This article was downloaded by: [Washburn University]On: 02 November 2014, At: 05:20Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH,UK

Aerosol Science and TechnologyPublication details, including instructions for authorsand subscription information:http://www.tandfonline.com/loi/uast20

Evaluation of Microbial Samplersfor Bacterial MicroorganismsChih-Shan LiPublished online: 30 Nov 2010.

To cite this article: Chih-Shan Li (1999) Evaluation of Microbial Samplers forBacterial Microorganisms, Aerosol Science and Technology, 30:2, 100-108, DOI:10.1080/027868299304705

To link to this article: http://dx.doi.org/10.1080/027868299304705

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all theinformation (the “Content”) contained in the publications on our platform.However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, orsuitability for any purpose of the Content. Any opinions and views expressedin this publication are the opinions and views of the authors, and are not theviews of or endorsed by Taylor & Francis. The accuracy of the Content shouldnot be relied upon and should be independently verified with primary sourcesof information. Taylor and Francis shall not be liable for any losses, actions,claims, proceedings, demands, costs, expenses, damages, and other liabilitieswhatsoever or howsoever caused arising directly or indirectly in connectionwith, in relation to or arising out of the use of the Content.

This article may be used for research, teaching, and private study purposes.Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly

forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Dow

nloa

ded

by [

Was

hbur

n U

nive

rsity

] at

05:

20 0

2 N

ovem

ber

2014

Aerosol Science and Technology 30:100–108 (1999)cs 1999 American Association for Aerosol Research

Published by Taylor and Francis0278-6826/99 $12.00 + .00

Evaluation of Microbial Samplers for BacterialMicroorganisms

Chih-Shan Li * , Mei-Ling Hao, Wen-Hai Lin, Ching-Wen Chang,and Chiu-Sen Wang

GRADUATE INSTITUTE OF ENVIRONMENTAL HEALTH, COLLEGE OF PUBLIC HEALTH,NATIONAL TAIWAN UNIVERSITY, TAIPEI, TAIWAN, R.O.C. (C.-S.L., M.-L.H., W.-H. L., ANDC.-W.S.), INSTITUTE OF OCCUPATIONAL SAFETY AND HEALTH, COUNCIL OF LABOR,

TAIPEI, TAIWAN, R.O.C. (C.-S.W.)

ABSTRACT. The in� uences of physical factors, sampling time, and � ow rates on to-tal recovery rate of airborne bacteria were evaluated in a chamber by three samplingmethods: AGI-30 all-glass impingers, Nuclepore � ltration and elution method, andgelatin � lters. The in� uence of the biological factor, microorganism hardiness, wasassessed by using sensitive Escherichia coli (E. coli) and hardy Bacillus subtilis (B. sub-tilis) spores. It was found that colony or relative survival of sensitive E. coli was muchlower than those of hardy B. subtilis spores by the evaluated three sampling methods.Sampling � ow rate did not have a signi� cant in� uence on the bacterial recovery by thethree sampling methods. However, the relative survival of sensitive E. coli collected onNuclepore and gelatin � lters became lower as the sampling time increased because ofdehydration effects. For hardy B. subtilis spores, it was demonstrated that recoveryby gelatin � lter was comparable to that by impinger. In conclusion, it was found thatimpingers could perform much better than � ltration methods for sampling airbornebacterial bioaerosols.

INTRODUCTIONCurrently, bioaerosol characterization has be-come an important issue because of their re-lated health effects. Therefore, it is neces-sary to assess composition and concentrationof airborne microorganisms in contaminatedindoor/outdoor or occupational environments(Henningson and Ahlberg 1994). In general,� eld evaluations in the natural environment arelikely to provide very limited information withuncontrolled or unknown natural aerosol con-

* Corresponding author.

centrations, particle sizes, and � ora (Henning-son and Ahlberg 1994; Nevalainen et al. 1993).Thus, it is necessary to use a test aerosol in cham-ber to assess bioaerosol sampler performance(Grif� ths and DeCosemo 1994).

