Mineral Properties and Their Contributionsto Particle ToxicityGeorge D. Guthrie Jr.Geology and Geochemistry Group, Los Alamos National Laboratory, LosAlamos, New Mexico

It has been recognized since at least as early as the mid-1500s that inhaled minerals (i.e.,inorganic particles) can pose a risk. Extensive research has focused on the biological mechanismsresponsible for asbestos- and silica-induced diseases, but much less attention has been paid tothe mineralogical properties and geochemical mechanisms that might influence a mineral'sbiological activity. Several important mineralogical characteristics control a mineral's reactivity ingeochemical reactions and are likely to determine its biological reactivity. In addition to thetraditionally considered variables of particle size and shape, mineralogical characteristics such asdissolution behavior, ion exchange, sorptive properties, and the nature of the mineral surface(e.g., surface reactivity) play important roles in determining the toxicity and carcinogenicity of aparticle. Ultimately, a mineral's species (which provides direct information on a mineral's structureand composition) is probably one of the most significant yet most neglected factors that must beconsidered in studies of toxicity and carcinogenicity. Environ Health Perspect 105(Suppl 5):1003-1011 (1997)

Key words: particles, minerals, toxicity, carcinogenicity, mechanisms

Introduction

In his 1556 treatise De Re Metallica,Georgius Agricola noted that minersexposed to dust from some mines hadincreased risks for various diseases, includ-ing consumption (1). Although it is notclear from Agricola's description whetheror not consumption refers specifically totuberculosis or more generally to pneumo-coniosis, it is interesting to note that thelinks both between dust exposure andtuberculosis and between dust exposureand pneumoconiosis were borne out inlater studies, induding work 400 years laterby King and co-workers (2-8). By the timeof King and co-workers, it was generally

This paper is based on a presentation at The SixthInternational Meeting on the Toxicology of Naturaland Man-Made Fibrous and Non-Fibrous Particlesheld 15-18 September 1996 in Lake Placid, NewYork. Manuscript received at EHP 26 March 1997;accepted 20 May 1997.

thank Eugene llton for permission to use Figure4; Art Langer for reminding me of Nagelschmidt'sbackground as a clay mineralogist; and Bill Carey andtwo anonymous reviewers for critical reviews of themanuscript. This work was supported by theDepartment of Energy through contract W-7405-ENG-36 to Los Alamos National Laboratory.

Address correspondence to Dr. G. Guthrie, 1187Rood Hall, Dept. of Geology, Western MichiganUniversity, Kalamazoo, Ml 49008. Telephone:(616) 387-5343. Fax: (616) 387-5513. E-mail:guthrie@wmich.edu

recognized that inhaled minerals can initiatea number of responses, including theformation of ferruginous bodies (9-12),fibrosis (3,4,6-8,13-17), and shortlythereafter carcinogenesis (18-21).

Even (perhaps especially) at these earlystages of research on mineral-inducedpathogenesis, it was recognized that the keyto understanding why some minerals aretoxic or carcinogenic is to link mineralogicalproperties with biological processes. Muchof this insight came from the work by Kingand co-workers, through their collabora-tions with G. Nagelschmidt, a mineralogist,and to a lesser extent V.M. Goldschmidt,who is considered by many to be thefounder of modern geochemistry. Throughthese collaborations, King and co-workersinvestigated the biological activity of a list ofminerals that is truly impressive: olivine(22), kaolin minerals (4,6,14), micas (2),various silica polymorphs (induding quartz,tridymite, cristobalite, and amorphoussilica) (5,7,23-25), various forms of alu-minum and iron oxides and hydroxides(including boehmite or y-AIOOH, corun-dum, or a-AI203-which is isostructuralwith hematite, y-A1203, goethite, ora-FeOOH, and lepidocrocite ory-FeOOH)(3,8), and berlinite or AIPO4 (8).

As demonstrated by their use ofmineral species names to describe their

materials, King and co-workers must haverecognized that the structure and composi-tion of a material (the two characteristicsthat define a mineral species*) are criticalto determining the way in which a materialinteracts with its environment. This is afundamental principle in the geosciences,where it has long been recognized that eachmineral species possesses unique properties(derived from its crystal structure and/orcomposition) and that these propertiesdetermine how a mineral interacts withits environment.

In this paper, I address a number ofmineralogical properties that affect how amineral interacts with its environment.Some of these properties have been shownto affect toxicity and carcinogenicity, andsome are known to be important in geolog-ical processes but have not been exploredwith respect to biological processes. Anunderlying principle throughout this paperis that pathogenesis originates at the min-eral-fluid-cell interface, so interactionsbetween a mineral and fluid or a mineraland a cell may ultimately lead to disease.These interactions range from indirectinteractions between a mineral surface andextracellular or intracellular fluids (includ-ing fluids associated with phagosomesand/or lysosomes) to direct interactionsbetween a mineral surf&ce and cell-surfacereceptors or other components of a cell'smembrane. To gain insight into what min-eralogical properties are important for amineral's role, we can borrow from thegeosciences where a large range of min-eral-fluid interactions have been and con-tinue to be studied. I discuss briefly severalproperties that are commonly addressed-particle size/shape, mineral species (struc-ture/composition), dissolution, and surfaces.I also discuss two properties that are seldomaddressed-cation exchange and oxida-tion/reduction. All these properties areknown to affect the way in which a min-eral interacts with a geological fluid andare likely to play roles in mineral-fluidinteractions in the lung.

*Mineral species are applied in much the same wayas animal/plant species: A mineral species is themost specific distinct division within the classificationscheme for minerals. It defines a specific crystalstructure and a composition or compositional range.Sometimes subspecies (termed varieties) are definedbased on characterstics such as morphology or crys-tal habit (e.g., crocidolite is the varietal term forasbestiform riebeckite).

Environmental Health Perspectives a Vol 105, Supplement 5 * September 1997 1 003

G.D. GUTHRIE JR.

Mineralogical PropertiesImportant in ToxicitySolids can be divided into two broadcategories based on the property of transla-tional periodicity: crystalline and noncrys-talline (or amorphous). Translationalperiodicity is the characteristic that allowsthe extended structure existing throughouta single-phase particle to be represented bya smaller subunit that is translated alongnon-coplanar vectors in three dimensions,in the same way that a wall might be repre-sented by a brick or cinderblock that istranslated in two dimensions. Translationalperiodicity is necessary for a structure todiffract X-rays constructively (which is whyproteins must be crystallized for structuralanalysis), but it also imparts a wide rangeof properties to minerals that differentiatethem from amorphous materials. In fact,crystallinity appears to be an importantfactor in toxicity/carcinogenicity, as exem-plified by the higher biological reactivity ofsome types of crystalline silica comparedwith noncrystalline silica (26).

