A neutral xenon-containing radical, HXeO

Article abstract of DOI:10.1021/ja029024r

We report an open-shell species containing xenon, HXeO (²Σ), prepared by UV photolysis of H₂O/Xe or N₂O/HBr/Xe solid mixtures at 7 K and subsequent thermal mobilization of oxygen atoms at ≥30 K. The H-Xe stretching absorption of HXeO in solid Xe is at 1466.1 cm^-1, and it shifts to 1070.3 cm^-1 upon deuteration. The extensive ab initio calculations indicate that HXeO is intrinsically stable, owing to significant ionic and covalent contributions to its bonding. The formation of HXeO (²Σ) radicals in these experiments suggests extensive stabilization and thermal mobility of singlet (¹D) oxygen atoms in solid Xe and holds promises for the stability of the HKrO and HArO species. Copyright

Full text of DOI:10.1021/ja029024r

Published on Web 01/15/2003
A Neutral Xenon-Containing Radical, HXeO
Leonid Khriachtchev,*,† Mika Pettersson,† Jan Lundell,† Hanna Tanskanen,† Tiina Kiviniemi,†
Nino Runeberg,‡ and Markku Ra¨sa¨nen†
Laboratory of Physical Chemistry, P.O. Box 55, FIN-00014 UniVersity of Helsinki, Finland, and
CSC-Scientific Computing Ltd, P.O. Box 405, FIN-02101 Espoo, Finland
Received October 21, 2002 ; E-mail: Leonid.Khriachtchev@Helsinki.Fi
In 1962, Bartlett reported the first xenon-containing chemical
compound, Xe+[PtF6]-,1 and a year later a molecule with krypton,
KrF2, was identified.2 It took four decades until argon was
conquered by low-temperature synthesis of HArF,3 and serious
activity is obvious nowadays in the field of rare-gas chemistry.4-6
The characterization of HArF followed the extensive studies of
HRgY molecules where Rg is a rare-gas atom and Y is an
electronegative fragment (HXeBr, HXeOH, HKrCN, etc.).7 These
HRgY molecules are truly chemically bound species with closed-
shell electronic configuration.8 Here we report the identification
of an open-shell species containing xenon, HXeO (2Σ), prepared
by UV photolysis of H2O/Xe or N2O/HBr/Xe solid mixtures at 7
K and subsequent thermal mobilization of oxygen atoms at g30
K. The infrared (IR) absorption spectroscopic experiments are
supported by extensive ab initio calculations suggesting the intrinsic
stability of HXeO.
Figure 1. IR absorption bands formed after 193 nm photolysis and posterior
annealing of Xe matrixes containing water. The band at 1466 cm-1 marked
with an asterisk in the upper trace (annealing 45 K) is assigned to HXeO
radicals. This species is formed upon annealing at 35 K (the middle and
lower traces). Upon deuteration, the band of DXeO is at 1070 cm-1. The
spectra in the 4000-400 cm-1 spectral region were measured at 7 K with
The H2O/Xe matrixes (∼1/2000) were prepared and photolyzed
at 193 nm as described elsewhere.9 As a result of the irradiation,
the matrix mainly contains O and H atoms, OH radicals, and
residual H2O molecules, full spectroscopic evidence for all of them
being available.9-13 In the previous studies of HXeY molecules,
the photolyzed HY/Xe samples were usually annealed at 40-48
K, which mobilized H atoms and led to diffusion-controlled
formation of the HXeY molecules in reactions of H atoms with
neutral Xe-Y centers.14 The result of such “high-temperature”
annealing is presented by the upper trace in Figure 1 for a
photolyzed H2O/Xe sample. In agreement with the previous reports,
the obtained IR spectrum shows formation of neutral HXeH and
HXeOH molecules.7,9,10
In addition to the known spectral features, we noticed systematic
annealing-induced build-up of an unknown band at 1466.1 cm-1
(for H216O precursor), and this is the key observation of the present
work. It was found that this band appears practically alone already
under annealing at 30-35 K (see the middle trace in Figure 1),
which is characteristic for selective mobilization of atomic oxygen
in solid Xe.15 Moreover, the 1466.1 cm-1 band partially decreases
under mobilization of H atoms at higher temperatures (g38 K),
indicating reaction of the absorber with mobile H atoms. No
analogous band with a normal matrix shift is seen upon photolysis
and annealing of H2O/Kr samples, which suggests participation of
a Xe atom in this species. Upon deuteration, the analogous band is
at 1070.3 cm-1 (H/D ratio of 1.370) showing direct participation
of a hydrogen atom in the absorbing vibration. It was verified that
this absorber does not form from the natural matrix impurities (O2
and N2). By analyzing all possible candidates for this absorber and
on the basis of our experience in preparation of HRgY molecules,
a Nicolet 60 SX FTIR spectrometer using resolution of 1 cm-1
.
