The origin of palladium and silver

Context. In recent years, progress has been made in understanding the rapid neutron-capture process (r-process) and current studies suggest the need for a second, so-called weak r-process. Observational proof of this possible second branch of the r-process may come from a detailed knowledge of the formation and evolution of silver (Ag) and to some extent palladium (Pd) abundances in halo stars. Aims. We study the silver and palladium abundances of a large sample of stars (only a few measurements have been made so far) to gain insight into the formation process of lighter neutron-capture elements. Methods. We analysed a sample of stars (including both dwarfs and giants), for which we determined in a consistent manner stellar parameters and abundances. The abundances were derived via spectrum synthesis, making use of the MOOG synthetic spectrum code (1D LTE) and MARCS stellar model atmospheres (1D LTE). Results. We derived the Ag and Pd abundances of 56 stars, the largest sample to date for which both Pd and Ag have been studied. The stars span a metallicity range of 2.5 dex, from the metal-poor giants at [Fe/H] similar to-3.2 to the more metal-rich dwarfs at [Fe/H] similar to -0.6. Both elements display rather flat trends with metallicity, with some dispersion being present at the lowest metallicities. Conclusions. The similarity detected in the evolutionary trends of both Ag and Pd is indicative of a common origin. Qualitative comparisons with some theoretical calculations seem to confirm the need for a second r-process, responsible for their formation. Further abundance studies (exploring more light n-capture elements) and comparisons with a wider variety of theoretical models are needed to fully characterise the details and site of the formation mechanism(s).