Tobacco Documents Online

Proceedings of Workshop on Asbestos: Definitions and Measurement Methods Proceedings of a Workshop on Asbestos held at the National Bureau of Standards, Gaithersburg, Maryland, July 18-20, 1977 Edited by C. C. Gravatt, Philip D. LaFleur, and Kurt F. J. Heinrich National Measurement Laboratory National Bureau of Standards Washington, D.C. 20234 Sponsored by Nationat Bureau of Standards of the Department of Cemmerce and occupational Safety and Health Administration of the li.S. f?epartmen[ of IzMv U.S. DEPARTMENT OF COMMERCE, Juanita M. Kreps, Secretary Dr, Sadney Harman, Under Secretary Jordan J. Baruch, Assistant Secretary for Science and Technology NATIONAL BUREAU OF STANDARDS, Ernest Ambler, Director Issued November 1978 1.0

Library of Congress Catalog Card Number: 78-600109 National Bureau of Standards Special Publication 506 Nut. Bur. Stand. (U.S.), Spec. Publ. 506,490 P:agcs (Nov. 1978) COQEN:XNBSAV ~ U.S. GOVERNMENT PRINTING OFFICE WASEffiVGTON: 1978 ~ {4 O Cn W r+ O a J b CN

FOREWORD Asbestos is a generic name used to describe a variety of hydrated silicate materials which exist as fibers. Because "asbestos" resists heat and acids, is noncombustible, and can be woven into fabrics, it is a valuable industrial material. Asbestos has been known and used since ancient times. Today, it is used in some 3000 cortmercial applications, from potholders, to brake linings, to construction materials. Concern over the use of asbestos has arisen from studies which indicate an increased incidence of various serious diseases among people who work with it. Meaningful regulation requires proper definitions of workplace air concentrations of asbestos and effective measurement methods for these minerals. This Workshop was organized to evaluate the existing state-of- the-art in measuring "asbestos° and is part of an interagency program dealing with definitions and measurement methods for asbestos between the National Bureau of Standards of the Department of Commerce and the Occupational Safety and Neaith Administration of the Department of Labor. philip D. LaFleur, Chief Center for Analytical Chemistry iii

ers PREFACE This Workshop was organized to provide a forum for representatives of industrial corporations, trade associations, regulatory and other federal agencies, state and local agencies and other researchers to discuss asbestos definitions and measurement methods. The Workshop was divided into four topical areas: Mineralogical Aspects, the Relationships Between Chemical and Physical Properties and Health Effects, Analytical Methods, and Regulatory Aspects. The format of the Workshop included presentations of technical papers by invited experts, followed by verbal discussions. At the conclusion of each session there was a general discussion of the material presented. The general discussions served to define those factors for which there is general agreement, what points of controversy exist, and to identify additional research that is required to resolve the remaining problems. The following protocol was employed for the preparation of these proceedings. Each author/speaker submitted a written manuscript based on and containing the material given in the oral presentation. The questions, answers, and comments which followed each talk have been transcribed from the tape recordings made of the Workshop, edited both to remove extraneous material and to improve readability, but without changing the meaning. These discussion sections are printed immediately following the manuscript. The general discussions which followed each session have been similarly transcribed, edited, and printed at the end of each topic section. In addition, any questions, answers, comments, or discussion material which was submitted to the editors in writing has been inserted in the appropriate section of the Proceedings and the material has been designated as "submitted in writing - not in recording of Workshop." I wish to express my gratitude to all those who, through participation in the Workshop or preparation of these proceedings, made this undertaking a success. These proceedings were expertly typed and prepared by Mrs. Joy Shoemaker and members of her Text Editing Facility and the assistance of Mrs. Betty Garrigues in correcting proofs was invaluable. The able assistance of Ors. Ryna Marinenko and John Small in editing the Analytical Methods Session is gratefully acknowledged. It is hoped that these Proceedings will provide useful information to those currently involved in formulating measurement methods, definitions, and regulatory positions with respect to asbestos and other fibrous materials. C. C. Gravatt, Chief Gffice of Environmental Measurements iv

ABSTRACT This document contains invited papers which were given at a workshop on "Asbestos: Definitions and Measurement Methods" which was jointly sponsored by the National Bureau of Standards of the U. S. Department of Commerce and the Occupational Safety and Health Administration of the U. S. Department of Labor. The discussion portions of the Workshop also have been included as has written material appropriate to the topics under consideration which was submitted to the editors at a later date. The Workshop covered four major topics: Mineralogical Aspects, the Relationship Between Chemical and Physical Properties and Health Effects, Analytical Methods, and Regulatory Aspects. Also included in these Proceedings is a summary of each of these topics. These summaries serve to define those factors for which there was general agreement at the Workshop, identify remaining points of controversy, and, in some cases, describe additional research required to resolve remaining problems. Key Words: Amphibole; asbestos; fibers; light microscopy; mineralogical terminology; scanning electron microscopy; serpentine; talc; transmission elec- tron microscopy. v

TABLE OF CONTENTS PAGE FOREWOPO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . i i i PREFACE ........................................ iv ABSTRACT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v SECTION I. MINERALOGICAL ASPECTS Chairman - Brian Mason, Smithsonian Institution, Washington, D. C. HISTORY OF ASBESTOS-RELATED MINERALOGICAL TERMINOLOGY . . . . . . . . . . . . . . . 1 Tibor Zoltai FIBROUS AND ASSESTIFORM MINERALS . . . . . . . . . . . . . . . . . . . . . . . . . 19 James R. Kramer THE CRYSTAL STRUCTURES OF AAIPHIBOLE AND SERPENTINE MINERALS. . . . . . . . . . . . 35 Jack Zussman THE "ASBESTOS" MINERALS: DEFINITIONS, DESCRIPTION, MODES OF FORMATION, PHYSICAL AND CHEMICAL PROPERTIES, AND HEALTH RISK TO THE MINING COMMUNITY ......... 49 Malcolm Ross GENERAL DISCUSSION OF MINERALOGICAL ASPECTS . . . . . . . . . . . . . . . . . . . . 65 SESSION II. RELATIONSHIP BETWEEN CHEMICAL AND PHYSICAL PROPERTIES AND HEALTH EFFECTS Chairman - Marvin Schneiderman, National Cancer Institute of the National Institutes of Health, Bethesda, Maryland EPIDEMIOLOGICAL EVIDENCE ON ASBESTOS . . . . . . . . . . . . . . . 71 W. J. Nicholson, A. M. Langer, and I. J. Selikoff MEASUREMENT OF ASBESTOS RETENTION IN THE HUMAN RESPIRATORY SYSTEM RELATED TO HE4LTH EFFECTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 J. Bignon, P. Sebastien, and A. Gaudichet EPIDEMIOLOGIC EVIDENCE OF THE EFFECT OF TYPE OF ASBESTOS AND FIBER DIMENSIONS ON THE PRODUCTION OF DISEASE IN MAN . . . . . . . . . . . . . . . . . . . . . . . . 121 W. Clark Cooper PATHOPHYSIOLDGY IN RELATION TO THE CHEMICAL AND PHYSICAL PROPERTIES OF FIBERS. . . 133 Paul Kotin THE CARCINOGENICITY OF FIBROUS MINERALS . . . . . . . . . . . . . . . . . . . . . . 143 Mearl F. Stanton and Maxwell Layard NIEHS ORAL ASBESTOS STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153 John A. Moore EPA STUDY OF BIOLOGICAL EFFECTS OF ASBESTOS-LIKE MINERAL FIBERS. ......... 163 0. L. Coffin and L. 0. Palekar A STUDY OF AIRBORNE ASBESTOS FIBERS IN CONNECTICUT . . . . . . . . . . . . . . . . 179 Leonard Bruckman GENERAL DISCUSSION OF RELATIONSHIP BETWEEN CHEMICAL AND PHYSICAL PROPERTIES AND HEALTH EFFECTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191 vi

Table of Contents Continued SESSION`III. ANALYTICAL METHODS Chairman - K. Heinrich, Analytical Chemistry Division, National Bureau of Standards PAGE IDENTIFICATION OF SELECTED SILICATE MINERALS AND THEIR ASBESTIFORM VARIETIES . . . 201 William J. Campbell AN OVERVIEW OF ELECTRON MICROSCOPY METHODS . . . . . . . . . . . . . . . . . . . . 221 Clayton 0. Ruud IDENTIFICATION OF ASBESTOS BY POLARIZED LIGHT MICROSCOPY . . . . . . . . . . . . . 235 Walter C. McCrone MINERAL FIBER IDENTIFICATION USING THE ANALYTICAL TRANSMISSION ELECTRON MICROSCOPE . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249 0. R. Beaman and H. J. Walker TRANSMISSION ELECTRON MICROSCOPICAL METHODS FOR THE DETERMINATION OF ASBESTOS. . . 271 Ian M. Stewart STATISTICS AND THE SIGNIFICANCE OF ASBESTOS FIBER ANALYSES . . . . . . . . . . . . 281 J. P. Leineweber SELECTION AND CHARACTERIZATION OF FIBROUS AND NONFIBROUS AMPHIBOLES FOR ANALYTICAL METHODS DEVELOPMENT . . . . . . . . . . . . 295 J. C. Haartz, B. A. Lange, R. G. Draftz, and R. F. Scholl ASGESTIFORM MINERALS IN INDUSTRIAL T.RLCS: COMMERCIAL DEFINITIONS VERSUS INDUSTRIAL HYGIENE REALITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313 John M. Dement THE DETECTION AND IDENTIFICATION OF ASBESTOS AND ASBESTIFORM MINERALS IN TALC. . . 325 Harold 0. Stanley MISIDENTIFICATION OF ASBESTOS IN TALC . . . . . . . . . . . . . . . . . . . . . . 339 Jerome B. Krause and William H. Ashton AMBIENT AIR MONITORING FOR CHRYSOTILE IN THE UNITED STATES . . . . . . . . . . . . 355 Richard J. Thompson ENVIRONMENTAL PROTECTION AGENCY INTERIM METHOD FOR DETERMINING ASBESTOS IN WATER. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 365 Charles H. Anderson INTER-LABORATORY MEASUREMENTS OF AMPHIBOLE AND CHRYSOTILE FIBER CONCENTRATION IN WATER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 377 K. S. Chopra THE STANDARD FOR OCCUPATIONAL EXPOSURE TO ASBESTOS BEING CONSIDERED BY ASTM COMMITTEE E-34. . . . . . . . . . . . . . . . . . . . . . . . . . . . 381 M. Cossette and A. A. Winer IDENTIFICATION AND COUNTING OF MINERAL FRAGMENTS . . . . . . . . . . . . . . . . . 387 R. J. Lee, J. S. Lally, and R. M. Fisher PRACTICAL ASPECTS OF TALC AND ASBESTOS . . . . . . . . . . . . . . . . . . . . . . 403 C. J. Parmentier and G. J. Gill GENERAL DISCUSSION OF ANALYTICAL METHODS . . . . . . . . . . . . . . . . . . . . . 413 Vii 20631.04801

Table of Contents Concluded PAGE SESSION IV. REGULATORY ASPECTS Chairman - John Martonick, Occupational Safety and Health Administration, Washington, D. C. I NTRODUCTI ON . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 421 John Martonick THE MINING ENFORCEMENT AND SAFETY ADMINISTRATION - REGULATIONS AND METHODS .... 423 Aurel Goodwin QCCUPATIGNAL SAFETY AND HEALTH ADMINISTRATION METHDDS . . . . . . . . . . . . . . . 431 Willard C. Dixon FDA PROJECTS AND METHODS . . . . . . . . . . . . . . . . . . . 441 J. A. Wenninger, I. M. Asher, and P. McGrath CPSC REGULATION OF NON-OCCUPATIONAL EXPOSURE TO ASBESTOS IN CONSUMER PRODUCTS. . . 451 Robert M. Hehir, Steven P. Bayard, and June Thompson IMPACT OF ASBESTOS REGULATIONS ON THE MINING INDUSTRY . . . . . . . . . . . . . . . 461 C. S. Thompson GENERAL DISCUSSION OF REGULATORY ASPECTS . . . . . . . . . . . . . . . . . . . . . 469 LIST OF ATTENDEES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 479 Disclaimer: Certain trade names and company products are identified in order to adequately specify the experimental procedure. In no case does such identification imply recoomend- ation or endorsement by the National Bureau of Standards, nor does it imply that the pro- ducts are necessarily the best available for the purpose. N w ~ 0 N

National Bureau of Standards Special Publication 506. Proceedings of the Workshop on Asbestos: Definitions and Measurement Methods held at NBS, Gaithersburg, MD, July 18-20, 1977. (Issued November 1978) HISTORY OF ASBESTOS-RELATED MINERALOGICAL TERMINOLOGY Tibor Zoltai Department of Geology and Geophysics University of Minnesota Minneapolis, Minnesota 55455 Abstract Asbestos-related mineralogical terms such as fiber, fibrous, orm asbestos-like, and asbestos have been misinterpreted and ing the last few years in the literature of environmental MNAand public health studies. The new definitions are inadequate for the proper description and study of various mineral particles and, at the same time, are causing considerable confusion in interdisciplinary communication. The meaning of these terms is traced through the history of mineralogy. It is demonstrated that: the use of the•term fiber has always required some resemblance to organic fibers; fibrous has been the term describing a crystallization habit in which the mineral appears to be composed of fibers; asbestiform has been used, without exception, to describe a special fibrous habnt in which the fibers have higher tensile strength and flexibility than crystals in other habits of the same mineral; asbestos was initially the name of an independent mineral species and gradually became a collective term applied to all asbestiform varieties of minerals. Ney words: Acicular; amphibole; asbestiform; asbestos; fiber; fibrous; fragments; mineralogical; serpentine; terminology. Introduction Until a few years ago there was no problem with the asbestos-related mineralogical terminology. Mineralogists knew exactly what other mineralogists meant when they used terms like asbestos, asbestiform, fibrous, and acicular, even if some of these terms, like asbestiform, are not always defined in textbooks. The last syllable of asbestiform (that is, -form) is consistent with several adjectives used for the description of textures or crysta~Tazation habits (e.g., reniform, filiform, dentiform, coiloform). Consequently, it is understood, without question, that asbestiform is a descriptive term for a certain texture or crystallization habit. This situation of content was suddenly changed less than five years ago, when through the focusing of public and scientific attention on asbestos pollution this portion of the mineralogical terminology was picked up by environmental and public health scientists, by engineers and by lawyers. Unfortunately, they did not adopt the terminology as used by mineralogists but have introduced a redefinition of most of the critical expressions, in spite of the objection of leading mineralogists. The most important of these arbitrary changes of definitions included: (1) Asbestos is understood by mineralogists as a collective term referring to the unusual crys~Tization of certain minerals in the form of long, strong, and flexible fibers, aggregated in parallel or radiating bundles from which the fibers can easily be 2063104803

separated. The definition accepted by the Minnesota District Court during the trial of Reserve Mining Co. [63, p. 24],t however, was a different one: Asbestos is a generic term for a number of hydrated silicates that, when crushed or processed, separate into flexible fibers made up of fibrils. (emphasis by the author) By this definition ali amphiboles and a number of other minerals became possible candidates for inclusion in the term asbestos. Because of the perfect prismatic cleavage, upon crushing, amphiboles always produce acicular fragments. Of course, acicular fragments are not fibers, are not flexible and are not composed of fibrils. However, they may not be distinguishable from asbestos fibers in routine electron microscopic examination. In order to get around that problem the term fiber had to be defined in a more practical sense. (2) The redefinition of fiber (U.S. District Court, District of Minnesota, Fifth Division, Fall, 1973) that was soon adopted by most environmental and public health scientists [28, p. 5] states that a fiber is: a mineral which is at least three times as long as it is wide.2,a This definition of fiber eliminated the difficult task of testing the flexibility and the presence of fibril composition of submicroscopic particles, and retained only the shape of the particle as a decisive criterion. Accordingly, all acicular amphibole cleavage fragments became fibers and as indirectly implied, all amphibole minerals became asbestos. (3) Leading mineralogists objected to calling amphibole cleavage fragments, asbestos fibers and amphiboles, asbestos minerals. In order to overcome that objection two less frequently used terms, "asbestiform" and "asbestos-like", were redefined in line with the new definitions of asbestos and fiber. The new definitions were introduced in the Minnesota courtroom [63], and subsequently in the language of the news media and the environmental literature: Asbestiform became a prefix added to the name of any mineral which is known to occur on occasion andlor produce "fibers" when crushed. Asbestos-like was defined as any hydrous silicate particle which is at least threeT nger than wide, that is, which is a°fiber". Thus, all amphiboles became asbestiform minerals,° instead of asbestos minerals, and amphibole fragments became asbestos-like fibers, underscoring its implied relationship with asbestos. These new definitions provided a simplified mineralogical interpretation for the complex and not fully resolved problem of asbestos mineralogy. It simplified the identification of mineral particles by eliminating the need for distinction between asbestos fibers and acicular cleavage fragments. A fiber can simply be identified by its shape (>3:1 aspect 1Figures in brackets indicate the literature references at the end of this paper. 2The 3>1 aspect ratio limitation In the description of fibers was used before by some British and American regulatory agencies. However, this was the first incident when this fiber description became an asbestos fiber identification, as the use of the term fiber im~plfed an identity between appropriately shaped amphibole fragments and amphibole asbestos fibers. This implicative use of the 3>1 aspect ratio is apparent in most current environmental studies. 311; should be noted that sedimentologists use the term acicular for the description of particles "whose length is more than three times its width"T2-f7.p_5]. 4The expression "asbestiform amphiboles" is basically valid. However, in the context of the new definitions it is erroneous as it includes all amphiboles. According to the proper mineralogical terminology the same expression 7s limited to those amphibole crystals'which actually grew in the asbestiform habit. 2

ratio) and if its composition and lattice matches that of an amphibote, that particle can be called ®asbestiform" amphibole, or simply, "asbestos". Consequently, all available data on the health hazards caused by the inhalation of asbestos fibers can be applied to acicular amphibole fragments, thus eliminating the need for the extensive job of determining the nature and the extent of the health effects of the actual particles, that is, the acicular amphibole fragments. On the other hand, the new definitions created serious problems, probably not forseen by the promoters of the new definitions. For example, jade became an asbestos in spite of the fact that jade is the toughest known natural substance [8]. One type of jade (nephrite) is mineralogically actinolite-tremolite, and according to the new definitions, it is an "asbestiform mineral" and its acicular fragments are "asbestos-like fibers". The other type of jade is jadeite, a pyroxene. Pyroxenes are similar to amphiboles as far as both are chain silicates and break into acicular fragments. The only major difference between these two groups of minerals, in terms of their qualifications for "asbestos", is that pyroxene is not "hydrated". Consequently, in terms of the new definitions jadeite is not an asbestos. However, one could argue whether the presence of OH is really necessary in the definition of asbestos.s At the same time the new definitions include many non-asbestiform mineral varieties in the rank of asbestos, they also exclude a number of other minerals, (e.g., non-hydrous silicates) which in fact may also crystallize occasionally in asbestiform habit. Most of these minerals are rare and are not known to constitute commercial deposits. Nevertheless, a mineralogical definition should not be tied to commercial criteria. The new definitions, of course, magnify the extent of the potential asbestos pollution problem by an exponential factor. If all amphiboles are "asbestiform" and their fragments are °'asbestos-like" then every state in the union has some asbestos in the soils, drifts, and bedrocks. Kryvial, Wood, and Barrett show [44, p. 13] the distribution of "high con- centration of asbestiform phases" of rocks in the continental United States. Only a few of the amphibole-bearing rocks included in that survey contain even a minor fraction of known, true asbestiform varieties of amphiboles. the new definitions are not only contrary to mineralogical traditions but are inadequate for crystal chemical descriptions. They also can lead to ambiguity and contradiction. For example, Kryvial, et al. in their monograph [44, p. 5] wish to exclude hornblende from the 11asbestiform" category of amphiboles; apparently because hornblende seldom crystallizes in asbestiform habit. However, the new definition of "asbestiform" does not allow them to use it in that sense, or to express the same concept in any other non-ambiguous way. They try to get around the problem by using the term fibrous in an ambiguous way by stating that hornblende is '°seldom seen in a fibrous form". Yet in page 3 of the paper they state that all amphiboles "fragment into fibers", whether they are products of an "acicular form of a fibrous crystal" (7) or not. They admit that at a microscopic scale the fragments of hornblende are no different from that of other amphiboles. Apparently, what they are trying to say is that fibrous is not always fibrous, but the new terminology does not allow them to distinguish between these two types. The proper mineralogical terminology can do that. Most mineralogists object to the misuse of the mineralogical terminology. Some mineralogists, however, have found themselves in situations where compromise was necessary and they used the new definition of fiber (>3:1 aspect ratio) and the term "fibrous" in an accordingly loose context [42,45,71], sometimes with comments on the disciplinary restric- tions of that terminology [11].e slt is not entirely impossible that the minute acicular crystals of jade may turn out to possess some asbestos properties. In that case minute fragments of jade could appropri- ately be called asbestiform fibers. sThe authors of [11] accept the new definition of fiber "in the context of studies of health hazards". 3 2063104805

The true background and character of mineralogical concepts and the seemingly complex definition of the asbestos-related mineralogical terminology can be best illuminated through a historical analysis of the relevant terms and expressions. That will be attempted in the following pages. Historical Review Asbestos 9n history. Asbestos is probably the most unique substance in the mineral kingdom. To begin with, it does not even look like a stone, but looks more like some organic wool or cotton. Good quality asbestos is more elastic than other minerals and its high tensile strength is unique. Asbestos is not only stronger than organic fibers but it is also more durable, is fireproof, and for all practical purposes, amphibole asbestos is chemically inert. The peculiar properties of asbestos have attracted the attention of man throughout history. In early times the use of asbestos was restricted either to the households of powerful and rich royalties or to special geographic areas. There are records that Egyptians, Greeks, Romans and even earlier civilizations had knowledge of asbestos and used it for special purposes. The Eqyptians sometimes used coarse asbestos cloth to protect the embalmed bodies of Pharaohs from the ravages of time. The Romans made cremation wrappings to collect the unspoiled ashes of emperors. The lamps of the Vestal Virgins were furnished asbestos wicks which lasted forever. There are also some questionable records that the Romans threw asbestos and other toxic substances in the river flowing through the besieged city of Auxium, in order to break the resistance of the defendants. According to legend, Charlemagne had an asbestos tablecloth which he threw in the fireplace after dinner for the purpose of cleansing it, to the amusement of his company (fig. 1). Figure 1. Ge Boot's (6] illustration of the fire-proof property and the making of asbestos cloth. As it can be expected, in addition to the practical uses there were some less logical and more mysterious applications of asbestos in early and in superstitious civilizations. In medieval times, for example, asbestos was used as a major ingredient in an ointment (fig. 2) intended to cure a number of diseases. Loosely translated De Boot's prescription reads: "Multiple application, miraculous asbestos ointment for juvenile tinea (head-fungus?) and shinbone (skin?) ulcer. Take 4 oz. asbestos, 12 oz. lead (oxide?), 2 oz. zinc oxide, and calcinate, thereupon pulverize into glass while adding vinegar, and agitate it daily for a month; after a month boil it for a quarter hour and let it cure until it becomes clear; thereafter add some vinegar, mix it with rose-petal oil until it becomes a homogeneous ointment: then go and smear it over the infant's head, to promote healing: for itches and shinbone ulcer smear it over the 4

affected area in the evening, for healing. The same mineral, mixed with aqua vitae and bamboo syrup, when applied in small quantities in the morning will sooth the pain of female white-menstruation (leukorrhea?), and will soon heal." Aa,a.. £t Amianta linimeutamadtrineampuerorum, 6!•s.r• & ad uleera tibiaru miraculoCum fit Cequenn mo- '"i^"" do. Accipiuntur Amiand une.quatuora plumbi uncix ta,mtizuncizdue,aculcinancur, deinde srmrsei. pulverifauinvitromaurantureumaeeco,acquo- tidicpermenCemmateriaagicamrfemoapo(tmen; fcm ebulliendaoR unius harm quadrante, ac quie- fcere fudtut, donec indarefcar. deinde illius ueti clariquandras, cumpari.quantirateolei roGcei. miCacmr,donec bona 6atunio Hnimentiforma: co innngiturceput pncrf tomtn ut eiro fanemc ad fea- bicm, & ulcera tibiarum refperi pucee unguntur, .rd.ln- donee Ganentur. Si laapis hie cum aqua vit.T, & Cae-t& :haro Colvatur,ac exigea portio mane quoridiemu-e f~tla licri albo monRruo laboranti decur,moc fanacur. mt,~a# Figure 2. De Boot's prescription for the miraculous asbestos ointment [6, p. 384-5]. The industrial revolution opened an era yielding rich rewards for imaginative inven- tions. Asbestos, as other unique minerals, did not escape attention and a large number of applications were discovered. Some of these were practical and were adopted, such as fire- proof suits and other products (fig. 3). Some, on the other hand, were not well received by the public, like the refillable asbestos cigarette paper introduced in England during the 1880's [38]. If Figure 3. Illustration of some early asbestos products, Jones [38]. 5

Actually, industry was rather slow in adopting asbestos. Even after the discovery of the extensive and high-quality Canadian chrysotile deposits, asbestos-industrialists spent more time promoting their product than manufacturing it,7 at least for a few decades. After the turn of the century they began to succeed and asbestos soon became one of the most widely used industrial minerals.- Asbestos in mineratoriv Although there were several references to asbestos in the ancient literature, the first scientific-type descriptions were offered relatively late by Dioscorides [20] and Plinius [55]. Dioscorides called it auiavrco amiantos (meaning immaculate, unpolluted) and Plinius added a comment that the Greeks used to call it aosaatoo, asbestos (meaning incombustible, unquenchable, inextinguishable). Plinius also used the Latin name of linum vivum for the same mineral as he believed it to be a plant from India; a plant which grew in a part of the earth burned completely by the sun, thus became accustomed to that environment and learned to survive in the flame of fire. During the scientifically dormant Middle Ages the nomenclature of Dioscorides and Plinius was neither challenged nor modified. Of the two names of asbestos, amiant appeared to be the more popular. Almost two centuries ahead of the era of the scientific revival, Agricola [1] offered in 1546 the first criteria for mineral identification. According to Werner [66], Agricola recognized several basic categories of mineral properties such as color, transparency (translucida), resplendence (fulgor), luster (mitor), weight (gravitas), hardness (durities), flexibility (flexibilitas), cleavage (fissio), etc., and used descriptive terms as: globular (figura globi), cyclindrical (figura cyclindrica), conical (figura metae), hair-like (figura capillorum), star-like (figura stellarum), etc. It is easy to recognize in Agricola's expressions the prototypes of some modern terms. Although he did not expand our knowledge of asbestos, he did introduce the descriptive term capillary (haarf6rmig, hair- like) which was adopted later for the description of the shape of asbestos fibers. The next stage of development in mineralogy was during the 18th century when scientists began the development of a mineralogical system divided into orders, classes and species on the basis of common and distinct external properties. These followed the natural history concepts and criteria used in botany and zoology. The first and still relatively crude classification was offered by Walerius [65] and by Cronstedt [15] and the first significant improvement of Agricola's list of external characters of minerals, and its application in mineral classification, was offered by Linneaus [47]. The term fibrous (fibrosum) appeared in his list of descriptive terms for minerals composed of parallel fibers. This period was closed by Werner who published the first comprehensive and consistent system of mineralogtr in 1774 [66]. He exerted unparalleled influence on the future development of mineralogy.8 His influence extended from the wide-spread acceptance of his svstem of mineralogy to the establishment of mining schoois in many countries and to the practice of naming new minerals after personal names. In his classification system, capillary is to be used for the description of asbestos fibers and fibrous is used to describe the breakage of bundles of fibers into small fibers.e He constructed a complete system of minerals of about 300 species. Although never published, it was spread by his students and fellow m neralogists [34,26], and was.adopted all over the 7Jones' book [38] may have been inspired by similar interests. He writes [38, p. VI] that "he hopes by this means (writing the book) to ... tend to develop the uses of" asbestos. °Werner used the name orvctoanosy for determinative mineralogy, after the Greek opvtoo (fossil) and yvaota (to know). sNote that there is very little difference between saying that (1) fibrous crystals grow in bundles of fibers (as we would say it today) and (2) a bundle of fibers breaks down to fibrous crystals. However, the use of fibrous as a description of breakage was soon changed to a description of texture or habit by &lum (5, p. 30], Thomson [61, p. 2567, Phillips [54, p. XXXVI and LXXII], etc. 6

civilized world, even before his death in 1817. In his system he recognized one asbestos species, with four subspecies, and two other [26] subspecies, asbestartioer or asbestarticher, one of actinolite and one of tremolite.so Werner recognized a number of other fibrous, but not "asbestartiger" mineral varieties. He called those strahliger or fasriger. Jameson [36,37] translated Werner's terminology into English as asbestous, rism tic, and fibrous,x' respectively for asbestartiger, strahliger, and fasriger. The comparison of the appropriate portion of Werner's and Jameson's classification are given in Table 1. Table 1. Comparison of Warner's and Jameson's classification of asbestos and some fibrous minerals. Werner, after Fretesleben [26] Jameson [36] ERSTE KLASSE: ERDLICHE FOSSILIEN 3. Kiesel Geschlecht Spischaft des Pistazits 52. Anthophyllite a. strahlicher Sipschaft des Zeoliths 75. Prehnit a. fasricher 77. Zeolith b. Faser - Zeolith 5. Talk Gaschlecht Sipchaft des Talks CLASS II. ORDER VI. SPAR Genus I. Schiiier Spar S. Prismatic Schiller Spar or Anthophyllite Genus IV. Prehnite 1. Axotomous Prehnite 2d Subsp, Fibrous Genus IV. Zeolite 7. Prismatic Zeolite lst Subsp. Fibrous Genus VIII. Augite 137, Asbest 2. Hemiprismatic Augite a. b. Bergkork Amianth 4th Subsp. Actynolite lst Kind Asbestous c. d. gemeiner Asbest Bergholz 2d 3d -- Common -- Glassy Sipschaft des Strahlsteins 5th Subsp. Tremotite 138. Strahlstein 1st Kind Asbestous a. asbestarticher 2d -- Common b. gemeiner 3d -- Glassy c. d. giasicher kbrnicher 6th Subsp. Asbestus lst Kind Rock-Cord 141. Tremolit 2d -- Flexible Asbestus a. asbestarticher 3d -- Common Asbestus b. gemeiner 4th -- Rock-Wood c. glesicher 'oThe actual names of classification units vary from author to author. In order to avoid the lengthy comparison of the expressions used by different authors only "species", "subspecies", or "varieties" are used in the text (instead of "subspecies" and "kinds" of Jameson, for example). "Phillips and Allen [54, p. LXXII] used asbestiform as well as a special term fasciculated for minerals composed of fibers or acicular crystals occurring in bundles. 7 2063104809

i Jameson's asbestous was soon changed to asbestiform in English mineralogy textbooks: Thomson in 1836 61, L, p. 22;, Phillips and Allan in 1838 [54, p. 58] and QanaSZ in 1857 [17, p. 153]. Ha"uy [36] also adopt2d Werner's basic system and terminology, and translated most of his German terms into French, although he practiced more flexibility than Jameson did as he introduced a more chemical classification scheme. However, he makes no apparent distinction between asbestos and other fibrous varieties and uses the term fibreux for all. He trans- lated Werner's asbestartiger Strahlstein and Tremolit as Actinot~breux and Grammatite fibreuse,13 fasriger Prehnite as Prehnite fibreuse, and straliger Antophyllit as anthophyl- lite aciculaire. Although he does not distinguish between asbestos and other fibrous crystals, he seem to restrict the use of fascicle, and to a lesser degree fibre, to asbestos fibers. The term filamenteux was introduced into the French mineralogical 7iterature by Brard [7], asbestoid by Beudant [4, p. 369] and asbestiforme by Cloizeaux [13, I, p. 80] as equivelent expressions for the German asbestartig and the English asbestiform. It should be noted that all these early mineralogists, including Werner and his followers, used the term fibrous in a general sense and considered asbestiform (asbestous, asbestartig, feinfaserig, asbestoid) as a special class of fibrosity. Although none of them have defined the uniqueness of asbestiform fibrosity, the reason for that distinction was implied in their recognition of the unique properties and appearance of asbestos, including the unusual strength of asbestos fibers. Hoffmann (and Breithaupt) [33, IIb, p. 307], for example, pointed out that the asbestiform variety of tremolite is less brittle, that is, stronger than the common prismatic or acicular variety. i r t Figure 4. Handcolored illustrations of Schilletnder Asbest (Amianth) and gemeiner (comewn) Tremolit in Schmidt's Mineralienbuch [60]. i2At the bottom of this page Dana gives some exercise questions like: What is the crystal- lization of hornblende? Mention the characters of the varieties of actinolite - (i.e., glassy, radiated, asbestiform, massive). L3occasionally, however, he used the German word "asbestartiger" in the French text without translation. a

Werner's historical system of mineralogy was used without fundamental modifications for over a century, especially in popularized mineralogy books like that of Schmidt's [60]. Werner's strong influence on mineralogy resisted, for some times, the acceptance of the proposals of a new breed of mineralogists who advocated to change the system of mineralogy from the "natural history" type to a more chemical one. Mineralogists like Thomson [61], Beudant [4], Berzelius [57], Rammelsberg [56], and others believed that the chemical properties of minerals are much more important than their external and physical properties. Thomson was a strong opponent of the classification of minerals on the principles of natural history. He was especially critical of Mohs [49] who carried the natural history approach to such an extreme that it almost became free of chemistry. Thomson came out to say [61, p. 8]: "It appears to me, that mineralogy is so closely connected with chemistry, and so dependent on it for its specific distinctions that it would be highly injurous to it, and therefore, very unwise to attempt to deprive it of so important an ally." In line with the emphasis on chemistry came a new classification and the redefinition of mineral species. All those former species which had no distinct chemical composition were discredited. This included the degradation of Werner's one asbestos species to the rank of variety. Of course, asbestiform actinolite and tremolite were already considered variations (or subspecies) by Werner himself. Asbestos and its subspecies became classified, on the basis of relatively poor and inconsistent chemical analysis, as variations of amphiboles, epidote, pyroxenes, talc, and tourmaline by Beudant [4, p. 837], for example. Rammelsberg expressed this philosophy of the reclassification of the former asbestos species" [56, Part II, p. 313] as: "Mineral substances described by the name of asbestos (or amiant) do not appear to constitute an independent species. As their chemical composi- tions indicate, and it may be more appropriate as noted by Breithaupt, that the name asbestos represents a condition which can be obtained by several, thoroughly different kinds of minerals." (emphasis by author) Berzelius, the Swedish chemist-mineralogist (also the major promoter of the "blowpipe analysis" which became one of the major mineralogical techniques for more than acentury), was one of the most ardent pioneer advocates of this "scientific" system of mineralogy. In his 1846 publication [54, p. 213-214] Berzelius states that mineral s ec~ies as previously defined don't exist. He proposed that instead of species, m nerals should be identified and classified on the basis of: "ingredients and different chemical proportions...as well as their definite bonding relationships." (Yerbindungsverhtiltnisse = 2= crystal structure.) The transfer of mineralogy from Natural History to Chemistry did not take place as proposed by Berzelius and his compeers. Instead, mineralogy developed gradually in that direction and assumed a unique position among the sciences, a status of transition between natural history and physical sciences. The concept of species was not fully abandoned either. In fact, with the meaning redefined in a chemical context, "mineral species" is still used today by some mineralogists, like Berry and Mason [3, p. 272-274]. The classification of minerals was also changed during the second half of the 19th century from categories of "common external properties" to groups of chemical units. Mineral species or individual minerals were defined by their chemical composition and crystal structure. Of course, crystal structures were not known at that time. Consequently, they had to be substituted for by the observable consequences of the crystal structure: the crystallography and physical- chemical properties of minerals. That is, if two minerals had the same composition but had different crystallography and physical properties they were considered to be two distinct minerals. That criterion was readily applicable to minerals which occurred in good crystal forms. However, the same could not be used for asbestos where there was not crystallographic Y4In the same book Rammelsberg recognized Krokydolite (crocidolite), named by Hausmann in 1831, as an independent species. That may, at first, look like a contradiction in his philosophy. However, it is not, as crocidolite's parent mineral riebeckite was only discovered many years later, in 1888, by Sauer. 9 2063104811

data and the only non-chemical information available was the difference in the tensile strength and flexibility of the asbestiform versus the compositionally equivalent non- asbestiform mineral. That difference was considered by many mineralogists to be sufficiently distinct to warrant the recognition of some asbestiform varieties as independent minerals. Several dozen asbestos mitaerals were proposed and accepted during this stage of evolution.15 The chemical compositions of most of these asbestos minerals were known and their chemical identity with other minerals were recognized. The compositional equivalence of chrysotile and serpentine was realized since Kenngott's publication [40] in 1853. That was sufficient for some mineralogists to declare chrysotile as a variety of serpentine. Others, however, still considered the differences in physical properties sufficiently significant to recognize chrysotile (under various names, like: metaxite, schweizerite, etc.) and serpentine as two distinct minerals. The other two major asbestos minerals, byssolite and crocidolite, were known to match amphibole compositions, byssolite since Scheerer's 1851 analysis [59] and crocidolite since Delasse's 1847 analysis [i9]. Crocidolite was first believed to be an asbestiform variety of arfvedsonite [50, p. 461] in spite of minor chemical differences. However, as soon as riebeckite was discovered by Sauer in 1888, crocidolite was reclassified by Naumann (and Zirkel) [51, p. 707], as its asbestiform (Asbestform in German) variety. As a consequence of the undecisive significance of differences in physical properties versus compositional identities, the classification of asbestos minerals as independent minerals or as varieties was a function of the individual interpretation of mineralogists. For example, Hintze [317, Groth [297, and Naumann (and Zirkel) [51] recognized chrysotile as a variety of serpentine and crocidolite as a variety of riebeckite; E. S. Dana [16] classified both as independent species; Klockmann [41] and Rogers [58] recognized chrysotile as an independent mineral and crocidolite as a variety of riebeckite. The asbestos nomenclature was further complicated during the last decades of the 19th and first decades of the 20th century when asbestos became a major industrial material. The industrially useful properties of asbestos obtained from certain deposits differed somewhat from that of others, and on the basis of that some asbestos were given distinct mineral names, usually reflecting the name of a mining company or district (for example, bostonite: Boston Asbestos Packing Co.; amosite: Asbestos Mines of South Africa; montasite: Montana mine, South Africa; prieskaite: Westerburg mine, Prieska, South Africa . he use of these distinct mineral names, of course, provided some promotional advantages. The majority of these commercial mineral names never got into mineralogy text books, and those few which did were subsequently eliminated or discredited. Amosite, for example, was formally discredited in 1946 [2]. The discovery of x-ray diffraction produced a tool available for crystal structure determination. As the basic crystal structures of the former asbestos minerals were proven to be identical with that of compositionally equivalent major minerals they were all degraded to the rank of varieties, without further arguments. For example, the final decision on crocidolite's mineralogical identity with riebeckite was provided by Whittaker in 1949 [67] and by Drysdall and Newton in 1960 [22]. The asbestos varieties of minerals were consequently identified by the prefix of fibrous or asbestos-like [24, p. 578] or asbestiform (see Table 2 for details). Fibrous16 was used as a more general term to include both asbestiform and non- asbestiform fibrous minerals. However, asbestiform was always restricted to asbestos varieties, as that was done consistently since Werner s time, 200 years ago. tsThese asbestos mineral names included: Adigenite, agalite, antholite, baltimorite, beaconite, cyclopeite, dermatite, fibrolite, griquanlandite, hydrophite, ishkyldite, karachaite, kolskite, kymantine, metaxite, nemalite, picrolite, retinalite, rezhikite, rhoduzite, schweizerite, vorhausite, williamsite, zermattite, zillerite, xylotite, etc. 1eFord [24, p. 204] gave a more liberal definition of fibrous than usual. He states that "fibres may or may not be separable" in a fibrous mineral. 10

e Z3 Table 2. Descriptive term~ used by mineralogists to distinguish between asbestos and other types of fibrous textures. (Frequently in conjunction with fibrous.) Page number of an example is given. Werner (Friesleben) asbestartich Naumann [50, p. 324] asbestartig [66, p. 107 Naumann (Zirkel) Asbestform HaDy [30] no distinction [51, p. 707] Hoffmann (Breithaupt) asbestartig Nicol [52, p. 152] asbestiform [33, 2b, p. 306] Tschermak [62, p. 444] ? feinfaserig Jameson [36, II, p. 22] asbestous Groth [29, p. 151] asbestartig Phillips (Allan) asbestiform [54, p. 58] E. S. Dana [16, p. 384 asbestiform Thomson [61, I, p. 481] asbestiform Hintze [31, II, p. 1195] ? feinfaserig Mohs (Haidinger) asbestous Klockmann [41, p. 567] ? feinfaserig [49, II, 27] Doelter [21, II, p. 589] asbestartia Beudant [4, p. 387] asbestoide Rogers [58] no distinction Brard [7, p. 206] filamenteux Ford [24, p. 578] asbestos-like Bium [5, p. 242] ? feinfaserig Hurlbut [35, p. 446] asbestiform Rammelsberg [56, p. 358] asbestartig Kraus, Hunt, Ramsdell asbestiform Schmidt [60, p. 358] asbestartig [43, P. 392] Bristow [9, p. 85] asbestiform Berry, Mason asbestiform [3, P. 527] Cloizeaux [13, p. 81] asbestiforme Deer, Howie, Zussman asbestiform J. D. Dana [17, p. 153] asbestiform [18, II, P. 2431 a French: fibreux; German: faserig. The term fiber, in reference to asbestiform fibers, was equivalent to the concept of organic fibers because the early natural historians believed that asbestos was actually a vegetable. Mineralogists from the 18th century on did not specifically state that the term fiber is used because of its resemblance with organic fiber. However, that reasoning is apparent in their description of asbestos fibers as hair-like or capillary or thread-like, and in the types of names they have given to asbestos minerals, such as mineral-wood, rock- cotton, mountain cork, rock-wood. Jones [38] provided extensive details in the description of the similarity between asbestos and organic fibers (fig. 5) and concluded [38, p. 221] that: "The nature of the asbestos fibre is thus so far identical in structures with the organic fibres." - 11 N 0 a w 0 A 00 M+ W

C;w Fg. 1. Bibrv of 9hoo '* p WaaL Magti@°d Fig. 2. Filnmenh B00 of Naw Cattan. di.mchn. No. 1.-TLOtfmd On. Fig. 5. 8pun Ghw. Nn. 7. 13etfo~d O». Figure 5. Jones' [38] comparison of asbestos and organic fibers. Although the use of the term fiber has not been restricted to asbestos and included a number of other minerals they all had some characteristics reminiscent of organic fibers. In any case, the term fiber has never been used as a description of the elongated shape of crystals. For that acicular is the proper mineralogical expression. The term asbestos was first a species name, as noted earlier it was introduced by Werner and his school. Later it became a collective term, like c1Tys or gems, in reference to asbestiform varieties of a number of otherwise unrelated minerals. Parallel with the mineralogical terminology asbestos also became an industrial term for a category of mineral products containing asbestiform varieties of silicates. However, some commercial asbestos may be mixed with non-asbestiform acicular crystals or cleavage fragments. The quality of asbestos is related to: (a) the extent of the development of the preferred asbestos character (high tensile strength, flexibility, length of fibers) of the asbestiform fibers, and (b) the percentage of the less desirable non-asbestiform, acicular crystals or cleavage fragments present in the product. That is, the mineralogical and industrial definitions of asbestos are not fully coincident. The unusual properties of the asbestiform fibers were always recognized by the early users of asbestos as well as by mineralogists. These properties included high tensile strength [for example, 32,33,38,64], increased flexibility (noticed by all mineralogists), unexpected optical properties [for example, 53,69] and differences in surface properties, like surface charges [for example, 28,42,45,72]. VO Fig. 3. Fig. 4. Fibro of p,adint'Chro.d of HsvSilk Gmden Spid.. . 12

.4 With the introduction of high-power electron microscopy, a new tool for mineralogical research and a new area of applied mineralogy was established. Electron microscopes permitted the examination of extremely small mineral particles and the study of the fiber of fibril structures of various asbestiform crystals. The long suspected cylindrical (tubular, scroll-like) structure of the chrysotile fibrils [68] was directly observed by Maser et al. in 1960 [48] and a more detailed record was offered by Yada [70]. In addition to th`e cylindrical fibril structure, Cressey and Zussman [14] reported on a polygonal chrysotile structure which appears to be the dominating fibril structure in the so called "schweizerite" and "Provlen-type" chrysotile varieties. Comparable work, although with less spectacular results, was done on asbestiform amphiboles by several investigators, for example, Chisholm [12], Franco, Hutchinson, Jefferson, and Thomas [25]. The asbestiform amphibole fibril structure appears to be more subtle than that of chrysotile. The increased tensile strength and flexibility may be due to the presence of systematic defects such as faults, dislocations and twinning, and/or to the lack of surface defects. Of course, we know that defects can interfere with the cleavage and fracture of solids and are frequently introduced artifically in alloys and other crystals to enhance their strength as it is elaborated on in the textbook of Kelly and Nicholson [39]. Undoubtedly, time and extensive research will be needed before the structural causes of the unusual properties of asbesti- form amphiboles will be fully explained. Conclusions Several conclusions can be drawn from this review of the history of asbestos-related mineralogical terminology and its current misuse in environmental sciences: (1) Terms such as fiber, fibrous, asbestiform, and asbestos, have distinct meanings in mineralogy whether or not we can offer a complete crystal structural explanation for the development of the properties, reflected by these terms. (2) The asbestos-related mineralogical terminology is adequate and clear, and is not in need of revision. However, its full understanding requires a relatively comprehensive knowledge of mineralogy. Consequently, a set of detailed and unambiguous definitions should be prepared for inter- disciplinary use. (3) The asbestos-related mineralogical terms have been grossly misinterpreted in most of the recent literature of environmental sciences. The implied definitions are inadequate for the description and discussion of the crystal chemical and crystal physical properties of minerals, and endanger the success of coordinated, interdisciplinary studies aimed at the understanding and the solution of the health hazards created by asbestos pollution. The presence of any forei n particle in air and waters in excessive quantities is undesirable and is potentially harmful. It is imperative that all efforts be made to clean up the environment starting with one of the most dangerous mineral pollutants: asbestos. This job requires extensive interdisciplinary cooperation and the establishment of an unambiguous interdisciplinary language. The extensive list of definitions offered in the recent U.S. Bureau of Mines Information Circular [10] are comprehensive and consistent with mineralogical traditions. The adoptation of these definitions for the interdisciplinary language of asbestos studies should be considered. The following definitions of the four most critical asbestos-related mineralogical terms are based on their historical review. 13 0 w ~ 0 A Go ,.r U

FIBER An acicular single crystal, or a similarly elongated polycrystalline aggregate, which displays some resemblance to organic fibers. Examples for criteria of "resemblance to organic fibers" are: circular cross section, `flexibility, silky surface luster, axial lineation, threaded appearance, etc. Most of these fiber characteristics cannot be observed at electron-microscopic scale. Consequently, any elongated particle may be called a fiber (when fiber used as a shape-descriptive expression) provided that it displays parallel edges and apparently equidimensional cross section. That is, elongated triangular-shaped or irregular particles cannot be considered to have the shape of a fiber. FIBROUS The descriptive term used for a mineral which is composed of parallel, radiating or interlaced aggre9ates offbers, from which the fibers are usually separable. That is, the crystalline aggregate may be referred to as fibrous even if it is not composed of separable fibers, but has that distinct appearance. ASBESTIFORM A special type of fibrous habit in which the fibers are separable, and are more flexible and osp sess higher tensi~ strength than crystals in other habits of the same mineral. Increased flexibility and higher tensile strength are, apparently, the most distinct qualities of asbestiform fibers. These properties are undoubtedly due to certain structural variations and can justifiably be included in the definition. ASBESTOS A collective mineralo ical term which includes the asbestiform varieties of various si icate minerals. lT•T$ - The justification for restricting asbestos to silicate minerals may be questionable from the mineralogical point of view, as non-silicate minerals may also crystallize in fibrous habit and the fibers may possess asbestiform properties. However, these properties are expected to be different in magnitude from those of the asbestiform silicates and, therefore, from the health study's point of view, are justifiably excluded from the category of asbestos. The dl nt of fibrous habits must be due to certain unusual conditions which existed at the time of the mineral's crystallization. These conditions may be accompanied by structural modifications and by consequent changes in the mineral's physical properties. These changes, however, are usually not as conspicuous as they are in silicate asbesti- form fibers. In fibrous gypsum, for example, the only readily observable change is in the mineral's fracture pattern. The usually absent ((111)) cleavage plane is perfect in fibrous gypsum and is responsible for its acicular rather than platy fragments. This change in the cleavage pattern is probably due to some structural modification. On the other hand, the conditions of crystallizations may be such that no change in the mineral's structure and properties is necessary. For example, if a fibrous mineral is altered to another, the new mineral may show pseudomorphic fibrous appearance. Dana [16, p. 678] believes that the appearance of fibrous talc is due to its alteration from enstatite. laThe industrial quality of asbestos depends, in part, on the degree of development of the asbestiform fiber structure in the mineral. That is, if more crystals have the scroll- like structure in chrysotile, or the crystals have higher density of defects or twinning in asbestiform amphiboles or have fewer surface defects, the asbestiform fibers are stronger and more flexible, and thus they are more desirable. A similar relationship may exist between the degree of development and the density of asbestiform fibers in the bundles, and their biological activity. That is, the gradation of asbestiform development in a mineral, from acicular cleavage fragments to asbestiform fibers, may constitute dif- ferent health hazards. , 14

The critical reviews offered by Drs. Edwin Roedder and James H. Stout are acknowledged with appreciation. This study was in part supported by a grant from the Regional Copper- nickel Study, Environmental Quality Council, Minnesota. References [1] Agricola, Georgius (Bauer), De Natura Fossilium, (Basilea, 1546). [2] Amosite discredited, Am. Mineral. 34, 339 (1946). [3] Berry, L. G. and Mason, Brian, Mineralogy, (Freeman and Co., San Francisco, 1959). [4] Beudant, F. S. , Traite Elementaire de Mineralogie, (Verdipre, Paris, 1824). [5] Blum, J. R. , Lehrbuch der Oryktognosie, (G. Schweizerbart, Stuttgart, 1833). [6] Boot, Anselmus Boetius, de, Gemmarum et Lapidum Historia, (Lugduni Bataborum, 1647). [7] Brard, C. P., Nouveau elemens de Mineralogie, 2e ed., (Meguinon-Marvis, Paris, 1824). [8] Bradt, R. C., Newnham, R. E. , and Biggers, J. V., The toughness of jade, Am. Mineral. 58, 727-732 (1973). [9] Bristow, H. W. , Glossary of Mineralogy, (Longman et al. , London, 1861). [10] Campbell, W. J., Blake, R. L., Brown, L. L., Cather, E. E., and Sjoberg, J. J., Selected silicate minerals and their asbestiform varieties; mineralogical definitionand identification-characterization, (U.S. Bureau of M nes. Int. Circ. No. 8751, 1977). [11] Champness, P. E., Lorimer, G. W., and Zussman, J., Fibrous cummingtonite in Lake Superior: Discussion, Canad. Mineral, 14, 394 (1976). [12] Chisholm, J. E., Planar defects in fibrous amphiboles, J. Material Sci. 8, 475-483 (1973). [13] Cloizeaux, A., des Manuel de Min€ralogie, 2 vols., (Dunond, Paris. 1862). [14] Cressey, B. A. and Zussman, J. , Electron microscopic studies of serpentinites, Canad. Mineral. 14, 307-313 (1976). , [15] Cronstedt, A. F., FBrsoktill en Mineralogie, eller Mineral Rikets Uppstdlning, (Stockholm, 1758). [16] Dana, Edward S., The System of Mineralogy of James Dwight Dana, Descriptive Mineralogy, 6th ed.,(J- . Wiley and Sons, New York, 1914)-. [17] Dana, James D., Manual of Mineralogy, (Henry H. Peck, New Haven, 1857, new ed. (1871). [18] Deer, W. A., Howie, R. A., and Zussman, J., Rock-Forming Minerals, 5 vols., (J. Wiley and Sons, 1962, 1963). [19] Delasse, A., (analysis of crocidolite) Compt. Rend. 44, 766 (1847). _ [20] Dioscorides, P., IIepl vans 1ezplKns (Materia Medica), (Approx. 50 A.D.). [21] Doelter, C., Handbuch des Mineralchemie, 3 vols., (T. Steinkopff, Dresden and Leipzig, 1912, 1914, 1918). [22] Drysdall, A. R. and Newton, A. R., Blue asbestos from Northern Rhodesia and its bearing on the genesis and classification of this type of asbestos, Am. Mineral. 45, 53-59 (1960). 15 2063104817

[23] Engleston, T., A Catalogue of Minerals and 5 ny onyms, (J. Wiley and Sons, New York, 1892). [24] Ford, W. E. , Dana's Textbook of Mineralogy, 4th ed., (J. Wiley and Sons, New York, 1951). [25] Franco, M. A., Hutchinson, J. L., Jefferson, 0. A., and Thomas, J. R., Structural imperfection and morphology of crocidolite (blue asbestos), Nature 266, 520-521 (1977). [26] Freiesleben, J. C., A. G. Werner's Letztes Mineral-System, (Craz and Gerold, Freyberg and Wien, 1817). [27] Gary, M. , McAfee, R. , Jr., and Wolf, C. L. , (editors), Glossary of Geology, (Am. Geol. Inst., Washington, D. C., 1972). [28] Great Lakes Research Advisory Board, Asbestos in the Great Lakes Basin, (February, 1975). [29] Groth, P., Tabellarische Ubersicht des Mineralien, 4te Aufl., (Friedrich Vieweg und Sohn, Braunschwe g, 1898 . [30] HaOy, M. L'Abbe, Traitd de Mineralogie, 5 vols., (Bachelier, Paris, 1801, 2e edition, 1822). [31] Hintze, C., Handbuch der Mineralogy, Bd. 2, (Von Veit, Leipzig, 1897). [32] Hodgson, A. A., Fibrous silicates, Lect. Ser. 1965 No. 4, (The Royal Inst. of Chem., 1966). [33] Hoffmann, C. A. S., Handbuch der Mineral6 ie, (after Bd. 2a by A. Breithaupt), 4 vals., (Craz and Gerold, Freyberg, 1811-1818 . [34] Hoffmann, C. A. S., (about Werner's Mineral System), 17th ed., 80rgm. 369 (1789). [35] Hurlbut, C. S., Jr., Dana's Manual of Mineralogy, (J. Wiley and Sons, New York, 1959). [36] Jameson, Robert, System of Mineralogv, (Archibald Constable, Edinburgh, 1816). [37] Jameson, Robert, Manual of Mineralogv, (Archibald Constable, Edinburgh, 1821). (38] Jones, R. H., Asbestos, its Properties, Occurrence and Uses, (Crosby Lockwood, London, 1890). [39] Kelly, A. and Nicholson, R. B., Strengthening Methods in Crystals, (Appl. Sci. Publ., London, 1971). [40] Kenngott, Adolf, Das Mohs'sche Mineral System, (C. Gerold, Wein, 1853). [41] Klockmann, F., Lehrbuch der Mineralogie, (F. Enke, Stuttgart, 1903). [42] Kramer, J. R. , Fibrous cuamingtonite in Lake Superior, Canad. Mineral. 14, 91-98 (1976). (43] Kraus, E. D., Hunt, W. F., and Ramsdell, L. S., Mineralogy, (McGraw-Hill, New York, 1959). [44] Kryvial, R. J., Wood, R. A., and Barrett, R. E., Identification and assessment of asbestos emissions from incidental sources of asbes os, (U.S. Eriv. Prot. Agency,: €PKTbT2-7 -iT,-WasTingtonep't. 19T4j- [45] Langer, A. M., Approaches and constraints to identification and quantitation of asbestos fibers, env. Health Presp. 9, 133-136 (1974). 16

Cia [46] Light, W. G. and Wei, E. T., Surface charge and asbestos toxicity, Nature 265, 537- 539 (1977). [47) Linneaus, C. A., 5 sy tema Naturae, (Holm, 1768). [48] ~Maser, M., Rice, R. V., and Mug, H. P., Chrysotile morphology, Am. Mineral. 45, 680-688 (1960). [49] Mohs, Frederich, GrUndriss der Mineralogie, 2 vols., (Dresden, 1822). English: Treatise on Mineralogy, Transl. by W. Haidinger, 2 vols., (Hurst and Robinson, London, 1825 . ' [50] Naumann, Carl F., Elemente der Mineralogie, 8te Aufl., (W. Engelmann, 1871). [51] Naumann, Carl F., Eiemente der Mineralogie, 13te Aufi. by Ferdinand Zirkel, (W. Engelmann, 1898). [52] Nicol, James, Elements of Mineralogy, (A. and C. Black, Edinburgh, 1873). [53] Peacock, M. A. , The nature and origin of the amphibole asbestos of South Africa, Am. Mineral. 13, 241-285 (1928). [54] Phillips, William, An Elementary Introduction to Mineralogy (1823), 4th ed. by R. Allan, (Longman, et al., London, 1837). [55] Plinius Secundus, C., Historia Naturalis, 27 books (77 A.D.) [56] Rammelsberg, C. F., Worterbuch des Chemischen Theils der Mineralogie, (C. G. Luderitz, Berlin, 1841). (57] Rammelsberg, C. F. , J. J. Berzelius' Neues Chemisches Mineral System (J. L. Schrag, NUrnberg, 1847). [58] Rogers, A. F. , Introduction to the Study of Minerals, (McGraw-Hill, New York 1937). [59] Scheerer, Th., (analysis of byssolite) POgg. Ann. 84, 389 (1851). [60] Schmidt, F. A., Mineralienbuch, (Krais and Hoffman, Stuttgart, 1855). [61] Thomson, Thomas, Outline of Mineralogy, Geology, and Mineral Analysis, 2 vols., (Baldwin and Cradock, London, 1836). [62] Tschermak, Gustav, Lehrbuch der Mineralogie, (Alfred Holder, Wein, 1884). [63] U.S. District Court, District of Minnesota, 5th Division, Supplemental Memorandum, (No. 5-72, Civil 19, Appendix 5, Judge Miles Lord, May 11, 1974). [64] Vermaas, F. H. 5., The amphibole asbestos of South Africa, Trans. Proc. Geol. Soc. South Africa 55, 199-229 (1952). [65] Walerius,I. A., Mineral-riket, (Hotm, 1847). [66] Werner, A. F. , Ausserlichen Kennzeichen des Fossilien, (Leipzig, 1774), English: On the external character ransl A. V. Carozzi (U of I11.. Press, 1962). [67] Whittaker, E. J. W., The structure of Bolivian crocidolite, Acta Crystallogr. 2, 312- 317 (1949). [68] Whittaker, E. J. W., The structure of chrysotile, Acta Crystallogr, 6, 747 (1953). [69] Wylie, A., Optical properties of asbestiform amphiboles and their non-asbestiform analogues, U.S. Bureau of Mines, I.R. (in press). 17 2063104819

C7 [70] Yada, Keiji, Study of chrysotile asbestos by a high resolution electron microscope, Acta Crystallogr. 23, 704-707 (1967). [71] Zoltai, T. and Stout, J. H., Comments on asbestiform and fibrous mineral fragments (Minn. Poll. Contr. Agency, 1935 W. Co. Road B-2, Roseville, MN 55113, 1976). [72] Zoltai, T., Veres, I., Wagner, M. J., and Hammer, R. F., Surface charges of asbestiform amphibole fibers, (in manuscript). Discussion M. COSSETTE: Could you tell me if the use of the word asbestoid implies that it is not quite asbestos? T. ZOLTAI: Brard and Beudant used it in lieu of asbestartich or asbestiform, that is, the expression is equivalent to asbestiform. A. BOHMER: Are you suggesting that if a mineral has an asbestiform habit in its varieties and it has a three-to-one ratio it is asbestos? That is, should we limit our classification of asbestiform to those minerals? ZOLTAI: Yes. 18

C 5 National Bureau of Standards Special Publication 506. Proceedings of the Workshop on Asbestos: Definitions and Measurement Methods held at NBS, Gaithersburg, MD, July 18-20, 1977. (Issued t,ovember 1978) FIBROUS AND ASBESTIFORM MINERALS James R. Kramer Department of Geology McMaster University Hamilton, Ontario LBS 4M1 Canada Abstract Asbestiform minerals may be differentiated from other elongate minerals by comparing their length and aspect ratio distributions in the greatest percentile level. Individual fiber analyses of UICC and other well-characterized samples suggest a possible 20-40 percent intensity ratio variation relative to Si of major cations. There is a very small amount of evidence to suggest that fibers other than asbestos are toxic. Key words: Acicular; asbestiform; asbestos; elemental composition; fibers. Introduction I The ability to differentiate between acicular minerals, fibrous minerals, and asbesti- form minerals is most significant to the work of analysts, health researchers, and mineralogists. Zoltai [21, p. 13-31]1 defines the terms carefully, discusses the history of the relevant terms, and shows how the discrepancies in the use of the terms found today has evolved. In short, the differentiation of the terms asbestos, asbestiform, fiber, fibrous, and acicular has been obscured in many cases and in different applications. One reason for a large part of the overlap of the usage of the terms is the difficulty in separation of one term from another on an analytical basis at the scale of the transmission electron microscope. When enumerating elongate particles at the micrometer scale, in many cases a cleavage fragment can appear similar to a fiber or an asbestiform mineral. Thus the advent of the transmission electron microscope to identify and enumerate particles on an environmental monitoring basis has brought certain ambiguities. The Glossary of Geolo [1] defines some of the pertinent terms as follows: "ASBESTOS: (a) a commercial term applied to a group of highly fibrous silicate minerals that readily separate into long, thin, strong fibers of sufficient flexibility to be woven, are heat resistant and chemically inert, and possess a high electric insulation, and therefore are suitable for uses (as in yarn, cloth, paper, paint, brake linings, tiles, insulation cement, fillers, and filters), where incombustible, nonconducting, or chemically resistant material is required. (b) a mineral of the asbestos group, principally chrysotile (best adapted for spinning) and certain fibrous varieties of amphibole (ex. tremolite, actinolite and crocidolite). (c) a term strictly applied to the fibrous variety of actinolite. syn: asbestus, amianthus, earth flax, mountain flax. ASBESTIFORM: said of a mineral that is fibrous, i.e., that is like asbestos. ACICULAR: (cryst) said of a crystal that is needlelike in form. cf. fascicular, sagenitic. 'Figures in brackets indicate the literature references at the end of this paper. 19 N O ~ w ~ 0 a m N H+

FIBROUS: said of the habit of a mineral, and of the mineral itself (e.g. asbestos), that crystallizes in elongated, needlelike grains or fibers." The more restrictive definition of asbestos (c) is not presently used. Thus asbestiform is a restricted usage of fibrous pertaining to asbestos. In the general field of mineralogy, asbestiform has not been a' commonly used term. Taken as a whole, one can easily imagine that overlap at an analytical level among the definitions of acicular, asbestiform, and fibrous definitions could occur. In a bulk sample, the distribution (bundle, fibrils, splitting), fiber length, and concentration of fibers would be used to distinguish between asbestiform and fibrous in most cases. Acicular would be distinguished from fibrous and asbestiform in that the properties of a fiber (flexibility, bundles, splitting) are not present. When minerals are dispersed, occur separately and are examined at the micrometer scale, the distinguishing characteristics for these terms disappear or are highly obscured. At this microscopic level, it is most difficult to distinguish among cleavage fragments, acicular minerals, and fibers. In no cases, however, are cleavage fragments considered to fall into the definitions of asbestiform, fibrous, or acicular. In another discussion of asbestiform [20, p. 19], it is considered "a type of mineral fibrosity in which the fibers and fibrils possess high tensile strength and flexibility." Spiel and Leineweber [18] point out that all asbestos minerals have overlapping tensile strengths; and methods of measurement are difficult "with large variations in results using the same and different techniques." Furthermore, there are virtually no tensile strength data on other fibers and cleavage fragments. Flexibility is related to the "harshness" or flexural modulus of fibers [18]. It is not clear what differences exist between asbestiform fibers and cleavage fragments of amphiboles. There is found considerable variation in the flexural modulus of chrysotile which may be due to the water content, mineral impurities, or orthorhombic and monoclinic crystal forms in the fibers. Another approach to obtain a working definition and differentiation of asbestos fibers and other elongate (length/width > 3) mineral fragments is to consider the definitions in terms of their health significance. Length and aspect ratios within certain defined limits have been proposed as the only important mineral parameters to be considered in respiratory disease. If one accepts this argument with no additional caveats, one could easily extend the length factor considerations to any elongate particle provided that the length and length/width criteria are met. This argument would then demote the analytical differentiation of the terms to a mineralogical wrangle; furthermore, there would be little necessity to distinguish among the various minerals in most cases. Following the extension of the length argument further, one then becomes faced with the conclusion that many minerals commonly occurring in rocks and soils on the earth's surface would be considered a health risk. Cralley [8] suggested that the ubiquity of occurrence of elongate mineral and non- mineral particles in autopsies may be related to the ubiquity of occurrence in the modern environment. He suggests that variable response in the lung may depend upon the chemical and physical characteristic of the fibers, but he does not state what specific characteristics should be studied. One might therefore conclude that all, or certain sizes of, elongate particles might be considered with variable response in the lung depending upon the mineralogy and surface properties. Lists of some of these common fibrous or acicular minerals are given in Kramer [12] and Zoltai [20]. There is very little epidemiological, animal, or cytotoxicity data on elongate and fibrous minerals other than asbestos. Table 1 summarizes the results obtained for studies on elongate/fibrous minerals other than asbestos from searching TOXLINE, MEDLINE, and Chemical Abstracts for the past few years. Almost all of the few elongate/fibrous minerals tested showed some toxicity, and there is some suggestion for endemic lung conditions related to soils. Many equidimensional minerals were not active or as active as the elongate/fibrous minerals in hemolytic studies. Almost all of the minerals tested were silicates, so it is not possible at present to generalize to all minerals. 20

Table 1. Toxicity of fibrous minerals other than asbestos. Mineral System and effect Reference General soils and endemic pleural plaques 13 Sepiolite-palygorskite increased enzyme activity 11 lactic acid Inhibition hemolytically active 17 endemic pleural calcifications 4 and soils tumors in rats, i.p. injection 15 Amphiboles amphibole in soil and 5 pleural plaques Arfvedsonite i.p. carcinogenicity in rats 16 Vermicullite i.p. carcinogenicity in rats 10 Apatite-nepheline dust effects 3 i.t. effect, rat lungs 9 Talc (tremotite) hemolytically active 17 Nemalite hemolytically active 17 Gypsum allergic reactions 14 chronic bronchitis 14 There is no specific information on the nature of the surfaces of the minerals, except that in one study of Schnitzer and Pundsack [17], the hand cut specimens of asbestos and other fibrous minerals were not hemolytically active. Interestingly, amphibole asbestos is not hemolytically active. However, there are very little data available to arrive at any definitive conclusions. In addition, Webster [19] has noted that in animal studies with monkeys, non-fibrous nepheline dust has produced interstitial fibrosis. This suggests that other factors besides fibrosity are responsible for the development of fibrosis. In summary, there is difficulty at the sub-micrometer level to differentiate asbesti- form, fibrous, and acicular minerals. Furthermore, there is no health evidence which might be used in an alternate classification of elongate particles. The relative response of different fibrous minerals is not clear. Since definitive animal studies and epidemiological information exist for asbestos minerals only, it is pertinent to investigate parameters which might be used to differentiate between asbestos minerals, other fibers, and cleavage fragments found in the environment. Length and aspect ratio distributions are examined for occupational asbestos samples and for environmental samples, and the composition of fibers and intra-fiber composition are examined to ascertain variations within a sample. N 21 a 0 ~ N W

Fiber Morphology The fiber length and often the fiber width are characterized in virtually all toxicity studies, and the length is often considered the most important factor in health aspects of fibers. More often than not, the mass-median length or median length is stated in reports of research. It is not uncemmon for the median length of occupational exposures to coincide with the median length of environmental measurements. Median length does not, however, provide information of the entire length distribution. Therefore, it is worthwhile to consider what variations if any exist for the entire length distribution of fibers measured in an occupational exposure and in an environmental exposure. Figure 1 summarizes length data for both chrysotile and amphibole fibers. Figure la compares the length distribution of 300 environmental samples of fibers measured in air and water environments by this laboratory to the length distribution of UICC amosite and chrysotile measured by this laboratory and to surface and underground mine dusts compiled from du Toit [9]. Figure lb compares fiber tailings from Lake Superior to UICC amosite analysis and to occupational exposures fram du Toit, and figure lc compares the distribution of the longest chrysotile sample measured in urban air in Ontario to UICC and occupational measurements of du Toit. All three cumulative length plots show that distributions for occupational exposure converge with environmental distributions at the 50 percentile level and that the fiber length from occupational exposures are greater than that from environmental exposures at the 99 percentile level. 50 95 99 g9,g Figures la and lb. (caption on the next page). 22

C 5 Figure 1. Oistribution curves showing the difference in asbestos length and other fiber length. (a) About 600 analyses of elongate fibers compared to length distri- bution of UICC samples and from underground (U) and surface (S) dusts from South African Mines. (b) Comparison of amphibole in taconite tailings from Lake Superior to UICC amosite and surface and underground mine dust from South Africa. (c) Comparison of UICC chrysotile and surface and underground mine dusts with sample from atmospheric environment containing largest fibers. See figures lb and lc for figure la labelling of individual distributions. Occupational and environmental samples show a broad length distribution over almost three orders of magnitude, and the only apparent differences in the length distributions are for the longest fraction. Therefore, characterization of the entire length distribution is mandatory for all studies. Campbell, et al. [6, p. 44 ff] have carried out a similar analysis for aspect ratios. They show a great deal of overlap of aspect ratio for the milled asbestos form and the milled non-asbestos form of anthophyllite and tremolite. Furthermore, they show a distinct difference for a commercial milled chrysotile sample and an ambient air sample. In both cases, the distributions overlap, but the milled asbestos form has a small distribution of large length/width aspect ratios that is not found in the milled non-asbestos form, and the commercial milled chrysotile has a small distribution of larger aspect ratios that is not found in the ambient air sample. The aspect ratio distribution of hornblende is very similar to the aspect ratio distribution of the non-asbestos amphiboles. The difference in the aspect ratios between milled asbestos and milled non-asbestos minerals is found for the upper five percent or less. This difference in aspect ratio parallels the difference in length distributions of the largest percentile discussed above for occupational and environmental samples. In fact, the very large aspect ratios would be measured on the fibers of largest length. It may well be that the differences in aspect ratio and of length of the longest fibers will be most significant in health studies. Figure 2 shows the morphology of six different samples of cummingtonite-grunerite from the Wabush Lake, Labrador, area. The bulk composition and the mineralogy are the same for all six samples, and all of the samples were taken within about 500 meters of each other. Figure 2a is clearly an asbestiform sample and figure 2f is clearly equidinensional. The detailed morphology of these samples may show some significant toxicological differences. They are now being studied in detail .ineralogically, and for hemolytic and cell activity responses. 23 0 a w r+ 0 ~ N U

a IIF Mf 4 I5 R A: d r. i~ • a ~ ii ~ • ~ ,,. .~ ,• S• ~ ~ FA Figure 2. Asbestiform and equidimensional cummingtonite-grunerite from Labrador. (a) Asbestiform cummingtonite scale units are in cm. (b) - (f) Variations in fibers, cleavage fragments, and equidimensional cummingtonite-grunerite sampled within 500 meters of each other and the asbestiform variety. Each numbered scale unit is 0.1 cm. fm . 24

CS Fiber Composition Asbestos and other fibers vary in major element composition due to the substitution of octahedral coordinating cations (typically Mg, Fe2 , Fe3 , Al), tetrahedrally coordinated cations (Si, Al), and coordination of larger cations (Ca, Na, K). Chrysotile is a silicate sheet structure of nearly fixed composition, Mg3Si20s(OH)*, but the amphibole asbestos minerals show more substitution of ma~or ions. Hsnce the anthophyllite-gedrite series develops with substitution of Mg for Fe2 , Al for Fe2 , and Mg with substitution of Al for Si makp the charge balance; the cummingtonite-grunerite series with supstitution of Mg for Fe2 ; the tremolite-actinolite series with substitution of Mg for Fe2 and substitution of Fe3+ for Al. "Amosite" is an asbestos acronym for a cummingtonite-grunerite of variable composition, and crocidolite is the asbestos variety for a glaucophane-riebeckite of variable composition. There are other less common amphiboles With asbestiform habit. In addition, there is substitution of trace elements and in some cases other elements (for example, Mn) may substitute in the amphibole structure to a large extent. Therefore, one may not conclude that there is any fixed composition for one asbestos mineral, and it is possible to have variations in composition within one sample depending upon the history of formation of the mineral. In addition, other asbestos minerals and other mineral impurities can and do often occur in asbestos samples. Normally a fibrous sample from an occupational setting or known single source can be identified and characterized quite well even at the micrometer size range. This is possible because there would generally be a limited number of minerals to consider. An environmental sample, however, poses a most difficult analytical task if available in small amounts. There can be many common minerals, each of which may have a variable composition, and the net result is that many minerals may occur with overlap in composition, gross crystallographic properties and optical properties. Health researchers often use well characterized samples from specific locations for their experiments. These samples have been chemically analyzed in bulk, but often individual fibers and variations in composition along a fiber have not been analyzed. UICC samples of amosite and crocidolite as well as one sample of tailings from Lake Superior and one asbestiform cummingtonite-grunerite sample from Labrador (fig. 2a) were subjected to analysis using energy-dispersive fluorescence spectroscopy in conjunction with a transmission electron microscope. The analytical procedure is similar to that of Beaman and File [2]. Isolated fibers between 0.2 - 0.8 pm in width were subjected to analysis with an excitation voltage of 80 kV and a take-off angle of 36 degrees. The excited area was estimated to be about 0.2 pm when considering scattering effects. Counts were recorded and areas under peaks were estimated using a computer routine which also adjusted for background. Ratios of peak area of Mg, Fe, Na, and Ca relative to Si were calculated, and these ratios were adjusted for areal ratios determined on an adjacent blank portion of the grid. This later correction was normally negligible. In the following discussion, ratios of areal peaks to Si corrected for background of analyzer and grid background are reported. Champness et al. [7] have noted that the use of intensity ratios should correct for fluorescence variaTtTons due to specimen thickness variations. Figure 3a shows elemental intensity ratios relative to Si for UICC amosite for 58 analyses on 15 fibers, whereas figure 3b shows similar results for 51 analyses on 15 fibers of UICC crocidolite. In both figures, the results are given for increasing Fe/Si intensity ratios, and the values between horizontal lines represent the intensity ratio value for the particular element. Both samples show a marked variation in elemental intensity ratios with between 30-60 percent variation about the mean f+or the corrected values. With reference to amosite, assuming all of the Fe is structural Fe2 , there should be a parallel decrease in the Mg/Si ratio as Fe/Si increases. This is obviously not apparent for the bulk analysis. Although the surfaces of all fibers were examined prior to analysis for optical density continuity so that surficial material such as Fe-oxides might be excluded, it is possible that some of the variation in the Fe/Si ratio is due to surface oxidation of Fe. But this would not explain the variation in Mg/Si ratios for amosite which, with the exception of two extreme analyses, varies about 20 percent about the mean of the ratio. 25

AmOSiT E (UICC) Mg 9 0.8 Fe FIBERS y O H a -J a ~ z w 2 w -J W CROCIDOLITE (UICC) 1.9 1.8 1.7 6 1.5 1.4 1.3 1.2 1.1 1.0 0.9 0.8 0.7 0.6 n ni u F. nj~ r't_I I U~I A IJ Y . n c B FI BERS Figure 3. Corrected intensity ratios for UICC amosite (a) and UICC crocidolite (b). Intensity ratios are cumulative with the value for each element depicted as the difference between adjacent horizontal bars. 26 I

,4 CS Crocidolite UICC. samples show similar variations with the Fe/Si ratios with a deviation of about 15 percent about the mean; Mg/Si varies about 15 percent about the mean, and Na/Si varies about 25 percent about the mean. There is no relationship between Mg or Na ratios and Fe ratios, but there is an apparent correlation between Na/Si and Mg/Si. This correlation could be due to radiation from Na, Mg, and Al. Figure 4 shows elemental intensity ratios for one fiber of UICC crocidolite (4a) and oqk fiber of UICC amosite (4b). Variation of intensity ratios along the fiber length is between 5-10 percent, and this is much less than for the range in variation for all mineral fibers. This is true for all of the eight mineral fibers tested at multiple locations. CROCIDOLITE (UICC) Ti • K Ca • • • • L • • • ANALYZED SPOTS (CROCIDOLITE FIBER 14rm. diameter 0.3Nm.) Figure 4a. Variation along a fiber of UICC crocidolite. 27

l 1-- Coz F- w zcn ww Cn w a- Ti K Na ~ o_ w Ca ~ Q Mn 0.050 / i 0 !0 ~ ~ 01S ~ M 0090 z Si 0080 w 2 Fe/ 1.000 LL! i AMOSITE (UICC) • • • • I • • • ANALYZED SPOTS (AMOSITE FIBER 16rm dia. 03rm.) Figure 4b. Variation along a fiber of UICC amosite. • d Figure 5 shows the results of 64 analyses on 14 fibers of asbestiform cummingtonite from Labrador. There is an approximate 50 percent variation of Fe/Si intensity ratios about the mean, and there appears to be a decrease in the Mg/Si intensity ratio with increasing Fe/Si ratio with only a few exceptions. t 0 • 28

CS Cummingtonite FIBERS Figure 5. Corrected intensity ratios for analysis of 64 asbestiform fibers from Labrador (figure 2a). Figure 6 shows two examples of the analysis of different locations on the same fiber for the Labrador cummingtonite-grunerite sample. Once again there is a much smaller variation (<10 percent) of intensity ratios along an individual fiber with the exception of one location which showed an extremely high Fe/Si ratio. This very large ratio may be due to surficial Fe-oxide, although there was no anomalous electron density visible. N O 29 w ~ 0 w ~+

0 F- Mn N Z Ti ~ z w ~ K 0 ~ ~ Na H a w a w Ca 0200 ar Mg~ 0180 J a /S' 0160 F- 0200 a F~ 0.750 ~ S i 0.700 w 0:650 J w CUMMfNGTONITE ~ ' • • • • • • • • ~ • c-~ • 0 0 • 0 _0 ~ ~ Mg 021 019 51i 017 J 0.15 a ~ Z F e 1.0 w i Q9 w Q8 J w 0.7 ANALYZED S POTS (CUMMINGTONITE FIBERS 16rm. dia. 0.4rm.) CUMM I NG TON ITE 19.6 ._..--.r i J ANALYZED SPOTS (CUM MING TONI T E FIBE RS I6ym. dia. 0.2rm.) Figure 5. Variation in intensity ratios for analyzed spots along two asbestiform cummingtonite-grunerite fibers from Labrador. I 0 30

There appear to be two possible reasons for variations in areal intensity ratios. There can be a real variation in the composition of individual fibers in an apparently homogeneous phase, and/or the differences can be due to x-ray adsorption and secondary radiation especially from Fe in these samples. The fact that analysis on spots on a specific fiber gives an intensity variation less than 10 percent (with one exception in 200 analyses) compared to a 30-50 percent variation in bulk is strongly suggestive that the difference in the two' variations (20-40 percent) is the approximate absolute variation in intensity ratio due to compositional variation that exists in these samples. If the coefficient relating intensity ratios to compositional ratios is not dependent upon other factors, one would anticipate a real variation in fiber composition of 20-40 percent maximum for the major elements. One assumes generally that the composition of fibers within a relatively pure mineral- ogical phase is reasonably constant in composition. This assumption must be tested by detailed analysis of many fibers within a specific sample. Conclusions It appears that asbestos morphology differs from other elongate acicular-fibrous minerals and from environmental exposures in the largest percentile group. Therefore, the entire size distribution should be characterized before carrying on toxicity studies. The composition of fibers within a well characterized sample may vary in composition. Hence analysis on individual fibers must always be carried out. Finally the health significance of fibers other than asbestos should be studied. Primary cytotoxicity and mutagenicity testing of hydrated silicates, anhydrous silicates and non-silicates may well provide clues for more extensive studies. Work supported in part by Inland Waters Directorate, Environment Canada. Microscope analytical work by 0. Mudroch, and field work by R. Marttila are gratefully acknowledged. References [1] American Geological Institute, Glossary of Geology, Washington, D.C., 1972. [2] Beaman, D. R. and File, D. M., Quantitative determination of asbestos fiber concen- trations, Anal. Chem., 48, 101-110, 1976. [3] Borschchevskii, Y. M. and Konikova, T. S., Apatite and apatite-nepheline dust, Nauch. Tr., Leningrad Inst. Usoversk. Vrachei, 115, 108-116, 1973. (Chemical, Abst: 080: 078900). [4] Burilkov, T. and Michailova, L., Sepiolite content of the soil in regions with endemic pleural calcifications, Int. Arch. Arbeitsned, 29, 95-101, 1972. [5] Buritkov, T., Michailova, L., and Babadjov, L., Amphibole asbestos in the soil and its significance for the endemic occurrence of pleural plaques, Zh. Gesamte Hyg. Grenzgeb, 18, 802-809, 1972. (HEEP:74/02556). [6] Campbell, W. J., Blake, R. L., Brown, L. L., Cather, E. E., and Sjoberg, J. J., Selected silicate minerals and their asbestiform varieties, Bur. of Mines Info. Circ. 8251, 1977. [7] Champness, P. E., Cliff, G., and Lorimer, G. W., The identification of asbestos, J. of Microscopy, 108, 231-249, 1976. [8] Cralley, L. J., Inhalable fibrous materials. in H. A. Shapiro (ed), Pneumoconiosis, Oxford U. Press, 1970, p. 70-74. ~. 0 A ~ W W

[9] du Toit, R. S. J., Dust in South African asbestos mines and fiberizing plants, in reference 8, p. 13-17. [10] Hunter, B. and Thomson, C., Evaluation of the tumorigenic potential of vermiculite by intrapleural injectiDn in rats, Brit. J. Indust. Med., 30, 167-173, 1973. [11] Koshi, Kimiko, Hayashi, Hisato, and Sakabe,, Hiroyuki, Cell toxicity and hemolytic action of asbestos dust, Ind. Health. 6, 69-79, 1968. [12] Kramer, J. R., Asbestos: Nomenclature, occurrence and redistribution in water, in Drinking Water and Health, U.S. Nat. Res. Counc., 1977 (in press). [13] Mikhailova-docheva, L., Hygienic evaluation of the mineral composition of soils in regions of the People's Republic of Bulgaria marked by the revelence of endemic pleural calcifications, Gig. Tr. Prof. Zabol., 16, 30-33, 1972. (HEEP: 73/04741). [14] Owinski, J., Effect of dust pollution and thermal microclimate on the incidence of chronic non-specific diseases of the respiratory system with regard to the gypsum workers industry in Gacki, part IV, Radiographic results and symptoms of chronic bronchitis, Przegl Lek, 32, 265-268, 1975. (MEDLINE). [15] Pott, F., Huth, F., and Friedrichs, K. H., Tumorigenic effect of fibrous dust in experimental animals, Environ. Health Perspect., 9, 313-316, 1974. [16] Pylev, L. N. and Iankova, G. D., Carcinogenic activity of magnesia arfvedsonite administered intrapleurally to nonbred rats, Vopr. Onkol., 21, 71-76, 1975. (TOXBIB: 75/104450). [17] Schnitzer, R. J. and Pundsack, F. L., Asbestos hemolysis, Environ. Res., 3, 1-13, 1970. [18] Spiel, S. and Leineweber, J. P., Asbestos minerals in modern technology, Environ. Res., 2, 166-208, 1969. [19] Webster, I., Commentary, in reference 8, p. 133. [20] Zoltai, Tibor, History of asbestos-related mineralogical terminology, This Proceed- ings, Paper 1. [21] Zoltai, Tibor, Comments on asbestiform and fibrous mineral fragments relative to Reserve Mining Company taconite deposits, Minn. Pollut. Contr. Agen., Minneapolis, 37 p + III append. 1 32

SS Discussion C. RUUD: What was the accelerating voltage of your electron beam in all of these microanalyses? J. KRAMER: We tried some studies varying it, but the value we used routinely was 80 kv ' There are a lot of details of these findings on the analytical part which suggest problems. I would be happy to discuss these with individuals. M. COSSETTE: Are you aware of any work with high pressure mercury porosimetry to differentiate between fibrous length groups? KRAMER: No, do you have some data or know of some? COSSETTE: No, I know of some people doing work in the area but nothing published. A. WILEY: Do you use the polarizing microscope, and, if so, do the clino-amphiboles show parallel extinction? KRAMER: Yes, within analytical error, but some of the cummingtonite fibers from Labrador may not show parallel extinction. They may have a small angle (5-10'). WILEY: Your ordinary varieties do, though? KRAMER: Yes, I think that this is a very important point to consider; this apparent optical difference and its significance to fiber morphology. D. BEAMAN: 0.3 pm is not particularly large for an amphibole. I wonder to what extent you feel some of these trends may be due to the difference in the size of your fibers. KRAMER: Yes, there may well be a size factor. 0.3 pm width is at the threshold of size effect upon intensity ratios according to your study published in Analytical Chemistry. F. MUMTON: I'd like to ask you about your ion exchange measurements of these two types of materials; you didn't show any data, but yet you say there are differences. What range are you talking about? What did you do? KRAMER: First of all, the ion exchange differences will depend upon the composition of the material. We worked mostly with cummingtonite from Labrador. What we are using basically are these minerals (see figure 2) as an exchange medium to compete against a copper-organic ligand. The procedure is analogous to an ion exchange column but we are using the minerals. We calibrate the system against known associations such as copper- glycine. We carried out the analyses using equidimensional, fibrous and asbestiform varieties and found little differences in conditional stability constants for the different varieties of the same composition. In addition, the exchange capacities appear to be very similar and typical of all silicate minerals (about 3-4 micro-equivalents/meter2). W. EISENBERG: Have you modified your definition of a mineral species as a result of the data you've obtained? KRAMER: No, you noticed I didn't give any definitions. I just quoted other people. Seriously, I am trying to point out that there are either analytical problems or variations in composition, or both, at the micrometer scale of a fiber. See Science, 198, 359-365 for some possible reasons. 33 1 N 0 A ~ Ln

CS National Bureau of Standards Special Publication 506. Proceedings of the Workshop on Asbestos: Definitions and Measurement Methods held at NBS, Gaithersburg, MD, July 18-20, 1977. (Issued November 1978) THE CRYSTAL STRUCTURES OF AMPHIBOLE AND SERPENTINE MINERALS Jack Zussman Department of Geology University of Manchester England Abstract The crystal structures of the two main asbestos-forming minerals, the amphiboles and serpentines, are surprisingly very different. The amphiboles are "chain silicates" in which Si04 tetrahedra are linked to, form bands four tetrahedra wide and of very great length. These bands run parallel to the asbestos fiber axis and are linked laterally by cations, mainly Ca and Mg in tremolite; Na, Mg and Fe in crocidolite; Mg and Fe in amosite and anthophyllite. The tempting correlation of the chain unit of crystal structure with asbestiform nature is, however, too facile. Many amphiboles are not asbestiform, and as the serpentine minerals show, some asbestiform minerals do not have a chain structure. The serpentine minerals are "layered silicates" in which Si0* tetrahedra are linked to form thin sheets of great lateral extent. The tetrahedra all point in the same direction and their apical oxygens are part of an (O,OH)-Mg-(OH) sheet which is itself formed by Mg-(O,OH) octahedra. Thus the fundamental serpentine layer is polar and has a tetrahedral and octahedral component. The mismatch in dimensions of these two components generally leads to curvature of the layers and in chrysotile asbestos the layers form either scrolls or concentric cylinders with very high length/breadth ratio and with length parallel to the fiber axis. Other forms of serpentine, however, with chemistry very similar to that of chrysotile, do not exhibit asbestiform morphology. For all minerals, the physical and chemical properties are impor- tant both for industrial usage and environmentally in determining the nature of the dusts produced in manufacturing processes and in subsequent abrasion. Factors which may influence properties in addition to the basic chemistry and "average" x-ray structure are the crystal morphology and mode of aggregation, and also the abundance and nature of structural defects. Keywords: Amphibole; asbestos; chemistry; cleavage; defects; dusts; environment; fibers; morphology; serpentine; structure. In this review I would like to describe briefly the crystal structures of the two main asbestos-forming minerals, the amphiboles and serpentines, to consider what they have in common and what are their differences, to identify if possible what are the fundamental criteria that lead to asbestiform habit, and to observe the crystallographic features that may contribute to their physiological behavior. Preceding page blank 35 N ~ 0 ~ w a

,. Amphiboles These minerals are "chain silicates" in which SiO,i tetrahedra are linked so as to form chains with composition 5i,011 as shown in figure 1. The chains are four tetrahedra wide, of very great length,and they lie parallel to the fiber axis in the asbestiform amphiboles. one might say that amphibole asbestos is finely fibrous because of the chain structure, but this is an over simplification. Some amphiboles are not fibrous at all, let alone asbestiform. -0] il. lI L2 ~ ~ E--5~3 A--)-- 0 Oxygen ® OH Cc Figure 1. Plan and end-view of an idealized Si4011 amphibole chain together with additional (OH) ions. No minerals could be formed from Si*011 chains alone and in the amphiboles there are cations linking chains laterally as shown in figure 2. The cations vary from one amphibole to another. In tremolite Mg ions link chains by means of a strip of Mg(O,OH) octahedra. The oxygens of this strip are the apices of the Si-0 tetrahedra and the OH ions occur as in figure 1. Calcium ions link the chains across the bases of the tetrahedra. An alternative view of the structure is one of almost continuous sheets of Mg and Ca polyhedra linked by Si ions. 36

x 4^ A ~ ~ 4^ V ~ Y ^ ^ ^ ^ ^ ^ ^ ^ Figure 2. Schematic end-view of amphibole chains linked by cations in K and Y positions. In some amphiboles the site A is occupied. The amphibole formula can thus be written Ao_1X2Y5(5i,A1)8o22(OH)2. Other important amphiboles are: - anthophyllite in which largely Mg ions play the role of both Ca and Mg in tremolite; the cummingtonite - grunerite series, which contain Mg and Fe in varying proportions; and riebeckite, in which Mg, Fe and Na are the principal cations in addition to Si (see Table 1). The above-mentioned compositions are those most relevant to the consideration of asbestos, since in addition to the less common varieties of asbestos, tremolite and anthophyllite, there are the two more abundant and commercially more important varieties -'amosite,' a form of cummingtonite - grunerite, and "crocidolite" (blue asbestos), a form of riebeckite.

