Millimeter methanol emission in the high-mass young stellar object G24.33+0.14

Abstract

1 Introduction

Recent high angular resolution observations have made significant contributions to the understanding of the high-mass star-formation process. Remarkable recent observational results are the discovery of accretion burst events in two high-mass young stellar objects (HMYSOs), S255IR NIRS 3 (Caratti o Garatti et al. 2017; Cesaroni et al. 2018; Liu et al. 2018, 2020; Szymczak et al. 2018b; Uchiyama et al. 2020; Hirota et al. 2021) and NGC 6334I-MM1 (Hunter et al. 2017, 2018, 2021; Brogan et al. 2018; MacLeod et al. 2018; Chibueze et al. 2021). Multi-wavelength observations in these two sources showed that the continuum emission from millimeter to infrared wavelengths increased due to the change in accretion luminosity of HMYSOs by a factor of 5.5 (S255IR NIRS3) and 16.3 (NGC 6334I-MM1). The estimated mass accretion rate is as high as >10⁻³ M_⊙ yr⁻¹ over a much shorter timescale (i.e., of the order of months or years) than the entire lifetime of HMYSOs (∼10⁴ yr). This can be interpreted as episodic mass accretion caused by disk fragmentation. Theoretical models suggest that HMYSOs undergo a number of such episodic mass accretion bursts during their formation (Meyer et al. 2017, 2019, 2021; Elbakyan et al. 2021). Despite their importance for the understanding of the high-mass star-formation process, observed examples of accretion bursts are still rare.

The accretion events in S255IR NIRS3 and NGC 6334I-MM1 were both accompanied by sudden flaring of the methanol (CH₃OH) maser line of the 5₁–6₀ A⁺ transition at 6.7 GHz (hereafter referred to as the 6.7 GHz methanol masers) as reported by many monitoring and mapping observations (Fujisawa et al. 2015; Moscadelli et al. 2017; Hunter et al. 2018; MacLeod et al. 2018; Szymczak et al. 2018b). Methanol masers are divided into two classes. Class I masers are collisionally pumped within the shocked gas by the interaction of the molecular outflows with the ambient gas at a variety of distances from protostars, while class II masers are radiatively excited by the strong infrared radiation field (Cragg et al. 2005) located only in close proximity of HMYSOs (Breen et al. 2013) tracing circumstellar gas of disks or outflow cavities (Menten 1991). The 6.7 GHz methanol emission is the representative transition of the class II methanol maser line. Thus, the above results suggest that the 6.7 GHz methanol maser flare is caused by a luminosity change due to an accretion burst; the increased accretion luminosity heats the surrounding dust which leads to increased emission of infrared photons, and hence increased 6.7 GHz methanol maser emission via the radiative pumping mechanism. In addition, it has also been reported that some pre-existing maser features disappear during the accretion bursts, due to the introduction of unfavorable conditions of maser pumping (Moscadelli et al. 2017; Hunter et al. 2018). Since maser emission is a non-linear phenomenon and is very sensitive to the physical environment, it can be used as a probe to detect the onset of accretion burst events.

With this in mind, a collaboration team “Maser Monitoring Organization (M2O)”¹ was organized in 2017 September during the International Astronomical Union (IAU) Symposium 336 “Astrophysical Masers” in Cagliari, Italy. One of the purposes of the M2O is to detect maser flares using a global collaborative network of radio telescopes. Indeed, the M2O activity succeeded in identifying the third candidate of a possible mass accretion event in an HMYSO, G358.93−0.03-MM1, in 2019 January based on the 6.7 GHz methanol maser flare (Sugiyama et al. 2019), followed by monitoring observations to obtain detailed maser light curves (MacLeod et al. 2019; Volvach et al. 2020).

