AAAR 37th Annual Conference October 14 - October 18, 2019 Oregon Convention Center Portland, Oregon, USA
Abstract View
Extending the Stolzenburg DMA Transfer Function to the Scanning Electrical Mobility Spectrometer (SEMS)
YUANLONG HUANG, John Seinfeld, Richard Flagan, California Institute of Technology
Abstract Number: 153 Working Group: Instrumentation and Methods
Abstract The differential mobility analyzer (DMA) has been widely used in particle size distribution measurement. To increase the measurement time resolution, the high voltage applied to the DMA is ramped exponentially (SEMS) rather than stepped through a sequence of discrete voltages. The transfer function for this scanning DMA describes how counts of particles of a given size are distributed among the successive time bins after classification within the time-varying electric field of the DMA. While a transfer function has previously been derived for the scanning DMA in the limit of non-diffusive particles, diffusional effects, which are very important for particles at the lower end of the sizing domain, have only been taken into account through numerical simulation, often a high computational cost. The Stolzenburg analysis that considers diffusion of particles from the ideal, nondiffusive trajectories provides a good approximation to the diffusional transfer function for a DMA that is operated at constant voltages. Here, we extend that method to scanning-mode operation of the cylindrical DMA. The kernel of this semi-analytical method is a contour plot of particle’s residence time distribution inside the DMA in the inlet-outlet space, from which one can estimate the contribution of diffusion to article transmission from inlet to outlet along different particle trajectories. This analysis reproduces the Stolzenburg transfer function in the asymptotic limit of constant-voltage operation. The scanning DMA transfer function derived by this method is further validated by comparison with that calculated by numerical simulations. By mapping the transmission probability between particle entrance and exit trajectories within the DMA, this semi-analytical method enables rapid calculation of both static and scanning DMA transfer functions for idealized DMA geometries; when applied to numerical simulations of nondiffusive trajectories, it further facilitates efficient determination of transfer functions with real DMAs.