Challenges in determining atmospheric organic aerosol volatility distributions using thermal evaporation techniques. Issue 8 (2nd August 2020)

• Record Type: Journal Article
• Authors: Cain, Kerrigan P.; Karnezi, Eleni; Pandis, Spyros N.
• Is Part Of: Aerosol science and technology. Volume 54:Issue 8(2020)
• Page Start: 941
• Page End: 957
• Publication Date: 2020-08-02
• Subjects: Jim Smith; Aerosols -- Periodicals; Aerosol Propellants -- Periodicals; Aerosols -- Periodicals; 660.294515
• DOI: 10.1080/02786826.2020.1748172
• Languages: English
• ISSNs: 0278-6826

Abstract

Abstract: Volatility is one of the most important physical properties of organic aerosol (OA), as it determines the partitioning of its components between the vapor and particulate phases. Despite their atmospheric importance, multicomponent OA volatility estimates remain quite uncertain. This study combined thermodenuder (TD) and isothermal dilution measurements to characterize secondary OA (SOA) generated from the ozonolysis of α-pinene and cyclohexene. The SOA from both precursors evaporated similarly in the TD, but behaved quite differently when isothermally diluted by similar amounts. The α-pinene ozonolysis SOA evaporated by only 20% after 2 h of dilution by a factor of around 20, while 65% of the cyclohexene ozonolysis SOA evaporated at the same conditions. The volatility distributions were first estimated by fitting only the evaporation in the TD. This approach resulted in similar volatility distributions for the two systems. Then, the model was used to fit both the evaporation in the TD and the dilution chamber. This technique estimated drastically different volatility distributions with the α-pinene ozonolysis SOA consisting of mostly low-volatility compounds and the cyclohexene ozonolysis SOA consisting of mostly semi-volatile compounds. In the next stage of analysis, the model was updated to account for vapor-phase wall-losses occurring in the dilution chamber. This approach resulted in slightly less volatile SOA and provided some information about the losses of vapors to the walls, but the results were fairly uncertain. These results show the necessity of combining thermal measurements with other techniques to accurately estimate OA volatility. Copyright © 2020 American Association for Aerosol Research