Table of contents

In this review, we introduce the production methods and applications of carbon nanotubes. Carbon nanotubes are now attracting a broad range of scientists and industries due to their fascinating physical and chemical properties. Focusing on the chemical vapor deposition (CVD) method, we will briefly review the history and recent progress of the synthesis of carbon nanotubes for the large-scale production and double-walled carbon nanotube production. We will also describe effective purification methods that avoid structural damage, and discuss the electrochemical, composite, and medical applications of carbon nanotubes.

The effect of the calcination process for cerium carbonate, a precursor of ceria, on the degree of synthesis and colloidal properties of ceria particles in aqueous media, which greatly influence shallow trench isolation (STI) chemical mechanical planarization (CMP) performance, was investigated. A two-step calcination process, consisting of decarbonation and crystal growth, resulted in a higher degree of synthesis for the same crystal size, a narrower particle size distribution, and better dispersion of the ceria particles, than conventional one-step calcination. These properties enhanced certain aspects of STI CMP performance, leading to a higher oxide removal rate, a better selectivity between oxide and nitride, and fewer defects, including remaining particles.

We have proposed an approach for growing robust ultrathin oxynitride using conventional thermal processes with the capability of preventing boron penetration. In this method, we obtain oxynitride with high nitrogen concentration (≈13 at. %) on the top and low interface state density (D_it=2×10¹⁰ cm^-2 eV^-1). The films demonstrate excellent properties in terms of low D_it, low leakage current, high endurance in stressing and superior boron diffusion blocking behavior. This method does not involve any additional capital equipment [such as decoupled plasma nitridation (DPN) or remote plasma nitridation (RPN)] or gas (NO or N₂O). In addition, it obtains high-quality oxynitride film with low thermal budget. Most importantly, this process is simple and fully compatible with current process technology. It would be important and interesting for process engineers engaged in the field of gate dielectrics. It is suitable for the next generation of ULSI technology.

The effect of nitrogen doping on the profiles of oxygen precipitates in Czochralski (CZ) silicon subjected to conventional annealing and rapid thermal annealing (RTA) has been investigated. After conventional high-low-high three-step annealing (1150 °C, 4 h + 650 °C, 128 h + 1050 °C, 16 h), it was found that an M-like oxygen precipitate profile existed in the cross section of a nitrogen-doped CZ silicon (NCZ-Si) wafer; however, this was not the case in conventional CZ silicon (CZ-Si) wafers. In contrast to conventional annealing, after rapid thermal annealing (RTA) followed by a low-high two-step annealing (800 °C, 8 h + 1050 °C, 16 h), an M-like oxygen precipitate profile existed in the cross section of the CZ-Si wafer but not in the NCZ-Si wafer. It is suggested that not only vacancies but also nitrogen doping have an influence on the oxygen precipitate profile. In addition, the relevant mechanism is discussed.

An InGaP/GaAs dual-emitter heterojunction phototransistor (DEPT) with a voltage-biased emitter is compared with a conventional heterojunction phototransistor (HPT) with a voltage-biased base. There are four (three) operating regions in the photocurrent–voltage characteristics of the DEPT (HPT): negative-saturation, negative-tuning, positive-tuning, and positive-saturation (cutoff, positive-tuning, and positive-saturation) regions. The power- and voltage-tunable optical gains are obtained in the DEPT, while only the voltage-tunable optical gain in the HPT. In addition, the maximum optical gain (32.3) obtained in the DEPT is 20-fold that (1.64) in the HPT. Experimental results show that 1) the voltage-tunable optical gain ranges from 0.8 to 1.64 with a gain-tuning efficiency of only 4.4 V^-1 for the HPT and that 2) the DEPT exhibits a voltage-tunable optical gain ranging from 12.9 to 32.3 with a maximum gain-tuning efficiency of 43.4 V^-1. Furthermore, a simple circuit model is developed to describe well the optical performance tuned by a voltage for both the DEPT and HPT. Experimental and modelling results indicate that the DEPT is very promising for optoelectronic applications when a low optical power is used.

A recess-gate-type n-channel depletion mode metal–semiconductor field-effect transistor (MESFET), a metal–oxide–semiconductor field-effect transistor (MOSFET) prepared by UV and ozone oxidation, and metal–insulator–semiconductor field-effect transistors (MISFETs) prepared by oxidation followed by nitrogen plasma treatment for different time durations with an electron cyclotron resonance system (oxinitridation) were fabricated using an ex-situ process and equally current controlled recessed wafers. The MISFETs, especially the longest-nitrided one, showed the highest pinch-off gate voltage (-1.5 V) and the highest peak transconductance (170 mS/mm) at a 0.9 V gate voltage. The MOSFET showed the largest drain current, whereas the longest-nitrided MISFET showed the smallest drain current, owing to the difference in applicable forward gate voltage. These were clearly confirmed with statistical data. The transconductance obtained reproduces that of our previous experiment, and its increase is due to the improvement of crystallographic order in the vicinity of the insulator/semiconductor interface. The different applicable gate voltage implies that the subsequent nitridation weakens the barrier effect of the oxidized layer. An AlGaAs layer, as the mother material for oxinitridation, improves it.

The properties of GaAs highly doped with Ge grown by molecular-beam epitaxy (MBE) and migration-enhanced epitaxy (MEE) have been studied. We report here that, although Ge doped GaAs films grown by conventional MBE at temperatures lower than 450 °C tend to become semi-insulating materials, a conductive Ge-doped GaAs film with an electron concentration as high as 2×10¹⁹ cm^-3 can be grown even at 300 °C by MEE. In MBE grown samples, a transition from n-type conductivity to p-type conductivity is observed when the Ge concentration of the films exceeds 1×10²⁰ cm^-3, whereas in MEE grown layers at 300 °C only n-type conductivity is observed regardless of the Ge concentration. The lattice constants of these MEE-grown samples are much larger than those of MBE-grown samples with equivalent Ge concentrations. These results indicate that Ge incorporation mechanism is quite different in MEE at low temperatures from that at high temperatures. The large lattice constant of as-grown MEE-grown layers suggests that a substantial fraction of Ge atoms occupy interstitial sites and act as donors. Postgrowth annealing results in marked reductions in lattice constant and electron concentration in MEE samples. The resulting lattice constants are even smaller than those of the n- and p-type GaAs films grown by MBE at high temperatures. This observed result can be explained in terms of (GaAs)_1-x(Ge₂)_x alloy formation in MEE samples during annealing.

The dynamics of zinc and oxygen adatoms deposited as atomic zinc and atomic oxygen on the C-terminated 6H–SiC(0001) surface are investigated using first-principles calculations. The results reveal that the original L site is unstable for the zinc adatom on the C-terminated 6H–SiC(0001) surface. However, the results reveal that the original L site is stable for the oxygen adatom on the C-terminated 6H–SiC(0001) surface. The migration barrier energy of an oxygen adatom on the C-terminated 6H–SiC(0001) surface is higher than that of a zinc adatom. Thus, the optimized growth conditions for the growth of ZnO crystals on the C-terminated 6H–SiC(0001) are Zn-polarity ZnO crystals grown under the stoichiometric growth condition.

The effect of a silicon substrate surface pretreatment on epitaxial iron silicide film formation on a Si(100) substrate with ion beam sputter deposition (IBSD) was investigated. In this study, two surface pretreatment methods namely, thermal etching (TE) and sputter etching (SE) with subsequent thermal annealing, were employed. The interface structure between a β-FeSi₂ film and a Si substrate was analyzed by cross-sectional observation using a transmission electron microscope (XTEM). High-resolution XTEM images showed that the lattice of the Si substrate is an almost perfect crystal immediately after the TE treatment. However, the TE treatment results in an undulated interface, and the deposited silicide contained coalesced β-FeSi₂ islands. On the other hand, the dislocations and stacking faults produced by radiation damage were observed near the Si substrate surface for the SE treatment. Even though this treatment produced defects, the interface of the SE-treated epitaxial β-FeSi₂(100) film had a smooth interface after the deposition at 973 K. It can be concluded that a moderate disorder of the silicon substrate surface treated by SE may well enhance the mixing of Fe and Si atoms for the epitaxial growth of β-FeSi₂.

Fluoride resonant tunneling diodes (RTDs) composed of CaF₂/CdF₂ heterolayers were fabricated in the V-grooved structures surrounded by (111) faces, which were preferred for the growth of the fluoride layers. The structures were formed by anisotropic etching with potassium hydroxide (KOH) or tetramethyl ammonium hydroxide (TMAH) on Si(100) substrates. In the case of KOH etching, RTDs with a high peak to valley current ratio (PVCR) over 10⁵ were obtained. Although the yield of RTDs whose active region was grown directly on the etched surface was very low because of surface roughness, an improved RTD structure, in which the Ca_xCd_1-xF₂ separation layer was inserted under the active region, exhibited higher yield. In the case of TMAH etching, the yield was higher than for KOH etching for directly grown RTD structures due to the presence of a smoother etched surface. These results show that the new method proposed in this work is an effective process to fabricate fluoride RTDs on Si(100) substrates.

