Infrared matrix isolation and theoretical study of the initial intermediates in the reaction of ozone with cis-2-Butene

Article abstract of DOI:10.1021/jp912253t

Matrix isolation studies combined with infrared spectroscopy of the twin jet codeposition of ozone and cis-2-butene into argon matrices have led to the first observation of several early intermediates in this ozonolysis reaction. Specifically, evidence is presented for the formation and identification of the long sought-after Criegee intermediate, as well as confirming evidence for earlier reports of the primary and secondary ozonides. These species were observed after initial twin jet deposition, and grew upon annealing to 35 K. Extensive isotopic labeling (¹⁸O and ^16,48O mixtures) experiments provided important supporting data. Detailed theoretical calculations at the B3LYP/6-311++G(d,2p) level were carried out as well to augment the experimental work, Merged jet (flow reactor) experiments followed by cryogenic trapping in solid argon led to the formation of "late", stable oxidation products. Photochemical, reactions of ozone with cis-2-butene was studied as well, as was the photochemical behavior of the primary and secondary ozonides.

Full text of DOI:10.1021/jp912253t

J. Phys. Chem. A 2010, 114, 2799–2805
2799
Infrared Matrix Isolation and Theoretical Study of the Initial Intermediates in the Reaction
of Ozone with cis-2-Butene
Mary Clay and Bruce S. Ault*
Department of Chemistryl, UniVersity of Cincinnati, P.O. Box 210172, Cincinnati, Ohio 45221-0172
ReceiVed: December 30, 2009; ReVised Manuscript ReceiVed: January 21, 2010
Matrix isolation studies combined with infrared spectroscopy of the twin jet codeposition of ozone and cis-
2-butene into argon matrices have led to the first observation of several early intermediates in this ozonolysis
reaction. Specifically, evidence is presented for the formation and identification of the long sought-after Criegee
intermediate, as well as confirming evidence for earlier reports of the primary and secondary ozonides. These
species were observed after initial twin jet deposition, and grew upon annealing to 35 K. Extensive isotopic
labeling (18O and 16,18O mixtures) experiments provided important supporting data. Detailed theoretical
calculations at the B3LYP/6-311++G(d,2p) level were carried out as well to augment the experimental work.
Merged jet (flow reactor) experiments followed by cryogenic trapping in solid argon led to the formation of
“late”, stable oxidation products. Photochemical reactions of ozone with cis-2-butene was studied as well, as
was the photochemical behavior of the primary and secondary ozonides.
SCHEME 1
Introduction
An important goal of atmospheric chemistry is to understand
the sources of different trace gases as well as their removal
mechanisms, that is, their atmospheric budget. Alkenes from
both biogenic and anthropogenic sources constitute an important
class of volatile organics in the troposphere, and reaction with
ozone is a primary removal mechanism.1-6 The importance of
the reaction of O3 with alkenes in the troposphere has led to
many experimental and theoretical studies of the kinetics and
mechanism of these reactions.7-14 Important early work was
carried out by Criegee, leading to the first proposal of a
mechanism for the reaction.15,16 The Criegee mechanism is
supported by extensive indirect experimental evidence as well
as high level theoretical calculations and is generally now
accepted. The mechanism, shown in Scheme 1, involves the
initial formation of a primary ozonide by a 1,3 polar addition
across the double bond of the olefin, forming a 1,2,3-trioxolane
species 1. This reaction is very exothermic, leading to further
reaction and decomposition into an aldehyde and the proposed
Criegee intermediate 2, a carbonyl oxide. The Criegee inter-
mediate and the aldehyde may then recombine to form a
secondary ozonide (1,2,4-trioxolane) 3, or react further to form
a range of products.
to isolate many novel and important short-lived species. Several
groups have applied matrix isolation in an attempt to characterize
the Criegee intermediate without success.19 All of these have
employed twin or dual jet deposition approaches. Recently,
novel deposition techniques, including merged jet and variable
length concentric jet approaches have been developed in our
laboratories and may provide potential new routes to the
formation and stabilization of the initial intermediates in these
important reactions. A study of the ozonolysis of cis-2-butene
is reported here.
A number of studies, employing a range of experimental
techniques, have sought to form and characterize each of the
initial intermediates in the reaction of alkenes with O3 and have
had limited success to date. Although the primary ozonide has
been tentatively identified for a few systems, and the secondary
ozonide in more studies, the Criegee intermediate has not been
observed experimentally. However, theoretical calculations
support the proposed mechanism and the existence of this
intermediate.
Experimental Section
All of the experiments in this study were carried out on a
conventional matrix isolation apparatus that has been de-
scribed.20 cis-2-Butene was introduced into the vacuum system
from a lecture bottle and was purified by freeze-pump-thaw
cycles at 77 K. Ozone was produced by Tesla coil discharge of
O2 (Wright Brothers) and trapped at 77 K to remove residual
O2 and trace gases. Isotopically labeled 18O3 was produced in
the same manner from 18O labeled O2 (94%, Cambridge Isotope
Laboratories). 2-Butanone was introduced as the vapor above
the room temperature liquid after purification by freeze-thaw
cycles at 77 K, and then diluted in argon. Argon (Wright
Brothers) was used as the matrix gas without further purification.
