An SiO Toroid and Wide-angle Outflow Associated with the Massive Protostar W75N(B)-VLA2

We have carried out Atacama Large Millimeter/submillimeter Array observations of the massive star-forming region W75N(B), which contains the massive protostars VLA1, VLA2, and VLA3. Particularly, VLA2 is an enigmatic protostar associated with a wind-driven H2O maser shell, which has evolved from an almost isotropic outflow to a collimated one in just 20 yr. The shell expansion seemed to be halted by an obstacle located to the northeast of VLA2. Here we present our findings from observing the 1.3 mm continuum and H2CO and SiO emission lines. Within a region of ∼30″ (∼39,000 au) diameter, we have detected 40 compact millimeter continuum sources, three of them coinciding with VLA1, VLA2, and VLA3. While the H2CO emission is mainly distributed in a fragmented structure around the three massive protostars, but without any of the main H2CO clumps spatially coinciding with them, the SiO is highly concentrated on VLA2, indicating the presence of very strong shocks generated near this protostar. The SiO emission is clearly resolved into an elongated structure (∼0.″6 × 0.″3; ∼780 au×390 au) perpendicular to the major axis of the wind-driven maser shell. The structure and kinematics of the SiO emission are consistent with a toroid and a wide-angle outflow surrounding a central mass of ∼10 M ⊙, thus supporting previous theoretical predictions regarding the evolution of the outflow. Additionally, we have identified the expected location and estimated the gas density of the obstacle that is hindering the expansion of the maser shell.


Dust continuum emission
Figure 1 shows the continuum image obtained towards the central part of the W75N(B) region.The millimeter cores MM1, MM2, and MM3 (previously observed with a beam of ≃ 1 ′′ -2 ′′ : Minh et al. 2010;Rodríguez-Kamenetzky et al. 2020;van der Walt et al. 2021), are now resolved into multiple compact millimeter sources.All known radio continuum sources at centimeter wavelengths in this region (labeled and indicated by crosses in Fig. 1; from Rodríguez-Kamenetzky et al. 2020), except Bc[E], Bc[W], Bd, and VLA4, have associated continuum emission at millimeter wavelengths, further reinforcing that VLA1, VLA2, VLA3, VLA [SW], and VLA [NE] are embedded protostars.The full width at half maximum (FWHM) of these continuum sources is unresolved by our beam.On the other hand, the fact that Bc[E], Bc[W], Bd, and VLA4 have no continuum emission at millimeter wavelengths (Fig. 1) gives support to the interpretation by Carrasco-González et al. (2010) and Rodríguez-Kamenetzky et al. (2020), through proper motion measurements, that these four objects are not embedded protostars, but obscured Herbig-Haro (HH) objects excited by the massive protostar VLA3.
In order to better distinguish weak compact mm continuum sources from extended emission present in the region (see Fig. 1), we obtained images that exclude baselines shorter than 200 kλ, thus filtering out large-scale structures, in a similar way as in Girart et al. (2018) and Busquet et al. (2019).With this baseline restriction, 40 unresolved continuum sources have been identified in a region of ∼30 ′′ (∼39,000 au) diameter centered on VLA2, and with flux densities in the range of ∼0.4-90 mJy beam −1 (Table A1, Fig. A1).We are proposing, in particular, that one of these detected sources (ID 32 in Table A1) is the powering source of the highly collimated bipolar jet that we have detected ∼6 ′′ (∼7,800 au) northeast of VLA2 (see Appendix C and Fig. C1), and that will be discussed in detail in a forthcoming paper.

