EDP Sciences logo
Subscriber Authentication Point
Search
Menu
Home All issues Volume 675 (July 2023) A&A, 675 (2023) A139 Full HTML
Open Access
Issue
A&A
Volume 675, July 2023
Article Number A139
Number of page(s) 9
Section Interstellar and circumstellar matter
DOI https://doi.org/10.1051/0004-6361/202346034
Published online 13 July 2023

© The Authors 2023

Licence Creative CommonsOpen Access article, published by EDP Sciences, under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

This article is published in open access under the Subscribe to Open model.

Open Access funding provided by Max Planck Society.

1 Introduction

Low- to intermediate-mass stars (0.8–8 M⊙) evolve to the asymptotic giant branch (AGB) when they are close to the end of their lives. These objects go through intense mass loss forming a cir-cumstellar envelope (CSE) around them that contains dust and molecules (Höfner & Olofsson 2018). Many of the 241 molecules that so far have been identified in astronomical sources were detected in the CSEs of AGB or red supergiant stars (McGuire 2022), which therefore efficiently enrich the interstellar medium (ISM).

The abundances of carbon and oxygen at the stellar surface reflect the interplay of nucleosynthesis and convection in the stellar interior. Stars with C/O > 1 are defined as carbon-rich, those with C/O < 1 as oxygen-rich, and those with C/O ≈ 1 are the S-type stars. The abundance of these atoms is mainly due to the third dredge-up, where the material is brought up to the stellar surface and subsequently ejected into the envelope thanks to thermal pulses (TPs) forming in response to instabilities in the helium-burning shell. The third dredge-up occurs after each TP (except for a few initial ones, depending on the model, e.g. Weiss & Ferguson 2009), influencing the C/O ratio in the envelope (Karakas & Lattanzio 2014, further references therein), while the number of TPs depends on the mass loss experienced by the star before it leaves the AGB.

IRC+10216 (= CW Leonis) is the archetypal carbon-rich AGB star located close to us at a distance between 120 and 140 pc (Crosas & Menten 1997; Groenewegen et al. 2012) and losing its mass at a high rate of 2–4 × 10−5 M⊙ yr−1 (Crosas & Menten 1997; De Beck & Olofsson 2018; Fonfría et al. 2022). The star has a luminosity of 8600 L⊙ determined from the Very Large Array (VLA) imaging of the optically thick radio photosphere (Menten et al. 2012) that, for a photospheric effective temperature of 2750 K (assumed from spectral energy distribution (SED) modelling, Men’shchikov et al. 2001), entails an optical photospheric diameter of 3.8 AU, while measurements of lunar occultations in the H and K bands estimate a near-infrared diameter of 7.1 AU (Richichi et al. 2003).

The star has a very extended molecular envelope moving through the ISM. Mauron & Huggins (1999) and Dharmawardena et al. (2018) also show that the dust continuum emission extends out to ~200″ or more from the central star. The interphase between the envelope and the ISM is seen in the ultraviolet images taken with the Galaxy Evolution Explorer (GALEX) satellite (Sahai & Chronopoulos 2010), and at far-infrared wavelengths with the Photodetector Array Camera and Spectrometer (PACS) and Spectral and Photometric Imaging Receiver (SPIRE) instruments aboard Herschel (Ladjal et al. 2010). A similar interaction between the two environments is also seen in Mira A as the star shows a bow-shock feature in the southward direction, and a tail extending in the north. The star has a space velocity of 130 km s−1 (Martin et al. 2007), which is larger than what is seen in IRC +10216 (~91 km s−1 in Sahai & Chronopoulos 2010). More than 80 molecular species have been detected in the CSE of IRC +10216 so far (see Cernicharo et al. 2000; Agúndez et al. 2014; De Beck & Olofsson 2018; Pardo et al. 2022, and many other publications cited therein). These molecules are important probes of the chemical processes at work in the envelope as different molecules at various levels of excitation trace distinct regions in the CSE: parent molecules such as HCN and C2H2 form in the inner, hotter part of the envelope (e.g. Cernicharo et al. 2015; Agúndez et al. 2020, further references therein), whereas the daughter species (i.e. mainly photodissociation products of the former) are found to be present in the outer parts (see Fig. 1 in Li et al. 2014; Millar & Herbst 1994; Millar et al. 2000; Van de Sande et al. 2018 for a further understanding of the circumstellar chemistry).

The distribution of the CO molecule, which is an important tracer of the history of mass loss, is quite widespread in the star’s CSE: Observations with the Institut de radioastronomie millimétrique (IRAM) 30 m telescope (Cernicharo et al. 2015) and with the Submillimter Array (SMA) and Atacama Large Millimeter Array (ALMA) interferometers (Guélin et al. 2018) show almost concentric shells, with strong CO emission up to the photodissociation radius of 180″ after which there is a sudden drop (best discernible in the single-dish maps, Cernicharo et al. 2015) owing to the molecule’s photodissociation yielding atomic carbon which is further ionised to form C+ (Morris & Jura 1983; Mamon et al. 1988; Schöier & Olofsson 2001; Groenewegen 2017; Saberi et al. 2019).

