Noble-gas difluoride complexes of mercury(II): The syntheses and structures of Hg(OTeF5)2·1.5NgF2 (Ng = Xe, Kr) and Hg(OTeF5)2

Article abstract of DOI:10.1021/ja412193z

The synthesis of high-purity Hg(OTeF₅)₂ has resulted in its structural characterization in the solid state by Raman spectroscopy and single-crystal X-ray diffraction (XRD) and in solution by ¹⁹F NMR spectroscopy. The crystal structure of Hg(OTeF₅)₂ (-173 C) consists of discrete Hg(OTeF₅)₂ units having gauche-conformations that interact through long Hg - -O and Hg - -F intramolecular contacts to give a chain structure. The Lewis acidity of Hg(OTeF₅)₂ toward NgF₂ (Ng = Xe, Kr) was investigated in SO₂ClF solvent and shown to form stable coordination complexes with NgF₂ at -78 C. Both complexes were characterized by low-temperature Raman spectroscopy (-155 C) and single-crystal XRD. The complexes are isostructural and are formulated as Hg(OTeF₅) ₂·1.5NgF₂. The Hg(OTeF₅)₂ units of Hg(OTeF₅)₂·1.5NgF₂ also have gauche-conformations and are linked through bridging NgF₂ molecules, also resulting in chain structures. These complexes represent the only examples of coordination compounds where NgF₂ coordinates to mercury in a neutral covalent compound and the only example of mercury coordinated to KrF₂. Moreover, the Hg(OTeF₅)₂·1. 5KrF₂ complex is the only KrF₂ complex known to contain a bridging KrF₂ ligand. Energy-minimized gas-phase geometries and vibrational frequencies for the model compounds, [Hg(OTeF₅) ₂]₃ and [Hg(OTeF₅)₂] ₃·2NgF₂, were obtained and provide good approximations of the local environments of Hg(OTeF₅)₂ and NgF₂ in the crystal structures of Hg(OTeF₅)₂ and Hg(OTeF₅)₂·1.5NgF₂. Assignments of the Raman spectra of Hg(OTeF₅)₂ and Hg(OTeF ₅)₂·1.5NgF₂ are based on the calculated vibrational frequencies of the model compounds. Natural bond orbital analyses provided the associated bond orders, valencies, and natural population analysis charges.

Full text of DOI:10.1021/ja412193z

Article
pubs.acs.org/JACS
Noble-Gas Difluoride Complexes of Mercury(II): The Syntheses and
Structures of Hg(OTeF5)2·1.5NgF2 (Ng = Xe, Kr) and Hg(OTeF5)2
John R. DeBackere, Helene P. A. Mercier, and Gary J. Schrobilgen*
́
̀
Department of Chemistry, McMaster University, Hamilton, Ontario, L8S 4M1, Canada
S
* Supporting Information
ABSTRACT: The synthesis of high-purity Hg(OTeF5)2 has resulted in its
structural characterization in the solid state by Raman spectroscopy and
single-crystal X-ray diffraction (XRD) and in solution by 19F NMR
spectroscopy. The crystal structure of Hg(OTeF5)2 (−173 °C) consists of
discrete Hg(OTeF5)2 units having gauche-conformations that interact through
long Hg---O and Hg---F intramolecular contacts to give a chain structure. The
Lewis acidity of Hg(OTeF5)2 toward NgF2 (Ng = Xe, Kr) was investigated in
SO2ClF solvent and shown to form stable coordination complexes with NgF2
at −78 °C. Both complexes were characterized by low-temperature Raman
spectroscopy (−155 °C) and single-crystal XRD. The complexes are
isostructural and are formulated as Hg(OTeF5)2·1.5NgF2. The Hg(OTeF5)2
units of Hg(OTeF5)2·1.5NgF2 also have gauche-conformations and are linked
through bridging NgF2 molecules, also resulting in chain structures. These
complexes represent the only examples of coordination compounds where NgF2 coordinates to mercury in a neutral covalent
compound and the only example of mercury coordinated to KrF2. Moreover, the Hg(OTeF5)2·1.5KrF2 complex is the only KrF2
complex known to contain a bridging KrF2 ligand. Energy-minimized gas-phase geometries and vibrational frequencies for the
model compounds, [Hg(OTeF5)2]3 and [Hg(OTeF5)2]3·2NgF2, were obtained and provide good approximations of the local
environments of Hg(OTeF5)2 and NgF2 in the crystal structures of Hg(OTeF5)2 and Hg(OTeF5)2·1.5NgF2. Assignments of the
Raman spectra of Hg(OTeF5)2 and Hg(OTeF5)2·1.5NgF2 are based on the calculated vibrational frequencies of the model
compounds. Natural bond orbital analyses provided the associated bond orders, valencies, and natural population analysis
charges.
