Secondary organic aerosol (SOA) is an important constituent of the
atmosphere where SOA particles are formed chiefly by the condensation or
reactive uptake of oxidation products of volatile organic compounds (VOCs).
The mass yield in SOA particle formation, as well as the chemical
composition and volatility of the particles, is determined by the identity
of the VOC precursor(s) and the oxidation conditions they experience. In
this study, we used an oxidation flow reactor to generate biogenic SOA from
the oxidation of Scots pine emissions. Mass yields, chemical composition
and volatility of the SOA particles were characterized and compared with SOA
particles formed from oxidation of α-pinene and from a mixture of
acyclic–monocyclic sesquiterpenes (farnesenes and bisabolenes), which are
significant components of the Scots pine emissions. SOA mass yields for
Scots pine emissions dominated by farnesenes were lower than for α-pinene but higher than for the artificial mixture of farnesenes and
bisabolenes. The reduction in the SOA yield in the farnesene- and
bisabolene-dominated mixtures is due to exocyclic C=C bond scission in
these acyclic–monocyclic sesquiterpenes during ozonolysis leading to smaller
and generally more volatile products. SOA particles from the oxidation of
Scots pine emissions had similar or lower volatility than SOA particles
formed from either a single precursor or a simple mixture of VOCs. Applying
physical stress to the Scots pine plants increased their monoterpene, especially monocyclic β-phellandrene, emissions, which further decreased SOA particle volatility and increased SOA mass yield. Our results
highlight the need to account for the chemical complexity and structure of
real-world biogenic VOC emissions and stress-induced changes to plant
emissions when modelling SOA production and properties in the atmosphere.
These results emphasize that a simple increase or decrease in relative
monoterpene and sesquiterpene emissions should not be used as an indicator of
SOA particle volatility.
Introduction
Secondary organic aerosol (SOA) formed from oxidation of volatile organic
compounds (VOCs) comprises a large fraction of the total aerosol mass in
the boreal forests of the Northern Hemisphere (Hallquist et al., 2009;
Jimenez et al., 2009; Riipinen et al., 2012). The chemical transformation of
primary VOC emissions to SOA particles, which have an important climate
impact (Hallquist et al., 2009), is a complicated cascade of gas-phase
oxidation and multiphase ageing reactions. The physical properties of SOA
are dictated by the chemical complexity of the initial VOC emissions and the
oxidative conditions they experience (Glasius and Goldstein, 2016).
The formation and growth of SOA particles are often described by the
absorptive partitioning of organic vapours (e.g. terpenoid oxidation
products) between the gas and particle phase (Donahue et al., 2011; Pankow,
1994). The main property determining how readily organic molecules enter and
stay in the particle phase is their volatility, usually expressed as
saturation vapour pressure (Psat) or saturation mass concentration (C*) in air. The
volatility of a specific compound is in turn determined by both its molar
mass and functional group composition (Capouet and Müller, 2006; Pankow
and Asher, 2008). Oxidation of a single VOC precursor produces a wide
variety of semi- and low-volatility compounds, which are able to condense
into the particle phase according to their C*. The uptake of oxidation
products may also involve or be facilitated by heterogeneous reactions.
In boreal forest environments, VOC emissions are dominated by monoterpenes.
As the globally most important monoterpene (Spanke et al., 2001), α-pinene has been most commonly used as a model species in laboratory
SOA studies. Consequently, α-pinene has been used as a proxy for all
other monoterpenes in atmospheric models using its properties to describe
the atmospheric oxidation and contribution to SOA particles of all other
monoterpenes
(Holopainen
et al., 2017; Yassaa et al., 2012).
However, the VOC emission patterns of vegetation vary significantly
depending on locations, environmental conditions and even genotypes of each
plant. Moreover, differences in emissions can even be found between two
plants of the same species with the same age located in the same environment
(Bäck et al., 2012; Hakola et al., 2017; Holopainen and Gershenzon,
2010). Even though the VOC emissions from boreal forests are thought to
mostly consist of monoterpenes, Hellén et al. (2018) recently showed a
significant contribution of sesquiterpenes to the total VOC budget in boreal
forests. This finding is in line with an earlier study by Hakola et al. (2017) that showed high emissions of sesquiterpenes measured from branch
enclosures in a coniferous forest in Finland during early spring and late
autumn. Mixtures of different VOCs as well as different trace gases like
NOx, CO and sulfuric acid also influence the oxidant reactivity and
oxidant product distribution compared to single VOC precursor oxidation
(McFiggans et al.,
2019; Ng et al., 2017). These results highlight the need to use real
emissions as precursors while exploring the physical and chemical properties
of SOA particles.
In this study, we present results from a state-of-the-art suite of
instruments used to investigate the VOC emission profile of Scots pine
saplings (Pinus sylvestris) and subsequent SOA formation from these VOCs in
an oxidation flow reactor (OFR). The central instrument in this work is a
Filter Inlet for Gases and AEROsols, coupled to a high-resolution
time-of-flight chemical ionization mass spectrometer (FIGAERO-CIMS; Aerodyne Research,
Inc. and Tofwerk AG; Lopez-Hilfiker et al., 2014), which allowed us to
characterize the composition and thermal desorption behaviour of the SOA
particles. For comparison, we performed the same measurements for SOA formed
from α-pinene in the same OFR. Prompted by a strong contribution of
farnesenes to the emissions from our plants, we also examined SOA formation
from a mixture of acyclic–monocyclic sesquiterpenes with major contributions
from isomeric farnesenes and bisabolenes. This is the first volatility
measurement of SOA particles formed from the oxidation of farnesene and
bisabolene since earlier studies involving these sesquiterpenes have focused
on the gas-phase chemistry (Kim et al., 2011; Kourtchev et al., 2009, 2012)
or the chemical composition of the SOA particles (Jaoui et al., 2017). Our
results show surprising impacts of the VOC mixture on the SOA particle
volatility compared to SOA particles formed from a single VOC.
Materials and methodsVOC measurements and SOA production
We conducted experiments with SOA generated from the ozonolysis and
photo-oxidation of VOCs by hydroxyl radicals (OH) in a potential aerosol
mass (PAM) OFR (Kang et al., 2007; Lambe et al., 2011) in the absence of
seed particles. The experimental setup is similar to that in our previous
study (Buchholz et al., 2019). A schematic of the setup is shown in
Fig. 1, and all experimental conditions are listed
in Tables 1 and 2. We
provide a very brief description of the experimental setup here, and more
detailed information can be found in the Supplement (Sect. S2). A
flow containing 200 to 400 ppb of the investigated VOCs was mixed with an
O3-containing flow directly before entering the OFR. With two UV
lamps (254 nm), O3 was photolysed to O(1D) which reacted with
water vapour to produce OH. A wide range of OH exposure was achieved
by adjusting the voltage of 254 nm UV lamps in the OFR and changing the
O3 concentration. Overall, the integrated OH exposure in the OFR ranged
from approximately 6.6×1010 to 2.5×1012 molec. cm-3 s across all experiments as calculated according to methods
described by Peng et al. (2015, 2016). This range of OH exposure corresponds
to 0.5 to 19 equivalent days of atmospheric ageing at an OH concentration of
1.5×106 molec. cm-3 (Palm et al., 2016). In all
experiments, the operation temperature of the OFR was 25 or 27 ∘C
and relative humidity was between 40 % and 60 %. For the Scots
pine experiments, VOCs were introduced by flushing purified air through a
plant enclosure (Tedlar®) containing a 6-year-old Scots pine
sapling. In the α-pinene (Sigma-Aldrich, 98 % purity) and
sesquiterpene mix (mixture of acyclic–monocyclic sesquiterpenes, Sigma-Aldrich) experiments the VOCs were introduced into a flow of clean nitrogen
by using a diffusion source or a dynamic dilution system
(Kari et al., 2018). For Scots
pine experiment 4, the plant was injured by making four 0.5–1 cm2
cuts into the bark of the plant exposing resin pools and, thus, increasing
the VOC emissions.
Measurement setup used in the experiments. Abbreviations in the
picture are explained in the main text (Sect. 2), except for
ozone-containing air (O3) and humidified air (RH).
List of different Scots pine experiments with corresponding PAM 1
reactor conditions. Monoterpenes are referred to as MT and sesquiterpenes as
SQT. Collected mass is estimated from SMPS data assuming a particle density of
1.3 g cm-3.