H 2 CO line emission
The emission of the formaldehyde molecule is known to be a good tracer of dense gas (n[H 2 ] ≳ 5 × 10 5 − 10 6 cm −3 ) over a wide range of temperatures (T K ≃ 10 − 300 K) in the cores of molecular clouds (e.g., Mangum & Wootten 1993;Shirley 2015).The integrated emission image obtained from the H 2 CO [3(0,3)-2(0,2)] line in the velocity range V LSR ≃ −4 to +23 km s −1 (Fig. 2a) shows a clumpy structure over a plateau of relatively weak emission covering a region of ∼ 3 − 4 ′′ (∼ 3, 900 − 5, 200 au).None of the main H 2 CO clumps coincide with the position of any of the massive protostars VLA1, VLA2, or VLA3.The velocity field of the H 2 CO emission (Fig. 2b) shows that the greatest velocity differences are found along the northwest-southeast direction at angular scales of ∼3 ′′ (∼3,900 au), with blueshifted (V LSR ≃ +6 km s −1 ) and redshifted velocities (V LSR ≃ +13 km s −1 ) to the northwest and southeast of  A1 lists the positions of all the mm continuum sources detected in the region (see §3.1, §A, and Fig. A1).
VLA2, respectively (ambient velocity V LSR ≃ +10 km s −1 , e.g., Surcis et al. 2023).This direction also coincides with the orientation of the strongest H 2 CO emission (Fig. 2b) and consistent with the velocity field previously observed with the Submillimeter Array (SMA) and ALMA through different molecular species (e.g., H 2 CO, CS, SO 2 , CH 3 OH), at scales of ∼ 4 ′′ and angular resolutions of ∼ 1 ′′ − 2 ′′ (Minh et al. 2010;Rodríguez-Kamenetzky et al. 2020;van der Walt et al. 2021).In particular, van der Walt et al. (2021) suggested that the observed velocity shift is due to infalling gas onto a disk-like structure surrounding one of the massive protostars, but without being able to specify which one, probably due to an insufficient angular resolution in their observations (∼1 ′′ ).More recently, Zeng et al. (2023) conclude that the gas infall proceeds along the magnetic field lines, as the gravitational force dominates over the magnetic field.
We suggest that the distribution of the clumpy H 2 CO structure near but around VLA1, VLA2, and VLA3 could be due either to residual dense molecular gas fragments after the formation of these massive protostars, or to material still infalling toward the center of the region.This is supported by the fact that none of the main H 2 CO clumps observed around the massive protostars have an associated mm continuum source (see Figs. A1, 2a, Table A1).The clumps To obtain this moment-1 image, we applied a 4σ threshold (4 mJy beam −1 ) to the individual spectral channels.The white line traces the half-power contour of the MOM0 image in left panel.Ringed cross symbols for VLA1 and VLA2 indicate that these are the activity centers of strong H2O maser emission in the W75N(B) star-forming region (see, e.g., Surcis et al. 2023).The five-pointed star in both panels marks the location of the northeastern masers components in the VLA2 H2O maser shell, where the expansion appears to have halted (Surcis et al. 2023).
observed in H 2 CO have a relatively low mass of gas.For example, for the most intense clump (located ∼ 0.2 ′′ − 0.3 ′′ northeast of VLA2), assuming a density n(H 2 ) ≳ (5 × 10 5 ) − 10 6 cm −3 , and adopting a diameter size of 0.5 ′′ (650 au) from the emission image (Fig. 2a), we roughly estimate a molecular mass M (H 2 ) ≳ 4 − 8 × 10 −4 M ⊙ .In addition, given its proximity to VLA2, we consider it plausible that this particular dense gas could be related to the obstacle predicted by Surcis et al. (2023) to halt the expanding motions of the H 2 O masers observed to the northeast of VLA2 (see Fig. 2 and discussion in §4.2).
We also identify an H 2 CO disk-like structure centered on VLA3 with a velocity gradient in the east-west direction (see Fig. 2), therefore perpendicular to the VLA3 radio jet oriented in the north-south direction (Carrasco-González et al. 2010;Rodríguez-Kamenetzky et al. 2020).In addition, the H 2 CO line is seen in absorption toward VLA3 (absorption is also observed in SiO at this location; see Fig. B1).These results on VLA3 will be presented and discussed in a forthcoming paper.

