We present results from a high-resolution chemical ionization time-of-flight mass spectrometer (HRToF-CIMS), operated with two different thermal desorption inlets, designed to characterize the gas and aerosol composition. Data from two field campaigns at forested sites are shown. Particle volatility distributions are estimated using three different methods: thermograms, elemental formulas, and measured partitioning. Thermogram-based results are consistent with those from an aerosol mass spectrometer (AMS) with a thermal denuder, implying that thermal desorption is reproducible across very different experimental setups. Estimated volatilities from the detected elemental formulas are much higher than from thermograms since many of the detected species are thermal decomposition products rather than actual SOA molecules. We show that up to 65% of citric acid decomposes substantially in the FIGAERO–CIMS, with ∼20% of its mass detected as gas-phase CO2, CO, and H2O. Once thermal decomposition effects on the detected formulas are taken into account, formula-derived volatilities can be reconciled with the thermogram method. The volatility distribution estimated from partitioning measurements is very narrow, likely due to signal-to-noise limits in the measurements. Our findings indicate that many commonly used thermal desorption methods might lead to inaccurate results when estimating volatilities from observed ion formulas found in SOA. The volatility distributions from the thermogram method are likely the closest to the real distributions.

Partitioning of gas-phase organic compounds to the walls of Teflon environmental chambers is a recently reported phenomenon than can affect the yields of reaction products and secondary organic aerosol (SOA) measured in laboratory experiments. Reported time scales for reaching gas-wall partitioning (GWP) equilibrium (τGWE) differ by up to 3 orders of magnitude, however, leading to predicted effects that vary from substantial to negligible. A new technique is demonstrated here in which semi- and low-volatility oxidized organic compounds (saturation concentration c* < 100 μg m–3) were photochemically generated in rapid bursts in situ in an 8 m3 environmental chamber, and then their decay in the absence of aerosol was measured using a high-resolution chemical ionization mass spectrometer (CIMS) equipped with an “inlet-less” NO3– ion source. Measured τGWE were 7–13 min (rel. std. dev. 33%) for all compounds. The fraction of each compound that partitioned to the walls at equilibrium follows absorptive partitioning theory with an equivalent wall mass concentration in the range 0.3–10 mg m–3. Measurements using a CIMS equipped with a standard ion–molecule reaction region showed large biases due to the contact of compounds with walls. On the basis of these results, a set of parameters is proposed for modeling GWP in chamber experiments.

Gas-phase low volatility organic compounds (LVOC), produced from oxidation of isoprene 4-hydroxy-3-hydroperoxide (4,3-ISOPOOH) under low-NO conditions, were observed during the FIXCIT chamber study. Decreases in LVOC directly correspond to appearance and growth in secondary organic aerosol (SOA) of consistent elemental composition, indicating that LVOC condense (at OA below 1 μg m–3). This represents the first simultaneous measurement of condensing low volatility species from isoprene oxidation in both the gas and particle phases. The SOA formation in this study is separate from previously described isoprene epoxydiol (IEPOX) uptake. Assigning all condensing LVOC signals to 4,3-ISOPOOH oxidation in the chamber study implies a wall-loss corrected non-IEPOX SOA mass yield of ∼4%. By contrast to monoterpene oxidation, in which extremely low volatility VOC (ELVOC) constitute the organic aerosol, in the isoprene system LVOC with saturation concentrations from 10–2 to 10 μg m–3 are the main constituents. These LVOC may be important for the growth of nanoparticles in environments with low OA concentrations. LVOC observed in the chamber were also observed in the atmosphere during SOAS-2013 in the Southeastern United States, with the expected diurnal cycle. This previously uncharacterized aerosol formation pathway could account for ∼5.0 Tg yr–1 of SOA production, or 3.3% of global SOA.