ExperimentScots pine 1Scots pine 2Scots pine 3Scots pine 4VOC mixing ratio (ppb from PTR-ToF-MS)MT: 125±1 SQT: 179±1 Sum: 304±2MT: 213±14 SQT: 179±1 Sum: 391±14MT: 127±4 SQT: 151±1 Sum: 278±4MT: 153±3 SQT: 66±2 Sum: 218±3SQT / MT ratio (by molar ratio)1.430.851.20.4SQT / MT ratio (by mass ratio)2.151.261.780.65OH exposure (molecules cm-3s)6.41×10116.49×10116.45×10117.31×1011PAM residence time (s)300300300300O3 mixing ratio (ppm)4.9555.6Collected mass on FIGAERO filter (ng)1000120013001000
List of different α-pinene and sesquiterpene experiments
with corresponding PAM reactor conditions. Collected mass is estimated from
SMPS data assuming a particle density of 1.3 g cm-3.
PAM 1 PAM 2 Experimentα-pineneα-pineneα-pineneSesquiterpeneα-pinenelow exposuremedium exposurehigh exposuremixturereferenceVOC mixing ratio (ppb from PTR-ToF-MS)199±2198±2196±2327±5270±10OH exposure (molecules cm-3 s)2.54×10116.85×10112.45×10128.2×10102.6×1011PAM residence time (s)120120120160160O3 mixing ratio (ppm)6.6252513.313.2Collected mass (ng)9608801350550925
The VOC mixing ratios entering the OFR were continuously monitored using a
proton-transfer-reaction time-of-flight mass spectrometer (PTR-MS; PTR-TOF
8000, IONICON Analytik Inc.) directly upstream of the OFR inlet but before
the addition of O3 to the system. All reported mixing ratios were
corrected for this dilution and represent the conditions at the inlet of the
OFR. In addition, to resolve the mixture of terpenoid emissions emitted by
the Scots pine sapling, we collected two cartridge samples (Markes International, Inc.) at the beginning of Scots pine experiment 1 and at the end of
Scots pine experiment 4 for off-line analysis using a thermal desorption gas
chromatograph mass spectrometer (TD-GC-MS; TD – PerkinElmer, ATD 400, USA;
GC-MS – Hewlett-Packard, GC 6890, MSD 5973, USA). The PTR-MS calibration
procedure is described in Sect. S3.
The measured emission profiles from the Scots pine showed a strong
contribution of α- and β-farnesene (Fig. 2), which are
acyclic sesquiterpenes. Therefore, we conducted follow-up experiments under
similar oxidative conditions using a commercially available mixture of
acyclic–monocyclic sesquiterpenes to investigate the effect of such
biogenic, unsaturated, acyclic–monocyclic VOCs on SOA properties. This
sesquiterpene mixture consisted of isomers of farnesenes and bisabolene,
both of which are found in the emissions of coniferous trees as well as the
emissions of Scots pines in this study
(Blande et al., 2009;
Holopainen and Gershenzon, 2010). For a detailed description of the mixture
see Tables S1 and Fig. S1 in the Supplement. For those follow-up experiments, a nominally
identical OFR was used (PAM 2). However, to recreate the same OH
exposure and particle composition (as characterized by the particle oxidation
ratio; see below), a different combination of light intensity, residence
time, and O3 concentration was necessary in the follow-up experiments.
Thus, the results are presented separately and marked PAM 1 or PAM 2.
Panel (a) shows the structures of the most abundant monoterpenes
and sesquiterpenes in Scots pine emissions, as measured with a TD-GC-MS.
Relative mass concentrations of each compound are shown for Scots pine experiments 1 (b) and 4 (c), the latter following
deliberate damage to the plant's stem. The whiskers show the standard
deviation of the measurements.
SOA particles characterization
SOA particles were examined with a suite of instruments sampling from the
outlet of the OFR: a high-resolution time-of-flight aerosol mass
spectrometer (AMS; Aerodyne Research, Inc.), a scanning mobility particle
sizer (SMPS; TSI Inc., Model 3082 with TSI Model 3775 condensation particle
counter) and a FIGAERO-CIMS using the iodide ionization scheme
(Lee et al., 2014). The AMS and SMPS were
used to continuously monitor the output SOA particle mass and size
distribution from the OFR to determine the point when the particle
concentrations and distributions had stabilized for a given OFR condition.
Then the filter collection for FIGAERO-CIMS was started so that only
steady-state SOA was sampled. More information about these other instruments
is given in Sect. S4.
The FIGAERO-CIMS was used to characterize the volatility and chemical
composition of the SOA particles for those organic compounds sensitive to
iodide cluster ionization. Briefly, in the ionization region of the
instrument, operated at 100 mbar pressure, an I- anion preferably
clusters with a neutral molecule M which contains hydroxy, hydroperoxy,
carboxyl or peroxycarboxyl groups in their structure. The neutral molecule
is then observed as [M+I]- in the mass spectrometer
(Iyer et al.,
2017; Lee et al., 2014). In the presence of water, collision of an
[H2O+I]- anion cluster with M may produce the same result. In some
cases, the [M+I]- cluster breaks apart leading to deprotonation of the
neutral molecule, which is then observed as [M-H]-, and possibly to other ion
fragments. This phenomenon has also been observed in earlier studies (e.g.
Lee et al., 2014). For the further analysis, we assume that deprotonation is
the only mechanism of declustering, as it is one known to potentially follow
from even relatively soft collisions. Additional declustering may happen
by more energetic collisions in the lower-pressure regions of the mass
spectrometer (Passananti et al., 2019).
In the FIGAERO inlet, the aerosol particles were sampled through 2 m
stainless steel tubing (outer diameter 6 mm) onto a Teflon filter (Zefluor 2 µm polytetrafluoroethylene (PTFE) membrane filter, Pall Corp.) for 2–5 min with a collection
flow of 2 L min-1. The particles were then evaporated into the
instrument with a gradually heated nitrogen flow with a heating rate of 11.6 K min-1 heated up to 200 ∘C over a period of 20 min and then kept at
200 ∘C for an additional 10 min to evaporate any residual compounds.
This results in temperature-dependent ion signals for each observed mass
spectrum peak (thermograms) that can be related to the volatility of the
collected organic compounds (Lopez-Hilfiker et al., 2014, 2015). In this study, two slightly different FIGAERO inlets were
used: one in conjunction with the initial experiments using the PAM 1
reactor (FIGAERO 1) and the other when using the PAM 2 reactor later on
(FIGAERO 2). The CIMSs themselves were nominally identical, and both of
the FIGAERO inlets followed the identical principles and were operated
identically. The differences were in the detailed design of the FIGAERO
inlets, e.g. the shape of the filter collection tray and exact positioning of the
temperature sensor. These changes led to apparent shifts in the measured
thermograms. To account for this, we performed instrument-specific
calibrations for both FIGAERO inlets, which are described in more detail in
Sect. 2.2.2.
A blank measurement, meaning a measurement with no particles collected on
the filter, was performed before each measurement to make sure the filter
was clear of residual compounds and to determine any instrument artefacts.
To ensure that the filter did not contain any particles, collection flow leading
to the filter was shut down between actual measurements. These blank
measurements were also considered in the data analysis. The relatively high
collected particle mass loading (between 500 and 1350 ng) on the filter
ensured that the majority of the signal came from the evaporating SOA
particles and that the instrument background and artefacts were neglectable.
The FIGAERO filter was also visually inspected daily and replaced when
needed.
FIGAERO-CIMS data analysis
The FIGAERO-CIMS data were analysed with the MATLAB-based tofTools (Junninen
et al., 2010) software, including the identification of elemental
compositions (creation of peak lists) based on the high-resolution (HR) peak
fitting with mass accuracy of 5 ppm. All data shown here are based on the
HR fitting of the mass spectra using peak lists covering the full spectra.
The presented mass values are those of the neutral composition (M), derived
by subtracting the molecular mass of I- when [M+I]- was observed or
by adding the molecular mass of H when [M-H]- was observed. The assigned
formulas were constrained to contain only C, H and O elements; any signal
peak appearing to contain other elements was considered background and
excluded from further analysis. This background consisted mostly of fluorine-containing compounds, which we assume to originate from the FIGAERO inlet
manifold or the collection filter, both of which are made of PTFE.
