Infrared spectroscopic and theoretical investigations of the OUF 2 and OThF2 molecules with triple oxo bond character

Article abstract of DOI:10.1021/ic3009128

The terminal oxo species OUF₂ and OThF₂ have been prepared via the spontaneous and specific OF₂ molecule reactions with laser ablated uranium and thorium atoms in solid argon and neon. These isolated molecules are characte

Full text of DOI:10.1021/ic3009128

Article
pubs.acs.org/IC
Infrared Spectroscopic and Theoretical Investigations of the OUF2
and OThF2 Molecules with Triple Oxo Bond Character
Yu Gong,† Xuefeng Wang,†,‡ Lester Andrews,*,† Tobias Schloder,§ and Sebastian Riedel§
̈
†Department of Chemistry, University of Virginia, Charlottesville, Virginia 22904-4319, United States
‡Department of Chemistry, Tongji University, Shanghai, P.R. China 200092
§Institut fur Anorganische und Analytische Chemie, Albert-Ludwigs Universitat Freiburg, Albertstrasse 21, D-79104 Freiburg i. Br.,
̈
̈
Germany
S
* Supporting Information
ABSTRACT: The terminal oxo species OUF2 and OThF2
have been prepared via the spontaneous and specific OF2
molecule reactions with laser ablated uranium and thorium
atoms in solid argon and neon. These isolated molecules are
characterized by one terminal M-O and two F−M−F (M = U
or Th) stretching vibrational modes observed in matrix
isolation infrared spectra, which are further supported by
density functional frequency calculations and CASPT2 energy
and structure calculations. Both molecules have pyramidal
structures with singlet (Th) and triplet (U) ground states. The
molecular orbitals and metal−oxygen bond lengths for the OUF2 and OThF2 molecules indicate triple bond character for the
terminal oxo groups, which are also substantiated by NBO analysis at the B3LYP level and by CASPT2 molecular orbital
calculations. Dative bonding involving O2p → Th6d and Udf interactions is clearly involved in these oxoactinide difluoride
molecules. Finally, the weak O−F bond in OF2 as well as the strong U−O, U−F and Th−O, Th−F bonds make reaction to form
the OUF2 and OThF2 molecules highly exothermic.
reports can be found for molecules with terminal ThO
groups.22
INTRODUCTION
■
Molecules with terminal metal oxo ligands are considered to be
important intermediates in a number of chemical and
biochemical reactions.1−3 Considerable experimental effort
has been devoted to the synthesis and characterization of
these molecules especially those containing late transition metal
centers, which are helpful in understanding their roles in the
corresponding catalytic reactions.4−7 Considering the impor-
tance of terminal oxo species in transition metal chemistry,
their actinide analogues are also interesting because applications
of uranium and thorium in a wide range of catalytic reactions
have been the subject of much scientific research.8−11 As the
most widely studied actinide element, uranium is known to
Besides experimental studies focused on synthetic methods,
reactions of uranium and thorium atoms with simple precursor
molecules in cryogenic matrixes have also yielded several
products with terminal oxo ligands.23−26 In these cases
structural identification can be determined by the agreement
of experimental and theoretical vibrational frequencies for a
specific calculated structure. Our recent studies on actinide and
NF3 reaction products provided experimental evidence for the
formation of terminal nitride species as the final products.27,28
Hence, similar reaction product molecules with terminal oxo
ligands are expected to be formed when OF2 is used as the
reactant, since the OF bond is even weaker than the NF bond
in NF3, and much weaker than the OH bond in H2O.23,29 In
this paper, we report the matrix infrared spectra of two
oxoactinide difluoride molecular species with terminal oxo
groups, namely, OUF2 and OThF2. These product molecules
are formed via the spontaneous, specific reactions of laser
ablated uranium or thorium atoms and OF2 in argon and neon
matrixes. Three bond stretching mode absorptions are observed
in the infrared above 400 cm−1 for each molecule, and they
compare favorably with density functional calculations of
vibrational frequencies. We also include a CASPT2 bonding
+
form a number of uranyl compounds containing UO2 and
2+
UO2 fragments, the structures of which have been fully
characterized by experimental and theoretical methods.12−15
Compared with the rich studies on these uranyl species
containing uranium oxygen multiple bonds, molecules with
terminal mono oxo groups are not as well structurally
characterized,16−18 which is partly related to their high
reactivity toward the formation of bridged oxo species.16
−
Although species like UOCl5 and UOF4 have also been
characterized to have terminal UO groups in their solid
states,19,20 they have been classified as uranyl analogues because
of preference for the almost linear X−U−Y arrangement
around uranium,16 where the inverse trans influence plays an
important role.13,21 For thorium, however, few structural
Received: May 4, 2012
Published: May 24, 2012
© 2012 American Chemical Society
6983
dx.doi.org/10.1021/ic3009128 | Inorg. Chem. 2012, 51, 6983−6991

Inorganic Chemistry
Article
analysis which describes triple bonding in these early actinide
oxo subunits.
EXPERIMENTAL AND THEORETICAL METHODS
■
The experimental apparatus and procedure for preparation and
characterization of metal atom reaction product molecules in excess
argon at 4 K have been described previously.30 The Nd:YAG laser
fundamental (1064 nm, 10 Hz repetition rate with 10 ns pulse width)
was focused onto a freshly cleaned uranium or thorium (Oak Ridge
National Laboratory, high purity) target mounted on a rotating rod.
Laser-ablated uranium or thorium atoms were codeposited with 3−4
mmol of argon (Matheson, research) containing 1.0% OF2 (Ozark-
Mahoning) onto a CsI cryogenic window for 60 min. The 18OF2
sample (91% 18O enriched) was synthesized and kindly provided by
Arkell and co-workers.31 Both OF2 and 18OF2 samples were used
without further purification in a stainless steel vacuum manifold. FTIR
spectra were recorded at 0.5 cm−1 resolution and 0.1 cm−1 frequency
accuracy on a Nicolet 750 FTIR instrument with a HgCdTe range B
detector. Matrix samples were annealed at different temperatures and
cooled back to 4 K for spectral acquisition. Selected samples were
subjected to broadband photolysis by a medium-pressure mercury arc
street lamp (Philips, 175W) with the outer globe removed.
Figure 1. Infrared spectra of laser-ablated uranium atoms and OF2
reaction products in solid argon at 4 K: (a) U + 1.0% OF2 deposition
for 60 min; (b) after annealing to 20 K; (c) after λ > 220 nm
irradiation; (d) after annealing to 30 K; (e) after annealing to 35 K.
reactions of uranium with mixed F2 and O2 in argon.41 New
product bands were observed at 834.8, 522.2, and 487.2 cm−1,
which increased by ∼10% during the first annealing to 20 K
(Figure 1, trace b). Subsequent broad band irradiation had
essentially no effect on these new bands while the uranium
fluoride absorptions increased (Figure 1, trace c). Further
sample annealing to 30 and 35 K decreased the 834.8, 522.2,
and 487.2 cm−1 absorption set (Figure 1, traces d and e), and
the UF6 band at 619 cm−1 became the dominant absorption in
the infrared spectra (not shown).41 In addition, spectra are
compared in Supporting Information, Figure S1 for U atom
reaction products with OF2 and with F2.
