Kinya Kobayashi

Density functional MO calculation for stacked DNA base-pairs with backbones☆

In order to elucidate the effect of the sugar and phosphate backbones on the stable structure and electronic properties of stacked DNA base-pairs, we performed ab initio molecular orbital (MO) calculations based on the density functional theory and Slater-type basis set. As a model cluster for stacked base-pairs, we employed three isomers for the dimer unit of stacked guanine-cytosine pairs composed with backbones as well as base-pairs. These structures were fully optimized and their electronic properties were self-consistently investigated. By including the backbones, the difference in total energy among the isomers was largely enhanced, while the trend in relative stability was not changed. The effect of backbones on the electronic properties is remarkable: the MOs with the character of the PO4 parts of backbones appear just below the highest-occupied MO. This result indicates that the PO4 parts might play a rule as a reaction site in chemical processes concerning DNA. Therefore, we conclude that the DNA backbones are indispensable for investigating the stability and electronic properties of the stacked DNA base-pairs.

Molecular structures, binding energies and electronic properties of dopyballs C59X (X=B, N and S)

Abstract When optimized by a molecular orbital method with Harris functional and spin-restricted approximations, the molecular structures of C 59 B and C 59 N are found to be almost the same as that of C 60 , while that of C 59 S is different from that of C 60 . The binding energies calculated self-consistently are 6.05 (C 60 ), 6.03 (C 59 B), 6.00 (C 59 N) and 5.91 (C 59 S) eV/atom. The energy gaps between the LUMO and the HOMO or the half-occupied molecular orbital are 1.50 (C 60 ), 1.06 (C 59 B), 0.30 (C 59 N) and 0.63 eV (C 59 S), and the Mulliken charges of the doped B, N and S atoms are 4.5, 7.2 and 15.5, respectively, meaning the N atom exists as an acceptor, while the B and S atoms exist as donors in these dopyballs.

Non-local density functional calculations of binding energies of carbon fullerenes Cn, with n = 10, 12, 16, 20, 24, 28, 32, 36, 50, 60, 70, 80, 90, 100, 110 and 120

Abstract Using a molecular orbital method based on the non-local density functional formalism, we have calculated the binding energies and HOMO—LUMO gaps of carbon fullerenes C n assuming that all bond lengths in each fullerene are the same. The results are as follows: (1) The fullerenes containing quadrilateral rings are less stable than those with only pentagonal and hexagonal rings. (2) With a larger number of hexagonal rings in a fullerene, the binding energy is larger. (3) C 60 , C 70 , C 100 , C 90 and C 32 have especially large HOMO—LUMO gaps and are expected to be less reactive.

EFFICIENCY OF THE MO METHOD USING A SLATER-TYPE BASIS SET AND NON-LOCAL DENSITY FUNCTIONAL FORMALISM FOR DESCRIBING DNA BASE STACKING ENERGY

Abstract We have developed a molecular–orbital (MO) method based on a Slater-type basis set and non-local density functional theory (DFT) for describing the DNA base stacking properties, and its efficiency has been confirmed by investigating the stacking energy of cytosine dimer. Our DFT method can reproduce the dependence of stacking energy on the stacking conformation obtained by the ab initio second-order Moller–Plesset (MP2) calculation. The stable structures of hydrogen-bonded Watson–Crick base pairs and (PO 4 H 2 ) −1 ion have been investigated and the structures obtained by our DFT method are comparable with those from the MP2 and DFT methods in Gaussian94. Therefore, our DFT method may be applicable for investigating the stable structures of more realistic models for stacked DNA base pairs including backbones.

Bonding and Electronic Properties of Substituted Fullerenes C58B2, C58N2 and C58BN

Abstract When optimized by a molecular orbital (MO) method with Harris approximation, structures of substituted fullerenes C58B2, C58N2 and C58BN are almost the same as that of 0, and the self-consistently calculated binding energies of C58B2 and C58BN are similar to that of C60, while that of C58N2 is much smaller. The energy gaps between the highest-occupied MO and the lowest-unoccupied MO depend on the positions of substituted atoms, being 0.49-0.54 (C58B2), 0.23-0.28 (C58N2) and 0.10-0.86 eV (C58BN).