Follow-up target-of-opportunity (ToO) observations of G358.93−0.03-MM1 have identified the possible candidate HMYSO undergoing the accretion burst from the Atacama Large Millimeter/submillimeter Array (ALMA) observation in Director’s Discretionary Time (DDT) (Brogan et al. 2019). Very-long-baseline interferometry (VLBI) imaging reveals a change in the radius of the shell-like maser distribution from 260 to 520 au within one month (Burns et al. 2020), implying that the location of the maser emitting region changes due to the heating from the central protostar. This finding, as well as the detection of a far-infrared luminosity increase (Stecklum et al. 2021), confirmed the presence of an accretion burst. In addition, a number of methanol maser lines were detected toward G358.93−0.03-MM1 including very rare and first-time detected transitions (Breen et al. 2019; Brogan et al. 2019; Chen et al. 2020), suggesting changes in the physical properties of the surrounding media as reported for the 6.7 GHz methanol masers in S255IR NIRS3 and NGC 6334I-MM1 (Moscadelli et al. 2017; Hunter et al. 2018). Another 6.7 GHz methanol maser flare which is proposed to have been caused by an accretion event has been reported in the HMYSO G323.459−0.079 (Proven-Adzri et al. 2019; MacLeod et al. 2021), as verified by the 2.2 μm near-infrared light curve (V. Wolf in preparation).

In addition, the 22.2 GHz water (H₂O) maser line of the 6_{1, 6}–5_{2, 3} transition (hereafter the 22.2 GHz water maser) is known to be highly variable, and sometimes to show strong flares associated with shocks induced by outflows in accretion burst sources (e.g., NGC 6334I-MM1; Brogan et al. 2018; MacLeod et al. 2018; Chibueze et al. 2021). According to variabilities in the 22.2 GHz water maser and mid-infrared continuum emission, an HMYSO M 17 MIR is also suggested to have experienced multiple accretion burst events (Chen et al. 2021). Thus, the number of possible candidates for accretion bursts in HMYSOs is increasing. Remarkably, the intermediate-mass YSO G107.298+5.639 shows joint periodic infrared and methanol maser flares (Stecklum et al. 2018), possibly due to binary-induced modulation of the accretion rate.

On 2019 September 5, the M2O reported a new 6.7 GHz methanol maser flare in the star-forming region G24.33+0.14 using the Torun 32 m radio telescope as shown in figure 1 (Wolak et al. 2019). The multi-frequency studies characterized the spatial structure and evolution of the methanol and water masers during this flare (A. Kobak, in preparation) . The integrated flux density increased from 20 to 100 Jy between ∼ 2019 June and October (between ∼58600 and 58800 in MJD), and the main contribution to the velocity-integrated flux density comes from two features near 113.5 km s⁻¹ and 115.4 km s⁻¹ (Wolak et al. 2019). This maser has previously been observed to show flaring activity between 2010 and 2012 (between ∼55500 and 56000 in MJD), in which almost all spectral features had shown similar flare profiles but with significant time delays between features/velocities (Wolak et al. 2018).

Fig. 1.

Light curves of G24.33+0.14. (a) Flux density of the 6.7 GHz methanol maser line integrated over the whole velocity range (Wolak et al. 2019, A. Kobak, in preparation). Observations were conducted with the Torun 32 m radio telescope. (b) Multi-epoch photometry data of WISE/NEOWISE Band 2 at the wavelength of 4.6 μm (Wright et al. 2010; Mainzer et al. 2011). Vertical lines indicate the dates of our ALMA observations in ALMA Cycle 3 and Cycle 6.

Open in new tabDownload slide

Long-term monitoring observations toward G24.33+0.14 revealed that the 6.7 GHz methanol maser shows variability with a strong flare lasting 200–400 d (Szymczak et al. 2018a). Since the WISE/NEOWISE infrared fluxes rose along with the onset of the recent maser flare (figure 1), an accretion burst origin seems likely. It is worth mentioning that G24.33+0.14 clearly shows a repeating nature of the 6.7 GHz methanol maser flare (figure 1). Monitoring observations of maser flares can provide insight into the physical mechanisms of the 6.7 GHz methanol maser flares and their periodicity.

