Document QgDwyGbr9XDgddmkJvVy5yk76

S.T0036565 Asbestos monitoring By Walter C. McCrone sbestos is considered by most authorities to yield filter samples that can be confidently evalu Abe one of the major occupational health and safety problems in the world today. Al ated by phase contrast microscopy (PCM). Methods exist for the identification and quantita though asbestos is no longer used in newly construction of asbestos in both bulk and air samples. Stain ted buildings, most old buildings contain large ing methods, x-ray diffraction, infrared absorption, quantities of asbestos. The first problem a building and differential thermal analysis have all been used owner faces today is to determine if a potential haz for asbestos identification, but these nonmicrosco- ard exists; in other words, whether there are asbes pical methods usually require orders of magnitude tos-containing materials within the building. This larger samples than microscopical methods -- involves bulk sample analysis by polarized light mi micrograms versus nanograms. This may not be a croscopy (PLM) on insulation and other building problem for bulk samples, but it would be difficult materials from all likely sources. Based on these re for air filter samples. Thus it seems likely that most sults, decisions can be made regarding sealing of the analytical chemists would agree that one of the asbestos source to prevent contamination of the microscopes (polarizing, scanning, or transmission building air or removal of the asbestos. If the choice electron) would be the method of choice. There are is removal, that operation must be carried out with strong advocates in each of these camps and it is great care to avoid hazard to the workers and sur very difficult, especially for a building owner or re rounding population and to make certain that once moval contractor, to evaluate the claims made by removal is completed, the building is "clean." each. Again, air filter samples and the monitoring of Personnel at McCrone Associates in Chicago and those filters for the presence of asbestos are re Atlanta and at McCrone Research Associates in quired at this stage. "Aggressive" air sampling will London over the past 15-20 years have analyzed several hundred thousand samples of insulation. Dr. McCrone is Director, McCrone Research Institute. Figures 1-5 show typical asbestos samples. Most of ; APRIL 1985 ASBESTOS continued 100 i Figure 3 Crocidollle fibers ms in Figure 1. Figure 4 UInert! wool fibers es in Figure 1. these were bulk insulation samples, though thou sands of water and air samples were also analyzed. We have no question concerning the analytical method for the water samples since nearly all of the fibers in these samples are too tiny to be seen by light microscopy. The identification and quantita tion of these fine fibers can only be done by trans mission electron microscopy (TEM). Only that technique can routinely identify and quantitate the percentage of the different asbestos fibers in most water samples. Even the TEM, however, needs the capability for elemental analysis in order to differ entiate between the various amphiboles on a routine basis. This capability is only available on analytical TEMs costing several hundred thousand dollars. On the best routine basis the cost of an analysis by TEM will be high, several hundred dollars per sam ple. However, since neither the light microscope nor the scanning electron microscope (SEM) has suffi cient resolution to do the job, TEM is mandated for water samples. The several hundred thousand bulk insulation samples analyzed in our laboratories have been done almost entirely by PLM. The only exceptions to this are when clients specify confirmation by a second method, and we tend in those cases to use xray diffraction. The identification of the individual types of asbestos in bulk insulation as well as the identification of all other fibrous components of those materials is rapid and certain by PLM, partic ularly when dispersion staining is used. Although we have in our laboratories several scanning elec tron microscopes, several transmission electron microscopes, and a number of microprobes for ele mental analysis, we have elected to analyze many air samples also by PLM. The nature of the sample site determines whether we use PLM or TEM. The fibrils of asbestos are distinctive in shape and size and can be differentiated by a trained light micro- : APRIL 1985 scopist from glass fibers, mineral wool, paper fibers, and other synthetic fibers often present in air samples. IfUtere is any question it is a simple matter for the trained microscopist to select individual fibers from the filter and determine their identity by PLM. Any properly taken air sample in an appro priate area will show at least a few fibers large enough to be easily selected and identified. We would not, however, attempt to analyze ambient air samples for asbestos by PLM. We acknowledge that the filter will hold many fibers