A combined matrix isolation and ab initio study of bromine oxides

Article abstract of DOI:10.1021/jp060586x

Bromine oxides have been generated by passing a mixture of Br ₂/O₂/Ar through a microwave discharge. The products were stabilized at 6.5 K in an excess amount of argon. Infrared spectroscopy was used to analyze the species formed; experiments with enriched ¹⁸O ₂ and ab initio calculations were carried out to assist in the assignment of the spectra. Besides the known species BrO, OBrO, and BrBrO, spectroscopic evidence for BrOBrO, BrBrO₂, and a new isomer of Br₂O₃ is reported for the first time. Extensive comparisons are drawn between the present studies and previous experimental and theoretical works. The chemistry involved in the production of the observed compounds is discussed. The assignments are corroborated by the good correlation between observed and calculated band positions and intensities.

Full text of DOI:10.1021/jp060586x

6472
J. Phys. Chem. A 2006, 110, 6472-6481
A Combined Matrix Isolation and ab Initio Study of Bromine Oxides
Oscar Ga´lvez,† Anja Zoermer,† Aharon Loewenschuss,‡ and Hinrich Grothe*,†
Institute of Materials Chemistry, Vienna UniVersity of Technology, A-1210 Vienna, Austria, and Department of
Inorganic & Analytical Chemistry, The Hebrew UniVersity of Jerusalem, Jerusalem 9190, Israel
ReceiVed: January 27, 2006; In Final Form: March 22, 2006
Bromine oxides have been generated by passing a mixture of Br2/O2/Ar through a microwave discharge. The
products were stabilized at 6.5 K in an excess amount of argon. Infrared spectroscopy was used to analyze
the species formed; experiments with enriched 18O2 and ab initio calculations were carried out to assist in the
assignment of the spectra. Besides the known species BrO, OBrO, and BrBrO, spectroscopic evidence for
BrOBrO, BrBrO2, and a new isomer of Br2O3 is reported for the first time. Extensive comparisons are drawn
between the present studies and previous experimental and theoretical works. The chemistry involved in the
production of the observed compounds is discussed. The assignments are corroborated by the good correlation
between observed and calculated band positions and intensities.
I. Introduction
HgO,21,38 photolysis with subsequent gas-phase reaction,34,35 and
thermolysis of precursors.42 Because of the instability of bromine
oxides, the matrix isolation technique is a necessary prerequisite
for stabilization of the synthesized compounds. In this context,
we want to highlight the experiments of Tevault et al., in which
several bromine oxygen compounds (BrO, OBrO, BrBrO, and
BrOBr) were formed in an argon matrix and studied by infrared
spectroscopy.34 Subsequently, more studies have been carried
out to gain insight into the nature of these oxides, mainly
applying matrix isolation techniques,35-38 but in some cases,
measurements in gas phase were also successful.39-41 However,
these methods demand a certain stability of the new compound
in question, and in particular, higher oxides are difficult to
access. Thus, a rather harsh oxidation process and rapid
stabilization at low temperatures presents itself as a viable
method and results in a broad variety of products, also including
hitherto unknown higher bromine oxides. Attempts to generate
higher bromine oxides by passing a mixture of bromine and
oxygen through a microwave or glow discharge have already
been made by Pflugmacher et al. in the 1950s, resulting in
different bromine species.43 More recent experiments have
produced several bromine oxides by means of the microwave
discharge of Br2/O2 mixtures.36,39
The halogen oxides, especially those containing chlorine and
bromine, have received considerable attention in the last two
decades, since they are involved in atmospheric processes and
play a leading role in stratospheric1,2 and tropospheric3-5 ozone
depletion reactions. Chlorine oxides have been investigated
extensively for many years,6,7 but studies on bromine oxides
are much scarcer, because they are less abundant in the
atmosphere and are thermodynamically less stable than the
chlorine oxides. Nevertheless, during the past decade it has been
shown that the bromine species have a higher ozone depletion
potential than their chlorine analogues.8,9 Enhanced concentra-
tions of bromine species have been associated with arctic
tropospheric ozone depletion episodes,3-5 probably caused by
catalytic cycles mainly involving the radicals Br and BrO,
whereas the corresponding chlorine compounds were less
significant in these processes.10 Consequently, the interest in
characterizing atmospheric bromine species has been stimulated
in the past few years.
Most of the spectroscopic and structural information available
on bromine oxides concern BrO, BrO2 (OBrO and BrOO), and
Br2O (BrOBr and BrBrO), which are the simplest and most
stable species.11-16 The first attempts to identify these com-
pounds date back to the 1930s, when UV-vis and IR spec-
troscopy helped in the identification.17-23 For other bromine
species, ab initio calculations are the major source of
information.24-30 For instance, the (BrO)2 oxides, which due to
kinetic data should be formed in the BrO self-reaction, are only
suggested by theoretical studies which have presented respective
structures and vibrational spectra;24,25 however, they were not
yet observed experimentally. Several studies, mainly on solid
phases, refer to higher oxides of bromine, BrOx (x ) 3-4),
Br2Oy (y ) 3-7), and Br3Oz (z ) 6, 8), but for some of them,
their existence was not even definitely confirmed.13,31-33
Regarding the production of halogen oxides, different meth-
odologies have been employed: heterogeneous reactions on
In this work, we have generated bromine oxides by passing
Br2/O2/Ar mixtures in different ratios through a microwave
discharge located close to the cold surface of a cryostat kept at
6.5 K. Thus, the wide range of highly unstable bromine oxide
species can, after their formation in the plasma afterglow and
in the flight path to the cold surface, be stabilized in an excess
amount of argon. Since the matrix layer contains a high
concentration of unreacted bromine and especially oxygen
atoms, further reactions of bromine species with the highly
mobile oxygen is facilitated by thermal cycles, resulting in
pronounced formation of higher bromine oxides. The matrix
isolated species were studied by infrared spectroscopy, and the
observed bands were assigned on the basis of both isotope
effects due to the natural abundance of bromine isotopes and
18-oxygen enrichment as well as by comparison with results
of ab initio calculations. The experimental design and the ab
* Corresponding author. E-mail: grothe@tuwien.ac.at. Tel. +43-1-
25077-3809. Fax. +43-1-25077-3890.
† Vienna University of Technology.
‡ The Hebrew University of Jerusalem.
10.1021/jp060586x CCC: $33.50 © 2006 American Chemical Society
Published on Web 04/29/2006

Bromine Oxides
J. Phys. Chem. A, Vol. 110, No. 20, 2006 6473
Figure 1. Matrix isolation setup for bromine oxide formation and chemistry study.
initio methods employed are addressed in the next two sections,
and then, our results are reported and discussed.
was found not to be dependent on the flow rate in the range
1-10 mmol/h. Therefore, the flow rate was kept at a value
which ensured optimum working conditions for the microwave
cavity. The incident microwave power was 35 W in all
experiments, and the reflected power was between 4 and 9 W,
depending on the gas mixture employed. The deposited amount
of discharged Br2/O2/Ar mixture was ca. 35 mmol, together with
another ca. 3 mmol of pure argon. Spectra were taken at a
resolution of 0.3 cm-1, and 400 scans were coadded to obtain
a good signal-to-noise ratio despite the low IR intensities of
the bromine oxide species. All depositions were done at 6.5 K,
and each matrix was subjected to a temperature cycle of 15
min at 33 K with a heating rate of 1 K/min; subsequent
measurement was done at 6.5 K.
