Oligomerization reaction of the Criegee intermediate leads to secondary organic aerosol formation in ethylene ozonolysis

Article abstract of DOI:10.1021/jp408672m

Ethylene ozonolysis was investigated in laboratory experiments using a Teflon bag reactor. A negative ion chemical ionization mass spectrometer (NI-CIMS) using SO₂Cl^- and Cl^- as reagent ions was used for product analysis. In addition to the expected gas-phase products, such as formic acid and hydroperoxymethyl formate, oligomeric hydroperoxides composed of the Criegee intermediate (CH₂OO) as a chain unit were observed. Furthermore, we observed secondary organic aerosol (SOA) formation from the ethylene ozonolysis, and the particle-phase products were also analyzed by NI-CIMS. The CH₂OO oligomers were also observed as particle-phase components, suggesting that the oligomeric hydroperoxides formed in the gas phase partition into the particle phase. By adding methanol as a stabilized Criegee intermediate scavenger, both the gas-phase oligomer formation and SOA formation were strongly suppressed. This indicates that CH₂OO plays a critical role in the formation of oligomeric hydroperoxides followed by SOA formation in ethylene ozonolysis. A new formation mechanism for the oligomeric hydroperoxides, which includes sequential addition of CH₂OO to hydroperoxides, is proposed.

Full text of DOI:10.1021/jp408672m

Article
pubs.acs.org/JPCA
Oligomerization Reaction of the Criegee Intermediate Leads to
Secondary Organic Aerosol Formation in Ethylene Ozonolysis
Yosuke Sakamoto,*,† Satoshi Inomata,‡ and Jun Hirokawa*,†
†Faculty of Environmental Earth Science, Hokkaido University, Sapporo 060-0610, Japan
‡Center for Global Environmental Research, National Institute for Environmental Studies, Tsukuba, Ibaraki 305-8506, Japan
ABSTRACT: Ethylene ozonolysis was investigated in laboratory experiments using a
Teflon bag reactor. A negative ion chemical ionization mass spectrometer (NI-CIMS)
using SO2Cl− and Cl− as reagent ions was used for product analysis. In addition to the
expected gas-phase products, such as formic acid and hydro-
peroxymethyl formate, oligomeric hydroperoxides composed of the Criegee
intermediate (CH2OO) as a chain unit were observed. Furthermore, we observed
secondary organic aerosol (SOA) formation from the ethylene ozonolysis, and the
particle-phase products were also analyzed by NI-CIMS. The CH2OO oligomers were
also observed as particle-phase components, suggesting that the oligomeric
hydroperoxides formed in the gas phase partition into the particle phase.
By adding methanol as a stabilized Criegee intermediate scavenger, both the gas-
phase oligomer formation and SOA formation were strongly suppressed. This
indicates that CH2OO plays a critical role in the formation of oligomeric
hydroperoxides followed by SOA formation in ethylene ozonolysis. A new formation
mechanism for the oligomeric hydroperoxides, which includes sequential addition of CH2OO to hydroperoxides, is proposed.
intermediate, CH2OO*. While some of the CH2OO* subse-
quently decomposes to produce CO, CO2, H2, H2O, H, and
OH or isomerizes to formic acid (HCOOH), the other part is
collisionally stabilized.
In ethylene ozonolysis, the stabilized Criegee intermediate
(SCI), CH2OO, has been shown to react with other pro-
ducts.2,4−8,10,11,14−16 For example, it can react with the initially
produced HCHO to produce a secondary ozonide:
INTRODUCTION
■
Alkene ozonolysis is an important source of free radicals and
secondary organic aerosols (SOAs) in the Earth’s atmosphere.1
Although the reaction mechanism has been investigated, it is
not yet understood in full detail. Most mechanistic studies of
alkene ozonolysis have focused on ozonolysis of the simplest
alkene, ethylene.2−15 Figure 1 briefly summarizes the initial
steps involved in ethylene ozonolysis. First, ozone adds across
the ethylene double bond to form a primary ozonide, which
decomposes to formaldehyde (HCHO) and an excited Criegee
Additionally, the SCI may also react with H2O and HCOOH to
produce hydroxymethyl hydroperoxide (HOCH2OOH,
HMHP) and hydroperoxymethyl formate (HCOOCH2OOH,
HPMF), respectively.
CH2OO + H2O → HO − CH2OO − H
(2)
Neeb et al. estimated the relative rate constants for these
reactions and concluded that the reaction of CH2OO with H2O
is the dominant process under tropospheric conditions.5,15
While reactions with other atmospheric species such as SO2 and
NO2 have been reported for a variety of SCIs including CH2OO,
reaction with H2O is predicted to consume most of the SCIs
Received: August 29, 2013
Revised: November 2, 2013
Published: November 7, 2013
Figure 1. Brief summary of initial steps of ethylene ozonolysis.
© 2013 American Chemical Society
12912
dx.doi.org/10.1021/jp408672m | J. Phys. Chem. A 2013, 117, 12912−12921

The Journal of Physical Chemistry A
Article
Figure 2. Schematic diagram of the experimental setup.
under tropospheric conditions based on indirectly estimated
rate constants.1 Recently, however, field observations suggest
that SCIs may contribute to the oxidation of SO2 and other
trace gases,17 implying that SCIs including CH2OO may be
significant participants in tropospheric chemistry.
with RH well below 1% with an initial total volume of approx-
imately 0.85 m3.
Ozone and gas-phase products were monitored using a
commercial ozone monitor (model 1150; Dylec Inc., Ibaraki,
Japan) and a custom-built NI-CIMS, respectively. The NI-CIMS is
described in the following subsection. All tubes and tube
connectors used for NI-CIMS were made of Teflon because
some reaction products disappeared when stainless steel tubing
was used, probably due to heterogeneous reactions on the
metal walls. The aerosol size distribution was measured using a
scanning mobility particle sizer (SMPS, DMA model 3080,
CPC model 3775; TSI Inc., Shoreview, MN, USA). To analyze
the aerosol composition, nearly all the air remaining in the bag
after a reaction time of 1 h was extracted using a diaphragm
pump, and any particles in the extracted air were collected on a
PTFE filter (ADVANTEC PF020; Toyo Roshi Kaisha, Ltd.,
Tokyo, Japan). This particle collection process took approx-
imately 30 min. Then, the filter was heated at 403 K in a flow
of synthetic air to vaporize components of the collected
particles, and this vapor was introduced into the NI-CIMS for
analysis.
