10th International Aerosol Conference
September 2 - September 7, 2018
America's Center Convention Complex
St. Louis, Missouri, USA

Abstract View


Estimating Secondary Organic Aerosol Yield from Size Distribution Measurements in Chamber Experiments

WEIMENG KONG, Sophia Charan, Yuanlong Huang, Huajun Mai, John Seinfeld, Richard Flagan, California Institute of Technology

     Abstract Number: 375
     Working Group: Instrumentation

Abstract
In order to determine secondary organic aerosol (SOA) yields in chamber experiments, size distribution measurements obtained with the differential mobility analyzer must be translated from number to mass distributions. Any uncertainty in the large particle tail of the size distribution significantly impacts the SOA estimates. Size distributions are typically obtained using the scanning electrical mobility analyzer (SEMS; also known as the scanning mobility particle sizer, SMPS). Most interpretation of SEMS data is performed using the constant-voltage differential mobility analyzer (DMA) transfer function derived using a highly idealized model of the DMA. To address biases associated with the finite time response of the CPC detector, SEMS/SMPS systems are generally operated with a relatively slow voltage scan, and convoluting the DMA transfer function with an empirically fit model of the delay-time distribution introduced by the CPC. Depending upon the quality of the delay time model employed and the details of the inversion, the large-particle tail of the particle size distribution can introduce substantial uncertainty in the estimated yield. Moreover, recent derivation of the actual transfer function for the SEMS/SMPS based on detailed numerical modeling of flows, fields, and particle transport within the TSI long column DMA during voltage scanning reveals differences from that derived for the constant voltage DMA. This study reports on the integration of this SEMS/SMPS transfer function into the data inversion process for determination of SOA yields. A number of previous chamber experiments have been reanalyzed with the new transfer function as well to explore the effects on actual experimental data, as are new experiments performed under very slow scans that minimize these instrumental effects. Simulated measurements from modeled experiments demonstrate the reduction of uncertainty in the SOA yields that results from use of the proper transfer function.