JNIOSH

Abstract of Special Research Report (RR-99)

National Institute of Occupational Safety and Health, Japan

Survey on Improving Occupational Environments in the Rapidly Aging Society Part 2 --Conditions of Care Work in Xursing Homes and the Prospects for Elderly Care Workers--

RR-99-1
Hisao NAGATA and Sunyoung LEE

: Physically handicapped, dementia and bedridden elderly people are estimated to account for 2.8 million in 2000 and then 5.2 million in 2025. In the rapidly aging society, it is necessary to facilitate the participation of elderly people in activities of our society by integrating them with other generations. In order to survey the conditions of care work in nursing homes for elderly people and to examine the possibilities of future elderly care work, mail questionnaires were sent to 2000 nursing homes in March, 1999. Managers of the care workers in 969 nursing homes completed questionnaires. The response rate was 48.5%.
    The questions concerned existing facilities, feature of users, number of staffs, a shift work system, breaks, arrangements of working space in bathing room, dining room etc., equipment required to decrease work loads, physically or mentally heavy tasks in nursing care, physical conditions of elderly care workers, maximum possible age for care workers, and characteristic of elderly care workers etc.
  (1) Work loads were heavily burdened in various aspects of nursing care work. Night shifts were worked at the average rate of 3.6 times a month with working hour extending to 16-17. Only 43.2 percent of the care workers could take a nap of 1-3 hours during the scheduled working time.
  (2) Physically heavy work loads in nursing care were mostly due to bathing care, the transfer of elderly people from beds to wheel chair. Peak loads were associated with manual handling loads in lifting, holding or carrying elderly people requiring nursing care.
  (3) In the current conditions of elderly care services in nursing homes, it seemed very difficult for care workers of age above 60 to participate due to poorly arranged working environments, mainly because of the lack of support measures for heavy manual lifting in bathing care and bed-to-wheel chair transfer.
  (4) As suggested from personal opinion freely written in the questionnaires, aged care workers were estimated to be very suitable for mental care work based on their good understanding of the needs of elderly people requiring nursing care.

Ignition Characteristics of Zirconium Dust

RR-99-2
Toei MATSUDA and Masaaki YASHIMA

: Zirconium metal dust is pyrophoric, highly combustible in its dry state, and has a great affinity for oxygen at elevated temperatures. The wide use of zirconium has been found in industry and the most important use of the material is in nuclear reactors for cladding fuel rods. It has been considered that fire and explosion risks of zircaloy (98 wt.% Zr) may be expected when cutting nuclear fuel rods in a reprocessing plant.
    Zirconium sample dusts, prepared by dehydriding process from hydrogen-absorbed metal zirconium, have been used to examine some ignition sensitivity properties.
    A thermo-gravitational analysis was made in air with variation of heating rates. The sample dusts showed S-character type weight increasing curves to the final oxide product, but when the heating rate was mostly higher, a small ignition was observed in the oxidation process. The ignition seems to be caused by oxidation of the upper layer of the dust in an open cell, following that of the lower part of the dust. Accordingly, such a thermal analysis using micro-quantity of sample dust in the capsule-type cell wouldn't be appropriate to evaluate actual ignition hazards of pyrophoric metal dusts. For layered dust, macroscopic ignition temperatures were then measured on a hot plate with a variation of depth of layer. The data have been interpreted with a thermal theory.
    It is found in the proceeding literature that the ignition of a dust cloud of zirconium is capable to occur at a room temperature. This fact gives us a question whether a quiescent dust cloud of zirconium will be ignited spontaneously at normal temperature. Using a Godbert-Greenwald furnace apparatus, ignition temperatures for dust clouds of zirconium were measured in air for water-contained sample dusts and for mixtures with zirconium oxide dust, respectively. Effect of oxygen contents on the ignition temperature was also studied in nitrogen-diluted atmospheres. All these suppressing factors of water, the inactive dust (zirconium oxide) and nitrogen abruptly raise the ignition temperatures from room temperature to higher temperatures. From these findings, it was estimated that there are two ignition modes involved in the ignition of zirconium dust clouds. After successful ignition tests of a dust heap of small quantity by a blast of compressed air, it is concluded that ignition of zirconium cloud at room temperatures is caused by friction in the furnace apparatus and that thermal ignition of the dust cloud could occur at a temperature of 300 to 400°C in air. As far as the furnace apparatus is used for the measurement of ignition temperature for a dust cloud, the ignition temperature could be found to be a room temperature. Dry zirconium dust is very sensitive to a spark, for ignition energies both for a cloud and a layer are too low, in accordance with previous values. Although the dust has a comparatively low ignition temperature, it will not ignite spontaneously at a room temperature for a form of dust clouds.

