Atmospheric Chemistry of CH3CH2OCH3: Kinetics and Mechanism of Reactions with Cl Atoms and OH Radicals

Article abstract of DOI:10.1002/kin.21051

The atmospheric chemistry of methyl ethyl ether, CH₃CH₂OCH₃, was examined using FT-IR/relative-rate methods. Hydroxyl radical and chlorine atom rate coefficients of k(CH₃CH₂OCH₃+OH) = (7.53 ± 2.86) × 10^?12 cm³ molecule^?1 s^?1 and k(CH₃CH₂OCH₃+Cl) = (2.35 ± 0.43) × 10^?10 cm³ molecule^?1 s^?1 were determined (297 ± 2 K). The Cl rate coefficient determined here is 30% lower than the previous literature value. The atmospheric lifetime for CH₃CH₂OCH₃ is approximately 2 days. The chlorine atom–initiated oxidation of CH₃CH₂OCH₃ gives CH₃C(O)H (9 ± 2%), CH₃CH₂OC(O)H (29 ± 7%), CH₃OC(O)H (19 ± 7%), and CH₃C(O)OCH₃ (17 ± 7%). The IR absorption cross section for CH₃CH₂OCH₃ is (7.97 ± 0.40) × 10^?17 cm molecule^?1 (1000–3100 cm^?1). CH₃CH₂OCH₃ has a negligible impact on the radiative forcing of climate.

Full text of DOI:10.1002/kin.21051

Atmospheric Chemistry of
CH CH OCH : Kinetics and
3
2
3
Mechanism of Reactions
with Cl Atoms and OH
Radicals
MADS P. SULBAEK ANDERSEN, SISSEL BJØRN SVENDSEN, FREJA FROM ØSTERSTRØM,2
1
,2
2
OLE JOHN NIELSEN2
1Department of Chemistry and Biochemistry, California State University Northridge, Northridge, CA, 91330-8262
2Copenhagen Center for Atmospheric Research, Department of Chemistry, University of Copenhagen,
2100, Copenhagen, Denmark
Received 15 August 2016; revised 20 September 2016; accepted 21 September 2016
DOI 10.1002/kin.21051
Published online in Wiley Online Library (wileyonlinelibrary.com).
ABSTRACT: The atmospheric chemistry of methyl ethyl ether, CH3CH2OCH3, was examined
using FT-IR/relative-rate methods. Hydroxyl radical and chlorine atom rate coefficients of
−
12
3
−1 −1
k(CH3CH2OCH3+OH) = (7.53 ± 2.86) × 10
cm molecule s
and k(CH3CH2OCH3+Cl)
were determined (297 ± 2 K). The Cl rate co-
−10
3
−1 −1
=
(2.35 ± 0.43) × 10
cm molecule s
efficient determined here is 30% lower than the previous literature value. The atmospheric
lifetime for CH3CH2OCH3 is approximately 2 days. The chlorine atom–initiated oxidation of
CH3CH2OCH3 gives CH3C(O)H (9 ± 2%), CH3CH2OC(O)H (29 ± 7%), CH3OC(O)H (19 ± 7%),
and CH3C(O)OCH3 (17 ± 7%). The IR absorption cross section for CH3CH2OCH3 is (7.97 ±
−17
−1 −1
cm molecule (1000–3100 cm ). CH3CH2OCH3 has a negligible impact on the
0.40) × 10
radiative forcing of climate. ꢀC 2016 Wiley Periodicals, Inc. Int J Chem Kinet 1–11, 2016
INTRODUCTION
cause of the widespread industrial use of ethers, it is im-
portant to know their atmospheric and environmental
impact.
Ethers make up a class of versatile organic compounds
that have a wide range of practical applications, in-
cluding as fuel additives, alternative fuels, and in the
chemical industry [1,2]. They are directly emitted from
both biogenic and anthropogenic sources [2,3]. Be-
The kinetics and atmospheric oxidation mech-
anisms of the symmetric short n-chain ethers,
dimethyl ether, CH3OCH3, and diethyl ether,
CH3CH2OCH2CH3, have been the subject of numer-
ous studies in the past [1,2,4–21]. By contrast, only one
kinetic study of methyl ethyl ether (CH3CH2OCH3),
by Starkey et al. [22], exists in the literature. No study
of the end products of the atmospheric oxidation of
Correspondence to: M. P. Sulbaek Andersen; e-mail: mpsa@
csun.edu. O. J. Nielsen; e-mail: ojn@kiku.dk.
ꢀ
2016 Wiley Periodicals, Inc.
C

2
SULBAEK ANDERSEN ET AL.
CH3CH2OCH3 exists. Unfortunately, there are several
inconsistencies in the data of Starkey et al. For exam-
ple, one would expect the Cl atom rate coefficients for
the short chain length ethers to follow the sequence
of kCH3OCH3 < kCH3CH2OCH3 < kCH3CH2OCH2CH3. How-
ever, the value of k(CH3CH2OCH3 +Cl) reported by
Starkey et al. [22] is greater than the recommended
value for k(CH3CH2OCH2CH3+Cl) [23]. It is diffi-
cult to understand why CH3CH2OCH3 should react
faster than CH3CH2OCH2CH3, with Cl atoms. It ap-
pears that either there are differences in the mech-
anisms of the chlorine atom reactions for the two
compounds (unlikely), or that there are errors in the
kinetic data. In any event, our knowledge of the at-
mospheric chemistry of the short chain ethers appears
incomplete.
