Table of contents

The electrical conductivities of organic single crystal heterointerfaces are investigated. Electron transfer at the rubrene/7,7,8,8-tetracyanoquinodimethane (TCNQ) interface from the highest occupied molecular orbital of rubrene to the lowest unoccupied molecular orbital of TCNQ imparts conductivity to the interface. A conducting layer is formed at a rubrene-on-TCNQ heterointerface, but not at a TCNQ-on-rubrene heterointerface fabricated on a rigid SiO₂/Si substrate. The formation of an interfacial conducting layer requires a good contact between two single crystals; therefore, the experimental results might be explained by the poor adherence of rather thick TCNQ crystals to the rigid rubrene/SiO₂/Si system.

We report on the coherent control of terahertz (THz) waves emitted from coherent longitudinal optical (LO) phonons in a GaAs/AlAs multiple-quantum-well structure at room temperature using two-pulse excitation. When the time difference between the first pump pulse and second weak control pulse is increased, the intensity of THz waves from the coherent phonons is cancelled and enhanced by the control pulse, and the phase of THz waves shows periodic changes accompanied by a phase jump. The time-difference dependences of the intensity and phase of THz waves are explained in terms of the superposition of two coherent oscillations.

A variant tunnel field effect transistor structure called the binary tunnel field effect transistor (BTFET) for low voltage and near ideal switching characteristics is proposed. The BTFET relies on a binary tunneling distance (HIGH and LOW) for its operation to achieve a steep sub-threshold swing with a predicted range of 5 mV/dec. The transition of tunneling distance from HIGH to LOW state is a step-function of the gate voltage with the threshold voltage as a transition voltage. BTFET has a high on-current due to the high gate electric field and a large tunneling cross section area. An orientation dependent non-local band-to-band tunneling model was used to analyze the DC characteristics of the device.

In atomic force acoustic microscopy (AFAM) one exploits the contact resonances of the micro-fabricated cantilever, either as an imaging method where the contrast arises from the local variation of elastic and anelastic properties, or for local quantitative measurements. In this work we used AFAM in its spectroscopic mode, where contact-resonance curves are recorded as a function of the loading force P. From such curves the local contact stiffness and the contact damping were determined. The contact damping was found to have a power dependence on the load P when measured on the two different samples quartz-glass and nanocrystalline aluminum. We attribute this effect to micro-sliding of the sensor tip on the sample described by Mindlin for macroscopic contacts. We calculate the friction coefficient from the dissipated energy per cycle in the contact area to be 0.1, while the amplitude of the oscillatory sliding velocity was in the range of 0.05 m/s.

The polarization effects on the elastic dust grain collisions are investigated in unmagnetized complex dusty plasmas. The result shows that the polarization effect enhances the scattering cross section in dusty plasmas. In addition, it is found that the polarization effect on the scattering cross section increases with an increase of the collision velocity. The polarization effect increases with increasing Coulomb radius of interaction between thermal ions and the dust grain in high collision velocities.

The properties of a nitrogen (N)-related hole trap HC2, located approximately 0.15 eV above the valence band maximum of GaAsN, and their relationship with the density of ionized acceptors (N_A) in p-type GaAsN grown by chemical beam epitaxy are investigated using deep level transient spectroscopy and on the basis of the temperature dependence of the junction capacitance. At room temperature, N_A is found to show a linear dependence on N concentration under N- and H-rich growth conditions. Furthermore, a N-dependent sigmoid increase in junction capacitance is observed in a specific temperature range from 70 to 100 K, which is the same as in the case where HC2 is recorded. Such behavior is explained by the thermal ionization of HC2, whose density affects in great part the magnitude order of N_A, essentially for a N concentration higher than 0.15%. Concerning its origin, HC2 is strongly considered to act as N–H related acceptor state.

Transparent conducting oxide films composed of ZnO and SnO₂ were prepared on glass substrates by co-sputtering method. After surveying the electrical properties of the films according to the cationic composition and process conditions, we investigated the influences of hydrogen doping on the films' characteristic properties. With a moderate addition of H₂ in sputtering gas, carrier concentration of the films increased from 3.32×10¹⁹ to 5.22×10¹⁹ cm^-3, and the resistivity decreased from 7.23×10^-3 to 5.29×10^-3 Ω·cm. The increase in carrier concentration with H₂ can be attributed to the presence of hydrogen shallow donors as well as the formation of oxygen vacancies. However, the hydrogen addition contributed to the formation of SnO local states in Zn–Sn–O films, resulting in the decreases in carrier mobility and optical transmittance. Furthermore, changes in the electrical properties of the films upon annealing in vacuum or reducing atmosphere were investigated to elucidate the state of hydrogen atoms incorporated in Zn–Sn–O films.

Free-exciton luminescence dynamics at the surface and deep inside of a ZnO single crystal are investigated by one- and two-photon excitation mothods. The free-exciton lifetime is carefully evaluated at each position by considering the excitonic diffusion effect and the trapping process inside the sample. The obtained temperature dependence of the lifetime indicates that the photon recycling effect dominates the exciton lifetime inside the sample while the K-selection rule determines the lifetime at the surface.

NiO films were grown on a Pt substrate by radio frequency (RF) magnetron sputtering using a NiO ceramic target. A crystalline NiO phase with the [111] preferred orientation was formed for the films grown above 100 °C. Resistance switching behavior was not observed in the NiO films annealed in the air or in ambient O₂ after film deposition. However, the NiO films annealed in ambient N₂ exhibited resistance switching properties. The stability of the switching voltage was considerably influenced by the oxygen to argon ratio during film growth. In particular, the NiO film grown under an 8.0 mTorr oxygen partial pressure exhibited stabilized switching voltages (V_set∼1.45±0.20 V and V_reset∼0.62±0.09 V). Therefore, the control of the ambient gas pressure during the growth and annealing of the NiO films was important for obtaining good resistance switching properties.

One of the fundamental objectives for research and development of space solar cells is to improve their radiation resistance. InGaP solar cells with low base carrier concentrations under low-energy proton irradiations have shown high radiation resistances. In this study, an analytical model for low-energy proton radiation damage to InGaP subcells based on a fundamental approach for radiative and nonradiative recombinations has been proposed. The radiation resistance of InGaP subcells as a function of base carrier concentration has been analyzed by using the radiative recombination lifetime and damage coefficient K for the minority-carrier lifetime of InGaP. Numerical analysis shows that an InGaP solar cell with a lower base carrier concentration is more radiation-resistant. Satisfactory agreements between analytical and experimental results have been obtained, and these results show the validity of the analytical procedure. The damage coefficients for minority-carrier diffusion length and carrier removal rate with low-energy proton irradiations have been observed to be dependent on carrier concentration through this study. As physical mechanisms behind the difference observed between the radiation-resistant properties of various base doping concentrations, two mechanisms, namely, the effect of a depletion layer as a carrier collection layer and generation of the impurity-related complex defects due to low-energy protons stopping within the active region, have been proposed.

GaAs solar cells with the lower base carrier concentration under low energy proton irradiations had shown experimentally the better radiation-resistance. Analytical model based on fundamental approach for radiative and non-radiative recombination has been proposed for radiation damage in GaAs sub-cells. The radiation resistance of GaAs sub-cells as a function of base carrier concentration has been analyzed by using radiative recombination lifetime and damage coefficient for minority carrier lifetime. Numerical analysis shows good agreement with experimental results. The effect of carrier concentration upon the change of damage constant and carrier removal rate have been studied.