Complementary density functional theory (DFT) calculations were
performed using the Gaussian 09 program system.32 The hybrid
B3LYP density functional was employed primarily in our calculations
along with the M06 functional.33 The 6-311+G(d) basis set was used
for oxygen and fluorine atoms, and the 60 electron core SDD
pseudopotentials were employed for thorium and uranium atoms.34 All
of the geometrical parameters were fully optimized, and the harmonic
vibrational frequencies were obtained analytically at the optimized
structures. The nature of the terminal oxo bonds in the OUF2 and
OThF2 molecules were further analyzed using NBO 3.1 as
implemented in Gaussian 09.35 After the DFT work, benchmark
CASPT2 calculations were done for OUF2 [using (6,6) and (8/8)
CAS] and for OThF2 [(6,6) CAS].36 The MOLPRO program
package36e used for the latter calculations employs a slightly different
code (RS2) and gives slightly different results from the MOLCAS
program of Roos et al.
Infrared spectra from the reactions of laser-ablated thorium
atoms and OF2 are shown in Figure 2. Weak absorptions
RESULTS AND DISCUSSION
■
Reactions of uranium or thorium atoms with OF2 were studied
using different OF2 concentrations as well as laser energies, and
only the best experimental results using 1.0% OF2 are presented
here. A common absorption due to the OF free radical was
observed at 1028.1 cm−1 after sample deposition in all
experiments,37 and this absorption intensity depended strongly
on the laser energy employed in the experiment. Broad band
irradiation (λ > 220 nm) increased the intensity of the OF
radical band, which decreased when the sample was annealed,
as observed previously.37c In addition, weak absorptions for
CF4, CO, CO2, O3, and FOO38 were also present in the matrix
infrared spectra. Very weak CO and CO2 contaminant
absorptions are always observed in these experiments, and
CF4 is a common impurity in fluorine and OF2 samples from
the reactions of highly oxidative F2 or OF2 with residual carbon
in the steel sample cylinders.
Figure 2. Infrared spectra of laser-ablated thorium atoms and OF2
reaction products in solid argon at 4 K: (a) Th + 1.0% OF2 deposition
for 60 min; (b) after annealing to 20 K; (c) after λ > 220 nm
irradiation; (d) after annealing to 30 K; (e) after annealing to 35 K.
Infrared Spectra. Figure 1 shows infrared spectra from
codeposition of laser-ablated uranium atoms with 1.0% OF2 in
+
argon. In addition to weak UO2 and UO2 absorptions as well
previously assigned to ThO and ThO2 molecules were observed
right after sample deposition,42 and their amounts are too small
to observe any possible reaction products. Three new product
absorptions at 806.6, 511.1, and 482.3 cm−1 increased
substantially when the sample was annealed to 20 K because
of the lower laser energy used for Th relative to that for U in
the uranium experiments. But these bands decreased upon
annealing to 30 and 35 K (Figure 2, traces b, d and e). No
as bands due to UFx species,39,40 several other new metal
dependent (i.e., these bands depended on the particular metal
source ablated) absorptions were observed after sample
deposition. These were not present in U with O2 or F2 argon
matrix experiments. Weak 940.5 and 871.7 cm−1 absorptions
(not shown) were assigned to the antisymmetric O−U−O
stretching modes of the UO2F2 and UO2F molecules from
6984
dx.doi.org/10.1021/ic3009128 | Inorg. Chem. 2012, 51, 6983−6991

Inorganic Chemistry
Article
change was found for the 806.6, 511.1, and 482.3 cm−1
absorption set upon broad band irradiation while the ThF4
absorption increased (Figure 2, trace c).43 The latter two
absorptions are clearly in the region for Th−F stretching
vibrations.
S2 with the product absorptions given in Table 1. Absorptions
due to uranium and thorium oxide molecules were observed as
well.42b,44 It is significant to note that these neon matrix
counterparts are blue-shifted slightly more than the typical
amount (here about 20 cm−1) from the argon matrix values
described above.45 These argon−neon shifts can be compared
to those observed for the CH2-LnF2 complexes in recent work
from this laboratory.46 Since the Ln-F bonds are almost purely
ionic, the argon−neon matrix shifts are large (13−16 cm−1) for
the ionic Ln-F bond stretching modes and appropriately small
(∼1 cm−1) for the more covalent Ln-C bond stretching modes.
Hence, the argon−neon matrix shifts for the subject oxo species
verify experimentally the polar character of these bonds which
are revealed by the natural (NPA) atomic charges from the
DFT calculations [O(−0.81) U(2.08) F(−0.63 × 2) and
O(−1.03) Th(2.38) F (−0.68 × 2)] (see Supporting
Information, Table S1).
To facilitate product identifications, experiments with 18OF2
sample (91% enriched) were also carried out, which are
diagnostic for identifying the UO and ThO stretching modes in
the new molecules. Additionally, the metal fluoride stretching
modes might also exhibit very small oxygen-18 shifts because of
coupling with the metal oxide stretching modes. Infrared
spectra from the reactions of uranium or thorium with 1.0%
18OF2 in argon are shown in Figure 3 (traces c and d), and the
newly observed product absorptions are listed in Table 1.
The additional observation of the best yield of the UO
diatomic molecule in solid neon here from the single oxygen
OF2 precursor at 899.2 cm−1 is in agreement with the earlier
observation at 889.5 cm−1 from dioxygen reactions.44 The
U18O counterpart is observed here at 841.6 cm−1 with the
U16O/U18O frequency ratio of 1.05656. We maintain that UO
is trapped in different electronic states in solid neon and solid
argon where the vibrational mode absorption is 819.8 cm−1.39
OMF2 (M = U, Th) Molecules. The 834.8, 522.2, and 487.2
cm−1 absorptions exhibited identical behavior throughout the
experiments with uranium, suggesting that they arise from
different vibrational modes of the same new molecule. The
834.8 cm−1 band shifted to 790.5 cm−1 with the 16O/18O
isotopic frequency ratio of 1.0560. Both the isotopic frequency
ratio and band position (1.0571 and 819.8 cm−1 for UO in solid
argon39) show that the 834.8 cm−1 band is due to the U−O
stretching mode of a new molecule (the smaller 1.0529 isotopic
frequency ratio is the signature of an antisymmetric O−U−O
vibration.39). In addition the 18OF2 sample contains 9% 16OF2,
and the relative intensities of the 16OF2 and 18OF2 isotopic
products in the 18O-enriched sample provide information on
the oxygen stoichiometry of the new product: with a single O
atom the relative 16/18 product band intensities would be
(0.09)/(0.91) = 1/9, but with two O atoms, the relative 16,16/
18,18 band intensities would be (0.09)2/(0.91)2 or 1/102. The
measured integrated relative 16/18 uranium product band
intensity is 1/5 where the 18O-enriched product area is difficult
to measure as a shoulder on the very intense 18OF2 precursor
band. However, the measured integrated relative 16/18
thorium product band intensity is 1/12 [here the 16O-product
Figure 3. Infrared spectra of laser-ablated uranium and thorium atom
and isotopically substituted OF2 reaction products in solid argon at 4
K. All spectra were recorded after sample deposition followed by 20 K
annealing: (a) U + 1.0% 16OF2; (b) Th + 1.0% 16OF2; (c) U + 1.0%
18OF2 (91% enriched); (d) Th + 1.0% 18OF2 (91% enriched).
