Invited Review Paper
7909
Akihisa Matsuda
Growth processes of hydrogenated amorphous silicon (a-Si:H) and microcrystalline silicon (µc-Si:H) from SiH4 and H2/SiH4-glow discharge plasmas are reviewed. Differences and similarities between µc-Si:H and a-Si:H growth reactions in the plasma and on the film-growing surface are discussed, and the nucleus-formation process followed by the epitaxial-like crystal growth process is explained as being processes unique to µc-Si:H. The governing reaction of dangling-bond-defect density in the resulting a-Si:H and µc-Si:H films is also discussed in order to obtain a clue to improve the optoelectronic properties of these materials to enable device applications, particularly to thin-film silicon-based solar cells. Material issues concerning the realization of low-cost high-efficiency solar cells are described, and finally, recent progress in those issues is presented.
Regular Papers & Short Notes
Semiconductors
7921
Joseph Der-Son Deng, Chih-Yuan Lee, Chin-Hsing Kao and Tung-Sheng Chen
This paper presents the realization of a phenomenal Q-value increment as high as 122% for 4 nH spiral inductors on a silicon substrate. The loss mechanisms of parasitic effects on the Q-factor have been studied to improve Q design. The dramatic improvement in the Q value resulted from optimizing the doping level and film thickness of a poly shield layer combined with proton implantation. The shielding effect of poly-silicon and the semi-insulating characteristics of a proton-bombarded substrate have added 37% and 58% increments to the Q value of the inductors, respectively. The combination of the two methods has multiplied their individual contributions rather than just adding them. The technique proposed in this work shall become a critical measure for putting inductors on silicon substrates with superior performance for silicon-based radio frequency integrated circuit (RFIC) applications.
7926
Hyunjun Sim, Chandan B. Samantaray, Taeho Lee, Hanwoong Yeom and Hyunsang Hwang
In this study, the electrical and structural characteristics of Gd2O3 gate dielectrics with an epitaxial Si3N4 interfacial layer grown on Si(111) were investigated. Compared with control Gd2O3 gate dielectrics deposited on HF-last treated Si (111), the Gd2O3 gate dielectrics with an epitaxial Si3N4 interfacial layer exhibited excellent electrical characteristics such as low leakage current density and low interface state density. These characteristics are due to a high-quality interfacial layer formation on Si. Transmission electron microscopy (TEM), X-ray photoelectron spectroscopy (XPS) and Fourier transform infrared (FTIR) spectroscopy were employed to analyze the structures of the gate dielectrics and interfacial layer. High-k gate dielectrics with an epitaxial Si3N4 interfacial layer have considerable potential for future use in sub-0.1 µm metal oxide semiconductor field-effect transistors (MOSFETs).
7929
Hiroya Kirimura, Kiyoshi Kubota, Eiji Takahashi, Shigeaki Kishida, Kiyoshi Ogata, Yukiharu Uraoka and Takashi Fuyuki
As a novel direct deposition method of microcrystalline silicon, we have developed the high-density and low-potential plasma-enhanced silane generating chemical vapor deposition (CVD) system. We have studied a two-step deposition process which consists of the silicon nucleation step using atomic hydrogen (radicals) and the microcrystalline growth step using silane plasmas at low temperature. Transmission electron microscopy (TEM) and micro-ultraviolet Raman spectrometry (UV-Raman) analyses reveal that silicon films crystallize with a low defect density starting from the interface between the SiO2 substrate and the film. Furthermore, the electron mobility determined on the basis of the TFT characteristics indicates that this method is highly effective for the direct deposition of microcrystalline silicon.
7934
Jong-Won Lim, Ho-Kyun Ahn, Hong-Gu Ji, Woo-Jin Chang, Jae-Kyoung Mun and Haecheon Kim
A plasma-enhanced chemical vapor deposited (PECVD) silicon–nitride-assisted process has been used to fabricate 0.12 µm T-shaped AlGaAs/InGaAs/GaAs PHEMTs. A two-step SiNx etch was performed to define the T-gate footprint. The SiNx was etched either by dry etching alone or using a combination of wet and dry etching. A structure for the top of the T-gate which consists of a wide head part and a narrow lower layer part has been employed, taking advantage of the large cross-sectional area of the gate and its mechanically stable structure. The gate recessing was carried out in two steps including a wet etching for removal of the damaged surface layer and a dry etching for the narrow recess. The cut-off frequency of the device fabricated by the two-step etch process is higher than that of the device fabricated by dry etching alone. We have observed the increase of the cut-off frequency fT and the maximum oscillation frequency fmax, by 39% and 16%, respectively. This is believed to be due to considerable decreases of the gate–source and gate–drain capacitances, by 25% and 18.3%, respectively. This improvement in RF performance can be understood in terms of the decrease in parasitic capacitances which is due to the use of the dielectric and the gate recess etching method. The minimum noise figure is 1.6 dB, with a 6.9 dB associated gain at 39 GHz.
7939
Makoto Miyoshi, Masahiro Sakai, Subramaniam Arulkumaran, Hiroyasu Ishikawa, Takashi Egawa, Mitsuhiro Tanaka and Osamu Oda
Different-Al-content AlxGa1-xN/GaN (0.26≤x≤0.52) high-electron-mobility transistor (HEMT) structures were grown on large-area 100-mm-diameter sapphire substrates by metalorganic vapor phase epitaxy (MOVPE). A low sheet resistance of approximately 380 Ω/sq. and an in-wafer total variation of 7.6% with a standard deviation of 1.6% were obtained for the sample with the highest Al content of 0.52. The two-dimensional electron gas (2DEG) density and room-temperature Hall mobility of the Al0.52Ga0.48N/GaN HEMT structures were 1.76×1013/cm2 and 971 cm2/Vs with in-wafer total variations of 7.8% and 8.1%, respectively. HEMTs were successfully fabricated and they exhibited good dc characteristics. The dependence of maximum source–drain current density on Al content was consistent with the Al content dependence of sheet carrier concentration. The highest source–drain current density of 1033 mA/mm with a maximum extrinsic transconductance of 228 mS/mm was obtained for 2-µm-gate-length Al0.52Ga0.48N/GaN HEMTs.
7944
Hiroaki Koga, Tsuyoshi Miyazaki, Satoshi Watanabe and Takahisa Ohno
We examine the feasibility of migration-enhanced epitaxy (MEE) of cubic BN (cBN) on diamond(001) by means of ab initio calculation. In particular, we are concerned with the question as to whether B deposition or N deposition should be performed first. We show that N deposition should not be performed first, since it is difficult to form a monolayer of N on diamond(001) due to the desorption of N addimers. We then show that MEE is feasible if B deposition is performed first, since the first and second deposition layers tend to assume a flat monolayer structure, without being hindered by N2 desorption, and since the subsequent growth process is free of the substrate effect, apart from the small lattice mismatch effect.
7947
Shinya Ono and Yoshinao Kobayashi
Voltage-controlled pixel circuit with four thin-film-transistor (TFT) for amorphous-silicon (a-Si) active-matrix organic light-emitting diode (AM-OLED) displays has been designed, simulated and evaluated. Deviation and aging of the properties of the driver-TFT cause image sticking or degradation of image quality. These problems require compensation for high-quality display applications, and a pixel-level compensation circuit is one of the solutions. In this paper, we propose a novel and suitable circuit for a-Si AM-OLED displays. The circuit is composed of four TFTs and a storage capacitor, and its driving scheme has four periods: preparation, programming threshold voltage (VT), writing data, and emission. It is possible to set a long detection time for the threshold-voltage of driver-TFT in each pixel, so it can compensate precisely for deviations and the aging of the threshold voltage.
7953
Yoshinori Ide, Yuji Saito, Akira Yamada and Makoto Konagai
Microcrystalline silicon (µc-Si:H) thin films and solar cells were prepared by a hot-wire cell (HW-cell) method. A maximum conversion efficiency of 6.0% (Voc: 0.50 V, Jsc: 19.69 mA/cm2, F.F.: 0.61, active area: 0.086 cm2, AM1.5) was obtained for µc-Si:H solar cells prepared at a low filament temperature of 1800°C. The influence of filament temperature on µc-Si:H film properties was investigated. As a result, it was found from secondary ion mass spectroscopy (SIMS) results that tungsten (W) and aluminum (Al) concentrations largely increase with increasing filament temperature from 1800°C to 2100°C. Next, we found that a considerable degradation of solar cell performances occurs with aging. The degradation of intrinsic µc-Si:H absorber layers with aging was investigated, and it was found that the peak at 1000–1100 cm-1 originating from Si–O–Si bonds and spin density markedly increase. SIMS results also showed that O and C atoms in our µc-Si:H films are as low as 2×1018 cm-3 in the initial state and the films can easily undergo post oxidation. However, it was also found that a deposition at a high substrate temperature of more than 165°C is very effective in realizing high-stability µc-Si:H films and that a 3-µm-thick a-Si:H cap layer can prevent post oxidation.
7960
Haijun Jia, Masanobu Nakajima, Aiko Nakao and Hajime Shirai
The effects of chamber wall heating and quartz window on oxygen contamination and film crystallinity have been demonstrated in a high-density microwave plasma of a SiH4 and H2 mixture. With increasing chamber wall temperature, the film crystallinity deteriorates with higher concentration of oxygen impurity. The content of the oxygen impurity from the quartz window in the films is also significant with increasing working pressure and flow rate ratio of H2 to SiH4. The combination of higher flux of SiH4 and appropriate pressure of 150–300 mTorr is effective in suppressing the oxygen impurity while maintaining high film crystallinity. As a result, µc-Si film with a low defect density of 1–1.2×1016 cm-3 has been fabricated at a deposition rate of ∼50 Å/s.