II. Experimental Section
The matrix isolation setup (see Figure 1) consisted of a two-
stage closed-cycle cryostat (Leybold RDK 6-320) fitted with
an ultrahigh vacuum-tight shroud and rotary feedthrough for
turning the sample support by 180°, i.e., from the plasma-tube
side to the IR-spectrometer side and vice versa. The system
allowed for extended deposition without the risk of contamina-
tion of the matrix. The base pressure of the system was below
10-8 mbar even during rotation. The cold surface was equipped
with a silicon diode for temperature measurement and a heating
unit managed by a Lakeshore 330 controller. The infrared
spectrometer (Bruker 113v) was fitted with a deflection optic
in one of the sample chambers to guide the beam out of the
spectrometer and through a KBr window to the gold-plated
sample support, thus facilitating measurement in reflection-
absorption mode. The microwave discharge was effected by
means of a microwave generator (EMS Microtron 200) con-
nected to an Evenson cavity (Sairem) enclosing a quartz tube.
An orifice of ca. 1.2-mm diameter downstream of the cavity
ensured suitable pressures in both the plasma and the cryostat
compartment. Gas samples were premixed in a separate glass
vacuum line. The resulting gas consisted of 0.5% bromine
(Aldrich; purified by fractioned condensation) and 0.5-1.5%
oxygen (16O2, Linde, purity 4.5; 18O2, Spectra Gases, 95%
enriched; both gases were used without further purification) in
argon (Messer, purity 6.0; used without further purification).
The flow of the gas mixtures was controlled by a needle valve.
Pure argon was deposited through separate inlets (gas inlets 1
and 2 in Figure 1) and controlled by a mass flow controller
(MKS M330). It was found that the optical quality of the
resulting matrix layers was improved when argon was co-
deposited at a lower flow rate with the bromine mixtures after
finishing a base layer of pure argon.
III. Calculations
Quantum calculations were performed with the hybrid DFT
method B3LYP, which consists of a mixture of Hartree-Fock
(HF) exchange with Becke’s three parameters exchange func-
tional plus the nonlocal correlation functional of Lee, Yang,
and Parr. We chose the aug-cc-pVTZ basis set as a fair
compromise between size and reliability. DFT calculations with
a sufficiently flexible basis have been found well-suited to obtain
geometries,25 spectra,27 and heats of formation26 of bromine
oxides. Equilibrium geometries were fully optimized using
analytic gradients without symmetry constraints. The energies
for every structure used in the calculations of reaction energies
include the zero-point energy correction (ZPEC). Geometries,
frequencies, and energies were determined with the Gaussian
03 package.44
IV. Results
Matrix Isolation Experiments. Different ratios of Br2/O2
(1:1, 1:2, and 1:3) were discharged and co-deposited in an argon
matrix. After deposition, infrared absorptions were measured
in the spectral range 500-4000 cm-1. However, most bromine
oxides can be observed in the region 650-950 cm-1 as shown
in Figure 2. When the matrix is subjected to an annealing
process, concentrations of the species previously observed
undergo strong changes, and new absorption bands are also
recorded. The amount of ozone (band labeled O in the spectra)
In a typical sample preparation run, a base layer of ca. 1 mmol
argon was deposited at a flow rate of 1.5 mmol/h. Subsequently,
the argon flow rate was turned down to 725 µmol/h, and the
plasma was ignited after establishing a suitable flow of the
premixed gases. The flow rate of the discharged gas was
maintained at ca. 8.5 mmol/h. The quality of the resulting sample

6474 J. Phys. Chem. A, Vol. 110, No. 20, 2006
Ga´lvez et al.
Figure 2. Infrared spectra observed following the deposition of microwave-discharged bromine and 16-oxygen mixture in argon matrixes. (a, c,
e) Br2/16O2 ratio of 1:1, 1:2, and 1:3, respectively. (b, d, f) Previous matrixes after annealing for 15 min at 33 K.
and the bands N2 and N5 increase. We use the label “N” for
addressing species containing nitrogen atoms (its origin is
explained in the discussion section) and “B” to refer to bromine
oxides. Concerning the bromine species, B1 decreases and B6
vanishes after annealing. The bands B3, B5, and B8 increase,
and six new bands appear: B2, B2′, B2′′, B4, B7, and B7′
(labels with the same number refer to different vibrational modes
of the same molecule). When the ratio of bromine/oxygen is
changed, different abundances of the species can be observed.
Consequently, the concentration of ozone formed during an-
nealing increases with the amount of oxygen in the sample.
Species which contain nitrogen atoms also show larger con-
centration when the amount of oxygen in the mixture is higher.
The bromine species (B1 to B8 in the spectra) show different
behaviors when the ratio Br2/O2 is altered. The group of bands
B1, B3, B5, B6, B7, B7′, and B8 increase in intensity when the
ratio changes from 1:1 to 1:2, but stay nearly constant from
1:2 to 1:3. B2 remains almost invariable despite the different
amounts of oxygen in the mixture, and the band B4 increases
from 1:1 to 1:2 and decreases when the ratio of bromine/oxygen
is 1:3. The low intensity of the bands B2′ and B2′′ makes it
rather difficult to properly analyze their variations.
To help in the assignment of the bands, similar experiments
with 18-oxygen were carried out, and the spectra are shown in
Figure 3. All the positions of bands shown in Figure 2 are red-
shifted in different proportions, verifying the absence of species
not containing oxygen. Because of the different red-shifts, the
pattern of the infrared bands shows some differences with
respect to those with 16-oxygen. Band N4 is shifted to the
higher-wavenumber side of B3, at 773.0 cm-1, overlapping with

Bromine Oxides
J. Phys. Chem. A, Vol. 110, No. 20, 2006 6475
Figure 3. Infrared spectra observed following the deposition of microwave-discharged bromine and 18-oxygen mixture in argon matrixes. (a, c,
e) Br2/18O2 ratio of 1:1, 1:2, and 1:3, respectively. (b, d, f) Previous matrixes after annealing for 15 min at 33 K.
B4, while N5 and B6 are also in the same frequency range in
the spectra. Regardless of these differences, the 18-oxygen bands
show variations similar to those of the 16-oxygen bands
described before when the matrix is annealed or the ratio Br2/
18O2 is changed. All band positions shown in Figures 2 and 3
are listed in Table 1.
