Characterizing oxidative flow reactor SOA production and OH radical exposure from laboratory experiments of complex mixtures (engine exhaust) and simple precursors (monoterpenes)

Recent commercialization of the Oxidative Flow Reactor (OFR, occasionally described in the literature as a "Potential Aerosol Mass") has created the opportunity for many researchers to explore the mechanisms behind OH-driven aerosol formation on a wide range of oxidative timescales (hours to weeks) in both laboratory and field measurements. These experiments have been conducted in both laboratory and field settings, including simple (i.e. single component) and complex (multi-component) precursors. Standard practices for performing OFR experiments, and interpreting data from the measurements, are still being developed. Measurement of gas and particle phase chemistry, from oxidation products generated in the OFR, through laboratory studies on single precursors and the measurement of SOA from vehicle emissions on short atmospheric timescales represent two very different experiments in which careful experimental design is essential for exploring reaction mechanisms and SOA yields. Two parameters essential in experimental design are (1) the role of seed aerosol in controlling gas-particle partitioning and SOA yields, and (2) the accurate determination of OH exposure during any one experiment. We investigated the role of seed aerosol surface area in controlling the observed SOA yields and gas/particle composition from the OH-initiated oxidation of four monoterpenes using an aerosol chemical ionization time-of-flight mass spectrometer and scanning mobility particle sizer. While the OH exposure during laboratory experiments is simple to constrain, complex mixtures such as diesel exhaust have high estimated OH reactivity values, and thus require careful consideration. We developed methods for constraining OH radical exposure in the OFR during vehicle exhaust oxidation experiments. We observe changes in O/C ratios and highly functionalized species over the temperature gradient employed in the aerosol-CIMS measurement. We relate this observed, speciated chemistry to the volatility of the aerosol, and compare observed SOA yields to other OFR and smog chamber SOA generation methods. Additionally, estimates of OH radical exposure in the OFR during different vehicle experiments of varying fuel type and speed were observed to vary as determined from a high-NOx and variable humidity calibration set.