Document a1eVZ7L795MzbXD84o7eOMg2e

FILE NAME: Asbestos in Hair Dryers (HD) DATE: 1979 Apr 10 DOC#: HD062 DOCUMENT DESCRIPTION: J-M Internal Report - Emissions of Asbestos Fiber from Hair Dryers Containing Asbestos Paper Insulation 1 i HEALTH, SAFETY & ENVIRONMENT JOHNS-MANVILLE CORPORATION HAIRDRYER ASBESTOS EMISSIONS APR 10, 1979 j(o yy'<yi'Z^n y* ' & * * * * " f' ' 2 y '^ L e T ^ -y c.. ` r ' - f s x^ t-<i c~ 0 ~h y </Z - O ?_). / y S-. ^O.-^.^. Tri ,,Emissions of Asbestos Fiber From Hair Dryers Containing Asbestos Paper Insulation As a result of a petition filed by the Environmental Defense Fund with the Consumer Product Safety Commission, on March 29, 1979 the Johns-Manville Industrial Hygiene Laboratory commenced a series of tests to measure the levels of fiber emission from hand-held hair dryers containing asbestos paper insulation- The new hair dryers used in this series of tests were purchased on the open market, while the "old" models were borrowed from employees. One model which did not contain asbestos paper was used to investigate the background level of asbestos fiber for the sampling and analytic procedures used in the laboratory. Ideally, such laboratory tests are conducted under "clean room" conditions so that the background fiber counts would be negligible The volume of air emitted from the dryers and other considerations however, made this impossible. Under any conditions it is good science to investigate background levels. The tests were run under three different conditions. In one case, the air was sampled by passing through pre filtered water in a large plastic container. After the designated running time, equal aliquots of this water were filtered through both Millipore and Nuciepore filters. These filters were then submitted to the microscopic lab oratory for examination by both transmission and scanning electron microscopy. , The second method of the sampling was direct sampling of the air from the dryer using a high volume sampler. The I dryer was attached to a tube and all the air emitted from the dryer passed through the tube. In this case a known fraction of the air emitted from the dryer was drawn through a Millipore filter for a specified period of time. The air volumes were measured with a thermal anamometer and corrected for temperature. The sampling was also done under near isokinetic conditions so that fibers in all size ranges would be captured. In this case it was not possible to v> sample on Nuclepore filters, therefore, the only analysis which was carried out was by transmission electron micros copy. The third method was to sample the air, with a high volume sampler, in the immediate vicinity of a simulated normal use test. For analysis by scanning electron microscopy, a 1 centimeter diameter section of a Nuclepore filter was cut out and mounted on an aluminum sample stub. This sample was then coated with carbon in a vacuum evaporator and then examined directly in the scanning electron microscope at 1500X magnification. Some of the p i c t u r e s included in the report were taken at either higher or lower magnifications. Lower magnifications were used in order to show a lai'ger and more rrepresentative area of the filter surface, whereas higher magnifications were used to show more details of the asbestos fibers encountered. All of the fibers encountered were chrysotile asbestos. This was confirmed by energy dispersive x-ray analysis, In several of the pictures there are very low contrast arti-- facts that appear on the surface of the filter. These are the desicated remains of bacteria which grew in the water between the time the sample was taken and the time that the aliquot was filtered. These images ar"'*not to be confused with asbestos fibers. For transmission electron microsfcopy, the Millipore filters were first treated as recommended by the Millipore Corpo ration to collapse their structure, and then coated with carbon in a vacuum evaporator to "glue down" particles re maining on the surface. A three millimeter diameter portion of the filter was then placed on a carbon coated electron microscope grid and the filter material cleared using the cold finger acetone extraction method. Proper pretreatment of the filter media prior to the extraction allows complete clearing to take place within 30 minutes so that the possible loss of particulates is held to a minimum. After clearing, the electron microscope grid is ready for examination. For this analysis 15 complete grid squares were examined at a magnification of 48000x. The presence of chrysotile fibers was verified by both electron diffraction and by energy dispersive x-ray analysis. In order to quantify the amount of chrysotile fiber