Atmospheric chemistry of i-butanol

Article abstract of DOI:10.1021/jp107950d

Smog chamber/FTIR techniques were used to determine rate constants of k(Cl + i-butanol) = (2.06 ± 0.40) × 10^-10, k(Cl + i-butyraldehyde) = (1.37 ± 0.08) × 10^-10, and k(OH + i-butanol) = (1.14 ± 0.17) × 10^-11 cm³ molecule^-1 s^-1 in 700 Torr of N₂/O₂ diluent at 296 ± 2K. The UV irradiation of i-butanol/Cl ₂/N₂ mixtures gave i-butyraldehyde in a molar yield of 53 ± 3%. The chlorine atom initiated oxidation of i-butanol in the absence of NO gave i-butyraldehyde in a molar yield of 48 ± 3%. The chlorine atom initiated oxidation of i-butanol in the presence of NO gave (molar yields): i-butyraldehyde (46 ± 3%), acetone (35 ± 3%), and formaldehyde (49 ± 3%). The OH radical initiated oxidation of i-butanol in the presence of NO gave acetone in a yield of 61 ± 4%. The reaction of chlorine atoms with i-butanol proceeds 51 ± 5% via attack on the α-position to give an α-hydroxy alkyl radical that reacts with O₂ to give i-butyraldehyde. The atmospheric fate of (CH₃)₂C(O)CH ₂OH alkoxy radicals is decomposition to acetone and CH₂OH radicals. The atmospheric fate of OCH₂(CH₃)CHCH ₂OH alkoxy radicals is decomposition to formaldehyde and CH ₃CHCH₂OH radicals. The results are consistent with, and serve to validate, the mechanism that has been assumed in the estimation of the photochemical ozone creation potential of i-butanol.

Full text of DOI:10.1021/jp107950d

1
2462
J. Phys. Chem. A 2010, 114, 12462–12469
Atmospheric Chemistry of i-Butanol
†
,‡
†
V. F. Andersen, T. J. Wallington,* and O. J. Nielsen
Copenhagen Center for Atmospheric Research, Department of Chemistry, UniVersity of Copenhagen,
UniVersitetsparken 5, DK-2100 Copenhagen, Denmark, and Systems Analytics and EnVironmental Sciences
Department, Ford Motor Company, Mail Drop RIC-2122, Dearborn, Michigan 48121-2053, United States
ReceiVed: August 22, 2010; ReVised Manuscript ReceiVed: October 21, 2010
Smog chamber/FTIR techniques were used to determine rate constants of k(Cl + i-butanol) ) (2.06 ( 0.40)
-
10
-10
-11
×
10 , k(Cl + i-butyraldehyde) ) (1.37 ( 0.08) × 10 , and k(OH + i-butanol) ) (1.14 ( 0.17) × 10
3 -1 -1
2 2 2 2
cm molecule
s
in 700 Torr of N
/O
diluent at 296 ( 2K. The UV irradiation of i-butanol/Cl
/N
mixtures
gave i-butyraldehyde in a molar yield of 53 ( 3%. The chlorine atom initiated oxidation of i-butanol in the
absence of NO gave i-butyraldehyde in a molar yield of 48 ( 3%. The chlorine atom initiated oxidation of
i-butanol in the presence of NO gave (molar yields): i-butyraldehyde (46 ( 3%), acetone (35 ( 3%), and
formaldehyde (49 ( 3%). The OH radical initiated oxidation of i-butanol in the presence of NO gave acetone
in a yield of 61 ( 4%. The reaction of chlorine atoms with i-butanol proceeds 51 ( 5% via attack on the
R-position to give an R-hydroxy alkyl radical that reacts with O
2
to give i-butyraldehyde. The atmospheric
OH alkoxy radicals is decomposition to acetone and CH OH radicals. The atmospheric
OH alkoxy radicals is decomposition to formaldehyde and CH CHCH OH radicals.
fate of (CH
3
)
2
C(O)CH
2
2
fate of OCH
2
(CH
3
)CHCH
2
3
2
The results are consistent with, and serve to validate, the mechanism that has been assumed in the estimation
of the photochemical ozone creation potential of i-butanol.
1
. Introduction
Recognition of the importance of energy security and climate
chemistry and environmental impact is needed. There is a
substantial kinetic and mechanistic database for smaller alcohols
such as methanol, ethanol, and propanol and the atmospheric
change has led to growing interest in the use of biofuels in
transportation. Biofuels are typically used in blends with diesel
or gasoline. The Renewable Fuel Standard established by the
Energy Independence and Security Act in 2007 in the U.S.
mandates the use of 36 billion gallons (136 billion liters) of
renewable fuel by 2022. This corresponds to replacement of
approximately 17% of the projected gasoline use for light-duty
8
chemistry of these compounds is well understood. The kinetic
and mechanistic database for larger alcohols is sparse and the
details of their atmospheric oxidation are unclear. There have
been two studies of the kinetics of the reaction of OH radicals
with i-butanol9 and one study of the kinetics of the reaction
,10
9
of chlorine atoms with i-butanol. The mechanism of the
1
atmospheric degradation of i-butanol has yet to be studied. To
improve our understanding of the atmospheric chemistry of
alcohols, we conducted a study of the kinetics and the oxidation
mechanism of i-butanol initiated by chlorine atoms and OH
radicals.