Fortunately, G24.33+0.14 was observed in the pre-flare phase in ALMA Cycle 3 at |${\sim}{0{^{\prime \prime }_{.}}3}$| resolution in 2016 August, which corresponds to a linear scale of 2200 au at the estimated distance of 7.2 kpc (see subsection 1.1). Thus, this is an ideal target to compare the millimeter continuum emission and molecular lines both before and during the flare phases. In this paper, we report observational studies of G24.33+0.14 using ALMA Cycle 3 and newly obtained ALMA Cycle 6 DDT observations at Band 6 (230 GHz or λ = 1.4 mm).

1.1 Target source

G24.33+0.14 is an HMYSO located at an estimated distance of D = 7.20 ± 0.76 kpc (Reid et al. 2016) according to their Bayesian distance estimator.² It has been identified as a hot molecular core in an infrared dark cloud (IRDC) associated with a compact millimeter core MM1 observed with the Plateau de Bure Interferometer (PdBI) of Institut de Radioastronomie Millimétrique (IRAM) and the Submillimeter Array (SMA) at resolutions of 1″–2″ and |${0{^{\prime \prime }_{.}}6}$|⁠, respectively (Rathborne et al. 2007, 2011). The estimated luminosity, diameter, and mass of the core are 10^4.8 L_⊙, 0.0128 pc, and 34 M_⊙ (Rathborne et al. 2011).³ The spectral energy distribution (SED) from millimeter to mid-infrared wavelengths at lower resolution gives a similar luminosity of 10^4.7 L_⊙ and a larger clump mass of 3088 M_⊙ (Sánchez-Monge et al. 2013),⁴ which is also consistent with the presence of (a) HMYSO(s). Single-dish mapping observations of the SiO and HCO⁺ lines with the IRAM 30 m telescope with resolutions of 28″ at 90 GHz and 10″ at 230 GHz revealed a molecular outflow along the north–south or north-west–south-east direction in which the northern lobe shows a redshifted component (Sánchez-Monge et al. 2013). G24.33+0.14 is associated with an EGO (extended green object) detected by SPITZER with emission at 4.5μm of shocked H₂ molecules (Cyganowski et al. 2008; Chambers et al. 2009).

The 6.7 GHz methanol masers were observed using the Australia Telescope Compact Array (ATCA) to measure their accurate positions (Caswell 2009; Xu et al. 2009; Breen et al. 2015) at absolute positional accuracy of |${\sim}{0{^{\prime \prime }_{.}}4}$|⁠. The 6.7 GHz methanol masers were also imaged by the European VLBI Network (EVN; Bartkiewicz et al. 2016), and the Jansky Very Large Array (JVLA; Hu et al. 2016) in the pre-flare phase. There are three spatial groups of maser features where the size of the distributions is |${\sim}{0{^{\prime \prime }_{.}}5}$| or 3600 au at a distance of 7.2 kpc. Another class I methanol maser emission at 95.2 GHz (8₀–7₁ A⁺ transition) was detected, but the spatial distribution has not yet been determined (Chen et al. 2011).

The 22.2 GHz water masers were detected in the pre-flare phase (Caswell & Green 2011; Cyganowski et al. 2013). After the onset of the 6.7 GHz methanol maser flare, this source was observed with ATCA, and the 22.2 GHz water, the 23.4 GHz methanol maser (10₁–9₂ A⁻) and the 4.8 GHz H₂CO maser (1_{1, 0}–1_{1, 1}) were also detected in 2019 November (McCarthy et al. 2022). The 23.4 GHz methanol maser and the 4.8 GHz H₂CO maser are the fourth and 11th example of these rare maser transitions, respectively, and hence they could suggest maser pumping during the accretion event similar to G358.93−0.03-MM1 (Breen et al. 2019; Brogan et al. 2019; Chen et al. 2020). The positions of the H₂O and H₂CO masers are coincident with those of the 6.7 GHz methanol maser within their astrometric accuracy (⁠|${0{^{\prime \prime }_{.}}5}$| and 3″ in right ascension and declination, respectively), while the 23.4 GHz methanol masers are shifted to 7″ north and 2″ west (McCarthy et al. 2022).