To improve our understanding of the atmospheric
chemistry of ethers and, in particular, the chlorine rad-
ical kinetics for CH3CH2OCH3, the chlorine atom–
initiated oxidation of CH3CH2OCH3 was investi-
gated. The OH radical kinetics and the IR spectrum
of CH3CH2OCH3 was also measured, and an esti-
mate for the atmospheric lifetime and global warm-
ing potential (GWP) for CH3CH2OCH3 is provided
herein.
HO2 + O3 → OH + 2O2
HO + HO → H O + O
(6)
(7)
2
2
2
2
2
H O + hv → 2 OH
2
2
(8)
OH + H2 → H + H2O
OH + O3 → HO2 + O2
(9)
(10)
(11)
(12)
OH + H2O2 → HO2 + H2O
OH + HO2 → H2O + O2
Ozone was produced using a commercially available
ozone generator from O3-Technology. The ozone was
preconcentrated using a silica gel trap, reducing the
amount of O2 introduced into the chamber. Reaction
(3) and (4) results in not only OH radicals in the ground
state but also in excited vibrational states with ν ࣘ 4
and ν ࣘ 9 for reaction (3) and (4), respectively [25–27].
However, it is safe to assume that excited OH will react
exclusively with N2 and O2 and not any of the organ-
ics, since the rate coefficients for vibrational relaxation
EXPERIMENTAL
The photochemical reactor consists of a 101-dm3
quartz chamber interfaced with a Bruker (Karlsruhe,
Germany) IFS 66v/s FTIR spectrometer [24]. The chlo-
rine atoms were produced in situ by photolysis of
molecular chlorine (Air Liquide, Denmark):
−15
of OH by N2 and O2 are on the order of 10
and
−1 −1
cm molecule s , respectively [28], and the
−13
3
10
mixing ratios of O2 and N2 are 4−5 orders of magni-
tude larger than those of the organic reactants.
1
The rate coefficients for O( D) reaction with refer-
Cl2 + hv(λ < 400 nm) → 2 Cl
(1)
ence compounds and H2 at 298 K are all of comparable
−
10
3
−1 −1
magnitude, ꢀ10 cm molecule s [29–32] (there
1
Hydroxyl radicals (OH) were generated effectively
by photolysis of O3 using UVB lamps (Waldmann
F85/100 UV6, Wavelength region 280–360 nm) in the
presence of H2:
is no rate coefficient available for the O( D) reactions
with the ethers). However, the mixing ratios of H2 are
typically 2–3 orders of magnitude larger than those of
1
the organics and the reaction of O( D) with the or-
ganics can be ignored. Similarly, the reactions of HO2
radicals formed in reactions (5), (10) and (11), with any
of our reactants are too slow to be of any concern [33].
The concentrations of reactants and products were
monitored using IR spectra derived from 32 coadded
ꢀ
ꢁ
1
O3 + hv → O D + O2
(2)
(3)
(4)
(5)
ꢀ
ꢁ
1
H2 + O D → OH + H
−1
interferograms with a spectral resolution of 0.25 cm
and an analytical path length of 53.94 m. Analysis of
the spectra was carried out by a process of spectral
stripping in which scaled reference spectra were sub-
tracted from the sample spectrum. The IR bands used
for analysis of C2H6, C3H8, n-C4H10, and n-C5H12
were centered at 1480, 1471, 1383, and 1465, and
O3 + H → OH + O2
H + O2 + M → HO2 + M
International Journal of Chemical Kinetics DOI 10.1002/kin.21051

ATMOSPHERIC CHEMISTRY OF CH3CH2OCH3
3
1
462 cm−1, respectively. A selection of CH3CH2OCH3
RESULTS AND DISCUSSION
IR bands (see Fig. 6) were used in the kinetic stud-
ies and in the product studies. Reference spectra were
obtained by expanding known volumes of reference
compounds into the reaction chamber.
Relative Rate Study of Cl + CH3CH2OCH3
The rate of reaction (15) was measured relative to re-
actions (16) and (17):
All reagents except CH3CH2OCH3 were ob-
tained from commercial sources with purities >99%.
CH3CH2OCH3 is not commercially available and was
synthesized in the laboratory through a Williamson
ether synthesis [34] using diethyl sulfate and sodium
methoxide as reagents. The resulting sample was as-
sured to be devoid of any impurities by FTIR analysis.
Several freeze–pump–thaw cycles were performed on
all reagents before use.
Dark experiments were performed ensuring no un-
wanted loss of reference compounds, reactants, or ref-
erence compounds within the chamber. Mixtures of re-
actants were irradiated with UV and subsequently left
in the dark for 30 min. No loss (<2%) was observed.
We conclude that loss of reactants or references due to
photolysis or heterogeneous reactions are not a com-
plication.
