Gas-phase ozonolysis: Rate coefficients for a series of terpenes and rate coefficients and OH yields for 2-methyl 2-butene and 2,3-dimethyl-2-butene

Article abstract of DOI:10.1002/kin.10063

The kinetics of the gas-phase reactions of O₃ with a series of selected terpenes has been investigated under flow-tube conditions at a pressure of 100 mbar synthetic air at 295 ± 0.5 K. In the presence of a large excess of m-xylene as an OH radical scavenger, rate coefficients k(O₃+terpene) were obtained with a relative rate technique, (unit: cm³ molecule^-1 s^-1, errors represent 2σ): α-pinene: (1.1 ± 0.2) × 10^-16, ³ δ-carene: (5.9 ± 1.0) × 10^-17, lirnonene: (2.5 ± 0.3) × 10^-16, myrcene: (4.8 ± 0.6) × 10^-16, trans-ocimene: (5.5 ± 0.8) × 10^-16, terpinolene: (1.6 ± 0.4) × 10^-15 and α-terpinene: (1.5 ± 0.4) × 10^-14. Absolute rate coefficients for the reaction of O₃ with the used reference substances (2-methyl-2-butene and 2,3-dimethyl-2-butene) were measured in a stopped-flow system at a pressure of 500 mbar synthetic air at 295 ± 2 K using FT-IR spectroscopy, (unit: cm³ molecule^-1 s^-1, errors represent 2σ): 2-methyl-2-butene: (4.1 ± 0.5) × 10^-16 and 2,3-dirnethyl-2-butene: (1.0 ± 0.2) × 10^-15. In addition, OH radical yields were found to be 0.47 ± 0.04 for 2-methyl-2-butene and 0.77 ± 0.04 for 2,3-dimethyl-2-butene.

Full text of DOI:10.1002/kin.10063

Gas-Phase Ozonolysis: Rate
Coefficients for a Series
of Terpenes and Rate
Coefficients and OH Yields
for 2-Methyl-2-butene
and 2,3-Dimethyl-2-butene
2
1
1
1
1
¨
M. WITTER, T. BERNDT, O. BOGE, F. STRATMANN, J. HEINTZENBERG
1Institut fu¨r Tropospha¨renforschung, Permoserstraße 15, 04318 Leipzig, Germany
2Fachhochschule Magdeburg, FB Chemie/Pharmatechnik, Breitscheidstraße 2, 39114 Magdeburg, Germany
Received 19 September 2001; accepted 15 February 2002
DOI 10.1002/kin.10063
ABSTRACT: The kinetics of the gas-phase reactions of O3 with a series of selected terpenes has
been investigated under flow-tube conditions at a pressure of 100 mbar synthetic air at 295
0.5 K. In the presence of a large excess of m-xylene as an OH radical scavenger, rate coefficients
k(O3+terpene) were obtained with a relative rate technique, (unit: cm3 molecule 1 s 1, errors
represent 2 ): -pinene: (1.1 0.2) 10 16, 31-carene: (5.9 1.0) 10 17, limonene: (2.5
0.3) 10 16, myrcene: (4.8 0.6) 10 16, trans-ocimene: (5.5 0.8) 10 16, terpinolene: (1.6
0.4) 10 15 and -terpinene: (1.5 0.4) 10 14. Absolute rate coefficients for the reaction
of O3 with the used reference substances (2-methyl-2-butene and 2,3-dimethyl-2-butene) were
measured in a stopped-flow system at a pressure of 500 mbar synthetic air at 295 2 K using
FT-IR spectroscopy, (unit: cm3 molecule 1 s 1, errors represent 2 ): 2-methyl-2-butene: (4.1
16
0.5) 10
and 2,3-dimethyl-2-butene: (1.0 0.2) 10 15. In addition, OH radical yields
were found to be 0.47 0.04 for 2-methyl-2-butene and 0.77 0.04 for 2,3-dimethyl-2-butene.
C
2002 Wiley Periodicals, Inc. Int J Chem Kinet 34: 394–403, 2002
penes [1]. Their atmospheric degradation process in the
gas phase is initiated either by the attack of OH or NO3
radicals or O3. The contribution of the three different
oxidants to the overall removal process is spatially and
temporally variable and depends on NOx availability,
meteorological parameters, and the local terpene emis-
sion itself. In the past two decades laboratory studies
have been performed concerning the kinetics and the
product distribution of terpenes oxidation [3,4]. Espe-
cially in the last few years, interest has been focused
INTRODUCTION
Terpenes are released into the atmosphere from vegeta-
tion with an estimated global emission rate in the order
of 1014 g yr 1 [1,2]. ␣-Pinene (35%), ␤-pinene (23%),
andlimonene (23%) arethe most important compounds
inthisprocessaccountingforca. 80%oftheemittedter-
Correspondence to: T. Berndt; e-mail: berndt@tropos.de.
2002 Wiley Periodicals, Inc.
c

GAS-PHASE OZONOLYSIS
395
on the contribution of terpene oxidation products to
the formation and growth of particles [5]. This idea
was hypothesized first in the sixties [6]. Furthermore,
the ozonolysis of alkenes including terpenes represents
a significant OH radical source [3]. For a rural area in
the southeast of the United States, 10–15% of the to-
tal radical production was attributed to the ozonolysis
of alkenes [7]. Generally, for the modeling of atmo-
spheric processes concerning terpene degradation as
well as for laboratory studies a high quality data set of
kinetic parameters for the reaction of O3 with terpenes
is desired.
