Document r6peOznmnXJr9DqbjoKnQ5nnE

FILE NAME: Shipbuilders Council of America (SHIP) DATE: 1972 May DOC#: SHIP048 DOCUMENT DESCRIPTION: Conference Presentation by NIOSH - Dust Exposure in Asbestos Processing MD-006371 ^ ~t& <* & C J j t ( & ) t - 1 9 1 1 c/ l <c t * ECTQStn?-g Hi AS53STQ5 E80CESSIWS Eo-hert B. Waidner and Howard E- Ayer Kartiaiial in s titu te f o r Occupational-Safety and Health &'& Department of Health, Education, and Welfare C inciriitati, Ohio 45202 IKIRQBTJOTION Asbestos is a generic term applied to a number of fibrous crystalline siiliaate minerals. Although chemically they are grouped as hydrated allicatcs, they vary greatly* in degree of hydration., complex coiapoaxticox of the silicate, and physical properties of the crystalline ftilj.ccaui. structure.* Deference.-was. made to asbestos as early as several centuries B.C. l b was not established as a mineral of industrial importance, how ever,; until around the end of the last century. Asbestos has a will variety of uses in m o d e m industrial technology because of its aa'!rn;andin:4 properties af chemical resistance, fibi'ous structure and ineombuetibility. It is used in the manufacture of textiles, felts, rapes, brake linings, clutch facings, roofing and flooring materials, asbestos-cement products, gaskets, insultation materials, srpsrci&l papers-, filters, and plastic products. These are used in pm-o.hic&lly every major industry. Uifrtirtics from the Bureau of Mines "Mineral Pacts and Problems, Bulletin 650, I97CV* indicate that the United States is the woxid *3 largest, consumer of asbestos, using more than one-fifth of the 3.5 million short tens of world production. Most of the asbestos pro duction is chiysotile asbestos, but approximately 200,000 tons of the. world's production each year consists of aaosite sand crocidolite. lir. 1963 the United States consumed about 235 of the total world pro duction. Abbestos workers, have been the subject of numerous studies and re view articles since asbestosis was first recognized as an occupa tional disease in the late 1920*s; however, many questions remained \uiEhcwsred concerning the clinical aspects of the disease and its efaciogy. Information on the prevalence and clinical characteristics af asbestosis had been derived primarily from medical studies of actively employed asbestos workers. These were supplemented by s-pacial clinical studies and autopsy data. little definitive data regarding the nature and magnitude of concurrent environmental ex posures had been obtained. Even though an estimated 3*5 million workers are exposed annually to asbestos, this may represent only a part of a much larger emerg ing base resulting from advancements in industrial technology and affecting a. far greater population, the rapidly expanding field of industrial fiber technology. There is evidence that respirable fi bers, whether they are of natural mineral, vegetable, or synthetic origins- may behave qrzite differently in the lunge than particulates af the same chemical composition. 104- A comprehensive study was started in order to obtain data which could be utilized to accomplish the following objectives: 1. To determine the health status of the asbestos workers in this industry with special attention to diseases of the respiratory tract. 2. To determine the relationship between occupational exposure and cause of death of asbestos workers. 3. To develop medical and environmental criteria and procedures for the control of health risks identified by the study, 4. To determine which environmental factors have an adverse effect upon the health of asbestos workers in the asbestos--products industry. The National Institute for Occupational Safety and Health (NIOSH) (formerly Division of Occupational Health) in the U.5. Public Health Service has been conducting an epidemiological study of the asbestos processing industry since 1964. This paper will confine itself to the environmental phase of this study which was designed to define the working environment of the various plants. Particular emphasis was placed on levels and nature of exposure to asbestos dust. PRESENT 'EPIDEMIOLOGIC STUDY The present epidemiologic study of the asbestos products industry was started for the following reasons: 1. The substantial changes in the size and technology^of the indus try have created new exposure situations in need of evaluation. 2. Over-all reductions in asbestos dust levels now make it possible to examine the relation of asbestos to asbestosis in workers with long-term lower level exposures. 