The matrix isolation technique17,18 was developed and has
been employed for the isolation and characterization of a wide
range of reactive intermediates, including radicals, ions, and
molecular complexes. This approach has been used successfully
* To whom correspondence should be addressed. E-mail: bruce.ault@
uc.edu.
10.1021/jp912253t  2010 American Chemical Society
Published on Web 02/08/2010

2800 J. Phys. Chem. A, Vol. 114, No. 8, 2010
Clay and Ault
TABLE 1: Product Band Positions and Assignments for Early Intermediates in the Ozonolysis of cis-2-Butene
experimental positiona and shifts
16O 16,18O
720
calculated position and shifts
16,18O
∆
∆18O
16O
∆
∆18O
assignment
-10
680
894
962
1044
1126
1139
1413
1487
1729d
-6, -6, -9, -11, -19, -26
-5, -6, -9, -11, -13, -16
-4, -14
0, -1, -1 -2, -2, -3
-1, -2, -3, -11, -12, -13
0, -2, -2, -5, -5, -6
0, 0, 0, 0, 0, 0
-31
-21
-19
-4
primary ozonide
secondary ozonidec
Criegee intermediate
primary ozonide
secondary ozonide
secondary ozonide
secondary ozonide
secondary ozonide
acetaldehyde
b
890, 895
922
-18
-18
-7
-14
-6
0
-4, -13
-4
-2, -13
1066
1116
1144
1394
1471
1720
-15
-7
b
0
0
-0
0
-30
0, 0, 0, 0, 0, 0
-0
-29d
a Band positions in cm-1 b Broad peak without resolved maxima. c From ref 30. d Experimental band positions for unperturbed (isolated)
.
acetaldehyde, from ref 26 and experiments run in this lab.
Matrix samples were deposited in 3 different modes: twin
jet, merged jet, and a new variable length concentric jet mode.
In the first, the two gas samples were deposited from separate
nozzles onto the 14 K window, allowing for only a brief mixing
time prior to matrix deposition. The distance from the tip of
each nozzle to the cold window was approximately 3 cm.
Several of these matrices were subsequently warmed to 33-35
K to permit limited diffusion and then recooled to 14 K and
additional spectra were then recorded. In addition, many of these
matrices were irradiated for 1.0 or more hours with either the
H2O/Pyrex-filtered or the H2O/quartz-filtered output of a 200
W medium-pressure Hg arc lamp, after which additional spectra
were recorded.
A number of experiments were conducted in the merged jet
mode,21 in which the two deposition lines were joined with an
UltraTorr tee at a distance from the cryogenic surface, and the
flowing gas samples were permitted to mix and react during
passage through the merged region. To more effectively model
the gas phase, the merged region was constructed of Teflon FEP,
either 1/4 or 1/8 in. O.D. in a number of experiments. The length
of the merged region (or reaction zone) was varied from as short
as 5 cm to as long as 2.0 m.
Figure 1. Infrared spectrum of a newly deposited matrix formed by
the twin jet codeposition of a sample of Ar/O3 ) 200 with a sample of
Ar/cis-C4H8 ) 250 (upper, blue trace) compared to blank spectra of
Ar/O3 (green trace) and Ar/cis-C4H8 (bottom, red trace).
B3LYP functional were used to locate energy minima, determine
structures, and calculate vibrational spectra. Final calculations
with full geometry optimization employed the 6-311 ++g(d,
2p) basis set, after initial calculations with smaller basis sets
were run to approximately locate energy minima. Thermody-
namic functions for the reactants and potential intermediates
were also calculated.
Twin jet and merged jet deposition probe different time scales
for reaction: very short for twin jet and somewhat longer for
merged jet. As will be reported below, quite different results
were obtained with these two modes. To probe the intermediate
time scale between twin and merged jet, a new concentric jet
device was developed. In this approach, a 1/8 in. O.D. Teflon
FEP tube was inserted inside of a larger, 1/4 in. O.D. tube, also
Teflon FEP. The length of the 1/8 in. tube could be adjusted to
be shorter, longer, or the same as the outer tube. The distance
between the outlet ends of the two tubes is referred to as ∆d )
(position of inner tube - position of outer tube). d > 0 indicates
that the inner tube extends beyond the outer tube, and d < 0
indicates that the inner tube is shorter than the outer tube (more
like merged jet). d ) 0 indicates that the ends of the two tubes
are at the same distance from the cold window. Mixing of the
two samples begins at the outlet of the inner tube and continues
until deposition onto the cold window (typically 2-3 cm). In
this manner, the time scale available for reaction could be
adjusted from nearly that of merged jet to nearly that of twin
jet. In all three deposition modes, matrices were deposited at
the rate of 2 mmol/h from each sample manifold onto the cold
window. Final spectra were recorded on a Perkin-Elmer
Spectrum One Fourier transform infrared spectrometer at 1 cm-1
resolution.
