John R. Van Wazer

On pseudopotential and effective‐potential SCF theory and its application to compounds of heavy elements

A formulation of pseudopotential and effective‐potential theory is developed within the framework of the Hartree–Fock formalism. It is shown that one‐electron pseudopotentials may be used for many‐valence‐electron atoms and molecules. An SCF computational procedure (the NOCOR method) is described that requires a minimal amount of parameterization. Sample calculations are performed on the four halogen atoms and ten dihalogen molecules. These results indicate that the ’’frozen‐core’’ approximation is quite accurate at least through the fourth row of the Periodic Table.

Application of effective potentials to relativistic hartree—fock calculations

Abstract The Phillips—Kleinman pseudopotential and the operator equivalence techniques for calculating effective potentials are applied directly to the Dirac—Hartree—Fock calculation of molecular properties. The results based on both relativistic and nonrelativistic core functions are compared for the PbO molecule.

The structure of molecular effective potentials in compounds of heavy elements, with application to I2

A new expression of the Phillips–Kleinman pseudopotential for many‐valence‐electron molecules is proposed. All remaining terms in the valence Hamiltonian, represented by a local potential, are evaluated from ab initio expressions based on the form of the atomic Fock operator. Sample calculations are reported for the orbital energies, equilibrium bond length, and vibrational frequency of the ground state of the I2 molecule.

Molecular pseudopotential calculations on transition-metal complexes: Ni(CO)4, Pd(CO)4, and Pt(CO)4

Abstract The electronic structures of the transition-metal complexes Ni(CO) 4 , Pd(CO) 4 , and Pt(CO) 4 are studied using ab initio molecular pseudopotential theory. We report the valence-orbital energies for each compound. For Ni(CO) 4 , we also present the calculated equilibrium Niue5f8C bond length and the A 1 vibrational stretching frequency. The Ni(CO) 4 results are compared with conventional SCF calculations and with experiment and are found to be in excellent agreement.

Pseudopotential SCF–MO studies of hypervalent compounds. I. XeF2 and XeF4

New evidence bearing upon the anomalous properties of xenon hexafluoride has been obtained via the ab initio molecular orbital approach applied successfully to the di‐ and tetrafluorides in paper I. Structures of both XeF+5 and XeF6 are governed by a stereochemically active lone pair. In the case of the square–pyramidal cation the Fax–Xe–Feq angle calculated for the bare ion is within 2° of the value observed in the crystalline complex. For the hexafluoride, however, the calculated deformation from Oh symmetry is appreciably greater than that deduced from electron diffraction intensities. Nevertheless, the results of calculations are in sufficient conformity with the Bartell–Gavin, Pitzer–Bernstein interpretation and at variance with the ’’electronic‐isomers’’ interpretation to leave little doubt about the answer. With increasing fluorination in the XeFn series the HOMO–LUMO energy difference decreases and the second‐order Jahn–Teller effect is enhanced. Increasing fluorination (and increased positive cha...

The geometric and electronic structures of I3− and I5− from effective-potential calculations

Abstract The (ab initio) effective-potential theory developed by Ewig et al. is applied to the structures of the polyiodide ions, I 3 − and I 5 − . The bare ions I 3 − and I 5 − are found by optimization of the geometry to be symmetric and linear. The counterion environment, however, greatly influences the equilibrium structure. A symmetric, flexible counterion environment produces only a slightly altered symmetric, linear equilibrium structure for the I 3 − anion ; whereas an asymmetric, rigid counterion frame leads to unequal bond lengths and bending of the anion. For the I 5 − anion, the potential energy calculated for bending of the central I—I—I angle, α, is very small so that a slight interaction with the lattice will readily lead to the experimental bent structure with α = 94°. At this value of α, the two outer angles are found to be equal and close to the experimental value. The electronic structures of the I 3 − and I 5 − ions are also discussed.

Rotational Barrier and Electronic Structure of Monomethylphosphine from Ab Initio LCAO–MO–SCF Calculations

Ab initio LCAO–MO–SCF calculations have been carried out on three conformations of monomethyl‐phosphine, CH3PH2, using an uncontracted, moderate‐sized Gaussian basis set, with and without d orbitals. The relative stability of the three forms (staggered, semieclipsed, and eclipsed) was studied. The effect of allowing d character to the phosphorus atom in this molecule was investigated in detail, with special reference to the carbon–phosphorus bond. Three‐dimensional electron‐density plots are presented and discussed in this context.

