Table of contents

Cross-sectional transmission electron microscopy was used to study the intrinsic breakdown spots of 10-nm-thick gate oxides thermally grown on (001) Si substrates. At a breakdown spot with a breakdown field of 15 MV/cm, (111) twin planes are confirmed by electron diffraction to exist in the surface region of the Si substrate. High resolution observation also reveals Si lattice fringes in the gate oxide, penetrating from the Si substrate and gate electrode of polycrystalline Si, and significant roughening of the anode interface between oxides and Si substrates. Also, similar growth twins and anode interface roughening are observed for a lower field breakdown spot with a breakdown field of 7 MV/cm. The presence of growth twins indicates that a Si substrate is locally molten during dielectric breakdown. These results are well explained by the local thermal breakdown model of intrinsic oxides. The interface roughening, presumably caused by deoxidization at the SiO₂/Si interface in a local hot spot, is indicative of an initial stage of intrinsic dielectric breakdown of thin oxide.

The substitution of Czochralski (CZ)-silicon (Si) wafers into p^-(n^-)/p^-(n^-) ( p^- or n^- layer on p^- or n^- Si substrate: resistivity of approximately 100 m Ω m) thin-film epitaxial Si wafers used as starting materials has been investigated with respect to application to metal-oxide-semiconductor (MOS) large-scale-integrated circuits (LSIs). The optimum epitaxial layer ( p^-/p^- structure) thickness for MOS-LSIs was determined to be approximately 1 µ m from the viewpoints of gate oxide integrity (GOI) improvement and cost effectiveness. With increasing epitaxial layer thickness from 0.1 to 0.3 µ m, the oxide defect density was greatly reduced and leveled off at approximately 1/30 that of a CZ-Si layer if the layer thickness is above 0.3 µ m. This is because microdefects in CZ-Si represented by crystal originated particles (COP) which cause weak spots in the gate oxide layer are covered by an excellent Si epitaxial layer on the CZ-Si surface. The p^-/p^- thin epitaxial structure results in very controlled resistivity for electrically active regions in the device, resulting in a lower cost of growth.

Annealing properties of defects in F⁺- and B⁺-implanted Si were studied using monoenergetic positron beams. For F⁺-implanted specimen with a dose of 2×10¹³ F/cm², before annealing treatment, the mean size of the open volume of defects was estimated to be close to the size of divacancies. After rapid thermal annealing (RTA) at 700° C, vacancy-fluorine complexes and vacancy clusters were formed. The mean size of the open volume for the vacancy-fluorine complexes was estimated to be close to the size of monovacancies, and their annealing temperature was determined to be 800° C. For F⁺-implanted specimen with a dose of 4×10¹⁵ F/cm², complexes between vacancy clusters and fluorine atoms were introduced during solid-phase epitaxial growth of the amorphous region, and they were observed even after RTA at 1100° C. Effects of additional B⁺ implantation on annealing properties of defects are also discussed.

Spectroscopic ellipsometry has been used to characterize SIMOX (separation by implanted oxygen) wafers. High-dose SIMOX wafers with 450 nm thick buried SiO₂ were measured using a four-zone null spectroscopic ellipsometer in a 230–840 nm range. The measured spectra of Ψ and Δ were analyzed based on a multilayer structure of air/surface- SiO₂/top-Si/buried- SiO₂/Si in which the optical constants of the top-Si layer were analyzed using the optical constants of bulk-Si and also using model dielectric functions proposed for Si. In the fitting of Ψ and Δ spectra, a thickness fluctuation of buried- SiO₂ was considered with an arbitrary thickness distribution function. Measured Ψ and Δ spectra could be well fitted under the assumption of thickness distribution in the buried- SiO₂. To examine the thickness dependence of the optical properties of the top-Si layer, the thickness of this layer was systematically reduced by repeating oxidation and etching with HF. It was found that the optical constants of the top-Si layer were equal to those of Si for a thickness above 5 nm.

The effects of growth temperature on Er-related photoluminescence (PL) have been investigated in Er-doped InP (InP:Er) and GaAs (GaAs:Er) grown by organometallic vapor phase epitaxy (OMVPE) using tertiarybutylphosphine (TBP) and tertiarybutylarsine (TBAs). In InP:Er, the PL spectra exhibit strong dependence on the growth temperature, and the intensity increases drastically in specimens prepared at temperatures lower than 550° C. Similar dependence of PL intensity on the growth temperature is observed in GaAs:Er. The activation energy obtained is about 3 eV in both materials. The activation energy is discussed based on atomic configurations of Er-related luminescence centers.

Using dimethylhydrazine (DMHy) as a group V source, we grew hexagonal GaN layers on (111)B GaAs substrates by low-pressure metalorganic vapor phase epitaxy. The surface morphology of the hexagonal GaN layers and the carbon incorporation in them strongly depend on the V/III ratio and the reactor pressure. A flat GaN surface can be obtained at the V/III ratio of 60 and the substrate temperature of 850° C. The carbon concentration decreases with increasing reactor pressure and the minimum concentration is 2×10¹⁹ cm^-3 for hexagonal GaN grown at 300 Torr. Low-temperature photoluminescence measurements reveal that the band edge emission for the GaN grown at 300 Torr is dominant compared with that of a deep level.

We have theoretically studied the response of quantum well intersubband photodetectors to high-frequency modulated infrared radiation. The small-signal responsivity dependent on the modulation frequency of infrared radiation and device parameters has been derived using the proposed analytical model. It has been shown that with increasing modulation frequency the roll-off of the small-signal responsivity is associated with the electron transit and capture effects which limit the device bandwidth.

The initial stage of stress-induced migration in aluminum interconnections on semiconductor devices was studied through measurements of the electric resistivity and the residual-stress, as well as observations of grain growth during high temperature storage. It was shown that characteristics of the initial stage varied markedly depending on the storage temperature and period. The variation of the characteristics was affected not only by (1) increase in grain size or decrease in dislocation density upon recrystallization, but also by (2) degeneration at the grain boundary in the aluminum interconnection, which was considered to be a pile-up of dislocations. In this paper, we propose a model that the initial stage of the stress-induced migration is determined by competition between phenomena (1) and (2), and discuss the possibility that the open failure rate of the interconnection is correlated qualitatively with the change in the electrical resistivity at the initial stage.

