Library

Notes: A new version of solutions in the form of an exponentially weighted power series is constructed for the two-dimensional circularly symmetric quartic oscillators, which reflects successfully the desired properties of the exact wave function. The regular series part is shown to be the solution of a transformed equation. The transformed equation is applicable to the one-dimensional problem as well. Moreover, the exact closed-form eigenfunctions of the harmonic oscillator can be reproduced as a special case of the present wave function. © 1996 John Wiley & Sons, Inc.

Notes: The wave function of a system may be expanded in terms of eigenfunctions of the N -1 electron Hamiltonian times one-particle functions known as generalized overlap amplitudes (GOAS). The one-electron operator whose eigenfunctions are the GOAS is presented, without using an energy-dependent term as in the one-particle Green function or propagator approach. It is shown that this operator and the extended Koopmans' theorem (EKT) one-electron operator are of similar form, but perform complementary roles. The GOA operator begins with one-electron densities and total energies of N -1 electron states to generate the two-matrix and total energy of an N-electron state. The EKT operator begins with the two-matrix of an N-electron state to generate one-electron densities and ionization potentials (or approximations thereto) for N -1 electron states. However, whereas the EKT orbitals must be linearly independent, no such restriction applies to the GOAS. © 1996 John Wiley & Sons, Inc.

Notes: Interactions of DNA bases frequently involve the DNA base amino groups. In contrast to the empirical force fields, the ab initio calculations predict nonplanar DNA base amino groups. The same conclusion also follows from the density functional theory (DFT) calculations. Both local and nonlocal density approximations were used. Optimized geometries of two other molecules with nonplanar amino groups (aniline, formamidine) are presented for comparison. The influence of nonplanar DNA base amino groups on the conformational variability of DNA is discussed. © 1996 John Wiley & Sons, Inc.

Notes: The conformational behavior and the stability of thiazolium and thiamin diphosphate (ThDP) adducts formed by the C2 addition of the substrates pyruvate and glyoxylate to the corresponding thiamin systems are investigated within the force-field version PIMM90 as well as the semiempirical AM1 and PM3 methods. Moreover, the reaction coordinate of the decarboxylation process of the adducts with respect to the C2α(SINGLEBOND)COO- bond are calculated by PM3 and AM1. The calculations on the key intermediates of the Breslow mechanism are performed in order to study the steric aspects of both substrate adducts that show a different pathway in the catalytic cycle. The alternative structural findings for the decarboxylation products are compared with first 6-31C* studies on the corresponding thiazolium model systems. Especially, the PM3 calculations show that the elimination of CO2 is favored if the arrangement of the carboxylate group is nearly perpendicular to the plane of the thiazolium ring. These results support the least-motion maximum-overlap mechanism in the enzymatic decarboxylation reaction, proposed by Kluger. The most stable conformers of the ThDP adducts and its decarboxylation products are characterized by V-like structures and the formation of a significant intramolecular hydrogen bonding under participation of the 4′-aminopyrimidine ring. © 1996 John Wiley & Sons, Inc.

Notes: Ab initio calculations were performed on the tautomers and conformers of N-hydroxyurea using a 6-31G** basis set. The minimum-energy structures have been found and the importance of the intramolecular hydrogen bond as the stabilizing factor was pointed out. © 1996 John Wiley & Sons, Inc.

Notes: Many biological processes are characterized by an essentially quantum dynamical event, such as electron or proton transfer, in a complex classical environment. To treat such processes properly by computer simulation, allowing nonadiabatic transitions involving excited states, we recently developed a density matrix evolution (DME) method [H. J. C. Berendsen and J. Mavri, J. Phys. Chem, 97, 13464 (1993)] which simulates the dynamics of quantum systems embedded in a classical environment. The formalism of the method is presented and an overview of the applications ranging from collisions of a quantum harmonic oscillator with noble gas atoms to proton tunneling in a double-well hydrogen bond is given. The methodology for treatment of proton-transfer processes with inclusion of excited states is presented. Future application of the method on biologically interesting processes, such as proton transfer in enzymatic reactions, is discussed. © 1996 John Wiley & Sons, Inc.

Notes: For theoretical reasons, and on account of the development of a new interpolation technique, it is useful and important to examine the asymptotic behavior of the solution to the one-dimensional Hubbard model. In this article, it is shown how perturbative expansions for the energy can be developed in the asymptotic region of the relevant coupling for all the relevant excited states of cyclic polyene rings. © 1996 John Wiley & Sons, Inc.

Notes: The simple BSSE-free SCF method (CHA/F) introduced in the previous parts of this series is extended to the case of three subsystems, which may be either three weakly interacting molecules or a bimolecular system described by using bond functions. The CHA/F formalism is formulated in a more transparent manner, indicating also a straightforward way for generalization to the case of an arbitrary number of subsystems. The illustrative calculations show the viability of using the CHA/F scheme for three-component systems. © 1996 John Wiley & Sons, Inc.

Notes: One is often led, in quantum mechanics, to a perturbative solution of an eigenvalue problem, which is defined by a given Hamiltonian. The perturbative series for the energy which results will be a function of a coupling constant which appears in the Hamiltonian. In this article, the perturbative series for the energy of a state of a cyclic polyene ring which are valid for the small and large coupling limit of the model are used to construct algebraic functions. These algebraic functions are defined in terms of polynomials which are given as a function of the energy variable and coupling parameter and can be solved to give the energy as a function of coupling. It is found that relatively small polynomials give very good agreement with the exact values and that the accuracy of the results increases rapidly as the degree of the polynomial increases. The final goal of this and subsequent articles is to study energy levels in PPP models of planar conjugated hydrocarbons. In this article, we test an interpolant technique on the case of the one-dimensional Hubbard model, where an exact solution can be obtained by solving a system of nonlinear equations. In the case of the Hubbard model, the correlation effects are overestimated. Therefore, if the technique works for the Hubbard model, it is reasonable to assume that the technique would work even better for the PPP model. © 1996 John Wiley & Sons, Inc.

Notes: The Ising, small-bipolaron (ISB) theory is a strong-coupling theory of cuprate superconductivity which is based on the negative-U, Hubbard Hamiltonian. Its ground state is composed of (small) bipolarons and (small-bipolaron) holes with a vibronically induced, bipolaron-hole exchange interaction, JBH, between them. The energy gap, Δ(0), is taken to be equal to the dissociation energy of a small bipolaron and which, since it is defined spectroscopically, is not an order parameter. The application of the Ising mean-field theory to the highly degenerate ground-state yields a second-order phase change with kTC/2 = JBH and a real order parameter, Ω(T), which is valid over the entire temperature range from zero to TC. Near TC, the Ising free-energy functional takes the same form as does the Landau. In the presence of an electromagnetic field, the Ising functional is a generalization of the Ginzburg-Landau functional which employs a complex order parameter and which is invariant under the electromagnetic gauge transformation. The breaking of the gauge invariance yields the London theory of superconductivity. © 1996 John Wiley & Sons, Inc.

Notes: The solvent shift of the π* ← n transition of acetone in water, acetonitrile, and tetrachloromethane was calculated in a combined quantum mechanical - classical mechanical approach, using both dielectric continuum and explicit, polarizable molecular solvent models. The explicit modeling of solvent polarizability allows for a separate analysis of electrostatic, induction, and dispersion contributions to the shifts. The calculations confirm the qualitative theories about the mechanisms behind the blue shift in polar solvents and the red shift in nonpolar solvents, the solvation of the ground state due to electrostatic interactions being preferential in the former, and favorable dispersion interaction with the excited state, in the latter case. Good quantitative agreement for the solvent shift between experiment (+1,700, +400, and -350 cm-1 in water, acetonitrile, and tetrachloromethane, respectively) and the explicit solvent model (+1,821, +922, and -381 cm-1) was reached through a modest Monte Carlo sampling of the solvent degrees of freedom. A consistent treatment of the solvent could only be realized in the molecular solvent model. The dielectric-only model needs reparameterization for each solvent. © 1996 John Wiley & Sons, Inc.

