Pop III Hypernova Nucleosynthesis and Abundances in Very Metal-Poor Halo Stars

We calculate evolution and nucleosynthesis in massive Pop III stars with M = 13 ∼ 270M⊙, and compare the results with abundances of very metal-poor halo stars. The observed abundances can be explained by the energetic core-collapse supernovae with M ∼ < 130M⊙ (“hypernovae”) but not by pair-instability supernovae (PISNe) with M ∼ 140 − 270M⊙. This result constrain the IMF for the Pop III and very metal-poor Pop II stars. Observed Abundance and Hypernova Nucleosynthesis The observed abundances of metal-poor halo stars show quite interesting trends. There are significant differences between the abundance patterns in the iron-peak elements below and above [Fe/H]∼ −2.5. For [Fe/H] ∼ < −2.5, the mean values of [Cr/Fe] and [Mn/Fe] decrease toward smaller metallicity, while [Co/Fe] and [Zn/Fe] increases (McWilliam et al. 1995; Ryan et al. 1996; Primas et al. 2000; Blake et al. 2001; see also Fig.4). The Galaxy was not well mixed in such a early stage, and the abundance pattern of each supernova (SN) may be kept in the very metal-poor stars. However, these trend could not be explained with the previous results of nucleosynthesis by core-collapse SNe such as Woosley & Weaver (1995), Nomoto et al. (1997) and Limongi et al. (2000). Therefore, we need reconsideration for the nucleosynthesis of very metal-poor stars. We found that these trends are explained simultaneously in the context of Fe core-collapse SNe, if the mass from complete Si-burning is relatively large compared with that from incomplete Si-burning (Nakamura et al. 1999). However, all previous model calculations including Nakamura et al. (1999) underproduced the Zn/Fe ratio significantly (Fig. 1 left). In Umeda & Nomoto (2002, UN02 hereafter), we find that [Zn/Fe] is larger for deeper mass-cuts, smaller neutron excess, and larger explosion energies. Among them the large explosion energy is most important to realize the large [Zn/Fe] ratio (Fig. 1 right). The observed trends of the abundance ratios among the iron-peak elements are better explained with this high energy (“Hypernova”) models than the simple “deep” mass-cut effect, because the overabundance of Ni can be avoided in the hypernova models. Pair Instability Supernovae We also investigate the yields of pair-instability supernova explosions of M ≃ 140 − 300M⊙ stars. In Fig.2 we compare the abundance pattern of a hypernova and a PISN model. As can be seen [Zn/Fe] of PISNe are always small, because relatively large incomplete Si-burning region is formed in the explosion (UN02). Therefore, the abundance features of very metalpoor stars cannot be explained by pair-instability supernovae.