This paper presents a simple, complete, and analytical drain current model for submicrometer silicon-on-insulator metal-oxide-semiconductor field-effect-transistor (SOI MOSFET). The model applicable for digital/analog circuit simulation contains the following advanced features: precise description of the subthreshold, near threshold, and above-threshold regions of operation by one single expression; precise description of I–V and G–V characteristics in the saturation region; single-piece drain current equation smoothly continuous from the linear region to saturation region; considering the source/drain resistance; inclusion of important short channel effects such as velocity saturation, drain induced barrier lowering and channel length modulation; self-heating effect due to the low thermal conductivity of the buried oxide; impact-ionization of MOS devices and the parasitic bipolar junction transistor (BJT) effect associated with drain breakdown. The model predicts that the parasitic resistances are important for submicron and deep submicron SOI MOS devices, the effects of impact-ionization and parasitic BJT are important in saturation region at small gate source voltage V _GF, and self-heating effect is important in saturation region at large V _GF. The present model agrees well with experimental results of various dimensions.

Room temperature surface-illuminated InAs_xP_1-x-ySb_y/InAs photodiodes with an external quantum efficiency as high as 50–86% in a 1.83–3.53 µm wavelength range have been fabricated for the first time. Lattice matched heterostructures with a wide energy gap InAsPSb cap layer were grown on the InAs substrate using the liquid phase epitaxy technique. According to temperature dependence measurements for a 1 mm diameter photodiode, peak responsivities of 1.83–2.5 A/W have been realized in a temperature range of 296 to 200 K. The Johnson noise limited room temperature detectivities D^* are deduced to be 1\endash6 ×10⁹ cm·Hz^1/2/W at zero bias. It is demonstrated that the only loss of external quantum efficiency is from the reflection of the entrance face.

An interesting AlGaAs/InGaAs/GaAs step-doped channel negative-differential-resistance transistor (SDCNDRT) has been fabricated. The SDCNDRT is based on a previously reported step-doped-channel field-effect transistor (SDCFET). The N-shaped negative-differential resistance (NDR) are three-terminal-controlled NDR phenomena. We believe that the NDR phenomena can be attributed to the real space transfer (RST). A maximum drain current and peak-to-valley current ratio (PVCR) of 59 mA and 6.6 are obtained, respectively. The high drain current and PVCR indicate the potential of the SDCFET for practical circuit applications.

We propose cellular automaton circuits that use single-electron-tunneling circuits (SET-CA). The unit cell consists of four intrinsic semiconductor islands and four single-electron-tunneling junctions. The dielectric constant of the intrinsic semiconductor is much larger than that of the junction insulator. The unit cell is charged with two single electrons. Polarization states of the two single electrons in the unit cell can be used to encode a binary signal. We designed various binary logic SET-CA circuits, and analyzed their operation by computer simulation. It was demonstrated that the SET-CA circuits having appropriate arrangements of the cells can perform correct signal transmission and elemental logic operations such as NOT, NAND, and NOR.

The impacts of the two major high temperature treatments on the characteristics of thin film transistors (TFT's) in the back-end process of ultra large scale integrated circuit (ULSI) technology have been investigated. TFT's without any high temperature treatment show poor characteristics. High temperature furnace annealing and rapid thermal annealing (RTA) which are performed for boron phosphorous tetra ethyl ortho silicate (BPTEOS) planarization can improve the characteristics of low temperature recrystallized TFT. Then, the technologies used for contact annealing result significant difference on the characteristics of TFT's. High temperature furnace contact annealing can still greatly improve the characteristics of TFT's. However, after furnace annealing for BPTEOS planarization, RTA contact annealing deteriorates the characteristics of TFT. High temperature furnace annealing for BPTEOS planarization improves the charge to breakdown (Q _bd) value of the gate oxide of TFT and so does the furnace contact annealing. However, RTA contact annealing performed after the furnace planarization flow deteriorates that of the gate oxide.

This paper reports the necessity of dedicating a reactive ion etching system for SiO₂ according to etching materials. Silicon wafers etched with TiN-, Ti- or Al-coated wafers were analyzed using total reflection X-ray fluorescence and secondary ion mass spectroscopy. The amount of contamination varied with distance from the metal coated wafers. Si etched after removing the metal-coated wafers were then analyzed. Contamination was also detected, though it decreased with increasing discharge time. The contamination is believed to be due to the collisions of reaction products with reactive gas molecules and redeposition of those products onto the cathode. The contamination is decreased to detection limit by dipping into a diluted HF solution. However, the carrier lifetime of the etched wafer is severely affected by contamination less than detection limit.

Carbon (C) heavily doped AlAs has been grown by metalorganic chemical vapor deposition (MOCVD) using trimethylaluminum (TMAl) and tertiarybutylarsine (TBA) without any additional dopant sources. The hole concentration was controlled by changing only the V/III ratio. The highest hole concentration was 2.5×10¹⁹ cm^-3. The decrease in the lattice constant of C-doped AlAs shows that the acceptor activation ratio is close to unity.

This paper describes the fabrication of 0.2 µ m hole resist patterns in KrF excimer laser lithography. By using a SiN_x single-layer halftone phase-shifting mask (PSM) and an in-house chemically amplified positive resist, 0.2 µ m hole resist patterns can be obtained with sufficient depth of focus (DOF). Furthermore, a 0.15 µ m hole resist pattern can also be fabricated by using the PSM.

We prepared in-plane-aligned La_2-xSr_xCuO₄(100) films using LaSrGaO₄(100) substrates by KrF laser ablation. The φ scan of X-ray diffraction and the large anisotropy of resistivity indicate a high degree of in-plane epitaxy. The T _c of the in-plane-aligned La_2-xSr_xCuO₄ (x=0.15) was 21.6 K. The critical temperatures measured along the c-axis were usually higher than those measured along the a-axis. We also observed voltage peaks at just above T _c, corresponding to the abrupt decrease of the anisotropy of resistivity at T _c.

Epitaxial growth of CeO₂ films on Al₂O₃ (102) substrates and the effect of CeO₂ buffer layers on the growth orientation and superconducting properties of EuBa₂Cu₃O_7-δ (EBCO) were investigated. CeO₂ and EBCO films were prepared by rf and dc magnetron sputtering, respectively, and were characterized by X-ray diffraction (θ-2θ scan and ϕ scan), atomic force microscopy (AFM) and high resolution scanning electron microscopy. Epitaxial (001) CeO₂ films were obtained at an off-center distance (D _on-off) of 3.5 cm and a substrate temperature of 660°C. The structural and superconducting properties of EBCO films deposited at 650°C depended on the thickness of the CeO₂ buffer layer. The EBCO films deposited on CeO₂ 50-400-Å-thick buffer layers had T _ce's of 90 K or above. The high-T _c EBCO films had in-plane epitaxial orientation relationships of Al₂O₃ [110]\varparallelCeO₂ [100]\varparallelEBCO[110]. The EBCO films on the thin CeO₂ buffer layers had rectangular grains similar to those on MgO(001) substrates. The critical current density of the EBCO films with T _ce=90 K was about 6 ×10⁵ A/cm² in zero field at 77.3 K. The T _ce varied largely and decreased with increasing CeO₂ buffer layer thickness above 500 Å. AFM observation of a 1000-Å-thick CeO₂ film showed growth of bamboo-like crystal grains 1700 Å long and 300 Å wide along the direction of CeO₂ [110]. The ravine depths were about 100 Å. The EBCO films on the thick CeO₂ buffer layer (>500 Å) exhibited poor superconducting behavior and gave (103) or (110) diffraction peaks.