reported. In contrast, examples of KrF2 coordination complexes
are very rare. The complex, [BrOF2][AsF6]·2KrF2, represents
the first KrF2 coordination complex to have been structurally
characterized by X-ray crystallography and contains two
terminally coordinated KrF2 molecules.15 Several terminally
coordinated NgF2 complexes of group 6 d0 transition-metal
oxide tetrafluorides, i.e., XeF2·nMOF4 (n = 1−3, M = W,16,17
Mo;17 n = 4, M = Mo17) and KrF2·nMOF4 (n = 1, M = W,18
Mo,18 Cr19); n = 2−3, M = Mo18) have also been synthesized
and characterized in the solid state by Raman spectroscopy
and/or in solution by 19F and 129Xe multi-NMR spectroscopy.
A low-precision, room-temperature X-ray crystal structure of
XeF2·WOF4 has also been reported.20
The pentafluoro-orthotellurate group, −OTeF5, may be
regarded as a bulky fluorine analogue having a group
electronegativity (3.8821 and 3.8722) comparable to that of
fluorine (3.98, Allred-Rochow scale). Negative charge dispersal
over its five fluorine and an oxygen atom results in a ligand
group of low nucleophilicity and high oxidative resistance.23
The steric bulk and propensity of the −OTeF5 group not to
extensively oxygen bridge but to bond in a monodentate
INTRODUCTION
■
The fluoride-ion donor behavior of noble-gas difluorides
toward strong fluoride ion acceptors such as AsF5 and SbF5
is well established, with several examples of NgF+ and Ng2F3
+
(Ng = Xe, Kr) salts having been characterized in the solid state
and in solution.1−3 Avoidance of “complete” fluoride transfer
requires the corresponding Lewis acid center to be weak to
moderate in strength and oxidatively resistant. In the case of
XeF2, two coordination modalities, terminal and bridging, have
been observed.1 Both fluorine atoms of XeF2 may coordinate to
two Lewis acid centers to give a bridging XeF2 ligand, or a
single fluorine atom of XeF2 may coordinate to give a terminal
XeF2 ligand. Examples of both coordination modalities are
known with XeF2 coordinated to a nonmetal ([BrOF2][AsF6]·
2XeF2)4 and to metal cation centers (Li+, Mg2+, Ca2+, Cu2+,
Zn2+, Sr2+, Ag+, Cd2+, Ba2+, La3+, Nd3+, and Pb2+).5−9 The
majority of these complexes have been structurally charac-
terized by single-crystal X-ray diffraction. Among the group 12
metal cation complexes of XeF2 that have been structurally
characterized are Cd(XeF2)(BF4)2,7 Cd2(XeF2)10(SbF6)4,9
Cd(XeF2 )4 (AsF6 )2 , 1 0 Cd(XeF2 )5 (PF6 )2 , 1 1 and
Zn(XeF2)6(SbF6)2.12 Preliminary X-ray crystal structures and
Raman studies of the Hg2 + cation complexes
Received: November 29, 2013
Published: February 3, 2014
13
14
Hg(XeF2)5(AsF6)2 and Hg(XeF2)5(SbF6)2 have also been
© 2014 American Chemical Society
3888
dx.doi.org/10.1021/ja412193z | J. Am. Chem. Soc. 2014, 136, 3888−3903

Journal of the American Chemical Society
Article
fashion,24−26 is expected to result in molecular species that have
smaller mercury coordination numbers in the solid state than
its fluorine analogue, HgF2, which possesses a three-dimen-
sional, networked structure (CN = 8, fluorite structure).27
Overall, the less crowded coordination environment of Hg(II)
in Hg(OTeF5)2 provides more space for donor ligand
molecules such as NgF2 to coordinate to Hg(II).
