Carbon-enhanced metal-poor stars: Relics from the dark ages

We use detailed nucleosynthesis calculations and a realistic prescription for the environment of the first stars to explore the first episodes of chemical enrichment that occurred during the dark ages. Based on these calculations, we propose a novel explanation for the increased prevalence of carbon-enhanced metal-poor (CEMP) stars with decreasing Fe abundance: the observed chemistry for the most metal-poor Galactic halo stars is the result of an intimate link between the explosions of the first stars and their host minihalo's ability to retain its gas. Specifically, high-energy supernovae produce a near solar ratio of C/Fe, but are effective in evacuating the gas from their host minihalo, thereby suppressing the formation of a second generation of stars. On the other hand, minihalos that host low-energy supernovae are able to retain their gas and form a second stellar generation, but, as a result, the second stars are born with a supersolar ratio of C/Fe. Our models are able to accurately reproduce the observed distributions of [C/Fe] and [Fe/H], as well as the fraction of CEMP stars relative to non-CEMP stars as a function of [Fe/H] without any free parameters. We propose that the present lack of chemical evidence for very massive stars (≳ 140 M ⊙) that ended their lives as a highly energetic pair-instability supernova does not imply that such stars were rare or did not exist; the chemical products of these very massive first stars may have been evacuated from their host minihalos and were never incorporated into subsequent generations of stars. Finally, our models suggest that the most Fe-poor stars currently known may have seen the enrichment from a small multiple of metal-free stars, and need not have been exclusively enriched by a solitary first star. These calculations also add further support to the possibility that some of the surviving dwarf satellite galaxies of the Milky Way are the relics of the first galaxies.