Classification of particles from the farm environment by automatedsizing, counting and chemical characterisation with scanning electronmicroscopy-energy dispersive spectroscopy†

Asbjørn Skogstad,* Lene Madsø and Wijnand Eduard

National Institute of Occupational Health, P.O. Box 8149 Dep, 0033 Oslo, Norway.E-mail: asbjorn.skogstad@stami.no; Fax: +47 23 19 52 06; Tel: +47 23 19 53 25

Received 13th April 1999, Accepted 28th May 1999

About 60 000 particles in 288 aerosol samples collected during farm work have been characterised with automatedparticle analysis using scanning electron microscopy (SEM) and energy dispersive X-ray spectrometry (EDS).Based on EDS-analysis of materials with known composition (potato flour, a-quartz, K-feldspar andb-wollastonite), criteria were developed for classification of particles as: (1) organic, (2) silicon-rich (silica), and(3) other inorganic particles. The reproducibility of the relative mass proportions in dust samples collected duringfarm work was 0.078 when approximately 200 particles were characterised per sample. Field samples from the farmenvironment showed clear differences in composition. Generally, inorganic particles dominated the particle mass.The proportion of the organic particle mass was highest for tending of swine and poultry, 55 and 38% respectively.Silica particles amounted to 10 to 20% of the total mass during handling crops, e.g. grain, straw, hay, potatoes, andonions. It seems likely that the results can be used in etiologic studies, but further validation would be needed forquantitative purposes.

automated particle analysis program for SEM-EDS was usedAim of investigationin the farmer study. Such instrumentation has been commer-

In a study of farmers’ respiratory health and exposure to cially available for several years. However, a literature searchbioaerosols (Melbostad et al.1), exposure to organic dust and revealed only a few studies (Hoflich3 and Gunst4) on inorganicsilica was considered to be of interest. Measurement of these particles collected in the work environment and no studies oncomponents by ashing and X-ray diffraction or infrared spec- organic particles or particles from the farm environment. Thetrometry was not feasible as sample sizes were expected to be present paper therefore describes classification of particles bysmall, typically <0.5 mg. A method based on chemical charac- chemical type in samples collected during different types ofterisation of particles by combined scanning electron farm work.microscopy and energy dispersive spectroscopy (SEM-EDS)was therefore considered.

Materials and methodsThe analytical scanning electron microscope can provideboth chemical and morphological data of individual particles. SamplingIt is frequently used for qualitative characterisation of aerosols,

The farmers carried battery powered pumps (AFC 123, Casellae.g., in the work environment. However, particle mass esti-Ltd., London, UK) connected to a closed-faced, graphite-mation by microscopy is inaccurate. Particles with surfacefilled polypropylene filter holder with a 50 mm extension tubestructure may not be recognised as single particles but counted(Gelman 25 mm Air Monitoring Cassette, Gelman Sciences,as several particles with smaller size. The recognition ofAnn Arbor, MI, USA) loaded with a polycarbonate filterindividual particles in complex aggregates can be difficult and(Poretics, Osmonics, Livermore, CA, USA) with pore sizeparticles may be obscured. The volume of a particle is estimated0.4 mm. The airflow through the filters was approximatelyfrom the microscopic dimensions by assuming that the particle1 L min−1 which was measured with a calibrated Brooks 2-65height is half of the minimum observed diameter and the massMM rotameter (Emerson Electric Co., Hatfield, PA, USA).is estimated from the estimated volume and the assumedSampling time was dependent on the work task, and variedparticle density. If these errors are systematic and comparablebetween 2 to 120 min.for all particle types their influence will be reduced when the

relative composition of particle types is estimated and thecombined mass is measured by another method, e.g., gra- Preparationvimetry. Finally, the counting process induces an error as a

