Slowing down atomic diffusion in subdwarf B stars: mass loss or turbulence?

Abstract

Subdwarf B (sdB) stars show chemical peculiarities that cannot be explained by diffusion theory alone. Both mass loss and turbulence have been invoked to slow down atomic diffusion in order to match observed abundances. The fact that some sdB stars show pulsations give upper limits on the amount of mass loss and turbulent mixing allowed. Consequently, nonadiabatic asteroseismology has the potential to decide which process is responsible for the abundance anomalies. We compute for the first time seismic properties of sdB models with atomic diffusion included consistently during the stellar evolution. The diffusion equations with radiative forces are solved for H, He, C, N, O, Ne, Mg, Fe and Ni. We examine the effects of various mass-loss rates and mixed surface masses on the abundances and mode stability. It is shown that the mass-loss rates needed to simulate the observed He abundances (10 −14 Ṁ /M yr −1 10 −13 ) are not consistent with observed pulsations. We find that for pulsations to be driven the rates should beṀ 10 −15 M yr −1 . On the other hand, weak turbulent mixing of the outer 10 −6 M can explain the He abundance anomalies while still allowing pulsations to be driven. The origin of the turbulence remains unknown but the presence of pulsations gives tight constraints on the underlying turbulence model.

The blue-edge problem of the V1093 Herculis instability strip revisited using evolutionary models with atomic diffusion

“…5). In Hu et al (2011) it was shown that a weak stellar wind or a mixing process in the envelope, effects which are not accounted for in the grid presented in Sect. 3.2, both tend to reduce the build up of Fe and Ni in the envelope.…”

“…The most important changes are the implementation of gravitational settling, thermal diffusion and concentration diffusion by Hu et al (2010), the implementation of radiative levitation by Hu et al (2011) and the coupling of the STARS code to the non-adiabatic pulsation code MAD (Dupret 2001) as described in Hu et al (2008). The evolutionary tracks are computed starting from red giant branch models from which most of the envelope was stripped off.…”

“…The authors showed that the resulting evolutionary models can excite low-degree g-modes at relatively high effective temperatures, suggesting that the blue-edge problem could be resolved. In this work, we present a grid of models computed using the codes of Hu et al (2011) to readdress the instability strip issue.…”

“…Van Grootel et al 2010;Fontaine et al 2012, and references therein;. Asteroseismology of sdBs is not only useful to constrain the evolutionary scenarios, but also allows us to test our knowledge of physical processes such as atomic diffusion and stellar winds (Charpinet et al 2009b;Hu et al 2011), as well as to test tidal theory through spin-orbit synchronizations of sdBs in short period binaries (Van Grootel et al 2008;Charpinet et al 2008;Pablo et al 2011Pablo et al , 2012. Recent reviews of sdB asteroseismology can be found in Fontaine et al (2006b), Østensen (2009, 2010), Charpinet et al (2009aCharpinet et al ( , 2013 and Kawaler (2010).…”

“…Recently, Hu et al (2011) added diffusion due to radiative levitation to their evolution code, next to the already implemented gravitational settling, thermal diffusion and concentration diffusion, and solved the diffusion equations for H, He, C, N, O, Ne, Mg, Fe and Ni. Because of the inclusion of radiative levitation, Fe and Ni enhancements are built up in the pulsation driving region, therefore eliminating the need to include artificial abundance enhancements.…”