MeerKAT and ALMA view of the AGAL045.804 − 0.356 clump

ABSTRACT

This study presents a detailed analysis of the GAL045.804 − 0.356 massive star-forming clump. A high-angular resolution and sensitivity observations were conducted using MeerKAT at 1.28 GHz and ALMA interferometer at 1.3 mm. Two distinct centimetre radio continuum emissions (source A and source B) were identified within the clump. A comprehensive investigation was carried out on source A, the G45.804 − 0.355 star-forming region (SFR) due to its association with Extended Green Object (EGO), 6.7 GHz methanol maser and the spatial coincidence with the peak of the dust continuum emission at 870 µm. The ALMA observations revealed seven dense dust condensations (MM1–MM7) in source A. The brightest (S_ν ∼ 87 mJy) and massive main dense core, MM1, was co-located with the 6.7 GHz methanol maser. Explorations into the kinematics revealed gas motions characterized by a velocity gradient across the MM1 core. Furthermore, molecular line emission showed the presence of an extended arm-like structure, with a physical size of 0.25 pc × 0.18 pc (∼ 50 000 au × 30 000 au) at a distance of 7.3 kpc. Amongst these arms, two arms were prominently identified in both the dust continuum and some of the molecular lines. A blue-shifted absorption P-Cygni profile was seen in the H₂CO line spectrum. The findings of this study are both intriguing and new, utilizing data from MeerKAT and ALMA to investigate the characteristics of the AGAL45 clump. The evidence of spiral arms, the compact nature of the EGO and < 2 km s⁻¹ velocity gradient are all indicative of G45.804 − 0.355 being oriented face-on.

1 INTRODUCTION

Massive star formation process occurs over a wide range of angular resolution. Associated phenomena over this dynamic range are seen in observations made over an extended range of wavelengths and in both continuum and spectral line observations. An all-encompassing approach is required to better understand the environment and processes in a star-forming region (SFR; Zinnecker & Yorke 2007). Such an approach can identify the kinematics within the dense molecular complexes (n > 5 × 10³ cm⁻³; Kurtz 2002), e.g. infall, outflow, rotation, expansion, etc. Advancements in interferometers make possible high-angular resolution and sensitivity observations of massive young stellar objects (MYSOs).

The AGAL045.804 − 0.356 star-forming clump (hereafter AGAL45 clump), is a bright infrared (IR) source located within the well-known Aquila star-forming complex (Rivera-Ingraham et al. 2010). It has a parallax distance of 7.3 kpc (Wu et al. 2019). Based on ¹³CO (1-0) line emission, Wienen et al. (2012) estimated the systemic velocity (V_sys) of the clump to be 59.46 ± 0.08 km s⁻¹ and older observations by Pandian, Menten & Goldsmith (2009) determined it to be 60.3 km s⁻¹. Using the James Clerk Maxwell Telescope (JCMT) spectral survey project [¹³CO/C¹⁸O Heterodyne Inner Milky Way Plane Survey (CHIMPS)], the V_sys of the clump derived from C¹⁸O (3–2) line emission was found to be 59.94 ± 0.33 km s⁻¹ (Rigby et al. 2016; Mège et al. 2021). Rivera-Ingraham et al. (2010) estimated the mass of the entire clump to be 450 ± 70 M_⊙ and the temperature 32.5 ± 1.6 K. However, Urquhart et al. (2018) derived a clump temperature of 26.5 K from the Spectral Energy Distribution (SED) fitting of the IR clump. From the IR fluxes, the clump luminosity function estimated by Rivera-Ingraham et al. (2010) was L_bol ∼ 1.9 × 10⁴ L_⊙, corresponding to a B0.5–B0 zero-age main sequence spectral-type star.

Fig. 1 shows a false colour composite image extracted from the Spitzer-Galactic Legacy Infrared Mid-Plane Survey Extraordinaire (Spitzer-GLIMPSE) survey. The AGAL45 clump contains two main IR components within the sub-millimetre dust emission from the Atacama Pathfinder Experiment (APEX) Telescope Large Survey of the Galaxy (ATLASGAL) at 870 µm (cyan contours). One of the IR components coincides with the Spitzer 4.5 µm emission (colour coded as false green colour) towards the SFR, G45.804 − 0.355. The second component (G45.806 − 0.35) has excess emission of the 8 µm (colour coded as red) and is associated with the IR bubble, MWP1G045810 − 003500S (Simpson et al. 2012). The G45.806 − 0.35 region is seen to have a bow-shock cometary-like structure in the Spitzer IR image (Rivera-Ingraham et al. 2010).

