time, and the N column density decreases with the luminosity-to-mass ratio of the targets, a well known evolutionary indicator (e.g., López-Sepulcre et al., 2011). This suggests a destruction of N with protostellar activity. Therefore, the use of the N/N₂H⁺ ratio as evolutionary indicator is not yet fully clear.

Other proposed chemical clocks involving N₂H⁺ are N₂H⁺/, and N₂H⁺/CCS. The N₂H⁺ ratio was investigated by Yu and Wang (2015) using data of the Millimetre Astronomy Legacy Team 90 GHz (MALT90) survey. MALT90 used the Mopra 22 m single-dish telescope of the Australia Telescope National Facility to map 2000 dense molecular clumps hosting different stages of the high-mass star formation, from HMSCs to UC H_II’s, in 16 lines of dense gas tracers with frequencies in between and GHz, among which the transitions of , C, HCN, HNC, C, and N₂H⁺ (Jackson et al., 2013). To derive the N₂H⁺/ ratio, Yu and Wang (2015) used the optically thin isotopologue C. They report a marginal decrease of this ratio from the HMPO to the UC H_II stage, possibly due to depletion of N₂H⁺ due to the formation of the ionised expanding region. Chen J. L. et al. (2025) proposed that the N₂H⁺/CCS ratio increases with evolution, perhaps owing to the simultaneous increase of N₂H⁺ production and decrease of CCS abundance. CCS is indeed a species thought to be formed in the gas-phase by atomic carbon of sulphurated ions at early stages, although Fontani et al. (2023) did not find a significant decrease of its abundance with the evolution in high-mass star-forming cores. However, the studies of Chen J. L. et al. (2025) and Fontani et al. (2023) probe sources with different average heliocentric distances, and hence potentially different linear scales.

Fontani et al. (2021) suggested that the chemistry of HC, a molecular ion rare but believed to be the main gas-phase precursor of HCN and HNC (upon dissociative recombination), is different in HMSCs and more evolved cores. In particular, the abundance ratios HC/HCN and HC/ are both below 0.01 in HMPOs and UC H_II’s, and higher than this threshold in HMSCs. Chemical modelling suggests that this difference is due to the different formation of HC: in HMSCs, the main progenitors should be and HN, while in the later stages, where the abundance of and HN drops by three orders of magnitude, should be and HCN/HNC (Fontani et al., 2021). Therefore, these ratios can be useful tools to identify HMSC candidates.

2.2.2 Isotopic fractionation

The process indicated under the name of isotopic fractionation refers to the chemical reactions that affect the distribution of the different isotopes of an element into molecules. Under specific physical conditions, the isotopes can be unequally distributed with respect to their elemental ratio. For example, deuterated molecules in cold and dense cores can have abundances higher than the D/H elemental ratio (Zavarygin et al., 2018) even by several orders of magnitude, owing to the exothermic forward reaction (Watson, 1974):If the gas kinetic temperature, , is lower than K and is mainly in the para-form², this reaction can proceed only from left to right, producing high / ratios. Moreover, at densities higher than , CO, the main destruction partner of , is heavily depleted (namely, the gaseous CO molecules get frozen onto dust grains, and hence the CO gaseous molecular abundance decreases, see e.g., Bergin and Tafalla, 2007). CO depletion boosts reactions of with other non-depleted neutrals such as, for example, . This explains why the deuterium fraction, , namely, the column density ratio between a deuterated species and its hydrogenated counterpart, of non depleted molecules is claimed to be a powerful chemical diagnostic to distinguish between cold/early and warm/late evolutionary stages in star-forming cores (e.g., Caselli and Ceccarelli, 2012; Ceccarelli et al., 2014a).

Observations of high-mass star-forming cores in different evolutionary stages have indeed highlighted that the /N₂H⁺ and the / ratios decrease from the early phases to the more evolved stages (e.g., Fontani et al., 2011; Gerner et al., 2015; Giannetti et al., 2019; Li et al., 2022; Pazukhin et al., 2023; Sabatini et al., 2024). Figure 5 shows the observed /N₂H⁺ ratios measured by Fontani et al. (2011) (panel (a)), and by Giannetti et al. (2019) and Sabatini et al. (2024) (both in panel (b)). In each work, the classification of the cores into different evolutionary stages is based on different criteria. For example, Fontani et al. (2011) separate the HMSC and HMPO classes based on the presence/absence of outflows and any embedded mid-infrared source, while Sabatini et al. (2024) consider as protostellar objects sources detected at 70 m but still undetected at 24 m. Fontani et al. (2011) make a further distinction in the HMSC group between cold and warm cores based on a lower or higher than 20 K, respectively (Figure 5). Nevertheless, the /N₂H⁺ fraction shows a decreasing trend with evolution in all works, regardless of the criteria used to classify the sources, and to the analysis performed to derive the molecular column densities (e.g., using a local thermodynamic equilibrium (LTE) or non-LTE approach). Interestingly, the very high /N₂H⁺ ratio in HMSC candidates is inferred from both single-dish observations (e.g., Fontani et al., 2011; Gerner et al., 2015; Sabatini et al., 2024), and interferometer images (e.g., Tan et al., 2013; Kong et al., 2016; Li et al., 2022), suggesting that this behaviour does not depend on the probed angular scale.

FIGURE 5

Figure 5 also compares the /N₂H⁺ ratio with o-/, DNC/HNC, D/, and DOH/OH. While the evolutionary decreasing trend seen in the o-/ ratio clearly resembles that obtained in /N₂H⁺, the DNC/HNC ratio shows a marginal decrease with evolution, D/ shows no significant decrease, and DOH/OH seems to have the largest value at the HMPO stage. The natural explanation for the similar decreasing of the two molecular ions is the gas temperature enhancement with evolution, which causes the progressive destruction of and of its daughter species via the backward reaction 2. The increase in the ionisation fraction during protostellar evolution can also contribute to regulate the abundance of these ions (e.g., Socci et al., 2024). The other species in panel (a) of Figure 5 are believed to form predominantly in the gas (DNC and HNC), or partially (D and ) or totally (DOH and OH) on grain ice mantles. Ammonia, methanol, and their deuterated isotopologues, are produced efficiently on grain mantles during the early dense and cold phases through hydrogenation of N and CO, respectively. Then, when the nascent protostellar object heats up the surrounding gas, the evaporation of the grain mantles progressively releases these molecules into the gas. Finally, as the HMPO evolves towards a UC H_II region, the deuterated species are expected to be gradually destroyed because of the higher efficiency of the backward endothermic gas-phase reactions. However, while can also have an efficient formation route in cold gas (e.g., Herbst, 2021), OH is expected to be a pure surface chemistry product. The trend shown in Figure 5 for the DOH/OH ratio is overall consistent with this scenario, while that of the D/ ratio suggests that the different formation and destruction mechanisms balance each other out along evolution. As for the behaviour of the DNC/HNC ratio, although DNC is believed to be a gas-phase product like , the destruction of DNC into the warm gas appears to be slower than that occurring to (Sakai et al., 2012; Fontani et al., 2014), and this could explain the flatter decrease of its with time.

Are other isotopic ratios (e.g., ¹⁴N/¹⁵N, ¹⁶O/¹⁸O, and ¹²C/¹³C) also evolutionary indicators? In principle, all backward fractionation reactions as that in Equation 2 are endothermic and have temperature thresholds due to the lower zero-point energy of the molecules containing the heavier, less abundant isotope (Watson, 1974). The zero-point energy depends on the reduced molecular mass, therefore the difference between reactants and products is more pronounced for itotopologues that have remarkable mass differences, like those containing D and H (Equation 2). In contrast, the exothermicities of the reactions that drive the isotopic fractionation of oxygen, carbon, and nitrogen, are of an order of magnitude lower than that in Equation 2, owing to the smaller mass difference between reactants and products (e.g., Mladenović and Roueff, 2014; 2017). Therefore, the greatest effect is expected in the D/H ratio, and the observational studies mentioned above confirmed this. Nevertheless, several studies have investigated whether some of the other isotopic fractions are evolutionary indicators. In this respect, the ¹⁴N/¹⁵N ratio is particularly intriguing because it shows a variation of an order of magnitude in low- and high-mass starless and protostellar cores, which even depends on the molecule used (e.g., Caselli and Ceccarelli, 2012; Bizzocchi et al., 2013; Fontani et al., 2015b; Daniel et al., 2016; Colzi et al., 2018a; De Simone et al., 2018; Redaelli et al., 2020). Fontani et al. (2015b) and Colzi et al. (2018a) investigated the ¹⁴N/¹⁵N ratio in N₂H⁺ and HCN, respectively, in the evolutionary sample divided in HMSCs, HMPOs, and UC H_IIs previously studied in deuterated molecules by Fontani et al. (2011) and Fontani et al. (2015a). Both studies did not show a clear trend of ¹⁴N/¹⁵N with evolution. Moreover, the ¹⁴N/¹⁵N ratios measured from either N₂H⁺ and HNC both show a limited dispersion (¹⁴N/¹⁵N and , respectively) around the protosolar nebula value (Marty et al., 2011). This indicates that the chemical evolution does not seem to play a role in the fractionation of nitrogen. Rather, the ¹⁴N/¹⁵N ratio seems more sensitive to nucleosynthesis processes, that can change the isotopic ratios across the Galaxy because of the varying stellar yields (e.g., Colzi et al., 2018b; Colzi et al., 2022; Romano, 2022).

2.2.3 Carbon chains and complex organic molecules

Carbon chains and COMs are essential species in astrochemistry. Both are important interstellar reservoirs of carbon, and precursors of prebiotic molecules (e.g., Caselli and Ceccarelli, 2012; Jørgensen et al., 2020), but while carbon chains are thought to be formed from atomic (neutral or ionic) carbon, abundant in the early stages of star formation when C is not yet locked almost totally in CO, the emission of COMs characterises especially the evolved stages. Observations of some carbon chain species (H, CCS, c-, N, N) were carried out by Taniguchi et al. (2018) and Taniguchi et al. (2019) using the Nobeyama 45m telescope towards a sample of HMSCs and HMPOs. As mentioned in Sect. 2.2.1, they proposed the N₂H⁺/N as evolutionary tracer because associated with a decreasing trend with evolution, due to simultaneous N₂H⁺ destruction upon reaction with CO, and N production through warm carbon chain chemistry (Sakai and Yamamoto, 2013). Fontani et al. (2025) studied the evolution of several carbon-bearing species through the observational project ’CHemical Evolution in MassIve star-forming COres (CHEMICO)’. The project, through an unbiased spectral survey obtained with the IRAM 30m telescope, aims at investigating any aspect of the chemical evolution of high-mass star-forming cores by observing representatives of the three main evolutionary categories. They found that carbon chains and hydrocarbons tend to trace gas at lower temperatures than cyanide species such as N, CN, and N. Moreover, the only species whose measured fractional abundances show a significant enhancement with evolution are the COMs CCH and CN.

COMs are believed to be formed either through gas-phase chemical reactions (e.g., Duley and Williams, 1985; Vasyunin and Herbst, 2013a; Balucani et al., 2015; Skouteris et al., 2018), and surface chemistry processes (e.g., Hasegawa et al., 1992; Ruffle and Herbst, 2000; Garrod et al., 2008; Ruaud et al., 2015). These processes are not completely independent and can interplay (e.g., Hasegawa et al., 1992; Balucani et al., 2015), as we will discuss in Section 3. However, no matter what the formation pathway is, their abundance in the gas-phase is expected to increase with temperature, and thus evolution, because higher temperatures trigger endothermic gas-phase reactions and cause grain mantle evaporation. The observational studies conducted so far agree with this general picture, although COMs have also been detected in IRDC cores (e.g., Vasyunina et al., 2014; Beaklini et al., 2020). Gerner et al. (2014) studied the chemical content of a sample of 59 high-mass star-forming cores divided in the evolutionary groups in Figure 1, but with the further distinction, in the HMPO group, between early chemically poorer objects and later HMCs characterised by richer chemistry. The work focussed on simple molecules, but included also the COMs OHOH and CN, and found for both a progressive increase in the molecular abundance from the HMSC to the HMC stage. Coletta et al. (2020) surveyed 39 high-mass star-forming cores. This sample includes the 27 sources selected and studied by Fontani et al. (2011), and additional nine objects classified as UC H_IIs or intermediate between the HMPO and UC H_II stage. The species studied were OCHO, , CN, and CHO. The fractional abundances of all these species clearly increase with evolution, covering 6 orders of magnitude in the ratio. Using ALMA, the project ’Complex Chemistry in hot Cores with ALMA’ (CoCCoA, Chen et al., 2023; Chen Y. et al., 2025) surveyed 14 high-mass chemically rich star-forming regions in several oxygen-bearing COMs. Among these, acetone () is found to be a peculiar species, because the abundance ratio acetone-to-methanol is lower in the gas phase than in ices by an order of magnitude (Chen et al., 2024). This suggests gas-phase destruction after sublimation from grain mantles, and indicates that care needs to be taken when using COMs as evolutionary indicators, because important gas-phase destruction pathways in the late evolutionary stages could be overlooked.

2.2.4 Shock tracers

Protostellar outflows are ubiquitous in the formation process of stars of all masses (Lee, 2020), which can locally attain velocities greater than 100 km . The shock due to the passage of the outflow can destroy both the ice mantles (grain sputtering) and the refractory cores (grain shattering) of dust grains. Molecules efficiently formed on ice mantles via surface chemistry, or synthesised in the warm gas from material released from the refractory cores, are indeed greatly enhanced in abundance. As stated in Sect. 2.1, typical selective shock tracers are Silicon- (SiO, SiS), sulphur- (SO, , S), and phosphorus-bearing (PO, PN) species. But also species that are efficiently produced on ice mantles (O, OH, OH, CO, , HCN), increase their abundance by orders of magnitude in shocks (e.g., Bachiller and Pérez Gutiérrez, 1997; Jørgensen et al., 2004; Lee, 2020; Codella et al., 2020).

SiO is certainly believed to be the most selective indicator of protostellar shocks. López-Sepulcre et al. (2011) observed with the IRAM 30m telescope the emission in 32 infrared dark cloud clumps, and detected emission at high velocities in most of them . Moreover the line luminosity is found to drop with the ratio, which suggests a decline of protostellar jet activity with evolution. A different result was achieved by Rodríguez et al. (2023), who observed with the VLA the SiO line towards ten jet candidates associated with high-mass protostars, and found no correlation between the line luminosity and the ratio. Similarly, Liu et al. (2021) observed with ALMA the SiO line in 32 infrared dark cloud clumps, and found no (anti-)correlation between line intensity and the ratio. These studies mention the possibility that the lack of an anti-correlation between line intensity, or line luminosity, and the ratio is due to a very similar age of the sources studied.

Some sulphur-bearing molecules, in particular and S, are also typical tracers of protostellar outflows and jets, as indicated by the correlation with SiO both in abundance (e.g., Luo et al., 2024) and in spatial emission (e.g., Fontani et al., 2024). Despite a long history of study, sulphur chemistry in the interstellar medium still faces the problem that its main reservoir is unknown. Observations of gaseous S-bearing molecules (Woods et al., 2015; Vastel et al., 2018; Rivière-Marichalar et al., 2020; Bouscasse et al., 2022; Fontani et al., 2023) indicate a sulphur abundance in the molecular gas of approximately percent of the elemental cosmic value (, Lodders, 2003), with a strong dependence on the environment (Fuente et al., 2023). However, observations of S-bearing molecules on ices provide abundances that are too low to solve the problem of the depletion seen in the gas (e.g., Boogert et al., 2015; McClure et al., 2023). Therefore, any attempt to study the chemical evolution of S-bearing molecules is challenging. From a pure observational point of view, several S-bearing species were proposed as evolutionary indicators in surveys of high-mass star-forming cores. Fontani et al. (2023) observed simple S-bearing species towards a limited sample of HMSCs, HMPOs, and UC H_IIs, previously studied in various isotopic fractions (Fontani et al., 2011; Fontani et al., 2014; Fontani et al., 2015a; Fontani et al., 2015b; Colzi et al., 2018a). They propose that species as CS, CCS, , and NS, trace preferentially quiescent (i.e., less evolved) gas, while species as OCS, and trace more turbulent (i.e., more evolved) material. They investigated how the molecular abundances vary as a function of the evolutionary indicators, in particular and the ratio, and found that the best positive correlations are found for SO and , and in general for oxygen-bearing species (see also Martinez et al., 2024), perhaps due to the larger availability of atomic oxygen with evolution produced by photodissociation of water (e.g., van Dishoeck et al., 2021). The total sulphur abundance in molecules is also found to increase with evolution (Figure 6), likely due to the increasing amount of S that is sputtered from dust grains owing to the increasing protostellar activity. In fact, the enhancement is significant especially from the earliest phase, classified in Fontani et al. 2023 as cold HMSC with kinetic temperatures K, to the later stages, and the total sulphur gaseous abundance is at most . A similar trend was found by Tang et al. (2024), who computed a total S-bearing molecular abundance attaining at most in the protostellar phase. However, these total sulphur molecular abundances are two orders of magnitude lower than the elemental one, confirming that sulphur is highly depleted from the gas phase even in the evolved stages.

FIGURE 6

Several molecular abundance ratios of sulphurated species have been investigated to find trends with evolution. One of the most promising candidates is probably the SO/ ratio (e.g., Herpin et al., 2009; Fontani et al., 2023; Martinez et al., 2024). This ratio was proposed to decrease with time in massive cores (Wakelam et al., 2011). The abundance can increase with respect to that of SO either via surface chemistry (by oxygenation of SO), or via the gas-phase barrierless reaction (Equation 3, Vastel et al., 2018):which is favoured in the late warm stages by the enhanced presence of OH in the gas. This decrease with evolution is confirmed in the single-dish observational surveys of Fontani et al. (2023) and Martinez et al. (2024), but not in Herpin et al. (2009). Other ratios proposed as possible evolutionary indicators involve CS, such as SO/CS, OCS/CS, and CS/S (Herpin et al., 2009; Li et al., 2015; el Akel et al., 2022; Fontani et al., 2023). In particular, el Akel et al. (2022) proposed the SO/CS ratio as a suitable tool to distinguish between warm (SO/CS) and cold (SO/CS) chemistry within the same source. However, we must bear in mind that in many of these works the sample size is small (e.g., Herpin et al., 2009; Fontani et al., 2023) and the dispersion of values is large, and therefore these proposed chemical clocks must be taken with caution. A particularly interesting case is the /OCS ratio, since and OCS are the only S-bearing species detected so far in interstellar ices (e.g., Palumbo et al., 1997; Boogert et al., 2022; McClure et al., 2023; Rocha et al., 2024). Both molecules are commonly detected in the gas, but while is a well-known outflow tracer and it is believed to have an important gas-phase formation route (Vastel et al., 2018,see above), OCS is more likely a surface chemistry product when CO ice is abundant on grains (e.g., Ferrante et al., 2008; el Akel et al., 2022) and is not commonly found in outflows (e.g., Drozdovskaya et al., 2018). In the gas, Herpin et al. (2009) found a /OCS ratio that increases with evolution. Santos et al. (2024) indicate that OCS is more strongly linked to OH, a pure surface chemistry product, than . This finding is in agreement with models (Vidal and Wakelam, 2018) that predict a reprocessing of upon desorption more efficient than for OCS, and an increase of the /OCS ratio at K on core timescales yrs. The negligible contribution of gas-phase processes in the formation of OCS is corroborated by gas-phase models, which are not able to reproduce observed OCS abundances (Loison et al., 2012).

