A series of experiments was designed and conducted in the Manchester
Aerosol Chamber (MAC) to study the photo-oxidation of single and mixed
biogenic (isoprene and α-pinene) and anthropogenic (o-cresol) precursors in the presence of NOx and ammonium sulfate seed particles. Several online techniques (HR-ToF-AMS, semi-continuous GC-MS, NOx and O3 analyser) were coupled to the MAC to monitor the gas and particle mass concentrations. Secondary organic aerosol (SOA) particles were collected onto a quartz-fibre filter at the end of each experiment and analysed using liquid chromatography–ultrahigh-resolution mass spectrometry (LC-Orbitrap MS). The SOA particle chemical composition in single and mixed precursor systems was investigated using non-targeted accurate mass analysis of measurements in both negative and positive ionization modes, significantly reducing data complexity and analysis time, thereby providing a more
complete assessment of the chemical composition. This non-targeted analysis
is not widely used in environmental science and has never been previously used in atmospheric simulation chamber studies. Products from α-pinene were found to dominate the binary mixed α-pinene–isoprene system in terms of signal contributed and the number of particle components detected. Isoprene photo-oxidation was found to generate negligible SOA particle mass under the investigated experimental conditions, and isoprene-derived products made a negligible contribution to particle composition in the α-pinene–isoprene system. No compounds uniquely found in this system sufficiently contributed to be reliably considered a tracer compound for the mixture. Methyl-nitrocatechol isomers (C7H7NO4) and methyl-nitrophenol (C7H7NO3) from o-cresol oxidation made dominant contributions to the SOA particle composition in both the o-cresol–isoprene and o-cresol–α-pinene binary systems in negative ionization mode. In contrast, interactions in the oxidation mechanisms led to the formation of compounds uniquely found in the mixed o-cresol-containing binary systems in positive ionization mode. C9H11NO and C8H8O10 made large signal contributions in the o-cresol–isoprene binary system. The SOA molecular composition in the o-cresol–α-pinene system in positive ionization mode is mainly driven by the high-molecular-weight compounds (e.g. C20H31NO4 and C20H30O3) uniquely found in the
mixture. The SOA particle chemical composition formed in the ternary system
is more complex. The molecular composition and signal abundance are both
markedly similar to those in the single α-pinene system in positive
ionization mode, with major contributions from o-cresol products in negative ionization mode.
IntroductionOrganic aerosols and their impacts
Atmospheric aerosols affect climate directly through scattering or absorbing
solar radiation (Novakov and Penner, 1993; Andreae and Crutzen, 1997) and
indirectly by acting as cloud condensation nuclei (CCN) (Mcfiggans et al., 2006). Exposure to particulate matter has also been directly linked to adverse impacts on human health (WHO, 2016). Organic aerosol significantly contributes to fine particulate matter (PM) in the atmosphere (Fiore et al., 2012; Jimenez et al., 2009) and can affect human health through the deep penetration of small aerosol particles into the lungs through inhalation and the deposition of larger particles in the upper respiratory tract (Burnett et al., 2014). Fine PM has a wide variety of primary (e.g. agricultural operations, industrial processes, and combustion processes) and secondary sources. In addition to secondary inorganic contributions from nitrate and sulfate, secondary organic aerosol (SOA) formed from the oxidation of atmospheric volatile organic compounds (VOCs) can make a major contribution (Hallquist et al., 2009).
SOA and its formation pathways
The chemical diversity of volatile organic compounds (VOCs) and their
oxidation pathways substantially influence SOA chemical composition (Lim
and Ziemann, 2009). VOCs can be both anthropogenic and biogenic in origin
(Li et al., 2018). Common and abundant anthropogenic
VOCs include aromatic hydrocarbons such as benzene, toluene, and cresol, which are
emitted from a wide variety of human activities, e.g. cooking and
biomass burning (Atkinson and Arey, 2003), with the latter being an
oxidation product of the former two compounds
(Schwantes et al., 2017). Biogenic VOCs,
including isoprene and monoterpenes (e.g. α-pinene), are emitted in
large quantities by vegetation as well as oceanic macroalgae and microalgae
(Bravo-Linares et al., 2010; Atkinson and Arey, 2003). Once
emitted into the atmosphere, VOCs undergo oxidation by the prevailing
atmospheric oxidants: the hydroxyl radical (OH) during daytime, the nitrate
radical (NO3) at night-time, and the unsaturated fraction by ozone
during both day and night (Atkinson, 1997). The oxidation of VOCs can
result in the formation of both more and less volatile organic products
(Jimenez et al., 2009). Low-volatility organic products can condense onto
existing particles or form new particles through nucleation if sufficiently
low in volatility, as described by the gas–particle partitioning framework
(Schervish and Donahue, 2020; Donahue et al., 2011). VOC oxidation can
result in a range of multifunctional products. Multiple generations of gas-phase oxidation results in continually evolving chemical speciation either
in the gas or particulate phase (Mcneill, 2015; Shrivastava et al., 2017),
and owing to the complexity of gaseous and particulate-phase oxidation
pathways, SOA formation mechanisms remain unclear and require further
investigation.
Prior studies of using offline techniques
Whilst techniques for online or semi-continuous SOA compositional measurements have recently become more widely adopted (Zhang et al.,
2011; Ahlberg et al., 2017; Schwantes et al., 2017; Hamilton et al., 2021;
Lopez-Hilfiker et al., 2014; Decarlo et al., 2006), offline techniques
generally provide more detailed insight into molecular composition. Offline
techniques such as gas chromatography–mass spectrometry (GC-MS)
(Ono-Ogasawara et al., 2008; Saldarriaga-Noreña et al., 2018; Cropper et al., 2018) and liquid chromatography–mass spectrometry (LC-MS) (Coscollà et al., 2008; Buiarelli et al., 2017; Pereira et al., 2015) can identify the chemical composition for thousands of organic compounds, with some of the techniques revealing information about a compound's structure, alluding to potential sources and formation mechanisms (Liu et al., 2007; Singh et al., 2011; Ono-Ogasawara et al., 2008; Carlton et al., 2009; Kroll et al., 2005a; Ng et al., 2008; Nestorowicz et al., 2018; Eddingsaas et al., 2012). LC-MS has been widely employed for the chemical characterization of laboratory-generated SOA and ambient SOA. For example, targeted analysis of SOA products using high-performance liquid chromatography time-of-flight mass spectrometry (HPLC-ToF-MS) illustrated a new pathway for the formation of 3-methyl-1,2,3-butane-tricarboxylic acid (MBTCA) through the further oxidation of nopinone, a known product in the oxidation of β-pinene, by OH Mutzel et al. (2016). Hamilton et al. (2021) used targeted LC-Orbitrap MS analysis of ambient Beijing filter samples to identify tracers of isoprene nitrate formation pathways in both gas and particle phases, indicating a strong dependence on nitrate radicals from early afternoon onwards. These targeted approaches are somewhat limited by their inability to comprehensively account for the entire mass of SOA components, though it is impractical to extract the non-targeted chemical information by manual data processing in complex ambient systems. Non-targeted screening tools have been widely employed in metabolite and protein analysis to reduce data analysis time but are uncommon in environmental science applications. Non-targeted analysis extracts the chemical information of all detected compounds in a sample dataset, providing tentative identification of unknown compounds via library screening, while allowing the rapid chemical characterization of complex mixtures through the chemical classification of detected compounds in a given sample (Place et al., 2021; Pereira et al., 2021). Mezcua et al. (2011) reported that 210 pesticides were successfully detected and identified in 78 positive samples of fruit and vegetable by using an automatic non-targeted screening method in LC-ToF analysis. Non-targeted screening analysis based on high-resolution accurate mass spectrometry (HRAM-MS) was applied in chemical characterized of tobacco smoke and successfully identified a total of known 331 compounds and 50 novel compounds as being present in the sample (Arndt et al., 2019). Chromatographic separation coupled with Fourier transform mass spectrometers (e.g. Orbitrap) have sufficient mass resolution to characterize the chemical composition of
complex particulate matter with the ability to distinguish structural isomers. Exploiting this capability, a methodology for automated non-targeted screening was presented by Pereira et al. (2021) using ultrahigh-performance liquid chromatography–Orbitrap MS data. This non-targeted screening tool has been rigorously tested using authentic standards and provides molecular formula assignments and plausible structure information (among other information) for all detected compounds within a sample dataset. Moreover, the accurate mass spectrometry employed has a mass resolution of 70 000 at m/z 200, leading to a substantial increase in the signal-to-noise ratio and enhanced quantification of low-concentration species. However, non-targeted screening methods are not infallible and rigorous testing of autonomous platforms must be performed to understand potential limitations of these tools. Moreover, it is challenging to make semiquantitative or quantitative measurements of unknown compounds in complex matrices. It is worth noting that quantitative measurements of unknown compounds are general limitations of ESI operation not directly attributed to the non-targeted screening method, but they arguably become more important. It is difficult to perform quantitative measurement of unknown compounds due to the analytical standards for SOA products being limited, and only a few molecules out of the thousands of detected compounds might be known. Therefore, it is also challenge to determine sample extraction recoveries during sample extraction procedures. The approach of using the normalized abundance of compounds in the sample does not consider different compound electrospray ionization (ESI) efficiencies, which can be influenced by the molecular structure among other parameters (Priego-Capote and Luque de Castro, 2004). For example, Cech and Enke (2000) found that ESI response increased for peptides with a more extensive non-polar region. Cech and Enke (2001) examined this further and concluded that analytes with a higher polar portion have a lower ESI response than the more non-polar analytes. Differences in ESI efficiencies of individual compounds may impact the normalized abundance of chemical groupings, particularly when comparing sample compositions which differ appreciably.
In one of the few studies applying an automated non-targeted method in
environmental matrices, Mehra et al. (2021) used this approach for
LC-Orbitrap MS data to characterize the SOA from the low-NOx oxidation
of 1-methylnaphthalene, propylbenzene, and 1,3,5-trimethylbenzene in
laboratory measurements, alongside characterizing the SOA from filters
collected in an urban area. The aim is to study the anthropogenic and
biogenic contributions to organic aerosol. This study also compared the
result with an online technique using a time-of-fight chemical ionization mass
spectrometer with an iodide ionization system (I-CIMS), which showed good
agreement between observations for online I-CIMS results and results of
offline LC-Orbitrap MS in negative ionization mode.
Wang et al. (2021) also used a non-targeted method
for LC-Orbitrap MS data to characterize particulate products on filters
collected from three cities located in northeastern, eastern, and southeastern
China, namely Changchun, Shanghai, and Guangzhou. This study suggested that
anthropogenic emissions are the dominate source of urban organic aerosol in
all three cities. Also, they found that samples from Shanghai and
Guangzhou shared considerable chemical similarity but significantly differed
from Changchun. In our present study, for the first time, we will apply this
automated non-targeted screening tool for the compositional analysis of SOA
generated in an aerosol chamber from single and mixed precursor experiments.
Summary of studies on similar SOA systems
There are numerous studies investigating SOA formation from the oxidation of
biogenic VOCs, particularly for terpenoid compounds (Stroud et al., 2001;
Surratt et al., 2006; Dommen et al., 2006; Carlton et al., 2009; Camredon et
al., 2010; Surratt et al., 2010; Henry et al., 2012; Ahlberg et al., 2017;
Hoffmann et al., 1997; Odum et al., 1996). Isoprene (C5H8) is the
most abundant biogenic VOC emission, and α-pinene (C10H16)
is one of the most abundant and widely studied biogenic monoterpenes
(Hallquist et al., 2009). Whilst oxidation products from these two biogenic precursors are both considered to substantially contribute to the global SOA budget, there are marked differences in their SOA particle mass yield; α-pinene has a yield in the range of 17 % to 45 % (Mcvay et al., 2016; Ng et al., 2007; Eddingsaas et al., 2012), while isoprene has
a much lower yield in the range of 0 % to 5 % (Dommen et al., 2006; Kroll et al., 2005a, 2006; Pandis et al., 1991; Carlton et al., 2009). The reason for the low isoprene SOA yield is in part a result of the high volatility of oxidation products. However, the yield of isoprene SOA is strongly acid-dependent and closely related to the particle-phase acidity
due to the impact on the amount of heterogenous uptake, which is the reason
for higher isoprene SOA mass concentration when increasing aerosol acidity (Surratt et al., 2007a). Xu et al. (2021) demonstrated that over 98 % of isoprene-oxidized organic molecules by mole were classified as semi-VOCs (SVOCs) and intermediate VOCs (IVOCs) with a volatility (log10C*, µ m-3) range of -0.5 to 5, while about 1.3 % of isoprene oxidation products were considered low VOCs (LVOCs). Conversely, the larger C10 monoterpene skeleton of α-pinene typically results in the formation of less volatile oxidation products. Lee et al. (2011) reported that the SOA from α-pinene ozonolysis required 80 ∘C for complete volatilization, and the volatility of α-pinene SOA strongly depended on the VOC / NOx ratios, forming volatile nitrate-containing species under high-NOx conditions.
There are many studies reporting the chemical characterization of SOA formed
in smog chambers from α-pinene and isoprene using liquid
chromatography–mass spectrometry (LC-MS)(Yasmeen et al., 2012; Surratt
et al., 2006; Kahnt et al., 2014; Pereira et al., 2014; Winterhalter et al.,
2003). Winterhalter et al. (2003) used LC-MS to demonstrate the major
particulate-phase compounds from the O3 and OH oxidation of α-pinene, such as cis-pinic acid, cis-pinonic acid, hydroxy-pinonic acid
isomers, and possibly hydroxy-carboxylic acid. It is worth noting that this
study suggested that ozonolysis reaction is the main pathway of
aerosol formation regarding its performance in various experiments.
Similarly, Surratt et al. (2006) studied isoprene
photo-oxidation under various NOx conditions. The chemical composition
of isoprene SOA products was analysed by a series of online and offline
techniques (including LC-MS) and indicated that oligomerization plays an
important role in SOA formation pathways, especially under high-NOx
conditions, forming acidic products.
SOA can also be produced from anthropogenic VOCs (e.g. o-cresol), although
global biogenic SOA production (∼ 88 TgC yr-1) is thought
to dominate over the anthropogenic SOA production (∼ 10 TgC yr-1) (Hallquist et al., 2009). Schwantes
et al. (2017) studied the formation of low-volatility products from
o-cresol photo-oxidation under various NOx conditions using chamber
experiments with chemical ionization mass spectrometry (CIMS) and direct
analysis with real-time mass spectrometry (DART-MS). This study identified
several o-cresol oxidation products, including a first-generation product
(methyl-catechol), second-generation products (trihydroxy-toluene and
hydroxy-methyl-benzoquinone), and third-generation products
(tetrahydroxy-toluene and dihydroxy-methyl-benzoquinone), indicating
successive addition of OH radicals onto the aromatic ring during the
oxidation, following expected mechanistic pathways (Atkinson and
Aschmann, 1994; Olariu et al., 2002)
Despite the wealth of knowledge of gaseous and particulate-phase product
formation from the oxidation of single VOC precursors using chamber experiments, there is a comparative lack of understanding in the real
atmosphere. Online measurements of the OA composition by an Aerodyne high-resolution aerosol mass spectrometer (HR-ToF-AMS) and VOCs by an Ionicon proton transfer reaction mass spectrometer (PTR-MS) during the CARES campaign in the vicinity of Sacramento, California, indicated that the mixing of
anthropogenic emissions from Sacramento with isoprene-rich air from the
foothills enhances the production of OA (Shilling et al., 2013). This study
suggested that anthropogenic–biogenic interactions enhance OA production from biogenic species, suggesting that the amount of isoprene SOA strongly depends on the VOC / NOx ratio. However, the physical and chemical reasons for such interactions remain unclear and warrant further investigation. There have been several laboratory studies investigating SOA formation in mixed VOC systems. Ahlberg et al. (2017) investigated SOA from VOC mixtures including biogenic (α-pinene, myrcene, and isoprene) and
anthropogenic VOCs (m-xylene) in an oxidation flow reactor (OFR) equipped
with high-resolution time-of-flight aerosol mass spectrometry (HR-ToF-AMS).
Their results showed that the SOA mass yield formed from a VOC mixture
containing myrcene was higher than expected, possibly a result of myrcene-nucleating particles leading to an increased condensation sink under the conditions of the OFR. This study also found that the SOA particle size was larger in VOC mixtures with isoprene and unlimited oxidant supply. However, other studies indicate that isoprene could inhibit new particle formation by scavenging oxidants and forming relatively high-volatility organic products rather than nucleating precursors (Kiendler-Scharr et al., 2009, 2012). McFiggans et al. (2019) reported a reduction in SOA mass and yield from the VOC mixture of α-pinene and isoprene with an increasing fraction of isoprene in the mixture. This was attributed to
isoprene acting as an OH scavenger and its radical oxidation products reacting with those formed from α-pinene, enhancing the overall volatility of the products in the mixture. This study indicates that
interactions between VOC products should be considered to enable a mechanistic understanding of SOA formation in the ambient atmosphere. Shilling et al. (2019) reported that freshly formed isoprene SOA did not fully mix with pre-existing SOA in an isoprene–α-pinene mixture system (e.g. aged isoprene SOA and aged α-pinene SOA) over the 4 h experimental timescale in a sequential condensation experiment without observing notable suppression of SOA formation in the α-pinene–isoprene mixture system.