3. The relationship between types and magnitudes of exposures and the occurrence of lung cancer needs to be defined. SAMPLING- METHODS The types of samples collected in the environmental phase can be divided into four categories: * 1. Bulk samples - Several pounds of each of the different'types of crude fiber, bag house waste, and treating and lubricating oils used were collected at each plant. - 2 . Long period gravimetric samples - Several instruments for obtain ing mass samples of total and respirable dust were run simul taneously in the representative area in the midst of each opera tion. These instruments included a high volume sampler, a Hex-- hiet modified to take a membrane filter, and a set of four mem brane filter field monitors with and without 10 millimeter cyclone presampl.ers, and a membrane filter field monitor with a horizontal aVatri?tor v.ps bream. 3 Simultaneous impinger-filter samples - Breathing zone samples taken with an impinger and a membrane filter simultaneously were used to obtain a comparison of these methods of dust . enumeration# This method of sampling was used because the in itial exposure limits were based on samples evaluated with the impinger and it was hoped that a correlation existed between the methods so that a meaningful use could be made of all the past data collected. Other sets of simultaneous impinger- filter samples, with and without horizontal elutriator pre samplers, were taken in general room air near specific opera tions. 4. .. Personal Samplers - Only the personal samples collected using membrane filters will be discussed because it gives a better index of airborne fiber concentration. Lynch, et all found that the impinger method of sampling v;as very inefficient with relation to capturing fibers to be counted later. The lack of fibers in the impinger samples was not due to the absence of fibers in the air. The impinger, an impaction device, has a collection efficiency related to the aerodynamic size of par ticles and is not efficient for particles with falling speeds less than that of one micrometer, unit density spheres. Fur ther, the 10X objective, light field, counting technique used does not resolve particles much smaller than one micrometer (pm). Laboratory experiments showed that the impinger passed fibers and from field sampling it is estimated that only about one out ox one hundred fibers in the air as seen in electron micro graphs was seen in the impinger samples. Since it is important that any method used be able to measure that factor in the environment that is most relevant to the diseasecausing mechanism and will, therefore, yield a significant correla tion between health and exposure, it was determined that the mem brane filter method met this criteria. Fibers have long been im plicated as the causative agent in asbestosis and may be signifi cant in cancer. Timbre1 1 * work showed that because of the peculiar aerodynamic properties of fibers (i.e., that their falling speed is dependent on diameter only when aspect ratios are greater than 10) it is possible for much larger objects (longer and heavier) to penetrate deep into the lung in the shape of fibers than in the shape of grains. This appears to be a most significant biologic property considering that the parent minerals involved (amphibole and serpentine) are not considered biologically active. Based on methods developed in Great Britain^, a method of collecting and counting fibers on membrane filters was developed^. These fil ters have pore sizes of 0.8 pm but are almost 100^ efficient down to several hundredths of a micrometer because of surface effects. They are dissolved by means of a mounting medium which is prepared by dissolving 0.05 g of membrane filter per ml of 1 to 1 mixture of diethyl oxalate and dimethyl phthalate and has an index of re fraction (ND) equal to 1.47. The samples are counted with a 4 m i objective (430X) under phase contrast illumination. By examining a suitable number of fields, it is possible to count enough fibers to make a statistically useful estimate of fibrous dust concentra tion. In order to collect a sample representative of airborne dust -106- which is likely to enter the subject's respiratory system, it is necessary to position a collection apparatus near the nose and mouth of the subject (i.e., in his "breathing zone"). An exploded view of the sampler is shown in Figure 1. The sample is collected on a 37 TM millipore type AA filter mounted in an open--face field monitor. The monitor is fastened