CoNb/Cu multilayers with amorphous CoNb magnetic layers were prepared by rf-sputtering method and then annealed at various temperatures in a high vacuum. A maximum magnetoresistance ratio of 2.83% at room temperature was obtained under a relatively low saturation field of about 200 Oe for the optimum annealing temperature of 400° C. The structural, magnetic and transport properties were strongly influenced by the annealing temperature, and the transport properties were explained on the basis of granular multilayered structures. The magnetoresistance ratio also depends on the thickness of CoNb and Cu layers.

LiFeWO₄Cl has been prepared by reductive intercalation of lithium into FeWO₄Cl. FeWO₄Cl crystallizes into a tetragonal system (space group P4/nmm) with cell dimensions of a=6.677(5) Å and c=5.270(5) Å, while lithium intercalation gives rise to the formation of a monoclinic phase ( P2₁/m) with cell dimensions of a=7.050(0) Å, b=6.926(2) Å, c=5.043(0) Å and β=92.54(2)°. Infrared (IR) spectroscopic analyses of FeWO₄Cl and LiFeWO₄Cl show that the symmetry of the tetrahedral WO₄^2- group in FeWO₄Cl becomes reduced as the lithium intercalate into its two-dimensional lattice, indicating an alternately ordered lithium occupancy of half of the octahedral interlayer sites. The effective magnetic moments of FeWO₄Cl and LiFeWO₄Cl were estimated to be 5.95 and 5.10 B.M., respectively, which can be attributed to the selective reduction of a high-spin ferric ion to a high-spin ferrous one upon lithium intercalation. It was also found that the two-dimensional magnetic property of FeWO₄Cl was changed to a three-dimensional one in LiFeWO₄Cl owing to the c-axis contraction.

A new mutually pumped phase conjugator with a photorefractive copper-doped potassium sodium strontium barium niobate crystal is demonstrated, in which the beams travel along a geometric path similar to the outline of a sickle. The maximum reflectivity is as high as 168%. Phase-conjugate images are obtained. The two-interaction region theory, based on the coupled plane wave amplitude, is used to account for the main characteristics of the new phase conjugator. The threshold coupling strength which depends on the reflection loss and the incident beam ratio, is also investigated and calculated.

Uniform fabrication of highly reliable 50–60 mW-class 685 nm laser diodes (LDs) with a window-mirror structure has been realized by using selective solid-phase Zn diffusion and three-inch full wafer processing. A window-mirror structure at the LD mirror is formed by Zn-induced disordering of an ordered GaInP multiple quantum-well (MQW) active layer. High uniformity of characteristics such as the operating current and the far-field pattern has been obtained by realization of highly uniform Zn diffusion. A small astigmatic distance (ΔZ≦3 µ m), a low relative intensity noise (RIN≦-135 dB/Hz) and a high speed response (T _r, T _f≦1.2 ns) are obtained in addition to the high-power and high-temperature characteristics (70 mW, 80° C) in spite of the existence of the window structure. The LDs have exhibited reliable 6,000–10,000 h operation under the conditions of 60° C and 50–60 mW for the first time.

We have achieved the growth of a GaInNAs lattice matched to GaAs by metalorganic chemical vapor deposition using dimethylhydrazine (DMHy) as a nitrogen source for the first time. We demonstrate the room-temperature operation of GaInNAs/GaInP double-heterostructure laser diodes. Lasing operation can be achieved at a wavelength of 1.26 µ m under pulsed operation. The threshold current characteristic temperature of 1.17 µ m laser diodes is found to be 96 K (at ambient temperatures between 10 and 50° C) and 69 K (at ambient temperatures between 50 and 70° C). Light emission for light-emitting diodes (LEDs) grown on a GaAs substrate is also demonstrated at a wavelength range from 1.2 to 1.45 µ m. A wavelength of 1.45 µ m is the longest reported to date for a GaInNAs lattice matched to GaAs. These results indicate the potential of GaInNAs for application to laser diodes without thermal cooling because they are more stable at ambient temperatures than conventional GaInPAs laser diodes at wavelengths from 1.3 to 1.55 µ m.

A new approach for reducing a beam divergence angle vertical to the junction plane was proposed. Low-refractive-index layers were introduced adjacent to the active region in a real-refractive-index-guided high-power laser diode. The effects of the low-refractive-index layers on vertical near-field patterns were studied in detail through numerical calculations. The calculations indicated that the beam divergence angle vertical to the junction plane can be reduced with respect to the low-refractive-index layers without significant increase in the threshold current. The fabricated device exhibited a vertical beam divergence angle as small as 15.3° and threshold current as low as 22.7 mA. The actual vertical near-field pattern was found to be wider than calculated, which is attributed to the asymmetrical refractive-index profile in the vertical direction outside the stripe.

A new axial Zeeman laser whose beat frequency varies from 0.7 MHz to 3.4 MHz was developed. The practical apparatus for realizing the beat-frequency tuning and the stabilization of the absolute frequency of the Zeeman laser is described. The capability of the Zeeman laser we have developed is verified by measurement of the transient response, long-term stability of the beat frequency, and relationship between the applied current for generating the magnetic field and the resultant beat frequency. The applications of this new Zeeman laser are discussed.

Resistance to the optical feedback of complex-coupled DFB lasers with current blocking grating was studied experimentally. The measured relative intensity noise of the lasers was as low as -160 dB/Hz even in the presence of an external feedback level of -15 dB. The modulation chirping of the device was quite small with a linewidth enhancement factor α of 1.9 measured using a fiber-response peaks method. A transmission experiment was performed over a 235 Km standard nondispersion-shifted fiber at 2.488 Gbps by direct modulation without using an optical isolator, and the power penalty was only 1.55 dB at a bit error rate of 10^-12.

