Dichotomous Samplers Modified for Use with Electron Microscopy

Large sulfate artifacts up to 2 μm in diameter were observed by scanning electron microscopy for the fine particle fraction collected in dichotomous samplers. The artifacts were attributed to small liquid particles that piled up on the filter, coalesced, and later dried as larger particles. Such artifacts were eliminated when particles were collected in a modified dichotomous sampler in which 80% to 90% of the airflow was diverted from the fine fraction filter. This airflow diversion technique was used successfully with both virtual-impactor and tandem-filter types of dichotomous samplers.     

Inlet characteristics of bioaerosol samplers

The inlet sampling characteristics of several commercial bioaerosol samplers operating in indoor and outdoor environments have been analyzed by use of available and newly developed equations for sampling efficiency. With a focus on the physical aspects of sampling efficiency, the aspiration and transmission efficiencies have been calculated for the bioaerosol particle size range 1–30 μm, which represents single bacteria, bacteria aggregates, bacteria carrying particles, fungal spores, yeast, and pollen. Under certain sampling conditions, the bioaerosol concentration was found to be significantly over- or underestimated. At wind velocities between 0 and 500 cm s⁻¹, calculations show that the AGI-30 would sample 1–10 μm particles with an inlet sampling efficiency of 20–100%. The entrance efficiency of the 6-stage Andersen viable sampler is 90–150% when sampling isoaxially with respect to horizontal aerosol flows, and 8–100% when oriented vertically at a right angle to the horizontal aerosol flow. For the Burkard portable air sampler, an even wider range of deviation may occur. The bioaerosol samplers used for large particles such as pollen are even less accurate: e.g. 10 times the ambient concentration of Lycopodium spores has been calculated to be aspirated by the Lanzoni sampler when operated at 0.5 1 min⁻¹ facing the wind at wind velocity of about 500 cm s⁻¹.

The actual bioaerosol concentration can be calculated from the measured data by use of the indicated procedures. The sampling efficiency graphs presented can be used to bracket the sampling conditions that enable the investigator to avoid or minimize significant sampling biases for each sampler. The findings can also be used for the design of new samplers or for improving commercially available samplers.     

Evaluation of total and inhalable samplers for the collection of carbon nanotube and carbon nanofiber aerosols

A growing number of carbon nanotubes and nanofibers (CNT/F) exposure and epidemiologic studies have utilized 25- and 37-mm open-faced cassettes (OFCs) to assess the inhalable aerosol fraction. It has been previously established that the 37-mm OFC undersamples particles >20?µm in diameter, but the size-selective characteristics of the 25-mm OFC have not yet been fully evaluated. This article describes an experimental study conducted to determine if the 25- and 37-mm OFCs performed with relative equivalence to a reference inhalable aerosol sampler when challenged with CNT/F particles. Side-by-side paired samples were collected within a small Venturi chamber using a 25-mm styrene OFC, 37-mm styrene OFC, 25-mm aluminum OFC, and button inhalable aerosol sampler. Three types of CNT/F materials and an Arizona road dust were used as challenge aerosols for the various sampler configurations. Repeated experiments were conducted for each sampler configuration and material. The OFC samplers operated at flow rates of 2 and 5?L/min. Results showed that the 25-mm OFC performed comparably to the button sampler when challenged with CNT/F aerosols, which was demonstrated in five of the six experimental scenarios with an average error of 21%. Overall, the results of this study indicate that the sampling efficiency of the 25- and 37-mm OFCs adequately followed the ISO/ACGIH/CEN inhalable sampling convention when challenged with CNT/F aerosols. Past exposure and epidemiologic studies that used these OFC samplers can directly compare their results to studies that have used other validated inhalable aerosol samplers.     

High-Volume Dichotomous Virtual Impactor for the Fractionation and Collection of Particles According to Aerodynamic Size

A prototype dichotomous virtual impactor (DVI) using a single acceleration nozzle, operating at approximately 500 1/min, and having an aerodynamic particle outpoint diameter of about 2–3 μm has been constructed and tested. Under these conditions the flow through the acceleration nozzle is calculated to be turbulent. This sampler was calibrated with a monodisperse aerosol, and the measured particle size-dependent collection efficiencies demonstrate that the sampler size fractionates atmospheric particulate matter as efficiently as the low-volume dichotomous virtual impactors. Analysis of test data indicates that the high-volume sampler can be described by classical impaction theory. These data also indicate that over the range of Reynolds numbers from 24,000 to 81,000 there is little, if any, dependence of inferred acceleration nozzle turbulence on the performance characteristics of the sampling system. A comparison of the concentration of atmospheric particulate matter, sulfate, and calcium on the fine filter samples collected with colocated high- and low-volume virtual impactors also shows that the two samplers are operating with similar performance characteristics. Additionally, the high-volume DVI collects at least 10–30 times the mass of particulate matter that the presently available virtual impactors collect and thus allows one to obtain improved precision in the measurement of those airborne species that are near the minimum detectable level of current analytical methods.     