Results
Prior to any codeposition experiments, blank experiments
were run on each of the reagents used in this study. These blanks
served as an authentic spectrum of the pure starting materials,
allowing for clear identification of product bands as well as an
internal standard for the extent of reaction. In each case, the
blanks were in good agreement with literature spectra23,24 and
with blanks run previously in this laboratory. Each blank
experiment was then irradiated by the H2O/Pyrex-filtered output
of a 200 W Hg arc lamp for 1.0 h and no changes were noted.
cis-2-Butene + O3, Merged Jet. Merged jet experiments
were conducted with samples of Ar/cis-2-butene and Ar/O3 with
merged regions (reaction zone) 5, 20, and 200 cm in length. In
an initial experiment with Ar/cis-2-butene ) 250 and Ar/O3 )
250 and a 5 cm reaction zone, a number of weak bands were
observed. These are listed in Table 1, shown in Figure 1, and
are hereafter called set A. Bands at 720 and 1066 cm-1 were
the most intense, with absorbances of 0.08 and 0.07 O.D.,
respectively. The remaining bands were less intense but were
nonetheless distinct. This experiment was repeated twice, and
similar results were obtained in each case. The same bands were
observed and no additional bands were noted.
Theoretical calculations were carried out on likely intermedi-
ates in this study, using Gaussian 03 and 03W suite of
programs.22 Density functional calculations using the hybrid
A number of experiments were conducting using a 20 cm
1/8 in. O.D. reaction zone, with similar sample concentrations.
In an initial experiment, the set A bands were observed although

Reaction of Ozone with cis-2-Butene
J. Phys. Chem. A, Vol. 114, No. 8, 2010 2801
TABLE 2: Bands and Assignments of Products Formed in
the 2.0 m Merged Jet Reaction of O3 with cis-2-Butene
band position
16O
I
assignment
507
636
969
w
w
vw
m
CH3CHO
CH3COOH or HCOOH
CH3CHO
1112
1127
1168
1184
1246
1349
1381
1427
1499
1709
1729
1742
1758
1765
1778
2142
2149
2718
2736
2799
2817
2840
2962
2980
3465
3497
3532
3538
3554
m
w
w
vw
m
vw
m
m
vw
m
s
w
s
m
vw
vw
vw
w
H2CO
H2CO
CH3CHO
CH3COOH?
CH3CHO
H2CO
Figure 2. Infrared spectrum of a matrix formed by merged jet
deposition of a sample of Ar/O3 ) 200 with a sample of Ar/cis-C4H8
) 250 (upper, blue trace) compared to a blank spectrum of Ar/cis-
C4H8 (bottom, red trace). Note the growth of bands between 1700 and
1800 cm-1 and the reduction of bands due to parent C4H8 between
1350 and 1450 cm-1
.
CH3CHO
H2CO
HCOOH?
CH3COOH?
with somewhat reduced intensity. In addition, a distinct new
somewhat broad band appeared at 1729 cm-1 with and intensity
of 0.04 O.D. as well as a band at 1426 cm-1 on the high energy
side of a parent band of cis-2-butene at 1420 cm-1. In addition,
numerous very weak bands (<0.008 O.D.) were seen in different
regions of the spectrum in these experiments. In subsequent
experiments with different sample concentrations, quite similar
results were obtained. The same main bands were observed (i.e.,
the set A bands from the 5 cm experiments, above, along with
the 1426 and 1729 cm-1 bands). Very weak bands were noted
as well. However, their low intensity made accurate determi-
nation of their position and reproducibility difficult. In all of
the above merged jet experiments, the parent band intensities
were comparable to the blank experiments, suggesting that only
a small percentage of the parent species were reacting. This is
consistent with the low product band intensities.
When the merged region or reaction zone was lengthened
substantially, to 2.0 m of 1/4 in. tubing, the resulting spectra
were quite different. Parent band intensities were greatly
reduced, indicating that substantial reaction had occurred as
shown in Figure 2. In addition, the bands seen in the 5 cm
experiments (set A) were absent. The 1426 and 1729 cm-1 bands
seen in the 20 cm experiments were much more intense and a
large number of additional product bands were noted. These
included bands at 663 and 2344 cm-1 due25 to CO2 [CO2 is
always present to some degree in experiments using O3,
including blanks, due to the production process of O3. However,
in these 2.0 m merged jet experiments, the amount of CO2 was
much greater than in the blanks]. In subsequent experiments,
the ratio of O3 to cis-2-butene was varied from the initial 1:1 to
2:1 and 1:2. In these experiments, bands due to the less
concentrated reactant were completely destroyed while bands
of the more concentrated reactant were still present. The main
product bands were seen in all of these experiments, with high
intensities. Finally, a merged jet experiment was conducted with
the 2.0 m reaction zone and sample concentrations of 1000/1
for each of the reactants. In this more dilute experiment, product
bands intensities were still high and all of the product bands
were sharpened relative to the more concentrated experiments.
Product band positions from this experiments are listed in
Table 2.
CO
H2CO
CH3CHO
H2CO
m
vw
m-w
vw
vw
vw
vw
w
H2CO
w
w
HCOOH?
to 1701 cm-1, while the 1116 cm-1 band shifted to 1102 cm-1
and the 1066 cm-1 band shifted to 1059 cm-1. All of the 18O
band shifts for the set A bands are listed in Table 1.
cis-2-Butene + O3, twin jet. Several twin jet experiments
were run with samples of Ar/cis-2-butene ) 100 and Ar/O3 )
250. In an initial experiment, the same set A bands were
observed that had been seen in the 5 cm merged jet experiment,
at 720, 890, 896, 922, 1066, 1116, 1144, 1394, and 1471 cm-1
.