present, the dimension of- each fiber bundle, fibril or clump was recorded so that the mass could be calculated. The mass was calculated assuming a cylindrical cross section for fiber bundles and a specific gravity of 2.60 grams per cubic centimeter. The mass of the clumps W-- inf i udfid_ in _the calculation because , in most cases, they were mixtures ofchrysotile fibers and other unidenti- Itj/as not possible^ in most, cases.to count o_f_all fibexs._j,n__aL.c Tump. The va ^ j n a j p r i t X_-_of__these _clumps contained only,. short chrysotile flbii-Which..would, contribute very little, to _the total mass of fiber emitted,_ - It is imperative to understand the various ways numbers are used to describe airborne asbestos. Current OSHA standards for occupational exposures to asbestos are: " Standard Effective J uly 1, 1976. The 8-hour time-weighted ,1 O'1'y average airborne concentrations of asbestos fibers to which y any employee may be exposed shall not exceed 2 fibers, longer than 5 micrometers, per cubic centimeter of air, as determined by the method prescribed in paragraph (e) of this section. geiljjia-cogc.entration. No employee shall be exposed at any time to airborne concentrations of asbestos fibers in excess of 10 fibers, longer than 5 micrometers, per cubic centimeter of air, as determined by the method prescribed in paragraph () this section." The extremely heavy occupational exposures of the past which are responsible for the morbidity and mortality experience of o c c u pational cohorts have been orders of magnitude (i.e. 10X, 100X, 1000X ) higher than permitted by present OSHA standards. However, to put fiber levels measured in the hair dryer tests into perspective with occupational exposures, even the highest' airborne levels predicted in laboratory experiments by EMV ant J-M would not be statistically detectable using t h e .sampling and, analytical techniques for .occupational exposures pursuant to the OSHA asbestos standard. 4}/ l x 4'" For example, in the membrane filter method for analyzing * * occupational exposures to asbestos, it is acceptable to count a few fibers on unused, new blank filters taken directly from the box. A count of up to 5 fibers on a blank is con sidered acceptable according to the NIOSH document P&CAM 239, which describes the analytical method (Asbestos Fibers In Ai r ) . If a personal sample is taken for two hours in the workplace (assuming a 2 liter/minute sampling rate) and a . 2 counting area of .005 mm (a size m the recommended range) is used on the microscope, each fiber counted "represents" 0.007 f/cc if 100 fields of view are counted. Even if the maximum 2-hour bundle count estimate in the EDF petition is assumed.to be ..entirely airborne in 18 of space, that would giver. = 0.53 X 10~2 == 0.005 f/cc : which is. .less than, what a.si ngle-f iber "represents" in the above example and in which is predictably found on blank, never used filters.. The extremely small numbers used here arc for purposes of illustration only, as it is widely known that the, occupational sampling/analyti cal procedure is not capable of reliably measuring at these extremely low levels. ` The purpose of the discussion of occupational exposures is not to suggest that non-occupational exposures should be permitted to approach OSHA's permissible occupational exposure levels, but rather_to_address_and refute the...EDF...data which iffiply that fiber levels emitted from hair dryers may approach O SHA's pexmissible occupational,, levels,.....Such an inference is^ scientifica!ly specious.. More sensitive sampling/analytidal techniques utilizing electron microscopy must be used for the very low levels encountered from hair dryers, in contrast to the method for measuring occupational exposures to asbestos. Even these more sensitive techniques have limitations and studies to improve the accuracy and precision of electron microscope techniques are on-going. For example, two quotes from EPA Technology Series, EPA-600/278-038 (June 1978) are appropriate for this discussion: "The presence of a few large bundles of fibers strongly j influence mass concentration estimates." j and "The conventional transmission electron microscope is superior to the scanning electron microscope for detecting and identifying chrysotile fibrils." To paraphrase the above, cw fiber bundles generally strongly bias the mass concentration (i.e. too high) and TEM is superior for the evaluation of the smaller fibrils. SincG EMV Associates reported_such extremely^smaXljTiass-quantities, and particularly since they disturbed the asbestos liners before conducting a test for the release of asbestos fibers, serious doubts must be -raised- as-to the.,validity o f .their data .end...the COr^ A - - i---- reached.,.therefrom,. Specifically, the EDF petition describes the EMV tests as follows: "All size dryers had asbestos liners. This was confirmed by examining a small section of the liner by the SEM/EDXA method described before. When' this confirmation took place/ before the test for asbestos release, the cut edges of the liner were sealed with glue." average bundle size of 2X2X30 microns i s q u i t e large. Assuming a 2X2X30_ micon_size,,.. if ,,a.,2X2X30..box. shape is assumed, that gives a volume of 120X10 3 JHEL A specific gravity of 2.6 gives a mass of 312XlO~^3gm or .312 ng,. 