Chemically-Resolved Volatility of Biomass Burning Emission

JUN ZHANG, David Bell, Tiantian Wang, Kun Li, Mihnea Surdu, Sophie Bogler, Imad El Haddad, Jay G. Slowik, André S. H. Prévôt, Paul Scherrer Institute

     Abstract Number: 344
     Working Group: Aerosol Physics

Abstract
Primary organic aerosol (POA) emitted from biomass burning contributes a large fraction of carbonaceous aerosol. These emissions play an important role in air quality, climate, and human health. Volatility is an important property of aerosols because it dictates the partitioning of compounds between the gas and particle phase, thereby affecting their atmospheric fate. The volatility of a compound can be described by effective saturation concentration (C*) at a given temperature and pressure. Many parameterizations are available for estimating the C* of organic aerosol (OA) on the basis of molecular formula.However, there are large uncertainties when applying parameterizations to estimate the volatility of compounds having the same molecular formula but deriving from different OA sources. The dissimilarity between estimated and true values has variable causes including the difference in particle size, phase state, and the activity coefficient of a compound in the OA matrix. Therefore, the volatility of biomass burning OA, as a large contributor to global OA, should be further clarified. Here, we investigate the evaporation behavior of emissions from beech log combustion using a thermodenuder (TD), with real time measurement of the molecular formulae of OA constituents measured by an extractive electrospray ionization time-of-flight mass spectrometer (EESI-TOF). Nearly all the POA measured with the EESI-TOF evaporated in the TD at 100 ℃ with the effective residence time of 6 s. An evaporation-kinetics model was used to retrieve the C*, which was constrained by the empirical relationship between C* and Hvap. The log₁₀C* of compounds from beech logs burning varied from -4 to 1 combined with Hvap in the range of ~110 to ~170 kJ mol^-1.

We acknowledge the support of the SNF grant MOLORG (200020_188624).