The absorption of incident laser beam is important when the quality of the laser machining is concerned, since the surface plasma acts as a medium between the workpiece and a laser beam. In the present study, the measurement of electron number density and temperature is carried out using a Langmuir probe. Effective absorption coefficients due to electron-ion and electron-neutral collisions as well as photoionization processes are dealt with analytically using the Maxwell's equations. The validity of classical treatment for the electron-neutral collision in relation to absorption is introduced and discussed. Consequently, the absorption coefficients due to these processes are computed within the frame of the data obtained from the present work. The study is extended to include the transmittance of He–Ne laser beam through the surface plasma generated during laser beam-workpiece interaction. It is found that the electron-ion collision is the dominant mechanism in the absorption process. The present data indicates that about 85% of the reference He–Ne is absorbed at the location close to the irradiated spot center in the plasma, therefore, the attenuation, scattering and deflection of the reference beam is mainly due to vapor atoms than the charged particles.

The generation of transform-limited 30-fs pulses from a laser-diode pumped, Kerr-lens mode-locked Cr:LiSAF laser is reported. Good mode-locking was achieved by use of a simple cavity configuration with a laser diode pumping source which had a separate section of a laser diode amplifier. The 30-fs optical pulse width was confirmed by use of an autocorrelator having little internal dispersion due to its composition of only reflecting mirror optics. A time-bandwidth product of 0.318 was obtained for the laser output pulse by assumption of a sech² waveform.

We show theoretically that mode splitting of a vertical coupled-cavity is very effective for realizing high contrast and lossless operation in asymmetric Fabry-Perot all-optical ultrafast time-division switches having a multiple quantum well saturable absorber. The merit of the coupled-cavity is mode splitting, which makes pump and signal pulse frequencies independently adjustable without lengthening of cavity; the short cavity feature is indispensable for fast response with high sensitivity. With 7.7 fJ/µ m², a 0.2 dB internal loss is achieved for a coupled-cavity, while a conventional single cavity has a loss of 17 dB. The maximum extinction ratio for a coupled-cavity reaches 32 dB, while that for a single cavity saturates at around 22 dB.

Amplification of high signal gain (36 dB, approximately 4000 times) and high output power (1200 W) was achieved in a Rhodamine B (RB)-doped graded-index (GI) polymer optical fiber amplifier (POFA) based on a prediction in which absorption and emission cross section data were used. The absorption and emission cross section spectra of RB, Rhodamine 6G, and 4-Dicyanmethylene-2-methyl-6-(p-dimethylaminostyryl)-4H-pyran (DCM) in poly(methyl methacrylate) (PMMA) bulk were determined. The cross section data enable us to predict gain characteristics of RB, R6G, and DCM-doped GI POFA. The wide spectral coverage in the visible is achievable with the GI POFAs doped with several selected dyes.

A new design of a (Pb,La)(Zr,Ti)O₃ (PLZT) polarization-plane rotator is proposed and tested. The electrodes, parallel to the light propagation di rection, are buried from the front to the rear surface of the rotator. Owing to the uniform distribution of the electric field inside the entire volume of the rotator, the half-wavelength voltage can be reduced to less than 100 V in the mid-infrared region. This is much lower than the voltage for a conv entional rotator with plane electrodes, of the order of 600–800 V. By combin ing the rotator with two polarizers, a mid-infrared electro-optical shutter, which is applicable to an infrared absorption-type gas monitor, is fabricated .

Exchange-coupled double-layered (ECDL) films were adopted to increase density of magnetically induced super-resolution (MSR) media by using center aperture detection scheme, and to improve sensitivity of switching field simultaneously. A carrier to noise ratio of 40 dB was obtained at 0.5 µ m recording marks produced by a switching field of 75 Oe. Temperature dependence of interface wall energy affects recording/playback characteristics of ECDL MSR disks considerably.

Jones matrix method was used to optimize the twisted nematic liquid crystal cell parameters for reflective direct-view and projection displays. Twist angles ranging from 0 to 90° were investigated. From these analyses, two operation modes are found to have particularly important applications. The mixed-mode 90° twisted nematic cell exhibits an excellent contrast ratio for normally white direct-view and projection displays whereas the improved 45° twisted nematic cell shows an impressive light efficiency for normally black projection display.

Evidence from deep level transient spectroscopy (DLTS) and Fourier transform infrared spectroscopy (FTIR) measurements strongly supported the assertion that the degradation mechanism of ZnS:Mn AC thin-film electroluminescent (ACTFEL) devices is mainly due to the deep electron trap, E _t, which comes from the Mn activators reacting with water molecules. The photoluminescence measurements revealed that the Mn-related E _t trap behaves like a nonradiative center. Furthermore, the X-ray diffraction experiments indicate that the crystallinity of ZnS:Mn phosphor films were destroyed in some degree by the E _t trap. As a result, poor brightness characteristics including lower brightness and higher threshold voltage were obtained when samples become aged.

Synergism in photocurrent observed with vacuum-sublimed films of zinc tetraphenylporphyrin (ZnTPP) is described in detail. The photocurrent synergism appears when a ZnTPP thin film sandwiched between two dissimilar metals is illuminated simultaneously with visible (VIS) and near-infrared (NIR) light. The magnitude of the synergism (η) at a fixed VIS light depends on the wavelength of the NIR light, which is not appreciably absorbed by the ZnTPP film, whereas the response of η to varying wavelengths of VIS light at a fixed NIR light exhibits a peak at 430 nm, where the ZnTPP film absorbs light most intensively. In addition to these η spectrum measurements, dependences of η on the intensities of the 430 nm and NIR light, film thickness, and bias voltage are examined for clarification of the mechanism of the synergistic effect.

Hole drift mobilities were measured in hydrazone-doped polycarbonate films under various concentrations, temperatures and electric fields. Hydrazone molecules used were 9-ethylcarbazole-3-carbaldehyde diphenylhydrazone (CTM1), 9-ethylcarbazole-3-carbaldehyde methylphenylhydrazone (CTM2), and 1-pyrenecarbaldehyde diphenylhydrazone (CTM3). Charge transport in the films obeyed the nonadiabatic small-polaron hopping formalism with almost the same activation energies at concentrations investigated. CTM2 exhibited a different concentration dependency and relatively small mobility especially at high concentrations. Analysis by disorder formalism indicated that the energetic disorder parameter (σ) is almost the same and the non-energetic disorder parameter (Σ) is different. We found some correlations between the mobilities of the CTMs and molecular overlap and orientation in the single crystalline state of CTMs.