vehicles in 2022. In Europe, the Renewable Energy Directive
calls for use of 10% renewable energy in the transportation
sector by 2020.2
To reduce competition with food crops and to increase the
yields per hectare, research is focusing on the development of
second-generation biofuels. Second-generation alcohol biofuels
3
can be produced from biomass via either biotic routes (e.g.,
pretreatment of cellulose and hemicellulose to release sugars
that can be fermented to give ethanol, butanol, or higher
alcohols) or abiotic routes (e.g., gasification followed by
2. Experimental Section
4
Experiments were performed in a 140 L Pyrex reactor
1
1
thermochemical synthesis giving a mixture [typically C
1
-C
4
]
interfaced to a Mattson Sirius 100 FTIR spectrometer. The
reactor was surrounded by 22 fluorescent blacklamps (GE
F40T12BLB), which were used to photochemically initiate the
experiments. Chlorine atoms were produced by photolysis of
molecular chlorine.
of alcohols). There is interest in the use of i-butanol as a second
generation biofuel. The chemical and physical properties of
i-butanol are similar to those of gasoline. Unlike smaller alcohols
such as ethanol, modest amounts (5-20%) of i-butanol can be
blended into gasoline without substantially changing the energy
density, water-induced phase separation performance, or volatil-
ity of the fuel blend.5
-7
Cl + hν f Cl + Cl
(1)
2
The use of alcohol biofuels will result in their release into
the atmosphere. In the atmosphere, alcohols undergo photo-
chemical oxidation initiated by the OH radical. Prior to the large
scale use of alcohols, an assessment of their atmospheric
OH radicals were produced by photolysis of CH
3
ONO in air.
(2)
*
To whom correspondence should be addressed. E-mail: twalling@
ford.com.
†
University of Copenhagen.
Ford Motor Company.
‡
CH ONO + hν f CH O + NO
3
3
1
0.1021/jp107950d  2010 American Chemical Society
Published on Web 11/04/2010

Atmospheric Chemistry of i-Butanol
CH O + O f HO + HCHO
J. Phys. Chem. A, Vol. 114, No. 47, 2010 12463
(3)
(4)
3
2
2
HO + NO f OH + NO
2
2
Relative rate techniques were used to measure the rate con-
stant of interest relative to a reference reaction whose rate
constant has been established previously. The relative rate
method is a well established technique for measuring the
reactivity of Cl atoms and OH radicals with organic compounds.
Kinetic data were derived by monitoring the loss of i-butanol
relative to one or more reference compounds. The method
assumes that the reaction under investigation is the only
significant loss process for the reactant and reference. The
following relation is then valid:
[
i-butanol]t0
[reference]t0
[reference]t
kreactant
ln
)
ln
(I)
(
)
(
)
[
i-butanol]t
kreference
where [i-butanol]to, [i-butanol]
t
, [reference]t0 and [reference]
t
are
the concentrations of i-butanol and the reference compound at
times t
reactions of Cl atoms, or OH radicals, with i-butanol and the
reference compound. Plots of ln([i-butanol]t0/[i-butanol] ) versus
ln([reference]t0/[reference] ) should be linear, pass through the
0
and t, and ki-butanol and kreference are the rate constants for
t
t
origin, and have a slope of ki-butanol/kreference. Unless stated
otherwise, quoted uncertainties represent the precision of the
measurements and include two standard deviations from the
regression analyses and uncertainties in the IR analysis (typically
(
1-2% of the initial reactant concentrations).
CH ONO was synthesized by the dropwise addition of
concentrated sulphuric acid to a saturated solution of NaNO
3
Figure 1. Loss of i-butanol and i-butyraldehyde vs ethene (triangles)
2
and c-hexane (circles) in the presence of chlorine atoms in 700 Torr of
in methanol. i-Butanol and i-butyraldehyde were obtained
from Sigma-Aldrich at purities of >99.5% and >99%,
respectively. Experiments were conducted in 700 Torr total
air or N at 296 ( 2 K.
2
pressure of high purity O
2
/N
2
diluent at 296 ( 2 K.
Concentrations of reactants and products were monitored by
FTIR spectroscopy. IR spectra were derived from 32 coadded
interferograms with a spectral resolution of 0.25 cm and
an analytical path length of 26 m. To check for unwanted
loss of reactants and reference compounds via heterogeneous
reactions, reaction mixtures were left to stand in the chamber
or C
2
H
4
2 2
, and 150-250 mTorr Cl in 700 Torr of air or N diluent.
-
1
The loss of i-butanol and i-butyraldehyde are plotted versus the
reference compounds in Figure 1. Linear least-squares analysis of
the data in Figure 1 gives k
0.16, k /k ) 0.42 ( 0.02, and k
rate constants (at 700 Torr total pressure) of k
5 7
/k
5 8
) 0.61 ( 0.07, k /k ) 2.24 (
6
7
6 8
/k ) 1.47 ( 0.06. Using reference
-10 12
for 60 min. With the exception of (CH
3
)
2
CHCH(Cl)OH, there
7
) 3.30 × 10
-
11 13
-10
was no observable (<2%) loss of any of the reactants or
products in the present work.
and k ) 9.29 × 10
gives k ) (2.02 ( 0.23) × 10 , k )
8
5
5
-10
-10
(
2.08 ( 0.15) × 10 , k
6
3
) (1.38 ( 0.07) × 10 , and k
6
)
-
10
-1 -1
(
1.36 ( 0.06) × 10 cm molecule s . We cite final values
that are averages of the individual determinations together with
error limits that encompass the extremes of the determinations, k
3
. Results and Discussion
3
.1. Measurement of k(Cl + i-Butanol) and k(Cl +
5
i-Butyraldehyde). The rates of reactions 5 and 6 were measured
-10
-10
3
)
(2.05 ( 0.26) × 10 and k
molecule s . Results from the current work are compared with
previous measurements in Table 1. Our result for k is in excellent
agreement with that reported by Wu et al. Our result for k is in
6
) (1.37 ( 0.08) × 10 cm
relative to reactions 7 and 8.