In the present study, Hg(OTeF5)2 has been synthesized in
high purity and yield, along with its isomorphous noble-gas
difluoride coordination complexes, Hg(OTeF5)2·1.5NgF2 (Ng
= Xe, Kr), and structurally characterized by low-temperature
single-crystal X-ray diffraction, Raman spectroscopy, and
quantum-chemical calculations. Mercury(II) bis(pentafluoro-
orthotellurate(VI)) was also characterized in solution by 19F
NMR spectroscopy in the present and past28,29 studies. The
present work provides a significant extension of the little
studied coordination chemistry of KrF2 by providing the only
example of a bridging KrF2 molecule that is currently known
and insight into the coordination behavior and Lewis acidity of
Hg(II) in the neutral Hg(OTeF5)2 molecule.
1J(123Te−19FB) = 2982 Hz). The δA and δB chemical shifts and
2J(19FA-19FB) coupling constant are in good agreement with the
previously reported values: −38.228 [−38.1],29 −40.128
[−40.2]29 ppm and 18628 [185]29 Hz in CH2Cl2 and −27.828
[−27.4],29 −40.428 [−40.3]29 ppm and 18028 [180]29 Hz in
1
CH3CN. Accurate J(123,125Te−19FA,B) couplings are reported
here for the first time; only one coupling was previously
reported (3560 Hz in CH2Cl2 and 3580 Hz in CH3CN) which
was described as a J(125Te−19F) coupling but was not
specifically assigned to FA or FB.28
Synthesis of Hg(OTeF5)2·1.5XeF2. The Hg(OTeF5)2·
1.5XeF2 complex was synthesized by adding a 1:1.7 molar
excess of XeF2 to Hg(OTeF5)2 at −140 °C. Sulfuryl chloride
fluoride was condensed onto the mixture at −78 °C, followed
by warming to 0 °C for 5 min, whereupon the solid dissolved.
The mixture was maintained at −78 °C for 5 days prior to
removing the solvent under dynamic vacuum at −78 °C,
leaving behind a white solid corresponding to Hg(OTeF5)2·
1.5XeF2. When the solid complex was warmed to room
temperature under 1 atm of dry N2, slow dissociation into
Hg(OTeF5)2 and XeF2 occurred and was complete within 6
days (eq 2).
RESULTS AND DISCUSSION
■
The purities of all products and progress of all reactions were
routinely monitored by periodically quenching the reactions at
−196 °C and recording the low-temperature Raman spectra
(−150 °C) of the reaction mixtures in the solid state or in
frozen solutions.
Hg(OTeF) ·1.5XeF → Hg(OTeF) + 1.5XeF
(2)
5 2
2
5 2
2
Synthesis of Hg(OTeF5)2·1.5KrF2. The Hg(OTeF5)2·
1.5KrF2 complex was synthesized by the reaction of
Hg(OTeF5)2 and KrF2 (1:2.2 molar ratio) in SO2ClF solvent.
The mixture was warmed to −20 °C for 2 min and then
maintained at −78 °C for 3 h. Removal of the solvent at −78
°C resulted in Hg(OTeF5)2·1.5KrF2 as a white solid. When
warmed to 0 °C under 1 atm of dry N2 for 3 h, the
Hg(OTeF5)2·1.5KrF2 complex underwent a redox decomposi-
tion to form HgF2, F5TeOOTeF5, KrF2, and Kr according to eq
3. The formation of HgF2 and F5TeOOTeF5 was confirmed by
quenching the decomposition reaction in the Raman
spectrometer and recording its spectrum at −150 °C. The
Raman spectrum showed two strong bands at 254 (HgF2) and
464 cm−1 (KrF2) as well as bands characteristic of
F5TeOOTeF5, in particular, a strong band corresponding to
the O−O stretch of F5TeOOTeF5 appeared at 899 cm−1.
