Thorium fluorides ThF, ThF2, ThF3, ThF4, ThF3(F2), and ThF5- characterized by infrared spectra in solid argon and electronic structure and vibrational frequency calculations

Article abstract of DOI:10.1021/ic401107w

Reactions of laser-ablated Th atoms with F₂ produce ThF ₄ as the major product based on agreement with matrix spectra recorded of the vapor from the solid at 800-850 °C. Weaker higher-frequency bands at (567.2, 564.8), (575.9, 575.1), and (531.0, 528.4) cm^-1 in argon are assigned to ThF, ThF₂ and ThF₃, ThF ₃(F₂) on the basis of their chemical behavior upon increasing reagent concentrations, annealing, and irradiation, the use of NF₃, OF₂, and HF as F-atom sources, and a comparison with frequencies calculated at the DFT/B3LYP and CCSD(T) levels with a large segmented + ECP basis set on Th and the aug-cc-pVTZ basis set on F. An additional broader band at 460 cm^-1 is assigned to the ThF ₅^- anion. The trigonal-bipyramidal ThF₅ ^- anion (calculated electron detachment energy of 7.17 eV) increases at the expense of ThF₃(F₂) and F₃^- on full mercury arc irradiation. [ThF₃⁺][F ₂^-] is shown by calculations to be an ionic complex with a side-bound F₂^- subunit. This paper reports the first evidence for novel pentacoordinated thorium species including the unique [ThF₃⁺][F₂^-] ionic radical-ion pair molecule and its electron-capture product, the very stable ThF₅ ^- anion.

Full text of DOI:10.1021/ic401107w

Article
pubs.acs.org/IC
−
Thorium Fluorides ThF, ThF2, ThF3, ThF4, ThF3(F2), and ThF5
Characterized by Infrared Spectra in Solid Argon and Electronic
Structure and Vibrational Frequency Calculations
Lester Andrews,*,† K. Sahan Thanthiriwatte,‡ Xuefeng Wang,† and David A. Dixon*,‡
†Department of Chemistry, University of Virginia, Charlottesville, Virginia 22904-4319, United States
‡Department of Chemistry, The University of Alabama, Shelby Hall, Tuscaloosa, Alabama 35487-0336, United States
S
* Supporting Information
ABSTRACT: Reactions of laser-ablated Th atoms with F2 produce ThF4 as the major
product based on agreement with matrix spectra recorded of the vapor from the solid at
800−850 °C. Weaker higher-frequency bands at (567.2, 564.8), (575.9, 575.1), and
(531.0, 528.4) cm−1 in argon are assigned to ThF, ThF2 and ThF3, ThF3(F2) on the
basis of their chemical behavior upon increasing reagent concentrations, annealing, and
irradiation, the use of NF3, OF2, and HF as F-atom sources, and a comparison with
frequencies calculated at the DFT/B3LYP and CCSD(T) levels with a large segmented
+ ECP basis set on Th and the aug-cc-pVTZ basis set on F. An additional broader band
at 460 cm−1 is assigned to the ThF5 anion. The trigonal-bipyramidal ThF5 anion
(calculated electron detachment energy of 7.17 eV) increases at the expense of
−
−
−
+
−
ThF3(F2) and F3 on full mercury arc irradiation. [ThF3 ][F2 ] is shown by calculations to be an ionic complex with a side-
bound F2 subunit. This paper reports the first evidence for novel pentacoordinated thorium species including the unique
[ThF3 ][F2 ] ionic radical-ion pair molecule and its electron-capture product, the very stable ThF5 anion.