Ab initio Calculations. The infrared spectra recorded are
very congested in the region of interest for the identification of
bromine oxide species. For several of these species, no previous
infrared spectroscopic information is available, so quantum
chemical methods can be of appreciable help in understanding
the observed spectra. We provide in Table 2 B3LYP calculations
of the geometrical parameters and harmonic frequencies for
bromine species supposed to be formed in our experiments along
with the available theoretical and experimental data found in
the literature. High-level ab initio results from other authors,
mostly obtained with MP2 and CCSD(T) methods, have been
included in Table 2 for the purpose of comparison. In some
instances, DFT data similar to ours were selected when other
methods were not available. The experimental data included
for comparison are primarily from measurements in the gas
phase, although frequency values observed in argon matrix were

6476 J. Phys. Chem. A, Vol. 110, No. 20, 2006
Ga´lvez et al.
TABLE 1: Wavenumbers (cm-1) of the Products of a
Microwave-Discharged Mixture of Molecular Bromine and
Molecular 16- or 18-Oxygen, Respectively, in an Excess of
Argona
the matrix during thermal cycles, confirmed by the respective
changes in band intensities and the appearance of new bands,
supports our assignments, as will be discussed in the following
sections.
IR (Ar matrix)
Spectra after Deposition of the Discharged Mixture. The
spectra recorded immediately after deposition of the discharged
mixture (Figures 2 and 3) show characteristic bands, which can
be classified into pairs of bands and single bands. Bromine has
two natural isotopes in almost equal amounts (50.54% for 79Br
and 49.46% for 81Br), and consequently, with sufficient resolu-
tion, band pairs are expected for modes which involve move-
ments of one bromine atom. The origin of the single bands
recorded will be discussed in detail later. Bromine and oxygen
atoms are formed in the plasma tube after the microwave
discharge and react to generate the different compounds
observed in the spectra. Besides IR-inactive species such as Br2
and O2, bromine monoxide could be expected as the major
product in this reaction, although Br2O, BrO2, and different
dimers of BrO are other probable species.
band
C
16O2
18O2
species
CO2
663.5
661.9
703.6
653.6
651.9
664.0
O
O3
BrO
NO2
N2O4
N2O4
X-O-O-Y
B1 730.1 (731.6)
N1 751.0
748.7
695.5 (697.1)
724.0
722.5
721.8
718.0
714.0
747.4
X
763.9
760.3
755.5
706.6
N2 769.5 (770.5)
N3 782.7
741.2 (742.7)
755.3
Br(ONO)x
BrNO2
OBrBrO2
B2 785.3 (787.0)
B2′ 854.7 (856.0)
B2′′ 913.5 (915.9)
B3 802.9 (804.6)
801.9 (803.6)
748.5 (750.2)
812.7 (814.2)
873.2 (875.8)
765.2 (766.9)
764.4 (766.0)
772.9
BrBrO
Four pairs of bands related to bromine oxides are observed
before annealing: B1, B3, B5, and B6. In former experiments
by Tevault et al.,34 bands similar to B1, B3, and B6 were
observed and assigned to BrO, BrBrO, and OBrO molecules,
respectively. The frequency values recorded by Tevault for these
bands deviate from our values at less than 1 cm-1 for BrO and
BrBrO, but 7 cm-1 for OBrO. Ko¨lm et al.37 and Maier and
Bothur59 observed the asymmetric stretching mode of OBrO at
845.2 and 846.6 cm-1, respectively, differing from our observa-
tion by less than 1 cm-1, so the value of Tevault for OBrO
should presumably be considered with some caution. The
measured and calculated frequency values and intensities for
these compounds are summarized in Table 3. Although anhar-
monic corrections are not included in the calculations, frequen-
cies deviate less than 3% from the measurements. The sym-
metric stretching mode of OBrO was observed at 795.7 cm-1
by Ko¨lm37 and at 794.1 cm-1 by Maier and Bothur.59 According
to our B3LYP calculation (results not shown), this normal mode
is 10 times less intense than the asymmetric stretching, so this
band may be too weak to be observed in our spectra. The
calculated and observed shifts due to the natural bromine
isotopes show exactly the same values: 1.5, 1.7, and 2.3 cm-1
for the B1, B3, and B6 species, respectively. The analogous
shifts caused by 16- and 18-oxygen isotopes for these species
are 34.5, 37.5, and 37.0 cm-1, which is also in good concordance
with our calculated values of 35.2, 39.2, and 38.5 cm-1. B1
shows a shoulder at ca. 723 cm-1 (689 cm-1 with 18-oxygen)
which might be caused by a complex X‚‚‚BrO. This band
decreases when the matrix is annealed, so a small molecule,
which can move inside the matrix, could be involved in this
complex. Our ab initio calculations for hydrates of BrO show
a red-shift of 10 cm-1 for a complex H2O‚‚‚BrO (publication
in process). Typical absorptions due to the O-H stretch in water
molecules are also recorded in the region between 3700 and
3800 cm-1, so we tentatively assign the shoulder to this
complex. In the region of the stretching -Br-O mode of
BrBrO, four bands (labeled B3) with different intensities can
be observed in the experiments with 18-oxygen, along with a
group of five bands in the 16-oxygen spectra due to the
absorption of species N4, which will be discussed toward the
end of this section. There are four different possibilities of
combining the two natural bromine isotopes to yield BrBrO,
which all have almost the same probability. Nevertheless, the
Br isotope splitting is not expected for ν(Br-O), since the BrO
N4 800.5
BrONO2
B4 810.0, 811.6, 812.5, 814.2 774.6, 776.3, 778.3, 779.6 BrBrO‚‚‚X
B5 829.7 (831.5)
N5 835.9
791.0 (792.8)
807.6
BrOBrO
BrONO
OBrO
B6 843.0 (845.3)
B7 880.3 (881.7)
B7′ 931.9 (934.4)
B8 1485.1
806.0 (808.3)
837.1 (838.6)
890.8 (893.4)
1401.3
BrBrO2
BrOO
a When pairs of bands are recorded, the absorption caused by the
79-bromine isotope is given in parentheses.
selected when gas-phase data do not exist. Our calculated
geometrical parameters yield results similar to those of ab initio
calculations from other authors, showing deviations less than
0.05 Å and 8° (3° in most cases) for bond lengths and angles,
respectively. The only exception is our calculated Br-O bond
length for the BrOO molecule, which shows large differences
from the calculations of other authors.29 In this case, the
bromine-oxygen interaction is weak, and consequently, the
bond length is large, making proper calculations difficult and
resulting in discrepancies among different ab initio methods.
The B3LYP results are in good agreement with the experimental
data available, with differences less than 0.05 Å and 3° in bond
lengths and angles, respectively. Ab initio results generally
overestimate the vibrational frequency values due to the neglect
of anharmonic corrections. Nevertheless, in the majority of
cases, our B3LYP calculations yield smaller values than MP2,
showing better agreement with the experimental values obtained
by others groups, with deviations typically less than 3%.