Synthesis of Hg(OTeF5)2. The synthesis of Hg(OTeF5)2 is
based on a modification of prior synthetic procedures28,29 and
was accomplished by the reaction of high-purity HgF2 (see
Supporting Information) with a small molar excess of HOTeF5
(1:2.05) at 50 °C for several hours (eq 1). The Raman
spectrum of pure Hg(OTeF5)2 is provided in Figure S1.
HgF + 2HOTeF → Hg(OTeF) + 2HF
5
5 2
(1)
2
The literature procedures for the synthesis of Hg(OTeF5)2
call for purification of the product by sublimation at 18028 or
200 °C.29 In the present work, the Raman spectrum of
sublimed Hg(OTeF5)2 (135−165 °C) showed several addi-
tional weak bands (Figure S2) that do not appear in the Raman
spectrum of unsublimed Hg(OTeF5)2 (Figure S1), indicating
that some decomposition had occurred during sublimation;
however, the decomposition products have not been identified.
The Raman bands associated with the decomposition products
may have been too broad and weak at room temperature to be
observed in the prior studies. The main bands observed in the
presently reported low-temperature Raman spectrum of the
product are in agreement with those previously obtained for
Hg(OTeF5)2 by room-temperature Raman spectroscopy
(CH2Cl2 solution)28 and room-temperature infrared spectros-
copy (solid in a CsBr pellet28 or in a nujol mull29).
Hg(OTeF) ·1.5KrF
5 2
2
1
→ HgF + F TeOOTeF + Kr +
KrF
2
2
5
2
(3)
2
The absence of Hg(OTeF5)2 as a decomposition product
indicates that oxidative fluorination of Hg(OTeF5)2 by KrF2
rather than dissociation of the complex occurs at elevated
temperatures. A previous computational study predicted that
HgF2 and F5TeOOTeF5 will be the dominant products
resulting from the decomposition of F2Hg(OTeF5)2 (eq 4).30
Thus, the formation of F2Hg(OTeF5)2 as an intermediate in
the decomposition pathway of the complex (eqs 4 and 5)
cannot be ruled out.
The current synthetic procedure produces Hg(OTeF5)2 in
high yield and purity without the need for further purification
and has allowed its full characterization in the solid state by
low-temperature single-crystal X-ray diffraction, Raman spec-
troscopy, and in solution by 19F NMR spectroscopy.
1
Hg(OTeF) ·1.5KrF → F Hg(OTeF) + Kr +
KrF
2
5 2
2
2
5 2
2
(4)
The 19F (470.568 MHz) NMR spectrum of Hg(OTeF5)2 was
obtained in CD2Cl2 at 25 °C. The spectrum was second order
and consisted of an AB4 pattern with the A resonance (δA =
F Hg(OTeF) → F TeOOTeF + HgF
2
5 2
5
5
(5)
2
2
1
−38.4 ppm; J(19FA−19FB) = 185 Hz; J(125Te−19FA) = 3363
Hz; 1J(123Te−19FA) = 2794 Hz) occurring to high frequency of
the B4 resonance (δB = −42.1 ppm; 1J(125Te−19FB) = 3603 Hz;
X-ray Crystallography. Details of the data collection
parameters and other crystallographic information for
Hg(OTeF5)2 and Hg(OTeF5)2·1.5NgF2 (Ng = Xe, Kr) are
3889
dx.doi.org/10.1021/ja412193z | J. Am. Chem. Soc. 2014, 136, 3888−3903

Journal of the American Chemical Society
Article
provided in Table 1 and the bond lengths and angles are listed
in Tables 2 and 3.
Another interesting structural feature is the O−Hg−O angle,
which deviates significantly from linearity (170.5(4)°) in the
crystal structure. The deformation, albeit smaller, is reproduced
in the calculated gas-phase structure of [Hg(OTeF5)2]3
(176.0°) which also reproduced the Hg---O secondary contacts.
It is therefore likely that this feature predominantly results from
crystal packing in addition to the steric demands of the
−OTeF5 ligands. Moreover, the experimental structure contains
additional Hg---F(Te) contacts that may further contribute to a
decrease in the O−Hg−O angle owing to their steric demands.
These contacts originate from Hg---F(Te) interactions between
adjacent Hg(OTeF5)2 units within a chain and approach the
Hg(II) atom in a direction opposite to the direction toward
which the O−Hg−O angle is bent.
