Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 322, 554-564 (1997)

Previous Section Next Section Title Page Table of Contents

1. Introduction

HD 5980 = AB 5 (Azzopardi & Breysacher 1979 ) is an eclipsing binary (Hoffmann et al. 1978 ) with an orbital period of 19.266 days derived from its light-curve elements (Breysacher & Perrier 1980 , 1991 ). It lies at the eastern border of the giant H II region NGC 346 or N 66 (Henize 1956 ), the most active site of star formation in the Small Magellanic Cloud (Niemela et al. 1986 , Massey et al. 1989 ) containing a large number of massive O stars. The spectral classification of HD 5980 has changed according to authors or epochs of observation. The oldest mention of a spectral type goes back to Henize (1956 ) who noted a rough type of Oa (his star S 28). Later, Feast et al. (1960 ) including this star (R 14) in their list of the brightest stars of the Magellanic Clouds (V = 11.61), labelled it as Wp and noted spectral line variations. Smith (1968b ) classified HD 5980 as OB + WN, Walborn (1977 ) proposed OB? + WN3, Breysacher & Westerlund (1978 ) confirmed the existence of spectral variations and assigned a spectral type WN3p + OB. Subsequently, Breysacher et al. (1982 ) classified the system as WN4 + O7I:. Spectroscopic observations by Niemela (1988 ), conducted during 1981-1983, showed that the emission lines arising from N v ions moved peculiarly in antiphase with those originating from N iv, whereas the absorption lines did not seem to participate in the 19.3 days orbit. This result led her to the conclusion that the system contains a WN4.5 + WN3 pair. Massey et al. (1989 ), based on point spread function fitting, detected a faint component lying at [FORMULA] 1 [FORMULA] from the main star at a position angle of [FORMULA] [FORMULA]. The presence of a third component contributing to the total light of the system was also supported by the solution of the visual light curve of HD 5980 (Breysacher & Perrier 1991 ).

Recently, drastic variations in the spectrum of HD 5980 were reported (Koenigsberger et al. 1994 ). Between 1978 and 1990 the spectrum of HD 5980 became cooler and evolved from an early type WN3-4 star to a late-type WN6. Furthermore, according to Barbá & Niemela (1995 ), HD 5980 showed an extraordinary spectral change in 1993 November, when the star brightened by about 1 mag and a WN8 type spectrum appeared, but two months later, the Wolf-Rayet (hereafter W-R) spectrum turned again to WN 6. After this event, HD 5980 underwent an even more spectacular change evolving abruptly into a sort of Luminous Blue Variable (LBV) type (Barbá et al. 1995 , Koenigsberger et al. 1995 ). This was accompanied by a visual brightening of HD 5980 by [FORMULA] mag in August 1994 (Bateson et al. 1994 , Bateson & Jones 1994 ). High dispersion IUE spectra obtained shortly after the discovery of the eruption show characteristic lines of a B1.5Ia [FORMULA] type (Koenigsberger et al. 1996 ). These authors report also that about one year later HD 5980 returned to its WN6 spectrum. The latter finding is also confirmed from optical spectra by Barbá et al. (1996 ) and from infrared data by Eenens & Morris (1996 ) who notice a significant increase of the ionization of the wind of HD 5980 from their K -band spectra obtained in 1995 May and November.

All these facts underline the exceptional character of HD 5980 as to the evolutionary status of the W-R stars of the nitrogen sequence and the advent of LBV characteristics. We have observed this puzzling SMC object at ESO using high-resolution imaging and spectroscopy. Since HD 5980 is a binary and a third component has been suspected, it is of prime importance to know how many components are involved and which of the objects has undergone eruption. We used NTT+SUSI and the 3.6 m telescope equipped with an adaptive optics system to take sub-arcsecond images of HD 5980. On the other hand, to study the spectrum of the object, we used NTT+EMMI in echelle and slit spectroscopy modes. Our observing run of September 1994 was carried out nearly one month before the observations of Barbá et al. (1995 ). The present observations are unique in the sense that they cover a crucial period in the evolution of HD 5980. The data presented here will therefore be helpful for better understanding this peculiar object in particular and the evolution of W-R stars in general.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1997

Online publication: June 5, 1998