A new approach to detection of highly oxidized products from ethylbenzene oxidation using matrix assisted ionization in vacuum-mass spectrometry

Abstract

Aromatic hydrocarbon oxidation by hydroxyl radicals (OH) efficiently forms secondary organic aerosol (SOA) particles, likely through formation of low volatility highly oxidized products which form and grow these particles. However, the nature of these oxidation products and their mechanisms of formation are not sufficiently well characterized to be included in models with confidence, leaving gaps in predicting SOA particle formation and growth. Challenges for highly oxidized product characterization include the small concentrations, low volatility, propensity for loss on surfaces and instability upon heating which is often part of the analytical detection techniques. A new approach is reported here in which oxidation products generated from the OH oxidation of ethylbenzene (C₈H₁₀) are condensed on glutaric acid particles. These seed particles provide a large condensation surface area and also act as the matrix for detection of surface-bound products via matrix assisted ionization in vacuum-mass spectrometry (MAIV-MS). Series of oxidation products consistent with the formation of peroxides (ROOH and ROOR) from several formation pathways were detected, indicating that they were scavenged from the gas phase. For comparison, gas-phase measurements using nitrate chemical ionization mass spectrometry (NO₃⁻ CIMS) showed highly oxidized C₈ products with up to 8 oxygen atoms, and weaker signals from highly oxidized C₁₆ products with up to 12 oxygen atoms. MAIV-MS of the condensable products showed relatively higher signal intensities for larger mass ROOR products compared to NO₃⁻ CIMS, supporting the importance of ROOR in the formation and growth of particles. Two product series with unexpectedly low numbers of hydrogens that have not been reported previously, C₁₆H₁₆O_5–8 and C₁₆H₁₈O_5–8, were detected with MAIV. Prominent C₁₆ accretion products were observed from a range of formation pathways, including multiple OH attacks to the aromatic ring and contributions from RO₂ isomerization. The use of MAIV-MS for condensable product detection provides direct insights that complement other gas phase measurements and leads to a more comprehensive picture of SOA particle formation and growth.

---

Environmental significance

The oxidation of aromatic hydrocarbons leads to secondary organic aerosol (SOA) particles that have adverse effects on health, visibility and climate. However, there are challenges associated with detecting the highly oxidized products most likely to participate in forming and growing particles, in large part due to their losses to walls and other surfaces that cause them to go undetected. A new approach is presented in which oxidation products are condensed on high surface area matrix particles, minimizing these losses. The matrix allows soft ionization of the condensed products to lend insight into a number of products and pathways that contribute to new particle formation and growth.

Introduction

Atmospheric aerosol particles have negative impacts on human health, air quality and visibility, and significantly impact Earth's radiative balance.^1–10 Oxidation of volatile organic compounds (VOC) leads to low volatility gas-phase products that can form secondary organic aerosol (SOA) particles and/or contribute to the growth of existing particles.^11,12 Oxidation of anthropogenic compounds, including aromatic hydrocarbons, is a major contributor to SOA formation globally, especially in urban regions where the contribution of anthropogenic SOA is believed to be underestimated.^13–17

Highly oxygenated organic molecules (HOM) comprise a subset of VOC oxidation products that contribute to SOA formation and growth.^18,19 For aromatic systems, the main oxidation pathway is OH addition to the ring, followed by O₂ addition to a neighboring carbon on the ring. This forms RO₂ products in which aromaticity is broken (Scheme S1, RO₂-A). These RO₂ can subsequently undergo isomerization in which the –O–O– moiety forms a second bond with the ring, and O₂ is added to give a bicyclic peroxy radical (Scheme S1, C₈H₁₁O₅·RO₂).¹⁷ Intramolecular H-shifts (autoxidation) from an alkyl substituent have also been proposed that can propagate the radical chain reaction and lead to more O₂ additions.^17,20–22 Termination reactions of RO₂ with HO₂ yield hydroperoxides, whose large numbers of oxygen atoms relative to carbon result in products with high molecular weights and low volatilities.^{18–21,23–26} Low-volatility products can also be generated via RO₂–RO₂ accretion reactions. For example, Berndt et al.²⁵ reported increasing rate constants (>10⁻¹¹ cm³ per molecule per s) for gas-phase ROOR formation as R groups became more oxygenated, and suggested isomerization and O₂ addition as the mechanism leading to the large number of oxygen atoms in accretion products.

Additional pathways have been proposed that can lead to large numbers of oxygen atoms in the products.^{17,21,26–32} For example, OH addition to the aromatic ring forms products with the aromaticity retained (e.g., phenol products, Scheme S1 and S2a).¹⁷ Polyhydroxylated products have been reported as evidence for multiple OH attacks during aromatic oxidation.^28,30,33 Schwantes et al.²⁸ also observed a number of quinone products arising from multiple OH attacks, leading to reduced numbers of hydrogen atoms relative to the precursor. These highly oxygenated hydroxy and quinone products are expected to have low volatilities. A less often considered pathway is that of an additional attack of OH on a double bond remaining in an ROOH formed through RO₂–HO₂ termination (e.g. Scheme S2b).^21,29 Upon each new OH attack, O₂ addition and isomerization again become viable pathways to add oxygen atoms, along with formation of closed-shell products and additional OH attacks.

Given the “mixing” of pathways that can occur (e.g. Scheme S1–S3), the term highly oxidized products, HOP, is used here to refer to the mixture of products formed through either (i) isomerization and O₂ addition, (ii) the phenolic path, or (iii) multiple OH attacks on closed-shell ROOH products, while HOM is used to refer to products from isomerization and autoxidation pathways.²⁶ While a number of pathways have been investigated or proposed, questions remain regarding which products are most likely to participate in forming new particles,³¹ in large part due to challenges in detecting low volatility precursors.

Chemical ionization mass spectrometry (CIMS) techniques were the first to detect HOM and other HOP using reagent ions (e.g. NO₃⁻, I⁻, CH₃CO₂⁻, CF₃O⁻, and NH₄⁺) that form adducts with HOP, and in some cases, with RO₂ radicals.^{18,19,24,28,34} Major advances have been made with CIMS toward identifying new gas phase species and understanding their evolution under a variety of conditions. CIMS is also being expanded to detect HOP in particles.^30,35–37 However, there are potential complications such as decomposition during analysis due to the need to vaporize particles for MS analysis.³⁶ Another challenge is the rapid loss of gas-phase oxidation products, and particles formed from them, to surfaces before detection.^28,38,39

This study demonstrates the use of an emerging technique, matrix assisted ionization in vacuum-mass spectrometry (MAIV-MS),^40,41 for detection of a wide range of HOP. Solid matrix particles are used to provide a large condensation surface area for gas-phase HOP formed from OH oxidation of ethylbenzene. In MAIV-MS, matrix ions along with co-adsorbed species are spontaneously emitted from the surface of charged matrix particles and detected in the absence of an external ion source. The soft conditions and low heat present at the MS inlet result in low background signals and facilitate the detection of condensed HOP, including thermally labile peroxides.⁴⁰ For comparison, gas-phase products are also characterized using high resolution nitrate-CIMS to obtain molecular formulae and confirm the formation of HOP with up to 12 oxygen atoms. These combined particle-gas data provide unique insights into the oxidation pathways leading to new particle formation and growth from the OH oxidation of ethylbenzene, and highlight the utility of MAIV-MS for low volatility products that are otherwise difficult to detect.

Experimental methods

Flow reactor

A schematic of the flow system used for the oxidation studies is shown in Fig. 1. The flow reactor is a 95 cm long borosilicate glass and quartz reactor with a volume of 1.6 L (4.6 cm ID). Ethylbenzene (Sigma Aldrich, 99.8%), or in some experiments toluene (Fisher, 99.5+%), and benzene (EMD, 99.7%), were introduced as a liquid from a syringe pump and evaporated into a flow of 0.7 L min⁻¹ clean air entering the flow reactor. Hydroxyl radicals (OH) were generated in situ from the ozonolysis of tetramethylethylene (2,3-dimethyl-2-butene, (TME), Sigma-Aldrich, >99%).^42–45 Liquid TME was introduced with a syringe pump and evaporated into the same air flow of 0.7 L min⁻¹ air. Ozone was prepared by flowing oxygen (Praxair, 99.993%) through a mercury Pen-Ray lamp (UVP, model 11SC-2) at 0.3 L min⁻¹. A third flow of 3.5 L min⁻¹ of either matrix particles or particle-free air was introduced, depending on the type of mass spectrometry measurement to be carried out. For either case, the total reactor flow was 4.5 L min⁻¹, giving a residence time of 20 s. Initial concentrations in the reactor were 4–150 ppm for each aromatic and ∼7 ppm for both TME and O₃. The O₃ concentration was measured with UV/visible spectroscopy (Ocean Optics, HR4000). The steady-state concentration of OH was estimated to be ∼(1–2) × 10⁹ cm⁻³. In a subset of experiments, methanol (Fisher, HPLC grade) was introduced to increase HO₂ concentrations by injecting the liquid with an additional syringe pump and evaporating it into the same 0.7 L min⁻¹ air flow as the ethylbenzene and TME.