0 Table 1. Cation distribution in idealised formulae of the amphibole minerals. Asbestos-forming amphiboles are marked*. - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - A X Y Z Cummingtonite-Grunerite* ., - S (Mg,Fe)Z (Mg,Fe)s Sig Anthophyllite ' * Gedrite - (Mg,Fe)2 (Mg,Fe)3A12 Si6A12 Tremolite*-Actinolite - Ca2 (Mg,Fe)s Sig Common Hornblende - Ca2 (Mg,Fe)4A1 Si,A1 Tschermakite - Ca2 (Mg,Fe)3A12 SieA12 Edenite Na Ca2 (Mg,Fe)s Si7A1 Pargasite-Hastingsite Na Ca2 (Mg,Fe)sA1 Si6A12 Richterite Na NaCa (Fe)s Sig Katophorite Na NaCa (Mg,Fe)+A1 5i7A1 Mboziite Na NaCa (Mg,Fe)sA12 Si6A12 Glaucophane-Riebeckite* - Na2 (Mg,Fe)3A12 Sig Eckermannite-Arfvedsonite Na Na2 (Mg,Fe)4A1 Sig CALCIUM AMPHIBOLES ALKALI AMPHIBOLES For the sake of completeness at least, though it may also have some indirect importance, it should be noted here that the amphibole asbestos-forming minerals are monoclinic in sym- metry except for anthophyllite which is orthorhombi c. Cell parameters are given in Table 2. Table 2. Amphiboles. Cell parameters.a aA bA cA p Tremolite [1]1 9.82 18.05 5.28 104°39' Actinolite [2] 9.89 18.20 5.31 104°38' Grunerite [3] 9.56 18.30 5.35 101°50' Crocidolite [4] 9.74 17.95 5.30 103°54' Anthophyllite [5] 18.56 18.01 5.28 900 a These cell parameters relate to particular specimens. Variations in chemical composition, particularly in Fe/Mg ratio, can be expected to yield a range of values but usually within 1 or 2 percent of those given. N 'Figures in brackets indicate the literature references at the end of this paper. ~ w 38 o A W W ~

A well-known physical property of amphiboles is that they generally cleave readily along (110) planes. Assuming that the Si-0 chain is a strong structural unit, and that the strongest inter-chain bonding is across the strips of octahedra joining tetrahedral apices, the probable paths of weakness can be traced as on figure 3, which on a macroscopic scale results in cleavages intersecting at approximately 1200, as observed. / r Figure 3. Schematic view of amphibole structure as seen down z axis, showing likely paths of weakness leading to cleavages intersecting at 57°. Although the good prismatic cleavages explain the readiness of amphibole crystals to splinter into elongated particles, this is not necessarily relevant to the unusual physical nature of asbestos. It would be so if a block of asbestos was a single crystal and the production of hair-like fibers was the process of splitting off cleavage fragments. However, a block of asbestos, even when very small, is not a single crystal but an aggregate of single- crystals all lined up parallel to the fiber axis but with a range of azimuthal orientations. The process of stripping fibrils from asbestos is thus more likely to be one of breaking crystallites away from the aggregate at the grain boundaries across which there is weak cohesion. 39 2063104840

Cot The asbestiform nature of certain amphiboles is thus a consequence of the crystallite morphology which in turn is influenced by the conditions of crystal growth as well as the inherent chemical and physical features of a single crystal. The production of a fiber aggregate as in asbestos must depend upon independent nucleation of each fibril and its preference for growth along z rather than at right angles to it. It is perhaps significant that the group of amphiboles loosely referred to as "hornblendes" occur in roughly equidimensional crystal habits and not as asbestos. Table 1 shows a simplified scheme for the chemical compositions of amphiboles and it is seen that the hornblendes are characterized chemically by having appreciable substitution of Al for Si. Asbestiform amphiboles show little substitution of this kind. The minerals richterite and eckermannite, which also have little Al for Si substitution, are not known to occur naturally as asbestos, but synthetic products have been so described and are at least extremely fibrous [6,7]. I would suggest therefore that the substitution of Al for Si might be responsible for increased potential for growth of prism faces relative to growth in the z direction. Even if true, the above suggestion cannot be the only criterion that governs asbestos formation since tremolite itself can occur in asbestiform or non-asbestiform habit, each variety having the same major element chemical composition. In such circumstances other parameters such as the pressure and temperature conditions, rates of cooling or heating, or minor or trace element concentrations may be critical factors. The mechanical properties of asbestos and related minerals are of importance both for the desirable physical attributes of articles made from asbestos, and environmentally in determining the nature of the dusts produced during the processes of manufacture or during subsequent abrasion. Factors which can give different mechanical properties are the nature of the fundamental particles and their state of aggregation (bundles of fibers versus single crystals). For single fibrils or crystals in the {110} cleavages, and the resistance to breakage across other planes (roughly perpendicular to fiber length), will help to determine the morphology of the dust particles produced. Structural defects may also have an influence on physical properties. Structural defects It should be emphasized here that the published crystal structures of amphiboles (and serpentines), determined by x-ray diffraction, are the content of the "average" unit cell, the volume of specimen investigated consisting of something like 1015 unit cells. In real crystals the unit cells do not repeat perfectly and several kinds of defects may occur. These departures from the perfect structure are no doubt important in questions concerning crystal growth and they may well influence physical properties and physiological effects. The two principal kinds of imperfection in amphibole structures are stacking defects and Wadsley defects. Stacking defects are illustrated schematically in figure 4. In the normal monoclinic amphibole, slabs of structure parallel to (100) are stacked alongside one another with regular displacements. In a faulted structure occasional errors in the direction of this displacement occur and the frequency of such faults varies from one specimen to another. When the faults are relatively infrequent the result can sometimes be described as a twinned crystal. Figure 5 shows a high resolution electron micrograph displaying twin components. Such defects are also seen in lower magnification electron micrographs and they have important effects on diffraction patterns. When the faults are frequent and regularly repeating, they are no longer really faults but are the regular displacement of a structure with a super-cell and perhaps different symmetry. The latter describes approximately the relationship between the orthorhombic and monoclinic amphiboles. 40

CS monoclinic monocUnic twin WO orthorhombic monoclinic with stacking fauLt Figure 4. Illustration of the stacking of blocks of amphibole structure to forni a) regular monoclinic, b) monoclinic twinned, c) monoclinic faulted, and d) orthorhombic structures. The fault plane is (100). Figure 5. High-resolution electron micrograph of amosite showing faulted and twinned struc- tures. (Electron beam parallel to Y) Figure from J. L. Hutchinson et a1. [8]. 41 2063104842

Figure 6 illustrates the Wadsley defect by showing how parts of an amphibole crystal might contain occasional triple or single Si-0 chains distributed among the normal double chains. In low magnification electron micrographs such defects are seen as linear features parallel to (010) (fig. 7). PYROXENE- AMPHIBOLE /-'a7Ma A/LJ\A ~ single chains double chains I triple chain double chain double chain double chain double chain single chain Figure 6. Schematic illustration of a) pyroxene structure, b) amphibole structure, c) amphibole with triple chain Wadsley defect, and d) amphibole with single chain defect. Figure 7. Electron micrograph of amphibole with beam perpendicular to y showing Wadsley defects on (010). Figure from J. E. Chisholm [9]. 42

C3 For environmental health considerations we do not yet have a causative understanding of the harmful effects of asbestos and do not know which properties of asbestos are involved. It is conceivable therefore that more subtle structural factors than those described above might be important. Although the structure described is broadly correct for all amphiboles, minor differences in atomic coordinates occur from one amphibole to another. Structure determinations have been performed for non-asbestiform tremolite, actinolite, anthophyllite and grunerite, but not for asbestiform specimens because of technical difficulties. For crocidolite, a fiber approaching a single crystal was used rather than a hair-like strand of asbestos. As part of the details of structure, variations can occur in the way in which Fe and Mg atoms are distributed amoQg similar but not strictly equivalent octahedral sites. In some amphiboles the role of Fe3 may be significant in oxidation-reduction processes, and there is the possibility of Na (or Ca) having a degree of cation exchange capacity. Serpentine Chrysotile, another important variety of asbestos is not an amphibole but is a member of the serpentine group of miner;1s. Because of its asbestiform character, and repeat distance in the unit cell of about 5.3 A parallel to the fiber axis (similar to that in amphiboles), it was once thought to have a chain-like crystal structure. Later work, however, showed it to be a layered silicate with structure analogous to that of the clay mineral kaolinite, but with Mg instead of Al in its composition. The paradox of how a layered mineral could have asbestiform habit was solved largely by Whittaker [10,11,12] who deduced from x-ray diffraction patterns that layers are rolled to form concentric cylinders or scrolls with their long axes parallel to the fiber. This indirect evidence was supported by electron microscopy of transverse sections of chrysotile, culminating in the spectacular high-resolution photographs published by Yada (fig. 8). Figure 8. High resolution electron micrograph of transverse section of chrysotile asbestos. Figure from K. Yada [13]. 43

0 The reason for the curving of the fundamental layers in chrysotile can be seen by examination of their chemical composition and structure. Each layer has two components, one a sheet of linked Si-0 tetrahedra, and the other (,joined to the first by sharing apical oxygens), a sheet of (Mg-O,OH) octahedra. A plan and elevation view of the composite layer, Mg3SizOs(0H)*, is shown iP figure 9. In order to form a flat-layered composite the dimensions of each component would need to match fairly closely. Reasonable estimates of the repeat distance of each show that the tetrahedral Si sheet has smaller dimensions than the octahedral Mg sheet, and this mis-match can be overcome by curvature, with the Mg sheet outermost, or by some other means of relieving structural strain. This leads to a number of strange structural configurations in serpentines, one of which is the tube-like character of chrysotile. W. O -{JT! I 1 . I b~ ~I I I ~ ~ I , I I 1 I I I j I I I I oti o e I ~ b O O O o 0 Me x x x x x x x x x x O.OH O O 0 O O 0 Si 0 e ~-b~o9~2A ~ Ol4 O O O O O O O O O O O cd•2A Figure 9. Plan and elevation views of idealized serpentine structure. Electron micrograph studies of chrysotile asbestos show that diameters of natural fibrils are of the order of 100 to 500 A, and the length/breadth ratios are of the order of 100 to 1 or greater. The limitation on growth in the radial direction is more easily understood for chrysotile than for amphiboles in that as successive layers are added during the growth process, the radius of curvature increases, eventually deviating too far from its ideal strain-free value to be energetically favorable. Thus chrysotile asbestos probably forms by multiple nucleation, usually on the walls of veins in massive fine-grained serpentinite rock, with relatively rapid growth in the fiber direction and limited growth at right angles to it. It is pertinent in the context of possible environmental problems to consider the structure and morphology of other serpentine minerals which have very similar composition but are not asbestiform. One such mineral is antigorite. It too has a curved sheet structure, but the layers are corrugated rather than rolled (fig. 10). The corrugations have a rather regular wavelength so that quite well-formed crystals result. Sometimes they are equidimensional but they have a tendency to be thin, and lath-like parallel to y. Antigorite 44

Q.S is often found associated with other serpentine minerals; it does have a small but distinct difference in chemistry and is known to form under higher temperature conditions than the others [15]. © da ® __ea: -nn -a ----------0~;~-vd7~ OF v var v w Figure 10. The "corrugated sheet" structure of antigorite viewed along the y axis. After G. Kunze (14]. A third kind of serpentine is the mineral lizardite, which in spite of the difficulties mentioned above does manage to achieve a more or less flat-layered structure [16]. The accompanying strain however means that crystals contain imperfections and usually grow only to very small dimensions. Thus a high proportion of apparently massive serpentine is composed of lizardite grains too small for optical resolution, but seen by the electron microscope to have platy morphology. The stacking of successive serpentine layers in lizardites can lead to 1,2,3,6 and even 9-layer repeats, and whereas lizardite platelets are usually not elongated, some of the multi-layer varieties yield lath-like crystals, and again, like antigorite a coarse splintery fiber. Cell parameters of serpentine minerals are given in Table 3. clino-chrysotile [10] ortho-chrysotile [11] para-chrysotile [12] lizardite [17] antigorite [14] Table'3. Serpentines. Cell parameters.a aA bA cA ~ Fiber Axis 5.34 9.25 14.65 93°16' x 5.34 9.2 14.63 900 x a5.3 9.24 14.7 900 y -5.3 '_~9.2 z7.3 x nb 90° x when fibrous 43.3c 9.23 7.27 91.6° Y when fibrous a These cell parameters relate to particular specimens. Variations in chemical composition, particularly in Fe/Mg ratio, can be expected to yield a range of values, but usually within 1 or 2 percent of those given. For antigorite markedly different a values occur. b Lizardites with n= 1, 2, 3, 6, and 9 have been described. c Other large values of a are found. 45

Yet another strange morphology for a serpentine mineral has been discovered recently [18], and although it has not yet been studied extensively, it does appear to be quite common in occurrence. In this variety, flat lath-like serpentine layers are arranged to form polygonal prisms, sometimes surrounding a core of tubular chrysotile. A cross-,section is illustrated in figure 11. Typical diameters are of the order of 1000 to 2000 A. Serpentine specimens in which this structure seems to be prevalent are those which have a coarse splintery fibrous texture. Their fracture fragments are expected to have, and indeed show, lath-like morphology. Whether this material should be classed as a form of chrysotile or of lizardite is a moot point, and it may be better to call it "polygonal serpentine" with our present state of knowledge. Figure 11. Electron micrograph of an ion-thinned serpentine specimen showing cross-sections of chrysotile tubes and polygonal serpentine. Figure from Cressey and Zussman [18]. Although for antigorite there is clearly a distinct chemical composition, there is no consistent chemical difference between chrysotiles and lizardites. The latter two can therefore be regarded as polymorphs and would be expected to have distinct (P,T) stability fields. Attempts to define these have not so far been successful. Examination of the mineralogy and textures of large numbers of serpentinite rocks have led to the conclusion that the chrysotile asbestos is formed secondarily from lizardite or antigorite and not directly from olivine and pyroxene, and that it is formed in a relatively low but rising temperature regime [19,20]. Concluding Remarks There have not as yet been extensive tests comparing the physiological activities of asbestiform and non-asbestiform varieties of amphiboles or of serpentines. It would clearly be useful to know what, in addition to morphology, are the essential chemical and physical differences between asbestiform and non-asbestiform varieties. These differences might be 46

CS either consequences or causes of the contrasting morphology. Since, for any mineral, different specimens show variations in properties even when morphology does not change significantly, it is not easy to determine which, if any, are absolutely specific to asbestos. Such differences, if established, might be quite subtle but nevertheless important for physiological effects, but since the mechanism of the latter is unknown, we have no clues from this quarter to aid us in the search. If particle size and shape are the only important factors, then we need not trouble to look further (except as a fascinating geological problem). If other factors are important, and we do not know what they are, then any material to which people are exposed on a large scale needs to be tested for its physiological effects. References [1] Papike, J. J., Ross, M., and Clark, J. R., Crystal-chemical characterization of clinoamphiboles based on five new structure refinements, in Pyroxenes and Am hib~oles: Cr stal Chemistry and Phase Petrology, Min. Soc. Amer. 5pecial Publ. No. 2 117-136 1969 . [2] Mitchell, J. T., Bloss, F. D., and Gibbs, G. V., Examination of the actinolite structure and four other C2/m amphiboles in terms of double bonding, Zeit. Krist., 133, 273-300 (1971). [3] Ghose, S. and Hellner, E., The crystal structure of grunerite and observations on the Mg-Fe distribution. J. Geol. , 67, 691-701 (1959). [4] Whittaker, E. J. W., The structure of Bolivian crocidolite, Acta Cryst., 2, 312-317 (1949). [5] Finger, L. W., Refinement of the crystal structure of an anthophyllite, Ann. Rep. Dir. Geophys. Lab., Carnegie Inst. Yr. Bk., 68, 283-288 (1970). [6] Fedoseev, A. D., Makarova, T. A., and Kosulina, G., Synthesis of fibrous richterite under hydrothermal conditions. Zap. Vses. Min. Obshch. , 97, 722-725 (1968). [7] Goncharov, Yu, I., Balitskii, V. S., Khadzhi, I. P., and Popova, N. P., Replacement of phlogopite by amphibole asbestos of the eckermannite-arfvedsonite series in alkaline hydrothermal solutions, ZaQ. Vses. Min. Obshch. , 103, 716-718 (1974). [8] Hutchison, J. L., Irusteta, M. C., and Whittaker, E. J. W., High resolution electron microscopy and diffraction studies of fibrous amphiboles. Acta Cryst., A31, 794-801 (1975). [9] Chisholm, J. E. , Planar defects in fibrous amphiboles, J. Mat. Sci., 8, 475-483 (1973). [10] Whittaker, E. J. W., The structure of chrysotile, II., Clinochrysotile, Acta Cryst. 9, 855-862 (1956). [11] Whittaker, E. J. W., The structure of chrysotile, III, Orthochrysotile, Acta Cryst. 9, 862-864 (1956). [12] Whittaker, E. J. W., The structure of chrysotile, IV, Parachrysotile, Acta Cryst. 9, 865-867. [13] Yada, K., Study of chrysotile asbestos by a high resolution electron microscope, Acta Cryst. 23, 704-707 (1967). [14} Kunze, G., Die gewellte Struktur des Antigorits, I, Zeit. Krist. 108, 82-107 (1956). [15] Evans, B. W., Johannes, W., Oterdoom, H., and Trommsdorff, V., Stability of chrysotile and antigorite in the serpentinite multisystem. Schweiz min. etro r. Mitt. , 56, 79-93 (1976). N 47 gw~ ~

[16] Rucklidge, J. C. and Zussman, J., The crystal structure of the serpentine mineral, lizardite Mg35iz0s(OH)s, Acta Cryst. , 19, 381-389 (1965). [17] Whittaker, E. J. W. and Zussman, J., The characterization of serpentine minerals by x- ray diffraction, Mine?al. Mag., 31,107-126 (1956). [18] Cressey, B. A., and Zussman, J., Electron microscopic studies of serpentines, Canad. Mineral., 14, 307-313 (1976). [19] Wicks, F. J. and Zussman, J., Microbeam X-ray diffraction patterns of the serpentine minerals, Canad. Mineral., 13, 244-258 (1975). [20] Wicks, F. J. and Whittaker, E. J. W. , Serpentine textures and serpentinization, Canad. Minera1..15, In press (1977). Discussion NOTE: Discussion of this paper was included in the General Discussion at the end of this session. 48

National Bureau of Standards Special Publication 506. Proceedings of the Workshop on Asbestos: Definitions and Measurement Methods held at NBS, Gaithersburg, MD, July 18-20, 1977. (Issued November 1978) THE "ASBESTOS" MINERALS: DEFINITIONS, DESCRIPTION, MODES OF FORMATON, PHYSICAL AND CHEMICAL PROPERTIES, AND HEALTH RISK TO THE MINING COMMUNITY Malcolm Ross U.S. Geological Survey National Center, 959 Reston, VA 22092 Abstract The mineralogical description of "asbestos" given here is based on a very special feature common to all forms of commercial "asbestos" - the property that permits the minerals to separate into long tubes or fibrils only a few tens of nanometers thick. This separation can be accomplished by very light grinding or agitation; the common non-fibrous amphiboles do not separate into such fibrils even after intense grinding. The ease of such fibril separation may be caused by the special nature of the crystal structures of the commercial "asbestos" minerals. Repeated twinning oh (100) in amosite and crocidolite, the curling of layers of chrysotile to form tubes, and the presence of triple, quadruple, n-tuple chains ("Wadsley" defects) in amosite, crocidolite, anthophyllite, and tremolite, are the structural features that probably promote the formation of thin fibrils. Stability diagrams in the system Mg0-Si02-H2O indicate possible geochemical processes by which commercial "asbestos" can form. The relative health risk posed by exposure to the "asbestos" minerals may be related to the fibril composition, crystal structure, size, shape, and total surface area. The relative chemical reactivity of the fibril surface is predicted to be chrysotile < anthophyllite < amosite < crocidolite on the basis of the types of oxidation-reduction and exchange reactions that may occur. According to epidemiological studies, the relative health risk appears to be anthophyllite < chrysotile < amosite < crocidolite. "Asbestos" health risks in the mining and milling industry and environs are reviewed. Health studies done in the chrysotile mining district of Quebec, Canada, have presented good evidence that realistic "asbestos" dust standards can be set that not only protect the workers and residents of the mining areas from undue health risks but probably allow the industry to operate economically. Key Words: Actinolite; ambient air; amosite; amphibole; amphibolite; anthophyllite; asbestos; asbestos stability; chrysotile; chrysotile emissions; chrysotile mining; crocidolite; cummingtonite; dust levels; grunerite; health risk; Homestake Mines, S.D.; hornblende; Hunting Hill Quarry, Rockville, Md.; lung cancer; mesothelioma; serpentinite, surface chemistry; talcbole; Thetford Mines, Quebec, Canada; tremolite; Urals; U.S.S.R.; and Wadsley defects. 49

0 Introduction It is generally a rather straightforward, though often time-consuming mineralogical task to describe the physical and chemical properties of amphiboles and serpentines including those varieties referred to as "asbestos". Exceptions are minerals such as fibrous tremolite and fibrous talc that to date do not have adequate mineralogical descriptions. Defining minerals that constitute an "asbestos" health hazard is an entirely different and a much more complex problem, for it involves many factors not included within the science of mineralogy. This commentary is concerned with the various definitions of "asbestos" as they relate to: (1) the medical profession, which must determine which types of mineral particles constitute an "asbestos" health hazard; (2) the legal and regulatory professions, which must enact and enforce the laws relating to "asbestos" use; (3) the mineralogical profession, which must describe the chemical, structural, and physical properties of such minerals; and (4) the mining and quarrying industries, which may be affected by these definitions. What is "Asbestos"? Three definitions of "asbestos" found in the Glossary of Geology [9, p. 41]1 are quoted as follows: "asbestos (a) A commercial term applied to a group of highly fibrous silicate minerals that readily separate into long, thin, strong fibers of sufficient flexibility to be woven, are heat resistant and chemically inert, possess a high electric insulation, and therefore are suitable for uses (as in yarn, cloth, paper, paint, brake linings, tiles, insulation, cement, fillers, and filters) where incombustible, nonconducting, or chemically resistant material is required. (b) A mineral of the asbestos group, principally chrysotile (best adapted for spinning) and certain fibrous varieties of amphibole (esp. tremolite, actinolite, and crocidolite). (c) A term strictly applied to the fibrous variety of actinolite." The term "asbestos", from a geoscientist's point of view, applies only to the minerals chrysotile (one of the serpentine polymorphs), "amosite" (a variety of grunerite), "crocidolite" (a variety of riebeckite), anthophyllite, tremolite, and actinolite when they are present in sufficient quantity to be commercially valuable for their special physical and chemical properties, which include fibrous habit, insulation qualities, low electrical conductivity, fire resistance, and suitability for weaving. Many other minerals sometimes possess habits described variously as acicular, asbestiform, elongate, fibrous, bladed, lamellar, filiform, prismatic, or cotumnaor example-m-fnerals of the zeo ite group having accular habit, fibrous calcite an3 quartz, acicular wollastonite, prismatic pyroxenes, elongate chrystallites of attapulgite, and filiform sepiolite. Since these minerals are not exploited for the commercially valuable propert es listed above, they are not called "asbestos" by geoscientists. At present, the most widely used definition of "asbestos" by various groups concerned with environmental health problems, including the,U.S. Environmental Protection Agency (EPA) and the U.S. Mining Enforcement and Safety Administration (MESA), is from the notice of proposed rule-making for "Occupational Exposure to Asbestos" published in the Federal Register (Oct. 9, 1975, p. 47652, 47660) by the U.S. Occupational Safety and Health Administration (OSHA). In this notice, the naturally occurring minerals chrysotile, amosite, crocidolite, tremolite, anthophyllite, and actinolite are classified as "asbestos" if the individual crystallites or crystal fragments have the following dimensions: length - greater than 5 micrometers, maximum diameter - less than 5 micrometers, and a length to diameter ratio of 3 or greater. Any product containing LnX of these minerals in this size range are also defined as "asbestos". The crushing and milling of ~any rock usually produces some mineral particles that are within the size range specified in the OSHA rules. Thus, these regulations present a formidable problem to those analyzing for "asbestos" minerals in the multitude of materials and products in which they may be found in some amount, for not only must the size and shape of the "asbestos" particles be determined, but also an exact mineral identification must be- made. 'Figures in brackets indicate the literature references at the end of this paper. 50

C3 A wide variety of amphiboles is found in many types of common rocks; many of these amphiboles might be considered "asbestos" depending upon the professional training of the person involved in their study and the methods used in mineral characterization. Campbell et al. [3] have carefully described the differences between the relatively rare fibrous varieties of the amphiboles and the common nonfibrous forms. - If the definition of "asbestos" from the point of view of a health hazard does include the common nonfibrous forms of amphibole, particularly the hornblende and cummingtonite varieties, then we must recognize that "asbestos" is present in significant amounts in many types of Igneous and metamorphic rocks covering perhaps 30 to 40 percent of the United States. Rocks within the serpentinite belts; rocks within the metamorphic belts higher in grade than the greenschist facies, including amphibolites and many gneissic rocks; and amphibole-bearing igneous rocks such as diabase, basalt, trap rock, and granite would be considered "asbestos" bearing. Many iron formations and copper deposits would be "asbestos" bearing, including deposits in the largest open-pit mine in the world at Bingham, Utah. "Asbestos" regulations would thus pertain to many of our country's mining operations, including much of the construction industry and its quarrying operations for concrete aggregate, dimension stone, road metal, railroad balast, riprap, and the like. The "asbestos" regulations would also pertain to the ceramic, paint, and cement industries, and to many other areas of endeavor where silicate minerals are used. We do not know whether health investigators will consider other minerals that commonly possess a fibrous or acicular habit to be health hazards; minerals such as wollastonite, the fibrous forms of calcite and quartz, acicular minerals of the zeolite mineral group, the pyroxenes, the sepiolite minerals including attapulgite, and the calcium silicates found in Portland cement. Certainly if the common amphiboles such as hornblende, tremolite, actinolite, gedrite, and cummingtonite with their typical prismatic cleavage are considered health hazards, the common pyroxenes having similar habits should also be considered health hazards. A Mineralogical Description of Commercial "Asbestos" The commercial deposits of "asbesto~s" contain one of the following minerals: chrysotile, Mg35i204.(0H)4; Imosite, (Fe2 ,Mg)rSi80zQ(OH)2 (a variety of grunerite); crocidolite, Na2(Fe2 ,Mg)3FeJ Sig022(OH)y (a variety of riebeckite); "fibrous" anthophyl- lite, (Mg,Fe)1Sis022(OH)z; and "fibrous" tremolite and actinolite, Ca2(Mg,Fe)s5ia022(OH)2. Tremolite and actinolite are now, as they were in the past, of little economic importance; anthophyllite is of little economic importance now. About 95 percent of the commercial asbestos now used in the United States is chrysotile, of which about 90 percent is imported from Canada. No commercial amosite or crocidolite has ever been mined in the United States. In addition to being compositionally different, the five amphibole forms of commercial "asbestos" have completely different crystal structures from that of chrysotile. The structure of chrysotile consists of double layers, each consisting of a layer of linked 5104 tetrahedra that is coordinated to a second layer of linked M902(OH)4 octahedra through the sharing of oxygen atoms; the composite double layer rolls up, like a window shade, to form long hollow tubes. The diameters of the individual tubes are on the order of 25 nm; the length-to-diameter ratio can vary from 5 or 10 to well over 10,000. The structures of the amphibole minerals, on the other hand, are composed of strips or ribbons of linked polyhedra, which join together to form the three-dimensional crystal. The individual strips are composed of three elements-two double chains of linked (Si,Al)04 tetrahedra that form a "sandwich" with a strip of linked MgOa, Fe08, or A108 octahedra. The structural relationship of the upper double tetrahedral chain to the octahedral part of the strip is shown in figure 1. The three-dimensional arrangements of these strips or "I-beams" [26] orthoamphibole (anthophyllite) and in clinoamphibole (tremolite, amosite, actinolite, and crocidolite) are shown in figure 2. N 51 ~ 3 U1 N

d, Pnmo ® ® b ® ® ® ® ® Figure 2. Arrangement of the amphibole strips or "I-beams" in (A) orthoamphibole (space group Pnma) and (B) clinoamphibole (space group C2/m). The "I-beams" are viewed end-on (parallel to the fiber c-axis). The central portion of the "I-beam" is composed of (Mg,Fe,AI)06 octahedra; the upper and lower portions are composed of double chains of (Si,A10*) tetrahedra. The "I-beams" are stacked in two ways: (1) +++... (clinoamphibole), and (2) + - + -... (orthoamphibole). Figure modified from Papike and Ross [26]. One feature is common to the six "asbestos" minerals: their ready separation into long fibrils or tubes only a few tens of nanometers in diameter. This separation can be accomplished by very light grinding or by agitation in water by means of an ultrasonic separator. The common nonfibrous amphiboles do not separate into such fibrils even after intense grinding; instead, they break up along cleavage planes into rather short stubby prisms-though the length-to-diameter ratio may still be greater than 3:1. Figure 1. Structural relationship between the upper double chain of linked (Si,AI)o4 tetrahedra and the octa- hedra part of the amphibole strip of "I-beam." The circles represent Mg, Fe, or Al atoms in octahedral coordination; at the apices of the polyhedra are oxygen atoms. Tetra- hedral Si and Al atoms are not shown. The "I-beams" extend infinitely in a direction parallel to the c-axis (the fiber axis). The width of the "I-beam" in the b-direction is three octahedra. Figure is modified from Papike and Ross [26]. 52

CS t What causes the special type of fibril separation found in commercial forms of "asbestos" but generally not in the nonfibrous amphiboles? Three observations are pertinent: (1) Chrysotile, which forms individual hollow tubes, can separate into fibrils as thin, as the diameter of the individual tube. The chemical bonding between tubes is very weak and perhaps is due only to van der Waals forces; thus, the tubes are easily separated from one another. (2) Amosite and crocidolite "asbestos" from South Africa is repeatedly twinned on (100) as has been observed in electron microscope studies [4,15,25,34]. This "poly- synthetic" twinning, which produces repeated planar faults parallel to (100), is extremely rare in the nonfibrous calcium-rich amphiboles (tremolite, hornblende) and uncommon in nonfibrous amphiboles of the cummingtonite-grunerite series [30,31,32]. (3) Amosite, crocidolite, fibrous anthophyllite, and fibrous tremolite have been shown to possess chain defects, also called "Wadsley" defects [8,15,36,37,38]. These defects are caused by the formation of expanded "I-beams" that are composed of triple, quadruple...etc. chains of linked (Si,AI)0.k tetrahedra rather than the double chains found in all amphibole crystal structures. If these "I-beams" are expanded indefinitely, the resulting strip becomes identical with the single talc layer of composition MgaSi802o(OH).; recall that the composition of anthophyllite is Mg7Si80ZZ(OH)z. These expanded "I-beam" units can intermix with the regular amphibole "I-beams" to form a variety of minerals that I refer to as "talcboles" in allusion to their hybrid character-between talc and amphibale. Veblen [38] has described the detailed structures of four of these "talcboles" obtained from specimens originally described as "fibrous anthophyllite." In these crystal struc- tures, "I-beams" of one or two types form an ordered three-dimensional structure. Veblen [38] showed evidence, as did Hutchison et al. [15], that disordered arrangements of these structural units also occur. Hutchison et al. [15] reported the presence of expanded "I-beam" structures in fibrous tremolite, and Franco et al. [8] reported the apparent presence of triple-chain lamellae, seen as planar faults on (010), in crocidolite from Western Australia. Formation of "Asbestos" How do chrysotile and the "talcboles" form? Modes of origin can be inferred from the stability relationships among talc, anthophyllite, enstatite, forsterite, antigorite, and chrysotile given by Hemley et al., (13]. Their mineral stability fields at 1 kbar H20, in terms of crystallization temperature and molality of aqueous silica, are given in figure 3. This figure shows a number of relationships pertinent to the problem of formation of "asbestiform" minerals. As the temperature decreases, forsterite (Mg-rich olivine) can react to form antigorite or chrysdtile depending on the silica concentration in the aqueous solutions to which the olivine-bearing rock is exposed. One chemical reaction that may lead to the formation of brucite-bearing serpentinite is: 2 Mg2Si04 + 3H20 4 Mg3Si205(OH)4 + Mg(OH)2 . fosterite chrysotile brucite This reaction may explain the origin of the very long brucite needles, referred to as "nemalite," that are found in various serpentinites. Thirty-centimeter-long needles of this mineral were collected by C.E. Brown (U.S. Geol. Survey) from a Quebec serpentinite locality and were examined by single-crystal x-ray methods (Malcolm Ross, unpub. data). The brucite needles show hexagonal symmetry, a= 0.315 nm, c = 0.474 nm, and the long direction of the needles are parallel to the brucite a-direction. The rather marked line broadening that appears in the x-ray pattern suggests that the brucite needles are composed of many small crystallites oriented so that their a-axes are parallel to the fiber direc- tion. The brucite needles are intergrown with chrysotile, for chrysotile x-ray reflections are superimposed on the diffraction pattern of brucite, and extremely long chrysotile fibrils remain when the brucite needles are dissolved by dilute HN03. 53 2063104854

6~1 T C ---b- Figure 3. Mineral stability relations in the system Mg0-Si02-H20 as a function of log of molality of aqueous silica and temperature, at 1 kilobar H20 pressure. Figure modified from Hemley et al. [13]. At higher concentrations of aqueous silica, forsterite may alter to talc by the reaction: 3Mg2Si04 + 5(H4Si04)aq -• 2Mg3Si4070(OH)2 + 8H2O . At silica concentrations near the quartz saturation curve, anthophyllite can alter directly to talc by the reaction: • 3Mg7Si8022(OH)2 + 4(H2Si04)aq ~ 7Mg3Si4010(OH)2 + 4H2O. This reaction may be of importance for the form#tion of fibrous anthophyllite and talc. As the temperature decreases and the H20, Mg2 , and silica activities remain within geologically reasonable limits, one probable reaction sequence is: enstatite + anthophyllite + talc. If. the alteration of a chain silicate to talc proceeds by an intragranular reaction, "talcbole-type" phases may form as intermediates between anthophyllite and talc during low-temperature alteration [36,37,38]. Figure 4 shows the stability fields of forsterite, enstatite, anthophyllite, and talc in terms of temperature and molality of aqueous silica 54

[13]. A stability (or metastability) field for the "talcboles" (labelled "asbestos") is superimposed on this diagram, overlapping the fields of talc and anthophyllite. The fibrous nature of the "talcboles" can be explained if the alteration process of a chain silicate (anthophyllite) to a sheet silicate (talc) proceeds by reforming the double chairys at the unit-cell level. In figure 4, the phase boundary between enstatite (a pyroxene having the formula Mg2SiY0a) and anthophyllite suggests the possibility of having mixed single chain (pyroxene) and double chain (amphibole) structures. 1.21 (1 kbar H20) OG' -•- QTZ SAT. -•-•-,;- ~ , r- y%r~.,; •-•- TA LC ~'A N T H. FORSTER f TE 2.0 600 640 0 680 720 TC} Figure 4. Adaptation of the enstatite-anthophyllite-talc-forsterite stability relationships at 1 kbar H20 to show a possible stability or metastability field of "talcbole asbestos" (strippled). Figure modified from Hemley et al. [13]. anth. = anthophyllite. The fibrous nature of commercial amosite and crocidolite appears to be related to the crystal growth mechanism; perhaps the crystallites nucleate at many centers and grow as individual fibers only a few tens of nanometers thick (see Franco et al. [8, figures 1,2]). The presence of (100) twinning and "Wadsley" defects may be the result of rapid growth and, in addition, may hinder growth in a direction perpendicular to the fiber axis. Properties of "Asbestos" That May Be Related to Health Risk Health studies suggest that of the four economically important forms of "asbestos," crocidolite has been responsible for the greatest health risk, followed by amosite, then chrysotile, and lastly anthophyllite [11). If we assume that the health hazard caused by the commercial "asbestos" minerals is due to some combination of their chemical, structural, and physical properties, we can make some predictions about their relative biological activity. All commercial "asbestos" minerals separate into very thin fibrils; possible reasons for this have been discussed previously. The thickness, length, and flexibility of the fibrils apparently is important in determining how the fibrils lodge in human tissue and 55 2063104856

how readily they are cleared from the lung areas. The straight fibrils of small diameter, particularly those of crocidolite, can more readily move to the periphery of the lung, where they are in a position to penetrate the pleura and thus produce mesotheliomas [11]. That curly fibrils, especially those of chrysotile, are more readily arrested in the upper respiratory tract is given as a reason for the low incidence of mesotheliomas in chrysotile miners and millers [11,1~J,23]. Assessment of the role of fibril size in relation to lung cancer is less clear [11]; however, Gross [12] cited evidence that "asbestos" fibers less than 5 pm long cause negligible pathogenicity, both of the lung and pleura. The problem of fibril size in relation to cancer incidence is of some importance, for the average ambient airborne "asbestos" fiber is shorter than the average fiber in the whole rock. Brulotte [2] reported that the average concentration of airborne dust particles in the chrysotile mining district of Thetford Mines, Quebec, was 80,500 ng/m3 during active mining and 39,600 ng/m3 during a 5-month period when the mines were closed. If we assume that the rock contains 4 weight percent chrysotile, these measurements suggest a minimum chrysotile dust concentration in the ambient air of 3220 and 1584 ng/m3.2 The total surface area of the inhaled fibrils and the chemical reactivity of this surface may have an important influence in the production of cancer. Researchers have not yet determined whether this surface plays a direct part In the formation of cancerous tissue, or whether a carcinogenic chemical adheres to the mineral surface and the chemical itself later reacts with the tissue or in some way catalyzes the carcinogenic process. The high incidence of lung cancer in men who worked In the "asbestos" trades (textiles, brake-lining fabrication, insulating) and who also smoked [33] indicates that carcinogenic chemicals in the tobacco smoke may somehow interact with the "asbestos" fibrils. If many of the fibrils are not easily cleared from the lung, they may adsorb these chemicals and hold them indefinitely. Injection of "asbestos" fibrils directly into the pleura of animals causes a high incidence of mesothelloma [40]. These experiments suggest a direct relationship between the active fibril surface and production of pleural cancer. However, other dissimilar substances injected into animals also cause tumors; for example, nonfibrous hematite (Fea03), sanidine (KA1Si30s), and corundum (A1203) [27]. As a generalization, the relative chemical reactivity of the exposed fibril surfaces of the four important forms of commercial "asbestos" in aqueous solutions is: chrysotile < anthophyllite < amosite < crocidolite. Chrysotile, the least reactive of the four, is composed of rolled-up layers that possess no broken chemical bonds except where the edges of the layers are exposed at the ends of the tubes. The three amphiboles, on the other hand, have broken chemical bonds on all surfaces of the fibrils. Anthophyllite can alter to various other silicates in aqueous solutions, as has been explained above. Similar alteration mechanisms might also exist for crocidolite and amosite although, to my knowledge, these have not been documented. However, studies of the zCOnversion of these figures (nanograms chrysotile per cubic meter of air) to numbers of "fibers" per cubic centimeter of air (the value usually given in health studies) is estimated by using the following relations: (5) density of chrysotile s 2.5g/cmg = 2.5x19sng/cma volume of I ng chrysotile = 4x10 1Ocma = 400 pm8 volume of chrysotile fibers in umg/cma = nS--~ if a fiber having dimensions 1 pm x 1 pq x 5 pm (5 pm3) is designated as a "standard fiber," then 1 ng chrysotile = 80 "standard fibers" number of chrysotile "standard fibers"/cros = inv~/m66s) . 56

es geochemistry of silicates indicate that the exposed surfaces of these two amphiboles present some interesting possibilities for chemical change. Amosite (and also crocidolite) can undergo oxidation-reduction reactions of the type, FeT+Si8022(OH)2 F Fe5+Fe2+Si802202 + H2 Ernst and Wai [6] have demonstrated that this reaction takes place in iron-bearing sodic amphiboles at 705 °C. The complete reversibility of such a reaction in the chemically similar silicate mineral biotite, has been demonstrated by Wones [42] and by Takeda and Ross [35]. In the experiments of Wones, auto-oxidation was accomplished in a neutral atmosphere (flowing argon) at 500-700 °C. Reduction was accomplished by passing hydrogen gas over the crystals. Analogous reactions can take place at much lower temperatures but also at much lower rates. Cation exchange reactions take place in the amphiboles known as richterites [14]; exchange is accomplished within the A-site of the amphibole structure at 775-850 °C by the reaction: (Na)CaNaMg5si8022(OH,f)2 + K+ « (K)CaNaMg5Si8022(OH,F)2 + Na+. Crocidolite having a partially filled A-site such as that from Bolivia [41] can also undergo exchange reactions with potassium being replaced by sodium and possibly by oxonium and ammonium ions. Crocidolite with a partially or completely vacant A-site may undergo exchange reactions coupled with oxidation-reduction, e.g.: 0 Na2Fe3+Fe23+si8022(oH)2 + R+ + e a(R+)Na2Fe42+Fe +si8022(OH)2 where R+ = K+, Na+, H30+, or NH4, and o= a vacant site. Whether such reactions can take place within aniRal tissue is not known, but the charged and reactive surfaces of croci.dolite and amosite fibrils appear to offer excellent sites or templates for the initiation of complex chemical changes. The surface area available for adsorption is, of course, directly related to fibril thickness or diameter. The specific surface of chrysotile, as measured both by nitrogen adsorption and permeability, is about twice that of amosite and crocidolite [28]. Because chrysotile forms hollow tubes, this larger area for adsorption in chrysotile is predictable if the average fiber thickness is similar for all three minerals. The strain-free layer of chrysotile has a radius of curvature of about 8.8 nm [5]; thus, the minimum diameter of the tube should not be much less than 17 nm. The most frequently measured tube diameter is about 26 nm. Bates and Comer [1] found in a study of chrysotile from Arizona and Quebec a range of diameters from 11.4 to 85 nm; the average diameter was 25 nm. The fiber size ranges in the other forms of commerciat "asbestos" have not come to ry attention, although some crocidolite fibers from Western Australia [8] appear to be on the order of 50 nm wide. "Asbestos" Health Risks in the Mining and Milling Industry and Environs Although a significant health risk for those who work in the "asbestos" trades, particularly for those who smoke, has been well documented, the risk appears to be much lower for those in the chrysotile mining and milling industry and for those who reside in areas of such activity. The most detailed study of an "asbestos" mining community is that of the chrysotile mining areas of Quebec, Canada; the studies were started in 1966 and continue to the present [20-23]. Similar studies of chrysotile miners on a smaller scale have been undertaken by Kogan et al. [16] in the Urals, U.S.S.R., and by Vigliani [39] in Italy. According to McDonald [17,18] these other studies casH: to the same conclusions on health risk as the Quebec studies, the latter of which have led the way in making some assessment of the health risk relative to the amount of dust to which the workers were exposed. Health-risk studies of workers in the "asbestos" trades, for the most part, have not given reliable dust-exposure figures, or even the relative amounts and types of "asbestos" inhaled. 57 2063104858

I Chrysotile has been mined in the Thetford Mines, Black Lake, and asbestos localities of Quebec for nearly a century, beginning in 1886. Production has increased steadily since then, reaching 907,000 metric tons in 1956 and 1,500,000 metric tons in 1976. A tremendous amount of ambient dust has been generated over the years both by mining activities and by the winds blowing over the huge tailings piles. Even in 1974, when dust-emission controls had much improved over those'of the earlier years (72 million particles per ft3 in 1950 to 4 million particles per ft3 in 1975 [20]) as a result of wet drilling, watering of haul roads, etc., emissions of particles from chrysotile mining and milling operations in the Province of Quebec amounted to 140,000 metric tons, of which about 4 percent (5600 metric tons) was "asbestos" dust [2). The ambient dust levels for this region have already been discussed. Is there a high incidence of cancer of the lung and pleura among the 35,000 residents of the Thetford Mines area of Quebec, 10 percent of whom are employed in the chrysotile industry? According to McDonald et al. [17-23], the cancer incidence for the male employees in the Quebec chrysotile industry is similar to that for Canada as a whole and only detectably raised in those with moderate to high levels of exposure. In Table 1 is given the proportional mortality from lung cancer and mesothelioma for the Quebec and North Italian chrysotile miners and millers, and also for the entire populations of various countries in the year 1970. In the period 1936-1973, seven cases of mesothelioma have been reported in the Quebec mining and milling industry [19, Table 12]. The worldwide incidence of inesothelioma in those who worked in the chrysotile mining and milling industry for the period 1958 to 1976 is 11 cases [19, Table 4]. The Canadian studies do show an increased incidence (2.1 to 3.6 times) ff lung cancer for those workers exposed to the highest concentrations of dust -- 400 to 800 mpcf- yr3, but little evidence of health risk from this disease at levels below 200 mpcf-yr. An unusually high number of deaths caused by lung cancer in Homestake gold miners during the period 1960 to 1973 has been reported by Gillam et al. [10]. The cohort consisted of 440 individuals who In 1960 had worked 5 years or more underground. Gillam et al. attributed the high incidence of lung cancer to inhalation of cummingtonite amphibole. They did not specify whether the hornblende amphibole, also present in the rock being mined, contributed to health risk. In rebuttal to this work, McDonald et al. [24] reported on a health analysis of a cohort of 1321 Homestake miners whose working period was from as far back as 1937 to the end of 1973; each of the miners had more than 21 years mining service. Deaths resulting from malignant neoplasm were very close to those expected (93 observed, 90.5 expected); this includes the subcategories of malignant neoplasm -- respiratory, gastro-intestinal, and "other" cancers. The excess death found in the Homestake miners was due in fact to silicosis, silico-tuberculosis, and heart disease. McDonald et al. [24] stated, "The pattern of mortality of men with long employment in this industry indicates a serious pneumoconiotic hazard characteristic of hard rock miners, but not of cancer." Fears [7] has made an epidemiological study of cancer risk, including respiratory cancer, in 97 U.S. counties in 22 states known to be mining chrysotile or amphibole "asbestos." He found no excess of cancer mortality compared with cancer mortality rates in 194 demographically matched counties in which such minerals are not known to be mined; cancer mortality in both groups of counties was significantly below the national average. 3This unit expresses (in millions) the average number of particles (including approximately 4 percent chrysotile) contained in each cubic foot of air inhaled during a worker's career in the mines or mills times the number of years the worker was employed. If the dust is assumed to contain 4 percent chrysotile, then working for 50 years at a dust level of 16 m cf (800 m cf- r) is roughly equivalent to inhaling 23 chrysotile particles for every cma of air taken into the lungs during the employment lifetime. A figure of 200 cf- r is roughly equivalent to 6 particles of chrysotile/cm3. Conversion from dust part c e measurements to chrysotile fibers per cm3 is difficult because chrysotile abundance varies from place to place. 58

Table 1. Proportional mortality from lung cancer and mesothelioma for selected male populations. Cohort Deaths Group No. men All causes % lung cancer % mesothetioma General population a Canada (1970) 82,052 5.3 0.03 USA (1970) 988,620 5.1 0.03 Finland (1970) 22,332 7.1 0.04 Italy (1970) 252,795 4.7 -- England - Wales (1970) 278,617 8.9 0.06 Chrysotile mining-mitling b Quebec (1936-73) 10,951 3,938 5.7 0.18 N. Italy (1932-70) 1.098 270 2.2 0 Anthophyllite mining-milling c Finland (1936-67) 900 216 9.7 0 "Asbestos" trades d Insulators 26,505 2,137 19.6 6.7 Asbestos factory 10,781 1,422 15.0 3.1 a Entire male population over 24 years of age [19, Table 13]. b [19, Table 12; 20, p. 525]. c [19, Table 12]. d Composite figures [19, Table 12]. At present, people are concerned about the possible health hazards associated with the quarrying of serpentine rock at Hunting Hill quarry near Rockville, MD, and its use as a surface material for roads, playgrounds, and parks. The rocks being quarried here are very similar geologically to those of the chrysotile mining localities of Quebec, except that they contain much less chrysotile - about 0.5 weight percent. Rohl et al. [29] from Mount Sinai Hospital reported chrysotile fiber abundances of 500 to 4700 ng/m$ of air sampled adjacent to roads and a parking lot paved with loose crushed stone from the Hunting Hill quarry. The highest figures were measured during "moderate" motor vehicle use. The Mt. Sinai figures are equivalent to 0.2 to 1.9 yms of chrysotile per cm3 of air or 0.04 to 0.4 "standard fibers" per cm3 of air. Air samples taken near the perimeter of the Hunting Hill quarry gave chrysotile mass concentrations of from 0.02 to 64 ng/mg or 2 x 10 to 5 x 10 3"standard fibers" per cros of air (U.S. Bureau of Mines, State of Maryland, and McCrone Assoc. , unpublished data). The present U.S. Government limits for "asbestos" content of air are 2 fibers/cmo (05HA) and 5 fibers/cm3 (MESA) where a fiber is defined as longer than 5 pm, less than 5 ym wide, and having a length-to-width ratio of 3:1 or greater. The publicity about the possible health risk because of dust emission from the Hunting Hill quarry and its rock products had caused the quarry to lose about 30 percent of its business by July 1, 1977. Montgomery County, MD, expected to pay about $2.3 million in its initial effort to seal the roads so as to reduce dust emissions and to remove loose stone from the parks The Council Reoort, Montgomery County, vol. 6, no. 22, July 1, 1977). Apparently, other mining and quarrying operations along the "serpentine belt" of the eastern U.S. from Maine to Alabama also will be considered health risks to the general public [29]. Rohl et al. [29] suggested that exploitation of crushed amphibolite rock also raises the possibility of contamination of the air by "asbestos"-like minerals. 59 2063104860

Discussion The cancer incidence among those employed in the chrysotile mining and milling industry does not appear to be excessive when compared to national populations (Table 1). However, the incidence of cancer among those employed in the "asbestos" trades is very high (Table 1); incidence of lung cancer being 3 to 4 times that of the average population, incidence of mesothelioma being 130 to 220 times that of the average population. The "asbestos" trades generally utilized a variety of "asbestos" minerals including amosite and/or crocidolite, sometimes mixed into a paste for lagging. If we consider that about 90 percent of all the commercial "asbestos" ever mined was chrysotile, and that there is a low incidence of cancer in the chrysotile mining industry, we are led to conclude that either amosite and crocidolite are very hazardous or that there is an additional factor relating to health risk in the "asbestos" trades which has not yet been discovered. Previously, I have discussed some reasons why these two minerals may be more chemically reactive than chryaotile. Definitive epidemiological studies of the amosite mining regions of South Africa and the crocidolite mining regions of South Africa, Bolivia, and Australia appear to be lacking; such studies are needed in order to understand the high cancer incidence in certain trades utilizing these minerals. It is important to point out that the "asbestos" minerals should be considered separately when analyzing their effects on the worker's health. Reasoning by analogy is dangerous; high cancer incidence associated with one form of "asbestos" in a particular occupation does not necessarily mean that there will be the same incidence when utilizing another form of "asbestos" in that or another occupation. Unfortunately, this type of reasoning has led many to assume that any amphibole in any environment will cause high cancer mortality. The operational problems in defining and characterizing fine mineral particles and the unknown health effects on humans by minerals not generally regarded as "asbestos" appear to be causing more and more investigators to accept rather broad definitions for "asbestos." The present analytical techniques used by the EPA and OSHA do not distinguish between amphibole cleavage fragments and the minerals geoscientists generally consider to be true "asbestos." In fact, if electron diffraction is not used expertly, many pyroxenes might be called "asbestos." For example, bronzite, a common orthopyroxene having the composition (Mg,Fe)aSieOa4, is very similar chemically to amphiboles of the cummingtonite- grunerite series, (Mg,Fe)zSie02Y(OH)g. Also, orthopyroxene gives an electron diffraction pattern similar to that of cummingtonite--both patterns possess 0.26 nm spacings between the diffraction row lines in the hoR reciprocal lattice net. A full interpretation of the patterns is necessary for positive identification. Similarly, calcic pyroxenes might be confused with amphiboles of the tremolite-actinolite series or with hornblende. Cumming- tonite (and possibly hornblende) is considered an "asbestos" health hazard by health investigators from the National Institute of Occupational Safety and Health (OSHA), as reported by Gillam et al. [10]. The Mt. Sinai group [29] suggested that crushed amphibole- bearing rocks (amphibolite) used as road-surfacing material may result in widespread "asbestos" contamination of community air. Along with the general use of broader definitions of "asbestos" is a trend toward setting lower and lower limits on the acceptable amount of "asbestos" permitted in the environment (at present the OSHA standard is 2 fibers/cros; the MESA standard is 5 fibers/cm3, but it will soon be changed to the OSHA value). A more stringent "asbestos" health standard is presently being proposed by the National Institute for Occupational Safety and Health (Reexamination and Update of Information on the Health Effects of Occu ational Ex osure to Asbestos, December 1976; document prepared by N OSH or transm ttal to OSHA, as requested by the Assistant Secretary of Labor). This document states (p. 92-93): "Evaluation of all available human data provides no evidence for a threshold or for a safe level of asbestos exposure." "In view of the above, the standard should be set at the lowest level detectable by available analytical techniques----." "Since phase contrast microscopy is the only generally available and practical analytical technique at the present time, this level is defined as 100,000 fibers >5 pm in length/m3 (0.1 fibers/cc)----." 60

es A definition of "asbestos" to include many amphiboles, chrysotile, and possibly other minerals that appear fibrous or acicular in the electron microscope coupled with a fiber- concentration standard of 0.1 fibers/cros should serve to shut down a large number of our hard rock mines and quarries. Also, nothing has yet been said about the effect of such standards on construction workers building highways, tunnels, bridges, or dams on amphibole- bearing rock, nor of the agricultural workers who are exposed to fiber-containing dust while working the croplands. If the present concept of low or "zero threshold" health risk and broad use of "asbestos" definitions continue, much of the crust of the earth could be considered a health hazard. A way of minimizing the effect on the mining industry of the present and proposed "asbestos" standards, yet still maintaining a good level of health safety, is presented by the Canadian studies of the Quebec chrysotile workers. Here J. C. McDonald and his colleagues G. W. Gibbs, A. D. McDonald, M. R. Becklake, J. Siemiatycki, C. E. Rossiter, F. D. K. Liddell, 0. A. El Attar, A. Harper, and many others [17-23] have undertaken not only to delineate areas of health risk in the Quebec environment but also to assess the exposure limits of rock dust where the incidence of cancer and other diseases is at an acceptably low level. No occupation can be considered to have a zero health risk. It would seem that similar studies in this field would be of value in the United States. References [1] Bates, T. F., and Comer, J. J., Further observations on the morphology of chrysotile and halloysite, Clays Clay Min., 6, 237-247 (1959). [2] Brulotte, Raynald, Study of atmospheric pollution in the Thetford Mines area, cradle of Quebec's asbestos industry, ~Atmos heric Pollution, M. M. Benarie, ed., Elsevier Sci. Pub., Amsterdam, 447-458 (1976). [3] Campbell, W. J., Blake, R. L. , Brown, L. L., Cather, E. E., and Sjoberg, J. J., Selected silicate minerals and their asbestiform varieties: mineralogical definitions and identification-characterization, U.S. Bureau of Mines Information Circular 8751, 56 pp. (1977). [4] Champness, P. E., Cliff, G., and Lorimer, G. W., The identification of asbestos, J. Microscopy, 108, 231-249 (1976). [5] Deer, W. A., Howie, R. A., and Zussman, Jack, Rock-formin Minerals, vol. 3, Sheet Silicates, Longmans, Green, and Co. Ltd., London (1 62 . [6] Ernst, W. G., and Wai, G. M., MSssbauer, infrared, x-ray, and optical study of cation ordering and dehydrogenation in natural and heat-treated sodic amphiboles, Am. Mineral., 55, 1226-1258 (1970). [7] Fears, T. R., Cancer mortality and asbestos deposits, Am. J. Epidemiology, 104, 523- 526 (1976). [8] Franco, M. A., Hutchison, J. L., Jefferson, D. A. , and Thomas, J. M. , Structural imperfection and morphology of crocidolite (blue asbestos), Nature, 266, 520-521 (1977). [9] Gary, Margaret, McAfee, Robert, Jr., and Wolf, C. L., Glossar of Geology, (Am. Geological Inst., Washington, D. C., 1972). [10] Gillam, J. D., Dement, J. M., Lemen, R. A., Wagoner, J. K., Archer, V. E., and Blejer, H. P. , Mortality patterns among hard rock gold miners exposed to an asbesti- form mineral, Annals. N.Y_ Acad. Sci., 271, 336-344 (1976). [71] Gilson, J. C., Asbestos cancers as an example of the problem of tomparative risks, Inserm Symposia Series, 52, IARC Scientific Publications No. 13, Environmental aion and Carc ; genic RTsks, 107-116 (1976). 61 2063104862

[12] Gross, Paul, Is short-fibered asbestos dust a biological hazard? Arch. Environ. Health, 29, 115-117 (1974). [13] Hemley, J. J., Montoya, J. W., Shaw, 0. R., and Luce, R. W., Mineral equilibria in the Mg0-Si0Z-H20 system:Il Talc-antigorite-forsterite-anthophyllite-enstatite stability relations and some geologic i.plications in the system, Am. J. Sci., 277, 353-383 (1977). [14] Huebner, J. S. and Papike, J. J., Synthesis and crystal chemistry of sodium-potassium richterite, (Na,K)NaCaMgsSisOa$(OH,F)2: A model for amphiboles, Am. Mineral., 55, 1973-1992 (1970). [15] Hutchison, J. L., Irusteta, M. C., and Whittaker, E. J. W., High-resolution electron microscopy and diffraction studies of fibrous amphiboles, Acta Cryst. , A31, 794-801 (1975). [16] Kogan, F. M., Guselnikova, N. A., and Gulevskaya, M. R., The cancer mortality rate among workers in the asbestos industry of the Urals, aig. Sanit. 37, 29-32 (1972). [17] McDonald, J. C., Cancer in chrysotile mines and mills, Biological Effects of Asbestos, Lyon, International Agency of Res. on Cancer, 189-194 (1973a). [18] McDonald, J. C., Asbestosis in chrysotile mines and mills, Biological Effects of Asbestos, Lyon, International Agency of Res. on Cancer, 155-159 (1973b). [19] McDonald, J. C., and McDonald, A. D., Epidemiology of mesothelioma from estimated incidence, Preventive Med., 6, 426-446 (1977). [20] Mc0onald, J. C., and Becklake, M. R; , Asbestos-related disease in Canada, Hefte z. Unfallheilkunde, 126, 2. Oeutsch-Osterreichisch-Schweizerische, Unfalltagung in er i$`nT~ pringer-Verlag, Berlin, 521-535 (1976). [21] McDonald, A. 0., Harper, A., El Attar, 0. A., and McDonald, J. C., Epidemiology of primary malignant mesothelfal tueors in Canada, Cancer, 26, 914-919 (1970). [22] McDonald, J. C., McDonald, A. 0., Gibbs, G. W., Siemiatycki, J., and Rossiter, C. E., Mortality in the chrysotile asbestos mines and mills of Quebec, Arch. Environ. Health, 22, 677-686 (1971). [23] McDonald, J. C., Becklake, M. R., Gibbs, G. W., McDonald, A. D., and Rossiter, C. E., The health of chrysotile asbestos mine and mill workers of Quebec, Arch. Environ. Health, 28, 61-68 (1974). [24] McDonald, J. C., Gibbs, G. W., Liddell, F. 0. K., and McDonald, A. 0., Mortality after long exposure to cummingtonite-grunerite (abstr.), Am. Rev. Resp. Disease, Supp., 115, No. 4, 230 (1977). [25] Nord, G. L., Jr., "State-of-the-Art" of the analytical transmission electron micro- scope, in Proc. 5 osium on Electron Microscopy and x-ray a lications to environ- mental and occupat oni al health a~es, Ann Arbor Sc~Publ., in press 1978). [26] Papike, J. J., and Ross, Malcoi., Gedrites: Crystal Structures and intracrystalline cation distributions, Am. Mineral. 55, 1945-1972 (1970). [27] Pott, F. and Friedrichs, K. H., Tumoren der Ratten Nach I. P. Infektion faser formiger Staube, Naturw., 59, 318 (1972). [28] Rendall, R. E. G., The data sheets on the chemical and physical properties of the U.I.C.C. standard reference samples, in Pneumoconiosis, H. A. Shapiro, ed., Oxford U. Press (1970). [29] Rohl, A. N., Langer, A. M., and Selikoff, I. J., Environmental asbestos pollution related to use of quarried serpentine rock, Science, 196, 1319-1322 (1977). 62

CS [30] Ross, Malcolm, Papike, J. J., and Weiblen, P. W., Exsolution in clinamphiboles, Science, 159, 1099-1102 (1968a). [31] Ross, Malcolm, Smith, W. L., and Ashton, W. H., Triclinic talc and associated amphiboles from the Gouverneur Mining District, New York, Am. Mineral., 53, 751-769 (1968b). [32] Ross, Malcolm, Papike, J. J. , and Shaw, K. W., Exsolution textures in amphiboles as indicators of subsolidus thermal histories, in Mineral. Soc. Am. special paper no. 2, J. J. Papike, ed., pp. 275-299 (1969). [33] Selikoff, 1. J., Hammond, E. C., and Churg, Jacob, Asbestos exposure, smoking, and . neoplasia, J. Am. Med. Assoc., 204, 106-112 (1968). [34] Seshan, K. and Wenk, H.-R., Identification of faults in asbestos minerals and applica- tion to pollution studies, in Proc. Electron Microscope Soc. Am., 34th annual meeting, G. W. Bailey, ed., 616-617 (ClaT-to-F's Pub. Div., Baton Rouge, LA, 197-6-T [35] Takeda, Hiroshi and Ross, Malcolm, Mica polytypism: Dissimilarities in the crystal structures of coexisting 1M and 2M1 biotite, Am. Mineral., 60, 1030-1040 (1975). [36] Veblen, D. R. and Burham, C. W., Triple-chain biopyriboles: Newly discovered inter- mediate products of the retrograde anthophyllite-talc transformation, Chester, Vt., (abstr.), Trans. Am. Geophys. Union, 56, 1076 (1975). [37] Veblen, D. R. and Burnham, C. W., Biopyriboles from Chester, Vermont: The first mixed-chain silicates (abstr.), Geol. Soc. Am. Abstracts with Programs 8, 1153 (1976). [38] Veblen, D. R. , Triple-and mixed-chain biopyriboles from Chester, Vermont, Ph.D. thesis, Harvard University, Cambridge, Mass. (1976). [39] Vigliani, E. C., Asbestos exposure and its results in Italy, in Proc. International Conf. on Pneumoconiosis, Johannesburg, Oxford U. Press, 192-196 (1970F- [40] Wagner, J. C., Berry, G. , and Trimbrell, V., Mesotheliomata in rats after inoculation with asbestos and other materials, Brit. J. Cancer, 28, 173-185 (1973). [41] Whittaker, E. J. W., The structure of Bolivian crocidolite, Acta Cryst., 2, 312-317 (1949). [42] Wones, 0. R., Physical properties of synthetic biotites on the join phlogopite annite, Am. Mineral. 48, 1300-1321 (1963). Discussion NOTE: Discussion of this paper was included in the General Discussion at the end of this session. 63

CS National Bureau of Standards Special Publication 506. Proceedings of the Workshop on Asbestos: Definitions and Measurement Methods held at NBS, Gaithersburg, MD, July 18-20, 1977. (Issued November 1978) GENERAL DISCUSSION OF MINERALOGICAL ASPECTS L. SWENT: Homestake Mining Company management is very aware of the implications that the Homestake mine study referred to by Mr. Ross will have for industry. We believe that we have a serious responsibility to see that a study is done and that it is a properly done study. The first study, done by NIOSH without consultation with Homestake Mining Company, was published in June 1976, and contained a number of serious defects of procedure, assumptions, and reasoning, which make its conclusions invalid. As a result, NIOSH and Homestake Mining Company have entered into a cooperative arrangement for a second study. The mortality analysis part of the study has been con- tracted to SRI International. NIOSH has begun the environmental sampling work in the mine, and SRI has started reviewing the Homestake personnel records for the mortality study. Anyone interested in reading a critique setting forth the defects which invalidated the conclusions of the first NIOSH study may obtain a copy by writing to: L. W. Swent, Vice President-Engineering, Homestake Mining Company, 650 California Street, San Francisco, California 94108. W. DIXON: I wanted to ask Malcolm Ross if he has studied fibers which are inter- mediate between talc and anthophyllite in their characteristics and composition? M. ROSS: Yes, that is the work of Veblen, Buseck, and Burnham; their papers on this will be coming out within the next few months (Science, Vol. 198, p. 359-365). These minerals are intermediate chemically and structurally between anthophyllite and talc. They have been found in two or three places; I'm sure we'll probably find more. DIXON: I'd like to make a general request that if anyone participating in this conference has comments to make on the toxicity of those types of materials mentioned above I would be glad to hear of any information that might be available. NOTE: No response was received to this request. (CCG). , R. LEE: I would like to make a comment on a couple of things. First is the outward morphology of amosite versus cleavage fragments; it's generally been written in the literature, which I've seen, that they're indistinguishable. This is, I think, the way a lot of people look at it. We've been doing some studies on amosite, penge amosite versus grunerites, and we find that indeed in the amoslte it's generally a (100) face when you get a single crystal diffraction pattern near 0,0 on your microscope. In the grunerites, they tend to lie about 28° away from this, which puts them on a (110) face, in other words a cleavage plane. The second comment is that our studies on the size distributions of airborne particles show that the aspect ratio of airborne serpentines and very fibrous amphiboles tend to be much, much larger than the size distribution of the corresponding cleavage fragments which were airborne. Something like a minimum of 30 to 1, or an average of 30 to 1 for the particles we observed in an electron microscope, versus about 7 or 8 to 1 for amphibole fragments. But the point I want to make is that we should not only be looking at the health effects, we should be making sure that we know whether we are looking at cleavage fragments or at amosite. ROSS: To add to this, Ann Wiley brought up one clue as to whether amosite or grunerite is really similar to the penge amosite from South Africa. Do the minerals have parallel extinction at the very highest optical magnification? Most of the garden variety cummingtonite-grunerite minerals have inclined extinction; even for the individual crystallites. The parallel extinction is caused by small lamellae randomly oriented about the fiber axis. Optically the specimen looks orthorhombic; optical observation is the Preceding page blank 65 2063104865

first technique to use in order to get an idea whether an amphibole may be similar to the known commercial asbestos. N. TATE: I wonder if you know of Judge Bowder's investigation among the miners in Quebec, where he's found very heavy incidence of disease which was not previously reported. Figures range-from 45 percent among workers, nonsmokers with low exposures, up to 70 percent lung changes in workers with heavy exposures. I also had the opportunity of talking to Prof. McDonald just before I left London. He has a new study which will be published shortly; he says he's found excess disease among the miners at Thetford, half of it the normal asbestos diseases and half of it shows that asbestos workers have lower resistance to all disease. These are two studies which I think should be taken into account. ROSS: Certainly, that's why I want to bring out the Canadian work. It should be taken into account; but you have to recall that these men have been exposed to heavy dust. Friends of mine who go there on geological field trips tell me that up until recently people would hose down the windows in the morning to see out of them, that's how thick the dust was up there. They have, in the past, gotten tremendous amounts of dust in their lungs. Now what the Canadian study is attempting to do is to divide the workers into what they consider low, intermediate, heavy, and very heavy exposure levels to see If they can see a difference in health risk. Now the reports I've seen indicate that below 200 mpcf- yr there's a very low health risk, but all I know is what I read in their papers. I want to point out that somewhere we have to find a tolerable health risk or we'll have to close down the surface of the earth. A. SUNDARAM: Dr. Ross, I'm wondering how you graded the various types of asbestos in relation to the toxicity or pathogenicity? There are at least four distinct types of pathogenicity arising from asbestos exposure: asbestosis, lung cancer, mesothelioma, and cancer of the gastrointestinal track; also it is claimed in other organs. When you graded it so easily: crocidolite, amosite, anthophyllite, and chrysotile, did you do the gradation yourself or are you quoting any paper? ROSS: I'm quoting Gilson. SUNDARAM: And is the gradation based on animal data or epidemological data? ROSS: I can give you the reference (Inserm Symposia Series 52, p. 107-116 (1976)); it's a summary paper by Gilson where he suggested this generalization. Perhaps you can find something wrong with it, but it was a generalization. I made an additional generali- zation that the chemical activity of these four minerals seemed to be similar in that crocidolite can undergo on the surface more chemical reactions than amosite, and amosite more than anthophyllite, and chrysotile being the least chemically reactive. I'm just pointing this out as a generalization, something to start from; maybe it might give some clues for the formation of cancer, I don't know. It may not be that it is the only factor, because the shape and the aerodynamics are apparently very important, and the lung clearance functions are very important, so there are many parameters that have to be taken into consideration. The chemical reactivity of the surface is one of them. I believe that the chemical reactivity of the surface is important. Consider a standard fiber lxlx5 pm in size. There will be 100 times more surface area if you divide a standard fibeir into 10,000 smaller fibers. So one big fiber might be a 100 times less effective, as far as the surface chemistry is concerned, than 10,000 small ones - yet they both would have the same weight in nanograms. SUNDARAN: So you mean to say that the gradation is based on chemical reactivity and not on any toxic parameter? ROSS: Well, I'm saying chemical reactivity may enter into the toxic parameters. What causes lung cancer? Does the fiber interreact with a chemical such as in tobacco smoke and then with the human tissue, and so forth? Does the fiber interreact directly with the human tissue chemically? I'm basically getting down to a chemical answer in the end. 66

B. WHITE: As you know we are in the process of putting together so called emergency regulations, relative to the Rockville Quarry. Now these regulations deal primarily with the containment of the crushed stone. You're inferring that you feel that this sort of approach is not indicated based on the Canadian work? ROSS: The Canadian work would suggest there is not a health danger with this level of asbestos dust. Now all the data are not in. What we would need is ambient air measurements in the Rockville area. Dr. Selikoff suggested, at the National Institutes of Health hearing on this a few weeks ago, 45 nanograms is the limit in ambient air. What level of ambient air do you want to have for chrysotile? I haven't seen an ambient air figure for the Washington, D.C. area. I don't know what it is. I'm really pointing out that we can shut down all the serpentinite quarries on the East Coast. If it's serpentinite it is going to have some chrysotiie in it. But then, where do we go from there? We also have tremolite; we can shut down other mines because of tremolite or because of fibrous hornblende and on and on and on. Now I think that I'm pointing out, from a mineralogical and geological point of view, that this is an immense problem. EPA is now getting set up to get crusher runs on mines and quarries all down the East Coast. It's going to run into millions of dollars. It's already running into millions in the Montgomery County area. Now I think that the health people have got to get together and decide what they're going to call asbestos, what dust levels are going to be considered dangerous, and what sort of mining operations they think they are going to have to shut down. You can shut down a mining operation very easily by putting so many requirements on it that the contractors say, "heck with it, I'll go to Frederick and get carbonate rock." I'm pointing out it's an immense problem, it's economic, it's political, it's health, and so forth. WHITE: I agree with you very much; our intent is certainly not to close down the mine, and I agree also that the health people must come to grips with the issue of the ambient air. Now obviously since there are no standards, our approach is purely on the mechanical side of this, which is trying to reduce the dust emission as much as possible and, quite frankly, I feel until there is more data on the amount that can be floating around in the air that this is a very sensible approach, a preventive approach actually of dealing with the problem. Even though there is nothing that one can hang the hat on from the health side, I personally think that to allow the crushed rock to be used indiscriminately is just simply not a good approach to preventive medicine. Thank you. R. DAVIS: We live in a complex world and you pointed out that contractors might use carbonate stone. A number of the state highway departments have shown that carbonate contributes to lower skid resistance. We are faced with the problem of how many people are going to die from cancer from the chrysotile type of material and how many are going to die from lowered skid resistance on the highways. These are very complex problems. ROSS: What makes it so frightening is if you pull a string and all of a sudden a lot more string comes out, you don't know whether you've increased health risk or decreased it. You've decreased it in one area and perhaps increased it in the other. One possibility is that people would be so scared of asbestos, they won't use it for anything. Asbestos has saved many lives when used for fireproofing. We could carry on with fiberglass which has a lot of similarities to asbestos, or we can get rid of fiberglass, and we can insulate with organic chemicals, like some that form carbon monoxide and HCN when they burn. The total picture is a big one and I think that we all should try to get a feeling for the entire situation, and consider some of the problems that could arise. E. COX: I'd like to ask M. Ross or Dr. Zoltai if you could tell us when the first commercial mining of asbestos took place, what type it was, and where it was? T. ZOLTAI: About a couple of thousand years ago; on a cowserciai scale the major mines started in the late 19th century. COX: About 1880-1890? ZOLTAI: Yes. COX: And where were they, sir? 67 N ~ ~ a v

ZOLTAI: In Canada. COX: In Canada, and what were they? ZOLTAI: Chrysotile.7 K. HEINRICH: I'd like to ask Malcolm Ross if you have information of the size distribution of chrysotile in Thetford and if it is similar to that in Montgomery County? ROSS: Tom Bates did a size study of chrysotile from the Thetford area, Canada, and also on the beautiful chrysotile from Arizona. I have the figures in my paper but I think in the Canadian chrysotile he had a minimum o 110 R outside diameter and a maximum of several hundred with an average of about 250 ~. I think the Arizona chrysotile had a generally larger diameter. You meant length, I'm sorry, I was thinking of width. I don't know that, I don't have that figure. Some of the Canadian chrysotile was in beautifully long fibers. This material set the chrysotile industry off, because in 1886 they found these exceptionally good types of asbestos. I imagine some of it was very long fiber material, but of course much of it would be short fiber also as the Rockville chrysotile is. J. ZUSSMAN: I have two comments and one question. One is the point about when some commercial use of asbestos started. I believe there is some record of something industrial in Italy with products like asbestos paper. There is also mention of the manufacture of asbestos socks and gloves at a place in Russia. These were both before the start of large scale mining at Thetford. Another comment is in connection with Or. Ross's remarks about the reactivity of various forms of asbestos, in which he put chrysotile low down on that scale. In one sense perhaps chrysotile is high up in the scale of reactivity in that it is less resistant to acid, and quite dilute acids can attack and start to dissolve away chrysotile. It has a rather exposed layer of magnesium hydroxide and this is obviously going to be quite reactive to dilute acids. I am not sure whether its reactivity in this sense makes chrysotile less or more physiologically harmful. I'd like to ask one question of Dr. Ross about the synthesis. I was very interested to hear of his colleague Or. Hemley's work on stability fields of the serpentine and amphibole minerals, and I would like to ask whether or not the chrysotile or amphibole formed was asbestiform or not. Quite a lot of work has been done on the stability fields of amphiboles and serpentines in general, but rather little pinpointing when long thin chrysotile fibers form and when other serpentines like lizardite and antigorite form; also when asbestiform and when non-asbestiform amphiboles form. I wonder if the products of those experiments were identified as asbestiform or not. ROSS: Yes, Dr. Hemley's work, I think, was really one of the outstanding contribu- tions we've had in this area of geochemistry this year. These experiments were very difficult; they are run at relatively low temperatures, so his run times were many weeks duration. Concerning the stabilities of the individual polymorphs of serpentine, he attempted to define an antigorite and chrysotile field. He did some electron microscopy, I believe, and found platy-serpentine, which he called antigorite. I asked Julian Hemley - "if you injected some chrysotile into the human blood stream or into the lung, what would you expect to happen?" He thought about the various parameters in the human body that might affect that system and he said, "I don't think anything would happen." Nevertheless, chrysotile is very soluble in dilute acids, and Dr. Langer will agree ingested chrysotile in the stomach should decompose quite readily. Hemley did not think that the pH range of the human body, other than the stomach, would contribute to any appreciable dissolution of the chrysotile. 0. MENIS: He suggested it would last and last. Being a chemist, I would like to ask the mineralogist why they have neglected would like materials, the to and OH group, the hydroxylation process. I wonder if Prof. Zussman and others comment an the role of the OH, the potential of local pH values of these the ease of the hydroxylation which is known from thermal data where you N have a great difference between the various amphiboles and chrysotile. .~i 68

:IO CS ZUSSMAN: I look to my colleagues because I really don't know much about it, and I don't know that very much is known about the comparative effects of the hydroxyl in these different minerals. Certainly in the amphiboles, and crocidolite in particular, some work has been done on oxidation-reduction phenomena, because there you have not only the hydroxyl but you have the ferric ion and the combination of the two is conducive to chemical reac- tions going on. I don't know of any work which has examined the effect of OH in the grunerites or very much in serpentine except with regard to decomposition. If you heat them, then they break down at different temperatures, and you mentioned the question of differential thermal analysis giving different results. I think one has to be very cautious about this because it's notoriously easy for a variety of results to be obtained in the decomposition temperatures by DTA methods, which may or may not be significant. In the amphiboles, hydroxyl is there, and so is fluorine (I hadn't mentioned that because there was a limit to the complication that one could go into in the time available for my paper), but it's quite possible that the ratio of hydroxyl to fluorine, the presence of fluorine or the presence of chlorine could be relevant. These are all minor variables and there has not been much systematic study of how many of these variables are relevant to the comparison of asbestos and non-asbestos amphiboles or serpentines, and their effects. Malcolm Ross may have some comments on this. ROSS: If you pass hot, inert gas over grunerite crystals, hydrogen will be removed and you'll get two atoms of trivalent iron. This is quite reversible, at least in other similar phases. Ernst and Wai have done this experiment with sodic amphiboles. Repeated experiments on biotite by Wones shows complete reversibility of this oxidation-reduction reaction. In amosite as well as crocidolite, the iron may be oxidized by removal of hydrogen. This can go very readily at higher temperatures. It is unknown whether this can go on in the human lung, but it is a possible chemical reaction. Also another reaction is ion exchange in crocidolite. You can oxidize or reduce the iron, and exchange oxonium, ammonium, potassium, or sodium in the vacant site. Thus there are some very interesting possibilities for chemical change on the surface of these crystals. J. KRAMER: I might make a comment. I think back to the original question on chemical reactivity. One of the ideas of looking at surface reactions in the amphiboles originally was that these crystallites forming the asbestos form of the amphibole may be hooked together with 0H0 bonds, and we thought we might see some differences here. Our type of measurements which I quickly alluded to, are crude. They're gross and are in no way domain measurements. We didn't find any differences. The other thing is of course that chrysotile versus the amphiboles has a much different zero point of charge, quite a bit different double layer in terms of surface reactions. One might want to compare these two groups in order to look at reactions involving the hydroxyl groups. But I think maybe Dr. Zoltai may like to comment upon some of his surface charge measurements because I think these are much more specific to the individual fiber. I'd like to hear your comments. ZOLTAI: Actually we haven't done any sophisticated work to be able to answer a question of that level. All I can say is that what we were trying to do was to detect surface charges at the level of single fibers rather than in bulk quantities of fibers. By using distilled water containing positive or negative labelling sols in suspension, we tried to detect the surface charges of amosite from South Africa and non-asbestiform cummingtonite. In other words, there was only one experiment, and in that case the asbesti- form material appeared to have much higher negative surface charge. However, the two specimens came from two different localities, besides being only one test that could not be considered very meaningful. Actually, the reason we did that was to see whether the technique is applicable to asbestos. It would be very nice to have a technique where you can get an indication of the surface charge at the scale of single fibers. KRAMER: Did you notice any domains pertinent to your technique? ZOLTAI: Occasionally, yes. UNKNOWN: I'd like to ask Dr. Zussman a question. Have you any way of estimating what fraction of the total amphibole structure might be defective, what are the length dimensions of the defects, and how much of a chemical variation would you expect to be associated with the defects that you outlined? 69 2063104869

ZUSSMAN: The little work that has been done on this shows the frequency of defects in the limited number of samples that have been looked at and, in some of the ones I can remember, the defect occurred about one every 50 cells, so it was a small proportion in that particular sample. Other samples may show a much higher density of defects, but I think just not enough samples have been looked at in that respect. As to the importance of defects, they coulb be very important in terms of crystal growth, and in terms of mechanical properties. Perfect crystals without defects have very different tensile strengths and other mechanical properties compared with crystals from the same substance but with defects, and it's conceivable that chemical reactivity may be concentrated at the sites of defects. It's an area which is not being looked into to my knowledge; perhaps somebody else can say otherwise. Added after meetin : My answer above about the density of defects was related to Wadsley defects. I omitted to say that the other kind of defect (stacking and twinning) have been reported as very abundant in crocidolite, amosite and tremolite asbestos. Only the Wadsley type of defect would have a direct effect on chemical composition, but it would be rather small if there are relatively few of them. SUMMARY: Dr. Mason, the session chairman, indicated that he felt the General Discussion provided a very adequate summary of the mineralogical aspects. 70

CS National Bureau of Standards Special Publication 506. Proceedings of the Workshop on Asbestos: Definitions and Measurement Methods held at NBS, Gaithersburg, MD, July 18-20, 1977. (Issued November 1978) EPIDEMIOLOGICAL EVIDENCE ON ASBESTOS William J. Nicholson, Arthur M. Langer, Irving J. Selikoff Environmental Sciences Laboratory Mount Sinai School of Medicine of the City University of New York One Gustave L. Levy Place New York, New York 10029 Abstract Data on the human health effects from occupational and environmental exposure to asbestos will be presented with special emphasis on the role of different asbestos minerals. Further, human tissue burdens of fibers and their association with asbestos related diseases will be discussed. Experimental animal data from various species and utilizing different routes of administration will also be presented, again with emphasis on differing fiber types. Key Words: Asbestos; cancer; epidemiology; fibers; mesothelioma; occupational exposure. PART I. HUMAN HEALTH EFFECTS We have already heard in the session on mineralogical aspects of asbestos considerable comment and speculation about health effects. What I would like to do here is present some data on human health effects associated with different forms of asbestos, and to discuss briefly some of their meaning in terms of ambient air concentrations. The modern history of asbestos disease dates from the turn of the century, when two reports were published documenting the effects of uncontrolled conditions in asbestos textile factories. One, the testimony of Dr. H. Montague Murray at a compensation hearing, described severe pulmonary fibrosis found at autopsy, in 1900, in the last survivor of a group of ten workers first employed 14 years previously in a carding room [1]i. The second was the description by Auribault of deaths during the early years of operation of an asbestos weaving mill established at Conde-sur-Noireau, France, in 1890 [2]. During this period 50 men died, including 16 of 17 recruited from a cotton textile mill previously owned by the factory director. Subsequently, cases of pulmonary fibrosis following inhalation of asbestos were published in the medical literature, including one by Cooke, who gave the disease its cur- rent name, asbestosis [3]. A 1929 study of asbestos textile operations by the British Factory Inspectorate revealed the existence and extent of a continuing problem [4]. In a clinical survey of mill employees, 80 percent of those employed for 20 years or more had x-ray evidence of asbestos disease. This finding stimulated the Factory Inspectorate to require the introduction of extensive environmental control technology in the industry and the establishment of an ongoing medical surveillance program. Conditions in the United States were not improved significantly until the 1960's and in recent years the prevalence of abnormal x-rays among workers with 20 or more years of occupational exposure to asbestos has been high. Table 1 lists data of such abnormalities found among insulation workmen employed in the New York and New Jersey area prior to 1960 [5]. Most x-rays of the group were normal until 20 years, and if abnormal usually showed 1Figures in brackets indicate references at the end of each part of this paper. There is also a set of references following the discussion. 71 2063104871

Table 1. X-ray changes in asbestos insulation workers. Asbestosis (grade) Onset of Percent Percent exposure (yrs.) ~ No. normal abnormal 1 2 3 40+ 121 5.8 94.2 35 51 28 30-39 194 12.9 87.1 102 49 18 20-29 77 27.2 72.8 35 17 4 10-19 379 55.9 44.1 158 9 0 0-9 346 89.6 10.4 36 0 0 Total 1,117 366 126 50 changes only of minimal extent. However, after 20 years most had abnormal x-rays and, when abnormal, often of significant degree. Thus, long term observations are required to obtain a valid assessment of lung scarring associated with asbestos exposure. Analysis of short- term data can be highly misleading. Asbestosis was the only disease known to be present among occupationally exposed workers until 1935, when it was suggested that lung cancer might be associated with asbestos exposure. In that year and again in 1936 a clinical report was published of lung cancer in an asbestos worker who had died with evidence of pulmonary fibrosis [6,7]. While such reports were not sufficient to causally relate asbestos exposure to lung cancer, the pos- sibility was raised. In 1947 it was confirmed by substantial data, which showed that 13 percent of individuals who died with asbestosis in Great Britain also had bronchogenic carcinoma [8]. Mesothelioma, a rare tumor of the lining of the abdomen or chest, was described in an asbestos worker in 1953 [9], found frequently to have followed potential asbestos exposure in 1960 [10], and unequivocally related to such exposure in 1965 [11]. Gastronintestinal cancer also was found to be in excess among asbestos insulation workers in the United States [12]. In 1975, three-quarters of a century after the first identification of asbestos-related deaths, society continues to be plagued by their presence, unfortunately, in ever increasing numbers. Moreover, the population at risk from the several asbestos-related cancers has expanded from those directly handling the mineral to those working nearby the application or removal of asbestos materials, and, finally, to those who simply live in the vicinity of an asbestos operation or in the household of an asbestos worker. Hiah Exposure Effects The full spectrum of disease from asbestos exposure is best manifest in the data of Selikoff, Hammand, and Seidman on the mortality experience of 17,800 asbestos insuiation workmen [13]. Table 2 shows the expected and observed deaths among this group of workers from January 1, 1967, through December 31, 1976. Among those individuals who have died, one in five deaths was due to lung cancer, about 5 percent to gastrointestinal cancer, approxi- mately 7 percent to mesothelioma (a tumor so rare in the general population that it may account for only one in ten thousand deaths in the absence of exposure to asbestos), 10 percent to other cancers, and 7 percent to asbestosis, the disease first characterized seven decades earlier and wished away numerous times subsequently. The data on the mortality experience of this group of workmen are also sufficient to suggest that cancer at sites other than those mentioned above may also be increased from asbestos exposure. Here, however, the malignancies are less common. Overall, comparing the frequencies of deaths from the cancers and asbestosis with those among the general population, nearly 40 percent of the deaths in this group of workers can be attributed to their occupational exposure to. asbestos. N , W r ~ N 72

Table 2. Deaths among 17,800a asbestos insulation workers in the United States and Canada. January 1, 1967 - December 31, 1976 Number of men Man-years of observation Expected 17,800 166,855 Observed Ratio Total deaths, all causes 1,660.96 2,270 1.37 Total cancer, all sites 319.90 994 3.11 Lung cancer 105.97 485 4.58 Pleural mesothelioma b 66 -- Peritoneal mesothelioma b 109 -- Cancer of esophagus 7.01 18 2.57 Cancer of stomach 14.23 22 1.55 Cancer of colon, rectum 37.86 59 1.56 All other cancer 154.83 235 1.52 Asbestosis All other causes b 1,351.06 162 1,114 0.82 a Expected deaths are based upon white male age specific mortality data of the U. S. National Center for Health Statistics for 1967-1975 and extrapolation to 1976. b These are rare causes of death in the general population. From: Selikoff, I. J., Hanmond, E. C., and Seidman, H., Mortality experience of insulation workers in the United States and Canada, 1943-1977, to be published, Ann. N.Y. Acad. Sci. Asbestos related disease has also resulted from exposures in asbestos factories. A study of production employees of the largest asbestos products manufacturing facility in the United States again demonstrated the presence of significant excess disease [14]. In this study, the mortality experience of all 689 individuals who were working on January 1, 1959, and who were first employed prior to 1939, was analyzed. From 1959 to 1976, it was expected that 188 deaths would have occurred in this group. Instead, 274 died, 46 percent more than anticipated. About 40 cancers were expected; 99 were observed. As shown in Table 3, the anticipated asbestos-related tumors were found in excess - bronchogenic carcinoma, mesothelioma, and gastrointestinal cancer. 73

Table 3. Expected and observed deaths among 689 factory workers, employed before January 1, 1939, during the seventeen years from January 1, 1959 through December 31, 1975. ~ -- -- Observed - 1959 - 1975 Expected -- - -- Obs. Ex . All causes 274 188.19 1.46 Cancer, all sites 99 39.93 2.47 Lung cancer 35 12.53 3.91a Pleural mesothelioma 14 n.a. -- Peritoneal mesothelioma 12 n.a. -- Cancer of esophagus, stomach, colon, and rectum 15 7.99 1.88 Cancer all other sites 23 19.40 1.19 All respiratory disease 42 12.16 3.45 Asbestosis 35 n.a. -- Other respiratory 7 b -- A11 other causes 133 136.11 0.98 Person-years of observation 9,646 a Pleural mesothelioma inctuded with cancer of bronchus in calculating ratio since expected rates are based upon "cancer of lung, pleura, bronchus, trachea." b This rate is virtually identical with that of "all respiratory disease." n.a. = not available. From: Nicholson, W. J., Case Study 1: Asbestos-the TLV approach, Ann. N.Y. Acad. Sci., 271, 152-169 (1976). Time Effects - Lapsed Period If one considers the time from onset of exposure to the clinical evidence of disease, one finds, just as with asbestosis, that there is a long-lapsed period from first exposure to appearance of asbestos related cancers. Data from the group of insulators illustrate this point in figure 1, where the excess cancer risk, calculated for equal but not aged standard- ized populations within each ten-year time interval, is plotted. A significant increase 9n risk is seen only after 25 years for lung cancer and after 30 years for mesothelioma. An increase in the ratio of observed to expected cases of the various asbestos cancers occurs prior to 20 years, but the total number of such cancers is small, as the population is relatively young. This long-lapsed period creates significant difficulties in attempting to establish dose-response relationships. The disease seen today is from exposures decades past when few .easurements were made of asbestos concentrations. Thus, we can only estimate past expo- sures, based on current knowledge. Further, such estimates can be unreliable, and the determination of the efficacy of standards based upon them cannot be made with certainty, until further decades have past. If we then find serious misjudgments have been made, asbestos disease will continue to plague us well into the twenty-first century.