EXPERIMENTAL
The experiments have been performed using two dif-
ferent approaches: (i) a stopped-flow system for mea-
suring absolute rate coefficients for the reaction of O3
with 2-methyl-2-butene and 2,3-dimethyl-2-butene as
well as the corresponding OH radical yields and (ii) a
quartz flow-tube (If T-LFT: Institute for Tropospheric
Research–Laminar Flow Tube) for the determination
of relative rate coefficients for the reaction of O3 with
a series of terpenes.
The aim of the present study was kinetic investiga-
tions for the reaction of O3 with a series of terpenes
(␣-pinene, 31-carene, limonene, myrcene, trans-
ocimene, terpinolene, and ␣-terpinene) under flow-
tube conditions using a relative rate technique.
Stopped-Flow System
The schematic diagram of the stopped-flow system is
shown in Fig. 1a. These experiments have been per-
formed at a pressure of 500 mbar at 295 2 K. The
temperature of the whole laboratory and hence for the
equipment was controlled by an air conditioning sys-
tem. O3 was produced by means of a low pressure Hg-
lamp using pure O2 and analyzed downstream flush-
ing the resulting O3/O2 mixture through a gas cell
for UV-absorption measurements (path length: 10 cm,
Lambda 2, Perkin-Elmer). The O3 concentration was
determined at ␭ = 254 nm with a cross section of
17
1
␴ = 1.16 10
cm2 molecule (base e) [18]. Af-
ter further dilution of the O3/O2 mixture with O2, just
before the entrance to the reaction cell the organics
For the kinetic parameters, several investigations exist
in the literature [8–15]. The reported room tempera-
ture rate coefficients show a high variability and differ
by a factor of three (31-carene and limonene) up to
an order of magnitude (terpinolene and ␣-terpinene)
making a reinvestigation necessary. So as to set the rel-
ative rate coefficients on an absolute scale, the absolute
rate coefficients of the reference substances have to be
known with high accuracy. Therefore, for the refer-
ence substances, 2-methyl-2-butene and 2,3-dimethyl-
2-butene, used in the present study, the absolute rate
coefficients were also measured. As a by-result of the
experimental evaluation the yield of OH radicals was
determined for both reference substances. It has to be
noted that the formation pathways of OH radicals from
the gas-phase ozonolysis of alkenes and the result-
ing OH radical yields, applicable under atmospheric
conditions, are still subject of intensive research
[16,17].
Figure 1a Scheme of the experimental set-up of the
stopped-flow system.

396
WITTER ET AL.
(2-methyl-2-butene or 2,3-dimethyl-2-butene) pre-
mixed in N2 were added to the O3/O2 mixture. The
several flows were set in a way that in the final gas
mixture the carrier gas consisted of 20% O2 and 80%
resulting differential equations derived from pathway
(1)–(4) were solved numerically by an extrapolation
method connected with a Newton-technique for para-
meter estimation [20].
1
N2. The total gas flow was 10.8 l min and the bulk
residence time of the final gas mixture before the en-
trance to the reaction cell was approximately 0.3 s.
The reaction cell (volume: 2.05 l, quartz glass) was
equipped with a White mirror system (path length:
10 m) allowing FT-IR analysis of the reaction gas
(Magna 750, Nicolet). The O3 decay was followed by
the absorption centered at 1055 cm 1. For starting the
experiment the reaction cell was closed with the help of
solenoid valves. The pressure was controlled by a ca-
pacitive manometer (Baratron) and all flow rates were
set by mass flow controllers (MKS 1259). The initial
Flow-Tube
Theexperimentsunderflowconditionswereperformed
at a pressure of 100 mbar synthetic air at 295 K using a
quartz flow-tube (If T-LFT) with a length of 505 cm and
an inner diameter of 8.0 cm, cf. Fig. 1b. The If T-LFT is
surrounded with a thermo-jacket controlling the tem-
perature with a precision of 0.5 K. O3 was produced
in pure O2 outside the If T-LFT using a low pressure
Hg-lamp. After dilution with the carrier gas, the mix-
ture was introduced into the flow tube through an inlet.
Three symmetrically distributed ports on the top served
as the entrance for the organics (terpene and reference
substance) which were also diluted in the carrier gas.
At the outlet of the If T-LFT a small gas stream was
pumped continuously through a GC loop for on-line
GC-FID analysis (HP 5890) of the organics. To reduce
wall effects, the GC-loop and the transfer line were
concentrations of O3 and the organics were kept nearly
3
constant with 2.78 1013 molecule cm
.