7966
Manabu Ishimaru, Mitsuhiro Yamaguchi and Yoshihiko Hirotsu
Thermally-induced structural relaxation of amorphous silicon-germanium (a-Si1-xGex) alloys has been examined using molecular-dynamics (MD) simulations with the Tersoff interatomic potential. The a-Si1-xGex networks were successfully observed in the present MD simulations. Major changes in the topological short-range order upon annealing occurred within the second coordination shell, while the chemical short-range order was not altered. It was found that the decrease in potential energy associated with the change in the standard deviation of bond angles becomes large as the Ge composition increases. This result was consistent with the heat release due to structural relaxation, which was observed experimentally. We also examined differences in the phonon densities of states between the as-quenched and annealed a-Si1-xGex alloys.
7971
Kannan L Narayanan, Tetsuo Soga, Tomoharu Tokunaga and Masafumi Yamaguchi
Microcrystalline and nanocrystalline diamond thin films deposited on Si substrates by microwave plasma chemical vapor deposition are bombarded by 40 keV B+ ions. Raman scattering studies of the ion implanted diamond films reveal the structural modification irrespective of the nature of the crystalline quality. X-ray diffraction results indicate the possibility of formation of borocarbide phase on ion irradiation. Ion beam current dependence and post ion implantation annealing studies on the micro and nano-crystalline thin films are also presented.
7976
Akihiro Takano and Tomoyoshi Kamoshita
Production technologies for light-weight, large-area and flexible solar cells are reviewed. Roll-to-roll processes including an originally developed stepping-roll deposition process are used to fabricate large-area hydrogenated amorphous silicon-based solar cells in succession on plastic film substrates. A unique monolithic device structure for the insulating film substrates having through-hole contacts was developed and applied to simplify the production processes and to generate high voltage using connection in series. Newly developed deposition methods of (1) plasma excitation frequency controlled chemical vapor deposition, (2) low substrate temperature selective reactive sputtering, and (3) large-area uniform transparent conductive oxide film deposition are reviewed. Finally, our recently developed 4 m-long, light-weight, large-area building-integrated modules are introduced.
7984
Joan Marc Rafí, Abdelkarim Mercha, Eddy Simoen, Kiyoteru Hayama and Cor Claeys
In this study, we analyze the impact of back-channel radiation-induced leakage and back-gate bias on switch-off drain current transients of thin-gate-oxide partially depleted (PD) silicon-on-insulator (SOI) n-channel metal–oxide–semiconductor field-effect transistors (MOSFETs). The presence of radiation-induced positive trapped charges in the buried oxide after 60 MeV proton irradiation is found to reduce the “switch-off” transient times for gate voltages above and below the front-gate threshold voltage for body-to-gate electron valence band tunnelling. An increase in steady-state drain current and an increase in the amplitude of weak inversion drain current transients are observed. A similar effect is observed when applying a positive bias to the back-gate, which is found to generate an “irradiation-like” subthreshold leakage. The observed switch-off drain current transient behavior is explained by taking into account an edge parasitic back-channel transient component, which is added to the conventional front-gate drain current transient.
7993
Chun-Hsing Shih, Yi-Min Chen and Chenhsin Lien
In this paper, we present an analytical short-channel threshold voltage model without any fitting parameter for the metal–oxide–semiconductor field-effect Transistor (MOSFET) with insulated shallow extension (ISE). Excellent agreements between the numerical simulated results and this model are obtained. Very good suppression of the short-channel effect is observed for this ISE MOSFET. Both the sidewall-oxide thickness and shallow-extension depth play a major role in containing the short-channel effect. The threshold-voltage equation is found by using the effective-doping model. The effective doping is derived from the knowledge of the channel potential. The channel potential is obtained by the scale-length approach to solve 2D Poisson's equation.
7997
Byung Woo Jeoung, Gun Young Hong, Bo Yong Han and Jae Soo Yoo
The charging tendency of a Zn2SiO4:Mn2+ phosphor surface was modified in order to improve the discharging characteristics of green cell in an ac-plasma display panel (ac-PDP). The zinc silicate (Zn2SiO4:Mn2+) green-emitting phosphor was coated with magnesium oxide (MgO), which has a positive charge on the surface. After fabricating a green cell with MgO-coated Zn2SiO4:Mn2+, the electrical and optical properties were examined. It was found that the dynamic voltage margin could be increased while decreasing address time. This may be ascribed to a change in the charging tendency of Zn2SiO4:Mn2+ phosphor by the MgO coating, which makes stable wall-charge accumulation possible. When Zn2SiO4:Mn2+ phosphor was coated with 1.3 wt% MgO, the address time was decreased to 1.2 µs and the address voltage lowered to 25 V without any misfiring problem, compared with those of a noncoated Zn2SiO4:Mn2+ phosphor layer. The luminescence intensity of the green cell using the MgO-coated phosphor layer was also improved by 10%.
8002
Tetsuya Okamoto, Tetsuya Ide, Atsushi Sasaki, Kazufumi Azuma and Yukihiko Nakata
Irradiation damage in an SiO2/Si system induced by photons and/or ions was evaluated. The flat-band voltage (Vfb) was not shifted and the interface density of states (Dit) was increased by irradiation of 172-nm-wavelength light, which corresponds to a photo-oxidation condition. Surface wave plasma (SWP) excited using microwave with Kr gas has an electron temperature less than 1 eV. However, Vfb was shifted and Dit was increased by the irradiation of photons from Kr plasma, which corresponds to a plasma-oxidation condition. Vfb was not shifted and Dit was increased by the irradiation of ions from Kr plasma. However, the SiO2/Si interface formed by photo-oxidation had a Dit of 3×1010 cm-2eV-1, which is equal to that of a SiO2/Si interface formed by a thermal-oxidation. This means that photo-oxidation forms the SiO2/Si interface by a concerted reaction of the photo-oxidation and the irradiation.
8007
Joo-Wan Lee, Jun Ki Kim, Soo-Hyun Kim, Ho-Jung Sun, Hong-Seon Yang, Hyun Chul Sohn and Jin Woong Kim
We attempted to explain the phenomenon that the electric resistivity of tungsten film increases as the thickness decreases and that physical vapor-deposited (PVD) tungsten shows a much lower resistivity than chemical vapor-deposited (CVD) tungsten. The crystallinity and electric conductivity of an under-layer does not affect the electric resistivity of tungsten film. The low resistivity of PVD tungsten originates from a large grain size. PVD tungsten with large grains is free from grain boundary scattering, while CVD tungsten with small grains exhibits grain boundary scattering. As film thickness decreases down to the mean free path of tungsten, the surface scattering effect surpasses the grain boundary scattering effect. Consequently, the resistivity of PVD and CVD tungsten becomes equal because surface scattering increases the resistivity of both large- and small-grained films. The same rule is adaptable for the resistance change of a narrow line structure. CVD tungsten shows a high resistance because of grain boundary scattering originating from a small grain size. However, if the line-width is reduced to the mean free path, grain boundary scattering disappears, making surface scattering only the factor that increases electric resistivity. Thus, CVD tungsten shows the same resistance as PVD tungsten in a very narrow line structure.
8013
Wolfgang Henke
In dynamic imaging static lens distortion has a detrimental impact on the image contrast of dense mask patterns and leads to blurring of images of isolated features, thus decreasing process window in lithographic applications. A new method is proposed here for combining transversal vibrations caused by the stage synchronization error of reticle and wafer stages and the lens distortion in lithographic step-and-scan exposure systems. Merging the descriptions of both effects into a unique characteristic function, the combined probability density function, allows us to quantify the individual contributions to the total result, to assess the mutual interaction of both effects and finally to calculate the impact of the combined effects on the projected dynamic image. The characteristic function characterizes the resulting image blur and is used here to study the impact of specific lens distortion fingerprints. The method presented here is also applicable to other types of transversal vibrations and can be extended to combine several vibration effects into one characteristic function.
8019
Masahiro Sakai, Takashi Egawa, Maosheng Hao and Hiroyasu Ishikawa
We have systematically studied the effect of various interlayers on epiwafer bowing in AlGaN/GaN high-electron-mobility transistor (HEMT) structures on sapphire substrates by using an interlayer insertion method. The interlayers including aluminum increase the epiwafer bowing independently of growth temperature. However, both InGaN-based interlayers and low-temperature-deposited GaN interlayers suppress the epiwafer bowing. When multilayers are used as interlayers, the effect on epiwafer bowing is enhanced compared with that in the case of using single interlayers. The minimum bowing values were lower than 10 µm, which were accomplished by using multilayers. The crystalline qualities of AlGaN/GaN HEMTs with interlayers were satisfactory and the performance of these HEMTs was better than that of conventional AlGaN/GaN HEMTs without interlayers. Epiwafers with reduced bowing exhibited a reduced strain in GaN layers compared with that of GaN layers with larger bowing, which indicates that strain control in the epitaxial layers is important for reducing the epiwafer bowing.
8024
Yoshiyuki Sato, Isao Sakaguchi and Hajime Haneda
In order to understand the possibility of using gallium for modern silicon device fabrication in a low-thermal-budget era, we investigated the diffusion characteristics of ion-implanted gallium in silicon during 850°C annealing. We obtained the following results. (1) About 70% of 5×1013 cm-2 implanted gallium remains in silicon after the annealing. (2) Transient enhanced diffusion occurs during the early stage of the annealing. (3) The intrinsic diffusion characteristic obtained is consistent with that extrapolated from a higher temperature region. (4) The sheet resistance of the gallium-doped layer is roughly the same as that of the boron-doped one.
8026
Takenobu Yoshino, Nobuhiro Hata and Takamaro Kikkawa
The evidence of copper (Cu) ion drift in low-dielectric-constant (low-k) interlayer films was obtained for the first time by transient capacitance measurements. The charge states of the electronic states related to Cu+ ions drifted into the low-k layer of Cu/low-k/SiO2/Si metal–insulator–semiconductor capacitors were altered by photo?electron injection from the p-Si substrate by applying a positive bias to the Cu layer under UV light illumination. A decrease in the time-dependent transient capacitance C(t) was attributed to the thermal emission of electrons from Cu-related electronic states. The trap level of the Cu-related electronic states was estimated to be approximately 0.9 eV below the conduction band edge in the bandgap of the low-k material.