TIME FROM ONSET OF EXPOSURE (YEARS) Figure 1. The excess, asbestos-related mortality rates for lung cancer and mesothelioma according to time from onset of asbestos disease. 75

Another aspect of time in the identification of carcinogens is seen in the data from the study of New York and New Jersey insulation workers over the period 1943 through 1973 [15]. Table 4 shows the mortality experience of 623 insulators, all with 20 years since first exposure in different time periods. One notable feature in these data is the deficit of deaths of all causes-in the first 10-year observation period; an excess of total mortal- ity appears only after several years from first observation (and 30 years from onset of exposure). It is common to observe such a deficit, often as great as 25 percent, in studies comparing the mortality experience of working groups with that of the general population the "healthy worker effect"). This results in part because identified groups of workmen are healthier than a corresponding age group in the general population, which would include terminally ill individuals and others unable to hold a job because of disability. However, even in these early years, the excess asbestos cancers can be seen, although they are not yet the dominant contribution to total mortality. Synergistic Effects A second important concern is increasing evidence that many cancers may have a multiple factor etiology. For example, lung cancer in asbestos workers is strongly associated with cigarette smoking. In the large cohort of 17,800 insulators observed by Selikoff and Hammond, the smoking habits were obtained on the majority of workers in 1967 [16]. Table 5 illustrates the effect of cigarette smoking on lungcancer mortality of these workers. Among 2,066 non-cigarette smokers, only eight lung cancers were seen in a ten-year period, where 1.82 were expected, based on American Cancer Society data on the risk of lung cancer death in non-smokers. Inhalation of asbestos by insulators appears to multiply the risk by four or five times. Considering the data for men with a history of smoking, among 9,591, 325 deaths were observed versus 66.78 expected, also a fivefold increase. However, since cigarette smokers already have a ten to twenty times greater risk of lung cancer deaths than non-smokers (depending on cigarette consumption), the multiplicative effect of the asbestos exposure increases the lung cancer risk up to 100 times for smoking asbestos workers compared to non-smokers unexposed to asbestos. This was also shown by the experiences of a cohort of New York and New Jersey insulators [17]. Hence, it was estimated that the risk of dying of lung cancer for cigarette smoking asbestos workers was more than 90 times that of individuals who neither smoked nor worked with asbestos. Indirect Asbestos Exposure In 1968 it was pointed out by Harries that shipyard workers other than insulators were at risk from asbestos disease [18]. Among Oevonport Dockyard employees, five cases of mesothelioma were found among men who had not been "asbestos workers" but had followed other trades in the yard. These men presumably had been inadvertently exposed to asbestos merely by working in the same shipyard areas where asbestos had been used. Continuing to follow this group, Harries later documented 55 cases of mesothelioma in this shipyard alone, only two of which occurred in asbestos workers [19], one, a man who had previously sprayed asbestos. A study of the distribution of all verified cases of mesothelioma found in Scotland between the years 1950 and 1967 is also revealing (20]. Of 89 cases available for study, 55 were in shipyard employees, dockers, or naval personnel. Of the 55, again only one was an asbestos insulation worker. A third important study of workers in British shipyards is that of John Edge, who reviewed x-rays of former shipyard workers in Barrow [21]. A prospective study was conducted of 235 men whose x-rays, taken between 1955 and 1969, showed abnormalities char- acteristic of asbestos exposure (pleural plaques, scarring of the covering of the lung or lining of the chest), but no parenchymal fibrosis (scarring of the lung tissue). Most of these x-rays were of individuals (riggers, welders, carpenters, electricians, machinists, steamfitters, etc.) who had not worked directly with asbestos, but who could have sometimes been nearby when asbestos was used. In tracing the individuals who had such x-ray changes, it was found that 70 had died froA 1970 to 1973_ Of these 70 deaths, 13 were of lung cancer, two and one-half times the number expected, and 17 were of mesothelioma (none, of_ course, were anticipated). N 9 h+ 76

Table 4. Expected and observed number of deaths among 623 New York-New Jersey asbestos insulation workers, i Janqary 1943 - 31 December 1973, twenty or more years after onset of first exposure to asbestos. Total 1943-1952 1953-1962 1963-1973 1943-1973 Exp. Obs. Ratio Exp. Obs. Ratio Exp. Ohs. Ratio Exp. Obs. Ratio Total deaths, all causes 88.22 82 0.94 111.05 170 1.53 101.38 191 1.88 300.65 444 1 48 . Cancer, all sites 13.02 30 2.30 18.75 65 3.47 19.49 103 5.28 51.26 198 3.86 Lung cancer 1.83 13 7.10 4.20 29 6.90 5.65 47 8.32 11.68 89 7.62 Pleural mesotheliona n.a.a 1 -- n.a. 2 -- n.a. 7 -- n.a. 10 -- Peritoneal mesothelioma n.a. 1 -- n.a. 3 -- n.a. 21 -- n.a. 25 -- Cancer of stomach 2.13 2 0.94 1.87 10 5.35 1.10 6 5.45 5.10 18 3.53 v Cancer of colon, rectum 2.22 7 3.15 2.74 9 3.28 2.54 6 2.36 7.50 22 2.93 Asbestosis n.a. 1 -- n.a. 11 -- n.a. 25 -- n.a. 37 All other causes 75.20 52 0.69 92.30 94 1.02 81.89 63 0.77 249.39 209 0.84 632 members were on the union's rolls on 1 January 1943. Nine died before reaching 20 years from first employment. All others entered these calculations upon reaching the 20-years-from-onset-of-first-exposure point. Expected deaths are based upon white male age-specific death rate data of the U.S. National Office of Vital Statistics from 1949 - 1971. Rates were extrapolated for 1943 - 1948 from rates for 1949 - 1955, and for 1972 - 1973 from rates for 1967 - 1971. a U. S. death rates not available, but these are rare causes of death in the general population. From: Reference [28]. LL,8b0i£9aZ

Table 5. Deaths of lung cancer among asbestos insulation workers in the United States and Canada, 1967-1976; influence of cigarette smoking. Expected deathsa Observed deaths U. S.b Smoking specificc 1. History of cigarette smoking 325 60.07 66.78 Current smokers 228 31.87 39.69 Ex smokers 97 23.29 13.34 2. No history of cigarette smoking 8 14.11 1.82 Never smoked 5 8.49 0.98 Pipe/Cigar 3 5.63 0.84 3. Unknown history of cigarette smoking 152 31.80 11.93 Total 485 105.97 66.78 a Age, year and sex specific. b Based upon age, specific data of the U. S. National Center for Health Statistics, cigarette smoking not considered. c Based upon American Cancer Society's Cancer Prevention Study, 1967-1972. From: Hammond, E. C., Selikoff, I. J., and Seidman, H., Cigarette smoking and mortality among U. S. asbestos insulation workers, to be published in Ann. N.Y. Acad. Sci. Environmental Asbestos Disease In 1960 Wagner reviewed 47 cases of mesothelioma found in the Northwest Cape Province, South Africa, in the previous five years [10]. Of this number, roughly half were in people who had worked with asbestos. Virtually all of the rest, however, were in individuals who had, decades before, simply lived or worked in an area of crocidolite asbestos mining (one lived along a roadway in which asbestos fibers were shipped). This germinal observation demonstrated that asbestos exposure of limited intensity, often intermittent, could cause mesothelioma. The hazard was further pointed by the findings of Newhouse[11], who showed that mesothelioma could occur among people whose potential asbestos exposure consisted of their having resided near an asbestos factory or in the households of asbestos workers. Twenty of 76 cases from the files of the London Hospital were the result of such exposure, 31 were occupational in origin, and asbestos exposure was not identified for 25. A recent extensive study of the effects of household exposure has been conducted by Dr. Henry Anderson and his colleagues of the Mount Sinai School of Medicine [22]. In a clinical survey of 489 family contacts of former factory workers, it was found that the x-rays of 36.2 percent of these individuals showed abnormalities characteristic of asbestos exposure. It did not matter greatly what the relationship to the worker was; the asbestos dust in the household could affect any resident - wife, sons, daughters, parents. While almost all were currently asymptomatic, and while most would perhaps suffer no impairment from their past exposure, others may be stricken with an asbestos-related cancer as a result of past household asbestos exposure. During the initial phase of the survey of deaths, mesothelioma had been identified in this group of family contacts. 78

Asbestos Fiber Types: Relation to Disease Canadian asbestos mine workers by the McGill group has already been mentioned earlier in these proceedings. In the initial publication of their mortality study [23], a favorable mortality experience was reported with lung cancer and gastrointestinal cancer being found in excess only in the higher exposure categories. While this study was comprised of 11,788 individuals, it should be noted that nearly half (4,818) were in the lowest dust category (virtually no exposure) or had been employed in the mines and mills for less than one year. Further, many others would have had relatively recent employment. Thus, the potential for dilution of asbestos-related health effects exists. A concomitant study of x-ray changes among mine and mill employees may suffer even more from the dis- advantage of short-term periods of observation [24]. Overall, 12.5 percent of 11,207 individuals were found to have abnormal x-rays. However, many of these had less than 10 years of employment and the x-ray that was read was the last maintained by the company of employment. We have also conducted studies of Canadian mine and mill employees, but of individuals who had been employed for at least 20 years [257. Table 6 lists the x-ray abnormalities found among 1,120 such individuals. As can be seen, extensive asbestos-related x-ray changes were present in this group of currently employed workers. Overall, 61 percent had abnormal x-rays. Table 7 presents the mortality experience of 535 men who were first employed in the mines and mills before 1941 and followed from 1961 [26]; 16 percent of the deaths were from asbestosis and 15 percent from lung cancer. One case of inesothelioma was found, considerably less than would have been expected on the experience of U. S. insulation workers or factory employees. The reason for this is unclear at this time. It may be related in part to the physical characteristics of the chrysotile fibers in the mine and mill environment, the fibers here being of a longer length than that encountered in manufacturing and end product use. Table 6. X-ray changes among 1,120 Quebec asbestos mine and mill employees by time from onset of exposure. Time from onset of Percent abnormal exposure (years) Normal x-ray Abnormal x-ray within category 20 - 24 83 46 35.7 25 - 29 99 104 51.2 30 - 34 122 182 57.6 35 - 39 76 170 69.1 40+ 58 180 75.3 Total 438 682 79

Table 7. Expected and observed deaths among 544a asbestos miners who were at least 20 years from onset of asbestos mining work at start of observation, 1961 through August 1977, by calendar years. - - - - - Total, 1961-77 - - - - - Expected Observed Ratio 0 E Total deaths 159.92 178 1.11 Total cancer, all sites 36.73 49 1.33 Lung cancer 11.10 28 2.52 Pleural mesothelioma b I -- Peritoneal mesothelioma b -- -- Cancer of stomach 3.65 4 -- Cancer of colon, rectum 5.03 6 1.19 Cancer of esophagus 0.87 -- -- A11 other cancers 16.08 10 0.62 Asbestosis b 26 -- Other non-infectious respiratory 6.69 4 0.60 All other causes 116.50 99 0.85 Man years 7,408 a Expected deaths are based upon age-specific death rate data for Canadian white males. b Death rates not available but these have been rare causes of death in the general population. Data are also available on exposure to amosite asbestos. From 1941 to 1954 a factory producing amosite insulation materials operated in Paterson, New Jersey. The mortality experience of individuals employed at any time between 1941 and 1945 is shown in Table 8. The usual asbestos diseases are seen to be present. Lung cancer is six times expected and 10.of 298 deaths are from pleural or peritoneal mesothelioma. An important aspect of this study is that individuals with relatively short exposures are shown to have an increased risk of death from asbestos-related causes. Table 9 shows the expected and observed deaths from lung cancer, mesothelioma, gastrointestinal cancer, and asbestosis according to time of employment in the plant. All time categories less than one year are elevated, and while a single one-month category does not have statistical significance, the longer periods up to six months do. ~ppp 80 ~

Ca Table 8. Deaths among 933a workers employed in an amosite asbestos factory, starting five years from onset of work 1941-1945 to December 31, 1974. - - - - - Deaths 1946-1974 - - Cause of death Expected Observed Ratio All causes 285.62 483 1.69 Cancer, all sites 50.10 157 3.13 Lung cancer 12.45 83 6.67 G.I. cancer 12.05 24 1.99 Pleural mesothelioma b 5 -- Peritoneal mesothelioma b 5 -- "Asbestos" cancer 24.50 117 4.78 Other cancer 25.60 40 1.56 Asbestosis b 28 -- All other causes 235.52 298 1.27 a Expected deaths are based upon white male age-specific death rate data of the U. S. National Office of Vital Statistics, 1949-1972. Rates were extrapolated for 1946-1948 from rates for 1949-1955 and for 1973-1974 from rates for 1968-1972. 128 workers were omitted from these calculations: 33 had prior asbestos exposure; 38 died in the first five years after onset of employment. 49 were not completely traced; and eight had other asbestos employment after the five year from onset point. b U. S. death rates not available but these are rare causes of death in the general population. 81

Table 9. Deaths of all "asbestos disease" among 933a workers employed in an amosite asbestos factory, starting five years from onset of work 1941-1945 to December 31, 1974. Effect of duration of exposure. - Death of "asbestos disease" 1946-1974 Duration of employment No. Expected Observed Ratio <1 month 62 3.47 6 1.73 1 month 92 3.73 8 2.14 2 months 79 3.73 11 2.95 3-5 months 145 5.98 17 2.84 6-11 months 129 4.15 21 5.06 1 year 105 3.74 20 5.35 2 years 77 2.91 24 8.25 3-4 years 51 2.36 15 6.36 5+ years 65 2.88 34 11.81 Total 805 32.95 156 4.73 a "Asbestos disease": asbestosis and chronic pulmonary insufficiency, lung cancer, pleural, and peritoneal mesothelioma, cancer of esophagus, stomach, colon-rectum. 128 workers were omitted from these calculations: 33 had prior asbestos exposure; 38 died in the first five years after onset of employment. 49 were not completely traced; and eight had other asbestos employment after the five year from onset point. Finally, if one considers the fiber type that insulation workers were exposed to, data from manufacturers have indicated that it was only to chrysotile and amosite. No crocido- lite was ever used as thermal insulation materials [27]. Further amosite was used in significant quantities only from 1940 through the early 1960's. As neither the period of use nor the incidence of mesothelioma among amosite workers listed above can account for the high frequency of this cause of death among insulation workers, it is clear that exposure to chrysotile asbestos is of importance here as well. Summary Accumulated human health data indicate that all major commercial varieties of asbestos, chrysotile, amosite, and crocidolite, produce significant disease. Lung cancer, asbestosis, mesothelioma, and gastrointestinal cancer are in significant excess among factory workers and insulators, while lung cancer and asbestosis are dominant causes of death among mine and mill employees. Further, evidence exists that environmental exposures, such as in the homes of workers or in the vicinity of mines and factories, have been sufficient to produce mesothelioma. Workers indirectly exposed to asbestos in their work, as shipyard workers, can be at significant risk. Currently no data exist that would indicate a threshold for asbestos related cancers. Prudence would suggest that exposures to all asbestos fibers be reduced to the minimum commensurate with feasible environmental controls. Considerable data exist that most work environments can maintain concentrations well below the current asbestos standard. I believe the issue is not that reduction of standards will result in the closing down of the _ surface of the earth, as was suggested earlier in this symposium, but that reduction in standards, with feasible control measures, will allow us to use the surface of the earth safely. 82

CS References [1] Murray, H. M., Report of the Departmental Committee on Compensation for Industrial Disease, London, H. M. Stationary Office, p. 127 (1907). [2] Autibault, M., Bull. del'Inspect. du Travail, p. 126 (1906). [3] Cooke, W. E. , Pulmonary asbestosis, Br. Med. J. , 2, 1024-1025 (1927). [4] Merewether, E. R. A. and Price, C. V., Report on effects of asbestos dust suppression in the asbestos industry, Part I., London, H. M. Stationary Office (1907). [5] Selikoff, I. J., Personal communication. [6] Lynch, K. M. and Smith, W. A. , Pulmonary asbestosis III: Carcinoma of lung in asbesto- silicosis, Am. J. Cancer, 24, 56-64 (1935). [7] Gloyne, S. R., A case of oat cell carcinoma of the lung occurring in asbestosis, Tubercle, 18, 100-101 (1936). [8] Merewether, E. R. A., Annual Report of the Chief Inspector of Factories, London, H. M. Stationary Office (1907). [9] Weiss, A., Pleurakrebs bei lungenasbestose, in vivo morphologisch gesichert, Medi- zinische, 3, 93-94 (1953). [10] Wagner, J. C., Sleggs, C. A., and Marchand, P. , Diffuse pleural mesothelioma and asbestos exposure in the northwestern Cape Province, Br. J. Ind. Med. , 17, 260-271 (1960). [11] Newhouse, M. L. and Thompson, H., Mesothelioma of pleura and peritoneum following exposure to asbestos in the London area, Br. J. Ind. Med., 22, 261-269 (1965). [12] Selikoff, I. J., Churg, J., and Hammond, E. C., Asbestos exposure and neoplasia, JAMA, 188, 22-26 (1964). [13] Selikoff, I. J., Hammond, E. C., and Seidman, H., Mortality experience of insulation workers in the United States and Canada, 1943-1977, to be published, Ann. N.Y. Acad. Sci. [14] Nicholson, W. J., Selikoff, I. J., Seidman, H., and Hammond, E. C., Mortality experi- ence of asbestos factory workers: effect of differing intensities of asbestos exposure, Environ. Res. (in press). For earlier data, see: Nicholson, W. J., Case study 1: Asiesths - the TLV approach, Ann. N.Y. Acad. Sci., 271, 152-169 (1976). [15] Selikoff, I. J., Hammond, E. C., and Seidman, H., Cancer risk of insulation workers in the United States, In Biolo ical Effects of Asbestos, Bogovski, Gilson, Timbrell, and Wagner (eds.), Lyon, IA ,RC pp. 209 2 97$f. [16] Hammond, E. C., Selikoff, I. J., and Seidman, H., Cigarette smoking and mortality among U. S. asbestos insulation workers, to be published in Ann. N.Y. Acad. Sci. For earlier data see: Hammond, E. C. and Selikoff, I. J., Relation of ciharette smoking to risk of death of asbestos-associated disease among insulation workers in the United States, in Biolo ical Effects of Asbestos, Bogovski, Gilson, Timbrell, and Wagner (eds.), Lyon, IAFC31'F3i7TM). [17] Selikoff, I. J., Hammond, E. C., and Churg, J., Asbestos exposure, smoking, and neoplasia, J. Am. Med. Assoc., 204, 106-112 (1968). [18] Harries, P. G., Asbestos hazards in naval dockyards, Ann. Occu .~., 11, 135 (1968). [19] Harries, P. G. et al., Radiological survey of men exposed to asbestos in naval dock- yards, Br. J. Ind. Med., 29, 274-279 (1972). 83 2063104883

[20] McEwen, J., Finlayson, A., Mair, A., and Gibson, A. A. M., Mesothelioma in Scotland, Br. Med. J., 4, 575-578 (1970). [21] Edge, J., Asbestos-related disease in Barrow-in-Furness, J. Environ. Res., 11, 244-247 (1976). [22] Anderson, H. A. et al., Household contact asbestos neoplastic risk, Ann. N.Y. Acad. Sci., 271, 311-323 (1976). [23] McDonald, J. C., McDonald, A. D., Gibbs, G. W., Siemiatycki, J., and Rossiter, C. E., Mortality in the chrysotile asbestos mines and mills of Quebec, Arch. Environ. Health, 22, 677-686 (1971). [24] Rossiter, C. E., Bristol, L. J., Cartier, P. H., Gilson, J. G., Grainger, T. R., Sluis-Cremer, G. K., and McDonald, J. C., Radiographic changes in chrysotile asbestos mine and mill workers of Quebec, Arch. Environ. Health, 24, 388-400 (1972). [25] Comite d'etude sur la salubrite dans 1'industrie de 1'amiante: rapport preliminaire, Beaudry, R. , Lagace, G. , and Juteau, L. (eds.) (1976). [26] Nicholson, W. J., Seidman, H., and Selikoff, I. J., (to be published) Ann. N.Y. Acad. Sci. (Proceedings of Science Week). [27] Selikoff, I. J., Hammond, E. C., and Churg, J., Mortality experiences of asbestos insulation workers, 1943-1968, in: Shapiro, H. A. (ed.), Pneumoconiosis, Proceedin s of the International Conference, Johannesburg 1969, Cape Town, x ord Un versity ress, PP._80-1 ~/d(1 T- PART II. EXTRAPOLATION TO OTHER INORGANIC FIBERS Current Status of the Asbestos Problem Part I of this contribution discusses essential elements and factors related to asbestos fiber exposure and associated human disease. The historical perspective presented, in conjunction with recent data, may help define the emerging problem area concerned with the biological potential of inorganic fibers as a class of compounds. These may be outlined as follows: The Time Required to Define the Asbestos Problem was Decades Long: Asbestosis, the disease characterized by scarred lungs due to the inhalation of asbestos fiber, was first described over 70 years ago [1]. It was not until the 1930's and 1940's that an accumulation of evidence suggested that asbestos fiber inhalation was also associated with increased neoplastic risk, specifically carcinoma of the lung [2-5]. This effect was not anticipated, and was overlooked for extraordinarily long time periods. Problems ~focussing on the ~acti~vity of mineral fibers, other than asbestos, ~ re u~ire _lenat~vt me ~e~riads to define. This ~ he or ung scarring, an obviou~fect associated with inha~a== tion, and esoecally so~'or neoplasms. The Different Asbestos Fiber Types Produce Similar Disease Patterns: The disease stigmata produced by asbestos fiber inhalation are similar for the different fiber species. Inhalation of all commercial asbestos, chrysotile [6], amosite [7], crocidolite [8], anthophyllite [9), and mixtures of these fibers [10] produce both scarring and various forms of malignant disease. A range of mineral species with different physical and chemical properties can produce disease patterns in humans which are similar and occasionally indistinguishable. It should be stressed that the major difference in biologi- cal effects noted are the relative risks associated with each fiber type for each disease entity. Because the ~man,y varietal forms of asbestos fibers Pro.d.uce- disease, and the non-asbestosbrous minerals are sim ar structural and chemicaliv, an eral entlLy which fi can be inhafed shou 3e studied for hea th effects. 84

C Extra-Pulmonary Organs in Humans are Involved in Asbestos Disease: The disease patterns associated with asbestos exposure are complex. Although inhala- tion is the primary route of exposure to the individual in the workplace, extra-pulmonary organs may be affected as well. For example, asbestos fiber exposure has been associated with the development of intra-abdominal and gastrointestinal tumors [11,12]; excess malignancies of the buccal cavity, pharynx, larynx, esophagus, and stomach have also been reported [13,22]. Therefore, multiple organs and cell types are targets of asbestos fiber action. Importantly, hundreds of thousands of man-years of observation were required to statistically verify that excesses of less common tumors occurred in these workers. Or9ans other than lungs should be considered targets for other mineral fibers as welt. Occasionally, Multiple Primary Tumors May Simultaneously Occur in the Same Host; Multiple primary tumors may occur in the same individual who had been occupationally exposed to asbestos fiber. Contributino causes of death, as well as the cause, are important in defining the extent of disease associatedwith miner f~er exposure. The Clinical Latency Period for Asbestos Disease is Extensive: There exists a long latency period between the onset of exposure to asbestos fiber and the first clinical appearance of neoplastic disease. These stigmata have different lapse time intervals for manifestation, e.g., mesothelioma is greater (30-40 years) than for lung cancer (20-30 years) [15,16]. This time lapse works against the establishment of an etiological link between the agent and the disease; it may confound exposure history by implicating several "agents." It therefore re ui~res ~many years of retros ective- ros ective stud to determine, gualitatively and guantitativety, the reTat onsh p betweem m nera exposure and disease. Fiber Exposure Continues Throughout the Life of the Exposed Individual: Although a long time period may elapse between the cessation of exposure to asbestos fiber and the appearance of disease, these materials tend to be retained in both lung parenchyma and extra-pulmonary tissues of exposed workmen [17-21]. Therefore, exposure in these individuals continues for their lifetime in that particles are often present, continuously interacting on the cellular level. Removal of an individual from immediate exposure to mineral fiber does not ~similarly remove him from o~ r a~n exposure This concept hd_Tds forwa 5 inorgamTitiers whdh are not readily solubfe in vivo. Fiber Dose-Response is a Function of both Duration and Intensity of Exposure: Both the duration and intensity of exposure to asbestos fiber appear to influence the relative risk of developing the different asbestos diseases, and markedly influence the length of the clinical latency period in which the disease becomes manifest [22]. For example, a study of workers employed in an asbestos factory utilizing amosite fiber demon- strated that exposure to high concentrations of amosite for as little as three months significantly increased the relative risk of developing lung cancer (3.87x=SMR) [13,22]. Exposures in this instance were extremely high. However, if one were to establish an average threshold limit value based on man-years of exposure (average fiber levels multiplied by number of years employed at such levels), such levels would be only 0.1 to 0.2 f/mL, generally considered to be a "safe" level for prevention of asbestosis by today's OSHA standard. This would essentially ignore short-term, high-level exposures which evidently carry significant disease potential. As counterpart, those workers employed for short time periods (less than one year) required longer clinical latency period before their diseases became manifest. This dose-response relationship is likely to hold for mineral fibers other than amosite. Peak exposures may be more importarnt than long-term ex osures and, on the other hand, low exposures m~ r uire~ger ep riods~observation to u y ~ine ne~asrisk. 85 N O ~ W H+ ~ w

Co-Factors Exist in Asbestos Oisease: Cocarcinogenic and other synergistic factors are important in the production of asbestos disease. The importance of cigarette smoking has been demonstrated by evidence that carcinoma of the lung synergistically increases in cigarette-smoking asbestos workers [23- 25]. However, present data indicate that cigarette smoking is important only for carcinoma of the lung, not for other malignancies. Lu~_n9 cancer, cigarette ~smokin and inhalation of other inorganic particles, e.., uranium min_ing__, has been shown to be interrelated in the past. 7herefore, such a synergism n~ exst for mineral- er than asbestos. The asbestos problem required decades of time to define through hundreds of thousands of man-years of observations. A range of materials produces similar disease patterns, acting singularly or in concert with other biologically active agents. The clinical latency period is long, target organs are many, and exposure related in part to fiber retention. No known safe level of exposure exists for the prevention of malignant disease. It may be logical to assume at present, that lessons learned from the study of the asbestos problem may be applied to other inorganic fibers as well; that these findings may be used as a model to guide and delineate in new and important areas. The Nature of Mineral Fibers Asbestos is the term which categorizes a specific group of natural silicate minerals which occur in fiber form. The term fiber indicates, by definition, that the mineral species grew with this morphology. It also indicates, by definition, that the plane surfaces which define the external symmetry of the mineral are crystal faces resulting from growth. Asbestos fiber consists of a polyfilamentous bundle of intergrown crystal units. The breaking open of such a fiber purportedly is brought about by separation along the juxtaposed crystal faces. The sane mechanical treatment, as during grinding, of a non- asbestos, single crystal fiber, produces acicular cleavage fragments. The surfaces so formed are cleavage planes rather than crystal faces. It is generally considered that the majority of cleavage surfaces normally follow crystal face morphological development, in that both tend to occur parallel to "low energy" planes within the mineral [28]. Some investi- gators, however, considered that there may be significant physical-chemical differences between crystal faces and cleavage planes (see T. Zoltai, this Conference). If so, such differences betweeo crystal faces and cleavage planes may result in different biological activities of these materials. This fundamental difference prevents direct extrapolation from asbestos fiber (bound by crystal faces) to fibrous rock-forming silicates (bound by cleavage planes). In addition to the differences in surface character, some difference in other proper- ties may exist as well. Asbestos minerals possess physical-chemical properties which are unique. Some of these properties, such as high fiber tensile strength and flexibility, are not observed in other mineral or synthetic fibers. These properties have been described in a number of recent documents [26,27]; their mineralogical character is detailed by others at this meeting (see, e.g., M. Ross, T. Zoltai, A. Goodwin). It is of very great importance to note that differences between asbestos fiber and other silicate fibers are based on mega- scopic properties, that is, those physical properties determined on bulk samples. The question arises as to whether or not these characteristics which distinguish asbestos from non-asbestos mineral fibers are derived from molecular properties (e.g., twinning). If they are, then these characteristics must also exist on the submicroscopic level as well. On the other hand, if these characteristics are determined by the physical nature of fiber bundles, that is, derived by properties related to the manner in which the units are inter- grown, then separation of these units upon comminution destroys the "unique characteristics." Single fibers on the submicroscopic level, of asbestos or other mineral fibers, are often indistinguishable on the basis of morphology, structural characterization (by selected area electron diffraction), and chemistry (as determined by an electron microprobe technique). Since mechanical properties cannot be measured on the microscopic or submicroscopic levels, it is unknown at the present time if the "asbestos properties" carry through to the sub- microscopic level. This focuses directly on the issue concerning the disease potential of fibrous silicates other than asbestos. If the "asbestos property" is only megascopic in nature, then size reduction of asbestos produces fibers essentially identical to acicular cleavage fragments of rock-forming silicates. The nature of the mineral fiber entity, on the submicroscopic level, prevents direct extrapolation concerning the biological activity 86

of other fibrous silicates. However, some extrapolation is currently possible on the basis of existing data. Data Which Suggest Inorganic Fibers Other than Asbestos are Biologically Active -Small fibers of various chemical compositions may form stable aerosols, persist in the work environment (with an accompanying increased inhalation potential), penetrate deep into the alveolar portions of the lung, and tend to be retained in tissues for long time periods. It has been suggested that such factors as fiber chemistry, trace metals, adsorbed hydro- carbons, etc. are not important in terms of carcinogenic potential. It has also been suggested that any fiber species in contact with the mesothelial lining of the chest, or lung, may produce mesothelioma, possibly by means of an "Oppenheimer" effect [297. Experi- mental work conducted with such materials as fibrous glass has demonstrated that even these man-made fibers may induce tumors when implanted at the mesotheliat surface [30,31]. Clinical human evidence suggests that all varieties of asbestos fibers can produce disease, and that any sub-species of a single variety can also produce disease. If certain forms of mineral species, commonly referred to as asbestos, are active biologically, what factors are responsible for this activity? Currently, only the size and shape of fiber are common to all mineral species which have been demonstrated to produce disease. It has been suggested that amphibole "fibers" observed in some industrial talcs are "acicular cleavage fragments" and therefore not asbestos per se.2 This argument carries with it the unsupported argument that since these particles are not asbestos, they are therefore not biologically active. However, a literature exists which implicates "fibers" in talc as a factor in human disease. These fibers are commonly asbestiform fibers (acicular cleavage fragments). Although these latter forms cannot be easily distinguished from each other, studies have indicated that these common contaminants of industrial grade talcs are the agents responsible for human disease. The disease stigmata are as follows: fibrosis, with patterns identical to asbestosis [34-38]; occurrence of uncoated fibers and asbestos bodies in lung tissues of workmen with interstitial lung scarring, and accompanied by other asbestosis stigmata (e.g., pleural plaques in workers with "talcosis") [37,39-41]; and excess malignancies, some of which are markers for asbestos exposure, e.g., mesothelioma [42]. One may cautiously accept that there are biologically active fibers contaminating industrial grade talcs. This might also carry with it, with some caution, that crystal faces and cleavage planes have the same biological potential in terms of producing human disease. Current Status It has taken 70 years to define the asbestos problem. The work of defining the human hazards associated with exposure to fibrous minerals, other than asbestos, will require at least as much effort and time. The varietal nature of asbestos, its broad range of mineralogical properties, suggests that other non-asbestos silicate fibers may be active as well. The argument centering on crystal face and cleavage plane difference extrapolated to biological potential requires study. The fact that a mineral fiber is non-asbestos does not extrapolate to its being non-active biologically. 2True asbestos, defined on the basis of mineral phase and its physical-chemical properties (flexibility and high tensile strength) does occur occasionally in talc deposits [32]. Asbestiform is defined as "formed-like or resembling asbestos...." This term refers to rock-forming fibrous silicates which are not flexible, do not have high tensile strength, yet when comminuted are identical to size-reduced asbestos. The term fiber in the present context is used to mean a morphological form, not necessarily the result of conditions of growth and therefore not necessarily bound by crystal faces. Since these characteristics cannot be easily measured on submicroscopic "fibers," the distinction if presently academic. 87 2063104887

References [1] Murray, H. M., In: Re ort of the Oepartmental Committee on Com ensation for Industrial Disease, Minutes of~ence ~ppendices and nd.d. 3495 - c.d. 349 , L~ ondon, Wyman and Sons (1907). [2] Lynch, K. M. and Smith, W. A., Pulmonary asbestosis. III. Carcinoma of lung in asbestos-silicosis. Amer. J. Cancer, 24, 56-64 (1935). [3] Gloyne, S. R., Two cases of squamous carcinoma of the lung occurring in asbestosis, Tubercle, 17, 5-10 (1935). [4] Wedler, H. W. , Asbestose und Lungenkrebs, Dtsch med. Wschr. , 69, 575-576 (1943a). [5] Wedler, H. W. , Uber den Lungenkrebs bei Asbestose, Dtsch. Arch. klin, Med. , 191, 189- 209 (1943b). [6] McDonald, J. C., Becklake, M. R., Gibbs, G. W., McDonald, A. D., and Rossiter, C. E., The health of chrysotile asbestos mine and mill workers of Quebec, Arch. environm. Hlth., 28, 61-68 (1974). [7] Selikoff, I. J., Hammond, E. C., and Churg, J., Carcinogenicity of amosite asbestos, Arch. environm. H1th., 25, 183-186 (1972b). [8] Wagner, J. C. and Berry, G., Mesotheliomas in rats following inoculation with asbestos, Brit. J. Cancer, 23, 567-581 (1969). [9] Meurman, L. 0., Koviluoto, R. , and Haka®a, M., Mortality and morbidity among the working population of anthophyllite asbestos miners in Finland, Brit. J. industr. Med., 31, 105-112 (1974). [10] SeliKoff, I. J., Hammond, E. C., and Seidman, H., Cancer risk of insulation workers in the United States, In: Bogovski, P., Gilson, J. C. Timbl, V., and Wagner, J. C., es.__o.~log~_ciTTffects of Asbestos, Lyon, International Agency for Research on Cancer (IARC Sc- ientiffc PuicatTone No. 8, pp. 209-216 (1973). [11] Keal, E. E. , Asbestosis and abdominal neoplasms, Lancet, pp. 1211 (1960). [12] Selikoff, I. J., Churg, J., and Hammond, E. C., Asbestos exposure and neoplasia, J. Amer. med. Assn., 188, 22-26 (1964). [13] Selikoff, I. J., Epidemiology of gastrointestinal cancer, Environm. Hlth. Perspect., 9, 299-305 (1974). [14] Dohner, V. A. , Beegle, R. G. , and Miller, W. T., Asbestos exposure and multiple primary tumors, Amer. Rev. resp. Ois., 112, 181-199 (1975). . [15] Selikoff, I. J., Hammond, E. C., and Churg, J., Mortality experiences of asbestos ~ insulation workers, 1943-1968, In: Shapiro, H. A., ed., Pneumoconiosis; Proceedings of the nternat~ Con erence, Johannesburg 1969, Cape Tawn, Oxford University Press, ( Pp. 1 -9 M [16] Selikoff, I. J., Lung cancer and mesothelioma during prospective surveillance of 1249 ~ asbestos insulation workers, 1963-1974, Ann. N.Y. Acad. Sci., 271, 448-456 (1975). ! [17] Pooley, F. 0., An examination of the fibrous mineral content of asbestos lung tissue from the Canadian chrysotile mining industry, Environm. Res., 12, 281-298 (1976). [18] Langer, A. M., Rubin, I. B., Selikoff, I. J., and Pooley, F. D., Chemical characteriza- • tion of uncoated asbestos fibers from the lungs of asbestos workers by electron micro- probe analysis, J. Histochem. Cytochem., 20, 735-740 (1972b). 88

[19] Fondimare, A. and Desbordes, J., Asbestos bodies and fibers in lung tissues, Environm. Hlth. Perspect. , 9, 147-148 (1974). [20] Sebastein, P., Fondimare, A., Bignon, J., Monchaux, G. , Desbordes, J., and Bonnaud, G., Topographic distribution of asbestos fibers in human lung in relation with occupational and non-occ_ - ex_ op ure, In: Wa~ton, T(.~., -,edIia eT d article and Vapours, ~, New York, ergamon (in press). [21] Langer, A. M., Inorganic particles in human tissues and their association with neoplastic disease, Environm, H1th. Perspect. , 9, 229-233 (1974). [22] Selikoff, I. J., Cancer risk of asbestos ex._ op sure, In: Hiatt, H. H., Watson, J. D., and Winsten, J. A., eds., ri 1784 (1977) gins of Human Cancer, Cold Spring Harbor, N.Y., pp. 1765- . [23] Selikoff, I. J., Hammond, E. C., and Churg, J., Asbestos exposure, smoking and neoplasia, J Amer. med Ass., 204, 106-112 (1968). [24] Doll, R., The age distribution of cancer: implications for models of carcinogenesis, J. ro- stat. Soc., 134, 133-166 (1971). [25] Berry, G. , Newhouse, M. L. , and Turok, M., Combined effect of asbestos exposure and smoking on mortality from lung cancer in factory workers, Lancet, ii, 476-479 (1972). [26] IARC Monograph Series: Evaluation of Carcinogenic Risk of Chemicals to Man, V. 14, Asbestos WH0-IARC Lyon, 1977, 106p. [27] Langer, A. M. and Wolff, M. S., Asbestos Carcino enesis, In: Schrauzer, G., ed., Inorganic and Nutritional Aspects of aer, P enum ress, N.Y., pp. 29-55 (1977). [28] Langer, A. M., Rohl, A. N., and Wolff, M. S., The nature of mineral surfaces and their role in biological interactions, Soc. Occup. and Environ. H1th., Wash., 1977 (in press). [29] Oppenheimer, B. S., Oppenheimer, E. T., Stout, A. R., Danishefsky, I., and Eirich, F. R., Malignant tumors and high polymers, Science, 118, 783-784 (1953). [30] Stanton, M. F. and Wrench, C., Mechanisms of mesothelioma induction with asbestos and fibrous glass, J. nat. Cancer Inst., 28, 797-821 (1972). [31] Stanton, M. F., Some etiolo ical considerations of fibre carcino enesis, In: Bogovski, P., Giison~ C., imbre ,., an~gner, S C. eds., ological ffects of Asbestos, Lyon, International Agency for Research on Cancer (IARC Scientifi~c PubT3ca- Lions No. 8, pp. 289-294 (1973). [32] Ford, W. E. , Dena's Textbook of Mineralogy, N.Y. Wiley, 4th Ed., 678p (1957). ' [33] Bureau of Mines, Dictionary of Mining, mineral and related terms, ed. P. W. Thrush, Washington, D. C. , U. S. Government Printing Office (1968). [34] Oreessen, W. C., Effects of certain silicate dusts in the lungs, J. Indust. Hyg., 15, 66-78 (1933). [35] Dreessen, W. C. and Dalla Valle, J. M., The effects of exposure to dust in two Georgia talc mi11s and mines, Publ Health REpts., 50, 1405-1415 (1935). [36] Siegal, W., Smith, A. R., and Greenburg, L., The dust hazard in tremolite talc mining, including roentgenological findings in talc workers, Am. J. Roentgenol., 4, 11-29 (1943). [37] Daymon, H., Latent silicosis and tuberculosis, Am. Rev. Tuberculosis, 53, 554-559 (1946). 89 2063104889

[38] Porro, F. W. and Levine, N. M., Pathology of talc pneumoconiosis with report of an autopsy, North. N.Y. State Med. J., 3, 23-25 (1946). [39] Hobbs, A. A., A type of pneumoconiosis, Am. J. Roentgenol. Radiol. Therap., 58, 488-497 (1950). - [40] McLaughlin, A., Rogers, E. , and Dunham, K. C., Talc pneumoconiosis, 8r. J. Indust. Med., 6, 184-194 (1949). [41] Porro, F. W. , Patton, J. R. , and Hobbs, A. A., Pneumoconiosis in the talc industry, Am. J. Roentgenol., 47, 507-524 (1942). [42] Kleinfeld, M., Messite, J., Kooyman, 0., and Zaki, M. H., Mortality among talc miners and millers in New York State, Arch, Environ. Health, 14, 663-667 (1967). Discussion M. SCHNEIDERMAN: You talked about short term exposures and problems of peak exposures. Then you divided an exposure by 20 years and that came out to some very small number, and you said that small number is substantially below the standards now set. Have you any information on the difference between biological results from peak exposures and long term exposures or should we consider only integrated exposures totaled over time and not consider problems of peak exposure? A. NICHOLSON: We don't have good data on the effects of peak exposures per se. They may in fact be proportionally greater than an amount averaged over a longer period of time. Insulation workers' exposures are very pealty-like. That is, they tend to spend most of their time working In conditions that would have very low ambient air concentra- tions. The material is wet or else they're not using asbestos, but at times when they were mixing cement, cutting block, or doing something like that they had very high concen- trations. This may be a factor. We just have no way of obtaining data on that particular item separately from the integrated exposures that we can make some estimate of. E. CDX: Dr. Nicholson, you mentioned an amosite exposure study of very short term nature and then went on to correlate that to the safe exposure over a long period of time. I believe your figures were three deaths, contrasted with an expected 1.34, from lung cancer for a person who was employed in that plant for one month or less. Was there any correlation with smoking done in that study? NICHOLSON: No, there was not, and the number of deaths in each single category were small. The consistency over each of those month by month categories, though, was strong. That is, if you looked at all months together over the period of time for less than one month through five months, the results are of statistical significance. In terms of cigarette smoking, we know it is strongly correlated with asbestos exposure. What asbestos does, in essence, is multiply whatever existing risk of death from lung cancer that is already present. If an individual has a high risk from cigarette smoking, then additional asbestos exposure can multiply that from five to ten times. If he has a very low risk of death from lung cancer because he's a non-smoker, it can be increased perhaps five times by the asbestos exposure. COX: Thus, there wasn't any correlation done. Now the other question would be with Dr. Langer's work, and perhaps you could answer it. It deals with the concentration of uranium involved in the mining danger. Langer had a chart, that went by rather rapidly, of the different things that are particularly dangerous and one was mining where uranium was involved. NICHOLSON: Uranium mining produces a very high risk of lung cancer. COX: Yes, I wonder if you could speak about the concentration of uranium? The amount of uranium in the are body is the thing of interest to me. 90

Ca NICHOLSON: Well, most ore bodies in the Southwest have two or three percent uranium oxide. COX: Well, let me be more specific. Phosphate mining in Florida where the yield is one pqund of uranium per ton of H3P04, would that be dangerous? NICHOLSON: It would depend on what the air concentration of the material is. I couldn't answer the question directly. COX: All right, thank you. L. SWINT: I'd like to clear up that question on uranium mining. Actually the cancer is caused by radon daughters which come from the radium, which is a decomposition product of uranium. The amount of uranium in the ore has nothing to do with the lung cancer. It's really a function of the exposure to radon daughters rather than the amount of uranium present. Although radon gas and radon daughters are decay products in the uranium decay series which, when in equilibrium, would be present in direct proportion to the amount of uranium present, for practical purposes they are independent because there are many events which occur that keep equilibrium of randon daughters and uranium from being established. Uranium and radium, the direct parent of radon, may be out of equilibrium due to differential leaching by groundwaters, since uranium is much more leachable than radium. The porosity and permeability of the rock affect the rate at which the rock will release radon gas into a mine atmosphere. Thus, the amounts of radon gas and radon daughters present in a mine atmosphere are not completely controlled by the amount of'radium or uranium in the rock. The grade of uranium ore mined in the U. S. through 1973 averaged between two and three tenths of one percent U308, but since 1973 this grade has steadily declined to fifteen hundredths of a percent in 1976. SCHNEIDERMAN: In fact, in some of those studies, the hard rock miners who would have similar exposures to the kinds of things that Dr. Nicholson was talking about were used as controls so that one might measure whether it was the radioactive material or the fibrous material that was of consequence. Those might have been inappropriate controls now that we know better, but hard rock miners were used as controls. NOTE: The following notes were sent following the meeting and were not part of the verbal discussion at the end of the session. P. GROSS: Dr. Langer's presentation suggested that fiberglass is carcinogenic to man. Epidemiologic studies as well as experimental studies in which animals inhaled fiberglass or were injected intratracheally with it have provided evidence that glass fibers were not carcinogenic. Only when glass fibers of a special thinness and length are placed in the chest cavity (not the lungs) or injected into the abdomen of rats do cancers develop. According to a recent publication (Money Causes Cancer: Ban It, by G. E. Moore and W. N. Palmer, JAMA 238, 397, August 17, 1977), sterilized dimes placed into the abdomen of rats caused more than'~5 percent of them to develop cancer within 14 months. The proclivity of certain rodents to develop cancers in response to various insoluble, solid materials embedded in their tissues is well recognized as "Solid-State Carcinogenesis" and should not be extrapolated to man. A. LANGER: During my presentation I voiced concern that among fibers other than asbestos, synthetic insulation fibers, e. g. , fibrous glass, when inhaled, may be biologi- cally active. This concern has been raised by a number of investigators, in different laboratories, based on observations made during more than 20 years of experimental work. As early as 1955, Schepers and Delahant [1], utilizing the inhalation route of administration, exposed guinea pigs and rats to 6-micron diameter fibrous glass. These animals were serially sacrificed for time periods up to two years, and progressive pulmonary changes followed. Guinea pigs were observed to develop pneumonia, lung abscesses, emphysema, and systemic neoplasms. Rats, in addition to these alterations, also formed pleural plaques, a stigma normally associated with asbestos fiber inhalation. "Severe parenchymal changes" were 91 2063104891

observed in both animal populations. In the same year, Schepers [2] published additional experimental data concerning the biological effects of intratracheally injected and inhaled glass wool. He observed persistence of glass in animal lung for up to 18 months after cessation of exposure. Glass wool fibers were observed in multinucleated giant cells and in areas of incipient atrophic emphysema. Epithelial hyperplasia was commonly observed. Inhalation experiments, cdnducted simultaneously with the same animals, produced epithelial hyperplasia and cellular desquamation; papillomas were observed in bronchioles. Focal cellular pneumonitis and other effects, such as alveolar wall thickening, were noted. Schepers considered some of these as "remarkable lesions" and suggested that ....... glass is not ~fibro9e~ni..c when retained in lu~g tissue. At the same time the gravity of the type a ronchiole lesion prov-oke~-necesitates caution in ~issTng qTas as tnnocuous. ndeed it shou~b ebe regarded as a ootential y harmf7 substance in circumstances leading to the inhalati- o of-lar e uan~ties of the t e of r~oducts studied in these experiments7 Yt shou d be strss-ed that t ~s ear~y study did not provide a contron group; however, one amy cautiously accept these findings considering the nature of the diseases and the extent to which the animal colony succumbed. In a number of animal studies which followed (e.g., Gross et al., 1959 [3]; Gross et al. , 1970 [4]), pulmonary changes from a variety of synthetic-7-i ers, in a number of an mi al models, appeared to occur. However, faced with some experimental caveats, these workers interpreted the results as not exclusivel indicative of biological activity of the fibers. It was not until 1972 that tanton an Wrench [5] demonstrated the ability of fibrous glass to induce malignant mesothelioma in experimental animals with appropriately vigorous control groups. This was substantiated further by Stanton in 1973 [6] and Pott et al. , 1974 [7]. Further work by Wright and Kuschner, in 1976 [8] unequivocally demon- stra ed the ability of fibrous glass to act in a manner similar to asbestos fibers in animal tissues (formation of scar tissue). Finally, Wagner et al., 1976 [9], were able to produce mesotheliomas in Wistar rats, after intrapleural inotuTation of glass fiber into chest cavities. The extent and even the histopathic nature of induced lesions may not be so marked as those from asbestos; nevertheless, many reports in the experimental pathology literature unequivocally demonstrate the potent activity of synthetic fibers in animal models [10]. Dr. Gross is correct in suggesting that the ability to induce tumors in the experimental model may well be related to the "Oppenheimer effect" (solid state carcinogenesis). Extrapolating to humans, this may indeed be the very same reactions which evokes mesothelial tumors. Hence, it has often been said that mesothelioma merely requires the hp ysical ,r~e_sence of a fiber at the pleural surface. If this is so, then the chemistry of the fiber, and its physical state, are secondary in terms of this particular biological response. Therefore, if inhalation of thin asbestos fibers (of any variety) produce mesotheliomas, the inhalation of thin glass fibers, which may also penetrate to the mesothelial lining of the lung, may produce the same response. The subject is still one which requires animal studies, and certainly human studies. It is an open issue. Discussion References [1] Schepers, G. W. H. and Delahant, A. B., An experimental study of the effects of glass wool on animal lungs, Arch. Ind. Health, 12, 276-279 (1955). [2] Schepers, G. W. H., The biological action of glass wool, Arch. Ind. Health, 12, 280- 287 (1955). [3] Gross, P., Westrick, M., and McNerney, J. M., Glass dust: a study of its biologic effect, Arch. Ind. Health, 16, 10-23 (1959). [4] Gross, P., Kaschak, M., Tolker, E., Bobyak, M., and deTreville, R. T. P., The pulmonary reaction to high concentrations of fibrous glass dust, Arch. Environ. Health, 20, 696-704 (1970). [5] Stanton, M. and Wrench, C., Mechanisms of mesothelioma induction with asbestos and fibrous glass, J. Nat. Cancer Inst., 48, 797-821 (1972). 92

Cs [6] Stanton, M., Some etiological considerations of fiber carcinogenesis, In: IARC Mono., Biolog. Effects of Asbestos, Bogovski, P., et al., Eds, Sci. Pub. No. 8, Lyon, pp 289-294 (1973). [7] Pott, F., Huth, F., and Friedrichs, K. H., Tumorigenic effect of fibrous dusts in -experimental animals, Env. Health Perspect., 9, 313-315 (1974). [8] Wright, G. and Kuschner, M., The influence of varying lengths of gtass and asbestos fibers on tissue response in guinea pigs, In: BOHS-Inhaled Particles and Vapours, IV, Edinburgh (1976), in press. [9] Wagner, J. C., Berry, G., and Skidmore, J. W., Studies of the carcinogenic effects of fiber glass of different diameters following intrapleural inoculation in experimental animals, In: Occupat. Exp. to Fibrous Glass, A Symposium - HEW - NIOSH, Univ. Maryland, June, 1974, U. S. Govt. Printing Office, Wash. p. 193-204 (with discussion). [10] Occu a~tional Ex to Fibrous Glass: A Sgposium, HEW - NIOSH, Univ. Maryland, June, 1974, U. Govt. Printing ffi'ce, Was6. 404 p. 93 ~ $ ~ w

CS National Bureau of Standards Special Publication 5D6. Proceedings of the Workshop on Asbestos: Definitions and Measurement Methods held at NBS, Gaithersburg, MD, July 18-20, 1977. (Issued November 1978) MEASUREMENT OF ASBESTOS RETENTION IN THE HUMAN RESPIRATORY SYSTEM RELATED TO HEALTH EFFECTS J. Bignon, P. Sebastien, and A. Gaudichet Universit€ Paris-Val de Marne - Institut de Recherche sur 1'Environnement Centre Hospitalier Intercommunal 94010 Creteil, France and Laboratoire des Particules Inhal4es Direction Departmentale de l'Action Sanitaire et Sociale 75013 Paris, France Abstract The retention pattern of asbestos fibers in the human respiratory system is related to four mechanisms: penetration into the respiratory tract deposition on the surface of respiratory epithelium, clearance, and intra-tissular translocation of asbestos fibers. Knowledge of such retention pattern for people exposed to asbestos dusts could provide useful information concerning the role of these mechanisms and the pathogenicity of fibers. So, asbestos fibers content has been assessed by light and electron microscopy in different samples from the respiratory tract: sputum, broncho-alveolar washing fluid, lung parenchyma, parietal pleural, and mediastinat lymph nodes from people diversely exposed to asbestos dusts and affected by various asbestos- related diseases. In each sample, asbestos fibers, identified as chrysotile or amphibole, have been counted and measured (length and diameter). It has been shown that asbestos fibers found in sputum and in broncho-alveolar washing fluid by light and electron microscopy were reliable for the assessment of inhaled asbestos fibers in the workplace or in the environment. Analytical data concerning asbestos burden in respiratory tissues can be summarized as follows: - despite the fact that most of the consumed asbestos is of chrysotile type, amphibole was more frequently found in lung parenchyma than chrysotile, in most cases; - most of the fibers retained in lung tissues were less than 0.20 pm in diameter and shorter than 5 pm. The intra-alveolar fibers were shorter (3.3 pm) than fibers found in lung parenchyma (4.9 pm). Fibers encountered in mediastinal lymph nodes were shorter (2.5 pm) and of amphibole type, whereas fibers encountered In parietal pleura were the shortest (2.3 Nm), and thinnest (0.06 pm in diameter) and mostly of chrysotile type. The signification of these data concerning the topographic variation in the fiber type and size are discussed in relationship with adverse health effects, particularly carcinogenesis. Key Words: Asbestos; carcinogenesis; fibers; pathogenicity; respiratory tract. Preceding page blank 95

Introduction The factors relevant to the assessment of public health risks of exposure to asbestos have been recently reviewed in two documents [1,2]1. It is now well documented that exposure to asbestos dust can lead to the development of lung fibrosis, bronchogenic carcinoma, pleural plaques, pleurisy, mesothelioma, gastro-intestinal tumors, and perhaps other unexpected diseases. The most critical point today is the establishment of dose- response relationship. Regarding cancer, adequate data to establish a threshold limit are not yet available. "The existence of a theoretical no-effect level may even be doubted; however, there may exist ; a rac4cT no-e ect level-6eTow wTicF any excess incidence cannot be adequately establ ished . As far as asbestos is concerned, because of the various possibilities of exposure, it is difficult to define retrospectively sharp conditions of exposure. So, the exposure- effect relationships are not very reliable and greater reliance should be put upon biological monitoring. Asbestos metrology in human samples could provide information about the most important questions arising for the assessment of dose-effect relationships and for the subsequent definition of prevention practices: A. Is there any relationship between one or several body-burden parameters at autopsy and the cause of death, sex, age, and possibilities of exposure? The problem is that the latency period of asbestos-induced diseases can be very long (up to 30 or 40 years). As the accumulation of fibers in man occurs in a dynamic way (related to inhalation and clearance mechanisms), only the residue-burden can be investigated at autopsy. Research is needed to establish eventual relationships between autopsy residue- burden and burden at the time of disease onset. B. What is the most suitable external indicator of body-burden during life? Such a contamination indicator, if it exists and if available for monitoring, could be very helpful for the detection or the survey of exposed people. If relationships could be established with related diseases or with any biological test, this kind of survey should be specifically relevant to biological monitoring. C. What is the biological significance of physical and chemical properties of fibers (length, diameter, elemental composition, associated pollutants...) regarding the induction of diseases (particularly tumors)? Recent experimental data using intrapleural implantation [3] or intraperitoneal injection [4] of fibers of different sizes indicated clearly that the size parameters are the most important for inducing cancer and that the most carcinogenic fibers, whatever the chemical composition, are those with diameters of less than 0.5 or 0.25 pm, and length more than 5 or 8 pm [5]. How can information provided by asbestos measurements in human respiratory tissues be correlated with these recent findings? 0. These studies on body-burden correlated to environmental monitoring could lead to more appropriate standards or quality guides for the future, in relation to the prevention of asbestos-related cancers. General Considerations Related to Asbestos Retention A. What Could Be the Definition of Body-Burden for Asbestos. The actual amount of pollutants in humans at any time is called retention. The retention of particles In humans occurs in a dynamic way and reaches an equilibrium level depending on the relative rate constants of deposition and clearance processes. The model of lung retention, based on the ICRP Task Group report [6], is suitable for describing the general scheme of deposition, clearance, penetration, and translocation of fibers in humans, as shown in figure 1. So far, the penetration and retention of asbestos fibers through the gastro-intestinal tract have not been intensively investigated [7]. 1Figures in brackets indicate the literature references at the end of this paper. 96

CS Dust in inspired and expred ai- Nasopharynx cofnpartment 1' Blood Tracheo - bironchial compartment Alveolar compartment ~' ~. Parenchymal , tissue Sputum Gastro- intestinal tract ~ ~ I Pulmonary lymph ~ ~ vessels and node '~s Pleural tissue 4 : Feces Figure 1. General scheme for deposition, clearance, translocation and retention of fibers, derived from the ICRP lung model [6]. (Heavy arrows : deposition; light dotted arrows : clearance pathway; light arrows : translocation pathways.) ' As asbestos measurement in tissues requires a destructive process, the retention of asbestos fibers cannot be controlled continuously. Measurement of asbestos in organs will provide information on asbestos retention at a very definite time: time of death for autopsic material, or time of surgical intervention for biopsic samples. So far, few attempts have been made for monitoring asbestos retention in alive people either by means of external magnetic procedure involving no sampling [8], or by means of relating body- burden to the amount of asbestos in sputum [9,10], in gastric juice [11], and in feces [12]. B. Deposition. Distinction has to be made between the two pathways for human exposure to asbestos: the pulmonary tract (PT) and the gastrointestinal tract (GIT). Timbrell has reviewed the mechanisms by which particles deposit in the respiratory system and has addressed specifically to the problem of fibers deposition [13]. He identified settling, inertial impaction and Brownian diffusion as deposition mechanisms which operate for both compact particles and fibers. In addition, he listed a fourth mechanism, direct interception, which is of little significance for compact particles but which may be of marked importance for fibers. In this view, a model for deposition of fibers in the human respiratory system has been described [14]. The effectiveness of these deposition mechanisms depends on the anatomy of the respiratory tract, the effective aerodynamic diameter of the particles (size, shape, density), and the breathing pattern. Asbestos fibers can also deposit in the gastrointestinal tract (GIT) either directly (because of the presence of asbestos in water, beverages and food) or indirectly (fibers coming from the respiratory airways and being swallowed). So far, there is little or no direct information regarding the way of fiber deposition at the surface of the human GIT. 97 2063104896

It, is obvious that accurate quantitative information on the deposition of asbestos fibers in humans is difficult to be obtained because of clearance and translocation mechanisms occurring simultaneously during lifetime. What we measure in the human body results from all these associated mechanisms! C. Clearance. Fibers which are deposited on the muco-ciliated blanket of the trachea and bronchi move toward the pharynx. The clearance of inhaled particles by this mechanism is believed to be more than 98 percent effective for most deposited particles [6]. However, the direct toxic effect of asbestos on the ciliated cells, as shwon recently [15], must impair the effectiveness of this clearance mechanism. The fibers deposited at the surface of the alveoli are either taken by alveolar macrophages or entrapped within the alveolar lining film. From there, some of them are cleared towards the ciliated airways while others should penetrate the alveolar membrane. The clearance is different according to the type of asbestos; for chrysotile, the clearance is important, since Wagner et al. [16], Morgan et al. [17] found that a large percentage of chrysotile asbestos entering the lungs of rats may be removed from the lungs within 58 days; but we do not know the mechanisms involved. However, most of the cleared fibers must reach the GIT as demonstrated by the study of Evans et al. [18] using inhaled neutron activated asbestos; up to 73 percent of this asbestos was found in the feces within 30 days. Measurements related to clearance in human have been carried out in several kinds of samples: sputum [10,19,20], gastric juice [11], and feces [12]. Generally, the finding of asbestos in such samples was related to past exposure, pulmonary burden or pathological features. The feasibility of using such samples as indicators of body-burden will be discussed later. D. Penetration and Translocation of Asbestos Fibers in the Human Body. Measurements in tissues using the transmission electron microscope (TEM) have revealed the presence of numerous fibers and fibrils far more than was ever imagined when the fiber population was evaluated by light microscopy alone. These findings, occurring even in case of moderate exposure and long elapsed time from last exposure, suggest a very high penetration and retention rate for TEM size fibers. In humans, asbestos fibers have been found by TEM in lung parenchyma by many authors [21,22,23, 24,25,26,27] and also in bronchial tissue, lymph nodes [28], parietal pleura [25,26], pleural fluid (29,30], peritoneum [31], liver [24], stomach [32,33], bowel walls [34], and colon [353. These findings suggest the penetration of asbestos in the human tissues and their migration throughout the whole body. Experimentally, penetration of fibers across the alveolar epithelium has been described in TEM by Suzuki (36]. The extreme tendency of asbestos fibers to migrate has also been demonstrated experimentally after subcutaneous injection [37], intrapleural or intraperitoneal inoculation [38,39,40], or after ingestion [41]. However, the penetration of ingested fibers through the wall of the gastrointestinal tract is still in discussion. This point is mostly relevant to asbestos-related extra- thoracic cancers, such as peritoneal mesothelioma, ovarian carcinoma, kidney carcinoma, etc. Some authors pointed out that there was no penetration [42]. However, an experiment In progress in our laboratories has shown that ingested chrysotile and crocidolite fibers did cross the intestinal barrier in the rat, being recovered in the lymph of the thoracic duct [43]. 98

Analytical Data Related to Asbestos Body-Burden in Humans A. Sam les Studied as Indicators of Asbestos Body-Burden. ' So far, most of the samples studied in this laboratory for estimating asbestos body- burden in humans were collected from the respiratory tract. We will only focus on data obtained from measurements in 3 kinds of samples: lung washing fluid (LWF) obtained by broncho-alveolar lavage (BAL), sputum collected on alive people, and respiratory tissues (lung parenchyma (LP), parietal pleura (PP), and mediastinal lymph nodes (LN) sampled at autopsy). According to the model shown in figure 1, it has been assumed that asbestos fibers found in LWF were related, on one hand to the intra-alveolarly deposited fraction of inhaled fibers, and on the other hand to the fraction cleared from the deep lung whereas those found in sputum must be related to the fibers cleared from the deep lung and from the tracheo-bronchial compartment [20]. The fibers detected by destroying lung parenchyma correspond to intra-alveolarly deposited fibers and intra-tissularly retained fibers at the time of autopsy. The point is to know if LWF and sputum can be used as external indicators of asbestos body-burden. In this view, a systematic comparative study of fibers encountered in LWF, in sputum and in lung tissue has been carried out and is still in progress. B. Analytical Procedures. For this study, the patients were classified according to their past asbestos exposure. A meticulous history was obtained by questioning each patient in detail about their successive occupations since leaving school. When a history of asbestos exposure was found, the duration of this exposure and the lapse-time since last exposure was recorded (expressed in years). Thus, the degree of exposure was estimated on one hand in terms of its duration and on the other hand according to the type of work done by the patients. All the biological samples were collected within 10 percent formalin. For autopsic lungs, the formalin was injected intratracheally. Pieces of tissue samples were cut and their volume measured. Typically, 1 cc of tissue was prepared for analysis. Each sample to be analyzed was put in a glass vessel containing sodium hypochlorite. This digestive procedure was performed at room temperature during one or two hours. Then, the mixture was directly filtered through a 0.4 pm pore size Nuclepore membrane filter previously coated with a carbon layer. At this stage the filter was scanned under the light microscope looking for ferruginous bodies. For TEM study, a second carbon layer was deposited upon the filter and the particles, entrapped in a double carbon-film, were transferred to TEM grids. The preparations were scanned at X 30,000 direct magnification, looking for fibers. Each fiber encountered was identified on the basis of its morphological features and its electron diffraction pattern and was called chrysotile, amphiboles, or non-asbestos fiber. The length and diameter of each asbestos fiber was measured using a calibrated mark on the viewing screen. For each grid square scanned, the data (number, mineralogical type, and size of fibers) were recorded directly on a computer. Several grid squares were scanned until the variation around the mean calculated for numerical concentrations was less than 30 percent. Concentrations of fibers were expressed in terms of number per sputum, number per total lung washing fluid recovered, and number per cc of tissue. Identification of associated non-fibrous particles has been assessed by means of electron microprobe analysis [44], but quantitative information concerning numerical or mass concentration of such particles has not been obtained. 99 2063104898

An intercomparison study between two laboratories (The University College of Cardiff - F. D. Pooley and Laboratoire des Particules Inhalees, Paris - P. Sebastien) has yielded very similar results concerning the assessment of asbestos fibers in tissues, using the procedure previously described [45]. C. Lung Washing Flui'd (LWF). The possibility of assessing the asbestos endo-alveolar content by means of broncho- alveolar lavage is now under investigation in diversely exposed people. Such a technique has been used by different workers in order to collect free cells and proteins from the human lung [46,47] and it has been shown in the baboon that pulmonary washing was an efficient procedure for the recovery of particles deposited in the alveolar compartment of the lung [48]. 1. Material and Method Up to date, this type of investigative procedure has been used in 26 cases (Table 1). The cases studied were divided in 4 groups: Table 1. Groups of 26 patients investigated by broncho-alveolar lavage. Nb Cases Asb. Exposure Diseases Definite Asb : 9 Group 1 9 Heavy Pl Pl : 5 Br Ca : 1 Definite P1 P1 : 2 Group 2 5 Moderate Silico-Asb : 1 Sm irr op : 1 Chr bronch : 1 Suspected Fibrosis + Group 3 3 Moderate P1 P1 : 1 P1 P1 : I Chr bronch : 1 None Lar Ca : 1 tuberculosis : 1 Controls 9 fibrosis : 1 histiocyt, x : 1 Chr bronch : 5 Abbreviations: Nb - number; Asb = asbestosis; P1 P1 - pleural plaques; Br Ca : bronchogenic carcinoma; Sm irr op • small irregular x-ray opacities; Chr bronch = chronic bronchitis; Lar Ca = larynx carcinoma. 100

CS Group 1 included 9 cases with definite heavy asbestos exposure (OH), subdivided into 7 insulation workers, 1 asbestos-cement worker, and 1 asbestos-textile worker. Lung asbestosis from 0/1 to 2/2 was diagnosed by x-ray according to the IL0 U/C International classification of radiographs of pneumoconiosis 1971. Asbestosis was associated with bronchial carcinoma in one case and with pleural plaques in 5 cases (Table 1). Group 2 included 5 cases with definite moderate asbestos exposure (DM), confirmed by minutious occupational inquiries. The occupation and associated diseases are indicated in Tables 1 and 2. Table 2. Occupations, associated diseases, and mineralogical results in cases of Group 2. (definite moderate exposure) Results LWF in Group 2(definite moderate exposure) Years Years since Nb coated Nb Cases Occupation occupation asb. exp. Diseases fibers fibers % A MOU... Boiler Fitter 10 19 P1 P1 10 + 0 GAN... Glass Blower 27 0 P1 P1 0 0 - MAR... Asbestos 19 11 Silicosis 0 + 50 Plate Cutting ± Asbest. ESS... Plumber with 18 3 Small Irr 0 0 - Welding, Brazing Opacities BOD... Isolation of 3 24 Chronic 0 0 - Central Heating Bronchitis Abbreviations: P1 P1 = pleural plaques; Years occupation = years of occupational exposure; Nb = number; LWF = lung washing fluid; % A= ratio of amphiboles number/ amphiboles number + chrysotile number. (See Table 1 a1so.) Group 3 included 3 cases with suspected (but not proven) moderate asbestos exposure (SM) according to the past occupational history of the patients. The occupation and associated diseases are indicated in Tables I and 3. Table 3 Results LWF in Group 3 (suspected moderate exposure) Years Nb coated Nb Cases Occupation occupation Diseases fibers fibers % A ABD... Autonabile 10 Chronic 0 + 0 Worker Bronchi ti s MON... Wood 10 Fibrosis 0 + 0 Worker + P1 P1 DEC... Plumber 25 P1 P1 0 0 - 'Abbreviations: See Table 2. 101 2063104900

CPS The 9 control cases included patients without specific dust exposure. The method used for broncho-alveolar lavage (BAL) has been extensively described elsewhere [49]. It was assumed that the volume of the lung washed by this procedure corresponded to about one segment. For mineralogical analysis, a 10 mL sample was taken from the whole lavage before the centrifugation was performed for cells recovery. 2. Results No asbestos fibers have been detected by LM and TEM in the LWF of the 9 control cases. Some other no fibrous mineral particles have been encountered in 50 percent of these cases, identified as chlorite, calcite, quartz, aragonite, phlogopite, magnetite, and Al metal. In the group I of heavily exposed patients (Table 4), the mean number of fibers was 12.1x106 per lavage. The mean number of alveolar macrophases (AM) was simultaneously estimated to be 12.6x108 per lavage. However, there was no correlation between the number of fibers and the number of AM. Asbestos fibers were mainly of the amphibole type in insulation or asbestos cement workers. The highest fiber count (50x10 ), only of the amphibole type, was observed in the patient working in an asbestos-cement plant. By contrast, in the case of having worked in an asbestos-textile plant, all the fibers were of the chrysotile type. The percentage of coated fibers was low, less than 1 percent in 7 out of 9 cases. The mean length and diameter were 3.3 and 0.13 pm respectively. Table 4. Mineralogical studies of lung washing fluid (LWF). Results LWF in Group 1(definite heavy exposure) Nb Nb % Mean Mean Exp. Yrs Yrs since A.M. fibers coated length diam Cases type exp. last exp. Diseases 106 106 fibers % A µm um CHA... I 16 2 A 7.6 21 5 100 3.9 0.15 KRE... 1 10 4 A 24.6 5 0.3 100 4.04 0.12 FRA... 1 11 3 A 26.1 6 0.5 100 3.02 0.14 CHE... I 10 11 A + B, CA - 2.4 0.15 100 2.9 0.10 BEN... I 15 4 A - 3.8 0.9 99 3.2 0.15 LAI... I 11 0 A 9.7 11.4 2 90 3.05 0.12 MAA... I 14 3 A+ p1 pl 7.3 7 0.8 100 2.07 0.15 MAR... AC 19 0 A 10.7 50 0.001 100 2.07 0.15 FAL... AT 4 1 A 2.4 3 0.02 0 5.6 - Average 12.2 3.1 12.6 12.1 1 3.3 0.13 ±4 ±3.1 ±8.4 ±14.4 t1.5 ±1 ±0.1 Abbreviations: Exp type = type of exposure; I• insulator; AC = asbestos-cement plant worker; AT a asbestos-textile plant worker; NB A.M. - number of alveolar macrophages per tavage; Nb fibers = number of asbestos fibers per lavage; % A• see Table 2; diam = diameter. 102

CS In the group 1, two parameters, duration of exposure in years and lapse-time since the last exposure, have been assessed and correlated with the fiber count in the LWF. The two curves show that the number deposited within the alveolus increases with duration of exposure, whereas this number decreases when the time since the last exposure increases (figure 2). t + Yrs of exposure • Yrs since the last exposure 5 10 15 20 Yrs I I I I Figure 2. Relationship between fiber count in lung washing fluid and exposure patterns for cases of group 1(definite heavily exposed people). The fiber count increases with the duration (years) of exposure; it decreases when the delay since the last exposure increases. In this group, the fiber yield obtained by BAL and by collecting one sputum has been compared (Table 5). The numbers of coated and uncoated fibers were one or two orders of magnitude higher in LWF than in sputum. Moreover, the fibers were shorter in LWF (mean length 3 pm) than in sputum (5 Nm). Elsewhere, the proportion of amphibole type fibers was less in sputum. - By contrast, in groups 2 and 3, with moderate exposure, the asbestos fiber count in LWF yielded less significant results (Tables 2 and 3). In some cases, both LM and TEM analysis were negative. In others, only a few fibers were found, but at a level not allowing a significant count to be expressed. 103 2063104902

COS Table S. Comparison of asbestos fibers in sputum and lung washing fluid (LWF) from cases of Group 1 (9 cases). Coated 4ibers Uncoated fibers % amphibole type fibers Mean length um Mean diameter um Sputum (one sample) 7.102 1.105 65 5 0.16 LWF (whole lavage) 3.104 5.106 88 3 0.13 In groups 2 and 3, the comparison of asbestos fibers found in sputum and LWF yielded the following results: in many cases, the numerical concentration was low or null; in other cases, one or the other sample showed some fibers. The asbestos content either in sputum or in LWF was similar, within the ranges: 0 to 10 for coated fibers and from not detectable to 5x10s for TEM size fibers, mostly of chrysotile type. D. S utum. It has been demonstrated in this laboratory [9,11] and by others [10] that the amount of coated fibers or ferruginous bodies (FB) in the sputum was significantly related to the asbestos exposure and to the amount of FB in lung parenchyma further measured at the autopsy time [11]. This test is very simple and can be used as a retrospective proof of asbestos exposure, even in the case of long lapse time after the end of exposure. Another advantage is that the coating around the fibers is the evidence that the fibers have stayed in the lung. The study of sputum can also be good in the case of light exposure if the TEM is used. As an example, in this laboratory the sputum has been studied from 45 people working inside buildings insulated with sprayed asbestos containing material. The TEM examination has shown the presence of TEM size asbestos fibers, only of the chrysotile type, in 13 cases (29 percent) (Table 6). The influence of duration of exposure on the presence or not of fibers in sputum has not been demonstrated. Chrysotile fibers were mostly short microfibrils (0.5 to 2 pm long) and forming clumps, probably entrapped in mucus (figure 3). Table 6. Sputum monitoring for asbestos in 45 people working in asbestos-sprayed buildings. TEM study Nb Percent Mean duration of exposure (yrs) Presence of Fibers 13 29 8.3 Absence of Fibers 32 71 8.1 ~ w ~ g 104 8 w

(CS Figure 3. Electron micrographs showing chrysotile type fibers isolated from sputum in people resident inside asbestos sprayed buildings. E. Respiratory Tissues. 1. Lung Parenchyma Lung parenchyma samples from 27 autopsic cases diversely exposed to asbestos and with different malignancies have been studied by TEM. Four blocks of parenchyma were sampled in different sites of the same lung: central upper lobe, peripheral upper lobe, central lower lobe, and peripheral lower lobe, as described elsewhere [25]. The geometric mean of fiber count in the 4 sites has been calculated and then the cases have been classified in groups according to the asbestos lung burden (Table 7). The proportion of cases having more than 106 fibers/cc of lung parenchyma was 8 out of 10 for the asbestosis + respiratory cancer group, 5 out of 11 for the mesothelioma group, 0 out of 2 for the lung cancer (without associated lung fibrosis) group, and 2 out of 4 for the other malignancies group. 105 2063104904

e Table. 7. Asbestos fibers burden in lung parenchyma according to pathological features. Pathological Fiber concentration in the lung, Nb cm 3 features ~ <106 106 - 107 >107 Total Asbestosis t Respiratory 2 5 3 10 Cancer Mesothelioma 6 3 2 11 Lung Cancer 2 0 0 2 Others Malignancies 2 2 0 4 Total 12 10 5 27 The mineralogical type of fibers encountered in lung parenchyma has been assessed by TEM and the results are expressed in Table 8 by the percentage of amphibole/all asbestos fibers. The parenchyma retention of amphibole type fibers has been found important in most cases, the amphibole proportion increasing with fiber concentration in all pathological groups. Moreover, whatever the fiber concentration in lung parenchyma, the highest mean proportion of amphibole type fibers was observed in the mesothelioma group. Table 8. Mineralogical type of fibers in lung parenchyma: ratio amphiboles/(amphiboles + chrysotile) x 100. Pathological Fiber concentration in the lung, Nb an 3 features <106 106 - 107 >107 Average Asbestosis t Respiratory 38 59 69 58 Cancer Mesothelioma 53 70 89 64 Lung Cancer 4 4 Others Maligancies 12 41 26 Several size parameters have been assessed: mean length, mean diameter, and proportion of fibers longer than 8 pa. The results are shown in Tables 9, 10, and 11 respectively. The main figures are: 1) the size of fibers increases when the concentration increases; 2) the mean diameter never exceeds 0.16 Nm; 3) the mean percentage of fibers longer than 8 pm does not exceed 20.8 percent. 106

.T, Table 9. Size of fibers in lung parenchyma: mean length (um). Pathological Fiber concentration in the lung, Nb cm 3 features <106 106 - 107 >107 Average Asbestosis ± Respiratory 3.7 5.4 5.5 5.1 Cancer Mesothelioma 4.8 5.7 4.1 4.9 Lung Cancer 1 1 Others Maiignancies 2.8 2.3 2.6 Table 10. Size of fibers in lung parenchyma: mean diameter (pm). Pathological Fiber concentration in the lung, Nb cm 3 features <106 106 - 107 >107 Average Asbestosis t Respiratory 0.11 0.13 0.16 0.13 Cancer Mesothelioma 0.09 0.13 0.12 0.11 Lung Cancer 0.05 0.05 Others Malignancies 0.09 0.13 0.11 Table 11. Size of fibers in lung parenchyma: than 8 pm (%). proportion of fibers longer Pathological Fiber concentration in the 1ung, Nb cm 3 features Asbestosis t <106 106 - 107 >107 Average Respiratory Cancer 11.6 20.1 20.8 18.6 Mesothelioma 13.1 20.5 11.4 15 Lung Cancer Others 0.7 0.7 Malignancies 1.6 6.3 4.1 107

2. Asbestos Fiber Parameters A~_ccordin to ~Samp~lin_q Sites in Respiratory Tissues: ar~ enhymaar et~TP leura, ~iastinaT Lymp~i Nes. Besides lung parenchyma samples „ parietal pleura samples were available in 13 cases and mediastinal lymph node.samples in 4 of these cases. The comparison of fiber concentration in lung parenchyma and parietal pleura is indicated on figure 4. The absenceoF-correlation between asbestos fiber content in parenchymal and pleural tissue is emphasized. It is noteworthy that in some mesothelioma cases, even with high concentration inside lung parenchyma, the fiber concentration in the parietal pleura was very low. By contrast, a correlation seemed to appear between the fiber concentration in parietal pleura and in lymph nodes (figure 5). 107 108 10 106 10 f LUNG PARENCHYMA M A AM M M A A 105 104 M M M M M FIBERS IN LUNG AND PLEURA (Nb.cw3) M. Masetheliom. A. Asbsstesis ± Luny Cancer Figure 4. Correlation between asbestos fiber concentration in lung and in parietal pleura (see text for comments). 108

CS A A y W PARIETAL PLEURA 105 106 10 7 A I I I FIBERS IN PLEURA AND LYMPH NODES (Nb.cm3) Figure 5. Correlation between asbestos fiber concentration in parietal pleura and mediastinal lymph nodes. The comparison of mineralogical types has been carried out in the same way. The most striking features were: a) Most of TEM fibers encountered in parietal pleura were of chrysotile type even when the proportion of amphibole/amphibole + chrysotile type fibers was higher than 0.5 in the lung parenchyma (figure 6). b) By contrast, so far in the few cases studied, most of the fibers encountered in lymph nodes were of amphibole type (figure 6). The fiber size has been compared in the different sampling sites (Table 12). The longest ftT ers were found in the lung and the thinnest in the parietal pleura. Mean fiber length was of 4.9 pm for lung parenchyma, 2.3 pm for parietal pleura, and 2.5 pm for lymph nodes. N ~ 109 W f+ ~ ~ 00

COS Ratio (AmPhiielas/Amp hib olms + ChrYsotilo) x 100 100 E-50 A 50 1g0 1~ _ AA A • LUNG PARENCHYMA 100 Dp 0 F-5• 50 100 PARIETAL PLEURA LYMPH NODES Figure 6. Ratio of amphiboles count/total asbestos fibers count in lung parenchyma coopared to the ratio in parietal pleura (top) and to the ratio in lyvph nodes (bottom). 110

CS Table 12. Fiber size in lung parenchyma, parietal pleura, and lymph nodes. Lung parenchyma Parietal pleura Lymph nodes Mean Length um 4.9 2.3 2.5 Mean Diameter um Proportion of Fibers 0.13 0.06 0.16 Longer than 8 um percent 15 2 3 Discussion { The contribution of this metrologic study of asbestos dusts in the human PT is relevant to three major points relating to the pathophysiology of fibrous particles: - It allowed a check of the reliability of monitoring asbestos in sputum and lung washing fluid for the assessment of asbestos exposure. - It provided a better understanding of the partition of fibers in the different compartments of the respiratory system, which allows hypothesis about the translocation of fibers in the PT. - It yielded quantitative data concerning the actual fiber dimensions in humans in different diseases, including pleural mesotheliomata, which have to be discussed in view of recent experiments concerning the mesothelial response in relation to fiber dimension. A. External Indicators of Asbestos Lung Burden. The present work demonstrated that the study of sputum and LWF by LM and TEM was very reliable for the assessment of asbestos exposure in heavily exposed people. The advantage of LWF over sputum is that it yields a greater amount of fibers which are most representative of the alveolarly deposited fraction. This technique, which requires that the patient accept a fiberoptic bronchoscopy, might help to diagnose asbestos-related diseases. However, this possibility has some limitation. Indeed, the information provided by BAL carried out in moderately exposed people was much less reliable than the study of lung parenchyma. This can be easily understood since we will discuss later on that the percentage of intraalveolar fibers is very low compared to the fibers retained in lung parenchyma. However, it seems that LM and TEM study of sputum is an excellent tool for detecting and following exposed people [9,11]. A cytological control of the sputum looking for, AM is needed to be sure that it represents the mineral content of the deep lung. It is possible that the measurement of asbestos fibers in other biological samples could be better indicators of asbestos body-burden, as discussed elsewhere [50]. Thus the search for asbestos fibers in feces appeared to be a very sensitive method, allowing detection of low intake of asbestos fibers [12]. B. Translocation of Asbestos Fibers in the Respiratory System. The figure 7 summarizes all the mean data concerning number, length, and diameter of fibers in four sites of the respiratory system: alveoli, LP, PP and LN. Moreover, the figure 8 gives the distribution of length fibers in these four sites. 111 2063104910

Figure 7. Diagram comparing the mean number (Nb), mean length (L) and mean diameter (d) of asbestos fibers in 4 sites of the respiratory system. Numbers have been estimated for the whole lung for parenchyma (Par) and alveoli (Alv), while they are given per cc of tissue for pleura and lymph nodes (LN). 112

percent 30 20 t0 30 20 10 30 20 10 30 20 10 LUNG PARENCHYMA 1 2 4 6 8 16 32 µm PARIETAL PLEURA LYMPH NODES ALVEOLUS Figure 8. Distribution of fibers length in parenchyma, parietal pleura, lymph nodes and alveoli. Note that long fibers, more than 4{rm in length, are less frequent in pleura, lymph nodes and alveoli than in parenchyma. 113 1

,-- For LP and alveoli, the fiber counts have been integrated for the whole lung, distinguishing intra-alveolar fibers assessed by the BAL and intra-parenchymal fibers assessed by destroying LP. Thus, the fraction corresponding to LP totalizes fibers entrapped in the pulmonary interstitial tissue (plus fibers inside blood vessels?) and fibers within the alveolar compartment. For that estimation, the volume of total lung has been assumed to be 5000 mL and the fraction of alveolar spaces washed by the BAL to be 1/20 of the whole lung volume. Thus, the figure 7 shows that the intra-alveolar fraction of all intra-parenchymal fibers would only represent about 1 percent of all the fibers retained in lung tissue, when assumed that BAL recovered all intra-alveolarly deposited fibers. The mineralogical type of alveolar and interstitial asbestos dusts did not differ significantly, as indicated on one hand by the electron diffraction pattern and on the other hand by the measurement of fiber diameters, identical in both sites (0.13 pm in mean diameter). Elsewhere, it is noteworthy that the intra-alveolar fibers were significantly shorter (3.3 pm in mean length) than the interstitial fibers (4.9 pm in mean length for LP fibers). This difference must even be more important, because the mean length of interstitial fibers is probably reduced by adding the 1 percent of short alveolar fibers to the interstitial fibers when LP is studied; on the other hand, it is possible that the mean length of intra-alveolar fibers is increased by the addition of longer fibers deposited at the surface of the peripheral airways and washed out during the BAL. Indeed, the mean length of fibers in sputum was found to be 5 pm (Table 5). These results clearly indicate a shorter length of fibers inside alveoli compared to pulmonary interstitial tissue. This can be related to two mechanisms, more or less associated (figure 9); either long fibers might penetrate more easily across the alveolar membrane or small fibers are more easily cleared from the interstitial tissue toward the alveolar spaces? As will be discussed, sizing of fibers in pleura and in lymph nodes brings a clue in the favor of the last hypothesis. Indeed, in these two sides (PP and LN), the asbestos fibers were significantly shorter than in lung parenchyma (2.3 pm in PP; 2.5 pm in LN compared to 4.9 pm in LP). These findings are additional clues to the greatest translocation effectiveness of short fibers. The migration of fibers was found even more selective in this study, since mostly chrysotile fibers were found inside the PP, with a mean diameter of 0.06 pm, whereas mostly amphibole type fibers with a mean diameter of 0.16 pm were found in mediatinal LN. This selective migration of fibers might be mostly related to their dimension, as if only short and very thin fibers could be entrapped in the PP tissue (figure 9). C. Fibers Dimension Related to Carcinogenicity. The aforementioned recent animal experiments after implantation of fibers in the pleura [3,5] reinforced the idea that the carcinogenicity of fibers depends only on dimension of fibers, whatever the chemical composition is, in such a way that the probability to induce pleural cancer reaches 100 percent when all the fibers are less than 0.25 Nm in diameter and more than 8 Nm in length (see Stanton et al. , this meeting). In humans, as demonstrated by this work and by others [24,26,57], all the asbestos fibers encountered in different sites of the respiratory system were found to have a diameter less than 0.25 pt. By contrast, the present study has clearly demonstrated that the mean length of fibers was always less than 8 pm in all sites (figure 7). However, a certain percentage of fibers was longer than 8 pm, especially in lung parenchyma (15 percent) (see Table 12 and figure 8). The point is to understand how such few fibers, distant from the parietal pleura, might induce the carcinogenetic transformation of mesothelial cells, or if other mechanisms specific to humans are to be considered. 114

PLEURA LP t // ~ ~ A>C` lzt Alv. Figure 9. Diagram showing the hypothetic different selective translocation pathways of fibers in the respiratory system. The longest fibers are retained within the lung parenchyma (LP) with more amphibole-type fibers than chrysotile-type fibers (A > C). The shortest fibers migrate either towards the parietal pleura (Par P1) and mostly of chrysotile-type (C), or towards the lymph nodes (LN) and mostly of amphibole-type (A). The fibers are shorter within the alveoli (Alv) than in lung parenchyma (LP); this must be due to the selective translocation of short fibers from the pulmonary interstitial tissue (1). The microprobe analyses have been carried out in the Laboratoire de Biophysique Medicale (Pr P. Galle) in collaboration with J. P. Berry. Part of this work has been supported by the Minist4re de la Qualite de la Vie and by the Institut National de la Sant4 et de 1a Recherche M€dicale. N 4 115 ~ ~ ~ r ~

0% References [1] IARC Monographs on the evaluation of carcinogenic risk of chemical to man, Asbestos, 14, WHO, IARC Publications (1977). [2] Zielhuis, R. L., Public Health risks of exposure to asbestos, European Economic Community, Directorate of Socials Affairs, Health and Safety Directorate, Pergamon Press (1977). [3] Stanton, M. F., Layard, M., Tegeris, A., Miller, E., May, M., and Kent, E., Carcinogenicity of fibrous glass: pleural response in the rat in relation to fiber dimension, J. Natl. Cancer Inst. 58, 587-603 (1977). [4] Pott, F., Friedrichs, K. H., and Huth, F., Ergebnisse aus Tierversuchen zur kanzerogenen Wirkung faserfgrmiger StaUbe und ihre Deutung im Hinblick auf die Tumorentstehung being Menschen, Zbl. Bakt. ~Zq. I Abt. Ori . B. 162, 467-505 (1976). [5] Meeting on the biological relevance of fiber parameters, IARC June 30-July 1, Lyon (1977). [6] Task group on lung dynamics, Deposition and retention models for internal dosimetry of the human respiratory tract, Health Phys. 12, 173-207 (1966). [7] Pontefract, R. 0. and Cunningham, H. J., Penetration of asbestos through the digestive tracts of rats, Nature, 243, 352-353 (1973). [8] Cohen, 0., Ferromagnetic contamination in the lungs and other organs of the human body, Science, 180, 745-748 (1973). [9] Bignon, J., Depierre, A., Bonnaud, G., Goni, J., and Brouet, G., Mise en evidence des corps ferrugineux par microfiltration de 1'expectoration, Nouv. Presse Med. 2, 1697- 1700 (1973). [10] Sluis-Cremer, G. K., Asbestosis in South Africa. Certain geographical and environmental considerations, Ann. N.Y. Acad. Sci. 132, 215-234 (1965). [i1] Bignon, J., Sebastien, P., Jaurand, M. C., and Hem, B., Microfiltration method for quantitative study of fibrous particles in biological specimens, Environ. Health Persp. 9, 155-160 (1974). [12] Cunningham, H. M., Pontefract, R. D., and O'Brien, R. C., Quantitative relationship of fecal asbestos to asbestos exposure, J. Toxicol. Environ. Health, 1, 377-379 (1976). [13] Timbrell, V., The inhalation of fibrous dusts, Ann. N.Y. Acad. Sci. 132, 255-273 (1965). [14] Harris, R. L. and Fraser, 0. A., A model for deposition of fibers in the human respiratory system, Aat. Industr. Hyg. Assoc. J. 3, 73-89 (1976). [15] Mossman, B. T., Kessler, J. B., Ley, B. W., and Craighead, J. E., Interaction of crocidolite asbestos with hamster respiratory mucosa in organ culture, Lab. Invest. 36, 131-139 (1977). [16] Wagner, J. C., Berry, G., Skidmore, J. W., and Timbrell, V., The effect of the inhalation of asbestos in rats, Br. J. Cancer 29, 252-269 (1974). [17] Morgan, A., Evans, J. C., Evans, R. J. , Hounam, R. F., Holmes, A., and Doyle, S. G., Studies on the deposition of inhaled fibrous material in the respiratory tract of the rat and its subsequent clearance using radioactive tracer techniques II. Deposition of the UICC standard reference samples of asbestos, Environ. Res. 10, 196-207 (1975). 116 S

[18] Evans, J. C., Evans, R. J., Holmes, A., Hounam, R. F., Jones, D. M., Morgan, A. and Walsh, M., Studies on the deposition of inhaled fibrous material in the respiratory tract of the rat and its subsequent clearance using radioactive tracer techniques, Environ. Res. 6, 180-201 (1973). [19] Stumphuis, J., Epidemiology of mesothelioma on Walcheren Island, Br. J. Industr. Med. 28, 59-66 (1971). [20] Bignon, J., Sebastien, P., Monchaux, G. and Bonnaud, G., L'Epuration ii long terme des particules fibreuses chez 1'homme, Collogues de 1'INSERM, 29, 205-218 (1974). [21] Langer, A. M., Selikoff, I. J., and Sastre, A., Chrysotile asbestos in the lungs of persons in New York City, Arch. Environ. Health 22, 348-360'(1971). [22] Pooley, F. D., Electron microscope characteristics of inhaled chrysotile asbestos fiber, Br. J. Industr. Med. 29, 146-153 (1972). [23] Pooley, F. D., Methods for assessing asbestos fibers and asbestos bodies in tissue by electron microscopy. In Biological effects of asbestos, Proceedings of a Working Conference held at the IARC, Lyon, France, 2-6 October 1972. IARC Scientific Publications n° 8, pp. 50-53 (1973). [24] Fondimare, A. and Desbordes, J., Asbestos bodies and fibers in lung tissues„__ _ Environ. Health Persp. 9, 147-148 (1974). [25] Sebastien, P., Fondimare, A., Bignon, J., Monchaux, G. , Desbordes, J., and Bonnaud, G. , Topographic distribution of asbestos fibers in human lung in relatian with occupational and non-occupational exposure. Fourth International Symposium on Inhaled Particles and Vapours of the BOHS, Edinburgh (22-26 September 1975). [26] Le Bouffant, L., Bruyere, S., Martin, J. C., Tichoux, G., and Normand, C., Quelques observations sur les fibres d'amiante et les formations minerales diverses rencontr4es dans les poumons asbestosiques, Rev. Fr. Mal. R2sp. 4, su 1. 2, 121-140 (1976). [27] Gross, P., Harley, R. A., Davis, J. M. G., and Cralley, L. J., Mineral fiber content of human lungs, Am. Industr. ~Xq. Assoc. J. 35, 148- (1974). [28] Gross, P., Davis, J. M. G. , Harley, R. A., and de Treville, R. T. P., Lymphatic transport of fibrous dust from the lungs, J. Occup. Med. 15, 186-189 (1973). [29] Pooley, F. D., Personal communication (1976). [30] Bignon, J. and Sebastien, P., Unpublished data. [31] Hourihane, 0. 0. B., A biopsy series of mesothelioma and attempts to identify asbestos within some of the tumors, Ann. N.Y. Acad. Sci. 132, 647-673 (1965). [32] Henderson, W. J., Evans, D. M. D., Davies, J. D., and Griffiths, K., Analysis of particles in stomach tumors from Japanese males, Environ. Res. 9, 240-249 (1975). [33] Chatel, A., Mignon, F., Sebastien, P., Hirsch, A., Bignon, J., Bader, J. P., and Chretien, J., Exploration. oeso-gastrique avec recherche de fibres d'amiante d'une s6rie de malades exposes is 1'amiante, Arch. Mal. 8pp. Dii. (in press). [34] Pooley, F. D., Locating fibers in the bowel wall, Environ. Health Persp. 9, 235 (1974). [35] Rosen, P., Savino, A., and Melamed, M., Ferruginous (asbestos) bodies and primary carcinoma of the colon, A.J.C.P. 61, 135-138 (1974). rsp. 9, (36] 241u252 (1974)nteraction of asbestos with alveolar cells, Environ. Health Persp. 117 2063104916

