An experimental study of the gas-phase reactions of no3 radicals with a series of unsaturated aldehydes: Trans -2-hexenal, trans -2-heptenal, and trans -2-octenal

Article abstract of DOI:10.1021/jp3071234

Rate constants for the gas-phase reactions of the NO₃ radical with a series of unsaturated aldehydes, trans-2-hexenal, trans-2-heptenal, and trans-2-octenal, have been measured using absolute rate method at 294 ± 3 K and atmospheric pressure. This work was performed to clarify discrepancies found in the literature and thus led to a clearer view of the effect of the increasing carbon chain length on the reactivity of trans-2-alkenals. The rate constants were determined to be (4.7 ± 1.5) × 10^-15, (5.3 ± 1.6) × 10^-15, and (5.6 ± 2.3) × 10^-15 cm³ molecule^-1 s^-1 for trans-2-hexenal, trans-2-heptenal, and trans-2-octenal, respectively. These results clearly indicate that the carbon chain lengthening of the trans-2-alkenals does not significantly affect the rate constant. In addition, the mechanism for the reaction of NO₃ with these unsaturated aldehydes was also investigated. Unsaturated peroxynitrate-type compounds that are exclusively formed through the abstraction channel were observed as the main products.

Full text of DOI:10.1021/jp3071234

Article
pubs.acs.org/JPCA
An Experimental Study of the Gas-Phase Reactions of NO3 Radicals
with a Series of Unsaturated Aldehydes: trans-2-Hexenal, trans-2-
Heptenal, and trans-2-Octenal
Jamila Kerdouci, Benedicte Picquet-Varrault,* Regine Durand-Jolibois, Cecile Gaimoz,
́
́
́
́
and Jean-Franco̧ is Doussin
Laboratoire Interuniversitaire des System
Paris Diderot (UPD), Institut Pierre Simon Laplace, 61 Avenue du General de Gaulle, 94010 cedex Creteil, France
̀
es Atmospher
́
iques, UMR CNRS 7583, Universite
́
Paris Est Cret
́
eil (UPEC) et Universite
́
́
́
́
S
* Supporting Information
ABSTRACT: Rate constants for the gas-phase reactions of the NO3
radical with a series of unsaturated aldehydes, trans-2-hexenal, trans-2-
heptenal, and trans-2-octenal, have been measured using absolute rate
method at 294
3 K and atmospheric pressure. This work was
performed to clarify discrepancies found in the literature and thus led to
a clearer view of the effect of the increasing carbon chain length on the
reactivity of trans-2-alkenals. The rate constants were determined to be
(4.7 1.5) × 10−15, (5.3 1.6) × 10−15, and (5.6 2.3) × 10−15 cm3
molecule−1 s−1 for trans-2-hexenal, trans-2-heptenal, and trans-2-octenal,
respectively. These results clearly indicate that the carbon chain
lengthening of the trans-2-alkenals does not significantly affect the rate
constant. In addition, the mechanism for the reaction of NO3 with these
unsaturated aldehydes was also investigated. Unsaturated peroxynitrate-
type compounds that are exclusively formed through the abstraction channel were observed as the main products.
unsaturated aldehydes. Several kinetic studies have been carried
out for the reactions of C4−C7 trans-2-alkenals with nitrate
radicals22−26 and are reported in Figure 1. One can observe that
INTRODUCTION
■
Carbonyl compounds are emitted into the atmosphere from
anthropogenic and biogenic sources. They are emitted by the
incomplete combustion of fuels.1−4 Moreover, in addition to
the emissions of isoprene and terpenes, a large variety of
oxygenated compounds have been observed as plant
emissions.5,6 This is in particular the case of C6 oxygenates,7−9
such as trans-2-hexenal (also called leaf aldehyde), which are
emitted from wounded leaves. Other aldehydes such as trans-2-
heptenal and trans-2-octenal have also been identified in the
emissions of plants.10−13 Besides these primary sources,
carbonyl compounds are also formed in the atmosphere from
the photo-oxidation of organic compounds. The oxidation of
terpenes and conjugated dienes produces a number of
unsaturated carbonyl compounds.14−16 For these reasons,
unsaturated aldehydes play a key role in the tropospheric
chemistry as a source of HOx radicals and as possible
intermediates in the formation of secondary organic aero-
sols.17,18
These unsaturated aldehydes are mainly removed from the
atmosphere by photolysis and by the gas-phase reaction with
OH and NO3 radicals and O3.16,17,19 Ozone and NO3 radical
are known to be the main oxidants of VOC during the
night.20,21 In particular, NO3 oxidation of unsaturated VOC,
which proceeds mainly by an electrophilic addition on the
double bond, is often a fast reaction. Consequently, oxidation
by NO3 is expected to be a significant loss process of
Figure 1. Rate constants for gas-phase reactions of NO3 radical with a
series of trans-2-alkenals (in logarithmic scale).
Received: July 18, 2012
Revised: September 16, 2012
Published: September 24, 2012
© 2012 American Chemical Society
10135
dx.doi.org/10.1021/jp3071234 | J. Phys. Chem. A 2012, 116, 10135−10142

The Journal of Physical Chemistry A
Article
there are very large discrepancies between published data.
distillation and could be kept for several weeks in the cold
trap under vacuum.
Indeed, the values reported by Cabanas et al.24 reveal a strong
̃
increase in the rate constant with increasing alkyl chain length,
whereas the values published by Atkinson et al.22 and more
recently by Zhao et al.27 do not confirm this large increase in
the rate constants.