Fig. 1 Schematic of the flow system used for OH oxidation of ethylbenzene in the presence or absence of glutaric acid matrix particles. The inset is an enlarged view of highly oxidized gasphase products, HOP, either (1) nucleating to form new particles that are not detected with MAIV or CIMS, but are detected with the SMPS (2) condensing to the matrix particle surface where they are detected using MAIV, or (3) remaining in the gas phase where they are detected using NO₃⁻ CIMS.

Oxidation products were measured using two different types of mass spectrometry, one for condensed-phase products and one for gas-phase products. For condensed-phase product measurements using MAIV, polydisperse glutaric acid (GA) matrix particles were added to the flow reactor. GA particles were generated by atomizing a solution of 10–80 mM glutaric acid (Sigma Aldrich, 99%) in 18.2 MΩ cm water (Nanopure, Barnstead, Thermo Scientific) with air using a constant output atomizer (TSI, Model 3076) at a flow rate of 3.5 L min⁻¹. GA particles flowed through two silica gel diffusion dryers before entering the flow reactor, resulting in low relative humidity (<5% RH) as measured in the exhaust flow (Vaisala, HMP234). For gas-phase product measurements using CIMS, a flow of 3.5 L min⁻¹ air was used in place of matrix particles. All experiments were carried out at low RH and at room temperature (22 ± 1 °C).

Particle size distributions were measured using a Scanning Mobility Particle Sizer (SMPS), which included an electrostatic classifier (TSI, model 3082) with a long DMA (TSI, model 3081A) and an ultrafine condensation particle sizer (TSI, model 3756). A Po-210 neutralizer (NRD, model P-2021) was included at the sample inlet as required for charge equilibration. The sample flow was typically 0.3 L min⁻¹ with a sheath air flow of 3.0 L min⁻¹ to measure diameters in the range ∼15–650 nm. In some cases, sample and sheath flows of 1.5 L min⁻¹ and 15 L min⁻¹, respectively, were used to measure diameters of ∼6–220 nm. AIM® software was used to record the data. Lognormal fitting was carried out to account for bimodal size distributions in cases where new particles were formed in addition to the matrix particles used for MAIV.

Matrix assisted ionization in vacuum-mass spectrometry (MAIV-MS)

MAIV-MS was carried out using a triple quadrupole mass spectrometer (Waters, Xevo TQ-S) in which the ion source had been removed as described previously.⁴¹ Briefly, the use of matrix particles, which are initially partially charged through atomization,^46–49 leads to soft ionization of adsorbed species in the surface layers of the dried particles as they enter the sub-atmospheric pressure region of the mass spectrometer. Molecules that are co-adsorbed with the matrix are also ejected as ions, including particularly labile species such as peroxides.⁴⁰ Glutaric acid was used as it has been shown in our previous studies to be the most effective matrix material.⁴¹ Recent molecular dynamics modeling of several diacids demonstrated that the low melting point of GA, relative to other diacids, leads to reduced diffusion coefficients in the surface layers which may be responsible for increased mobility, faster proton transfer at the surface, and efficiency as an ionization matrix.⁵⁰

Oxidation of aromatic compounds occurred in the presence of GA matrix particles so that low-volatility products could condense onto the particles. In some cases, new particles nucleated (Fig. 1 inset), which was minimized by increasing the GA matrix surface area. The flow from the flow reactor, including gases and particles, was guided to the MS through Teflon tubing and flowed through a ceramic tube (6 cm length, 1/4 in. diameter) and a metal elbow (Vapur® Interface, ∼5 cm length) at atmospheric pressure leading up to the MS inlet cone. The flow into the mass spectrometer inlet was 1.5 L min⁻¹, and the excess flow from the flow reactor was directed to an exhaust. The conditions at the MS inlet were as follows: the temperature-controlled ion block (source temperature) was set to 70 °C, cone voltage at 30 V, and source offset at 50 V. Although the ion block temperature was set to 70 °C, particles experience a lower measured temperature of ∼40 °C and the residence time in this heated region is ∼100 ms.⁴¹

Spectra were collected in positive ion mode using multichannel analysis (MCA), in which spectra were summed for 15 minutes for increased sensitivity. Oxidation products formed ammonium ion adducts, [M + NH₄]⁺, due to the presence of ammonia in laboratory air and the strong tendency for oxygenated compounds to form ammonium adducts.⁵¹ Some protonated adducts were observed for lower molecular weight, less oxygenated compounds, e.g. [M + H]⁺ for glutaric acid at m/z 133, but only ammoniated peaks were considered for HOP. Peak intensities for a series of products were collected for experiments with initial aromatic concentrations of 4–150 ppm. Intensities were normalized to the peak intensity of glutaric acid at m/z 133 [GA + H]⁺ from the initial spectrum with no product coating (I_GA,0). Note that the resolution of the mass spectrometer is not sufficiently high to provide accurate masses. Data were collected using MassLynx (Waters) and processed using IgorPro (Wavemetrics, Inc.).

Chemical ionization mass spectrometry (CIMS)

Analysis of gas phase oxidation products was carried out using a high-resolution time-of-flight mass CIMS (LTOF, Aerodyne Research, m/Δm ∼8000) with nitrate reagent ions, which are selective for HOM and HOP detection.^34,52 Nitrate reagent ions were generated by flowing 10 cm³ min⁻¹ N₂ (AirGas, UHP) over a vial of concentrated nitric acid (Sigma-Aldrich, 70 wt%, redistilled). The HNO₃ vapor was then passed through a Po-210 radioactive source (NRD, model P-2021) to generate the nitrate ions. An additional N₂ flow of 1 L min⁻¹ was added to adjust the reagent ion concentration and the combined flow was directed into a custom-built transverse ionization inlet orthogonal to the orifice of the mass analyzer.⁵³ Sample inlet flow was adjusted to 4.1 ± 0.1 L min⁻¹, with excess flow exhausted.

Data were analyzed using Tofware version 4.0.1 (Aerodyne Research). Spectra were collected every 1 s and pre-averaged in 10 s intervals. Oxidation products were detected as nitrate adducts, [M + NO₃]⁻, and m/z calibrations utilized the ions NO₂⁻, NO₃⁻, HNO₃·NO₃⁻, (HNO₃)₂·NO₃⁻, and HSO₄⁻. Results of m/z calibrations were auto-improved within Tofware to maintain mass accuracy within ±10 ppm, providing accurate masses and elemental formulae. Intensities were normalized to the reagent ion (NO₃⁻) intensity from the same spectrum, and background intensities were subtracted using a spectrum of clean air collected before each experiment. No corrections were made for wall losses. CIMS data were also not corrected for sample dilution that occurred due to the N₂ flow in the CIMS inlet, which would be applied uniformly to all peaks (∼25% correction). Since only relative intensity comparisons were made between MAIV and CIMS, dilution corrections were unnecessary.

Kinetics modeling

Gas-phase kinetics modeling was carried out at the concentrations and residence time in the flow reactor using the Master Chemical Mechanism (MCM)^54–56 to predict radical concentrations.