[37] Roe, F. J. C., Carter, R. L., Walters, M. A., and Harington, J. 5., The pathological effects of subcutaneous injections of asbestos fibers in mice: migration of fibers to submesothelial tissues and induction of inesotheliomata, Int. J. Cancer 2, 628-638 (1967). [38] Karacharova, V. N., Olshvang, R. A., and Kooan, F. M., On changes of certain organs after intraperitoneal administration of asbestos containing dust in experiment, Byull. EksQ. Biol. Med. 67, 117-120 (1967). [39] Morgan, A., Holmes, A., and Gold, C., Studies of the solubility of constituents of chrysotile asbestos in vivo using radioactive tracer techniques, Environ. Res. 4, 558-570 (1971). [40] Friedrichs, K. H., Hilscher, W. , and Sethi, 5., Staub und gewebe untersuchungen an rattennach intraperitonealer injektion von asbest, Int. Arch. Arbeits. Med. 28, 341- 354 (1971). [41] Cunningham, H. M. and Pontrefact, R. 0., Asbestos fibers in beverages, drinking water, and tissues: their passage through the intestinal wall and movement through the body, J. AOAC J6, 976-981 (1973). [42] Gross, P., Harley, R. A., Swinburne, L. M., Davis, J. M. G., and Green, W. B., Ingested mineral fibers, Arch. Environ. Health 29, 341-347 (1974). [43] Bignon, J. , Masse, R., and Sebastien, P. , Unpublished data. [44] Berry, J. P., Henoc, P., Galle, P., and Pariente, R. , L'empoussierage pulmonaire: etude par microscopie electronique, microanalyse par sonde ectronique, microdiffraction d'6lectrans, J. Microscopie 17, 11-18 (1975). [45] Pooley, F. 0. and Sebastien, P., Asbestos fibers from mesothelioma cases, Results of an intercomparison program (1976). [46] Reynolds, H. Y. and Newball, H. H., Analysis of proteins and respiratory cells obtained from humans lung by bronchial lavage, J. Lab. Clin. Med. 84, 559-573 (1974). [47] Daniele, R. P., Altose, M. 0., and Rowlands, 0. T., Immunocompetent cells from the lower respiratory tract of normal human lungs, J. Clin. Invest. 56, 986-995 (1975). [48] Nolibe, D., Metivier, H., Masse, R., and Lafuma, J., Therapeutic effects of pulmonary lavage, in vivo, after inhalation of insoluble radioactive particles, Fourth International Symposium on Inhaled and Vapours of the BOHS, Edinburgh, 22-26 September 1975 (in press). [49] Basset, F. , Soler, P., Jaurand, J. C., and Bignon, J. , Ultrastructural examination of broncho-alveolar lavage for diagnosis of pulmonary histiocytosis X, Thorax 32, 303- 306 (1977). [50] Bignon, J. , Sebastien, P., and Bientz, M., Review of some factors relevant to the assessment of exposure to asbestos dusts, International workshop on biological specimen collection, Luxembourg 18-22 April 1977 (in press). [51] Pooley, F. 0. , Oral communication, Meeting on the biological relevance of fiber parameters, IARC June 30-July 1, Lyon (1977). 118

Discussion ~: FISHER: I noticed you used the term amphibole in your tables. Since I believe these were insulation workers, you mean amosite rather than the general mineral group? J. BIGNON: The identification of asbestos fibers has been done only by the morphology in TEM and by electron diffraction. As we did not use microanalysis to identify the different type of amphibole, and as we did not get accurate inquiries about the material used by patients, I cannot answer your question. FISHER: But these were insulation workers, am I correct? BIGNON: Yes. These workers sprayed a mixture of asbestos and other material; but as the material used by these workers changes from time to time, it is difficult to identify by a questionnaire the type of asbestos fibers to which the patients have been exposed. FISHER: The type of amphibole used would be one that would be considered a commercial form of asbestos and would only be useful for that purpose if it did have the long fiber length that you showed in your tables. I am trying to distinguish between this type of amphibole and the more general, more widely occurring forms. I think that's an important point. M. SCHNEIDERMAN: Is your question related to the fact that the type of amphibole used by the insulation workers is in some manner different from what one has in some other kinds of general exposures; is that what you're driving at? FISHER: Exactly, yes. SCHNEIDERMAN: Yes, I think Prof. Bignon agrees with you. G. WRIGHT: I have one question which is becoming increasingly bothersome. In looking at old materials from autopsies, the question of whether or not the material that was used for fixing the lung contains asbestos fiber is beginning to be raised. I would ask whether the materials you used in fixing the lung had been demonstrated to be asbestos fiber-free? The other is a comment, because your study, I think, demonstrates rather well the fol- lowing: the lung apparently is a concentrator of long fibers. In most occupational exposures, the ratio of fibers longer than 5 pm to those that are shorter is of the order of 20 to as much as 50 or 100 to 1. So if you find 17 percent of the residual fibers in the parenchyma are longer than 8 pm, this strongly suggests that the lung preferentially concentrates the, long fibers. There is very recent evidence by Arthur Morgan, in experi- mental animals, of precisely what you've shown. In acute experiments lasting for several months, the animal rather rapidly clears the short fibers and retains the long ones. So it's a very nice confirmation of your observations. BIGNON: The liquids we have used for lung fixation and processing were constantly filtered through 0.5 pm Millipore filters. N ~ 119 W H+ ~ ,.+ ~

-% C 3 National Bureau of Standards Special Publication 506. Proceedings of the Workshop on Asbestos: Definitions and Measurement Methods held at NBS, Gaithersburg, MD, July 18-20, 1977. (Issued November 1978) EPIDEMIOLOGIC EVIDENCE OF THE EFFECT OF TYPE OF ASBESTOS AND FIBER DIMENSIONS ON THE PRODUCTION OF DISEASE IN MAN W. Clark Cooper Equitable Environmental Health, Inc. Berkeley, California 94704 Abstract There is epidemiologic evidence to indicate that all types of commercial asbestos, i.e., chrysotile, crocidolite, amosite, tremolite asbestos, and anthophyllite asbestos, when inhaled, can cause pulmonary fibrosis and increase the risk of 1ung cancer. All but anthophyllite asbestos have been associated with malignant mesothelial tumors. There is also strong evidence to support a decreasing gradient of pathogenicity as one proceeds from crocidolite to amosite to chrysotile, but this evidence does not clearly rule out the interrelated influence of fiber dimension, shape, and co-factors. Clear-cut epidemiologic evidence related to differing fiber dimensions is scanty. Such information is critically needed. The most pressing need is to determine the pathogenicity of ultrafine fibers in the electron-microscope size range, and for fibers shorter than 5 micrometers, whether inhaled or ingested. It is suggested that there be expanded epidemiologic studies of populations which have been exposed to such fibers, without the presence of long fibers. This will probably occur where the exposures are incidental to operations other than commercial asbestos production. It is also recommended that there be systematic study of the fiber content of human lungs and other tissues, as related to causes of death. Key Words: Asbestos; asbestosis; carcinoma; epidemiology; fine particles; mesothelioma. When the seriousness of the problem of asbestos-related disease became generally recognized 15 to 20 years ago, it was regarded as arising solely from commercially-produced asbestos. Most evidence had been obtained from workers exposed during the mining, process- ing, or use of commercial chrysotile, amosite, crocidolite, anthophyilite asbestos, or tremolite asbestos, so studies logically focused on these types. The scientific and practical importance of determining whether all these types of asbestos were equally hazardous became apparent. One of the first recommendations made by the Working Group on Asbestos and Cancer, under the auspices of the International Union against Cancer, meeting in New York City in October, 1964, was "that the importance of fiber type on the risk of developing asbestosis, carcinoma of the lung, and mesothelial and other tumors be investigated" [1]1. Eight years later, meeting in Lyon, the Advisory Committee on Asbestos Cancers to the International Agency for Research on Cancer [2], the successor to the subcommittee that arose out of the 1964 Working Group, answered its own question: "Are all commercial types of asbestos able to cause lung carcinoma?" as follows: 'Figures in brackets indicate the literature references at the end of this paper. Preceding page blank - N

"Yes. Since 1964 the evidence of a causal relationship has been increased by epidemiological studies showing exposure-response relations for the incidence of lung carcinomas. The production of lung carcinomas in certain animals by all types of asbestos supports this conclusion. The epidemiological evidence in man, however, shows that there are clear differences in risk, with type of fibre and nature of exposure." With respect to mesothelioma, the Committee's report stated that, "There is evidence that all commercial types of asbestos except anthophyllite may be responsible. Evidence for an important difference in risk in different occupations and with the type of asbestos has increased. The risk is greatest with crocidolite, less with amosite, and apparently less with chrysotile. With amosite and chrysotile there appears to be a higher risk in manufacturing than in mining and milling." The Committee then made specific recommendations for projects assessing excess cancer risks following exposure to only one type of fiber, mentioning chrysotile, amosite, and chrysotile, with special emphasis on differences between those engaged in mining and milling and those engaged in the manufacture and use of these types of commercial asbestos. It was further recommended that there be investigation of "talc-exposed groups in mining and manufacturing to establish any differences in morbidity or mortality which might be related to the amount and shape of the fine respirable particles." In a related recommendation pertaining to experimental work, recognition was given to the need for more information about the role of fine particles, especially the influence of fiber size in the induction of tumors: "These studies should be extended to include fibres other than asbestos. A subcommittee should be established to review the need for, and arrange the distribution of, standard samples of asbestos and other fibres in addition to the UICC reference samples." Another pertinent recommendation was: "There is an urgent need for the quantitative assessment, size analysis, and characterization of particles and fibres in the lungs and other organs." Participants in the present workshop are engaged in the continuing search for answers to the foregoing questions, and it is apparent definitive answers are not easy to obtain. There is an expanded appreciation of the ubiquity of mineral fibers with shapes resembling those of commercial asbestos, with diameters extending into a range below detectability by light microscopy, and with lengths below 5 micrometers (pm), now arbitrarily used as the lower limit for occupational standards. Decisions on pathogenicity for man are urgently needed with respect to these, the asbestiform varieties of many minerals, and for all durable fibers in the range below light microscopic detection, i.e., below 0.4 or 0.5 pm in diameter, and which are very short, i.e., less than 5 pm in length. How can epidemiologic evidence contribute to these decisions? Epidemiologic studies cannot stand alone. They fit into a network of observations from many sources, including theoretical and observed information on the aerodynamic properties of particles, in vitro tests, studies in experimental animals, and isolated clinical observations. They are nevertheless, by definition, the final source for quantita- tive information in man, and ultimately must be the basis for establishing and evaluating environmental controls. Some of the effects in man which lend themselves to quantitative study and correlation with occupational or non-occupational exposures include: (1) Evidence of asbestosis, such as fibrosis of the lung parenchyma, fibrosis or thickening of the pleura, calcification of the pleura, and other non- malignant reactions as demonstrated by radiography, functional tests, physical examination, or study of tissues. 122

(2) Evidence of malignancy, notably carcinoma of the lung, mesothelioma of the pleura or peritoneum, or cancer of the gastro-intestinal tract, larynx, or other sites. (3) Evidence of past exposures, as demonstrated by fibers in various tissues, sputum, or urine. It is generally accepted that fiber characteristics probably operate differently with respect to different pathologic effects, so that asbestosis, lung cancer, mesothelioma, and other malignancies will follow differing dose-response curves as we consider different types and dimensions of fibers. Hopefully, we can obtain useful epidemiologic evidence by considering the patterns of disease, as related to different types and dimensions of mineral fiber, in groups identified as follows: (1) Populations whose preponderant exposure has been to one type of asbestos or the asbestiform variety of a mineral, whether by inhalation, ingestion, or both, which can be observed for periods of at least 30 years and preferably 50 years after exposures began, and which can be compared with groups having little or no exposure to the same or related fibers; (2) Populations with suspect diseases, whose past exposures can be reconstructed by history, records, place of residence, or body burdens of fibrous particles, and which can be compared with a matched series having some disease unlikely to be asbestos-related. This case-study method is most useful in relatively rare diseases, such as mesothelioma. (3) Populations having differing concentrations, types, and sizes of mineral fibers demonstrated at autopsy, to determine whether or not the patterns of pathology and causes of death correlate with differing tissue burdens of fibers. What evidence have we gathered to date, using the foregoing approaches? Types of Asbestos Used Commercially There is unequivocal evidence that chrysotile, amosite, crocidolite, tremolite asbes- tos, and anthophyllite asbestos can produce asbestosis and increase the risk of lung cancer. All but anthophyllite have been associated with an increased risk of mesothe- lioma. Grading the relative biologic activity of these several types of asbestos, in terms of the production of each type of asbestos-related disease, is more difficult. As Margaret Becklake [3] pointed out in her excellent review, it is not easy to control precisely for dosage and cofactors. Fiber diameter, length, and shape are highly interrelated with asbestos type and may be more important than chemical and crystal structure. The consensus that crocidolite is the most hazardous commercial asbestos has been derived from a number of studies, but these do not all rule out an influence of shape and size. Emphasis on crocidolite as being particularly hazardous arose from its early associa- tion with mesothelioma in the Northwestern Cape Province of South Africa, as first described by Wagner et al. [4]. Although the relative absence of mesothelioma in the crocidolite areas of the Transvaal reported by Sluis-Cremer [5] was at first questioned because of the exclusion of black and colored miners, Webster [6] has confirmed that there is a much lower incidence of mesothelioma in the Transvaal. Timbrell [7,8] has offered as an explana- tion the fact that crocidolite in the Northwest Cape is of smaller diameter (therefore more respirable) and shorter (therefore more likely to avoid interception in the airways) than the crocidolite of the Transvaal. It should be emphasized that although the Transvaal fibers averaged three times as long as the Northwest Cape fibers, both samples had many fibers above 5 pm in length. Webster [6] on the basis of pathologic observations of the distribution of fibers in the lungs has questioned the foregoing explanation. He has suggested that possibly concurrent exposures to iron and manganese in the Northwest Cape may have an influence. 123 2063104921

C With respect to lung cancer, Enterline and Henderson [9] compared the experience of workers making asbestos cement pipe, where both crocidolite and chrysotile were used, with that of others exposed only to chrysotile. Those whose exposures included crocidolite had 6.1 times the expected number of deaths due to lung cancer, while those exposed only to chrysotile had 1.4 times the number expected. Weill et al. [10] carried out comparative studies of two populations of workers, one making asbestos shingles containing chrysotile, and the other making shingles, flooring, and asbestos-cement pipe and exposed to both chrysotile and crocidolite. Those exposed to crocidolite had more small irregular opacities by x-ray, more pleural thickening, and significantly greater reduction in pulmonary function. Despite the consensus that crocidolite is probably the most hazardous type of commercial asbestos, the evidence does not appear strong enough to support a 10-fold stricter standard for a time-weighted average, or a 60-fold stricter standard for 10- minute exposures, as applied in the United Kingdom [11]. Amosite has been positively identified as responsible for pulmonary fibrosis, lung cancer, anTmesothelioma. Selikoff et al. [12] found a 10-fold excess of lung cancer, as well as 5 deaths from mesothelioma, in a population of 230 men who had been previously employed in an amosite-using plant, during the period 1960 to 1971. This has been one of the few opportunities in the United States to study workers without mixed exposures. The high rates of asbestosis, lung cancer, and mesothelioma in asbestos insulation workers have been in men with mixed exposures, to both amosite and chrysotile. The foregoing experience in an amosite-using industry is in striking contrast to that reported in the amosite mines in South Africa. Webster [6] states that of 232 confirmed cases of mesothelioma diagnosed in South Africa between 1956 and 1972, 78 had been in miners, but practically all had been exposed to Cape Blue crocidolite, with only two having had exposures only to amosite. As pointed out earlier, the fact that Transvaal amosite shared with Transvaal crocidolite the property of being thicker and longer than Northwest Cape crocidolite makes it impossible to ascribe the difference to type alone. Men exposed to crocidolite in the Transvaal also had relatively few mesotheliomas. Chrysotile has been rated the least pathogenic type of the three major forms of commercially-produced asbestos on the basis of relatively few studies in which exposures were limited to this type. Most such studies have been in workers engaged in the mining and milling of chrysotile, in Canada, Italy, Russia, and Cyprus. A report by Braun and Truan [13] indicated that the incidence of lung cancer in chrysotile miners and millers in Quebec, while slightly elevated, was not nearly as great as had been described in asbestos workers in the United Kingdom or in U.S. Insulators. These studies have been criticized for methodologic flaws, but it would now appear that they reflected a lower risk in chrysotile miners. More recent studies of Quebec miners and millers by McDonald et al. [14] show an excess of lung cancer, 5 times expected, only in the highest exposure group. Only 5 deaths from mesothelioma were found among 3,270 deaths. A more recent estimate by McDonald [15] gives the proportion of mesothelioma deaths as 8 out of 4,000 deaths. This is far less than the proportion found in U.S. insulation workers, where, for example, Selikoff found 77 of 1,092 deaths due to mesothelioma. Weiss [16] has recently studied the mortality in a group of 264 employees hired during the period 1935-1944 in a plant manufacturing chrysotile products, and who worked one year or more. The Standard Mortality Ratio (5MR) for lung cancer was only 0.93. Although the design of the overall study did not permit strict comparison with the study by Selikoff et al. (17] in an asbestos insulation material producing plant, comparison of groups with similar intervals from first exposure to end of operation indicated a significantly lower lung cancer risk in the Weiss study. These reports, combined with those of Weill et al. [10] and Enterline and Henderson [9] previously reported, suggest that chrysotile is less pathogenic than crocidolite or amosite. But, as Timbrell [8] has pointed out, the curliness of chrysotile fibers influences their deposition and transmigration, so shape and size may be more important than chemical composition per se. The evidence on anthophyllite asbestos comes almost entirely from Finland, where this form of asbestos was commercially developed until recently, and where there have been widespread non-occupational exposures. The extraordinary incidence of pleural calcification associated with low level exposures is well-documented (Kiviluoto) [18]. Kiviluoto and Meurman [19] and Nurminen [20] have shown in studies of workers exposed to 124

C anthophyllite asbestos that they have an increased risk of asbestosis and lung cancer, but. mesotheliomas have not been reported. Meurman et al. (21] analyzed 248 deaths in 1,092 anthophyllite miners and millers. There were 21 deaths from lung cancer, where 12.6 were expected; no mesotheliomas were reported. Studies of workers exposed to tremolite asbestos without associated exposures to other fibers are not sufficiently well documented to permit placing them in a gradient of response with other commercial types of asbestos. The same is true for actinolite asbestos. ' . Other Asbestiform Minerals What is the evidence for the pathogenicity of mineral fibers other than the types of asbestos commercially exploited? It is almost non-existent because, in the absence of commercial development and occupational exposures, contacts have been incidental to other operations and have been poorly documented and usually of less magnitude. The best of such studies have been associated with commercial talc operations. The presence of tremolite asbestos, anthophyllite asbestos, and chrysotile in many talc deposits has confirmed the potential of these types to produce fibrosis, pleural plaques, and to increase the incidence of lung cancer. There are no studies to indicate that (ibrous talc, In the absence of asbestos of the types mentioned, can produce disease in man, but one would predict that such fibers in the right size ranges would be pathogenic. Non- fibrous talc is apparently hazardous only if there is concurrent silica exposure. Rubino et a1. [22] reported on the mortality pattern in 1,346 talc miners and 438 talc millers, in which there were 931 deaths. Although there was an'increased incidence of si]icosis and silico-tuberculosis, they reported no excess in cancer. • They did nqt indicate any fibrous talc being present. . A promising source of information on a non-commercial asbestiform variety of mineral has been the population of the Homestake gold mine in South Oakota, where there have been exposures to amphibole fibers, described as predominantly in the grunerite series similar to those found in the Mesabi range of Minnesota, extending back for over 100 years. Unfortunately, results to date are far from conclusive, despite a published mortality analysis by Gillam et al. [23] and an environmental report by Dement et al. [24]. Gillam et al. reported a statistically significant excess of iung cancer deaths (10 contrasted with 2.7 expected) in 440 gold miners identified by the Public Health Service in a 1960 silicosis study. However, a more recent report by McDonald et al. (1977) covering deaths between 1937 and 1973 in 1,321 employees of the same mine who were members of the Homestake Veterans Club, and had worked 21 years or more, showed no excess lung cancer deaths. There were 660 deaths for analysis. There was an excess of deaths from pneumoconiosis and pulmonary tuberculosis. This, and the excess of non-malignant respiratory disease deaths reported by Gillam et al. is not surprising, since 39 percent quartz had been demonstrated in settled dust. Records kept by the mines since 1937 showed dust concentrations ranging from 11 to 25.5 million particles per cubic foot (mppcf) before 1952, greatly exceeding standards for free silica. The miners who died of non- malignant respiratory disease had begun work as early as 1916. Even if an excess of lung cancer were proven in the Homestake mine, attributing it to low concentrations of mineral fibers would not be justified without careful consideration of what is known of smoking histories and concurrent exposures to arsenic and radon daughters. Asbestiform minerals almost certainly cannot be held responsible for the excess deaths from non-malignant respiratory disease, in view of quartz exposures and death certificates which in most cases had diagnoses of silicosis. It is absurd to attribute fatal pneumoconiosis in such a situation to grunerite fibers at levels approximating one-tenth the current standard for asbestos. Swent [25] has critically reviewed the Gillam study and documented ventilation back to 1916 and dust counts to 1937 which show that the assumption that past exposures to silica, arsenic, radon daughters, and fibers were the same as those found in a 1972 survey is untenable. As matters now stand, the Homestake study cannot be regarded as supporting the pathogenicity of grunerite fibers. One awaits the results of new studies being supported 125 2063104923

s by NIDSN, which may establish the mortality patterns with more certainty and hopefully will permit more accurate estimates of past exposures. Influence of Fiber Dimensions Throughout consideration of types of asbestos, it is apparent that type cannot be separated from shape and size. This is true even when exposures are characterized solely on the basis of fibers in the light microscopic range (i.e., with diameters greater than 0.4-0.5 pm) and those greater than 5 pm in length. It has been demonstrated in recent years, however, that neither in standard reference samples of commercial asbestos (Langer) [26], nor in air and water samples, nor in lung tissue, are fibers mainly in the light microscopic size range. Furthermore, as Pooley [27] has shown, even chrysotile miners and millers contain large numbers of amphibole fibers, most of them in the microfiber range, in their lung tissues, so their exposures are mixed. When we turn to consideration of epidemialogic evidence on fiber dimensions, either within a given species of commercially used asbestos, or in the asbestiform varieties of minerals not used commercially, there is relatively little to report. There is suggestive but not conclusive evidence from South Africa (7] that relatively short and fine fibers are more likely to produce mesotheliomas than longer and thicker fibers, but these are within the range of light microscopy and longer than 5 micrometers. There are no conclusive studies in man to support the strong evidence from animal studies that very short fibers (under 5 pm) are non-pathogenic. In considering the influence of fiber size, the question of the ultrafine fiber must be separated from the question of the very short fiber. The ultrafine fiber is defined as one below the level of resolution by the light microscope, i.e., less than about 0.4 pm in diameter, down. to the size of the smallest chrysotile fibril, of the order of 0.025 pm or 250 Angstrom units. Evaluation of such ultrafine fibers is of great importance because: 1) diameter has a strong inverse relationship to falling speed, so such fibers remain airborne for long periods and are highly respirable, although their capture and retention will vary not only with diameter, but also with length; 2) they are found in large numbers in lung tissues, both in individuals occupationally exposed and those without such exposures, but seldom to the exclusion of large fibers [28]; 3) they have been found to be widespread in coamunity air [29] and in association with the quarrying and use of serpentinite rock [30]; 4) they are not included in fibers counted by the methods currently recommended for monitoring work environments, and are not covered by current standards; 5) data are not being systematically collected on the numbers of ultrafine fibers in the air nor how their concentrations relate to the concentrations of larger fibers found in various occupational and environmental situations. There are no epidemiologic studies in which ultrafine fibers are an isolated variable. All studies of populations exposed to commercial asbestos have involved heavy exposures within the light microscope range, i.e., to fibers larger than 0.5 um in diameter, so the contribution of ultrafine fibers cannot be determined. On the evidence from studies in animals, it is likely that such fibers, when longer than 5 or 10 pm, would be pathogenic. 126

The problem of the very short fiber is more critical: `1) studies in animals strongly suggest a decreasing gradient of fibrogenic risk and carcinogenic potential (at least for mesothelioma) for fibers shorter than 5 to 10 micrometers; 2) samples of naturally occurring chrysotile, amosite, and crocidolite have been shown to contain a majority of fibers shorter than 5 Ym in length (28); 3) lung tissue contains a high proportion of short fibers; 4) samples of ambient air in many areas, such as those collected near taconite mining operations in Minnesota, and associated with crushed rock in Montgomery County, Maryland, are predominantly short fibers (303; 5) since current monitoring methods for the occupational environment exclude fibers shorter than 5 pm, data are not being systematically collected. The biologic activity of short fibers in man is not known. By analogy with studies in animals one would not expect fibers shorter than 5 pm or 10 pm in length to produce asbestosis or mesothelioma. The only epidemiologic study in which fibrosis and excess lung cancer has been attributed to exposures which were predominantly too short, ultrafine fibers is that of Gillam et al. (23] in the Homestake mine. Here 94 percent of fibers were less than 5 pm in length, the median diameter was 0.13 pm, and the median length was 1.1 Nm. For reasons pointed out earlier, these exposures, which were described as' consisting largely of grunerite with some fibrous cummingtonite and hornblende, are inconclusive. Neither the actual mortality experience nor the past exposures a'r,e well enough defined to be used as scientific evidence. ' The. case report by Miller et al. [31] in which a 53-year old man who died with extensive interstitial pulmonary fibrosis was found to have had large numbers of ultrafine, short fibers in his lungs cannot in itself establish a causal relationship, nor does it indicate how often such an association might occur. It is analogous to an earlier report by Miller et al. [32] who made a somewhat similar finding in a man who had been exposed for many years to talc in a rubber products plant and whose lungs showed enormous numbers of submicroscopic talc particles (non-fibrous). Both reports suggest that overwhelming concentrations of a reactive dust may in some individuals produce generalized interstitial fibrosis. It does not tell us how often such might occur, nor provide any information on relationships with malignancy. The essentially negative evidence as to health effects from the airborne fibers associated with taconite mining operations in Minnesota, and the negative evidence from Duluth (Masson et al.) [33] with respect to the ingestion of ultrafine, short fibers in Lake Superior water are reassuring, but it is too soon to rule out effects with long latent periods, i. e. , 25 years or more. In summary, no populations whose exposures have been confined to ultrafine fibers, short fibers, or fibers which are both ultrafine and short, have been defined or studied long enough to permit epidemiologic evaluation. There have been several studies in recent years in which the concentrations of fibers in lung tissue have been quantitated and described, with some attempt at correlation with pathologic changes. That of Ashcroft and Hepplestone (1973) [34] was limited to 35 individuals with asbestos bodies detected in histological sections, and all but one had definite or probable histories of occupational exposure. The authors found that from 12 to 30 percent of the fibers were optically visible, the rest being detectable only by electron microscopy. (They did not describe fiber lengths.) There was a general correlation between fiber concentration and asbestosis, up to the level of moderate asbestosis. Another study, by Doniach et al. [35], was limited to optically visible asbestos bodies in a London necropsy series. The study by Pooley [27] of the lungs of 127 2063104925

individuals with asbestosis who had been employed in the chrysotile mining industry in Canada, and in 30 individuals who died with mesothelioma, provided valuable information on the relative proportions of chrysotile and amphibole fibers and on the large numbers of EM-sized fiber present, but no detailed data on lengths and diameters of fibers were presented. Its most interesting feature was the large number of amphibole fibers that were found in chrysotile miners. In short, we know of no large series of cases in which the numbers and sizes of fibers in tissues have been correlated with causes of death. Studies Which Are Needed How can the necessary epidemiologic evidence be obtained? It can be accepted without reemphasis that infection and inhalation studies in animals, testing various types of asbestos and asbestiform varieties of other minerals in appropriate size ranges, must be done. It is not likely that further study of individuals who mine, mill, process, or use commercial asbestos will do more than tune more finely what we already know. Even though this is desirable and necessary, it is not likely to answer questions about very fine or very short fibers, since the nature of commercial asbestos is such that long fibers are always present. Only if dust control measures preferentially increase very greatly the proportion of short fibers in the electron microscope range would studies in commercial asbestos operations provide useful information regarding fiber size. We must turn to other populations, where exposures have been incidental to non- asbestos industrial operations but which liberate or disperse asbestiform varieties of minerals in the electron microscope range below 5 pm in length. The Homestake mine has had this type of population, but here a positive finding would lead to a need to consider several confounding variables. On the other hand, an absence of serious risk would be . highly reassuring, if past exposures were found to have been high. Other populations which might be studied are those in association with taconite mining and milling operations, where, in some areas, the airborne mineral fibers are predominantly less than 3 pm in length and do not represent any form of commercial asbestos. There are many sections of the United States where chrysotile and amphibole fibers are present in the natural rock and have been present in air or drinking water for long periods of time. Careful search should be made for areas which might permit comparisons of malignancy patterns as related to such exposures. The work of Fears (1976) [36], who found no excess of cancer in U.S. counties with known asbestos deposits, needs to be refined to concentrate on census tracts contiguous to operations which actually increase fiber concentrations in the air or water. A second approach which should be expanded is the large scale study of the fiber content of human lungs and other tissues, with determination of fiber concentrations and fiber dimensions, for comparison with causes of death. This has been periodically suggested but never actively pursued. Stanton (1974) [37] stated, "There is perhaps one way to determine the hazards of fibers without waiting the many years necessary for the effects of even massive exposure to become evident. Unlike most carcinogens, fibers that are a threat are sufficiently durable to remain in the tissues from which cancers are derived. Since carcinogenic response can be related to doses of sized fibers in experimental animals, it may be possible to equate the number and size distribution of fibers in human tissues to cancer in man. Although much has been accomplished in assessing large, protein-coated fibers in human lungs, surprisingly little has been done in assessing the size distribution and total quantity of all fibers in human tissues. This would be a tedious job, but it might determine the true significance of fibers as carcinogens in man." It is believed that the design and organization of such a major study is long overdue. Without the information it might provide, environmental decisions involving ultrafine and ultrashort asbestos fibers or the asbestiform varieties of other minerals will continue on a very uncertain and often emotional basis. When one considers the tremendous outlays involved in containing or capturing such fibers in mining and quarrying operations, as well as in asbestos-using industries and in waste disposal, the cost of such studies would 128

appear a prudent investment. As Rohl, Langer, and Selikoff observed in their recent report [30] providing data on fibers found near Montgomery County roads where serpentinite rock had been used, "The evaluation of the possible health hazard that may be associated with this exposure requires information that is not yet known in the scientific community: (i) the biological activity of short chrysotile fiber, (ii) the level of exposure to asbestos which is safe insofar as human cancers are concerned, if a safe level exists, and (iii) the biological activity of asbestiform silicates, not necessarily asbestos." The same comment applies to numerous other environmental situations currently under scrutiny. We do not know what fiber concentrations expressed in nanograms per cubic meter or in total fibers per unit volume, when detected by electron microscopy, mean in terms of human health. Unfortunately, epidemiology does not yet provide the answers. Summary and Conclusion There is epidemiologic evidence to indicate that all types of commercial asbestos, i.e., chrysotile, crocidolite, amosite, tremolite asbestos, and anthophyllite asbestos, when inhaled, can cause pulmonary fibrosis and increase the risk of lung cancer. All but anthophyllite asbestos have been associated with malignant mesothelial tumors. There is also strong evidence to support a decreasing gradient of pathogenicity as one proceeds from crocidolite to amosite to chrysotile, but this evidence does not clearly rule out the interrelated influence of fiber dimension, shape, and co-factors. Clear-cut epidemiologic evidence related to differing fiber dimensions is scanty. Such information is critically needed. The most pressing need is to determine the pathogenicity of ultrafine fibers in the electron-microscope size range, and for:fibers shorter than 5 micrometers, whether inhaled or ingested. It is suggested that there be expanded epidemiologic studies of populations which have been exposed to such fibers, without the presence of long fibers. This will probably occur where the exposures are incidental to operations other than commercial asbestos production. It is also recommended that there be systematic study of the fiber content of human lungs and other tissues, as related to causes of death. References [1] Report and Recommendations of the Working Group on Asbestos and Cancer, Ann. N. Y. Acad. Sci., 132, 706-721 (1965). - - - [2] Report of the Advisory Committee on Asbestos Cancers to the Director of the Interna- tional Agency for Research on Cancer, in Biological Effects of Asbestos, Proceedings of a Workin Conference, Lyon, 1972, P. Bogovski, V. mbrell,ll, J. C. Gilson, and .7. C. Wagner, eds. , IARC Scientific ublications No. 8 Lyon, pp. 341-346 (International Agency for Research on Cancer, 1973). j3] Becklake, M. R., Asbestos-related diseases of the lung and other organs: their epidemiology and implications for clinical practice, Amer. Rev. Resp. Dis., 114, 187- 227 (1976). [4] Wagner, J. C., Sleggs, C. A., and Marchand, P., Diffuse pleural mesothelioma and asbestos exposure in North Western Cape Province, Brit. J. Ind. Med., 17, 260-271 (1960). [5] 3lu3}oC319me~1,97G0) K., Asbestosis in South African asbestos miners, Environ. Res., [6] Webster, I., Malignancy in relation to crocidolite and amosite, in Biological Effects of Asbestos, Proceedings of a Workina Conference, L jyon, 1972, P. Bogovski, V. TimbreTS, J. 77-GiTs-on, and J. C. Wagner, eds., I~enti ic -hublications No. 8, Lyon, pp. 195-198 (International Agency for Research on Cancer, 1973). 129 2063104927

[7] Timbrell, V. , Griffiths, 0. M., and Pootey, F. D., Possible biological importance of fiber diameters of South African amphiboles, Nature, 232, 55-56 (1971). [8] Timbrell, V., Physical factors as etiological mechanisms, in Biolo ical Effects of Asbestos, Proceedin s of a Working Conference, L9n, 1972, P. ogovski, V. imbrelT, J.- C.ilson, and .ff. Wagner, eds.,, A C cientific Publications No. 8, Lyon, pp. 295-303 (International Agency for Research on Cancer, 1973). [9] Enterline, P. E. and Henderson, V., Type of asbestos and respiratory cancer in the asbestos industry, Arch. Environ. Health, 27, 312-317 (1973). [10] Weill, H., Ziskind, M. M., Waggenspack, C., and Rossiter, C. E. , Lung function consequences of dust exposure in asbestos cement manufacturing plants, Arch. Environ. Health, 30, 88-97 (1975). [11] Guidance Note from the Health and Safety Executive: Asbestos-hygiene standards and measurement of airborne dust concentrations, Environ. Hvaiene/10 (London, HM Factory Inspectorate, December 1976). [12] Selikoff, I. R., Hammond, E. C., and Churg, J., Carcinogenicity of amosite asbestos, Arch. Environ. Health, 25, 183-186 (1972). [13] Braun, 0. C. and Truan, T. 0., An epidemiological study of lung cancer in asbestos miners, Arch. Ind. Health, 17, 634-653 (1958). [14] McDonald, J. C., Becklake, M. R., Gibbs, G. W., McDonald, A. D., and Rossiter, C. E., The health of chrysotile asbestos mine and mill workers of Quebec, Arch. Environ. Health, 28, 61-68 (1974). [15] McDonald, A. D. and McDonald, J. C., Etudes epidemiologiques sur les maladies dues a 1'amiante au Canada, Rev. Fr. Ma1 Reso., 4 su 1 2), 25-38 (1976). [16] Weiss, W.; Mortality of a cohort exposed to chrysotile asbestos in a manufacturing plant (in manuscript, 1977). [17] Selikoff, I. J., Hammond, E. C., and Seidman, H., Cancer risk of insulation workers in the United States, in Biolo i~_cal Effects of Asbestos, Proceedings of a Working Conference, Lyon, 1972, P. og8 ovski, V. TimbreIlTT Gi so'1 n, and J. C. Wagner, eds., Scientiff-cPublications No. 8, Lyon, pp. 209-216 (International Agency for Research on Cancer, 1973). [18] Kiviluoto, R., Pleural calcification as a roentgenologic sign of non-accupatianal endemic anthaphyltite-asbestosis, Acta Radiol. Suppl., 194, pp. 1-67 (1960). [19] Kiviluoto, R. and Meurman, L., Results of asbestos exposure in Finland, in, Pneumo- coniosis: Proceedin s of the International Conference, Johannesbura, 1969, H. A. Shapiro, e., ape Town, tSxfor University ress, 1970). [20] Nurminen, M., The epidemiologic relationship between pleural mesothelioma and asbestos exposure, Scand. J. Work Environ. Health, 1, 128-137 (1975). [21] Meruman. L. 0., Kiviluoto, R., and Hakama, M., Mortality and morbidity among the working population of anthophyllite asbestos miners in Finland, Brit. J. Tnd. Med., 31, 105-112 (1974). [22] Rubino, G. F., Scansetti, G., Piolatto, G., and Romano, C. A., Mortality study of talc miners and millers, J. Occup. Med., 18, 186-193 (1976). [23] Gillam, J. C., Dement, J. M., Lemen, R. A., Wagoner, J. K., Archer, V. E., and Blejer, H. P., Mortality patterns among hard rock gold miners exposed to asbestiform mineral, Ann. N. Y. Acad. Sci., 271, 336-344 (1976). 130

Cl [24] Dement, J. M., Zumwalde, R. D., and Wallingford, J. M., Discussion paper: Asbestos fiber exposures in a hard rock gold mine, Ann. N. Y. Acad. Sci., 271, 345-353 (1976). (25] Swent, L. W., Herrin, G. R., Waterland, J. K., and Bell, R. F., Mortality patterns among hard rock gold miners exposed to an asbestiform mineral - a critique. Presented at American Industrial Hygiene Conference, New Orlenas, LA, May 26, 1977. [26] Langer, A. M., Mackler, A. D., and Pooley, F. D., Electron microscopical investigation of asbestos fibers, Environ. Health Perspectives, 9, 63-80 (1974). [27] Pooley, F. D., An examination of the fibrous mineral content of asbestos lung tissue from the Canadian chrysotile mining industry, Environ. Res., 12, 281-298 (1976). [28] Langer, A. M. and Selikoff, I. J., Chrysotile asbestos in lungs of residents of New York City, International Union of Air Pollution Prevention Association's Second International Clean Air Congress, Washington, D.C., December 6-11, 1970. [29] Nicholson, W. J. and Pundsack, F. L., Asbestos in the environment, in Biolo ical Effects of Asbestos, Proceedings of a Workin Conference, on, 1972, P. ogovsk , im re31, J. C. Gilson, and 3-C. Wagner, eds., RC c ent~f9c Publications No. 8, Lyon, pp. 126-130 (International Agency for Research on Cancer, 1973). [30] Rohl, A. N., Langer, A. M., and Selikoff, 1. J., Environmental asbestos pollution related to use of quarried serpentine rock, Science, 196, 1319-1322 (1977). [31] Miller, A., Langer, A. M., Teirstein, A. S., and Selikoff, I. J., "Nonspecific" interstitial pulmonary fibrosis: association with asbestos fibers detectbd by electron microscopy, N. Enal. J. Med., 292, 91-93 (1975). ' [32] Miller, A., Teirstein, A. S., Bader, M. E., Bader, R. A., and Selikoff, I. J.; Talc pneumoconiosis: Significance of sublight microscopic mineral particles, Am. J. Med., 50, 395-402 (1971). (33] Masson, T. S., McKay, F. W., and Miller, R. W. , Asbestos-like fibers in Duluth water supply. Relation to cancer mortality, J. Amer. Med. Assn. , 228, 1019-1020 (1974). [34] Ashcroft, T. and Heppleston, A. G. , Asbestos fibre concentration in relation to pulmonary reaction, in, Biolo ical Effects of Asbestos, ~Proceedin _s of a W~~orkin Conference, Lyon, 1972, P. ogovs i,-O-Timbell,~T i soGan .~ C A a gner, eds. , TA1(C 3cientf?T-c Publications No 8, Lyon, pp. 236 (International Agency for Research on Cancer, 1973). [35] Doniach, I., Swettenham, K. V., and Hathorn, M. K. S. , Prevalence of asbestos bodies in a necropsy series in East London: Association with disease, occupation, and domiciliary address, Brit. J. Ind. Med., 32, 16-30 (1975). [36] Fears, T. R., Cancer mortality and asbestos deposits, Amer. J. Epidemiol., 104, 523- 526 (1976). [37] Stanton, M. F., Fiber carcinogenesis: Is asbestos the only hazard? Editorial, J. Nat. Cancer Inst. , March 1974, 633-634. oN W '.+ 131 a° N b

t DISCUSSION J. DEMENT: I'.d like to make several observations dealing with a couple of points. First of all, Dr. Cooper pointed out that the Homestake mine study dealt with exposure to very short fiber lengths, and that's certainly true. However, you failed to point out that in most industrial settings, as high as 99 percent of the fibers, of chrysotile especially, are shorter than five pm in length with very typical lognormal distributions which follow closely to the Homestake study. Secondly, a couple of comments with respect to the Homestake study. In its publication, NIOSH did in fact recognize the possible contributory effects of free silicate exposures for non-malignant respiratory disease. Our study ascribed the cancers predominantly to fibrous grunerite exposures. With regard to the McDonald study, I'd also like to make a couple of comments. First of all, it was a group from a Veterans Association with 21 years minimum employment at Homestake, but not necessarily underground mining. The copy of the Homestake paper, which I have been given, does not indicate whether or not they were miners or surface workers. Homestake operates several above-ground facilities. One must question whether or not 21 years requirement isn't a selective population, especially with regard to the data we saw from Dr. Nicholson today where he indicated that even one month carries with it an excess risk. Thirdly, we at NIOSH of course do realize the importance of the study as evidenced by our increase in the scope of the study, mainly to get a larger cohort to study. I would like to express a bit of gratitude for your pointing out that lack of evidence should not be taken as lack of effect. W. COOPER: With respect to the proportion of individuals underground, I can't answer that question. I think that the criticism of limiting it to individuals who had worked for 21 years or more is not a valid criticism. Actually, these were not retired miners, and even a study of retired miners is not necessarily a bad study; Enterline has developed the arguments pro and con very well. The fact that members of this club had been there for 21 years is not very different from, the basis that Selikoff and Hammond used in selecting their population of insulating workers, which was limited to those who had their initial exposures, or rather had been insulators, 20 years or more. The same processes of selection which keep an insulating worker working for 20 years keep a miner working 21 years; I do not think that this is a valid criticism. As to whether or not the paper ascribes the non-malignant respiratory disease to asbestos or to silica, I think it is unmistakable. The paper, as I recall, does not use the word "silicosis," except in describing the population as having come from a Public Health Service study of silicosis. I will read from the conclusion: "The observed excess of malignant respiratory disease can therefore be attributed to asbestos, singly or in combination with cigarette smoke, and that of non-malignant respiratory disease can therefore be ascribed to asbestos with a possible additive role from low exposures to free silica dust." That's a direct quotation from the report, so I think the implication is that the non-malignant respiratory disease is asbestosis. 132

CS National Bureau of Standards Special Publication 506. Proceedings of the Workshop on Asbestos: Definitions and Measurement Methods held at NBS, Gaithersburg, MD, July 18-20, 1977. fIssued November 1978) PATHOPHYSIOLOGY IN RELATION TO THE CHEMICAL AND PHYSICAL PROPERTIES OF FIBERS Paul Kotin Johns-Manville Corporation Denver, Colorado 80217 Abstract The array of asbestos-related diseases are reviewed in relation to their pathogenesis, pathology, and natural history. Biological avail- ability following host entry is especially critical for the biological effect of asbestos. Experimental data consistently demonstrate that hazard is related to the geometry of fibers, with fiber diameter and fiber length being primary determinants. Controversy exists as to the extent of influence of the two major classes of asbestos fiber: chrysotile and amphibole. Considerations affecting the anatomic and metabolic fate of asbestos fibers are also discussed. Key Words: Asbestosis; fibers; lung cancer; mesotheliomas, pathophysi- ology; toxicology. Any postulated role for exogenous agents in the etiology (cause) and pathogenesis (development) of tissue change or clinical disease is critically dependent on the biotogical. availability of the agent. Biological availability is defined here as, "possessing chemical, physical, and steric properties that allow reaction with receptor sites in the living system at the host, organ, tissue, cell, and macromolecule levels." In consequence, the environmental presence of a potentially toxic agent need not inevitably assume an adverse biological effect. For example, fly ash, no less than soot, contains carcinogenic hydrocarbons; yet the latter may be carcinogenic to man whereas the former is harmless since it cannot be respired. A low dose of a chemical may be metabolized to a harmless metabolite, while by an alternative biochemical pathway a higher dose may yield a proximate carcinogen, as, for example, with vinyl chloride. Perhaps nowhere does biological availability play a greater role in the pathogenesis of disease than in relation to fibers. Clinical and epidemiological studies describing the asbestos-related diseases have already been presented, and later in this workshop Dr. Mearl Stanton will report on his elegant experimental studies on fibers. My presentation will attempt to describe in an omnibus and therefore relatively superficial fashion the continuum of environmental and host factors that result in pathology and disease due to exposure to excessive concentra- tions of asbestos. To accomplish this, I will formulate a series of questions and let the answers provide the desired overview. Before doing this, however, let me emphasize now, and elaborate throughout my remarks, that the adverse effects of asbestos, like those of all environmental agents, occur in accordance with recognized toxicological principles. The chronic effects of asbestos exposure-asbestosis, mesothelioma, lung cancer, and possibly gastrointestinal cancer, if it indeed is truly related to asbestos exposure-are characterized by four rela- tively common aspects of environmental response: 1. A long latent period ensues following onset of exposure for either stigmata of exposure or clinical disease to appear. For the latter, time is measured in decades or segments of the total life span. 133

2. Exposure is in accord with recognized principles of dose-response in relation to disease development and appearance. Dose, the product of concentration or intensity of exposure multiplied by duration of exposure (time), is clearly the indispensable element in any current hazard analysis and in future projection. Dose-response considerations apply at all levels of response from the single cell to the intact host. 3. A no-effect level of exposure or threshold (if that particular word does create argument) exists for asbestos-related disease. 4. Multifactorial etiology plays a role for some of the asbestos-related diseases. The issue of the determinant and the modifier in a multicausation situation is a critical one. It appears to be that cigarette smoking is the determinant for lung cancer. The data on the role of cigarette smoking in the development of asbestosis, though a factor, are too recent to permit any conclusions even though a modifier role appears reasonable. Now, the questions that can be used to provide an overview of our subject are: 1. Inasmuch as asbestos is a generic term for a group of fibrous crystalline hydrated silicates, which of the spectrum of characteristics of this group are of relevance to the initiation of asbestos- related disease? The mineralogy and chemistry of asbestos have been reviewed in detail in this morning's session. Of the two major sets of characteristics, chemical and physical, fiber chemistry appears at this time to play only a minor role, if in fact any role at all, in relation to asbestos-associated disease. Physical characteristics, specifically fiber size, surface character, internal architecture and substructure, are all related in varying degrees to biological effect. Prior to addressing the second question, a brief description of the gross and micro- scopic anatomy and the physiology of the respiratory tract is necessary. As can be seen in figure 1, fibrous particles enter the lungs via the trachea following inhalation through the nose or mouth and are distributed throughout the tracheobranchial tree, ultimately reaching the alveoli or air sacs. These air sacs are like the spaces in sponges and are lined by thin membranes in which the capillaries and venules flow. The entry and penetration of fibers into the lung is governed by physical laws. For those particles which get into the tracheobronchial tree, some will settle on the lining and they will move upward (on the mucociliary escalator) where they will be unconsciously swallowed or spit out. Particles small enough to reach the alveoli will settle out on the lining of the air spaces where they may be engulfed by phagocyte cells (macrophages) which may neutralize them or carry them up to the mucociliary escalator so they can be removed. The mechanism whereby uningested fibers penetrate the lining of the tracheobronchial tree or the air spaces so that they may reach the pleura is largely unknown. Thus it is the mucociliary escalator and the macrophages that are the primary defense mechanisms of the lung. Of particular interest is the fact that cigarette smoke is the most potent and ubiquitous of all inhalants in its capability to neutralize or destroy the effectiveness of the lung defense mechanisms. 134

Figure 1. Anatomy of the respiratory tract. 135

The second question in relation to fiber effect is: 2. Following host entry, which of the anatomic and physiologic character- istics of the respiratory tract that I have just described affects the anatomic fate of the inhaled fiber? What is the algebraic summation of depositior, retention, parenchymal localization and mobilization as factors governing lung clearance? With respect to the chemical characteristics of fiber, there appears to be no consistent identifiable effect of__cemica com-1 position after host entry of fibers by inhalation. With respect to hp ysical characteristics the following effects are noted: Size. Fibers greater than 5 pm in diameter are virtually entirely lodged in the nose an do not penetrate the respiratory tract. Fibers greater than 3 Nm and less than 5 pm in diameter enter the trachea but do not reach the conducting airways deeply enough to be retained in the lung. Fibers less than 3 pm and more than one micron can penetrate to the smaller bronchi. Fibers in the millimicron range in diameter are deposited in the peripheral airways and air spaces through Brownian movement. All these dimensions are very close approximations. Length is probably less critical than diameter in relation to anatomic localization but it is of great importance in relation to biological effect. One possible measure of localization is the length of fibers found in the lungs in both experimental animals and man following environmental exposure. There are few fibers longer than 100 microns. There are virtually none longer than 200 pm. The majority are less than 50 Nm in length. We can conclude that only fibers thinner than 3 pm and shorter than 200 pm are of significance in eliciting a biological response in intact animals. ShaAe. Chrysotile asbestos is curly and spiral, whereas amphibole is harsh and rigid. ~t is imperative to emphasize that in relation to interception and deposition on the wall of air-conducting passages a curly or a spiral fiber behaves like a straight fiber having the diameter of the spiral fiber's maximum dimensional curl. Timbrell [1]1 concludes from his studies that chrysotile fibers (curly, pliable) do not penetrate into the deeper and more peripheral portions of the lung to the extent that the more rigid fibers of crocidolite and amosite do. More recently, using isotopically labeled fibers, Morgan [2] has obtained data that tend to question this generalization. It seems, then, that as a major determinant of biological localization and effect, shape is still an open question. Surface Character and Internal Architecture. Surface charge and leaching characteristics ai~t been i ed nEied to at~ as 6eing of major importance in relation to question two. Time may change this. In contrast, internal architecture has been shown to be relevant. In fact, chrysotile stands in sharp contrast to the amphiboles. The long, pliable fibers are readily split longitudinally into progressively finer fibrils and this feature may be critically related to biological effect. An unanswered yet crucial question is the one of durability of fibers in living systems. Quantitative data on the splitting of fibers and their solubility in relation to persistence of fibers are an urgent need. In sumnary, size and shape are the major determinants of anatomical localization and retention. iFigures in brackets indicate the literature references at the end of this paper. 136

C Now let's move on to question three and see what can or may follow when fibers set up residence in the lung: 3. During and subsequent to anatomic localization, what characteristics affect the biological fate of asbestos fibers at physiological, pharmacological, and biochemical levels, and what is the sequence of the morphogenetic events and altered morphology resulting from asbestos exposure at cell, tissue, and organ level sites? It is clear that as desirable as data from man might be in assessing the importance of the chemical and physical variables of asbestos in relation to asbestos-associated disease, reality forces the conclusion that observations on humans, alive or dead, are incapable of providing all the information necessary for this purpose. Most of our current knowledge is derived from laboratory experimentation, and it is to this resource that we must turn for our needs. Experimental data have been derived from research in which animal models have been exposed (a) in chambers to clouds of asbestos fibers (the most physiological method and the most analogous to human environmental exposure experience); (b) by intratracheal installation of the test material (less physiologic but highly useful and informative); or (c) by intracavitary installation (the least physiologic and the most artifactitious inasmuch as this method "forces" biological availability where, in fact, in the human situation none may exist; this method is useful as a tool for studying in-site cellular responses and mechanisms). Chemical composition of the several forms of asbestos can be dismissed as a major factor in the pathophysiology of asbestos, not because fiber chemistry may not indeed play a role, but because at our current level of ignorance we have no proven concept Cf what such a role might be. In support of eliminating chemical composition as a factor ,is the consistent observation that in experimental models all forms of asbestos can produce asbestosis, lung cancer, and mesothelioma depending on the mode of exposure. The report on the federally supported asbestos feeding study, to be presented later during this workshop, may shed more light on this mode of exposure. The size of the fiber, in sharp contrast to chemical composition, is the most clearly documented pFysical characteristic that determines biological effect. Data on the fiber size and cause-and-effect relationship are virtually entirely derived from the laboratory, since, in human experience, exposure has been in a mixed length and diameter milieu, thereby rendering epidemiological data worthless for assessing single size fiber effect. If only one axiom were permissible in my remarks it would be that on the basis of the d namic_s and kinetics of the behavior of airborne fibers, and in accordance with our know- e1 d e of iologl^cT~vailabiTi-tyTTo-taanatomi~f and at6o h s,a o ua Tibers thicker t an 3 microns and oi nger thanZu6 microns, or ti•i'i-cker t an . microns anasFo_rterkhan 7microns are dev-0oid of biolo ica]- e`ect. TnhaTat on experiments iave confirmed thTs anatomica l, and intrapTeuraT st'udies support the conclusion pathophysiologically. Three studies can be cited to challenge this statement: 1. Holt [3] reported the production of pulmonary fibrosis in animals exposed in chambers to a cloud of predominantly short fibers. However, his own data record the contamination of his sample with long fibers (greater than 10 microns). 2. King [4] is silent on the percent of longer fibers contaminating his short fiber sample when he reports fibrosis produced in animals. He says the sample was almost all short fibers, but he used a technique for sample preparat oi fhat in the hands of others (experienced fiber researchers) consistently fails to yield a "pure" short sample. 3. Pott and Friedrichs [5] recently reported the production of peritoneal mesothelioma with samples made up of fibers shorter than 2 microns. This is a serious challenge to the long-thin concept. I can suggest possible 137

factors confounding their experiment and conclusions, but at present suffice it to say that we are reviewing their findings in great detail. This controversy would be rhetorical were it not that, except for the above, all physiological studies and research reports on biological mechanisms are compatible with experimental bioassay in relation to the role of fiber size. Briefly, the sequence of events is as follows: Respired particles can settle at levels of the tracheobronchial tree which are covered by a mucous blanket that fs constantly being propelled cephalad toward the pharynx by the ciliated cells. Clearance of the particles from the lung by this mechanism is brisk, rapid (minutes to hours), and effective. Particles can also penetrate to the distal bronchioli and air sacs (the nonciliated regions). They can be cleared here also, provided they do not penetrate but remain on the surface. This clearance is slow (days to years), moderately effective, and the particles may need help via phagocytosis to decrease penetration and to migrate up to the mucous escalator. The importance of size can be demonstrated at this stage. Allison [6] and others have shown that short fibers (those less than 5 pm) appear to be readily and completely phagocytosed, whereas long fibers are not, even when simultaneously attacked by more than a single macrophage. This process may lead to cell fusion and the formation of giant cells which are usually found in abundance at the site. Estimates as to the efficiency of the combined clearance mechanisms range up to 95 to 98 percent. It is especially noteworthy, though, that mucociliary clearance is minimally affected during exposure to fibers, even in patients with asbestosis, while it is maximally affected by cigarette smoke. The swallowing of fibers subsequent to their escalation to the throat is postulated as the mechanism for the reported low-level increased risk to gastrointestinal cancer in asbestos workers. When the term "ingestion" is used in relation to occupational risk to gastrointestinal cancer, it is this passive form of ingestion that is meant. I will say nothing about penetration of asbestos through the wall of the gastrointestinal tract because the data are meager and are truly conflicting. The next step in the sequence of events depends on what happens to the retained fiber. One of two things may occur: 1. The short fibers, and to a certain degree the long fibers, are engulfed by pulmonary phagocytes or macrophages, the latter often fusing to engulf large fibers. These fibers then become coated with an iron/protein complex. On the basis of animal studies, coating is now believed to be an intracellular process and follows the engulfing of particles by macrophages to which they adhere. These coated fibers are what have traditionally been called "asbestos bodies"; now they are called "ferruginous bodies" because they are not necessarily limited to asbestos exposure and they take a positive iron stain due to the iron/protein complex coating the fiber. There is evidence to suggest that the coating of a fiber renders it nonfibrogenic. 2. A majority of the fibers, approximately 75 percent, will remain uncoated, which can facilitate effective penetration and retention of thin fibers, or the breakdown of thicker fibers into thinner fibrfls. Ttfese fibers tend to accumulate in the peripheral regions of the lower lobes, the site of early fibrosis (asbestosis). The fibers remain in situ (static) for long periods of time. Some may migrate nakedly tgr-ougTi the lymphatic channels, while others may follow the migration paths of the cells they have entered. There is no entirely satisfactory or universally accepted explanation for fibro- genesis. Suggested mechanisms have included (a) simple irritation, (b) leaching out of metal ions or silicic acid, and now (c) the immune mechanism. 138

C There are, however, cellular data that suggest a reasonable mechanism, and this mechanism assumes that fibrogenesis is evoked through the macrophage response. Such an explanation is attractive since: 1. It is compatible with the observation that long-thin fibers are the hazardous ones. 2. Macrophages tend to aggregate in the peribronchial area, site of the earliest fibrosis. 3. The cumulative effect of exposure is nicely explained by the repetitive and constant response of macrophages to asbestos exposure. The sequence of fibrosis and its relation to other asbestos-associated diseases is unknown except for the mechanical impairment of cardiopulmonary function by the scarring. Fibrosis produces interference with lung function through replacement of the air spaces (alveolar septa) with scar tissue and by restricting the normal excursion of the lining during breathing. Asbestos may affect anatomical sites in the following ways: 1. First and foremost, the gas exchange area or distal segments of the tracheobronchial-alveolar tree of the lung may be partially replaced by scar tissue, with resulting decreased lung function, x-ray changes, changes in physical findings, and blood gas changes. 2. The pleura (visceral and parietal) may thicken with the formation of plaques; pleural effusion may fill the chest cavity with fluid; or mesothelioma may spread and infiltrate all layers of the lung and chest wall. The peritoneum may also be affected, although how the fibers reach this site is unknown. 3. Lung cancer or bronchogenic cancer may result. The role of cigarette smoking and its impact on the mucociliary apparatus is a critical factor in the development of lung cancer. 4. Gastrointestinal cancer may occur through entry of fibers into the gastrointestinal tract by pharyngeal transpassage from the trachea. The development of cancer, or carcinogenesis, is a multistage process in which the chemical interaction between the carcinogenic agent and the DNA is a necessary but certainly an insufficient step in itself for the development of clinical cancer. The issue of dose-response and no-effect level cannot be pursued in appropriate depth here, but suffice it to say that a synthesis of experimental and epidemiological data clearly supports a no-effect level. The experience with asbestos has, very appropriately, given rise to concern that other fibers to which man is exposed may also represent a potential hazard to health. Organic fibers and manmade mineral fibers are in common use. I will limit my comments to manmade mineral fibers: 1. The dynamics of fiber entry, clearance, retention, and localization apply to manmade mineral fibers as they do to asbestos. 2. The concept of long-thin fibers as the source of potential hazard, as given for asbestos, also appears to be applicable to the chronic biological effect of manmade mineral fibers. 3. In relation to chemistry, however, manmade mineral fibers differ from asbestos. While chemistry may be dismissed in relation to asbestos, solubility, fiber integrity, fiber fracture, and fiber persistence in manmade mineral fibers are most logically related to the chemistry of manmade mineral fibers. For example, glass does not seem to split N 139 ~ w ~ ~ ~ w ~

vertically; rather it fractures horizontally. It Is soluble, and in some exposure studies it seems to have disappeared from predicted sites of localization. A natural fibrous material like gypsum disappears so rapidly that it cannot be detected even at the site of administration after a very short interval. These facts are well recognized. Lest one become overly sanguine as to the ease or speed with which critically necessary information about manmade mineral fibers can be obtained, it is sobering to reflect that despite our extensive knowledge of asbestos and asbestos-related disease, the following issues are still unresolved and subject to controversy: 1. Relation of fiber type to asbestos-associated disease. 2. ' The role of host factors (immunological state; peculiarities of respiratory tract architecture; concurrent or antecedent disease) in susceptibility to asbestos-related disease. 3. Progression of asbestos-related disease subsequent to cessation of exposure to asbestos and the specific etiological influence on cancer of the lung or gastrointestinal tract in the absence of asbestosis or other anatomic evidence of exposure to asbestos. I can best conclude by reiterating that there are special characteristics of asbestos that, though specific and not unique, to the best of our knowledge, invoke no mystique. The principles of asbestos-related disease are those of environmental biology, specifically toxicology and carcinogenesis. References [1] Timbrell, V., Aerodynamic Considerations and Other Aspects of Glass Fiber, in Occu ational .Ex o~sure to Fibrous Glass- Proceedin s of a Symposium, pp. 33-50 (HEW pu cat on (NIOSH) 76-T$1, wash ngton,_1976). [2] Morgan, A., Progress report to IOEH/QAMA, Montreal, Canada (1976). [3] Holt, P. F., Mills, J., and Young, D. K., Experimental asbestosis with four types of fibers: Importance of small particles, N.Y. Acad. Sci. 132, 87 (1965). [4] King, E. J., Clegg, J. W., and Rae, V. M., The effect of asbestos, and of asbestos and aluminum, on the lungs of rabbits, Thorax 1, 188 (1946). [5] Pott, F., Friedrichs, K.-H., and Huth, F., Results of animal experiments concerning the carcinogenic effect of fibrous dusts and their interpretation with regard to the carcinogenesis in humans, Zbl. Bakt. Hvg., I. Abt. Orig. B 162, 467-505 (1976). [6] Allison, A. C., Lysosomes and the toxicity of particulate pollutants, Arch. Internal Med. 128, 131-139 (1971). Discussion A. SUNDARAN: Dr. Kotin, I really enjoyed your talk. I would appreciate it if you could answer two simple questions that bother me. One, do you believe that fibrogenesis or fibrosis is an essential process that has to occur as a precarcinogenic lesion before you could find cancer? Two, do you think that fibers actually have to reach a parietal pleura before pleural mesothelio.a can occur, or do you think it can be an indirect outcome of other toxic efforts? P. KOTIN: Let .e answer your second question first. I would say that the occurrence of parietal sw3sothelioaa does not inevitably demand the presence of asbestos fibers. 140 I

A. WILEY: Could you state again the fiber sizes, length, and width that you felt were of biological importance? KOTIN: I will say it in microns; it took Dr. George Wright a year to get me to say micrometers. Fibers thicker than 3.5 p in diameter and longer than 200 are nonpathogenic, and that is an arbitrary number. The only reason I say 200 is because that is the maximum length of fibers that have been detected in lungs. Up to 200 p and thinner than 3.5 {i is the critical size range. If the diameter is thicker than 3.5 p length is irrelevant, because the fiber is not going to get down to the lower airways and air sacs. WILEY: Question was inaudible. KOTIN: What she is saying Is, I am not convinced that 3 to 1 is necessarily the right ratio. I agree. While 3 to 1 is a very handy rubric, there is nothing sanctified about it. 142

CS National Bureau of Standards Special Publication 506. Proceedings of the Workshop on Asbestos: Definitions and Measurement Methods held at NBS, Gaithersburg, MD, July 18-20, 1977. (Issued November 1978) THE CARCINOGENICITY OF FIBROUS MINERALS Mearl F. Stanton Laboratory of Pathology and Maxwell Layard Biometry Branch National Cancer Institute National Institutes of Health Public Health Service Department of Health, Education, and Welfare Bethesda, Maryland 20014 Abstract The carcinogenicities of 37 different dimensional distributions of seven different durable fibrous materials were correlated with fiber dimension. Optimum correlation was attained with fibers that measured <0.25 pm x >8 pm. Morphologic studies suggested that fibers in this dimensional range lie free in interstitial tissues, while fibers of smaller dimension are readily phagocytosed and fibers of larger dimension are sequestered by adherent phagocytes and fused phagocytic giant cells. Fibers that are fine and long may be more carcinogenic than others, simply because they are uncompromised by phagocytic activity. Keywords: Aluminum oxide; asbestos; carcinogenicity; Dawsonite; fibers; fibrous glass; phagocytosis; potassium octatitanate. For the past several years we have been interested in the question of how asbestos causes cancer once a fiber reaches susceptible tissues. We have approached t;~s problem with the simple device of introducing various types of particles into the pleural space of rats and observing the resultant tumors during the subsequent two years. The methods that we have used can be summarized briefly [1,2,37.1 A standard 40 mg dose of particles is applied by open thoracotomy directly to the left pleural surface of young female Osborne-Mendel rats. In each experiment, 30 to 50 rats are followed for two years and those surviving at two years are killed. All rats are necropsied and all pathological lesions examined histologically. Tumors that resemble the mesenchymal mesotheliomas of man generally develop after the first year. For the sake of precision we have called these tumors pleural sarcomas. During the second year, rats die at various times with and without pleural sarcomas; consequently, we have used actuarial computation to arrive at a valid estimate of the incidence of pleural sarcomas which takes into account differences in life-span [4]. Probability of pleural sarcoma has ranged from 0 to 100 percent depending on the materials used. Pleural sarcomas have not been observed in several thousand untreated controls; however, pleural sarcomas have occurred in rats treated only by simple thoracotomy. Our best estimate of these non-specific, background pleural sarcomas in treated controls is In the range of 2-4 percent. This is important to keep in mind since it makes interpretation of low level response unreliable with small numbers of animals. 'Figures in brackets indicate the literature references at the end of this paper. 143

There are two separate features of asbestos particles that merit consideration as potentially carcinogenic. First, the chemical nature of their constituents and contami- nents, especially those with a known potential for carcinogenicity such as the polycyclic hydrocarbons and heavy metals. Secondly, the physical structure of asbestos particles which in their fibrous fineness are somewhat unique in the natural world. It is our contention that it is the latter property, namely the simple quality of being an excep- tionally fine, long, durable fiber, that is most critical to carcinogenicity. The sup- porting evidence for this hypothesis is derived primarily from the type of experiments described above, as carried out by us and others [1,2,3,5,6,7]. It can be summarized briefly as follows: (1) Vigorous extraction of natural and contaminating hydrocarbons from asbestos does not alter its carcinogenicity. (2) Hand-cobbed, hand-milled asbestos that is free of metallic mill contamination is no less carcinogenic than machine-milled asbestos. (3) Naturally occurring or contaminating carcinogenic metals such as nickel, cobalt, chromium, iron, magnesium, and silica, or hydrocarbons such as benzo(a)pyrene, of comparable quantity to that in asbestos, when attached to inert non-fibrous particles of a size comparable to asbestos, do not show the carcinogenicity of asbestos. (4) The carcinogenicity of asbestos is greatly reduced if implanted as whole unseparated sheets of fibers or implanted as very short submicroscopic fibrils. (5) The carcinogenicity of asbestos is greatly reduced if asbestos is heated sufficiently to increase its fragility or if pulverized to non-fibrous particles. (6) Finally, between the various types of asbestos, particularly crocidolite, amosite, tremolite, anthophyllite, and chrysotile, there are wide variations in chemical, crystalline, and molecular structure. Nevertheless, when similar dimensional distri- butions of these asbestoses are applied directly to the pleura their carcinogenic response is similar_ If the carcinogenicity of asbestos depends on its dimensional configuration, two corollary hypotheses are suggested. First, durable fibers of other materials if in the same dimensional range as asbestos should be as carcinogenic as asbestos. Secondly, there should be an optimal dimensional range of fibers relevant to carcinogenicity. The data which I would like to present today relates to these two corollaries. Table 1 lists 37 experiments with seven different durable fibrous materials, each of differing dimensional distributions, but at or near the size distribution of asbestos. We have listed these in order of their probability of inducing pleural sarcomas, and as you can see the range runs the gamut from 0 to 100 percent. Asbestos fibers of a standard size characterized by a working group of the Unio Internationale contra Cancer would fall in the range of 65-80 percent [8]. The problem that follows is that of determining the dimensional distribution of the particles in each sample. To do this we used the straightforward method of measuring length and diameter from montage photographs of typical samples of the particles implanted. A minimum of 1000 particles were tabulated at magnification of 3000X to 29,000X. Subse- quently, in samples containing large particles, magnifications of 1000X were used to tabulate fibers inadequately represented on electron micrograph grids. The proper ratio of microscopic to submicroscopic particles yielded a representative sum of measured particles which was then entered into an IBM System 370 computer. From the density and the sum of the calculated volumes, the weight of the counted samples could be obtained and the distribution of particles per microgram of the sample estimated. For convenience, the numbers of particles per microgram were grouped into 34 dimensional ranges as indicated in Table 2. Table 2 illustrates the tabulation of six of the experiments with different samples of glass. By simple inspection the six examples show an apparent relationship between tumor probability and particle distribution that has held for all of the fibers tested thus far. The examples suggest that particles in relatively thin (diameter <0.25 pm) and long (length >8 }rm) dimensional categories are associated with the higher tumor probabilities. 144

Y l\ i [W \r! Table 1. Cumulative list of experiments arranged by percent probability of pleural sarcoma. Percent 1. DIHYDROXY SODIUM ALUMINUM CARBONATE V. . . . . . . . . . 100 2. POTASSIUM OCTATITANATE I . . . . . . . . . . . . . • . . 100 3. POTASSIUM OCTATITANATE II . . . . . . . . . . . . . . . . 100 4. SILICON CARBIDE GTC #1 . . . . . . . . . . . . . . . . . 100 5. DIHYDROXY SODIUM ALUMINUM CARBONATE I. . . . . . . . . . 95 6. BOROSILICATE GLASS (MOL) . . . . . . . . . . . . . . . . 85 7. BOROSILICATE GLASS (M6D) . . . . . . . . . . . . . . . . 77 8. BOROSILICATE GLASS + BINDER (KL) . . . . . . . . . . . . 74 9. BOROSILICATE GLASS (M6L) . . . . . . . . . . . . . . . . 72 10. ALUMINUM OXIDE -HC . . . . . . . . . . . . . . . . . . . 70 11. BOROSILICATE GLASS + BINDER (KW) . . . . . . . . . . . . 69 12. DIHYDROXY SODIUM ALUMINUM CARBONATE VII. ........ 68 13. DIHYDROXY SODIUM ALUMINUM CARBONATE IV .....'. ... 66 14. DIHYDROXY SODIUM ALUMINUM CARBONATE III. ........ 66 15. BOROSILICATE GLASS (M6W) . . . . . . . . . . . . . . . . 64 16. ALUMINUM OXIDE M3 . . . . . . . . . . . . . . . . . . . . 44 17. ALUMINUM OXIDE #4a . . . . . . . . . . . . . . . . . . . 41 18. ALUMINUM NITRIDE + OXIDE #6a . . . . . . . . . . . . . . 28 19. ALUMINUM OXIDE k2 . . . . . . . . . . . . . . . . . . . . 22 20. BOROSILICATE GLASS + BINDER (KCP). . . . . . . . . . . . 21 21. BOROSILICATE GLASS - BINDER (KUP). . . . . . . . . . . . 19 22. BOROSILICATE GLASS (MBL) . . . . . . . . . . . . . . 14 23. ALUMINUM OXIDE N4 . . . . . . . . . . . . . . . . . . . . 13 24. DIHYDROXY SODIUM ALUMINUM CARBONATE VI . . . . . . . . . 13 25. DIHYDROXY SODIUM ALUMINUM CARBONATE II . . . . . . . . . 12 26. BOROSILICATE GLASS (MOS) . . . . . . . . . . . . . . . . 8 27. BOROSILICATE GLASS + BINDER (K2P). . . . . . . . . . . . 8 28. MINERAL WOOL (Hi-Ca, Mg)(02P) . . . . . . . . . . . . . . 7 29. BOROSILICATE GLASS + BINDER (KFP). . . . . . . . . . . . 6 30. BOROSILICATE GLASS + BINDER (Y2P). . . . . . . . . . . . 6 31. HIGH CA-NA (P2P) . . . . . . . . . . . . . . . . . . . . 6 32. BOROSILICATE GLASS (M8S) . . . . . . . . . . . . . . . . 5 33. ALUMINUM OXIDE M5 . . . . . . . . . . . . . . . . . . . . 5 34. ALUMINUM OXIDE-LC (non-fibrous) . . . . . . . . . . . . . 3 35. BOROSILICATE GLASS (YW) (vehicle) (n=270). . . . . . . . 2 36. BOROSILICATE GLASS (M6S) . . . . . . . . . . . . . . . . 0 37. NI CKEL TITANATE . . . . . . . . . . . . . . . . . . . . . 0 145 N 0 a w r+ ~ ~ a. w

4 C Table 2. Six example experiments illustrating fiber distribution into 34 dimensional categories by comnon log of the number of particles per microgram in each size category. MOL KI 85.3% 73.9% >0.8 >4.0-8.0 0.67 >2.5-4.0 0.67 1.52 0.97 >1.5-2.5 1.45 0.67 2.03 2.19 >.50-1.5 2.23 2.53 3.23 3.08 2.95 2.40 3.42 2.74 >.25-.50 3.08 3.95 2.55 3.16 3.63 >.10-.25 2.93 3.35 4.53 3.03 3.16 3.33 3.03 >.05-.10 3.93 4.79 2.85 4.09 3.76 3.25 .01-.05 3.46 4.65 3.03 3.73 3.03 3.03 CP M K BL 21.5% 14.3% >8.0 . 0.97 >4.0-8.0 1.81 2.01 1.81 0.17 >2.5-4.0 1.44 2.05 1.81 0.77 1.73 2.02 >1.5-2.5 2.05 2.17 2.31 0.97 1.49 1.12 2.11 2.27 >.50-1.5 3.00 3.59 3.17 2.85 2.01 2.60 1.45 2.35 2.42 >.25-.50 3.24 3.38 3.10 2.55 1.85 2.38 >.10-.25 3.24 3.28 3.10 2.50 1.85 >.05.-.10 2.50 2.90 1.85 .01-.05 2.20 2.98 2.67 MOS YW 8.3% 2.8% >8.0 0.89 0.34 >4.0-8.0 2.46 2.72 0.92 0.40 >2.5-4.0 2.97 2.99 1.76 0.80 0.30 >1.5-2.5 3.43 2.37 1.17 T.00 0.11 >.50-1.5 3.88 3.91 2.76 2.46 T.10 0.41 >.25-.50 4.34 4.02 3.69 3.37 >.10-.25 4.43 3.88 >.05-.10 5.90 4.19 .01-.05 6.77 4.63 .01-1 >1-4 >4.8 >8-64 >64 .01-1 >1.4 >4.8 >8-64 >64 Length um Diameter um 146

<0.25 E,m) and long (length >8 pm) dimensional categories are associated with the higher tumor probabilities. Statistical regression techniques afford a method of analysis that can use a variety of explanatory variables to determine the best correlations between tumor probability and size distribution. The logit transformation [9] was applied to the estimated tumor probabilities (p) according to the formula: logit - log W-5 Then, linear regression methods which find the best fitting function of the form logit = a + blxl+....bkxk were used to compare the common logarithm of the number of particles per microgram in various size categories to the probability of pleural sarcoma. After analyzing various dimensional ranges that might have narrowed the optimum tumor inducing size range, it was determined that the best fit was with the dimensions <0.25 pm x >8 pm. The estimated regression equation was: logit = log (Q, _ -1.31 + .424x with a correlation coefficient of 0.9. The regression curve for this dimensional range is illustrated in figure 1. Figure 1 also indicates clearly that none of the seven different types of fibers show consistently greater deviations from the curve than any othbr, and that the curve's steepest slope is between 3-4 log particles per microgram. There was no correlation with particles less than 8 pm in length, but relatively good correlatiohs were also noted with numbers of fibers >8 pm in length and up to 1.5 vm in diameter (correlation coefficient 0.52 to 0.74). Figure 2 illustrates the 34 parameters used for carcinogenicity correlation and those categories in which relatively good correlation was obtained. It should be remembered that absence of correlation does not preclude a low level of tumor response outside these ranges. Histologic observations suggest the reason for the difference in response to fine, long fibers and those fibers that are either very short or very thick. The lesions in those experiments with a low probability of pleural sarcoma were highly cellular, being composed primarily of fibroblast-laden vascular granulation tissue with a relatively low collagen content and an abundance of macrophages. In lesions from low tumor probability groups in which virtually all fibers were less than 10 pm in length, the fibers seemed completely contained within macrophages. On the other hand, in those lesions from low tumor probability groups in which the fibers were virtually all of large diameter, the fibers seemed sequestered from adjacent tissue by both macrophages and multinucleated giant cells that closely invested the fiber surface. In contrast, the high tumor proba- bility lesions were relatively acellular, with an abundance of collagen at the site of implantation and the fine long fibers lay free in the interstitial tissues unaffected by phagocytes. In ancillary experiments with non-fibrous particles that stimulate collagen such as talc, silica, or carrageenin, collagen deposition equal to that of the high tumor probability lesion has been observed without the subsequent development of tumors. There- fore it seems evident that collagen itself is oot the critical factor in carcinogenesis. However, the fact that the fine, long fibers were unaffected by phagocytic activity in the high tumor probability groups suggests that fibers that are fine and long may be more carcinogenic than others simply because they are uncompromised by phagocytic activity. N ~ W 147 0 A b a ue

C~b 0 0 1 2 4 8 6 LOG NUMBER PARTICUES /MICROGRAM . ~ SIC • ~ GLASS • . AL2o2 o ~ DAWSONITE o - POTASSIUM OCTATfTANATE  - NICKEL TfTANATE PARTICLE SIZE, -5 026 {m x >8 Nn CORRELATION . 0$ COEFFICIENT Figure 1. Regression curve relating tumor probability to the comnon logarithm of the number of particles per microgram with diameters <0.25 um and lengths >8 pm. 148

~ »r~ Figure 2. Graphic display of the 34 size categories in scale with phagocytic and mesothelial cells. The starred fibers are those that correlate with pleural sarcoma probability. 149

cd References [1] Stanton, M. F., Layard, M., Tegeris, A., Miller, E., May, M., and Kent, E., Carcinogenicity of fibrous glass: Pleural response in the rat in relation to fiber dimension, J. Natl. Cancer Inst. 58, 587-603 (1977). [2] Stanton, M. F. and Wrench, C., Mechanisms of mesothelioma induction with asbestos and fibrous glass, J. Natl. Cancer Inst. 48, 797-821 (1972). [3] Stanton, M. F., Some etiological considerations of fibre carcinogenesis in Biolo ical Effects of Asbestos (Bogovski, P., Timbrell, V., Gilson, J., et al. , e3s. . yon, rance, t~orjT-FleaTth Organization, International Agency for Research on Cancer, Publication No. 8, 1973, pp. 289-294. [4] Armitage, P., Statistical Methods in Medical Research, New York, John Wiley & Sons, 1971, pp. 376-377, 410-414. [5] Wagner, J. C., The pathogenesis of tumors following the intra pleural injection of asbestos and silica in Mor of Ex er_ _ Res ratory Carcina enesis (Nettesheim, P., Hanna~, M.-G-De-atherage, J W, eds.) Oak Ri ge ationa Laboratory, Oak Ridge, Tennessee, Atomic Energy Commission Symposium Series No. 21, 1970, pp. 347-358. [6] Smith, W. E., Miller, L., and Elasser, R. E., Tests for carcinogenicity of asbestos, Ann. N.Y. Acad. Sci. 132, 456-488 (1965). [7] Timbrell, V., Physical factors as etiological mechanisms in Biolo ical Effects of Asbestos (Bogovski, P., Timbrell, V., Gilson, J. C., et aT-, e s. yon, i•rance, or dealth Organization, International Agency for Research on Cancer, Publication No. 8, 1973, pp. 295-303. [8] Timbrell, V., Gilson, J. C., Webster, I., UICC standard reference samples of asbestos, Int. J. Cancer 3, 406-408 (1968). [9] Cox, 0., Regression models and life tables, J. R. Stat. Soc. r6 34, 187-220 (1972). Discussion A. SUNDARAM: These pleural sarcomas, are they localized? If you leave them for a duration of time do they metastasize? M. STANTON: Yes, this is real cancer, but they do not metastasize early. A. LANGER: You allow your animals to live only two years, whereas Wagner allows his rats to live three years. Have you had any control groups run for the duration of the animals' lives? In those animals which you are reporting here, you are reporting frank malignancy? How many of the other animals had hyperplastic lesions? STANTON: It is difficult to detect precancerous lesions in mesenchymal tissues, so that It really is difficult to say what might be precancerous. We find that after two years plus the twenty weeks the animal has aged by the time we treat them, most rats are in their terminal stages. These rats do not have normal lung capacity and so they do not survive as long as the untreated rat does. By the end of the two years there is only a small percentage of the rats left. P. KOTIN: This business of assuming that hyperplasia carries with it as a corollary, or is even indicative of subsequent malignant transformation or neoplasia, particularly in mesenchymal tissue, has no substance at all. Now I am speaking as a pathologist. The other thing is, there is a tendency to denigrate experiments which are terminated for the reasons you said, when in fact the confounding findings which arise in the last six to eight months of a rat or any experimental animals, are such as to muddy up the results. I'd like your comment on this. The elegance of the observation occurs before you get into 150

CS the agonal state, where the exposure of the animal probably has little as anything to do with what he ultimately gets and dies from. STANTON: No further comment on that. J. WARREN: Our firm recently completed a report for OSHA, "The Economic and Infla- tionary Impact Study for the Effects of a Proposed Standard for Asbestos in Construction," and in the process of doing this report for OSHA we had to talk to a lot of your firms, universities - everybody from environmentalists to producers, maybe not some of you in here personally. This type of meeting is needed. We need people not just talking to each other, but with each other, and I think you have seen this today. You got one group over here saying, "Lookout! we are going to put out of business x number of people." Another group says you have to protect the worker - the worker comes first even at zero exposure. The only way we are going to resolve this problem, and it is a very sticky problem, is by everybody talking together. So that is just a comment; we found there is not enough talking to each other; even if you don't agree you can still talk. Has anyone looked at the possibility of using experimental animals other than rats, say a primate? Would this change results or give better data if we had an animal that would live longer? Has anyone used a rat that has been exposed to cigarette smoke at the same time? STANTON: Yes, other species have been used. Dr. Bi11 Smith is here today who has been using hamsters for many years and has some very elegant data with hamsters. We ourselves have used mice, and have been successful with mice as well as rats. I don't think that unless there was an exotic species that we would particularly contribute a great deal more by using another species. Chickens have been used and various other types of birds with some interesting results.