At 4.5 µm, Cyganowski et al. (2008) identified ‘fuzzy green’ objects in Spitzer images for different SFRs. Consequently, these sources were categorized as Extended Green Objects (EGOs). EGOs are likely to trace outflowing activities and are most times accompanied by the presence of class I methanol (CH₃OH) masers as well as water (H₂O) masers (Chen, Ellingsen & Shen 2009; Cyganowski et al. 2009; Voronkov et al. 2014; Urquhart et al. 2022). The AGAL45 clump is associated with the EGO G45.80 − 0.36 and has been classified as a ‘possible’ MYSO outflow candidate (Cyganowski et al. 2008). The detection of the 6.7 GHz class II methanol maser is evidence of the presence of a MYSO (Minier et al. 2003; Xu et al. 2008; Pandian et al. 2011; Breen et al. 2013). Other masers such as CH₃OH masers at 12.2 GHz, 44.4 GHz, and 95 GHz and 22 GHz H₂O masers have been detected towards the EGO (Chen et al. 2011; Cyganowski et al. 2013; Breen et al. 2016). This suggests that the clump is undergoing active star formation processes (Pandian et al. 2011; Yang et al. 2017).

This work presents intriguing findings of the AGAL45 clump, derived from high-angular resolution interferometric observations across various cloud scales, ranging from 7000 au (0.7 arcsec) to ≥ 73 000 au (10 arcsec). In addition, the study discusses the physical characteristics and gas kinematics of the surroundings, as well as the primary powering source within the G45.804 − 0.355 SFR.

2 OBSERVATIONS

Observational and archival data from various surveys used for this study are described as follows.

2.1 Centimeter data: MeerKAT observations

The AGAL45 clump was observed in 2018 as part of the SARAO MeerKAT 1.3 GHz Galactic Plane Survey (SMGPS). This was a high-sensitivity continuum survey of the Milky Way Galaxy. The observation was completed at a central frequency of 1.28 GHz, on the 64 (13.5 m) element, MeerKAT interferometer. The angular resolution was in the range of 7.5–8 arcsec, and continuum sensitivity had typical values between 10 and 15 µJy per beam. The MeerKAT data was reduced using the obit package developed by Cotton (2008). All the data analysis, the early results, and the survey itself are described in Goedhart et al. (2023). The |$\textrm {H}\scriptstyle \mathrm{I}$|/OH/Recombination (THOR) galactic plane survey is the deepest survey at 1–2 GHz in the first Quadrant of the Milky Way Galaxy. In comparison to the THOR, the accuracy of the flux calibration performed on the MeerKAT observation was around 4 per cent.

2.2 Millimeter data: ALMA band 6 observations of the EGO

The EGO source, G45.804 − 0.355, was observed in 2016 March at 1.3 mm using ALMA. The observation was part of the ALMA Cycle 3 science products (project code 2015.1.01312.S, PI: Gary A Fuller). 38 of the 12-m telescopes participated in the observation. Four spectral windows (SPWs 0–3) were covered at frequencies 224.23 to 226.11 GHz (SPW3), 226.12 to 227.99 GHz (SPW 2), 238.83 to 240.69 GHz (SPW 0) and 240.87 to 242.75 GHz (SPW 1). The continuum and molecular lines were observed simultaneously. The maximum baseline length was 460 m and the beam size was 0.7 arcsec. The angular resolution was 0.33 arcsec and the recovered large-scale structure was about 10 arcsec.

The calibrators used in the observation were J1751+0939 (Bandpass calibrator), Titan (flux calibrator), and J1922 + 1530 (phase calibrator). A single pointing was performed on the target source and the phase centre was (⁠|$\alpha _{\textnormal {2000}}$|⁠, δ₂₀₀₀) = (19^h16^m31.2^s, + 11°16′12.″0). Data calibration and imaging were performed using the ALMA pipeline and software package, Common Astronomy Software Application (casa), version 5.6.1–8 (McMullin et al. 2007). The imaging was done in Stokes I, using a Briggs weighting robust parameter of 0.5. The H|$\ddot{\textnormal {o}}$|gbom CLEAN algorithm was employed (Högbom 1974). Self-calibration was subsequently done on the calibrated data, improving the signal-to-noise ratio. The data was corrected for primary beam attenuation. A dust continuum emission map was obtained from the final averaged emission data cube after excluding channels showing line emission. The deconvolved continuum sensitivity was 0.446 mJy per beam.