Notes: In density functional theory (DFT), a many-electron problem for the electron density in atoms may be reduced, according to the Kohn-Sham scheme, to a one-electron problem. In the present work, a variational model is proposed which leads, within some assumptions, to the set of equations describing the change of the electron density ρ and energy ε during the ionization process. It is shown that the one-electron density contributions are not necessarily spherically symmetric, but assume the symmetry which depends upon the symmetry of the positive field. A few nonspherically symmetric potentials are studied in the present article. The nonlinear differential equation for density r is formulated and solved for Coulombic, Fues-Kratzer, and Hartmann potentials. The solutions and some particular examples are presented. © 1996 John Wiley & Sons, Inc.

Notes: The cycloaddition reaction of FCH(double bond)C(double bond)O and NH2CH(double bond)NH leading to 2-azetidinone was studied theoretically at the level of RHF/6-31G and RHF/6-31G*. Two possible mechanisms via a gauche or trans intermediate were compared. The obtained results show that the reaction proceeds in a gauche manner much easier than in a trans one. © 1996 John Wiley & Sons, Inc.

Notes: The role of the asymptotic behavior of approximating sequences of electron densities ρn(r) in the calculation of one-electron properties is studied. Rigorous mathematical results in the frame of Hilbert spaces are used to prove the following facts: (i) Both the L2 convergence of wave functions ψn and the E convergence of the corresponding energies En guarantee the correctness of the limiting procedure limn→x ∫Ω s((overline)x(/overline)|ψn|2 d(overline)x(/overline) = ∫Ω s((overline)x(/overline))|ψ|2 d(overline)x(/overline) for the most frequently used operators s(x), Ω being any bounded region of the n-particle configuration space R3N; and (ii) the uniform boundedness of the sequence {ρn} together with both the L2 and E convergencies is sufficient to guarantee the correctness of the limiting procedure limn→x ∫∞0 s(r)ρnr2dr = ∫x0 s(r)ρr2 dr for most one-electron operators s(r) including the power moment operators rk which, for large k, are representative of the class of operators not relatively form-bounded by the Hamiltonian. The mathematical concept of uniform boundedness is used to give a characterization of the capability of {ρn} to reproduce the asymptotic behavior of the true electron density ρ and it is shown by means of numerical examples how a sequence {ρn} that does not reproduce the correct asymptotic behavior is not uniformly bounded and can give divergent expectation values of one-electron operators s(r) not relatively form-bounded by the Hamiltonian. © 1996 John Wiley & Sons, Inc.

Notes: We performed electronic ab initio calculations of ion clusters simulating a piece of ionic crystal. Our main interest here is to estimate the (direct) energy band gap from the cluster energy levels. The model consists of a central cation surrounded by four shells of ions, with additional point charges embedding the cluster. We present results for the series of alkali halides with rock-salt structure. © 1996 John Wiley & Sons, Inc.

Notes: We report on an isolated cluster approach to determine local electronic structures of TiO2 surfaces before and after formation of intrinsic defects, i.e., oxygen vacancies, at different crystallographic sites. In particular, isolated oxygen vacancies at bridging sites, isolated oxygen vacancies at in-plane sites and aggregated oxygen vacancies at bridging sites have been treated which lead to changes in the coordination of the adjacent Ti atoms. We find that electronic band gap states are only formed in the presence of fourfold coordinated Ti surface atoms. © 1996 John Wiley & Sons, Inc.

Notes: The quantum dynamics of prototypical three-level metastable system under external perturbations, both time-independent and time-dependent, are studied numerically as well as analytically. The regular Rabi oscillations generated in the system under a monochromatic perturbation are shown to develop additional features when the system has a metastable state. The autocorrelation function of ψ(t) reveals nondecaying, almost quasiperiodic, behavior in such systems. The Rabi oscillations turn more complex when a bichromatic perturbation with two incommensurate frequencies is allowed to interact with the three-level system, irrespective of the presence or absence of a metastable state. A rapid decay in the autocorrelation function of the wavefunction [ψ(t)] is observed in such cases. © 1996 Wiley & Sons, Inc.

Notes: The present knowledge of the monotonicity properties of the spherically averaged electron density ρ(r) and its derivatives, which comes mostly from Roothan-Hartree-Fock calculations, is reviewed and extended to all Hartree-Fock ground-state atoms from hydrogen (Z = 1) to uranium (Z = 92). In looking for electron functions with universal (i.e., valid in the whole periodic table) monotonicity properties, it is found that there exist positive values of α so that the function go(r; α) = ρ(r)/rα is convex, and g1(r;α) = -ρ′(r)/rα is not only monotonically decreasing from the origin but also convex. This is, however, not the case for the function g2(r; α) = ρ′(r)/rα. Additionally, the conditions which specify values for β such that the function gn(r; β) = (-1) ′ρ(n)(r)/rβ is logarithmically convex are obtained and numerically calculated for n = 0,1 in all neutral atoms below uranium. The last property is used to obtain inequalities of general validity involving three radial expectation values which generalize all the similar ones known to date, as well as other relationships among these quantities and the values of the electron density and its derivatives at the nucleus. © 1996 John Wiley & Sons, Inc.

Notes: Pekeris' solution for vibrational-rotational levels of the Morse potential includes the unphysical domain - ∞ 〈 r 〈 0 of the internuclear separation, just like Morse's solution for vibrational levels. In this article, the method of confinement in a box is implemented to obtain the vibrational-rotational J = 1 levels of Morse potentials in the physical domain 0 〈 r 〈 ∞, as well as the corresponding Franck-Condon factors. Applications to Li2 and AIO are presented as illustrations. © 1996 John Wiley & Sons, Inc.

Notes: Potential energy curves are calculated for O2-, O2, and O2+ at the CCSD, QCISD, CCSD(T), and QCISD(T) levels of theory using aug-cc-pVDZ and aug-cc-pVTZ basis sets with electron correlation built onto inversion symmetry constrained and relaxed UHF wave functions. The spectroscopic constant re, we, we, xe, Dj, and αe, are determined from the potential curves using standard second-order perturbation theory expressions and are compared with experimental values to assess the relative accuracy of the theoretical approaches. Comparison of corresponding symmetry-constrained and symmetry-relaxed calculations indicates that the CCSD method is generally superior to CCSD(T), QCISD, and QCISD(T) in recovering from a symmetry-broken reference function. © 1996 John Wiley & Sons, Inc.

Notes: Ab initio and density functional theory (DFT) methods have been applied to generated geometries of chemical systems built only from nitrogen and oxygen. The results of hybrid, local, and nonlocal DFT methods were compared with ab initio HF and MP2 methods. The suitability of DFT methods for modeling nitrogen-oxygen chemical systems is discussed. © 1996 John Wiley & Sons, Inc.

Notes: The alternative decomposition reactions CH2(OH)2 → CH2O + H2O and CH2(OH)2 + H2O → CH2O + 2H2O are investigated using the semiempirical PM3 as well as the ab initio HF/3-21G, HF/6-31G, HF/6-31G**, and MP2/6-31G** calculations. Reactants, products, and appropriate transition states are located on corresponding potential energy surfaces and compared with those reported in earlier studies. © 1996 John Wiley & Sons, Inc.

Notes: The simplest prototypical hydrogen transfer reaction, i.e., Ḣ+ H2 → H2 + Ḣ, is studied by the quantum-mechanical ab initio methods. Results reveal that during this reaction free valence which almost equals the square of the spin density develops on the migrating hydrogen atom. Bond orders are calculated using Mayer's formalism. Both the variations of bond orders and bond lengths along the reaction path are examined. Our analysis reveals that the bond formation and bond cleavage processes in this reaction are not perfectly synchronous. The bond cleavage process is slightly more advanced on the reaction path. © 1996 John Wiley & Sons, Inc.