8028
Chao-Min Cheng and Ren-Haw Chen
A novel work chamber cover design for a spin coater that can reduce the effect of airflow on spin coating is presented. The experimental results reveal that superior uniformity and film area can be obtained by utilizing the modified design instead of the conventional design.
8030
Katsushi Akita, Takao Nakamura and Hideki Hirayama
We fabricated InAlGaN quaternary ultraviolet (UV)-light-emitting diodes (LEDs) on GaN substrates with a low dislocation density and on GaN templates consisting of n-GaN on sapphire substrates, and compared the characteristics of these LEDs. A UV LED on a GaN substrate showed a considerably higher output power than that on a GaN template and no saturation even at an injection current of 500 mA. Cathodeluminescence images indicated the advantages of using GaN substrates in InAlGaN UV LEDs.
Magnetism
8032
Kuoshe Li, Jing Xu, Hongchuan Yang, Yongqiang Yuan, Dunbo Yu, Qiming Ying and Shengen Zhang
The effect of solidification rate on the preferred orientation, microstructure and magnetostriction of (Tb0.3Dy0.7)Fe1.95 alloys was investigated using a new “one-step” process. The functional relation between the crystal growth rate and the traction velocity was determined by the analysis of heat transfer. Results demonstrate that the crystal grows synchronously with the traction velocity as the traction velocity U≤360 mm/h. When the thermal gradient GL≈100°C/cm and the solidification rate V=70 mm/h, samples with a <110> preferred orientation can be obtained. Samples with the <110> preferred orientation show a large low-field magnetostriction and an apparent compressive stress effect. With an increase in solidification rate, the crystal growth mode will transform gradually from the cellular and primary dendrite morphology into the developed dendrite morphology. Crystals growing in the cellular morphology have a <110> preferred orientation. In the primary dendrite, the preferred orientation is <113>. However, a poor magnetostrictive performance and failed preferred orientation are observed in the developed dendrite.
8036
Keiichi Koyama, Masanari Sakai, Takeshi Kanomata and Kazuo Watanabe
The structural transformation of a ferromagnetic Mn1.07Co0.92Ge compound was studied by magnetization measurement in magnetic fields up to 13 T and powder X-ray diffraction measurement in fields up to 5 T. The Curie temperature is determined to be 275 K. At room temperature, the compound has a hexagonal Ni2In-type structure, and it transforms diffusionlessly into an orthorhombic TiNiSi-type structure at MS=210 K with hysteresis. The cell volume expands by 5.3% and the magnetic moment increases by 24% at MS, accompanied with this martensitic structural transformation. In addition, we confirmed for the first time that magnetic field induces the structural transformation in Mn1.07Co0.92Ge at the vicinity of MS.
8040
Yuji Itoh, Takehiro Aoyagi, Arata Tsukamoto, Katsuji Nakagawa, Akiyoshi Itoh and Toshikazu Katayama
We fabricated FePt films on thermal-oxidized Si substrate by rapid thermal annealing (RTA), and obtained (001)-oriented FePt/L10 small grains. By varying the total thickness of as-deposited Pt/Fe layers, we obtain isolated FePt/L10 crystalline grains after RTA. The smallest average grain size was only 7 nm in diameter. These as-deposited Pt/Fe layers were uniform and have a smooth surface; however after RTA, cohesion occurred and FePt/L10 small crystalline particles appeared. After annealing, the FePt grains are embedded in the SiO layer. The films showed high Hc and high remanence ratio in the direction perpendicular to the film plane. The FePt Curie temperature is appoximately 400°C.
Optics and Quantum Electronics
8044
Jung-Il Kim, Yanyu Wei, Gun-Sik Park, Dong-Wook Kim and In-Kyeong Yoo
In e-beam projection lithography (EPL) systems, the relativistic focus condition was derived and compared with the nonrelativistic focus condition to investigate the effect of a large acceleration field related to the initial energy and emission angle of an electron emitted from a cathode. The difference in confusion disk between the relativistic and nonrelativistic limits was 180 nm when 1 eV was used as the initial energy of the electron. The variation in the initial energy of the electron was more sensitive to change the confusion disk than the acceleration field. In this comparison, the focusing parameters are used as follows: axial magnetic field, 0.5 T, distance between the mask layer and the workpiece, 2 mm and emission angle of the electron, 30°.
8048
Kyo Inoue and Kaoru Shimizu
The generation of quantum-correlated photon pairs by spontaneous four-wave mixing (FWM) in an optical fiber is studied. After demonstrating the time correlation between signal and idler photons, the influence of spontaneous Raman scattering on extracting photon pairs is experimentally investigated. It is shown that a short pulse and a narrow filtering bandwidth are preferable for obtaining correlated photon pairs with a better signal-to-noise ratio via FWM in an optical fiber.
8053
Kejian Yang, Shengzhi Zhao, Guiqiu Li and Hongming Zhao
By using GaAs as both an output coupler and a saturable absorber, we demonstrate a laser-diode end-pumped passively Q-switched Nd:GdVO4 laser for the first time to our knowledge. Pulses with widths ranging from 22.8 ns to 119.2 ns and repetition rates ranging from 333 kHZ to 761 kHZ are generated. At the maximum incident pump power of 4.77 W, the maximum average output power of 1.23 W is obtained, corresponding to an optical conversion efficiency of 25.7% and an average slope efficiency of 37.9%. A rate equation model is introduced to theoretically analyze the results obtained in the experiment, in which the Gaussian spatial distribution of the intracavity photon density and the longitudinal distribution of the photon density along the cavity axis are taken into account. The results of numerical calculations of the rate equations are consistent with the experimental results. A further discussion on the pulse repetition rates is presented.
8059
Montian Tianprateep, Junji Tada, Tomotaka Yamazaki and Fumihiko Kannari
The influence of a frequency chirp induced in an excitation laser pulse by varying the second- and third-order group-delay dispersions with liquid-crystal spatial light modulators in a Fourier pulse shaper on the supercontinuum (SC) spectra generated from microstructured fibers is experimentally studied. In close-loop control experiments, an input laser pulse is adaptively shaped by referring to the SC spectrum. Optimum input laser pulse shapes for obtaining the highest power at one or two narrow bands in the SC spectrum are successfully obtained using a self-learning adaptive pulse shaping technique.
8064
Tetsuo Sato, Norio Nagayama and Masaaki Yokoyama
Entirely transparent polysilane films, which were modified by ultraviolet (UV) light exposure to decrease their refractive index, were used as phase masks. The phase modulation amount of the transmitted laser, which was estimated using a Mach–Zehnder interferometer, can be modified linearly by UV exposure dose, indicating that the polysilane films have a very useful property of controllability by UV light exposure. Optical correlation using the polysilane phase masks as the information-carrying medium was then demonstrated. A matched-filter four-wave-mixing optical correlator using a photorefractive (PR) polymer as a real-time holographic data storage medium was used. A large correlation signal difference of over 3.6-fold was obtained for the striped polysilane phase masks that were arranged orthogonally or in parallel. This is due to the distinct encoded images in the polysilane phasemasks originating from the large refractive index modification in polysilane films of approximately Δn=-0.14 at λ=633 nm. The high transparency in the entire visible region and such large refractive index modification make polysilane films difficult to forge, and allow us to use phase masks in pattern recognition systems for security verification.
8069
Junji Hirohashi, Kazuhiro Yamada, Hiroyuki Kamio and Shiro Shichijyo
A compact integrated folded Šolc filter using periodic 90°-domain structures in KNbO3 single crystals is proposed. The periodic 90°-domain structures with a period of 200 µm are fabricated on 0.7-mm-thick (101)-cut KNbO3 single-crystal plates by applying an electric field of 200 V/mm. Using the refractive index change between 90°-domains, we can construct the same arrangement for a folded Šolc filter. The measured filtering properties agree well with the calculated ones. This is the first demonstration of an optical filter using periodic domain structures other than 180°-domain structures. It is now possible to fabricate a compact folded Šolc filter without inconvenient alignment.
8075
Yung-Shun Yang, Yi-Pai Huang, Han-Ping D. Shieh, Meng-Chang Tsai and Ching-Yu Tsai
A novel transflective liquid crystal display (LCD) with a microtube array structure to make the most effective use of backlight was demonstrated. By redefining the area ratio of transmissive and reflective regions, high light efficiency in both transmissive and reflective subpixels can be attained. Based on the principle of light collection, a simulation model was established to characterize the features and optimize the microtube array structure. Furthermore, a typical TFT-LCD process was utilized to fabricate the optimized structure. A prototype microtube array was fabricated, and the measured enhancement of backlight utilization efficiency can be as high as a factor of 1.81.
8080
Yasuo Kokubun, Atsushi Miura and Sumihiro Inokuchi
We propose a new type of photonic gate-array technology (photonic GA) for the flexible design of photonic integrated circuits. It consists of busline waveguides with a cross-grid configuration and fundamental elements, such as microring resonators, bends, branches and switches, stacked on the crossing points of the busline waveguides. In this study, vertically coupled waveguide bends and branches (VCWBs) were designed and demonstrated. Very compact devices with 20–50 µm coupler length and 25–40 µm bending radius were realized using high-index contrast waveguides. Polarization-independent compact bends and branches with low loss (<1.6 dB) were demonstrated. Their wavelength dependence was also confirmed to be very small in the wavelength range of 1500–1580 nm.
8085
Takao Saito
To design a dual-layered disk with a high reflectivity, we selected the organic recording material. However, organic recording materials for violet lasers have low sensitivity. To solve this problem, we developed new methods of designing a translucent reflective layer. One of these methods is the control of the extinction coefficient k from 0.1 to 0.3. In addition, a method of controlling k is the utilization of the intermediate. The intermediate used in this study is the state between a metal and a compound of saturation. These methods show a potentiality of the design for a dual-layered disk with organic recording materials.
