2025MNRAS.538..537M

A large number of $\bf{R}$-matrix calculations of electron-impact excitation for heavy elements ($Z\gt 70$) have been performed in recent years for applications in fusion and astrophysics research. With the expanding interest in heavy ions due to kilonova (KN) events such as AT2017gfo and AT2023vfi, these new data can be utilized for the diagnosis and study of observed KN spectra. In this work, recently computed electron-impact excitation effective collision strengths are used, for the first three ionization stages of tungsten (W, $Z=74$), platinum (Pt, $Z=78$), and gold (Au, $Z=79$), to construct basic collisional radiative models tailored for the late stage nebular phases of KN. Line luminosities are calculated at a range of electron temperatures and densities and the strengths of these lines for a representative ion mass are compared. For the case of W III, these optically thin intensities are additionally used to constrain the mass of this ion in both AT2017gfo and AT2023vfi. Comparing with theoretical predictions of nucleosynthesis yields from neutron star merger simulations, broad agreement with the inferred ion masses of W is found. Furthermore, we highlight the value of W measurements by showing that the abundance of other groups of elements and outflow properties are constrained by exploiting theoretically motivated correlations between the abundance of W and that of lanthanides or third r-process peak elements. Based on simple estimates, we also show that constraints on the distribution of tungsten in the ejecta may be accessible through the line shape, which may also yield information on the neutron star merger remnant evolution.