Investigation of possible non-additive behavior in the formation of secondary organic aerosol from a mixture of anthropogenic and biogenic VOC precursors

Secondary organic aerosol (SOA) mass yields were tested for six combinations of two anthropogenic volatile organic compounds (AVOC); m-xylene & toluene, and three biogenic volatile organic compounds (BVOC); α-pinene, myrcene & isoprene. The purpose was to investigate any non-additive results in SOA formation from mixtures of AVOCs and BVOCs by comparing experimental yield with corresponding two-product model yields. No seed particles were used. Mass and number size distributions from generated SOA was investigated. The measurements were performed by aging the VOC gases using a potential aerosol mass (PAM) chamber and measuring the resulting SOA mass with a scanning mobility particle sizer (SMPS), and to a limited extent an aerosol mass spectrometer (AMS). AMS data was used to calculate an average SOA density of ≈ 2.0 g cm^−3 , although for calculations a value ≈ 1.4 g cm^−3 was used. The VOC air flow was at RH = 30.6±0.3 %, and PAM chamber O3 concentrations were for all experiments 5400 ± 880 ppb. Differences in experimental and model yields were observed with a maximum difference of 30 % and a minimum of 4 %, although for all experiments model yields were mostly or completely inside experimental uncertainty limits, and so no non-additive effects could be concluded. Aging of a mixture with isoprene added showed a factor ≈ 4 decrease in overall number concentration, and a slight shift of mass size distribution to larger particle sizes. The experiment set-up used was also investigated, and showed several shortcomings. Most notably was the uncertainty regarding VOC emission rate, to a part caused by a lack of instrumentation measuring diffusion chamber parameters.