Typical initial concentrations of ethylene and ozone were
2.9 and 4.5 ppmv, respectively, unless otherwise noted. Although
these concentrations are several orders of magnitude higher
than ambient levels, we used them due to the sensitivity
limitations of our instruments. In some of the experiments,
excess cyclohexane (99.8%, Wako Pure Chemical Industries,
Ltd., Osaka, Japan) was added as an OH radical scavenger,9 and
excess methanol (99.9%, Wako Pure Chemical Industries, Ltd.)
was added as an SCI scavenger.11,21 These scavenger compounds
were used without further purification. They were injected as
liquids into the Teflon bag using a syringe and allowed to
vaporize prior to initiating the reaction. A typical concentration
of cyclohexane was 350 ppmv. On the basis of the reported rate
constant of 7.9 × 10−12 cm3 molecule−1 s−1 (at 1 bar air) for
ethylene + OH22 and that of 7.0 × 10−12 cm3 molecule−1 s−1 for
cyclohexane + OH,23 approximately 99% of any product OH
radicals will be scavenged by cyclohexane under the present
To expand our understanding of CH2OO chemistry in the
troposphere, we investigated ethylene ozonolysis in laboratory
experiments under dry conditions using a Teflon bag reaction
enclosure. Gas-phase products were analyzed by a negative ion
chemical ionization mass spectrometer (NI-CIMS) using
SO2Cl− or Cl− as the reagent ion. SO2Cl−-CIMS has the ability
to ionize acidic species through a chloride ion transfer reaction
because of their high chloride affinities.18,19 We also found that
a variety of hydroperoxides can be detected using SO2Cl−- and
Cl−-CIMS. In addition, during the ethylene ozonolysis
experiments, SOA formation was observed. This is intriguing
because the known products of the ethylene ozonolysis
are thought to be too volatile to form SOAs.20 Thus, we
conducted off-line NI-CIMS analyses of the particle-phase
products by collecting the SOAs on a PTFE filter, which
was then heated to vaporize the particle-phase components.
By comparing the gas- and aerosol-phase products, we pro-
posed a new mechanism for SOA formation from ethylene
ozonolysis.
EXPERIMENTAL SECTION
■
Experiments were performed using a 175 cm × 155 cm pillow-
shaped bag made of FEP Teflon. Figure 2 depicts the experi-
mental setup. Synthetic air (composed of N2 and O2 with other
contaminants below 1 ppmv, Japan Fine Products, Kawasaki,
Japan) and ethylene (C2H4, 1% in N2, Japan Fine Products)
were introduced into the Teflon bag.
The ozonolysis reaction was initiated by introduction of
ozone (O3), which was produced by irradiating the synthetic
air with vacuum ultraviolet light at 185 nm from a low-
pressure Hg lamp, into the Teflon bag containing synthetic
air and ethylene. The reaction was investigated at room
temperature and atmospheric pressure under dry conditions
12913
dx.doi.org/10.1021/jp408672m | J. Phys. Chem. A 2013, 117, 12912−12921

The Journal of Physical Chemistry A
Article
Figure 3. Schematic diagram of the NI-CIMS instrument.
experimental conditions. The products of cyclohexane with OH
radical include cyclohexanol (c-C6H11OH), cyclohexanone
(c-C6H10O), cyclohexyl hydroperoxide (c-C6H11OOH), and
other products formed via ring-opening.9 A typical concen-
tration of methanol was 1300 ppmv. It should be noted that in
experiments with the addition of both cyclohexane and
methanol, a fraction of the OH radicals was scavenged by
methanol as well as cyclohexane. On the basis of the rate
constant of 0.93 × 10−12 cm3 molecule−1 s−1 for methanol +
OH,22 approximately 33% of the OH radicals will be scavenged
by methanol and 67% by cyclohexane. The reaction of methanol
with OH produces HCHO and HO2 with a yield of unity.22
Negative Ion Chemical Ionization Mass Spectrometer
(NI-CIMS). In the NI-CIMS analysis, either SO2Cl− or Cl− was
used as a reagent ion. A schematic diagram of the NI-CIMS
instrument is shown in Figure 3.
chemical ionization directly. In Cl−-chemical ionization, species
were ionized through a chloride attachment reaction involving a
third body, M, as shown below:
A + Cl− + M → ACl− + M
(5)
Through reaction 5, species with lower chloride affinities than
SO2 could be also ionized. Thus, Cl−-CIMS can detect a greater
variety of species than SO2Cl−-CIMS. However, it is difficult to
obtain quantitative information from the Cl−-CIMS method
because the reagent Cl− is largely consumed by reaction with
the analyte species.
As shown in eqs 4 and 5, analyte molecules are converted to
Cl− adducts in both chemical ionization schemes. Because
chlorine has two stable isotopes, 35Cl and 37Cl, at the ratio of
approximately 3:1, the Cl− adduct ions are expected to be
detected as doublet peaks within the mass spectrum. Hereafter,
we will refer to the mass-to-charge ratio (m/z) of the ions
containing 35Cl− and will include the (m/z) of the ions
containing 37Cl− in parentheses, for each set of doublet peaks.
The SO2Cl−-CIMS has been described in detail previously,19
so that only a brief description focused on relevant modi-
fications for the present study will be given here. The reagent
ion, SO2Cl−, was produced through the addition of SO2 to Cl−.
While Cl− was generated by discharge of a CH3Cl/Ar gas
mixture in previous studies, it was generated by the electron
impact from a resistively heated filament onto the CH3Cl/Ar
gas mixture herein. Gas in the Teflon bag was sampled into the
NI-CIMS at a flow rate of approximately 0.8 L min−1 through a
0.25 mm diameter orifice. The sampled gas flow from the bag
was mixed with the reagent ion flow, resulting in a chloride ion
transfer reaction from the reagent SO2Cl− to a selected species,
A, in the sampled air as follows:
RESULTS AND DISCUSSION
■
Analysis of Gas-Phase Products. First, the gas-phase
products of ethylene ozonolysis with the addition of cyclo-
hexane as an OH scavenger were analyzed by NI-CIMS. Figure 4a
shows SO2Cl−-CI mass spectra.
The orange line in this figure shows a mass spectrum obtained
before the injection of ozone into the bag. The reagent ion,
SO2Cl−, is observed at m/z 99 (101). Mass peaks at m/z 163
(165) are assigned to (SO2)2Cl−, which was produced through
a further addition of SO2 to the reagent ion. No other peaks
were observed probably because ozone, ethylene, and cyclo-
hexane have much lower chloride affinities than SO2 and are
not efficiently ionized by SO2Cl−. The blue line in Figure 4a
shows a mass spectrum obtained at a reaction time of 1 h. By
comparing the two mass spectra, peaks attributable to products
of the ethylene ozonolysis can be identified. The peaks at m/z
81 (83) are assigned to the Cl− ion adduct of formic acid (FA),
which has a molecular weight (MW) of 46 and is reportedly
formed through isomerization of an excited Criegee inter-
mediate, CH2OO*,2−4 and/or through decomposition of the
secondary ozonide produced in eq 1.5 In similar fashion, the
A + SO2Cl− → ACl− + SO2
(4)
Through reaction 4, species with higher chloride affinities than
SO2 are ionized effectively. Although many organic acids and
hydroperoxides (ROOH) could be ionized and detected, a
portion of the hydroperoxides produced from ethylene ozonolysis
could not be ionized. To detect those hydroperoxides, Cl−
chemical ionization was also employed. The Cl−-CIMS is
similar to the SO2Cl−-CIMS except that SO2 is not introduced
into the ion source region. Therefore, Cl− ions generated from
electron impact on the CH3Cl/Ar mixture can participate in
12914
dx.doi.org/10.1021/jp408672m | J. Phys. Chem. A 2013, 117, 12912−12921

The Journal of Physical Chemistry A
Article
attributed to a compound composed of HPMF and two units of
CH2OO. As discussed later, the compounds with MW of 92 +
46 n (n = 1, 2) are assigned to oligomeric hydroperoxides
composed of CH2OO as a chain unit. Hereafter, these species
are denoted as OS1 (oligomer series 1). The peaks at m/z 145
(147) cannot be definitively assigned from the SO2Cl−-CI mass
spectrum alone, but comparison with the Cl−-CI mass
spectrum allows their assignment, as described below.
