Silicate carbon stars are carbon-rich evolved stars whose circumstellar dust shows the 10 and 18 micrometre emission of oxygen-rich amorphous silicates, a combination that single-star asymptotic giant branch evolution cannot explain, since dust condensing in a carbon-rich wind should itself be carbon-bearing. A 2006 cross-identification of the third General Catalog of Galactic Carbon Stars with the expanded IRAS Low-Resolution Spectra database identified nine new silicate carbon stars, raising the Galactic census to 35, but left six candidate identifications ambiguous because more than one USNO/2MASS source of compatible magnitude lies inside each arcminute-scale IRAS positional error ellipse. The coupled open questions that follow from that work and its successors are: first, completing the census, that is, determining unique, spectroscopically confirmed counterparts for the ambiguous and unconfirmed candidate silicate carbon stars; and second, establishing the population-wide physical origin, most prominently whether every silicate carbon star is a binary whose oxygen-rich material, shed while the primary was an oxygen-rich M giant, survives in a long-lived circumbinary or circum-companion disk, versus transition-object or single dual-dust-chemistry alternatives. Later literature enlarged the census to 44 collected objects (29 confirmed) and strengthened binary and disk evidence for individual stars, but as of 2020 neither the candidate confirmations nor the population-wide origin is settled; the problem is open.

Background

Stars of roughly 0.8 to 8 solar masses spend their last nuclear-burning stage on the asymptotic giant branch (AGB), a phase of large radius, cool pulsating photosphere, and slow, dense mass loss. During the AGB, dredge-up (convective mixing that carries interior material to the surface) can raise the photospheric ratio of carbon to oxygen above unity, moving the star through the classical spectral sequence M (oxygen-rich) to S (roughly equal abundances) to C (carbon-rich), where carbon-bearing molecules dominate the spectrum. The condensing dust normally follows the photospheric chemistry: oxygen-rich AGB stars drive winds containing amorphous silicate dust with its characteristic 10 and 18 micrometre emission bands, while carbon stars form carbonaceous dust. The Infrared Astronomical Satellite (IRAS), which surveyed the whole sky in 1983, obtained Low-Resolution Spectra (LRS), mid-infrared spectra covering roughly 8 to 22 micrometres, for about five thousand bright point sources; Kwok, Volk and Bidelman (1997) classified an expanded database of 11,224 LRS spectra into groups by dust-feature type, in which Group E denotes 10 micrometre amorphous silicate emission, the signature of oxygen-rich dust.

Silicate carbon stars (SCSs), first recognized by Little-Marenin (1986) and by Willems and de Jong (1986) in the IRAS data, are carbon stars whose infrared spectra show exactly this oxygen-rich silicate emission. Their discovery poses a challenge to the theory of stellar evolution in the late stage: the currently observed carbon-rich wind cannot have produced oxygen-rich dust, so the dust must be older material surviving around a star that has since become carbon-rich. The most widely accepted model, as summarized by Chen and Zhang (2006), holds that such a star has a low-luminosity companion, and that the oxygen-rich material was shed by mass loss while the primary was an M giant and is stored in a circumbinary disk (a long-lived disk orbiting both stars, proposed by Morris 1987 and later work cited therein) or in a disk around the companion itself. Competing interpretations exist: the transition-object picture, in which the star is caught in the brief switch from oxygen-rich to carbon-rich while the old oxygen-rich shell is still present, and single-star dual-dust-chemistry scenarios, in which a single star simultaneously condenses both oxygen-rich and carbon-rich dust; later authors (Zhao et al. 2020) frame the binary-versus-dual-chemistry distinction as the deciding question.

Chen and Zhang (2006) searched for new SCSs by cross-identifying the newer entries of the third General Catalog of Galactic Carbon Stars (CGCS3; Alksnis et al. 2001, listing 6,891 Galactic carbon stars) with the expanded LRS database of Kwok, Volk and Bidelman (1997). The practical difficulty is astrometric: CGCS3 positions are inhomogeneous and typically accurate only to 10-30 arcseconds, while IRAS sources carry arcminute-scale 95 percent positional error ellipses. Chen and Zhang therefore refined the IRAS associations using the USNO-B1.0 photographic astrometric catalog (about 1 arcsecond positions) and the 2MASS Two Micron All Sky Survey near-infrared catalog, requiring magnitude consistency between the optical, near-infrared, and IRAS fluxes, and also checked the candidates in the standard IRAS two-color diagram used for evolved-star classification. Of about 1,000 new CGCS3 entries they found 469 IRAS associations; 243 had LRS spectra, of which 15 showed Group E silicate emission. Nine of these were unique high-confidence identifications of new silicate carbon stars, raising the Galactic total to 35. For the remaining six, more than one USNO/2MASS source with compatible magnitudes lies within the IRAS error ellipse, so the true counterpart, and hence the status of these six candidates as silicate carbon stars, is undetermined. The six cases are counted but not individually named in the published tables.