All experiments were performed in the dark at 294 3 K and
at atmospheric pressure with synthetic air (80% N2 + 20% O2).
In a first step, the trans-2-alkenal was injected into the chamber,
and its concentration was measured for approximately one hour
to check for adsorption on the chamber walls. Initial
concentrations of trans-2-alkenals were (3.3−8.4) × 1013
molecules cm−3. Then, N2O5 was introduced by stepwise
injections in order to ensure measurable consumption of the
VOC on a reasonable time scale. For each experiment, a total,
of two to three additions of N2O5 were performed (each
addition corresponding to an initial concentration of NO3 in
the chamber of approximately (3−40) × 1010 molecules cm−3).
The mixing of N2O5 in the reactor is ensured by a mixing pipe
(a pyrex line equipped with many small radial holes which is
installed all across the simulation chamber). N2O5 is introduced
and flushed with N2 as carrier gas into the line that allows
introducing it in the whole length of the reactor. The efficiency
of this mixing system has been checked in the past and allows a
good mixing at the few seconds’ scale which is very fast in
comparison to the duration of visible and infrared spectra
(respectively 1 min and 3−4 min). During the course of the
experiment, time-resolved concentrations of organic reactants
and products were monitored from their infrared spectral
absorptions. Integrated band intensities (IBIs) used to quantify
these species are: IBItrans‑2‑hexenal (1190−1115 cm−1) = (2.6
0.3) × 10−18 cm/molecule, IBItrans‑2‑heptenal (1015−945 cm−1) =
(2.4 0.1) × 10−18 cm/molecule, and IBItrans‑2‑octenal (1025−
930 cm−1) = (3.6 0.9) × 10−18 cm/molecule (base e). NO3
was monitored from its visible absorption using the cross
section of its main absorption band at 662 nm equal to 2.1 ×
10−17 cm2/molecule (base e) which is the value recommended
by IUPAC.32 UV−visible spectra were treated using DOASIS
software developed by IUP Heidelberg group (https://doasis.
iup.uni-heidelberg.de/bugtracker/projects/doasis/)). Figure S1
of the Supporting Information shows a typical plot of an
experiment where concentrations of reactants (trans-2-hexenal
and NO3), products (peroxynitrate (PAN) analogues and CO),
and NO2 were monitored by IR and UV−visible spectrometric
devices. Unfortunately, due to a saturation of its absorption
bands, N2O5 could not be monitored during the experiments.
The sources of chemicals used and their purities were: trans-
2-hexenal (>99%) from Acros Organics; trans-2-heptenal
(>97%) from Sigma-Aldrich; trans-2-octenal (>94%) from
Sigma-Aldrich. Dry synthetic air was generated using N2 (from
liquid nitrogen evaporation, purity >99.995%, H2O < 5 ppm,
MESSER) and O2 (Alphagaz 1, purity > 99.995%, H2O < 3
ppm, CnHm < 0.5 ppm, Air Liquide).
Thus in this work, we have performed kinetic and
mechanistic studies for the gas-phase reactions of NO3 radicals
with trans-2-hexenal (trans-CH3(CH2)2CHCHCH(O)),
trans-2-heptenal (trans-CH3(CH2)3CHCHCH(O)), and
trans-2-octenal (trans-CH3(CH2)4CHCHCH(O))
NO3 + trans‐2‐hexenal → products k1
(1)
NO3 + trans‐2‐heptenal → products k2
(2)
NO3 + trans‐2‐octenal → products k3
(3)
The aim was here to determine the effect of the alkyl chain
length on the reactivity of the trans-2-alkenals. This study on a
series of unsaturated aldehydes is also essential for the
development of reliable structure activity relationships (SAR)
for this class of compounds. In particular, it will be used to
complete the SAR which was developed recently in our
group.28
EXPERIMENTAL SETUP AND PROCEDURES
■
Experiments were carried out in an evacuable chamber,
consisting of a Pyrex reactor of 977 L coupled to two multiple
reflection optical systems interfaced to a FTIR spectrometer
and a UV−visible grating spectrometer. The main features of
this chamber and its spectroscopic devices have previously been
described;29,30 thus only the relevant details are given here. The
quantification of the organic compounds in the reactor was
performed by using FTIR spectroscopy. The FTIR spectrom-
eter is a BOMEM DA8-ME equipped with a glowbar as mid-
infrared source. Infrared spectra were acquired by averaging
100 scans at a resolution of 0.5 cm−1 and with an optical path
length of 204 m. For absolute rate determination, NO3 was
monitored by its visible absorption at 662 nm using UV−visible
spectrometry. This spectrometer consists of a source (high-
pressure xenon arc lamp, Osram XBO, 450WXe UV), a
multipass optical system inside the reactor (path length of 72
m), a monochromator (HR 320 Jobin-Yvon), and a 1024 × 58
pixel charge-coupled device (CCD) camera (CCD 3000, Jobin-
Yvon) as a detector. In this configuration, the CCD covers a
spectral range of ca. 60 nm, and the instrument has a maximum
resolution of 0.15 nm.
