SrTiO3 thin films by mocvd for 1 gbit DRAM application

Abstract

Three important aspects of the preparation of SrTiO₃ thin films by MOCVD are discussed in detail in view of the application of these films as the capacitor dielectric of Gbit-scale DRAMs: CVD reactions in the Sr(DPM)₂-Ti(i-OC₃H₇)₄-O₂ system, step coverage and relations between microstructure and electrical properties. The effect of the substrate temperature on the Sr and Ti deposition rates was first investigated for thermal and ECR CVD SrTiO₃ films. SrO and TiO₂ deposition by thermal CVD above 550°C were found to be controlled by the surface reaction and gas transport, respectively, whereas both SrO and TiO₂ deposition are controlled by gas transport for ECR CVD at 450 to 600°C. The influence of the Sr and Ti deposition regimes on the step coverage of SrO, TiO₂ and SrTiO₃ were then assessed. SrO films prepared by thermal CVD at 600°C exhibited the best step coverage, indicating that a relation exists between reaction controlled deposition and good step coverage. The effect of the film composition and film thickness on the microstructure of SrTiO₃ thin films were finally investigated and correlations were made to other analyzed physical and electrical properties. Polycrystalline perovskite phase SrTiO₃ films were obtained for a composition 0.7 ≤ Sr/Ti ≤ 1.2. The best crystallinity, maximum permittivity and maximum refractive index were obtained for Sr/Ti = 0.95. Titanium rich films are thought to be composed of a mixture of a titanium rich amorphous phase and crystalline SrTiO₃, and strontium rich films are believed top correspond to a (SrTiO₃)_m (SrO)_n structure. The dielectric constant slowly decreased as the film thickness was reduced. The sharp decrease observed near 400–500 Å could be due to the existence of some perturbed layer at the interface with one or both of the electrodes     

Fabrication of a microelectromechanical system mirror array and its drive electrodes for low electrical interference in wavelength‐selective switches

This article describes the fabrication of a microelectromechanical system (MEMS) mirror array and mirror‐drive electrodes with high‐aspect‐ratio gold walls that reduce electrical interference in a wavelength‐selective switch (WSS). The MEMS mirror array, in which a lot of closely spaced adjacent mirrors are electrostatically operated, can be fabricated with a high yield by encapsulating the mirrors with an organic film that protects them from process damage. The gold walls with a high‐aspect ratio are formed in the narrow space between adjacent mirror‐drive electrodes by using thick‐multilevel interconnection technology. Because of these walls between adjacent electrodes, each MEMS mirror operates with low electrical interference. © 2011 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.     

Modeling and compensation for angular transmission error in harmonic drive gearings

This paper presents a novel modeling and compensation approach for the angular transmission error in harmonic drive gearings. In the modeling, physical phenomena of the transmission error due to nonlinear elastic deformations in micro-displacement region are especially dealt with, as well as the synchronous component which has been discussed in a variety of conventional studies. On the basis of the analyses of the phenomena, the nonlinear elastic component is mathematically modeled by applying a modeling framework for the rolling friction with hysteresis attributes. The proposed transmission error model has been adopted to the positioning system as a model-based feedforward compensation manner. Experimental results using a prototype show the effectiveness of the proposed modeling and compensation. Copyright © 2009 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.     

Biological stereoselective reduction of ­4‐methylcyclohexanone and ­4‐ethylcyclohexanone by anthracnose ­fungi

The biotransformations of 4‐methylcyclohexanone and 4‐ethylcyclohexanone were investigated using 10 kinds of anthracnose fungi as biocatalysts. 4‐Methylcyclohexanone and 4‐ethylcyclohexanone were reduced to the corresponding cis‐ and trans‐alcohols respectively. In the case of 4‐methylcyclohexanone, it was transformed to mainly trans‐4‐methylcyclohexanol by all the fungi examined. In particular, the ratio of cis‐ and trans‐alcohol products was shown to be 1:81 with high stereoselectivity by Colletotrichum lagenarium after a 7‐day incubation period. The biotransformation of 4‐ethylcyclohexanone by C lagenarium, C dematium MAFF410046, C trifolii MAFF305389, C fragariae, C atramentarium MAFF712102, C lindemuthianum (C‐1), C lindemuthianum (C‐3) and C lindemuthianum (C‐13) produced mainly trans‐4‐ethylcyclohexanol. On the other hand, cis‐alcohol was formed with stereoselectivity by Glomerella cingulata and C graminicola MAFF305460. © 2000 Society of Chemical Industry     

Size-dependent mixing characteristics of volatile and nonvolatile components in diesel exhaust aerosols

Mixing characteristics of particles of different volatilities from a diesel engine were studied with two tandem differential mobility analyzers (TDMAs) and an aerosol particle mass analyzer (APM). In both TDMA systems, a heater was located in the aerosol path between the first and second DMAs. Diesel exhaust particles that were size-selected in the first DMA were passed through the heater, and the change in particle size due to loss of volatile components was determined by the second DMA. On the basis of the volatility measurements, the particles could be separated into two overlapping modes that varied in peak diameter and magnitude depending on the engine operating conditions. Particles in the smaller size mode were almost completely volatile, while those in the larger size mode contained a nonvolatile core. The TDMA data inversion technique used here allowed accurate determination of the mixing ratios of the two types of particles. These data were in turn used to validate a simple fitting method that uses two log-normal curves to obtain the mixing ratios. In some experiments, the APM was used downstream of a TDMA to directly measure the particle mass loss due to evaporation. The loss determined bythe TDMA-APM system was significantly greater than that calculated from mobility size changes measured solely with the TDMA. The TDMA-APM results were used to calculate the size-dependent mass concentrations of volatile and nonvolatile components for particles in the size range from 70 to 200 nm.     

