HIDDEN SECONDARY OR A SHADOW IN THE CIRCUMSTELLAR MATTER?

• lithium resonance lines

  • Half of FS CMa stars have resonance doublet 6 708 Å of the neutral lithium in spectra. This is taken as an unambiguous evidence of the secondary, because lithium has very low ionization potential, about (~5.4 eV). Therefore, spectral lines of Li I can not be present in the spectra of hot stars.
    an unambiguous evidence ?
    If these stars were just ordinary binaries, the detection of neutral lithium lines would be proof of the binarity.
    However, FS CMa are not ordinary binaries, they are surrounded by a geometrically thick disk.
  if observed in FS CMa stars, Li I resonance doublet 6 708 Å is
  • always in the absorption
  • always at the velocity of [OI] λ λ 6 300, 6 364 Å
  • He I 5 876 Å has very sharp central absorption
  ⇒ Li I lines may be formed in the disk
  • A similar situation is in Herbig Ae/Be stars. Polarimetric observations of Alecian et al. (2013). show that Li I does not come from the secondary.

• hot and cold components of the spectra

  • Li I resonance lines do not have to be the only lines formed in the disk. For example, MWC 623 spectrum is composed from B4III and K2II-Ib spectral types. (Zickgraf, 2001) . However, the Hα bisector variability is not possible to explain by the secondary (Polster, 2018 .

• central quasiemission peak

  Central quasiemission peak is taken as the evidence of the binarity. It is assumed to originate near Lagrange point L1. ✘ Hanuschik (1995) showed that this small emission peak is formed through the circumstellar disk, which is optically thin in the continuum and thick in the line center. It may also be a coincidence. Moving bumps across the Hα are frequently observed.

• spectroastrometry

  • The method detects the shift of the spectra on the CCD chip due to the source near the star. An asymmetric bright envelope may cause the spectropolarimetric signal similarly as the secondary.
  • The spectroastrometric signal was detected in FS CMa, HD 50138 and HD 85567 (Baines, 2006) . The observations were interpreted as the evidence of the secondary. However, later Kluska et al. (2016) were not able to interpret the VLTI/PIONIER data of HD 50138 using the secondary. The fit with the observations provided a disk with a spot.

• large mass-loss rate

  • Large mass-loss rate is another argument for the binarity. The found values ( Ṁ= $2.5.{\mathrm{10}}^{-7}-\; 1.5.{\mathrm{10}}^{-6}M$_⊙ ) can not produce a single B-type star. However, the Ṁ can be determined only based on the comparison with the synthetic spectra based on a guess of the properties of the circumstellar disk. Ṁ was determined for HD 87643, AS 78 and IRAS 00470+6429.
    • i) All three models assumed a freely expanding disk. However, the decelerated expanding layers were observed (Kučerová et al., 2013). In this situation, where the matter is accumulated around the star, such description of the wind speed gives a highly overestimated values of Ṁ.
    • ii) These three stars are among the most massive FS CMa stars, however most FS CMa stars are rather late B-type stars. Therefore, this three objects do not represent the group well. Especially because the radiative pressure driving the wind strongly decrease toward late B-type stars.
    • iii) MHD simulations ( Moranchel-Basurto et al., 2023, 2024) show that the disk structure under the influence of the magnetic field from the star is far from a simple prescription. The disk is wide and large matter accumulations and gaps are created.