National Bureau of Standards Special Publication 506. Proceedings of the Workshop on Asbestos: Definitions and Measurement Methods held at NBS, Gaithersburg, MD, July 18-20, 1977. (Issued November 1978) NIEHS ORAL ASBESTOS STUDIES John A. Moore National Institute of Environmental Health Sciences P. 0. Box 12233 Research Triangle Park, North Carolina 27701 Abstract Epidemiologic data clearly associate inhalation of asbestos with an increased incidence of cancer. In addition to pulmonary and thoracic neoplasia, there are data which associate an increased incidence of gastrointestinal and peritoneal tumors. Controversy exists as to whether these latter types of neoplasia result from asbestos fibers that were ingested subsequent to clearance from the respiratory system. Exposure to ingested asbestos does occur in the general population through the presence of fibers in water and food. The NIEHS oral asbestos studies in rats and hamsters represent a systematic attempt to assess the biological effects associated with primary ingestion of selected asbestos fibers. The objectives of the studies include: assessment of biological (carcinogenic) effects as a consequence of exposure to one of several types of asbestos; assess if an interaction may exist between a chemical carcinogen which is known to produce bowel cancer, and ingestion of asbestos. The specific experimental design of this series of ongoing studies will be presented. Key Words: Asbestos; bowel cancer; cancer; epidemiology; fibers. There is strong evidence that associates occupational exposure to chrysotile amosite, and crocidolite to a resulting high incidence of lung cancer. Exposure to these forms of asbestos has also been observed to result in an increased incidence of pleural and peri- toneal mesothelioma and an excess risk of gastrointestinal cancer. Environmental exposure to asbestos through living in the neighborhood of asbestos factories or mines or through residing in households of asbestos workers also correlates with increased mesotheliomas [1].1 It is plausible to speculate that the increased incidence of gastrointestinal cancer in occupationally exposed populations may be a consequence of asbestos fiber ingestion. Fiber ingestion in these circumstances may result through the swallowing of fibers cleared from the nasal or tracheobronchial tree. Direct ingestion of fibers deposited in the oral cavity also occurs. Exposures of the general population to asbestos occurs through ingestion of materials and substances that contain fibers. For example, several million fibers per liter were found in Canadian tap water [2]; Great Lakes and St. Lawrence River water showed average concentrations of about 1.7 million asbestos fibers per liter [3]; water collected from the north shore of Lake Superior in the Silver Bay/Duluth region were found to have even higher fiber levels. A number of studies have reported the appearance of asbestos fibers in commercial beverages such as beer, vermouth, and soft drinks [2]. The fibers found in these products may be a result of the use of asbestos filters used in their preparation [2]. Food may contain asbestos through the use of asbestos filters or the use of talc, which has an asbestos impurity [4,5]. 1Figures in brackets indicate the literature references at the end of this paper. Preceding page Ylank 153 2063104950

In response to a growing concern about the possible biological effects of ingested asbestos, a conference was held in 1973, co-sponsored by the National Institute of Environmental Health =5ciences and the Environmental Protection Agency. The meeting confirmed that the preponderance of biological data concerning exposure to asbestos focused on the inhalation and not the ingestion route of exposure. A consensus of that interna- tional conference was that research was needed on health effects associated with asbestos ingestion. A Subcommittee of the DHEW Committee to Coordinate Toxicology and Related Programs (CCTRP) subsequently reviewed the existing data, recommended that additional research be undertaken, and prepared a draft research protocol that it felt would be responsive to the scientific needs. This protocol was widely distributed for comments both within and outside the Government. Based on the comments received, a final protocol was developed and submitted as part of its final report. In response to the Subcommittee's report, Congress appropriated specific funds directing the National Institute of Environ- mental Health Sciences to research the effects of oral asbestos ingestion. The NIEHS is conducting this research primarily through its research contracts program. The Environ- mental Protection Agency also contributed funds for these studies. The design of these studies is in concert with the recommendations of the CCTRP Subcommittee. The basic design of the studies provides for an evaluation of chrysotile, a serpentine asbestos; and amosite and crocidolite, fibers representative of amphibole asbestos; plus a non-fibrous tremolite, which does contain low levels of asbestiform fibers. The studies call for asbestos to be fed continuously in the diet over the entire lifespan of the test animal. Each form of asbestos is contained at a one-percent level in a pelleted rodent diet of constant ingredient formulation (NIH Feed 31). The proposal to incorporate the asbestos within a pelleted diet form was approved only after studies indicated that the pelleting process did not alter the physical integrity of the fiber. The utilization of an asbestos diet in a pelleted form has obvious advantages: it minimizes fiber aerosols which would occur with greater ease in a non-pelleted form; it minimizes variations of asbestos concentration in the diet due to segregation of fibers that would occur during shipping, handling, and feeding. Incorporating asbestos into food rather than water eliminates settling and subsequent uneven distribution. All materials are being fed to the F-344 strain of rat; whereas two forms of asbestos, chrysotile and amosite, are also to be tested in hamsters. Golden Syrian hamsters represent a second test species and are being fed a serpentine or amphibole form of asbestos. All studies encompass the lifespan of the animal, which is defined as the age at which the animal begins eating solid food until its death. To insure asbestos ingestion at a young age, these studies are initiated by feeding the asbestos diet to a nursing mother, which is removed once the pups are weaned. These latter animals that begin eating asbestos at two weeks of age constitute the test generation. In the basic studies, the test group size is 500, composed of equal numbers of males and females. In each of the rat and hamster studies, there is a composite total of 1000 animals that receive diet which does not contain asbestos and serve as controls. The experimental group size allows one to detect a statistically significant increase in gastrointestinal tumors in the treated groups at a two percent increase above the control population. In another rat experiment, two subsets of 200 animals each are to receive asbestos from the first to the 28th day of life by gastric intubation. The rat pups received 2.35 mg of an aqueous asbestos suspension daily. At weaning, the rats are placed on the appropriate asbestos diet for the remainder of their lifespan. One subset of 200 animals is to receive chrysotile while another subset is to receive amosite. The objective of these experiments is to see if a possibility exists that neonates may be a special risk population. There is also scientific interest in determining if asbestos in the diet alters the expression of intestinal neoplasms induced by a known chemical carcinogen. Studies of this type are performed in rats that are fed either chrysotile or amosite. A similar study will be conducted in hamsters receiving the chrysotile diet. There are 350 animals in each of these three groups. The chemical carcinogen to be utilized was selected after a series of dose-ranging experiments of one-year's duration was performed in each species. In these dose-ranging studies, both dimethylhydrazine and methylazoxymethanol were evaluated. The 154

results indicated that dimethylhydrazine was the chemical carcinogen of choice due to lower toxicity and greater specificity of intestinal tumor response. The dose selected is one that will produce approximately a 10 percent incidence of intestinal tumors. That dose for hams-ters is 4 mg/kg, whereas for rats it was 7.5 mg/kg and 15 mg/kg in male and female rats, respectively. The dimethylhydrazine is administered by gavage once every fourteen days until five doses have been administered. The initial dose was administered at six weeks of age. Tables 1 and 2 summarize the design of the animal study. The animal testing phase of the experiments commenced in late 1975. Since the natural lifespan of the F-344 rat is 26-30 months and 18-23 months for the hamster, definitive interpretation of these studies is several years away. Table 1. NIEHS oral asbestos study. Golden Hamster Chrysotile Intermediate Chrysotile Short Range Amosite Asbestos diet 500a 500 500 Asbestos diet plus dimethylhydrazinea 350 NO ND Control dietb 500 250 250 Control diet plus dimethylhydrazinea 250 ND ND a Number of animals (equal numbers of each sex). b Control allocations are descriptive only. Experimental response will be evaluated against total controls (1000). Subsets of control will reflect temporal differences in commencing phases of study which is expected to be aggregates of 250-350. ND - Not done. Studies conducted by Illinois Institute of Technology Research Institute, Chicago, Illinois. I

Table 2. NIEHS oral asbestos study. Chrysotile Intermediate F-344 Rat Chrysotile Short Range Amosite Intermediate Crocidolite Tremolite Asbestos diet 500a 500 500 500 500 Asbestos diet plus dimethylhydrazinea 350 ND 350 ND ND Preweaning asbestos gavage plus asbestos diet 200 ND 200 ND ND Control dieth 175 175 175 175 175 Control diet plus a dimethylhydrazine 250 ND 250 ND ND a Number of animals (equal numbers of each sex). b Control allocations are descriptive only. Experimental response will be evaluated against total controls (1000). Subsets of control will reflect temporal differences in commencing phases of study which is expected to be aggregates of 250-350. ND - Not done. Studies conducted by Hazleton Research Laboratories, Vienna, Virginia. All animals receive a thorough pathologic evaluation at time of autopsy. In con- formance with the NCI Carcinogen Bioassay protocol, some thirty tissues in addition to any gross lesions will be examined under light microscopy. The rat studies are being performed through a contract with Hazleton Research Laboratories, Vienna, Virginia; whereas the hamster experiments are being performed by the Illinois Institute of Technology Research Institute, Chicago, Illinois. As a biologist, I wish to emphatically state that the most difficult decision in the design of these studies was determining the types and specific forms of asbestos that were to be fed. The literature clearly indicated that some previous studies were flawed due to unwitting physical violence imposed upon the asbestos during its preparation. In some cases, there was concern about contamination by organic chemicals. In medical research circles, the issue still rages with respect to the size of fiber that may be associated with observed neoplastic response. It is necessary to relate size that produces optimal biological response to the distribution of fiber sizes to which there is general human population exposure. The common fiber found in municipal water supplies represents one of serpentine origin. From a numerical standpoint, the preponderance of these fibers is of the low micron and submicron lengths. To accommodate to these circumstances, it was decided that there would be two chrysotile asbestos materials used in the rat and hamster studies. These are referred to as the NIEHS short-range chrysotile and the NIEHS intermediate-range chrysotile. NIEHS short-range chrysotile was mined from the New Idria deposits in California. This chrysotile is of very small fiber length and diameter. It is a single lot produced by Union Carbide and is referenced by them as COF-25. The NIENS intermediate- range chrysotile originated from the Johns-Manville Jeffrey Mine in Canada. This material has general analogies to their Plastobest 20. 156

S5 One method of comparing these two chrysotile samples is by comparing surface area determinations. Table 3 presents the results of such tests; the UICC chrysotile surface area values are listed for comparison. The UICC samples have been the asbestos source for the majority of biological studies over the past several years. As can be seen from the table, the NIEHS intermediate-range chrysotile compares quite favorably with the UICC Canadian chrysotile. The •two-fold increase in surface area of the NIEHS short-range chrysotile compared to its intermediate-range counterpart reflects the much smaller fiber size found in this sample. Table 3. Comparison of UICC and NIEHS chrysotile samples. Asbestos Identification UICC Value (m2/g) IITRI Value (m2/g) UICC Rodesian Chrysotile 21.3 ± 1.5 22.35 UICC Canadian Chrysotile 26.8 ± 0.7 27.7 NIEHS Intermediate Range Chrysotile 27.8 ± 2.7 NIEHS Short Range Chrysotile 59.0 ± 6.2 The amphibole samples, amosite and crocidolite, were prepared by the Ontario Research Foundation under the direction of the U.S. Bureau of Mines' College Park Laboratory. This asbestos, purchased commercially, has been processed by air jet milling to better stan- dardize the range of fiber size contained in the material. The tremolite sample was mined and milled to -325 mesh by the R. T. Vanderbilt Company, Balmat, New York. It was subsequently blended by the U.S. Bureau of Mines personnel to insure homogeneity of the sample. All test materials are being extensively characterized as to.chemical and fiber size characteristics. The characterization data include x-ray diffraction parameters, chemical composition, DTA, TGA, optical constants, density, and surface area. These studies are being performed by the U.S. Bureau of Mines. Exhaustive electron microscopic characteriza- tion of each material as to fiber length, fiber diameter, surface area, distribution of fiber size, and selective pore volume measurements are being performed by the Fine Particles Laboratory of the Illinois Institute of Technology Research Institute, Chicago, Illinois. The characterization studies on tremolite and the short-range and intermediate-range chrysotile are nearly complete. The characterization of amosite and crocidolite are scheduled for completion by the end of the year. Two recent ingestion studies that have been reported within the past year yielded variable results. In a British study, a group of 32 Wistar rats were fed 100 mg per day of UICC Canadian chrysotile prepared in milk powder on a five-day-a-week schedule for a total of 100 days of ingestion. There were 16 control animals which were fed only the malted milk. The animals were then allowed to live out their lifetime, which was a mean survival of 619 days for those animals on chrysotile versus 641 days for the controls. One gastric leiomyosarcoma was observed in the chrysotile group. No tumors of this type were found to occur in the controls [6]. In a study reported in the East German literature, a statistically significant (p <0.01) increased incidence of malignant tumors occurred in rats that received asbestos filter material in the diet [7]. The exact composition of the asbestos filter material was not given in this paper. In this study, 25 male and 25 female Wistar rats were given 50 mg/kg body weight per day of asbestos filter material which contained approximately 52 percent chrysotile asbestos. This asbestos containing filter material had been previously 157 2063104954

powdered and added as a water suspension to the diet. In the group of animals which received the asbestos filter material, the average survival time was 441 days. Untreated controls had an average_survival time of 702 days. Of the 42 treated rats available for pathologic evaluation, i2 malignant tumors were found. This is to be compared to seven tumors (two liver cell carcinomas and five mammary fibroadenomas) observed in 49 control animals. The tumor types observed in the animals fed the asbestos filter material included four kidney carcinomas, one lung carcinoma, three reticulum cell sarcomas, and four liver cell carcinomas. Two mammary fibroadenomas, as well as a lung adenoma, two cholangiomas, and two forestomach papillomas were also observed. The NIEHS Oral Asbestos Studies should provide controlled data from large enough sample sizes to allow for initial formulation of basic principles as to the biological effects of exposure to ingested asbestos. References [1] IARC Monographs on the Evaluation of Carcinogenic Risk of Chemicals to Man, Asbestos, Volume 14. International Agency for Research on Cancer, Lyon, France (1977). [2] Cunningham, H. M. and Pontefract, R. D., Asbestos fibers in beverages and drinking water, Nature Lond. , 232, 332-333 (1971). [3] Cook, P. M., Glass, G. E., and Tucker, J. H., Asbestiform amphibole minerals: detec- tion and measurement of high concentrations in mu0icipal water supplies, Science, 185, 853-855 (1974). [4] Merliss, R. R., Talc and asbestos contaminant of rice, J. Amer. Med. Assoc., 216, 2144 (1971). [5] Wolff, A. H. and Oehme, F. W., Carcinogenic chemicals in food as an environmental health issue, J. Amer. Vet. Med. Assoc., 164, 623-629 (1974). [6] Wagner, J. C., Berry, G., Cooke, T. J., Hill, R. J., Pooley, F. D., and Skidmore, J. W., Animal experiments with talc, in Inhaled Particles and Va ours, IV, W. C. Walter, ed., New York, Pergamon (in press). [7] Gibel, W., Lohs,..K., Horn, K.-H., Wildner, G. P., and Hoffmann, F., Tierexperimentelle Untersuchungen Uber eine kanzerogene Wirkung von Asbestifiltermaterial nach oraler Aufnahme, Arch. Geschwulstforsch, 46, 437-442 (1976). Discussion . M. SCHNEIDERMAN: Dr. Moore completed his paper considerably earlier than the time allotted. Are there some questions concerning this particular elaborate set of experiments, and the experiment design? Are there some suggestions that people would have? When the results come in from these experiments, what kinds of doubts will exist in your mind? What sorts of things would you like to see answered that these are not going to answer? I hope that there are people here that have thought about these and might have some questions. V. WOLKADOFF: Evidently Dr. Moore has brought up chrysotile, the short size range sample from Edra and the intermediate size sample from Jeffry, and an evaluation of the size by specific surface area. Amosite was size fractionated by airjet milling, and tremolite, evidently by milling of some type, to minus 325 mesh. The chrysotile more or less has been characterized by specific surface area. Do you have information, within each of these categories, as to the crystallinity of the individual fibers, the four categories versus degradation of the crystallinity of individual fibers by the method of preparation, or is it too early to say? You mentioned data by x-ray diffraction, OTA, and optical microscopy. I also wanted to know if you are going to include the electron diffraction results in your studies. 158

S? J. MOORE: I think I mentioned the electron diffraction work as part of the study, and I'd rather let the Bureau of Mines personnel, who are here, or the IITRI personnel, answer your question with regard to the crystallinity. I'm sure it has been looked at, but I don't know if the data are in such a stage to make any comment about it. WOLKADOFF: What about air-jet milling of amosite, do you have any data now? MOORE: No, if I did, I would have presented it. WOLKADOFF: The tremolite data also, you don't have anything then? MOORE: No sir, it's not in complete form. As I mentioned in the paper, the characterization of the two chrysotile samples and the tremolite sample should be available within the next couple of months, and we would expect that the similar types of studies characterizing the amosite and the crocidolite will be done by the end of the year. WOLKADOFF: Thank you very much. W. CAMPBELL: All the data has been completed on tremolite, including optical micro- scopy, SEM, TEM, chemical data, and surface area. On chrysotile, the optical data is finished, the SEM data is about completed, and the TEM is about completed. So in answer to your question there are very extensive data available on the optical properties, the morphology, the crystallinity, the trace metals, and so forth. Surface area is just one of the many parameters being investigated. MOORE: I may have misled you in my presentation by only showing the slide on surface area; I did that one because it did show a distinction between the two chrysqtiles. I would point out that we do not wish to infer that there is a clear separation of fibers between these two materials. Certainly the intermediate range chrysotile sample does have fibers that are well into the size range of fibers that are found in the short range sample. The distinction between the two is the proportion of fibers that may exceed, with respect to length and diameter, those that were found only in the short range. G. WRIGHT: You quite properly pointed out that the kind of occupational exposure which has led to what we know about tumor incidence is quite different from what's found in water supplies. In fact, the differences are very striking. On the other hand, in occupational exposure generally, and I say would say almost without exception, the percent of the total fibers that exceeds eight to ten and even five pm in length is of the order of less than five percent, and in many situations is only one or two percent. In the animal experiments that have been done by inhalation of asbestos, in general, the clouds created contained only one or two percent of what we call long fibers. For this reason, I think that to look at your samples in terms of percent, inferring that one, two, or three percent of long fibers present in a sample is acceptable when you're talking about short fiber samples is erroneous. We need to get around to the number of long fibers, not the percent. Now also I would like to ask if these experiments are designed to look at the occupational experience or at the water experience? MOORE: I would hope that they would have relevance in both areas. WRIGHT: What percent or what number of long fibers are still present in your so- called short sample? MOORE: Well, at what length do you want to consider as a long fiber? WRIGHT: Anything over 5 Nm, because in water, you've said, it is under 5. MOORE: I recall the raw data that are available on that; about 90 percent of the samples in the short range chrysotile would be below that. WRIGHT: In other words, ten percent are still above 5 pm? MOORE: Right. 159 N ~ w ~. ~ e ~ a

C WRIGHT: Well, that's essentially what the occupational exposure is. I don't think you're looking at water related exposures. W. BANK: I'll change the subject slightly. There have been some animal nutrition studies going on since 1965 in Japan, and more recently in the U.S., in which fibrous material, namely certain zeolites, have been fed to these animals. The results were that the animals gained weight faster, certai,n diseases seemed to disappear, and so on. It's recognized, however, that there is a possible long-range pathological effect that might be Involved because of the fibrous materials. Have you heard or do you know of any such information? MOORE: I'm not aware of that work coming out of Japan. C. COOPER: I strongly support the observation made by George Wright that the 10 percent or even 5 percent of long fibers in your short fiber samples would leave serious doubts as to whether the results of these experiments would be applicable to water supplies. Another question that has bothered a number of people is whether or not consideration was given, in the design of this experiment, in the choice of samples, to actually including a sample of the material that has contributed a great deal to this whole controversy. That is, the amphiboles that are found in Lake Superior water, in the size distribution in which they were found. I can see the difficulties in doing this, but I wonder just what the course of reasoning was that led to this type of material not being included? MOORE: We were advised, and I must say we also subscribe to the opinion, that with regard to injestion studies we might, in an initial series of experiments, be better off by using materials that are known to have biological effects in test species. This is where we opted to go with the amosite, for example; it's probably the closest thing we have to being representative of a cummingtonite/grunerite which is the Lake Superior type of sample. The other problem that we had when we did discuss the possibility of using something from the water in that area, was the complete lack of agreement among people we talked to with regard to what actually should be the sample that would come out of that area. In addition there was the logistics of trying to get that sample; it was just that simple. I would also state that with regard to fiber size, maybe having too many fibers above 5 pm to persiit direct relevance or extrapolation to municipal water supply samples, as was pointed out by Dr. Wright and yourself, I assume you gentlemen would accept a negative. A. SUNDARAX: You quoted Dr. Gibel's paper from East Germany. Do you believe that study was well conducted, showing a significant effect of asbestos by oral ingestion? MOORE: All I can comment on is the information which was available in the reprint, which brings questions to mind which certainly aren't explained in the materials and methods. For example, how they prepared the material, actually what was the other 48 percent since they inferred that 52 percent of the material was chrysotile. I think this sample size data may have some problems as well. SUNDARAM: I agree to that, but in addition there is a significant point worth noting. The paper never mentions the number of animals affected over the control. There should be a significant increase in the number of animals that had tumors, not a significant increase in the number of tumors, because it may be one animal that had twelve tumors or it may be 12 animals that had 12 tumors. So It is the number of animals that were affected that's more important than the number of tumors. This paper has been quoted many times even though it just appeared in 1976. So many people quote it, and I thought it's better to point out this significant question that we should not miss. MOORE: I thank you for your point, because the paper does not indicate as to whether, for exa.ple, the 12 malignant tumors found in 42 treated animals came from 12 separate individuals, or whether it was one or more animals which may have had multiple tumors. N ~ W r 160 ~ N -7

CS 0. ALTON: I am really wondering whether a dosage of 100 milligrams per day per unit weight-of rat for the lifetime of a rat is comparable to the ordinary ingestion of asbestos fibers by man. Is there any relationship between those two figures for rat and man. MOORE: I don't remember quoting 100 milligrams per rat per day, but suffice it to say that the level of asbestos that's in a diet at the 1 percent level certainly is in the high range of exposure. COOPER: I have a very crass and practical question to ask. I know it's a matter of public record what a study of this magnitude costs, what kind of investment it involves, but I think the group would be interested in knowing just how much a major study like this represents in cost. MOORE: It's estimated that by the time the studies that I have outlined are completed, which will include the characterization of the materials as well, it will probably be somewhere around 3-4 million dollars. N 161 ~ ~ ~ ~ m

CS National Bureau of Standards Special Publication 506. Proceedings of the Workshop on Asbestos: Definitions and Measurement Methods held at NBS, Gaithersburg, MD, July 18-20, 1977. (Issued November 1978) EPA STUDY OF BIOLOGICAL EFFECTS OF ASBESTOS-LIKE MINERAL FIBERS D. L. Coffin EPA Health Effects Research Laboratory Research Triangle Park, North Carolina 27709 and L. D. Palekar Health Effects Research Laboratory Northrop Services, Inc. Environmental Sciences Group Research Triangle.Park, North Carolina 27709 Abstract A large amount of the earth's crust is composed of rock containing mineral fibers which resemble asbestos to varying degrees in their physical and chemical properties. Consequently, such materials are likely to be encountered inadvertently during the extraction of various ores, the extraction of rock for commercial purposes, and even from rock moving operations encountered during highway construction, and the like. Because the air and water may become contaminated by these fibers, it is of interest from the standpoint of environmental protection to know how the biological effect of such material compares with that of asbestos. Consequently, a study has been instituted by EPA to investigate the relative biological potency of such materials. The project is being approached on both in vivo and in vitro levels. The minerals being studied at the outset are fibrous ampriboles from a taconite mine, but it is the intent to broaden these studies as soon as possible. The animal studies are being conducted in pathogen-free rats by intratracheal instillation (with and without interacting organic carcinogens) and by intrapleural injections. The end points are tumor induction and other chronic diseases. Attention is also being given to early pathogenic sequences. The in vitro studies consist of red cell lysis, pulmonary macrophage systems, and~various biological and chemical studies connected with the influence of these agents on cell membranes and interaction with mutagens and carcinogens. The prime objective is to compare the biological effect of the minerals studied to the corresponding asbestos species to determine the comparative influence of such co-variables as fiber length, trace element content, surface area, zeta potential, and the like, on the biological outcome. Thus, the study will relate biological activity to mineralogical characterization so that generalization can be made on the basis of such factors. Key Words: Alveolar macrophages; hemolysis; intrapleural injections; intratracheal instillation; multinucleated giant cell; PMP I; PMP II; Polyp. Preceding page blank 163 L

The hazards for human health associated with the extraction and handling of various members of the commercial asbestos series are now well known. However, a new issue has recently come to the forefront of environmental toxicology concerning the possible health hazard from inhalation or ingestion of fibrous silicate minerals, not asbestos per se, that contaminate the air and water. Such silicate materials are ubiquitous in the earth's crust where amphibole-bearing rocks may serve as a potential source for a number of mineral species, for example, fibers from the cummingtonite-grunerite series, hornblende, etc. When the above-mentioned facts became known, there was a tendency to class all of these materials as "asbestos" and to try to make inferences concerning their potential health effects in man merely on the basis of supposed analogy to commercial asbestos. We know now, however, that there is an enormous variation in these materials; some closely resemble the corresponding asbestos, and others do not. It would be folly, therefore, to base the threat to human health solely on such a crude determinant. This is particularly true since, despite the great number of epidemiological and biological studies carried out with asbestos, much remains to be learned concerning the exact causal mechanisms of the various lesions attributed to such exposure. For instance, one cannot safely postulate a common etiological mechanism for the usual lesions of asbestos exposure such as pulmonary fibrosis, carcinoma of the lung, and mesothelioma, and the possible role of asbestos for tumors in other locations which at this time is largely unexplored. Because of these issues, the Environmental Protection Agency (EPA) has taken the initiative to study these matters to determine if a threat to health exists from non- asbestos minerals, and if it does, by means of its quantification, to determine how best to control it on the basis of health benefit versus cost. EPA is conducting a study of the relative pathogenic potential of such minerals compared to asbestos, silica, and other particulate substances of known toxicity. The prime purpose of these experiments is to relate biological effects to the physiochemical properties of the minerals. Beginning with the convening of an advisory committee, the following approach evolved, which includes mineralogical as well as biological studies. Mineralogical Studies Intensive study was made from 50 large rock specimens removed from a taconite mine. After preliminary lithological examinations, two of these were selected for employment in biological experiments, which are designated as PMP I and PMP II. Fibers were separated from the rock by such means as mechanical vibration, hand cobbing, air jet milling, spinning, and riffling. The final specimens were subjected to a detailed analysis by means of optical and electron microscopy, x-ray emission spectroscopy, and x-ray diffraction. Computations of surface area and determination of extractable organics were made. Comparisons were also made an the basis of the above parameters with UICC amosite (fibrous grunerite) and airborne material collected in the vicinity of the mine and the ore processing plant. On the basis of the above measurements, a decision was made to prepare a large amount of this material suitable for biological experimentation. Figures 1 through 8 and Tables I-III illustrate various mineralogical characteristics of the samples chosen from the mine for biological studies, as well as samples from the airborne material in the vicinity of the mine and ore-processing area. Figures 1 and 2 represent electron micrographs of air samples from mine and processing areas respectively. The chemical analysis of air samples revealed that in addition to magnetite and quartz particles there were predominantly two other types of minerals in both areas. The electron microscope x-ray analysis revealed the presence of Mg, Si, Ca, Mn, and Fe in one sample (fig. 3), whereas the second sample contained only Mg, Si, and Fe (fig. 4). Data from a careful analysis of size distribution of the air samples are presented in Table 1, showing two samples from each of the processing and mine areas. The majority of the particles in both areas were found to be less than 5 pm in length and less than 1 pm in diameter. A small percentage of particles were between 5 and 10 pm in length, with varying diameters. Air samples from the processing areas contained 66 to 70 percent fibers with diameters less than 0.5 pm as compared to 52 to 55 percent in the mine area. This may suggest that further fibrillation of the rock occurs during the processing. 164

. Figure 1. Air sample from mine area showing long and straight fibers (10,000x). Figure 2. Air sample from the area of processing plants also showing long and straight fibers (10,000x). z /z ® ® En 00170 Ca Q17O3! 165 Figure 3. Electron microscope x-ray spectra of air sample indicating the presence of Mg, Si, Ca, Mn, and Fe.

Figure 4. Electron microscope x-ray spectra of air sample indicating the presence of Mg. Si, and Fe. Table 1. Sumnary data of size distribution of mineral fibers in ambient air samples. - - - - - - - - - Lengths by Percent Number in Microns - - - - - - - - - <1 1-5 5-10 >10 Total Air Sample No. 1 <0.50 9 71 5 0 0.51-1.00 0 12 2 0 >1.00 1 0 0 0 Total 10 83 7 0 100 -93% 7% + - Air Sample No. 2 <0.50 8 66 2 1 0.51-1.00 0 13 1 0 >1.00 0 2 6 1 Total 8 81 9 2 100 89% . 11% 1 f . Air Sample No. 3 <0.50 5 55 2 0 0.51-1.00 0 21 4 0 >1.00 0 4 8 1 Total 5 80 14 1 100 85% . 15% Air Sample No. 4 <0.50 9 52 5 0 0.51-1.00 0 21 3 0 >1.00 0 3 5 2 Total 9 76 13 2 100 85% 15% Below 5 um Diameter by Percent Number in Microns 166 Above 5 Wn

CS Figure 5. Electron micrograph of PMP I showing long and straight fibers with acicular particles (1000x). Figure 6. Electron micrograph ~of PMP II indicating long and straight fibers and particles (1000x). The electron microscope x-ray emission spectra of the fibers collected from the two rock samples revealed the presence of Mg, Si, Ca, Mn, and Fe on PMP I (fig. 7); and Mg, Si, and Fe on PMP II (fig. 8). The size distribution of the samples is given in Tables 2 and 3. The data indicate that the majority of the fibers are less than 5 pm in length and less than 0.5 pm in diameters in both samples. Figure 7. Electron microscope x-ray spectra of PMP I showing the presence of Mg, Si, Ca, Mn, and Fe. N Q a 167 ~ g ~ w

Figure 8. Electron microscope x-ray spectra of PMP II showing the presence of Mg. Si, and Fe. Table 2. Size distribution of PMP I sample. - - - - - - - - - - - - - Lengths in Microns (lun) ------------- 0.00 - 0.50 0.51 - 1.00' 1.01 - 5.00 5.01 - 10.00 10.01 - 25.00 Total 0.00 - 0.50 1.47 8.09 68.38 2.94 0.73 81.61 0.51 - 1.00 0.00 0.00 5.88 2.94 0.00 8.82 1.01 - 2.00 0.00 0.00 4.41 0.73 0.73 5.87 2.01 - 5.00 0.00 0.00 0.00 0.00 2.94 2.94 5.01 - 10.00 0.00 0.00 0.00 0.00 0.73 0.73 88.23 --• 11.74 99.97 Below 5 um Above 5 ym Diameter by Percent Number in Microns Table 3. Size distribution of PMP-2 sample. Lengths by Percent Number in Microns <1 1- 5 5- 10 10 - 15 0.00 - 0.50 27.06 41.53 0 0 0.51 - 1.00 0 18.01 5.50 1.80 >1.00 - 10 0 0.80 3.90 1.80 27.06 60.34 9.40 3.60 .-- 87% -. «-~ 13% -+ Be1ow 5 um Above 5 1¢a Diameter by Percent Number in Microns 168

Figure 9. Fibrous grunerite (UICC amosite) showing the general shape of the particle which is long and straight (1000x). Sim-e-the air samples and the rock samples seem to be representative of the grunerite Tamiiy, a fibrous grunerite, namely UICC standard reference amosite, with known biological properties, was selected as a possible control for the studies, and characterized. The electron microscope x-ray analysis of amosite indicates the presence of Mg, Si, and Fe (fig. 10). Size distribution data for this material are presented in Table 4. Eighty- seven percent of the fibers were found to be less than 5 Nm in length and 1.5 Nm in diameter. I V Figure 10. Electron microscope x-ray spectra of fibrous grunerite (UICC amosite) indicating the presence of Mg, 51, and Fe. En Qi.IU Ca =7Q39 Table 4. Size distribution data of UICC amosite by IITRI method. - - - - - - - Lengths Distribution (by percent number), in Microns - - - - - - - 0.2-0.5 0.5-1 1-2 2-5 5-10 10-25 25-50 50-100 100-200 Total 0.00-1.10 15.90 3.48 1.64 1.80 0.57 0.20 -- -- -- 23.39 0.10-0.40 8.69 13.49 18.24 16.40 5.16 1.68 0.41 0.20 0.01 64.28 0.40-1.50 -- -- 2.54 4.75 1.31 1.84 2.69 0.20 -- 12.93 87% 7% Diameter by Percent Number in Microns 169 ~ 6%

The air samples, ttie fibers obtained from rocks, and amosite fibers were examined by electron microscope for their general shape. All samples contained straight and long fibers and acicular particles (figs. 5, 6, 9). These photographs are not representative of the size distribution. Biological Studies Toxicity evaluations are proceeding both in vivo and in vitro. Whole animal experi- ments are being carried out to determine the comparative ef e~ ct of the above-mentioned mineral fibers in inducing lesions such as pulmonary fibrosis, lung cancer, and pleural mesothelioma. Basically, a comparison between a test amphibole of the cummingtonite- grunerite family, UICC amosite, and an inert particle is intended. These studies are being conducted in Fisher 344 pathogen-free rats during their life span. The particles are administered to the animals by intratracheal instillation and intrapleural injections. In vitro studies are conducted on sheep blood erythrocytes and rabbit alveolar macropFages. The cytotoxicity is evaluated by quantitation of red cell hemolysis and cell death respectively. - In Vivo Studies The doses for the intratracheal instillations were determined by an initial range- finding study. Several doses of the particulates were administered to the animals and the highest tolerated dose was determined. Two series of intratracheal studies are planned. Innoculation of the animals in the first series is complete. The second series will be initiated in the near future. Chronic Intratracheal Testing of PMP Amphibole The first series will determine whether the particles alone cause significant toxicity to animals. The regimen for this series is as follows: Series I: Unknown Sample - PMP I Amphibole ........600 animals Asbestos Control - UICC Amosite..........200 animals Negative Control - Saline and Gel ........ 200 animals Chronic Interaction Studies by Intratracheal Instillations The purpose of the second series is to determine whether the particles will interact with a known carcinogen to produce a higher incidence of tumors. A knpwn amount of benzo(a)pyrene (BaP) will be coated on the particles to compare the synergistic effect of the carcinogen with amosite, the test amphibole, and hematite. The regimen of this series is as follows: Series II: PMP I Amphibole + Bap ....................300 animals UICC Amosite + BaP .......................300 animals Iron Oxide + BaP .........................300 animals PMP I Amphibole ..........................200 animals Iron Oxide ...............................200 animals BaP ......................................200 animals Chronic Intrapleural Testing of PMP Particles Intrapleural studies employing 20 mg of particles injected once into the pleural cavity are being carried out as follows: Series III: Unknown Sample - PMP I Amphibole......... 150 animals Asbestos Control - UICC Amosite.......... 150 animals Negative Control - Saline ................150 animals 170

Ip addition to the lifetime experiments, exploration of the pathological sequences induced by these materials in the lung is in progress by experiments in which sequential sacrifices are being carried out. Figures 11 and 12 demonstrate epithelial polyps and fiber-containing giant cells observed in the parenchyma of rats treated with 12 weekly Injections of 1 mg of amosite or the test sample PMP I, 50 days after the last innocula- tion. The polyps essentially consist of several multi-nucleated giant cells covered with columnar epithelium. Figure 11. Epithelial polyps observed in the bronchi (250x). Figure 12. Multinucleated giant cell containing fibers (1000x). N ~ 171 W ~ ~ J

In Vitro Studies The second part of the biological studies consists of in vitro investigation to determine cytotoxicity of the particles. Two techniques are empTed, namely, sheep erythrocyte hemolysis and rabbit alveolar macrophage destruction. A comparison was made between several commercial asbestos samples of known biological properties, PMP I and non- fibrous grunerite. The data presented in figure 13 suggest that the amphiboles are not as hemolytic as chrysotile fibers, requiring large doses to achieve 50 percent hemolysis. Among the amphiboles, anthophyllite, PMP I, and tremolite are similar in their effect. Crocidolite and amosite seem to be less hemolytic. In contrast, non-fibrous grunerite is non-hemolytic. In the rabbit alveolar macrophage study, amosite and PMP I caused marked depression of cellular viability, whereas non-fibrous grunerite showed no significant change in cellular viability (fig. 14). The sample PMP II is not yet tested. A second advisory committee was convened to consider further investigations to increase our understanding of the mechanisms of mineral interactions with the biological systems. It was the opinion of the committee that the comparative study of minerals should be started as soon as possible. On the basis of the existing data, produced by different laboratories throughout the world, the problem of contamination of the environment with inorganic fibers may pose a significant health threat. Indeed, it may shed significant light on existing problems, e.g., asbestos in potable water supplies, asbestos released from degraded asbestos cement water pipes, natural sources, etc. The selection of minerals and bioassays are as follows: fibrous and non-fibrous grunerite will be collected from different geological localities and their biological properties will be compared. The careful mineralogical analysis and bioassays may indicate whether there is some influence in terms of the crushing process that may create new fiber surfaces not present when communiting materials from other areas. 100 N } 80 QJ 2 = 60 ~ 40 20 0 0 CHRYSOTILE • CROCIDOLITE p (RHODESIA-UICC) TREMOLITE (INDIA) V (S. AFRICA-UICC) GRUNERITE • PMP (NON FIBROUS) O ANTHOPHYLLITE (S. AFR ICA-UICC)  AMOSITE (S. AFRICA-UICC) I I p....q_..~. ~-..q 0.01 0.02 0.05 0.10 0,20 0.50 1 2 5 CONCENTRATION (mg/ml) 10 20 Figure 13. Hemolysis of sheep erythrocytes by various minerals. 172

3 100 80 60 40 20 0 0.01 GRUNERITE (NON FIBROUS) ~ PMP , AMOSITE (UICC-S. AFRICA) I 1 1 111111 1 1 1 111111 1 1 1 111111 1 I t I°II1I 0.05 0.10 0.20 0.50 1 2 5 CONCENTRATION (mg/ml) Figure 14. Cytotoxic effect caused by various minerals when exposed to rabbit alveolar macrophages. For a proper comparison, a standard reference sample of fibrous grunerite (UICC amosite) containing particles of mixed sizes, and another sample specially prepared with short fibers will be used. Since relatively short fibers are observed in Lake Superior, the information obtained from these fibers will be useful. Fibrous cummingtonite with a high magnesium content from several geological local- ities will also be studied for comparison to determine if different processing methods may alter surface properties and, in turn, affect the biological properties. In addition, minerals of known biological properties, such as UICC anthophyllite, UICC chrysotile A, chrysotile RG 144, UICC crocidolite, Indian tremolite, UICC actinolite, antigorite, fibrous glass, and quartz will be studied for comparison. Several assays will be employed to evaluate the biological properties of the inerals. The direct toxicity of the particles will be tested by hemolysis of sheep red blood cells, viability of rabbit alveolar macrophages, human lung fibroblasts such as strain WI-38 and perhaps the mouse ascitis tumor cell line P3881. The possible mutagenic effects of these materials will be evaluated in well-established mutagenesis test systems, such as the Ames test and the L5178Y mouse lymphoma cell assay. Neoplasm induction will be tested by the use of tracheal transplants, as well as transformation of Syrian hamster embryo (SHE) cells, or mouse fibroblasts, such as the C3H 10T>f or BALB/c 3T3 celi lines. 173

Conclusion Preliminary In vitro tests show that both fibrous grunerite and PMP I amphibole are lytic to sheep erythrocytes and depress the viability of rabbit alveolar macrophages, while non-fibrous grunerite is inactive in both systems. The biological significance of these studies Is at this time unclear. Hopefully the proposed investigation will contribute sufficient information to correlate mineral properties to health hazards associated with inhalation and/or ingestion of minerals other than the known commercial asbestos. Mineralogical characterization was done by Illinois Institute of Technology, Chicago, Illinois. Contract #68-02-2451. Discussion C. COOPER: I want to congratulate Dr. Palekar for the description of what is getting under way, and the great care that has been taken apparently to obtain test materials that at least resemble some of the fibers in the taconite areas. I think an important question is how representative this is of the entire Mesabi range and I personally don't have figures available to me as to whether or not the size distributions found at the Peter Mitchell pit are representative of a larger area. I wonder if anybody in the audience here, or Dr. Palekar herself, have data on other areas in the Mesabi range to answer the question as to whether or not 15 percent, approximately, of the fibers are longer than 5 micrometers in length, because the representativeness of this sample is going to be, I think, an important issue in the future, and I wonder if anybody could address themselves to that? L. PALEKAR; I don't have a clear-cut answer to your question, but if somebody in the audience wants to answer that... A. LANGER: You mean the representativeness of the Peter Mitchell fibers? COOPER: Yes. LANGER: It's unlike anything in the rest of the Mesabi. COOPER: Are there air samples in other areas with this same distribution? LANGER: No there are not. Unfortunately for the Reserve Mining Company, the situation at the Peter Mitchell pit is unique for the Mesabi range. The mineral fibers have been originated through contact metamorphism with the Duluth Gabbro, which metamorphose the pre-existing materials here. Now Malcolm Ross is here, who has done work on the amphiboles in the area. He knows a great deal about the geochemistry of the amphibole/pyroxene phases; this is a high temperature metamorphic assemblage, while the rest of the Mesabi range, the rest of the Biwabik iron ore formation, are generally considered to be low temperature iron silicates. They do have problems with fibers, but these may not be as important biologically as the asbestiform amphiboles in the Peter Mitchell pit. This is just unique for that particular area. W. NICHOLSON: In looking at the fiber distribution in other than the Reserve Mining areas, they are of a smaller size distribution and tend, rather than being regular fibers, (that is with collinear sides) to be chips of fibrous length. They are irregular fragments rather than the natural fibers that we've been hearing of, and they are in general of a size distribution somewhat smaller than that which has been described here, but there are many fibers (that is defined by a 3 to 7 length to width ratio) that are present in other areas. P. GROSS: I would like to comment on the two microphotographs of tissue which Dr. Palekar showed. I was most interested in the visualization of fibers at that magnifi- cation, which indicated that the fibers were quite long, much longer than 5 microns. As a matter of fact, one of the fibers that I saw, where one of the giant cells was, was as 174

long as a giant cell, which probably was in the neighborhood of 100 pm in length. Also the photomicrograph of the bronchial-polyp, this sort of picture has been produced in my laboratory with long fibers of any kind: glass, silicon carbide, aluminum silicate, as well as asbestos. Again, it suggests the presence of a fairly considerable number of long fibers, and it seems to me that may be a reflection of an exceedingly high dosage administered even though your long fibers were less than 15 percent of the total. PALEKAR: Yes sir, we administered the highest tolerable dose. The animals received twelve weekly injections of 1 mg. A. WILEY: Since there seems to be a good deal of controversy about what's a fiber and what's not a fiber, I was interested in your characterization of a grunerite sample as non-fibrous and I'd like to know what you meant by it. PALEKAR: The particles are not completely characterized at this time and it was presumptuous on my part to present the data. This is really a very preliminary study and no conclusions can be drawn at this time. We have asked our colleagues from IITRI to analyze this properly. Thus far I have just taken their word for the non-fibrous nature of the particles. G. NORD: Yesterday we saw a great deal about the mineralogy of amphiboles. One of the things that was brought up was the defect structure of amphiboles. Amosite has a very high defect density; it's polysynthetically twin on a unit cell scale. The grunerite that you used, or I should say the minerals that you used from the Mesabi range sample, may have an entirely different defect population. Is there going to be any attempt to characterize this defect population? That could also go for the characterization of the samples discussed by the previous speaker. I have one other comment: It's not enough to characterize a fibrous mineral strictly by an energy-dispersive analysis. You cannvt tell the difference between a low calcium pyroxene and a low calcium amphibole. It tis not enough to characterize a low calcium amphibole merely by knowing its chemistry. It also has a different structure; you have orthorhombic amphiboles and you have monoclinic amphiboles. Grunerite/cummingtonites are monoclinic. You also have anthophyllites which are orthorhombic. If one is to characterize these samples adequately so one can separate out the very small differences, perhaps in the experimental data, you will have to do a great deal more work. PALEKAR: Well, this paper is by no means the entire story. I never said that this is it, that this is the only thing we are going to do. We are open to ideas and we are going to characterize many more minerals more thoroughly; this is just the beginning and we intend to do further analyses. B. SMITH: Dr. Palekar, I believe you said that the EM measurements that you had on a standard reference sample of amosite, UICC amosite, was showing about 87 percent of the particles shorter than 5 pm, and that the measurements that you had on the preparation, the PMP preparation that you made from taconite rock, showed about 85 percent of fibers running below 5 pm. Now, as I looked at the photographs you showed, the photographs of the taconite preparation had a micron scale on them, so we were looking at fibers that were being compared with a 1-micron scale. They didn't seem to be more than, or only a little bit more than the scale. They looked to me about 2 or 3 times the size of the scale, so I guess they were fibers that were about 2 or 3 Nm long. In comparison, the photograph you showed of the UICC amosite was fitted with a 10-micron scale and there were an enormous number of fibers visible in that photograph that were much longer than the 10- micron scale. This presents a problem that has puzzled me many times in samples that I've looked at, where we've gotten electron microscopy measurements that are telling us that two samples really are about the same as far as the mean fiber length is concerned. When I look at them with an optical microscope, it's very apparent to me that there are an enormous number of long fibers that I can easily see at say 400X in one sample, and with the other sample that electron microscopy figures are telling me is about the same, I have a tough time seeing any fibers. Now how do we get around this problem? PALEKAR: Yes, I agree with you wholeheartedly and I had the same questions to our mineralogist. The electron micrographs of the fibers are not representative. It is known that there is a tremendous variation between samples. One must make an effort to use the 175 2063104971

4 . same sample for mineralogical analysis and biological evaluations to establish a proper relationship between the two. 0. WALIA: I don't have a question but I'd like to address myself to some of the comments regarding the preparations and characterizations that we did for Or. Palekar. The comment that electron microscopy is not the only criteria to distinguish one fiber from another is true, and we did not depend only on that. Instead we picked up the fibers from the filter samples, mounted them on glass fibers, and then performed x-ray diffraction studies on them. We then compared the data with the known fibers from the taconite mines, and also with the ASTM standards, and from that we were able to identify or pinpoint their identity as to the mineral species. Second, regarding the size distribution comments, if you remember the tables Dr. Palekar showed, in the case of UICC amosite, where we have compared our size distribution data, which is both by diameter and by length, you get a comparison within ±6 percent. I believe this is a good comparison and from the table you see that UICC amosite has fibers which are as long as 200 pm. When we look at the taconite samples, which we have prepared, and the size distribution data, you see that there is no fiber greater than 20 pm. To my knowledge, from all the taconite rock samples I've seen, I've never come across any mineral fiber which, even using this ambiguous three-to- one aspect ratio criteria, that I can say is 200 pm in length. Another comment I'd like to address myself to is about the non-fibrous grunerite we used. This non-fibrous grunerite, which has some preliminary results that Or. Palekar showed, was the one we got from bawabush iron ore formations in Canada, and the non-fibrous nature of this is based on the lack of flexibility of the fibers which you commonly see in UICC amosite type materials. NOTE: The following notes were sent following the meeting and were not part of the verbal discussion at the end of the session. GROSS: Dr. Palekar's description of the bronchial lesions that develop in animals following the intratracheal injections of long-fibered asbestos as "polyps" deserves explanation. A polyp is generally conceived to be a tumor - a neoplasm. The intrabronchial lesions developing in animals after intratracheal injections of asbestos are not tumors. The lesions are composed of inframmatory tissues that surrounds impacted, aggregated asbestos. The inflammatory tissue extends (often in a finger-like manner) into the bronchial lumen and, in time, becomes covered by normal-appearing bronchial epithelium - hence its resem- blance to a polyp. R. BLEIFUSS: The reports submitted by the Illinois Institute of Technology Research Institute (IITRI) regarding the origin of the sample materials to be used in these biological studies indicates that the source material represents an unusual situation within the Peter Mitchell Pit (PMP) of Reserve Mining Company. The original sample material represents a unique occurrence within the PMP in the same sense that the PMP may be said to be unique to the rest of the Mesabi Range. IITRI personnel collected more than 100 samples in their initial survey an which they carried out extensive mineralogy studies to characterize the ore. Based on this initial information the sample location from which they extracted the fibers for the biological study was selected as described below.i "On October 2, 1975, approxiwtately 750 lbs of high fibrous content ore were located and collected. It was found that the ore containing rich fibrous veins was a very localized phenomenon. Such samples were avail- able only near the incursion of the Duluth Gabbro and occurred only in two very localized areas within approximately 100 m of each other." lIITRI Report No. C6321C02-11, Final Report, Contract No. 68-02-1687, "Amphibole Mineral Study to Complement the Ongoing Characterization of Finely Particulate Environmental Contaminants for Biological Experimentation." 176

"Fibers were separated from fiber-rich rocks using several methods. Both hand and vibratory cobbing were used to separate fibrous material (~1.5 kg) in veins. Several rocks were found to consist almost entirely of soft, light green or brown fibrous material. These rocks were crushed, ground, and sieved (<35 mesh) to produce a material (~3 kg) with a high fibrous-to-non-fibrous ratio." "These separated fibrous materials are not necessarily representative in all respects of the majority of the fibers in the ore in the Reserve Mine or In the tailings from the magnetite extraction at Silver Bay, Minnesota. However, this method was used as large quantities of materials with a large fibrous fraction could be produced more easily than by separating fibers from the ore or the tailings." The mineral composition of the sample prepared from this "high fibrous" ore, which has been encapsulated for the biological studies, was determined by x-ray diffraction. The minerals "definitely present" include cummingtonite, riebeckite, and rich(t)erite. Minerals "present as trace material" were tremolite and crocidolite. However, the basic mineralogy studies on the 100 original samples include no mention of riebeckite, richterite, or crocidolite. Both the riebeckite2 and crocidolite3 have been described in the literature and are present only in trace amounts in the Peter Mitchell Pit. The sodium in these two minerals is considered to be of metasomatic origin. Richterite was not reported by previous workers in the area which suggests that it may be the result of local hydrothermal activity. Thus the sample prepared for these biological studies contains three minerals which were either unreported or considered to be present in trace amounts by previous authors. These minerals are all commonly reported to be of inetasomatic origin, meaning that some of the critical elements (sodium) for their formation was introduced from outside the iron forma- tion. The occurrence of these minerals in veins further suggests that they are related to metasomatism. _ The sample which was finally selected and processed to produce the fibers for'biologi- cal studies appears to have a unique metasomatic origin, or at least some of the minerals in that sample are related to metasomatism. The sample is certainly not representative of the potential tailings from the PMP. It cannot be classified as typical since three of the finer most important mineral components are certainly atypical in the PMP area. The sample was selected to provide a high "fibrous" to "non-fibrous" ratio that was unobtainable from representative taconite samples. Biological experiments on this sample will contribute little to the resolution of the problem pertaining to the possible carcinogenic nature of taconite tailings. The argument that it is a means of establishing a bridge between a known carcinogen (amosite) and a possible, or suspected carcinogen (cummingtonite in taconite tailings) is not realistic. The direction of the sampling program was to obtain a fibrous sample as analogous to amosite as possible. In so doing it is so far removed from being representative, or typical, of taconite tailings as to make the final outcome essentially meaningless. 2Gundersen, J. N. and Schwartz, G. M., The Geology of the Metamorphosed Biwabik Iron- Formation, Eastern Mesabi District, Minnesota. Geological Survey Bulletin No. 43, 1962. 3White, D. A., The Stratigraphy and Structure of the Mesabi Range, Minnesota, Minnesota Geological Survey Bulletin No. 38, 1954, 92 pp. 177 2063104973

(C National Bureau of Standards Special Publication 506. Proceedings of the Workshop on Asbestos; Definitions and Measurement Methods held at NBS, Gaithersburg, MD, July 18-20, 1977. (Issued November 1978) A STUDY OF AIRBORNE ASBESTOS FIBERS IN CONNECTICUT Leonard Bruckman Air Compliance Unit - Engineering Section Connecticut Department of Environmental Protection Hartford, Connecticut 06115 Abstract The following discussion describes actions taken by the Connecticut Air Compliance Unit for the purposes of studying the danger to public health associated with excessive airborne asbestos fiber concentrations. In Connecticut, the criteria of mesothelioma was selected as the basis for developing an ambient air quality standard for asbestos (i.e., 30 9g/m3 or 30,000 fibers/m3, 30-day average) and compatible mass emission standard (i.e., 24 g/day) in lieu of EPA's qualitative asbestos regulations. An ambient air asbestos survey indicated that asbestos concentrations contiguous to manufacturing sources of asbestos emissions exceed Connecticut's proposed standard. Furthermore, asbestos levels adjacent to toll plazas were also elevated relative to levels removed from manufacturing sources, implicating vehicle brake lining decomposition as a significant source of airborne asbestos fibers. In addition to the aforementioned air asbestos survey, a preliminary study of mesothelioma was conducted. There were 133 Connecticut residents diagnosed with mesothelioma between 1935 and 1972. Although subject to diagnostic error, available statistics suggest that the combined sex age-adjusted mesothelioma incidence rate (AAR) per 100,000 Connecticut population has exhibited a possible 10-fold increase since 1935, rising from 0.02 during 1940 to 0.25 from 1960 to 1969. The trends for both men and women also showed sharp increases over the same time period (1940 to 1970). The rapid rise in Connecticut's mesothelioma incidence rate closely follows the increase in the State's cumulative asbestos consumption and suggests a linearly increasing cause-effect relationship which warrants further investigation. Key Words: Air pollution; air quality data; air quality monitoring; air quality standards; asbestos; health effects; toxic substances. Introduction In 1973 the Federal EPA, recognizing the need to control the emission of asbestos fibers into the ambient air, promulgated National Emission Standards for Hazardous Air Pollutants (NESHAPS) - asbestos, mercury, and beryllium [1,2]1. After an extensive review, Connecticut's Air Compl an~ ce Unit found EPA's asbestos regulation to be inadequate for the purposes of protecting public health in Connecticut and, consequently, developed its own asbestos regulation [3,4]. While EPA's asbestos regulation was written in rather general terms (i.e., "...no visible emissions or application of the best available control technology..."), Connecticut proposed a numerical ambient air quality standcrd of 30 qg/m3 or 30,000 total asbestos fibers (determined by electron microscopy) per cubic meter of air, 30-day average, and a compatible mass emission standard of 24 g/day, at public hearings held in July of 1973. In the judgment of the Connecticut Air Compliance Unit a "no visible iFigures in brackets indicate the literature references at the end of this paper. 179 Preceding page blank 2063104974

emission" asbestos air quality standard does not provide the State's residents with an adequate degree of protection from this carcinogenic substance. In addition, Connecticut also proposed to more stringently control the demolition of asbestos-containing structures. In order to define the magnitude of the environmental hazards posed by airborne asbestos fibers in Connecticut, prior to the promulgation of the State's asbestos standard, the Air Compliance Unit conducted an ambient air asbestos survey along with a study of asbestos-induced mesothelioma Incidence (5,6]. The following discussion describes actions taken by the Connecticut Air Compliance Unit for the puuposes of studying the danger to public health associated with excessive airborne asbestos fiber concentrations. Sources of Airborne Asbestos Fibers in Connecticut Outdoors, the principal source of airborne asbestos fibers in Connecticut is the manufacture of the many asbestos-containing products (e.g., friction products, gaskets). It is estimated that almost 10 tons of asbestos fibers might be released into the Connecticut atmosphere annually as a result of manufacturing operations, assuming reasonably efficient (i.e., 95% asbestos removal efficiency or greater) control equipment is employed. Another major source of airborne asbestos fibers is the erosion of asbestos- containing brake linings and clutch facings. This accounts for approximately two additional tons of airborne asbestos fibers each year [3,4]. Notwithstanding EPA's current regulations covering the demolition of asbestos containing structures, perhaps the largest potential future source of asbestos emissions might be the demolition of buildings which have been insulated and/or fireproofed with asbestos materials. The portion of the NESNAPS regulation pertaining to the demolition of asbestos-containing structures does not clearly state what requirements a demolition operator must meet in order to ascertain whether a structure to be demolished does or does not contain friable asbestos materials. The inherent difficulty in determining wheifiir_a Tiu lding to be demolished contains any asbestos materials, and the associated costs involved in removing such materials if present, necessitate some type of formalized testing procedure. Briefly, such a test might entail taking samples from the walls, the insulation covering load-supporting structural members and the floor and ceiling tile, from at least one floor of the candidate structure, in addition to the insulation covering the boiler and pipes. A composite sample could then be created and analyzed to determine its asbestos content using relatively inexpensive techniques (x-ray diffraction). It is important that the asbestos content of floor and ceiling tiles be ascertained since these non-friable asbestos materials might be pulverized during the demolition of a structure creating a potentially serious asbestos air pollution problem, especially if the technique known as "explosive demolition" is used. The amount of asbestos fiber dust released into the outdoor air during the demolition of an asbestos-containing structure is unknown at this time, but would appear to be potentially large since there are over 2,000 demolitions in the State each year, and should thus be quantified as soon as possible. Indoors, many do-it-yourself home projects create asbestos dust due to the mixing of dry, loose asbestos with water and subsequent application of such mixtures for the purposes of insulating and/or fireproofing boilers, pipes, etc..., and the cutting and sawing of asbestos-containing wallboard, ceiling, and floor tile. Perhaps the most serious public health hazard posed at this time by excessive asbestos fiber exposure has been created by the release of asbestos fibers from asbestos-containing surface coatings, which were applied indoors to walls, ceilings, exposed structural steel, air ducts, plenums, return air spaces, for insuiating, decorating, and fireproofing purposes indoors. As a result of such activities, appreciable amounts of asbestos fibers may be released into the air indoors, during the application, again as the surface coating deteriorates, and finally, when the building is demolished. The asbestos fibers resulting from the spraying operation itself, as well as those released from the coating over a period of time due to its friable nature, should be of primary health concern. At least one state (i.e., New Jersey) and one local municipality (i.e., New Haven, Connecticut) have already promulgated regulations for the purposes of controlling and/or prohibiting the future use of spray-on asbestos surface coatings indoors. NESHAPS currently prohibits the use of such asbestos-containing spray-on insulation and fireproofing materials outdoors; a recent amendment to NESHAPS proposes to prohibit the future use of any type of spray-on asbestos coating indoors [7]. 180

Ics Ambient Air Asbestos Standard The approach taken in developing Connecticut's proposed ambient air quality standard for asbestos was to derive a numerical standard which should not be exceeded at this time. In other words, all assumptions were made such that the standard could not be criticized as being too strict. Setting standards should be viewed as a dynamic process in that any value must be reviewed and revised periodically as additional pertinent information becomes available. Even a preliminary air quality standard is valuable because it provides some quantitative idea as to what health risk is associated with varying pollutant levels. Such a standard can be especially useful in developing a set of priorities for correcting environmental problems created by certain pollutants. For example, areas which are well below the standard need no immediate attention, while areas well above the standard require that some sort of corrective action be taken as soon as possible. Such an approach is particularly needed for toxic multi-media environmental pollutants, such as asbestos. In this manner limited resources can be effectively directed at solving the more serious aspects of the problem and, at the same time, actions based solely on emotional decisions by poorly informed administrators can be minimized. Connecticut's proposed asbestos standard should be viewed in this light; i.e., this standard is a first attempt at quantifying the adverse health effects posed to the general public by excese airborne asbestos fibers. Hopefully, any questions raised by the rationale used in developing this standard will be answered by future studies using varied approaches. Mesothelioma incidence was selected as the foundation for developing Connecticut's proposed air quality standard for asbestos for the following reasons [8-10]: The high frequency of lung cancer in the general population makes it difficult to relate a given case of bronchiogenic carcinoma to asbestos exposure with the high degree of probability that exists for mesothelioma. Some investigators suggest that the smaller asbestos fibers (i.e., those less than 5 p in length) most likely encountered in the ambient air may be incapable of inducing lung cancer, however, it has not been demonstrated that these shorter asbestos fibers are incapable of producing mesothelioma. Most of the information available on the adverse health effects caused by excessive asbestos fiber exposure has been collected in occupational environments (Table 1) [11-17]. Table 1. Incidence of inesothelioma and asbestos concentrations in occupational environments [11]. Industry Cohorta number of individuals Mesothelioma incidence percent Reference Highestb concentration fiber/cm3 Lowestb concentration fiber/cm3 Insulation 689 2.18 [11] 74.4 0.1 Shipyards 3000 0.73 [11],[14] 8.7 0.3 Construction 632 0.63 [11],[15] 7.1 0.9 Textile plants 716 1.50 [113,[l3] 29.9 0.1 1300 1.00 [11],[13],[16] 29.9 0.1 ti1300 1.20 [11],[12],[17] 29.9 0.1 a Most of the individuals in these studies had been followed for 20 years or longer. Concentrations for NIOSH document [18]. $ 181 ti ~ b J a

.C Unfortunately, quantitative dose-response relationships concerning environmental asbestos exposures and mesothelioma incidence in different industrial settings are not available. In 1973, the National Institute for Occupational Safety and Health (NIOSH) monitored asbestos concentrations in a number of occupational environments [18]. While these short-term fiber concentrations are of recent origin and, therefore, cannot be directly related to epidemiological studies of mesothelioma incidence, they can be used to obtain an estimate of the range of occupational asbestos exposure likely encountered in different industrial settings. For example, Selikoff and co-workers reported that for workers in the construction industry (followed for 20 years or longer) 0.63 percent contracted mesothelioma [15]. The variation in asbestos fiber exposure for the construction industry from the NIOSH study ranged from 0.1 to 29.9 fibers/cc which corresponds to a hypothetical probability of contrac- ting mesothelioma of 63/10,000 (i.e., 0.63%). In a like manner, occupational mesothelioma incidence (provided by studies appearing in the open literature) and corresponsing estimates of the range of asbestos fiber exposure (provided by the aforementioned NIOSH report) were used to construct a first generation occupational asbestos fiber exposure-mesothelioma incidence envelope (Figure 1). 108 107 LL 104 103 10-g 10 e 10-4 10-3 10'2 10-1 Hypothetical Probability Of Contracting Mesothelioma Figure 1. Expected incidence of contracting mesothelioma as a function of industrial air asbestos exposure (8 hr day. 5-day week). 182

CS Only asbestos fibers greater than 5 p in length with an aspect ratio of 3:1 (as viewed by phase contrast light microscopy; 430X magnification) are monitored in industrial environments. These longer asbestos fibers account for approximately two percent of all asbestos fibers present (by number) [19]. Expressed in another manner, there are approximately 50 asbestos fibers for every 5 p size fiber. Furthermore, it has been estimated that there are approximately 1,000 asbestos fibers per nanogram of asbestos [3,20,21]. Consequently, 20 "industrial size" asbestos fibers are equivalent to approximately one nanogram of asbestos. Other investigators have reported similar relationships between industrial size asbestos fibers, total asbestos fibers and their weight equivalents [3,19]. In addition, occupational exposure concentrations based on a 8-hour day, 5-day week should be related to general population ambient exposure levels. This can be accomplished by dividing occupational concentrations by 4.2 (i.e., 24-hour/8- hour x 7 day/5 day = 4.2) [22]. Now the occupational mesothelioma incidence envelope dep'icted in Figure 1 can be converted to a general population eesothelioma incidence envelope (as a function of both weight and number of asbestos fibers per volume of air), from which an ambient air quality standard for asbestos can be selected (see Figure 2). Using the minimum line a level of 30 ryg/m8 or 30,000 fibers/mg, which is projected to induce 150 mesotheliomas nationwide or 2 in Connecticut, was chosen. The use of the minimum line, which reflects the smallest probability of an individual contracting mesothelioma for a given exposure level, is consistent with the aforementioned objective of developing an asbestos standard which' would be difficult to criticize as being too strict; the use of either the maximum or some average line would have yielded an asbestos standard some 2 orders of magnitude more restrictive (lower) than the proposed standard for the same response. The chosen standard should result in about 1/10 the yearly fatalities from airplane accidents and approximately the same number of deaths as from train mishaps (see Figure 3) [3]. F 103 C ! 0 V1 O U) ~ ~ Jd E Q 10 2 100 10 Proposed A-Min Standard Z of 30 qg/m3 ~h • . 30,000 total tib.rs/ms i I 102I I 103 l0, 150 Nationwide Expected / 104 105 Cases Of Mesothelioma 105 0 ~ Figure 2. Nationwide expected cases of mesothelioma as a function of ambient air asbestos exposure (assumed population of United States was 230 million people). 183 106 2063104978

, 105 , } .~ ~ a v m ~ a °- 104 c aa s H ~ O t 5 3 R 10 m CC Q w Q H L Activity Figure 3. Nationwide mortality statistics due to different modes of travel and expected cases of inesothelioma. The subject asbestos standard is equivalent to an occupational asbestos level of 0.0025 fibers (>5 p)/cc, well below the newly proposed occupational standard of 0.5 fibers (>5 N)/cc [23]. This strongly suggests that the aforementioned proposed occupational asbestos standard is not yet low enough to adequately protect the worker exposed to asbestos fibers froo contracting mesothelioma. Connecticut's ambient air quality standard for asbestos is based on a 30-day average sampling period instead of the more common 24-hour duration because a 1-month averaging time is more manageable from a monitoring standpoint and is not sensitive to short-term perturbations in air asbestos emissions, but at the same time provides the public with a high degree of protection from the adverse health effects caused by excessive asbestos fiber concentrations. Compliance with the proposed standard can be easily and accurately evaluated using Connecticut's low-volume particulate sampler (lo-vol) [6,24]. 184

C? In certain instances it may be necessary to impose asbestos emission standards on manufacturing and other sources of airborne asbestos fibers in order to attain the desired ambient air asbestos standard. A mass emission standard of 24 g/day (for an isolated point source of asbestos emissions) is consistent with the 30 qg/mg (30,000 fibers/m3) proposed standard. The development of this emission standard, in addition to a possible stack sampling train, are explained elsewhere [3,4]. Mesothelloma Incidence in Connecticut The mesothelioma incidence trend In Connecticut men mounted through the 10 year period covering 1960 to 1969 from an age-adjusted rate (AAR), obtained using the indirect method, of 0.04/100,000 Connecticut population for the interval between 1940 and 1949 to 0.37/100,000 from 1960 to 1969. No mesotheliomas were diagnosed in Connecticut women until the period 1950 to 1959 when 12 were reported yielding an AAR of 0.1/100,000. The trend for females increased slightly to 0.15/100,000 in 1960 to 1969 (Figure 4). The combined sex AAR 0.4 I a i d a U C ~ U ~ i i 0.3 0.2 0.1 0.0 L• I 1940-49 1950-59 Period 1960-69 Figure 4. Connecticut mesothelioma incidence by 10-year period. 185

rose from 0.021100,000 during 1940 t0 0.25/100,000 frdm 7960 to 1969, over a 10-fold increase. The increa3e in cases over the years may in part reflect an increased awareness of this type of tumor and an attempt by pathologists to classify all malignancies. Though increases in both occupational and non-occupational asbestos fiber exposure are expected to have occurred over the last 40 years, only four people were reported with known exposure to asbestos. Eight others were felt to have experienced some exposure. Occupation at the time of diagnosis was obtained from hospital admission records and the usual occupation from death certificates. It was found that 44 individuals (33.0%) worked in the home or in like occupations. Thirty-six (27.1%) were reported to have worked in manufacturing industires. Nineteen (14.3%) worked in offices as professionals or clerical employees. Of the remaining individuals it is interesting to note that one person was listed as a toll collector. Unfor- tunately, complete occupational histories of each of those individuals afflicted with mesothelioma are not available at this time [57. Cumulative United States asbestos consumption has increased rapidly since the beginning of the 20th century and is projected to exceed 60 million tons by 1980; [25) Connecticut's asbestos consumption has been estimated by proportionally allocating total U. S. consumption using the appropriate Connecticut to United States population ratio. A plot of both cumulative U. S. and Connecticut (estimated) asbestos consumption and Connecticut's combined- sex mesothelioma AAR/100,000 population as a function of time suggests that the sharp increase in mesothelioma incidence closely followed the rapid rise in the State's cumulative asbestos consumption for comparable intervals (i.e., 1940 to 1970) (Figure 5). This apparent cause-effect relationship warrants further investigatian. 10e 10` l0' 10° . lOs 11~ ~ ~ 11 1 a ~ 10-: ~CJp~A~N~tr)~d'g1Ao?tD Period Figure 5. Cumulative asbestos consumption and Connecticut mesothelioma incidence as a function of time. 186

C. 2 Air Asbestos Survey An ambient air asbestos survey was conducted during late 1975 and early 1976 to define the magnitude of the health hazard posed by airborne asbestos fibers in Connecticut prior to the promulgation of the State's asbestos standard. The newly developed low volume particulate sampler (lo-vol) (see figure 6), which operates continuously for a 30- day interval at an air sampling flow rate of approximately 4 cfm, was used to collect ambient TSP samples. The io-vol was equipped with special membrane filters (8" x 10", Gelman Metricel GN-6 0.45 p pore size, non-nylon reinforced). The filters were analyzed for chrysotile asbestos by the Battelle-Columbus Laboratories using transmission electron microscopy in conjunction with electron diffraction (to confirm a minimum of 10 chrysotile asbestos fibers) [6]. Filter Transition Piece Stalnless Steel Adapter Transition Piece ~`emperature t"_1511__~ High-Volume Sampler Compensating Dry Gas Meter Figure 6. High volume (hi-vol) and low volume (1o-vol) TSP samplers. Approximately 30 monitoring sites were selected; locations included "typical" urban sites removed from known sources of asbestos emissions, rural-background sites and stations contiguous to four industrial users of asbestos (i.e., manufacturers of friction products, insulated wire and cable, ammunition and molding compounds, respectively) and three toll plazas situated at various locations along Interstate 95. Ambient chrysotile asbestos levels removed from asbestos emission sources in both urban and rural location were below 10 ng/m3. However, chrysotile asbestos concentrations above the 30 rg/m3 proposed standard were measured near each of the industrial users of asbestos (i.e., 32 rig/m3 at a public works building located near the friction products manufacturer; 33 ng/m3 at a junior high school located adjacent to the insulated wire and cable and ammunition manu- facturer combination; 33 rlg/m3 at a private home near the molding compounds manufacturer). 187 ~ u G r-

Each of the subject point sources are in compliance with NESHAPS and other existing state and federal air quality regulations. Ambient asbestos levels adjacent to the three toll plazas on 1-95 were also elevated (in the 10 qg/m3 to 25 Ig/m3 range), implicating asbestos emissions from vehicle brake lining decomposition as a significant source of airborne asbestos fibers. Asbestos concentrations at the rural toll plaza (11,000 cars/day eastbound lane; 12,000 cars/day westbound lane) were 10 qg/m3 (eastbound lane) and 14 rig/m3 (westbound lane); there are no known industrial users of asbestos near this rural toll station. Asbestos levels at one of the urban toll plazas (28,000 cars/day eastbound lane; 27,500 cars/day westbound lane) were 3 qng/m3 (Administration Building, south side of highway) and 25 ng/m3 (westbound lane). The asbestos concentration at the other urban toll plaza (27,000 cars/day eastbound lane; 28,000 cars/day westbound lane), which is also located near one of the largest industrial users of asbestos in Connecticut (i.e., the aforementioned friction products manufacturer), was 41 qg/m3 (Administration Building, south side of highway); this was the highest concentration measured during the subject survey. The ratio of the maximum asbestos concentration measured at the first urban toll plaza to that at the rural toll station was approximately equal to the ratio of the number of cars/day passing through each toll plaza (i.e., 1.8 versus 2.3) during the sampling interval. All of the aforementioned measured asbestos levels were 30-day average values, except the 41 qg/m3 concentration, which was approximately a 20-day average value (due to a sampler malfunction). In addition to the ambient air asbestos survey described above, asbestos levels were also measured indoors at the boy's swimming pool located in the University of Connecticut's field house. The ceiling covering this pool was sprayed with an asbestos-containing insu- lating compound in 1955 and then re-sprayed some 10 years later. Chunks of this coating have been falling from this exposed ceiling for some two years. Analyses of a bulk sample of the ceiling material by the Connecticut State Department of Health revealed evidence of asbestos fibers (between 10-30%) within fiberglass and binding material. Subsequent electron microscopic analyses of the ceiling material by the Battelle-Columbus Laboratories indicated that the asbestos was of the amphibole variety. Four (4) long-term (i.e., 30-day) air ' samples were collected at various locations at the pool. Identical sampling techniques ~ were used for both the indoor and outdoor air asbestos surveys. These indoor samples are ; being analyzed for amphibole asbestos using transmission electron microscopy and energy ~ dispersive electron-diffraction by Walter C. McCrone Associates, Inc. The results of this : indoor asbestos survey will be reported at a later data [26]. Conclusions and Recommendations Connecticut's studies to-date indicate the existence of a potential health hazard posed by airborne asbestos fibers which warrants further investigation. Firstly, additional ambient asbestos monitoring should be performed as soon as possible to: 1) define the month-to-month variations in ambient asbestos levels at various locations, primarily in densely populated areas contiguous to manufacturing sources of asbestos emissions and especially those locations which already exhibited asbestos concentrations in excess of Connecticut's standard, 2) further quantify, asbestos levels near toll stations, the relation- ship between traffic counts and ambient asbestos concentrations, and determine how asbestos levels decline with increasing distance fraa a toll plaza, 3) define ambient asbestos concentrations contiguous to different types of demolition operations and how rapidly these levels approach background concentrations after the demolition activity is completed, and 4) quantify the hazard posed by asbestos concentration indoors where it is suspected that asbestos-containing spray-on materials are fraying and flaking. 188

C Secondly, the relationship between asbestos consumption and mesothelioma incidence in Connecticut should be investigated in more detail. A thorough epidemiological study of the 133 reported cases of mesothelioma (as of 1972) should be performed as soon as possible to i6entify those cases which are likely associated with non-occupational asbestos fiber exposure. A prospective study of school children exposed to asbestos fibers indoors as a result of the spray-on application and deterioration of asbestos-containing surface coatings should be conducted to accurately quantify the health hazard posed by this type of asbestos fiber exposure. . It is recommended that Connecticut's standard be promulgated and applied both outdoors and indoors. The routine monitoring of asbestos levels should be initiated as soon as possible. The resulting measured concentrations (along with the populations exposed) should be compared to the standard so that a rational program and set of priorities can be formulated to minimize the health hazard posed by airborne asbestos fibers. This seems to be the most logical way to objectively determine how best to allocate the people's money in implementing sensible ways of controlling contamination of the environment by airborne asbestos fibers. References [1] National emission standards for hazardous air pollutants (asbestos, beryllium, and mercury), Federal Register, 38, 66 (April 6, 1973). [2] Asbestos and mercury (proposed amendments to national emission standards), Federal Register, 39, 208 (October 25, 1974) [3] Bruckman, L. and Rubino, R. A., Rationale behind a proposed asbestos air quality standard, J. Air Poll. Control Assoc. 25, 1207 (1975). [4] Bruckman, L., The Environmental Impact of Asbestos in Connecticut, Internal report issued by the Connecticut Department of Environmental Protection, Air Compliance Unit, Engineering Section, March 12, 1973. [5) Bruckman, L., Rubino, R. A., and Christine, B., Asbestos and mesothelioma incidence in Connecticut, J. Air Poll. Cntr. Assoc. , 27, 121 (1977). [6] Bruckman, L. and Rubino, R. A. , Monitored asbestos concentrations in Connecticut, paper presented at the 70th annual meeting of the Air Pollution Control Association, Toronto, Ontario, June 20-24, 1977. [7] National emission standards for hazardous air pollutants; proposed amendments to asbestos standard, Federal Register, 42, 41 (March 2, 1977). [8] Health effects and recommendations for atmospheric lead, cadmium, mercury, and asbestos, Illinois Institute for Environmental Quality, Environmental Health Resources Center, Report No. EQ-73-2, Chicago, Illinois, 1973. [9] Gross, P., Is short-fibered asbestos dust a biological hazard, Arch. Environ. Health, 29, 115 (1974). [10] Gross, P., deTreville, R. T. P., and Haller, M. N., Asbestos versus nonasbestos fibers, Arch. Environ. Health, 20, 571 (1970). [11] Health effects and recommendations for atmospheric lead, cadmium, mercury, and asbestos, Environmental Health Resources Center, State of Illinois Institute for Environmental Quality, Report #IIEQ-73-2, Chicago, Illinois, 1973. [12] Lewinsohn, H. C., The medical surveillance of asbestos workers, !a. Soc. Health J., 92, 69-77 (1972). (13] Newhouse, M. L. , Berry, G. , Wagner, J. C., and Turok, N. E. , A study of the mortality of female asbestos workers, Brit. J. Ind. Med., 29, 134 (1972). 189

[14] Stumphuis, J., Epidemiology of mesothelioma on Walcheren Island, Brit. J. Ind. Med., 28, 59 (1971). [15] Selikoff, I. J., Churg, J., and Hammond, E. C., Asbestos exposure and neoplasia, J. Am. Med. Assoc., 188, 22 (1964). [16] Newhouse, M. L., A study of the mortality of workers in an asbestos factory, Brit. J. Ind. Med., 26, 294 (1969). [17] Knox, J. F., Holmes, S., Doll, R., and Hill, I. D., Mortality from lung cancer and other causes among workers in an asbestos textile factory, Brit. J. Ind. Med. 25, 293 (1968). (18] Criteria for a recommended standard for occupational exposure to asbestos, U. S. Department of Health, Educationa and Welfare, Public Health Service, Health Services and Mental Health Administration, National Institute for Occupational Safety and Health, HSM N72-10267, Washington, 0. C. 1973. [19] Lynch, J. R., Ayer, H. E., and Johnson, 0. L., The interrelationships of selected asbestos exposure indices, Amer. Ind. Hyg. J., 31, 598 (1970). [20] Thompson, R. J., personal communication, preprint R. J. Thompson and G. B. Morgan, Determination of asbestos in ambient air, May 2, 1973. [21] Wesolewski, J. J., Asbestos in the California environment, Air and Industrial Hygiene Laboratory Report, AIHL #164, California State Department of Health, Berkeley, California, May, 1974. [22] Fulkerson, W. and Goeller, W. E. , (eds.), Cadmium: the dissipated element, Oak Ridge National Laboratory, Report #ORNL-NSF-EP-21, Oak Ridge, Tennessee, 1973. [23] Occupational exposure to asbestos; notice of proposed rulemaking, Federal Reaister, 40, 197 (October 9, 1975). [24] Bruckman, L., Hyne, E., and Norton, P., A low volume particulate ambient air sampler, paper presented at the Speciality Conference: Measurement Accuracy as it Relates to Regulation Compliance, New Orleans, Louisiana, October 1975. [25] Clifton, R. A., Asbestos, preprint from the 1972 Bureau of Mines Minerals Yearbook, U. S. Department of the Interior, Washington, D. C. 1975. [26] Bruckman, L. , Monitored asbestos concentrations indoors, paper presented at the Fourth Joint Conference on Sensing of Environmental Pollutants, November 6-11, 1977, New Orleans, Louisiana. Discussion NOTE: Discussion of this paper was included in the General Discussion at the end of this session. 190

5 National Bureau of Standards Special Publication 506. Proceedings of the Workshop on Asbestos: Definitions and Measurement Methods held at NBS, Gaithersburg, MD, July 18-20, 1977. (Issued November 1978) GENERAL DISCUSSION OF RELATIONSHIP BETWEEN CHEMICAL AND PHYSICAL PROPERTIES AND HEALTH EFFECTS Editor's Note: This session was actually conducted on two days. The papers through Dr. M. Stanton's were given the first day and were followed by a general discussion. The remaining papers were presented the next day, followed by a second general discussion. These two general discussions have been combined below and are followed by a summary given by the Session Chairman, Dr. S. Schneiderman, at the start of the second day's papers. (CCG) A. SUNDARAM: I would like to address this question to Dr. Kotin. He mentioned that he believes that there exists a no-effect level for asbestos. Assuming that he is right, what are the future steps that industry is going to take? Are they going to conduct animal studies at various dose levels, come up with a no-effect level, extrapolate to the human situation, and have a TLV? Alternatively, are they going to do more epidemiological studies and come up with a no-effect level which can directly apply to humans? If we do these two types of studies, we are still faced with the problem of variability of suscepti- bility between different groups of humans as well as between animals and humans. P. KOTIN: The answer to the first part is obviously that industry has a respons{bility to support studies at all levels, from fundamental mechanisms to bioassay. My own bias is that there is a no-adverse effect level. The question of what will demonstrate the no- adverse effect level is one that is going to require a fundamental understanding of carcin- ogenesis. I think there are occupational asbestos exposures of sufficient duration where, to the best of our knowledge as of now (and you always have to put that in), there seems to be a level of exposure to asbestos not associated with asbestos-related disease. I will say, in advance, I am aware of and accept all the caveats that just as the rats have not lived three years, these people have not been exposed for forty years, and maybe at that end of the distribution curve some evidence of some response may come. There is no answer to your question, and I wish that there were, other than to say that industry has a responsi- bility and would be incredibly shortsighted and incredibly stupid if it were not on the leading edge of supporting all research in relation to fiber and its relation to any adverse human effect. M. SCHNEIDERMAN: Dr. Nicholson, would you care to comment on the no-adverse level problem since you presented information on individuals exposed one month? W. NICHOLSON: In fact, I was going to ask Dr. Kotin to elaborate on that. I recall seeing a quote from you that was made sometime in the late sixties before some congressional committee, when you were Director of the NIH. You felt at that time that there was no evidence that would indicate that a threshold exists. If you could elaborate on that, particularly on the hard data that exist for asbestos. As one knows, you need enormous populations in order to see what the dose-response is at lower levels of exposure. I am in complete agreement with you that there is a dose-response effect at the levels we are speaking of; as you go down in exposure and dose, you certainly go down in effect, but to my knowledge the difficulty of finding the existence of a threshold exceeds our capability either in animals or in man. KOTIN: The answer to your second part is the degrees of reliability that you are willing to accept in terms of the totality of any response. Let me elaborate a little more. First of all, indeed I did say that, not only before a congressional committee but before numerous congressional committees. I only have two comments: (a) I'm smarter now, and (b) I will send you reprints of three articles published in 1954 where I say, on the basis of what is now known, air pollution is infinitely more iAportant to the evolution of bronchogenic cancer than cigarette smoking. If I am dumb initially, at least give me credit 191 2063104986

for not being cast in concrete in opinion. No, the answer to your question is that there are no absolute data that a no-adverse effect level exists, because of the heterogeneity of man. So what I have chosen to do is look at the sequence of events that are necessary for the evolution of a cancer, and I have not used asbestos as a model but I have used other carcinogenic agents, such as aromatic amines and hydrocarbons. There is no such thing as a threshold for carcinogenesis, there are a series of thresholds. I am prepared to say that at the molecular level you may have a threshold, but in terms of clinical cancer and par- ticularly in the laboratory, one can quantify the exposure to carcinogenic agents and predictably get a carcinogenic response, including no tumor formation within the normal life span of the animal, with no evidence of any abnormality. This is a mumbojumbo answer because it is not a clear thing, otherwise we would just have to go to the blackboard and make this a seminar on just chemical carcinogenesis, which I would be delighted to do, and then get down to specifics rather than these generalizations. Cancer is not a simple process. It is a highly complex sequential process, with sequential steps dependent on the antecedent step, and the sequential steps capable of occurring or not occurring on the basis of what happened in the immediate antecedent step and this can be quantified beautifully. NICHOLSON: I don't think I want to pursue this, except to make one comment. Some of the extrapolation and theoretical predictions that one might make on the basis of chemical carcinogenesis, as opposed to asbestos carcinogenesis, may not be that direct. Let me ask a question of Dr. Stanton which has to do with relative risks of fibers of different lengths (as with the issue of threshold; it is a relative risk at different doses): finding in human tissue and in air samples the vast preponderance of fibers of the shorter sizes, less than 5 pm (we have had some air exposures where 99.5% are under 5 pm in length, others may be 98% or 95% depending upon the particular process), at what level can you say, or at what length can you say, that the shorter fibers are ten times or fifty times, or some rough estimate, less carcinogenic than the longer ones. Certainly your 8 pm value is not a sharp cut off. How might it go down with length, in other words just how does the response go down with dose? M. STANTON: Any correlation data are simply that. It doesn't say that only certain fiber sizes are carcinogenic, nor does it say that short fibers are not carcinogenic. Cor- relation suggests that, long, fine fibers are more carcinogenic than short fine fibers. There is no sharp demarkation line. I think, if anything, one should go back to the patho- logical data again, and I've been impressed by the fact that fibers up to 30 Nm in length can be picked up and effectively handled by a phagocyte. So it may be that we are far under what can be considered very hazardous. Maybe only fibers over 30 Nm in length are more hazardous. Now, what happens if you overload phagocytes? What happens if there are no phagocytes or they are inadequate to handle these fibers in individuals who have compromised the reticuloendothelial system? It may be the short fibers in such situations can be just as carcinogenic as long fibers. There is some suggestion that if the reticuloendothelial system is overwhelmed by foreign bodies, then perhaps short fibers can also be highly carcinogenic. What we are saying sisyly is that long, fine fibers seem to be the most carcinogenic; we are not saying that any fiber is non-carcinogenic. W. SMITH: Question for Dr. Stanton: The experiments that we have had a chance to hear about this afternoon certainly present an animal model for asking questions and gathering information that would be extremely hard to get at through more complicated procedures such as inhatation exposures; but, Dr. Stanton, what do you think about extrapolation of informa- tion gained from intrapleural studies over to situations more comparable to human exposures that could be approached by inhalation studies? We have done a number of experiments by intrapleural exposure of another species, the hamster, to different kinds of minerals. With long thin fibers we have been getting tumors, and with short fibers we have not. One of the materials that has given us a great many tumors has been a preparation of long, thin glass fibers that have dimensions approximately like those that induce tumors in some of the experiments that you just described. However, Dr. Gross, I believe, has exposed rats to some very similar types of fibers by inhalation exposures, and these fibers gave him no tumors at all on the inhalation tests. So here we have a problem of how to extrapolate data from the intrapleural situation, where the fibers are trapped, to the inhalation type of exposure, where they are subject to physiologic clearing mechanisms. 192

STANTON: Clearly, our experiments are designed to find out what happens once the fibers' get to the tissue that is going to respond. It doesn't take into consideration all the extraneous problems that might arise in the fiber getting to that tissue, which is what Bill Smith is saying. What about inhalation? There is no doubt about it; inhalation studies are the only ones that will really give us some reasonable means of extrapolating to human experience. Those experiments have not been adequately done, and there are not enough of them to really get a good handle on what's happening. Dr. Gross has done about the only experiments that have been done up to this point, with the exception of some that Chris Wagner did; he has shown that tumors develop in the lung from various types of asbestos inhalation. Glass has not been studied, or only studied as a large fiber or as non-fibrous material. Dr. Gross is in the audience and I am certain he would be pleased to tell us about his experiments with glass fibers. P. GROSS: We exposed rats and hamsters to fibrous glass dust for a period of two years. The fibrous particles had an average diameter of 0.5 Nm and a range of lengths 5 to 20 pm. Inasmuch as the average fiber length was 10 pm, one-half of the fibers were 10 to 20 pm long and the rest were shorter. Since the dust concentration was ti10g mg/M3, the exposure included ~50 mg/Ms of fibers 10 to 20 pm in length. Thin mineral fibers of this length have been found carcinogenic when implanted in the abdomen or thorax of rats. However, long-term exposure by inhalation of these long, thin glass fibers resulted neither in pulmonary fibrosis, lung cancer, nor mesothelioma in ahy of our animals. These were allowed to live out their lives. I. ASHER: We are concerned about parenteral drugs, and we are wondering if anyone has any information about subcutaneous or intravenous injection of solutions that contain asbestos or fiberglass fibers? EDITOR'S NOTE: No reply was received to the above question. (CCG) ' R. LEE: Questions for Dr. Palekar: First, what is the unknown amphibole,' PMP 1? Second, I'd like to point out that there were at least three or four people very familiar with scanning microscopy who picked up a possible trace quantity of potassium in what you called a non-calcium amphibole. It would be very surprising if that particular non-calcium amphibole x-ray spectrum looked just like that, and it was a grunerite! Next, I was wondering if PMP 1 is a mineral characteristic of the Peter Mitchell pit, and is that a fibrous or non-fibrous variety of material? Finally, what was the set of aspect ratios measured and particle sizes measured for the non-fibrous "grunerite?" Were they cleavage fragments or typical of amosite? L. PALEKAR: Yes, PMP 1 is the unknown sample. We did some analyses of the air samples in the taconite mine and it happens to be Peter Mitchell pit; that's correct. There were two samples, one had calcium and the other didn't. The first sample I believe had calcium and the second didn't. I wasn't aware of the fact that there was a potassium peak on it. According to our mineralogist colleagues from IITRI, the studies were done by using several other techniques, and they didn't find any potassium. LEE: In that particular spectrum you showed something which had at least, on a con- servative estimate, one percent and possibly two percent potassium. PALEKAR: I will have to take that into consideration. Your second question is whether we did any analysis on non-fibrous minerals. So far, we have not, but we intend to do it in the near future. LEE: Was the sample identified as PMP 1 characteristic of the grunerite minerals that are found in the Peter Mitchell pit? PALEKAR: Yes. K. HEINRICH: I would suggest that there is a subject that hasn't been discussed, although it is of great practical importance. We frequently characterize particles by their shape, and grinding is a very common industrial process. This process will change the shapes, and the question is this: I have heard isolated statements here which range from the suggestion that a massive material on grinding acquires characteristics equal to natural 193 2063104988

fibers, to the statement that you have to be careful in grinding asbestos because it loses its properties. Could we have a discussion of what the biological implications of grinding are and how one has to handle this situation? A. LANGER: Dr. Heinrich has touched upon an extremely important problem which concerns the biological activity of small particles. The origin of the theory concerning grinding and subsequent alteration of the activity of minerals dates back some 25 years to Great Britain, to its Pneumoconiosis Research Unit. At that date, this unit boasted of having the finest laboratory of its kind in the world. They are remembered for their fine work. At that time, the pathologists in the group observed that the smaller the size of quartz particles, the more biologically active the dust was. Indeed, a 5 Nm quartz particle was relatively "inert," if you can use that word, but a 3 pm particle of the same composi- tion was a thousand times more active. A 1 pm quartz particle was a thousand times more active than the 3 pm particle, and a 0.1 pm quartz particle was yet more active. At that time this unit was interested in the interaction mechanism of the silica particles in biological systems. One such proposed mechanism involved the generation of silicic acids in tissue. These acids were thought to be the agent in the production of the response called silicosis. Production of silicic acid is enhanced as quartz is made soluble. Grinding of quartz produces a more "soluble" material. To "prove" this theory, workers ground quartz in a mortar. The ground powder was split into two equal parts. One aliquot was then washed in hydrofluoric acid and a strong alkali, removing all of the surface layers, including the Beilby layer, which is the surface disrupted layer. This disrupted layer on the surface may be demonstrated by x-ray diffraction techniques. There was x-ray line-broadening produced in the ground material, without "treatment," and a very sharp x-ray pattern generated by the material that was acid and alkali "washed." These two preparations, both quartz, were then instilled into animals. According to theory, the solubility theory, the materials which had not been "washed" should have been more active. The reverse was found to be the case. It was found that the materials that had the amorphous layers on the surface had less biological activity as compared to those materials which had been "washed." They observed the "fresh" surface to be more biologically active. This has been re-established in many experimental models. If we carry this concept into the asbestos problem, one sees the extrapolation to the different sizes of the asbestos fibers and their different biological activities. The early investigators in this field were divided into two camps. One group demonstrated biological activity with short asbestos fiber; the other group demonstrated a lack of activity. The question may be asked as to how the same animal model, the same route of administration, and the same laboratory could produce conflicting sets of data? When one examines the process by which the experimental pathologists size reduced their materials, the explanation is there. These pathologists mechanically milled these materials to shorten the fiber length. They are not only dealing with short fiber, but also with milled fiber. We have looked at these reports in the literature, dating back to the 60's, many of which indicate that milling was used to reduce fiber length. Milling of chrysotile fiber produces a material with a disrupted surface. We have observed this with x-ray diffraction and electron microscopic studies. We have taken chrysotile asbestos so prepared and have examined it by x-ray diffraction step scan technique. We've followed the line-broadening and decreased crystallinity. We've looked at this material by infrared spectroscopy for specific structural changes corresponding to different molecular groups within the struc- ture. We have examined the material in hemolytic test systems for altered membrane activity. We have looked at these materials in regard to the ability to reduce free radicals. We've looked at these milled fibers by many, many techniques and have observed that those fibers that are produced as "short" fibers show a progressive decrease in surface activity. I think that it is the preparation technique which alters the surface of the material. The experimental pathologist may indeed be working with materials that are not "truly" asbestos. The circumvention of the problem may be brought about by, instead of using mechanically milled materials, using air-jet milling, or if not air-jet milling, water sedimentation techniques to separate small fibers. Wagner's group in Penarth uses sonification methods, air-jet milling, and water fractionation to separate and collect small fibers. They produce biologically active small fibers. G. WRIGHT: The inference has been made by Dr. Langer that the experiments using short fibers have no validity because the surface has been altered by grinding. I would like to report that Dr. Kuschner and I have used contrasting fibers prepared synthetically 194

and not involving any grinding. The short fibers produced no fibrosis, but from the same batch" permitted to grow long, we got well developed, extensive pulmonary fibrosis from intratracheal injection into guinea pigs. LANGER: Several years ago we ordered synthetic chrysotile from a company in Pennsyl- vania. The materials were obtained for animal work. We examined these materials very carefully; the material was half talc and half poorly crystallized chrysotile. I think when one talks about chrysotile grown in a thermal bomb in someone's laboratory, one has got to characterize it extremely well because the crystallization process is very difficult and very often one does not produce chrysotile. I see Julie Yang here in the audience who's done a great deal of work at Johns-Manville growing chrysotile. They had to use a number of compounds to grow really good chrysotile fibers. It is extremely difficult to do. J. LEINEWEBER: I would just like to comment that the synthetic chrysotiles that were made in our laboratory were the ones referred to by Dr. Wright. I've also had the oppor- tunity to see the samples that were made by Tempress. Julie Yang can comment on the great divergence in quality between the two samples. Ours were good. I did want to say that the synthetic chrysotiles that were prepared in our laboratory were of good quality crystals and this is absolutely important. J. YANG: I worked for Johns-Manville making synthetic chrysotile. The synthetic chrysotile we made for Dr. Wright is the pure synthetic chrysotile; there was no mineralizer added. I think the electron micrograph shows the size distribution; it's all fibrous material. LANGER: Julie, didn't you use cobalt or nickel in the preparation of those materials? YANG: No, that's for a different purpose. When we put nickel or cobalt or iron into it, at that time, was for a different group of tests where we were trying to figpre out whether or not any heavy metal substitution would cause carcinogenic effects. We also prepared the pure ones with no additives. ' LANGER: I think that another important issue should be raised. There were many discussions of a number of studies in which short fibers produced no biological signs of activity. There is for every study which shows no activity, another one which does indeed show that sma11 particles are active. As a matter of fact one of the first studies of short chrysotile fiber, which is cited extensively in the literature, is probably the most unread paper in the field today (Durkan, Vorwald, and Pratt on the biological activity of small fibers). These workers were interested in fiber length as related to biological activity. At that time they were impressed with the work to come out of Great Britain demonstrating that the small silica particles were far more active than the large silica particles. They of course used various size fractionated materials of chrysotile and in their paper stated that, although they saw no "increased effect" of short fiber they reported "more limited" activity of_ the short fiber. Mineralogical analyses of the dusts used experimentally showed the "short dust material" consisted of only some 17 percent chrysotile, the rest being other materials. J. MOORE: I want to raise a question. Dr. Wright, is It possible for you to give me a reference for that work or to provide the audience with the data if it is not published? G. WRIGHT: With regard to the comment that Dr. Langer made about the work of Vorwald and others at the Saranac Lake Laboratory - I was working there at the time and, in the samples which produced fibrosis, at least five percent of the fibers were of the long, or greater than 10-micrometer, variety. In answer to the question for a reference to the work by Dr. Kuschner and myself, this has been published recently, in part, in Proceedings of an International Symposium on Inhaled Particles, IV, held at Edinburgh in September of T$7~ t s~f ed t~by Walton and-pub~Tished by- er~gamonress. N 195 ~ ~ I

