Atmospheric Chemistry of Cyclohexane: UV Spectra of c-C6H11. and (c-C6H11)O2. Radicals, Kinetics of the Reactions of (c-C6H11)O2. Radicals with NO and NO2, and the Fate of the Alkoxy Radical (c-C6H11)O.

Article abstract of DOI:10.1021/jp984195x

pulse radiolysis technique was used to measure the UV absorption spectra of c-C₆H₁₁^. and (c-C₆H₁₁)O₂^. radicals over the ranges 230-290 and 220-300 nm, σ(c-C₆H₁₁^.)250nm = (7.0 ± 0.8) × 10^-18 and σ((cC₆H₁₁)O₂^.)_{250 nm} = (5.7 ± 0.6) × 10^-18 cm² molecule^-1. The rate constant for the self-reaction of c-C₆H₁₁^. radicals was k₃ = (3.0 ± 0.4) × 10^-11 cm³ molecule^-1 s^-1. The addition reaction of c-C⁶H₁₁^. radicals with O₂ proceeds with a rate constant k₂ = (1-3 ± 0.2) × 10^-11 cm³ molecule^-1 s^-1. Rate constants for reactions of (c-C₆H₁₁)O₂ radicals with NO and NO₂ were k₄ = (6.7 ± 0.9) × 10^-12 and k₅ = (9.5 ± 1.5) × 10^-12 cm³ molecule^-1 s^-1, respectively. FTIR-smog chamber techniques were used to record the IR spectrum of the peroxynitrate (c-C₆H₁₁)O₂NO₂, determine that the reaction between (c-C₆H₁₁)O₂^. radicals and NO produces a (16 ± 4)% yield of the nitrate (c-C₆H₁₁)ONO₂, and study the atmospheric fate of cyclohexoxy radicals. Decomposition via C-C bond scission and reaction with O₂ are competing fates of the cyclohexoxy radical. In 700-750 Torr total pressure at 296 ± 2K, the rate constant ratio k_decomp/k_O2 = (8.1 ± 1.5) × 10¹⁸ molecule cm^-3. At 296 K in 1 atm of air, 61% of cyclohexoxy radicals decompose and 39% react with O₂. These results are discussed with respect to the literature data concerning the atmospheric chemistry of cyclohexane and analogous compounds.

Full text of DOI:10.1021/jp984195x

2688
J. Phys. Chem. A 1999, 103, 2688-2695
•
•
Atmospheric Chemistry of Cyclohexane: UV Spectra of c-C6H11 and (c-C6H11)O2 Radicals,
•
Kinetics of the Reactions of (c-C6H11)O2 Radicals with NO and NO2, and the Fate of the
Alkoxy Radical (c-C6H11)O•
†
,‡
J. Platz, J. Sehested, and O. J. Nielsen*
Department of Atmospheric Chemistry Plant Biology and Biogeochemistry, Risø National Laboratory,
DK-4000, Roskilde, Denmark
T. J. Wallington*,§
Ford Motor Company 20000 Rotunda DriVe, Mail Drop SRL-3083, Dearborn, Michigan 48121-2053
ReceiVed: October 27, 1998; In Final Form: December 30, 1998
•
11)O •
A pulse radiolysis technique was used to measure the UV absorption spectra of c-C
6
H
11 and (c-C
6
H
2
•
-18
radicals over the ranges 230-290 and 220-300 nm, σ(c-C
6
H
11
)
250 nm ) (7.0 ( 0.8) × 10
and σ((c-
•
-18
2
-1
•
C
6
H11)O
2
)
250 nm ) (5.7 ( 0.6) × 10 cm molecule . The rate constant for the self-reaction of c-C
radicals was k 11 radicals with O
) (3.0 ( 0.4) × 10 cm molecule s . The addition reaction of c-C
proceeds with a rate constant k
6 11
H
-
11
3
-1 -1
11
•
3
6
H
2
-
3
-1 -1
2
) (1.3 ( 0.2) × 10 cm molecule s . Rate constants for reactions of
-12 -12 3
2 4 5
•
(
c-C
6
H
11)O
2
radicals with NO and NO
were k
) (6.7 ( 0.9) × 10 and k
) (9.5 ( 1.5) × 10 cm
-
1
-1
molecule s , respectively. FTIR-smog chamber techniques were used to record the IR spectrum of the
•
peroxynitrate (c-C
a (16 ( 4)% yield of the nitrate (c-C
Decomposition via C-C bond scission and reaction with O
In 700-750 Torr total pressure at 296 ( 2K, the rate constant ratio kdecomp/kO2 ) (8.1 ( 1.5) × 10 molecule
cm . At 296 K in 1 atm of air, 61% of cyclohexoxy radicals decompose and 39% react with O . These
6
H
11)O
2
NO
2
, determine that the reaction between (c-C
, and study the atmospheric fate of cyclohexoxy radicals.
are competing fates of the cyclohexoxy radical.
6 2
H11)O radicals and NO produces
6 2
H11)ONO
2
1
8
-
3
2
results are discussed with respect to the literature data concerning the atmospheric chemistry of cyclohexane
and analogous compounds.
1. Introduction
O2 to form an alkyl peroxy radical:
Alkanes constitute a significant fraction of automotive fuels,
vehicle emissions, and air pollution in urban areas. Atmospheric
oxidation of alkanes is initiated primarily via attack by OH
radicals giving alkyl radicals which then add O2 to give alkyl
peroxy radicals. Subsequent reactions of the alkyl peroxy
radicals determine the atmospheric degradation mechanism, and
hence ozone forming potential, of organic compounds. There
are two reasons why we are interested in the atmospheric
chemistry of cyclohexane. First, cyclohexane is present in small,
but significant (0.1-1.0 wt %), amounts in typical gasoline
We have used pulse radiolysis coupled with time-resolved UV-
vis absorption spectroscopy to determine the UV absorption
•
•
spectra of the c-C H and (c-C H )O radicals, and the
6
11
6
11
2
kinetics of the reactions 2-5.