We have theoretically evaluated phase stability and electronic structure of Cu₂ZnSiSe₄ and Cu₂ZnGeSe₄ and compared the results with those of Cu₂ZnSnSe₄. The enthalpies of formation for kesterite (KS), stannite (ST), and wurtz-stannite (WST) phases of Cu₂ZnSiSe₄, Cu₂ZnGeSe₄, and Cu₂ZnSnSe₄ (CZTSe) were calculated by first-principles calculations. In these three compounds, the KS phase is more stable than the ST and WST phases. The theoretical band gaps of KS-type Cu₂ZnSiSe₄ (1.48 eV) and Cu₂ZnGeSe₄ (1.10 eV) are wider than that of KS-type Cu₂ZnSnSe₄ (0.63 eV). The valence band maximum (VBM) of KS-type Cu₂ZnIVSe₄ consists of antibonding orbital of Cu 3d and Se 4p, while the conduction band minimum (CBM) consist of antibonding orbital of IV ns and Se 4p. The VBMs of Cu 3d + Se 4p in Cu₂ZnSiSe₄ and Cu₂ZnGeSe₄ are similar to that in Cu₂ZnSnSe₄. Therefore, the energy levels of VBMs in Cu₂ZnIVSe₄ (IV = Si, Ge) do not change so much compared with that of CZTSe. On the other hand, the energy levels of CBMs of IV ns + Se 4p in Cu₂ZnSiSe₄ and Cu₂ZnGeSe₄ become higher than that in Cu₂ZnSnSe₄. These trends in the electronic structures are explained by the schematic molecular orbital diagrams of tetrahedral CuSe₄^7-, ZnSe₄^6-, and IVSe₄^4- (IV = Si, Ge, Sn) clusters.

Single crystals for 300 mm wafer are grown by horizontal magnetic Czochralski method. 300 mm wafers are made from the vertical samples cut from crystal along ingot axial direction. Micro defects in various defect regions are investigated with various measurement methods. In order to investigate the size of the defects, the locations of defects are identified, the wafers with the defects are cut in cross-sectional direction, and the sizes of the defects are measured by transmission electron microscopy (TEM). The voids more than 20 nm size exist in vacancy-rich region. Any as-grown defect is not observed by any available measurement tools in the region having nuclei of oxidation-induced stacking fault (P-band), pure silicon in a vacancy-dominant crystal region (Pv), and pure silicon in an interstitial-silicon-dominant crystal region (Pi). High sensitive laser scattering tomography system with the detection limit of 20 nm size is used to investigate as-grown defects in P-band, Pi, and Pv regions. It is concluded that there are no as-grown defects more than 20 nm size in P-band, Pi, and Pv regions.

The microcrystalline phase obtained by adopting a two-step rapid thermal annealing (RTA) process for rf-sputtered silicon films deposited on thermally durable glass was characterized. The optical properties, surface morphology, and internal stress of the annealed Si films are investigated. As the thermally durable glass substrate allows heating of the deposited films at high temperatures, micro-polycrystalline silicon (micro-poly-Si) films of uniform grain size with a smooth surface and a low internal stress could be obtained after annealing at 750 °C. The thermal stress in the Si films was 100 times lower than that found in the films deposited on conventional glass. Uniform grains with an average grain size of 30 nm were observed by transmission electron microscopy (TEM) in the films annealed at 800 °C. These micro-poly-Si films have potential application for fabrication of uniform and reliable thin film transistors (TFTs) for large scale active-matrix organic light emitting diode (AMOLED) displays.

Defect generation was usually predicted by using the V/G (where V is growth rate and G is axial temperature gradient at the interface of melt/solid) theory, but it was hard to get appropriate critical V/G value and the value could not show the distribution of grown-in defects. Otherwise, direct defect simulation is a very useful method of interpreting initial point defect behavior and micro void generation. In this research, the direct defect simulation was preformed with variable process parameters and optimized by comparing with experiment results. With optimized direct defect analysis, the critical V/G value was modified as 0.00155 cm² min^-1 K^-1. The critical pulling rate range was defined as that has low residual point defect concentration in silicon crystal, thus a high-quality wafer can be obtained at the critical pulling rate. The initial point defect distribution and the critical pulling rate range were analyzed by using direct defect model. Additionally, the generation of micro void density was also calculated with variable pulling rates and compared with experiment results. In this research, the initial point defect incorporation and the tendency of micro void generation were well explained by using direct defect model.

A 0.5 V six-transistor static random access memory (6T-SRAM) with ferroelectric-gate field-effect-transistors (Fe-FETs) is proposed and experimentally demonstrated for the first time. During the read and the hold, the threshold voltage (V_TH) of Fe-FETs automatically changes to increase the static noise margin (SNM) by 60%. During the stand-by, the V_TH of the proposed SRAM cell increases to decrease the leakage current by 42%. In case of the read, the V_TH of the read transistor decreases and increases the cell read current to achieve the fast read. During the write, the V_TH of the SRAM cell dynamically changes and assist the cell data to flip, realizing a write assist function. The enlarged SNM realizes the V_DD reduction by 0.11 V, which decreases the active power by 32%. The proposed SRAM layout is the same as the conventional 6T-SRAM and there is no area penalty.

We investigated sol–gel-derived hafnium dioxide (HfO₂) films on silicon substrates fired in air at 350, 450, 550, and 700 °C for 30 min using either formic acid (HCOOH) or nitric acid (HNO₃) solutions as a catalyst. At less than 450 °C, both films are amorphous and approximately 8–10 nm thick. Crystallization into the monoclinic structure (111) was found to occur at 560 °C in the HCOOH sol. In the HNO₃ sol, the crystallization into the monoclinic structures (111) and (111) occurs at 470 °C. The temperature-programmed desorption curves of the sol–gel-derived HfO₂ thin films using each sol solution are separated into five distinct H₂O desorption components caused by physically adsorbed H₂O, chemically adsorbed OH, and/or Hf–OH bonds in the HfO₂ film. On the basis of these components, a model is proposed to explain the H₂O desorption mechanism. The dielectric constant (relative permittivity: ε_HfO2) of the sol–gel-derived HfO₂ film was calculated to be 11 and the EOT was estimated to be 2.1 nm, which need to be improved. As an alternative gate insulator in advanced integrated complementary metal–oxide–semiconductor (CMOS) devices, the amorphous state of the sol–gel-derived HfO₂ film is promising for both sol solutions, if H₂O desorption can be accomplished and other defects eliminated.

The negative-bias temperature instability (NBTI) and positive-bias temperature instability (PBTI) of HfSiON/SiO₂ metal–oxide–semiconductor field-effect transistors (MOSFETs) with and without an ultrathin SiN cap layer were investigated. For the PBTI of n-channel MOSFETs, the dominant degradation mechanism is the electron tunneling from the Si channel and electron trapping in the pre-existing traps in HfSiON. The SiN cap layer does not make a significant difference in PBTI. For the NBTI of p-channel MOSFETs, on the other hand, both the electron trapping in HfSiON and the dissociation of Si–H bonds at the SiO₂/channel-Si interface (i.e., the interface trap generation) play a role and the SiN cap layer makes a significant difference in NBTI: the dominant degradation mechanism for the devices without the SiN cap layer is the electron trapping in HfSiON, whereas that for the devices with the SiN cap layer is the interface trap generation. This indicates that the interfacial SiN cap layer can effectively suppress the electron tunneling from the polycrystalline silicon (polySi) gate to HfSiON under the NBT stress.