−
+
−
−
INTRODUCTION
RESULTS AND DISCUSSION
■
■
IR spectra recorded from the reaction products of laser-ablated
Th atoms codeposited with molecular fluorine (F2) using
established experimental methods16−19 are illustrated in Figure
1 for 0.5% F2 (scans a−d) and 1.0% F2 (scans e−h). First, the
strong absorptions at 521.0, 519.2, and 514.5 cm−1 in the
region of the 520 cm−1 gas-phase absorption for ThF4 are in
agreement with argon matrix spectra of ThF4 vapor from the
solid at 800−850 °C.6,7 New absorptions of interest here are
the unresolved doublet at (575.9, 575.1) cm−1, the resolved
doublet at (567.2, 564.8) cm−1, the 531.0 cm−1 band, and the
528.4 cm−1 band, which survives 20 K annealing, decreases
upon >220 nm irradiation, and increases slightly upon final 30
K annealing. Meanwhile, absorption appears at 460 cm−1 upon
deposition, decreases upon 20 K annealing, increases slightly
and sharpens upon >220 nm irradiation, and decreases again
Uranium hexafluoride is the basis for the important uranium
isotope separation process by gaseous diffusion.1 The UF6
molecule is also interesting in its own right for vibrational
spectroscopy (laser isotope separation)2 and its large electron
affinity with values of 4.9 0.5,3 >5.1,4 and 5.3 eV.5 However,
the corresponding highest thorium fluoride, ThF4, has received
much less attention. The vapor in equilibrium with solid ThF4
at 1200 °C exhibited an IR band at 520 cm−1, which was
assigned to the antisymmetric Th−F stretching mode.6 Spectra
of the vapor generated from the solid at 800−850 °C and
trapped in argon and neon matrixes produced matrix site-split
bands in the 520 cm−1 region.7 Vapor-pressure data for thorium
tetrahalides support the tetrahedral structure expected for the
4+ oxidation state of Th.8 A high-resolution spectroscopic
investigation of ThF and ThF+ together with high-level
electronic structure calculations has recently appeared.9 The
gaseous species ThF, ThF2, ThF3, and ThF4 have been
generated in a heated effusion beam source and observed as
cations by mass spectrometry.10
The laser-ablation technique has been employed to generate
both Th and U atoms for reactions with small molecules and to
trap the novel intermediate reaction products in noble gas
solids in this laboratory. Most recently, we have investigated the
OUF2 and OThF2 molecules with polarized triple oxo bond
character from reactions with OF2.11 Earlier work provided
characterization of N−ThF3, thorium oxides, the NThO
molecule, and thorium hydrides.12−15
−
upon final 30 K annealing. The F3 absorption at 510 cm−1 is
destroyed by >220 nm irradiation.19 Note the reversal of the
531.0 and 528.4 cm−1 band intensities with increasing F2
concentration, where more F2 favors the lower frequency band.
Additional laser-ablated thorium experiments were done in
collaboration with the Riedel group at Freiburg using higher
fluorine concentrations. Spectra for 2% F2 in argon are
illustrated in Figure SI-1 in the Supporting Information. Several
points in comparison with the 1 and 2% F2 experiments are
noteworthy: (1) the same absorptions were observed within 0.1
Received: May 3, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ic401107w | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Full mercury arc irradiation substantially decreased the (567.2,
564.8) cm−1 and 531.0 cm−1 peaks, doubled the intensity of the
new (575.9, 575.1) cm−1 and ThF4 bands, and enhanced a weak
absorption at 460 cm−1. Next annealing to 30 K slightly
increased the 567.2 and 531.0 cm−1 peaks and a 528.4 cm−1
satellite feature and decreased the lower-frequency band, as
observed with the F2 reagent. A final >220 nm irradiation
enhanced the broad 460 cm−1 absorption. The weak 511.1 and
482.3 cm−1 peaks are due to OThF2 owing to the trace ThO
impurity in the system.11
Matrix spectra from the reaction of Th with 18OF2 are
illustrated in Figure 2 (scans f−k), which includes regions not
illustrated previously.11 As reported, weaker bands were
observed for ThF4 and stronger new bands at 510.8 and
482.1 cm−1 for the major 18OThF2 product.11 Additional weak
bands were observed at 575.2, 567.2, 564.8, 531.0, and 528.2
cm−1. The (567.2, 564.8) cm−1 peaks and the major product
bands increased upon annealing to 20 K, as before. Full arc
photolysis increased the 575.2 cm−1 band and decreased the
other weaker peaks. A final annealing to 30 K restored some of
the 564.8 cm−1 band intensity. In this case, features in the 460
cm−1 region show small oxygen isotopic shifts and are due to
the precursor and its aggregates.11 It is significant that the
575.2, 567.2, 564.8, 531.0, and 528.2 cm−1 product bands did
not shift in a similar experiment with Th and natural isotopic
OF2. Hence, their observation from Th reactions with F2, NF3,
and OF2 with no O-18 shift shows that these are due to
molecular thorium fluoride species.
Figure 1. IR spectra of laser-ablated Th atom and F2 molecule reaction
products in solid argon: (a) Th + 0.5% F2 deposition for 60 min; (b)
after annealing to 20 K; (c) after λ > 220 nm irradiation; (d) after
annealing to 20 K [spectra absorbance scale expanded two times]; (e)
Th + 1% F2 deposition for 60 min; (f) after annealing to 20 K; (g)
after λ > 220 nm irradiation; (h) after annealing to 30 K.
cm−1; (2) the 528.4 cm−1 band was slightly stronger than the
530.1 cm−1 band, photolysis destroyed them both, and
annealing restored only the 528.4 cm−1 band; (3) the 460
cm−1 absorption was stronger, and it increased upon UV
irradiation.
