Physical Conditions for the r-process. I. Radioactive Energy Sources of Kilonovae

Abstract

Abstract Radioactive energies from unstable nuclei made in the ejecta of neutron star mergers play principal roles in powering kilonovae. In previous studies, power-law-type heating rates (e.g., ) have frequently been used, which may be inadequate if the ejecta are dominated by nuclei other than the A ∼ 130 region. We consider, therefore, two reference abundance distributions that match the r-process residuals to the solar abundances for A ≥ 69 (light trans-iron plus r-process elements) and A ≥ 90 (r-process elements). Nucleosynthetic abundances are obtained by using free-expansion models with three parameters: expansion velocity, entropy, and electron fraction. Radioactive energies are calculated as an ensemble of weighted free-expansion models that reproduce the reference abundance patterns. The results are compared with the bolometric luminosity (> a few days since merger) of the kilonova associated with GW170817. We find that the former case (fitted for A ≥ 69) with an ejecta mass 0.06 M ⊙ reproduces the light curve remarkably well, including its steepening at ≳7 days, in which the mass of r-process elements is ≈0.01 M ⊙. Two β-decay chains are identified: 66Ni 66Cu 66Zn and 72Zn 72Ga 72Ge with similar halflives of parent isotopes (≈2 days), which leads to an exponential-like evolution of heating rates during 1–15 days. The light curve at late times (>40 days) is consistent with additional contributions from the spontaneous fission of 254Cf and a few Fm isotopes. If this is the case, the GW170817 event is best explained by the production of both light trans-iron and r-process elements that originate from dynamical ejecta and subsequent disk outflows from the neutron star merger.