-1 -1
5
Cl + (CH ) CHCH OH f products
(5)
(6)
(7)
(8)
9
3
2
2
6
14
good agreement with those reported by Ullerstam et al. and Le
Cr aˆ ne et al.,15 but is approximately 25% lower than the results
reported by Th e´ venet et al.16
Cl + (CH ) CHCHO f products
3
2
3
.2. Kinetics of the OH + i-Butanol Reaction in 700 Torr
of Air. The rate of reaction 9 was measured relative to reactions
0 and 11:
Cl + c-C H f products
6
12
1
Cl + C H f products
2
4
Reaction mixtures consisted of either 30-60 mTorr i-butanol
or 11-15 mTorr i-butyraldehyde, 5-30 mTorr of either c-C
OH + (CH ) CHCH OH f products
(9)
3
2
2
H
6 12

1
2464 J. Phys. Chem. A, Vol. 114, No. 47, 2010
Andersen et al.
TABLE 1: Literature Data for k(Cl + i-Butanol) and
k(Cl + i-Butyraldehyde)a
-CH, -CH
of k
, and -OH groups, respectively. The SAR estimate
2
9
9
is consistent with the experimental determinations of k .
3
.3. Mechanism of the Cl + i-Butanol Reaction Studied
in 700 Torr of N . The mechanism of the reaction of Cl atoms
with i-butanol was investigated by irradiating mixtures of 29
mTorr of i-butanol and 110 mTorr of Cl in 700 Torr of N
diluent. The reaction of Cl atoms with i-butanol can proceed
via four channels
T
k(Cl + i-butanol) reference k(Cl + reference) citation
2
2
96
96
95
95
2.02 ( 0.23
2.10 ( 0.33
1.94 ( 0.04
2.11 ( 0.12
c-hexane
ethene
propane
c-hexane
3.30
0.929
1.40
3.30
this work
this work
2
2
2
9
2
2
Wu et al.
Wu et al.
9
k(Cl +
i-butyraldehyde)
k(Cl +
reference)
T
reference
citation
this work
Cl + (CH ) CHCH OH f (CH ) CHCHOH + HCl
3
2
2
3 2
2
96
96
98
98
98
97
98
98
1.38 ( 0.07
1.36 ( 0.06
1.78 ( 0.25
1.73 ( 0.30
1.70 ( 0.35
1.50 ( 0.30
1.38 ( 0.14
1.24 ( 0.13
c-hexane
ethene
ethane
propane
n-butane
propene
ethene
3.30
0.929
0.59
1.40
2.05
2.30
0.929
3.30
(
5a)
2
2
2
2
2
2
2
this work
1
1
1
6
Th e´ venet et al.
Th e´ venet et al.
Th e´ venet et al.
Ullerstam et al.
Le Cr aˆ ne et al.
6
6
Cl + (CH ) CHCH OH f (CH ) CCH OH + HCl
3
2
2
3 2
2
(
5b)
1
4
1
5
5
1
cyclohexane
Le Cr aˆ ne et al.
Cl + (CH
3
)
2
CHCH
2
OH f (CH
3
)(CH
2
)CHCH OH + HCl
2
(5c)
a
All studies were performed using the relative rate method. Rate
constants are in units of 10-10 cm molecule
3
-1 -1
s .
Cl + (CH ) CHCH OH f (CH ) CHCH O + HCl
3
2
2
3 2
2
(
5d)
Hydrogen abstraction from the -OH group (reaction 5d) in
-
1 8
alcohols is endothermic by approximately 5 kJ mol
and is
not considered further. The radicals generated in reactions 5a-c
react with molecular chlorine via reactions 12-14
Cl + (CH ) CHCHOH f (CH ) CHCH(Cl)OH + Cl
2
3 2
3 2
(12)
Cl + (CH ) CCH OH f (CH ) CClCH OH + Cl
2
3 2
2
3 2
2
(
13)
Cl + (CH )(CH )CHCH OH f
2
3
2
2
(
CH )(CH Cl)CHCH OH + Cl (14)
Figure 2. Loss of i-butanol vs ethene (triangles) and propene (circles)
in the presence of OH radicals in 700 Torr of air at 296 ( 2 K.