W. DIXON: I would like to ask about the toxic activity of several kinds of fibers: (1) partially coated asbestos fibers, for example asbestos fibers which have an organic coating, (2) talc fibers (I have seen true talc fibers, just as fibrous looking as any asbestos fibers), (3) fibers which are intermediate between talc and anthophyllite asbestos in composition, (4) substitute mineral fibers such as wollastonite which are used in place of asbestos. EDITORS NOTE: No response was made to the above question by anyone in attendance or in writing. (CCG) P. LEBER: I was interested in the macrophage work of Dr. Palekar. Do you have any information on the mechanisms of the site of toxicity? I'm thinking particularly whether you have any information supporting the cell membrane puncture ideas of Dr. Kotin, with the release of lysozymal enzymes or any organal changes that might occur after ingestion of these particles, or whether ingestion of particles is actually necessary for cytotoxicity? L. PALEKAR: Well, the data that I presented was very preliminary and I don't want to make any conclusions. We performed some standard tests for acid phosphatose and lactate dehydrogenase and we did find release of these two enzymes into the medium as well as within the cell itself. J. KRAMER: I have two questions. The first one is addressed to the taconite study. There were various comments earlier voicing concern about the characterization of the sample. I would like to add a few additional comments. First of all, I think that you will find that there is a large variation in the composition of both the tremolites-actinolites and the cummingtonites (Bonnichsen, 1969, Mineral. Soc. Amer. Spec. Paper 2; Kramer, 1976, Canad. Mineral, 14, 91-98), and I believe that you must be aware of these variants when you characterize your sample. You may wish to determine the cell constants, and there is literature relating cell volume to composition (Finger, L. , 1967, The ~cr,, crystal structures and cr stal chemistry of ferromagnesian amphiboles, PhD thesis, Umv Minnesota . here are other factors to consider. The cummingtontites contain variable amounts of manganese, for example. There are a large number of mineralogical factors that you may wish to consider prior to your animal studies. Also I would suggest that if you look at the tailings you will be able to ascertain these mineralogical variations. My second question regards the Connecticut survey. I think that there is one assump- tion that needs careful consideration, and that is the constant relationship between fiber number and mass. If this assumption is not valid, then your mass basis is not valid. Fibers appear to have size distributions over about two orders of magnitude. Therefore, the mass can be determined by a very small percentage of the fibers. In other words, if you consider one-100 pm fiber out of 100-1 pm fibers, you change your count by only one percent, but you change your mass by a factor of five or more times. Therefore, the size distribu- tion of the largest few percentile of fibers will be most significant in your mass/fiber ratio. Why are you using a mass basis and not a count basis? L. BRUCKMAN: There are many problems in developing that envelope besides what you just said, which are obviously important. What we were trying to do was to take today's information and develop some type of standard and again try and make it such that it would not be criticized as being too strict, and while we were studying and refining the rela- tionships between dose-reponse, we'd at least have a standard. Now we have places in Connecticut which are above that level, and pretty much everybody has said that there are some problems with it, but the level looks basically reasonable and I think that it should be promulgated as a first step. It's a lot better than a no-visible-emission standard. I forgot the second part of your question. KRAMER: No, it was basically related to why you used a mass standard rather than a count standard. BRUCKMAN: At the time that we were doing our analysis, the procedures available which were basically developed by Dr. Thompson at EPA, were based on mass measurements of chryso- tile. When we went out and did our ambient survey back then, and it took some time to get it done, that was the technique that was readily available. As we continued on, in order to get comparative numbers, in other words to say whether the levels were twice as high or ~ 196 w ~ ~ r

twice as low, we continued doing the same type of analysis through Battelle. I'm probably not the, one to comment on which way is the best way to do it, but when Battelle did the work for us, their mass analysis, based on activated chrysotile samples, was ±50 percent. It's a kind of reproducible, gross measurement of the amount of asbestos in the air, but it doesn't give you any information at all about fiber count. But mass was one way of relating back to our standard. The standard could also be expressed in terms of total asbestos fibers; I believe it's 30,000 total asbestos fibers for a cubic meter of air sampled. So if you do do a number determinations, you could still relate that back to the standard. Battelle does a mass analysis and that was the way we have been doing it all along. C. COOPER: I also want to comment on Dr. Bruckman's very practical approach to an environmental problem. I'm not going to comment on the audacious assumptions that went into it, because I think he'd be the first one to say that was the case. My comments are two- fold, that is, I saw two important things. One was that the bottom line (and it was the literal bottom line in his graphs) was a probability of certain events occurring. This cut right back to the dialogue between Dr. Nicholson and Dr. Kotin yesterday afternoon. It assumed a no-threshold response; a straight line relationship, but it acknowledged that at some point that the straight-line relationship reached a probability, or a level of risk, that was very, very low. There's a great deal of difference between a 1 in 10 risk of getting something, and a I in 100 million risk. I think Dr. Bruckman at least faced up to this important question, regardless of the validity of the assumptions that went into determining the actual values. The second comment I wanted to make was that using his 30 nanogram limit, the levels of 12 and 25 did not seem particularly alarming. Since he was basing his original case on 168 hours of exposure during a week, probably what one might call a time-weighted average would be well within the 30 nanograms that was proposed. I was struck by how low these observed concentrations were, using the assumptions in scale. R. BLEIFUSS: I want to return to the Peter Mitchell mine again and a sample p'repared by IITRI for the EPA. If you read the IITRI reports, it is apparent that the samp,le site selected represents unique geological situation within the Peter Mitchell mine, in the same sense that the Reserve operation is unique on the Mesabi Range. It does not really appear to be typical of the taconite in that area. The sample represents a local segregation of a rather.unusual mineral suite and it is doubtful that we should use such a sample on health studies. I really think we should go back and provide you with a better starting material for the kind of work you are proposing. 0. MENIS: I would like to address my question to Dr. Bruckman. I appreciate the advance of this mass measurement and simplification. I just have a question about the total volume of sample in which this was determined, and what kind of weight basis that was. What was the total sample of your low volume sampler that was used to establish the 40 nanograms or 10 nanogram levels that you distinguish between borderline cases and significantly high. BRUCKMAN: If I understand you right, it's just a different type of sampling equipment that we developed for this purpose. If you wanted to get a 30-day average sample with a high-volume sampler, which only runs for one day, you'd have to collect 30 samples. Thirty samples at $500 a throw is a lot of money. MENIS: My question was, what was the total weight of the collected dust during that period of time? BRUCKMAN: We didn't do that determination, because there are problems in getting total weight with cellulose nitrate membrane filters. They are very hygroscopic and that presents a lot of difficulty, but that would not affect the amount of asbestos there. So there were no total weight measurements made, only chrysotile asbestos determinations. We don't know what the total weights were. We did do total weight for one sample. It looked like we were getting reasonable numbers, therefore we didn't continue it. M. COSSETTE: I have a comment that I'd like to address to Dr. Bruckman. One author, Mr. Rutner, has published a paper on 19 cases of mesothelioma in Switzerland. And of these, only two were related to asbestos exposure. Also, in experimental animal studies, mesothe- lioma has been produced with many other materials. In the case of your survey of mesothe- lioma in Connecticut, did you make any attempt to relate mesothelioma to anything besides asbestos? 197 2063104992

0 BRUCKMAN: We only did a very preliminary study based an hospital records and death certificates. We'd like to get some money to do a detailed epidemiological study, a complete case history, occupational exposure, and whether these were relatives of people who worked in asbestos industries. We aren't able to do that. We haven't got any funds at all to do any of these studies, and it's impossible to carry them on without funding. We just haven't been able to get into it. Hopefully, the data that I reported on concerning mesothelioma incidence will be updated. My study was only up to 1972. It should be updated and maybe other types of potential causes, like fiberglass exposure or something like that, will come out of this. COSSETTE: Thank you. The data that you showed indicates that the number of inesothe- lioma cases has gone up dramatically in the last few years. Do you think this may be partially due to the fact that it's more easily found now, that we have better determina- tion techniques. BRUCRMAN: I think that's definitely a contributing factor. I would say yes. M. ROBERTS: A question for Dr. Palekar: Going back to the presentation of the slides, the slide with the 1 pm scale showed an electron micrograph of ambient air at the process plant and at the mine, as compared to the slide with the 10 Nm scale showing the preparation, that you have apparently prepared for your inhalation studies. On the slide from the ambient air at the plant and mine there was very few fibers more than 1 pm long, which was the scale shown on that slide, whereas the second preparation, on the 10 Nm slide, showed considerable material that was over 10 to 15 pm. You have replied to a previous question that the rock selected to be used in your preparation was representative, and I would like to ask how this was selected? Can you give a complete history as to the location and selection of this material? Further, if these studies are to reflect the pulmonary response of exposure to the dust from these ores, should not the rock be prepared from a blind selection of typical mine ore? The principal question here is how the sample was selected, and can you give some detailed history of where and how this was selected? PALEKAR: The purpose of this study was to evaluate the biological effects of the fibers which were emitted in the taconite mine. The air samples were procured from the mine area and the processing areas. Several samples were collected on filter papers and proper size distributions were made. I can understand the confusion here between the size distribution tables presented and the electron micrographs, and I would like to emphasize again that the electron micrographs are not truly representative of size. The tables presented are more accurate. Quite a few fibers were counted, and I think that the fiber size that I presented in those tables are more representative. Now, originally we selected air samples and characterized them, then we went back to the rocks. Several rocks were collected, about 50 or so, out of which we selected two rocks which represented the air samples and the processing area, as well as in the mining areas. We are studying the biological properties of these two samples. Currently we are not doing inhalation studies, we are doing intratracheal studies and intrapleural studies. In the future we plan to do inhalation studies. LANGER: I wonder if I could add something to this. I think that everyone is missing a very obvious point: It appears that the regulatory agencies operate in a "management by crisis" mode, and everytime some new material is dumped into a lake or a river or is thrown into the air, a few million dollars is then invested in investigating the biological activity of that particular substance. It is the consensus of workers in the field that something should be known concerning the properties of fibers in terms of the mechanisms of interaction. Whether or not one could get pure Peter Mitchell pit fiber, whatever that is, is an academic point. There are many lithologies in this mine, as described in Gunderson and Schwart and the Seven French monographs. Whether a "representative" fiber exists is probably unlikely. It was then decided that the Environmental Protection Agency should investigate a fibrous rock-forming silicate which was not asbestos per se. The materials which were fibrous and "pure," yet not exactly characteristic of the cummingtonitel grunnerite within the Peter Mitchell pit, occurred in localized veins. They were fibrous on a megascopic level and when comminuted they resembled asbestos fibers. But they were not asbestos per se. These were rock-forming fibrous amphiboles. I think that if these 198

materials induce changes in biological test systems, then we shall go further and investi- gate'-others. We must know something about the mechanisms of interaction. If they are not active, then everything else is academic. SCHNEIDER,MAN: RECAP OF SESSION. The session yesterday afternoon seemed to me vigorous and active and ended on quite a high note. This morning's session is a continuation of that, and in view of the speakers we have this will be at least as exciting and as inter- esting as yesterday's session. At this time I would like to give you a very short summary of what I thought happened yesterday. In the instructions that were given to the Chairmen, we were asked to summarize what things people agreed on, what things were learned or said or now accepted as fact, what things were questioned, and where further work should be done. I made some notes on this during the course of the day and I made some notes yesterday evening after having gone out to Wolftrap to hear the Preservation Hall "Jazz Band." The last number they always play is "As the Saints Go Marching In" and I think that if anyone tries to tell you what people fully agree on he has to be a saint, or as you know, fools walk is where angels fear to tread. I'm going to be foolish and try to tell you what people agreed upon. But as I looked at my list, I discovered that that list of agreements was really quite small, and my list of disagreements was quite long and, therefore, the list of further work to be done is even longer. Any of you that are involved in the funding agencies, I want you to hear that work to be done is quite long. It seemed to me, in the agreements, from the notes I have for myself, are that asbestos, whatever it might be, in many of its subclasses and subdivisions, whatever they are called, is a material which can have adverse health effects. We talked a lot about the carcinogenic effects and talked about how some of these might be different or have less intensity for certain forms of this mineral than for others. The discussions were tempered by the fact that some people said what looked like very sharp differences in the past don't look like such sharp differences any longer, and these materials have effects that now appear to be closer to eaGi other. All through that, there was an undercurrent that we really don't know this because;we have great problems of determining doses to which people were exposed. There was also the under- current, although a great deal of emphasis was on cancer, that there are other health effects, and we have to talk about those. There were questions during the day, as you may recall, as to whether the cancer effects were dependent upon some of these other effects having occurred. Whether these were independent, or whether these ran parallel with each other. Do you have to have hyperplasia, for example, as a necessary component? Was it a pre-cancerous condition? The major questions that people raised during the course of the afternoon were questions concerning two things: first, questions concerning particle size. What are the particle size variables with respect to health effects? What are the particle sizes necessary in order to produce health effects? Are there particle sizes that are safe? Are there particles that don't produce these kinds of effects? To address themselves to these questions, Dr. Bignon of Paris showed us information on distribution of particle size found in the lungs and tissues of individuals with various diseases associated with asbestos and showed for us - at least in the trapped particles, the remaining particles, the particles that are still there - a tremendous overlap of the particle size in persons with illness and persons without illness. This is not necessarily indicating that these particle sizes that he found (by the way you will recall he found rather smaller particle sizes than most people have indicated) were necessary to induce certain of these illnesses. He made it clear, this was not to say that these smaller particles were the ones that induce the illness. It may very well be these were the only ones that remained, these were the ones that were trapped, but that is what he found. Dr. Kotin, in a rather elegant lecture that he labeled as a kind of lecture in pathology that one would give to sophomore medical students (I rather think it was more elegant than one would give to sophomore medical students, having taught sophomore medical students myself), gave us a lovely theo- retical discussion of physiology of the lung and a lovely theoretical discussion on what might be going on in the pathogenesis of illness induced by, supported by, and/or stimulated by asbestos particles. Or. Kotin remarked that he would attempt to be controversial; he succeeded at least in asserting the existence of thresholds, with which, as you know, there is a great deal of difference of opinion. He in turn was challenged on this by Dr. Nicholson, who had earlier presented data showing relatively very low levels of exposure. He was also challenged by Dr. Sunderlin of Canada and also a gentleman from the State of Maryland. The discussion, seemed to me, at one point got really highly theoretical, and I think Dr. Kotin and other people indicated that there would certainly be a need for a full scale discussion of this issue. There was one, by the way, in 199 2063104994

Heidelburg last year; a whole meeting devoted to the problems of threshold. In relation to problems of particle size, Dr. Stanton and his colleague Dr. Layard described certain experiments they had done to see whether the carcinogenic effect that we found in these various materials was a carcinogenic effect peculiar to asbestos or whether it was an effect one would get from any particles of that size and of the same dimensions. The animal studies that Dr. Stanton described would seem to indicate that the very long, thin particles, longer than 8 N(I did not make a note of the diameter), but long thin particles, were the most carcinogenic. Stanton very carefully, it seemed to me, said these are (in answers to questions) the most carcinogenic, but he did not find a line below which you find materials which are not carcinogenic, that you could be certain that they are not carcinogenic. He said no he could not find such a line. It was just that these were more carcinogenic than others; the carcinogenicity fell off as the particles got shorter and stubbier, but he did not find any sharp line of of demarcation. Now, this is a problem for the regulatory agencies because they have set a measure relating to the size of the particle. There was then discussion concerning the sort of thing that Stanton had done, because he installs these particles where they can have their effect. People raised many questions about asbestos in the ambient air and problems that would be associated with such things as ubiquitous asbestos, most of which are smaller particles than the ones people are industrially exposed to. The questioning addressed - what about inhalation studies? The remark was made that with very few exceptions, the inhalation studies were not particularly well done. A nice reference was made to Dr. Gross saying that his studies were well done, and the remark further carried that the inhalation studies had not shown the same sorts of effects as the installation studies had shown. This bring to my mind the similar problem we have with tobacco carcinogeneSis, where again in the inhalation studies, unless done in some very peculiar way, by slitting the trachea in the neck of the dog and having the dog smoke through the slit, nobody has produced, so far as I know, lung cancers in any of the experimental animals. So the inhalation studies still have some serious difficulties with them. A question was raised by Dr. Ross, a geologist, about these ambient materials and the problems that strict standards would raise for small businesses. I think Dr. Ross' hope is that one could establish that there were particles sizes or materials or levels that were in some sense absolutely safe. These economic problems might not be loaded on the small businesses. It seemed to me what we had was a general agreement on the carcinogenesis of these materials, and their capability of causing other illnesses and a very large set of statements of all kinds of things we just don't know, and all kinds of things that we still need to have some work on. I have tried to list those for you. 200

C3 National Bureau of Standards Special Publication 506. Proceedings of the Workshop on Asbestos: Definitions and Measurement Methods held at NBS, Gaithersburg, MD, July 18-20, 1977. (November 1978) IDENTIFICATION OF SELECTED SILICATE MINERALS AND THEIR ASBESTIFORM VARIETIES William J. Campbell Particulate Mineralogy Unit, Bureau of Mines College Park Metallurgy Research Center College Park, Maryland 20740 Abstract The problem of asbestiform particulates with its environmental and health implications has been compounded by the lack of precision with which the term "asbestos" has been used. In many instances, non- asbestiform mineral particles have been identified as microscopic fibers of asbestos-related minerals. This lack of precision in identifying these particulates not only works to the disadvantage of the minerals industry, but is also a handicap to rational science-based decision making by regulatory agencies. This presentation summarizes methods and terminology suggested by the Bureau of Mines for the identification and characterization of asbestiform minerals and also sharpens the distinction between common serpentine and amphibole minerals and their relatively rare asbestiform varieties.l The continuing effort of the Bureau's Particulate Mineral- ogy Unit is to characterize mineral particles by morphological, compositional, and structural data using various instrumental analytical techniques and by developing new methods for identification and characterization. Key Words: Asbestos; cleavage fragments; fibers; silicate minerals. Introduction The objective of this paper is to present a general introduction on the identification and characterization of asbestos-related minerals. Detailed discussions of specific analytical techniques are given in other papers presented at this workshop. At present there are three types of identification-characterization to supply the needs of regulatory agencies, medical researchers, and mineral scientists. It is hoped that through inter- actions such as this workshop a common mineralogical-based procedure can be developed that meets the needs of all concerned groups. Until recently, emphasis in the United States was placed on occupational exposure of employees manufacturing or using asbestos products for insulation and other applications. Regulatory procedures were adopted from those used in Great Britain. The industrial- hygiene identification procedures were acceptable to industry, health, and regulatory organizations because the concern was restricted to several mineral products known collectively as asbestos. Although light optical microscopic procedures counted only the larger particles collected on the air filters, the procedure was adequate for correlating IThis paper is an abbreviated version of the sections on mineral identification and charac- terization in Bureau of Mines Information Circular 8751 - Selected Silicate Minerals and Their Asbestiform Varieties: Definitions and Identification-Characterization, 1977, 56 pp, authored by W. J. Campbell, R. L. Blake, L. L. Brown, E. E. Cather, and J. J. Sjoberg. Copies of IC 8751 are available upon request to W. J. Campbell. 201 2063104996

health effects to the number of fibers observed. Exact definitions for asbestos-related mineralogical terms were essential since all three groups (industry, health, and regulatory) clearly understood what was being counted and regulated. The light optical microscopic procedures used by industrial hygienists were designed for control of asbestos-processing operations in which the chrysotile and asbestiform amphiboles are present as bundles of fibers as well as individual fibers [1]2. These bundles may have an average diameter of 0.75 to 1.5 pm for chrysotile and 1.5 to 4.0 pm for the amphibole asbestos [2]. Particulates of these sizes can be readily observed at a magnification of X 450 to X 500. In contrast, samples from ambient air and personnel air monitors may consist of individual fibrils or small bundles of chrysotile 0.02 to 0.1 pm in diameter, and/or amphiboles 0.1 to 0.2 pm in diameter [3]. Fibrils and small fibers in this size range are not visible with the conventional light optical microscopic procedures. Therefore, the identificaiton procedures currently used for regulating the U. S. mineral producing and consuming industries must be reexamined to insure that they are both mineralogically correct and applicable to the size range of the particles being regulated. This discussion will be limited to the selected silicate minerals and their asbestiform varieties listed in Table 1. The objective is to point out the particle size at which the minerals can be identified and characterized by various analytical techniques [4]. Oetailed descriptions of the various analytical and characterization techniques are available in numerous publications and textbooks. Table 1. Selected silicate minerals and their asbestiform varieties. Mineral Asbestiform variety AMPHIBOLE GROUP Anthophyllite: Anthophyllite asbestos. (Mg, Fe+2)7 Sia022(OH,F)2 Cunmingtonite-grunerite: Cummingtonite-grunerite asbestos. (Mg,Fe+2)7 SiB022(OH)2 Tremolite-actinolite: Tremolite-actinoiite asbestos. Ca2(Mg,Fe+2)5 SiB022(OH,F)2 Riebeckite: Crocidolite. Na2Fe32 FeZ3 Sia022(OH,F)2 SERPENTINE GROUP Serpentine: Chrysotile. M96S14070(OH)8 2Figures in brackets indicate the literature references at the end of this paper. 202

(C I A crystalline mineral is defined primarily by its crystal structure and by its definite composition or range of compositions. Therefore, any system of mineral identification should be based principally on crystal structure and chemical criteria. Additional characteristics have to be determined to distinguish varieties. These varieties have similar basic. crystal structures and composition, but are usually differentiated macroscopically by the characteristic habits and/or other specific features of the varieties. The objective is to summarize the methodology for identifying the mineral first by mineral group (such as serpentine and amphibole), then by mineral (actinolite, anthophyllite, or chrysotile), and finally by mineral variety. Macroscopic Samples At the macroscopic level (easily visible by the unaided eye), the obvious feature of the asbestiform varieties is the presence of fibers that can be easily separated, while the nonasbestiform varieties have a massive, blocky, bladed, or columnar appearance. Although chrysotile does occur very rarely in a nonasbestiform habit, in general the distinction between chrysotile and serpentine can be based on the presence or absence of separable fibers. In some serpentine samples where an obvious asbestos texture is not displayed, the distinction between serpentine varieties may require more specialized techniques [5,6]. The distinction between serpentine and amphibole minerals at the macroscopic level can be made by elemental analysis, differential thermal analysis, and x-ray diffraction techniques. For essentially pure samples, these techniques should also be sufficient to identify the individual amphibole minerals based on the elemental composi- tion corresponding to the various members of the solid solution series. Many macroscopic samples of interest to the occupational and environmental health personnel may contain low percentages of asbestiform minerals (for example, chryfotile in serpentine and tremolite asbestos in talc). As a supplement to optical microscopy, the presence or absence of serpentine or amphibole minerals can be determined in 10- ~0 100-mg samples by instrumental techniques such as x-ray diffraction, differential' thermal analysis, or infrared spectrophotometry. In general, the sensitivity of these instrumental methods is approximately 1.0 weight-percent. Sensitivity is significantly affected by the presence of other minerals that give a response at or near the response peak of the serpentine and amphibole minerals. It is important to note that these methods usually only distinguish between mineral groups; light optical or electron optical microscopy is required to obtain morphological characteristics necessary to identify varieties of the same material. Chemical characterization is generally necessary to assign a specific mineral name to an amphibole whose structure is known. The amphiboles have been described [7] using ~he structural formula Wo_ix2YsZa022(OH,O,F)2. Generally, W = Na, K; X = Na, Ca, Mg, Fe 2, Mn; Y= Al, Fe 3, Ti; and Z = Si, Al. In addition to the variation implied by the structural formula, a chemical analysis must take into account inclusions of other minerals that may be present. In contrast to the more formidable task of chemical characterization of amphiboles, the serpentine minerals generally show little deviation from the formula Mg3SiZ0s(OH).. For either structural or chemical characterization of a macroscopic sample, sufficient time must be spent in sample preparation to insure that relatively pure minerals are being examined. Microscopic Samples The petrographic microscope provides a general method by which particles larger than 5 pm can be characterized. By observing the optical properties characteristic of the structure and chemistry of a mineral, an experienced microscopist can distinguish amphiboles from serpentines and, in some cases, distinguish individual minerals within these groups [8]. The refractive indices are sufficiently different for the serpentine and amphibole groups to make a distinction between groups by using the appropriate index oil (Table 2). There is significant overlap in the range of the three refractive indices among the amphiboles, but a specific index (for example, a, P, or y) can be determined to aid in identifying the amphibole species. Optical relationships can be confused, however, if the particle consists of fiber bundles or is some other form of crystalline aggregate. 203 2063104998

f• a Table 2. Refractive indices for the serpentine group and selected amphibole minerals. Refractive index Range of values Chrysotile a 1.493 - 1.560 0 1.504 - 1.550 y 1.517 - 1.562 Antigorite-lizardite a 1.538 - 1.564 y 1.546 - 1.573 Anthophyllite a 1.596 - 1.652 S 1.605 - 1.662 y 1.615 - 1.676 Actinolite-tremolite a 1.599 - 1.668 0 1.612 - 1.680 y 1.622 - 1.688 Cummingtonite-grunerite a 1.635 - 1.696 p 1.644 - 1.709 y 1.655 - 1.729 Riebeckite a 1.654 - 1.701 ~ 1.662 - 1.711 y 1.668 - 1.717 The well-known parallel extinction of the commerciai asbestos known as Amosite can be used to distinguish that variety from the nonasbestiform varieties of cummingtonite and actinolite. A method of using extinction angles and cleavage directions to distinguish specific asbestiform and nonasbestiform amphiboles has been described [97; however, this technique is limited to particles with diameters greater than about 5 Nm and cannot be universally applied to all amphiboles. There are many other optical parameters such as pleochroism, sign of the elongation, and color that are easy to obtain. Other parameters such as optic axial angle, optical orientation, and optic sign are relatively more difficult to obtain. Except for the asbestiform variety, serpentines are usually massive, while amphiboles range from fine-grained massive to columnar or radiating aggregates of prismatic or acicular crystals. Amphiboles in acicular habit may appear to grade into the asbestiform varieties. The characteristic features of this habit may still be seen by electron microscopy. Terms such as "acicular" or "prismatic" may still be applied when seen, but the term "asbestiform" begins to lose its usefulness. For example, how may flexibility be demonstrated in a 2-pm bundle of fibers? As particle size decreases, the inability to manipulate the mineral grains restricts the use of the term "asbestiform" without altering the original sense of the word. High magnification necessitates the use of strictly dimensional terms such as size and aspect ratios to accurately describe the morphology of the amphiboles and serpentines. The degree of morphologic characterization possibly will depend on the magnification being used. An asbestos particle being described as a single fiber at low magnification may be seen to be a bundle of fibers at some high magnification. Therefore, the magnification must be stated in the description. Morphologic characterization using light microscopy can be accomplished on particles as small as a few micrometers. Electron optics can be used to characterize a wide range of sizes extending down to a few angstroms. Morpholo4ic characterization alone will not identify a mineral without supplemental structural or chemical data. Structural information on individual particulates can be obtained by use of a transmission electron microscope (TEM) in the selective area electron diffraction mode (SAED). The inclination of the single crystal fragments to the electron beam is very critical since a slight tilt of the crystal may change a relatively simple reciprocal 204

lattice pattern into a very complex one. Consequently, a special goniometer or tilting stage•-..is necessary to obtain easily interpretable diffraction patterns. For the identification of the mineral, a goniometer or tilting stage is even more essential since dependable conclusions cannot be made from measurements on one reciprocal lattice plane. The quality of the SAED pattern is a function of fiber diameter. The larger diameter fibers (>0.5 pm) strongly absorb the 60- to 100-keV electrons used in a conventional TEM, while the very small-diameter fibers (<0.2 pm) do not give sufficient electron-diffraction intensity. A second problem with small-diameter fibers is the degradation of the single- crystal pattern by diffraction lines from nearby particles. A higher energy TEM, with the resultant greater penetration of the electron beam, can be utilized for large-diameter particles. However, these costly instruments are not widely available. Although the magnitude of the characteristic C, the distance between the conspicuous layer lines for chrysotile and the amphiboles, is similar in direct space (dool - 5.3A), the chrysotile pattern has very prominent streaks on these layer lines compared with the spot pattern for the amphiboles [10]. Researchers indicate that the ability to distinguish between the fibrous and nonfibrous variety of amphibotes by SAED is still to be resolved. At the very high magnification available with a TEM, chrysotile's hollow-tube (scroll-like) structure, approximately 5 nm in diameter, is visible (fig. 1). This hollow-tube structure, together with chemical and structural data regarding the sample, is sufficient to identify the mineral variety. However, the hollow-tube structure is only visible for individual fibrils; fibers (composed of several fibrils) will not display this characteristic because of stacking of the fibrils. Figure 1. Chrysotile, showing individual fibrils, at two magnifications: X 18,000 (left) and X 35,000 (right). The hollow-tube structure is visible at the higher magnification. tTEM microphotographs.) 205 N O a w ... 0 w 0 0 0

The elemental composition of microscopic grains is determined by either wavelength or energy-dispersive x-ray spectrography in conjunction with scanning or transmission electron microscopy. Extreme care must be taken in the calculation of elemental con- centrations from x-ray spectral intensities because the spectral line intensities (FeKa, MgKa, CaKu, relative to SiKn) are dependent on particle diameter for small fibers [3]. Energy-dispersive x-ray spectral calibration data for each scanning or transmission electron microscope must be made using relatively pure standard minerals analyzed by accepted chemical-instrumental techniques. The analyst should be aware that other nearby grains may be contributing to the characteristic x-ray lines because of either penetration of the electron beam through the particles or secondary excitation of nearby particles from primary x-rays generated in the particle being measured. Modern electron optical instruments have electron beam diameters of 0.1 to 0.01 pm; however, the sphere of excitation can be several micrometers in diameter as a result of scattered electrons and primary x-rays generated in this particle. Conversion of intensity into concentration using accepted computer programs such as "MAGIC" is limited in accuracy because these programs are designed for use with grains or particles several micrometers in diameter or larger, whereas the average mineral fiber diameter is less than 0.5 pm for chrysotile. A good example is the diameter size distribution of chrysotile fibers in ambient air samples (Table 3). The important point to note is that approximately 95 percent of these chrysotile fibers are 0.12 pm or less in diameter. Therefore, quantitative correction procedures applicable to large particles will be of limited value in mineral-fiber identification because the relative x-ray spectral intensities are dependent on fiber diameter below 0.2 pm. Table 3. Frequency distribution of the lidth of chrysotile fibers in ambient-air samples, percent. Diameter of chrysotile - - - - - Sample----- fibers, pm 1 2 3 4 5 6 0.02 -<0.04 10 70 57 17 15 17 0.04 -<0.06 47 24 28 29 33 49 0.06 -<0.08 24 5 8 28 20 15 0.08 -<0.10 14 1 2 12 26 6 0.10 -<0.12 2 0 1 7 3 6 0.12 -<0.14 0 0 2 3 1 1 0.14 -<0.15 1 0 1 2 1 1 0.16 -<0.18 0 0 0 1 0 1 0.18 -<0.20 0 0 0 0 1 1 0.20 -<0.22 1 0 0 0 0 1 0.22 - 0.24 0 0 1 0 0 1 >0.24 1 0 0 1 0 1 a Samples were collected 1-2 miles from a serpentine rock quarry. Another problem with the elemental characterization of very small particles is the poor signal-to-background ratio. Longer counting times will help to improve the reliability of the measurement, but the best approach is to minimize the continuum background resulting from the interaction of the electron beam and the sample substrate. 206

Applying Mineral Terminology to the Identification and Characterization of Particulates This section addresses the practical considerations and limitations encountered when applying nomenclature and identification-characterization procedures to regulatory and environmental samples. Applying Morphological Terminology One of the obvious features of minerals and their particulates is their morphology or shape. The need for precise definitions of terms such as "asbestiform," "fiber," "cleavage fragment," and "fibril" was explained in IC 8751. These definitions were carefully structured to eliminate ambiguity and to be technically correct. Applying the definitions to samples requires careful thought as to what limits must be placed on interpretations resulting from the use of these terms and other mineralogical concepts. The underlying problem, recognized by both medical and regulatory personnel, is clas- sifying the mineral particle as the asbestiform or nonasbestiform variety. In a mineralogical sense, the source of the mineral particulates must be considered, as explained in the following discussion. Particulates From A Known Asbestiform Serpentine or Amphibole Source The definition of asbestiform minerals includes three aspects: morphology, structure, and chemistry. Morphologically, asbestiform mineral varieties separate into flexible fibers or flexible bundles of fibers. Flexible fibers bend readily and only break across the fibers into distinct pieces with some difficulty. Structurally, the asbestiform minerals are limited, in common practice, to the serpentine and amphibole mineral groups. Chemically, these minerals are all hydroxylated silicates; the term "hydroxylited" is preferred over "hydrated" because these minerals contain OH ions rather than water of crystallization. The serpentines contain approximately 13 weight-percent waCer; the amphiboles, approximately 2.5 weight-percent water. For the purpose of this discussion, assume that a hand specimen meeting these requirements is correctly identified as an asbestiform mineral. If this sample is crushed and its fragments examined at various magnifications, its fibrous nature would be apparent. These elongated fragments would be termed "fibers" and "bundles of fibers," and with the other available information would be called "asbestiform." As these asbestiform particles are examined at increasing magnification, smaller particles become visible, while the image of large fibers and fiber bundles may exceed the field of the microscope. At increasingly smaller sizes, while fibers or bundles of fibers are still the predominant shape, a few of the fibers are observed to have broken into shorter and shorter segments. These very short fiber segments are no longer described as fibers, but would be classified as fragments of fibers, or cleavage fragments if one or more cleavage planes govern their shape. Therefore, a known asbestiform sample would show an increase in the ratio of fiber fragments to fibers with a decrease.in particle size. Particulates From A Known Nonasbestiform Serpentine or Amphibole Source If the hand specimen discussed previously does not separate into flexible fibers or bundles of fibers, the mineral would not be considered asbestiform. However, the specimen would be classified as serpentine or amphibole if the specific mineral is identified on the basis of optical properties, chemistry, and structure. If crushed fragments of this known nonasbestiform mineral are examined at various magnifications, the particles would be primarily cleavage fragments, or irregularly broken fragments if cleavage does not govern breakage. However, a few elongated particles may resemble a fiber in appearance to the degree that they may be indistinguishable morphologically from fibers derived from an asbestiform mineral sample. What can be stated morphologically about particles derived from crushing a known nonasbestiform mineral is that most of the particles are cleavage fragments with non- asbestiform texture; a few are fibrous in appearance, particularly at low magnification; and all of the particles are known to be derived from a nonasbestiform source. 207 2063105002

Comparison of Particulates From Known Serpentine and Amphibole Minerals and Their Asbestiform Varieties The appearance of particles generated by milling known serpentine and amphibole minerals and their asbestiform varieties is shown in figures 2 to 5. The samples shown in figures 2 to 4 were photographed using light optical microscopy at three magnifications to show that, at decreasing size (depicted by increasing magnification), the original habit generally persists. For the nonasbestiform amphibole minerals, there were a few elongated particles from the riebeckite and tremolite. Elongated particles of this type are typical of the prismatic cleavage of amphiboles. To increase optical contrast, the serpentine group samples were dispersed in an immersion oil considerably below the refractive indices for the serpentine. Figure 2. Light optical photomicrographs of chrysotile and antigorite-lizardite at three magnifications. Chrysotile (left) at A, X 100; B, X 500; and C. X 950. Antigorite-lizardite (right) at 0, X-100; E, X 500; and F. X 950. 208

.4u Figure 3. Light optical photomicrographs of crocidolite and riebeckite at three magnifications; Crocidolite (left) at A, X 100; 8, N 500; and C, X 950. Riebeckite (right) at D, X 100; E, X 500; and F, X 950. N 209 w r 0 ~ ~ ~ C 3

Figure 4. Light optical photamicrographs of tremolite asbestos and tremolite at three magnifications. Tremolite asbestos left) at A, X 100; B, X 500; and C, X 950. Tremolite right) at D, X 100: E, X 500; and F, X 950. 210

Figure 5. SEM photomicrographs of crocidolite and riebeckite at three magnifications: Crocidolite (left) at A. X 500; B, X 2,500; and C, X 10,000. Riebeckite (right) at D, X 500; E, X 2,500; and ~, X 10,000. Rectangles indicate the area shown at the next higher magnifications. 211

Riebeckite and crocidolite particles are compared at higher magnifications in figure 5. The outlined areas in the scanning electron micrographs indicate the area displayed at the next higher magnification. Again, note the presence of a few elongated cleavage fragments of riebeckite visible at the higher magnification. In contrast, the aspect ratio of the crocidolite will decrease with decreasing particle size because the individual fibers cannot cleave further along the fiber axis; they can only break into shorter segments. Aspect Ratio Existing regulatory standards are based on counting specific mineral particulates with aspect ratios of 3 to I or greater. The aspect ratio has little mineralogical significance for individual particulates but is applicable statistically to a large number of particles. A few relatively long thin particles are produced as cleavage fragments from the crushing and grinding of many nonasbestiform minerals. Conversely, similar milling treatment will result in a few short segments of true fibers from the asbestiform varieties. However, statistically, the length-to-width characteristics of the milled amphiboles and serpentine and their asbestiform varieties are significantly distinct, as shown by the data in figures 6 to 9. 1:1 3:1 MILLED AN71iOPHYWTE -Anthophyllite --MthophyHRe aabeetae 5:1 10:1 20:1 50:1 ASPECT RATIO 100:1 200:1 500:1 .a.M a Yr M N1Y O Figure 6. Frequency polygons for the aspect ratios of anthophyllite and anthophyllite asbestos. N O a7 1.+ 212 O $ v

C3 70 f 60 50 E 40 r ~ 30 ~ 20 10 MILLED TREMOLITE -Tremoiite --Tremolite asbestos Figure 7. Frequency polygons for the aspect ratios of tremolite and tremolite asbestos. 1:1 3:1 5:1 10:1 ~ 20:1 50:1 100:1 200:1 ASPECT RATIO a..w c. u~. H ie.w. v MILLED HORNBLENDE Figure 8. Frequency polygons for the aspect ratio of hornblende. 5:1 10:1 20:1 50:1 100:1 ASPECT RATIO x .ac. o 213

CHRYSOTILE Commercial milled chrysotile ,_A~ / ~ --Ambient air sample 1:1 3:1 6:1 10:1 20:1 50:1 ASPECT RATIO 100:1 200:1 500:1 1000:1 «Me.u er ..w. N10Wp Figure 9. Frequency polygons for the aspect ratios of commercial-grade chrysotile and chrysotile in ambient air. Figures 6, 7, and 8 show the frequency polygons of the aspect ratio distribution for milled samples of the normal nonasbestiform variety of three amphiboles--anthophyllite, tremolite, and hornblende, respectively. Note that in all three examples, approximately 70 percent of the particles have an aspect ratio of less than 3 to 1, and 95 percent of the particles have a length-to-width ratio of less than 10 to 1. The frequency distri- bution maxima of the aspect ratios for milled anthophyllite asbestos and tremolite asbestos are significantly higher than those for the normal, nonasbestiform variety. Thirty to forty percent of the asbestiform particulates are in the 10-to-l-or-langer class, with a significant number of particles having an aspect ratio greater than 20 to 1. Figure 9 shows the distribution frequencies for a milled commercial grade of chrysotile asbestos and for chrysotile particulates collected on ambient air filters in the vicinity of a serpentine rock quarry. For the commercial-grade chrysotile, over 50 percent of the particles have an aspect ratio greater than 50 to 1, whereas the frequency distribution for the ambient air sample has a maximum between 10 to I and 20 to 1. These results are anticipated because the higher aspect ratios for the commercial-grade chrysotile are characteristic of the significantly longer starting material. All of the aforementioned samples except the ambient air were milled, then dispersed in water for collection on a suitable substrate. The samples were then measured using electron microscopy at magnifications of 5,000 to 10,000. The ambient air sample, collected near a serpentine rock quarry, was measured using a TEM with magnifications of X 5,000 to X 32,000. Based on these data, one test for distinguishing the presence or absence of the asbestiform variety of a mineral could be an examination of the frequency distribution of the aspect ratio for that mineral. Assuming positive Identification of the mineral type, then the designation of variety would be based both on particle morphology and the frequency maximum of the aspect ratio. Cleavage fragments will generally have a frequency maximum less than 3 to 1, whereas the asbestiform varieties will fall between 10 to 1 and 20 to 1 or higher, depending on the characteristics of the mineral and the history of the sample, particularly the type and degree of milling. If a sha e or size limits are pl aced on characterizing mineral articulates, such limits shobe based on medicai evidence or on some limitation of the charing techm Que ang -sostated. 214

1C 3 Particulates From Unknown Sources Samples such as environmental airborne or waterborne mineral particulates collected at a considerable distance from a possible source are examples of particulates from an unknown source. The samples could have been collected at a location so distant from a known source that other mineral particulates originating from other sources compose most of the sample. The source of the particulates in an environmental sample may be located by taking additional samples at selected intervals in the direction of, and closer to, the suspected source. However, several factors must be considered: The direction of air and water currents with respect to the suspected source, and the proximity to and direction of other sources with regard to the suspected source. One study found very low concentration of airborne chrysotile upwind from a source compared with a concentration two orders of magnitude greater downwind [11]. Another important consideration is the level of natural or human disturbances of particulates; for example, strong versus weak winds, or heavy versus light vehicle traffic. In some instances, it may be possible to identify the source if the mineral particulates of interest have unique trace elements or combinations of elements that are specific to the probable mining or milling operation emitting the particulates. Detailed elemental analysis using the X-ray spectral capabilities of an SEM or TEM is required on both the suspected source and the particulates. Applications The following examples illustrate the application of mineral terminology and identification-characterization procedures to three types of problems: (1) chrysotile determination in ambient-air samples collected near a serpentine rock quarry, (2) iden- tification of asbestiform minerals in ceilings and walls of public buildings,'sand (3) characterization of a mineral product. These examples illustrate, in order, the *need for higher magnification than available with the light optical microscope, the use of~ various characterization techniques to screen and identify asbestiform minerals, and the 'fudgment of the analyst in distinguishing cleavage fragments and asbestiform particles. Ambient-Air Samples Collected Near Serpentine Rock Quarry The Bureau of Mines is working with State and Federal officials to measure mineral particulates in ambient-air samples collected in the vicinity of a serpentine rock quarry. Optical microscopic procedures at about X 500 are limited to the identification of mineral particulates longer than 5 pm with an aspect ratio of 3 to 1 or larger (criteria set by the Mining Enforcement and Safety Administration and the Occupational Safety and Health Administration). The mineralogist can further identify the particles as belonging to the serpentine, amphibole, or other mineral group with index oils (Table 2). The serpentine rock in the quarry is interlaced with small veins of chrysotile (figure 10). Optical microscopic procedures used for industrial hygiene are adequate for the detection of large chrysotile fiber bundles. These fiber bundles of commercial-grade chrysotile can be several micrometers or larger in diameter. In contrast, the mining and crushing operations in the quarry plus transport of particulates over a distance breaks bundles of fibers down to fibers or fibrils with diameters of 250 to 1,000A (Table 3). 215

Figure 10. Macrophotograph showing chrysotile veins in serpentine rock (X 1). Figure 11 is a series of SEM photomicrographs of a mixture of chrysotile and non- asbestiform serpentine handpicked from a small vein in the serpentine rock quarry. Note that at X 450 (corresponding to the optical microscope magnification), only one or two bundles of chrysotile are faintly visible; the predominant particles are the nonasbestiform serpentine. As the magnification is increased, the high concentration of chrysotile fibers becomes readily visible. The fiber diameter size data in Table 3 indicate that more than 95 percent of the chrysotile fibers in these ambient-air samples are below the limit of resolution of the optical microscope. Although many other scientists have pointed out the limitation of the o tical ~rocedures for chr s~otile in ambi~'ent a)r, there is neetc for cont~em ~has that higher magnifc-ation techniques are necessary for env7ronmental and reg~u ato~ry samples. 216

vl-f , . T~.,--; ---,~-~ r . . U . ..i f +F~" .~ I ~ ~ 0~ +! .. , .. r~ ;• ~ :~ 'A ~ ~ Figure 11. Mixture of nonasbestiform serpentine and chrysotile at five magnifications: A. X 450; 8, X 2,250; C, X 1,800; D, X 9,000; and E, X 18,000. Rectangles indicate t~ie area shown in the next panel. 217

Asbestos in Ceiling and Wall Materials A possible environmental hazard is the release of asbestos from ceilings and walls in homes, churches, schools, and various other public and commercial buildings. Because of the very high number of potential samples to be examined by various State or Federal agencies, a rapid and reliable screening procedure is necessary to identify those samples that warrant further test. Three complementary analytical methods for screening, identifi- cation, and semi-quantitative estimate of the asbestiform mineral concentration are x-ray diffractometry, differential thermal analysis, and microscopy (light optical and scanning electron). The screening Identification procedures can be relatively simple because chrysotile is the principal asbestos mineral used for building insulation materials, with amosite used to a much lesser extent. In 18 samples from a midwestern municipal health department, chrysotile was a major constituent (>50 weight-percent) in 2 samples, a minor constituent (1 to 10 weight-percent) in 12 samples, and not detected in 4 samples. Other minerals present in various concentrations in these samples were calcite, quartz, gypsum, and mica. Amosite was found as a major constituent in the ceiling of an older building located on a university campus. The presence of either serpentine or amphibole minerals in the insulation materials can be used as a probable indication of asbestos. Therefore, screening tests are based on the presence or absence of characteristic differential thermal analysis or x-ray diffraction peaks of either serpentine or amphibole minerals. For the positive samples, confirmation of the presence of the asbestiform variet reguires some type of microscopic examination because the thermall~ and x-ray '~rifd f action methods do not identify the mineral variety. Some samples will be composed of a mixture of synthetic and natural fibers, such as the mixture of fiberglass and chrysotile shown in figure 12. Generally, it is not difficult to identify the synthetic fibers based on their larger diameter and the more ~'niform appearance. Figure 12. Sample from university building, showing a mixture of chrysotile and fiberglass (X 140). Amphiboles and Talc Asbestos-related health regulations are having a significant impact on the domestic talc industry from occupational exposure at the mines and mills and at various manufacturing plants that use talcs in their operations. Certification that the talc does or does not contain asbestiform minerals is important because the occupational health 218

(03 requirements are much more restrictive if the talc is designated as containing asbestiform serpentine or amphibole minerals. Talc is both the name of a specific mineral, Mg3Si*01o(OH)2, and a commercial term for a mixture of minerals ranging from essentially 100 percent talc to blends where the mineral talc is a minor constituent [12,13]. Semi-quantitative estimation of the serpentine and/or amphibole mineral concentration, if present, can be obtained by x-ray diffraction and differential thermal analysis. Several talc deposits contain a variable amount of tremolite. Therefore, the essential question faced by the analyst is whether or not the tremolite is fibrous. Judgment required of the analyst is illustrated by the sample shown in figure 13. This sample consists of platy talc, cleav- age fragments of an amphibole, and minor to trace amounts of fibrous amphibole. For this latter sample, the 3-to-1 aspect-ratio criteria would greatly overestimate the number of fibrous amphibole particles collected on air filters or other monitors. Figure 13. Platy talc, tremolite cleavage fragments, and a fibrous tremolite particle (A) (X 400). References [1] Journal of the American Industrial Hygiene Association. Recommended Procedures for Sampling and Counting Asbestos Fibers. Vol. 36, pp. 83-90 (February 1973). [2] Berger, H., Asbestos Fundamentals. (Chemical Publishing Co., New York) 171 pp. (1963). [3] Beaman, 0. R. and File, D. M. , Quantitative Determination of Asbestos Fiber Concentrations. Anal. Chem. , 48, pp. 101-110 (January 1976). [4] Langer, A. M., Approaches and Constraints to Identification and Quantification of Asbestos Fibers. Environmental Health Perspectives, 9, pp. 133-136 (1974). [5] Cressey, B. A. and Zussman, J. , Electron Microscopic Studies of Serpentinites. Canadian Mineralogist, 14, pp. 307-313 (1976). [6] Mumpton, F. A. and Thompson, C. S., Mineralogy and Origin of the Coalinga Asbestos Deposit. Clays and Clay Minerals, 23, pp. 131-143 (1975). [7] Ernst, W. G., Earth Materials (Prentice-Hall, Inc., New York, 1969). 219

[8] Deer, W. A., Howie, H. A., and Zussman, J., Rock Forming Minerals. (John Wiley & Sons, Inc., New York, 1963) 5 v. [9] Wylie, A., Optical Properties of Asbestiform Amphiboles and Their Nonasbestiform Analogs. Available from A. Wylie, Bureau of Mines, College Park, Maryland 20740. [10] Ruud, C. 0. , Barrett, C. S., Rissell, P. A., and Clark, R. L., Selected Area Electron Diffraction and Energy Dispersive X-Ray Analyses for the Identification of Asbestos Fibres, a Comparison, Micron, 7, pp. 115-132 (1976). [11] John, W., Berner, A., Smith, G., and Wesolowski, J. J., Experimental Determination of the Number and Size of Asbestos Fibers in Ambient Air. Calif. State Department of Health, Rept. AIHL/SP-1, 36 pp. (January 1976). [12] Hamer, D. H., Rolle, F. R., and Schelz, J. P., Characterization of Talc and Associated Minerals. J. American Industrial Hvgiene Association, 37, pp. 296-304 (May 1976). [13] Rohi, A. N., Langer, A. M., Selikoff, I. J., Tordini, A., Klimentidis, R., Bowes, D. R., and Skinner, 0. L. , Consumer Talcums and Powders--Mineral and Chemical Characteriz- ation. J. Toxicology and Environmental Health, 2, pp. 255-284 (1976). Discussion J. LEINEWEBER: You brought up the question of cleavage fragments vs fibers, and asbestiform vs non-asbestiform varieties. I would like to ask why you attach so much significance to this. I think Dr. Kotin couched it most directly yesterday: the body doesn't have a dictionary. When we see fibers, if they are in the size range and if we accept this philosophy, does it matter where they come from? W. CAMPBELL: I think all health data has been based on commercial asbestos, correct? LEINEWEBER: Not necessarily commercial asbestos, but fibers of one type or another. CAMPBELL: OK, fibers. LEINEWEBER: Man-made mineral fibers or natural mineral fibers. CAMPBELL: There has been little medical studies made upon cleavage fragments. Now these may be just as harmful as fibers, but until you find this out you should call them by their proper names. To call a cleavage fragment a fiber does not help anybody. LEINEWEBER: I don't see any reason for muddying the waters with the semantic dif- ferences. CAMPBELL: I think there is some dispute whether or not there is a difference between a fiber, based on surface properties and a much larger length-to-width, and a cleavage fragment. Until you find this out you should call it either a fiber or a cleavage fragment. They may be equally harmful if they are both 20:1 and 0.5 pm in diameter, but this really has not been studied. The whole problem with the Lake Superior region was the debate whether or not the cummingtonite fragments were the same as the amosite asbestos. LEINEWEBER: This, in that context, was an argument based on the shenanigans that normally take place in the court of law, and here we are in a scientific environment. CAMPBELL: I am not a medical scientist. Obviously I don't know if a cleavage fragment is the same harmful particle as an asbestiform particle, but until you find this out; you just call it by the proper name. It does not help to call them both the same when they may be different. 220

National Bureau of Standards Special Publication 506. Proceedings of the Workshop on Asbestos: Definitions and Measurement Methods held at NBS, Gaithersburg, MD, July 18-20, 1977. (Issued November 1978) AN OVERVIEW OF ELECTRON MICROSCOPY METHODS Clayton 0. Ruud Denver Research Institute University of Denver Denver, Colorado 80208 Abstract According to a recent National Academy of Sciences Report, animal deposition model studies have shown the fiber size has some effect upon the toxicity of mineral microfibers, the long thin ones appearing to be most active [1]1. However, the extrapolation of these results to the relative carcinogenicity in humans must be tempered by the consideration that an experimental animal model has not been established. Moreover, the size range to be considered long, thin microfibers is not clearly defined, that is to say, the shortest length may be on the order of one micrometer or ten micrometers. For this and other reasons most scientists in the field consider that it is necessary to obtain data on length and width, as well as on concentration and species of mineral fiber fragments in the environment. Due to these considerations, microscopy methods are necessary for mineral fiber analysis, and because of the small size of the particles, electron microscopy is necessary. This paper will describe the methods and techniques of electron microscopy which are most generally applied. These are the transmission electron microscope-selected area electron diffraction (TEM-SAED) and the scanning electron microscope-energy dispesive x-ray spectroscopy (SEM-EDXS) methods. The advantages and disadvantages of these two techniques will be discussed, including their relative proficiency in detecting sub-micrometer fiber fragments. Their ability to identify the species of mineral, sample preparation techniques, statistical considerations and the cost of analysis will also be reviewed. The application of various techniques and methods based upon the TEM-SAED or SEM-EDXS systems will be discussed, including situations where one or the other is the optimum method. The advantages of combined systems, scanning transmission electron microscopy with SAED and EDXS, will be discussed. Also new approaches of combination and computer controlled methods using both TEN and SEM will be described. In conclusion, the state of the art will be discussed in terms of general considerations necessary for the selection of an electron microscopy technique for mineral fiber analysis. Key Words: Amphibole asbestos; asbestos; chrysotile; electron diffraction; energy-dispersive x-ray spectroscopy; mineral microfibers; scanning electron microscopy; selected-area electron diffraction; transmission electron microscopy. 1Figures in brackets Indicate the literature references at the end of this paper. 221

Background Collection of mineral particles for identification and counting is usually done by filtering the medium, air or water, through cellulose ester membrane (Millipore) or perforated polycarbonate (Nuclepore) filters, thereby concentrating them through deposition on the filter's surface. The effective minimum particle collection size is always less than one half a micrometer. The optical microscope is used extensively for counting mineral fibers collected from occupational environments, but it is generally agreed that this is a matter of expedience and not due to adequacy. By far the greatest number of asbestos mineral fibers found in the environment, including occupational environments, are below the resolving power of the optical microscope. Since neither epidemiology nor animal studies on the relative toxicity of mineral microfiber have shown conclusively that those less than 0.5 pm in diameter or width are innocuous, it has been considered prudent to count, size, and identify all particles with an aspect ratio greater than 3 to 1 which are tens of micrometers in length and shorter. Although the long thin fibers seem to be more active in animal deposition- model studies, the shortest active fiber length has not been established [1]. Also, because a number of mineral and man-made microfibers are suspected of producing varying degrees of adverse health effects, the identification or classification of a mineral fiber as to species is important. Until the effect of size, morphology, species and other properties of microfiber can be related to toxicity, it will be necessary for the analyst to characterize the distribution of a number of these parameters from environmental samples. Electron Microscopy The group of analytical instruments which provides more of what are considered the important parameters mentioned above is that of the electron microscopes. Both. transmission and scanning electron microscopy have been used extensively for mineral fiber identification, sizing and counting, and both types of instruments and their related characterization techniques have their place. The transmission electron microscope (TEM) with selected area electron diffraction (SAEO) Is considered the most widely applicable instrument, although it has some disadvantages which will be discussed. This technique requires that the image forming electrons travel through the sample and therefore the sample matrix must be transparent to the high kinetic energy (usually about 100 KeV) electrons. SAED also requires that the electrons travel through the matrix as well as some part of the microfiber to be iden- tified. SAED is used to characterize the crystal structure of the particle of interest and is valuable for the identification of the type or class of fiber, e.g., serpentine asbestos, amphibole asbestos, non-crystalline or non-asbestos. Scanning electron microscopy (SEM) can be compared with reflected light microscopy. However, images are formed electro-optically, usually by secondary electrons produced by a focused electron beam in the sample. The technique usually employed for species identification is energy dispersive x-ray spectroscopy (EDXS) which determines the energy of x-rays emitted from the sample. This emitted x-ray energy spectrum is caused by the electron beam interaction with the sample and can be used to qualitatively and semi- quantitatively identify the elemental content of a microfiber. A third type of instrument which combines the advantages of both the TEM and SEM is the scanning transmission electron microscope (STEM). This instrument has been used by a number of laboratories, most of which have procured it specifically for asbestos microfiber counting and identification. Essentially it is a transmission electron microscope equipped with scanning and focusing coils so that a focused beam of electrons can be scanned over the sample or pinpointed in a particular area. The most general mode of application is to obtain a shadow image as with the TEM, then perform SAED and/or EDXS as desired. The focused beam should produce a brighter SAED pattern for particle identification than in the TEM, and if an elemental analysis is desired this may be obtained from the same particle without transferring the specimen to another electron beam instrument. 222

There is another type of electron microscope which has been used only sparsely for asbestos mineral fiber analysis. This instrument is an SEM with an electron detector below- the specimen for transmission imaging. This allows a transmitted electron image to be formed and the instrument might be called a transmission scanning electron microscope (TSEM). Application of this technique will be discussed in a subsequent section. Needless to say, combinations of SEM and TEM instruments have been and are being used for microfiber analysis also. Applications There are four important considerations in the selection of an electron microscopy method for the counting and characterization of microfibers. These are: observability, specificity, sample preparation and analysis cost. Observability Observability is concerned with the sharpness and contrast of the microfiber image against the matrix. This controls the relative ability of the microscopist to find microfibers, measure them, and characterize their morphology. Flinckinger and Standridge [2] compared fiber counts with SEM and TEM from water samples and concluded that for small fibers TEN gave much higher counts, about an order of magnitude or greater. Ruud et al. [3] showed the relative clarity of SEM and TEM images illustrating the superior contrast of the latter (see figure 1). The highly magnified shadowgraph obtained in transmission electron microscopy is for the most part an accurate representation of the length and width or diameter of the fiber. Chrysotile fibers are usually circular bundles of fibrils or round single fibrils. Often the fibrils can be distinguished in a TEM image by the fact that they are tubular and the hollow center can be seen in the electron microscope image [3]. While this tubular appearance is characteristic of chrysotile, it is not always present so that if a fiber does not appear to be hollow this does not rule Nt out as chrysotile. Amorphous material can be attached to the surface and fill the tubes, thereby giving the appearance, as far as density is concerned, that the fiber is solid [4]. At any rate it is well to have an identification method in addition to morphology for chrysotile and it is imperative for the amphibole minerals since non-asbestos material can appear in the electron microscope to be fibrous, i.e., they may have a 3:1 length-to- width ratio. Also, many chain silicate non-asbestos minerals fracture in the same general way as the asbestos minerals so that morphology does not lead to a reliable identification. See figure 2 from Ruud et al. [3]. The most effective additional identification method is selected area electron diffraction, which will be discussed subsequently. 223

Figure 1. Comparison of SEM and TEM image clarity for a microfiber fonn an environmental sample. Top is SEM image and bottom is TEM image. The marks in the upper left corner of each micrograph are 1 micrometer apart. 224

Figure 2. A TEM micrograph of the mineral wollastonite. The superior image contrast of small microfibers and the clarity of internal voids in the TEM can be understood when the mechanism of image production and resolution of the two types of instrumentation is compared. The TEM relies upon the electron opacity of the microfibers which depends upon the thickness but which is invariably several times higher than that of the specimen substrate. SEM relies upon the production of secondary electrons for imaging and the relative difference of their efficiency of production between microfibers and substrate is often rather slight. In spite of these considerations, a recent report issued by the EPA [5] judged the two techniques as equal with respect to fiber counting. However, the sample type and analytical procedure covered In that report were very specific and not what may be generally expected or applied in environmental samples. The sample source was a laboratory prepared and dispersed Canadian chrysotile. The TEM sample preparation was one which is se}dom if ever used in TEM preparation because it is complicated and prone to fiber loss and contamination. This report therefore cannot be used as justification for the general use of SEM for microfiber sizing and counting. The electron microscope magnification used to locate and measure microfibers is an important concern and generally varies from 4000 to 20,000 times. It should be obvious that the lower the magnification used to find microfibers consistent with sharp contrast, the higher the likelihood of missing very fine ones. At 10,000X a 0.1 micrometer fiber would appear to be 1 mm wide and at 4000X it would be 0.4 mm wide. On the other hand, the lower the magnification used to search for microfibers the larger the area of electron microscope specimen observed, thereby improving counting statistics for a given amount of analysis time. Specificity Specificity is concerned with the identification of a microfiber species. In the SEM, clues as to the elemental content may be obtained by EDXS, and these can sometimes be used to identify the microfiber. With the TEM, SAED is usually employed for speciation. SAED produces a pattern which is indicative of the crystal structure of a microfiber. This crystal structure can then be related to the type or species of fiber. Usually only classification is possible, but in the case of chrysotile asbestos it is usually readily identified by SAED. 225 2063105020

The basis for SEM-EDXS is that electron beam microchemical analysis may sometimes be used to distinguish particles of various minerals (6,7,8]. The most common method presently in use is the energy dispersive x-ray system (EDXS) attached to an SEM. X-ray wavelength dispersive analyzers and the conventional electron microprobe have been used; however, their routine application is negligible in asbestos microfiber analysis because the high electron beam currents required may damage the specimen and the microanalysis procedure is relatively time-consuming. Semi-quantitative electron beam x-ray microchemical analysis in the electron microscope is based on the fact that a beam of high energy electrons incident upon a particle generates x-rays with energies that are characteristic of the elements present in that particle. Only those elements heavier than sodium (atomic number 11) can be practically detected. An EDXS detector placed in the electron microscope sample chamber close to the specimen converts the energy of the x-ray photons to voltage pulses which are amplified, digitized and stored in a multichannel analyzer or a minicomputer. In the EDXS identification of microfibers, ambiguities can arise from x-rays produced by adjacent or adhering particles, from instrumental uncertainties in determining the exact chemical composition of a particle [9], or from the fact that a given mineral can exist over a wide range of compositions (10]. Rs much as a 10 percent variation in the element x-ray intensity can be expected from any one mineral sample [71 or even a single microfiber (11]. To further confuse the matter we have observed many mineral particles that are often associated with asbestos materials which show a 3:1 length-to-width ratio and give EDXS spectra that cannot be distinguished from the asbestos types. Figure 3 shows an example of SEM-EDXS data from an anthophyllite microfiber and a lizardite cleavage fragment with a greater than 3:1 aspect ratio. Anthophyllite is an amphibole asbestos mineral and lizardite is a non-asbestos polymorph of chrysotile. However, the EDXS spectrum from the two are indistinguishable. A number of examples of this type of possible misidentification of mineral microfiber appear in Ruud et al. [3]. In spite of the above considerations, a number of researchers have surmised that each of the asbestos minerals can give x-ray spectra that usually are characteristic enough, when combined with fiber morphology, to allow their mineral identification [6,7,12]. Visual observation of the semi-quantitative fiber x-ray spectrum is the usual method of fiber identification; however, three-component diagrams have been used after subtracting the continuous background from the semi-quantitative x-ray spectrum for further extrapolation of the data [6]. For these analyses, matrix corrections are rarely used. Typically, iron, magnesium, and silicon are plotted on the three component diagram and compositional boundaries for the asbestos minerals established. In addition to the major shortcomings mentioned in the previous paragraph, this added refinement suffers from its failure to use all compositional data obtained such as presence or absence of sodium, calcium, aluminum. and manganese which might aid in identification [6]. As has already been discussed, observation of proper elemental intensities by energy- dispersive x-ray analysis is generally not sufficient for positive identification of fibers. For example, chrysotile, anthophyllite, and fibrous talc, which have similar elemental compositions, may be difficult to differentiate (3,6]. These considerations make the sole use of SEM-EDXS unreliable in its general appli- cation to the identification of fibers and microfibers. There are specific cases where the source of the sample is well characterized and the absence of particles of nearly similar chemical composition has been confirmed that it may be useful. Considering the uncertainties in SEM application to the identification of micro- fibers, it is understandable that transmission electron microscopy coupled with selected area electron diffraction has been selected by many researchers as the most viable method for identifying and counting asbestos fibers [1]. Although this method has some disadvantages, the overriding advantage is that usually it is specific with respect to the identification of chrysotile or amphibole microfibers and it permits accurate size measurement of particles even when that size is on the order of fractions of micrometers in diameter. 226

Figure 3. SEM-EDXS spectra from anthophyllite asbestos (top) and lizardite (bottom) samples. ' Selected area electron diffraction can be readily accomplished on a modern transmission electron microscope and a pattern observed in about 10 seconds and recorded usually in less than two minutes. However it usually requires an experienced microscopist and some fine manipulation of the specimen in the SAED mode for production of a clear pattern. The two-dimensional SAED pattern of diffraction spots has the advantage, in the case of some asbestos microfibers, that it contains certain outstanding characteristics that can be recognized at a glance. This is particularly true for the more common type of asbestos, the serpentine mineral chrysotile [4,13], figure 4. 227 2063105022

I Figure 4. Chrysotile asbestos TEM micrograph (top) and SAED pattern (bottom). The SAED pattern of a single chrysotile fiber or fibril is analogous to a rotating or oscillating crystal x-ray diffraction pattern in which the long dimension of the fiber tends to lie parallel or nearly parallel to the supporting membrane and therefore is perpendicular to the incident beam corresponding to the axis of rotation being normal to the beam in the usual type of rotating crystal x-ray exposure. This analogy is also 228 ~

cv, (CS partially true for amphibole fibers. In x-ray patterns the spots are arranged in lines, universally called "layer lines," with the spacing between the lines dependent upon the periodicity of the crystal structure in the direction of the axis of rotation (see, for example, Barrett and Massalski [14]). The analogous layer lines in SAED are also very prominent and their spacing reveals the crystal periodicity in the direction of the fiber axis. From a quick view of the layer line spacing one cannot distinguish between chrysotile, tremolite, and amosite which all have layer line spacings corresponding to a periodicity of approximately 0.53 nm, but this group of materials can often be distinguished from some others of interest, for example wollastonite, lizardite, antigorite, albite, hedenbergite, or diopside [3]. Fortunately there is no need for a detailed study of the pattern in order to positively Identify chrysotile. The chrysotile diffraction pattern has very prominent streaks on layer lines other than the central one, and some streaking also may be seen on the central one [13]. Some spots of normal sharpness also occur; these are on the central layer line and alternate ones (2nd, 4th, etc.). The streaks are seen on the pattern in figure 4 and can also be seen on the fluorescent screen of the electron microscope. The geometry of the pattern is known for orthochrysotile, clinochrysotile, parachrysotile and mixed ortho plus clino varieties [15], and the origin of the streaks is now well understood as resulting from disorder in the stacking of the prominent layers in the crystal (the hydroxyl, magnesium oxygen-hydroxyl, silicon and oxygen layers). The series of researches beginning with Warren in 1941 and extending through many studies by Whittaker in 1956, have shown that the layered structure is curved cylindrically around the axis of the fiber, the axis with 0.53 nm periodicity in clino and ortho varieties. This is called the c axis in some of the papers [16], but is called the a axis on others [15]. There is x-ray evidence [16] that the layers are wrapped in a helical cylindrical manner and this is confirmed by electron microscopic views of the cross-section of the chrysotile tubes by Yada [17]. This curvature of the structure accounts for the preqence of the prominent layer lines, which are perpendicular to the length direction of• the fiber. Amphibole minerals exist in both asbestiform and massive varieties. Numerous names have been given to varieties of the amphibole groups, and the many different types of atoms substituted in the different members of the groups [18] add to the natural difficulties of identifying them. It is not surprising that the Joint Conmittee on Powder Diffraction Standards x-ray powder data file contains many cards of diffraction patterns differing from each other by small amounts. SAED patterns prepared in this laboratory of known samples of the amphibole asbes- tiform minerals tremolite, crocidolite and amosite have prominent rows of spots which resemble the layer lines of rotating crystal x-ray patterns and which we will also call 'layer lines.' There are especially closely spaced spots on each of these layer lines, far more closely spaced than they are in the rows of spots from the minerals hedenbergite, albite or wollastonite, for example [3]. We have rarely observed any non-asbestos material exhibiting the characteristic layer line spacing and spot patterns within the layer lines displayed by asbestos mineral fibers. However, this author has recently been informed that pyroxenes have been observed to produce asbestos-like SAED patterns. Although chrysotile is usually readily distinguished from the asbestiform varieties of amphibole (crocidolite, amosite2, anthophyllite, tremolite and actinolite), it is not easy to distinguish one variety of these amphiboles from another because the spacing of prominent rows of spots in these are the same, and the differences occur only in the arrangement of spots along the rows. However, an experienced microscopist can learn to distinguish on sight a pattern usually characteristic of an asbestos fiber from the patterns of most non-asbestos minerals commonly associated with them. Crystalline materials that exist in the form of thin plates also produce SAED patterns with many spots, but these in general are arranged in a two-dimensional array in which there are not such prominent layer lines in a single direction. ZAmosite - a discredited term. N 229 ~ r+ 0 w a N A

As mentioned above, SAED is used extensively as the major criterion for the identification of mineral microfibers [1,2,3]. However, it should be mentioned that the method is empirical and has not been rigorously tested. The possibility exists that some species of non-asbestos mineral fibers or microfibers may produce a high incidence of SAED patterns characteristic of chrysotile or the asbestos amphiboles. An example of this, which has been mentioned, is pyroxenes. Transmission electron microscopes and STEM equipped with an energy dispersive x-ray detector are available which allow simultaneous observation of morphology, crystal structure and elemental composition. These microscope systems have been used to study fibers of known asbestos origin as well as environmental and material samples [12,19]. It would be highly advantageous if a thorough crystallographic examination of the SAED pattern could be performed in the few seconds in which patterns are now cursorily examined. This is technologically possible, but requires the building of a TEM or STEM with a television camera in place of the fluorescent screen coupled to a computer programmed to index and classify the pattern with respect to standard or calculated patterns. These facilities are extremely expensive and few laboratories will be so equipped in the near future. However, studies of the patterns with respect to mineral type, cleavage and fiber orientation are needed. Sample Preparation As previously mentioned, samples for electron microscopy analysis of microfibers are generally collected on cellulose ester membrane (Millipore) or perforated polycarbonate (Nuclepore) filter media [5,19]. For analysis by the SEM the latter medium, due to its smooth surface, is preferred. SEM preparation is usually done by coating the surface directly with an electrical conducting material, e.g. gold, silver, carbon or silicon monoxide [5]. More complicated methods have been used for SEM preparation of samples collected on Millipore [9]. These filters with their rough surface are not generally suitable for direct coating for SEM because small fibers may be masked by protrusions of the surface. In TEM, STEM, and TSEM analysis, the matrix must be nearly electron transparent to electrons of about 100 KeV energy. This requires that the filtrate (particles) be mounted upon electron microscope grids with very thin, on the order of 100 Angstroms, carbon or metallic substrates and the filter material dissolved away. Several dissolution techniques are used, including the Jaffe wick and condensation washing. Generally these techniques are relatively simple and maintain the original particle size distribution and relative particle location. Some investigators have reported particle losses as high as 60 percent with the condensation washing technique compared with less than 10 percent with the Jaffe wick method [20]. Coating the filter and filtrate with a conductive layer prior to dissolution has been proposed as a technique to minimize particle loss [19,21]. Also, careful control of the condensation washer can reduce filtrate loss to much less than 10 percent. Most laboratories apply a second carbon, metallic or silicon monoxide coating to the filtrate after filter dissolution to reduce the probability of particle loss. The choice of conduction coating is varied; however, many laboratories have been considering fine grained metallic coatings because of superior contrast and the fact that a reference pattern is provided on the SAEg patterns. The general preparation technique discussed in the previous paragraph is known as the "direct transfer" method. A variety of more complicated techniques include the direct transfer procedure as the last few steps. This includes ashing of the sample which is required when a considerable amount of organic material is collected with the inorganic microfibers or sometimes is used as a preliminary step to redistributing the filtrate for a more uniform or more suitable concentration. Dissolution of the collection filter substrate and subsequent refiltering has also been used. Needless to say, whether TEM or STEM is performed, the particulates must be distributed as uniformly as possible on the filter sample. This is a vital consideration in the statistics of analysis which will be covered by another author in this publication. Ashing can be performed in a low temperature oxygen plasma device or at high temperatures in a muffle or tube furnace. There are pros and cons to all redistribution procedures which must be considered by the analyst; however, it is always highly desirable to process control specimens, i.e., 230

blanks, when preparing samples for fiber counting and analysis. These blanks confirm a clean preparation environment or bear witness to laboratory contamination. Another preparation technique which has been used off and on is the so-called "rub- out" technique. This was used early in the electron microscopy analysis of microfibers and has been applied by the Mount Sinai group [22]. High particle losses and the destruction of the true particle size distribution to produce only a mass concentration are cited as disadvantages with this technique. Other techniques have also been cited as viable, including that in a recent EPA report [5]. However, most have been discarded in favor of the direct transfer method alone or preceded by ashing only when necessary. The added specimen handling necessary for transmission electron analysis has often been cited as a serious disadvantage to TEM, STEM and TSEM analysis. However, experienced laboratories have developed preparation routines and techniques which make particle losses, contamination and labor time negligible. The usual amount of time lag in preparation of a transmitted electron sample is about four hours. Analysis Cost The amount of electron microscope time necessary for an analysis is the major consideration affecting cost, and is dependent upon many factors, not the least of which is the sample from which the specimen was produced. The size distribution, particle loading and uniformity of distribution are just three of these. If a very limited amount of microscope time requires that the analyst use only a low magnification, e.g., 4000X, then the small microfibers may be missed. Computer image analysis has been used by a few laboratories [9] and can be applied directly on an electronic image as produced in the SEM, STEM and TSEM or on photomicrographs produced by an electron microscope. Direct computer image analysis is also possible with suitably modified imaging devices mounted into TEM's. This technique can greatly reduce the amount of microscope time requireil for microfiber searches but is prone to certain errors, especially where high concentrations of microfibers and other particles are present. The application of microfiber identification techniques affects the microscope time as well as imaging. The TEM image is essentially instantaneous, whereas an SEM image must be acquired with time and takes several seconds to form. Furthermore, on a typical SEM the time for one EDXS analysis is 100 or more seconds. As a conse- quence most analysts working with SEM and STEM only obtain analyses from selected microfibers, not all of those found. SAED usually requires 10 to 30 seconds to form an image suitable for recognition by the microscopist and is usually performed on all microfibers found. Recording of this image is done selectively on a few microfibers and usually requires 100 to 200 seconds. The beam focusing feature available on all STEM and some TEM can reduce the recording time by producing a brighter SAED pattern. Technique Development A number of laboratories are evaluating the various electron microscopy techniques used in the analysis of microfibers. That this is necessary is evident from the wide discrepancy in results produced on similar samples by different laboratories and/or microscopists (23]. No two laboratories perform sample preparation or microfiber analysis exactly the same and some are markedly different. However, over the past three years a number of laboratories have markedly improved their analytical reliability in spite of the overwhelming statistical uncertainties. This author is aware of some new approaches to the identification, counting and measurement of microfibers. United States Steel Research Laboratories are applying a specially equipped TSEM with an EDXS detector located for a very high x-ray take-off angle, higher than possible in a standard unit. This sytem is computer controlled using criteria from the transmitted electron image data at 1g,000X magnification processed through an image analyzer to locate microfibers. The geometry of the system and the sample and x-ray detector distance (less than 1 cm) are such that a very adequate EOXS spectrum can be accumulated an order of magnitude faster than with standard electron microscopes, SEM or STEM. After a statistically significant number 231 2063105026

of microfibers are found and EDXS data obtained from each, they are classified with respect to aspect ratio and EOXS spectrum. The specimen is then transferred to a TEM, a 1200 KeV instrument in this case, where some microfibers in each EDXS clas- sification are selected and an SAED pattern obtained for identification. It is recog- nized that a 1200 KeV TEM is not' readily available; however, the SAED could be performed on most TEN instruments with 80 to 100 KeV. The advantage in the transmitted image over that usually produced in an SEM is greater visibility of particles, as has previously been stated. Moreover, the tech- nique has a great advantage over those presently applied from the standpoint that a large number of microfibers are analyzed at least through classification and this is a tremendous statistical advantage. Conclusion In conclusion there are a few points that should be made. 1. The transmitted electron image is generally accepted as being superior for counting and measuring microfibers as compared with a secondary or backscatter electron image. 2. Selected area electron diffraction is generally accepted as the best criterion for the identification of asbestos mineral microfibers, although a few non-asbestos minerals may be mistaken for asbestos. 3. The statistical consideration affecting electron microscopy of microfibers is a source of considerable error and new techniques are being and must be developed to relieve these problems. 4. There are a few specific situations where the SEM can be applied to the counting of microfibers, especially where the source and species mixture are well characterized. 5. Although the TEN-SAED method of asbestos mineral microfiber counting and identifica- tioh is not absolute, it is the best compromise of accuracy and cost available. The author would like to thank C. S. Barrett and J. M. Dement for their contribution to this paper. References [1] National Academy of Sciences, "Drinking Water and Health," Part I, 1977. [2] Flinckinger, J. and Standridge, J., Identification of fibrous material in two public water supplies, Env. aci. and Tech. 10, No. 10, 1028-1032 (Oct. 1976). [3] Ruud, C. 0., Barrett, C. S., Russell, P. A., and Clark, R. L., Selected area electron diffraction and energy dispersed x-ray analysis for the identification of asbestos fibers, a comparison, Micron 7, 115-132 (1976). [4] Mumpton, F. A., Characterization of chrysotile asbestos and other members of the serpentine group .inerals, Siemens Review XLI, 7th Special Issue, 75-84 (1974). [5] Gerber, R. M. and Rossi, R. C., Evaluation of electron microscopy for process control in the asbestos industry, EPA-600/2-77-059 (Feb. 1977). [6] Rubin, I. B. and Maggiore, C. J., Elemental analysis of asbestos fibers by means of electron probe techniques, Env. Health Perspect. 9, 81-84 (1974). 232

[7] Ferrell, Jr., R. E., Paulson, C. G., and Walker, C. W. , Evaluation of an SEM-ES method for identification of chrysotile, In Proc. 8th Annual SEM Symposium, Johari, 0. and Corvin, E. (eds.). IIT Research InstStute, Chicago, I11., 537-546 (1975). [8] Langer, A. M. , Rubin, I., and Selikoff, I. J., Electron microprobe analysis of asbestos bodies, Histochem. Cytochem. J. 20, 735-740 (1975). [9] Pattnaik, A. and Maakin, J. D., Development of scanning electron microscopy for measurement of airborne asbestos concentration, U.S.E.P.A., Office of R and 0 Pub. No. EPA 650/2-75-029. Research Triangle Park, N.C., (1975). [10] Speil, S. and Leineweber, J. P., Asbestos minerals in modern technology, Env. Res. 2, 166-208 (1969). [11] Pooley, F. D., The identification of asbestos dust with an electron microscope microprobe analyzer, Norelco Reporter, 23, No_2, 5-9 (Oct. 1976). [12] Dement, J. M. , Zumwalde, R. D., and Wallingford, K. M. , Asbestos fiber exposures in a hard rock gold mine, In Proc. N.Y. Acad. of Scci. Conf. Occup. Carcinogenesis. (1975). [13] Clark, R. L. and Ruud, C. 0., Transmission electron microscopy standards for asbestos, Micron 5, 83-88 (1974). [14] Barrett, C. S. and Massalski, T. R., Structure of Metals, (McGraw-Hill, New York, 654, 1966). [15] Zussman, J., Brindley, G. W., and Comer, J. J., Electron diffraction studies of serpentine minerals, Amer. Mineralogist 42, 133-153 (1957). [16] Whittaker, E. J. W. , The structure of chrysotile, Acta Cryst. 27, 659-664 (1956). r [17] Yada, K., A study of microstructures of chrysotile asbestos by high resolution microscopy, Acta Cr st. 9, 855-857 (1971). [18] Deer, W. A., Howie, R. A., and Zussman, J., Rock Forming Minerals, (Wiley, New York, Vol. 3, p. 270, 1963). [19] Cook, P. M., Rubin, J. B., Maggiore, C. J., and Nicholson, W. J., X-ray diffraction and electron beam analysis of asbestiform minerals in Lake Superior waters. In Trans. Inst. Electrical Electronic Eng. (in press) (1974). [20] Beaman, D. R. and File, D. M., The quantitative determination of asbestos fiber concentrations. The Dow Chemical Company, unpublished report (1975). [21] Oriz, L. W. and Loom, B. L., Transfer technique for electron microscopy of membrane filter samples, Am. Ind. ~jg. Assoc. J., 423-425 (1974). [22] Rohl, A. N., Langer, A. M., and Selikoff, I. J., Environmental asbestos pollution related to use of quarried serpentine rock, Science, 196, 1319-1322 (17 June 1977). [23] Brown, A. L., Taylor, W. F., and Carter, R. E., The reliability of measures of amphibole fiber concentrations in water, Env. Res. 12, 150-160 (1976). N ~ Y O 233 0 ~

Discussion J. LEINEWEBER: I would like to make one comment with regard to Clay Ruud's remark about using the central channel of the chrysotile fiber for identification. This is good a reasonable percentage of the time, but you can run into chrysotile fibers such that this channel is not very visible and may be pretty well filled up with the non-cystalline material. So, it cannot be used as positive identification. C. RUUD: I know. R. FISHER: I want to get clarification whether you advocate visual identification from the diffraction patterns and visual counts in contrast to recording micrographs. It seems to me desirable to have your data in a form that others can confirm, look at your diffraction patterns, look at your counts, and not rely on visual observations that are just recorded in a pad or notebook. RUUD: I hear what you say, and I would like to record every pattern or every micro- graph that is projected on the screen, but I can't afford to do this; my sponsor won't stand for it. So, what we do is, we record typical SAD patterns we see in particular samples or sets of samples. When we see something different than that, something unusual, strange, we record it. I agree that it would be nice to have everything recorded for posterity, but it takes too much time. FISHER: Well, at this stage it is essential to have records that can be accepted by others, I am afraid. I agree the costs are high, and people will have to pay them, but I think that any data that are not recorded for confirmation and detailed examination are going to be challenged in all kinds of situations. RUUD: As I say, we record typical ones; we save the samples and since the samples are on finder grids any grid can be found and the data confirmed. J. ZUSSMAN: I'd like to make three comments concerning Or. Ruud's paper. One concerning electron diffraction patterns. I think he has very much underplayed the varia- tions and variability one can get in electron diffraction patterns, depending very much upon the orientation of the grain, the way it lays on the stage. If you look at these patterns carefully, you see enormous numbers of different effects; I would have made this comment anyway; I make it still much more strongly now, having heard a lot of judgments are made perhaps without even taking photographs. From looking down on the screen you can certainly not see the subtle variations which are nevertheless important, produced by orientation effects. Secondly, you mention the scanning electron microscope as being best for chemical analytical purposes. I don't think it is capable of an accuracy that can be obtained by the transmission electron microscope with suitable attachments, or STEM, which brings me to the third point. You showed that lizardite and anthophyllite were not distinguishable from their x-ray fluorescence spectra, and this is surprising. The magnesium-to-silicon ratio for lizardite is 1.5 to I in atomic ratio, the other ratio is 0.9 to 1, and I think there is a detectable difference. The reason why we may not pick up this difference is that your crystal has the wroog kind of thickness so that the crude ratios of peak height are not indicative of concentration. The crystal has to be of a suitable thickness for this to be so. RUUD: Regarding the last comment, we can rotate the fiber or change the position in the microscope, and get different ratios, and, as someone pointed out yesterday, just be going along the fiber you may get different ratios. So, that's one reason why I do not have too much confidence in energy-dispersive x-ray spectroscopy. The first comment had to do with selected-area diffraction and the variability of patterns. We do not study them that carefully. We do not try to distinguish between the various amphiboles, amphibole asbestos materials. We do not have the time to study individual patterns that carefully. We looked at the possibilities of trying to get good d-spacings from them; it seems like it is a good possibility if we could connect the computer into a vidicon or a camera tube in the bottom of a TEM or STEM and put it directly into a computer; I think that would be great. But, so far I know of only one microscope equipped that way, and it is not used for asbestos analysis. 234

V National Bureau of Standards Special Publication 506. Proceedings of the Workshop on Asbestos:- Definitions and Measurement Methods held at NBS, Gaithersburg, MD, July 18-20, 1977. (Issued November 1978) IDENTIFICATION OF ASBESTOS BY POLARIZED LIGHT MICROSCOPY Walter C. McCrone (Paper presented by John G. Delly) McCrone Associates, Inc. Chicago, Illinois 60616 Abstract A number of analytical tools can be used to characterize and identify asbestos: infrared absorption, x-ray diffraction, DTA, SEM, TEN, and the light microscope. Each has advantages and limitations. The polarized light microscope (PLN) has many advantages, and the only disadvantages are 1) the asbestos particles must be at least a micrometer in largest dimension, and 2) considerable training in optical crystallography is needed. PLM, on the other hand, is very sensitive (ppm range), extremely rapid (1-5 minutes to identify all components of most samples) and, of all the methods, only PLM will identify the individual amphiboles. Key Words: Amphiboles; asbestos; dispersion staining; microscopy. There are a number of analytical methods useful for the identification of asbestos. These include infrared absorption (IR), x-ray diffraction (XRD), differential thermal analysis (DTA), scanning electron microscopy (SEM), transmission electron microscopy (TEM) with or without electron microprobe analyzer (EMA), and polarized light microscopy (PLM) with or without dispersion staining (OS). Each has advantages and disadvantages. Every analyst uses and should use the techniques in which he has the required training and with which he feels confident. At the same time, every sample should ideally be analyzed by the most suitable technique. Occasionally, of course, it may be wise to use two or more techniques and this is certainly true for asbestos. We would like to summarize our attitude toward the various techniques for asbestos and describe in more detail the technique we feel has many advantages and is under-utilized; this is polarized light microscopy, especially when supplemented by dispersion staining. First, however, the advantages and disadvantages of each technique: TEM is most useful for the detection and identification of asbestos fibers smaller than the resolving power limit of the PLM. This is usually the case for water or beverages in general. Quantitative procedures are available so that the number, size, and identity of asbestos fibers per unit volume can be accurately determined. Identification by TEM depends on selected area electron diffraction (SAED). Occasionally energy or wavelength dispersive detectors are fitted to the TEM to make possible elemental analysis of individual fibers. Nothing can compete with TEM for the analysis of samples containing subpicogram particles. SEM has no advantage over TEM except that it takes prettier pictures. It will also fail To- see the smallest fibers, and lacking SAED it cannot identify all fibers. The energy dispersive detector on most SEMs is not as effective as the wavelength dispersive detectors on some TEMs and SEMs. XRD is a useful method since it can be made quantitative. However, it cannot tell size or shape, is not very sensitive (about 1 percent or a bit better), and does not differentiate between most of the amphiboles. At best it supplements other techniques. 235 2063105030

IR and DTA can also be dismissed for all except routine samples containing high percentages oi-asbestos. This brings us to PLM and 5 on which we wish to spend more time because of our conviction that, of all the eicroanalytical techniques for asbestos, it is by far the most effective. It is the only method depending on the unique optical crystallographic properties of the various crystal phases in the sample. These properties - refractive indices, dispersion of refractive indices, birefringence, sign of elongation and extinc- tion angle - are unique to the crystalline state and therefore unequivocally identify chrysotile, anthophyllite, tremolite, actinolite, grunerite, cummingtonite, etc. The background for dispersion staining has been adequately covered elsewhere [13i. Very briefly, it imparts color to any transparent particle mounted in a liquid whose dispersion curve intersects the dispersion curve for the particle in the visible. The colors, related to this matching wavelength, characterize and identify any given sub- stance. With polarized light, isotropic substances show a single characteristic color, but anisotropic substances show different colors corresponding to the different refractive indices in different orientations. Chrysotile, for example, shows blue and blue-magenta colors, crosswise and lengthwise respectively, for each needle crystal when mounted in Cargille high dispersion liquid nps = 1.550. The colors shown by the various types of asbestos and a few other associated minerals are indicated in Figures 1-20 by the wavelengths on each crystal view. These are the wavelengths at which the liquid indicated and that direction in the crystal have the same refractive index. This matching wavelength, ao, determines the dispersion staining colors. TALC Mg8(S18020)(OI)4 In 1.55 (H.D.): n's (D, H, Z) * 360 497 y m 1.589-1.600 5990.05(-) This sample (Vermont): a II 1.546 0= 1.588 y= 1.589 8 = 0.045 (-) * Deer, Bovie and Zussman. Figure 1. Dispersion staining colors shown by talc crystals in Cargille high dispersion liquids nD = 1.550 and n0 = 1.580. 'Figures in brackets indicate the literature references at the end of this paper. p 9K 360 _, MQo_, «A i•a=645nm C=i.oaa-i.ba4 N O w 236 ~. 0 ~ 0 w ~

CHLORITE (Mg, Al, Fe)12((Si' AI) 8020) (0R)16 In 1.55 (H.D.): pale yellow to golden yellow In 1.580 (H.D.): 529 nm = 420 nm 513nm a= 529 nm ~+~-- 420 nm p = 513 nm Figure 3. Chrysotile. IL n's (D, H, Z) aa 1.57-1.66 S ~ 1.57-1.67 y ~ 1.57-1.67 5 = 0-0.01 Figure 2. Chlorite. The sample (Caiifornia): a=1.586 $=1.587 y = 1.598 5 = 0.010 (+) CHRYSOTILE Mg3[S12O51(O1i)4 In 1.550 (H.D.); a= 660 nm FEj ~ 530 ma , '(y p- 558nm n'e(D,$,Z) a- 1.5324.549 S = 1.540-1.553 y s 1. 545-1.558 5 = 0. 013-0. 007 (-) Kingte Mine, Quebec Sample -a= 1.5444 9-1.5525 y= 1. 5555 5 = 0. 0111 ANTIGORITE Mg3(Si205)(OH)4 In 1.550 (H.D.); F3; •Y$1W--r =3na...,.~~~ Figure 4. Antigorite. n's (D, H, Z) a~ 1.558-1.587 ~ ~ 1. 56 -1.57 y s 1. 562-1.574 5 = 0.004-0. 007 ( ) This sample: a=1.555 A= 1.559 y= 1.561 N ~ 5 = 0.008 (-) W 237 O S W N

LIZARDTTE Mg3(53205) (OH)4 In 1.55 (H.D.); n's (D, H, Z) 520 494 nm ~= 690 nm ~494 nm undulose extinction Figure 5. Lizardite. TREMOLTTE Ca2(Mg, Fe ~5(Sl 8022)(OH, F)2 (low Fe) y = 427 nm 530 nm 91460 nm Figure 6. Tremolite. a= 1.538-1.544 Q° y = 1. 546-1.560 6 = 0. 016-0. 008 This sample: a= 1.545 9911.555 y-1.557 6=0.012 (-) n'a (D, H, Z) a = 1.604-1.619 y = 1.627-1.642 8 = 0. 021-0.023 (-) This sample: a= 1.599 y= 1.621 8=0.022() 238

ACTINOLITE Ca2(MB, Fe)5(Si8022)(OH' F)2 (20-80% Fe & 80-20% Mg) In 1.605 (H.D.): In 1.640 (H.D.): n's (D, H, Z) o•= 436 nm a= 1.619-1.668 9= y = 1. 642-1.687 5 - 0. 023-0. 019 (-) This Sample (Virginia): a- 1.633 9= 1.841 y=1.847 5 = 0.014 (-) Figure 7. Actinolite. ANTHOPHYLLTTE (Mg, Fe2+)7(S18022)(OH, F)2 ffi 1.580 (H.D.): n'a (D, H. Z) In 1.605 (H.D.): y=395am p = 370 am y ~ 300 nm a= 421 nm a- 598 nm Figure 8. Anthophyllite. a= 1.596-1.694 S = 1.605-1. 710 y= 1.615-1.722 5 = 0.013-0.028 (+) (-) This sample (Pine Mt., Ga. ) o:= 1.601 p = 1.618 y=1. 628 5=0.027 (-) 239

ANTHOPHYLLITE (Mg. Fe2+)7(Si8 22)(OH"~2 (Mg> Fe) In 1.640 (H.D.): In 1.67: ,, , p = 633 nm 557 am n's (D, H, Z) ~= 1.596-1.694 ~ = 1. 605-1.710 y = 1.615-1.722 6 = 0.013-0.028 (+) (-) This sample (Connecticut): n= 1.659 p = 1.666 y=1.674 5a0.015(+) Figure 9. Anthophyllite. GRUNERITE (Fe}2, Mg)7(Bi8022) (OH)2 (high Fe) ffi 1.670: In 1.700: a = 627 nm Figure 10. Grunerite. n'e (D. H, Z) a= 1.663-1.686 S 21 1.681-1.707 y= 1. 697-1.729 5 = 0.034-0.043 (-) Thie sample: a=1.669 5=1.684 y=1.697 b = 0. 026 (-) 240

CROCIDOLITE Na2Fe32Fe23(Si8022)(OH)2 (no Mg, contains Fe+2) In 1.670: 0 =386nm 0=418 = 380 nm nm In 1. 700: w= 408 nm E= 1. 624-1.666 a=1.688 0= 1.703 y=1.708 6 = 0.010 (-) Figure 11. Crocidolite. APATTTE Ca5(P04)3(OH, Cl, F) ln 1.606 (H.D.): w = 408 = 448 nm Tn 1.64 E=715nm Onm n's (D, H, Z) Figure 12. Apatite. n's (D, H, Z) tt= 1.654-1.701 P - 1.662-1.711 y - 1.668-1.717 6 = 0.006-0.016 (-) This sample (Orange River, South Africa): w - 1.628-1.667 w= 408 nm 6= 0. 001-0. 007 (-) This sample: E = 1. 6285 w = 1.6357 6 = 0.0062 (-) 241

FORSTERITE Mg2SLO4 ln 1. 640 (H.D. ): n's (D, H, Z) r ** a= 1.635 -1. 827 ~ = 1.651 -1.869 y=1.670 -1.879 6 = 0. 035 -0.052 (+) (-) *pure forsterite **plns Fe2t replacing Mg giving fayalite, Fe2Si04 This sample: a= 1.643 0 = 1.663 y= 1.682 6 = 0.039 (-) Figure 13. Forsterite. HORNBLENDE )(OH,F7 (Na, K)0.1-0.7Os2(Mg, F 2{, F 3+, Al) 5 (Si 6_7, At 2-1O22 2 ffi 1.605 (H.D): , _ _ ,,,, -,,, n's (D, H, Z) a= 570 nm Is. 1.640 (H.D): , _ 518nm 475 Figure 14. Hornblende. a= 1.615-L 705 ~ a 1.618-1.714 y a 1.632-1.730 6 - 0.014-0.026 (+) This sample: a= 1.843 ~ = 1.650 y= 1.660 6=0.017 (+) 242

es In 1.580 (H.D.): In 1.605: WOLLASTONITE Ca(8103) n's (D, H, 4 a a= 412 am a= 1.616-1.640 = 1 628-1 650 p~340nm S . . y = 1.631-1.653 y s< 300 nm 6 = 0. 015-0.013 (-) This sample: a= 532 am Figure 15. Wollastonite. CALCITE CaCO3 a= 1. 612 9=1. 628 y=1. 632 6 = 0. 020 (-) In 1.64 (H.D.): n's (D, H, Z) s - 1.486-1.550 m - 1.658-1.74 6 - 0.172-0.190 (-) Tbie sample: s'= 1.525 c=1.486 ~=1.653 6=0.167(-) Figure 16. Calcite. 243

