Skip to main content

A detailed study of the chemistry, physics, and structure in AGB stars

Final Report Summary - STUDYSTARS (A detailed study of the chemistry, physics, and structure in AGB stars)

The goal of the project was to constrain the chemical and physical properties of evolved stars on the asymptotic giant branch (AGB), and their circumstellar envelopes (CSEs), and on the structure and dynamics of the mass loss that creates the CSE. AGB stars contribute a significant fraction of new elements to the medium between the stars, and are important for star formation and the chemical evolution of galaxies.

In this project I described the chemical complexity of AGB CSEs through observations covering a broad frequency range, and identifying the typical chemical reaction paths in the CSE. I further described the creation and evolution of detached shells and for the first time observationally constrained the thermal-pulse cycle directly.

The majority of all stars in the Milky Way will evolve along the AGB. During its AGB evolution, a star periodically undergoes rapid helium burning in a shell around the core. This phenomenon is known as a thermal pulse (TP), and lasts for only a few hundred years every 10000 - 100000 years. The release of extra energy causes the star to restructure, leading to the nucleosynthesis of heavy elements inside the star (mainly carbon and s-process elements). The new elements are mixed to the stellar surface, incorporated into the stellar wind, and released into the interstellar medium (ISM), leading to the chemical evolution of galaxies. Depending on the stellar mass, an AGB star will undergo ~15–25 TPs on the AGB. A consequence of a TP is the formation of a detached shell of dust and gas around the central star.

The amount of new elements created depends critically on the physical parameters of the star during subsequent TPs. Owing to their short duration and the long timescales between pulses, it is extremely unlikely to observe an AGB star during a TP directly. As a consequence stellar synthesis models of nucleosynthesis and TPs have been essentially independent of observations. In particular the evolution of the mass-loss rate during and after a pulse is unknown. Constraining this is essential to determine the lifetime on the AGB, and hence the number of pulses a star experiences.

The mass loss from the surface of an AGB star builds up a large CSE. Nearly 80 molecules have to date been detected in AGB CSEs, a large fraction of these exclusively in these environments. In broad terms, one expects the carbon to oxygen (C/O) ratio in the stellar atmosphere to determine the domi- nating chemistry. The quick formation and high stability of the CO molecule leave ample amounts of O in the C/O<1 (M-type, or O-rich) case, and C in the C/O>1 (C-rich) case. These can take part in chemical processes that produce O- and C-bearing molecules, respectively. A suspected intermediate case are the S-type AGB stars, with C/O≈ 1.

However, the chemical composition of the CSE is continuously affected by, e.g. shockwaves caused by stellar pulsations, dust-gas reactions, and penetration of interstellar UV photons, leading to a non-equilibrium chemistry. It is currently not clear which of these processes dominate.

Therefore, the chemical formation networks, and hence also the yields from AGB stars, are still uncertain. It is important to know which molecules and dust particles are formed, and how these affect the physical structure of the CSE. This is significant for astrophysical contexts also beyond AGB stars, e.g. grain-surface reactions, the survival of complex molecules on dust grain surfaces, and the formation of organic molecules.

Circumstellar chemistry:
At the beginning of the funding period I led a team of scientists to work with pilot observations to observe the molecular emission from the CSEs from three AGB stars with the IRAM 30m telescope. The receivers of the 30m telescope cover a large frequency range, and hence allow to detect a large number of transitions from several molecules. Unbiased surveys like this give a comprehensive inventory of the molecular content in the CSEs around AGB stars. Radiative transfer modelling of the detected transitions allows us to determine molecular abundances, in turn constraining chemical models of the CSEs. The pilot observations included one AGB star of each chemical type. While we indeed detected a large number of lines in the carbon AGB star, the number of detections is less in the M-type star, and even less in the S-type star. However, technical difficulties has delayed the further investigation of this data. Instead the team is part of an ALMA project to observe spectral scans towards a sample of carbon AGB stars. ALMA is by far the most powerful telescope in the world in this frequency range, both in terms of sensitivity and resolution. Analysis of the data is ongoing, and in the end will contain a catalogue of spatially resolved molecular emission lines. In particular the spatial information, together with modelling of multiple emission lines, will effectively constrain the chemical models. I am further involved in a project to observe the entire submillimeter range towards the nearby M-type AGB star R Doradus. The work is currently in preparation and will be submitted to the scientific journal Astronomy & Astrophysics by the end of the year. In addition to the spectral scans I determined the H2O abundances in a sample of M-type AGB stars. The abundances were determined through radiative transfer modelling of H2O emission lines observed with Herschel/HIFI. I constrain the amount of H2O in the CSEs, the velocity structure, and temperature profiles. These are the most detailed observations of H2O towards AGB stars. Additionally, I demonstrate the difficulty of fully including molecular line cooling through H2O, and the need for updated radiative transfer codes. The results have important implications for the circumstellar chemistry and physical structure. Finally, we defined a PhD student project to investigate the chemical networks in AGB CSEs using theoretical models. In particular the student will study the role of internal UV-radiation fields on the chemical structure of the CSEs.

The projects show the chemical complexity of the envelopes around AGB stars, signifying their importance in the cosmic cycle of matter. In particular the spatial distribution of molecules is important for understanding the chemistry in these sources. In order to derive a complete understanding of the yields from AGB stars, it is important to include all effects (e.g. internal UV radiation fields) on the chemistry on the theoretical models.

Thermal pulses:
Just before the beginning of the funding period I published a paper in Nature presenting ALMA observations of the CO emission towards the carbon AGB stars R Sculptoris. The observations show the detached shell around the star that was created in a thermal pulse about 2000 years ago, as well as a spiral structure shaped by previously undetected binary companion. The shell combined with the spiral structure allowed us to measure the evolution of the mass-loss rate and expansion velocity during and after a thermal pulse observationally for the very first time.

The CO observations measure the position of the molecular gas around the star. However, in order to understand the formation and evolution of the detached shells fully, it is important to know how the dust around the star is distributed in relation to the circumstellar gas. The dust drives the mass loss of the star, and the evolution of the dust and gas directly relates to the evolution of the mass loss. In addition, the dust properties constrain the contribution of dust to the interstellar medium from evolved stars. The dust is effectively probed through observations of dust scattered, stellar light in the optical. Light that is scattered off dust grains is polarised, and observations of polarised light directly constrain the distribution of the dust. I published optical polarisation observations with our own instrument PolCor of the circumstellar structure around the carbon stars R Sculptoris and V644 Sco. The observations clearly show the detached shells of dust around the stars. In the case of R Sculptoris I show that the dust coincides exactly with the molecular gas, indicating that the dust and gas must have evolved together (see the attached figure). For V644 Sco this is the first time that the structure of the shell has been observed directly.

Additionally, I published the full set of CO observations with ALMA towards R Sculptoris. The spatial resolution of ALMA allows to analyse the shell and the inner CSE separately, hence allowing to constrain the evolution during the thermal pulse (i.e. when the shell is created), and the post-pulse evolution. I find that much more mass is lost during the thermal pulse cycle than predicted in theoretical models. This may significantly reduce the number of thermal pulses a star experience, and hence the yields of new elements the star produces.