Results and discussion

Fig. 2 shows typical MAIV mass spectra of GA particles in positive ion mode in the presence of ethylbenzene (EB) + OH oxidation products over a range of initial concentrations ([EB]_o), as well as the spectrum of glutaric acid matrix particles in the absence of EB and OH. The GA particles act both as a matrix to promote ionization of adsorbed species^{41,50,57–60} as well as providing a condensation surface for low volatility oxidation products. Peaks are seen over a wide mass range, representing a large number of products. Detailed examination of a set of products with eight (C₈) or sixteen (C₁₆) carbon atoms and their behavior in the presence of MAIV seed particles was carried out here. For reference, a MAIV spectrum of OH-oxidized GA particles in the absence of ethylbenzene is also shown (Fig. 2e, orange), indicating that the peaks appearing in the presence of ethylbenzene are not explained by GA or TME oxidation by OH (with the exception of m/z 208, discussed below).^40,61

Fig. 2 Typical MAIV (+) mass spectra of glutaric acid particles coated with ethylbenzene + OH oxidation products for initial ethylbenzene concentrations of (a) 75 ppm, (b) 19 ppm, and (c) 4 ppm. Part (d, black) is GA alone with no OH or EB present; part (e, orange) is GA alone with OH present. Grey vertical lines mark where products are on each spectrum. Oxidation products are detected as ammonium adducts.

The most intense peaks in MAIV-MS are those from C₁₆, formed in bimolecular reactions of radicals formed in the oxidation. Addition of OH to EB can result in a phenolic product. Alternatively, a carbon-centered radical can form and add O₂ to generate RO₂, which have a number of possible fates, summarized in a simplified mechanism as follows:⁶²

RO₂ + HO₂ → ROOH + O₂ (1)

RO₂ + RO₂ → ROOR + O₂ (2a)

→ R=O + ROH (2b)

→ RO + RO + O₂ (2c)

Reaction (1) generates hydroperoxides with 8 carbon atoms while reaction (R2a) generates peroxides with 16 carbons. Closed-shell products with low oxygen content can be generated by reaction (R2b) and RO radicals are formed in (R2c). Reaction (R3) can lead to new peroxy radicals, QO₂, with incorporation of many oxygen atoms while preserving the number of hydrogen and carbon atoms.^{17,20,21,25,29,30,32} The focus here is on pathways (R1) and (R2a) that lead to HOP, for which examples have been illustrated in Schemes S1–S3.

To confirm that the source of the smaller C₈ compounds is one aromatic unit and two for the larger C₁₆ compounds, MAIV spectra from benzene and toluene oxidation were also collected (Fig. S1). A number of peaks above m/z 300 separated by Δ m/z = 28 (two CH₂ groups) were observed, which is consistent with products containing two C₈ precursor units. Below m/z 300, some series of peaks showed Δm/z = 14 (one CH₂ group), indicative of products containing one C₈ unit. Some benzene peaks did not follow the homologous trends, likely because it lacks a reactive alkyl side chain where H-migration can occur.²⁰

To identify the gas-phase product distribution and establish molecular formulae, oxidation products of ethylbenzene were measured with high resolution NO₃⁻ CIMS under the same initial conditions but in the absence of matrix particles. As discussed in the SI (Text S1 and Fig. S2 and S3), CIMS is not very sensitive to self-nucleated particles that are formed at higher reactant concentrations. Fig. 3 shows typical NO₃⁻ CIMS spectra for initial EB concentrations of 4 and 38 ppm in the absence of GA particles, in which the same set of products was examined as for MAIV. The CIMS measurements provide accurate masses and molecular formulae, showing C₈ products with a range of 3 to 8 oxygen atoms (i.e. C₈H₁₀O_3–5, C₈H₁₂O_5–8, and C₈H₁₄O_6–8), and C₁₆ products with 6 to 12 oxygen atoms and different numbers of H atoms. In contrast to the MAIV spectra, the most intense CIMS peaks are seen at smaller m/z, primarily corresponding to C₈ hydroperoxides (ROOH), with less intense peaks at higher masses corresponding to C₁₆ peroxides (ROOR).

Fig. 3 Typical nitrate-CIMS spectra of OH oxidation products of EB at initial concentrations of 4 and 38 ppm (in the absence of glutaric acid particles). The bottom panel includes an inset with the y-axis scale increased for the C16 region. Oxidation products are detected as nitrate ion adducts.

Fig. 4 compares the normalized peak intensities for CIMS (relative to NO₃⁻) and MAIV (relative to I_GA,0), showing that the relative product intensity distributions observed using these two techniques are very different. The larger relative C₁₆ intensities for MAIV and higher C₈ signals in CIMS are consistent with MAIV detecting primarily lower volatility species that condense on GA particles, and CIMS detecting lower molecular mass gas-phase products. It should be noted that the signals in Fig. 4 do not represent relative concentrations because the relative ionization efficiencies and hence sensitivities of CIMS and MAIV differ. For example, recent comparisons of the selectivity of product-ion adducts have shown that NO₃⁻ generally forms stable adducts with molecules containing approximately 5–10 oxygen atoms, while NH₄⁺ adducts form with less oxidized species, but a wider range containing approximately 2–9 oxygen atoms.^34,63 However, the greatly enhanced signals at low masses for CIMS, and at higher masses for MAIV, clearly reflect differences in the relative amounts in the gas phase and those condensed on GA surfaces. Table S1 and Fig. S4 summarize a common set of peaks observed in MAIV and CIMS, as well as their calculated volatilities and carbon oxidation states.^26,64–66

Fig. 4 Relative intensities of ROOH and ROOR species from OH oxidation of ethylbenzene detected in (a) NO₃⁻ CIMS (hatched) and (b) MAIV-MS (solid) at an initial concentration of 38 ppm EB. Grey shaded regions separate groups of products by their number of hydrogens.

Even as [EB]_o is lowered, which decreases the relative importance of RO₂-RO₂ reactions (R2a), the MAIV intensities for most C₁₆ ROOR remain larger than those for ROOH (Fig. S5). The observation of a similar set of products in CIMS (but at smaller intensities) is consistent with ROOR forming in the gas phase followed by adsorption on GA particles. There was no evidence for cross reactions with glutaric acid in MAIV spectra. (As discussed in the SI, Text S2, this is expected since the loss of OH to EB was much faster than that to GA.) Since most products observed here have been reported in previous gas-phase aromatic oxidation studies in the absence of seed particles, the products likely formed in the gas phase and were subsequently scavenged to the matrix particle surface. While more efficient adsorption of ROOR is expected given their higher molecular weights relative to ROOH, this surface-sensitive measurement provides a method of directly detecting the adsorbed products. Fig. S5 is organized by the number of hydrogen atoms in each series, and it can be seen that several less intense C₁₆ series in CIMS tend to be more prominent in MAIV, which demonstrates that many products are more efficiently scavenged to the matrix particles in MAIV. Fig. S5 also shows that as [EB]_o is lowered, the accretion products with the highest intensities in CIMS are those formed through unimolecular isomerization (22 hydrogen atoms), indicating a reduction in RO₂–RO₂ reactions.

To provide structural insight into the products, HO₂ was increased by adding methanol:⁶⁷

This increases the importance of reaction (R1) of RO₂ with HO₂ to form ROOH relative to (R2) and (R3). While the structures cannot be confirmed without standards, the presence of CH₃OH does indeed lead to increases in the relative intensities of C₈ products (Fig. S6), suggesting they are primarily ROOH products. Most C₁₆H₂₂O_x products decreased upon addition of CH₃OH, consistent with an increase in RO₂ termination (R1) and decrease in ROOR formation (R2a). Several C₁₆ products increased upon CH₃OH addition, including C₁₆H₁₆O_5–8, C₁₆H₁₈O_5–8, C₁₆H₂₀O_7–12, and C₁₆H₂₆O_10–12. This suggests their units contain both –OOH and –O–O– functional groups that respond in complex ways to HO₂, and given their low numbers of hydrogen atoms relative to ethylbenzene (C₈H₁₀), are formed from paths involving multiple OH attacks on ROOH. For example, HOP that contain 26 hydrogens, C₁₆H₂₆O_10–12, can form through hydroperoxide pathways (Scheme S2a and b) to first generate ROOH that can undergo a second OH addition. Formation pathways for the other C₁₆ products that increased upon CH₃OH addition, C₁₆H₁₆O_5–8, C₁₆H₁₈O_5–8, and C₁₆H₂₀O_7–12, are yet unknown, and those with 16 and 18 hydrogens have not been reported.

