ALMA Observations of the Extraordinary Carina Pillars: A Complementary Sample

The HH 666 object, known as the axis of evil, was first reported by Smith et al. (2004) using ground-based optical and near-infrared (NIR) observations that revealed the bipolar flow emerging from the head of a dusty globule pointing in the direction of η Carinae and Tr 16. A protostar (listed as a Class I object) embedded within the dense globule is proposed as the possible driving source of the optical jet (detected at the center of the globule at the position in the sky of α = 10^h43^m513, δ = −59°55'212, (J2000), see Smith et al. (2004). In a later study, Smith et al. (2010) reported new Hα observations from the HST, corroborating the jet structure associated with the HH object, the extension of the jet component (∼45 or ∼3 pc), and the detection of a weak object located at the same position as the previous study. This object was also detected in Spitzer NIR observations (Smith et al. 2010)

In recent work, Reiter et al. (2015a) presented NIR spectroscopic studies of [Fe ii] obtained with the Folded-Port InfraRed Echelette spectrograph on the Baade/Magellan Telescope. These observations reveal that the [Fe ii] emission is tracing a highly collimated protostellar jet embedded in the dense pillar of gas and dust with an extension of about 40''. They also found a clear connection between the previously revealed jet-driving source and the extent of the collimated jet. Finally, their studies of [Fe ii] and Hα showed differences, especially in areas very close to the protostellar source, where the images in Hα did not reveal any emission.

In Figure 1, we present the optical image of the HST/ACS presented in Smith et al. (2010) overlaid with our resulting ALMA 1.3 mm continuum and line images of the globule HH 666. The optical image (green) corresponds to Hα emission and overlaps with the ALMA high-velocity CO(2−1) molecular emission (blue and red) and the 1.3 mm continuum emission (white contours). The CO emission is tracing the bipolar outflow very close to the exciting source, which moves at an LSR system velocity of −10 km s⁻¹. The angular size is 16, which at 2.3 kpc represents a size of 3680 au for HH 666.

The 1.3 mm continuum emission is only located in the axis of the optical jet detected in Hα at α = 10^h43^m515, $\delta =-59^\circ 55^{\prime} 21\buildrel{\prime\prime}\over{.} 1$ (J2000), perfectly matching the NIR source detected by Smith et al. (2004; see Figure 1) and reveals a compact source corresponding to an embedded structure in the central zone of the globule. We think that this object is composed of a dusty envelope and circumstellar disk corresponding to the driving source of the protostellar optical jet, and for this millimeter continuum emission, we obtained an integrated flux of 14.94 ± 0.57 mJy and an observed rms noise of 0.071 mJy (see Table 1) for the continuum image, which is very close to the theoretical noise value for this configuration (43 antennas), frequency (230 GHz), bandwidth (4.3 GHz), and integration time (12.5 minutes), which is 0.046 mJy. From a Gaussian fit in CASA for the continuum source and the synthesized beam size (007 × 005), we obtained a deconvolved size of 00852 ± 00046 × 00686 ± 00035 with a PA of 31° ± 12°, corresponding to an approximate physical size of 200 au × 150 au, which accords with the physical sizes of protoplanetary disks (∼100 au).

From this detected dust emission, we can estimate the total mass for these structures, even for the case of component B, which is much fainter than component A. At millimeter wavelengths, the emission can be assumed to be optically thin, and we can then estimate the dust mass from the following expression:

$$\begin{eqnarray}&&{M}_{d}=\displaystyle \frac{{D}^{2}\,{S}_{\nu }}{\left.{\kappa }_{\nu }\,{B}_{\nu }({T}_{d}\right)},\end{eqnarray} \tag{ 1 }$$

where S_ν is the flux density, D is the distance to the considered source, where we assume that it is the same as the Carina Nebula (2.3 ± 0.1 kpc; Smith & Brooks 2008), κ_ν = 0.015 cm² g⁻¹ is the dust mass opacity for a dust-to-gas ratio of 100 appropriate for 1.3 mm (Ossenkopf & Henning 1994), and B_ν (T_d ) is the Planck function for dust temperature T_d . Given the observing frequency of our data (∼225 GHz), we considered the Rayleigh–Jeans regime. We compute the values for the masses of this (and the others) source and show them in Table 2 from the measured fluxes presented in Table 1. For this source, we found that the dust mass associated with the detected continuum structure (∼15 mJy) is between 0.3 and 0.7 M_⊙, assuming temperature values between 20 and 50 K as in Cortes-Rangel et al. (2020).

