Explosive Nucleosynthesis in GRB Jets Accompanied by Hypernovae

Two-dimensional hydrodynamic simulations are performed to investigate explosive nucleosynthesis in a collapsar using the model of MacFadyen and Woosley (1999). It is shown that {sup 56}Ni is not produced in the jet of the collapsar sufficiently to explain the observed amount of a hypernova when the duration of the explosion is {approx} 10 sec, which is considered to be the typical timescale of explosion in the collapsar model. Even though a considerable amount of {sup 56}Ni is synthesized if all explosion energy is deposited initially, the opening angles of the jets become too wide to realize highly relativistic outflows and gamma-ray bursts in such a case. From these results, it is concluded that the origin of {sup 56}Ni in hypernovae associated with GRBs is not the explosive nucleosynthesis in the jet. We consider that the idea that the origin is the explosive nucleosynthesis in the accretion disk is more promising. We also show that the explosion becomes bi-polar naturally due to the effect of the deformed progenitor. This fact suggests that the {sup 56}Ni synthesized in the accretion disk and conveyed as outflows are blown along to the rotation axis, which will explain the line features of SN 1998bw and double peaked line features of SN 2003jd. Some fraction of the gamma-ray lines from {sup 56}Ni decays in the jet will appear without losing their energies because the jet becomes optically thin before a considerable amount of {sup 56}Ni decays as long as the jet is a relativistic flow, which may be observed as relativistically Lorentz boosted line profiles in future. We show that abundance of nuclei whose mass number {approx} 40 in the ejecta depends sensitively on the energy deposition rate, which is a result of incomplete silicon burning and alpha-rich freezeout. So it may be determined by observations of chemical composition in metal poor stars which model is the proper one as a model of a gamma-ray burst accompanied by a hypernova