2 Observations and data analysis

2.1 ALMA observations in Cycle 3 and Cycle 6

The observations in the pre-flare phase were conducted with ALMA in Cycle 3 (2015.1.01571.S), as a part of a survey of millimeter methanol maser lines in high-mass star-forming regions. Key parameters in the observations are listed in table 1. There were three individual observing sessions on 2016 August 15, 19, and 22 using 38, 39, and 40 12 m antennas, respectively. The baseline length ranged from 15.1 m to 1.5 km in each of the three sessions and the corresponding uv coverage is shown in figure 2. The tracking center position of G24.33+0.14 was taken to be |$\rm {RA} = {18^{\rm h}35^{\rm m}07{_{.}^{\rm s}}800}$| and |$\rm {Dec} = -{07^{\circ }35^{\prime }06.^{\prime \prime }00}$| (J2000.0). The net on-source time on G24.33+0.14 was 18 min, comprising 18 1 min scans (six scans in each session). Amplitude calibration was obtained by observing J1733−1304 or J1924−2914. As discussed in the Appendix, the absolute flux calibration accuracy is estimated to be |$10\%$|⁠. Bandpass and phase calibrators were J1751+0939 and J1851+0035, respectively. The ALMA Band 6 receiver was used to cover portions of the frequency range from 213 GHz to 231 GHz (see table 1 for details). The primary beam size of the 12 m antenna is 25″–27″ at the observing frequency.

Fig. 2.

uv coverage of the ALMA observations carried out in Cycle 3 (purple) and Cycle 6 (green). Small and large black circles indicate the uv range used in the imaging between 25 kλ and 1000 kλ, respectively.

Open in new tabDownload slide

Just 20 days after the onset of the maser flare, additional data was obtained in Cycle 6 (2018.A.00068.T) as an ALMA DDT project. Details of observations are also summarized in table 1. Observations were conducted from 2019 September 25 to 26 consisting of two sessions. The number of 12 m antennas was 45, with baseline lengths from 15.1 m to 2.5 km. The uv coverage in the observation, along with that in Cycle 3, is plotted in figure 2. Based on the previous observations in Cycle 3, the tracking center position of G24.33+0.14 was shifted to the continuum peak of |$\rm {RA} = {{18^{\rm h}35^{\rm m}08{_{.}^{\rm s}}140}}$| and |$\rm {Dec} = -{07^{\circ }35^{\prime }04.^{\prime \prime }15}$| (J2000.0) by (⁠|${+5.^{\prime \prime }06}$|⁠, |${+1.^{\prime \prime }85}$|⁠) with respect to that in the Cycle 3 observations. The net on-source time was 96 min, consisting of 12 8 min scans (six scans in each session). Amplitude and bandpass calibration were performed by observing J1924−2914 and phase calibration was performed using J1832−1035. The ALMA Band 6 receiver was used to cover frequency ranges similar to those employed in the Cycle 3 observations but at higher spectral resolution, and hence narrower bandwidths (for details see table 1).