Hg(OTeF5)2·1.5NgF2 (Ng = Xe, Kr). The coordination
complexes, Hg(OTeF5)2·1.5NgF2 are isomorphous and belong
to the P21/n space group. The crystal structures of
Hg(OTeF5)2·1.5NgF2 are analogous, containing Hg(OTeF5)2
units that are linked to one another through bridging NgF2
molecules (Figures S4 and 2). As expected, the Hg(OTeF5)2·
1.5XeF2 complex has a slightly larger unit cell (Table 1),
reflecting the larger covalent radius of xenon (1.40 Å) relative
to that of krypton (1.16 Å).33
Table 1. Summary of Crystal Data and Refinement Results
for Hg(OTeF5)2 and Hg(OTeF5)2·1.5NgF2 (Ng = Xe, Kr)
Hg(OTeF5)2·
Hg(OTeF5)2·
chem formula
space group
Hg(OTeF5)2
1.5XeF2
1.5KrF2
C2/c
P21/n
P21/n
a (Å)
b (Å)
c (Å)
β (°)
V (Å3)
18.681(1)
8.6489(4)
5.8008(3)
96.215(3)
931.74(1)
4
9.0574(2)
17.8943(3)
9.1285(2)
114.548(1)
1345.78(6)
2
8.7981(4)
17.4153(9)
9.2243(4)
113.788(3)
1293.3(2)
2
Z (molecules/unit
cell)
M (g mol−1
)
677.79
4.832
931.74
4.599
860.49
4.419
ρcalcd (g cm−3
)
T (°C)
−173
22.80
−173
19.55
−173
21.57
μ (mm−1
)
a
R1
0.0461
0.1232
0.0282
0.0595
0.0284
0.0716
b
wR2
a
b
R1 is defined as Σ∥Fo| − |Fc∥/Σ|Fo| for I > 2σ (I). wR2 is defined as
[Σ[w(Fo − Fc2)2]/Σw(Fo )2]1/2 for I > 2σ(I).
2
2
The mercury coordination spheres of Hg(OTeF5)2·1.5NgF2
are comparable to that of Hg(OTeF5)2 (vide supra), consisting
of very distorted pseudo-octahedra (Figure 3 and Table 3). The
Hg(OTeF5)2 units likewise have gauche-conformations with
Hg−O bonds trans to one another. The contacts with mercury
include two shorter Hg---F contacts (Xe, 2.606(5) and
2.623(4) Å; Kr, 2.664(3) and 2.675(3) Å) that are trans to
one another and two slightly longer Hg---F and Hg---O
contacts (Xe, 2.701(5) and 2.749(4) Å; Kr, 2.741(3) and
2.725(3) Å, respectively) that are trans to one another. The
Hg---F contacts originate from the fluorine ligands of three
nonequivalent NgF2 molecules and the Hg---O contact from an
oxygen atom of an adjacent Hg(OTeF5)2 group. These contacts
are significantly less than the sums of their respective van der
Waals radii and indicate significant covalent interactions. The
Hg---F and Hg---O contacts within the NgF2 complexes are
shorter and longer, respectively, than those within the chain
networks of Hg(OTeF5)2 (Hg---F, 2.810(7) and Hg---O,
2.641(7) Å). These secondary bonding interactions, which
are similar to the secondary bonding interactions found in the
crystal structure of Hg(OTeF5)2, presumably favor the
observed gauche-conformation of the Hg(OTeF5)2 unit.
