Analysis of Secondary Organic Aerosol Compounds from the Photooxidation of d-Limonene in the Presence of NO X and their Detection in Ambient PM 2.5 MOHAMMED JAOUI,* ,† E. CORSE, † TADEUSZ E. KLEINDIENST, ‡ JOHN H. OFFENBERG, ‡ MICHAEL LEWANDOWSKI, ‡ AND EDWARD O. EDNEY ‡ Alion Science and Technology, P.O. Box 12313, Research Triangle Park, North Carolina 27709 and U.S. Environmental Protection Agency, Office of Research and Development, National Exposure Research Laboratory, Research Triangle Park, North Carolina 27711 Chemical analysis of secondary organic aerosol (SOA) from the photooxidation of a d-limonene/NO X /air mixture was carried out. SOA, generated in a smog chamber, was collected on Zefluor filters. To determine the structural characteristics of the compounds, the filter samples were solvent extracted and derivatized using analytical techniques that characterize functional groups contained in the compound: BF 3 -methanol derivatization was used for carboxylic groups, BSTFA for acidic and nonacidic hydroxyl groups, and PFBHA for ketone and aldehyde groups. The resulting derivative compounds were analyzed by GC-MS in the methane CI and EI modes. GC-MS analysis showed the occurrence of 103 oxygenated organic compounds in the filter extracts, 28 of which were identified. The major components include five tracer compounds previously identified from the photooxidation of R-pinene/NO X or -pinene/NO X systems, C 4 -C 6 linear dicarboxylic acids, ketolimononaldehyde, limonic acid, and ketolimonic acid. Time profiles, yields, and proposed reaction schemes are provided for selected compounds. The laboratory SOA yield was 0.51 at a SOA concentration of 1470 µgm -3 . To determine the contributions of SOA products from d-limonene to ambient PM 2.5 , an analysis was performed for eight ambient PM 2.5 samples collected in the southeastern United States in summer 2003. GC-MS analysis showed the occurrence of 21 d-limonene SOA compounds, indicating the impact of d-limonene on the regional aerosol burden. Based on our analysis, two compounds (nos. 55 and 69), not observed from the photooxidation of R-pinene or -pinene, are candidate tracers for d-limonene in atmospheric particulate matter. Introduction Biogenic compounds constitute a significant fraction of hydrocarbons emitted into the atmosphere (1). The primary mechanism of removal of these hydrocarbons from the atmosphere is through gas-phase reaction with the hydroxyl and nitrate radicals and ozone to produce nonvolatile and semivolatile compounds that can condense onto particulate matter (PM) and contribute to ambient secondary organic aerosol (SOA) and tropospheric ozone production. A number of studies have shown the relationship between exposure to PM2.5 (PM with diameter <2.5 µm) and adverse health effects (2). As the understanding of the toxicology associated with these particles develops, more accurate compositional data may be required. A wide range of organic compounds from primary and secondary sources, including polar oxygenated compounds (POCs) bearing one or more functional groups (e.g., hydroxyl (-OH), carboxylic (-COOH), ketone (>CO), and aldehyde (-HCO)), have been identified in PM2.5, although less than 25% of organic mass has been accounted for. Monoterpenes, particularly R- -pinene, and d-limonene, are emitted in relatively large quantities into ambient atmosphere. While SOA formation from R- and -pinene has been investigated in many laboratories, few studies have reported d-limonene SOA, one of the most frequently emitted monoterpenes from vegetation (3, 4), notwithstanding its high SOA yield (5-13). Recently, an analytical technique was developed in our laboratory for the characterization of POCs bearing one or more of the following groups: -OH, -COOH, >CO, and -HCO (14). This method is based on derivatizing -COOH groups using BF3-methanol as the derivatizing reagent, -COOH/-OH groups using BSTFA as the silylation reagent, and >CO/-HCO groups using PFBHA. GC-MS analysis in electron impact (EI) ionization or positive chemical ionization (CI) mode was used for identification and/or quantification of POCs. In the present study, the chemical composition of SOA from the photooxidation of a d-limonene/NOX/air system was determined using the new method (14). In addition, a similar chemical analysis of eight PM2.5 field samples (FS) was carried out to determine the occurrence of d-limonene tracer compounds. To account for some observed SOA reaction products, a chemical mechanism is proposed in this study. Materials and Methods All chemicals, including derivatization reagents (BF3- methanol, BSTFA 1% trimethylchlorosilane, and PFBHA), were purchased from Aldrich Chemical Co. (Milwaukee, WI) at the highest purity available and were used without further purification. All solvents (GC 2 quality) were from Burdick and Jackson (Muskegon, MI). The experiments were carried out in a 14.5-m 3 paral- lelepiped, Teflon-coated stainless steel chamber. Details of the chamber, its operation, and the experimental methods are described elsewhere (15). The derivatization procedures used in this study were reported in our previous paper (14), which included a discussion of the optimization and possible artifacts associated with each procedure. Two sets of experi- ments were conducted (Table 1). A dynamic experiment (DE) was conducted with a 6-h residence time, chosen to ensure that a highly oxidized aerosol would be produced. SOA was collected for 24 h and multiple derivatization techniques were used to identify compounds. In addition, a static experiment (SE) was carried out for d-limonene, in which five 90-min samples were collected, and product time profiles were examined using BSTFA derivatization. No seed aerosol was used in these experiments and d-limonene SOA is assumed to be new particle growth. * Corresponding author phone: (919) 541-7728; e-mail: jaoui.mohammed@epa.gov. † Alion Science and Technology. ‡ U.S. Environmental Protection Agency. Environ. Sci. Technol. 2006, 40, 3819-3828 10.1021/es052566z CCC: $33.50  2006 American Chemical Society VOL. 40, NO. 12, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 3819 Published on Web 05/17/2006