CH3CH2OCH3 + Cl → products
C H + Cl → products
(15)
(16)
(17)
2
6
C3H8 + Cl → products
Initial reaction mixtures consisted of 10.4–10.5 mTorr
CH3CH2OCH3, 52.2-52.4 mTorr Cl2, and either 11.1
mTorr C2H6 (ethane) or 10.7 mTorr C3H8 (propane)
in a total of 700 Torr air. The loss of CH3CH2OCH3
is plotted against the loss of the reference compounds
in Fig. 1. Linear least squares analysis of the data in
Fig. 1 gives k15/k17 and k16/k17 of (4.25 ± 0.46) and
The relative rate method is a well-established tech-
nique for measuring the rate coefficients of Cl atoms
or OH radicals with organic compounds. For two com-
peting reactions:
(
1.57 ± 0.16), respectively. Using k16 = (5.9 ± 0.06)
−
−
11
3
−1 −1
×
×
±
10
10
cm molecule
s
and k17 = (1.4 ± 0.06)
10
3
−1 −1
cm molecule
s
[35] gives k15 = (2.51
−10
3
−1 −1
0.27) × 10
cm molecule
s
and (2.20 ±
−10
3
−1 −1
0
.24) × 10
cm molecule s , respectively. We
Cl/OH + CH3CH2OCH3 → Products
(13)
(14)
cite a final rate coefficient of k15 = (2.35 ± 0.43) ×
−
10
3
−1 −1
1
0
cm molecule s , which is the average of
Cl/OH + Reference → Products
The decay of CH3CH2OCH3 can be expressed as
ꢂ
ꢃ
[
CH3CH2OCH3]t0
ln
[
CH3CH2OCH3]t
ꢂ
ꢃ
kreactant
[Reference]t0
=
ln
(I)
kReference
[Reference]t
where
[CH3CH2OCH3]t0,
[CH3CH2OCH3]t,
[Reference]t0, and [Reference]t are the concentrations
of CH3CH2OCH3 and the reference compound at
times t0 and t. Equation (I) describes a straight line
with a slope equal to the ratio of the rate coefficients
for the reaction of CH3CH2OCH3 with either Cl atoms
or OH radicals and that of the corresponding reference
compound with either Cl atoms or OH radicals.
Quoted uncertainties for the relative rate data and
product yields are two standard deviations from the
least squares regressions (forced through zero), to-
gether with our estimated errors (± 1% of the initial
concentration) associated with the IR spectral analysis
of the relative reactant and reference concentrations.
Figure 1 The loss of CH3CH2OCH3 versus the loss of
C2H6 (circles) and C3H8 (triangles) in the presence of Cl
atoms in 700 Torr total pressure of air, 297 ± 2 K.
International Journal of Chemical Kinetics DOI 10.1002/kin.21051

4
SULBAEK ANDERSEN ET AL.
Table I Summary of Short n-Chain (2 ࣘ C ࣘ 4) Ether +
1.4
Cl Rate Coefficient Data
C H
3
8
010 × k(Ether + Cl)
1
1
0
.2
.0
.8
1
C H
4
10
3
−1
)
Reaction
(cm s
Reference
CH3OCH3 + Cl
1.51 ± 0.08
3.49 ± 0.67
Nelson et al. [1]
Starkey et al. [22]
This work
CH3CH2OCH3 + Cl
2
.35 ± 0.43
CH3CH2OCH2CH3
2.54 ± 0.17
Nelson et al. [1]
C H
5 12
+
Cl
0.6
0
0
.4
.2
the two determinations with a stated uncertainty that
encompasses the extremes for the two determinations.
Starkey et al. [22] reported a value of k15 = (3.49
−10
3
−1
0.0
±
0.67) × 10
cm s based on experiments us-
0
.0
0.1
0.2
0.3
0.4
0.5
0.6
ing the relative rate technique. The reason for the
large discrepancy between our result and that from
Starkey et al. is unclear, but as mentioned by Calvert
et al. [23] there are additional internal inconsisten-
cies in the work of Starkey et al. Our determination
of k6 is significantly lower than the one by Starkey
et al. [22], as expected it is lower than the liter-
ature value for k(CH3CH2OCH2CH3+Cl), and it is
consistent with the expected trend of reactivity for
CH3OCH3, CH3CH2OCH3, and CH3CH2OCH2CH3
with Cl atoms: k(CH3OCH3+Cl) = (1.51 ± 0.08) ×
Ln ( [Reference] / [Reference] )
t0
t
Figure 2 The loss of CH3CH2OCH3 versus the loss of
C3H8 (upward triangles), n-C4H10 (circles), and n-C H12
(downward triangles) in the presence of OH radicals in 700
Torr total pressure of air, 297 ± 2 K.
5
312.7–319.0 mTorr H , and either 11.6 mTorr propane,
2
11.2 mTorr butane, 12.1 mTorr pentane, or 15.8 mTorr
diethyl ether in a total pressure of 700 Torr air. The
loss of CH CH OCH is plotted against the loss of the
3
2
3
−10
3
−1 −1
s
10
cm molecule
k(CH3CH2OCH2CH3+Cl) = (2.54 ± 0.17) × 10
(see Table I). Reaction occurs
< k15(CH3CH2OCH3 +Cl)
reference compounds in Fig. 2.
−10
<
Linear least squares analysis gives k /k = 5.87
18
19
3
−1 −1
s
cm molecule
± 0.55, k /k = 3.18 ± 0.31, and k /k = 2.17 ±
1
8
20
18 21
−12
mainly at the CH2 groups, which explains the trend
of increasing reactivity with increasing alkyl chain
lengths (e.g., the rate coefficient for the reaction of di-n-
0.14. Using k = (1.11 ± 0.17) × 10 , k = (2.36
1
9
20
−
1
2
−12
3
± 0.35) × 10 , and k = (3.96 ± 0.79) × 10 cm
21
−
1
−1
−12
cm3
molecule
s
[36] gives k = (6.51 ± 1.17) × 10
18
,
−10
3
−1 −1
−12
−12
propyl ether with Cl is 3.7 × 10 cm molecule
23].
s
(7.50 ± 1.33) × 10 , and (8.59 ± 1.8) × 10
−
1
−1
[
molecule
s . The three determinations of k span
18
a significant range, notably the results using n-C5H12
as a reference are 30% larger than those obtained using
C3H8 as a reference. We are at odds explaining why this
would be. Still, the three values of k18 obtained here
using three different references agree within their range
of uncertainties. We choose to cite a final value which is
the average of the individual determinations together
with error limits which encompass the extremes of
the determinations, hence, k18(CH3CH2OCH3+OH) =
Relative Rate Study of OH + CH CH OCH
3
2
3
The rate of reaction (18) was measured relative to re-
actions (19)–(21):
CH3CH2OCH3 + OH → products
C3H8 + OH → products
(18)
(19)
(20)
(21)
−12
3
−1 −1
(7.53 ± 2.86) × 10
cm molecule s . Within
the reported ranges of uncertainty, the value of k18,
obtained in the present work, is consistent with the
+
3.88
2.62
n-C4H10 + OH → products
value obtained by Starkey et al. (k = (6.60
) ×
−
−12
3
−1 −1
s [22]).