Additional chemical support can be derived from the fact that
the same 576, 575, 567.2, 564.8, and 531.0 cm−1 absorption
peaks were observed with the same annealing and photolysis
behavior in analogous argon matrix experiments with Th and
HF performed at Freiburg. Spectra from an HF experiment are
shown in Figure SI-2 in the Supporting Information for
comparison.
IR spectra using the NF3 reagent instead of F2 are shown in
Figure 2 (scans a−e). As reported previously,12 the major N−
Evidence will be presented for the identification of thorium
fluoride molecules in the order of increasing numbers of F
atoms. The experimental assignments together with our
computational results are given in Tables 1 and 2. First, the
two sharp bands at 567.2 and 564.8 cm−1 and a weaker matrix
site splitting at 569.2 cm−1 can be assigned to the ThF
molecule. These bands are photosensitive, and they are restored
upon annealing, more than any other new product bands in this
system. Thus, a straightforward reaction must be available to
form this product, and that reaction is the direct exothermic
combination of atoms to form ThF, whose energy is given for
reaction 1 in Table 3.9,10 UV irradiation by the laser-ablation
plume provides F atoms from precursor photodissociation.
Second, these bands are observed with the F-atom precursors
NF3, OF2, and HF with relative product band intensities even
more prominent than those with F2 itself. Third, an
Figure 2. IR spectra of laser-ablated Th atom and NF3 or OF2
molecule reaction products in solid argon: (a) Th + 0.4% NF3
deposition for 60 min; (b) after annealing to 20 K; (c) after λ >
220 nm irradiation; (d) after annealing to 30 K; (e) after λ > 220 nm
irradiation. (f) Th + 1% 18OF2 (91% enriched) deposition for 60 min;
(g) after annealing to 20 K; (h) after λ > 220 nm irradiation; (i) after
annealing to 30 K.
approximate gas-phase fundamental of 605
15 cm−1 was
deduced from electronic band spacings, which is in reasonable
agreement with a multireference configuration interaction
calculation (574 cm−1).9 Fourth, our calculations20,21 at the
spin-unrestricted density functional theory (DFT) level with
the B3LYP exchange-correlation functional22,23 and at the
correlated molecular orbital theory coupled cluster CCSD-
(T)24−27 level produce frequencies in this region appropriately
slightly higher than the above argon matrix values and
consistent with the experimental gas-phase result (Table 1).
[R/UCCSD(T) has a restricted open-shell Hartree−Fock
calculation for the starting wave function with the spin
ThF3 product can clearly be identified, and its strongest band,
the antisymmetric ThF3 stretching mode, appeared at 525.3
cm−1. Here, we note first that exactly the same frequency, but
weaker bands, was observed for ThF4, and for the new products
at (575.9, 575.1), at (567.2, 564.8), and at 531.0 cm−1.
Annealing to 20 K increased the latter three band groups
slightly with the (567.2, 564.8) cm−1 peaks increasing more.
B
dx.doi.org/10.1021/ic401107w | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
−
+
−
−
Table 1. Calculated Thorium Fluoride Geometry
Parameters, Frequencies (cm−1), IR Intensities (km/mol),
and Observed Argon Matrix Frequencies (cm−1)
Table 2. Calculated ThF4 , [ThF3 ][F2 ], and ThF5
Geometry Parameters, Frequencies (cm−1), and IR
Intensities (km/mol), and Observed Argon Matrix
Frequencies (cm−1)
CCSD(T)/aug-cc- B3LYP/aug-cc-
a
b
parameter
pVTZ/ECP
pVTZ/ECP
obsd
B3LYP/aug-
ThF(2Δ)
CCSD(T)/aug-
cc-pVTZ/ECP
cc-pVTZ/
a
parameter
r(Th−F) (Å)
ECP
obsd
r(Th−F) (Å)
Th−F str
2.0319
2.037
ThF4−(D4h A1g)
2
c
a
598.9
587.2 (78)
567.2, 564.8,
d
(605 15)
2.180
511.1
2.179
a
ThF2(C2v 1A1)
Th−F sym str (a1g)
504.0 (0)
r(Th−F) (Å)
2.061
129.2
2.061
142.3