In this work, we investigate the electronic contribution to local dielectric property of La₂O₃ and HfO₂ using cluster models. The relation between the coordinate number of metal atoms and their bonding energy shows a hint that hafnia takes the cubic structure by the incorporation of La₂O₃ in HfO₂. The local properties of polarizability and dielectric constant of La₂O₃ and HfO₂ are closely similar to each other. It is considered to be one of the reasons why the incorporation of lanthanum atoms does not lower the permittivity of HfO₂. We confirm this by the study of the local dielectric property of the HfLaO_x cluster model. We compare the dielectric properties around an oxygen atom and that between the oxygen atom and a next metal atom. Our results show that the contribution to the dielectric response from the bond regions is not so large.

Through electric-field-induced second-harmonic generation (EFISHG) measurement, we studied carrier injection and accumulation in a pentacene layer under the effect of dipole reversal in a ferroelectric poly(vinylidene fluoride–trifluoroethylene) [P(VDF–TrFE)] layer. The change in the electric field in the pentacene layer due to polarization reversal in the P(VDF–TrFE) layer was determined by direct probing. Using the value of remnant polarization of P(VDF–TrFE) obtained from displacement current measurement (DCM), we showed that the EFISHG response closely reflected the hole injection and accumulation at the pentacene/P(VDF–TrFE) interface induced by spontaneous polarization in P(VDF–TrFE).

The thermal diffusivities of tris(8-hydroxyquinoline)aluminum (Alq₃) and N,N'-di(1-naphthyl)-N,N'-diphenylbenzidine (α-NPD) films have been characterized quantitatively using a "rear heating/front detection-type" nanosecond thermoreflectance system. Alq₃ or α-NPD films sandwiched by Al films, Al/(Alq₃ or α-NPD)/Al three-layered films, were prepared by vacuum evaporation. Al acted as a reflective layer for pulse lasers in the thermoreflectance system. The nominal thicknesses of Alq₃ and α-NPD layers varied roughly from 30 to 100 nm. The thermal diffusivities of Alq₃ and α-NPD films were found to be (1.4–1.7)×10^-7 and 1.4×10^-7 m² s^-1, respectively.

The low-angle bent-shaped molecules with 1,7-naphthalene central core and alkylthio tails can form a novel hexagonal columnar phase and a dark B4 phase. The columnar phase has a large two-dimensional hexagonal lattice with edges of 65–75 Å and exhibits polar switching with spontaneous polarization along the column axis. Calculated from the density (∼1 g·cm^-3) and unit volume, the number of molecules that are necessary to fill a 4.6-Å stratum of each column were found to be ∼11. Such a large number of molecules can be accommodated only in the tube-like assembly, which may be the first example as formed by the usual bent-shaped molecule with a single alkylthio tail.

We propose a liquid crystal (LC) alignment method that can align LC molecules homogeneously on ion-beam-exposed indium–tin-oxide (IB-ITO) surface without coating alignment layer. To achieve high quality alignment conditions, the ion beam exposure energy and time have been found to be in the range from 50 to 1500 eV and from 1 to 300 s, respectively. On the IB-ITO surface treated under these conditions, a groove pattern is formed to be parallel to the ion beam exposure direction and optical anisotropy is induced to be parallel to the groove pattern, which causes the LC molecules to align. Moreover, thermal stability is ensured by baking the IB-ITO surface, due to effects such as an increase in polar anchoring energy. According to this method, the coating process of alignment layer is eliminated as well as alignment material is saved, which reduce fabrication cost.

Photoelectrical measurements were taken on InGaP (p⁺)-GaAs/InGaP-InGaP (n⁺) multilayers structures, formed by a sequence of nominally undoped InGaP/GaAs quantum wells, interposed between two p⁺ and n⁺ InGaP cladding layers. The heterostructures were grown through Low Pressure Metal Organic Vapour Phase Epitaxy, with liquid precursors for the III–V elements and growth conditions optimized for obtaining sharp interfaces and negligible ordering effects in InGaP. The experimental temperature dependence of the photoelectrical signal intensity exhibited peculiarities and anomalies which could lead to erroneous analysis of the perpendicular transport mechanisms, so that they are here critically discussed in the light of a partial depletion of the nominally intrinsic superlattice region of the p–i–n structure.

We propose a numerical model to estimate the self-sustained pulsation frequency (SSP frequency) for a two-section distributed feedback laser (TS-DFB). A modulation transfer function is derived from the rate equations for carriers and photons. The SSP frequency can be obtained from the singularity condition of the transfer function. A useful but simple systematic design procedure is proposed for investigating the effects of various structural parameters on the SSP condition. The device parameters varied in the analysis include carrier density, section length ratio, grating coupling coefficient, and the refractive index change caused by adding a shift-layer. The device structure used for the SSP experiments and analysis is a TS-DFB laser with a shift-layer. This type of lasers can have stable lasing mode with large side-mode suppression ratio such that the inherent mode instability in a conventional index-guide DFB laser can be eliminated in this structure. This can simplify the operation of the SSP laser. The analysis model can also be applied to other laser structures. The results of numerical analysis match well with the experimental data.

We fabricated InGaN-based resonant-cavity light-emitting diodes (RC-LEDs) in which one of the reflectors forming the cavity was made of Ag metal and the other was a ZrO₂/SiO₂ dielectric distributed Bragg reflector (DBR). Ag metal was deposited onto an InGaN epitaxial layer grown on a sapphire substrate. A ZrO₂/SiO₂ DBR was deposited onto n-GaN after a Ag/InGaN/sapphire sample was bonded to a silicon substrate using Au bonding metal and the sapphire substrate was removed by a laser lift-off process. Multiple emission modes with a narrowed spectrum were observed with corresponding dips in the reflectance of the resonant cavity, which was examined by microreflectance measurements. The output intensity of the RC-LED was higher than that of a conventional LED and the turn-on voltage increased with a slight decrease in series resistance. The output power of the RC-LED in a 1 ×1 mm² chip increased linearly with the injection of current up to a 500 mA.

We report the fabrication and characterization of an evanescent semiconductor optical isolator. The fabricated device exhibited 7.4 dB/mm optical isolation and improved gain saturation characteristics. The effect of the gain saturation on the semiconductor optical isolator was clarified by experimental results and comparison with previously reported propagation characteristics. The demonstrated device is appropriate for monolithically integrated semiconductor optical isolators, which will be useful for all-optical signal processing.

In this study a novel p-emitter/window capping configuration design applied to a p⁺–n In_0.5Ga_0.5P solar cell is developed. By grading the Ga and Al compositions in the interface between the p-In_0.5Ga_0.5P emitter and p-In_0.5Al_0.5P window layers, the output characteristics of the p⁺–n In_0.5Ga_0.5P solar cell are improved. It is found that the photoluminescence (PL) intensity is increased and the minority carrier lifetime obtained from room-temperature time-resolved (TR) PL measurement can be increased from 5.3 ns of the typical design to 7.0 ns, indicating that the application of compositional grading can improve crystal quality and the interface becomes smoother, thus reducing the nonradiative recombination losses. Both the short-circuit current and open-circuit voltage are increased correspondingly and the conversion efficiency is improved from 14.57% of the typical design to 15.32% of the new p-emitter/window configuration under one-sun air-mass 1.5 global illumination.