Notes: A general scheme for theoretical treatment of organometallic reactivity is proposed. It is based upon the notion that the reactivity of a molecule is strongly affected by its coordination to metal-containing fragments. Based upon this idea we describe the large-scale organometallic reactions as reactions of the ligands in the coordination spheres of transition metal complexes. We propose here a quantum mechanical framework for analysis of effects of coordination on the reactivity and give several examples of qualitative energy profiles for reactions in the ligand spheres of transition metal complexes. © 1996 John Wiley & Sons, Inc.

Notes: Silica sodalite is an ideal model system to establish base-line computer requirements of ab initio periodic Hartree-Fock (PHF) calculations of zeolites. In this article, we investigate the effect of various basis sets on the structural and electronic properties of bulk silica sodalite. We also study the interaction of He, Ne, and Ar with the sodalite cage. Our work shows that basis-set superposition errors (BSSE) in calculations using STO-3G and 6-21G(*) basis sets are as large as the interaction energies, leading to poor confidence in the results. To cure this problem, we present high-quality basis sets for Si, O. He, Ne, and Ar, optimized for use with PHF methods, and demonstrate that the new basis set greatly reduces BSSE. The theoretical barriers for transfer of the rare gases between sodalite cages are 5.6, 13.2, and 62.1 kcal/mol for He, Ne, and Ar. © 1996 John Wiley & Sons, Inc.

Notes: A study of π-electron systems confined by impenetrable surfaces is presented. The study results in a nonempirical-based approach to obtain confinement-adapted semiempirical π-Hamiltonians including repulsive terms (PPP or Hubbard). The impenetrable surface confinement of a physical system involves changes in the boundary conditions that the eigenvectors of its differential Hamiltonian operator have to fulfill, while the Hamiltonian itself remains unchanged. However, if this Hamiltonian is written in second quantization language, then confinement only involves changes of the Hamiltonian scalar factors (integrals). Semiempirical Hamiltonian integrals are replaced by parameters; therefore, confinement involves only changes of these parameters. It is shown that confinement changes Coulomb (αi) and exchange (βij), while repulsion (γij) parameters remain unaffected. Next, the influence of confinement upon the electron correlation of (i) π-electron molecular systems, (ii) atoms, and (iii) an electron gas is discussed. The behaviour of the correlation energy vs. the confinement size is found to be different for each type of system. A neat explanation of this variety is given in terms of the Coulomb attractive fields of the systems. Some chemical confinement effects such as an increase in the reactivity of π-electron systems is also outlined. © 1996 John Wiley & Sons, Inc.

Notes: The Cl2CO … Cl2 complex was studied using ab initio post-Hartree-Fock theory at the MP2 and MP4 levels and, for comparison, the DFT method with 6-311G(2d), 6-311 + G(2d), and Sadlej's medium-size polarized (MSPBS) basis sets. A potential energy search recovered a planar minimum-energy structure characterized by a bent conformation. For this weakly bound complex, the interaction energy corrected for the basis set superposition error amounted to - 0.88, - 1.09, - 1.43, and - 0.38 kcal/mol at the MP4(SDTQ)/6-311G(2d), MP4(SDTQ)/6-311 + G(2d), MP4(SDTQ)/MSPBS, and DFT(Becke3LYP)/6-311 + G(2d) levels of theory, respectively. Two highly symmetrical forms, linear and T-shaped, correspond to transition-state conformers. The analysis of harmonic vibrational frequencies and potential energy distribution was performed at the MP2 and DFT levels with the 6-311 + G(2d) basis set. © 1996 John Wiley & Sons, Inc.

Notes: A method of selection of basis functions for the quantum chemical description of atoms and molecules inside spherical and cuboidal cavities is presented. The method consists of two independent steps. First, the appropriate GTO or STO exponents are chosen by fitting the basis functions to cutoff orbitals. The criterion of maximal orbital density inside the cavity is next applied to form the final model orbital space. The method presents the fundamental advantage over many other methods of being trivially applicable in standard quantum chemical progam packages. As an illustration, the method is applied to the hydrogen atom in a spherical box. The use of contracted basis functions is also discussed. © 1996 John Wiley & Sons, Inc.

Notes: Electrostatic and hydrophobic complementarities between chymotrypsin and its inhibitor, avian ovomucoid third domains, were evaluated for eight species, which have different amino acid sequences, using molecular electrostatic potential (MEP) and MEP correlation, and the enzyme-inhibitor interaction was analyzed. The changes in the electrostatic and hydrophobic complementarities caused by the amino acid replacements were reflected clearly in the calculated MEP correlation, and it explained the observed binding association constants correctly. The electrostatic complementarity due to arginine at P′3 strongly promotes the binding process of the inhibitor, while the hydrophobic complementarity in the P1 and P′2 positrons also affects the binding process. It was demonstrated that our method is an effective molecular modeling tool in drug design and protein engineering. © 1996 John Wiley & Sons, Inc.

Notes: The electron localization function (ELF) was calculated for the atoms Li to Sr. The ELF maxima reveal the atomic shell structure for all these atoms. The shells are separated from each other by ELF minima. The integration of the electron density in a shell gives electron numbers. For the valence shell those are in good agreement with the ones expected from the Periodic Table of Elements. © 1996 John Wiley & Sons, Inc.

Notes: The variational method proposed earlier has been generalized, using a trial function of the form: ψ = (Λ0 + ΣsΛsms)ψ0 in which ms = r2p+1Cml, s standing for a triplet (p, l, m) and implemented into the program Hondo-8.4. The second-order density matrices are used to take into account the mono and bi-excited states (DM1 and DM2 matrices, GUGA Algorithm). This allows us to compute the dipole (α), dipole-quadrupole (A) and quadrupole (C) polarizability tensors at the CISD level. The results obtained for a series of test molecules: CO, HF, NH3, and methane with various gaussian basis sets are compared with experimental results (when available for A and C) and those obtained with other theoretical methods. The correlation is found to lower the values of the dipole polarizability which was generally too high when computed by the variational method at the RHF level and the values obtained here are in good agreement with the experimental ones. © 1996 John Wiley & Sons, Inc.

Notes: The problems of semiempirical quantum chemical calculations of (a) spin densities in paramagnetic organometallics, (b) hydrogen bonds, and (c) bond energies and the structure of transition-metal compounds are discussed. Some modifications of the existing semiempirical quantum chemical method are presented. An extended NDDO approximation has been developed. This scheme includes explicit symmetric orthogonalization of the core Hamiltonian and the use of Hellmann's effective core potential for core-electron interaction. © 1996 John Wiley & Sons, Inc.

Notes: A simplified LCAO-DFT-LDA scheme for calculations of structure and electronic structure of large molecules, clusters, and solids is presented. Forces on the atoms are calculated in a semiempirical way considering the electronic states. The small computational effort of this treatment allows one to perform molecular dynamics (MD) simulations of molecules and clusters up to a few hundred atoms as well as corresponding simulations of condensed systems within the Born-Oppenheimer approximation. The accuracy of the method is illustrated by the results of calculations for a series of small molecules and clusters. © 1996 John Wiley & Sons, Inc.

Notes: We describe enhancements to the method of partial retention of diatomic differential overlap (PRDDO). The new method, denoted PRDDO/M, employs a basis set of not quite orthogonal atomic orbitals (NQOAOS) and utilizes sparse matrix techniques to greatly increase the computational efficiency for large molecules. Other modifications, including a complete reparametrization of the method against ab initio STO-3G calculations and implementation of integral screening/damping algorithms, are described. The method is an order of magnitude or more faster than are STO 3G single-point calculations using modern ab initio codes, with little loss in accuracy. © 1996 John Wiley & Sons, Inc.

Notes: No Abstracts.

