Infrared band intensities of bromine nitrate, BrONO2

Article abstract of DOI:10.1016/j.cplett.2008.04.089

The integrated band intensities at 296 K of gaseous bromine nitrate (BrONO₂) in the spectral range 500-2000 cm^-1 have been measured using Fourier-transform absorption spectroscopy. The absorption spectra were calibrated to absolute a

Full text of DOI:10.1016/j.cplett.2008.04.089

Chemical Physics Letters 458 (2008) 44–47
Contents lists available at ScienceDirect
Chemical Physics Letters
journal homepage: www.elsevier.com/locate/cplett
Infrared band intensities of bromine nitrate, BrONO2
b
b
Johannes Orphal a, , Mireille Morillon-Chapey , Guy Guelachvili
*
a Laboratoire Interuniversitaire des Systemes Atmospheriques (LISA), CNRS UMR 7583, Universite Paris-Est, 94010 Creteil Cedex, France
b Laboratoire de Photophysique Moleculaire (LPPM), CNRS UPR 3361, Universite de Paris-Sud, 91405 Orsay Cedex, France
a r t i c l e i n f o
a b s t r a c t
Article history:
The integrated band intensities at 296 K of gaseous bromine nitrate (BrONO2) in the spectral range
500–2000 cmÀ1 have been measured using Fourier-transform absorption spectroscopy. The absorption
spectra were calibrated to absolute absorption cross-sections using the integrated band strength of the
m3 band around 803 cmÀ1 (2.70 0.50 Â 10À17 cm moleculeÀ1) published previously. In this way, inte-
grated band intensities were determined for the first time for the m1 band around 1709 cmÀ1 (2.43
0.48 Â 10À17 cm moleculeÀ1), the m2 band around 1286 cmÀ1 (2.99 0.60 Â 10À17 cm moleculeÀ1), the
weak m4/m8 bands between 720 and 760 cmÀ1 (0.16 0.04 Â 10À17 cm moleculeÀ1) and the m5 band around
559 cmÀ1 (1.02 0.20 Â 10À17 cm moleculeÀ1). The results are compared with the infrared band
strengths of chlorine nitrate and with ab initio calculations.
Received 19 March 2008
In final form 21 April 2008
Available online 26 April 2008
Ó 2008 Elsevier B.V. All rights reserved.
1. Introduction
the integrated band intensities of the m1 and m2 fundamental bands
that are located above 1000 cmÀ1 [6,8] and to provide experimental
Bromine nitrate (BrONO2) is an important bromine and NOx res-
ervoir in the Earth’s atmosphere; it represents a large fraction of
the total bromine in the stratosphere and is also involved in tropo-
spheric chemistry [1–3]. It was first investigated in the 1960s [4]
but its atmospheric relevance was only discovered in the early
1970s [5].
data for comparisons with the recent ab initio calculations, unpub-
lished spectra of BrONO2 that were originally measured in 1993 in
the spectral region 500–2000 cmÀ1 [8] were calibrated to absolute
absorption cross-sections using the integrated band strength of
the m3 band around 803 cmÀ1 [7]. In this way, integrated band inten-
sities of the m1, m2, m4/m8 and m5 fundamentals were obtained.
Pure samples of bromine nitrate are rather difficult to synthe-
size and to handle, and decompose rapidly at ambient conditions.
Therefore, only very few spectroscopic studies of BrONO2 have
been reported in the past [5–9]. However, its structural parameters
have been obtained from electron diffraction [10], and several the-
oretical calculations of its structure, vibrational and electronic en-
ergy levels, dipole moment and dipole moment derivates have
been published [11–13]. Although the most important chemical
reactions of BrONO2 have been investigated, too (see [14,15] and
references therein), the infrared band intensities of BrONO2 have
not been measured, except for the m3 band around 803 cmÀ1 [7].