The present study
In this study, we designed a series of chamber experiments using single,
binary, and ternary VOC systems, expanding on the work performed by McFiggans
et al. (2019), with the aim of better understanding the chemical composition and interactions during SOA formation in mixed VOC systems. We move beyond the consideration of SOA formation from anthropogenic VOC precursors to consider the effect of their mixture with biogenic VOC. Ortho-cresol (o-cresol) was chosen as an anthropogenic precursor with a moderate SOA yield between that of isoprene and α-pinene. o-Cresol has reactivity toward the hydroxyl radical (OH) that is comparable to those of the chosen biogenic VOCs (Atkinson et al., 2004) and a negligible reactivity towards ozone. Hence, the oxidation products from each precursor are likely to be of comparable abundance in a mixed systems. We retained the two biogenic precursors studied in McFiggans et al. (2019), with isoprene being the dominant VOC emitted from plants globally, but with modest SOA formation potential and alpha-pinene (α-pinene), similarly widely emitted in lower amounts, but a more efficient SOA precursor.
The objectives of the present study are to investigate, using offline analysis
of SOA chemical composition, whether (i) high-yield precursors dominate the
contribution to SOA formation of mixture systems and (ii) whether cross-products
from mechanistic interactions in the oxidation of precursors feature
strongly in the mixed precursor systems. A series of photochemical oxidation
experiments was designed and conducted to produce SOA from the selected
VOCs (α-pinene, isoprene, and o-cresol) and their mixtures in the
presence of neutral seed particles (ammonium sulfate) and NOx. The
experimental programme included three single precursor systems, three binary
precursor mixtures, and one ternary mixture of precursors. The aerosol
samples were collected onto a filter from each experiment and analysed
offline using liquid chromatography–ultrahigh-resolution mass spectrometry
with an automated non-targeted data processing methodology recently
described in Pereira et al. (2021).
MethodChamber description
All experiments were performed in the 18 m3 Manchester Aerosol Chamber
(MAC). Briefly, the MAC operate as a batch reactor to study the atmospheric
processing of multicomponent aerosols under controlled conditions. The
chamber comprises an FEP Teflon bag mounted on three rectangular extruded
aluminium frames housed in an air-conditioned enclosure. Two 6 kW Xenon arc
lamps (XBO 6000 W/HSLA OFR, Osram) and a bank of halogen lamps (Solux 50
W/4700 K, Solux MR16, USA) are mounted in the inner aluminium wall of the
enclosure, which is lined with reflective “space blanket” material to
provide maximum and homogenous light intensity to simulate the realistic
daytime atmospheric environment. To remove unwanted radiation flux below
300 nm, a quartz filter was mounted in front of each arc lamp. Removal of
unwanted heat from the lamps as well as temperature and relative humidity control
of the chamber were assisted by conditioned air introduced between the bag
and the enclosure at 3 m3 s-1 as well as active water cooling of the
mounting bars of the halogen lamps and of the filter in front of the arc
lamps. Regular steady-state actinometry experiments were conducted through
the entire campaign and indicated a photolysis rate of NO2 (JNO2) in the range of 1.83–3 × 10-3 s-1 during the
experimental period. Photolysis of NO2 leads to O3 formation,
which further photolyses to produce OH radicals in our moist experiments.
Humidity and temperature are controlled by the humidifier and by controlling
the air-conditioning set point during the experiment. It is continuously
monitored using a dew-point hygrometer and a series of thermocouples and
resistance probes throughout the chamber. Additional online instruments
included a semi-continuous gas chromatography mass spectrometer (GCMS) for
VOC measurement (Minaeian, 2017), a water-based condensation particle
counter, a differential mobility particle sizer (DMPS), and an aerosol mass
spectrometer (AMS) for particulate-phase compound measurement
(Canagaratna et al., 2007). The
filter collection, extraction, measurement, and analysis techniques are
described below. Full details of the MAC characterization, experimental
procedure, and instrumentation payload are provided in
Shao et al. (2022).
Experimental strategy
The experimental programme was conceived using a concept of “initial
iso-reactivity” towards OH, with the intention of allowing a reasonably
comparable contribution of oxidation products from each VOC at the chosen
concentration and experimental conditions. Clearly this does not take into
account consumption by oxidants other than OH formed during the experiment
(notably ozone) and also neglects the reactivity of the subsequent oxidation
products. The injected precursor mass was therefore chosen according to its
reactivity towards OH (Atkinson et al., 2004). SOA composition was determined using analysis of chamber filter samples by liquid chromatography–ultrahigh-resolution mass spectrometry (LC-Orbitrap MS) and automated non-targeted data processing for all single precursor and mixed VOC systems.
Experimental procedure
Programmed “pre-experiment” and “post-experiment” procedures were
routinely conducted before and after each SOA experiment to minimize the
possible contamination in the chamber. The pre-experiment and
post-experiment are comprised of multiple automated fill–flush cycles
with an approximate airflow rate of 3 m3 min-1 for cleaning the
chamber. The upper and lower frames were free to move vertically to expand
and collapse the bag during the fill–flush cycle. Filtered air was
sequentially injected into and extracted from the bag, reducing contaminants
in the bag with each cycle. Several instruments (e.g. WCPC, model 49C
O3 analyser, Thermo Electron Corporation; model 42i
NO–NO2–NOx analyser, Thermo Scientific) were continuously
connected to the chamber during the pre-experiment to monitor the
concentration of particles and the concentration of ozone and NOx, as well as to
ensure the bag was sufficiently clean (with all aforementioned factors close
to zero) to conduct the chamber background procedure. When conducting this
procedure, there were no reactants in the bag and the bag was stabilized for
at least an hour for the instruments to establish the baseline of the clean
chamber. In the following stage, VOC precursor(s), NOx, and seed
particles were injected into the chamber sequentially in dark conditions and
the chamber remained steady for an hour for the instruments to obtain a
baseline of the initial chamber conditions (e.g. experimental background)
before the SOA experiment. The baselines of the chamber background and
experimental background were subsequently subtracted from the experimental
measurements.
Ammonium sulfate seed particles were generated via atomization from
ammonium sulfate solution (Puratonic, 99.999 % purity) using a Topaz model ATM 230 aerosol generator. The concentration of seed particles in the
chamber was controlled by altering the injection time and concentration of
the prepared solution (0.01 g mL-1). The accumulating seed particles are injected into the stainless-steel residence chamber for 1 min, then diverted into the main chamber injection flow for 30 s during the final fill cycle of the pre-experiment procedure. The liquid α-pinene, isoprene, and o-cresol (Sigma Aldrich, GC grade ≥ 99.99 % purity) were injected as required through the septum of a heated glass bulb and evaporated into an N2 carrier flow into the chamber during this final fill along with NOx as NO2 from a cylinder, also carried by N2. The injected VOC mass was calculated using the “initial OH iso-reactivity” approach described above. Photochemistry was initiated by irradiating the VOC at a moderate VOC / NOx ratio using the lamps as described above. The temperature and relative humidity conditions were controlled at 50 % ± 5 % and 24 ± 2∘, respectively, during the experiment. The concentration of NOx and O3, the particle number concentration, and particle mass concentration were monitored during the experiment using the online instruments. SOA particles were collected on a blank filter (Whatman quartz microfiber, 47 mm) mounted in a bespoke holder built into the flush pipework by flushing the remaining chamber contents after a 6 h experiment. The filters were then wrapped in foil and stored at -18 ∘C prior to analysis. Quartz-fibre filters were pre-conditioned by heating in a furnace at 550 ∘C for 5.5 h. It is noted that both positive (conversion of gas-phase organics to particulate form) and negative (volatilization of particulate organic compounds) artefacts are possible during collection of particulate matter during filter sampling, resulting in overestimation and underestimation of particulate organic carbon, respectively. The samples were rapidly collected in our experiments (emptying the chamber through the filter in 5 or 6 min), precluding the ability to effectively denude gases at the flow rate. Whilst gases may be adsorbed or adsorbed on the filters, it is challenging to quantify these impacts. Formation of products of reactions in the particles themselves could also occur after gas–particle collisions during the experiment with a much longer residence time in the chamber.
Du et al. (2021) combined online (FIGAERO-CIMS) and offline mass spectrometric (LC-Orbitrap MS) techniques to characterize the chemical composition in the same systems. It was reported that the distribution of particle-phase products is highly consistent between the I–CIMS and LC-Orbitrap MS negative ionization mode for the α-pinene SOA products, suggesting nearly negligible (or at least comparable) gas-phase absorption artefacts introduced during filter collection in both techniques.
Actinometry and off-gassing experiments were conducted regularly after
several of SOA experiments to establish the consistency of the chamber's
performance, evaluate the effectiveness of the cleaning procedure, and
confirm cleanliness of the chamber. “Background” filters were collected
from the actinometry and off-gassing experiments. A summary of experimental
conditions is given in Table 1.
Experimental descriptions, VOC mixing ratios, VOC : NOx ratio, and
mass concentration of seed particles in the chamber.
Filter samples of SOA particles were extracted using the following procedure. Each filter was cut into small pieces into a pre-cleaned 20 mL scintillation vial. A total of 4 mL (Fisher Scientific FB15051) of methanol (Optima LC-MS grade, Thermo Fisher Scientific) was added to the vial. The sample was then wrapped in foil and left for 2 h at ambient temperature. It was then sonicated for 30 min and the extractant filtered through a 0.22 µm pore size polyvinylidene difluoride (PVDF) filter using a BD PlasticPak syringe. An additional 1 mL of methanol was added to the vial and filtered through the same syringe membrane to minimize sample loss. The filtered extractant was then evaporated to dryness using solvent evaporator (Biotage, model V10) at 36 ∘C and 8 mbar pressure and redissolved in 1 mL of 90 : 10 water–methanol (Optima LC-MS grade) for LC-Orbitrap MS analysis. The efficiency of the aerosol extraction procedure using non-targeted analysis in this study is difficult to determine owing to the limitation of unknown compound identification. Few molecules of the thousands detected can be identified in the analytical standards for SOA products. It is also difficult to determine sample extraction recoveries since compounds have different recovery efficiencies determined by their molecular structure
(Priego-Capote and Luque de Castro, 2004). Much further work on the recovery efficiency is required to quantify potential losses and provide insights into the quality of the extraction procedure.
Liquid chromatography–mass spectrometry analysis
Samples were analysed using ultra-performance liquid chromatography–ultrahigh-resolution mass spectrometry (Dionex 3000, Orbitrap QExactive, Thermo Fisher Scientific). A reverse-phase C18 column (aQ Accucore, Thermo Fisher Scientific) of 100 mm (long) × 2.1 mm (wide) with a 2.6 µm particle size was used for compound separation. The flow rate was set to 0.3 mL min-1, with 2 µL sample injection volume. The autosampler temperature was set to 4 ∘C and the column at 40 ∘C. The mobile-phase solvent included (a) water and (b) methanol that both contain 0.1 % (v/v) formic acid (Sigma Aldrich, 99 % purity). Gradient elution was performed starting at 90 % (a) with a 1 min post-injection hold, decreasing to 10 % (a) over 26 min, before returning to the initial mobile-phase conditions at 28 min, followed by a 2 min column re-equilibration. Electrospray ionization (ESI) was used with a mass-to-charge (m/z) scan range of 85 to 750. The ESI parameters were set as follows: 320 ∘C for capillary and auxiliary gas temperature as well as 70 (arbitrary units) and 3 (arbitrary units) flow rate for sheath gas and auxiliary gas, respectively (Pereira et al., 2021). Compound fragmentation was achieved using higher-energy collision-induced dissociation (MS2). A fragmentation spectrum is generated for each selected precursor, which allows structural identification through the elucidation of fragmentation patterns (Mcluckey and Wells, 2001). These fragmentation spectra can aid in the structural identification of isomeric species (i.e. compounds with the same molecular formula but different structural arrangement). Accurate mass calibration was performed prior to sample analysis in positive and negative ESI mode using the manufacturer recommended calibrants (Thermo Scientific). A procedural
control (i.e. pre-conditioned blank filter subject to the same sample
extraction procedure) was analysed, along with solvent blanks (consisting of
90 : 10 water–methanol), which were frequently run throughout the sample
analysis sequence, allowing any instrument or extraction artefacts to be
detected. Automated non-targeted data analysis was performed using Compound
Discoverer version 2.1 (Thermo Fisher Scientific). Full details of the data
processing methodology can be found in Pereira et al. (2021). Briefly, the chemical information on all detected compounds in each sample data file is extracted. The method provides molecular formulae assignment of detected compounds using the following elemental restrictions: unlimited carbon, hydrogen, and oxygen atoms, up to 5 nitrogen and sulfur atoms, and in positive ionization mode 2 sodium and 1 potassium atom are also allowed (sodium and potassium are typically introduced into the samples via
glassware). Molecular formulae were attributed if the mass error < 3 ppm, signal-to-noise ratio > 3, and the isotopic intensity tolerance was within ±30 % of the measured and theoretical isotopic abundance. Instrument artefacts and compounds detected in the background filter were removed from sample data if the same detected
molecular species had a retention time within 0.1 min and sample / artefact or background peak area ratio < 3. Any compounds detected in the sample and background data with a sample / background peak area ratio > 3 were conserved in the sample dataset after subtracting the background peak area (new peak area = sample peak area – background peak area). The automated Python programme generates a list of detected compounds, assigned molecular formulae, and tentatively assigned mass spectral library identifications (see Pereira et al., 2021, for further information). The mass spectra for both ESI modes from each VOC system are shown in the Supplement (Figs. S1 and S2).
To provide confidence in the components in each system detected by the
non-targeted method, only those compounds found in all three replicate
experiments (two in the single precursor isoprene and binary
o-cresol–isoprene systems) and not found in any background “clean”
experiments were attributed to a particular single precursor or mixed
system. The approach taken thus ensures the most conservative assignment of
compounds to a particular precursor system. Where quantities are analysed
and presented from “representative” experiments, only those relating to
compounds found in all replicate experiments are confidently attributed to
this particular system. Compounds that were found above the detection limit in
only a subset of the experiments in a single system were not attributed to
the system and were considered “inconclusive”. Moreover, the common
compounds were only considered to be the same detected molecular species if
they had a retention time within 0.1 min and sample / artefact peak area
ratio > 3 in all replicate experiments. Section 3.2.1 and
3.2.3 only consider the compounds which can be confidently attributed to a
particular system. For the elemental characterization in Sect. 3.2.2, both
the confident and inconclusive components are presented, with only the
compounds confidently attributed analysed according to carbon number.
Results and discussionSOA particle mass formation in the experiments
The formation of SOA particle mass in the seven experimental systems is
shown alongside the VOC concentration as well as NOx and O3 mixing ratio
time series in Fig. 1. As shown in Fig. 1a, the particle wall-loss-corrected SOA mass in all α-pinene-containing systems reaches a
maximum value within the 6 h experimental timeframe. α-Pinene
produced the highest SOA particle mass (∼ 400 µg m-3)
of all systems at nominal “full” VOC reactivity with the most rapid onset
and rate of mass formation. The SOA particle mass continued to increase at
the end of the experiment in the single VOC o-cresol and binary
isoprene–o-cresol systems. No measurable SOA particle mass above background
(∼ 0 µg m-3) was produced within the 6 h duration
in any single VOC precursor isoprene experiment.
As shown in Fig. 1b, NOx was observed to decay in all systems wherein
significant SOA mass was formed, but little NOx consumption was
observed in the single isoprene system or in the binary isoprene–o-cresol
mixture. The reduction of NOx will result from (i) reaction between OH
radicals and NO2, leading to HNO3 formation with subsequent loss
to the chamber walls or particles as inorganic nitrates. It will also result from (ii) termination
reactions between NO and RO2 radicals or NO2 and RO2 radicals,
leading to formation of nitrogen-containing organic (NOROO2 and
NO2ROO2) compounds (Atkinson, 2000).
Noting that there was no O3 initially in any experiment, Fig. 1c
illustrates ozone concentration time series in each system. Ozone can be
seen to increase during the initial stage in most experiments,
with most modest rises in the single o-cresol and binary isoprene–o-cresol
systems. An initial rise is expected owing to the fairly rapid photolysis of
NO2 tending towards a photo-stationary state (PSS) between NO2, NO,
and O3. The onset of VOC oxidation will result in the consumption of
O3 when unsaturated α-pinene and isoprene are present. At
the same time, NO will react with RO2 and HO2 radicals formed in
the VOC degradation, resulting in NO2 and OH radical formation. The
reduction in the proportion of NO reacting with O3 and photolysis of
the NO2 produced results in net O3 production and deviation from
PSS.
The time profiles of the VOC concentration from experiments in all single and
mixed precursor systems are shown in Fig. 1e–f. A rapid and pronounced
onset of VOC consumption in each system is observed after illumination, which is
attributable to reaction with OH radicals and O3 in α-pinene-
and isoprene-containing systems. Panels (d) to (f), which are plotted logarithmically
for clarity, show the VOC decay profile in each experiment, reflecting their
differences in reactivity and the variable oxidant regime in each
experiment. Individual VOCs have comparable decay rates in each mixture
except for (i) α-pinene in the binary α-pinene–o-cresol
system, which had a significantly lower decay rate than it had in other
α-pinene-containing systems, and (ii) isoprene, which had a
faster decay rate in the binary α-pinene–isoprene system than in
other isoprene-containing systems. No VOC was entirely consumed in any
system by the end of the 6 h experiments, with consumption continuing
until the end.
