Using Environmental Forensic Microscopy in Exposure Science

www.nature.com/jes

Using environmental forensic microscopy in exposure science

JAMES R. MILLETTE, RICHARD S. BROWN AND WHITNEY B. HILL

MVA Scientiﬁc Consultants, 3300 Breckinridge Blvd, Suite 400, Duluth, Georgia, USA

Environmental forensic microscopy investigations are based on the methods and procedures developed in the fields of criminal forensics, industrial hygiene and environmental monitoring. Using a variety of microscopes and techniques, the environmental forensic scientist attempts to reconstructthe sources and the extent of exposure based on the physical evidence left behind after particles are exchanged between an individual and the environmentshe or she passes through. This article describes how environmental forensic microscopy uses procedures developed for environmental monitoring, criminal forensics and industrial hygiene investigations. It provides key references to the interdisciplinary approach used in microscopic investigations. Case studies dealing with lead, asbestos, glass fibers and other particulate contaminants are used to illustrate how environmental forensic microscopy can be veryuseful in the initial stages of a variety of environmental exposure characterization efforts to eliminate some agents of concern and to narrow the field of possible sources of exposure. Journal of Exposure Science and Environmental Epidemiology (2008) 18, 20–30; doi:10.1038/sj.jes.7500613; published online 7 November 2007

Keywords: dust, particulate, soil, indoor environment, exchange principle.

Introduction microscopy procedures used in comparing physical evidence including dust and specific particle types such as hairs, fibers Dust can be used as a metric for residential and building and mineral grains have been presented by Palenik (1988), exposure assessment and source characterization (Lioy et al., Saferstein (1995, 2006) and Petraco and Kubic (2003). 2002a). By using forensictechniques that have been In describing how dust can be used as a metric for developed by criminologists (now called forensic scientists) residential and building exposure and source characteriza- over many years, the field of environmental forensic tions, it has been suggested that from microscopic dust microscopy has come a long way in identifying sources of analyses, information is frequently obtained from one or indoor and outdoor contaminants and in improving how the more different microenvironments occupied or passed total human exposure is assessed in residential and office through by an individual or individuals over the course of environments. a day (Lioy et al., 2002a). This same concept is often referred Environmental forensic microscopy is the use of micro- to as the exchange principle by criminologists (forensic scopy following forensic procedures to characterize particles scientists) when describing the premise of their forensic trace and materials involved in environmental studies. The evidence studies. The Locard Exchange Principle was procedures used in environmental forensic investigations are proposed in the 1930s by Edmund Locard, the director of generally the same as the criminal forensic science approach the crime laboratory in Lyon, France. Locard’s Exchange to trace evidence analysis. Samples of the particles (dust, dirt, Principle states that whenever two objects come into contact, soil or suspensions in liquid) in question are collected and a transfer of material will occur (Locard, 1930). By analyzed by different microscopy techniques to identify the recognizing, documenting and examining the nature and particles and to determine possible sources. The procedure extent of this evidentiary exchange, Locard observed that then involves comparing the results of the analyses with criminals could be associated with particular locations, items cataloged information about particles that are emitted or of evidence and victims (Chisum and Turvey, 2000). In his released from suspect sources. In some situations, a writings on the subject of physical evidence and crime comparison is made of sample particle types to particles reconstruction, Locard wrote, ‘‘ythe criminologist recreates obtained from the suspect source(s). Guidelines for basic the criminal from traces the latter leaves behind, just as the archaeologist reconstructs prehistoric beings from his finds’’ 1. Address all correspondence to: Dr. J. R. Millette, MVA Scientific (Locard, 1930). On the basis of the Exchange Principle, Consultants, 3300 Breckinridge Blvd, Suite 400, Duluth, Georgia 30096, Locard built the first crime lab in the attic of Lyons Police USA. Tel.: 770-662-8509. Fax: 770-662-8532. Department, which later became the Institute of Criminalis- E-mail: [email protected] Received 5 March 2007; accepted 22 June 2007; published online 7 tics at the University of Lyons. The Exchange Principle is November 2007 widely regarded as a cornerstone of the forensic sciences and Environmental Forensic Microscopy Millette et al

