Kinetics and mechanisms of OH-initiated oxidation of small unsaturated alcohols

Article abstract of DOI:10.1002/kin.20475

Smog chamber relative rate techniques were used to measure rate coefficients of (5.00 ± 0.54) × 10^-11, (5.87 ± 0.63) × 10^-11, and (6.49 ± 0.82) × 10 ^-11 cm³ molecule^-1 s^-1 in 700 Torr air at 296 ± 1 K for reactions of OH radicals with allyl alcohol, 1-buten-3-ol, and 2-methyl-3-buten-2-ol, respectively; the quoted uncertainties encompass the extremes of determinations using two different reference compounds. The OH-initiated oxidation of allyl alcohol in the presence of NO_x gives glycolaldehyde in a molar yield of 0.85 ± 0.08; the quoted uncertainty is two standard deviations. Oxidation of 2-methyl-3-buten-2- ol gives acetone and glycolaldehyde in molar yields of 0.66 ± 0.06 and 0.56 ± 0.05, respectively. The reaction of OH radicals with allyl alcohol, 1-buten-3-ol, and 2-methyl-3-buten-2-ol proceeds predominately via addition to the >C=CH₂ double bond with most of the addition occurring to the terminal carbon.

Full text of DOI:10.1002/kin.20475

Kinetics and Mechanisms of
OH-Initiated Oxidation of
Small Unsaturated Alcohols
1
2
2
KENSHI TAKAHASHI, MICHAEL D. HURLEY, TIMOTHY J. WALLINGTON
1Pioneering Research Unit and Research Institute for Sustainable Humanosphere, Kyoto University,
Gokasho, Uji, 611-0011, Japan
2
System Analytics & Environmental Sciences Department, Ford Motor Company, P. O. Box: 2053,
Dearborn, MI 48121-2053
Received 11 August 2009; revised 8 October 2009, 19 October 2009; accepted 20 October 2009
DOI 10.1002/kin.20475
Published online in Wiley InterScience (www.interscience.wiley.com).
ABSTRACT: Smog chamber relative rate techniques were used to measure rate coefficients of
5.00 ± 0.54) × 10− , (5.87 ± 0.63) × 10 , and (6.49 ± 0.82) × 10
11
−11
−11
cm molecule s
3
−1 −1
(
in 700 Torr air at 296 ± 1 K for reactions of OH radicals with allyl alcohol, 1-buten-3-ol,
and 2-methyl-3-buten-2-ol, respectively; the quoted uncertainties encompass the extremes of
determinations using two different reference compounds. The OH-initiated oxidation of allyl
alcohol in the presence of NOx gives glycolaldehyde in a molar yield of 0.85 ± 0.08; the quoted
uncertainty is two standard deviations. Oxidation of 2-methyl-3-buten-2-ol gives acetone and
glycolaldehyde in molar yields of 0.66 ± 0.06 and 0.56 ± 0.05, respectively. The reaction of OH
radicals with allyl alcohol, 1-buten-3-ol, and 2-methyl-3-buten-2-ol proceeds predominately
via addition to the >C CH2 double bond with most of the addition occurring to the terminal
carbon. ꢀC 2010 Wiley Periodicals, Inc. Int J Chem Kinet 42: 151–158, 2010
INTRODUCTION
of allyl alcohol (CAS registry # 107-18-6), 1-buten-3-
ol (#598-32-3), and 2-methyl-3-buten-2-ol (#115-18-
4) in 700 Torr air at 296 ± 1 K.
Volatile organic compounds (VOCs) from natural and
industrial sources are important precursors of tro-
pospheric O3 [1–3]. Unsaturated alcohols are emit-
ted into the atmosphere in signiÞcant quantities from
natural sources, for example, 2-methyl-3-buten-2-ol
OH + CH2 CHCH2OH → Products, k1
(1)
OH + CH2 CHCH(CH3)OH → Products, k2 (2)
(MBO232) is emitted from certain pine species [4–6].
OH + CH2 CHC(CH )2OH → Products, k3 (3)
3
To assess the impact of VOCs emitted from industrial
sources, it is necessary to understand the kinetics and
mechanisms of the atmospheric oxidation of naturally
emitted VOCs. To contribute to this understanding, we
have conducted smog chamber studies of the kinetics
and mechanisms of the OH radical initiated oxidation
There have been six studies of reaction (1) (Papagni
et al. [7], Orlando et al. [8], Upadhyaya et al. [9],
Holloway et al. [10], Le Person et al. [11], and Parker
and Espada-Jallad [12]), and values of k1 in the range
−
11
3
−1 −1
(
3.7–6.8) × 10
cm molecule s have been
reported at room temperature. There has been one
study of reaction (2); Papagni et al. [7] measured
Correspondence to: Kenshi Takahashi; e-mail: tkenshi@rish.
kyoto-u.ac.jp; Timothy J. Wallington; e-mail: twalling@ford.com.