Notes: The electronic structure and bonding of UF6 and UF6- are studied within a relativistic framework using the MOLFDIR program package. A stronger bonding but more ionic molecule is found if one compares the relativistic with the nonrelativistic results. The first peak in the photoelectron spectrum of Karlsson et al. is assigned to the 12γ8u component of the 4t1u orbital, in agreement with other theoretical and experimental results. Good agreement is found between the experimental and theoretical 5f spectrum UF6-. Some properties, like the dissociation energy and electron affinity, are calculated and the necessity of a fully relativistic framework is shown. The Breit interaction has an effect on the core spinors and the spin-orbit splitting of these spinors but the influence on the valence spectrum is negligible. © 1996 John Wiley & Sons, Inc.

Notes: No Abstracts.

Notes: Quantum chemical and molecular modeling computations on large molecular systems are defined for the computational facilities assumed to be available from now to the next 4 years. We considered a few topics which are requiring much attention. The correlation energy is discussed in some detail and we have presented two new functionals, called the J-functional and the K-functional, which make use of Coulomb or exchange-type integrals. In addition, we report new computational results for the Coulomb-Hole-Hartree-Fock approximation. Very brief summaries on new developments in relativistic Dirac-Fock computation and in density functional theory, on the advantages gained by using different basis sets in the same computation, and on the promises of parallel computing conclude the article. © 1996 John Wiley & Sons, Inc.

Notes: We investigated the dependence of the electrostatic potential on the size and the shape of various cluster models of the MgO(100) surface. Both Mg2+ and O2- adsorption sites have been considered. The clusters were embedded in a large array of point charges to provide a representation of the Madelung potential. We found that the electrostatic potential in the adsorption region shows a marked dependence on the size of the cluster, in particular, for non-stoichiometric clusters where the number of cations and anions differs considerably. These oscillations are due to (a) the different contribution to the electrostatic potential given by a point charge or by an extended ion, and (b) by the polarization of the ions at the cluster border. The effect of the oscillations in the electrostatic potential on the chemisorption properties was investigated for the case of CO2 interacting with surface and defect O2- sites of the MgO surface. © 1996 John Wiley & Sons, Inc.

Notes: Electronic properties of a water molecule embedded in a water droplet are studied in the framework of the generalized self-consistent reaction field approach, using ab initio Hartree-Fock and configuration interaction wave functions. Electrostatic and inductive effects of the surrounding water molecules were calculated with the help of configurations drawn from a classical molecular dynamics simulation. Basis-set effects and solute-solvent interaction operator representation are examined. Embedding energies and liquid-phase multipole moments obtained from the present mixed quantum-classical model are compared with corresponding quantities for purely classical water models. © 1996 John Wiley & Sons, Inc.

Notes: This article describes the present status of the program system ASTERIX and its move toward an efficient implementation on distributed memory computers with 128 processors or more. The ASTERIX program is currently used for scf and ci calculations on what we consider to be truly large systems; polyoxometallates, metallocarbohedrenes, organometallic clusters, and polypeptides. A brief review of our projects in those fields, either recently achieved or in progress, is provided. © 1996 John Wiley & Sons, Inc.

Notes: Various approaches for surface simulation are described. They are based on free, saturated, and embedded clusters, as well as periodic models. The features of semiempirical methods are reviewed and ideas for their improvement are discussed. Special features of the structure and stability of clusters are presented which are suitable for the transition to the solid state. Linear and nonlinear relationships for binding energies and bond lengths in dependence of the average coordination number are discussed. Finally, a systematic way for the simulation of adsorption at ion crystal surfaces by model clusters is suggested. © 1996 John Wiley & Sons, Inc.

Notes: Theoretical models will be presented in which the internal energy of (a) a ferromagnet and (b) a pyroelectric material is expressed in terms of magnetization and electric polarization, respectively. For the ferromagnet, simple models of elementary excitations (e.g., spin wave theory in an insulator, to which Stoner excitations must be added in a metal) lead to formulas for the internal energy at low temperature as power laws in the change of magnetization from its saturation value. An unconventional use of two order parameters, the sublattice magnetization plus the metallic discontinuity in momentum distribution at the Fermi surface, allows the phase transition between metallic and insulating states of antiferromagnets to be treated at T = 0, the three-dimensional transition-metal dichalcogenides being an example here. The treatment of the internal energy of the ferromagnet is then extended to include the entropy also, using specifically the Ising model in nonzero external magnetic field. Its relevance to the Landau theory of phase transitions will be emphasized. Some comments will finally be made about the analogue for pyroelectrics. © 1996 John Wiley & Sons, Inc.

Notes: Density functional theory (DFT) calculations of the electronic structure and properties of Fe, Ni, and Cu carbonyls with one and five transition-element atoms are reported, due to their importance in heterogeneous catalysis (especially in Fischer-Tropsch synthesis). The local density approximation (LDA) with the Vosko-Wilk-Nussain (VWN) correlation functional as well as the generalized gradient approximation (GGA) using Becke's 1988 exchange functional and Perdew's 1986 correlation functional (BP method) were applied to the study of the clusters using a triple-zeta plus polarization (TZP) basis set. Triplet and quintuplet states for FeCO, singlet and triplet for NiCO, and doublet for CuCO were considered, and optimization of energy at the LDA/BP level for these cases was done. Dissociation energies in M and CO fragments, frequencies, and population analysis were performed. Geometry optimization of the distance between the C atom and the clusters M5 (with geometries taken from the crystals) were carried out and the dissociation energy in fragments M5 and CO as well as population analysis at the optimized geometries were also done. Comparison between our results and some others published using ab initio and density functional methods were performed. © 1996 John Wiley & Sons, Inc.

Notes: The fractional quantum Hall effect (FQHE) occurs at a certain magnetic field strengths B*(n) in a two-dimensional electron gas of density n at strong magnetic fields perpendicular to the plane of the electron gas. At these magnetic fields strengths, the system is incompressible, i.e., there is a finite cost in energy for creating charge density fluctuations in the bulk, while the boundary of the electron gas has gapless modes of density waves. The bulk energy gap arises because of the strong electron-electron interactions. While there are very good models for infinite homogeneous systems and for the gapless excitations of the boundary of the electron gas, computational methods to accurately model finite, inhomogeneous systems with more than about 10 electrons have not been available until very recently. We will here review an ensemble density functional approach to studying the ground state of large inhomogeneous spin-polarized FQHE systems. © 1996 John Wiley & Sons, Inc.

Notes: In a positively charged metallic cluster, surface tension tends to enhance the ionic density with respect to its bulk value, while surface-charge repulsion tends to reduce it. Using the stabilized jellium model, we examine the self-expansion and compression of positively charged clusters of simply metals. Quantal results from the Kohn-Sham equations using the local density approximation are compared with continuous results from the liquid drop model. The positive background is constrained to a spherical shape. Numerical results for the equilibrium radius and the elastic stiffness are presented for singly and doubly positively charged aluminum, sodium, and cesium clusters of 1-20 atoms. Self-expansion occurs for small charged clusters of sodium and cesium, but not of aluminum. The effect of the expansion or compression on the ionization energies is analyzed. For Al6, we also consider net charges greater than 2+. The results of the stabilized jellium model for self-compression are compared with those of other models, including the SAPS (spherical averaged pseudopotential model). © 1996 John Wiley & Sons, Inc.

Notes: In this work, we examine several continuum models for estimating the free energy of solvation. We find statistically that the best overall one-parameter fit depends only on whether there is a hydrogen bond or not. Within the non-hydrogen-bonded set of molecules, a three-parameter fit including molecular volume, molecular surface area, and the electrostatic component from any reaction field model is quite successful. We do not find a strong bias for any of the dielectric models, although the PCM model of Tomasi and co-workers seems to slightly surpass the others. Within the hydrogen-bonded set of compounds, a good fit can be obtained simply by considering the surface accessible area of each atom, together with a dispersion and electrostatic interaction descriptor (a six-parameter fit). Such a model, containing primarily nonquantum chemical descriptors, extrapolates very well to the solvation energies of the non-hydrogen-bonded set with an overall R2 = 0.9334. © 1996 John Wiley & Sons, Inc.