Si spheres for spherical Si solar cells are produced by a dropping method. In the current state, the Si spheres are generally multicrystalline. In the dropping method, some impurities, mainly iron, which act as minority carrier recombination centers can be introduced in the Si spheres. To remove the impurities, phosphorous diffusion gettering (P-gettering) has been performed on spherical Si solar cells at 925 °C for 40 min. The effect of P-gettering was evaluated by solar cell performance and external quantum efficiency (EQE). The increase in the EQE of a long wavelength region was confirmed, which indicates that the deleterious impurities in a bulk region (not only the surface region) would be effectively reduced. Also, a minority carrier diffusion length estimated by the surface photovoltage method increased after P-gettering. Consequently, the efficiency of the spherical Si solar cell was improved. These results confirmed that P-gettering is effective to improve the solar cell performance of spherical Si solar cells.

The purpose of this study is to investigate the effects of electrical stress on the 1/ f noise behavior in n-channel metal–oxide–semiconductor transistors with ultrathin gate oxides. Even under a weak electrical stress, the drain current noise (S_id) of the device with a 1.4-nm-thick oxide was found to increase abruptly beyond a certain critical gate bias. This deteriorated noise property was proven to be from simultaneous increases in gate current noise (S_ig) and the correlation between S_id and S_ig, which were directly related to oxide trap generation and gate/drain current (I_g/I_d) ratio, respectively. Meanwhile, the increase in S_id in the device with a 2.3-nm-thick oxide after stress, with a comparable transconductance degradation, was relatively insignificant because of the device's smaller I_g/I_d ratio, even if the measured S_ig was comparable to that of the thinner oxide device. Consequently, the 1/ f noise degradation could be much more significant than the accompanying DC characteristic degradations in the thin gate oxide below 1.5 nm.

We have fabricated a highly efficient inverted bottom-emission organic light-emitting diode (IBOLED) based on an indium–tin oxide (ITO) bottom cathode deposited with an ultrathin 1 nm layer of Mg to promote electron injection. The threshold voltage of this IBOLED with a structure of ITO/Mg/Alq₃/NPB/WO₃/Al was 4.2 V and an efficiencies of 4.66 cd/A and 1.51 lm/W were achieved at an operational voltage of 8.9 V and a brightness of 940 cd/m². In comparison with an ITO/Alq₃ bottom cathode composition, a reduction in drive voltage from 13.8 to 7.8 V in voltage was obtained at 1 mA/cm². A charge-transfer dipole model is proposed to rationalize the enhanced electron injection.

A novel memory effect in metal–chalcogenide–metal structures was observed using chalcogenide films connected by two Al electrodes. A bias polarity-dependent switching between a high (RESET) and low (SET) resistance state was observed in current–voltage characteristics of the structure when sweeping voltage in forward and backward directions. The reversible SET/RESET switching was induced by voltage pulses and their polarity. The resistances of the structure changed over a range of approximately two orders of magnitude. In readouts of the resistance, we obtained its retention time for data over 17 days at room temperature.

Unique NAND flash memory cells with a booster line structure were designed to increase the channel voltage of a program-inhibited cell during program cycles. When a program voltage was applied to the selected word line, booster-line voltage coupled with control gate potential induced a high voltage in a program-inhibited channel. Because the turning on of the unselected cells was initiated by the booster line during programming, an unselected word line was maintained in the floating state without applying a pass voltage. Program disturbance in the NAND flash memory cell was decreased using a booster-line boosting scheme, and the cell's pass disturbance was effectively eliminated. The proposed unique NAND flash memory cell with a booster line can be used to improve the reliability of nanoscale NAND flash memories.

The frequency dependencies of the drain conductance (G_d) and the responsivity of InAlAs/InGaAs high electron mobility transistors were investigated using a network analyzer. The G_d exhibited the same frequency dependence as responsivity under 1.3-µm-wavelength laser illumination, indicating that the frequency dependence of G_d arises from the recombination of holes that have accumulated in the source region with the two-dimensional electron gas (2DEG). The holes were generated in the drain region by impact ionization under the strong electric field and drifted toward the source region. The drain conductance due to impact ionization can be expressed as Lorentz frequency dependence with f_3dB being the 3-dB-down frequency. The minority carrier (hole) lifetime, τ, was estimated using the relation 1/2 πf_3dB. The frequency dependence of G_d at several drain-to-source and gate-to-source voltages was investigated. The recombination lifetime for a system in which both electrons and holes co-exist was theoretically estimated, taking account of self-consistent solutions of both the Schrödinger and Poisson equations and the nonradiative Auger recombination mechanism. The experimental and theoretically estimated results clearly show that the frequency dependence of G_d is caused by the accumulation of holes in the source region and their recombination with the 2DEG. Based on the experimental and theoretical results, the distribution of holes in the channel and the multiplication factor due to the impact ionization are discussed in details.

Transmission characteristics of integrated linear dipole antennas fabricated on Si substrates were investigated for inter-chip signal transmission of wireless interconnections of Si ultra-large scale integrated circuits. Linear dipole antennas 1–6 mm long were fabricated on oxidized P-type Si substrates with resistivities of 10, 79.6, and 2290 Ω·cm. The transmission characteristics were investigated in the frequency and time domains. The transmission gain of a 4-mm-long dipole antenna fabricated on a 2.29 kΩ·cm resistivity Si substrate was -10 dB for a distance of 3 mm. A Gaussian monocycle pulse with a pulse width of 70 ps and a bandwidth of 20 GHz was transmitted and received successfully between Si chips with an air gap.

In this paper, we propose a novel technique for achieving high-density, high-speed and low-power on-chip bus lines using differential transmission lines. The feasibility of this technique is discussed with a two-dimensional electromagnetic simulator (Ansoft 2D Extractor) and time-domain measurements. Results show that the proposed bus line can transmit at over 12 Gbps. The proposed bus line can reduce wiring area by 30% compared with a conventional co-planar line.

A study of frequency dependent AC surface photovoltages (SPV) in n-type silicon (Si) wafers contaminated with an aqueous solution containing chromium (Cr) was carried out. Immediately after rinsing (5 min exposure to air), the Cr (Cr³⁺+3e^- →Cr) deposited on the surfaces of the wafers had already been converted into Cr(OH)₃ and/or Cr₂O₃. This Cr(OH)₃–Si contact forms a Schottky barrier on n-type Si. This gives rise to depletion and/or an inversion layer formed at the surface, resulting in the appearance of a frequency-dependent AC SPV. With exposure to air, the AC SPV in n-type Si wafers was reduced. This happens because Cr is oxidized and the Cr(OH)₃ becomes Cr₂O₃ through the reaction 2Cr(OH)₃ →Cr₂O₃+3H₂O, thus reducing the Schottky barrier height. Another factor is compensation by the positive oxide charge formed when the surface goes into accumulation, resulting in a suppression of the AC SPV in n-type Si. A schematic band diagram of the Schottky barrier was developed, and the barrier height was calculated to be 0.75 eV for n-type Si.

We report for the first time on the RF performance of a low-threshold AlGaN/GaN metal–oxide–semiconductor heterostructure field transistor (MOSHFET) with zirconium dioxide as the gate dielectric. Low gate leakage current of 5×10^-7 A/mm² and a threshold voltage which was only 1 V higher than that of an HFET were achieved. The RF power of these devices at 2 GHz was 14.32 W/mm at 50 V drain bias.

Single crystalline In₂O₃ nanowires have been synthesized by thermal evaporation of ball-milled In₂O₃ powders without any catalysis. The diameter and length range of the synthesized In₂O₃ nanowires are about 25 nm and 20–30 µm, respectively. Their X-ray diffraction pattern is indexed to bcc structure with a lattice constant of a=1.0126 nm. High-resolution transmission electron microscopy image shows that the inner part of the In₂O₃ nanowires is free of dislocations, and that any amorphous layers are not formed on the surface of the nanowires. In their PL spectrum, two peaks are observed in the ultraviolet region centered at 380 nm and in the visible region centered at 550 nm.