NO3 radicals were generated by thermal decomposition of
N2O5,31 which was synthesized in a vacuum line by the reaction
of NO2 with excess of O3 according to reaction 4 and followed
by reaction 5
RESULTS AND DISCUSSION
■
Kinetic Study. A minimum of three experiments were
performed for each aldehyde. The NO3 oxidation rate constants
were obtained by using the absolute rate technique. If it is
assumed that the loss of the trans-2-alkenal is due solely to the
reaction 6 with nitrate radicals
NO2 + O3 → NO3 + O2
(4)
NO3 + NO2 ↔ N2O5
(5)
In the first stage of the synthesis, NO2 was trapped at 193 K
in a cold tube. Ozone was then flushed onto the NO2 crystals,
whereas the cold trap was moved to a second tube where
formed dinitrogen pentoxide crystals were collected. At the end
of the synthesis, the bulb containing dinitrogen pentoxide
crystals was evacuated down to remove remaining NO2 at room
temperature. N2O5 was hence purified by trap-to-trap
trans‐2‐alkenal + NO3 → products
(6)
the second-order kinetic equation for reaction 6 is
‐[trans‐2‐alkenal]
= k6[trans‐2‐alkenal][NO3]
(I)
dt
10136
dx.doi.org/10.1021/jp3071234 | J. Phys. Chem. A 2012, 116, 10135−10142

The Journal of Physical Chemistry A
Article
This equation can be approximated for small time intervals
−Δ[trans‐2‐alkenal] = k6[trans‐2‐alkenal][NO3]Δt
(II)
Where −Δ[trans-2-alkenal] corresponds to the consumption
of trans-2-alkenal during the time interval Δt and [trans-2-
alkenal] and [NO3] are averaged concentrations during this
interval. By plotting −Δ[trans-2-alkenal] vs [trans-2-alkenal] ×
[NO3] × Δt, a straight line with a slope corresponding to k6 is
obtained. Data from all experiments are exhibited in Figures
2−4 for trans-2-hexenal, trans-2-heptenal, and trans-2-octenal,
Figure 4. Absolute rate plot for the gas-phase reaction of trans-2-
octenal with NO3 radicals. Different symbols indicate independent
experiments.
Table 1. Rate Constants for the Oxidation of a Series of
Unsaturated Aldehydes by the NO3 Radical
kNO × 1014
3
compound
trans-2-
hexenal
(cm3 molecule−1 s−1
)
method
relative rate Zhao et al.27
ref
1.36 0.29
5.49 0.95
absolute rate Cabanas et
̃
al.24
Figure 2. Absolute rate plot for the gas-phase reaction of trans-2-
hexenal with NO3 radicals. Different symbols indicate independent
experiments.
1.21 0.44
relative rate Atkinson et
al.22
0.47 0.15
absolute
rate
this work
trans-2-
heptenal
2.31 0.36
9.59 0.19
relative rate Zhao et al.27
absolute rate Cabanas et
̃
al.24
0.53 0.16
0.56 0.23
absolute
rate
this work
this work
trans-2-
octenal
absolute
rate
Table 1 summarizes the rate constants measured by this work
with data from literature for the reactions of the NO3 radical
with the series of trans-2-alkenals. The rate constant for trans-2-
hexenal has already been measured by three studies. Two of
them are relative rate studies (Atkinson et al.22 and Zhao et
al.27). The third one is an absolute determination in a fast-flow
discharge system using laser-induced fluorescence (LIF) for the
NO radical detection (Cabanas et al.24). Concerning the trans-
̃
3
2-heptenal, two values have been published by Cabanas et al.24
Figure 3. Absolute rate plot for the gas-phase reaction of trans-2-
heptenal with NO3 radicals. Different symbols indicate independent
experiments.
̃
and Zhao et al.27 For trans-2-octenal, this work provides the
first experimental determination.
Our k1 value differs significantly from those obtained by the
previous determinations by factors from 3 to 12. Similarly, our
k2 value obtained for trans-2-heptenal is also much lower than
respectively. It can be observed that all individual experiments
are in good agreement. As a consequence, all data points were
combined, and a linear regression, which takes into account
errors on both abscissa and ordinate scales, was performed
using the program developed by Brauers and Finlayson-Pitts.33
The errors on both abscissa and ordinate scales were estimated
from the error on NO3 and trans-2-alkenal concentrations,
which were taken as 7 × 109 and 7.38 × 1011 molecules cm−3,
respectively. These errors correspond to the uncertainty on the
estimation of band intensities in visible and infrared spectra.
The uncertainties on NO3 cross sections and on Δt were
considered to be negligible here and thus were not taken into
account. The quoted uncertainties on the rate constants are the
2σ, where σ is the standard deviation on the linear regression.
those measured by Zhao et al.27 and Cabanas et al.24 by factors
̃
of 4 and 18, respectively.
Moreover, it can be noticed that the rate constants reported
by Cabanas et al.24 are systematically higher than the other
̃
determinations. In their experiments, the rate constants were
determined by monitoring the decay of NO3 radical and trans-
2-alkenals in a flow-tube system. However the studies of
Ullerstam et al.34 and D’anna et al.35 on the reactivity of
saturated aldehydes have shown that secondary reactions of
NO3 radicals with RCO radicals (produced after H-atom
abstraction from the −CHO groups) occurs with this type of
experimental devices and thus may lead to higher rate
10137
dx.doi.org/10.1021/jp3071234 | J. Phys. Chem. A 2012, 116, 10135−10142

The Journal of Physical Chemistry A
Article
constants. Moreover as suggested by Zhao et al.,27 heteroge-
neous wall reactions involving the VOC might have occurred.