The particle mass on the filters was measured by gravimetrylimited number of the particles are analysed. The latter errorusing an ultra-microbalance (Satorius model S4, Goettingen,and other random errors decrease if the number of analysedGermany) with a detection limit of 3 mg. Samples were resus-particles is increased. Therefore 1000 to 5000 particles havepended in filtered (<0.45 mm) Tween 80 solution (0.05% w/v)been characterised in samples from outdoor air (Hanna et al.,2by treatment for 3 min in an ultrasonic bath with a frequencyand personal communication with the author).of 35 kHz and a high-frequency power of 225 W (SonorexAs manual counting would be extremely time consuming anRK510H, Bandelin Electric, Berlin, Germany). Sub-volumesof the particle suspension containing 20–40 mg were filteredthrough a second polycarbonate filter with pore size 0.4 mm†Presented at AIRMON ’99, Geilo, Norway, February 10–14, 1999.

J. Environ. Monit., 1999, 1, 379–382 379

Publ

ishe

d on

01

Janu

ary

1999

. Dow

nloa

ded

on 2

7/10

/201

4 09

:02:

27.

View Article Online / Journal Homepage / Table of Contents for this issue

using a funnel with a 15 mm internal diameter. This step was fractions were done in Microsoft Excel for Windows 95version 7.0.necessary to achieve an appropriate density of particles on the

filter without overlap and aggregates were also dispersed to The precision was estimated by one-way analysis of variance(ANOVA) of samples that had been analysed twice. Thesome extent. Specimens of 8 mm×8 mm were cut from the

filter and gently fixed on carbon specimen stubs with double- composition of the dust during different work operations wascompared by one-way ANOVA and by the Kruskal-Wallissided carbon adhesive discs. In addition spots of carbon

cement were applied at each corner of the specimen to secure test on the proportions of the chemical types. Statisticalanalyses were carried out with SystatA 8.0 (SPSS Inc., Chicago,good conductivity to earth. Finally the sample was coated

with approximately 30 nm carbon film in a sputter coater IL, USA).(Balzers SCD 050, Balzers, Liechtenstein) equipped with acarbon thread flashing unit. Results

Analytical procedureAnalysisInorganic particles and most organic particles were observedThe specimens were analysed with a Jeol (Akishima, Tokyo,with good contrast in the backscattered electron image mode.Japan) JSM-6400 scanning electron microscope (SEM) con-However, some organic particles appeared smaller in size ornected to a Series II energy dispersive X-ray spectrometerwere counted as several smaller particles, Fig. 1A and B.(EDS)-system (NORAN Instruments, formerly Tracor

Northern Middleton, WI, USA) equipped with a thin window Classification criteriadetector allowing detection of elements with atomic number

As the main focus of the project was to study organic dust�6 (carbon). The automated analysis was performed usingexposure, classification into three categories was consideredthe software package ‘Particle Recognition andsufficient for our purpose: 1, organic, 2, silicon-rich, andCharacterisation’ (PRC, NORAN Instruments Inc.,3, other inorganic particles. To define the particle types,Middleton, WI, USA) run on the Series II EDS-system. Aclassification criteria were developed from PRC analysis of amodified version of the PRC program (made locally atparticle mixture of quartz, feldspar, wollastonite and starch.NORAN Instruments B.V., Naarden, The Netherlands) wasFig. 2 shows the number of particles (total number 363) as aused which was similar to a previous version of the programfunction of silicon and carbon content. Since specimens were(Fritz et al.5) where the spectrum was acquired while theflashed with carbon, the silicon counts were corrected forelectron beam was scanning over 8 rotational cords throughcarbon counts by the formulathe centroid of the particle. X-ray counts were more reproduc-

ible with this acquisition mode as chemical inhomogeneitiesSicorr=

SicountTotalcount−Ccount

and geometric and absorption effects are averaged over theentire particle during the scan (Johnson6).