As mentioned earlier, we also show the molecular CO(2−1) emission corresponding to the outflow in the HH 666 object. In Figure 1, we can see the molecular emission plotted in blue and red colors corresponding to the blueshifted and redshifted gas, respectively. From these maps, we only detected CO(2−1) emission (blueshifted+redshifted) toward the western side of the ALMA 1.3 mm continuum source, revealing a monopolar outflow, which is associated with the extended HH 666 object. We did not detect molecular CO(2−1) emission associated with the large globule itself, maybe because its emission is either extended and complex, so our ALMA observations could not recover quite well the extension of the globule. Another possibility is that this is an old globule with a quite small quantity of molecular gas. Klaassen et al. (2019) using ALMA Compact Array observations of pillar 3 (corresponding to the HH 666 globule) reported clumpy and dispersed CO(2−1) emission. Despite the larger beam size of their observations with respect to ours (i.e., 6'' or about 20 times larger), the extended emission of the globule was poorly recovered; as such, we favor the interpretation that our observations are not capturing most of the extended emission.

Since the detected CO emission is faint, it is difficult to say anything about the orientation, extension, or morphology of the molecular outflow. From Figure 1, this appears to have a west-to-east orientation with most of the compact emission arising toward the western side. In addition, we cannot give an estimation for the mass of the globule, and its respective outflow due to not tracing their complete structure.

HH 900 is a dark globule located very close to the central zone of the Carina Nebula, a region strongly affected by the extreme-ultraviolet (EUV) radiation generated by nearby massive stars to the star cluster Tr 16, and the massive and evolved variable star η Carinae located about ∼3 pc from HH 900. It has an angular size of the order of 1'' (equivalent to a spatial scale of ∼0.01 pc). This source is also known as the tadpole due to its unusual morphology that simulates a baby frog, made up of a dense globule (tadpole head) followed by an elongated and compact tail that extends in the same direction as the axis of the optical jet revealed by images of the HST (see Figure 2). Using these observations, Smith et al. (2010) studied the components of the bipolar structure of the optical jet and globule traced by the gas emission of Hα corresponding to externally ionized gas. The asymmetric morphology revealed at optical wavelengths has not been explained since it differs from typical jet morphologies that present a collimated and very symmetrical bipolar structure (Reipurth & Bally 2001; Reipurth et al. 2019; Noriega-Crespo et al. 2020). The EUV radiation field is illuminating throughout the length of the protostellar jet with a projection close to the plane of the sky and an inclination angle for the jet ∼10° (Reiter et al. 2015b).

Reiter et al. (2015b) studied the high-velocity component of the protostellar jet through NIR observations for the line of [Fe ii] and that, unlike the observations of Hα, showed an external collimated jet structure that is extending from east to west of the globule, presenting a remarkable symmetry that argues for the fact that the driving source would have to be located inside the center of the globule although they did not find IR emission from the protostar inside the globule, suggesting the fact that the protostar may be embedded inside remaining invisible due to the high column density. Shinn et al. (2013) detected two sources associated with the HH 900 object through observations of [Fe ii] at 1.64 μm but until now, no study could corroborate that these detected sources were generating the jets because no direct connection was found between the extended flows and the compact sources detected by this study (YSOs also cataloged by Spitzer; Povich et al. 2011).

Subsequently, Reiter et al. (2019) studied the HH 900 source at optical wavelengths with Very Large Telescope (VLT)/MUSE observations, detecting several emission lines (hydrogen recombination lines and some forbidden emission lines) that also trace the outflow. Their observations were also unable to detect the protostar embedded within the tadpole, although the emission was detected for different lines that are tracing the collimated bipolar jet at different densities (for example [Fe ii], [Ca ii], and [Ni ii]); the emission emanating from the globule reinforces the idea that the protostellar source must be housed inside it.

More recently Reiter et al. (2020a) reported using ALMA observations a bipolar molecular outflow traced by CO emission arising from the protostar HH 900 YSO. The outflow is biconical and matches very well the overall morphology of the optical jet. The estimated mass for the total globule is about 2 M_⊙ (from the dust continuum), which results in a lifetime of 4 Myr (Reiter et al. 2020a), with the assumption of a constant photoevaporation rate.

