Exploring wind-driving dust species in cool luminous giants I. Basic criteria and dynamical models of M-type AGB stars

Context. The heavy mass loss observed in evolved asymptotic giant branch stars is usually attributed to a two-stage process: atmospheric levitation by pulsation-induced shock waves followed by radiative acceleration of dust grains, which transfer momentum to the surrounding gas through collisions. In order for an outflow to occur the two stages of the mass-loss scheme have to connect, i.e., the radiative acceleration can only be initiated if the levitated gas reaches a distance from the stellar photosphere where dust particles can condense. This levitation distance is limited by the kinetic energy transferred to the gas by the shock waves, which imposes strict constraints on potential wind-driving dust species. Aims. This work is part of an ongoing effort aiming at identifying the actual wind-drivers among the dust species observed in circumstellar envelopes. In particular, we focus on the interplay between a strong stellar radiation field and the dust formation process. Methods. To identify critical properties of potential wind-driving dust species we use detailed radiation-hydrodynamical models which include a parameterized dust description, complemented by simple analytical estimates to help with the physical interpretation of the numerical results. The adopted dust description is constructed to mimic different chemical and optical dust properties in order to systematically study the effects of a realistic radiation field on the second stage of the mass loss mechanism. Results. We see distinct trends in which combinations of optical and chemical dust properties are needed to trigger an outflow. Dust species with a low condensation temperature and a near-infrared absorption coefficient that decreases strongly with wavelength will not condense close enough to the stellar surface to be considered as potential wind-drivers. Conclusions. Our models confirm that metallic iron and Fe-bearing silicates are not viable as wind-drivers due to their near-infrared optical properties and resulting large condensation distances. TiO2 is also excluded as a wind-driver due to the low abundance of Ti. Other species, such a SiO2 and Al2O3, are less clear-cut cases due to uncertainties in the optical and chemical data and further work is needed. A strong candidate is Mg2SiO4 with grain sizes of 0.1-1 mu m, where scattering contributes significantly to the radiative acceleration, as suggested by earlier theoretical work and supported by recent observations.