Primary and secondary organic aerosol from heated cooking oil emissions

Cooking emissions have been identified as a source of both primary organic aerosol (POA) and secondary organic aerosol (SOA). To examine the characteristics of SOA from cooking emissions, emissions from seven vegetable oils (sunflower, olive, peanut, corn, canola, soybean, and palm oils) heated at 200°C were photooxidized under high-NO_x conditions in a smog chamber. OA was characterized using a high-resolution time-of-flight aerosol mass spectrometer (HR-TOF-AMS). Sunflower, peanut, corn, canola, and soybean oil generated relatively low concentrations of POA (≤ 0.5 µg m^-3) in the chamber. For palm and olive oil, positive matrix factorization (PMF) analysis separated POA and SOA better than the residual spectrum method. Temporal trends in concentrations of POA from heated palm oil were accurately predicted assuming first-order POA wall loss. However, this assumption overestimated POA concentrations from heated olive oil, which was attributed to the heterogeneous oxidation of POA. The mass spectra of the PMF resolved POA factor for palm oil, and the average POA from sunflower, peanut, corn, and canola oils were in better agreement (θ = 8-12º) with ambient cooking organic aerosol (COA) factors resolved in select Chinese megacities than those found in given European cities in the literature. The mass spectra of SOA formed from heated cooking oils had high abundances of m/zs 27, 28, 29, 39, 41, 44, and 55 and displayed limited similarity (θ > 20º) with ambient semi-volatile oxygenated OA (SV-OOA) factors. The entire OA data set measured herein follows a linear trend with a slope of approximately -0.4 in the Van Krevelen diagram, which may indicate oxidation mechanisms involving the addition of both carboxylic acid and alcohol or peroxide functional groups without fragmentation and/or the addition of carboxylic acid functional groups with fragmentation.