Emission in the two fine-structure lines of neutral atomic carbon, C0, from the 3P1 → 3P0 and 3P2 → 3P1 transitions near 492 and 809 GHz, respectively, was first detected in molecular clouds by Phillips et al. (1980) and Jaffe et al. (1985). [C I] emission1 is recognised as a valuable tracer for environments characterised by a range of molecular gas fractions occurring in photo-dissociation regions, while either C+ or CO trace them only partially (Papadopoulos et al. 2004, further references therein). However, for circumstellar envelopes, very few studies exist of these [C I] lines. IRC +10216 has been targeted in the 3P1 →3P0 line with the Caltech Submillimeter Observatory (CSO) 10.4 telescope and the James Clerk Maxwell 15 m Telescope (JCMT; Keene et al. 1993; van der Veen et al. 1998). As suggested by Keene et al. (1993), another important carrier of carbon in C-rich circumstellar envelopes is acetylene (C2H2), contributing to the shielding against interstellar UV radiation, and acting as a precursor molecule for the subsequent photo-chemistry (Santoro et al. 2020; Siebert et al. 2022).

Thanks to the triple bond between carbon and oxygen, CO has a high bond energy, 11109.2 ± 4.1 meV (Darwent 1970) that is slightly below the ionisation energy of C0, 11260.2880 ± 0.0011 meV (Glab et al. 2018). In an environment characterised by internal or external radiation fields and by gas density gradients, the production of free carbon by photodissociation of CO is therefore accompanied by the production of C+ (of which minor amounts might arise already before, from the photolysis of carbon-bearing species that dissociate easier than CO). Since the ionisation threshold of carbon falls significantly below that of hydrogen (13.598 eV; Jentschura et al. 2005), the 2P3/2 → 2P1/2 transition from the C+ fine-structure singlet originates from both the warm ionised gas phase of the interstellar medium and from cold, neutral, and partially molecular gas (T ~ 10 000 K and ~100 K, respectively; Wolfire et al. 2003, with further references therein). Its emission from diffuse gas is therefore widespread. However, despite the interest in the C+ emission for the analysis of the carbon budget in circumstellar environments, detections of the [C II] line were restricted to post-AGB stars (e.g. Cerrigone et al. 2012; Bujarrabal et al. 2016), before it was found in IRC +10216 (Reach et al. 2022).

Given the importance of atomic and ionised carbon in the circumstellar envelope of IRC +10216, we want to constrain the spatial distribution of the atom. In this publication, we discuss the observations we conducted and the data reduction methods in Sect. 2, followed by the presentation of the obtained spectra and their analysis (Sect. 3). We then discuss the physical constraints we obtained and contextualise them in view of the existing literature in Sect. 4 and conclude our work in Sect. 5. A more detailed model, including dedicated photo-chemical network and radiative transfer calculations, will be presented in a follow-up study (Wiesemeyer et al., in prep., hereafter Paper II), along with the analysis of the variable [C II] emission.

2 Observations and data reduction

2.1 Observations of the [C I] line

We performed observations towards IRC+10216 in the 3P1 → 3P0 fine structure line of atomic carbon with the Atacama Pathfinder Experiment (APEX) 12 m sub-millimetre telescope2. The star was observed under the project ID M-0108.F-9515C-2021 between 2021 November 15 and 30 with the nFLASH460 receiver, a new facility frontend built by the Max-Planck-Institut für Radioastronomie (MPIfR) for APEX, while a fast Fourier transform spectrometer (FFTS; Klein et al. 2006) was used as the backend, providing a spectral resolution of 61 kHz. nFLASH460 is a dual sideband receiver with two polarisations and covers an intermediate frequency (IF) range from 4–8 GHz. We tuned the lower sideband (LSB) to the studied [C I] line at a frequency of 492.160700 GHz (Haris & Kramida 2017). The observations took place in good weather with a precipitable water vapour level below 0.6 mm, translating to sightline system temperatures ranging from 400 to 1200 K. We adopted a forward efficiency (ηf) of 0.95 and a main beam efficiency (ηmb) of 0.483 to calibrate the spectra to the main-beam brightness temperature scale (Tmb). At the frequency of 492 GHz, the telescope has a beamwidth of 12.7″ full width at half maximum (FWHM). The integrations were carried out in a cross along right-ascension and declination offsets, comprising 33 positions, for a total of 9 h equally shared between the target and a reference 1000″ away, safely outside of the astrosphere. The observed sky-plane positions are spaced by 6.5″ out to 26″ from the star, and by 13″ beyond, up to a maximum distance of 78″. Frequent line pointings were performed on the CO(4−3) line, with the secondary wobbling with a throw of 120″ at a frequency of 1.5 Hz.

We used the CLASS software from the GILDAS4 package (Pety 2005) to reduce and further analyse the data. The individual steps consist of masking irrelevant parts of the spectra, subtracting linear baseline fits (with baseline noise figures ranging from 38 mK to 120mK on Tmb scale), and averaging adjacent spectral channels, so as to increase the signal-to-noise ratio while preserving an appropriate velocity resolution of 0.2 km s−1, corresponding to five spectral channels in the unprocessed data. Spectra from equivalent positions (or distances from the centre) are averaged with 1/σ2 weighting (i.e. the inverse of the squared baseline noise).