Evolution of gas and total SOA particle mass measurements during
the photo-oxidation of VOCs after chamber illumination. (a) The SOA mass was
measured using a high-resolution time-of-flight aerosol mass spectrometer
(HR-ToF-AMS) during single, binary, and ternary experiments. (b–c) Concentration of NOx and O3 against time in all of the single, binary, and
ternary experiments. (d–f) Decay rate of VOCs across all systems for
α-pinene (b), isoprene (c), and o-cresol (d) in single, binary, and
ternary experiments, respectively.
Characterization of components by LC-Orbitrap MSCharacterization by number of discrete compounds in each system
The number of discrete peaks extracted using the Compound Discoverer
software from the LC-Orbitrap MS data for all experiments in each SOA system
is listed in Table 2 and illustrated using Venn diagrams showing the
compounds found in more than one system (henceforth referred to as
“common” compounds) and those found solely in a single system (referred to
as “unique”) in Figs. 2 and 3 (in negative and positive ionization modes,
respectively).
As seen in Table 2, all α-pinene-containing systems were found to
contain a greater number of compounds than any system not containing α-pinene. The binary α-pinene–isoprene system contained the
highest number of all systems, with 377 in negative ionization mode and 441
in positive ionization mode. A total of 644 total compounds were seen in the
single VOC α-pinene system across both negative and positive
ionization modes, which is fewer than in the binary α-pinene–isoprene system with 818 compounds but higher than the α-pinene–o-cresol system with 483 compounds. The total number of discrete products in the ternary system is lower than in the single α-pinene and binary α-pinene system. The single VOC isoprene system generated the lowest total number of products of all systems above the detection limit. This is unsurprising, since undetectable mass concentration was found by the online instrumentation in these experiments. Multifunctional compounds can be detected in both negative and positive ionization mode. Negative ionization mode typically exhibits high sensitivity to compounds containing alcohol and
carboxylic acid functionalities, whereas positive ionization mode typically
has a greater affinity for compounds with functional groups that are readily
protonated (e.g. -NH, -O-, or -S-, -CH2-,-C=O,-SO2 group) (Glasius
et al., 1999; Steckel and Schlosser, 2019).
Number of compounds detected in an SOA sample in negative and positive
ionization mode from single, binary, and ternary precursor systems.
Number of detected compounds ExperimentNegativePositivemodemodeα-Pinene282362Isoprene2868o-Cresol8453α-Pinene–isoprene377441α-Pinene–o-cresol339144o-Cresol–isoprene7287α-Pinene–isoprene–o-cresol112188(a) Negative ionization mode
Figure 2 shows a Venn diagram of the number of discrete compounds identified
in negative ionization mode in each of the individual and binary precursor
experiments. Figure 2a and b show that the number of discrete
compounds from α-pinene dominated those found in the binary mixture
system compared to those from the other precursors. 182 compounds found in
all α-pinene single precursor experiments were also found in the
binary α-pinene–isoprene mixed system, which is approximately 45 times
greater than the four compounds also found in all single isoprene experiments.
Similarly, 99 common compounds were found between the single precursor
α-pinene experiments and those found in the binary α-pinene–o-cresol system, which is roughly 3 times higher than the number of
o-cresol-derived products that were also found in binary mixed system. More
than half of the total number of compounds in the α-pinene–isoprene and α-pinene–o-cresol binary systems were unique to the
mixtures and not observed in any of single precursor experiments. In the
isoprene–o-cresol system a lower total number of compounds were detected in
every repeat experiment, with more compounds in the mixture also found in
the o-cresol system than the isoprene system (Fig. 2c).
Number of common discrete compounds and unique compounds in single
and binary precursor mixed experiments detected by negative-ionization-mode
LC-Orbitrap MS. Products are considered identical in the mixed and single
precursor systems if the compound has the same empirical formula and a
retention time difference < 0.1 min.
(b) Positive ionization mode
Figure 3 shows the number of discrete SOA compounds identified in positive
ionization mode in the single and binary systems. There are 226 compounds
found in all α-pinene single precursor experiments that were also
found in the binary α-pinene–isoprene system, which is about 32 times more
than also found in the isoprene-only experiments (Fig. 3a). A total of 48 α-pinene-derived compounds were also found in the binary α-pinene–o-cresol
system, which is 16 times greater than those also found in all o-cresol-only
experiments (Fig. 3b). In both α-pinene-containing binary mixtures,
around or more than half of all detected compounds were unique to the
mixture. In the binary isoprene–o-cresol system shown in Fig. 3c,
o-cresol-derived compounds were more numerous than those in the isoprene
experiments, with 23 compounds observed.
Number of common discrete compounds and unique compounds in single
and binary precursor mixed experiments detected by positive-ionization-mode
LC-Orbitrap MS. Products are considered identical in mixed and single
precursor systems if a compound has the same empirical formula and the
retention time difference < 0.1 min.
The Venn diagrams for both ionization modes indicate the importance of
α-pinene oxidation products in both binary systems, with a large
number of binary SOA compounds found to be present in the single precursor
α-pinene system. In contrast, there are few common compounds
observed between single isoprene and binary systems, possibly a result of
the majority of isoprene-derived products remaining in the gas phase or the
isoprene products participating in cross-product formation in mixed
precursor systems.
Characterization of organic particulates by elemental groupsNegative ionization mode
Elemental groupings are used here to provide insights into the SOA chemical
composition in each system. All detected molecular formulae in each system
were classified into four categories based on their elemental compositions, which are CHO, CHON, CHOS, and CHONS (C, H, O, N, and S corresponding to
the atoms in the molecule), and separated into seven carbon number categories. The measured peak area of each compound was normalized to the
total sample peak area as shown in Fig. 4 and described in Pereira et al. (2021). Figure 4 presents the signal fraction of compounds in representative experiments that can be confidently attributed as found in each of the systems (i.e. that are found in every repeat experiment in this system) in the coloured stacked bars according to their carbon number and classified according to their elemental groupings. The fractional contributions of compounds that are confidently stated are similar for each experiment in a particular system. Also shown in the grey bar is the signal fraction of compounds that are inconclusively found in the experiment in each system classified by elemental grouping, but not found in all repeat experiments or the chamber background experiment.
The normalized signal intensity distribution of different compound
categories (CHO, CHON, CHOS, and CHONS) for various single and mixed
precursor systems in negative-ionization-mode ESI (–) by LC-Orbitrap MS. The
grey bar (inconclusive compounds) indicates the signal attributed to compounds that were
not universally found in all repeat experiments.
(a) α-Pinene
As shown in Fig. 1a, the SOA particle mass produced in the single
precursor α-pinene system was greater than in any other system at
∼ 362 µg m-3. In the α-pinene single
precursor representative experiment, ∼ 55.6 % of the signal was
found in molecules containing only C, H, and O atoms, with the majority
consisting of 6 to 10 carbon atoms (47.5 %). Larger compounds were also
observed with carbon numbers ranging from C16 to C20
(representing 6.1 % of the total signal fraction; Fig.4a). Compounds
confidently found in the system in the CHON, CHONS, and CHOS groupings
represented 10.8 %, 5.2 %, and 3.2 % of the signal abundance,
respectively, again concentrated at C9–C10 and C16–C20.
C11–C15 molecules represent 2.2 % and 1.1 % of the signal in
the CHONS and CHOS categories, respectively. Inconclusively attributed
compounds contributed 25 % of the total signal abundance, with 47.6 % of
inconclusive compounds containing only C, H, and O atoms.
C6 to C10 compounds will include those produced through both
functionalization (addition oxygenated function group) and fragmentation
(cleavage of C–C bond) pathways during α-pinene oxidation (Eddingsaas et al., 2012). It has been further suggested that particle-phase dimerization and oligomerization reactions (e.g. alcohol + carbonyl to form hemiacetals and acetals, hydroperoxide + carbonyl to form peroxyhemiacetals and peroxyacetals, carboxylic acid + alcohol to form esters, and aldehyde self-reactions to form aldols) can play an important role in α-pinene oxidation (Ziemann and Atkinson, 2012; Gao et al., 2004a, b), resulting in formation of large molecules (nC > 10) and potentially accounting for the C16 to C20 abundance. Recent studies have additionally identified gas-phase autoxidation as playing a pivotal role in formation of highly oxygenated organic molecules (HOMs) (Tomaz et al., 2021; Crounse et al., 2013; Bianchi et al., 2019; Zhao et al., 2018). HOMs may condense on exiting seed particles or lead to new particle formation, depending on their vapour pressure (Tröstl et al., 2016). Autoxidation of RO2 radicals in the gas phase occurs rapidly via intermolecular and/or intramolecular hydrogen abstraction, leading to formation of R radicals with subsequent O2 addition (Mentel et al., 2015; Jokinen et al., 2014). The new RO2 radicals can undergo further autoxidation reaction or react with RO2 to generate dimer accretion products (Zhao et al., 2018; Berndt et al., 2018), leading to so-called highly oxygenated organic molecules (HOMs) with very low volatilities (Bianchi et al., 2019). Autoxidation may therefore contribute to CHO products with carbon numbers 16–20 in α-pinene oxidation (Berndt, 2021; Ehn et al., 2014). It has also been
found that the uptake of α-pinene oxidation products on ammonium
sulfate particles can lead to formation of organosulfate and nitrooxy organosulfate (Eddingsaas et al., 2012; Iinuma et al., 2009), contributing to the CHOS and CHONS groupings.
(b) Isoprene
As also seen in Fig. 1a, negligible SOA particle mass was generated in the
single precursor isoprene system (0.1 µg m-3, close to our chamber
background), and the total signal in Fig. 4b therefore corresponds to
extremely low SOA particle mass. Nevertheless, the presence of compounds in
all repeat experiments but not on any filters taken in background
experiments allows identification and attribution to isoprene products.
Similar to the α-pinene system, compounds found in all repeat
experiments containing CHO were the most abundant in the single precursor
isoprene experiment shown in Fig. 4b, with a normalized sample abundance
of 14.3 %, mainly comprising compounds with 4 or 5 carbon atoms.
Similarly, compounds in the CHONS classification can be confidently stated
to make a non-negligible contribution to the total signal fraction, with a
normalized abundance of 4.1 %, also mainly comprising C4–C5
compounds. CHON (1 %) and CHOS (1.5 %) each contributed significantly
less than the other molecular groupings. Figure 4b shows that the majority of
the signal in this single isoprene representative experiment was composed of
compounds (78.9 % of the normalized total signal fraction) that were not found
in all isoprene experiments (and/or were also detected in background
filters) and are therefore inconclusively assigned.
The presence of C4–C5 CHO compounds in a single isoprene
photo-oxidation system can be readily explained by established oxidation
pathways. For example, it is well-known that the double bond in isoprene is
oxidized to form C4 and C5 compounds, such as methacrolein (C4) and C5-hydroxycarbonyls as first-generation products, as well as 2-methylglyceric acid (C4) and isoprene tetrol (C5) as
second-generation products (Wennberg et al., 2018; Stroud et al., 2001;
Carlton et al., 2009). However, it is less clear how such small compounds
readily partition to the particle phase owing to their relatively high
vapour pressures, though it has been suggested that small compounds such as
glyoxal (CHOCHO) have extremely high activity coefficients when partitioning to aqueous particles, leading to low effective vapour pressures (Volkamer et
al., 2009) The possibility that small detected molecules were formed in the
filter sample extraction process cannot be ruled out. For example,
degradation of organic compounds can be induced by ultrasonic extraction of
particulate matter from filters (Miljevic et al., 2014; Mutzel et al.,
2013).
The negligible SOA particle mass formed in the isoprene single precursor
system is consistent with the literature observations (Kroll et al.,
2005a, b, 2006; Carlton et al., 2009).
However, condensed-phase reactions on acidic seeds would be expected to
appreciably increase this yield (Surratt et al., 2010,
2007a; Carlton et al., 2009). The large normalized signal contribution
corresponds to the high number of inconclusively assignable compounds
detected in this system. Most of these inconclusive compounds contained a
large number of carbon atoms (nC > 15). These compounds are likely
to have been formed via particle-phase accretion reactions, such as
oligomerization and organosulfate formation, even in the absence of acidity
in our experiments, leading to low-volatility higher-molecular-weight
accretion products (Berndt et al., 2019; Carlton et al., 2009). Whether
these products are formed on the filter medium or are present in the
suspended particle mass requires investigation. While these components are
the most abundant, this still corresponds to a very small mass compared to
all other systems and they were not found in all repeat experiments.
(c) o-Cresol
The particle wall-loss-corrected SOA mass concentration at the end of the
presented o-cresol experiment was approximately 101 µg m-3 (Fig. 1a). Figure 4c shows ∼ 26.6 % of the normalized signal
abundance in inconclusively assigned compounds, mainly in the CHON
classification. However, the key characteristic in the single precursor
o-cresol system is that the most abundant compounds that are confidently
found in all repeat experiments were found in the CHON category with between
6 and 8 carbon atoms (Fig. 4c) with around 72.1 % of the normalized
signal. CHO, CHONS, and CHOS groupings comprised around 1 % of the total
sample signal abundance. These three groups of compounds should not be
completely neglected since the SOA particle mass concentration of this
system was appreciable compared to other systems. It might be expected to
find a significant contribution of CHO compounds arising from formation of
organic acids (e.g. acetyl acrylic acid and glyoxylic acid) under high-NOxo-cresol photo-oxidation (Schwantes et
al., 2017).
Nitro-compounds retaining the carbon number of the parent VOC dominated the
CHON grouping. The C6–C8 components were identified as
methyl-nitrocatechol, C7H7NO4, isomers (see Table S1). OH
reaction with o-cresol forms various dihydroxytoluene isomers via addition
of the OH group to different positions on the ring (Olariu et al., 2002). Subsequent hydrogen abstraction followed by NO2 addition on the
ring at the moderate NOx concentrations of our experiments was a likely
dominant fate of dihydroxytoluene in the current study to form the observed
dihydroxy nitrotoluene. Further discussion of these isomers is presented in
Sect. 3.2.3: “Negative ionization mode (c)”. Schwantes et al. (2017) reported that H abstraction was not the dominant pathway in dihydroxy toluene oxidation, with dihydroxy nitrotoluene only detected at low concentrations by CIMS, with a significant number of highly oxygenated multigenerational products (mainly CHO compounds) detected by offline direct analysis in real-time mass spectrometry (DART-MS). It should be noted that the high signal contribution of CHON compounds, dominated by nitro-aromatics in o-cresol photo-oxidation (Kitanovski et al., 2012), in Fig. 4c may be influenced by their high negative-mode
sensitivity using electrospray ionization (Kiontke et al., 2016; Oss et
al., 2010).
(d) Binary α-pinene–isoprene mixture
The binary α-pinene–isoprene mixture generated considerable
particle wall-loss-corrected SOA particle mass in all experiments
(∼ 101 µg m-3 in the representative one shown here), which is
lower than in the single precursor α-pinene system but much higher
than in the isoprene system. The distribution of elemental categories of the
particle-phase products in this system was very similar to that in the
single precursor α-pinene experiments, with CHO compounds dominating
the total signal, mainly with between 6 and 10 carbon atoms or between 16
and 20 (Fig. 4d). The normalized signal contribution of compounds
confidently found in each repeat in the CHON group was slightly increased in
the binary α-pinene–isoprene system (14.4 %) compared to the single
α-pinene system (10.8 %) with a modest enhancement of compounds
with greater than 15 carbon atoms (from 3.3 % to 6.0 %). In addition,
the contribution of large compounds (nC > 15) was enhanced in the
CHON and CHONS categories in the binary system compared to the single VOC
α-pinene system.
This profile is consistent with the domination of the chemical composition
in the mixture by α-pinene products, which is unsurprising since
α-pinene is established as a much higher-yield SOA compound than
isoprene, especially under neutral seed conditions (Ahlberg et al., 2017;
Eddingsaas et al., 2012; Henry et al., 2012).
(e) Binary systems containing o-cresol
As shown in Fig. 1a, the isoprene–o-cresol system produces a low particle
mass concentration (∼ 22 µg m-3), whilst the α-pinene–o-cresol mixture generated the second-highest particle wall-loss-corrected SOA mass concentration (∼ 150 µg m-3).
Compounds found across repeat experiments in these mixtures containing
o-cresol show the same dominance of the CHON signal as the single precursor
o-cresol experiment (α-pinene–o-cresol, isoprene–o-cresol) (Fig. 4e and f). The contribution of CHON compounds to the total SOA increased to
approximately 87.8 % and 96.0 % when α-pinene and isoprene were
introduced into the mixed precursor systems, respectively. Moreover, the
contribution of CHO signal intensity increased in both binary o-cresol mixed
systems compared to the single precursor o-cresol system. Also, the
o-cresol–isoprene binary mixture (Fig. 4f) showed a slightly increased
proportion of signal in CHONS compounds at 1.0 % (compared with 0.6 %
in the single precursor o-cresol system; Fig. 4c), though noting that the
total mass concentration in the mixed system at the end of the experiment
was a factor of 5 lower than in the single VOC o-cresol system.