Our results obtained from the ALMA observations for the continuum emission at 1.3 mm reveals two compact sources called A and B (white contours in Figure 2). These two compact objects were already reported in previous ALMA observations presented in Reiter et al. (2020a). The millimeter emission is shown in Figure 2 as white contours and is likely associated with dust emission from a circumstellar envelope and flatted disk (see for example Reiter et al. 2020a, 2020b). We note that the emission is located at the center of the tadpole's head from which emerges the bipolar molecular outflow, also detected thanks to the high angular resolution of ALMA (Reiter et al. 2020a). Our observations also revealed a second component traced by the continuum emission (component B) located in the southeastern part of the optical jet in the area strongly illuminated by external UV radiation. From the figure, we can see that there is no molecular outflow associated with this component, maybe because this source is prestellar, as already discussed in Reiter et al. (2020a). Reiter et al. (2019) also rejected the premise that this detected object (named PCYC 838) was driving a microjet. They dismissed the idea by arguing that the symmetric [Fe ii] jet is centered on the globule and not the star, and, in addition to the lack of evidence of different velocity components in the reported spectra, concluded that a likely explanation is that this object is coincidentally aligned with the observed jet.

From the continuum emission, we obtained an integrated flux of 8.37 ± 0.61 mJy and 1.09 ± 0.09 mJy for components A and B respectively, and an observed rms noise of 0.043 mJy (see Table 1). From the synthesized beam size (034 × 027), we obtained a deconvolved size for component A of 0576 ± 0052 × 0396 ± 0042 with a PA of 155° ± 12°, with an approximate physical size of 1330 × 910 au.

From the millimeter continuum emission, we estimate the dust mass associated with the structure detected by our observations, using Equation (1) described in the previous subsection, giving values between 0.17−0.43 M_⊙, assuming temperatures between 20 and 50 K for component A, which is the component associated with the detected CO(2–1) bipolar molecular outflow shown in red and blue colors in the Figure 2. This line emission at 230 GHz is tracing perfectly the tail of the tadpole (blueshifted CO gas) extending along the axis of the optical jet, just like its counterpart (redshifted CO gas). We can deduce that the protostar is a Class 0/I object from the mass-loss rate (∼5 × 10⁻⁶ M_⊙ yr⁻¹) reported by Reiter et al. (2015b). The continuum emission of this source possibly reveals emission from the disk and the envelope around the driving source. For component B we estimated a dust mass between 0.02 and 0.05 M_⊙ under the same assumptions.

We note in Figure 2 that the molecular outflow is aligned with the optical jet, and expands at velocities between −32 and −36 km s⁻¹. The flux detected in CO is tracing the innermost parts of the dark globule connected to the extent of the outflow with the driving source detected by the continuum emission. We integrate LSR velocities from −29.27 to −33.08 km s⁻¹ for the redshifted emission oriented to the northwest, and from −34.99 to −40.70 km s⁻¹ for the blueshifted emission oriented to the southeast, for the CO outflow in the HH 900 object. The molecular outflow revealed by the CO is only evident for one component denominated component A.

In Figure 3, we show the emission by molecular gas for the CO(2−1) (top panel) and C¹⁸O(2−1) (bottom panel) detected lines. The moment zero of CO is shown in black contours for both panels of Figure 3, as a comparison between the emission of the molecules. The moment one emission is shown in color scale for CO (top panel) and for C¹⁸O (bottom panel). It is observed that CO is perfectly tracing the globule, including the tail of the globule hosting the compact source detected by the continuum (gray circle) and corresponding to the bipolar molecular outflow of CO also observed. The bipolar molecular outflow is also evident in these maps for both panels, with a northwestern orientation for the redshifted emission and a southeastern orientation for the blueshifted emission. If we analyze the velocities for the molecular outflow in this map, we note that the width velocity is ∼5 km s⁻¹, which suggests that the molecular flow is not a high-velocity one.

Zoom In Zoom Out Reset image size

The outflow extends no further than the emission detected for the rest of the HH 900 globule (bounded by the moment zero of CO). The different velocity components detected vary between the LSR range of −29.27 to −40.70 km s⁻¹, velocities that were used to integrate the moment zero (contours) and one (colors) of the CO emission. The average LSR velocity of the cloud is ∼−34 km s⁻¹. The CO molecule best traces the globule as a whole, while the C¹⁸O is evidencing more internal zones of the cloud where the protostar and envelope are localized, and the density is higher. This image (bottom panel) was obtained by integrating over the LSR velocity range of −33.23 to −35.23 km s⁻¹. The system LSR velocity of the cloud is ∼−33.90 km s⁻¹. For this case, the molecular emission shows a velocity width ∼1 km s⁻¹, which confirms the result obtained with the CO map, showing that these flows are not high speed.