10
cm molecule
n-C5H12 + OH → products
Cl Atom Product Study
The experiments were performed with initial mix-
tures of 9.9–11.5 mTorr CH3CH2OCH3, 2–3 Torr O3,
The atmospheric degradation mechanism was stud-
ied using the UV irradiation of CH3CH2OCH3/Cl2/air
International Journal of Chemical Kinetics DOI 10.1002/kin.21051

ATMOSPHERIC CHEMISTRY OF CH3CH2OCH3
5
Figure 3 IR spectra of a mixture of 10.4 mTorr CH3CH2OCH3 and 53.1 mTorr Cl2 in 700 Torr air diluent before (A) and after
B) 15 s of UV irradiation. The consumption of CH3CH2OCH3 was 42%. Subtraction of features attributed to CH3CH2OCH3
(
from panel B gives the product spectra shown in panel C. Comparison of reference spectra, panels D–G, shows formation of
these products. Subtraction of IR features due to products shown in panel D–G gives the residual product spectrum shown in
panel H.
mixtures. Reaction mixtures consisted of 9.5–
and 53.1 mTorr Cl2 in 700 Torr of air to UV irradia-
tion for 15 s. Subtraction of IR features attributed to
CH3CH2OCH3 from panel B gives the product spec-
trum in panel C. Comparison of the product features
in panel C with the reference spectrum in panel D,
E, F, and G shows the formation of CH3C(O)H (Q
1
1
0.6 mTorr CH3CH2OCH3, 52.1–54.8 mTorr Cl2, and
40, 350, and 700 Torr O2 in a total pressure of 700
Torr made up with N2. Figure 3 shows IR spectra ob-
tained before (panel A) and after (panel B) subject-
ing a mixture containing 10.5 mTorr CH3CH2OCH3
International Journal of Chemical Kinetics DOI 10.1002/kin.21051

6
SULBAEK ANDERSEN ET AL.
0
0
0
0
0
0
.30
.25
.20
.15
.10
.05
give yields of 9 ± 2%, 19 ± 7%, 17 ± 7%, and 31
± 7% for CH3C(O)H, CH3OC(O)H, CH3C(O)OCH3,
and CH3CH2OC(O)H, respectively.
Panel A
CH CH OC(O)H
3
2
The reaction of Cl atoms with CH CH OCH oc-
3
2
3
curs via hydrogen abstraction from one of three sites:
˙
Cl + CH3CH2OCH3 → CH3CHOCH3 + HCl
CH OC(O)CH
3
3
(22a)
0
0
.00
.20
Panel B
˙
CH OC(O)H
Cl + CH CH OCH → CH CH OCH + HCl
0
0
0
0
.15
.10
.05
.00
3
3
2
3
2
2
3
(
22b)
CH C(O)H
˙
3
Cl + CH3CH2OCH3 → CH3CH2OCH2 + HCl
0.0
0.2
0.4
0.6
0.8
1.0
(
22c)
Δ[CH CH OCH ] / [CH CH OCH ]
3 t0
3
2
3
3
2
The alkyl radicals formed in reaction (22a) will
add O2 to give peroxy radicals, which will un-
dergo self and cross reaction with other peroxyrad-
icals to mainly produce the corresponding alkoxy
Figure 4 Formation of CH3CH2OC(O)H (panel A, upward
triangles), CH3OC(O)CH3 (panel A, circles), CH3OC(O)H
(panel B, diamonds), and CH3C(O)H (panel B, downward
triangles) versus the loss of CH3CH2OCH3 during the Cl
atom–initiated oxidation. The shades indicate the O2 partial
pressure: 700 Torr (black), 350 Torr (gray), and 140 Torr
˙
˙
radicals, CH3CH(O)OCH3, OCH2CH2OCH3, and
˙
CH2CH2OCH2O:
(open). Dashed lines indicate a small dependence on oxygen
partial pressure for CH3OC(O)H and CH3OC(O)CH3. The
line through the data for CH3C(O)H is a fit based on an
expression, derived by Meagher et al. [37]; see the text for
details.
