Astron. Astrophys. 347, 524-531 (1999)

1. Introduction

It is often claimed that stellar pulsations are one of the main causes of the onset and the variability of mass-loss in OB-stars. The most detailed effort to show that this is the case has recently been presented by Rivinius et al. (1998a, 1998b) for the Be star Cen. They suggest that the beat-periods of the short-term line-profile variability govern the occurrence of the line emission outbursts in this star. Detailed modelling of the observed line profiles in terms of non-radial-pulsation theory (or any other physical model) is undertaken at present (Rivinius et al., in preparation; see also Rivinius 1998 for the preliminary results). Our work originates from the lack of a general firm connection between photospheric pulsations and mass-loss in early-type stars.

A division between the hot O and very-early B stars on the one hand, and later B stars on the other hand, is necessary. The effect of radiatively driven winds to infer mass-loss decreases rapidly for stars later than B 0. The status of ongoing theoretical efforts to understand the wind structure of O stars is summarised by Owocki (1998). At present, he did not yet succeed in taking into account the two most commonly suggested perturbation mechanisms, namely magnetic fields and non-radial pulsations, in the calculations that simulate the wind structure. The models are currently restricted to a driving modulation induced by the occurrence of bright and dark spots in the stellar photosphere. Improvements towards the full inclusion of the two perturbation mechanisms mentioned will probably be achieved in the near future. From an observational point of view, the mass-loss of stars earlier than B 0 can best be studied from UV resonance lines. An extended observational study of O-star winds is presented by e.g. Fullerton et al. (1996) and Kaper et al. (1997). They give an overview of the observed line-profile variability among O-type stars. Photometric studies of O stars reveal the occurrence of microvariations with a large range of periods (see e.g. Van Genderen 1985). For a recent study of the photometric variability in O-star runaways and in Wolf-Rayet stars based on Hipparcos data, we refer to Marchenko et al. (1998). They find that about half of the selected stars are variable while many of them were considered to be constant before the Hipparcos mission. The question if the observed photometric and spectroscopic periods are caused by the same mechanism is still open.

Regarding the Be stars, it is not clear what mechanism causes the presence of the circumstellar disk. Non-radial pulsations have been suggested as a prime cause of the onset of the disk formation. The question whether or not Be stars exhibit pulsations remains, however, unanswered (see e.g. Baade & Balona 1994). As far as the maintenance of the circumstellar disk around rapidly rotating B stars is concerned, promising physical mechanisms, such as the Wind-Compressed Disk model (Bjorkman & Cassinelli 1993), have been challenged seriously as a theoretical explanation (Owocki et al. 1996).

As recently shown by Cohen et al. (1997), wind attenuation of X-rays decreases abruptly beyond B 0 and becomes negligable around spectral type B 2. These authors also tried to investigate if pulsating B stars have a different X-ray behaviour compared to normal B stars, but failed to make any conclusion about this. The same remark is true for the Be stars in their sample.

Optical spectroscopy, polarisation measurements, and IR excesses are often used to study the mass-loss in stars cooler than B 0. In this paper, we consider IR excesses and Hobservations of mainly B-type stars. In order to study the role of non-radial pulsations in the creation of a circumstellar environment, we have undertaken an infrared study of OB-type stars that are confirmed pulsators. Our intention is to see if these stars have a larger probability of having circumstellar material compared to non-pulsators. If so, then one would expect to find a high percentage of pulsating (rapidly rotating) stars with excess fluxes at infrared wavelengths.

For our purpose we have considered the infrared data gathered by IRAS of confirmed non-radial pulsators, and we have compared them to the infrared fluxes of constant stars (see Waters et al. 1987). We describe the selection of the sample of pulsating stars in Sect. 2. It turns out that none of the very few considered O-type pulsators show an appreciable excess flux in the infrared, while four pulsating B stars do exhibit an unexpected IR excess. In Sects. 3 and 4 we study the nature of the excess flux at infrared wavelengths for respectively the Cep stars and the slowly pulsating B stars. We discuss our findings in Sect. 5.

© European Southern Observatory (ESO) 1999

Online publication: June 30, 1999
helpdesk.link@springer.de