Figure 4b shows typical Cl−-CI mass spectra. The orange line
in the figure shows a mass spectrum obtained before ozone
injection. No peaks were observed in the m/z 75−250 range.
The blue line in the figure shows a mass spectrum obtained at a
reaction time of 1 h under the same experimental conditions as
Figure 4a. The peaks at m/z 145 (147) observed in Figure 4a
are seen to disappear in Figure 4b, which indicates that the
presence of SO2 is required for their observation. Thus, we
assign them as an adduct ion, SO2Cl−·HCOOH. Similarly, the
peaks at m/z 191 (193) are assigned to SO2Cl−·HPMF.
All peaks observed in the blue line mass spectrum in Figure 4b
are also observed in Figure 4a, except for the peaks at m/z 151
(153) and m/z 197 (199). The peaks at m/z 151 (153) are
attributed to a compound with MW of 116. The reaction
between OH and cyclohexane can produce cyclohexyl hydro-
peroxide (c-C6H11OOH) with MW of 116, and thus, we
assigned this compound to c-C6H11OOH. Similarly, the peaks
at m/z 197 (199) can be attributed to a compound composed
of CH2OO and c-C6H11OOH. These species were not observed
in the SO2Cl−-CIMS possibly because their chloride affinities
are lower than SO2. Using either reagent ion method, NI-CIMS
could not detect formaldehyde, which is a major product of the
ethylene ozonolysis.2−5,12,14,25 This suggests that formaldehyde
has too low of a chloride affinity to be ionized, even through
the chloride attachment reaction. For the same reason, other
expected products such as CO2, CO, and H2 could not be
detected. The observed SO2Cl−- and Cl−-CIMS peaks are
summarized in Table 1.
Figure 4. Typical mass spectra of gas-phase products in the ethylene
ozonolysis with cyclohexane obtained by SO2Cl−-CIMS (a) and
Cl−-CIMS (b) with concentrations of 2.9 ppmv C2H4, 4.5 ppmv O3,
and 350 ppmv cyclohexane. Blue line, 1 h after the initiation of the
reaction; orange line, before reaction initiation. The vertical axis in
panel a represents signal count rates normalized by that of SO235Cl at
m/z 99. FA and HPMF denote formic acid and hydroperoxymethyl
formate, respectively. OS1 denotes the oligomer series 1 (see text).
molecular weight of each analyte can be obtained by subtracting
35 (37), the mass number of chlorine, from the mass-to-charge
ratio at which analyte ion signals are detected. In addition to
signals attributed to formic acid, a series of mass peaks starting
from m/z 127 (129) with an interval of 46 Da were observed.
The strong peaks at m/z 127 (129) with MW of 92 are
attributed to HCOOCH2OOH (HPMF), which formed via
the reaction between formic acid and CH2OO (eq 3).5,8,10,24
The peaks at m/z 173 (175) are attributed to a compound with
MW of 92 + 46, suggesting that this compound is composed of
HPMF and CH2OO. Similarly, the peaks at m/z 219 (221) are
Then, the gas-phase products of the ethylene ozonolysis were
investigated in the absence of an OH scavenger. Figure 5a
shows a typical SO2Cl−-CI mass spectrum obtained without an
OH scavenger (pink line).
For comparison, we replot a mass spectrum obtained with
the addition of cyclohexane as the blue line. Note that these are
difference spectra and are obtained by subtracting the mass
spectrum before adding O3 from that obtained at a reaction
Table 1. Summary of the Peaks Observed in the Gas-Phase Analysis by NI-CIMS 1 h after Initiation of the Reaction
a
b
b
b
c
m/z
MW
w/ cyclohexane
w/o OH scavenger
w/ cyclohexane and methanol
attribution
81(83)
46
64
78
92
○
nd
nd
○
○
○
nd
○
○
○
nd
○
○
○
nd
○
○
○
○
○
○
○
○
○
nd
nd
nd
nd
nd
nd
nd
FA
d
d
99(101)
CH3OH
d
d
113(115)
127(129)
145(147)
151(153)
159(161)
173(175)
191(193)
197(199)
205(207)
219(221)
251(253)
2-HEHP, CH3OCH2OOH
HPMF
d
d
110
116
124
138
156
162
170
184
216
FA, CH3OCH2OOH
d
d
d
○
c-C6H11OOH
nd
○
○
OS2 (n = 1)
OS1 (n = 1)
HPMF
d
d
○
c-C6H11OOCH2OOH
nd
○
nd
OS2 (n = 2)
OS1 (n = 2)
OS2 (n = 3)
a
b
c
d
Obtained by subtracting 35 (37) from m/z. nd: not detected. OS1: m/z = 127 + 46 n. OS2: 113 + 46 n. Observed in the Cl−-CIMS mode only.
12915
dx.doi.org/10.1021/jp408672m | J. Phys. Chem. A 2013, 117, 12912−12921

The Journal of Physical Chemistry A
Article
comparison of these two mass spectra, OS2 production is
clearly seen in the absence of OH scavengers. Additionally, the
peaks at m/z 151 (153) and 197 (199) disappear in the absence
of cyclohexane. This is consistent with the mass assignments in
the presence of cyclohexane, in which the peaks at m/z 151
(153) are attributed to cyclohexyl hydroperoxide (c-C6H11OOH)
produced from the reaction of OH and cyclohexane, and the
peaks at m/z 197 (199) to an addition product of CH2OO to
c-C6H11OOH. The observed product peaks in the absence of an
OH scavenger are also summarized in Table 1.
To confirm that SCI was involved in the gas-phase oligomer
formation, experiments with the addition of methanol as an
SCI scavenger were performed. Figure 6a shows a typical
Figure 5. Typical mass spectra of gas-phase products in the ethylene
ozonolysis obtained by SO2Cl−-CIMS (a) and Cl−-CIMS (b) 1 h after
the initiation of the reaction. The pink lines in both panels show the
spectra without OH scavenger ([C2H4] = 2.9 ppmv and [O3] = 4.5
ppmv), and the blue lines with cyclohexane (same as that in Figure 4).
