samplers as freestanding devices if it can be shown there is no effect from the mannequin body [16]. However, if the sampler is to be used at higher windspeeds, which frequently occur outdoors, personal aerosol samplers should be evaluated on a mannequin in a wind tunnel [16]. Respirable aerosol samplers generally do not have problems with inlet effects because the particles being sampled have low enough inertia and settling velocity. However, Cecala et al. [6] found that the 10-mm nylon cyclone operated free-standing in a wind tunnel oversampled at external air velocities greater than 4 m/s when the inlet faced the wind and undersampled at 90/ and 180/ to the wind at velocities greater than 1 m/s. The maximum sampling error of about 40% was observed at 10 m/s. It is expected that these errors would be reduced if the sampler were located on a person, because the air velocity decreases near the body surface. It should be noted that the Cecala et al. work was conducted in the context of aerosol sampling in underground mining environments. Here, windspeeds of the magnitude quoted are not uncommon. However, in industrial workplaces more generally, windspeeds are much lower. Two surveys of a wide range of workplaces [17,18] revealed that actual indoor windspeeds rarely exceeded 0.2 to 0.3 m/s and more typically were less than 0.1 m/s. The EPA PM-10 standard for environmental sampling specifies a sampler that has a 50% cutpoint at 10.6 :m, approximately the same as that for the thoracic sampler [19]. Although the requirements for environmental PM-10 samplers stipulate wind tunnel testing, similar work on personal PM-10 and thoracic samplers has yet to be performed. It is expected that these samplers will be more susceptible to wind effects than respirable samplers because larger particles are more susceptible to inertial and gravitational effects. The most extensive comparison of available inhalable aerosol samplers was that carried out under the auspices of the European Commission (EC) [9]. Eight samplers were tested: CIP-10 (foam-based, French); 37-mm closed-face cassette (Spain and US) (Figure 2a); 37-mm openface cassette (Sweden) (Figure 2a); PAS-6 (Netherlands); PERSPEC (Italy); GSP (Germany, sold as CIS sampler in US; BGI, Inc. Waltham MA); IOM (United Kingdom; (Figure 2b); and the Seven-Hole Sampler (United Kingdom; Casella CEL, Inc., UK) (Figure 2c). Conditions of the experiment included measurement of sampler collection efficiencies on a mannequin for aerosol particles with diameters as large as 100 :m at a wind speed of 0.5 m/s, 1.0 m/s, and 4 m/s. Samplers were positioned on a mannequin rotating within a wind tunnel. The samplers were all conductive; the 37-mm cassette samplers were painted with an external conductive coating. The aerosol was, however, not neutralized. The results of this experiment indicated high inter-sampler variability, but permitted estimates of bias relative to the inhalable convention. The EC study also indicated that most samplers work reasonably well at low wind speeds (<1 m/s) for particle median diameters below 25 :m [20]. The study indicated that experiments of this type were difficult, expensive, and generally had poor precision. Perhaps better understanding of the flow field near the body may lead the way to improved and simplified sampler testing. Recent work suggests ways of making the wind tunnel testing of inhalable samplers simpler and less expensive, e.g., by using a compact body to simulate the chest of a mannequin [21] and by using miniaturized mannequins and samplers that are calculated to be aerodynamically equivalent [22]. The orientation and diameter of an inhalable sampler inlet may affect the collection of very large particles (generally >100 µm), since these may be thrown into the inlet as projectiles. The current definition of inhalable aerosol only covers particles up to 100-:m aerodynamic diameter.
3/15/03
186
NIOSH Manual of Analytical Methods