QUARTZ In 1.550 (H.D.): W= 682 = 682 nm 510 2 n's (D, H, Z) m=1.544 e = 1.553 682 5 = 0.009 (+) = 553 nm This eamplee m=1.544 [=1.553 Figure 19. Quartz. ORGANIC FIBERS (1. 550 liquid) ny- 630 nm Paper fiber figure 20. Organic fibers. Although we speak of dispersion staining colors as specific for a given substance in a given liquid (at a given temperature) we sometimes observe closely similar colors for other substances. We must, especially when this possibility exists, make sure that we observe enough data to be able to state with certainty that the substance is, say, chrysotile. It is not sufficient to observe the proper color in one direction - both chrysotile and paper fibers can show the same blue color perpendicular to their lengths. Nor is it sufficient to observe the two colors on a single view of a crystal - both quartz aad chrysotile can have two colors in common. If all colors shown by the crystal in all orientations correspond to the known data for a given substance, and if the crystal morphology shows the colors to be oriented properly, there is then very little chance of misidentification. 245 2063105040

Another serious complication, especially with minerals, is the eff-ect of substi+ tutionat solid s9lution op the optical properties. The substitution of F for OH , Fe2 for Mg2 , or Ca2 for 2Na can drastically change the optical properties of many minerals. One of the most serious in this respect is anthophyllite. Nominally Mg75is0Y2(OH)Z, anthophyllite forms a continuous series of solid solutions with iron replacing magnesium (Table 1) with corresponding changes in the refractive indices and dispersion staining colors. Anthophyllite can also have up to 14 percent MnO, 10 percent ZnO, or 15 percent A1203 with corresponding variations in the optical properties. Figures 8 and 9 show the dispersion staining properties for two different anthophyllites, one from Connecticut and the other from Georgia. In spite of the wide differences between these two anthophyllites, both samples show parallel extinction, a unique characteristic among the asbestos minerals, and the birefringence values, y-p, S-a, and y-a, as well as the optic axial angle remain quite uniform or change progressively and uniformly as the composition changes. If, for example, one observes refractive indices in the anthophyllite range, the possibility of tremolite, actinolite, ferroactinolite, or cummingtonite should be considered. The index range will tell which is present, and all of the latter differ from anthophyllite in that they show oblique extinction, usually about 20° rather than parallel extinction. In other words, anthophyllite is orthorhombic; all other amphiboles (and chrysatile) are monoclinic. a Table 1. Optical properties in Jhe anthaphyllite solid solution series. - - - - Refractive indices - - - - X Fe a A y 2V 0 1.596 1.608 1.615 120(-) 20 1.622 1.632 1.642 91(-) 40 1.641 1.650 1.665 685+) From Deer, Howie, and Zussman, "An Introduction to the Rock-Forming Minerals," Longmans, London (1966), pages 156-7. Many interfering substances are just not fibrous, hence they can be ignored if only asbestos is the target. Quartz has only two refractive indices, 1.544 (w) and 1.553 (s), but these fall within the range of chrysotile, a = 1.544 and y= 1.558. However, chrysotile is very fibrous whereas quartz is usually flakes or chips. Chrysotile shows three refrac- tive indices a, S, and y and a low 2V = 30-35° (+) and always shows nearly the maximum birefringence, 0.014 or 0.012. Quartz can show any birefringence value between 0.000 (w-w) and 0.009 (s-w) depending on orientation. Even a thin sliver of quartz oriented to show e and w (and therefore chrysotile colors) can be bounced into other more nearly isotropic orientations by tapping on the coverslip with a needle. Organic fibers are not generally confused with asbestos because they have obvious morphological differences, e. , pits, twists, central lumens, nodes, cross-over marks, etc. However, if mechanically broken down into tiny fibrils they lose this obvious morphology and some, e. , wool and other animal hairs, may closely resemble chrysotile in optical properties. A carefui exa.ination of such fibers morphologically and optically will usually, however, end any confusion and permit certain identification. Glass or mineral wool may happen to show a color near the chrysotile range but these, of course, are isotropic and morphologically quite distinctive. With careful application, dispersion staining is capable of rapid certain identifica- tion of any transparent substance whose optical and morphological properties are known. It also quickly differentiates between fibrous and nonfibrous minerals and detects traces of any substance in extraneous mixtures. 246 I

Reference [1] Brown, K.M., McCrone, W.C., Kuhn, R. and Forlini, L., Microscope 13 311; 14 39 (1963). Discussion J. ZUSSMAN: I enjoyed this very beautiful demonstration of the method. This is an academic question, but I think I remember a phenomenon called "form birefringence" which is supposed to be effective in giving peculiar results for very fine particles of small dimensions. If you have a very fine piece of an isotopic material, there is a shape factor which can make it appear to be anisotropic. I wonder if you get any anomalies with this method coming up, particularly with chrysotile, with fine fibrils, because of form. I think it is called form birefringence. J. DELLY: To answer your question, yes, there is an effect, but we don't apply this technique to a single isolated fiber, so there is not really much chance of being wrong on that. I agree with you, it is extremely fascinating academically, but in a practical sense with a bulk sample there are so many fascinating things associated with it that one spends actually a great deal of time with any one sample playing with colors. R. DRAFTZ: We have been using some of the techniques, and run into a problem with paper fibrils, especially with parenteral contaminants. I wonder if you tried the dispersion technique with chrysotile and with paper fibrils and perhaps found some similarities in color since the refractive index range is about the same. DELLY: You will see that the highest reported value of y of chrysotile (Deer; Howie, Zussman) is 1.556. ,1, in 1.550 HO refractive index liquid is about 515'nm. Figure 20 of the article shows that paper fibers in liquid 1.550 will show a>to of 450 nm parallel to the fiber. This wide difference in wavelengths should be easily discerned by most people. In any case, the microscopist is in the enviable position of settling the matter finally by resorting to the familiar cuoxam test to detemine whether a given fibril is cellulose or not. J. LEINEWEBER: I appreciate your very elegant description of the technique, and it has aroused a lot of questions in my mind about how the dispersion staining really works, but I would also appreciate a comment or two on the advantages of this technique over ordinary petrographic techniques for fiber identification, and also the size limits that you are confined to in working with particular particles. DELLY: Those are a couple of very good questions. First one: The major advantage is speed. For somebody who does primarily dispersion staining, he can complete an analysis in, probably, under five minutes. It is cheap and it is fast. It is a very quick survey type of thing, a very quick confirmation. I think that is probably the primary advantage of the technique. But the lower limit is a bit tricky. The abstract says that the major dimension should be one micrometer, which, if you are going to use 3:1, makes it about 0.3 pm or 0.25 Nm for the minimum. This technique does not depend on resolving power. It could not; otherwise you would not put all these stops in the back focal plane that deliberately destroy the resolving power. But, the spread of the light is all you are really looking for. You don't want to see the particle. So, that the lower limit is probably nominally around 0.3x1 pm. The reason I say nominally is, as with any other technique, when you go to the limits of any instrumental technique, the art starts coming in as well as the science. There is no reason though, why you could not apply this technique with higher-aperture objectives as well and still carry it further down. I have not personally done it. V. WOLK000FF: I just cannot see the advantage of this particular technique compared to classical techniques. For example, even if crocidolite does or does not show the blue color, you can pick it up immediately under crossed polars. We have no difficulty whatever using classical methods for the time element or whatsoever the case may be. And as one gentleman pointed out, for paper fibers or textile fibers we can pick that up instantaneously. Also, we are looking for the resolution, and, as you well know, materials 247 2063105042

containing asbestos fibers contain other materials as well. I must agree that the slides are extremely glamorous and picturesque, but I really believe that there is just no substitute for the classical petrographic or optical mineralogy when it comes to solid solutions that exist in several of these asbestos series. I just want to go on record on that. DELLY: Dispersion staining methods do not exclude classical methods; indeed, they are used simultaneously. The commercial form of the dispersion staining objective has three positions of use: a central stop, an annular stop for dispersion staining, and a position free of any stops which is used for classical methods in conjunction with dispersion staining. 248

National Bureau of Standards Special Publication 506. Proceedings of the Workshop on Asbestos: Definitions and Measurement Methods held at NBS, Gaithersburg, MD, July 18-20, 1977. (Issued November 1978) MINERAL FIBER IDENTIFICATION USING THE ANALYTICAL TRANSMISSION ELECTRON MICROSCOPE D. R. Beaman and H. J. Walker The Dow Chemical Company Midland, Michigan 48640 Abstract In a transmission electron microscope equipped with an energy dispersive spectrometer (EDS), It is possible to obtain the high resolu- tion morphology, crystal structure, and elemental composition of sub- micron mineral fibers, particulate, and thin films. The reliability of fiber analysis is enhanced when fiber identification is based on the nearly simultaneous determination of these three characteristics because each of the individual modes can yield ambiguous information. Energy dispersive spectrometer data can be converted to elemental fiber compositions using known standard spectra or relative sensitivity factors which can be calculated or experimentally determined for a given , instrumental configuration. Calculated and experimental sensitivity : factors are found to agree within 15 percent for photon energies above . 1.5 keV. The relative error in composition calculated from EDS spectra ; will generally be better than 10 percent, but only if the TEM column and ' components have been properly modified to reduce the effects of extraneous x-ray generation and electron scattering. The sources of these problems are described and a procedure for minimizing the effects outlined. Proper aperturing, collimation, selection of materials of construction, and operating conditions can provide useful mineral spectra. It is often necessary to correct for x-ray absorption even in fine mineral fibers, and this may be done using reference standards or sensitivity factors corrected for absorption. The effect of absorption increases rapidly as the difference between the mass-absorption coefficients of the elemental constituents of the mineral increases. Carbon contamination which degrades both EDS spectra and electron diffraction patterns can be minimized by using low current density and short analysis times. Less than 15 percent of the chrysotile fibrils in a standard provided positive selected area electron diffraction patterns (SAED), but up to 50 percent did have the correct layer line spacing. The fraction of fibers providing good diffraction increases rapidly as the number of fibrils in a fiber increases. The reported differences in SAED quality arise primarily because investigators use differing criterion for defining a positive SAED pattern and the fiber size distribution examined varies. Sample preparation methods were reviewed and it was found that condensation washing is only reliable if loss corrections are applied, particularly in the case of amphibole fibers. In spite of the many problems, inter-laboratory end multiple sample reproducibility in the measurement of fiber concentrations can be t30 percent when using good procedures. Key Words: Carbon contamination; electron diffraction; mineral fibers; transmission electron microscope; x-ray spectroscopy. 249

Introduction The need to identify and determine the concentration of small mineral fibers in environmental samples provided motivation for the development of the analytical trans- mission electron microscope (ATEM) which consists of a conventional transmission electron microscope (CTEM) equipped with energy dispersive spectroscopy (EDS) and possibly scanning transmission electron microscopy (STEM) capabilities. In such an instrument it is possible to obtain from very small volumes of material high resolution morphology in the TEM or STEM mode, elemental data using the EDS, and structural information for crystalline materials in the selected area electron diffraction (SAED) mode. When identification is based on the nearly simultaneous determination of three quantities-morphology, elemental composition, and crystal structure-the reliability of the analysis is significantly improved because the individual modes sometimes yield ambiguous information. The limita- tions of each mode have been discussed previously [1,2]1. All modes are adversely affected by the presence of adjacent non-fibrous debris and overlaying films. Fibers that are too thin or too thick do not provide sufficiently good SAED patterns for positive identification by comparison with standards. Less than 15 percent of the chrysotile fibrils in a particular standard gave positive SAED patterns. Chrysotile diffraction is further degraded by electron beam bombardment and instrumental contamination. Energy dispersive spectrometry is not a panacea because there are different minerals with similar compositions and elemental substitution is common. Morphology is often compromised by the environment and interfering solids. The hollow-core or tubular appearance of chrysotile is distinctive but often absent and degraded during analysis. It is difficult to establish a protocol for basing identification on three criteria, but when this is done the quality of the analysis is significantly improved. This paper describes some of the difficulties associated with fiber counting in the ATEM with the goal of circumventing the problems. The data from an energy dispersive spectrometer can be converted to chemical concentrations but there is a need to calibrate the instrument and correct for x-ray absorption even in very fine fibers. There are instrumental limitations which degrade EDS spectra but can, to some extent, be avoided. Contamination seriously affects both the EDS spectra and SAED patterns, but there is little that can be done to avoid it in existing instruments other than to understand the problem. The reasons for the controversy concerning the quality of SAED patterns from mineral fibers are examined and criteria suggested for classifying chrysotile SAED patterns. Sample preparation methods are reviewed and some results of inter-laboratory reproducibility are presented. Sample Preparation The three methods of water sample preparation that are commonly used are summarized in table 1 and references 1-6. Water is vacuum filtered through 0.22 pm Millipore or 0.1 pm Nuclepore filters. Nuclepore has the advantage of being smooth and therefore not generating a replicated structure when carbon coated; it has the disadvantages of being prone to fiber loss during handling and sporadic occurrences of non-uniform solids deposition during filtration. Millipore retains fibers well but generates a structured background if carbon coated prior to destruction of the filter structure. In the method of condensation washing [1,2,6], TEM grids with carbon-coated Formvar films are positioned on the Ni support screen of the cold finger in a condensation washer. A piece of Whatman filter paper placed between the TEM grid and the Ni support screen has been shown to reduce fiber loss during solvent extraction [7]. The grids are preconditioned by the application of a few drops of acetone beneath the Ni support screen to prevent warping of the filter section. The filter sections are placed, sample side down, on the TEM grid immediately following pre-conditioning. The Millipore is removed in 10-50 minutes of acetone vapor extraction. The complete procedure and sources of errors are described elsewhere [1,2]. 1Figures in brackets indicate the literature references at the end of this paper. 250 E F.n

Table 1. Method of preparing liquids for ATEM analysis. Method + reference filter medium pre-treatment fiber fixation by vacuum evaporation of carbon Jaffe-fusion 3,4 0.22 71m Millipore fused in acetone vapor for 5-10 minutes yes pre-conditioning none extraction filter section on configuration grid on polyurethane in enclosed petri dish Jaffe-wick 5,6 0.1 um Nuclepore none yes 10 uL droplet of solvent onto sample positioned on grid filter section on grid on wire mesh on several layers of filter paper in enclosed petri dish Condensation washing 1,2,6 0.22 pm Millipore none no acetone wetting of grid without filter filter section on grid on cold finger in reflux column solvent acetone chloroform acetone duration of 12 hours 10-24 hours 10-50 minutes extraction In the Jaffe-wick method [5,6], the Nuclepore filter is carbon coated after filtration to fix the solids in place prior to filter extraction. The TEM grid is positioned on a wire mesh placed on several layers of filter paper in a,:tri dish. The carbon coated filter section is positioned on a grid and a 10 p1 droplet of chloroform is added to prevent warping. The layers of filter paper are,saturated with chloroform and the Nuclepore extracted slowly (10-24 hours) in the covered petri dish. In the Jaffe-fusion method [3,4], a portion of the Millipore filter is attached to a glass slide and placed for 5-10 minutes in acetone vapor. This short pre-treatment in acetone destroys the structure of the Millipore and therein avoids the formation of a replicated network structure during carbon coating which would interfere with fiber counting. The fused Millipore on glass is carbon coated and then extracted using acetone in the same manner as in the case of the Jaffe-wick method. One of the prime sources of error in the analysis is the fiber loss which occurs during sample preparation. Condensation washing is a popular method of preparation, but it introduces variability in the results and yields higher fiber losses than Jaffe-type methods [1]. While some investigators have obtained good results with condensation washing [8,9], there are a sufficient number of technique problems [1,2] so that serious differences occur in inter-laboratory comparisons. It is possible to correct for the losses associated with condensation washing using partially-extracted Jaffe samples to determine the total fiber concentration [1]. This requires additional preparation time and TEM analysis. Fortunately the chrysotile losses associated with condensation washing are usually below 20 percent [1] and can be considered insignificant if the duration of wash is less than an hour in a properly controlled washer. We have obtained reproducible results using Jaffe extraction of carbon-coated Nuclepore [2] and loss corrections in conjunction with condensation washing. 251

All of the above discussion refers to water samples. In preparing air samples it is preferable to low-temperature ash the filter because of the heavy filter loading associated with air sampling. The ash is then suspended in water and processed as a water sample. Because the ash tends to be clumped, it is necessary to subject the suspended ash to ultrasonic treatment. Instrumental Limitations Instrumental problems arise when using energy dispersive spectrometers, because TEMs were never intended to be used in quantitative chemical analysis and ATEMs have been constructed by retrofitting EDS and STEM capabilities to existing systems. There are two prime sources of the instrumental problem: 1) the EDS is not a focusing spectrometer and is Insensitive to the location of the x-ray source and, thus, will detect all x-rays with a line-of-sight path to the detector [3]; 2) in a typical CTEM column there is, in a confined volume, a high density of hardware such as pole pieces, apertures, anti-contamination surfaces, sample grids, samples holders and associated clips. These two features combine to yield remote x-ray generation, i.e., x-radiation originating from regions outside of the volume excited by the primary electron beam. This causes: 1) spectral peaks unrelated to the sample to appear in the EDS spectrum leading to quantitative inaccuracy and errors in identification; 2) increased background radiation which raises the detectability limits; and 3) a loss in spatial resolution. The sources of the problem are secondary fluorescence by characteristic and continuous radiation generated in the column apertures, backscattered electrons from the sample and its support, and scattered primary electrons. The use of high voltages to penetrate thin samples and retain good spatial resolution leads to the generation of characteristic and continuous radiation in column apertures. The second condenser (C2) variable aperture, which is the last aperture above the sample, poses the most serious problem. The maximum in the generated continuum at a bevrt energy of 100 keV and PtKa characteristic radiation both have wavelengths of abou~ 0.2A and are readily transmitted by thin Pt apertures, e.g., over 40 percent of the 0.2 Pt radiation is transmitted by an 100 pm thick Pt aperture. Most of this radiation will be dissipated by absorption in the column but any that does reach the sample area can generate secondary fluorescence at and near the sample which is unrelated to primary electron beam excitation. Because almost all primary electrons are transmitted by thin films and small particles, the backscattered electron fraction is small as indicated for Au films in figure 1 [11]. If the beam voltage is high and the sample thin, less than 5 percent of the incident electrons will be backscattered. Any electrons that are backscattered toward the detector can penetrate the 7.5 pm Be window of the EDS because they will, for the most part, have energies close to the incident beam energy. Eighty percent of the 100 keV electrons can penetrate 7.5 pm of Be and in so doing lose less than 5 percent of their energy. Most backscattered electrons do not reach the detector because they are confined by the strong objective lens field. They can, however, excite remote particulate matter and the support grid. 252

Acceleration Potential In keV Figure 1. The percentage of backscattered electrons as a function pf incident electron energy for two different thicknesses of Au. The data are from Philibert and Tixier [11]. Scattered electrons in the column cause electron beam tailing [12] which leads to excitation of areas in the sample immediately adjacent to the region of primary beam excitation. This effect is due to improper alignment and scattering by column components and increases in severity as the beam voltage is lowered. The following list indicates some steps that may be taken to alleviate these instrumental problems. The magnitude of the problem and, therefore, the effectiveness of these alterations will vary appreciably from one instrument to another because of differences in electron optical configurations, alignment procedures, column cleanliness, aperturing (sizes, materials, thicknesses, and location), and operating mode (TEM vs. STEM). I. Reduce the generation in and transmission of radiation by column apertures. a) Use thick apertures [13] b) Use Pt apertures rather than Mo or Ta [12,14] c) Use column inserts somewhere between C2 and the sample [15] d) The use of low acceleration potential reduces this problem, but promotes beam tailing, backscattering, and absorption effects e) Determine if performance depends upon the emission current for the instrument being used and the type of sample being studied II. Reduce the excitation of material remote to the sample. a) Specimen holders, specimens clamps, and support grids should be made of low atomic number materials (Be, graphite, or polymer) or coated with such materials [1,13,16] b) Use support grids with maximum open area [13] 253 2063105048

c) Coat components near the specimen such as anticontamination devices and sample support rods with low atomic number materials (Aquadage) d) The objective aperture must be removed during EDS data acquisition e) The sample support film should be as thin and have as low an atomic number as possible f) Operate at as low a tilt angle as will provide adequate EDS intensities (less area of grid exposed to excitation) III. Optimize the EDS detector configuration. a) Use the greatest Si(Li) crystal-to-sample distance that will provide adequate count rates [17] b) Collimate the detector with a low atomic number material c) The collimator should be thick enough or shielded with sufficient material (high z) to absorb any stray radiation [18] IV. Minimize electron scattering a) b) c) Use a smail (100 pm) condenser aperture [14] Operate at high acceleration potential Have the column clean and properly aligned These effects of extraneous radiation can best be examined by comparing spectra obtained on and off the edge of a thin film or fiber or by comparing the spectra obtained with the beam positioned in a hole (hole-count) [12] with spectra obtained on the sample. In performing on- and off-film measurements on a Sn-Cu-Cr film, 3 percent of the Cr intensity was attributable to Cr plating on the sample hold-down clip while the Cu TEM grid was responsible for 15 percent of the Cu signal. Insertion of an aperture just beneath the variable C2 aperture on a Philips EM300 operated in the TEM mode increased the Cu peak-to-background ratio and reduced the off-film Cu by 35 percent. The maximum peak- to-background ratios have been achieved using a column insert (1 mm ID x 2.57 ms OD x 3mm thick) in the lower end of the vacuum tube through which the variable C2 aperture passes. Kyser and Geiss [18] have found that operation in the STEM mode reduces the extraneous background by about a factor of two. Even after these precautions have been taken, it is still advisable to subtract the off-fiber spectrum from the fiber spectrum and to use as dilute a sample as feasible. A high density of solids on the grid may reduce the analysis time required to find fibers, but it seriously degrades the quality of SAEO patterns and.EDS spectra. Quantitative Analysis There are two aspects to quantitative fiber analysis of environmental samples in the ATEM, namely, the proper identification of the fibers coupled with the accurate determina- tion of the number of fibers per unit area. When the concentration of a specific mineral is sought the best procedure is to compare unknown spectra and diffraction patterns with those obtained from well-characterized standards in the same instrument using constant operating conditions. When unknown samples are encountered, it is advisable to compare ATEM data with the results of x-ray diffraction, infrared spectroscopy, and x-ray fluores- cence in conjunction with a careful consideration of the mineralogy of the problem. When the fibers, particles, or films of interest are thin, the following expression, originally proposed by Duncumb [19] and pursued by Cliff and Lorimer [20] and Russ [21], can provide good results; 254

CA IA _ (p-B)A ZB SAS r SAS T"F B (1) where I is the net peak intensity corrected for background and peak overlap and SAB is a relative sensitivity factor, i.e., the ratio of the detected intensities (IB /IA ) for two pure thin standards of the same mass thickness. Absorption, secondary fluorescence, and backscattering effects must be negligible for eq. (1) to be applicable. SAB is most easily measured on multi-element thin standards of known composition. There are not many experimental data and the bulk of what is available has been pub- lished by Cliff and Lorimer [20] and Sprys and Short [22]. SAB can be calculated from the following expression which is fully discussed elsewhere [21,23,24]: SAS 4 AA C10 +ZT) GB 1 n CE~ / EC,A exp' o~ 8e 13.9x10 4, AB C10-~ GA 1 n CE~ ) EC,B exp `- p I Be 13.9x10 4 (2) The subscripts A and B refer to the elements A and B. A is the atomic weight, z is the atomic number, G is the fractional emission in the line of interest, e.g., G(Ka12) °Ko.2 intensity/(Ka12 intensity + KS intensity), Eo is the acceleration energy in keV, Ec ie the excitation energy in keV, and p/plBe is the mass absorption coefficient for A or B radiation by the 7.5 pm Be window on the EDS detector. Note that this expression shows no dependence on the instrumental configuration. However, SAS values determined in different instruments may differ from each other and from theoretical values because: 1) the contribution of secondary fluorescence, back- scattering, and beam tailing may be vastly different in different instruments; 2) the Be window thickness and detector efficiencies may be different and, in some instances, the Si dead layer and Si crystal thickness may be significant; and 3) the samples used to measure SAB may not be truly thin with respect to absorption. Figure 2 compares the values calculated from eq. (2) obtained using the Reed and Ware [25] values for G with the experimental values of Cliff and Lorimer [20]; the ratios are relative to Si, i.e., B = Si. As noted by Goldstein et al. [23] the agreement is poor below 2 keV and good above 2 keV. Table 2 also compares calculated and experimental SAB values. For SMg Si' SA1 Si' STi Si' and SF.e Si, the agreement in the experimental values is generally better than 13 percent (fractional standard deviation or coefficient of variation), notwithstanding the variation in experimental configuration and conditions. With the exception of the SNa Si and SMg Si, the agreement between theory and experiment is better than 15 percent. The SMg Si value determined from eight different mineral fiber standards using the data of Beaman and File [1] was 1.7 ± 0.2 (± 14 percent). This varia- tion is primarily due to inaccuracies in the bulk chemical analysis of the mineral fibers. If IC = 1 and the S values are all relative to Si, n SA,Si1A'i~A Si,Si1i ' 255 (3)

Table 2. Calculated and experimental values of the relative sensitivity factor, SA-Si for Ka radiation. Investigator and - - - - - - - - Experimental SA-Si Values--------- Conditions 5Ma-Si SMg-Si SA1-Si STi-Si SFe-Si SCu-Si Cliff & Lorimer[13] EMMA-4 100 kY 5.77 2.07 1.42 1.08 1.27 1.58 0=0° T=45° amphibole particles Beaman & Fiie[2] EM300 80 kV 1.7 ± 0.2 1.4 ± 0.2 1.25 0-39° 7=26° asbestos fibers=0.1 um Sprys & Short[41] EM300 100 kV 7.22 1.08 1.30 silicide particles Morgan et al.[30] EM300 80 kV 3.92 1.55 1.16 1.13 1.38 f<42° 3 ym iso-atomic drops Suzuki et al.[42] JEOL 100C 400 kY 1.7 1.3 2.5 0-0° mineral fibers ---------Calcu7atedSA-SiValues--------- Goldstein et al.[22] 100 kY. 1.66 1.25 1.12 1.16 1.33 1.59 This report Eq.[11] 100 kY 1.52 1.13 1.09 1.07 1.22 1.46 Russ[4] 700 kY 2.01 1.39 1.12 0.95 1.12 1.34 0= tilt angle T - x-ray take-off angle 256

Figure 2. Relative sensitivity factors, SA Si, for Ka radiation as a function of the atomic number of element A. The curves are calculated fr.om eq. (2) and the points are experimental values from Cliff and Lorimer [20]; from Beaman [24,i. Other relative sensitivity factors can be calculated from the Si values because SA8/SC8 = SAC. If the 5 values are not relative to Si n CA ° IA/(IA + 1. 5i AIi) . (4) iB ~ e We measured the composition of a 3000A thick Cu-Sn-Cr film on a Cu TEM grid using Philips EM300 CTEM at 80 keV and a Cameca electron probe operated at 25 keV. The results are shown in Table 3 and compared with bulk chemical results. The ATEM results are seriously degraded by the secondary fluorescence and electron scattering as evidenced by the high Cu value resulting from the use of a Cu TEM grid. Off-film spectra were subtracted from the film measurements. The Cr/Sn ratio which is independent of the scattering problems is in good agreement with the chemical data (relative error = 11 percent). The Cu grid was used to demonstrate the difficulties associated with quantitation in the ATEM. As indicated previously, the results will be improved by using low atomic number grids and grids that do not contain any of the elements present in the sample. The results obtained in the electron probe, where scattering problems are minimized by the instrumental configura- tion and the use of low acceleration potential, are excellent (relative error <10 percent). From these limited data and other reported results on thin films [20,26], we conclude that the thin film model of eq. (1) is valid and capable of providing relative errors of less than 10 percent when using experimentally determined 5., values. This represents reasonably good performance when compared with the 5 percent relative error obtained using EDS systems and bulk samples [27]. However, it rust be stressed that this will only be attained in CTEMs after taking the precautions described previously. The accuracy will be best when measuring concentration ratios. The presence of oxide films or organic contamina- tion on the surface and the tendency for surface segregation and particle inhomogeneity to occur complicates and degrades quantitative results. 257 2063105052

C7 a Table 3. Experimental composition of a 3000 A thick Cu-Sn-Cr film. Method Element Neutron activation ATEM at 80 keV with SAB values Electron probe at 25 keV with SAB values and absorption corrected Electron probe at 25 keV with 5AB values but no absorption correction Composition in weight percent ~ Cu Sn Cr Cr/Sn 14.6 77.6 7.8 0.101 27 67 6 0.090 15.6 76.7 7.6 0.099 16.4 76.3 7.3 0.096 Correction of Quantitative Data It has generally been assumed that if the sample was transparent to electrons, i.e., structure was visible in the TEM image, then the sample was sufficiently thin so that the only consideration necessary in quantitative analysis was the variation in x-ray generation by the primary electron beam. The loss of ionization through backscattering will generally be negligible for sub-micro diameter mineral fibers, if the acceleration potential is above 80 keV. From figure 1, it is seen that for an 1000A film of Au the voltage could be as low as 50 keV and the backscatter fraction still below 10 percent, whereas over 50 percent would be backscattered by a bulk material. Philibert and Tixier [11] have found that continuous fluorescence is negligible and that characteristic fluorescence will be negligible if p/p ' B line t«l. p/p is the mass Ialloy absorption coefficient for the exciting radiation, B, by the material. It is not presently clear how significant the characteristic fluorescence correction is for thin films because the limited accuracy of the analysis in most CTEMs obscures the effect of characteristic fluorescence. In order to make any corrections to the data, it is necessary to know the thickness which certainly complicates the analysis and detracts from the simplicity of standardless correction. However, for particles and fibers the thickness can often be accurately estimated from the TEM image. Absorption effects in the analysis of mineral fibers were reported by Beaman and File [1] and figure 3 shows the dependence of Ix/ISi on fiber size for various minerals. The ratio of intensity ratios at one fiber radius (rl) to those at another fiber radius (r2) can be determined from Beers law. 258

1.0 0.9 0.8 0.7 0.6 0.5 0.4 • • • • • ChrYSOtile Mg/Si Ferroactinolite Ce/Si F• 0.1 Grunerite M Si • n 0.3 0.2 0.1 Ok 0.9 • 0'8 Amosite Fe/Si 0.7 • • • • • 0.3 0.6 • x Hombiende AI/Si 02 •• 0.5 0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.B~ Figure 3. Elemental intensities ratioed to the Si intensity as a function of mineral fiber diameter. The scales for chrysotile, grunerite, and amosite are on the left and on the right for ferroactinolite and hornblende. / x \ 0 exp ' plm pm rl csc y I Ix \,J Ix \, \ S /i \ ~/ rl y C~/ rl eXp (- Plm pm r2 csc ~/ \10 / x \Ix / CIx / 2 exp ` pl pm r2 csc ~~ Si r m Si r2 exp '- pl pm r2 csc \ m (5) where p/p Im is the mass absorption coefficient for x or Si radiation by the mineral, pm is the mineral density, W is the x-ray take-off angle and (IX /ISi°)r /(Ix°/ISi°)r is the ratio of the generated intensities which is independent of r. The intensity is assumed to be generated at the center of the fiber. Rearranging yields Im ~ a pm cscV(r2-rl) (plm plmi) . 2 259 (6) N 0 a w F+ 0 ~ 0 ~ A

This expression provides a satisfactory fit (± 10 percent) to the experimental data in figure 3 except in the case of contamination at small fiber diameters [1]. Equation 6 illustrates that it is the difference between the mass absorption coefficients that deter- mines the magnitude of the absorption effect. When p/p'mineral >>'p/plmineral' a decrease in Iw/ISi occurs with decreasing size because the relative increase in emission will be greater for the element with the larger absorption coefficient. Thus, in grunerite there is a greater relative increase in Si emission (N/pl9~unerite = 1455) than in Fe emission (1'/plgrunerite - 65) and a subsequent 25 percent decrease in I(Fe)/I(Si) as the diameter decreases from 1.5 to 0.15 pm. When N/plmineral << N/p jmineral' Ix/ISi increases with decreasing size because the relative increase in emission is greater for x than for Si. Thus in grunerite, where N/plMg = 3460 and p/pISi = 1455, there is a greater grunerite grunerite relative increase In Mg emission and a subsequent 50 percent increase in I(Mg)/I(SI) as the size decreases from 1.5 to 0.15 pm. The easiest way of correcting for such effects is to use calibration curves of the type shown in figure 3. Combining eqs. (1) and (5) shows that (SA8)til(SA8)tZ = R2/Rl where t is the film thickness (r = t/2). In the case of a very thin film or fiber, taking the limit in eq. (6) as t approaches zero gives: AB not-so-thin , t PB u_ A In A8(thi--~ ~ -pfilm cscy ~ ( plfilm - plfilm ) (7) which is in accord with the expression published recently by Goldstein et al. (23]. The SCu Si' S5n Si and 5Cr Si values used to calculate the Cu-Sn-Cr values were corrected for absorption using SA8 (not-so-thin) values from eq. (7), and in all cases the relative error in concentration decreased as shown in Table 3. Figure 4 can be used as a guide to determine when an absorption correction is advisable. When the absorption coefficient difference for a given particle radius or film thickness is above the line, the absorption correction will be greater than 10 percent and should be taken into account. Many of the amphibole fibers with diameters of 0.2 pm and over require absorption corrections (1]. 260

10,000 5000 500 100 0 I 100 200 pfXmtflim/2 in µg/cm2 300 Figure 4. A(pIp) = u/p Ifilmne u/P Ifilmne (pt)film = film mass thickness. When the value of o(u/p) for a particular film thickness is below the line, the absorption correction will be less than 10 percent. The absorption correction will exceed 10 percent for values above the lines. The values shown for amosite and crocidolite indicate that the absorption correction is significant for relatively thin fibers. Instrumentally Induced Contamination Superimposed on the absorption effects just described is the sample contamination which occurs when the hydrocarbons from the vacuum pump fluids are decomposed by the electron beam and deposited on the sample surface [10]. The deposited thickness can, in time, represent an appreciable portion of the total sample thickness. The magnitude of the problem depends upon; 1) the cleanliness of the vacuum system; 2) the electron beam current density; 3) the duration of the analysis; and, 4) the difference in absorption by carbon for the x-ray lines of interest. The magnitude of the latter effect can be estimated from the following expression: with In {Ix/I5i)Woithoutnation = pc tc csc>y pICiK - plC (Ix/ISi)contamination 261 0.2 µm diameter amosite (Mg/Si) **-0.4 pm diameter crocidolite (Fe/Si) I (8)

where pC is the density of carbon and tC is the thickness of the carbon deposit in cm. Figure 5 shows the observed variation of IMg/ISi in chrysotile with time for different current densities. The analysis of small (300-400A) chrysotile fibers often requires a small electron beam (higher current density) and a longer analysis time (>5 minutes) to generate credible counting 'statistics. Even though p/pl~~ - N/pI~g is 800, the rapid decrease in IMg/ISi can only be partially accounted for by contamination implying other electron beam induced effects. When the difference in absorption coefficients Is small, contamination is not a serious problem as indicated in figure 5 for the Cu-Cr-Sn film. µ CuKa CrKa SnLa MgKa SiKa p o 5 14 51 1170 360 1.0 018 0.8 0.4 0. I 10 Chrysotlle Fibers 3.7 Km Beam a Chrysotile Fibril 0.6 pm Beam Cu-Sn-Cr Thin Film 0.7 pm Beam i I 20 30 Time In Minutes 1 40 50 Figure 5. Elemental intensity ratios as a function of the duration of electron bombardment in an ATEM operated at 80 keV. IM9JISi and ICr/ISn are plotted for chrysotile asbestos fibers and a Cu-Sn-Cr thin film respectively. The beam diameter for each analysis is indicated on the curves. The mass absorption coefficients for the indicated radiation by carbon are also shown. Optimu. Conditions for Analysis In thin films, theory predicts [24] that the peak-to-background ratio should vary approximately as In U with E , increasing rapidly at low U and then more slowly, where U is the over-voltage ratio, °acceleration potential/excitation potential. This is not always observed experimentally as shown in Table 4. The failure to increase continuously with voltage is, in part, due to the background contribution from extraneous radiation which varies from instrument to instrument. The superiority of the STEM (vs. TEM) configura- tion is indicated in Table 4 where the two STEM instruments have their best peak-to- background ratios at the highest voltage. Unfortunately, fiber or particle counting in the STEM mode is not practical [2]. When column modifications are completed, the optimum operating conditions should be experimentally determined for each instrument. Note that low voltage operation will promote absorption and backscatter effects and reduce the effectiveness of SAEO on thicker fibers. 262

Table 4. Experimental determinations of the acceleration potential providing the maximum peak-to-background ratios in the ATEM. E in keV for maximum Investigator Instrument and mode X-ray line opeak to background This report EM 300-TEM CuK 60 This report EM 300-TEM SnL 40 Russ[39] EM 300-TEM FeK 50 Joy & Maher[25] JEOL 100B-STEM M9K 100 Mizuhira[29] JEOL 100C-TEM Na-CiK 20-40 Ga11e et al.[19] Cameca-TEM A1K, Au 20 Geiss & Kyser[27] EM 301-STEM Fe and CuK 100 While there are some mineralogical ambiguities that cannot be resolved by EDS, a well-designed ATEM with the appropriate column modifications used in conjunction with good analytical procedure can provide distinctive mineral spectra that are of great utility in fiber identification. ' Selected Area Electron Diffraction Vastly differing claims have been published as to the utility of SAED in the identifi- cation of mineral fibers: Ampian [28] finds that positive identification using SAED is only forthcoming from carefully indexed patterns yielding accurate lattice parameters. Ross [29] found SAED patterns of asbestos minerals difficult to obtain and interpret and that 200 keV was required to have distinct patterns. Beaman and File [1] reported that only about 10 percent of the chrysotile fibrils examined in a standard gave distinct patterns (40 percent were crystalline). Biles and Emerson [30] reported that most chrysotile fibers in beer did not give identifiable patterns. Samudra [31] reported that 99 percent of the chrysotile fibers in the size range of 200-1200 A provided good patterns. Much of this variation can be accounted for. A distinctive SAED pattern for chrysotile: 1) has a characteristic layer line spacing; 2) is streaked in alternate layer lines; and 3) shows some characteristic reflections, e.g., those in the second row from center are often quite distinctive. We classify as positive only those fibers exhibiting all of these characteristics. Fibers showing only the correct layer line spacing as determined visually on the fluorescent screen are clas- sified as ambiguous; the streaking or characteristic reflections are not sufficiently distinctive to permit positive identification. Patterns without systematic reflections or distinctive layer lines are classified as unknown and the sum of positive, ambiguous, and unknown is termed crystalline. The percentage of fibers in each category has been deter- mined as a function of fiber size using different instruments, standards, and sample preparation methods. Droplets of 10 pL volume, prepared from the dispersion of a high purity chrysotile standard [32] in water, were placed on carbon-coated formvar films on TEM grids. The samples were examined at 00 tilt in a Philips EM300 at 80 keV and a JEOL 100B at 60 and 100 keV. Fiber searching was carried out in the selected area mode with the diffraction aperture in position and focused to minimize the time lapse between finding a fiber and obtaining a SAED pattern. The aperture size at the specimen level was 1-2 pm, the camera length was minimized, and the SAED patterns were focused with the diffraction and objective lens controls. 263 2063105058

Figure 6 shows that less than 15 percent of the individual chrysotile fibrils (300- 400 A in diameter) provide positive SAED patterns. A significantly larger portion (20-50 percent) do exhibit the correct layer line spacing (positive + ambiguous) as observed on the fluorescent screen. For the fraction of positive fibers to exceed 50 percent, the fibers must contain over 3 fibrils. 100 80 60 40 20 0 I 2 3 4 5 No. of Chrysotile Fibrils in Fiber e 7 Figure 6. The percentage of chrysotile fibers in a standard providing the indicated quality of the SAEO pattern is shown to depend upon the number of fibriis in the chrysotile fiber. The results obtained on two different instruments are plotted along with previously reported results [1j. All samples were prepared using 10 uLl water droplets containing suspended chrysotile. The results obtained in instrument B were similar at 60 and 100 keV. The lower two curves in figure 6 compare the present results with earlier work [1]. The differences are due to the present use of slightly more stringent requirements for positive identification and possibly to the use of different standards (Wards in reference 1 vs. Union Carbide). Figure 7 illustrates that the percentage of fibers providing diffraction patterns in every category is lower when using samples prepared by the Jaffe extraction of carbon-coated Nuclepore as compared to water droplets. This is presumably due to the carbon coating and/or the presence of some residual Nuclepore. Note that the positive fiber category is not significantly affected by sample preparation. 264

as 80 20 0 100 2 Positive + Ambiguous Positive In Water Drop After Jaffe Extraction Of Nuclepore 3 4 5 No. of Chrysotile Fibrils In Fiber a Figure 7. The percentage of chrysotile fibers providing the indicated SAEg pattern quality is shown to depend, to some extent, on the method of sample preparation. The results for 10 pL water droplets are compared with those obtained after Jaffe extraction of a Nuclepore filter in chloroform. All samples were examined in instrument A. 7 The primary reasons for the differing claims are the use of different criterion for classifying a pattern as positive and differences in the fibril content of the fibers being examined. A rigorous definition of positive SAED is needed if identification errors are to be avoided and interlaboratory agreement achieved. Figure 6 shows that over 70 percent of the fibers containing. three fibrils show the correct layer lines spacing (positive + ambiguous category). Most published SAED patterns are not from single fibrils as indicated by the presence of partial rings and diffraction spot smearing or multiplicity [28,33]. To a lesser extent, the reported variation is due to differences in: 1) standard source and treatment; 2) sample preparation methods; 3) instrumental capabilities; 4) operator judgment; and 5) diffraction technique. In the river, tap water, and lake samples we have studied, the chrysotile has consisted predominantly of fibers with 3 or less associated fibrils with single fibrils appearing most frequently. The fibers in 50 percent NaOH produced from chlorine cells using chrysotile asbestos diaphragms are predominantly fibrils and 80 percent have lengths less than 2 pm and 95 percent have lengths less than 5 Ns. Identification based on morphology or SAED alone in these cases has not been particularly reliable because less than 20 percent of the chrysotile fibers had a tubular appearance and only 5-30 percent gave positive SAED patterns. Those fibers identified as chrysotile had EDS spectra and fibril diameters characteristic of chrysotile. 265 2063105060

In counting fibers with the ATEM, searching with the diffraction aperture in place is not practical because the field diameter is decreased from about 7 pm to 1 pm. When counting in the TEM mode, the fiber is subjected to more electron beam bombardment before a diffraction pattern can be obtained. When searching with the diffraction aperture in position, the SAED patterns from chrysotile fibers containing three or less fibrils generally fade within 30 seconds to such an extent as to be unidentifiable. This electron beam induced change is due to dehydroxylization [28] and carbon contamination. Reliability of the Method If a sufficient number (typically 60-100) of fibers are analyzed [1,2], the method will generally provide concentrations that are accurate within a factor of two. The reproducibility is considered to be represented by the coefficient of variation or 100o/mean fiber concentration. Inter-laboratory reproducibility between two different Dow laboratories measuring chrysotile in 50 percent NaOH, which is a relatively clean sample, has recently been better than 20 percent (see Table 5). This is reasonably good performance for the small, amount of material being detected as shown in Table 5. The idgntification of an 1000A long chrysotile fibril corresponds to the detection of 3 x 10 18 grams of material [24]. The results will not be this good for a series of labora- tories using a variety of sample preparation techniques and differing criteria for fiber identification. Table 5. Experimentally measured asbestos concentrations. Sample Concentration in millions of fibers per liter Mass of asbestos in parts per billion by weiqht Midland, MI Tap Watera 0.6 0.001 Waste Water Effluenta 10-400 0.2-10 50% NaOHa 50-5000 0.5-40 Duluth Tap Waterb 25 25 50% NaOHa sample 1 Dow Lab A Dow Lab B 380 380 50% NaOH sample 2 380 300 50% NaOH sample 3 530 520 50% NaOH sample 4 1900 1500 a Chrysotile b Amphibole In order to achieve good reproducibility, we adhere to the following: 1. Use a sample preparation method with proven low fiber loss such as the extraction of carbon-coated Nuclepore [2,5,6] or apply a fiber loss correction to each sample [1,2]. 2. Count only samples that have a uniform distribution of solids on the TEM grid, i.e., the fibers per unit area should not fluctuate widely [1,2]. 266

C2 3. Count until a sufficient number of fibers (generally 60-100) have been detected so that number of fibers per unit area does not change significantly with additional counting [1,2]. 4. Use a sample volume that provides a particulate density with minimum inters ferences from non-fibrous solids. 5. Modify the TEM column to reduce electron scattering and secondary fluorescence. 6. Subtract off-fiber EDS spectra from fiber spectra. 7. Correct for absorption, when present, using standards or relative sensitivity factors. 8. Minimize contamination rates, when possible, by the use of low current density and short analysis times. 9. Experimentally determine the optimum acceleration potential which often differs for EDS and SAED performance, necessitating a compromise. 10. Use a reasonable and consistent scheme for classifying fibers. The authors wish to thank L. Sturkey and W. A. Knox of The Dow Chemical 6mpany, Walnut Creek, California, for helpful discussions concerning selected area electron dif- fraction and R. H. Geiss and D. F. Kyser of IMB, San Jose, California, for their komments and critical review of the manuscript. The assistance of D. J. Peterson of Dow Chemical of Canada Ltd. and E. B. Bradford of Dow, Midland, in performing the experimental measure- ments is also gratefully acknowledged. References [1] Beaman, D. R. and File, D. M., Anal. Chem. 48, 101 (1976), also in Proceedings Microbeam Analysis Society igth AnnualZ'onference, paper 31 (1975). [2] Beaman, D. R. and Walker,. H. J. , in FDA Symposium on Electron Microscopy of Microfibers (Aug. 1976) in press. [3] Ortiz, L. W. and Isom, B. L. , in 32nd Annual Proceedings of EMSA 554 (1974). [4] Zumwalde, R., In FDA symposium on Electron Microscopy of Microfibers (Aug. 1976) in press. [5] Cook, P. M., Rubin, I. B., Maggiore, C. J., and Nicholson, W. J., in Proceedings of International Conference on Environmental Sensing and Assessment Section 34-1 I.E.E.E. Las Vegas (1976). [6] Anderson, C. H. and Long, J. M., Preliminary Interium Procedure for Fibrous Asbestos, U.S. Environmental Protection Agency, Athens, GA (1977). [7] Benefield, D., The Dow Chemical Company, Freeport, Texas, private communication (1977). [8] Millette, J. R., in FDA Symposium on Electron Microscopy of Microfibers (Aug. 1976) in press. [9] Stewart, I., in FDA Symposium on Electron Microscopy of Microfibers (Aug. 1976) in press. 267 2063105062

[10] Beaman, 0. R. and Isasi, J. A., Electron Beam Microanalysis, 5TP506, American Society for Testing and Materials, Philadelp ia 9722T [11] Philibert, J. and Tixier, R., in Physical As ects of Electron Microscopy and Microbeam Analysis; Seigel, B.M. and Beaman, D.R., ds John Wiley and Sons New York, 333 1975). [12] Bolon, R. B. and McConnell, M. D., in Scanning Electron Microscopy/IITR[/SEM/76, Part 1 (1976). [13] Russ, J. C., in Scanning Electron Microscopy/IITRI/SEM/77 1 335 (1977). [14] Joy, 0. C. and Maher, 0. M., in Scanning Electron Microscopy IITRI/SEM/77 1 325 (1977). [15] Zaluzec, N. J. and Fraser, H. L., in Proceedings Microbeam Analysis Society, 11th Annual Conference, paper 14 (1976). (16] Packwood, R. H., Laufer, E. E. , and Roberts, W. N., in Proceedings Microbeam Analysis Society, 12th Annual Conference, paper 115 (1977). [17] Geiss, R. H. and Huang, T. C., X-ray Spectrometry 4 196 (1975). [18] Kyser, D. F. and Geiss, R. H., in Proceedings Microbeam Analysis Society, 12th Annual Conference (1977) paper 110; also private communication with R.H. Geiss (1977). [19] Ouncumb, P., J. de Microscopie 7, 581 (1965). (20] Cliff, G. and Lorimer, G. W., J. Microscopy 103, 203 (1975). [21) Russ, J. C., in Proceedings Microbeam Analysis Society, 8th Annual Conference (1973) paper 30; also in Edax Editor, 5, 11 (1975); also J. Submicr. Cvtol., 6, 55 (1974). [22] Sprys, J. W. and Short, M. A., in Proceedings Microbeam Analysis Society, 11th Annual Conference (1976) paper 9; also private communication with Sprys, J.W. (1977). [23] Goldstein, J. I., Costley, J. L., Lorimer, G. W., and Reed, 5. J. B., in Scanning Electron Microscopy IITRI/SEM/77 1, 315 (1977). [24] Beaman, 0. R., in Modern Techniques for the Detection and Measurement of Environmental Pollutants, 10th Rochester International Conference on Environmental Toxicity (May 1977) in press. [25] Reed, S. J. B. and Ware, N. G., X-ray Spectrometry 3, 149 (1974). [26] Rao, P. and Lifshin, E., in Proceedings Microbeam Analysis Society, 12th Annual Conference, paper 118 (1977). [27] Beaman, D. R. and Solosky, L. F., in Proceedings Microbeam Analysis Society, 9th Annual Conference, paper 26 (1974). [28] Ampian, S. G., in FDA Symposium on Electron Microscopy of Microfibers (Aug. 1976) in press. (29] Ross, M., in FDA Symposium on Electron Microscopy of Microfibers (Aug. 1976) in press. [30] Biles, B. and Emerson, T. R. , Nature, 219, 93 (1968). [31] Samudra, A. V., in Scanning Electron Microscopy IITRI/SEM/77 1(1977). 268

SS [32] Union Carbide chrysotile standard provided by K.S. Chopra, Union Carbide Corp., Niagara Falls, NY (1977). [33] Mueller, P. K., Alcocer, A. E., Stanley, R. L., and Smith, G. R., Asbestos Fiber Atlas, Environmental Protection Technology Series, EPA-650/2-75-036 (1975). Discussion K. HEINRICH: When you showed the variation of intensity with fiber diameter, was the scale in micrometers? D. BEAMAN: Yes. P. McGRATH: What can be done to develop criteria to reduce the energy-dispersive interferences so that we can develop criteria for asbestos? BEAMAN: We can do much better with the EDS spectra than in the past by making column modifications and by subtracting background spectra from the fiber spectra. Question (inaudible): BEAMAN: You can make an identification in the STEM mode, but you cannot count fibers easily. It would be difficult to continuously switch from TEM to STEM. C. PARMENTIER: I would like to make a comment concerning TEM-SAED and the lack of d-spaces and difficulty in measuring them for single-fiber chrysotile or amphibole asbestos in small particulates; we run into the same problem of rapidly decreasing signal intensity. We have used a cold finger with liquid nitrogen which allows d-spacings to be resolved on the screen, photographed, and subsequently measured and indexed directly on the negative, so we come up with very accurate d-spacings. The second point I'd like to make is in the spectrometric measurement of Mg-Si ratios. Have you seen varying Mg-Si ratios from chrysotiles of different locals, and is this taken into account in your analysis? BEAMAN: We have used two chrysotile standards, but the chemical differences are smaller than data reproducibility. We could not detect any trend. We, of course, use a cold finger but still observe the rapid deterioration of SAED patterns in the case of chrysotile. Amphibole patterns on the other hand do not tend to fade. N O ~ W 269 0 un ~ .P

Cs National Bureau of Standards Special Publication 506. Proceedings of the Workshop on Asbestos: Definitions and Measurement Methods held at NBS, Gaithersburg, MD, July 18-20, 1977. (Issued November 1978) TRANSMISSION ELECTRON MICROSCOPICAL METHODS FOR THE DETERMINATION OF ASBESTOS Ian M. Stewart Walter C. McCrone Associates, Inc. Chicago, IL 60616 Abstract Three 'criteria are given for the identification of a mineral fragment as asbestos: morphology, crystallography, and chemistry. The derivation of this information in the transmission electron microscope is discussed. Quantification of asbestos fiber content in an environmental sample is considered and currently practiced techniques for quantification both by mass and by number are reviewed. Key Words: Analysis; amphibole; asbestos; electron diffraction; electron microscopy; fibers; transmission electron microscopy; x-ray energy analysis. The first meeting on methodology for determination of asbestos by electron microscopy was held almost exactly seven years ago. Sponsored by, as it then was, the National Air Pollution Control Administration, it was attended by about a dozen people. The explosion of interest in asbestos has led to a series of methodology meetings, particularly over the last two or three years, culminating in the massive attendance at the present meeting. It is clear, therefore, that there is considerable interest in asbestos and in particular, asbestos methodologies. There is thus no need to reiterate the reasons for this interest here. What may be less obvious however, is why there should be such a necessity for the development of electron microscopical methods. Figure I shows an electron micrograph of a standard suspension of an ultrasonerated chrysotile sample which has been prepared to simulate material shed from asbestos filters used for parenteral drugs. The size range represented is quite wide and very closely approximates that which has been found in liquids filtered through an asbestos filter. If such a sample were to be characterized entirely by light microscopical methods, much of the material which can be seen in the electron microscope, for example fibers A and B in Figure 1, would be completely omitted. Figure 2 is an environmental sample, taken approximately three miles down stream from an asbestos plant and here again we have material below the detection range of the light microscope. The level of asbestos fibers determined by electron microscopy in this case was of the order of 108 - 10s fibers/liter, several orders of magnitude higher than would have been determined if the light microscope was used. Again, in water samples from the Duluth and Silver Bay areas, the number of asbestos fibers that were identified by light microscopy was virtually zero, fewer than one dozen fibers being detected in over fifty samples by this method. Nevertheless, transmission electron microscopy, as shown in Figure 3, established that there were indeed high levels of fibrous amphiboles in these samples. Clearly then, in order to satisfactorily characterize the asbestos content of such samples, electron microscopy is a necessity. Preceding page blank .~ 0 ~ 0 ~

Figure 1. Ultrasonerated chrysotile suspension simulating size distribution of fibers shed from asbestos filters used for parenteral drugs - 3200 X. 272

Figure 2. Filtered river water 3 miles downstream from an asbestos processing plant - 20.000 X. N 0 273 w r. 0 ~ 0 4 1

Figure 3. Water from western arm of Lake Superior - 12,600 X. Before discussing methods of preparing samples for examination in the electron micro- scope or for counting them, it is necessary to be sure what information we need to derive from the electron microscope in order that we can characterize a particular particle as an asbestos fiber. If one accepts the Federal Register definitions of asbestos and, from a legal standpoint, that is all that one can use at the present time, then to determine an asbestos fiber, one must show first that the material is fibrous, that is, that it has an aspect ratio of greater than 3:1 and, second, that it is a mineral of the type which is classed as asbestos by the Federal Register. The determination of the aspect ratio is quite straightforward. One measures the length and the width of the particle. The determination that the particle is indeed asbestos, however, is not so straightforward. There are basically two criteria which must be satisfied for a positive identification, certainly on the amphiboles, although for chrysotile perhaps only one of these criteria will suffice. These criteria are, firstly, that the particle in question belongs to the correct crystallographic system and has the correct crystallographic parameters for one of the asbestos minerals. Because of the unique structure of chrysotile, which will not be discussed here, the diffraction pattern of chrysotile can be regarded as sufficiently definitive without the addition of chemical information (Figure 4). In the case of the amphiboles, the diffraction patterns are less characteristic and careful diffraction work must be performed to establish that the particle is indeed an amphibole. Having established that it is an amphibole, one must then differentiate which of the several 274

S2 amphibole types it may be. This can best be performed by chemical analysis in the electron microscope. At the present time the most popular method of determining this analysis is by use of an energy dispersive x-ray analyzer, fitted to the transmission electron microscope. Figures 5 and 6 show, respectively, the electron diffraction pattern and the energy dispersive spectrum of an amphibole fiber which can be tentatively identified as the commercial asbestos "amosite" - actually a fibrous grunerite. The word 'tentatively' is used deliberately since there are -many problems associated with the interpretation of both the diffraction pattern and the energy dispersive spectrum. Thus, in general, it is prudent only to classify an amphibole as being within a certain series, such as the tremolite-actinolite series, or the cummingtonite-grunerite series. Figure 4. Chrysotile diffraction pattern. 275

Figure 5. "Amosite" diffraction pattern. © F. 4 o f Yf. . jY ~I ~~~ e.+ na.aa Co navo1e Figure 6. "Amosite" energy dispersive x-ray spectrum. 276

SS In order to determine these parameters simultaneously in the electron microscope, some idea should be given of how this information is derived. The morphology is obvious; this follows from the normal operation of the microscope as an image producing instrument. However, like all optical systems, the laws of diffraction apply in the transmission electron microscope. Thus, given an object with a periodic structure, the image of this object in the back focal plane of the objective lens will be a diffraction image related to the periodicity of the structure. In the transmission electron microscope, this image may be observed at higher magnification by adjusting the strength of one of the projection lenses, such that the back focal plane of the objective lens is in focus at the final viewing screen. As was stated above, the chemical nature of the particle under investigation can also be determined in the microscope by an energy dispersive x-ray system. This is because striking a target with a high energy electron beam will result in the emission of x-rays whose wavelengths or energies are characteristic of the chemical species at the point of impact. By suitably focusing the incident beam it is possible to isolate individual particles in the microscope and to either analyze their energies by an energy dispersive spectrometer or their wavelengths by a wavelength spectrometer. In practice the energy dispersive spectrometers are more common. They have the advantage that they detect all elements simultaneously from about sodium upwards in atomic number and they are also considerably cheaper to install on an instrument than the wavelength dispersive system which, although having a better signal to noise ratio, suffers a major disadvantage for rapid analysis in that it is sequential, analyzing only one element at a time. There are many factors which may interfere with or disturb the energy dispersive signal; factors such as particle size, shape, geometry, scatter from the instrument, and so forth, confuse the already complex chemistry of the amphiboles. These have been discussed in many other sources and will not be discussed in detail here. One should, however, be aware that such complications do occur and should interpret the spectia with appropriate caution. Having settled on criteria by which one would identify the fibers, the next problem is, "What does one wish to count or measure?" There are two philosophies which are current. One is that the important factor is to determine the number and size distribution of the fibers present as they exist in the sample. The other philosophy is that the mass concentration is important. We should discuss a little why these two schools of thought have arisen. It would seem to the lay observer, that, as yet, there is no sound medical reason in favor of determining one or the other. There are sound analytical reasons for suggesting either. The most attractive feature of the fiber number, size distribution and shape philosophy is that as well as giving information on levels that exist in the material, it also gives the size range, which may or may not be important, and much recent work has suggested that it is. It is also possible, by factoring in a geometric factor together with a density factor, to determine the mass of fiber present. One of the major drawbacks of such a method, however, is the tendency of fibers to overlap each other and also to overlap other material in the sample. It is particularly common in quarry samples, for example, to find intergrowths of chrysotile with the related serpentine mineral antigorite. Unless a good separation between the antigorite and the chrysotile is obtained, it may not be possible to positively identify the asbestos fibers and hence they will not be included in the count. Repeated over many fibers, and bearing in mind the multiplication factors which exist by virtue of the difference in area examined in the microscope relative to that represented by a membrane filter area, this can lead to quite dramatic differences in fiber counts or mass levels detected. In addition, the presence of one or two massive fibers can drastically skew the mass number, again because of the multiplication factors involved. Mass concentrations have been determined by several workers, and several methods exist for preparation of samples to determine mass reasonably accurately. These methods, developed principally by Battelle, Mt. Sinai, and Johns-Manville, and ideally, applicable only to chrysotile asbestos, all involve the reduction of more massive fibers to the so- called unit fibril of chrysotile. In some methods these fibrils are then individually measured for length, and by geometric calculations the mass is deteimined. In the Battelle method, the intercepts of fibrils along a line are counted and compared to similar counts performed on a standard mass concentration sample. The advantage claimed for such methods is that they will separate the fibrils from interfering material. One disadvantage is the several preparation steps which may be involved in preparing the sample and which may lead to either cross contamination of the sample or loss of material 277 2063105071

from the sample, leading to high and low readings, respectively. Additionally, such methods may liberate fibers which would not normally be considered free fibers and therefore presumably not hazardous. Details of these procedures have been published previously and will not be reiterated here. As regards methods for sample preparation for fiber counting without destroying the identity of the fibers, one might say there are as many variations of sample preparation methods as there are electron microscopists working in this field. The state of the art does, however, seem to have boiled down to two basic direct transfer methods, one using condensation washing and one using a wicking technique. These methods will be discussed by Dr. Anderson, who has prepared an excellent document entitled "A Preliminary Interim Procedure for Determining Fibrous Asbestos", which spells out the basic steps in preparing samples and the criteria for asbestos identification. I believe this document represents the most acceptable state of the art on asbestos determination by transmission electron microscopy at the present time. Although there have been other methods proposed, these have not received as wide favor as the direct transfer methods. These other methods include placing a drop of the fluid suspected to contain asbestos on an electron microscope grid with a calibrated micro pipette. Assuming that all the material from the drop is deposited on the grid uniformly, and knowing the volume of the micro pipette, it is possible to derive the number of fibers per unit volume of fluid. In a similar method a calibrated micro pipette is not used, but a small drop of the liquid is placed on a grid and the diameter of the area occupied by the deposited solids after the droplet has dried is measured. It is assumed that the diameter of the evaporated circle represents the diameter of the original drop and hence the volume of the drop may be calculated and again the number of fibers per unit volume determined. One of the major drawbacks of many of these direct drop emplacement methods is the difficulty in holding the liquid in such a manner that none of the drop is transferred off the grid to its surroundings, for example by wicking up between the arms of a pair of tweezers or by contact with the substrate on which the grid may be supported. An additional disadvantage is the tendency for size separation to occur within the drying drop, resulting in an uneven distribution of fibers on the grid. - In any event, in any method involving direct transfer either from a liquid or from a filter it should be borne in mind that due to the overlapping nature of the particulate species present, the possible ambiguities of interpretation of diffraction patterns and/or chemistry due to such overlaps and the inability to see many of the fibers, the number of fibers counted will, in all cases (with the exception of bad housekeeping resulting in contamination), result in a minimal number for the total fiber loading per unit volume. A truer estimate of the loading per unit volume may be made by applying corrections for such overlaps or by additionally counting those ambiguous fibers which cannot be directly identified. There is, however, no hard and fast rule as to the magnitude of such corrections. In the case of methods reducing fibers to unit fibrils and estimating mass, these will again be minimal numbers if the criterion used is that the fiber must be positively identified, as, it is more difficult in general to obtain a positive identification of a small fiber than a large one either by electron diffraction or by chemical characterization. Although this may paint a rather pessimistic picture in terms of establishing a standard using electron microscopy, some positive suggestions may be put forward. For example, if it is decided that the standard should be a certain number of asbestos fibers per unit volume, then it should be possible to set up the microscope parameters such that the microscopist can determine all fibers in a unit area quite rapidly. This can, in turn, be calibrated in terms of fibers per unit volume of the sample source. If none of these fibers are asbestos and the number is still below the statutory limit, then clearly it is not necessary to perform any identification on the fibers to determine if they are asbestos or not. Such a procedure could well be used for screening purposes. Again, a subjective opinion could be made by the microscopist as to what percentage of those fibers are asbestos. If the total fiber content was 2 or 3 times that which is permitted by the regulation but the asbestos content is clearly, say 10 percent, of the total fiber content, then again there should be no major problem. This would dramatically reduce the number of marginal cases in which the total asbestos content may be close to or exceed the statutory limit. Only in such cases would it be necessary to perform a complete and detailed analysis. It would be necessary, of course, to ensure valid documentation uf the 278

data in- those cases where it is said that the level does not exceed the statutory limit. A similar approach could also be applied to the mass method and indeed may be more readily applied if one is already estimating mass on the basis of number intercepts per unit area. In the foreseeable future it is quite conceivable that automated methods for deter- mining asbestos in the electron microscope may come to be a reality. The application of computer solution to the electron diffraction pattern as described by Fisher and Lee in these proceedings could be combined with the capability for electronically recording such diffraction patterns which is offered by the technique of scanning electron diffraction. This could then be integrated in one instrument with an x-ray energy dispersive x-ray system, and electron energy loss analysis system operating in the scanning transmission mode to provide a valuable and powerful tool for automating the asbestos identification process. It is unlikely, however, that such a tool would be applied on a routine basis, in view of the capital cost which would be involved. Thus, there remains the major problem of characterizing asbestos particles in the submicroscopic size range and doing this economically. Work is currently in hand to effect separation of asbestos from other mineral species; separation from organic material may already be achieved by such techniques as low temperature ashing. Assuming that such separation can be both successful and complete the analytical procedures may well be simplified. Until such time, however, transmission electron microscopy must remain primarily a technique applicable to the research situation and is not presently an economically viable tool for monitoring and control programs on an extensive scale. References In trying to put together specific references to techniques mentioned in this pa~er, I realized how much of my information had been absorbed through discussions both privately and at meetings such as this -- a sort of mental osmosis. The following list is therefore not complete and should be more properly regarded as suggestions for further reading. I apologize in advance to those who may feel slighted by the omission of references to their work. Descriptions of the mass method can be found in the following: Leineweber, J. P., "A Method for Determination of the Fiber Content of Water", Johns- Manvilie Research and Engineering Center, Report No. E 404-37, August 1968. Thompson, R. J. and Morgan, G. B., "Determination of Asbestos in Ambient Air," Proc. International Symposium on Identification and Measurement of Environment Pollutants, p. 154, June 1971. Descriptions of direct transfer methods from filters are given in: Anderson, C. M., "Preliminary Interim Procedure for Determining Fibrous Asbestos," July 1976. Available from Dr. C. M. Anderson. See also Dr. Anderson's paper in these proceedings. Overlap, loss, and similar problems: Knight, G., "Overlap Problems in Counting Fibers." AIHA Journal, p. 113-114, February 1975. Beaman, D. R. and File, D. M., "Quantitative Determination of Asbestos Fiber Concentrations," Anal. Chem., 48, No. 1, p. 101-110, 1976. Energy dispersive x-ray analysis is discussed in: Beaman's paper cited above and Ruud, C. 0., Barrett, C. S., Russel l, P. A., and Clark, R. L. ,"Selected Area Electron Diffraction and Energy Dispersive X-ray Analysis for the Identification of Asbestos Fibers, A Comparison," Micron, 7, p. 115-132, 1976. 279 2063105073

Several general papers on the characterization, identification, and quantification of asbestos also appear in the Proceedings of the First FDA Office of Science Summer Symposium on Electron Microscopy of Microfibers, August 1976, currently in press. Discussion C. ANDERSON: Ian, it strikes me that to determine mass and to determine the number of fibers at a certain period of time are entirely incompatible for the reason that you state-that 10 percent or even less of the fibers contribute to 90 percent of the mass. I. STEWART: That's exactly right. ANDERSON: Therefore, for any kind of precision of mass you must count possibly 100 large fibers. STEWART: Or 1000 fibrils or you just look at your intercept. But I wasn't putting it forward as being a way that we should go. You see the big problem is that you're like me, you're an analyst too, and the medical people haven't decided what they want from us -- mass data or fiber counts and sizes. If that problem is resolved, so too will many of the analytical problems. ANDERSON: I wonder if you agree that determining mass and the number of fibers in the same amount of time is almost incompatible within a certain precision? STEWART: Yes and no. You can get a mass number out. If you're too lazy to look at the statistics of the size distribution, the mass will give you an idea of whether you've got a lot of big fibers there; not always, but sometimes. Written comments by Prof. J. Zussman to Dr. Stewart's paper. J. ZUSSMAN: Dr. Stewart mentioned that fiber counts by electron microscopy would be expected to be in error on the low side, especially through overlapping particles. This effect can be lessened, of course, if specimen preparation is such as to produce not too dense a fiber population on the elm grid. I would also like to mention that there are two factors leading to erroneously high fiber counts - the use of the rub out technique, and the process of ultrasounding if too vigorous. STEWART: I agree in part. However, in the case of overlaps due to other suspended particulates, dilution may produce too low a fiber population for the data to be statisti- cally valid. I also agree on the comments an erroneously high fiber counts. The rub out technique is only valid for mass data although I know that fiber counts produced by this technique have been quoted by some people. 280

(cs National Bureau of Standards Special Publication 506. Proceedings of the Workshop on Asbestos: Definitions and Measurements Methods held at NBS, Gaithersburg, MD, July 18-20, 1977. (Issued November 1978) STATISTICS AND THE SIGNIFICANCE OF ASBESTOS FIBER ANALYSES J. P. Leineweber Johns-Manville Research & Development Center Denver, Colorado Abstract The analysis of asbestos fibers by electron microscope methods involves many operations, each of which can affect the final results. Normal random fluctuations can be described by the Poisson distribu- tion, which applies to any truly random process. Deviations from normal statistics, sample preparation losses, identification errors, and laboratory contamination are sources of error which are difficult to quantify. Each, however, can cause variations which will be greater than predicted by the Poisson distribution. The significance of each of the sources of error are discussed together with recommendations for experimental techniques, which should minimize•the errors. Key Words: Analysis; asbestos; electron microscope; errors; fiber; statistics. Introduction The counting of asbestos fibers by the "membrane filter" method, approved by the National Institute of Occupational Safety and Health, has been studied in considerable detail [1,2,3,4]1. The procedures to be followed are specified in detait, and the precision and accuracy of the results have been analyzed by competent statisticians. The background data are based on several controlled experiments designed to describe the variations which can occur between operators in a given laboratory, as well as the variations which can occur between laboratories. Although there is considerable debate over the lower limit of fiber concentrations that can be accurately determined, the fluctuations that can occur with standard samples have been described to a reasonable degree. In recent years, there has been increasing emphasis on the quantitative determination of fiber concentrations in the environment [5,6,7]. Analysis of these samples is much more difficult because of the extremely low fiber concentrations, the very small fiber dimensions involved, and the high concentrations of extraneous materials in the sample. Traditional methods of analysis cannot be used, so the analyst must rely upon the electron microscope to resolve, identify, count, and measure the fibers. This requires the intro- duction of several additional sample preparation techniques. Furthermore, the fraction of the sample actually examined is extremely small and there is much more latitude for operator interpretation. The objective of this paper is to review the various sources of error in the counting of asbestos fibers by electron microscope methods, discuss how they might influence the results, and finally, suggest steps which might be taken to minimize these errors. 'Figures in brackets indicate the literature references at the end of this paper. 281

Electron Microscopic Fiber Analysis Procedures The techniques used to determine asbestos fiber concentrations with the electron microscope have gone through several evolutionary changes during the past decade. Although a"standard" procedure has yet to be agreed upon, all use most of the following steps (8,9]. Sample collection Deposition on Filter Ashing and refiltration . Clearing of the filter Scanning and counting Each of these steps involves manipulation of the sample in the field or in the laboratory. Errors can be introduced with each step, and, as in any sequential system, the errors will be accumulative. The following are the principal factors which can influence the accuracy and precision of the analysis. Normal statistical fluctuations Deviations from normal statistics Sample preparation losses Identification errors Laboratory contamination The significance of each of these sources of error will be discussed in more detail in the following sections together with recommendations for experimental techniques designed to minimize the errors. Normal Statistical Fluctuations - The Poisson Distribution In environmental systems such as air and water, it is reasonable to assume, as a first approximation, that the fibers are distributed in a purely random manner. Furthermore, it is also reasonable to assume that the random distribution will be maintained during the deposition of the sample on a filter. If this is the case, the variations to be expected can be described in terms of the Poisson distribution [10]. The distribution function can be represented as: f(x, m) _ mxe-m XT- where: m= the mean value of a parameter for a series of trials x = the actual value for a specific event e = the base for natural logarithms f = the probability of occurrence for a specific value. Figure 1 is a plot of the probability of occurrence for specific events for a Poisson distribution with a medn value of 10.0. The Poisson distribution is actually a limiting case of the more general binomial distribution. It has the unique characteristics that: - the variance is equal to the mean - the standard deviation is equal to the square root of the mean. For the fiber counting problem, the most significant characteristic is that the variance will be dependent on the total number of fibers counted-regardless of the number of fields that were examined to obtain the results. 282

Cos : 0.130 1 Io ~ x 0.120 0.110 0.100 0.090 0.060 0.070 0.060 0.050 0.040 0.030 0.020 0.010 0.000 0.0 10.0 20.0 COUNT Figure 1. Poisson distribution mean = 10. N O 283 W 0 w 0 V v

C;b The consequences of the foregoing characteristics of the Poisson distribution are best illustrated by using the "two sigma" limits to define the range within which the results might be expected to fall for given total fiber counts. The "two sigma" limits are chosen on the basis of the hypothesis that about 95 percent of the results should be within two standard deviations of the mean value. Table 1 lists the "two sigma" limits for total counts ranging from 1 to 100. Figure 2 is a plot of the range (upper limit/lower limit) for various total counts. This plot shows very dramatically how large the range can be for small total counts. Only when the total fiber count is 20 or greater does the range fall to a factor close to 2. It is also significant to note that the range decreases relatively slowly for total fiber counts in excess of 20. Table 1. Two sigma limits for various fiber counts. Two Sigma Limits Total Count Lower Upper 1 0.00 3.00 2 0.00 4.83 3 • 0.00 6.46 4 0.00 8.00 5 0.53 9.47 10 3.68 16.32 20 11.06 28.94 30 19.05 40.95 40 27.35 52.65 50 35.86 64.14 60 44.51 75.49 70 53.27 86.73 80 62.11 97.89 90 71.03 108.97 100 80.00 120.00 284

cS 20.0 ,, 18.0 j 14.0 J. 12.0 10.0 8.0 1 0.0 20.0 40.0 60.0 TOTAL FIBER COUNT Figure 2. Range of 2 sigma limits. 285 80.0 100.0

The final, and most important point to be made in regard to this theoretical discus- sion is that the Poisson distribution can only be considered to be a limiting case. It represents the best that can be achieved under ideal circumstances. If the fibers are not deposited in a truly random manner, the variations will be larger than predicted. As a matter of fact, all available experimental data indicates that real world samples do not follow the Poisson distribution [11]. Although there is much more data available for optical counting, there is no reason to believe that electron microscope samples should be any better. Causes for Non-Random Distribution - Experimental Results The obvious causes for non-random distribution of fibers on a filter surface are inadequate mixing, eddy currents in the filter, and fiber clustering. With water samples, the first two of these can probably be controlled by good experimental technique. In the case of airborne samples, the operator will have little or no influence over the initial distribution and only some control over air currents which may influence the deposition. Recently, an experiment was designed to test the validity of the Poisson distribution under reasonably ideal conditions. We had available a small amount of very well charac- terized glass fiber, 1.5 micrometers in diameter and 30 micrometers long. A carefully weighed quantity, calculated to contain one million fibers, was dispersed in one liter of water. One hundred (100) mL of this dispersion was filtered on a 25 mm membrane filter. The filter was then clarified and examined by phase contrast microscopy. Figure 3 shows a typical area near the center of the filter. The distribution appears reasonably random, but there also appears to be too many fibers lying closely parallel to each other to say that the distribution is completely random. Figure 4 shows the configuration near the edge of the filter. The lower right hand corner is the region closest to the edge of the filter. Here the fibers show a tendency to align circumferentially. Next, there is a complete ring in which very few fibers are deposited. In the next few hundred micrometers, the fibers tend to be radially oriented. As we proceed toward the center of the filter, the distribution becomes more random, as was shown in the first photo in this series. Obviously, there are eddy currents near the side of the filter funnel which have strong influence on the fiber distribution. Continuing the experiment as originally designed, 1000-80 micrometer square fields were counted. The expected number of fibers per field was 2.58. The average found was 3.18. This calculates back to 1.28 million fibers per liter. An excellent correlation, considering all the possible sources of error, including the original characterization of the fibers. Figure 5 shows the actual distribution of the number of fibers per field versus the theoretical Poisson distribution for a mean of 3.18. Even in this well-controlled experi- ment, the distribution is significantly broader than predicted. 286

20b3105051 I

Figure 4. Glass fiber dispersion. Area near edge of filter. Nominal dimensions of fibers are 1.5 x 30 micrometers. Phase contrast. 288

x 0.200 0.190 0.180 0.170 0.160 0.150 0.140 0.130 0.120 0.110 0.100 0.090 0.080 0.070 0.060 0.050 0.040 0.030 0.020 0.010 0.000 0.0 COUNT THEORETICAL Figure S. Actual versus theoretical fiber distribution. 289 ® ACTUAL

Table 2 shows the results of actual electron microscope counts from some typical water and air samples. The fourth water sample and the fourth air sample are of particular interest. In the water sample, 8 grid squares were counted with a mean value of 12.13. The probability of finding a grid square with only 2 fibers is calculated to be about 4 in 10,000. Likewise, in the water sample 20 grid squares were counted with a mean value of 2.9, the probability of finding 11 fibers in one grid square is 2 in 10,000. These are both good examples of serious deviations from the theoretical Poisson distribution which will lead to greater than expected uncertainties. Table 2. Typical counting results. Grid Opening Water Samples Air Samples 1 0 2 4 15 5 0 8 1 2 0 0 1 15 6 0 3 11 3 0 2 2 10 7 0 12 2 4 0 7 2 16 3 0 18 6 5 0 3 0 13 0 0 3 6 6 1 4 1 11 4 0 4 3 7 0 0 1 15 1 0 7 1 8 0 1 1 2 2 1 8 1 9 0 1 3 4 1 8 3 10 0 0 1 4 0 3 3 11 0 5 0 0 2 12 0 1 0 1 0 13 0 4 0 0 3 14 0 5 0 1 2 15 0 3 0 1 0 16 0 3 4 2 0 17 0 5 1 0 3 18 0 1 2 0 1 19 0 2 0 0 3 20 0 7 1 1 6 Figure 6 is a typical clump of fibers and other material found in a water sample. One can only speculate on whether such an agglomerate actually existed in the original sample or is an artifact caused by sample preparation. In any event, its occurrence can have serious consequences on the final results. 290

CO 3 Figure 6. Fiber clump found in water sample. Transmission electron micrograph. 91 N O W H 0 ~ 0 ~ tA

"' Sample Preparation Errors After a sample has been collected on a filter surface, additional processing is necessary prior to examination in the electron microscope. A variety of methods can be used and each can be the source of significant errors. Perhaps the most serious of all is the loss of a significant number of fibers during the clearing or dissolution of the filter. The "cold finger" apparatus is commonly used to clear cellulose ester (Millipore) membranes, and the Jaffe wick method is used for clearing Nuclepore membranes. Both depend on dissolving the polymer in solvent vapors with the subsequent deposition of the entrapped particles on the carbon substrate. Some particles will always be washed away as the polymer is removed. How many and how consistently are very difficult to quantify. Beaman et al. [8], estimate that the losses can be as high as 50 percent for amphibole fibers. Extreme care must be exercised to avoid flooding when using the Jaffe wick method and to control the rate of boiling when clearing by the "cold finger" method. In many cases, a sample might be contaminated with excessive organic material which interferes with the examination of the sample. Removal of the organic material can be accomplished by low temperature ashing followed by redispersion and deposition on a second membrane filter. Although this may be a necessary step, it can lead to serious clumping of fibers. Furthermore, the redispersion can alter the size distribution of the fibers. Chrysotile asbestos, for example, is extremely sensitive to dispersing agents such as Aerosol OT. Another technique that is sometimes used in conjunction with low temperature ashing is the so-called rub-out method. This is useful for reducing the size of large extraneous particles, but does result in a radical change in the fiber dimensions. This method should not be used if the analyst is required to report fiber counts and fiber dimensions. It can only be used to estimate the total mass of fiber present. In general, sample preparation errors lead to an understatement of the number of fibers present in a sample and can distort the size distribution. Some analysts multiply the counts by a factor which was established on the basis of a few controlled experiments. This•practice could only be considered valid if the factor was determined for conditions identical to the reported analysis. This would require the analysis of a standard sample along with each group of unknown samples. Fiber Identification Errors The identification, or mis-identification, of the fiber spgcies present can lead to either positive or negative errors in total fiber counts. With extremely fine fibers positive identification using electron beam techniques is very difficult. Diffraction patterns may have only a few discernible spots and can also be quite fugative. Elemental analyses by x-ray emission can also be erroneous due to the influence of nearby particles. Fiber identification errors can be minimized by adequate operator training. Cer- tainly, critical samples should be analyzed only by experienced operators. Laboratory Contamination Because of the extremely low levels of fibers encountered in environmental samples and the very small sample size, contamination of the specimens can be a serious source of error. Most laboratories concerned with fiber analysis have handled bulk fibers for many reasons. Fibers can also be present in the other media used to process the samples. Good housekeeping practices can keep laboratory contamination to a minimum. It is advisable to handle all samples in an isolated area. A clean air hood equipped with HEPA Filters is most desirable. Obviously, no bulk fibers should be handled in this area. Finally, all solvents should be filtered immediately prior to use. Never rely on the fact that distilled water or other solvents, regardless of their purity, will be fiber free. Finally, it is advisable to run a blank sample through all of the steps of the procedure, along with each group of samples being analyzed. 292

Work to be Done It is obvious from the foregoing discussion that the analysis of environmental samples for asbestos fiber is far from precise. Large errors can be the result of normal random variations and also the manipulations required for sample preparation. It is further obvious that additional work should be done to establish techniques which will minimize the controllable errors. First, and foremost, among the tasks to be accomplished is to establish an acceptable standard procedure for fiber analysis. Work of this type is currently underway in several laboratories. This should be pursued with vigor so that methodology can be specified as soon as possible. Second, and concurrent with the methodology development, should be a systematic study of filter clearing techniques. The objectives of this task would be to better describe the losses which can occur, and to seek imporvements which might give smaller and more consistent losses. Finally, serious consideration should be given to the preparation of a standard dispersion which could be used for comparative studies between laboratories. Such a standard dispersion would also be useful to assist in the quantification of the errors introduced by the various analytical steps. Reporting Results Because of the variety of procedures currently employed and the magnitude of the errors, it is important that as much information as possible be included with fiber analysis reports. This information should include: Sampling conditions Volume filtered Sample preparation method Number of fibers and fields counted Blank counts Identification problems Fiber dimensions This information is absolutely essential. Too many reports are published which show only the number of fibers found in an environmental sample without any background information. Without this information, it is impossible to evaluate the true significance of any and all fiber analyses. References [i] Beckett, S. T. and Attfield, M. D., Inter-Laboratory Comparison of Asbestos Fibers Samples on Membrane Filters, Ann. Occup. HIg. 17, (1974). [2] Curtis, P. A. and Bierbaum, P. J., Technological Feasibility of the 2 Fibers/cc Asbestos Standard in Asbestos Textile Facilities, Amer. Ind. Hyg. Assn. J., 115-125 (February 1975). [3] Rajhans, G. S. and Bragg, G. M., A Statistical Analysis of Asbestos Fiber Counting in the Laboratory and Industrial Environment, Amer. Ind. 1!yg. Assn. J., 909-915 (December 1975). [4] Walton, W. H., Attfield, M. D., and Beckett, S. T., An International Comparison of Counts of Airborne Asbestos Fibers Sampled on Membrane Filters, Ann. Occup. ~Yg. 19, 215-224 (1976). 293 2063105087

[5] Cunningham, H. M. -and Pontrefract, R., Asbestos Fibers in Beverages and Drinking Water, Nature London , 232, 332-333 (1971). [6] Durham, R. W., and Pang, T., Asbestos Fibers in Lake Superior, Water ualit Param- meters, ASTM STP573, American Society for Testing and Materials, pp 5-13 (19 [7] Rohl, A. N., Langer, A. M., and Selikoff, I. J. , Environmental Asbestos Pollution Related to Use of Quarried Serpentine Rock, Science 196, 1319 (June 1977). [8] Beaman, 0. R. and File, 0. M., Quantitative Determination of Asbestos Fiber Concen- trations, Anal. Chem. 48, 101-110 (1976). [9] Anderson, C. H. and MacArthur Long, J. Preliminary Interim Procedure for Fibrous Asbestos, Analytical Chemistry Branch, USEPA, Athens, Georgia (July 31, 1976). [10] Miller, I. and Freund, J. E., Probability and Statistics for Engineers, Second Edition, pp. 77-82 (Prentiss-Hall, New Jersey, 1977 . [11] Brown, A. L. Jr., Taylor, W. F., and Carter, R. E. , The Reliability of Measures of Amphibole Fiber Concentration in Water, Environmental Research 12, 150-160 (1976). Discussion D. SARVADI: Are you familiar with the NIOSH proficiency analytical testing program, and do you have any feel for the inter- and intra-laboratory work they are doing on asbestos counts? J. LEINEWEBER: They have done a fairly credible job on making inter- and intra- laboratory comparisons on standard samples, and even within one laboratory in attempting to compare the results of a group of operators. They have come a lot farther with optical counting than we have with EM counting. There are still problems, but I think they have their situation under a little better control than we do. 294

National Bureau of Standards Special Publication 506. Proceedings of the Workshop on Asbestos: Definitions and Measurement Methods held at NBS, Gaithersburg, MD, July 18-20, 1977. (Issued November 1978) SELECTION AND CHARACTERIZATION OF FIBROUS AND NONFIBROUS AMPHIBDLES FOR ANALYTICAL METHODS DEVELOPMENT J. C. Haartz and B. A. Lange U.S. Department of Health, Education, and Welfare Public Health Service, Center for Disease Control National Institute for Occupational Safety and Health 4676 Columbia Parkway, Cincinnati, Ohio 45226 and R. G. Draftz and R. F. Scholl IIT Research Institute (IITRI) 10 West 35th Street Chicago, Illinois 60616 Abstract More than 50 mineral specimens of fibrous and prismatic (nonfibrous) amphibole species, including tremolite, grunerite, and cummingtonite, were collected and characterized to determine their suitability for use as reference materials in the development of analytical methods. These methods will be used for the detection and measurement of hazardous materials which are found as workplace contaminants. The specimens have been characterized using light microscopy, x-ray diffraction (XRD), and differential thermal analysis (DTA). Some of these specimens have been purified by appropriate physical or chemical techniques and then ground to provide a material with a mass median particle size of less than 10 Nm (major) diameter. The results of characterization studies of the minerals, including a comparison of the properties determined for each of the specimens, are presented. Differences in physical properties of the fibrous and prismatic tremolite specimens are indicated by the data obtained from OTA and XRD studies. While the prepared quantity of each mineral is quite limited, the source of each of the specimen materials and the appropriate methods of sample preparation have been carefully documented should additional quantities be desired. Key Words: Amphibole asbestos; cummingtonite; grunerite; thermal analysis; tremolite; x-ray diffraction. Introduction= Under the provisions of the Federal Occupational Safety and Health Act of 1970 (PL 91-596), the National Institute for Occupational Safety and Health (NIOSH) is charged with the responsibility for research related to occupational health, including the develop- ment and evaluation of analytical methods for the determination of hazardous workplace contaminants. To meet this charge, the Measurements Research Branch of NIOSH has a program concerned with the development of new analytical methods as well as with the iMention of product or trace names does not constitute endorsement by the Public Health Service. 295 2063105089

evaluation and improvement of existing methods. Many mineral dusts, such as those of the silica polymorphs, talc, and asbestos minerals, are included in the hazardous materials for which analytical methods are needed. Earlier work in the NIOSH laboratory showed that it was feasible to quantitatively determine by x-raydiffraction techniques (XRD) chryso- tile, amosite, and crocidolite using either samples of the bulk material or of airborne dust collected on filters [1]2. However, further work rather graphically demonstrated the fact that specimens of a mineral originating from different deposits often exhibit signif- icant variations in impurity content and crystallinity [2], and consequently also exhibit vast differences in their response to analytical measurement techniques. It was obvious that reference materials were needed for the development of analytical methods, that these materials should be from natural sources, and that they be selected on the basis of purity, especially as to an absence of other similar minerals. Pure minerals could then be mixed with other materials to simulate the mixtures found in samples collected from occupational environments. For asbestos, the International Union Against Cancer (UICC) Standard Reference Samples [3] are available as reference materials for chrysotile, amosite, anthophyllite, and crocidolite. These samples have been well characterized with respect to overall chemical composition (elemental weight o/o) and fiber length distribution [4]. There are also some data relating to sample response to heat treatment, and the electron and x-ray diffraction properties [4,5]. However, since these materials were collected and prepared to provide reference samples for inhalation and injection experiments, they were chosen not for phase purity but to be representative of the various types of asbestos used by industry. Further, the UICC samples do not include specimens of the prismatic (nonfibrous) forms of the minerals. . Other reference materials were also needed by NIOSH for the methods development and evaluation program. Consequently, an effort to collect and characterize at least four representative specimens of each of eighteen minerals from different geographical locations was initiated. Table 1 lists the minerals sought and the techniques used for preliminary characterization of the samples. Following the preliminary evaluation and characterization of these samples, the "best" source specimens were chosen for beneficiation, grinding to a respirable size range, and for further characterization and analysis for impurities. A one kilogram quantity of the ground material was established as the final, processed amount to be prepared of each mineral. It was expected that this amount would suffice as reference material for NIOSH analytical research; the source of selected specimens and the appropriate methods for sample preparation were carefully documented should additional quantities be desired. The following discussion will cover the selection, preliminary separation techniques, beneficiation, grinding, and characterization of some of the amphibole species. Details concerning the other minerals will be published separately. Selection of Minerals More than 80 sources were contacted to obtain the approximately 50 samples of mineral specimens containing amphiboles which were received and inspected. Of these samples, 12 were discarded based on macroscopic examination; 38 were carried through the preliminary characterization steps prior to the final selection of the eleven "best" amphibole samples. Since the final quantity of each mineral needed was large (one kilogram), speci- mens were chosen based on (1) the least contamination by other minerals and the contrast- ing habit, and, (2) the amenability of the specimen to beneficiation for removal of contaminant phases. ZFigures in brackets indicate the literature references at the end of this paper. 296

C3 Table 1. Reference materials sought. Mineral Characterization Techniques Silica -Quartz -Cristobalite -Tridymite X-ray Diffraction Beryl Infra-red Spectroscopy Bunsenite (Ni0) Thermal Analysis Fluorite (TG and DTA) Talc Fibrous Serpentine -Chrysotiie Platy Serpentine -Antigorite Fibrous Amphiboles -Crocidolite Macroscopic Habit -Grunerite ("Amosite") Light Microscopy -Anthophyllite X-ray Diffraction -Tremolite Thermal Analysis Prismatic Amphiboles -Riebeckite -Grunerite -Cunmingtonite -Anthophyllite -Tremalite N O 297 °i w .+ 0 (A 0 e .,