˙
˙
CH3CHOCH3 + O2 + M → CH3C(OO)HOCH3 + M
(23)
˙
˙
CH3C(OO)HOCH3 + RHO2
˙
˙
→
CH3C(O)HOCH3 + RHO + O2 (24a)
−
1
branch at 1745 cm ), CH3CH2OC(O)H (Q branch at
− −1
1
1
750 cm ), CH3OC(O)H (Q branch at 1755 cm ),
−1
and CH3C(O)OCH3 (R branch at 1777 cm ). Panel
H shows the residual spectrum after subtracting IR
features due to these products. The small residual IR
features shown in panel H are due to HC(O)H (Q
CH3C (OH) HOCH3 + RH(O) + O2
(24b)
(24c)
−
1
CH3C(O)HOCH3 + RH (OH) + O2
branch at 1746 cm ) and HC(O)OH (Q branch at
−1
1
776 cm ). Unfortunately, formation of small
amounts of HC(O)H, and HCO(O)H, from the oxi-
dation of CH3CH2OCH3 is difficult to quantify in our
system, as HC(O)H is a minute, but pervasive, back-
ground contaminant from the chamber walls. Thus
we cannot unequivocally say that the small, but sharp
HC(O)H and HC(O)OH IR features, are due to prod-
ucts of the chemical system under investigation. Figure
˙
˙
CH2CH2OCH3 + O2 + M → OOCH2CH2OCH3 + M
(25)
˙
˙
H
OOCH CH OCH + R O2
2
2
3
˙
˙
→ OCH2CH2OCH3 + RHO + O2 (26a)
4
shows the formation of products versus the loss of
CH3CH2OCH3. A small dependence of oxygen par-
tial pressure was observed for CH3C(O)OCH3 (pos-
itive) and CH3OC(O)H (negative), whereas the yield
of CH3CH2OC(O)H was found to be independent of
O2 partial pressure. Linear fits to the product data
HOCH2CH2OCH3 + RH(O) + O2
(26b)
(26c)
H(O)CCH2OCH3 + RHOH + O2
International Journal of Chemical Kinetics DOI 10.1002/kin.21051

ATMOSPHERIC CHEMISTRY OF CH3CH2OCH3
7
˙
˙
˙
CH3CH2OCH2 + O2 + M
CH3C(O)HOCH3 + M → CH3C(O)H + OCH3 + M
(34)
˙
→
CH3CH2OCH2OO + M
(27)
˙
˙
˙
˙
CH3 + O2 → OOCH3
(35)
CH3CH2OCH2OO + RHO2
→
˙
˙
CH3CH2OCH2O + RHO + O2 (28a)
˙
˙
˙
˙
OOCH3 + RHO2 → OCH3 + RHO + O2 (36)
CH3CH2OCH2OH + RH(O) + O2
CH3CH2OCHO + RHOH + O2
(28b)
˙
OCH3 + O2 → HC(O)H + HO2
(37)
(28c)
and
˙
The alkoxy radicals, OCH2CH2OCH3
CH3OC(O)H was observed in a significant yield
and is formed though reaction (33). The observation
of minor, detectable amounts of HC(O)H indicates
˙
CH3CH2OCH2O, formed in reactions (26a) and
28a), are expected to react with O2 to yield the
(
˙
corresponding aldehyde/formate:
CH3CH(O)OCH3 may also undergo some decompo-
sition via reactions (33) and (34) (note, as discussed
above, we cannot rule out that HC(O)H is a contami-
nant of the chamber walls, of no relevance to the ether
oxidation). However, there are other possible pathways
that can yield HC(O)H in addition to CH3C(O)H, i.e.,
˙
OCH2CH2OCH3 + O2 → CH3OCH2C(O)H + HO2
(29a)
˙
CH3CH2OCH2O + O2 → CH3CH2OC(O)H + HO2
˙
CH3CH2OCH2O + M
(30a)
˙
→
CH3CH2O + HC(O)H + M
(30c)
(38)
In addition, these kinds of radicals have been shown
to undergo direct hydrogen atom elimination, which
would yield the same end products [20]:
˙
CH3CH2O + O2 → CH3C(O)H + H2O
˙
OCH2CH2OCH3 + M → CH3OCH2C(O)H + H + M
Since the yield of CH3CH2OC(O)H shows no O2 par-
tial pressure dependence, reactions (30c) and (38) are
probably of minor importance. Cleavage of the R–OR
bond is also been shown to be very unlikely [21]. In-
stead, reactions (30a) and (30b) appear to be the only
(29b)
˙
CH3CH2OCH2O + M → CH3CH2OC(O)H + H + M
30b)
(
˙
fate of CH3CH2OCH2O, resulting in the aforemen-
tioned absence of any effect of O2 partial pressure
CH3CH2OC(O)H. For the CH3C(O)H data shown in
Fig. 4, panel B, a nonlinear fit was used, as derived by
Meagher et al. [37], which describes the conversion of
CH3CH2OCH3 to CH3C(O)H as a function of time (t),
yield of CH3C(O)H (α), and the rate coefficient ratio
of the reaction of Cl with CH3C(O)H over the Cl re-
action with CH3CH2OCH3 (k (CH3C(O)H+Cl) = 8 ×
The alkoxy radical generated in reaction (24a),
CH3C(O)HOCH3, has four possible fates:
˙
˙
CH3C(O)HOCH3 + O2 → CH3C(O)OCH3 + HO2
(31)
−11
3
−1 −1
10
cm molecule
s
[21]/k15 = 3.40):
˙
CH3C(O)HOCH3 + M → CH3C(O)OCH3 + H + M
32)
(
[
CH3C(O)H]t
α
=
(1 − x)
[
CH3CH2OCH3]t0
1 − 3.40
ꢄ
ꢅ
(
3.40−1)
˙
˙
CH3C(O)HOCH3 + M → CH3OC(O)H + CH3 + M
× (1 − x)
− 1
(33)
(II)
International Journal of Chemical Kinetics DOI 10.1002/kin.21051

8
SULBAEK ANDERSEN ET AL.