ꢀ
c 2010 Wiley Periodicals, Inc.
k2 = 5.9 × 10− cm molecule
11
3
−1 −1
s at 296 K. Nine

1
52
TAKAHASHI, HURLEY, AND WALLINGTON
experimental studies of the kinetics of reaction of OH
radicals with 2-methyl-3-buten-2-ol have been con-
ducted (Papagni et al. [7], Rudich et al. [13], Fantechi
et al. [14], Ferronato et al. [15], Alvarado et al. [16],
Reisen et al. [17], Imamura et al. [18], Baasandorj and
Stevens [19], and Carrasco et al. [20]). Values of k3 at
constants for reactions of OH radicals with the alco-
hol and reference, respectively. Plots of ln([alcohol]t /
0
[alcohol]t ) versus ln([reference]t /[reference]t ) should
0
be linear, pass through the origin and have a slope of
kunsaturated alcohol/kreference.
The loss of the reactant and reference compounds
was monitored by FT-IR spectroscopy using an in-
frared path length of 27 m and a resolution of
−11
3
room temperatures in the range (4.3–6.6) × 10
cm
−1
s have been reported. Acetone, glyco-
−1
molecule
−1
laldehyde, and 2-hydroxy-2-methylpropanal have been
identiÞed as major products of the OH-initiated oxida-
tion of 2-methyl-3-buten-2-ol [7,14–17,21].
0.25 cm . Infrared spectra were derived from 32 coad-
ded interferograms. Analysis of the IR spectra was
achieved through spectral stripping, in which small
fractions of the reference spectrum were subtracted
incrementally from the sample spectrum. The concen-
trations of reactant and reference compounds were de-
termined with a precision of ± 1% of their initial con-
centrations. All experiments were performed in 700
Torr synthetic air diluent at 296 ± 1 K. Alcohols were
obtained from commercial sources at >99% purity and
were subjected to repeated freeze/pump/thaw cycling
before use. CH3ONO was synthesized by the dropwise
addition of concentrated H2SO4 to a saturated solution
of NaNO2 in methanol.
In this paper, we report the results of a smog
chamber/FT-IR study of the kinetics and mechanism
of OH-initiated oxidation reactions (1)–(3) in 700 Torr
air. Relative rate techniques were used to measure the
rate coefÞcients k1–k3, at 296 ± 1 K. In addition, prod-
uct studies of the OH radical initiated oxidation of
allyl alcohol and 2-methyl-3-buten-2-ol are reported.
The results are compared with the literature data.
EXPERIMENTAL
Experiments were performed using the smog cham-
ber system at Ford Motor Company, consisting of a
RESULTS AND DISCUSSION
140-L Pyrex reactor interfaced to a Mattson Sirus 100
spectrometer [22]. The reactor was surrounded by 22
ßuorescent blacklamps (GE F15T8-BL). OH radicals
were produced by photolysis of CH3ONO in the pres-
ence of NO in synthetic air.
Kinetics of the Reactions of OH with
Unsaturated Alcohols
The kinetics of reactions (1)–(3) were measured rela-
tive to reactions (8) and (9).
CH3ONO + hν → CH3O + NO
CH3O + O2 → HO2 + H2CO
HO2 + NO → OH + NO2
(4)
(5)
(6)
OH + C3H6 → Products (8)
OH + 1, 3, 5-Trimethylbenzene → Products (9)
Figure 1 shows the observed loss of allyl alco-
hol versus propene and 1,3,5-trimethylbenzene follow-
ing UV irradiation of allyl alcohol/reference/CH3ONO
mixtures in 700 Torr air diluent. Consistent with
expectations (see the preceding section) plots of
Relative rate techniques were used to measure rate con-
stants for reactions of OH radicals with unsaturated al-
cohols relative to reference reactions whose rate con-
stants have been established previously. The relative
rate method is a well-established technique for mea-
suring the reactivity of OH radicals with organic com-
pounds [23]. Kinetic data were derived by monitoring
the loss of the alcohol relative to reference compounds
ln([allyl alcohol]t /[allyl alcohol]t ) versus ln([1,3,5-
TMB]t /[1,3,5-TMB]t ) were linear and extrapolated
0
0
to the origin. Linear least-squares analyses of the
data in Fig. 1 give k1/k8 = 1.85 ± 0.14 and k1/k9 =
0.908 ± 0.063. The quoted uncertainties include two
standard deviations from linear regression analyses
and uncertainties in the analysis of the spectra. Us-
(
propene and 1,3,5-trimethylbenzene). The decays of
the alcohol and reference were plotted using the fol-
lowing expression:
−
11
−11
[25]
ing k8 = 2.6 × 10
[24] and k9 = 5.7 × 10
ꢀ
ꢁ
ꢀ
ꢁ
−11
[
alcohol]t0
kalcohol
[reference]t0
[reference]t
at 700 Torr, we derived k = (4.81 ± 0.37) × 10
1
ln
=
ln
(7)
−11
3
−1 −1
and k1 = (5.18 ± 0.36) × 10
cm molecule s .