In our experiments, rate constants were determined by
monitoring the decay of the VOC, which may be less subject
to secondary reactions than the NO3 radical (which reacts with
many other species in the mixture). In addition, it was checked
that aldehydes were stable in the chamber in absence of NO3
radicals. These hypotheses may explain the large discrepancies
Table 2. Comparison of Rate Constants for the Reaction of
NO3 Radical with trans-2-Alkenals Obtained in This Work
with Those for the Corresponding Alkenes and Saturated
Aldehydes
k
NO3
compound (cm3 molecule−1 s−1
)
ref
trans-2-alkenals
trans-2-
hexenal
(4.7 1.5) × 10−15 this work
(5.3 1.6) × 10−15 this work
(5.6 2.3) × 10−15 this work
observed with the data from Cabanas et al.24 However, the
̃
trans-2-
heptenal
reasons for the disagreements with the relative rates studies of
Zhao et al.27 and Atkinson et al.22 can not be explained by this
secondary chemistry. To identify the origin of these
discrepancies, we have carefully looked for potential sources
of artifacts in our experiments. In particular, we have checked
that the “low” resolution of our UV−visible spectrometer did
not induce an error in the quantification of NO3. To do that,
concentrations obtained with DOASIS software were compared
to those obtained by calculating the spectra in absorbance
(ln(I0/I), integrating the main absorption peak of NO3 at 662
nm, and deducing the concentration from the Integrated Band
Intensity of NO3. We observed that the concentrations
obtained by the two methods were the same within
uncertainties.
trans-2-
octenal
corresponding
saturated
hexanal
(1.7 0.1) × 10−14 Noda et al.44
(1.6 0.2) × 10−14 Noda et al.44
(1.5 0.2) × 10−14 D’Anna et al.45
aldehydes
(1.8 0.4) × 10−14 Cabanas et al.46
̃
(1.5 0.1) × 10−14 Papagni et al.47
(1.7 0.2) × 10−14 D’Anna et al.48
(1.8 0.1) × 10−14 Noda et al.44
(1.9 0.2) × 10−14 Noda et al.44
(2.1 0.3) × 10−14 Noda et al.44
heptanal
octanal
(2.4 0.4) × 10−14 Cabanas et al.46
̃
(1.5 0.2) × 10−14 Noda et al.44
(2.0 0.3) × 10−14 Noda et al.44
In conclusion, we did not identify any artifact in our
experiments and the reason of the disagreement between our
data and those of Zhao et al.27 and Atkinson et al.22 remains
unclear.
corresponding
alkenes
trans-2-
alkenes
3.9 × 10−13
Pfrang et al.36
and IUPAC32
Considering the rate constants obtained in this work, we can
deduce that the alkyl chain length of trans-2-alkenals has little
or no effect on the rate constants.
the reactivity of both groups. Electronic structure calculations
would be helpful to clarify the cause of these deactivating
effects.
Effect of the Structure on the Reactivity. The reaction
of NO3 radicals with unsaturated aldehydes can proceed by two
main pathways: an H-atom abstraction from the carbonyl group
and an addition to the double bond. It is considered here that
the H-abstraction on the alkyl chain is negligible. It is thus
interesting to compare the reactivity of the unsaturated
carbonyls toward NO3 radicals with the corresponding alkenes
and saturated aldehydes when available (see Table 2) to
determine the effect of each group on the reactivity of the other
group. For the parent alkenes, no data is available for trans-2-
heptene and trans-2-octene. Only rate constants for smaller
trans-2-alkenes (trans-2-butene, trans-2-pentene, and trans-2-
hexene) have been published. Nevertheless, since no significant
trend was observed along the series trans-2-butene, trans-2-
pentene, and trans-2-hexene,32,36 a single rate constant of 3.9 ×
10−13 cm3 molecule−1 s−1 has been considered for the trans-2-
alkenes. By confronting this value with the rate constants for
the unsaturated aldehydes, it can be observed that alkenes are
much more reactive (by 2 orders of magnitude) than the
alkenals. This suggests that the carbonyl group strongly
deactivates the addition on the double bond. This could be
explained by the electron-withdrawing effect of the carbonyl
group in α position.
Moreover, depending on the rate constants, which were
chosen for the trans-2-alkenals (this work or previous studies),
it can be observed that the rate constants for the unsaturated
aldehydes are twice as small or similar to those of the
corresponding saturated aldehydes suggesting that the double
bond may slightly deactivate the H-abstraction on the carbonyl
group. This effect has already been observed by Ullerstam et
al.,26 in their study on the reactivity of crotonaldehyde with
NO3 radicals. Therefore this comparison clearly shows that the
association of an aldehyde function and a double bond lowers
Product Analysis. Product studies were also carried out for
the trans-2-alkenals. An example of the FTIR data for one
typical experiment on the reactivity of trans-2-hexenal with
NO3 radical is given in Figure 5. One can observe the quite
complex nature of the IR spectra obtained (see spectrum b).
However the absorption bands of the trans-2-alkenal between
900 and 1200 cm−1 remain identifiable and allow us to quantify
it precisely.