We compute the gas mass using the observed CO transition J = 2 → 1 from the following equation:

$$\begin{eqnarray}\begin{array}{rcl}\left[\displaystyle \frac{{M}_{{{\rm{H}}}_{2}}}{{M}_{\odot }}\right] & = & 1.2\times {10}^{-15}\,{T}_{\mathrm{ex}}\,{e}^{\tfrac{16.59}{{T}_{\mathrm{ex}}}}{X}_{\tfrac{{{\rm{H}}}_{2}}{\mathrm{CO}}}\left[\displaystyle \frac{\displaystyle \int {I}_{\nu }{dv}}{\mathrm{Jy}\,\mathrm{km}\,{{\rm{s}}}^{-1}}\right]\\ & & \times \,\left[\displaystyle \frac{{\theta }_{\mathrm{maj}}\,{\theta }_{\min }}{{\mathrm{arcsec}}^{2}}\right]{\left[\displaystyle \frac{D}{\mathrm{pc}}\right]}^{2},\end{array}\end{eqnarray} \tag{ 2 }$$

where ${X}_{\tfrac{{{\rm{H}}}_{2}}{\mathrm{CO}}}$ ∼ 10⁴ (Scoville et al. 1986) is the abundance ratio between the molecular hydrogen (H₂) and carbon monoxide (CO), I_ν is the average intensity (I_ν ) for the pillar/globule/outflow, dv is the velocity range for the pillar/globule/outflow, D is the distance to the source (2.3 ± 0.1 kpc), and T_ex is the excitation temperature, taken as 50 K; this is the highest temperature in the range used previously for the dust temperature, and it provides an upper limit for our mass estimation.

We obtained a value for the gas mass of the tadpole globule of 4 × 10⁻³ M_⊙ using the emission lower than ±1 km s⁻¹ from the system velocity, and the mass of the outflow was 1.4 × 10⁻³ M_⊙ for velocities larger than ±1 km s⁻¹ (see Table 3). The estimated values for both masses should be considered as lower mass limits because the medium could be optically thick.

Table 3. Gas Masses Obtained by the Emission Detected for the CO(2−1) Emission at 230 GHz for the Pillars, Globules, and Outflows Associated with Each HH Object Studied

Source	Pillar/Globule				Outflow a
	Iν b	dvc	Size d	Mass	${I}_{\nu }^{\mathrm{red},}$ e ${I}_{\nu }^{\,{blue}}$	dvred, f dvblue	Sizered Sizeblue	Mass g
	(mJy beam−1]	(km s−1)	(arcsec2)	(M⊙)	(mJy beam−1)	(km s−1)	(arcsec2)	(M⊙)
HH 900	60	2.7	5.2	4 × 10−3	46.9 23.5	3 1	2 2	1.4 × 10−3
HH 1006	50	3.2	11.4	7 × 10−3	44 17.6	3 2.5	3.2 1.2	2.1 × 10−3
HH 1010	50	2.2	170	7 × 10−2	3.3 4	4 2	2.5 0.25	1.5 × 10−3

Notes. We estimate the mass using an excitation temperature of T_ex = 50 K, an abundance ratio between the molecular hydrogen and CO of ${X}_{\tfrac{{{\rm{H}}}_{2}}{\mathrm{CO}}}$ ∼ 10⁴, and a distance to the Carina Nebula of 2.3 kpc. We only included the sources for which there was CO molecular detection.

^a Red and blue superscript refer to detected redshifted and blueshifted gas, respectively. ^b Average intensity for the pillar or globule. ^c Velocity range for the pillar or globule flux. ^d We obtain the size by multiplying ${\theta }_{\max }\times {\theta }_{\min }$. ^e Average intensity for the outflow. ^f Velocity range for the outflow flux. ^g It is the total mass of the outflow, that is, the sum of the mass of the redshifted and the blueshifted outflow.

