An Exploration of Recombination of Uranium with application to Kilonovae Spectra

Dielectronic recombination (DR) is expected to be the dominant recombination process during the non-local thermodynamic equilibrium (non-LTE) phase of kilonovae, yet reliable DR data remain unavailable for most heavy ions. Current spectral models therefore rely on simplified recombination prescriptions, introducing significant uncertainties into predicted spectra. We present an optimization strategy for open f-shell ions using \texttt{AUTOSTRUCTURE}, targeting uranium ions U II--U IV relevant to kilonova ejecta. As a benchmark case, calculations are performed for Nd III to validate the treatment of the f-shell structure and its impact on DR. The resulting DR rate coefficients are of order $10^{-10}$--$10^{-12}$ cm$^{3}$s$^{-1}$ over temperatures relevant to kilonova plasmas. The optimized rates are intended for implementation in radiative-transfer calculations with \texttt{SUMO} to assess the sensitivity of kilonova spectra to improved recombination physics. The Nd III benchmark demonstrates that refinements to the atomic structure can produce measurable changes in spectral features, motivating similar calculations for actinide ions.