Th−F asym str (eu)
Th−F asym, sym str (b1g)
456.4
452.5 (503)
418.4(0)
∠(F−Th−F)
428.3
ThF4−(Td 2A1)
(deg)
F−Th−F bend
67.1
106.1 (2)
r(Th−F) (Å)
Th−F sym str (a)
Th−F asym str (t)
2.149
2.161
(a1)
511.7(0)
481.8 (268)
Th−F sym str
585.4
573.4
581.8 (17)
(a1)
[ThF3 ][F2−](Cs 2A″)
+
Th−F asym str
577.2 (186)
ThF3(D3h 2A1)
575.1, 575.9
(b1)
r(Th−Fbonded) (Å)
r(Th−Fnonbonded) (Å)
2.106
2.108
2.356
2.363
r(Th−F) (Å)
out-of-plane
bend (a2″)
2.103
28.5
2.105
r(Fnonbonded−Fnonbonded) (Å)
∠Fbonded−Th−Fbonded (deg)
1.908
1.962
65.6 (5)
106.6/101.5
108.4/102.1
109.1
∠Fbonded (on Cs)−Th−Fnonbonded
110.2
F−Th−F bend
105.6
537.2
570.8
96.9 (20)
539.3 (394)
571.9 (0)
(deg)
(e′)
Th−F sym str (a′)
Th−F asym str (a″)
Th−F asym str (a′)
F−F str (a′)
580.2 (55)
534.3 (206)
531.1 (166)
457.4 (44)
Th−F asym str
(e′)
531.0
528
528
Th−F sym str
(a1′)
ThF5−(D3h 1A1′)
2.194
ThF4(Td 1A1)
r(Th−Fax) (Å)
r(Th−Feq) (Å)
Th−F sym str (a1′)
Th−Feq asym str (e′)
Th−Fax asym str (a2″)
Th−F sym str (a1′)
2.194
r(Th−F) (Å)
2.117
112.0
2.112
2.183
2.190
F−Th−F asym
101.7 (0)
bend (e)
534.4
523.8 (0)
452.5 (242)
446.1 (323)
415.6 (0)
Th−F sym bend
113.2
528.3
563.3
93.9 (90)
463.5
460
460
(t2)
451.7
Th−F asym str
537.5 (627)
521.0, 519.2,
514.5
426.7
(t2)
a
Values in parentheses are IR intensities (km/mol).
Th−F sym str
588.6 (0)
(a1)
a
b
Table 3. Calculated Reaction Enthalpies (kcal/mol) at 298 K
for the Lowest-Energy Conformer Involved in These Matrix
Isolation Experiments
IR intensity (km/mol) in parentheses. Argon matrix frequencies.
ωexe = 2.1 cm−1. Gas phase.9
c
d
constraint relaxed in the coupled cluster calculation where both
use the aug-cc-pVTZ basis set on F28 and the small-core
relativistic effective core potential (ECP) from the Stuttgart
group with a corresponding segmented basis set on Th.29 We
denote this basis set as aug-cc-pVTZ/ECP.] The CCSD(T)
calculations correlated 12 electrons on Th and 7 electrons on F.
Furthermore, the CCSD(T) harmonic value for the asymmetric
Th−F stretch of ThF4 is within 8 cm−1 of the experimental
anharmonic matrix value (Table 1). The calculated bond
dissociation energy for ThF requires a spin−orbit correction of
8.8 kcal/mol for the Th atom30 and a correction of 0.38 for the
F atom.31 The site-split ThF4 absorptions were 8 cm−1 higher
in solid neon than argon,7 so we expect a higher neon matrix
value for ThF. From these observations, we predict a gas-phase
fundamental in the 580−600 cm−1 range, which is in
accordance with the calculations and the estimate from the
band structure.9
reaction
no.
B3LYP/aug-cc-
pVTZ/ECP
CCSD(T)/aug-cc-pVTZ/
ECP
−153.1 (−155 2)9,10
−289.2
reaction
1
2
Th + F → ThF −156.1
Th + F2 →
ThF2
−292.6
−164.0
−153.5
−273.1
−156.0
−16.3 [0.71 eV]
134.6
3
ThF + F →
ThF2
−162.9
4
ThF2 + F →
ThF3
−153.4
5
ThF2 + F2 →
ThF4
−275.4
6
ThF3 + F →
ThF4
−157.9
7
ThF4 + e−
→
−9.8 [0.42 eV]
127.6
−
ThF4
8
ThF5 → ThF3
+ F2
+
9
ThF5 → ThF3
204.3
211.1
−
+ F2
The matrix site-split doublet at (575.9, 575.1) cm−1 observed
in all experiments with F2, NF3, OF2, and HF increased slightly
in intensity upon 20 K annealing and increased substantially in
intensity upon photolysis. This band is the most prominent
product after ThF4. Our CCSD(T)/aug-cc-pVTZ/ECP calcu-
10
11
12
ThF5 → ThF4
+ F
15.0
5.7
ThF5 + e−
→
→
−158.1
−165.4 [−7.17 eV]
−94.7
−
ThF5
[−6.86 eV]
ThF4 + F−
−91.8
−
ThF5
1
lation predicts a C2v A1 state to be the ground state for ThF2
and the strongest mode, the antisymmetric stretch, to be within
C
dx.doi.org/10.1021/ic401107w | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
2 cm−1 of the matrix doublet (Table 1). The weaker symmetric
stretch is computed to be 4 cm−1 higher with 10% of the
strongest band intensity, which is too weak to observe here.