Development of a new procedure for precise determination of viral aerosol lethal dose (ALD50) for birds

This paper describes a new method for precise determination of viral aerosol lethal dose (ALD₅₀) and reports the results obtained for birds exposed by H5N1 Avian influenza A strain under controlled laboratory conditions. The reported method utilizes our recently developed personal bioaerosol sampler capable of monitoring viable virus concentration in the ambient air. Up to six laboratory animals could be located in a specially designed aerosol chamber and exposed in parallel by an airborne strain of interest to generate the amount of data sufficient for representative statistical analysis. A concentration of viable airborne virus was measured by the personal aerosol samplers directly in the breathing zone of each particular bird. The results show a very low inter-sampler variation used for each particular run, which was confirmed by a single factor ANOVA test undertaken for all six personal samplers involved in each experiment. As was shown, the difference in amount of viruses collected by all samplers during each particular run was not statistically significant. It was found that the ALD₅₀ for approximately 400-g birds exposed to H5N1 Avian influenza A strain A/Chicken/Suzdalka/Nov-11/2005 was around 26.5 FFU (focus forming units).     

Measurements of the total and regional deposition of inhaled particles in the human respiratory tract

The total and regional deposition of monodisperse aerosols in the human respiratory tract has been measured in 12 healthy subjects breathing through the mouth. Radioactively labelled polystyrene particles in the aerodynamic diameter range 3.5–10.0 μm were employed. The total deposition results are similar to those reported by Stahlhofen et al. (1980), showing only a slight progressive increase with particle size, from a mean fraction of 0.79 of the inhaled aerosol at 3.5 μm, to 0.88 for 10 μm particles. The extrathoracic airways show a very marked deposition at all sizes, predominantly in the throat. The throat values rise rapidly from a mean of 0.09 at 3.5 μm to 0.36 at 10 μm particle diameter. Two intrathoracic fractions were also obtained by the widely accepted method of measuring the relative amounts of activity cleared from the thorax as a function of time. Alveolar deposition was apparently still some 0.06 of the inhaled aerosol at 10 μm particle diameter. Tracheo-bronchial deposition showed little change at any particle size except at 3.5 μm, when it was 0.24 of the inhaled aerosol.     

A wind tunnel evaluation of the physical sampling efficiencies of three bioaerosol samplers

The physical sampling efficiency of three commonly used samplers for bioaerosols has been determined under controlled conditions in a wind tunnel. Non-biological, monodisperse test aerosols of aerodynamic diameters up to 23 μm were used in a range of wind speeds up to 5 m s⁻¹. The performance of each sampler type was different. For the Andersen Microbial Sampler and the Casella Slit Sampler, sampling efficiency dropped both with increasing wind speed and particle size, while for the Aerojet General Glass Cyclone, performance was generally independent of windspeed and particle size. This work is part of a larger study to determine both the physical and the biological sampling efficiencies of currently used samplers for bioaerosols. The results highlight the importance of understanding the performance of aerosol monitoring equipment, if results obtained in the field are to be interpreted correctly.     

Ultrafine Particle Sampling with the UNC Passive Aerosol Sampler

A model is presented to describe the collection of ultrafine particles by the UNC passive aerosol sampler. In this model, particle deposition velocity is calculated as a function of particle size, shape and other properties, as well as a function of sampler geometry. To validate the model, deposition velocities were measured for ultrafine particles between 15 and 90 nm in diameter. Passive aerosol samplers were placed in a 1 m ³ test chamber and exposed to an ultrafine aerosol of ammonium fluorescein. SEM images of particles collected by the samplers were taken at 125 kX magnification. Experimental values of deposition velocity were then determined using data from these images and from concurrent measurements of particle concentration and size distribution taken with an SMPS. Deposition velocities from the model and from the experiments were compared and found to agree well. These results suggest that the deposition velocity model presented here can be used to extend the use of the UNC passive aerosol sampler into the ultrafine particle size region.     