•New algorithms to find and assign peaks in series of complex mass spectra with overlapping peaks.•Derive both molecular and bulk chemical information from atmospheric spectra.•Develop algorithm to directly extract bulk chemical information without any peak fitting.The resolution of mass spectrometers is often insufficient to conclusively identify all peaks that may be present in recorded spectra. Here, we present new methods to extract consistent molecular and bulk level chemical information by constrained fitting of series of complex organic mass spectra with multiple overlapping peaks. Possible individual peaks in a group of overlapping peaks are identified by both defining a chemical space and by free peak fitting. If simply all possible formulas from the chemical space would be used to fit each peak, the result would not be well constrained. The free peak fitting algorithm provides information about likely peak locations. A new algorithm then reconciles the results of both methods and produces a final peak list for use in subsequent fitting, while using all available experimental constraints. Comparison to ultra-high resolution data suggests that the real peak density is substantially higher than can be resolved with the instrument resolution. Bulk chemical properties such as carbon number (nC) and carbon oxidation state (OSC) can be calculated from the fit results. For mixtures of compounds dominated by C, H, O and N, bulk properties can be reliably extracted, even though some formula assignments may remain uncertain. This ability to retrieve correct bulk parameters even if not all assigned formulas are correct originates from the relationship between mass defects of individual peaks and the chemical parameters under our CHON composition assumptions. Retrieving consistent bulk parameters across series of many mass spectra is essential for extracting time trends, e.g. for field measurements taking place over several weeks. We illustrate the fitting method using a sample data set from a chemical ionization mass spectrometer with a resolution of approximately 4000 (M/dM), operated using acetate reagent ions. Spectral simulation experiments validate the analysis method by showing good agreement of intensities for many specific ions, as well as for bulk chemical parameters. An alternative method to directly extract bulk chemical information from the raw spectra without the need of any peak assignment or peak fitting is also introduced, which shows good agreement with the peak fitting results. The latter method can be applied very rapidly without the need for complex analysis procedures, e.g. as a quick online diagnostic during data acquisition.

Oxidation flow reactors (OFRs) containing low-pressure mercury (Hg) lamps that emit UV light at both 185 and 254 nm (“OFR185”) to generate OH radicals and O3 are used in many areas of atmospheric science and in pollution control devices. The widely used potential aerosol mass (PAM) OFR was designed for studies on the formation and oxidation of secondary organic aerosols (SOA), allowing for a wide range of oxidant exposures and short experiment duration with reduced wall loss effects. Although fundamental photochemical and kinetic data applicable to these reactors are available, the radical chemistry and its sensitivities have not been modeled in detail before; thus, experimental verification of our understanding of this chemistry has been very limited. To better understand the chemistry in the OFR185, a model has been developed to simulate the formation, recycling, and destruction of radicals and to allow the quantification of OH exposure (OHexp) in the reactor and its sensitivities. The model outputs of OHexp were evaluated against laboratory calibration experiments by estimating OHexp from trace gas removal and were shown to agree within a factor of 2. A sensitivity study was performed to characterize the dependence of the OHexp, HO2/OH ratio, and O3 and H2O2 output concentrations on reactor parameters. OHexp is strongly affected by the UV photon flux, absolute humidity, reactor residence time, and the OH reactivity (OHR) of the sampled air, and more weakly by pressure and temperature. OHexp can be strongly suppressed by high OHR, especially under low UV light conditions. A OHexp estimation equation as a function of easily measurable quantities was shown to reproduce model results within 10% (average absolute value of the relative errors) over the whole operating range of the reactor. OHexp from the estimation equation was compared with measurements in several field campaigns and shows agreement within a factor of 3. The improved understanding of the OFR185 and quantification of OHexp resulting from this work further establish the usefulness of such reactors for research studies, especially where quantifying the oxidation exposure is important.

In this study, we apply several recently proposed models to the evolution of secondary organic aerosols (SOA) and organic gases advected from downtown Mexico City at an altitude of ∼3.5 km during three days of aging, in a way that is directly comparable to simulations in regional and global models. We constrain the model with and compare its results to available observations. The model SOA formed from oxidation of volatile organic compounds (V-SOA) when using a non-aging SOA parameterization cannot explain the observed SOA concentrations in aged pollution, despite the increasing importance of the low-NOx channel. However, when using an aging SOA parameterization, V-SOA alone is similar to the regional aircraft observations, highlighting the wide diversity in current V-SOA formulations. When the SOA formed from oxidation of semivolatile and intermediate volatility organic vapors (SI-SOA) is computed following Robinson et al. (2007) the model matches the observed SOA mass, but its O/C is ∼2× too low. With the parameterization of Grieshop et al. (2009), the total SOA mass is ∼2× too high, but O/C and volatility are closer to the observations. Heating or dilution cause the evaporation of a substantial fraction of the model SOA; this fraction is reduced by aging although differently for heating vs dilution. Lifting of the airmass to the free-troposphere during dry convection substantially increases SOA by condensation of semivolatile vapors; this effect is reduced by aging.

