Gas-phase reactions of nopinone, 3-Isopropenyl-6-oxo-heptanal, and 5-Methyl-5-vinyltetrahydrofuran-2-ol with OH, NO3, and Ozone

Article abstract of DOI:10.1021/es980530j

In the troposphere, α-pinene, β-pinene, limonene, and linalool are mainly oxidized to pinonaldehyde, nopinone, 3-isopropenyl-6-oxoheptanal (IPOH), and 5-methyl-5-vinyltetrahydrofuran-2-ol (MVT), respectively. The rate constant of the reactions of nopinone, IPOH, and MVT with OH, NO₃, and O₃ were determined by long path FT-IR spectroscopy, and the oxidation products from the reactions between the OH radical and pinonaldehyde, nopinone, IPOH, and MVT were investigated using GC-MS and HPLC. The reaction rate constants (k) for the reactions have been determined at 740 ± 5 Torr and 298 ± 5 K, and a number of reaction products were identified. The rate constants obtained for the reactions with nopinone were k_OH = (1.7 ± 0.2) x 10^-11, k_NO3 < 2 x 10^-15, and k_O3, < 5 x 10^-21; for the reactions with IPOH were k_OH = (1.1 ± 0.3) x 10^-10, k_NO3 = (2.6 ± 0.8) x 10^-13, and k_O3, = (8.3 ± 2.2) x 10^-18; and for the reactions with MVT were k_OH = (7.4 ± 0.9) x 10^-11, k_NO3 = (2.0 ± 0.9) x 10^-14, and k_O3, = (3.8 ± 0.8) x 10^-18 (all units are in cm³ molecule^-1 s^-1, and uncertainties are given as two σ on the experimental data). From the results obtained in this investigation and previous studies, it was concluded that a typical atmospheric lifetime with respect to chemical reactions was only a few hours for pinonaldehyde, IPOH, and MVT but was much longer·for nopinone with a lifetime of about 10 h.

Full text of DOI:10.1021/es980530j

Environ. Sci. Technol. 1999, 33, 453-460
products such as formaldehyde and acetone. In the case of
Gas-Phase Reactions of Nopinone,
3-Isopropenyl-6-oxo-heptanal, and
5-Methyl-5-vinyltetrahydrofuran-2-ol
products with low vapor pressure such as those that have
been investigated in the present work, the impact on
atmospheric chemistry will depend on to what extend they
react further in the gas phase and possibly contribute to
ozone formation or, alternatively, are removed from the gas
phase by adsorption on surfaces.
with OH, NO , and Ozone
3
R-Pinene, â-pinene, limonene, and linalool (Figure 1) are
among the most abundant terpenes emitted into the
troposphere (2-4, 7, 8). Their gas-phase oxidation reactions
have been the subject of several investigations (9-19), and
pinonaldehyde (cis-3-acetyl-2,2-dimethylcyclobutylethanal)
was found as the major product in the gas-phase oxidation
of R-pinene, similar to nopinone (6,6-dimethylbicyclo[3.1.1]-
heptan-2-one) from â-pinene, 3-isopropenyl-6-oxo-heptanal
(IPOH) from limonene, and 5-methyl-5-vinyltetrahydrofuran-
2-ol (MVT) from linalool (see Figure 1).
A G G E L O S C A L O G I R O U ,
N I E L S R . J E N S E N , * C L A U S J . N I E L S E N , †
D I M I T R I O S K O T Z I A S , A N D J E N S H J O R T H
European Commission, Joint Research Centre, Environment
Institute, TP 272, I-21020 Ispra (VA), Italy
In the troposphere, R-pinene, â-pinene, limonene, and
linalool are mainly oxidized to pinonaldehyde, nopinone,
3-isopropenyl-6-oxoheptanal (IPOH), and 5-methyl-5-
vinyltetrahydrofuran-2-ol (MVT), respectively. The rate
constant of the reactions of nopinone, IPOH, and MVT with
OH, NO , and O were determined by long path FT-IR
Very little is known about the atmospheric fate of these
terpene oxidation products. Only few rate constants of
terpene oxidation products have previously been investigated,
and in those studies no products were identified: Hallquist
et al. (20) and Glasius et al. (21) both used a relative rate
method and FT-IR as an analytical technique and found a
fast rate constant for the reaction of pinonaldehyde with OH
of (8.7 ( 1.1) × 10-11 cm3 molecule-1 s-1 and of (9.1 ( 1.8)
× 10-11 cm3 molecule-1 s-1, respectively, at 298 ( 5 K and
760 ( 20 Torr. For the reaction between pinonaldehyde and
NO3, they found k values of (2.4 ( 0.4) × 10-14 and (5.4 ( 1.8)
× 10-14 cm3 molecule-1 s-1, respectively. The rate constant
for the reaction between pinonaldehyde and O3 has only
been investigated by Glasius et al. (21) where a slow value
of (8.9 ( 1.4) × 10-20 cm3 molecule-1 s-1 was determined.
Atkinson and Aschmann (22) used GC-FID to study the rate
constant for the reaction of nopinone with OH at 296 ( 2 K
and at about 735 Torr total pressure. For the reaction between
nopinone and the OH radical, a value of (1.4 ( 0.4) × 10-11
cm3 molecule-1 s-1 was found. No values for the reaction
between nopinone and the NO3 radical or ozone were given.