The presence of biogenic precursors leads to additional formation of CHO
compounds, while the relative signal contribution of CHON compounds is
reduced in each binary system compared to the single VOC o-cresol system. A
plausible explanation for this observation could be the increase in O3
generated in the binary mixture, increasing the ozonolysis of first-generation o-cresol products with double bonds and hence a higher CHO contribution
than in the sole o-cresol system. Overall, the negative-ionization-mode signal
from the SOA components in a binary mixture containing both biogenic and
anthropogenic precursors in our systems was dominated by categories of
components found in the single anthropogenic precursor system, specifically
the CHON group dominated by nitro-aromatics. This may be considered somewhat
surprising in the case of the mixture with α-pinene, since α-pinene (as widely reported and shown in Fig. 1) produces higher SOA mass
concentration than o-cresol under the same initial conditions as the mixture
experiment.
(f) Ternary α-pinene–isoprene–o-cresol mixture
Figure 4g shows the group contribution of the signals in the ternary
mixed VOC system corresponding to its moderately high SOA particles mass
concentration (∼ 85 µg m-3) shown in Fig. 1a. Across
the compounds found in all repeat experiments, whilst not as completely
dominant as in the o-cresol-containing binary systems, the substantial
(65.7 %) C6–C8 CHON contribution again shows that the
o-cresol-derived nitrocatechols play a significant role. CHO compounds make a
significant contribution with normalized abundance ∼ 14 %.
Whilst the CHON compounds mainly consist of C6–C8 compounds, the
CHO compounds comprise both C6–C8 and C9–C10 compounds.
SOA production in the ternary system appears not to be entirely driven by
any single precursor, and additionally, the overwhelming negative-mode CHON
dominance, which may be controlled by sensitivity of the electrospray method,
does not appear to the same degree in the ternary system as it does in the
o-cresol-containing binaries.
There was a small contribution to the CHO group from compounds with more
than 15 C atoms. Whilst relatively low in normalized signal contribution,
they were found in all ternary repeat experiments and can be presumed to be
accretion products. As an indication of the relative contribution of
accretion products to the SOA particle mass in each system, Table S2 shows
the signal-attributed mass concentration of molecules with nC > 21
that were observed confidently in all repeat experiments by scaling the
fractional signal contribution to the measured PM mass at the end of the
experiment. The signal-attributed mass concentration of these large
molecules is around 6, 575, and 80 times lower in the single VOC isoprene
system (0.002 µg m-3) than in the isoprene–o-cresol (0.013 µg m-3), α-pinene–isoprene (1.15 µg m-3), and ternary
(0.16 µg m-3) mixtures, respectively.
This section describes average properties of the SOA PM mass using a variety
of chemical metrics including molar carbon number (nC), molar hydrogen to
carbon ratio (H / C), oxygen to carbon ratio (O / C), average oxidation state (OS‾c), double bond equivalent (DBE), and double bond equivalent to carbon ratio (DBE / C). The molar carbon number reflects to the average size of SOA particle components, and often the major condensed-phase products retain the same carbon number as the precursor (Romonosky et al., 2015). H / C and O / C provide summary information about chemical composition of bulk organics, and OSc‾ corresponds to the average degree of oxidation of carbon in the organic species (value of OS‾c increasing upon oxidation) (Daumit et al., 2013; Safieddine and Heald, 2017). The OSc‾ values were calculated by using 2*O / C-H / C for CHO, CHONS, and CHOS compounds due to the low measured abundance fractions of two species in the oxidation products we observed in Sect. 3.2.2 Negative ionization mode and Sect. 3.2.2: “Positive ionization mode”. For CHON compounds, the equation OSc‾= 2*O / C-H / C–(OSN*N / C) was used to determined the OS‾c. OSN=+5 if nO >= 3 and OSN=+ 3 if nO < 3 for CHON compounds (Kroll et al., 2011). It is common to use DBE and DBE / C to quantify the unsaturated bonds (and aromaticity) in a molecule. The DBE corresponds to the sum of unsaturated bonds (including aromatic and cycloalkene rings), and increasing DBE / C ratios indicate an increasing contribution of the signal from molecules containing aromatic rings (Koch and Dittmar, 2006).
Table 3 shows the signal-weighted chemical metrics from compounds detected
in all repeat experiments in each system. All properties were normalized to
the total detected compound abundance. All parameters in the single VOC
α-pinene and binary α-pinene–isoprene systems are similar,
consistent with the dominance of α-pinene-derived particle mass in
the binary system. In contrast, the H / C value decreased from 1.46 to 1.03 and the O / C value remained constant (∼ 0.5) in binary α-pinene–o-cresol compared to the single VOC α-pinene system. Indeed, the signal-intensity-weighted average values of all chemical
parameters show that the o-cresol single VOC system aggregate properties are very similar to those in both o-cresol-containing binary systems, with an understandably high level of aromaticity (DBE / C >=0.67) (Koch and Dittmar, 2006), indicating that oxidation and partitioning to the particles in the single and binary o-cresol systems are largely ring-preserving. The OSc‾ value decreased from -0.55 to -0.63 in α-pinene–o-cresol compared to the single VOC o-cresol system, which suggests that less oxidized products were formed when introducing α-pinene precursors into the single o-cresol system. The abundance-weighted average values of all chemical parameters in the particles in the ternary mixture do not show common features with any single precursor system, with the coincidental exception of the nC and O / C value that are similar to that in the o-cresol system.
Intensity-weighted average values from negative-ionization-mode
LC-Orbitrap MS for O / C, H / C, OSc‾, DBE / C, DBE, and the number of carbons present (nC) for SOA filter extracts from single and mixed precursor experiments.
The weighted average number of carbons in the α-pinene experiment
(∼ 11) indicated that a modest accretion reaction (including
oligomerization and functionalization) occurred in oxidation and that the
α-pinene particle-phase oxidation products had a significant impact on the
α-pinene–isoprene binary system. The average carbon number of
isoprene SOA particles was larger than the isoprene precursor (C5),
implying particle-phase accretion reactions such as organosulfate formation
though forming very little particle mass in the current study. The
similarity of properties between the single VOC o-cresol system and its
binary mixtures suggests that common compounds dominate the signals, and from
Fig. 4c, e, and f it can be seen that these are compounds in the CHON
elemental category. In addition, the DBE / C values indicate dominance of the
major oxidation products in these o-cresol-containing systems by condensed
aromatic structure, consistent with the finding in Ahlberg et
al. (2017).
Positive ionization mode
Figure 5 presents the positive-ionization-mode signal fraction of compounds
in representative experiments that can be confidently stated as found in
each of the systems (i.e. found in every repeat experiment in this system)
in the coloured stacked bars according to their carbon number and classified
according to their elemental CHO, CHON, CHOS, and CHONS categories. Also
shown in the grey bar is the signal fraction of compounds that are
inconclusively found in the experiment in each system classified by
elemental grouping, but not found in all repeat experiments or the chamber
background. The fractional contributions of confidently stated products
are similar for each experiment in a particular system.
It is evident that there is a generally greater fraction of the positive-ionization-mode signal that is inconclusive than in negative ionization mode
as shown in Fig. 4. This indicates larger variability in composition
between repeat experiments, with some compounds not found in some repeats
experiments, or a larger fraction of the signal from compounds also found on
chamber background filters. Moreover, the greater fraction of
“inconclusive” compounds in positive ionization mode might also be attributed
to automated non-targeted method programming. For example, the automated
non-targeted method is programmed such that a compound will be removed from the final
detected molecule peak list when it has a signal-to-noise ratio below 3
and low measured signal abundance close to the signal-to-noise cut-off
values in the replicate experiment. The automated non-targeted method also
programmed the molecular formula assignment base on the isotopic pattern,
wherein the isotopic intensity tolerance was within ±30 % of the
theoretical isotopic abundance. Consequently, it becomes a challenge to
accurately assign a molecular formula to compounds with “large” molecular
weights due to a rising number of possible formulas. The large
compound could have different molecular formula assignments in
“representative” and replicate experiments, though it has a
similar retention time and molecular weight in both experiments.
The normalized signal intensity distribution of different compound
categories (CHO, CHON, CHOS, and CHONS) for various single and mixed
precursor systems in positive ionization mode by LC-Orbitrap MS. The grey
bar (inconclusive compounds) is the signal attributed to compounds that were not
universally found in all repeat experiments.
(a) α-Pinene
In the single precursor α-pinene system (Fig. 5a), 33.5 % of
the total signal abundance was from CHO compounds found in each repeat
experiment, with the majority of molecules containing between 6 and 10 carbon
atoms. The compounds confidently found in the CHOS category provided 7.4 %
of the signal fraction, also mainly comprising compounds with 6 to 10 carbon
atoms. The remainder of the signal was observed in CHON (14.6 %) and CHONS
(3.5 %) categories, which were found in all repeat experiments in this
system mainly comprised large compounds, with some nC < 11 molecules
in the CHON category.
The contribution of C9 to C10 molecules in the CHO and CHONS
categories is consistent with previous studies of α-pinene
ozonolysis and OH oxidation in the presence of NOx and seed particles
(Winterhalter et al., 2003, Yasmeen et al., 2012). The signal contribution
of CHO compounds and CHOS with carbon numbers 6 to 8 suggested that
fragmentation plays an important role. It is likely that these compounds
formed from fragmentation of alkoxy radicals (RO2+ NO → RO + NO2) (Pullinen et al., 2020).
The CHOS and CHONS compounds may be attributed to esterification of α-pinene SOA. Experimental results from Surratt et al. (2007b) reported that sulfate esters and/or their derivatives have a
significant contribution to SOA formation of α-pinene
photo-oxidation in the presence of ammonium sulfate seed. The large
molecules in CHON and CHONS groups suggest the occurrence of accretion
reactions between peroxy-peroxy radicals containing nitrogen and sulfur
(RO2+ R'O2→ROOR'+O2)
(Pullinen et al., 2020).
(b) Isoprene
Considering only the compounds found in all repeat experiments and not on
the background filter, the dominant contribution in the isoprene signal in
the single VOC photo-oxidation system was observed in the CHO category with
a normalized signal of 14.8 %, with molecules mostly comprising 5–8 carbon
atoms or larger molecules with carbon number greater than 9 (Fig. 5b).
CHONS compounds are the next-largest constituent, with 11.7 % of the total
signal. More than half of the CHONS signal is from large molecules
(nC > 11), and the rest of the CHONS compounds mainly comprise
molecules with carbon numbers of 6 to 10. The remainder of the signal was found
in the CHON and CHOS categories with a carbon number greater than 11.
Compounds which could not be confidently attributed to the isoprene system
owing to their sole presence in every repeat experiment made a significant
contribution (∼ 65 %) to the total signal, though an even
greater fraction of inconclusive signal (78.9 %) was observed in negative
ionization mode (likely resulting from the extremely low total mass yield).
Clearly, accretion reactions dominated the isoprene system in the positive
(as well as negative) ionization modes. The contribution of CHONS compounds to
total SOA is consistent with the formation of organosulfate and nitrooxy
organosulfate by uptake of isoprene oxides on ammonium sulfate particles
(Surratt et al., 2007b, a). Moreover, the presence of
C4–C5 molecules in CHO categories could be simply explained by the
gas-phase oxidation pathway of isoprene, though as with the negative-mode
samples, it is unclear why such small molecules partition to the particle
phase. A possible interpretation is that weakly bound large molecules
fragment during LC-Orbitrap MS analysis and/or due to the possibly gas-phase filter absorption (Lopez-Hilfiker
et al., 2016).
(c) o-Cresol
Figure 5c shows that approximately 20 % of the signal is in the CHON category
in the single VOC o-cresol system. The majority (18.3 %) of the signal from
confidently attributable molecules found in all repeat experiments in this
CHON category contains 6 to 8 carbon atoms. The signal in the CHO and CHOS
categories is similarly dominated by compounds containing 6 to 8 carbon
atoms with fractional contributions of 5.4 % and 1.7 %, respectively.
Specifically, the C7 compounds have fractional signal contributions of
3.8 %, which is approximately 3 times higher than C6 (1.1 %) and
about 12 times higher than C8 molecules (0.3 %) in CHO categories.
In the CHOS categories, C6 organic species (1.3 %) made the dominant
contribution compared to C7 (0.3 %) and C8 (0.05 %) species.
The CHONS category in this system almost entirely comprised molecules that
were not found in all repeat experiments and are therefore considered inconclusive in
this analysis.
Compounds found in all repeat experiments with between 11 and 15 carbon
atoms in the CHON category account for 1.6 % of the signal (with C14 1.5 % and C11-12 0.1 %). It is likely that the majority of the
C11 to C15 signal is attributed to C7 dimers.
C6-8 CHON compounds are likely to be second-generation o-cresol
oxidation products such as dihydroxy nitrotoluene, which are also detected in
negative ionization mode as a result of being both protonated and
deprotonated. The CHO compound present in this study might have some
contribution from multigenerational products generated from decomposition of
bicyclic intermediate compounds formed from OH oxidation of o-cresol, as
reported by Schwantes et al. (2017), but they are
probably mainly dihydroxy toluene compounds, which have been reported with a
70 % yield from o-cresol oxidation (Olariu et al., 2002).
Decomposition of bicyclic intermediate compounds leading to formation of
unsaturated carbonyl molecules could form oligomeric species, resulting in
formation of the C11–15 molecules in the CHO and CHON groups.
(d) Binary α-pinene–isoprene mixture
The elemental categories in the binary α-pinene–isoprene samples
shown in Fig. 5d indicate high similarity to the single VOC α-pinene system (Fig. 5a), with CHO compounds dominating the total signal
and predominantly containing 9 to 10 carbon atoms, but with some
fragmentation to C6–C8. The signal intensity of CHOS compounds
was reduced by 0.7 % of the total signal in the binary system (Fig. 5d)
compared to the single VOC α-pinene system (Fig. 5a), mostly in
the C6–C8 signal. In contrast, the signal intensity of CHON
components is 19.7 % of the total in the binary system, which is 5.1 % higher
than in the single VOC α-pinene system, with enhancement in
molecules with a carbon number > 16.
The similarity in the elemental categorization between the single VOC α-pinene and binary α-pinene–isoprene system again supports the
contention that α-pinene components dominate the total signal in the
binary system. However, the enhancement of CHON compound intensity in the
binary system possible implies an increase in the RO2/ NO2 or
RO2/ NO termination pathways, leading to stronger organic nitrate
formation. A large fraction of the signal from molecules with a carbon number
greater than 16 in this binary system might be attributed to dimerization of
gas-phase nitrated highly oxidized molecules.
(e) Binary systems containing o-cresol
The distribution of SOA products from α-pinene–o-cresol (Fig. 5e)
and isoprene–o-cresol binary systems (Fig. 5f) shows obvious differences
compared to the corresponding single precursor systems. In the α-pinene–o-cresol binary system, the dominant signal intensity was
contributed by CHON compounds, and they mainly comprise molecules with more
than 16 carbon atoms. The rest of the signal was found in the CHO (9.0 %),
CHOS (1.6 %), and CHONS (3.3 %) categories, while compounds with
nC >= 9 made up a significant proportion. In the
isoprene–o-cresol system, most of the compounds were in the CHON category
(17.2 %), and the majority of them were composed of 6 to 10 carbon atoms.
C9–C15molecules also made a non-negligible contribution in CHON
compounds (7.4 %). The remainder of the signal was found in the CHO
(12.9 %), CHONS (1.3 %), and CHOS (5.8 %) categories, again
concentrated at C6-8.
Lack of similarities between o-cresol-containing binary systems and the
corresponding sole precursor systems in the positive ionization mode
suggests a significant contribution to the signal from the unique compounds
shown in Fig. 3 exerting some control over the elemental composition of
SOA in binary systems. For instance, cross-products from α-pinene
and o-cresol gas- or particle-phase oxidation probably contribute to the high-carbon-number compounds in the binary system. In the isoprene–o-cresol system,
high C6–C8 contributions in all categories were likely from
o-cresol, though the other contributions were dissimilar to the individual
precursor systems.
(f) Ternary α-pinene–isoprene–o-cresol mixture
The distribution of SOA products in the ternary system (Fig. 5g) was very
similar to the single precursor α-pinene experiments (Fig. 5a). The
dominant compounds were found in the CHO categories with a signal intensity of
21.1 %, most of them with 6 to 10 carbon atoms. The 17.5 % signal
contribution of molecules with carbon numbers greater than C16 in CHON
is 7.2 % higher than the signal intensity of CHON molecules with carbon number > 16 in the single precursor α-pinene system (10.3 %).
The most notable difference between the positive-mode signal in the ternary
system and the single precursor systems was the high contribution of
molecules with nC > 21 in the CHON category. As an indication of the relative contribution of accretion products to the SOA particle mass in each system, Table S2 shows that the signal-attributed mass concentration of molecules (nC > 21) in the single VOC isoprene system, at 0.016 µg m-3, is significantly lower than in the α-pinene–o-cresol binary (2.85 µg m-3) and is about 8 times less than in the isoprene–o-cresol binary (0.14 µg m-3) and 70 times less than the ternary (1.10 µg m-3) systems, which is comparable to the single
precursor α-pinene system (1.34 µg m-3). The SOA particle products of the ternary system are mainly attributable to α-pinene oxidation and accretion reactions, possibly across different precursor products, leading to high-carbon-number nitrogen-containing compounds.