Tmax-to-saturation-concentration calculations
Typically, the thermograms collected from the heat-induced SOA evaporation
in the FIGAERO filter comprise a clearly defined peak, i.e. a temperature
at which the largest signal is observed (Tmax). It has been shown that
Tmax correlates with the volatility of the desorbing organic compound,
typically expressed as saturation concentration C*, at least for systems with a
limited number of compounds
(Bannan
et al., 2019; Lopez-Hilfiker et al., 2014). We determined the relationship
between Tmax and C* for both FIGAERO inlets independently, based on the
results of calibration experiments using a series of polyethylene glycols
(PEGs) with known saturation pressures Psat (Pa) at reference
temperature of 298.15 K
(Bannan et al., 2019;
Krieger et al., 2018). All reported C* and Psat values are thus contrasted
with this temperature. The calibration method and resulting parameters are
described in the Supplement Sect. S5.
(a) SOA yield vs. condensed organic mass (Coa) from Scots pine
and α-pinene experiments. Each point represents a single SMPS scan. In these
cases, error bars are omitted for clarity. The colour scale of the Scots
pine experiment results corresponds to the sesquiterpene-to-monoterpene
(SQT / MT) ratio by mass ratio (crosses; showing all yield measurements,
reflecting the variability in plant emission rates). Each α-pinene
experiment is averaged to one point (black triangles) and labelled. Error
bars shown are standard deviations of multiple scans. The orange curve is
fit with Eq. (2) to the Scots pine experiments 1–3 (SQT / MT >1) with the following fit parameters: α1=0.3783, K1=0.0035, α2=0.0029 and K2=0.0035. (b)α-pinene reference and sesquiterpene
mixture SOA yields versus COA and Odum fits to both datasets. The fit
parameters are α1=0.3219 and K1=0.0195 for α-pinene and α1=0.1164 and K1=0.0062 for
the sesquiterpene mixture. Circles
show the points of FIGAERO measurements. Titles PAM 1 (a) and
PAM 2 (b) refer to different PAM reactors used in the experiments.
SOA yield calculations and theoretical yields
From the amount of consumed precursor VOC (ΔVOC) and the condensed
(i.e. particulate) organic mass (ΔCOA) one can calculate the
effective SOA mass yield Y by
Y=ΔCOAΔVOCMT+SQT.
The subscripts here refer to monoterpenes (MT) and sesquiterpenes (SQT),
which were the dominant contributors to SOA formation in these experiments.
The ΔVOCMT+SQT (µgm-3) was calculated from the
difference in the total concentrations of monoterpenes and sesquiterpenes
entering the OFR and the VOC concentration exiting the OFR, which was
quantified by the PTR-MS. Measured mixing ratios of both MT and SQT
exiting the OFR were <1 ppb. The condensed organic mass ΔCOA (µgm-3) was monitored using the SMPS number size
distribution and assuming a particle density of 1.3 g cm-3 for all
measurements (Faiola et al.,
2018). No wall-loss correction was applied to the VOC and particle
measurements. However, because the VOC measurement was made almost
immediately before the inlet of the OFR and formed SOA size distribution was
roughly identical between different experiments, we assume that possible
sampling line wall losses have a minor impact on our SOA yield calculations.
The SOA yield can also be expressed using the approach of
Odum et al. (1996), which allows for
formally breaking down SOA formation into the contribution of several VOC
species:
Y=ΔCOA∑iαiKi1+KiΔCOA,
where αi is a proportionality constant relating the amount of
reacted VOC (precursor) to the total concentration of species i (oxidation
product) and Ki is the partitioning coefficient for species i. We
use a two-product version of Eq. (2), which is common in modelling
applications.
Scots pine sapling treatment
Due to scheduling constraints, the experiments were conducted in early
autumn outside of the most active emission time for Scots pines. To
circumvent this, a Scots pine sapling (6 years old) was stored in a cold
room throughout the summer to maintain winter dormancy. It was removed
from the cold room ∼1 month before experiments to initiate
spring phenology and metabolism – the time of year when Scots pines are
active with new shoot growth. At least 24 h before the SOA experiments,
the sapling was transported to the laboratory and a dynamic Teflon plant
enclosure was installed around the plant foliage. The Teflon enclosure
(custom-built; Jensen Inert Products, Inc.) was secured to the plant stem
with two cable ties. The plant enclosure was flushed with 5 L min-1
purified compressed air and operated under positive pressure to push
enclosure air into the flow reactor through perfluoroalkoxy alkane tubing. The flow of plant
enclosure air into the OFR was regularly measured in-line and maintained at
1.7–1.9 L min-1. Three LED lamps were placed around the plant to
provide photosynthetically active radiation (PAR) similar to ambient levels
in Finland.
Results and discussionScots pine emission patterns
The VOC emissions from the Scots pine sapling were mainly composed of
monoterpenes and sesquiterpenes. The emissions were monitored on-line with
the PTR-MS and off-line with the TD-GC-MS, the latter allowing us to better speciate
the VOCs by distinguishing between isomers. In Fig. 2 we show the chemical
structures of the most abundant compounds (see Table S2 for full list),
along with the relative concentrations of the individual monoterpenes and
sesquiterpenes, as measured by the TD-GC-MS.
Among the monoterpenes, β-phellandrene had the highest concentration,
followed by 3-carene, d-limonene, and β- and α-pinene, in that
order. These five monoterpenes accounted for 90 % of all monoterpenes in
all experiments. Among the sesquiterpenes, β- and α-farnesene and α-bisabolene were the most abundant species,
accounting for about 95 % of all sesquiterpenes. All sesquiterpenes
combined accounted for 55 %–70 % of the total VOC mass concentration
in the Scots pine experiments 1–3 and for 40 % in Scots pine
experiment 4, based on PTR-MS measurements. Relating these results to the
ambient pine emission characterizations by
Bäck et al. (2012), our Scots pine sapling may be classified as a 3-carene
chemotype, due to the higher emission of 3-carene. The fraction of
sesquiterpene emissions in our study was considerably higher than expected
from the ambient pine emission measurements of the SQT / MT ratio of typically
∼0.1 by mixing ratio
(Hellén et al., 2018).
The used Scots pine sapling was infested with herbivores, creating biotic
stress for the plant and changing the VOC emission pattern. This type of biotic
stress is natural, especially in a changing climate where insect outbreaks
are predicted to become more frequent (Bale et al.,
2002; Jactel et al., 2019).
During Scots pine experiments 1–3, the VOC levels slowly decreased
over time. Therefore, after the end of Scots pine experiment 3, we made
four 0.5–1 cm2 cuts to the Scots pine sapling's stem, with the goal of
increasing VOC emissions to continue the production of sufficient SOA. The
wounds exposed plant resin pools in the stem and increased the measured
monoterpene concentrations compared to pre-cutting concentrations by roughly
doubling them. Sesquiterpene concentrations, however, were not significantly
affected. The compound that increased most was β-phellandrene,
emissions of which had also been shown to increase with bark beetle
infestation damage
(Amin
et al., 2012; Faiola et al., 2018). Such damage essentially consists of cuts
into the stem as the bark beetle feeds on the plant; hence that observation
is consistent with expectations.
SOA mass yields
In Fig. 3, we plot SOA yields against condensed organic mass COA from
each experiment, split into two panels according to which of the two PAM
reactors was used (Fig. 3a, PAM 1; Fig. 3b, PAM 2). Note that a
similar organic mass range was covered in all experiments.
Scots pine experiments 1–3 featured sesquiterpene-to-monoterpene ratios
(SQT / MT) by mass ratio between 1.2 and 2 (colour-coding in Fig. 3a). The SOA
yields obtained during these experiments are consistent with each other in the
sense that all the data points are well described by the same two-product
model (orange line). The SOA yields resulting from Scots pine experiment 4
are about 30 % larger, even though SQT / MT was substantially smaller than
in experiments 1–3. An earlier study by Faiola et al. (2018)
reported a positive correlation with SOA yield and the SQT / MT ratio of VOCs
measured from Scots pine seedlings in a flow tube experiment similar to our
study.
To understand the unexpected suppression of SOA yield at higher relative concentrations of SQT, we also measured SOA yields for a synthetic sesquiterpene
mixture containing isomers of farnesene and bisabolene. A different PAM
reactor had to be used; therefore we compare those results to a reference
α-pinene experiment using the same reactor (Fig. 3b). The ingoing VOC
concentration was varied for both precursors, and yield measurements were
conducted in order to cover the COA range of the Scots pine
experiments. A single FIGAERO measurement was made for both precursors, and
the corresponding sampling conditions are marked with circles in Fig. 3b.