Although both structures are similar, the secondary bond
distances between mercury and the fluorine atoms of NgF2 are
somewhat shorter in Hg(OTeF5)2·1.5XeF2 than in the KrF2
analogue (Table 3), which is consistent with the more polar
characters of the Xe−F bonds.3
The Te−F bond lengths (Xe, 1.814(5)−1.845(4) Å; Kr,
1.824(3)−1.848(3) Å) and F−Te−F angles (Xe, 85.5(2)−
91.8(2)° and 172.1(2)−173.6(3)°; Kr, 86.2(2)−91.3(2)° and
172.5(1)−174.3(1)°) are comparable in the krypton and xenon
analogues and to those of Hg(OTeF5)2 (vide supra). The
Hg−O (Xe, 2.015(5) and 2.037(5) Å; Kr, 2.017(3) and
2.029(3) Å) and Te−O (Xe, 1.811(6) and 1.815(5) Å; Kr,
1.819(3) and 1.836(3) Å) bond lengths are all equal (within
3σ) to those of Hg(OTeF5)2. The effects of NgF2
coordination to mercury are reflected in the O−Hg−O (Xe,
173.0(2)°; Kr, 173.3(1)°) and Hg−Te−O (Xe, 132.8(3) and
127.2(3)°; Kr, 129.6(2) and 126.5(1)°) angles, which are
slightly larger (within 3σ) than those of Hg(OTeF5)2, and in
Hg(OTeF5)2. Unlike the three-dimensional network structure
of HgF2, the structure of Hg(OTeF5)2 consists of discrete
Hg(OTeF5)2 units that interact through long Hg---O and
Hg---F intramolecular contacts, resulting in chains that run
parallel to the c-axis of its crystallographic unit cell (Figures 1a
and S3A). The adjacent chains form isolated layers along the a-
and b-axes with no F···F distances between them that are less
than the sum of twice the fluorine van der Waals radii (2.94 Å,
Figure S3b).
The Hg(II) coordination sphere is a distorted octahedron
consisting of secondary Hg---O (2 × 2.641(7) Å) and slightly
longer Hg---F contacts (2 × 2.810(7) Å) in addition to the
primary Hg−O bonds (2 × 2.016(6) Å), as illustrated in Figure
1b. The Hg−O bonds are trans to one another, whereas the
Hg---O and Hg---F contacts are cis to one another and to the
primary Hg−O bonds. The secondary contacts are significantly
less than the sums of the van der Waals radii (3.05 Å for Hg···O
and 3.06 Å for Hg···F),31 indicating significant covalent
interactions between Hg(II) and the −OTeF5 groups of
neighboring Hg(OTeF5)2 molecules. Among the secondary
contacts, the Hg---O contacts are the strongest. The Te−O
(1.842(7) Å) and Te−F (1.819(6)−1.839(6) Å) bond lengths
are comparable to those of Xe(OTeF5)2 (1.842(11) and
1.843(11) Å; 1.823(9)−1.855(11) Å).32 An interesting feature
of the Hg(OTeF5)2 structure is the gauche-conformation
adopted by the two −OTeF5 groups in the solid state, with a
dihedral Te−O−Hg−O−Te angle of 53.7(3)°. The gauche-
conformation is attributed to crystal packing and to the
aforementioned Hg---O and Hg---F secondary contacts with
adjacent −OTeF5 groups. This was verified computationally by
showing that the calculated gas-phase geometry of the
unknown trimeric [Hg(OTeF5)2]3 molecule, which possesses
two long Hg---O contacts (2.737 Å), retains the experimental
gauche-conformation (Te−O−Hg−O−Te, 55.8°; see Compu-
tational Results). In contrast, gas-phase monomeric
Hg(OTeF5)2 optimizes to an anti-conformation (dihedral
Te−O−Hg−O−Te angle, 139.1°).