where x is defined as the conversion of the ether,
CH3CH2OCH3:
It is possible that small amounts of either hydroxy
compounds (through reaction (24b), (26b), or (28b)
or hydroperoxy compounds (through reaction (39a),
(40a) or (41a) could be formed in our system and ac-
count for the remainder of the carbon balance. It is
also possible that intramolecular H-shift(s) are im-
portant in the chemistry of the CH CH OCH de-
[
CH3CH2OCH3]t
x ≡ 1 − [
CH3CH2OCH3]t0
(III)
It is based on this fit we quote a yield of 9 ± 2% for
CH3C(O)H. As seen in Fig. 4, the yield of CH3C(O)H
did not show any discernible dependence on oxygen
partial pressure. This is surprising since the main path-
way outlined above, reaction (34), is occurring in com-
petition with O2-dependent pathways. The absence of
any discernible dependence oxygen partial pressure
may be indicative of a different mechanism for pro-
duction of CH3C(O)H than the one outlined above, or
possibly the small yield and significant scatter in the
data for CH3C(O)H may cloud the experimental obser-
vation of an O2 dependence. A close inspection of the
initial CH3C(O)H data shows the data points falling
significantly below the fit line, rising toward the line
as conversion of CH3CH2OCH3 approaches comple-
tion. While the scatter in the data is significant, one
could be lead to the conclusion that CH3C(O)H is in
fact a second-order product, not formed directly from
CH3CH2OCH3.
3
2
3
rived peroxy radicals [21], which could allow for
the formation of multifunctional hydroperoxides, e.g.,
CH C(OOH)HOC(O)H. These peroxides would likely
3
have low-vapor pressures and may partition to the walls
of the photoreactor, thus offering a possible explana-
tion for the missing carbon balance.
Figure 5 shows a simplified oxidation mechanism
for CH CH OCH We find that the Cl atom–mediated
3
2
3.
hydrogen abstraction preferentially occurs at the β-
carbon (ꢀ40%) followed closely by the γ -carbon
H-abstraction (30%). We estimate that pathway 1 (α-
hydrogen abstraction) is likely not important. Thus,
30% of the carbon balance is unaccounted for. Includ-
ing upper limits to account for the incomplete carbon
balance our experiments, we can say that less than 70%
of the reaction must proceed though pathway 2, while
the upper limit for pathway 3 is 60%.
It is of interest to compare the Cl atom oxidation
end product results to those expected for the OH-
initiated oxidation. Spectral interference (large satu-
Our observed carbon balance is significantly less
than 100% (total ꢀ70%). Evidence for IR bands in the
product spectra due to formation of CH3OCH2C(O)H
was sought for, but not found. The peroxy radicals gen-
erated in reactions (23), (25), and (27) can react with
HO2 radicals as these are generated in the chamber,
yielding hydroperoxides:
rated O bands) in the product analysis prohibited an
3
analogous experiment to be concluded for the OH oxi-
dation. Empirical structure–activity relationships [38]
can be used to predict OH radical reaction rate co-
efficients for molecules in the gas phase. From these
calculations, we expect 90% of OH-initiated oxida-
tion of CH CH OCH to occur through pathway 2
˙
CH3C(OO)HOCH3 + HO2
3
2
3
producing CH3COC(O)H, and CH3C(O)OCH3, and
0% to occur through pathway 3 producing mainly
→
CH3C (OOH) HOCH3 + O2
(39a)
(39b)
1
CH3CH2OC(O)H. Pathway 1 is expected to be of little
importance in the OH-initiated oxidation. This picture
is broadly consistent in the Cl atom–initiated oxidation
as determined above, although it appears that pathway
CH3C(O)OCH3 + O2 + H2O
3
(H-abstraction from the γ -carbon) is more important
˙
OOCH2CH2OCH3 + HO2
in the Cl atom reaction than in the OH radical reaction.
→
HOOCH2CH2OCH3 + O2
(40a)
(40b)
ATMOSPHERIC LIFETIME AND GLOBAL
WARMING POTENTIAL
H(O)CCH2OCH3 + O2 + H2O
The atmospheric lifetime depends on the rate coeffi-
cients for reactions with Cl atoms and OH radicals
and the atmospheric concentrations of these radicals.
Because the concentration of OH radicals is approxi-
˙
CH3CH2OCH2OO + HO2
→
CH3CH2OCH2OOH + O2
(41a)
(41b)
3
mately a factor 10 higher than the concentration of Cl
atoms [33], the lifetime of CH3CH2OCH3 is dictated
6
−3
CH3CH2OC(O)H + O2 + H2O
by reaction with OH radicals. Using [OH] = 10 cm
International Journal of Chemical Kinetics DOI 10.1002/kin.21051

ATMOSPHERIC CHEMISTRY OF CH3CH2OCH3
9
Figure 5 Possible degradation pathways for the Cl atom–initiated oxidation of CH3CH2OCH3. Proposed stable products are
indicated by the boxes. Based on the experimental results, pathway 1 appears to be of minor importance with pathways 2 and 3
˙
being approximately equal. The data also suggest that CH3C(O)HOCH3 decomposition may in fact not be an important source
of CH3C(O)H (see the text for details). It should be noted that under most ambient atmospheric conditions, the peroxy radicals
will react with NO forming the corresponding alkoxy radicals and NO2.
−
17
−1
and assuming τ(CH3CH2OCH3) = 1/(kOH[OH]), the
atmospheric lifetime of CH3CH2OCH3 is estimated to
be approximately 2 days. It should be noted that in the
marine boundary layer the Cl atom concentrations can
reach concentrations of 10 –10 cm making the Cl
atom H-abstraction an additional and possibly signifi-
cant degradation pathway.