The comparison shows a clearly lower yield for the sesquiterpene mixture
than for α-pinene, which is in line with results shown in Fig. 3a.
Note that the medium- and low-exposure α-pinene measurements in PAM 1
show similar yield values to those in the reference experiment in PAM 2, but
the yield of the high-exposure experiment is significantly lower. This is
consistent with the extensive fragmentation expected inside the PAM reactor
under strong oxidative conditions, which reduces the effective SOA yield
(Lambe et al.,
2012).
We conclude that the increase in SOA yield in Scots pine experiment 4,
compared to Scots pine experiments 1–3, is likely due to the large
relative increase in emitted monoterpenes, especially β-phellandrene,
caused by cutting the sapling (Fig. 2). We surmise that β-phellandrene must have a high SOA yield, possibly comparable to that of
d-limonene, which has been reported in the range of 50 %–60 %
(Berg
et al., 2013; Lee et al., 2006; Surratt et al., 2008) and which shares some
structural similarity with β-phellandrene (Fig. 2).
Mackenzie-Rae et al. (2017) measured SOA yields of α-phellandrene, an isomer of β-phellandrene with an endocyclic C=C bond, and found the SOA yield to be
around twice that of α-pinene, reaching up to 100 %. While those
yield numbers are not directly comparable to our results, they qualitatively
indicate that SOA yields from monocyclic monoterpenes, with endocyclic C=C
bonds, could be higher than those from the more commonly studied bicyclic
monoterpenes, such as α-pinene.
It is instructive to compare our SOA yield results to results of earlier
experiments that used (stressed) Scots pines. Faiola et al. (2018) studied
SOA formation from emissions of herbivore-stressed Scots pines in a
custom-made OFR where the main SQT type was β-caryophyllene. There the
increasing SOA yields with increasing SQT contribution to the precursor mix
were explained by the much higher SOA yield of β-caryophyllene
(Faiola et al., 2018). In
chamber studies with emissions of aphid-stressed Scots pines in which farnesenes
were the dominant SQT species, SOA yields decreased with increasing SQT
contribution for ozonolysis reaction while no change was observed for pure
OH reaction experiments (Faiola et al., 2019). In our
experiments in the PAM reactor, the ratio between O3 and OH exposure
was in the range of 105 which is comparable to ambient levels (Kang et
al., 2007). Due to their very fast reaction with O3, more than 80 % of
any SQT are expected to react with O3 under these conditions (calculated
with methods described by
Peng
et al., 2015, 2016). Thus, the SOA yield should decrease with increasing
amounts of farnesene as described for the ozonolysis reaction pathway in
Faiola et al. (2019).
The reason for this different behaviour of the two SQT types is based on
their molecular structure. β-caryophyllene is a bicyclic compound
with one endo- and one exocyclic C=C bond, whereas farnesene isomers are
acyclic compounds with four C=C bonds. In the (photo-)oxidation process, both
ozone and OH break the C=C bonds and thus the carbon backbone of
farnesene into small fragments. Just a single oxidation step decreases
the number of carbons from 15 to 5–12
(Kourtchev
et al., 2009, 2012). In the case of the bicyclic β-caryophyllene, such
fragmentation is expected to be much less prominent
(Jaoui et al., 2003, 2013).
There may be further interactions between the small, most likely volatile,
farnesene reaction products and the other oxidation products, suppressing the
particle formation further as recently described by
McFiggans et al. (2019).
SOA composition
Figure 4 shows mass spectra integrated over the heating period of
FIGAERO-CIMS measurements normalized to the maximum signal and molecular
mass adjusted to neutral compositions (Sect. 2.2.1). The results are grouped
into two portions as two different PAM reactors and FIGAERO inlets were used
in the measurements (Sect. 2.2). The α-pinene measurements (Fig. 4, top row)
show a decrease in the average molecular weight with increasing OH exposure,
which is consistent with the decrease in SOA mass yields as a function of OH
exposure observed in Fig. 3 and also observed in an earlier study
(Hall et al., 2013).
Mass spectra integrated over the whole FIGAERO desorption cycle
for each experiment. Each spectrum is normalized to the peak height of the
most abundant ion.
The distribution of molecular weights of compounds in the sesquiterpene
mixture SOA particles (Fig. 4, right) shifted more towards smaller molecular
masses than it did in any other experiment. We explain this by the acyclic
molecular structure of the sesquiterpene compounds leading to their more
efficient splitting into smaller products (fragmentation) during oxidation
(see Sect. 3.2). The α-pinene reference spectra differ slightly from other
α-pinene measurements, even though OH exposure falls between medium and high
exposure (Table 2).
The difference might be due to the different residence time inside the PAM
reactor (120 vs. 160 s) or to the higher mixing ratio of VOCs in the
experiment performed with the PAM 2 reactor. The FIGAERO mass spectra distribution
of SOA formed from the Scots pine emissions had the most similarities to
high-exposure α-pinene SOA, with almost all compounds appearing within a
single mode, roughly centred on the monomer mode of the other α-pinene experiments.
Table 3 shows the average carbon oxidation states
(OSC; Kroll et al., 2011) and the
average O:C ratio, calculated from both AMS and FIGAERO data. The average
chemical composition is also calculated from FIGAERO data. All FIGAERO data
values are weighted averages, using the integrated signal strength of each
ion thermogram as weights. The range of values calculated from the FIGAERO
data represents the spread of different compounds in the mass spectrum,
which we will investigate below.
Average O:C ratio, OSC and chemical composition from each
experiment calculated from AMS and FIGAERO data. Values calculated from
FIGAERO are weighted by the integrated signal strength. All FIGAERO data are
shown with standard deviation to highlight the spread of different
compositions in the mass spectrum.
Experiment〈O:C〉 AMS〈O:C〉FIGAERO〈OSC〉AMS〈OSC〉FIGAERO〈CxHyOz〉α-pinene low exposure0.530.65±0.28-0.46-0.3±0.56C9±4H14.2±4O5.4±4α-pinene medium exposure0.690.75±0.3-0.050.02±0.65C8.3±3.5H12.1±3.5O5.7±3.5α-pinene high exposure0.960.9±0.330.630.47±0.73C7.5±3H9.9±4.5O6±1.9Scots pine experiment 10.850.79±0.430.370.02±1.1C8.5±4H13.6±8O5.4±2Scots pine experiment 20.890.77±0.420.43-0.04±1.1C8.4±4H13.5±8.3O5.4±2.1Scots pine experiment 30.90.79±0.430.550.02±1.1C8.7±4H14±7.9O5.4±2.1Scots pine experiment 410.81±0.430.750.06±1.1C8.2±3.7H13.1±7.5O5.3±2Sesquiterpene mixture0.820.89±0.370.260.47±0.83C7.5±3.7H9.7±5.1O5.6±1.8α-pinene reference0.770.82±0.330.060.28±0.71C7.4±3H10±4.4O5.3±1.5
In the α-pinene experiments with PAM 1, the average O:C ratio and OSC
increase with the increasing oxidative strength as expected while the
average carbon chain length decreases. In terms of the average O:C ratio or OSC,
the Scots pine experiments most closely correspond to the medium exposure
α-pinene experiment. Interestingly, all Scots pine experiments appear
broadly similar from this viewpoint, even though experiment 4 is associated
with emissions dominated by monoterpenes (i.e. much smaller SQT / MT ratio) and
clearly higher SOA yields (Figs. 2 and 3a). However, the Scots pine SOA
particles differ from each other and from other experiments in other ways,
as we will see in Sect. 3.4.
SOA volatility
In Fig. 5, we show normalized “sum” thermograms, calculated by summing
up the thermograms from all ions that contain only C, H and O atoms. With
these sum thermograms, the average thermal desorption behaviour can be
compared across multiple experiments. While viewing the thermograms, it is
important to know that the thermal decomposition during the aerosol
desorption from the FIGAERO filter often manifests as signal at relatively
high desorption temperatures, which may appear as shoulders to the main
peak, but it can appear as simple peaks as well
(e.g.
Lopez-Hilfiker et al., 2015; Schobesberger et al., 2018).
Normalized sum thermograms (all observed ions containing C, H and
O atoms) from (a)α-pinene experiments, (b) Scots pine experiments 1–4 and (c) sesquiterpene mixture experiment together with reference α-pinene experiment. Tmax values of the sum thermograms are shown with
dashed lines. In panel (b) only the Tmax values of Scots pine experiment 1
and Scots pine experiment 4 are shown for clarity.