On the other hand, infrared band intensities are important refer-
ence data for studying chemical kinetics, for example using tropo-
spheric reaction chambers (see e.g. [16–18]). Furthermore, in a
recent study of the European Space Agency (ESA), BrONO2 was in-
cluded in the list of target species for future satellite missions focus-
ing on atmospheric chemistry [19]. However, the only infrared band
for which the absolute intensity is known [7] is located below
1000 cmÀ1, where windows made of CaF2 that are typically used in
such experiments are no longer transparent, which might be a prob-
lem for future laboratory and field studies. Therefore, to determine
2. Experimental
Pure samples of bromine nitrate were synthesized using the
reaction between pure ClONO2 and Br2 with ClONO2 in excess.
The ClONO2 was obtained using the method of Schmeisser [20],
i.e. the reaction between pure Cl2O and N2O5. Cl2O was synthesized
using the reaction of gaseous Cl2 with dry yellow HgO, and N2O5
was obtained by reaction of freshly synthesized HNO3 with P2O5.
All reactions were carried out under vacuum (10À2 mbar) avoiding
organic materials and metallic surfaces that could induce chemical
decomposition. Several low-temperature distillations were carried
out in order to separate reactants and products. Pure BrONO2 at
low temperatures forms yellow crystals and sublimates into the
gas phase when warming it up slowly to about À30 °C.
Absorption spectra of BrONO2 at different partial pressures
were recorded using the step-scan high-resolution Fourier-trans-
form spectrometer at LPPM. For the experiments described here,
a maximum spectral resolution of 0.03 cmÀ1 was used. The absorp-
tion cell was made of Pyrex glass and equipped with AgCl windows
and a Teflon (Young type) valve. Other window materials than AgCl
(NaCl, KBr) were observed to react with ClONO2 and BrONO2 form-
ing solid deposits (NaNO3, KNO3) on the windows’ surfaces. The
only impurity observed in the BrONO2 spectra was HNO3, either
* Corresponding author.
E-mail address: orphal@lisa.univ-paris12.fr (J. Orphal).
0009-2614/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.cplett.2008.04.089

J. Orphal et al. / Chemical Physics Letters 458 (2008) 44–47
45
from the hydrolysis of BrONO2 with residual H2O in the gas trans-
2.5x10-18
2.0x10-18
1.5x10-18
1.0x10-18
fer line (although no HOBr was observed in the absorption spectra)
or as residual impurity from the synthesis. Typically, the recording
time of one spectrum was about 30 min. All experiments were
carried out at room temperature (296 K). More details of the exper-
imental setup and procedure are given in Ref. [8]. The spectra were
corrected for the HNO3 absorption using a reference spectrum of
pure HNO3 vapour recorded under the same conditions as the
BrONO2 spectra. Note that the overlap of the bands of HNO3 and
of BrONO2 can be a source of systematic errors both in this work
and future quantitative studies using the infrared spectrum of
BrONO2.
5.0x10-19
0.0
The uncertainty of the absolute values for the integrated band
intensities of this study (20%) is mainly due to the uncertainty of
the measurement employed as Ref. [7], with an estimated uncer-
tainty of 0.50 Â 10À17 cm moleculeÀ1 at 296 K, corresponding to
18.5% of the integrated band intensity of the m3 band around
803 cmÀ1 (2.70 Â 10À17 cm moleculeÀ1). The relative band intensi-
ties of this study are more accurate, with an estimated uncertainty
of 5–10% except for the weak m4/m8 bands between 720 and
780
790
800
810
820
Wavenumber in cm-1
Fig. 2. The m3 band of BrONO2 corresponding to the Br–O stretching vibration. This
band has been previously observed by Burkholder et al. [7] and was used to dete-
rmine the absolute amount of BrONO2 in the absorption cell (containing a mixture
of BrONO2 and Br2).
760 cmÀ1
3. Results
An overview spectrum of the infrared bands of BrONO2 ob-
.
