Abstract. The current understanding of secondary organic aerosol (SOA) formation within biomass burning (BB) plumes is limited by the incomplete identification and quantification of the non-methane organic compounds (NMOCs) emitted from such fires. Gaseous organic compounds were collected on sorbent cartridges during laboratory burns as part of the fourth Fire Lab at Missoula Experiment (FLAME-4) and analyzed by two-dimensional gas chromatography–time-of-flight mass spectrometry (GC × GC–ToFMS). The sensitivity and resolving power of GC × GC–ToFMS allowed the acquisition of the most extensive data set of BB NMOCs to date, with measurements for 708 positively or tentatively identified compounds. Estimated emission factors (EFs) are presented for these compounds for burns of six different vegetative fuels, including conifer branches, grasses, agricultural residue, and peat. The number of compounds meeting the peak selection criteria ranged from 129 to 474 among individual burns, and included extensive isomer groups. For example, 38 monoterpene isomers were observed in the emissions from coniferous fuels; the isomeric ratios were found to be consistent with those reported in relevant essential oils, suggesting that the composition of such oils may be very useful when predicting fuel-dependent terpene emissions. Further, 11 sesquiterpenes were detected and tentatively identified, providing the first reported speciation of sesquiterpenes in gas-phase BB emissions. The calculated EFs for all measured compounds are compared and discussed in the context of potential SOA formation.

Abstract. Multiple trace-gas instruments were deployed during the fourth Fire Lab at Missoula Experiment (FLAME-4), including the first application of proton-transfer-reaction time-of-flight mass spectrometry (PTR-TOFMS) and comprehensive two-dimensional gas chromatography–time-of-flight mass spectrometry (GC × GC-TOFMS) for laboratory biomass burning (BB) measurements. Open-path Fourier transform infrared spectroscopy (OP-FTIR) was also deployed, as well as whole-air sampling (WAS) with one-dimensional gas chromatography–mass spectrometry (GC-MS) analysis. This combination of instruments provided an unprecedented level of detection and chemical speciation. The chemical composition and emission factors (EFs) determined by these four analytical techniques were compared for four representative fuels. The results demonstrate that the instruments are highly complementary, with each covering some unique and important ranges of compositional space, thus demonstrating the need for multi-instrument approaches to adequately characterize BB smoke emissions. Emission factors for overlapping compounds generally compared within experimental uncertainty, despite some outliers, including monoterpenes. Data from all measurements were synthesized into a single EF database that includes over 500 non-methane organic gases (NMOGs) to provide a comprehensive picture of speciated, gaseous BB emissions. The identified compounds were assessed as a function of volatility; 6–11 % of the total NMOG EF was associated with intermediate-volatility organic compounds (IVOCs). These atmospherically relevant compounds historically have been unresolved in BB smoke measurements and thus are largely missing from emission inventories. Additionally, the identified compounds were screened for published secondary organic aerosol (SOA) yields. Of the total reactive carbon (defined as EF scaled by the OH rate constant and carbon number of each compound) in the BB emissions, 55–77 % was associated with compounds for which SOA yields are unknown or understudied. The best candidates for future smog chamber experiments were identified based on the relative abundance and ubiquity of the understudied compounds, and they included furfural, 2-methyl furan, 2-furan methanol, and 1,3-cyclopentadiene. Laboratory study of these compounds will facilitate future modeling efforts.

During the summer and fall of 2005 in Riverside, California, the seasonal volatility behavior of submicrometer aerosol particles was investigated by coupling an automated thermodenuder system to an online single-particle mass spectrometer. A strong seasonal dependence was observed for the gas/particle partitioning of alkylamines within individual ambient submicrometer aged organic carbon particles internally mixed with ammonium, nitrate, and sulfate. In the summer, the amines were strongly correlated with nitrate and sulfate, suggesting the presence of aminium nitrate and sulfate salts which were nonvolatile and comprised approximately 6-9% of the average particle mass at 230 degrees C. In the fall, 86 +/- 1% of the amines volatilized below 113 degrees C with aminium nitrate and sulfate salts representing less than 1% of the particle mass at 230 degrees C. In the summer, a more acidic particle core led to protonation of the amines and subsequent formation of aminium sulfate and nitrate salts; whereas, in the fall, the particles contained more ammonium and thus were less acidic, causing fewer aminium salts to form. Therefore, the acidity of individual particles can greatly affect gas/particle partitioning of organic species in the atmosphere, and the concentrations of amines, as strong bases, should be included in estimations of aerosol pH.

Biomass burning (BB) is a large source of reactive compounds in the atmosphere. While the daytime photochemistry of BB emissions has been studied in some detail, there has been little focus on nighttime reactions despite the potential for substantial oxidative and heterogeneous chemistry. Here, we present the first analysis of nighttime aircraft intercepts of agricultural BB plumes using observations from the NOAA WP-3D aircraft during the 2013 Southeast Nexus (SENEX) campaign. We use these observations in conjunction with detailed chemical box modeling to investigate the formation and fate of oxidants (NO 3 , N 2 O 5 , O 3 , and OH) and BB volatile organic compounds (BBVOCs), using emissions representative of agricultural burns (rice straw) and western wildfires (ponderosa pine). Field observations suggest NO 3 production was approximately 1 ppbv hr −1 , while NO 3 and N 2 O 5 were at or below 3 pptv, indicating rapid NO 3 /N 2 O 5 reactivity. Model analysis shows that >99% of NO 3 / N 2 O 5 loss is due to BBVOC + NO 3 reactions rather than aerosol uptake of N 2 O 5. Nighttime BBVOC oxidation for rice straw and ponderosa pine fires is dominated by NO 3 (72, 53%, respectively) but O 3 oxidation is significant (25, 43%), leading to roughly 55% overnight depletion of the most reactive BBVOCs and NO 2 .