[
alcohol]t
kreference
The results for k1 obtained using the two different
reference compounds are indistinguishable within the
experimental uncertainties. We choose to cite a Þnal
value which is the average of the two determinations
where [alcohol]t0 , [alcohol]t , [reference]t0 , and
reference]t are the concentrations of alcohol and ref-
erence at times t0 and t, kalcohol and kreference are the rate
[
International Journal of Chemical Kinetics DOI 10.1002/kin

OH-INITIATED OXIDATION OF SMALL UNSATURATED ALCOHOLS
153
−
11
cm3
k(OH + 3-buten-1-ol) = (5.87 ± 0.63) × 10
−1
−1
molecule
s is in excellent agreement with the pre-
2
1
1
0
0
.0
.5
.0
.5
.0
t
vious measurement by Papagni et al. [7].
Our determination of k(OH + 2-methyl-3-buten-2-
−
11
3
−1 −1
C H
ol) = (6.49 ± 0.82) × 10
cm molecule s is in
3
6
excellent agreement with the relative rate studies con-
ducted at, or near, atmospheric pressure by Ferronato
et al. [15], Papagni et al. [7], Imamura et al. [18], and
Carrasco et al. [20], and the absolute rate study by
Rudich et al. [13] in 90–100 Torr helium diluent. The
measurement by Fantechi et al. [14] is approximately
t
1,3,5-TMB
35% below the results from the present and all other
previous determinations. In the study by Fantechi et
al. [14], OH radicals were generated by the 254-nm
photolysis of H2O2 using mercury lamps. In the rela-
tive rate studies by Ferronato et al. [15], Papagni et al.
0
.0
0.4
0.8
1.2
[7], Imamura et al. [18], Carrasco et al. [20], OH rad-
ln ([Reference]t0 / [Reference]t)
icals were generated via photolysis of alkyl nitrites or
HONO at wavelengths >280 nm. It seems likely that
the underestimation of k3 by Fantechi et al. [14] was
caused by complications (e.g., ozone formation) asso-
ciated with the short wavelength UV radiation from the
mercury lamps used.
Figure 1 Loss of allyl alcohol versus C3H6 (diamonds)
and 1,3,5-trimethylbenzene (hexagons) following exposure
to OH radicals in 700 Torr air at 296 K.
with error limits that encompass the extremes of
the individual determinations: k1 = (5.00 ± 0.54) ×
Pfrang et al. [26] have proposed a method to pre-
dict the gas-phase rate coefÞcients for reactions of
−10
3
−1 −1
10
cm molecule s . We estimate that uncer-
3
tainties in the reference rate coefÞcients contribute an
additional 10% uncertainty to k1.
OH, NO , O , and O( P) with unsaturated alcohols
3
3
based on structure–activity relations (SAR). Pfrang
et al. [24] noted what appeared to be a surprising
discrepancy for allyl alcohol with the experimental
value based on the measurements by Upadhyaya et al.
[9] and Papagni et al. [7] being substantially lower
than that predicted by the SAR. As discussed above,
the measurements by Upadhyaya et al. [9] were con-
ducted at low pressure and are an underestimate of
the rate coefÞcient at atmospheric pressure. The value
1
As seen from Table I, the rate coefÞcient ratios mea-
sured for allyl alcohol in the present work are indistin-
guishable from those reported by Papagni et al. [7], Or-
lando et al. [8], and Hollowell et al. [10] in experiments
conducted in 1 atm of air, 700 Torr air, and 100–300
Torr helium, respectively. Our rate coefÞcients are con-
sistent with the relative and absolute rate results from
Le Person et al. [11] and Parker and Espada-Jallad
[
12]. The pulsed laser photolysis—laser-induced ßuo-
rescence (PLP-LIF) measurements by Upadhyaya et al.
9] in 10–20 Torr argon give a rate coefÞcient that is
approximately 25% lower than our measurement in
00 Torr air. The magnitude of this apparent pressure
of k predicted by Pfrang et al. [26] from the SAR
−
11
(5.44 × 10 ) is in good agreement with the value of
−11
3
−1 −1
[
k = (5.50 ± 0.54) × 10
cm molecule s mea-
1
sured in the present work (see Table I). Similarly, the
−11
−11
7
values of k = 5.83 × 10
and k = 6.45 × 10
2
3
3
−1 −1
effect is comparable to that observed in the reaction
of OH radicals with propene [24] and presumably re-
ßects the pressure dependence of the addition channel
in reaction (1).
cm molecule
s
predicted by Pfrang et al. [26] are
consistent with our experimental determinations.