Halogen substitution in methyl ethyl ether produces
a range of analogous hydro-fluoro-ethers or hydro-
chloro-fluoro-ethers, of which some have commer-
cial importance. For these halogenated compounds, the
C–F bond stretches impart a strong absorption of long-
wave infrared radiation and tend to result in signif-
icant radiative efficiencies. Literature values for the
is (7.97 ± 0.40) × 10
cm molecule (1000–3100
−1
cm ). The radiative efficiency of CH CH OCH can
3
2
3
be estimated by the narrowband model as described by
Pinnock et al. [40] with an addition of a correction
factor for incomplete vertical mixing (calculated to be
0.0227 [41]). Using the IR spectrum shown in Fig. 6,
4
6
−3
−3
we calculate a radiative efficiency of 1.18 × 10
W
−2 −1
m
ppb of CH CH OCH . Using the atmospheric
3 2 3
lifetime determined above and the radiative efficiency
of CH CH OCH , the GWP can be calculated as
3
2
3
follows:
ꢀ
ꢀ ꢁꢁ
ꢁ
GWP CH3CH2OCH3 t
ꢁ
o
t
−t/τ(CH3CH2OCH3)
∫
F
× e
1
00-year time-horizon GWPs for halogen-substituted
methyl ethyl ethers range from 1 (CF3CH2OCH3) to
450 (CF3CHFOCF3) [39]. As part of the present
CH3CH2OCH3
=
(IV)
ꢁ
t
∫ F
× R (t)
o
CO2
6
ꢁ
work, the IR spectrum of CH3CH2OCH3 was mea-
sured in 700 Torr of air diluent at 297 ± 2 K. The inte-
grated IR absorption cross section for CH3CH2OCH3
where t is 20, 100, or 500 year time horizons,
FCH3CH2OCH3 and τ(CH CH OCH ) are the radiative
efficiency and atmospheric lifetime of CH CH OCH ,
3
2
3
3
2
3
International Journal of Chemical Kinetics DOI 10.1002/kin.21051

1
0
SULBAEK ANDERSEN ET AL.
Figure 6 IR spectrum of 2.07–12.5 mTorr CH3CH2OCH3 in 700 Torr of air (297 ± 2 K). Inset shows the linearity of the
absorptions.
FCO2 is the the radiative forcing per unit mass of CO2,
and R(t) is the response function, which describes the
decay of an instantaneous pulse of CO2. The denomina-
tor in expression (IV) makes up the absolute GWP for
CO2 which has been evaluated by the World Meteoro-
the mechanisms of reactions of Cl atoms and OH with
CH3CH2OCH3 are quite similar.
A lifetime of approximately 2 days was determined
based on the reaction with OH radicals. The calcu-
lated GWPs shows that CH3CH2OCH3 would have a
negligible impact on the radiative forcing of climate.
−2
logical Organization as 0.192, 0.676, and 2.223 W m
−1
ꢁ
ppm for t = 20, 100, and 500 year time horizons,
respectively [42]. Substituting the radiative efficiency
and atmospheric lifetime values determined above into
expression (IV) gives GWPs for CH3CH2OCH3 of 0
for any time horizon.
We acknowledge Martyn Jevric and Søren Broman (Univer-
sity of Copenhagen) for their help in synthesis and purifica-
tion of CH3CH2OCH3.
BIBLIOGRAPHY
CONCLUSION
1. Nelson, L.; Rattigan, O.; Neavyn, R.; Sidebottom, H.;
Treacy, J.; Nielsen, O. J. Int J Chem Kinet 1990, 22,
A substantial body of kinetic and mechanistic data
for the atmospheric chemistry of CH3CH2OCH3
is presented here. Using the relative rate method,
the OH radical and Cl atom reaction rate coef-
ficients for CH3CH2OCH3 were determined. Our
k(CH3CH2OCH3+Cl) is significantly lower than what
has previously been reported; however, we believe that
the previous literature value was erroneously large.
Both the OH radical and the Cl atom rate coeffi-
cients determined in the present work are consistent
with the expected sequence of reactivities: kCH3OCH3
1111–1126.
2
3
. Atkinson, R.; Arey, J. Chem Rev 2003, 103, 4605–4638.
. Mellouki, A.; Le Bras, G.; Sidebottom, H. Chem Rev
2003, 103, 5077–5096.
4. Perry, R. A.; Atkinson, R.; Pitts, J. N. J Chem Phys
1997, 67, 611–615.
5
6
7
. Tully, F. P.; Droege, A. T. Int J Chem Kinet 1987, 19,
251–259.
. Wallington, T. J.; Liu, R.; Dagaut, P.; Kurylo, M. J. Int J
Chem Kinet 1988, 20, 41–49.
. Wallington, T. J.; Andino, J. M.; Skewes, L. M.; Siegl,
W. O.; Japar, S. M. Int J Chem Kinet 1989, 21, 993–
<
kCH3OCH2CH3 < kCH3CH2CH2OCH3. Cl-initiated oxida-
1
001.
. Arif, M.; Dellinger, B.; Taylor, P. H. J Phys Chem A
997, 101, 2436–2441.
tion gives CH3C(O)H, CH3OC(O)H, CH3C(O)OCH3,
and CH3CH2OC(O)H. Minor amounts of HC(O)H and
HC(O)OH were also detected. Comparison with the
expected OH radical oxidation products suggests that
8
1
9. DeMore, W. B.; Bayes, K. D. J Phys Chem A 1999, 103,
2649–2654.
International Journal of Chemical Kinetics DOI 10.1002/kin.21051

ATMOSPHERIC CHEMISTRY OF CH3CH2OCH3
11
1
0. Bonard, A.; Daele, J.-L.; Vovelle, C. J Phys Chem A
002, 106, 4384–4389.