Complementary neon matrix experiments were performed to
ascertain the magnitude of the argon to neon matrix shifts and
thus determine the influence of matrix atoms on the trapped
species. The successful trapping of reactive molecules in solid
neon made by reactions of laser ablated metal atoms
necessitates lower laser energy and less heat load to the
condensing sample because the 4 K substrate is much closer to
the freezing point of neon (24.5 K) than that of argon (84.0 K).
Hence, the product yield is typically less in the neon matrix,
and such spectra are shown in Supporting Information, Figure
Table 1. Vibrational Frequencies Observed in Solid Argon and Solid Neon [in brackets] and Calculated Frequencies (Infrared
a
Intensities) for the OUF2 and OThF2 Molecules Prepared from the 16OF2 and 18OF2 Reagents
16OF2
18OF2
b
mode
U−O str.
sym. F−U−F
antisym. F−U−F
Th−O str.
obsd.
calc.
obsd.
calc.
OUF2
834.8 [855.0]
522.2 [529.0]
487.2 [507.6]
806.6 [mask]
511.1 [522.4]
482.3 [500.8]
862.4 (217)
521.3 (126)
498.4 (185)
825.9 (235)
513.1 (126)
484.6 (202)
790.5 [809.5]
522.1 [528.8]
487.2 [507.6]
763.9 [782.7]
510.8 [522.0]
816.3 (194)
521.2 (126)
498.3 (184)
781.9 (209)
512.9 (127)
484.5 (201)
OThF2
sym. F−U−F
antisym. F−U−F
482.1 [500.5]
a
b
Only frequencies above 400 cm−1 are listed. Numbers in parentheses are infrared intensities (km/mol). Our calculations employed the B3LYP
density functional, and for comparison the M06 functional gave 859.2(260), 512.6 (129), and 483.8 (207) for OThF2. Ref 47a gave 838.1 cm−1
(PBE) and 889.0 cm−1 (PBE0) for the U−O stretching mode of OUF2 using SC-RECP basis set for uranium.
6985
dx.doi.org/10.1021/ic3009128 | Inorg. Chem. 2012, 51, 6983−6991

Inorganic Chemistry
Article
intensity is from Figure 3 (spectrum d minus spectrum c) since
the 18O-enriched sample contains weak impurity bands in this
region]. Both of these measured integrated relative 16/18
product band intensities are much closer to the 1/9 ratio
expected for a single O atom containing product, in agreement
with the isotopic frequency ratio and band position.
B3LYP calculations were also performed on the OUF4
molecule offered as a possible assignment in our earlier
experiments.41 Consistent with most previous investigations,47
a C3v structure is calculated to be most stable for the OUF4
molecule. Frequency calculations reveal four infrared active
modes above 400 cm−1: the axial U−O and U−F stretches at
918.7 and 619.0 cm−1 as well as the equatorial U−F stretching
modes at 554.6 and 533.6 cm−1 with relative intensities of
160:87:432:20. Clearly, assignment of the new 834.8, 522.2,
and 487.2 cm−1 absorptions to the OUF4 molecule is not a
reasonable possibility.
The new 522.2 and 487.2 cm−1 absorptions are observed in
the region of uranium fluoride absorptions.40 Experiments with
18OF2 isotopic sample revealed that the former band red-shifted
slightly to 522.1 cm−1 while no obvious isotopic shift was
observed for the latter band. Since these two bands were not
observed in the argon, U and F2 experiments,40 they are
probably due to the uranium-fluoride stretching modes of the
new oxygen containing molecule. Hence we assign the three
new absorptions to the U−O stretching, symmetric and
antisymmetric F−U−F stretching modes of the OUF2
molecule. Note that the 834.8 cm−1 absorption corresponds
to the weak 834.7 cm−1 band observed in previous argon matrix
U/O2/F2 experiments, which was assigned incorrectly to the
U−O stretching mode of the OUF4 molecule as a minor
product in those experiments.41 However, as mentioned in that
paper, it is not possible to distinguish unambiguously between
the OUF2 and the OUF4 molecule possibilities based on
observation of only the uranium-oxide stretching band without
further experimental information like that provided here from
the higher yield concerted U and OF2 reaction. The
observation of two additional uranium fluoride stretching
modes along with the terminal U−O stretching mode in our
new experiments validates the identification of OUF2.
Structure and Bonding. Geometry optimizations on the
OUF2 and OThF2 molecules give pyramidal structures (Figure
4). These are related to other OUX2 and OThX2 species (X =
Figure 4. Optimized structures (bond lengths in angstroms and bond
angles in degree) for the OUF2 and OThF2 molecules at the B3LYP/
6-311+G(d) level of theory. Structural parameters were almost the
same with those given in ref. 47a. Structural parameters are also similar
using the CASPT2 method (Supporting Information, Table S1).
H, CH3) identified in earlier matrix isolation experiments,23−26
where the geometries are believed to involve 6d orbitals as
exemplified in the H2ThO molecule.24 The U−O bond length
of the OUF2 molecule is calculated here by B3LYP to be 1.830
Å, almost the same as that for OU(CH3)2 (1.833 Å) and OUH2
(1.823 Å) at the same level of theory.26 This is in line with the
small changes in the U−O stretching frequencies for these
molecules.23,26 A similar trend is found for OThF2 frequencies
and its analogues.24,26
Selected molecular orbitals for OUF2 from B3LYP
calculations are shown in Supporting Information, Figure S3
so as to understand better the multiple bonding interactions
involved in this molecule. The two unpaired electrons are left in
the highest occupied molecular orbital (HOMO) and HOMO-
1, which are mostly U 5f orbital in character. The HOMO-2 is
apparently a σ bonding orbital formed mainly via the U 6d and
O 2p interactions. Both HOMO-3 and HOMO-4 are π
bonding orbitals arising from the overlap between oxygen 2p
lone pair and uranium 6d orbitals with the former contributing
more. Calculations reveal that the U−O bond distance in the
OUF2 molecule is 1.830 Å at the B3LYP level, 1.8312 Å with
CASPT2 (6,6), and 1.8314 Å using the (8,8) active space
(Supporting Information, Table S1), which are close to the
values of recently characterized terminal uranium oxo
complexes.16,17,48 The size of the (6,6) active space is in
principle large enough because of the mostly ionic U−F bonds
formed. Therefore, one only needs six active orbitals with six
electrons to describe the multiple bonding character of the O−
U bond.