3 2 2
By analogy to the behavior of other alkyl radicals,20 it is
OH + C H f products
(10)
(11)
3
6
expected that the rate constants for reactions 12-14 will be
-
12
-11
3
-1 -1
on the order of 10
to 10
cm molecule s . In the
in 700 Torr of
present experiments with 110 mTorr of Cl
2
OH + C H f products
2
4
nitrogen, reactions 12-14 will be the dominant fate of the
radicals produced by reaction 5. It has been observed in
previous experiments using the chamber at Ford that R-chloro-
Reaction mixtures consisted of 29-31 mTorr of i-butanol,
ONO, and 7 mTorr of either C or C
8
7-88 mTorr of CH
3
H
3 6
H
2 4
21
21
21
24
alcohols such as CH
2
ClOH,
CHCl
2
OH, CCl
3
3
OH,
in 700 Torr total pressure of air diluent. Figure 2 shows the
22
,23 and CH
CH
3
CHClOH, CH
3
CClOHCH
3
(CH ) CHClOH
2 2
loss of i-butanol versus loss of the reference compounds. Linear
decompose heterogeneously in the chamber via elimination
of HCl to give the corresponding carbonyl compounds on a
time scale typically of a few minutes. Similar behavior was
observed in the present study with the formation of i-
butyraldehyde (complete within 5-10 min) on allowing
reaction mixtures to stand in the dark. The i-butyraldehyde
least-squares analysis gives k
9
/k10 ) 0.41 ( 0.04 and k
9
/k11
)
1
.41 ( 0.10. Using reference rate constants (at 700 Torr total
-11 17
-12 17
-11
pressure) of k10 ) 2.63 × 10
and k11 ) 8.52 × 10
11
-
gives k
9
) (1.08 ( 0.11) × 10 and (1.20 ( 0.09) × 10
3
-1 -1
cm molecule s . We cite a final value that is the average of
the individual determinations together with error limits that
is generated via decomposition of (CH
provides a marker for reaction 5a.
3
)
2
CHCHClOH and
encompass the extremes of the determinations, k ) (1.14 (
9
-11
3
-1 -1
0
.17) × 10 cm molecule s . As seen from Table 2, the
9
results from the three independent studies of k are in good
(
CH ) CHCHClOH f (CH ) CHCHO + HCl
3 2 3 2
agreement.
Structure Activity Relationship (SAR) methods18,19 can be
(15)
used to estimate k(OH + i-butanol). Using the substituent factors
1
9
proposed by Bethel et al. and group rate constants given by
Figure 3 shows the formation of i-butyraldehyde versus
loss of i-butanol following the UV irradiation of mixtures
Kwok et al.18 and Bethel et al., the SAR method provides an
19
-
12
3
-1 -1
estimate of k
3
9
) 8.9 × 10
cm molecule
s
with 4, 57,
of i-butanol and chlorine in 700 Torr of N
2
. The decrease in
7, and 2% of reaction occurring via abstraction from -CH ,
3
the yield of i-butyraldehyde for experiments employing high

Atmospheric Chemistry of i-Butanol
J. Phys. Chem. A, Vol. 114, No. 47, 2010 12465
TABLE 2: Literature Data for k(OH + i-Butanol) near Ambient Temperaturea
T
k(OH + i-butanol)
techniqueb
reference
k(OH + reference)
citation
this work
this work
2
96
96
95
95
98
98
98
96
1.08 ( 0.11
1.20 ( 0.09
0.91 ( 0.04
0.99 ( 0.05
0.85 ( 0.01
0.98 ( 0.01
0.92 ( 0.04
1.05c
RR
RR
RR
RR
RR
RR
PLP-LIF
PLP-LIF
propene
ethene
propane
c-hexane
1-butanol
1,3-dioxalane
2.63
0.85
0.11
0.72
0.85
1.11
2
2
2
2
2
2
2
9
Wu et al.
Wu et al.
9
1
1
1
0
0
0
Mellouki et al.
Mellouki et al.
Mellouki et al.
Mellouki et al.
10
a
Rate constants are in units of 10-11 cm molecule s-1. b RR, relative rate; PLP-LIF, pulsed laser photolysis-laser induced fluorescence.
3
-1
c
Average of two determinations.
Figure 3. Formation of i-butyraldehyde vs loss of i-butanol following
UV irradiation of i-butanol/Cl mixtures in 700 Torr of nitrogen diluent.
2
(
>60%) consumption of i-butanol reflects the loss of i-
butyraldehyde via reaction with chlorine atoms. The data in
Figure 3 contain information on the initial yield of i-
butyraldehyde (which we equate to k5a/k
constant ratio k /k . Assuming that the formation and loss of
i-butyraldehyde are determined by reactions 5, 12, 15, and
5
) and the rate
6
5
Figure 4. IR spectra obtained before (A) and after (B) 30 s of
2
5
irradiation of 31 mTorr of i-butanol and 129 mTorr of Cl in 700 Torr
6
, then it can be shown that
2
of air. Panel (C) is the product spectrum obtained by subtracting 65%
of panel (A) from panel (B). Panels (D) and (E) are reference spectra
for i-butyraldehyde and acetone.
[
(CH ) CHCHO]
3 2 t
R
6 5
k /k -1
)
(1 - x)[(1 - x)
- 1]
[
(CH ) CHCH OH]
t)0
k6
3
2
2
1
- k5
3.4. Products of Cl Atom Initiated Oxidation of i-Butanol
in 700 Torr of N
oxidation of i-butanol was investigated by irradiating mixtures
of 30-31 mTorr of i-butanol, 117-1296 mTorr of Cl , and
5-150 Torr of oxygen in 700 Torr total pressure of N
diluent. Figure 4 shows spectra acquired before (A) and after
B), a 30 s irradiation of a mixture of 31 mTorr of i-butanol
and 129 mTorr of Cl in 700 Torr of air. The consumption
2 2
/O . The mechanism of Cl atom initiated
(II)
2
1
2
Where R is the yield of (CH
of chlorine atoms with (CH CHCH
constant for reaction 6, and x is the fractional loss of
3
)
2
CHCHO following reaction
3
)
2
2
OH (k5a/k ), k is the rate
5
6
(
2
(
3 2 2
CH ) CHCH OH defined as:
of i-butanol in this experiment was 35%. Panel (C) shows
the product spectrum derived by subtracting the IR features
of i-butanol from the spectrum in panel (B). Comparison with
the reference spectrum in panel (D) shows the formation of
i-butyraldehyde.