1. Wallington, T. J.; Skewes, L. M.; Siegl, W. O.; Wu,
C.-H.; Japar, S. M. Int J Chem Kinet 1988, 20, 867–
30. Burkholder, J. B.; Sander, S. P.; Abbatt, J.; Barker,
J. R.; Huie, R. E.; Kolb, C. E.; Kurylo, M. J.; Orkin,
V. L.; Wilmouth, D. M.; Wine, P. H. Chemical Kinet-
ics and Photochemical Data for Use in Atmospheric
Studies, Evaluation No. 18, JPL Publication 15-10; Jet
Propulsion Laboratory: Pasadena, CA, 2015.
31. Baasandorj, M.; Hall, B. D.; Burkholder, J. B. Atmos
Chem Phys 2012, 12, 11753–11764.
2
1
875.
1
2. Jenkin, M. E.; Hayman, G. D.; Wallington, T. J.; Hurley,
M. D.; Ball, J. C.; Nielsen, O. J.; Ellermann, T. J Phys
Chem 1993, 97, 11712–11723.
1
3. Langer, S.; Ljungstrom, E.; Wangberg, I.; Wallington,
T. J.; Hurley, M. D.; Nielsen, O. J. Int J Chem Kinet
32. Dillon, T. J.; Vereecken, L.; Horowitz, A.; Khamaganov,
V.; Crowley, J. N.; Lelieveld, J. Phys Chem Chem Phys
2011, 13, 18600–18608.
33. Finlayson-Pitts, B. J.; Pitts, J. N. Jr., Chemistry of the
Upper and Lower Atmosphere; Academic Press: Lon-
don, 1999.
34. Williamson, A. W. J Chem Soc 1852, 4, 229–239.
35. 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–
4055.
36. Calvert, J. G.; Derwent, R. G.; Orlando, J. J.; Tyndall,
G. S.; Wallington, T. J. Mechanisms of Atmospheric Ox-
idation of the Alkanes; Oxford University Press: Oxford,
UK, 2008.
1
996, 28, 299–306.
4. Wallington, T. J.; Japar, S. M. Environ Sci Technol 1991,
5, 410–415.
1
1
1
1
1
1
2
5. Japar, S. M.; Wallington, T. J.; Richert, J. F. O.; Ball,
J. C. Int J Chem Kinet 1990, 22, 1257–1269.
6. Mellouki, A.; Teton, S.; Le Bras, G. Int J Chem Kinet
1995, 27, 791–805.
7. Michael, M.; Nava, D. F.; Payne, W. A.; Stief, L. J. J
Chem Phys 1979, 70, 3652–3656.
8. Park, J.-Y.; Slagle, I. R.; Gutman, D. J Phys Chem 1983,
8
7, 1812–1818.
9. Semadeni, M.; Stocker, D. W.; Kerr, J. A. J Atmos Chem
993, 16, 79–93.
1
2
2
0. Orland, J. J. Phys Chem Chem Phys 2007, 9, 4189–4199.
1. Wang, S.; Wang, L. Phys Chem Chem Phys 2016, 18,
37. Meagher, R. J.; McIntosh, M. E.; Hurley M. D.;
Wallington, T. J. Int J Chem Kinet 1997, 29, 619–
625.
7707–7714.
2
2. Starkey, D. P.; Holbrook, K. A.; Oldershaw, G. A.;
Walker, R. W. Int J Chem Kinet 1997, 29, 231–236.
3. Calvert, J. G.; Mellouki, A.; Orlando, J. J.; Pilling,
M. J.; Wallington, T. J. The Mechanisms of Atmospheric
Oxidation of the Oxygenates; Oxford University Press:
Oxford, UK, 2011.
38. Kwok, E. S. C.; Atkinson, R. Atmos Environ 1995, 29,
1685–1695.
2
39. IPCC, 2013: Climate Change 2013: The Physical Sci-
ence Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on
Climate Change, Stocker, T. F.; Qin, D.; Plattner, G.-K.;
Tignor, M.; Allen, S. K.; Boschung, J.; Nauels, A.; Xia,
Y.; Bex, V.; Midgley, P. M., Eds.; Cambridge University
Press: Cambridge, UK, 2013.
40. Pinnock, S.; Hurley, M. D.; Shine, K. P.; Wallington,
T. J.; Smith, T. J. J. Geophys Res 1995, 100, 23227–
23238.
41. Hodnebrog, Ø.; Etminan, M.; Fuglestvedt, J. S.;
Marston, G.; Myhre, G.; Nielsen, C. J.; Shine, K. P.;
Wallington, T. J. Rev Geophys 2013, 51, 300–378.
42. World Meteorological Organization (WMO), Scien-
tific Assessment of Ozone Depletion: 2006, Global
Ozone; Research and Monitoring Project–Report 50:
Switzerland, Geneva, 2007.
2
2
2
2
2
4. Nilsson, E. J. K.; Eskebjerg, C.; Johnson, M. S. Atmos
Environ 2009, 43, 3029–3033.
5. Aker, P. M.; Sloan, J. J. J Chem Phys 1986, 85, 1412–
1417.
6. Huang, Y.; Gu, Y.; Liu, C.; Yang, X.; Tao, Y. Chem Phys
Lett 1986, 127, 432–437.
7. Streit, G. E.; Whitten, G. Z.; Johnston, H. S. Geophys
Res Lett 1976, 3, 521–523.
8. D’Ottone, L.; Bauer, D.; Campuzano-Jost, P.; Fardy,
M.; Hynes, A. J. Phys Chem Chem Phys 2004, 6, 4276–
4282.
29. Matsumi, Y.; Tonokura, K.; Inagaki, Y.; Kawasaki, M. J
Phys Chem 1993, 97, 6816–6821.
International Journal of Chemical Kinetics DOI 10.1002/kin.21051