Iterative natural orbitals for configuration interaction using perturbation theory

Approximate natural orbitals are determined iteratively from CI expansions constructed using first-order perturbation theory in order to investigate the possibility of eliminating the complete transformation of MO integrals on each iteration. Results on LiH and H₂O are compared with fully variationally determined NO's to assess questions of convergence.     

Basis set dependence of localized orbitals

The effects of Gaussian basis set contraction and addition of polarization functions on H₂O localized orbitals have been studied at the experimental geometry. It is shown that the electric moments and moment features of localized orbitals are not influenced very much by basis set quality variations, as going from medium size to enlarged basis sets. The difference between bond pair and lone pair charge densities was found to be larger on approaching the Hartree-Fock limit. A minimal basis set, however, does not suitably characterize the localized charge distributions.     

Transferability of some properties of localized molecular orbitals

The transferability of the first and second order electric moments of localized orbitals have been shown. Some moment characteristics have also been investigated. The standard deviation of first order localized moments is less than 2.0%. The standard deviation of second order electric moment components are larger: although they do not exceed 4.0% for bond orbitals, for lone pair ones they vary about 1.5–10.1%.     

Transferability of some properties of localized molecular orbitals

The localized molecular orbitals of some related ten- and eighteen-electron systems have been studied. The transferability of the kinetic, self-interaction, Coulomb and exchange interaction energies on localized orbitals have been shown. The standard deviation of the kinetic and of the interaction energies (including exchange) are less than 2.5% except for lone pair orbitals of the oxygen atoms where the standard deviation is close to 4%.     

Molecular orbital calculations based on linear combinations of fragment orbitals

A semiempirical MO method based on localized fragment orbitals has been developed, which is particularly suited for the construction of orbital correlation diagrams for the discussion of the electronic structure of complex molecules in terms of fragments and their interactions. The method allows for the inclusion of experimental ionization potentials and electron affinities of the fragments within the calculation of the Fock matrix elements and may thus form the basis of an interpretation of photoelectron spectra, comparable to the interpretation of UV spectra by means of the MIM method of Longuet-Higgins and Murrell. Several levels of approximation are discussed using the acrolein molecule as an example.     

Moment characterization of localized orbitals

The parallel between orbital first and second electric moments and statistical first and second central moments is noted. Three measures of orbital spatial distribution in terms of their moments are proposed, and applied to the LMO's in a series of ten-electron hydrides. Consistent differences between bond and lone pair distributions are found. Using the statistical interpretation, for each LMO an effective solid angle around the central atom is postulated.     

On the use of pseudopotentials in molecular calculations

A series of tests was performed of the Kahn-Goddard-Melius-Topiol pseudopotentials in view of their utilization with small contracted basis sets in molecular computations. The effects of inner-shell separability and of basis set contraction are underlined. The utilizability of Topiol's valence least-squares fitted Gaussian basis sets is studied.     

Linear combination of hybrid orbitals: Acyclic alkanes

A LCHO SCF-CI semiempirical scheme is formulated for compounds of carbon and hydrogen, and far-UV transition energies computed for a number of acyclic alkanes. The type of spectral bands (CH CH*, etc.) and symmetries of subbands are assigned, and their dependence on conformation and branching discussed. It is argued that, for pentanes and higher alkanes, spectral quantities can be estimated from data on the lower alkanes.Multi-Conformational Compounds, part X. For part IX, see Ref. [1].     

An optimum minimal basis of exponential atomic orbitals

An analysis of some of the energetic properties of the conventional minimal STO basis is used to suggest a new optimum set of exponential functions for use in molecular calculations.     

Configuration interaction methods for improving unrestricted Hartree-Fock spin densities

Limited Configuration Interaction wave functions based on Unrestricted Hartree-Fock natural orbitals are found to be easy to compute and to give much more satisfactory spin densities than are provided by techniques currently in use.     

Delocalization corrections to the strictly localized molecular orbitals: A linearized SCF approximation

An explicit formula is derived for calculating the delocalization corrections (tails) to be added to the strictly localized bond orbitals. It was obtained by solving analytically the SCF problem for the interbond interactions in a linearized approximation. The model calculations at the CNDO/2 level show that this simple approach is sufficient to account for the molecular conformations.     

