An incisive look at the symbiotic star SS Leporis

Milli-arcsecond imaging with PIONIER/VLTI^⋆,⋆⋆

N. Blind¹, H. M. J. Boffin², J.-P. Berger², J.-B. Le Bouquin¹, A. Mérand², B. Lazareff¹ and G. Zins¹

¹ UJF-Grenoble 1/CNRS-INSU, Institut de Planétologie et d’Astrophysique de Grenoble (IPAG) UMR 5274, Grenoble, France
e-mail: nicolas.blind@obs.ujf-grenoble.fr
² European Southern Observatory, Casilla 19001, Santiago 19, Chile
e-mail: hboffin@eso.org

Received: 7 September 2011
Accepted: 20 November 2011

Abstract

Context. Determining the mass transfer in a close binary system is of prime importance for understanding its evolution. SS Leporis, a symbiotic star showing the Algol paradox and presenting clear evidence of ongoing mass transfer, in which the donor has been thought to fill its Roche lobe, is a target particularly suited to this kind of study.

Aims. Since previous spectroscopic and interferometric observations have not been able to fully constrain the system morphology and characteristics, we go one step further to determine its orbital parameters, for which we need new interferometric observations directly probing the inner parts of the system with a much higher number of spatial frequencies.

Methods. We use data obtained at eight different epochs with the VLTI instruments AMBER and PIONIER in the H and K bands. We performed aperture synthesis imaging to obtain the first model-independent view of this system. We then modelled it as a binary (whose giant is spatially resolved) that is surrounded by a circumbinary disc.

Results. Combining these interferometric measurements with previous radial velocities, we fully constrain the orbit of the system.We then determine the mass of each star and significantly revise the mass ratio. The M giant also appears to be almost twice smaller than previously thought. Additionally, the low spectral resolution of the data allows the flux of both stars and of the dusty disc to be determined along the H and K bands, and thereby extracting their temperatures.

Conclusions. We find that the M giant actually does not stricto sensus fill its Roche lobe. The mass transfer is more likely to occur through the accretion of an important part of the giant wind. We finally rise the possibility for an enhanced mass loss from the giant, and we show that an accretion disc should have formed around the A star.