These are shown in Figure 1. Their intensities were somewhat
lower than in the short merged jet experiments. When this
sample was then annealed to 35 K and then recooled to 16 K
and an additional spectrum recorded, this set of bands grew
substantially, by approximately a factor of 5, as shown in Figure
3. In addition, a weak broad product band began to grow in at
1720 cm-1. No other new product bands were observed. This
experiment was repeated using similar sample concentrations,
and nearly identical results were obtained.
Similar experiments were run using twin jet deposition, and
samples of samples of Ar/cis-2-butene ) 100 and Ar/18O3 )
250. 18O counterparts of most of the set A bands were seen, at
875, 904, 1059, 1102, 1138, 1394, and 1471 cm-1. These are
shown in Figure 4.
An 18O counterpart could not be detected for the 720 cm-1
band due to interference by a nearby parent band of cis-2-butene.
Upon annealing, these bands grew substantially, just as their
16O counterparts had done. Also, a weak broadband became
apparent at approximately 1690 cm-1 as shown in Figure S1 of
the Supporting Information. Finally, multiple experiments were
run in which samples of Ar/cis-2-butene ) 250 were codeposited
with samples of Ar/16,18O3 ) 250, with scrambled isotope
Merged jet experiments were also conducted using the 20
cm reaction zone, with samples of Ar/cis-2-butene and Ar/18O3,
using 94% enriched 18O. In most cases, 18O analogs were
identified for the product bands seen in the normal isotopic
experiments. For example, the product band at 1729 cm-1 shifted

2802 J. Phys. Chem. A, Vol. 114, No. 8, 2010
Clay and Ault
Figure 5. Infrared spectrum from 902 to 930 cm-1 arising from the
twin jet codeposition of a sample of Ar/cis-2-butene ) 250 with a
sample of Ar/16,18O3 ) 250 (50% 18O, scrambled), after annealing. The
feature denoted with an asterisk (*) at 916 cm-1 was apparent in the
background scan before deposition and is not due to a product.
Figure 3. Infrared spectrum of a newly deposited matrix formed by
the twin jet codeposition of a sample of Ar/O3 ) 200 with a sample of
Ar/cis-C4H8 ) 250 (lower, blue trace) compared to a spectrum of the
same matrix after annealing to 36 K and recooling.
constant or decreased slightly. In a similar experiment with 18O3,
the same results were obtained overall, with shifting of many
of the product bands to lower energy. In addition, the yield of
the 18O counterpart of the 922 cm-1 band, at 905 cm-1, was
greater. As a result, it was clear that this band did increase as
a result of irradiation (although it was still relatively weak),
and additional bands were noted at 909 and 918 cm-1 corre-
sponding to mixed 16O-18O species analogous to the twin jet
experiment with scrambled 16,18O3.
In addition to merged and twin jet experiments, several
concentric jet experiments were conducted to explore the
intermediate twin regime between the two. Experiments were
conducted with ∆d ) -5 and -20 cm. The results of these
two experiments were very similar. The set A bands were
present, with intensities comparable to the twin jet and 5.0 cm
merged jet experiments. In addition, a band of weak-to-medium
intensity was observed as 1729 cm-1. However, the most intense
product bands seen in the 2.0 m merged jet experiments (e.g.,
1742 and 2799 cm-1) were not observed. This suggests that
the reaction proceeded slightly further under concentric jet
conditions than the twin jet experiments, but not nearly as far
as in the 2.0 m merged jet experiments. No additional product
bands were observed as a result of concentric jet deposition.
Figure 4. Infrared spectrum of an annealed matrix formed by the twin
jet codeposition of a sample of Ar/16O3 ) 200 with a sample of Ar/
cis-C4H8 ) 250 (lower, blue trace) compared to a spectrum of an
annealed matrix form by the twin jet codeposition of a sample of Ar/
18O3 ) 200 with a sample of Ar/cis-C4H8 ) 250 (upper, red trace).
labeling, approximately 50% 18O. In these experiments, a new
band was observed at 710 cm-1, while the region below 700
was obscured by parent bands. The doublet at 889, 895 cm-1
(16O) appeared as a broad band with weak maxima between
895 and 875 cm-1. The 922 cm-1 band (16O) appeared as a
distinct, nearly symmetric quartet at 904, 908, 918, and 922
cm-1, as shown Figure 5.
Results of Theoretical Calculations
Since the final products of the ozonolysis of cis-2-butene are
well-known, the focus of this study is on the initial intermediates
in the reaction sequence. In particular, the primary ozonide is
believed to form initially, then the Criegee intermediate,
followed by the secondary ozonide. The potential energy surface
for this reaction has been studied extensively.11-14 Although the
energetics are well-known, calculated vibrational frequencies
for these species have not been reported, including 18O isoto-
pomers. Thus, such calculations were carried out here using
the B3LYP hybrid functional and basis sets as high as
6-311G++(d,2p). All of these species optimized to energy
minima on their respective potential energy surfaces, with all
positive vibrational frequencies. The calculations carried out here
also reproduced the known experimental reaction energetics
quite well. Tables 2 and S1-S3 list the calculated bands, 18O
isotopic shifts and intensities for these three species, while
Figure 6 shows the optimized structures.