To further demonstrate scavenging of HOP by GA particles, the effects of particle surface area were explored by varying the concentration of GA in the atomizer, which changes the particle size distribution. Fig. S7 shows the size distributions of uncoated matrix particles resulting from three different solution concentrations, along with their average mode diameters and total number and surface area concentrations. Fig. 5 shows the size distributions in the absence and presence of the EB-OH products for the three surface areas. At low surface areas and higher [EB]_o, gas-phase HOP formed new particles from products that failed to condense on the matrix. Taking the highest [EB]_o as an example, the formation of HOP in the presence of the lowest surface area matrix particles led to an increase in particle number concentration from GA alone ((3.8 ± 0.2) × 10⁶ particles cm⁻³) to GA + products ((5.8 ± 0.1) × 10⁶ particles cm⁻³), an increase of ∼50% (Fig. 5a, pink). In contrast, for the highest surface area (and largest number) of matrix particles, the number concentration increased by only ∼10% during the EB-OH reaction ((7.9 ± 0.4) × 10⁶ to (8.7 ± 0.2) × 10⁶ particles cm⁻³). Thus, more efficient scavenging of products on the GA and much less new particle formation occurred when a large condensation sink was available (Fig. 5c, pink). This is consistent with the work of Garmash et al.²⁹ who observed that introducing seed particles during benzene oxidation led to a significant scavenging of highly oxidized products that had been undetected in the gas phase.

Fig. 5 Particle size distributions as a function of [EB]_o = 0–150 ppm generated by atomizing solutions of (a) 10 mM GA for low surface area, (b) 20 mM GA for medium surface area, and (c) 80 mM for high surface area. Bimodal lognormal fits to size distributions were carried out with the following: for mode 1 the area of the GA particles was held constant to within ± 20% to represent a constant GA number concentration, but its mode diameter was unconstrained to allow increases due to condensational growth. For mode 2, the smaller new particle mode, the fit was unconstrained in area and diameter. Initial GA particle number and surface area concentrations are provided in Fig. S7. Total particle number concentrations at 0 ppm and 150 ppm EB, respectively, are (a) (3.8 ± 0.2) × 106 and (5.8 ± 0.1) × 106 particles cm⁻³, (b) (5.6 ± 0.3) × 106 and (6.7 ± 0.1) × 106 particles cm⁻³, (c) (7.9 ± 0.4) × 106 and (8.7 ± 0.2) × 106 particles cm⁻³. Uncertainties are 1 s of the average of N = 3–5 samples.

The bimodal lognormal fitting of SMPS data (Fig. 5) also shows that when the number and diameter of matrix particles is increased, a thinner product coating on GA is formed. For example, the mode for low surface area matrix particles increases from 80 to 109 nm as [EB]_o increases from 0 to 150 ppm (Fig. 5a), giving a coating thickness of ∼14 nm. The high surface area matrix particles grow from 113 to ∼118 nm (Fig. 5c), giving a coating of only ∼2 nm. A decrease in GA signal intensity (m/z 133) upon reaction is also indicative of the product coating thickness, as shown in Fig. S8. At high surface area, the GA intensity decreases by only ∼2× for all [EB]_o, i.e. thin product coatings. However, at low surface area, GA intensity decreases by a factor of 30 due to the thick condensed HOP coating that inhibits GA ion ejection and lowers all signals, as previously observed.^41,68 Further detail on the effect of product coatings is presented in Text S3 and Fig. S9–S10.

Products and pathways of formation

While the focus of this study is on the utility of MAIV-MS for scavenging and detecting HOP, it also provides some insight into HOP most likely to participate in new particle formation and growth. Fig. S11 shows the effects of the initial EB concentration and GA surface area on the products. Estimated vapor pressures and carbon oxidation states (Table S1 and Fig. S4) show that C₁₆ products generally fall in the categories of extremely low VOC (ELVOC, −8.5 ≤ log₁₀ C⁰ ≤ −4.5) and ultra low VOC (ULVOC, log₁₀ C⁰ < −8.5) and are expected to more readily condense than C₈ products (mostly semi-VOC (SVOC, −0.5 ≤ log₁₀ C⁰ ≤ 2.5)).²⁶ However, previous studies have shown that in some cases particle growth is not accurately predicted by volatility and OS_C.^31,69–71 Thus, some C₈ intensities increase with the surface area of the scavenger GA particles, and also increase with decreasing [EB]_o due to decreased competition from the RO₂ + RO₂ reaction (R2a).

From Fig. 5, low surface area leads to new particle formation, where many products go undetected. At high [EB]_o, as the surface area increases, the relative distribution of C₁₆ products changes, leaving a few select products at high intensity (Fig. S11a–c). Notably, C₁₆H₂₂O₈ is relatively high intensity (Fig. S11c, f and i). This product results from an RO₂-RO₂ reaction (R2a) of the first bicyclic peroxy radical formed. As shown in Scheme S1 (red box), the bicyclic peroxy radical is formed after one OH attack, two O₂ additions, and one isomerization step. It is not surprising that accretion plays a role at high [EB]_o given that MCM modeling^54–56 indicates a ΣRO₂/HO₂ ratio of ∼10 (Fig. S12b).

As [EB]_o is lowered, the relative importance of C₁₆H₂₂O₈ slightly decreases and two series that have not been reported before become prominent as the surface area is increased to suppress new particle formation. The C₁₆H₁₆O_5–8 becomes prominent as surface area is increased (Fig. S11g–i). This series also becomes more prominent as [EB]_o is decreased, i.e. Fig. S11c, f and i. As discussed above, the formation pathways for the C₁₆H₁₆O_5–8 and C₁₆H₁₈O_5–8 products that increased upon CH₃OH addition are unknown, but are suggested to have both –OOH and –O–O– moieties formed from multiple OH additions to ROOH products and accretion. One path for generating products containing two precursor units with this low number of hydrogens relative to the initial reactant (C₈H₁₀) is through the formation of quinones in one or both precursor units. Quinone products containing one precursor unit were reported in toluene-OH oxidation using CF₃O⁻ CIMS, albeit with very low intensities, and were proposed to form from di- and trihydroxytoluene that undergoes additional OH addition.²⁸ They may also form following reaction (R2c) when RO + O₂ leads to R=O + HO₂.⁶² The data reported here suggest that oxidized C₁₆ products with low numbers of hydrogens, likely quinones, are scavenged by high surface area matrix particles where new particle formation and wall loss are minimized. Most other highly oxidized C₁₆ products become less intense at this high surface area. This points to a prominent role for the C₁₆H_16,18O_5–8 series in new particle formation and growth and highlights the complementary nature of MAIV for detecting highly oxygenated products.

Summary and conclusions

Condensing highly oxidized products on MAIV matrix particles provides a new approach to detecting the low volatility precursors to SOA and insight into the mechanisms and specific products contributing to new particle formation and growth. The matrix particles provide a condensation surface as well as soft ionization that allows detection of gas-phase HOP in the absence of heat or external ionization source. This includes labile peroxides with 3–12 oxygen atoms and a range of hydrogen atoms that suggest multiple formation pathways for HOP. Increasing the matrix particle surface area facilitated adsorption of HOP that, in the absence of a condensation sink, would go undetected because of formation of new particles or loss to the walls.

The observation of a plethora of products and pathways in Schemes S1–S3 indicates the importance of multiple OH additions in forming HOP that condense to surfaces and/or form new particles, depending on the experimental conditions. The high relative intensities of C₁₆ products from multiple pathways points to their importance in new particle formation and growth, despite many series having smaller intensities in gas-phase CIMS measurements. MCM model results predict modeled ΣRO₂/HO₂ ratios of ∼3–10 (Fig. S12b). While the experiments were carried out under higher reactant concentrations than found in air, this ratio of peroxy radicals is within the range of ∼0.3–10 reported in the atmosphere,^72–78 suggesting that both RO₂ and HO₂ will be important determinants for the fate of surface radicals on organic particles in air.

Author contributions

LMW, BJFP, and EMW wrote the manuscript. LMW and EMW processed, visualized, and carried out formal analysis of the data. YQ, EMW, LMW, MD, JNS, and BJFP conceptualized the experiments. LMW and BJFP acquired the funding. JNS and MD provided expertise in chemical ionization mass spectrometry. All authors read and approved the manuscript.

Conflicts of interest

There are no conflicts to declare.