Much of our insight into how a mineralinteracts with a fluid comes from observa-tions on geological systems. Reactionsinvolving minerals in geological environ-ments are often mediated by fluids; conse-quently, mineral-fluid interactions areamong the most widely studied phenom-ena in mineralogy, geochemistry, and otherbranches of the geosciences. Our currentunderstanding of these phenomena has ledto several important geological tenets thatare equally important in mineral toxicity/carcinogenicity:* a mineral affects its environment;* a mineral is affected by its environment;* mineral species is a critical descriptor of

a material's overall characteristics;* the properties of a mineral species can

vary between samples.Although these tenets may appear self-

evident (as they undoubtedly did to King,Nagelschmidt, Goldschmidt, and their co-workers), they serve as important remindersthat mineralogy is an integral component ofmineral toxicity. As such, care must betaken to ensure that mineralogical issues ina study are as adequately addressed as bio-logical issues. However, it also means thatthe mineralogical approaches to toxicitystudies can provide important new insightsinto the molecular processes that occur. Forexample, the first tenet embodies thenotion that a mineral can induce reactionsin a cell or physiological fluid. Hence, byappropriate manipulations of mineralogical

variables, one can alter a biological responsein a systematic way to reveal the mineralog-ical and biological molecular mechanisms(in much the same way that one manipu-lates the biological variables to reveal mech-anisms). The second tenet embodies thenotion that the way a mineral is changed bya reaction preserves information about thereaction. Hence, one can learn somethingabout a mineral-fluid-cell interaction byobserving not only how the cell and fluidrespond but also by observing how themineral responds.

These principles form the motivationfor the remainder of this paper, which willattempt both to address some of the miner-alogical properties important in toxicityand to illustrate how a combined miner-alogical and biological approach canimprove our understanding of these com-plex processes. The first three topics (min-eral species, particle size and shape, andsample history) cover properties that ingeneral should be determined for everysample studied. The remaining four topics(ion exchange, oxidation/reduction, disso-lution, and surfaces) are additional miner-alogical factors that are key components ofa mechanistic model for mineral-inducedpathogenesis. I focus primarily on the firsttwo topics because the last two topics arecovered extensively elsewhere (includingother articles in this volume that addresspartide surfaces).

Mineral Speces: Structureand CompositionA determination of the mineral speciesused in a specific study is the bare mini-mum required in terms of sample charac-terization. Although mineral species oftentakes a back seat to particle size and shape(discussed below) in most studies on toxic-ity and carcinogenicity, it is, in fact, one ofthe most critical characteristics because itprovides information on the bulk structureand composition of a material-the twomost basic characteristics of a material thatultimately have a profound effect on manyof the other mineralogical propertiesimportant in pathogenesis.

Both the structure and composition areneeded to define a mineral species becauseneither alone is sufficient to describe theproperties of a material; this is well illus-trated by the minerals quartz, stishovite, andrutile. Figure 1 shows polyhedral representa-tions * for the structures of these materials.Quartz and stishovite are compositionallyidentical (SiO2) but have markedly differentstructures (Figure 1A-D). This structural

difference imparts different solubilities(important in biodurability and possibly tox-icity), different functional groups on the sur-face (related to different bonding strengthsfor various surface oxygen sites, which trans-lates, for example, into different dissociationconstants for protons on the surface), anddifferent tolerances for various trace elemen-tal contaminants (to name a few differencesimportant in toxicity). Stishovite and rutileare structurally identical (Figure 1B,D) buthave different compositions (SiO2 andTiO2, respectively). This compositional dif-ference affects solubilities, surface functionalgroups, and bulk oxidation-reduction char-acteristics, to name a few. (Interestingly, sti-shovite and rutile are both nonfibrogenicand noncarcinogenic, suggesting that thisstructure type may not elicit a pathogenicresponse. Although this observation has beenalluded to often, no one has tested thismechanistic hypothesis by investigating sys-tematically the many other materials withthis structure such as pyrolusite or MnO2,cassiterite or SnO2, argutite or GeO2, andmarcasite or FeS2.)

The asbestos literature has evolved to thepoint where the use of terms like crocidolite(the asbestiform variety of the mineralspecies riebeckite), amosite (a commercialterm mostly referring to the asbestiformvariety of the mineral series cumming-tonite-grunerite), asbestiform tremolite, andchrysotile is commonplace. Similarly,studies on the oxides of silicon (typicallySiO2 or silica) often use proper mineralspecies terms, like quartz, tridymite, andcristobalite. Nevertheless, since the days ofKing and co-workers (who were faithful tothe use of mineral species names for silicapolymorphs as well as for a vast array ofother minerals in their studies), the use ofterminology has become much less rigorous.Now the use of the nondescript term silica isnot uncommon and the use of chemicalterms (and not mineralogical terms) forother materials is the norm-e.g., studies onthe oxides of titanium almost always use

*Most rock-forming minerals (i.e., minerals that arecommon in rocks found at Earth's surface) arelargely composed of oxygen coordinating cations in asmall set of regular shapes (termed polyhedra).Hence, portions of mineral structures are commonlysimplified by replacing some of the atoms with poly-hedra (26-28) for discussions on the graphical repre-sentations of mineral structures), an approachpioneered by Linus Pauling (29), who began his sci-entific career as a mineralogist. More detailed pre-sentations of minerals and their structures can befound in a general mineralogy text such as theManual of Mineralogy (30).

Environmental Health Perspectives * Vol 105, Supplement 5 * September 19971 004

MINERALOGICAL PROPERTIES AND PARTICLE TOXICITY

B

D

Figure 1. (A) Polyhedral representation of the left-handed quartz structure (94) projected nearly down the c-axis.Each tetrahedron consists of Si4+ at the center and 02- at each of the four apices. (B) An isolated double helix fromthe quartz structure. (C,D) Polyhedral representation of the stishovite and rutile structure (95) viewed down the b-axis. Each octahedron consists of a metal ion (Si4+ for stishovite and Ti4+ for rutile) at the center and 02- at eachof the six apices.

the term titanium dioxide to describe amaterial. This lax approach to sampledescription can lead to a false sense of con-fidence in the ability to interpret resultsfrom various studies. For example, one isled to believe that the results for titaniumdioxide in one study can be compared withresults for titanium dioxide in anotherstudy. In fact, TiO2 or titania crystallizes inat least seven different polymorphs (i.e.,different structures), including rutile(Figure 1C,D), anatase, brookite, andTiO2 (B) (31) for figures of these last threepolymorphs). In addition, titanium oxideswith stoichiometries different from TiO2occur (i.e., where the oxidation state of Tiis not uniformly Ti4+). Each of these formsof titanium oxide has different properties(31). For example, anatase is used as a cat-alyst (32,33) and photocatalytic (34-36).Differences in biological activities have alsobeen noted for the polymorphs of TiO2(37-39). Nevertheless, it is commonplaceto read that a particular study used TiO2 asa negative control. Clearly, in the absenceof information on the crystal structure of a

material, the use of a chemical term such astitania or titanium dioxide is inadequateinformation to provide for a sample usedin a toxicity study.