The band at 1066 cm-1 (16O) showed a very broad single
maximum at about 1062 cm-1, while the 16O feature near 1116
cm-1 now appeared as a broad doublet at 1103 and 1114 cm-1
.
The 1144 cm-1 band (16O) appeared as a single broad feature
between 1138 and 1144 cm-1, while the 1394 and 1471 cm-1
bands showed no isotopic shift. All of these bands were weak
due to the distribution of intensity over several different isotopic
species.
In a separate set of experiments, samples of Ar/cis-2-butene
) 100 and Ar/O3 ) 250 were codeposited using twin jet
deposition and then irradiated after deposition was complete.
The parent bands of O3 and cis-2-butene decreased slightly as
a result of irradiation, while a number of new bands were seen,
the most intense of which was at 1746 cm-1. Most of the product
bands seen upon irradiation were at different locations than the
bands seen in the merged jet experiments, indicating that
different products were forming. In these experiments, the set
A bands were seen prior to irradiation. Upon irradiation, the
bands at 890, 922, 1116, 1144, 1394, and 1471 cm-1 increased
in intensity, while the bands at 720 and 1066 cm-1 remained
Discussion
Merged jet and twin jet deposition probe somewhat different
time and temperature regimes with respect to the mixing and

Reaction of Ozone with cis-2-Butene
J. Phys. Chem. A, Vol. 114, No. 8, 2010 2803
more intense; and in the 5 cm merged jet the bands at 890,
898, 1116, 1144, 1394, and 1471 cm-1 were relatively more
intense. Second, the different bands did not grow at the same
rate upon annealing, indicating that more than one species is
involved. Third, different subgroups of bands showed different
intensity behavior upon irradiation.
These latter bands (890, 898, 1116, 1144, 1394, and 1471
cm-1) have tentatively been assigned by previous researchers
to the secondary ozonide of cis-2-butene (SO), a dimethyl-1,2,4-
trioxolone.30 This previous study did not have the benefit of
isotopic labeling or theoretical calculations. The data here
support and confirm this identification. As shown in Table 2,
theoretical calculations predict vibrational modes of the SO near
the bands at 890, 898, 1116, 1144, 1394, and 1471 cm-1. In
addition, the calculated isotopic shift of each vibrational mode
for the fully 18O isotopomer is very close to the observed shift.
Due to the unsymmetric nature of the molecule, six mixed
isotopic species are anticipated. For two of the modes, these
six are predicted to be relatively evenly spaced between the
position of the “pure 16O” species and the “pure 18O” species.
Since the overall shifts are not large, this leads to the prediction
of one broad band for each of these two modes, with not-well-
resolved maxima. This matches well the observations for the
890, 898, and the 1144 cm-1 bands. For two modes (1394 and
1471 cm-1), no 18O shift is predicted (CH2 bending modes), so
that the mixed isotopomers should all come at the same position.
The final band, at 1116 cm-1, shifted to 1102 cm-1 with pure
18O3, and appeared as a broad doublet at 1103 and 1114 cm-1
in the scrambled isotopic experiment. This was initially puzzling
given the 6 mixed 16,18O species that should have been observed.
However, the theoretical calculations predict that the all-16O
band and three of the mixed species should lie within 3 cm-1
of one another, leading to a broad, unresolved band at 1114
cm-1. The all-18O band and the bands of the remaining three
mixed isotopic species should likewise lie close to one another,
leading to a broad band at 1103 cm-1. The physical reason for
this is that this vibration is a stretching mode of the unique
oxygen in the molecule (the O atom in the 4 position in the
1,2,4-trioxolone, or between the two carbon atoms). In the mixed
experiments, this is either a 16O or an 18O. The nature of the
isotopic label for the other two oxygens (in the 1 and 2 positions)
has only a minimal affect on the position of this vibrational
mode. So, the isotopic labeling data, combined with density
functional calculations, supports and confirms assignment of
these bands to the secondary ozonide of cis-2-butene.
Figure 6. Calculated structures of the (a) primary ozonide of cis-2-
butene, (b) secondary ozonide of cis-2-butene, and (c) Criegee
intermediate derived from cis-2-butene.
reacting of cis-2-butene and O3 prior to deposition onto the cold
window. Twin jet deposition allows for only a very brief mixing
time on the surface of the condensing matrix surface, at
temperatures below room temperature but above the 14 K
temperature of the rigid matrix. Merged jet deposition allows
for room temperature mixing and flow through the reaction zone
before condensation into the matrix. The time available for
reaction, on the order of milliseconds, is much longer than in
twin jet. For cis-2-butene and O3, quite different products were
seen upon merged jet as compared to twin jet deposition.
Concentric jet deposition was designed to probe the time regime
intermediate between twin and merged jet deposition.