to the worker's lapel and air is drawn through the filter by means of the battery powered personal sampler pump, similar to those approved by NI05H under the provisions of 30 CFR 74. A support pad is placed between the field monitor and the membrane filter which aid3 in controlling the distribution of air through the filter. Flow rate of the personal sampler pump waa maintained at 1,7 liters per minute (1pm). Sample time varied from 15 minutes up to 4 hours. The majority of the samples were collected between 1 1/2 to 2 1/2 hours. Actual sampling time depended upon many variables (e.g., concentration of fibers, types of operation, amount of background material present, etc.). GENERAL CHARACTERISTICS OF PARTICULATES IN ASBESTOS PLANTS The sources of airborne particulates in an asbestos processing plant may be classified as follows: ` 1. Fibrous asbestos 2. Asbestos type minerals (serpentine, antigorite, etc.) 3. Non-asbestos type mineraLs occurring with asbestos (magnetite, periodotile, chromite, nickel, quartz). 4. Other process materials (cotton, rayon, glass fiber, etc.) 5. Non-process materials (air pollution). An air sample from an asbestos plant taken by any efficient sampling device and examined microscopically reveals that most particles do not appear fibrous. They resemble, in fact, the particles common in air samples from virtually any operation producing mineral dust. If such a sample is magnified several thousand diameters by electron microscopy many particles less than 0.5 micrometers (jam) in diameter are also visible. These small "air pollution" particles occur in any air sample, indoor or outdoor, and the number varies depending on time, season, community heating fuel, location of plant air inlets relative to nearby smokestacks, and general atmospheric conditions. Air pollution particles bear little if any relationship to the dust generated from asbestos processing. * However, many fibers including rayon and cotton which are identifi able by their large diameters are also observed in such an air sample. The presence of asbestos fibers is the unique characteristic of the sample. Figure 2 is a typical electron micrograph of dust from as bestos textile processing. Figure 3 is an electron micrograph at a higher magnification of extremely clean bulk asbestos treated to open the fibers. Both Breesen^ and Fulton^ reported that the usual diameter of airboms asber>tf>o fibers was less than 0.5?mi, Electron micrc-rvnphv ira z 0,,L.pm di.arae.ter. I;f a fiber is defined. 31$ having an aspect ratio greater than- three to-one, most-of the fibers, by number, are shortj- median, lengths fr-om electron micrographs are between 1 and (LUO*, ......... ....... CURRENT STAICDARDS ' Asbestos has- been listed- as one of-- the five target health hazards by,- the= Icaupational Safety -and Health Administration of the U.S. Department; of' labor* As of December 7, 1971* an emergency stan-- 2hrd_f:0.r asbestos was invoked by the" federal government under the auspices of. the Occupational Safety;and' Health Act. The standard state-a-:. ' --- . ...... '*Thft 8-hpur time-weighted average airborne concentration erf a@bes.tos dust to which employees are exposed shall not exceed, fiv.e. fibers per-ndlliliter greater than five mi crons in length, as-'determined by the membrane filter method at 400-450X magnification '(4 millimeter objective) phase contrast illumination. Concentration above five fibers per milliliter.but, not to exceed'ten fibers per milliliter, may- be permitted up to a total of 15 minutes in an hour for- up. to,, five hours, in an B^hour day." The National Institute for Occupational Safety and Health (NIOSH) in its ''Criteria for a Recommended "Standard.. ...Occupational Exposure to Asbestos," recommendsi " ^Occupational exposure to airborne asbestos dust shall be controlled so that no workers shall be exposed to ore than 2,0 asbestos fibers per cubic centimeter (cc) of air based on a count of fibers greater than five micrometers (>5 urn) in length Cdetermined by the mem brane filter method at 400-450X magnification (4 milli meter objective) phase contrast illumination, ....3, determined as a timeH/veighted average (TWA) exposure for an 8-rOiour work day, and no peak concentration of asbestos to which workers' are exposed shall exceed 10.0 fiperp/cc" greater than 5 pm as determined by a minimum sampling time of 15 minutes.