Finally, Li et al. (2017) report an enhancement of c– and deuterated species such as D, , and DCN toward shocked regions, as suggested by the presence of SiO and/or SO emission, in their spectral line mapping observations towards four massive star-forming regions (Cepheus A, DR21S, S76E, and G34.26 + 0.15).

2.2.5 Masers

Although strictly speaking maser lines are not chemical tracers, they are excellent signposts of high-mass star forming regions (e.g., Ellingsen et al., 2010) and were proposed as evolutionary indicators because of their quick variations with the change in local physical properties. Therefore, without going into the details of the mechanisms responsible for the formation of interstellar masers (see, e.g., Elitzur et al., 1989), we mention some studies that highlighted the appearance of particular maser lines along the evolution of the high-mass star formation process. Masers of methanol, water, and OH are typically observed towards both the HMPO and UC H_II stages. Methanol is likely the species towards which the highest number of maser lines were detected. Menten (1991) suggested to classify them into class I and class II types based on their position: class I masers are thought to be offset from bright infrared sources and UC H_II regions, while class II masers are spatially associated with them in projection. For this reason, class I masers are believed to be collisionally pumped, while class II masers radiatively pumped. In both cases, they tend to appear very early because of the high column density of OH they need to be excited, but differences were highlighted between specific lines. A tentative evolutionary sequence was proposed by Breen et al. (2010), based on observations of the two class II methanol masers at 6.7 GHz and 12.2 GHz, towards dusty clumps with or without an embedded UC H_II region detectable in the radio continuum. They proposed that the less evolved clumps are associated only with the 6.7 GHz maser line, and then the 12.2 GHz line appears when the OH column density decreases significantly, that is after the formation of the UC H_II region. Several studies (Breen et al., 2010; Beltrán, 2018, e.g.,) proposed a coarse evolutionary sequence in which class I OH masers appear in the earliest protostellar phases, followed by class II OH masers and then O masers. Water masers would tend to appear later because they need significant outflows or winds to develop well. The latest stages are characterised by the appearance of OH masers (e.g., Garay and Lizano, 1999; Breen et al., 2018), a molecule believed to be abundantly produced via the photodissociation of water by UV photons. However, other studies suggest that there are no significant evolutionary differences between class I and II OH masers (e.g., Fontani et al., 2010), and that O masers tend to appear even before class II OH masers (Ladeyschikov et al., 2022; Yang et al., 2023). Therefore, it is very difficult to define an evolutionary sequence only with masers because the different types can be found almost in all evolutionary stages (e.g., Yang et al., 2020).

2.2.6 Summary, open questions, and future prospects

The results obtained so far from source surveys towards specific lines or species indicate that the clearest evolutionary indicators are deuterated fractions of a few molecules (e.g., N₂H⁺,). Tentative trends are proposed for some carbon-chain molecules, COMs, and sulphur-bearing species, as well as for some abundance ratios (e.g., N₂H⁺/, N₂H⁺/CCS, SO/), although with remarkable diversity among different surveys.

The main difficulty in studies of this kind is that source surveys are never perfectly homogeneous. Sources in similar evolutionary stages can have significantly different physical properties, making it difficult to disentangle the effect of individual parameters on the observed chemical abundances. Even when objects are coeval, variations in temperature, density, irradiation from cosmic rays and UV photons, and/or initial elemental abundances, can produce markedly different observational chemical signatures. Environmental effects, such as isolated versus crowded regions, can also affect the chemical composition especially in the external envelope of star-forming cores. Moreover, source surveys often contain objects at different distances, which implies that the observed emission may arise from significantly different linear scales. To solve these problems, observations of objects with both homogeneous heliocentric distances and very similar (and well-measured) physical parameters are critical to constrain the parameter space, and help isolating the effect of time on the observed molecular abundances.

The results summarised above concern solely the evolution of gaseous species. Therefore, a relevant open question that still needs to be addressed is whether the molecular composition of iced dust grain mantles also evolves with time. During the star-formation process, especially in the early cold phases, atoms and molecules freeze-out on dust grains forming ice mantles, in which surface chemistry contributes to the molecular complexity. Because of the continuous adsorption on and sublimation from the ice mantles, the mutual exchange between gaseous and iced species plays an essential role in regulating the evolution of molecular abundances. Therefore, studying both the composition of gas and ices is critical to have a complete picture of molecular formation processes, since many species can form via both gas-phase and surface chemistry processes (e.g., Caselli and Ceccarelli, 2012; Ceccarelli et al., 2023).

Ice mantles can be identified and studied through vibrational modes observed in absorption against near- and mid-infrared continuum background sources. The James Webb Space Telescope (JWST) is the best facility currently available to perform such comparative studies of gas and ices. Observations from JWST programs like Ice Age (McClure et al., 2023) and JOYS (van Dishoeck et al., 2025) have begun to give insights into the composition of the two phases. Both programs have identified a wide range of iced molecules including simple and complex organic molecules. However, none of these programs are focussed on high-mass objects, nor on their evolution. Future JWST surveys of high-mass cores in well-defined evolutionary stages will play a fundamental role in attacking this problem. In particular, sources with well-constrained physical parameters and gas-phase abundances of key species will be eminently suitable to unveil the role of evolution in shaping the molecular composition of ice grain mantles.

3 Theory

In this Section, we will review theoretical studies devoted to the chemistry in high-mass star-forming regions. Cold chemistry in HMSCs is followed by warm-up, and warm chemistry in HMPOs, and evolves into photon-dominated chemistry in H_II regions. Since timescales of dynamical evolution of high-mass protostars that also have complex morphology are short (Zinnecker and Yorke, 2007), and comparable to or even shorter than those of chemical evolution, chemistry appears to be linked to dynamics more tightly than in low-mass protostars. Although, as stated in Sect. 1, distinction between the evolutionary stages is not always clear, it is nevertheless possible to identify chemical species and processes related to the phases proposed according to Figure 1.

Chemical modelling is a powerful tool not only for exploring chemical abundances, but also for estimating key physical properties of star-forming regions, such as gas density, dust and gas temperatures, cosmic-ray ionisation rate, ionisation fractions, and dynamical evolution timescales. As can be seen, some studies employ a simplistic physical setup, but explore chemical processes in great detail. Other studies are focussed on construction of complex models that combine chemistry, dynamics, and radiative transfer. In such studies, chemistry often serves as a tool for investigating physical characteristics of massive star-forming regions.

3.1 Chemistry in HMSCs and IRDCs

High-mass starless cores are often embedded in infrared dark clouds (IRDCs). Those are characterised by low temperatures and high densities, and likely represent the earliest stages of development of high-mass protostars. Although physical conditions in HMSCs and IRDCs are mostly similar to that in low-mass prestellar cores, there are indications that some differences may exist, including somewhat higher temperature (15–20 K in IRDCs vs. 10 K in low-mass prestellar cores). Given the exponential dependence of grain surface-related chemical processes on temperature, such small temperature difference may impact observed chemical composition of IRDCs. This possibility was explored with chemical models.

Vasyunina et al. (2012) studied chemical evolution of infrared dark clouds (IRDCs) utilizing a pseudo-time-dependent, zero-dimensional chemical modelling approach. Their simulations considered three chemical networks: a purely gas-phase network, a network combining gas-phase reactions with accretion and desorption processes, and a complete gas-grain network incorporating detailed grain-surface chemistry. They specifically analysed two representative IRDCs: the colder IRDC013.90–1 (temperatures 15 K) and the warmer IRDC321.73–1 (temperature 25 K). The authors compared model-predicted abundances against observed abundances for several simple molecular species, including , N, HNC, , HCN, H, , and CS. Chemistry of those species is mainly governed by gas-phase chemical reactions. Interestingly, the complete gas-grain chemical network consistently demonstrated the best agreement with observational data. It was found that at the dust temperature range between 20 K and 30 K grain surface chemistry impacts abundances of certain simple species in the gas phase, such as . At these intermediate temperatures, grain-surface chemical reactions becomes notably efficient in removing gaseous CO. At dust temperature above 20 K, the CO molecules do not freeze-out efficiently. They rather involved in an accretion-desorption cycle due to collisions of CO molecules with interstellar grains. Vasyunina et al. (2012) showed that despite low residence time on grain surface, CO participate in surface reactions with OH and S, forming and OCS (prefix (gr) denotes reactions on grains):Those reactions (Equations 4, 5) effectively decrease the abundance of the gas-phase CO at the temperature range 20–30 K at which CO freeze-out is not efficient. This depletion of gas-phase CO significantly decreases its ability to destroy via the reaction (Equation 6):Consequently, the abundances of become notably elevated in environments within this temperature window, where gaseous that has similar binding energy to that of CO but is inert in surface chemistry, remains abundant, yet CO is efficiently depleted onto grains. The is formed through the reaction (Equation 7):The described chemical interactions illustrate how the temperature range of 20–30 K is crucial for distinguishing between various types of star-forming regions. Warm IRDCs in this temperature regime demonstrate noticeably different chemical behavior compared to colder low-mass starless cores, where extensive freeze-out dominates, and warmer massive protostellar objects, where higher temperatures keep CO primarily in the gas phase.

Vasyunina et al. (2014) investigated how complex organic molecules form in cold, dense Infrared Dark Clouds (IRDCs). They surveyed 43 IRDCs both in Northern and Southern hemispheres. Estimates of dust temperature in this set of IRDCs vary in a range of 15–25 K. Those temperatures were thought to be too cold for efficient COM formation, yet observations revealed species such as acetaldehyde (CHO), methyl formate (OCHO), and dimethyl ether (). Estimated abundances for CHO, OCHO, and in IRDCs were reported for the first time, typically ranging from to . COM abundances in IRDCs were found to be higher than in low-mass prestellar cores but lower than in evolved hot cores or high-mass protostellar objects (HMPOs). In the observed set of cores, the highest gas temperature was found in IRDC028.34 + 0.06: Sanhueza et al. (2012) estimated it as 30 K. This is the only core in the set where methyl formate (OCHO) was detected.

To understand how the COMs typically found in hot cores with temperatures higher than 100 K can be formed in much colder environments of IRDCs, using the astrochemical model, the authors tested several scenarios for IRDC028.34 + 0.06. They found that models with fixed temperatures (15–120 K) and density () could not reproduce observed COM abundances, underestimating them by 2–10 orders of magnitude. A sort of a warm-up scenario which was originally proposed by Garrod and Herbst (2006) for hot cores was more successful in reproducing the abundances of COMs in IRDC028.34 + 0.06. The model considers two stages. During the cold stage, chemistry develops during years at 10 K and constant gas density of . In the warm-up phase, the temperature of gas and dust in the model increases from 10 K to 30 K over years. This modest warm-up is sufficient to make radicals on grain surface mobile and to form COMs. In warm-up models for hot cores, temperature eventually increases to 200 K or higher resulting in complete grain mantle evaporation including newly formed COMs. In contrast, in the warm-up model applied to IRDC028.34 + 0.06, the icy grain mantle remains intact as final temperature is only 30 K. The surface-formed COMs are released to gas at the onset of their formation via reactive desorption. In Vasyunina et al. (2014), the assumed efficiency of reactive desorption for COMs is 1%. Although the assumption of efficient reactive desorption for complex molecules may be doubtful, COMs may be delivered to gas via other non-thermal desorption mechanisms. For example, Dartois et al. (2019) demonstrated that cosmic-ray-induced sputtering is an efficient non-thermal desorption mechanism for COMs such as methanol (OH) and methyl acetate (). Unlike VUV photodesorption, which is inefficient for large COMs due to photolysis, cosmic-ray sputtering effectively releases unfragmented molecules. The need for an evolutionary chemical model for the explanation of observed molecular content supported the claim that IRDCs are the early stages of massive star formation.

Entekhabi et al. (2022) carried out astrochemical modelling of the infrared dark cloud G28.37 + 00.07. The authors aimed to constrain the cosmic-ray ionisation rate (CRIR) and the chemical age in various regions of the IRDC. By combining observations of molecular lines such as O, C, , and with gas-grain astrochemical models, the authors constrain the CRIR and chemical age across ten positions within the cloud that sample various levels of star formation activity. The analysis finds typical conditions in the cloud characterised by densities nH-, temperatures around 10–15 K, and CO depletion factors between 3 and 10. Results indicate relatively low cosmic-ray ionisation rate ( to ). Notably, no systematic variation in CRIR was observed as a function of star-formation activity in different locations of the cloud. Chemical ages of the regions were generally estimated to be at least several free-fall times ( years), although modelling with additional species like HCN, HNC, HNCO, OH, and CO tended to suggest somewhat older ages and higher CRIRs.

A notable work by Sabatini et al. (2021) introduce a time-dependent modelling method to estimate evolutionary timescales in massive star formation, improving upon previous static or statistical approaches. They combine a simplified isothermal collapse phase with a detailed one-dimensional warm-up phase, incorporating radiative transfer simulations to accurately track temperature, density, and chemical evolution simultaneously. The model is applied to a representative sample of massive clumps from the ATLASGAL-TOP100 survey. Using chemical tracers such as CCH, CN, CO, and OH as chemical clocks, the study successfully reproduces observed molecular abundances, particularly for CCH and CN, though less accurately for CO and OH. The estimated total duration for massive star formation is approximately years, with the individual evolutionary phases clearly distinguished. The specific phases were found to last approximately years for the earliest (70-m weak), years for mid-IR weak, years for mid-IR bright, and years for the most evolved H_II-region phase. The novelty of this study is primarily in its combination of radiative transfer and chemical modelling in a time-dependent manner, moving beyond previous static density and temperature assumptions. The results indicate that molecular abundances significantly depend on the thermal evolution of the clumps, notably displaying clear evaporation fronts within the modelled clumps.

Deuterated molecules can serve as another powerful diagnostic tool for the earliest evolutionary stages of both low- and high-mass star forming regions. Kong et al. (2015) investigated how rapidly deuterium enrichment builds up in the dense, cold gas of star-forming cores by modelling the ratio ]/() with a new, expanded chemical network that keeps track of deuterium chemistry and the spin states of and isotopologues. Their network also contained reactions involving and its deuterated forms. Using this framework the authors carry out the first uniform exploration of how deuterium fractionation depends on four key physical controls–volume density, temperature, cosmic-ray ionisation rate and the gas-phase depletion factor of heavy elements–over ranges that cover both low-mass and massive starless cores. The simulations show that reaching a deuterium fraction of 0.1 – the level often observed in both low- and high-mass starless cores—requires at least several local free-fall times. Because these chemical timescales match the ambipolar-diffusion time rather than the dynamical free-fall time, the authors argue that significant magnetic support must be retarding collapse in highly deuterated cores. The deuterium fraction of 0.1 are favoured by densities above a few , temperatures below about 17 K, depletion factors above 6 and cosmic-ray ionisation rate below ;

Kong et al. (2016) performed similar study of two massive starless cores, C1-N and C1-S, in the IRDC G028.37 + 00.07. They used multi-line observations of and to measure deuterium fractionation and to perform chemodynamical modelling aimed at estimating the collapse rate of the cores. The authors find high deuterium fractions [0.2–0.7], far exceeding the cosmic [D]/[H] ratio, and use these results in conjunction with chemodynamical models to infer the cores’ evolutionary states. The models indicate that such high deuteration levels can only be achieved if the cores are collapsing slowly–at less than one-tenth the free-fall rate–implying near-virial equilibrium likely supported by magnetic fields.

3.2 Chemistry in HMPOs and hot cores

This stage of protostellar development is characterised by the richest chemical composition. In particular, high abundances of complex organic and prebiotic molecules are observed (Sect. 2). Here, we will review recent advances in modelling of hot core chemistry. The development during the last decades of models of hot core chemistry from a basic two-step “cold collapse – warm-up” approach into advanced tools for detailed understanding of chemical evolution that combines chemical models with radiative transfer, and results of radiation-magnetohydrodynamical simulations.

The pivotal role of gradual increase in dust temperature during the development of a low- and high-mass protostars on the formation of complex organic molecules (COMs) in grain-surface chemistry was first recognised in the seminal paper by Garrod and Herbst (2006). In hot cores, saturated organic molecules, such as methyl formate (), formic acid (HCOOH), and dimethyl ether () cannot be formed in observed abundances via gas-phase chemistry. Therefore, the authors employed a two-stage, time-dependent gas-grain chemical model that simulates both the collapse of a molecular cloud and the subsequent warm-up phase caused by protostellar heating. The novelty of this study lies in its detailed treatment of the intermediate warm-up period, during which the temperature of the gas and dust rises gradually from 10 K to more than 100 K over a time span of a hundred thousand years. In contrast to earlier models that focussed solely on cold collapse or hot gas-phase chemistry, this work emphasises the chemistry occurring during this transitional phase, particularly grain-surface reactions involving relatively immobile but reactive heavy radicals such as HCO, O, and OH. These radicals are produced via photolysis of water and methanol ices. Garrod and Herbst (2006) showed that the formation of methyl formate, formic acid, and dimethyl ether is significantly enhanced by these surface reactions at temperatures between 20 K and 40 K via radical-radical grain-surface reactions such as Equations 8–10:andoriginally proposed by Allen and Robinson (1977). At these temperatures, atomic hydrogen (H) evaporates from grain surfaces, freeing radicals such as OH and HCO from hydrogenation, which then react to produce complex organic molecules. Further warm-up beyond the sublimation temperatures of COMs, release them to the gas phase. It was also demonstrated that certain gas-phase routes previously considered inefficient become important at lower intermediate temperatures after evaporation of large amounts of reactants from grains, further enhancing molecular complexity.

Garrod and Herbst (2006) introduced an important concept relevant for both low-mass and high-mass star formation that links physical development of a protostar with its chemical complexity. The key role of radical-radical reactions and photolysis in icy mantles of dust grains in the formation of complex organic and prebiotic molecules was revealed. The warm-up model was further expanded on a large number of complex organic and prebiotic species in the following papers by Garrod et al. (2008) and Garrod (2013).

Choudhury et al. (2015) extended a simplistic physical models of first warm-up works into a self-consistent modelling framework to simulate the chemical and spectral evolution of hot molecular cores during high-mass star formation. The authors coupled a gas-grain chemical code Saptarsy with radiative transfer calculations done using RADMC-3D code to track the spatio-temporal evolution of key complex organic molecules (COMs) such as OH, OH, , and . They explore how variations in density structure, protostellar bolometric luminosity evolution, and cosmic-ray ionisation rate affect COM formation, desorption, and emission. Synthetic (sub-)mm spectra and emission maps are generated under LTE conditions to enable comparison with observations. Unlike previous studies that used static or simplified warm-up prescriptions, Choudhury et al. (2015) considered dynamic and spatially-resolved treatment of physical conditions, integrating time-dependent temperature structures derived from protostellar bolometric luminosity models with detailed gas-grain chemistry. It is also one of the first studies to simulate the spectral evolution of COMs in high-mass cores. Choudhury et al. (2015) reported modelled COM abundances (e.g., OH –, –) to be consistent with observed values, especially for higher cosmic-ray ionisation rate. Simulated spectra reproduce observed trends, including increasing line density and intensity with time, and predict that the spatial extent of emission from more weakly bound molecules (e.g., ) exceeds that of more strongly bound species like OH–offering a potential observational diagnostic of desorption energies.

