Articles | Volume 17, issue 8
Research article
25 Apr 2017
Research article |  | 25 Apr 2017

Secondary organic aerosol formation from in situ OH, O3, and NO3 oxidation of ambient forest air in an oxidation flow reactor

Brett B. Palm, Pedro Campuzano-Jost, Douglas A. Day, Amber M. Ortega, Juliane L. Fry, Steven S. Brown, Kyle J. Zarzana, William Dube, Nicholas L. Wagner, Danielle C. Draper, Lisa Kaser, Werner Jud, Thomas Karl, Armin Hansel, Cándido Gutiérrez-Montes, and Jose L. Jimenez

Abstract. Ambient pine forest air was oxidized by OH, O3, or NO3 radicals using an oxidation flow reactor (OFR) during the BEACHON-RoMBAS (Bio–hydro–atmosphere interactions of Energy, Aerosols, Carbon, H2O, Organics and Nitrogen – Rocky Mountain Biogenic Aerosol Study) campaign to study biogenic secondary organic aerosol (SOA) formation and organic aerosol (OA) aging. A wide range of equivalent atmospheric photochemical ages was sampled, from hours up to days (for O3 and NO3) or weeks (for OH). Ambient air processed by the OFR was typically sampled every 20–30 min, in order to determine how the availability of SOA precursor gases in ambient air changed with diurnal and synoptic conditions, for each of the three oxidants. More SOA was formed during nighttime than daytime for all three oxidants, indicating that SOA precursor concentrations were higher at night. At all times of day, OH oxidation led to approximately 4 times more SOA formation than either O3 or NO3 oxidation. This is likely because O3 and NO3 will only react with gases containing C  =  C bonds (e.g., terpenes) to form SOA but will not react appreciably with many of their oxidation products or any species in the gas phase that lacks a C  =  C bond (e.g., pinonic acid, alkanes). In contrast, OH can continue to react with compounds that lack C  =  C bonds to produce SOA. Closure was achieved between the amount of SOA formed from O3 and NO3 oxidation in the OFR and the SOA predicted to form from measured concentrations of ambient monoterpenes and sesquiterpenes using published chamber yields. This is in contrast to previous work at this site (Palm et al., 2016), which has shown that a source of SOA from semi- and intermediate-volatility organic compounds (S/IVOCs) 3.4 times larger than the source from measured VOCs is needed to explain the measured SOA formation from OH oxidation. This work suggests that those S/IVOCs typically do not contain C  =  C bonds. O3 and NO3 oxidation produced SOA with elemental O : C and H : C similar to the least-oxidized OA observed in local ambient air, and neither oxidant led to net mass loss at the highest exposures, in contrast to OH oxidation. An OH exposure in the OFR equivalent to several hours of atmospheric aging also produced SOA with O : C and H : C values similar to ambient OA, while higher aging (days–weeks) led to formation of SOA with progressively higher O : C and lower H : C (and net mass loss at the highest exposures). NO3 oxidation led to the production of particulate organic nitrates (pRONO2), while OH and O3 oxidation (under low NO) did not, as expected. These measurements of SOA formation provide the first direct comparison of SOA formation potential and chemical evolution from OH, O3, and NO3 oxidation in the real atmosphere and help to clarify the oxidation processes that lead to SOA formation from biogenic hydrocarbons.

Short summary
Ambient forest air was oxidized by OH, O3, or NO3 inside an oxidation flow reactor, leading to formation of particulate matter from any gaseous precursors found in the air. Closure was achieved between the amount of particulate mass formed from O3 and NO3 oxidation and the amount predicted from speciated gaseous precursors, which was in contrast to previous results for OH oxidation (Palm et al., 2016). Elemental analysis of the particulate mass formed in the reactor is presented.
Final-revised paper