The overall sampling ef� ciency of bioaerosolsamplers with different designs may differ sig-ni� cantly from one another because of the differ-ent physical collection ef� ciency and the stressimparted to the microorganisms. The selectionof sampler, microorganism hardiness, samplingtime, and sampling � ow rate are considered tobe the most important factors to affect micro-bial collection and survival in bioaerosol sam-

Dow

nloa

ded

by [

Was

hbur

n U

nive

rsity

] at

05:

20 0

2 N

ovem

ber

2014

Aerosol Science and Technology30:2 February 1999

Evaluation of Samplers for Bacteria 101

plers (Macher and Willeke 1992; Nevalainen etal. 1993).

In this study, the performance of the com-monly used bioaerosol sampling methods AGI-30 impingers, Nuclepore � ltration and elutionmethod, and gelatin � lters in highly contami-nated environments was evaluated. In addition,the in� uences of sampling time and � ow rates,microorganism hardiness, and aerosol concen-trations on the recovery of the viable bacterialaerosols were also investigated.

MATERIALS AND METHODS

Test Microorganisms

Cultures used in this evaluation included E. coliand B. subtilis from Graduate Institute of Mi-crobiology, College of Medicine, National Tai-wan University. The recovery of gram-negativeand nonspore-forming E. coli, representative ofa sensitive bacterial strain, could be used to as-sess the effect of the microorganism hardinesson sampler performance in comparison with thatof gram-positive, endospore-forming, and hardyB. subtilis.

An active E. coli culture was inoculated intoapproximately 10 TSA (trypticase soy agar,Difco) plates and incubated for 24 h at 37°C(Jensen et al. 1992). The colonies were lateraseptically washed to a 15-ml sterile coni-cal centrifuge tube (SARSTEDT, Germany),capped, and centrifuged at 2,500 rpm for 5 min.For B. subtilis, the cells were initially inoculatedon 10 TSA plates for sporulation at 37°C for 7days. Bacterial growth was harvested into ster-ile distilled water, agitated at 45 rpm for morethan 24 h at room temperature, and heated for10 min in an 80°C water bath to kill vegeta-tive cells. Microscopic examination of the sus-pension of B. subtilis demonstrated motile, sin-gle cells with an endospore intact and a minuteamount of cell debris. The resulting spore sus-pension was centrifuged at 3,500 rpm for 5 min.The supernatants of both strains were discardedand the pellets were resuspended in sterile dis-

tilled water. This washing process was repeatedtwo more times and the cells/spores were thenresuspended in approximately 70 ml of steriledistilled water for high-concentration genera-tion. The resuspended solution was diluted 100times for medium-concentration generation.

Aerosol Generation and Test system

The test system was a modi� cation of the sys-tem previously developed by Thompson et al.(1994). The test chamber is 12.5 cm in diame-ter with a height of 27 cm. A Collison three-jetnebulizer (BGI Inc., Waltham, MA) was used fornebulization of the microbial E. coli and B. sub-tilis suspension at 3 L/min of dry, � ltered, andcompressed laboratory air, then passed througha Kr-85 particle charge neutralizer (model 3077,TSI). The aerosolized suspension was then di-luted with � ltered and compressed air at 50L/min.

The Aerosizer, an aerodynamic particle sizer(Model 8000, API Inc., Hadley, MA) was usedto determine real-time number concentration(C0, #/m3) and size distribution of bacterialaerosol throughout the run in order to determinetotal recovery (TR, Ctest/C0) of E. coli and B.subtilis by the test samplers (Ctest, CFU/m3).The Aerosizer is capable of measuring parti-cles larger than 0.5 m m at an air� ow rate of6 L/min (Qian et al. 1995). To determine therelative survival (RS, Ctest/Cref), the evaluatedsampler was placed in the test chamber togetherwith the reference sampler, the AGI-30 impingerat a � ow rate of 12.5 L/min (Cref, CFU/m3),which was widely applied in � eld measurement.During the tests, the bacterial concentrationswere maintained at 106 (medium) and 108 (high)particles/m3, which represented the medium tohighly contaminated environments, for at least3 h. The entire system was placed in a safetycabinet to eliminate the uncollected bioaerosolsby the samplers. During the course of the ex-periments, the temperature was in the range of20–24°C and relative humidity was maintainedat 60%.

