Millimetre Observations of Maser-Emitting Planetary Nebulae

Abstract

Observations in the millimetre bands of maser-emitting planetary nebulae (PNe) are crucial to study their circumstellar molecular gas at the beginning of the PN phase. Maser-emitting PNe are in the earliest phases of PN formation; therefore, these sources are key objects to study the molecular content during the early evolution of PNe. These circumstellar envelopes are active sites for the formation of molecules. We present preliminary results of millimetre observations with the IRAM 30 m telescope towards one PN (IRAS 17393−2727) of a sample of five maser-emitting PNe, where we detect ${}^{12}$CO and ${}^{13}$CO lines in both $J=1\to 0$ and $J=2\to 1$ transitions.

1. Introduction

Planetary nebulae (PNe) are one of the last phases in the evolution of low- and intermediate-mass stars (≤8 M${}_{\odot }$). Their immediate precursors are stars in the asymptotic giant branch (AGB), followed by a short (≃100–10,000 yr) transitional post-AGB phase.

The circumstellar envelopes of O-rich evolved stars provide optimal conditions to pump different species of masers. During the AGB phase, maser emission from the following molecules has been detected: SiO, H${}_2$O, and OH [1]. Masers are potentially important tools in the study of PNe formation due to their short life times in evolved objects. For example: SiO, H${}_2$O, and OH masers are expected to disappear ≃10, 100, and 1000 yr, respectively, after the end of the AGB mass-loss [2,3]. Considering that the post-AGB phase is short, PNe showing H${}_2$O and OH maser emissions are expected to be very young. Very few members of these special types of H${}_2$O- and/or OH-emitting-PNe are known, including K3-35, IRAS 17347−3139, IRAS 18061−2505, IRAS 16333−4807, IRAS 15103−5754, JaSt 23, IRAS 17393−2727, IRAS 19219+0947 [4,5,6,7,8,9,10,11]

Thus, H${}_2$O- and OH-emitting PNe could be among the youngest PNe, and therefore, they are key objects in the study of circumstellar molecular gas in very recently formed PNe.

The post-main sequence envelopes are very active sites for the production of molecules [12]. For example, HCN and HCO${}^{+}$ were detected in 13 of the 17 PNe with an age range from 800 to 13,000 yr, at least in one transition [13]. Nine of them were common to both molecules. Later, [14] detected HNC in the same sample of PNe. In a recent molecular line survey in PNe, CN emission was common in most of the sources together with the molecules mentioned above [15]. Moreover, a robust dependence of the HNC/HCN line ratio on the UV luminosity of the PN central star was found. Furthermore, a marginal correlation between the HNC/HCN line ratio and PN age was identified [15].

The survival of molecules in the shells of PNe has been a subject of theoretical debate. Earlier models by [16] suggested that dusty cometary-like globules could shield molecular material and preserve it through the evolution of proto-PNe into PNe. Regarding maser-emitting PNe, K3-35 was the only source of this group where both HCO${}^{+}$ and ${}^{12}$CO had been detected until 2018 [17,18]. Now we know that IRAS 15103−5754 also harbours HCO${}^{+}$, C${}^{18}$O, and ${}^{12}$CO, that was mapped with ALMA [19].

2. Observations

We carried out millimetre (mm) observations with the Institut de Radioastronomie Millimétrique (IRAM) 30 m telescope in August 2014 towards a sample of five maser-emitting PNe. We used the Eight MIxer Receiver (EMIR) for all the observations. We selected simultaneous observations of ${}^{12}$CO $J=1\to 0$ (115.271 GHz), $J=2\to 1$ (230.538 GHz), and ${}^{13}$CO $J=1\to 0$ (110.201 GHz), and $J=2\to 1$ (220.399 GHz). The frequency resolution was ∼200 kHz, corresponding to a velocity resolution of 0.5 km s${}^{-1}$ and 0.25 km s${}^{-1}$ at 3 mm and 1 mm, respectively. The bandwidth was 4 GHz. The rms (one-sigma) noise was between 5 mK and 92 mK at 3 mm and between 17 mK and 130 mK at 1 mm.