2.2 Observations of the [C II] line

The 2P3/2 → 2P1/2 fine-structure line of ionised carbon, [C II], at 1900.5369 GHz was observed with the upGREAT 7-pixel, dual polarisation receiver array for THz spectroscopy (Risacher et al. 2018) aboard the Stratospheric Observatory For Infrared Astronomy (SOFIA). The [C II] line was tuned to the upper sideband, so as to avoid a contamination by telluric features from the image band of the mixer (operation at THz frequencies precludes sideband separation). The data were acquired on a total of twelve SOFIA flights, spanning 1.4 cycles of the stellar surface pulsation of 630 days (Menten et al. 2012). Two more flights targeting IRC +10216 were conducted in 2021 and 2022. The typical (median) on-sky observing time was 1 h (including calibrations and updates of the gyroscope settings), of which typically 40 min were used for integrating towards the target, equally shared between the on-position and the off-positions on either side (chop-and-nod method, with a throw of 2′). The data were taken on subscan granularity (40 s including both chop phases) at elevations ranging from 23° to 52°. Median water vapour columns of 10 µm (as referred to zenith) and single-sideband receiver temperatures of 1770K resulted in a median single-sideband system temperature of 1870 K.

The spectroscopic analysis was performed with fast Fourier transform spectrometers (XFFTSs, Klein et al. 2012), covering the instantaneous 4 GHz wide bandpass with 16384 channels, resulting in a channel separation of 244 kHz. Data processing followed steps similar to those applied to the [C I] data; here we removed second-order spectral baselines and averaged them to a spectral resolution of 1.5 km s−1, resulting in a baseline noise of 16 mK (rms value, in Rayleigh-Jeans equivalent main-beam brightness temperature). The conversion from backend count rates to forward-beam antenna temperatures was done using calibration loads at ambient and cold temperature (typically around 295 and 100 K, respectively), and adopting a forward efficiency of 0.97. The beam efficiency was calibrated by frequent observations of Mars, and determined to a coupling ranging from ηmb = 0.65 to 0.70 (i.e. to the physically possible limit set by the design of the telescope’s optics). This interval reflects the inherent calibration uncertainty. At the frequency of the [C II] line, the main beam of the instrument’s Airy pattern has a width of 14″ (FWHM), which is merely 10% wider than the beam at the frequency of the 3P1 → 3P0 [C I] transition and thus facilitates the comparison of the line profiles.

3 Results

3.1 [C I]emission

Since the CO emission from the envelope is largely centro-symmetric (Cernicharo et al. 2015), and C0 and C+ are products of photochemistry, we averaged the spectra from the half-beam sampled, equivalent offsets from the star (Fig. 1). Evidence for deviations from this symmetry is discussed below; individual spectra, extending to up to 78″ from the star, are shown in Fig. A.1. We have analysed the spectra using various methods, from a simple phenomenological description assuming constant excitation to a radiative transfer model of the non-local thermo-dynamic equilibrium (non-LTE) excitation of [C I] throughout a homogeneous envelope, characterised by constant gradients in temperature and density, and expanding at a constant velocity.

3.2 [C I] emission: Line profile fitting

The emission from the centre and the innermost sightlines (up to a 13″ projected distance) displays well-separated blue- and red-shifted peaks (horn-shaped line profiles) from the hemispheres expanding towards and away from the observer, respectively. Beyond the 19.5″ offset, the components start to merge because the sightline-projected expansion velocity decreases. We have fit the observations (up to 26″ offsets) with individual Gaussian profiles by masking the emission component at the respective opposite velocity. The resulting parameters are reported in Table A.1. At 39″ significant emission can only be detected after smoothing these spectra to a resolution of 3 km s−1 with a rather flat-topped profile of 0.08 K and rms noise of 31.8 mK. For closer inspection, we provide the spectrum of the sightline in Fig. A.2. At the 52″ offset and beyond, we see no significant emission anymore.

The individual spectra (Fig. A.1) observed at the same distance from the star show no strikingly remarkable differences. The central sightline features a slightly asymmetric double-peak profile, which can be attributed to a weak absorption in the blue wing, because in the corresponding hemisphere the sight-line crosses layers of decreasing excitation. We note, however, that the blue- and red-shifted fractions of the line area display the opposite behaviour. The same holds for the observations of Keene et al. (1993) at 15″ (FHWM) resolution, who reported peak temperatures of less than 0.4 K, falling 0.3 K below ours. While the discrepancy may be partly due to the individual coupling of the [C I] emission to the different main beams or due to residual pointing or calibration errors, we note that our line flux of 3.1 Kkm s−1 is in agreement with that of Keene et al. (1993). We concede that with the ≃30 yr time span between these observations and ours, the [C I]-emitting shell has expanded by almost 90 AU (or 0.7″ at the adopted distance of 130 pc). While such a distance is too short to be noticeable in the beams of single-dish sub-millimetre telescopes, it may be long enough (17 pulsation periods of 630 days, Menten et al. 2012) to possibly alter the line profiles if the shell and its photochemistry are not in a steady state. From the asymmetry of the line profile, which is weak at most, we conclude that opacity effects cannot be very important in the formation of the line. At offsets of −19″ and −26″ (in either right ascension or declination), the spectra show more centre-filling emission than the direct fits do (i.e. around the systemic velocity of −26.5 km s−1).