Although mineral species is one of themost critical characteristics to be deter-mined for a sample used in a toxicity, thereare cases for which the use of a mineralspecies name (which defines the ideal com-position and bulk structure) is insufficientinformation for describing a sample. Withrespect to composition, this can occurwhen the mineral species is defined for arange in composition or when the compo-sition of the sample deviates from the idealstoichiometry. The first case is well illus-trated by the asbestiform amphiboles. Forexample, asbestiform riebeckite (or croci-dolite) has an ideal end-member composi-tion of Na2Fe3+Fe2+ Si8O22(OH)2, but themineral species riebeckite is actuallydefined over a much broader range of com-position, which allows for a) potassiumand sodium to partially occupy the "A"site, which is omitted in the ideal formula;b) up to 50% replacement of the iron by

magnesium; and c) limited other substitu-tions for sodium, iron, silicon, andhydroxyl (40). These compositional varia-tions can have profound affects on thesample's properties, including toxicity. Forexample, one can easily imagine that a sig-nificant replacement of iron by magnesiumwould have an impact on the particle'sability to drive the Fenton-type reactionsthat are currently believed to explain croci-dolite's extreme biological activity (41,42).Hence, a compositional analysis must beprovided for the particular crocidolite sam-ple used in a toxicity study. The secondcase can occur even for well behaved min-erals like quartz, which can have up to afew wt-% of elements like Al and Fe (28).These minor and trace elements can have asignificant impact on the biological reactivityof quartz (43).

In some cases, mineral speciesinadequately describes a material's struc-ture because the presence of defects (whichare deviations from the ideal structure)introduce significant variation into thematerial's properties. For example, ostensi-bly non asbestiform riebeckite and asbesti-form riebeckite (crocidolite) share the samestructure. However, samples of crocidolitegenerally have a large proportion of chain-width defects, which imparts uniquemechanical properties on crocidolite (i.e.,crocidolite is flexible whereas nonasbesti-form riebeckite is not) (27). These defectscan additionally impart other differences inthe properties between the two materials(e.g., the diffusion of cations within andthe dissolution properties of amphibole areaffected by chain-width defects, both ofwhich will affect the release of iron to thefluid). Hence, the differences in biologicalactivities noted for these materials (44,45)cannot be uniquely attributed to particlemorphology, as is often done.

Partide Size and ShapeParticle size and shape are universallyconsidered important factors in pathogene-sis and are faithfully reported in moststudies. There are several understood (orpartially understood) mechanisms bywhich size and shape may influence toxic-ity and carcinogenicity, including fate ofthe particle (from deposition to physicaltranslocation to cell-mediated transloca-tion), surface area, types of reactive sites,particle-cell interactions, and catalysis.

Particle size and shape exert a majorcontrol on deposition, translocation, andclearance (i.e., the fate of a particle followinginhalation). Deposition is affected by a

Environmental Health Perspectives * Vol 105, Supplement 5 * September 1997

A

C

1005

G.D. GUTHRIE JR.

combination of physical limitationsimposed by the constricting airways-which reduce to approximately 50 pm bythe time they reach the alveolar ducts(46)-and aerodynamic and gravitationalfactors-which control processes such asimpaction, settling, interception, and diffu-sion (47). These processes lead to heteroge-neous particle-size deposition throughoutthe conducting airways and lungs. Forexample, larger particles (>0.2 pm) aredominantly deposited in the nasopharyn-geal region (47), whereas smaller particlesare deposited in the respiratory tract. Thistranslates into approximately 10 pm as aneffective maximum size for respirable parti-cles in humans and approximately 5 pm asan effective maximum size for respirableparticles in rats (48). Translocation (partic-ularly from the airways through theparenchyma to the pleura) is also affectedby particle size and shape as demonstratedby the observation that fibrous particles arecommonly found in the pleural space.Finally, clearance mechanisms-e.g., disso-lution rate, which is an important clearancemechanism for rapidly dissolving materialslike chrysotile (49), and cellular clearancemechanisms such as phagocytosis andtranslocation-are strongly limited byparticle size and shape.

Size and shape also determine thesurface area of a particle and, perhaps moreimportantly, the surface area per unit vol-ume or per unit mass of the sample. Particlevolume (and hence mass) scales with thecube of a particle size, whereas the surfacearea of a smooth particle scales with thesquare of particle size. In other words, smallparticles have larger surface areas per unitmass than larger particles, which means thatsmaller particles have more reactive surfaceavailable on a per-mass basis. Consequently,a number of researchers have argued infavor of comparing toxicity of materials ona per-fiber (or per-particle) basis rather thanon a per-mass basis (50). "Per surface area"is probably a more defensible basis onwhich to compare results, but only a fewstudies have endorsed this approach (51).

Another important aspect of particlemorphology relates to the nature of reactivesites on the particle surface (as discussedbelow), because a particle's morphologydetermines the exposure of various reactivesites. For example, the active sites associ-ated with the ends of crocidolite fibers(which differ dramatically from the activesites associated with the sides of the fibers)have lower exposed areas than they wouldin a case where the crocidolite formed

Figure 2. Schematic representation of a particle with afibrous versus platy morphology. If the fiber were a cro-cidolite fiber 0.1 pm x 0.1 pm x 0.1 pm (i.e., 0.1 pm3),the fiber ends-the shaded end that correspondsapproximately to the (001) plane-would have a sur-face area of 2 x 0.01 pm2; whereas a hypotheticalplaty crocidolite of the same volume (1 pm x 1 pm x0.1 pm) and with the plates parallel to (001) wouldhave a (001) surface area (shaded) of 2 x 1 pm2, i.e.,two orders of magnitude greater. Hence, particle mor-phology impacts the exposed surface areas of a material.

sheet-like particles normal to the fibers(Figure 2).