1
blends sold in the U.S., Germany, and the U.K. An understand-
ing of the atmospheric chemistry of cyclohexane is needed to
assess the impact of its release into the environment. Second,
cyclohexane has a unique unstrained cyclic structure and
information concerning the reactivity of the radicals generated
during its oxidation provides insight into the likely behavior of
analogous secondary radicals formed during the oxidation of
other alkanes.
The atmospheric degradation of c-C6H12 is initiated by
reaction with OH radicals giving an alkyl radical which adds
†
Current address: Haldor Topsoe A/S, Nymoellevej 55, DK-2800
Lyngby, Denmark. E-mail: jss@topsoe.dk.
‡
FTIR product studies were performed to determine the fate
of the cyclohexoxy radical (c-C6H11)O formed in reaction 4a.
E-mail: ole.john.nielsen@risoe.dk.
§
•
E-mail: twalling@ford.com.
1
0.1021/jp984195x CCC: $18.00 © 1999 American Chemical Society
Published on Web 03/11/1999

Atmospheric Chemistry of Cyclohexane
J. Phys. Chem. A, Vol. 103, No. 15, 1999 2689
2. Experimental Section
The two experimental systems used are described in detail
2
,3
elsewhere.
•
2
.1. Pulse Radiolysis System. c-C6H11 radicals were gener-
ated from c-C6H12 by the pulsed radiolysis of SF6/c-C6H12
gasmixtures in a 1 L stainless steel reaction cell with a 30 ns
pulse of 2 MeV electrons from a Febetron 705 field emission
accelerator. SF6 was always in great excess and was used to
generate fluorine atoms:
2
MeV e-
SF6
8 F + products
(6)
The radiolysis dose was varied by insertion of stainless steel
attenuators between the accelerator and the reaction cell. In this
work we refer to the radiolysis dose as a fraction of the
maximum dose that is achievable. The fluorine atom yield was
calibrated by monitoring the transient absorption at 260 nm by
CH3O2 radicals produced by pulse radiolysis of SF6/CH4/O2
4
mixtures, as described previously. Using σ(CH3O2)260 nm ) 3.18
-
18
2
-1 5
×
10
cm molecule , the F atom yield was determined to
15
-3
be (3.18 ( 0.32) × 10 molecules cm at full radiolysis dose
and 1000 mbar of SF6 [4]. The quoted error includes 10%
uncertainty in σ(CH3O2)260nm and two standard deviations from
the experimental absorption measurements. Unless otherwise
stated, all uncertainties reported in this paper are two stand-
ard deviations and standard error propagation methods were
used.
Figure 1. Observed maximum transient absorbance (base 10) at 250
nm versus radiolysis dose following pulsed radiolysis of mixtures of
either (A) 5 mbar c-C
6 6 6 12
H12 and 995 mbar SF or (B) 5 mbar c-C H ,
1
5 mbar O , and 980 mbar SF
2
6
. The solid lines are linear regressions
of the low dose data (filled circles). The inserts show typical transient
absorption traces using (A) 5 mbar c-C H and 995 mbar SF , dose )
6
12
6
32% of maximum, UV path length ) 80 cm and (B) 5 mbar c-C
6
H
12
,
1
5 mbar O
2 6
, and 980 mbar SF , dose ) maximum, UV path length )
The analysis light was provided by a pulsed 150 W
Xenon arc lamp and was multipassed through the reaction
cell using internal White cell optics to give total optical path
lengths of 80, 160, or 200 cm. After leaving the cell, the light
was guided through a monochromator and detected by a
photomultiplier (to record absorption transients) or by a di-
ode array (to record absorption spectra). All absorption tran-
sients were derived from single pulse experiments. Using the
photomultiplier detector, the spectral resolution was 0.8 nm
while it was 1 nm when the diode array was used to record
spectra. Spectral calibration was achieved using a Hg pen ray
lamp.
8
0 cm. The smooth line in the insert in part A is a second-order decay
fit.
the rate of NO2 formation following radiolysis of SF6/c-C6H12/
O2/NO mixtures was used to measure the rate constant for the
reaction of (c-C6H11)O2 radicals with NO. Finally, k5 was
determined by monitoring the rate of NO2 loss following
radiolysis of SF6/c-C6H12/O2/NO2 mixtures:
•
Reagent concentrations used were SF6, 980-995 mbar;
c-C6H12, 5 mbar; O2, 0.95-15.0 mbar; NO, 0.32-1.16 mbar;
and NO2, 0.34-1.07 mbar. All experiments were performed at
2
96 ( 2 K and 1000 mbar total pressure. Chemicals were
2.2. FTIR-Smog Chamber System. The FTIR system was
interfaced to a 140 L Pyrex reactor. Radicals were generated
by the UV irradiation (22 blacklamps) of gas mixtures contain-
ing 20-96 mTorr of c-C6H12, 100-250 mTorr of Cl2, 50-595
Torr of O2, 50-100 mTorr of NO, and 9.21 mTorr of NO2 in
700 Torr total pressure with (N2 + O2) diluent at 296 ( 2 K
(760 Torr ) 1013 mbar ) 101.3 kPa). Reactant loss and product
formation were monitored by FTIR spectroscopy, using an
analyzing path length of 27 m and a resolution of 0.25 cm .
Infrared spectra were derived from 32 coadded spectra. Refer-
ence spectra were acquired by expanding known volumes of
authentic reference compounds into the chamber.
supplied by SF6 (99.9%), Gerling and Holz; c-C6H12 (>99%)
Aldrich; O2, (ultrahigh purity) L’Air Liquide; NO2 (>98%),
Linde Technische Gase; and NO (99.8%), Messer Griesheim.
All chemicals were used as received.
Five sets of experiments were performed using the pulse
•
radiolysis system. First, the UV absorption spectra of c-C6H11
•
and (c-C6H11)O2 radicals were obtained using the diode array
-
1
camera to capture the UV absorption following radiolysis of
SF6/c-C6H12 and SF6/c-C6H12/O2 mixtures. Second, the rate
constant for reaction 3 was determined from the rate of the decay
of the absorption at 250 nm, following the radiolysis of SF6/
c-C6H12 mixtures.