V. Discussion
Contrary to other experiments described in the literature,
which were designed to yield some specific bromine oxide,37,38
our method of production generates a high number of different
species. As mentioned before, this fact could complicate the
assignment of the IR spectra; nevertheless, our method provides
the opportunity to form species not previously studied, for
instance, the BrO dimers. Bromine oxides usually show less
infrared activity than the analogous chlorine compounds; e.g.,
BrO has an absorption coefficient two times lower than that of
ClO.57,58 Although this fact introduces yet another complication
to getting high-quality spectra, infrared spectroscopy remains
the method of choice, because it provides information about
the geometry of the compounds and thus assists the proper
assignment of the bands. The observation of reactions inside

Bromine Oxides
J. Phys. Chem. A, Vol. 110, No. 20, 2006 6477
TABLE 2: Ab Initio Geometric Parameters and Vibrational Frequencies in Comparison with Calculated and Experimental
Literature Valuesaa
B3LYP
ab initio
B3LYP
ab initio
species
BrO
parameter
this work other authors
exptl
1.717c
723.4d
species
parameter
r(O-Br-)
this work other authors
exptl
r(Br-O)
stretch BrO
r(Br-O)
1.729
1.743a
OBrBrO2
1.695
2.763
1.631
1.635
108.5
101.9
99.1
-46.3
797.9
860.5
922.9
2.057
1.188
114.0
131.9
180.0
804.7
742.0
784b
1.634
1.604e
112.4
r(-Br-Br-)
BrO3
r(-Br-Ocis)
θ(O-Br-O)
stretch sy Br-O3 (A)
stretch asy Br-O3 (E)
r(Br-Br)
114.0e
r(-Br-Otrans)
801.8
788.0
2.427
1.674
111.3
827.3
1.858
1.868
1.663
116.8
815f
800g
828g
θ(O-Br-Br-)
810f
θ(-Br-Br-Ocis
)
)
BrBrO
2.45
θ(-Br-Br-Ocis
r(Br-O)
θ(Br-Br-O)
stretch -Br-O
1.64h
112.8h
1056h
1.846j
1.904j
1.630j
111.6j
112.0j
70.3j
τ(O-Br-Br-Ocis
)
stretch O-Br-
804.6i
stretch sy -Br-O2
stretch asy -Br-O2
r(Br-N)
BrOBrO r(Br-O-)
r(-O-Br-)
Br-NO2
2.040q
1.201q
2.012r
1.196r
r(-Br-O)
r(-N-O)
θ(Br-O-Br-)
θ(-O-Br-O)
τ(Br-O-Br-O)
stretch -Br-O
θ(Br-N-O)
114.3q
114.5r
110.3
78.4
850.8
θ(O-N-O)
131.4q
180.0q
799q
131.0r
180.0r
782.9s
τ(Br-N-O-O)
scissors -NO2 (A1)
1082b
1.656a
117.5a
877b
OBrO
BrOO
r(O-Br)
1.652
1.649k
114.4k
799.4l
trans-BrONO r(Br-O-)
r(-O-N-)
1.836
1.503
1.150
111.1
108.5
180.0
1.849q
θ(O-Br-O)
stretch sy BrO2 (A1)
stretch asy BrO2 (B2)
r(Br-O-)
113.9
828.5
869.1
1.529q
1.158q
r(-N-O)
898b
848.6l
θ(Br-O-N-)
108.4q
2.512
1.199
118.4
1585.6
1.829
1.916
1.615
114.8
2.258m
1.168m
118.3m
1581m
1.840n
1.971n
1.623n
113.6n
101.2n
θ(-O-N-O)
τ(Br-O-N-O)
108.1q
180.0q
835q
r(-O-O)
θ(Br-O-O)
stretch -O-O
bend -O-N-O (A′) 865.1
835.9t
1485.1i
BrONO2
r(Br-O-)
r(-O-N-)
r(-N-Ocis
1.837
1.821u
1.829V
BrOBrO2 r(Br-O-)
r(-O-Br-)
1.485
1.189
1.191
1.502u
1.190u
1.193u
1.456V
1.205V
1.205V
)
r(-Br-O)
r(-N-Otrans)
θ(Br-O-Br-)
θ(-O-Br-O)
τ(Br-O-Br-O)
θ(Br-O-N-)
115.1
113.0u
113.9V
104.1
55.9
905.3
θ(-O-N-Ocis
)
118.0
109.2
0.0
117.4u
108.6u
0.0u
119.5V
106.6V
0.0V
θ(-O-N-Otrans
)
stretch sy -Br-O2 (A′)
911°
966°
τ(Br-O-N-Ocis)
stretch asy -Br-O2 (A′′) 953.8
stretch Br-O- (A′)
scissors -NO2 (A′)
751.5
815.1
753u
750w
BrBrO2
r(Br-Br-)
2.495
1.620
104.4
111.3
112.0
887.2
942.8
2.488j
1.595j
104.1j
112.1j
780u
803.3u
r(-Br-O)
θ(Br-Br-O)
θ(O-Br-O)
τ(Br-Br-O-O)
stretch sy -Br-O2 (A′)
stretch as -Br-O2 (A′′)
893p
950p
aa Bond distances, r, in Å; angles, θ and τ, in deg; and frequencies in cm-1. For the harmonic frequencies calculation, 79Br, 16O, and 14N isotopes
were used. a MP2/6-311G* in ref 30. b MP2/6-31G* in ref 30. c Ref 11. d Gas phase in ref 45. e MP2/6-311G(2df) in ref 46. f B3LYP/6-311G(2df)
in ref 46. g Deduced from an assumed geometry for BrO3 in ref 47. h MP2/6-31G(3d1f) for O and (9s,6p,2d) contraction of a (14s,11p,5d) primitive
set with additional set of d and f polarization functions for Br, in ref 38. i Ar matrix in ref 38. j MP2/AREP/TZ(2df) in ref 24. k Ref 48. l Ref 49.
m MP2/AREP/TZ(2df) in ref 29. n B3LYP/DZP++ in ref 28. o B3LYP/6-311+G(3df) in ref 27. p B3LYP/6-311++G(3df,3dp) in ref 25. q CCSD(T)/
TZ2P in ref 50. r Ref 51. s Ar matrix in ref 52. t Ar matrix in ref 53. u MP2/6-311++G(3df) in ref 54. V Ref 55. w Ne matrix in ref 56.
TABLE 3: Observed and Calculated IR Frequencies of the Br-O Stretching Modes of the Bromine Oxides Studied in This
Worka
calculation (B3LYP this work)
81Br, 16 79Br, 18 81Br, 18
IR Ar matrix
81Br, 16 79Br, 18
730.1 697.1
species
BrO
BrBrO
BrOBrO
OBrO
OBrBrO2
79Br, 16
O
O
O
O
intensityb 79Br, 16
O
O
O
81Br, 18
O
assignment
742.0
827.3
850.8
869.1
797.9
860.5
922.9
887.2
942.8
740.5
825.6
849.0
866.8
796.3
859.2
920.4
885.8
940.2
706.8
788.1
810.7
830.6
760.0
817.6
882.0
842.8
900.8
705.3
786.4
808.9
828.2
758.3
816.2
879.3
841.4
898.2
1
61
731.6
695.5
765.2
791.0
806.0
748.5
812.7
873.2
837.1
890.8
stretch Br-O
804.6
831.5
845.3
787.0
856.0
915.9
881.7
934.4
802.9
829.7
843.0
785.3
854.7
913.5
880.3
931.9
766.9
792.8
808.3
750.2
814.2
875.8
838.6
893.4
stretch Br-O
65
49
stretch -Br-O
asym stretch BrO2
stretch O-Br-
131
78
sym stretch -BrO2
asym stretch -BrO2
sym stretch -BrO2
asym stretch -BrO2
86
76
BrBrOO
97
a Frequencies in cm-1 and intensities in km/mol. b Calculated at B3LYP/aug-cc-pVTZ.
vibration is weakly coupled with the low-lying Br-BrO mode.