The rate coefÞcient ratios k2/k8, k2/k9, k3/k8, and
k3/k9 measured in the present work are given in Table I.
As seen from Table I, when placed on an absolute ba-
Products and Mechanism of OH-Initiated
Oxidation of 2-Methyl-3-buten-2-ol
(MBO232)
−11
−11
sis using k8 = 2.6 × 10
[24] and k9 = 5.7 × 10
−1 −1
s [25], the rate coefÞcients k2 and k3
3
cm molecule
The mechanism of OH-initiated oxidation of MBO232
was investigated by irradiating mixtures of 46.3–51.3
mTorr MBO232, 71.3–75.8 mTorr methyl nitrite and
10.3–47.9 mTorr NO in 700 Torr air. Infrared ab-
sorption spectra obtained before (Fig. 1A) and after
(Fig. 1B) a 2-min irradiation of a mixture containing
derived using the two different reference compounds
are in agreement. Taking averages of the individual
determinations together with error limits that encom-
pass the extremes of the individual determinations
gives the results listed in Table I. Our determination of
International Journal of Chemical Kinetics DOI 10.1002/kin

1
54
TAKAHASHI, HURLEY, AND WALLINGTON
Table I Rate Coefficients for OH Reactions with Allyl Alcohol, 3-Buten-1-ol, and 2-Methyl-3-butene-2-ol at Room
Temperature
Rate CoefÞcient
−
11
3
(
× 10
cm
s
−
1
−1
Experimental Methoda
RR (propene)
Alcohol
Rate Ratio
molecule
)
References
This work
This work
This work
Allyl alcohol
1.85 ± 0.13
4.81 ± 0.34
5.18 ± 0.36
0
.908 ± 0.063
RR (1,3,5-trimethylbenzene)
b
5
.00 ± 0.54
c
1
.72 ± 0.12
4.5 ± 0.6
RR (propene)
RR (1,3,5-trimethylbenzene)
PLP-LIF
PLP-LIF
PLP-LIF
RR (butyl vinyl ether)
RR (isopropenyl acetate)
RR (1,3,5-trimethylbenzene)
Orlando et al. [8]
Papagni et al. [7]
Upadhyaya et al. [9]
Holloway et al. [10]
Le Person et al. [11]
Le Person et al. [11]
Le Person et al. [11]
Parker and
Espada-Jallad [12]
Parker and
Espada-Jallad [12]
This work
This work
This work
Papagni et al. [7]
This work
This work
d
0
.958 ± 0.06
5.46 ± 0.35
e
3
.7 ± 0.5
f
4
.60 ± 0.19
g
5
.0 ± 0.5
h
0
0
1
.48 ± 0.01
.77 ± 0.01
.04 ± 0.12
4.8 ± 0.1
h
5.4 ± 0.1
i
6.13 ± 0.69
6.77 ± 0.94
i
1
.31 ± 0.18
RR (α-pinene)
3
2
-Buten-1-ol
2.17 ± 0.14
.07 ± 0.07
5.64 ± 0.36
6.10 ± 0.40
RR (propene)
RR (1,3,5-trimethylbenzene)
1
b
5
.87 ± 0.63
d
0
.965 ± 0.035
5.93 ± 0.23
6.73 ± 0.57
6.24 ± 0.51
RR (1,3,5-trimethylbenzene)
RR (propene)
RR (1,3,5-trimethylbenzene)
-Methyl-3-butene-2-ol
2.59 ± 0.22
1
.09 ± 0.09
.38 ± 0.08
b
6
.49 ± 0.82
This work
Rudich et al. [13]
Fantechi et al. [14]
j
6.4 ± 0.6
PLP-LIF
RR (isoprene)
RR (propene)
RR (ethene)
RR (propene)
RR (1,3,5-trimethylbenzene)
RR (n-butylether)
RR (propene)
k
k
l
0
3.85 ± 0.80
4.27 ± 1.80
1
8
2
.6 ± 0.7
.5 ± 0.9
.4 ± 0.2
6.89 ± 0.73
6.96 ± 0.58
Ferronato et al. [15]
l
d
0
.995 ± 0.022
5.7 ± 0.1
Papagni et al. [7]
Imamura et al. [18]
m
2
2
0
.32 ± 0.10
.46 ± 0.12
.56 ± 0.02
6.69 ± 0.73
6.46 ± 0.72
m
n
5.6 ± 0.6û
RR (isoprene)
Carrasco et al. [20]
a
RR: Relative rate method, PLP-LIF: pulsed laser photolysis–laser-induced ßuorescence detection. Chemicals in parentheses are the
references used in the relative rate method.
b
Average value of our relative rate measurements using two different reference compounds.
c
−11
3
−1 −1
Relative to k (propene) = 2.63 × 10 cm molecule
s in 700 Torr synthetic air at 298 K. Quoted uncertainties include two standard
deviations and uncertainty in the reference rate coefÞcient (estimated at 10%).
d
−11
3
−1 −1
s
Relative to k(1,3,5-trimethylbenzene) = 5.70 × 10 cm molecule
in 740 Torr puriÞed air at 5% relative humidity at 296 ± 2 K.