Recent work presented at this confer-ence suggests that fiber length may affectparticle-cell interactions by causingmechanical stresses on the cell surface (52).Mijailovich and co-workers hypothesizethat the contact of a long fiber with alveo-lar epithelium that is in cyclic motionbecause of tidal breathing causes stresses onthe cell that trigger a response.

Finally, a number of materials becomeeffective catalysts when their particle sizebecomes extremely small (i.e.,

MINERALOGICAL PROPERTIES AND PARTICLE TOXICITY

in the colony-forming efficiency of ratlung epithelial cells exposed to a suite ofcation-exchanged erionites (Na, K, Ca, andFe3+) all derived from the same parentmaterial. However, using the same suite oferionite samples, we have found that cationexchange may have an effect on cytotoxic-ity, gene response (as measured by steady-state levels of mRNA for c-jun), andapoptosis in some rat pleural mesothelialcells (Guthrie G, Timblin C, Mossman BT,unpublished data). For example, prelimi-nary results suggest that Na- and K-exchanged erionite may be more cytotoxicat 72 hr than Ca- and Fe-exchanged erion-ite (Figure 3). Initially, one might condudethat cation exchange, which happens rapidlyin simple aqueous solutions, would only besignificant immediately after exposure.

A

- - - - 8-hr exposure72-hr exposure

B

co

0.8 -

0.6 -

Dose, jg/cm2

Figure 3. Effect of cation exchange on (A) Na-exchanged erionite; (B) Ca-exchanged erionite] erionitecytotoxicity on rat pleural mesothelial cells (Guthrie,unpublished data). Erionite from Eastgate Nevada wascation exchanged in chloride solutions and washedthoroughly to remove remaining salt. Details of theexperiments are in Timblin et al. (unpublished data).

However, the kinetics of cation exchangein a complex biological fluid which con-tains molecules that could inhibit exchangeby interfering with ion channels on themineral surface are not known.

CatalysisMinerals-particularly but not exclusivelyzeolites and some clays-are exploited exten-sively as catalysts. The mechanisms by whichminerals function as catalysts generally relateto their ability to donate or accept electronsor protons, to provide a stabilizing surface(a template) for reacting components, andto exclude molecules of a specific shape orsize from the catalytic sites. In other words,minerals can function in a manner similarto that of traditional enzymes.

The proton and electron donor/acceptorsites (the acid/base sites) of the mineral arecommonly exploited and are responsible forthe widespread use of zeolites as catalysts inthe cracking of hydrocarbons. The acid/base characteristics of framework silicatessuch as zeolites are strongly influenced bythe substitution of aluminum for silicon inthe tetrahedral framework. This substitu-tion can be charge compensated for in anumber of ways, including the associationof a proton with the underbonded oxygensaround the aluminum. Hence, these sites(like many other surface sites on minerals)function in a manner similar to that ofenzymes in that they can alter the apparentlocal activity (or thermodynamic concentra-tion) of a species like hydrogen at a specificreactive site. A number of different suchsites can exist on the surface of zeolite, andrange from silanol groups (Si-O-H) to thealuminum equivalent (Al-O-H) to protonsgenerally associated with an aluminum-exchanged tetrahedral site, e.g.,

H

AI-O-Si

The protons associated with each of thesesites have different pK values. There recentlyhas been much success in applying ab initiocalculation methods to calculate the reactiv-ity associated with such surface sites (72).

It has long been argued that hydro-genated surface sites (e.g., silanol groups)are responsible for the toxicities of the sil-ica polymorphs because they function ashydrogen donors. In support of this is thefact that the hemolytic reactivity of quartzcan be diminished (73) by treatment withpolyvinyl-pyridine-N-oxide (PVPNO), apolymer that binds to proton donor sites.

Oxidation-ReductionThe transfer of electrons between a mineraland fluid drives a number of geochemicalprocesses. In general, silicates and manyother minerals are considered to be insula-tors (i.e., they do not conduct electronsrapidly). At higher temperatures (hundredsof degrees Celsius), some silicates begin toconduct electrons sufficiently rapidly toallow their electrical properties to be stud-ied somewhat routinely. The oxidation/reduction properties of the amphiboleasbestos minerals crocidolite and amositehave been studied extensively at high tem-peratures, beginning with the pioneeringstudies ofAddison and co-workers (74-76).Although they focused on higher tempera-tures (450-6150C), they studied the kinet-ics of the reaction down to 350°C andnoted that oxidation can occur even at0°C. At lower, physiological temperatures,the rates may be too low to measure effec-tively in a laboratory experiment, but theymay be sufficiently high to provide achronic source (or sink) of electrons forreduction (or oxidation) of fluid species(e.g., to form free radicals).

Interestingly-and predictably based onmineralogy-the resistance of amphibolesvaries strongly with crystallographic direc-tion. Electron conduction occurs mostrapidly along the length of the fibers (i.e.,along the octahedral strips that containiron). Crystallographically similar electronconduction pathways occur within the octa-hedral sheets of phyllosilicates like biotite(i.e., at the edges of the sheets readilyformed by the dominant cleavage directionin micas), which is why micas are effectiveinsulators normal to their sheets (they areexploited this way in capacitors) but havemuch lower resistance along their sheets.

Although most investigations of oxida-tion-reduction and conduction in silicateshave focused on high temperatures, theseprocesses are known to be important for anumber of reactions at lower temperatures(e.g., < 50°C). For example, electron transferreactions have been shown to be importantin the weathering at ambient temperaturesof minerals such as amphibole (77) andmagnetite (78), in the formation of copperore deposits (79), and in the sorption ofmetals such as Cr to the mineral surface at250C (80). Clearly, the transfer of elec-trons between minerals and fluids is impor-tant at physiological temperatures and, asshown in Figure 4, this redox process isstrongly controlled by the crystal structure.

The propensity for a redox reaction tooccur can be assessed by comparing the EH

Environmental Health Perspectives * Vol 105, Supplement 5 * September 1997

I

1007

G.D. GUTHRIE JR.

