Reply to "Comment on 'Taming multiple valency with density functionals: A case study of defective ceria'"

We address the issues raised in the preceding Comment by discussing the effects of an explicit account of the nonlinear core correction (NLCC) into the Ce pseudopotential employed in our previous calculations of reduced ceria [Phys. Rev. B 71, 041102(R) (2005)]. At the plain density-functional-theory (DFT) level, such an account brings our pseudopotential results in good agreement with all-electron ones, which do not predict the insulating character of reduced ceria. At the DFT+U level, the inclusion of the NLCC has no effect on the previously reported electronic structure and equilibrium geometries. The dependency of the DFT+U energy on the choice of the projector functions defining the Hubbard-U functional is discussed in further detail.