In experiments with the longest merged jet region (2.0 m),
hence the longest reaction time, parent bands of cis-2-butene
and O3 were nearly destroyed, indicating extensive reaction had
occurred. Numerous product bands were observed in these, some
of which were quite intense. As shown in Table 1, many of
these bands can be assigned to well-known species,25-28
including CH3CHO, CH2O, and CO2. Additional bands in the
CdO stretching region and the O-H stretching region strongly
suggest the formation of carboxylic acids, through secondary
reactions in the reaction zone. The majority of these products
are consistent with previous kinetics studies of this reaction as
well as a matrix isolation study in which the two reagents were
allowed to reaction for 1 h in a static reactor prior to depo-
sition.29
A direct correlation was observed between the length of the
merged region and the extent of reaction based on decreasing
intensity of parent bands and increasing intensity of product
bands (e.g., CH3CHO and CH2O) with increasing length. In the
shorter merged jet experiments, the set A were first observed
with low intensity. This same set of bands were the only product
bands seen in the twin jet experiments, indicating that they are
formed in the earliest stages of the reaction. This is consistent
with the observation that these bands increased in intensity upon
annealing matrices deposited by twin jet deposition as seen in
Figure 3. The primary interest of this study, then, is the different
species responsible for the set A bands. There are three primary
sources of experimental evidence that support the hypothesis
that more than one species is responsible for these bands. First,
the intensity ratio of these bands was not constant between the
twin jet, short merged jet, and concentric jet experiments. In
the twin jet, the bands at 720 and 1066 cm-1 were relatively
Heicklen et al.31 and Kuhne et al.30 both have reported two
infrared absorptions near 715 and 1062 cm-1 in their earlier
studies. Heicklen et al. tentatively assigned these two bands,
along with several other bands, to the primary ozonide of cis-
2-butene, cis-dimethyl-1,2,3-trioxolone. Kuhne et al. later
showed that the several additional bands were, in fact, due to
the secondary ozonide. They concluded that the 715 and 1062
cm-1 bands might be due to the primary ozonide. Data obtained
in the present study adds both isotopic shift information and
theoretical calculations to the analysis. These calculations predict
moderately intense absorptions for the primary ozonide near
680 and 1044 cm-1, consistent with the two observed experi-
mental bands within the normal error associated with the
B3LYP/6-311++g(d,2p) level of theory. A few primary ozo-
nides of other alkenes have been identified32,33 (e.g., for ethene,
in a xenon matrix). The antisymmetric stretch of the O-O-O
linkage was observed in these systems near 700 cm-1 with a
strong 18O isotopic shift. The 720 cm-1 band observed here
shifted strongly when 18O3 was employed. Unfortunately, the

2804 J. Phys. Chem. A, Vol. 114, No. 8, 2010
Clay and Ault
18O counterpart could not be observed due to overlap with
intense parent bands. However, one mixed 16,18O isotopomer
was observed in the scrambled isotopic experiments at 710 cm-1,
indicative of a strong 18O shift. The symmetric C-O ring
stretching mode of the P.O. was calculated at 1044 cm-1, not
far from the second band, at 1066 cm-1. This band was
calculated to have a 4 cm-1 18O shift, compared to the 7 cm-1
shift that was observed. Moreover, the band positions for the
mixed isotopic species are evenly distributed between those of
the all-16 and all-18 species, leading to one broad band
approximately at the midpoint between the position of the 16O
isotopomer and the 18O isotopomer. This prediction agrees very
well with the observation of a single broad maximum at 1062
cm-1 in the scrambled isotopic experiments. Finally, these two
vibrational modes of the primary ozonide were calculated to
be the two most intense modes by at least a factor of 2, providing
a rationale for why additional modes were not observed. The
evidence, then, leads to the conclusion that the 720 and 1066
cm-1 bands are due to the primary ozonide of cis-2-butene,
based on the isotopic shifts, theoretical calculations including
intensities, comparison to similar systems,32,33 and the tentative
identification by earlier researchers all point to this conclusion.
Thus, identification of and assignment to the primary ozonide
is made.
report an intense O-O stretching mode at 933 cm-1 while
Sheridan et al. reported35 the formation of a bis-carbonyl oxide
with an intense O-O stretch at 929 cm-1. Both reported
significant 18O red shifts. These data are in excellent agreement
with the observations presented here.
Overall, the several arguments presented here support forma-
tion and observation of the elusive Criegee intermediate, cage
paired with acetaldehyde. Identification of such a species through
one vibrational band is not ideal, even taking into account the
fact that only one mode with substantial intensity is predicted.
However, based on the confirming presence of acetaldehyde
and the excellent agreement of the one observed band and its
three isotopic Variants with calculations, the 922 cm-1 band
observed in these experiments is assigned to the Criegee
intermediate of cis-2-butene, namely H3C(H)CdOsO. This very
likely marks the first observation of a Criegee intermediate
formed from an acyclic alkene and strongly supports the Criegee
mechanism for the ozonolysis of cis-2-butene.
It is interesting to note the significant growth of the product
bands on annealing to 35 K. This result infers a nonzero rate
constant at cryogenic temperatures which in turn requires a very
low activation barrier to reaction. The known activation barrier
for the ozonolysis of cis-2-butene of 1.95 kcal/mol. Certainly,
while an Arrhenius-like treatment of reaction rates in a semirigid
argon lattice at 35 K is not quantitatively appropriate, this
activation barrier is consistent with the growth of bands due to
the primary ozonide, Criegee intermediate and secondary
ozonide. The relative amounts and growth of these three species
would then be determined by the rate at which the matrix
material can remove the roughly 55 kcal/mol exothermicity of
the initial reaction to form the primary ozonide. Finally, it is
noteworthy that bands due to the secondary ozonide grew
significantly upon broadband irradiation while the band of the
Criegee intermediate grew somewhat and those of the primary
ozonide decreased slightly. This suggests an interconversion of
the two ozonide species via the Criegee intermediate upon UV
excitation in the argon matrix.