2.2 Data analysis

For both the Cycle 3 and Cycle 6 projects, the pipeline script provided by the EA-ARC (East-Asian ALMA Regional Center) was used to calibrate the visibility data. Synthesis imaging and self-calibration were done using the CASA (Common Astronomy Software Applications) software package (McMullin et al. 2007) versions 4.7.0 and 5.6.1 for Cycle 3 and Cycle 6 data, respectively. The pipeline versions employed in the data analysis were r38377 and r42866, respectively. Continuum images were produced using line-free channels in all spectral windows (spws) and were used in self-calibration. Both phase and amplitude solutions were determined in self-calibration, and these solutions were applied to both the continuum and continuum-subtracted spectral line data. The spectral line images were made by smoothing the velocity resolution to 1.5 km s⁻¹ and 0.5 km s⁻¹ for Cycle 3 and Cycle 6 data, respectively, to investigate velocity structures of detected methanol transitions. However, image cubes were made using the same velocity resolution of 1.5 km s⁻¹ to compare the Cycle 3 and Cycle 6 results. In order to compare the flux densities and peak intensities between Cycle 3 and Cycle 6 as accurately as possible, a common uv range of (25–1000) kλ was employed (figure 2), for both the Cycle 3 and Cycle 6 data in the synthesis imaging. The shortest uv distance on 25 kλ corresponds to the spatial scale of 8″. Using a Briggs robust weighting parameter of 0.5, the resultant beam sizes of the continuum images were |${0{^{\prime \prime }_{.}}26}\times {0{^{\prime \prime }_{.}}24}$| with a position angle of −87° and |${0{^{\prime \prime }_{.}}22}\times {0{^{\prime \prime }_{.}}17}$| with a position angle of −74° for Cycle 3 and Cycle 6 data, respectively. These were convolved to the same beam size of |${0{^{\prime \prime }_{.}}3}$| using the same velocity resolution of 1.5 km s⁻¹ in the following discussion. Part of the image analysis and plotting was done using the National Radio Astronomy Observatory (NRAO) Astronomical Image Processing System (AIPS) software package (Greisen 2003).

For the ALMA Cycle 3 and Cycle 6 data, a visibility amplitude calibration error is known to impact some observations.⁵ This was caused by the combined effect of correlator spectral normalization and Tsys (system noise temperature) calibration. We checked the possible effect of this amplitude calibration error and found that the differences are less than |$1\%$| (typically |$0.1\%$|–|$0.5\%$|⁠) for both continuum and target spectral lines in both Cycle 3 and Cycle 6 observations. Thus, the effect on our analysis is negligible and we have utilized the originally delivered data before correction of this amplitude calibration error.

3 Results

3.1 Continuum emission

Figure 3 depicts the continuum emission measured in ALMA Band 6 in both Cycle 3 and Cycle 6 using the same uv ranges and convolved beam size of |${0{^{\prime \prime }_{.}}3}$|⁠. The continuum emission shows almost the same spatial structure with marginally resolved peak emission and a diffuse component extending toward the eastern side of the core. The maps reveal multiple continuum components. We conducted multiple 2D Gaussian fitting to the images of both Cycle 3 and Cycle 6 data. As a result, we identify three continuum peaks, C1, C2, and C3. Their properties are summarized in table 2.

Fig. 3.

ALMA Band 6 continuum (λ = 1.4 mm) images for (a) Cycle 3, (b) Cycle 6, and (c) their differential image. Contour levels are −4, 4, 8, 16, ..., 512 times the rms noise level of 0.45 mJy beam⁻¹ and 0.32 mJy beam⁻¹ for the Cycle 3 and Cycle 6 data, respectively. Dashed contours indicate negative levels. The rms noise level of the differential image is 0.55 mJy beam⁻¹. All panels employ the same color scale. Magenta crosses indicate the positions of the continuum peaks, C1, C2, and C3. Yellow X symbols represent the positions of the 6.7 GHz methanol maser cloudlets observed with the VLBA and a gray square indicates the position of the flaring maser feature (A. Kobak in preparation). The (0, 0) position corresponds to |$\rm {RA}={{18^{\rm h}35^{\rm m}08{_{.}^{\rm s}}140}}$| and |$\rm {Dec}=-{07^{\circ }35^{\prime }04.^{\prime \prime }15}$| (J2000.0) .

