Super asymptotic giant branch stars. I – Evolution code comparison

• Published in: Monthly notices of the Royal Astronomical Society Vol. 401; no. 3; pp. 1453 - 1464
• Main Authors: Doherty, C. L., Siess, L., Lattanzio, J. C., Gil-Pons, P.
• Format: Journal Article
• Language: English
• Published: Oxford, UK Blackwell Publishing Ltd 21-01-2010; Wiley-Blackwell; Oxford University Press
• Subjects: abundances; Astronomy; Astrophysics; Earth, ocean, space; Exact sciences and technology; Metals; methods: numerical; nuclear reactions; nuclear reactions, nucleosynthesis, abundances; nucleosynthesis; Stars & galaxies; stars: AGB and post-AGB; stars: evolution; stars: AGB and post-AGB; nuclear reactions, nucleosynthesis, abundances; methods: numerical; stars: evolution; Carbon burning; Nucleosynthesis; Abundance; Stellar evolution; Numerical method; Critical mass; Star models; Stellar structure; Algorithms; Stellar mass; Metallicity; Asymptotic giant branch
• ISSN: 0035-8711; 1365-2966
• DOI: 10.1111/j.1365-2966.2009.15772.x

We present an extensive set of detailed stellar models in the mass range 7.7–10.5 M⊙ over the metallicity range Z= 10−5–0.02. These models were produced using the Monash University version of the Mount Stromlo Stellar Structure Program (monstar) and follow the evolution from the pre-main sequence to the first thermal pulse of these super asymptotic giant branch stars. A quantitative comparison is made to the study of Siess. Prior to this study, only qualitative comparisons and code validations existed in this critical mass range, and the large variations in the literature were largely unexplained. The comparison presented here is particularly detailed due to the standardization of the input physics, where possible. The minimum initial mass of star which ignites carbon, Mup, was found to agree within 0.2 M⊙ between the codes over the entire metallicity range. We find exceptional agreement in the model results between these two codes for all stages of evolution up to and including carbon burning. For additional comparison, we also present results from the evolve code, a modified version of the iben code as described in Gil-Pons, Gutiérrez & García-Berro for some important variables during the carbon burning phase. Several numerical tests showed that the carbon burning phase is weakly dependent on the spatial resolution but that inadequate temporal resolution alters the behaviour of the convective zones. We also discovered that stars just below Mup may experience a carbon flash that is not followed by the development of the flame. Such aborted carbon burning models thus preserve a CO core surrounding by a 0.2–0.3 M⊙ shell of partially burnt carbon material. We present a simplified algorithm for calculating carbon burning that only relies on tracking two species, 12C and 16O, but which tests show works quite accurately for the a wide range of initial masses and compositions.