3
. Results and Discussion
.1. UV Absorption Spectrum of the c-C6H11 Radical.
•
3
Following the pulse radiolysis of SF6/c-C6H12 mixtures, a rapid
(complete within 0.5-1.5 µs) increase in absorption was
observed at 250 nm, followed by a slower decay. The insert in
Figure 1A shows the transient absorption at 250 nm following
Third, the rate constant for the association reaction between the
•
c-C6H11 radical and O2 was measured by monitoring the
•
formation of (c-C6H11)O2 radicals in the presence of O2. Fourth,

2690 J. Phys. Chem. A, Vol. 103, No. 15, 1999
Platz et al.
the radiolysis of a mixture of 995 mbar SF6 and 5 mbar c-C6H12.
No absorption was observed when either 995 mbar of SF6, or
5
mbar of c-C6H12, were radiolyzed separately. We ascribe the
absorption shown in the insert Figure 1A to formation of
•
c-C6H11 radicals and their subsequent loss by self-reaction. In
this work we assume that F atoms react exclusively with c-C6H12
•
by H atom abstraction to give c-C6H11 radicals. To work under
conditions where 100% of the F atoms are converted into alkyl
radicals, it is necessary to consider potential interfering radical-
radical reactions such as reactions 3 and 8:
To check for such complications, experiments were performed
using mixtures of 995 mbar of SF6 and 5 mbar of c-C6H12, with
the maximum transient absorption at 250 nm measured as a
function of radiolysis dose. Figure 1A shows the observed
maximum absorbance as a function of the dose. As seen from
Figure 1A, the absorption observed in experiments using
maximum dose (open circle) was significantly less than that
expected based upon a linear extrapolation of the low dose data
(filled circles). This observation suggests that unwanted radical-
radical reactions such as reactions 3 and 8 become important
at higher radical concentrations.
The line through the experimental data in Figure 1A is a linear
least-squares fit to the low dose data and has a slope of 0.769
•
Figure 2. (A) UV-absorption spectrum of the c-C
6
H
11 radical re-
6
ported by Schuler and Patterson (circles) and measured herein (solid
•
6 2
line). (B) UV-absorption spectrum of the (c-C H11)O radical reported
(
(
0.044. Combining this slope with the F atom yield of (3.18
13
by Rowley et al. (circles) and measured in the present work (solid
line).
1
5
-3
0.32) × 10 molecules cm (full dose and [SF6] ) 1000
•
-18
mbar), we obtain σ(c-C6H11 )250 m ) (7.00 ( 0.81) × 10
2
-1
cm molecule . The quoted uncertainty includes both statistical
and potential systematic uncertainties and so reflects the
accuracy of the measurement.
•
The spectrum of the c-C6H11 was measured by recording
the initial absorbance following the pulsed radiolysis of SF6/
c-C6H12 mixtures using a diode array. The delay was 0.1 µs,
the integration time was 1 µs, and the spectral resolution was 1
nm. The initial absorbances were scaled to that at 250 nm and
•
placed on an absolute basis using σ(c-C6H11 )250nm ) (7.00 (
-
18
2
-1
0
2
.81) × 10
cm molecule . The result is shown in Figure
•
A. For comparison the UV absorption spectrum of the c-C6H11
radical reported by Schuler and Patterson is also given in Figure
A. This spectrum was obtained in liquid cyclohexane and, for
6
2
reasons which are unknown, is substantially less intense than
the gas-phase spectrum reported herein.
•
3
.2. Kinetic Study of the Self-Reaction of c-C6H11 Radi-
•
cals. The insert in Figure 1A shows a typical absorption trace
obtained by pulse radiolysis of a mixture of 995 mbar SF6 and
6 11
Figure 3. Plot of the reciprocal half-life for self-reaction of c-C H
radicals as a function of the maximum absorbance at 250 nm.
5
mbar c-C6H12. The trace shows a rapid increase in absorption
tion was varied by varying the initial F atom concentration. A
linear least-squares fit of the data in Figure 3 gives a slope of
at 250 nm followed by a slower decay. We ascribe the decay
to the self-reaction of the c-C6H11 radicals, via reaction 3, The
•
5
-1
(
2.47 ( 0.05) × 10 s ) (k3 2 ln 10)/(σ(alkyl)L), where σ-
•
(alkyl) is the absorption cross section of the c-C6H11 radical
and L is the optical path length (80 cm). The intercept of the
5
-1
linear regression of the data (0.06 ( 0.08) × 10 s is not
decay was fitted with the second-order decay expression A(t)
significantly different from zero. The slope of (2.47 ( 0.05) ×
5
-1
•
)
(A0 - Ainf)/(1 + 2k3(A0 - Ainf)t) + Ainf, where A(t) is the
10 s can be combined with σ(c-C6H11 )250 nm ) (7.00 ( 0.81)
-
18
2
-l
-11
3
measured absorbance at time t and A0 and Ainf are the
absorbances extrapolated to time zero and infinite time,
respectively. The fit is shown in the insert Figure 1A. No
residual absorption was observed, indicating the absence of any
long-lived products which absorb significantly at 250 nm.
The reciprocal half-lives, 1/t1/2 (derived from the fits), are
× 10
cm molecule to give k3 ) (3.0 ( 0.4) × 10
cm
-1 -1
molecule s . The quoted error includes statistical uncertainties
•
associated with both the slope in Figure 3 and σ(c-C6H11 )250 nm.