8089
Yoshio Hayasaki, Hayato Takagi, Akihiro Takita, Hirotsugu Yamamoto, Nobuo Nishida and Hiroaki Misawa
We investigated the processing of a human fingernail surface using a tightly focused femtosecond laser pulse. The processed structure in the fingernail surface is strongly dependent on the focus position and irradiation energy of the single laser pulse. We observed a ring, a simple pit, a small pit with a surrounding uplift, an irregular jagged surface, and a swell containing a void, depending on the focus position. We also observed a sudden change in the size of the processed structure according to the irradiation pulse energy. From a linear theoretical estimation based on the diffraction of the laser beam, we found that the sudden change is primarily due to the diffraction pattern generated by the circular aperture of the objective lens. We also describe the processing features by comparing the structures processed in a fingernail with those processed in glass.
8094
Tatsuo Uchida, Takahiro Ishinabe, Takanori Nakayama, Makoto Suzuki and Tetsuya Miyashita
Reflective liquid crystal displays (LCDs) are attracting a great deal of interest for use in portable information systems because of their extremely low power consumption, very thin profile and light weight; also, their color does not wash out in outdoor use. We proposed a new type of reflective LCD composed of a light-diffusing film over the display and mirror electrode inside of the LC cell. This reflective LCD has the advantages of high resolution, a wide viewing angle, fast response and full color capability, and is promising as a future high-quality reflective full-color AM-LCD.
8101
Peng-Chun Peng, Wei-Ren Peng, Wen-Piao Lin and Sien Chi
A novel dynamic encoder/decoder based on fiber Bragg gratings for optical security systems is presented in this paper. The encoded and decoded optical communication signals can be realized by controlling the optical switches and fiber Bragg grating arrangement. We describe the system operation principle and set up an experimental encoder/decoder to demonstrate the feasibility of the proposed scheme.
Optical Properties of Condensed Matter
8103
Seongtae Park, Yonghwa Chung, Kiwan Jang, Hong-Guo Liu, Yong-Ill Lee and Changdae Kim
Sm-doped Mg0.5Sr0.5FCl0.5Br0.5 mixed crystals were prepared by three different methods and their optical properties were studied through emission spectroscopy and the hole-burning technique. While the sample prepared in air (MA) shows the transition lines due to Sm3+-ions, the sample prepared in hydrogen (MH) shows Sm2+ emission lines. We could also observe the reduction effects of Sm3+ in the X-ray irradiated sample (MX). The spectral holes were burnt within the 7F0–5D1 transition band and the hole-burning efficiencies in the MH and MX samples were measured at room temperature and 9 K, using the one color and two color hole-burning techniques. The hole-burning efficiency decreased with the lowering of the temperature, and almost no hole signal was found at 9 K in the MX sample. We thus performed a site selective excitation experiment to reveal the origin of the decrease of hole-burning efficiency with decreasing temperature. On the basis of these results, we found that the energy used to excite the Sm2+-ions in the hole-burning process is also simultaneously used to excite the residual (or unreduced) Sm3+-ions, and we also found that the excitation efficiency of the Sm3+, compared to that of Sm2+, increased with decreasing temperature. As a result, we consider that the residual Sm3+-ions strongly affect the hole-burning efficiency and the photoluminescence properties of the Sm2+ by depriving the Sm2+-ions of excitation energy.
8107
Isuke Ouchi, Ikuo Nakai, Mitsumasa Ono and Shin-ichi Kimura
Out of two species of fluorescence of polyethylene terephthalate (PET), intrinsic fluorescence (I), which was caused by short-wavelength (<320 nm) excitation, was dominant for very thin films (less than a few microns), while, trap fluorescence (II), excited at longer wavelengths (>320 nm), was much more intense for thick films. This fact was confirmed, and explained by an elementary formula. Effects of polymerization catalysts onto the fluorescence spectra were found to be indirect; namely, peak positions of the fluorescence were not affected by catalyst systems, while fluorescence intensity was sensitive not only to catalyst systems but also to minute differences in local surroundings. Effects of catalyst systems on the weak absorption bands at 338 nm and 354 nm were examined. Integrated excitation spectra were resolved; therein, a hidden n→π* state was found. Most likely, this is related with the weak absorptions at 338 nm and 354 nm, which are the major origins of fluorescence (II).
8115
Yuan-Fong Chau, Tzong-Jer Yang and Din-Ping Tsai
A series of finite difference time domain (FDTD) methods for obtaining more insight on the near-field distribution of subwavelength aperture and fiber probes are numerically investigated. The treatment of dispersive materials in a time-domain-dependent fashion is considered. Several significant factors are also considered successively, e.g., the near-field distribution of subwavelength aperture in an infinite aluminum plane with sample interactions, the characteristics between the near-field distribution and the depolarization phenomenon, different types of three-dimensional tip (noncoated and metal-coated) used to illuminate a photosensitive sample and the polarization of an incident electromagnetic field. Moreover, the FDTD designs of two types of improved probe are illustrated and a suggestion for fabricating an optimal probe is given. Our proposed structures will yield useful information and guidelines for designing high-performance near-field probes.
Electrical Properties of Condensed Matter
8126
Eun Jung Lee, Jaill Jeong and Young Ho Han
The effects of Dy addition on the electrical properties of BaTiO3 have been studied by analyzing equilibrium electrical conductivity as a function of temperature, oxygen partial pressure (Po2) and composition. In(Ba1-xDyx)TiO3+0.5x, the conductivity follows the -1/4th power dependence of Po2 at high oxygen activities, which supports the notion that cation vacancies are generated to compensate for Dy donors, and is nearly independent of Po2 where electron compensation prevails at a low Po2. The incorporation of “Dy3+” ion into Ti sites up to 2.0 mol% showed the typical acceptor-doped behavior of the Dy3+ ion. Curie temperature (Tc) was systematically decreased when the Dy3+ ion was incorporated into Ti sites.
8130
Jixiao Li, Yewen Zhang, Feihu Zheng and Changshun Wu
In this paper, laser induced pressure pulse (LIPP) method and electrostatic force microscopy (EFM) method are utilized to investigate the distribution of space charge in cross-linked polyethylene (XLPE), and the action on the groups of XLPE by the accumulative charge are investigated by infrared spectroscopy (IR) method. It was found that space charge in the sample has obvious influence on the vibration of chemical group, especially on group OH, group CH3 and group CH2. Group vibration affected considerably by space charge in XLPE sample locates on the interface between crystalline and amorphous domains. The experimental results also indicate that positive charge compared to negative charge has a different effect on bands.
8135
Jianding Yu, Paul-François Paradis, Takehiko Ishikawa and Shinichi Yoda
BaTiO3 samples were aerodynamically levitated and laser melted under containerless conditions and quenched with a twin-roller apparatus. The microstructure of the BaTiO3 samples became more refined with increasing undercooling level prior to quenching. A two-phase microstructure consisting of perovskite BaTiO3 and hexagonal BaTiO3 was observed in the samples. The dielectric constant of the samples decreased due to the presence of hexagonal BaTiO3. The formation of the two-phase microstructure was explained by a significant release of crystallization heat during rapid solidification. Roller-quenching experiments also revealed that BaTiO3 is not easily formed into glass.
8139
Na-Fu Wang, I-Tseng Tang, Chen-I Hung and Mau-Phon Houng
A novel GHz-band low-loss large-bandwidth microwave microstrip, surface-acoustic-wave (SAW) filter used in spread spectrum communication systems was designed and fabricated. The 16 µm input/output-interdigital-transducer (IDT) LiNbO3 61.00 MHz SAW filter, with microwave square open-loop resonators, has an insertion loss of -3.987 dB. This device can also be used as a 1.064 GHz microwave microstrip SAW filter with four cross-coupled microstrip square open-loop resonators and two planar interdigital capacitors. It was found that this novel device has an insertion loss (S21) of -2.962 dB, a reflection loss (S11) of -25.497 dB and a 3 dB bandwidth of 800 MHz. The characteristics of these low-loss large-bandwidth (BW=80%) microwave microstrip SAW filters are influenced by the interaction between electromagnetic waves and piezoelectric SAWs. To confirm this claim, devices of identical design were fabricated on the three types of substrate: i) 128°-rotated YX-cut lithium niobate (LiNbO3) with a high electromechanical coupling coefficient (K2=5.3%), ii) GaAs S-I photo-piezoelectric material with a low electromechanical coupling coefficient (K2=0.07%) and iii) non-piezoelectric SiO2/Si material with a thickness of 9000 Å. The performances of these devices were significantly different. We will apply the principle of SAWs and the equivalent circuit of IDT to prove our experimental results.
8146
Richard E. Eitel, Thomas R. Shrout and Clive A. Randall
Recently, a new family of piezoelectric perovskite materials based on the solid solution (1-x)BiMeO3–xPbTiO3 (where Me is one or a combination of cations with sum valance +3) have been developed. Specifically, the (1-x)BiScO3–xPbTiO3 (BS–PT) system is noted for its high Curie temperature (TC=450°C) and large piezoelectric coefficient (d33>460 pC/N). Lanthanum additions, as a donor dopant in the piezoelectric system Pb(Zr,Ti)O3 and other lead based perovskite materials, are commonly utilized to both lower Curie temperature and increase piezoelectric coefficients. In the current work, the effect of lanthanum substitution in the BS–PT system is reported. The results of lanthanum additions in the BS–PT system show no large enhancement of the piezoelectric properties. However, fine grain sized materials (<1 µm), desirable for improved mechanical properties, with Curie temperature of 365°C and high piezoelectric coefficient (d33=465 pC/N) were successfully produced. The results of this study suggest that for low lanthanum additions, <5 mol%, internal compensation mechanisms in the BS–PT system dominate piezoelectric performance. Further grain size effects in lanthanum doped BS–PT compositions provide the first experimental evidence of significant extrinsic contributions to the piezoelectric properties in this system.
