草部 浩一

For a zigzag crystal structure of calcium in the phase V, we estimated the superconducting transition temperature T c by the use of the Allen-Dynes formula. If we set the effective screened Coulomb repulsion constant μ* at 0.1, we obtain T c =16.37K at 120 GPa and T c =17.15K at 140 GPa. In order to clarify the origin of such high values of T c , we analyzed a partial electron-phonon coupling constant at each phonon mode. As the result we found that an optical mode at the G point strongly interacts with electrons and it induces the high T c in the phase V. The phonon mode can exist only in the particular structure like the zigzag structure. © 2009 IOP Publishing Ltd.

Stable oxygen sites on a PdO film over a Pd(100) thin structure with a (root 5 x root 5)R27 degrees surface unit cell are determined using the first-principles electronic structure calculations with the generalized gradient approximation. The adsorbed monatomic oxygen goes to a site bridging two twofold-coordinated Pd atoms or to a site bridging a twofold-coordinated Pd atom and a fourfold-coordinated Pd atom. Estimated reaction energies of CO oxidation by reduction of the oxidized PdO film and N2O reduction mediated by oxidation of the PdO film are both exothermic. Motion of the adsorbed oxygen atom between the two stable sites is evaluated using the nudged elastic band method, where an energy barrier for a translational motion of the adsorbed oxygen may become similar to 0.45 eV, which is low enough to allow fluxionality of the surface oxygen at high temperatures. The oxygen fluxionality is allowed by the existence of twofold-coordinated Pd atoms on the PdO film, whose local structure has a similarity to that of Pd catalysts for the Suzuki-Miyaura cross-coupling. Although NOx (including NO2 and NO) reduction is not always catalyzed by the PdO film only, we conclude that continual redox reactions may happen mediated by oxygen-adsorbed PdO films over a Pd surface structure, when the influx of NOx and CO continues, and when the reaction cycle is kept on a well-designed oxygen surface.

In this paper, we propose a diamond structure including stacking faults as a candidate atomicscale model for nanocrystalline diamond (NCD) with a small amount of non-se3 bonded region. The previous work reveals that NCD thin films show unusual elastic behavior, where the diagonal elastic constants decrease and the off-diagonal elastic constants increase as the grain size decreases. Using ab-initio calculations, the stable stacking-fault structures are obtained, whose partial density of states (PDOS) of the 2px orbital of the carbon atoms facing stacking faults behave as those of non-sp3 bonded atoms. Elastic constants of these structures are calculated from changes in the total energy by applying strains, which consistently explain the unusual elastic behavior of NCD films. This result shows that a combination of measurement and theoretical calculation is effective to analyze the relationship between microscopic structures and macroscopic property.

We theoretically investigate electronic transport properties of a distorted carbon nanotube and estimate conductance shift for the sub-band conduction. In this simulation nanotubes are distorted in such a way that they make shapes of various vibrational modes. We find that the shift depends on the way of distortion. Thus, if the distortion is induced by selective excitation of a phonon mode by a monochrome light, we may expect clear response in the electronic current through the biased nanotube, which implies its possible application as a current switching device. (C) 2009 The Japan Society of Applied Physics

To construct an optimization scheme for an extension of the Kohn-Sham approach, I introduce an operator form of the Coulomb interaction. This form is the sum of quadratic form pairs, which can be redefined in a self-consistent calculation of a multi-reference density functional theory. A detailed derivation of the form is given. A fluctuation term introduced in the extended Kohn-Sham scheme is expressed in this form for regularization. The present procedure also provides an exact derivation of effective negative interactions in charge fluctuation channels. Relevance to high-temperature superconductors is discussed.

Vibrations in ultrathin metallic films excited by ultrafast light pulses have been studied based on continuum mechanics. However, this paper shows that they are Γ-point phonon vibrations of plate-phonon modes. Ab initio and lattice dynamics calculations are made to compare Γ-point phonon vibrational frequencies with measurements obtained by the picosecond ultrasound spectroscopy. The standing-wave frequencies of specific Γ-point phonon modes of the slab model show good agreement with measurements without any fitting parameters. This study informs us of a limitation of the continuum-mechanics theory for explaining the mechanics of ultrathin metallic films. © 2009 American Institute of Physics.

