Secondary organic aerosol formation during the oxidation of large aromatic and other cyclic anthropogenic volatile organic compounds

Volatile (VOCs) and intermediate volatility organic compounds (IVOCs) can undergo atmospheric oxidation, forming secondary organic aerosol (SOA) as their low volatility oxidation products condense in the particulate phase. Recent research has suggested that IVOCs, which have been neglected for decades, may have an important role in atmospheric SOA formation (Tkacik et al., 2012). Most of the work until now, has focused on SOA formation from VOCs with 5 to 10 carbon atoms.

The main goal of this work is to study the SOA production from the reactions of individual anthropogenic large VOCs and IVOCs with hydroxyl radicals (OH), under high NO_x conditions often encountered in urban areas. The organic compounds that were studied include cyclic alkanes of increasing size (amylcyclohexane, hexylcyclohexane, nonylcyclohexane and decylcyclohexane) and also aromatic compounds (1,3,5-trimethylbenzene, 1,3,5-triethylbenzene and 1,3,5-tri tri-tert-butylbenzene). The effects of the structure of the compound (alkylic cycle and aromatic ring) and the size of the molecule on the SOA yields is also investigated.

Photo-oxidation experiments were carried out in the atmospheric simulation chamber of the Foundation for Research and Technology-Hellas (FORTH-ASC). The instrumentation used included a scanning mobility particle sizer (SMPS) to measure the particle size distribution, a high-resolution aerosol mass spectrometer (AMS) to quantify the particle mass concentration and composition, and a proton transfer reaction mass spectrometer (PTR-MS) to monitor the organic vapor concentrations. Thermal desorption gas chromatography was also used for offline analysis of the gas-phase products of the reactions. The volatility distribution of the produced SOA was quantified combining thermodenuder and isothermal dilution measurements with the SOA yields.

In each experiment the basic procedure was to fill the chamber, which is a 10 m³ Teflon reactor, with dry, clean air, introduce dry ammonium sulfate particles, inject d9-butanol and the VOC, add the nitrous acid (HONO) and turn on the UV lights to initiate the SOA formation. The injection of the cyclic alkanes demanded heating the injection lines. Because the 1,3,5-tri-tert-butylbenzene is solid at room temperature, it was introduced with a vaporizer. Before each experiment the chamber was cleaned with dry, clean air for a full day.

The total SOA concentration in the chamber was calculated after the data were corrected for particle losses to the chamber walls. The AMS measurements were corrected also for the collection efficiency (CE) that was estimated in each experiment using the algorithm of Kostenidou et al. (2007). From the same algorithm the density of SOA was also estimated.

All the compounds were found to form a considerable amount of SOA. The cyclohexanes were found to have higher yields than the aromatic compounds. Our experiments indicated that aromatic precursors produce a more oxidized SOA than the cyclohexanes. The results of this study can be used in atmospheric chemical transport models for more accurate simulation of anthropogenic SOA formation.

 

REFERENCES

Kostenidou, E., Pandis, S. N., Pathak, R. K., Pandis, S. N., Kostenidou, E., and Pandis, S. N. (2007). Aerosol Science and Technology, 41, 1002–1010.

Tkacik, D. S., Presto, A. A., Donahue, N. M., and Robinson, A. L. (2012). Environmental Science and Technology, 46, 8773–8781.