Figure 4. A secondary electron image of a biotite crystal that has interacted with a fluid containing silver sulfatesuch that electrons have been transferred from the mineral to the Ag+ in the solution. Crystals formed at 25°C overthe course of several days. Note that the silver crystals form at the edge of the biotite crystals where the octahe-dral sheets surface. Photograph used with permission of Eugene Ilton (Lehigh University).

or pe values for the individual half reactions,which ultimately relate to the electrochemi-cal potentials (EO) for the half reactions.Electrochemical potentials for aqueous reac-tions can be readily determined and havebeen tabulated for a number of half reac-tions. For example, [Fe3+]aq + e -* [Fe2+]aqhas an E° of 0.77 V (81). Unfortunately,the electrochemical potential for variousredox reactions in minerals are poorlyknown. White and Yee (77) bracketed theelectrochemical potential for Fe3+o-Fe2+ ina hornblende amphibole at between 0.33and 0.52 V under their experimental con-ditions. White and co-workers (78) deter-mined the electrochemical potential forFe3+o-Fe2+ in a magnetite sample (-0.27 Vat pH 7) directly by measuring the self-induced potential of a magnetite electrode(a procedure that requires a large singlecrystal). Ilton [(79,82); personal communi-cation] has found that the electrochemicalpotential for Fe3+->Fe2+ in biotite is proba-bly close to the values determined byWhite and Yee for hornblende, based onthe observation that Ag+ reduces vigorously(E = 0.80 V; even at a concentration of 10ppm), whereas Cu2+ (EO = 0.34 V) reducesbut much less vigorously.

The mechanisms by which a mineralcan transfer electrons from within the crys-tal to the surface also must be known toevaluate mineral-catalyzed redox reactionsin a fluid. An important component of themineral-catalyzed redox is that the crystalmust remain charge balanced (or at least

close to neutral charge). Hence, for theoxidation of crystal-bound iron to reduce asolution-bound species, the reaction can bebroken down into several steps:

(Fe2+)crystal -e (Fe3+)crystal + (e)surface [1]

(OH-)crystal -e (02-)crystal + (H+)surface [2a]

(R+)crystal -e (R+)surface [2b]

(02-)surface + ELcrystal -* (02-)crystal [2c]

where O designates a vacancy or unoccu-pied crystallographic site and R+ designatesa cation site. The reactions are written withrespect to changes within the mineral andwhere the surface represents the interfacebetween the mineral and the fluid (e.g., anelectron at the surface can transfer to thefluid). The electron-exchange reaction(Equation 1) is written involving iron,because this is the most common polyva-lent cation in minerals. Equation 2 sum-marizes three possible mechanisms formaintaining charge balance within thecrystal (74). Equations 2a and 2b maintaincharge balance by diffusing a chargedspecies out of the crystal. Hydrogen,because of its small size, would be the easi-est of the possible cations to diffuse out ofminerals such as amphiboles and, in fact,

Addison and co-workers (74) proposed thismechanism for crocidolite oxidized at 450to 615°C. Scott and Amonette (83)endorsed a slightly different mechanism inweathering conditions (i.e., low tempera-tures) whereby charge balance is maintainedby dissolution of iron after oxidation(Equation 2b). In some materials (e.g.,magnetite or Fe3O4) diffusion of hydrogenis not an option, so mineralogical changesoccur. For magnetite, oxidation occursthrough the formation of maghemite (78).The mechanism by which a material oxi-dizes has a profound effect on the rate ofoxidation (i.e., on the rate of sustainedelectron transfer).

How, then, might electron transferprocesses play a role in pathogenesis?Electron transfer involving the surfaceof minerals has been attributed to theincreased biological activity and heightenedformation of free radicals associated withfreshly fractured quartz compared to agedquartz (54,84,85). Such a process producesa transient or acute burst in free radicalsthat ceases once the particle surface hasbeen passivated (i.e., once the surface radi-cals have equilibrated). Electron transferinvolving the internal regions of the crystal(through transfer between the surface andinterior) has the potential to produce a sus-tained or chronic redox condition to driveformation of radicals in the fluid. In addi-tion, once electron transfer to the fluid hasoccurred, iron release to the fluid (to main-tain charge balance) could provide anothermechanism for driving Fenton-type reac-tions. Several lines of evidence support thenotion that electron transfer processes areimportant in pathogenesis. Fubini and co-workers (86) reported that magnetite willbreakdown hydrogen peroxide, whereashemnatite (Fe2O3) will not, which suggeststhat Equation 1 plays an important role inthe formation of free radicals. Figure 4shows a mica crystal that has reduced sil-ver from solution to cause its precipitationat the redox-active edges. Similarly, in hisdescriptions of ferruginous bodies, Roggli(87) shows several particles of mica recov-ered from human lung (his Figures 3-18);the particles have become coated with aferruginous material only at the redox-active edges of the crystals. Although themechanism of ferruginous body formationis still not understood in its entirety, it isbelieved to relate to the breakdown of aniron protein such as ferritin (which can bedenatured by a redox mechanism). It isinteresting to note that asbestos bodiestypically have more precipitate at their

Environmental Health Perspectives a Vol 105, Supplement 5 * September 19971 008

MINERALOGICAL PROPERTIES AND PARTICLE TOXICllY

ends, which are crystallographicallysimilar to the edges of mica and which areknown to be the redox-active areas athigher temperatures.

Dissolution BehaviorDissolution can be a significant componentof particle clearance mechanisms and cancause the release to the lung fluid of ionssuch as iron, other metals, or other toxicelements. Dissolution is often used as abasis for differentiating nonhazardous frompotentially hazardous minerals, where non-hazardous minerals have a low biodurabil-ity and, hence, do not remain in the lungfor long periods of time.

Unfortunately, there are few data onthe kinetics of mineral dissolution in bio-logical fluids that are also based on mecha-nistic dissolution models for minerals. Onesuch study was recently conducted byHume and Rimstidt (49), who based theirmodel on the release of silica as the rate-limiting step for chrysotile dissolution.Their dissolution rate model predicts that

chrysotile fibers will dissolve completely indays to months under lunglike conditions.

It has long been recognized that mineralburdens in human lungs are not exactly rep-resentative of the dusts to which an individ-ual is exposed [reviewed by Churg (88)].For example, chrysotile miners typicallyhave a nearly steady-state level of chrysotilein their lungs but a continuously increasinglevel of tremolite (a minor contaminant ofthe chrysotile ore). These observations areconsistent with the rate model of Humeand Rimstidt, in that the steady-state levelsobserved represent competition betweendeposition and dissolution. Dissolution ofminerals under these conditions is discussedin detail by Hochella (89).