Twin dimers, mOAM5AMOm, form characteristic smectic phases, single-layer SmCA^s and antiferroelectric bilayer SmCA^b phases. Structural transformation from SmCA^s to SmCA^b was examined in two series of binary mixtures of 8OAM5AMO8/4OAM5AMO4 and 8OAM5AMO8/16OAM5AMO16. In this transformation, the frustrated SmCA^f with density modulation along the layer is formed and the size of modulation increases from 50 to 250 Å with variations in mixture contents from an edge near SmCA^s to another edge near SmCA^b. From polymorphisms shown by some mixtures, it was found that SmCA^s and SmCA^b are located at high-temperature and low-temperature edges, respectively, and that the SmCA^f phase is set between them.

Eddy current testing utilizing a cooled normal pickup coil and a high-T_c superconducting quantum interference device (SQUID) picovoltmeter was performed both experimentally and analytically. In the experiment, we successfully detected a small crack on the back surface of a Cu plate by moving the coil in an unshielded environment. First, we developed a method of avoiding the drift of the detected signal that was caused by the variance of lift-off. Next, we clarified the dependences of the detected signal on the excitation frequency and the thickness of the Cu plate. It was shown that an optimum frequency that maximizes the detected signal exists. Because this frequency changed with the thickness of the Cu plate, the frequency dependence could be used to estimate the depth of the crack from the surface of the Cu plate. The experimental results were analyzed taking into account the phase and amplitude of the signal field caused by the crack. Good agreement was obtained between experiment and analysis.

The sputtering pressure dependence of magnetic switching volume and the nature of magnetic intergranular interaction in CoSm/Cr films have been investigated using several magnetic characterization techniques. As the results, we found that magnetic switching volume strongly correlates to the nature and strength of magnetic intergranular interaction determined on the basis of dc demagnetization remanence and isothermal remanence curves. Also the sign of interaction field factor is closely related to the nature of magnetic intergranular interaction.

We have studied the magnetic domain structures and reversal processes of Fe thin films on two-dimensionally arranged hexagonal land-and-groove substrates by using spin-polarized secondary electron microscopy (SP-SEM). The coercivity difference between the land and groove was partly due to the difference in surface roughness between these areas. Weak domain wall pinning at the boundary between the land and groove areas occurred in the magnetization reversal process. The magnetic domain structures and reversal processes were found to depend on film geometry. It was also found that the anisotropy induced by film geometry for the hexagonal land-and-groove structure is weaker than that for the rectangular one. These results suggest that symmetry of film geometry and surface roughness play important roles for the magnetic domain structures and reversal processes in thin magnetic films.

A sensor using bending sensitive fiber (BSF) was studied for real-time long-distance sensing of submersion. The structure of the submersion sensor effectively used the relationship between buoyant force and bending loss of the BSF. A three-dimensional finite-difference beam propagation method (3D FD-BPM) was used to select the bending dimension of the proposed submersion sensor. The operation of the submersion sensor utilizing the BSF was also investigated using 3D FD-BPM. As results of experiments, the submersion sensor fabricated showed a change of optical power from -17 to -1 dB at 1550 nm when the selected sensing region was submerged. In addition, the BSF was easily connected to a single mode fiber (SMF) by fusionsplicing. The BSF exhibited a -1 dB loss when connected to a SMF at both ends.

We describe the operational characteristics of an ultrabroadband light source using a fiber Raman laser based on the stimulated Raman scattering in a single-mode optical fiber pumped by a self Q-switched Ti:sapphire laser for application in optical coherence tomography. Ultrabroadband output light covering the range from 770 to 1650 nm was obtained using a 500-m-long single-mode P-doped SiO₂ optical fiber pumped by the Ti:sapphire laser with a spectral width of 13 nm. When a white-light Michelson interferometer was constructed using the fiber Raman laser as a light source, a longitudinal spatial resolution of 1.7 µm was achieved.

Gd₂O₃:Eu phosphor particles with cubic and monoclinic phases were directly prepared by high-temperature spray pyrolysis from aqueous, polymeric precursor and colloidal solutions. The colloidal solution with nanosized silica was effective for the preparation of Gd₂O₃:Eu phosphor particles with a cubic phase, a high photoluminescence intensity and a good morphology. The phosphor particles prepared from colloidal solutions with nanosized silica obtained by hydrolysis had a small size, a narrow size distribution, a spherical shape and a filled morphology. The optimum doping concentration of nanosized silica particles was 0.5 wt % of the product particles. The phosphor particles prepared by spray pyrolysis from colloidal solutions at temperatures below 1580 °C had the photoluminescence spectrra of the cubic phase and a maximum photoluminescence intensity at a temperature of 1560 °C.

Transmission volume holograms recorded in methacrylate photopolymer films codoped with benzyl n-butyl phthalate (BBP) and silica nanoparticles are studied. It is found that BBP, which is a well-known plasticizer, can be directly mixed with methacrylate monomer and that a refractive index modulation as high as ∼0.006 is recorded with a BBP concentration of 36 vol %. It is also found that the additional dispersion of silica nanoparticles substantially suppresses polymerization shrinkage without increasing optical scattering loss.

Magnesium fluoride thin films were prepared by electron-beam evaporation and ion-assisted deposition (IAD). The effects of ion assistance and substrate temperature during deposition on the optical properties and microstructure were studied. The grain size, the crystallinity and the surface roughness of MgF₂ films deposited without ion assistance all decreased with substrate temperature. MgF₂ films deposited with IAD exhibited small grains, rough surfaces, fluorine deficiencies and large optical losses in the 200–500 nm wavelength range when bombarded with argon ions.

In this paper, we present a high-performance electrostatically actuated optical switch with a multiswitching function using stress-induced bending micromirrors and a seesaw structure. The use of a curved polysilicon seesaw structure and a torsional beam substantially lowers the electrostatic operating voltage of the optical switch and provides a multi-switching function. Large mirror deflection angles of 15° (with a mirror elevation of 310 µm) are obtained at a low operating voltage of 17 V. Furthermore, the device demonstrates a submillisecond switching time (<900 µs) and a low optical insertion loss (0.65 dB). The developed optical switch, which uses mirrors in an N ×N array, could redirect 4N optical inputs to 4N outputs. The compact nature of the optical multiswitch element renders it ideal for the construction of optical crossconnects (OXCs) with a large number of ports on a chip.

Cylindrical convex lenses are commonly used in fabricating fiber Bragg gratings (FBGs) by the phase mask method to increase the energy fluence of UV laser light and thereby enhance FBG productivity. We have found that lens aberrations can affect the efficiency of FBG fabrication by this method. Comparisons of the experimental measurements of FBG transmission minimum as a function of fiber position during irradiation by focussed excimer laser light with ray tracing calculations demonstrate the effect of spherical aberration, particularly for lenses with shorter focal lengths.

We proposed a novel method of constructing a chip polarization beam splitter (PBS) based on two-dimensional metallic photonic crystal (2D-MPC) structures that can be realized using the industrial-based semiconductor IC Cu-interconnect technology. The performance of such a PBS at wavelengths for optical telecommunications is evaluated by the finite-difference-time-domain (FDTD) simulation method. This 2D-MPC PBS device uses three rows of one-dimensional (1D) arrays (3×N-MPC-2D-MPC) as a basic building block to construct the PBS. Using this architecture, the incoming transverse magnetic (TM) and transverse electric (TE) waves can be splitted with very high splitting efficiency (polarization ratio over 1000 for both the TM and TE modes). The implication of our scheme is that the fabrication of such a 2D-MPC PBS is totally compatible with current commercial complementary metal oxide semiconductor (CMOS) fabrication technology. This makes it a ready-to-use technology for the fabrication of a chip-base 2D-MPC PBS.

A scanning laser line source (SLLS) technique is simulated numerically using a finite element method in this study, and changes in amplitude and frequency content have been observed for ultrasound signals generated by laser scanning over a large aluminum block containing a small surface notch. The laser generation of surface acoustic waves with the frequency range from 4 to 15 MHz in an elastic material is modeled using a thermoelastic model for finite element analysis, where a transient heat source is employed to represent a pulsed laser source. The experimentally observed SLLS amplitude and spectral signatures are shown to be captured very well by this model. In addition, the possibility of utilizing the SLLS technique to size surface notches that are sub-wavelength in depth is explored.