too small to detea by light microscopy unless we go to ultramicroscopy or darkfield illumination. This latter technique, which, incidentally, requires little time, would detea individual fibrils as fine as any standard SEM could show, in other words, as small as about 250 A (0.025 ^m) in diameter. This is, notably, the size of the finest asbestos fibrils. We almost never go to this extreme because available evidence indicates that when such tiny fibrils are present there are reasonably well-accepted values for the percentage of finer fibers which cannot be deteaed by PLM. It is extremely unlikely that the polarized light miaoscopist would miss finding as bestos when it is present in a close-to-the-source air sample. It is extremely important however, that the sampling be performed properly and the microscop ist be properly trained. The terra used to describe the sampling operation is "aggressive''; in other words, the air must be aaively agitated in the work area during sampling. Proper training for the microscopist analyzing as bestos-containing dust includes knowing how to use the light microscope and skill at phase contrast, dis persion staining, and polarized light microscopy. The microscopist must be familiar with the optical crystallography and morphology of all types of as bestos and other fibers, e.g., wollastonite, brurite, fibrous talc, paper, cotton, glass, polyester, polya- ST0036566 if 4- * t: ASBESTOS continued 101 ST0036567 Figura S Paper Ubera aa In Figure 1. mide, Kevlar, and other man-made fibers. He or she must be able to differentiate between fibrous and nonfibrous amphiboles and serpentines and not be confused by striations on vermiculite, talc, and mica plates. These are just a few examples of the background areas a microscopist should know to do a dependable job on asbestos samples. (Proper use of the light microscrope for the identification of as bestos and other harmless substitutes is a problem. A background in mineralogy is helpful. The McCrone Research Institute in Chicago teaches a "crash course" in asbestos identification. Other training materials are available.') A microscopist with these skills wQI not be misled by ideas and procedures presented almost daily by nonmicroscopists (or microscopists not trained in asbestos monitoring). One recent paper presented at the American Industrial Hygiene Conference, held May 25, 1983, recommended differentiation of as bestos fibers from nonbirefringent fibers (only glass) by crossing the polars, stating that if the fi bers are invisible they are not asbestos. A simple calculation based on diameter and birefringence of asbestos shows that asbestos fibers smaller than about 1.0 pm in diameter will be invisible with crossed polars. This procedure would lead to nondetection of asbestos in many air samples otherwise easily detected by a trained microscopist. Such mis informed statements are dangerously irresponsible. Another ill-advised idea, point-counting, has been suggested for the quantitation of asbestos in bulk insulation. This is a standard and very valid procedure for well-mixed and nonfibrous compo nent samples for which high accuracy is needed. Anyone who has examined more than a few insula tion samples, however, realizes these samples are poorly mixed and very nonuni form in composition. Grab samples of 1 g each from different portions of an insulation sample will nearly always vary by 10-25% of the fibers present. Smaller samples still too large for a PLM sample often vary from 0-100% fiber. If a large enough number of PLM samples of insulation are mounted and point- counted, the analyst might, with hick, analyze 2-4 samples a day rather than the 20-50 samples now possible--and the answer would be no more accur ate, significant, or useful than the 2-min method with the stereobinocular light microscope to be de scribed. In our asbestos courses we recommend quanti tating large samples (e.g., 1 oz.) using the stereo binocular microscope at 30-40 x. It is easy with that instrument to see how uniform the mixture is . and to estimate in 2-4 min with more than sufficient accuracy the percentage of each individual fibrous component. Two different individuals will usually agree within about 15%, e.g., 17-23%, 34-46%, or 51-69%. This may not sound "accurate," but it is more than accurate enough and as accurate as point-counting could be without hours of counting. What would be the difference in deciding what to do if the insulation contained 60% asbestos rather than 50% or even 75% rather than 50%? Analysis based on an intelligent estimation looking at the en tire sample is 20-50 x faster and probably as accur ate as point-counting on such samples. It is certain ly the only method justified by the sample and the accuracy needed to decide how to proceed next. Jean Prentice of our London laboratory has been very active in all areas concerning asbestos for about 2Q years. She states,1 "With respect to the analysis of insulating materials for