The spectra in panel a are obtained by subtracting spectra before
adding O3 to initiate the reaction. FA, 2-HEHP, and HPMF denote
formic acid, 2-hydroxyethyl hydroperoxide, and hydroperoxymethyl
formate, respectively. OS1 and OS2 denote the oligomer series 1 and
2, respectively (see text).
time of 1 h. All peaks observed in the presence of cyclohexane
(blue line) were also observed without the OH scavenger (pink
line). This confirms that these peaks are attributable to pro-
ducts from the ethylene ozonolysis. In addition, a series of peaks
starting from m/z 113 (115) were observed. In the absence of a
scavenger, the OH radicals produced during ozonolysis can
contribute to the ethylene oxidation reaction. The OH oxida-
tion of ethylene can produce HOCH2CH2OOH (2-hydrox-
yethyl hydroperoxide, 2-HEHP) with MW of 78 as follows:7
Figure 6. Typical mass spectra of gas-phase products in the ethylene
ozonolysis obtained by SO2Cl−-CIMS (a) and Cl−-CIMS (b) 1 h after
initiation of the reaction. The green lines in both the panels show the
spectra with an SCI scavenger ([C2H4] = 2.9 ppmv, [O3] = 4.5 ppmv,
[cyclohexane] = 350 ppmv, and [methanol] = 1300 ppmv), and the
blue lines with cyclohexane (same as that inFigure 4). The spectra in
panel a are obtained by subtracting spectra before adding O3 to initiate
the reaction. FA and HPMF denote formic acid and hydroperoxymethyl
formate, respectively. OS1 denotes the oligomer series 1 (see text).
CH2CH2 + OH → HO−CH2CH2
(6)
HO−CH2CH2 + O2 → HO−CH2CH2−OO
(7)
HO−CH2CH2−OO + HO2
→ HO−CH2CH2−OOH + O2
SO2Cl−-CI mass spectrum obtained with the addition of both
cyclohexane and methanol as the green line.
(8)
where HO2 in eq 8 can be produced by reactions such as O2
reacting with an H atom, which can be formed from decom-
position of an excited Criegee intermediate, CH2OO*. Thus,
the peaks at m/z 113 (115) can be attributed to 2-HEHP. The
series of peaks at m/z 113 (115) + 46 n (n = 1 − 3) are
attributed to compounds with MW of 78 + 46 n, another series
of oligomers that are composed of 2-HEHP and n CH2OO.
Hereafter, these species are denoted as OS2.
For comparison, a mass spectrum with only the addition of
cyclohexane is shown in blue. Again, both spectra are taken as
the difference between the mass spectra before and after
reaction. By comparing these two mass spectra, the peaks
attributed to HPMF and OS1 observed with the addition of
cyclohexane only (blue line) are found to disappear in the
presence of methanol (green line), clearly indicating that the
CH2OO species plays a predominant role in oligomer
formation. It is also found that the peaks attributed to formic
acid were suppressed by adding methanol, implying that at least
some fraction of the formic acid is produced from SCI through
Figure 5b shows a typical Cl−-CI mass spectrum obtained
without an OH scavenger (pink line) together with that
obtained with the addition of cyclohexane (blue line). From the
12916
dx.doi.org/10.1021/jp408672m | J. Phys. Chem. A 2013, 117, 12912−12921

The Journal of Physical Chemistry A
Article
decomposition of the CH2OO reaction products, such as the
secondary ozonide formed in eq 1.
Figure 6b shows a typical Cl−-CI mass spectrum obtained under
the same experimental conditions. In addition to the disappearance
of the peaks corresponding to HPMF and OS1 described above,
the peaks at m/z 151 (153) were found to be suppressed by adding
methanol. This is because OH radicals produced from ethylene
ozonolysis are partly scavenged by methanol as well as by cyclo-
hexane. However, new peaks at m/z 113 (115) were found to
appear with the addition of methanol. As described above, peaks
at m/z 113 (115) were observed also in the total absence of
OH scavengers and were attributed to 2-HEHP produced from
OH oxidation of ethylene. However, 2-HEHP can be ruled out in
this case because the OH radicals are sufficiently scavenged by
cyclohexane and methanol. Instead, these peaks can be attributed
to another hydroperoxide, which is produced from the reaction
between methanol and CH2OO as follows:6,26
The first step of this mechanism is the direct addition of the
Criegee intermediate, CH2OO, to a hydroperoxide, ROOH,
which is analogous to the reaction of CH2OO with hydrogen
peroxide.5 Importantly, the first step generates a hydroperoxide,
ROOCH2OOH, to which another CH2OO can add. In this
way, CH2OO adds sequentially to resultant hydroperoxides as
shown in eq 14, and finally, oligomeric hydroperoxides are
formed. The formation of OS1 can be explained based on this
mechanism with HPMF as ROOH:
CH2OO + CH3OH → CH3O−CH2OO−H
(9)
This same mechanism can be used to explain not only the
formation of OS1 but also the formation of OS2. In the OS2
formation, 2-HEHP plays the role of ROOH:
The product, CH3OCH2OOH, has MW of 78 and is expected to
be detected as a Cl− adduct at m/z 113 (115) in CIMS. In the
presence of methanol, peaks at m/z 99 (101) and those at m/z 145
(147) were also observed. They are assigned to (CH3OH)2Cl− and
(CH3OCH2OOH)(CH3OH)Cl−, respectively. These cluster ions
were observable because of the extremely high concentration of
methanol. These peaks are summarized in Table 1.
This mechanism can also be invoked to explain the formation
of other products observed by NI-CIMS. For example, the
peaks at m/z 197 (199) observed in the Cl−-CIMS spectra with
the addition of cyclohexane can be attributed to the addition
compound of CH2OO to cyclohexyl hydroperoxide through
eq 13. Thus, the newly proposed mechanism can fully explain
the formation of all of the hydroperoxides and oligomers
observed as gas-phase products. However, the radical chain
reaction mechanism26,27 shown in eqs 10−12 cannot be ruled
out. Very recently, the radical chain reaction mechanism was
supported by ab initio calculations, where the sequential
addition reactions proceed over a submerged energy barrier.29
SOA Formation from Ethylene Ozonolysis. In our
investigation of ethylene ozonolysis, SOA formation was
observed. Figure 7 shows time profiles of the particle volume
concentrations obtained from the SMPS measurements.
Proposed Mechanisms for the Oligomer Formation in
the Gas Phase. As described above, our results suggest that
oligomeric species composed of CH2OO as a chain unit are
produced in the gas phase. Oligomer formation involv-
ing Criegee intermediates from alkene ozonolysis has been
reported by Sadezky et al.27,28 They observed oligomers in the
aerosol phase using ESI-MS analysis techniques and suggested
that these oligomers are produced in the gas phase. Our results
represent the first direct observation of oligomers formed in the
gas phase and support their suggestion.
