Comment on "Going beyond the frozen core approximation: development of coordinate-dependent pseudopotentials and application to Na2(+)" [J. Chem. Phys. 138, 054110 (2013)].

The new coordinate-dependent pseudopotential for Na2(+) by Kahros and Schwartz [J. Chem. Phys. 138, 054110 (2013)] is assessed and compared to the pseudopotential approach by Fuentealba et al. [Chem. Phys. Lett. 89, 418 (1982)] which incorporates the coordinate-dependent core-polarization potential by Müller and Meyer [J. Chem. Phys. 80, 3311 (1984)]. In contrast to the latter approach, the one by Kahros and Schwartz does not reproduce the accurately known experimental data and∕or high level theoretical results for Na2(+). The treatment of core polarization by Kahros and Schwartz neglects the dynamic polarization of atomic cores which is much more important for Na2(+) than the static one. On the other hand, the Kahros and Schwartz method heavily overestimates frozen-core corrections at the Hartree-Fock level by compounding them with artifacts of a superposition of non-norm-conserving pseudopotentials.