A b i n i t i o model potential study of the equilibrium geometry of alkaline earth dihalides: MX2 (M=Mg, Ca, Sr, Ba; X=F, Cl, Br, I)

Abstract

Ion model and equilibrium configuration of the alkaline-earth dihalide dimers. The Mg2Cl4 molecule J. Chem. Phys. 66, 3920 (1977); 10.1063/1.434468Geometry and vibrational spectra of the alkaline earth dihalides. III. MgCl2, CaCl2, SrCl2, and BaCl2The ground state equilibrium geometry of alkaline earth dihalides MX 2 (M = Mg, Ca, Sr, Ba; X = F, CI, Br, I) has been optimized at the Hartree-Fock (HF) level using the Cowan-Griffin relativistic ab initio model potential method and a uniformly good, extended, spd valence basis set. The results show that, according to the method, all magnesium dihalides and CaCI 2 , CaBr 2 , and Calz are linear, SrF2 and all barium dihalides are bent, and CaF2 and SrCI 2 , SrBr 2' and SrI z are quasilinear molecules. The alkaline earth (n -I) d orbitals are shown to be responsible for the bending of the heavier molecules while their (n -l)p orbitals contribute considerably to the final quantitative prediction of the apex angle and the relative stability of the bent structures. Relativistic effects are shown to be very small on the bond distances and vibrational frequencies; they are important on the size of the bending barrier of the bent molecules. The results obtained are compared to previous theoretical studies and provide some insight in the interpretation of the contradictory experimental conclusions. 3762

The abinitio model potential method. Cowan–Griffin relativistic core potentials and valence basis sets from Li (Z = 3) to La (Z = 57)

“…Dlffuse functions for Rydberg states and for anions can be taken from ref. 14, and, for those not available, the same obtention procedure can be followed, i.e., p-type GTF with exponents 0.043 and 0.039 for Br and I provide an excellent agreement with the numerical HF atomic electron affinities and they can be used as diffuse functions for anions (9). Dzffuse d-functions able to provide a balanced description of s'd", sld"+ I, and sodnf2 configurations of first-row and second-row transition metal elements are available in refs.…”

“…In addition to the Cowan-Griffin relativistic core model potentials and valence basis sets already published for the second series transition metal elements, those for Sr, Ba, Br, and I have been obtained prior to conducting theoretical Can studies of the equilibrium geometry of alkaline earth dihalides (9). The latter are presented in this paper together with those of their respective periods, in order to facilitate the systematic performance of CG-AIMP molecular calculations.…”

1994 Polanyi Award Lecture Concept of active electrons in chemistry

“…[20] is certainly a one-electron operator but the evaluation of the expectation values involves compytation of so-called two-electron energy integrals because of Jc and Kc in Vco,, This is not a welcome feature because we are actually interested in economizing the molecular calculation by dealing only with valence electrons from the constituent atoms. It is desirable to replace QCor,,, with something more economical computationally.…”

“…This version, the AIMP method of Seijo and Barandiarhn, was applied to study the equilibrium geometry of alkaline earth dihalides MX2 (M = Mg, Ca, Sr, Ba: X = F, C1, Br, I) (20), the hydrides XH2, XH4 (X = Si, Ge, Sn, Pb), and also the oxides XO (X = Ge, Sn, Pb). These are mostly without the reference all-electron calculation and emphasis was placed on the comparison between the NR-AIMP and the QR-AIMP to gauge the relativistic effect.…”

“…[21] 623 is expressed as a local spherical function of r in the form, This is the origin of the term "model potential," which probably has survived too long in view of many later developments in the method. The model potential (mp) parameters {Al, al) were optimized through a series of atomic H F calculations for the valence electrons using the one-electron Hamiltonian [20] in such a way that the orbital energies and the radial orbital functions {EFP, RFP} will be as close as possible to those of the reference all-electron calculation, {&:ef1~:ef). At this stage, the process of optimization was not automated.…”