Hatakeyama et al. (11) report a value of 1.5 × 10-11 cm3
molecule-1 s-1 for the reaction between OH and nopinone,
but no experimental details were given. To our knowledge,
no rate constant measurements have been published previ-
ously for the reactions of IPOH and MVT with OH, NO3, and
O3 and no degradation product investigations have been
performed with pinonaldehyde, nopinone, IPOH, or MVT.
3
3
spectroscopy, and the oxidation products from the reactions
between the OH radical and pinonaldehyde, nopinone,
IPOH, and MVT were investigated using GC-MS and HPLC.
The reaction rate constants (k) for the reactions have
been determined at 740 ( 5 Torr and 298 ( 5 K, and a
number of reaction products were identified. The rate
constants obtained for the reactions with nopinone were
kOH ) (1.7 ( 0.2) × 10-11, kNO < 2 × 10-15, and kO < 5 ×
3
3
10-21; for the reactions with IPOH were kOH ) (1.1 (
0.3) × 10-10, kNO ) (2.6 ( 0.8) × 10-13, and kO ) (8.3 (
3
3
2.2) × 10-18; and for the reactions with MVT were kOH
) (7.4 ( 0.9) × 10-11, kNO ) (2.0 ( 0.9) × 10-14, and kO
3
3
) (3.8 ( 0.8) × 10-18 (all units are in cm3 molecule-1
s-1, and uncertainties are given as two σ on the experimental
data). From the results obtained in this investigation and
previous studies, it was concluded that a typical atmospheric
lifetime with respect to chemical reactions was only a
few hours for pinonaldehyde, IPOH, and MVT but was much
longer for nopinone with a lifetime of about 10 h.
To elucidate the atmospheric degradation pathways of
these compounds, the rate constants of their reactions with
OH, NO3, and O3 were measured, and some of their reaction
products were identified.
Introduction
Emissions of natural non-methane hydrocarbons (NMHC)
have been estimated to exceed anthropogenic hydrocarbons
by a factor of about 10, with biogenic NMHC emissions being
about 1000 Tg of C yr-1, compared to anthropogenic emission
of about 100 Tg of C yr-1 (1-4). An important fraction of the
natural NMHC is the monoterpenes (C10H16) emitted by
vegetation, which contributes between 10 and 50% of the
total natural NMHC emission (3, 4). In the troposphere, they
react with OH and NO3 radicals as well as with O3, which
leads to the formation of a variety of products. Some of these
products are nonvolatile or semivolatile and contribute to
the formation of organic aerosol mass in the troposphere
(5). Further, terpene oxidation products may act as precursors
for ozone (6); this is certainly the case for the very volatile
Experimental Section
All of the experiments were performed in purified air at 740
( 5 Torr and 298 ( 5 K in a 480-L Teflon-coated reaction
chamber surrounded by photolysis lamps (black lamps, λ g
300 nm) and equipped with a multiple reflection mirror
system for on-line fourier transform infrared spectroscopy
(FT-IR) with a total optical path length of 81 m. A detailed
description of the experimental system has been published
previously (23). The spectra were obtained with a Bruker IFS
113v spectrometer at 1 cm-1 nominal resolution by co-adding
20-50 scans.
The OH radicals were formed by photolysis of CH3ONO,
which was synthesized by adding H2SO4 (50 wt % aqueous
solution) to sodium nitrite dissolved in a methanol/ water
mixture (24). NO3 radicals were formed by thermal dissocia-
* Corresponding author e-mail: niels.jensen@jrc.it; phone: +39-
0332-789225; fax: +39-0332-785837.
‡ On leave from the Department of Chemistry, University of Oslo,
Oslo, Norway.
9
10.1021/es980530j CCC: $18.00
Published on Web 12/24/1998
1999 Am erican Chem ical Society
VOL. 33, NO. 3, 1999 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 4 5 3

FIGURE 1. Major oxidation products from the atmospheric degradation of four terpenes.
TABLE 1. Summary of Reaction Rate Constants, k, Measured in This Investigation at 298 ( 5 K and 740 ( 5 Torri
a
b
Data from Glasius et al. (21). Measured relative to isoprene and 1,3-butadiene using kisoprene
) 1.01 × 10-10 and k1,3-butadiene
) 6.66
+
OH
+
OH
× 10-11
.
Measured relative to 1-butene using k1-butene + NO3 ) 1.25 × 10-14
.
Measured relative to isobutene using kisobutene OH ) 5.13 × 10-11
.
c
d
+
e
Measured relative to ethene and propene using kethene
3 ) 2.0 × 10-16 and kpropene
3 ) 9.45 × 10-16
.
f Measured relative to isobutene using
+
NO
+ NO
g
h
kisobutene
) 3.4 × 10-13
.
Measured relative to isobutene using kisobutene
) 1.21 × 10-17
.
