Search

Negative substrate-bias effects on diamond heteroepitaxial nucleation have been investigated using microwave plasma-assisted chemical vapor deposition. The density and the percentage of azimuthally ordered particles grown on carburized Si(001) or on high-quality β-SiC(001) have been estimated by varying the bias-treatment conditions (reaction pressure, microwave power and bias voltage). The percentage of azimuthally ordered particles increases under the following conditions: pressure of 50 Torr, substrate temperature of ∼900° C and bias voltage of -50 V. Smooth heteroepitaxial diamond films of ∼4 µ m thickness have been successfully synthesized on high-quality β-SiC(001) by the bias treatment under the above conditions followed by a two-step growth process: fast growth and fast growth. The initial growth of these films has been governed by the Volmer-Weber mode, where [1̄10]- or [110]-directed protrusions have been formed on β-SiC(001). Diamond islands grew along the grooves between the β-SiC protrusions in the initial stage.

Distortion of SiC X-ray mask substrate during anodic bonding has been found to occur due to the difference in the thermal expansions of SD glass frame and Si wafer deposited with SiC, and the distortion characteristics can be explained by the bi-metal effect. The reduction in the difference in the expansion is required to obtain lower substrate distortion, and has been realized by using the SD-2 glass, which has the most accurate matching of the expansion to Si, and by lowering the bonding temperature. Any other parameters than the bonding temperature, such as the applied voltage and the bonding time, have been found not to cause the distortion. Homogeneously concave mask substrate bonded at 230°C, using 4-mm-thick SD-2 glass frame and 1-mm-thick wafer, has the flatness values of 2.18 µm and 0.26 µm for the wafer and the membrane area, respectively.

The mechanism of X-ray mask membrane displacement induced by synchrotron radiation (SR) has been discussed. Silicon nitride (SiN) and silicon carbide (SiC) membranes were irradiated by SR in a 1 atm helium ambient. SR-induced displacement for both membranes was 25-97 nm (σ). Oxygen concentration in both SiN and SiC was below 0.01 in O/Si atomic ratio. Although an increase in dangling bond density of SiN was observed, no remarkable increase in spin density was detected in SiC. Moreover, the most important finding was that thin oxides were grown on the membrane surface after SR irradiation. From these results, it is considered that the oxide growth on SiC membrane surfaces, and both the oxide growth and the increase of dangling bond density in SiN play an important role in the SR-induced displacement for the X-ray mask membranes.

We have obtained a smoother surface of SiC membranes than that of as-deposited SiC using etch-back and polishing methods, and investigated the influence of surface roughness of SiC films on internal stress and film structure of X-ray absorber films and optical transmittance of SiC membrane. Surface roughness for the as-deposited SiC surface is 15 nm (Ra), whereas that for etch-backed SiC is 6.9 nm (Ra) and that for polished SiC is 4.3 nm (Ra). Optical transmittance of as-deposited SiC membrane is 48% (λ=633 nm) and increases by a few percent with decrease in the Ra value. Internal stress of sputtered Ta tends to show gentler change with increasing surface roughness of SiC film. Film structure of Ta film is largely influenced by the surface roughness of SiC films at low Ar pressure. Sub-half micron absorber patterns with vertical walls, however, have been obtained even on rough surface SiC.

A morphological defect with the shape of a torus was observed at the heterointerfaces between GaAs and Al x Ga 1- x As layer grown by liquid phase epitaxy. The structure of the toroidal defect was investigated, and it was found that the defect was composed of a region with aluminum content different from that of the surrounding normal region. The toroidal defect is believed to be caused by inhomogeneous mixing of solutions during sliding due to oxygen contamination.