Structure and Mechanical and Thermal Properties of Condensed Matter
8151
Michinori Honma and Toshiaki Nose
A liquid crystal (LC) cell is fabricated by combining a uniformly rubbed substrate and a periodically rubbed substrate. The LC cell is applied to an LC reflective beam deflector. It is theoretically derived that the sawtooth phase profile is achieved under optimum LC molecular orientation conditions. That is, the retardation of the LC cell is one-half the wavelength and the polarization direction of the incident ray makes an angle of ±π/4 with respect to the uniformly rubbed direction. Optical properties of the fabricated LC cell are experimentally investigated. It is revealed that the diffraction efficiencies of two first-order light beams (m=±1, m denotes diffraction order) and the zeroth-order light beam (m=0) can be controlled by the applied voltage. Since the strong diffraction of the first-order light beams (m=±1) can be selected at will, the fabricated LC cell functions as an electrically controllable beam deflector.
8156
Minhua Lu
It has been discovered that liquid crystal alignment on an inorganic surface such as SiO2 depending on the dielectric anisotropy of the liquid crystal materials. The liquid crystals with positive dielectric anisotropy tend to align parallel or parallel with a tilt to the SiO2 substrates, while the liquid crystals with negative dielectric anisotropy tend to align perpendicular or perpendicular with a tilt to the SiO2 substrates. The pretilt angle of liquid crystals with positive dielectric anisotropy aligned by single evaporation of SiO2 tends to be at a complementary angle to that of liquid crystals with negative dielectric anisotropy. Van der Waals interaction is responsible for such orientation when all the other types of interactions between liquid crystal molecules and surfaces are absent or negligible. In addition, this work provides a simple and reliable method for high quality homeotropic and tilted homeotropic liquid crystal alignment.
8161
Hideo Fujikake, Hiroto Sato, Fumito Isaka, Takeshi Murashige, Hiroshi Kikuchi, Taiichiro Kurita and Fumio Sato
In this paper, we discuss a self-restoration phenomenon affecting smectic layer deformation and molecular alignment in monostable ferroelectric liquid crystals used for flexible displays. First, the mechanical stability of tilted `bookshelf' structures of smectic layers anchored on substrates using alignment layers was examined by precisely shearing two substrates. The microscopic texture of a monostable ferroelectric liquid crystal showed tolerance to shearing and the self-restoration was thought to be due to a smectic layer reconnection phenomenon, whereas a conventional bistable ferroelectric liquid crystal film generated alignment defects due to the tilting of the stable molecular direction. We then fabricated a flexible monostable device containing fine polymer fiber networks, where anchoring of molecularly aligned polymer fibers led to the observation of monostable liquid crystal switching. A uniform liquid crystal alignment was maintained in a 100 mm×100 mm device even after bending more than 10000 times at a minimum radius of curvature of 20 mm, due to the enhanced shearing tolerance of the smectic layer structure.
8168
Arvind K. Gathania, Buta Singh and Kuldeep K. Raina
Switching dynamics have been investigated in homogeneously aligned 5 µm thin cell filled with (FLC-6980) ferroelectric liquid crystal mixture using field reversal method and Sawyer Tower method. The effect of applied voltage and temperature on switching time has been studied. It has been observed that switching time decreases with the increase of applied voltage and also with the rise of temperature. The tortional viscosity has been calculated. It shows Arrhenius type behavior. Hysteresis loop has been studied at different temperatures and at different frequencies. It has been observed that coercive field (Ec) increases with the increase of frequency. Ec has non-zero value at zero frequency, which is important characteristics of ferroelectrics. Ec increases near the transition temperature.
8173
Kazuyuki Hiraoka, Taku Maruyama and Yoko Noguchi
The dielectric and switching behaviors have been investigated using cells of varying thickness of an antiferroelectric liquid crystal, (S)-4-(1-methylheptyloxycarbonyl) phenyl 4'-octylbiphenyl-4-carboxylate (S-MHPBC) which shows antiferroelectric and ferrielectric subphases, designated as the AF and Sm-Cγ* phases, respectively, in a temperature region above the Sm-CA* phase. While the tristable switching with a characteristic dc threshold and hysteresis is clearly observed in the AF phase in a cell thicker than about 8 µm with homogeneous alignment, a thresholdless bistable switching is confirmed in a cell thinner than 3.9 µm. The emergence of thresholdless bistable switching means that confinement of the antiferroelectric liquid crystal to the thin cell induces the phase transformation from the AF and Sm-Cγ* phases to the frustrated ferroelectric Sm-C* phase. In addition, a novel dielectric relaxation appears in the temperature region of the AF and Sm-Cγ* phases in the liquid crystal confined to a cell 3.9 µm thick. We concluded that the novel dielectric relaxation appearing in the confined antiferroelectric liquid crystal is associated with a Goldstone mode of the frustrated ferroelectric Sm-C* phase transformed from the AF and Sm-Cγ* phases.
8179
Jeoung-Yeon Hwang, Ki-Hyung Nam, Jong-Hwan Kim, Dae-Shik Seo and Dong-Hack Suh
We have investigated the generation of pretilt angle for nematic liquid crystal (NLC) alignment by the rubbing alignment method in two types of polyimide (PI) surfaces using polymer substrates. The NLC pretilt angle generated by the rubbing alignment method on a polymer substrate is about 3 deg, and it tends to decrease as rubbing strength increases. The pretilt angle measured is about 1.7 deg smaller on the glass substrate than on the polymer substrate. We obtain an atomic force microscope (AFM) image of the rubbed PI surface with a polymer film that has formed the micro-groove structure at a low curing temperature (120°C). However, no grooves were obtained on the glass substrate at the same temperature. It is considered that this alignment may be attributed to the roughness of the micro-groove substrate. Therefore, LC alignment is affected by the type of topographical structure. The pretilt angle was larger on the rubbed soluble-type PI surface with a higher imidization rate using the polymer film; the pretilt angle on the polyimide surface is attributed to high imidization rate of the PI. Consequently, flexible LCDs required at the low-temperature processing should use the soluble-type PI as the alignment layer. Also, good eletrooptical (EO) curves were observed for the TN-LCD using the polymer substrate at a low curing temperature.
8185
Hidemi Kato, Akihisa Inoue and Ho-Sou Chen
A model calculation based on the fictive stress is conducted on the creep deformation of a Zr55Al10Ni5Cu30 bulk metallic glass near the glass transition temperature under a constant applied load. The model successfully reproduces qualitatively the main features associated with creep viscoelastic behavior at various initial applied stresses, σ0. Whenever the value of σ0 exceeds the critical flow stress, σc (=80 MPa), the flow stress curve, log
σ–log t, shows a dip due to stress-induced softening, and the stress dependent viscosity, η(σ), initially decreases. It attains a minimum value, ηmin(σmin), as the glass attains an equilibrium structure corresponding to the imposed stress, σ. For stress, σ, less than the viscosity minimum stress, the η(σ) curves all decrease and tend to merge together, and may be fit with a master curve established previously. This reveals that the model curves depend mainly on the choice of structural relaxation time, λfic, and little on the value of Young's modulus used in a simple Maxwell model calculation. We found a systematic deviation between the model and experimental curves, such that the model tends to overestimate the structural softening in case with smaller initial applied stress. The cause of this deviation is unclear and is a subject for future studies.
8190
Mai Pham Thi, Henri Hemery, Olivier Durand and Hichem Dammak
Textured (1-x)PbMg1/3Nb2/3–xPbTiO3, (PMN–xPT), ceramics were synthesised by the homo template grain growth (HTGG) technique using PMN–PT nanoparticles and cubic single crystals seeds (templates). The orientation of the templates was obtained during tape casting. Sintering was performed under controlled O2/PbO partial pressure. The fraction of textured ceramics was estimated by the Lotgering method at RT and above Tc. Textured tape cast ceramics display a quasi complete (001) texture ( f=0.99) in the cubic phase. The (011) pole figure reveals a (001) fibre texture in accordance with the optical observation. The texture was obtained in all volumes of ceramics and only a small orientation distribution was observed (<6°).
8195
Makoto Iwai, Makoto Ohmori, Takashi Yoshino, Shoichiro Yamaguchi and Minoru Imaeda
Epitaxial MgO-doped lithium niobate [LiNbO3 (LN)] single-crystal thin films with a good surface morphology and crystallinity were grown on lithium tantalate [LiTaO3 (LT)] substrates by low-pressure metalorganic chemical vapor deposition (MOCVD). Dipivaloylmethanatolithium [Li(C11H19O2); Li(DPM)], pentaethoxyniobium [Nb(OC2H5)5; Nb(OEt)5] and bisacetylacetonatomagnesium [Mg(C5H7O2)2; Mg(Acac)2] were used as volatile metalorganic precursors. The crystallinity of the films was investigated by X-ray diffraction measurements. Reciprocal space mapping indicated that good epitaxial growth had been achieved. The film surface was sufficiently smooth to act as a waveguide. The ordinary and extraordinary refractive indices (no and ne) of these as-grown MgO-doped LN thin films were measured to be no=2.2844 and ne=2.1908 at 633 nm, respectively, which were close to those of 3 mol% MgO-doped stoichiometric LN.
8199
Toshihide Ito, Hiromitsu Kato, Tomohiro Nango and Yoshimichi Ohki
Amorphous hafnium silicate films with several composition ratios were deposited on Si substrates by plasma-enhanced chemical vapor deposition, and their energy band profiles were studied by X-ray photoelectron spectroscopy. The band gap energy estimated from the energy loss spectrum of O 1s electrons decreases monotonically and then approaches a constant value with an increase in hafnium content. The valence band offset and the conduction band offset were estimated using the sample sandwiched between an evaporated Au electrode and the Si substrate. Although the two offsets decrease until they become almost constant with an increase in hafnium content at both silicate/Si and silicate/Au interfaces, they hold values higher than that necessary for a high-k dielectric material to maintain a good insulating performance for all the deposited silicates.
8203
Tetsu Mitsumata, Akiyasu Nagata, Kazuki Sakai and Takashi Taniguchi
The effects of magnetization direction on the complex modulus of magnetic gels, which consist of magnetized barium ferrite and poly vinyl alcohol, have been investigated. The modulus after magnetization showed anisotropy depending on the geometry of magnetization and strain direction. It was also confirmed that the modulus change occurred without volume change. This strongly suggested that the modulus change of magnetic gels originates not from a decrease in the volume of the magnetic gels but from the magnetic dipole interactions between ferrites dispersed in the gels.