SurfacUltimately, the surface is that part of amineral that interacts with a fluid or cell.For some materials, the structure at thesurface can differ substantially from thestructure exhibited by the bulk (89). Thesedifferences between the surface and the

bulk can range from simple distortionalrelaxation of surface atoms to a completelydifferent material on the surface. Frequently,a dissolving mineral will form a precipitateat the surface with a composition/structurethat differs from the bulk material. Evenchemically simple minerals such as quartzcan have surfaces that are structurally dif-ferent from the bulk (90,91); at anotherextreme would be a fiber of amphiboleasbestos, which likely has much of its sur-face covered by a phyllosilicate-like materialthat is both compositionally and structurallydistinct from the bulk amphibole (27).Clearly, there is a large range of surface-related factors that can change the activesites on the surface, can affect binding/sorp-tion processes on the surface, can affect dis-solution characteristics, and can generallyhave an impact on a mineral's pathogenicpotential. A detailed discussion of surfacecharacteristics important in pathogenesis isbeyond the scope of this paper, but someof these aspects are addressed in Guthrie(26,28) and Hochella (89).

REFERENCES

1. Agricola G. De Re Metallica (translated from the first Latinedition, 1556). New York:Dover Publications, 1950.

2. King EJ, Gilchrist M, Rae MV. Tissue reaction to sericite andshale dusts treated with hydrochloric acid: an experimentalinvestigation on the lungs of rats. J Pathol Bacteriol59:325-327(1947).

3. King EJ, Harrison CV, Mohanty GP. The effect of variousforms of alumina on the lungs of rats. J Pathol Bacteriol69:81-93 (1955).

4. Attygalle D, Harrison CV, King EJ, Mohanty GP. Infectivepneumoconiosis. BrJ Ind Med 11:245-259 (1954).

5. Attygalle D, King EJ, Harrison CV, Nagelschmidt G. Theaction of variable amounts of tridymite, and of tridymite com-bined with coal, on the lungs of rats. Br J Ind Med 13:41-50(1956).

6. Hale LW, Gough J, King EJ, Nagelschmidt G. Pneumoconiosisof kaolin workers. BrJ Ind Med 13:251-259 (1956).

7. Englebrecht FM, Yoganathan M, King EJ, Nagelschmidt G.Fibrosis and collagen in rats' lungs produced y etched andunetched free silica dusts. AMA Arch Ind Health 17:287-294(1958).

8. Stacy BD, King EJ, Harrison CV. Tissue changes in rats' lungscaused by hydroxides, oxides, and phosphates of aluminiumand iron. J Pathol Bacteriol 77:417-426 (1959).

9. Cooke WE. Pulmonary asbestosis. Br Med J 2:1024-1025(1927).

10. McDonald S. Histology of pulmonary asbestosis. Br Med J2:1025-1026(1927).

11. Gloyne SR. The formation of the asbestosis body in the lung.Tubercle 12:399-401 (1931).

12. Gloyne SR, Leeds MD. The asbestos body. Lancet 1:1351-1355(1932).

13. Vorwald AJ, Durkan TM, Pratt PC. Experimental studies ofasbestosis. Arch Ind Hygiene Occup Med 3:1-43 (1951).

14. King EJ, Harrison CV. The effects of kaolin on the lungs ofrats. J Pathol Bacteriol 60:435-440 (1948).

15. Kettle EH. Experimental pneumoconiosis: infective silicatosis. JPathol Bacteriol 38:201-208 (1934).

16. Gardner LU, Dworski M, Delahant AB. Aluminum therapy insilicosis. J Ind Hygiene Toxicol 26:211-223 (1944).

17. Cooke WE. Fibrosis of the lungs due to inhalation of asbestosdust. Br Med J ii:147 (1924).

18. Sluis-Cremer GK Asbestosis in South Africa-certain geographi-cal and environmental considerations. Ann NY Acad Sci132:215-234 (1965).

19. Selikoff IJ, Churg J, Hammond EC. Relation between exposureto asbestos and mesothelioma. N EnglJ Med 272:560-565 (1965).

20. Selikoff IJ, Churg J, Hammond EC. Asbestos exposure andneoplasia. JAMA 188:22-26 (1964).

21. Wagner JC, Sleggs CA, Marchand P. Diffuse pleural mesothe-lioma and asbestos exposure in the north western CapeProvince. Br J Ind Med 17:260-271 (1960).

22. King EJ, Rogers N, Gilchrist M, Goldschmidt VM,Nagelschmidt G. The effect of olivine on the lungs of rats. JPatol Bacteriol 57:488-491 (1945).

23. Zaidi SH, King EJ, Harrison CV, Nagelschmidt G. Fibrogenicactivity of different forms of free silica. Arch Ind Health13:112-121 (1956).

24. King EJ, Mohanty GP, Harrison CV, Nagelschmmidt G. Theaction of different forms of pure silica on the lungs of rats. Br JInd Med 10:9-17 (1953).

25. King EJ, Mohanty GP, Harrison CV, Nagelschmmidt G. Theaction of flint of variable size injected at constant weight andconstant surface into the lungs of rats. Br J Ind Med 10:76-92(1953).

26. Guthrie GD Jr. Mineralogical factors affect the biological activ-ity of crystalline silica. Appl Occup Environ Hygiene10:1126-1131 (1995).

27. Veblen DR, Wylie AG. Mineralogy of amphiboles and 1:1 layersilicates. In: Health Effects of Mineral Dusts, Vol 28 (GuthrieGD Jr, Mossman BT, eds). Washington:Mineralogical SocietyofAmerica, 1993;61-137.

Environmental Health Perspectives * Vol 105, Supplement 5 * September 1997 1009

G.D. GUTHRIE JR.

28. Guthrie GD, Jr., Heaney PJ. Mineralogical characteristics ofthe silica polymorphs in relation to their biological activities.Scand J Work Environ Health 21:5-8 (1995).

29. Pauling L. The Nature of the Chemical Bond. 3rd ed.Ithaca:Cornell University Press, 1960.

30. Klein C, Hurlbut CS Jr. Manual of Mineralogy. 21st ed. NewYork:John Wiley & Sons, 1993.

31. Heaney PJ, Banfield JA. Structure and chemistry of silica, metaloxides, and phosphates. In: Health Effects of Mineral Dusts,Vol 28 (Guthrie GD Jr, Mossman BT, eds). Washington:Mineralogical Society ofAmerica, 1993; 185-233.

32. Carrisoza I, Munuera G, Castature and chemistry of silica,metal oxides, and phosphates. In: Health Effects of MineralDusts, Vol 28 (Guthrie GD Jr, Mossman BT, eds).Washington:Mineralogical Society, 1993;584 pp.