Assignments of the 806.6, 511.1, and 482.3 cm−1 absorptions
to the OThF2 molecule are straightforward following the
uranium example. The 806.6 cm−1 band shifted to 790.5 cm−1
on oxygen-18 substitution, and the isotopic frequency ratio of
1.0559 for a typical Th−O stretching mode was observed for
the 806.6 cm−1 band (1.0563 and 876.4 cm−1 for ThO in solid
argon42). The 511.1 and 482.3 cm−1 absorptions due to
symmetric and antisymmetric F−Th−F stretches exhibited 0.3
and 0.2 cm−1 red shifts upon 18O substitution, which clearly
demonstrates the slight involvement of oxygen in these two
lower frequency vibrational modes.
To confirm further our experimental assignments and to get
detailed insight into the structures of the product molecules,
theoretical calculations using the B3LYP functional were carried
out on the OUF2 and OThF2 molecules. The triplet state for
OUF2 is found to be lowest in energy while the OThF2
molecule possesses a closed shell singlet ground state. The
U−O stretching mode calculated for OUF2 at 862.4 cm−1 shifts
to 816.3 cm−1 with 18O substitution, and the computed isotopic
frequency ratio of 1.0565 is quite close to the experimental
value of 1.0560. The other two vibrational frequencies above
400 cm−1 are predicted at 521.3 and 498.4 cm−1, which are due
to symmetric and antisymmetric F−U−F stretching modes,
respectively. Note that the slight participation of oxygen in
these two modes results in very small 0.1 cm−1 calculated 18O
shifts, which is in line with the negligible shifts observed
experimentally (Table 1). Similar agreements are found
between the calculated and observed frequencies for the
OThF2 molecule (Table 1). Both the calculated band position
and 16O/18O isotopic frequency ratio for the Th−O stretching
mode fit the experimental values well. The thorium fluoride
stretching modes 0.2 and 0.1 cm−1 red shifts upon 18O
substitution as predicted by our calculations, which are also
consistent with the observed shifts of 0.3 and 0.2 cm−1.
The computed U−O bond distance in the OUF2 molecule is
0.12 Å longer than the tabulated UO triple bond length
proposed from triple bond radii,49a but near the experimental
bond length for UO (1.838 Å).49b However, this proposed
triple bond radius is based on the uranyl dipositive cation, which
clearly has shorter bond lengths than a neutral species, and is
not a good comparison for multiple bond lengths in neutral
6986
dx.doi.org/10.1021/ic3009128 | Inorg. Chem. 2012, 51, 6983−6991

Inorganic Chemistry
Article
species. Hence, the UO bond in our OUF2 molecule has
considerable triple bond character, the same as suggested for
other terminal UO containing species.47,48 The OUF2 molecule
can be compared with the isostructural OUH2 molecule.23 The
computed dipole moment of the former is 5.61 D, and the
Mulliken atomic charges are O(−0.63) U(1.60) F2(−0.49)2
and of the latter is 5.17 D with Mulliken atomic charges of
O(−0.58) U(1.22) H2(−0.32)2.
Additional support for the multiple bonding in the OUF2
molecule can be obtained from natural bond orbital (NBO)
analysis.35 As shown in Table 2, three U−O bonding orbitals
Table 2. Composition of the Bonding Molecular Orbitals in
a
the OUF2 and OThF2 Molecules from NBO Calculations
metal atomic orbital content in this
bond,%
Figure 5. CASPT2-Molecular orbitals for the oxo bond in OUF2.
bond
type
% of the NBO
on each atom
s
p
d
f
2.894, which characterizes a triple oxo bond in OUF2. Since
additional π bonding comes from dative electron donation from
oxygen, while the unpaired electrons of uranium occupy the
two f orbitals, the formal oxidation state of uranium is still best
described as IV. The singly occupied CASPT2 f orbitals are
illustrated in Supporting Information, Figure S4.
OUF2
π
π
σ
O (80.42)
(83.85)
1.85
1.48
0.32
0.26
0
98.10
98.48
0.82
0.05
0.05
U (19.58)
(16.15)
48.62
53.06
0.05
50.23
0.85
45.83
O (82.00)
(84.87)
99.95
99.95
1.47
0
0.05
The bonding interactions in the singlet OThF2 molecule are
very similar to those for OUF2 except that the Th−F bond is
more ionic. It is also reasonable to consider triple bond
character for the ThO bond in the OThF2 molecule with the
existence of one σ and two π bonds, the latter of which can be
mainly viewed as dative bonds with lone pair electron densitiy
donated from the oxygen center. The Th−O distance (1.886
Å) in the OThF2 molecule is also consistent with the value for
the recently characterized thorium oxo metallocenes (1.929
Å),22 and almost the same as the ThO triple bond length (1.89
Å) taken from atomic radii,49a but longer than the measured
bond length for ThO (1Σ+) itself (1.840 Å).49c The B3LYP
calculated bond length is 1.835 Å, and the harmonic frequency
U (18.00)
(15.13)
0
47.23
47.01
0.04
51.30
0
1.51
51.48
O (75.58)
(78.72)
13.04
17.30
0.49
0.71
0.69
1.55
16.84
0.23
0
86.91
82.66
2.47
0.04
U (24.42)
(21.28)
46.67
55.21
0.04
50.37
3.00
41.09
OThF2
π
σ
π
O (85.00)
Th (15.00)
O (82.01)
Th (17.99)
O (86.67)
Th (13.33)
99.27
2.23
61.10
0.02
35.12
39.20
41.20
83.13
4.35
56.21
0.04
99.96
0
2.08
56.73
a
3
Numbers in bold type are for β spin orbitals in A′′ OUF2.
is 896.1 cm−1 (224 km/mol intensity) for ThO.41 As can be
b
found in Table 2, the Th−O bond is a triple bond as well
because of the three bonding orbitals in spite of the slightly
larger participation of oxygen 2p orbitals, which is similar with
the uranium case. The natural charge on the thorium center is
+2.38, slightly higher than the charge on uranium. This is
consistent with the lesser contribution from thorium to the
Th−O bond, giving rise to a more ionic character. The triple
bond character for the terminal oxo ligand in OThF2 is also
supported by the B3LYP NBO calculated bond order of 2.94,
close to the value of 2.96 computed here for the ThO diatomic
molecule. Additionally, the Th 5f orbitals are found to be
involved in the Th−O bonding from NBO analysis, although
Th does not have 5f valence electrons. In conclusion, the ThO
bond in the OThF2 molecule also has considerable triple bond
character, which is illustrated nicely by the CASPT2 molecular
orbitals, Figure 6, and the computed 2.892 bond order.