The time-of-flight aerosol mass spectrometer (ToF-AMS) determines particle size by measuring velocity after expansion into vacuum and analyzes chemical composition by thermal vaporization and electron ionization mass spectrometry (MS). Monitoring certain dynamic processes requires the ability to track changes in aerosol chemistry and size with sub-second time resolution. We demonstrate a new ToF-AMS data acquisition mode capable of collecting high-resolution aerosol mass spectra at rates exceeding 1 kHz. Coupled aerosol size and MS measurements can be made at approximately 20 Hz. These rates are about 1/10 of the physically meaningful limits imposed by the ToF-AMS detection processes. The fundamentals of the time-of-flight MS (TOFMS) data acquisition system are described and characterized with a simple algebraic model. Derived expressions show how improvements in data acquisition and computer hardware will translate into rates approaching the physical limits. Conclusions regarding limits of performance can be extended to other TOFMS that use analog signal detection in a high-speed application outside of aerosol science. The high-speed acquisition mode of the ToF-AMS enables speciated aerosol eddy covariance flux measurements, which demand precise, 10-Hz synchronization of the MS with a sonic anemometer. Flux data acquired over a forest during the BEARPEX-1 campaign are presented as an example of this new technique. For aircraft measurements, faster acquisition translates to higher spatial resolution, which is demonstrated with data from the recent NASA ARCTAS field campaign in Alaska. Finally, the fast acquisition mode is used to measure the rapid fluctuations in particle emissions of a controlled biomass burn during from the FLAME-2 experiment. To our knowledge this is currently the fastest system for acquisition of chemically resolved aerosol data.Graphical abstractResearch highlights▶ Characterization of TOFMS data acquisition system. ▶ Chemically resolved aerosol analysis at greater than 1000 Hz. ▶ Aircraft-based measurements with high temporal resolution. ▶ Observe dynamic changes in aerosol emissions from biomass burning. ▶ Demonstrate new eddy covariance flux methods.

Organonitrates (ON) are important products of gas-phase oxidation of volatile organic compounds in the troposphere; some models predict, and laboratory studies show, the formation of large, multifunctional ON with vapor pressures low enough to partition to the particle phase. Organosulfates (OS) have also been recently detected in secondary organic aerosol. Despite their potential importance, ON and OS remain a nearly unexplored aspect of atmospheric chemistry because few studies have quantified particulate ON or OS in ambient air. We report the response of a high-resolution time-of-flight aerosol mass spectrometer (AMS) to aerosol ON and OS standards and mixtures. We quantify the potentially substantial underestimation of organic aerosol O/C, commonly used as a metric for aging, and N/C. Most of the ON-nitrogen appears as ions in the AMS, which are typically dominated by inorganic nitrate. Minor organonitrogen ions are observed although their identity and intensity vary between standards. We evaluate the potential for using fragment ratios, organonitrogen ions, ions, the ammonium balance of the nominally inorganic ions, and comparison to ion-chromatography instruments to constrain the concentrations of ON for ambient datasets, and apply these techniques to a field study in Riverside, CA. OS manifests as separate organic and sulfate components in the AMS with minimal organosulfur fragments and little difference in fragmentation from inorganic sulfate. The low thermal stability of ON and OS likely causes similar detection difficulties for other aerosol mass spectrometers using vaporization and/or ionization techniques with similar or larger energy, which has likely led to an underappreciation of these species.