Accordingly, the four maxima shown in the absorption of B3
are presumably caused by matrix splitting due to different
environments. From our calculated intensity values (see Table
3), BrBrO and OBrO are about 50 times stronger IR absorbers
than BrO. Despite this, the band B1 shows higher intensity in
the spectra than B3 and B6, so the concentration of BrO in the
matrix must be much higher, in accord with our assumption of
BrO being the dominant species. In the region of the O-O
stretching vibrations, a band (B8) at 1485.1 cm-1 was recorded.
This band was previously observed by different groups and
assigned to the O-O stretching mode of the BrOO molecule.34,37
Another pair of bands is observed around 830 cm-1 (792
cm-1 for the experiments with 18O2) before the annealing
process, and was formerly tentatively assigned by Tevault et
al.34 to (BrO)2. This compound has been proposed to exist as
an important intermediate in the BrO self-reaction
BrO + BrO f [(BrO)2*] f Br2 + O2
(1)

6478 J. Phys. Chem. A, Vol. 110, No. 20, 2006
Ga´lvez et al.
which is the key step in the catalytic destruction of tropospheric
ozone occurring in the arctic vortex during springtime (see ref
10 and references therein). Despite the importance of (BrO)2,
no conclusive spectroscopic information is available so far to
confirm its existence. Since BrO is the major bromine oxide
species formed after reaction of bromine and oxygen atoms,
reaction 1 could probably take place, and the dimer could be
stabilized in the argon matrix. Three different dimers can be
proposed as the product of this reaction
Our predicted intensity values for the nitrogen-containing species
(with the exception of NO2) are 2 orders of magnitude higher
than bromine monoxide. Although the intensities of these bands,
especially N3 and N4, appear comparable with those of the
bromine oxides, the concentrations of these species in our matrix
are rather low. Experiments with mixtures enriched in CO and
water were also done, but no relevant features were found in
the spectral region under study, revealing that impurities of
carbon and hydrogen are not responsible for the formation of
the bands shown in the figures. When the amount of oxygen in
the mixture is increased, a low-intensity group of bands grows
(bands labeled “X” in the spectra), with the strongest absorption
at 760.3 cm-1, which is red-shifted to 718.0 cm-1 in the
experiments with 18-oxygen. These bands neither show the
natural bromine isotopic splitting nor increase when contami-
nants are deliberately added to the mixture, so we tentatively
suggest that O-O stretching vibrations in higher bromine
peroxides could be the modes responsible for these absorptions.
Spectra after Annealing Process. When an ideal and
extremely dilute matrix is subjected to an annealing process,
the structure relaxes and the argon atoms are reordered to yield
the face-centered cubic structure of the equilibrium rare-gas
solids.63 This fcc structure is usually not completely achieved
in a real matrix due to the disorder generated by the guest
molecules. During the process of annealing, small species can
move inside the matrix and react with larger molecules
embedded in it. In this process, an excess of oxygen atoms
produced in the microwave discharge and deposited in the argon
matrix could react with bromine species, yielding higher bromine
oxides. This idea is supported by the increase of ozone formed
by the reaction of O2 and oxygen atoms during annealing (see
Figure 2), revealing the presence of unreacted oxygen atoms in
the matrix environment. The voluminous bromine compounds
are not expected to become mobile, so reactions are facilitated
by small mobile species, in particular, oxygen atoms.
BrO + BrO f BrOOBr ∆E ) -16.93 kcal/mol (2)
f BrOBrO ∆E ) -6.08 kcal/mol (3)
f OBrBrO ∆E ) +13.44 kcal/mol (4)
The values for ∆E are calculated including the ZPEC as outlined
in section III. While the formation of BrOOBr (reaction 2) and
BrOBrO (reaction 3) are thermodynamically favored, the
production of OBrBrO (reaction 4) is an endothermic process.
The calculated vibrational frequencies, with a correction factor
of 0.986 for taking anharmonic effects into account, show an
infrared absorption around 830 cm-1 for BrOOBr and BrOBrO.
In the case of BrOOBr, this IR absorption is caused by the O-O
stretching mode, so bromine isotopic splitting is not expected.
Our calculations also predict another intense IR absorption at
586 cm-1 for this species, which is not observed in our spectra.
As a consequence of these facts, we tentatively assign the band
B5 to the stretching -Br-O normal mode of BrOBrO, in good
accordance with our predicted value (see Table 3). The
calculated bromine and oxygen isotopic splittings for this normal
mode are 1.8 and 40.1 cm-1, which is in very good agreement
with the measured values of 1.8 and 38.7 cm-1. The other
stretching and bending modes of this compound are below the
low-frequency limit of our setup. The presence of (BrO)2 with
the structure BrOBrO agrees with recent experiments of Odeh
et al. studying the reaction BrCl + ClO2 .
- 60 The authors propose
The concentration of BrO (band B1) decreases during
annealing, most probably due to reactions inside the matrix
leading to other compounds. Assuming that BrO is reacting with
oxygen atoms, two possible products could be generated
that the reaction proceeds through BrOXO intermediates with
a chainlike structure rather than a Y-shaped form. In high
theoretical level calculations, the nonplanar linear conformation
of BrOBrO yields the highest interaction energy, with differ-
ences larger than 3 kcal/mol with respect to the other conforma-
tions.30 These findings support our structural assignments. The
compound BrBrO2, which presents a Y-shaped structure, is also
found in our experiments (see bands B7 and B7′ in the spectra),
but it is probably not formed in the course of the BrO self-
reaction, as will be discussed in more detail later.
OBr + O f OBrO
BrO + O f BrOO
(5)
(6)
Reaction 5 does not seem to occur to a large extent, because
the OBrO band (B6) nearly disappears during annealing.
Nevertheless, a strong increase in the intensity of the O-O
stretching vibration of BrOO (band B8) is observed, supporting
the second reaction channel. A higher excess of oxygen atoms
is expected when the amount of oxygen in the mixture is
elevated, and consequently, a larger decrease of BrO should be
observable in mixtures rich in oxygen. This assumption agrees
with the results obtained by the evaluation of the BrO band
areas in Figures 2 and 3, which show a 30% larger decrease of
this band for a ratio of Br2 to O2 of 1:3 in comparison to 1:1.
