Evolution and nucleosynthesis of extremely metal-poor and metal-free low- and intermediate-mass stars

Abstract

Context. Models of primordial and hyper-metal-poor stars that have masses similar to the Sun are known to experience an ingestion of protons into the hot core during the core helium flash phase at the end of their red giant branch evolution. This produces a concurrent secondary flash powered by hydrogen burning that gives rise to further nucleosynthesis in the core. Aims. We aim to model the nucleosynthesis occurring during the proton ingestion event to ascertain if any significant neutron-capture nucleosynthesis occurs. Methods. We perform post-process nucleosynthesis calculations on a one-dimensional stellar evolution calculation of a star with mass 1 M and a metallicity of [Fe/H] = −6.5 that suffers a proton ingestion episode. Our network includes 320 nuclear species and 2366 reactions and treats mixing and burning simultaneously. Results. We find that the mixing and burning of protons into the hot convective core leads to the production of 13 C, which then burns via the 13 C(α, n) 16 O reaction, releasing a large number of free neutrons. During the first two years of neutron production the neutron poison 14 N abundance is low, allowing the prodigious production of heavy elements such as strontium, barium, and lead via slow neutron captures (the s process). These nucleosynthetic products are later carried to the stellar surface and ejected via stellar winds. We compare our results with observations of the hyper-metal-poor halo star HE 1327-2326, which shows a strong Sr overabundance. Conclusions. Our model provides the possibility of self-consistently explaining the Sr overabundance in HE 1327-2326 together with its C, N, and O overabundances (all within a factor of ∼4) if the material were heavily diluted, for example, via mass transfer in a wide binary system. The model produces at least 18 times too much Ba than observed, but this may be within the large modelling uncertainties. In this scenario, binary systems of low mass must have formed in the early Universe. If this is true, it puts constraints on the primordial initial mass function.

S-process in extremely metal-poor, low-mass stars

“…Hollowell et al 1990). In contrast, Campbell et al (2010) presented a model with a metallicity of [Fe/H] = −6.5 where Δt PIE ∼ 20 yrs, which is about a factor of 50 larger than our findings. According to our calculations, differences in Δt PIE imply differences in the production of s-process elements.…”

“…According to our calculations, differences in Δt PIE imply differences in the production of s-process elements. Therefore, the large variations found in Δt PIE as described, which span up to three orders of magnitude, are probably the reason why results for nucleosynthesis of s-process elements from different authors range from no production at all, as seen for example in Suda & Fujimoto (2010), to extremely large production factors, as seen in Campbell et al (2010), and described in Sect. 1.…”

“…Suda et al (2004) claim that intermediate-mass stars are the polluting stars, which are responsible for the observed enhancements in carbon and s-process elements in CEMP-s stars. Campbell et al (2010) have calculated models of low-mass EMP stars during the core He-flash and found that the PIE leads to the production of a large amount of neutrons (n n > 10 14 cm −3 ) lasting for ∼0.2 years. Consequently, they have found production of s-process elements during the core He-flash of this star.…”

“…We present evolutionary and s-process calculations for the core He-flash of 1 M stellar models. Our approach differs from Campbell et al (2010) in that we use a more complete network that includes a neutron sink in the evolutionary calculations and a post-processing code for a posteriori calculations of s-process nucleosynthesis. We have also been able to follow the evolution of our models through the entire PIE and subsequent dredge-up phase.…”

“…In contrast, Campbell et al (2010) estimated the surface enrichment from previous simulations (using a smaller network) due to numerical problems in following the subsequent dredge-up. We hope to close the gaps left by Campbell et al (2010) by presenting models that include the evolution of the star during the dredge-up episode. In this way the necessity of extrapolating the resulting envelope abundances for all elements is prevented.…”