8208
Tsuyoshi Takami, Hiroshi Ikuta and Uichiro Mizutani
We studied the thermoelectric properties of polycrystalline samples of the n=1–5 members of the homologous series An+2Con+1O3n+3, where A represents Ca, Sr, or an appropriate combination of Sr and Ba. The electrical resistivity (ρ) decreased monotonically with increasing temperature in all samples. With increasing temperature, the Seebeck coefficient (S) of the n=1 sample decreased, while S of the n=2–5 samples increased up to about 900 K. The power factor (=S2/ρ) increased with temperature in all samples. The n=2 sample exhibited the largest power factor, which was 0.76 µW/K2cm at 1070 K. The thermal conductivity (κ) of this sample was very small, 1.3 W/K·m at 773 K, and the dimensionless figure of merit (ZT=S2T/ρκ) increased with temperature. By extrapolating κ to high temperatures, ZT was estimated to reach 0.054 at 1070 K in the case of the n=2 sample.
8213
Wenyong Duan, Masao Fujiwara and Yoshifumi Tanimoto
In situ observation of thermal convection of a benzene solution of a photochromic compound, 1,2-dicyano-1,2-bis(2,4,5-trimethyl-3-thienyl)ethene (CMTE), was carried out in vertical high magnetic fields. When a CMTE solution is irradiated using a UV laser from the bottom of a vessel, CMTE undergoes photoisomerization to its photoisomer (PI) and the color of the PI solution changes from pale yellow to red. At zero field, the PI solution formed on the bottom surface of the vessel is removed at a 5 s delay after laser excitation and then moves upward. At -1300 T2m-1, it is removed at a 9 s delay and moves upward, whereas at +1000 T2m-1, it does at a 3 s delay, moves upward and then finally moves downward. All these results are discussed in terms of the magnetic force in the CMTE/PI solution.
8217
Hitoshi Kato, Hideki Katayanagi, Yoshikata Koga and Keiko Nishikawa
A novel method of determining a liquid–liquid phase boundary was developed. This method is based on our discovery that a nascent low-density phase is attracted to the center of a Rankine vortex at the onset of phase separation. Thus a liquid–liquid phase boundary is detected easily, rapidly, and accurately. The phase diagrams of the ternary systems NaCl–H2O–1-propanol and NaCl–H2O–1-butanol were obtained by this method. The results matched well with literature values.
Surfaces, Interfaces, and Films
8219
Hisanori Matsumoto, Kengo Asai, Nobuyasu Kobayashi, Shiro Nagashima, Atsushi Isobe, Nobuhiko Shibagaki and Mitsutaka Hikita
Highly c-axis-oriented aluminum nitride (AlN) films were deposited by reactive rf magnetron sputtering on molybdenum (Mo) films for film bulk acoustic wave resonators (FBARs). We investigated the effect of underlayer materials on the crystalline qualities of AlN/Mo bilayers. AlN/Mo bilayers were sputter-deposited on silicon, silicon dioxide and AlN underlayers under the same sputtering conditions. The results revealed that the AlN underlayer markedly improves Mo film quality, and this greatly affects the orientation of the upper AlN film. By applying the AlN underlayer, we obtained a full width at half maximum of 2.3° from the X-ray diffraction rocking curve of the AlN film grown on a Mo film. We propose that the large surface energy of AlN is the reason the AlN underlayer improves the quality of the Mo films. In the FBAR with the AlN underlayer, a quality factor of 800 and an effective electromechanical coupling coefficient of 5.7% were achieved.
8223
Hiroyuki Kageshima, Masahi Uematsu, Kazuto Akagi, Shinji Tsuneyuki, Toru Akiyama and Kenji Shiraishi
The excess Si emitted from the Si-oxide/Si interface is studied using the first-principles calculations. It is shown that the excess Si can have many (meta-) stable positions around the interface. In addition, some positions in the oxide do not have any dangling bonds or floating bonds in contrast to those in the bulk crystalline Si. The results indicate that the emitted Si can be located in the oxide layer but they do not necessarily cause charge traps in the oxide. The emitted Si atoms are thought to just be oxidized and absorbed into the oxide while a portion of them cause the E' centers, the Pb centers or charge traps.
8227
Chi-Jung Chang, Shinn-Jen Chang, Feng-Mei Wu, Mei-Wen Hsu, Wanda W. W. Chiu and Kevin Chen
UV curable jet inks were ejected by a bubble-jet cartridge onto a glass substrate and fixed by a UV curing reaction to prepare LCD color filters. The rheology of jet inks was adjusted by changing the surfactants and monomers, in order to achieve good jetting directionality, uniform drop size and excellent wetting on the glass substrate. The surface energy of the substrate was modified by plasma treatment. The printed aqueous UV curable ink tends to move toward the O2-plasma treated ribs and this causes them to swell. This lowers the nanoindentation hardness of the ribs and deteriorates the uniformity of the printed color strips. Both the surface smoothness and color uniformity of the color strips were improved tremendously when the substrates were treated by CF4 plasma.
8234
Shingo Ohno, Daisuke Sato, Masato Kon, Yasushi Sato, Masato Yoshikawa, Peter Frach and Yuzo Shigesato
Titanium dioxide (TiO2) films were deposited on unheated nonalkali glass substrates by reactive midfrequency (mf) magnetron sputtering using dual cathodes with two Ti metal targets. In order to maintain the very high deposition rate, the depositions were successfully carried out stably in the “transition region” between the metallic and reactive (oxide) sputter modes using plasma control units (PCU). Very high-rate depositions of 12–70 nm/min were successfully achieved in the deposition of TiO2 films throughout the whole “transition region”. The as-deposited films deposited in the “oxide mode” had a polycrystalline anatase structure and exhibited both photoinduced hydrophilicity and photodecomposition of acetaldehyde. Whereas, all the as-deposited TiO2 films deposited in the “transition region” had amorphous structure, which did not exhibit photocatalytic activity. Such amorphous films deposited in the transition region were crystallized by postannealing in air at 200°C or 300°C and were then shown to have photocatalytic activity. Very thin TiO2 films with a thickness of 25 nm deposited in the transition region and postannealed at 300°C exhibited photoinduced hydrophilicity, whereas there was a clear thickness dependence of photodecomposition and the rather thick films of 300 nm exhibited pronounced photodecomposition.
8242
Shinichi Urabe, Kazuo Nishimura, Satoru Morita and Mizuho Morita
The reaction of hydrogen-terminated Si(100) surfaces with water during heating is analyzed using a combination of heating and cooling in thermal desorption spectroscopy. The reaction of the Si surface with water after hydrogen desorption is observed even at about 400°C at a low concentration of water molecules. An estimation method of surface hydrogen coverage by the combined measurement is proposed and the surface coverage as a function of temperature is estimated for quantitative understanding of hydrogen desorption and the subsequent reaction during heating. The combined measurement in thermal desorption spectroscopy is useful for revealing the reaction after hydrogen desorption during heating and for estimating the surface coverage.
8248
Motonobu Yukawa, Shunsuke Tatsumi, Hiroki Kitagawa and Shoji Iida
Although a Si(100) surface forms dimers to produce a stable structure, it contains defects such as steps which are thought to affect surface resistance. We measured the surface resistance of step-containing Si manufactured by an ordinary process, and discovered that Si exhibits angular characteristics (anisotropy). After performing measurements at 10° intervals over two revolutions of 360°, we found that the results are reproducible between the first and second turns, that the resistance is periodic with periods of 180° and 360°, and that the sheet resistances measured by the four-point-probe method are 7–13-fold greater than those calculated from the bulk resistance and thickness of the samples. Similar measurements were performed on step-free Si surfaces, which were found to exhibit almost no periodicity and had a sheet resistance is of 4–10-fold the bulk value. It was suggested that the periodicity of resistance is caused by the presence of many steps on the Si surface.
8253
Young-Hoon Shin and Yukihiro Shimogaki
We report on the deposition of Ti1-xAlxN films by chemical vapor deposition (CVD) system. Titanium tetrachloride (TiCl4), dimethylethylamine alane (DMEAA) and ammonia gas were used as sources. Chemical composition, microstructure and electrical resistivity were investigated at various deposition conditions by X-ray photoelectron spectroscopy, X-ray diffraction, scanning electron microscopy and 4 point probe. Al content, x, in Ti1-xAlxN film was varied from 0.04 to 0.79 in our experimental conditions. As deposition temperature increased in the temperature range from 220°C to 410°C, Al content decreased. Also, Al content in Ti1-xAlxN films shows a linear relationship with the partial pressure of DMEAA. Up to x=0.15, glancing angle XRD peaks showed only B1 (NaCl) cubic structures but further increasing of Al content leads to amorphous structure. For Ti0.76Al0.24N film, Cu(50 nm)/Ti0.76Al0.24N(20 nm)/Si substrate stack was prepared to examine the barrier property against Cu diffusion. Cu diffused into Si substrate through the Ti0.76Al0.24N(20 nm) film from 700°C, 30 min vacuum annealing. This result is higher than that of CVD-TiN(50 nm) barrier which failed at 400°C.
8258
Muh-Wang Liang, Tsung-Eong Hsieh, Chi-Chin Chen and Yuan-Te Hung
In this work, we studied the application of electroless copper/nickel (Cu/Ni) films deposited on aluminum nitride (AlN) substrates to high-frequency power GaAs device packaging. Experimental results showed metal films deposited on polished AlN surfaces possess a flatter surface, a finer grain structure, and a lower resistivity than those on unpolished surfaces. On unpolished AlN substrates, rough-surface-induced voids appear in the film interface during grain clustering, therefore deteriorating the electrical conductivity of the deposited layers. Pull-off tests revealed that the Cu/Ni films strongly adhered on both types of AlN substrates and that the adhesion strength exceeded 761 kg/cm2. The Ni film remained a mixture of amorphous and microcrystalline structures, and the Cu film was polycrystalline. The resistivity of the Cu/Ni film was decreased by annealing process, which in turn decreased the number of crystal defects in the films. Subsequent Pb-Sn solder bumping experiments indicated the amorphous Ni(P) film was a good diffusion barrier layer since Sn could not diffuse through it.