For the α-pinene experiments, the sum thermograms reveal a clear
increase in Tmax (dashed vertical lines in Fig. 5a) with increasing
oxidative exposure. As we have shown above, this change is concurrent with a
decrease in molecular size and an increase in average O:C-ratio (Fig. 4 and
Table 3). Together, these observations imply that α-pinene
photo-oxidation successively forms compounds with lower volatility.
Evidently, the oxidation reactions are both fragmenting, which generally
increases volatility, as well as functionalizing, which generally decreases
volatility
(Capouet
and Müller, 2006; Pankow and Asher, 2008). The clearly increasing
Tmax values observed with FIGAERO suggests a net decrease in volatility
due to these processes overall, for the condensed-phase constituents,
consistent with the results of isothermal evaporation experiments performed
using the same (but size-selected) aerosol (Buchholz et al.,
2019).
The sum thermograms for the Scots pine experiments gradually shift towards
yet higher desorption temperatures (Fig. 5b); all their Tmax values are
higher than those of any of the α-pinene experiments (Fig. 5a). In
particular, SOA particles from Scots pine experiment 4 are the most
resistant to thermal desorption. That experiment was also the only one with
plant emissions clearly dominated by monoterpenes (Fig. 2), specifically
with β-phellandrene being the most abundant, which we associated
above with relatively high observed SOA yields (Sect. 3.1). That specific
VOC mix in Scots pine experiment 4 is unique within our study here. It
is plausible that this mix is also directly responsible for producing SOA
with the lowest effective C* as suggested by the results of our FIGAERO
measurements (Fig. 5, as well as our more detailed discussion below).
We suggest that the increased desorption temperatures for the Scots pine
experiments 1–3 relative to the α-pinene experiments are due to the
large contribution of acyclic sesquiterpenes (in particular farnesenes, Fig. 2) to the plant emissions in those experiments. We tested this hypothesis
via our follow-up experiments using the PAM 2 reactor and FIGAERO 2 (Fig. 5c). These experiments yielded sum thermograms for SOA particles formed from
the farnesene-dominated sesquiterpene mixture, and those from α-pinene using the same setup, for reference. The reference α-pinene
SOA particles had a 〈O:C〉AMS of 0.77, similar to
those in the medium- and high-exposure α-pinene case. Indeed, the sum
thermogram Tmax value for the sesquiterpene mix case is about 10 ∘C higher than for the reference α-pinene case,
confirming that the strong contribution of farnesene and sesquiterpenes with
similar structures leads to the effectively lower-volatility aerosol
particles (as measured by FIGAERO) in the Scots pine experiments 1–3
compared to α-pinene SOA particles. As mentioned earlier (Sect. 2.2), a quantitative comparison of thermograms between Fig. 5c and a–b, including comparison of Tmax values, is not straightforward, due
to the differences in the respective experimental setups. However, we will
deal with this issue below.
We provide a more extended discussion of our examination of SOA
volatilities, which includes graphical depictions of the individual ion
thermograms, in the Supplement (Sect. S6). When looking at
individual experiments, the Tmax value for each ion broadly depends on
the molecular weight of the ion, as expected. However this dependency seems
to differ in scale between experiments. Some signatures of thermal
decomposition are visible as well, but overall this appears to play a minor
role, with very small effects on the thermogram in most individual ion
cases.
The resistance to thermal desorption at each unit mass appears to increase
with increasing strength of oxidation in the α-pinene experiments,
as is observed in the sum thermograms (Fig. 5), while the contribution of
thermal decomposition appears to increase concurrently. The Scots pine
experiments show similar effects between Scots pine experiments 1–4, but the
change is not as pronounced as with α-pinene experiments and cannot
be as clearly attributed to a single factor such as oxidative strength.
Previous studies have indicated that the Tmax value of individual
thermograms largely remains controlled by the C* of the respective compound,
even when a substantial fraction of the signal is the result of thermal
decomposition of different, larger structures
(e.g. Schobesberger et al., 2018).
The reason is that this decomposition typically occurs only at temperatures sufficiently
higher than the desorption temperatures for most compounds.
Measured Tmax values can therefore be used as a fairly robust estimation
of C* of the respective composition. Thus, we performed calibration
experiments, in order to establish the Tmax–C* relationship for both
FIGAERO inlets used in this study (see Sect. S5), and accordingly derived
C* for each measured organic composition from its respective Tmax value.
Note that in cases where thermogram peaks are affected by thermal
decomposition (which we determined play a relatively minor role in this
study), we implicitly assign upper-limit C* values. We also note that
relatively high collected aerosol mass (on the order of 1 µg) might
induce so-called matrix effects in the evaporation process
(Huang et al., 2018), which might in turn shift the observed
Tmax to higher temperatures and could consequently lead to a slight
systematic underestimation of C*.
The results from the Tmax to C* conversion are shown in Figs. 6 and 7,
summarizing them as volatility basis set (VBS) bins of 1 order of
magnitude of C*, as defined by Donahue et al. (2011; SVOC is semi-volatile organic compound, green volatility range; LVOC is low-volatility organic compound, red volatility range; ELVOC is extremely
low volatility organic compound, grey volatility range). These VBS
distributions represent only the compounds found in the particle phase. Note
that having translated from Tmax to C* using the instrument-specific
calibration parameters, all results become comparable to each other. Also
note that the abundance of SVOCs might generally be underestimated
due to the fast evaporation of particulate matter during collection to the
FIGAERO filter and switching from collection phase to desorption phase.
VBS bins and corresponding normalized integrated mass spectra from
α-pinene low, medium and high experiment; sesquiterpene mixture; and
α-pinene reference experiment. Top row: VBS bins determined from
Tmax with background colours corresponding to different volatility
classes (defined in the text). Bottom row: integrated FIGAERO signal
normalized to maximum signal and coloured by corresponding volatility class.
LVOC-class compounds (red) are plotted on an inverted y axis to more easily distinguish them from other volatility classes.
A clear shift from higher to lower volatility for α-pinene SOA is
observed with increasing OH exposure (Fig. 6). This shift manifests itself
in the reduction in SVOC while the amount of LVOC
increases. For sesquiterpene mixture SOA particles, a major part of the
observed compounds falls into the LVOC class, while SVOC-class compounds are
almost non-existent. This is also seen in the α-pinene reference
experiment, where most of the compounds are attributed to a single VBS bin.
Note also the different y scale in the α-pinene reference results
compared to other results. A small number of signals in the sesquiterpene
mixture results can confidently be categorized as thermal decomposition
products, namely the ones that fall into the ELVOC volatility range but have
relatively small molecular masses (<200 Da). These compounds are
C3–C7 compounds with relatively high oxygen content (O3–O7) and comprise
about 11 % of the total integrated signal.
In the Scots pine results (Fig. 7), a similar shift from higher to lower
volatilities can be seen, with Scots pine experiment 4 having the lowest
volatility overall. Differences in the Scots pine VBSs are probably due to
evolving VOC emissions from the sapling. Note that simple SQT / MT ratio
calculated from the total concentration of monoterpenes and sesquiterpenes (Table 1;
insets in Fig. 7, top row) cannot explain the change in the volatility.
VBS bins and corresponding normalized integrated mass spectra in
the Scots pine experiments. Top row: VBS bins determined from Tmax with
background colours corresponding to different volatility classes. Bottom
row: integrated FIGAERO signal normalized to maximum signal coloured by
corresponding volatility class. LVOC class is plotted on reverse y axis to more easily distinguish them from other volatility classes. Text in the upper row
panels shows measured amounts of monoterpenes and sesquiterpenes and
calculated SQT / MT ratio per mass ratio.
Number of oxygen molecules versus number of carbon molecules for
each compound and each experiment. Colours correspond to the saturation
ratio C* as derived from the measured Tmax. The size of a marker
corresponds to the signal of the compound. The two black lines in the
figures correspond to O:C=0.5 and O:C=1 ratios, for reference.
Finally, we examine the compositional information present in the FIGAERO
datasets. In Fig. 8, we plot the carbon numbers versus oxygen numbers for
each compound for each experiment, coloured according to the
Tmax-derived saturation concentration (C*). The panels reveal the spread
of compounds across the ranges of O:C ratios. In the α-pinene and
sesquiterpene mixture experiments, the compounds fall roughly into similar
areas (i.e. range of O:C ratios is 0.5–1.0), but in the Scots pine
experiments the spread across different O:C ratios is notably wider.