Locard’s successful use of the Principle with scientific acterization of soil minerals is just one part of the methods led to the creation of forensic laboratories in many examination of dust and dirt particle samples. Dusts always countries. In environmental forensic particulate studies, the contain particles other than soil minerals. Among these may goal is to recognize, examine and document the nature and be biological substances such as pollen grains, skin cells, extent of the particles left behind after an interaction between plant fragments, combustion products (soot), building two environments or an environment and an individual. In materials such as paint flakes, construction debris and metal other words, the environmental forensicscientist attempts to flakes; and fibers (both natural and man-made). Microscopic reconstruct the sources and the extent of exposure based on forensicexaminations of hairs and fabricfibers have been the physical evidence left behind. long established procedures in tracing criminal evidence An illustration of the Exchange Principle and environ- evaluations (Saferstein, 1995). In the environmental and mental forensic microscopy is the case of dust that had industrial hygiene areas, standard procedures for the analysis accumulated on surfaces in an attic several feet below an of one particular fiber, asbestos, have been well documented asbestos-containing fireproofing. Investigators had concluded for a variety of media; air, dust, water and bulk building that a fallout from the fireproofing was the primary source of materials (Millette and Bandli, 2005). Some of these methods the asbestos fibers found in the dust based on the facts that have been adapted for the characterization and monitoring of the dust was the same color as the fireproofing and no other various types of glass fibers such as fiberglass, mineral wool, asbestos-containing products were apparent in the attic. It slag wool and rock wool. Ceramic fibers are also the subject was also apparent that chunks of debris had fallen from the of health concerns, and microscopy methods to characterize fireproofing to the surfaces in some areas. Although it them in some media are available. In the investigation of appeared fairly straightforward, another investigator insisted sooty deposits, the American Society for Testing and that the asbestos fibers in the dust were the result of Materials Method 6602 (ASTM, 2003) is a standard infiltration of fibers from other sources from inside and procedure that serves as an excellent basis for environmental perhaps outside of the building. Samples of the dust were sent forensic microscopy investigations where soot might be to a laboratory, and an environmental forensic microscopy involved. This ASTM standard was designed primarily for analysis of the dust was performed. The analysis showed that the determination of carbon black among soot particles and the dust contained particles of gypsum, vermiculite and other dark particles but provides the framework for the chrysotile asbestos including some where the three particle microscopy studies necessary to determine the identity of all types were combined in small aggregates. The aggregated particles and possible sources of surface contamination. particles were consistent with the composition of the Microscopic examination first by stereomicroscope and then fireproofing: gypsum binder, vermiculite filler and chrysotile polarized light microscope is used to sort out the various asbestos. The analysis did not find evidence of other asbestos- particle types present and to determine the approximate containing products such as floor tile fragments. The relative percentages by volume of the different components. conclusion of the forensic analysis was that the primary Transmission electron microscope analysis is used in source of asbestos in the dust on the attic surfaces was the conjunction with energy-dispersive x-ray analysis (TEM– asbestos-containing fireproofing. An exchange had taken place EDS) to identify the ‘‘fine’’ (small) size fraction of the dark between the building product and the attic environment. particulate and is especially useful in differentiating between various carbon soots. If a significant amount of fine (small) particulate such as soot is determined by polarized light Background: combining disciplines microscopy (PLM), additional analysis by TEM–EDS is necessary to confirm the presence of aciniform soot and to The methods used in environmental forensicinvestigations identify a particular type or source. are based on those used in criminal forensics, industrial Environmental forensic microscopy often involves the hygiene and environmental monitoring. The investigator analysis of dust samples where there is little information combines aspects of these disciplines to develop a strategy about the dust except that there is a suspicion it may be that fits the novel and often complex environmental situation. involved with a complaint. This is in contrast to criminal The microscopy techniques used in the field of sedimentary forensicinvestigations, where there is often some ‘‘intelli- petrography to characterize soils (Krumbein and Pettijohn, gence’’ associated with the sample. For example, a criminal 1938) were used as a basis for forensicsoil particle investigation may focus on matching glass fragments found classification (Graves, 1979; Murray, 2004). The use of on a suspect with a pane of broken glass located at the crime various microscopy techniques allows the identification of scene. Likewise, it may involve the comparison of hairs or individual (detrital) grains of soil and has been used to link fibers found on a victim with hairs and fibers associated with suspects to the scene of the crime and to aid in determining a suspect. The initial environmental forensic investigations the victim’s location (Palenik, 1979; McCrone, 1992; Hopen, usually require a broader, often less routine approach than 2004). In environmental forensicinvestigations, the char- in the criminal arena, because there are so many possibilities