The 575 cm−1 band was labeled ThF2 in Figure 1 of our report
on the reaction of Th with NF3 where N−ThF3 was the major
product with a strong antisymmetric ThF3 subunit stretching
absorption at 525 cm−1.12 At that time, it was thought that the
much weaker (calculations predict 8%) symmetric mode might
be covered by the ThF2 band and 575 cm−1 was also listed in
the tables for this weaker symmetric mode. Further
examination of our Th + NF3 reaction spectra reveals a weaker
band at 566 cm−1, which is tentatively assigned to this weaker
mode for N−ThF3. Observation of the same (575.9, 575.1)
cm−1 band in Th experiments with F2, NF3, OF2, and HF and
our frequency calculations demonstrate that this absorption is
due to ThF2. The slight increase of these bands upon annealing
to 20 K suggests that the Th + F2 reaction is spontaneous,
which is consistent with the large negative energy for reaction 2
(Table 3).
Our CCSD(T) calculations predict ThF3 to be a trigonal-
planar radical with a very strong antisymmetric stretching
fundamental at 537 cm−1 and no symmetric-mode intensity as
long as the radical remains planar. Thus, both the 531.0 and
528.4 cm−1 bands, just above this mode at 525 cm−1 for N−
ThF3,12 merit consideration for ThF3 itself. Note that the latter
band is a weak shoulder on the former band using 0.5% F2 and
the former band is a weak shoulder on a much stronger latter
band with 1% F2 (Figure 1). Then note that only the 531.0
cm−1 band is observed upon sample deposition with the NF3,
OF2 and HF reagents, where the ablation plume photolysis
byproduct F2 concentration is very low; however, final
annealing produces a weak 528.2 or 528.4 cm−1 satellite peak
(Figure 2). This and the understanding that molecular
perturbations usually red shift isolated molecule frequencies
lead us to assign the 531.0 cm−1 band to isolated ThF3 and the
528.4 cm−1 band for ThF3 perturbed by a fluorine molecule, the
most abundant molecule present in the Ar/F2 matrix samples.
Our B3LYP calculations predict a 3 cm−1 red shift for this
interaction. The ThF3 radical is probably made here by the F-
atom reaction with ThF2, reaction 4 in Table 3. The major
product in these experiments, ThF4, can be made by both the
F-atom and molecule reactions 5 and 6 in Table 3. It is
interesting that the first F-atom elimination energies for
ThF1,2,3,4 are nearly the same, but this is highest for ThF2
(Table 3), as is its stretching frequency.
Figure 3. (a) Structures and natural charges for the lowest-energy
−
+
−
conformers of ThF4 , [ThF3 ][F2 ], and ThF5− and (b) spin densities
−
+
−
for ThF4 (D4h) (0.95 on Th), [ThF3 ][F2 ] (0.51 on each F in the
−
−
F2 fragment and −0.02 on Th), and ThF4 (Td) (0.98 on Th),.
half-maximum) to the strong antisymmetric stretching mode(s)
+
−
for the ThF3 subunit in the [ThF3 ][F2 ] complex, which are
predicted by DFT to be separated by 3 cm−1 and would be
degenerate under 3-fold symmetry. Although the ThF3 subunit
+
−
in the [ThF3 ][F2 ] complex is nonplanar, the symmetric mode
associated with the ThF3 subunit is expected to have a small IR
intensity, and this weaker mode is not observed. Note that our
calculations predict the symmetric mode to be near 580 cm−1
with ∼15% of the total antisymmetric-mode intensities at 534
and 531 cm−1 (Table 2). The weaker F−F stretching mode
calculated at 457 cm−1 is not observed because this band may
be broad and matrix-shifted too low for detection.