Dow

nloa

ded

by [

Was

hbur

n U

nive

rsity

] at

05:

20 0

2 N

ovem

ber

2014

102 C.-S. Li et al. Aerosol Science and Technology30:2 February 1999

Test Sampler

The AGI-30 (Ace Glass Inc., Vineland, NJ) isan all-glass impinger with 30 mm jet-to-platedistance. It was developed from the Porton im-pinger for sampling microorganisms (May andHarper 1957) and proposed as a reference sam-pler (Brachman et al. 1964). Twenty ml of steriledeionized water with 1% peptone, 0.01% Tween80, and 0.005% antifoam A (Sigma ChemicalCo., St. Louis, MO) was put into each auto-claved AGI-30 impinger (Thorne et al. 1992).The antifoam was added to reduce foaming andto prevent excessive � uid loss (Lembke et al.1981).

A Nuclepore � lter (Costar, Cambridge, MA)is a track-etched polycarbonate � lter consist-ing of a polycarbonate membrane with straight-through pores of a uniform size (0.01–14 m mpore size). In this study, � lters with 0.4 m mpore size and 37 mm diameter supported by cel-lulose pads were loaded into open-face, three-piece plastic cassettes. Filters and support padswere autoclaved and plastic cassettes were ster-ilized with ethylene oxide before sampling.

The gelatin � lter (Sartorius, Gottingen, Ger-many) has been specially designed for detectionand analysis of airborne microbes. Each sterilegelatin � lter (3.0 m m pore size, 80 mm diameter)was placed in a sterile � lter holder by carefullyletting the � lter slide out of the pocket onto the� lter support of the aluminum � lter holder.

Sampling time and � ow rates

To evaluate the in� uence of sampling time onthe collection of viable airborne bacterial par-ticulates, the selected sampling times included1, 2.5, 5, 15, 30, 45, and 60 min. For the AGI-30impinger, the sampling � ow rates consisted of4.7, 7.1, 9.4, 12.5, and 14 L/min and the cor-responding velocities through the nozzle were100, 150, 200, 265, and 300 m/s, respectively.Regarding Nuclepore � lters, the four sampling� ow rates were selected to be 1, 1.5, 2, and 4L/min and the corresponding sampling veloci-ties were 2, 3, 4, and 8 cm/s, respectively. In ad-

dition, the four air � ow rates used for the gelatin� lters were 10, 15, 20, and 30 L/min (4, 6, 8, and12 cm/s), which were widely used by industrialhygienists and matched with the sampling ve-locity selection for track-etched polycarbonate� ltration.

Sample processing

After sampling, the bacterial particles on the� lters were removed. The Nuclepore sampleswere eluted by injecting 1 ml of � ltered and ster-ile water with 0.1% peptone and 0.01% Tween80 into the support pad and 5 ml of the samemedium onto the � lter surface (Thorne et al.1992). The cassettes were recapped and vor-texed in rotator (Vortex-2 Genie, G-560, Scien-ti� c Industries Inc.) for 30 s. The gelatin � lterwas moved carefully into a 90 mm petri dish anddissolved in 10 ml of sterile deionized water ina 40°C water bath for 5–15 min. The suspen-sions from both � lters were then vortexed andduplicate 0.2 ml samples of serial 10-fold dilu-tion were plated onto TSA. The agar plates wereincubated for 24 h at 37°C.

Colony forming unit (CFU) counting wasdone on plates containing between 30 and 300colonies (Thorne et al. 1992). Airborne bacte-rial concentrations were determined by multi-plying the CFU by the dilution factor and by theelute volume and dividing by the volume of se-rial dilution material plated (0.2 ml) and the vol-ume of sampled air which was calculated fromsampling time and � ow rate.

