Abstract
Particle size classifiers must be used for analyzing PM2.5. Particle rebound may occur when using a traditional impactor or cyclones. The use of virtual impactors would prevent particle rebounds. However, typical virtual impactors have complicated structures, and are operated at a relatively low flow rate in many cases. In this study, a simple single-nozzle middle-volume PM2.5 virtual impactor was developed, and its performance evaluation was carried out. The classification characteristics of the VI developed in this study were similar to those of WINS, which was a USEPA-approved PM2.5 inlet, but the 50% cut-point of the VI was lower than that of WINS. Changes in the minor flow rate and the distance between the inlet and collection nozzles did not significantly affect the separation characteristics of the VI. On the contrary, a wider inlet nozzle diameter produced higher 50% cut-points of the VI. The mass and chemical compositions of PM2.5 collected using the VI were compared to those collected using WINS. The concentrations of several elements in the particles collected using the VI were different from those collected using WINS due to the difference in the classification characteristics of the VI and WINS. However, the PM2.5 mass obtained using these two devices were similar to each other.