Open in new tabDownload slide

All the components including the marginal structure of C3 can be fitted consistently in both Cycle 3 and Cycle 6 data. The separations of C2 and C3 with respect to C1 are 5000 au (⁠|${0{^{\prime \prime }_{.}}7}$|⁠) and 4000 au (⁠|${0{^{\prime \prime }_{.}}6}$|⁠), respectively. The continuum positions measured in Cycle 6 are offset to the south by |${0{^{\prime \prime }_{.}}02}$|–|${0{^{\prime \prime }_{.}}04}$| compared with those in Cycle 3. These differences can be regarded as uncertainty in the astrometry of the ALMA data. The brightest peak, C1, carries |$80\%$| of the total continuum flux density, while the two weaker peaks, C2 and C3, are identified to the east and south of C1, respectively, with signal-to-noise ratios greater than 50. The position of C1 is coincident with the 6.7 GHz methanol maser feature at (⁠|${0{^{\prime \prime }_{.}}0361}$|⁠, |${0{^{\prime \prime }_{.}}2258}$|⁠) or the absolute coordinate of (⁠|$\rm {RA} = {{18^{\rm h}35^{\rm m}08{_{.}^{\rm s}}14243}}$|⁠, |$\rm {Dec} = -{07^{\circ }35^{\prime }03.^{\prime \prime }9242}$|⁠) measured through phase-referencing VLBI astrometry with the Very Long Baseline Array (VLBA) (A. Kobak in preparation). C2 is well resolved from C1, while C3 is seen as a spur extending from the southern part of C1. The nature of C3 is unclear because of the limited spatial resolution. After subtracting three Gaussian components corresponding to C1, C2, and C3 from the images, there still remain two residual components both in the Cycle 3 and Cycle 6 data, located at |${0{^{\prime \prime }_{.}}5}$| north and |${0{^{\prime \prime }_{.}}5}$| west of C1 with the maximum intensities of 19 mJy beam⁻¹ and 8 mJy beam⁻¹, respectively.

A previous interferometer observation at 1.3 mm with PdBI in 2005 reported a flux density of 284.0 mJy with an uncertainty of |$10\%$| (Rathborne et al. 2007). While the spatial resolution with the PdBI (⁠|${1.^{\prime \prime }5}\times {0{^{\prime \prime }_{.}}8}$|⁠) was much lower than that of the ALMA observations (⁠|${0{^{\prime \prime }_{.}}3}$|⁠), and hence only a single component was identified, the flux density measured with the PdBI is |$78\%$| of the ALMA result of the brightest component C1 in the pre-flare phase in 2013. The difference could be attributed to the larger missing flux in the PdBI data because its shortest baseline length of 48 m, which corresponds to the angular scale of 6″ (Rathborne et al. 2007), is slightly longer than that employed in our ALMA imaging of 35 m (25 kλ). Since the upper-side-band frequency of the PdBI of 219.560 GHz was only lower by |$0.2\%$| than our average frequency of 220.010 GHz for the three spws, the frequency difference will not significantly affect the flux mismatch.

The difference in the continuum emission between Cycle 3 and Cycle 6 is shown in panel (c) of figure 3. The difference map shows a peak at |${0{^{\prime \prime }_{.}}1}$| south of the brightest component, C1. Between the 2016 August and 2019 September ALMA observations, both the flux density and peak intensity of C1 appear to increase by a factor of 1.16 ± 0.01 (Cycle 6|$/$|Cycle 3). C3 also shows a flux difference between Cycle 3 and Cycle 6, although the measured flux densities could not be fully distinguished from those of C1. If we consider C3 as a part of a single source, the flux ratio for C1 + C3 between Cycle 3 and Cycle 6 is calculated at 1.20 ± 0.01, which is slightly higher than that for only C1 (1.16 ± 0.01). On the other hand, the flux density of the more isolated component C2 appears to be unchanged. As noted above, the residual components at the north and east of C1 are unchanged in intensities between Cycle 3 and Cycle 6, and hence there is no corresponding component in the differential image (figure 3c). We have checked for possible contamination or residual from line emission by constructing synthesized images of the continuum-subtracted visibility using only line-free channels and find that it is less than 0.015 mJy beam⁻¹ or |$0.2\%$|⁠. Considering additional contributions of absolute flux calibration uncertainties of |$10\%$| and a slight difference in the uv coverage between Cycle 3 and Cycle 6 data, the flux increase in C1 from Cycle 3 to Cycle 6 is marginal.⁶