Role of sulfur d orbitals in isothiocyanic acid and an electronic‐structure comparison with isocyanic acid

The electronic structure of isothiocyanic acid, HNCS, was determined by an ab initio LCAO‐MO‐SCF calculation. Isocyanic acid, HNCO, was also calculated and the two molecules are compared. The basis set consisted of uncontracted, atom‐optimized Gaussian functions, with nine s plus five p exponents to describe the sulfur and five s plus two p apiece for the carbon, nitrogen, and oxygen. The hydrogen was represented by three s‐type functions. An additional calculation of HNCS was carried out by adding a d set to the description of the sulfur. The molecular orbitals of HNCS are analyzed using simple perturbation theory applied to the localized valence‐bond functions and the SCF orbitals are displayed using planar electron‐density plots. The electronic structure of HNCS is compared with that of HNCO and it is concluded that (1) the π system in HNCS involves a nitrogen lone pair stabilized by a higher‐lying C–S π bond while the π system of HNCO consists of a C–O π bond stabilized by the higher‐energy nitrogen l...

Carbodiimide-intermediated esterification of the inorganic phosphates and the effect of tertiary amine base☆

Detailed analysis of appropriate 31P nuclear-magnetic-resonance spectra shows that under the usual laboratory conditions, carbodiimide-induced condensation of orthophosphoric acid in a number of solvents leads to condensation only slightly beyond the metaphosphate composition in the presence of strong tertiary amines; whereas in the absence of amine, the condensation proceeds into the ultraphosphate region about halfway between the metaphosphate and phosphoric anhydride compositions. With amine, the principal product consists of the cyclic trimetaphosphate anion, with one of the nonbridging oxygen atoms substituted by the urea resulting from hydration of the carboiimide, i.e., (O2-) P-O-P(O2-) -O-P(O) [N[CH(CH3)2] see article [C(O)NHCH(CH3)2]] for the condensation with diisopropylcarbodiimide. Without amine, the major product is the 1,5-mu-oxotetrametaphosphate anion see article. The well-known carbodiimide-mediated phosphorylation of alcohols with orthophosphoric acid is shown to be directly attributable to the high reactivity of the phosphate branch groups of the carbodiimide-generated ultraphosphates.

31P nuclear magnetic resonance studies of phosphate condensations with trichloroacetonitrile. I: Condensation to the meta composition

Abstract The condensation of phosphoric acids by trichloroacetonitrile in tetramethylurea, with or without tri- n -butylamine, was studied in considerable detail, using 31 P and 1 H nuclear magnetic resonance for quantitatively analyzing the various compositions studied as a function of time. It was found that the only phosphate species observed in the condensation process were the ortho-, pyro-, tripoly-, tetrapoly-, and trimetaphosphates, with occasional small amounts of the two larger cyclic species tetrameta- and hexametaphosphate. The results could be accounted for adequately in terms of the following reactions: (1) two orthophosphates condensing to a pyrophosphate, (2) ortho and pyro to tripoly, (3) ortho and tripoly to tetrapoly, (4) two pyros to tetrapoly, (5) cyclization of tripoly to trimeta, and (6) cyclization of tetrapoly to trimeta plus ortho. With a stoichiometric amount (or an excess) of trichloroacetonitrile, the condensation of any of these phosphates occurs in a stepwise manner to give all of the phosphorus in the form of the cyclic trimetaphosphate. The amine acts primarily to speed up the reaction, by approximately 10,000-fold, when going from no amine to 1 equivalent of amine for each equivalent of phosphorus. When the mole ratio of amine to phosphorus was increased beyond unity, however, the added amine did not cause the reactions to accelerate. When the concentrations of the various phosphate species were plotted as a function of time in the condensation of orthophosphoric acid, it was found that pyrophosphate was first produced at a high rate with the concentration going through a maximum. This concentration maximum was followed by another for the tripolyphosphate, and then by yet another for the tetrapolyphosphate. Throughout the course of the reaction the trimetaphosphate was found to increase continuously. Detailed kinetic studies were carried out on these rate curves according to the mechanism of Cramer and Weimann [ Chem. Bio. 94, 966 (1961)]. The condensations are interpreted as involving chainphosphate terminal groupings (including the orthophosphate) reacting with the trichloroacetonitrile. At the conclusion of the condensations, a much slower reaction sets in. This slower reaction involves the exchange of parts between the various phosphate species to approach a redistribution equilibrium.