An optical frequency measurement system based on an octave-spanning optical frequency comb generated by a chirped-mirror-dispersion-controlled mode-locked Ti:Al₂O₃ laser and a photonic-crystal fiber is developed. All of the modes of the octave-spanning optical frequency comb are frequency-stabilized to a microwave frequency standard, where the carrier-envelope offset frequency is phase-locked with self-referencing of the comb. We investigate the methods of controlling carrier-envelope offset frequency in a chirped-mirror-dispersion-controlled mode-locked laser. The rotation of a pair of chirped mirrors is useful for setting the bias of carrier-envelope offset frequency. Although our mode-locked laser has a low pulse-repetition frequency of 150 MHz, a high signal-to-noise ratio in beats results in the direct measurement of beat frequency with a laser to be measured using a frequency counter, and enables us to phase lock carrier-envelope offset frequency merely by using a mixer analogously without the need for a prescaler, with a servo bandwidth at approximately 500 kHz. The uncertainty of our optical frequency measurement system, besides the uncertainty of microwave reference frequency, is 4×10^-14, and is limited by the uncertainty of the rf synthesizer used for phase locking and by that of the beat frequency measurement. Frequency measurements of an iodine-stabilized frequency-doubled Nd:YAG laser at 532 nm, an iodine-stabilized He–Ne laser at 633 nm and a rubidium two-photon-absorption stabilized extended-cavity laser diode at 778 nm are conducted. The results contributed to the revision of the practical realization of the metre adopted by the International Conference on Weights and Measures (CIPM) in 2001.

We investigate how miscellaneous parameters influence the velocity of the disclination line in a bistable chiral splay nematic liquid crystal (BCSN LC) device, using a model derived from the relationship of the free energy loss rate with dissipation energy. The disclination velocity is consistently coupled with the free energy difference between two stable states, rotational viscosity, and damping factor, in which the free energy difference is the most influential. The relationships between the energy difference and the parameters of the BCSN LC device were elucidated by examining the portrait of bistable curves. The calculated velocities of the disclination motion were also confirmed by comparing them with the measured results.

This paper proposes a wavelet-based prepossessing method to improve the detecting capacity of a blob-Mura-defect-detecting algorithm. The non-uniformity of the background region is eliminated by replacing the approximation coefficients with a constant value, and the brightness difference between the background region and defect regions is increased by multiplying the detail coefficients and a weighting factor. The proposed method can perfectly control the detectable defect level by properly selecting the defect detecting level. Experimental results demonstrate that the proposed method can effectively enhance blob-Mura defects in thin film transistor liquid crystal display panels.

The reflectivity, threshold energy, and Brillouin shift of a stimulated Brillouin scattering (SBS) were measured in new SBS liquid media. The new liquids consist of four heavy fluorocarbon (FC) liquids and six perfluoropolyether (HT) liquids. Among them, FC-77 and HT-70 were found to be good candidates to replace FC-75 which has been known to be a good SBS medium up to now.

An optical isolator with a TiO₂/magnetic garnet waveguide obtained using a nonreciprocal phase shift was studied. The optical isolator can be operated in a unidirectional magnetic field owing to an interferometer with distinct layer structures.

Violet-blue photoluminescence (β-band emission) of twofold-coordinated tin (O–Sn⁰–O, >Sn⁰) atoms in silica (SiO₂) glasses, synthesized via a sol–gel route, was observed with a large band tail to longer wavelength. The presence of >Sn⁰ centers more easily enabled us to produce Sn-associated E'-centers with X-ray irradiation, which is expected for a photosensitivity similar to that in the case of Ge-doped SiO₂ glasses fiber. In the initial state of the new fabrication process, stannic oxide (SnO₂) nanocrystals were precipitated in SiO₂ glasses and an adequate annealing of SnO₂–SiO₂ glass ceramics in H₂ gas led to the decomposition of SnO₂ nanocrystals and concurrently the production of >Sn⁰ point defects. Although the quantum efficiencies of the photoluminescence were very low to be 0.95% (external) and 1.3% (internal) under near-ultraviolet excitation owing to the triplet nature of the emissive level of >Sn⁰ center, we could clarify the photoluminescence properties, by means of time-resolved fluorescence spectroscopy, and the >Sn⁰→Sn-E' conversion.

The broadband Raman conversion of femtosecond-light pulses to the anti-Stokes side is demonstrated. A 130 fs light pulse from a Ti:sapphire-based regenerative amplifier system is divided into two light pulses. They are focused on a SrTiO₃ crystal with a finite cross angle. Over 24 higher-order coherent anti-Stokes Raman scattering signals are observed, which cover from near infrared to green. These higher-order coherent anti-Stokes Raman scattering (CARS) signals come from the resonant excitation of two-phonon waves. The results offer a possibility of an all solid-state anti-Stokes Raman shifter covering an extremely broadband tunable range.

Thin films of bismuth titanate are deposited on indium–tin-oxide (ITO)/glass substrates by RF magnetron sputtering for 60 min at room temperature using a Bi₄Ti₃O₁₂+4 wt % Bi₂O₃ ceramic target, and they are annealed at various temperatures ranging from 600 to 725 °C by a rapid thermal processing for 10 min. The experimental results indicate that the Bi₄Ti₃O₁₂ films annealed at 675–700 °C exhibit superior electrical characteristics compared to those of other ferroelectric materials such as BaTiO₃, PbTiO₃, and SrTiO₃ films, including a higher dielectric constant, a higher polarization charge density, and a more stable leakage current density. In addition, a high optical transmittance of the film was obtained at an annealing temperature of 650 °C. The results suggest that Bi₄Ti₃O₁₂ thin film is a candidate worthy of consideration as the insulating layer material of AC thin film electroluminescence devices.

We investigated a series of bulk, ceramic samples of Si-added bismuth layered-structure compounds, SBTS, prepared by alloying SrBi₂Ta₂O₉ with Bi₂SiO₅ or Bi₄Si₃O₁₂ using a solid-state reaction method. X-ray and neutron powder diffraction were carried out to determine the crystal structures. Structural information is correlated with the measured data of the ferroelectric properties. We find that remanent polarization and coercive electric field increase with the addition of Bi₂SiO₅ or Bi₄Si₃O₁₂ in SrBi₂Ta₂O₉. The dielectric constants of SBTS increase markedly with increasing temperature. It is found that the ferroelectric behavior is related to the distortion and tilting of the TaO₆ octahedron that are effected by variations in Bi and Si contents, respectively, in the SBTS samples.

The optimum ferroelectric film thickness in metal–ferroelectric–insulator–semiconductor (MFIS) structures is investigated, in which 80- to 560-nm-thick (Bi,La)₄Ti₃O₁₂ (BLT) films are deposited on HfO₂ buffer layers using a sol–gel spin-coating method. It is found from electrical characteristics of MFIS diodes as well as MIS diodes that the HfO₂ layers act as excellent barriers for suppressing both leakage current and atom interdiffusion when they are annealed in a rapid-thermal-annealing furnace at 900 °C for 1 min in O₂ flow. In MFIS diodes, the memory window width in capacitance–voltage (C–V) characteristics is found to increase from 0.2 to 1.6 V, as ferroelectric film thickness increases from 80 to 560 nm. On the basis of these results, the relationships among memory window width, ferroelectric film thickness, and the optimum applied voltage are discussed. Finally, it is shown from the capacitance change measured over 24 h that data retention characteristics are excellent in samples with BLT films thicker than 240 nm.

Pb(Zr,Ti)O₃ (PZT) films were prepared on Si substrates at 600 °C by metalorganic chemical vapor deposition (MOCVD). Crystallinity of tetragonal PZT film was dramatically increased by making Pt bottom electrode layer on the Si substrate. Based on this observation, we succeeded in directly patterning PZT films having quite different crystal quality, i.e., well-crystallized (100)-/(001)-preferentially-oriented ones, and poorly-crystallized randomly-oriented ones, by pre-patterning Pt bottom electrodes. The achievement of the patterning of PZT films down a size of 1 µm was determined by scanning probe microscopy using a piezoresponse technique. As a result, we demonstrated the direct patterning of PZT films on Si substrates for the first time and introduced the possibility of directly patterning PZT films with good piezoelectric properties without post-deposition processes such as etching.

The characteristics of epitaxial γ-Al₂O₃ film deposited by molecular beam epitaxy (MBE) on a Si substrate have been studied for its application to quantum devices with different thicknesses. The epitaxial growth properties and surface morphology of the film were studied by in situ reflection high-energy electron diffraction (RHEED) and atomic force microscopy (AFM). Epitaxial γ-Al₂O₃ films on Si substrates with film thicknesses ranging from 2 to 10 nm exhibited an appropriate surface flatness. We observed that the epitaxial γ-Al₂O₃ films exhibit appropriate dielectric properties (6–12 MV/cm) and very low leakage currents. From the electrical characteristics, we observed Fowler–Nordheim (F–N) tunneling phenomena with a large band offset. The same properties in thick epitaxial γ-Al₂O₃ films may be suitable for quantum tunneling applications.