Large Particle Size Distribution in Five U.S. Cities and the Effect on a New Ambient Participate Matter Standard (PM10)

A mobile aerosol-sampling system was used to determine the large particle ambient aerosol size distribution (up to approximately 100 μm particle diameter) in five cities across the United States: Birmingham, Alabama; Research Triangle Park, North Carolina; Philadelphia, Pennsylvania; Phoenix, Arizona; and Riverside, California. A mobile wide range aerosol classifier (WRAC) developed at the University of Florida was used. The study shows that any measurement of ambient particulate matter with a size-fractionating inlet sampler will be influenced by the ambient particle size distribution.

Mass distribution measurements determined by the WRAC were compared with mass measurements obtained simultaneously using TSP Hi-Vol and 15 μm cut-size inhalable particulate network samplers. Aerosol size-classification results showed the presence of a large particle mass mode at all sites sampled. The position and magnitude of the large particle mode varied and was not a simple function of concentration. The percentage of the total aerosol mass collected by the present EPA reference method high-volume air sampler varied from about 85 to 95%. The percentage of total aerosol mass less than 10 μm varied from about 50 to 90%, depending on the sampling location and sampling condition.     

Effect of Semivolatile Material on PM 2.5 Measurement by the PM 2.5 Federal Reference Method Sampler at Bakersfield,California

The chemical composition of fine particulate material was determined for samples collected in Bakersfield, CA, during February-March, 1998 using several diffusion denuder samplers, including the PC-BOSS, which measures both semivolatile fine particulate nitrate and organic material. An average of 56% of the fine particulate carbonaceous material was lost from the filters of the Particle Concentrator-Brigham Young University Organic Sampling System (PC-BOSS). A comparable amount of fine particulate semivolatile organic material was also lost from collected particles with single filter samplers, such as the PM 2.5 Federal Reference Method. The fraction of nitrate lost from collected particles was a function of temperature and humidity, with the biggest effect being due to temperature. The fraction of nitrate lost was comparable for conventional annular denuder samplers and the PM 2.5 FRM, averaging 33%. The nitrate loss from particles for the PC-BOSS was smaller, averaging 11%, possibly due to the concentration of particulate material prior to collection with this sampler. The loss of nitrate and semivolatile organic material during sample collection resulted in the PM 2.5 FRM sampler giving PM 2.5 mass that was an average of 30% (7.3 w g/m 3 ) lower than the true value and different from the true value from negligible to 20 w g/m 3 .     

Note on the movement of insecticide droplets through and out of a crop canopy after emission below canopy level

Coarse aerosol sprays with number median diameters in the range 30–40 μm were emitted at 0.25 m above the ground within the canopy of a cotton crop planted at 0.93 m row separation and standing 1.0 m tall. Movement of spray was monitored by vertical stepped paper tower samplers with narrowest section of width 0.13 cm, giving collection by inertial impaction down to 10 μm in a 1 m/sec wind. Horizontal paper targets were used to sample droplets depositing by sedimentation. Droplet number counts indicated that under conditions of moderate instability and associated convective turbulence, droplets were drawn rapidly out of the crop to a height of 2 m (the upper level sampled). The peak concentration of the spray cloud was centred at a height of between 1.0 and 2.0 m beyond 1.0 m downwind of the emission point in the first test, using droplet spectrum with volume median diameter of 41 μm, and beyond 2.5 m downwind of the emission point in the second test with spectrum of 57 μm volume median diameter. Inertial impaction at the 0.25 and 0.5 m levels was slight, except on the samplers within 0.5 m of the sprayer, and sedimentation was only significant at and below 0.5 m in the first two rows sampled. The moderate wind speed of 2.5 m/sec at canopy level produced marked inertial impaction over the first 4.5 m downwind of the sprayer at the 1.0 m level.     