Measured relative to isobutene using kisobutene
+
NO
+
O
+
3
O
) 1.21 × 310-14
.
i Units are cm 3 m olecule-1 s-1. Uncertainties are presented with 2 σ on the experim ental data. kinetic data of pinonaldehyde
3
is taken from Glasius et al. (21). The values for the reference com pounds were obtained from refs 33 and 44.
tion of N2O5, which was prepared in solid form according to
a previously described procedure (25). Ozone was prepared
by silent discharge of pure oxygen. In a typical experiment,
the temperature was 295 ( 3 K at the beginning; but when
CH3ONO was used as the precursor for OH radicals, the
temperature increased by 3-5 K due to the heating from the
UV lamps.
The OH, NO3, and O3 reaction rate constants were
determined mainly with the “relative rate” method, but the
O3 + nopinone reaction was measured under pseudo-first-
order conditions. Both methods have been previously
described in detail (21). Reference compounds used in the
relative rate method were isobutene, 1-butene, propene, and
ethene. In the footnote of Table 1, it is written which reference
compounds were used for which reaction. By the relative
FIGURE 2. Plot of ln {[nopinone]0/[nopinone] } corrected for wall
t
rate method, the ratio of the rate constants kA/ kB are found
as the slope of a plot of ln (A / A ) - kwt versus ln (B0/ Bt) -
losses and photolysis vs ln {[isobutene]0/[isobutene] } for the decay
t
0
t
of nopinone and isobutene during photolysis of methylnitrite,
producing OH radicals; kw ) losses on wall and/or by photolysis.
kwt, where A and B0 are the initial concentrations, A and Bt
0
t
are the concentrations at time t, and kwt is an additional
first-order loss process, if any additional loss process is
present, described by rate constant kw for Aand B, respectively
(see Figure 2). In purified air, nopinone showed a first-order
decay in our reaction chamber with the UV lamps turned on
(λ g 300 nm); this additional loss process was attributed to
9
4 5 4 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 33, NO. 3, 1999

wall loss and losses due to photolysis, and a value of kw
)
9.5 × 10-6 s-1 was measured for nopinone. For the other
compounds, kw ) 1.7 × 10-4 s-1 for IPOH and 7.9 × 10-5 s-1
for MVT, but in the case of the reference compounds
(isobutene, 1-butene, propene, and ethene), no additional
loss process could be identified in the reaction chamber with
UV light and values of <5 × 10-6 s-1 were assumed for kw,
which means that no data adjustment is needed for those
compounds in this investigation. Without UV light, IPOH
showed a first-order decay in our reaction chamber; this
additional loss process was attributed to wall loss, and a
value of kw ) 5.5 × 10-5 s-1 was measured for IPOH. For the
other compounds, kw ) 8.4 × 10-5 s-1 for MVT, but in the
case of nopinone and the reference compounds (isobutene,
1-butene, propene, and ethene), no additional loss process
could be identified in the reaction chamber without UV light,
and values of <5 × 10-6 s-1 were assumed for kw, which
again means that no data adjustment is needed for those
compounds in this investigation. It should be mentioned
that these additional losses are relatively small because the
time range for an experiment is normally between 5 and 15
min. The additional loss process (wall loss + photolysis) was
always below 15% of the overall decay during an experiment
for all of the experiments performed in this investigation.
For the measurements of the reactions of IPOH and MVT
with O3, cyclohexane (270 and 200 ppmV, respectively) was
added to scavenge OH radicals, which are believed to be
formed by the ozonolysis of these olefinic species. For the
reaction of nopinone with O3, no scavenger was used since
nopinone does not contain olefinic double bonds and
because these experiments were performed under pseudo-
first-order conditions without a reference compound present
(1 ppmV ) 2.46 × 1013 molecules cm-3 at 298 K and 760
Torr).
FIGURE 3. Plot of ln {[IPOH]0/[IPOH]t} vs ln {[isobutene]0/
[isobutene]t} (9). Plot of ln {MVT]0/[MVT]t} vs ln {[isobutene]0/
[isobutene] } (b). During photolysis of methylnitrite, producing OH
t
radicals; kw ) losses on wall and/or photolysis.
eter (HP-5972) operated in the chemical ionization (isobutane
as ionizing agent) mode. The compounds trapped in the
aqueous solution of the impinger were derivatized with
O-(2,3,4,5,6-pentafluorobenzyl)hydroxylamine using 0.5 mL
of a 1 M phosphate buffer (27) and extracted twice with
solvent (first dichloromethane and then methyl tert-butyl
ether). The analysis was carried out by chemical ionization
GC-MS in the same system used to analyze the Carboxen
tubes as described by Lahaniati et al. (28). The analysis of the
DNPH cartridges was done by solvent extraction (acetonitrile)
followed by high-performance liquid chromatography (HPLC,
Kontron) with ultraviolet detection.