3.3 C0 excitation modelling

Given the ad hoc approach of the direct fitting, we now use a more sophisticated method to model the excitation of [C I] throughout an envelope characterised by density, abundance, and temperature gradients. We performed radiative transfer modelling to describe the observed spectral lines and determine the abundance and temperature profiles. We used the ID radiative transfer code RATRAN5 (Hogerheijde & van der Таk 2000), where a spherically symmetric model is assumed. For the model, we take a distance of 130 pc to the star with a mass-loss rate of 2 × 10−5 M⊙ yr−1 (Cernicharo et al. 2015; Reach et al. 2022; Fonfría et al. 2022). Furthermore, the gas is expanding at a velocity of 14 km s−1. The line profile function is Gaussian with a 1/e width of 2.0 km s−1. Under a turbulent contribution of 1.5 km s−1 (based on observations of CO and C2H, De Beck et al. 2012), the resulting thermal width of () km s−1 would be too large to be accounted for by our temperature profile. We note, however, that non-thermal line broadening in circumstellar envelopes is poorly constrained, and that magnetohydrodynamic turbulence cannot be ruled out on grounds of CN Zeeman measurements revealing significant deviations from a homogeneous magnetic field structure (Duthu et al. 2017). In such a picture, the fact that the emissions of CO and [C I] trace spatially distinct envelope regions could then reconcile the different estimates of turbulent line broadening. The excitation model accounts for far-infrared pumping by the dust emission (with an assumed gas-to-dust ratio of 100), and assumes that the temperatures of gas and dust are tightly coupled.

Reach et al. (2022) assume that the abundances of C+ and C0 in the outer envelope derive entirely from the photo-dissociation of CO (see their Fig. 3). The abundance of the latter is parameterised by the photo-destruction radius rphot and steepness parameter α, (1)

Here we adopt their model with rphot = 3.5 × 1017 cm, and α = 3.0 (cf. Saberi et al. 2019), for an underlying elemental abundance of [C/H2] = 8 × 10−4, of which at the inner boundary of the model 75% is contained in CO, while the remainder accounts for acetylene (C2H2) and HCN. The model parameters are summarised in Fig. 4 and Table 1. While this model provides a satisfactory description of the azimuthally averaged spectra (Fig. 1, baseline noise 67–128 mK rms) or on individual positions (Fig. A.1), we have to test whether it entails uniqueness or not. We therefore overlay the corresponding spectra obtained from adjusting the spectra obtained from a constant-abundance model to the observed ones. A reasonable fit can be obtained with n(C0)/n(H) = 4 × 10−5 and rin=4 × 1016 cm, but otherwise the model has the same parameters. This abundance falls an order of magnitude below that of elemental carbon (e.g. [C/H] = 2.1 × 10−4 from young B-type stars, Nieva & Przybilla 2012), because it attributes elemental carbon to C0 at distances from the star where it should still be taken up by mainly CO.

Figure 1 shows both models overlaid on the observations, reproducing a weakly self-absorbed blue-shifted peak towards the centre and the overall distribution of the [C I] emission. On the innermost sightlines, both models reproduce the observed spectral profiles reasonably well. At 19.5″ from the star, the variable-abundance model somewhat underestimates the emission arising closer to the systemic velocity, by 1–2 σrms, whereas the constant abundance model works well. Beyond this offset, both models agree with each other fairly well. We infer that the model of Reach et al. (2022) can be used to describe the inner boundary of the C0 shell; whereas, in the constant-abundance model, a C0 abundance that is a factor two higher could be compensated for by displacing the inner boundary from 2.3 × 1016 cm outwards to 4 × 1016 cm. Such a degeneracy reflects the scaling expected from the adopted density and temperature profiles and the spherical symmetry. Shifting the inner boundary even more towards the photo-destruction radius of CO indeed brings the abundance of C0 required for a good fit closer to the cosmic, elemental carbon abundance. However, an authoritative chemical model cannot be deduced from these observations alone. We, therefore, continue with a closer look at the [C II] emission from the central sightline.

thumbnail Fig. 1

[C I] 3P1 → 3P0 averaged spectra towards nine offsets from the star (in purple) overlaid with the RATRAN model for variable abundance (in green) and constant abundance (in orange).
Table 1

Input parameters for the RATRAN model of the [C I] emission.

thumbnail Fig. 2

As Fig. 1, except that the [C I] spectra are overlaid with those from the contant-abundance, thick-shell model (in orange), and a thin-shell model (in green). The latter clearly fails on all sightlines except for the two innermost ones.
thumbnail Fig. 3

[C II] 2P3/2 → 2Pı/2 spectrum (2015–2021 average) towards the central line of sight. The average off-centre spectrum (at 32″ from the centre) is overlaid in green. The systemic velocity (Vsys) is marked with a red dotted line, and Vsys+Vexp and Vsys−Vexp values are marked with an orange and blue dotted line, respectively, where Vexp is the gas terminal expansion velocity. The emission feature at 15 km s−1 could be from the HNC J = 21−20 transition, and is marked as such.