The uncertainties are two standard deviations and do not include uncertainties in the reference rate constant (estimated at approximately ± 10%).
e
In 10–20 Torr argon. The errors are one standard deviation.
In 100–300 Torr helium diluent. The quoted uncertainty is two standard deviations and does not include an estimate of possible systematic
f
uncertainties.
g
In 33–300 Torr helium at 298 K. The uncertainty includes two standard deviations and an estimate of systematic uncertainties (5%).
h
−10
3
−1 −1
−11
3
−1 −1
Relative to k(butyl vinyl ether) = 1.0 × 10 cm molecule
of synthetic air at 298 ± 3 K. The uncertainties are two standard deviations and do not include uncertainties in the reference rate constant.
Relative to k(1,3,5-trimethylbenzene) = 5.88 × 10 cm molecule and k(α-pinene) = 5.18 × 10 cm molecule in 1 atm
of puriÞed air at 300 K. The uncertainties are one standard deviation and do not include uncertainties in the reference rate constants.
s
and k(isopropenyl acetate) = 6.96 × 10 cm molecule
s in 1 atm
i
−11
3
−1 −1
−11
3
−1 −1
s
s
j
2 2
In 90–100 Torr total pressure with He/O diluents. The O partial pressure was 2–7 Torr. The quoted error bars include systematic and
precision errors at the 95% conÞdence level.
k
−10
3
−1 −1
−11
3
−1 −1
s
Relative to k(isoprene) = 1.01 × 10 cm molecule
air at 298 ± 2 K. The uncertainties are two standard deviations and do not include uncertainties in the reference rate constants.
Relative to k(ethene) = 8.1 × 10 cm molecule and k(propene) = 2.9 × 10 cm molecule in 700 Torr synthetic air at 295
s
and k(propene) = 2.63 × 10 cm molecule
in 740 ± 5 Torr puriÞed
l
−11
3
−1 −1
−11
3
−1 −1
s
s
±
1 K. The uncertainties are two standard deviations and do not include uncertainties in the reference rate constants.
m
Relative to k(n-butyl ether) = 2.89 × 10−11 cm molecule
3
−1 −1
−11
3
−1 −1
s
and k(propene) = 2.63 × 10 cm molecule
s in 760 Torr puriÞed
air at 298 K. The uncertainties include two standard deviations and uncertainties in the reference rate constants (assumed to be 10%).
n
−10
3
−1 −1
Relative to k(isoprene) = 1.01 × 10 cm molecule
s
in 1 atm of synthetic air at 298 ± 2 K. The uncertainty was calculated using
the method described by Brauers and Finlayson-Pitts [29] and includes uncertainties in the rate constant ratio and k(isoprene).
International Journal of Chemical Kinetics DOI 10.1002/kin

OH-INITIATED OXIDATION OF SMALL UNSATURATED ALCOHOLS
155
2
2
5
0
0
0
0
0
.6
.4
.2
.0
A: Before UV
0
0
0
.6
.4
.2
B: After UV
15
10
5
0
0
.0
.3
C: B-0.575*A
0
0
0
.2
.1
.0
D: CH C(O)CH
3
3
0
0
5
10
15
20
25
30
35
0
.4
Δ [CH =CHC(CH ) OH] (mTorr)
2
3 2
Figure 3 Formation of acetone and glycolaldehyde follow-
ing the UV irradiation of mixtures of 46.3–51.3 mTorr of
MBO232, 71.3–75.8 mTorr methyl nitrite, and 10.3–47.9
mTorr NO in 700 Torr air. The lines are linear least-squares
Þts.
0
0
.0
.3
E: HOCH2CHO
0
.0
7
40
750
1200
1300
1400
are primary products, and there are no signiÞcant losses
for the consumptions of MBO232 employed (<61%).
The lines through the experimental data are linear least-
squares Þts that give molar yields of 0.66 ± 0.06 and
Figure 2 Spectra acquired before (A) and after (B) a 2-
min irradiation of a mixture of 51.3 mTorr MBO, 75.8 mTorr
CH3ONO, and 47.9 mTorr NO in 700 Torr air. Panel C shows
the result of subtracting features attributable to MBO from
panel B. Reference spectra of acetone and glycolaldehyde
are shown in panels D and E, respectively.
0.56 ± 0.05 for acetone and glycolaldehyde formation,
respectively. The formation of acetone and glycolalde-
hyde in the OH-initiated oxidation of MBO232 has
been reported previously by several groups. As seen
from inspection of Table II, our results are consistent
with the results of previous product studies within the
combined uncertainties.