'which does^not_agree with the O .52 ng given in the footnote to Table 1. ^ volume of 94.25 cm3 and a rnajss^calculation _o^0 .245 jng instead Since the EMV study: 1. Found few bundles, which bias the mass- on the high side, 2. Used questionable scientific procedures, and 3 . Presented mass calculations which are incorrect., ^the EMV results are of questionable validity and emphasize the need for additional testing. The water sampling (duplicating the EMV test) done by J-M involved four dryers (three with asbestos paper) and four tests (one of the dryers with asbestos paper was run twice.) Twenty grids were counted .for,,each .test, .with only., a. single bundle I \ (quite, large and non-respirable) observed in total in all^the, * tests. This is in contrast to .the numbers of bundles given^ in Table 1 of the None of these., numbe-rs.-.should be viewed as definiti v e . Rather, the lower SEM counts in the J-M data and questionable scientific procedure in the EMV tests point to the need for additional studies. In addition to the water analyses conducted by J-M by SEM, TEM work on the water samples and TEM work on the air samples were done by J-M, which provides us with additional data not obtained by EMV. The mass determinations by TEM are significantly lower [ than SEM mass estimates in the EDP Table 1. (P <..03 using 2 1 tailed Wilcoxon 2 sample ranks test). When one considers the conclusions in EPA's Technical Series, EPA-600/2-/8-038 (June 1978) regarding the significance of a low count of bundles and the fact that moisture is normally present in a bathroom which suppresses dust, that exhaust fans or windows are always present to remove fibers and that routine vacuuming ' and cleaning will also remove fiber, the lower order of magnitude of the mass (as determined in J - M 's TEM analyses) which was found to be emitted from the dryers lead us to conclude that ^ n-- liable-- 3-- of-- the-- RDF- petition (which Table was contructed from the highest reported finding) are at least two S i -- 51--90.4,tude^toO--.h.igh_. Therefore the concentrations of W, -------- asbestos speculated in Table 3 of the EDF petition should be r o u g e d, bo the range of the EPA survey^ of ^urban.,.and.. rural,,..air t forth in Table 4 in the EDF petition. At the same time, it should be noted that we question the accuracy of the numbers reported in Table 4 and their validity and application in assessing potential risk. 0 The quantitative analysis of asbestos fibers by electron microscopy is not an exact science. The precision is ex tremely poor when compared to chemical analyses in general. In addition, because of the very poor precision very little is known regarding the accuracy of analyses by electron microscopy. Finally, there is very little agreement among workers in this field regarding the methods that should be used to collect and prepare the samples for this kind of analysis. However, most workers do agree on one major point. The transmission electron miscroscope, and not the scanning electron microscope, is the instrument of choice for this type of analysis. Only by the former method can the ex tremely small asbestos fibers normally found in the environ ment be detected and characterized reliably. The results obtained by the two methods in this series of tests demon strate this difference quite dramatically. One of the principle sources of error in quantitative electron microscope analysis can be traced to the extremely small fraction of the sample which is actually examined. In many cases the detection of a single fiber requires a multi plication by a factor as large as 1,000,000 to determine the final concentration. Even if there were no errors involved in the collection or preparation of the sample for examinatic there is irreducible statistical variation in'counting that can be a major factor. In this type of analysis it must be recognized that other errors are unavoidable, there!ore we II *> must be content with the fact that errors as large as a factor of 10 will be commonplace, m consequence, very little