Spectroscopic modelling of astrophysical mergers and kilonovae

Second, we comprehensively modelled the entire spectral sequence, using realistic composition profiles from theoretical simulations, to deduce the properties of the ejecta across its entire lifetime. We found that the early spectra (≲ 7 days) can be well-reproduced with a combination of first <i>r</i>-process peak elements, and lanthanides. We found we must invoke a disjoint in composition to obtain agreement with the early spectra, indicative of a ‘two-component’ model for AT2017gfo.

We also modelled the late spectra, and identified a number of prominent emission features. We were able to reproduce the feature at ∼ 1𝜇m across all phases with the strong Sr II NIR triplet, although we require clumped material at late times. We also modelled the two NIR emission features at ∼ 1.58 and ∼ 2.07𝜇m, both as P-Cygni and pure emission features, but concluded that without complete atomic data that contains NIR line information for the heavy elements, we cannot uniquely identify their source.

Finally, we modelled the fast-evolving transient, AT2018kzr. Its lightcurve declined comparably rapidly to AT2017gfo, indicating a compact merger scenario. Our analysis determined that it did not synthesise any <i>r</i>-process material, and that it was the result of the merger of an oxygen-neon white dwarf with a compact binary – either a neutron star or a stellar mass black hole.

Although this transient is unrelated to <i>r</i>-process nucleosynthesis, understanding these types of merger systems is important for developing our understanding of fast-evolving transients that occupy the same parameter space as kilonovae.