Absolute Rate Coefficients
Using the described stopped-flow system absolute rate
coefficients were determined for 2-methyl-2-butene
and 2,3-dimethyl-2-butene. Because of the low time
resolution of the FT-IR measurements, the insuffi-
cient detection limit for O3 and a distinct wall loss
of O3,experiments under pseudo-first-order conditions
(initial [organic]/[O3] > 20) were impossible. Under
second-order conditions (here initial [organic]/[O3] =
1) the disappearance rate of the organics was influ-
enced by produced OH radicals [19]. Because the O3
absorption was influenced by absorptions arising from
products of the OH radical scavenger, experiments in
the presence of OH radical scavenger gave no useful
results. Therefore, the following generalized reaction
sequence has to be considered:
O3 + organic → xOH + y product + other products
(1)
OH + organic → other products
OH + product → other products
O3 → wall
(2)
(3)
(4)
For a better description of the OH radical balance,
besides the reaction of OH radicals with the parent or-
ganic, thereactionofOHradicalswiththe“main”prod-
uct from the ozonolysis (acetaldehyde from 2-methyl-
2-butene and acetone from 2,3-dimethyl-2-butene) was
also included. The formation yield y and the rate co-
efficients k2 and k3 were taken from literature. The
rate coefficient of the first-order wall loss for O3 (k4)
was obtained experimentally. For the determination of
the free parameters k1 and the OH radical yield x, the
Figure 1b Scheme of the experimental set-up of the flow
tube (If T-LFT: Institute for Tropospheric Research–Laminar
Flow Tube).

GAS-PHASE OZONOLYSIS
397
heated to approximately 373 K. All flows were set by
mass flow controllers (MKS 1259) and the pressure
in the tube was measured by a capacitive manometer
(Baratron). The residence time inthe flow tube variedin
the range 140–280 s assuming an ideal plug flow (bulk
velocity: 0.018–0.036 m s 1). m-Xylene was used as
the OH radical scavenger and the amount of added
m-xylene was adjusted so that more than 90% of pro-
duced OH radicals reacted with this substance. The
initial concentrations of the terpenes and the reference
substances were (1.9–4.1) 1012 molecule cm 3, of
Chemicals
The gases used had stated purities as follows: synthetic
air (99.999%), N2 (99.999%) and O2 (99.9996%)
(Linde). The organics were used as purchased:
m-xylene (>99.5%), 2-methyl-2-butene (>99.5%), ␣-
pinene ( 99.5%), 31-carene (99%), limonene
(>99%), myrcene (90%), trans-ocimene (ca.70%,
main impurity: limonene), terpinolene (>97%), ␣-
terpinene (97%) (Fluka), and 2,3-dimethyl-2-butene
( 99%) (Aldrich). Purities were checked by GC-
MS/FID.
m-xylene (3.1–5.3) 1014 molecule cm 3, and of O3
(3.0–46) 1012 molecule cm
.
3
RESULTS AND DISCUSSION
Relative Rate Technique
O3 + 2-Methyl-2-butene/2,3-Dimethyl-
Using the If T-LFT rate coefficients were obtained with
a relative rate technique, i.e. the consumption of a ter-
pene relative to that of a reference substance with a
well known rate coefficient was measured.
2-butene
First, in separate experiments the disappearance of the
individual reactants in the reaction cell of the stopped-
flow system was measured. For O3, the observed dis-
appearance was described with first-order kinetics,
k4 = (5.65 0.16) 10 4 s 1. In Fig. 2 the measured
data (open triangles) and the modeling curve are given.
In a further experiment at 950 mbar k4 was obtained to
O3 + terpene (2,3-dimethyl-2-butene) → products
(5)
O3 + reference → products
(6)
4
1
be (2.89 0.18) 10
s
indicating clearly a dif-
fusion controlled O3 disappearance to the walls and
not a reaction of O3 with possible impurities from the
For the concentrations in the absence of O3 (index “0”)
and in the presence of O3 (index “t”) the following
expression is valid:
[terpene]0
[terpene]t
k5
k6
[reference]0
[reference]t
ln
=
ln
(I)
The consumption of the organics was varied by chang-
ing the light intensity of the Hg-lamp for O3 forma-
tion by means of an aperture. As the reference sub-
stance, 2-methyl-2-butene was applied for the reac-
3
tion of ␣-pinene, 1-carene, limonene, myrcene, and
trans-ocimene. Because terpinolene and ␣-terpinene
are more reactive towards O3, the usage of 2-methyl-2-
butene was found to be inappropriate and 2,3-dimethyl-
2-butene (TME) was chosen as the reference sub-
stance for these experiments. To check the accuracy
of the measured absolute rate coefficients of the ref-
erence substances, the relative rate coefficient of 2,3-
dimethyl-2-butene (TME) with respect to 2-methyl-2-
butene was investigated additionally. The given errors
in this study take into consideration the experimental
errors 1(k5/k6) and the errors of the reference rate co-
efficient 1k6 according to Eq. (II) (2␴ limits).
s
Figure 2 Measured O3 profiles in the absence and presence
of 2-methyl-2-butene (MeBut). The curves result from the
reaction scheme given in Table I.
2
2
1k5
1(k5/k6)
k5/k6
1k6
=
+
(II)
k5
k6

398
WITTER ET AL.
carrier gas. For 2-methyl-2-butene and 2,3-dimethyl-2-
butene no significant loss was found in a time period of
1200 s.