The molecular line observation covered methanol (CH₃OH) transitions J = 20₋₂–19₋₃ at 224.699 GHz, 21₁–21₀ at 227.095 GHz, J = 14₁–13₂ at 242.446 GHz and a series of J = 5–4 lines. Other lines such as carbon monoxide [C¹⁷O (J = 2–1)] at 224.71437 GHz, carbon sulphide [C³⁴S (J = 5–4)], formaldehyde [H₂CO (J = 3–2)], cyano radical [CN (N = 2–1 J = 5/2–3/2 F = 5/2–3/2)], acetaldehyde [CH₃CHO (13_{−1, 13}–12_{−1, 12})], sulfur dioxide [SO₂ (J = 13_{2, 12}–13_{1, 13})], thioformaldehyde [H₂CS (J = 7–6)], and cyanoacetylene [HC3N (J = 25–24)] were observed. These lines were obtained from the final CLEANed data cubes after the continuum had been subtracted from the calibrated measurement set. The recovered velocity resolution for the molecular line observation was 1.41 km s⁻¹ and the frequency resolution was 1.128 MHz [Avison et al. (2023) and Asabre Frimpong et al. (submitted)].

2.3 Complemented data: continuum emission at IR wavelengths

The MeerKAT and ALMA observations were complemented with auxiliary data from the Spitzer-GLIMPSE survey at 3.6, 4.5, and 8 µm and ATLASGAL at 870 µm (Benjamin et al. 2003; Urquhart et al. 2014). The continuum sensitivity of the ATLASGAL observation was ∼50 mJy per beam and the angular resolution was 19.2 arcsec. The angular resolutions of the (Spitzer-GLIMPSE) bands at 3.6, 4.5, and 8 µm were 1.66, 1.72, and 1.98 arcsec, respectively.

3 RESULTS

3.1 MeerKAT radio continuum emission

The left panel of Fig. 2 shows an image of the radio continuum emission obtained from MeerKAT. Two radio sources were identified towards the clump, labelled source A for G45.804 − 0.355, centred at (⁠|$\alpha _{\textnormal {2000}}$|⁠, δ₂₀₀₀) = (19^h16^m30.912^s, + 11°16′13.″076) and source B for G45.806-0.35 centred at (⁠|$\alpha _{\textnormal {2000}}$|⁠, δ₂₀₀₀) = (19^h16^m30.1^s, + 11°16′28″). In the following subsections, each of the identified radio sources is described.

3.1.1 Properties of source A

Compared to Fig. 1, the position of source A overlaps with that of the EGO and the 6.7 GHz CH₃OH maser. The size of the full deconvolved width at half-maximum (FWHM) major axis is 7.27 ± 0.43 arcsec (∼ 0.26 pc) and FWHM minor axis size is 5.48 ± 0.43 arcsec (∼ 0.19 pc) at a position angle (PA) of 42.6° for G45.804 − 0.355. The peak intensity is 172.5 ± 1.5 µJy per beam, and the integrated flux is 281 ± 11 µJy.

3.1.2 Properties of source B

The peak flux density of source B is 770.4 ± 7.7 µJy per beam and its integrated flux density is 1.25 ± 0.02 mJy. The estimated Gaussian-fitted size is 10.06 arcsec (0.36 pc) and 9.85 arcsec (0.35 pc) at a PA of 114°.

3.2 Millimeter dust continuum emission

The ALMA 1.3-mm dust continuum results are presented in Fig. 2 (right panel). The Astrodendrogram technique using python package astrodendro (see Rosolowsky et al. 2008) was employed to identify the individual core components. A dendrogram graphically represents the hierarchical structure of a molecular clump or core. Bright components in the clump or core are classified as ‘leaves’ by the dendrogram while features with extended structures are classified as ‘branches’. The minimum intensity value was set to I_min (5× σ, where σ ∼ 0.39 mJy per beam). To identify each separate core entity as a detection, the minimum significance was ΔI_min ∼1σ. Only separate cores with 5σ and a minimum number of pixels of 55 were listed as ‘detections’. This minimized the selection of spurious cores or noise in the analysis.

Multiple dust cores were identified employing this technique. The dendrogram algorithm combined the MM1 and MM2 cores into a single-core structure, hence, Gaussian fitting was done in casa to derive the physical parameters of each core. Herewith, seven cores (labelled as MM1–MM7 in the right panel of Fig. 2) were identified over a region of 10 arcsec including the MM2 core which is located southeast of MM1.

The physical parameters such as deconvolved major and minor axes, the PA, the peak flux, and the integrated flux density of the ALMA dust continuum emission as well as the fitting errors are listed in Table. 1. The fluxes were determined from the ALMA primary beam corrected map. Three cores (MM1, MM2, and MM3) were found at the position of the MeerKAT radio continuum emission of source A; with MM1 having the highest peak flux. The MM2 core was poorly fitted, hence, only the peak flux was listed in Table. 1. Core MM7 had a relatively small size and could not be fitted with a 2D Gaussian. Despite core MM5 being close to the edge of the primary beam map (∼ 2.5 arcsec away from the edge of the beam), the signal-to-noise ratio was ∼ 10, hence, was included.