Table 4 shows the intensity-weighted average values for compounds detected
in positive ionization mode in all repeat experiments of individual SOA
systems. All properties were normalized to the total detected compound
abundance. Clearly, the nC values in all three single VOC systems were
higher than their precursor's carbon number. For example, the nC value in
isoprene SOA is 11.73, which is 2 times higher than carbon number of isoprene
(C5). In the binary α-pinene–isoprene system, the nC (11.90) was slightly higher than in the single α-pinene system (11.55) and in
the single isoprene system, suggesting a contribution from each. The OSc‾ values seem comparable in both single systems and binary α-pinene–isoprene systems. The average value of nC in the binary α-pinene–o-cresol system (17.88) was significantly higher than in the single VOC α-pinene (11.55) and o-cresol systems (7.61). The O / C values in the binary α-pinene–o-cresol system were approximately 0.15 lower than the sole α-pinene and o-cresol system, while the H / C values in the binary α-pinene–o-cresol system are comparable to single α-pinene and about ∼ 0.4 times higher than the sole o-cresol system. The average value of nC in the binary isoprene–o-cresol system (8.43) was lower than sole isoprene systems (nC = 11.73) but higher than the single o-cresol system (nC = 7.61). The signal-intensity-weighted values for all chemical parameters in the ternary mixture show no obvious similarity to those in any sole precursor system, with the exception of the DBE / C parameter.
It is apparent in the positive mode that accretion reactions occurred, and
its products play an essential role in single isoprene systems, binary
α-pinene-containing systems, and the ternary system. It cannot be
discounted that chemical transformation may occur during filter sample
preparation, which might impact the intensity-weighted average values of
various chemical properties. Moreover, although some of the chemical
parameters in the binary system show similar values compared to single
precursor systems, the significant differences between mixed systems and
those of the individual precursors imply that categories of components in
the mixed systems were controlled by the compounds that were unique to the
mixture and not found in the single precursor systems.
Intensity-weighted average values obtained from positive-ionization-mode LC-Orbitrap MS for O / C, H / C, OSc‾, DBE, and the number of carbons (nC) present for SOA filter extracts from the single and mixed precursor experiments.
Chemicalα-PineneIsopreneo-Cresolα-Pinene–Isoprene–α-Pinene–α-Pinene–parametersisopreneo-cresolo-cresolisoprene–o-cresolnC11.5511.737.6111.908.4317.8813.69H / C1.561.651.091.541.251.551.52O / C0.320.360.360.290.460.170.23OS‾c-0.95-1.00-0.66-1.03-0.50-1.38-1.17DBE / C0.320.320.640.330.540.310.33DBE3.723.154.823.864.285.494.55Insights from the combination of positive- and negative-mode elemental
categorization of signal contribution
Considering the results of both negative and positive ionization modes, the
α-pinene-derived compounds unsurprisingly dominate the elemental
categorization of the binary α-pinene–isoprene binary system, since
α-pinene produced a much greater mass concentration than isoprene.
The average carbon number in positive ionization mode (Tables 3 and 4) shows
that SOA formation in the binary α-pinene–isoprene binary system
involves similar accretion products as found in the single VOC α-pinene system. Whilst o-cresol generated appreciable SOA particle mass
concentration, this was still significantly lower than in the single α-pinene system. However, the negative-mode analysis suggests that
o-cresol oxidation products can make a more significant contribution than
α-pinene products, notwithstanding the particularly high sensitivity
to aromatic nitro-compounds, which make a high contribution to the
o-cresol CHON category. The positive mode, being sensitive to a different
subset of the compounds, differs to the observations in negative ionization
mode. The observation in positive mode reveals that SOA elemental
composition in the binary o-cresol–α-pinene system is not driven by any
single precursor's oxidation products but by the new compounds that appear
to be o-cresol–α-pinene large molecular cross-products.
Approximately half of the compounds were unique in the binary
o-cresol–α-pinene system in both positive and negative modes (Figs. 2b and 3b). In the o-cresol–isoprene system, it may be expected that
the elemental composition was driven by the o-cresol since the isoprene
oxidation produced very little particulate mass compared to that of
o-cresol. Negative ionization results were consistent with this, though
positive mode indicated an additional significant contribution from
o-cresol–isoprene large molecular cross-products. Overall, SOA particle
formation in binary systems can be seen to be mainly dependent on high-yield precursors but is also influenced by the interaction between products of
the individual precursors, with the unique compounds making a greater
contribution than any sole precursor's products in positive ionization mode
of the o-cresol–α-pinene system. In the ternary system, the
elemental composition has a striking resemblance to the single α-pinene
system in positive ionization mode, but in negative ionization mode, there
was little similarity with any single precursor system, with all three
precursors contributing. On the other hand, the elemental grouping results
clearly show that the compounds that were not present in all repeat
experiments and are hence inconclusively attributable in all precursor systems
made non-negligible contributions in both modes (especially in positive
ionization), suggesting that the repeatability of SOA chemical composition in
each system is not ideal. This may be an artefact of the inherent difficulty
of precisely replicating operating process during chamber experiments. It
should not affect the analysis of SOA chemical characterization between
single and mixture precursor systems, since only the confidently
attributable compounds between repeat experiments were employed for
comparison.
Molecular characterization of particulate organicsNegative ionization mode
This section aims to investigate whether the components in mixtures were
also present at significant fractional abundance in particles or absent from
any of the single VOC photo-oxidation systems. The absence in the single VOC
systems of those components making a substantial contribution to the
mixtures may be indicative of interactions during the photochemistry and
multiphase processing giving rise to tracers of the combinations of VOC
precursors in multicomponent particles that may be of use in SOA source
attribution in future ambient studies. The normalized peak area of 15
selected compounds in the binary mixed system and 20 selected compounds in the ternary
mixed system is shown in Fig. 6. In all mixed systems, the five compounds
with the highest signal fraction that were also present in each
corresponding single precursor system are shown alongside the top five
compounds uniquely found in the mixture but absent from any single precursor
system.
Only compounds found in all repeat experiments in each system were chosen
for this analysis, so there is confidence in the component identification.
The dominant compounds in terms of their normalized peak area in
the mixed VOC systems shown in the bars: (a) binary α-pinene–isoprene system, (b) binary α-pinene–o-cresol, (c) binary isoprene–o-cresol, (d) ternary system. The normalized peak areas of
these selected compounds in a mixed precursor system are also presented if
they existed in the corresponding single precursor system (yellow: single
o-pinene, green: isoprene, orange: o-cresol) . The compounds are considered
identical in the mixed system and single VOC systems if they have the same
empirical formula and a retention time difference of < 0.1 min in
negative ionization mode.
(a) The binary α-pinene–isoprene system
The components in the binary mixture system that were also found in the single
precursor α-pinene system were found to have a larger signal
fraction than those found in the single isoprene system and the unique
compounds. In particular, it was found that C9H14O4 made the
greatest signal contribution (Fig. 6a). This is also the case in the single
precursor α-pinene system, and it is likely to be pinic acid due to
this peak having a similar fragmentation pattern (Figure S3) compared to results
reported in Yasmeen et al. (2010). C8H14O2 and
C9H12N2O8 made a non-negligible contribution in the
binary mixture system, with a normalized molecular abundance of 4.4 % and
3.4 %, and were also found to be conserved in both single precursor systems.
Compounds that were only present in the binary mixture had relatively low
abundance, with the highest contribution from
C17H14N2O17 with only 0.7 % of the total signal fraction.
Clearly, the SOA particle composition in the binary α-pinene–isoprene system was dominated by α-pinene components and partially
contributed by isoprene, but not those from cross-products from their
interaction.
(b) Binary α-pinene–o-cresol system
As shown in Fig. 6b, the four most abundant peaks (three
C7H7NO4 isomers (i to iii) and C7H7NO3) in the
mixture were found to be present in the single precursor o-cresol system.
C7H7NO4 isomers in the binary mixture had signal
contributions of 51.7 % (iii), 8.6 % (ii), and 2.5 % (i),
with the C7H7NO3 contribution of 24.3 %.
C9H14O4 is present in both the single VOC α-pinene
system and mixed system, with a relatively high (0.68 %) signal contribution
in the mixed system compared to the other four compounds common to the
mixture and α-pinene alone. The top five unique compounds in the binary
mixture system had a small normalized signal fraction compared to the total sample
abundance in the range of 0.12 % to 0.17 %.
The four dominant compounds in the binary mixture are all nitro-aromatic
compounds formed in the oxidation of o-cresol (Schwantes et al., 2017;
Kitanovski et al., 2012). C7H7NO4 includes multiple isomers of
methyl-nitrocatechol with the methyl, hydroxyl, and nitro groups at various
positions on the aromatic rings. C7H7NO3 was identified as
methyl-nitrophenol. (Details of deprotonated species of
C7H7NO4 and C7H7NO3 in Table S1). As with the
group categorization, care must be taken with the interpretation of the
molecular contributions to the signal owing to the enhanced sensitivity of
electrospray ionization.
(c) Binary isoprene–o-cresol
system
Figure 6c shows that only one compound in the binary isoprene–o-cresol
system was unequivocally observed in all repeat experiments in the single
isoprene precursor system. Components present in the single o-cresol system make a
higher contribution in the binary mixture system than isoprene-derived compounds
and those unique to the mixture, with one C7H7NO3 and three
C7H7NO4 isomers making the most significant contribution.
According to the deprotonated molecular species fragmentation (Table S1),
three C7H7NO4 isomers were found at retention times of 9.14, 4.52,
and 7.53. These three C7H7NO4 isomers have similar fragmentation
ions that relate to loss of the NO ion (m/z= 138) and NOH ion (m/z= 137). The
five compounds that were unique to the mixture were found to make negligible
contributions to total sample abundance (between 0.05 % to 0.2 %).
As in the α-pinene–o-cresol binary mixture, the compounds found in
the o-cresol system dominate the SOA particles in the binary isoprene–o-cresol system. Isoprene-derived compounds were found to make a negligible
contribution; all dominant compounds in the binary system were found in
the single VOC o-cresol system and only one compound in the single VOC
isoprene system. There is no evidence to suggest that a compound has a high
enough contribution to act as a tracer for the binary mixture. The three
dominant compounds (C7H7NO4 isomers) were uniquely identified
as o-cresol oxidation products (methyl-nitrocatechol isomers) with similar
retention time and fragmentation ions as the C7H7NO4
compounds that were found in the binary α-pinene–o-cresol system.
As with the group categorization the consideration of enhanced sensitivity
of electrospray ionization must be borne in mind in the isoprene–o-cresol
and α-pinene–o-cresol mixtures.
(d) The ternary mixture
In the ternary system (Fig. 6d), the top three largest contributing signals
(C7H7NO4 isomers) are from an o-cresol oxidation product, and the other
two o-cresol compounds have a comparable normalized peak area (∼ 1.2 %). Also, the α-pinene SOA makes a non-negligible
contribution in the range of 0.7 % to 2.2 % in the ternary mixture system,
though this is significantly lower than o-cresol-derived compounds. Five isoprene-derived compounds (0.28 % to 0.76 %) make comparable signal contributions
to the five unique compounds (0.25 % to 1.1 %).
o-Cresol SOA and α-pinene SOA clearly significantly influenced the
chemical composition in the ternary system, while the isoprene SOA and
unique compound contributions are modest. A unique potential tracer
compound (C21H34O6) was only observed in this ternary
mixture of α-pinene, o-cresol, and isoprene with a 1.1 %
contribution and was found in all repeat experiments.
Positive ionization mode
Figure 7 compares the normalized peak area of selected compounds in mixed
and single precursor systems in positive ionization mode. As the negative
ionization mode, Fig. 7 shows 15 selected compounds in each binary mixed
system and 20 compounds in ternary mixed system, following the same
selection criteria.
The dominant compounds in terms of their normalized peak area in
the mixed VOC systems shown in the bars: (a) binary α-pinene–isoprene system, (b) binary α-pinene–o-cresol, (c) binary isoprene–o-cresol, (d) ternary system. The normalized peak areas of these selected compounds in a mixed precursor system are also presented if they existed in the corresponding single precursor system (yellow: single α-pinene, green: isoprene, orange: o-cresol). The compounds are considered identical in the mixed system and single VOC systems if they have the same empirical formula and a retention time difference of < 0.1 min in positive ionization mode.
(a) The binary α-pinene–isoprene system
Figure 7a indicates that α-pinene-derived compounds dominated the
binary α-pinene–isoprene system: C10H14O2 with the
highest normalized peak area of 15.5 %, followed by
C20H31NO4 at 8.6 %. The contribution of isoprene-derived
compounds (0.5 %–1.4 %) is lower than that of those derived from α-pinene but higher than compounds unique to the mixture, the highest
fractional abundance of which was 0.32 % (C8H10O).
The particle components in the binary α-pinene–isoprene system were
substantially driven by the α-pinene components as found in negative
ionization mode, likely resulting from the low SOA yield of isoprene
oxidation under the conditions of our experiment. The isoprene components
had little influence on the composition in this system. There is
insufficient information to suggest that a compound has a high enough
contribution to act as a tracer for the binary mixture, but compounds unique
to this mixture with seed particles under moderate NOx conditions were
found to be sulfur-containing.
(b) Binary α-pinene–o-cresol system
Two α-pinene-derived compounds dominated this system
(C21N33NO3 and C20H31NO3) in positive
ionization mode (Fig. 7b). The other three α-pinene-derived
compounds were found at levels comparable to the top two derived from
o-cresol (C8H11NO and C10H13NO2) at approximately
2.5 % of the total molecular signal. The contribution of compounds unique to
the mixture were lower than all five α-pinene-derived compounds but
higher than most o-cresol SOA. The highest contribution from these unique
compounds was C21H33NO4 with 1.8 % signal intensity.
Although both α-pinene and o-cresol oxidation products contributed to
this system, the most abundant peaks (C21H33NO3 and
C20H31NO3) were only found in the single precursor α-pinene system but not in the single precursor o-cresol system. The
nitrogen-containing compound (C21H33NO4) might act as a tracer
compound for the binary system, which is possibly driven by further oxidation of the
C21H33NO3 compound.
(c) Binary isoprene–o-cresol system
Figure 7c shows that o-cresol-derived compounds controlled the particulate chemical
composition in this system. The fractional contributions of
C7H8O4 and C7H7NO2 from the o-cresol system
were 15.5 % and 10.5 %, respectively, in the binary mixture. One isoprene-derived compound (C9H13NO2) made a considerable contribution
(6.6 %) in the binary mixture system. Compounds unique to the mixture were
C9H11NO (1.3 %), C8H8O10(1.0 %),
C13H29NO5 (0.9 %), C7H18N2O2S2
(0.7 %), and C16H30O6S2(0.6 %).
The higher-yield o-cresol made a much more significant contribution to the
SOA components than the lower-yield isoprene. The significant abundance of
two unique compounds (C9H11NO, and C8H8O10 ) may
result from interactions in the mixture, and their exploration for use as
tracers of the mixed system might prove useful.
(d) The ternary mixture
From Fig. 7(d), the dominant compounds of the ternary system in positive
ionization mode were derived from α-pinene with a fractional
contribution range of 0.39 % to 14.9 %. The highest peak was
C10H14O with a signal intensity 14.9 %. The top o-cresol-derived
compounds were C10H13NO2, C8H11NO, and
C6H10O2 with 2.7 %, 1.6 %, and 1.2 % signal intensity. The top three isoprene-derived compounds were
C8H16O6S (2.0 %), C13H26N2OS2
(1.6 %), and C13H26N2OS2 (1.2 %) respectively.
C20H28N4O4S was unique to the ternary mixture with a
fractional contribution of approximately 1.1 %, which could
be the products from dimerization of α-pinene products or from
interactions in the mixture.
The highest-SOA-yield α-pinene clearly dominated the product
distribution of the ternary mixture in positive ionization mode. Isoprene-
and o-cresol-derived as well as unique-to-mixture components made little
contribution.
Further insight from companion papers
This study probes the chemical composition and interactions during SOA
formation in mixed VOC systems using the offline LC-Orbitrap MS technique.
The complete instrument description and experimental design are given in
Voliotis et al. (2022b), along with the data
from online techniques (e.g. SMPS, semi-continuous GCMS, HR-ToF-AMS, and
FIGAERO-CIMS). Comprehensive analysis of FIGAERO-CIMS and HR-ToF-AMS data is
provided in Voliotis et al. (2022a), Voliotis et al. (2021), and Du et al.
(2021). Voliotis et al. (2022a) and
Voliotis et al. (2021) investigated the
volatility distribution of products in mixed systems using the FIGAERO-CIMS
and a thermal denuder coupled with an SMPS and HR-ToF-AMS.
Voliotis et al. (2021) reported FIGAERO-CIMS
measurements showing an abundance of products uniquely found in the α-pinene–o-cresol mixture, with the majority in the nC = 5–10 and
nC > 10 classes. This result is consistent with the finding in
this study that unique compounds were found in the α-pinene–o-cresol
mixture obtained from LC-Orbitrap MS measurement, likely the cross-products
from α-pinene and o-cresol oxidation in the particle phase.