'3` After a macroscopic inspection of the specimens as received, using a hand magnifier, portions were hand ground in an agate or diamonite mortar and pestle. The ground samples were dry sieved to pass, a_325 mesh screen and were further characterized using polarized light microscopy, qualitative x-ray diffraction (XRD), and qualitative differential thermal analysis (OTA). The types and quantities of impurities were noted for each of the specimens, and careful scrutiny was given to the mineral morphology, especially for the samples needed for the fibrous and prismatic (or nonfibrous) habits. For macroscopic specimens, the mineralogical criteria distinguishing the fibrous from the prismatic habit are unequivocal. This is illustrated by the samples of tremolite which are shown in figures 1 through 4. The origin of the fibrous tremolite shown in figure 1 is Alaska, while that of figure 2 is a small sample from Italy which was collected in approximately 1890 and has since been in the collection of the Field Museum of Natural History in Chicago, IL. It was not possible to locate a contemporary source of fibrous tremolite in Italy. The prismatic tremolite in figure 3 is from South Dakota and is a fairly pure sample with an acicular radiated structure which is quite evident in the hand specimens. The sample shown in figure 4 contains interlaced prismatic tremolite, talc and other impurities. Although the individual tremolite "needles" are colorless, the sample has a lavender color which may be due to manganese substitutions [G]. Figure 1. Fibrous tremolite: Alaska, 1X. Figure 2. Fibrous tremolite: Tuscany, Italy, 1X. 298

Figure 3. Prismatic tremolite with calcite: South Dakota, 0.57X. Figure 4. Prismatic tremolite with talc and other impurities, 0.5X. Distinguishing between the fibrous and prismatic habits is less straightforward with microscopic specimens. The photomicrographs of tremolite (figures 5 and 6) illustrate the appearance of fibrous and prismatic tremolite specimens ground to a mean particle size of 3.1 pm and 1.7 pm respectively. Similarities in particle shape are evident, although the mean aspect ratio of the fibrous tremolite particles is greater than that of the cleavage fragments of the prismatic material. 299 2063105093

Figure 5. Fibrous tremolite: Rajasthan State, India, 407X. Figure 6. Prismatic tremolite: Gouverneur, New York, 407X. Table 2 lists the amphiboles, and their sources, which were chosen for any necessary beneficiation and final grinding. The impurities listed are those contaminants determined by microscopic analysis of the hand-separated portions of the desired phase. Some of the amphiboles, including the samples of prismatic and fibrous tremolite as well as crocidolite, were obtained as nearly pure, single phase specimens. Others, such as the prismatic grunerite, anthophyllite, and cummingtonite were intermixed with accessory minerals. Hand specimens of the amphiboles selected for preparation as reference materials are illustrated in figures 7-14. 8 00 a w r+ 0 ~ 0 e A

CIS Figure 7. Fibrous tremolite: Rajasthan State, India, IX. Figure 8. Prismatic cumningtonite with associated minerals: Homestake Mine, Lead, South Dakota, O.8X. 302

a. I Figure 9. Fibraus grunerite ("Amosite"): Lydenburg District, Transvaal, South Africa, 0.8X. Figure 10. Prismatic grunerite with quartz: Luce No. 1 Mine, Newfoundland, O.SX. 303

COS Figure 11. Fibrous anthophyllite: Bozeman, Montana, IX. Figure 12. Prismatic anthophyllite with quartz: Bamble. Norway, 0.8X. 04 N O W ~. 0 w 0 b OC

Figure 13. Crocidolite (fibrous riebeckite): South Africa, 0.57X. Figure 14. Prismatic riebeckite (black) with quartz and feldspar: St. Peter's Dome, El Paso County, Colorado, 1X. 305

Separation and Grinding Techniques For those samples which required beneficiation to produce the pure minerals, separation techniques were chosen ,which would adequately liberate the desired phases and least adversely affect their purity. In order to conserve the selected minerals, techniques were chosen which could be applied to material varying widely in size. The preliminary size reduction necessary for beneficiation and grinding of the fibrous amphiboles was accomplished using a rock saw with diamond-impregnated blades. For the nonfibrous amphiboles, a large mortar and pestle were fabricated from strongly magnetic stainless steels so that metals abraded from the equipment during crushing could be removed from the ground material using a magnet. All beneficiation steps were done before the final grinding to allow efficient use of the mineral extraction methods, which are severely limited if the particle size is too smali. To avoid chemical alteration of the desired phases, beneficiation was generally limited to physical methods [7]. The final grinding was designed to produce nonfibrous materials which had a mass median aerodynamic diameter between 0.5 and 5.0 pm, and a maximum size of 10 pm. For the fibrous materials, the desired median length was the range 2-10 pm, with a maximum length of 200 pm. Beneficiation Methods Simple, primarily physical methods of mineral extraction were employed. Three types of hand separation were used: (1) With a mason's hammer and chisels, the available specimen material was "high-graded" to obtain pieces with the greatest concentration of the desired phase; from these, the larger masses of impurities were cobbed. (2) The rock saw was used to cut cross-fiber vein materials into slabs one centimeter thick measured along the fiber length. The slabs were then chipped into small pencils of fibers for further beneficiation and/or preparation for milling. The saw was also used to cut wall rock from the margins of cross-fiber vein specimens of fibrous grunerite, anthophyllite, and crocidolite. (3) Hand picking, or for ferromagnetic minerals a powerful hand magnet, was used to remove small quantities of obvious contaminants at any stage in the size reduction procedure. Only two beneficiation techniques were used in which mineral specimens were exposed to the risk of chemical alteration. Slow dissolution of carbonate minerals from specimens of tremolite and actinolite was accomplished by digestion in dilute (ti3 N) acetic acid. Bromoform and tetrabromoethane were used for density separations of quartz, micas, and other silicates from tremolite, cumsingtonite and grunerite. After separation, the samples were rinsed repeatedly, with acetone or ethanol and then distilled water, to remove residues of the organic liquids. Grinding Techniques Research has shown that some grinding mechanisms degrade the crystalline structure of minerals, particularly asbestiform species, to a considerable degree. Shearing and cutting (in the sense of pinching) actions are reported to be very destructive to crystallinity [8]. Initial attempts in this program to grind asbestos in ball mills equipped with lifter bars confirmed this observation. Impact between air-suspended particles and/or impact of elongate fragments on cutting edges accomplished size reduction with much less reduction in crystallinity, as shown by x-ray diffraction studies. Therefore, grinding tests were made to identify milling devices which exploit the free impact principle and which could efficiently produce large quantities of respirable size particles. For size reductions of fibrous amphiboles, a fiber mill (Retsch Ultracentrifugal-mill, Type ZM-1) was chosen. In this device a rotor with vertical pins at the periphery spins at 10,000 or 20,000 rpm impelling fibers outward against the perforated wall of the grinding chamber (sieve ring) on which cutting edges are angled toward the oncoming particles. The non-fibrous amphiboles were ground using a jet mill (Micron-master Jet Pulverizer) in which tangentially inward-directed jets of dry, filtered air (50 scfm at 90 psig) circulate the feed material in an annular grinding chamber. Size reduction is accomplished by impact between particles; the air stream minimizes particle contact with the walls of the grinding chamber. Additional advantages of the fiber mill and jet mill for this work are: (1) the carrying air stream controls heat build-up in the equipment 306

C2 thereby reducing the risk of thermal degradation of the material being milled. (2) Virtually all particles are subjected to size reduction with each pass of material through the mill. (3) Each mill is provided with a cyclone collector, thus providing coarse and fine fractions. (4) The continuous processes permit efficient size reduction of kilogram quantities of fibrous and nonfibrous amphiboles to the specified size by iterative milling without additional size classification steps. Table 3 presents particle size distributions for fibrous and prismatic tremolite reduced to final size by the respective milling devices. Table 3. Particle sizesa of "reference" tremolite samples after grinding. Fibrous (India) Prismatic (New York) Size Range (um) Number Percentage Size Range (lim) _ Number Percentage <2 37.4 <1 28.0 2-6 35.8 1-3 47.0 6-10 15.4 3-5 18.3 10-20 6.9 5-7 5.9 20-80 3.6 7-10 0.8 80-160 0.9 >10 0.0 >160 0 3.1 pm Geometric Mean 1.7 pm Geometric Mean a Particle sizes determined using optical microscopy. For fibrous tremolite, fiber length is reported; for prismatic tremolite, Feret's diameter. Analytical Studies Analytical studies have been initiated using two of the "reference" materials from this program, the fibrous and prismatic tremolite samples. In addition to these "reference" samples, which were processed by IITRI, and which were carefully characterized as to identity, source, and particle size, a number of samples from the NIOSH mineral collection were used. These samples were included in the analyses to allow comparisons of tremolite specimens from various sources and geographical locations to determine if general characteristics of tremolite specimens could be delineated by obtaining additional experimental data. The NI05H specimens were ground in a SPEX freezer mill at liquid nitrogen temperatures, sieved through a 10 pm sieve, and sized using electron microscopy techniques. The ground material had a mean particle length or diameter of <3.0 pm. The following sections summarize the preliminary results obtained in the studies of tremolite. Chemical Analyses The relative iron, magnesium, and calcium content of several of the specimens used in these studies was determined in order to confirm the designation of these amphiboles as tremolite. To minimize contamination which could occur from contact with metallic surfaces during grinding, pieces of the hand specimens instead of ground material were used for the analyses. These pieces were dissolved by heating in a mixture of HF and concentrated HC1. Blind replicate analyses were done for each of the specimens using 307 2063105101

atomic absorption spectrophotometry. The results (table 4) for the ratio (Fe + Mg : Ca) and the calculated weight percent Fe0 indicate that all of, the samples fall within the empirical composition limits for tremolite [9], including a specimen previously identified as prismatic actinolite. In general, the specimens of fibrous tremolite contain more iron than the prismatic form although the South Korean sample of fibrous tremolite was an exception. Table 4. Chemical analyses. Atom Ratio Atom Ratio Amphibole Fe : Mg : Ca Fe + Mg : Caa Wt. % FeO Prismatic Tremolite Gouverneur, N.Y.b 1: 205 : 78 5.3 : 2.0 0.21 South Dakota 1: 44 : 21 4.3 : 2.0 1.03 Fibrous Tremolite Rajasthan, Indiab 1: 13 : 6 4.7 : 2.0 2.87 Alaska 1: 13 : 6 4.7 : 2.0 3.03 Korea 1: 33 : 16 4.3 : 2.0 0.69 Italy I : 31 : 14 4.6 : 2.0 1.43 Prismatic Actinolitec South Dakota 1: 15 : 7 4.6 : 2.0 2.63 a Theoretical limit of ratio = 5:2. b"Reference" material, supplied by IITRI. c Classification based on color and location of source. X-Ray Diffraction Studies For the x-ray powder diffraction studies of the "reference" tremolites, both bulk powder samples (packed in cups) and thin layers on silver membrane filters were used. For the filter studies, homogeneous suspensions of known tremolite concentration in isopropanol were prepared using ultrasonic agitation to ensure dispersion. Aliquots of this suspension were filtered through 25 m., 0.45 Nm pore size silver membrane filters. The calculated weight of treeolite deposited was confirmed by weighing, using a m'cro- balancp. For both fibrous and prismatic tremolite the 310 and 110 peaks (3.14 A and 8.38 A, CuKa radiation) were step scanned to determine the integrated peak intensities. The calibration curves (figure 15) were prepared by plotting the net normalized integrated intensities of these peaks versus the amount of tremolite on the filters. The data clearly indicate that quantitation of pure samples as small as 20 Ng is feasible. However, the ratios of the reflections, I131o1•I(31o) , age dif~erent for filter deposits of fibrous and prismatic habits. The peak r3tio (8.38 A:3.4 ) for prismatic tremolite is approximately 1.0 while that for the fibrous tremolite is approximately 0.40. Packed bulk samples of both tresalite habits give the same peak ratio, the value of which is 0.20. Information in the Powder Diffraction File [10] indicates a peak ratio of 1.0 for tremolite from St. Gotthard, Switzerland. The morphology is described as "white 308

.40 40 30 10 0 0 2 0 40 60 80 100 120 140 MICROGRAMS OF TREMOLITE Figure 15. Calibration curves for fibrous and prismatic tremolite: n a Fibrous tremolite: 03.14 A; O 8.38 A Prismatic tremolite: •3.14 A;  8.38 0 160 C4 2 180 309

radiating fine fibrous~masses," but the term "radiating" suggests it may be a prismatic form. Data obtained for specimens of.tremolite from other geographical locations indicate that, for material deposited on filters, the samples of prismatic tremolite in general show a larger ratio for these peaks than do samples of fibrous tremolite (table 5). Table 5. Ratio of XRD peaks observed for fibrous and prismatic tremolite.a Amphibole No. of Replicates Ratio (8.38A:3.14A) Prismatic Tremolite Gouverneur, N.Y.b 7 1.04 South Dakota 5 1.08 Newburyport, Mass. 10 1.45 Fibrous Tremolite Rajasthan, Indiab 7 0.39 Alaska 10 0.55 Korea 5 0.35 Italy 5 1.25 a 150 pg on 0.45 pm pore size silver filters. b"Reference" material, supplied by IITRI. At this point no explanation can be advanced to account for the differences in peak ratios, although the effects observed may be due at least in part to preferred orientation of the particles in some or all of the samples. Regardless of the reason, the effect is seen for a variety of samples and for a wide range of filter loadings as demonstrated by the calibration curves. The distinctions observed using this technique may prove useful in analytical attempts to ascertain the type of material to which a worker is being exposed. Thermal Analysis Studies Preliminary differential thermal analysis (DTA) studies on tremolite samples have been completed. These studies included an evaluation of the feasibility of this technique for the quantitative analysis of tremolite and, while good calibration curves were obtained, DTA was not sensitive enough to detect microgram quantities of tremolite. The samples were heated in platinum cups to a temperature of 1150 °C at a heating rate of 10°/min in dry air flowing at 5.7 L/hr; the instrument was calibrated using SrCO3, an NBS- ICTA Standard Reference Material. In parallel with the XRD studies of the "reference" tremolite samples, differences between these samples (table 6) were observed during the thermal studies of fibrous and prismatic tremolite samples. These differences in peak position and the color of the decomposition product were observed for samples from other geographical locations as well as for the "reference samples." Similar differences were observed by IiTRI for those specimens considered for selection as "reference" materials. All samples displayed the strong endotherm which is associated with the loss of structural water and the breakdown of the amphibole structure, which subsequently recrystallizes to a monoclinic pyroxene j11]. However, the data indicate that in general the fibrous tremolite samples dehydrate and recrystallize at a lower temperature than do the prismatic tremolite samples. This 310

behavior is analogous to that noted for serpentine, i.e., chrysotile loses structural water at a lower temperature than does antigorite [12]. Although it is recognized that differences in particle size, grinding techniques and experimental conditions can affect the position of a DTA peak [13], data obtained in both the NIOSH and IITRI laboratories are consistent in showing that the endotherm of fibrous tremolite is lower by approximately 50 °C than'that of the prismatic tremolite. It was also observed that the pyroxenes formed from fibrous tremotite were always brown to tan in color while the pyroxenes formed from the prismatic tremolite were always white in color. However, XRD scans of the pyroxenes were virtually the same regardless of color or origin of the specimen and indicated that the final decomposition material was primarily diopside. Table 6. Thermal analysis of tremolite.a Amphibole No. Samples DTAa Endotherm, °C Color of Pyroxene Fibrous Tremolite NIOSHb 4 1026 ± 27 tan IITRI 1 1002 not determined Prismatic Tremolite NIOSHC 5 1078 ± 20 white IITRI 4 1053 ± 11 white a NIOSH samples included those listed in Table 5 as well as two additional samples from the Gouverneur, N.Y. area; IITRI samples include those screened as potential "reference" materials. b Geometric mean particle length <3.0 pm. c Geometric mean particle maximum dimension <3.0 pm. Summary and Conclusion The analytical studies planned for the reference materials have been initiated using the tremolite specimens. These studies have indicated that x-ray diffraction may turn out to be an even more useful tool than expected. The detection limits obtained and the differences in peak ratios observed for samples of fibrous and prismatic tremolite on silver filters have potential for applications to analyses of hazardous, workplace contaminants. The authors gratefully acknowledge the guidance and assistance received during this program from J. V. Crable of NIOSH and B. G. Woodland of the Field Museum. References [1] Crable, J. V., Am. Ind. !!jq. Assoc. J., 27, 293 (1966). [2] Nenadic, C. M. and Crable, J. C., Am. Ind. Hyg. Assoc. J., 32, 539 (1971). [3] Timbrell, V., Gibson, J. C. and Webster, I., Int. J. Cancer, 3, 406 (1968). 311

[4] Timbrell, V., "P'neumonoconiosis. Proceedings Int. Conf.," Johannesburg, 1969, . 28-36. [5] Skikne, M. I., Talbot, J. H., and Rendall, R. E. G., Env. Res., 4, 141 (1971). [6] Deer, W. A., Howie, R. H., and Zussman, J., "Rock-Forming Minerals, Vol. 2, Chain Silicates," J. Wiley & Sons, N.Y., N.Y. 1963 p. 253. [7] Muller, L. D., "Laboratory Methods of Mineral Separation," CH. 1, in J. Zussman, ed., Physical Methods in Determinative Mineralogy, Academic Press, London & N.Y., 1967. [8] Ocella, E. and Maddalon, G., Med. Lavoro, 54 (#10), 628 (1963). [9] See, for example, Deer, W. A., Howie, R. A., and Zussman, J., "Rock-Forming Minerals, Vol. 2, Chain Silicates," J. Wiley & Sons, N.Y., N.Y. 1963, p. 251. [10] Powder Diffraction File. Joint Committee on Powder Diffraction Standards, 1601 Park Lane, Swarthmore, PA 19081. [11] Deer, W. A., Howie, R. H., and Zussman, J., "Rock-Forming Minerals, Vol. 2, Chain Silicates," J. Wiley & Sons, N.Y., N.Y. 1963, pg. 254. [12] Deer, W. A., Howie, R. H., and Zussman, J., "Rock-Forming Minerals, Vol. 3, Sheet Silicates," J. Wiley & Sons, N.Y., N.Y. 1962, pg. 180. [13] Martinez, E., Am. Miner., 46, 901 (1961). Discussion I. STEWART: Both OTA and x-ray diffraction are very sensitive to packing, and, of course, this can be related to shape. Did you do any tests to determine whether packing or repacking would change the relative ratios of peak heights or peak positions? J. HAARTZ: No, we haven't. STEWART: Or spinning the sample in x-ray diffraction perhaps? HAARTZ: The relative ratios of the peaks in x-ray diffraction were the same for the bulk samples. For the samples that were deposited on a silver filter, that is a very thin layer; we did see the differences in the peak ratios. This was the case not only with samples of different origins, but with a great many replicas of the same material. STEWART: I see. So, it was purely the fact that it was fibrous, you think? I didn't quite catch what you meant by your bulk sample. By bulk, I was equating that with "massive." You mean a bulk fiber sample. HAARTZ: By a bulk sample, I mean a milligram or more, of either the massive or fibrous, showed th® same diffraction pattern: identical. When these samples are deposited as a thin layer on a silver membrane filter and the pattern taken, we do see differences in the peak ratios. 312 Cj

Nationat Bureau of Standards Special Publication 506. Proceedings of the Workshop on Asbestos: Definitions and Measurement Methods held at NBS, Gaithersburg, MD, July 18-20, 1977. (Issued November 1978) ASBESTIFORM MINERALS IN INDUSTRIAL TALCS: COMMERCIAL DEFINITIONS VERSUS INDUSTRIAL HYGIENE REALITY John M. Dement U.S. Department of Health, Education and Welfare Public Health Service, Center for Disease Control National Institute for Occupational Safety and Health Division of Surveillance, Hazard Evaluations and Field Studies Cincinnati, Ohio 45226 Abstract As part of its industry-wide study of the talc industry, the National Institute for Occupational Safety and Health (NIOSH) has conducted Detailed industrial hygiene studies of mine and mill operations processing talcs contaminated with asbestiform minerals. The principal analytical method used for studies of asbestiform minerals in talc bulk samples and airborne dust samples is analytical transmission electron microscopy utilizing selected area electron diffraction and microchemical analysis for fiber identification. This presentation includes a discussion of the methods of analysis being used by NIOSH and comparisons of results of analysis with other analytical techniques. Also included are results of NIOSH industrial hygiene studies in asbestiform talc operations and comparisons of airborne fiber characteristics (fiber length, diameter, aspect ratios, etc.) in these operations with other industrial processes using asbestos fibers. Key Words: Amphiboles; anthophyllite, asbestiform minerals; industrial talc; occupational health; tremolite. Introduction The mineral talc is a pure hydrous magnesium silicate Mgs(Sie022)(OH)4 which has a theoretical chemical composition of 63.5 percent Si02, 21.7 percent MgO, and 4.8 percent Hz0 [1,2]1. However, this ideal chemical structure is rarely found in nature due to ionic substitution in the talc structure and due to common association with other minerals such as tremolite, anthophyllite, calcite, magnesite, quartz, dolomite, diopside, and serpentines (chrysotile, antigorite, and lizardite) [1,2]. Most talcs, as mined, are associated with varying proportions of some of these minerals [1] and sold as industrial talcs. In 1974 over 1.4 million short tons of talc were produced in the United States with major uses being in ceramics, elastomers, foundry facings, insecticides, paints, paper, roofing and toilet preparations [3]. The National Institute for Occupational Safety and Health (HIOSH) in cooperation with the Mining Enforcement and Safety Administration has underway an industry-wide study of the talc mining and milling industry. These studies include both epidemiological studies of exposed worker populations to determine health effects which may be attributed to occupa- tional exposures and detailed industrial hygiene studies to characterize the various agents to which workers have been exposed. 1Figures in brackets indicate the literature references at the end of this paper. N 313 a° w r+ 0 ~ r+ 0 ~

Since many talc deposits contain asbestiform amphiboles and in some cases chrysotile (a serpentine), a large portion of the NIOSH environmental studies is directed toward determining mineral fiber exposure patterns and characteristics. For such studies, the primary method used is analytical transmission electron microscopy. This report includes a description of the equipment and procedures used by NIOSH for its environmental studies of industrial talc exposures and results of industrial hygiene studies in a talc mine and mill producing talcs containing asbestiform amphibole minerals. Also discussed are com- mercially employed definitions of what constitutes asbestos and the relationship of these definitions to observed industrial asbestos exposure characteristics. Analytical Methods Equipment A number of methods are available and have been used to identify and quantitate asbestos concentrations in environmental samples. These methods include x-ray diffraction, differential thermal analysis, phase contr,ast and bright field optical microscopy, petrographic microscopy, scanning electron microscopy, and transmission electron microscopy. Each of these methods have certain advantages and disadvantages [4,5]. However, many researchers today consider analytical electron microscopy to be the method of choice for studies of occupational and environmental asbestos exposures. For NIOSH studies of industrial talc exposures, analytical transmission electron microscopy is employed along with other standard mineralogical techniques such as x-ray diffraction and petrographic microscopy. The analytical system consists of a combination transmission-scanning electron microscope with a side entry stage equipped with an energy dispersive x-ray detector which is fitted through a port in the microscope column parallel to the specimen holder. The specimen-to-detector distance is approximately 10 mm with the specimen tilted 39 degrees to the electron beam for optimum x-ray collection. The energy dispersive x-ray detector has an actual energy resolution of less than 170 electron volts, and spatial resolutions of less than 0.5 micrometers are easily realized. This combination of analytical instrumentation permits visual characterization of particulate morphology such as fiber shape, length, and diameter as well as fiber identification using both selected area electron diffraction and x-ray microchemical analysis. In addition, surface topography may be further studied with this instrument by use of the scanning mode of operation using secondary electron images. Procedures Either bulk quantities of materials of interest, such as talcs, or environmental samples collected on membrane filters are routinely analyzed. The majority of samples studied consists of airborne particulates collected in industrial operations for the purpose of determining occupational exposure patterns. These samples are routinely col- lected on Millipore AA, 37 mm diameter membrane filters at sample rates of 1.5-2.0 liters per minute. Sample durations may vary from 15 minutes in very dusty operations to six hours for operations with little visible dust. The method presentty used by NIOSH for preparation of membrane filter samples for electron microscopic analysis is a modification of a direct clearing method first described by Ortiz and Isom [6]. The NIOSH method has been described in detail elsewhere [4]. Briefly, this method consists of the following steps: 1. A section of the membrane filter is cut with a cork bore (8 mm diameter) or a scalpel. This section is removed and placed sample side up on a clean microscope slide with the edges fastened to the slide with either a gummed binder ring or tape. 2. The slide assembly containing the sample is placed in a glass petri dish on top of four Whatman filters which have been saturated with acetone and covered. The acetone vapors destroy the microporous structure of the filter by slow dissolution, producing a fused, microscopically smooth surface on 314

S5 the sample side of the membrane filter. A 10-minute fusion time has been found to be generally acceptable for Millipore AA filters. 3. After fusion of the filter surface, the slide assembly is placed in a vacuum evaporator on a rotary stage where the sampled side of the filter receives a fairly heavy (~200 A) carbon coat. This carbon coat aids in retaining particles during subsequent filter dissolution and also provides for greater thermal stability during microscopic examination. 4. The final step is dissolution of the membrane filter and deposition of the particles onto electron microscope grids. A modified Jaffe Wick method is used whereby four Whatman filter papers are saturated with acetone. Two- hundred mesh carbon filmed grids are used and the coated filters are placed sample side down on them. The petri dish is then covered. Complete filter dissolution takes 8 to 16 hours. Acetone is replaced as necessary. Using this method, many filters may be prepared as a "batch". Particle losses have been low and estimated at less than 10 percent [6]. Samples prepared by the preceding'method are analyzed using analytical transmission electron microscopy whereby three pieces of data are gathered and used to identify each fiber (3 to 1 aspect ratio particles) observed. These Include: (1) visual identification of single fiber electron diffraction patterns, (2) visual identification of semi quanti tati ve elemental analysis spectra using x-ray microchemical techniques, and (3) observation of morphological characteristics, such as diffraction fringes, which may aid in identification. In addition, fiber length and diameter are also recorded. For most studies an accelerating voltage of 100 kilovolts is used with a screen magnificition of approximately 17,000X. Beam currents are usually fixed at 100 microamps (not to be confused with specimen current). Fiber concentrations are estimated using the average grid opening area as the cali- brated counting area. To optimize statistical accuracy of the analysis while keeping analysis time to acceptable limits, 10 grid openings or 50 fibers are analyzed for each sample with a minimum of 5 grid openings. Analysis times range from 90 minutes to 3 hours per sample. Using this counting criterion for a typical 90 minute sample collected at 2 liters per minute, the lower limit of detection is estimated to be less than 0.1 fibers/cc. Precision and accuracy estimates from studies of the NIOSH phase contrast method [7] are considered generally applicable with a coefficient of variation of approximately ±25 percent for most samples. Environmental Studies of Talcs Containing Asbestiform Minerals Methods As previously mentioned, a large portion of the NIOSH industry-wide study of the talc industry involves industrial hygiene studies of worker exposures, including exposures to asbestiform minerals. One such operation recently studied involved a mine and mill producing industrial talcs certified by the mining concern to be free of asbestos. Apparently, the prime analytical methods relied upon by this company to conclude that its products were asbestos free were gross methods such as observation with a common hand lens or at best low power stereomicroscopy both of which were claimed to be sufficient and proper mineralogical techniques. In order to evaluate these claims, a detailed industrial hygiene study was conducted at the mine and mill in question to evaluate worker exposures using best available sampling and analytical technology. Although a number of different sampling and analysis methods were employed, only results of the fiber samples are presented in this report. In order to evaluate fiber exposures and exposure characteristics, personal, breathing zone samples were collected from workers in the mine and mill using 37 mm diameter, Millipore AA membrane filters operated at a flow rate of 1.7 liters per minute. Sample 315 2063105109

filters were changed periodically throughout the work shift to prevent filter overloading. During the study, more than 220 such samples were collected and used to determine both peak and time-weighted-average exposures. All samples were analyzed for fiber concentrations (>5 pm) using the standard phase contrast method recommended by NIOSH [7]. In addition, approximately 15 percent of these samples were analyzed by the electron microscopic methods previously described. Results Results of the fiber concentrations in the mine and mill as determined by phase contrast optical microscopy are shown in Table 1. Highly elevated fiber concentrations were observed in both mine and mill operations with time-weighted-average exposures ranging from 0.8 to 9.8 fibers >5 pm/cc in the mine and 0.2 to 16.0 fibers >5 pm/cc in the mill. Peak exposures as high as 29.1 fibers >5 pm/cc were observed. Table 1. Summary of fiber exposures in talc mine and mill operations as determined by optical microscopy. Operation - - - - - Fiber Concentrations (fibers >5 um/cc) - - - - - Time-Weighted Averages Highest Peak Mean ± SE Median Range Conc. Observed Mine (N=54) 4.5 ± 0.8 4.4 0.8- 9.8 18.2 Mill (N=168) 5.0 t 0.5 4.3 0.2-16.0 29.1 N= Number of individual samples collected SE = Standard Error Time-Weighted averages represent full shift determinations While the above fiber concentrations, determined by phase contrast microscopy, may include some fiber types other than asbestos (e.g., talc "fibers"), they nevertheless represent minimum estimates of true exposures to asbestiform minerals as most asbestiform fibers are less than 5 pm in length and, in addition, some fibers, although longer than 5 pm, may escape detection due to resolution limits of optical microscopy. These facts are demonstrated in Table 2, which show concentrations of positively identified asbestiform mineral fibers as determined by electron microscopy. Time-weighted-average exposures were found to range from 9.5 to 25.0 fibers/cc in the mine and 7.3 to 102.7 fibers/cc in the mill. The highest concentration observed on a single sample was 102.7 fibers/cc. Table 2. Summary of asbestiform mineral fiber exposures in talc mine and mill operations as determined by electron microscopy. - - Fiber Concentrationsa (fibers (all lengths)/cc) - - Operation Time-Weighted Averages Highest Peak Mean ± SE Median Range Conc. Observed Mine (N-8) 16.4 ± 0.9 15.3 9.5- 25.0 25.0 Mill (N=19) 30.0 ± 1.4 24.1 7.3-102.7 102.7 N= Number of air samples randomly chosen and analyzed by electron microscopy SE = Standard Error a Concentrations reported include only those fibers positively identified as one of the asbestos minerals by analytical electron microscopy. 316 (C

A typical electron photomicrograph of fibers in these operations is shown in figure 1 demonstrating the fibrous morphology of these particulates. The asbestiform habit of many of these fibers is evidenced by the "fiber bundle" effect. Results of the electron dif- fraction and microchemical studies on these fibers clearly demonstrated the presence of two amphibole fiber types; these being tremolite and anthophyllite. Analytical data for typical tremolite and anthophyllite fibers are shown in figures 2 and 3, respectively. The anthophyllite is seen to be low in iron content. Figure 1. Electron photomicrograph of particles in talc certified as asbestos-free. 317

I tiectron photomicrographs Diffraction pattern X-ray spectrum Figure 2. Analytical data for tremolite fibers in talc certified as asbestos-free. 318

Electron photomicrographs Diffraction pattern Figure 3. Analytical data for anthophyllite fibers in talc certified as asbestos-free. Tabulations of results of the fiber identification studies by electron microscopy are shown in Table 3. Of all airborne fibers (3:1 aspect ratio particles), 12-19 percent and 38-45 percent were found to be tremolite and anthophyllite, respectively, while 38-39 percent remained unidentified due to unrecognizable diffraction patterns. Tremolite fibers were observed to be generally shorter in length than anthophyllite fibers as demonstrated in Table 3 when only fibers longer than 5 pm were considered. Only 7 percent of the fibers longer than 5 pm were identified as tremolite whereas 65 percent were anthophyllite. This may also be observed in Table 4 where summary statistics of fiber length are given. While all median fiber lengths were found to be similar and not statistically different, the proportion of anthophyllite fibers longer than 5 Nm in length was significantly (P<0.05) greater than tremolite (8-10% for anthophyllite versus only 3% for tremolite). 319 2063105113

Table 3. Summaryof airborne fiber types in talc mine and mill operations as determined by analytical electron microscopy. Fiber Length Percent of all Airborne Fibersa Not Positively Tremolite Anthophyllite Nonasbestos Identified All Fibers 12-19 38-45 1-2 38-39 Fibers > 5 ps 7 65 3 25 a Total number of fibers analyzed was approximately 1850. Table 4. Summary of airborne fiber lengths for positive amphiboles in talc mine and mill operations as determined by electron microscopy. Operation and Fiber Type Median Length Nm Geometric Standard Deviation %< 5 pm in Length Mine Tremolite (W83) 1.6 1.8 97 Anthophyllite (N=164) 1.5 2.6 90-92 Mill Tremolite (N=160) 1.5 1.9 97 Anthophyllite (N=687) 1.4 2.9 90 N = Number of individual fibers analyzed Inasmuch as the NIO5H recommended phase contrast counting method defines countable fibers only on the basis of fiber length and aspect ratio, much controversy has arisen with various industrial and mining groups claiming that this liberal criterion would define many mineral fragments as being asbestos. In this regard, fiber aspect ratios for positively identified amphibole fibers in the talc mine and mill under study are shown in Table 5 for all fiber lengths, and similar data for fibers longer than 5 pm are given in Table 6. These tables demonstrate that anthophyllite fibers in these talcs have larger aspect ratios than tremolite fibers and by comparison of Tables 5 and 6, aspect ratios increase with fiber length. Of interest is the fact that less than two percent of the positively identified amphiboles longer than 5 ps in length had aspect ratios 5 to 1 or smaller. Table 5. Aspect ratios (length to width) for airborne amphibole fibers (all lengths) in mine and mill operations as determined by electron eicroscopy. Fiber Type Median Ratio %< 5 to 1 % c 10 to 1 Tremolite (N--164) 7.5 23-24 70 Anthophyllite (N=687) 9.5 15-17 70 N = Number of individual fibers identified and sized 320

r , (CS Table 8. Aspect ratios (length to width) for airborne fibers > 5 pm in length in mine and mill operations as determined by electron microscopy. Fiber Type %< 5 to 1 %< 10 to 1 Positively Identified Amphiboles <2 37-38 Non-Asbestos or Unidentified Fibers 18 80 Approximately 1850 fibers analyzed Discussion Results of an industrial hygiene study of talc operations producing industrial talcs certified by the company under study to be asbestos free have been presented. Contrary to claims of this company that its products do not conta n asbestos, this study demonstrated excessive exposures to airborne fibers of which more than 70 percent of the fibers >5 pm in length could be identified as positive asbestiform amphiboles by best available analytical techniques. Repeated requests have been made of this company to clarify analytical methods and definitions of asbestos used to arrive at the conclusion that its products were free of asbestiform minerals. Apparently, the analytical method used was observation of hand ore specimens with a hand lens or, at best, use of low power stereomicroscopy. The definition of "asbestos" employed is less clear. Apparently the definition used is one which might best be termed a"commercial definition"; that is, in order for an amphibole to be consid- ered to be asbestos it must have commercial value due to its fibrous.shape. This same company also operates another nearby talc mine and mill producing talc products which the company acknowledges as containing anthophyllite asbestos and labels these products with the warning required by the Occupational Safety and Health Administra- tion. The determination made by the company that these talcs should be labeled was again based on macroscopic observation of hand specimens. Having observed such elevated exposures as were presented in this report in operations considered by this company to be "asbestos free", it would seem logical to evaluate air- borne fiber characteristics in this other operation acknowledged as containing asbestos. Such a study has been conducted using 10 airborne dust samples collected by the Mining Enforcement and Safety Administration during a 1975 survey. These samples were analyzed by identical electron microscopic methods which have been previously described and results are given in Table 7 along with comparisons with the other mine and mill operations producing products certified to be "asbestos free". Table 7 clearly demonstrates that all airborne fiber characteristics between these two operations are remarkably the same. In fact, the mine and mill producing "certified" talcs were found to have a statistically (P<0.05) significantly higher proportion of positive amphiboles based largely on a higher tremoiite fiber content. Considerations for what constitutes an "asbestos fiber" from an industrial health point of view warrants further discussion. Many researchers continue to promote unusable defini- tions based on the microscopic world whereas microscopic .ineral fibers are of real concern for the health scientist. The data shown in Tables 4 and 7 demonstrate that more than 90 percent of all airborne amphibole fibers in the talc operations studied were shorter than 5 pm in length. Some individuals might argue that these fibers were mineral fragments and not "asbestos", however, it must be pointed out that all industrial operations using or processing asbestos generate airborne fibers similar to those seen in this study. This fact is demonstrated in Table 8 which compares airborne fiber lengths in various operations. 321 20631Q5115

Table 7. Comparison of airborne fiber characteristics between two operations of the same company, one producing asbestos talcs and the other producing talcs certified by the company as asbestos free. Mine and Mill Mine and Mill Airborne Fiber Characteristics Producing Producing Statistical Labeled Talcs Unlabeled Talcs Significance Proportion Positive Amphiboles 0.50 0.58 P<0.05 Proportion Anthophyllite 0.47 0.45 NS Proportion Tremolite 0.03 0.13 P<0.001 Median Fiber Length Anthophyllite 1.61 pm 1.45 pm NS Tremolite ---a 1.55 pm -- Median Fiber Diameter Anthophyllite 0.16 pm 0.13 pm NS Tremolite ---a 0.19 pm -- Median Fiber Aspect Ratio Anthophyllite 9.9 9.5 NS Tremolite ---a 7.5 -- X of Fibers < 5 Nm in Length Anthophyllite 92 90-92 NS Tremolite a --- 97 -- a Insufficient number of fibers observed for calculation of size distribution. NS = Not significantly different at 0.05 level Table 8. Comparison of airborne fiber length distribution in various asbestos operations. Operation Fiber Type Median Length %< 5 um Textilea fiber preparation and carding chrysotile 1.4 4 spinning, twisting, weaving 1.0 2 Frictiona mixing chrysotile 0.9 2 finishing 0.8 2 Asbestos-cement pipea mixing chrysotile 0.9 2 finishing 0.7 1 Study Talc Mine and Mi11 tremolite and 1.4 to 1.6 3-10 anthophyllite a Taken from reference B. N 0 322 w ~ '.. a

Conclusions Based on the preceding discussion, the following conclusions are drawn. 1. Commercial definitions of asbestos, whereby asbestos fibers are defined on a micro- scopic scale, have little or no relevance to actual airborne fiber exposures where fibers of microscopic scale are of concern. Furthermore, those mineralogical or geological methods such as examination of ore specimens with a hand lens or low power microscopy are of limited value for routine identification of asbestiform mineral contamination in minerals or mineral products. 2. Users of products containing asbestos have a right to know that they have potential for exposures to asbestos or asbestiform minerals such that proper precautions may be taken to eliminate or reduce exposures. Producers of these products have an obligation to provide these data based on appropriate analytical techniques. Regulatory agencies must insist that appropriate techniques be employed and monitor results. 3. Inasmuch as considerable quantities of data are available suggesting that many fibrous materials may be biologically active [8], consideration should be given for establishing exposure standards for "mineral fibers" as a class of materials with similar health effects. The lives and health of American workers, America's most valuable resource, should not be compromised while the health scientist and the mineralogist disagree over definitions. As Dr. Paul Kotin of the Johns-Manville Corporation stated so well at this conference, the body has not read the asbestos regulations to decide which fibers should cause a biological response. Similarly, neither has the body read a mineralogy text to determine which particles of fibrous minerals should be considered "asbestos" or only mineral fragments. , References [1] Rohl, A. N. and Langer, A. M., Identification and qualitative of asbestos in talc, Env. Health Persp. 9, 95-109 (1974). [2] Stanley, H. D. and Norwood, R. E., The detection and identification of asbestos and asbestiform minerals in talc in Proceedings of the Symposium on Talc, U.S. Bureau of Mines, Washington, 0. C., May 8, 1973. [3] Minerals in the U.S. Economy: Ten Year Supply - Demand Profiles for Mineral and Fuel Commodities (1965-74), United States Department of Interior Bureau of Mines (1975). [4] Zumwalde, R. 0. and Cement, J. M., Review and Evaluation of Analytical methods for Environmental Studies of Fibrous Particulate Exposures, CHEW (NIOSH) Publication No. 77-204, May (1977). [5] Keenan, R. G. and Lynch, J. R. , Techniques for the detection, identification and analysis of fibers, Amer. Ind. Hyq. J., 31, 587-597 (1970). [6] Ortiz, L. W. and Isom, B. L. , Transfer technique for electron microscopy of membrane filter samples, Amer. Ind. Lg. Assoc. J., 423-425 (1974). [7] Leidel, A. L., Bayer, S. G., and Zumwalde, R. D., USPHS/NIOSH Membrane Filter Method for Evaluating Airborne Asbestos Fibers, NIOSH, November (1975). [8] Oement, J. M., Zumwalde, R. D., and Wallingford, K. M., Asbestos fiber exposures in a hard rock gold mine, Ann. N.Y. Acad. of Science, 271, 345-352 (1975). Discussion NOTE: Discussion of this paper was included in the General Discussion at the end of this session. 323 2063105117

National Bureau of Standards Special Publication 506. Proceedings of the Workshop on Asbestos: Definitions and Measurement Methods held at NBS, Gaithersburg, MD, July 18-20, 1977. (Issued November 1978) THE DETECTION AND IDENTIFICATION OF ASBESTOS AND ASBESTIFORM MINERALS IN TALC Harold 0. Stanley Pfizer Inc., MPM Divisionl Easton, Pennsylvania 18042 Abstract Concern with the health hazards associated with the presence of chrysotile asbestos and/or the asbestiform minerals in talc has prompted widespread investigation of methods of analysis which would be consistent with good analytical practices. Of all the currently available techniques examined and evaluated, the two most reliable have been found by us to be Step Scanning X-ray Diffraction and Transmission Electron Microscopy (TEM), with Selected Area Electron Diffraction (SAED). The Step Scanning X-ray Diffraction technique allows quantitative detection and identification of tremolite and the asbestiform minerals down to 0.1 percent by weight_ In the absence of chlorite it can detect and quantitatively determine chrysotile asbestos at the 0.5 percent level. Chlorite, however, is often associated with talc ore bodies. When present, chlorite will mask most of the main x-ray diffraction peaks of chrysotile. Additionally, the x-ray diffraction technique cannot distinguish between fibrous and non-fibrous forms of the asbestiform minerals_ TEM is ideally suited to determinations of this type because of its high resolution and magnification capabilities, the morphological nature of the problem, and the mineralogical identification capability through SAED. Key Words: Asbestiform; asbestos; chrysotile; detection; fiber; identi- fication; light microscopy; selected area electron diffraction; talc; transmission electron microscopy; tremolite; x-ray diffraction. Introduction The pneumoconiotic and cancer-inducing health hazards of exposure to the asbestos and asbestiform minerals sometimes found associated with talc have been appropriately identified by recent research, the mass media publications [1-8]2, and the papers heard earlier today. Because of Pfizer's position as a supplier of talc to many industries, we felt that a reliable method of detecting and identifying asbestos and asbestiform minerals possibly present in talc had to be developed. Prior to 1970, we were looking for ,fust such a method. Previous investigators had addressed themselves to the problem of identifying asbestos in bulk form or in airborne samples. We concerned ourselves with detecting and identifying the various forms of asbestos in the bulk talc matrix. As we were to later discover, this is indeed a hostile environment for the analyst. 'Now with Degussa Corp., Rt. 46 at Hollistor Rd., Teterboro, N.J. 07608. 2Figures in brackets indicate the literature references at the end of this paper. Preceding page blank 325

Our goals were: 1. Identifying the mineralogy of our products, specifically, that of our talcs. 2. The unambiguous determination of the crystal habit and crystal structure of the mineralogical species present. Ideally, we were looking for a technique that would be simple and direct, but above all, it was mandatory that the technique be positive and unambiguous. The mineralogical and chemical nature of talc and that of the amphiboles or asbestiform minerals and chrysotile have been adequately described previously at this session. Currently available methods and methodology for detecting asbestos, tremolite, and the asbestiform minerals in the presence of talc were reviewed. Types of analyses which we tried included the following: 1. Infrared spectroscopy 2. Thermal analysis including TGA and DTA 3. X-ray diffraction 4. X-ray fluorescence 5. Adsorption from solution 6. Light microscopy including phase contrast, interference contrast, polarized light, and dispersion staining 7. Electron microscopy including transmission electron microscopy and scanning electron microscopy After initial investigation, the three most likely candidates were; 1. Light microscopy 2. X-ray diffraction 3. Transmission electron microscopy In order to determine which of the above would meet all criteria for the test, we secured samples of pure talc and tremolite from various deposits owned by Pfizer. Samples of pure and carefully characterized asbestos minerals were obtained from the International Union Against Cancer, (UICC), Pneumoconiosis Research Unit, Llandough Hospital, Penarth Glamorgan, United Kingdom. The talcs and asbestiform minerals were examined in the pure or as-received state, their characteristics noted and mixtures made to determine if detection of asbestos minerals was possible at low levels and, if so, what the minimum detection levels might.be. Experimental X-ray diffraction patterns were obtained for all the minerals and mixtures used in this study employing the conventional technique of scanning at rates of 0.5 to 1.0 degrees 2 theta per minute. The samples were then subjected to scrutiny by optical and electron microscopy. During this procedure it was discovered that certain mixtures and mineral species shown to be free of asbestiform minerals by the conventional x-ray diffraction and light microscopy techniques exhibited fairly large percentages (5% or more) of fibrous tremolite and/or asbestiform minerals when viewed by transmission electron microscopy. Oelineation of the reasons for this paradox enabled us to develop reliable techniques for detecting tremolite and the asbestiform minerals at the 0.2 percent level in most talcs by x-ray diffraction. Even lower levels of these minerals are detectable by transmission electron microscopy. Light Microscopy Techniques employing the optical microscope have been used to identify mineral speci- mens for a long time. Techniques that we have examined include polarized light microscopy, transmission light microscopy, phase contrast, and dispersion staining. The difficulty which we encountered in applying these techniques to the problem at hand is that while they work well with pure samples of fairly massive fiber length (3 to 5 microns 326

S45 and larder), observations by transmission electron microscopy have shown that naturally occurring asbestiform minerals often lie below the working resolution of the light microscope. While massive fiber bundles can often be observed by either light or electron microscopy, the observation of individual fibers smaller than approximately 1 micrometer long by 0.02 micrometers wide requires the high resolution capability of the transmission electron microscope. In addition, the limit of detection is confounded by the presence of "apparent fibers" formed when thin talc plates curl up at the edge and roll into a cylindrical morphology. The limit of positive detection and identification of fibers is felt by us to be too high to be of any commercial value. X-ray Diffraction The d-spacings for talc, chlorite, tremolite, and the asbestiform minerals are seen in Table 1. The values given in Table 1 are averaged for pure materials and can shift as much as ±0.02 to ±0.03 nanometers depending upon sample preparation, the level at which the constituent is found in the parent matrix, and the specimens conformity to the idealized chemical composition. While attempting to detect tremolite and the asbestiform minerals in talc at concentrations of two to five percent or below, we found that the normal scanning rate of 0.5 to 1 degree 2 theta per minute was not satisfactory for the following reasons: 1. The noise level was too high providing a detection limit of only a few percent. 2. It was difficult to accurately quantify data from the high noise tracing obtained. Table 1. Principal lattice spacings of talc and related minerals by x-ray diffraction Cu K alpha. - - - - Principal d-spacings in angstroms - - - - Mineral Species 1 2 3 4 5 6 Talc - 9.51 4.73 3.14 2.61 2.50 4.62 Chlorite 14.00 7.03 4.70 3.53 2.82 Tremolite 8.38 3.38 3.12 2.94 2.71 3.27 2.81 2.59 2.53 Chrysotile 7.38 4.55 3.66 2.45 1.54 Amosite 8.26 3.27 3.07 2.77 Anthophyllite 9.50 8.40 4.58 3.25 3.13 3.06 Crocidolite 8.43 4.51 3.43 3.11 2.72 In order to avoid these difficulties, an automated step-scanning method was employed in which the diffractometer was moved in increments of 0.05 degrees 2 theta, and the intensity of x-ray radiation at each step measured for-a total of two minutes. An intensity versus degrees 2 theta plot over the area of interest of 9 degrees to 11 degrees 2 theta was made. Figure 1 shows this step-scan method plotted for a talc which showed no evidence of any asbestos or asbestiform content. Calibration curves were established by integrating the area under the appropriate x-ray diffraction peak of mixtures of I to 10 percent of the species under investigation, the remainder being a sample of talc shown to be tremolite and asbestiform mineral free by the method of transmission electron microscopy to be outlined below. Figure 2 shows this step-scan plot for the one and five percent addition of tremolite to the base talc matrix. Figure 3 shows the calibration curve obtained by this technique for asbestos in talc, and figure 4 shows the same type of plot 327 2063105120

8000 y i ~ ~ ~ ~ ~ 7000 ' , , , , ~ H 6000 p N z 0 5000 i 0 2000 : j 11° 10.50 10° 0.50 9° Degrees 2 Theta Figure 1. Step-scan plot of intensity versus degrees 2 theta for Pfizer, Inc. Montana Talc. 328

8000 7000 6000 E D a .+ ~ 04 5000 y 4 z 0 0 U 4000 3000 2000 5% addition of tremolite , } , i i 11° 10.50 10.0° 9.5° 9.00 Degrees 2 Theta i11L_ Figure 2. Step-scan plot of intensity versus degrees 2 theta showing the effect of adding 1% and 5% tremolite to talc. 329

38,000 34,000 30,000 26,000 22,000 18,000 14,000 10,000 6,000 2,000 0 1 2 3 4 5 6 7 8 9 PERCENT ASBESTOS 10 Figure 3. Percent asbestos as a function of intensity of diffracted x-rays. I / Figure 4. Percent tremolite as a E 000 / function of intensity L 4, ( / I fj of diffracted x-ra s. z 0 $ -3,000 330

e g 5 for t`remolite in talc. The minimum detection limit was calculated as that equivalent to three times the square root of the background. For tremolite in talc, the minimum detec- tion limit was found to be approximately 0.2 percent. For chrysotile and the other asbestiform minerals, the minimum detection level obtained by this method is approximately 0.5 percent. This can only be achieved in the absence of chlorite, however. Attempts to remove chlorite by careful acid wash succeeded only in rendering the chrysotile amorphous to the x-ray beam with the result that no x-ray spectrum was obtained in the chrysotile region. Further experimentation revealed that the presence of tremolite at fairly low levels tended to mask or interfere with the detection of some of the other asbestiform minerals. It was thus clear that another technique would be required in these special cases in order to be able to achieve the unambiguous analysis originally required. Electron Microscopy By virtue of its ability to examine individual particles in minute detail and at very high magnifications, the transmission electron microscope has been found by us to provide the technique, ancillary to x-ray diffraction, that is needed to complete the unambiguous detection and identification of asbestiform minerals in talc. The morphology of the asbestiform minerals and tremolite is generally described as acicular or fibrous. This immediately serves to isolate them from the platy talc matrix even in the presence of chlorite, since the chlorite morphology closely resembles that of the talc. If the sample, made into a specimen for the transmission electron microscope, is or can be made homogenous, and a careful examination of approximately 100 different fields of view fails to reveal any fibrous material, then that talc is felt by us to be free of tremolite, chrysotile, and the other asbestiform minerals. The lower detection limit of this technique is difficult to assess since one is often dealing with individual crystals. Figure 5 shows a typical field of view of thp fiber free Montana talc used as a basis of comparison in this study. In order to obtain some idea of the amount of fibrous material in a talc, we carefully counted the number of fibers present in each of 100 fields of view of samples contaminated with 0.1, 0.5, and 1.0 percent by weight of fibrous asbestos. The average number of fibers in each field of view is then plotted as a function of the weight percent of fibers added. A linear relationship is seen to exist between the average number of fibers and the weight percent, as illustrated in figure 6. Table 2 shows the results of the fiber count and the raw data for the calibration curve construction. In the range of 0.1 to 1.0 percent, the linear relationship shows an excellent correlation coefficient [9]. We have plotted data of other investigators up to as high as five percent and found that this linear relationship still holds. An interesting point to note at this time is that the standard deviation for 0.1 weight percent of fibers is more than half of the value of the average number of fibers in the same field of view. Further investigations in our laboratories have Table 2. Fiber count - calibration curve. Weight % fibers Total fibers/100 FOVa Avg. # fibers/FOV Std. deviation 1.0 1183 11.83 7.07 0.5 634 6.34 2.49 0.1 206 2.06 1.39 a FOV = field of view. y= mx + b 3~_b = 2.92 fibers/FOV m= 10.86 Correlation coefficient = 0.99997 b = 0.95 331 V

Cp Figure 5. Pfizer, Inc. Platy Montana Talc. Bar is one micron. N O 01 332 W f.+ 0 ~ ..

w w 12 10 4 2 1 I I I I I I I I__ 0.1 0.2 0.3 0.4 0.5 0.6 0.9 0.8 0.9 1.0 Weight Percent - Fibers in Talc Figure 6. Fibers in talc. 9ZiS0i£90Z

convinced us that this linear relationship does not hold much below 0.1 percent. This is intuitively obvious upon an examination of figure 6 which, you will remember, does not pass through the origin. Somewhere below 0.1 weight percent of fibers in the talc, the linear relationship no longer holds, and the line curves down through the origin. Repeated examinations have confirmed the fact that the Montana talc used in this study is fiber free. Below 0.1 weight percent the data must become so scattered as to be meaningless on a statistical basis. A typical field of view of a Montana talc which was doped with 1.0 percent chrysotile fibers is seen in figure 7. A semi-qualitative estimate of the weight percent fiber content can be easily obtained by reference back to the calibration curve. It is mandatory, however, that the samples under investigation be prepared in exactly the same manner as the samples used in the original calibration curve construction. It is also mandatory that one be certain of the homogeneity of the calibration samples and the Figure 7. Commercial talc product with a 1% addition of chrysotile asbestos. Bar is one micron. 334

sample under investigation. Great care must be exercised in the sample preparation, or the results become totally meaningless. Figure 8 shows a commercial talc in which approximately one percent of naturally occurring chrysotile was obscured from detection by the method of x-ray diffraction because of the presence of chlorite. w Figure 8. Commercial talc with naturally occurring co-deposits of chlorite and chrysotile asbestos. The asbestos is present at approximately the 1% concentration level. Bar is one micron. 335

1 Selected area electron diffraction was used in conjunction with the examination of morphology. Using this combined method, a single crystal or particle can be selected and analyzed. Single particles usually yielded spot patterns, but if a group or bundle of fibers was found and would transmit electrons, a polycrystalline ring type pattern would result. The use of selected area electron diffraction is mandatory to prove that the pseudo fibers of talc caused by plate-edge curling and talc plates on edge were actually talc, and not tremolite or an asbestiform mineral. A comparison of selected area electron diffraction patterns of these pseudo-fibers to that of the talc platelets showed that the identical compound, talc, was the only species present. Table 3 lists the principle electron diffraction maximum for talc, tremolite, and the asbestiform minerals [10]. In almost all cases, many more spots or rings were observed than are reported here. In Table 3, only the strongest lines which are the ones most likely to be observed have been tabulated. Table 3. Selected area electron diffraction maxima for talc and related mineralsa (in angstroms). Talc Tremolite Chrysatile Amosite Anthophyllite 4.60 4.51 4.58 3.88 4.58 2.62 2.59 3.67 3.45 2.65 2.32 2.53 2.61 3.00 2.27 1.74 2.32 2.14 2.64 1.75 1.59 2.27 1.70 1.74 1.55 1.53 2.04 1.55 1.61 1.33 1.33 1.86 1.34 1.55 1.28 1.28 1.69 1.29 1.32 1.23 1.65 a The data for chrysotile, amosite, and anthophyllite were taken from reference [11]. Conclusions The present work has shown that properly prepared samples of talc can be examined by x-ray diffraction to detect tremolite at levels down to 0.2 percent and chrysotile at the 0.5 percent level in the absence of chlorite. In the presence of chlorite, and at concen- tration levels lower than those stated above, the transmission electron microscope was found to provide reliable detection and identification of fibrous tremolite and the asbestifors minerals. The transmission electron microscope is the most sensitive we have found, and appears to be a more or less referee technique since, when morphology observations are coupled with selected area electron diffraction studies, there are no known interferences. light microscopy was helpful only in screening samples with large particles and high concentrations of objectionable fibers. Using the above techniques, we have been able to screen large numbers of talc speci- mens. We have been able to detect chrysotile and/or tremolite and the asbestiform minerals at levels down to 0.1 weight percent of fiber. We have been able to detect the asbestiform minerals in low concentration specifically by transmission electron microscopy with selected area electron diffraction, when the presence of the asbestos was masked by the presence of chlorite (which was also present at less than 5% concentration). We, there- fore, feel that we have a technique that allows us to detect and identify chrysotile fibrous tremolite, and asbestiform minerals at concentrations down to 0.1 percent by weight. 336

c, S References [1] Cralley, L. J., Key, M. M., Groth, D. H., Lainhart, W. S., and Ligo, R. M., Fibrous and mineral content of cosmetic talcum products, J. Amer. Indus. Hyq. Assoc. , 29, 350 (1968). [2] Hogue, W. L. Jr. and Mallette, L. S., A study of workers exposed to talc and other dusting compounds in the rubber industry, J. Indus. Ea. Toxical. , 31, 359 (1949). [3] Smith, K. W., Plumonary disability in asbestos workers, Arch. Ind. Health, 12, 198 (1955). [4] Schepers, G. W. H. and Durkan, T. M., The effects of inhaled talc-mining dust on the human lung, AMA Arch. Indus. Health, 12, 182 (1955). [5] Brodeur, P., The magic mineral, New Yorker Ma aZine, p. 12, October 1968. [6] Merliss, R. R., Talc-treated rice and Japanese stomach cancer, Science, 173, 1141- 1142 (1971). [7] Sax, N. Irving, Ed., "Talc", dangerous properties of industrial materials, Reinhold Publishing Corp., New York, 1963, p. 1217. [8] Sax, N. Irving, Ed., "Talc", dangerous properties of industrial materials, Reinhold Publishing Corp., New York, 1963, p. 469. [9] Stanley, H. D., The detection and identification of asbestos and asbestiform minerals in talc, 34th Annual Proceedings of the Electron Microscopy Society of America, p. 618-619, August, 1976. [10] Timbrell, V., Characteristics of the International Union Against Cancer Standard Reference Samples of Asbestos, Proc. Int. Pneumoconiosis Conf., Johannesburg, 1969. 337

Discussion ,1. SCHELTZ: As the spokesman for the Cosmetic, Toiletry, and Fragrance Association, I would like to make several comments. First: In a survey conducted recently by that organization among its member companies, some thirty-four hundred samples of cosmetic talc from both domestic and international sources were analyzed and not a single sample was found to contain chrysotile asbestos. We are aware that the spiking of chrysotile asbestos into talc can be analyzed effectively by x-ray diffractometry. These samples of talc are cosmetic which, by definition, means that they contain at least 90 percent of the actual talc mineral species. I would also like to comment on quantitative analysis of amphibole minerals, by x-ray diffractometry. While x-ray diffractometry is a good technique to detect amphibole minerals, one needs to be very cautious in attempting to perform a quanti- tative analysis. I think Dr. Haartz from NIOSH just pointed out that there are major differences based not only on compositional variations, but also morphological character- istics that make not only peak heights but also integrated peak intensity variable. So, while x-ray diffractometry is a good method for detection, it is not necessarily good for quantitative analysis. I would also like to point out that the Cosmetic, Toiletry, and Fragrance Association is currently undertaking an extensive analysis of consumer talcum products for the traces of amphibole minerals. H. STANLEY: As I understand it, your first point is that x-ray diffraction is not particularly quantitative for determination of amphiboles in talc. We haven't found that to be the case in our laboratory, and I think there are a number of people here that I have been talking to the last several days that have had the same experience. The x-ray diffraction is good if you want to know, for example, the total amount of tremolite present, but if you want to know if some of that tremolite is fibrous, then as I attempted to point out, you have to go to transmitted electron microscopy with selected area diffraction. SCHELTZ: That's exactly my point. ..... (rest inaudible) ..... As to the second point, we were talking about cosmetic grade talc of at least 90 per- cent purity, the purity of the Montana talc is in excess of 96 percent, so I understand your point. 338

C V [ National Bureau of Standards Special Publication 506. Proceedings of the Workshop on Asbestos: Definitions and Measurement Methods held at NBS, Gaithersburg, MD, July 18-20, 1977. (Issued November 1978) MISIDENTIFICATION OF ASBESTOS IN TALC Jerome B. Krause Colorado School of Mines Research Institute Golden, Colorado 80401 and William H. Ashton Johnson & Johnson Raritan, New Jersey 08869 Abstract Both optical microscopy and x-ray diffraction (XRD) are widely used to detect minerals associated with talc. Optical microscopy can determine the morphology of a particle, but cannot always fully identify the specific mineral. Although XRD is an excellent screening technique for the detection of minerals associated with talc, the method can misidentify minerals due to interferences, interpretive errors, and the inability to determine morphology. Methods for reduction or elimination of these problems include special techniques of sample preparation and x-ray diffraction, combined with microscopic examination (both optical and electron). Key Words: Amphiboles; asbestos; chlorite; electron microscopy; fiber; morphology; optical microscopy; x-ray diffraction; talc. Introduction There are many ways to analyze and study any naturally occurring material. The conclusions reached will often vary widely depending on the expertise and specific interest of the investigator. That situation sums up the present status of "asbestos"; it is also the status of minerals which are associated with "asbestos"; and it is becoming the status of other minerals which can be naturally associated with talc. Popular methods of analysis can give the wrong answer - namely that asbestos is present when it certainly is not. That problem (misidentification) is not so much one of limitations of the methods, but rather one of misinterpretation of data, and failure to recognize the mineralogical background required to certify mineral purity,,for example, when analyzing sheet silicates for asbestos. Unfortunately, one main factor is that asbestos has now developed variable definitions, depending on whether the point of view is mineralogical, industrial, medical, or regulatory. The medical definition is most concerned with whether or not the particles are biologically active; the industrial definition is dependent upon flexibility and weavability; the mineralogical definition upon crystallography; and the regulatory definition upon size and aspect ratio. The word "asbestos" stems from ancient Greek and has always referred to a very fibrous industrial mineral product. Since asbestos has historically related to a mineral exploited as an important industrial commodity, we think a combined mineralogical and industrial definition should take precedence [1,2]1. Other presentations during this IFigures in brackets indicate the literature references at the end of this paper. 339 2063105132

I workshop have amply covered the aspects of asbestos terminology, and it is not our intent to provide comprehensive coverage of that subject. Our primary objective is to review some of the basic principles of analysis, and to point out problem areas where identification of "asbestos" has been abused. Analysis Methods and Misidentification of Asbestos It is useful to categorize the various analytical methods which have been applied to talc to highlight inherent principles which lead to misidentifying asbestos as being present. We offer the following general comments on the three principle determinative properties (chemical composition, morphology, structure). Chemical Composition It is well known that every mineral has a specific chemical composition, and that each mineral has an ideal theoretical chemical formula (configuration). Unfortunately, many investigators overlook the fundamental point that chemical composition does not identify a specific mineral. A simple example will bring that point into focus: A pearl, an oyster shell, a slab of marble, a piece of chalk, and the minerals aragonite and calcite are obviously different materials, and . yet each will be identified as calcium carbonate. That is to say, chemical analyses will identify them all as the same substance, where everyone knows that a pearl is not a piece of chalk. The same situation exists in certain phases of asbestos analysis. For example, chrysotile, antigorite, lizardite, sepiolite, chlorite, and talc are all hydrous magnesium silicates. But a Meerschaum pipe (sepiolite) is certainly not chrysotile asbestos in spite of the fact that chemical analysis alone could lead to that misidentification. Accordingly, chemistry alone does not identify a mineral, nor do those sophisticated instrumental methods which are based on chemical principles, such as: Wet Chemical Analysis Classical (gravimetric, volumetric) Instrumental (atomic absorption, flame emission) Microprobe (electron and ion) Emission Spectrograph Mass Spectrograph X-Ray Fluorescence Morphology Although the shape of a mineral particle is one of the key characteristics in the identification of a mineral, shape alone cannot be the sole determinant of a specific mineral species. There are hosts of minerals in different mineral classes whose particles have' the same shape. They exist across the spectrum of all classes of minerals and the possibilities are beyond comprehension. Even if we limit ourselves to minerals which occur in the true fibrous state, we would estimate there are up to 100. There have been instances where nonasbestos particles have been misidentified as chrysotile in talc because shape alone was the index used. Methods based on morphology include: Optical Microscopy Automated Image Analyzers Electron Microscopy (SEM and TEM) 340

Structure The configuration of atoms in the crystal lattice of a mineral does not necessarily determine a mineral species. The atomic arrangement at the molecular level does not always carry through to the external visible physical form. That is to say that methods based on molecular structure can misidentify a mineral. For example, chrysotile asbestos is classified with the sheet silicates because of its crystal structure arrangement, but it certainly does not occur in flat sheets like the micas or its sibling, antigorite. Methods of identification which relate to molecular structure are: Infrared Spectroscopy Differential Thermal Analysis X-ray Diffraction Electron Diffraction In general then, no single property defines a mineral, and no single method which depends on one property can identify a specific mineral. Conversely, methods which depend on a single factor or characteristic of a mineral can give misidentifications. Two Popular Methods Optical microscopy and x-ray diffraction methods require some additional discussion primarily because they have received widespread attention by industry and gpvernment laboratories as possible monitoring techniques. Although both these methods are fundamental to the science of mineralogy and are highly reliable in the hands of experts, complications arise when shortcuts are taken in the professional procedures. Optical Microscopy When an experienced optical mineralogist or crystallographer identifies a mineral with a petrographic microscope, he can come to a remarkably accurate conclusion. The reason for high accuracy is that not one but several specific properties are determined, such as refractive indices, extinction angle, birefringence, and optical orientation. Specific training and wide mineralogical background are required to get the right answer. In contrast, current optical methods in federal regulatory proposals relating to asbestos presume that asbestos is present in the first place. The analyst then merely observes the mineral particle for size/shape. Consequently, those methods which depend solely on aspect ratio give misidentification. They misidentify the presence of asbestos by such simple oversights as looking at a platelet on edge and counting it as an asbesti- form particle. It is not necessary to elaborate on the other shortcomings of those methods in view of the recent NBS report on the analysis of 80 industrial talcs [3] evaluating that methodology. The same shortcomings were also recently corroborated in a study conducted by Harvard University and NIOSH [4]. However, there are a few rare cases where abnormal crystal habit can be misleading and subtly can lead to a misidentification. Optical microscopy is most vulnerable to this type of misidentification. For example, talc normally occurs as micaceous plates, but rare acicular talc does exist, and one must be very careful to avoid misidentifying the rare occurrence as asbestos. As an example, our XRD examination of an industrial acicular talc sample has identified the presence of significant amphibole (probably tremolite). However, when the material was subjected to thorough petrographic examination it was found to be composed of free grains of columnar amphibole and acicular talc and composite talc- amphibole. The significance is that an erroneous conclusion could be reached by misidentifying such a rare talc variety as asbestos, if only aspect ratio and simple optical microscopy were used. 341 2063105134

Thus, simple optical microscopy can determine the morphology of a particle, but if used alone it cannot always fully identify the specific mineral observed. . X-Ray Diffraction Although x-ray diffraction (XRD) is a valuable technique, it cannot determine the physical shape of a mineral particle, and for that reason it cannot determine whether or not a sample is asbestos. Furthermore, it cannot distinguish between two mineral varieties in the same mineral class in cases such as the asbestos minerals and their nonasbestiform analogues. It is surprising that such a basic shortcoming continues to be overlooked by responsible investigators alleging to have identified asbestos by XRD. One result of the inability of powder XRD to differentiate between the asbestiform and nonasbestiform varieties of a mineral is the potential error of prejudging an XRD detected phase to be the asbestiform variety. For example, preparing calibration standards of mixtures of talc plus chrysotile could have the effect of causing a serpentine peak in an unknown sample to be prejudged as the asbestiform variety, i.e., chrysotile. A mixture of talc spiked with the serpentine mineral chrysotile will give the same XRD pattern as a mixture of talc spiked with the very common platy serpentine mineral antigorite. It should be obvious that an unknown talc showing a serpentine peak cannot be pre,7udged or branded as containing chrysotile asbestos under such circumstances. Unfortunately, the literature has articles by responsible authors who have overlooked that error in logic [5,6,7]. For research purposes only, single crystal XRO can provide information as to whether or not the specimen could be asbestos. However, due to the difficulty of handling minute specimens, single crystal XRD is inadequate for particles smaller than about 20 x 5 pm, and, of course, is also inadequate for routine monitoring procedures. Amphiboles Each of the five amphibole minerals, anthophyllite, cummingtonite-grunerite, riebeckite, tremolite, and actinolite has an asbestiform variety, namely anthophyllite asbestos, amosite, crocidolite, tremolite asbestos, and actinolite asbestos, respectively. Tremolite asbestos is quite rare, and actinolite asbestos is so rare that a recent NIOSH project to prepare reference standard minerals has been unable to locate a source of pure actinolite asbestos [8]. The amphiboles (named from the Greek "amphibolos," meaning ambiguous) are characterized by similar crystal structure and wide variation in chemical composition and appearance. All amphiboles have XRD patterns which are similar° and are characterized by having their (110) or (210) diffraction peaks occur within ±0.2A of each other (Table 1, Figure 1). Reliable identification of individual amphibole species is difficult in the absence of confirming composition data. Examination of Table I and Figure 1 illustrates that attempted identification of a specific amphibole on the basis of d plo) or d Q10) has good potential for being in error. For example, selection of Joint Committee on Powder Diffraction Standards (JCPDS) card 13-437 as being definitive of tremolite presents serious problems. Twenty-nine additional JCPDS amphiboles have their (110) or (210) peaks within ±0.1°20 of this tremolite (110) peak at 10.56°20. Identification of an amphibole as tremolite on the basis of a peak at 10.56°20 is obviously an identification with very low reliability. In other words, a peak at that location is not necessarily the mineral tremolite since it could be one of 29 other minerals. 342

Table 1. CS or d(210) peak position, and Pmphibole JCPDS Card No's., d(110) relative intensity. JCPDS card # Aa 29 Cu I Name Z3-118 8.58(1) 10.31 100 prieskaite 10-456 8.550) 10.35 100 richterite 20-734 8.53(1) 10.37 70 mboziite 20-378 8.52(1) 10.38 100 dashkesanite 14-633 8.51( )1 10.39 70 arfvedsonite 21-149 8.51(1) 10.39 55 hornblende 19-467 8.50(1) 10.41 100 ferropargasite, syn 20-982 8.50(1) 10.41 65 richterite, syn 23-665 8.48(1) 10.43 45 richterite, calcian, syn 23-664 8.47(1) 10.44 ' 35 edenite, sodian, syn 23-667 8.47(1 10.44 45 richterite, calcian, syn 23-663 8.46(1 10.46 40 eckermanite, calcian, syn 9-434 8.45(1 10.47 50 hornblende 13-499 8.45 1 10.47 100 ma9nesioriebeckite 20-656 8.45 1 10.47 100 magnesioriebeckite 20-470 8.44 1) 10.48 100 crossite 23-666 8.44(1) 10.48 40 tremolite, sodian, syn 20-469 8.43( )1 10.49 100 hastingsite 23-1405 8.43(1) 10.49 80 edenite 23-1406 8.43( )1 10.49 40 paragasite 20-1310 8.43(1) 10.49 40 tremolite, syn 10-428 8.42(1) 10.51 100 richterite, fluor, syn 23-603 8.42(1) 10.51 100 tirodite 10-431 8.41( )1 10.52 8D edenite, fluor, syn 19-1061 8.40(1) 10.53 100 riebeckite 20-481 8.40 (1) 10.53 100 hornblende 20-1390 ( 8.40 )1 10.53 90 winchite 23-302 ( 8.40 1 10.53 100 cummingtonite, mangoan 19-1063 8.39(1 10.54 70 richterite 13-437 8.38(1 10.56 100 tremolite 17-478 8.38(1 10.56 65 kaersutite 23-495 8.38(1 10.56 80 eckermanite 9-330 8.37(1) 10.57 100 tremolite, fluor, syn 17-750 8.36(1) 10.58 25 richterite, ferrian 20-386 8.35(1) 10.59 40 eckermanite, syn 22-531 8.35(1 10.59 30 joesmithite 16-401 8.33(2 10.62 70 anthophyllite, magnesian, syn 17-725 8.33(1 10.62 100 grunerite 17-745 8.33(1 10.62 100 grunerite 20-376 8.31(1) 10.65 100 crossite 17-726 8.30(1) 10.66 100 cunmingtonite 20-484 8.29(1) 10.67 100 richterite 13-506 8.27(2) 10.70 80 gedrite 23-679 8.27(1) 10.70 90 glaucophane 9-455 8.26(2) 10.71 55 anthophyllite 20-453 8.26(1) 10.71 100 glaucophane 11-253 8.23(2) 10.75 100 ferrogedrite 23-310 8.20(1) 10.79 75 richterite, ferrian 13-401 8.11(2) 10.91 100 holinquistite a (1100)2. Maximum 428(Cu) = 10.91° - 10.31° - 0.6° Table 1 illustrates the very close proximity of the (210) or (110) XRD-peak of all amphiboles, showing the inability to identify a specific amphibole on the basis of d(210) or d(110)' 343

~ so 70 60 RI 50 TREIdOL1TE 20-656 19•1 I 13-437 ~ 13-499 40 ~ ANTXOAryLLITE 20 16•401 9-455 l0 n 100 90 60 70 60 50 40 30 20 10 0 10.4 10.6 10.6 DEGREES 28 (CuK(C) 10.7 10.9 Figure 1. Anlphibole d(110 or d(210) - peak positions (20 for CuK ) and relative intensity. An additional problem further affecting the reliability of identification by XRD is the effect of shift in peak position caused by slight mispositioning of the sample surface in the instrument. For example, a 100 pn6 mispositioning of the specimen surface will result in a shift of approximately 0.6-0.7 A in d-spacing at low 20 angles [9]. A slight shift in the position of the peak (fro. a different amphibole or mispositioning of the sample surface, for example) could go unnoticed, resulting in misidentification of an amphibole that is not even present. In order to conclusively identify an amphibole by XRO, it is necessary to have an essentially complete diffraction pattern. In order to obtain such an XRD pattern, the sample must have a relatively high amphibole content and the pattern must be acquired with a time-consuming slow scan. Acquisition and interpretation of such patterns is time- consuming, and discourages proper application of the full procedure, especially for routine monitoring where large numbers of samples require analysis. Shortened procedures, such as single peak identification of amphiboles, provide good opportunity for misidentification. The shortened procedure of single peak identification was apparently used in a 1972 paper [7], where our examination of some of the same samples disagreed with 0 ~ identifications of serpentine, tremolite-actinolite anthophyllite, and anhydrite. w 0 344 ~ r+ w a

aa t~? Chlorite-Serpentine Chlorite is one of the most common accessory minerals found associated with talcs. The chlorite group of minerals are somewhat analogous to amphiboles in that they exhibit a wide 'variation in chemical composition and all have a similar crystal structure. The diagnostic Fplor te basal xRD peaks (001), (002), and (004) are characteristic, and occur at about 14A, 7A, and 3.5tt, respectively. As in the case for the amphiboles, specific identification of a particular chlorite species by XRO is difficult. The XRD problem with chloritic talcs is that the serpentine first order basal peak overlaps the chlorite (002) peak, and the corresponding serpentine second order basal peak overlaps the chlorite (004) peak. Generally, however, the chlorite (004) and serpentine second order peaks are separate enough to allow unambiguous determination of the presence of both phases when present in adequate amounts to give definable peaks. Tables 2 and 3 and Figures 2, 3, and 4 are compilations of JCPDS data for the positions of the (004) basal peak for chlorites and (002), (004), or (0012) basal peak for serpentines, respectively. Table 2. Chlorite JCPDS Card No's., d(004) peak positions, and relative intensity. JCPDS , card Y A 2e Cu I Name 10-183 3.60 24.73 100 penninite 20-671 3.60 24.73a 90 k8mmererite 16-351 3.59 24.80 70 chlorite lb 12-185 3.57 24.94 85 kotschubeite 7-160 3.58 24.87 60 kotschubeite 19-749 3.56 25.01 80 clinochlore 7-77 3.558 25.03 50 sheridanite 16-362 3.55 25.08 80 chlorite la 19-751 3.55 25.08 65 sudoite 22-712 3.55 25.08 45 nimite 7-165 3.545 25.12 60 grochauite 7-78 3.541 25.15 60 thuringite 7-171 3.541 25.15 80 diabantite 12-242 3.54 25.16 100 leuchtenbergite 7-76 3.537 25.18 50 ripidolite 13-29 3.53 25.23 80 thuringite 7-166 3.523 25.28 50 daphnite 12-243 3.52 25.30 92 aphrosiderite 21-1227 3.52 25.30 100 thuringite 3-67 3.49 25.52 100 thuringite a d(115)' Table 2 illustrates variation in position of the chlorite d(004) XRD peak. Table 2 should be compared with Table 3 to see that the chlorite and serpentine XRD peaks overlap and interfere with each other. Identification and quantification of serpentine in the presence of chlorite is extremely difficult at best. 345

(oo ~ eo TD 60 50 40 30 20 io 0 4 W A 0 ------ ------------- --~-- - - -- - ---- - - Emm ----------- --- ---- ---- a _. .. _ ___. H0 ----- - _-~ ------_ TO CHLORITE ~ ~ Fi 40 d -.-___ 30 10 i . O a DEGREES 28 (Cu Ka) Figure 2. Chlorite d(004) - peak positions and relative intensity. The data of Table 2 are presented in graphical form showing the variation in position of the d(004) XRD peaks for different chlorites. Selection of JCPDS card 16-362 as diagnostic for chlorite can obviously result in misidentification. 6£TSOI£90Z

cs Table 3. Serpentine, Kaolinite, Halloysite, and Dickite JCPDS Card Nos., peak position, miller index (hkl), and relative intensity. JCPOS Card # , A 2e Cu I hkl 18 -1779 3.67 24.25 80 (002) 9-444 3.66 24.32 100 (0012 21-543 3.65 24.39 70 (004) 7-417 3.63 24.52 300 (102) 11-386 3.62 24.59 60 (002) 21-963 3.61 24.66 80 (002) 12-583 3.56 25.01 80 (0012) 13-4 3.56 25.01 70 (0012) 7-339 3.55 25.08 100 (002) 11-388 3.55 25.08 100 (0012) 7-315 3.52 25.30 100 (002) 9-493 3.52 25.30 100 (004) 6-221 3.58 24.87 100+ (002) 14-164 3.579 24.88 80 (002) 12-447 3.56 25.01 50 (002) 9-453 3.63 24.52 90 (002) 10-446 3.58 24.87 100+ (004) Chlorite 29 Range: 24.73 - 25.52 Serpentines lizardite, 1M antigorite, 60 chrysotile, 2M antigorite, 6M lizardite, 10, aluminian antigorite, 6M antigorite, 60, aluminian antigorite, 60, aluminian berthierine antigorite, 60, syn berthierine amesite Kaolinites kaolinite, 1Md kaolinite, 1T kaolinite, 1T Halloysite halloysite, dehydrated Dickite dickite 2M, Table 3 illustrates variation in position of XRD peaks of serpentine, kaolinite, halloysite, and dickite. The XRD patterns of these minerals interfere with each other and with chlorite (see Table 2). N O ~ W F+ 347 ~ .+ A O

~ SERPENTINE dpapj qp4l.(OOIZ) ®Kl,OLINITE dpp21 ~HALLOYSITE d10m oDICKITE dppsy w P ~ YEGflEES 28 /tuMI Figure 3. Peak positions and relative intensities. The data of Table 3 are presented in graphical form to illustrate the variation in position and interferring overlap of XRD peaks of serpentine, kaolinite, halloysite, and dickite. rrrsorsvoz C4 ,

w A b Z6TSOT£90Z - CHLORITE d(oo.) ® KAOLI NITE dM) I SERfENTINE dVy,(oo4),taa2) HALLOYSITE d(00z) ° DICKITE dW4) Figure 4. Peak positions and relative intensities. The data of Tables 2 and 3 are presented combined, illustrating the problems of XRD identification when chlorite and serpentine, and possibly kaolinite, halloysite, or dickite are also present.

Three essential features are demonstrated in Tables 2 and 3, and Figures 2, 3, and 4: 1. The diagnostic peaks show considerable variation in the position in which thV occur (a2e=0.79° for chlorites and 1.05° for serpentines). 2. The chlorites and serpentines overlap and interfere with each other. 3. Basal peaks of the clay minerals kaolinite, halloysite, and dickite overlap the positions of the chlorite and serpentine peaks, and will interfere when present. The significance of the chlorite-serpentine interference is increased by the fact that chlorite is a very common accessory mineral associated with talcs, whereas serpentine is much less commonly associated. In spite of the chlorite-serpentine problem, numerous investigators have performed XRD identification and/or quantification of serpentine in chloritic talcs. It is obvious to us that they have misidentified asbestos as being present by overlooking the chlorite/serpentine interference and by misconcluding that a chlorite peak was serpentine. Other Methods Infrared Spectroscopy (IR) The infrared absorption spectrum of a material results from vibrational and bending frequencies of various atomic bonds within the structure. For example, Si-0 stretching frequencies produce similar IR peaks for all silicate minerals. As a result, IR spectra are not particularly useful for identifying the minerals present in a mixture, and the method certainly is not capable of determining whether or not a detected mineral is the asbestiform variety. Differential Thermal Analysis (DTA) The rearrangement or decomposition of mineral crystal structures due to thermal heating is a characteristic and reproducible reaction. It follows that DTA can identify specific minerals in a mixture but the method is not capable of determining morphology. Therefore, any DTA data which might point to the presence of a serpentine mineral could lead to misidenfying chrysotile asbestos in a talc when the mineral could well be a normally occurring platy antigorite having the same DTA pattern. Electron Microscopy Electron microscopic techniques of identification of asbestos have been amply covered in other presentations during this workshop. We do not intend to cover that subject again, but rather to point out some areas where asbestos can be misidentified. The high magnification attainable with electron microscopy is, in itself, inadequate as the sole index of mineral identity. For example, chrysotile is often identified by the presence of a hollow central core and streaked electron diffraction spots. But the clay mineral halloysite also crystallizes in that form and will produce a similar electron diffraction pattern. Therefore, In the absence of exact chemical composition, halloysite can be misidentified as asbestos. Similar care must be exercised to avoid misidentifying other fibrous clay inerals as asbestos, e.g., attapulgite and alpha sepiolite. In addition, talc ribbons can be mistaken to be asbestos, especially when some talcs have particles which roll up into spiral tubes giving the appearance of a chrysotile particle. Selected area electron diffraction is routinely used to identify a mineral particle as amphibole. Many investigators simply observe the electron diffraction pattern in the microscope and decide on the basis of general pattern geometry whether or not the particle is an amphibole. This can lead to misidentification, since numerous other minerals can give electron diffraction patterns with amphibole pattern geometry [10,11]. Careful measurement of an electron diffraction pattern is required in order to identify the type 350

of mineral which produced the pattern. Chemical composition is further required in order to have a chance at identifying the particular species when the mineral is a member of a complex group such as the amphiboles. Otherwise, misidentification will result. Cosmetic Talc Free from Asbestos ~ In the United States, we have a self-regulating association known as the Cosmetic Toiletry and Fragrance Association. In certifying the purity of the talcs which they use, they are aware that no single method can identify asbestos and their most recent spec- ification for cosmetic talc [12] combines two methods (XRD and optical microscopy) for monitoring their types of talc. The rationale is that a talc is first examined by XRD, and if even the smallest amount of amphibole is indicated, then the test proceeds into optical microscopy using a dispersion staining technique to determine whether or not the material contains asbestiform particles in the amphibole group. Summary This paper has categorized the main methods which have been used for detection of asbestos in talcs. The basic principles of the various methods were categorized to explain how asbestos has been and can be misidentified in talc. Generally, misidentifications arise by jumping to a conclusion from a single mineral characteristic, when, in fact, many characteristics are required to fully identify a mineral species and/or its variety. Both optical microscopy and XRD required a more detailed review than other methods since they have received the most attention from a monitoring point of view. This review is presented with the hope that our guidelines will enable analysts to avoid the misidentification of asbestos in talcs. References [1] Ampian, S. C., Asbestos minerals and their nonasbestos analogs. Mineral Fibers Session, Electron Microscopy of Microfibers Symposium, Penn State Univ., August, 1976. [2] Thompson, C. S., Discussion of the mineralogy of industrial talcs. U.S. Bureau of Mines Information Circular 8639, Proceedings of the Symposium on Talc, Washington, D.C., May 8, 1973. [3] National Bureau of Standards Staff, A report on the fiber content of eighty industrial talc samples obtained from, and using the procedures of, the Occupational Safety and Health Administration, 51 pp. (1977). [4] Bowndy, M. G., Gold, K., Burgers, W. A., and Dement, J. M., Exposure to industrial talc in Vermont talc mines and mills: AIHA Conference presentation, May 1977. [5] Rohl, A. N., Langer, A. M., Selikoff, I. J., Tordini, A., Klimentidis, R., Bowes, D. R., and Skinner, D. L. , Consumer talcums and powders: mineral and chemical characteriza- tion, Jour. of Toxicology and Environmental Health, 2, 255-284 (1976). [6] Rohl, A. N. and Langer, A. M., Identification and quantification of asbestos in talc, Environmental Health Perspectives, 9, 95-109 (1974). [7] Snider, D. W., Pfeiffer, D. E., and Mancuso, J. J., Asbestos form impurities in commercial talcum powders, Compass of Sigma Gamma Epsilon, 49, 65-67 (1972). [8] Scholl, R. and Drafts, R., 1977, XRD characterization of asbestiform reference minerals. Symposium on Electron Microscopy and X-Ray Applications to Environmental and Occupational Health Analyses, April 1977. 351 2063105144

0-9 [9] Jenkins, R., A review of x-ray diffraction procedures as related to the quantitative analysis of air particulates. Symposium on Electron Microscopy and X-Ray Applications to Environmental and Occupational Health Analyses, April 1977. [10] Zoltai, T. and Stout; J. H., Comments on asbestiform and fibrous mineral fragments, relative to Reserve Mining Company taconite deposits: Report to Minnesota Pollution Control Agency, 89 pp. (1976). [11] Lee, R., Electron optical identification of particulates: Symposium on Electron Microscopy and X-Ray Applications to Environmental and Occupational Health Analyses, April 1977. [12] CTFA Specification - COSMETIC TALC Issued 10/7/76, The Cosmetic, Toiletry and Fragrance Association, Inc. Discussion A. WILEY: You said that instantaneous recognition of SAD patterns is difficult. Could you give some examples as to what kind of confusions could exist in this? Can you confuse amphibole with serpentine or amphibole with talc, or is that kind of a gross mistake possible? J. KRAUSE: Those kinds of mistakes probably would not generally happen if you are looking at pyroxenes or olivine. Electron diffraction is not one of my areas of real expertise, but I think that you could possibly get feldspars that would give confusing patterns, depending upon their orientation in the microscope. L. MADSEN: We are using all the methods that have been talked about today for identi- fication for asbestos materials and do not in any way limit ourselves to fiber length and aspect ratios. J. WAGMAN: I would like to comment that it is possible by x-ray diffraction and through a special technique to identify and measure the presence of asbestos fibers even when they are in the presence of their non-fibrous counterparts. About two years ago this was demonstrated in a study which we supported at the Naval Research Laboratory in which samples were pre-treated so that fibers were first aligned and then the x-ray diffraction intensities measured at two different orientations with respect to the x-ray beam and in this way the intensity due to the non-fibrous counterparts could be subtracted from the total diffraction intensities. KRAUSE: You were putting the fibers in some specific preferred orientation in the sample and then looking for those orientations by XRD. WAGMAN: That is correct, and this had the advantage of not only making possible corrections, that is correcting for the non-fibrous material present, but also it greatly enhances the detectability for the fibers themselves. KRAUSE: Is this method being currently used? WAGMAN: This is a method whose feasibility was demonstrated and there are two publica- tions on this in the literature. Actually our objective was to apply this method to airborne samples, which is a much more difficult application incidently, I should think than in the case of talc. The problem here is a preparative problem in that an air sample usually has a lot of organic material, sticky material present which interferes with the ability to orient the fibers. This is a preparative problem which will have to be overcome. But I should think that in the case of talc samples you probably would not have that problem. 352

K. HEINRICH: Would the talc plates interfere just as well with the orientation of the fibers? WAGMAN: The orientation of the fibers is accooplished in an electric field, and the platy material does not preferentially orient itself. 'HEINRICH: I mean, just in the sense of a passive restraint to the movement of the fibers. WAGMAN: This of course would have to be tested experimentally. A. LANGER: We heard today from a representative of one of the member organizations of the Cosmetic, Fragrance, and Toiletry Associations, that of 3800 consumer talcs examined none contained chrysotile. Today you presented some interesting information on the identification of crocidolite in talc. Have you seen crocidolite in many talcs you have examined? . KRAUSE: No I have not seen it, nor did I say that I have. LANGER: It does not occur in consumer talcs, or is it industrial talc. I just do not see why the crocidolite issue was raised; have you seen it? KRAUSE: Just because I have not seen it certainly does not mean that it could not conceivably exist. All I was trying to do was point out that choosing a specific amphibole peak as being representative and definitive for giving a good identification of a particular amphibole species has great potential for error. There are many, many other minerals that could fall within that same two theta region. LANGER: I would agree with you that even though talcs occur in nature and they have great mineralogical variability they are still bound by the physical and chemical laws involving calcium-silicate rock systems. A mineral phase such as you described would not occur normally. N O 353 W ~ 0 ~ t

National Bureau of Standards Special Publication 506. Proceedings of the Workshop on Asbestos: Definitions and Measurement Methods held at NBS, Gaithersburg, MD, July 18-20, 1977. (Issued November 1978) AMBIENT AIR MONITORING FOR CHRYSOTILE IN THE UNITED STATES Richard J. Thompson Analytical Chemistry Branch Environmental Monitoring Support Laboratory Environmental Protection Agency Research Triangle Park, North Carolina 27711 Abstract The only continuing national air monitoring has been conducted by the National Air Surveillance Network. The objective is long term trend assessment of air quality. The information has proven of value in setting standards, in consideration of health effects, in estimation of economic effects, and in showing patterns of pollutant distribution in both urban and non-urban areas. In order to provide samples which could be analyzed for constituents not determinable in particulate matter samples collected with glass- fiber filters, a membrane sampling network was instituted. The only analyses of the samples conducted thus far has been for airborne asbestos using in part a method developed under contract which provides for the determination of the mass of chrysotile in the particulate samples. A viewpoint will be presented on the method needed for air monitoring and an assessment of the mass method as the most suitable for this purpose. Data obtained will be examined which will include information on inter- and intra-laboratory replication. Key Words: Air monitoring; airborne particulate; asbestos; chrysotile; filters. National air monitoring had its inception in a Public Health Service survey of protein in airborne particulate matter conducted at seventeen sites in 1953-54. Sufficient amounts of material were collected using glass fiber filters to permit chemical analyses as well as the determination of total suspended particulate matter. In 1955 the Federal Air Pollution Research and Technical Assistance Act, Public Law 159, 84th Congress, was passed. The Network was expanded to 66 station