3.2.1 Derived properties of the individual cores

Each core mass was calculated by the following equation:

where S_ν is the integrated flux density at ν ∼225.119 GHz, the parallax distance d (∼ 7.3 kpc), and the ratio of the gas-to-dust mass R (assumed to be 100). The dust coefficient, κ_ν, is taken from Ossenkopf & Henning (1994); with values ranging between 0.64 to 1.0 cm² g⁻¹. The optical depth correction and dust temperature are denoted by C_τ and T_d, respectively. The and Planck function J_ν(T|$_{\textnormal {d}}$|⁠) is defined as

where k_B is the Boltzmann constant. The optical depth correction, C_τ, is estimated as

and the dust grain optical-depth (τ_dust) is derived as

From rotational diagram analysis, the excitation temperature (T_ex) was derived and used as the temperature of the dust. The analysis was performed using MM1 methanol molecular line transitions from the torsional ground state (ν_t = 0), under the assumption of local thermodynamic equilibrium. The lines that were considered in the analysis had upper energies ranging from 34 to 803 K and are shown in Fig. 3. For most of the transitions, the methanol line was thermal at around 280 K. For the MM1 core, the estimated excitation temperature was 242.98 ± 57.68 K and the column density (log N) was found to be 16.94 (±16.24) cm⁻².

Regarding cores MM3–MM6, a dust temperature value of 26.5 K was adopted (Urquhart et al. 2018). The derived core masses ranged from ∼ 2.0 M_⊙ to 9.2 M_⊙. From Table. 1, MM1 and MM2 cores were found to be more massive than the other cores using the high temperature estimated from the rotational diagram analysis. Adopting the dust temperatures, the beam-averaged column density for the individual cores was estimated using the equation,

where Ω is the solid angle of the observing beam, µ is the mean molecular weight of the gas, assumed to be 2.8 and m_H is the mass of hydrogen. All other symbols have the same definitions as stated above. It was noticed that beam-averaged column density for the MM1 dust core was higher compared to the other cores. However, deriving individual core temperatures would be useful in estimating the correct column density as well as the mass for each core. With core MM1, MM2 and MM3 being associated with the MeerKAT radio continuum emission of source A, the total and mean masses for the continuum core were ∼ 23.81 M_⊙ and 7.8 M_⊙, respectively. This implies that G45.804 − 0.355 is a massive dense core.

From the right panel image of Fig. 2, the dust cores are surrounded by large-scale extended emission. To distinguish between the morphology of the extended emission and the compact dust emission, we convolved the ALMA continuum emission with a larger beam (2.0 arcsec × 1.8 arcsec). This made it possible to recover the structure of the faint extended dust continuum. Two pronounced molecular gas arms were identified in the extended emission; one protruding to the North-West of the MM1 core and the other protruding to the south-east direction.

3.3 Molecular line emission from ALMA data

Molecular line emission transitions from CH₃OH, C¹⁷O, C³⁴S, H₂CO, and CN radical were used to further study the gas kinematics. These lines are useful for investigating both dense gas and large-scale structures as well as bulk motions of the gas surrounding the forming star. An overview of each of the molecular line species has been described in Sections 3.3.1–3.3.3.

3.3.1 Dense core tracer: CH₃OH

A number of CH₃OH lines were identified in MM1, some tracing the dense core and others tracing the enveloping gas. Lines such as CH₃OH (20_{−2, 19}–19_{−3, 17}) at 224.699 GHz, CH₃OH (21_{1, 20}–21_{0, 21}) at 227.095 GHz, CH₃OH (5_{−1, 4}–4_{−1, 3}) at 241.238 GHz and CH₃OH (5_{0, 5}–4_{0, 4}) at 241.268 GHz were found to be bright towards the MM1 core and had isolated line spectra, hence, had no or less contamination from nearby molecular lines. Using these lines, the derived average peak velocity of the source was found to be 60.9 ± 0.2 km s⁻¹.

Fig. 4 shows the MM1 core velocity-integrated intensity map (moment 0; left panel), velocity-weighted map (moment 1; middle panel), and velocity dispersion (moment 2; right panel) of various CH₃OH lines at different frequencies.

With respect to the moment 1 map, the velocity range for each molecular line was zoomed in between 58 and 63 km s⁻¹, making it easier to search for any associated velocity gradient. With the systemic velocity being 60.9 km s⁻¹, a slight velocity gradient was observed along the east-to-west direction. The velocity gradient can also be seen in the moment 1 map of CH₃CHO (bottom panel) towards MM1. A counterpart feature was found east of the MM1 core of the CH₃CHO image. This will not be discussed in this work.