For calcium in the phases IV and V, we estimated the superconducting transition temperature Tc by the use of the Allen-Dynes formula. Setting the effective screened Coulomb repulsion constant * at 0.1 in the formula, we obtained Tc=23.42K at 100GPa for Ca-IV and Tc=15.87K at 120GPa for Ca-V. In order to clarify the origin of such high values of Tc, first, we investigated the band character of electrons and found that the high Tc is not necessarily related to the so called s-d transfer. Then we analyzed the electron-phonon coupling at each phonon mode in Ca-V where the highest Tc in elements has been experimentally observed. As a result, we discovered that an optical mode at the point has the strongest electron-phonon coupling. Such phonon mode can exist only in the complex crystal structure of Ca-V, and the result shows that the high Tc seems to be closely linked with the complex crystal structures like Ca-IV and Ca-V.

Following the general rule on the negative effective U system, we have derived a microscopic mechanism of effective negative interaction from the first principles. The charge-excitation induced negative effective U is exemplified by s(1) electronic configuration, where the charge excitation by pure Coulomb interaction results in an energy gain for s(2) electronic configuration. Due to less positive nature of s(1) orbitals, we obtain a negative effective U system. In addition, the strategy gives us a method to find an effective local scattering channel with a negative interaction parameter. This purely Coulomb-originated mechanism for appearance of the effective negative interaction enables realization of charge-fluctuation-induced stabilization and enhancement of the high-temperature superconductivity.

By means of first-principles calculations, we studied stable lattice structures and estimated superconducting transition temperature of CaSi 2 at high pressure. Our simulation shows stability of the AlB 2 structure in a pressure range above 17 GPa. In this structure, doubly degenerated optical phonon modes, in which the neighboring silicon atoms oscillate alternately in a silicon plane, show prominently strong interaction with the conduction electrons. In addition there exists a softened optical mode (out-ofplan motion of silicon atoms), whose strength of the electron-phonon interaction is nearly the same as the above mode. The density of states at the Fermi level in the AlB2 structure is higher than that in the trigonal structure. These findings and the estimation of the transition temperature strongly suggest that higher Tc is expected in the AlB2 structure than the trigonal structures which are known so far. ©2008 The Physical Society of Japan.

A new mechanism of nano-catalyst generation based on the spinodal nano-decomposition in self-regenerating perovskite catalysts for automotive-emissions control is proposed. To demonstrate existence of the spinodal nano-decomposition in real perovskite catalysts, we performed first-principles calculations to evaluate the free energy of La(Fe(1-x)Pd(x))O(3) and La(Fe(1-x)Rh(x))O(3). The result indicates appearance of a spinodal region in the phase diagram of each material. Formation of nano-catalyst particles in the perovskite host matrix is crucial for the self-regeneration of perovskite catalyst. Based on the spinodal nano-decomposition model, possible materials are designed for new three-way catalyst with no contents of precious metal. (c) 2008 The Japan Society of Applied Physics

Based on the microscopic mechanisms of (1) charge-excitation-induced negative effective U in s(1) or d(9) electronic configurations, and (2) exchange-correlation-induced negative effective U in d(4) or d(6) electronic configurations, we propose a general rule and materials design of negative effective U system in itinerant (ionic and metallic) system for the realization of high-T-c superconductors. We design a T-c-enhancing layer (or clusters) of charge-excitation-induced negative effective U connecting the superconducting layers for the realistic systems. (C) 2008 The Japan Society of Applied Physics.

The anisotropic elastic constants of Co/Pt superlattice thin films were measured by resonant ultrasound spectroscopy and picosecond laser ultrasound. The experimental results were compared with the ab initio calculations for perfect superlattices. The calculated out-of-plane elastic constant C 33 is consistent with the measured value within 4%. However, the measured in-plane elastic constant C11 is smaller than the calculated value by 15-20%. © 2008 The Japan Society of Applied Physics.

Theoretical methods to determine the work function of metal surfaces using the density functional theory are reviewed. Evaluation of the Fermi energy relative to the vacuum level is usually done by utilizing rapid convergence of a local scalar potential obtained by subtracting the exchange-correlation potential from the total effective potential. We checked dependence on thickness d<Sub>eff</Sub> of the vacuum layer, which is 1/d<Sub>eff</Sub>, both for the Fermi energy and the vacuum level in finite slab models. Agreement between the theory and the experiments is shown for some typical metals.

We have reinvestigated the method to determine the work function of metals by the first-principles electronic structure calculation using the plane-wave expansion method with pseudo-potentials. Changing width of a vacuum layer of a slab model, we obtained the Fermi level measured from the effective Kohn-Sham potential at the center of the vacuum layer. By extrapolating the values of the Fermi level to the infinite vacuum-layer limit, the Fermi energy is determined. We propose a fitting function for the extrapolation. This function works well for determination of the highest occupied level and the lowest unoccupied level of materials with a surface relative to the vacuum level in the Kohn-Sham scheme. The obtained work functions for [100], [110] and [111] surfaces of three noble metals, Rh, Pd and Pt are comparable to the known theoretical estimation as well as the experimental values. © 2008 The Surface Science Society of Japan.