−
The related alkali-metal (M+)(F2 ) species were observed by
−
matrix Raman spectroscopy, and the F2 stretching modes
ranged from 452 to 475 cm−1 for different alkali-metal reactions
with fluorine.33 We calculated the isolated F2 fundamental to
−
be at 435 cm−1 (ωe = 446.6 cm−1 and ωexe = 6.0 cm−1) at the
CCSD(T)/aug-cc-pVTZ/ECP level with r(F−F) = 1.927 Å.34
It is typical for (cation)(anion) paired species to have higher
anion frequencies than isolated anions, as found for [M+][F2 ]
−
+
−
and the current [ThF3 ][F2 ] species. As another example,
[M+][F3 ] has the antisymmetric F−F−F stretching funda-
−
mental at 550 cm−1 for M = K, Rb, Cs relative to an isolated
17,35
F3 value at 510 cm−1 in solid argon.
It is useful to note a
−
Our DFT B3LYP/aug-cc-pVTZ-ECP calculations showed
that the approach of F2 to ThF3 on the C3 axis leads directly to
ThF4 and an F atom, but if the F−F molecule is not restricted
to this axis, it converges without barrier to the Cs side-bound
structure. We found no minimum for an end-bound structure.
A sideways approach forms a (Cs 2A″) (ThF3)(F2) complex
very exothermically, reaction 8 in Table 3. On the basis of the
geometry parameters, natural charge analysis,32 and the spin
weak F3 band at 510 cm−1 in our solid argon experiments,
which verifies the presence of laser-ablated electrons and F− in
fluorine bearing argon condensed with metal ablation.19
−
+
−
Further support for our observation of the [ThF3 ][F2 ]
complex in these fluorine-rich samples is the high stability of
ThF3(F2) with respect to dissociation to ThF3 + F2 (reaction 8,
Table 3). The Th−F bond in ThF4 is very strong (158 kcal/
mol; reaction 6, Table 3), so it is not surprising that this ThF5
isomer is stable by only 6 kcal/mol with respect to the ThF4 +
F asymptote (reaction 10, Table 3). Thus, ThF5 can only be
stabilized at low temperatures, such as those in our matrix
isolation experiments.
+
−
density (Figure 3), (ThF3)(F2) is best written as [ThF3 ][F2 ].
The calculated frequencies for the ground state (Cs 2A″)
+
−
[ThF3 ][F2 ] complex are given in Table 2, and the structure is
illustrated in Figure 3. We used the B3LYP/aug-cc-pVTZ/ECP
calculation frequency predictions for this larger open-shell
molecule because of the computational cost at the CCSD(T)
level. The good agreement between the CCSD(T) and DFT/
B3LYP results for the frequencies in Table 1 and for the
geometries in Table 2 shows that the DFT frequencies are
reliable. We assign the 528.4 cm−1 band (3 cm−1 full-width at
Our calculations further show that electron capture by
+
−
[ThF3 ][F2 ] is a highly exothermic process (reaction 11,
Table 3), leading to the trigonal-bipyramidal ThF5− anion with
strong antisymmetric Th−F stretching modes predicted in the
446−464 cm−1 range (Table 2). The broader 460 cm−1 band
D
dx.doi.org/10.1021/ic401107w | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
shows the decrease upon annealing expected for an isolated
molecular ion, and its increase upon UV irradiation, with
destruction of the F3− band at 510 cm−1, is expected for the more
ACKNOWLEDGMENTS
■
We gratefully acknowledge financial support from the DOE
Office of Science, Basic Energy Sciences, supported by DOE
Grant DE-SC0001034 (L.A.) and the BES SISGR program in
actinide sciences (D.A.D.). D.A.D. thanks the Robert Ramsay
Fund at the University of Alabama and Argonne National
Laboratory for partial support. L.A. appreciates valuable
assistance with the higher-fluorine-concentration experiments
and helpful discussions with J. Metzger, T. Vent-Schmidt, and
S. Riedel at the University of Freiburg.
−
stable anion. Thus, the 460 cm−1 band is assigned to the ThF5
anion, and it probably contains the absorptions for both
antisymmetric Th−F stretching modes. Another highly
exothermic process possible here is fluoride-anion capture by
ThF4, reaction 12 in Table 3. As noted above, observation of
−
the F3 band provides evidence for F− reactions in these
experiments because the fluoride affinity of F2 is 23 kcal/mol.17
Similar experiments with U and F2 produced strong UF6 and
36,37
weaker UF6 anion absorptions at 620 and 520 cm−1.
The
−
REFERENCES
electron affinity of ThF4 is much smaller than that of UF6, only
0.42 eV (reaction 7 in Table 3), and there are two ThF4
■
−
(1) Grenthe, I.; Drozdzyn
ski, J.; Fujino, T.; Buck, E. C.; Albrecht-
̇
̇
structures that are very close in energy. The lowest-energy
structure is the planar D4h structure.38 The D4h conformer has a
predicted very strong eu mode at 456 cm−1. It is possible that
Schmitt, T. E.; Wolf, S. F. Uranium. In The Chemistry of the Actinide
and Transactinide Elements; Morss, L. R., Edelstein, N. M., Fuger, J.,
Eds.; Springer: Dordrecht, The Netherlands, 2006; pp 253−698.
