Characterization of secondary organic aerosol particles using aerosol laser time-of-flight mass spectrometer coupled with FCM clustering algorithm

Experiments for formation of secondary organic aerosol (SOA) from photooxidation of 1,3,5-trimethylbenzene in the CH3ONO/NO/air mixture were carried out in the laboratory chamber. The size and chemical composition of the resultant individual particles were measured in real-time by an aerosol laser time of flight mass spectrometer (ALTOFMS) recently designed in our group. We also developed Fuzzy C-Means (FCM) algorithm to classify the mass spectra of large numbers of SOA particles. The study first started with mixed particles generated from the standards benzaldehyde, phenol, benzoic acid, and nitrobenzene solutions to test the feasibility of application of the FCM. The FCM was then used to extract out potential aerosol classes in the chamber experiments. The results demonstrate that FCM allowed a clear identification of ten distinct chemical particle classes in this study, namely, 3,5-dimethylbenzoic acid, 3,5-dimethylbenzaldehyde, 2,4,6-trimethyl-5-nitrophenol, 2-methyl-4-oxo-2-pentenal, 2,4,6-trimethylphenol, 3,5-dimethyl-2-furanone, glyoxal, and high-molecular-weight (HMW) components. Compared to offline method such as gas chromatography mass spectrometry (GC-MS) measurement, the real-time ALTOFMS detection approach coupled with the FCM data processing algorithm can make cluster analysis of SOA successfully and provide more information of products. Thus ALTOFMS is a useful tool to reveal the formation and transformation processes of SOA particles in smog chambers. (C) 2012 Elsevier Ltd. All rights reserved.