Evaluation of a personal and microenvironmental aerosol speciation sampler (PMASS)

In this study, an all-aluminum sampler was reconfigured to a lighter weight by using plastic for structural components not contacted by aerosol. Because the sampler body was made of plastic, the aluminum size-selective cyclone inlet was redesigned into a separate, removable unit. The resulting personal and microenvironmental aerosol speciation sampler (PMASS*) is a small, compact sampler designed for assaying concentrations of each major chemical constituent of particulate matter less than or equal to 2.5 pm in aerodynamic diameter (PM2.5) as well as its mass. The sampler consists of a miniature cyclone inlet and two parallel sampling channels, each of which may accommodate a denuder, particle filter, and backup filter. One channel serves to measure mass and inorganic ions; the other, organic and elemental carbon. Six of these samplers were first evaluated in the laboratory to verify the 50% cutpoint of the redesigned cyclone inlet near the target flow rate of 4.0 L/min and to test the effect of sampler orientation, aerosol type, and flow rate on the 50% cutpoint. The six samplers were fabricated at two different times with slight modifications in construction that yielded slightly different 50% cutpoints. The units were then evaluated as microenvironmental samplers under field conditions against reference samplers in Fresno, California, and Baltimore, Maryland. Testing in Maryland also included deployment as a personal sampler. In the laboratory, the mean (+/- SD) 50% cutpoint at a 4.1 L/min flow rate was 2.52 +/- 0.08 pm. Sampler orientation did not affect performance of the cyclone. Small differences in penetration efficiency were found with different challenge aerosols. For flow rates between 3.1 and 4.9 L/min, the 50% cutpoint varied from 3.3 pm to 2.1 pm, respectively. As a microenvironmental sampler, the PMASS performed well against relevant reference samplers for elemental carbon, organic carbon, and nitrate measurements but not for mass measurements, for which the PMASS showed a significant bias of -34%. Precision estimates for PMASS elemental carbon, organic carbon, nitrate, and mass, respectively, were 6.8%, 9.0%, 11.9%, and 12.7% in Fresno and 6.6%, 8.7%, 13.6%, and 17.5% in Baltimore. For personal sampling, PMASS measurements of mass again showed a significant bias of -24% compared with the reference sampler. Precision estimates for the PMASS as a personal sampler for elemental carbon, organic carbon, nitrate, and mass were 8.1%, 19.3%, 19.1% and 24.2%, respectively. Sulfate data were not analyzable because of the large magnitude and variability of the sulfate blanks. Field testing revealed some deformation of components of the PMASS filter pack. These problems may have affected the integrity of the particle filter.