the presence of asbestos fibres, low magnifications can be used be cause of the natural distribution of fibres in any bulk material. In other words, if fine fibres are present, large fibres will also be present. Similarly in the analysis of an airborne cloud (providing the air samples are taken close to source) there will be a significant number of fibres which are visible by phase contrast microscopy at 500 x and an assess ment of airborne contamination can be based on this. However, there are times when the optical microscope is totally inadequate, such as looking at ambient levels of fibres in the atmosphere when the source of pollution is a long way from the sampling site and the heavier fibres have dropped out." She continues: "Within the U.K., the standard method of testing for airborne respirable asbestos fibre dur ing or after asbestos removal work is the standard phase contrast-membrane filter technique and, this has in the main been accepted. This technique has been adopted because it is relatively cheap and the results can be quickly obtained. On-site counting is often requested." _______ ' continued AMERICAN LABORATORY : ASBESTOS continued The views of a number of other authorities in this field reinforce this conclusion. Most of these refer* ences are from the Proceedings of the Fourth Inter- national Colloquium on Dust Measuring Technique and Strategy held in Edinburgh. Scotland, in Sep tember 1982.' Dr. J.M.G. Davis, of the Institute of Occupational Medicine in Edinburgh* introduced this meeting with the-fair background statement: "There is a further problem relating to the methods of fibre estimation. Up till now, all routine factory dust counts have been undertaken using light mi croscopy yet this cannot detect fibres below about 0.25 pm in diameter. With asbestos at least, fibres below this detection limit can still be over 10 pm in length and are, therefore, potentially very danger ous. Fibres with this small diameter can only be de tected by electron microscopy. While these tech niques are too costly and time-consuming for rou tine use it would be desirable to undertake occa sional electron microscope counts to make certain that no production process produced large numbers of fine fibres but only small numbers of fibres of the sizes at present counted by light microscopy." Hwang and Gibbs' show that, assuming a visibil ity limit of 0.2 pm using a well-set-up phase contrast microscope at 400 x, the number of amosite fibers missed by an optical microscope survey would be of the order of 20ft or less. If the same experiments are carried out with chrysotile asbestos, one would miss about 50ft of the fibers and for crocidoh'te one would miss about 75 ft of the fibers. The significant point, however, is that one would see not less than 25 ft of the fibers in the worst case, croddolite, and one would know, with sufficient certainty, how many fibers had been missed. Marconi* of the Instituto Supcriore di Sanita Laboraiorio in Rome compared TEM and PCM counts on 30 samples of airborne dust from a rail way repair and maintenance facility and from an auto brakeshoe and clutch operation. The PCM al ways found asbestos when the TEM found it. The TEM found more fibers/mm] but only (on average) 20ft more. Cherrie' of the Institute of Occupational Medi cine in Edinburgh found that carefully done PCM and SEM fiber counts agreed well (especially for amosite) but that the slowness of scan for the SEM to see fibers less than 0.3 pm in diameter makes the use of SEM impractical. Middleton1 of TBA Industrial Products Ltd. in Rockdale, England, summarizes: "the use of slow scan speeds (by SEM) gave improved visibility of fine fibers but it would not be possible to use slow scan rates during routine counting." Robock' of the Asbestos Institute, in Neuss, Ger many, states, "Regarding the necessity to apply the electron microscope for the occupational environ ment it may be stated by means of the comparative measurements conducted within the DMAP and particularly by the work of Carton/France, that this cannot be justified in view of the time con sumed and the result related to the loss in counting thinner fibers (diameter below 0.2 pm) when apply ing the light microscope. Subsequently, (sic: conse quently?) application of electron microscopy for routine measuring of work places is not essential." Tillman,1* from Sweden's National Board of Oc cupational Safety and Health, states in evaluating SEM work, "In investigating the SEM negatives it soon becomes obvious that despite the higher mag nification counting was even more complex than in the phase contrast microscope. In the SEM because more details were visible, it was often more difficult for the microscopist to know how many fibres should be counted." Good indications of these dif ficulties are the results of comparisons of fiber counts by two different laboratories on the same series of 19 air samples (from a variety of manufac turing operations) by both SEM