Sadezky et al.28 proposed an oligomer formation mechanism
involving initiation by the reaction of an alkylperoxy radical and
an SCI, sequential addition of SCIs, and termination by reac-
tion with the HO2 radical to produce an oligomeric hydro-
peroxide. According to their mechanism, oligomer formation
from the ethylene ozonolysis can be described as follows:
The pink, blue, and green circles show the particle volume
concentrations without scavengers, with cyclohexane, and with
cyclohexane and methanol, respectively. It should be noted that
our Teflon bag experiments potentially underestimate volume
concentrations because of possible particle deposition onto the
inner surface of the bag. Thus, only a relative change in the mea-
sured concentrations is discussed here. First, the result shown by
the blue circles suggests that either SOAs are formed via ethylene
ozonolysis in the absence of OH radicals or that the reac-
tion between OH and cyclohexane may contribute to the SOA
formation. This will be discussed on the basis of the particle-phase
analysis in the next subsection. It is seen in Figure 7 that particle
concentrations for the reactions without scavengers are higher
than those with the addition of cyclohexane. This indicates that
OH oxidation of ethylene serves as an additional contributor to
SOA formation in ethylene ozonolysis. Finally, Figure 7 shows
that aerosol formation is strongly suppressed with the addition of
methanol, as shown by the green circles. This clearly indicates that
the SCI, CH2OO, plays a critical role in SOA formation, as well as
gas-phase oligomer formation, in ethylene ozonolysis.
RO2 + CH2OO → ROO−CH2OO
(10)
where RO2 denotes alkylperoxy radicals. The formation of OS2
species observed in our studies can be rationalized by the above
mechanism by assigning RO2 in eq 10 as the 2-hydroxyethylper-
oxy radical, HOCH2CH2OO, which is produced from the reaction
between C2H4 and OH7 as shown in eq 7. However, the for-
mation of OS1 cannot be explained by the above mechanism since
the formation of radicals with MW of 45 or 91, which are unlikely
produced in ethylene ozonolysis, would be required. Instead, we
propose a new mechanism for the oligomer formation, in which
CH2OO reacts with hydroperoxides as follows:
Analysis of Particle-Phase Components. As described in
the experimental section, particles formed during the reaction
ROOH + CH2OO → ROO−CH2OO−H
(13)
12917
dx.doi.org/10.1021/jp408672m | J. Phys. Chem. A 2013, 117, 12912−12921

The Journal of Physical Chemistry A
Article
Table 2. Summary of Peaks Observed in the Particle-Phase
Analysis by NI-CIMS 1 h after Initiation of the Reaction
w/o OH
a
b
b
c
m/z
MW w/ c-hexane
scavenger
attribution
81(83)
46
○
nd
○
○
○
FA
97(99)
62
d
d
d
99(101)
113(115)
115(117)
127(129)
129
64
○
HMHP
78
nd
nd
○
○
nd
○
○
nd
○
nd
○
○
nd
nd
nd
nd
○
○
nd
○
nd
nd
nd
nd
nd
○
○
○
○
○
○
nd
○
nd
○
nd
nd
○
○
○
○
nd
nd
○
nd
○
○
○
○
○
2-HEHP
80
92
HPMF
94
143(145)
145(147)
151(153)
159(161)
165(167)
173(175)
181(183)
183(185)
187(189)
189(191)
203(205)
205(207)
211(213)
213(215)
217(219)
229(231)
233(235)
235(237)
249(251)
251(253)
297(299)
108
110
116
124
130
138
146
148
152
154
168
170
176
178
182
194
198
200
214
216
262
OS3 (n = 0)
CH3O(CH2OO)2H
Figure 7. Time profiles of the particle volume concentration
measured by SMPS. Pink circles, without scavengers ([C2H4] = 2.9
ppmv and [O3] = 4.5 ppm); blue circles, with cyclohexane ([C2H4] =
2.9 ppmv, [O3] = 4.5 ppmv, and [cyclohexane] = 350 ppmv); and
green circles, with cyclohexane and methanol ([C2H4] = 2.9 ppmv,
[O3] = 4.5 ppmv, [cyclohexane] = 350 ppmv, and [methanol] =
1300 ppmv).
e
cyclohexane−OH
OS2 (n = 1)
e
cyclohexane−OH
OS1 (n = 1)
e
cyclohexane−OH
cyclohexane−OH
e
OS3 (n = 1)
OS1 (n = 2)
e
cyclohexane−OH
cyclohexane−OH
e
e
cyclohexane−OH
OS3 (n = 2)
OS2 (n = 3)
OS2 (n = 4)
a
b
c
Obtained by subtracting 35 (37) from m/z. nd: not detected. OS1:
d
m/z = 127 + 46 n. OS2: 113 + 46 n. OS3: 143 + 46 n. Observed in
e
the Cl−-CIMS mode only. Peaks are attributed to the OH oxidation
products of cyclohexane.
Note that all the spectra in Figure 8a are obtained by
subtracting blank spectra recorded in the same setup without a
PTFE filter.
The pink line in Figure 8a shows a SO2Cl−-CI mass spectrum
obtained without an OH scavenger. Among the peaks that were
observed in the gas-phase mass spectrum, those attributed to
formic acid, HPMF, and OS2 were clearly observed, while those
attributed to OS1 had very weak intensities. It is noteworthy
that the chain length of OS2 species detected in the particle-
phase analysis (n = 1−4) was longer than those detected in the
gas-phase analysis (n = 1− 3). This suggests that OS2 produced
in the gas phase can contribute to SOA formation through
partitioning into the particle phase because of their low vapor
pressures. In addition, many peaks that were not observed in
the gas-phase analysis were observed in the particle-phase
spectra. While it is presently difficult to assign most of the peaks
solely from the observed mass-to-charge ratios, we found a
series of the peaks starting from m/z 143 (145) with an interval
of 46 Da. This series is attributed to compounds with MW of
108 + 46 n (n = 0−2), which is referred to as OS3. One can
suppose that it might result from thermal decomposition of
oligomeric species in the particle phase upon heating at 403 K
for vaporization. Alternatively, it might be produced from an
addition reaction of formaldehyde to 2-HEHP and OS2 in the
particle phase, considering that the compound with MW of 108
has a mass 30 Da higher than 2-HEHP. A possible route would
Figure 8. Typical mass spectra of particle-phase components in the
ethylene ozonolysis with and without an OH scavenger obtained by
SO2Cl−-CIMS (a) and Cl−-CIMS (b) 1 h after initiation of the
reaction. The pink lines in both panels show the spectra without an
OH scavenger ([C2H4] = 2.9 ppmv, [O3] = 4.5 ppm), and the blue
lines with cyclohexane ([C2H4] = 11.6 ppmv, [O3] = 18 ppmv, and
[cyclohexane] = 1400 ppmv). The spectra in panel a are obtained by
subtracting the blank spectra. OS2 and OS3 denote the oligomer series
2 and 3, respectively (see text).
were collected on a PTFE filter, which was then heated at
403 K. The vaporized species from the collected particles
were introduced into NI-CIMS and analyzed similarly to the
gas-phase products. Figure 8a,b shows mass spectra for the
particle-phase products, and Table 2 lists the observed peaks
and their possible attribution.