The effects of changing a unit time length of a planning horizon from a month to a week on the optimum planning horizon were examined by calculating the optimum planning horizon through the methods proposed by Nagasawa, Nishiyama, and Hitomi (1982). It was found that the optimum planning horizon decreased by 20-30% in calendar time when the unit time length was changed from a month (monthly scheduling) to a week (weekly scheduling). However, the degree of this decrease was much smaller than the (65% shown by Bernardo (1978), and it followed that the optimum planning horizon largely increased in the number of periods with this change of the unit time length. It was also clarified that the large amount of the decrease shown by Bernardo was derived on the basis of the erroneous analysis of the relation between cost coefficients and the unit time length. Consequently, weekly scheduling was not always preferred to monthly scheduling

A generalized concept for ‘Planning Horizon’ is introduced based upon a property of an analytical solution to a multiperiod production planning problem converging into an extreme as a length of planning horizon increases. Two algorithms for determining the value of this new ‘Planning Horizon’ are proposed. One is for a model with stationary demand, and the other is for a model with trigonometric seasonal demand. A numerical example is given to show that the recommended decision algorithms can be employed simply for practical use.

The V-grooved GaAs substrate was found to be deformed before starting LPE growth. The groove became noticeably shallow and wide after heat treatment at 850°C for l h in H 2 atmosphere. Vapor-phase transport of GaAs from a convex surface to a concave surface on the V-grooved GaAs substrate was found to be responsible for the deformation. The transport is caused by the dependence of an equilibrium vapor pressure of the substance on the curvature of the surface. A dissociation of GaAs and an evaporation of As at a convex surface may start the transport. Near-congruent evaporation seems to occur in GaAs placed in a closed space, even at a temperature as high as 850°C.

全般的生産計画において,単位加工時間や費用係数が見積もり誤差を伴う場合に,この誤差が計画生産量の最適性にいかなる影響を及ぼすかを解析した.その結果,計画生産量に許容される最適生産量からのずれ量が需要量の変動幅の 1% の場合,単位加工時間で±1%,費用係数で±5% の見積もり誤差ならその影響を無視できること,また計画生産量の許容ずれ量が上記値より小さければ,見積もり誤差の影響をほとんど無視しえないことを示した.

The recent paper of Teny and Kochhar (1983) concluded that the solution calculated by using the functional space analysis technique presented by Hitomi and Nakamura (1976) is inferior to that rendered by the quadratic programming approach (Bcale 1968). This paper shows the contradiction included in their discussion and denies the above conclusion. Functional space analysis is also developed to obtain a mixed-integer solution.

Deficient mitochondrial acetoacetyl-CoA thiolase in fibroblasts from four patients with 3-ketothiolase deficiency was studied using immunochemical methods. We also examined fibroblasts from two heterozygotes, the mother and the brother of the case 1 patient, identified on the basis of the results of the enzyme activity measurements, using 2-methylacetoacetyl-CoA as substrate. The results were as follows: 1 ) in fibroblasts from all four patients, the thiolase activity using acetoacetyl-CoA was not activated by K+, although that of the controls and the heterozygotes was activated about 2-fold. 2) by immunoblot analyses, mitochondrial acetoacetyl-CoA thiolase was not detectable in fibroblasts from cases 2 and 3, although a very faint band was seen in tissues from cases 1 and 4. However, the band of mitochondrial 3-ketoacyl-CoA thiolase was clearly detected in all patients to the same extent as in the controls. 3) mitochondrial acetoacetyl-CoA thiolase was observed after pulse labeling for 1- h and a 72-h chase period in three cell lines (cases 1, 2, and 4), but was fainter compared to the controls. In another cell line (case 3), a fluorographic band at the same position was detected following a 1-h pulse, but disappeared following a 6-h chase. These results demonstrate heterogeneity in the enzyme defect resulting in a deficiency of mitochondrial acetoacetyl-CoA thiolase in fibroblasts from patients with 3-ketothiolase deficiency.

Temperature-dependent Hall effect investigations in the channel of lateral 3C-SiC LDDMOSFETs with nitrogen(N)-implanted source/drain regions are conducted. The free electron concentration and the electron Hall mobility are independently determined. A maximum electron Hall mobility of 75 cm2/Vs is observed. The gate oxide withstands electric field strengths up to 5 MV/cm. A high density of interface states of a few 1013 cm-2eV-1 close to the 3C-SiC conduction band edge still lowers the performance of the MOS device.