Chem icals. Pinonaldehyde was synthesized and identified
as described previously (21). Nopinone was commercially
available from Aldrich (98% pure). IPOH and MVT were
synthesized as described in refs 15 and 29. IPOH, IR spectrum
(gas phase, cm-1): 900 (m), 1161 (m), 1367 (m), 1740 (s),
2714 (m), 2814 (m), 2945 (m), and 3081 (m). EI-MS m/ z 168
(M+, 0), 107 (25), 67 (23), 58 (15), 55 (18), 43 (100), and 41
(32); CI-MS (methane) 169 (M + H+, 20), 151 (100), and 107
(26). MVT, IR spectrum (gas phase, cm-1): 845 (m), 924 (m),
975 (s), 995 (s), 1036 (s), 1210 (m), 1459 (m), 2881 (m), 2935
(m), 2984 (s), 3019 (m), 3102 (m), 3650 (m). EI-MS m/ z 128
(M+, 2), 113 (44), 110 (10), 95 (20), 71 (90), 67 (94), 55 (78),
43 (100), and 41 (53); CI-MS (methane) 127 (M + H+ - H2,
3), 111 (100), 93 (36), and 81 (9). The other chemicals used
in this study were all commercial samples used without
further purification: isobutene (Matheson, >99% pure),
1-butene (99.9% pure, Ucar), propene (Ucar, >99.5% pure),
ethene (g99.5% pure, Air Liquide), synthetic air (80% N2 and
20% O2: mixture g99.95% pure, SIO), and O2 (g99.9% pure,
SIO).
The terpene oxidation products used in this investigation
have a low vapor pressure and become unstable when heated.
Therefore, instead of the normal manometric method, the
following method was applied: Each of the compounds was
introduced into the reaction chamber by applying 10-20
µL of the compound to a filter and gently blowing the vapor
into the reaction chamber by a stream of air. Infrared spectra
were obtained for all four compounds.
Typical initial experimental conditions were 15 ppmV <
[CH3ONO] < 30 ppmV, 10 ppmV < [N2O5] < 100 ppmV, 10
ppmV < [O3] < 300 ppmV, and 1 ppmV < [A] < 10 ppmV
(where A is one of the terpene oxidation products). The
concentration of A was an estimate, based on known cross
sections of similar compounds. All of the kinetics data are
based on relative measurement of the decay of A and the
reference compound, B; therefore, it is not necessary to know
the exact concentrations. The concentrations of the reference
compound [B] were in the range 4 ppmV < [B] < 25 ppmV.
Samples of reaction products were taken in the following
ways: (a) on stainless steel adsorbent tubes (Perkin-Elmer)
filled with 200 mg of Tenax-TA at a sampling flow of 150 mL
min-1 (sampling volume ) 0.3 L); (b) on glass adsorbent
tubes filled with 150 mg of Carboxen 563 at a sampling flow
of 600 mL min-1 (sampling volume ) 12 L); (c) through an
impinger filled with 15 mL of a solution of H2O/ t-BuOH (3:1)
at a flow rate of 800 mL min-1 (sampling volume ) 12 L) (d)
on 2,4-dinitrophenylhydrazine (DNPH) coated C18-silica gel
cartridges at a sampling flow of 1 L min-1 (sampling volume
) 1 L).
Results and Discussion
Kinetic Studies. In this investigation, we have determined
the rate constants for the reactions of OH, NO3, and O3 with
nopinone, IPOH, and MVT. The values measured are
presented in Table 1 together with the values for pinonal-
dehyde. Typical examples of the results obtained by the
relative rate measurements of the OH, NO3, and O3 reactions
with nopinone, IPOH, and MVT can be found in Figures2-6.
For the relative rate measurement of the reaction between
nopinone and the NO3 radical, in the presence of the reference
compounds ethene and propene, only a very small decay of
nopinone could be observed, and only an upper limit for
that reaction rate constant has been given; no figure is shown
here. For the only pseudo-first-order measurements that were
performed (nopinone + O3), no reaction could be detected,
and thus only an upper limit for that reaction rate constant
has been given; no figure is shown.
The analysis of the Tenax tubes was done by thermal
desorption (ATD 400, Perkin-Elmer) gas chromatography
(Fisons HRGC Mega 2 Series)-mass spectrometry (Fisons
Trio1000) as described in detail elsewhere (26). The analysis
of the Carboxen tubes was done by solvent extraction
(dichloromethane) followed by on-column injection into a
gas chromatograph (HP-5890) coupled to a mass spectrom-
The major oxidation product of linalool (4-hydroxy-4-
methyl-5-hexenal) is a γ-hydroxyaldehyde (15, 18) and exists
9
VOL. 33, NO. 3, 1999 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 4 5 5

reaction of pinonaldehyde with OH of (8.7 ( 1.1) × 10-11 and
of (9.1 ( 1.8) × 10-11 cm3 molecule-1 s-1, respectively, at 298
( 5 K and 760 ( 20 Torr. As can be seen, the two values are
in good agreement. For the reaction between pinonaldehyde
and NO3, they found a k value of (2.4 ( 0.4) × 10-14 and of
(5.4 ( 1.8) × 10-14 cm3 molecule-1 s-1, respectively, which
is about a factor of 2 difference. The relatively high value
obtained for the reaction between pinonaldehyde and the
NO3 radical is in agreement with a recent study, where for
aliphatic aldehydes an increasing rate constant with an
increasing number of carbon atoms was observed (31). The
reaction rate constant for the reaction between pinonalde-
hyde and O3 has only been investigated by Glasius et al. (21)
where a value of (8.9 ( 1.4) × 10-20 cm3 molecule-1 s-1 was
determined. The reaction of O3 with aldehydes is believed
to be very slow (32), which is in agreement with the value
obtained.