Our recent studies identified the CH2UF2 molecule, which is
isoelectronic with the OUF2 molecule.51 The CH2UF2
molecule is predicted to have some CH-U agostic interaction
with the uranium center in a pyramidal geometry. Although one
of the fluoride stretching modes of the CH2UF2 molecule was
not resolved, the symmetric and antisymmetric F−U−F
stretching vibrational frequencies of the CD2UF2 methylidene
were observed at 528.3 and 507.4 cm−1, only 6 and 20 cm−1
higher than those of the OUF2 molecule. For the CH2ThF2
are found, in which the oxygen atom (mainly 2p orbital)
contributes about 80% to the U−O bond while the remaining
20% contributions come from the equivalent hybridization of U
6d and 5f orbitals. This contrasts the bonding situation in the
Cp2UO molecule, which is considered to be Cp2U(+)-O(−) with
a U−O single bond because of the localized nature of the six
electrons on oxygen.50 Similar with the triple U−O bond in the
OUF2 molecule, the U−O σ bond in the Cp2UO molecule also
contains 80% oxygen participation. The natural charge on the
uranium center in the Cp2UO molecule is +2.49,50 which is
higher than our value for OUF2 (+2.11) and consistent with
more covalent character in the latter molecule. Since fluorine is
highly electronegative, it is expected that the electrons on
oxygen tend to overlap more with the uranium orbitals, which
is different from the case with the cyclopentadienyl ligand.
The B3LYP computed bond order of 2.95 for OUF2 from
NBO analysis also supports the triple bond character for the
terminal oxo bond. For comparison, similar calculations for
quintet UO at the B3LYP level of approximation gave a 2.97
bond order. To confirm this oxo bonding scheme, benchmark
level CASPT2 level calculations were done, which have been
used to describe multiple bonding in a number of simple
uranium bearing species including NUF3.27 The CASPT2
molecular orbitals shown in Figure 5 reveal a bond order of
6987
dx.doi.org/10.1021/ic3009128 | Inorg. Chem. 2012, 51, 6983−6991

Inorganic Chemistry
Article
Figure 7. In-plane π orbitals for the isoelectronic OThF2 and
CH2ThF2 molecules.
several decades ago. The uranium oxide tetrafluoride
compound (UOF4) was characterized to have axial fluorine
and oxygen atoms with four bridged and one nonbridged
equatorial fluorine atom around the uranium center.20 The
presence of nonbridged oxygen atom was supported by the
observation of a strong presumably U−O stretching band at
885−881 cm−1, the absence of bridged UOU stretching modes,
as well as structural parameters determined by neutron
diffraction, which showed a UO distance of 1.870 Å.54 This is
slightly longer than the UO bond length computed for the
OUF2 molecule (1.830 Å) and several structurally characterized
terminal UO containing compounds.16−18
The existence of the OUF2 compound has been proposed in
its water complex OUF2·H2O without direct structural
information.55 However, all of the heavier analogues, OUX2
(X = Cl, Br, I), have been characterized to have bridged oxo
structures.56,57 Similarly, structural studies on the thorium
oxyfluorides (OThF2) as well as other oxyhalides suggested that
only bridged oxygen atoms are present in these solid
compounds, which possess longer ThO distances as well as
bridged ThOTh vibrational modes.57−59
Several aspects of the solution chemistry of the OUF2
molecule are noteworthy.55 As the pH is increased to 1.7, the
color changes from green to blue to brown and ultimately a
black precipitate is formed. This black material was found to be
76.25% U by mass, which suggested the stoichiometry
(UF2)(OH)2 or (OUF2)(H2O). The NMR spectrum of this
material is sharp compared to that of (UF4)(H2O)1.5, 2.5, which
suggested that the H2O ligand in the (OUF2)(H2O) complex is
mobile in solution. We suspect a dipole bound complex for the
two polar molecules OUF2 and H2O, with computed dipole
moments of 5.61 and 2.16 D, respectively (The experimental
value for H2O is 1.854 D, and our B3LYP value is 20%
higher.).60 Accordingly, B3LYP calculations were done for this
complex, the structure is illustrated in Supporting Information,
Figure S5, and its chemistry is discussed in our Supporting
Information.
Reactions in the Matrix. As shown in Figures 1 and 2,
ground state uranium and thorium atoms react with OF2 on
low temperature annealing in solid argon, which results in the
spontaneous formation of the OUF2 and OThF2 molecules. No
M(OF2) complexes nor FOMF (M = U,Th) insertion product
intermediates were observed in our experiments. Our B3LYP
calculations reveal that geometry optimizations of the M(OF2)
and FOMF molecules (quintet for U, triplet and singlet for Th)
give rise to the OMF2 structures, suggesting that neither
M(OF2) nor FOMF is stable along the reaction coordinates of
uranium and thorium with OF2. Since both Th−F and U−F
bonds are about three times as strong as the O−F bond,29,61,62
Figure 6. CASPT2-Molecular orbitals for the oxo bond in OThF2.
molecule, an approximately symmetrical structure was found by
the B3LYP functional, and it also showed a slightly pyramidal
geometry around Th.52 Compared with the fluoride stretching
frequencies of OThF2, the symmetric stretching mode of the
CH2ThF2 methylidene complex was almost the same, but the
antisymmetric mode was about 15 cm−1 higher, which agree
with the small changes in bond lengths. In addition to the
methylidene complexes, the first actinide borylene complex,
FBThF2, was recently identified in solid argon from the
reactions of BF3 and thorium.53 Since only three valence
electrons are available for boron, the FB fragment can only
form one σ and one π bond with the corresponding thorium
orbitals, which results in a ThB double bond. To further
compare the geometric changes for the XThF2 (X = BF, CH2,
O) molecules, the structural parameters of the CH2ThF2 and
OThF2 molecules were optimized using the B3LYP functional
and the larger 6-311+G(3df) basis set, which is the same as the
one used for FBThF2. The changes in Th−F bond lengths are
within 0.01 Å and the changes in bond angles are less than 2
degrees compared with the results using the smaller 6-
311+G(d) basis set. The borylene FBThF2 still possesses the
shortest computed Th−F bond (2.119 Å) followed by
CH2ThF2 (2.129 Å) and OThF2 (2.147 Å), which is also
consistent with its highest fluoride stretching frequencies
observed near 540 and 525 cm−1,53 about 30 and 40 cm−1
higher than those of the OThF2 molecule. The calculated FThF
bond angle for FBThF2 is 126.9°, larger than that for CH2ThF2
(120.6°) and OThF2 (106.5°) all calculated with large basis sets
while the difference in XThF bond angle is smaller with OThF2
(112.3°) followed by FBThF2 (108.5°) and CH2ThF2 (106.5°).