Journal of Exposure Science and Environmental Epidemiology (2008) 18(1) 21 Millette et al Environmental Forensic Microscopy to consider. For most environmental forensic investigations wavelength, low-energy radio waves that are meters in length simple chain of custody documents are sufficient to transfer to short wavelength, high energy, g-rays that are in the sub- samples. Once the investigation has been narrowed by the nanometer range. Visible light corresponds to a wavelength initial light microscopy examination, specific standard range of 400–700 nanometers (nm) or 0.4–0.7 micrometers methods developed through groups like the American (mm). Society for Testing and Materials (ASTM-International) A polarized light microscope is a compound light may be used depending on the contaminant of interest and microscope that contains a piece of polarizing material in the type of microscope involved. Standards to calibrate the the light path below the sample and then another in the light different types of microscopes for magnification and particle path above the sample. As light passes through a crystal analysis are available from the National Institute of between the two polarizing sheets, it reveals information Standards and Technology (NIST). One combination about its optical properties. The PLM identification of calibration standard for TEM with NIST traceable compo- particles depends on characterizing particles according to nents is available from the authors (Few and Millette, 1996). several optical crystallographic properties such as refractive indices, dispersion staining, birefringence, sign of elongation and extinction angle (McCrone and Delly, 1972). Methods An infrared microscope makes use of the electromagnetic wavelengths that are longer than those of visible light. By Collection recording the absorption of specific infrared wavelengths, Collecting samples of airborne particles for environmental different types of molecules can be characterized. forensicinvestigations is usually done using membrane filter The electron microscopes, scanning and transmission, use sampling cassettes. Various types of membrane filters are used electromagnetic coils as lens to form magnified images with in environmental and industrial hygiene sampling. These an electron beam in a way that is similar to the way a light include filters made of mixed cellulose ester polymer, microscope uses glass lenses and a light beam to form polycarbonate (PC) polymer, glass fiber filters, quartz fiber magnified images. Because electrons can be accelerated with filters, silver membrane filters and polyvinyl chloride (PVC) high potential energies to produce a beam with a wavelength filters. Microscopists prefer air samples collected on PC that is much smaller than the wavelengths of visible light, the filters, because the PC filters have a flat surface. electron microscope can resolve particles that are smaller Surface dust particle collection techniques vary according than those that can be seen with the light microscope. A to situation, are chosen to maximize probative value and, scanning electron microscope produces images of the surfaces when possible, to preserve a portion for future analysis. of particles similar to those seen with reflected light. A TEM Common particle collection techniques for environmental produces images that are similar to those seen with forensic microscopy investigations of surface particulate transmitted light. The TEM images are essentially ‘‘shadow include: scraping, brushing, taping with cellophane tape on pictures.’’ In addition to forming images, both types of a slide or post-it notes, scoop and bag techniques, wipe electron microscopes make use of the fact that the electron sampling, vacuuming with either a microvac (air cassette) or beam interacts with the atoms of a particle to produce x-rays. large-scale vacuuming (Millette and Few, 2001). Sample The energies of the x-rays are different for each element and collection procedures with cotton or polyester balls, wipes are related to the atomicnumber of the element. Both and tape slides for outdoor surfaces are described in the scanning electron microscopy (SEM) and TEM instruments ASTM Standard Practice D6602 (ASTM, 2003). Because commonly use energy-dispersive x-ray spectroscopy analysis case objectives and dust and particulate characteristics differ (EDS or EDXA) detectors to acquire the x-ray information from site to site, there is no single sampling method of choice that allows the determination of the elemental composition of for all situations. When the concern is about exposure to a particle. In addition to imaging and x-ray analyses, the particles that can be readily entrained into the air during dust TEM can also be used in the selected area electron diffraction disturbance, the microvac sampler is preferred. For situations (SAED) mode to gather information about the crystal where the concern is darkening agents or stains on property, structure of the particle. The SEM–EDS analysis is generally a combination of tape lift and cotton ball provides the best used to compliment the PLM analysis, because the SEM can media for the analyst to examine. In most cases, it is best to be used to study particles smaller than those examined by collect several samples of dust particles from the same area PLM. Of equal importance is the SEM–EDS capability to with more than one sampling medium and let the laboratory provide information about opaque particles that do not show choose which is best for analysis. optical properties in the light microscope. The SEM–EDS generates information about the elemental composition of the Analysis Equipment: Different Types of Microscopy particles seen during the light microscope examination. This Different types of microscopes use different portions of the information is especially useful for identifying particles electromagnetic spectrum that spans a range from long containing metals such as lead, cadmium and chromium.

22 Journal of Exposure Science and Environmental Epidemiology (2008) 18(1) Environmental Forensic Microscopy Millette et al

TEM–EDS–SAED analysis is generally used to characterize are different based on their optical properties using the the fine fraction of particles (less than 5 mm) and those polarized light microscope. The optical properties of particles that require crystallographic identification such as thousands of crystals and other particles have been cataloged asbestos. to aid an analyst in identifying the particles in a sample of Initial forensicinvestigations of environmental dusts dust by PLM (McCrone and Delly, 1972; Petraco and usually begin with an examination of the particles using a Kubic, 2003). Although environmental forensic microscopic stereobinocular microscope and PLM. In some situations particle analysis of dusts may seem daunting due to the many when dealing with the investigation of unknown dust types of particles possible, it has been found that for many particles, the PLM analysis is sufficient to answer a question common situations such as normal residential and building about the presence of a specific contaminant. In other dust, the particles can be classified by PLM examination situations, other microscopy techniques such as infrared using approximately 20 particle classes (Millette et al., 2003; microscopy or electron microscopy may be needed to Turner et al., 2005). complete the analysis. The choice of which additional When the light microscopy results suggest that the dust microscopy tool to use is dependent on the results of the may contain fragments of plastics, resins or polymers, initial PLM analysis. For identifying particles by PLM, the infrared microscopy, using Fourier transform infrared micro- microscopist studies their shape and optical properties such spectrophotometry (micro-FTIR), is very useful. Figure 2 as birefringence and refractive index, using various techni- shows a comparison of the FTIR spectra from a nylon fiber ques including dispersion staining. Figure 1 shows how foundinofficedustwithaknownreferencenylonfiberfrom particles that come from two similar-looking white powders our polymer database. In the same sample, the FTIR was

Figure 1. PLM images showing the difference between two particles types that are white powders (a is starch and b is zinc arsenate as viewed in reflected light, dark-field mode). Images c (starch) and d (zinc arsenate) show that the white materials have different morphologies when viewed in transmitted bright-field light. Images e (starch) and f (zinc arsenate) show very different optical properties for the two white particle types when viewed using crossed polarizers with a polarizing light microscope.

Journal of Exposure Science and Environmental Epidemiology (2008) 18(1) 23 Millette et al Environmental Forensic Microscopy instrumental in identifying fragments of aramid fibers. The Hundreds or thousands of particles from each sample can be source of the aramid particles was determined to be individually sized (typically down to 0.5 mm in diameter) and deterioration of an elevator cable made of aramid fibers classified according to elemental composition. (e.g., Kevlar created by DuPont, or Twaron produced by TEM also allows the analyst to see particles that are Teijin). smaller than those that can be seen with light microscopy and SEM is used to investigate particles that are smaller than when equipped with electron diffraction capabilities and an those that can be seen with light microscopy or are opaque. x-ray analysis unit, allows the analyst to determine the crystal When equipped with an x-ray analysis unit (SEM–EDS), the structure and elemental composition of the particles. Figures SEM allows the analyst to determine the elemental 5a, b and cshow a TEM image, SAED pattern and TEM composition of the particles. Figures 3 and 4 show an image x-ray elemental spectrum from an aggregate of nano-silicon and an elemental spectrum from a stainless steel particle that particles. was found in indoor dust as part of an environmental forensic source investigation, following the finding of elevated chromium in the dust during chemical testing. Automated particle analysis is done with a scanning electron microscope, operating in automated mode under the control of an x-ray analysis system utilizing a standard operating procedure.