Gieser et al. (2019) used the MUSCLE (MUlti Stage CLoud codE) framework, which combines static 1D physical structure of a modelled core with a time-dependent gas-grain chemical model ALCHEMIC (Semenov et al., 2010) to model AFGL 2591 VLA 3 hot core. By comparing modelled and observed column densities, the authors find that the abundances of 10 out of 14 species are reproduced at a chemical age of 21,100 years. While observations suggest a core density power-law index of 1.7, the model prefers a shallower gradient (1.4), indicating limitations of the 1D assumption. This study improves on earlier work by integrating high-resolution molecular data with detailed chemical modelling tailored to a specific source, offering a physically grounded estimate of chemical age and structure in a high-mass protostar.

Bonfand et al. (2019) combined a warm-up chemical model with a detailed radiative transfer calculations and the results of radiation-magnetohydrodynamic simulations, and applied it to four molecular cores (N2, N3, N4, N5) in Sagittarius B2(N) object close to the Galactic center. The ALMA data from the EMoCA spectral line survey (Belloche et al., 2016) has been used to benchmark models against observational data. The analysis reveals distinct chemical compositions among the sources: cores N3 and N5 share similar abundances relative to methanol, while the core N2 differs significantly, particularly in its higher abundance of ethyl cyanide (CN) and formamide (CHO). Chemical modelling with the MAGICKAL code (Garrod et al., 2008), coupled with radiative transfer calculations (RADMC-3D) and radiation-magnetohydrodynamic simulations, demonstrated that the production of complex organic molecules (COMs) is highly sensitive to environmental conditions. A cosmic-ray ionisation rate of 7 (50 times the standard interstellar value) best reproduced the observed abundances of ten COMs relative to methanol. This elevated rate accounts for the Galactic center environment while still allowing efficient COM formation. The models further constrained the minimum dust temperature during the prestellar phase in the Galactic center region to 25 K, with 15 K enabling efficient COM formation on grains through radical addition reactions. It was found that higher temperatures (25 K) suppress crucial grain-surface chemistry, particularly for cyanides and oxygen-bearing species like methyl formate (OCHO) and dimethyl ether (). The authors stress that the chemical differentiation between sources, especially the anomalous composition of N2, underscores the importance of individual physical histories. A combined hydrodynamic and gas-grain chemical model of a hot core is also presented in Barger et al. (2021). In that work, gas-grain chemical model is coupled with physical evolution of the core computed using a radiation hydrodynamics model. Interestingly, despite employing an advanced physical model of a hot core, the model exhibited somewhat limited agreement with observational data on Sgr B2(N2).

Barger and Garrod (2020) explored the influence of cosmic-ray ionisation rate (CRIR) and warm-up timescales on the chemical evolution of massive hot cores. Using the MAGICKAL code, the authors simulated a grid of 81 models varying in both CRIR and warm-up timescale, and incorporated LTE radiative transfer to produce simulated molecular emission spectra. These are then compared against observed data for four well-studied hot cores: NGC 6334 IRS 1, NGC 7538 IRS 1, W3 (O), and W33A. The results suggest that elevated cosmic-ray ionisation rate – order of magnitude higher than the canonical value of commonly used in previous models, as well as relatively rapid warm-up timescales ( years) are necessary to reproduce the observed molecular abundances. Cosmic rays drive radical production via photodissociation and thereby enable radical-radical reactions on grain surfaces, but also contribute to gas-phase destruction of COMs through ion–molecule reactions. Interestingly, best-fit CRIR is higher than that estimated by Entekhabi et al. (2022) for IRDCs (see Section 3.1).

As it can be seen, the development of models is headed towards creating self-consistent packages that combine chemistry with radiative transfer and hydrodynamics. However, models of chemistry themselves also undergone significant improvement, directed towards inclusion of new complex molecules, isomers and new mechanisms of solid-state reactivity. Several works related to the EMoCA line survey of Sgr B2(N2) hot core (Belloche et al., 2016) are the examples of such approach.

Following the first discovery of a branched molecule in the ISM, iso-propyl cyanide in Sgr B2(N) (Belloche et al., 2014), Garrod et al. (2017) used the three-phase chemical kinetics model MAGICKAL to simulate the formation of branched carbon-chain molecules in the hot-core source Sgr B2(N2). The updated chemical network explicitly treated radical isomerism and incorporated key grain-surface reactions, particularly the addition of the CN radical to unsaturated hydrocarbons with low activation energy barriers. The modeling identified the dominant formation mechanism for branched isomers. The simulations predict that branching becomes more pronounced with molecular size. The results highlight that efficient CN-addition channels enhance the production of straight-chain nitriles, suggesting that other molecular families without such low-barrier pathways may exhibit even greater degrees of branching. Willis et al. (2020) introduced a single-stage “collapse-warmup” chemical model of isocyanides formation in Sgr B2(N). A significantly expanded chemical network for the first time included new isocyanides like NC, HCCNC, NC and NC, with reactions often inferred from their cyanide analogs. A key finding is the critical importance of using an enhanced and extinction-dependent cosmic-ray ionization rate () for the reproducing of the NC:CCN ratio.

In a recent comprehensive paper by Garrod et al. (2022), a major revision of the MAGICKAL gas–grain chemical kinetics code aimed at explaining the origin of complex organic molecules (COMs) in hot molecular cores is presented. They extend the standard diffusive three-phase framework by embedding non-diffusive mechanisms of surface reactivity first introduced in Garrod and Pauly (2011) and later generalized in Jin and Garrod (2020) to Eley–Rideal encounters, photodissociation-induced reactions, three-body (follow-on) reactions and its excited-formation variant. In this way, radicals and stable species can react immediately after they come into contact on grain surfaces or inside the ice mantle without waiting for thermal diffusion. This is coupled to other upgrades, including revised treatment of diffusion and desorption barriers, new proton-transfer reactions with ammonia that prolong gas-phase COMs lifetimes, and an expanded network of reactions. Using a standard collapse followed by fast, medium and slow warm-up tracks, the authors show that nondiffusive chemistry moves the production of many COMs from the classical 20–50 K diffusive regime to much earlier times and to dust temperatures as low as 10 K. Significant fractions of molecules such as methyl formate, dimethyl ether, ethanol and even glycine are synthesised while the ice mantle is still growing, either as by-products of CO hydrogenation or by cold UV photolysis of nascent ices. Additional bursts of synthesis occur when water and other strongly bound ices desorb at 120–160 K, liberating trapped radicals that react on hot surfaces. Because ammonia efficiently neutralises protonated COMs in the gas, the modelled abundances remain high long after ice desorption. The model simultaneously reproduces the gas-phase abundances of dozens of COMs across both low- and high-mass protostars, including previously problematic methyl formate: glycolaldehyde: acetic acid ratio.

Although prebiotic importance is usually attributed to carbon-bearing complex organic molecules, phosphorus (P), which is synthesised in massive stars, is another crucial element for development of life on Earth. It is part of DNA and RNA molecules, phospholipids, and the adenosine triphosphate molecule. However, chemistry of phosphorus-bearing molecules in star-forming regions is not well understood. Nevertheless, several studies of P-bearing molecules were published during the last decade.

Fontani et al. (2016) presented detections of phosphorus nitride (PN) in a sample of massive dense cores at different stages of high-mass star formation, ranging from starless cores to protostellar objects and ultracompact H_II regions. Interestingly, for the first time PN was found in relatively cold (20–60 K) and quiescent environments, in contrast to previous studies that associated PN predominantly with hot (100 K) and turbulent regions influenced by shocks. The authors employed a time-dependent chemical model MONACO based on the work of Vasyunin and Herbst (2013b) to understand the formation of detected PN. The model comprises a two-phase evolutionary scenario: an initial cold phase (10 K, n () = ) that lasts 1 Myr, followed by a warm-up phase (final temperatures are T = 40, 50 or 60 K, n () = ) that also lasts 1 Myr. This framework was used to compute the abundance ratios of several species (e.g., CN/PN, OH/PN, /PN, and HNC/PN), which were then compared with observationally derived ratios. The models reproduced the observed abundance ratios for most species within the expected timescales and temperatures, except for CN/PN, which was likely affected by differing spatial distributions of the molecules. Thus, the possibility of formation of observed abundance of PN in a relatively cold quiescent gas via gas-phase chemistry was confirmed.

Rivilla et al. (2016) presented first detection of another key phosphorus-bearing prebiotic molecule, phosphorus monoxide (PO) in two massive star-forming regions, W51 e1/e2 and W3(OH). The observational results are supplemented by chemical modelling that clarifies the formation of PO alongside phosphorus nitride (PN). The modelling setup is similar to that utilised in Fontani et al. (2016) with the exception final temperature of the warm-up phase, which is equal to 200 K. The modelling indicated that both PO and PN shall be formed via gas-phase ion–molecule and neutral–neutral reactions already during the cold phase of protostellar development and freeze to grains. The observed PO and PN are likely released to the gas phase from grains at the onset of protostellar warm-up at 35 K. The observed abundance of PO to as well as PO/PN abundance ration of 1.8–3 are successfully reproduced by the model. Interestingly, a relatively high initial abundance of atomic phosphorus of , 25 times higher than the “low metal” P-abundance typically used in dark cloud chemical models is required to reproduce observations. In a later work by Rivilla et al. (2020), an alternative scenario of PO formation in shocked environments is proposed. The scenario implies the formation of PO in gas-phase photochemistry of phosphine (). The latter is previously formed on dust grains via hydrogenation of P atoms and then released to the gas due to shocks.

3.3 Chemistry in HC and UC H_II regions

These regions follow hot cores, and represent the final stage of the development of a massive protostar (Churchwell, 2002). At the same time, they have a strong impact on their surrounding molecular clouds via feedback mechanisms (Giannetti et al., 2012; Moscadelli et al., 2018), and thus in setting the initial conditions for the next-generation of forming stars. Complex physical and chemical evolution occurs there on small spatial scales which makes its study challenging.

Pilleri et al. (2013) investigated the spatial distribution and chemistry of small hydrocarbons, specifically CH, CCH, and c-, in the vicinity of the ultra-compact H_II region Monoceros R2 (Mon R2), with a focus on photo-dissociated regions (PDRs) exposed to high ultraviolet (UV) radiation fields. The modelling work combines extensive observational data from the IRAM 30 m telescope and the Herschel Space Observatory, which are interpreted using both steady-state gas-phase chemical models (via the Meudon PDR code, Le Petit et al. (2006)) and time-dependent gas-grain chemical models (via UCL_CHEM code, Viti et al. (2004)). The core modelling approach consists of dissecting the Mon R2 region into distinct physical components: a highly UV-irradiated thin PDR layer (LPDR), a denser, more molecular high-density PDR shell (LHD), and a surrounding low-density molecular envelope (Lenv). For the PDR layers, the Meudon PDR code is used to simulate steady-state gas-phase chemistry under intense UV irradiation (), while for the cooler, shielded envelope, UCL_CHEM is employed to capture grain surface processes and time-dependent effects, particularly relevant during cloud collapse and subsequent UV illumination. A key finding of the modelling is that gas-phase steady-state chemistry can reproduce the observed abundances of CH and CCH relatively well in the high-UV illuminated PDR, but it fails–by about an order of magnitude–to account for the c- abundance. The inclusion of grain-surface chemistry and non-equilibrium effects, modelled through UCL_CHEM, is essential to reproduce c- abundances in the lower-density envelope. The study shows that the N (c-)/N(CCH) ratio increases from 0.004 in the envelope to as high as 0.03 near the PDR peaks, and this ratio correlates spatially with 8m polycyclic aromatic hydrocarbon (PAH) emission, suggesting that considered hydrocarbons could be a product of photoprocessing of PAHs.

Stéphan et al. (2018) investigated hot cores with embedded hypercompact (HC)/ultracompact HII (UCHII) regions with a focus on the chemical structure and evolution of internal photo-dissociated regions (PDRs) that surrounds deeply embedded HC and UCHII regions. The aim was to identify specific chemical tracers that can differentiate these evolutionary phases of massive star formation from hot molecular cores (HMCs). The authors used the astrochemical code Saptarsy coupled with radiative transfer simulations (RADMC-3D) to compute spatio-temporal evolutions of chemical abundances and to derive synthetic spectra for different atomic and molecular species. It was found that common molecular tracers like C, , CN, and HCO are not reliable indicators for distinguishing between internal HII regions/PDRs and hot molecular cores because their distributions do not uniquely trace the inner core regions influenced by strong UV fields. On the other hand, atomic species such as and O robustly trace these internal PDRs, but unobservable with existing facilities due to the very narrow size of these internal PDRs (typically less than 100 AU thick).

An attempt to estimate duration of every phase of development of high-mass protostars using astrochemical modelling is made in Gerner et al. (2014). In this work, a comprehensive modelling effort to trace the chemical evolution of high-mass star-forming regions across four evolutionary stages: infrared dark clouds (IRDCs), high-mass protostellar objects (HMPOs), hot molecular cores (HMCs), and ultra-compact HII (UCHII) regions is presented. The MUSCLE framework was utilised to simulate the chemistry within each evolutionary phase. This approach integrates radial variations in density and temperature, and incorporates an advanced chemical network that includes gas-phase and grain-surface processes, as well as cosmic-ray and UV-induced effects. Using the IRAM 30 m telescope, the authors conducted a molecular line survey at 1 mm and 3 mm of 59 regions spanning all evolutionary stages of massive protostars, the authors fit observed molecular column densities with models by varying physical parameters such as central density, temperature, and chemical age. Each evolutionary stage is treated as a distinct model, with physical and chemical properties constrained independently. The iterative modelling yields chemical ages of approximately 10,000 years for the IRDC stage, 60,000 years for HMPOs, 40,000 years for HMCs, and 10,000 years for UCHII regions, aligning with theoretical predictions of a years of total duration for massive star formation.

3.4 Summary, open questions, and future prospects

Chemical evolution of high-mass star-forming regions exhibit many similarities to their low-mass counterparts. However, shorter timescales of dynamical evolution and stronger radiation fields imply that chemistry in high-mass protostars should be considered as tightly coupled to their dynamics. Chemical modeling is used not only to predict molecular abundances but also to constrain physical properties such as density, temperature, cosmic-ray ionization rates, and evolutionary timescales.

In HMSCs and IRDCs, chemical models reveal that even small temperature differences (e.g., 15–30 K) significantly affect molecular abundances. For instance, at 20–30 K, CO is efficiently depleted from the gas phase via grain-surface reactions, enhancing abundances. Complex organic molecules (COMs) are detected in cold IRDCs. While early models suggested that a modest warm-up phase and non-thermal desorption mechanisms–such as reactive desorption or cosmic-ray sputtering–are responsible for the formation and release of these molecules into the gas without full mantle evaporation, recent modelling highlight the importance of non-diffusive surface reactivity in the formation of COMs (e.g., Jin and Garrod, 2020; Garrod et al., 2022; Borshcheva et al., 2025). Deuterium fractionation, particularly the ]/() ratio, serves as a sensitive chemical clock, indicating slow collapse rates and significant magnetic support in prestellar cores.

For HMPOs and hot cores, chemical models have evolved from simple two-stage (collapse followed by warm-up) approaches to sophisticated frameworks combining gas-grain chemistry, radiative transfer, and radiation-magnetohydrodynamic simulations. The gradual warm-up (from 10 K to over 100 K) is critical for forming COMs via radical-radical reactions on grain surfaces. Recent models also incorporate non-diffusive reaction mechanisms, allowing COM formation at temperatures as low as 10 K, and highlight the importance of cosmic-ray-induced chemistry and gas-phase destruction pathways for complex species.

In hyper- and ultra-compact H_II regions, chemistry becomes dominated by UV irradiation, resembling photo-dissociated regions (PDRs). Models highlight the role of atomic lines and recombination lines, and the difficulty of tracing internal PDRs with common molecular tracers. Time-dependent models combining chemistry and radiative transfer provide estimates for the duration of each evolutionary phase, with total massive star formation timescales around years.

Overall, theoretical advances are moving toward more self-consistent frameworks that integrate chemistry, dynamics, and radiative transfer, while also incorporating new grain-surface reaction mechanisms–such as non-diffusive processes–that allow COM formation at colder temperatures and earlier evolutionary stages. Although analysis of ice composition in some high-mass star-forming regions is already published (see, e.g., Nakibov et al., 2025), chemical modelling of the evolution of icy mantles of interstellar grains in high-mass star forming regions is yet to be done comprehensively.

4 Comparison with other astrophysical environments

In this section, we briefly highlight the main differences and similarities among high-mass star-forming cores and other astrophysical environments from the point of view of the chemical evolution.

4.1 Low-mass star-forming cores

As stated in Sect. 1, the low-mass star formation process follows a theoretical scenario significantly different from that of high-mass stars: high-mass star-forming cores are characterised by larger temperatures and densities and evolve in much shorter timescales. These physical differences are expected to impact the evolution of molecular abundances, and chemical evolutionary studies that compare low- and high-mass star-forming cores are starting to highlight similarities and differences.

Probably, the chemical evolutionary indicator that most closely links low- and high-mass star-forming cores, both from the theoretical and observational point of view, is deuterium fractionation. The abundance of deuterated molecules, especially and , is found to be greatly enhanced in pre-stellar cores (Caselli et al., 2003; Crapsi et al., 2005; Pagani et al., 2007), and has a clear similar decreasing trend with increasing temperature, and hence protostellar evolution, as in the high-mass case, particularly in N₂H⁺ (Emprechtinger et al., 2009; Friesen et al., 2013; Ceccarelli et al., 2014a). As in the high-mass case, evolution does not seem to play a role in the fractionation of nitrogen, since the ¹⁴N/¹⁵N ratio does not vary significantly with time in low-mass star-forming cores (e.g., De Simone et al., 2018), except for a depletion of ¹⁵N in N₂H⁺ in pre-stellar cores (e.g., Bizzocchi et al., 2013; Redaelli et al., 2018), which causes ¹⁴N/¹⁵N to attain values , never measured so far in high-mass star-forming regions.

Another possible chemical link between low- and high-mass star formation is represented by sulphurated species, COMs, and carbon chains. Sulphur-bearing molecules are detected at any stage of the low-mass star formation process (e.g., Lefloch et al., 2018). Buckle and Fuller (2003) reported a tentative decrease of the SO/ abundance ratio from the class 0 to the class I protostellar stage, confirmed by Ghosh et al. (2024). This trend is similar to what predicted (Wakelam et al., 2011) and observed (e.g., Herpin et al., 2009; Martinez et al., 2024) in massive cores, although other works show different results. For example, studying a sample of class I protostars, Le Gal et al. (2020) found no anti-correlation between the SO/ abundance ratio and the disk-to-envelope mass ratio, expected to increase with evolution. A similar result was obtained with ALMA in a sample of class 0 and class I objects in Perseus (Artur de la Villarmois et al., 2023). The link between the low- and high-mass case is even more uncertain in the evolution of COMs. While in the high-mass star formation process all COMs keep increasing their abundance with evolution (Coletta et al., 2020), in low-mass cores the situation is different. It is well known that warm regions ( K) around low-mass protostars are enriched by COMs. These compact cores, called hot corinos (e.g., Ceccarelli et al., 2007), have physical and chemical properties similar to HMCs and are detected both in early (class 0) and evolved (class I) low-mass protostars. Some observational studies propose an abundance peak of COMs in the class 0 phase followed by a decrease in the class I phase (e.g., Bhat et al., 2023), due to the massive release in the gas of molecules produced on ice mantles. However, similar studies do not highlight clear evolutionary trends, and show that the abundance of COMs do not substantially evolve from the class 0 to the class I protostellar phase (e.g., Mercimek et al., 2022; Ceccarelli et al., 2023). Similarly to COMs, observations of carbon chains toward low-mass young stellar objects revealed that these species are as commonly detected as in high-mass protostars (Taniguchi et al., 2024). However, their formation around low-mass objects would be favoured by the processes know as ’warm carbon chain chemistry’ (WCCC, Sakai and Yamamoto, 2013), a chemistry initiated by the evaporation of methane from icy grain mantles, but observations of cyanopolyynes towards high-mass star-forming cores indicate that their chemistry occurs differently and requires much higher temperatures (Taniguchi et al., 2021).