General distribution of the SiO line emission
The spatial and kinematic distribution of silicon monoxide emission is completely different from that of formaldehyde discussed in the previous section ( §3.2).In Figure 3a we show the image of the integrated intensity of the SiO [v=0, emission line covering the same spatial region as that of H 2 CO in Figure 2a.From the comparison of these two images (Figs.2a and 3a), we see that, while the H 2 CO emission is distributed in several bright molecular clumps close, but not coinciding with any of the three massive protostars (VLA1, VLA2, VLA3), the SiO emission is strongly and almost exclusively concentrated around VLA2.This remarkable behavior related to the strong concentration of SiO emission on VLA2 had already been reported by Minh et al. (2010) through SMA observations, but without resolving its spatial distribution.With our ALMA observations we now resolve the SiO emission around VLA2 into a northwest-southeast elongated structure with a size of ∼ 0.6 ′′ × 0.3 ′′ (∼780 au×390 au; PA ≃ −53 • ), as seen in Figure 3b.The range of velocities with SiO emission observed within this elongated structure surrounding VLA2 is V LSR ≃ −20 to +28 km s −1 , which is significantly larger than that of H 2 CO emission throughout all the region containing the three massive protostars ( §3.2).The SiO emission is a well-known tracer of shocked gas, since the molecule is present in the gas phase in regions undergoing shocks with velocities of ≳ 25 km s −1 (e.g., Schilke et al. 1997).Moreover, the SiO lines are high-density tracers and, for instance, the critical density of excitation for the SiO(5-4) transition is ≃ 1.5 × 10 6 cm −3 (Huang et al. 2022).These extreme properties of the SiO excitation, together with the distribution of its emission described above, lead us to conclude the presence of very strong shocks around VLA2.We think that this is a consequence of the high outflow activity that VLA2 has, with very intense H 2 O masers expanding at several tens of km s −1 driven by strong repetitive, short-lived ionized winds from the central source (Torrelles et al. 2003;Kim et al. 2013;Carrasco-González et al. 2015;Surcis et al. 2023).All these properties observed in VLA2 are what make this massive protostar unique in comparison with the other two massive protostars in the region (VLA1 and VLA3).In §4.1 we discuss and interpret the SiO structure with its kinematics in terms of a molecular toroid surrounding VLA2, together with a wide-angle outflow.