2 Fiber from office dust

1.5

Absorbance Nylon reference spectrum

0.5

0 4000 3500 3000 2500 2000 1500 1000 Wavenumber (cm-1) Figure 2. A comparison of the FTIR spectra from a ﬁber found in Figure 3. SEM image of a stainless steel particle found in an indoor ofﬁce dust with a known reference nylon polymer. dust sample.

Full scale counts: 1074 Fe

1000

800

600

400

Cr Fe 200 Fe Si C Cr

0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 Kev Figure 4. The elemental x-ray spectrum from a stainless steel particle shown in Figure 3. The spectrum shows iron, chromium, carbon and silicon.

24 Journal of Exposure Science and Environmental Epidemiology (2008) 18(1) Environmental Forensic Microscopy Millette et al

Figure 5. (a) TEM image of nano-sized silicon carbide particles in an aggregate. (b) TEM diffraction pattern from nano-sized silicon carbide particles shown in (a). (c) TEM x-ray analysis spectrum from nano-sized silicon carbide particles shown in (a).

Examples of Applications of Environmental Forensic duties admitted to bringing her cat with her and allowing it to Microscopy roam throughout the office suite. Environmental forensic microscopy can be very useful in a The second example concerns municipal maintenance new situation when trying to determine the initial direction of workers in a tunnel in California. After receiving complaints an environmental exposure assessment effort. It can be used of symptoms of respiratory distress, samples of the dust in to eliminate some agents of concern and to narrow the field of the tunnels were sent to the municipal laboratory for possible sources of exposure. The following are two examples chemical analysis. The chemical analysis showed no obvious that illustrate the usefulness of starting with a careful toxic elements, and a sample of dust collected by scraping microscopic examination of the dust collected at the site into a plastic vial was sent for microscopic testing. The PLM where a person is complaining of symptoms that may be examination quickly showed that there were thousands of caused by a variety of agents. microscopic mold spores in the dust. With this knowledge, proper worker protection measures were determined. Having shown that the direction of the environmental exposure Application of environmental forensic microscopy in an assessment should be biological rather than chemical, the office and tunnel The first case is that of a lady suffering samples could be analyzed further to determine the specific from apparent allergicreactionsymptoms when she started type of the mold spores. workeachweekinhernewoffice.Testingformoldand organicvapors showed negative results. A sample of dust Previously published examples of building complaints collected using a transparent tape lift was sent for investigated by microscopy Some examples of environmental microscopic testing. The analysis showed cat hairs forensic microscopy investigations have been published prevalent in the office dust. The lady was allergic to cats previously. Applications of microscopic techniques as an thus solving the question of her environmental exposure and approach in understanding issues of exposure intensity and symptoms. Eventually, another lady who came into the office sourceidentificationwereprovidedinanarticleaboutusing on Sundays after the cleaning staff had performed their dust analysis as a metricfor exposure evaluation (Lioy et al.,

Journal of Exposure Science and Environmental Epidemiology (2008) 18(1) 25 Millette et al Environmental Forensic Microscopy