3.4 [C II] emission

The average [C II] 2P3/2 → 2P1/2 spectra from the central sight-line and at 32″ from the star are shown in Fig. 3. Unlike the [C I] 3P1 → 3P0 emission, the [C II] emission is only detected towards the central pixel of the upGREAT array. The outer ring of the array’s hexagonal layout displays no detectable [C II] emission. This finding, and the broader components of the strongly asymmetric double-peak profile (FWHM 8.1 and 13.4 km s−1 in the blue- and red-shifted peak, respectively), which are less pronounced than that of [C I], indicate that the emission cannot solely originate from the outer boundary of the circumstellar envelope, exposed to the interstellar radiation field. Additionally, there is an emission feature present at 15 km s−1 velocity, which could be from the HNC J = 21−20 transition. This line’s small width (~5 km s−1), which is much smaller than the IRC+10216 CSE’s terminal velocity, suggests an origin close to the stellar photosphere, following the conclusions Patel et al. (2009, 2011) drew for similarly narrow vibrationally excited lines from various molecules. Consistent with Reach et al. (2022), we do not detect any emission from the [13C II] ion. Furthermore, the [C II] emission was found to be variable on a timescale corresponding to that of the stellar pulsation (paper II). This and the signature of self-absorption in the blue-shifted line component corroborate the presence of a contribution from within the envelope. Although with time the [C II] line profile shown in Fig. 3 varies both in strength and shape, its overall appearance (broad, strongly asymmetric peaks) is persistent.

thumbnail Fig. 4

Model profiles of (left) kinetic temperature, (centre) densities of C0 and H2 (the latter taken from Reach et al. 2022), and (right) C0 abundance (from Reach et al. 2022), used as input for the RATRAN model.

4 Discussion

The [C I] 3P1 → 3P0 lines display profiles that are characteristic for optically thin emission from a geometrically extended, expanding envelope. The profiles towards the star, and towards the envelope’s limb, could be readily explained by a spatially resolved optically thin shell, whose atomic carbon is produced from the dissociation of CO by the interstellar radiation field: towards the central position, the double-peaked “horn profile” is attributed to the front and rear part of the expanding envelope (the slight asymmetry between the blue- and red-shifted peaks is commented on below), while the single-peaked, flat-topped spectra beyond 26″ from the star are dominated by sightline elements of which the expansion velocity vector is mostly parallel to the sky plane. The interstitial positions, however, show that the shell cannot be geometrically thin, thanks to the emission close to the systemic velocity filling the double-horn profile. Indeed, a model consisting of a thin, constant-abundance shell extending from 1017 to 1.6 × 1017 cm fits only the central spectrum and that at a 6.5″ distance; however, from a 13″ to 39″ distance, it cannot provide the emission observed close to systemic velocity (Fig. 2). At 52″ and beyond, where the shell starts to emit at a zero sightline-projected velocity, that is −26.5 km s−1 in local standard of rest (LSR) velocity, the double-peak profile is still too pronounced and strong, while the observed emission starts to fall below the detection limit. Likewise, a thin shell extending from 1.2 × 1016 to 1.6 × 1016 cm, as advocated by Reach et al. (2022) to explain the [CII] observation of Herschel/HIFI, fails to reproduce our observations as well (see paper II).

Curiously, the [C II] 2P3/2 → 2P1/2 line towards the star (Fig. 3) does not show a narrow double-peak profile either (components ~10 km s−1 wide at a corresponding halfmaximum); although in a picture where C+ is produced by the exposure to an external radiation field, the line should originate outwards from the [C I] emitting layers. On the other hand, producing the [C II] emission in a layer adjacent to the stellar photosphere is not easily understandable either, because the beam dilution there would require an unusually high emissivity.

The absence of detectable [C II] emission at a projected distance of ≳20″–30″ from the star and its presence towards it (Fig. 3) suggest that the environment of the inner shell harbours an additional source of ionisation. As a matter of fact, at a 700 AU distance, the star cannot drive any photo-chemistry (1016 cm, Fig. 5). This led Reach et al. (2022) to suggest that mass transfer from the star to a nearby companion locally enhances the UV radiation field, thanks to the shock forming in response to the accretion flow. The presence of such a companion, on a highly eccentric orbit, was indeed advocated by Cernicharo et al. (2015), searching for an explanation for the multiple, partial shells discovered in the optical by light scattering (Mauron & Huggins 1999) and, subsequently, in thermal dust emission (beyond 1′ from the star, Decin et al. 2011) and CO emission (Cernicharo et al. 2015; Decin et al. 2015; Guélin et al. 2018). The depth and extent of the CO studies demonstrate that a solar-like companion star would provide an explanation for the multiple shells, which form neither in response to thermal pulses, nor to the stellar pulsation, which would result in larger or shorter spacings, respectively. The binary hypothesis was also adopted by Siebert et al. (2022, further references therein), in support of a photo-chemical model of cyanoacetylene (HC3N) emission from the inner envelope, which would explain the presence of other molecules closer to the star than expected (CH3CN, C4H2) or typical of O-rich stars (H2O). With the data at hand, and the restriction of our models to spherical symmetry, we cannot provide an argument favouring a binary system. We note that Matthews et al. (2018) confirm the proper motion determined for IRC +10216’s by Menten et al. (2012) with longer time span data and find no evidence of statistically significant astrometric perturbations as would be expected from a binary companion, despite claims to the contrary. As discussed above, some asymmetry seen in the [C I] profiles is suggestive of a fingerprint of the multiple-shell structure. Given that it is evident in CO (2–1) emission observed at 11″ resolution (FWHM, Cernicharo et al. 2015), it should also be discernible in our 13″ resolution [C I] data. The Fourier transforms analysis of Guélin et al. (2018) reveals a 16″ shell spacing in all quadrants except in the south-eastern one, where our direct line-profile fits display larger residuals. Whether this observation is coincidental or not can only be settled by a full, critically sampled [C I] map.

