The correlation, relativistic, and vibrational contributions to the dipolemoments, polarizabilities, and first and second hyperpolarizabilities of ZnS, CdS, and HgS

The dipole moments, dipole polarizabilities, and the first and second hyper polarizabilities of the Group IIb sulfides have been calculated by using di fferent high-level-correlated methods and including both the relativistic a nd vibrational contributions. The electron correlation effects have been st udied at the levels of the second-order Moller-Plesset perturbation theory and the coupled-cluster methods. The relativistic contributions and the int erference relativistic-correlation effects have been accounted for by using the spin-averaged Douglas-Kroll approximation. The vibrational properties (pure vibrational contributions and the zero-point vibrational averaging co rrections) have been computed using CCSD(T) theory with and without relativ istic corrections. The present pure electronic nonrelativistic results exhi bit essentially the same pattern as that observed for similar molecules stu died earlier. Most of the relativistic effects on dipole moments and dipole polarizabilities is accounted for at the level of the SCF approximation an d rapidly increases with the nuclear charge of the heavy atom. The contribu tion of the relativistic-correlation interference terms has been found to b e quite significant for axial components of the first and second dipole hyp erpolarizabilities. All the properties reported here are static. This is th e first study which reports on the relativistic contributions to hyperpolar izabilities as well as on vibrational effects upon both polarizabilities an d hyperpolarizabilities of heavy metal (Group IIb) involving compounds. Thu s the reported results add to the knowledge and understanding of the import ance of the electron correlation, relativistic, and vibrational effects on electric properties of heavy molecules and extend the corresponding data be yond the linear response approximation. The reliability of the computed dat a is discussed in terms of the underlying approximations and limitations of methods used in this study. (C) 1999 American Institute of Physics. [S0021 -9606(99)30440-2].