The SEM was operated under the following conditions:accelerating voltage 20 keV, working distance 8 mm, electronprobe current ca. 3 nA, solid-state backscattered electrondetection, magnifications 500× for particles 1–5 mm and 100×for particles >5 mm, and X-ray acquisition time 6 s per par-ticle. The elements selected for chemical characterisation were:C, Na, Mg, Al, Si, P, S, K, Ca, Ti, Cr, Mn, Fe, Cu and Zn.The total net X-ray counts varied between 1000 and 5000. Thenet peak area for a given element was expressed as thepercentage of the total X-ray counts.

Classification criteria

Criteria for classification of particles were based on PRCanalysis data of well defined particle types: K-feldspar(Norfloat NGP 325), a-quartz (Fyle-quartz), b-wollastoniteand potato flour (commercial domestic quality), prepared andanalysed following the same protocol as the field samples.

Quantification of field samples

Particle volume was estimated using particle area×smallestdiameter/2. A minimum of two frames per magnification wereanalysed. The analysis was terminated when 100 particles or4 frames per magnification had been counted. The particlemass was estimated assuming that the particle density ofinorganic particles, including quartz, was 3 times the densityof organic particles. The proportions of three particle typeswere computed by division by the total estimated mass.

Data analysis

The particle data were transferred in ASCII format from theFig. 1 Scanning electron micrographs of organic particles from theEDS-system to a PC using the terminal-emulator programfarm environment (poultry house) as seen in (A), secondary electronProcomm Plus 3.0 (Datastorm Technologies, Inc., Columbia,image mode, SEI, and (B), backscattered electron image mode, BEI.MO, USA). A tailor-made Basic-program further made the Particles which are underestimated are indicated by arrows.

data accessible for Microsoft-Excel. Computations of the mass

380 J. Environ. Monit., 1999, 1, 379–382

Publ

ishe

d on

01

Janu

ary

1999

. Dow

nloa

ded

on 2

7/10

/201

4 09

:02:

27.

View Article Online

combinations as the independent variable, Table 1. The overallprecision of the proportions was 0.078. The precision ofobservations where both replicate analyses had found pro-portions <0.25 or >0.75 was 0.045, whereas the precision forobservations with at least one proportion between 0.25 and0.75 was 0.10. The difference was statistically significant,F36,39=5.14 and p<0.001. There was a tendency to a betterprecision when more than 200 particles were classified (arith-metic mean 270) compared to samples with less than 200classified particles (arithmetic mean 127), 0.071 and 0.090respectively, F24,51=1.61 and p=0.08.

Characterisation of field samples

Results from the classification of organic, silica and otherinorganic particles in samples from different farm work areFig. 2 Distributions of relative X-ray counts from carbon (CcountM )shown in Table 2. One-way ANOVA showed that the pro-and silicon corrected for carbon (SicorrU ) of 363 particles from a

mixture of potato starch, quartz, feldspar and wollastonite. portions between tasks were significantly different for allparticle types, F213,9=4.4 to 15.8, p=0.01 to <0.001. However,the conditions for ANOVA are not satisfied as the proportionsThe particles were classified by the following criteria. Firstwere not normally distributed and the variances were notwere particles with Ccount�80 classified as organic. Of thehomogeneous. The non-parametric Kruskal-Wallis test wasremaining particles were particles with Sicorr>90 classified astherefore also used, which also showed that the differencesquartz. The following particles were classified as other inor-between tasks were significant for all particle types, p<0.001.ganic particles: 1, particles ∏5 mm with Sicorr>35 and

Sicorr∏90; 2, particles >5 mm with Sicorr>40 and Sicorr∏90;and 3, particles with at least one element count >10 except Discussioncarbon. When these criteria were applied approximately 65%

