hide
Free keywords:
Astrophysics, High Energy Astrophysical Phenomena, astro-ph.HE, Astrophysics, Solar and Stellar Astrophysics, astro-ph.SR, Physics, Atomic Physics, physics.atom-ph
Abstract:
Kilonova spectra provide us with information of r-process nucleosynthesis in
neutron star mergers. However, it is still challenging to identify individual
elements in the spectra mainly due to lack of experimentally accurate atomic
data for heavy elements in the near-infrared wavelengths. Recently, Domoto et
al. (2022) proposed the absorption features around 14500 A in the observed
spectra of GW170817/AT2017gfo as Ce III lines. But they used theoretical
transition probabilities (gf-values) whose accuracy is uncertain. In this
paper, we derive the astrophysical gf-values of the three Ce III lines, aiming
at verification of this identification. We model high resolution H-band spectra
of four F-type supergiants showing the clear Ce III absorption features by
assuming stellar parameters derived from optical spectra in literatures. We
also test the validity of the derived astrophysical gf-values by estimating Ce
III abundances in Ap stars. We find that the derived astrophysical gf-values of
the Ce III lines are systematically lower by about 0.25 dex than those used in
previous work of kilonovae, while they are still compatible within the
uncertainty range. By performing radiative transfer simulations of kilonovae
with the derived gf-values, we find that the identification of Ce III as a
source of the absorption features in the observed kilonova spectra still
stands, even considering the uncertainties in the astrophysical gf-values. This
supports identification of Ce in the spectra of GW170817/AT2017gfo.