RESULTS AND DISCUSSION

Characteristics of aerosolized E. coliand B. subtilis spores

The aerodynamic particle diameter of the targetaerosol is one of the most important physicalfactors which determine the stage collection ef-� ciencies of inertial and � ltrationdevices. Itwasfound that geometric mean aerodynamic diam-eter (GMAD) of E. coli and B. subtilis spores

Dow

nloa

ded

by [

Was

hbur

n U

nive

rsity

] at

05:

20 0

2 N

ovem

ber

2014

Aerosol Science and Technology30:2 February 1999

Evaluation of Samplers for Bacteria 103

were in the range of 0.7–0.9 m m with geomet-ric standard deviation (GSD) of 1.2. In addi-tion, the distributions of CFUs collected by theAndersen six-stage sampler for E. coli and B.subtilis spores indicated that more than 98% ofthe recovered colonies were # 2.1 m m (data notshown).

In addition to size characteristics, the viabil-ity of the generated bacteria is another importantfactor to affect sampling performance. In ourcurrent investigation, it was observed that the vi-ability of prepared B. subtilis spores was lowerthan that of prepared E. coli in the suspension so-lution by evaluating the ratio of Csusp (CFU/ml insuspension) to Co (as shown in Table 1). The vi-ability of the prepared bacterial strains dependson preparation process, which might vary fromone to another investigation. Therefore, TRscould not be a good indicator for sampler bioef-� ciency comparison among different bacterialstrains. It will be better to consider the bac-terial viability in the liquid suspension, whichwas used as the source of the generated bacterialbioaerosols. It is recommended that Ctest/Csusp

(colony survival, CS) rather than TR could beused as an indicator to evaluate sampler perfor-mance among bacteria with different viability.

Bioef� ciency of test samplers

AGI-30 impinger. Bioef� ciency analysis atdifferent sampling times and � ow rates indicatedthat the average values of CS for sensitive E. coliwere observed to be 0.053, 0.039, 0.041, 0.047,0.041, 0.045, and 0.04 at the sampling times of 1,

2.5, 5, 15, 30, 45, and 60 min, respectively. Forhardy B. subtilis, the average values of CS (%)were 0.144, 0.134, 0.124, 0.117, 0.095, 0.101,and 0.104 at the sampling times of 1, 2.5, 5, 15,30, 45, and 60 min, respectively (as shown inFigure 1). It was demonstrated that sampling� ow rates and time and aerosol concentrationsdid not have a signi� cant in� uence on the CSperformance of the impinger. However, bacte-rial hardiness was observed to play an impor-tant role in impinger collection. From the pre-vious report (Lin et al. 1997), it was found thatboth sampling time and � ow rates can affect thecollection ef� ciency, bubble pattern (Grinshpunet al. 1997), and the culturability of collectedbacteria for the AGI-30 impinger. Concerningsampling time, it was indicated that collectionef� ciency of 1.06 m m PSL did not change signif-icantly throughout the 60 min sampling period(Lin et al. 1997). Regarding sampling � ow rates,the calculated cutoff diameter decreases and col-lection ef� ciency increases as � ow rate increases(Nevalainen et al. 1992, 1993). Another impor-tant issue with which we are concerned was theevaporation and reaerosolization effect. In thecurrent evaluation, an antifoam agent was addedto reduce the evaporation rate of the collectionsolution. In addition, a wetting agent, Tween80, was added into the collection solution forenchancing the bacterial suspension during im-pingement. However, the effect of evaporationand reaerosolization on collection ef� ciency ofthe impinger still needs further evaluation.