Ignition and Explosion Characteristics of Tantalum Dust

RR-99-3
Toei MATSUDA and Mizuki YAMAGUMA

: Elemental tantalum powder is mainly used for a material of a compact high performance capacitor called the tantalum electrolytic capacitor. As the miniaturization of electronic devices, such as cellular phones, has progressed remarkably, the demand for tantalum has rapidly increased in recent years. In the spring of 1997 a manufacturer of tantalum powder experienced a tantalum dust deflagration in its bag filter dust-collecting device which resulted in a fatal accident, in which one worker was killed and the other seriously wounded.
    The fire and explosion hazard properties of light metal powders such as aluminum and magnesium are well documented, but information concerning the flammability hazards of heavy metals in the finely divided forms is limited. Then, tantalum ignition and explosibility data have been obtained by using modern testing apparatuses.
    Dust explosibility data were obtained in a 20-L spherical explosion chamber. The data showed that the magnitude of the tantalum dust explosion is classified as severe (normalized rate of pressure rise : Kst = 273), contrary to the classification found in the preceding literature. In the explosion tests, it was found that the sample dust of larger particle sizes was crushed into smaller sizes during injection from the high-pressure air dust container. Accordingly, these test methods for dust explosibility are inappropriate to evaluate the dust samples, which are susceptible to fragmentation or brittlement.
    The minimum ignition energies both for a dust cloud and a dust layer of the tantalum powder were also far lower than literature values. In the layer form, especially, the electrostatic discharge channel was formed through the layer, and a small heat spot was created on its surface when it was ignited. The spot spread gradually in the beginning, then a bright flash was formed several seconds after the ignition. It is thought that the direct heat-up of a part of the heap by the electric current passing through the layer is the main cause of tantalum powder's sensitivity to an electrostatic discharge. A thin, nonconductive oxide layer of the tantalum powder surface has high resistivity and generates electrostatic charge when rubbed with conductive materials like the wall of the collecting device.
    Judging from the observation of the surface with an SEM, the coral-like structure of each particle of the tantalum powder can enhance its fire and explosion hazards and affect its sensitivity to electrostatic sparks by increasing in particle surface area. The authors conclude that the possible cause of the ignition was electrostatic discharge resulting from charging electrostatically.

Analysis and Experimental Study on Labor Accidents Related to Communication in Construction Work

RR-99-4
Yoshiyuki EGAWA, Takahiro NAKAMURA, Takuro SHOJI, Kiyoshi FUKAYA, Shigeo HANAYASU and Yoshimi SUZUKI

: The causes of accidents in construction work were investigated based on the labor accident reports. As the result, about 10% of them were responsible for communication errors between workers. By the analysis on the accidents caused by communication errors, it was cleared that they were classified into six models having three factors of time, place and purpose. From the examination of these three factors in the cooperation work in which two workers were injured or nearly injured 80% of accidents were occupied in three models of six ones. One model of them was experimentally verified. This model was that purpose was different and both time and place were same. In the experiment, an assembly work of some parts was performed by two teams which consisted of two members, respectively. During the work, "error" was defined to occur when the team members entered one of the restricted areas at the same time. As the result of the experimental data analysis, four error patterns were observed. The number of errors of one pattern was more than that of the others. This pattern was that a member entered the restricted area before the other left. Communication was effective when two members' position was face to face each other. However, communication was ineffective when their position was back to face. A lot of errors occurred at this position of two members.