The residual spectrum c obtained after subtraction of
characteristic bands of reactants, H2O, N2O5, NO2, and
HNO3, allows the product identification. The absorption
band between 2000 and 2200 cm−1 indicates the formation
of CO. This spectrum also shows several intense absorption
bands at around 796 cm−1 (NO2 deformation), 1300 cm−1
(NO2 symmetric stretch), and 1740 cm−1 (NO2 asymmetric
stretch), which are characteristic of PAN compounds. More-
over, an absorption band around 1820 cm−1 corresponding to
CO stretching modes in peroxyacyl nitrates has also been
observed. These absorption bands can be attributed to an
unsaturated PAN analogue. This observation is supported by
the comparison with Figure 5d, which corresponds to the
supposed crotoperoxynitrate (CPAN) spectrum obtained by
Orlando et al.,37 as these two spectra have several similar bands
around 1080, 1300, 1740, and 1820 cm−1. In addition, a weak
absorption at 845 cm−1, which is characteristic of alkyl nitrates
and dinitrates, has been observed in the residual spectra for the
experiments on trans-2-heptenal and trans-2-octenal. However,
the absorption band was so weak that it could not be quantified.
This absorption feature was not observed for trans-2-hexenal.
The formation of formic acid is also observed. The origin of
this minor product is not known, but may be heterogeneous.
Complementary analyses of carbonyl compounds were
performed by sampling the mixtures in impingers containing
10138
dx.doi.org/10.1021/jp3071234 | J. Phys. Chem. A 2012, 116, 10135−10142

The Journal of Physical Chemistry A
Article
Figure 5. Spectrum (a) is a standard spectrum of trans-2-hexenal. Spectrum (b) was obtained after the beginning of the experiment. Spectrum (c)
shows a typical product obtained after subtraction of the spectral features belonging to water, trans-2-hexenal, N2O5, NO2, and HNO3. Spectrum (d)
corresponds to the presume spectrum of CPAN obtained by Orlando et al.37
2,4-dinitrophenylhydrazine solution according to the procedure
described by Ferrari et al.38 The chemical derivatization
products were analyzed by high-performance liquid chromatog-
raphy (HPLC) coupled to UV detection at 360 nm. This
technique gives access to a detection limit of ca. 10 ppb for
most carbonyls for a sampling volume of 10 L used here. For
trans-2-hexenal, we did not observe any formation of saturated
carbonyl compounds during the course of the experiment. For
the trans-2-heptenal and trans-2-octenal, several peaks have
been observed on the chromatograms. Comparison with
standards allows us to suspect the formation of acetaldehyde.
However, its formation yield could not precisely be calculated
and was estimated to be below 3%. Some other peaks which
have been detected remain unidentified, but their intensities are
very weak suggesting that they might be very minor products.
Despite the fact that no standard spectra of PAN analogues
were available, we can estimate their concentration by
considering that the IBIs at 796 and 1300 cm−1 of our PAN
analogues are equal to those determined by Monedero et al.39
for the CPAN. Therefore, for the detected products that could
be quantified (CO and PAN analogues), yields were derived
from initial slopes of plots [product] vs −Δ[trans-2-alkenal]
and are reported in Table 3. Plots of the concentration of these
compounds as a function of trans-2-alkenal consumption are
given in Supporting Information. The quoted errors take into
account the uncertainty in the linear regression (2), the
uncertainties in reactant and products IR absorption bands
calibration and those in the quantification of these bands in the
“mixture” spectra. For comparison, Table 3 also lists the result
of mechanistic studies carried out for others α,β-unsaturated
aldehydes, i.e., acrolein, methacrolein, and crotonaldehyde.25,40
PAN and CO yields for the reaction of NO3 with the trans-2-
Table 3. Product Yields Derived from Initial Slopes in the
trans-2-Alkenals + NO3 Reactions
a
reaction
product
APAN
yield (%)
47
ref
acrolein + NO3
3
Salgado et al.25
CO
20.9 1.9
31.6 2.0
HCHO
methacrolein + NO3 MPAN
CO
observed
Canosa-Mas et
al.40
observed
observed
methylglyoxal
CPAN
crotonaldehyde +
NO3
93.6 4.3 Salgado et al.25
8.3 1.1
CO
trans-2-hexenal +
NO3
PAN analogue
CO
100 25
6.4 4.2
this work
formic acid
PAN analogue
CO
observed
trans-2-heptenal +
NO3
64 12
this work
5.3 1.8
organic nitrates observed
formic acid
PAN analogue
CO
observed
66 26
4.9 3.3
trans-2-octenal +
NO3
This work
organic nitrates observed
formic acid observed
a
Product yields for acrolein, methacrolein, and crotonaldehyde are
included for comparison.
respectively. Thereby for these two compounds, all the
products have not been identified. Finally, it can be deduced
from this study that PAN-type compounds were the main
oxidation products for the three trans-2-alkenals. Furthermore,
it is interesting to notice that these resulting PAN analogues are
stable at room temperature, since no significant decrease in the
absorption bands was observed during the experiment but also
during a period of one hour after the end of the reaction.
Moreover, for some experiments, several ppm of NO were
added to the mixture at the end of the reaction in order to
promote the decomposition of these PAN, but it has no effect
on it (with NO/NO2 concentration ratios including between 1
and 2).
hexenal are found to be 100
25% and 6.4
4.2%,
respectively. The carbon balance is thus complete for this
compound. However, it is not the case for the reaction of NO3
radicals with the trans-2-heptenal and trans-2-octenal. Indeed
for the trans-2-heptenal (trans-2-octenal), PAN and CO yields
are of 64 12% (66 26%) and 5.3 1.8% (4.9 3.3%),
10139
dx.doi.org/10.1021/jp3071234 | J. Phys. Chem. A 2012, 116, 10135−10142

The Journal of Physical Chemistry A
Article
As mentioned above, the reaction of NO3 with trans-2-
alkenals is expected to proceed mainly by both NO3 addition to
the double bond and H-abstraction of the aldehydic hydrogen,
the abstraction from the alkyl group being negligible. The
formation of a PAN analogue can easily be explained by the
abstraction of the aldehydic hydrogen (see channel 3 of Scheme
1). On the contrary, we did not find any mechanism that could
formed by the channel 6 and could probably result from the
addition channel.