In order to associate the ${}^{12}$CO and ${}^{13}$CO line emission with the PN, we performed observations towards the source position and four off-source positions located 24 arcsec away from it, in the N, S, E and W directions. These 24 arcsec correspond to the half-power beam width (HPBW) at 3 mm, and twice the HPBW at 1 mm. This observational strategy was used previously by [20] while observing these molecular lines towards a particular type of post-AGB star known as water fountains (WFs).

3. Analysis

Data reduction and analysis were performed using the GILDAS software1. We used the shell method available in GILDAS to fit the observed ${}^{12}$CO and ${}^{13}$CO lines in our sample. This method fits horn-type profiles for circumstellar envelopes and provides as outputs the parameters listed in

Table 1. Parameters of the ${}^{12}$CO and ${}^{13}$CO line emission in IRAS 17393−2727.

Line	${\mathit{V}}_{\mathbf{pk}}$ (km s${}^{-1}$)	${\mathit{V}}_{\mathbf{exp}}$ (km s${}^{-1}$)
${}^{12}$CO ($1\to 0$)	−105.5 (1.0)	18.9 (1.4)
${}^{12}$CO ($2\to 1$)	−106.0 (1.5)	17.5 (1.5)
${}^{13}$CO ($1\to 0$)	−106.9 (0.5)	15.4 (0.7)
${}^{13}$CO ($2\to 1$)	−107.0 (1.0)	16.2 (0.3)

Note. One-sigma errors within parentheses.

for the PN IRAS 17393−2727. In particular, we obtained the velocity of the emission peak ($V_{\mathrm{pk}}$) at the centre of the line profile and the expansion velocity ($V_{\exp }$) deduced from the full width at zero level.

Moreover, we used the code shapemol that is included in the software SHAPE [21,22] to estimate some physical parameters of the molecular emission associated with this source. shapemol produces synthetic spectral profiles of ${}^{12}$CO and ${}^{13}$CO lines to be compared with mm-range observations. See the next section for more details.

4. First Results

Circumstellar ${}^{12}$CO and ${}^{13}$CO line emission has been detected towards four of five sources in the sample (Uscanga et al., in prep). The emission from these molecules is only present in the on-source position and not in the other four off-source positions (N, S, E, and W).

Here we present our first results in one PN, IRAS 17393−2727. The line emission is broad, with a width of ∼35 km s${}^{-1}$ in the four lines observed (see Figure 1). All the intensities are given in a scale of main-beam temperature ($T_{\mathrm{MB}}$), assuming main-beam efficiencies of 0.78 and 0.59 at 3 mm and 1 mm, respectively. Velocities are given with respect to the kinematic definition of the Local Standard of Rest (LSR). The intensities of the $J=1\to 0$ and $2\to 1$ transitions of ${}^{12}$CO emission in our spectra are similar to those reported by [23], also observed with the IRAM 30 m telescope, although our observations have a better signal-to-noise ratio, and we detected for the first time the ${}^{13}$CO $J=1\to 0$ and $2\to 1$ transitions in this source (Uscanga et al., in prep).

The emission of ${}^{12}$CO and ${}^{13}$CO lines is located within the velocity range defined by the OH maser features at 1612 MHz with a velocity dispersion of ∼34 km s${}^{-1}$ [24]. The average velocity of the emission peak of the four emission lines (weighted by the inverse square of the errors) is ≃$-106.6\pm 0.4$ km s${}^{-1}$ (Table 1). We used this value as the systemic velocity for the model of the molecular envelope of the PN IRAS 17393−2727 and compared the synthetic spectra, generated with shapemol, against the observed ones (Figure 1).

