New models for the evolution of post-asymptotic giant branch stars and central stars of planetary nebulae

Abstract

Context. The post-asymptotic giant branch (AGB) phase is arguably one of the least understood phases of the evolution of low-and intermediate-mass stars. The two grids of models presently available are based on outdated micro-and macrophysics and do not agree with each other. Studies of the central stars of planetary nebulae (CSPNe) and post-AGB stars in different stellar populations point to significant discrepancies with the theoretical predictions of post-AGB models. Aims. We study the timescales of post-AGB and CSPNe in the context of our present understanding of the micro-and macrophysics of stars. We want to assess whether new post-AGB models, based on the latter improvements in TP-AGB modeling, can help us to understand the discrepancies between observation and theory and within theory itself. In addition, we aim to understand the impact of the previous AGB evolution for post-AGB phases. Methods. We computed a grid of post-AGB full evolutionary sequences that include all previous evolutionary stages from the zero age main sequence to the white dwarf phase. We computed models for initial masses between 0.8 and 4 M and for a wide range of initial metallicities (Z 0 = 0.02, 0.01, 0.001, 0.0001). This allowed us to provide post-AGB timescales and properties for H-burning post-AGB objects with masses in the relevant range for the formation of planetary nebulae (∼0.5−0.8 M ). We included an updated treatment of the constitutive microphysics and included an updated description of the mixing processes and winds that play a key role during the thermal pulses (TP) on the AGB phase.Results. We present a new grid of models for post-AGB stars that take into account the improvements in the modeling of AGB stars in recent decades. These new models are particularly suited to be inputs in studies of the formation of planetary nebulae and for the determination of the properties of CSPNe from their observational parameters. We find post-AGB timescales that are at least approximately three to ten times shorter than those of old post-AGB stellar evolution models. This is true for the whole mass and metallicity range. The new models are also ∼0.1−0.3 dex brighter than the previous models with similar remnant masses. Post-AGB timescales only show a mild dependence on metallicity. Conclusions. The shorter post-AGB timescales derived in the present work are in agreement with recent semiempirical determinations of the post-AGB timescales from the CSPNe in the Galactic bulge. The lower number of post-AGB and CSPNe predicted by the new models might help to alleviate some of the discrepancies found in the literature. As a result of the very different post-AGB crossing times, initial final mass relation and luminosities of the present models, the new models will have a significant impact on the predictions for the formation of planetary nebulae and the planetary nebulae luminosity function. In particular, the new models should help to understand the formation of low-mass CSPNe as inferred from asteroseismic and spectroscopic d...

On the production of He, C, and N by low- and intermediate-mass stars: a comparison of observed and model-predicted planetary nebula abundances

“…In this circumstance TDU leads to the mixing of M TDU ≃ 0.003M ⊙ from the H-free core into a H-rich envelope of M H env ≃ 0.027M ⊙ , significantly increasing the surface carbon abundance of the star. This example shows why it is necessary to keep in mind that final AGB thermal pulses coupled with low envelope masses can significantly change the surface abundances from those predicted by AGB stellar evolution models 8 We note that oxygen may not be a reliable metallicity indicator if significant amounts of O are dredged up to the surface or destroyed by hot bottom burning during the TP-AGB as predicted by some models -see section 3.1.1 in Di Criscienzo et al (2016) and Table 3 in Miller Bertolami (2016).…”

“…Central star luminosities and effective temperatures in all but three cases were taken from Frew (2008), and central and progenitor star masses were inferred by plotting these values in L-T diagrams containing evolutionary tracks from Wood (1994) andMiller Bertolami (2016).…”

“…We now briefly review the treatment of these key ingredients in the four grids adopted here for the comparison: the MONASH grid (Karakas 2014;, the LPCODE grid (Miller Bertolami 2016), the ATON grid (Ventura et al 2015;Di Criscienzo et al 2016) and the FRUITY database (Cristallo et al 2011(Cristallo et al , 2015. While all the models discussed here include an upto-date treatment of the microphysics, and all of them neglect the impact of rotation, the theoretical models discussed in this section have some key differences in the modeling of winds and convective boundary mixing processes.…”

“…In fact, trying to fit all available observational constraints by means of a simple overshooting prescription might not be even possible [see Weiss & Ferguson (2009); Karakas & Lattanzio (2014);Miller Bertolami (2016)]. This fact, together with the lack of compelling theoretical arguments and the lack of a common observational benchmark for AGB theoretical evolution models has led authors to the adoption of very different approaches.…”

“…Finally we note that convection in the ATON code is computed with the full spectrum of turbulence convection, which leads to stronger HBB (Ventura & D'Antona 2005) when compared with models that adopt the standard mixing length theory [Cristallo et al (2011), and Miller Bertolami (2016)]. …”