3890
dx.doi.org/10.1021/ja412193z | J. Am. Chem. Soc. 2014, 136, 3888−3903

Journal of the American Chemical Society
Article
Table 2. Experimental Geometrical Parameters for Hg(OTeF5)2 and Calculated Geometrical Parameters for [Hg(OTeF5)2]3
a
b
exptl Hg(OTeF5)2
calcd [Hg(OTeF5)2]3
Bond Lengths (Å)
Hg1−O1
Te1−O1
Te1−F1
Te1−F2
Te1−F3
Te1−F4
Te1−F5
Hg1---O1A
Hg1---O1C
Hg1---F4D
Hg1---F4H
2.016(6)
1.842(7)
1.819(6)
1.824(6)
1.830(6)
1.833(6)
1.839(6)
2.641(7)
2.641(7)
2.810(7)
2.810(7)
Hg16−O22
Te17−O22
Te17−F18
Te17−F21
Te17−F23
Te17−F19
Te17−F20
Hg16---O14
Hg16---O30
2.008
1.865
1.832
1.839
1.848
1.837
1.855
2.737
2.737
Bond Angles (°)
O1−Hg1−O1B
Hg1−O1−Te1
O1−Te1−F1
O1−Te1−F2
O1−Te1−F3
O1−Te1−F4
O1−Te1−F5
F1−Te1−F2
F1−Te1−F3
F1−Te1−F4
F1−Te1−F5
F2−Te1−F4
F4−Te1−F5
F5−Te1−F3
F3−Te1−F2
F2−Te1−F5
F3−Te1−F4
O1−Hg1---F4D
O1−Hg1---F4H
O1−Hg1---O1A
O1−Hg1---O1C
170.5(4)
124.1(3)
178.3(3)
91.0(3)
92.9(3)
90.7(3)
91.8(3)
89.0(3)
88.8(3)
87.6(3)
88.2(3)
89.7(3)
89.1(3)
89.5(3)
91.5(3)
177.0(3)
176.2(3)
105.9(3)
78.3(3)
78.7(3)
93.8(3)
78.3(3)
105.9(3)
155.7(3)
81.3(3)
129.6(3)
O22−Hg16−O37
Hg16−O22−Te17
O22−Te17−F18
O22−Te17−F21
O22−Te17−F23
O22−Te17−F19
O22−Te17−F20
F18−Te17−F21
F18−Te17−F23
F18−Te17−F19
F18−Te17−F20
F21−Te17−F19
F19−Te17−F20
F20−Te17−F23
F23−Te17−F21
F21−Te17−F20
F23−Te17−F19
176.0
122.0
179.6
90.2
91.7
91.6
91.6
89.4
88.2
88.5
88.8
90.2
90.5
88.7
90.5
178.0
176.6
O22−Hg16---O14
O22−Hg16---O30
73.0
103.8
O1A---Hg1---F4D
O1A---Hg1---F4H
O1C---Hg1---F4H
O1A---Hg1---O1C
O14---Hg16---O30
79.7
55.8
F
4D---Hg1---F4H
Dihedral Angles (°)
Te1−O1−Hg1−O1B−Te1B
53.7(3)
Te17−O22−Hg16−O37−Te32
a
b
Atom labeling scheme corresponds to that used in Figure 1b. Calculated at the PBE1PBE/def2-TVZPP level of theory. Atom labeling scheme
corresponds to that used in Figure 6b. Only the parameters associated with the central Hg(OTeF5)2 unit are reported.
the dihedral Te−O−Hg−O−Te angles (Xe, 45.2(5)°; Kr,
50.3(3)°), which are smaller than in Hg(OTeF5)2 (53.7(3)°).
There are two crystallographically inequivalent bridging
NgF2 molecules within the asymmetric units of Hg(OTeF5)2·
1.5NgF2; one Ng atom is positioned on an inversion center,
providing two symmetry-equivalent Ng−F bonds (Xe, 1.981(4)
Å; Kr, 1.883(3) Å), whereas the other Ng atom is on a general
position, giving rise to two symmetry-inequivalent Ng−F bonds
(Xe, 1.991(4) and 2.012(4) Å; Kr, 1.897(3) and 1.885(3) Å).
The Ng−F bond lengths are equal within 3σ to those
observed in free NgF2 (XeF2, 1.999(4) Å;3 KrF2, 1.894(5) Å 2).
The Xe−F bond lengths are comparable to those observed for
the bridging XeF2 molecule in Cd(XeF2)4(AsF6)2 (Xe−Fb,
1.995(5) and 2.017(5) Å)10 and in Cd(XeF2)5(PF6)2
(Xe−Fb, 1.999(6) and 2.016(6) Å).11 The structure of
Hg(OTeF5)2·1.5KrF2 currently represents the only example
of KrF2 coordinated to a neutral metal center and of a bridging
KrF2 molecule. The [BrOF2][AsF6]·2KrF2 complex is the only
other KrF2 complex reported to date that has been
characterized by X-ray crystallography. The Kr−F bridge
bonds of the present complex are shorter than the Kr−F
bridge bonds of the terminally coordinated KrF2 molecules in
[BrOF2][AsF6]·2KrF2 (1.943(4) and 1.933(4) Å),15 indicating
that they are more covalent and more weakly coordinated than
those of [BrOF2][AsF6]·2KrF2.