(2) (a) McDowell, R. S.; Asprey, L. B.; Paine, R. T. J. Chem. Phys.
1974, 61, 3571−3580. (b) Grzybowski, J. M.; Andrews, L. J. Chem.
Phys. 1978, 68, 4540−4545.
(3) Beauchamp, J. L. J. Chem. Phys. 1976, 64, 929−935.
(4) Compton, R. N. J. Chem. Phys. 1977, 66, 4478−4485.
(5) de Jong, W. A.; Nieuwpoort, W. C. Int. J. Quantum Chem. 1996,
58, 203−216.
(6) Buchler, A.; Berkowitz-Mattuck, J. B.; Dugre, D. H. J. Chem. Phys.
1961, 34, 2202−2203.
(7) Buchmarina, V. N.; Gerasimov, A. Yu.; Predtechenskii, Yu. B.;
Shklyarik, V. G. Opt. Spectrosc. (USSR) 1992, 72, 38−40.
(8) Konings, R. J. M.; Hildenbrand, D. L. J. Alloys Compd. 1998,
271−273, 583−586.
(9) Barker, B. J.; Antonov, I. O.; Heaven, M. C.; Peterson, K. A. J.
Chem. Phys. 2012, 136, 104305.
−
broad absorption near 450 cm−1 is due to ThF4 , formed
through electron capture by the major reaction product ThF4.
It is interesting to note that this possible frequency ratio for the
strongest mode of the D4h ThF4 anion and the corresponding
neutral, 450/521 = 0.86, is almost the same as that for the UF6
anion and the neutral, 520/620 = 0.84. The Td structure of
−
ThF4 is 2.9 kcal/mol higher in energy at the CCSD(T)/aug-
cc-pVTZ/ECP level and only 0.5 kcal/mol higher at the DFT/
B3LYP level. The intense band for Td ThF4− is predicted to be
at 480 cm−1 at the DFT/B3LYP level.
CONCLUSIONS
■
We have shown that reactions of laser-ablated Th with F2
produce the expected small thorium fluoride molecules ThF,
ThF2, ThF3, and ThF4 on the basis of the behavior of their
argon matrix IR spectra upon increasing sample concentration,
annealing, and irradiation, the use of complementary NF3, OF2,
and HF precursor reactions, and a comparison with frequencies
calculated by the CCSD(T) and DFT/B3LYP levels using the
aug-cc-pVTZ/ECP basis set. Additional lower-frequency
absorptions are assigned to the thorium pentafluoride species
(10) Lau, K. H.; Brittain, R. D.; Hildenbrand, D. L. J. Chem. Phys.
1989, 90, 1158−1164.
(11) Gong, Y.; Wang, X. F.; Andrews, L.; Schloder, T.; Riedel, S.
̈
Inorg. Chem. 2012, 51, 6983−6991.
(12) Wang, X. F.; Andrews, L. Dalton Trans. 2009, 9260−9265.
(13) Andrews, L.; Gong, Y.; Liang, B.; Jackson, V. E.; Flamerich, R.;
Li, S.; Dixon, D. A. J. Phys. Chem. A 2011, 115, 14407−14416.
(14) Zhou, M. F.; Andrews, L. J. Chem. Phys. 1999, 111, 11044−
11049.
+
−
−
[ThF3 ][F2 ] and ThF5 , which are formed in highly
exothermic reactions. This is the first characterization of new
pentacoordinated thorium species including the stable molec-
(15) Souter, P. F.; Kushto, G. P.; Andrews, L.; Neurock, M. J. Phys.
Chem. A 1997, 101, 1287−1291.
+
−
(16) Andrews, L.; Citra, A. Chem. Rev. 2012, 102, 885−911.
(17) Andrews, L. Chem. Soc. Rev. 2004, 33, 123−132.
(18) Andrews, L.; Cho, H.-G. Organometallics 2006, 25, 4040−4053.
ular cation−molecular anion [ThF3 ][F2 ] and the isolated
−
ThF5− anion. We believe that it will be possible to prepare the
ThF5 anion with an appropriate stabilizing cation in a solid-
state compound.
(19) Riedel, S.; Kochner, T.; Wang, X. F.; Andrews, L. Inorg. Chem.
̈
2010, 49, 7156−7164.
