Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 350, 587-597 (1999)

Previous Section Next Section Title Page Table of Contents

8. How does helium content vary with metallicity?

Our isochrones are computed with the assumption that the helium content varies in proportion with metallicity, namely:


Any unevolved star presenting a significant deviation with respect to the isochrone corresponding to its metallicity may indicate that its helium content deviates from this simple, naive assumption. Actually there is already an indication that the Hyades do not follow this rule (Perryman et al. 1998), Hyades stars being metal-rich by 0.15 dex with respect to the Sun, without being correspondingly helium-rich. On the contrary, the metal-rich stars of the [FORMULA] Cen system were found also helium-rich (Y=0.30) by Fernandes & Neuforge, (1995) in agreement with Eq. (1).

In Fig. 6 we have plotted the most unevolved stars ([FORMULA]), labeled by their value of [Fe/H], with a solar ZAMS and a grid of isochrones of 8 Gyr, of metallicities -0.3, 0. and [FORMULA]. Only stars of metallicity corresponding to this interval are plotted. We admit that the age of these thin disk stars is between 0 and 8 Gyr. So if the error bars intersect either the right metallicity ZAMS or a right metallicity isochrone there is no compelling evidence that the scaled helium value is wrong. Globally all stars within the metallicity range -0.3 [FORMULA] [Fe/H] [FORMULA] 0.3 are within the corresponding isochrones. But inside the lane they define, some stars are not located exactly where one would expect to see them. We note that 5 stars (HD 37394, HD 131977, HD 166620, HD 201091 A, HD 219134) with metallicity very close to zero or even slightly negative, stand on the [Fe/H] = +0.3, 8 Gyr isochrone. The non-LTE correction for their iron abundance is negligible at this metallicity.

[FIGURE] Fig. 6. This figure is a large scale HR diagram for the subsample of unevolved stars in the LTE metallicity range -0.3 [FORMULA] [Fe/H] [FORMULA] 0.3. Each star is labeled by its LTE [Fe/H] value, followed by the HD number of the star. The full line is a standard isochrone for the metallicity [Fe/H] = 0. The dashed line is the standard isochrone corresponding to the LTE metallicity of the sample [Fe/H] = -0.3 The dot-dashed line corresponds to an isochrone of metallicity +0.3. All isochrones are for an age of 8 Gyr, but evolutionary effects are very small, and age is almost irrelevant. If stars stand between the limit isochrones, they do not always closely match the isochrone corresponding to their metallicity, leaving the impression that [FORMULA] is not constant in the sample.

As disk stars are younger than halo ones, sedimentation is less important. Its effect has been estimated, using a 5 Gyr, 0.8 [FORMULA] star of solar composition as a template. In this star, the shift in the HR diagram due to microscopic diffusion is of approximately 50 K in effective temperature and 0.02 dex in luminosity. These values are comparable to the observational error bars on these parameters for individual objects. Only for the oldest and more massive stars, the shift can reach more important values, around 120 K. Unfortunately, ages are most of the time unknown for individual unevolved stars of the disk.

Though certainly responsible for a scatter in the HR diagram (due to mass and age differences), and for a systematic very small shift towards lower effective temperature, any clear signature of this effect in the data seems impossible to decifer presently. Further work is planned on this point.

Young stars should not show this effect. Indeed, in another paper (Perrymann et al. 1998, the content in helium of the Hyades, sufficiently young to forget about diffusion, has been determined, and found to be solar like or slightly below, but certainly not enhanced at the level of the 0.15 dex level of [Fe/H]. Also the very young active stars as [FORMULA] Eridani (HD 22049) or HD 17925 do not significantly deviate from the solar ZAMS.

Another possibility for the 5 stars close to the +0.3 dex isochrone is that these stars have less helium than the scaled value given by Eq. (1). Interpolation of ZAMSs corresponding to various (Y,Z) combinations could provide an estimate of Y, for a star of known Z. However we must remember that the error bar of 0.1 dex on [Fe/H] translates into an error of 0.015 in Y, because of the (Y,Z) degeneracy in the HR diagram. Therefore we have decided to refrain from giving an helium content for individual unevolved stars until the effects of sedimentation are properly quantified at high metallicities.

Summarizing, our data are consistent with the scaled helium content relationship at a coarse level of accuracy, of the order of a factor of two in metallicity, but suggest that a scatter exists at a finer resolution.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1999

Online publication: October 4, 1999