Detection of particles with the PRC program relies on a highof the particles were classified. The remaining particles werecontrast, preferably binary image, with the particles as whiteclassified as organic particles as the Sicorr and the counts ofstructures against a black background. As backscattered elec-single inorganic elements were too low for positive recognitiontron production is strongly correlated with atomic number,as an inorganic particle. Of these particles 60–80% hadinorganic particles are easily recognised using the signal fromCcount�70 and 90–96% had Sicount∏10 for particles >5 andthe back-scattered electron detector. Organic particles have a∏5 mm respectively.high content of light elements and their average atomic numberis close to that of the filter. It can therefore be difficult toReproducibilityobtain sufficient contrast between organic particles and back-ground. A satisfactory contrast was achieved by fixing theTwenty five specimens were reanalysed. A total of 50 to 500

particles were classified per analysis. The standard deviation backside of the filter to the specimen stub using double-sidedcarbon adhesive discs. Problems with static electric chargingwas estimated by one-way ANOVA with sample-particle typeof the specimen were absent and a smooth background witha good signal-to-noise ratio was obtained. Although it is likelyTable 1 One-way ANOVA of replicate counting of 25 samples col-that some particles are not correctly sized or recognised aslected during farm work. The precision was estimated from the

residual variance two or more smaller particles, the underestimation of organicparticles was minimised by the improved preparation

Source of variance df Mean square F-test p technique.The PRC program classifies particles with low X-ray countsBetween all combinations of

as non-integratable and does not store the element counts forsample and particle type 74 0.142 23.5 %0.001such particles. As non-integratable particles from the farmResidual 75 0.00604aenvironment are likely to be of organic origin, carbon was

aPrecision=√0.00604=0.078.included in the X-ray acquisition. Very few non-integratable

Table 2 Composition of the dust that farmers are exposed to during different tasks

Particle type

Organic Silica Other inorganic

Task Na AMb Median sc AM Median s AM Median s

Threshing 21 0.26 0.24 0.21 0.17 0.15 0.12 0.58 0.56 0.28Handling of hay 28 0.18 0.14 0.15 0.12 0.08 0.13 0.71 0.79 0.20Handling of silage 15 0.12 0.08 0.10 0.06 0.03 0.07 0.82 0.87 0.13Sorting of potatoes 20 0.02 0.01 0.02 0.11 0.10 0.06 0.88 0.88 0.06Sorting of onions 16 0.03 0.02 0.02 0.12 0.11 0.03 0.85 0.86 0.04Tending of—

Dairy cows and cattle 23 0.32 0.31 0.20 0.08 0.02 0.14 0.61 0.62 0.20Sheep/goats 26 0.22 0.14 0.27 0.15 0.10 0.15 0.64 0.71 0.28Poultry 31 0.39 0.33 0.26 0.02 0.01 0.03 0.60 0.64 0.25Swine 25 0.54 0.51 0.15 0.07 0.05 0.07 0.40 0.40 0.14

Handling of liquid manure 18 0.16 0.07 0.23 0.15 0.11 0.16 0.70 0.80 0.23

aSample size. bArithmetic mean. cSample standard deviation.

J. Environ. Monit., 1999, 1, 379–382 381

Publ

ishe

d on

01

Janu

ary

1999

. Dow

nloa

ded

on 2

7/10

/201

4 09

:02:

27.

View Article Online

particles were observed thereafter. Analysis of the test materials Large differences between the proportions of the particletypes during different tasks were found. Except for tending ofshowed that potato starch particles could clearly be classified

on the basis of the carbon count. Silicon counts were primarily swine and poultry, the proportion of organic particles waslow. Although many work operations in farming involveused for chemical classification of particles. Inorganic particles

showed substantial and variable carbon counts due to the handling of organic materials, the generated dust is dominatedby inorganic particles. A possible explanation is that thecarbon coating which complicated the classification. The

carbon count was therefore subtracted from the total X-ray surface of farm materials like grain and hay is contaminatedwith particles from the soil which can settle from air or depositcount after which quartz could be clearly distinguished from

the silicates wollastonite and feldspar (Fig. 2). As the total by rain-splash. During handling of farm materials particles onthe surface are more likely to be dispersed into the air thanX-ray count excluding carbon could be very small for organic