we experimentally assign the band at 1466.1 cm-1 (1070.3 cm-1
)
to the H-Xe (D-Xe) stretching mode of the HXeO (DXeO)
radical.
The assignment of HXeO was supported by experiments with
18O-substituted water. Small but definite blue shifts of the HXeO
bands were found for the heavier oxygen isotope in numerous
experiments without any exclusion. Summarizing the data, the band
centers are shifted by (0.27 ( 0.04) cm-1 for H218O and (0.19 (
0.04) cm-1 for D218O. In the same experiments, the oxygen-free
rare-gas molecules such as HXeD have absorption frequencies
independent of oxygen isotope substitution (with accuracy of (0.02
cm-1), which makes the reported oxygen-isotope shifts of HXeO
confident. In agreement with our assignment, these data show that
oxygen is not involved in the absorbing mode directly but that it
participates in the species. The observed oxygen-isotope shift is
“anomalous”, meaning blue shift for the larger reduced mass. We
have recently found experimentally similar oxygen-isotope shifts
for HXeOH and HXeOD molecules and described the effect
theoretically in terms of anharmonic coupling between the normal
modes.16 The low sensitivity of our apparatus in the far-IR spectral
range (<400 cm-1) did not allow us to detect the Xe-O stretching
vibration that should be more sensitive to oxygen isotope substitu-
tion.
According to the proposed image, HXeO is formed in reactions
of mobile oxygen atoms with H-Xe centers, i.e., the process needs
H and O atoms to be present in a Xe lattice. To verify this model,
we performed a series of experiments on Xe matrixes doped with
HBr and N2O. Irradiation at 193 nm efficiently decomposes both
HBr and N2O, producing H and O atoms from the separate sources.
Upon annealing of the photolyzed matrixes at 30-35 K, the 1466.1
† University of Helsinki.
‡ CSC-Scientific Computing.
9
1454
J. AM. CHEM. SOC. 2003, 125, 1454-1455
10.1021/ja029024r CCC: $25.00 © 2003 American Chemical Society

C O M M U N I C A T I O N S
cm-1 band assigned to HXeO grew similarly to the experiments
with water. It was shown that the formation of the 1466.1 cm-1
band requires simultaneous doping with both HBr and N2O. The
additional species seen in the IR absorption spectra after annealing
are HXeH, HXeBr, HO2, and O3. These experiments with HBr/
N2O/Xe samples strongly confirm that the species absorbing at
1466.1 cm-1 is formed from H and O atoms.
2.31 Å.23 The exact boundaries between chemical compounds and
van der Waals complexes are not absolutely clear, and accurate
analysis is always needed.4 It is worth noting in this respect that
the computational Xe-F distances are 1.904 Å in XeF+ and 2.146
Å in HXeF,24 which are clearly shorter than in the XeF complex.
The identification of HXeO (2Σ) suggests that analogous radicals
with other rare-gas elements should be searched for. Both HKrO
and HArO are computationally stable species. The H-Ar distance
in HArO is 1.343 Å at the CCSD(T) level of theory, which is
slightly shorter than the corresponding value of 1.355 Å for HArF,24
and the Ar-O distance is 2.021 Å. In HKrO, the H-Kr and Kr-O
distances are 1.490 and 2.087 Å [CCSD(T)]. Direct preparation of
HKrO and HArO using the method described here is problematic
due to the questionable stability of O (1D) atoms in Kr and Ar
solids. However, their probable existence is intriguing, especially
for HArO that might be found in the atmospheres.