The SiO toroid surrounding W75N(B)-VLA2
The elongated SiO structure (∼780 au×390 au, PA ≃ −53 • ; §3.3, Fig. 3b) that surrounds VLA2 is approximately perpendicular to the major axis of the expanding H 2 O maser shell (PA ≃ 45 • ; Surcis et al. 2014Surcis et al. , 2023)), which, in its turn, is shock-excited and driven by the VLA2 ionized jet (PA ≃ 65 • ; Carrasco-González et al. 2015).We interpret this SiO molecular structure as the toroid whose existence was hypothesized and modeled by Carrasco-González et al. (2015), as the collimating agent of an episodic, originally isotropic ionized wind associated with VLA2.These authors explain the observed evolution of a non-collimated outflow to a collimated outflow in VLA2 as due to a toroidal environmental density stratification, seen almost edge-on, with densities ∼5×10 7 cm −3 at distances of ∼28 au from the central massive protostar.This seems to be consistent with the fact that SiO(5-4) emission, which traces densities ≳ 1.5×10 6 cm −3 , is observed up to distances of ∼390 au from VLA2.Assuming this critical density of SiO as the minimum density for the entire toroid-like structure and given its observed size, a lower limit for the molecular mass M (H 2 ) ≳ 2 × 10 −3 M ⊙ is estimated.An independent estimate for a lower limit of the total mass (gas+dust) in this region can be obtained from the observed dust continuum emission of the unresolved VLA2 source (see §A), under the assumption of optically thin emission and using Eq.A2 (see §A).Adopting T d = 300 K (a reasonable value for gas temperatures near the massive object, considering the presence of strong H 2 O maser emission nearby, which implies that the gas should be at high-temperatures), we obtain a total mass (gas+dust) ≳ 5 × 10 −2 M ⊙ (≳ 51 M J ; Table A1).
Disk-like structures have been also reported in SiO emission surrounding other massive protostars (e.g., Gómez et al. 1999;Matthews et al. 2010;Maud et al. 2018).In particular, Maud et al. (2018), through ALMA 1.3 mm observations, report a rotating SiO disk and a disk wind from the massive protostar G17.64+0.16(probably an O-type star).The radius of the SiO disk structure in that object extends up to ∼600 au, similar to that of the SiO toroid observed around VLA2 (in terms of nomenclature we refer as toroids to disk-like structures with a size of ∼1,000 au).The very large velocity components observed in SiO within the VLA2 toroid (V LSR ≃ −20 to +28 km s −1 ) are consistent with the velocities of the shocks needed to produce SiO as proposed by Maud et al. (2018) for the disk (toroid) of G17.64+0.16.Similarly to G17.64+0.16, the shocks in VLA2 may originate at the inner edge and surface interface of the toroid by the fast expansion of the ionized wind and outflow (≳ 30 km s −1 , Surcis et al. 2014;Carrasco-González et al. 2015).
Figure 4a shows the position velocity (PV) diagram of the SiO emission along the major axis of the toroid-like structure (PA = −53 • ) with an averaging width of 0.14 ′′ (selected to optimize the visualization of the SiO kinematics).A velocity gradient of ∼6 km s −1 is observed in the inner and most intense part of the SiO structure at angular scales of ∼0.1 ′′ (∼130 au), with the more blueshifted velocities to the northwest and the more redshifted velocities to the southeast.This velocity shift is in the same sense as the velocity shift observed northwest-southeast in H 2 CO at larger scales (∼ 4 ′′ , Fig. 2b; see also Minh et al. 2010) and in other molecular species (e.g., SO 2 , CH 3 OH; Rodríguez-Kamenetzky et al. 2020).If the observed velocity shift of 6 km s −1 is due to rotating motions at the inner parts (∼130 au) of the toroid, those motions could be gravitationally bound by a central mass of ∼10 M ⊙ assuming Keplerian conditions, consistent with the expected mass of a B1-B0.5 espectral type star for VLA2 (Shepherd 2001).However, it seems clear that observations at submillimetre wavelengths, with higher angular resolution than currently available in this work, are needed to resolve in more detail the kinematic behavior of molecular gas in the internal parts of the toroid (at angular scales of ≲ 0.1 ′′ ).
The weakest SiO emission displays high blueshifted velocities, reaching up to V LSR ≃ −17 km s −1 , which appear as two rabbit-ear-shaped structures in the PV diagram along the major axis of the toroid (Fig. 4a), and located symmetrically on both sides (∼ ±0.2 ′′ ) of VLA2.Additionally, there is high redshifted velocity emission, with velocities up to V LSR ≃ +37 km −1 , which is more apparent in the PV diagram along the minor axis of the SiO structure with an averaging width of 0.6 ′′ (Fig. 4b).To explain these motions, which amount to approximately 27 km s −1 relative to the ambient velocity (V LSR = +10 km s −1 ; e.g., Surcis et al. 2023), and considering the distances of 0.2 ′′ (260 au), a central mass of around 210 M ⊙ would be required.However, such a mass is not observed, leading us to conclude that these high velocities traced by the weakest SiO emission represent unbound motions of an outflow originating from the outer parts (∼ ±0.2 ′′ ) of the toroid.
Figure 4c shows contour maps of the blueshifted (V LSR = −16.6 to −3.1 km s −1 ) and redshifted (V LSR = +23.9 to +37.4 km s −1 ) SiO integrated emission, with the blueshifted lobe being stronger than the redshifted one.This figure allows us to identify the source of the two rabbit-ear-shaped structures observed in Figure 4a as originating from the two peaks seen in the blueshifted lobe.These two peaks are separated by ∼0.4 ′′ , with the VLA2 source located between them.4c) with an averaging width of 0.14 arcsec.The first contour level and the increment step are 3σ, where σ = 1.4 mJy beam −1 (the rms of the diagram).The maximum of the diagram is 44 mJy beam −1 .The two 'rabbit-ear-shaped" structures tracing unbound blueshifted motions are indicated.These structures originate from two peaks seen in the high-velocity blueshifted lobe (see §4.1 and Fig. 4c).The dotted line indicates a possible inner velocity gradient in the SiO toroid.Upper right panel (b): same as upper left panel, but for the minor axis (PA ≃ −143 • ; dashed line in Fig. 4c) with an averaging width of 0.6 arcsec.
The first contour level is 3σ and the increment step is 6σ, where σ = 0.38 mJybeam −1 .The maximum of the diagram is 32 mJy beam −1 .Offset positions are with respect to the maximum SiO integrated intensity position (Fig. 3b).Lower panel (c): Integrated intensity image (moment of order 0) of the blueshifted and redshifted SiO in the high velocities ranges VLSR = −16.6 to −3.1 km s −1 and VLSR = +23.9 to +37.4 km s −1 , respectively.The first contour level is 4σ and the increment step is 6σ, where σ = 5 mJy beam −1 km s −1 .Symbols are as in Fig. 2.
It is worth noting that the SiO blueshifted lobe is slightly displaced towards the northeast of VLA2, while the redshifted lobe is slightly displaced towards the southwest (Fig. 4c).This displacement is consistent with the direction observed in the CO bipolar outflow at parsec scales (blueshifted to the northeast, redshifted to the southwest; e.g., Davis et al. 1998a,b;Shepherd et al. 2003), indicating that the SiO outflow is being incorporated into the larger-scale molecular outflow, similar to the case of G17.64+0.16(Maud et al. 2018).Furthermore, the morphological distribution of the two SiO outflow lobes (Fig. 4c), which spatially surround VLA2, indicates the presence of a wide-angle outflow associated with this massive protostar.
Finally, we have considered the possibility that the two rabbit-ear-shaped structures that can be distinguished in Figure 4a are due to the presence of two protostars embedded in the SiO toroid, one of which being VLA2.However, the fact that no other continuum source is detected, despite the high sensitivity of the data, together with the distribution of the two SiO outflow lobes around VLA2, makes this possibility very unlikely.