Table 1
Integrated band intensities of BrONO2 compared to ab initio calculations
Band Band
This work,
km molÀ1
Ref. [11],
km molÀ1
Ref. [12],
km molÀ1
Ref. [13],
km molÀ1
served in this study is presented in Fig. 1. The strong m3 band
around 803 cmÀ1 that has been measured previously by Burkholder
et al. [7] is shown in Fig. 2. The band contour is different from the
corresponding band in ClONO2 [8,21], and the integrated cross-sec-
tion at 296 K of the m3/m4 bands of ClONO2 (interacting by a strong
Fermi resonance [22]) is 3.16 Â 10À17 cm moleculeÀ1 [8,23,24], i.e.
about 17% larger. The principal rotational constants of bromine
nitrate are relatively small [10,11]; this and the nearly equal natu-
ral abundances of the two main bromine isotopomers, together
with the existence of several low-lying vibrational states that are
significantly populated at room temperature, lead to a highly con-
gested spectrum that is too dense to show rotationally resolved
structure.
In Table 1, the integrated band intensities of this study are com-
pared to the results of several recent ab initio calculations. One can
state that the agreement of the relative intensities (taking e.g. the
m2 band as reference) is relatively good for the m2, m4/m8 and m5 fun-
damentals, but that there are significant differences between the
experiment and the theoretical predictions for the m1 (antisymmet-
center,
cmÀ1
m1
m2
m3
m4/m8
m5
1709
1286
803
722
559
146 29
180 36
162 32
9.6 2.5
61 12
325
353
297
204
16
352
284
196
16
272
204
11
85
93
101
ric N–O stretching vibration) and m3 (NO2 angle bending vibration)
bands. Furthermore, the theoretical band intensities of the stron-
gest bands are too high compared to the experimental values, pos-
sibly due to the depletion of the population of the vibrational
ground state at room temperature, that needs to be taken into
account in the theoretical calculations, or because of anharmonic
effects. Also in contrast to the calculations, the m2 band of BrONO2
around 1286 cmÀ1 (see Fig. 3) is observed to be the strongest band
(having a similar shape as the corresponding band of chlorine
nitrate [25]) with an integrated absorption cross-section of
ν3
2.5x10-18
2.0x10-18
1.5x10-18
1.0x10-18
ν2
2.0x10-18
1.5x10-18
ν1
1.0x10-18
ν5
5.0x10-19
5.0x10-19
ν4/ν8
0.0
0.0
600
800 1000 1200 1400 1600 1800 2000
Wavenumber in cm-1
1260
1270
1280
1290
1300
1310
Wavenumber in cm-1
Fig. 1. Absorption cross-sections of the infrared bands of BrONO2. The spectral
resolution is 0.075 cmÀ1. The high-frequency structure at the baseline is from the
correction for residual H2O and HNO3 absorption, and due to noise.
Fig. 3. The m2 band of BrONO2 corresponding to the symmetric N–O stretching
vibration. This is the strongest infrared band of bromine nitrate, its shape and
intensity are similar to the corresponding band of chlorine nitrate [25].

46
J. Orphal et al. / Chemical Physics Letters 458 (2008) 44–47
2.99 Â 10À17 cm moleculeÀ1
.
This value is about 27% lower
of the m4 band of BrONO2 illustrates the strong coupling between
the m3 (809 cmÀ1) and m4 (780 cmÀ1) bands of ClONO2 [22] where
the m4 band is much stronger due to anharmonic interactions.
Concerning the atmospheric detection of BrONO2, Burkholder et
al. [7] have estimated that the m3 band of bromine nitrate around
803 cmÀ1 would have a peak absorbance of 0.04% at an air mass
factor of 10, assuming a column abundance of 2 Â 1013 cmÀ2 at
night. Since the present study shows that the m3 band has the stron-
gest peak cross-section (see Fig. 1) and since it is the only band that
falls into an atmospheric absorption window, it will indeed be very
difficult to detect BrONO2 in the atmosphere by infrared spectros-
copy, even using instruments with high sensitivity, but it might
neverthelesss be useful to look for BrONO2 in atmospheric spectra
with very high signal-to-noise ratio.