Notes: In recent experimental studies of infrared (IR) spectra of hydrated phenol clusters, the Mikami and Ebata group at Tohoku University attributed apparently two very different spectra to PhOH(H2O)4. The first spectrum has a wide transparent region between 3500 and 3700 cm-1, which they named a window region. The second spectrum has several absorption bands in this window region. Our previous study revealed that the first spectrum was assigned to the isomers which have a single-ring structure of a hydrogen-bonding network of OH's of phenol and waters. The experimentalists suggested that the second spectrum might be identified to a proton-transferred phenol(SINGLEBOND)water cluster. In the present article, the geometrical structures of proton-transferred hydrated phenol clusters were determined with the ab initio molecular orbital method and their IR spectra were calculated. The proton-transferred phenol(SINGLEBOND)water cluster is at a stable local minimum, but the energy is much higher than that of the most stable nontransferred cluster. The calculated IR spectrum has a still wider window region and is far different from the experimental spectrum of the second type. © 1996 John Wiley & Sons, Inc.

Notes: Density functional calculations for ethylene glycol (CH2OHCH2OH) in the gas and in a dielectric medium are reported. The condensed-phase calculations are based on the self-consistent reaction field approach and the environment has the dielectric constant of liquid methanol. NPT Monte Carlo simulations of ethylene glycol (ETG) in liquid methanol are also reported. The simulations were carried out for three conformers of ETG (tGg′, gGg′, and tTt). Comparison between SCRF results for the conformational equilibrium in the gas and in the dielectric suggests that the tGg′ conformer is slightly stabilized relative to the gGg′ conformer in the solvent. However, the energy difference between them is less the 1.0 kJ/mol, which indicates that frequent interconversions between the tGg′ and gGg′ conformers are expected in the condensed phase. The all-trans conformer (tTt) is higher than the most stable conformer in the gas by 14 kJ/mol. Monte Carlo simulations predict that the tGg′ and gGg′ conformers have very similar energies in the solvent. However, the simulations also show, in agreement with experimental data, that the tTt conformer is stabilized in liquid methanol, relative to the gas phase. The microscopic mechanism leading to the stabilization of the tTt conformer in the liquid is related to the differential hydrogen-bonding formation between the ETG conformers and the methanol molecules. © 1996 John Wiley & Sons, Inc.

Notes: In intense laser fields, atoms and molecules can absorb many more photons than required for ionization. This phenomenon is called above-threshold ionization (ATI) and it shows up in kinetic energy spectra of ionized electrons. The presence of very high energy photoelectrons necessitates the use of extremely large grids in numerical simulations based on the full time-dependent Schrödinger equation. We show that by using a wave function splitting scheme one can circumvent the problem of large grids and thus obtain accurate multiphoton photoelectron spectra. This scheme will be very useful in studies of ATI spectra generated by molecules in intense laser fields. We illustrate this method for a one-dimensional model of the H atom in intense laser fields. © 1996 John Wiley & Sons, Inc.

Notes: He(I) ultraviolet (UV) photoelectron spectroscopy and ab initio, self-consistent field (SCF) calculations with the 6-31G basis set have been employed to characterize the valence electronic structures of anionic 2′-deoxyadenosine-5′-phosphate (5′-dAMP-). Theoretical ionization potentials (IPs) of 5'-dAMP-, of the neutral model compounds 9-methyladenine (9-MeA) and 3-hydroxytetrahydrofuran (3-OH-THF), and of the model anion CH3HPO4- have been obtained by applying Koopmans' theorem to ab initio SCF results. The ionization potentials predicted from the SCF calculations have been compared to He(I) photoelectron spectra of 9-MeA and 3-OH-THF. The SCF calculations predict a value (8.45 eV), for the highest occupied π orbital in 9-MeA which agrees well with the experimental vertical IP (8.39 eV). However, IPs for the highest occupied lone-pair orbitals in 3-OH-THF are predicted to be more than 1.52 eV higher than the experimental IPs. Results from recently reported [H. S. Kim and P. R. LeBreton, Proc. Natl. Sci. USA 91, 3725-3729 (1994), and N. S. Kim and P. R. LeBreton, J. Am. Chem. Soc., 118, 3694 (1996)] second-order Møller-Plesset perturbation (MP2) calculations and configuration interaction calculations using the configuration interaction singles (CIS) method indicate that configuration interaction effects strongly influence the energies of the first five ionization events arising from removal of electrons from the closed-shell model anion CH3HPO4-. Results from the 6-31G SCF calculations of 5′-dAMP-, 9-MeA, 3-OH-THF, and CH3HPO4- indicates that valence orbital electron distributions in the nucleotide and in the model compounds and anion are similar. The correspondence between the orbital structure of the nucleotide, and the model compounds and anion makes it possible, employing experimental photoelectron data and MP2/CIS computational results for the model compounds and anion, to individually correct IPs calculated for the nucleotide at the 6-31G SCF level. Here, this approach has provided values for the 13 lowest IPs of 5′-dAMP- and indicates that the first IPs of the base, sugar, and phosphate groups are 6.1, 7.8, and 5.5 eV, respectively. © 1996 John Wiley & Sons, Inc.

Notes: A conformational search was carried out for five opioid peptide homologues and for angiotensin II. Density of states versus energy plots were obtained for each peptide, and the occurrence of common main-chain conformations was investigated by searching homologies between strings of four, five, and six contiguous main-chain amino acid residues rotamers. The results were compared to rates of hydrolysis by endooligopeptidase (EOP) 24.15, known for its specificity for substrate conformations. A catalytic assay of the hydrolysis of angiotensin II was also performed. The two best substrates of EOP 24.15 were found to share unique main-chain conformations and the two worst substrates of EOP 24.15 were found to be nonstructurally homologous to each other and the remaining peptide chains. The conformational search is compared to previous experimental and theoretical results. © 1996 John Wiley & Sons, Inc.

Notes: Irregularities in the connection of the high- and low-density forms of the Perdew-Zunger correlation energy functional lead to computational difficulties that are removed by a new parametrization. These parametrizations and a number of other correlation functionals are compared and discussed. © 1996 John Wiley & Sons, Inc.

Notes: In this article, an attempt was made to develop an activation process model. The average energy of the translational motion of the atoms, taking part in the elementary activation process and being in the thermodynamic equilibrium with thermal radiation, was obtained using the quantum canonical Gibbs distribution and the model principles of elementary activation. The degeneracy and exclusion of some excited vibrational levels were taken into consideration, the result being a strong dependence of the probability of surmounting the activation barrier on the behavior of the excited vibrational states of the quantum subsystems. As an application of the development model, the formulas of the preexponential factor for solid-state atomic diffusivity and first-order chemical reaction rate constants were derived. Quantitative analysis of the atomic diffusion in solids in the framework of our model has made it possible to describe the diffusion processes in metals, covalent semiconductors, as well as diffusion anomalies, connected with the “nonclassical” behavior of the empirical Arrhenius dependence. A possible physical essence of a kinetic compensation effect is discussed. It was shown that compensation may be caused by only changing the degeneracy of the vibrational levels of the quantum subsystems. © 1996 John Wiley & Sons, Inc.

Notes: The general expression for the sixth-order Møller-Plesset (MP6) energy, E(MP6), has been dissected in the principal part A and the renormalization part R. Since R contains unlinked diagram contributions, which are canceled by corresponding terms of the principal part A, E(MP6) has been derived solely from the linked diagram terms of the principal part A. These have been identified by a simple procedure that starts by separating A into connected and disconnected cluster operator diagrams and adding terms associated with the former fully to the correlation energy. After closing all open disconnected cluster operator diagrams, one can again distinguish between connected and disconnected energy diagrams, of which only the former lead to linked diagram representations and, therefore, contributions to E(MP6). The connected diagram parts of A have been collected in four energy terms E(MP6)1, E(MP6)2, E(MP6)3, and E(MP6)4. The sum of these terms has led to an appropriate energy formula for E(MP6) in terms of first- and second-order cluster operators. © 1996 John Wiley & Sons, Inc.