8267
Jae-Won Lim, Kouji Mimura and Minoru Isshiki
Glow discharge mass spectrometry (GDMS) was used to analyze a Ta target and Ta films for trace impurities. The Ta films were deposited on Si (100) substrate at substrate bias voltages of 0 V and -125 V using a non-mass separated ion beam deposition system. Although both Ta films were contaminated by impurities during the deposition, the Ta film deposited at a substrate bias voltage of -125 V showed lower impurity content than the Ta film deposited without the substrate bias voltage, which means that applying a negative bias voltage to the substrate decreased the total concentration of impurities. Furthermore, the concentration change of individual impurities in the Ta film is related to their ionization ratio in the argon discharge plasma. Considering the effect of the ionization potential of an individual impurity on the ionization ratio, purification by applying a negative bias voltage to the substrate results from Penning ionization and an ionization mechanism proposed in this study, as well as from the difference between the kinetic energies of Ta neutral atoms and Ta+ ions accelerated toward the substrate with/without a negative substrate bias voltage.
8273
Soohyon Phark, Zheong G. Khim, Beom Jin Kim, Byoung Jin Suh, Seokwon Yoon, Jinkwon Kim, Jin Mook Lim and Youngkyu Do
A film of Mn12O12(O2CC4H3S)16(H2O)4 molecules was chemically deposited onto an atomically flat Au(111) surface. The surface morphology of the film and the elastic property of the molecules were investigated using tapping-mode atomic force microscopy. The molecules were determined to be randomly arranged with an abnormal height distribution due to the anisotropic shape and bonding probability of the molecules. We also measured the amplitude-distance curves for the molecules and determined that the mechanical stiffness ks of the molecules adsorbed on the Au(111) surface is about 0.2 N/m.
Nuclear Science, Plasmas, and Electric Discharges
8278
Tetsuya Shimoyama, Eido Shibamura, Tsuguo Nishikawa and Mitsuhiro Miyajima
In order to identify a daughter ion of Ba+ with laser-induced fluorescence just after a candidate event of double beta decay of 136Xe is found in a liquid xenon ionization drift chamber, the size of the laser beam directed to the vertex position is determined to be 1.6 mm in diameter by simulating the signal amplitudes of scintillation and ionization due to two beta rays with the EGS4 simulation code. The detection efficiencies of the apparatus and the trigger threshold of scintillation signals, which are obtained with the simulation, are also discussed in detail.
8285
Majid Masnavi, Mitsuo Nakajima, Akira Sasaki, Eiki Hotta and Kazuhiko Horioka
The characteristics of the plasma conversion efficiency of 4d75p→4d8 transitions of xenon have been theoretically investigated using the quasi-steady-state collisional radiative model. Results indicate that a large conversion efficiency is achieved for electron densities between 1018 and 1019 cm-3. In relation to rapid plasma heating in capillary discharges, the influence of nonequilibrium ionization on the conversion efficiency has been investigated. The study clearly shows that the ionization state distribution is a function of the history of plasma evolution, and a rapid heating process under appropriate plasma conditions allows us to obtain higher conversion efficiency values compared with the steady-state calculations.
8292
Yasuyuki Yagi, Haruhisa Koguchi, Hajime Sakakita, Yoichi Hirano and Toshio Shimada
We report that the thermal wall load due to locked modes can be decreased by means of gas puffing in the reversed-field pinch (RFP) device, TPE-RX [Y. Yagi et al.: Fus. Eng. Design 45 (1999) 409]. Under standard RFP operation in TPE-RX, the phase- and wall-locked modes, i.e., spatially localized and stationary magnetic perturbations, respectively, appear after the current-rising phase. The locked modes cause a severe thermal wall load of the order of 100 MW/m2 on the first wall. It is estimated that the temperature on the plasma-facing wall at the mode-locking position sometimes exceeds the melting temperature of stainless steel. The thermal wall load increases with plasma current, Ip. We have succeeded in mitigating the high thermal wall load by gas puffing during discharges. It is shown that the maximum temperature increment of the vacuum vessel wall is halved when a gas-puffing rate of 200 Pa m3/s is applied at Ip=350 kA. The characteristics of the temperature increase of the vacuum vessel are presented, and they indicate the nature of the locked mode in RFP discharges in TPE-RX.
8300
Kyung Hwan Jang, Won Jeong Lee, Hyung Rae Kim and Geun Young Yeom
In this study, using inductively coupled Cl2/Ar plasmas (ICP), the effects of various process conditions such as source power, bias power, Cl2/Ar gas ratio, and ultraviolet (UV) ray were investigated to obtain high Cu etch rates without remaining any nonvolatile etch products. Due to the formation of nonvolatile copper chloride, copper film was not etched and, instead, a thick copper chloride residue was formed on the copper surface when Cl2/Ar ICP plasma was used. However, the residue could be removed and the copper film could be etched when high ICP source power was used with high intensity of ultraviolet rays and high bias power. Especially, when the ICP source power higher than 300 W and the bias power higher than 70 W were used, no UV irradiation was required to etch copper films, and which is more desirable for the etching of thin film transistor liquid crystal display (TFT-LCD) substrates. The maximum copper etch rate obtained was 300 nm/min with inductive power of 600 W, bias power of 75 W, and 0.5 of Cl2/Ar gas ratio at the pressure of 7 mTorr without applying UV.
8304
Zhibiao Hao, Shuo Han, Fan Ren, Bing Xiong, Changzheng Sun and Yi Luo
The inductively coupled plasma (ICP) etching of a GaAs/GaInP/AlGaInP heterostructure is studied using various gas mixtures. It is found that gas chemistry has a significant influence on the etch profile of the heterostructure, which is crucial for the fabrication of ridge/rib structures on AlGaInP-based materials. The self-masking effects of both aluminum oxide and InClx cause rough etched surfaces and sidewalls. Using Cl2/BCl3/CH4 chemistry, the smooth and vertical etching of a GaAs/GaInP/AlGaInP heterostructure has been realized at a moderate etch rate.
8308
Hiroyuki Yoshiki and Yasuhiro Horiike
The load impedance of the on-chip atmospheric capacitively coupled microplasma (ACCµP) source was evaluated from circuit elements in a correctly tuned matching network by the network theory. The resultant load impedance showed a high resistance of more than 500 Ω and a capacitive reactance with an equivalent capacitance of about 2 pF, implying that the L-type matching network is preferable to the on-chip ACCµP source. In the matching network, the coil inductance was determined by the excitation frequency, and the capacitance was very sensitive to the load impedance and the stray capacitance.
8310
Wei Xu, Ming-Wei Li, Gen-Hui Xu and Yi-Ling Tian
An experimental investigation has been conducted to decompose pure CO2 into CO and O2 via dc corona discharge at atmospheric pressure. In the tested range, the most conversion of CO2 is 10.91% and the energy efficiency is approximately 1.56–6.73%.
Atoms, Molecules, and Chemical Physics
8312
Akihiko Fujii, Hiroyoshi Mizukami, Tokiyoshi Umeda, Tomoki Shirakawa, Yuuki Hashimoto and Katsumi Yoshino
Optical and photovoltaic properties of a photovoltaic cell with a structure of indium-tin-oxide(ITO)/C60/poly(2-methoxy-5-(2'-ethylhexyloxy)-1,4-phenylenevinylene) (MEH-PPV)/Au have been investigated. The C60/MEH-PPV heterojunction of this cell was fabricated by spin-coating the solution of MEH-PPV onto the C60 thin film formed on ITO. Fabricated by this method, the photovoltaic cell has demonstrated a high efficiency, depending on the organic solvents used in the spin-coating process. From the optical properties and the direct observation of the C60/MEH-PPV interface, it has been clarified that such a high photovoltaic performance originates from the interpenetrating interface of MEH-PPV and C60, the morphology of which depends on the solvent used in the fabrication process of the MEH-PPV film.
8316
Gyeong Bok Jung, Toshiki Mutai, Ryushi Fujimura, Satoshi Ashihara, Tsutomu Shimura, Koji Araki and Kazuo Kuroda
We studied the temperature dependencies of electric field-induced birefringence, conductivity, and photorefractive properties of two poly(n-vinylcarbazole) (PVK)-based composites with different nonlinear optical chromophores: 2,5-dimethyl-4-(4'-nitrophenylazo)anisole (DMNPAA) and its derivative, 4-butoxy-3-propyl-1-(4'-nitrophenyazo) benzene (m3p4). Both composites showed similar temperature dependencies; the photorefractive responses were faster at higher temperatures and the diffraction efficiencies increased with temperature and then began to decrease. Such behavior is consistent with the measured temperature dependencies of the field-induced birefringence and the conductivity. We found that the m3p4 composite exhibited a faster photorefractive response rate than the DMNPAA composite when compared at the same reduced temperature T-Tg, where T is the temperature of the measurement and Tg is the glass transition temperature. The results indicate that one can enhance the total photorefractive response rate by changing the nonlinear optical chromophore.
Instrumentation, Measurement, and Fabrication Technology
8322
Shinji Yamashita, Teruyuki Baba and Ken Kashiwagi
We propose and demonstrate a novel active multiwavelength fiber Bragg grating (FBG) laser sensor system. Simultaneous multiwavelength oscillation by the frequency-shifted feedback technique markedly enhances output power and signal-to-noise ratio (SNR) compared with a conventional passive FBG sensor system. The principle of the simultaneous measurement of the dynamic strain signal from wavelength multiplexed FBGs is experimentally confirmed by employing an arrayed waveguide grating (AWG) as a wavelength demodulator.
