A. D. McLean

Contracted Gaussian basis sets for molecular calculations. I. Second row atoms, Z=11–18

Contracted Gaussian basis sets for molecular calculations are derived from uncontracted (12,8) and (12,9) sets for the neutral second row atoms, Z=11–18, and for the negative ions P−, S−, and Cl−. Calculations on Na...2p63p, 2P and Mg...2p63s3p, 3P are used to derive contracted Gaussian functions to describe the 3p orbital in these atoms, necessary in molecular applications. The derived basis sets range from minimal, through double‐zeta, to the largest set which has a triple‐zeta basis for the 3p orbital, double‐zeta for the remaining. Where necessary to avoid unacceptable energy losses in atomic wave functions expanded in the contracted Gaussians, a given uncontracted Gaussian function is used in two contracted functions. These tabulations provide a hierarchy of basis sets to be used in designing a convergent sequence of molecular computations, and to establish the reliability of the molecular properties under study.

Accurate calculation of the attractive interaction of two ground state helium atoms

Calculations were performed on the Van der Waals unteraction of two ground state He atoms. At each R, three calculations were done to obtain orbitals. One SCF calculation un the guven one-particle basis, and two atomic CI calculations carried out in one-and two-center basis sets. The effect of intra- atomic correlations was studied. (AIP)

Molecular orbital predictions of the vibrational frequencies of some molecular ions

Recent spectroscopic advances have led to the first determinations of infrared vibration-rotation bands of polyatomic molecular ions. These initial detections were guided by ab initio predictions of the vibrational frequencies. The calculations reported here predict the vibrational frequencies of additional ions which are candidates for laboratory analysis. Vibrational frequencies of neutral molecules computed at three levels of theory, HF/3-21G, HF/6-31G*, and MP2/6-31G*, were compared with experiment and the effect of scaling was investigated to determine how accurately vibrational frequencies could be predicted. For 92% of the frequencies examined, uniformly scaled HF/6-31G* vibrational frequencies were within 100 cm-1 of experiment with a mean absolute error of 49 cm-1. This relatively simple theory thus seems suitable for predicting vibrational frequencies to guide laboratory spectroscopic searches for ions in the infrared. Hence, the frequencies of 30 molecular ions, many with astrochemical significance,were computed. They are CH2+, CH3+, CH5+, NH2+, NH4+, H3O+, H2F+, SiH2+, PH4+, H3S+, H2Cl+, C2H+, classical C2H3+, nonclassical C2H3+, nonclassical C2H5+, HCNH+, H2CNH2+, H3CNH3+, HCO+, HOC+, H2CO+, H2COH+, H3COH2+, H3CFH+, HN2+, HO2+, C3H+, HOCO+, HCS+, and HSiO+.

Theory of Molecular Polarizabilities

Buckinghams theory of the interaction between a polarizable charge distribution and an external electric field is presented and extended in a unified way. Transformations of components of the molecular polarizability tensors under change of the coordinate origin are derived. Relationships between tensor components in systems of axial and spherical symmetry are given.

Relativistic effects on Re and De in AgH and AuH from all‐electron Dirac–Hartree–Fock calculations

All‐electron relativistic self‐consistent‐field calculations on AgH and AuH show bond length concentrations of 0.08 and 0.25 A and dissociation energy increases of 0.08 and 0.42 eV from nonrelativistic values. With correlation estimates from nonrelativistic calculations, the experimental Re and De of AgH are completely accounted for. In AuH, this is not the case, and we conclude that relativistic correlation effects are markedly different from nonrelativistic in this molecule. Comparison of effective core potential calculations on AuH with the present all‐electron calculations shows good agreements for De’s but significant discrepancies for Re’s.

Classification of configurations and the determination of interacting and noninteracting spaces in configuration interaction

The n‐particle space of a configuration interaction (CI) calculation, spanned by configuration state functions (CSFs), is partitioned into subspaces defined, relative to a set of zeroth‐order CSFs, by the type of orbital excitation and by the lowest order of the Rayleigh‐Schrodinger perturbative correction contributed by the subspace. A method is given for determining CSFs spanning the ith‐order subspace which contains those and only those functions that have a nonzero interaction through the Hamiltonian with some member of the (i−1)th‐order subspace. These subspaces are anticipated to be of considerable use in structuring and interpreting configuration interaction wavefunctions.

The interacting correlated fragments model for weak interactions, basis set superposition error, and the helium dimer potential

We report the LM‐2 helium dimer interaction potential, from helium separations of 1.6 A to dissociation, obtained by careful convergence studies with respect to configuration space, through a sequence of interacting correlated fragment (ICF) wave functions, and with respect to the primitive Slater‐type basis used for orbital expansion. Parameters of the LM‐2 potential are re=2.969 A, rm=2.642 A, and De=10.94 K, in near complete agreement with those of the best experimental potential of Aziz, McCourt, and Wong [Mol. Phys. 61, 1487 (1987)], which are re=2.963 A, rm=2.637 A, and De=10.95 K. The computationally estimated accuracy of each point on the potential is given; at re it is 0.03 K. Extrapolation procedures used to produce the LM‐2 potential make use of the orbital basis inconsistency (OBI) and configuration base inconsistency (CBI) adjustments to separated fragment energies when computing the interaction energy. These components of basis set superposition error (BSSE) are given a full discussion.

Symmetry breaking in molecular calculations and the reliable prediction of equilibrium geometries. The formyloxyl radical as an example

A systematic approach to symmetry breaking in molecular calculations, based on MCSCF and multireference CI (MRCI) wave functions, is presented. A series of MCSCF expansions is generated by successively incorporating resonance effects and size effects into the wave functions. The character of the potential surface obtained at each level is analyzed. As an example, the potential energy curves of the ground state (σ) and the first excited state (π) of the formyloxyl radical (HCO2) are characterized. The σ and π equilibrium structures are shown to be symmetric, with an adiabatic σ−π excitation energy of 9.2 kcal/mol. Unlike earlier theoretical studies, our MCSCF model produces a qualitatively correct potential surface. Therefore, we are able to extract reliable vibrational frequencies from the MRCI potential surface.

Computed Ground‐State Properties of FH and CH

The accuracy of existing comprehensive calculations of wavefunctions of first‐ and second‐row hydrides in close to the Hartree—Fock approximation is confirmed by performing test calculations, with larger Slater‐function basis sets, on FH and ClH.Dipole moments, quadrupole moments, magnetic susceptibilities, rotational g factors, forces on the nuclei, field gradients at the nuclei, and polarizabilities are then discussed and estimated from either computation alone, or by combining computed expectation values with observed properties. Excellent agreement between computation and observation is demonstrated for the quadrupole coupling constant for 35Cl and 37Cl nuclei in ClH. The following estimates of molecular properties not available elsewhere are made: rotational g factor is −0.033 for 19FH and +0.006 for 35ClH; diamagnetic anisotropy (ξ∥−ξ⊥) is −2.1 × 10−6 erg/G2·mole for FH, and −3.3 × 10−6 erg/G2·mole for ClH; eqQ(2H) = 0.34 Mc/sec in FH and 0.18 Mc/sec in ClH. For FH the axial components of the molecu...

Improved collisional excitation rates for interstellar water.

Theoretical rate constants among the lowest 45 para and 45 ortho rotational levels of water in collisions with He atoms have been calculated for temperatures between 20 and 2000 K using a recently improved theoretical interaction potential. These values are about 30%-40% larger than those reported previously (Palma et al. 1988b) but relative sizes of different state-to-state rates have not changed significantly. Successive improvements to the theoretical description of this system now appear to have converged.