A wind emerging from the toroid, interacting with a high-density gas obstacle
As metioned above, Carrasco-González et al. (2015) showed that both, the VLA2 radio continuum source and the maser shell around it evolved in ∼20 years from a compact source into an elongated one in the northeast-southwest direction.They successfully modeled this behavior as due to the interaction of an episodic, short-lived isotropic ionized wind with a toroidal density stratification around the source.And we now find this toroidal structure, traced by the SiO emission.
Recently, Surcis et al. (2023) found that the northeast part of the shell traced by the H 2 O masers has stopped, and they suggest that this is due to the interaction of the shell with a high-density gas obstacle.An analytical and numerical simulation of such a situation by Cantó et al. (2006) show that, indeed, the interaction of a moving shell with a very dense medium can slow down the shell and stop its expansion completely.The detection of a strong H 2 CO clump to the northeast of VLA2 as reported in the present paper ( §3.2, Fig. 2a) seems to be consistent with the suggestion of Surcis et al. (2023).
Following this idea, we can estimate the required gas density of the obstacle by assuming a balance between the thermal pressure of the gas in the clump and the ram pressure of the wind that drives the shell.That is, where n is the density of the obstacle, T k its kinetic temperature, k the Boltzmann constant, Ṁw and V w the mass-loss rate and terminal velocity of the isotropic wind, respectively, and R the distance between the observed halted H 2 O masers and the powering source of the wind.Solving for the density of the clump we find, (2) Assuming for Ṁw and V w the values quoted above from Carrasco-González et al. (2015), together with T k = 300 K and R = 260 au (∼ 0.2 ′′ ; Surcis et al. 2023), we found n ≃ 3.5 × 10 7 − 3.0 × 10 8 cm −3 .This is a very high density, but consistent with the fact that this obstacle is located within a very high-density region, close to the observed H 2 CO clump to the northeast of VLA2 (Fig. 2a) and to the outer parts of the SiO toroid (Fig. 3b).In addition, we should also consider that the density we estimated for the obstacle to halt the expanding motions of the H 2 O masers to the northeast of VLA2 is likely to be an upper limit to the actual density.This is so because: 1) we have calculated the gas pressure (left expression of Eq. 1) from the kinetic temperature only, but if the clump has turbulent motions (which is most likely the case; see Surcis et al. 2023) the actual pressure will be higher; 2) the ram pressure (right expression in Eq. 1) assumes that the shock is normal to the direction of the wind, but any deviation from this condition will produce a lower ram pressure; 3) to estimate the density in Eq. 2, we have used the projected distance between the obstacle and the source, but the actual linear distance might be longer.