Modulated structures in phosphorous and iodine at high pressure have been studied by ab-initio calculations. In phosphorous the electrostatic interaction between the charge density, i.e. Ewald energy Eew and Hartree energy Eh play an important role for the creation of the modulated structure, whereas such roles of Eew and Eh are not observed in iodine. Results of a detailed study of the electronic properties are shown and discussed for phosphorous in comparison with those of iodine. Superconductivity in these materials has been discussed in relation to those properties.

We searched for the structure of calcium in phase V by a metadynamics based on the first-principles calculation, and found a new structure, where the positions of the calcium atoms in the unit cell have a zigzag pattern. Calculating the x-ray diffraction pattern of the new structure and comparing it with the experimental patterns of Ca-IV and Ca-V, we conclude that the new zigzag structure is a candidate of the structure of Ca-V.

Using the first-principles lattice dynamics, we have studied physical origin of enhancement in the superconducting transition temperature T-c of CaSi2 with structural phase transition. Optimization results show that CaSi2 has the AlB2 structure as an optimized structure above 17GPa. The electron-phonon interaction is enhanced, when CaSi2 takes the AlB2 structure compared to phase III. Especially, an E-2g Einstein mode and a softened optical B-2g mode are important.

We searched for the structure of calcium in phases IV and V by a metadynamics simulation based on ab initio calculations, and found two structures. One is a tetragonal lattice which consists of two helical chains along the c axis. The other is an orthorhombic lattice of four zigzag chains. We have calculated the x-ray diffraction patterns and enthalpies of the two structures discovered by our simulation. From comparisons of the patterns with the experimental x-ray patterns of the phases IV and V and from the pressure dependence of the enthalpies, we conclude that the structure with a helical pattern corresponds to phase IV and the structure with a zigzag pattern to phase V.

A self-consistent calculation scheme for correlated electron systems is created based on the density-functional theory (DFT). Our scheme is a multi-reference DFT (MR-DFT) calculation in which the electron charge density is reproduced by an auxiliary interacting fermion system. A short-range Hubbard-type interaction is introduced in a rigorous manner with a residual term for the exchange-correlation energy. The Hubbard term is determined uniquely by referencing the density fluctuation at a selected localized orbital. This strategy to obtain an extension of the Kohn-Sham scheme provides a self-consistent electronic structure calculation for the materials design. Introducing two approximations for the residual exchange-correlation energy functional, we have the LDA+U energy functional. Practical self-consistent calculations are exemplified by simulations of hydrogen systems, i.e.a molecule and a periodic one-dimensional array, which is a proof of existence of the interaction strength U as a continuous function of the local fluctuation and structural parameters of the system. © IOP Publishing Ltd.

We have studied the. xi-phase of solid oxygen using the generalized gradient approximation in the density functional approach. Calculations of total energies and pressures have been carried out for the prototype of diatomic. xi-phase and other hypothetical monoatomic crystal structures. The diatomic phase was found to be stable over a wide range of pressure (100-2000 GPa). The stacking of molecular layers is discussed in comparison with the available experimental data.

Encouraged by recent experimental facts of synthesizing materials which do not exist in nature, we introduce a type of ferromagnets, CaN and SrN, which have been proposed by first-principles calculations. These are half-metallic ferromagnets and they have magnetic moments of 1 mu(B) per chemical formula unit. Out of typical structures of binary compounds we have calculated, the rocksalt structure is the most stable form for both CaN and SrN. The majority of the magnetic moment of these compounds originates from the nitrogen sites since the p states of nitrogen are spin polarized. The mechanism of the ferromagnetism is discussed. Their formation energies have been calculated and the results show that it should be feasible to synthesize these materials. The structural stability of CaN has been confirmed by performing first-principles molecular dynamics simulations. We propose a synthesis process for CaN using the high pressure and temperature environment.

We have studied the magnetic properties of LaCo2 in a MgCU2 structure by first-principles electronic structure calculation. The Korringa-Kohn-Rostoker method is utilized to evaluate the spin-moment dependence of total energy at zero temperature. When the external magnetic field is zero, the ground state has no net spin as determined by local-spin density approximation (LSDA). The obtained result is the same when the energy functional of a generalized gradient approximation is used. The fixed spin moment method, which successfully confirms the metamagnetic transition Of YCo2 in LSDA, is applied to LaCo2. Our calculation suggests the occurrence of metamagnetic transition in LaCo2.