Bonding criteria for diatomic molecular orbitals and inter-relations among them

Bonding criteria for molecular orbitals in diatomic molecules are discussed. An orbital force criterion is shown to have several conceptual and practical advantages, providing a basis for the investigation of inter-relations among many of the commonly employed criteria.It is found that interconsistency among those criteria is guaranteed, within the framework of Koopmans' Theorem, if the orbital energies are monotonic in the range (R _e, ).The application of the orbital force criterion to the second row homonuclear diatomics exhibits reasonable chemical trends concerning the valence-shell orbitals, as well as indications of a slightly antibonding nature of the inner orbitals.Based on a section of a thesis to be submitted by Y.T. to the Senate of the Technion-Israel Institute of Technology, in partial fulfilment of the requirements for the D.Sc. degree     

Calculations of the electronic structures of cage molecules using free-electron orbitals as a basis

A non-empirical molecular orbital method, particularly suitable for calculations on cage-like molecules, is described. The method uses as basis functions the set of free-electron functions which are the solutions of Schrödinger's equation for an electron confined between two concentric, spherical potential energy barriers. Application of the theory to the SCF calculation of the energies of the delocalized electrons in benzene and tetrasulphur tetranitride shows that the model is capable of interpreting the properties of such systems. However, it does highlight a difficulty in the calculation of excited state energies with one-centre models which appears to be largely unrecognized.Extension of the method to a consideration of all the valence electrons, using P₄ as an example, reveals problems the origin of which is an inadequate treatment of the core electrons. It is suggested that these problems may best be dealt with by use of a suitable pseudo potential.     

Band structure of nonclassical polymers

It is shown that fully conjugated alternant nonclassical polymers are comparatively stable systems as a result of the considerable delocalisation energy.The energy characteristics of the polymers are strongly determined by the electron spin distribution of the degenerate nonbonding molecular orbitals; the full spin configuration,S>0, is favoured. The spin densities depend on the electron correlation and alternate; this corresponds to a ferrimagnetic state of the polymer at 0°K.     

A non-empirical molecular orbital study on the relative stabilities of adenine and guanine tautomers

The relative stabilities of a series of adenine and guanine tautomers have been calculated using anab initio Hartree-Fock-Roothaan SCF MO method. The calculated relative stabilities agree in general with the results of earlier semiempirical studies. According to the present study, tautomeric forms with regular Kekulé structure for the six-membered purine ring are the most stable. The amine-imine tautomerization of purine bases is not likely to be responsible for spontaneous mutations in DNA.     

Band structure of even and ions of odd polyenes

The analysis of experimental data for singlet transitions (E _n) of even polyenes (I), cations (II) and anions (III) of odd polyenes show that for infinite chains E (I)/E (II)=E (I)/E (III) = 2:1. It is shown that the energy gap is equal for the three systems. In cases (II) and (III) there is a level (NBMO) in the gap which is vacant in (II) and occupied in (III). That is why the first optical transition in (II) and (III) depends on the semiwidth of the gap.     

A theoretical study of electrophilic reactions

The electronic structures of protonated formyl and acetylium cations and their deprotonation paths leading to HCO⁺, COH⁺ and CH₃CO⁺have been studied by means of ab initio calculations. The results support Olah's theory that dipositive species can be the de facto reagents in electrophilic reactions.     

On the configuration interaction method for singlet states

The direct CI method, which avoids explicit calculation of the Hamiltonian matrix, is presented in a new form. The method is linked with Davidson's algorithm for iterative evaluation of the ground state eigenvector. The viability of the method is indicated by the test calculations on water which are described.     

Theory of isomer shift in hemin

Using the self-consistent charge extended Hückel procedure, the charge density difference at Fe⁵⁷ nucleus, between hemin and Fe⁺³ ion is calculated. This is combined with the recent value of the calibration constant, –0.23±0.02a ₀ ³ mm/sec to obtain an isomer shift of –0.374 mm/sec between hemin and Fe⁺³ in good agreement with the value –0.392 mm/sec derived from experimental data and the calculated value of the isomer shift of Fe⁺³ with respect to K₃FeF₆ from first principle covalency investigations in the latter compound. is composed of contributions from core and valence electrons of the same order of magnitude, with the latter being more than one-half of the former. The core contribution is composed of a number of terms of comparable magnitude and differing signs, whose significance is discussed.Supported by grant HL 15196-02 from the Heart and Lung Institute of National Institute of Health.     

An examination of UHF methods for atoms in the first three rows

The reliability and usefulness of UHF methods of calculating hyperfine parameters has been examined. Comparisons with CI and SEHF methods are made.A detailed analysis of OPHF calculations for first and second row atoms has shown that various one-electron properties are accurately described by physically reasonable functions of the atomic number Z. In addition there is a strong correlation between these properties. This has led to a method whereby UHF spin densities across a row of atoms can be obtained from UHF calculations of only two atoms in that row.A strong correlation between experimental and UHF spin densities is shown to exist for atoms of the first three rows. This is used to predict experimental spin densities for atoms in these rows which have not yet been measured experimentally.