As expected for such structurally similar radicals, k3 is
indistinguishable from the rate constants for the self-reaction
of alkyl radicals derived from 1,4-dioxane and 1,3,5-trioxane,
•
-11
-11
3
-1
plotted as a function of A0 in Figure 3. The c-C6H11 concentra-
(3.3 ( 0.4) × 10
and (3.1 ( 0.6) × 10
cm molecule

Atmospheric Chemistry of Cyclohexane
J. Phys. Chem. A, Vol. 103, No. 15, 1999 2691
•
3
.4. Spectrum of (c-C6H11)O2 Radicals. Following the pulse
radiolysis of mixtures of 980 mbar SF6, 5 mbar c-C6H12, and
5 mbar O2, a rapid increase (complete within 0.5-2.0 µs) in
1
the UV absorbance was observed, followed by a slower decay.
An example is shown in the insert in Figure 1B. The observed
absorption does not decay to zero, indicating the formation of
a reaction product which absorbs at 250 nm. We ascribe the
initial increase in absorption to the formation of the (c-C6H11)-
•
O2 via reaction 2. To work under conditions where a known
•
fraction of the F atoms are converted into (c-C6H11)O2 radicals,
it is necessary to consider potential interfering secondary
chemistry. Potential complications include unwanted radical-
radical reactions such as reactions 3 and 8-10.
Figure 4. Pseudo-first-order rate constants for the formation of
•
(
(
c-C
6
H
11)O
2
radicals as a function of the O
2
concentration in 500
, diluent. The
opened circles) and 1000 (filled circles) mbar of SF
6
insert shows an experimental transient obtained using a mixture of
mbar c-C 12, 5.1 mbar O , and 990 mbar SF . The UV path
5
6
H
2
6
length was 80 cm and the radiolysis dose was 32% of maximum.
The smooth curve in the insert is the first-order fit (starting at
t ) 0.2 µs).
and loss of F atoms by reaction with molecular oxygen, reaction
11:
-
1 7,8
F + O + M f FO + M
(11)
s . The kinetics of reaction 3 have been studied previously
2
2
6
by Schuler and Patterson. For reasons which are unknown, the
value of k3 reported by Schuler and Patterson is a factor of 16
times less than that reported herein.
6
-10
cm molecule s-1 9 and using
3
-1
-1 -l 12
Assuming k ) 1 × 10
7
-
13
3
k ) (1.9 ( 0.3) × 10
l3
cm molecule s , we calculate
•
3.3. Association Reaction between O2 and c-C6H11 Radi-
that 0.6% of the F atoms are converted into FO2 radicals and
99.4% into (c-C6H11)O2 radicals. No corrections due to forma-
•
cals. The kinetics of reaction 2 were studied by monitoring the
absorbance at 280 nm following pulse radiolysis of mixtures
of 495-995 mbar of SF6, 5.0 mbar of c-C6H12, and 0.95-6.8
tion of FO2 radicals was necessary.
There are no literature data concerning the kinetics of
reactions 8-10, hence, we cannot calculate their importance.
To check for these unwanted radical-radical reactions, the
transient absorbance at 250 nm was measured using mixtures
of 980 mbar SF6, 5 mbar c-C6H12, and 15 mbar O2 with the
radiolysis dose varied by an order of magnitude. The UV path
length was 80 cm. Figure 1B shows the observed maximum
absorbance of the experimental transients at 250 nm as a
function of dose. As seen from Figure 1B, the absorption
observed in experiments using maximum dose (open circle) was
significantly less than that expected upon linear extrapolation
of the low dose data (filled circles). We ascribe this to
-
10
3
mbar of O2. Assuming k7 to be approximately 1 × 10
cm
-
1
-1 9
molecule s , it follows that in the presence of 5.0 mbar of
c-C6H12 the lifetime of F atoms with respect to conversion into
•
•
c-C6H11 radicals is 0.08 µs. Assuming that c-C6H11 radicals
behave like other alkyl radicals and add O2 with a rate constant
of the order 10 -10
c-C6H11 radicals with respect to conversion into (c-C6H11)O2
radicals will be of the order of a microsecond. Consistent with
expectations, the pulsed radiolysis of SF6/c-C6H12/O2 mixtures
produced absorption transients consisting of a initial rapid
increase in absorption (complete within 0.1 µs) followed by a
slower increase which occurred over a time scale of 1-4 µs.
We ascribe the initial rapid increase in absorption to the rapid
-
12
-11
3
-1 -1
cm molecule s , the lifetime of
•
•
•
incomplete conversion of F atoms into (c-C6H11)O2 radicals
caused by secondary radical-radical reactions 3 and 8-10 at
high radical concentrations. The solid line drawn through the
data in Figure 1B is a linear least-squares fit of the low dose
data. The slope is 0.62 ( 0.03. From this value and the yield
•
formation of c-C6H11 radicals and the subsequent “slow”
increase to their conversion into (c-C6H11)O2 radicals. The
•
“slow” increase in absorption was observed to follow first-order
1
5
-3
kinetics. The insert in Figure 4 shows a typical absorption
transient together with a first-order fit. As seen from Figure 4,
the pseudo-first-order rate constant k increased linearly with
O2 concentration. Experiments were performed in the presence
of either 500 or 1000 mbar total pressure using SF6 as diluent;
there was no discernible effect of total pressure suggesting that
reaction 2 is at, or near, the high-pressure limit in 500-1000
mbar of SF6. This behavior is consistent with the available
database for the addition of O2 to large (gC3) alkyl radi-
of F atoms of (3.18 ( 0.32) × 10 molecule cm (full dose
•
and [SF6] ) 1000 mbar), we derive σ((c-C6H11)O2 )250nm ) (5.72
first
-18
2
-1
( 0.64) × 10
cm molecule . The quoted uncertainty
includes both statistical and potential systematic uncertainties
and so reflects the accuracy of the measurement.
