Investigation of void formation beneath thin AlN layers by decomposition of sapphire substrates for self-separation of thick AlN layers grown by HVPE

Void formation at the interface between thick AlN layers and (0 0 0 1) sapphire substrates was investigated to form a predefined separation point of the thick AlN layers for the preparation of freestanding AlN substrates by hydride vapor phase epitaxy (HVPE). By heating 50–200 nm thick intermediate AlN layers above 1400 °C in a gas flow containing H₂ and NH₃, voids were formed beneath the AlN layers by the decomposition reaction of sapphire with hydrogen diffusing to the interface. The volume of the sapphire decomposed at the interface increased as the temperature and time of the heat treatment was increased and as the thickness of the AlN layer decreased. Thick AlN layers subsequently grown at 1450 °C after the formation of voids beneath the intermediate AlN layer with a thickness of 100 nm or above self-separated from the sapphire substrates during post-growth cooling with the aid of voids. The 79 μm thick freestanding AlN substrate obtained using a 200 nm thick intermediate AlN layer had a flat surface with no pits, high optical transparency at wavelengths above 208.1 nm, and a dislocation density of 1.5×10⁸ cm⁻².     

An ab initio-based approach to the stability of GaN(0 0 0 1) surfaces under Ga-rich conditions

Structural stability of GaN(0 0 0 1) under Ga-rich conditions is systematically investigated by using our ab initio-based approach. The surface phase diagram for GaN(0 0 0 1) including (2×2) and pseudo-(1×1) is obtained as functions of temperature and Ga beam equivalent pressure by comparing chemical potentials of Ga atom in the gas phase with that on the surface. The calculated results reveal that the pseudo-(1×1) appearing below 684–973 K changes its structure to the (2×2) with Ga adatom at higher temperatures beyond 767–1078 K via the newly found (1×1) with two adlayers of Ga. These results are consistent with the stable temperature range of both the pseudo-(1×1) and (2×2) with Ga adatom obtained experimentally. Furthermore, it should be noted that the structure with another coverage of Ga adatoms between the (1×1) and (2×2)-Ga does not appear as a stable structure of GaN(0 0 0 1). Furthermore, ghost island formation observed by scanning tunneling microscopy is discussed on the basis of the phase diagram.     

A numerical study of 3D natural convection in a cube: effects of the horizontal thermal boundary conditions

A high-resolution, three-dimensional finite-difference numerical study of natural convection flows of a viscous fluid in a differentially heated cubical box is reported. The vertical sidewalls of the enclosure are maintained at constant temperatures of different values. The other vertical walls (the end walls) are thermally insulated. For the horizontal walls, two kinds of thermal boundary conditions are specified: adiabatic and perfectly conducting. Computations have been performed for an air-filled cavity for Rayleigh numbers of 10⁵ and 10⁶. The specific effects of the horizontal thermal boundary conditions on the flow structure are examined in detail. In the case of conducting walls, heat transfer through the horizontal walls enhances the convective flow activities. The numerically predicted velocity and temperature profiles in the symmetry planes are consistent with previous experimental measurements and computations.     

Electronic structures of the highest occupied molecular orbital bands of a pentacene ultrathin film

We report the energy dispersions of the highest occupied molecular orbitals (HOMO)-derived bands of a pentacene (Pn) thin film, whose in-plane structure resembles closely that of the ab plane of a low-density bulk Pn phase. Our present photoemission result indicates that the overlap of the pi-orbitals of adjacent Pn molecules is larger than what was expected from theoretical calculations. Further, of the two HOMO-derived bands, the large dispersion width of the band with higher binding energy suggests that this one mainly contributes to the bandlike charge transport in a Pn crystal.     

Polyaniline‐Induced CH Arylation of Arenes with Arenediazonium Salts

A reduced form of polyaniline has been shown to induce direct arylation of an arenediazonium salt with an arene (Gomberg–Bachmann reaction) to give the cross‐coupling product in moderate to good yields under mild conditions. Various arenediazonium salts and arenes, including heteroarenes such as furans, thiophenes, and pyrroles, are employed for the reaction. The most favorable combination of substrates is an electron‐poor arenediazonium salt with an electron‐rich heteroarene. Investigation of the mechanism by reactions with radical scavengers and experiments on kinetic isotope effects indicated the occurrence of a radical chain reaction initiated by one‐electron reduction of an arenediazonium salt by the polyaniline. Only 1 mol % (based on aniline tetramer) of the polyaniline is required for the cross‐coupling reaction to occur. This reaction proceeds under metal‐free conditions and with no need for photonic activation.     