Download table as:  ASCIITypeset image

We calculate the theoretical value of the mass photoevaporation rate ($\dot{M}$) for the globule HH 900 associated with the driving source following Cortes-Rangel et al. (2020; see their Equation (3)). The size of the globule was calculated using a Gaussian fitting in CASA, assuming a distance of 2.3 kpc. In this case, the globule is affected by UV radiation originating from Tr 16 with a UV luminosity of (Q_H ) = 9 × 10⁵⁰ photons s⁻¹ (Smith & Brooks 2008). This luminosity (Q_H ) translates into a flux F_EUV = 8.35 × 10¹¹ photons s⁻¹ cm⁻² assuming that the distance from the HH 900 globule to the cluster is ∼3 pc. We obtained a mass photoevaporation rate of 2.6 × 10⁻⁵ M_⊙ yr⁻¹. This estimated value is consistent with the methodology described by Reiter et al. (2015b; 7 × 10⁻⁶ M_⊙ yr⁻¹), taking into account that we have used the gas mass of the globule (instead of the dust mass), which translates into a difference of at least an order of magnitude in mass and a factor of ∼4 in size, so our value differs by an order of magnitude than the one presented by them. In this calculation we do not consider the optical depth value, and since the globule is very optically thick this may account for the difference. We can estimate the photoevaporation timescale of the globule HH 900 by dividing the estimated gas mass for the globule (4 × 10⁻³ M_⊙) by this mass photoevaporation rate, giving ∼150 yr, which will be the time in which the globule HH 900 will be photoevaporated by the massive stellar cluster Tr 16 (see Table 4). Using the dust mass we estimated for this globule (0.21 M_⊙) that corresponds to the whole continuum emission at 1.3 mm, the photoevaporation timescale would be ∼8000 yr, ∼two orders magnitude less than the reported value reported by Reiter et al. (2020a; 4 Myr), although it should be noted that our estimated dust mass differs by a factor of ∼10 and the photoevaporation rate by a factor of ∼100.

Table 4. The Sizes of Each Pillar/Globule Were Calculated Using a Gaussian Fitting in CASA to the Extended Gas Emission Using CO (2−1), Assuming a Distance of 2.3 kpc to Each of the Studied Sources

Source	Ra	$\dot{M}$ b	tc
	(au)	(M⊙ yr−1)	(yr)
HH 900	8050	2.6 × 10−5	153
HH 1006	8740	5.7 × 10−6	1228
HH 1010	39,100	3.1 × 10−5	2258

Notes. Each pillar/globule is affected by EUV radiation (F_EUV) originating from the nearby clusters, Tr 14 and Tr 16, for which UV luminosities of approximately (Q_H ) = 2 × 10⁵⁰ photons s⁻¹ for Tr 14 and (Q_H ) = 9 × 10⁵⁰ photons s⁻¹ for Tr 16 (Smith & Brooks 2008) are found.

^a Radius of the pillar/globule. ^b Mass photoevaporation rate. ^c Photoevaporation timescale.

Download table as:  ASCIITypeset image

This source is located in the southern region of the Carina Nebula, specifically within the South Pillars. The pillar identified by Smith et al. (2010) exhibits a highly elongated structure pointing toward the Tr 16 stellar cluster. This displays a clear optical bipolar jet at the tip of the pillar, which is being strongly irradiated by the UV radiation field emanating from the OB stars present in Tr 16. Further south of the pillar, the presence of another bright jet is evident, expanding in the southwest direction opposite to that of the stellar cluster. There are shock structures further away that may potentially be associated with the same jet phenomenon, although considering only the innermost part of the detected optical jet, a jet length of approximately 0.4 pc is quantified.

In Figure 4, the combination of observations made with HST/ACS for the HH 1004 pillar and our observations obtained with ALMA for the continuum and CO emission are shown. The optical image (green) corresponds to Hα emission and overlaps with the ALMA high-velocity CO(2−1) molecular emission (blue and red) and the 1.3 mm continuum emission (white contours). The CO(2−1) molecular emission was obtained by integrating over LSR velocities from −15.55 to −4.75 km s⁻¹ for the redshifted emission, and from −40.31 to −25.07 km s⁻¹ for the blueshifted emission. The ALMA 1.3 mm continuum image reveals three compact sources (A, B, and C) distributed in the northern region of the pillar (at the head of the pillar, see Table 1). Components A and B appear to be associated with the optical jet detected in Hα.

Zoom In Zoom Out Reset image size

For this millimeter continuum emission, we obtained an integrated flux of 7.42 ± 0.38 mJy for source A, 0.90 ± 0.15 mJy for source B, and 1.49 ± 0.09 mJy for source C. The observed rms noise was 0.053 mJy for the three sources, which is very close to the theoretical noise value (see Table 1). From the synthesized beam size (011 × 006), we obtained a deconvolved size similar to HH 666; for example, for source C we obtained a deconvolved size of 0036 ± 0014 × 0017 ± 0018 with a PA of 178° ± 49°, which corresponds to an approximate physical size of 83 au × 40 au, consistent with the typical physical sizes of protoplanetary disks (Andrews 2020).

