Experimental Study of an Annular Impactor

It is desirable to calibrate impactors in order to gain quantitative information from the measurements made with them. Calibration of an impactor requires high reproduceability of the experimental conditions. In particular, the nature of the surface on which the deposit of particles is formed must be such that each impact is effective: the best results are obtained by coating the plates with an adhesive mixture of dioctylphthalate and Canada balsam. With this procedure, the shape of the curves for yield of stages is similar to that of theoretical curves. The geometry of the impactor with which the present study deals is such that certain dimensional relationships remain unchanged from one stage to the next. Furthermore, the curves for each stage can be reduced to a single curve by adopting a system of reduced co-ordinates. From this curve the different stages can easily be proportioned so as to localize at the desired point of the spectrum the cut-offs made by each stage. The same curve also indicates the effect of the rate at which air passes through the instrument on the stopping power of each stage. In the present case a throughput of 24 litres/min was selected; for four stages it corresponds to a pressure-drop of approximately 200 mm/Hg. The first and last cut-offs are at 0.5 and 4 μm, respectively. Depending on the use of the sample and the number of fractions desired, one or two intermediate stages can be eliminated. (author)