Volatile flavour components of Allium grayi regal

The volatile flavour components of Allium grayi Regal have been investigated by a combination of chromatographic and spectroscopic methods. Among the more than 90 component peaks observed by capillary gas chromatography of the oil obtained by ether extraction of the steam distillate, 47 compounds were positively identified. They included 11 sulphur compounds, 5 alcohols, 7 aldehydes, 3 ketones, 2 furanones and 19 miscellaneous compounds.     

C16 hydroxylation of 3beta-hydroxy-delta5-steroids during the early neonatal period

Temporal changes of the serum levels of 16-hydroxypregnenolone (3beta,16alpha-dihydroxy-5-pregnen-20-one) 3-sulfate (16-OH-Preg S) and 16-hydroxydehydroepiandrosterone (3beta,16alpha-dihydroxy-5-androsten-17-one) 3-sulfate (16-OH-DHEA S) were investigated by analyzing the levels of their precursor steroids, pregnenolone (3beta-hydroxy-5-pregnen-20-one) 3-sulfate (Preg S) and dehydroepiandrosterone (3beta-hydroxy-5-androsten-17-one) 3-sulfate (DHEA S), respectively, in the early neonatal period. The serum levels of these steroids were measured by GC-MS in full-term (gestational age: 37-41 weeks), pre-term (gestational age: 28-36 weeks) and extremely immature (gestational age: 24-27 weeks) infants. The changes in 16-hydroxysteroid production were also investigated by analyzing the ratios of the serum levels of 16-OH-Preg S and Preg S (16-OH-Preg S/Preg S ratio), and 16-OH-DHEA S and DHEA S (16-OH-DHEA S/DHEA S ratio). It was confirmed that the 16-hydroxylation of DHEA S and Preg S increased after birth, and the 16-OH-Preg S/Preg S ratio in full-term infants was significantly higher than in pre-term and extremely immature infants at days 0, 1-6 and 7-13. On the other hand, there were no significant differences between the 16-OH-DHEA S/DHEA S ratios of the three groups at days 0, 1-6 or 7-13. The mechanism of differences in the 16-hydroxylation of Preg S and DHEA S is also discussed.     

Use of LNG‐Oxygen Gas Mixture in RH Degassing Unit

Mixed gas of liquefied natural gas (LNG) and oxygen was injected into the vacuum vessel during non‐decarburization period using a multifunctional lance. Tthis study reports the experimental results when applying this process in a 100‐ton piece of equipment and reports the changes in temperature of the refractory surface and molten steel. Using a multifunctional lance allowed heat supplementation by injecting LNG‐oxygen mixed gas into the vacuum vessel of a RH device and then burning the gas near the outlet of the injection nozzle. The following was found: The minimum surface temperature required so that the skull would not adhere to the RH vessel was calculated as about 1643K (1370°C). When the interior of the RH vacuum vessel was preheated with a multifunctional lance so that the temperature would be higher than this, skull adhesion to the vessel did not occur. Calculation of heat balance for burning under atmospheric pressure showed that about 70% of the heat quantity of burning was supplied to the refractory. When mixed gas was burned during the RH process, it was shown from the temperature change of molten steel that about 52% of the burning heat was transferred to the steel. Heat quantity removed through lance cooling water was proportional to the charge length of the lance, and emissivity ε was equivalent to heat transfer quantity of 0.38.     

Burning Characteristic of LNG‐Oxygen Mixed Gas used for Vacuum Degassing Process

This study was conducted using LNG (Liquefied Natural Gas) as fuel which has methane gas as its main component, and theoretical and experimental evaluation was made of the burning conditions of LNG‐oxygen mixed gas under atmospheric and subatmospheric pressure in a vacuum chamber used for the RH process. Conclusions reached in this study of the conditions involved in injecting LNG‐oxygen mixed gas in a vacuum chamber of the RH device and burning it near the nozzle outlet to raise the surface temperature of brick were as follows: (1) Laminar burning velocity of mixed gas of LNG and oxygen was fastest when the mixture ratio was 1:2.3, and when the velocity was proportionally related to the square of temperature and 0.5 root of pressure. (2) Turbulent burning velocity u_t was calculated from the research results of Andrews. (3) It is possible to calculate the turbulent burning velocity and to evaluate u_t/u_m which is the index of stability of the burner flame by experimentally obtaining laminar burning velocity.