Thin PbZrO₃ (PZ) films were investigated as a buffer layer to improve fatigue endurance of Pb(Zr_0.45Ti_0.55)O₃ (PZT) thin films. The PZ thin films were deposited on Pt₍₁₁₁₎/Ti/SiO₂/Si₍₁₀₀₎ substrates by RF magnetron sputtering from a loose powder target containing a mixture of PbZrO₃ and ZrO₂ powders. The PZT thin films on the PZ buffer layer were obtained by sol–gel spin coating. The PZ buffered PZT films had well crystallized, uniform and dense microstructure with partial (111) orientation. The P–E hysteresis measurements indicated comparable or slightly lower remnant polarization values (2P_r=50–70 µC/cm²) and higher coercive field levels compared to unbuffered film (2P_r=66 µC/cm²). The PZ buffered films displayed an asymmetric leakage current. The fatigue endurance of PZT thin films with PZ buffer layers was superior compared to the unbuffered counterpart with fatigue-free behavior observed up to 10⁹ switching cycles.

To embody an infinite memory time for the bistable chiral splay nematic liquid crystal (BCSN LC) device, we propose a multidomain structure in which a π/2 twist domain is formed around the BCSN LC device. We investigate how the multidomain structure of the BCSN LC device affects the propagation of a disclination line using a subpixel model. By analyzing the multidomain structure using the subpixel model, the free energy is minimized when the disclination line is located at the boundary between the two different domains. We also demonstrate experimentally that, by forming a π/2 twist domain around the pixel of the BCSN LC device, the propagation of the disclination line can be prevented and an infinite memory time can be achieved.

We investigate the solvent effects of water, methanol and ethanol on the anionic (–SO₃^-) and acid (–SO₃H) forms of the Nafion side chain using ab initio density functional calculations. The polarizable continuum model (PCM) and the explicit solvent molecule additive are used to represent the solvent. The alcohol molecule can more strongly interact with the proton of the acid form than a water molecule, indicating that the proton tends to trap the alcohol molecule.

In this study, two types of ZnO-based multilayer varistor (MLV) with two different dielectric layers (12 and 24 µm), sintered from 900 to 1000 °C for 2 h were prepared to investigate the effects of microstructure such as the grain size and number of grain boundaries between two adjourn electrodes on electrical properties. The results show that the grain size linearly increases with sintering temperature, which results in an increase in the capacitance of ZnO-based MLVs. In contrast, the number of grain boundaries between two adjourn electrodes linearly decreases with sintering temperature associated with a decrease in breakdown voltage, leakage current and nonlinear coefficient of ZnO-based MLVs. The energy absorption capabilities determined from the peak current (PC) measurements of ZnO-based MLVs with sintering temperature are reported. The optimum peak currents of ZnO-based MLVs can be obtained by sintering at 950 °C.

Sol–gel method followed by calcination at temperatures of 300–900 °C was used to obtain a series of LiMn₂O₄ samples with varying amounts of chemical and structural defects while preserving a constant Li:Mn atomic ratio. The physicochemical and structural properties of the samples were characterized by X-ray diffraction (XRD), thermal gravimetry analysis–mass spectrometry (TGA–MS), differential scanning calorimetry (DSC), and Raman spectroscopy techniques. These results were correlated with separately performed high-frequency EMR measurements on the same samples. The oxidation number of Mn ions changes as compared with the stoichiometric spinel composition LiMn₂O₄. These changes, induced by the removal or incorporation of oxygen, were analyzed and correlated with the appearance or disappearance of the cubic to orthorhombic phase transition at around room temperature. The phase transition occurs only if the concentration of vacancies (cationic or anionic) is below the limit necessary to form sufficient majority of perfect MnO₆ octahedra of the high symmetry, O_h⁷. Disturbance of this high local symmetry, being a condition for Jahn–Teller distortion, seems to also be the decisive factor in suppressing the phase transition that is regarded to be responsible for electrical capacity fading during cell cycling.

Lead-free sodium–potassium–bismuth–titanate (Na_1-xK_x)_1/2Bi_1/2TiO₃ was prepared by a traditional solid-state reaction method. The temperature dependences of dielectric constants and piezoelectric properties were measured. The diffuse phase transition behavior of the solid solutions with increasing potassium content was also investigated. When x is lower than 0.2, the solid solutions exhibit a first-order phase transition below the temperature corresponding to the dielectric constant maximum. The shift behavior of the dielectric constant maximum in perovskite structure materials containing titanate was related to the diffuse phase transition behavior, which is due to the ionization of oxygen vacancies. The upshift of the dielectric constant maximum with increasing content of x shows ferroelectric properties, but it's downshift shows an antiferroelectric phase transition in the (Na_1-xK_x)_1/2Bi_1/2TiO₃ system.

The feasibility of applying ZrO₂·H_x thin films as solid electrolytes in solid-state ionic energy systems, such as solid oxide fuel cells and supercapacitors was studied. ZrO₂·H_x thin films were deposited on Pt/Ti/SiO₂/Si substrates by radio-frequency reactive sputtering with various hydrogen volume fractions in reactive gas. With a variation in hydrogen volume fraction, the surface roughness of the as-deposited films increased. In addition, the structure of the as-deposited films grew in the [111] direction with an increase in hydrogen volume fraction. By Rutherford backscattering spectrometry (RBS) and secondary ion mass spectrometry (SIMS) studies, the Zr/O ratio and hydrogen distribution were evaluated. On the basis of a sample structure of Pt/ZrO₂·H_x/Pt/Ti/SiO₂/Si for measuring an electrochemical property, an impedance measurement conducted at room temperature revealed an ionic conductivity of 1.67 ×10^-6 S/cm, suggesting that ZrO₂·H_x thin films can possibly be used as solid oxide thin film electrolytes in all solid-state ionics power devices requiring a hydrogen conducting electrolyte.

The conventional crystal rotation method to evaluate the pretilt angles of calamitic liquid crystals has been applied to determine the pretilt angles of a discotic nematic liquid crystal. We found that pretilt angle increases with decreasing rubbing strength and temperature.

A commercially available resistor paste was embedded into a calcium borosilicate glass ceramic substrate. The fabricated packages were sintered in the 700–900 °C temperature range. The interactions between the embedded resistor and the substrate were studied. Two consecutive phase separations were observed. First, at 800 °C, the diffusion of Ca²⁺ ions from the substrate into the embedded resistors occurred, resulting in the formation of two separated glass phases with high and low SiO₂ content. When the reaction temperature increased to 850–900 °C, the separated glass phase composition reached consistent values and the second phase separation stage occurred.

Two types of electrodes for electrochemical generation of ozone were fabricated by thermal decomposition and sputtering. Ozone generation was observed using both the electrodes. The catalyst layer of the electrode, fabricated by thermal decomposition, functions as an insulator comprising of tantalum oxide and platinum. It had many cracks, and some of them extended up to the platinum-buffered layer. However, we have not explained the ozone generation by the current concentration in the platinum-buffered layers due to the passage of an electrolyte through the cracks alone. On the other hand, the catalyst layer of the electrode fabricated by sputtering was also an insulator comprising only of tantalum oxide; this layer was also crack free. Because of the thin catalyst layer (ca. 20 nm), electrons drifted through it thereby causing electrolysis (i.e., ozone was generated). The results show that tantalum oxide functions as a catalyst of ozone generation. Therefore, with regard to electrolysis using the electrode fabricated by thermal decomposition, ozone generation by an electrochemical reaction is possible in both the thin layer of tantalum oxide, in which cracks on the inner wall extend up to the interior of the catalyst layer, and the platinum-buffered layer.

We propose a Schur-type algorithm that includes spectral factorization of covariance matrix using circulant matrix factorization to design optical multimirror filters. The Schur algorithm is the method used for a fast Cholesky factorization of the Toeplitz matrix, which can determine the reflection coefficient of optical multimirror structures. Circulant matrix factorization is a very powerful tool used for spectral factorization from the covariance polynomial using matrix manipulation in vector space that can be found in the minimum phase polynomials without using the polynomial root finding method. We present a detailed description of the circulant matrix factorization for the reciprocal polynomial approximation of an arbitrary curve (or spectrum). The Schur algorithm can, in turn, be applied to obtain the desired reflection coefficient of the optical filters. We also verify the performance of the proposed method by comparing it with the polynomial root finding method.

Aluminum nitride (AlN) thin films were prepared on Inconel 600 superalloy diaphragms by rf magnetron sputtering for the first time to our knowledge. The crystal structure of the AlN films is hexagonal, and the c-axis of the AlN films orients perpendicular to the diaphragm surfaces. The full-width at half-maximum (FWHM) of the X-ray rocking curves of the AlN films is 5.7°, and the piezoelectric constants d₃₃ and d₃₁ are 2.0 and 0.7 pC/N, respectively. We have investigated the influence of the diaphragm structure on the piezoelectric response to pressure of the AlN films. The AlN films sensitively generate electric charges to pressure changes, and the generated charges show an excellent linearity with increasing pressure. The AlN films indicate a high sensitivity of 723 pC/N. The sensitivity of the AlN films agrees with the result calculated using a method in which the electroelastic energy is differentiated from the voltage in AlN films for unimorph circular diaphragms.

