Low abundances of heavy elements in the solar outer layers: comparisons of solar models with helioseismic inversions

¹ Laboratoire d'Astrophysique de Toulouse et Tarbes – UMR 5572 – Université Paul Sabatier Toulouse III – CNRS, 14 Av. E. Belin, 31400 Toulouse, France e-mail: mcastro@ast.obs-mip.fr
² GRAAL – UMR 5024 – Université Montpellier II – CC072, Place E. Bataillon, 34095 Montpellier, France

Received: 31 August 2006
Accepted: 11 November 2006

Abstract

Context.Recent solar photospheric abundance analyses have led to a significant reduction of the metal abundances compared to the previous determinations. is 30% smaller, down to 0.0165. The solar models computed with standard opacities and diffusion processes using these new abundances give poor agreement with helioseismic inversions for the sound-speed profile, the surface helium abundance, and the convective zone depth.

Aims.We attempt to obtain a good agreement between helioseismic inversions and solar models that present the “old mixture” in the interior and new chemical composition in the convective zone. To reach this result, we assume an undermetallic accretion at the beginning of the main sequence.

Methods.We compute solar models with the Toulouse-Geneva Evolution Code, in which we simulate an undermetallic accretion in the early stages of the main sequence, in order to obtain the new mixture in the outer convective zone. We compare the sound-speed profile, the convective zone depth, and the surface helium abundance with those deduced from helioseismology.

Results.The model with accretion but without any mixing process inside is in better agreement with helioseismology than the solar model with the new abundances throughout. There is, however, a spike under the convective zone which reaches 3.4%. Furthermore, the convective zone depth and the surface helium abundance are too low. Introducing overshooting below the convective zone allows us to recover the good convective zone radius and the addition of rotation-induced mixing and tachocline allows us to reconcile the surface helium abundance. But in any case the agreement of the sound-speed profile with helioseismic inference is worse than obtained with the old abundances.