Several contributions to soil compactibility induced by cyclic loading test

To investigate the soil compactibility during the cyclic loading, three different kinds of paddy field soils were prepared so as to clarify interrelationships among stress amplitude ratio, bulk density, soil water content and pore water pressure. The presetting values of specimen include the soil water content(percent dry basis) and bulk density of 25% d.b. and 1.1 Mg/m³ respectively. The relation between the number of cyclic loading and axial strain exhibited an asymptotically increasing trend, converging toward a specific value for each experimental condition. Possible effect caused by elastic–plastic characteristics could be recognized, when axial strain for 0.5 Hz excitation becomes greater than the one of 1.0 Hz under same stress amplitude ratio. When the stress amplitude ratio took 0.1, the absolute value of axial strain of 0.5 Hz was greater than the one of 1.0 Hz, whereas its decreasing trend was recognized in the sequence of silt, clay and silty sand. The qualitative relations between pore water pressure and number of cyclic loading were also examined to scrutinize the effect of effective water pressure to the soil compactibility.     

Buffers to suppress sodium dodecyl sulfate adsorption to polyethylene oxide for protein separation on capillary polymer electrophoresis

Although polyethylene oxide (PEO) offers several advantages as a sieving polymer in SDS capillary polymer electrophoresis (SDS-CPE), solution properties of PEO cause deterioration in the electrophoresis because PEO in solution aggregates itself, degrades into smaller pieces, and forms polymer-micelle complexes with SDS. We examined protein separation on SDS-CPE with PEO as a sieving matrix in four individual buffer solutions: Tris-CHES, Tris-Gly, Tris-Tricine, and Tris-HCl buffers. The solution properties of PEO as a sieving matrix in those buffers were examined by dynamic light scattering (DLS) and by surface tension. Preferential SDS adsorption onto PEO disturbed protein-SDS complexation and impaired the protein separation efficiency. Substantial adsorption of SDS to PEO was particularly observed in Tris-Gly buffer. The Tris-CHES buffer prevented SDS from adsorbing onto the PEO. Only Tris-CHES buffer achieved separation of six proteins. This study demonstrated efficient protein separation on SDS-CPE with PEO.     

A free-energy perturbation method based on Monte Carlo simulations using quantum mechanical calculations (QM/MC/FEP method): application to highly solvent-dependent reactions

This study describes the framework of the quantum mechanical (QM)/Monte Carlo (MC)/free‐energy perturbation (FEP) method, a FEP method based on MC simulations using quantum chemical calculations. Because a series of structures generated by interpolating internal coordinates between transition state and reactant did not produce smooth free‐energy profiles, we used structures from the intrinsic reaction coordinate calculations. This method was first applied to the Diels–Alder reaction between methyl vinyl ketone and cyclopentadiene and produced ΔG values of 20.1 and 21.4 kcal mol^?1 in aqueous and methanol solutions, respectively. They are very consistent with the experimentally observed values. The other two applications were the free‐energy surfaces for the Cope elimination of N,N‐dimethyl‐3‐phenylbutan‐2‐amine oxide in aqueous, dimethyl sulfoxide, and tetrahydrofuran solutions, and the Kemp decarboxylation of 6‐hydroxybenzo‐isoxazole‐3‐carboxylic acid in aqueous, dimethyl sulfoxide, and CH₃CN solutions. The calculated activation free energies differed by less than 1.8 kcal mol^?1 from the experimental values for these reactions. Although we used droplet models for the QM/MC/FEP simulations, the calculated results for three reactions are very close to the experimental data. It was confirmed that most of the interactions between the solute and solvents can be described using small numbers of solvent molecules. This is because a few solvent molecules can produce large portions of the solute–solvent interaction energies at the reaction centers. When we confirmed the dependency on the droplet sizes of solvents, the QM/MC/FEP for a large droplet with 106 water molecules produced a ΔG value similar to the experimental values, as well as that for a small droplet with 34 molecules. © 2010 Wiley Periodicals, Inc. J Comput Chem, 2011