8325
Fumiaki Takahashi, Yukihiro Hamada, Takatoshi Mori and Ichiro Hatta
To consider the thermal characteristics of a Bi2Te3/Sb2Te3 superlattice, we analyzed the phase lag of ac temperature waves between front and rear surfaces when an ac heat flux was applied to the front surface. In this analysis, we examined two typical models. In one model, a sharp interface exists between the adjacent layers in the superlattice where two crystals abruptly face each other at the interface. In the other model, an interfacial region composed of a disordered structure exists between the adjacent layers where the lattice constant and/or the chemical composition of the two adjacent layers are relaxed in the interfacial region. We found that both models could explain the experimental results almost equivalently. So far it has been known that in many cases the interface between adjacent crystals has a relaxed region where a disordered structure exist, and its width is comparable to phonon mean free path at room temperature. To understand the characteristics of the thermophysical properties at an interface, it is important to know the structure of the interfacial region.
8331
Taichiro Fukumori, Koji Futagami and Kiyohiro Kuroki
Reflection X-ray topography using double-crystal arrangement has been applied to the observation and the evaluation of lattice distortions in the near-(100)-oriented surface of silicon crystals. A series of X-ray topographs were taken with extremely asymmetric 311 reflection of CuKα1 radiation at a glancing angle of approximately 0.20° near the critical angle of total reflection. The experimental condition is realized using wafers with the (100) lattice plane inclined to the surface at an angle Δα (approximately 2.6°), which usually are used as substrates for epitaxially grown films. Both the local lattice misorientation induced at the boundary of oxide-deposited films of 4–5 nm thickness on a silicon wafer and the surface distortions arising through the processes of mechanochemical polishing and epitaxial growth are clearly visualized as a result of low background noise from the bulk layer due to a shallow X-ray penetration depth of several tens nanometers. From a series of topographs taken by changing the glancing angle from a low angle to a high angle of the Bragg peak at intervals of 5 arcsec, the average tilt angle of lattice planes localized at the oxide film edges was estimated.
8335
Takashi Ohshima and Makoto Kudo
The brightness of negative-electron-affinity photocathodes was measured to determine their applicability as an electron beam source for high-resolution, low-voltage scanning electron microscopes. The smallest electron source size was 3 µm, and the largest angular current density was 41 mA/sr at an extraction voltage of 2 kV. The resulting reduced brightness of 5.9×106 A·sr-1·cm-2·V-1 was about one order of magnitude higher than that of an LaB6 emitter and about one order lower than that of a Zr-O/W Schottky emitter. The limited brightness was due to the space charge effect at the GaAs surface. The brightness was improved using a GaAs film with a thinner depletion layer and using a pulsed excitation beam to prevent charge accumulation on the surface. The use of these techniques will enable the use of an NEA electron gun in small electron probes, which is expected to be a powerful tool for low-voltage scanning electron microscopy.
8341
Hironobu Sato, Yoshitaka Houshi, Toshiharu Otsuka and Shuichi Shoji
This paper presents the fabrication methods of polymer and metal three-dimensional (3-D) micromesh structures. The photoresist micromesh structures were formed using multiple inclined UV exposure from the backside of the SU-8 coated glass substrate with metal patterns. A few micrometer-size pillars and pores were realized by optimizing the UV lithography conditions. The 3-D macroporous-mesh Ni structures were also fabricated using electroplating. The inverse-mesh photoresist structures, fabricated by multiple inclined backside exposure, were used as molds for Ni electroplating. Ni meshes of about 3 µm in diameter were obtained by this method.
8345
Takahisa Mitsui
We artificiality generate a ripplon on the surface of a liquid droplet adhered to the tip of an optical fiber by using laser light, and detect it by Michelson interferometry. By this method, we observe the vibrations of the liquid droplet surface as a function of time and separately detect the real and imaginary parts of the susceptibility.
8347
Myung-Soon Lee, Seong-Cheol Hong and Don Kim
We studied a simple and easy way to fabricate gold electrodes on various insulating substrates through the direct photochemical lithography of self-assembled monolayers (SAMs) of alkanethiols (hexadecanethiol: HDT, octadecanethiol: ODT, pentaerythritol-tetrakis(3-mercaptopropionate): PTMP) on gold films. A SAM of PTMP was used as a resistor for positive (+) lithography. An array of gold islands, 33×33 µm, could be extended up to the centimeter scale.
Applied Bioscience
8349
Sehyun Shin, Yunhee Ku, Myung-Su Park and Jang-Soo Suh
The suspension of hardened erythrocytes (red blood cells) differs from the suspension of normal erythrocytes with respect to their rheological behavior. The deformability of normal and hardened erythrocytes (obtained by heating blood at 49°C or by incubating erythrocytes in a solution of hydrogen peroxide) was measured with a slit diffractometer and erythrocyte suspension viscosity was measured with a rotational viscometer. We found that when erythrocytes were heated at 49°C, with much less deformability than normal erythrocytes, their suspension viscosity was almost the same as the normal blood viscosity at high shear rates, whereas when erythrocytes were incubated in a solution of hydrogen peroxide, with an intermediate decrease of deformability, their suspension viscosity was greatly increased. The present study concluded that not all rigid cells cause an increase in blood viscosity at high shear rates, and therefore that decreased cell deformability is not always a good indicator of high-shear blood viscosity.
Nano Science and Technologies
8354
Chien-Chon Chen, Chin-Guo Kuo, Jung-Hsuan Chen and Chuen-Guang Chao
A novel and convenient method of thermo-immersion was used to fabricate of a large of lead–bismuth (Pb–Bi) eutectic nanoparticles on an uniform size anodic aluminum oxide (AAO) template. The AAO with a 60 nm pore diameter was used as a template for nanoparticles nucleation and growth. Both Pb–Bi and AAO were placed in a glass tube in vacuum at 10-6 Torr, controlled by a molecular turbo pump then heated up to the melting point of the alloy. Nanoparticles nucleated and grew on the AAO template when the molten Pb–Bi solidified. The nanoparticles nucleated on the pore corners of the AAO template initially, grew simultaneously and assembled into a big nanoparticle on top of the AAO template later. Additionally, the size diameters of the initial nanoparticles were approximately 15 nm on the pore corners of the AAO template, and 80 nm when the initial nanoparticles formed a large nanoparticle. Latent heat was released when the molten Pb–Bi solidified to form nanoparticles. The size diameters distribution of nanoparticles was observed by scanning electron microscopy (SEM) such that, the diameters of the nanoparticles were 80 nm on top of the AAO template, and 200 nm in depths of the concave.
8360
Chia-Chieh Chang and Chen-Shiung Chang
Porous silicon (PS) technology is, for the first time, used to grow ZnO nanowires on the surface of a PS substrate with a rough morphology without any catalyst. The characteristics of these nanowires were investigated by field-emission scanning electron microscopy (FESEM), transmission electron microscopy (TEM), energy dispersive X-ray (EDX), grazing-angle X-ray diffraction (XRD) and photoluminescence (PL) measurements. Zn vapor condenses easily on the PS surface and forms a wetting layer, but not on the flat Si surface. The PS surface provides a rough surface morphology to form a wetting layer by decreasing the surface energy so that ZnO nanowires can grow without any catalyst. The Zn-rich composition changed during growth; the ratio of zinc to oxygen was near to one at the top part of the nanowires. The probable growth mechanism was the vapor-solid (VS) process. In principle, the selective growth of ZnO nanowires on Si-base devices for optoelectronic application is possible.
8365
Masato Ohkohchi, Xinluo Zhao, Sakae Inoue and Yoshinori Ando
Single-walled carbon nanotubes (SWNTs) were synthesized in a mixed gas of H2 and Ar by AC arc discharge. The catalyst used in the synthesis was a single Fe catalyst. SWNTs containing a negligible number of nanoparticles were obtained. A novel purification method, in which synthesized soot is heated and oxidized in situ, was adopted. As a result, high-quality SWNTs with excellent crystallinity containing very few coexisting nanoparticles and fine particles from the catalyst were obtained.
8369
Heon Lee and Gun-Young Jung
Due to its ability to produce nanosized patterns on large area substrates with high throughput, UV curing nanoimprinting lithography is a promising technology for nanosized pattern fabrication. However, for it to become a technology for the mass production of nanosized patterns, we must first show that UV curing imprint lithography can produce a uniform layer with minimal residue. In this study, a pressurized chamber imprinting system was used. Isotropic pressure was applied to insure effective delivery of pressing force and uniform pressing. A UV curable resin, consisting of a perfluorinated acrylate and a photo-initiator, was used. The factors determining residual layer thickness in UV curing imprint lithography were investigated. To obtain imprinted patterns with no residual layer, a thin and uniform initial resin layer and pressurized imprinting must be used to rearrange the resin effectively. The high fidelity transfer of various patterns as small as 150 nm with no residual layer was successfully demonstrated by controlling the initial resin thickness at an imprint pressure of 15 atm.
8374
Hitoshi Koizumi, Yuki Itoh and Tsuneki Ichikawa
Carbon nanotubes dispersed in cyclohexane migrated toward the cathode under electric field and inserted themselves into pores in a track etch membrane on the cathode. The efficiency of the insertion depends on the frequency of the applied electric field. Application of rectangular electric pulses with a frequency of 0.1 Hz gave a maximum efficiency of 22%. Prolonged application of the field did not increase the fraction of occupied pores. The fraction saturates at 30 min, since the inserted tubes attract subsequent tubes and prevent the insertion of the tubes into pores.
Cross-Disciplinary Areas
8376
Fujio Shimizu
Quantum computer that is composed of Mott insulator atoms trapped in an optical lattice and a control atom(s) that moves with an independent optical lattice is discussed. A three-dimensional array of Mott insulator atoms functions as memory qubits. The control atom that moves with the second lattice mediates one and two qubit operations among memory qubits with the help of resonant optical beams. Explicit examples of general quantum computational operations are described. In this system all memory qubits are equivalent, and the system is scalable. It is also not necessary to address individual memory qubits by optical means.