To map out the spectrum of the (c-C6H11)O2 radical, the
maximum transient absorbance between 220 and 300 nm
following the pulsed radiolysis of SF6/c-C6H12/O2 mixtures was
recorded using a diode array. The delay was 0.1 µs, the
integration time was 1 µs, and the spectral resolution was 1
nm. The initial absorbances were scaled to that at 250 nm and
•
9
cals. Linear least-squares analysis of the data in Figure 4
-
11
3
-1 -1
gives k2 ) (1.3 ( 0.2) × 10
cm molecule s . This re-
•
sult is in excellent agreement with that of k2 ) (1.4 ( 0.2)
placed on an absolute basis using σ((c-C6H11)O2 ) ) 5.72 ×
-
11
3
-1 -1
10
-18
2
-1
×
10
cm molecule
s
measured by Wu et al. in a
10
cm molecule . The result is shown in Figure 2B. The
•
previous gas phase study. In contrast, a value of (5.6 ( 1.0) ×
UV spectrum of (c-C6H11)O2 radicals reported by Rowley et
-
12
3
-1 -1
13
1
0
cm molecule
s
has been measured in liquid cyclo-
al. is also given in Figure 2B. For reasons which are unclear,
the spectrum measured in the present study is slightly (10%)
11
hexane by Bjellqvist and Reitberger.

2692 J. Phys. Chem. A, Vol. 103, No. 15, 1999
Platz et al.
•
C6H11)O2 radicals reacted via reaction 4a, which we ascribe to
a minor contribution from reaction channel 4b.
•
3
.6. Kinetics of the Reaction between (c-C6H11)O2 Radi-
cals and NO2. The kinetics of reaction 5 were studied by
monitoring the decrease in absorbance at 400 nm following
the pulsed radiolysis of mixtures of 10 mbar c-C6H12, 10
mbar O2, 0.34-1.07 mbar NO2, and SF6 added to 1000
mbar total pressure. The insert in Figure 5B shows the ab-
sorbance as a function of time after the radiolysis of a mixture
containing 0.64 mbar of NO2. The rate of decay of the ab-
sorbance at 400 nm increased linearly with the NO2 concentra-
tion. We ascribe the decrease in absorbance to loss of NO2 in
the system. Three reactions could be responsible for the NO2
loss:
However, the time scale of the decay shown in the insert (Figure
5
B) is only consistent with reaction 5 causing the loss of NO2
•
(the lifetimes of both F atoms and c-C6H11 radicals are <1 µs,
see section 3.3).
Figure 5. (A) Pseudo-first-order rate constants for formation of NO
following pulsed radiolysis of SF /cyclohexane/O /NO mixtures versus
NO]. The insert shows the transient absorbance at 400 nm observed
following pulsed radiolysis of a mixture of 0.40 mbar NO, 10 mbar
cyclohexane, 10 mbar O , and 980 mbar SF . (B) Pseudo-first-order
rate constants for decay of NO following pulsed radiolysis of SF
cyclohexane/O /NO mixtures versus [NO ]. The insert shows the
transient absorbance at 400 nm observed following pulsed radiolysis
of a mixture of 0.64 mbar NO , 10 mbar cyclohexane, 10 mbar O
and 980 mbar SF
2
The smooth curve in the insert (Figure 5B) is the first-order
first
6
2
fit which gives a pseudo-first-order rate constant, k ) 1.55
[
5
-1
×
10 s . To allow sufficient time for conversion of F atoms
into the alkyl peroxy radicals the analysis of the decay traces
2
6
first
was performed starting 2 µs after the radiolysis pulse. k values
2
6
/
for eight experiments with different concentrations of NO2 are
shown in Figure 5B. Linear least-squares analysis of the data
2
2
2
-
12
3
-1
2
2
,
in Figure 5B gives k5 ) (9.5 ( 1.5) × 10
cm molecule
6
.
s . The observed loss of NO2 was 88 ( 10% of the initial
-1
•
concentration of (c-C6H11)O2 radicals.
less intense and is shifted to the red by approximately 10 nm
3
.7. IR Absorption Spectrum of (c-C6H11)O2NO2. Prior to
when compared to that of Rowley et al.13
our study of the atmospheric oxidation mechanism of cyclo-
hexane in the presence of NOx, the IR spectrum of the per-
oxynitrate, (c-C6H11)O2NO2, was recorded using the FTIR-
smog chamber system. (c-C6H11)O2NO2 was prepared by the
UV irradiation of a mixture of 34.7 mTorr c-C6H12, 9.21
mTorr NO2, and 167 mTorr Cl2 in 700 Torr total pressure of
O2 diluent. Twenty seconds of UV irradiation resulted in
the consumption of 16% of the cyclohexane and the forma-
•
3
.5. Kinetics of the Reaction between (c-C6H11)O2 Radi-
cals and NO. The kinetics of reaction 4 were studied by
monitoring the increase in absorption at 400 nm, ascribed to
the formation of NO2, following the pulse radiolysis of c-C6H12/
O2/NO/SF6 mixtures. The initial conditions were 10 mbar
c-C6H12, 10 mbar O2, 0.34-1.16 mbar NO, in 1000 mbar of
SF6 diluent. The insert in Figure 5A shows a typical absorption
transient observed following the pulsed radiolysis of a mixture
containing 0.40 mbar of NO.
tion of product features at 791, 919, 1042, 1295, 1457, and
-1
1
715 cm . Figure 6 shows the product spectrum obtained
after subtraction of features attributable to cyclohexane which
we assign to (c-C6H11)O2NO2 formed in the following re-
actions:
The absorption transients were fitted using a first-order kinetic
expression, and the resulting pseudo-first-order rate constants
are plotted versus the initial NO concentration in Figure 5A.