2002a). One case concerned the investigation of ‘‘ghosting’’ Environmental forensic microscopy has shown instances also known as ‘‘black soot deposition.’’ In this case, where a source of lead other than the generally recognized microscopic examination of the black material by TEM principal sources was important. Subsequent to their early and x-ray elemental analyses showed that the black material studies of 1992, Hunt et al. (1998) used similar SEM was soot particles consistent with carbon soot from paraffin procedures to explain the source of lead recontamination burning, but it was established that candles were not used in after a lead paint abatement. They found that particulate lead the residence. Further investigation showed that the cause of was consistent in form with lead scraped from the interior of the ghosting problem was soot from paraffin logs that had one of the radiators, which had been removed from the been used in the fireplace. A backdraft was apparently complex some time after the abatement. Millette et al. (1991) carrying the soot particles from the interior chimney walls used similar SEM–EDS procedures in the investigation of throughout the house. In the second case, black and white lead in dust in a residence located in Georgia, USA. particles were found in a governmental office building several Following a finding that a child who had an elevated level days in a row on a desk. The particles were found during the of lead in her blood, a lead-based paint abatement was fall months of the year. The desk was located under an air planned for the residence. However, testing of the paint in the system duct grate. Light and scanning electron microscopies home showed no lead. Samples of the dust were collected on showed that the particles were black clusters of fungal spores adhesive notes and sent to the laboratory for environmental and white fragments of galvanized metal corrosion product. forensicanalysis by SEM. The analysis included x-ray Apparently, the seasonal change in humidity caused the analysis of the elements in the dust particles. The results release of fungal material and corrosion product particles showed the source of the lead in dust was lead-containing fly from the duct. In Case 3, residents complained of eye ash. A search of the neighborhood determined that there irritation and general itchiness in a convent in Oklahoma. were piles of material at a nearby industrial site. The SEM– Glass fibers from the duct insulation were suspected of EDS showed lead-containing fly ash consistent with that causing the problem. Air samples collected on PC filters and found within the home. Several steel mills in the southeastern examined by light microscopy did not show the presence of United States had paid Sogreen Corp. to take their electric- glass fibers or other particles that are considered normal arcfurnacefly-ash waste dust. The waste material contained irritants. Analysis of particles associated with the duct about 10% zincand 3.6% lead. Sogreen mixed the fly-ash insulation sent as a reference showed a high concentration waste with limestone and sold it as a fertilizer. Unfortunately, of mites. Additional testing for mite antigens was the fertilizer business was not keeping up with the shipments recommended. of fly ash, and large piles were accumulating around the facility. Lead-containing particles were apparently trans- ported by air and foot trafficinto the residence. The residence Examples of environmental forensic microscopic analyses was cleaned with wet wiping and a high-efficiency vacuum involving metals Electron microscopes equipped with x-ray system. The site was ultimately declared a Superfund site and elemental detectors have been involved in several of the cleanup measures initiated. This example illustrates the published accounts of environmental forensic investigations Exchange Principle, where leaded particles were left behind of metallic particles. Automated electron microscopy x-ray after an interaction between the indoor environment and the microanalysis on individual particles on more than 120 environment of the contaminated industrial site. samples collected at two sampling stations in central Siberia The SEM–EDS analysis procedures have played an provided evidence that industrial complexes in the former important role in environmental forensicinvestigations of Soviet Union contributed to Arctic pollution (Malderen various metal contaminants including copper, tin, zinc and et al., 1995). Automated SEM particle analysis has been used lead arsenate (Rybicka et al., 1994; Thornton et al., 1994; extensivelyintheinvestigationofleadedparticlesinsoil/ Brown et al., 1995; Millette et al., 1995; Pirrie et al., 2003). house dust. Hunt et al. (1991) have described a study of the In these published studies, the chemical composition and apportionment of lead in the dusts of six households in sometimes morphology as determined by SEM with x-ray London using automated SEM with energy-dispersive x-ray analysis on individual particles played a role in determining analysis. The authors concluded that road dust was the the source or distribution of metal particles in environmental highest contributor to lead, with paints and soils being the samples. For example, microscopic analysis of the dust in a next largest contributors. This work and additional computer facility showed that metallic fibers or whiskers were microscopic studies lead Hunt et al. (1992) to conclude present. Additional investigations showed that zincwhiskers that the generally recognized principal sources of house-dust actually grew in-place in the false floor environment in a lead were paint lead, soil lead and road-dust lead. The likely computer facility (Millette et al., 2004). When metals such sources of soil and road dust lead were industrial emissions as zinc, tin, copper, cadmium, chromium and lead are and previous depositions of vehicular lead, although exterior associated with other elements in individual particles the lead paint may also play a part. x-ray spectrum which shows these associations can be used

26 Journal of Exposure Science and Environmental Epidemiology (2008) 18(1) Environmental Forensic Microscopy Millette et al as a signature to track particle dispersion from a particular some nanomaterials may have the potential to cause harm to type of source. people and the environment (Maynard et al., 2006). Environmental forensic microscopy was also used in an As the toxicology of certain types of nanoparticles investigation of lead particles found in the spaces between the becomes clear (Oberdorster et al., 2005) exposure evaluation walls of an office building that had been affected by the will be necessary to address the health concerns. Microscopy collapse of the World Trade Center (WTC) towers on 11 will be an important tool in assessing exposure, because some September 2001. One laboratory that used SEM–EDS to nanoparticles, such as nanotubes (Figure 6) and clusters of study the wall cavity dust particles theorized that particles fullerenes, molecules composed of 60 carbon atoms, are best containing lead and chlorine could only have come from the characterized by their morphology on a nanoscale. Although WTC dust, because they were not consistent with lead it is not clear which aspects of nanomaterials should be particles from paint or other recognized sources. However, measured F number, surface area or mass concentration, a SEM–EDS studies by another laboratory of leaded particles combination of these, or something else entirely F it is clear found that in the same wall cavity dust, the most likely source that environmental forensic microscopy can play an im- of lead–chlorine particles was lead used in the plastic portant role in the identification of which nanoparticles are insulation on wires that ran through the wall cavities (Ghazi present. There is currently no reliable source of information and Millette, 2004). This conclusion is supported by an about how frequently nanoparticles are used today in home interesting report in the scientific literature by the Centers for and personal products. However, research has begun to Disease Prevention and Control (CDC) about a case of lead understand how the particles may be dispersed in the intoxication associated with chewing plastic wire coatings environment. Hyung et al. (2007) has shown that natural (Kelley et al., 1993). In December 1991, a venous blood lead organicmatter stabilizes carbon nanotubes in aqueous level of 50 mg/dl was detected in a 46-year-old Ohio man environments; an interaction that increases the potential for during a routine preemployment examination. He reported their transport downstream. no history of exposure to lead from soldering or welding. He Microscopy is already used to identify nanoparticles of had no activities or hobbies associated with exposure to lead welding fumes (Figure 7), and aggregates of nanoparticles of or lead products, no previous bullet or birdshot wounds, and aciniform soot are generated by various sources of combus- he denied drinking illicitly distilled alcohol or using lead tion. When the burning conditions are uncontrolled, soots additives in his car. His residence was built in 1974 (after lead of a variety of sizes are generated when the heated vapors was banned from use in residential paint), and household condense. Figure 8 shows that nano-sized primary soot water was obtained from a well that was low in lead content. spheres are not found normally as nanoparticles but as larger Eventually, he mentioned that for approximately 20 years, he aggregated soot particles. Carbon blacks, lampblacks and had habitually chewed on the plastic insulation that he furnace blacks are commercially produced soot particles that stripped off the ends of electrical wires. Samples of the wire areusedinplasticsandtiresandasblackpigmentsinpaints. coatings were obtained and analyzed for lead content. The coatings contained 10,000–39,000mgofleadpergramof coating. He was being exposed to the lead added as a stabilizer in PVC wire coatings. Environmental forensic microscopy is especially useful when assessing exposures to fibers. The morphological characteristics of a fiber, its length and width, can best be determined by microscopic examination. Other techniques such as bulk chemical analysis or x-ray diffraction can provide some information about the elemental or mineralogical makeup of a dust sample, but only microscopy can provide full information about the fibrous nature of the dust components. Microscopic analyses for glass fibers and asbestos fibers are an important part of the investigation of the components of the dust produced by the WTC collapse (Lioy et al., 2002b; Millette et al., 2002; Yiin et al., 2004) and in the proposed assessment of which residences will be part of the cleanup effort (USEPA, 2005).