We estimated the total masses for these structures using Equation (1) and we obtained that the masses for components A, B, and C are between 0.15−0.38, 0.01−0.04, and 0.03−0.07 M_⊙, respectively, assuming temperature values between 20 and 50 K (see Table 2).

We show in the same Figure 4 the molecular CO (2−1) emission plotted in blue and red colors corresponding to the blueshifted and redshifted gas, respectively. We observe a bipolar outflow emerging from the compact continuum source A, with the blueshifted portion extending southwest along the axis of the optical jet detected in Hα. Very little redshifted emission is apparent, and due to the weak flux detected at these velocities, it is not clear whether this emission expands in the same northeast direction along the axis of the optical jet. Source B also appears to overlap with a portion of the blueshifted molecular outflow detected, although it is difficult to discern whether the detection of this molecular outflow is generated by both sources A and B, or if it is solely attributed to source A, which is larger (∼230 au × 140 au) and more massive (0.38 M_⊙) compared to source B (∼140 au × 100 au, 0.04 M_⊙ mass). Compact source C exhibits a CO redshifted molecular outflow that is more extended than the ones detected for sources A and B, although it does not display a blueshifted counterpart. This can be explained by inhomogeneities in the cloud. As far as we know this is the first molecular detection for this outflow component. This source is north of the [Fe ii] emission that goes through the pillar (see Figure 6 in Reiter et al. 2016). This looks to be in the right place to drive the extended Hα emission seen to the left of the pillar that Smith et al. (2010) identify as HH 1005.

This bipolar jet emerges from a small dark globule, pointing in the northern direction of η Carinae, and is located in the central region of the South Pillars in the zone where these pillars are strongly irradiated by the massive stars of Tr 16. The HH 1006 jet is formed by a chain of knots that extend along its axis, and have an extension of ∼0.2 pc measured at optical wavelengths by Smith et al. (2010), and by Sahai et al. (2012), who presented CO (3−2) single dish observations, although for our study we have focused only on the characterization of the central jet, that is, the part closest to the globule.

In Figure 5 we show an optical HST image in Hα (green) and our results obtained with ALMA for the continuum emission at 1.3 mm (white contours in the figure) and the CO molecular emission (red and blue colors). The continuum emission reveals a compact source at the center of the globule with an integrated flux of 9.41 ± 0.73 mJy and a PA of 43° ± 17° (see Table 1). From the synthesized beam size (023 × 020), we obtained a deconvolved size of 1139 ± 0089 × 0945 ± 0075, which corresponds to an approximate physical size of 2600 au × 2200 au. The compact source is generating the molecular outflow that extends along the same axis of the globule. We estimate that the mass of dust, using Equation (1), is between 0.19 and 0.48 M_⊙, assuming temperature values between 20 and 50 K. This value is an order of magnitude lower than the estimation of 0.02 M_⊙ obtained by Mesa-Delgado et al. (2016), which can be explained by the fact that we report a larger source size. Perhaps, we are detecting more of the envelope than they did.

Zoom In Zoom Out Reset image size

The molecular outflow is detected aligned with the axis of the optical jet and extends in the same direction of the globule axis at velocities between −21 and −24 km s⁻¹. This CO flux is tracing the innermost parts of the globule connected to the extent of the outflow with the driving source. We integrate in radial velocities from −17.88 to −21.05 km s⁻¹ for the redshifted emission oriented to the south, and from −24.23 to −27.40 km s⁻¹ for the blueshifted emission oriented to the north of the globule.

In Figure 6, we show the moment zero map obtained from CO(2−1) in contours overlaid with moment one map obtained from CO(2−1, upper panel) and C¹⁸O(2−1, bottom panel), respectively. It is observed that CO is perfectly tracing the globule hosting the compact source detected by the continuum (gray circle), corresponding to the center of the bipolar molecular outflow, which is evident in the top panel. The width velocity is ∼5 km s⁻¹, which tells us that the molecular outflow is not a high-velocity one. The CO moment zero for the top panel image was obtained by integrating over the LSR velocity range of −17.88 to −27.40 km s⁻¹. The system LSR velocity of the cloud is ∼−23 km s⁻¹. The CO is the best tracer of the globule, while the C¹⁸O traces more internal zones of the cloud where the protostar and envelope are localized. The bottom panel image was obtained by integrating over the LSR velocity range of −21.93 to −24.60 km s⁻¹ with a velocity width ∼1 km s⁻¹.