3.3.2 Large-scale tracers: H₂CO (3–2), C¹⁷O (2–1), and C³⁴S (5–4)

Here, we refer to the H₂CO, C|$^{\textnormal {17}}$|O, and C³⁴S molecular lines without specifying the transitions since they have been stated in Section 2.2. These molecular lines trace large-scale gas flows and arms of extended emission as shown in Fig. 5. From the H₂CO image, the physical size of the extended emission is 0.25 pc × 0.18 pc (∼ 50 000 au × 30 000 au). The identified main arms have been labelled as Arm 1 (in the north-west direction) and Arm 2 (in the south-east direction). There were other mini-arms seen to be trailing in the east and west directions.

The exploration of the mini-arms was done using channel maps of the C³⁴S line emission as presented in Fig. 6. From the figure, multiple mini-arm gas flow patterns were detected around the MM1 core at velocities ranging from 57 to 61 km s⁻¹. The mini-arms were more noticeable at velocities 58.3 and 59.5 km s⁻¹. A high-velocity feature was also noticed in velocities ranging from 63.2 to 66.8 km s⁻¹. This new red-shifted feature had no detectable blue-shifted counterpart.

3.3.3 Cyano radical [CN (N = 2-1 J = 5/2–3/2 F = 5/2–3/2)]

The CN radical emission is one of the large-scale structure tracers. Fig. 7 shows the moment 0 map of the CN line at 226.87419 GHz. A combination of emission and absorption features was seen in the continuum map from the moment 0 map. However, the central dense core was found in absorption (see the left panel of Fig. 7). This study verified the reliability of the results to ensure the CLEANed images were not compromised by the calibration and imaging process. This was achieved by comparing the study’s final images with ones obtained from the science archive. The reprocessed data reassembled the ones from the archive, hence, the absorption features were not attributable to the CLEANing process. Furthermore, the CN line image was compared with the ALMA dust continuum image in Fig. 2. All the ALMA dust continuum cores were found in absorption in the CN line image.

The spectrum of the CN molecular line was extracted using a Gaussian profile fitting around the MM1 absorption feature in the right panel of Fig. 7. To find the peak of the absorption and emission features, a double Gaussian profile was fitted to the spectrum at velocities between 47.8 and 65.0 km s⁻¹. The FWHM of the dip was estimated to be ∼ 1.85 km s⁻¹ while the velocity was ∼ 60.30 km s⁻¹. This velocity is similar to the systemic velocity of G45.804 − 0.355 as estimated by Pandian et al. (2009). The emission features that were not fitted in the spectrum between velocities 20 to 47 km s⁻¹ and 66 to 91 km s⁻¹ correspond to other CN line transitions.

The dip in the spectrum is seen in the blue-shifted velocity. Based on the classifications of Erkal et al. (2022), the blue-shifted profile is known as a regular P-Cygni. The profile of the dip is comparable with the CN profiles identified towards various protostars by Paron et al. (2021).

In Fig. 8, the moment 1 map of the enveloping gas surrounding the MM1 core is shown. In the left panel, the velocity variation of the gas is noticeable while in the right panel, the velocity structure of the mini-arms is visible. From the figure, Arm 1 is red-shifted and Arm 2 is blue-shifted. The velocity patterns of the two images are similar, except for the presence of the red-shifted high-velocity feature in the CH₃OH line image. Investigating the velocity gradient in more detail will require higher resolution observation than the one used in this study.

4 DISCUSSIONS

4.1 Outflow orientation

According to Cyganowski et al. (2008), EGOs are good tracers of outflows emanating from MYSOs. Such systems have a central disc in the plane of the sky which has either a face-on or edge-on orientation. The implication of this is that the outflow lobe will be projected either away or towards an observer. Since the AGAL45 clump has an EGO towards source A, this study investigates the orientation of the outflow on a small-scale using SO₂ (J = 13_{2, 12}–13_{1, 13}) at 225.154 GHz, H₂CS (J = 7–6) at 240.549 GHz and HC3N (J = 25–24) at 227.419 GHz. Wing structure on either side of each spectrum can be inferred from the spectra shown in Fig. 9. There are no distinct separations between the peaks of the red and blue lobes presented in images of the SO₂ [J = 13(2,12)–13(1,13)] and H₂CS (J = 7–6) line emission in Fig. 10. This is an indication that the apparent disc of the young stellar object associated with G45.804 − 0.355 has a face-on geometry. The clump scale analysis of Yang et al. (2018) also showed no distinct separations of the red and blue wing outflow lobes.