The chemical complexity in low-mass star-forming cores appears to be more affected by local environmental conditions than by evolution (e.g., van Gelder et al., 2020; Scibelli et al., 2024). In particular, it is now clear that energetic phenomena associated with high-mass star, such as high UV irradiation and enhanced cosmic-ray ionisation rates, have a significant impact on the chemistry of some species, like carbon chains and hydrocarbons. This brings us to the link between the chemical evolution in high-mass star-forming cores and the chemistry inherited by protostellar envelopes forming in high-mass clusters, including the protosolar nebula (Sect. 1). The prototypical source where this effect is apparent is OMC-2 FIR4, an intermediate-mass protocluster in Orion north to the Trapezium OB star cluster (Chini et al., 1997). The comparison of the high abundances of the N and N cyanopolyynes (Fontani et al., 2017) and of c- (Favre et al., 2018) with the predictions of astrochemical models indicates that OMC-2 FIR4 is irradiated by a Far-UV field times larger than the interstellar one, and associated with a cosmic-ray ionisation rate three orders of magnitude larger than the canonical interstellar value (). The dose of energetic particles responsible for this observed high ionisation rate, causing the abundance enhancements in cyanopolyynes and hydrocarbons, would be comparable to that experienced by the young Solar system (Ceccarelli et al., 2014b). Therefore, it is tempting to speculate that such high energetic phenomena may have had an important impact in the budget of pre-biotic material inherited from the protosolar nebula. Indeed, cyanopolyynes were detected in on Titan’s atmosphere (e.g., Vuitton et al., 2007) and comets (e.g., Mumma and Charnley, 2011), the continuous rain of which may have enriched the primitive Earth of carbon usable for synthesising biological material.

4.2 Sub-solar metallicity and extragalactic environments

Interferometric observations are now able to resolve individual high-mass star-forming regions both in the outer periphery of the Milky Way and in nearby galaxies. Both environments are characterised by metallicities lower than Solar. Understanding chemical evolution in star formation at sub-Solar metallicity can give insights into chemical processes at work in the early Galaxy or in high-redshift galaxies, where the gas was not enriched in metals yet by stellar nucleosynthesis. Although evolutionary studies have not been performed either in outer Galaxy or in extragalactic high-mass star forming regions, observational works have highlighted an interesting chemical similarity with regions in the immediate vicinity of the Sun.

In the outer Galaxy the expected scenario would be that the formation and survival of molecules is less efficient than in regions close to the Sun, owing to both a lower abundance of metals and less shielding from UV radiation. However, a hot core, WB89-789, associated with emission of COMs (such as OH, CCH, OCHO, and ), was detected at kpc from the Galactic centre (Shimonishi et al., 2023), where metallicity is estimated to be a factor of 4 lower than the Solar value according to the Galactic oxygen abundance gradient (e.g., Arellano-Córdova et al., 2021). Moreover, several studies indicate that the environmental metallicity does not affect the formation efficiency of most of the molecules studied so far, but it appears to act mostly as a scaling factor for the molecular abundances relative to the elemental ones (Bernal et al., 2021; Fontani et al., 2022; Gigli et al., 2025). However, the targets observed in the mentioned works all harbour protostellar massive objects whose relative evolutionary stages have not been determined precisely yet. Significant progresses in this direction can be made in the future by observing the centimetre-continuum emission at high-angular resolution to distinguish between HMPOs, HCHIIs and UCHIIs. In this vein, the unprecedented sensitivity observations that can be performed in the near future with the Square Kilometre Array (SKA) and the next-generation Very Large Array (ngVLA) will be of paramount importance.

In line with these findings, there are now several examples of extragalactic hot cores in the magellanic clouds: (Sewiło et al., 2018; 2022; Shimonishi et al., 2016; 2023). Considering that the Large Magellanic Cloud (LMC) and Small Magellanic Cloud (SMC) have metallicities smaller than Solar (by a factor 0.5 and 0.2 , respectively), these findings indicate that the formation of HMCs are common phases of the high-mass star formation process also in low metallicity environments. Interestingly, these extragalactic HMCs show molecular abundances that in some species, like , are consistent among them considering only a metallicity scaling factor, while others, like OH, are not consistent. Such difference suggests that the abundance of OH, and in general organic molecules (Shimonishi et al., 2023), is influenced less by metallicity than by other physical parameters (e.g., visual extinction, cosmic-ray ionisation rate, UV illumination, etc.).

Moving further away from the Milky Way, Sutter et al. (2024) have observed with JWST a sample of 19 nearby galaxies from the Physics at High Angular Resolution in Nearby GalaxieS (PHANGS) survey, focussing on the presence of PAHs on 10–50 parsec-scale clouds. The PAH fraction steeply decreases in H_II regions, revealing the destruction of these small grains in regions of ionised gas. Outside H_II regions, the PAHs abundance is constant across the PHANGS sample.

5 Summary and conclusions

A tentative chemical evolutionary sequence is illustrated in Figure 7: Simple abundant molecules and deuterated molecules are the best tracers of the early HMSC phase, while COMs and Si-/S-/P-bearing molecules characterise the HMPO and UCHII stages. The main chemical difference between HMPOs and UCHIIs is an enhanced abundance in UCHIIs of simple carbon-chains and rare ions. UCHIIs are also characterised by the presence of bright RRLs.

FIGURE 7

However, from an observational point of view one of the key takeaway messages from this review is that identifying good evolutionary indicators remains a significant challenge. A major issue arises from the fact that it is still not easy to understand if objects with similar observational properties are in different evolutionary stages. In this respect, to establish the elusive presence/absence of embedded indicators of a specific phase is critical. For example, an embedded centimetre continuum source can distinguish between a HMPO and a young H_II region with a similar molecular envelope. Similarly, the presence of an embedded young outflow can distinguish between a HMSC and an early HMPO in which the very embedded protostellar object has not significantly affected the collapsing envelope yet. Another major observational problem mirroring the previous one lies on the fact that source surveys often contain objects almost coeval but with different physical properties (in particular density, temperature, or external irradiation), making it difficult to isolate the role of time in determining the molecular abundances.

Next-generation radio and millimetre interferometers (SKA, ngVLA, the new ALMA receivers) are likely the best facilities to solve these problems, because they can unveil essential source properties, such as faint embedded centimetre continuum sources and/or young outflows, thanks to their unprecedented sensitivity. They can also perform source surveys of many sources and derive their detailed physical structure in reasonable amounts of observation time.

Finally, a relevant open question is how the chemical composition of ice mantles evolves with time in high-mass star forming regions, and how it is linked to the gas-phase composition and evolution. JWST observations of ices towards high-mass star-forming cores in well-established evolutionary stages and with well-constrained gas-phase abundances, combined with models specifically tailored for such environments, will be crucial to shed light on this essential and fascinating open question.

References

• 1

  AdamsF. C. (2010). The birth environment of the solar system. ARA&A48, 47–85. 10.1146/annurev-astro-081309-130830

  • CrossRef
  • Google Scholar

• 2

  AllenM.RobinsonG. W. (1977). The molecular composition of dense interstellar clouds. ApJ212, 396–415. 10.1086/155059

  • CrossRef
  • Google Scholar

• 3

  AndersonL. D.MakaiZ.LuisiM.AndersenM.RusseilD.SamalM. R.et al (2019). The origin of [C II] 158 μm emission toward the H II Region Complex S235. ApJ882, 11. 10.3847/1538-4357/ab1c59

  • CrossRef
  • Google Scholar

• 4

  Arellano-CórdovaK. Z.EstebanC.García-RojasJ.Méndez-DelgadoJ. E. (2021). On the radial abundance gradients of nitrogen and oxygen in the inner Galactic disc. MNRAS502, 225–241. 10.1093/mnras/staa3903

  • CrossRef
  • Google Scholar

• 5

  Artur de la VillarmoisE.GuzmánV. V.YangY. L.ZhangY.SakaiN. (2023). The Perseus ALMA Chemical Survey (PEACHES). III. Sulfur-bearing species tracing accretion and ejection processes in young protostars. A&A678, A124. 10.1051/0004-6361/202346728

  • CrossRef
  • Google Scholar

• 6

  BachillerR.Pérez GutiérrezM. (1997). Shock chemistry in the young bipolar outflow L1157. ApJL487, L93–L96. 10.1086/310877

  • CrossRef
  • Google Scholar

• 7

  BalucaniN.CeccarelliC.TaquetV. (2015). Formation of complex organic molecules in cold objects: the role of gas-phase reactions. MNRAS449, L16–L20. 10.1093/mnrasl/slv009

  • CrossRef
  • Google Scholar

• 8

  BargerC. J.GarrodR. T. (2020). Constraining cosmic-ray ionization rates and chemical timescales in massive hot cores. ApJ888, 38. 10.3847/1538-4357/ab5b0d

  • CrossRef
  • Google Scholar

• 9

  BargerC. J.LamK. H.LiZ.-Y.DavisS. W.HerbstE.GarrodR. T. (2021). Combined hydrodynamic and gas-grain chemical modeling of hot cores. I. One-dimensional simulations. A&A651, A43. 10.1051/0004-6361/202039226

  • CrossRef
  • Google Scholar

• 10

  BeakliniP. P. B.MendozaE.CaneloC. M.AlemanI.MerelloM.KongS.et al (2020). Sulphur-bearing and complex organic molecules in an infrared cold core. MNRAS491, 427–439. 10.1093/mnras/stz3024

  • CrossRef
  • Google Scholar

• 11

  BellM. B.AveryL. W.WatsonJ. K. G. (1993). A Spectral-Line Survey of W51 from 17.6 to 22.0 GHz. ApJS86, 211. 10.1086/191776

  • CrossRef
  • Google Scholar

• 12

  BellocheA.MentenK. M.ComitoC.MüllerH. S. P.SchilkeP.OttJ.et al (2008). Detection of amino acetonitrile in Sgr B2(N). A&A482, 179–196. 10.1051/0004-6361:20079203

  • CrossRef
  • Google Scholar

• 13

  BellocheA.GarrodR. T.MüllerH. S. P.MentenK. M.ComitoC.SchilkeP. (2009). Increased complexity in interstellar chemistry: detection and chemical modeling of ethyl formate and n-propyl cyanide in Sagittarius B2(N). A&A499, 215–232. 10.1051/0004-6361/200811550

  • CrossRef
  • Google Scholar

• 14

  BellocheA.MüllerH. S. P.MentenK. M.SchilkeP.ComitoC. (2013). Complex organic molecules in the interstellar medium: IRAM 30 m line survey of sagittarius B2(N) and (M). A&A559, A47. 10.1051/0004-6361/201321096

  • CrossRef
  • Google Scholar

• 15

  BellocheA.GarrodR. T.MüllerH. S. P.MentenK. M. (2014). Detection of a branched alkyl molecule in the interstellar medium: iso-propyl cyanide. Science345, 1584–1587. 10.1126/science.1256678

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 16

  BellocheA.MüllerH. S. P.GarrodR. T.MentenK. M. (2016). Exploring molecular complexity with ALMA (EMoCA): deuterated complex organic molecules in sagittarius B2(N2). A&A587, A91. 10.1051/0004-6361/201527268

  • CrossRef
  • Google Scholar

• 17

  BellocheA.GarrodR. T.MüllerH. S. P.MentenK. M.MedvedevI.ThomasJ.et al (2019). Re-exploring molecular complexity with ALMA (ReMoCA): interstellar detection of urea. A&A628, A10. 10.1051/0004-6361/201935428

  • CrossRef
  • Google Scholar

• 18

  BeltránM. T. (2018). “Perspectives on star formation: the formation of high-mass stars,”. Astrophysical masers: unlocking the mysteries of the universe (IAU Symposium). Editors TarchiA.ReidM. J.CastangiaP. (Cambridge: Cambridge University Press), 336, 193–200. 10.1017/S1743921317010742

  • CrossRef
  • Google Scholar

• 19

  BeltránM. T.CesaroniR.NeriR.CodellaC.FuruyaR. S.TestiL.et al (2005). A detailed study of the rotating toroids in G31.41+0.31 and G24.78+0.08. A&A435, 901–925. 10.1051/0004-6361:20042381

  • CrossRef
  • Google Scholar

• 20

  BeltránM. T.CodellaC.VitiS.NeriR.CesaroniR. (2009). First detection of glycolaldehyde outside the galactic center. ApJL690, L93–L96. 10.1088/0004-637X/690/2/L93

  • CrossRef
  • Google Scholar

• 21

  BeltránM. T.OlmiL.CesaroniR.SchisanoE.EliaD.MolinariS.et al (2013). A Hi-GAL study of the high-mass star-forming region G29.96-0.02. A&A552, A123. 10.1051/0004-6361/201321086

  • CrossRef
  • Google Scholar

• 22

  BeltránM. T.CesaroniR.RivillaV. M.Sánchez-MongeÁ.MoscadelliL.AhmadiA.et al (2018). Accelerating infall and rotational spin-up in the hot molecular core G31.41+0.31. A&A615, A141. 10.1051/0004-6361/201832811

  • CrossRef
  • Google Scholar

• 23

  BerginE. A.TafallaM. (2007). Cold dark clouds: the initial conditions for star formation. ARA&A45, 339–396. 10.1146/annurev.astro.45.071206.100404

  • CrossRef
  • Google Scholar

• 24

  BerginE. A.PhillipsT. G.ComitoC.CrockettN. R.LisD. C.SchilkeP.et al (2010). Herschel observations of EXtra-Ordinary sources (HEXOS): the present and future of spectral surveys with Herschel/HIFI. A&A521, L20. 10.1051/0004-6361/201015071

  • CrossRef
  • Google Scholar

• 25

  BernalJ. J.SephusC. D.ZiurysL. M. (2021). Methanol at the edge of the galaxy: new observations to constrain the galactic habitable zone. ApJ922, 106. 10.3847/1538-4357/ac27a6

  • CrossRef
  • Google Scholar

• 26

  BeutherH. (2007). “Physics and chemistry of hot molecular cores,”. Triggered star formation in a turbulent ISM (IAU Symposium). Editors ElmegreenB. G.PalousJ. (Cambridge: Cambridge University Press), 237, 148–154. 10.1017/S1743921307001378

  • CrossRef
  • Google Scholar

• 27

  BeutherH.ThorwirthS.ZhangQ.HunterT. R.MegeathS. T.WalshA. J.et al (2005a). High spatial resolution observations of NH₃ and CH₃OH toward the massive twin cores NGC 6334I and NGC 6334I(N). ApJ627, 834–844. 10.1086/430735

  • CrossRef
  • Google Scholar

• 28

  BeutherH.ZhangQ.GreenhillL. J.ReidM. J.WilnerD.KetoE.et al (2005b). Line imaging of orion KL at 865 μm with the submillimeter array. ApJ632, 355–370. 10.1086/432974

  • CrossRef
  • Google Scholar

• 29

  BeyerM.HölschN.HusselsJ.ChengC.-F.SalumbidesE. J.EikemaK. S. E.et al (2019). Determination of the Interval between the Ground States of Para- and Ortho-H_{2}, Phys. Rev. . Interval Between Ground States Para- Ortho-H₂. 123, 163002. 10.1103/PhysRevLett.123.163002

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 30

  BhatB.KarR.MondalS. K.GhoshR.GoraiP.ShimonishiT.et al (2023). Chemical evolution of some selected complex organic molecules in low-mass star-forming regions. ApJ958, 111. 10.3847/1538-4357/acfc4d

  • CrossRef
  • Google Scholar

• 31

  BizzocchiL.CaselliP.LeonardoE.DoreL. (2013). Detection of ¹⁵ NNH⁺ in L1544: Non-LTE modelling of dyazenilium hyperfine line emission and accurate ¹⁴N/¹⁵N values. A&A555, A109. 10.1051/0004-6361/201321276

  • CrossRef
  • Google Scholar

• 32

  BlakeG. A.SuttonE. C.MassonC. R.PhillipsT. G. (1986). The rotational emission-line spectrum of orion A between 247 and 263 GHz. ApJS60, 357. 10.1086/191090

  • CrossRef
  • Google Scholar

• 33

  BonfandM.BellocheA.GarrodR. T.MentenK. M.WillisE.StéphanG.et al (2019). The complex chemistry of hot cores in sgr B2(N): influence of cosmic-ray ionization and thermal history. A&A628, A27. 10.1051/0004-6361/201935523

  • CrossRef
  • Google Scholar

• 34

  BonnellI. A.BateM. R.ClarkeC. J.PringleJ. E. (2001). Competitive accretion in embedded stellar clusters. MNRAS323, 785–794. 10.1046/j.1365-8711.2001.04270.x

  • CrossRef
  • Google Scholar

• 35

  BoogertA. C. A.GerakinesP. A.WhittetD. C. B. (2015). Observations of the icy universe. ARA&A53, 541–581. 10.1146/annurev-astro-082214-122348

  • CrossRef
  • Google Scholar

• 36

  BoogertA. C. A.BrewerK.BrittainA.EmersonK. S. (2022). Survey of ices toward massive young stellar objects. I. OCS, CO, OCN⁻, and CH₃OH. ApJ941, 32. 10.3847/1538-4357/ac9b4a

  • CrossRef
  • Google Scholar

• 37

  BorshchevaK.FedoseevG.PunanovaA. F.CaselliP.Jiménez-SerraI.VasyuninA. I. (2025). Formation of complex organic molecules in prestellar cores: the role of nondiffusive grain chemistry. ApJ990, 163. 10.3847/1538-4357/adea73

  • CrossRef
  • Google Scholar

• 38

  BouscasseL.CsengeriT.BellocheA.WyrowskiF.BontempsS.GüstenR.et al (2022). Sulphur-rich cold gas around the hot core precursor G328.2551-0.5321. An APEX unbiased spectral survey of the 2 mm, 1.2 mm, and 0.8 mm atmospheric windows. A&A662, A32. 10.1051/0004-6361/202140519

  • CrossRef
  • Google Scholar

• 39

  BreenS. L.EllingsenS. P.CaswellJ. L.LewisB. E. (2010). 12.2-GHz methanol masers towards 1.2-mm dust clumps: quantifying high-mass star formation evolutionary schemes. MNRAS401, 2219–2244. 10.1111/j.1365-2966.2009.15831.x

  • CrossRef
  • Google Scholar

• 40

  BreenS. L.ContrerasY.EllingsenS. P.GreenJ. A.WalshA. J.AvisonA.et al (2018). The 6-GHz multibeam maser survey - III. Comparison between the MMB and HOPS. MNRAS474, 3898–3911. 10.1093/mnras/stx3051

  • CrossRef
  • Google Scholar

• 41

  BuckleJ. V.FullerG. A. (2003). Sulphur-bearing species as chemical clocks for low mass protostars?A&A399, 567–581. 10.1051/0004-6361:20021816

  • CrossRef
  • Google Scholar

• 42

  BusquetG.PalauA.EstalellaR.GirartJ. M.Sánchez-MongeÁ.VitiS.et al (2010). The NH₂D/NH₃ ratio toward pre-protostellar cores around the UC H II region in IRAS 20293+3952. A&A517, L6. 10.1051/0004-6361/201014866

  • CrossRef
  • Google Scholar

• 43

  CarpenterJ. M. (2000). 2MASS observations of the perseus, orion A, orion B, and monoceros R2 molecular clouds. AJ120, 3139–3161. 10.1086/316845

  • CrossRef
  • Google Scholar

• 44

  CaselliP.CeccarelliC. (2012). Our astrochemical heritage. A&A Rev.20, 56. 10.1007/s00159-012-0056-x

  • CrossRef
  • Google Scholar

• 45

  CaselliP.van der TakF. F. S.CeccarelliC.BacmannA. (2003). Abundant H₂D⁺ in the pre-stellar core L1544. A&A403, L37–L41. 10.1051/0004-6361:20030526

  • CrossRef
  • Google Scholar

• 46

  CeccarelliC.CaselliP.HerbstE.TielensA. G. G. M.CauxE. (2007). “Extreme deuteration and hot corinos: the earliest chemical signatures of low-mass star formation,”. Protostars and planets V. Editors ReipurthB.JewittD.KeilK., 47. 10.48550/arXiv.astro-ph/0603018

  • CrossRef
  • Google Scholar

• 47

  CeccarelliC.CaselliP.Bockelée-MorvanD.MousisO.PizzarelloS.RobertF.et al (2014a). “Deuterium fractionation: the ariadne’s thread from the precollapse phase to meteorites and comets today,” in Protostars and planets VI. Editors BeutherH.KlessenR. S.DullemondC. P.HenningT.859–882. 10.2458/azu_uapress_9780816531240-ch037

  • CrossRef
  • Google Scholar

• 48

  CeccarelliC.DominikC.López-SepulcreA.KamaM.PadovaniM.CauxE.et al (2014b). Herschel finds evidence for stellar wind particles in a protostellar envelope: is this what happened to the young sun?ApJL790, L1. 10.1088/2041-8205/790/1/L1

  • CrossRef
  • Google Scholar

• 49

  CeccarelliC.CodellaC.BalucaniN.Bockelee-MorvanD.HerbstE.VastelC.et al (2023). “Organic chemistry in the first phases of solar-type protostars,”. Protostars and planets VII. Editors InutsukaS.AikawaY.MutoT.TomidaK.TamuraM.San Francisco, CA: Astronomical Society of the Pacific, 534, 379.

