10th International Aerosol Conference September 2 - September 7, 2018 America's Center Convention Complex St. Louis, Missouri, USA
Abstract View
Prediction of the Chamber Wall Process of Gaseous Semivolatile Organic Compounds Using a Linear Solvation Energy Relationship
HUANHUAN JIANG, Myoseon Jang, Sanhee Han, University of Florida
Abstract Number: 504 Working Group: Aerosol Modeling
Abstract The chamber wall process of gaseous and semivolatile organic compounds has been known to significantly influence studies of secondary organic aerosol (SOA) formation. For example, the wall process of gaseous semivolatile organic compounds leads considerable underestimation of SOA yields (e.g., 30% for toluene SOA) and furthermore affects the credibility of the predictive SOA model derived using chamber data. In this study, a mathematical model is built to predict the of uptake coefficient of a gaseous semivolatile organic compound by the integration of the deposition kinetics of organic compounds with a Linear Solvation Energy Relationship (LSER). The deposition process of organic compounds is approached using a conservation balance via on- (adsorption) and off-kinetic mechanism (desorption) between the gaseous phase and the chamber wall phase. The adsorption rate constant (kon) of organic compounds is processed using a conventional mass transfer process. The desorption rate constant (koff), which is linked to both an off-gassing rate constant and the gas-wall partitioning coefficient of an organic compound, i, (Ki,wall), is semiempirically predicted by fitting the measured decay rate of compound i to the thermodynamic physicochemical parameters based on LSER. Various model compounds (i.e., long-chain deuterated alkanes, alcohols, carboxylic acids, and carbonyls, substituted phenols) are introduced into the large outdoor smog reactor (UF-APHOR chamber). The deposition rate constants of these compounds are measured using the traditional denuder sampling integrated with a gas chromatography-mass spectrometer (GCMS) in the course of a 12-hour wall-loss process. The resulting deposition rate constants are applied to the derivation of the model. Our recent exploratory data proved that the deposition process was well-predicted using the newly derived LSER base model. In this study, the resulting wall process model will be applied to the estimation of SOA yields and improve the quality of explicit SOA models (i.e., UNIPAR).