A series of correlation consistent basis sets have been developed for the post-d group 16-18 elements in conjunction with small-core relativistic pseudopotentials of the energy-consistent variety. The latter were adjusted to multiconfiguration Dirac-Hartree-Fock data based on the Dirac-Coulomb-Breit Hamiltonian. The outer-core (nϪ1)spd shells are explicitly treated together with the nsp valence shell with these PPs. The accompanying cc-pVnZ-PP and aug-cc-pVnZ-PP basis sets range in size from DZ to 5Z quality and yield systematic convergence of both Hartree-Fock and correlated total energies. In addition to the calculation of atomic electron affinities and dipole polarizabilities of the rare gas atoms, numerous molecular benchmark calculations ͑HBr, HI, HAt, Br 2 , I 2 , At 2 , SiSe, SiTe, SiPo, KrH ϩ , XeH ϩ , and RnH ϩ) are also reported at the coupled cluster level of theory. For the purposes of comparison, all-electron calculations using the Douglas-Kroll-Hess Hamiltonian have also been carried out for the halogen-containing molecules using basis sets of 5Z quality.

Scalar-relativistic pseudopotentials and corresponding spin-orbit potentials of the energy-consistent variety have been adjusted for the simulation of the [Ar]3d(10) cores of the 4d transition metal elements Y-Pd. These potentials have been determined in a one-step procedure using numerical two-component calculations so as to reproduce atomic valence spectra from four-component all-electron calculations. The latter have been performed at the multi-configuration Dirac-Hartree-Fock level, using the Dirac-Coulomb Hamiltonian and perturbatively including the Breit interaction. The derived pseudopotentials reproduce the all-electron reference data with an average accuracy of 0.03 eV for configurational averages over nonrelativistic orbital configurations and 0.1 eV for individual relativistic states. Basis sets following a correlation consistent prescription have also been developed to accompany the new pseudopotentials. These range in size from cc-pVDZ-PP to cc-pV5Z-PP and also include sets for 4s4p correlation (cc-pwCVDZ-PP through cc-pwCV5Z-PP), as well as those with extra diffuse functions (aug-cc-pVDZ-PP, etc.). In order to accurately assess the impact of the pseudopotential approximation, all-electron basis sets of triple-zeta quality have also been developed using the Douglas-Kroll-Hess Hamiltonian (cc-pVTZ-DK, cc-pwCVTZ-DK, and aug-cc-pVTZ-DK). Benchmark calculations of atomic ionization potentials and 4d(m-2)5s(2)-->4d(m-1)5s(1) electronic excitation energies are reported at the coupled cluster level of theory with extrapolations to the complete basis set limit.

No abstract

New relativistic energy-consistent pseudopotentials have been generated for the 5d transition metals Hf-Pt. The adjustment was done in numerical two-component multiconfiguration Hartree-Fock calculations, using atomic valence-energy spectra from four-component multiconfiguration Dirac-Hartree-Fock calculations as reference data. The resulting two-component pseudopotentials replace the [Kr]4d(10)4f(14) cores of the 5d transition metals and can easily be split into a scalar-relativistic and a spin-orbit part. They reproduce the all-electron reference energy data with deviations of approximately 0.01 eV for configurational averages and approximately 0.05 eV for individual relativistic states. Full series of correlation consistent basis sets from double to quintuple-zeta have also been developed in this work for use with the new pseudopotentials. In addition, all-electron triple-zeta quality correlation consistent basis sets are also reported in order to provide calibration for the pseudopotential treatment. The accuracy of both the pseudopotentials and basis sets are assessed in extensive coupled cluster benchmark calculations of atomic ionization potentials, electron affinities, and selected excitation energies of all the 5d metal atoms, including the effects of spin-orbit coupling.

A coupled cluster composite approach has been used to accurately determine the spectroscopic constants, bond dissociation energies, and heats of formation for the X1(2)II(3/2) states of the halogen oxides ClO, BrO, and IO, as well as their negative ions ClO-, BrO-, and IO-. After determining the frozen core, complete basis set (CBS) limit CCSD(T) values, corrections were added for core-valence correlation, relativistic effects (scalar and spin-orbit), the pseudopotential approximation (BrO and IO), iterative connected triple excitations (CCSDT), and iterative quadruples (CCSDTQ). The final ab initio equilibrium bond lengths and harmonic frequencies for ClO and BrO differ from their accurate experimental values by an average of just 0.0005 A and 0.8 cm-1, respectively. The bond length of IO is overestimated by 0.0047 A, presumably due to an underestimation of molecular spin-orbit coupling effects. Spectroscopic constants for the spin-orbit excited X2(2)III(1/2) states are also reported for each species. The predicted bond lengths and harmonic frequencies for the closed-shell anions are expected to be accurate to within about 0.001 A and 2 cm-1, respectively. The dissociation energies of the radicals have been determined by both direct calculation and through use of negative ion thermochemical cycles, which made use of a small amount of accurate experimental data. The resulting values of D0, 63.5, 55.8, and 54.2 kcal/mol for ClO, BrO, and IO, respectively, are the most accurate ab initio values to date, and those for ClO and BrO differ from their experimental values by just 0.1 kcal/mol. These dissociation energies lead to heats of formation, DeltaH(f) (298 K), of 24.2 +/- 0.3, 29.6 +/- 0.4, and 29.9 +/- 0.6 kcal/mol for ClO, BrO, and IO, respectively. Also, the final calculated electron affinities are all within 0.2 kcal/mol of their experimental values. Improved pseudopotential parameters for the iodine atom are also reported, together with revised correlation consistent basis sets for this atom.