  • Google Scholar

• 50

  CernicharoJ.AgúndezM.KaiserR. I.CabezasC.TerceroB.MarcelinoN.et al (2021). Discovery of benzyne, o-C₆H₄, in TMC-1 with the QUIJOTE line survey. A&A652, L9. 10.1051/0004-6361/202141660

  • CrossRef
  • Google Scholar

• 51

  CesaroniR.HofnerP.ArayaE.KurtzS. (2010). The structure of hot molecular cores over 1000 AU. A&A509, A50. 10.1051/0004-6361/200912877

  • CrossRef
  • Google Scholar

• 52

  ChenY.van GelderM. L.NazariP.BroganC. L.van DishoeckE. F.LinnartzH.et al (2023). CoCCoA: Complex chemistry in hot cores with ALMA. Selected oxygen-bearing species. A&A678, A137. 10.1051/0004-6361/202346491

  • CrossRef
  • Google Scholar

• 53

  ChenY.RochaW. R. M.van DishoeckE. F.van GelderM. L.NazariP.SlavicinskaK.et al (2024). JOYS+: the link between the ice and gas of complex organic molecules: comparing JWST and ALMA data of two low-mass protostars. A&A690, A205. 10.1051/0004-6361/202450706

  • CrossRef
  • Google Scholar

• 54

  ChenJ. L.ZhangJ. S.GeJ. X.WangY. X.YuH. Z.ZouY. P.et al (2025a). The chemical clock of high-mass star-forming regions: N2H+/CCS. Astron. J.170, 74. 10.3847/1538-3881/addf35

  • CrossRef
  • Google Scholar

• 55

  ChenY.GarrodR. T.RachidM.van DishoeckE. F.BroganC. L.LoomisR.et al (2025b). CoCCoA: Complex chemistry in hot cores with ALMA: the chemical evolution of acetone from ice to gas. A&A696, A198. 10.1051/0004-6361/202453389

  • CrossRef
  • Google Scholar

• 56

  ChiniR.ReipurthB.Ward-ThompsonD.BallyJ.NymanL.-Å.SieversA.et al (1997). Dust filaments and star formation in OMC-2 and OMC-3. ApJL474, L135–L138. 10.1086/310436

  • CrossRef
  • Google Scholar

• 57

  ChoudhuryR.SchilkeP.StéphanG.BerginE.MöllerT.SchmiedekeA.et al (2015). Evolution of complex organic molecules in hot molecular cores. Synthetic spectra at (sub-)mm wavebands. A&A575, A68. 10.1051/0004-6361/201424499

  • CrossRef
  • Google Scholar

• 58

  ChurchwellE. (2002). Ultra-compact HII regions and massive star formation. ARA&A40, 27–62. 10.1146/annurev.astro.40.060401.093845

  • CrossRef
  • Google Scholar

• 59

  CodellaC.CeccarelliC.BianchiE.BalucaniN.PodioL.CaselliP.et al (2020). Seeds of life in space (SOLIS). V. Methanol and acetaldehyde in the protostellar jet-driven shocks L1157-B0 and B1. A&A635, A17. 10.1051/0004-6361/201936725

  • CrossRef
  • Google Scholar

• 60

  ColettaA.FontaniF.RivillaV. M.MininniC.ColziL.Sánchez-MongeÁ.et al (2020). Evolutionary study of complex organic molecules in high-mass star-forming regions. A&A641, A54. 10.1051/0004-6361/202038212

  • CrossRef
  • Google Scholar

• 61

  ColziL. (2020). Isotopic fractionation study towards massive star-forming regions across the galaxy. 10.36253/978-88-5518-380-2

  • CrossRef
  • Google Scholar

• 62

  ColziL.FontaniF.CaselliP.CeccarelliC.Hily-BlantP.BizzocchiL. (2018a). Nitrogen and hydrogen fractionation in high-mass star-forming cores from observations of HCN and HNC. A&A609, A129. 10.1051/0004-6361/201730576

  • CrossRef
  • Google Scholar

• 63

  ColziL.FontaniF.RivillaV. M.Sánchez-MongeA.TestiL.BeltránM. T.et al (2018b). Nitrogen fractionation in high-mass star-forming cores across the galaxy. MNRAS478, 3693–3720. 10.1093/mnras/sty1027

  • CrossRef
  • Google Scholar

• 64

  ColziL.RivillaV. M.BeltránM. T.Jiménez-SerraI.MininniC.MelossoM.et al (2021). The GUAPOS project. II. A comprehensive study of peptide-like bond molecules. A&A653, A129. 10.1051/0004-6361/202141573

  • CrossRef
  • Google Scholar

• 65

  ColziL.RomanoD.FontaniF.RivillaV. M.BizzocchiL.BeltranM. T.et al (2022). CHEMOUT: CHEMical complexity in star-forming regions of the OUTer galaxy. III. Nitrogen isotopic ratios in the outer galaxy. A&A667, A151. 10.1051/0004-6361/202244631

  • CrossRef
  • Google Scholar

• 66

  ComitoC.SchilkeP.PhillipsT. G.LisD. C.MotteF.MehringerD. (2005). A molecular line survey of orion KL in the 350 micron band. ApJS156, 127–167. 10.1086/425996

  • CrossRef
  • Google Scholar

• 67

  CorbyJ. F.JonesP. A.CunninghamM. R.MentenK. M.BellocheA.SchwabF. R.et al (2015). An ATCA survey of sagittarius B2 at 7 mm: chemical complexity meets broad-band interferometry. MNRAS452, 3969–3993. 10.1093/mnras/stv1494

  • CrossRef
  • Google Scholar

• 68

  CrapsiA.CaselliP.WalmsleyC. M.MyersP. C.TafallaM.LeeC. W.et al (2005). Probing the evolutionary status of starless cores through N₂H⁺ and N₂D⁺ observations. ApJ619, 379–406. 10.1086/426472

  • CrossRef
  • Google Scholar

• 69

  CrockettN. R.BerginE. A.WangS.LisD. C.BellT. A.BlakeG. A.et al (2010). Herschel observations of EXtra-Ordinary sources (HEXOS): the terahertz spectrum of orion KL seen at high spectral resolution. A&A521, L21. 10.1051/0004-6361/201015116

  • CrossRef
  • Google Scholar

• 70

  CsengeriT.BontempsS.WyrowskiF.MegeathS. T.MotteF.SannaA.et al (2017). The ATLASGAL survey: the sample of young massive cluster progenitors. A&A601, A60. 10.1051/0004-6361/201628254

  • CrossRef
  • Google Scholar

• 71

  CsengeriT.BontempsS.WyrowskiF.BellocheA.MentenK. M.LeuriniS.et al (2018). Search for high-mass protostars with ALMA revealed up to kilo-parsec scales (SPARKS). I. Indication for a centrifugal barrier in the environment of a single high-mass envelope. A&A617, A89. 10.1051/0004-6361/201832753

  • CrossRef
  • Google Scholar

• 72

  CsengeriT.BellocheA.BontempsS.WyrowskiF.MentenK. M.BouscasseL. (2019). Search for high-mass protostars with ALMA revealed up to kilo-parsec scales (SPARKS). II. Complex organic molecules and heavy water in shocks around a young high-mass protostar. A&A632, A57. 10.1051/0004-6361/201935226

  • CrossRef
  • Google Scholar

• 73

  CumminsS. E.LinkeR. A.ThaddeusP. (1986). A survey of the millimeter-wave spectrum of sagittarius B2. ApJS60, 819. 10.1086/191102

  • CrossRef
  • Google Scholar

• 74

  DanielF.FaureA.PaganiL.LiqueF.GérinM.LisD.et al (2016). N₂H⁺ and N¹⁵NH⁺ toward the prestellar core 16293E in L1689N. A&A592, A45. 10.1051/0004-6361/201628192

  • CrossRef
  • Google Scholar

• 75

  DartoisE.ChabotM.Id BarkachT.RothardH.AugéB.AgnihotriA. N.et al (2019). Non-thermal desorption of complex organic molecules. Efficient CH₃OH and CH₃COOCH₃ sputtering by cosmic rays. A&A627, A55. 10.1051/0004-6361/201834787

  • CrossRef
  • Google Scholar

• 76

  De SimoneM.FontaniF.CodellaC.CeccarelliC.LeflochB.BachillerR.et al (2018). Deuterium and ¹⁵N fractionation in N₂H⁺ during the formation of a Sun-like star. MNRAS476, 1982–1990. 10.1093/mnras/sty353

  • CrossRef
  • Google Scholar

• 77

  DrozdovskayaM. N.van DishoeckE. F.JørgensenJ. K.CalmonteU.van der WielM. H. D.CoutensA.et al (2018). The ALMA-PILS survey: the sulphur connection between protostars and comets: IRAS 16293-2422 B and 67P/Churyumov-Gerasimenko. MNRAS476, 4949–4964. 10.1093/mnras/sty462

  • CrossRef
  • Google Scholar

• 78

  DuleyW. W.WilliamsD. A. (1985). Book-review - Interstellar chemistry. Science227, 786.

  • Google Scholar

• 79

  EganM. P.ShipmanR. F.PriceS. D.CareyS. J.ClarkF. O.CohenM. (1998). A population of cold cores in the galactic plane. ApJL494, L199–L202. 10.1086/311198

  • CrossRef
  • Google Scholar

• 80

  el AkelM.KristensenL. E.Le GalR.van der WaltS. J.PittsR. L.DulieuF. (2022). Unlocking the sulphur chemistry in intermediate-mass protostars of cygnus X. Connecting the cold and warm chemistry. A&A659, A100. 10.1051/0004-6361/202141810

  • CrossRef
  • Google Scholar

• 81

  ElitzurM.HollenbachD. J.McKeeC. F. (1989). H 2O masers in star-forming regions. ApJ346, 983. 10.1086/168080

  • CrossRef
  • Google Scholar

• 82

  EllingsenS. P.BreenS. L.CaswellJ. L.QuinnL. J.FullerG. A. (2010). Masers associated with high-mass star formation regions in the Large Magellanic Cloud. MNRAS404, 779–791. 10.1111/j.1365-2966.2010.16349.x

  • CrossRef
  • Google Scholar

• 83

  EmprechtingerM.CaselliP.VolgenauN. H.StutzkiJ.WiednerM. C. (2009). The N{2}D+̂/N{2}H⁺ ratio as an evolutionary tracer of Class 0 protostars. A&A493, 89–105. 10.1051/0004-6361:200810324

  • CrossRef
  • Google Scholar

• 84

  EntekhabiN.TanJ. C.CosentinoG.HsuC. J.CaselliP.WalshC.et al (2022). Astrochemical modelling of infrared dark clouds. A&A662, A39. 10.1051/0004-6361/202142601

  • CrossRef
  • Google Scholar

• 85

  EspluguesG. B.TerceroB.CernicharoJ.GoicoecheaJ. R.PalauA.MarcelinoN.et al (2013). A line confusion-limited millimeter survey of orion KL. III. Sulfur oxide species. A&A556, A143. 10.1051/0004-6361/201321285

  • CrossRef
  • Google Scholar

• 86

  FavreC.CeccarelliC.López-SepulcreA.FontaniF.NeriR.ManigandS.et al (2018). SOLIS IV. Hydrocarbons in the OMC-2 FIR4 region, a probe of energetic particle irradiation of the region. ApJ859, 136. 10.3847/1538-4357/aabfd4

  • CrossRef
  • Google Scholar

• 87

  FerranteR. F.MooreM. H.SpiliotisM. M.HudsonR. L. (2008). Formation of interstellar OCS: radiation chemistry and IR spectra of precursor ices. ApJ684, 1210–1220. 10.1086/590362

  • CrossRef
  • Google Scholar

• 88

  FontaniF.CesaroniR.FuruyaR. S. (2010). Class I and class II methanol masers in high-mass star-forming regions. A&A517, A56. 10.1051/0004-6361/200913679

  • CrossRef
  • Google Scholar

• 89

  FontaniF.PalauA.CaselliP.Sánchez-MongeÁ.ButlerM. J.TanJ. C.et al (2011). Deuteration as an evolutionary tracer in massive-star formation. A&A529, L7. 10.1051/0004-6361/201116631

  • CrossRef
  • Google Scholar

• 90

  FontaniF.SakaiT.FuruyaK.SakaiN.AikawaY.YamamotoS. (2014). DNC/HNC and N₂D⁺/N₂H⁺ ratios in high-mass star-forming cores. MNRAS440, 448–456. 10.1093/mnras/stu298

  • CrossRef
  • Google Scholar

• 91

  FontaniF.BusquetG.PalauA.CaselliP.Sánchez-MongeÁ.TanJ. C.et al (2015a). Deuteration and evolution in the massive star formation process. The role of surface chemistry. A&A575, A87. 10.1051/0004-6361/201424753

  • CrossRef
  • Google Scholar

• 92

  FontaniF.CaselliP.PalauA.BizzocchiL.CeccarelliC. (2015b). First measurements of ¹⁵N fractionation in N₂H⁺ toward high-mass star-forming cores. ApJL808, L46. 10.1088/2041-8205/808/2/L46

  • CrossRef
  • Google Scholar

• 93

  FontaniF.RivillaV. M.CaselliP.VasyuninA.PalauA. (2016). Phosphorus-bearing molecules in massive dense cores. ApJL822, L30. 10.3847/2041-8205/822/2/L30

  • CrossRef
  • Google Scholar

• 94

  FontaniF.CeccarelliC.FavreC.CaselliP.NeriR.SimsI. R.et al (2017). Seeds of life in space (SOLIS). I. Carbon-chain growth in the solar-type protocluster OMC2-FIR4. A&A605, A57. 10.1051/0004-6361/201730527

  • CrossRef
  • Google Scholar

• 95

  FontaniF.ColziL.RedaelliE.SipiläO.CaselliP. (2021). First survey of HCNH⁺ in high-mass star-forming cloud cores. A&A651, A94. 10.1051/0004-6361/202140655

  • CrossRef
  • Google Scholar

• 96

  FontaniF.ColziL.BizzocchiL.RivillaV. M.EliaD.BeltránM. T.et al (2022). CHEMOUT: CHEMical complexity in star-forming regions of the OUTer galaxy. I. Organic molecules and tracers of star-formation activity. A&A660, A76. 10.1051/0004-6361/202142923

  • CrossRef
  • Google Scholar

• 97

  FontaniF.RoueffE.ColziL.CaselliP. (2023). The evolution of sulphur-bearing molecules in high-mass star-forming cores. A&A680, A58. 10.1051/0004-6361/202347565

  • CrossRef
  • Google Scholar

• 98

  FontaniF.MininniC.BeltránM. T.RivillaV. M.ColziL.Jiménez-SerraI.et al (2024). The GUAPOS project: G31.41+0.31 unbiased ALMA sPectral observational survey. IV. Phosphorus-bearing molecules and their relation to shock tracers. A&A682, A74. 10.1051/0004-6361/202348219

  • CrossRef
  • Google Scholar

• 99

  FontaniF.RivillaV. M.RoueffE.Martín-CaballeroH.BizzocchiL.ColziL.et al (2025). CHemical evolution in MassIve star-forming COres (CHEMICO). I. Evolution of the temperature structure. arXiv:2507.01519 doi. 10.48550/arXiv.2507.01519

  • CrossRef
  • Google Scholar

• 100

  FriedZ. T. P.El-AbdS. J.HaysB. M.WenzelG.ByrneA. N.MargulèsL.et al (2024). Rotational spectrum and first interstellar detection of 2-methoxyethanol using ALMA observations of NGC 6334I. ApJL965, L23. 10.3847/2041-8213/ad37ff

  • CrossRef
  • Google Scholar

• 101

  FriedelD. N.LooneyL. W. (2017). CARMA λ = 1 Cm spectral line survey of Orion-KL. AJ154, 152. 10.3847/1538-3881/aa8865

  • CrossRef
  • Google Scholar

• 102

  FriedelD. N.Widicus WeaverS. L. (2012). Complex organic molecules at high spatial resolution toward ORION-KL. II. Kinematics. ApJS201, 17. 10.1088/0067-0049/201/2/17

  • CrossRef
  • Google Scholar

• 103

  FriedelD. N.SnyderL. E.TurnerB. E.RemijanA. (2004). A spectral line survey of selected 3 millimeter bands toward sagittarius B2(N-LMH) using the national radio astronomy observatory 12 meter radio telescope and the Berkeley-Illinois-Maryland association array. I. The observational data. ApJ600, 234–253. 10.1086/379765

  • CrossRef
  • Google Scholar

• 104

  FriesenR. K.KirkH. M.ShirleyY. L. (2013). An analysis of the deuterium fractionation of star-forming cores in the perseus molecular cloud. ApJ765, 59. 10.1088/0004-637X/765/1/59

  • CrossRef
  • Google Scholar

• 105

  FuenteA.Rivière-MarichalarP.Beitia-AnteroL.CaselliP.WakelamV.EspluguesG.et al (2023). Gas phase elemental abundances in molecular cloudS (GEMS). VII. Sulfur elemental abundance. A&A670, A114. 10.1051/0004-6361/202244843

  • CrossRef
  • Google Scholar

• 106

  FuruyaR. S.WalmsleyC. M.NakanishiK.SchilkeP.BachillerR. (2003). Interferometric observations of FeO towards sagittarius B2. A&A409, L21–L24. 10.1051/0004-6361:20031304

  • CrossRef
  • Google Scholar

• 107

  GarayG.LizanoS. (1999). Massive stars: their environment and formation. PASP111, 1049–1087. 10.1086/316416

  • CrossRef
  • Google Scholar

• 108

  García de la ConcepciónJ.ColziL.Jiménez-SerraI.MolpeceresG.CorchadoJ. C.RivillaV. M.et al (2022). The trans/cis ratio of formic (HCOOH) and thioformic (HC(O)SH) acids in the interstellar medium. A&A658, A150. 10.1051/0004-6361/202142287

  • CrossRef
  • Google Scholar

• 109

  GarrodR. T. (2013). A three-phase chemical model of hot cores: the formation of glycine. ApJ765, 60. 10.1088/0004-637X/765/1/60

  • CrossRef
  • Google Scholar

• 110

  GarrodR. T.HerbstE. (2006). Formation of methyl formate and other organic species in the warm-up phase of hot molecular cores. A&A457, 927–936. 10.1051/0004-6361:20065560

  • CrossRef
  • Google Scholar

• 111

  GarrodR. T.PaulyT. (2011). On the formation of CO₂ and other interstellar ices. ApJ735, 15. 10.1088/0004-637X/735/1/15

  • CrossRef
  • Google Scholar

• 112

  GarrodR. T.Widicus WeaverS. L.HerbstE. (2008). Complex chemistry in star-forming regions: an expanded gas-grain Warm-up chemical model. ApJ682, 283–302. 10.1086/588035

  • CrossRef
  • Google Scholar

• 113

  GarrodR. T.BellocheA.MüllerH. S. P.MentenK. M. (2017). Exploring molecular complexity with ALMA (EMoCA): simulations of branched carbon-chain chemistry in sgr B2(N). A&A601, A48. 10.1051/0004-6361/201630254

  • CrossRef
  • Google Scholar

• 114

  GarrodR. T.JinM.MatisK. A.JonesD.WillisE. R.HerbstE. (2022). Formation of complex organic molecules in hot molecular cores through nondiffusive grain-surface and ice-mantle chemistry. ApJS259, 1. 10.3847/1538-4365/ac3131

  • CrossRef
  • Google Scholar

• 115

  GerinM.CombesF.WlodarczakG.JacqT.GuelinM.EncrenazP.et al (1992). Interstellar detection of deuterated methyl cyanide. A&A259, L35–L38.