An Aerosol Wind Tunnel for Evaluation of Massive-Flow Air Samplers and Calibration of Snow White Sampler

Massive-flow air samplers are being deployed around the world to collect aerosol samples for analysis of radioactivity as a result of nuclear tests and nuclear accidents. An aerosol wind tunnel capable of an 1100 m³ min^?1 flow rate was built at Lovelace Respiratory Research Institute (LRRI) to test the sampling efficiency of these samplers. This aerosol wind tunnel uses a stationary air blender to enhance mixing, and therefore it achieves the required uniform distribution of wind speed and aerosol concentration in the test section. The test section of the wind tunnel has a cross section that is 4.3 m × 3.7 m. The aerosol wind tunnel was tested for performance in terms of distribution of wind speed, turbulent intensity, SF₆ tracer gas concentration, and aerosol concentration. Test criteria consistent with U.S. Environmental Protection Agency (EPA) and American National Standards Institute (ANSI) standards were adopted as the guidelines for the aerosol wind tunnel. Additional criteria for aerosol wind tunnel were also recommended. Initial test of the aerosol wind tunnel showed that the wind tunnel could be operated in a wind speed range of 2 to 24 km h^?1. Within this range, the distribution of wind speed SF⁶ trace gas concentration and aerosol concentration in two-thirds of the central area of the test section showed coefficient of variances (COVs) of less than 10% for the range of wind speeds. This met the stringent guidelines for aerosol wind tunnel performance set by EPA and ANSI standards.

The LRRI wind tunnel was used to evaluate the collection efficiency of the sampling head of massive-volume air samplers, including the Snow White sampler. The sampler was tested in this aerosol wind tunnel for particles between 2 and 20 μm. The sampling flow rates were 500 and 700 m³ h^?1 for the tested wind speeds of 2.2 and 6.6 m S^?1, respectively. The results showed that sampling efficiency was influenced by both sampling flow rate and wind speed. The sampling efficiency decreased with an increase in particle size of between 2 and 20 μm. The sampling efficiency also decreased as the wind speed was increased from 2.2 to 6.6 m S^?1.     

Simulation of Particle Size-Selective Air Samplers Placed in a Polydisperse Aerosol

The results of a numerical simulation of several air sampling instruments are presented. They are assumed to sample the same aerosol, with a log-normal particle-size distribution. Four instruments were studied: the 10-mm nylon cyclone, the MRE 113A gravimetric sampler, the CPM 3, and the CIP 10. The experimental data of particle collection efficiency were reduced by a model for each instrument. The model used combines two cumulative log-normal distribution functions, in order to have a good degree of flexibility necessary for representing the data of some devices that exhibit a maximum in efficiency (CPM 3, CIP 10). The concentrations “measured” by several air samplers were compared with each other; the differences were analyzed as functions of the aerosol parameters: mass median aerodynamic diameter and σ_g. The results that were obtained and those calculated from standard collection efficiencies, defining the conventional alveolar fraction of the aerosol, were also taken into account. This simulation method can be extended to any type of instrument and aerosol, and enables the prediction of the maximal deviations that could be observed between different instruments, or between one instrument and some reference standards.     

Atmospheric Carbonaceous Species Measurement Methods Comparison Study: General Motors Results

The General Motors Research Laboratories participated in both the field sampling and round-robin portions of the Carbonaceous Species Intercomparison study that was held in Glendora, CA, during the summer of 1986. Five samplers were operated during the field study. The average particulate elemental carbon (EC) concentrations determined from the five samplers agreed to within 12%. Large differences were observed in the concentrations of particulate organic carbon (OC) determined from the five samplers. Some of the differences are attributed to losses of OC from the filters due to volatilization during the collection period. The amount of volatilization varies with the length of the sampling time and the filter face velocity. In addition, the adsorption of gas-phase organic compounds caused a significant positive interference in the determination of OC. Our OC and EC results for the round-robin samples were compared to the values obtained by the other participating laboratories. The average ratio of our results to the mean of the other laboratories was 0.97 for all the OC data and 1.23 for EC from all but the ambient wood-burning and organic aerosol samples. The ratios for the latter samples were 1.9 or greater. It is concluded that EC can be collected and analyzed with high precision; however, the accuracy of the measurements is unknown since standards for EC in atmospheric particulate do not exist.     