12918
dx.doi.org/10.1021/jp408672m | J. Phys. Chem. A 2013, 117, 12912−12921

The Journal of Physical Chemistry A
Article
be via an acid-catalyzed reaction producing a peroxyhemiacetal
or hemiacetal compound. The NI-CIMS analysis shows the
production of formic acid, which could act as a catalyst for the
reactions. Peroxyhemiacetal formation has been proposed as a
particle-phase reaction in SOAs from 1-tetradecene (C13H26
CH2) ozonolysis.30 Sato et al. also suggested that hemiacetal
and peroxyhemiacetal oligomers are formed in SOA from the
photooxidation of toluene.31
radical scavengers, the oligomers formed from reactions
between CH2OO produced from ethylene ozonolysis and
2-HEHP produced from OH oxidation of ethylene contribute
to SOA formation through partitioning into the aerosol phase.
This indicates that coupling between OH oxidation and
ethylene ozonolysis plays an important role in SOA formation.
In studies where cyclohexane was added as an OH scavenger,
aerosol-phase products, which would arise solely from the
ethylene ozonolysis, could not be detected. However, we
cannot rule out the possibility that ethylene ozonolysis could
contribute to SOA formation even without OH radicals because
there may be key species that cannot be detected by CIMS
employed in this study. Meanwhile, SOA components derived
from cyclohexane were detected, indicating that OH radical
scavengers could affect SOA formation through partitioning of
their oxidation products into the aerosol phase.
The pink line in Figure 8b shows the Cl−-CI mass spectrum
obtained without an OH scavenger. The peaks observed in the
SO2Cl−-CI mass spectrum were also observed in the Cl−-CI mass
spectrum. In addition, peaks at m/z 99 (101) were observed.
These peaks are possibly attributed to HMHP, which has been
reported to be produced from the reaction between CH2OO and
water as shown in eq 2. Because it was not observed as a gas-phase
product, HMHP is not likely formed from a gas-phase reaction,
but instead from a heterogeneous reaction on/in the particles.
The peaks at m/z 145 (147) were also newly observed and are
attributed to a compound composed of HMHP and CH2OO.
The blue lines in Figure 8a,b display typical SO2Cl−- and
Cl−-CI mass spectra for aerosol-phase products obtained with
cyclohexane as an OH scavenger. It should be noted that these
spectra were obtained from experiments with much higher con-
centrations to obtain sufficient signal intensities of particle com-
ponents: The concentrations of ethylene, ozone, and cyclohexane
were 12, 18, and 1400 ppmv, respectively, which are approxi-
mately four times higher than the typical conditions used in this
study. Under this higher concentration condition, the volume con-
centration reaches 0.8 × 109 nm3 cm−3 after a reaction time of 1 h,
which is comparable to that measured in the experiments without
OH scavengers under the typical conditions.
It is found from the comparison of the blue and pink lines
in Figure 8a that OS2 and OS3 peaks almost disappear in
the presence of cyclohexane. This supports the proposed
mechanism for formation of OS2, which involves gas-phase
OH oxidation of ethylene. Also, it is consistent with the
presumption that OS3 might be produced from 2-HEHP and
OS2. Conversely, several peaks not observed in the absence of
OH scavengers were observed with the addition of cyclohexane
(see Table 2). Although individual peak assignment is difficult,
we believe these peaks do not arise from ethylene ozonolysis,
but come instead from unidentified processes, including the
reaction between cyclohexane and OH. Recently, Sato et al.
detected aerosol components derived from OH radical
scavengers in their investigation of SOA formation in iso-
prene ozonolysis.32 In the presence of cyclohexane as an OH
scavenger, they identified hexanedioic acid (C6H10O4, MW 146) and
detected three other compounds with molecular formulas
C6H12O4 (MW 148), C7H12O5 (MW 176), and C8H14O5 (MW
190) in the LC/TOF-MS analysis. The particle-phase analysis
in the present study also detected mass peaks attributed to
compounds with MW of 146, 148, and 176 at m/z 181 (183),
183 (185), and 211 (213), respectively. These peaks are
consistent with the LC/TOF-MS analysis by Sato et al. and
confirm that cyclohexane itself is involved in SOA formation
during ethylene ozonolysis. In addition, several peaks attributed
to OH oxidation products of cyclohexane were observed with
an interval of 46 Da, implying that species derived from OH
oxidation of cyclohexane may interact with CH2OO.
CONCLUSIONS
■
This study examined ethylene ozonolysis using a Teflon bag
reactor. SOA formation was observed by SMPS, and gas- and
particle-phase products were analyzed by SO2Cl−- and
Cl−-CIMS. Oligomeric hydroperoxides composed of CH2OO
as a chain unit were observed as both gas- and particle-phase
products, suggesting that the oligomeric hydroperoxides formed
in the gas phase are incorporated into the particle phase. The
addition of methanol as a stabilized Criegee intermediate scav-
enger strongly suppressed both the gas-phase oligomer for-
mation and SOA formation, indicating that CH2OO plays a key
role in the formation of oligomeric hydroperoxides, followed by
SOA formation, in ethylene ozonolysis.
We propose an oligomerization mechanism initiated by the
reaction of CH2OO with ROOH, while Sadezky et al.27,28
proposed another mechanism initiated by the reaction of
CH2OO with RO2. In the latter mechanism, the CH2OO +
RO2 reaction may be limited to the daytime when photo-
chemistry is active and by the low concentration of RO2 in the
atmosphere. Conversely, the CH2OO + ROOH reaction does
not have these limitations because atmospheric concentrations
up to several ppbv have been reported for some selected
hydroperoxides,33 although our understanding of organic
hydroperoxides in the troposphere is still lacking.
As described in the introduction, recent observations17 suggest
that SCIs including CH2OO may be involved in chemical processes
in the troposphere through bimolecular reactions with species
other than H2O. However, we cannot discuss the importance of the
oligomer and SOA formation from CH2OO to tropospheric
chemistry at the present stage because information on reaction rates
and SOA yields were not obtained in this study. Further investi-
gations on the kinetics of CH2OO with atmospheric species such as
hydroperoxides are required to estimate accurately the atmospheric
impact of the oligomer and SOA formation observed in this study.
Meanwhile, the findings obtained in this study for ethylene
ozonolysis will be extended to ozonolysis of other unsaturated
hydrocarbons relevant to SOA formation in the atmosphere.