Voliotis et al. (2021) observed a dominant
contribution of nitrogen-containing compounds to the total signal in all
o-cresol-containing systems, similar to the results obtained from negative
ionization mode in LC-Orbitrap MS in this study. This is unsurprising owing
to the high sensitivity of the iodide CIMS to o-cresol photo-oxidation-produced nitro-aromatic compounds with hydroxyl groups, such as
methyl-nitrocatechol and methyl-nitrophenol (Lee et al., 2014; Iyer et
al., 2016).
Conclusion
In this study, the SOA chemical composition formed from the photo-oxidation
of α-pinene, isoprene, o-cresol, and their binary and ternary mixtures
in the presence of NOx and ammonium sulfate seed particles was
determined by non-targeted LC-Orbitrap MS. SOA particle mass from isoprene
was almost negligible under our experimental conditions; o-cresol generated
more and α-pinene the highest and exhibited the highest yield in
our experiments.
The number of detected SOA compounds and their molecular composition
indicated that α-pinene oxidation products have a dominant influence
on the SOA particle composition in the binary α-pinene–isoprene
system, which can involve oligomerization–accretion reactions forming
products such as C20H31NO4. The major products in this system
show that SOA composition is clearly driven by the high α-pinene
yield, with isoprene oxidation products observed to make a minor
contribution. The nitrogen-containing compound
C17H14N2O7 might be a potential tracer in binary
α-pinene–isoprene systems in the presence of ammonium sulfate seed.
The compositional analysis in negative ionization mode reveals that
o-cresol products dominate SOA particle composition in the α-pinene–o-cresol system, with major contributions from methyl-nitrocatechol
isomers (C7H7NO4) and methyl-nitrophenol
(C7H7NO3), though this will be influenced by the high
sensitivity in the employed electrospray ionization method. There is a
relatively high contribution to the elemental composition from
unique-to-mixture products in positive ionization mode, indicating the
significant prevalence of interactions between the oxidation products in
this system. The molecular analysis in both ionization modes also indicated
that both α-pinene and o-cresol influenced the product distribution
in their binary mixture.
Similarly, o-cresol oxidation heavily influenced SOA particle composition in
the binary isoprene–o-cresol system in negative ionization mode, but
unique-to-mixture products made considerable contributions in the
positive ionization mode. The molecular analysis in both modes suggested
that higher-yield o-cresol products were present in greater abundance than
those from isoprene. Two unique compounds (C9H11NO and
C8H8O10) in positive mode were identified that could behave
as tracers in this system.
SOA composition in binary mixtures was therefore generally strongly
determined by the oxidation products of the higher-yield precursors, but
interactions leading to cross-product formation also play an important role,
especially in o-cresol-containing systems.
In the ternary system, the elemental category composition analysis presented
in positive ionization mode suggested that the chemical composition of SOA
strongly depends on sole α-pinene oxidation, with products from
the oxidation of α-pinene and o-cresol identified as important in
negative ionization mode. The molecular analysis shows that products from
both α-pinene and o-cresol strongly influence the composition of SOA
particles with very few isoprene oxidation products making a major
contribution, indicating a limited role for isoprene oxidation. Moreover,
cross-products C21H34O6 and C20H28N4O4S were identified as potential tracers in
the ternary system.
This study did not examine the molecular structure of the unique
compounds or potential tracers in the mixture precursor systems. It is suggested that future
studies focus on identifying the molecular structure of
unique-to-mixture components, which will help researchers better understand the detailed
mechanisms of interactions involved in ambient SOA formation from mixture
VOC oxidations.
Data availability
All the data used in this work are available upon request from the corresponding authors.
The supplement related to this article is available online at: https://doi.org/10.5194/acp-22-9799-2022-supplement.
Author contributions
GM, MRA, AV, YW, and YS conceived the study. AV, YW, YS, and MD conducted the
experiments. KP provided on-site LC-Orbitrap MS training for filter analysis
and provided the automated non-targeted method for LC-Orbitrap MS
analysis. YS conducted the data analysis and wrote the paper with
contributions from all co-authors.
Competing interests
At least one of the (co-)authors is a member of the editorial board of Atmospheric Chemistry and Physics. The peer-review process was guided by an independent editor, and the authors also have no other competing interests to declare.
Disclaimer
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Acknowledgements
We acknowledge support from the NERC Atmospheric Measurement and Observational Facility (AMOF) in providing the SMPS instrument. We thank Jacqui Hamilton and Kelly Pereira, who offered valuable assistance and guidance on LC-Orbitrap MS instruments in the laboratory and data analysis.
Financial support
The Manchester Aerosol Chamber received funding support from the European Union's Horizon 2020 research and innovation programme under grant agreement no. 730997, which supports the EUROCHAMP2020 research programme. Instrumentational support was provided by the NERC Atmospheric Measurement and Observational Facility (AMOF). Yu Wang was supported by the joint scholarship of The University of Manchester and the Chinese Scholarship Council. M. Rami Alfarra was funded by the UK National Centre for Atmospheric Sciences (NACS). Aristeidis Voliotis received funding support from the Natural Environment Research Council (NERC) EAO Doctoral Training
Partnership.
Review statement
This paper was edited by Tao Wang and reviewed by two anonymous referees.
ReferencesAhlberg, E., Falk, J., Eriksson, A., Holst, T., Brune, W. H., Kristensson,
A., Roldin, P., and Svenningsson, B.: Secondary organic aerosol from VOC
mixtures in an oxidation flow reactor, Atmos. Environ., 161,
210–220, 10.1016/j.atmosenv.2017.05.005, 2017.
Andreae, M. O. and Crutzen, P. J.: Atmospheric aerosols: Biogeochemical
sources and role in atmospheric chemistry, Science, 276, 1052–1058, 1997.Arndt, D., Wachsmuth, C., Buchholz, C., and Bentley, M.: A complex matrix
characterization approach, applied to cigarette smoke, that integrates
multiple analytical methods and compound identification strategies for
non-targeted liquid chromatography with high-resolution mass spectrometry,
Rapid Commun. Mass Sp., 34, e8571, 10.1002/rcm.8571, 2019.
Atkinson, R.: Gas-phase tropospheric chemistry of volatile organic
compounds: 1. Alkanes and alkenes, J. Phys. Chem.
Ref. Data, 26, 215–290, 1997.Atkinson, R.: Atmospheric chemistry of VOCs and NOx, Atmos. Environ., 34, 2063–2101,
10.1016/S1352-2310(99)00460-4, 2000.
Atkinson, R. and Arey, J.: Gas-phase tropospheric chemistry of biogenic
volatile organic compounds: a review, Atmos. Environ., 37, 197–219,
2003.Atkinson, R. and Aschmann, S. M.: Products of the gas-phase reactions of
aromatic hydrocarbons: Effect of NO2 concentration, Int. J.
Chem. Kinet., 26, 929–944, 10.1002/kin.550260907, 1994.Atkinson, R., Baulch, D. L., Cox, R. A., Crowley, J. N., Hampson, R. F., Hynes, R. G., Jenkin, M. E., Rossi, M. J., and Troe, J.: Evaluated kinetic and photochemical data for atmospheric chemistry: Volume I – gas phase reactions of Ox, HOx, NOx and SOx species, Atmos. Chem. Phys., 4, 1461–1738, 10.5194/acp-4-1461-2004, 2004.Berndt, T.: Peroxy Radical Processes and Product Formation in the OH
Radical-Initiated Oxidation of α-Pinene for Near-Atmospheric
Conditions, J. Phys. Chem. A, 125, 9151–9160, 10.1021/acs.jpca.1c05576, 2021.Berndt, T., Scholz, W., Mentler, B., Fischer, L., Herrmann, H., Kulmala, M.,
and Hansel, A.: Accretion Product Formation from Self- and Cross-Reactions
of RO2 Radicals in the Atmosphere, Angew. Chem. Int. Edit.,
57, 3820–3824, 10.1002/anie.201710989, 2018.Berndt, T., Hyttinen, N., Herrmann, H., and Hansel, A.: First oxidation
products from the reaction of hydroxyl radicals with isoprene for pristine
environmental conditions, Communications Chemistry, 2, 21, 10.1038/s42004-019-0120-9, 2019.Bianchi, F., Kurtén, T., Riva, M., Mohr, C., Rissanen, M. P., Roldin,
P., Berndt, T., Crounse, J. D., Wennberg, P. O., Mentel, T. F., Wildt, J.,
Junninen, H., Jokinen, T., Kulmala, M., Worsnop, D. R., Thornton, J. A.,
Donahue, N., Kjaergaard, H. G., and Ehn, M.: Highly Oxygenated Organic
Molecules (HOM) from Gas-Phase Autoxidation Involving Peroxy Radicals: A Key
Contributor to Atmospheric Aerosol, Chem. Rev., 119, 3472–3509, 10.1021/acs.chemrev.8b00395, 2019.
Bravo-Linares, C., Mudge, S., and Loyola-Sepulveda, R.: Production of
volatile organic compounds (VOCs) by temperate macroalgae. The use of Solid
Phase Microextraction (SPME) coupled to GC-MS as method of analysis, J.
Chil. Chem. Soc., 55, 227–232, 10.4067/S0717-97072010000200018, 2010.Buiarelli, F., Di Filippo, P., Pomata, D., Riccardi, C., and Bartocci, M.: A
liquid chromatography tandem mass spectrometry method for simultaneous
analysis of 46 atmospheric particulate-phase persistent organic pollutants
and comparison with gas chromatography/mass spectrometry, Int.
J. Environ. An. Chem., 97, 797–818, 10.1080/03067319.2017.1369535, 2017.Burnett, R., Pope, C., Ezzati, M., Olives, C., Lim, S., Mehta, S., Shin, H.,
Singh, G., Hubbell, B., Brauer, M., Anderson, H., Smith, K., Balmes, J.,
Bruce, N., Kan, H., Laden, F., Prüss-Ustün, A., Turner, M., Gapstur,
S., and Cohen, A.: An Integrated Risk Function for Estimating the Global
Burden of Disease Attributable to Ambient Fine Particulate Matter Exposure,
Environ. Health Persp., 122, 397–403, 10.1289/ehp.1307049, 2014.Camredon, M., Hamilton, J. F., Alam, M. S., Wyche, K. P., Carr, T., White, I. R., Monks, P. S., Rickard, A. R., and Bloss, W. J.: Distribution of gaseous and particulate organic composition during dark α-pinene ozonolysis, Atmos. Chem. Phys., 10, 2893–2917, 10.5194/acp-10-2893-2010, 2010.Canagaratna, M. R., Jayne, J. T., Jimenez, J. L., Allan, J. D., Alfarra, M. R., Zhang, Q., Onasch, T. B., Drewnick, F., Coe, H., Middlebrook, A., Delia,
A., Williams, L. R., Trimborn, A. M., Northway, M. J., DeCarlo, P. F., Kolb, C. E., Davidovits, P., and Worsnop, D. R.: Chemical and microphysical characterization of ambient aerosols with the aerodyne aerosol mass spectrometer, Mass Spectrom. Rev., 26, 185–222, 10.1002/mas.20115, 2007.Carlton, A. G., Wiedinmyer, C., and Kroll, J. H.: A review of Secondary Organic Aerosol (SOA) formation from isoprene, Atmos. Chem. Phys., 9, 4987–5005, 10.5194/acp-9-4987-2009, 2009.Cech, N. and Enke, C.: Practical Implications of Some Recent Studies in
Electrospray Ionization Fundamentals, Mass Spectrom. Rev., 20,
362–387, 10.1002/mas.10008, 2001.Cech, N. B. and Enke, C. G.: Relating Electrospray Ionization Response to
Nonpolar Character of Small Peptides, Anal. Chem., 72, 2717–2723, 10.1021/ac9914869, 2000.Coscollà, C., Yusa, V., Marti Requena, P., and Pastor, A.: Analysis of
currently used pesticides in fine airborne particulate matter (PM2.5) by
pressurized liquid extraction and liquid chromatography-tandem mass
spectrometry, J. Chromatogr. A, 1200, 100–107, 10.1016/j.chroma.2008.05.075, 2008.Cropper, P. M., Eatough, D. J., Overson, D. K., Hansen, J. C., Caka, F., and
Cary, R. A.: Use of a gas chromatography–mass spectrometry organic aerosol
monitor for in-field detection of fine particulate organic compounds in
source apportionment, J. Air Waste Manage.,
68, 390–402, 10.1080/10962247.2017.1363095, 2018.Crounse, J. D., Nielsen, L. B., Jørgensen, S., Kjaergaard, H. G., and
Wennberg, P. O.: Autoxidation of Organic Compounds in the Atmosphere, The
J. Phys. Chem. Lett., 4, 3513–3520, 10.1021/jz4019207,
2013.Daumit, K. E., Kessler, S. H., and Kroll, J. H.: Average chemical properties
and potential formation pathways of highly oxidized organic aerosol, Faraday
Discuss., 165, 181–202, 10.1039/c3fd00045a, 2013.DeCarlo, P. F., Kimmel, J. R., Trimborn, A., Northway, M. J., Jayne, J. T.,
Aiken, A. C., Gonin, M., Fuhrer, K., Horvath, T., Docherty, K. S., Worsnop,
D. R., and Jimenez, J. L.: Field-Deployable, High-Resolution, Time-of-Flight
Aerosol Mass Spectrometer, Anal. Chem., 78, 8281–8289, 10.1021/ac061249n, 2006.Dommen, J., Metzger, A., Duplissy, J., Kalberer, M., Alfarra, M., Gascho,
A., Weingartner, E., Prevot, A., Verheggen, B., and Baltensperger, U.:
Laboratory observation of oligomers in the aerosol from isoprene/NOx
photooxidation, Geophys. Res. Lett., 33, 338–345, 10.1029/2006GL026523, 2006.Donahue, N. M., Epstein, S. A., Pandis, S. N., and Robinson, A. L.: A two-dimensional volatility basis set: 1. organic-aerosol mixing thermodynamics, Atmos. Chem. Phys., 11, 3303–3318, 10.5194/acp-11-3303-2011, 2011.Du, M., Voliotis, A., Shao, Y., Wang, Y., Bannan, T. J., Pereira, K. L., Hamilton, J. F., Percival, C. J., Alfarra, M. R., and McFiggans, G.: Combined application of Online FIGAERO-CIMS and Offline LC-Orbitrap MS to Characterize the Chemical Composition of SOA in Smog Chamber Studies, Atmos. Meas. Tech. Discuss. [preprint], 10.5194/amt-2021-420, in review, 2021.Eddingsaas, N. C., Loza, C. L., Yee, L. D., Chan, M., Schilling, K. A., Chhabra, P. S., Seinfeld, J. H., and Wennberg, P. O.: α-pinene photooxidation under controlled chemical conditions – Part 2: SOA yield and composition in low- and high-NOx environments, Atmos. Chem. Phys., 12, 7413–7427, 10.5194/acp-12-7413-2012, 2012.Ehn, M., Thornton, J. A., Kleist, E., Sipilä, M., Junninen, H.,
Pullinen, I., Springer, M., Rubach, F., Tillmann, R., Lee, B.,
Lopez-Hilfiker, F., Andres, S., Acir, I.-H., Rissanen, M., Jokinen, T.,
Schobesberger, S., Kangasluoma, J., Kontkanen, J., Nieminen, T., Kurtén,
T., Nielsen, L. B., Jørgensen, S., Kjaergaard, H. G., Canagaratna, M.,
Maso, M. D., Berndt, T., Petäjä, T., Wahner, A., Kerminen, V.-M.,
Kulmala, M., Worsnop, D. R., Wildt, J., and Mentel, T. F.: A large source of
low-volatility secondary organic aerosol, Nature, 506, 476–479, 10.1038/nature13032, 2014.Fiore, A. M., Naik, V., Spracklen, D. V., Steiner, A., Unger, N., Prather,
M., Bergmann, D., Cameron-Smith, P. J., Cionni, I., Collins, W. J.,
Dalsøren, S., Eyring, V., Folberth, G. A., Ginoux, P., Horowitz, L. W.,
Josse, B., Lamarque, J.-F., MacKenzie, I. A., Nagashima, T., O'Connor, F.
M., Righi, M., Rumbold, S. T., Shindell, D. T., Skeie, R. B., Sudo, K.,
Szopa, S., Takemura, T., and Zeng, G.: Global air quality and climate,
Chem. Soc. Rev., 41, 6663–6683, 10.1039/C2CS35095E, 2012.Gao, S., Keywood, M., Ng, N. L., Surratt, J., Varutbangkul, V., Bahreini,
R., Flagan, R. C., and Seinfeld, J. H.: Low-Molecular-Weight and Oligomeric
Components in Secondary Organic Aerosol from the Ozonolysis of Cycloalkenes
and α-Pinene, J. Phys. Chem. A, 108, 10147–10164, 10.1021/jp047466e, 2004a.Gao, S., Ng, N. L., Keywood, M., Varutbangkul, V., Bahreini, R., Nenes, A.,
He, J., Yoo, K. Y., Beauchamp, J. L., Hodyss, R. P., Flagan, R. C., and
Seinfeld, J. H.: Particle Phase Acidity and Oligomer Formation in Secondary
Organic Aerosol, Environ. Sci. Technol., 38, 6582–6589, 10.1021/es049125k, 2004b.