SUMMARY
Observations of the W75N(B) massive star-forming region using ALMA at a wavelength of 1.3 mm (beam = 0.17 ′′ × 0.08 ′′ , PA = −2 • ) are presented, focusing on the continuum, H 2 CO, and SiO lines.This region contains several massive protostars, including VLA1, VLA2, and VLA3, which are clustered within an area of ∼2 ′′ in size (∼2,600 au).The most relevant results are summarized as follows: • We detected 40 compact 1.3 mm continuum sources within an area of ∼ 30 ′′ diameter (∼ 39, 000 au), with three of them coinciding with VLA1, VLA2, and VLA3.The majority of these sources are new detections and are likely indicative of low-or intermediate-mass YSOs.
• The H 2 CO emission is distributed in a clumpy structure around VLA1, VLA2, and VLA3.In contrast, the SiO emission is highly concentrated almost exclusively on VLA2, implying the presence of intense shocks in the vecinity of this particular massive protostar.The SiO emission outlines an elongated structure that is perpendicular to the major axis of the wind-driven shell of H 2 O masers and its associated radio jet.We identify this structure as the toroidal structure that was previously theorized to explain the outflow's evolution from an almost isotropic outflow to a collimated one.
• The observations provide a plausible scenario in which we are observing an SiO toroid and a wide-angle outflow surrounding VLA2.The wide-angle SiO outflow originates from the outer regions of the toroid, being incorporated into the large-scale bipolar molecular outflow.The gas located in the inner regions of the toroid might be rotating, although more precise observations are required to confirm this.
• The location of the obstacle that was previously predicted to stop the expansion of the wind-driven H 2 O maser shell has been identified.It is situated between the outer parts of the SiO toroid and one of the main H 2 CO clumps detected nearby, toward the northeast of VLA2.Additionally, this determination involves estimating the necessary gas density to impede the shell's expansion toward the northeast.We report here the detection of an extremely highly collimated bipolar SiO jet located ∼5 ′′ (∼6,500 au) northeast of the three massive protostars (Fig. C1).This bipolar jet extends in the southeast-northwest direction, over an angular scale of ∼ 8 ′′ (∼ 10, 400 au) and velocities in the range V LSR ≃ −3 to +43 km s −1 .From the detected compact mm continuum sources in the W75N(B) region ( §3.1), we identify one of them (source ID 32 in Table A1) that is wellcentered between the two jet lobes (Fig. C1), as the most probable powering source, most likely a low-mass protostar.Since it is located in a region outside the main focus of our observations on VLA2, we will present and discuss the main properties of this remarkable bipolar SiO jet in a forthcoming paper.

Figure 1 .
Figure 1.ALMA 1.3 mm continuum emission image of the high-mass star-forming region W75N(B).The intensity range of the image has been saturated at 15 mJy beam −1 (beam = 0.17 ′′ × 0.08 ′′ , PA = −2 • ; rms ≃ 0.08 mJy beam −1 ) to distinguish the faintest continuum sources, given the high dynamic range of the image (∼1,100).The maximum of the image (coincident with VLA3) is 97 mJy beam −1 .The position of the radio continuum sources detected in the region at cm wavelengths, including the massive protostars VLA1, VLA2, and VLA3, are labeled and indicated by crosses, while dotted rectangles indicate the regions MM1, MM2, and MM3 where continuum emission was previously reported at 1.3 mm with ALMA, but with a beam of 1.73 ′′ × 0.86 ′′ , PA = −4 • (Rodríguez-Kamenetzky et al. 2020).TableA1lists the positions of all the mm continuum sources detected in the region (see §3.1, §A, and Fig.A1).