This method of measuring the kinetics of the reactions of alkyl
peroxy radicals with NO has been used previously in our
laboratory and is discussed in detail elsewhere.1 As seen from
4
first
Figure 5A, the pseudo-first-order rate constants k increased
linearly with the NO concentration. Linear least-squares analysis
The y-axis scale in Figure 6 was placed upon an absolute
basis by equating the observed loss of cyclohexane to formation
of peroxynitrate. NO2 absorbs strongly in the region 1500-
-
12
3
of the data in Figure 5A gives k4 ) (6.7 ( 0.9) × 10
cm
-
1
-1
molecule s . The NO2 yield calculated from the observed
-
1
maximum absorbance was 86 ( 10% of that expected if all F
1660 cm , and we are unable to provide spectral information
•
-1
atoms were converted into (c-C6H11)O2 radicals and all the (c-
for (c-C6H11)O2NO2 in this region. The IR feature at 1295 cm

Atmospheric Chemistry of Cyclohexane
J. Phys. Chem. A, Vol. 103, No. 15, 1999 2693
6 2 2
Figure 6. IR spectrum of peroxynitrate (c-C H11)O NO .
is assigned to the NO2 symmetric stretching mode which is a
characteristic feature in the IR spectra of organic nitrates. The
-
17
integrated absorption cross section of this band is 2.48 × 10
-
1
cm molecule and is comparable to those reported for the
-
17 15
analogous features in CH3ONO2 (2.90 × 10
), C2H5ONO2
-
17 15
-17 15
(
2.97 × 10
), and CH3C(O)O2NO2 (2.58 × 10
) pro-
viding confidence that the spectrum shown in Figure 6 is indeed
that of (c-C6H11)O2NO2.
•
3
.8. Nitrate Yield in the (c-C6H11)O2 + NO Reaction and
Rate Constant Ratio k17/k16. To determine the branching
ratio k4b/(k4a + k4b) and the rate constant ratio k17/k16, the
FTIR-smog chamber system was used to study the products
resulting from the Cl atom initiated oxidation of cyclohexane
in the presence of NO. Experiments were conducted using
c-C6H12/NO/Cl2/O2/N2 mixtures in 700 Torr total pressure at
Figure 7. IR spectra acquired before (A) and after (B) a 90 s irradi-
ation of a mixture of 19.6 mTorr c-C
NO, and 595 Torr O in 700 Torr total pressure of N
During the irradiation 47% of the c-C 12 was consumed. Part C
shows a spectrum of cyclohexanone. Subtraction of IR features
attributable to c-C 12, cyclohexanone, HONO, HNO , and (c-C
NO from part B gives part D. The prominent feature at 1281
6
H12, 90 mTorr Cl
2
, 46 mTorr
2
2
diluent.
2
96 ( 2 K.
6
H
6
H
3
6
H11)-
O
2
2
-
1
cm is a typical IR feature of an organic nitrate (see text for
details).
gous to reaction 4b involving the peroxy radical derived from
the alkyl radical generated in reaction 17. For the experiment
shown in Figure 7 ([O2] ) 595 Torr), the cyclohexanone yield
is 70%, the dominant fate of cyclohexoxy radicals is reaction
with O2 and complications caused by nitrates formed after
decomposition of cyclohexoxy radicals should be of minor
importance. The c-C6H11ONO2 yield was estimated using an
integrated absorption cross section for the IR band centered at
Cyclohexane loss and product formation were monitored
using FTIR spectroscopy. Figure 7 shows IR spectra acquired
before (A) and after (B) a 90 s irradiation of a mixture of 19.6
mTorr c-C6H12, 90 mTorr Cl2, 46 mTorr NO, and 595 Torr O2
in 700 Torr total pressure with N2 diluent. Part C shows a
reference spectrum of cyclohexanone. Comparison of part B
with part C shows that cyclohexanone is a significant product.
-
1
-17
-1
1281 cm of 2.7 × 10
cm molecule (average of values
listed in previous section for organic nitrates). Figure 8 shows
c-C6H11ONO2 formation versus c-C6H12 loss (corrected for the
formation of a small amount of c-C6H11O2NO2 via reaction 5)
from experiments conducted using [O2] ) 595 Torr. Linear
least-squares regression of the data in Figure 8 gives k4b/(k4a +
k4b) ) 0.197 ( 0.010. Incorporating an estimated uncertainty
-
1
The infrared features at 1121 and 1221 cm were used to
quantify the cyclohexanone yield. Experiments were performed
using [O2] ) 50-595 Torr. The cyclohexanone yield increased
with increasing [O2] reflecting a competition between reactions
-
17
of 15% in the integrated cross section of 2.7 × 10
gives
1
6 and 17 for the available cyclohexoxy radicals produced in
k4b/(k4a + k4b) ) 0.197 ( 0.031. It should be noted that this
approach overestimates the branching ratio because of the
possible contribution of nitrates formed after decomposition of
the cyclohexoxy radicals via reaction 17. This point is addressed
at the end of this section.
reaction 4a. Subtraction of features attributable to c-C6H12,
cyclohexanone, HONO, HNO3, and c-C6H11O2NO2 from part
B gives the residual spectrum shown in part D which has a
-
1
prominent feature at 1281 cm characteristic of an organic
-
1
nitrate. Several nitrates could be responsible for the 1281 cm
Reactions 16 and 17 compete for the cyclohexoxy radi-
cals produced in reaction 4a. If we define R as the yield of
cyclohexoxy radicals from reaction 4, R ) k4a/(k4a + k4b),
feature in part D; the possibilities include: c-C6H11ONO2 from
reaction 4b and HC(O)(CH2)4CH2ONO2 from a reaction analo-

2694 J. Phys. Chem. A, Vol. 103, No. 15, 1999
Platz et al.
Figure 8. Yield of (c-C
c-C /NO/O /N mixtures as a function of the c-C
corrected for formation of (c-C NO ) at constant total pressure
700 or 750 Torr) and 296 ( 2K. The fit through the data is a linear
11)ONO yield of 19.7 (
6
H
11)ONO
2
following the UV irradiation
Figure 9. Reciprocal yield of cyclohexanone following the UV
6
H12/Cl
2
2
2
6
H12 loss
irradiation of c-C H /Cl /NO/O /N mixtures as a function of the
6
12
2
2
2
(
(
6
H11)O
2
2
reciprocal oxygen concentration at a total pressure of 700 or 750 Torr
and 296 ( 2K. Closed circles are data obtained in this work; the open
circle is taken from Aschmann et al.18 The open triangle and square
are the limiting cases calculated using a (c-C H )ONO yield of 0.16
least squares regression which gives a (c-C
.1%.