*' RESULTS OF ENVIRONMENTAL SAMPLING Although many different types c samples have been collected only the results of the membrane filter method will be discussed because this is the method enumerated in the existing 'Federal Standard on asbes tos . All of the data listed are the values of fibers greater than 5 micrometers -in length. All concentrations are listed as the num ber of fibers greater than 5 micrometers in length per cubic centi meter of air sampled (hereinafter, as fibers/cc). The results listed are based on approximately 10,000 samples that were collected and counted. ~ " ' "" liable 1 lists the arithmetic means of -the over-all asbestos concen trations by types of plants for those that were in the cohort. In some cases there were three surveys .conducted, for example, in the -textile and friction plants. A minimum of two surveys was conducted f a.n each plant. The -cycle on the resurvoya was approximately every -108- 2 l/2 years. This table provides an overview of the various types of plants and allows only general conclusions to be made. It can be seen that there is a definite difference in exposure levels among the various types of plants. On the surface it would appear that the insulation plants have the highest exposures, followed by the textile, friction, construction materials and cement pipe plants. The values for the insulation plants are high because two of the plants had very high values as shown in Table No. 2. This table shows the asbestos concentrations by various operations within the insulation plants. The plants are arranged according to the over all individual plant average for all surveys conducted there. The wide variation between the highest plants and lowest plants can be attributed to the amount of controls that each plant has* It is evident that regardless of the over-all plant average the mixing operations have the highest exposure of all operations in all the plants and the inspection and packing operations the lowest. If these values are compared with the emergency standard of 5.0 fibers/ cc (TWA) and 10.0 fibers/cc (peak), it can be seen that all means for all operations at the two highest plants not only exceed the TWA value but the peak value as well. Whereas the plant average for the lowest plant for all operations was equal to or less than the proposed standard of 2.0 fibers/cc. Listed in Table No. 3 are the asbestos concentrations for the insu lation plants by operations broken down according to the over-all means by survey conducted. The over-all values at each operation are high as a result of the very high values from two of the plants. There was very little difference between the results of the two surveys conducted except at the mixing operations. All other op erations had similar values for both surveys. The mixing operations had the highest exposures. Table No. 4 lists the textile plants by individual plants according to their over-all plant average for all surveys conducted. It ap pears that a plant which has a high over-all average will generally have a high average for each of the various operations. That is not to say that the highest averaged plant over-all will have the highest average for each and every operation, but it i3 true most of the time. If their values are compared with the emergency standard of 5 fibers/cc, then neither of the two highest averaged plants have any operations where the mean value is below it. The second lowest--averaged plant has two operations which fall below the emer gency standard. Whereas the lowest-averaged plant has virtually every operation below the emergency standard. The various asbestos concentrations by operations within the textile plants for each survey conducted are shown in Table No. 5 The fi ber preparation, carding, spinning and twisting operations all ap pear to be problem areas based on the present emergency standard. Only the winding and weaving operations appear to be near the stan dard. The average concentrations for the friction plants are listed by the over-all individual plant average in Table-No. 6. When considering the mean values of the various operations with respect to t-^ o r gency s tandard of five fibors/cc, it is appar-nt plant hasn't any operations whieh &re below the ,standard, She seerond highest averaged plant hay foyir operations wh-igh are just" ~' slightly above the emergency standard' and two operations jiyhiph gre below it. The two lowest avorgged plants have mgan y a lues for all the operations below the emergency standard, '//hen ppaiparing these two plants using the proposed standard of two fibers per pc", it can be seen that the second lowest pliai-t has-pwp operations which have mean