Regional self-interaction correction of density functional theory

We propose a new simple scheme for self-interaction correction (SIC) of exchange functionals in the density functional theory. In the new scheme, exchange energies are corrected by substituting exchange self-interactions for exchange functionals in regions of self-interaction. To classify the regions of self-interaction, we take advantage of the property of the total kinetic energy density approaching the Weizs?cker density in the case of electrons in isolated orbitals. The scheme differs from conventional SIC methods in that it produces optimized molecular structures. Applying the scheme to the calculation of reaction energy barriers showed that it provides a clear improvement in cases where the barriers are underestimated by conventional "pure" functionals. In particular, we found that this scheme even reproduces a transition state that is not given by pure functionals.     

Thermodynamic properties of some N-alkyl-N-methylpiperidinium chlorides and N-alkylpiperidine hydrochlorides in water

Densities and heat capacities at 25°C were measured for N-octyl-, N-decyl- and N-dodecyl-N-methylpiperidinium chlorides and for N-octyl- and N-dodecylpiperidine hydrochlorides in water as functions of concentration. Enthalpies of dilution at 25°C and osmotic coefficients at 37°C of the N-methyl-N-alkylpiperidinium chlorides were also measured as functions of concentration. The partial molar volumes, heat capacities, relative enthalpies, nonideal Gibbs energies and entropies at 25°C were derived as functions of the surfactant concentration. By increasing the alkyl chain length of the surfactant, both the apparent molar volume vs. concentration curves are shifted toward greater values while the corresponding ones for the heat capacity are moved toward more negative values. These results are consistent with the higher hydrophobicity the longer the alkyl chain of the surfactant is. In the micellar region, the entropy and enthalpy vs. log m/m _cmc curves increase in a parallel manner by decreasing the alkyl chain length of the surfactant. Consequently, the negligible effect of the hydrophobicity of the surfactant on the Gibbs energy vs. log m/m _cmc trends is due to the enthalpy-entropy compensative effect. The thermodynamic functions of micellization were graphically evaluated on the basis of the pseudo-phase transition model. The absolute values of both the volume and heat capacity of micellization increase with an increasing number of carbon atoms in the alkyl chain (n _c ). The enthalpy and entropy of micellization vs. n _c are convex curves. Comparisons are also made between the present data and those of some alkylpyridinium chlorides reported elsewhere.     

Chromophore Systeme, 2. Konformation und Lichtabsorption von (2-Alkoxyvinyl)ethandionen

Chromophoric Systems, 2. – Conformation and Absorption of Light in (2-Alkoxyvinyl)ethanediones The unusual color properties of yellow 1,2-bis(4,5-dihydrofuran-3-yl)ethane-1,2-dione ( 1 ) and colorless 1,2-bis(5,6-dihydro-4H-pyran-3-yl)ethane-1,2-dione ( 2 ) in the solid state and in solution are investigated by crystal structure analysis, UV-Vis, PE, and ¹³C-NMR spectroscopy as well as by reparameterized force-field calculations. Dione 1 takes an antiperiplaner CO/CO conformation of its chromophore in the solid state whereas dione 2 is twisted in this respect by 102°. In solution both chromophores are not planar, but 1 is less distorted (ca. 142°) than dione 2 (ca. 126°). The calculated rotational barrier for CO/CO twisting amounts to 0.9 kcal mol⁻¹ for 1 and to 2.2 kcal mol⁻¹ for 2 .     

Analysis of fourth-order Møller–Plesset limit energies: the importance of three-electron correlation effects

Fourth-order M?ller–Plesset (MP4) correlation energies are computed for 28 atoms and simple molecules employing Dunning's correlation-consistent polarized-valence m-zeta basis sets for m=2, 3, 4, and 5. Extrapolation formulas are used to predict MP4 energies for infinitely large basis sets. It is shown that both total and partial MP4 correlation energies can be extrapolated to limit values and that the sum of extrapolated partial MP4 energies equals the extrapolated total MP4 correlation energy within calculational accuracy. Therefore, partial MP4 correlation energies can be presented in the form of an MP4 spectrum reflecting the relative importance of different correlation effects. Typical trends in calculated correlation effects for a given class of electron systems are independent of the basis set used. As first found by Cremer and He [(1996) J Phys Chem 100:6173], one can use MP4 spectra to distinguish between electron systems with well-separated electron pairs and systems for which electrons cluster in a confined region of atomic or molecular space. MP4 spectra for increasing size of the basis set reveal that smaller basis set calculations underestimate the importance of three-electron correlation effects for both classes by overestimating the importance of pair correlation effects. The minimum size of a basis set required for reliable MP4 calculations is given by a valence triple-zeta polarized basis, which even in the case of anions performs better than a valence double-zeta basis augmented by diffuse functions. Received: 14 June 2000 / Accepted: 16 June 2000 / Published online: 24 October 2000     

Fourier transform spectroscopy of the CO-stretching band of O-18 methanol

The high-resolution Fourier transform spectrum of the ν₈ CO-stretching band of CH₃¹⁸OH between 900 and 1100 cm⁻¹ has been recorded at the Canadian Light Source (CLS) synchrotron facility in Saskatoon, and the majority of the torsion-rotation structure has been analyzed. For the ν_t = 0 torsional ground state, subbands have been identified for K values from 0 to 11 for A and E torsional symmetries up to J values typically well over 30. For ν_t = 1, A and E subbands have been assigned up to K = 7, and several ν_t = 2 subbands have also been identified. Upper-state term values determined from the assigned transitions using the Ritz program have been fitted to J(J + 1) power-series expansions to obtain substate origins and sets of state-specific parameters giving a compact representation of the substate J-dependence. The ν_t = 0 subband origins have been fitted to effective molecular constants for the excited CO-stretching state and a torsional barrier of 377.49(32) cm⁻¹ is found, representing a 0.89% increase over the ground-state value. The vibrational energy for the CO-stretch state was found to be 1007.49(7) cm⁻¹. A number of subband-wide and J-localized perturbations have been seen in the spectrum, arising both from anharmonic and Coriolis interactions, and several of the interacting states have been identified.     