In Fig. 2 the experimental findings of the O3 pro-
file in the absence and presence of 2-methyl-2-butene
(MeBut) are shown, [O3]0 = [MeBut]0 = 2.78 1013
molecule cm 3. Open and filled circles represent two
series of measurements. The considered pathways for
the description of the measured O3 profile from the re-
action of O3 with 2-methyl-2-butene are compiled in
Table I. The formation yield of acetaldehyde in path-
way (1a) was taken from the literature [22]. Because of
the relatively low reactivity towards OH radicals, ace-
tone formed from pathway (1a) with a yield of 0.3 was
disregarded for the consumption of OH radicals. The
16
parameter estimation yielded k1a =(4.1 0.5) 10
cm3 molecule 1 s 1 and an OH radical yield of 0.47
0.04, error limit: 2␴. The line in Fig. 2 represents the
calculated O3 profile using the estimated values for k1a
and the OH radical yield. For a time period of 820 s, it
was calculated that 85% of the produced OH radicals
were consumed in the reaction with 2-methyl-2-butene
via pathway (2a).
Figure 3 Measured O3 profiles in the presence of 2,3-
dimethyl-2-butene. The curve results from the reaction
scheme given in Table II.
In Fig. 3 the obtained O3 profiles from two series
of measurements (open and filled circles) of the reac-
tion of O3 with 2,3-dimethyl-2-butene (TME) are plot-
ted, [O3]0 = 2.77 1013 molecule cm 3, [TME]0 =
2.78 1013 molecule cm 3. The corresponding path-
ways for this reaction system are given in Table II.
The acetone formation yield of 1.0 in pathway
(1b) was taken from Ref. [24]. As a result of the
parameter estimation k1b = (1.0 0.2) 10 15 cm3
molecule 1 s 1 and an OH radical yield of 0.77 0.04
were obtained, error limit: 2␴. For a time period of
375 s, it was calculated that more than 99% of the pro-
duced OH radicals reacted with 2,3-dimethyl-2-butene
via pathway (2b) reflecting the low reactivity of the
main product acetone towards OH radicals.
To check the accuracy of the obtained absolute
rate coefficients, the relative rate coefficient of 2,3-
dimethyl-2-butene (TME) with respect to 2-methyl-
2-butene (MeBut) was measured in the If T-LFT;
p = 100mbar, [TME]0 = [MeBut]0 = (2.8–4.1) 1012
molecule cm 3, [m-xylene]0 = 4.7 1014 molecule
cm 3. In Fig. 4 the experimental findings are plot-
ted according to Eq. (II). The intercept was found to
be insignificant with 0.022 0.041. The relative rate
Table II Pathways Describing the Reaction of O3 with
2,3-Dimethyl-2-butene
k295 K
(cm3 molecule
1
1
Reaction
s
)
Table I Pathways Describing the Reaction of O3 with
2-Methyl-2-butene
O3 +
estimated
(1b)
(2b)
k295 K
1
1
Reaction
(cm3 molecule
s
)
→ x OH +
+ ..
→ ..
→ ..
O3 +
estimated
(1a)
(2a)
10a
OH +
1.1 10
→ x OH + 0.685
+ ..
11a
13b
4c
OH +
→ ..
→ ..
8.8 10
OH +
2.1 10
(3b)
(4)
11a
4b
OH +
1.6 10
(3a)
(4)
O3 → wall
5.65 10
O3 → wall
5.65 10
a Ref. [21].
b Ref. [23].
a Ref. [21].
c This study.
b This study.

GAS-PHASE OZONOLYSIS
399
[25] and Cox and Penkett [26] are approximately twice
as high as the value from the present study. An ex-
planation for this discrepancy cannot be given. The
rate coefficient from Japar et al. [27] is 20% higher
than that of this study. Although the given error limit
indicates a significant difference there is a reasonable
agreement. The rate coefficients for 295 K derived from
the Arrhenius equations given by Huie and Herron [28]
and Treacy et al. [29] are in excellent agreement with
the value reported in this study.
For 2,3-dimethyl-2-butene, only two experimental
studies exist in the literature so far. The reported value
from Japar et al. [27] is again higher as pointed out for
2-methyl-2-butene. The derived rate coefficient from
the Arrhenius equations reported by Huie and Herron
[28] is full in line with the value of this study.
Because the OH radical yield from the reaction of
O3 with alkenes is not the main topic of this work, only
a brief discussion is given here. For 2-methyl-2-butene,
the OH radical yield of 0.47 0.04 from the present
studyissignificantlylowercomparedtovaluesreported
in the literature (yields: 0.81–0.93) [30–32]. The data
from the literature were obtained using the OH radi-
cal scavenger technique measuring the products from
the OH attack on the scavenger (cyclo-hexane [30] or
2-butanol [31]) or measuring the disappearance of the
scavenger (mesitylene [32]). In the present study, no
scavenger was added and the produced OH radicals re-
acted with the parent organic (85%) and the main prod-
uct acetaldehyde (15%). If other products are formed in
considerable amounts and these products are very re-
active towards OH radicals, the OH radical yield can be
underestimated using the reaction scheme from Table I.
But, there is no experimental evidence in the literature
for the occurrence of such reactive products.