Recent laboratory and field studies suggest that dimer esters
generated from ozonolysis of monoterpenes, an important class
of biogenic hydrocarbons, participate in new particle formation
in forested regions.34−38 Dimeric species of an SCI combined
with other reaction products have also been proposed21,39 and
identified as SOA components by laboratory studies on mono-
terpene ozonolysis.40,41 The oligomer formation mechanism pre-
sented in this study may operate for larger SCIs than CH2OO
including those from monoterpene ozonolysis. The NI-CIMS
SOA Formation Chemistry from Ethylene Ozonolysis.
Our experimental results demonstrate that the SCI, CH2OO,
produced from ethylene ozonolysis plays a central role in gas-
phase oligomer and subsequent SOA formation. Without OH
12919
dx.doi.org/10.1021/jp408672m | J. Phys. Chem. A 2013, 117, 12912−12921

The Journal of Physical Chemistry A
Article
(14) Niki, H.; Maker, P. D.; Savage, C. M.; Breitenbach, L. P. A FT
IR Study of a Transitory Product in the Gas-Phase Ozone-Ethylene
Reaction. J. Phys. Chem. 1981, 85, 1024−1027.
(15) Neeb, P.; Sauer, F.; Horie, O.; Moortgat, G. K. Formation of
Hydroxymethyl Hydroperoxide and Formic Acid in Alkene Ozonolysis
in the Presence of Water Vapour. Atmos. Environ. 1997, 31, 1417−
1423.
(16) Hatakeyama, S.; Kobayashi, H.; Akimoto, H. Gas-Phase
Oxidation of Sulfur Dioxide in the Ozone−Olefin Reactions. J. Phys.
Chem. 1984, 88, 4736−4739.
(17) Mauldin, R. L., III; Berndt, T.; Sipila, M.; Paasonen, P.; Petaja,
T.; Kim, S.; Kurten, T.; Stratmann, F.; Kerminen, V. M.; Kulmala, M.
A New Atmospherically Relevant Oxidant of Sulphur Dioxide. Nature
2012, 488, 193−196.
́
(18) Hirokawa, J.; Kato, T.; Mafune, F. Uptake of Gas-Phase Nitrous
Acid by pH-Controlled Aqueous Solution Studied by a Wetted Wall
technique is expected to be applied to gas- and particle-phase
analysis in that it can detect acids and hydroperoxides that play
a critical role in SOA formation in the atmosphere.
AUTHOR INFORMATION
■
Corresponding Authors
*(Y.S.) E-mail: y_sakamoto@ees.hokudai.ac.jp.
*(J.H.) E-mail: hirokawa@ees.hokudai.ac.jp. Tel: +81-11-706-
4528. Fax: +81-11-706-4528.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by Grant-in-Aid for Scientific
Research on Innovative Areas (Grant No. 20120003) from
the Ministry of Education, Culture, Sports, Science, and
Technology in Japan. Y.S. was supported by a research
fellowship from the Japan Society for the Promotion of Science
for young scientists.
Flow Tube. J. Phys. Chem. A 2008, 112, 12143−12150.
(19) Hirokawa, J.; Kato, T.; Mafune, F. In Situ Measurements of
́
Atmospheric Nitrous Acid by Chemical Ionization Mass Spectrometry
Using Chloride Ion Transfer Reactions. Anal. Chem. 2009, 81, 8380−
8386.
(20) Seinfeld, J. H.; Pankow, J. F. Organic Atmosphric Particulate
Material. Annu. Rev. Phys. Chem. 2003, 54, 121−140.
(21) Lee, S.; Kamens, R. M. Particle Nucleation from the Reaction of
α-Pinene and O3. Atmos. Environ. 2005, 39, 6822−6832.
(22) Atkinson, R.; Baulch, D. L.; Cox, R. A.; Crowley, J. N.;
Hampson, R. F.; Hynes, R. G.; Jenkin, M. E.; Rossi, M. J.; Troe, J.
Evaluated Kinetic and Photochemical Data for Atmospheric
Chemistry: Volume II, Gas Phase Reactions of Organic Species.
Atmos. Chem. Phys. 2006, 6, 3625−4055.
REFERENCES
■
(1) Johnson, D.; Marston, G. The Gas-Phase Ozonolysis of
Unsaturated Volatile Organic Compounds in the Troposphere.
Chem. Soc. Rev. 2008, 37, 699−716.
(2) Su, F.; Calvert, J. G.; Shaw, J. H. A FT IR Spectroscopic Study of
the Ozone-Ethene Reaction Mechanism in Oxygen-Rich Mixtures. J.
Phys. Chem. 1980, 84, 239−246.
(23) Atkinson, R. Kinetics of the Gas-Phase Reactions of OH
Radicals with Alkanes and Cycloalkanes. Atmos. Chem. Phys. 2003, 3,
2233−2307.
(3) Herron, J. T.; Huie, R. E. Stopped-Flow Studies of the
Mechanisms of Ozone-Alkene Reactions in the Gas Phase. Ethylene.
J. Am. Chem. Soc. 1977, 99, 5430−5435.
(24) Horie, O.; Moortgat, G. K. Gas-Phase Ozonolysis of Alkenes.
Recent Advances in Mechanistic Investigations. Acc. Chem. Res. 1998,
31, 387−396.
(4) Horie, O.; Moortgat, G. K. Decomposition Pathways of the
Excited Criegee Intermediates in the Ozonolysis of Simple Alkenes.
Atmos. Environ. 1991, 25, 1881−1896.
(25) Grosjean, E.; Grosjean, D. Carbonyl Products of the Gas Phase
Reaction of Ozone with Symmetrical Alkenes. Environ. Sci. Technol.
1996, 30, 2036−2044.
(5) Neeb, P.; Horie, O.; Moortgat, G. K. The Ethene−Ozone
Reaction in the Gas Phase. J. Phys. Chem. A 1998, 102, 6778−6785.
(6) Wolff, S.; Boddenberg, A.; Thamm, J.; Turner, W. V.; Gab, S.
Gas-Phase Ozonolysis of Ethene in the Presence of Carbonyl-Oxide
Scavengers. Atmos. Environ. 1997, 31, 2965−2969.
̈
(26) Klotz, B.; Barnes, I.; Imamura, T. Product Study of the Gas-
Phase Reactions of O3, OH and NO3 Radicals with Methyl Vinyl
Ether. Phys. Chem. Chem. Phys. 2004, 6, 1725−1734.
(7) Hasson, A. S.; Orzechowska, G.; Paulson, S. E. Production of
Stabilized Criegee Intermediates and Peroxides in the Gas Phase
Ozonolysis of Alkenes: 1. Ethene, trans-2-Butene, and 2,3-Dimethyl-2-
butene. J. Geophys. Res. 2001, 106, 34131−34142.
