Quantitative Characterization of the Volatility Distribution of Organic Aerosols in a Polluted Urban Area: Intercomparison Between Thermodenuder and Molecular Measurements

To quantify the volatility of organic aerosols (OA), a comprehensive campaign was conducted in the Chinese megacity. Volatility distributions of OA and particle-phase organic nitrate (pON) were estimated based on five methods: (a) empirical method and (b) kinetic model based on the measurement of a thermodenuder (TD) coupled with an aerosol mass spectrometer; (c) Formula-based SIMPOL model-driven method; (d) Element-based estimations using molecular formula measurements of OA; and (e) gas/particle partitioning. Our results demonstrate that the ambient OA volatility distribution shows good agreement between the two heating methods and the formula-based method when assuming ambient OA was mainly composed of organic nitrate (pON), organic sulfate and acid groups using the SIMPOL model. However, the element-based method tends to overestimate the volatility of OA compared to the above three methods, suggesting large uncertainties in the parameterizations or in the representativeness of the molecular measurements that need further refinement. The volatility of ambient OA is generally lower than that of the laboratory-derived secondary OA, emphasizing the impact of aging. A large fraction at the higher and lower volatility ranges (approximately log C* <= -9 and >= 2 mu g m-3) was found for pON, implying the importance of both extremely low volatile and semi-volatile species. Overall, this study evaluates different methods for volatility estimation and gives new insight into the volatility of OA and pON in urban areas. Volatility, which controls the gas/particle partitioning of organic compounds, is one of most the important physiochemistry properties of organic aerosols (OA). Multiple methods have been used to estimate OA volatility, whereas the accuracy of each method is still unclear. The purpose of this study is to verify the volatility of ambient OA and its key component (i.e., particulate organic nitrate, pON) in urban areas by comparing the estimated results using different methods. With the help of state-of-art mass spectrometers, the estimation methods, including two heating methods, one formula-based method, one element-based method, and one gas/particle partitioning method, were achieved. In general, we found the heating methods show good agreement with the formula-based method with a reasonable assumption of functional groups contained in OA and pON. In general, we observed a large faction of extremely low volatile compounds exist in OA (30%-40%), and both extremely low volatile and semi-volatile species exist in pON, suggesting the complex evolution of atmospheric particles. The results provide a better understanding of the volatility of OA and pON in urban areas and benefit accurate volatility estimation that facilitates the model simulations of OA. The volatility of ambient organic aerosols (OA) and organic nitrate (pON) estimated from five methods were systematically compared The OA volatility from SIMPOL-driven model under organic sulfate, pON and acid assumption are consistent with those from heating method The volatility distribution of pON indicates both extremely low volatility compounds as oligomers and semi-volatile species exist in pON