In the current investigation, it was observedthat the sampling � ow rate did not signi� cantly

TABLE 1. Comparison of viability of prepared E. coli and B. subtilis in suspension solutions

E. coli B. subtilis

Aerosol Concentration Medium High Medium High

Suspension Concentration 8.09 1590 1.38 143(Csusp , 106 CFU/ml)Aerosol Concentration 2.02 149 2.24 112(C0 , 106 #/m3)Csusp /C0 4.00 10.7 0.62 1.28(CFU/ml/#/m3 )

Dow

nloa

ded

by [

Was

hbur

n U

nive

rsity

] at

05:

20 0

2 N

ovem

ber

2014

104 C.-S. Li et al. Aerosol Science and Technology30:2 February 1999

FIGURE 1. The average colony survival (CS: CFU/m3/CFU/ml, CFU/m3 concentration divided by CFU/ml in sus-pension) of AGI-30 for E. coli and B. subtilis at different sampling times and � ow rates. In the upper � gure, each errorbar represents one standard deviation on the mean of � ve different sampling � ow rates. In the lower � gure, each errorbar represents one standard deviation on the mean of seven different sampling times.

in� uence the biological recovery of the collectedbacteria. This should be related to a highersampling � ow rate resulting in higher biolog-

ical stress, which affects the bacterial survival.For AGI-30 impingers, the biological stress dur-ing sampling consists of the impinging action

Dow

nloa

ded

by [

Was

hbur

n U

nive

rsity

] at

05:

20 0

2 N

ovem

ber

2014

Aerosol Science and Technology30:2 February 1999

Evaluation of Samplers for Bacteria 105

through the nozzle of the inlet tube, impactiononto the bottom of the collection bottle, and con-tact effect with collection solution (Terzieva etal. 1996). The biological stress during impinge-ment was found to be so destructive that the re-coveries of sensitive E. coli were much lowerthan those of hardy B. subtilis.

Nuclepore � lter. The average values of RS (%)for sensitive E. coli were observed to be 55%,44%, 43%, 18%, 10%, 9%, and 5% at the sam-pling times of 1, 2.5, 5, 15, 30, 45, and 60 min,respectively. For hardy B. subtilis, the averagevalues of RS (%) were 64%, 51%, 51%, 47%,48%, 47%, and 42% at the sampling time of 1,2.5, 5, 15, 30, 45, and 60 min, respectively (asshown in Figure 2). To be compared with theperformance of the reference AGI-30 impinger,RSs of Nuclepore � lter were found to be lowerthan 100%. In addition, it was clearly demon-strated that � ow rates did not signi� cantly in-� uence the performance of Nuclepore � lters forcollecting bacterial bioaerosols. Furthermore,the recoveries of hardy B. subtilis spores weredemonstrated to be higher at high rather thanmedium concentrations. In addition, samplingtime was indicated to play a role in RS (from55% to 5%) determination of sensitive E. colirecovery rather than hardy B. subtilis (50%).

For the � lter, the particle diameter of bacterialaerosols was much larger than the 0.4 mm poresize and the bacterial penetration through � ltershould be negligible. Therefore, it can be indi-cated that RS values are primarily related to thebiological stress during � ltration. The biolog-ical stress during � ltration includes impactionand dehydration effects caused by the large vol-ume of air over the collected bioaerosols (Jensenet al. 1992; Crook 1995). In the current evalu-ation, sampling time prolongation was found toreduce RS for sensitive E. coli, but not for B.subtilis. It was demonstrated that dehydrationduring sampling did signi� cantly affect viabil-ity of sensitive bacteria which are not as resistantto desiccation as hardy ones.

Gelatin � lter. The average RS (%) values ofgelatin � lters for E. coli were observed to be4%, 7%, 4%, 3%, 1%, 0.2%, and 0.2% at thesampling times of 1, 2.5, 5, 15, 30, 45, and 60min, respectively. Moreover, the average RS(%) values of B. subtilis were observed to be128%, 93%, 84%, 78%, 108%, 88%, and 72%at the sampling times of 1, 2.5, 5, 15, 30, 45, and60 min, respectively (as shown in Figure 3). Itwas demonstrated that the recovery of B. subtilisby gelatin � lter was as good as that by impinger.However, the recovery of E. coli by gelatin � l-ter was found to be close to zero, which agreedwith those reported by Macher and First (1984).Similarly to the performance of the Nuclepore� lter, the gelatin � lter performance was not sig-ni� cantly in� uenced by sampling � ow rates.

