Enhanced Chamber Wall Loss of Intermediate Volatility Polar Organic Compounds at High Water Vapor Concentrations

KENNETH S. DOCHERTY, Diya Yang, Olson David, Mohammed Jaoui, Michael Lewandowski, U.S. Environmental Protection Agency

     Abstract Number: 292
     Working Group: Instrumentation and Methods

Abstract
Results of a series of experiments conducted to investigate the role of water vapor in enhancing chamber wall loss of intermediate volatility polar organic compounds (IVPOC) are reported. Smog chamber investigation and wall loss characteristics of IVPOC are of interest, in part, due to the potential for volatile consumer products to contribute to ambient secondary organic aerosol formation. In the current study, a range of alcohols including C4-C9 linear alcohols, aromatic alcohols, and 1,2-ethanediol were used as surrogates of IVPOC and were introduced individually into the EPA continuous flow smog chamber along with propane as an internal standard. Chamber relative humidity (RH) was actively controlled in stages to vary from a minimum of ~3% to a maximum of ~40% while both inlet manifold and chamber gas phase concentrations were monitored continuously by on-line gas chromatography with flame ionization detection. Care was taken to rule out transmission losses throughout the system as the underlying cause of measurement variation. The extent of wall loss for each surrogate was calculated as the ratio of compound concentration measured from the inlet manifold upstream of the chamber to that measured from the chamber itself. Dilution ratios across the range of surrogates increase with chamber RH and the magnitude of this increase trends opposite to compound vapor pressure on average. Monitoring the decay rate of surrogate chamber concentrations under both dry and humid conditions suggests that high water vapor concentrations may also moderate the short-term reversibility of IVPOC wall losses under conditions like those here by limiting the repartition of these compounds back to the gas phase. A mechanism consistent with these observations is proposed and the implications of these results to both chamber and ambient IVPOC measurements are discussed.