The fatigue effect and the time dependence of the current transient during polarization reversal in 8/65/35 lead lanthanum zirconate titanate (PLZT) ceramics were studied as a function of the electric field. The comparison to the Avrami expression has revealed that the nucleation process in a thick PLZT film is constant during polarization reversal and the dimensionality of the domain growth is D ∼0.5, indicating the existence of needlelike domains. Also, it was found that the dimensionality of the domain growth in bulk PLZT ceramics is D ∼1.5 and that the nucleation process is latent during polarization reversal.

Dry etching of PbZr_XTi_1-XO₃ (PZT) thin film capacitors with RuO_X/Pt multilayered electrodes was studied to examine the etching effects. PZT films were deposited on RuO_X/Pt/Ti/SiO₂/Si substrates by sol–gel process and Pt films were prepared by DC magnetron sputtering. PZT and Pt thin films were etched with Cl₂/C₂F₆/Ar gas combination in an inductively coupled plasma (ICP) by varying the etching parameters such as coil RF power, dc-bias voltage to wafer susceptor, and gas pressure. Etching effects were investigated in terms of etch rate, etch selectivity, etch profiles, and electrical properties of etched capacitors. Quantitative analysis of the etching damage was obtained by calculating the shift of the coercive field and the switchable polarization in hysteresis loops. Finally, the etching damage mechanism was discussed and the optimization of etching processes for the fabrication of PZT capacitors was attempted to minimize the etching damage to ferroelectric capacitors.

Using a mode-locking Nd:YAG laser, we measured the second harmonic generation (SHG) of polymer-stabilized cholesteric texture (PSCT) films. The SHG observed in these samples is due to the dielectric gradient which arises from the interfaces between the cholesteric liquid crystal/polymer and those between the adjacent cholesteric focal conic domains having a different helical axis. Both the SHG intensities and the electro-optical characteristics of PSCT films were measured as a function of the polymer concentration. In addition, polymer network structures formed in PSCT films were investigated. Good correlations among the SHG measurements, electro-optical characteristics, and the polymer network morphologies were found. This means that SHG measurements could provide us with another qualitative, nondestructive method to analyze the PSCT films.

A novel method of determining the cell thickness and the twist angle of twisted nematic liquid crystal (TNLC) cells by measurement of the Stokes parameters of light transmitted through the cell is presented. The relationships of the Stokes parameters of transmission light to the cell thickness and the twist angle are reduced by use of Jones matrix representation. The cell thicknesses and twist angles of some TNLC cells are determined practically by this simple method. The experimental results indicate that this method is valid. Furthermore, some properties of this method are discussed and an accuracy analysis by theoretical calculation is undertaken.

Use of liquid crystal (LC) microlenses prepared from various types of LC materials is demonstrated for clarification of the influence of material parameters on the optical performance experimentally. Spherical aberration in the LC microlens is reduced by optimization of structure parameters, and the correlation between astigmatic aberration and elastic constants is investigated. The astigmatic aberration in the LC microlens strongly depends on the elastic constants and is explained by use of molecular orientation models calculated by a finite element method.

Reverse mode properties have been achieved in novel polymer dispersed liquid crystal (PDLC) cells using a UV curable liquid crystal (LC) as a polymer matrix. A homogeneously aligned PDLC cell is composed of a nematic LC with positive dielectric anisotropy and the UV curable LC. Another type of the PDLC cell where the molecular alignment is homeotropic is prepared using a LC with negative dielectric anisotropy. These PDLC cells have a very wide viewing angle both in a highly transparent off-state and in a light scattering on-state due to the refractive index anisotropy of the matrix. Fast response and recovery properties of a few ms can be obtained in the homeotropically aligned PDLC cell.

Maxwell displacement currents (MDCs) across mesogenic liquid crystal monolayers of 4-cyano-4'-n-alkyl-biphenyl (nCB) with positive anisotropy, a mixture of 4-alkoxyphenyl 4-alkyl-cyclohexanecarboxylates of k-17-N-73-I (DON103) with negative dielectric anisotropy, and LVI-035L without dielectric anisotropy on a water surface were measured by compression of the monolayer film in the range of immeasurably low surface pressures by means of a MDC-measuring technique. It is shown that the MDCs generated in the Langm\ddot uir films in the range just after the phase transition from the planar surface alignment isotropic phase to the polar orientational alignment phase mainly depend on the orientational change of the long-axis-orienting dipole moment of organic molecules. A novel theoretical explanation was introduced for study of this dependence. That is, a biaxial molecule model expressible with two order parameters S_∥ and S_⊥ was developed for the first time, and applied to the analysis of the aforementioned experimental results. Here S_∥ and S_⊥ are the orientational order parameter in the direction parallel and perpendicular to the molecular long-axis, respectively. It is found that the monolayer films of LVI-035L without dielectric anisotropy exhibit no MDC generation as the Langmüir films are compressed, and the extent of MDC generation in monolayers with positive dielectric anisotropy is much larger than that in monolayers with negative dielectric anisotropy. We concluded that the developed mathematical expressions based on the biaxial molecule model are useful for the analysis of the MDC generation across biaxial monolayers.

The velocity and attenuation of longitudinal and transverse ultrasonic waves in four different compositions of (V₂O₅)_x(GeO₂)_y(P₂O₅)_1-x-y glass system with P₂O₅ concentration ranging from 2.85 to 15.44 mol% have been measured at temperatures between 80 and 300 K using the ultrasonic pulse echo technique at 8 MHz. The results indicate that for low concentration of GeO₂ (y<0.5) the velocities decrease slowly and monotonically throughout the entire range of temperature but for higher concentration (y>0.5) velocity variation shows a minimum. The attenuation shows broad high temperature relaxation peak for all samples.

The optical propagation loss of rib waveguides fabricated on magnetic garnet films increased upon annealing in H₂ ambient during wafer direct bonding. The heat treatment in wafer direct bonding between InP and Gd₃Ga₅O₁₂ was investigated with the aim of circumventing the loss increase. The bonding was achieved by heat treatment in H₂ ambient at temperatures of ≦330° C or in N₂ ambient.

The problem of SiC precipitation during micro-pulling-down growth of Si–Ge mixed crystals has been analyzed as an important constraint of the technique. Three mechanisms of SiC microcrystal formation have been discussed. The solute transport in the melt and the dependence of the rate of Si–Ge melt evaporation on melt composition have been investigated. The troublesome influence of CO on SiC formation has been reduced by modification of the micro-pulling-down (µ-PD) growth assembly with optimization of argon flow. As a result the effective lifespan of the crucible was increased at least fivefold.