FIGURE 4. Plot of ln {[IPOH]0/[IPOH] } corrected for wall losses vs
t
ln {[isobutene]0/[isobutene] } for the decay of IPOH and isobutene
t
during reaction with the NO radical.
3
Atkinson and Aschmann (22) used GC-FID to study the
rate constant for the reaction of nopinone with OH at 296
( 2 K and at about 735 Torr total pressure. For the reaction
between nopinone and the OH radical, a value of (1.4 ( 0.4)
× 10-11 cm3 molecule-1 s-1 was found, which is in good
agreement with our value of (1.7 (0.2) ×10-11 cm3 molecule-1
s-1. The reactions of ketones with NO3 radicals are very slow
(33), and the reactions of ketones with O3 are believed to be
below a measurable rate (22). Both statements are in
agreement with the values obtained in this investigation for
the rate constants of the reactions of nopinone with NO3 and
with O3 (see Table 1).
FIGURE 5. Plot of ln {MVT]0/[MVT] } corrected for wall losses vs
t
To our knowledge, no rate constants measurements have
been published previously for the reactions of IPOH and
MVT with OH, NO3, and O3 (Table 1), but the k values are
similar to those obtained from alkene reactions (33, 34).
ln {[isobutene]0/[isobutene] } for the decay of MVT and isobutene
t
during reaction with NO .
3
It has been suggested that the reactivity of the OH radical
toward organics, including the oxygenated hydrocarbons
presently studied, may be estimated from a simple structure-
reactivity relationship (35). This incremental reactivity
method predicts a rate constant for the OH reaction of 2.4
× 10-11 cm3 molecule-1 s-1 for pinonaldehyde, of 4.6 × 10-12
cm3 molecule-1 s-1 for nopinone, of 7.8 × 10-11 cm3
molecule-1 s-1 for IPOH, and of 7.4 × 10-11 cm3 molecule-1
s-1 for MVT as shown in Table 2, which also includes the
predicted rate constants for these reactions obtained by
applying a structure-reactivity method based upon molec-
ular orbital calculations (36-38). As seen in Table 2, the
agreement between the measured values and those estimated
by the incremental reactivity method (35) is acceptable in
the case of MVT and reasonable in the case of IPOH, but in
the case of pinonaldehyde and nopinone, the estimated value
is about 3-4 times lower than the measured one. There is
no obvious explanation for this discrepancy, but the fact
that disagreement is found only for the two molecules that
contain a 1,1-dimethylcyclobutane skeleton suggests that
more data concerning the reactivity of such compounds are
needed. For the structure-reactivity method based upon
molecular orbital calculations (36-38), there is a good
agreement in the case of nopinone + OH, but for the other
three reactions the predicted rate constants are a factor of
2-3 lower than the measured ones.
FIGURE 6. Plot of ln {[IPOH]0/[IPOH]t} vs ln {[isobutene]0/
[isobutene]t} (9). Plot of ln {MVT]0/[MVT]t} vs ln {[isobutene]0/
[isobutene] } (b). During reaction with O in the presence of
t
3
cyclohexane; kw ) losses on wall.
therefore in equilibrium with its cyclic form MVT (5-methyl-
5-vinyltetrahydrofuran-2-ol), a thermodynamically favored
lactol (in this section under Products Studies, the two forms
can be seen in Figure 8). MVT is mentioned in the literature
also as lavender lactol (30). Shu et al. (18) suggested on the
basis on FT-IR data that the compound must be present
entirely in the cyclic form. We confirm the observation done
by Shu et al. (18), because we observed a very weak carbonyl
signal in the FT-IR and 1H NMR spectra, indicating the
presence of the open chain form. On the basis of the
integration of the aldehyde signal and the single hydrogen
Product Studies. Reaction between Pinonaldehyde and
OH. Norpinonaldehyde (cis-3-acetyl-2,2-dimethylcyclobu-
tylmethanal, [I] in Figure 7) was identified by CI-GC-MS
(methods b and c in the Experimental Section). In addition,
acetone and a carbonyl compound, tentatively identified as
methylglyoxal, were identified as their hydrazones by HPLC
(method d). Norpinonaldehyde has previously been observed
as a product from the reaction between R-pinene and O3 (9,
39). The formation of norpinonaldehyde results from known
reaction mechanisms as shown in Figure 7, and a rear-
1
of the vinyl group in the H NMR spectrum of MVT, a ratio
of 18:1 of closed to open chain form was estimated. This
equilibrium is of particular interest when discussing the
reaction pathways of MVT.
Only a few rate constants of terpene oxidation products
have previously been investigated: Hallquist et al. (20) and
Glasius et al. (21) both used a relative rate method and FT-IR
as the analytical technique and found a rate constant for the
9
4 5 6 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 33, NO. 3, 1999

FIGURE 7. Reaction scheme proposed for the reaction between pinonaldehyde and the OH radical.