In conclusion, the presented data allow us to claim the identifica-
tion of an open-shell triatomic chemical compound containing
xenon, HXeO (2Σ). The HXeO species is prepared by using UV
photolysis of H2O/Xe or N2O/HBr/Xe solid mixtures at 7 K and
subsequent thermal mobilization of oxygen atoms at g30 K. The
calculated stabilization barrier is 213 meV at the MRCISD/aug-
cc-pVTZ level of theory. The HXeO (2Σ) radical forms presumably
in the H + Xe + O (1D) reaction, and it is higher in energy (by
∼1 eV) than the H + Xe + O (3P) triad.
The experimental assignment of HXeO is fully supported by ab
initio quantum chemical calculations performed at the MP2/LJ18/
6-311++G(2d,2p) and CCSD(T)/LJ18/6-311++G(2d,2p) levels of
theory.7 HXeO is computationally a true energy minimum exhibiting
a linear structure with H-Xe and Xe-O distances of 1.652 and
2.171 Å (MP2) and 1.694 and 2.153 Å [CCSD(T)], respectively.
These bond lengths are slightly shorter than the corresponding
distances in HXeOH, which are 1.740 and 2.218 Å at the same
CCSD(T) level.10 The H-Xe and Xe-O distances in HXeO are
close to the corresponding values in XeH+ (1.603 Å)17 and in XeO
(1Σ+) (2.06 Å).18 The natural bond orbital analysis indicates a strong
ionic character of HXeO, with a positive charge on the Xe atom
(+0.88) and a negative charge on the O atom (-0.90). The H-Xe
stretching frequency is 1681 cm-1 at the CCSD(T) level of theory.
The agreement with the experimental value of 1466 cm-1 is very
reasonable, taking into account similar overestimates for H-Rg
stretching frequencies of other known HRgY molecules by this
harmonic theory.7 The bending modes are computationally at 580
cm-1, and the Xe-O stretching mode is at 382 cm-1. Estimated at
the MP2 theory level, the intensity of the H-Xe stretching mode
is 309 km/mol, which is ∼5 times smaller than that of HXeOH.
This would suggest that the concentrations of HXeO and HXeOH
are similar in the H2O/Xe experiments (see Figure 1), and the
formation of HXeO makes an important product channel upon
annealing of photolyzed H2O/Xe samples. The MP2 calculations
yield a much smaller intensity for the bending modes (6 km/mol),
which explains their invisibility in the experiments similarly to the
situation with HXeOH,9,10 and the Xe-O stretching has an intensity
of 185 km/mol.
Acknowledgment. We thank V. Feldman for helpful discus-
sions. The Academy of Finland supported this work. CSC (Espoo,
Finland) is thanked for computing resources.
References
(1) Bartlett, N. Proc. Chem. Soc. 1962, 218.
(2) Turner, J. J.; Pimentel, G. C. Science 1963, 140, 974-975.
(3) Khriachtchev, L.; Pettersson, M.; Runeberg, N.; Lundell, J.; Ra¨sa¨nen, M.
Nature (London) 2000, 406, 874-876.
(4) Pyykko¨, P. Science 2000, 290, 64-65.
(5) Christe, K. O. Angew. Chem., Int. Ed. 2001, 40, 1419-1421.
(6) Li, J.; Bursten, B. E.; Liang, B.; Andrews, L. Science 2002, 295, 2242-
2245.
The dissociation of HXeO (2Σ) to the atomic products correlates
with the singlet oxygen (1D) potential surface meaning that the H
+ Xe + O (1D) reaction can lead to HXeO which is 1.52 eV lower
in energy [CCSD(T)]. In contrast, the H + Xe + O (3P) triad is
much lower in energy than HXeO (by ∼1.01 eV), which suggests
that the O (3P) atoms are energetically incapable of forming the
HXeO radical. The diffusion-controlled formation of HXeO at 30-
35 K indicates the presence of O (1D) atoms in the Xe lattice. It
follows that solid Xe can provide storage for the O (1D) atoms,
and they move upon thermal annealing at g30 K. This agrees with
the discussions on favorable stabilization of the O (1D) atoms in
Xe lattice compared with O (3P).19
(7) Lundell, J.; Khriachtchev, L.; Pettersson, M.; Ra¨sa¨nen, M. Low Temp.