The collection ef� ciency has been observedto be 99.9% for particles of 0.5–3 m m despitethe fact that the � lter pore size is as large as 3m m (Rotter and Koller 1973) and is compara-ble to that of cellulose acetate membrane � lters(Macher and First 1984). Therefore, the in� u-ence of physical factors (such as sampling � owrate and time) on collection ef� ciency of gelatin� lters should be negligible.

Generally speaking, gelatin � lters are consid-ered to be unsatisfactory for collecting airbornebacteria because during extended sampling thegelatin dried out, placing additional dehydra-tion stresses on the collected microorganisms(Crook 1995). In the current evaluation, sam-pling time was still found to affect RS of gelatin� lter, more for sensitive E. coli than for hardy B.subtilis spores. Our data clearly demonstratedthat the hardy spores could be retrieved morereadily than the sensitive strains on gelatin andNucelpore � lters.

The physical collection ef� ciency of � ltrationwas higher than that of impingers for collectingbacterial bioaerosols. Therefore, the observedlower RS values of � ltration compared to im-pingement should be associated with the higherbiological stress by � ltration, which was relatedto dehydration stress.

Dow

nloa

ded

by [

Was

hbur

n U

nive

rsity

] at

05:

20 0

2 N

ovem

ber

2014

106 C.-S. Li et al. Aerosol Science and Technology30:2 February 1999

FIGURE 2. The average relative survival (RS in %: CFU/m3 concentration divided by CFU/m3 concentration mea-sured simultaneously by an AGI-30 at a � ow rate of 12.5 L/min) of Nucelpore � lters for E. coli and B. subtilis atdifferent sampling times and � ow rates. In the upper � gure, each error bar represents one standard deviation on themean of four different sampling � ow rates. In the lower � gure, each error bar represents one standard deviation onthe mean of seven different sampling times.

CONCLUSIONSThe observed lower level of total bacterial re-covery on � lter as compared to impinger is pri-marily due to the higher biological stress duringthe sampling process of � ltration. Furthermore,

the sampling � ow rate did not have a signi� -cant in� uence on the bacterial recovery. Sam-pling time related to dehydration effect did playa role on the bacterial recoveries, especially forthe sensitive strain. In summary, impingers are

Dow

nloa

ded

by [

Was

hbur

n U

nive

rsity

] at

05:

20 0

2 N

ovem

ber

2014

Aerosol Science and Technology30:2 February 1999

Evaluation of Samplers for Bacteria 107

FIGURE 3. The average relative survival (RS in %: CFU/m3 concentration divided by CFU/m3 concentration mea-sured simultaneously by an AGI-30 at a � ow rate of 12.5 L/min) of gelatin � lters for E. coli and B. subtilis at differentsampling times and � ow rates. In the upper � gure, each error bar represents one standard deviation on the mean offour different sampling � ow rates. In the lower � gure, each error bar represents one standard deviation on the meanof seven different sampling times.

likely to perform better than � ltration methodsfor sampling airborne bacterial microorganisms.

This work was supported by grant IOSH85-H301 from theInstitute of Occupational Safety and Health, Council of La-bor, Republic of China.

ReferencesBrachman, P. S., Ehrich, R., Eichenwald, H. F.,

Gabelli, V. J., Kethley, T. W., Madin, S. H., Malt-man, J. R., Middlebrook, G., Morton, J. D., Silver,I. H., and Wolfe, E. K. (1964). Standard Samplerfor Assay of Airborne Microorganisms, Science144:1295.

Crook, B. (1995) Non-Inertial Samplers: BiologicalPerspectives. In Bioaerosols Handbook, edited by

Dow

nloa

ded

by [

Was

hbur

n U

nive

rsity

] at

05:

20 0

2 N

ovem

ber

2014

108 C.-S. Li et al. Aerosol Science and Technology30:2 February 1999

C. S. Cox and C. M. Wathes. CRC Lewis Publish-ers, Boca Raton, FL, pp. 269–283.

Grif� ths, W. D., and DeCosemo, G. A. L. (1994). TheAssessment of Bioaerosols: A Critical Review, J.Aerosol Sci. 25:1425–1458.