[
(CH ) CHCH OH]
3 2 2
t
x ) 1 -
(III)
[
(CH ) CHCH OH]
3 2 2
t0
As with other R-hydroxyalkyl radicals, reaction with O
2
giving i-butyraldehyde will be the sole fate of the radical
The best fit was obtained with R ) 0.53 ( 0.03 and k
.64 ( 0.06. Quoted errors correspond to two standard devia-
/k ) 0.64 ( 0.06
is consistent with the results reported in section 3.1; k /k
/k ) k /k ) 0.69 ( 0.09 and k /k /k ) k /k ) 0.66 (
× k
.05. From the value of R we conclude that 53 ( 3% of the
6 5
/k )
generated in reaction 5a.
0
tions from the fit. The rate constant ratio k
6
5
(
CH ) CHCHOH + O f (CH ) CHCHO + HO
×
3 2 2 3 2 2
6
7
(16)
k
7
5
6
5
6
8
8
5
6
5
0
reaction of chlorine atoms with i-butanol occurs via H-atom
abstraction from the R-carbon (reaction 5a).
The circles in Figure 5 show the formation of i-butyraldehyde
versus loss of i-butanol observed following the UV irradiation

1
2466 J. Phys. Chem. A, Vol. 114, No. 47, 2010
Andersen et al.
2
0-60% there was a substantial increase in the observed yield
of acetone and HCHO, and we conclude that these com-
pounds are formed as secondary products. For i-butanol
consumptions of >60% the yield of HCHO reaches a
maximum and then declines; in contrast, the yield of acetone
continues to increase with i-butanol consumption. Acetone
is relatively unreactive towards chlorine atoms (k(Cl +
-
12 8
acetone) ) 2.2 × 10
), whereas HCHO is reactive k(Cl
-
11 8
+
HCHO) ) 7.2 × 10 . The acetone yield for i-butanol
consumptions >90% tends towards a value that is similar to
the initial i-butyraldehyde yield (see previous sections). The
simplest explanation for the acetone formation shown in
Figure 6 is that it is formed via chlorine-initiated oxidation
of i-butyraldehyde:
(
CH ) CHCHO + Cl f (CH ) CHCO + HCl
3 2 3 2
Figure 5. Formation of i-butyraldehyde vs the loss of i-butanol
following UV irradiation of i-butanol/Cl mixtures with 15 Torr of O
open symbols) or 150 Torr of O (closed symbols) in 700 Torr total
pressure of nitrogen diluent in the presence (triangles) and absence
circles) of NO
(6a)
2
2
(
2
(
CH ) CHCO + O f (CH ) CHC(O)O
(17)
3 2 2 3 2 2
(
x
.
(
CH ) CHC(O)O + RO f (CH ) CHC(O)O + RO + O
3 2 2 2 3 2
2
(18)
(
CH ) CHC(O)O f (CH ) CH + CO
(19)
(20)
3
2
3 2
2
(
CH ) CH + O f (CH ) CHO
3 2 2 3 2 2
(
CH ) CHO + RO f (CH ) CHO + RO + O
3 2 2 2 3 2 2
(21)
(
CH ) CHO + O f CH C(O)CH + HO
(22)
3 2 2 3 3 2
The dashed curve in Figure 6 shows the expected yield of
acetone based on the values of k /k ) 0.70 ( 0.05 and R )
) 0.85. As seen from Figure
, the observed formation of acetone is consistent with its
formation as a secondary product following reaction of chlorine
with i-butyraldehyde. In contrast to the situation for acetone,
there are several potential secondary reactions that could give
HCHO in the system and it is difficult to point to a precise
source.
Figure 6. Formation of acetone (circles) and formaldehyde (triangles)
vs the loss of i-butanol following UV irradiation of i-butanol/Cl
mixtures with 15 Torr of O (open symbols) or 150 Torr of O (closed
symbols) in 700 Torr total pressure of nitrogen diluent in the absence
of NO . The dashed line shows the expected yield of acetone based on
the mechanism discussed in Section 3.4.
6
5
2
15
0
6
.48 ( 0.03 (see above) and k6a/k
6
2
2
x
of i-butanol/Cl
experiments with 150 Torr O
with 15 Torr O . There was no discernible impact of [O
the i-butyraldehyde yield. A fit of eq II to the data gives k
0.70 ( 0.05 and R ) 0.48 ( 0.03. These results are consistent
with those presented in Section 3.3 obtained in 700 Torr N
The rate constant ratio k /k ) 0.70 ( 0.05 is consistent with
the results presented in Section 3.1.
2
/O
2
/N
2
mixtures. The filled circles are for
, open circles are for experiments
] on
/k
2
We attribute the absence of acetone and formaldehyde as
observable primary products in the experiments performed in
2
2
6
5
)
the absence of NO
x
to significant loss of (CH
3 2 2
) C(OO)CH OH
2
.
peroxy radicals via reactions with HO
2
and other peroxy radicals
in reactions that do not lead to the formation of the alkoxy
radical. For example,
6
5
As seen from comparing panels C and E in Figure 4, a
small amount (0.84 mTorr) of acetone was observed. The
sharp feature at 1746 cm- evident to the left of the i-butyr-
aldehyde Q-branch in Figure 4C is attributable to the for-
mation of HCHO (1.05 mTorr). Figure 6 shows the formation
of acetone and formaldehyde versus the loss of i-butanol.