values below and two operations which gre"reasonably " close to the proposed standard, whereas, the lowest averageA~'plant has four operations which have mean values below -the proposed standard and two of which are reasonably clp.se -to it*' ' Table No. 7 has the arithmetic means by .operations for all -the friction plants surveyed according tP the' three' s u ^ e y s "ppndueted at these plants. There appears to be a certain amount of flucua- tion among the various operations for &aph .survey, with tbe~over all means being highest in the mixing, grinding hnd sLand-ing, ana cutting and drilling operations. s' ' ' ' " The concentrations by operations fp-r the .ons-t;rp.ction material plants are listed according to overfall individual plant average in Table No.. 8, The range of -values is small. The' reason that the value of the average of all pamples for the mixing operation is higher than either value listed is that there was m o t h e r -plant in the cohort which had .a high value. That -plant wap not listed because it had low values -for the forming and finishing operations which brought the over--all plant .average' below the highest plant. There were only two .surveys conducted in the construction mater ials plants (Table .No. 9)* There -was a .decrease in mean values in all operations between the .original survey and the resurvey. Only the mixing .operations during the original purvey exceeded the emergency standard of five fibers/pc.. Kesults fox the cement pipe -plants &re -in .Table No.. ;10.. .Only one of the arithmetic means of the asbestos concentrations by opera tions exceeds the .emergency standard of five i-ibers/cc that., in coupling finishing.. This .high mean value -is -the result of one b ample with an .extremely high concpnfration. .Since .there was no immediate apparent reason .for suspecting -that -it -,was not a valid isamplfi,, it was included,. However, it should ,b.e pointed out that all <ohe.r operations at that plant and ail .operations at the other plants have concentrations that are relatively low.. jBased on the proponed .standard of two fib exa/cc, there'.was .only one operation, mixing., in the -two lowest aveiiaged plants .which exceeded that standard,. " ` .Table No. 11 lists the asbestos concentrations by operations for the cement pipe plants .according .to -.the .surveys conducted at each plant,. .There were only -two a.urveys .conducted .at each -plant. .All. mean values for .the .resurvey were .below those from the original survey., .All operations in dement pipe plant .have -mean values which ta.ro below the cmc.i'gen.cy .o.tahdPhd o f j> fib,ers/cc,. -- 110-- SAMPLING METHODS EVALUATION ' A special study was conducted to evaluate the possible valuations between: 1) different campling positions, and 2) repeatability be tween adjacent samples. In the past, it was standard operating pro cedure to ascertain whether a worker was right-handed or left-handed, and then the field monitor would be hung from his left lapel and right lapel, respectively. The first phase of this study consisted of having a worker simul taneously wear two samplers with one field monitor positioned at his left lapel, and one at his right lapel (left-right paired sam ples). The pumps were hung on the worker's belt and the hoses put over his shoulders and attached to the field monitors on each lapel. Table No. 12 lists the results of the left-right paired samples that were collected simultaneously on each worker. There were a total of 23 pair's of samples collected (46 total samples). The results we re tabulated according to which side had the higher concentration,- no matter how small. The total number of samples listed on this table is 22 rather than the 23 pairs which were collected since one pair of samples had the same concentration. The fact that an equal number of samples had higher concentrations on each side does not of itself tell the complete story, since it doesn't show the variability in the differences. A more complete evaluation can be realized if a range of differences is calculated. This was accom plished by subtracting the lower concentration value from the higher concentration value. Where the right side had the higher concentra tion, the range of differences was from less than 0.1 fibers/cc up to 1.0 fibers/cc, with a mean difference equal to 0.3 fibers/co. Where the left side had the higher concentrations, the range of differences was from less than 0.1 fibers/cc up to 1.1 fibers/cc, with a mean difference equal to 0.5 fibers/cc. Thus, it is