Modelling of the scandium abundance evolution in AmFm stars

TL;DR

This study models the surface abundance of scandium in AmFm stars using atomic diffusion, mixing, and mass loss processes to match observed under-abundances, providing insights into stellar surface composition evolution.

Contribution

It introduces detailed models of scandium surface abundance in AmFm stars incorporating new atomic data, mixing, and mass loss, advancing understanding of chemical peculiarities.

Findings

Models with full mixing match observed scandium abundances.

Mass loss rates of 10^{-13} to 10^{-14} solar masses per year are compatible with some observations.

Mixing prescriptions can reproduce observed abundances in the most massive models.

Abstract

Scandium is a key element of the Am star phenomenon since its surface under-abundance is one of the criteria that characterise such stars. Thanks to the availability of a sufficiently complete set of theoretical atomic data for this element, reliable radiative accelerations for Sc can now be computed, which allows its behaviour under the action of atomic diffusion to be modelled. We explore the required conditions, in terms of mixing processes or mass loss, for our models to reproduce the observed surface abundances of Sc in Am stars. The models are computed with the Toulouse-Geneva evolution code, which uses the parametric single-valued parameter method for the calculation of radiative accelerations. Fingering mixing is included, using a prescription that comes from 3D hydrodynamical simulations. Other parameter-dependent turbulent mixing processes are also considered. A global mass loss is also implemented. When no mass loss is considered, the observed abundances of Sc are rather in favour of the models whose superficial layers are fully mixed down to the iron accumulation zone, although other mixing prescriptions are also able to reproduce the observations for the most massive model presented here ($2.0 M_\odot$). The models including mass loss with rates in the range of $[10^{-13};10^{-14}] M_\odot$/yr are compatible with some of the observations, while other observations suggest that the mass-loss rate could be lower. The constraints brought by the modelling of Sc are consistent with those derived using other chemical elements.