As discussed in previous studies (e.g., Ferronato
et al. [15] and Alvarado et al. [16]), the observation
of acetone and glycolaldehyde as products in similar
yields is consistent with the majority of the reaction of
OH radicals with MBO232 proceeding via addition of
the OH radical to the terminal carbon of the >C CH2
bond. As with all other alkenes [24], the dominant at-
mospheric fate of the resulting >C(O)CH2OH alkoxy
radical is decomposition via C C bond scission. The
reactions leading to the products reported from the
OH-radical initiated oxidation of 2-methyl-3-buten-2-
ol are shown in the mechanism given in Fig. 4.
5
1.3 mTorr MBO232, 75.8 mTorr CH3ONO, and 47.9
mTorr NO are shown in Fig. 2. Panel C in Fig. 2
shows the result of subtracting features attributable to
MBO232 from panel B. Reference spectra of acetone
and glycolaldehyde are shown in panels D and E, re-
spectively. Comparison of panel C with the reference
spectra of acetone and glycolaldehyde shows the for-
mation of these products.
Figure 3 shows a plot of acetone and glycolalde-
hyde formation versus the MBO232 loss following
irradiation of MBO232/methylnitrite/NO mixtures in
7
<
00 Torr air. Small corrections (<1% for acetone and
10% for glycoaldehyde) have been applied to the
data in Fig. 3 to account for secondary loss via re-
action with OH radicals using IUPAC-recommended
rate coefÞcients for k(OH + acetone) and k(OH +
HOCH2CHO) [27] and the procedure described else-
where [28]. As seen from Fig. 3, the concentrations of
acetone and glycolaldehyde increased linearly with the
loss of MBO232. This behavior suggests these species
Products and Mechanism of OH-Initiated
Oxidation of Allyl Alcohol
Glycolaldehyde was observed using its distinct absorp-
tion feature at ∼745 cm− as a product of the OH
1
International Journal of Chemical Kinetics DOI 10.1002/kin

1
56
TAKAHASHI, HURLEY, AND WALLINGTON
Table II Products from OH-Initiated Oxidation of MBO232 in the Presence of NOx
Product
Acetone
Yield
Experimental Technique
Reference
a
0.66 ± 0.06
FTIR
FTIR
FTIR
FTIR+GC
FTIR
FTIR
This work
b
0
0
0
.67 ± 0.05
Carrasco et al. [20]
Ferronato et al. [15]
Alvarado et al. [16]
This work
Carrasco et al. [20]
Ferronato et al. [15]
Alvarado et al. [16]
Reisen et al. [17]
Chan et al. [21]
Carrasco et al. [20]
Ferronato et al. [15]
Alvarado et al. [16]
Carrasco et al. [20]
Alvarado et al. [16]
Reisen et al. [17]
Chan et al. [21]
c
.52 ± 0.05
d
a
b
.58 ± 0.04
Glycolaldehyde
0.56 ± 0.05
0
0
0
0
.78 ± 0.20
c
.50 ± 0.05
FTIR
d
e
.61 ± 0.09
FTIR
g
.67 ± 0.11
SPME
f
b
CIMSh
FTIR
FTIR
FTIR
FTIR
GC
0
.66 ± 0.03
Formaldehyde
0.33 ± 0.08
c
0
0
.35 ± 0.04
d
b
.29 ± 0.03
2
-Hydroxy-2-methylpropanal (HMPr)
0.31 ± 0.11
d
e
0
0
0
.19 ± 0.07
.31 ± 0.04
SPME
CIMS
CIMS
FTIR
f
f
d
.37 ± 0.13
Dihydroxynitrate
0.10 ± 0.07
Chan et al. [21]
Alvarado et al. [16]
0
.05 ± 0.02
a
The errors are two-standard deviations from the regression analysis.
b
c
d
The errors include two standard deviations from the regression analysis and uncertainties in calibrations of the analytical techniques.
The errors were from a least-squares Þt analysis of the data. It was not speciÞed whether the errors are one or two standard deviations.
Errors are two least-squares standard deviations combined with the estimated overall uncertainties in the GC-FID or FT-IR calibration
factors for 2-methyl-3-butene-2-ol and products of ± 5% each (± 8% for the FT-IR analysis of glyocolaldehyde).
e
Errors are two standard deviations.
Average of three sets of the experiments. No exact description on the sources of errors.
Sampling by solid-phase microextraction (SPME) with on-Þber derivitization followed by gas chromatographic analyses.