Zoom In Zoom Out Reset image size

The gas mass of the globule is 4 × 10⁻³ M_⊙ and the mass of the outflow is 2.1 × 10⁻³ M_⊙ (see Table 3). The globule HH 1006 is affected by UV radiation coming from the cluster Tr 16, which has a UV luminosity of (Q_H ) = 9 × 10⁵⁰ photon s⁻¹, and given the distance at which the source of the cluster is located (∼16 pc), we obtain a flux F_EUV = 3.09 × 10¹⁰ photons s⁻¹ cm⁻². We obtained a mass photoevaporation rate of 5.7 × 10⁻⁶ M_⊙ yr⁻¹ and the photoevaporation timescale of the globule HH 1006 given the estimated mass of gas (7 × 10⁻³ M_⊙) is ∼1230 yr (see Table 4).

This source emerges from the tip of a giant dark pillar located at the western periphery of the Carina Nebula, and pointing toward the evolved star η Carinae and Tr 16. Smith et al. (2010) detected optical emission within the pillar, although their detection could not be corroborated. From the optical images, the presence of a bipolar jet becomes evident, also reported by McLeod et al. (2016), with its jet axis perpendicular to the pillar axis. This geometrical configuration is commonly observed in Carina Pillars.

In Figure 7, we show an optical HST image in Hα (green), and our results obtained with the ALMA observations for the continuum emission at 1.3 mm (white contours) and the CO (2−1) molecular emission (red and blue colors). The continuum emission reveals a compact source at the center of the pillar head, and that is exciting the molecular outflow. This outflow extends along the jet axis at velocities between −9 and −27 km s⁻¹. This continuum emission has an integrated flux of 4.12 ± 0.17 mJy. We were not able to resolve the source because the angular resolution was not sufficient (see Table 1).

Zoom In Zoom Out Reset image size

The dust mass for the continuum source is between 0.08 and 0.21 M_⊙, assuming temperature values between 20 and 50 K. The CO is tracing the outflow and was obtained by integrating the emission in radial velocities from −9.54 to −15.25 km s⁻¹ for the redshifted emission oriented to the south, and from −18.43 to −27.32 km s⁻¹ for the blueshifted emission oriented to the north of the pillar. The red/blueshifted distributions are consistent with McLeod et al. (2016)'s observations.

In Figure 8, we show the emission by molecular gas for the CO (2−1) and C¹⁸O (2−1) transitions. The moment zero (black contours) represents the CO emission in both panels. The moment one is presented (in colors) for the CO (top panel) and C¹⁸O (bottom panel). For the top panel, the CO moment zero image was obtained by integrating in the velocity range of −9.54 to −27.32 km s⁻¹, with the average LSR velocity of the cloud ∼−16 km s⁻¹. The bottom panel image was obtained by integrating over the velocity range of −15.65 to −18.65 km s⁻¹, with the average LSR velocity of the cloud ∼−17 km s⁻¹. We note that the CO is tracing perfectly the pillar in all its extension and the accurate shape of the molecular outflow is still evident even considering all the CO emission. The C¹⁸O is again tracing zones deep inside the cloud, where the protostar is forming and therefore the density is higher.

Zoom In Zoom Out Reset image size

The estimated gas mass for the HH 1010 object is 7 × 10⁻² M_⊙ and the mass of the outflow is 1.5 × 10⁻³ M_⊙ (see Table 3). The globule HH 1010 is affected by UV radiation coming from the cluster Tr 14. Given the distance at which the source of the cluster is located (∼13 pc), we obtain a flux of F_EUV = 1.02 × 10¹⁰ photons s⁻¹ cm⁻², and a mass photoevaporation rate 3.1 × 10⁻⁵ M_⊙ yr⁻¹, yielding a photoevaporation timescale ∼2260 yr (see Table 4).