credence should be placed on the absolute value of any numbers reported by these methods. They are only indicative of the approximate quantities of material present. * , * ^ The petition filed by the Environmental Defense Fund asking for a recall of all hair dryers containing asbestos in sulation, 'assumes the data provided by EMV Associates were accurate. Furthermore, the petition assesses the potential risk involved by making comparisons with data obtained by other laboratories by entirely different methods. Because of the intrinsic inaccuracies of the method ologies, one cannot make meaningful comparisons of data obtained by different methods in different laboratories. The only way in which any degree of consistency might be obtained would be in comparing results within a single laboratory by the same analyst. In any event, such data cannot, and should not, be used as has been done by EDF as a means of comparing the risk associated with exposure to these low levels of asbestos fiber. Those who do make such risk assess ments are doing so with the apriori assumption that any exposure whatsoever to asbestos fiber can cause disease. However, there is no evidence to support such an assumption. # All available evidence leads to the opposite conclusion. There is an increasing body of evidence which indicates that O exposure to very low levels of asbestos fiber, as found in non-occupational and non-paraoccupational situations, are not responsible for disease in man. Several arguments in this regard are presented in the attached addendum to the testimony of Dr. J. P. Leineweber before the House of Representatives Subcommittee on Elementary, Seco'iYdary and Vocational Education. This document, together with the documentation in the form of published papers by recognized authorities in the field, is now part of the record of * HR 3282 now before the House of Representatives. There are two additional concepts which are put forth in the EDF petition which must be corrected. First, they imply that all of the fiber which is released from a hair dryer will accumulate and remain in the room in which the hair dryer is used. What they did not consider is the fact that homemakers normally clean their homes on a regular basis. Each time a room is cleaned with a vacuum cleaner one would have to assume that at least 90% of the accumulated fiber would be removed from the environment. Wet mopping of a bathroom floor would probably remove an even greater per centage of any accumulated fiber. In any event, periodic cleaning will remove a significant portion of any dust. Secondly, from reading the EDF petition, one might get the impression that all of the fiber released from the dryer will be inhaled by the user. This is manifestly not so. Based on sxistmcr knowledae, looselv bonded asbestos naner as an insulator in the barrel of hand held hair dryers represents an inappropriate use of asbestos. We believe that non-locked-in uses of asbestos fiber, which result in the release of fiber to the ultimate user, should be'-replaced with appropriate substitutes. However, it must also be recognized that there is no evidence to indicate that the use of hair dryers containing these asbestos papers poses a health hazard to anyone. The fiber concentrations emitted are, in fact, consistent with those found in urban environ ments for which there is absolutely no evidence of risk. In conclusion, the manufacturers of hair dryers should be requested to discontinue the use of asbestos paper. However, the general public should also be informed that there is no evidence to indicate that the continued use of hair dryers containing asbestos paper will pose a health hazard to any one. Therefore, such hair dryers need not be recalled. We conclude from our review of the EMV data as well as from the results of our own studies that asbestos fibers released incidental to the usual use of hair dryers creates no health hazard. This is entirely compatible with epidemiological data derived from persons exposed to ambient concentrations of asbestos (non-occupationa1 and non-paraoccupational). The absolute numbers in the EDF petition and in this J-M study are meaningless from a health hazard point of view. What is meaningful is that small numbers of fibers are being observed and that most of those are either non-respirable or too short to be biologically active. The evidence suggests that the EMV results overstate the mass of asbestos emitted by hair dryers. Furthermore, even the highest airborne fiber levels that can be calculated using unrealistic assumptions (e.g. all fibers remain airborne)* predict airborne fiber levels orders of magnitude below current permitted occupational levels.