and PCM.1' The two sets of SEM counts varied by factors of 0.27 to 1.86. The two sets of PCM counts varied by 0.66 to 3.30. Tillman's actual data show that the PCM missed 15ft on average of the smallest asbestos fibers in the comparison of SEM versus PCM on 30 different air samples. A number of investigators have found that fibers shorter than 3-5 pm in length have very low, if any, carcinogenicity. The first work in this area by King et al." in 1946 showed that fibers 2.5 pm long showed much less fibrosis than fibers 15 pm long. Similar results were obtained by Vorvold et aL" in 1951. In 1968, Klosterkotter" found much greater reaction with long chrysotile and long croddolite fi bers. Subsequently, studies by HQscher et al.,'* Davis," Wright and Kuschner," Timbrdl and Skid more,1' and Webster" confirmed the general view that fibers shorter than about 5 pm are, at least, much less cardnogenic. The most detailed data on the effects of fiber length were by Stanton and his colleagues." They concluded that the most dangerous fibers were longer than 8 pm and less than 1.5 pm in diameter. These data have been confirmed by Pott" and most recently by Ferguson." These quotations yield considerable confidence in the conclusion that the light microscope, properly used on proper samples, is the only practicaJ ap proach to the detection and identification of asbes- : APRIL 1985 For Particle and Fiber Identification to obtain the most reliable results Dispersion staining, double variation refractometry. focal masking, and polariscopic immersion techniques are among those used all over the world in conjunction with Cargille Retractive Index Liquids in optical analysis and identi fication of solids, glasses, crystals, stress and strain effects, and flow patterns. Custom liquids are developed continuously to meet the specific needs of particular research environments. A phone all mef be enltghtsrung--optically speekmg. toil <J. I A STMEINS3 COMMITMENT A full line of lab stlrrars backed by the best guarantee In the business Try this on for size: 1) 16 lab stirrers ranging from 1/75 tol/8hp and 2) a lifetime guarantee on all parts lor every stirrer (abuse and normal maintenance parts such as brushes and gears excepted). Our catalog contains photos and complete technical information on every stirrer model, plus data on speed controls, paddles, stands, chucks, extension sleeves and bars. Write or calf tUboys Engineering Cor* i ST0036569 Circle Reader Service Card Na 142 Circle Reader Service Card No. 34 ASB ESTOS continued tos in bulk and most air samples. We feel that the inability to detea the very finest fibrils is not a sig nificant handicap in determining the degree of haz ard represented by properly taken and properly ex amined air samples. Finally, the analytical (capable of elemental analysis for Mg, Na, Ca, and Fe) TEM should be used for ail water samples and for all air samples except those close to the source and where aggressive sampling can ensure re-entrainment of the coarser fibers. PCM should be used to evaluate the latter air samples and PLM to identify the mem brane- or Nuclepore -filtered air samples and all bulk samples. Roloroncos 1. MC CRONE. W.C., The Asbestos Allas: IAnn Arbor Science Publishers (available from McCrone Research Institute. 2104 S. Michigan Ave. Chicago, Illinois 60616)]. A slide tape program, "Identification of Asbestos," is available from F.l. Scott A Associates. Check. VA 2-1072. 2. PRENTICE. J.. Personal Communication. 3. Proceedings of the Fourth International Colloquium. ''Dust Measuring Technique and Strategy." Edinburgh, Scotland. September 20-23, 1982. 4. See Ref. 3.. p. 13. 3. HWANG. C.Y. and GIBBS. C.W.. Ann. Ocatp. Hyg. 24, 23-11 (1981). : APRIL 1985 6. MARCONI. A., see Ref. 3., p. 171. 7. CHERR1E, J.. sec Ref. 3., p. 369. I. MIDDLETON, A.P.. see Ref. 3.. p.459. 9. ROBOCK. K.. see Ref. 3., p. 33. 10. TILLMAN. C.. see Ref. 3.. p. 39J. 11. XING. E.J.. CLEGG, J.W., and RAE, V.M.. Thorax I. 188-197 (1946). 12. VORVOLD. AJ.. DURKAN. T.M.. and PRATT. P.C.. AMA Arch. fnd. Hyg. Occ. Med. 3. M3 (1957). 13. KLOSTERKOTTER, W.. Proc. International Conference. Dresden, Deutchcs Zentral institul fur Arbeiismedidn. Ber lin 47-52(1968). 14. HIL5CHER. W.. SETHI. S.. and FRIEDRICKS, K.H.. Nantrwissenschaften 37. 356-357 (1970). 15. DAVIS. J.M.G.. Brit. J. Exp. Path. S3. 190-204 (1972). 16. WRIGHT. G.w. and KUSCHNER. M,, in Inhaled Particles IV (Pentagon Press. New York. 1977), pp. 4J5-472. 17. TtMBRELL, V. and SKIDMORE. J.W., set Ref. 12., pp. 52-56. IS. WEBSTER. L, Proc. of International Conference on Pneumonoconiosa, Johnnesburg (Oxford Univ. Press, 1970), pp. 117-119. 19. STANTON. M.F. et al.. National Bureau of Standards Special Publication 506:143-131. 20. POTT. F,, Staub-Beinhalf Luft 38. 486-490-U978). 21. FERGUSON. J-S-. "Scanning and transmission electron microscopy applications in an asbestos inhalation toxicology study." paper presented at INTER/MICRO-84. Chicago. July 16-20,1984.