This object was detected to the west of the Carina Nebula very close to the Tr 14 star cluster just above the HH 901 and 902 sources, which all together form the structure known as The Mystic Mountain. ⁶ The detected jet is approximately 0.05 pc in extent and is associated with what appears to be a small, bright bow shock (see Smith et al. 2010) although without an obvious optical jet counterpart. Through optical observations (Hα) Smith et al. (2010) were able to detect a weak point source of emission located at the tip of the dark cloud and within the axis of the optical jet, which was also detected with IR observations of Spitzer (YSO detected along the optical jet; see Povich et al. 2011) without a clear association with the driving source. Further optical and IR observations by Reiter & Smith (2013) made evident the bipolarity of the collimated jet detected in Hα and [Fe ii] but again without being able to evidence the complementary bow shock.

In Figure 9, we show an optical HST image (in Hα, in green) and our results obtained with ALMA observations for the continuum emission (at 1.3 mm in white contours) and the C¹⁸O molecular emission (red and blue colors). In this case, it was not possible to highlight the outflow with CO(2−1) emission due to contamination by ambient gas and the structures near the globule. Although we tried to reproduce the innermost zone (blue and red colors) of the C¹⁸O molecular outflow (in the previous cases where we have used CO), the emission plotted in Figure 9 only represents the innermost zones of the molecular cloud, without clearly outlining the outflow. The emission in this case was integrated over the LSR velocity range of −7.51 to −8.18 km s⁻¹ for the redshifted gas and −9.84 to −10.18 km s^–1 for the blueshifted gas.

Zoom In Zoom Out Reset image size

From the continuum emission, we detected a compact dust source associated with the envelope and disk structure, where the protostellar source is embedded (component A, see Figure 9), and a compact emission source (component B). For component A, we obtained an integrated flux of 3.86 ± 0.66 mJy, and an rms noise of 0.05 mJy (see Table 1). From the synthesized beam size (026 × 018), we obtained a deconvolved size of 0373 ± 0085 × 0211 ± 0092, with a PA of 170° ± 26°, which corresponds to an approximate physical size of 850 × 490 au. We estimated the dust mass from the continuum emission and obtained values between 0.07 and 0.19 M_⊙ by varying the temperature between 20 and 50 K. These values are in agreement with the estimated values by Mesa-Delgado et al. (2016; they estimated a mass ∼0.04 M_⊙); thus, we clearly are recovering some extended emission. Additionally, we report for the first time the detection for component B, whose integrated flux of 2.9 mJy and peak flux of 0.8 mJy, are about an order of magnitude larger than the rms value. There was no gas detection associated with this component.

In Figure 10, we show a moment zero image obtained from the C¹⁸O (2−1) emission (contours) combined with moment one image obtained from the C¹⁸O (2−1) (upper panel) and N₂D⁺(3−2) (bottom panel) emissions. This is the only source in this study where we detected N₂D⁺(3−2) emission. The moment zero image of C¹⁸O perfectly traces the globule HH 1066, and it does not extend beyond the boundary delimited by the optical emission detected in Hα (see Figure 9). The moment one reveals gas kinematics that for C¹⁸O evidence a small velocity gradient with ranges between −7 and −10 km s⁻¹, with higher velocities toward the outer walls of the globule. In the case of N₂D⁺ moment one, the velocities vary over a smaller range from −8 to −9 km s⁻¹, and the structure traced by the gas is located in the innermost regions of the globule, close to the protostellar source detected by the continuum emission (gray circle).

Zoom In Zoom Out Reset image size

We can examine if the results obtained for the compact disks+envelopes revealed in our 1.3 mm continuum maps for the HH objects are in principle consistent with the formation of planets. If we consider the lowest value obtained for the masses of the disks+envelopes (0.01 M_⊙ for HH 1004 component B) which corresponds to ∼10 M_Jup, we can see that is above the minimum required mass (10 M_Jup) for a prestellar nebula to develop a planetary system (Weidenschilling 1977). However, we remark that there is probably contamination by the dusty envelope, so the mass of the disk could be lower. However, if we consider the maximum mass value obtained (0.77 M_⊙ for HH 666), corresponding to ∼800 M_Jup, we see that it is well above the minimum required, which is the case for many other sources detected with our observations, so at least in most of the sources this minimum mass condition is fulfilled.

On the other hand, a second important factor to consider is the average age of the Carina population, which is ∼1–4 Myr (Smith & Brooks 2008), consistent with the minimum timescale required to form planets ∼1–2 Myr (see Hubickyj et al. 2005; Najita & Kenyon 2014). The external UV irradiance from Tr 14 and Tr 16 could photodisassociate the material of the disks. However, these disks are still very embedded within the globules and pillars, implying that they are not strongly irradiated yet. All these conditions indicate the possibility that there are young planets formed or in formation in these detected disks.