From these results, the branching ratio of the abstraction
pathway can be estimated as follow: 1
0.25 for trans-2-
hexenal > 0.64 for trans-2-heptenal and >0.66 for trans-2-
octenal. The observations made here are similar to the one
made by Salgado et al.25 in their study on the reactivity of
crotonaldehyde and acrolein with NO3 radicals. Indeed these
latter authors mainly detect the formation of products resulting
from the abstraction channel (see Table 3) and in particular
APAN and CPAN. Canosa-Mas et al.40 have studied the
reactivity of methacrolein with NO3 radicals. Even if the
authors could not provide formation yields, they also observed
that MPAN (methacryloylperoxynitrate) is a major product.
These conclusions are quite unexpected considering the
significant reactivity of the substituted double bonds compared
to the −CHO function. Indeed, according to Table 2, trans-2-
alkenes are up to 20 times more reactive than saturated
aldehydes. Thereby all these studies confirm the strong
deactivating effect of the aldehydic group on the double bond
since the main reaction pathway is the abstraction of the
aldehydic hydrogen. Notwithstanding, to confirm and explain
these observations further chamber experiments (including the
study of the kinetic isotope effect) and computational studies
would be useful.
Scheme 1. Proposed Mechanism for the H-Atom Abstraction
from the Carbonyl Group on Aliphatic Unsaturated
Aldehydes
CONCLUSIONS AND ATMOSPHERIC
IMPLICATIONS
■
Rate constants for three unsaturated aldehydes have been
measured using an absolute rate method. The rate constants
obtained in this study led to a significant re-evaluation of
previously published values. The rate constants measured here
for the trans-2-hexenal and trans-2-heptenal are in particular
much lower than those obtained by Cabanas et al.24 using
̃
absolute rate technique and to a lesser extent to those by
relative rate technique.22,27
Thereby, our study reveals that the alkyl chain of the trans-2-
alkenals has no significant effect on their reactivity with the
nitrate radical. The aldehydic function appears to decrease the
reactivity of the neighboring double bond, and unexpectedly, a
double bond appears also to lower the reactivity of the
aldehydic hydrogen atom. Thus, this work provides crucial
information for the extension to the aldehydes of the SAR
developed recently in our group.28
Furthermore the mechanisms of the reactions between the
nitrate radical and the three trans-2-alkenals have been studied,
and the PAN-type compounds have been identified as major
products. Therefore this work indicates that the abstraction of
the aldehydic hydrogen is the dominant pathway at room
temperature for the NO3 radical reactions with these three
unsaturated carbonyl compounds.
The typical tropospheric lifetimes of the trans-2-alkenals have
been estimated using the rate coefficients determined in this
study and the rate constants for their reactions with OH and O3
obtained from literature. Typical tropospheric concentrations
were used for the oxidants: 1.6 × 106 molecules cm−3 for OH
radicals,41 5 × 108 molecules cm−3 for NO3 radicals,42 and 7 ×
1011 molecules cm−3 for ozone.43 These lifetimes are given in
Table 4. It can clearly be seen that tropospheric lifetimes
toward OH radicals, being on the order of a few hours, are
significantly shorter than those estimated with respect to the
NO3 and ozone reactions which have been found to be around
few days. Therefore, the daytime OH-oxidation is the dominant
lead to the primary formation of PAN after the addition of NO3
radicals on the double bond (see the addition mechanism
proposed in Figure S8 of the Supporting Information). This
suggests that the abstraction of the aldehydic hydrogen is the
major reaction pathway for the three unsaturated aldehydes. It
should be mentioned that PANs yields obtained here are not
necessarily representative of the tropospheric conditions as
experiments were carried out with high NO2 levels (several
ppm) whereas tropospheric concentrations do not exceed a few
tens of ppb. The consequence is that PANs yields are much
higher than those that should be obtained under real
tropospheric conditions. However, this does not change the
main conclusion of this mechanistic study which is that the H-
abstraction is the major pathway. The primary formation of CO
may be explained by the scission of the alkoxy radical which
results from the reduction of the corresponding peroxy (see
channel 6 of Scheme 1). This pathway leads also to the
formation of a saturated aldehyde: butanal, pentanal, and
hexanal for trans-2-hexenal, trans-2-heptenal, and trans-2-
octenal, respectively. However, none of these carbonyl
compounds was detected as a primary product by FTIR or
by HPLC analyses. Even if we consider a concentration of 10
ppb, which correspond to the detection limit of these
compounds, their yields would be of 0.5%. Thereby, their
yields would be much lower than CO yields, which mean that
they cannot be coproducts. Therefore, CO obtained is not
10140
dx.doi.org/10.1021/jp3071234 | J. Phys. Chem. A 2012, 116, 10135−10142

The Journal of Physical Chemistry A
Article
(5) Guenther, A.; Geron, C.; Pierce, T.; Lamb, B.; Harley, P.; Fall, R.
Atmos. Environ. 2000, 34, 2205−2230.
(6) Wildt, J.; Kobel, K.; Schuh-Thomas, G.; Heiden, A. C. J. Atmos.
Chem. 2003, 45, 173−196.