The Effect of Heat Conduction on the Stress Measurement of the Sheet Specimen by Means of Thermoelastic Effect

RR-99-5
Takashi HONDA, Tetsuya SASAKI and Teruhito OHTSUKA

: The local stresses and strains near notches or cracks are very important for fracture analyses of cyclically loaded structure components. Thus, a number of experimental methods such as a photo-elasticity method and numerical methods such as a finite element analysis have been developed to evaluate stresses and strains near notches or cracks. Among those methods, an infrared stress analysis technique is relatively a new stress analysis technique based on the measurement of infrared radiation emitted from the surface of a body. However, when this technique was used to evaluate the stress concentration factors for several kinds of notches in sheet specimens and to estimate stress intensity factors for compact tension and center cracked tension specimens, the determined values tended to be somewhat lower than those of other comparable numerical or theoretical values, and the error increased in proportion to those values. Thus, in this study, the effect of heat conduction in a body of sheet specimens under cyclic loading was investigated by a finite differential analysis to examine the source of the error, and the results were compared with experimental results.
    The material of the specimens used in this study is JIS S 45 C steel. Three types of center hole sheet specimens and CCT specimens were prepared. Those are 3 mm thickness machined from a 3.3 mm thick sheet material. The proportion of diameter to width is constant for center hole sheet specimens in order to have the same stress concentration factors. The unsteady heat conduction analysis under cyclic loading was carried out by a finite differential method (FDM). Prior to the unsteady heat conduction analysis, the stresses of specimens were analyzed by the finite element method (FEM) under the plane stress condition using a two-dimensional mesh. Considering the symmetrical configuration, one fourth of the hole specimen was analyzed in order to determine the temperature range of each node.
    The stress measurements by thermoelastic technique were carried out using an infrared stress analysis system developed by JEOL. Each specimen was cyclically loaded at different levels of frequency of 1, 5, 10 and 15 Hz using a servo-hydraulic fatigue machine of 196 kN capacity.
    The comparison between numerical calculations and experiments showed that there were three sources of error : (1) inability to achieve the adiabatic condition due to heat conduction around the notch root and the crack tip, (2) the edge effect of an infrared camera and (3) the resolution limit of an infrared camera.

Influence of SO2 Gas on Insulation Failure of Wiring Board Caused by Electrochemical Migration

RR-99-6
Tatsuo MOTOYAMA and Kenji ICHIKAWA

: With progress of science and technology, highly compacted electronic equipments with high density wiring having multi-functions and high quality have been developed. And according to the change of the using environment, electronic equipments have become easy to receive the adverse effect from environments involving dew condensation or pollutant. SOx in the environment is increasing due to automotive exhaust gases and acid rain, etc., while SOx that discharged from factories is decreasing. Increase of SOx is one of the main causes of insulation troubles of electronic equipments which may affect the control system in systematized facility. Consequently this insulation trouble causes the malfunction of the equipment, and may cause the danger to workers. The purpose of this study is to prevent insulation trouble caused by the electrochemical migration under the SO2 gas environment. Experiments are carried out to clarify 1) the effect of SO2 gas on the generation of migration, and 2) the generation mechanism of migration under the SO2 gas environment.
   The experiments were carried out using the circuit board set in the cycle testing tank controlling temperature and humidity. The voltage applied to the test wirings was 50 V. SO2 gas concentrations were 0, 0.3, 12 and 85 ppm. Distances between wirings were 0.16, 0.4, 1, 2.5 and 6.3 mm. The one cycle of the cycle testing was first (5°C, 60%RH, 20 min.) and second (25°C, 90%RH, 20 min.). The results are as follows :
  1) SO2 gas accelerates the progress of the migration and causes the migration with wirings of 6.3 mm distance.
  2) The cause of the affection of SO2 gas is attributed to the dissolution in the water adsorbed on the circuit board.
  3) For preventing the generation of migration under the SO2 gas environment, prevention of the adhesion of the water is important.

Thermomechanical and Isothermal Fatigue Behaivor of Stainless Steel Type 316 Weldments