It is important to reduce the series resistance of a dye-sensitized solar cell (DSC) to improve its conversion efficiency. Here, a porous titanium (Ti)/dense Ti/aluminum (Al)/glass composite counter electrode for a DSC was fabricated. A porous and a dense Ti layer were deposited by a DC magnetron sputtering process. The dense Ti film was used as a protective layer against corrosion of the Al layer, which reduces the sheet resistance of the counter electrode. The solution treatment furthermore roughened the surface of the porous Ti layer. The composite counter electrode increased the fill factor and short-circuit current, resulting in the improvement of the conversion efficiency of the DSC. Our results indicate that it is important to introduce a porous material to the counter electrode of the DSC to improve the performance of the DSC and that the porous Ti film with rough surfaces is a promising material.

The generation and homodyne detection of continuous terahertz (THz) waves using a single photoconductive antenna excited by dual-wavelength continuous-wave laser light have been carried out. The THz waves emitted from the photoconductive antenna are returned and focused onto the same photoconductive antenna via a mirror. Sinusoidal variation of the homodyne current corresponding to the mirror displacement was observed. The homodyne current is caused by THz waves with the frequency of 1028 ±5 GHz. The homodyne current amplitude of 0.5–1.5 nA with a signal-to-noise ratio of 10–25 is obtained with the incident laser power of 4 mW and bias voltage of 10–30 V applied to the photoconductive antenna. The homodyne current amplitude is proportional to the square of the incident laser power up to 2 mW, which can be explained by considering the dependences of the dc photocurrent and dc photoconductance on the irradiating laser power.

A novel design for a broadband single-polarization single-mode photonic crystal fibers is proposed and numerically analyzed by using a finite-element method. It is shown that the single-polarization single-mode operation spectral region of a highly birefringent photonic crystal fiber can be strongly enhanced by introducing a suitably down-doped silica rod inside the fiber core. In the proposed design, the single-polarization single-mode guidance can be realized in the wavelength range from 1.0 to 1.8 µm with the confinement loss less than 0.12 dB/km.

Phase-type diffraction gratings, inscribed using an ultraviolet laser beam, were grown in a thin film of a photoreactive monomer base mixture over a period of several hours. The photopolymerization of monomers and subsequent molecular migrations generated both surface relief gratings and inner refractive index modulation. The changes in the diffraction intensity over time revealed the pronounced behavior of grating formation, which can be explained using a time-dependent model based on the Raman–Nath diffraction theory. Homogeneous irradiation with ultraviolet light during the growth enabled us to fix the diffraction characteristics of the gratings at an arbitrary moment in time, allowing us to tailor the resultant diffraction gratings according to a desired specification. This sequential process can have practical applications.

Slot waveguide, of which the optical field profiles mainly in the slot region, is attractive for sensing application by optical absorption because it has a high portion of optical field inside the slot region. A slot waveguide itself can be structured by utilizing buried hetero, ridge, rib, strip, and high-mesa structures. Among them, slot waveguide by using double high-mesa structure is attractive because it offers higher portion of optical field profile inside the slot region compared to the others. One critical issue is its propagation condition, as the slot waveguide structure may allow "directional-coupler" mode, which prevents proper utilization especially to the sensing application. To get rid of the directional-coupler mode propagation, here we propose a scheme by utilizing leaky mode for each high-mesa part. Under the condition of leaky mode for each high-mesa part, it is successfully confirmed that there is a propagation condition of which optical field mainly profiles in the slot region. Based on this propagation condition, more than 85% of the optical field profile can be achieved in the slot region. Moreover, under this condition, it is also confirmed that a very small radius of curvature of 10 µm is available, which is useful for long length integration into a compact area. We have also discussed about a mode-size converter to inject optical field into this double high-mesa slot waveguide from regular single high-mesa waveguide. Only 1.08 dB excess loss has been reported at mode-size converter.

Precise parameters are investigated for rectangular transmission gratings fabricated in a LiNbO₃ substrate for high-power terahertz-wave generation. Here, a simple setup is considered, where the pumping laser is injected into the grating surface from air or fused silica. The maximum diffraction efficiencies are 0.4 for air and 0.9 for fused silica. The designed gratings have deep grooves with a depth of about 0.5 µm and a width of 0.12–0.14 µm. It was found that the diffraction efficiency is much less sensitive to the groove depth than to the groove width.

A new violet phosphor was synthesized by sintering under the reduction condition from the Sr–Al–O:Eu powder oxidized from an ethylenediaminetetraacetic (EDTA) complex on a polycrystalline alumina plate. The crystalline phase of violet phosphor was determined to be Sr₇Al₁₂O₂₅ by the change in the crystallinity of the product with a change in the composition of the raw material. The emission wavelength of the phosphor was 410 nm for Sr₇Al₁₂O₂₅:Eu²⁺. For the composition of Sr:Al:Eu = 6.86:8:0.14 in the EDTA complex with sintering at 1400 °C, Sr₇Al₁₂O₂₅:Eu²⁺ was obtained only in the crystalline phase of Sr₇Al₁₂O₂₅ and exhibited the highest emission intensity of 410 nm in our experiment. The sintering temperature at sintering and composition of metals in the EDTA complex affected the emission wavelength change from 410 to 460 nm because of change in crystallinity. These changes indicate that the emission color in the range from violet to green is controlled by composition change or sintering temperature.

Ce:LuLiF₄ is evaluated as a fast scintillator using a storage ring free-electron laser operating in the deep-ultraviolet region. The relatively flat spectral response across its absorption bands makes this material an attractive scintillator for various excitation sources. Its decay time is acceptable for ignition timing in fast-ignition, inertial confinement nuclear fusion using laser. This report is the first systematic study on Ce:LuLiF₄ as a scintillator, wherein the storage ring free-electron laser is also shown as a powerful tool for material survey.

We report the growth of ytterbium-doped yettrium aluminum garnet (Yb:YAG) crystal fibers by the laser-heated pedestal growth (LHPG) method. The characteristics of Yb:YAG crystal fibers were analyzed by electron probe microanalyzer (EPMA) and X-ray diffraction measurements. The small crystal fiber diameter makes it effective for heat removal from a high-Yb³⁺-concentration and quasi-three-level gain medium. The length of this crystal-fiber-based high absorption can be shorter, which is eminently suitable for a multipass ring cavity to maintain cavity stability and mode symmetry. We successfully demonstrated a two-mirror multipass ring laser with 54.7% slope efficiency, which is higher than the 50.3% for a bulk Yb:YAG laser.

Since it was noted that quantum computers could break public key cryptosystems based on number theory, extensive studies have been undertaken on quantum cryptography, which offers unconditionally secure communication based on quantum mechanics. We investigate a quantum key distribution (QKD) scheme using macroscopic coherent light with optically pre-amplified direct differential detection. A transmitter "Alice" sends a series of two macroscopic nonorthogonal coherent states that partially overlap due to quantum noise. A receiver "Bob" amplifies and receives it with direct differential detection followed by a thresholding process. To avoid difficulties in detection, our scheme uses conventional direct differential photodetection, not single-photon detection or homodyne detection as in previous QKD protocols. System performance assuming some eavesdropping is evaluated, the results of which suggest that our scheme is usable for short or medium distance.

Mandel dip measurement was performed at 1550 nm using photon pairs created via spontaneous parametric down conversion in two separately located periodically poled lithium niobate waveguides. A visibility of 77.6% was obtained in fourfold coincidence counting. We experimentally clarified that the simultaneous generation of three pairs (one pair created in one waveguide and two pairs created in another waveguide) is the main source of accidental coincidence counts. Propagation loss before mixing two photons at a beamsplitter for observing photon bunching plays an important role in the determination of visibility. A large loss improves visibility, but also reduces the number of fourfold coincidence counts.