For the known, brighter SCSs, a series of studies obtained evidence bearing on the binary and disk picture. Szczerba et al. (2006) used MERLIN (a linked network of UK radio telescopes acting as an interferometer) to resolve the 22 GHz water-vapour maser emission (masers are extremely bright, narrow radio lines from compact masing gas, standard tracers of oxygen-rich circumstellar disks) in V778 Cyg, finding it offset from the carbon star, which they interpreted as an oxygen-rich disk around an unseen companion; Babkovskaia et al. (2006) modelled that emission as a warped maser disc and inferred a companion near 1.7 solar masses with a disc tens of astronomical units across. Ohnaka et al. (2006) resolved the mid-infrared dust of IRAS 08002-3803 with the VLTI and its MIDI mid-infrared instrument (an interferometer combining unit telescopes of the European Southern Observatory's Very Large Telescope), finding a disk-like geometry that standard spherical models fail to reproduce. Ohnaka and Boboltz (2007) mapped the water masers of EU And with the VLBA (a continent-scale US radio interferometer), obtaining a disk-like distribution consistent with an oxygen-rich disk not centred on the carbon star. Izumiura et al. (2008) detected in BM Gem a Balmer continuum excess (extra continuum light from hydrogen recombination, the mark of a hot source) and P Cygni-type line profiles (blueshifted absorption plus redshifted emission, the signature of an outflow) with roughly 400 km/s terminal velocity, evidence for a hot companion with an accretion disk; Ohnaka et al. (2008) resolved an asymmetric silicate dust distribution around the same star. Ohnaka et al. (2013) surveyed ten SCSs in 22 GHz water masers with the VLA (the US Very Large Array radio telescope), detecting five. A conference review by Szczerba, Schmidt and Pulecka (2007) situates the SCS mixed-chemistry puzzle among related late-stellar-evolution objects.

The census itself was subsequently extended and audited. Chen and Shan (2011) cross-matched all Group E LRS sources with CGCS3 and other catalogs and infrared criteria, identifying five new SCSs (raising the census to 39) plus additional candidates and probable candidates, many lacking optical or near-infrared spectroscopic confirmation, and noting that at least one previously claimed SCS had been reclassified as an oxygen-rich M star. Kwon and Suh (2014) compiled a new catalog of 44 SCSs, judged 29 confirmed, and left 15 unconfirmed, some with doubtful carbon status, some suspected of interstellar-cirrus contamination (Galactic foreground dust emission that can mimic a far-infrared source), and three with position offsets larger than 30 arcseconds from the AKARI infrared satellite positions, which they state need more reliable observational data. Zhao et al. (2020) observed 28 SCSs in radio molecular lines, including the first 6 centimetre formaldehyde (H2CO) absorption detections in seven of them, with double-peaked formaldehyde or CO line profiles in six of the seven, which the authors describe as suggesting the possible presence of a companion, and used the water-maser-versus-formaldehyde behaviour to argue for an evolutionary sequence among SCSs. The WISE (Wide-field Infrared Survey Explorer satellite) photometric study of carbon-star classes by Chen, Liu and Shan (2020) further documents the variety of dust features among carbon stars but does not resolve the candidate list. Thus both halves of the problem, the unique identification and spectroscopic confirmation of candidate SCSs, and the population-wide establishment of the binary, long-lived oxygen-rich-reservoir origin against its alternatives, remain open.

Problem Statement

Chen and Zhang (2006) leave a problem with two coupled parts, one identification and one physical, whose connection is intrinsic: only a reliably confirmed census of silicate carbon stars can support population-level conclusions about their origin. The identification part follows directly from that source; the population-wide physical part is framed as the open deciding question by the later literature, most explicitly Zhao et al. (2020), rather than by Chen and Zhang (2006) itself, which summarized the binary, oxygen-rich-reservoir model as the most widely accepted picture.