particles, the relative silicon count could be high even with a the organic substrate itself. The soil as a source of inorganiccontamination is also plausible since the mineral soils whichlow silicon count. Therefore carbon counts >80% of the total

count were used as a primary criterion for classification of dominate in the region of the study have an organic contentwhich is often less than 5% by weight (Singh et al.7). Theorganic particles. About 35% of the particles that were not

classified as silica or inorganic were finally classified as organic silica content of the dust varied from 2 to 17% which alsomay point at soil as an important source of dust as soil isas they were relatively high in carbon and low in silicon. It is

possible, however, that some of these particles were not organic probably the only likely source of silica dust in the farmenvironment.or were aggregates of both organic and inorganic particles.

The proportion of organic particles may therefore have beenoverestimated. However, this error is counteracted by under- Conclusionsestimation of size and number of some organic particles.

Accuracy was further studied by the reproducibility of the An automated method for sizing, counting and chemicalcharacterisation of particles collected in the farm environmentestimated proportions of the particle types. The overall stan-

dard deviation was 0.078 when on average 200 particles were by scanning electron microscopy-energy dispersive spec-troscopy was developed. It was possible to classify particles asanalysed per analysis. The precision was sufficient for the

comparison of field samples from the farm environment where organic, silicon-rich (silica) and other inorganic particles.Large differences in composition of the airborne dust werehighly significant differences between tasks were found. The

standard deviation was smaller in samples where a larger observed during different types of farm work. Most of thedust was inorganic and probably due to contamination fromnumber of particles had been analysed as should be expected.

However, a much larger number of particles need to be the soil. It therefore seems likely that the results can be usedin etiologic studies, but further validation is needed for quanti-classified to improve the precision substantially. Hanna et al.2

mention ranges of 5–11% and 26–37% for the proportions of tative purposes.two chemical particle types in samples collected in outdoorair. These data suggest standard deviations of 0.015 to 0.03 Referenceswhich were found when 1000 or 5000 particles where classifiedper analysis (personal communication with the author). Low 1 E. Melbostad, W. Eduard and P. Magnus, Scand. J. Work Environ.

Health, 1997, 23, 271.(<0.25) and high (0.75) proportions were estimated with2 R. B. Hanna, K. J. Karcich and D. L. Johnson, Scanning Electronbetter precision which is similar to the binomial distribution.

Microsc, 1980, 1, 323.The results probably have a bias due to many assumptions 3 B. Hoflich, Diploma Thesis, Technical University of Darmstadt,that have been made in the quantification. In addition soluble 1997, (in German).particles as salts and sugars were lost as samples needed to be 4 S. Gunst, Diploma Thesis, Technical University of Darmstadt, 1997,

(in German).resuspended to obtain an optimal particle density on the SEM5 G. S. Fritz, J. J. McCarthy and R. J. Lee, in Proceedings of the 16thspecimen. If the bias does not depend on the type of farm

Annual Conference of the Microbeam Analysis Society, ed.work, i.e., particle types are similar, the results may still beR. H. Geiss, San Francisco Press, CA, 1981, pp. 57–60.useful in etiologic research. As the farmer study aims at 6 D. L. Johnson, Scanning Electron Microsc, 1983, 3, 1211.

identifying the most important components in airborne dust 7 B. R. Singh, T. Børresen, G. Uhlen and E. Ekeberg, in Managementfor respiratory effects among farmers, further validation of of Carbon Sequestration in Soil, CRC Press, Boca Raton, New

York, 1988, pp. 195–208.the method, e.g., by comparison to standard methods, waspostponed until more was known about the importance oforganic dust. Paper 9/02941I

382 J. Environ. Monit., 1999, 1, 379–382

Publ

ishe

d on

01

Janu

ary

1999

. Dow

nloa

ded

on 2

7/10

/201

4 09

:02:

27.

View Article Online