There is only one estimation for the distance to this source ∼1.2 kpc, based on statistical methods [25]. The kinematic distance is ∼8 kpc. There is no counterpart to this source, neither in Gaia DR2 nor in Gaia Early DR3; the closest source does not have a measured parallax [26,27]. We build models for three different distances: the statistical one (∼1.2 kpc), the kinematic one (∼8 kpc), and an intermediate value (∼5 kpc). We obtained a molecular mass range from 0.007 M${}_{\odot }$ at 1.2 kpc to 0.6 M${}_{\odot }$ at 8 kpc. However, the best-fit model for the emission in ${}^{12}$CO and ${}^{13}$CO is at 5 kpc with the physical parameters listed in

Table 2. Best-fit model physical parameters for the molecular emission of IRAS 17393−2727.

Parameter	Value
Assumed Distance	5 kpc
Envelope Radius	$4.5\times {10}^{17}$ cm
Systemic Velocity ($V_{\mathrm{LSR}}$)	$-106.6$ km s${}^{-1}$
Expansion Velocity, Linear Pattern, $V_{\max }$	19 km s${}^{-1}$
Microturbulence Velocity (${\delta }_V$)	2 km s${}^{-1}$
Temperature (T)	190 K
Molecular Mass	0.28 M${}_{\odot }$
Total Density (n)	$3.1\times {10}^2$ cm${}^{-3}$
X(${}^{12}$CO)	$9.0\times {10}^{-5}$
X(${}^{13}$CO)	$1.7\times {10}^{-5}$

.

Given the scarcity of information on the spatial distribution of ${}^{12}$CO and ${}^{13}$CO line emission, with only an on-source pointing and four off-source positions, we resorted to a model as geometrically simple as possible, that is, a filled sphere with a radius of $4.5\times {10}^{17}$ cm, a typical size for AGB envelopes, with constant values for density and temperature. The only spatial dependence of the model is that on the expansion velocity, for which we considered a linear dependence on radius similar to that found in many young PNe, a homologous expansion. The maximum value of the velocity was 19 km s${}^{-1}$ (see Table 2). This is consistent with the average expansion velocity of the envelope obtained from the four emission lines (weighted by the inverse square of the errors) of ∼16.2 ± 0.3 km s${}^{-1}$ (Table 1).

5. Discussion and Future Work

In the model, the total size of the envelope at the assumed distance is ∼12 arcsec. This is about the size of the HPBW at 1 mm in our observations. In addition, an HST image of IRAS 17393−2727 shows a collimated, bipolar outflow of ∼2.2 arcsec in extent, with the two parts of the outflow separated by a dark region of ∼0.6 arcsec. Therefore, the assumed molecular envelope has a larger size than the outflow observed at optical wavelengths.

We have found a molecular mass of 0.28 M${}_{\odot }$ for IRAS 17393−2727. This value is quite similar to the masses reported in other maser-emitting evolved objects, i.e., WFs [20]. These sources are in a previous evolutionary stage, the post-AGB phase. They present high-velocity jets traced by H${}_2$O maser emission [28].

The abundance X of the species ${}^{12}$CO and ${}^{13}$CO is listed at the end of Table 2. The ${}^{12}$CO/${}^{13}$CO ratio is ∼5.3, which is quite similar to other PNe (especially O-rich) such as NGC 6302 [22,29]. However, another C-rich nebula, NGC 7027, may reach a ${}^{12}$CO/${}^{13}$CO ratio ∼50 [30] even though it is a young PN. On the other hand, O-rich AGB stars present ${}^{12}$C/${}^{13}$C ratios ∼10–35 [31], while in post-AGB stars such as WFs, this ratio is ∼7–30 [20]. We are still working in the analysis of the abundances of IRAS 17393−2727 and the other sources of our sample to reach a conclusive result, keeping in mind the uncertainties of these estimations.

For future work, we plan to study the molecular content of very young PNe, especially PNe with maser emission, and determine the physical parameters of the molecular gas, namely $v,T,M,n$. We will also measure the fractional abundances of HCN and HCO${}^{+}$ and compare them with the values obtained in more evolved PNe. Finally, we will carry out high-resolution observations of these sources to define the structures associated with these molecular emissions (envelope, torus, and outflow).