5 Conclusion

Following the pioneering observations of atomic carbon in IRC +10216 by Keene et al. (1993), we present more sensitive and more extensive observations of the 3P1 → 3P0 emission from the [C I] fine structure triplet. Accounting for differences in the data quality, and allowing for an underlying variability which might show after the 30 yr time lapse between the observations, we notice an overall good agreement. We present a model employing an abundance gradient that best reproduces the lines profiles at increasing distance from the star; however, even a model employing a constant C0 abundance would work. We infer that the inner boundary of the [C I]-emitting shell is at ~1016 cm from the star, but cannot yet locate it more precisely. The fact that the [C II] line is seen only towards the inner sightlines through the circumstellar envelope (≲4 000 AU), while the outer envelope, ranging to up to ~ 10 000 AU, is void of [C II] emission, remains an enigma. Both findings call for a full map of the [C I] emission and deeper observations of the [C II] line. The line profiles obtained so far from both tracers exclude a picture in which atomic carbon is only produced in a thin low-density shell fully exposed to the interstellar radiation field.

thumbnail Fig. 5

Comparison of cross-sections from the Leiden Observatory database for photochemistry (Heays et al. 2017) for the photo-ionisation of СI and the photo-dissociation of CO (gold curves top and bottom, respectively, with right-hand ordinates) with SEDs of the attenuated stellar photosphere (green) and the interstellar radiation field (purple) at 1016 cm from the star (with left-hand ordinate). The corresponding dotted curves refer to the unattenuated and undiluted radiations fields. Dust extinction coefficients were calculated from Rouleau & Martin (1991). The frequency scale is linear, so as to enhance its high-frequency end.

Acknowledgements

This publication is based on data acquired with the Atacama Pathfinder Experiment (APEX). APEX is a collaboration between the Max-Planck-Institut für Radioastronomie, the European Southern Observatory, and the Onsala Space Observatory. The authors thank the anonymous referee for thoughtful comments. MJ. thanks Ankit Rohatgi for making the WebPlotDig-itizer tool open source (https://automeris.io/WebPlotDigitizer/). MJ. also thanks Ivalu Barlach Christensen for her help in fixing some python scripts. This research has made use of NASA’s Astrophysics Data System. This work has also made use of Python libraries including NumPy (https://www.numpy.org/) (van der Walt et al. 2011), SciPy (https://www.scipy.org/) (Jones et al. 2001), and Matplotlib (https://matplotlib.org/) (Hunter 2007).

1

We denote neutral and ionised carbon as C0 and C+, respectively, while their fine-structure line emission is labelled [C I] and [C II], respectively.

2

This publication is based on data acquired with APEX, which is a collaboration between the MPIfR, the European Southern Observatory, and the Onsala Space Observatory.

Appendix A Individual results for directly fitted and modelled [C I] line profiles

Table A.1

Parameters deduced from direct fits to the observed spectra.

thumbnail Fig. A.1

Observed [C I] spectra (in black) and line profiles modelled with RATRAN (in purple) at the sampled positions.
thumbnail Fig. A.2

[C I] emission smoothed to 3 km s−1at 39″ from the star.

