The Be Star Newsletter, Volume 40 - April 2010
[Previous Article] [Table of Contents] [Next Article]
Periodic behavior of the He I 6678 emission in δ ScoErnst Pollmann
δ Sco is a binary system with an eccentric orbit which is exhibiting a strong mass loss that has resulted in a circumstellar gaseous disk formation. This study of a correlation between Hα and He I 6678 equivalent widths (EWs) and the He I 6678 line profiles has been performed at the observatory of the "Vereinigung der Sternfreunde Köln" (Germany) with a 0.4 m Schmidt-Cassegrain telescope C14. A classical slit-grating spectrograph with a spectral resolving power R ∼ 14000 and a CCD camera (768 × 512 pixel, pixel size 9 μm) provides spectra within the range from 6500 to 6700 Å. The spectra have been reduced manually with the programs VSpec and MK32. A high signal-to-noise ratio (as a rule > 400) is necessary to obtain simultaneous information about the strength and line profile behaviour of Hα and the double-peak emission line of He I 6678. Following the generally accepted assumption that the disk of this binary system is being fed in a kind of outbursts (Miroshnichenko et al., 2003), and since the He I 6678 line forms near the photosphere of the primary component one can expect a correlation between the equivalent width of the Hα and He I 6678 lines (Fig. 1). Such a correlation might be interpreted as a result of a disk feeding process. However on the other hand we can not exclude, that this reflects only contemporaneous density variations within the line formation zones.
Figure 1. Correlation between the equivalent width of He I 6678 and Hα from 04/2005 to 08/2009. Since April 2005, during every observing season, the observed correlation impressively supports the existence of this disk-feeding process, in which the slope of the linear fit shown in Fig. 1 reflects the quantitative correlation. In addition to the analysis of the EWs measured in the same spectra, the He I 6678 line double-peaked profile exhibits a variable V/R ratio. For the first time it was possible to monitor the entire cycle of the V/R variations in 2009 (Fig. 2). In the earlier seasons, merely the descent could be measured. On this occasion I would like to emphasize particularly that, amongst others, members of the French group, ARAS (http://www.astrosurf.com/aras), made a significant contribution to the frequent observations.
(a) 
(b) 
(c) Figure 2. Plots of the periodogram analysis with program AVE: (a) V/R-time behaviour (b) Periodogram of the data shown in Fig. 2a (c) Phase plot of the data shown in Fig. 2a The V/R measurements of the five cycles presented here permitted an analysis of its possible periodicity (Fig. 2). Thom Gandet, Thomas Rivinius and Ernst Pollmann independently calculated the following periods:
Gandet: 535 d ± 5 d
The ephemerides of the V/R variability are the following:
JD 2453420 (± 8–12) + 535 × E (Gandet)
The V/R-ratio has been measured only in the spectra for which both peaks are apparent. In the observations reported here, the triple peak structures were observed on the following dates:
2006: 04/02, 05/02, 05/11, 06/02, 06/23, 06/29, 07/02, 08/13, 08/18
An inspection of the V/R phase curve shows that the third emission component was observed only within the phase interval ∼0.67 to ∼1.1 (see Fig. 3). This might be due to the presence of a density enhancement in front of the star and hidden behind it at other phases.
Figure 3. Examples of the He I 6678 line profile labeled with observing dates and the V/R cycle phases. Reference Miroshnichenko, A. S., et al. 2003, A&A, 408, 305 |