Abstract. Western US wildlands experience frequent and large-scale wildfires which are predicted to increase in the future. As a result, wildfire smoke emissions are expected to play an increasing role in atmospheric chemistry while negatively impacting regional air quality and human health. Understanding the impacts of smoke on the environment is informed by identifying and quantifying the chemical compounds that are emitted during wildfires and by providing empirical relationships that describe how the amount and composition of the emissions change based upon different fire conditions and fuels. This study examined particulate organic compounds emitted from burning common western US wildland fuels at the US Forest Service Fire Science Laboratory. Thousands of intermediate and semi-volatile organic compounds (I/SVOCs) were separated and quantified into fire-integrated emission factors (EFs) using a thermal desorption, two-dimensional gas chromatograph with online derivatization coupled to an electron ionization/vacuum ultraviolet high-resolution time-of-flight mass spectrometer (TD-GC × GC-EI/VUV-HRToFMS). Mass spectra, EFs as a function of modified combustion efficiency (MCE), fuel source, and other defining characteristics for the separated compounds are provided in the accompanying mass spectral library. Results show that EFs for total organic carbon (OC), chemical families of I/SVOCs, and most individual I/SVOCs span 2–5 orders of magnitude, with higher EFs at smoldering conditions (low MCE) than flaming. Logarithmic fits applied to the observations showed that log (EFs) for particulate organic compounds were inversely proportional to MCE. These measurements and relationships provide useful estimates of EFs for OC, elemental carbon (EC), organic chemical families, and individual I/SVOCs as a function of fire conditions.

Organosulfate species have recently been identified as a potentially significant class of secondary organic aerosol (SOA) species, yet little is known about their behavior in the atmosphere. In this work, organosulfates were observed in individual ambient aerosols using single particle mass spectrometry in Atlanta, GA during the 2002 Aerosol Nucleation and Characterization Experiment (ANARChE) and the 2008 August Mini-Intensive Gas and Aerosol Study (AMIGAS). Organosulfates derived from biogenically produced isoprene were detected as deprotonated molecular ions in negative-ion spectra measured by aerosol time-of-flight mass spectrometry; comparison to high-resolution mass spectrometry data obtained from filter samples corroborated the peak assignments. The size-resolved chemical composition measurements revealed that organosulfate species were mostly detected in submicrometer aerosols and across a range of aerosols from different sources, consistent with secondary reaction products. Detection of organosulfates in a large fraction of negative-ion ambient spectra − ca. 90−95% during ANARChE and ∼65% of submicrometer particles in AMIGAS − highlights the ubiquity of organosulfate species in the ambient aerosols of biogenically influenced urban environments.

After smoke from burning biomass is emitted into the atmosphere, chemical and physical processes change the composition and amount of organic aerosol present in the aged, diluted plume. During the fourth Fire Lab at Missoula Experiment, we performed smog‐chamber experiments to investigate formation of secondary organic aerosol (SOA) and multiphase oxidation of primary organic aerosol (POA). We simulated atmospheric aging of diluted smoke from a variety of biomass fuels while measuring particle composition using high‐resolution aerosol mass spectrometry. We quantified SOA formation using a tracer ion for low‐volatility POA as a reference standard (akin to a naturally occurring internal standard). These smoke aging experiments revealed variable organic aerosol (OA) enhancements, even for smoke from similar fuels and aging mechanisms. This variable OA enhancement correlated well with measured differences in the amounts of emitted volatile organic compounds (VOCs) that could subsequently be oxidized to form SOA. For some aging experiments, we were able to predict the SOA production to within a factor of 2 using a fuel‐specific VOC emission inventory that was scaled by burn‐specific toluene measurements. For fires of coniferous fuels that were dominated by needle burning, volatile biogenic compounds were the dominant precursor class. For wiregrass fires, furans were the dominant SOA precursors. We used a POA tracer ion to calculate the amount of mass lost due to gas‐phase oxidation and subsequent volatilization of semivolatile POA. Less than 5% of the POA mass was lost via multiphase oxidation‐driven evaporation during up to 2 hr of equivalent atmospheric oxidation.

Volatile organic compounds (VOCs) emitted from residential wood and crop residue burning were measured in Colorado, U.S. When compared to the emissions from crop burning, residential wood burning exhibited markedly lower concentrations of acetonitrile, a commonly used biomass burning tracer. For both herbaceous and arboraceous fuels, the emissions of nitrogen‐containing VOCs (NVOCs) strongly depend on the fuel nitrogen content; therefore, low NVOC emissions from residential wood burning result from the combustion of low‐nitrogen fuel. Consequently, the emissions of compounds hazardous to human health, such as HNCO and HCN, and the formation of secondary pollutants, such as ozone generated by NOx, are likely to depend on fuel nitrogen. These results also demonstrate that acetonitrile may not be a suitable tracer for domestic burning in urban areas. Wood burning emissions may be best identified through analysis of the emissions profile rather than reliance on a single tracer species.