Measurements of relative humidity-dependent bounce and density for atmospheric particles using the DMA-impactor technique

The dependence of particle bounce on relative humidity in an inertial impactor was measured for atmospheric aerosols in Meadview, AZ, and Minneapolis, MN. An aerosol composed of 0.25 μm atmospheric particles, selected by a differential mobility analyser, was drawn through a single stage of a MOUDI impactor having an aerodynamic cut point of 0.112 μm. The impactor collection efficiency was determined by using two condensation nucleus counters to simultaneously count particles upstream and downstream of the impactor. Bounce measurements were done using uncoated aluminum substrates, and provision was made to control relative humidity of the sampled aerosol. We found that for both locations bounce increased sharply as relative humidities decreased below 60-70%, but little bounce occurred when the relative humidity was maintained above 70 or 80%. Because errors caused by particle bounce and flow-induced relative humidity changes are minimal for relative humidities from 70 to 80%, this is the optimal range for atmospheric sampling with cascade impactors having uncoated stages. Surface loading was found to have a negligible effect on particle bounce for atmospheric particles. A similar apparatus was also used to measure densities of particles in the 0.06-0.18 μm diameter range using the technique described by Kelly and McMurry (Aerosol Sci. Technol. 17, 199-212, 1992). We found that the average density varied between 1.60 and 1.79 gcm^-3 (typical uncertainty 4%), and tended to decrease with increasing relative humidity. Measured dry particle densities' exceeded the value of 1.48 gcm^-3 that was estimated from size-resolved chemical composition measured with MOUDIs.