It is interesting to note that the in-plane π bonding orbital for
the CH2ThF2 molecule is mainly a Th−C nonbonding orbital,
but it has some C−H bonding character, which is obviously
different from the in-plane π bonding orbital for OThF2 (Figure
7). This results in a true ThC double bond for the CH2ThF2
molecule. In the case of FBThF2, only one σ and one out-of-
plane π orbital were found while the in-plane orbital does not
exist because of only three valence electrons for boron. As a
result, the FThF bond angle is larger when the ThX bond
involves in-plane π interaction, the presence of which is
believed to increase the repulsive interactions between terminal
Th−F σ bond and this π bond. Hence, the FThF bond angles
are sensitive to the existence of the second π bonding orbital in
such a system with multiple ThX bonds.
OUF2/OThF2 in the Condensed Phase. Both uranium
and thorium oxyfluorides have been reported in the solid state
6988
dx.doi.org/10.1021/ic3009128 | Inorg. Chem. 2012, 51, 6983−6991

Inorganic Chemistry
Article
formation of the OUF2 and OThF2 molecules is expected to be
highly exothermic, and these molecules are further stabilized
through triple bonds in the UO and ThO linkages. Our B3LYP
calculations reveal that the bond dissociation energies for the
U−F and Th−F bonds in the OMF2 molecules are 137 and 143
kcal/mol, while those for the U−O and Th−O bonds are 160
and 185 kcal/mol. The overall reaction exothermicites for the
formation of the OUF2 and OThF2 molecules are quite large,
377 and 419 kcal/mol, respectively, for reactions 1 and 2,
computed at the B3LYP level of theory without spin orbit
corrections. Although no UO was detected in the argon matrix
experiments, U16O is masked by the intense OF2 absorption
and weak U16O2 bands may cover any weak U18O absorption.39
However, UO was observed in the neon matrix experiment,
which may arise from the slower condensation and relaxation
rate of energized product molecules formed in neon freezing at
4 K.
CONCLUSIONS
■
Reactions of laser ablated uranium and thorium atoms with
OF2 have been investigated using matrix isolation infrared
spectroscopy. Uranium and thorium oxyfluoride species, OUF2
and OThF2, are identified by the observations of terminal M-O
and two F−M−F (M = U, Th) stretching vibrational modes as
well as density functional vibrational frequency calculations.
The OUF2 and OThF2 molecules are produced via the
reactions of uranium and thorium atoms with OF2 upon
sample annealing, during which negligible activation energy is
required. This is the first example of an oxide-fluoride molecule
prepared in a single concerted reaction. All aspects of uranium-
fluorine chemistry are important: first, for the investigation of
chemical bonding, and second, for better understanding of
applications for uranium fluorides such as the well-known
separation of U isotopes through the diffusion of UF6
isotopomers.
B3LYP and CASPT2 calculations predict pyramidal
structures for these two oxyfluoride products. The closed
shell singlet is found to be the ground state for the thorium
product while triplet OUF2 is calculated to be lowest in energy.
Bonding analysis (NBO) reveals that the UO and ThO
fragments in OUF2 and OThF2 contain considerable triple
bond character because of the existence of an in-plane π bond,
which is also reflected in the F−M−F bond angles as compared
to molecules with similar structures. Benchmark CASPT2
calculations reveal 2.9 bond orders for each of these new oxo
species and demonstrate the importance of dative bonding
involving O2p → Th6d and U6d5f interactions. The highly
exothermic character of the OF2 reactions suggest that
numerous metal containing species with terminal oxo groups
can be prepared using these fluorine transfer reactions.
U + OF → OUF
(1)
(2)
2
2
Th + OF → OThF
2
2
A series of recent investigations on the reactions of fluorine
substituted molecules with uranium and thorium has
demonstrated that molecules containing as many metal fluorine
bonds as possible were usually the final observed prod-
ucts.27,28,51,52,63,64 This leads to the formation of a series of
uranium carbide and nitride molecules as well as the terminal
oxo species with multiple bonds identified here.27,64 Formation
of the FC÷ThF3 and N÷ThF3 molecules with two singly
occupied π bonds was still favorable28,52 even if the C−F bond
is stronger than the N−F bond.29 However the strongest B−F
bond makes the BF3 and thorium reactions terminate at the
borylene complex FBThF2 without further fluorine transfer to
give the B÷ThF3 molecule.53
ASSOCIATED CONTENT
* Supporting Information
■
S
OF2 is known to decompose upon absorbing UV photons.65
As a result, photodecomposition of OF2 is expected during
sample deposition due to irradiation from the plume produced
by laser ablation of metal target, which contains UV and
vacuum UV photons.66 The intensity of the plume radiation
can be changed by the laser energy, which accounts for the laser
energy dependence of the OF band intensity. A similar process
occurs when the sample is exposed to UV irradiation, both here
and in earlier work, where Arkell et al. first prepared OF.65
Along with the decomposition of OF2, the OF radical band
appeared. Additionally, atomic fluorine and oxygen were
produced at the same time, which were evidenced by the
presence of metal fluoride as well as weak ozone absorptions.
Previous studies on the reactions of uranium with F2 and O2
mixtures suggested that the UO2F2 and UO2F molecules were
major products while a weak band due to a single oxygen
containing species was detected.41 Uranium fluoride absorp-
tions especially UF6 became dominant when more concen-
trated fluorine samples were used. In contrast, the dioxo
uranium fluoride species were stronger in the experiments with
more O2, which favors the formation of uranium dioxide as well
as its fluorine adducts. In the case of OF2, the fluorine transfer
reactions make it possible for the formation of the molecules
with one oxygen atom, and the fluoro and dioxo species are
weak. Finally, these matrix reactions can be used to make an
interesting series of metal containing molecules with terminal
oxo groups.
Complete references 32 and 36e, computed data table,
comparison of U and OF2 vs F2 reaction product argon matrix
infrared spectra, and neon matrix spectra for Th and U vs OF2
reaction products. This material is available free of charge via
the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*E-mail: lsa@virginia.edu.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We gratefully acknowledge financial support from DOE Grant
DE-SC0001034 and NCSA computing Grant CHE07-0004N.
We thank Al Arkell of Texaco Research for the enriched 18OF2
sample and our colleague Sergei Egorov for translation and
helpful discussion of the work in ref 55. We appreciate Bjorn
Roos for his many contributions to our understanding of
chemical bonding. S.R. and T.S. gratefully acknowledge
financial support from Fonds der Chemischen Industrie
(FCI). The authors are grateful to the BWGrid cluster for
providing computational resources.
REFERENCES
■
(1) Nugent, W. A.; Mayer, J. M. Metal-Ligand Multiple Bonds; Wiley:
New York, 1988.