evident that even the differences (which are minimal) are reasonably ' uniform regardless of which side had the higher value. The second phase of this study was conducted similar to the first' phase m that two samples were collected simultaneously from an in dividual worker. Instead of hanging the field monitors one on each lapel, the field monitors were hung side-by-side on the same lapel. The results of the side-by-side paired samples that were collected simultaneously are listed on Table No. 13- For identification pur poses the field monitor located closer to the center of the worker was called the inside sample, and conversely, the other was -inferred to as the outside sample. There were a total of 85 paired samples collected for this phase of the study. The total number of samples shown on Table No. 13 is 84. This is the result of one pair of samples having the same con centration for each sample. The distribution on these samples shows that 38 of the samples had higher concentrations on the inside fil ter, whereas, 46 had higher concentrations on the outside filter. The binomial probability test v/as applied to the data and the re sults indicated that this distribution of 38 and 46, for the inside and outside samples, respectively, could have easily occurred by chance. There is no significance in the number of samples on the outside which had higher concentrations as compared with those on the inside. A check of the range of differences for the samples that had a higher concentration on the inside filter would reveal values of less than 0.1 fiber/cc up to 1*5 fibers/cc. The mean of these differences was 0.4 fibers/cc. Where the 46 outside samples had higher concentrations than the inside, the range of differences varied from less than 0.1 fibers/cc to 4.9 fibers/cc. The varia tion between the high values of each side, 1.5 to 4.9, appears to be relatively large. This is the result of one pair o f samples. The mean of the differences is only 0.6 fibers/cc for the outside samples in contrast to 0.4 fibers/cc for the inside samples. The difference here is not significant. This phase of the study, the side-by-side paired samples, was also instituted to determine what the variation in the sampling proce dure was for samples that had values close to the proposed stan dard of 2.0 fibers/cc. The side-by-side paired samples were arbitrarily broken down into three separate groups: 1. Samples with mean concentrations less than 1.5 fibers/cc, 2. Samples with mean concentrations between 1.5 and 2.5 fibers/cc; and 3. Samples with mean concentrations greater than 2.5 fibers/cc. The results of the first group (samples with mean concentrations less than 1.5 fibers/cc) are shown in Table Wo. 14. There was an even distribution in the number of samples that had higher con centrations on either side, 28 and 27. The range of differences is the same, 0.2 fibers/cc, which is 10 of the proposed stan dard of 2.0 fibers/cc. The breakdown for the mean concentrations for these samples which had values between 1.5 and 2.5 fibers/cc is shown in Table No. 15There were 11 samples collected on the outside which had higher concentrations as compared with 5 samples collected on the inside. The outside samples had a higher range value, 1.6 fibers/cc, than did the inside samples, 0.9 fibers/cc. The mean of the differ ences was the same for both types of samples -- 0.6 fibers/cc. Table No. 16 presents the data for those samples with, mean con centrations greater than 2.5 fibera/cc. The inside samples to taled 6, with a range of differences between 0.1 and 1.5 fibers/cc. She outside samples, of which there were 7, had a range of differ ences between 0.1 and 4.9 fibers/cc. Since this group includes all samples whose mean concentrations are greater than 2.5 fibers/cc it would be expected that a wider variation would occur between ranges of difference and means of differences because of the larger values involved. The "hi^h" value of 1*5 fibers/cc for the inside samples was the result of a sample whose mean concentration was 4.7 fibers/cc. Whereas the "high*1 value of 4.9 fibers/cc for the outside samples resulted from a sample with a mean concentration of 12.3 fibers/cc. The mean of differences for the inside samples was 0.7 fibers/cc, whereas the outside samples had a value of 2.0 fibers/ cc. Once again, these values are influenced by several high sample concentrations which, as expected, had higher differences between paired samples. SUMMARY There is a wide variation among types of plants as well as among