Masahiro Iwasaki

Nonequilibrium atmospheric pressure plasma with ultrahigh electron density and high performance for glass surface cleaning

We produced a nonequilibrium atmospheric pressure plasma by applying an alternative current between two electrodes. The gas temperature and electron density were evaluated using optical emission spectroscopy. It was found that the plasma had gas temperatures from 1800to2150K and ultrahigh electron densities in the order of 1016cm−3. A remarkably high oxygen radical concentration of 1.6×1015cm−3 was obtained at a 1% O2∕Ar gas flow rate of 15slm (standard liters per minute). Contact angles below 10° were obtained in the process of glass cleaning with a plasma exposure time of 23ms.

Roles of oxidizing species in a nonequilibrium atmospheric-pressure pulsed remote O2/N2 plasma glass cleaning process

Atmospheric pressure plasma treatments have attracted attention for various application processes. The effect of O2 additions below 0.2% to N2 was investigated for the efficiency of removing organic contaminants on a glass surface using nonequilibrium atmospheric-pressure pulsed plasma. A remarkably high efficiency of cleaning was obtained by a plasma treatment with ca. 0.03% O2 additions to N2. The concentration of ozone (O3) and the ground-state oxygen radical [O(P32)] were measured using ultraviolet absorption spectroscopy and vacuum ultraviolet laser absorption spectroscopy, respectively. It was found that the key factors for surface cleaning were the scission of carbon bonds due to ultraviolet irradiation and subsequent oxidation due to O(P3), and that the surface cleaning proceeded in broad areas due to the photodissociation of O3.

Effect of Low Level O2 Addition to N2 on Surface Cleaning by Nonequilibrium Atmospheric-Pressure Pulsed Remote Plasma

The effect of O2 additions below 0.2% to N2 was investigated for the cleaning of organic contaminants on an indium tin oxide film using a nonequilibrium atmospheric-pressure pulsed plasma. A remarkably high cleaning efficiency was obtained for plasma treatment with additions from 0.025 to 0.1% O2 to N2. The concentration of the ground state oxygen radical [O(3P)] was measured using vacuum UV laser absorption spectroscopy. It was found that the key factor for the surface cleaning was the scission of carbon bonds due to UV emissions, and subsequent oxidation due to O(3P) and ozone.

Ion attachment mass spectrometry of nonequilibrium atmospheric-pressure pulsed remote plasma for SiO2 etching

An etching process for SiO2 that employs atmospheric-pressure pulsed remote plasma with a mixture gas (CF4∕Ar) has been demonstrated. The etch rate increased by increasing the pulse frequency and also increased rapidly with the addition of a small amount of O2 gas. A SiO2 etch rate of 400nm∕min was obtained without any bias supplied to the substrate, and more than 8μm∕min was obtained with H2O addition. The exhaust gas from the plasma source was investigated using ion attachment mass spectrometry (IAMS). With IAMS, the polymerized species were measured without the generation of fragment ions. The results suggest that HF molecules are generated by H2O addition to CF4 and that the etch rate of SiO2 depends on the density of HF molecules.

Self-creating and organizing neural networks with weight duplication

We propose a new self-creating and self-organizing neural networks utilizing weight duplication. The proposed model consists of competitive and input layers. Weights of the activated node are modified by competitive learning. A new node is created depending on the activation frequency. The weights of a daughter node are duplicated from the mother node. The mother node has refractory period just after the creation. The weights of the daughter node and those of her mother node will be similar after the refractory period. A mother-daughter relationship represents the hierarchical structure of input data. It is possible to organize the column by itself with similar features which has hierarchical structure. The represented features will become precise when descending the hierarchy. Some simulations are carried out to show the feasibility of the proposed model.

A study on with the bottom-up and the top-down procedure in hierarchical neural networks

This paper presents a new framework for hierarchical neural networks. It is based on the interaction between bottom-up and topdown procedure to realize human-like cognitive processes. The proposed system consists of 3 layers, input layer, feature extraction layer and symbol layer. Input layer is a kind of associative memory and associate by itself. When the association fails, feature extraction through bottom-up procedure and attention to the parts of input pattern through top-down procedure make it possible to associate target pattern. Some simulation results show the feasibility of proposed system.

Surface Modification Process of Contact Lens Using Three-Phase AC Excited Nonequilibrium Atmospheric Pressure Ar Plasma

A surface modification process on the contact lens using a nonequilibrium atmospheric pressure remote plasma excited by three-phase AC power source has been investigated. Two dimensional spatial distributions of optical emissions of Ar and OH were measured, and gas temperature and electron density were estimated using the optical emission spectroscopy. It has been found that the gas temperature in the upstream region of the remote plasma is lower than that in the downstream region due to the interfusion of ambient air. The heat-sensitive material such as plastic could be processed without thermal damage by the nonequilibrium atmospheric pressure remote plasma in the proper low-temperature region.

Determining field of visual attention in associative memories

A new method of determining field of visual attention using associative memory is proposed. Paying attention to the typical features of the patterns plays a great role in recognition scheme of human beings. In conventional approaches, the field of visual attention is determined by the designers a priori and fixed while the systems are running. The proposed model is capable of determining the field of visual attention automatically. The proposed model is hierarchical network which consists of associative memory layer and feature layer. The connections between the layers represent the features of the input patterns and are generated automatically. These connections are modified through the Hebbian learning rule. In the proposed model, features are extracted through bottom-up procedure. The connections corresponding to the most excited unit in the feature layer are used as attention points in the top-down procedure. Through these procedures, the system is capable of associating appropriate patterns.

A Novel Silicon-Dioxide Etching Process Employing Pulse-Modulated Electron-Beam-Excited Plasma

We developed a novel silicon-dioxide (SiO2) etching process employing a pulse-modulated electron-beam-excited plasma (EBEP) and investigated its etching characteristics. The pulse-modulated EBEP was found to have an excellent potential for resolving problems such as the plasma-induced thermal damage on a photoresist film on a substrate and the nonuniformity of etching in a conventional EBEP. Furthermore, the anisotropic SiO2 etching at a high rate of 375 nm/min was successfully performed without any thermal damage to the photoresist film. The etching rate of the pulse-modulated EBEP was a few tens times higher than those of conventional etching processes without any additional bias power supply. As a result, the microfabrication on the end surface of the optical fiber was demonstrated.

Extraction of features and attention areas in hierarchical neural networks

A new method of automatic extraction of features and attention areas from input data is proposed in this paper using a hierarchical neural network. Human-beings can distinguish useful information from those in ambiguous and noisy environments using attention functionality. It is said that human-beings may recognize information roughly at first, and pay attention to the detailed features to confirm what the information really means. The main difficulty in realizing the attention function in the computational model is to determine the part to which the system pays attention. To determine the attention areas automatically, the proposed model consists of three hierarchical neural networks, associative memory layer, middle layer and symbol layer. The associative memory plays a great role for finding out the features of the input pattern. The middle layer corresponds to the feature extracting layer. The units in each layer connect to the units in other layers. The connection weights between the layers are modified through a Hebbian learning rule. The connection weights between the input and the middle layers represent rough features of the input patterns, while those between the symbol and the input layers represent the attention areas. In the proposed model, these features and attention areas can be extracted automatically without the designers knowledge.