Theoretical prediction of the second to fourth actinide ionization potentials

Theoretical values for the experimentally mostly undetermined second to fourth ionization potentials of the actinide elements Ac to Lr are presented. One set of results is based on relativistic ab initio pseudopotential calculations at the complete active space self-consistent field and multi-reference averaged coupled-pair functional level of theory, corrected for spin-orbit effects, combined with extrapolations of the correlation contributions to the basis set limit. A second set of values also takes into account the errors of the applied pseudopotentials with respect to all-electron multi-configuration self-consistent field calculations applying the Dirac-Coulorrlb Hamiltonian. The quality of the predictions is estimated from related calculations on the lanthanide elements La to Lu, where the experimental values are available.