Aerosol and Air Quality Research, 13: 11451150, 2013 Copyright Taiwan Association for Aerosol Research ISSN:

: 1680-8584 print / 2071-1409 online doi: 10.4209/aaqr.2012.04.0094

Asbestos Imaging and Detection with Differential Interference Contrast Microscopy

Iuliia Zarubiieva1, Gi Byoung Hwang1, Ji Sok Lee1, Gwi-Nam Bae1, Yoo-Mi Oh2, Sun-Woo Park2, Taik Jin Lee1, Hyuk Jae Lee1, Deok Ha Woo1, Seok Lee1, Myoung-Ock Cho3, Jung Kyung Kim3, Seong Chan Jun4, Jae Hun Kim1*
1

Sensor System Research Center, Korea Institute of Science and Technology (KIST), Hwarangno 14-gil 5, Seongbuk-gu, Seoul 136-791, Korea 2 Department of Electrical and Computer Engineering, University of Seoul, Seoul 130-743, Korea 3 School of Mechanical Engineering, Kookmin University, Seoul 136-702, Korea 4 School of Mechanical Engineering, Yonsei University, Seoul 120-749, Korea ABSTRACT This study presents a new method for imaging and counting the concentration of asbestos fibers. This approach combines the principle of differential interference contrast (DIC) microscopy for imaging fibers with an imaging program to automatically assess their concentration. Asbestos is typically detected by optical microscopy or electron microscopy methods. While optical microscopy techniques (such as phase contrast microscopy and polarized light microscopy) are fast and inexpensive, they cannot ensure a thorough examination due to their low resolutions. In contrast, electron microscopy methods (such as transmission electron microscopy and scanning electron microscopy) provide high resolutions images, but are expensive and the related technology is not widely available. This work thus proposes the DIC method for detecting asbestos fibers, as it can overcome many of the disadvantages of existing methods. It also has good potential for use in portable measuring devices, which can detect asbestos right at the location of possible exposure. Keywords: Asbestos fibers; Differential Interference Contrast microscopy; Environmental monitoring.

INTRODUCTION Asbestos is used in a wide range of applications due to its desirable physical properties such as sound absorption, average tensile strength, and resistance to various harsh environmental conditions including fire, heat, electrical and chemical damage (Michaels and Chissick, 1979). However, it causes serious health problems to human beings. The influence of asbestos exposure on human health is one of the most widely studied subjects of modern epidemiology (Becklake, 1982; Doll and Peto, 1985; Higgson et al., 1992). Asbestosis is one of typical diseases generated by permanent asbestos exposure. Its severity depends on the amount of asbestos to which people are exposed and length of exposure time since their first exposure. Lung cancer caused by asbestos inhalation generally results from long-term high

Corresponding author. Tel.: +82-2-958-5782; Fax: +82-2-958-5709 E-mail address: jaekim@kist.re.kr

level exposures, and is usually associated with asbestosis. Another common disease is mesothelioma, which is a rare form of cancer of mesothelium (Whitwell et al., 1977). Previous studies showed that pathogenicity of airborne asbestos is more related to the number of long thin fibers than the total mass exposed (Davis and Jones, 1988, Lippmann, 1988, 1990). The hazard from airborne asbestos is the greatest in fiber of between 5 and 100 m in length with diameter less than 2 m, and aspect ratio more than 5:1 (Walton, 1982). Various individuals, who spend most of their time indoor, can be possibly exposed to airborne asbestos fibers. These include teachers, students, office workers, housekeepers or custodial employees who may come in contact with asbestos containing building materials (ACBM) or contaminated settled dust during their work activities, and construction workers who may disturb ACBM during repair or installation activities (US EPA, 2006). In order to safely protect human health from airborne asbestos, since 1980s, health organizations in the world such as World Health Organization (WHO), National Institute for Occupational Health and Safety (NIOSH) and Environmental Protection Agency (EPA) have made

