Mg abundances in metal-poor halo stars as a tracer of early Galactic mixing^*

¹ Dept. of Physics and Astronomy, The Open University, Milton Keynes, MK7 6AA, UK e-mail: s.g.ryan@open.ac.uk
² Dept. of Physics and Astronomy, University of Leicester, Leicester, LE1 7RH, UK e-mail: e.arnone@ion.le.ac.uk
³ Dept. of Physics and Astronomy, University of Basel, Basel, Switzerland e-mail: argast@astro.unibas.ch
⁴ Research School of Astronomy and Astrophysics, The Australian National University, Weston Creek ACT 2611, Australia e-mail: jen@mso.anu.edu.au
⁵ Dept. of Physics and Astronomy and JINA: Joint Institute for Nuclear Astrophysics, Michigan State University, East Lansing, MI 48824, USA e-mail: beers@pa.msu.edu

Received: 6 June 2004
Accepted: 27 September 2004

Abstract

We present results of a detailed chemical analysis performed on 23 main-sequence turnoff stars having -3.4 ≤ [Fe/H] ≤ -2.2, a sample selected to be highly homogeneous in T_eff and log(g). We investigate the efficiency of mixing in the early Galaxy by means of the [Mg/Fe] ratio, and find that all values lie within a total range of 0.2 dex, with a standard deviation about the mean of 0.06 dex, consistent with measurement errors. This implies there is little or no intrinsic scatter in the early ISM, as suggested also by the most recent results from high-quality VLT observations. These results are in contrast with inhomogeneous Galactic chemical evolution (iGCE) models adopting present supernova (SN) II yields, which predict a peak-to-peak scatter in [Mg/Fe] as high as 1 dex at very low metallicity, with a corresponding standard deviation of about 0.4 dex. We propose that cooling and mixing timescales should be investigated in iGCE models to account for the apparent disagreement with present observations. The contrast between the constancy and small dispersion of [Mg/Fe] reported here and the quite different behaviour of [Ba/Fe] indicates, according to this interpretation, that Mg and Ba are predominantly synthesised in different progenitor mass ranges.