Recently, a detailed matrix isolation study of the intermediates
formed during the ozonolysis of cyclopentadiene and cyclo-
pentene was reported.36 This study described the observation
of the primary ozonide, Criegee intermediate and secondary
ozonides of two cyclic alkenes, also marking the first observation
of such species. The spectral characteristics of these species
(experimental band positions and intensities, and theoretically
calculated positions) were in good agreement with the observa-
tions here. For example, the C-O-O stretch of the Criegee
intermediate of cyclopentadiene was observed at 966 cm-1 with
a 27 cm-1 18O shift, while DFT calculations at the same level
employed here predicted this band at 977 cm-1 with a 30 cm-1
shift. The O-O-O stretch of the primary ozonide of cyclo-
pentadiene was observed at 737 cm-1, similar to the 720 cm-1
band observed here, while this mode was observed at 706 cm-1
for the primary ozonide of cyclopentene. These results are
internally consistent and serve to further support the species
identification and assignments made in both papers. In general,
the level of calculations employed here (DFT B3LYP/6-
311++g(d,2p)) predicted the vibrational frequencies within 3%,
consistent with literature results,37 except for the C-O-O
stretch of the Criegee intermediate, where the calculations were
closer to 4% high.
Two bands remain from set A, the weak bands at 922 and
1720 cm-1. The 922 cm-1 band clearly arises from an early
intermediate in the ozonolysis reaction yet does not fit with
either the known spectrum of the secondary ozonide or the
calculated spectrum of the primary ozonide (indeed, it was not
reported by either previous group of researchers, as it is the
weakest of the bands observed here). It is noteworthy that the
band at 1720 cm-1 near the known26 CdO stretch of acetalde-
hyde, grows on annealing, while the 922 cm-1 band also
increases. A candidate for the species responsible for the 922
cm-1 is the long-sought after Criegee intermediate, a carbonyl
oxide. This species is formed from the primary ozonide along
with acetaldehyde in a 1:1 ratio and is destroyed by recombina-
tion to form the secondary ozonide. Since the primary and
secondary ozonides were both observed in these experiments it
is quite reasonable that the Criegee intermediate and acetalde-
hyde are formed as well. The concurrent growth of the band at
1720 cm-1 suggests formation of acetaldehyde, although slightly
perturbed from the known CdO stretching position of 1729
cm-1. This perturbation is anticipated, since the Criegee
intermediate and the acetaldehyde molecules must be trapped
in the same matrix cage. Moreover, the 18O shift of the 1720
cm-1 band, to 1690 cm-1, matches very closely the shift
anticipated for acetaldehyde.
DFT calculations for the Criegee intermediate indicate that
only one Vibrational mode should haVe significant intensity (i.e.,
I > 40 km/mol), a mode calculated to come at 962 cm-1
(unscaled) with an intensity of I ) 108 km/mol. This mode is
the antisymmetric stretch of the CdOsO linkage of the
carbonyl oxide, and has a predicted 18O shift of -19 cm-1. The
922 cm-1 band is in a reasonable position for this mode
(corresponding to a scale factor of 0.96, a value which is typical
for this level of theory), and shows an 18O shift of -18 cm-1
.
The calculated isotopic pattern for the mixed 16,18O species for
2 inequivalent O atoms has Cd16Os18O and Cd18Os16O shifts
of -4 and -14 cm-1, respectively. The experimental shifts were
-4 and -13 cm-1, in very good agreement. In addition, several
groups have observed carbonyl oxides in cryogenic matrices
(formed through different routes, not ozonolysis). For example,
Sander et al. isolated propinal O-oxide in argon matrices.34 They
Conclusions
The variation of conditions for the deposition of ozone and
cis-2-butene into argon matrices has led to the observation of

Reaction of Ozone with cis-2-Butene
J. Phys. Chem. A, Vol. 114, No. 8, 2010 2805
(18) Andrews, L.; Moskovits, M.; Eds. Chemistry and Physics of Matrix-
Isolated Species; Elsevier Science Publishers: Amsterdam, 1989.
(19) (a) Kohlmuller, C. K.; Andrews, L. J. Am. Chem. Soc. 1981, 103,
2578–2583. (b) Hawkins, M.; Kohlmuller, C. K.; Andrews, L. J. Phys.
Chem. 1982, 86, 3154–3166. (c) Andrews, L.; Kohlmuller, C. K. J. Phys.
Chem. 1982, 86, 4548–4557. (d) Clark, R. J. H.; Foley, L. J. J. Phys. Chem.
A 2002, 106, 3356–3364. (e) Feltham, E. J.; Almond, M. J.; Marston, G.;
Ly, V. P.; Wiltshire, K. S. Spectrochim. Acta, Part A 2000, 56A, 2605–
2616.
both initial reaction intermediates and stable reaction products.
Twin jet deposition followed by annealing led to product bands
which have been assigned to the primary ozonide, Criegee
intermediate and secondary ozonide of cis-2-butene, as well as
acetaldehyde cage-paired with the Criegee intermediate. These
results provide strong evidence that these ozonolysis reactions
follow the mechanism proposed by Criegee in the 1950s.
