Context. Observations of asymptotic giant branch (AGB) stars with increasing spatial resolution reveal new layers of complexity of atmospheric processes on a variety of scales. Aims. To analyze the physical mechanisms that cause asymmetries and surface structures in observed images, we use detailed 3D dynamical simulations of AGB stars; these simulations self-consistently describe convection and pulsations. Methods. We used the CO5BOLD radiation-hydrodynamics code to produce an exploratory grid of global "star-in-a-box" models of the outer convective envelope and the inner atmosphere of AGB stars to study convection, pulsations, and shock waves and their dependence on stellar and numerical parameters. Results. The model dynamics are governed by the interaction of long-lasting giant convection cells, short-lived surface granules, and strong, radial, fundamental-mode pulsations. Radial pulsations and shorter wavelength, traveling, acoustic waves induce shocks on various scales in the atmosphere. Convection, waves, and shocks all contribute to the dynamical pressure and, thus, to an increase of the stellar radius and to a levitation of material into layers where dust can form. Consequently, the resulting relation of pulsation period and stellar radius is shifted toward larger radii compared to that of non-linear 1D models. The dependence of pulsation period on luminosity agrees well with observed relations. The interaction of the pulsation mode with the non-stationary convective flow causes occasional amplitude changes and phase shifts. The regularity of the pulsations decreases with decreasing gravity as the relative size of convection cells increases. The model stars do not have a well-defined surface. Instead, the light is emitted from a very extended inhomogeneous atmosphere with a complex dynamic pattern of high-contrast features. Conclusions. Our models self-consistently describe convection, convectively generated acoustic noise, fundamental-mode radial pulsations, and atmospheric shocks of various scales, which give rise to complex changing structures in the atmospheres of AGB stars.

Context. The stellar winds of asymptotic giant branch (AGB) stars are commonly attributed to radiation pressure on dust grains, formed in the wake of shock waves that arise in the stellar atmospheres. The mass loss due to these outflows is substantial, and modelling the dynamical properties of the winds is essential both for studies of individual stars and for understanding the evolution of stellar populations with low to intermediate mass. Aims. The purpose of this work is to present an extensive grid of dynamical atmosphere and wind models for M-type AGB stars, covering a wide range of relevant stellar parameters. Methods. We used the DARWIN code, which includes frequency-dependent radiation-hydrodynamics and a time-dependent description of dust condensation and evaporation, to simulate the dynamical atmosphere. The wind-driving mechanism is photon scattering on submicron-sized Mg2SiO4 grains. The grid consists of ~4000 models, with luminosities from L⋆ = 890 L⊙ to L⋆ = 40 000 L⊙ and effective temperatures from 2200 to 3400 K. For the first time different current stellar masses are explored with M-type DARWIN models, ranging from 0.75 M⊙ to 3 M⊙. The modelling results are radial atmospheric structures, dynamical properties such as mass-loss rates and wind velocities, and dust properties (e.g. grain sizes, dust-to-gas ratios, and degree of condensed Si). Results. We find that the mass-loss rates of the models correlate strongly with luminosity. They also correlate with the ratio L*∕M*: increasing L*∕M* by an order of magnitude increases the mass-loss rates by about three orders of magnitude, which may naturally create a superwind regime in evolution models. There is, however, no discernible trend of mass-loss rate with effective temperature, in contrast to what is found for C-type AGB stars. We also find that the mass-loss rates level off at luminosities higher than ~14 000 L⊙, and consequently at pulsation periods longer than ~800 days. The final grain radii range from 0.25 to 0.6 μm. The amount of condensed Si is typically between 10 and 40%, with gas-to-dust mass ratios between 500 and 4000.

Context. The inclusion of molecular physics is an important piece that tends to be missing from the puzzle when modeling the spectra of supernovae (SNe). Molecules have both a direct impact on the spectra, particularly in the infrared, and an indirect one as a result of their influence on certain physical conditions, such as temperature. Aims. In this paper, we aim to investigate molecular formation and non-local thermodynamic equilibrium (NLTE) cooling, with a particular focus on CO, the most commonly detected molecule in supernovae. We also aim to determine the dependency of supernova chemistry on physical parameters and the relative sensitivity to rate uncertainties. Methods. We implemented a chemical kinetic description of the destruction and formation of molecules into the SN spectral synthesis code SUMO. In addition, selected molecules were coupled into the full NLTE level population framework and, thus, we incorporated molecular NLTE cooling into the temperature equation. We produced a test model of the CO formation in SN 1987A between 150 and 600 days and investigated the sensitivity of the resulting molecular masses to the input parameters. Results. We find that there is a close inter-dependency between the thermal evolution and the amount of CO formed, mainly through an important temperature-sensitive CO destruction process with O+. After a few hundred days, CO completely dominates the cooling of the oxygen-carbon zone of the supernova which, therefore, contributes little optical emission. The uncertainty of the calculated CO mass scales approximately linearly with the typical uncertainty factor for individual rates. We demonstrate how molecular masses can potentially be used to constrain various physical parameters of the supernova.