Nonideal Czochralski (CZ) silicon crystals were characterized systematically by double-crystal and triple-crystal diffractometry in the Bragg case for high energy synchrotron radiation. The crystals were either annealed at 780°C, 1000°C, and 1170°C, or lapped, with a mean abrasive particle sizes of 13 µm, 25 µm, and 42 µm. Triple-crystal diffractometry was performed at 60 keV of the BL14 vertical wiggler of the Photon Factory and double-crystal diffractometry by use of a Mo rotating-anode X-ray generator. Physical properties of the annealed and lapped crystals were studied quantitatively with respect to mosaic spread and lattice parameter fluctuation.

The author previously proposed diffusion equations which describe diffusion behavior of point defects in Si crystal growth ends during melt growth. The author would like to correct the mistakes made during the derivation of the diffusion equations, starting from the phenomenological diffusion equations.

Polycrystalline semiconducting β- FeSi₂ layers on Si (100) have been formed by ion beam synthesis. Results from two different annealing processes, either two-step (2SA) annealing up to 900° C or three-step annealing (3SA) up to 1100° C, are discussed. β- FeSi₂ grown by 3SA has shown a typical direct band-gap energy (E _g ^dir) of 0.88 eV and a high localized defect density (N₀) of 1.0×10¹⁸ cm^-3, the latter being due to crystallographic mismatches or relevant defects at grain boundaries introduced during the transformation process from β to α. On the contrary, β- FeSi₂ grown by 2 SA has shown a lower E _g ^dir of 0.80 eV and a smaller N₀ of 1.7×10¹⁷ cm^-3, the former arising from a deviation of the stoichiometric composition to the Si-rich side. Broad PL bands near 0.8 eV have been observed at 2 K from both 2SA and 3SA samples, and we assign these PL bands to optical radiative transitions intrinsic to β- FeSi₂.

Fe₃O₄/MgO superlattices are grown on MgO(001) substrates. Modulation coherency and sharp interface structures are observed by X-ray diffraction and reflection high energy electron diffraction (RHEED). The RHEED intensity oscillations of the specular and Bragg scattering provide direct evidence that structurally smooth interfaces can be maintained over many bilayers. Similar superlattices are also grown on an Fe-coated MgO(001) substrate for comparison and although the interface is relatively rough due to the misfit between the film and substrate, the modulation coherency is still maintained. RHEED intensity oscillations allow direct microscopic observations of the growth mode which is important for the further development of oxide superlattices.

We developed a radio-frequency-plasma-enhanced chemical vapor deposition (CVD) method with a positively self-biased electrode system. This method enables simultaneous ion-assisted deposition on both sides of a vertically suspended substrate at a grounded electrical potential with one rf generator. In the positively self-biased electrode system, the cathode sheath is located close to the grounded electrode. Accordingly, positive ions are accelerated toward the grounded electrode, producing an ion-assisted reaction on the surface of the substrate. The grounded substrate is advantageous when transporting a substrate holder suspended vertically in an in-line multilayer deposition system. An important feature of this electrode is the electrical capacitance between the hollow anode chamber and the ground which is parallel to the sheath capacitance. Application of the electrode to the deposition process of thin film magnetic recording disks in pure methane gas was investigated. Diamond-like carbon (DLC) films were deposited at a rate of 3 nm/s on the magnetic layer, with uniform film thickness within ±5% over a substrate measuring 95 mm in diameter. Analysis of this film showed that its characteristics are essentially the same as those of DLC film deposited by the conventional method using a cathode sheath on a high-voltage electrode.

An in situ optical observation method for investigations of fluoride surface modification induced by electron irradiation was developed for the purpose of improving the heteroepitaxy of semiconductors on fluorides. Spectra of reflected light after the electron irradiation and continuous intensity change of reflected light at a fixed wavelength during the electron irradiation were observed. It was found that a Ca colloidal absorption band at around 580 nm due to aggregation of Ca atoms appeared as the electron dose to the CaF₂ surface was increased. Progressive P atom adsorption during the electron irradiation was also observed in the case of use of a probing light of 400 nm wavelength. The amount of adsorbed P atoms increased as the electron dose increased and saturated at approximately one atomic monolayer on the CaF₂ surface. Effects of the acceleration energy of electrons on the surface modification were also investigated and it was found that the efficiency of P adsorption increased as the electron energy increased, while that of Ca aggregation decreased. A possible mechanism of this phenomenon is discussed.

NaOCl-polishing of a metal organic chemical vapor deposition (MOCVD)-grown GaAs surface on a Si substrate has been studied by spectroscopic ellipsometry (SE) and atomic force microscopy (AFM). The AFM results indicate that the root-mean-square (rms) roughness of the surface of polished samples is typically 0.3 nm which is about 1/10 that of unpolished samples. The SE data also clearly indicate that the maximum ε₂ (E₂) value is ∼24.5 for polished samples. The thickness of rough surface and void fraction (an effective medium of rough layer consisting of bulk GaAs and void) for unpolished samples are obtained by using the effective medium approximation (EMA) and the obtained dielectric functions from the respective polished samples. The surface thicknesses obtained using this method compare well with those obtained by AFM measurements.

We have investigated Cs-induced reconstruction on the Si(113)3×2 surface using low energy electron diffraction (LEED) and X-ray photoelectron spectroscopy (XPS). For Cs deposition at room temperature, the (3×1) LEED pattern was observed for a wide Cs coverage range. At high substrate temperatures, the (3×1), (1×5+2×) and (2×2) phases were observed with increasing Cs deposition time. The relative Cs saturation coverages of (3×1)-Cs at RT and (2×2)-Cs at 300°C were measured from Cs 3d/Si 2p core level XPS intensity ratios. The results are summarized in a phase diagram as a function of the Cs deposition time.

Thin films of AlN were prepared on sapphire C- and A-plane by chemical vapor deposition. Crystal orientations, oxygen impurity contents and surface acoustic wave properties (SAW) of the films were investigated. Under optimized conditions, epitaxial AlN films were deposited and the crystal orientation relationships were (001)[110]AlN//(001)[110]Al₂O₃ and (001)[110]AlN//(110)[110]Al₂O₃. The crystal orientation of the AlN films decreased as the thickness of the films increased slightly, and the oxygen impurity content was less than 1 atm% for both systems. The dependences of SAW velocity (V _s) and the temperature coefficient of the center frequency (τ _f) on the film thickness were measured, and τ _f was found to increase as the film thickness increased for all measurment directions; however, zero-temperature coefficient was not achieved.