Highly conformal hafnium silicate films were deposited by atomic-layer chemical vapor deposition (ALCVD) using a new combination of precursors: hafnium tetra-tert-butoxide [Hf(OC(CH₃)₃)₄] and tetrakis-ethylmethylaminosilane [Si(N(CH₃)(C₂H₅))₄]. The self-limiting nature of ALCVD film growth was demonstrated by showing the convergent growth rate at high concentrations of the precursors. The growth rate was 3.8 Å/cycle at 220 °C, which was relatively high compared with results using other precursors. It was also shown that we could control the Hf/(Hf+Si) composition ratio in the high Hf/(Hf+Si) ratio region. The carbon impurity concentrations of the films made were lower than the X-ray photoelectron spectroscopy (XPS) detection limit (<1 at. %). Hafnium silicate films with ∼80% HfO₂ were amorphous up to 700 °C. The hafnium silicate films deposited at 220 °C have an average dielectric constant of 9.8 with a flatband voltage (V_fb) and a hysteresis voltage in capacitance–voltage (C–V) measurements of 0 V and less than 0.18 V, respectively.

Ti–Cu–N films containing approximately 0–10 at. % Cu were deposited on Si(100) substrates with high-density low-energy ion flux irradiation by inductively coupled plasma (ICP)-assisted magnetron sputtering. The effects of Cu doping on film microstructures, morphology and properties were investigated. The addition of a small amount of Cu markedly modified the preferred orientation and morphology, and significantly increased film hardness. A Ti–Cu–N film containing 2 at. % Cu has a maximum hardness of approximately 42 GPa. This film was characterized as having a nanocomposite structure, consisting of nanocolumns of TiN crystallites with very small Cu crystallites inside the column boundaries. The hardness increase was attributed to a nanocomposite effect.

We could form a smooth DNA film using synthesized short oligomers (10-mer adenine oligomer and thymine oligomer). The DNA film was observed by atomic force microscopy (AFM) in air. We found from AFM images that the DNA film growth strongly depended on sample preparation temperature within the range of 20 to 40 °C. The DNA film formed at 30 °C was markedly smoother than those prepared at other temperatures. This marked variation in morphology was attributed to the temperature-dependent interaction between oligomers. We discussed the interaction in light of the melting temperature of the DNA oligomers in saturated solution.

In this paper, the oxygen sensitivity of gallium oxide thin films and single crystals at high temperatures is presented. To investigate the oxygen sensing mechanism at high temperature, we used sputtered β-Ga₂O₃ thin films and β-Ga₂O₃ single crystals with different electrode geometries. For β-Ga₂O₃ single crystals: a response time of about 10 s was achieved, while for β-Ga₂O₃ thin-film this was about 11 s. For single crystal samples the response time does not depend on the type of electrode. This can be explained by the combination of a greater influence of surface effects and smaller influence of bulk effects.

When an electron swarm in gas is composed of component electron swarms individually in drift equilibrium, the higher-order diffusion coefficients (HDCs) of the composite electron swarms are equal to those of the component electron swarms. We have derived this equality theoretically and have examined it by numerical simulation. The HDCs are the time derivatives of higher-order cumulants, which are quantities characterizing the shape of a spatial electron distribution. This fact seemingly indicates the dependence of the HDCs of the composite electron swarms on the arrangement of the component electron swarms. However, the equality holds irrespective of the relative positions and electron populations of the component electron swarms. We have given a consistent explanation to these facts. We have also discussed electron swarm development from dispersed or multiple electron sources that may appear in practical experiments.

The internal coil device Mini-RT (Ring Trap) has been constructed in order to study the extremely high beta plasma confinement predicted in two fluid relaxation theory. Plasma is produced by electron cyclotron heating with a 2.45 GHz, 2.8 kW microwave. Plasma experiments were carried out for two conditions: with a mechanically supported coil, and with a magnetically levitated one. In the case in which the internal coil was mechanically supported, plasma densities up to n_e=7×10¹⁶ m^-3 were produced with a filling gas pressure of p_n=4×10^-2 Pa, and plasma could not be available at a gas pressure less than 10^-2 Pa due to the loss of high energy electrons by the supporting structure. It has been clearly demonstrated that the levitation of the internal coil, which was followed by the removal of the supporting structure from the plasma confinement region, improves plasma parameters markedly. The levitation of the internal coil enables plasma production at a lower filling pressure ( p_n=1.5×10^-3 Pa). It also enables over-dense plasma production (n_e=1.5×10¹⁷ m^-3, while cutoff density is 7.6×10¹⁶ m^-3 for 2.45 GHz microwave). Since the calculation shows that the efficiency of mode conversion from fast X-mode wave to electron Bernstein wave (FX–SX–B conversion) is sufficiently high (∼87%) in the Mini-RT device, the over-dense plasma production is thought to be caused by heating with electron Bernstein wave (EBW).

A practical definition for the non-neutral/quasi-neutral plasma edge region is proposed and tested with the help of a fluid model. The numerical calculations show that the definition reproduce well the space-time behavior of the plasma sheath for single and dual frequency capacitively coupled discharges. The simulation results indicate that the velocity of the sheath expansion for a H₂ discharge sustained at 13.56 MHz frequency, 0.5 Torr pressure and 200 V applied rf voltage is about 2.5 ×10⁷ cm s^-1. For dual frequency discharges, the modulation of the sheath is stronger when the sheath thickness assumes its minimum.

This paper is a report of the ignitability of coating polymer powders due to an electrostatic spark in an electric field with a corona discharge. The ultrasonic vibration-type apparatus for measuring the ignitability of an electrostatic spark of coating polymer powders was used in this study. The dust concentration in all tests was always 0.60 kg/m³. Polyester powder with grains of 32 µm used in practical electrostatic powder-coating plants was employed as the powder material. The corona discharger, which was located in the explosion chamber, was controllable in the range of DC 5 to 8 kV (bipolar). As a result, the ignitability of coating polymer powders was confirmed to decrease in an electric field created by corona discharge. The reduction was dependent on the voltage applied to the wire electrode of the corona discharger. This result was attributed to the changes in the dispersion of the powder and the free ions produced by the corona discharge. Therefore, the influence of the corona discharge must be considered in the minimum ignition energies of powder paints when using electrostatic powder coating systems.

Carbon nanowalls (CNWs) are synthesized under pure methane gas (CH₄) using helicon plasma-enhanced chemical vapor deposition. CH₄ in the helicon discharge is effectively dissociated to hydrogen atoms and hydrocarbon radicals, resulting in the formation of CNWs on a Ni substrate only from CH₄. CNWs are grown up at a high growth rate of 18 µm/h.

The density of CN(X²Σ⁺) radicals, which are considered to be the precursor to the formation of amorphous carbon nitride films, was measured by laser-induced fluorescence (LIF) spectroscopy of the CN(A²Π_i–X²Σ⁺), 4-0, 5-1, and 7-2 bands. The CN(X²Σ⁺) radicals were produced by the dissociative excitation reaction of BrCN with microwave (MW) discharge flow of Ar under the pressure of 0.1 Torr, and the LIF spectra were observed just downstream of the nozzle tip from which BrCN was introduced. The LIF intensity for the individual transition was evaluated by spectral simulation analysis and by calibration against the Rayleigh scattering intensity by Ar atoms. The CN(X²Σ⁺) density was evaluated to be 9.9×10¹⁷ m^-3. This density was analyzed using the previously determined Ar⁺ density [H. Ito et al.: Jpn. J. Appl. Phys. 43 (2004) 7277], and it is suggested that the dominant production and loss processes of CN(X²Σ⁺) are BrCN⁺-e^- recombination and quenching with BrCN, respectively.

