TESTING OF THE OPTIMUM BASIS SET FOR MODELING THE HEXABROMOTELURATE ANION

Abstract

Using the PRIRODA 19 program, the bond lengths Te–Br have been calculated for the hexabromotelurate anion [TeBr6]2– within the frameworks of the GGA PBE density functional theory and the Møller–Plesset second order perturbation theory. In order to find the optimum basis set, the λXa (where X = 1-4) basis sets developed by D. Laikov, with diffusion functions, were tested. It must be noticed that in the case of the λXa (where X = 1-4) basis sets, the anion [TeBr6]2– is described by different numbers of basis functions: 261 for the double-zeta basic set λ1a, 422 for the triple-zeta basic set λ2a, 660 and 989 for the basis sets λ3a and λ4a, respectively. As the system under investigation contains relatively heavy atoms, the relativistic effects were taken into account using the four-component scalar-relativistic Hamiltonian by Dirac and using the new two-component AAA Hamiltonian by Laikov.
 The optimization of the geometric parameters of the hexabromotelurate anion was carried out with no symmetry restrictions. In all cases, the optimized [TeBr6]2– geometry corresponds to the nearly perfect octahedral symmetry; so the average Te–Br bond lengths have been considered. In the case of the PBE/λXa method (where X = 1-4), an increase in the size of the basis set results in a shortening of the Te–Br bond lengths. It has been shown that the difference between the four-component and two-component relativistic approximations increases with an increase of the basic set used. In turn, in the case of the MP2 method, the difference between the four-component and two-component relativistic approximations decreases with an increase of the basic set, and goes through the minimum for the quadruple-zeta λ3a basis set. Thus, in the post-Hartree–Fock approach, such as the Møller–Plesset perturbation theory and the coupled clusters theory, the agreement between the above relativistic approximations is reasonably good, and the two-component scalar-relativistic AAA Hamiltonian by Laikov (which is characterized by much lesser recourse demands) is more optimal for a variety of «heavy» and time-consuming calculations at the higher levels of theory. For modeling the geometric parameters of hexabromotelurate anion with an accuracy of 0.01 Å, we recommend the use of the triple-zeta basis set with the scalar-relativistic Hamiltonian.