AAAR 33rd Annual Conference
October 20 - October 24, 2014
Rosen Shingle Creek
Orlando, Florida, USA
Abstract View
Determining Aerosol Volatility Parameters using a ‘Dual Thermodenuder’ System: Application to Laboratory-generated Organic Aerosol
PROVAT SAHA, Andrey Khlystov, Andrew Grieshop, North Carolina State University
Abstract Number: 514 Working Group: Carbonaceous Aerosols in the Atmosphere
Abstract A thermodenuder (TD) is a widely used tool for measuring aerosol volatility in the lab and field. Although the TD is a simple system, extraction of the parameters dictating aerosol volatility from TD data is challenging due to long equilibration times and uncertainties concerning potential kinetics limitations to mass transfer and components’ enthalpies of evaporation (Hvap). To address these limitations, we have developed a ‘dual TD’ based experimental approach in which one line consists of a temperature-stepping TD (TS-TD) with a relatively long residence time and the other operates isothermally at variable residence time (VRT-TD). The TD system was characterized for axial temperature distributions, distribution of measured residence time using pulses of CO2 and temperature-dependent and size-resolved particle losses using non-volatile sodium chloride (NaCl) aerosols. An optimizing evaporation kinetics model based on the volatility basis set (VBS) was used to extract the values of parameters dictating volatility (e.g., distribution of materials at different volatility bins and associated Hvap and mass accommodation coefficient). Initial evaluation of the system measured the volatility of laboratory generated di-carboxylic acid aerosols (e.g., adipic acid, succinic acid, and glutaric acid); excellent agreement with previously published results was observed. For example, our results reconfirm the mass accommodation coefficient (alpha) of approximately order of 0.1 determined by the other groups via curve fitting approaches. The volatility of alpha-pinene ozonolysis secondary organic aerosols (SOA) in a smog chamber was then studied. Approximately 35-45% SOA mass evaporated at 50° C with RT of 50 s indicating that a substantial portion of SOA is highly semi-volatile, whereas about 8-10% SOA mass remained at 120° C indicating that some portion of SOA mass consisted of relatively low volatility material. A multi-component model was adapted to model SOA and modeling effort suggests that the VRT data provides additional constraint on the feasible Hvap and alpha values. This method will be applied in future studies to characterize the volatility of organic aerosol from different sources (e.g., cook stoves emission, near road environment, and ambient atmospheric conditions).