The observation of a radioactively powered kilonova AT 2017gfo associated with the gravitational wave event GW170817 from a binary neutron star merger proves that these events are ideal sites for the production of heavy r-process elements. The gamma-ray photons produced by the radioactive decay of heavy elements are unique probes for the detailed nuclide compositions. Based on the detailed r-process nucleosynthesis calculations and considering radiative transport calculations for the gamma rays in different shells, we study the gamma-ray emission in a merger ejecta on a timescale of a few days. It is found that the total gamma-ray energy generation rate evolution is roughly depicted as E ̇ ∝ t − 1.3 . For the dynamical ejecta with a low electron fraction (Y e ≲ 0.20), the dominant contributors of gamma-ray energy are the nuclides around the second r-process peak (A ∼ 130) and the decay chain of 132Te (t 1/2 = 3.21 days) → 132I (t 1/2 = 0.10 days) → 132Xe produces gamma-ray lines at 228, 668, and 773 keV. For the case of a wind ejecta with Y e ≳ 0.30, the dominant contributors of gamma-ray energy are the nuclides around the first r-process peak (A ∼ 80) and the decay chain of 72Zn (t 1/2 = 1.93 days) → 72Ga (t 1/2 = 0.59 days) → 72Ge produces gamma-ray lines at 145, 834, 2202, and 2508 keV. The peak fluxes of these lines are 10−9 ∼ 10−7 ph cm−2 s−1, which are marginally detectable with the next-generation MeV gamma-ray detector ETCC if the source is at a distance of 40 Mpc.

The association of GW170817/GRB 170817A/AT2017gfo provides the first direct evidence for neutron star mergers as significant sources of r-process nucleosynthesis. A gamma-ray transient (GRT) would be powered by the radioactive decay of the freshly synthesized r-process elements. By analyzing the composition and gamma-ray opacity of the kilonova ejecta in detail, we calculate the lightcurve and spectrum of the GRT for a range of spherically symmetric merger ejecta models with mass M ej = 0.001 to ∼0.05M ⊙ and expansion velocity v ej = 0.1c to ∼0.4c. It is found that the peak of the GRT lightcurve depends on M ej and v ej as t pk ≈ 0.5 days ( M ej / 0.01 M ⊙ ) 1 / 2 ( v ej / 0.1 c ) − 1 and L pk ≈ 2.0 × 10 41 erg s − 1 ( M ej / 0.01 M ⊙ ) 1 / 2 ( v ej / 0.1 c ) . Most radiating photons are in the 100–3000 keV band and the spectrum peaks at about 800 keV for different nuclear physics inputs. The line features are blurred out by the Doppler broadening effect. Adopting the ejecta parameters reported in the literature, we examine the detection probability of the possible GRT associated with AT2017gfo. We show that the GRT cannot be convincingly detected with either current or proposed missions in the MeV band, such as ETCC and AMEGO. The low gamma-ray flux, together with the extremely low event rate at local universe, makes a discovery of GRTs a great challenge.