33. Anderson MA, Gieselmann MJ, Xu Q. Titania and aluminaceramic membranes. J Membr Science 39:243-258 (1988).

34. Tunesi S, Anderson MA. Surface effects in photochemistry: Anin sito CIR-FTIR investigation of the effect of ring substituteson chemisorption onto TiO2 ceramic membranes. Langmuir8:487-495 (1992).

35. Tunesi S, Anderson MA. The influence of chemisorption of thephotodecomposition of salicylic acid and related compoundsusing suspended TiO2 ceramic membranes. J Phys Chem95:3399-2405 (1991).

36. Tunesi S, Anderson MA. Photocatalysis of 3,4-DCB in TiO2aqueous suspensions: Effects of temperature and light intensity:CIE-FTIR interfacial analysis. Chemosphere 16:1447-1456(1987).

37. Maata K, Arstila AW. Pulmonary deposits of titanium dioxidein cytogenic and lung biopsy specimens. Lab Invest33:342-346(1975).

38. Zitting A, Skytta E. Biological activity of titanium dioxides. IntArch Occup Environ Heafth 43:93-97 (1979).

39. Driscoll KE, Maurer JK. Cytokine and growth factor releasealveolar macrophages: potential biomarkers of pulmonary toxi-city. Toxicol Pathol 19:398-405(1991).

40. Hawthorne FC. The crystal chemistry of the amphiboles. CanMineral 21:173-480 (1983).

41. Mossman BT, Marsh JP. Evidence supporting a role for activeoxygen species in asbestos-induced toxicity and lung disease.Environ Health Perspect 81:91-94 (1989).

42. Aust AE, Lund LG. The role of iron in asbestos-catalyzed damageto lipids and DNA. Biol Oxidation Systems 2:597-605 (1990).

43. Daniel LN, Mao Y, Saffiotti U. Oxidative DNA damage bycrystalline silica. Free Radic Biol Med 14:463-472 (1993).

44. Hansen K, Mossman BT. Generation of superoxide (02-) fromalveolar macrophages exposed to asbestiform and nonfibrousparticles. Cancer Res 47:1681-1686 (1987).

45. Janssen YM, Heintz NH, Marsh JP, Borm PJ, Mossman BT.Induction of c-fos and c-jun proto-oncogenes in target cells ofthe lung and pleura by carcinogenic fibers. Am J of Respir CellMol Biol 11:522-530 (1994).

46. Fawcett DW. A Textbook of Histology. Philadelphia:WBSaunders Company, 1986.

47. Lehnert BE. Defense mechanisms against inhaled particles andassociated particle-cell interactions. In: Health Effects ofMineral Dusts, Vol 28 (Guthrie GD Jr, Mossman BT, eds).Washington:Mineralogical Society ofAmerica, 1993;427-469.

48. Davis JMG. In vivo assays to evaluate the pathogenic effects ofminerals in rodents. In: Health Effects of Mineral Dusts. Vol28 (Guthrie GD Jr, Mossman BT, eds). Washington:Mineralogical Society ofAmerica, 1993;471-487.

49. Hume LA, Rimstidt JD. The biodurability of chrysotileasbestos. Am Mineral 77:1125-1128 (1992).

50. Palekar LD, Most BM, Coffin DL. Significance of mass andnumber of fibers in the correlation of V79 cytotoxicity withtumorigenic potential of mineral fibers. Environ Res46:142-152 (1988).

51. Jolicoeur C, Roberge P, Fortier J-L. Separation of short fibersfrom bulk chrysotile asbestos fiber materials: analysis and physio-chemical characterization. Can J Chem 59:1140-1148 (1981).

52. Mijailovich SM, Ejaz K, Godleski JJ, Krishna Murthy GG,Tsuda A. Unpublished data.

53. Hochella MF, Jr. Atomic structure, microtopography, compo-sition, and reactivity of mineral surfaces. In: Mineral-WaterInterface Geochemistry, Vol 23 (Hochella MFJ, White AF,eds). Washington:Mineralogical Society of America,1990;87-132.

54. Vallyathan V, Shi X, Dalal NS, Irr W, Castranova V.Generation of free radicals from freshly fractured silica dust.Am Rev Respir Dis 138:1213-1219 (1988).

55. Helfferich F. Ion Exchange. Mineola, NY:Dover Publications,1995.

56. Ataman G. The role of erionite (zeolite) on the development ofpulmonary mesothelioma. Comptes Rendus Hebdomadairesdes Seances de l'Academie des Sciences Serie D 291:167-169(1980).

57. Bish DL, Guthrie GD, Jr. Mineralogy of clay and zeolite dusts(exclusive of 1:1 layer silicates). In: Health Effects of MineralDusts, Vol 28 (Guthrie GD Jr, Mossman BT, eds).Washington:Mineralogical Society ofAmerica, 1993; 139-184.

58. Baris YI, Artvinli M, Sahin AA. Environmental mesotheliomain Turkey. Ann NYAcad Sci 330:423-432 (1979).

59. Baris I, Simonato L, Artvinli M, Pooley F, Saracci R, SkidmoreJ, Wagner C. Epidemiological and environmental evidence ofthe health effects of exposure to erionite fibres: a four-yearstudy in the Cappadocian region of Turkey. Int J Cancer39:10-17 (1987).

60. Wagner JC, Skidmore JW, Hill RJ, Griffiths DM. Erionite expo-sure and mesotheliomas in rats. Br J Cancer 51:727-730 (1985).

61. Suzuki Y, Kohyama N. Malignant mesothelioma induced byasbestos and zeolite in the mouse peritoneal cavity. Environ Res35:277-292(1984).

62. Korkina LG, Suslova TB, Nikolova SI, Kirov GN,Velichkovskii BT. Mekhanizm tsitotoksicheskogo deistviiaprirodnogo tseolita klinoptilolita [Mechanism of the cytotoxicaction of the natural zeolite clinoptilolite]. Farmakologiia iToksikologiia 47:63-67(1984).

63. Pylev LN, Krivosheeva LV, Bostashvili RG. 0 vozmozhnoikantserogennoi opasnosti tseolita-klinoptilolita [Possible car-cinogenic hazard of zeolite-clinoptilolite]. Gigiena Truda iProfessionalnye Zabolevaniia Mar:48-51(1984).

64. Nikolova S, Diakovska R, Burkova T. Biologichna aktivnost nabulgarski zeolitni skali [Biological activity of Bulgarian zeoliterocks]. Problemi na Khigienata 10:94-100(1985).

65. Fomina AS, Domnin SG, Gerasimenko TI, Sukhanova MM,Belobragina GV. Fibrogennaia aktivnost' pylei tseolitov tipa Y imordenita [Fibrogenic activity of type Y zeolite and mordenitedusts]. Gigiena Truda i Professionalnye ZabolevaniiaSep:17-19(1986).

