Comment on "Nonadiabatic couplings from the Kohn-Sham derivative matrix: Formulation by time-dependent density-functional theory and evaluation in the pseudopotential framework"

Abstract

The issue of the ab initio evaluation of the first-order nonadiabatic couplings (NAC's) is re-examined. In particular, a recent derivation in Phys. Rev. A 82, 062508 (2010) of the NAC's is corrected by performing an extension of the derivation to the case of arbitrary values of the Kohn-Sham (KS) occupation factors. A single expression for the nonadiabatic couplings, valid both for integer and fractional values of the KS occupation factors, is provided.