Asymmetric Instrument Response Resulting from Mixing Effects in Accelerated DMA-CPC Measurements

Abstract

Several method have been proposed to accelerate the measurements made with differential mobility analyzers (DMA), including the Scanning Electrical Mobility Spectrometer (SEMS) and the Scanning Mobility Particle Sizer (SMPS). Wang and Flagan (1990) developed a data analysis procedure that accounts for the migration of the particles through a time-varying electric field and the delay associated with transport from the analyzer column outlet to the detection point. Experience using a variety of detectors and scan rates has indicated that the instrument response depends on the plumbing configuration if a condensation particle counter (CPC) is used as a detector. When a sharply peaked distribution is analyzed, the apparent breadth of the measured distribution depends on the detector used and on the scan rate. Size distributions measured with an increasing voltage scan exhibit a tail toward large particle sizes, while decreasing voltage scans produce a more pronounced tail on the small particle end of the distribution. The reduction in peak height with increasing scan rate can be attributed to particle retention in the plumbing between the outlet of the DMA analyzer column and the point in the CPC where the particles are detected optically. This paper examines the smearing of the transfer function of the SEMS as a result of flow non-idealities in the system. A model has been developed to predict the distortion of the transfer function in terms of the particle residence time distribution within the instrument. Results of this model are compared with calibration data as a function of the detector employed and the scan rate. We include laboratory observations of the phenomenon and examine the use of data inversion techniques to retrieve the true size distribution.