While it is difficult to measure continuum opacity only from our ALMA Band 6 data, the beam-averaged brightness temperature of ∼50–60 K would suggest a moderate or low opacity given the gas temperature of >100 K inferred from the methanol line brightness temperatures (table 3). If the dust continuum emission is optically thin, the mass of each continuum source can be estimated using M = FD²/[κ_νB_ν(T_dust)], where F is the integrated flux density listed in table 2, D is the distance of 7.2 kpc, and B_ν(T_dust) is the Planck function at the dust temperature of T_dust. For consistency, we assumed a dust opacity of κ_ν = 1.0 cm² g⁻¹ at 230 GHz (Ossenkopf & Henning 1994), a gas-to-dust ratio of 100, and T_dust of 100 K, as employed in Rathborne, Simon, and Jackson (2007) and Rathborne et al. (2011). If we assumed a higher dust temperature of 150–300 K, then the estimated masses would be reduced by a factor of 1.5–3. On the other hand, the core masses of C2 and C3 could be underestimated if they are colder prestellar cores, given the non-detection of the high excitation methanol line toward these positions (see subsection 3.3). Assuming the lower temperature of 20–50 K for these cores, the masses would become larger by a factor of 2–5. The estimated core masses are listed in table 2. Even considering the above uncertainties, these cores are potential sites of future and on-going high-mass star-formation with masses of about 10 M_⊙ or larger.

There is no other continuum emission source within the field of view, with the 3σ mass detection limit of 1 M_⊙ (assuming a dust temperature of 20 K).

3.2 Molecular line spectra

Observed spectra within the common frequency range in both Cycle 3 and Cycle 6 are shown in figure 4. The spectra were extracted by integrating the 2″ × 2″ region around the brightest continuum peak C1 of the channel maps. G24.33+0.14 is known as a hot molecular core, because higher-excitation lines of various complex organic molecules (COMs) consisting of more than six atoms (Herbst & van Dishoeck 2009) have been detected (Rathborne et al. 2011). Detected lines are identified based on the JPL Molecular Spectroscopy database (Pickett et al. 1998)⁷ and the Cologne Database for Molecular Spectroscopy (CDMS; Müller et al. 2001),⁸ as listed in table 3 for methanol and in table 4 for other molecules. Four lines, not identified with known molecular lines, are indicated as “unidentified” in table 4. We will not include these unidentified lines in the following statistical discussion. In the present paper, we focus on eight selected methanol lines including candidates of the class I and class II masers as summarized in table 3, including one torsionally excited line (v_t = 1) and one ¹³C isotopologue. We note that there is one more identified CH₃OH line at 229.864121 GHz (19₅–20₄ A⁺). However, this line is not included in the detailed analysis because it is partially blended with the nearby ³⁴SO₂ line (see figure 4). Further detailed chemical properties of G24.33+0.14 along with other HMYSO samples (Baek et al. 2022) will be reported in forthcoming papers.

Fig. 4.

All spectra observed within the common frequency range in both Cycle 3 (purple) and Cycle 6 (green). Flux densities are integrated over 2″ × 2″ around the brightest continuum source C1. Common uv range, convolved beam size, and spectral resolution are employed to produce image cubes for comparison (see subsection 2.2). “U” means an unidentified line (see table 4).

Open in new tabDownload slide