Table 4. Calculated Tropospheric Lifetimes of trans-2-
Hexenal, trans-2-Heptenal, and trans-2-Octenal from Gas-
Phase Reactions with OH Radicals, NO3 Radicals, and O3
(7) Arey, J.; Winer, A. M.; Atkinson, R.; Aschmann, S. M.; Long, W.
D.; Lynn Morrison, C. Atmos. Environ. 1991, 25, 1063−1075.
(8) Kirstine, W.; Galbally, I.; Ye, Y.; Hooper, M. J. Geophys. Res.
1998, 103, 10605−10619.
compound
τOH
3.9 h49
τNO
τO
3
8 day22
3
trans-2-hexenal
5.0 day
this work
4.4 day
trans-2-heptenal
trans-2-octenal
3.9 h49
4.3 h50
(9) Fall, R.; Karl, T.; Hansel, A.; Jordan, A.; Lindinger, W. J. Geophys.
Res. 1999, 104, 15963−15974.
this work
4.2 day
(10) Springett, M. B.; Williams, B. M.; Barnes, R. J. Food Chem. 1994,
49, 393−398.
this work
́
(11) Boue, S. M.; Shih, B. Y.; Carter-Wientjes, C. H.; Cleveland, T. E.
J. Agric. Food Chem. 2003, 51, 4873−4876.
(12) Sinyinda, S.; Gramshaw, J. W. Food Chem. 1998, 62, 483−487.
(13) Cramer, A.-C. J.; Mattinson, D. S.; Fellman, J. K.; Baik, B.-K. J.
Agric. Food Chem. 2005, 53, 7526−7531.
chemical removal pathway for these unsaturated aldehydes in
the troposphere.
(14) Calogirou, A.; Larsen, B. R.; Kotzias, D. Atmos. Environ. 1999,
33, 1423−1439.
(15) Jemma, C. A.; Shore, P. R.; Widdicombe, K. A. J. Chromatogr.
Sci. 1995, 33, 34−48.
(16) Finlayson-Pitts, B. J.; Pitts Jr., J. N. Chemistry of the Upper and
Lower Atmosphere; Academic Press: San Diego, 2000.
(17) Mellouki, A.; Le Bras, G.; Sidebottom, H. Chem. Rev. 2003, 103,
5077−5096.
(18) Kanakidou, M.; Seinfeld, J. H.; Pandis, S. N.; Barnes, I.;
Dentener, F. J.; Facchini, M. C.; Van Dingenen, R.; Ervens, B.; Nenes,
A.; Nielsen, C. J.; Swietlicki, E.; Putaud, J. P.; Balkanski, Y.; Fuzzi, S.;
Horth, J.; Moortgat, G. K.; Winterhalter, R.; Myhre, C. E. L.;
Tsigaridis, K.; Vignati, E.; Stephanou, E. G.; Wilson, J. Atmos. Chem.
Phys. 2005, 5, 1053−1123.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental plots of concentrations of reactants, nitrogen
species (trans-2-hexenal, NO2, PAN analogue), and CO
measured by FTIR and by UV−visible absorption spectrometry
(NO3) vs time (Figure S1), formation yields of the PAN
analogue during NO3 oxidation of trans-2-hexenal (Figure S2),
formation yields of CO during NO3 oxidation of trans-2-
hexenal (Figure S3), formation yields of PAN analogue during
NO3 oxidation of trans-2-heptenal (Figure S4), formation yields
of CO during NO3 oxidation of trans-2-octenal (Figure S5),
formation yields of PAN analogue during NO3 oxidation of
trans-2-octenal (Figure S6), formation yields of CO during
NO3 oxidation of trans-2-octenal (Figure S7), proposed
mechanism for the NO3 radical addition to the double bond
on aliphatic unsaturated aldehydes (Figure S8). This material is
available free of charge via the Internet at http://pubs.acs.org.
(19) Atkinson, R. Atmos. Environ. 2007, 41, 200−240.
(20) Wayne, R. P.; Barnes, I.; Biggs, P.; Burrows, J. P.; Canosa-Mas,
C. E.; Hjorth, J.; Le Bras, G.; Moortgat, G. K.; Perner, D.; Poulet, G.;
Restelli, G.; Sidebottom, H. Atmos. Environ. 1991, 25, 1−203.
(21) Atkinson, R. J. Phys. Chem. Ref. Data 1991, 20, 459−507.
(22) Atkinson, R.; Arey, J.; Aschmann, S. M.; Corchnoy, S. B.; Shu, Y.
Int. J. Chem. Kinet. 1995, 27, 941−955.
AUTHOR INFORMATION
Corresponding Author
*Fax: +33 1451 71564. Phone: +33 1451 71586. E-mail:
picquet@lisa.u-pec.fr.
■
(23) Atkinson, R.; Aschmann, S. M.; Goodman, M. A. Int. J. Chem.
Kinet. 1987, 19, 299−307.
(24) Cabanas, B.; Salgado, S.; Martin, P.; Baeza, M. T.; Martinez, E. J.
Phys. Chem. A 2001, 105, 4440−4445.
Notes
(25) Salgado, M. S.; Monedero, E.; Villanueva, F.; Martin, P.; Tapia,
A.; Cabanas, B. Environ. Sci. Technol. 2008, 42, 2394−2400.
The authors declare no competing financial interest.
(26) Ullerstam, M.; Ljungstrom, E.; Langer, S. Phys. Chem. Chem.
̈
ACKNOWLEDGMENTS
Phys. 2001, 3, 986−992.
