Abundance Analysis of Chemically Depleted Post-AGB/Post-RGB Binaries with Faint Discs (2510.16691v1)
Abstract: Post-AGB and post-RGB binaries with stable circumbinary discs provide key insights into late stellar and disc evolution, revealing how binary interactions shape disc structure and stellar surface composition. A defining trait of such systems is the observed underabundance of refractory elements in the stellar photosphere relative to volatile elements -- photospheric chemical depletion -- resulting from the star accreting volatile-rich circumstellar gas. In this study, we investigated the link between photospheric depletion and disc evolution by focusing on post-AGB/post-RGB binaries with low infrared excess (hereafter `faint disc'' targets). We analysed high-resolution optical spectra from HERMES/Mercator and UVES/VLT for 6 Galactic and 2 LMC targets. Using \texttt{E-iSpec}, we homogeneously derived atmospheric parameters and chemical abundances of 29 elements from carbon to europium, and included NLTE corrections for 15 elements from carbon to barium that we calculated using pySME and pre-computed grids of departure coefficients. All targets exhibitsaturated' depletion patterns, which we characterised using two-piece linear fits defined by three parameters: initial metallicity ([M/H]$0$), turn-off temperature ($T{\rm turn-off}$), and depletion scale ($\nabla_{\rm 100 K}$). Among several findings, we highlight the bimodal distribution of $T_{\rm turn-off}$ in faint disc targets, which allows classification into two subgroups analogous to full discs with continuous, optically thick dust ($T_{\rm turn-off}$ > 1 100 K), and transition discs with inner clearing ($T_{\rm turn-off}$ < 1 100 K). Our results imply that faint disc targets likely represent the final stages of disc dissipation, highlighting the diversity of depletion profiles, the complexity of disc-binary interactions, and the need to understand the rarity and evolution of faint disc systems.
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
Top Community Prompts
Collections
Sign up for free to add this paper to one or more collections.