Glasius, M., Duane, M., and Larsen, B. R.: Determination of polar terpene
oxidation products in aerosols by liquid chromatography-ion trap mass
spectrometry, J. Chromatogr. A, 833, 121–135, 1999.Hallquist, M., Wenger, J. C., Baltensperger, U., Rudich, Y., Simpson, D., Claeys, M., Dommen, J., Donahue, N. M., George, C., Goldstein, A. H., Hamilton, J. F., Herrmann, H., Hoffmann, T., Iinuma, Y., Jang, M., Jenkin, M. E., Jimenez, J. L., Kiendler-Scharr, A., Maenhaut, W., McFiggans, G., Mentel, Th. F., Monod, A., Prévôt, A. S. H., Seinfeld, J. H., Surratt, J. D., Szmigielski, R., and Wildt, J.: The formation, properties and impact of secondary organic aerosol: current and emerging issues, Atmos. Chem. Phys., 9, 5155–5236, 10.5194/acp-9-5155-2009, 2009.Hamilton, J. F., Bryant, D. J., Edwards, P. M., Ouyang, B., Bannan, T. J.,
Mehra, A., Mayhew, A. W., Hopkins, J. R., Dunmore, R. E., Squires, F. A.,
Lee, J. D., Newland, M. J., Worrall, S. D., Bacak, A., Coe, H., Percival,
C., Whalley, L. K., Heard, D. E., Slater, E. J., Jones, R. L., Cui, T.,
Surratt, J. D., Reeves, C. E., Mills, G. P., Grimmond, S., Sun, Y., Xu, W.,
Shi, Z., and Rickard, A. R.: Key Role of NO3 Radicals in the Production of
Isoprene Nitrates and Nitrooxyorganosulfates in Beijing, Environ.
Sci. Technol., 55, 842–853, 10.1021/acs.est.0c05689, 2021.Henry, K. M., Lohaus, T., and Donahue, N. M.: Organic Aerosol Yields from
α-Pinene Oxidation: Bridging the Gap between First-Generation Yields
and Aging Chemistry, Environ. Sci. Technol., 46, 12347–12354, 10.1021/es302060y, 2012.Hoffmann, T., Odum, J. R., Bowman, F., Collins, D., Klockow, D., Flagan, R.
C., and Seinfeld, J. H.: Formation of Organic Aerosols from the Oxidation of
Biogenic Hydrocarbons, J. Atmos. Chem., 26, 189–222, 10.1023/A:1005734301837, 1997.Iinuma, Y., Böge, O., Kahnt, A., and Herrmann, H.: Laboratory chamber
studies on the formation of organosulfates from reactive uptake of
monoterpene oxides, Phys. Chem. Chem. Phys., 11, 7985–7997, 10.1039/B904025K, 2009.Iyer, S., Lopez-Hilfiker, F., Lee, B. H., Thornton, J. A., and Kurtén,
T.: Modeling the Detection of Organic and Inorganic Compounds Using
Iodide-Based Chemical Ionization, J. Phys. Chem. A, 120,
576–587, 10.1021/acs.jpca.5b09837, 2016.Jimenez, J. L., Canagaratna, M. R., Donahue, N. M., Prevot, A. S. H., Zhang,
Q., Kroll, J. H., DeCarlo, P. F., Allan, J. D., Coe, H., Ng, N. L., Aiken,
A. C., Docherty, K. S., Ulbrich, I. M., Grieshop, A. P., Robinson, A. L.,
Duplissy, J., Smith, J. D., Wilson, K. R., Lanz, V. A., Hueglin, C., Sun, Y.
L., Tian, J., Laaksonen, A., Raatikainen, T., Rautiainen, J., Vaattovaara,
P., Ehn, M., Kulmala, M., Tomlinson, J. M., Collins, D. R., Cubison, M. J.,
Dunlea, J., Huffman, J. A., Onasch, T. B., Alfarra, M. R., Williams, P. I.,
Bower, K., Kondo, Y., Schneider, J., Drewnick, F., Borrmann, S., Weimer, S.,
Demerjian, K., Salcedo, D., Cottrell, L., Griffin, R., Takami, A., Miyoshi,
T., Hatakeyama, S., Shimono, A., Sun, J. Y., Zhang, Y. M., Dzepina, K.,
Kimmel, J. R., Sueper, D., Jayne, J. T., Herndon, S. C., Trimborn, A. M.,
Williams, L. R., Wood, E. C., Middlebrook, A. M., Kolb, C. E.,
Baltensperger, U., and Worsnop, D. R.: Evolution of Organic Aerosols in the
Atmosphere, Science, 326, 1525, 10.1126/science.1180353, 2009.Jokinen, T., Sipilä, M., Richters, S., Kerminen, V.-M., Paasonen, P.,
Stratmann, F., Worsnop, D., Kulmala, M., Ehn, M., Herrmann, H., and Berndt,
T.: Rapid Autoxidation Forms Highly Oxidized RO2 Radicals in the Atmosphere,
Angew. Chem. Int. Edit., 53, 14596–14600,
10.1002/anie.201408566, 2014.Kahnt, A., Iinuma, Y., Blockhuys, F., Mutzel, A., Vermeylen, R.,
Kleindienst, T. E., Jaoui, M., Offenberg, J. H., Lewandowski, M., Böge,
O., Herrmann, H., Maenhaut, W., and Claeys, M.: 2-Hydroxyterpenylic Acid: An
Oxygenated Marker Compound for α-Pinene Secondary Organic Aerosol in
Ambient Fine Aerosol, Environ. Sci. Technol., 48, 4901–4908, 10.1021/es500377d, 2014.Kiendler-Scharr, A., Wildt, J., Maso, M. D., Hohaus, T., Kleist, E., Mentel,
T. F., Tillmann, R., Uerlings, R., Schurr, U., and Wahner, A.: New particle
formation in forests inhibited by isoprene emissions, Nature, 461, 381–384, 10.1038/nature08292, 2009.Kiendler-Scharr, A., Andres, S., Bachner, M., Behnke, K., Broch, S., Hofzumahaus, A., Holland, F., Kleist, E., Mentel, T. F., Rubach, F., Springer, M., Steitz, B., Tillmann, R., Wahner, A., Schnitzler, J.-P., and Wildt, J.: Isoprene in poplar emissions: effects on new particle formation and OH concentrations, Atmos. Chem. Phys., 12, 1021–1030, 10.5194/acp-12-1021-2012, 2012.Kiontke, A., Oliveira-Birkmeier, A., Opitz, A., and Birkemeyer, C.:
Electrospray Ionization Efficiency Is Dependent on Different Molecular
Descriptors with Respect to Solvent pH and Instrumental Configuration, PLOS
ONE, 11, e0167502, 10.1371/journal.pone.0167502, 2016.Kitanovski, Z., Grgić, I., Yasmeen, F., Claeys, M., and Čusak, A.:
Development of a liquid chromatographic method based on ultraviolet–visible
and electrospray ionization mass spectrometric detection for the
identification of nitrocatechols and related tracers in biomass burning
atmospheric organic aerosol, Rapid Commun. Mass Sp., 26,
793–804, 10.1002/rcm.6170, 2012.Koch, B. P. and Dittmar, T.: From mass to structure: an aromaticity index
for high-resolution mass data of natural organic matter, Rapid Commun. Mass Sp., 20, 926–932,
10.1002/rcm.2386, 2006.Kroll, J. H., Ng, N. L., Murphy, S. M., Flagan, R. C., and Seinfeld, J. H.:
Secondary organic aerosol formation from isoprene photooxidation under
high-NOx conditions, Geophys. Res. Lett., 32, L18808,
10.1029/2005GL023637, 2005a.Kroll, J. H., Ng, N. L., Murphy, S. M., Varutbangkul, V., Flagan, R. C., and
Seinfeld, J. H.: Chamber studies of secondary organic aerosol growth by
reactive uptake of simple carbonyl compounds, J. Geophys. Res.-Atmos., 110, D23207, 10.1029/2005JD006004, 2005b.
Kroll, J. H., Ng, N. L., Murphy, S. M., Flagan, R. C., and Seinfeld, J. H.:
Secondary organic aerosol formation from isoprene photooxidation,
Environ. Sci. Technol., 40, 1869–1877, 2006.Kroll, J. H., Donahue, N. M., Jimenez, J. L., Kessler, S. H., Canagaratna,
M. R., Wilson, K. R., Altieri, K. E., Mazzoleni, L. R., Wozniak, A. S.,
Bluhm, H., Mysak, E. R., Smith, J. D., Kolb, C. E., and Worsnop, D. R.:
Carbon oxidation state as a metric for describing the chemistry of
atmospheric organic aerosol, Nat. Chem., 3, 133–139, 10.1038/nchem.948, 2011.Lee, B.-H., Pierce, J. R., Engelhart, G. J., and Pandis, S. N.: Volatility of secondary organic aerosol from the ozonolysis of monoterpenes, Atmos. Environ., 45, 2443–2452, 10.1016/j.atmosenv.2011.02.004, 2011.Lee, B. H., Lopez-Hilfiker, F. D., Mohr, C., Kurtén, T., Worsnop, D. R.,
and Thornton, J. A.: An Iodide-Adduct High-Resolution Time-of-Flight
Chemical-Ionization Mass Spectrometer: Application to Atmospheric Inorganic
and Organic Compounds, Environ. Sci. Technol., 48, 6309–6317, 10.1021/es500362a, 2014.Li, N., He, Q., Greenberg, J., Guenther, A., Li, J., Cao, J., Wang, J., Liao, H., Wang, Q., and Zhang, Q.: Impacts of biogenic and anthropogenic emissions on summertime ozone formation in the Guanzhong Basin, China, Atmos. Chem. Phys., 18, 7489–7507, 10.5194/acp-18-7489-2018, 2018.Lim, Y. B. and Ziemann, P. J.: Effects of molecular structure on aerosol
yields from OH radical-initiated reactions of linear, branched, and cyclic
alkanes in the presence of NOx, Environ. Sci. Technol., 43,
2328–2334, 2009.Liu, L.-b., Liu, Y., Lin, J.-m., Tang, N., Hayakawa, K., and Maeda, T.:
Development of analytical methods for polycyclic aromatic hydrocarbons (PAHs) in airborne particulates: A review, J. Environ. Sci., 19, 1–11, 10.1016/S1001-0742(07)60001-1, 2007.Lopez-Hilfiker, F. D., Mohr, C., Ehn, M., Rubach, F., Kleist, E., Wildt, J., Mentel, Th. F., Lutz, A., Hallquist, M., Worsnop, D., and Thornton, J. A.: A novel method for online analysis of gas and particle composition: description and evaluation of a Filter Inlet for Gases and AEROsols (FIGAERO), Atmos. Meas. Tech., 7, 983–1001, 10.5194/amt-7-983-2014, 2014.Lopez-Hilfiker, F. D., Mohr, C., D'Ambro, E. L., Lutz, A., Riedel, T. P.,
Gaston, C. J., Iyer, S., Zhang, Z., Gold, A., Surratt, J. D., Lee, B. H.,
Kurten, T., Hu, W. W., Jimenez, J., Hallquist, M., and Thornton, J. A.:
Molecular Composition and Volatility of Organic Aerosol in the Southeastern
U.S.: Implications for IEPOX Derived SOA, Environ. Sci. Technol., 50, 2200–2209, 10.1021/acs.est.5b04769, 2016.McFiggans, G., Mentel, T. F., Wildt, J., Pullinen, I., Kang, S., Kleist, E.,
Schmitt, S., Springer, M., Tillmann, R., and Wu, C.: Secondary organic
aerosol reduced by mixture of atmospheric vapours, Nature, 565, 587, 10.1038/s41586-018-0871-y, 2019.McFiggans, G., Artaxo, P., Baltensperger, U., Coe, H., Facchini, M. C., Feingold, G., Fuzzi, S., Gysel, M., Laaksonen, A., Lohmann, U., Mentel, T. F., Murphy, D. M., O'Dowd, C. D., Snider, J. R., and Weingartner, E.: The effect of physical and chemical aerosol properties on warm cloud droplet activation, Atmos. Chem. Phys., 6, 2593–2649, 10.5194/acp-6-2593-2006, 2006.McLuckey, S. A. and Wells, J. M.: Mass Analysis at the Advent of the 21st
Century, Chem. Rev., 101, 571–606, 10.1021/cr990087a, 2001.McNeill, V. F.: Aqueous Organic Chemistry in the Atmosphere: Sources and
Chemical Processing of Organic Aerosols, Environ. Sci. Technol., 49, 1237–1244, 10.1021/es5043707, 2015.McVay, R. C., Zhang, X., Aumont, B., Valorso, R., Camredon, M., La, Y. S., Wennberg, P. O., and Seinfeld, J. H.: SOA formation from the photooxidation of α-pinene: systematic exploration of the simulation of chamber data, Atmos. Chem. Phys., 16, 2785–2802, 10.5194/acp-16-2785-2016, 2016.Mehra, A., Bannan, T., Worrall, S., Bacak, A., Priestley, M., Liu, D., Zhao,
J., Xu, W., Sun, Y., Hamilton, J., Squires, F., Lee, J., Bryant, D.,
Hopkins, J., Elzein, A., Budisulistiorini, S., Cheng, X., Qi, C., and Coe,
H.: Using highly time-resolved online mass spectrometry to examine biogenic
and anthropogenic contributions to organic aerosol in Beijing, Faraday
Discuss., 226, 382–408, 10.1039/D0FD00080A, 2021.Mentel, T. F., Springer, M., Ehn, M., Kleist, E., Pullinen, I., Kurtén, T., Rissanen, M., Wahner, A., and Wildt, J.: Formation of highly oxidized multifunctional compounds: autoxidation of peroxy radicals formed in the ozonolysis of alkenes – deduced from structure–product relationships, Atmos. Chem. Phys., 15, 6745–6765, 10.5194/acp-15-6745-2015, 2015.Mezcua, M., Malato, O., Martinez-Uroz, M. A., Lozano, A., Agüera, A.,
and Fernández-Alba, A. R.: Evaluation of Relevant Time-of-Flight-MS
Parameters Used in HPLC/MS Full-Scan Screening Methods for Pesticide
Residues, J. AOAC Int., 94, 1674–1684, 10.5740/jaoacint.SGEMezcua, 2011.Miljevic, B., Hedayat, F., Stevanovic, S., Fairfull-Smith, K. E., Bottle, S.
E., and Ristovski, Z. D.: To Sonicate or Not to Sonicate PM Filters:
Reactive Oxygen Species Generation Upon Ultrasonic Irradiation, Aerosol
Sci. Tech., 48, 1276–1284, 10.1080/02786826.2014.981330, 2014.Minaeian, J. K.: Development and Deployment of an Airborne Gas
Chromatography/Mass Spectrometer to Measure Tropospheric Volatile Organic
Compounds, PhD, University of York, https://etheses.whiterose.ac.uk/18227/ (last access: 1 August 2022), 2017.Mutzel, A., Rodigast, M., Iinuma, Y., Böge, O., and Herrmann, H.: An
improved method for the quantification of SOA bound peroxides, Atmos. Environ., 67, 365–369, 10.1016/j.atmosenv.2012.11.012, 2013.Mutzel, A., Rodigast, M., Iinuma, Y., Böge, O., and Herrmann, H.:
Monoterpene SOA – Contribution of first-generation oxidation products to
formation and chemical composition, Atmos. Environ., 130, 136–144,
10.1016/j.atmosenv.2015.10.080, 2016.Nestorowicz, K., Jaoui, M., Rudzinski, K. J., Lewandowski, M., Kleindienst, T. E., Spólnik, G., Danikiewicz, W., and Szmigielski, R.: Chemical composition of isoprene SOA under acidic and non-acidic conditions: effect of relative humidity, Atmos. Chem. Phys., 18, 18101–18121, 10.5194/acp-18-18101-2018, 2018.Ng, N. L., Chhabra, P. S., Chan, A. W. H., Surratt, J. D., Kroll, J. H., Kwan, A. J., McCabe, D. C., Wennberg, P. O., Sorooshian, A., Murphy, S. M., Dalleska, N. F., Flagan, R. C., and Seinfeld, J. H.: Effect of NOx level on secondary organic aerosol (SOA) formation from the photooxidation of terpenes, Atmos. Chem. Phys., 7, 5159–5174, 10.5194/acp-7-5159-2007, 2007.Ng, N. L., Kwan, A. J., Surratt, J. D., Chan, A. W. H., Chhabra, P. S., Sorooshian, A., Pye, H. O. T., Crounse, J. D., Wennberg, P. O., Flagan, R. C., and Seinfeld, J. H.: Secondary organic aerosol (SOA) formation from reaction of isoprene with nitrate radicals (NO3), Atmos. Chem. Phys., 8, 4117–4140, 10.5194/acp-8-4117-2008, 2008.Novakov, T. and Penner, J.: Large contribution of organic aerosols to
cloud-condensation-nuclei concentrations, Nature, 365, 823, 10.1038/365823a0, 1993.