Chemical ionization mass spectrometry.
f
g
h
OH
12
C(CH ) OH
HOH2C
O2
C(CH3)2OH
3
2
H
1
0
O
NO
O
H
O
H
8
6
4
2
0
HOH2C
C(CH3)2OH
HOH2C
C(CH3)2OH
NO2
O
O2
+
CH(CH ) OH
CH C(O)CH + HO
3
2
3
3
2
HOH2C
O
H
O2
0
2
4
6
8
10
12
14
+
CH OH
HCHO + HO2
2
Δ [CH =CHCH OH] (mTorr)
H
C(CH3)2OH
2
2
Figure 5 Formation of glycolaldehyde versus loss of allyl
alcohol following the UV irradiation of mixtures of 13.2–
Figure 4 Mechanism showing the formation of glycolalde-
hyde (HOCH2CHO), acetone (CH3C(O)CH3), 2-hydroxy-
1
1
6.3 mTorr allyl alcohol, 100–106 mTorr CH3ONO, and
3–25 mTorr NO in 700 Torr air.
2
-methylpropanal (HC(O)C(CH3)2OH, and formaldehyde
(HCHO) in the OH-initiated oxidation of allyl alcohol.
International Journal of Chemical Kinetics DOI 10.1002/kin

OH-INITIATED OXIDATION OF SMALL UNSATURATED ALCOHOLS
157
at low pressure and are not appropriate for atmospheric
pressure, and the results from Fantechi et al. [14] ap-
pear to be in error. The remaining absolute and relative
rate studies listed in Table I are in good agreement.
Taking averages of these data give values of k(OH +
OH
CH OH
HOH
2
C
CH OH
2
2
H
O2
−11
O
allyl alcohol) = 5.32 × 10 , k(OH + 3-buten-1-ol) =
11
NO
−
O
5.90 × 10 , and k(OH + 2-methyl-3-buten-2-ol) =
11 3
O
−
−1 −1
cm molecule s . Combining these
6
.39 × 10
6
−3
HOH
2
C
CH
2
OH
HOH C
CH OH
values with a daytime average [OH] = 2.5 × 10 cm
2
2
H
H
gives lifetimes of approximately 2 h for these small
unsaturated alcohols with respect to reaction with OH
radicals. Clearly, these are very reactive compounds,
which will be oxidized in the atmosphere close to their
sources.
NO2
O
O2
+
CH OH
HCHO + HO2
2
HOH
2
C
H
Figure 6 Mechanism showing the formation of glycolalde-
hyde (HOCH2CHO) and formaldehyde (HCHO) in the OH-
initiated oxidation of allyl alcohol.
This study was partly supported by Program for Improve-
ment of Research Environment for Young Researchers from
Special Coordination Funds for Promoting Science and Tech-
nology (SCF) commissioned by the Ministry of Education,
Culture, Sport, Science and Technology (MEXT) of Japan
radical initiated oxidation of allyl alcohol, as shown in
Fig. 2. Figure 5 shows a plot of the observed formation
of glycolaldehyde (corrected for secondary loss via re-
action with OH) versus the loss of allyl alcohol follow-
ing the UV irradiation of mixtures of 13.2–16.3 mTorr
allyl alcohol, 100–106 mTorr CH3ONO, and 13–25
mTorr NO in 700 Torr air. Linear least-squares analy-
sis of the data in Fig. 5 gives a glycolaldehyde yield of
(
K T.). This work was also supported by Grants-in-Aid from
the MEXT for ScientiÞc Research on Innovative Areas “Im-
pacts of Aerosols in East Asia on Plants and Human Health”
(K. T.)
BIBLIOGRAPHY
0
.85 ± 0.08 (the uncertainty is two standard deviations
from the regression analysis). This result is consis-
tent with the previous product study by Orlando et al.
1
. Fehsenfeld, F.; Calvert, J.; Fall, R.; Goldan, P.; Guenther,
A. B.; Hewitt, C. N.; Lamb, B.; Liu, S.; Trainer, M.;
Westberg, H.; Zimmerman, P. Global Biogeochem Cy-
cles 1992, 6, 389.
[8] in which glycolaldehyde was observed in a molar
yield of 0.90 ± 0.12. As with MBO232, the reaction
of OH radicals with allyl alcohol proceeds via addition
to the terminal carbon of the >C CH2 bond leading
to the formation of an alkoxy radical whose predomi-
nant atmospheric fate is decomposition via C–C bond
scission. The reactions leading to the products reported
from the OH-radical initiated oxidation of allyl alcohol
are shown in the mechanism given in Fig. 6. As seen in
Fig. 6, HCHO is an expected coproduct of glyco-
laldehyde. The formation of large amounts of HCHO
associated with the OH radical source employed
2. Chameides, W.; Lindsay, R.; Richardson, J.; Kiang, C.
Science 1988, 241, 1473.
3
4
5
6
. Steiner, A. L.; Tonse, S.; Cohen, R. C.; Goldstein, A. H.;
Harley, R. A. Geophys Res Lett 2007, 34, L15806.
. Goldan, P. D.; Kuster, W. C.; Fehsenfeld, F. C.; Montzka,
S. A. Geophys Res Lett 1993, 20, 1039.
. Schade, G. W.; Goldstein, A. H.; Gray, D. W.; Lerdau,
M. T. Atmos Environ 2000, 34, 3535.
. Tarvainen, V.; Hakola, H.; Hell e´ n, H.; B a¨ ck, J.; Hari, P.;
Kulmala, M. Atmos Chem Phys 2005, 5, 989.