From the study of Cyganowski et al. (2008), a typical EGO has an angular extent of a few to about 30 arcsec in size. The major and minor axis lengths of EGO G45.804 − 0.355 are 12.3 and 11 arcsec from the Spitzer image. The ratio of the major and minor length of the clump is 1.1. Other examples of EGOs associated with 6.7 GHz methanol maser and Class I methanol masers are listed in Table 2. From these examples, the samples of Cyganowski et al. (2009) show EGOs with extended and compact morphologies. For instance, G19.01 − 0.03 and G35.03 + 0.35 are well known EGOs with extended 4.5 µm emission (Cyganowski et al. 2009; Williams et al. 2022). The ratios of their major and minor axis linear lengths are 2.56 and 2.33, respectively. The elongated morphology of these EGOs implies a possible orientation where the disc of the outflow is observed edge-on. EGOs G22.04 + 0.22 and G18.67 + 0.03 have compact morphology similar to that of G45.804 − 0.355 at 4.5 µm. The ratio of their major and minor axes length are 1.06 and 1.4, respectively. Comparing the ratios listed in Table. 2, both the extended and compact EGOs show the full width of the outflow. However, the full length of the outflow is only seen in the extended EGO.

Morgan et al. (2021) additionally proposed that a maser spectrum displaying a bunny-hop light curve is likely to have a face-on orientation. The authors further suggested that such a maser source might have its peak near the systemic velocity, and could possess a relatively narrow line-width. In the case of G45.804 − 0.355, Olech et al. (2022) showed that at 6.7 GHz, the maser exhibits a bunny-hop light curve. The methanol maser spots spread over a velocity range of 55.67 to 70.79 km s⁻¹ (Pandian et al. 2011; Hu et al. 2016; Nguyen et al. 2022). The velocity component with the highest peak intensity has a velocity of 60.1 km s⁻¹ for the 6.7 GHz and 60.0 km s⁻¹ for the 12.2 GHz methanol maser. This value is close to the systemic velocity. Furthermore, the moment one map of CH₃CHO shown in Fig. 4 reveals a minimal velocity gradient (≤ 2 km s⁻¹) across the MM1 core in source A. In the case of a source having a face-on orientation, the anticipated outcome is the absence of a velocity gradient. However, the observed slight velocity gradient could suggest that the G45.804 − 0.355 is slightly inclined. Given that the 6.7 GHz methanol masers typically form within the disc where the protostar resides (Norris et al. 1993; Bartkiewicz et al. 2009), the 6.7 GHz Very Long Array (VLA/VLBI) observations from Hu et al. (2016) were utilized. The distribution of the maser spots was offset from the centre of the ALMA methanol moment 0 maps presented in Fig. 4. The maser spots were shifted to the reference position from Hu et al. (2016) observations. A Keplerian fitting from MacLeod et al. (2021) was applied to the maser spots located towards the centre of the methanol moment 0 maps. The best fit, with an enclosed central mass of 8 M_⊙, yielded an inclination angle of ∼0°.

4.2 The nature of the driving sources in AGAL45 clump

Gerner et al. (2014) proposed four phases of the evolution of high-mass star formation based on the chemical characteristics of their regions. These phases are the quiescent IR dark clouds (IRDC), high-mass protostellar objects (HMPO), warmer hot molecular cores (HMC) and the formation of ultra-compact H ii regions (UC H ii). Once a MYSO is formed, it evolves from a starless IRDC to the HMPO. The presence of a central protostar increases the temperature and density of the SFR, leading to the HMC stage. One of the signatures of a hot core is the abundance of methanol. The G45.804 − 0.355 MM1 core has a significant number of detected methanol lines, other complex molecules such as H₂CO and CH₃CN and excitation temperature of ∼ 243 K, suggesting a HMC phase.

MYSOs are deeply embedded sources and actively influence their environment by producing UV photons. An indirect way to search for these MYSOs in highly obscured environments is to search for radio continuum emission. The compact radio continuum emission detected towards the 6.7 GHz methanol maser allows us to investigate the powering source responsible for the maser emission. First, the ionization photon rate (i.e. the number of Lyman ionizing photons needed to sustain the ionization in the source) was estimated from the MeerKAT continuum emission shown as contours in the right panel of Fig. 2. The equation of the ionization photon rate is

where the distance to the source is denoted by d, S_ν is the source integrated flux at the observing frequency (ν ∼1.28 GHz) and T_e is the electron temperature; assumed to be 6343 K (Khan et al. 2022).

The derived ionizing photon rate for G45.804 − 0.355 (sources A) and G45.806-0.35 (source B) were found to be 1.42 ± 0.05 × 10⁴⁵ s⁻¹ and 6.33 ± 0.10 × 10⁴⁵ s⁻¹, respectively. Billington et al. (2019) estimated the logarithm of the bolometric luminosity for the AGAL45 clump to be 3.95 which means that the source is ionized by spectral type B1 star. From the ionizing photon rate estimate of the radio continuum emission for sources A and B (log(N_Ly) ≈ 45.15 s⁻¹ and 45.8 s⁻¹, respectively), there is an ionized star of spectral type B1(III)/B0.5(III) (Panagia 1973; Vacca, Garmany & Shull 1996).

