Field tests were conducted to evaluate the PM2.5 sizing performance of preseparators for the NILU filter holder system. The PM2.5 mass concentrations measured by a TEOM with the preseparator was higher by 9% than those measured by another TEOM with a United States Environmental Protection Agency reference inlet for PM2.5 on average. This overestimation was probably due to the re-suspension of large particles. Another field test on the chemical composition of particles showed that daily concentrations of sulfate and ammonium in PM2.5 measured by a denuded filter-pack sampling system with the preseparator well corresponded to those obtained by a low-pressure cascade impactor, the reference method. The re-suspension seemed to affect the PM2.5 nitrate concentration, but it was difficult to estimate because the nitrate concentration by the cascade impactor was likely influenced by the artifact. The re-suspension can be avoided by cascading the preseparators.