  • Google Scholar

• 116

  GernerT.BeutherH.SemenovD.LinzH.VasyuninaT.BihrS.et al (2014). Chemical evolution in the early phases of massive star formation. I. A&A563, A97. 10.1051/0004-6361/201322541

  • CrossRef
  • Google Scholar

• 117

  GernerT.ShirleyY. L.BeutherH.SemenovD.LinzH.AlbertssonT.et al (2015). Chemical evolution in the early phases of massive star formation. II. Deuteration. A&A579, A80. 10.1051/0004-6361/201423989

  • CrossRef
  • Google Scholar

• 118

  GhoshR.DasA.GoraiP.MondalS. K.FuruyaK.TanakaK. E. I.et al (2024). Understanding the various evolutionary stages of the low-mass star-formation process by SO and SO2. Front. Astronomy Space Sci.11, 1427048. 10.3389/fspas.2024.1427048

  • CrossRef
  • Google Scholar

• 119

  GiannettiA.BrandJ.MassiF.TieftrunkA.BeltránM. T. (2012). Molecular clouds under the influence of massive stars in the galactic hii region G353.2+0.9. A&A538, A41. 10.1051/0004-6361/201116774

  • CrossRef
  • Google Scholar

• 120

  GiannettiA.BovinoS.CaselliP.LeuriniS.SchleicherD. R. G.KörtgenB.et al (2019). A timeline for massive star-forming regions via combined observation of o-H₂D⁺ and N₂D⁺. A&A621, L7. 10.1051/0004-6361/201834602

  • CrossRef
  • Google Scholar

• 121

  GieserC.SemenovD.BeutherH.AhmadiA.MottramJ. C.HenningT.et al (2019). Chemical complexity in high-mass star formation. An observational and modeling case study of the AFGL 2591 VLA 3 hot core. A&A631, A142. 10.1051/0004-6361/201935865

  • CrossRef
  • Google Scholar

• 122

  GigliD.FontaniF.ColziL.VermariënG.VitiS.RivillaV. M.et al (2025). “CHEMOUT: CHEMical complexity in star-forming regions of the OUTer galaxy,” in V. Chemical composition gradients as a function of the galactocentric radius. 10.48550/arXiv.2509.26556

  • CrossRef
  • Google Scholar

• 123

  GinardD.González-GarcíaM.FuenteA.CernicharoJ.Alonso-AlbiT.PilleriP.et al (2012). Spectral line survey of the ultracompact HII region monoceros R2. A&A543, A27. 10.1051/0004-6361/201118347

  • CrossRef
  • Google Scholar

• 124

  GinsburgA.BallyJ.GoddiC.PlambeckR.WrightM. (2018). A keplerian disk around orion SrCI, a - 15 M - YSO. ApJ860, 119. 10.3847/1538-4357/aac205

  • CrossRef
  • Google Scholar

• 125

  GinsburgA.McGuireB.PlambeckR.BallyJ.GoddiC.WrightM. (2019). Orion srci’s disk is salty. ApJ872, 54. 10.3847/1538-4357/aafb71

  • CrossRef
  • Google Scholar

• 126

  GoddiC.GreenhillL. J.HumphreysE. M. L.MatthewsL. D.TanJ. C.ChandlerC. J. (2009). A 42.3-43.6 GHz spectral survey of orion BN/KL: First detection of the v = 0 J = 1-0 line from the isotopologues ²⁹SiO and ³⁰SiO. ApJ691, 1254–1264. 10.1088/0004-637X/691/2/1254

  • CrossRef
  • Google Scholar

• 127

  GongY.HenkelC.ThorwirthS.SpezzanoS.MentenK. M.WalmsleyC. M.et al (2015). A 1.3 Cm line survey toward orion KL. A&A581, A48. 10.1051/0004-6361/201526275

  • CrossRef
  • Google Scholar

• 128

  GreenhillL. J.GezariD. Y.DanchiW. C.NajitaJ.MonnierJ. D.TuthillP. G. (2004). High angular resolution mid-infrared imaging of young stars in orion BN/KL. ApJL605, L57–L60. 10.1086/386544

  • CrossRef
  • Google Scholar

• 129

  HasegawaT. I.HerbstE.LeungC. M. (1992). Models of gas-grain chemistry in dense interstellar clouds with complex organic molecules. ApJS82, 167. 10.1086/191713

  • CrossRef
  • Google Scholar

• 130

  HerbstE. (2021). Unusual chemical processes in interstellar chemistry: past and present. Front. Astronomy Space Sci.8, 776942. 10.3389/fspas.2021.776942

  • CrossRef
  • Google Scholar

• 131

  HerpinF.MarseilleM.WakelamV.BontempsS.LisD. C. (2009). S-bearing molecules in massive dense cores. A&A504, 853–867. 10.1051/0004-6361/200811257

  • CrossRef
  • Google Scholar

• 132

  HirotaT.KimM. K.HonmaM. (2012). The first detection of the 232 GHz vibrationally excited H₂O maser in orion KL with ALMA. ApJL757, L1. 10.1088/2041-8205/757/1/L1

  • CrossRef
  • Google Scholar

• 133

  HollisJ. M.LovasF. J.JewellP. R. (2000). Interstellar glycolaldehyde: the first sugar. ApJL540, L107–L110. 10.1086/312881

  • CrossRef
  • Google Scholar

• 134

  HollisJ. M.JewellP. R.LovasF. J.RemijanA.MøllendalH. (2004). Green bank telescope detection of new interstellar aldehydes: propenal and propanal. ApJL610, L21–L24. 10.1086/423200

  • CrossRef
  • Google Scholar

• 135

  HollisJ. M.LovasF. J.RemijanA. J.JewellP. R.IlyushinV. V.KleinerI. (2006a). Detection of acetamide (CH₃CONH₂): the largest interstellar molecule with a peptide bond. ApJL643, L25–L28. 10.1086/505110

  • CrossRef
  • Google Scholar

• 136

  HollisJ. M.RemijanA. J.JewellP. R.LovasF. J. (2006b). Cyclopropenone (c-H₂C₃O): a new interstellar ring molecule. ApJ642, 933–939. 10.1086/501121

  • CrossRef
  • Google Scholar

• 137

  HoqS.JacksonJ. M.FosterJ. B.SanhuezaP.GuzmánA.WhitakerJ. S.et al (2013). Chemical evolution in high-mass star-forming regions: results from the MALT90 survey. ApJ777, 157. 10.1088/0004-637X/777/2/157

  • CrossRef
  • Google Scholar

• 138

  HuangM.BaniaT. M.BolattoA.ChamberlinR. A.IngallsJ. G.JacksonJ. M.et al (1999). Atomic carbon observations of southern hemisphere H II regions. ApJ517, 282–291. 10.1086/307194

  • CrossRef
  • Google Scholar

• 139

  HunterT. R.BroganC. L.MegeathS. T.MentenK. M.BeutherH.ThorwirthS. (2006). Millimeter multiplicity in NGC 6334 I and I(N). ApJ649, 888–893. 10.1086/505965

  • CrossRef
  • Google Scholar

• 140

  ImmerK.LiJ.Quiroga-NuñezL. H.ReidM. J.ZhangB.MoscadelliL.et al (2019). Anomalous peculiar motions of high-mass young stars in the scutum spiral arm. A&A632, A123. 10.1051/0004-6361/201834208

  • CrossRef
  • Google Scholar

• 141

  JacksonJ. M.RathborneJ. M.FosterJ. B.WhitakerJ. S.SanhuezaP.ClaysmithC.et al (2013). MALT90: the millimetre astronomy legacy team 90 GHz survey. PASA30, e057. 10.1017/pasa.2013.37

  • CrossRef
  • Google Scholar

• 142

  JewellP. R.HollisJ. M.LovasF. J.SnyderL. E. (1989). Millimeter- and submillimeter-wave surveys of orion A emission lines in the ranges 200.7–202.3, 203.7–205.3, and 330–360 GHz. ApJS70, 833. 10.1086/191359

  • CrossRef
  • Google Scholar

• 143

  Jiménez-SerraI.Báez-RubioA.RivillaV. M.Martín-PintadoJ.ZhangQ.DierickxM.et al (2013). A new radio recombination line maser object toward the MonR2 H II region. ApJL764, L4. 10.1088/2041-8205/764/1/L4

  • CrossRef
  • Google Scholar

• 144

  JinM.GarrodR. T. (2020). Formation of complex organic molecules in cold interstellar environments through nondiffusive grain-surface and ice-mantle chemistry. ApJS249, 26. 10.3847/1538-4365/ab9ec8

  • CrossRef
  • Google Scholar

• 145

  JohanssonL. E. B.AnderssonC.EllderJ.FribergP.HjalmarsonA.HoglundB.et al (1984). Spectral scan of orion A and IRC +10216 from 72 to 91 GHz. A&A130, 227–256.

  • Pubmed Abstract
  • Google Scholar

• 146

  JørgensenJ. K.HogerheijdeM. R.BlakeG. A.van DishoeckE. F.MundyL. G.SchöierF. L. (2004). The impact of shocks on the chemistry of molecular clouds. High resolution images of chemical differentiation along the NGC 1333-IRAS 2A outflow. A&A415, 1021–1037. 10.1051/0004-6361:20034216

  • CrossRef
  • Google Scholar

• 147

  JørgensenJ. K.BellocheA.GarrodR. T. (2020). Astrochemistry during the formation of stars. ARA&A58, 727–778. 10.1146/annurev-astro-032620-021927

  • CrossRef
  • Google Scholar

• 148

  KirsanovaM. S.Ossenkopf-OkadaV.AndersonL. D.BoleyP. A.BiegingJ. H.PavlyuchenkovY. N.et al (2020). The PDR structure and kinematics around the compact H II regions S235 A and S235 C with [C II], [¹³C II], [O I], and HCO⁺ line profiles. MNRAS497, 2651–2669. 10.1093/mnras/staa2142

  • CrossRef
  • Google Scholar

• 149

  KongS.CaselliP.TanJ. C.WakelamV.SipiläO. (2015). The deuterium fractionation timescale in dense cloud cores: a parameter space exploration. ApJ804, 98. 10.1088/0004-637X/804/2/98

  • CrossRef
  • Google Scholar

• 150

  KongS.TanJ. C.CaselliP.FontaniF.PillaiT.ButlerM. J.et al (2016). The deuterium fraction in massive starless cores and dynamical implications. ApJ821, 94. 10.3847/0004-637X/821/2/94

  • CrossRef
  • Google Scholar

• 151

  KurtzS. (2002). “Ultracompact HII regions,”. Hot star workshop III: the earliest phases of massive star birth. Editor CrowtherP. (San Francisco, CA: Astronomical Society of the Pacific), 81. 10.48550/arXiv.astro-ph/0111351

  • CrossRef
  • Google Scholar

• 152

  KurtzS. (2005). “Hypercompact HII regions,”. Massive star birth: a crossroads of astrophysics (IAU Symposium). Editors CesaroniR.FelliM.ChurchwellE.WalmsleyM. (Cambridge: Cambridge University Press), 227, 111–119. 10.1017/S1743921305004424

  • CrossRef
  • Google Scholar

• 153

  LadaC. J.LadaE. A. (2003). Embedded clusters in molecular clouds. ARA&A41, 57–115. 10.1146/annurev.astro.41.011802.094844

  • CrossRef
  • Google Scholar

• 154

  LadeyschikovD. A.GongY.SobolevA. M.MentenK. M.UrquhartJ. S.BreenS. L.et al (2022). Water masers as an early tracer of star formation. ApJS261, 14. 10.3847/1538-4365/ac6b43

  • CrossRef
  • Google Scholar

• 155

  Le GalR.ÖbergK. I.HuangJ.LawC. J.MénardF.LeflochB.et al (2020). A 3 mm chemical exploration of small organics in class I YSOs. ApJ898, 131. 10.3847/1538-4357/ab9ebf

  • CrossRef
  • Google Scholar

• 156

  Le PetitF.NehméC.Le BourlotJ.RoueffE. (2006). A model for atomic and molecular interstellar gas: the meudon PDR code. ApJS164, 506–529. 10.1086/503252

  • CrossRef
  • Google Scholar

• 157

  LeeC.-F. (2020). Molecular jets from low-mass young protostellar objects. A&A Rev.28, 1. 10.1007/s00159-020-0123-7

  • CrossRef
  • Google Scholar

• 158

  LeeE. J.MurrayN.RahmanM. (2012). Milky Way star-forming complexes and the turbulent motion of the galaxy’s molecular gas. ApJ752, 146. 10.1088/0004-637X/752/2/146

  • CrossRef
  • Google Scholar

• 159

  LeflochB.VastelC.VitiS.Jimenez-SerraI.CodellaC.PodioL.et al (2016). Phosphorus-bearing molecules in solar-type star-forming regions: first PO detection. MNRAS462, 3937–3944. 10.1093/mnras/stw1918

  • CrossRef
  • Google Scholar

• 160

  LeflochB.BachillerR.CeccarelliC.CernicharoJ.CodellaC.FuenteA.et al (2018). Astrochemical evolution along star formation: overview of the IRAM large program ASAI. MNRAS477, 4792–4809. 10.1093/mnras/sty937

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 161

  LerateM. R.BarlowM. J.SwinyardB. M.GoicoecheaJ. R.CernicharoJ.GrundyT. W.et al (2006). A far-infrared molecular and atomic line survey of the orion KL region. MNRAS370, 597–628. 10.1111/j.1365-2966.2006.10518.x

  • CrossRef
  • Google Scholar

• 162

  LiJ.WangJ.ZhuQ.ZhangJ.LiD. (2015). Sulfur-bearing molecules in massive star-forming regions: observations of OCS, CS, H₂S, and SO. ApJ802, 40. 10.1088/0004-637X/802/1/40

  • CrossRef
  • Google Scholar

• 163

  LiS.WangJ.ZhangZ.-Y.FangM.LiJ.ZhangJ.et al (2017). Millimetre spectral line mapping observations towards four massive star-forming H II regions. MNRAS466, 248–275. 10.1093/mnras/stw3076

  • CrossRef
  • Google Scholar

• 164

  LiS.SanhuezaP.LuX.LeeC. W.ZhangQ.BovinoS.et al (2022). The ALMA survey of 70 μm dark high-mass clumps in early stages (ASHES). VII. Chemistry of embedded dense cores. ApJ939, 102. 10.3847/1538-4357/ac94d4

  • CrossRef
  • Google Scholar

• 165

  LisD. C.PearsonJ. C.NeufeldD. A.SchilkeP.MüllerH. S. P.GuptaH.et al (2010). Herschel/HIFI discovery of interstellar chloronium (H₂Cl⁺). A&A521, L9. 10.1051/0004-6361/201014959

  • CrossRef
  • Google Scholar

• 166

  LiuM.TanJ. C.MarvilJ.KongS.RoseroV.CaselliP.et al (2021). SiO outflows as tracers of massive star formation in infrared dark clouds. ApJ921, 96. 10.3847/1538-4357/ac0829

  • CrossRef
  • Google Scholar

• 167

  LiuX.LiuT.ShenZ.QinS.-L.LuoQ.ChengY.et al (2022). A Q-band line survey toward orion KL using the tianma radio telescope. ApJS263, 13. 10.3847/1538-4365/ac9127

  • CrossRef
  • Google Scholar

• 168

  LiuX.LiuT.ShenZ.QinS.-L.LuoQ.GongY.et al (2024). The first Ka-band (26.1–35 GHz) blind line survey toward orion KL. ApJS271, 3. 10.3847/1538-4365/ad1601

  • CrossRef
  • Google Scholar

• 169

  LoddersK. (2003). Solar system abundances and condensation temperatures of the elements. ApJ591, 1220–1247. 10.1086/375492

  • CrossRef
  • Google Scholar

• 170

  LoisonJ.-C.HalvickP.BergeatA.HicksonK. M.WakelamV. (2012). Review of OCS gas-phase reactions in dark cloud chemical models. MNRAS421, 1476–1484. 10.1111/j.1365-2966.2012.20412.x

  • CrossRef
  • Google Scholar

• 171

  LoomisR. A.ZaleskiD. P.SteberA. L.NeillJ. L.MuckleM. T.HarrisB. J.et al (2013). The detection of interstellar ethanimine (CH₃CHNH) from observations taken during the GBT PRIMOS survey. ApJL765, L9. 10.1088/2041-8205/765/1/L9

  • CrossRef
  • Google Scholar

• 172

  López-GallifaÁ.RivillaV. M.BeltránM. T.ColziL.MininniC.Sánchez-MongeÁ.et al (2024). The GUAPOS project - V: the chemical ingredients of a massive stellar protocluster in the making. MNRAS529, 3244–3283. 10.1093/mnras/stae676

  • CrossRef
  • Google Scholar

• 173

  López-SepulcreA.WalmsleyC. M.CesaroniR.CodellaC.SchullerF.BronfmanL.et al (2011). SiO outflows in high-mass star forming regions: a potential chemical clock?A&A526, L2. 10.1051/0004-6361/201015827

  • CrossRef
  • Google Scholar

• 174

  LuoR.WangJ. Z.ZhangX.QuanD. H.JiangX. J.LiJ.et al (2024). Observational studies of S-bearing molecules in massive star-forming regions. A&A691, A359. 10.1051/0004-6361/202449477