(20) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; et al. Gaussian 09, revision B.01; Gaussian, Inc.:
Wallingford, CT, 2009.
ASSOCIATED CONTENT
* Supporting Information
■
S
Complete author lists for refs 20 and 21, figures of IR spectra
for Th with HF and higher-concentration F2 reaction products,
optimized Cartesian x, y, and z coordinates in angstroms for
ThF5, total energies at the CCSD(T) and DFT/B3LYP levels
in atomic units, and the complete list of vibrational frequencies
in reciprocal centimeters. This material is available free of
charge via the Internet at http://pubs.acs.org.
(21) Knowles, P. J.; Manby, F. R.; Schutz, M.; Celani, P.; Knizia, G.;
̈
Korona, T.; Lindh, R.; Mitrushenkov, A.; Rauhut, G.; Adler, T. B.;
et al. MOLPRO, version 2010.1, a package of ab initio programs. See
http://www.molpro.net.
(22) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652.
(23) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785−789.
(24) Purvis, G. D., III; Bartlett, R. J. J. Chem. Phys. 1982, 76, 1910−
1918.
AUTHOR INFORMATION
Corresponding Author
*E-mail: lsa@virginia.edu (L.A.), dadixon@bama.ua.edu
(D.A.D.).
(25) Raghavachari, K.; Trucks, G. W.; Pople, J. A.; Head-Gordon, M.
■
Chem. Phys. Lett. 1989, 157, 479−483.
(26) Watts, J. D.; Gauss, J.; Bartlett, R. J. J. Chem. Phys. 1993, 98,
8718−8733.
(27) (a) Deegan, M. J. O.; Knowles, P. J. Chem. Phys. Lett. 1994, 227,
321−326. (b) Rittby, M.; Bartlett, R. J. J. Phys. Chem. 1988, 92, 3033−
3036.
Notes
The authors declare no competing financial interest.
E
dx.doi.org/10.1021/ic401107w | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(28) Kendall, R. A.; Dunning, T. H., Jr.; Harrison, R. J. J. Chem. Phys.
1992, 96, 6796−6806.
(29) (a) Kuchle, W.; Dolg, M.; Stoll, H.; Preuss, H. J. Chem. Phys.
̈
1994, 100, 7535−7542. (b) Institut fur Theoretische Chemie
̈
Universitat Stuttgart, http://www.theochem.uni-stuttgart.de/
̈
pseudopotentials/index.en.html. (c) Cao, X.; Dolg, M.; Stoll, H. J.
Chem. Phys. 2003, 118, 487−496. (d) Cao, X.; Dolg, M. J. Mol. Struct.
(THEOCHEM) 2004, 673, 203−209.
(30) Gong, Y.; Andrews, L.; Jackson, V. E.; Dixon, D. A. Inorg. Chem.
2012, 51, 11055−11060.
(31) Sansonetti, J. E.; Martin, W. C. J. Phys. Chem. Ref. Data 2005,
34, 1559−2259. See also http://physics.nist.gov/PhysRefData/
Handbook/Tables/iridiumtable5.htm.
(32) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988, 88,
899−926. Reed, A. E.; Weinstock, R. B.; Weinhold, F. J. Chem. Phys.
1985, 83, 735−746. Weinhold, F.; Landis, C. R. Valency and Bonding;
Cambridge University Press: Cambridge, U.K., 2005. Glendening, E.
D., Reed, A. E., Carpenter, J. E., Weinhold, F. NBO, version 3.1;
Department of Chemistry, University of WisconsinMadison:
Madison, WI, 1988.
(33) Howard, W. F., Jr.; Andrews, L. J. Am. Chem. Soc. 1973, 95,
3045−3046.
(34) DFT/B3LYP gives a longer bond distance of 2.006 Å and a
lower ωe of 357 cm−1.
(35) Ault, B. S.; Andrews, L. J. Am. Chem. Soc. 1976, 98, 1591−1592.
(36) Hunt, R. D.; Thompson, C.; Hassanzadeh, P.; Andrews, L. Inorg.
Chem. 1994, 33, 388−391 and references cited therein.
(37) Kunze, K. R.; Hauge, R. H.; Hamill, D.; Margrave, J. L. J. Chem.
Soc., Dalton Trans. 1978, 433−440.
(38) There are two small imaginary frequencies at both the
CCSD(T) (46i and 52i) and DFT/B3LYP levels. Following these
frequencies led to a higher-energy C4v structure with a single imaginary
frequency or to the Td structure.
F
dx.doi.org/10.1021/ic401107w | Inorg. Chem. XXXX, XXX, XXX−XXX