Secondary organic aerosol from VOC mixtures in an oxidation flow reactor
h i g h l i g h t s SOA from VOC mixtures in a flow reactor was analysed in terms of mass yield. A higher than expected yield was found in mixtures containing myrcene. At low mass concentrations SOA condensation in the reactor is diffusion limited. Isoprene shifted the size distribution of mixtures to larger diameters. Isoprene was not found to affect the average OH exposure significantly. a b s t r a c t The atmospheric organic aerosol is a tremendously complex system in terms of chemical
... ms of chemical content. Models generally treat the mixtures as ideal, something which has been questioned owing to modelmeasurement discrepancies. We used an oxidation flow reactor to produce secondary organic aerosol (SOA) mixtures containing oxidation products of biogenic (a-pinene, myrcene and isoprene) and anthropogenic (m-xylene) volatile organic compounds (VOCs). The resulting volume concentration and chemical composition was measured using a scanning mobility particle sizer (SMPS) and a highresolution time-of-flight aerosol mass spectrometer (HR-ToF-AMS), respectively. The SOA mass yield of the mixtures was compared to a partitioning model constructed from single VOC experiments. The single VOC SOA mass yields with no wall-loss correction applied are comparable to previous experiments. In the mixtures containing myrcene a higher yield than expected was produced. We attribute this to an increased condensation sink, arising from myrcene producing a significantly higher number of nucleation particles compared to the other precursors. Isoprene did not produce much mass in single VOC experiments but contributed to the mass of the mixtures. The effect of high concentrations of isoprene on the OH exposure was found to be small, even at OH reactivities that previously have been reported to significantly suppress OH exposures in oxidation flow reactors. Furthermore, isoprene shifted the particle size distribution of mixtures towards larger sizes, which could be due to a change in oxidant dynamics inside the reactor.