“…Let us remember that, in the present relativistic calculation, only a one-component wave function is used through the MSformalism.21, 37,38 Consequently, the SO operator cannot be introduced self-consistently as in relativistic Hartree-Fock-Slater calculations39,40 or in Dirac-Slater SW-Xa calculations (which use the four components through the MS technique).41 Therefore, SO effects are taken into account by perturbation theory, once self-consistency has been reached.21,42 It has been shown on specific examples that this technique is a very good approximation of the Dirac-Slater SWformalism.43…”