Figure 4 Experimental data for the kinetics of the reaction
of O3 with 2,3-dimethyl-2-butene (TME) by using 2-methyl-
2-butene (MeBut) as the reference substance.
coefficient was k5/k6 = 2.53 0.10 (here 2-methyl-2-
butene was used as the reference substance) to be in ex-
cellent agreement with the ratio of the absolute rate co-
efficients from the present study, k1b/k1a = 2.4 0.6.
This fact indicates the usefulness of the obtained abso-
lute rate coefficients and the consistence of the results
by using different experimental approaches.
In Table III, absolute rate coefficients for the reac-
tion of O3 with 2-methyl-2-butene and 2,3-dimethyl-2-
butene available from the literature are compared with
the data from the present study. For 2-methyl-2-butene,
the reported rate coefficients by Bufalini and Altschuler
For 2,3-dimethyl-2-butene, the OH radical yield of
0.77 0.04 from the present study is on the lower
end of the reported literature data (yields: 0.7–1.0)
[24,30,31]. There is an excellent agreement with the
Table III Absolute Rate Coefficients for the Reaction of O3 with 2-Methyl-2-butene and 2,3-Dimethyl-2-butene
1
1
Reactant
Temperature (K)
303
k (cm3 molecule
s
)
Ref.
16a
16a
2-Methyl-2-butene
7.47 10
7.97 10
[25]
[26]
[27]
[28]
[29]
295
299
2
2
16
(4.93 0.16) 10
16b
295
295
295
299
295
295
3.86 10
3.89 10
16b
16
2
2
(4.1 0.5) 10
This study
[27]
[28]
This study
15
2,3-Dimethyl-2-butene
(1.51 0.08) 10
15b
1.04 10
15
2
(1.0 0.2) 10
a No errors given in the original work.
b Value taken from Arrhenius plot.

400
WITTER ET AL.
values of 0.7 0.1 derived from the stoichiometry
1[TME]/[O3] from Niki et al. [24] and 0.80 0.12
from Chew and Atkinson [31] using the scavenger
technique with 2-butanol. The reported yield of 1.0
from experiments with cyclo-hexane [30] is signifi-
cantly higher.
O3+ Terpenes
All these experiments were performed with a large ex-
cess of m-xylene scavenging more than 90% of pro-
duced OH radicals. The room-temperature rate coeffi-
cient for the reaction of O3 with m-xylene is reported
to be 2.9 10 21cm3 molecule 1 s 1 [35] making a
conversion of m-xylene by O3 negligible under the con-
ditions used. 2-Methyl-2-butene was used as the ref-
Furthermore, using a modeling study, it was tested
whether or not the wall loss of OH radicals was impor-
tant for the OH radical balance under the experimental
conditions. To account properly for gas losses to all
walls of the cell (top, bottom, side) 3-D simulations
were performed using the software package FLUENT
5.4 [33]. Input parameters for the calculations were
the actual cell geometry, the gas diffusion coefficient
for OH radicals (D(OH) = 0.6 cm2 s 1, p = 500 mbar,
T = 295 K [34]), a constant initial OH radical concen-
tration for the whole gas volume, and a zero OH radi-
cal concentration on the walls as the boundary condi-
tion. From the calculated OH time profile in the actual
measurement volume (volume of the FT-IR paths) a
first-order rate coefficient for the diffusion limited wall
3
erence substance for ␣-pinene, 1-carene, limonene,
myrcene, and trans-ocimene. For the more reactive
substances terpinolene and ␣-terpinene the more re-
active 2,3-dimethyl-2-butene was chosen as the refer-
ence substance. To set the relative rate coefficients on
an absolute scale, the absolute rate coefficients for 2-
methyl-2-butene and 2,3-dimethyl-2-butene measured
in the present study were used.
An assessment of the wall loss processes of O3 for
the different reactants was carried out in the follow-
ing way. The diffusion limited wall loss with a first-
order rate coefficient kwall
4 D/r2 (D: diffusion
1
coefficient, r: tube radius [36]) was calculated to
loss of OH radicals of 0.41 s was derived. Even, as-
be kwall(O3) 0.45 s (D(O3) = 1.8 cm2 s 1, p =
1
suming a 90% conversion of 2-methyl-2-butene (2,3-
dimethyl-2-butene) in the present experiments, at this
concentration the OH radical life time with respect to
reaction (2a) from Table I (reaction (2b) from Table II)
is 0.004 s (0.003 s) to be more than two orders of magni-
tude shorter compared to the diffusion limited wall loss
(life time: 2.4 s) making the wall loss of OH radicals
negligible.
100 mbar, T = 295 K [34]; r = 4 cm). Because both of
thereactioncellofthestopped-flowsystemandtheflow
tube were manufactured from quartz glass, it was as-
sumed that the effective wall loss for O3 in the flow tube
was also 1/425 of the diffusion limited value as pointed
out for the reaction cell of the stopped-flow system.