Notes: Based on a cluster operator formulation of sixth-order Møller-Plesset (MP6) perturbation theory equations for the calculation of MP6 in terms of spin-orbital two-electron integrals are derived. Efficiency has been gained by systematically using intermediate arrays for the determination of energy contributions resulting from disconnected cluster operators. In this way, the maximum cost factor of O(M12) (M being number of basis functions) is reduced to O(M9). The implementation of MP6 on a computer is described. The reliability of calculated MP6 correlation energies has been checked in three different ways, namely (a) by comparison with full configuration interaction (CI) results, (b) by using alternative computational routines that do not involve intermediate arrays, and (c) by taking advantage of relationships between fifth-order and sixth-order energy contributions. First applications of the MP6 method are presented. © 1996 John Wiley & Sons, Inc.

Notes: A comparison of sixth-order Møller-Plesset perturbation energies (MP6) with the corresponding full configuration interaction (FCI) energies shows that in the case of equilibrium geometries MP6 values differ by just 1.7 mhartree. MP6 correlation energies turn out to be important for systems with oscillatory convergence behavior as well as for systems with considerable multireference character. Contributions from pentuple (P) and hextuple (H) excitations are mostly positive and smaller than 1 mhartree in the cases investigated. Initial oscillations in the convergence behavior of a MPn series result from positive fifth-order T energy contributions to the correlation energy. In these cases, MP6 correlation energies are relatively large (T contributions dominate the total correlation energy) and absolutely necessary when estimating the convergence limit of the MPn series. MP6 is an O(M9) method and, therefore, can only be used for relatively small electron systems. More economic alternatives are given by the approximated MP6 methods MP6(M8) and MP6(M7), which involve O(M8) and O(M7) operations, respectively. According to calculated absolute and relative energies, MP6(M7) is an attractive alternative to full MP6 because it offers reasonable sixth-order results for cost comparable to those of MP4. © 1996 John Wiley & Sons, Inc.

Notes: The characteristic polynomial associated with π-electrons of conjugated molecules are discussed by using subgraphs derived from molecular graphs as a basis for their construction. A practical method has been developed for evaluating the coefficient aK of conjugated molecules. Applying this method, the general formulas of evaluating the coefficient aK for homologous conjugated molecules have been obtained. The approach is illustrated on a few simple conjugated systems, including also a few polymeric systems. © 1996 John Wiley & Sons, Inc.

Notes: A new series of general formulas to evaluate the electron-repulsion integral (ERI) can be derived from modifying the Gauss-Rys quadrature formula. These named as “accompanying coordinate expansion (ACE) formulas” are capable of evaluating very fast ERIs, especially for contracted Gaussian-type orbitals (GTOs). According to the degree of the contraction of GTOs, the optimum formula can be selected among these ACEs. Numerical examples are shown for (ps|ps) and (pp|pp) ERIs as typical examples. It is found that the present ACE algorithm is numerically stable and is most efficient among all algorithms in the literature in the floating-point-operation (FLOP) count for all varieties of the degree of contraction. © 1996 John Wiley & Sons, Inc.

Notes: All the possible rotamers of 2-aminoethanol and 2-amino-ethanethiol were fully optimized at the ab initio level using the 6-31G** basis with correlation energy inclusion and zero-point energy evaluation. Thirteen local minima for the former and 14 for the latter compound were found. In both molecules, the gauche′-gauche-gauche′ (g′Gg′) is the prevailing conformation, but in the sulfurated compound, it is accompanied by relevant percentages of other conformers, owing to the very low ΔE values. The stability of the g′Gg′ accommodation derives mainly from the presence of weak hydrogen bridges (O(SINGLE BOND)H···N and S(SINGLE BOND)H···N, respectively). The rotation barriers around the C(SINGLE BOND)C and C(SINGLE BOND)N bonds are higher than those around the C(SINGLE BOND)O and C(SINGLE BOND)S ones. © 1996 John Wiley & Sons, Inc.

Notes: That the bound energy eigenstates of one-dimensional quantum systems can be degenerate in the presence of specific singular or supersingular potentials is demonstrated by choosing a family of bistable and other oscillators. Relevance of our study to spectroscopic observations is noted. Quasi-degeneracy is found even in the absence of any singularity in the potential and the importance of tunneling is highlighted in this context to analyze the general nature of such potentials leading to double degeneracy. Additionally, the case of spiked oscillators is discussed with particular reference to the “Klauder phenomenon,” revealing clearly that the mere presence of singularity in the potential is not a sufficient criterion for the occurrence of degeneracy. © 1996 John Wiley & Sons, Inc.

Notes: We employ a dynamical configuration interaction approach to study the transport of charge along the polyacetylene chain axis when placed in an external electric field. The steady-state wave function in this approach is constructed as a linear combination of solitonic wave functions weighted by nuclear vibrational functions. We retain the Born-Oppenheimer approximation, i.e., we neglect the terms involving the differentiation of the electronic configurations with respect to the nuclear coordinates. We also use basis functions that follow the nuclei. A solution of the time-dependent Schrödinger equation in terms of the wave functions of the above kind is obtained, leading to a closed formula for estimating mobilities of the soliton transport. © 1996 John Wiley & Sons, Inc.

Notes: The impact of the choices of the complete model space (CMS) and of the orbital basis set on the existence, attainability, and properties of the nonstandard solutions of the valence-universal coupled-cluster (VU-CC) methods has been studied for the case of nonlinear equations corresponding to the atomically oriented form of these methods accounting for one- and two-electron excitations (VU-CCSD/R method) and applied to the Be atom. The results for five 1S states are discussed. In addition to the previously applied CMS defined by the orbital set (2s, 2p) and (2s, 3s), we have employed the CMSs defined by the (2s, 2p, 3s) set. For each of the CMSs several nonstandard solutions are documented. It is found that the energies of the individual states corresponding to standard and nonstandard solutions differ very little. These energies are also almost independent on the choice of CMS. On the other hand, the energies of excited states disclose a strong dependence on the radial structure of the orbital basis set. It is also demonstrated that the magnitudes of the cluster amplitudes representing a set of states depend both on the choice of the CMS and whether they correspond to the standard or nonstandard solutions. © 1996 John Wiley & Sons, Inc.

Notes: Two different approaches were used to explain the distribution of water coordinates in the crystallographically invisible part of the unit cell. Monte Carlo calculations were done starting from different initial water structures. Extended Monte Carlo calculations using equal initial structures were also used to obtain two further water structures. The differences between the Monte Carlo water structures were used to calculate the mean-square displacements of the water molecules. Monte Carlo calculations starting from different structures lead to a mean-square displacement of 0.58 Å2, whereas extended Monte Carlo moves using the same initial structure only show a mean-square displacement of 0.17 Å2. The mean-square displacement of 0.58 Å2 can be used to explain the experimental data. © 1996 John Wiley & Sons, Inc.