One-dimensionally oriented thin crystals of a phenylene oligomer, p-sexiphenylene (6P), are grown on the poly( p-phenylene) (PPP) film prepared by the friction-transfer method. The 6P molecules tend to grow epitaxially on the (001) planes of alkali halides and the molecular axis is parallel to the <110> directions of alkali halides. The needle crystals orient along two directions. On the other hand, 6P stands on quartz plates with random orientations. The slender crystals of 6P formed on the friction-transferred PPP on alkali halides and quartz plates confirm that the uniaxial orientation has anisotropic optical properties.

The effects of He addition on the growth of crystallized TiO₂ films and the change in both the substrate-incident ion current and plasma parameters in rf magnetron sputtering were investigated. The growth of rutile TiO₂ was achieved on a non-heated substrate in the case of a mixture of Ar (10 sccm) and He (40 sccm) at a rather high total pressure of 20 mTorr. He (50 sccm) sputtering brought about anatase TiO₂ film growth, improving radial inhomogeneity. Mass-resolved analyses of the substrate-incident ion current showed a significant change in species, such as Ar⁺, O₂⁺, O⁺ and Ti⁺, with increasing He. Studies using the Langmuir probe and gridded energy analyzer showed that high-energy electrons which contribute to excitation and ionization of reactant particles increased over the whole discharge space, while electron density decreased with increasing He. Metastable He atom densities were estimated by the optical absorption method and the effect of the Penning process was discussed quantitatively. We pointed out that the increase in high-energy electrons by He mixture is responsible for the change in the substrate-incident ion current and formation of anatase/rutile TiO₂. The Penning effect caused by the metastable He atom was thought to be rather small in the present experimental condition.

The energy loss structure in X-ray photoemission spectra was studied in detail for complicated band structure systems ( LiNbO₃ and LiTaO₃) and simple band structure systems (MgO and α- Al₂O₃). The energy loss structure on the lower kinetic energy side of core lines can be approximated by a sum of four components. The narrower peaks positioned at 7.0, 12.0 and 14.5 eV for LiNbO₃, those at 8.0, 13.4 and 15.8 eV for LiTaO₃, those at 11.3, 15.3 and 18.3 eV for MgO, and that at 14.5 eV for α- Al₂O₃ were assigned to interband transitions from the valence to the conduction bands. The peak positioned at 35.3 eV for α- Al₂O₃ was assigned to interband transition from the O2s level to the conduction band. The broader peaks positioned at 21.8 eV for LiNbO₃, 22.6 eV for LiTaO₃, 23.2 eV for MgO, and 25.2 eV for α- Al₂O₃ were ascribed to the single loss of the bulk plasmon excitation. The peak positioned at 49.9 eV for α- Al₂O₃ was ascribed to the double losses of the plasmon excitation. The observed plasmon energies are larger than the estimated one in the free electron gas model for each substance (10 eV for the complicated and 1 eV for the simple systems).

We measured the quantum-efficiency (QE) dependence of the spin polarization of photoemission from a GaAs-AlGaAs superlattice over a wide range of QE values. In the QE range from 10^-4 to 10^-9, a change in the polarization from 76% to 48% was observed at a wavelength of 778.5 nm. The maximum polarization at 778.5 nm was observed at a QE one tenth of the maximum value. We also observed a similar dependence at other wavelengths. Such a behavior suggests that the vacuum level and surface potential barrier play important roles concerning the spin polarization of photoemission.

The time-resolved optical technique has been applied to RF glow discharge to measure the expansion velocity of the sheath of the RF glow in conjunction with the applied RF voltage. In the technique, the movement of the edge of the negative glow is assumed to indicate the expansion of the electrode sheath thickness. We have proposed a model of the sheath expansion which takes into account acceleration, by RF voltage, of electrons initially in the region from the sheath edge to the floating potential surface relative to the plasma. The values deduced using the model compare well with the experimental results.

The delay time of discharge triggered by a laser-produced plasma is investigated in a mixture of oxygen and nitrogen. When oxygen exists in the mixture, the delay time of discharge has two styles of "late discharge" and "early discharge". The delay time of "late discharge" agrees with the transit time of oxygen negative ions between electrodes. The delay time of "early discharge" approaches the transit time of electrons between electrodes as the applied voltage between the electrodes increases. There exists a threshold voltage at which the delay time leaps from "late discharge" to "early discharge". Calculation of the electron detachment cross section reveals that the leap of the delay time is caused by changing of negative particle triggering discharge from oxygen negative ions to electrons due to an increase in electron detachment as the applied voltage becomes higher.

A two-dimensional self-consistent fluid simulation of capacitively coupled rf glow discharges of electronegative plasma with a cylindrically symmetric parallel plate electrode is presented. The model equations include continuity equations, a Poisson equation, and an electron energy balance equation. The two-dimensional distributions of charged particle densities, electric potential, electron temperature, and ionization rate are calculated. The effects of the applied rf voltage, the driver frequency, and the gas pressure on the discharge characteristics are investigated in detail.

The possibility of use of a sapphire disk as an output window of a high-power and long-pulse gyrotron was investigated using an RF test facility with high-power gyrotrons. The dielectric loss tangents, tan δ=1.8×10^-4 (T/300)^1.3±0.1 for 110 GHz and tan δ=2.6×10^-4 (T/300)^1.3±0.1 for 170 GHz (T: absolute temperature in K), were obtained in the temperature range of 300 K to 600 K. The experimental results showed that RF reflection from the window became large due to the permittivity change with increasing temperature, which limits the applicable power for the sapphire window rather than the thermal runaway. Moreover, it was found that a 170 GHz, 1 MW, CW gyrotron is feasible using a two-port window system consisting of sapphire double disks more than 18 cm in diameter with a flat RF power profile.

A new application of the atmospheric cold plasma torch has been investigated. Namely, the surface treatment of an air-exposed vulcanized rubber compound. The effect of plasma treatment was evaluated by the bondability of the treated rubber compound with another rubber compound using a polyurethane adhesive. The adhesion property was improved by treatment of the rubber compound with plasma containing oxygen radicals. Physical and chemical changes of the rubber surface as a result of the plasma treatment were analyzed by field emission scanning electron microscopy (FE-SEM) and fourier transform infrared attenuated total reflection (FTIR-ATR).

Wave propagation characteristics in a 70 GHz waveguide containing transversely magnetized p-type and n-type InSb slabs have been studied experimentally. Surface wave resonance occurs in this configuration. It is also shown that the magnetic field required for slow surface wave resonance is greatly reduced by the addition of the p-layer.