7. Papagni, C.; Arey, J.; Atkinson, R. Int J Chem Kinet
2001, 33, 142.
(CH3ONO photolysis) prevents the measurement of
HCHO formed in the oxidation of allyl alcohol.
8. Orlando, J. J.; Tyndall. G. S.; Ceazan, N. J Phys Chem
A 2001, 105, 3564.
9. Upadhyaya, H. P.; Kumar, A.; Naik, P. D.; Sapre, A. V.;
Mittal, J. P. Chem Phys Lett 2001, 349, 279.
CONCLUSIONS
1
1
0. Holloway, A. L.; Treacy, J.; Sidebottom, H.; Mellouki,
A.; Da e¨ le, V.; Le Bras, G.; Barnes, I. J Photochem Pho-
tobiol A 2005, 176, 183.
1. Le Person, A.; Solignac, G.; Oussar, F.; Da e¨ le, V.;
Mellouki, A.; Winterhalter, R.; Moortgat, G. K. Phys
Chem Chem Phys 2009, 11, 7619.
The results presented here serve to increase the kinetic
database concerning the reactions of hydroxyl radicals
with small unsaturated alcohols. As discussed above,
the experiments of Upadhyayaet al. [9] wereconducted
International Journal of Chemical Kinetics DOI 10.1002/kin

1
58
TAKAHASHI, HURLEY, AND WALLINGTON
1
1
1
1
1
1
1
2. Parker, J. K.; Espada-Jallad, C. J Phys Chem A 2009,
13, 9814.
3. Rudich, Y.; Talukdar, R.; Burkholder, J. B.;
Ravishankara, A. R. J Phys Chem 1995, 99, 12188.
4. Fantechi, G.; Jensen, N. R.; Hjorth, J.; Peeters, J. Int J
Chem Kinet 1998, 30, 589.
5. Ferronato, C.; Orlando, J. J.; Tyndall, G. S. J Geophys
Res 1998, D103, 25579.
6. Alvarado, A.; Tuazon, E. C.; Aschmann, S. M.; Arey, J.;
Atkinson, R. Atmos Environ 1999, 33, 2893.
22. Wallington, T. J.; Japar, S. M. J Atmos Chem 1989, 9,
399.
1
23. Atkinson, R.
Monograph 1.
J
Phys Chem Ref Data 1989,
24. Calvert, J. G.; Atkinson, R.; Kerr, J. A.; Madronich,
S.; Moortgat, G. K.; Wallington, T. J.; Yarwood, G.
The Mechanisms of Atmospheric Oxidation of Alkenes;
Oxford University Press: New York, 2000.
25. Calvert, J. G.; Atkinson, R.; Becker, K. H.; Kamens,
R. M.; Seinfeld, J. H.; Wallington, T. J.; G. Yarwood, G.
The Mechanisms of Atmospheric Oxidation of Aromatic
Hydrocarbons; Oxford University Press: New York,
2002.
26. Pfrang, C.; King, M. D.; Braeckevelt, M., Canosa-Mas,
C. E.; Wayne, R. P. Atmos Environ 2008, 42, 3018.
27. Atkinson, R.; Baulch, D. L.; Cox, R. A.; Crowley,
J. N.; Hampson, R. F.; Hynes, R. G.; Jenkin, M. E.; Rossi,
M. J.; Troe, J. Atmos Chem Phys 2006, 6, 3625.
28. Meagher, R. J.; McIntosh, M. E.; Hurley, M. D.;
Wallington, T. J. Int J Chem Kinet 1997, 29, 619.
29. Brauers, T.; Finlayson-Pitts, B. J. Int J Chem Kinet 1997,
29, 665.
7. Reisen, F.; Aschmann, S. M.; Atkinson, R.; Arey, J.
Environ Sci Technol 2003, 37, 4664.
8. Imamura, T.; Iida, Y.; Obi, K.; Nagatani, I.; Nakagawa,
K.; Patroescu-Klotz, I.; Hatakeyama, S. Int J Chem Kinet
2
004, 36, 379.
9. Baasandorj, M.; Stevens, P. S. J Phys Chem A 2007,
11, 640.
1
2
2
1
0. Carrasco, N.; Doussin, J. F.; O’Connor, M.; Wenger,
J. C.; Picquet-Varrault, B. J Atmos Chem 2007, 56, 33.
1. Chan, A. W. H.; Galloway, M. M.; Kwan, A. J.; Chhabra,
P. S.; Keutsch, F. N.; Wennberg, P. O.; Flagan, R. C.;
Seinfeld, J. H. Environ Sci Technol 2009, 43, 4647.
International Journal of Chemical Kinetics DOI 10.1002/kin