Notes: Continuing the search for a broader interpretation of hemeprotein behavior, we give preliminary results showing that there are electric and dynamic couplings between the heme group and amino acid residues within the protein matrix. EPR and X-ray absorption spectroscopy studies on azidometmyoglobin show that both magnetic and geometric properties of Fe—N3 evolve in the same nonlinear way as pH is increased and are tightly correlated to the strains on the helical segments of the protein. Flash photolysis of carbon monoxide hemoglobin, in the presence of ethanol or formamide, allows the study of cosolvent effects on geminate and nongeminate recombinations of the CO ligand trapped within the protein matrix. Data clearly show that cosolvents alter the statistic fluctuations of the protein, as well as the ligand partition between different protein matrix domains. From these studies, it is concluded that alterations occurring at particular sites give way to global protein perturbations. Then, each perturbated protein domain - binding site included - evolves with its own sensitivity to a new metastable state of the protein. The amplification of the initial perturbation which - instead of regressing - progressively propagates through the whole macromolecule is typical of a dissipative structure in the Prigogine sense. Biological properties of hemeproteins largely involve the surrounding solvent, via permanent or temporary exchanges of water molecules, protons, and small ligands. These fluxes along with their entropic corollary are not quite compatible with a conservative system. These works present the current trends developed in our laboratory in association with the European network “The Dynamics of Protein Structure.” In this framework, our laboratory collaborates with Dr. W. Doster and T. Kleinert (Munich, Germany) for the CO recombination studies in hemoglobin and with Dr. J. Hutterman (Homburg, Germany) and Drs. A. Bianconi and S. Della Longa (Rome and L'Aquila, Italy) for the magnetic and geometric properties of the myoglobin iron site. © 1996 John Wiley & Sons, Inc.

Notes: To investigate the backbone dynamics of proteins 15N longitudinal and transverse relaxation experiments combined with {1H, 15N{ NOE measurements together with molecular dynamics simulations were carried out using ribonuclease T1 and the complex of ribonuclease T1 with 2′GMP as a model protein. The intensity decay of individual amide cross peaks in a series of (1H, 15N)HSQC spectra with appropriate relaxation periods was fitted to a single exponential by using a simplex algorithm in order to obtain 15N T1 and T2 relaxation times. The relaxation times were analyzed in terms of the “model-free” approach introduced by Lipari and Szabo. In addition, a nanosecond molecular dynamics (MD) simulation of ribonuclease T1 and its 2′GMP complex in water was carried out. The angular reorientations of the backbone amide groups were classified with several coordinate frames following a transformation of NH vector trajectories. In this study, NH librations and backbone dihedral angle fluctuations were distinguished. The NH bond librations were found to be similar for all amides as characterized by correlation times of librational motions in a subpicosecond scale. The angular amplitudes of these motions were found to be about 10°-12° for out-of-plane displacements and 3°-5° for the in-plane displacement. The contributions from the much slower backbone dihedral angle fluctuations strongly depend on the secondary structure. The dependence of the amplitude of local motion on the residue location in the backbone is in good agreement with the results of NMR relaxation measurements and the X-ray data. The protein dynamics is characterized by a highly restricted local motion of those parts of the backbone with defined secondary structure as well as by a high flexibility in loop regions. Comparison of the MD and NMR data of the free liganded enzyme ribonuclease T1 clearly indicates a restriction of the mobility within certain regions of the backbone upon inhibitor binding. © 1996 John Wiley & Sons, Inc.

Notes: NMR techniques can give insight into a wide variety of motional events that occur in proteins over a range of timescales. In the first section of this article an overview of the results of dynamics studies, using NMR methods, on both small globular and larger multi-domain proteins is presented including the findings from investigations of non-native partly folded states. The second section of the article then concentrates on two topics where NMR can give residue specific quantitative data, namely coupling constant measurements and relaxation studies, including comparisons of these NMR data with results from crystallographic studies and theoretical molecular dynamics simulations. Finally the possible functional significance of the experimentally observed motions in proteins is discussed. © 1996 John Wiley & Sons, Inc.

Notes: We measured the νFe-His Raman band of horse heart deoxymyoglobin and human deoxyhemoglobin as a function of temperature between 10 and 300 K. A self-consistent spectral analysis of the deoxymyoglobin Raman band reveals that it is underlied by three different sublines with frequencies at Ο1 = 209 cm-1, Ο2 = 217 cm-1, and Ο3 = 225 cm-1 and an identical half-width of 13 cm-1. All these parameters were found to be independent of temperature. These sublines were attributed to different conformational substates of the Fe2+-His F8 linkage, which comprise different out-off-plane displacements of the heme iron and tilt angles of the Fe2+-Ng(His F8) bond. The intensity ratio I3/I2 exhibits a van't Hoff behavior between 150 and 300 K, bends over in a region between 150 and 80 K, and remains constant at lower temperature. In contrast, I2/I1 shows a maximum at 170 K and approaches a constant value at 80 K. These data can be fitted by a modified van't Hoff expression, which accounts for the freezing into nonequilibrium distributions of substrates in a temperature interval ΔT around a distinct temperature Tf and also for the linear temperature dependence of the protein's specific heat. The fits to the above intensity ratios yield a freezing temperature of Tf = 117 K and a transition region of ΔT = 55 K. The νFe-His Raman band of human deoxyhemoglobin was decomposed into seven sublines with frequencies at 195, 202, 211, 218, 226, 234, and 240 cm-1, with half-widths of 12 cm-1. While the low-frequency sublines are strong at 300 K, the high-frequency sublines dominate the band at cryogenic temperatures. In comparison, we also investigated the temperature dependence of the near-infrared band III at 760 nm. Band III of deoxymyoglobin can be decomposed into two subbands which are 165 cm-1 apart. The ratio of their absorption cross sections shows a temperature dependence which parallels that of the ratio I3/(I2 + I1) of the corresponding Raman sublines. Band III of deoxyhemoglobin was decomposed into three subbands, the absorption cross sections of which also depend on temperature, similar to what has been observed for the νFe-His subbands. These observations provide strong evidence that the frequency positions of the subbands of band III and the νFe-His sublines are governed by the same coordinates. For both proteins investigated, the frequency positions and the half-widths of the band III subbands depend significantly on temperature. This is rationalized in terms of an earlier proposed model (Cupane et al., Eur. Biophys. J. 21; 385 1993) which assumes that the corresponding electronic transition is coupled to a bath of low-frequency modes. Our data indicate that these modes are harmonic below 134 K but become anharmonic above this temperature. This onset of anharmonic motions is interpreted as resulting from conformational transitions within the protein which affect the prostethic group via heme-protein coupling. © 1996 John Wiley & Sons, Inc.

Notes: 131Xe-NMR longitudinal relaxation rates have been measured by the inversion recovery method for xenon in presence of lecithin vesicles or a small protein charybdotoxin. The transverse relaxation rates in the same spectra have been obtained by spectral deconvolution. The results show that xenon in lecithin vesicles is in a rapid exchange between free and bound sites and that averaging of the electric-field gradient at the Xe nucleus is a two-step process. From these results, estimates have been obtained for the parameters entering the 131Xe quadrupolar interaction. © 1996 John Wiley & Sons, Inc.

Notes: The interaction between the R2R3 DNA binding domain of the oncoprotein c-Myb and oligodeoxynucleotides was investigated by 1H-NMR spectroscopy and fluorescence anisotropy assays. Titration of 12 and 16 base-pair DNA fragments containing the TAACGGTC sequence with R2R3 revealed the presence of two complexed forms (in a 40/60 ratio): either two complexes or two conformations in slow exchange at the NMR chemical shift time scale. The largest variations of imino proton chemical shifts were observed for the imino proton of the base pairs 2, 3, 4 and 6 of the DNA sequence, suggesting a direct involvement of these base pairs in the interaction. Using fluorescence anisotropy measurements, a dissociation constant of 5.12 ± 1.49 nM for the specific DNA-R2R3 complex was found, whereas a value of 2.7 ± 0.1 μM was determined for the nonspecific complex. © 1996 John Wiley & Sons, Inc.

Notes: We show that the Ising free-energy functional f(T, X) yields a second-order phase transition. The value of X, a generalized order parameter, which minimizes the functional is the real-order parameter, Ω = tanh[Ω(TC/T)] where Tc = J/(2k) and J is the Ising coupling constant. The Ising theory is applicable to the second-order phase transitions in ferromagnets, ferroelectrics, and superconductors. The constant temperature slices of f(T, X) yields the Ising functional which is valid over the entire range of T and which is identical to the Landau functional for T/TC close to unity. © 1996 John Wiley & Sons, Inc.

