Carbon-enhanced metal-poor (CEMP) stars in the Galactic Halo display enrichments in heavy elements associated with either the s (slow) or the r (rapid) neutron-capture process (e.g., barium and europium respectively), and in some cases they display evidence of both. The abundance patterns of these CEMP-s/r stars, which show both Ba and Eu enrichment, are particularly puzzling since the s and the r processes require neutron densities that are more than ten orders of magnitude apart, and hence are thought to occur in very different stellar sites. We investigate whether the abundance patterns of CEMP-s/r stars can arise from the nucleosynthesis of the intermediate neutron-capture process (the i process), which is characterised by neutron densities between those of the s and the r processes. Using nuclear network calculations, we study neutron capture nucleosynthesis at different constant neutron densities n ranging from 10 7 to 10 15 cm −3 . Neutron densities on the highest side of this range result in abundance patterns that show an increased production of heavy s-and r-process elements but similar levels of the light s-process elements. With our i-process model, we are able to reproduce the abundance patterns of 20 CEMP-s/r stars that could not be explained by s-process nucleosynthesis.KEYWORDS: nucleosynthesis, stars: population II, stars: carbon
MethodThe nucleosynthesis tools that we use are NucNet Tools, a set of C/C++ codes developed by Bradley S. Meyer [1]. In this study, the codes are used to follow the nuclear processing of a single zone with given initial composition under conditions of fixed temperature, density and neutron density. The nuclear network used for this project contains the 5442 isotopes and 45831 reactions from the JINA Reaclib V0.5 database [2]. In more recent releases, neutron-capture reaction rates from KADoNiS v0.2 [3] are included and refitted to eliminate blow-ups at low temperatures and to match the theory at higher temperatures (JINA Reaclib label kd02). However, in some cases the fits underpredict the rates in the temperature regime relevant for the conditions in an AGB intershell region, with deviations up to two orders of magnitude, for example, in the case of 151 Eu(n,γ) 152 Eu. Such a large underprediction of the reaction rate introduces artificial bottlenecks on the neutron capture path. Because of this, we use the previous version of the the refitted rates (JINA Reaclib label ka02).The physical input conditions are adapted from the density and temperature profiles of the intershell region that [4] found for a low-metallicity AGB star. In particular, we present here the test case with T = 1.5 × 10 8 K and ρ = 1600 g cm −3 . To model the nucleosynthesis in the intershell region, the composition of the input zone is adapted from the intershell composition of [5]. In particular, we use the abundances of 320 isotopes from an AGB star model with metallicity Z = 10 −4 and initial mass 1