Trench and via filling profile simulations for copper electroplating process

We have developed a simulation method to calculate filling profile in sub-micron scale trenches and vias using Cu electroplating. The physical properties (diffusivity and surface reaction rate) of additives are directly incorporated in the boundary condition of this method so that the effects of physical properties of the additives on profile of filling can be evaluated. Using this method, we clarified that micro-knobs and overhangs on seed layer are important for creating a void or seam in the Cu film.

A plasma kinetics model: analysis of wall loss reactions in dry etching of silicon dioxide

Abstract Computational models for gas-phase chemical reactions in plasmas and for sticking reactions on metallic Al walls have been developed and applied to the plasma chemistry of dry etching of silicon dioxide. Dissociation paths and threshold energies of gases are determined by using an ab initio density functional molecular orbital method, and dissociation cross sections are approximated. The electron energy distribution function is determined by using a particle-in-cell model with the Monte Carlo collision method, and dissociation reaction rates are determined. Plasma densities, electron temperatures. and radical densities are calculated by a kinetic model which consists of the fluid equations for plasmas and rate equations for radicals. The model effectiveness was confirmed by results comparison for the Ar discharge in an radio-frequency device. The chemical compositions of the dry etching plasmas have been investigated for C 4 F 8 . Calculated electron temperatures and densities agree with experimental results within factors of three. Correlations could be found between the composition of radicals in the plasma and the etch selectivities in C 4 F 8 . Adsorption potentials of fluorocarbon radicals on Al (III) surface clusters have been calculated by using molecular orbital method, and sticking coefficients are estimated. Sticking coefficient of fluorine atom is the largest and decrease in order of F, C, CF, H, CF 2 , and CF 3 . Effects of hypervalence bonding at Al surface are discussed. Phenomena [1] of no film depositions in CF 4 RF plasma and film depositions in RF plasma of CF 4 mixed with H 2 were explained by hypervalence reactions of Al.

Dependence of Binding Energy on the Number of Atoms for Carbon Fullerences

Using a molecular orbital method with non-local-density and Harris-functional approximations, the bond lengths, binding energies and HOMO-LUMO energy gaps of the carbon fullerenes C n have been calculated on the assumption that all bond lengths in each fullerene are the same. The results are as follows. (1) The binding energies (per atom) of C n containing hexagonal and pentagonal rings are expressed as F–A/ n , where F and A are the binding energy (per atom) of an infinitely long carbon fullerene and the excess strain energy of pentagonal rings, respectively. (2) The values of A for C n ( n =24–36) are larger than those for C n ( n =60–120), because of the adjoining pentagonal rings in the former. (3) C 60 , C 70 , C 100 , C 90 , C 32 and C 120 , in decreasing order, have especially large HOMO-LUMO energy gaps and are expected to be less reactive.

Influence of space charge by primary and secondary streamers on breakdown mechanism under non-uniform electric field in air

We have clarified that the primary and secondary streamers have a significant impact on breakdown mechanism in air. In this paper, we focused on the highly-conductive channel progress breakdown, and considered the initiation condition of highly-conductive channel from the viewpoint of the secondary streamer initiation. We revealed that the highly-conductive channel initiation voltage in the case with the grounded electrode covered by a dielectric barrier was higher than that without the barrier. We suggested that the electric field relaxation effect of the residual positive space charges formed by the primary streamer is greater than that by the secondary streamer. The highly-conductive channel is initiated from the region where the electric field relaxation is weak.

Cross sections of charge transfer between uranium atoms and ions produced by autoionization

A technique to reduce charge transfer has been studied to cut enrichment loss during atomic vapor laser isotope separation. The charge transfer cross sections of uranium were experimentally examined. The cross sections of charge transfer for case A (between ground-state ions and atoms) and, cases B and C (between ions populated in some excited states and atoms) were relatively measured in the energy range of 100–2,000 eV by the cross-beam method. The charge transfer cross sections for cases A, B and C do not depend on impact energy, and the cross sections for case A are smaller than those for cases B and C. These results are in fairly good agreement agreed with theoretical predictions. It is confirmed experimentally and theoretically that the uranium charge transfer cross section depends on its initial ionic state, therefore ion loss by charge transfer will be reduced by controlling the initial state of the ions.