(1) Completing the census. For the cross-matches between carbon-star catalogs (CGCS3, the third General Catalog of Galactic Carbon Stars, and its successors) and IRAS Low-Resolution Spectra (LRS) Group E sources (the 10 micrometre amorphous-silicate emission class) whose status is undetermined, namely the 15 unconfirmed silicate carbon stars individually named by Kwon and Suh (2014) (Table 2 and Section 3), the 12 candidate and probable-candidate objects named by Chen and Shan (2011) (5 candidates and 7 probable candidates, re-listed in Kwon and Suh 2014, Table 3), and the six cases left ambiguous by Chen and Zhang (2006) because multiple USNO or 2MASS sources (entries of the USNO-B1.0 optical and the Two Micron All Sky Survey near-infrared all-sky astrometric catalogs) of compatible magnitude lie within the IRAS positional error ellipse, determine the unique optical/infrared counterpart and the spectroscopic classification of each candidate, and separate genuine carbon stars with oxygen-rich circumstellar silicate dust from oxygen-rich contaminants, interstellar-cirrus sources, and misclassified objects. The six ambiguous cases of Chen and Zhang (2006) are counted but not individually named in its published tables, so they fall within the scope of the census to the extent that they are recovered by the CGCS3 x LRS Group E cross-match (the 15 Group E matches minus the nine high-confidence objects of that paper's Table 1).

(2) Establishing the physical origin. For the confirmed population, determine whether the silicate carbon star phenomenon is produced by binaries with long-lived oxygen-rich reservoirs, that is, whether each object is (or is not) a binary system whose oxygen-rich material, shed by mass loss while the primary was an M giant, survives in a circumbinary disk (a disk orbiting both stars) or in a disk around a low-luminosity companion, establishing the binary fraction, the companion properties, the reservoir geometry, and the survival timescales; or whether some or all objects are instead transition objects caught during the oxygen-rich to carbon-rich switch, or single stars with dual dust chemistry (one star simultaneously condensing both oxygen-rich and carbon-rich dust). Determine which of the competing scenarios the evidence supports for the population as a whole.