Data availability

All data that support the findings of this study are included within the article and supplementary information (SI). Supplementary information: Schemes S1–S3, Ethylbenzene oxidation pathways; Fig. S1, MAIV spectra of benzene, toluene, and ethylbenzene oxidation products; Text S1, Response of CIMS and MAIV to self-nucleated particles; Fig. S2, Size distributions of nucleated ethylbenzene particles in the absence of glutaric acid seed particles; Fig. S3, Typical CIMS and MAIV mass spectra with gases or particles removed; Table S1, Set of oxidation products detected with MAIV-MS and CIMS; Table S2, Alternate assignments for new products; Fig. S4, Carbon oxidation state as a function of saturation concentration for products, Fig. S5, Relative intensities of highly oxidized products; Text S2, Comparison of OH loss to particles versus ethylbenzene; Fig. S6, Effect on MAIV intensities from the addition of methanol; Fig. S7, Size distributions of unreacted glutaric acid matrix particles; Fig. S8, Intensity of the base peak of GA for different particle surface areas; Text S3, Effects of product coatings on MAIV intensities; Fig. S9, Intensity of GA as a function of initial ethylbenzene concentration; Fig. S10, Intensity of the GA oxidation product as a function of initial ethylbenzene concentration; Fig. S11, Relative MAIV intensities of ROOH and ROOR products as a function of initial ethylbenzene concentration and GA particle surface area; Fig. S12, Kinetics modeling with the Master Chemical Mechanism. See DOI: https://doi.org/10.1039/d6ea00043f.

Acknowledgements

The authors gratefully acknowledge support from the National Science Foundation (Grant nos. AGS-2331523 and CHE-2505740). The authors would like to thank Jeremy Wakeen and Dr Madeline Cooke for guidance with the Nitrate Chemical Ionization Mass Spectrometer.

