2014A&A...565A..51C

Thanks to the heroic observational campaigns carried out in recent years we now have large samples of metal-poor stars for which measurements of detailed abundances exist. In particular, large samples of stars with metallicities -5<[Fe/H]←1 and measured abundances of Sr, Ba, Y, and Eu are now available. These data hold important clues on the nature of the contribution of the first stellar generations to the enrichment of our Galaxy. We aim to explain the scatter in Sr, Ba, Y, and Eu abundance ratio diagrams unveiled by the metal-poor halo stars. We computed inhomogeneous chemical evolution models for the Galactic halo assuming different scenarios for the r-process site: the electron-capture (EC) supernovae and the magnetorotationally driven (MRD) supernovae scenarios. We also considered models with and without the contribution of fast-rotating massive stars (spinstars) to an early enrichment by the s-process. A detailed comparison with the now large sample of stars with measured abundances of Sr, Ba, Y, Eu, and Fe is provided (both in terms of scatter plots and number distributions for several abundance ratios). The scatter observed in these abundance ratios of the very metal-poor stars (with [Fe/H]←2.5) can be explained by combining the s-process production in spinstars, and the r-process contribution coming from massive stars. For the r-process we have developed models for both the EC and the MRD scenarios that match the observations.With the present observational and theoretical constraints we cannot distinguish between the EC and the MRD scenarios in the Galactic halo. Independently of the r-process scenarios adopted, the production of elements by an s-process in spinstars is needed to reproduce the spread in abundances of the light neutron capture elements (Sr and Y) over heavy neutron capture elements (Ba and Eu). We provide a way to test our suggestions by means of the distribution of the Ba isotopic ratios in a [Ba/Fe] or [Sr/Ba] vs. [Fe/H] diagram.