FIGURE 8. Reaction scheme proposed for the reaction between MVT and the OH radical.
rangement of the intermediate cyclobutyl radical (40) could
explain the formation of acetone. Other formation pathways
to acetone could be hydrogen abstraction from either of the
two tertiary C-H sites and further reactions of the cyclic
radical formed.
ylbicylco[3.1.1]heptan-2-one [III] (see Chart 1) were identi-
fied, and 6,6-dimethyl-3-hydroxybicylco[3.1.1]heptan-2-
one [IV] was identified in traces by CI-GC-MS (method b).
6,6-Dimethylbicylco[3.1.1]heptan-2,3-dione was identified
by GC-MS [method a: 152 (M+, 1), 124 (26), 109 (80), 81
(63), 68 (84), and 55 (100)] and 3,7-dihydroxy-6,6-dimeth-
ylbicylco[3.1.1]heptan-2-one as its oxime (method c). Acetone
Reaction between nopinone and OH. 6,6-Dimethylbicylco-
[3.1.1]heptan-2,3-dione [II] and 3,7-dihydroxy-6,6-dimeth-
9
VOL. 33, NO. 3, 1999 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 4 5 7

TABLE 2. Measured and Estimated OH Reaction Rate Constantsa
a
The units are in cm 3 m olecule-1 s-1. Num bers in parentheses are reference num bers. An asterisk (*) m eans that the values were calculated
by the authors using the m ethod described by Kwok and Atkinson (35). The num ber sign (#) m eans that the values were calculated by Klam t and
Lohrenz (38) using the m ethod described in refs 36 and 37.
hydroxycarbonyl with mass 142, which we attributed to
5-methyl-4-oxo-5-vinyltetrahydrofuran-2-ol, were observed
in small amounts by GC-MS and CI-GC-MS (methods a
and b). The formation of 4-oxopentanal can be better
understood if a fast equilibrium of MVT with its open form
4-hydroxy-4-methyl-5-hexenal is supposed, see Figure 8.
4-Oxopentanal could be readily formed together with hy-
droxyacetaldehyde following the mechanism described in
Figure 8.
CHART 1
was observed as its hydrazone by HPLC (method d). 6,6-
Dimethylbicylco[3.1.1]heptan-2,3-dione and 6,6-dimethyl-
3-hydroxybicylco[3.1.1]heptan-2-one have previously been
observed as products from the reaction between R-pinene
and O3 (7, 39), and their formation can be explained by the
well-known chemistry of alkyl peroxy and oxy radicals. The
formation of 3,7-dihydroxy-6,6-dimethylbicylco[3.1.1]heptan-
2-one can be understood if a 1,5 hydrogen transfer takes
place after the formation of a 3-oxy or 7-oxo radical
intermediate in the oxidation of nopinone.
Reaction between IPOH and OH. 2-Isopropenyl-5-oxo-
hexanal was identified by CI-GC-MS (methods b and c).
Reaction between MVT and OH. 4-Oxopentanal ([V] in
Figure 8) was unambiguously identified by CI-GC-MS
(methods b and c) by comparison with an authentic standard
synthesized in the laboratory (37). 4-Oxopentanal has been
incorrectly identified in a previous study as 4-hydroxypen-
tanal (14). 5-Methyl-5-vinyltetrahydrofuran-2-one and a
Conclusions and Atmospheric Implications
Typical chemical lifetimes in the troposphere for the terpene
oxidation products (pinonaldehyde, nopinone, IPOH, and
MVT) have been estimated (Table 3) by applying the values
of the reaction rate constants from Table 1. Reaction with
the OH radical seems to be the dominant reaction pathway
in the troposphere for these compounds, reaction with the
NO3 radical is only of minor importance, whereas reaction
with O3 is negligible. The present results indicate a typical
atmospheric lifetime with respect to chemical reactions of
about 1-2 h for pinonaldehyde, IPOH, and MVT and about
9 h for nopinone.
From the available kinetic information, the ratios between
the concentrations of these terpene oxidation products and
their parent compounds in ambient air can be estimated,
taking only gas-phase chemical reactions into account. Such
9
4 5 8 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 33, NO. 3, 1999

TABLE 3. Estimated Chemical Lifetimes of Oxidation Products in the Troposphere under Typical Tropospheric Conditionsa
a
An asterisk (*) indicates that the concentration of OH used is considered to be a typical daytim e value (45), while that of NO3 is assum ed to
be a characteristic continental nighttim e concentration (46). The concentration given for O3 is a 24-h average (47).
a calculation requires knowledge not only about the relevant
rate constants but also about the yields of the oxidation
products. For daytime conditions, the NO3 radical concen-
tration can be considered as being negligible, and thus only
OH and ozone are taken into consideration. For this
calculation, the OH and ozone concentrations shown in Table
3 (0.07 pptV and 30 ppbV, respectively) have been used
together with the rate constant values presented in Table 1
and the rate constants for the OH and ozone reactions with
R-pinene, â-pinene, limonene, and linalool given by Atkinson
(34) and by Atkinson et al. (42). Assuming steady-state
conditions, i.e., that the formation rates equals the removal
of the terpene oxidation product A:
product studies of the atmospheric terpene degradation that
the mechanisms are not fully clarified. The present product
is of only qualitative character, and additional analytical
techniques are needed to give a more detailed reaction
mechanism.