(
CH ) C(OO)CH OH + HO f
3
2
2
2
1
(
CH ) C(OOH)CH OH + O (23)
3
2
2
2
The relative importance of reaction 23 probably reflects a
very slow rate of self-reaction of (CH C(OO)CH OH radicals
which would be consistent with the slow rate of self-reaction
of (CH
COO radicals.26
3.5. Products of Cl Atom Initiated Oxidation of i-Butanol
in the Presence of NO in 700 Torr of N /O . The mechanism
of chlorine atom initiated oxidation of i-butanol in the presence
of NO was investigated by irradiating mixtures of 29-30 mTorr
i-butanol, 118-225 mTorr Cl , 59-76 mTorr NO, and 15-150
Filled symbols were obtained using 150 Torr O
symbols were obtained using 15 Torr O . There was no
discernible effect of [O ] on the yield of acetone, but the
yield of HCHO was lower in the experiment using higher
]. For small consumptions (<20%) of i-butanol there was
2
, and open
)
3 2
2
2
2
3 3
)
[
O
2
2
2
little, or no, formation of either acetone or HCHO, and we
conclude that these compounds are not major primary
products (yields <5%). For i-butanol consumptions of
2

Atmospheric Chemistry of i-Butanol
J. Phys. Chem. A, Vol. 114, No. 47, 2010 12467
Figure 7. Formation of acetone(circles) and formaldehyde (triangles) vs the loss of i-butanol following UV irradiation of i-butanol/Cl
with 15 Torr of O (open symbols) or 150 Torr of O (closed symbols) in 700 Torr total pressure of nitrogen diluent in the presence of NO
solid lines are least-squares fits. The dashed line is a fit of an expression analogous to expression II, see Section 3.5.
2
mixtures
2
2
x
. The
Torr oxygen in 700 Torr of nitrogen. The observed products
were i-butyraldehyde, formaldehyde, and acetone. The presence
because the source of the HCHO is unclear we are not able to
derive kinetic data from the plot.
of NO ensures rapid removal of RO
and, to a small degree, organic nitrates, RONO
radicals (converted into OH radicals).
2
(converted into RO radicals
Formaldehyde and acetone can be generated by decom-
position of the hydroxyalkoxy radical generated by H-ab-
straction from the ꢀ-carbon atom, via reactions 26, 27 and
2 2
) and HO
2
8. The reaction sequence will result in the formation of equal
amounts of acetone and formaldehyde. Since no CH C(O)-
CH OH was observed (yield <5%), reaction 29 can be
excluded. The formation of acetone serves as a marker for
3
RO + NO f RO + NO
(24a)
(24b)
(25)
2
2
2
reaction 5b and the acetone yield provides a value of k5b/k
0.35 ( 0.03.
5
RO + NO + M f RONO + M
2
2
)
HO + NO f OH + NO
2
2
(
CH ) CCH OH + O f (CH ) C(O )CH OH
3 2 2 2 3 2 2 2
(26)
The presence of NO simplifies the mechanistic interpretation
of the product yields because it provides a more direct
conversion of peroxy into alkoxy radicals via reaction 24a and
(
CH ) C(O )CH OH + NO f (CH ) C(O)CH OH + NO
3 2 2 2 3 2 2
2
(
(
(
27)
removes the need to consider RO
2
+ RO
2
and RO
2
+ HO
2
reactions. As discussed in Section 4, the formation of OH
radicals does not appear to have a major impact on the chemistry
in these experiments.
(
(
CH ) C(O)CH OH f CH C(O)CH + CH OH
3 2 2 3 3 2
28)
29)
The triangles in Figure 5 show the formation of i-butyralde-
hyde in the presence of NO. There was no discernible effect of
[
O
2
] over the range studied and the yield of i-butyraldehyde
observed in the presence of NO was indistinguishable from
that in the absence of NO . This behavior is consistent with
reaction 16 being the source of i-butyraldehyde; NO is not
expected to compete with O for the available R-hydroxy alkyl
radicals. Fitting expression II to the NO data in Figure 5 gave
CH ) C(O)CH OH f CH C(O)CH OH + CH
3
3
2
2
3
2
x
x
2
The yield of formaldehyde exceeds that of acetone by 14 (
x
6% and probably reflects the chemistry associated with H-
abstraction from the γ -carbon atom indicated in the scheme in
Figure 8. Hydrogen abstraction from the γ-carbon atom in
i-butanol followed by oxygen addition and reaction with NO
forms the γ-hydroxyalkoxy radical which can then decompose
an initial i-butyraldehyde yield of 46 ( 3%.
Figure 7 shows the formation of formaldehyde and acetone
versus loss of i-butanol. Filled symbols are data obtained with
1
50 Torr of O
2
and open symbols were obtained with 15 Torr
] on the product
of O . There was no discernible impact of [O
2
2
3 2
to formaldehyde and CH CHCH OH. The reaction sequence is
yields. In marked contrast to the experiments performed in the
absence of NO (see Figure 6), both acetone and formaldehyde
are observed as primary products. The lines through the data in
Figure 7 are linear least-squares fits (for consumptions of
i-butanol <40% in the case of HCHO) and give molar yields of
similar to reactions 26-28 for the ꢀ-hydroxyalkoxy radical, and
it provides an explanation for the additional formaldehyde
formation. The ꢀ-hydroxyalkyl radical formed by decomposition
of (CH
to form the ꢀ-hydroxyalkoxy radical, CH
same alkoxy radical is formed in the oxidation of n-propanol
and is known to decompose27 to give CH
CHO and a CH OH
to give another molecule of
3
)(CH
2
O)CHCH
2
OH will react further with O
2
and NO
3
CH(O)CH
2
OH. The
3
5 ( 3% and 49 ( 3% for acetone and formaldehyde,
respectively. The dashed line through the HCHO data is a fit
of an equation analogous to that given in expression II. This fit
is provided as a guide for visual inspection of the data trend,
3
2
radical, which then reacts with O
formaldehyde.
