Abstract. We present the first results of a CO(2-1), (1-0), and 86 GHz SiO maser survey of AGB stars, selected by their weak near-infrared excess. Among the 65 sources of the present sample we find 10 objects with low CO outflow velocities, v exp < ∼ 5 km s −1 . Typically, these sources show (much) wider SiO maser features. Additionally, we get 5 sources with composite CO line profiles, i.e. a narrow feature is superimposed on a broader one, where both components are centered at the same stellar velocity. The gas mass-loss rates, outflow velocities and velocity structures suggested by these line profiles are compared with the results of hydrodynamical model calculations for dust forming molecular winds of pulsating AGB stars. The observations presented here give support to our recent low outflow-velocity models, in which only small amounts of dust are formed. Therefore, the wind generation in these models is dominated by stellar pulsation. We interpret the composite line profiles in terms of successive winds with different characteristics. Our hydrodynamical models, which show that the wind properties may be extremely sensitive to the stellar parameters, support such a scenario.

Abstract.We present a self-consistent time-dependent model for the oxygen-rich Mira variable IRC -20197. This model includes a consistent treatment of the interactions among hydrodynamics, thermodynamics, radiative transfer, equilibrium chemistry, and heterogeneous dust formation with TiO 2 nuclei. The model is determined by the stellar parameters, stellar mass4 L , stellar temperature T = 2400 K, and solar abundances of the elements. The pulsation of the star is simulated by a piston at the inner boundary where the velocity varies sinusoidally with a period of P = 636 d and an amplitude of ∆v p = 8 km s −1 . Based on the atmospheric structure resulting from this hydrodynamic calculation at different phases, we have performed angle-and frequency-dependent continuum radiation transfer calculations, which result in the spectral energy distributions at different phases of the pulsation cycle and in synthetic light curves at different wavelengths. These are in good agreement with the infrared observations of IRC -20197. The model yields a time averaged outflow velocity of 11.9 km s −1 and an average mass loss rate of 7.3 × 10 −6 M yr −1 which are in good agreement with the values derived from radio observations. Furthermore, the chemical composition of the resulting grains is discussed.

Y CVn is a carbon star surrounded by a detached dust shell that has been imaged by the Infrared Space Observatory at 90 microns. With the Nancay Radio Telescope we have studied the gaseous counterpart in the 21-cm HI emission line. New data have been acquired and allow to improve the signal to noise ratio on this line. The high spectral resolution line profiles obtained at the position of the star and at several offset positions set strong constraints on the gas temperature and kinematics within the detached shell; the bulk of the material should be at ~ 100-200 K and in expansion at ~ 1-2 km/s. In addition, the line profile at the central position shows a quasi-rectangular pedestal that traces an 8 km/s outflow of ~ 1.0 10^-7 Msol/yr, stable for about 2 10^4 years, which corresponds to the central outflow already studied with CO rotational lines. We present a model in which the detached shell results from the slowing-down of the stellar wind by surrounding matter. The inner radius corresponds to the location where the stellar outflow is abruptly slowed down from ~ 8 km/s to 2 km/s (termination shock). The outer radius corresponds to the location where external matter is compressed by the expanding shell (bow shock). In this model the mass loss rate of Y CVn has been set constant, at the same level of 1.0 10^-7 Msol/yr, for ~ 4.5 10^5 years. The gas temperature varies from ~ 1800 K at the inner limit to 165 K at the interface between circumstellar matter and external matter. Our modelling shows that the presence of a detached shell around an AGB star may not mean that a drastic reduction of the mass loss rate has occurred in the past. The inner radius of such a shell might only be the effect of a termination shock rather than of an interruption of the mass loss process.Comment: Accepted for publication in Monthly Notices of the Royal Astronomical Societ