The effects of the fabrication methods and different capped oxide (SiO₂ and TiO₂) layers on the microstructure and magnetism of FePt thin films were studied. Both structural ordering (S ∼0.7) from the fcc FePt phase to the fct FePt phase and magnetic hardening were observed in the annealed FePt/SiO₂ thin films with a low substrate rotation speed (S_r = 1 rpm). However, only the annealed FePt/SiO₂ thin films prepared with a high S_r (10 rpm) exhibited isolated FePt grains separated by the grain boundary SiO₂, as revealed by transmission electron microscopy and magnetometry. Furthermore, similar results in microstructures and magnetic properties were obtained after replacing the capped layer with TiO₂. However, an enhanced order parameter (S ∼0.85) and a smaller FePt grain size (∼6.8 nm), which are promising characteristics for ultrahigh-density magnetic recording, were achieved in the annealed FePt/TiO₂ thin films; however, the annealed FePt/SiO₂ thin films exhibited a larger grain size (∼15 nm). This indicates that TiO₂ inhibits the grain growth of FePt more effectively than SiO₂.

The characteristics of thin-film transistors (TFTs) fabricated on pseudo-single-crystal (PSX)-Si thin films were examined. The variations of mobility were more than the theoretical values derived from the crystallographic orientation dependence of a bulk Si metal–oxide–semiconductor (MOS) transistor. To clarify the origin of this discrepancy, the relationships between the TFT characteristics and the crystallographic orientation of Si films in the channel region were investigated by using an electron backscattering pattern (EBSP) method. It was found that the surface orientation dependence for the PSX-Si TFT was different from that for a bulk Si MOS transistor, especially for the p-channel mode. A group of TFTs having a nearly {100}-oriented nucleus had a mobility close to those of simultaneously processed silicon-on-insulator (SOI) devices in the p-channel mode as well as in the n-channel mode. In contrast, a group of TFTs having a nearly {110}-oriented nucleus had a low and widely scattered mobility. The reason for these results is that twin boundaries with dislocations are easily generated in a grain grown from a {110}-oriented nucleus in order to compensate for the difference of the growth rates in different directions.

In this study, the effects of the via number and the current direction on electromigration characteristics in the dual-damascene Cu lines have been investigated. The results reveal an interesting difference in electromigration behavior of electron up- and down-flow directions on the multi-via structures. Increasing the via number results in a higher electromigration failure time and then reaches saturation for electron up-flow case. As for electron down-flow direction, the failure time is independent of the via number. Moreover, the failure time of Cu lines with via structure is lower than that without via structure. A higher current density at the triple junction site in the inner via is the possible mechanism, resulting in a shorter failure time and via-number independent. These observed effects are specific to Cu dual-damascene structures and can provide great technological implications in electromigration improvement.

A novel cylindrical surrounding-gate metal–oxide–semiconductor field effect transistor with high-κ gate dielectric and tri-material gate stack (TMGCSG MOSFET) is presented. The performance of the new structure is studied by developing physics-based analytical models for surface potential, electric field, and threshold voltage. It is found that TMGCSG MOSFET can effectively suppress short-channel effects and hot-carrier effects, and simultaneously improve carrier transport efficiency. It is also revealed that threshold voltage roll-off for TMGCSG MOSFET can be significantly reduced by adopting both a small effective stack-gate oxide thickness and a small radius silicon channel. The accuracy of the analytical models is verified by its good agreement with the three-dimensional numerical device simulator DESSIS.

Gold nanorods have a strongly polarized light at their longitudinal plasmon frequency that can be utilized to characterize colloidal gold nanorods and monitor their rotational dynamics in a bulk sample by polarized light scattering microscopy. By monitoring the time trace of the scattering polarization contrast, we could measure the polarization anisotropy of nanorods and their aspect ratio. More, we could gain insights into the rotational dynamics of nanorods and measure the rotational diffusion time on the microsecond time scale, which is an important parameter for various biological phenomena. The effects of aspect ratio and solution viscosity on the rotational time were determined. To carry out the measurements in a robust way, two-color laser illumination schemes were used and the correlation between both results was figured out. Results demonstrate the possibility of using polarized light scattering from gold nanorods to analyze the diffusion dynamics/conformations of biomolecules on the nanoscale.

An assembly of nanoparticles using a colloidal solution is promising for the fabrication of future highly integrated electron and photoelectronic devices because of low manufacturing cost, flexible substrates, and alternative methods that can overcome the limitation of top-down technology. We have successfully prepared two-dimensional arrays of nanocrystalline silicon (nc-Si) quantum dots with a uniform size of 10 nm. However, the area of two-dimensional arrays has been limited because of the problems of dissolution in water and agglomeration of nc-Si due to a high surface reactivity. The key issue is the surface modification of nc-Si particles. In this study, we have demonstrated the evaluation of surface modification states of nc-Si QDs by zeta potential and particle size distribution measurements. As a result of the optimization of the surface modification process, we have successfully obtained a well-dispersed nc-Si QD solution, namely, nanosilicon ink. Furthermore, we have successfully fabricated a two-dimensional array of nc-Si QDs using the Langmuir-Blodgett film method in the entire 1 ×1 cm² silicon substrate.

The microwave absorption properties of single-wall carbon nanotubes (SWNCTs) and barium ferrite nanocrystalline (SWCNT/BaFe₁₂O₁₉) composites with different doping ratios are investigated in the frequency region of 2–18 GHz. The transmission line theory is used to calculate the reflection loss properties, and microwave absorptive mechanism of the SWCNT/BaFe₁₂O₁₉ composites is discussed. The experiment results reveal that the microwave absorption properties of composite are very sensitive to the volume percentage of SWCNTs. Owing to the multiple absorptive mechanisms, the microwave absorption properties of composite are evidently improved. When SWCNTs are doped with 6 vol % of the sample volume, the maximum reflection loss of the SWCNT/BaFe₁₂O₁₉ composite with a 3 mm thickness reaches 30.79 dB at 10.5 GHz, and the range of resonance absorption peak below -10 dB is about 6 GHz.

In this study, the laser-induced thermal effect on the sensitivity of a scanning near-field optical microscope (SNOM) tapered probe is analyzed. In the analysis, the thermal effect can be considered as an axial force and is dependent on the temperature distribution of the probe. The Rayleigh–Ritz method is used to determine the sensitivity of the probe. According to the analysis, the sensitivity of the first three vibration modes increases when the thermal effect is taken into account. When the contact stiffness is low, the thermal effect on the sensitivity of mode 1 is particularly significant. The sensitivity of mode 1 increases with increasing taper angle and coating thickness of the probe. In addition, the effect of a SNOM probe with three different coating materials, Al, Au, and Ag, on the sensitivity of mode 1 is studied. The result shows that the highest sensitivity is obtained for the probe with an Al coating, whereas it is the lowest with a Au coating.