1146

Zarubiieva et al., Aerosol and Air Quality Research, 13: 11451150, 2013

guidelines on the exposure limits of airborne fibers in workplace environments. According to International Labor Organization, the term respirable asbestos fibers means asbestos fibers with a diameter of less than 3 m and a length-to-diameter ratio greater than 3:1 (Safety in the use of asbestos: An ILO code of practice, 1984). Only fibers with a length greater than 5 m shall be taken into account for purposes of measurement. In different countries, exposure limits vary from 0.1 f/cc to 2 f/cc for 4- or 8-hour time-weighted-average exposure (Vitra, 2002). Currently, there are several techniques for imaging asbestos. In order to detect airborne asbestos, phase contrast microscopy (PCM) is commonly used; for detecting asbestos in soil - Polarized light microscopy (PLM). Transmission electron microscopy (TEM) and scanning electron microscopy (SEM) can be used in both cases (Perry, 2004). PCM is the most common method of imaging asbestos fibers even though it has a few disadvantages. Firstly, PCM cannot distinguish asbestos from other fiber-like particles such as gypsum, mineral wool, fiberglass, cellulose and others (Roggli et al., 2003). To do this, PLM is used. It can distinguish asbestos from non-asbestos, but does not give an image with good resolution. Secondly, phase contrast optics produce haloes around particles in the sample that can be often confused as fibers, which brings many inconveniences to an observer (Baron, 2001). Thirdly, poor contrast between the fibers and background image in PCM leads to the fact that many fibers are hardly distinguishable (Baron, 2001). TEM and SEM techniques have very high resolution (0.05 m for SEM and 0.0002 m for TEM), but are expensive ($50300 for SEM and $200600 for TEM on each measurement) and take a long time to get the image (VERSAR Inc, 1991). Also, they are not widely available, and require special skills from an operator. In this study, we find differential interference contrast (DIC) microscopy as an inexpensive, but fast method for possible usage as a portable measuring device. We combined the principle of DIC with an imaging program in order to image asbestos and count its concentration automatically. We suggest using proposed DIC method for detecting asbestos directly at the workplace as it is extremely important to protect workers and usual citizens from the risk of asbestos inhalation. For purpose of our investigation, we examined 3 different types of asbestos (amosite, crocidolite and chrysotile), which are most commercially used.

METHODS Principle of DIC Microscopy DIC is an optical illumination technique used to increase the contrast in transparent samples (Lang, 1968). It works on the principle of interferometry, as shown in Fig. 1. A light source (a), which is fiber optic light source from Nikon, is used. The beam initially polarized by 45 after passing through a polarizer (b) is spatially separated by the first Wollaston prism (c) into two beams that have orthogonal polarization to each other. These two beams are focused by the condenser (d) for passage through the sample (e). They are focused so they will pass through the sample with a very small distance to each other shear distance. Usually, it is slightly less than the resolution of the objective. As these two beams pass through a specimen, they experience different optical path lengths and refractive index changes since they target different spatial areas. After passing through the sample, the rays travel through the objective lens (f) and are focused for the second Wollaston prism (g), which removes the spatial separation between them. Then the analyzer (h) polarizes the beam at 135. Similar to an interferometer, now two beams can interfere to produce amplitude contrast. Therefore, a specimen such as asbestos can be imaged. The image with amplitude contrast is detected with a CCD array (i), which is STC-GEG152A from Sentech. Sample Preparation For the first part of experiment, 3 types of asbestos samples (amosite, crocidolite and crysotile), were prepared. 0.5 grams of each asbestos type were added to 40 milliliters of distilled water. Then, the solution was dispersed by an ultrasonic wave. A drop of each asbestos solution was placed on a 75 25 mm2 slide glass and covered with 18 18 mm2 cover glass. They were connected by epoxy. After drying the epoxy, the obtained samples were examined with DIC. For the second part of experiment, 0.5 grams of asbestos (amosite type) and 0.5 grams of milled glass fibers were added to 100 milliliters of distilled water. After that, the solution was dispersed by an ultrasonic wave. Slide glasses were prepared the same way, as in the first case. Asbestos Differentiation As it was mentioned before, Phase Contrast microscopy is the most common method of detecting asbestos and it

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

(i)

Fig. 1. Experimental DIC setup. a) Light source; b) Polarizer, c) Wollaston prism I (ICT-condenser prism K2), d) Condenser (Condenser head 0,90 S1), e) Specimen, f) Objective (HC PLFL 10x/0,30 f = 11 mm), g) Wollaston prism II (Objective prism-slider D1), h) Analyzer, i) Line CCD camera.