Notes: Calculations are reported on the dipole moments, transition energies, and hyperpolarizabilities of 1-amino-4-nitrobutadiene and its four aza-analogs using a semiempirical sum-over-states method. The results obtained suggest that the presence of nitrogen in the unsaturated chain can have a substantial beneficial effect on the molecular hyperpolarizability provided that it is placed in a position which assists charge transfer from the donor to acceptor group on excitation. The values calculated for the 3-azabutadiene are greater than those found for either the simple butadiene or for donor - acceptor aromatics or five-membered heterocycles. The results do not agree with those reported recently using an ab initio approach. © 1996 John Wiley & Sons, Inc.

Notes: In a novel approach to molecular shape representation, molecular electronic density functions are treated as 3D hypersurfaces in a 4D Euclidean space, spanned by the three spatial coordinates and an electronic density threshold variable. The general objects resulting from this approach are topological spheres with handles, modeling the electron density. Based on this model, a new, simple representation of molecular topology of electron density is proposed, using a tree describing the evolution of topological spheres with handles as function of a varying density threshold. The vertices of the tree are characterized by nonnegative integers (the genea of the topological objects). This representation is compared to earlier, detailed 3D topological descriptions of molecular shape as well as to the more conventional molecular network (bond skeleton) representation. A variety of relevant topological equivalence relations of molecules are discussed and a new, global, molecular similarity measure is proposed. © 1996 John Wiley & Sons, Inc.

Notes: An ab initio procedure for the calculation of atomic and molecular electronic wave functions, the Optimized-Basis-Set Multi-Configuration Spin-Coupled (OBS-MCSC) method, is generalized by introducing a separate linear combination of spin functions for each configuration, turning it into the OBS-GMCSC method. The ability to use a second-order minimization procedure in the computation of the wave function is maintained through appropriate generalization of the analytic expressions for the first and second derivatives of the energy with respect to the optimization parameters, as is the optional inclusion among the latter of the basis-function exponential parameters. The generalization, a variational improvement of the wave function, strengthens the connection with classical VB theory, of which the method can now be considered an optimized-orbitals variant, while maintaining the link with single-configuration Spin-Coupled theory, of which it may still be considered a multiconfiguration extension. The method can also be viewed as a nonorthogonal variant of the MCSCF approach. To demonstrate its practical feasibility and usefulness, the OBS-GMCSC method is applied to a study of the electronic structure and electron affinity of boron. © 1996 John Wiley & Sons, Inc.

Notes: Proton affinities (PAs) of a series of phosphorous compounds bearing the phosphoryl function have been calculated using AM1 and PM3, as well as lithium-cation affinities (LCAs) using the recent PM3 lithium parametrization. Sulfonyl derivatives PAs and LCAs have been also calculated using PM3. The Li+ cation can be bonded in a “chelate” form with the two oxygens of the sulfonyl group. Nevertheless, the “linear” adduct, with the lithium-oxygen bond collinear with one of the S=O bonds, is more stable. This is confirmed by ab initio calculations on Me2SO2—Li+ adducts. © 1996 John Wiley & Sons, Inc.

Notes: The reaction of Y+ + NH3 → Y+ NH + H2 was theoretically investigated by ab initio MO methods. Two possible pathways (1-1 H2 loss and 1-2 H2 loss) on the singlet potential energy surface and reaction mechanism were examined and discussed. The singlet and triplet PESs of this reaction system were compared to confirm the correctness of spin conservation concepts. © 1996 John Wiley & Sons, Inc.

Notes: The performance of a number of different local and nonlocal density functional theory (DFT) methods has been investigated for some small titanium - oxygen systems. Equilibrium geometries, ionization potentials, dipole moments, atomization energies, and harmonic vibrational frequencies have been calculated for the TiO, TiO2, and Ti2 molecules, and the results are compared with experimental data and ab initio calculations. It is shown that most DFT methods perform much better than the ab initio Hartree - Fock (HF), second-order perturbation theory (MP2), and configuration interaction including single and double excitations (CISD) treatments. For good agreement with experimental data, gradient corrections to the exchange part of the DFT functional are needed, as well as some type of correction for the errors in the calculated energy splittings between different atomic states of titanium. Hybrid methods including a mixture of HF exchange with DFT exchange correlation do not perform as well as “pure” DFT methods for the studied systems. © 1996 John Wiley & Sons, Inc.

Notes: The Feynman path integral method is applied to the many-electron problem of quantum chemistry. We begin with constructing new closure relations in terms of the linear combination of atomic orbital (LCAO) coefficients and investigate the transition amplitude and the partition function of the system in question; then a “classical path of electrons,” which is described by the time-dependent Hartree-Fock-Roothaan equation, is obtained by minimizing the action integral of the system with respect to the “electron coordinate.” The next order approximation is obtained by evaluating the deviation from this classical path, which is approximately written by a Gaussian integral. The result is expected to be the random-phase approximation. © 1996 John Wiley & Sons, Inc.

Notes: Configuration interaction (CI) spaces obtained from the full CI space by imposing arbitrary restrictions on the occupancies of molecular orbital (MO) groups are studied. It is proved that such restricted spaces are in a certain sense “closed.” Namely, in the course of the Hamiltonian matrix construction the excitations out of the chosen restricted CI space may be easily replaced by the excitations within this space. © 1996 John Wiley & Sons, Inc.

Notes: Formulas are derived for the density matrices belonging to an n-particle wave function built on the basis of single-center explicitly correlated Gaussian basis functions. An explicit formula for the first-order density matrix, P(r1, r′1), is obtained for computing the probability distribution P(r1, r1). Other formulas are derived for matrix elements of the first-order density operator P on a basis of single-particle Gaussian orbitals so that natural orbitals (NOs) can be expressed in such a basis. The method is illustrated for the case of the ground state of the helium atom using the 16-term (geminal) wave function by Singer and Longstaff (E = -2.90233 au) and a set of even-tempered Gaussian orbitals. The resulting natural orbitals compare favorably with natural orbitals from Cl expansions. The method is also applied to our 20 term (trimal) wave function for the ground state of dipositronium (E = -0.51560 au). Analysis is made in this case for pair correlation functions of both the electron-electron and the positron-electron pairs; results include the radial distributions of these pairs and their relative angular momentum. © 1996 John Wiley & Sons, Inc.

Notes: We examine the state of the art of the solvation procedure called the polarizable continuum model (PCM), focusing our attention on the basic properties: energy of the solute, solvation energy, and their derivatives with respect to nuclear coordinates. The PCM method is based on the use of an effective solute Hamiltonian, where the solute-solvent potential is described in terms of continuous response functions with boundary conditions given in terms of the solute cavity surface. This exposition is mainly based on recent progress, a large part of which is still in press. The new procedures are quite effective, at the ab initio quantum mechanical level, but cannot be applied to very large solutes for the limitations of computer hardware. We introduce then other methods, presented here for the first time, which make possible the classical calculation of the solvation energy also for very large solutes (a few thousand atoms). The strategy outlined here regards a new method to define cavity surfaces (supplemented with analytical definitions of its partition in tesserae) and of their derivatives, combined with a fast noniterative method to compute solvation energy. Finally, we discuss the introduction of this procedure in hybrid quantum mechanical/molecular mechanics descriptions of large solutes (enzymes), where the quantum description is limited to the reacting portion of the enzyme. © 1996 John Wiley & Sons, Inc.

Notes: A practical procedure (FUERZA) to obtain internal force constants from Cartesian second derivatives (Hessians) is presented and discussed. It allows a systematic analysis of pair atomic interactions in a molecular system, and it is fully invariant to the choice of internal coordinates of the molecule. Force constants for bonds or for any pair of atoms in general are defined by means of the eigenanalysis of their pair interaction matrix. Force constants for the angles are obtained from their corresponding two-pair interaction matrices of the two bonds or distances forming the angle, and the dihedral force constants are similarly obtained using their corresponding three-pair interaction matrices. © 1996 John Wiley & Sons, Inc.