Secondary Organic Aerosol Formation From Non-Traditional Sources

Traditional anthropogenic Secondary Organic Aerosol (SOA) research has focused on various combustion sources, including components of emissions from factories, vehicles, and biomass burning. Current models under-predict SOA formation and the oxidation products of non-traditional sources are likely an overlooked source. To this end we studied three major nontraditional sources of emissions for their SOA formation potential: oxygenated Volatile Chemical Products (VCP) and Intermediate Volatile Chemical Products (IVCP), Polycyclic Aromatic Hydrocarbons (PAH), and asphalt heating. 

Emissions of Volatile Chemical Products and Intermediate Volatility Chemical Products such as cleaning solvents, pesticides, coatings, and personal care products are now larger than emissions from more traditional sources, in part because of control measures on those sources and due to their large emission factors. Various I/VCPs containing oxygen groups such as glycol ethers, esters, and oxygenated aromatics have SOA formation potentials which have not been substantially studied. I/VCPs containing oxygen groups may have a greater probability of undergoing fragmentation and inhibiting intramolecular hydrogen shifts when oxidized, resulting in less polar, lower molecular weight products. As the volatility is inversely dependent on molecular weight and polarity, the addition of these groups may prevent SOA formation. 

We oxidized common I/VCPs containing oxygen groups under high and low NO_x conditions to determine their potential as SOA precursors. We oxidized four glycol ether I/VCPs, two oxygenated aromatic VCPs, and two ester VCPs using an Oxidation Flow Reactor at 50% RH with and without the presence of NO_x. All non-aromatic species had SOA mass yields below the detection limit. The two aromatic ring-containing compounds (2-phenoxyethanol and 1- phenoxy-2-propanol) had SOA mass yields of approximately 15%. CIMS data demonstrated this was due to the formation of lower volatility products from the aromatics. SOA composition varied with NO_x level despite similar mass yields. Higher oxidation occurred under low NO_x, though nitrogen content remained low under high NO_x. Overall, this supports the idea that oxygen groups can inhibit SOA formation and that some oxygenated VCPs may be used in consumer products without making SOA. 

We further tested the two most basic Polycyclic Aromatic Hydrocarbons: 2- methylnaphthalene and naphthalene as a comparison to oxygenated aromatic species. PAHs are of particular interest both for their widespread production from non-traditional sources such as plastic and asphalt heating as well as their health effects as carcinogens. As PAHs are aromatic and relatively high in molecular weight, they have strong potential as SOA precursors, which requires further study. 

We oxidized the two PAHs 2-methylnaphthalene and naphthalene in an OFR using similar low NO_x oxidation levels to the oxygenated VCPs at 50% RH. Both species demonstrated significant SOA yields (g OA/g >0.1) for each OH exposure measured. 2- methylnaphthalene formed SOA yields of ~0.35 g OA/g for all OH exposures, whereas naphthalene formed SOA yields over 0.2 g OA/g only under the two lowest OH exposure conditions, and only 0.11-0.12 g OA/g at higher OH concentrations. The primary differences in products were driven by naphthalene having higher oxidation exposure (> 2 days equivalent exposure) compared to 2-methylnaphthalene (< 2 days equivalent exposure). Nonetheless the additional methyl group in 2-methylnaphthalene did have substantial effects on product formation, primarily by lowering the O:C ratio and forming larger nC = 11 products not formed from naphthalene. Overall, this further demonstrates the ability of aromatic species to generate SOA as well as the importance of the oxidation exposure and overall structure for SOA precursors. 

Finally, we studied OA formation from asphalt heating. Asphalt is ubiquitous throughout urban areas worldwide. Asphalt is a source of organic compounds spanning a wide range of volatility, from volatile organic compounds to low-volatility species. The emission rate and composition depend strongly on the temperature. For example, Khare et al. showed that emission rates are significantly higher at application temperatures (~130-160°C) than summertime road surface temperatures (~50-70°C), and that the composition of the emissions depends on factors like the presence of UV radiation. Khare et al. also estimated that asphalt paving is a major missing source of intermediate and semivolatile organic compounds in southern California; however, additional studies are necessary to further constrain its contribution to urban particulate matter. 

In this study, we evaluate the production of primary (POA) and secondary organic aerosols (SOA) from asphalt-related emissions. Fresh roadway asphalt samples were collected from road-paving operations in Pittsburgh and stored in a chemical freezer. Samples were heated at either common application temperature (~130°C) or peak summertime road surface temperature (~55°C). The emitted gas-phase vapors were flushed into a smog chamber containing ammonium sulfate seed particles that served as a condensation sink. SOA was then generated via the photo-oxidation of the emissions under high-NO_x conditions typical of urban chemistry. 

 At application temperature, organic vapors were emitted and condensed into POA upon cooling, whereas POA was not formed at summertime temperature. SOA was formed under both conditions, with SOA formation at application temperature partially balanced by POA evaporation. Though application temperature experiments were able to generate substantially greater OA, SOA formation from passive asphalt heating would have a greater effect over time by comparison due to the long lifetime of emissions from asphalt pavements. We further compare the aerosol mass spectra of the laboratory-generated asphalt OA to different OA factors to see if asphalt heating may be an overlooked source in OA inventories. Overall, we demonstrate that asphalt heating both at immediate application and over time is capable of substantial SOA formation. Together, these non-traditional sources have the potential to be major sources of SOA and narrow the gap between models and observations. It is the hope of this work to encourage further study of these and other non-traditional sources of SOA.