6989
dx.doi.org/10.1021/ic3009128 | Inorg. Chem. 2012, 51, 6983−6991

Inorganic Chemistry
Article
(2) (a) Nam, W. Acc. Chem. Res. 2007, 40, 522−531. (b) Yoshizawa,
K. Acc. Chem. Res. 2006, 39, 375−382. (c) Gunay, A.; Theopold, K. H.
Chem. Rev. 2010, 110, 1060−1081.
(33) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652. (b) Lee,
C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785−789.
(34) (a) McLean, A. D.; Chandler, G. S. J. Chem. Phys. 1980, 72,
5639−5648. (b) Krishnan, R.; Binkley, J. S.; Seeger, R.; Pople, J. A. J.
Chem. Phys. 1980, 72, 650−654. (c) Kuchle, W.; Dolg, M.; Stoll, H.;
Preuss, H. J. Chem. Phys. 1994, 100, 7535−7542.
(3) Somorjai, G. A. Introduction to Surface Chemistry and Catalysis;
Wiley: New York, 1994.
(4) Rohde, J. U.; In, J. H.; Lim, M. H.; Brennessel, W. W.; Bukowski,
M. R.; Stubna, A.; Munck, E.; Nam, W.; Que, L., Jr. Science 2003, 299,
(35) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988, 88,
899 , and references therein..
̈
1037−1039.
(5) Poverenov, E.; Efremenko, I.; Frenkel, A. I.; Ben-David, Y.;
Shimon, L. J. W.; Leitus, G.; Konstantinovski, L.; Martin, J. M. L.;
Milstein, D. Nature 2008, 455, 1093−1096.
(36) (a) Roos, B. O. The Complete Active Space Self-Consistent
Field Method and its Applications in Electronic Structure Calculations.
In Advances in Chemical Physics; Ab Initio Methods in Quantum
Chemistry − II; Lawley, K. P., Ed.; John Wiley & Sons Ltd.: New York,
1987; Chapter 69, p 399. (b) Andersson, K.; Malmqvist, P.-Å.; Roos,
B. O. J. Chem. Phys. 1992, 96, 1218. (c) Roos, B. O.; Lindh, R.;
Malmqvist, P.-Å.; Veryazov, V.; Widmark, P.-O. J. Phys. Chem. A 2005,
(6) (a) Anderson, T. M.; Neiwert, W. A.; Kirk, M. L.; Piccoli, P. M.
B.; Schultz, A. J.; Koetzle, T. F.; Musaev, D. G.; Morokuma, K.; Cao,
R.; Hill, C. L. Science 2004, 306, 2074−2077. (b) Anderson, T. M.;
Cao, R.; Slonkina, E.; Hedman, B.; Hodgson, K. O.; Hardcastle, K. I.;
Neiwert, W. A.; Wu, S.; Kirk, M. L.; Knottenbelt, S.; Depperman, E.
C.; Keita, B.; Nadjo, L.; Musaev, D. G.; Morokuma, K.; Hill, C. L. J.
Am. Chem. Soc. 2005, 127, 11948−11949. (c) Cao, R.; Anderson, T.
M.; Piccoli, P. M. B.; Schultz, A. J.; Koetzle, T. F.; Geletii, Y. V.;
Slonkina, E.; Hedman, B.; Hodgson, K. O.; Hardcastle, K. I.; Fang, X.;
109, 6575 ANO-RCC-VTZP basis. (d) Karlstrom, G.; Lindh, R.;
̈
Malmqvist, P.-Å.; Roos, B. O.; Ryde, U.; Veryazov, V.; Widmark, P.-O.;
Cossi, M.; Schimmelpfennig, B.; Neogrady, P.; Seijo, L. Comput. Mater.
Sci. 2003, 28, 222. (e) Werner, H.-J.; Knowles, P. J. et al. MOLPRO,
2006.1 ed.; University of Birmingham: Birmingham, U.K., 2006.
(37) (a) Andrews, L. J. Chem. Phys. 1972, 57, 51−55. (b) Andrews,
L.; Raymond, J. I. J. Chem. Phys. 1971, 55, 3078−3086. (c) Arkell, A. J.
Am. Chem. Soc. 1965, 87, 4057−4062.
Kirk, M. L.; Knottenbelt,; Kogerler, P.; Musaev, D. G.; Morokuma, K.;
̈
Takahashi, M.; Hill, C. L. J. Am. Chem. Soc. 2007, 129, 11118−11133.
(7) Saouma, C. T.; Peters, J. C. Coord. Chem. Rev. 2011, 255, 920−
937.
(38) Jacox, M. E. J. Mol. Spectrosc. 1980, 84, 74−88.
(39) Hunt, R. D.; Andrews, L. J. Chem. Phys. 1993, 98, 3690−3696.
(40) Hunt, R. D.; Thompson, C.; Hassanzadeh, P.; Andrews, L. Inorg.
Chem. 1994, 33, 388−391.
(8) Arnold, P. L. Chem. Commun. 2011, 47, 9005−9010.
(9) Fox, A. R.; Bart, S. C.; Meyer, K.; Cummins, C. C. Nature 2008,
455, 341−349.
(41) Souter, P. F.; Andrews, L. J. Mol. Struct. 1997, 412, 161−167.
(42) (a) Gabelnick, S. D.; Reedy, G. T.; Chasanov, M. G. J. Chem.
Phys. 1974, 60, 1167−1171. (b) Andrews, L.; Gong, Y.; Liang, B.;
Jackson, V. E.; Flamerich, R.; Li, S.; Dixon, D. A. J. Phys. Chem. A
2011, 115, 14407−14416.
(43) (a) Buchmarina, V. N.; Gerasimov, A. Y.; Predtechenskii, Y. B.;
Shklyarik, V. G. Opt. Spektrosk. (USSR) 1992, 72, 69−74. (b)
unpublished infrared spectra from this laboratory of Th reaction
products with F2 in excess argon at 4 K.
(10) Andrea, T.; Eisen, M. S. Chem. Soc. Rev. 2008, 37, 550−567.
(11) Barnea, E.; Eisen, M. S. Coord. Chem. Rev. 2006, 250, 855−899.
(12) Fortier, S.; Hayton, T. W. Coord. Chem. Rev. 2010, 254, 197−
214.
(13) Denning, R. G. J. Phys. Chem. A 2007, 111, 4125−4143.
(14) Arnold, P. L.; Love, J. B.; Patel, D. Coord. Chem. Rev. 2009, 253,
1973−1978.
(15) Graves, C. R.; Kiplinger, J. L. Chem. Commun. 2009, 3831−
3853.
(16) Hayton, T. W. Dalton Trans. 2010, 39, 1145−1158.
(17) (a) Fortier, S.; Kaltsoyannis, N.; Wu, G.; Hayton, T. W. J. Am.
Chem. Soc. 2011, 133, 14224−14227. (b) Fortier, S.; Brown, J. L.;
Kaltsoyannis, N.; Wu, G.; Hayton, T. W. Inorg. Chem. 2012, 51, 1625−
1633.