We investigated the magnetism in Ca and Sr pnictides by using the first-principles calculations. These compounds are half-metallic and ferromagnetic (FM) when they assume the zinc-blende structure at the equilibrium lattice constant. Ferromagnetism is induced by the spin polarization of the p-orbitals of the pnictogen atoms; Ca and Sr atoms have no magnetic moments, which is different from that of CrAs or CrSb with a zinc-blende structure. To confirm the mechanism of the ferromagnetism, we have calculated a hypothetical crystal - fcc-As with two additional electrons - and have shown that fcc-As has the same magnetic moment as CaAs with a zinc-blende structure. This means that the role of Ca or Sr atoms is to provide electrons with As atoms at the fee site and to sustain the distances between the As atoms and crystal symmetry. The FM exchange interactions between the pnictogen atoms are considered to exist in these lattices, which is briefly discussed.

To determine the effective spin model of the nano scale ferrimagnetic ring [Mn(hfac)(2)NITPh](6) abbreviated as Mn6R6, we propose a new criterion that the model should reproduce spatial configuration of the local spin densities of the reference system calculated by the hybrid density functional theory. The new prescription to Mn6R6 is successfully applied to clarify that the effective spin model consists of not only the nearest neighbor antiferromagnetic interaction but also the next nearest neighbor antiferromagnetic interaction which causes frustration in the spin system. The present result suggests that the spatial configuration of the spin densities gives an essential information for the determination of the effective spin model. (c) 2006 Elsevier Ltd. All rights reserved.

We have designed new materials, CaN and SrN in a rock-salt structure (RS). They are ferromagnetic nitrides. Their ferromagnetic states are quite stable. Our calculations for the formation energy and first-principles MD simulations suggested that RS-CaN is at least metastable and stable in normal condition. Based on our results we have proposed the synthesis process of RS-CaN. Our proposed process is (t) heat up alpha-Ca3N2 until it is transformed into beta- Ca3N2, (2) compress it above 50 GPa until 2Ca(3)N(2) -&gt; 2CaN+Ca reaction occurs, (3) cool them down to room temperature and (4) decompress them into an ambient pressure. We consider that this high pressure and high temperature synthesis is one of the hopeful method to synthesize new materials.

Half-metallic zinc-blende superlattices of CrAs/GaAs, MnAs/GaAs and VAs/GaAs have been calculated by a first-principles techniques to investigate the dependence of electronic properties and related ones on transition metals. For all cases half-metallicity has been preserved. The chemical shifts have been investigated and the difference at the interface is caused by the difference of the electronic configuration. The potential barriers has been estimated by using the data of the chemical shift. (c) 2005 Elsevier B.V. All rights reserved.

To perform the first-principles calculation of a non-uniform electron system with both localized and delocalized electrons, we have developed a calculational method based on a newly developed density-matrix functional theory. Our scheme is based on a recent rigorous result on uniqueness of an effective impurity Anderson model which reproduces a positive-definite density-density correlation as well as the single electron density. A tractable algorithm to determine the U term using estimation of an impurity problem by the transcorrelated method is proposed. (c) 2006 Elsevier B.V. All rights reserved.

We explore the unknown structure of phosphorus in phase IV (P-IV phase) based on first-principles calculations using the metadynamics simulation method. Starting from the simple cubic structure, we find a new modulated structure of the monoclinic lattice. The modulation is crucial to the stability of the structure. Through refining the structure further by changing the modulation period, we find the structure whose x-ray powder diffraction pattern is in best agreement with the experimental pattern. We expect that the modulation period of the structure in the P-IV phase is very close to that found in this study and probably incommensurate.

The edge states that emerge at hydrogen-terminated zigzag edges embedded in dominant armchair edges of graphite are carefully investigated by ultrahigh-vacuum scanning tunneling microscopy (STM) measurements. The edge states at the zigzag edges have different spatial distributions dependent on the alpha- or beta-site edge carbon atoms. In the case that the defects consist of a short zigzag (or a short Klein) edge, the edge state is present also near the defects. The amplitude of the edge state distributing around the defects in an armchair edge often has a prominent hump in a direction determined by detailed local atomic structure of the edge. The tight binding calculation based on the atomic arrangements observed by STM reproduces the observed spatial distributions of the local density of states.