System design and test method for measuring respirator filter efficiency using mycobacterium aerosols

Recently, the protection of health care workers from tuberculosis-containing aerosols has been the subject of considerable debate. An experimental apparatus and test protocol were developed to measure the collection efficiency of surgical mask and respirator filter media using a microbial aerosol challenge. Mycobacterium chelonae (M. chelonae), used as a surrogate for Mycobacterium tuberculosis, was generated from liquid suspension using a Collison nebulizer. Upstream and downstream concentrations of viable aerosol particles were measured using Andersen cascade impactors, while total particle concentrations were measured with an aerodynamic particle sizer (APS). A monodisperse polystyrene latex (PSL) sphere aerosol (0.804 μm) was used in separate experiments to measure filter efficiency; concentrations were determined with the APS. The mycobacterial aerosol ranged in size from 0.65 to 2.2 μm when measured with the cascade impactor. A similar size range was found with the APS, yielding a count median diameter of about 0.8 μm. Samples of the mycobacterial aerosol were collected on glass slides, stained M. chelonae, as determined by environmental scanning electron microscope, were found to be rod shaped with an average length of 2 μm and average width of 0.3 μm. To evaluate the apparatus over a range of filter efficiencies (10–100%), different layers of fiberglass filter paper were tested for penetration using a 0.12 μm dioctyl phthalate (DOP) aerosol measured with a light scattering photometer, in addition to the mycobacterial and PSL aerosols. For the range of efficiencies tested it was shown that filter collection of DOP was linearly related to that of both mycobacterial and PSL sphere aerosols (r² = 0.99), demonstrating that an inert aerosol may be used to predict the collection of biological aerosols by such filter media.     

Aerosol Aspiration Efficiency of Blunt and Thin-Walled Samplers at Different Wind Orientations

The measured aerosol aspiration efficiency of a thin-walled probe and 5 blunt (thick-walled) samplers was used to compare the aspiration efficiency calculated using the models developed by Belyaev and Levin (1972, 1974) and by Vincent and his colleague (1987, 1995). At 0° wind direction, the model developed by Belyaev and Levin agrees quite well with the experimental value for both the thin-walled probe and the blunt samplers. In this case, the blunt sampler body diameter ( D b ) was used to replace the inlet diameter ( D i ) for the calculation of the Stokes number (St). At 90° and 180° wind directions, the model developed by Vincent and his colleague agrees fairly well with the experimental value. This study shows that the model developed by Belyaev and Levin, at 0° wind direction, can be modified to calculate the aerosol aspiration efficiency by using personal samplers as area samplers. For the model developed by Vincent and his colleague, in addition to personal sampler application, at 90° and 180° wind directions their model can also be used to estimate the aspiration efficiency of area samplers.     

Comparative study of elemental mass size distributions in urban atmospheric aerosol

Elemental mass size distributions in aerosols collected at four different urban sites with gradually increasing overall aerosol mass concentration are presented, compared and discussed in the present paper. The aerosol samples were collected with cascade impactor and stacked filter unit samplers, and were analyzed by particle-induced X-ray emission spectrometry and instrumental neutron activation analysis. Typical coarse-mode elements, i.e., Na, Mg, Al, Si, P, Ca, Ti, Fe, Ga, Sr, Zr, Mo and Ba, exhibited unimodal size distributions at all four urban locations studied, and the mass median aerodynamic diameters were increased with aerosol pollution level. Elements typically related to high-temperature or anthropogenic sources, i.e., S, Cl, K, V, Cr, Mn, Ni, Cu, Zn, Ge, As, Se, Br, Rb and Pb, either had a unimodal size distribution with most or their mass in the fine size fraction or clearly showed a bimodal size distribution at the urban background site. However, significant differences between the size distributions of four sampling sites were noted. There was a clear tendency for the accumulation mode to decrease and for the coarse mode to increase with increasing total aerosol mass concentration. A pronounced resuspension of the soil and roadway dust associated with the fine aerosol particles that were deposited on the ground surface previously, and the condensation process of volatile precursor gases on the already existing aerosol particles can explain the observed tendencies. The elemental mass size distributions derived for the polluted urban environments differ from those typically observed for industrial, combustion or automotive sources. A consequence of the diversity in the size distributions on the PM_2.5 speciation concept is also presented.     

Investigation of inherent and latent internal losses in liquid-based bioaerosol samplers