This paper studies the influences of the surface roughness and waviness on the static and dynamic flying performances of a thermal protrusion slider with flying height under 3 nm. Simulations show that the air bearing force and the contact force are proportional to the average roughness values of the surfaces, while the intermolecular force or the electrostatic force are the smallest for the smoothest surfaces when the minimum flying height is above a certain value. As a result, the total force on the slider is the largest on the smoothest surfaces in a certain minimum flying height region. When the minimum flying height is designed in this region, the flying ability of the slider is maximized. The energy analysis method is introduced to study the influences of the surface waviness on the dynamic performances of the slider. It is observed that the surface waviness may excite the slider to the bouncing state. The kinetic energy for maintaining the slider in the bouncing state comes mainly from the intermolecular force, while the air bearing force plays an important role in stabilizing the slider. The contributions of other forces, such as the friction force and the contact force, are also evaluated quantitatively. In order to improve the flying stabilities of the slider, the intermolecular force, the friction force and the electrostatic force should be reduced as much as possible.

Using thermodynamic analysis, the solid compositions of Mg in Mg_xZn_1-xO grown by halide vapor phase epitaxy (HVPE) are calculated in terms of the various input partial pressures of the gaseous species. It is revealed that the Mg composition shows temperature stability up to 1200 °C, while there is a strong dependence on the input VI/II ratio and the input H₂ partial pressure at high temperatures. Based on the thermodynamic calculations, the epitaxial growth of Mg_xZn_1-xO layers with different Mg compositions has been demonstrated. The thermodynamic model is found to accurately describe the experimentally observed dependence of Mg compositions on the input VI/II ratio that is used in the growth of Mg_xZn_1-xO.

Tin oxide (SnO₂) thin films were successfully electrodeposited on indium tin oxide (ITO) coated glass substrate from an acidic aqueous solution containing SnSO₄ at room temperature. Oxygen bubbling was employed so that dissolved oxygen serves as oxygen precursor. With O₂ bubbling and short deposition time, transparent films were obtained. The composition ratios of the films were measured by Auger electron spectroscopy. The n-type conductivity and the photosensitivity of the films were confirmed from photoelectrochemical measurement.

Ultrananocrystalline diamond (UNCD)/nonhydrogenated amorphous carbon (a-C) composite films were grown in vacuum using a coaxial arc plasma gun. From the X-ray diffraction measurement, the UNCD crystallite size was estimated to be 1.6 nm. This size is dramatically reduced from that (2.3 nm) of UNCD/hydrogenated amorphous carbon (a-C:H) composite films grown in a hydrogen atmosphere. The sp³/(sp³ + sp²) value, which was estimated from the X-ray photoemission spectrum, was also reduced to be 41%. A reason for it might be the reduction in the UNCD crystallite size. From the near-edge X-ray absorption fine-structure (NEXAFS) spectrum, it was found that the π^*C=C and π^*C≡C bonds are preferentially formed instead of the σ^*C–H bonds in the UNCD/a-C:H films. Since the extremely small UNCD crystallites (1.6 nm) correspond to the nuclei of diamond, we consider that UNCD crystallite formation should be due predominantly to nucleation. The supersaturated condition required for nucleation is expected to be realized in the deposition using the coaxial arc plasma gun.

In order to improve the uniformity of InGaAs lateral islands on Si grown by micro-channel selective area growth, we have investigated the dependences of initial InAs nucleation on the partial pressures of trimethylindium (P_TMIn) and tertiary butylarsine (P_TBAs) using the in situ monitoring of surface reflectivity. The high P_TMIn resulted in a short incubation period, a high density of nuclei, and vertical growth, suggesting that a high P_TMIn is suitable for obtaining single nuclei in each growth area, which is vital for the growth of single-domain crystals on Si. Laterally grown InAs nuclei, which are preferable for the lateral growth of InGaAs crystals that succeeds the initial nucleation of InAs, were obtained using either a low P_TMIn or a high P_TBAs, however, the effect of the latter was not significant. P_TBAs did not affect the incubation period. The density, uniformity, and shape of InAs nuclei can be controlled effectively by adjusting P_TMIn, but the uniformity and lateral shape could not be obtained simultaneously. We, therefore, devised a flow-modulated sequence and obtained InAs islands that grew in the lateral direction and almost filled the growth area with a single-crystal domain.

The alignment behavior of a crystal with a magnetic anisotropy of χ_c < χ_a under the imposition of a rotating magnetic field has been investigated by numerical calculation. The promotion of the crystal alignment when the projection of the magnetically hard axis on the magnetic field rotating plane is parallel to the magnetic field direction and its suppression when the magnetically hard axis is perpendicular to the magnetic field direction can be explained by the fact that the direction of the driving torque acting on the crystal minimizes the magnetic energy. Non dimensional alignment time normalized by the alignment time under the imposition of a static field is constant in the out-of-step region where the crystal cannot follow the magnetic field rotation during its alignment. The initial phase difference between the projection of the magnetically hard axis on the magnetic field rotating plane and its direction hardly affects the alignment time in the out-of-step region but strongly affects that in the synchronous region where the crystal rotation synchronous with the magnetic field rotation. A crystal aligns quickly if the initial phase difference is between 0 and 90° in the synchronous region. The minimum alignment time is the same as that under the imposition of a static field.

We have developed a top-gate type of field-effect transistor with a single-crystal SrTiO₃ channel and a DyScO₃ gate insulator stack consisting of an epitaxial interface layer and an amorphous breakdown barrier layer. We show that the zero-bias conductivity of the transistor channel is strongly affected by the presence of charged traps in the amorphous gate insulator. Low off-state current could only be achieved in devices that were fabricated at an oxygen ambient pressure of 10 mTorr. At lower pressures, metallic channel interfaces were obtained, even after post-annealing in air. When both epitaxial and amorphous DyScO₃ films were grown at 10 mTorr of oxygen, the on/off ratio of the field-effect transistors (FETs) reached 10⁶. We argue that when designing oxide FETs, it is necessary to consider not only breakdown characteristics, but also the charged trap density in wide-gap oxide insulators.

Al-doped ZnO (AZO) films with various Ga contents were prepared by magnetron co-sputtering in order to investigate the effect of Ga additions on the structural and optoelectronic characteristics of AZO films. The appropriate Ga doping level improved the crystallinity of the AZO films, investigated by X-ray diffraction analysis. The resistivity of AZO films decreased from 3.5 ×10^-3 to 8.1 ×10^-4 Ω cm by Ga doping at 2.1 at. %. The Hall mobility was improved by enhancing the polycrystalline growth of the films. The carrier concentration was increased by Ga doping, which was activated as an extrinsic donor. At a further increase in the Ga content of more than 2.1 at. %, the crystallinity and resistivity of the Ga-doped AZO films deteriorated. The optical band gap was increased, and the transmittance in the visible region was increased from 86.7 to 91.0% using the same level of Ga doping at 2.1 at. %.

Photoinduced current transient spectroscopy was used to investigate the defect states and capture kinetics of charge carriers for traps in low temperature polycrystalline silicon (poly-Si) films. A broad deep trap was found to be located 0.30 eV from the conduction band edge of poly-Si with capture cross section of 1.51×10^-15 cm². The variation of the trap capture kinetics with filling pulse time showed extended traps and linear arrays of traps, which might be grain boundary defects. Proton implantation and H-plasma treatment were used to improve poly-Si device characteristics, with traps more effectively suppressed by the former treatment. The ionized hydrogen atoms implanted into the poly-Si films are imputed to amorphize the defective poly-Si film with post-annealing enhancing re-crystallization, resulting films with fewer defects.