RR-99-7
Etsuji YOSHIHISA, Takashi HONDA and S.G.S.RAMAN

: In any installation, welded joints are of major concern as they are frequently the sites of localized damage. Type 316 stainless steels and their welds are commonly used for structural components designed to operate at elevated temperatures. Under these operating conditions, component life may be limited by various damage mechanisms such as fatigue, creep and oxidation. When stainless steel welds are exposed to elevated temperature, the δ-ferrite, which is introduced to reduce hot cracking and microfissuring, gets transformed to carbides and various intermetallic phases like brittle σ phase. This transformation shows to decrease ductility when high stress are applied at elevated temperature. In isothermal fatigue, an extensive transformation of δ-ferrite to σ phase was reported to be responsible for reduction in fatigue life. It was observed with a number of materials that the most conservative fatigue life may not be under isothermal fatigue test condition but thermomechanical fatigue (TMF) where combined cycles of both temperature and mechanical strain are applied. Though there are several studies on the isothermal fatigue of 316 stainless steel weldments, to the authors' knowledge there is not any work reported on the TMF behavior. In this study the TMF (inphase) behavior of type 316 stainless steel weldments with temperature range 573-973K was studied using cylindrical specimens machined from base metal, weld metal and weld joint and a comparison was made with isothermal fatigue behavior at 973K. The progress of δ-ferrite transformation in weld metal was examined at different levels of life and the effects on the fatigue behavior were also discussed. In all joint specimens, fatigue failure occurred in the weld metal region. The lives of the weld metal and joint specimens were nearly equal and were always inferior to those of base metal specimens. In base metal specimens, the effect of strain rate on isothermal fatigue life was not very much significant. Though TMF lives were always a little shorter than the isothermal fatigue lives in base metal, the difference was small at the same mechanical strain range and similar strain rate. This may be due to that fracture mode in both loading was similar-a mixed type. On the other hand, a drastic reduction in life was noticed in weld metal and joint specimens under TMF in comparison with isothermal fatigue. This was attributed to the additional damage due to many independent subsurface cracks at σ phase boundaries and linkage of these cracks with the surface crack leading to rapid crack propagation, δ-ferrite in weld metal rapidly transformed to σ phase in both isothermal fatigue and TMF tests. Difference of the stress amplitudes between both tests was considered to have caused the difference of fatigue lives.

Survey on Organizational Safety Policy and Activity in Construction Industry

RR-99-8
Takuro SHOJI, Yoshimi SUZUKI, Takahiro NAKAMURA, Yoshiyuki EGAWA, Kiyoshi FUKAYA, Shigeo HANAYASU, Mitsuhiro KOJIMA, Ayako HIROSE, Naoko HASEGAWA and Ken-ichi TAKANO

: Recently much attention has been paid to the human factor approach considering worker's psychological state and organizational factor, as well as development of new technology or improvement of working environment, in order to decrease labor accident rate further. The purpose of this study is to clarify the effect of organizational safety policy and activities on safety attitude and behavior of workers, by making use of information through questionnaire on the safety activities and systems of the several companies and by analyzing the relationship between these safety activities and safety attitude of workers. Self-rating questionnaire was developed for this study and delivered to the staffs of the safety and health division of 49 companies, 300 personnel who were working at construction sites in general contractors and 300 foremen of subordinate companies. The response rate of questionnaire obtained was about 82 %.
    The main results of this study were as follows :
  1) There found significant relationships among safety systems, activities, organizational policy and safety attitude of workers. Vigorous safety activities at work site might be supported by completion of organizational management in safety division and, furthermore, they would bring up safety attitude and behavior of workers.
  2) Close communication was taken among staffs of the safety and health division within the company ; on the other hand sometimes they had troubles in contacting to the workers of other sections of the company. Safety division staffs sometimes met difficulty in getting necessary information from work sites and telling their policy to other sections in the company.
  3) It was pointed out that much attention should be paid toward the organizational activities rather than individual worker/staff activities to prevent accident efficiently.