Application of environmental forensic microscopy to nanoparticles It is generally accepted that, in principle, Figure 6. TEM image of multiwalled nanotubes.

Journal of Exposure Science and Environmental Epidemiology (2008) 18(1) 27 Millette et al Environmental Forensic Microscopy

Figure 9. TEM image of glass ﬁber recovered from a human lung sample.

Figure 7. TEM image of welding fume particles. forensic microscopy. This particular area that overlaps with forensic pathology uses microscopes to characterize and identify particles to which an individual has been exposed and has inhaled (Abraham, 1979; Abraham et al., 1999, 2002; Fireman et al., 1999, 2004). A microscopic count of particles per amount of processed tissue can be used to determine a lung tissue particle burden, and the characterization of the types of particles can be used to determine the sources of exposure the individual had. Analysis of sputum can also be used to characterize exposure. Figure 9 shows an image taken with a TEM of a glass ﬁber found in a lung tissue. The tissue had been processed using a bleach digestion procedure developed for asbestos lung burden testing to eliminate the organic material and concentrate the particles for analysis (Dodson and Atkinson, 2006).

Strengths and limitations Figure 8. TEM image of an aciniform soot aggregate. Microscopy is most useful in examining, classifying and identifying particulate contaminants. The information deter- They are formed as aggregates of nanoparticles during mined may be valuable in its own right if the particulates are carefully controlled industrial processes. The ASTM Micro- a source of concern (Millette et al., 1994) and may also be scopy Practice D6602 (ASTM, 2003) for differentiating valuable in assessing the significance of environmental carbon black from other soots such as automobile and diesel transport and deposition processes (Lioy, 2006). Because exhausts, fireplace soots, industrial flare emissions and other forensic microscopy analysis is based on a particle-by-particle uncontrolled burning situations has been found to be useful examination, it can be more sensitive than other techniques in the assessment of soot particles in environmental situa- when determining minute amounts. Looking for different tions. Using this method, darkening agents that cause sources of a specific element such as cadmium in a dust, for concern to residents can be identified and their sources instance, when the element is present at a part-per-million determined. (ppm) level is not feasible by microscopy, unless there is a way to concentrate the particles of interest. Although Application of Environmental Forensic Microscopy to automated electron microscopy can provide chemical and Particles in Lung Tissue The examination for particles in size information in the order of thousands of particles in lung tissue and sputum is a special case of environmental a sample, and manual microscopy generally involves

28 Journal of Exposure Science and Environmental Epidemiology (2008) 18(1) Environmental Forensic Microscopy Millette et al

