The AT2017gfo has added to the growing interest in r-process elements, which are expected to be particularly abundant in the nucleosynthesis trajectories of neutron star mergers. With the choice of elements guided by nuclear physics at particular values of Y_e, our group calculates atomic data catered for modelling of the astrophysical objects without the use of local-thermodynamic-equilbirum using collisional radiative modelling. By enforcing observed luminosities, we are then able to make mass estimates of the candidate ions. This serves particularly as a test of the calculated atomic data and, based on the feasibility of the mass estimate, also the underlying nucleosynthesis theory. This event, as well as the GRB230307A last year sport features consistent with the fine structure lines of Te and W, which are particularly interesting to atomic and nuclear physicists likes – as these species lie at the second and third peaks respectively of the r-process abundance. These features also occur at the late stage collisionally dominated regime of the events, making an optically thin model suitable for their analysis. Collisional radiative modelling, and particularly mass estimation of these species in and out of LTE will be discussed.