Organonitrates (ON) are important products of gas-phase oxidation of volatile organic compounds in the troposphere; some models predict, and laboratory studies show, the formation of large, multifunctional ON with vapor pressures low enough to partition to the particle phase. Organosulfates (OS) have also been recently detected in secondary organic aerosol. Despite their potential importance, ON and OS remain a nearly unexplored aspect of atmospheric chemistry because few studies have quantified particulate ON or OS in ambient air. We report the response of a high-resolution time-of-flight aerosol mass spectrometer (AMS) to aerosol ON and OS standards and mixtures. We quantify the potentially substantial underestimation of organic aerosol O/C, commonly used as a metric for aging, and N/C. Most of the ON-nitrogen appears as ions in the AMS, which are typically dominated by inorganic nitrate. Minor organonitrogen ions are observed although their identity and intensity vary between standards. We evaluate the potential for using fragment ratios, organonitrogen ions, ions, the ammonium balance of the nominally inorganic ions, and comparison to ion-chromatography instruments to constrain the concentrations of ON for ambient datasets, and apply these techniques to a field study in Riverside, CA. OS manifests as separate organic and sulfate components in the AMS with minimal organosulfur fragments and little difference in fragmentation from inorganic sulfate. The low thermal stability of ON and OS likely causes similar detection difficulties for other aerosol mass spectrometers using vaporization and/or ionization techniques with similar or larger energy, which has likely led to an underappreciation of these species.

A newly modified fast temperature-stepping thermodenuder (TD) was coupled to a High Resolution Time-of-Flight Aerosol Mass Spectrometer for rapid determination of chemically resolved volatility of organic aerosols (OA) emitted from individual sources. The TD-AMS system was used to characterize primary OA (POA) from biomass burning, trash burning surrogates (paper and plastic), and meat cooking as well as chamber-generated secondary OA (SOA) from α-pinene and gasoline vapor. Almost all atmospheric models represent POA as nonvolatile, with no allowance for evaporation upon heating or dilution, or condensation upon cooling. Our results indicate that all OAs observed show semivolatile behavior and that most POAs characterized here were at least as volatile as SOA measured in urban environments. Biomass-burning OA (BBOA) exhibited a wide range of volatilities, but more often showed volatility similar to urban OA. Paper-burning resembles some types of BBOA because of its relatively high volatility and intermediate atomic oxygen-to-carbon (O/C) ratio, while meat-cooking OAs (MCOA) have consistently lower volatility than ambient OA. Chamber-generated SOA under the relatively high concentrations used in traditional experiments was significantly more volatile than urban SOA, challenging extrapolation of traditional laboratory volatility measurements to the atmosphere. Most OAs sampled show increasing O/C ratio and decreasing H/C (hydrogen-to-carbon) ratio with temperature, further indicating that more oxygenated OA components are typically less volatile. Future experiments should systematically explore a wider range of mass concentrations to more fully characterize the volatility distributions of these OAs.

Organic aerosol (OA) emissions from motor vehicles, meat-cooking and trash burning are analyzed here using a high-resolution aerosol mass spectrometer (AMS). High resolution data show that aerosols emitted by combustion engines and plastic burning are dominated by hydrocarbon-like organic compounds. Meat cooking and especially paper burning emissions contain significant fractions of oxygenated organic compounds; however, their unit-resolution mass spectral signatures are very similar to those from ambient hydrocarbon-like OA, and very different from the mass spectra of ambient secondary or oxygenated OA (OOA). Thus, primary OA from these sources is unlikely to be a significant direct source of ambient OOA. There are significant differences in high-resolution tracer m/zs that may be useful for differentiating some of these sources. Unlike in most ambient spectra, all of these sources have low total m/z 44 and this signal is not dominated by the CO2+ ion. All sources have high m/z 57, which is low during high OOA ambient periods. Spectra from paper burning are similar to some types of biomass burning OA, with elevated m/z 60. Meat cooking aerosols also have slightly elevated m/z 60, whereas motor vehicle emissions have very low signal at this m/z.