The mass analysis of positive ions has been carried out at a position 90 cm downstream from the center of the Ar/CF₄ plasma-generating area, the size of which is 5 cm in length and 5 cm in diameter in a cylindrical tube. As a result, it has been found that there are six series of adduct ions, C_nF_2n+1⁺ (n=2–7), C_nF_2n-1⁺ (n=3–8), C_nF_2n-3⁺ (n=3–9), C_nF_2n-5⁺ (n=6–10), C_nF_2n⁺ (n=2–6), and C_nF_2n-2⁺ (n=4–6), as well as CF⁺, CF₂⁺, and CF₃⁺ produced by (dissociative) ionization of CF₄ and its neutral fragments. The dependence of the intensities of Ar⁺, CF⁺, CF₂⁺, and CF₃⁺ on the CF₄ mixing ratio in a range of 0–0.3 agrees very well with that predicted by Kimura and Takai [Jpn. J. Appl. Phys. 43 (2004) 7240] using a global model for an electronegative plasma. This fact demonstrates that various chemical reactions advance even in the downstream region of the plasma. The logarithmic plots of the intensity in C_nF_2n+1⁺ at n≥2 with respect to the mass number decrease linearly as the mass number increases. This is the case for C_nF_2n-1⁺ at n≥3, and its slope is gentler than that in the C_nF_2n+1⁺ case. Quantum chemical calculations with GAUSSIAN 03 have been carried out to estimate the enthalpy change in the various reactions predicted to advance in the downstream region of the plasma. As a result, it has been found that C₂F₅⁺ is produced dominantly by the addition reaction of CF₃⁺ to CF₂. No peak is assigned to the C₂F₃⁺ ion in all the observed mass spectra. This finding indicates that the C₃F₅⁺ ion must be produced, not by the addition reaction between C₂F₃⁺ and CF₂, but by other reactions. C₃F₅⁺ is expected to be produced by the reactions, C₃F₇⁺+CF₂→C₃F₅⁺+CF₄, C₂F₅⁺+CF→C₃F₅⁺+F, and C₃F₇⁺+CF→C₃F₅⁺+CF₃, judging from the calculated enthalpy changes of these processes. The linearity of the logarithmic plots obtained experimentally can be described by considering the rate equations for the addition reactions of CF₂ with C_nF_2n+1⁺ at n≥2 and C_nF_2n-1⁺ at n≥3 under the steady-state approximation. The difference in the slope between the logarithmic plots of the intensity in the C_nF_2n+1⁺ and C_nF_2n-1⁺ series indicates that the formation of C_nF_2n-1⁺ at n≥4 is due to the addition reaction of CF₂ with C_n-1F_2n-3⁺. The lowest unoccupied molecular orbitals (LUMOs) of the C_nF_2n+1⁺ series are dominantly constructed from the –CF₂⁺ end included commonly in C_nF_2n+1⁺, and those of the C_nF_2n-1⁺ series are from the –C₃F₄⁺ end, regardless of the value of n, leading to the difference in the reactivity between C_nF_2n+1⁺ and C_nF_2n-1⁺.

Red phosphorescent polymer light-emitting diodes based on bis(2-(2'-pyridyl)benzo[b]thiophene-N,C³)iridium(III)acetylacetonate [btp₂Ir(acac)] and poly[(9,9-dioctylfluorene)-alt-(pyridine)] (PF8-Py) have been investigated. The btp₂Ir(acac):PF8-Py blend device emitted a rich red phosphorescence from the btp₂Ir(acac) complex. This is due to the effective energy transfer from the PF8-Py copolymer to btp₂Ir(acac) enhanced by the pyridine unit in the copolymer. By the addition of a hole transporting material, i.e., N,N'-bis(3-methylphenyl)-N,N'-diphenylbenzidine (TPD), to the btp₂Ir(acac):PF8-Py blend, the driving voltage of the device was dramatically lowered and luminance and efficiency were greatly increased. Device performance can be further improved by introducing a hole transporting layer (HTL). The multilayer device, with a poly[(9,9-dioctylfluorene)-alt-(triphenylamine)] (PF8-TPA) layer as a hole transporting layer, showed a maximum luminance of 2600 cd/m², a maximum efficiency of 3.78 cd/A and a maximum external quantum efficiency of 4.92%. These are supposed to be obtained by having a good carrier balance in an emitting layer, an effective energy transfer from PF8-Py to the btp₂Ir(acac) complex, and a highly-efficient triplet emitter.

Tetracene single-crystal field-effect transistors (FETs) were fabricated using fresh cleaved tetracene single crystals. In order to obtain a fine contact between the tetracene crystals and the substrate, a stress was applied using an Instron-type machine. Carrier mobility was measured as a function of applied stress and it was observed that 0.5 cm²/(V s) was the highest mobility under the optimum applied stress.

Polymer solar cells based on bulk heterojunction of poly(3-hexylthiophene) (PAT6) and C₆₀ with high efficiencies have been demonstrated. The efficiencies were markedly dependent on composite ratios and annealing temperatures, and energy conversion efficiencies of 2.2% under white light and 6.5% under monochromatic light were obtained in a cell with a PAT6:C₆₀ composite ratio of 1:0.5 annealed at 100 °C. Furthermore, the absorption of PAT6:C₆₀ composite films in a wavelength range related to the absorption of PAT6 depended on the C₆₀ composite ratios, and was changed significantly by annealing. These changes are discussed taking into consideration the hindrance of the crystallization of PAT6 by C₆₀ and the phase segregation between PAT6 and C₆₀ through the aggregation and crystallization of each molecule caused by annealing.

For fast and simultaneous measurement of both birefringence and azimuth angle of samples, a technique using transverse phase modulator is proposed. The phase modulator includes y-cut z-propagation LiNbO₃ crystal. Several experiments using rotating quarter-wave plates and LiNbO₃ crystal in presence of direct voltage have been performed, and the measurement results agree well with predicted theoretical data.

We have performed high sensitive imaging of the atomic arrangement of Ge buried in a Si crystal as Ge clusters using the internal detector scheme of X-ray fluorescence holography. The atomic arrangement of Ge buried in a Si crystal is expected to have the strained diamond structure. We have measured holograms of a thin Si/Ge multilayer film composed of thin Si layers and Stranski–Krastanov Ge islands using a crystal analyzer in combination with synchrotron radiation. The atomic arrangement reconstructed from the holograms shows that the Ge islands are indeed of the diamond structure. Comparison of atomic arrangements reconstructed from experimental and simulated holograms allows us to determine the average lattice constant of the Ge islands. The lattice constant obtained shows that the Ge islands are partially relaxed in Si matrix lattice.

The X-ray phase tomography of biological samples is reported, which is based on X-ray Talbot interferometry. Its imaging principle is described in detail, and imaging results obtained for a cancerous rabbit liver and a mouse tail with synchrotron radiation are presented. Because an amplitude grating is needed to construct an X-ray Talbot interferometer, a high-aspect-ratio grating pattern was fabricated by X-ray lithography and gold electroplating. X-ray Talbot interferometry has an advantage that it functions with polychromatic cone-beam X-rays. Finally, the compatibility with a compact X-ray source is discussed.

An electrostatic ion trap was designed for the study of vibrational states of molecular ions. The trap consists of two einzel lenses and two parallel-plate reflectors. An optical model analogous to this trap predicts that a range of focal lengths of lenses exists where ions can be stored irrespective of the reflector's focal length. Beam tests with 1.2-keV ion beams showed that ions can be stored in this range. That storage time is 0.1 s for an Ar⁺ beam at a vacuum pressure of 2×10^-8 Torr.

The control of airborne molecular contamination (AMC) plays an increasing role in semiconductor manufacturing processes. We conducted a parametric study of purging a front-opening unified pod (FOUP), a wafer box for handling 300 mm wafers, with nitrogen experimentally, analytically and numerically. Factor considered to affect purging include the coefficient of absorption and desorption of water vapor through the FOUP's polycarbonate material, the configuration of plenum injector, the best composition of plenum injector and the FOUP, and the selection on purge flow rate. This study clarified purge phenomena.

A convenient and precise method has been developed for the determination of Peltier coefficient, Seebeck coefficient and thermal resistance of the thermoelectric module (TM). The key point is the application of Joule heating by an AC current through the module in addition to a DC current that transfers heat by Peltier effect. A simple circuitry consisting of a signal generator and two digital multimeters is all that is required. The performance test was satisfactory. The temperature difference required for measurement is below 0.1 K, within which the temperature homogeneity is maintained in the equipment. This method is applicable under high pressure as well as in vacuum.

The analysis of isotope ratio in a material consisting of a single element was developed as a fundamental technique to determine a self-diffusion coefficient in a melt based on time-of-flight secondary mass spectrometry (TOF-SIMS). The self-diffusion coefficient for a pure Ge melt was measured using the stable isotope ⁷³Ge as a tracer under a homogeneous static magnetic field in order to evaluate the influence of thermal convection upon isotope distribution. The results obtained showed that the magnetohydrodynamic effect in the melt obviously damped the convection, but it was not strong enough for the self-diffusion measurement.

Microfocus X-ray generators have been extensively used for magnification radiography. In these applications, the sharpness of an image depends on the X-ray source size. Therefore, it is important to evaluate and optimize the X-ray spot quantitatively. Although several methods have been devised to date, some require complicated systems whereas others are not fully quantitative. Here, we introduce a simple, practical, and quantitative method that makes use of the penumbra of a knife edge. Because the method is very simple, it can be readily integrated as an auto focusing system of a focal spot.

