Self-consistent hydrodynamical models for stellar winds

The physical and chemical conditions in the atmosphere of pulsating AGB stars are not well understood. In order to properly model this region, which is packed with shocks arisen from the pulsational behaviour of the star, we aim to understand the interplay between spatial and temporal changes in both the chemical composition and the hydro/thermodynamical behaviour inside these regions. Ideal models require the coupling of hydrodynamics, chemistry and radiative transfer, in three dimensions. As this is computationally not yet feasible, we aim to model this zone via a bottom-up approach. At first, we build correct 3D hydrodynamical set-up without any cooling or heating. Omitting cooling hampers the mass-loss of the AGB star within the reasonable confines of a realistic parameter space. Introducing cooling will decrease the temperature gradients in the atmosphere, counteracting the mass-loss even more. However, cooling also ensures the existence of regions where the temperature is low enough for the formation of dust to take place. This dust will absorb the momentum of the impacting photons from the AGB photosphere, accelerate outward and collide with the obstructing gas, dragging it along. Moreover, since chemistry, nucleation and dust formation depend critically on the temperature structure of the circumstellar environment, it is of utmost importance to include all relevant heating/cooling sources. Efforts to include cooling have been undertaken in the last decades, making use of different radiative cooling mechanisms for several chemical species, with some simplified radiative transfer. However, often the chemical composition of these 1D atmosphere models is fixed, implying the very strong assumption of chemical equilibrium, which is not at all true for a pulsating AGB atmosphere. We wish to model these atmospheres making as few assumptions as possible on equilibrium conditions. Therefore, as a first step, we introduce H2 dissociative cooling to the hydrodynamical model, arguing this is the dominant cooling factor. Using dissociative H2 cooling allows the ratio of the H-H2 gas mixture to vary, making the cooling efficiency time and space dependent. This will affect local cooling, in turn affecting the hydrodynamics and chemical composition, hereby introducing a feedback loop. Secondly, most significant radiative heating/cooling sources will be introduced to obtain the most realistic temperature structure. Next, dust acceleration will be introduced in the regions cool enough for dust condensation to exists. Hereby laying the basis of our hydrodynamical chemistry model for stellar winds of evolved stars.