66. Pylev LN, Bostashvili RG, Kulagina TF, Vasil'eva LA,Chelishchev NF. Otsenka kantserogennoi aktivnosti tseolita-klinoptilolita [Evaluation of the carcinogenic activity of zeoliteand clinoptilolite]. Gigiena Truda i ProfessionalnyeZabolevaniia May:29-34(1986).

67. Nikolova S, Vinarova M, Kirov G, Bradvarova I. Izsledvane invitro vzaimodeistvieto na prirodni zeoliti s diploidni kletki otchoveshki embrionalen bial drob [In vitro research on the interac-tion of natural zeolites with diploid cells from the human embry-onic lung]. Problemi na Khigienata 12:133-141(1987).

68. Pylev LN, Kulagina TF, Grankina EP, Chelishchev NF,Berenshtein BG. Kantserogennos't tsiolita-fillipsita[Carcinogenicity of zeolite-phillipsite]. Gigiena i SanitariiaAug:7-10(1989).

69. Kruglikov GG, Velichkovskii BT, Garmash TI, VolkogonovaVM. Strukturno-funktsional'nye izmeneniia makrofagovlegkikh pri fagotsitoze prirodnogo tseolita-klinoptilolita [Thestructural-functional changes in pulmonary macrophages dur-ing the phagocytosis of a natural zeolite-clinoptilolite]. GigienaTruda i Professionalnye Zabolevaniia 11-12:44-46(1992).

70. Tatrai E, Ungvary G. Study on carcinogenicity of clinoptilolitetype zeolite in Wistar rats. Polish J Occup Med Environ Health6:27-34(1993).

1010 Environmental Health Perspectives * Vol 105, Supplement 5 * September 1997

MINERALOGICAL PROPERTIES AND PARTICLE TOXICITY

71. Guthrie G, McLeod K, Johnson N, Bish D. Effect of exchange-able cation on zeolite cytotoxicity. In: GoldschmidtConference, 8-10 May 1992 Reston, VA;A-46.

72. Lasaga AC. Abinitio methods in mineral surface reactions. RevGeophys 30:269-303(1992).

73. Nolan RP, Langer AM, Harington JS, Oster G, Selikoff IJ.Quartz hemolysis as related to its surface functionalities.Environ Res 26:503-520(1981).

74. Addison CC, Addison WE, Neal GH, Sharp JH. Amphiboles.Part I. The oxidation of crocidolite. J Chem Soc,London: 1468-1471(1962).

75. Addison WE, Neal GH, Sharp JH. Amphiboles. Part II. Thekinetics of the oxidation of crocidolite. J Chem SocLondon: 1472-1475(1962).

76. Addison WE, Sharp JH. Amphiboles. Part III: The reductionof crocidolite. J Chem Soc London:3693-3698(1962).

77. White AF, Yee A. Aqueous oxidation-reduction kinetics associ-ated with coupled electron-cation transfer from iron-containingsilicates at 25°C. Geochim Cosmochim Acta 49:1263-1275(1985).

78. White AF, Peterson ML, Hochella MF Jr. Electrochemistryand dissolution kinetics of magnetite and ilmenite. GeochimCosmochim Acta 58:1859-1875(1994).

79. Ilton ES, Earley D, Marozas D, Veblen DR. Reaction of sometrioctahedral micas with copper-sulfate solutions at 25°C and 1atmosphere-an electron microprobe and transmission electronmicroscopy investigation. Econ Geol 87:1813-1829(1992).

80. Ilton ES, Veblen DR. Chromium sorption by phlogopite andbiotite in acidic solutions at 25°C: insights from X-ray photo-electron spectroscopy and electron microscopy. GeochimCosmochim Acta 58:2777-2788(1994).

81. Lide DR, ed. CRC Handbook of Chemistry and Physics. BocaRaton:CRC Press, 1992.

82. Ilton ES, Veblen DR. Copper inclusions in sheet silicates fromporphyry Cu deposits. Nature 334:516-518(1988).

83. Scott AD, Amonette J. Role of iron in mica weathering. In:

Iron in Soils and Clay Minerals. NATO ASI Series C: Vol 217(Stucki JW, Goodman BA, Schwertmann U, eds).Dordrecht:D. Reidel Publishing, 1988;537-624.

84. Vallyathan V, Kang JH, Van Dyke K, Dalal NS, Castranova V.Response of alveolar macrophages to in vitro exposure to freshlyfractured versus aged silica dust: The ability of Prosil 28, anorganosilane material, to coat silica and reduce its biologicalreactivity. Toxicol Environ Health 33:303-315 (1991).

85. Fubini B, Bolis V, Giamello E. The surface chemistry ofcrushed quartz dust in relation to its pathogenicity. Inorg ChimActa 138:193-197(1987).

86. Fubini B, Mollo L, Giamello E. Free radical generation at thesolid/liquid interface in iron containing minerals. Free RadicRes 23:593-614(1995).

87. Roggli VL, ed. Asbestos Bodies and Nonasbestos FerruginousBodies. Boston:Little, Brown, and Company, 1992.

88. Churg A. Asbestos lung burden and disease patterns in man.In: Health Effects of Mineral Dusts. Vol 28 (Guthrie GD Jr,Mossman BT, eds). Washington:Mineralogical Society ofAmerica, 1993;409-426.

89. Hochella MF Jr. Surface chemistry, structure, and reactivity ofhazardous mineral dust. In: Health Effects of MineralDusts.Vol 28 (Guthrie GD, Jr., Mossman BT, eds).Washington:Mineralogical Society ofAmerica, 1993;275-308.

90. Altree-Williams S, Byrnes JG, Jordan B. Amorphous surfaceand quantitative X-ray powder diffractometry. Analyst106:69-75(198 1).

91. Jordan B, O'Connor BH, Deyu L. Use of Rietveld pattern fit-ting to determine the weight fraction of crystalline material innatural low quartz specimens. Powder Diffrac 5:64-69(1990).

92. LePage Y, Calvert LD, Gabe EJ. Parameter variation in low-uartz between 94 and 298K. J Phys Chem Solids

41:721-725(1980).93. Baur WH, Khan AA. Rutile-type compounds. IV: SiO2, GeO2

and a comparison with other rutile-type compounds. ActaCrystallog B27:2133-2139(1971).

Environmental Health Perspectives * Vol 105, Supplement 5 * September 1997 1011