In general, the uncertainty of the estimated integrated flux density of the MeerKAT radio continuum is |$\lt 4~{{\ \rm per\ cent}}$|⁠. The ionizing photon rate is dependent on the integrated flux density. The statistical deviation errors of the measured quantities of sources A and B, given an accuracy of 4 per cent are 1.42 ± 0.057 × 10⁴⁵ and 6.33 ± 0.25 × 10⁴⁵ s⁻¹, respectively.

4.3 The environment and properties of the central B-type star

This section discusses the size, age and dynamics of the surroundings of the central ionizing star. The size (radius) of an |$\textrm {H}\scriptstyle \mathrm{II}$|region can be determined using the derived ionizing photon rate for an optically thick nebulae (Tielens 2005). This is identified as the Str|$\ddot{\text{o}}$|mgren radius of the H ii region. It can be estimated using the equation of Khan et al. (2022),

where R_st is the str|$\ddot{\text{o}}$|mgren radius and |$n_{\mathrm{ H}_2}$| is the mean hydrogen density, assumed to be 10⁴ cm⁻³ in a dense region (Tielens & Hollenbach 1985; Rahner et al. 2017). For a star-forming molecular cloud, the hydrogen density range between |$n_{\mathrm{ H}_2}\sim 10^3$|-10⁶ cm⁻³. Within an H ii region where the temperature of the gas is about 10⁴ K, the maximum density of the gas is about 10⁴ cm⁻³. In a photo-dissociation region, the temperature ranges from 10 to 40 K and the density is 10³ cm⁻³ (Tielens & Hollenbach 1985).

The dynamical age of the |$\textrm {H}\scriptstyle \mathrm{II}$|region was estimated using

where t_dyn is the dynamical age of the H ii region and R_f is the final radius of the region for which the deconvolved sizes of 0.26 pc and 0.36 pc for sources A and B, were used respectively. The C_s is the isothermal sound speed of the H ii region. In the calculation, a sound speed of ∼10 km s⁻¹ through the ionized gas was adopted (Khan et al. 2022). The dynamical ages of sources A and B were found to be 2.02 ± 0.029 × 10⁵ and 2.38 ± 0.033 × 10⁵ yr. Given a calibration uncertainty of 4 per cent, the dynamical ages of sources A and B were 2.02 ± 0.08 × 10⁵ and 2.38 ± 0.09 × 10⁵ yr. Considering the uncertainties, the measurement has a low variability around each source’s central value. Source A is younger than source B by a factor of 0.85.

The environment of Source A (G45.804 − 0.355) includes an EGO which is known to have class I and II methanol masers and water masers. As mentioned in Section 1, class I methanol and water masers are indicators of outflows, most likely pumped by collision at the interface between molecular outflows and the surrounding ambient material. The 95 GHz class I methanol maser of G45.804 − 0.355 (BGPS6202) is offset from the position of the 6.7 GHz class II methanol maser (Yang et al. 2017). The detection of the Class I maser in the outflow region versus Class II masers within the emission of source A in Fig. 1 may imply that the gas of source B is shock-enhanced. Although, there is also a possibility that the class I methanol maser may not be associated with the 6.7 GHz methanol maser, a blue-shifted velocity dip is present in Fig. 7. This dip feature is an indication of an expanding shell of outflow material.

Spiral arms have been observed at both large-scale (∼0.6 pc) and small-scale within molecular clumps. For instance, Liu et al. (2015) identified clump scale mini-arms (size ∼ 0.1 pc) towards G33.92 + 0.11, a massive OB cluster-forming region at d ∼7.1 kpc. Through VLBI monitoring at a scale ranging from 50 to 900 au (< 0.005 pc), Burns et al. (2023) identified spiral arms traced by the distribution of maser spots in the high-mass protostar G358.93 − 0.03 (d ∼ 6.7 kpc). These spiral arms were extended, ranging from 50 to 900 au (< 0.005 pc) around the G358.93 − 0.03 protostar. It is worth noting that both G33.92 + 0.11 and G358.93 − 0.03 sources possess an almost face-on inclination, making them suitable for comparison with the arm-like structure observed in G45.804 − 0.355. Although the G45.804 − 0.355 ALMA 1.3 mm dust continuum unveils the presence of two distinct arms spiralling around the central massive dust core, our current study did not allow for an exploration of the velocity gradient along these individual arms. A more advanced observational approach with a higher resolution will be necessary to comprehensively examine and analyse the velocity gradient.