6
H
2
0
6
11
2
(
this work) and 0.156 (Aschmann et al.18 (see text for details).
then the molar yield of cyclohexanone Y(c-C6H10O) can be
expressed as
quoted error includes uncertainties associated with the correction
procedure. This result is in good agreement with the value of
k [O ]
18
16
2
k4b/(k4a + k4b) ) 0.156 ( 0.017 reported by Aschmann et al.
Y(c-C H O) ) R
6
10
((
)
but somewhat higher than that of k4b/(k4a + k4b) ) 0.09 ( 0.04
k [O ] + k )
16
2
17
2
0
reported by Tagagi et al. The open triangle and square in
Figure 9 are the limiting cases calculated using values of k4b/
and
1
8
(
k4a + k4b) ) 0.16 (derived above) and 0.156 and are consistent
k17
with expectations based upon extrapolation of cyclohexanone
1
1
1
1
R
)
+
yields observed using experiments with [O2] ) 50-592 Torr.
Y(c-C H O) R k [O ]
6
10
16
2
4
. Discussion
where k16 and k17 are rate constants for reactions 16 and 17.
The yield of cyclohexanone was measured in experiments
conducted using the UV irradiation of c-C6H12/NO/Cl2/O2/N2
mixtures with [O2] varied between 50 and 595 Torr. For each
experiment the cyclohexanone yield was obtained from a linear
least squares regression of a plot of cyclohexanone formation
versus c-C6H12 consumption. The rate constant for the reaction
The results presented here improve our understanding of the
atmospheric degradation mechanism of cyclohexane. Oxidation
is initiated via OH radical attack which occurs with a rate
-
12
3
-1 -1
21
constant of 7.4 × 10
cm molecule
s
at 298 K. While
the concentration of OH radicals in the atmosphere varies
considerably both spatially and temporally, a reasonable estimate
-
10
3
6
between Cl atoms and cyclohexanone is 3.4 × 10
cm
of the 24 h average OH radical concentration is 0.5-1.0 × 10
-
1
-1 16
-3
molecule s . Small corrections (4-25%) were applied to
account for loss of cyclohexanone via reaction with Cl atoms
using methods described previously.17
cm . Hence, the atmospheric lifetime of cyclohexane lies in
the range 1-3 days. Reaction of OH radicals with cyclohexane
gives cyclohexyl radicals, c-C6H11 , which add O2 with a rate
•
-
11
3
-1 -1
Figure 9 shows a plot of 1/Y(c-C6H10O) versus 1/[O2]. The
filled circles in Figure 9 are results from the present work, the
open circle is the cyclohexanone yield observed during cyclo-
constant k2 ) (1.3 ( 0.2) × 10
cm molecule s . In 1
•
atm of air at 298 K, the lifetime of c-C6H11 radicals with respect
•
to addition of O2 is 15 ns. The sole atmospheric fate of c-C6H11
is addition of O2 to give c-C6H11O2 radicals.
1
8
•
hexane oxidation in air reported by Atkinson et al. The fit
shown in Figure 9 is a linear regression to both data sets giving
an intercept of 1/R ) 1.11 ( 0.15 and a slope of (1/R)x(k17/
k16) ) 279 ( 20 Torr; hence, k17/k16 ) 251 ( 47 Torr. Quoted
errors in k17/k16 include uncertainties in both the slope and
intercept. There are no literature data to compare with this result.
At this point, it is appropriate to return to the estimation of
k4b/(k4a + k4b) ) 0.197 ( 0.043 above and consider the likely
impact of the formation of nitrates formed after decomposition
of the cyclohexoxy radicals. Using k17/k16 ) 251 Torr it follows
that, in the presence of 595 Torr of O2, 30% of the cyclohexoxy
radicals will decompose via reaction 17 while the remaining
•
We have shown here that c-C6H11O2 radicals react with NO
giving a (16 ( 4)% yield of the nitrate (c-C6H11)ONO2 with
the balance of reaction producing c-C6H11O radicals and NO2.
-
12
3
-1 -1
Using k4 ) (6.7 ( 0.9) × 10
cm molecule s , together
with an estimated background tropospheric NO concentration
8
-3 22
, gives a lifetime of c-C6H11O2•
of 2.5 × 10 molecule cm
•
radicals with respect to reaction with NO of 10 min. c-C6H11O2
radicals also react rapidly with NO2, but the peroxynitrate
product is thermally unstable and decomposes to regenerate
•
19
c-C6H11O2 radicals and NO2 and therefore is not an important
atmospheric oxidation product. Reaction with O2 and ring
opening via C-C bond scission are two competing atmospheric
fates of the c-C6H11O radical produced in the c-C6H11O2 +
NO reaction. In 700-750 Torr total pressure at 296 ( 2K the
7
0% will react with O2. On the basis of the literature database
19
•
•
for hexyl and pentyl peroxy radicals, it is likely that 10-20%
of the peroxy radicals formed following reaction 17 will react
with NO to give nitrates. Correcting for the formation of such
nitrates we arrive at k4b/(k4a + k4b) ) 0.16 ( 0.04, where the
18
rate constant ratio kdecomp/kO2 ) (8.1 ( 1.5) × 10 molecule
-
3
cm . At 296 K in 1 atm of air, 61% of cyclohexoxy radicals

Atmospheric Chemistry of Cyclohexane
J. Phys. Chem. A, Vol. 103, No. 15, 1999 2695
decompose and 39% react with O2 to give cyclohexanone.
Decomposition produces an alkyl radical which adds O2 to give
the peroxy radical HC(O)(CH2)5O2. Aschmann et al. have
shown that the reaction of this peroxy radical with NO leads to
a variety of different products including the nitrate HC(O)-
Huie, R. E.; Kurylo, M. J. J. Phys. Chem. 1996, 19, 3514.
(
67.
5) Wallington, T. J.; Dagaut, P.; Kurylo, M. J. Chem. ReV. 1992, 92,
6
1
8
(6) Schuler, R H.; Patterson, L. K. Chem. Phys. Lett. 1974, 27, 369.