Merged jet deposition, which allows for a longer gas phase
reaction time, led to stable, known species.
(20) Ault, B. S. J. Am. Chem. Soc. 1978, 100, 2426.
(21) Carpenter, J. D.; Ault, B. S. J. Phys. Chem. 1991, 95, 3502.
(22) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.;
Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.;
Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.;
Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.;
Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li,
X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.;
Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.;
Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.;
Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich,
S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.;
Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.;
Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz,
P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.;
Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson,
B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03,
reVision B.04; Gaussian, Inc.: Pittsburgh, PA, 2003.
Acknowledgment. The National Science Foundation is
gratefully acknowledged for support of this research through
grant CHE 0749109. Michael Hoops is acknowledged for his
assistance in the early stages of this research.
Supporting Information Available: Additional tables of
calculated vibrational frequencies isotopic shifts and intensities
are available. Specific tables are cited in the manuscript text.
This material is available free of charge via the Internet at http://
pubs.acs.org.
References and Notes
(23) Andrews, L.; Spiker, R. C., Jr. J. Phys. Chem. 1972, 76, 3208.
(24) Barnes, A. J.; Howells, J. D. R. J. Chem. Soc. Faraday Trans. 2
1973, 69, 532.
(25) Schriver, A.; Schriver-Mazzuoli, L.; Vigasin, A. A. Vib. Spectrosc.
2000, 23, 83.
(26) Della Vedova, C. O.; Sala, O. J. Raman Spectrosc. 1991, 22, 505.
(27) Nelander, B. J. Chem. Phys. 1980, 73, 1026.
(28) Diem, M.; Lee, E. K. J. Phys. Chem. 1982, 86, 4507.
(29) Feltham, E. J.; Almond, M. J.; Marston, G.; Ly, V. P.; Wiltshire,
K. S. Spectrochim. Acta, Part A 2000, 56A, 2605–2616.
(30) Kuhne, H.; Forster, M.; Hulliger, J.; Ruprecht, H.; Bauder, A.;
Gunthard, H. H. HelV. Chim. Acta 1980, 63, 1971.
(31) Hull, L. A.; Hisatsune, I. C.; Heicklen, J. J. Am. Chem. Soc. 1972,
94, 4856.
(32) Kohlmuller, C. K.; Andrews, L. J. Am. Chem. Soc. 1981, 103, 2578–
2583.
(33) Andrews, L.; Kohlmuller, C. K. J. Phys. Chem. 1982, 86, 4548–
4557.
(34) Wierlacher, S.; Sander, W.; Marquardt, C.; Kraka, E.; Cremer, D.
Chem. Phys. Lett. 1994, 222, 319.
(35) Subhan, W.; Rempala, P.; Sheridan, R. S. J. Am. Chem. Soc. 1998,
120, 11528.
(36) Hoops, M. D.; Ault, B. S. J. Am. Chem. Soc. 2009, 929, 22.
(37) Irikura, K. K.; Johnson, R. D., III; Kacker, R. N. J. Phys. Chem. A
2005, 109, 8430.
(1) Ravishankara, A. R. Chem. ReV. 2003, 103, 4505.
(2) Slanger, T. C.; Copeland, R. A. Chem. ReV. 2003, 103, 4731.
(3) Olzmann, M.; Kraka, E.; Cremer, D.; Gutbrod, R.; Andersson, S.
J. Phys. Chem. A 1997, 101, 9421–9429.
(4) Horie, O.; Moortgat, G. K. Atmos. EnViron., Part A 1991, 25A,
19881–1896.
(5) Grosjean, E.; Grosjean, D. EnViron. Sci. Technol. 1996, 30, 1321.
(6) Atkinson, R.; Tuazon, E. C.; Aschmann, S. M. EnViron. Sci.
Technol. 1995, 29, 1860.
(7) Atkinson, R.; Finlayson, B. J.; Pitts, J. N., J. Am. Chem. Soc. 1973,
95, 7592.
(8) Atkinson, R.; Carter, W. P. L. Chem. ReV. 1984, 84, 437.
(9) Johnson, D.; Marston, G. Chem. Soc. ReV. 2008, 37, 699.
(10) Gillies, J. Z.; Gillies, C. W.; Suenram, R. D.; Lovas, F. J. J. Am.
Chem. Soc. 1988, 110, 7991.
(11) Chan, W.-T.; Hamilton, I. P. J. Chem. Phys. 2003, 118, 1688–
1701.
(12) Hendricks, M. F. A.; Vinckier, C. J. Chem. Phys. A 2003, 107,
7574–7580.
(13) Ljubic, I.; Sabljic, A. J. Phys. Chem. A 2002, 106, 4745–4757.
(14) Anglada, J. M.; Crehuet, R.; Bofill, J. M. Chem.sEur. J. 1999, 5,
1809–1822.
(15) Criegee, R. Rec. Chem. Prog. 1957, 18, 111.
(16) Criegee, R. Angew. Chem. 1975, 87, 765.
(17) Cradock, S.; Hinchcliffe, A. J. Matrix Isolation; Cambridge
University Press: Cambridge, 1975.
JP912253T