The concentration of BrBrO (B3) increases during the thermal
cycle. This fact could be caused by the reaction of IR-inactive
molecular bromine with oxygen atoms. When a large excess of
oxygen is available (i.e., in mixture 1:3), more bromine is
expected to be consumed already in the microwave discharge,
depleting the final concentration inside the matrix. Consequently,
a minor increase of the BrBrO concentration is expected for
oxygen-rich mixtures. This correlation is found in our experi-
ments, offering support to our assumptions.
To elucidate the origin of the single bands, experiments with
mixtures enriched in possible contaminants (nitrogen, carbon,
and hydrogen) were carried out. When ca. 10% of N2 is added
to the mixture (results not shown), bands labeled “N” in our
spectra are dramatically increased in intensity. This fact reveals
that these absorptions belong to species containing nitrogen.
These contaminations most probably arise from impurities of
the gases used for the mixtures. The most likely source is O2,
because the amount of nitrogen species increases with the
amount of oxygen in the mixture. By comparison with IR spectra
recorded in argon matrix and reported in the literature, the bands
N1, N3, N4, and N5 could be assigned to bending modes of
-NO2 groups in the species NO2,61 BrNO2,52 BrONO2,56 and
trans-BrONO,53 respectively. These assignments agree with our
calculated frequencies for these molecules, displaying deviations
less than 4% (see Table 2). In the case of the NO2 band, several
maxima are observed. We assign the maximum at 751.0 cm-1
to pure NO2 and the maxima at 748.7 and 747.4 cm-1 to N2O4
species, according to the experiments of Louis and Crawford.62
The band belonging to BrOBrO (B5) dramatically increases
in intensity during annealing. Since BrO is immobilized after

Bromine Oxides
J. Phys. Chem. A, Vol. 110, No. 20, 2006 6479
it is embedded in the matrix, the BrO self-reaction is no viable
path for the formation of BrOBrO during the thermal cycle. A
possible reaction to form BrOBrO inside the matrix would be
discussion, the spectra are not shown, and we will only mention
features relevant for our assignments. The spectrum recorded
in the experiment with a mixture of 16O2 and 18O2 reveals a
new band which shows a red-shift of 13.4 cm-1 with respect to
B7′, in very good agreement with our simulated spectrum of
BrBr16O18O, which yields a red-shift of 13.8 cm-1 for this
absorption.
BrOBr + O f BrOBrO
(7)
The symmetric and asymmetric stretching modes of BrOBr
have been observed by Ko¨lm et al.37 in argon matrix at 622.2
and 525.3 cm-1, respectively. Because both the experimental
intensities for these modes are much lower than those of the
compounds described before and our setup shows very low
signal intensity below 650 cm-1, we do not observe this oxide
in our spectra. Our calculated intensities confirm the low values
for these two normal modes of BrOBr, being around 40 and 15
times lower than the -Br-O stretching mode of BrBrO and
BrOBrO, respectively. Thus, we can only assume reaction 7 as
the most probable path for the formation of BrOBrO.
The concentration of OBrO drastically decreases in the
annealing process, although reaction 5 would be a possible path
to form OBrO molecules. Following the chemistry described
for the other species, oxygen atoms could react with OBrO to
produce BrO3. While many authors have claimed to find
BrO3,43,47 no definite experimental evidence of this radical has
been reported so far. Several theoretical studies have been
performed to elucidate the structure and IR frequencies.24,46
Although this compound is expected to have two IR absorptions
in the region under study (see Table 2), we have not observed
any band which could confidently been ascribed to BrO3 with
the predicted C3V structure. Nevertheless, further experiments
have to be performed to consider other possible structures of
BrO3.
The bands around 786 cm-1 (B2) were already observed by
Tevault et al. and assigned to a complex BrBrO‚‚‚X. According
to our calculated spectra for OBrBrO2, the stretching O-Br-
mode of this molecule shows a predicted harmonic value of
797.9 cm-1, which compares well with the recorded value for
B2. A possible way for the formation of this compound in the
matrix involves the following sequence:
Br2 + O f BrBrO
BrBrO + O f BrBrO2
BrBrO2 + O f OBrBrO2
(9)
(8)
(10)
Unreacted molecular bromine as well as BrBrO are present
in the matrix in appreciable amounts before annealing and thus
provide the educts for the formation of OBrBrO2. The small
amount of BrBrO2 observed immediately after the microwave
discharge could also facilitate the formation of this compound.
Previous theoretical and experimental information has not been
found for OBrBrO2, but different studies proved the existence
of BrOBrO2.31,64 The latter, the mixed anhydride, can be
synthesized in solution and is stable in the solid phase and might
be present in minor amounts also in our experiments. However,
processes in matrix are dominated by kinetic and steric effects
rather than thermodynamics. Thus, unusual species are often
observed in matrix isolation spectroscopy. The band pair B2
shows a bromine isotopic splitting of 1.7 cm-1, which agrees
well with the calculated value for the stretching O-Br- mode
of OBrBrO2 of 1.6 cm-1. The oxygen isotopic splitting observed
for this band is 36.8 cm-1, which is also in good concordance
with the predicted value of 37.9 cm-1. OBrBrO2 further has
two intense normal modes in the region of the spectra shown
in the figures. The calculated frequency values for the symmetric
and asymmetric stretching -BrO2 modes are around 860 and
923 cm-1, which fit well with the recorded values for B2′ and
B2′′ (see Table 3). The predicted intensities for these modes
are noticeably lower than for the O-Br- stretching vibration,
a feature which is also observed in the recorded spectra. In
addition to these facts, the calculated bromine isotope splittings
for these modes are 1.3 and 2.5 cm-1, which is in good
agreement with the recorded values of 1.3 and 2.4 cm-1. The
calculated oxygen isotopic shifts of 42.9 and 40.9 cm-1 also
agree with the measured values of 41.8 and 40.1 cm-1. The
predicted pattern for the stretching O-Br- and -BrO2 modes
of OBrBrO2 in a mixture of 16O2/18O2 could not properly be
observed in our spectrum, since the low intensity of the bands
and the overlapping with other absorptions complicate the
analysis. We also like to mention that our calculated frequencies
for the symmetric and asymmetric stretching -BrO2 modes of
BrOBrO2, which was claimed to be a product of the reaction
of bromine with ozone at -50 to -60 °C,31 have similar values
to those from OBrBrO2 (see Table 2). Nevertheless, the
agreement of the calculations with the measured bands B2′ and
B2′′ is worse for BrOBrO2 than for OBrBrO2. Additionally, the
difference between the experimental and calculated values of
the symmetric and asymmetric stretching mode of -BrO2 would
point to different anharmonicities in the case of BrOBrO2, which
Six new pairs of bands (79/81Br) become visible after the
annealing process: B2, B4, B2′, B2′′, B7, and B7′. To elucidate
the origin of these bands, we have calculated the structures and
harmonic frequencies of different compounds which might be
formed in the reaction between the species previously identified
in the matrix and oxygen atoms, namely, BrO3, OBrBrO,
OBrOBrO, BrOBrO2, BrBrO2, BrBrO3, OBrBrO2, OBrOBrO2,
and BrOBrO3. We tentatively assign the set of bands B7 and
B7′ to the symmetric and asymmetric stretching -BrO2 modes
of BrBrO2. There are no previous experimental data for this
species, although several theoretical works have been done to
elucidate its structure and vibrational frequencies.24,25 In addition
to the small amount of BrBrO2 formed immediately after the
microwave discharge, probably produced by the reaction of Br2
and O2, a possible way to generate it during the annealing is
described by the following reaction:
BrBrO + O f BrBrO2
(8)
Thus, BrBrO is generated as well as consumed during the
thermal cycle. Our calculated frequencies for BrBrO2 deviate
less than 1% from the recorded values. The bromine isotopic
splittings observed are 1.4 and 2.5 cm-1 for B7 and B7′, which
is in very good concordance with our calculated values of 1.4
and 2.6 cm-1. The observed red-shifts introduced by 18-oxygen
are 43.1 and 41.0 cm-1 for B7 and B7′, in good agreement with
our calculated values of 44.4 and 42.0 cm-1, respectively. Our
predicted intensities (see Table 3) show a stronger absorption
for the asymmetric mode, as is observed in our spectra. To
evaluate the splitting caused by the break of the symmetry in
the stretching modes of -BrO2, an experiment with a mixture
of 16- and 18-oxygen was carried out. The spectrum recorded
presents a “forest of bands” in the region under study. Since
most of these bands do not add new information to our

6480 J. Phys. Chem. A, Vol. 110, No. 20, 2006
Ga´lvez et al.