(18) (a) Kraft, S. J.; Walensky, J.; Fanwick, P. E.; Hall, M. B.; Bart, S.
C. Inorg. Chem. 2010, 49, 7620−7622. (b) Evans, W. J.; Kozimor, S.
A.; Ziller, J. W. Polyhedron 2004, 23, 2689−2694.
(19) de Wet, J. F.; du Preez, J. G. H. J. Chem. Soc., Dalton Trans.
1978, 592−597.
(20) Paine, R. T.; Ryan, R. R.; Asprey, L. B. Inorg. Chem. 1975, 14,
1113−1117.
(21) (a) Denning, R. G. Struct. Bonding (Berlin) 1992, 79, 215−276.
(b) Kaltsoyannis, N. Chem. Soc. Rev. 2003, 32, 9−16.
(22) Ren, W. S.; Zi, G. F.; Fang, D. C.; Walter, M. D. J. Am. Chem.
Soc. 2011, 133, 13183−13196.
(23) Liang, B.; Hunt, R. D.; Kushto, G. P.; Andrews, L.; Li, J.;
Bursten, B. E. Inorg. Chem. 2005, 44, 2159−2168.
(24) Liang, B.; Andrews, L.; Li, J.; Bursten, B. E. J. Am. Chem. Soc.
2002, 124, 6723−6733.
(25) Gong, Y.; Andrews, L. Inorg. Chem. 2011, 50, 7099−7105.
(26) Gong, Y.; Andrews, L. Dalton Trans. 2011, 40, 11106−11114.
(27) Andrews, L.; Wang, X. F.; Lindh, R.; Roos, B. O.; Marsden, C. J.
Angew. Chem., Int. Ed. 2008, 47, 5366−5370.
(28) Wang, X. F.; Andrews, L. Dalton Trans. 2009, 9260−9265.
(29) CRC Handbook of Chemistry and Physics; CRC Press: Boca
Raton, FL, 1985−1986.
(44) Zhou, M. F.; Andrews, L.; Ismail, N.; Marsden, C. J. Phys. Chem.
A 2000, 104, 5495−5502.
(45) Jacox, M. E. Chem. Phys. 1994, 189, 149−159.
(46) Wang, X. F.; Cho, H. G.; Andrews, L.; Chen, M. Y.; Dixon, D.
A.; Hu, H. S.; Li, J. J. Phys. Chem. A 2011, 115, 1913−1921.
(47) (a) Shamov, G. A.; Schreckenbach, G.; Vo, T. N. Chem.Eur. J.
́
2007, 13, 4932−4947. (b) Kovacs, A.; Konings, R. J. M.
ChemPhysChem 2006, 7, 455−462.
(48) (a) Bart, S. C.; Anthon, C.; Heinemann, F. W.; Bill, E.;
Edelstein, N. M.; Meyer, K. J. Am. Chem. Soc. 2008, 130, 12536−
12546. (b) Kosog, B.; La Pierre, H. S.; Heinemann, F. W.; Liddle, S.
T.; Meyer, K. J. Am. Chem. Soc. 2012, 134, 5284−5289.
(49) (a) Pyykko, P.; Riedel, S.; Patzschke, M. Chem.Eur. J. 2005,
̈
11, 3511−3520. (b) Heaven, M. C. Phys. Chem. Chem. Phys. 2006, 8,
4497−4509. (c) Huber, K. P.; Herzberg, G. Molecular Spectra and
Molecular Structure of Diatomic Molecules; Van Nostrand: New York,
1979.
(50) (a) Zi, G.; Jia, L.; Werkema, E. L.; Walter, M. D.; Gottfriedsen, J.
P.; Andersen, R. A. Organometallics 2005, 24, 4251−4264. (b) Barros,
N.; Maynau, D.; Maron, L.; Eisenstein, O.; Zi, G.; Andersen, R. A.
Organometallics 2007, 26, 5059−5065.
(51) Lyon, J. T.; Andrews, L.; Hu, H. S.; Li, J. Inorg. Chem. 2008, 47,
1435−1442.
(52) Lyon, J. T.; Andrews, L. Eur. J. Inorg. Chem. 2008, 1047−1058.
(53) Wang, X. F.; Roos, B. O.; Andrews, L. Chem. Commun. 2010, 46,
1646−1648.
(54) Levy, J. H.; Taylor, J. C.; Wilson, P. W. J. Inorg. Nucl. Chem
1977, 39, 1989−1991.
(30) (a) Andrews, L.; Citra, A. Chem. Rev. 2002, 102, 885−911.
(b) Andrews, L. Chem. Soc.Rev. 2004, 33, 123−132.
(31) Reinhard, R. R.; Arkell, A. Int. J. Appl. Radiat. Isot. 1965, 16,
498−499.
(32) Frisch, M. J.; et al. Gaussian 09, Revision B.01; Gaussian, Inc.:
Wallingford, CT, 2010.
(55) Vdovenko, V. M.; Romanov, G. A.; Solntseva, L. V.
Radiokhimiya 1967, 9, 727−729.
(56) Levet, J. C.; Noel, H. J. Solid State Chem. 1979, 28, 67−73.
̈
(57) Bagnall, K. W.; Brown, D.; Easey, J. F. J. Chem. Soc. A 1968,
288−291.
6990
dx.doi.org/10.1021/ic3009128 | Inorg. Chem. 2012, 51, 6983−6991

Inorganic Chemistry
Article
(58) D’Eye, R. W. M. J. Chem. Soc. 1958, 196−199.
(59) Scaife, D. E.; Turnbull, A. G.; Wylie, A. W. J. Chem. Soc. 1965,
1432−1437.
(60) Xu, X.; Goddard, W. A., III J. Phys. Chem. A 2004, 108, 2305−
2313 , and references therein.
(61) Lau, K. H.; Hildenbrand, D. L. J. Chem. Phys. 1982, 76, 2646−
2652.
(62) Zmbov, K. F. J. Inorg. Nucl. Chem. 1970, 32, 1378−1381.
(63) Andrews, L.; Wang, X. F.; Roos, B. O. Inorg. Chem. 2009, 48,
6594−6598.
(64) Lyon, J. T.; Hu, H. S.; Andrews, L.; Li, J. Proc. Natl. Acad. Sci.
2007, 104, 18919−18924.
(65) Arkell, A.; Reinhard, R.; Larson, L. P. J. Am. Chem. Soc. 1965, 87,
1016−1020.
(66) (a) Flesch, R.; Schurmann, M. C.; Hunnekuhl, M.; Meiss, H.;
Plenge, J.; Ruhl, E. Rev. Sci. Instrum. 2000, 71, 1319−1324 , and
references therein. (b) Gong, Y.; Andrews, L. J. Phys. Chem. A 2011,
115, 3029−3033.
6991
dx.doi.org/10.1021/ic3009128 | Inorg. Chem. 2012, 51, 6983−6991