Variability in size distribution measurements obtained using multiple Andersen Mark II cascade impactors

Purpose. Andersen Mark H cascade impactors are commonly used for the testing of pharmaceutical aerosols. The reproducibility of size distribution measurements made with various Mark II impactors was assessed theoretically and experimentally. Methods. Stage cutpoints were calculated for fourteen Mark II impactors based on jet diameter measurements for each stage. The calculated cutpoints were used to predict differences in size distributions measured with various Mark II impactors. Five impactors were exposed to an atomizer-generated oleic acid test aerosol to experimentally verify predicted differences in size distribution measurements among impactors. Results. Stage cutpoints were calculated to vary by up to 0.45 micron for 'identical' stages of different Mark II impactors based on jet diameter measurements. The size distributions measured with various Mark II impactors were shown to be significantly different based on theoretical and experimental observations. For one particular experiment, the amount of material collected on Stage 6 ranged from 26.8 percent of the total sampled mass to 40.9 percent depending on which Mark II was used. Theoretical calculations predicted that the amount of material collected on Stage 6 would vary from 24.0 to 43.5 percent of the total sampled mass among impactors for this experiment. A Similarity Ratio, useful for interpreting the test results, was defined and discussed. Conclusions. Significant differences in stage cutpoints exist among Andersen Mark II impactors. These differences in stage cutpoints can result in large differences in size distribution measurements made with various Mark II impactors. By measuring jet diameters and calculating stage cutpoints, it is possible to predict the performance of a particular Mark II impactor.