We have developed an automatic air monitoring apparatus for measuring trace amounts of acidic or basic contaminants in clean room air. The apparatus includes an ion chromatograph and uses a new air sampling method. Its absorbing vessel is a tube made of quartz glass whose inner surface is cleaned with ultraviolet light so as to be highly hydrophilic. When an air sample is bubbled into 6 ml of absorbing water, the bubbles rise stably to the upper part of the tube and the inner surface is coated with water film. Since the air sample has sufficient contact with the absorbing water, analytical components can be effectively absorbed into the water. It takes 27 min to analyze a sample consisting of organic and inorganic acid of more than 0.1 μg/m3 in air.
Many organic compounds are present in cleanroom air of liquid crystal display (LCD) manufacturing factories, it is therefore important to remove organic compounds from the clean room air. The purpose of this study is to investigate the removal performances of air washer for monoethanolamine (MEA), acetic acid and dimethyl sulfoxide (DMSO), which are commonly found in the LCD cleanroom air. In the removal test for a single component, the removal efficiency of MEA by air washer was 57 % and that of acetic acid was 38 % at the washer water/air ratio (L/G) of 0.8. In the test with the mixed gas, the removal efficiency of MEA, acetic acid, and DMSO by the air washer and cooling coils was 95 %, 97 %, and 92 % respectively. In the case of an actual LCD cleanroom equipped with the air washer in circulation air conditioning units, the concentration of MEA, acetic acid, and DMSO is estimated to decrease to 9 %, 28 %, and 10 % of that without the air washer.
The influence of wafer storage environment on oxidation and organic contamination of Si surfaces has been investigated. And the electronic reliability of thin (2.8 nm) SiO2 films of metal-oxide-semiconductor (MOS) capacitors and MOS transistors has been measured as a function of the storage method. We found that shielding wafers from visible light is effective to prevent oxidation of silicon. Hydrogen terminated p-Si (100) (8 ∼ 12 Ωcm) wafers were stored in wafer boxes under various brightness levels. The oxidation rate in a dark box (∼ 0 lx) is found to be about one order of magnitude as small as that in a light box (∼ 1,000 lx). The MOS capacitors were fabricated by adding a storage process with various contamination levels before gate oxidation. It was indicated that for samples stored in the polyethersulfone (PES) box with a UV/photoelectron cleaning unit, the organic contamination level was substantially reduced, resulting in an improvement of the time dependent dielectric breakdown characteristics of the gate oxides. Furthermore we fabricated n-channel MOS transistors and investigated the influence of the organic contaminant before and after the gate oxidation on hot-electron degradation of the oxide. The wafer surfaces were contaminated with organic gases during the storage in a front opening unified pod (FOUP) made of polycarbonate for 6 h. In the result, the neutral traps were generated by hot-electron injection. It was shown that the density of the generated traps was larger for the pre-oxidation contamination than for the post-oxidation contamination. The model for the trap generation by the organic contamination was also discussed.
In the manufacturing processes of electric devices and systems, anti-electrostatic technology is applied to prevent production yield reduction due to electrostatic charge. Ionizers with corona discharge electrodes which emit air ions neutralize electrostatic charge. High voltage sources and cables of ionizers generate noises due to ESD (Electro-static Discharge), which cause deterioration and malfunctions of electric devices and systems. We have developed a low noise air ionizer with single crystal silicon emitters which generate corona-discharge at 6 kV AC voltage of 50/60 Hz. The count of particles with diameter larger than 0.05 μm measured at 22 cm away from the emitter was less than 70 count/m3 (2 count/ft3) during continuous 6-month operation. The deterioration rate of the Si emitters was about 0.8 μm /month.
Various high purity process gases are used in the semiconductor manufacturing process. Contamination of process gases by molecular contaminants from pipe wall and gas transport system is inevitable. In order to ensure the high purity of process gas, direct purification technique at the point of use is required. We propose a new separation technique for molecular contamination, which utilizes preferential ionization and electrical migration of ions. In the present work, gas flow containing volatile organic compound (VOC) vapor is divided into two flows while irradiating the flow with alpha-ray or soft X-ray under DC electric field. Ionized VOC vapor in one flow electrically migrates into the other flow causing VOC rich and poor flows. The maximum VOC separation efficiency is 67 % from the nitrogen gas containing toluene vapor of 0.15 ppm. The VOC separation efficiency increases with decrease in VOC concentration. The separation of VOC is affected by carrier gas components. The effective separation of VOC requires a given irradiation time for VOC molecules to acquire the electrical charge from carrier gas ions. When the irradiation zone is sufficiently large, the VOC separation efficiency is determined by the separation efficiency of cations.
Monitoring of particles generated in a process chamber is one of the effective means to control the semiconductor manufacturing process. A sampling probe is inserted into the process chamber during the reaction in order to monitor the particles generated by plasma enhanced chemical vapor deposition (PECVD) with a light scattering particle sensor. As a result, it is found that the factors such as plasma power, discharge time, process pressure, and material gas flow rate significantly influence the number of particles generated in the chamber. It is also shown that the particle size distribution varies drastically by the particle coagulation and that PECVD with modulation discharge plasma is effective in controlling gas phase particle generation without lowering the deposition rate.
In this study, we estimated the number of airborne microorganisms a patient might inhale while nurses dispersed particles during their care. Although the amenity in biological clean room (BCR) is of great importance for immunocompromised patients such as post bone marrow transplantation, taking care of a patient in BCR with direct contact or face-to-face is strictly limited in many facilities in Japan. Little was known about the actual risk of air contamination caused by nurses. We built a model of a patient's room with HEPA filter unit in order to conduct an experimental study of bed making, which seems to generate the most particles among all nursing care behaviors. While changing bed cloths, we samples aerosol by M Air TTM to find colonies of bacteria and measured the number concentration of particles ( > 0.1 μm) by a laser particle counter. There was no significant correlation between colonies and particles (p-value > 0.05), although the colony data followed poisson distribution. The air exchange rate of the affected part seemed to be more important than the correlation. These findings allowed nurses with care ful preparation to provide positive care to patient in BCR.