renders this compound even less likely. Taking all these findings
into account, we assign the bands B2, B2′, and B2′′ to the
molecule OBrBrO2; nevertheless, some uncertainty remains.
Finally, the band group B4, for which additional information
has not been found in the literature, was tentatively assigned to
a BrBrO complex. The four bands belonging to B4 suggest a
matrix splitting as for BrBrO (B3); in the 18-oxygen spectra,
the bands are overlapping with the absorption of BrONO2,
changing the original shape of this band group. Nevertheless,
we do not have enough information to complete a conclusive
assignment for these bands, and further investigations are
needed.
References and Notes
(1) Molina, L. T.; Molina, M. J. J. Geophys. Res. 1986, 91, 14501-
14508.
(2) Solomon, S.; Garcia, R. R.; Rowland, F. S.; Wuebbles, D. J. Nature
(London) 1986, 321, 755-758.
(3) Barrie, L. A.; Bottenheim, J. W.; Schnell, R. C.; Crutzen, P. J.;
Rasmussen, R. A. Nature (London) 1988, 224, 134-141.
(4) Bottenheim, J. W.; Fuentes, J. D.; Tarasick, D. W.; Anlauf, K. G.
Atmos. EnViron. 2002, 36, 2535-2544.
(5) Haussmann, M.; Platt, U. J. Geophys. Res. 1994, 99, 25399-25413.
(6) Escribano, R.; Mosteo, R. G.; Go´mez, P. C. Can. J. Phys. 2001,
79, 597-609.
(7) Wayne, R. P.; Poulet, G.; Biggs, P.; Burrows, J. P.; Cox, R. A.;
Crutzen, P. J.; Hagman, G. D.; Jenkin, M. E.; Le Bras, G.; Moorgat, G. K.;
Platt, U.; Schindler, R. N. Atmos. EnViron. 1995, 29, 2675-2883.
(8) Yagi, K.; Williams, J.; Wang, N. Y.; Cicerone, R. J. Science 1995,
267, 1979-1981.
VI. Conclusion
(9) Wennberg, P. O.; Cohen, R. C.; Stimpfle, R. M.; Koplow, J. P.;
Anderson, J. G.; Salawitch, R. J.; Fahey, D. W.; Woodbridge, E. L.; Keim,
E. R.; Gao, R. S.; Webster, C. R.; May, R. D.; Toohey, D. W.; Podolske,
J. R.; Chan, K. R.; Wofsy, S. C. Science 1994, 266, 398-404.
(10) Platt, U.; Ho¨nninger, G. Chemosphere 2003, 52, 325-338.
(11) Chase, M. W., Jr. J. Phys. Chem. Ref. Data 1996, 25, 1069-1111.
(12) Escribano, R.; Ortega, I. K.; Mosteo, R. G.; Go´mez, P. C. Can. J.
Phys. 2004, 82, 998-1005.
(13) Seppelt, K. Acc. Chem. Res. 1997, 30, 111-113.
(14) Hwang, I.-C.; Kuschel, R.; Seppelt, K. Z. Anorg. Allg. Chem. 1997,
623, 379-383.
(15) Mu¨ller, H. S. P.; Cohen, E. A. J. Chem. Phys. 1997, 106, 8344-
8254.
(16) Mu¨ller, H. S. P.; Miller, C. E.; Cohen, E. A. J. Chem. Phys. 1997,
107, 8292-8302.
(17) Vayda, W. M. Proc. Indian Acad. Sci. 1937, 6A, 122-128.
(18) Vayda, W. M. Proc. Indian Acad. Sci. 1938, 7A, 321-326.
(19) Schwarz, R.; Schmeisser, M. Ber. Dtsch. Chem. Ges. 1937, 70B,
1163-1166.
(20) Rabinowitch, E.; Wood, W. C. Trans. Faraday Soc. 1936, 32, 907-
917.
(21) Zintl, E.; Rienacker, G. Ber. Dtsch. Chem. Ges. 1930, 63B, 1098-
1104.
(22) Dixon, D. A.; Parrish, D. D.; Herschbach, D. R.Faraday Discuss.
Chem. Soc. 1973, 55, 385-387.
(23) Parrish, D. D.; Herschbach, D. R. J. Chem. Soc. 1973, 95 (18),
6133-6134.
(24) Pacios, L. F.; Go´mez, P. C. THEOCHEM 1999, 467, 223-231.
(25) Guha, S.; Francisco, J. S. J. Phys. Chem. A 1997, 101, 5347-
5359.
(26) Li, Z.; Francisco, J. S. Chem. Phys. Lett. 2002, 354, 109-119.
(27) Guha, S.; Francisco, J. S. J. Phys. Chem. A 1998, 102, 6702-
6705.
We have studied the bromine oxides formed after microwave
discharge of a mixture of bromine and oxygen in different ratios.
Bromine oxides are relevant compounds which are involved in
the process of stratospheric and tropospheric ozone depletion
and are less stable than the analogous chlorine oxides. The
species formed in the reaction were stabilized in an excess of
argon and deposited on a cold surface kept at 6.5 K by a cryostat.
When the matrix is subjected to a thermal cycle, unreacted
oxygen atoms embedded in the matrix become mobile and react
with trapped bromine species, thus facilitating further reactions
inside the matrix. The products embedded in the argon matrix
were studied by infrared spectroscopy. Our method of produc-
tion yields a large variety of bromine oxides. Despite the
complications of dealing with spectra with a high number of
absorptions, this method generates species not previously
reported. Mixtures deliberately enriched in contaminant species,
isotopic exchange experiments, and ab initio calculations were
employed to assist in the assignment of the spectra. Our
calculated frequencies for the compounds proposed are in good
agreement with the observed bands. The calculated shifts for
the two natural bromine isotopes and the 18-oxygen experiments
also are in very good concordance with our recorded values.