in dividual plants within a given type. This is a result of many factors, including, but not limited to: different processing methods (e.g., spinning and weaving operations in textile plants as apposed to v;et processes in the cement pipe plants); use of different types of as bestos (e.g., long fibers vs. short fibers); good houskeeping prac tices vs. bad housekeeping practices; properly designed and operated control systems; and m o d e m machinery vs. antiquated machinery. Whereas a number of plants have mean concentrations at most opera tions which exceed not only the proposed standards but also the ex isting emergency standard, there is a significant number of plants which have values for most operations that do meet this criteria. There are a number of controlled plants as evidenced by their abil ity to meet the lower proposed standard. An evaluation of the special study based on all the samples collected would indicate that there isn't any appreciable difference between samples that were collected on either the right or left sides or be tween samples that were collected side-by-side, especially for the samples that had low concentrations. As the concentrations got higher on the side-by-side samples, there might be a tendency for the outside sample to collect a higher concentration sample than the in side sample. When the proposed standard becomes law, the variation at the higher concentrations becomes moot. CONCLUSIONS The results of environmental surveys of the various asbestos plants, reveal that dust concentration as measured by the membrane filter me thod are generally below the existing emergency standard. Due to the natural variability of the environment, categories of signifi cantly different dust measurements must be broad. There are currently available methods by which dust concentrations to asbestos du3t can be maintained at or below the existing standard as evidenced by the number of controlled plants in this study. Based on the limited number of samples collected end analysed during the special phase of this study, it would appear that position of the field monitor on the wox'ker has no influence on the resulting concentration. There is little variability between samples collected side-by-side, especially at the lower concentrations. . REFERENCES 1. Lynch, J. r . and H. 3. Ayer, Measurement of Asbestos Exposure, Journal of Occupational Medicine, Vol. 10, No. T, 19bBT 1-4 ,, 2 Timbrell, V., The Inhalation of Fibrous Dusts, Ann. N. Y, Acad. Sci. 132: 1, 19^5: 3. AdcLingley, C G., Asbestos Dust and Its Measurement,'Ann. Occupa. Hygiene, 9: 2, 19657 4. Edwards, G. H. and J. R. lynch, The Method Used by the U. S, Public Health Service for Enumeration of Asbestos Duat on lierabrane filters, Ann. Occupa. Hygiene','*'XI': "T,` l9b7. ' 5. Dreesen, W. C,, J. M. Dalla Valle,!. I* Edwards, J. Y/. Miller, R. B. Sayers, H. F. Eason, and M. F. Trice, A Study of Asbestos in the Asbestos Textile Industry, Public Health Bull. 24 I, 1933. 6. Fulton, ;7. B., A. Dooley, J. 1,, Matthews, and R. L. Hautz, Asbestosis Fart II The Nature and Amount of Dust Encountered in Asbestos fabricating Plants on blie Ee'alth 'oi a Group of Workers, Pennsylvania Dept. of labor and" Industry BuXlV j2, 19 3 5 7 -- ? \ ; I . I r j. ! ! i ! ; I 114-- TABLE NO, 1 ASBESTOS CONCENTRATIONS BY TYPES OF PLANTS Arithmetic Means Type of Plants (Fibers >5jum in length per cc) Original First Survey Resurvey Second Resurvey Over-all Textile Friction Cement Pipe Construction Materials Insulation 4.4 4.9 2.6 4.0 13.9 9.6 2.4 1.7 1.3 12.9 9.0 7.6 4.9 4.3 - 2.2 - 3.2 -- 13.4 TABLE MO, 2 ASBESTOS CONCENTRATIONS BY OPERATIONS - INSOLATION PLANTS t Arithmetic Means (Fibers>5jmni in length per cc) Operation Over-all Individual Plant Average Highest Plant Second Highest Plant Second Lowest Plant Lowest Plant Average of all Samples Mixing 58.9 25.3 11.2 2.0 26.0 Forming 23.2 25.6 3.3 0.3 13.8 Finishing 34.9 27.1 5.1 0.8 16.0 Inspection & Packing 10.9 13 .0 4.0 0.4 5.8 Me ans 26.2 13.5 4.9 0.5 11.8 X TABLE WO. 3 ASBESTOS CONCENTRATIONS BY OPERATIONS . IiVNWSjUiL|iAt||<TION PLAN.Ti?S.mL A^i:tJunet4i:Q; Means Operations. ((EiLfe^>,5>yn in. length. pen <y$). Original purvey-' Re survey ( ^ 35-a.ll MiringForming Finishing Inspection & Peeking 3..4l 12.,3. 15,2 6,1. 3&..a 15,8 15.8 25.0, 5C r v 13..8 15.0 Means - 13.9 X2A % . \3..4 wTvABLE-- N.rOr.rvv-4ASBESTQS CONCENTRATIONS BY OPERATIONS TEXTILE PLANTS Arithmetic Means Operation (Fibers?5pJ in length per pc) Overfall Individual Plant Average Highest ' geeond Second- Lowest Plant Highest Lowest 'Plant pfaht "Plant Fiber Preparation Carding Spinning Twisting Weaving n>i 12,5 12.9 % 2,p r?r? g o ,.4 9.0 ^5.7: 8 * ?