This paper describes submicron-resolution X-ray topography using a Fresnel zone plate (FZP). A diffraction image was expanded by an FZP and focused onto a four-quadrant slit, through which X-ray intensity was measured. Topographs were obtained by moving a sample in two dimensions. A local strain analysis of oxide-patterned silicon provided a spatial resolution of less than 0.5 µm and a strain sensitivity of 7 ×10^-6 at a photon energy of 8.5 keV. Unlike microbeam X-ray diffraction analysis, this method is free from diffraction-limit constraint.

This paper describes resonant characteristics of spherically contoured AT-cut quartz-crystal resonators, and a novel fabrication process. The fabrication process is based on a photolithographic process suitable for the mass production. Further, the time required to finish quartz blanks into the spherically contoured shape is dramatically less than for conventional methods. Our measurements show that Q of the spherically contoured resonators fabricated with our method are an order of magnitude higher than those resonators without the spherical shape.

In this paper, we proposed a method of deriving equations of motion, which is free of Lagrange multipliers, for a constrained dynamical system with emphasis on its suitability for nonlinear nonholonomic constraints. A local coordinate space is established on the basis of a scaled acceleration-constrained flat surface and the corresponding normal space. Constraint forces are decomposed into components along this local coordinate space. The normal components of constraint forces are determined by the motion of the system and constraints, and the tangential components are arbitrary. Then Gauss' principle of least constraint is employed to minimize the tangential components so as to derive the equations of motion in explicit form. An example demonstrating the general equations derived here is provided.

We developed a new angle calibration system combining a pure silicon crystal in a natural state and an angle interferometer with a wide span of rotation. The uncertainty in the angle of the six-faced silicon polygon is approximately 10^-8 rad (0.004''), and the angle resolution of the angle interferometer is 0.005''. We evaluated the performance of the angle calibration system and summarized its budget of uncertainty. The uncertainty of the system was estimated to be ±0.07'' (k=2). To demonstrate the new calibration system, we used it to measure the angle error of an index table, and we compared the results with existing calibration methods.

A two-dimensional dosimeter composed of a charge coupled device (CCD) camera and a scintillating screen has been tested for measuring the dose distributions of therapeutic proton beams in situ. The dosimeter is the main component of the dosimetry system designed for the quality assurance of scanning beams in the course of patient treatments. A thin-wall parallel plate ionization chamber was built as part of the system to scale the beam currents, and a collimator and variable-thickness phantom being the other components. The system has been tested using a 40 MeV proton beam in a nontherapeutic beam line. The light output from the screen was linear within 0.1% at the typical dose rates of 1–2 Gy/min. The depth dose distributions were measured using the CCD camera system, and agreed well with the measurements by a calibrated ionization chamber. The CCD camera was also used to measure the beam intensity distributions produced with a Cu scatterer in the beam line. The comparison with the results of Monte Carlo simulations using MCNPX was useful for validating the code.

In the plasma flash X-ray generator, a 200 nF condenser is charged up to 50 kV by a power supply, and flash X-rays are produced by the discharging. The X-ray tube is a demountable triode with a trigger electrode, and the turbomolecular pump evacuates air from the tube with a pressure of approximately 1 mPa. Target evaporation leads to the formation of weakly ionized linear plasma, consisting of copper ions and electrons, around the fine target, and intense Kα lines are left using a 10-µm-thick nickel filter. At a charging voltage of 50 kV, the maximum tube voltage was almost equal to the charging voltage of the main condenser, and the peak current was about 16 kA. The K-series characteristic X-rays were clean and intense, and higher harmonic X-rays were observed. The X-ray pulse widths were approximately 300 ns, and the time-integrated X-ray intensity had a value of approximately 1.5 mGy per pulse at 1.0 m from the X-ray source with a charging voltage of 50 kV.

The formation of many carbon nanotube (CNT) tips with small diameters on the surface of thick CNTs was investigated to improve the field-emission uniformity of a CNT emitter. A CNT layer was first grown on a metal substrate by thermal-chemical vapor deposition (CVD). CNTs in the layer were multiwalled nanotubes with diameters of 30–50 nm and formed a weblike network structure. A metal catalyst, iron, was then vacuum-deposited on the CNT layer, and secondary thin CNTs with diameters of 5–10 nm were formed from the catalyst particles attached to the surface of primary CNTs by thermal-CVD. The field emission from the CNT layer with the secondary CNTs showed an improvement in emission current distribution uniformity.

We have fabricated nanostructures of diluted magnetic semiconductors (DMSs) integrated with high-purity glass thin films and have studied the giant magneto-optical properties, for the purpose of applying in microscopic magneto-optical devices. Zn_1-x-yCd_xMn_ySe quantum wells grown on GaAs substrates were fabricated into two-dimensional arrays of disks or wires with the lateral dimensions ranging from 10 to 500 nm using electron-beam lithography. These nanostructures have been covered with SiO₂ thin films, without a significant degradation of their optical quality, by ultrahigh vacuum sputtering with low plasma damage. The giant Zeeman effects of excitonic photoluminescence (PL) in the integrated nanostructures of DMS are studied in terms of the exchange interactions between exciton and magnetic ions. The magnetic-field dependence of the giant Zeeman shift of excitons shows a redshift beyond 25 meV for 5 T with a significant increase in PL intensity. This ensures characteristic magneto-optical functionalities of the present integrated nano-glass structures of DMS.

We have fabricated a new single-electron device (SED) that has many nanodots. Although SEDs have the great advantages of small size and low power consumption, they should have small dots on the order of a few nanometers, which makes them difficult to fabricate. The proposed device uses many nanodots aligned as an array, on which many gate electrodes are attached so as to couple capacitively to underlying nanodots. Some of the gates are used as input gates of a logic-gate device. The others are control gates that are used to change the logic function of the device, such as from an AND gate to an XOR (exclusive OR) one. The principal operations have been demonstrated using numerical simulations.

An extremely high-resolution, nanometer-resolution, solid-surface trace element detection has been demonstrated. Nanometer-thinned pulsed laser ablation was combined with extremely sensitive laser-induced fluorescence spectroscopy, and depth resolution of 3.6 nm was experimentally demonstrated for the first time on sodium detection in polymeric samples. An extremely high absolute detection limit of 25.2 fg was also obtained, and a theoretical calculation program was also generated to analyze the results.

Vertically aligned carbon nanofiber (CNF) films were successfully grown on glass substrates at 450 °C with metal buffer layers by inductively coupled plasma chemical vapor deposition (ICP-CVD). The diameter and number density of the aligned CNFs can be controlled by changing the type and thickness of the metal buffer layers deposited on the glass substrates. The metal buffer layers play an important role in reducing the thermal expansion coefficient difference between the catalyst metal film and the glass substrate, resulting in the enhancement of the formation of catalyst nanoparticles so as to grow the aligned CNFs with high number density.

Thermal-CVD was carried out for the low-temperature growth of carbon nanofibers (CNFs) using a CuNi alloy catalyst film with a thickness of 5 nm on Si in a gas mixture of C₂H₂ and He (C₂H₂/He=3/12 sccm). The experimental results obtained using the CuNi alloy catalyst film were compared with those obtained using the Fe, Ni, and FeNi catalyst films with the same thickness of 5 nm. It was shown that an amorphous CNF with a diameter of 20 nm can be grown even at 400 °C using the CuNi catalyst film, but not using the Fe, Ni and FeNi catalysts. A reduction in the growth temperature of CNFs was considered to be achieved using small CuNi catalyst particles with a comparatively smaller surface energy than FeNi catalyst particles.

A tabletop experimental system has been developed for the study of various collective effects in space-charge-dominated beams. It is based on the recently proposed idea that the dynamic motion of a one-component plasma in a trap can be made physically equivalent to that of a charged-particle beam propagating through a linear transport channel. In this paper, we report on the details of the system and on results of test experiments with a compact Paul trap that is divided into several independent sections. The trap design is carried out in consideration of practical constraints. A Maxwell equation solver is used to calculate the properties of the plasma confinement potential. Experimental observations are compared with numerical data obtained by a tracking simulation code that enables us to approximately predict the three-dimensional trajectories of particles in the system. Low-density N₂⁺ plasmas are employed to examine the basic performance of the multi-section trap. The initial temperature, density and lifetime of a confined plasma are estimated from experiments and simulations.

We developed a one-way quantum key distribution (QKD) system based upon a planar lightwave circuit (PLC) interferometer. This interferometer is expected to be free from the backscattering inherent in commercially available two-way QKD systems and phase drift without active compensation. A key distribution experiment with spools of standard telecom fiber showed that the bit error rate was as low as 6% for a 100-km key distribution using an attenuated laser pulse with a mean photon number of 0.1 and was determined solely by the detector noise. This clearly demonstrates the advantages of our PLC-based one-way QKD system over two-way QKD systems for long distance key distribution.