A simple model to inclusively understand radiated wave fields in slot-excited microwave plasmas has been developed. It is shown that the result of the simple model has a good agreement with that of a full wave model from the viewpoint of the mode spectra. The calculation cost of the simple model is considerably lower than that of a commonly used full wave model. In addition, it is also shown that the mode spectra exist in the broad region in the case of a few slots and in the narrow region in the case of many slots. The comparison with the experimental results supports that the physical phenomena such as mode jumps and hysteresis behavior observed in experiments are understood in terms of the mode spectra. Therefore, the simple model is useful to inclusively understand the antenna system and to design an antenna system with the complex slot geometry.

We present an experimental study of NO_x (NO and NO₂) formation from air and N₂/O₂/NO_x mixtures using a nonthermal microwave plasma device. The tests were performed considering the energy consumed to generate plasma gas and the flow rate of air. The results demonstrated that NO_x production was proportional to input power and the inverse of air flow rate. In the experiments that used N₂ and O₂ mixtures instead of air, the maximum NO_x concentration produced at equilibrium was 16.02 ×10¹⁶ molecules/J (0.66 ×10¹⁶ NO molecules/J and 15.36 ×10¹⁶ NO₂ molecules/J). Under the condition of the source gases consisting of NO and NO₂ of 970 ppm in the N₂ and O₂ mixture with a ratio of 79/21, the concentrations of NO_x generated were 5,571 and 5,320 ppm.

The plasma discharge in aqueous solution was scientifically studied and applied to template removal in mesoporous silica synthesis. Highly dispersed spherical mesoporous silica particles were synthesized by the ternary surfactant system containing the Pluronic P123 copolymer (EO₂₀PO₆₉EO₂₀), sodium dodecylbenzene sulfonate, and 1,1,2,2,3,3,4,4,4-nonafluoro-1-butane sulfonate, via the sol–gel method in acid solutions. The solution plasma process (SPP), instead of conventional thermal calcinations, was used to remove the template. The mechanism of the removal of the organic template occurred via oxidation by the hydroxyl radicals generated during discharge. The transformation of a mesopore structure from a disordered wormlike structure to a hexagonally arranged structure was observed by X-ray diffraction analysis and was confirmed by transmission electron microscopy. The results of the thermal analysis and functional group identification of mesoporous silica after SPP showed evidence of organic template removal. The surface area calculated using the Brunauer–Emmett–Teller (BET) theory and the mean pore diameter results could be used to evaluate the plasma efficiency, demonstrating that this method does not affect the pore size in the case of discharge in a solution of pH 3 compared with the results of thermal calcination. Hence, SPP was proved to be highly efficient for organic template removal, exhibiting short consumption time and less contamination.

Intense and reproducible Kr Kα X-rays (∼12.7 keV) have been generated via the interaction between a 3 TW laser pulse and a micron-sized Kr cluster target. A single-photon counting technique with an X-ray charge-coupled device (X-ray CCD) was used for measurements of the X-ray energy spectrum in a single shot. At a laser irradiance of 8 ×10¹⁶ W/cm², the averages and standard deviations of the total X-ray yield and Kα X-ray yield were equal to (6.6 ±1.1) ×10⁷ and (6.8 ±2.6) ×10⁵ photons/sr, respectively. When the X-ray energy spectrum was fit using a Maxwellian with an effective temperature, the temperature was estimated to be 1.62 ±0.08 keV. These results indicate that the X-ray energy spectrum was reproducible at 8 ×10¹⁶ W/cm².

A normally-off GaN-based metal–oxide–semiconductor field effect transistor (MOSFET) was fabricated using the Al_0.3Ga_0.7N/GaN heterostructure with a two-dimensional electron gas (2DEG) density of ∼1×10¹⁴ cm^-2 grown on a silicon substrate. The AlGaN layer in the gate region was fully recessed and the whole surface of the device was covered with a high-quality plasma-assisted atomic-layer-deposited (PAALD) Al₂O₃ layer, which plays the role of not only a gate insulator in the recessed gate region, but also a surface passivation layer in the ungated region between the source and the drain. The fabricated Al₂O₃/GaN MOSFET exhibited excellent device properties, such as a threshold voltage of 1.1 V extrapolated in the linear region at a drain voltage of 0.1 V, maximum drain current of 353 mA/mm, field-effect mobility of 225 cm²·V^-1·s^-1, and on-resistance of 9.7 Ω·mm, which are among the best values ever reported for GaN MOSFETs fabricated on silicon substrates.

This paper is concerned with the observation of phase defects in an extreme ultraviolet lithography (EUVL) mask using an EUV microscope developed by the University of Hyogo. It is very important to determine the type and size of defects on a substrate that are printable after deposition of a multilayer film. Thus, some mask blanks with programmed hole-pit defects with different widths and depths were fabricated by a new process. In addition, critical dimensions of a pit defect were investigated using the EUV microscope. As a result, 4.0-nm-deep hole-pit defects with widths larger than 35 nm were resolved. However, 4.0-nm-deep hole-pit defects with widths smaller than 25 nm were not resolved. On the other hand, 3.0- and 2.0-nm-deep hole-pit defects with widths larger than 60 nm were resolved. However, hole-pit defects with widths smaller than 40 nm were not resolved. Furthermore, the EUVM system was capable of clearly resolving 1.0-nm-deep hole-pit defects with widths larger than 70 nm. However, hole-pit defects with widths smaller than 60 nm were not resolved. From these results, we have determined the size of phase defects that are printable or not by observing phase defects that have various widths and depths on mask blanks utilizing the EUV microscope.

Electrical and physical characteristics of nickel disilicide (NiSi₂)-whisker defects in n-channel metal–oxide–semiconductor field-effect transistors (nMOSFETs) on Si(100) have been investigated. NiSi₂-whisker defects are easily generated in narrow-channel-width nMOSFETs with the <110> channel on Si(100) and anomalously increase the leakage current between the drain and the source. A NiSi₂ whisker elongates toward the <110> direction along the trench edge and pierces the channel region. These physical properties of NiSi₂-whisker defects were revealed by detailed failure analyses. The influence of the recessed depth of trench-fill oxides on NiSi₂-whisker defects was also investigated. Furthermore, it is found that trench-edge defects, such as Si(111) stacking faults, are generated in the <110> channel before the Ni silicide formation. These trench-edge defects were not observed in the <100> channel. We also propose a generation model for NiSi₂-whisker defects. The nucleation of NiSi₂ precipitates might be generated at trench-edge defects, and Ni atoms diffuse toward the <110> direction during the silicidation annealing. As a result, NiSi₂-whisker defects are generated toward the <110> direction at the trench edge.

To develop an ion milling system for multilayer mirrors in the extreme ultraviolet (EUV) wavelength region, the current density profile of a φ100-mm-ion beam has been measured by plural methods. The profile has been measured indirectly via milling thickness by an optical surface profiler (WYKO TOPO-2D) and an X-ray diffractometer. Direct current detection by a commercial Faraday cup and a newly developed array electrode sensor was also performed. These methods were compared and evaluated in terms of their experimental results. It was demonstrated that our array electrode sensor is useful for observing a wide ion beam profile with reasonable accuracy.

We studied the electrophoretic migrations of submicron particles in nonpolar inks sealed in narrow gap cells of 5.3 µm by using optical and current responses. We evaluated the mobility of particles by using the optical responses of the total reflection at interfaces between electrodes and the solvent in addition to simultaneously measuring current, from which the concentrations of ions and charged particles were analyzed. The mobility of the particles in the narrow gap was similar to that of the bulk ink, except for the case with no charge director and less dependence on the charge director concentration. We also analyzed how the mobility was distributed and how the particles interacted with the interface by using the optical responses.