No evidence for the existence of the bromine nitrate isomer
BrOONO (bromine peroxynitrite) [30] was found, in agreement
with previous studies. BrOONO is predicted to have strong charac-
teristic absorption bands in the mid-infrared, with the strongest
band in the region 1800–1900 cmÀ1 [31]. However, no absorption
was observed in this range (see Fig. 1), so that the relative amount
of BrOONO in the spectra of the present study must have been less
than 2% of BrONO2.
than the integrated band intensity at 296 K of the m2 band of
ClONO2 centered around 1292 cmÀ1 (3.80 Â 10À17 cm moleculeÀ1
)
[23–26].
The m5 band of BrONO2 centered around 559 cmÀ1 (see Fig. 4)
has an integrated cross-section of 1.02 Â 10À17 cm moleculeÀ1
and has a band contour that is very similar to the m5 band of chlo-
rine nitrate (centered at 563 cmÀ1 [27] with an integrated cross-
section of 1.16 Â 10À17 cm moleculeÀ1 at 296 K [28]). The m1 band
of BrONO2 around 1709 cmÀ1 (see Fig. 5) with an integrated band
intensity of 2.43 Â 10À17 cm moleculeÀ1 carries less than 50% of
the intensity of the m1 band in ClONO2 [23–26], 5.68 Â 10À17
cm moleculeÀ1, in disagreement with the ab initio calculations,
however the origin of this discrepancy remains unresolved. Further
experimental and theoretical work is required to investigate this
issue. The m8 and m4 bands of BrONO2 are very weak, with a progres-
sion of Q-branches around 722–725 cmÀ1 that is possibly due to
coupling of the m8 and m9 (torsional) modes. As already observed
for ClONO2 [29], the P-branch of the m8 band is extremely weak
due to a rather large Herman-Wallis effect. Note that the weakness
4. Conclusions
8.0x10-19
6.0x10-19
4.0x10-19
2.0x10-19
0.0
In this Letter, the integrated band intensities at 296 K of the m1,
m2, m4/m8 and m5 fundamental bands of bromine nitrate have been
determined for the first time. The results have been compared with
recent ab initio calculations and with the band intensities of
chlorine nitrate, ClONO2. No absorptions that would indicate the
presence of the isomer BrOONO (bromine peroxynitrite) were
observed. In conclusion, the data of this work are useful for future
laboratory studies of atmospheric chemistry and may be interest-
ing for the detection of atmospheric BrONO2 using infrared
spectroscopy.
530
540
550
560
570
580
References
Wavenumber in cm-1
[1] D.J. Lary, J. Geophys. Res. D 101 (1996) 1505.
[2] D.J. Lary, M.P. Chipperfield, R. Toumi, T. Lenton, J. Geophys. Res. D 101 (1996)
1489.
[3] D.J. Lary, J. Geophys. Res. D 102 (1997) 21515.
[4] M. Schmeisser, L. Taglinger, Chem. Ber. 94 (1961) 1533.
[5] J.E. Spencer, F.S. Rowland, J. Phys. Chem. 82 (1978) 7.
[6] W.W. Wilson, K.O. Christe, Inorg. Chem. 26 (1987) 1573.
[7] J.B. Burkholder, A.R. Ravishankara, S. Solomon, J. Geophys. Res. D 100 (1995)
16793.
Fig. 4. The m5 band of BrONO2 showing noise due to the limited sensitivity of the
infrared detector in this region. The position and overall contour of this band are
very similar to the corresponding fundamental band in chlorine nitrate [27].
1.2x10-18
1.0x10-18
8.0x10-19
6.0x10-19
4.0x10-19
2.0x10-19
[8] J. Orphal, Ph.D. Thesis, Université de Paris-Sud (Orsay), 1995.