We present the calculations of the phonon frequency and its line width over the Brillouin zone for simple substances based on ab-initio linear response theory and the evaluation of superconducting transition temperature. The substances we report on are FCC iodine, FCC lithium, and BCC tellurium at high pressures. In iodine the phonon dispersion reveals a phonon softening along the Sigma-line (Gamma-K direction) near the lower pressure boundary of the FCC phase. In lithium, the frequency softening is striking along the entire A-line (Gamma-L direction) and along the E-line (F-K direction) in the upper pressure boundary of the FCC phase. The FCC structure shows a phonon instability around 40 GPa and a turnover of the T-c before reaching this pressure. Similar phonon anomaly has been observed in BCC tellurium under high pressure. We discuss the origins of the soft phonon modes. The soft modes have much bigger line width than the other modes and contribute more to the electron phonon coupling constants. We have discussed the superconducting transition temperature on the basis of the Allen-Dynes formula, assuming phonon mediated superconductivity in those materials. The increasing phonon anomalies strongly dictate the direction of increasing T, in these materials. Calculated values of T, themselves show good agreement with experiments.

Size dependence of the magnetic properties in nanoscale ferrimagnetic rings is investigated by the numerical diagonalization of the Heisenberg model. The field derivative of the magnetization is drastically dependent on size of the ring if the magnetic system has frustration, although the character does not exist in the non-frustrated system. Our numerical data support this tendency that the effect of frustration causes the size dependence of the magnetic properties in the nanoscale ferrimagnetic ring. We also demonstrate that the size dependence caused by frustration is also found in behavior of the translational quantum number of the ground state. (c) 2005 Elsevier Ltd. All rights reserved.

The presence of structure-dependent edge states of graphite is revealed by both ambient and ultrahigh-vacuum (UHV) scanning tunneling microscopy and scanning tunneling spectroscopy observations. On a hydrogenated zigzag (armchair) edge, bright spots are (are not) observed together with a (root 3 X root 3)R30 degrees super-lattice near the Fermi level (Vs similar to-30 mV for a peak of the local density of states) under UHV. demonstrating that a zigzag edge is responsible for the edge states, although there is no appreciable difference between as-prepared zigzag and armchair edges in air. Even in the hydrogenated armchair edge, however, bright spots are observed at defect points, at which partial zigzag edges are created in the armchair edge.

Mechanical properties of fullerene encapsulated nanotubes called peapods are theoretically investigated. We compared two models of the peapods in which inter-molecular forces are estimated by the ab initio calculation or by the Lennard-Jones potential. The former model concludes stiffening in every vibrational normal mode, but the latter shows softening. Although stiffness change depends on the model, we are able to extract general tendency of each specific mode. The encapsulation tends to stiffen the compressing modes and radial breathing mode (RBM), while the bending mode and the stretching mode may be softened by the encapsulation, if fullerenes and a tube attract with each other.

Design of a nanometer-scale half-metallic wire made of a graphene structure on a diamond (100) surface was performed theoretically. The wire consists of a nano-graphene ribbon covalently bonded to the diamond surface. Due to spin polarization of edge states, the graphene structure shows a finite magnetic moment. Stability of the structure and existance of spin moment (1 mu(B) per an edge carbon atom) are confirmed by the first-principles calculation with the local-spin-density approximation. The half-metallicity appears when a shift (similar to 120meV) in the Fermi level is given.

Magnetization process of (S = 3/2, S = 1) ferrimagnetic spin chain with single-ion anisotropy D is investigated by numerical diagonalization. The magnetization process has a plateau at 1/5 of the saturation moment in the small D region. We find a quantum phase transition that the plateau once vanishes at a critical D and revives in the large D region. The critical value D, is estimated to be 0.9. In order to clarify the mechanism of this transition, we propose new trial states: a pentamer state in the small D phase and a trimer state in the large D phase. We find that representative ground state is changed from the pentamer state to the trimer one, while D increases across D,. This result suggests that the mechanism of the quantum phase transition is understood as the change of the representative ground state.

Edge effects on possible magnetism appearing in finite hydrogenated zigzag nanotubes are investigated using the DFT-LSDA calculations. Our result shows that two types of edges, i.e. a mono-hydrogenated zigzag edge (MHZE) and a di-hydrogenated zigzag edge (DHZE), can give magnetic moments in finite nanotubes. Due to curvature of the tube, however, the total magnetic moment Stot is smaller than an expected value given by a counting rule of 5tot, which holds for magnetic nanographite ribbons. In addition to magnetic nanotubes found by Okada and Oshiyama, we show that a (7, 0) tube and an (8,0) tube with both MHZE and DHZE are magnetic. The reactions to create hydrogenated (7,0) and (8,0) tubes are exothermic.