Circumstellar dust shells around long-period variables: VIII. CO infrared line profiles from dynamical models for C-stars

J.M. Winters, J.J. Keady, A. Gauger, and P.V. Sada

Astronomy & Astrophysics, 359, 651-662 (2000)

Gzipped PostScript version (593 kB)

Abstract:

We present simplified non-LTE radiative transfer calculations for CO (v 0-1) fundamental and (v 0-2) overtone lines, using the structure of dynamical models for dust shells around carbon-rich long-period variables, which result from a consistent treatment of hydro- and thermodynamics, radiative transfer, chemistry and carbon grain formation.

Owing to the saturation of the fundamental band absorption cores the most useful information is provided by the P-Cygni emission component, which can be used to constrain the dust optical depth in the 5 µm region. The first overtone lines clearly reflect the shocked, layered structure of the inner shell region by showing a multi-component structure of the profiles. At least the lines of higher rotational excitation (Jl > 10) are not saturated even for mass loss rates of the order of 10-5Msunyr-1 and thus can be used to constrain the mass loss rate. Furthermore, the dynamics of the dust formation process is reflected in the temporal evolution of the synthetic CO first overtone line profiles resulting from the models. The formation of a new dust layer in the inner shell region leads to a secondary (low velocity) absorption component which evolves on the time scale of a dust
formation event, usually longer than the pulsation period of the star.

The synthetic overtone line profiles calculated on the basis of one dust shell model resemble the time variations of corresponding lines in the observed high resolution CO line spectra of the carbon-Mira IRC+10216, which thereby are interpreted as result of a dust formation event taking place on a time scale of approx. 10yr corresponding to 6 pulsation periods of the star. However, the calculated fundamental line absorption cores are broader than observed and the first overtone line strengths are too high, indicating that the density and thus probably the mass loss rate of the model is too high by a factor of about 3 and should be closer to the value of 1 10-5 Msunyr-1 given for IRC+10216 in the literature.

Keywords:

Hydrodynamics -- Line: profiles -- Stars: circumstellar matter -- Stars: individual: IRC+10216 -- Stars: AGB and post-AGB -- Stars: winds, outflows