Resolving the ionized wind of the post-red supergiant IRC +10 420 with VLTI/AMBER^*

¹ School of Physics & Astronomy, University of Leeds, Woodhouse Lane, Leeds LS2 9JT, UK e-mail: w.j.m.dewit@leeds.ac.uk
² Institute for Astronomy, University of Leuven, Celestijnenlaan 200D, 3001 Leuven, Belgium
³ Laboratoire d'Astrophysique, Observatoire de Grenoble, BP 53, 38041 Grenoble Cedex 9, France 

Received: 1 October 2007
Accepted: 29 November 2007

Abstract

Aims.The paper investigates the milli-arcsecond scale structure of the present-day wind of the post-red supergiant IRC +10 420.

Methods.We use three telescopes of the VLT Interferometer in combination with the AMBER near-infrared beam combiner to measure spectrally dispersed correlated fluxes in the K-band around the Brγ transition. The resulting visibilities are compared with predictions of various simple models in order to infer the size of the observed emission region.

Results.The Brγ line is resolved by VLTI+AMBER on all three baselines, with the maximum projected baseline extending 69 m and a PA ranging between 10° and 30°. A differential phase between line and continuum is detected on the longest baseline. The best fitting model, a Gaussian flux distribution, implies a size of the Brγ emission region of 3.3 milli-arcsec (FWHM). The size of the continuum emission is not constrained by the observations. From a comparison of the AMBER equivalent width of Brγ with measurements from various epochs, we find that the stellar photosphere contributes about 60% of the total continuum light at 2.2 μm. The remaining 40% continuum emission originates from structures larger than the 66 mas AMBER field of view. This independently confirms similar results made by previous studies. If the Brγ emission is optically thin, then the observations do not allow to make any inferences about the shape of the line forming region. However, there is indirect evidence that the hydrogen recombination line emission is optically thick. In that case, using simple arguments, we find that the line emitting region is elongated. This is because the spectrum indicates that the projected line emitting area is about twice that of the stellar surface. A circular Brγ line emitting area, however, would have a size an order of magnitude larger using the size of the emission region measured along our baseline. We briefly mention the possibilities whether such a structure is due to a bi-polar flow or a circumstellar disk.