(7) Platz, J.; Sehested, J.; Møgelberg, T. E.; Nielsen, O. J.; Wallington,
T. J. J. Chem. Soc., Faraday Trans. 1997, 93, 2855.
(
CH2)5ONO2, the dialdehyde HC(O)(CH2)4C(O)H, and the
(8) Platz, J.; Christensen, L. K.; Sehested, J.; Nielsen, O. J.; Wallington,
T. J.; Sauer, C.; Barnes, I.; Becker, K. H.; Vogt, R. J. Phys. Chem. A 1998,
02, 4829.
hydroxydialdehyde HC(O)CH2CH(OH)CH2CH2C(O)H. At the
present time the absolute yields of these compounds are not
1
(9) Mallard, W. G.; Westley, F.; Herron, J. T.; Hampson, R. F. NIST
1
8
known. The rate constant ratio kdecomp/kO2 ) (8.1 ( 1.5) × 10
Chemical Kinetics Database, Version 6.0; NIST Standard Reference Data,
Gaithersburg, MD, 1994.
(10) Wu, D.; Bayes, K. D. Int. J. Chem. Kinet. 1986, 18, 547.
11) Bjellqvist, B.; Reitberger, T.; K. Tek. Hoegsk. Nucl. Sci. Abstr.
974, 30, 120.
12) Ellermann, T.; Sehested, J.; Nielsen, O. J.; Pagsberg, P.; Wallington,
-
3
molecule cm was measured at T ) 296 ( 2K. At the lower
temperatures characteristic of the upper troposphere the rates
of both reactions 16 and 17 will decrease substantially. Reaction
(
1
1
7 is a unimolecular decomposition and it is expected that k17
(
will be more sensitive than k16 to temperature and that the rate
constant ratio kdecomp/kO2 (k17/k16) will decrease with temperature.
Hence, the yield of cyclohexanone during the atmospheric
oxidation of cyclohexane is expected to increase significantly
with altitude. A study of the temperature dependence of kdecomp/
kO2 is beyond the scope of the present work but is clearly needed
to better define the atmospheric chemistry of cyclohexane.
At this point it is germane to note that it has been reported
recently that chemical activation can play an important role in
the atmospheric fate of alkoxy radicals formed via reaction of
T. J. Chem. Phys. Lett. 1994, 218, 287.
(13) Rowley, D. M.; Lightfoot, P. D.; Lesclaux, R.; Wallington, T. J. J.
Chem. Soc., Faraday Trans. 1991, 87, 3221.
(
14) Sehested, J.; Nielsen, O. J.; Wallington, T. J. Chem. Phys. Lett.
1
993, 213, 257.
(
(
15) Tuazon, E. C.; Atkinson, R., Int. J. Chem. Kinet. 1990, 22, 1221.
16) Wallington, T. J.; Guschin, A.; Hurley, M. D. Int. J. Chem. Kinet.
1998, 30, 309.
(17) Meagher, R. J.; Mcintosh, M. E.; Hurley, M. D.; Wallington, T. J.
Int. J. Chem. Kinet. 1997, 29, 619.
(
18) Aschmann, S. A.; Chew, A. A.; Arey, J.; Atkinson, R. J. Phys.
Chem. A 1997, 101, 8042.
19) Lightfoot, P. D.; Cox, R. A.; Crowley, J. N.; Destriau, M.; Hayman,
G. D.; Jenkin, M. E.; Moortgat, G. K.; Zabel, F. Atmos. EnViron. 1992, 26,
805.
(20) Takagi, H.; Washida, N.; Bandow, H.; Akimoto, H.; Okuda, M. J.
Phys. Chem. 1981, 85, 2701.
23-26
peroxy radicals with NO.
In systems where chemical
(
activation is important, plots of carbonyl yield versus [O2] (e.g.,
1
Figure 9) display y-axis intercepts which are substantially in
2
3,26
excess of unity.
There is no evidence from the present work
•
that chemical activation plays any role in the fate of c-C6H11O
radicals.
(21) Atkinson, R.; Aschmann, S. M. Int. J. Chem. Kinet. 1989, 21, 355.
(22) Atkinson, R. Scientific assessment of stratospheric ozone. World
Meteorological Organization Global Ozone Research and Monitoring
Project, Report No. 20; World Meteorological Organization: Geneva, 1989;
Vol. 2, p 167.
Acknowledgment. The work at Risø was supported by Ford
Forschungzentrum, Aachen.
(23) Wallington, T. J.; Hurley, M. D.; Fracheboud, J. M.; Orlando, J.
J.; Tyndall, G. S.; Sehested, J.; Møgelberg, T. E.; Nielsen, O. J. J. Phys.
Chem. 1996, 100, 18116.
References and Notes
(
24) Møgelberg, T. E.; Sehested, J.; Tyndall, G. S.; Orlando, J. J.;
(
1) Siegl, W. O. 1998. Private communication.
Fracheboud, J. M.; Wallington, T. J. J. Phys. Chem. 1997, 101, 2828.
(25) Bilde, M.; Wallington, T. J.; Ferronato, C.; Orlando, J. J.; Tyndall,
G. S.; Estupi n˜ an, E.; Haberkorn, S. J. Phys. Chem. 1998, 102, 1976.
(26) Orlando, J. J.; Tyndall, G. S.; Bilde, M.; Ferronato, C.; Wallington,
(
2) Hansen, K. B.; Wilbrandt, R.; Pagsberg, P. ReV. Sci. Instrum. 1979,
5
0, 1532.
(
(
3) Wallington, T. J.; Japar, S. M. J. Atmos. Chem. 1989, 9, 399.
4) Bilde, M.; Møgelberg, T. E.; Sehested, J.; Nielsen, O. J.; Wallington,
T. J.; Hurley, M. D.; Japar, S. M.; Dill, M.; Orkin, V. L.; Buckley, T. J.;
T. J.; Vereecken, L.; Peeters. J. J. Phys. Chem. A 1998, 102, 8116.