References

  1. Agúndez, M., Cernicharo, J., & Guélin, M. 2014, A&A, 570, A45 [Google Scholar]
  2. Agúndez, M., Martínez, J. I., de Andres, P. L., Cernicharo, J., & Martín-Gago, J. A. 2020, A&A, 637, A59 [Google Scholar]
  3. Bujarrabal, V., Castro-Carrizo, A., Alcolea, J., et al. 2016, A&A, 593, A92 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  4. Cernicharo, J., Guélin, M., & Kahane, C. 2000, A&As, 142, 181 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  5. Cernicharo, J., Marcelino, N., Agúndez, M., & Guélin, M. 2015, A&A, 575, A91 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  6. Cerrigone, L., Menten, K. M., & Kamiński, T. 2012, A&A, 542, A15 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  7. Crosas, M., & Menten, K. M. 1997, ApJ, 483, 913 [NASA ADS] [CrossRef] [Google Scholar]
  8. Darwent, B. 1970, Bond Dissociation Energies in Simple Molecules, NBS NSRDS (United States Department of Commerce) [Google Scholar]
  9. De Beck, E., & Olofsson, H. 2018, A&A, 615, A8 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  10. De Beck, E., Lombaert, R., Agúndez, M., et al. 2012, A&A, 539, A108 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  11. Decin, L., Royer, P., Cox, N.L.J., et al. 2011, A&A, 534, A1 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  12. Decin, L., Richards, A. M. S., Neufeld, D., et al. 2015, A&A, 574, A5 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  13. Dharmawardena, T. E., Kemper, F., Scicluna, P., et al. 2018, MNRAS, 479, 536 [NASA ADS] [Google Scholar]
  14. Duthu, A., Herpin, F., Wiesemeyer, H., et al. 2017, A&A, 604, A12 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  15. Fonfría, J. P., DeWitt, C. N., Montiel, E. J., Cernicharo, J., & Richter, M. J. 2022, ApJ, 927, L33 [CrossRef] [Google Scholar]
  16. Glab, W. L., Haris, K., & Kramida, A. 2018, J. Phys. Commun., 2, 055020 [Google Scholar]
  17. Groenewegen, M. A. T. 2017, A&A, 606, A67 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  18. Groenewegen, M. A. T., Barlow, M. J., Blommaert, J. A. D. L., et al. 2012, A&A, 543, A8 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  19. Guélin, M., Patel, N. A., Bremer, M., et al. 2018, A&A, 610, A4 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  20. Haris, K., & Kramida, A. 2017, ApJS, 233, 16 [Google Scholar]
  21. Heays, A. N., Bosman, A. D., & van Dishoeck, E. F. 2017, A&A, 602, A105 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  22. Höfner, S., & Olofsson, H. 2018, A&ARv, 26, 1 [Google Scholar]
  23. Hogerheijde, M. R., & van der Tak, F. F. S. 2000, A&A, 362, 697 [NASA ADS] [Google Scholar]
  24. Hunter, J. D. 2007, Comput. Sci. Eng., 9, 90 [Google Scholar]
  25. Jaffe, D. T., Harris, A. I., Silber, M., Genzel, R., & Betz, A. L. 1985, ApJ, 290, L59 [NASA ADS] [CrossRef] [Google Scholar]
  26. Jentschura, U. D., Kotochigova, S., Le Bigot, E.-O., Mohr, P. J., & Taylor, B. N. 2005, Phys. Rev. Lett., 95, 163003 [CrossRef] [Google Scholar]
  27. Jones, E., Oliphant, T., Peterson, P., et al. 2001, SciPy: Open source scientific tools for Python [Google Scholar]
  28. Karakas, A. I., & Lattanzio, J. C. 2014, PASA, 31, e030 [NASA ADS] [CrossRef] [Google Scholar]
  29. Keene, J., Young, K., Phillips, T. G., Buettgenbach, T. H., & Carlstrom, J. E. 1993, ApJ, 415, L131 [NASA ADS] [CrossRef] [Google Scholar]
  30. Klein, B., Philipp, S. D., Krämer, I., et al. 2006, A&A, 454, L29 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  31. Klein, B., Hochgürtel, S., Krämer, I., et al. 2012, A&A, 542, L3 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  32. Ladjal, D., Barlow, M. J., Groenewegen, M. A. T., et al. 2010, A&A, 518, L141 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  33. Li, X., Millar, T. J., Walsh, C., Heays, A. N., & van Dishoeck, E. F. 2014, A&A, 568, A111 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  34. Mamon, G. A., Glassgold, A. E., & Huggins, P. J. 1988, ApJ, 328, 797 [Google Scholar]
  35. Martin, D. C., Seibert, M., Neill, J. D., et al. 2007, Nature, 448, 780 [NASA ADS] [CrossRef] [Google Scholar]
  36. Matthews, L. D., Reid, M. J., Menten, K. M., & Akiyama, K. 2018, AJ, 156, 15 [NASA ADS] [CrossRef] [Google Scholar]
  37. Mauron, N., & Huggins, P. J. 1999, A&A, 349, 203 [NASA ADS] [Google Scholar]
  38. McGuire, B. A. 2022, ApJS, 259, 30 [NASA ADS] [CrossRef] [Google Scholar]
  39. Men’shchikov, A. B., Balega, Y., Blöcker, T., Osterbart, R., & Weigelt, G. 2001, A&A, 368, 497 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  40. Menten, K. M., Reid, M. J., Kamiński, T., & Claussen, M. J. 2012, A&A, 543, A73 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  41. Millar, T. J., & Herbst, E. 1994, A&A, 288, 561 [NASA ADS] [Google Scholar]
  42. Millar, T. J., Herbst, E., & Bettens, R. P. A. 2000, MNRAS, 316, 195 [NASA ADS] [CrossRef] [Google Scholar]
  43. Morris, M., & Jura, M. 1983, ApJ, 264, 546 [Google Scholar]
  44. Nieva, M. F., & Przybilla, N. 2012, A&A, 539, A143 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  45. Papadopoulos, P. P., Thi, W. F., & Viti, S. 2004, MNRAS, 351, 147 [NASA ADS] [CrossRef] [Google Scholar]
  46. Pardo, J. R., Cernicharo, J., Tercero, B., et al. 2022, A&A, 658, A39 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  47. Patel, N. A., Young, K. H., Brünken, S., et al. 2009, ApJ, 692, 1205 [NASA ADS] [CrossRef] [Google Scholar]
  48. Patel, N. A., Young, K. H., Gottlieb, C. A., et al. 2011, ApJS, 193, 17 [NASA ADS] [CrossRef] [Google Scholar]
  49. Pety, J. 2005, in SF2A-2005: Semaine de l’Astrophysique Française, eds. F. Casoli, T. Contini, J. M. Hameury, & L. Pagani, 721 [Google Scholar]
  50. Phillips, T. G., Huggins, P. J., Kuiper, T. B. H., & Miller, R. E. 1980, ApJ, 238, L103 [NASA ADS] [CrossRef] [Google Scholar]
  51. Reach, W. T., Ruaud, M., Wiesemeyer, H., et al. 2022, ApJ, 926, 69 [NASA ADS] [CrossRef] [Google Scholar]
  52. Richichi, A., Chandrasekhar, T., & Leinert, C. 2003, New A, 8, 507 [NASA ADS] [CrossRef] [Google Scholar]
  53. Risacher, C., Güsten, R., Stutzki, J., et al. 2018, J. Astron. Instrum., 7, 1840014 [NASA ADS] [CrossRef] [Google Scholar]
  54. Rouleau, F., & Martin, P. G. 1991, ApJ, 377, 526 [NASA ADS] [CrossRef] [Google Scholar]
  55. Saberi, M., Vlemmings, W. H. T., & De Beck, E. 2019, A&A, 625, A81 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  56. Sahai, R., & Chronopoulos, C. K. 2010, ApJ, 711, L53 [NASA ADS] [CrossRef] [Google Scholar]
  57. Santoro, G., Martínez, L., Lauwaet, K., et al. 2020, ApJ, 895, 97 [CrossRef] [Google Scholar]
  58. Schöier, F. L., & Olofsson, H. 2001, A&A, 368, 969 [Google Scholar]
  59. Siebert, M. A., Van de Sande, M., Millar, T. J., & Remijan, A. J. 2022, ApJ, 941, 90 [NASA ADS] [CrossRef] [Google Scholar]
  60. Van de Sande, M., Sundqvist, J. O., Millar, T. J., et al. 2018, A&A, 616, A106 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  61. van der Veen, W. E. C. J., Huggins, P. J., & Matthews, H. E. 1998, ApJ, 505, 749 [NASA ADS] [CrossRef] [Google Scholar]
  62. van der Walt, S., Colbert, S. C., & Varoquaux, G. 2011, Comput. Sci. Eng., 13, 22 [Google Scholar]
  63. Weiss, A., & Ferguson, J. W. 2009, A&A, 508, 1343 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  64. Wolfire, M. G., McKee, C. F., Hollenbach, D., & Tielens, A. G. G. M. 2003, ApJ, 587, 278 [Google Scholar]