Previous studies have analyzed collection efficiencies and reaerosolization rates of liquid-based bioaerosol collectors. However, these studies did not analyze latent internal losses in the samplers, i.e., the fraction of the particles that is aerosolized from the collection liquid, attach to the samplers’ inner surfaces and do not leave the sampler via outlet. Here, we investigated the internal losses and reaerosolization rates in BioSampler (SKC Inc., Eighty Four, PA) and AGI-30 (Ace Glass Inc., Vineland, NJ) bioaerosol collectors operated with different amounts of liquid as a function of particle type (polystyrene latex particles of 0.9 and 3.2 μm, B. subtilis bacteria and C. cladosporioides fungal spores), particle concentration (“low” and “high” differing by a factor of 100) and operating time (15 and 30 min). The samplers were filled with sterile deionized water containing known (reference) particle concentrations and were operated in a particle-free atmosphere for 15 or 30 min. The overall particle loss was determined by comparing the concentration of particles remaining in the collection liquid with the reference concentration. The reaerosolization rates were estimated by comparing the concentration of particles collected at each sampler's outlet with the reference concentration. The internal loss was determined as a fraction of particles remaining attached to the inner walls of a sampler, i.e., particles not in the collection liquid and not reaerosolized. All the investigated variables had a statistically significant effect on the overall particle loss, the reaerosolization rate and the internal loss. Averaged for all test conditions, the internal loss for BioSampler with 5 mL liquid, BioSampler with 20 mL liquid and AGI-30 was 37.7%, 29.6% and 22.5%, respectively. The observed reaerosolization rates were rather low and ranged from 0.2% to 6.9%. This study shows that depending on a particular set of sampling conditions a substantial fraction of already collected particles could leave the collection fluid, attach to the inner surfaces of the samplers and not be available for sample analysis thus affecting the accuracy of bioaerosol investigations.     

Aspiration Efficiency of IOM-Like Personal Aerosol Samplers from Experiments with a New Rapid Data Acquisition System

New automated instrumentation for the rapid acquisition of aerosol sampler aspiration efficiency data has been applied to an investigation of a range of personal aerosol samplers of the type developed during the 1980s at the Institute of Occupational Medicine (IOM) in Edinburgh, Scotland, U.K. The experimental research was carried out in a small wind tunnel, and the relation of the results for IOM-like samplers to full-scale life-size personal aerosol sampling scenarios—like those encountered in occupational aerosol exposure assessment—was investigated by reference to the scaling laws that have been developed based on familiar aerosol mechanics as they apply to the physics of aerosol sampling. In the small-scale experimental study, the IOM-like sampler was mounted centrally on a rectangular bluff body, simulating the wearing of the sampler on the body (e.g., as by a worker in an industrial setting). Scaling with respect to the corresponding, more-realistic full-scale system for a corresponding full-scale windspeed of 1.0 m/s was achieved by varying the inlet diameter, the windspeed and the sampling flowrate. The results for windspeeds in the scaled experiments of 1.5 m/s and lower were found to differ significantly from those for windspeeds of 2.0 m/s and higher. In particular, the measured aspiration efficiency values for the lower windspeeds were markedly higher than—and clearly not consistent with—the higher windspeed group of results. It is considered likely that such divergence may be associated with a characteristic of the small wind tunnel in which the experiments were conducted. However, the scaling laws developed were found to work well for windspeeds in the scaled experiments of 1.5 m/s and higher. The results confirm that the performance of the IOM personal inhalable aerosol sampler is in quite good general agreement with the inhalability criterion.     

Calibration,Intersampler Comparison,and Field Application of a New PM-10 Personal Air-Sampling Impactor

The MS&T personal impactor is designed to provide a particle size cut (d _50%) of 10 μm in aerodynamic diameter (PM-10) at a flow rate of 4 L/min. Data are presented that verify the designed particulate mass cut specifications of this impactor for personal sampling. These data are derived from three different analyses', laboratory calibration, intersampler comparison, and field application. Laboratory calibration using monodispersed liquid aerosol shows a sharp 10-μm particle cut size.

The performance of the personal impactor was tested using ambient and combustion-generated aerosols. Established PM-10 samplers (the Sierra/Andersen dichotomous and the MS&T indoor air sampler impactor) were run side by side with the personal impactor. The intra- and intersampler vari-abilities in PM-10 measurements were evaluated. Results showed good precision among personal impactors (CV = 3.2%). The PM-10 sampled by the personal impactor was found to be highly correlated with measurements made with the indoor air sampler impactor (r ² = 0.99) and the dichotomous sampler (r¹ = 0.97).

The impactor was subsequently employed for personal air sampling in the Total Human Environmental Exposure Study (THEES). The THEES sampling protocol entailed 24-hour sampling during a 14-day study interval. THEES field measurements included indoor, outdoor, and personal PM-10 samples. The personal impactor measurements for 13 participants were predicted by a time-weighted exposure model using indoor and outdoor PM-10 and specific activity variables (p < 0.01).