Odum, J. R., Hoffmann, T., Bowman, F., Collins, D., Flagan, R. C., and
Seinfeld, J. H.: Gas/particle partitioning and secondary organic aerosol
yields, Environ. Sci. Technol., 30, 2580–2585, 1996.Olariu, R. I., Klotz, B., Barnes, I., Becker, K. H., and Mocanu, R.: FT–IR
study of the ring-retaining products from the reaction of OH radicals with
phenol, o-, m-, and p-cresol, Atmos. Environ., 36, 3685–3697,
10.1016/S1352-2310(02)00202-9, 2002.Ono-Ogasawara, M., Myojo, T., and Smith, T. J.: A simple direct injection method for GC/MS analysis of PAHs in particulate matter, Ind. Health, 46, 582–593, 10.2486/indhealth.46.582, 2008.Oss, M., Kruve, A., Herodes, K., and Leito, I.: Electrospray Ionization
Efficiency Scale of Organic Compounds, Anal. Chem., 82, 2865–2872, 10.1021/ac902856t, 2010.Pandis, S. N., Paulson, S. E., Seinfeld, J. H., and Flagan, R. C.: Aerosol
formation in the photooxidation of isoprene and β-pinene, Atmos.
Environ. A-Gen., 25, 997–1008,
10.1016/0960-1686(91)90141-S, 1991.Pereira, K., Ward, M., Wilkinson, J., Sallach, J., Bryant, D., Dixon, W.,
Hamilton, J., and Lewis, A.: An Automated Methodology for Non-targeted
Compositional Analysis of Small Molecules in High Complexity Environmental
Matrices Using Coupled Ultra Performance Liquid Chromatography Orbitrap Mass
Spectrometry, Environ. Sci. Technol., 55, 7365–7375, 10.1021/acs.est.0c08208, 2021.Pereira, K. L., Hamilton, J. F., Rickard, A. R., Bloss, W. J., Alam, M. S., Camredon, M., Muñoz, A., Vázquez, M., Borrás, E., and Ródenas, M.: Secondary organic aerosol formation and composition from the photo-oxidation of methyl chavicol (estragole), Atmos. Chem. Phys., 14, 5349–5368, 10.5194/acp-14-5349-2014, 2014.Pereira, K. L., Hamilton, J. F., Rickard, A. R., Bloss, W. J., Alam, M. S.,
Camredon, M., Ward, M. W., Wyche, K. P., Muñoz, A., Vera, T.,
Vázquez, M., Borrás, E., and Ródenas, M.: Insights into the
Formation and Evolution of Individual Compounds in the Particulate Phase
during Aromatic Photo-Oxidation, Environ. Sci. Technol., 49,
13168–13178, 10.1021/acs.est.5b03377, 2015.Place, B. J., Ulrich, E. M., Challis, J. K., Chao, A., Du, B., Favela, K.,
Feng, Y.-L., Fisher, C. M., Gardinali, P., Hood, A., Knolhoff, A. M.,
McEachran, A. D., Nason, S. L., Newton, S. R., Ng, B., Nuñez, J., Peter,
K. T., Phillips, A. L., Quinete, N., Renslow, R., Sobus, J. R., Sussman, E.
M., Warth, B., Wickramasekara, S., and Williams, A. J.: An Introduction to
the Benchmarking and Publications for Non-Targeted Analysis Working Group,
Anal. Chem., 93, 16289–16296, 10.1021/acs.analchem.1c02660, 2021.Priego-Capote, F. and Luque de Castro, M. D.: Analytical uses of ultrasound
I. Sample preparation, TrAC-Trend. Anal. Chem., 23, 644–653,
10.1016/j.trac.2004.06.006, 2004.Pullinen, I., Schmitt, S., Kang, S., Sarrafzadeh, M., Schlag, P., Andres, S., Kleist, E., Mentel, T. F., Rohrer, F., Springer, M., Tillmann, R., Wildt, J., Wu, C., Zhao, D., Wahner, A., and Kiendler-Scharr, A.: Impact of NOx on secondary organic aerosol (SOA) formation from α-pinene and β-pinene photooxidation: the role of highly oxygenated organic nitrates, Atmos. Chem. Phys., 20, 10125–10147, 10.5194/acp-20-10125-2020, 2020.Romonosky, D. E., Laskin, A., Laskin, J., and Nizkorodov, S. A.:
High-Resolution Mass Spectrometry and Molecular Characterization of Aqueous
Photochemistry Products of Common Types of Secondary Organic Aerosols,
J. Phys. Chem. A, 119, 2594–2606, 10.1021/jp509476r, 2015.Safieddine, S. A. and Heald, C. L.: A Global Assessment of Dissolved Organic
Carbon in Precipitation, Geophys. Res. Lett., 44, 11672–611681,
10.1002/2017GL075270, 2017.Saldarriaga-Noreña, H., López-Márquez, R., Murillo Tovar, M. A.,
Arias-Montoya, M., Guerrero-Álvarez, J., and Vergara, J.: Recent
Advances for Polycyclic Aromatic Analysis in Airborne Particulate Matter,
in: Hydrocarbon Pollution and its Effect on the Environment, edited by: Muharrem, I. and Olcay Kaplan, I., IntechOpen, 10.5772/intechopen.79714, 2018.Schervish, M. and Donahue, N. M.: Peroxy radical chemistry and the volatility basis set, Atmos. Chem. Phys., 20, 1183–1199, 10.5194/acp-20-1183-2020, 2020.Schwantes, R. H., Schilling, K. A., McVay, R. C., Lignell, H., Coggon, M. M., Zhang, X., Wennberg, P. O., and Seinfeld, J. H.: Formation of highly oxygenated low-volatility products from cresol oxidation, Atmos. Chem. Phys., 17, 3453–3474, 10.5194/acp-17-3453-2017, 2017.Shao, Y., Wang, Y., Du, M., Voliotis, A., Alfarra, M. R., O'Meara, S. P., Turner, S. F., and McFiggans, G.: Characterisation of the Manchester Aerosol Chamber facility, Atmos. Meas. Tech., 15, 539–559, 10.5194/amt-15-539-2022, 2022.Shilling, J. E., Zawadowicz, M. A., Liu, J., Zaveri, R. A., and Zelenyuk,
A.: Photochemical Aging Alters Secondary Organic Aerosol Partitioning
Behavior, ACS Earth Space Chem., 3, 2704–2716, 10.1021/acsearthspacechem.9b00248, 2019.Shilling, J. E., Zaveri, R. A., Fast, J. D., Kleinman, L., Alexander, M. L., Canagaratna, M. R., Fortner, E., Hubbe, J. M., Jayne, J. T., Sedlacek, A., Setyan, A., Springston, S., Worsnop, D. R., and Zhang, Q.: Enhanced SOA formation from mixed anthropogenic and biogenic emissions during the CARES campaign, Atmos. Chem. Phys., 13, 2091–2113, 10.5194/acp-13-2091-2013, 2013.Shrivastava, M., Cappa, C. D., Fan, J., Goldstein, A. H., Guenther, A. B.,
Jimenez, J. L., Kuang, C., Laskin, A., Martin, S. T., Ng, N. L., Petaja, T.,
Pierce, J. R., Rasch, P. J., Roldin, P., Seinfeld, J. H., Shilling, J.,
Smith, J. N., Thornton, J. A., Volkamer, R., Wang, J., Worsnop, D. R.,
Zaveri, R. A., Zelenyuk, A., and Zhang, Q.: Recent advances in understanding
secondary organic aerosol: Implications for global climate forcing, Rev.
Geophys., 55, 509–559,
10.1002/2016RG000540, 2017.Singh, D. P., Gadi, R., and Mandal, T. K.: Characterization of
particulate-bound polycyclic aromatic hydrocarbons and trace metals
composition of urban air in Delhi, India, Atmos. Environ., 45,
7653–7663, 10.1016/j.atmosenv.2011.02.058,
2011.Steckel, A. and Schlosser, G.: An Organic Chemist's Guide to Electrospray
Mass Spectrometric Structure Elucidation, Molecules, 24, 611, 10.3390/molecules24030611, 2019.Stroud, C. A., Roberts, J. M., Goldan, P. D., Kuster, W. C., Murphy, P. C.,
Williams, E. J., Hereid, D., Parrish, D., Sueper, D., Trainer, M.,
Fehsenfeld, F. C., Apel, E. C., Riemer, D., Wert, B., Henry, B., Fried, A.,
Martinez-Harder, M., Harder, H., Brune, W. H., Li, G., Xie, H., and Young,
V. L.: Isoprene and its oxidation products, methacrolein and methylvinyl
ketone, at an urban forested site during the 1999 Southern Oxidants Study,
J. Geophys. Res.-Atmos., 106, 8035–8046,
10.1029/2000JD900628, 2001.Surratt, J. D., Lewandowski, M., Offenberg, J. H., Jaoui, M., Kleindienst,
T. E., Edney, E. O., and Seinfeld, J. H.: Effect of Acidity on Secondary
Organic Aerosol Formation from Isoprene, Environ. Sci. Technol., 41, 5363–5369, 10.1021/es0704176, 2007a.Surratt, J. D., Kroll, J. H., Kleindienst, T. E., Edney, E. O., Claeys, M.,
Sorooshian, A., Ng, N. L., Offenberg, J. H., Lewandowski, M., Jaoui, M.,
Flagan, R. C., and Seinfeld, J. H.: Evidence for Organosulfates in Secondary
Organic Aerosol, Environ. Sci. Technol., 41, 517–527, 10.1021/es062081q, 2007b.Surratt, J. D., Murphy, S. M., Kroll, J. H., Ng, N. L., Hildebrandt, L.,
Sorooshian, A., Szmigielski, R., Vermeylen, R., Maenhaut, W., Claeys, M.,
Flagan, R. C., and Seinfeld, J. H.: Chemical Composition of Secondary
Organic Aerosol Formed from the Photooxidation of Isoprene, J.
Phys. Chem. A, 110, 9665–9690, 10.1021/jp061734m, 2006.Surratt, J. D., Chan, A. W. H., Eddingsaas, N. C., Chan, M., Loza, C. L.,
Kwan, A. J., Hersey, S. P., Flagan, R. C., Wennberg, P. O., and Seinfeld, J.
H.: Reactive intermediates revealed in secondary organic aerosol formation
from isoprene, P. Natl. Acad. Sci. USA, 107, 6640, 10.1073/pnas.0911114107, 2010.Tomaz, S., Wang, D., Zabalegui, N., Li, D., Lamkaddam, H., Bachmeier, F.,
Vogel, A., Monge, M. E., Perrier, S., Baltensperger, U., George, C.,
Rissanen, M., Ehn, M., El Haddad, I., and Riva, M.: Structures and
reactivity of peroxy radicals and dimeric products revealed by online tandem
mass spectrometry, Nat. Commun., 12, 300, 10.1038/s41467-020-20532-2, 2021.Tröstl, J., Chuang, W. K., Gordon, H., Heinritzi, M., Yan, C., Molteni, U., Ahlm, L., Frege, C., Bianchi, F., Wagner, R., Simon, M., Lehtipalo, K., Williamson, C., Craven, J. S., Duplissy, J., Adamov, A., Almeida, J., Bernhammer, A.-K., Breitenlechner, M., Brilke, S., Dias, A., Ehrhart, S., Flagan, R. C., Franchin, A., Fuchs, C., Guida, R., Gysel, M., Hansel, A.,
Hoyle, C. R., Jokinen, T., Junninen, H., Kangasluoma, J., Keskinen, H., Kim, J., Krapf, M., Kürten, A., Laaksonen, A., Lawler, M., Leiminger, M., Mathot, S., Möhler, O., Nieminen, T., Onnela, A., Petäjä, T., Piel, F. M., Miettinen, P., Rissanen, M. P., Rondo, L., Sarnela, N.,
Schobesberger, S., Sengupta, K., Sipilä, M., Smith, J. N., Steiner, G., Tomè, A., Virtanen, A., Wagner, A. C., Weingartner, E., Wimmer, D., Winkler, P. M., Ye, P., Carslaw, K. S., Curtius, J., Dommen, J., Kirkby, J., Kulmala, M., Riipinen, I., Worsnop, D. R., Donahue, N. M., and Baltensperger, U.: The role of low-volatility organic compounds in initial particle growth in the atmosphere, Nature, 533, 527–531, 10.1038/nature18271, 2016.Voliotis, A., Wang, Y., Shao, Y., Du, M., Bannan, T. J., Percival, C. J., Pandis, S. N., Alfarra, M. R., and McFiggans, G.: Exploring the composition and volatility of secondary organic aerosols in mixed anthropogenic and biogenic precursor systems, Atmos. Chem. Phys., 21, 14251–14273, 10.5194/acp-21-14251-2021, 2021.Voliotis, A., Du, M., Wang, Y., Shao, Y., Bannan, T. J., Flynn, M., Pandis, S. N., Percival, C. J., Alfarra, M. R., and McFiggans, G.: The influence of the addition of a reactive low SOA yield VOC on the volatility of particles formed from photo-oxidation of anthropogenic – biogenic mixtures, Atmos. Chem. Phys. Discuss. [preprint], 10.5194/acp-2022-312, in review, 2022a.Volkamer, R., Ziemann, P. J., and Molina, M. J.: Secondary Organic Aerosol
Formation from Acetylene (C2H2): seed effect on SOA yields due to organic photochemistry in the aerosol aqueous phase, Atmos. Chem. Phys., 9, 1907–1928, 10.5194/acp-9-1907-2009, 2009.Voliotis, A., Du, M., Wang, Y., Shao, Y., Alfarra, M. R., Bannan, T. J., Hu, D., Pereira, K. L., Hamilton, J. F., Hallquist, M., Mentel, T. F., and McFiggans, G.: Chamber investigation of the formation and transformation of secondary organic aerosol in mixtures of biogenic and anthropogenic volatile organic compounds, Atmos. Chem. Phys. Discuss. [preprint], 10.5194/acp-2021-1080, in review, 2022b.Wang, K., Huang, R.-J., Brüggemann, M., Zhang, Y., Yang, L., Ni, H., Guo, J., Wang, M., Han, J., Bilde, M., Glasius, M., and Hoffmann, T.: Urban organic aerosol composition in eastern China differs from north to south: molecular insight from a liquid chromatography–mass spectrometry (Orbitrap) study, Atmos. Chem. Phys., 21, 9089–9104, 10.5194/acp-21-9089-2021, 2021.Wennberg, P. O., Bates, K. H., Crounse, J. D., Dodson, L. G., McVay, R. C.,
Mertens, L. A., Nguyen, T. B., Praske, E., Schwantes, R. H., Smarte, M. D.,
St Clair, J. M., Teng, A. P., Zhang, X., and Seinfeld, J. H.: Gas-Phase
Reactions of Isoprene and Its Major Oxidation Products, Chem. Rev.,
118, 3337–3390, 10.1021/acs.chemrev.7b00439, 2018.
WHO – World Health Organisation: Ambient air pollution: A global assessment of exposure and burden of disease, ISBN 9789241511353, 2016.Winterhalter, R., Van Dingenen, R., Larsen, B. R., Jensen, N. R., and Hjorth, J.: LC-MS analysis of aerosol particles from the oxidation of α-pinene by ozone and OH-radicals, Atmos. Chem. Phys. Discuss., 3, 1–39, 10.5194/acpd-3-1-2003, 2003.Xu, Z. N., Nie, W., Liu, Y. L., Sun, P., Huang, D. D., Yan, C., Krechmer,
J., Ye, P. L., Xu, Z., Qi, X. M., Zhu, C. J., Li, Y. Y., Wang, T. Y., Wang,
L., Huang, X., Tang, R. Z., Guo, S., Xiu, G. L., Fu, Q. Y., Worsnop, D.,
Chi, X. G., and Ding, A. J.: Multifunctional Products of Isoprene Oxidation
in Polluted Atmosphere and Their Contribution to SOA, Geophys. Res.
Lett., 48, e2020GL089276,
10.1029/2020GL089276, 2021.Yasmeen, F., Vermeylen, D., Maurin, N., Perraudin, E., Doussin, J.-F., and
Claeys, M.: Characterisation of tracers for aging of α-pinene secondary organic aerosol using liquid chromatography/negative ion electrospray ionisation mass spectrometry, Environ. Chem., 9, 236–246, 10.1071/EN11148, 2012.Yasmeen, F., Vermeylen, R., Szmigielski, R., Iinuma, Y., Böge, O., Herrmann, H., Maenhaut, W., and Claeys, M.: Terpenylic acid and related compounds: precursors for dimers in secondary organic aerosol from the ozonolysis of α- and β-pinene, Atmos. Chem. Phys., 10, 9383–9392, 10.5194/acp-10-9383-2010, 2010.
Zhang, Q., Jimenez, J. L., Canagaratna, M. R., Ulbrich, I. M., Ng, N. L.,
Worsnop, D. R., and Sun, Y.: Understanding atmospheric organic aerosols via
factor analysis of aerosol mass spectrometry: a review, Anal.
Bioanal. Chem., 401, 3045–3067, 2011.Zhao, Y., Thornton, J. A., and Pye, H. O. T.: Quantitative constraints on
autoxidation and dimer formation from direct probing of monoterpene-derived
peroxy radical chemistry, P. Natl. Acad. Sci. USA, 115, 12142-12147, 10.1073/pnas.1812147115, 2018.Ziemann, P. J. and Atkinson, R.: Kinetics, products, and mechanisms of
secondary organic aerosol formation, Chem. Soc. Rev., 41,
6582–6605, 10.1039/C2CS35122F, 2012.