Spectroscopic Measurements of a Spark Ignition in Methane-Oxygen Mixture

RR-99-9
Atsushi OHSAWA and Keiichi ISHIKAWA

: Spark ignition process was investigated by using spectroscopic techniques. The stoichimetric methane and oxygen mixture (33.3 kPa) was used. Optical emission spectroscopy was used to determine the species produced by discharge and combustion. Time variations of their optical emission spectra and timeresolved imaging of light emission were used for investigating the spark ignition. To measure the spatiotemporal evolution of OH radicals, 2 D-laser induced fluorescence (LIF) method was used.
    Ignition energy was 5.1 mJ and the transfer duration of it was 0.7μs, when the voltage of 8 kV and the capacitor of 500 pF were used. The distribution of the optical emission spectra showed that various species were produced by the discharge and the combustion. The main detected species were C2, CO, CO2, CH, CH3, CHO, CH2O, H, H2, H2O, O, O3, OH. The waveforms of the optical emissions from them consisted of three phases which corresponded to the discharge until --2μs, the induction period of the ignition and the explosion after --1μs. It seems that the radical species are produced by dissociation collisions with electrons in the discharge period and also produced by thermal reactions in the explosion period. The production of CO, CO2 and H2O implies that oxidation process occurs even in the discharge period. Then, the production of an initial flame kernel could be expected at the end of the discharge, because thermalization by electron collisions should be reached. Although the light emissions from the species between the peaks could not be detected by a photo-multiplier, a gate-intensified CCD camera could detect them. The profile of the image of the initial flame after the discharge did not change until --50μs. Moreover the light intensity of it gradually decreased and the images from 80 to 90μs almost disappeared. After the disappearance, the flame kernel suddenly propagated after --100μs. Therefore it was found that the ignition occurred between 90 and 100μs. The velocity of the flame propagation was approximately 20 m/s. The fluorescence signal from OH radicals could be detected after 30μs. The 2 D profile of OH radicals gradually expanded until the ignition and it also extremely propagated after the ignition. Furthermore it was found that OH concentration increased simultaneously at the ignition, because the fluorescence signal increased after 90μs. It was expected that the additional heating by exothermic reactions involving de-excitation collisions could propagate the initial flame.

Hydrolytic Rate of Acid Anhydrides in Heterogeneous Reaction

RR-99-10
Yasuhiro FUJIMOTO

: The Reaction Calorimeter RC 1 has been used to evaluate reaction hazards in laboratories under the batch process conditions-reaction temperature, type of reaction, etc. For the better result of evaluation, the conditions for evaluation are usually attempted to be the same as ones in actual chemical plants as well as possible. But all of the reaction conditions in actual plants may not be safe and available for laboratory evaluation. Some reactions may bring explosions or burnings easily, and may be too toxic to protect normally. In such cases, less hazardous model reactions are useful if their physical and chemical characteristics are roughly similar to the actual reactions. Results using the model need not to show the actual magnitude of reaction hazard, but need to show the trend of the hazard, that is, the evaluation result using one model reaction will be useful for a variety of reactions.
    Neutralization of acetic acid and sodium hydroxide was used as a model reaction for the evaluation of the effect of mode and rate of stirring on the thermal behaviors of exothermic chemical reaction-nitration. The neutralization is rather fast reaction (reaction rate constant of AcOH in benzene and NaOH in water : k == 58 x 10-3 s -1 at 300 K), so it seems suitable for a model of fast reactions, nitration etc.
    In this paper, the rate of hydrolysis of acid anhydrides was evaluated as a candidate of the model reaction with moderate rate. The hydrolytic rates of three anhydrides-acetic anhydride (C4H6O3), propionic anhydride (C6H10O3) and butyric anhydride (C8H14O3) were measured. The hydrlytic rates were k = 2.8 x 10-3 s -1 for hydrolytic rate of acetic anhydride at 300 K and k = 0.54 x 10-3 s -1 for propionic anhydride at 300 K. The rate for butyric anhydride was too small to measure. The hydrolytic rates for diluted anhydrides with benzene were k = 1.3 x 10-3 s -1 for acetic anhydride at 300 K and k = 0.54 x 10-3 s -1 for propionic anhydride at 300 K. The experiments show that the hydrolysis of acetic anhydride can be useful for the model with moderate rate reaction.
    The re-start experiment of stirring for acetic anhydride of benzene solution was conducted. The operating condition of stirring rate was 40 rpm while dosing acetic anhydride of benzene solution into water, then increased to 180 rpm. The reaction rate after increasing stirring rate was k = 3.2 x 10-3 s -1 at 320 K. It is about 50 % higher than the rate in the other experiment in which the stirring rate is 180 rpm throughout dosing and aging. This result is a little different from the similar experiment with neutralization. The rate of re-start was about ten times higher than the normal rate in the neutralization experiment. This shows the severity of the re-start hazard is sensitive to the reaction rate.

