Performance of the Effective Core Potentials of Ca, Hg, and Pb in Complexes with Ligands Containing N and O Donor Atoms.

Abstract

This paper presents a systematic study of the performance of the relativistic effective core potentials (RECPs) proposed by Dolg-Stoll-Preuss, Christiansen-Ermler, and Hay-Wadt for Ca(2+), Hg(2+), and Pb(2+). The RECPs performance is studied when these cations are combined with ethyleneglycol, 2-aminoethanol, and ethylenediamine to form bidentate complexes. First, the description of the bidentate ligands is analyzed with the Kohn-Sham method by using SVWN, BLYP, and B3LYP exchange-correlation functionals, and they are compared with the Møller-Plesset perturbation theory (MP2); for all these methods the TZVP basis set was used. We found that the BLYP exchange-correlation functional gives similar results like those obtained by the B3LYP and MP2 methods. Thus, the bidentate metal complexes were studied with the BLYP method combined with the RECPs. To compare RECPs performance, all the systems considered in this work were studied with the relativistic all-electron Douglas-Kroll (DK3) method. We observed that the Christiansen-Ermler and Dolg-Stoll-Preuss RECPs give the best energetic and geometrical description for Ca and Hg complexes when compared with the all-electron method. For Pb complexes the spin-orbit effect and the basis set superposition error must be taken into account in the RECP calculations. In general, the trend showed in the complexation energies with the all-electron method is followed by the complexation energies computed with all the pseudopotentials tested in this work.