A recently developed method to rapidly quantify the elemental composition of bulk organic aerosols (OA) using a high-resolution time-of-flight aerosol mass spectrometer (HR-ToF-AMS) is improved and applied to ambient measurements. Atomic oxygen-to-carbon (O/C) ratios characterize the oxidation state of OA, and O/C from ambient urban OA ranges from 0.2 to 0.8 with a diurnal cycle that decreases with primary emissions and increases because of photochemical processing and secondary OA (SOA) production. Regional O/C approaches ∼0.9. The hydrogen-to-carbon (H/C, 1.4–1.9) urban diurnal profile increases with primary OA (POA) as does the nitrogen-to-carbon (N/C, ∼0.02). Ambient organic-mass-to-organic-carbon ratios (OM/OC) are directly quantified and correlate well with O/C (R2 = 0.997) for ambient OA because of low N/C. Ambient O/C and OM/OC have values consistent with those recently reported from other techniques. Positive matrix factorization applied to ambient OA identifies factors with distinct O/C and OM/OC trends. The highest O/C and OM/OC (1.0 and 2.5, respectively) are observed for aged ambient oxygenated OA, significantly exceeding values for traditional chamber SOA, while laboratory-produced primary biomass burning OA (BBOA) is similar to ambient BBOA, O/C of 0.3–0.4. Hydrocarbon-like OA (HOA), a surrogate for urban combustion POA, has the lowest O/C (0.06–0.10), similar to vehicle exhaust. An approximation for predicting O/C from unit mass resolution data is also presented.

Ambient sampling was conducted in Riverside, California during the 2005 Study of Organic Aerosols in Riverside to characterize the composition and sources of organic aerosol using a variety of state-of-the-art instrumentation and source apportionment techniques. The secondary organic aerosol (SOA) mass is estimated by elemental carbon and carbon monoxide tracer methods, water soluble organic carbon content, chemical mass balance of organic molecular markers, and positive matrix factorization of high-resolution aerosol mass spectrometer data. Estimates obtained from each of these methods indicate that the organic fraction in ambient aerosol is overwhelmingly secondary in nature during a period of several weeks with moderate ozone concentrations and that SOA is the single largest component of PM1 aerosol in Riverside. Average SOA/OA contributions of 70−90% were observed during midday periods, whereas minimum SOA contributions of ∼45% were observed during peak morning traffic periods. These results are contrary to previous estimates of SOA throughout the Los Angeles Basin which reported that, other than during severe photochemical smog episodes, SOA was lower than primary OA. Possible reasons for these differences are discussed.

We report the quantification of ambient particle-bound polycyclic aromatic hydrocarbons (PAHs) for the first time using a real-time aerosol mass spectrometer. These measurements were carried out during the Mexico City Metropolitan Area field study (MCMA-2003) that took place from March 29 to May 4, 2003. This was the first time that two different fast, real-time methods have been used to quantify PAHs alongside traditional filter-based measurements in an extended field campaign. This paper focuses on the technical aspects of PAH detection in ambient air with the Aerodyne AMS equipped with a quadrupole mass analyzer (Q-AMS), on the comparison of PAHs measured by the Q-AMS to those measured with the other two techniques, and on some features of the ambient results.PAHs are very resistant to fragmentation after ionization. Based on laboratory experiments with eight PAH standards, we show that their molecular ions, which for most particulate PAHs in ambient particles are larger than 200 amu, are often the largest peak in their Q-AMS spectra. Q-AMS spectra of PAH are similar to those in the NIST database, albeit with more fragmentation. We have developed a subtraction method that allows the removal of the contribution from non-PAH organics to the ion signals of the PAHs in ambient data. We report the mass concentrations of all individual groups of PAHs with molecular weights of 202, 216, 226 + 228, 240 + 242, 250 + 252, 264 + 266, 276 + 278, 288 + 290, 300 + 302, 316 and 326 + 328, as well as their sum as the total PAH mass concentration.The time series of the Photoelectric Aerosol Sensor (PAS) and Q-AMS PAH measurements during MCMA-2003 are well correlated, with the smallest difference between measured PAH concentrations observed in the mornings when ambient aerosols loadings are dominated by fresh traffic emissions. The Q-AMS PAH measurements are also compared to those from GC–MS analysis of filter samples. Several groups of PAHs show agreement within the uncertainties, while the Q-AMS measurements are larger than the GC–MS ones for several others. In the ambient Q-AMS measurements the presence of ions tentatively attributed to cyclopenta[cd]pyrene and dicyclopentapyrenes causes signals at m/z 226 and 250, which are significantly stronger than the signals in GC–MS analysis of filter samples. This suggests that very labile, but likely toxic, PAHs were present in the MCMA atmosphere that decayed rapidly due to reaction during filter sampling, and this may explain at least some of the differences between the Q-AMS and GC–MS measurements.