2

1
2468 J. Phys. Chem. A, Vol. 114, No. 47, 2010
CH CH(O)CH OH f CH CHO + CH OH (30)
Andersen et al.
3
2
3
2
As shown in the scheme in Figure 8, the γ-hydroxyalkoxy
radical leads to the formation of two formaldehyde molecules.
As noted above, the absence of an effect of [O ] on the HCHO
yield shown in Figure 7 indicates that reaction with O does not
compete with decomposition as a fate of (CH )(CH O)CHCH
radicals. Geiger et al.28 have reported that reaction with O
2
2
3
2
2
OH
and
2
decomposition are competing fates for the structurally similar
isobutoxy radical, (CH )(CH O)CHCH . Further work is needed
to better understand the behavior of these alkoxy radicals.
.6. Products of OH Radical Initiated Oxidation of
3
2
3
3
i-Butanol in 700 Torr of Air. The products of OH radical
initiated oxidation of i-butanol in the presence of NO were
investigated by irradiating mixtures of 27-28 mTorr of i-
butanol, 125-132 mTorr of CH ONO and 0-13 mTorr of NO
3
in 700 Torr air diluent. Acetone was observed as a major
product. The formation of acetone is plotted versus loss of
i-butanol in Figure 9. The line through the data is a linear least-
squares fit that gives an acetone yield of 61 ( 4%. We looked
for IR features attributable to i-butyraldehyde in the product
Figure 9. Formation of acetone vs the loss of i-butanol following UV
irradiation of i-butanol/CH ONO/NO mixtures in 700 Torr of air diluent.
3
3
spectra, but absorption by CH ONO and its photolysis products
be formed in experiments conducted in the presence of NO,
some fraction of the observed i-butanol loss will be attributable
to reaction with OH, which would lead to an overestimation of
precluded the detection of i-butyraldehyde. The observation of
acetone in a yield of 61 ( 4% is consistent with expectations
based upon the SAR estimates (see Section 3.2) that 57% of
the reaction of OH radicals with i-butanol occurs at the ꢀ-
position and the conclusion (see Section 3.2) that decomposition
k
5b/k
5
. Although the fact that the presence of NO does not have
a discernible impact on the i-butyraldehyde yield (see Figure
) suggests that OH radicals do not make a major contribution
to chemistry in the system, we opt to cite an upper limit of
< 0.38.
We report the results of the first product study of the OH
5
3 2 2
via reaction 28 is the fate of (CH ) C(O)CH OH alkoxy radicals.
k5b/k
5
4
. Implications for Atmospheric Chemistry
We report a large body of self-consistent data concerning
radical initiated oxidation of i-butanol. From the observed yield
of acetone we conclude that 61 ( 4% of the reaction of OH
radicals with i-butanol occurs at the ꢀ- position. This branching
ratio is indistinguishable from values estimated using SAR
methods that have been used in calculations of the photochemi-
cal ozone creation potential (POCP) of i-butanol (e.g., by Jenkin
the kinetics and mechanism of the oxidation of i-butanol. The
kinetic data for reactions of Cl atoms and OH radicals reported
here are consistent with the results from previous studies, and
we conclude that the rates of reaction of chlorine atoms and
OH radicals with i-butanol at ambient temperature are well
established. The present work is the first study of the mechanism
of the reaction of chlorine atoms with i-butanol. Taking an
29
and Hayman ). As indicated by the SAR calculation in Section
3
-
.2, most of the remaining reaction is likely to proceed at the
CH - group. The present work provides experimental valida-
average of the i-butyraldehyde yields determined in N
in Section 3.3 and in N /O diluent in the absence of NO (and
hence OH radicals) discussed in Section 3.4, we quote a final
value of k5a/k ) 0.51 ( 0.05. From the yield of acetone
observed in N /O diluent in the presence of NO we derive an
estimate of k5b/k ) 0.35 ( 0.03. Given that OH radicals will
2
discussed
2
2
2
tion of the i-butanol mechanism used in such POCP calculations.
The atmospheric chemistry of i-butanol appears to be well
established.
5
2
2
5
Acknowledgment. V. F. A. and O. J. N. acknowledge
financial support from the Danish National Science Research
Council, the Villum Kann Rasmussen Foundation and EURO-
CHAMP2.
References and Notes
(
1) Anderson, J. E.; Baker, R. E.; Hardigan, P. J.; Ginder, J. M. ;
Wallington, T. J. Energy Independence and Security Act of 2007:
Implications for the US Light-Duty Vehicle Fleet; Society of Automotive
Engineers Technical Paper, 09FFL-0302; Society of Automotive Engineers:
2
009.
2) Directive 2009/28/EC of The European Parliament and of the
Council, 2009, Official Journal of the European Union, L 140/16, 2009.
3) Farrell, A. E.; Plevin, R. J.; Turner, B. T.; Jones, A. D.; O’Hare,
(
(
M.; Kammen, D. M. Science 2006, 311, 506.