Decomposition Explosion Properties of Ethylene Oxide under High Temperature and Pressure

RR-99-11
Takaaki MIZUTANI and Hidenori MATSUI

: Ethylene oxide (EO) is one of the important materials in chemical industries. There are many literatures on decomposition explosion properties of EO. However the knowledge of decomposition explosion limits and ignition energies under high temperature and high pressure have not been well known.
    In this study, stainless steel cylinder (5 cm inner diameter, 13 cm in height, 200 cm3 in volume) was used for a explosion vessel. The temperature inside the vessel was automatically well controlled by dipping the vessel into an oil bath. Concentration limit of EO diluted with nitrogen for decomposition explosion under high temperature up to 180°C and pressure up to 0.9 MPa was determined using a high voltage electric spark igniter (16 kV, 50 mA, 0.1 s). In higher initial pressure (over 0.5 MPa), pencil lead guided capacitance spark electrode was used for ignition. Minimum ignition energy (MIE) for decomposition explosion was determined using lead guided capacitance spark as well. The energy was evaluated based on the 0.5 CV².
    The experimental results are as follows ;
  (1) Decomposition pressure limit for pure EO was 15 kPa at 180°C.
  (2) No ignition occurred at any temperature and pressure in this experimental conditions, when EO were diluted into less than 40 vol.%.
  (3) Higher temperature and lower nitrogen dilution showed lower minimum ignition pressure limit.
  (4) The maximum explosion pressure was four to five times as high as initial pressure, and the pressure rising time was 0.4 to 1.5 s under the experimental conditions near the limit.
  (5) MIE of EO-nitrogen mixtures steeply increased with the nitrogen dilution.
  (6) MIE of pure EO showed the linear relation with the inverse of both initial pressure and temperature.
  (7) Estimated MIEs of pure EO at 150°C were about 0.4J and 0.1J at initial pressure of 0.1 and 1.0 MPa, respectively.
    These results suggest that usual static electricity sparks can not ignite even pure EO and that no explosion occur when EO is diluted with nitrogen less than 40 vol.% at this experimental temperature and pressure conditions.

Failure Mechanism of Tie-Rod Anchored Sheet Pile Wall in Centrifuge Tests

RR-99-12
Yasuo TOYOSAWA, Noriyuki HORII, Satoshi TAMATE and H.G.B.ALLERSMA

: Accidents due to collapse sometimes occur in excavation site. The need for decreasing these accidents is currently one of major concern. It is difficult to find out the causes of accident, because of the disturbance of site during the rescue and the collapse itself. To find out the scenario of an accident is very important to retrieve the information involved.
    In this paper, firstly, a case, history is described of a temporary earth support collapse. Secondly, the basic attempt for this approach by using centrifuge would be illustrated. The combination of both site investigation and centrifuge model tests would give more useful information to approach the scenario of accidents.
    Thanks to the low weight of the models (they can be conducted by one person) several configurations could be simulated in a short period. Six models, which consisted of different profiles, were tested to observe the tie-rod anchored sheet pile wall's collapse in centrifuge.
    The tests covered both sheet pile walls anchored in the soil and sheet pile walls supported by struts.
Performing the tests at 100 g simulated a prototype depth of approximately 6 m. To simulate a field accident, struts were removed in flight. The modes of failure were observed in the different model tests by means of an on-board video camera. It was possible, even in small models, to simulate complicated configurations, and detailed information could be obtained by using image-processing techniques. Based on the results of these centrifuge model tests, the deformation and failure mechanisms are discussed.
    The following conclusions are drawn :
  (1) A case history of the temporary earth support collapse shows that the buckling of the struts induced the sequence of collapse.
  (2) It appeared that the observed failure modes in the centrifuge tests show a good agreement with field observations.
  (3) Piles that are blown in the bottom of the excavation have a significant influence on the failure mechanism.
  (4) It has been shown that the anchoring of sheet pile walls does not seem to be very effective in several cases. Cracks are initiated from the anchoring wall of the sheet pile walls, which led to unstable behavior of the retaining wall.
  (5) Centrifuge model tests have proven to be helpful in obtaining a good understanding of the failure mechanism in the case history. For further information, water and three-dimensional effects, etc. should be considered.

Other Publications

Kiyose District Map

Umezono 1-4-6, Kiyose,
Tokyo 204-0024 Japan
TEL: +81-42-491-4512
FAX: +81-42-491-7846

Noborito District Map

Nagao 6-21-1, Tama-Ku,
Kawasaki 214-8585 Japan
TEL: +81-44-865-6111
FAX: +81-44-865-6124