Ion collection by a hollow probe has been studied in electron cyclotron resonance (ECR) microwave plasma under a divergent magnetic field where an ion beam was accelerated along the plasma axis. The ion current collected by the probe was found to consist of currents proportional to the geometric solid angle from the probe collector caused by the ion beam parallel to the probe axis. The effect of the ion beam energy and its injection angle to the probe has been also discussed.

An instrument which can be used to measure both sound velocity and attenuation in small specimens (about 5×5×5 mm³) is described. A dual path comparison method based on the ultrasonic pulse transmission method was used in the present work. For this measurement method, an easy-to-handle mercury delay line cell was devised. To test applicability of the instrument, sound velocities were measured at 10 MHz for synthetic silica glasses and a single crystal of sodium chloride along the [100] axis at room temperature. The accuracy of the measurements was better than 0.2%. As a further example, sound velocities and attenuation in the ceramic barium titanate, BaTiO₃, were measured in the temperature range from room temperature to 200° C. A large peak was detected in the attenuation near the Curie temperature.

We introduce an evanescent-wave-spectroscopic fiber optic pH sensor. This sensor has part of the polymer cladding doped with either congo red (CR), which responds to pH=3 to 5, or methyl red (MR), which responds to pH=5 to 7, as the pH sensor film. In this study we succeeded in expanding the measurable pH range by using two pH sensor probes with the different pH sensitivities. In addition making the pH fiber sensor into a probe structure enabled us to achieve high pH sensitivity and to monitor the pH value easily. Reducing the quantity of doped dye made it possible to use the sensor for objects highly sensitive to pH chemical equilibrium.

We report the use of a new technique to measure the X-ray detection efficiency of a charge-coupled device (CCD) with subpixel resolution. The new technique makes use of a parallel X-ray beam and metal mesh placed just in front of the CCD. The CCD camera we used is a conventional system using the TC213 (Texas Instrument Japan (TIJ)) whose pixel size is 12 µ m ×12 µ m with one million pixels. The mesh has 4 µm diameter holes spaced at 12 µm intervals. We produced an efficiency map within a typical pixel showing the gate structure in detail: a virtual gate, a clock gate and an antiblooming gate. The gate structure we measured is consistent with the manufacturer's design value. By selecting single pixel events, we detected a pixel boundary. Additional plans for application will also be discussed.

Electron beam (EB) direct writing systems have often been used for fabricating sub-half-micron advanced devices because EB direct writing is the most practical method for making the required patterns. Recently, the cell projection (CP) method has become indispensable for increasing the writing throughput in the EB direct writing system. However, it is considered that resist heating may be seriously aggravated below the quarter-micron level when the CP method is used, because the total deposited energy, which is irradiated by one CP EB shot, is almost the same as that irradiated by one variably shaped (VS) EB maximum size shot. Resist heating in the case of the CP method is calculated by a finite element method using the ANSYS (Ver. 5.0A: ANSYS, Inc.) program. In particular, thermal diffusion calculation is mainly carried out under the conditions of 50 kV acceleration voltage and 10 A/cm² current density for practical application to advanced device fabrication. The calculated results suggest that resist heating in the CP method is mainly caused by the horizontal thermal flux between plural EB shots within the area of one CP shot, by the same mechanism as proximity resist heating under the VS method. Therefore, CP EB writing causes horizontal-mode resist heating. In particular, when a low current density is used, this resist heating mode arises significantly. However, CP writing with high acceleration voltage causes a reduction in the rise of the resist temperature, which causes resist heating. When the EB irradiation time is longer than 1.0 µ s under practical EB writing conditions, the resist temperature increases proportionally to the decrease of writing pattern size in the case of the CP writing with a maximum shot size of 5.0×5.0 µ m. It is also shown that the larger the beam blur of an incident beam, the more serious is the resist heating. When a highly sensitive resist (10 µ C/cm²) is used under these practical conditions, however, resist heating in the CP method is prevented without writing throughput degradation regardless of the CP maximum shot size, because the resist temperature does not rise above the thermal denaturation temperature of standard EB resists. Accordingly, the maximum CP shot size, which affects the writing throughput, is determined by the proximity effect and the Coulomb interaction for fine pattern fabrication.

A new and simple expression developed by us has been used for the estimation of contrast of thin amorphous specimens in the tilted-beam mode of dark field (DF) imaging using a conventional transmission electron microscope (CTEM). DF contrasts calculated for thin amorphous films of some representative elements in the electron accelerating voltage range between 0.1 to 3.0 MV have been compared with the corresponding bright field (BF) contrasts. The magnitude of the gain in contrast in the DF mode compared with BF one, observed over a wide range of accelerating voltage (φ), microscope aperture (α) and atomic number (Z) of elements, has been highlighted.

Ultrasonic velocity and absorption of liposome suspensions of dimyristoyl phosphatidylcholine in the liquid crystal phase were measured in the megahertz region, varying the temperature and the size of liposomes. Temperature dependence of ultrasonic properties indicated a critical phenomenon due to the gel-liquid crystal transition at about 24° C, which is characterized by a dip in ultrasonic velocity and an anomalous increase in absorption. However, the anomalous frequency dependence of ultrasonic absorption remained significant even at 35° C where the critical phenomenon almost disappeared. This additional absorption anomaly strongly depended on the size of liposomes. The value of viscosity of multilamellar liposome suspensions estimated from the ultrasonic absorption was inversely proportional to the frequency of the ultrasound and diverged to infinity in the low-frequency limit, while this kind of anomaly was not observed for small unilamellar lipid vesicles. This liposome size dependence indicated clearly that the additional absorption anomaly in multilamellar liposomes was due to the fluctuation of their lamellar structure. The possibility of estimating the content of stacked membrane layers in biological tissue by ultrasonic absorption is discussed based on the ultrasonic measurements of liposomes.

For quantitative evaluation of body motion, a fully noncontact and unconstraining monitoring method was developed by introducing image sequence analysis. A spatiotemporal local optimization method was applied to determine optical flow in the image sequence. The optical flow visualized the apparent velocity field of the entire body motion, including both breast movement of respiration and posture changes in a bed. The experiment was carried out under regulated posture changes and under a sleeping condition by measuring heart rate, respiration and digitized image sequences using a video camera. A temporal increase in heart rate reflected the magnitude of physical activities. We proposed two candidate parameters for evaluation of respiratory and physical activities based on comparison among experimental results. The average of squared motion velocities reflected the magnitude of physical activities. The representative field-averaged component showed a waveform with periodic fluctuation corresponding to that of respiration obtained with a nasal thermistor.