Zarubiieva et al., Aerosol and Air Quality Research, 13: 11451150, 2013

1147

has well-established analytical counting protocol (OSHA, NIOSH). However, the protocols do not consider other particles aside of asbestos that can be present in a sample (glass fibers, cellulose, etc.). That is why it is important to distinguish asbestos from other fiber-like particles. Using DIC technique can help solving this problem. Fig. 2 illustrates the image of asbestos and glass fibers, detected with the DIC microscopy. The differences in shape and structure, which can be clearly visible in this image, allow easy differentiation. Asbestos fibers have long and thin shape, while glass fibers are thicker and shorter. Also, there is quite a bit of variability within asbestos fibers, e.g., amphiboles versus chrysotile. Therefore, distinction among different asbestos fibers is also possible. Image Processing and Concentration Measurement For calculating the concentration, we used image processing and analysis program (ImageJ; http://imagej.nih. gov/ij/). It allows defining fibers from other particles (spots, dust etc.), determining aspect ratio (length-to-width ratio, > 3:1) and length (> 5 m considering the magnification) and automatically counts a number of particles that fit the requirements (Cho et al., 2011). Automatic fibers counting is made in a few steps. Firstly, to produce average background brightness, we subtracted the image background (Fig. 3(b)) from a raw image (Fig. 3(a)). We considered the contrast to decide the background, which is lighter in contrast than the fibers. Then, we chose the automatic threshold level for correcting background illumination (Fig. 3(c)). In this case, threshold level was chosen as Auto Local Threshold (Cho et al., 2011). After that, we convert the file to binary image, and eroded or dilated unnecessary pixels, which may be different for each image. And finally, we analyzed the image. We set the circularity range from 0 to 0.3 (definition of circularity range from (Cho et al., 2011)) and size range from 50 to 5000

because countable fibers can be only those with aspect ratio 3:1 and fiber length > 5 m according to asbestos regulation requirements (OSHA, NIOSH). Particles with size circularity values outside the range specified in this field are ignored. Summarization and display of the results are the last step of analysis (Fig. 3(d)). Also, it is possible to use ImageJ program for distinguishing asbestos with glass fibers. Fig. 4 illustrates the steps of analysis. Due to bigger thickness of glass fibers compared to asbestos, defining such parameters as circularity and particle size allows as separating needed asbestos fibers from others. Circularity range varies from 0 (infinity elongated object) to 1 (perfect circle). So, setting this parameter in a range between 0 and 0.3 allows picking only thin particles, i.e. asbestos. RESULTS AND DISCUSSION We analyzed 3 different types of asbestos with DIC method. This method provides several advantages compared to PCM. DIC uses polarization change according to the path difference between two orthogonally polarized lights while PCM uses diffraction method. Therefore, the contrast between fibers is much better in DIC images. The main advantage of DIC using polarization change is that there is no halo around the fibers edges so that images are clearly visible. Therefore, there is much less chance for other particles and unnecessary images to be confused as fibers. Fig. 5 shows DIC images of amosite, crocidolite and chrysotile, which are most common kinds of asbestos currently used. Usually, chrysotyle fibers are long and slightly curved while amosite and crocidolite are straight and shorter. The proposed DIC method can detect asbestos fibers and its images are used for counting the concentration. Also, automatic imaging program can be used with DIC images for their differentiation with other particles, like glass

Fig. 2. Images of asbestos and glass fibers, taken by DIC technique.

1148

Zarubiieva et al., Aerosol and Air Quality Research, 13: 11451150, 2013

Fig. 3. a) Raw image; b) Subtract the background; c) Auto threshold; d) Overlay mask (Analyze particles process).

Fig.4. Differentiating asbestos from glass fibers. a) Raw image, b) Subtract background, c) Auto threshold, d) Analyze particles results. fibers etc. Further improvement in image resolution can be achieved by using objectives with different magnifications. Some other types of fibers, like glass fibers or cellulose, can have similar shape as asbestos fibers (Muhle et al., 1997). However, comparison of asbestos and glass fibers in current research showed that those two kinds of fibers can be distinguished by optical characteristic differences. First, glass fibers are nearly transparent that reflection intensity caused by asbestos and glass fibers are different. Second, DIC technique can also distinguish asbestos from nonasbestos because it uses polarized light. Rotating the sample can cause some pleochroism in asbestos fibers while nonasbestos fibers will not show any image variation (Introduction to polarized light microscopy, Nikon; OSHA). CONCLUSIONS We suggested the method of differential interference contrast (DIC) microscopy for detecting and counting airborne asbestos fibers as an alternative to already existing methods. We compared the results of our researches with asbestos images made by different techniques and offered applying DIC method for making portable measuring device. Also images made by DIC can be easily analyzed by different automated counting techniques, which greatly reduce the calculation time. In addition, asbestos fibers can be distinguished from other particles, so the asbestos concentration will not be distorted by the presence of other fiber-like particles. Proposed DIC method can be a useful

