Comparing VOC Oxidation Results from Atmospheric Chamber Studies and Explicit Chemical Mechanisms

Abstract:

Chamber experiments coupled with mass spectrometric measurements and chemical mechanistic models have greatly contributed to our community’s knowledge of VOC oxidation and subsequent SOA formation. However, both measurements and mechanisms have inherent limitations: mass spectra typically can not provide detailed molecular structures or deconvolute overlapping signals of constitutional isomers, and mechanisms must be validated against experimental data and are hindered by uncertainties in our understanding of the chemistry. Herein we explore how coupled measurement-mechanism analyses enable us to use the strengths of one approach to improve the other. Chamber studies were performed with a suite of instruments including a PTR-MS (Proton Transfer Reaction Mass Spectrometer), a CIMS (Chemical Ionization Mass Spectrometer), and an AMS (Aerosol Mass Spectrometer) to measure the vast majority of secondary compounds, and mechanisms were generated using GECKO-A (the Generator of Explicit Chemical Kinetics of Organics in the Atmosphere). GECKO-A uses structure-activity relationships (SARs) to predict reaction products and rates if they are not already explicitly known, and these SARs may be modified to affect classes of reactions (e.g. alkoxy radical decomposition) throughout the entire mechanism. The n-butane oxidation system was chosen for this study because its chemistry is well-characterized and it produces relatively few major products which are all measured by PTR-MS. Results show strong agreement between GECKO-A’s mechanism and the chamber data. Furthermore, this analysis elucidated PTR-MS ion chemistry including losses of water, nitrate, and PAN functional groups which has implications for interpreting mass spectra of functionalized species (including organonitrogen species). Analysis of more complex systems including 1,2,4-trimethylbenzene allow for further interpretation of mass spectra and validation of heretofore unvalidated GECKO-A mechanisms. The combined results of these analyses suggest that they can serve as a template for future studies to yield new insights into SOA systems while simultaneously improving our ability to interpret chamber data and construct detailed mechanisms.