4.4 Spontaneous or triggered star formation?

The formation of new-generation stars and cores can be triggered by the influence of an expanding H ii bubble. However, most EGOs are associated with the peak of the ATLASGAL dust continuum and found close to IR bubbles of known H ii regions. G45.806 − 0.35 and G9.62 + 0.20E are examples of such regions. In Fig. 1, the G45.806 − 0.35 region (source B) is an IR bubble which has a cometary morphology and photon-dominated region traced by the 8µm IR emission. The EGO G45.804 − 0.355 (source A) is located southeast of the border of the source B H ii region. EGO G19.01 − 0.03 is a classical example of a source at the borders of an IR bubble which is undergoing triggered star formation (see fig. 2 of Tackenberg et al. 2013). In a triggered star formation scenario, it is expected that source B will trigger the formation of new stars at its borders.

The separation between source A and B is about 3.99 arcsec (∼ 0.14 pc). Dale, Haworth & Bressert (2015) suggested that the 2D separation between the nearest ionization front and the location of a new generation of stars is estimated to be a few tens of pc to ∼ 10⁻² pc. The value of our separation distance between the G45.806 − 0.35 and G45.804 − 0.355 regions falls within the region where triggered star formation occurs from the nearest ionization front (see fig. 5 of Dale et al. 2015).

The comparison of the dynamical ages of sources A and B (2.02 × 10⁵ yr and 2.38 × 10⁵ yr, respectively), the G45.805 − 0.355 region is younger and formed 3.6 × 10⁴ yr later after the G45.806 − 0.35 SFR. In our discussion above we present the evidence that source A was triggered after the formation of source B. However, the ALMA observation did not extend to G45.806 − 0.35 (source B), hence, we reserve any further discussions on the triggered star formation for a future detailed study of the AGAL45 clump.

5 CONCLUSIONS

This study has presented the millimetre and centimetre observation of the AGAL45 dust clump (AGAL045.805 − 0.356) using ALMA and MeerKAT interferometers. The gas kinematics from molecular line analysis has also been reported. The main results are as follows;

• From the MeerKAT observation, two radio continuum emission sources were detected at 1.28 GHz towards the dust clump. These emissions were labelled as source A for G45.806 − 0.35 and source B for G45.804 − 0.355.

• From the estimate of the ionizing photon rate and luminosity, the main powering source in the H ii region was found to be a B1(III)/B0.5(III) main-sequence type star.

• Comparing the dynamical ages of the radio sources (A and B), source A was considered to be younger and is in the hot molecular phase. The association of a 6.7 GHz maser and EGO at the position of source A, suggest that source A is the main powering source in the region and hosts an embedded YSO or handful of B1/B0.5 type stars.

• At 1.3 mm wavelength, the source A radio continuum is resolved into multiple dust components, including the massive dense core, MM1.

• Based on the velocity-channel maps of the C³⁴S line, mini-spiral arms were identified surrounding the MM1 core.

• There is no apparent separation between the red- and blue-shifted lobes in the ¹³CO integrated intensity map of Yang et al. (2018). The ALMA SO₂ and H₂CS molecular lines identified towards the MM1 dense core also show no clear separations between the blue and red lobes. This supports the interpretation that the G45.804 − 0.355 source is viewed face-on.

To further reveal the detailed dynamics of the sources within the clump and resolve the individual dust condensations will require complementary high-resolution observations better than the resolution used in this work. The data used in this study were analyzed using python packages such as aplpy, matplotlib, numpy, scipy, and pandas.

ACKNOWLEDGEMENTS

We extend our gratitude to the SARAO team for the SMGPS data. The MeerKAT telescope is operated by the South African Radio Astronomy Observatory, which is a facility of the National Research Foundation, an agency of the Department of Science and Innovation. We also thank the African Astronomical Society (AfAS) for the research seed grant. We would like to extend our appreciation to the Centre for Space Research, North-West University, and the Development in Africa with Radio Astronomy (DARA) for their financial support. A special thanks goes to Prof Jesús Martín-Pintado for his assistance with using the madcuba software. We would like to express our gratitude to Dr Gordon MacLeod and the anonymous reviewer for their insightful inputs and suggestions. This paper makes use of the following ALMA data: ADS/JAO.ALMA|$\#$|<2015.1.01312.S>. ALMA is a partnership of ESO (representing member states), NSF (USA), and NINS (Japan), together with the NRC (Canada), MOST and ASIAA (Taiwan), and KASI (Republic of Korea), in cooperation with the Republic of Chile. The Joint ALMA Observatory is operated by ESO, AUI/NRAO, and NAOJ.

DATA AVAILABILITY

The data sets utilized in this study were directly requested from the PIs and are accessible from the ALMA archival website¹ and the MeerKAT data archive website.² The data supporting the findings of this study can be obtained from the corresponding author upon reasonable request.

Footnotes

References