![]() | ![]() |
Astron. Astrophys. 363, 629-639 (2000) 1. IntroductionSilicate carbon stars were first discovered in studies of IRAS LRS (Low-Resolution Spectrometer) spectra (Willems & de Jong 1986; Little-Marenin 1986). The LRS spectra of these carbon stars exhibit emission features of amorphous silicate dust at 10 and 18 µm, typical for an oxygen-rich environment. Subsequently, H2O and OH masers were detected from some silicate carbon stars (Nakada et al. 1987; Little-Marenin et al. 1988 , 1994; te Lintel Hekkert et al. 1991; Engels & Leinert 1994), confirming the O-rich chemistry in the circumstellar material of these stars. No detection of SiO or HCN pure-rotational lines has been reported (Little-Marenin et al. 1994). Further searches in the LRS data (Chan & Kwok 1991; Kwok & Chan 1993; Kwok et al. 1997) have resulted in the discovery of 19 stars in this peculiar class. The nature of the silicate carbon stars is still unclear.
Little-Marenin (1986) proposed that the objects are binaries
consisting of a carbon-rich and an oxygen-rich giant. This idea is now
thought to be unlikely, since there is no evidence of O-rich giants in
the objects from spectroscopic observations (Lambert et al. 1990) or
infrared speckle interferometry (Engels & Leinert 1994). Willems
& de Jong (1986; also Chan & Kwok 1991) suggested that these
objects have just experienced a transition from oxygen-rich to a
carbon star by a recent ( The most widely accepted idea nowadays to explain the phenomenon is the following (Morris 1987; Lloyd Evans 1990). Suppose these are all binaries with an unseen companion star, probably a main-sequence star. While the star was an oxygen-rich giant, some portion of the mass ejected was captured and stored in a disk around the companion star. Then, the central star experienced thermal pulses and evolved to a carbon star. The numerical calculations by Mastrodemos & Morris (1998) show that a permanent disk around the companion can actually be formed. On the other hand, Kahane et al. (1998) proposed a circum-binary reservoir, i.e. a disk surrounding the whole system. Radial velocity variations have been detected in two silicate carbon stars, BM Gem and EU And (Barnbaum et al. 1991), confirming the binarity of these objects. It is known that most, if not all, silicate carbon stars are J-type carbon stars, i.e. carbon stars with enhanced 13C and depletion of s-process elements (Willems & de Jong 1986; Lloyd Evans 1990). However, not all J-type carbon stars show silicate dust features. Observations with the Infrared Space Observatory (ISO; Kessler et
al. 1996) have revealed that many evolved stars show evidence of a
complex chemistry in their circumstellar material. Waters et al.
(1998) reported that the spectrum of a post-AGB binary system, the
Red Rectangle, taken by the Short-Wavelength Spectrometer (SWS;
de Graauw et al. 1996), shows both Polycyclic-Aromatic Hydrocarbons
(PAH) features and crystalline silicate features. Mid-infrared images
show that the PAH features arise from the bipolar mass flow, i.e.
oxygen-rich materials are probably located in the equatorial disk,
while PAHs are in the present-day stellar wind. Another remarkable
example is IRAS 09425-6040 (Molster et al. 2000). The nature of
this object is not well understood. The SWS spectrum up to
Thanks to its large wavelength coverage and improved spectral resolution, observations with the ISO/SWS are expected to provide further information to clarify the nature of the silicate carbon stars. Unfortunately, many silicate carbon stars bright enough for SWS spectroscopy were not visible during the life span of ISO. The only available star, V778 Cyg, was observed in the guaranteed time programme AGBSTARS (P.I. T. de Jong). V778 Cyg is of spectral type C4,5J (Yamashita 1975). H2O and OH masers have been detected (Nakada et al. 1987; Little-Marenin et al. 1988; Deguchi et al. 1988; Engels & Leinert 1994). In this paper, we present the result of the ISO/SWS observation of V778 Cyg both on the dust features (Sect. 3) and the molecular features (Sect. 4). A possible origin of the silicate emission is proposed (Sect. 5). The formation and stability of disks in binary systems (Sect. 6), and possible solutions for V778 Cyg and other objects are discussed (Sect. 7). Finally, we elaborate on the nature of the silicate carbon stars (Sect. 8).
© European Southern Observatory (ESO) 2000 Online publication: December 11, 2000 ![]() |