Specifically, there is an increased contribution of both relatively small
(#C ∼5) but highly oxygenated compounds and larger (#C
>10) though less oxygenated compounds (O:C<0.5),
both with relatively low effective volatility. This suggests that the bulk
O:C ratio is insufficient for comparing the expected properties of SOA
particles generated from mixtures of different VOC precursors (e.g. MT and
SQT; cf. average O:C and OSC values with standard deviations in Table 3).
Partially, the profoundly different spread of O:C ratios in the Scots pine
experiment results might also be due to more extensive thermal fragmentation
in the FIGAERO desorption process.
Summary and conclusions
In this study, we compared the physicochemical properties of SOA particles
generated from combined ozonolysis and photo-oxidation of (1) α-pinene, (2) a complex mixture of VOCs emitted from Scots pine saplings,
and (3) a mixture of farnesenes and bisabolenes which were observed to
comprise a significant fraction of the Scots pine emissions. Our
measurements examined the SOA mass yield, as well as the chemical
composition and the thermal desorption properties of SOA particulate
constituents to assess their volatility. In general, we found that all of
these quantities or properties are substantially controlled by the oxidative
strength, as expected, but crucially also by the identity of the isomers
making up the precursor VOC mixture.
The SOA mass yield from Scots pine emissions was in general lower than the
SOA yield from α-pinene. The notable exception is a single
experiment (Scots pine experiment 4) in which plant emissions were not
dominated by farnesene and sesquiterpenes with similar structure but by
β-phellandrene and other monoterpenes. That experiment had the
highest SOA yields among the Scots pine experiments, which is consistent
with the high yields for the specifically involved monoterpenes reported or
suggested in the literature (e.g. Faiola et al., 2018). For the other Scots
pine experiments (experiments 1–3), we attribute the low observed SOA yields
to the substantial contribution of acyclic sesquiterpenes, particularly
β- and α-farnesene, to the total terpenoid plant emissions in
those cases, ranging from 40 % to 70 % by mass. This is supported by our
additional experiments that resulted in a much lower SOA yield from a
mixture of monocyclic and acyclic sesquiterpenes than from α-pinene. The
fragmentation of acyclic sesquiterpenes likely results in a product
distribution containing a smaller amount of organic material with
sufficiently low volatility to partition from the gas to particle phase. These
acyclic sesquiterpenes might also go through multiple cycles of
auto-oxidation following the reaction with OH, before suffering substantial
fragmentation (Bianchi et al., 2019), which would explain
the relatively high (>5) number of oxygen atoms in a major
fraction of the products comprising SOA.
Indeed, our thermal desorption results indicate that the oxidation of
acyclic and monocyclic sesquiterpenes forms a substantial number of
relatively small compounds, compared to α-pinene (Fig. 4). The
average molecular weight of particle-phase α-pinene oxidation
products decreases with increasing OH exposure (Table 3), while their O:C ratio increases. At the same time, the thermal
desorption temperature of the SOA particulate constituents increases (Figs. 5–6), indicating a decrease in the effective SOA particle volatility.
Interestingly, our measurements showed that our Scots pine and sesquiterpene
SOA particles were of lower volatility than any of the α-pinene SOA
particles (even at higher oxidation exposure and comparable O:C and OSC
values). This result is indicated by both sum thermograms and supported by
their derived VBS distributions for the individual SOA particulate
constituents. It is also worth noting that, for the Scots pine experiments,
the lowest-volatility SOA is formed in the experiment resulting in the
highest SOA yields (Scots pine experiment 4), contrary to the observations
made for the α-pinene experiments.
The molecular composition and thermal desorption behaviour of SOA particles
observed with the FIGAERO instruments in experiment 4 were strikingly
similar to the other, sesquiterpene-dominated, Scots pine experiments 1–3.
It appears that a certain contribution of those acyclic sesquiterpenes was
sufficient to lower SOA volatility, whereas the monoterpenes that dominated
the mixture led to the efficient formation of SOA to start with.
Interestingly though, the monoterpenes did not seem to directly affect the
volatility of Scots pine SOA, at least for our experiments here, even though
their relative contribution to the precursor VOC mixture varied
substantially. Further experiments are clearly warranted to explore this
suggestion for a wider range of conditions and precursor mixtures than
covered by this study.
In conclusion, our results highlight the need to know the structural
identity of mixtures of VOCs, as typically encountered in real atmospheric
conditions, if one endeavours to make an accurate prediction of SOA yields and SOA
particle properties. In particular the emissions of sesquiterpenes need to
be considered more carefully in current atmospheric models. Importantly,
their effects on SOA yields can be both enhancing and suppressing, depending
on the isomers involved. Specifically, depending on which sesquiterpene
isomers are involved, the product distributions obtained from their
oxidation can differ substantially from each other in terms of the products'
volatility and of the subsequent SOA chemistry. At the very least, a
differentiation between cyclic and acyclic terpenes is desirable. This issue
is likely to become more relevant in the future, when biological and abiotic
plant-stressed events increase in frequency as is projected for a warming
climate (Bale et al., 2002; Jactel et al., 2019).
Such stresses will both increase biogenic VOC emissions and change the
composition of emitted mixtures
(Faiola et al., 2018,
2019).
Data availability
The data shown in the paper are available on request from corresponding
author and from Siegfried Schobesberger (siegfried.schobesberger@uef.fi).
The supplement related to this article is available online at: https://doi.org/10.5194/acp-20-5629-2020-supplement.
Author contributions
AV, CF, AB and TYJ designed the study. AY, AB, ZL, LB, AL, CF, EK and SN
performed the measurements. AY and SS led the paper writing, and all of the
co-authors participated in the interpretation of the results and paper
editing.
Competing interests
The authors declare that they have no conflict of interest.
Acknowledgements
The authors wish to thank James Blande, Minna Kivimäenpää and
Rajendra Ghimire (University of Eastern Finland, Department of Environmental
and Biological Sciences) for tending the Scots pine seedlings. Sergey A.
Nizkorodov acknowledges the Fulbright Finland Foundation and the
Saastamoinen Foundation for funding his visit to the University of Eastern
Finland.
Financial support
his research was supported by the Academy
of Finland (272041, 310682, 299544), European Research Council
(ERC-StQ QAPPA 335478) and University of Eastern Finland Doctoral
Program in Environmental Physics, Health and Biology.
Review statement
This paper was edited by Barbara Ervens and Jacqui Hamilton and reviewed by two anonymous referees.
ReferencesAmin, H., Atkins, P. T., Russo, R. S., Brown, A. W., Sive, B., Hallar, A. G.,
and Huff Hartz, K. E.: Effect of Bark Beetle Infestation on Secondary
Organic Aerosol Precursor Emissions, Environ. Sci. Technol., 46,
5696–5703, 10.1021/es204205m, 2012.Bäck, J., Aalto, J., Henriksson, M., Hakola, H., He, Q., and Boy, M.: Chemodiversity of a Scots pine stand and implications for terpene air concentrations, Biogeosciences, 9, 689–702, 10.5194/bg-9-689-2012, 2012.Bale, J. S., Masters, G. J., Hodkinson, I. A. N. D., Awmack, C., Bezemer, T.