References

1. A. Nel, Air Pollution-Related Illness: Effects of Particles, Science, 2005, 308(5723), 804–806,  DOI:10.1126/science.1108752.
2. S. Hameed, J. Zhao and R. N. Zare, Ambient PM particles reach mouse brain, generate ultrastructural hallmarks of neuroinflammation, and stimulate amyloid deposition, tangles, and plaque formation, Talanta Open, 2020, 2, 100013,  DOI:10.1016/j.talo.2020.100013.
3. V. Stone, M. R. Miller, M. J. D. Clift, A. Elder, N. L. Mills, P. Møller, R. P. F. Schins, U. Vogel, W. G. Kreyling, K. A. Jensen, T. A. J. Kuhlbusch, P. E. Schwarze, P. Hoet, A. Pietroiusti, A. D. Vizcaya-Ruiz, A. Baeza-Squiban, J. P. Teixeira, C. L. Tran and F. R. Cassee, Nanomaterials Versus Ambient Ultrafine Particles: An Opportunity to Exchange Toxicology Knowledge, Environ. Health Perspect., 2017, 125(10), 106002,  DOI:10.1289/EHP424.
4. J. L. Mauderly and J. C. Chow, Health effects of organic aerosols, Inhalat. Toxicol., 2008, 20(3), 257–288,  DOI:10.1080/08958370701866008.
5. C. A. Pope III and D. W. Dockery, Health effects of fine particulate air pollution: Lines that connect, J. Air Waste Manage. Assoc., 2006, 56, 709–742.
6. M. Hallquist, J. C. Wenger, U. Baltensperger, Y. Rudich, D. Simpson, M. Claeys, J. Dommen, N. M. Donahue, C. George, A. H. Goldstein, J. F. Hamilton, H. Herrmann, T. Hoffmann, Y. Iinuma, M. Jang, M. E. Jenkin, J. L. Jimenez, A. Kiendler-Scharr, W. Maenhaut, G. McFiggans, T. F. Mentel, A. Monod, A. S. H. Prévôt, J. H. Seinfeld, J. D. Surratt, R. Szmigielski and J. Wildt, The formation, properties and impact of secondary organic aerosol: current and emerging issues, Atmos. Chem. Phys., 2009, 9(14), 5155–5236,  DOI:10.5194/acp-9-5155-2009.
7. M. R. Heal, P. Kumar and R. M. Harrison, Particles, air quality, policy and health, Chem. Soc. Rev., 2012, 41(19), 6606–6630,  10.1039/C2CS35076A.
8. A. Singh, W. J. Bloss and F. D. Pope, 60 years of UK visibility measurements: impact of meteorology and atmospheric pollutants on visibility, Atmos. Chem. Phys., 2017, 17(3), 2085–2101,  DOI:10.5194/acp-17-2085-2017.
9. S. Fuzzi, M. O. Andreae, B. J. Huebert, M. Kulmala, T. C. Bond, M. Boy, S. J. Doherty, A. Guenther, M. Kanakidou, K. Kawamura, V. M. Kerminen, U. Lohmann, L. M. Russell and U. Pöschl, Critical assessment of the current state of scientific knowledge, terminology, and research needs concerning the role of organic aerosols in the atmosphere, climate, and global change, Atmos. Chem. Phys., 2006, 6(7), 2017–2038,  DOI:10.5194/acp-6-2017-2006.
10. M. Kanakidou, J. H. Seinfeld, S. N. Pandis, I. Barnes, F. J. Dentener, M. C. Facchini, R. Van Dingenen, B. Ervens, A. Nenes, C. J. Nielsen, E. Swietlicki, J. P. Putaud, Y. Balkanski, S. Fuzzi, J. Horth, G. K. Moortgat, R. Winterhalter, C. E. L. Myhre, K. Tsigaridis, E. Vignati, E. G. Stephanou and J. Wilson, Organic aerosol and global climate modelling: a review, Atmos. Chem. Phys., 2005, 5(4), 1053–1123,  DOI:10.5194/acp-5-1053-2005.
11. P. J. Ziemann and R. Atkinson, Kinetics, products, and mechanisms of secondary organic aerosol formation, Chem. Soc. Rev., 2012, 41(19), 6582–6605,  10.1039/C2CS35122F.
12. J. M. Krasnomowitz, M. J. Apsokardu, C. M. Stangl, L. Tiszenkel, Q. Ouyang, S. Lee and M. V. Johnston, Growth of Aitken mode ammonium sulfate particles by α-pinene ozonolysis, Aeros. Sci. Technol., 2019, 53(4), 406–418,  DOI:10.1080/02786826.2019.1568381.
13. N. L. Ng, J. H. Kroll, A. W. H. Chan, P. S. Chhabra, R. C. Flagan and J. H. Seinfeld, Secondary organic aerosol formation from m-xylene, toluene, and benzene, Atmos. Chem. Phys., 2007, 7(14), 3909–3922,  DOI:10.5194/acp-7-3909-2007.
14. J. A. de Gouw, C. A. Brock, E. L. Atlas, T. S. Bates, F. C. Fehsenfeld, P. D. Goldan, J. S. Holloway, W. C. Kuster, B. M. Lerner, B. M. Matthew, A. M. Middlebrook, T. B. Onasch, R. E. Peltier, P. K. Quinn, C. J. Senff, A. Stohl, A. P. Sullivan, M. Trainer, C. Warneke, R. J. Weber and E. J. Williams, Sources of particulate matter in the northeastern United States in summer: 1. Direct emissions and secondary formation of organic matter in urban plumes, J. Geophys. Res. Atmos., 2008, 113(D8), D08301,  DOI:10.1029/2007JD009243.
15. Y. Fan, C. Q. Liu, L. Li, L. Ren, H. Ren, Z. Zhang, Q. Li, S. Wang, W. Hu, J. Deng, L. Wu, S. Zhong, Y. Zhao, C. M. Pavuluri, X. Li, X. Pan, Y. Sun, Z. Wang, K. Kawamura, Z. Shi and P. Fu, Large contributions of biogenic and anthropogenic sources to fine organic aerosols in Tianjin, North China, Atmos. Chem. Phys., 2020, 20(1), 117–137,  DOI:10.5194/acp-20-117-2020.
16. D. Taraborrelli, D. Cabrera-Perez, S. Bacer, S. Gromov, J. Lelieveld, R. Sander and A. Pozzer, Influence of aromatics on tropospheric gas-phase composition, Atmos. Chem. Phys., 2021, 21(4), 2615–2636,  DOI:10.5194/acp-21-2615-2021.
17. Z. Yang, L. Du, Y. Li and X. Ge, Secondary organic aerosol formation from monocyclic aromatic hydrocarbons: insights from laboratory studies, Environ. Sci. Process Impacts, 2022, 24(3), 351–379,  10.1039/D1EM00409C.
18. J. D. Crounse, L. B. Nielsen, S. Jørgensen, H. G. Kjaergaard and P. O. Wennberg, Autoxidation of Organic Compounds in the Atmosphere, J. Phys. Chem. Lett., 2013, 4(20), 3513–3520,  DOI:10.1021/jz4019207.
19. M. Ehn, J. A. Thornton, E. Kleist, M. Sipilä, H. Junninen, I. Pullinen, M. Springer, F. Rubach, R. Tillmann, B. Lee, F. Lopez-Hilfiker, S. Andres, I.-H. Acir, M. Rissanen, T. Jokinen, S. Schobesberger, J. Kangasluoma, J. Kontkanen, T. Nieminen, T. Kurtén, L. B. Nielsen, S. Jørgensen, H. G. Kjaergaard, M. Canagaratna, M. D. Maso, T. Berndt, T. Petäjä, A. Wahner, V.-M. Kerminen, M. Kulmala, D. R. Worsnop, J. Wildt and T. F. Mentel, A large source of low-volatility secondary organic aerosol, Nature, 2014, 506(7489), 476–479,  DOI:10.1038/nature13032.
20. S. Wang, R. Wu, T. Berndt, M. Ehn and L. Wang, Formation of highly oxidized radicals and multifunctional products from the atmospheric oxidation of alkylbenzenes, Environ. Sci. Technol., 2017, 51(15), 8442–8449,  DOI:10.1021/acs.est.7b02374.
21. U. Molteni, F. Bianchi, F. Klein, I. El Haddad, C. Frege, M. J. Rossi, J. Dommen and U. Baltensperger, Formation of highly oxygenated organic molecules from aromatic compounds, Atmos. Chem. Phys., 2018, 18(3), 1909–1921,  DOI:10.5194/acp-18-1909-2018.
22. Z. Dong, R. Tang, H. Liu, Q. Zhang, W. Zong, J. Cheng and X. Shi, The formation mechanism of highly oxygenated organic molecules produced by toluene in the urban atmosphere, Atmos. Environ., 2023, 295, 119555,  DOI:10.1016/j.atmosenv.2022.119555.
23. R. Atkinson, Atmospheric chemistry of VOCs and NOx, Atmos. Environ., 2000, 34(12), 2063–2101,  DOI:10.1016/S1352-2310(99)00460-4.
24. T. Jokinen, M. Sipilä, S. Richters, V.-M. Kerminen, P. Paasonen, F. Stratmann, D. Worsnop, M. Kulmala, M. Ehn, H. Herrmann and T. Berndt, Rapid Autoxidation Forms Highly Oxidized RO2 Radicals in the Atmosphere, Angew. Chem., Int. Ed., 2014, 53(52), 14596–14600,  DOI:10.1002/anie.201408566.
25. T. Berndt, W. Scholz, B. Mentler, L. Fischer, H. Herrmann, M. Kulmala and A. Hansel, Accretion Product Formation from Self- and Cross-Reactions of RO₂ Radicals in the Atmosphere, Angew. Chem., Int. Ed., 2018, 57(14), 3820–3824,  DOI:10.1002/anie.201710989.
26. F. Bianchi, T. Kurtén, M. Riva, C. Mohr, M. P. Rissanen, P. Roldin, T. Berndt, J. D. Crounse, P. O. Wennberg, T. F. Mentel, J. Wildt, H. Junninen, T. Jokinen, M. Kulmala, D. R. Worsnop, J. A. Thornton, N. Donahue, H. G. Kjaergaard and M. Ehn, Highly oxygenated organic molecules (HOM) from gas-phase autoxidation involving peroxy radicals: A key contributor to atmospheric aerosol, Chem. Rev., 2019, 119(6), 3472–3509,  DOI:10.1021/acs.chemrev.8b00395.