-- 4 .4 *:? - 5^.7i 3 :S 8.3 3.1 v2.,W6 Average brail ^Samples it-- . 1i.2.1-- p1k, m1. .r .11. S8.. -. 7, 7.7 A-9 Means 34.."2 ' 9*.'9 4.5 3*4 *t J .7.3 | J -116- TABLE WO. 5 ASBESTOS CONCENTRATIONS BY OPERATIONS TEXTILE PLANTS f8l Arithmetic Means Operation (Fibers>5ym in length per co) Original Survey First Resurvey Second Resurvey Fiber Preparation Carding Spinning Twisting Winding Weaving Means 8.4 11.4 15 .0 5.3 15.5 13-3 4.6 12.1 9.0 4.1 8.5 9.6 4.6 4.5 5.3 3.1 5.8 6.3 4.4 S .6 9.0 Over-all 12.1 11.1 8.7 7.7 5.0 4.9 7.6 TABLE NO. 6 ASBESTOS CONCENTRATIONS BY OPERATIONS FRICTION PLANTS Arithmetic Means Operation (Fibers>!}um in length per cc) Highest Plant Second Highest Plant Second Lowest Plant Lowest Plant Mixing 11.8 4.1 3.7 1.4 Forming 7.1 5.5 1.4 2.5 Hot Pressing 8.7 2.5 1-3 1.4 Grinding & Sanding 8.6 5.7 2.6 2.4 Cutting & Drilling 8.7 5.6 2.2 1.8 Inspection & Packing 7.7 5.8 3-4 1.7 Means 8.5 4.5 2.1 1.7 Average of all Samples 5.3 3.6 *3.4 4.8 4.4 3.8 4.0 . FABLE N O . 7 ASBESTOS CONOSNTRATION^BY OPERATIONS FRICTION PLANTS Arithmetic Means (Fibers>5>m in length, per cc) Operation Original Survey First Resurvey Second Resurvey Waxing Perming 3,4 ?.8 9.2 > 5.1 2.4 Ret Pressing 4.? 4t7 24 grinding & Sanding 1.0 3., 5 4.6 ^ t i n g & Drilling 4,8 if 5.3 ^ , Inspection & Packing 5,0 1,5 3.2 w* ' Jleans 4,S 2 ?4 4.9 Over-all 5 .3 3.6 3.4 4.8 4.4 3.8 4.3 TABLE NO. 8 ASBESTOS CONCENTRATIONS BY OPERATIONS CONSTRUCTION MATERIAL PLANTS fc ri-p im e -ti c Means in length per .cc) Operation - pye r--all Individual----- -- Average -Plant Average of o-ll 'highest Plant ' Lowest Plant Samples Mixing 4.6 Forming A .$ Finishing 3.6 5.9 1.1 2.2 A-:3 3.4 4.-7 4 3.1 118- TABLB NO. 9 ASBESTOS CONCENTRATIONS BY OPERATIONS CONSTRUCTION MATERIALS Arithmetic Means (Fibers>5ium in length per cc) Operation Original Survey Resurvey 1Over-all Mixing 6.8 3.4 5-9 Forming 2.6 1.3 2.2 Finishing 4.8 0.9 3.4 Packing 2.6 0.2 1.9 Means 4.0 1.3 3.2 TABLE NO. 10 ASBESTOS CONCENTRATIONS BY OPERATIONS CEMENT PIPE PLANTS Arithmetic Means (Fibers>5j^m in length per cc) Operation Over-all Individual Plant Average Highest Plant Second Highest Plant Seoond Lowest Plant Lowest Plant Average of all Samples Mixing 4.9 3.5 4.2 1.6 . 3.4 Pipe Forming 2.0 1.4 0.5 1.1 1.9 Pipe Finishing 2.9 3.2 0.8 0.8 1.6 Coupling Finishing 9.5 2.9 1.1 1.1 2.5 Means 3.2 2.6 1.1 1.0 1.8 TABLE NO. 11 ASBESTOS CONCENTRATIONS B 'l OPERATIONS CEMENT PIPS PLANTS Arithmetic Means (Fibers>5pm in length per cc) Operation Mixing Pipe Forming Pipe Finishing Coupling Finishing Original Hesurvey Over-all Survey 4*4 2.2 3-4 2.0 1.6 1.9 1*7 1.4 1.6 3*3 1.7 2.5 _ Means 2.6 1.7 2.2 TABLE NO. 12 " SAMPLING- METHODS EVALUATION Left-Bight Paired Samples Collected Simultaneously Side with Higher Concentration Number Mean of of Differences Samples Fibers>5fun/cc Mean of Differences Pibers^5/um/cc Right Left Low High 11 *0.1 1.0 11 <0.1 1.1 120- TABLE NO. 13 SAMPLING METHODS EVALUATION Side by Side Paired Samples Collected Simultaneously Position y/ith Higher Concentration Inside Outside Number of Samples 38 46 Range of Differences Pibers>5jjot/cc Low <0.1 <0.1 Mgh 1.5 4.9 Mean of Differences Fibers>5jma/cc 0.4 0.6 TABLE NO. 14 SAMPLING METHODS EVALUATION Side by Side Paired Samples Collected Simultaneously ' Samples with Mean Concentrations Less than 1.5 Fibers>5pm in Length per cc Position -with Higher Concentration Inside Outside Number of Samples - 28 27 Range of Differences Pihera>5jum/cc Low <0.1 High 0.7 <0.1 0.8 Mean of Differences PibcrsT-^um/oc 0.2 0.2 TABLE NO. 15 SAMPLING METHODS EVALUATION Side by Side Paired Samples Collected Simultaneously Samples with Mean Concentrations Between 1.5 and 2.5 Fibers >5}-un in Length per cc Position with Higher Concentration Inside Outside Lumber of Samples 5 11 Range of Differences Fibers >5im / c o Low <0.1 0.9 <0.1 1.6 Mean of Differences Fibers >5pm/cc 0.6 0.6 TABLE NO. lb SAEPLING METHODS EVALUATION Side by Side Paired Samples Collected Simultaneously Samples with Mean Concentrations Gi'cater titan 2.5 Fibers >5pm in Length per cc Position with Higher Concentration Number of Samples Mean of Mean of Differences Differences Fibers >5wm/cc Fibers > 5 h w cc Low High - Inside 6 0.1 1.5 0.7 Outside 7 0.1 4.9* 2.0 *Sample Concentration = 12.3 fibers >5pra in length per cc -122FIGURE 1 FIGURE 2 -124- FIGURS 3 i - 122FIGURE 1 123. FIGURE 2 :h'.Sv. 124FIGORE 3