Calibrating convective-core overshooting with eclipsing binary systems
The case of low-mass main-sequence stars
1 INAF–Osservatorio Astronomico di Collurania, via Maggini, 64100 Teramo, Italy
2 INFN, Sezione di Pisa, Largo Pontecorvo 3, 56127 Pisa, Italy
3 Dipartimento di Fisica “Enrico Fermi”, Università di Pisa, Largo Pontecorvo 3, 56127 Pisa, Italy
Received: 17 September 2015
Accepted: 21 December 2015
Context. Double-lined eclipsing binaries have often been adopted in literature to calibrate the extension of the convective-core overshooting beyond the border defined by the Schwarzschild criterion.
Aims. In a robust statistical way, we quantify the magnitude of the uncertainty that affects the calibration of the overshooting efficiency parameter β that is owing to the uncertainty on the observational data. We also quantify the biases on the β determination that is caused by the lack of constraints on the initial helium content and on the efficiencies of the superadiabatic convection and microscopic diffusion.
Methods. We adopted a modified grid-based SCEPtER pipeline to recover the β parameter from synthetic stellar data. Our grid spans the mass range [1.1; 1.6] M⊙ and evolutionary stages from the zero-age main sequence (MS) to the central hydrogen depletion. The β estimates were obtained by generalising the maximum likelihood technique described in our previous works. As observational constraint, we adopted the effective temperatures, [Fe/H], masses, and radii of the two stars.
Results. By means of Monte Carlo simulations, adopting a reference scenario of mild overshooting β = 0.2 for the synthetic data, and taking typical observational errors into account, we found both large statistical uncertainties and biases on the estimated values of β. For the first 80% of the MS evolution, β is biased by about −0.04, with the 1σ error practically unconstrained in the whole explored range [0.0; 0.4]. In the last 5% of the evolution the bias vanishes and the 1σ error is about 0.05. The 1σ errors are similar when adopting different reference values of β. Interestingly, for synthetic data computed without convective-core overshooting, the estimated β is biased by about 0.12 in the first 80% of the MS evolution, and by 0.05 afterwards. Assuming an uncertainty of ±1 in the helium-to-metal enrichment ratio ΔY/ ΔZ, we found a large systematic uncertainty in the recovered β value, reaching 0.2 at the 60% of the MS evolution. Taking into account both the helium abundance indetermination and 1σ statistical uncertainty, we found that in the terminal part of the MS evolution the error on the estimated β values ranges from −0.05 to + 0.10, while β is basically unconstrained throughout the explored range at earlier evolutionary stages. We quantified the impact of a uniform variation of ±0.24 in the mixing-length parameter αml around the solar-calibrated value. The largest bias occurs in the last 5% of the evolution with an error on the estimated median β from −0.03 to + 0.07. In this last part, the 1σ uncertainty that addresses statistical and systematic error sources ranges from −0.09 to + 0.15. Finally, we quantified the impact of a complete neglect of diffusion in the stellar evolution computations. In this case, the 1σ uncertainty that addresses statistical and systematic error sources ranges from −0.08 to + 0.08 in the terminal 5% of the MS, while β is practically unconstrained in the first 80% of the MS.
Conclusions. The calibration of the convective core overshooting with double-lined eclipsing binaries – in the explored mass range and with both components still in their MS phase – appears to be poorly reliable, at least until further stellar observables, such as asteroseismic ones, and more accurate models are available.
Key words: binaries: eclipsing / methods: statistical / stars: evolution / stars: low-mass
© ESO, 2016