All Tables

Table 1

Input parameters for the RATRAN model of the [C I] emission.

In the text
Table A.1

Parameters deduced from direct fits to the observed spectra.

In the text

All Figures

thumbnail Fig. 1

[C I] 3P1 → 3P0 averaged spectra towards nine offsets from the star (in purple) overlaid with the RATRAN model for variable abundance (in green) and constant abundance (in orange).
In the text
thumbnail Fig. 2

As Fig. 1, except that the [C I] spectra are overlaid with those from the contant-abundance, thick-shell model (in orange), and a thin-shell model (in green). The latter clearly fails on all sightlines except for the two innermost ones.
In the text
thumbnail Fig. 3

[C II] 2P3/2 → 2Pı/2 spectrum (2015–2021 average) towards the central line of sight. The average off-centre spectrum (at 32″ from the centre) is overlaid in green. The systemic velocity (Vsys) is marked with a red dotted line, and Vsys+Vexp and Vsys−Vexp values are marked with an orange and blue dotted line, respectively, where Vexp is the gas terminal expansion velocity. The emission feature at 15 km s−1 could be from the HNC J = 21−20 transition, and is marked as such.
In the text
thumbnail Fig. 4

Model profiles of (left) kinetic temperature, (centre) densities of C0 and H2 (the latter taken from Reach et al. 2022), and (right) C0 abundance (from Reach et al. 2022), used as input for the RATRAN model.
In the text
thumbnail Fig. 5

Comparison of cross-sections from the Leiden Observatory database for photochemistry (Heays et al. 2017) for the photo-ionisation of СI and the photo-dissociation of CO (gold curves top and bottom, respectively, with right-hand ordinates) with SEDs of the attenuated stellar photosphere (green) and the interstellar radiation field (purple) at 1016 cm from the star (with left-hand ordinate). The corresponding dotted curves refer to the unattenuated and undiluted radiations fields. Dust extinction coefficients were calculated from Rouleau & Martin (1991). The frequency scale is linear, so as to enhance its high-frequency end.
In the text
thumbnail Fig. A.1

Observed [C I] spectra (in black) and line profiles modelled with RATRAN (in purple) at the sampled positions.
In the text
thumbnail Fig. A.2

[C I] emission smoothed to 3 km s−1at 39″ from the star.
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.