Grinshpun, S. A., Willeke, K., Ulevicius, V.,Juozaitis, A., Terzieva, S., Donnelly, J., Stelma,G. N., and Brenner, K. P. (1997). Effect of Im-paction, Bounce, and Reaerosolization on the Col-lection Ef� ciency of Impingers, Aerosol Sci. Tech-nol. 26:326–342.

Henningson, E. W., and Ahlberg, M. S. (1994). Eval-uation of Microbiological Aerosol Samplers: AReview, J. Aerosol Sci. 25:1459–1492.

Jensen, P. J., Todd, W. F., David, G. N., Scarpino,P. V. (1992). Evaluation of Eight Bioaerosol Sam-plers Challenged with Aerosols of Free Bacteria,Am. Ind. Hyg. Assoc. J. 53:660–667.

Lembke, L. L., Kniseley, R. N., Van Mostrand, R. C.,and Hale, M. D. (1981). Precision of the All-GlassImpinger and the Andersen Microbial Impactor forAir Sampling in Solid-Waste Handling Facilities,Appl. Environ. Microbiol. 42:222–225.

Lin, X., Willeke, K., Ulevicius, V., and Grinshpun,S. A. (1997). Effect of Sampling Time on the Col-lection Ef� ciency of All-Glass Impingers, Am. Ind.Hyg. Assoc. J. 58:480–488.

Macher, J. M., and First, M. W. (1984). PersonalAir Samplers for Measuring Occupational Expo-sures to Biological Hazards, Am. Ind. Hyg. Assoc.J. 45:76–83.

Macher, J. M., and Willeke, K. (1992). PerformanceCriteria for Bioaerosol Samplers, J. Aerosol Sci.23:S647–S650.

May, K. R., and Harper, G. J. (1957). The Ef� ciencyof Various Liquid Impinger Samplers in BacterialAerosols, Brit. J. Ind. Med. 14:287–297.

Nevalainen, A., Pastuszka, J., Liebhaber, F., andWilleke, K. (1992). Performance of BioaerosolSamplers: Collection Characteristics and Sam-pler Design Considerations, Atmos. Environ.26A:531–540.

Nevalainen, A., Willeke, K., Liebhaber, F., Pas-tuszka, J., Burge, H., and Henningson, E. (1993).Bioaerosol Sampling. In Aerosol Measurement:Principles, Techniques, and Applications, editedby K. Willeke and P. A. Baron. Van NostrandReinhold, New York, pp. 471–492.

Rotter, M., and Koller, W. (1973). Sampling of Air-borne Bacteria by Gelatin-foam-� lters, Zbl. Bakt.Hyg., I. Abt. Orig. B. 157:257–270.

Terzieva, S., Donnelly, J., Ullevicius, V., Grinshpun,S. A., Willeke, K., Stelma, G. N., and Brenner,K. P. (1996). Comparison of Methods for Detec-tion and Enumeration of Airborne MicroorganismsCollected by Liquid Impingement, Appl. Environ.Microbiol. 62:2264–2272.

Thompson, M. W., Donnelly, J., Grinshpun, S. A.,Juozaitis, A., and Willeke, K. (1994). Method andTest System for Evalaution of Bioaerosol Sampler,J. Aerosol Sci. 25:1579–1594.

Thorne, P. S., Kiekhaefer, M. S., Whitten, P., andDonham, K. J. (1992). Comparison of BioaerosolSampling Methods in Barns Housing Swine, Appl.Environ. Microbiol. 58:2543–2551.

Received November 17, 1997; accepted September14, 1998.

Dow

nloa

ded

by [

Was

hbur

n U

nive

rsity

] at

05:

20 0

2 N

ovem

ber

2014