It is necessary to form fine holes and grooves by machining in the manufacture of equipment in the medical or information field and the establishment of such a machining technology is required. In micromachining, the use of the ultrasonic vibration cutting method is expected and examined. In this study, I experimentally form microgrooves in stainless steel SUS304 by the ultrasonic vibration cutting method and examine the effects of the shape and material of the tool on the machining accuracy. As a result, the following are clarified. The evaluation of the machining accuracy of the straightness of the finished surface revealed that there is an optimal rake angle of the tools related to the increase in cutting resistance as a result of increases in work hardening and the cutting area. The straightness is improved by using a tool with low flexural rigidity. In particular, Young's modulus more significantly affects the cutting accuracy than the shape of the tool.

We derived the analytical formula of the motional capacitance C_a of a quartz-crystal tuning-fork tactile sensor by use of the conservation law of energy that the electrostatic energy in C_a is equal to the sum of the strain energy of its arm and the elastic energy of the material in contact with its base. In this analysis, the strain energy of the arm and the elastic energy of the material in contact with its base were obtained by applying the torsion spring model to the joint of the arm and the base, which were depicted as an L-shaped bar, and the elastic foundation to the materials in contact with its base, and the electrostatic energy of the arm was derived by applying the bimorph flexural model to the arm treated as a piezoelectric material. As a result of calculation, we found that the calculated value of the change in reciprocal motional capacitance Δ(1/C_a) of the quartz-crystal tuning-fork tactile sensor by our model is closer to the measured one for plastics than that given by the theoretical formula of motional capacitance of only one arm and Δ(1/C_a) is affected by the Young's moduli of materials in contact with the sensor's base.

There are a few concerns in dielectric modeling of biological cells by the finite-element method (FEM) to simulate their dielectric spectra. Cells possess thin plasma membranes and membrane-bound intracellular organelles, requiring extra fine meshes and considerable computational tasks in the simulation. To solve the problems, the "thin-layer" approximation (TLA) and the "effective medium" approximation (EMA) were adopted. TLA deals with the membrane as an interface of the specific membrane impedance, and therefore it is not necessary to divide the membrane region. EMA regards the composite cytoplasm as an effective homogeneous phase whose dielectric properties are calculated separately. It was proved that TLA and EMA were both useful for greatly reducing computational tasks while accurately coinciding with analytical solutions.

In this paper, we propose a microfluidic chip that measures the deformability of single cells by an impedance measurement method. The proposed chip is designed to differentiate the deformability of various cells by measuring the length of their stretched membrane indirectly according to the variation of the impedance after applying aspiration pressure to the cell membrane. The length of the stretched cell membrane is proportional to the applied pressure. Lengths of 18 and 21 µm were observed at the same suction pressure for human breast normal cells (MCF-10A) and caner cells (MCF-7), respectively. Electrical measurement was performed using an impedance analyzer at various frequencies. Results revealed that the impedance measurement method can be used to analyze the biomechanical characteristics of single cells, which indicates the state of malignancy of cells.

In order to realize a biophotonic device with biomimetic functions, we focus on the retinal in human eyes. A photoreceptive device was prepared using all-trans retinal immobilized in a gel film chitosan. This device was prepared with either indium tin oxide (ITO)-coated glass and gold-coated glass electrodes (ITO–Au), or using ITO electrodes for both ends (ITO–ITO). Each device was irradiated with 365 nm ultraviolet light for 20 s intervals at an applied voltage of 4 V. The photocurrent response was synchronized to the ON–OFF states of the ultraviolet light and was about 16 times higher for an ITO–ITO device than for an ITO–Au device. We searched for the optimum applied voltage because the ITO–ITO devices deteriorated on the second day at 4 V and found that 3 V produced a photocurrent response in both pre- and post-regenerated devices. Further ITO–ITO devices using three kinds of retinoids (all-trans retinal, all-trans retinoic acid, and all-trans retinol) were prepared. Photocurrent responses measured by a similar method persisted for 331 days using retinal, 313 days using retinoic acid, and 59 days using retinol. Furthermore, the photocurrent response was also observed in post-regenerated gel films of these ITO–ITO devices. These photoreceptive devices could be applied to bio-functional optical sensing or to future visual information processing devices.

The self-cleaning effect of super-hydrophobic surfaces has attracted the attention of researchers. Typical ways of manufacturing super-hydrophobic surfaces include the use of either dedicated equipment or a complex chemical process. In this study, a simple innovative filler-dissolved method is developed using mainly powder salt and rinsing to form hydrophobic surfaces. This method can produce large super-hydrophobic surfaces with porous and micro rib surface structures. It can also be applied to curved surfaces, including flexible membranes. The contact angle of the manufactured artificial hydrophobic surface is about 160°. Furthermore, water droplets roll off the surface readily at a sliding angle of less than 5°, resembling the nonwetting lotus like effect.

We proposed and demonstrated the single-cell isolation of bacteria using a microenclosure array with a structure composed of many micropillars. We fabricated the microenclosure array on a semiconductor wafer by electron beam lithography and the dry etching technique. The capturing frequency for single cells of Escherichia coli was approximately 50% using a 4-µm-width microenclosure array. We think that the proposed novel technique is very simple and useful for the single-cell isolation of many kinds of bacteria using a microenclosure array of optimum size.

The net polarization charge and induced internal electric field can positively or negatively affect the performance of InGaN-based optoelectronic devices, depending on the formation of the polarization charge. While positive polarization charges at the InGaN/n-GaN interface are supposed to affect solar efficiency positively by increasing the electric field, negative polarization charges at the InGaN/n-GaN interface negatively affect efficiency by forming an energy barrier. However, the solar performance is not improved in a beneficial case when the polarization charge reaches 10²⁰ cm^-3. In the detrimental case, the efficiency of p-GaN/i-In_0.1Ga_0.9N/n-GaN solar cells is seriously degraded from 2.75 to 0.49% as the polarization charge increases from 0 to 2.5×10¹⁸ cm^-3. The insertion of graded layers with a slowly increasing In content is proposed to relax the polarization, and shown to improve solar efficiency.

Based on the standard large-scale integrated circuit (LSI) process, sub-100 nm gate metal–oxide–semiconductor field-effect transistor (MOSFET) with thick gate oxide was fabricated. This was realized only by the modification of layout design, and no customization of the fabrication process was necessary. This unique designing technique is of great use in obtaining low-input-leakage MOSFET by advanced LSI process for high-performance analog applications.

The fabrication of ZnO nanopillars by electrodeposition and their application to an organic photovoltaic cell have been studied. The growth of ZnO nanopillars on indium-tin-oxide coated quartz substrates strongly depends on the conditions of electrodeposition. The use of a ZnO nanopillar in organic photovoltaic cells with a C₆₀/poly(3-hexylthiophene) interpenetrating interface resulted in improved performance.

Molecular orientation in a photo-crosslinkable liquid crystalline polymer (PLCP) film was explored by a thermal nanoinprinting technique using a SiO₂/Si mold with a line-and-space (L&S) pattern. A molecular oriented structure was confirmed by polarization optical microscopy and diffraction efficiency measurements, at which the reoriented direction was parallel to the mold lines. We demonstrated thermal nanoimprinting on prealigned PLCP films by linearly polarized UV (LPUV) and evaluated the optical characteristics of imprinted PLCP films. As a result, the prealigned PLCP films were reoriented by thermal nanoimprinting.