examinations in the order of hundreds of particles per misclassification of source attribution and reduce uncertain- sample, this is not very useful if the element is dispersed ties in exposure characterizations for risk estimation. uniformly throughout the sample at a level of 50 parts per million. However, microscopy can be very useful in source determination, if the element is concentrated in only a few Acknowledgements particles of nearly pure elemental composition that standout from the other materials in the sample. With the exception of We thank Frank Bull, the Atlanta architect, for his questions determining the concentration of various types of fibers in a in 1986 concerning the source of asbestos fibers in dusts that sample, quantifying the specific amount of a contaminant, be initiated a number of studies that lead to the development of it elemental (lead, cadmium, chromium, etc.), organic environmental forensic microscopy procedures in our labora- (hydrocarbons, etc.) or biological (mold, bacteria, etc.), tory for all types of environmental particulates. usually requires the use of other analytical methods. Determining the nature of a contaminant and measuring the amount of a contaminant are two fundamentally, References although often overlapping, questions in environmental forensic investigations. In most instances, the microscopic Abraham J.L. Documentation of environmental particulate exposures in humans using SEM and EDXA. In: Johari O. (Ed.) Scanning Electron Microscopy, examination is most useful when applied to the first task. 1979, II. Scanning Electron Microscopy Inc., Chicago, 1979, pp. 751–766. Abraham J.L., Hunt A., and Burnett B.R. Steel particles as a marker of welding fume exposure, frequency of occurrence and prevalence of steel particles in a series of over 400 human lungs. Ann Occup Hyg 1999: 38(Suppl. 1): 551–557. Conclusions Abraham J.L., Hunt A., and Burnett B.R. Quantification of non-fibrous and fibrous particulates in human lungs: twenty year update on pneumoconiosis Lioy (2006) points out that our indoor environment contains database. Ann Occup Hyg 2002: 46(Suppl. 1): 397–401. ASTM. Standard Practice D6602-03b, Sampling Testing of Possible Carbon Black many types of organicand inorganicmixtures that can be Fugitive Emissions or Other Environmental Particulate, or Both.American found simultaneously in air or deposited on surfaces. This is Society for Testing Materials International, West Conshohocken, PA, 2003. certainly true for the outdoor environment as well. Environ- Brown R.S., Millette J.R., and Mount M.D. Application of scanning electron microscopy for pollution particle source determination in residential dust and mental forensicmicroscopyallows an investigator to soil. Scanning 1995: 17(5): 302–305. characterize the particles in a sample in ways that are Chisum J.W., and Turvey B.E. Evidence dynamics: Locard’s Exchange Principle different from the results of instrumental analyses. These & crime reconstruction. J Behavioral Profiling 2000: 1(1): 25–38. Dodson R.F., and Atkinson M.A.L. Measurements of asbestos burden in tissues characterizations augment the information determined by F identifying and preventing hazards in the environment and at work. Ann other tests and allow the determination of contaminant NY Acad Sci 2006: 1076: 281–291. sources. With the information obtained through the environ- Few P., and Millette J.R. A combination calibration grid for transmission electron microscopes. Microscope 1996: 44(4): 175–179. mental forensic study of the particle types exchanged between Fireman E.M., Greif J., Schwarz Y., Man A., Ganor E., Ribak Y., and Lerman one or more different microenvironments occupied or passed Y. Assessment of hazardous dust exposure by BAL and induced sputum. through by an individual over the course of a day, the Chest 1999: 11: 1720–1728. Fireman E.M., Lerman Y., Ganor E., Greif J., Fireman-Shoresh S., Lioy P.J., investigator can more fully assess an individual’s exposure to Banauch G.I., Weiden M., Kelly K.J., and Prezant D.J. Induced sputum a particular contaminant. This article has shown how assessment in New York City firefighters exposed to World Trade Center dust. environmental forensic microscopy can be very useful in the Environ Health Perspect 2004: 112: 1564–1569. Ghazi A.M., and Millette J.R. Environmental forensicapplicationof lead isotope initial stages of an environmental exposure assessment effort ratio determination: a case study using laser ablation sector ICP-MS. where it is used to eliminate some agents of concern and to Environmental Forensics 2004: 5(2): 97–108. narrow the field of possible sources of exposure. It is clear Graves W.J. A mineralogical soil classification technique for the forensic scientist. JForensicSci1979: 24: 323–338. that microscopy plays an important role when concerns Hopen T.J. The value of soil evidence. In: Houck M.M. (Ed.). Trace Evidence over contaminants in the fibrous form are involved. It is also Analysis. Elsevier Academic Press, Boston, 2004, pp. 105–122. apparent that microscopic examination will have a critical Hunt A., Hawkins J., Gilligan E., and Bhatiab S. A comparison of the lead particle content of indoor dust before and after a lead paint abatement: a new role in the characterization of exposures to unwanted source of lead recontamination. Indoor Built Environ 1998: 7: 32–46. nanoparticles. The important challenge to be answered in Hunt A., Johnson D.L., and Thornton I. Descriptive apportionment of lead in the next few years is how to integrate environmental forensic house dust by automated SEM. Water Air Soil Pollut 1991: 57–58: 69–77. Hunt A., Johnson D.L., Watt J.M., and Thornton I. Characterizing the sources of microscopy into an overall assessment strategy involving particulate lead in house dust by automated scanning electron microscopy. inorganic chemical testing, biological monitoring (bacterial, Environ Sci Technol 1992: 26(8): 1513–1526. fungal, allergens) and organicanalyses (volatile and semi- Hyung H., Fortner J.D., Hughes J.B., and Kim J. Natural organicmatter stabilizes carbon nanotubes in the aqueous phase. Environ Sci Technol 2007: volatile chemicals). Environmental forensic microscopy is a 41(1): 179–184. tool that should be seriously considered when designing an Kelley M., Watson P., Thorton D., and Halpin T.J. Lead intoxication associated exposure characterization study that may have multiple with chewing plastic wire coating. MMWR Morb Mortal Wkly Rep 1993: 42: 465–467. sources or material with a complex mixture of components. Krumbein W.C., and Pettijohn F.J. Manual of Sedimentary Petrography. Microscopic analysis early in the study will minimize Appelton-Century-Crofts, New York, 1938.

Journal of Exposure Science and Environmental Epidemiology (2008) 18(1) 29 Millette et al Environmental Forensic Microscopy