References

  1. Pei-Sheng Chen and Pin Zhang, "Newly Identified Silicate Carbon Stars from IRAS Low-Resolution Spectra", Chinese Journal of Astronomy and Astrophysics 6, 697-704 (2006), DOI 10.1088/1009-9271/6/6/08, https://doi.org/10.1088/1009-9271/6/6/08
  2. I. R. Little-Marenin, "Carbon stars with silicate dust in their circumstellar shells", The Astrophysical Journal 307, L15 (1986), DOI 10.1086/184720, https://doi.org/10.1086/184720
  3. Frits J. Willems and Teije de Jong, "Carbon stars with oxygen-rich circumstellar dust shells. Observational evidence for the onset of the carbon star phase", The Astrophysical Journal 309, L39 (1986), DOI 10.1086/184756, https://doi.org/10.1086/184756
  4. Mark Morris, "Mechanisms for mass loss from cool stars", Publications of the Astronomical Society of the Pacific 99, 1115 (1987), DOI 10.1086/132089, https://doi.org/10.1086/132089
  5. Sun Kwok, Kevin Volk, and William P. Bidelman, "Classification and Identification of IRAS Sources with Low-Resolution Spectra", The Astrophysical Journal Supplement Series 112, 557-584 (1997), DOI 10.1086/313038, https://doi.org/10.1086/313038
  6. A. Alksnis, A. Balklavs, U. Dzervitis, I. Eglitis, O. Paupers, and I. Pundure, "General Catalog of Galactic Carbon Stars by C. B. Stephenson. Third Edition", Open Astronomy (Baltic Astronomy) 10, 1-318 (2001), DOI 10.1515/astro-2001-1-202, https://doi.org/10.1515/astro-2001-1-202
  7. R. Szczerba, M. Szymczak, N. Babkovskaia, J. Poutanen, A. M. S. Richards, and M. A. T. Groenewegen, "Oxygen-rich disk in the V778 Cygni system resolved", Astronomy & Astrophysics 452, 561-565 (2006), DOI 10.1051/0004-6361:20053782, https://doi.org/10.1051/0004-6361:20053782
  8. K. Ohnaka, T. Driebe, K.-H. Hofmann, Ch. Leinert, S. Morel, F. Paresce, Th. Preibisch, A. Richichi, D. Schertl, M. Schöller, L. B. F. M. Waters, G. Weigelt, and M. Wittkowski, "High angular resolution N-band observation of the silicate carbon star IRAS 08002-3803 with the VLTI/MIDI instrument", Astronomy & Astrophysics 445, 1015-1029 (2006), DOI 10.1051/0004-6361:20053474, https://doi.org/10.1051/0004-6361:20053474
  9. Natalia Babkovskaia, Juri Poutanen, Anita M. S. Richards, and Ryszard Szczerba, "A warped m=2 water maser disc in V778 Cyg?", Monthly Notices of the Royal Astronomical Society 370, 1921-1927 (2006), DOI 10.1111/j.1365-2966.2006.10626.x, https://doi.org/10.1111/j.1365-2966.2006.10626.x
  10. K. Ohnaka and D. A. Boboltz, "Imaging the oxygen-rich disk toward the silicate carbon star EU Andromedae", Astronomy & Astrophysics 478, 809-814 (2007), DOI 10.1051/0004-6361:20079030, https://doi.org/10.1051/0004-6361:20079030
  11. Hideyuki Izumiura, Kunio Noguchi, Wako Aoki, Satoshi Honda, Hiroyasu Ando, Masahide Takada-Hidai, Eiji Kambe, Satoshi Kawanomoto, Kozo Sadakane, Bunei Sato, Akito Tajitsu, Wataru Tanaka, Kiichi Okita, Etsuji Watanabe, and Michitoshi Yoshida, "Evidence for a Companion to BM Gem, a Silicate Carbon Star", The Astrophysical Journal 682, 499-508 (2008), DOI 10.1086/589424, https://doi.org/10.1086/589424
  12. K. Ohnaka, H. Izumiura, Ch. Leinert, T. Driebe, G. Weigelt, and M. Wittkowski, "Asymmetric silicate dust distribution toward the silicate carbon star BM Geminorum", Astronomy & Astrophysics 490, 173-178 (2008), DOI 10.1051/0004-6361:200810229, https://doi.org/10.1051/0004-6361:200810229
  13. R. Szczerba, M. R. Schmidt, and M. Pulecka, "Mixed chemistry phenomenon during late stages of stellar evolution", arXiv:0704.0557 (2007), https://arxiv.org/abs/0704.0557
  14. Pei-Sheng Chen and Hong-Guang Shan, "Identification of new silicate carbon stars", Astrophysics and Space Science 335, 457-467 (2011), DOI 10.1007/s10509-011-0631-z, https://doi.org/10.1007/s10509-011-0631-z
  15. Young-Joo Kwon and Kyung-Won Suh, "A new catalog of silicate carbon stars", Journal of the Korean Astronomical Society 47, 123-135 (2014), DOI 10.5303/jkas.2014.47.4.123, https://doi.org/10.5303/jkas.2014.47.4.123
  16. K. Ohnaka, D. A. Boboltz, G. Mulitz-Schimel, H. Izumiura, and M. Wittkowski, "VLA survey of 22 GHz H2O masers toward ten silicate carbon stars", Astronomy & Astrophysics 559, A120 (2013), DOI 10.1051/0004-6361/201322425, https://doi.org/10.1051/0004-6361/201322425
  17. Zhang Zhao, Xi Chen, Simon P. Ellingsen, Andrej M. Sobolev, and Xujia Ouyang, "A Radio Molecular Line Search for Evidence of an O-rich Environment Around Binary in Silicate Carbon Stars", The Astrophysical Journal 890, 34 (2020), DOI 10.3847/1538-4357/ab6594, https://doi.org/10.3847/1538-4357/ab6594
  18. P. S. Chen, J. Y. Liu, and H. G. Shan, "WISE observations for the variety of Galactic carbon stars", Astrophysics and Space Science 365, 156 (2020), DOI 10.1007/s10509-020-03872-9, https://doi.org/10.1007/s10509-020-03872-9
  19. Kyung-Won Suh, "A New Catalog of Asymptotic Giant Branch Stars in Our Galaxy", The Astrophysical Journal Supplement Series 256, 43 (2021), DOI 10.3847/1538-4365/ac1274, https://doi.org/10.3847/1538-4365/ac1274
  20. W. A. Weidmann, J. A. Ahumada, R. Gamen, O. Pintado, S. Paron, L. Gramajo, and F. Di Mille, "Wray 15-811: An unusual runaway star", New Astronomy 98, 101932 (2023), DOI 10.1016/j.newast.2022.101932, https://doi.org/10.1016/j.newast.2022.101932