27. K. P. Wyche, P. S. Monks, A. M. Ellis, R. L. Cordell, A. E. Parker, C. Whyte, A. Metzger, J. Dommen, J. Duplissy, A. S. H. Prevot, U. Baltensperger, A. R. Rickard and F. Wulfert, Gas phase precursors to anthropogenic secondary organic aerosol: detailed observations of 1,3,5-trimethylbenzene photooxidation, Atmos. Chem. Phys., 2009, 9(2), 635–665,  DOI:10.5194/acp-9-635-2009.
28. R. H. Schwantes, K. A. Schilling, R. C. McVay, H. Lignell, M. M. Coggon, X. Zhang, P. O. Wennberg and J. H. Seinfeld, Formation of highly oxygenated low-volatility products from cresol oxidation, Atmos. Chem. Phys., 2017, 17(5), 3453–3474,  DOI:10.5194/acp-17-3453-2017.
29. O. Garmash, M. P. Rissanen, I. Pullinen, S. Schmitt, O. Kausiala, R. Tillmann, D. Zhao, C. Percival, T. J. Bannan, M. Priestley, Å. M. Hallquist, E. Kleist, A. Kiendler-Scharr, M. Hallquist, T. Berndt, G. McFiggans, J. Wildt, T. F. Mentel and M. Ehn, Multi-generation OH oxidation as a source for highly oxygenated organic molecules from aromatics, Atmos. Chem. Phys., 2020, 20(1), 515–537,  DOI:10.5194/acp-20-515-2020.
30. M. Wang, D. Chen, M. Xiao, Q. Ye, D. Stolzenburg, V. Hofbauer, P. Ye, A. L. Vogel, R. L. Mauldin III, A. Amorim, A. Baccarini, B. Baumgartner, S. Brilke, L. Dada, A. Dias, J. Duplissy, H. Finkenzeller, O. Garmash, X.-C. He, C. R. Hoyle, C. Kim, A. Kvashnin, K. Lehtipalo, L. Fischer, U. Molteni, T. Petäjä, V. Pospisilova, L. L. J. Quéléver, M. Rissanen, M. Simon, C. Tauber, A. Tomé, A. C. Wagner, L. Weitz, R. Volkamer, P. M. Winkler, J. Kirkby, D. R. Worsnop, M. Kulmala, U. Baltensperger, J. Dommen, I. El-Haddad and N. M. Donahue, Photo-oxidation of Aromatic Hydrocarbons Produces Low-Volatility Organic Compounds, Environ. Sci. Technol., 2020, 54(13), 7911–7921,  DOI:10.1021/acs.est.0c02100.
31. V. Lannuque, B. D'Anna, E. Kostenidou, F. Couvidat, A. Martinez-Valiente, P. Eichler, A. Wisthaler, M. Müller, B. Temime-Roussel, R. Valorso and K. Sartelet, Gas–particle partitioning of toluene oxidation products: an experimental and modeling study, Atmos. Chem. Phys., 2023, 23(24), 15537–15560,  DOI:10.5194/acp-23-15537-2023.
32. X. Cheng, Y. J. Li, Y. Zheng, K. Liao, T. K. Koenig, Y. Ge, T. Zhu, C. Ye, X. Qiu and Q. Chen, Oxygenated organic molecules produced by low-NOx photooxidation of aromatic compounds: contributions to secondary organic aerosol and steric hindrance, Atmos. Chem. Phys., 2024, 24(4), 2099–2112,  DOI:10.5194/acp-24-2099-2024.
33. S. Nakao, C. Clark, P. Tang, K. Sato and D. Cocker Iii, Secondary organic aerosol formation from phenolic compounds in the absence of NOx, Atmos. Chem. Phys., 2011, 11(20), 10649–10660,  DOI:10.5194/acp-11-10649-2011.
34. M. Riva, P. Rantala, J. E. Krechmer, O. Peräkylä, Y. Zhang, L. Heikkinen, O. Garmash, C. Yan, M. Kulmala, D. Worsnop and M. Ehn, Evaluating the performance of five different chemical ionization techniques for detecting gaseous oxygenated organic species, Atmos. Meas. Tech., 2019, 12(4), 2403–2421,  DOI:10.5194/amt-12-2403-2019.
35. A. Mehra, Y. Wang, J. E. Krechmer, A. Lambe, F. Majluf, M. A. Morris, M. Priestley, T. J. Bannan, D. J. Bryant, K. L. Pereira, J. F. Hamilton, A. R. Rickard, M. J. Newland, H. Stark, P. Croteau, J. T. Jayne, D. R. Worsnop, M. R. Canagaratna, L. Wang and H. Coe, Evaluation of the chemical composition of gas- and particle-phase products of aromatic oxidation, Atmos. Chem. Phys., 2020, 20(16), 9783–9803,  DOI:10.5194/acp-20-9783-2020.
36. L. H. Yang, M. Takeuchi, Y. Chen and N. L. Ng, Characterization of thermal decomposition of oxygenated organic compounds in FIGAERO-CIMS, Aeros. Sci. Technol., 2021, 55(12), 1321–1342,  DOI:10.1080/02786826.2021.1945529.
37. J. Zhao, V. Mickwitz, Y. Luo, E. Häkkinen, F. Graeffe, J. Zhang, H. Timonen, M. Canagaratna, J. E. Krechmer, Q. Zhang, M. Kulmala, J. Kangasluoma, D. Worsnop and M. Ehn, Characterization of the Vaporization Inlet for Aerosols (VIA) for online measurements of particulate highly oxygenated organic molecules (HOMs), Atmos. Meas. Tech., 2024, 17(5), 1527–1543,  DOI:10.5194/amt-17-1527-2024.
38. M. Song, X. Li, H. Fuchs, A. Novelli, P. T. M. Carlsson, S. R. Zorn, G. I. Gkatzelis, M. Roska, R. Tillmann, F. Rohrer, B. Bohn, R. Wu, S. Wedel, H. Wang, S. He, Y. Liu, A. Wahner and Y. Zhang, Experimental Insights on the NO-Dependent Yields of Aromatic Hydrocarbon Oxidation Products, Environ. Sci. Technol., 2025, 59(42), 22688–22700,  DOI:10.1021/acs.est.5c06022.
39. F. Graeffe, K. Kupi, H. Timonen and M. Ehn, SOA yields from C10 alkanes and oxygenates, Atmos. Chem. Phys., 2025, 25(18), 10837–10851,  DOI:10.5194/acp-25-10837-2025.
40. Y. Qin, V. Perraud, B. J. Finlayson-Pitts and L. M. Wingen, Peroxides on the Surface of Organic Aerosol Particles Using Matrix-Assisted Ionization in Vacuum (MAIV) Mass Spectrometry, Environ. Sci. Technol., 2023, 57(38), 14260–14268,  DOI:10.1021/acs.est.3c02895.
41. Y. Qin, L. M. Wingen and B. J. Finlayson-Pitts, Toward a molecular understanding of the surface composition of atmospherically relevant organic particles, Proc. Natl. Acad. Sci. U. S. A., 2022, 119(35), e2209134119,  DOI:10.1073/pnas.2209134119.
42. R. Atkinson and S. M. Aschmann, Hydroxyl radical production from the gas-phase reactions of ozone with a series of alkenes under atmospheric conditions, Environ. Sci. Technol., 1993, 27, 1357–1363.
43. R. Atkinson, S. M. Aschmann, A. M. Winer and J. N. Pitts Jr., Rate constants for the reaction of OH radicals with a series of alkanes and alkenes at 299 ± 2 K, Int. J. Chem. Kinet., 1982, 14(5), 507–516,  DOI:10.1002/kin.550140508.
44. R. Atkinson and W. P. L. Carter, Kinetics and mechanisms of the gas-phase reactions of ozone with organic compounds under atmospheric conditions, Chem. Rev., 1984, 84(5), 437–470,  DOI:10.1021/cr00063a002.
45. A. T. Lambe, J. Zhang, A. M. Sage and N. M. Donahue, Controlled OH Radical Production via Ozone-Alkene Reactions for Use in Aerosol Aging Studies, Environ. Sci. Technol., 2007, 41(7), 2357–2363,  DOI:10.1021/es061878e.
46. B. Forsyth, B. Y. H. Liu and F. J. Romay, Particle Charge Distribution Measurement for Commonly Generated Laboratory Aerosols, Aeros. Sci. Technol., 1998, 28(6), 489–501,  DOI:10.1080/02786829808965540.
47. W. C. Hinds, Aerosol Technology: Properties, Behavior, and Measurement of Airborne Particles, John Wiley & Sons, Inc., 1999, p. 483.
48. V. V. Pervukhin, D. G. Sheven’ and Y. N. Kolomiets, Collison nebulizer as a new soft ionization source for mass spectrometry, Tech. Phys., 2016, 61(8), 1262–1268,  DOI:10.1134/S1063784216080211.
49. C.-J. Tsai, J.-S. Lin, C. G. Deshpande and L.-C. Liu, Electrostatic Charge Measurement and Charge Neutralization of Fine Aerosol Particles during the Generation Process, Part. Part. Syst. Charact., 2005, 22(5), 293–298,  DOI:10.1002/ppsc.200500961.
50. S. Consta, L. M. Wingen, Y. Qin, V. Perraud and B. J. Finlayson-Pitts, Modeling the ionization mechanism of amorphous solid particles without an external energy source coupled to mass spectrometry, Phys. Chem. Chem. Phys., 2024, 26(44), 28220–28233,  10.1039/D4CP03597F.
51. D. Rondeau, DART Mass Spectrometry: Principle and Ionization Facilities, in Direct Analysis in Real Time Mass Spectrometry, 2018, pp. 43–80.
52. N. Hyttinen, O. Kupiainen-Määttä, M. P. Rissanen, M. Muuronen, M. Ehn and T. Kurtén, Modeling the Charging of Highly Oxidized Cyclohexene Ozonolysis Products Using Nitrate-Based Chemical Ionization, J. Phys. Chem. A, 2015, 119(24), 6339–6345,  DOI:10.1021/acs.jpca.5b01818.