  • CrossRef
  • Google Scholar

• 175

  MartinezN. C.ParonS.OrtegaM. E.PetriellaA.ÁlamoA.BrookM.et al (2024). Sulfur-bearing molecules in a sample of early star-forming cores. A&A692, A97. 10.1051/0004-6361/202452350

  • CrossRef
  • Google Scholar

• 176

  MartyB.ChaussidonM.WiensR. C.JurewiczA. J. G.BurnettD. S. (2011). A ¹⁵N-Poor isotopic composition for the solar system as shown by genesis solar wind samples. Science332, 1533–1536. 10.1126/science.1204656

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 177

  MatthewsL. D.GreenhillL. J.GoddiC.ChandlerC. J.HumphreysE. M. L.KunzM. W. (2010). A feature movie of SiO emission 20-100 AU from the massive young stellar object orion source I. ApJ708, 80–92. 10.1088/0004-637X/708/1/80

  • CrossRef
  • Google Scholar

• 178

  McClureM. K.RochaW. R. M.PontoppidanK. M.CrouzetN.ChuL. E. U.DartoisE.et al (2023). An ice age JWST inventory of dense molecular cloud ices. Nat. Astron.7, 431–443. 10.1038/s41550-022-01875-w

  • CrossRef
  • Google Scholar

• 179

  McCutcheonW. H.SandellG.MatthewsH. E.KuiperT. B. H.SuttonE. C.DanchiW. C.et al (2000). Star formation in NGC 6334 I and I(N). MNRAS316, 152–164. 10.1046/j.1365-8711.2000.03487.x

  • CrossRef
  • Google Scholar

• 180

  McGuireB. A. (2018). 2018 census of interstellar, circumstellar, extragalactic, protoplanetary disk, and exoplanetary molecules. ApJS239, 17. 10.3847/1538-4365/aae5d2

  • CrossRef
  • Google Scholar

• 181

  McGuireB. A.BroganC. L.HunterT. R.RemijanA. J.BlakeG. A.BurkhardtA. M.et al (2018). First results of an ALMA band 10 spectral line survey of NGC 6334I: detections of glycolaldehyde (HC(O)CH₂OH) and a new compact bipolar outflow in HDO and CS. ApJL863, L35. 10.3847/2041-8213/aad7bb

  • CrossRef
  • Google Scholar

• 182

  McGuireB. A.LoomisR. A.BurkhardtA. M.LeeK. L. K.ShingledeckerC. N.CharnleyS. B.et al (2021). Detection of two interstellar polycyclic aromatic hydrocarbons via spectral matched filtering. Science371, 1265–1269. 10.1126/science.abb7535

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 183

  McKeeC. F.TanJ. C. (2002). Massive star formation in 100,000 years from turbulent and pressurized molecular clouds. Nature416, 59–61. 10.1038/416059a

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 184

  McKeeC. F.TanJ. C. (2003). The formation of massive stars from turbulent cores. ApJ585, 850–871. 10.1086/346149

  • CrossRef
  • Google Scholar

• 185

  MehringerD. M.SnyderL. E.MiaoY.LovasF. J. (1997). Detection and confirmation of interstellar acetic acid. ApJL480, L71–L74. 10.1086/310612

  • CrossRef
  • Google Scholar

• 186

  MentenK. M. (1991). “Methanol masers and submillimeter wavelength water masers in star-forming regions,” in Atoms, ions and molecules: new results in spectral line astrophysics. 16 of astronomical society of the Pacific conference series. Editors HaschickA. D.HoP. T. P.119–136.

  • Google Scholar

• 187

  MentenK. M.ReidM. J.ForbrichJ.BrunthalerA. (2007). The distance to the orion nebula. A&A474, 515–520. 10.1051/0004-6361:20078247

  • CrossRef
  • Google Scholar

• 188

  MercimekS.CodellaC.PodioL.BianchiE.ChahineL.BouvierM.et al (2022). Chemical survey of class I protostars with the IRAM-30 m. A&A659, A67. 10.1051/0004-6361/202141790

  • CrossRef
  • Google Scholar

• 189

  MininniC.FontaniF.RivillaV. M.BeltránM. T.CaselliP.VasyuninA. (2018). On the origin of phosphorus nitride in star-forming regions. MNRAS476, L39–L44. 10.1093/mnrasl/sly026

  • CrossRef
  • Google Scholar

• 190

  MininniC.BeltránM. T.RivillaV. M.Sánchez-MongeA.FontaniF.MöllerT.et al (2020). The GUAPOS project: G31.41+0.31 unbiased ALMA sPectral observational survey. I. Isomers C₂H₄O₂. A&A644, A84. 10.1051/0004-6361/202038966

  • CrossRef
  • Google Scholar

• 191

  MininniC.BeltránM. T.ColziL.RivillaV. M.FontaniF.LorenzaniA.et al (2023). The GUAPOS project. III. Characterization of the O- and N-bearing complex organic molecules content and search for chemical differentiation. A&A677, A15. 10.1051/0004-6361/202245277

  • CrossRef
  • Google Scholar

• 192

  MladenovićM.RoueffE. (2014). Ion-molecule reactions involving HCO⁺ and N₂H⁺: isotopologue equilibria from new theoretical calculations and consequences for interstellar isotope fractionation. A&A566, A144. 10.1051/0004-6361/201423733

  • CrossRef
  • Google Scholar

• 193

  MladenovićM.RoueffE. (2017). Isotope exchange reactions involving HCO⁺ with CO: a theoretical approach. A&A605, A22. 10.1051/0004-6361/201731270

  • CrossRef
  • Google Scholar

• 194

  MolinariS.SwinyardB.BallyJ.BarlowM.BernardJ. P.MartinP.et al (2010). Clouds, filaments, and protostars: the herschel Hi-GAL Milky Way. A&A518, L100. 10.1051/0004-6361/201014659

  • CrossRef
  • Google Scholar

• 195

  MöllerT.SchilkeP.SchmiedekeA.BerginE. A.LisD. C.Sánchez-MongeÁ.et al (2021). Herschel observations of extraordinary sources: full Herschel/HIFI molecular line survey of sagittarius B2(M). A&A651, A9. 10.1051/0004-6361/202040203

  • CrossRef
  • Google Scholar

• 196

  MoscadelliL.RivillaV. M.CesaroniR.BeltránM. T.Sánchez-MongeÁ.SchilkeP.et al (2018). The feedback of an HC HII region on its parental molecular core. The case of core A1 in the star-forming region G24.78+0.08. A&A616, A66. 10.1051/0004-6361/201832680

  • CrossRef
  • Google Scholar

• 197

  MotteF.BontempsS.LouvetF. (2018). High-mass star and massive cluster formation in the Milky Way. ARA&A56, 41–82. 10.1146/annurev-astro-091916-055235

  • CrossRef
  • Google Scholar

• 198

  MummaM. J.CharnleyS. B. (2011). The chemical composition of comets—emerging taxonomies and natal heritage. ARA&A49, 471–524. 10.1146/annurev-astro-081309-130811

  • CrossRef
  • Google Scholar

• 199

  NakibovR.KarteyevaV.PetrashkevichI.OzhiganovM.MedvedevM.VasyuninA. (2025). Solid and gaseous methane in IRAS 23385+6053 as seen with open JWST data. ApJL978, L46. 10.3847/2041-8213/ada290

  • CrossRef
  • Google Scholar

• 200

  NeckelT. (1978). UBV, VRI and hbeta observations of stars in the H II regions NGC 6334 and NGC 6357. A&A69, 51–56.

  • Google Scholar

• 201

  NeillJ. L.MuckleM. T.ZaleskiD. P.SteberA. L.PateB. H.LattanziV.et al (2012). Laboratory and tentative interstellar detection of trans-methyl formate using the publicly available green bank telescope primos survey. ApJ755, 153. 10.1088/0004-637X/755/2/153

  • CrossRef
  • Google Scholar

• 202

  NeillJ. L.BerginE. A.LisD. C.SchilkeP.CrockettN. R.FavreC.et al (2014). Herschel observations of extraordinary sources: analysis of the full Herschel/HIFI Molecular Line Survey of Sagittarius B2(N). ApJ789, 8. 10.1088/0004-637X/789/1/8

  • CrossRef
  • Google Scholar

• 203

  NeufeldD. A.ZmuidzinasJ.SchilkeP.PhillipsT. G. (1997). Discovery of interstellar hydrogen fluoride. ApJL488, L141–L144. 10.1086/310942

  • CrossRef
  • Google Scholar

• 204

  OliveiraC. M.HébrardG.HowkJ. C.KrukJ. W.ChayerP.MoosH. W. (2003). Interstellar deuterium, nitrogen, and oxygen abundances toward GD 246, WD 2331-475, HZ 21, and lanning 23: results from the FUSE mission. ApJ587, 235–255. 10.1086/368019

  • CrossRef
  • Google Scholar

• 205

  OsorioM.AngladaG.LizanoS.D’AlessioP. (2009). Collapsing hot molecular cores: a model for the dust spectrum and ammonia line emission of the G31.41+0.31 hot core. ApJ694, 29–45. 10.1088/0004-637X/694/1/29

  • CrossRef
  • Google Scholar

• 206

  OssenkopfV.RölligM.NeufeldD. A.PilleriP.LisD. C.FuenteA.et al (2013). Herschel/HIFI observations of [C II] and [¹³C II] in photon-dominated regions. A&A550, A57. 10.1051/0004-6361/201219837

  • CrossRef
  • Google Scholar

• 207

  PaganiL.BacmannA.CabritS.VastelC. (2007). Depletion and low gas temperature in the L183 (=L134N) prestellar core: the N2H+̂-N2D⁺ tool. A&A467, 179–186. 10.1051/0004-6361:20066670

  • CrossRef
  • Google Scholar

• 208

  PalumboM. E.GeballeT. R.TielensA. G. G. M. (1997). Solid carbonyl sulfide (OCS) in dense molecular clouds. ApJ479, 839–844. 10.1086/303905

  • CrossRef
  • Google Scholar

• 209

  PazukhinA. G.ZinchenkoI. I.TrofimovaE. A.HenkelC.SemenovD. A. (2023). Variations of the HCO⁺, HCN, HNC, N₂H⁺, and NH₃ deuterium fractionation in high-mass star-forming regions. MNRAS526, 3673–3696. 10.1093/mnras/stad2976

  • CrossRef
  • Google Scholar

• 210

  PeraultM.OmontA.SimonG.SeguinP.OjhaD.BlommaertJ.et al (1996). First ISOCAM images of the Milky Way. A&A315, L165–L168.

  • Google Scholar

• 211

  PillaiT.WyrowskiF.HatchellJ.GibbA. G.ThompsonM. A. (2007). Probing the initial conditions of high mass star formation. I. Deuteration and depletion in high mass pre/protocluster clumps. A&A467, 207–216. 10.1051/0004-6361:20065682

  • CrossRef
  • Google Scholar

• 212

  PillaiT.KauffmannJ.WyrowskiF.HatchellJ.GibbA. G.ThompsonM. A. (2011). Probing the initial conditions of high-mass star formation. II. Fragmentation, stability, and chemistry towards high-mass star-forming regions G29.96-0.02 and G35.20-1.74. A&A530, A118. 10.1051/0004-6361/201015899

  • CrossRef
  • Google Scholar

• 213

  PillaiT.KauffmannJ.ZhangQ.SanhuezaP.LeuriniS.WangK.et al (2019). Massive and low-mass protostars in massive “starless” cores. A&A622, A54. 10.1051/0004-6361/201732570

  • CrossRef
  • Google Scholar

• 214

  PilleriP.Treviño-MoralesS.FuenteA.JoblinC.CernicharoJ.GerinM.et al (2013). Spatial distribution of small hydrocarbons in the neighborhood of the ultra compact HII region monoceros R2. A&A554, A87. 10.1051/0004-6361/201220795

  • CrossRef
  • Google Scholar

• 215

  Portegies ZwartS. (2019). The formation of solar-system analogs in young star clusters. A&A622, A69. 10.1051/0004-6361/201833974

  • CrossRef
  • Google Scholar

• 216

  Portegies ZwartS.PelupessyI.van ElterenA.WijnenT. P. G.LugaroM. (2018). The consequences of a nearby supernova on the early solar system. A&A616, A85. 10.1051/0004-6361/201732060

  • CrossRef
  • Google Scholar

• 217

  QinS. L.SchilkeP.RolffsR.ComitoC.LisD. C.ZhangQ. (2011). Submillimeter continuum observations of sagittarius B2 at subarcsecond spatial resolution. A&A530, L9. 10.1051/0004-6361/201116928

  • CrossRef
  • Google Scholar

• 218

  RedaelliE.BizzocchiL.CaselliP.HarjuJ.Chacón-TanarroA.LeonardoE.et al (2018). ¹⁴N/¹⁵N ratio measurements in prestellar cores with N₂H⁺: new evidence of ¹⁵N-antifractionation. A&A617, A7. 10.1051/0004-6361/201833065

  • CrossRef
  • Google Scholar

• 219

  RedaelliE.BizzocchiL.CaselliP. (2020). First sample of N₂H⁺ nitrogen isotopic ratio measurements in low-mass protostars. A&A644, A29. 10.1051/0004-6361/202039303

  • CrossRef
  • Google Scholar

• 220

  Rivière-MarichalarP.FuenteA.Le GalR.BaruteauC.NeriR.Navarro-AlmaidaD.et al (2020). AB Aur, a rosetta stone for studies of planet formation. I. Chemical study of a planet-forming disk. A&A642, A32. 10.1051/0004-6361/202038549

  • CrossRef
  • Google Scholar

• 221

  RivillaV. M.Jiménez-SerraI.Martín-PintadoJ.Sanz-ForcadaJ. (2014). The role of low-mass star clusters in forming the massive stars in DR 21. MNRAS437, 1561–1575. 10.1093/mnras/stt1989

  • CrossRef
  • Google Scholar

• 222

  RivillaV. M.FontaniF.BeltránM. T.VasyuninA.CaselliP.Martín-PintadoJ.et al (2016). The first detections of the key prebiotic molecule PO in star-forming regions. ApJ826, 161. 10.3847/0004-637X/826/2/161

  • CrossRef
  • Google Scholar

• 223

  RivillaV. M.BeltránM. T.CesaroniR.FontaniF.CodellaC.ZhangQ. (2017a). Formation of ethylene glycol and other complex organic molecules in star-forming regions. A&A598, A59. 10.1051/0004-6361/201628373

  • CrossRef
  • Google Scholar

• 224

  RivillaV. M.BeltránM. T.Martín-PintadoJ.FontaniF.CaselliP.CesaroniR. (2017b). On the chemical ladder of esters. Detection and formation of ethyl formate in the W51 e2 hot molecular core. A&A599, A26. 10.1051/0004-6361/201628823

  • CrossRef
  • Google Scholar

• 225

  RivillaV. M.DrozdovskayaM. N.AltweggK.CaselliP.BeltránM. T.FontaniF.et al (2020). ALMA and ROSINA detections of phosphorus-bearing molecules: the interstellar thread between star-forming regions and comets. MNRAS492, 1180–1198. 10.1093/mnras/stz3336

  • CrossRef
  • Google Scholar

• 226

  RizzoJ. R.FuenteA.Rodríguez-FrancoA.García-BurilloS. (2003). Detection of reactive ions in the ultracompact H II regions monoceros R2 and G29.96-0.02. ApJL597, L153–L156. 10.1086/379867

  • CrossRef
  • Google Scholar

• 227

  RizzoJ. R.TerceroB.CernicharoJ. (2017). A spectroscopic survey of orion KL between 41.5 and 50 GHz. A&A605, A76. 10.1051/0004-6361/201629936

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 228

  RochaW. R. M.van DishoeckE. F.ResslerM. E.van GelderM. L.SlavicinskaK.BrunkenN. G. C.et al (2024). JWST observations of young protoStars (JOYS+): detecting icy complex organic molecules and ions. I. CH₄, SO₂, HCOO⁻, OCN⁻, H₂CO, HCOOH, CH₃CH₂OH, CH₃CHO, CH₃OCHO, CH₃COOH. A&A683, A124. 10.1051/0004-6361/202348427

  • CrossRef
  • Google Scholar

• 229

  RodónJ. A.ZavagnoA.BaluteauJ. P.AndersonL. D.PolehamptonE.AbergelA.et al (2010). Physical properties of the Sh2-104 H II region as seen by herschel. A&A518, L80. 10.1051/0004-6361/201014609

  • CrossRef
  • Google Scholar

• 230

  RodríguezT. M.HofnerP.EdelmanI.ArayaE. D.RoseroV. (2023). Searching for molecular jets from high-mass protostars. ApJS264, 30. 10.3847/1538-4365/aca4c6

  • CrossRef
  • Google Scholar

• 231

  RomanoD. (2022). The evolution of CNO elements in galaxies. A&A Rev.30, 7. 10.1007/s00159-022-00144-z

  • CrossRef
  • Google Scholar

• 232

  RuaudM.LoisonJ. C.HicksonK. M.GratierP.HersantF.WakelamV. (2015). Modelling complex organic molecules in dense regions: eley-rideal and complex induced reaction. MNRAS447, 4004–4017. 10.1093/mnras/stu2709

  • CrossRef
  • Google Scholar

• 233

  RubinR. H.SwensonG. W.Jr.BensonR. C.TigelaarH. L.FlygareW. H. (1971). Microwave detection of interstellar formamide. ApJL169, L39. 10.1086/180810

  • CrossRef
  • Google Scholar

• 234

  RuffleD. P.HerbstE. (2000). New models of interstellar gas-grain chemistry - I. Surface diffusion rates. MNRAS319, 837–850. 10.1046/j.1365-8711.2000.03911.x

  • CrossRef
  • Google Scholar

• 235

  SabatiniG.BovinoS.GiannettiA.GrassiT.BrandJ.SchisanoE.et al (2021). Establishing the evolutionary timescales of the massive star formation process through chemistry. A&A652, A71. 10.1051/0004-6361/202140469

  • CrossRef
  • Google Scholar

• 236

  SabatiniG.BovinoS.RedaelliE.WyrowskiF.UrquhartJ. S.GiannettiA.et al (2024). Time evolution of o-H₂D⁺, N₂D⁺, and N₂H⁺ during the high-mass star formation process. A&A692, A265. 10.1051/0004-6361/202451659

  • CrossRef
  • Google Scholar

• 237

  SakaiN.YamamotoS. (2013). Warm carbon-chain chemistry. Chem. Rev.113, 8981–9015. 10.1021/cr4001308

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 238

  SakaiT.SakaiN.FuruyaK.AikawaY.HirotaT.YamamotoS. (2012). DNC/HNC ratio of massive clumps in early evolutionary stages of high-mass star formation. ApJ747, 140. 10.1088/0004-637X/747/2/140

  • CrossRef
  • Google Scholar

• 239

  Sánchez-MongeÁ.SchilkeP.SchmiedekeA.GinsburgA.CesaroniR.LisD. C.et al (2017). The physical and chemical structure of sagittarius B2. II. Continuum millimeter emission of sgr B2(M) and sgr B2(N) with ALMA. A&A604, A6. 10.1051/0004-6361/201730426

  • CrossRef
  • Google Scholar

• 240

  SandellG. (2000). (Sub)mm continuum mapping of NGC 6334 I and I(N). A cobweb of filaments and protostars. A&A358, 242–256.