M., Brown, V. K., Butterfield, J., Buse, A., Coulson, J. C., Farrar, J.,
Good, J. E. G., Harrington, R., Hartley, S., Jones, T. H., Lindroth, R. L.,
and Press, M. C.: Herbivory in global climate change research: direct
effects of rising temperature on insect herbivores, Glob. Chang. Biol., 8, 1–16, 10.1046/j.1365-2486.2002.00451.x, 2002.Bannan, T. J., Le Breton, M., Priestley, M., Worrall, S. D., Bacak, A., Marsden, N. A., Mehra, A., Hammes, J., Hallquist, M., Alfarra, M. R., Krieger, U. K., Reid, J. P., Jayne, J., Robinson, W., McFiggans, G., Coe, H., Percival, C. J., and Topping, D.: A method for extracting calibrated volatility information from the FIGAERO-HR-ToF-CIMS and its experimental application, Atmos. Meas. Tech., 12, 1429–1439, 10.5194/amt-12-1429-2019, 2019.Berg, A. R., Heald, C. L., Huff Hartz, K. E., Hallar, A. G., Meddens, A. J. H., Hicke, J. A., Lamarque, J.-F., and Tilmes, S.: The impact of bark beetle infestations on monoterpene emissions and secondary organic aerosol formation in western North America, Atmos. Chem. Phys., 13, 3149–3161, 10.5194/acp-13-3149-2013, 2013.Bianchi, F., Kurten, 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.Blande, J. D., Turunen, K., and Holopainen, J. K.: Pine weevil feeding on
Norway spruce bark has a stronger impact on needle VOC emissions than
enhanced ultraviolet-B radiation, Environ. Pollut., 157, 174–180,
10.1016/J.ENVPOL.2008.07.007, 2009.Buchholz, A., Lambe, A. T., Ylisirniö, A., Li, Z., Tikkanen, O.-P., Faiola, C., Kari, E., Hao, L., Luoma, O., Huang, W., Mohr, C., Worsnop, D. R., Nizkorodov, S. A., Yli-Juuti, T., Schobesberger, S., and Virtanen, A.: Insights into the O : C-dependent mechanisms controlling the evaporation of α-pinene secondary organic aerosol particles, Atmos. Chem. Phys., 19, 4061–4073, 10.5194/acp-19-4061-2019, 2019.Capouet, M. and Müller, J.-F.: A group contribution method for estimating the vapour pressures of α-pinene oxidation products, Atmos. Chem. Phys., 6, 1455–1467, 10.5194/acp-6-1455-2006, 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.Faiola, C. L., Buchholz, A., Kari, E., Yli-Pirilä, P., Holopainen, J.
K., Kivimäenpää, M., Miettinen, P., Worsnop, D. R., Lehtinen, K.
E. J., Guenther, A. B., and Virtanen, A.: Terpene Composition Complexity
Controls Secondary Organic Aerosol Yields from Scots Pine Volatile
Emissions, Sci. Rep., 8, 1–13, 10.1038/s41598-018-21045-1, 2018.Faiola, C. L., Pullinen, I., Buchholz, A., Khalaj, F., Ylisirnio, A., Kari,
E., Schobesberger, S., Yli-juuti, T., Miettinen, P., Holopainen, J. K., and
Kivima, M.: Secondary Organic Aerosol Formation from Healthy and Aphid-Stressed Scots Pine Emissions, ACS Earth Space Chem., 3, 1756–1772, 10.1021/acsearthspacechem.9b00118,
2019.Glasius, M. and Goldstein, A. H.: Recent Discoveries and Future Challenges in Atmospheric Organic Chemistry, Environ. Sci. Technol., 50, 2754–2764, 10.1021/acs.est.5b05105, 2016.Hakola, H., Tarvainen, V., Praplan, A. P., Jaars, K., Hemmilä, M., Kulmala, M., Bäck, J., and Hellén, H.: Terpenoid and carbonyl emissions from Norway spruce in Finland during the growing season, Atmos. Chem. Phys., 17, 3357–3370, 10.5194/acp-17-3357-2017, 2017.Hall, W. A., Pennington, M. R., and Johnston, M. V.: Molecular
transformations accompanying the aging of laboratory secondary organic
aerosol, Environ. Sci. Technol., 47, 2230–2237, 10.1021/es303891q,
2013.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.Hellén, H., Praplan, A. P., Tykkä, T., Ylivinkka, I., Vakkari, V., Bäck, J., Petäjä, T., Kulmala, M., and Hakola, H.: Long-term measurements of volatile organic compounds highlight the importance of sesquiterpenes for the atmospheric chemistry of a boreal forest, Atmos. Chem. Phys., 18, 13839–13863, 10.5194/acp-18-13839-2018, 2018.Holopainen, J. K. and Gershenzon, J.: Multiple stress factors and the
emission of plant VOCs, Trends Plant Sci., 15, 176–184,
10.1016/j.tplants.2010.01.006, 2010.Holopainen, J. K., Kivimäenpää, M., and Nizkorodov, S. A.:
Plant-derived Secondary Organic Material in the Air and Ecosystems, Trends
Plant Sci., 22, 744–753, 10.1016/j.tplants.2017.07.004, 2017.Huang, W., Saathoff, H., Pajunoja, A., Shen, X., Naumann, K.-H., Wagner, R., Virtanen, A., Leisner, T., and Mohr, C.: α-Pinene secondary organic aerosol at low temperature: chemical composition and implications for particle viscosity, Atmos. Chem. Phys., 18, 2883–2898, 10.5194/acp-18-2883-2018, 2018.Iyer, S., He, X., Hyttinen, N., Kurtén, T., and Rissanen, M. P.:
Computational and Experimental Investigation of the Detection of HO2
Radical and the Products of Its Reaction with Cyclohexene Ozonolysis Derived
RO2 Radicals by an Iodide-Based Chemical Ionization Mass Spectrometer,
J. Phys. Chem. A, 121, 6778–6789, 10.1021/acs.jpca.7b01588, 2017.Jactel, H., Koricheva, J., and Castagneyrol, B.: ScienceDirect Responses of
forest insect pests to climate change: not so simple, Curr. Opin. Insect Sci., 35, 103–108,
10.1016/j.cois.2019.07.010, 2019.Jaoui, M., Leungsakul, S., and Kamens, R. M.: Gas and Particle Products
Distribution from the Reaction of β-Caryophyllene with Ozone, J. Atmos. Chem., 45, 261–287,
261–287, 2003.Jaoui, M., Kleindienst, T. E., Docherty, K. S., Lewandowski, M., and
Offenberg, J. H.: Secondary organic aerosol formation from the oxidation of
a series of sesquiterpenes: α-cedrene, β-caryophyllene,
α-humulene and α-farnesene with O3, OH and NO3
radicals, Environ. Chem., 10, 178, 10.1071/EN13025, 2013.Jaoui, M., Lewandowski, M., Offenberg, J. H., Docherty, K. S., and Kleindienst, T. E.: Ozonolysis of α/β-farnesene mixture: Analysis of gas-phase and particulate reaction products, Atmos. Environ., 169, 175–192, 10.1016/j.atmosenv.2017.08.065, 2017.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, E. 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., Worsnop, D. R., and Worsnop, D. R.: Evolution of organic aerosols in the atmosphere, Science, 326, 1525–1529, 10.1126/science.1180353, 2009.Junninen, H., Ehn, M., Petäjä, T., Luosujärvi, L., Kotiaho, T., Kostiainen, R., Rohner, U., Gonin, M., Fuhrer, K., Kulmala, M., and Worsnop, D. R.: A high-resolution mass spectrometer to measure atmospheric ion composition, Atmos. Meas. Tech., 3, 1039–1053, 10.5194/amt-3-1039-2010, 2010.Kang, E., Root, M. J., Toohey, D. W., and Brune, W. H.: Introducing the concept of Potential Aerosol Mass (PAM), Atmos. Chem. Phys., 7, 5727–5744, 10.5194/acp-7-5727-2007, 2007.Kari, E., Miettinen, P., Yli-Pirilä, P., Virtanen, A., and Faiola, C. L.:
PTR-ToF-MS product ion distributions and humidity-dependence of biogenic
volatile organic compounds, Int. J. Mass Spectrom., 430, 87–97,
10.1016/j.ijms.2018.05.003, 2018.Kim, D., Stevens, P. S., and Hites, R. A.: Rate constants for the gas-phase reactions of OH and O3 with β-ocimene,β-myrcene, and α-and β-farnesene as a function of temperature, J. Phys. Chem. A, 115, 500–506, 10.1021/jp111173s, 2011.Kourtchev, I., Bejan, I., Sodeau, J. R., and Wenger, J. C.: Gas-phase
reaction of (E)-β-farnesene with ozone: Rate coefficient and carbonyl
products, Atmos. Environ., 43, 3182–3190,
10.1016/j.atmosenv.2009.03.048, 2009.Kourtchev, I., Bejan, I., Sodeau, J. R., and Wenger, J. C.: Gas phase
reaction of OH radicals with (E)-β-farnesene at 296±2 K:
Rate coefficient and carbonyl products, Atmos. Environ., 46, 338–345,
10.1016/j.atmosenv.2011.09.061, 2012.Krieger, U. K., Siegrist, F., Marcolli, C., Emanuelsson, E. U., Gøbel, F. M., Bilde, M., Marsh, A., Reid, J. P., Huisman, A. J., Riipinen, I., Hyttinen, N., Myllys, N., Kurtén, T., Bannan, T., Percival, C. J., and Topping, D.: A reference data set for validating vapor pressure measurement techniques: homologous series of polyethylene glycols, Atmos. Meas. Tech., 11, 49–63, 10.5194/amt-11-49-2018, 2018.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.Lambe, A. T., Ahern, A. T., Williams, L. R., Slowik, J. G., Wong, J. P. S., Abbatt, J. P. D., Brune, W. H., Ng, N. L., Wright, J. P., Croasdale, D. R., Worsnop, D. R., Davidovits, P., and Onasch, T. B.: Characterization of aerosol photooxidation flow reactors: heterogeneous oxidation, secondary organic aerosol formation and cloud condensation nuclei activity measurements, Atmos. Meas. Tech., 4, 445–461, 10.5194/amt-4-445-2011, 2011.Lambe, A. T., Onasch, T. B., Croasdale, D. R., Wright, J. P., Martin, A. T.,
Franklin, J. P., Massoli, P., Kroll, J. H., Canagaratna, M. R., Brune, W.
