SpringerLink
Forum Springer Astron. Astrophys.
Forum Whats New Search Orders


Astron. Astrophys. 335, 959-968 (1998)

Previous Section Next Section Title Page Table of Contents

2. Lithium and rotation in the Hyades F-stars

Lithium (a fragile element which burns at relatively low temperature in stellar interiors) has traditionally been used as a very powerful tracer of particle transport processes. Many Li abundance determinations are available for stars of different spectral types in galactic clusters of various ages, together with observed rotational velocities in some cases (see for ex. Soderblom 1993, Balachandran 1995 and references therein).

Relying on [Li/Ca] observations in Hyades stars, Wallerstein et al. (1965) detected a drop-off in the lithium content of main sequence stars with a spectral index around (B-V)=0.4. It was clearly confirmed much later by Boesgaard & Tripicco (1986) that Li is indeed depleted in Hyades F-stars in a range of 300 K in effective temperature centered around 6600 K (cf. Fig. 1 top). On the blue side of the so-called "Boesgaard-dip", Li abundances drop sharply, while the rise on the red side is more gradual. Evidence of the same feature has been seen in all galactic clusters older than 108yr as well as in field stars (see Michaud & Charbonneau 1991 and Balachandran 1995 for a complete list of references).

[FIGURE] Fig. 1. (top ) Lithium versus effective temperature in F Hyades stars (Boesgaard & Tripicco 1986, Thorburn et al. 1993). Triangles denote upper limits. (bottom ) Projected rotational velocity versus effective temperature in F Hyades stars. Observational data are from Kraft (1965), Stauffer et al. (1987), and Mermilliod (1992).

The simplest explanation for this characteristic feature was proposed by Michaud (1986) who showed how chemical separation could shape the gap in the Hyades F stars. This model relied on well-known physics with two adjusted parameters: the mass loss rate needed to reduce the predicted over-abundances due to radiative acceleration on the hot side of the plateau and the ratio of the mixing length to the pressure scale height. Three observational facts however contradict the pure microscopic diffusion hypothesis. Firstly, the predicted width of the Li dip at the age of the Hyades is narrower than observed (Richer & Michaud 1993). Secondly, the carbon, oxygen and boron under-abundances expected in the case of pure diffusion (Michaud 1986, Turcotte et al. 1997) failed to be found in the Hyades F stars (Boesgaard 1989, Friel & Boesgaard 1990, García López et al. 1993) and in Li and Be deficient F field stars (Boesgaard et al. 1997). This indicates that a macroscopic process counteracts the effects of element segregation in these stars. Finally, in the pure diffusion model, Li settles and remains in a buffer zone below the convective envelope; it should then been dredged to the surface as soon as those stars leave the main sequence. Observations of lithium in M67's slightly evolved stars (Pilachowski et al. 1988; Balachandran 1995; Deliyannis et al. 1997) show however that the lithium depletion in stars formerly from the dip persists on the sub-giant branch. This strongly favors explanations relying on nuclear destruction of lithium.

Schramm et al. (1990) proposed an explanation relying on mass loss. The peeling of the outer layers of the stars could bring to the surface the regions where lithium has been depleted by pure nuclear destruction. However, the existence of Hyades and field F stars which still have some lithium but where some beryllium has been also depleted argues against that mechanism (Stephens et al. 1997).

Following Press' suggestion (1981) that gravity waves may induce shear mixing, García López & Spruit (1991) studied the transport of lithium as a function of spectral type. They found that low degree waves may become efficient sources of shear mixing for stars of the Li dip, as their production model is mainly dependent on the convective flux. A suitable choice of the mixing length then permits one to place the dip at the correct effective temperature (since it influences the disappearance of the convective zone). However, they needed to increase the efficiency of the wave generation by a factor of 15 over their estimation to correctly reproduce the dip. Furthermore, that model fails to reproduce the large abundance dispersion observed on the red side.

Boesgaard (1987) noticed that the Li dip in the Hyades coincides (in terms of effective temperature) with both a sharp drop in rotational velocities (see Fig. 1 bottom) and with the transition from high stellar activity to activity controlled by the stellar dynamo (Wolff et al. 1986). Rotation was then suggested to play a dominant role in the build up of the Li dip. Up to now, the different investigations of the possible connection between rotation and Li deficiencies in F stars have relied on highly simplified descriptions of the rotation-induced mixing processes. In the meridional circulation model of Tassoul & Tassoul (1982) used by Charbonneau & Michaud (1988), the feed-back effect due to angular momentum transport as well as the induced turbulence were ignored. Following Zahn (1992), Charbonnel et al. (1992, 1994) considered the interaction between meridional circulation and turbulence induced by rotation, but the transport of angular momentum was not treated self-consistently.

Here we go one step further by including in our models the most complete description currently available for rotation-induced mixing, and we compute simultaneously the transport of chemicals and the transport of angular momentum due to wind-driven meridional circulation. Let us stress again that this description has been used to successfully reproduce the slight over (under) abundances of C (N) observed in B type stars (cf. Talon et al. 1997) but failed to explain the flat rotation profile observed in the Sun (cf. Matias & Zahn 1997). Another process which efficiently transports angular momentum must be invoked in order to explain the helioseismic data. Here, we use the observations of lithium and rotation in the Hyades in order to get constraints on the onset of that process.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1998

Online publication: June 26, 1998
helpdesk.link@springer.de