Zarubiieva et al., Aerosol and Air Quality Research, 13: 11451150, 2013

1149

Fig. 5. Asbestos images obtained with DIC. a) Amosite; b) Crocidolite; c) Chrysotile. tool for on-place asbestos detection and concentration measurement, since sample preparation does not require much time or any special equipment; only slide glasses, cover glasses, pipette, mixing vessels and epoxy. It can be easily prepared outside of the laboratory. ACKNOWLEDGEMENTS This work was partially supported by the public welfare & safety research program (2010-0020796) and the Pioneer Research Center Program (2010-0019313) through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (MEST), Global Top Project (GT-11-E-02-002-1) funded by Korea Ministry of Environment, and IT R&D program of MKE/KEIT (10035610). REFERENCES Baron, P.A. (2001). Measurement of Airborne Fibers: A Review. Ind. Health 39: 3950. Becklake, M.R. (1982). Exposure to Asbestos and Human Disease. New Engl. J. Med. 306: 14801482. Cho, M.O., Yoon, S.H., Han, H.T. and Kim, J.K. (2011). Automated Counting of Airborne Asbestos Fibers by a High-Throughput Microscopy (HTM) Method. Sensors 11: 72317242 Davis, J.M.G. and Jones, A.D. (1988). Comparisons of the Pathogenicity of Long and Short Fibres of chrysotile Asbestos in Rats. Br. J. Exp. Pathol. 69: 7I7737. Doll, R. and Peto, J. (1985). Effects on Health of Exposure to Asbestos, HSE Books, London. Higginson, J., Muir, C.S. and Muos, N. (1992). Human Cancer: Epidemiology and Environmental Causes, Cambridge University Press, Cambridge, United Kingdom. Introduction to Polarized Light Microscopy, http://www. microscopyu.com/articles/polarized/polarizedintro.html Lang, W. (1968). Nomarski Differential InterferenceContrast Microscopy, Reprinted from Zeiss Information, No. 70, 16th, p. 114120. Lippmann M. (1988). Asbestos Exposure Indices. Environ. Res. 46: 86106 Lippmann, M. (1990). Effects of Fiber Characteristics on Lung Deposition, Retention, and Disease. Environ. Health Perspect. 88: 311317. Michaels, L. and Chissick, S.S. (1979). Asbestos: Properties, Applications, and Hazards, John Wiley & Sons, New York. Muhle, H., Ernst, H. and Bellmann, B. (1997). Investigation of the Durability of Cellulose Fibers in Rat Lungs. Ann. Occup. Hyg. 41: 184188. NIOSH: http://www.cdc.gov/niosh/docs/2003-154/. OSHA: http://www.osha.gov/dts/sltc/methods/inorganic/id 160/id160.html. Perry, A. (2004). A Discussion of Asbestos Detection Techniques for Air and Soil, U.S. Environmental Protection Agency, Office of Superfund Remediation and Technology Innovation, United States. Roggli, V.L., Oury, T.D. and Sporn, T.A. (2003). Pathology of Asbestos-Associated Diseases, 2nd ed., Springer, New York. Safety in the Use of Asbestos (1984). An ILO Code of

1150

Zarubiieva et al., Aerosol and Air Quality Research, 13: 11451150, 2013

Practice, International Labour Office, Geneva. Sampling and Analytical Methods (1992). Polarized Light Microscopy of Asbestos, 191, 18p, http://www.osha. gov/dts/sltc/methods/inorganic/id191/id191.html. US EPA (2006). Asbestos in Schools, Asbestos, (2006-0628) http://www.epa.gov/asbestos/pubs/ashome.html. VERSAR Inc. (1991). Asbestos Air Monitoring Results at Eleven Family Housing Areas throughout the United States, Final Report. Virta, R.L. (2002). Asbestos: Geology, Mineralogy, Mining, and Uses, U.S. Department of the Interior, U.S. Geological Survey, Open-File Report 02-149.

Walton, W.H. (1982). The Nature, Hazards and Assessment of Occupational Exposure to Airborne Asbestos Dust: A Review. Ann. Occup. Hyg. 25: 117119. Whitwell, F., Scott, J. and Grimshaw, M. (1977). Relationship between Occupations and Asbestos Fiber Content of the Lungs in Patients with Pleural Mesothelioma, Lung Cancer, and Other Diseases. Thorax 32: 377386. Received for review, April 20, 2012 Accepted, October 24, 2012