A&A Supplement series, Vol 121, February 1997, 275-299
Received October 31, 1995; accepted June 3, 1996
D. de Winter - M.E. van den Ancker
Send offprint request: D. de Winter (email@example.com)
Astronomical Institute ``Anton Pannekoek'', University of Amsterdam, Kruislaan 403, 1098 SJ Amsterdam, The Netherlands
Dpto. Física Teórica, C-XI, Facultad de Ciencias, Universidad Autónoma de Madrid, Cantoblanco, E-28049 Madrid, Spain
The photometric behaviour of the peculiar B[e] star HD 45677 over the last 25 years is investigated. We conclude that the photometric variations (V = 7 22-8 85) on such a time scale can be well explained by obscurations, possibly due to large (> 1 m) circumstellar (CS) dust grains which were created after an explosive event around 1950. Intermediate time scale variations are also identified and can be well explained by infall of CS material.
Evidence is also found for smaller, pulse-like, amplitude variations. The time scale of this ``flickering'' ranges from days down to hours and is explained by instabilities in accretion flows towards HD 45677. The accretion mechanisms can be the origin of the hypothesized existence of a bipolar flow.
HD 45677, seen edge-on, shows evidence for the presence of a circumstellar disk. The significant accretion flows in this disk probably increased some time after the 1950 event, either due to a fall-back of part of the ejected material by a blow-out around 1950 or either due to the explosive dissociation of a large cometary-like body. Dynamic effects of such infalling and outflowing material close to the stellar surface are also detected by high resolution spectroscopy in the H and HeI profiles. These short time scale variations are also seen in the cool material as detected by the variations in the violet part of the NaID profiles, which are probably due to collisions of the outflowing material with the outer disk.
The presence of a disk is often indicated and here by the emission of the sodium lines with a strong absorption component at the systemic velocity (about 20 km s). Most detected lines, except H, are well centered at this velocity. Apart from the red [SII] lines, all other nebular lines as well as many other emission lines, including the FeII emission spectrum, were seen in spectra taken far before and after the 1950 event as well as in the latest spectra. So, the situation of the gaseous stellar environment seems to be stable in spite of the large photometric variations. The FeII emission lines are double peaked with a velocity separation of about 32 km s. This could mean that the inner disk material, in which these lines are thought to be formed, rotates with about 16 km s. In our 1992 data this rotation velocity seems to range up to 30 km s, which could be due to material accelerated by the 1950 event.
We discuss in this paper the evolutionary status of HD 45677. Because of its rather isolated position in the sky and because of the very fast evolution of a B2 type star, we think that HD 45677 could be young, but not in the sense of being a pre-main sequence object. Options like an evolved object such as a LBV or PN are not suitable because of the probable luminosity class III, IV or V and its rather cool central source, respectively. HD 45677 shows no evidence of any companion. The slight possibility of it being symbiotic can be added as well as instabilities in its unknown post-main sequence phase and the option that HD 45677 could be a hot post-AGB star.
keywords: circumstellar matter -- stars: emission-line -- stars: HD 45677 -- stars: variables -- infrared: stars