Investigations into the alpha-decay of <Superscript>195</Superscript>At

The low-energy nuclear structure and decay properties of the neutron-deficient isotopes ¹⁹⁵At and ¹⁹¹Bi have been studied. ¹⁹⁵At was produced in the reaction ¹⁴²Nd(⁵⁶Fe,p2n)¹⁹⁵At and ¹⁹¹Bi as the daughter activity of ¹⁹⁵At. The activities were implanted in a position-sensitive silicon detector after being separated from the primary beam by a gas-filled recoil separator. The 1/2⁺ intruder state was determined to be the ground state in ¹⁹⁵At with an alpha-decay energy of E _α = 6953(3) keV and a half-life T _1/2 = 328(20) ms. Another state with an alpha-decay energy E _α = 7075(4) keV and a half-life T _1/2 = 147(5) ms was found to decay to a 148.7(5) keV excited state in ¹⁹¹Bi for which a spin and parity of 7/2^- were deduced. Consequently, the same 7/2^- character was assigned to the initial state at 32(7) keV in ¹⁹⁵At on the basis of unhindered alpha-decay. The 9/2^- state, being the ground state in heavier odd-mass astatine isotopes, was not observed. Received: 16 September 2002 / Accepted: 19 December 2002 / Published online: 25 March 2003 RID="a" ID="a"e-mail: heikki.kettunen@phys.jyu.fi RID="b" ID="b"Present address: Laboratory of Radiochemistry, P.O. Box 55, FIN-00014, University of Helsinki, Finland. RID="c" ID="c"Present address: Radiation and Nuclear Safety Authority, P.O. Box 14, FIN-00881 Helsinki, Finland. Communicated by W. Henning     

MO theoretical studies of electron pair decorrelation processes. I. Pair correlation energy and single bond dissociation

The homolytic dissociation of a single bond involves the decorrelation of one electron pair. Thus, the contribution of electron correlation to dissociation energies is large. In the present paper a new procedure is presented which allows the computation of the (within the given basis) complete correlation energy of one optimized electron pair. The method which requires only modest computational effort has been applied to the calculation of dissociation energies of a number of bonds of different types. The results show that the correlation of the electron pair of the bond which is broken contributes about 50–80% to the change of the total correlation energy occuring during the dissociation process which amounts to 20–70 kcal/mol. The fraction of correlation contributed by the bond electron pair as well as the relative importance of the left-right correlation within the bond depend very much on the type of the bond. In the case of CC and CH single bonds our method yields dissociation energies which are low by only about 5 kcal/mol. Thus, the method seems to be well suited for the calculation of potential surfaces of non-concerted organic chemical reactions which involve diradicals as intermediates.     

A trial wave function to study of the structure of the molecular dd μ ion

This paper provides a new effort to study of the ddμ structure. The present work is numerically performed using a new trial wave function to the ddμ system in configuration of coupled channels. The present results of energies are more accurate than those of our previous work. The obtained results of formation rates are close to results published by Yu.V. Petrov et al. and giving strong indications that the trial wave function is good enough in determining the resonance states of the mentioned ionic molecule.     

The C−NO2 bond dissociation energies of some nitroaromatic compounds: DFT study

The C−NO₂ bond dissociation energies in nitrobenzene; 3-amino-nitrobenze; 4-amino-nitrobenze; 1,3-dinitrobenzene; 1,4-dinitrobenzene; 2-methyl-nitrobenzene; 4-methyl-nitrobenzene; and 1,3,5-trinitrobenzene nitroaromatic molecules, are computed using B3LYP, B3PW91, B3P86 three-parameter hybrid Density Functional Theory (DFT) methods in conjunction with 6-31G^** basis set. By comparing the computed energies and experimental ones, it is found that B3P86/6-31G^** is not capable of predicting the satisfactory bond dissociation energy (BDE). The BDEs computed with both B3LYP/6-31G^** and B3PW91/6-31G^** for the nitroaromatic molecules are closer to the experimental ones than those obtained with B3P86/6-31G^**. But, when compared with the experimental one, the BDE from the B3LYP/6-31G^** has the maximum deviation, which is completely outside our desired target accuracy for chemical predictions (less than 2.00 kcal mol⁻¹). Therefore, we suggest B3PW91/6-31G^** method as a reliable method of computing the BDE for removal of the nitrogen dioxide group in the nitroaromatic compounds. In addition, the C−NO₂ BDEs for 2,4,6-trinitrotoluene (TNT), triaminotrinitrobenzene (TATB), diaminotrinitrobenzene (DATB), and picramide are studied with B3PW91/6-31G^** method.     

A rational method for probing macromolecules dissociation: the antibody-hapten system

The unbinding process of a protein-ligand complex of major biological interest was investigated by means of a computational approach at atomistic classical mechanical level. An energy minimisation-based technique was used to determine the dissociation paths of the system by probing only a relevant set of generalized coordinates. The complex problem was reduced to a low-dimensional scanning along a selected distance between the protein and the ligand. Orientational coordinates of the escaping fragment (the ligand) were also assessed in order to further characterise the unbinding. Solvent effects were accounted for by means of the Poisson–Boltzmann continuum model. The corresponding dissociation time was derived from the calculated barrier height, in compliance with the experimentally reported Arrhenius-like behaviour. The computed results are in good agreement with the available experimental data.