Phys. 2000, 26, 680-690.
(8) Berski, S.; Silvi, B.; Lundell, J.; Noury, S.; Latajka, Z. In New Trends in
Quantum Systems in Chemistry and Physics; Maruani, J. et al., Eds.;
Kluwer: Dordrecht, 2001; Vol. 1, pp 259-279.
(9) Khriachtchev, L.; Tanskanen, H.; Pettersson, M.; Ra¨sa¨nen, M.; Ahokas,
J.; Kunttu, H.; Feldman, V. J. Chem. Phys. 2002, 116, 5649-5656.
(10) Pettersson, M.; Khriachtchev, L.; Lundell, J.; Ra¨sa¨nen, M. J. Am. Chem.
Soc. 1999, 121, 11904-11905.
(11) Goodman, J.; Brus, L. E. J. Chem. Phys. 1977, 67, 4858-4865.
(12) Creuzburg, M.; Koch, F.; Wittl, F. Chem. Phys. Lett. 1989, 156, 387-
391.
(13) Lawrence, W. G.; Apkarian, V. A. J. Chem. Phys, 1992, 97, 2229-2236.
(14) Pettersson, M.; Nieminen, J.; Khriachtchev, L.; Ra¨sa¨nen, M. J. Chem.
Phys. 1997, 107, 8423-8431.
(15) Danilychev, A. V.: Apkarian, V. A. J. Chem. Phys. 1993, 99, 8617-
8627.
To check the energetic stability of HXeO with respect to the H
+ XeO (1Σ+) channel, computations at the valence MRCISD/aug-
cc-pVTZ level of theory were performed. It was found that HXeO
is lower in energy than H + XeO (1Σ+) by 137 meV, and it is
additionally stabilized by 76 meV due to crossing of the neutral
and charge-transfer potential surfaces, which means a stabilization
barrier of 213 meV (20.6 kJ/mol). This energetics qualitatively
corresponds to the computational results on HArF.20
(16) Khriachtchev, L.; Lundell, J.; Pettersson, M.; Tanskanen, H.; Ra¨sa¨nen,
M. J. Chem. Phys. 2002, 116, 4758-4761.
(17) Rogers, S. A.; Brazier, C. R.; Bernath, P. F. J. Chem. Phys. 1987, 87,
159-162.
(18) Yamanishi, M.; Hirao, K.; Yamashita, K. J. Chem. Phys. 1998, 108, 1514-
1521.
(19) Danilychev, A. V.; Apkarian, V. A. J. Chem. Phys. 1994, 100, 5556-
5566.
(20) Runeberg, N.; Pettersson, M.; Khriachtchev, L.; Lundell, J.; Ra¨sa¨nen M.
J. Chem. Phys. 2001, 114, 836-841.
(21) Stein, L.; Norris, J. R.; Downs, A. J.; Minihan, A. R. J. Chem. Soc., Chem.
Commun. 1978, 502-504.
HXeO is an open-shell species (2Σ), which makes it rather
exceptional among neutral rare-gas compounds. The known open-
(22) Filippi, A.; Troiani, A.; Speranza, M. J. Phys. Chem. A 1997, 101, 9344-
9350.
(23) Aquilanti, V.; Luzzatti, E.; Pirani, F.; Volpi, G. G. J. Chem. Phys. 1988,
89, 6165-6175.
shell species with rare-gas atoms are mainly radical cations, such
+ 21
as Xe2
and XeO+.22 A good candidate for a neutral open-shell
(24) Lundell, J.; Chaban, G. M.; Gerber, R. B. Chem. Phys. Lett. 2000, 331,
308-316.
species containing rare-gas atoms seems to be the XeF complex,
which has a dissociation energy of 145 meV and a bond length of
JA029024R
9
J. AM. CHEM. SOC. VOL. 125, NO. 6, 2003 1455