Particle bounce cascade impactors

Better collection methods are needed for stack measurements. Virtual Impactors or other devices that avoid the bounce-off problems of cascade Impactors should be developed. Methods for collection of very large amounts of size-segregated particles suitable for detailed organic analyses are needed. Better low-blank filter and collection-surface materials that can withstand high stack temperatures are needed. Careful attention should be given to the determination of vapor-phase species of volatile elements and compounds. 

Most of the published composition/size distribution data have been obtained by analyzing cascade impactor samples. Some of these data suffer from poor size classification as a result of particle bounce or reentrainment, seriously limiting size resolution. Even when this problem is overcome, the data obtained with conventional cascade impactors are not capable of resolving many details of the distribution of submicron particles. These instruments typically classify only those particles larger than 0.3-0.5 tam aerodynamic diameter. All smaller particles are collected on a filter downstream of the impactor. Some measurements of the variation of composition with size below this limit have been attempted by aerodynamically classifying resuspended ash ( ). These data suffer from incomplete disapregation as well as poor classification of the smaller particles. 

The principal problems in determining size distribution parameters with cascade impactors are wall losses, inefficient collection due to particle bounce, deposition of gas-phase species on impaction substrates, and deposition of fine particles from boundary layers. 

To minimize the particle bounce off effect, collection surfaces should also be selected carefully. Common types of impaction surfaces include membrane, fiberglass, silver membrane, Teflon and Nuclepore filter, and brass and stainless steel shim stock. Table 2.2 shows an example of the effect of selection of collection surface on the wall losses (Newton et al., 1990). In Table 2.2, the test aerosols are droplets of 1% CsCl plus 1% uranine. Three types of cascade impactors were used, including Mercer, Sierra Radial Slit Jet (SRSJ), and Lovelace Multi-Jet (LMJ). The occurrence of particle bouncing may be indicated by the presence of excess mass on the back-up filter. 

Cascade impactors—measures one-half to two-thirds of accumulation mode, all of coarse particle mode particle bounce is a serious 0 (filter) 0.3 (last stage) 30 8 1 pg/m /stage Inertial impaction... 

Dzubay, T.G., Hines, L.E., Stevens, R.K., 1976. Particle bounce errors in cascade impactors. Atmos. Environ. 10, 229—234. 

---