This assumption leads to an effective first-order rate co-
efficient for the O3 wall loss of 1 10 3 s 1 (life time
with respect to wall loss: 103 s). The initial concentra-
tions of the added terpenes and the reference substances
The same considerations, as pointed out for OH
radicals, were performed for O3, 2-methyl-2-butene
(MeBut), 2,3-dimethyl-2-butene (TME) and addition-
ally for ␣-pinene. This was needed to estimate wall loss
processes in the flow tube. For O3 a diffusion limited
were in the order of (2–4) 1012 molecule cm and
the corresponding rate coefficients of the ozonolysis
ranged from 5 10
resulting in initial life times of O3 with respect to the
ozonolysis in the range of 25–104 s. This indicates
that for the less reactive terpenes the wall loss of O3
represented an important process. Using the same ap-
proach as discussed for O3, diffusion limited first-order
rate coefficient for the wall loss of 2-methyl-2-butene
(MeBut), 2,3-dimethyl-2-butene (TME), and terpenes
(␣-pinene) were found to be kwall(MeBut) 0.24 s
kwall(TME) 0.2 s and kwall(␣-pinene) 0.15 s
Taking into account that the effective rate coefficients
are lower than 1/5000 of the diffusion limited values (as
derived from the stopped-flow system), for a residence
time of 140 s less than 0.7% of the injected organics
were deposited on the walls. Therefore, the wall loss
of the organics was negligible under the experimental
conditions used.
3
17
14
1
1
to 10
cm3 molecule
s
1
first-order rate coefficient of 0.24 s (D(O3) = 0.36
cm2 s 1, p = 500 mbar, T = 295 K [34]) was cal-
culated. Using the experimental rate coefficient k4 =
4
1
(5.65 0.16) 10
s
it was found that the effec-
tive first-order rate coefficient for the O3 wall loss is
1/425 of the diffusion limited value. For MeBut, TME,
and ␣-pinene the diffusion limited first-order rate co-
efficients at p = 500 mbar and T = 295 K were calcu-
1
1
1
lated to be 0.13 s (D(MeBut) = 0.19 cm2 s [34]),
,
.
1
1
0.11 s 1 (D(TME) = 0.16 cm2 s 1 [34]), and 0.08 s
1
(D(␣-pinene) = 0.12 cm2 s [34]), respectively. As
1
mentioned earlier, in a time period of 1200 s for 2-
methyl-2-butene and 2,3-dimethyl-2-butene no signif-
icant loss was found. The same behavior was observed
for ␣-pinene. Assuming a decay of the organics 2%
after 1200 s an “experimental” first-order rate coeffi-
5
1
cient 1.7 10
s
follows and the effective first-
The experimental findings for the relative rate coef-
ficients of the terpenes are plotted for the two reference
order rate coefficients for the wall loss of the organics
is 1/5000 of the diffusion limited rate coefficients.

GAS-PHASE OZONOLYSIS
401
Figure 5 Experimental data for the kinetics of the reac-
tion of O3 with selected terpenes by using 2-methyl-2-butene
(MeBut) as the reference substance.
Figure 6 Experimental data for the kinetics of the reac-
tion of O3 with ␣-terpinene and terpinolene by using 2,3-
dimethyl-2-butene (TME) as the reference substance.
substances 2-methyl-2-butene (MeBut) and 2,3-
dimethyl-2-butene(TME)inFigs. 5and6, respectively.
The results from the linear regression analyses are sum-
marized in Table IV. In all cases, the plots according to
Eq. (I) showed a straight line and the intercepts were
insignificant. In Table V the derived rate coefficients
of the present study are given together with literature
data for comparison.
Generally, the values from the present study are in
reasonable agreement with those from the Atkinson
group [11,13,14] and in the case of ␣-pinene with
those from Ripperton et al. [8], Nolting et al. [12],
and Khamaganov and Hites [15]. The reported absolute
rate coefficients from Grimsrud et al. [10] are all higher
by a factor of 1.3 for ␣-pinene up to 6.4 for terpino-
lene. This overestimation is probably caused by sys-
tematic experimental problems as pointed out earlier
[13]. The absolute rate coefficient from Japar et al. [9]
for ␣-pinene is higher compared to the value of the
present study by a factor of three. On the other hand,
for terpinolene the reported value from Japar et al. [9] is
somewhat lower compared to the other rate coefficients
published to date indicating no systematic errors in
the measurements of Japar et al. [9]. Neglecting the rate
3
coefficients from Grimsrud et al. [10], for 1-carene
and limonene the relative rate coefficients from the
Table IV Obtained Data from the Gas-Phase Reaction of O3 with a Series of Terpenes in the Presence of an OH
Radical Scavenger, T = 295 K, p = 100 mbar Synthetic Air
1
1
Reactant
Reference Substance
Intercept
k5/k6
k5 (cm3 molecule
s
)
16
17
16
16
16
15
14
␣-Pinene
2-Methyl-2-butene
2-Methyl-2-butene
2-Methyl-2-butene
2-Methyl-2-butene
2-Methyl-2-butene
2,3-Dimethyl-2-butene
2,3-Dimethyl-2-butene
0.005 0.016
0.002 0.015
0.003 0.012
0.041 0.071
0.008 0.062
0.008 0.024
0.012 0.014
0.270 0.022
0.143 0.018
0.622 0.024
1.17 0.06
1.35 0.08
1.60 0.06
14.7 1.7
(1.1 0.2) 10
(5.9 1.0) 10
(2.5 0.3) 10
(4.8 0.6) 10
(5.5 0.8) 10
(1.6 0.4) 10
(1.5 0.4) 10
31-Carene
Limonene
Myrcene
trans-Ocimene
Terpinolene
␣-Terpinene

402
WITTER ET AL.