Further work should clarify the importance of other
tropospheric sinks of these terpene oxidation products and
their degradation products. The importance of photolysis
and heterogeneous losses should be investigated. Recent
results indicate that the typical lifetime of pinonaldehyde
with respect to photolysis under clear sky conditions is 3.3
h at 50° N in July and 22 h at 50° N in January (20), when a
photolysis quantum yield of 1 was assumed. By comparing
these lifetime values with the values in Table 3, it can be
concluded that photolysis could be an important degradation
mechanism for pinonaldehyde in the atmosphere.
k1OH[OH][terpene]R + k1ozone[O3][terpene]â )
k2OH [OH][A]ss + k2ozone [O3][A]ss
Acknowledgments
where R an â are the fractional yields of A, and the ratio
between the concentration of the terpene and its oxidation
product A ([A]ss/ [terpene]) can be calculated. The reaction
of A with ozone is negligible in the cases of pinonaldehyde
and nopinone. The steady-state assumption is best in the
case where a ‘box model’ approach can be used and not
good in the cases where transport and dispersion are likely
to be important. However, it does give an insight into the
concentration levels of these compounds that should be
expected in a forest area. If R and â are set equal to 1, the
following ratios are obtained: [pinonaldehyde]/ [R-pinene]
) 1.0, [nopinone]/ [â-pinene] ) 5.1, [IPOH]/ [limonene] )
2.2, and [MVT]/ [linalool] ) 4.7. These ratios must then be
modified according to the yields of these oxidation products
(i.e., R and â), which are not known precisely; however,
laboratory experiments indicate that pinonaldehyde, nopi-
none, and MVT are formed in high yields, i.e., in the range
from 20 to more than 50% on a molar basis, by the oxidation
of their parent compounds by reaction with OH radicals or
ozone (9, 11, 13, 18).
Very few field measurements of these terpene oxidation
products are available; results of measurements performed
in Italy and Spain, reported by Calogirou et al. (43), show
ratios between oxidation products and parent compounds
that are lower than those calculated here even when yields
(R and â) as low as 0.2 are used for the calculation. More field
measurements of these compounds in ambient air will show
if atmospheric sinks other than gas-phase reactions with OH
are of importance.
A.C. would like to thank the European Commission for a
Ph.D. grant. C.J.N. acknowledges financial support from the
Department of Chemistry, University of Oslo, Norway, during
his stay at the EI, JRC (I). The authors gratefully acknowledge
the help of G. Ottobrini with the experimental work. The
authors acknowledge financial support from the European
Commission, DG XII, within the contract ENV4-CT97-0419
“RADICAL”.
Literature Cited
(1) Isaksen, I. S. A.; Hov, Ø. Tellus 1987, 39B, 271-285.
(2) Rasmussen, R. A. J. Air. Pollut. Control Assoc. 1972, 22, 537-
543.
(3) Zimmerman, P. R.; Chatfield, R. B.; Fishmann, J.; Crutzen, P. J.;
Hanst, P. L. Geophys. Res. Lett. 1978, 5, 679-682.
(4) Guenther, A.; Hewitt, C. N.; Erickson, D.; Fall, R.; Geron, C.;
Graedel, T.; Harley, P.; Klinger, L.; Lerdau, M.; McKay, W. A.;
Pierce, T.; Scholes, B.; Steinbrecher, R.; Tallamraju, R.; Taylor,
R J.; Zimmerman, P. J. Geophys. Res. 1995, 100, 8873-8892.
(5) Novakov, P.; Penner, J. Nature 1993, 365, 823-826.
(6) Trainer, M.; Williams, E. J.; Parrish, D. D.; Buhr, M. P.; Allwine,
E. J.; Westberg, H. H.; Fehsenfeld, F. C.; Liu, S. C. Nature 1987,
705-707.
(7) Hull, L. A. Atmospheric biogenic hydrocarbons; Bufalini, J. J.,
Arnds, R. R., Eds.; Ann Arbor Press: Ann Arbor, MI, 1980; Vol.
2, pp 161-186.
(8) Yokouchi, Y.; Ambe, Y. Atmos. Environ. 1985, 19, 1271-1276.
(9) Hatakeyama, S.; Izumi, K.; Fukuyama, T.; Akimoto, H. J. Geophys.
Res. 1989, 94, 13013-13024.
(10) Arey, J.; Atkinson, R.; Aschmann, S. M. J. Geophys. Res. 1989, 95,
18539-18546.
(11) Hatakeyama, S.; Izumi, K.; Fukuyama T.; Akimoto, H.; Washida
N. J. Geophys. Res. 1991, 96, 947-958.
A number of reaction products could be identified, part
of which have been previously detected as products from
their parent terpene. It is, however, clear from this and other
9
VOL. 33, NO. 3, 1999 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 4 5 9

(12) Arey, J.; Corchnoy, S. B.; Atkinson, R. Atmos. Environ. 1991,
25A, 1377-1381.
(13) Hakola, H.; Arey, J.; Aschmann, S. M.; Atkinson, R. J. Atmos.
Chem. 1994, 18, 75-102.
(31) D’Anna, B.; Nielsen, C. J. J. Chem. Soc., Faraday Trans. 1997,
93, 3479-3483.