(8) Neeb, P.; Horie, O.; Moortgat, G. K. The Nature of the
Transitory Product in the Gas-Phase Ozonolysis of Ethene. Chem.
Phys. Lett. 1995, 246, 150−156.
(27) Sadezky, A.; Chaimbault, P.; Mellouki, A.; Rompp, A.;
̈
Winterhalter, R.; Le Bras, G.; Moortgat, G. K. Formation of Secondary
Organic Aerosol and Oligomers from the Ozonolysis of Enol Ethers.
Atmos. Chem. Phys. 2006, 6, 5009−5024.
(28) Sadezky, A.; Winterhalter, R.; Kanawati, B.; Rompp, A.;
̈
Spengler, B.; Mellouki, A.; Le Bras, G.; Chaimbault, P.; Moortgat,
G. K. Oligomer Formation during Gas-Phase Ozonolysis of Small
Alkenes and Enol Ethers: New Evidence for the Central Role of the
Criegee Intermediate as Oligomer Chain Unit. Atmos. Chem. Phys.
2008, 8, 2667−2699.
(9) Alam, M. S.; Camredon, M.; Rickard, A. R.; Carr, T.; Wyche, K.
P.; Hornsby, K. E.; Monks, P. S.; Bloss, W. J. Total Radical Yields from
Tropospheric Ethene Ozonolysis. Phys. Chem. Chem. Phys. 2011, 13,
11002−11015.
(29) Vereecken, L.; Harder, H.; Novelli, A. The Reaction of Criegee
Intermediates with NO, RO2, and SO2, and Their Fate in the
Atmosphere. Phys. Chem. Chem. Phys. 2012, 14, 14682−14695.
(30) Tobias, H. J.; Ziemann, P. J. Thermal Desorption Mass
Spectrometric Analysis of Organic Aerosol Formed from Reactions of
1-Tetradecene and O3 in the Presence of Alcohols and Carboxylic
Acids. Environ. Sci. Technol. 2000, 34, 2105−2115.
(31) Sato, K.; Hatakeyama, S.; Imamura, T. Secondary Organic
Aerosol Formation during the Photooxidation of Toluene: NOx
Dependence of Chemical Composition. J. Phys. Chem. A 2007, 111,
9796−9808.
(32) Sato, K.; Inomata, S.; Xing, J.-H.; Imamura, T.; Uchida, R.;
Fukuda, S.; Nakagawa, K.; Hirokawa, J.; Okumura, M.; Tohno, S.
Effect of OH Radical Scavengers on Secondary Organic Aerosol
Formation from Reactions of Isoprene with Ozone. Atmos. Environ.
2013, 79, 147−154.
(10) Thamm, J.; Wolff, S.; Turner, W. V.; Gab, S.; Thomas, W.;
̈
Zabel, F.; Fink, E. H.; Becker, K. H. Proof of the Formation of
Hydroperoxymethyl Formate in the Ozonolysis of Ethene: Synthesis
and FT-IR Spectra of the Authentic Compound. Chem. Phys. Lett.
1996, 258, 155−158.
(11) Neeb, P.; Horie, O.; Moortgat, G. K. Gas-Phase Ozonolysis of
Ethene in the Presence of Hydroxylic Compounds. Int. J. Chem. Kinet.
1996, 28, 721−730.
(12) Copeland, G.; Ghosh, M. V.; Shallcross, D. E.; Percival, C. J.;
Dyke, J. M. A Study of the Ethene-Ozone Reaction with Photoelectron
Spectroscopy: Measurement of Product Branching Ratios and
Atmospheric Implications. Phys. Chem. Chem. Phys. 2011, 13,
14839−14847.
(13) Qi, B.; Sato, K.; Imamura, T.; Takami, A.; Hatakeyama, S.; Ma,
Y. Production of the Radicals in the Ozonolysis of Ethene: A Chamber
Study by FT-IR and PERCA. Chem. Phys. Lett. 2006, 427, 461−465.
12920
dx.doi.org/10.1021/jp408672m | J. Phys. Chem. A 2013, 117, 12912−12921

The Journal of Physical Chemistry A
Article
(33) Lee, M.; Heikes, B. G.; O’Sullivan, D. W. Hydrogen Peroxide
and Organic Hydroperoxide in the Troposphere: a Review. Atmos.
Environ. 2000, 34, 3475−3494.
(34) Muller, L.; Reinnig, M. C.; Warnke, J.; Hoffmann, T.
̈
Unambiguous Identification of Esters as Oligomers in Secondary
Organic Aerosol Formed from Cyclohexene and Cyclohexene/α-
Pinene Ozonolysis. Atmos. Chem. Phys. 2008, 8, 1423−1433.
(35) Camredon, M.; Hamilton, J. F.; Alam, M. S.; Wyche, K. P.; Carr,
T.; White, I. R.; Monks, P. S.; Rickard, A. R.; Bloss, W. J. Distribution
of Gaseous and Particulate Organic Composition during Dark α-
Pinene Ozonolysis. Atmos. Chem. Phys. 2010, 10, 2893−2917.
(36) Yasmeen, F.; Vermeylen, R.; Szmigielski, R.; Iinuma, Y.; Boge,
̈
O.; Herrmann, H.; Maenhaut, W.; 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.
2010, 10, 9383−9392.
(37) Gao, Y.; Hall, W. A.; Johnston, M. V. Molecular Composition of
Monoterpene Secondary Organic Aerosol at Low Mass Loading.
Environ. Sci. Technol. 2010, 44, 7897−7902.
(38) Kristensen, K.; Enggrob, K. L.; King, S. M.; Worton, D. R.; Platt,
S. M.; Mortensen, R.; Rosenoern, T.; Surratt, J. D.; Bilde, M.;
Goldstein, A. H.; et al. Formation and Occurrence of Dimer Esters of
Pinene Oxidation Products in Atmospheric Aerosols. Atmos. Chem.
Phys. 2013, 13, 3763−3776.
(39) Bonn, B.; Schuster, G.; Moortgat, G. K. Influence of Water
Vapor on the Process of New Particle Formation during Monoterpene
Ozonolysis. J. Phys. Chem. A 2002, 106, 2869−2881.
(40) Heaton, K. J.; Dreyfus, M. A.; Wang, S.; Johnston, M. V.
Oligomers in the Early Stage of Biogenic Secondary Organic Aerosol
Formation and Growth. Environ. Sci. Technol. 2007, 41, 6129−6136.
(41) Tolocka, M. P.; Heaton, K. J.; Dreyfus, M. A.; Wang, S.; Zordan,
C. A.; Saul, T. D.; Johnston, M. V. Chemistry of Particle Inception and
Growth during α-Pinene Ozonolysis. Environ. Sci. Technol. 2006, 40,
1843−1848.
12921
dx.doi.org/10.1021/jp408672m | J. Phys. Chem. A 2013, 117, 12912−12921