Table V Rate Coefficients for the Reaction of O3 with a Series of Terpenes
1
1
Reactant
Method
k295 K (cm3 molecule
s
)
Ref.
16a
␣-Pinene
A
A
A
A
1.69 10
(3.3 0.3) 10
1.46 10
(294 K)
[8]
[9]
16
(298 K)
16a
[10]
[11]
[12]
[13]
[15]
17
(8.3 1.3) 10
(8.6 1.3) 10
(9.71 1.06) 10
(8.41 0.74) 10
17
RR (cis-2-butene)
A
RR (1-butene and
2-methyl-propene)
RR (2-methyl-2-butene)
A
(297 K)
(296 K)
(298 K)
17
17
16
(1.1 0.2) 10
1.22 10
(5.20 0.56) 10
This study
[10]
[13]
[13]
This study
[10]
16a
31-Carene
17
A
(296 K)
17
RR (␣-pinene)
RR (2-methyl-2-butene)
A
(3.87 0.40) 10
(296 K)
17
(5.9 1.0) 10
6.49 10
(2.09 0.22) 10
(2.01 0.07) 10
(2.13 0.15) 10
16a
Limonene
16
RR (␣-pinene)
(296 K)
(296 K)
(298 K)
[13]
[14]
[15]
16
RR (cis-2-butene)
RR (cis-2-butene, 2-methyl-
propene and 1-butene)
RR (2-methyl-2-butene)
A
RR (limonene)
RR (2-methyl-2-butene)
RR (limonene)
RR (2-methyl-2-butene)
A
16
16
(2.5 0.3) 10
1.26 10
(4.85 0.78) 10
(4.8 0.6) 10
(5.56 0.85) 10
(5.5 0.8) 10
(7.3 1.1) 10
1.02 10
(1.41 0.25) 10
(1.88 0.08) 10
(1.6 0.4) 10
8.94 10
(8.66 2.33) 10
This study
[10]
[13]
This study
[13]
15a
Myrcene
16
(296 K)
16
16
trans-Ocimene
(296 K)
16
This study
[9]
16
14a
Terpinolene
(298 K)
A
[10]
[13]
[14]
15
15
RR (limonene)
RR (2,3-dimethyl-2-butene)
RR (2,3-dimethyl-2-butene)
A
RR (terpinolene)
RR (2,3-dimethyl-2-butene)
RR (␤-caryophyllene)
RR (2,3-dimethyl-2-butene)
(296 K)
(296 K)
15
This study
[10]
14a
␣-Terpinene
15
(296 K)
(296 K)
[13]
[14]
[14]
14
(2.69 0.90) 10
(2.11 0.22) 10
14
(296 K)
14
(1.5 0.4) 10
This study
A: absolute method; RR: relative rate technique, reference substance is given in parentheses.
a No errors given in the original work.
present study are on the upper end of the range of the
reported data. In most cases, however, the differences
are not significant. For myrcene and trans-ocimene
the agreement of the relative rate coefficients from the
present study with those reported from Atkinson et al.
[13] is excellent. Disregarding again the Grimsrud et
al. data [10] as well as the given value for terpinolene
from Japar et al. [9], for terpinolene and ␣-terpinene the
existing rate coefficients in the literature from Atkin-
son et al. [13] and Shu et al. [14] were confirmed by
rate coefficients from the present study.
pounds are reported using two different approaches: (i)
a stopped-flow system and (ii) a flow tube. As a result of
the stopped-flow experiments absolute rate coefficients
at 295 2 K for the reaction of O3 with 2-methyl-2-
16
1
butene, k1a = (4.1
0.5)
10
cm3 molecule
s
1, and 2,3-dimethyl-2-butene, k1b = (1.0 0.2)
10 15cm3 molecule
s
1, were determined. The
1
OH radical yields were found to be 0.47
0.04 for
2-methyl-2-butene and 0.77 0.04 for 2,3-dimethyl-2-
butene. The determination of relative rate coefficients
for the reaction of O3 with a series of terpenes was per-
formed under flow-tube conditions using 2-methyl-2-
butene and 2,3-dimethyl-2-butene as the reference sub-
stances. m-XylenewasusedforOHradicalscavenging.
To set the relative rate coefficients on an absolute
scale, the determined data for the reference substances
SUMMARY
In the present work, kinetic investigations concerning
the gas-phase reactions of O3 with unsaturated com-

GAS-PHASE OZONOLYSIS
403
resulting from the stopped-flow experiments were
chosen. The resulting rate coefficients are (T = 295
0.5 K, unit: cm3 molecule 1 s 1): ␣-pinene: (1.1 0.2)
13. Atkinson, R.; Hasegawa, D.; Aschmann, S. M. Int J
Chem Kinet 1990, 22, 871.
14. Shu, Y.; Atkinson, R. Int J Chem Kinet 1994, 26, 1193.
15. Khamaganov, V. G.; Hites, R. A. J Phys Chem 2001,
105, 815.
