The Andersen Impactor: Calibration, Wall Losses and Numerical Simulation

-Andersen cascade impactors are widely used to assess airborne particle size distributions. Two forms (MkI and MklI) have been calibrated using solid and liquid aerosols, measured by microscopy, fluorimetry, and aerodynamic particle sizer. A fluorescent tracer was used to examine qualitatively the details of the wall loss deposition patterns when using greased and ungreased plates, and to quantify their magnitude at different particle sizes. The resultant calibrations differ in some respects from the manufacturer's, but are broadly consistent with theory. Suggestions are made for improvements to the design. The results of these investigations were used to produce size-dependent curves for deposition on each plate, and between stages. An entry efficiency curve was assumed. A computer simulation was then performed of the sampling of various log-normal size distributions, so that measured distribution parameters can be corrected for the inevitable imperfections of the impactors. The assumption about entry efficiency is important, but there is little information on this variable, which remains to be determined for the various versions of the Andersen impactor. N O M E N C L A T U R E C Cunningham slip correction D i Jet diameter d Diameter of sphere da Aerodynamic diameter dso Diameter at which collection efficiency is 50% L Jet length of impactor N d Quantity of material of size d approaching plate n d Quantity of material of size d retained on plate R Fractional retention Rej Jet Reynolds Number S Jet exit to impaction plate distance Vj Air velocity in jet ~/ Fluid viscosity pp Density of particle Stokes Number ~1/2 Value of the square root of the Stokes Number at which efficiency is 50% 50 crg Geometric standard deviation I N T R O D U C T I O N ¢ Determination of airborne particle size distributions is frequently undertaken using cascade impactors. In all such devices, dust laden air is aspirated through a series of impaction stages of successively higher efficiency with respect to aerodynamic diameter. The ideal situation would be where an impaction stage was perfectly efficient in the removal of particles larger than a given diameter, and allowed all particles smaller than this diameter to pass on to the next impaction stage. In practice, for a variety of reasons such a sharp size collection is not possible, and the impaction efficiency of a given stage increases more gradually with increasing particle diameter. The parameter governing impaction phenomena is the Stokes Number (~b), given for round jet impactors by: o=c o~d2 vj (1) 18 Djvl © Crown copyright. 67