(32) Finlayson-Pitts, B. J.; Pitts, J. N., Jr. Atmospheric Chemistry;
Wiley: New York, 1986; pp 1033-1034.
(33) Wayne, R. P.; Barnes, I.; Biggs, P.; Burrows, J. P.; Canosa-Mas,
C. E.; Hjorth, J.; LeBras, G.; Moortgat, G. K.; Perner, D.; Poulet,
G.; Restelli, G.; Sidebuttom, H. The nitrate radical: Physics,
chemistry, and the atmosphere; Commission of the European
Communities: Air pollution research report 31; 1991; pp 129-
131.
(34) Atkinson, R. J. Phys. Chem. Ref. Data 1997, 26, 215-290.
(35) Kwok, E. S.; Atkinson, R. Atmos. Environ. 1995, 29, 1685-1695.
(36) Klamt, A. Chemosphere 1993, 26, 1273-1289.
(37) Klamt, A. Chemosphere 1996, 32, 717-726.
(38) Klamt, A.; Lohrenz, J. 1998, private communication.
(39) Barton, D. H. R.; Ozbalik, N.; Schmitt, M. Tetrahedron Lett. 1989,
30, 3263-3266.
(40) Jay, K.; Stieglitz, L. Proceeding of air pollution and ecosystems,
Grenoble, France, May 18-22, 1997; pp 542-547.
(41) Fruekilde, P.; Hjorth, J.; Jensen, N. R.; Kotzias, D.; Larsen, B.
Atmos. Environ. 1998, 32, 1893-1902.
(42) Atkinson, R.; Arey, J.; Aschmann, S. M.; Corchnoy, S. B.; Shu,
Y. Int. J. Chem. Kinet. 1995, 27, 941-955.
(43) Calogirou, A.; Duane, M.; Kotzias, D.; Lahaniati, M.; Larsen, B.
R. Atmos. Environ. 1997, 31, 2741-2751.
(44) Atkinson, R. J. Phys. Chem. Ref. Data 1989, Monograph No. 1,
1-246.
(45) Prinn, R. G.; Weiss, R. F.; Miller, B. R.; Huang, J.; Alyea, F.;
Cunnold, D.; Fraser, P.; Hartley, D.; Simmonds, P. Science 1995,
269, 187-192.
(46) Platt, U.; Heintz, F. Isr. J. Chem. 1994, 34, 289-300.
(47) Logan, J. A. J. Geophys. Res. 1985, 90, 10463-10482.
(14) The BEMA project. EUR report EUR 15955 EN; 1995.
(15) Calogirou, A.; Kotzias, D.; Kettrup, A. Naturwissenschaften 1995,
82, 288-289.
(16) Wa¨ngberg, I.; Barnes, I.; Becker, K. H. Environ. Sci. Technol.
1997, 31, 2130-2135.
(17) Berndt, T.; Bo¨ ge, O. J. Chem. Soc., Faraday Trans. 1997, 93,
3021-3027.
(18) Shu, Y.; Kwok, E. S. C.; Tuazon, E. C.; Atkinson, R.; Arey, J. Environ.
Sci. Technol. 1997, 31, 896-904.
(19) Grosjean, D.; Williams, E. L. II; Grosjean, E.; Andino, J. M.;
Seinfeld, J. H. Environ. Sci. Technol. 1993, 27, 2754-2758.
(20) Hallquist, M.; Wa¨ngberg, I.; Ljungstro¨m, E. Environ. Sci. Technol.
1997, 31, 3166-3172.
(21) Glasius, M.; Calogirou, A.; Jensen, N. R.; Hjorth, J.; Nielsen, C.
J. Int. J. Chem. Kinet. 1997, 29, 527-533.
(22) Atkinson, R.; Aschmann, S. M. J. Atmos. Chem. 1993, 16, 337-
348.
(23) Jensen, N. R.; Hjorth, J.; Lohse, C.; Skov, H.; Restelli. G. Atmos.
Environ. 1991, 25A, 1897-1904.
(24) Taylor, W. D.; Allston, T. D.; Moscato, M. J.; Fazekas, G. B.;
Kozlowski R. Takacs, G. A. Int. J. Chem. Kinet. 1980, 12, 231-
240.
(25) Schott, G.; Davidson, N. J. Am. Chem. Soc. 1958, 80, 1841-1853.
(26) Calogirou, A.; Larsen, B. R.; Brussol, C.; Duane, M.; Kotzias, D.
Anal. Chem. 1996, 68, 1499-1506.
(27) Yu, J.; Jeffries, H. E.; Le Lacheur, M. R. Environ. Sci. Technol.
1995, 29, 1923-1932.
(28) Lahaniati, M.; Calogirou A.; Duane, M.; Larsen, B.; Kotzias, D.
Fresenius Environ. Bull. 1998, 7, 302-305.
(29) Wolinsky, J.; Barker, W. J. Am. Chem. Soc. 1960, 82, 636-368.
(30) Klimes, I.; Lamparsky, D. Proceeding of the 16th symposium on
essential oil research, Germany, September 1986; pp 197-214.
Received for review May 21, 1998. Revised manuscript re-
ceived October 29, 1998. Accepted November 5, 1998.
ES980530J
9
4 6 0 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 33, NO. 3, 1999