■
(27) Zhao, Z. J.; Husainy, S.; Smith, G. D. J. Phys. Chem. A 2011, 115,
12161−12172.
This work has been supported by the French Ministry of
Ecology and Sustainable Development through the program
“Primequal” and by the European Community within the
seventh framework Programme, section “Support for Research
InfrastructureIntegrated Infrastructure Initiative”: EURO-
CHAMP-2, Contract No. 228335. The authors thank IUP
Heidelberg for free access to DOASIS software and Dr. T.
Brauers and Pr. B.J. Finlayson-Pitts for free access to the
routine developed for linear analysis. The authors also
acknowledge John Orlando for providing us the CPAN
spectrum.
(28) Kerdouci, J.; Picquet-Varrault, B.; Doussin, J. F. ChemPhysChem
2010, 11, 3909−3920.
(29) Doussin, J. F.; Ritz, D.; Durand-Jolibois, R.; Monod, A.; Carlier,
P. Analusis 1997, 25, 236−242.
(30) Picquet-Varrault, B.; Orphal, J.; Doussin, J. F.; Carlier, P.; Flaud,
J. M. J. Phys. Chem. A 2005, 109, 1008−1014.
(31) Atkinson, R.; Plum, C. N.; Carter, W. P. L.; Winer, A. M.; Pitts,
J. N. J. Phys. Chem. 1984, 88, 1210−1215.
(32) IUPAC. Subcommittee on Gas Kinetic Data Evaluation,
Evaluated Kinetic Data, http://www.iupac-kinetic.ch.cam.ac.uk/.
(33) Brauers, T.; Finlayson-Pitts, B., J. Int. J. Chem. Kinet. 1997, 29,
665−672.
REFERENCES
■
(34) Ullerstam, M.; Langer, S.; Ljungstrom, E. Int. J. Chem. Kinet.
̈
(1) Graedel, T. E. Chemical Compounds in the Atmosphere; Academic
Press: New York, 1978.
2000, 32, 294−303.
(35) D’Anna, B.; Langer, S.; Ljungstrom, E.; Nielsen, C. J.; Ullerstam,
M. Phys. Chem. Chem. Phys. 2001, 3, 1631−1637.
(36) Pfrang, C.; Martin, R. S.; Nalty, A.; Waring, R.; Canosa-Mas, C.
E.; Wayne, R. P. Phys. Chem. Chem. Phys. 2005, 7, 2506−2512.
(37) Orlando, J. J.; Tyndall, G. S. J. Phys. Chem. A 2002, 106, 12252−
12259.
(2) Grosjean, D.; Grosjean, E.; Gertler, A. W. Environ. Sci. Technol.
2000, 35, 45−53.
(3) Kean, A. J.; Grosjean, E.; Grosjean, D.; Harley, R. A. Environ. Sci.
Technol. 2001, 35, 4198−4204.
(4) Carlier, P.; Hannachi, H.; Mouvier, G. Atmos. Environ. 1986, 20,
2079−2099.
10141
dx.doi.org/10.1021/jp3071234 | J. Phys. Chem. A 2012, 116, 10135−10142

The Journal of Physical Chemistry A
Article
(38) Ferrari, C. P.; Durand-Jolibois, R.; Carlier, P.; Jacob, V.; Roche,
A.; Foster, P.; Fresnet, P. Analusis 1999, 27, 45−53.
(39) Monedero, E.; Salgado, M. S.; Villanueva, F.; Martin, P.; Barnes,
I.; Cabanas, B. Chem. Phys. Lett. 2008, 465, 207−211.
(40) Canosa-Mas, C. E.; Carr, S.; King, M. D.; Shallcross, D. E.;
Thompson, K. C.; Wayne, R. P. Phys. Chem. Chem. Phys. 1999, 1,
4195−4202.
(41) Prinn, R. G.; Weiss, R. F.; Miller, B. R.; Huang, J.; Alyea, F. N.;
Cunnold, D. M.; Fraser, P. J.; Hartley, D. E.; Simmonds, P. G. Science
1995, 269, 187−192.
(42) Shu, Y.; Atkinson, R. J. Geophys. Res. 1995, 100, 7275−7281.
(43) Logan, J. A. J. Geophys. Res. 1985, 90, 10463−10482.
(44) Noda, J.; Holm, C.; Nyman, G.; Langer, S.; Ljungstrom, E. Int. J.
̈
Chem. Kinet. 2003, 35, 120−129.
(45) D’Anna, B.; Andresen, O.; Gefen, Z.; Nielsen, C. J. Phys. Chem.
Chem. Phys. 2001, 3, 3057−3063.
(46) Cabanas, B.; Martin, P.; Salgado, S.; Ballesteros, B.; Martinez, E.
J. Atmos. Chem. 2001, 40, 23−39.
(47) Papagni, C.; Arey, J.; Atkinson, R. Int. J. Chem. Kinet. 2000, 32,
79−84.
(48) D’Anna, B.; Nielsen, C. J. J. Chem. Soc., Faraday Trans. 1997, 93,
3479−3483.
(49) Davis, M. E.; Gilles, M. K.; Ravishankara, A. R.; Burkholder, J. B.
Phys. Chem. Chem. Phys. 2007, 9, 2240−2248.
(50) Gao, T.; Andino, J. M.; Rivera, C. C.; Marquez, M. F. Int. J.
Chem. Kinet. 2009, 41, 483−489.
10142
dx.doi.org/10.1021/jp3071234 | J. Phys. Chem. A 2012, 116, 10135−10142