H., Worsnop, D. R., and Davidovits, P.: Transitions from Functionalization to
Fragmentation Reactions of Laboratory Secondary Organic Aerosol (SOA)
Generated from the OH Oxidation of Alkane Precursors, Environ. Sci.
Technol., 46, 5430–5437, 10.1021/es300274t, 2012.Lee, A., Goldstein, A. H., Kroll, J. H., Ng, N. L., Varutbangkul, V.,
Flagan, R. C., and Seinfeld, J. H.: Gas-phase products and secondary aerosol
yields from the photooxidation of 16 different terpenes, J. Geophys. Res.,
111, D17305, 10.1029/2006JD007050, 2006.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.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., Ehn, M., Rubach, F., Kleist, E., Wildt, J., Mentel, Th. F., Carrasquillo, A. J., Daumit, K. E., Hunter, J. F., Kroll, J. H., Worsnop, D. R., and Thornton, J. A.: Phase partitioning and volatility of secondary organic aerosol components formed from α-pinene ozonolysis and OH oxidation: the importance of accretion products and other low volatility compounds, Atmos. Chem. Phys., 15, 7765–7776, 10.5194/acp-15-7765-2015, 2015.Mackenzie-Rae, F. A., Liu, T., Deng, W., Saunders, S. M., Fang, Z., Zhang, Y., and Wang, X.: Ozonolysis of α-phellandrene – Part 1: Gas- and particle-phase characterisation, Atmos. Chem. Phys., 17, 6583–6609, 10.5194/acp-17-6583-2017, 2017.McFiggans, G., Mentel, T. F., Wildt, J., Pullinen, I., Kang, S., Kleist, E.,
Schmitt, S., Springer, M., Tillmann, R., Wu, C., Zhao, D., Hallquist, M.,
Faxon, C., Breton, M. Le, Hallquist, A. M., Simpson, D., Bergstro, R.,
Jenkin, M. E., Ehn, M., Thornton, J. A., Alfarra, M. R., Bannan, T. J.,
Percival, C. J., Priestley, M., Topping, D., and Kiendler-scharr, A.:
Secondary organic aerosol reduced by mixture of atmospheric vapours, Nature, 565, 587–593,
10.1038/s41586-018-0871-y, 2019.Ng, N. L., Brown, S. S., Archibald, A. T., Atlas, E., Cohen, R. C., Crowley, J. N., Day, D. A., Donahue, N. M., Fry, J. L., Fuchs, H., Griffin, R. J., Guzman, M. I., Herrmann, H., Hodzic, A., Iinuma, Y., Jimenez, J. L., Kiendler-Scharr, A., Lee, B. H., Luecken, D. J., Mao, J., McLaren, R., Mutzel, A., Osthoff, H. D., Ouyang, B., Picquet-Varrault, B., Platt, U., Pye, H. O. T., Rudich, Y., Schwantes, R. H., Shiraiwa, M., Stutz, J., Thornton, J. A., Tilgner, A., Williams, B. J., and Zaveri, R. A.: Nitrate radicals and biogenic volatile organic compounds: oxidation, mechanisms, and organic aerosol, Atmos. Chem. Phys., 17, 2103–2162, 10.5194/acp-17-2103-2017, 2017.Odum, J. R., Hoffmann, T., Bowman, F., Collins, D., Flagan Richard, C., and
Seinfeld, J. H.: Gas particle partitioning and secondary organic aerosol
yields, Environ. Sci. Technol., 30, 2580–2585, 10.1021/es950943+,
1996.Palm, B. B., Campuzano-Jost, P., Ortega, A. M., Day, D. A., Kaser, L., Jud, W., Karl, T., Hansel, A., Hunter, J. F., Cross, E. S., Kroll, J. H., Peng, Z., Brune, W. H., and Jimenez, J. L.: In situ secondary organic aerosol formation from ambient pine forest air using an oxidation flow reactor, Atmos. Chem. Phys., 16, 2943–2970, 10.5194/acp-16-2943-2016, 2016.Pankow, J. F.: An absorption model of the gas/aerosol partitioning involved in the formation of secondary organic aerosol, Atmos. Environ., 28, 189–193, 10.1016/1352-2310(94)90094-9, 1994.Pankow, J. F. and Asher, W. E.: SIMPOL.1: a simple group contribution method for predicting vapor pressures and enthalpies of vaporization of multifunctional organic compounds, Atmos. Chem. Phys., 8, 2773–2796, 10.5194/acp-8-2773-2008, 2008.Passananti, M., Zapadinsky, E., Zanca, T., Ehn, M., Attoui, M. and Vehkama,
H.: ChemComm How well can we predict cluster fragmentation, Chem. Commun., 55, 5946–5949,
10.1039/c9cc02896j, 2019.Peng, Z., Day, D. A., Stark, H., Li, R., Lee-Taylor, J., Palm, B. B., Brune, W. H., and Jimenez, J. L.: HOx radical chemistry in oxidation flow reactors with low-pressure mercury lamps systematically examined by modeling, Atmos. Meas. Tech., 8, 4863–4890, 10.5194/amt-8-4863-2015, 2015.Peng, Z., Day, D. A., Ortega, A. M., Palm, B. B., Hu, W., Stark, H., Li, R., Tsigaridis, K., Brune, W. H., and Jimenez, J. L.: Non-OH chemistry in oxidation flow reactors for the study of atmospheric chemistry systematically examined by modeling, Atmos. Chem. Phys., 16, 4283–4305, 10.5194/acp-16-4283-2016, 2016.
Riipinen, I., Yli-Juuti, T., Pierce, J. R., Petäjä, T., Worsnop, D. R., Kulmala, M. and Donahue, N. M.: The contribution of organics to atmospheric nanoparticle growth, Nat. Geosci., 5, 453–458, 10.1038/ngeo1499, 2012.Schobesberger, S., D'Ambro, E. L., Lopez-Hilfiker, F. D., Mohr, C., and Thornton, J. A.: A model framework to retrieve thermodynamic and kinetic properties of organic aerosol from composition-resolved thermal desorption measurements, Atmos. Chem. Phys., 18, 14757–14785, 10.5194/acp-18-14757-2018, 2018.Spanke, J., Rannik, Ü., Forkel, R., Nigge, W., and Hoffman, T.: Emission fluxes and atmospheric degradation of monoterpenes above a boreal forest: field measurements and modelling, Tellus B, 53, 406–422, 10.3402/tellusb.v53i4.16614, 2001.Surratt, J. D., Goìmez-Gonzaìlez, Y., Chan, A. W. H., Vermeylen, R.,
Shahgholi, M., Kleindienst, T. E., Edney, E. O., Offenberg, J. H.,
Lewandowski, M., Jaoui, M., Maenhaut, W., Claeys, M., Flagan, R. C., and
Seinfeld, J. H.: Organosulfate Formation in Biogenic Secondary Organic
Aerosol, J. Phys. Chem. A, 112, 8345–8378, 10.1021/jp802310p, 2008.Yassaa, N., Song, W., Lelieveld, J., Vanhatalo, A., Bäck, J., and Williams, J.: Diel cycles of isoprenoids in the emissions of Norway spruce, four Scots pine chemotypes, and in Boreal forest ambient air during HUMPPA-COPEC-2010, Atmos. Chem. Phys., 12, 7215–7229, 10.5194/acp-12-7215-2012, 2012.