Evolution, nucleosynthesis and yields of low mass AGB stars

The envelope of thermally pulsing AGB stars undergoing periodic third dredge-up episodes is enriched in both light and heavy elements, the ashes of a complex internal nucleosynthesis involving p, alpha and n captures over hundreds of stable and unstable isotopes. In this paper, new models of low-mass AGB stars (2 Msun), with metallicity ranging between Z=0.0138 (the solar one) and Z=0.0001, are presented. Main features are: i) a full nuclear network (from H to Bi) coupled to the stellar evolution code, ii) a mass loss-period-luminosity relation, based on available data for long period variables, and ii) molecular and atomic opacities for C- and/or N-enhanced mixtures, appropriate for the chemical modifications of the envelope caused by the third dredge up. For each model a detailed description of the physical and chemical evolution is presented; moreover, we present a uniform set of yields, comprehensive of all chemical species (from hydrogen to bismuth). The main nucleosynthesis site is the thin 13C pocket, which forms in the core-envelope transition region after each third dredge up episode. The formation of this 13C pockets is the principal by-product of the introduction of a new algorithm, which shapes the velocity profile of convective elements at the inner border of the convective envelope: both the physical grounds and the calibration of the algorithm are discussed in detail. The final surface compositions of the various models reflect the differences in the initial iron-seed content and in the physical structure of AGB stars belonging to different stellar populations. The agreement with the observed [hs/ls] index observed in intrinsic C stars at different [Fe/H] is generally good.

💡 Research Summary

This paper presents a comprehensive set of evolutionary models for low‑mass (2 M☉) asymptotic giant branch (AGB) stars covering a wide metallicity range from solar (Z = 0.0138) down to extremely metal‑poor (Z = 0.0001). The authors couple a full nuclear reaction network—extending from hydrogen to bismuth and encompassing several hundred stable and unstable isotopes—directly to a stellar evolution code. This integration allows the nucleosynthesis to be computed self‑consistently with the changing stellar structure at each time step.

A key methodological innovation is the implementation of a mass‑loss prescription that links the stellar pulsation period, luminosity, and mass‑loss rate. The relation is calibrated against observational data for long‑period variables, ensuring that the models reproduce realistic AGB lifetimes and wind properties across the metallicity spectrum. In addition, the authors compute molecular and atomic opacities for carbon‑ and nitrogen‑enhanced mixtures, which are essential for correctly modeling the envelope structure once third dredge‑up (TDU) events enrich the surface in C and N.

The central nucleosynthetic site in all models is the thin ¹³C pocket that forms in the transition region between the H‑burning shell and the convective envelope after each TDU episode. The formation of this pocket is driven by a newly introduced algorithm that shapes the convective velocity profile at the inner boundary of the envelope. Rather than imposing an abrupt drop to zero velocity, the algorithm imposes a gradual exponential decline, allowing a small amount of protons to diffuse into the He‑rich intershell and produce a localized ¹³C‑rich layer. This layer then serves as the primary neutron source through the ¹³C(α,n)¹⁶O reaction during the interpulse phase. The mass and ¹³C abundance of the pocket naturally adjust with metallicity, providing a physically motivated alternative to the often‑used parametric prescriptions.

For each metallicity the paper follows the full physical and chemical evolution through the thermally pulsing AGB phase, documenting the number of thermal pulses, the depth and efficiency of each TDU, the growth of the core, and the surface abundances after every dredge‑up. The authors then compute detailed yields for every isotope from H to Bi, presenting a uniform yield set that can be directly employed in galactic chemical evolution models.

The results show a clear metallicity dependence of the s‑process pattern. At low metallicity the reduced iron seed abundance leads to a higher neutron‑to‑seed ratio, favoring the production of heavy s‑process elements (hs) such as Ba, La, Ce, and Nd, and even lead (Pb) and bismuth (Bi). Consequently, the