Figure 2 .
Figure 2. Left panel (a): Integrated intensity image (moment of order 0) of the H2CO [3(0,3)-2(0,2)] emission line in the velocity range VLSR ≃ −4 to +23 km s −1 .Contour levels are 10, 20, 30, 40, 50, 60, 70, 80, 90 per cent of the maximum integrated intensity of the image, 0.54 Jy beam −1 km s −1 .The synthesized beam is plotted as yellow ellipse at the upper right corner of the image.Right panel (b): velocity field image (moment of order 1).To obtain this moment-1 image, we applied a 4σ threshold (4 mJy beam −1 ) to the individual spectral channels.The white line traces the half-power contour of the MOM0 image in left panel.Ringed cross symbols for VLA1 and VLA2 indicate that these are the activity centers of strong H2O maser emission in the W75N(B) star-forming region (see, e.g.,Surcis et al. 2023).The five-pointed star in both panels marks the location of the northeastern masers components in the VLA2 H2O maser shell, where the expansion appears to have halted(Surcis et al. 2023).

Figure 3 .
Figure 3. Left panel (a): Integrated intensity image (moment of order 0) of the SiO [v=0, 5-4] emission line in the velocity range VLSR ≃ -20 to +28 km s −1 .The displayed field of view is the same as the one shown in Fig. 2a for H2CO.Right panel (b): zoom-in showing the integrated SiO emission around VLA2. Contour levels are 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, and 0.9 Jy beam −1 km s −1 .All symbols used in this figure are the same as those shown in Fig. 2.

Figure 4 .
Figure 4. Diagrams and image showing the various velocity components of the VLA2 SiO toroid.Upper left panel (a): Positionvelocity diagram of the SiO emission along the major axis of the elongated structure (PA ≃ −53 • ; dashed line in Fig.4c) with an averaging width of 0.14 arcsec.The first contour level and the increment step are 3σ, where σ = 1.4 mJy beam −1 (the rms of the diagram).The maximum of the diagram is 44 mJy beam −1 .The two 'rabbit-ear-shaped" structures tracing unbound blueshifted motions are indicated.These structures originate from two peaks seen in the high-velocity blueshifted lobe (see §4.1 and Fig.4c).The dotted line indicates a possible inner velocity gradient in the SiO toroid.Upper right panel (b): same as upper left panel, but for the minor axis (PA ≃ −143 • ; dashed line in Fig.4c) with an averaging width of 0.6 arcsec.The first contour level is 3σ and the increment step is 6σ, where σ = 0.38 mJybeam −1 .The maximum of the diagram is 32 mJy beam −1 .Offset positions are with respect to the maximum SiO integrated intensity position (Fig.3b).Lower panel (c): Integrated intensity image (moment of order 0) of the blueshifted and redshifted SiO in the high velocities ranges VLSR = −16.6 to −3.1 km s −1 and VLSR = +23.9 to +37.4 km s −1 , respectively.The first contour level is 4σ and the increment step is 6σ, where σ = 5 mJy beam −1 km s −1 .Symbols are as in Fig.2.

Figure B1 .
Figure B1.H2CO (left panel) and SiO (right panel) lines seen in absorption toward the continuum source VLA3.

Figure C1 .
Figure C1.Left panel: Integrated intensity image (moment of order 0) of the SiO bipolar jet found ∼ 5 ′′ northeast of VLA2 in the velocity range VLSR ≃ −3 to +43 km s −1 .The position of the 1.3 mm continuum source proposed to power this SiO jet ( §C, Table A1) is indicated by a cross and labeled as number 32.The nearby radio continuum source VLA[NE] (Rodríguez-Kamenetzky et al. 2020) is also labeled.Other 1.3 mm continuum sources found nearby of the SiO jet are also indicated by crosses.The synthesized beam is marked as a filled yellow ellipse at the upper right corner.Right panel: Velocity field image of the SiO jet (moment of order 1).