[9] B. Deters, J.P. Burrows, J. Orphal, J. Geophys. Res. D 103 (1998) 3563.
[10] B. Casper, P. Lambotte, R. Minkwitz, H. Oberhammer, J. Phys. Chem. 97 (1993)
9992.
[11] S. Parthiban, T.J. Lee, J. Chem. Phys. 109 (1998) 525.
[12] S. Parthiban, T.J. Lee, J. Chem. Phys. 113 (2000) 145.
[13] P. Zou, A. Derecskei-Kovacs, S.W. North, J. Phys. Chem. A 107 (2003) 888.
[14] R. Soller, J.M. Nicovich, P.H. Wine, J. Phys. Chem. A 105 (2001) 1416.
[15] R. Soller, J.M. Nicovich, P.H. Wine, J. Phys. Chem. A 106 (2002) 8378.
[16] R. Bröske, Ph.D. Thesis, Fachbereich Chemie, Bergische Universität
Gesamthochschule (Wuppertal), 2000.
–
[17] B.J. Finlayson-Pitts, L.M. Wingen, A.L. Sumner, D. Syomin, K.A. Ramazan, Phys.
Chem. Chem. Phys. 5 (2003) 223.
[18] J.-F. Doussin, B. Picquet-Varrault, R. Durand-Jolibois, H. Loirat, P. Carlier, J.
Photochem. Photobiol. A 157 (2003) 283.
[19] H. Kelder et al. (Ed.), Operational Atmospheric Chemistry Missions, Final
Report for ESA Project 17237/03/NL/GS (CAPACITY), ESA-ESTEC, 2005.
[20] M. Schmeisser, W. Fink, K. Brändle, Angew. Chem. 69 (1957) 780.
[21] J. Orphal, M. Morillon-Chapey, A. Diallo, G. Guelachvili, J. Phys. Chem. A 101
(1997) 1062.
0.0
1670 1680 1690 1700 1710 1720 1730 1740
-1
Wavenumber in cm
[22] J.-M. Flaud, J. Orphal, W.J. Lafferty, M. Birk, G. Wagner, J. Geophys. Res. D 107
(2002), doi:10.1029/2002JD002628.
[23] A. Goldman, C.P. Rinsland, J.-M. Flaud, J. Orphal, J. Quant. Spectrosc. Rad.
Transf. 60 (1998) 875.
[24] G. Wagner, M. Birk, J. Quant. Spectrosc. Rad. Transf. 82 (2003) 443.
[25] J. Orphal, M. Morillon-Chapey, G. Guelachvili, J. Geophys. Res. D 99 (1994)
14549.
Fig. 5. The m1 band of BrONO2 corresponding to the antisymmetric N–O stretching
vibration with dominant b-type structure, and a less pronounced Q-branch from
weak a-type transitions. Part of the high-frequency structure is from the correction
for residual HNO3 absorption which is rather strong in this region. The overlap of
the bands of BrONO2 and HNO3 in this region is a potential source for systematic
errors, both for this work and for future quantitative studies.

J. Orphal et al. / Chemical Physics Letters 458 (2008) 44–47
47
[26] J. Ballard, A.B. Johnston, M.R. Gunson, P.T. Wassell, J. Geophys. Res. D 93 (1988)
1659.
[29] J.-M. Flaud, W.J. Lafferty, J. Orphal, M. Birk, G. Wagner, Mol. Phys. 101 (2003)
1527.
[27] D.G. Johnson, et al., Geophys. Res. Lett. 23 (1996) 1745.
[28] J.A. Davidson, C.A. Cantrell, R.E. Shetter, A.H. McDaniel, J.G. Calvert, J. Geophys.
Res. 92 (1987) 10921.
[30] S. Kovacic, A. Lesar, M. Hodoscek, J. Koller, Chem. Phys. 323 (2006) 369.
[31] A. Lesar, S. Prebil, M. Mühlhäuser, M. Hodoscek, Chem. Phys. Lett. 368 (2003)
399.