16. Kroll, J. H.; Clarke, J. S.; Donahue, N. M.; Anderson, J.
G.; Demerjian, K. L. J Phys Chem 2001, 105, 1554.
17. Kroll, J. H.; Sahay, S. R.; Anderson, J. G.; Demerjian,
K. L.; Donahue, N. M. J Phys Chem 2001, 105, 4446.
18. Molina, L. T.; Molina, M. J. J Geophys Res 1986, 91,
14501.
3
10 16, 1-carene: (5.9 1.0) 10 17, limonene:
16
(2.5 0.3) 10 16, myrcene: (4.8 0.6) 10
,
trans-ocimene:(5.5 0.8) 10 16, terpinolene:(1.6
14
0.4) 10 15, and ␣-terpinene: (1.5
0.4)
10
.
Generally, the presented set of rate coefficients for a
series of terpenes can help to evaluate existing data
and improve the reliability of recommended data.
19. Calvert, J. G.; Atkinson, R.; Kerr, J. A.; Madronich, S.;
Moortgat, G. K.; Wallington, T. J.; Yarwood, G. The
Mechanism of Atmospheric Oxidation of the Alkenes;
Oxford University Press: New York, 2000.
20. Stoer, J.; Bulirsch, R. Einfu¨hrung in die Numerische
Mathematik II; Springer-Verlag: Berlin, 1978.
21. Atkinson, R. Chem Rev 1986, 86, 69.
22. Grosjean, E.; Grosjean, D. Environ Sci Technol 1996,
30, 859.
The authors thank Thomas Conrath and Herwig
Macholeth for technical assistance.
BIBLIOGRAPHY
1. Gu¨nther, A.; Hewitt, C. N.; Erickson, D.; Fall, A.; Geron,
C.; Graedel, T.; Harley, P.; Klinger, L.; Lerdau, M.;
McKay, W. A.; Pierce, T.; Scholes, B.; Steinbrecher, R.;
Tallamraju, R.; Taylor, J.; Zimmermann, P. J Geophys
Res 1995, 100, 8873.
2. Mu¨ller, J. F. J Geophys Res 1992, 97, 3787.
3. Finlayson-Pitts, B. J.; Pitts, J. N. Jr. Chemistry of the Up-
per and Lower Atmosphere; Academic Press: San Diego,
2000 (and references therein).
23. Atkinson, R.; Baulch, D. L.; Cox, R. A.; Hampson, R.
F.; Kerr, J. A.; Rossi, M. J.; Troe, J. J Phys Chem Ref
Data 1997, 26, 521.
24. Niki, H.; Maker, P. D.; Savage, C. M.; Breitenbach, L.
P.; Hurley, M. D. J Phys Chem 1987, 91, 941.
25. Bufalini, J. J.; Altschuler, A. P. Can J Chem 1965, 43,
2243.
26. Cox, R. A.; Penkett, S. A. J Chem Soc, Faraday Trans1
1972, 68, 1735.
4. Calogirou, A.; Larsen, B. R.; Kotzias, D. Atmos Environ
1999, 33, 1423 (and references therein).
5. Hoffmann, T.; Odum, J. R.; Bowman, F.; Collins, D.;
Klockow, D.; Flagan, R. C.; Seinfeld, J. H. J Atmos
Chem 1997, 26, 189.
27. Japar, S. M.; Wu, C. H.; Niki, H. J Phys Chem 1974, 78,
2318.
28. Huie, R. E.; Herron, J. T. Int J Chem Kinet Symp 1975,
1, 165.
29. Treacy, J.; El Hag, M.; O’Farrell, D.; Sidebottom, H. Ber
Bunsenges Phys Chem 1992, 96, 422.
6. Went, F. W. Nature 1960, 187, 641.
7. Paulson, S. E.; Orlando, J. J. Geophys Res Lett 1996,
23, 3727.
30. Atkinson, R.; Aschmann, S. M. Environ Sci Technol
1993, 27, 1357.
8. Ripperton, L. A.; Jeffries, H. E.; White, O. Adv Chem
Ser 1972, 113, 219.
31. Chew, A. A.; Atkinson, R. J Geophys Res 1996, 101,
28649.
9. Japar, S. M.; Wu, C. H.; Niki, H. Environ Lett 1974, 7,
245.
32. McGill, C. D.; Rickard, A. R.; Johnson, D.; Marston, G.
Chemosphere 1999, 38, 1205.
10. Grimsrud, E. P.; Westberg, H. H.; Rasmussen, R. A. Int
J Chem Kinet Symp 1975, 1, 183.
11. Atkinson, R.; Winer, A. M.; Pitts, J. N. Jr. Atmos Environ
1982, 16, 1017.
12. Nolting, F.; Behnke, W.; Zetzsch, C. J Atmos Chem
1988, 6, 47.
33. FLUENT 5.4, Fluent Inc., Lebanon, New York, 1998.
34. Reid, R. C.; Prausnitz, J. M.; Poling, B. E. The Properties
of Gases and Liquids; McGraw-Hill: New York, 1987.
35. Kramp. F.; Paulson, S. E. J Phys Chem 1998, 102, 2685.
36. Seinfeld, J. H. Atmospheric Chemistry and Physics of
Air Pollution; Wiley: New York, 1986.