These results support our assignments and conclusions.
The major infrared-active compound formed in the reaction
is BrO, which might be overlooked due to its very low
absorption intensity. In addition, OBrO, BrOO, BrBrO, and
BrOBr are generated after the microwave discharge. Because
of the high concentration of BrO, the self-reaction of this
compound takes place before it is immobilized in the matrix,
generating BrOBrO as the dominant product. BrOBr is not
observed in our spectra because of its low infrared activity, but
is most probably present since BrOBrO is also formed during
thermal cycles. Absorptions due to BrOOBr and OBrBrO, which
have also been proposed as intermediates of the BrO self-
reaction, are not observed. Only small amounts of BrBrO2 have
been found, which are probably due to the direct reaction of
Br2 and O2 in the course of the discharge and to the reaction of
BrBrO and oxygen atoms during annealing. Furthermore, the
hitherto unknown oxide OBrBrO2 is most probably present in
the matrixes after annealing, resulting from the reaction of
oxygen atoms with BrBrO2.
(28) Pak, C.; Xie, Y.; Schaefer, H. F., III Mol. Phys. 2003, 101 (1-2),
211-225.
(29) Pacios, L. F.; Go´mez, P. C. J. Phys. Chem. A 1997, 101, 1767-
1773.
(30) Li, Z.; Jeong, G.-R. Chem. Phys. Lett. 2001, 340, 194-204.
(31) Kuschel, R.; Seppelt, K. Angew. Chem. 1993, 105, 1734-1735.
(32) Leopold, D.; Seppelt, K. Angew. Chem. 1994, 106, 1043-1044.
(33) Li, Z.; Jeong, G.-R.; Person, E. Int. J. Chem. Kinet. 2002, 34, 430-
437.
(34) Tevault, D. E.; Walker, N.; Smardzewski, R. R.; Fox, W. B. J.
Phys. Chem. 1978, 82, 2728-2733.
(35) Tevault, D. E.; Smardzewski, R. R. J. Am. Chem. Soc. 1978, 100,
3955-3957.
(36) Loewenschuss, A.; Miller, J. C.; Andrews, L. J. Mol. Spectrosc.
1980, 81, 251-262.
(37) Ko¨lm, J.; Engdahl, A.; Schrems, O.; Nelander, B. Chem. Phys. 1997,
214, 313-319.
(38) Ko¨lm, J.; Schrems, O.; Beichert, P. J. Phys. Chem. A 1998, 102,
1083-1089.
(39) Li, Z. J. Phys. Chem. A 1999, 103, 1206-1213.
(40) Mauldin, R. L., III; Wahner, A.; Ravishankara, A. R. J. Phys. Chem.
1993, 97, 7585-7596.
Acknowledgment. This research was supported by Hoch-
schuljubila¨umsstiftung der Stadt Wien, project H-847/2005, and
the European Union, INTAS project 03-51-5698. A.L. acknowl-
edges stipends from the Ausseninstitut, TU Vienna and VEWIS-
TA. O.G. acknowledges financial support from the Spanish
Ministerio de Educacio´n y Ciencia, MEC. The authors would
like to thank Philipp Baloh for his help concerning the
experiments and spectra acquisition.
(41) Knight, G.; Ravishankara, A. R.; Burkholder, J. B. J. Phys. Chem.
A 2000, 104, 11121-11125.
(42) Kopitzky, R.; Grothe, H.; Willner, H. Chem.sEur. J. 2002, 8,
5601-5621.
(43) Pflugmacher, A. Z. Anorg. Allg. Chem. 1953, 273, 41-47.
(44) 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.;

Bromine Oxides
J. Phys. Chem. A, Vol. 110, No. 20, 2006 6481
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; Gaussian, Inc.: Wallingford, CT, 2004.
(52) Scheffler, D.; Grothe, H.; Willner, H.; Frenzel, A.; Zetzsch, C.
Inorg. Chem. 1997, 36, 335-338.
(53) Scheffler, D.; Willner, H. Inorg. Chem. 1998, 37, 4500-4506.
(54) Zoug, P.; Derecskei-Kovacs, A.; North, S. W. J. Phys. Chem. A
2003, 107, 888-896.
(55) Casper, B.; Lambotte, P.; Minkwitz, R.; Oberhammer, H. J. Phys.
Chem. 1993, 97, 9992-9995.
(56) Wilson, W. W.; Christe, K. O. Inorg. Chem. 1987, 26, 1573-1580.
(57) Burkholder, J. B.; Hammer, P. D.; Howard, C. J.; Maki, A. G.;
Thompson, G.; Chackerian, C., Jr. J. Mol. Spectrosc. 1987, 124, 139-161.
(58) Orlando, J. J.; Burkholder, J. B.; Bopegedera, A. M. R. P.; Howard,
C. J. J. Mol. Spectrosc. 1991, 145, 278-289.
(59) Maier, G.; Bothur, A. Z. Anorg. Allg. 1995, 621, 743-746.
(60) Odeh, I. N.; Nicoson, J. S.; Hartz, K. E.; Margerum, D. W. Inorg.
Chem. 2004, 43, 7412-7420.
(61) Nakata, M.; Somura, Y.; Takayanagi, M.; Tanaka, N.; Shibuya,
K.; Uchimaru, T.; Tanabe, K. J. Phys. Chem. 1996, 100, 15815-15820.
(62) St. Louis, R. V.; Crawford, B., Jr. J. Chem. Phys. 1965, 42, 857-
864.
(45) Drouin, B. J.; Miller, C. E.; Mu¨ller, H. S. P.; Cohen, E. A. J. Mol.
Spectrosc. 2001, 205, 128-138.
(46) Pacios, L. F.; Go´mez, P. C. Chem. Phys. Lett. 1998, 298, 412-
418.
(47) Thirugnanasambandam, P.; Mohan, S. Indian J. Phys. 1978, 52B,
173-178.
(48) Mu¨ller, H. S. P.; Miller, C. E.; Cohen, E. A. Angew. Chem., Int.
Ed. Engl. 1996, 35, 2129-2131.
(49) Miller, C. E.; Nicolaisen, S. L.; Francisco, J. S.; Sander, S. P. J.
Chem. Phys. 1997, 107, 2300-2307.
(63) Hallamasek, D.; Babka, E.; Kno¨zinger, E. J. Mol. Struct. 1997,
408/409, 125-132.
(64) Pascal, J. L.; Pavia, A. C.; Potier, J.; Potier, A. C. R. Acad. Sci.,
Ser. C 1974, 279 (1), 43-46.
(50) Lee, M. J. Phys. Chem. 1996, 100, 19847-19852.
(51) Kwabia Tchana, F.; Orphal, J.; Kleiner, I.; Rudolph, H. D.; Willner,
H.; Garcia, P.; Bouba, O.; Demaison, J.; Redlich, B. Mol. Phys. 2004, 102
(14-15), 1509-1521.