(4) Atsumi, S.; Hanai, T.; Liao, J. C. Nature 2008, 451, 7174.
(5) Mueller, S. A.; Anderson, J. E.; Wallington, T. J.; Hammond, R. M.
J. Chem. Educ. 2009, 86, 1045.
6) Andersen, V. F.; Anderson, J. E.; Wallington, T. J.; Mueller, S. A.;
Nielsen, O. J. Energy Fuels 2010, 24, 2683.
7) Andersen, V. F.; Anderson, J. E.; Wallington, T. J.; Mueller, S. A.;
Nielsen, O. J. Energy Fuels 2010, 24, 3647.
8) 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.
(
(
(
Figure 8. Mechanism of the OH-initiated oxidation of i-butanol.

Atmospheric Chemistry of i-Butanol
J. Phys. Chem. A, Vol. 114, No. 47, 2010 12469
Phys. 2006, 6, 3625; IUPAC Subcommittee for Gas Kinetic Data Evaluation,
(20) National Institute of Standards and Technology. Standard Reference
Database 17, Version 7.0, Release 1.5; http://kinetics.nist.gov/kinetics/,
downloaded October 2010.
http://www.iupac-kinetic.ch.cam.ac.uk.
(
9) Wu, H.; Mu, Y.; Zhang, X.; Jiang, G. Int. J. Chem. Kinet. 2003,
5, 81–87.
10) Mellouki, A.; Oussar, F.; Lun, X.; Chakir, A. Phys. Chem. Chem.
Phys. 2004, 6, 2951.
11) Wallington, T. J.; Gierczak, C. A.; Ball, J. C.; Japar, S. M. Int.
J. Chem. Kinet. 1989, 21, 1077.
12) Calvert, J. G.; Derwent, R. G.; Orlando, J. J.; Tyndall, G. S.;
3
(21) Wallington, T. J.; Schneider, W. F.; Barnes, I.; Becker, K. H.;
Sehested, J.; Nielsen, O. J. Chem. Phys. Lett. 2000, 322, 97.
(22) Taatjes, C. A.; Christensen, L. K.; Hurley, M. D.; Wallington, T. J.
J. Phys. Chem. A 1999, 103, 9805.
(
(
(
23) Yamanaka, T.; Kawasaki, M.; Hurley, M. D.; Wallington, T. J.;
Schneider, W. F. Phys. Chem. Chem. Phys. 2007, 9, 4211.
24) Hurley, M. D.; Wallington, T. J.; Laursen, L.; Javadi, M. S.; Nielsen,
O. J.; Yamanaka, T.; Kawasaki, M. J. Phys. Chem. A 2009, 113, 7011.
25) Meagher, R. J.; McIntosh, M. E.; Hurley, M. D.; Wallington, T. J.
Int. J. Chem. Kinet. 1997, 29, 619.
(
(
Wallington, T. J. Mechanisms of Atmospheric Oxidation of the Alkanes;
Oxford University Press: 2008.
(
(
13) Wallington, T. J.; Andino, J. M.; Lorkovic, I. M.; Kaiser, E. W.;
Marston, G. J. Phys. Chem. 1990, 94, 3644.
14) Ullerstam, M.; Ljungstr o¨ m, E.; Langer, S. Phys. Chem. Chem. Phys.
001, 3, 986.
15) Le Cr aˆ ne, J.-P.; Villenave, E.; Hurley, M. D.; Wallington, T. J.;
Nishida, S.; Takahashi, K.; Matsumi, Y. J. Phys. Chem. A 2004, 108, 795.
16) Th e´ venet, R.; Mellouki, A.; Le Bras, G. Int. J. Chem. Kinet. 2000,
2, 676.
17) Calvert, J. G.; Atkinson, R.; Kerr, J. A.; Madronich, S.; Moortgat,
(
(
26) Wallington, T. J.; Dagaut, P.; Kurylo, M. J. Chem. ReV. 1992, 92, 667.
(
27) Orlando, J. J.; Tyndall, G. S.; Wallington, T. J. Chem. ReV. 2003,
2
1
03, 4657.
(
(28) Geiger, H.; Barnes, I.; Becker, K. H.; Bohn, B.; Brauers, T.; Donner,
B.; Dorn, H.-P.; Elend, M.; Freitas Dinis, C.; Grossmann, D.; Hass, H.;
Hein, H.; Hoffmann, A.; Hoppe, L.; H u¨ lsemann, F.; Kley, D.; Klotz, B.;
Libuda, H. G.; Maurer, T.; Mihelcic, D.; Moortgat, G. K.; Olariu, R.; Neeb,
P.; Poppe, D.; Ruppert, L.; Sauer, C. G.; Shestakov, O.; Somnitz, H.;
Stockwell, W. R.; Th u¨ ner, L. P.; Wahner, A.; Wiesen, P.; Zabel, F.; Zellner,
R.; Zetzsch, C. J. Atmos. Chem. 2002, 42, 323.
(
3
(
G. K.; Wallington, T. J.; Yarwood, G., The Mechanisms of Atmospheric
Oxidation of the Alkenes; Oxford University Press: 2000.
(
(
18) Kwok, E. S. C.; Atkinson, R. Atmos. EnViron. 1995, 29, 1685.
19) Bethel, H. L.; Atkinson, R.; Arey, J. Int. J. Chem. Kinet. 2001, 33,
(
29) Jenkin, M. E.; Hayman, G. D. Atmos. EnViron. 1999, 33, 1275.
3
10.
JP107950D