53. X. Li, S. Chee, J. Hao, J. P. D. Abbatt, J. Jiang and J. N. Smith, Relative humidity effect on the formation of highly oxidized molecules and new particles during monoterpene oxidation, Atmos. Chem. Phys., 2019, 19(3), 1555–1570,  DOI:10.5194/acp-19-1555-2019.
54. M. E. Jenkin, S. M. Saunders, V. Wagner and M. J. Pilling, Protocol for the development of the Master Chemical Mechanism, MCM v3 (Part B): tropospheric degradation of aromatic volatile organic compounds, Atmos. Chem. Phys., 2003, 3(1), 181–193,  DOI:10.5194/acp-3-181-2003.
55. C. Bloss, V. Wagner, M. E. Jenkin, R. Volkamer, W. J. Bloss, J. D. Lee, D. E. Heard, K. Wirtz, M. Martin-Reviejo, G. Rea, J. C. Wenger and M. J. Pilling, Development of a detailed chemical mechanism (MCMv3.1) for the atmospheric oxidation of aromatic hydrocarbons, Atmos. Chem. Phys., 2005, 5(3), 641–664,  DOI:10.5194/acp-5-641-2005.
56. Master Chemical Mechanism (MCM) v3.3.1. https://mcm.york.ac.uk (date last accessed December 30, 2025).
57. S. Trimpin, Magic Ionization Mass Spectrometry, J. Am. Soc. Mass Spectrom., 2016, 27(1), 4–21,  DOI:10.1007/s13361-015-1253-4.
58. S. Trimpin and E. D. Inutan, Matrix assisted ionization in vacuum, a sensitive and widely applicable ionization method for mass spectrometry, J. Am. Soc. Mass Spectrom., 2013, 24(5), 722–732.
59. S. Trimpin, B. Wang, C. B. Lietz, D. D. Marshall, A. L. Richards and E. D. Inutan, New ionization processes and applications for use in mass spectrometry, Crit. Rev. Biochem. Mol. Biol, 2013, 48(5), 409–429.
60. S. Trimpin, I. C. Lu, S. Rauschenbach, K. Hoang, B. Wang, N. D. Chubatyi, W.-J. Zhang, E. D. Inutan, M. Pophristic, A. Sidorenko and C. N. McEwen, Spontaneous charge separation and sublimation processes are ubiquitous in nature and in ionization processes in mass spectrometry, J. Am. Soc. Mass Spectrom., 2018, 29(2), 304–315,  DOI:10.1007/s13361-017-1788-7.
61. Z. R. Schiffman, Y. Yang, B. J. Finlayson-Pitts and L. M. Wingen, Surface-Sensitive Spectrometry: New Insights into Radical Reactions at Interfaces, Environ. Sci. Process Impacts, 2026, 28, 1276–1287,  10.1039/D6EM00197A.
62. B. J. Finlayson-Pitts and J. N. Jr. Pitts, Chemistry of the Upper and Lower Atmosphere - Theory, Experiments, and Applications, Academic Press, San Diego, 2000, p. 969.
63. D. Li, D. Wang, L. Caudillo, W. Scholz, M. Wang, S. Tomaz, G. Marie, M. Surdu, E. Eccli, X. Gong, L. Gonzalez-Carracedo, M. Granzin, J. Pfeifer, B. Rörup, B. Schulze, P. Rantala, S. Perrier, A. Hansel, J. Curtius, J. Kirkby, N. M. Donahue, C. George, I. El-Haddad and M. Riva, Ammonium CI-Orbitrap: a tool for characterizing the reactivity of oxygenated organic molecules, Atmos. Meas. Tech., 2024, 17(17), 5413–5428,  DOI:10.5194/amt-17-5413-2024.
64. J. H. Kroll, N. M. Donahue, J. L. Jimenez, S. H. Kessler, M. R. Canagaratna, K. R. Wilson, K. E. Altieri, L. R. Mazzoleni, A. S. Wozniak, H. Bluhm, E. R. Mysak, J. D. Smith, C. E. Kolb and D. R. Worsnop, Carbon oxidation state as a metric for describing the chemistry of atmospheric organic aerosol, Nature Chem, 2011, 3(2), 133–139,  DOI:10.1038/nchem.948.
65. N. M. Donahue, S. A. Epstein, S. N. Pandis and A. L. Robinson, A two-dimensional volatility basis set: 1. organic-aerosol mixing thermodynamics, Atmos. Chem. Phys., 2011, 11(7), 3303–3318,  DOI:10.5194/acp-11-3303-2011.
66. Y. Li, U. Pöschl and M. Shiraiwa, Molecular corridors and parameterizations of volatility in the chemical evolution of organic aerosols, Atmos. Chem. Phys., 2016, 16(5), 3327–3344,  DOI:10.5194/acp-16-3327-2016.
67. T. J. Held and F. L. Dryer, A comprehensive mechanism for methanol oxidation, Int. J. Chem. Kinet., 1998, 30(11), 805–830,  DOI:10.1002/(SICI)1097-4601(1998)30:11<805::AID-KIN4>3.0.CO;2-Z.
68. M. Schervish, H. G. Kang, L. M. Wingen, K. Nguyen, C. Begay, J. Wilson, M. Shrivastava, Y. Chen, J. E. Shilling, N. L. Ng, U. Pöschl, T. Berkemeier, A. Zelenyuk and M. Shiraiwa, Surface crust formation controls evaporation kinetics of secondary organic aerosols, Environ. Sci. Technol., 2026 DOI:10.1021/acs.est.5c11018 , ASAP Article, .
69. A. C. Vander Wall, V. Perraud, L. M. Wingen and B. J. Finlayson-Pitts, Evidence for a kinetically controlled burying mechanism for growth of high viscosity secondary organic aerosol, Environ. Sci. Process Impacts, 2020, 22(1), 66–83,  10.1039/C9EM00379G.
70. A. C. Vander Wall, L. M. Wingen, V. Perraud, Y. Zhao and B. J. Finlayson-Pitts, Enhanced Gas Uptake during α-Pinene Ozonolysis Points to a Burying Mechanism, ACS Earth Space Chem., 2020, 4(8), 1435–1447,  DOI:10.1021/acsearthspacechem.0c00163.
71. A. C. Vander Wall, L. M. Wingen, V. Perraud, Y. Zhao and B. J. Finlayson-Pitts, Correction to Enhanced Gas Uptake during α-Pinene Ozonolysis Points to a Burying Mechanism, ACS Earth Space Chem., 2024, 8(11), 2345,  DOI:10.1021/acsearthspacechem.4c00298.
72. Y. Baker, S. Kang, H. Wang, R. R. Wu, J. Xu, A. Zanders, Q. F. He, T. Hohaus, T. Ziehm, V. Geretti, T. J. Bannan, S. P. O'Meara, A. Voliotis, M. Hallquist, G. McFiggans, S. R. Zorn, A. Wahner and T. F. Mentel, Impact of HO₂/RO₂ ratio on highly oxygenated α-pinene photooxidation products and secondary organic aerosol formation potential, Atmos. Chem. Phys., 2024, 24(8), 4789–4807,  DOI:10.5194/acp-24-4789-2024.
73. B. Bottorff, M. M. Lew, Y. Woo, P. Rickly, M. D. Rollings, B. Deming, D. C. Anderson, E. Wood, H. D. Alwe, D. B. Millet, A. Weinheimer, G. Tyndall, J. Ortega, S. Dusanter, T. Leonardis, J. Flynn, M. Erickson, S. Alvarez, J. C. Rivera-Rios, J. D. Shutter, F. Keutsch, D. Helmig, W. Wang, H. M. Allen, J. H. Slade, P. B. Shepson, S. Bertman and P. S. Stevens, OH, HO₂, and RO₂ radical chemistry in a rural forest environment: measurements, model comparisons, and evidence of a missing radical sink, Atmos. Chem. Phys., 2023, 23(18), 10287–10311,  DOI:10.5194/acp-23-10287-2023.
74. C. M. Cho, H. Fuchs, A. Hofzumahaus, F. Holland, W. J. Bloss, B. Bohn, H. P. Dorn, M. Glowania, T. Hohaus, L. Liu, P. S. Monks, D. Niether, F. Rohrer, R. Sommariva, Z. F. Tan, R. Tillmann, A. Kiendler-Scharr, A. Wahner and A. Novelli, Experimental chemical budgets of OH, HO₂, and RO₂ radicals in rural air in western Germany during the JULIAC campaign 2019, Atmos. Chem. Phys., 2023, 23(3), 2003–2033,  DOI:10.5194/acp-23-2003-2023.
75. D. Mihelcic, F. Holland, A. Hofzumahaus, L. Hoppe, S. Konrad, P. Müsgen, H. W. Pätz, H. J. Schäfer, T. Schmitz, A. Volz-Thomas, K. Bächmann, S. Schlomski, U. Platt, A. Geyer, B. Alicke and G. K. Moortgat, Peroxy radicals during BERLIOZ at Pabstthum: Measurements, radical budgets and ozone production - art. no. 8254, J Geophys Res-Atmos, 2003, 108(D4), 8254,  DOI:10.1029/2001jd001014.
76. M. Schervish and N. M. Donahue, Peroxy radical kinetics and new particle formation, Environ. Sci. Atmos., 2021, 1(2), 79–92,  10.1039/d0ea00017e.
77. P. S. Stevens, J. H. Mather, W. H. Brune, F. Eisele, D. Tanner, A. Jefferson, C. Cantrell, R. Shetter, S. Sewall, A. Fried, B. Henry, E. Williams, K. Baumann, P. Goldan and W. Kuster, HO₂/OH and RO₂/HO₂ ratios during the Tropospheric OH Photochemistry Experiment: Measurement and theory, J Geophys Res-Atmos, 1997, 102(D5), 6379–6391,  DOI:10.1029/96jd01704.
78. Z. F. Tan, F. Rohrer, K. D. Lu, X. F. Ma, B. Bohn, S. Broch, H. B. Dong, H. Fuchs, G. I. Gkatzelis, A. Hofzumahaus, F. Holland, X. Li, Y. Liu, Y. H. Liu, A. Novelli, M. Shao, H. C. Wang, Y. S. Wu, L. M. Zeng, M. Hu, A. Kiendler-Scharr, A. Wahner and Y. H. Zhang, Wintertime photochemistry in Beijing: observations of RO_x radical concentrations in the North China Plain during the BEST-ONE campaign, Atmos. Chem. Phys., 2018, 18(16), 12391–12411,  DOI:10.5194/acp-18-12391-2018.

---