Lioy P.J. Employing dynamical and chemical processes for contaminant mixtures Murray R.C. Evidence From the Earth, Forensic Geology and Criminal outdoors to the indoor environment: the implications for total human exposure Investigation. Mountain Press Publishing Co., Missoula, MT, 2004. analysis and prevention. J Expos Analysis and Environ Epi 2006: 16: 207–224. Oberdorster G., Oberdorster E., and Oberdorster J. Nanotoxicology: an emerging Lioy P.J., Freeman N.C., and Millette J.R. Dust: a metricfor use in residential discipline evolving from studies of ultraﬁne particles. Environ Health Perspect and building exposure assessment and source characterization. Environ Health 2005: 113: 823–839. Perspect 2002a: 110(1): 969–983. Palenik S.J. The determination of geographical origin of dust samples. In: Lioy P.J., Weisel C.P., Millette J.R., Eisenreich S., Vallero D., Offenberg J., McCrone W.C., Delly J.G., Palenick S.J. (Eds.). TheParticleAtlas, Buckley B., Turpin B., Zhong M., Cohen M.D., Prophete C., Yang I., Stiles 2nd edn. Vol 5, Ann Arbor Science Publishers Inc., Ann Arbor, 1979 pp. R., Chee G., Johnson W., Porcja R., Alimokhtari S., Hale R.C., Weschler C., 1347–1361. and Chen L.C. Characterization of the dust/smoke aerosol that settled east of Palenik S.J. Microscopy and microchemistry of physical evidence. In: Saferstein R. the World Trade Center (WTC) in Lower Manhattan after the collapse of the (Ed.). Forensic Science Handbook, Vol 2, Prentice Hall, Englewood Cliffs, NJ, WTC 11 September 2001. Environ Health Perspect 2002b: 110(7): 703–714. 1988 pp. 161–208. Locard E. The analysis of dust traces. Part II. Am J Police Sci 1930: 1: 401–418. Petraco N., and Kubic T. Color Atlas, Manual of Microscopy for Criminalists, Malderen H., Van Grieken R., Bufetov N.V., and Koutzenogii K.P. Chemical Chemists, and Conservators. CRC Press, New York, 2003. characterization of individual aerosol particles in central Siberia. Environ Sci Pirrie D., Power M.R., Rollinson G., Camm G.S., Hughes S.H., Butcher A.R., Technol 1995: 30(1): 312–321. and Hughes P. The spatial distribution and source of arsenic, copper, tin and Maynard A.D., Aitken R.J., Butz T., Colvin V., Donaldson K., Oberdo¨ rster G., zincwithin the surfacesediments of the Fal Estuary, Cornwall, UK. Philbert M.A., Ryan J., Seaton A., Stone V., Tinkle S., Tran L., Walker N.J., Sedimentology 2003: 50(3): 579–595. and Warheit D.B. Safe handling of nanotechnology. Nature 2006: 444: 267–269. Rybicka E.H., Wilson M.J., and McHardy W.J. Chemical and mineralogical McCrone W.C. Forensic soil examination. Microscope 1992: 40: 109–121. forms and mobilization of copper and lead in soils from a Cu-smelting area in McCrone W.C., and Delly J.G. TheParticleAtlas. Ann Arbor Science Publishers Poland. J Environ Sci Health 1994: Part A A29(3): 531–546. Inc., Ann Arbor, 1972. Saferstein R. Criminalistics: An Introduction to Forensic Science. 5th edn. Prentice Millette J.R., and Bandli B.R. Asbestos identiﬁcation using available standard Hall, Englewood Cliffs, NJ, 1995. methods. Microscope 2005: 53(4): 179–185. Saferstein R. Criminalistics: An Introduction to Forensic Science. 9th edn. Prentice Millette J.R., Boltin R., Few P., and Turner Jr W. Microscopical studies of World Hall, Englewood Cliffs, NJ, 2006. Trade Center disaster dust particles. Microscope 2002: 50(1): 29–35. Thornton I., Watt J.M., Davies D.J.A., Hunt A., Cotter-Howells J., and Johnson Millette J.R., Brown R.S., Barnett J., and Mount M.D. Scanning electron D.L. Lead contamination of UK dusts and soils and implications for microscopy of post-it notes used for environmental sampling. NAC Journal childhood exposure: an overview of the work of the Environmental 1991: 1: 32–35. Geochemistry Research Group, Imperial College, London, England 1981– Millette J.R., Brown R.S., and Mount M.D. Lead arsenate. Microscope 1995: 1992. Environ Geochem Health 1994: 16: 113–122. 43(4): 187–191. Turner W.L., Millette J.R., Boltin W.R., and Hopen T.J. A standard approach to Millette J.R., Brown R.S., Turner Jr W., and Few P. Zincwhiskers in computer the characterization of common indoor dust constituents. Microscope 2005: facility dust. Microscope 2004: 52(2): 59–62. 53(4): 169–177. Millette J.R., and Few P. Sample collection procedures for microscopical United States Environmental Protection Agency (USEPA). Final Report on the examination of particulate surface contaminants. Microscope 2001: 49(1): World Trade Center (WTC) Dust Screening Method Study. U.S. Environ- 21–27. mental Protection Agency, Washington, DC, 17 August, 2005. Millette J.R., Hopen T.J., and Brown R.S. Investigating the particulate Yiin L-M., Millette J.R., Vette A., Ilacqua V., Quan C., Gorczynski J., Kendall component in indoor air quality concerns. Environmental Choices 1994: 3(2): M., Chen L.C., Weisel C.P., Buckley B., Yang I., and Lioy P.J. Comparisons 14–16. of the dust/smoke particulate that settled inside the surrounding buildings and Millette J.R., Lioy P.J., Wietfeldt J., Hopen T.J., Gipp M., Padden T., Singsank outside on the streets of southern New York City after the collapse of the C., and Lepow J. A microscopical study of the general composition of World Trade Center, September 11, 2001. J Air Waste Manage Assoc 2004: household dirt. Microscope 2003: 51(4): 201–207. 54: 515–528.

30 Journal of Exposure Science and Environmental Epidemiology (2008) 18(1)