  • Google Scholar

• 241

  SanhuezaP.JacksonJ. M.FosterJ. B.GarayG.SilvaA.FinnS. C. (2012). Chemistry in infrared dark cloud Clumps: a Molecular Line Survey at 3 mm. ApJ756, 60. 10.1088/0004-637X/756/1/60

  • CrossRef
  • Google Scholar

• 242

  SantosJ. C.van GelderM. L.NazariP.AhmadiA.van DishoeckE. F. (2024). SO₂ and OCS toward high-mass protostars: a comparative study of ice and gas. A&A689, A248. 10.1051/0004-6361/202450736

  • CrossRef
  • Google Scholar

• 243

  SchilkeP.BenfordD. J.HunterT. R.LisD. C.PhillipsT. G. (2001). A line survey of Orion-KL from 607 to 725 GHZ. ApJS132, 281–364. 10.1086/318951

  • CrossRef
  • Google Scholar

• 244

  SchilkeP.ComitoC.ThorwirthS.WyrowskiF.MentenK. M.GüstenR.et al (2006). Submillimeter spectroscopy of southern hot cores: NGC 6334(I) and G327.3-0.6. A&A454, L41–L45. 10.1051/0004-6361:20065398

  • CrossRef
  • Google Scholar

• 245

  SchmiedekeA.SchilkeP.MöllerT.Sánchez-MongeÁ.BerginE.ComitoC.et al (2016). The physical and chemical structure of Sagittarius B2. I. Three-dimensional thermal dust and free-free continuum modeling on 100 au to 45 pc scales. A&A588, A143. 10.1051/0004-6361/201527311

  • CrossRef
  • Google Scholar

• 246

  ScibelliS.ShirleyY.MegíasA.Jiménez-SerraI. (2024). Survey of complex organic molecules in starless and pre-stellar cores in the perseus molecular cloud. MNRAS533, 4104–4149. 10.1093/mnras/stae2017

  • CrossRef
  • Google Scholar

• 247

  SemenovD.HersantF.WakelamV.DutreyA.ChapillonE.GuilloteauS.et al (2010). Chemistry in disks. IV. Benchmarking gas-grain chemical models with surface reactions. A&A522, A42. 10.1051/0004-6361/201015149

  • CrossRef
  • Google Scholar

• 248

  SewiłoM.IndebetouwR.CharnleyS. B.ZahoreczS.OliveiraJ. M.van LoonJ. T.et al (2018). The detection of hot cores and complex organic molecules in the Large Magellanic Cloud. ApJL853, L19. 10.3847/2041-8213/aaa079

  • CrossRef
  • Google Scholar

• 249

  SewiłoM.CordinerM.CharnleyS. B.OliveiraJ. M.Garcia-BerriosE.SchilkeP.et al (2022). ALMA observations of molecular complexity in the Large Magellanic Cloud: the N 105 star-forming region. ApJ931, 102. 10.3847/1538-4357/ac4e8f

  • CrossRef
  • Google Scholar

• 250

  ShimonishiT.OnakaT.KawamuraA.AikawaY. (2016). The detection of a hot molecular core in the Large Magellanic Cloud with ALMA. ApJ827, 72. 10.3847/0004-637X/827/1/72

  • CrossRef
  • Google Scholar

• 251

  ShimonishiT.TanakaK. E. I.ZhangY.FuruyaK. (2023). The detection of hot molecular cores in the Small Magellanic Cloud. ApJL946, L41. 10.3847/2041-8213/acc031

  • CrossRef
  • Google Scholar

• 252

  SkouterisD.BalucaniN.CeccarelliC.VazartF.PuzzariniC.BaroneV.et al (2018). The genealogical tree of ethanol: gas-Phase formation of glycolaldehyde, acetic acid, and formic acid. ApJ854, 135. 10.3847/1538-4357/aaa41e

  • CrossRef
  • Google Scholar

• 253

  SocciA.SabatiniG.PadovaniM.BovinoS.HacarA. (2024). Parsec-scale cosmic-ray ionisation rate in orion. A&A687, A70. 10.1051/0004-6361/202449960

  • CrossRef
  • Google Scholar

• 254

  StéphanG.SchilkeP.Le BourlotJ.SchmiedekeA.ChoudhuryR.GodardB.et al (2018). Chemical modeling of internal photon-dominated regions surrounding deeply embedded HC/UCHII regions. A&A617, A60. 10.1051/0004-6361/201730639

  • CrossRef
  • Google Scholar

• 255

  SutterJ.SandstromK.ChastenetJ.LeroyA. K.KochE. W.WilliamsT. G.et al (2024). The fraction of dust mass in the form of polycyclic aromatic hydrocarbons on 10–50 Pc scales in nearby galaxies. ApJ971, 178. 10.3847/1538-4357/ad54bd

  • CrossRef
  • Google Scholar

• 256

  SuttonE. C.BlakeG. A.MassonC. R.PhillipsT. G. (1985). Molecular line survey of orion A from 215 to 247 GHz. ApJS58, 341–378. 10.1086/191045

  • CrossRef
  • Google Scholar

• 257

  SuzukiT.OhishiM.SaitoM.HirotaT.MajumdarL.WakelamV. (2018). The difference in abundance between N-bearing and O-bearing species in high-mass star-forming regions. ApJS237, 3. 10.3847/1538-4365/aac8db

  • CrossRef
  • Google Scholar

• 258

  TafallaM.MyersP. C.CaselliP.WalmsleyC. M.ComitoC. (2002). Systematic molecular differentiation in starless cores. ApJ569, 815–835. 10.1086/339321

  • CrossRef
  • Google Scholar

• 259

  TakanoS.MasudaA.HiraharaY.SuzukiH.OhishiM.IshikawaS.-I.et al (1998). Observations of (̂13)C isotopomers of HC_(3)N and HC_(5)N in TMC-1: evidence for isotopic fractionation. A&A329, 1156–1169.

  • Google Scholar

• 260

  TanJ. C.KongS.ButlerM. J.CaselliP.FontaniF. (2013). The dynamics of massive starless cores with ALMA. ApJ779, 96. 10.1088/0004-637X/779/2/96

  • CrossRef
  • Google Scholar

• 261

  TanJ. C.BeltránM. T.CaselliP.FontaniF.FuenteA.KrumholzM. R.et al (2014). “Massive star formation,” in Protostars and planets VI. Editors BeutherH.KlessenR. S.DullemondC. P.HenningT.149–172. 10.2458/azu_uapress_9780816531240-ch007

  • CrossRef
  • Google Scholar

• 262

  TanJ. C.KongS.ZhangY.FontaniF.CaselliP.ButlerM. J. (2016). An ordered bipolar outflow from a massive early-stage core. ApJL821, L3. 10.3847/2041-8205/821/1/L3

  • CrossRef
  • Google Scholar

• 263

  TangM.QinS.-L.LiuT.ZapataL. A.LiuX.PengY.et al (2024). A survey of sulfur-bearing molecular lines toward the dense cores in 11 massive protoclusters. ApJS275, 25. 10.3847/1538-4365/ad7df0

  • CrossRef
  • Google Scholar

• 264

  TaniguchiK.SaitoM.SridharanT. K.MinamidaniT. (2018). Survey observations to study chemical evolution from high-mass starless cores to high-mass protostellar objects. I. HC₃N and HC₅N. ApJ854, 133. 10.3847/1538-4357/aaa66f

  • CrossRef
  • Google Scholar

• 265

  TaniguchiK.SaitoM.SridharanT. K.MinamidaniT. (2019). Survey observations to study chemical evolution from high-mass starless cores to high-mass protostellar objects. II. HC₃N and N₂H⁺. ApJ872, 154. 10.3847/1538-4357/ab001e

  • CrossRef
  • Google Scholar

• 266

  TaniguchiK.HerbstE.MajumdarL.CaselliP.TanJ. C.LiZ.-Y.et al (2021). Carbon chain chemistry in hot-core regions around three massive young stellar objects associated with 6.7 GHz methanol masers. ApJ908, 100. 10.3847/1538-4357/abd6c9

  • CrossRef
  • Google Scholar

• 267

  TaniguchiK.GoraiP.TanJ. C. (2024). Carbon-chain chemistry in the interstellar medium. Ap&SS369, 34. 10.1007/s10509-024-04292-9

  • CrossRef
  • Google Scholar

• 268

  TaquetV.CharnleyS. B.SipiläO. (2014). Multilayer formation and evaporation of deuterated ices in prestellar and protostellar cores. ApJ791, 1. 10.1088/0004-637X/791/1/1

  • CrossRef
  • Google Scholar

• 269

  TerceroB.CernicharoJ.PardoJ. R.GoicoecheaJ. R. (2010). A line confusion limited millimeter survey of orion KL. I. sulfur carbon chains. A&A517, A96. 10.1051/0004-6361/200913501

  • CrossRef
  • Google Scholar

• 270

  ThompsonM. A.GibbA. G.HatchellJ. H.WyrowskiF.PillaiT. (2005). “SCAMPS: the SCUBA massive precluster survey,” in ESA special publication. Editor WilsonA.577 (Noordwijk, Netherlands: ESA Publications Division), 425–426.

  • Google Scholar

• 271

  ThorwirthS.WinnewisserG.MegeathS. T.TieftrunkA. R. (2003). “The twin-core system NGC 6334 I and I(N): high mass stars and stellar clusters in the making,” in Galactic star Formation across the stellar mass spectrum. Astronomical Society of the Pacific conference series. Editors De BuizerJ. M.van der BliekN. S.257–260.

  • Google Scholar

• 272

  Treviño-MoralesS. P.PilleriP.FuenteA.KramerC.RoueffE.González-GarcíaM.et al (2014). Deuteration around the ultracompact HII region monoceros R2. A&A569, A19. 10.1051/0004-6361/201423407

  • CrossRef
  • Google Scholar

• 273

  TurnerB. E. (1989). A molecular line survey of sagittarius B2 and Orion-KL from 70 to 115 GHz. I. The observational data. ApJS70, 539. 10.1086/191348

  • CrossRef
  • Google Scholar

• 274

  TurnerB. E. (1991). A molecular line survey of sagittarius B2 and Orion–KL from 70 to 115 GHz. II. Analysis of the data. ApJS76, 617. 10.1086/191577

  • CrossRef
  • Google Scholar

• 275

  van der WaltS. J.KristensenL. E.JørgensenJ. K.CalcuttH.ManigandS.el AkelM.et al (2021). Protostellar interferometric line survey of the Cygnus X region (PILS-Cygnus). First results: observations of CygX-N30. A&A655, A86. 10.1051/0004-6361/202039950

  • CrossRef
  • Google Scholar

• 276

  van DishoeckE. F.KristensenL. E.MottramJ. C.BenzA. O.BerginE. A.CaselliP.et al (2021). Water in star-forming regions: physics and chemistry from clouds to disks as probed by Herschel spectroscopy. A&A648, A24. 10.1051/0004-6361/202039084

  • CrossRef
  • Google Scholar

• 277

  van DishoeckE. F.TychoniecŁ.RochaW. R. M.SlavicinskaK.FrancisL.van GelderM. L.et al (2025). JWST observations of young protoStars (JOYS): overview of program and early results. :10.48550/arXiv.2505.08002

  • CrossRef
  • Google Scholar

• 278

  van GelderM. L.TaboneB.TychoniecŁ.van DishoeckE. F.BeutherH.BoogertA. C. A.et al (2020). Complex organic molecules in low-mass protostars on solar system scales. I. Oxygen-bearing species. A&A639, A87. 10.1051/0004-6361/202037758

  • CrossRef
  • Google Scholar

• 279

  VastelC.QuénardD.Le GalR.WakelamV.AndrianasoloA.CaselliP.et al (2018). Sulphur chemistry in the L1544 pre-stellar core. MNRAS478, 5514–5532. 10.1093/mnras/sty1336

  • CrossRef
  • Google Scholar

• 280

  VasyuninA. I.HerbstE. (2013a). A unified monte carlo treatment of gas-grain chemistry for large reaction networks. II. A multiphase gas-surface-layered bulk model. ApJ762, 86. 10.1088/0004-637X/762/2/86

  • CrossRef
  • Google Scholar

• 281

  VasyuninA. I.HerbstE. (2013b). Reactive desorption and radiative association as possible drivers of complex molecule formation in the cold interstellar medium. ApJ769, 34. 10.1088/0004-637X/769/1/34

  • CrossRef
  • Google Scholar

• 282

  VasyuninaT.VasyuninA. I.HerbstE.LinzH. (2012). Chemical modeling of infrared dark clouds: the role of surface chemistry. ApJ751, 105. 10.1088/0004-637X/751/2/105

  • CrossRef
  • Google Scholar

• 283

  VasyuninaT.VasyuninA. I.HerbstE.LinzH.VoronkovM.BrittonT.et al (2014). Organic species in infrared dark clouds. ApJ780, 85. 10.1088/0004-637X/780/1/85

  • CrossRef
  • Google Scholar

• 284

  VidalT. H. G.WakelamV. (2018). A new look at sulphur chemistry in hot cores and corinos. MNRAS474, 5575–5587. 10.1093/mnras/stx3113

  • CrossRef
  • Google Scholar

• 285

  VitiS.CollingsM. P.DeverJ. W.McCoustraM. R. S.WilliamsD. A. (2004). Evaporation of ices near massive stars: models based on laboratory temperature programmed desorption data. MNRAS354, 1141–1145. 10.1111/j.1365-2966.2004.08273.x

  • CrossRef
  • Google Scholar

• 286

  VuittonV.YelleR. V.McEwanM. J. (2007). Ion chemistry and N-containing molecules in Titan’s upper atmosphere. Icarus191, 722–742. 10.1016/j.icarus.2007.06.023

  • CrossRef
  • Google Scholar

• 287

  WakelamV.HersantF.HerpinF. (2011). Sulfur chemistry: 1D modeling in massive dense cores. A&A529, A112. 10.1051/0004-6361/201016164

  • CrossRef
  • Google Scholar

• 288

  WalmsleyC. M.BachillerR.Pineau des ForêtsG.SchilkeP. (2002). Detection of FeO toward sagittarius B2. ApJL566, L109–L112. 10.1086/339694

  • CrossRef
  • Google Scholar

• 289

  WalshA. J.ThorwirthS.BeutherH.BurtonM. G. (2010). Mopra line survey mapping of NGC 6334I and I(N) at 3mm. MNRAS404, 1396–1414. 10.1111/j.1365-2966.2010.16353.x

  • CrossRef
  • Google Scholar

• 290

  WangK.ZhangQ.WuY.ZhangH. (2011). Hierarchical fragmentation and jet-like outflows in IRDC G28.34+0.06: a growing massive protostar cluster. ApJ735, 64. 10.1088/0004-637X/735/1/64

  • CrossRef
  • Google Scholar

• 291

  WangK.ZhangQ.TestiL.van der TakF.WuY.ZhangH.et al (2014). Hierarchical fragmentation and differential star formation in the galactic ‘snake’: infrared dark cloud G11.11-0.12. MNRAS439, 3275–3293. 10.1093/mnras/stu127

  • CrossRef
  • Google Scholar

• 292

  WangY.AudardM.FontaniF.Sánchez-MongeÁ.BusquetG.PalauA.et al (2016). Ongoing star formation in the protocluster IRAS 22134+5834. A&A587, A69. 10.1051/0004-6361/201526637

  • CrossRef
  • Google Scholar

• 293

  WangY. X.ZhangJ. S.YuH. Z.WangY.YanY. T.ChenJ. L.et al (2023). A possible chemical clock in high-mass star-forming regions: N(HC₃N)/N(N₂H⁺)?ApJS264, 48. 10.3847/1538-4365/acafe6

  • CrossRef
  • Google Scholar

• 294

  WatanabeY.SakaiN.López-SepulcreA.FuruyaR.SakaiT.HirotaT.et al (2015). Spectral line survey toward the young massive protostar NGC 2264 CMM3 in the 4 mm, 3 mm, and 0.8 mm bands. ApJ809, 162. 10.1088/0004-637X/809/2/162

  • CrossRef
  • Google Scholar

• 295

  WatanabeY.NishimuraY.HaradaN.SakaiN.ShimonishiT.AikawaY.et al (2017). Molecular-cloud-scale chemical composition. I. A mapping spectral line survey toward W51 in the 3 mm band. ApJ845, 116. 10.3847/1538-4357/aa7ece

  • CrossRef
  • Google Scholar

• 296

  WatsonW. D. (1974). Ion-molecule reactions, molecule formation, and hydrogen-isotope exchange in dense interstellar clouds. ApJ188, 35–42. 10.1086/152681

  • CrossRef
  • Google Scholar

• 297

  WhiteG. J.ArakiM.GreavesJ. S.OhishiM.HigginbottomN. S. (2003). A spectral survey of the orion nebula from 455-507 GHz. A&A407, 589–607. 10.1051/0004-6361:20030841

  • CrossRef
  • Google Scholar

• 298

  WillisE. R.GarrodR. T.BellocheA.MüllerH. S. P.BargerC. J.BonfandM.et al (2020). Exploring molecular complexity with ALMA (EMoCA): complex isocyanides in sgr B2(N). A&A636, A29. 10.1051/0004-6361/201936489

  • CrossRef
  • Google Scholar

• 299

  WoodsP. M.OcchiogrossoA.VitiS.KaňuchováZ.PalumboM. E.PriceS. D. (2015). A new study of an old sink of sulphur in hot molecular cores: the sulphur residue. MNRAS450, 1256–1267. 10.1093/mnras/stv652

  • CrossRef
  • Google Scholar

• 300

  YangW.XuY.ChoiY. K.EllingsenS. P.SobolevA. M.ChenX.et al (2020). 44 GHz methanol masers: observations toward 95 GHz methanol masers. ApJS248, 18. 10.3847/1538-4365/ab8b5b

  • CrossRef
  • Google Scholar

• 301

  YangW.GongY.MentenK. M.UrquhartJ. S.HenkelC.WyrowskiF.et al (2023). ATLASGAL: 3 mm class I methanol masers in high-mass star formation regions. A&A675, A112. 10.1051/0004-6361/202346227

  • CrossRef
  • Google Scholar

• 302

  YuN.WangJ.-J. (2015). A molecular line study towards massive extended green object clumps in the southern sky: chemical properties. MNRAS451, 2507–2516. 10.1093/mnras/stv1058

  • CrossRef
  • Google Scholar

• 303

  ZaleskiD. P.SeifertN. A.SteberA. L.MuckleM. T.LoomisR. A.CorbyJ. F.et al (2013). Detection of E-Cyanomethanimine toward sagittarius B2(N) in the green bank telescope PRIMOS survey. ApJL765, L10. 10.1088/2041-8205/765/1/L10

  • CrossRef
  • Google Scholar

• 304

  ZavaryginE. O.WebbJ. K.DumontV.Riemer-SørensenS. (2018). The primordial deuterium abundance at z_abs = 2.504 from a high signal-to-noise spectrum of Q1009+2956. MNRAS477, 5536–5553. 10.1093/mnras/sty1003

  • CrossRef
  • Google Scholar

• 305

  ZernickelA.SchilkeP.SchmiedekeA.LisD. C.BroganC. L.CeccarelliC.et al (2012). Molecular line survey of the high-mass star-forming region NGC 6334I with Herschel/HIFI and the submillimeter array. A&A546, A87. 10.1051/0004-6361/201219803

  • CrossRef
  • Google Scholar

• 306

  ZinneckerH.YorkeH. W. (2007). Toward understanding massive star formation. ARA&A45, 481–563. 10.1146/annurev.astro.44.051905.092549

  • CrossRef
  • Google Scholar