Raman Spectroscopy. The low-temperature, solid-state
Raman spectra of Hg(OTeF5)2 and Hg(OTeF5)2·1.5NgF2 (Ng
= Xe, Kr) are shown in Figures S1, 4, and 5. Their assignments
are listed in abbreviated form in Tables 4−6 and in more detail
in Tables S1−S4 along with their experimental and calculated
frequencies and intensities.
3891
dx.doi.org/10.1021/ja412193z | J. Am. Chem. Soc. 2014, 136, 3888−3903

Journal of the American Chemical Society
Article
a
Table 3. Experimental Geometrical Parameters for Hg(OTeF5)2·1.5NgF2 (Ng = Xe, Kr)
Ng = Xe
Ng = Kr
Bond Lengths (Å)
Ng = Xe
Ng = Kr
Hg1−O1
Hg1−O2
O1−Te1
O2−Te2
Te1−F1
Te1−F2
Te1−F3
Te1−F4
Te1−F5
Te2−F6
Te2−F7
2.015(5)
2.037(5)
1.815(5)
1.811(6)
1.827(5)
1.831(4)
1.837(4)
1.841(4)
1.845(4)
1.835(5)
1.822(4)
2.017(3)
2.029(3)
1.819(3)
1.836(3)
1.837(3)
1.839(3)
1.844(3)
1.848(3)
1.835(3)
1.838(3)
1.829(3)
Te2−F8
Te2−F9
1.828(5)
1.827(5)
1.814(5)
2.749(4)
2.606(5)
2.623(4)
2.701(5)
1.981(4)
1.981(4)
2.012(4)
1.991(4)
1.836(3)
1.836(3)
1.824(3)
2.725(3)
2.664(3)
2.675(3)
2.741(3)
1.883(3)
1.883(3)
1.897(3)
1.885(3)
Te2−F10
Hg1---O2B
Hg1---F11
Hg1---F12
Hg1---F13B
Ng2−F11
Ng2−F11A
Ng1−F12
Ng1−F13
Bond Angles (°)
O1−Hg1−O2
Hg1−O1−Te1
Hg1−O2−Te2
O1−Te1−F1
O1−Te1−F2
O1−Te1−F3
O1−Te1−F4
O1−Te1−F5
O2−Te2−F6
O2−Te2−F7
O2−Te2−F8
O2−Te2−F9
O2−Te2−F10
F5−Te1−F2
F5−Te1−F3
F5−Te1−F4
F5−Te1−F1
F2−Te1−F4
F4−Te1−F1
F1−Te1−F3
F3−Te1−F2
F3−Te1−F4
F1−Te1−F2
F8−Te2−F9
F8−Te2−F6
173.0(2)
132.8(3)
127.2(3)
93.2(2)
94.7(2)
95.3(2)
91.1(2)
176.7(2)
95.0(2)
91.2(2)
178.0(2)
93.4(2)
92.7(3)
86.6(2)
87.7(2)
86.0(2)
85.5(2)
89.5(2)
89.8(2)
89.4(2)
90.4(2)
173.6(2)
172.1(2)
87.4(2)
86.9(2)
173.3(1)
129.6(2)
126.5(1)
93.0(1)
94.4(1)
95.0(1)
90.6(1)
176.8(1)
94.8(1)
90.8(1)
178.5(1)
92.7(1)
92.8(2)
86.4(1)
88.1(1)
86.3(1)
86.2(2)
90.5(1)
89.4(1)
89.1(1)
90.2(1)
174.3(1)
172.5(1)
87.4(2)
86.7(1)
F8−Te2−F7
F8−Te2−F10
F9−Te2−F7
F7−Te2−F10
F10−Te2−F6
F6−Te2−F9
F9−Te2−F10
F6−Te2−F7
F11A−Ng2−F11
F12−Ng1−F13
Ng2−F11---Hg1
Ng1−F12---Hg1
O1−Hg1---F13B
O1−Hg1---F11
O1−Hg1---F12
O2−Hg1---F13B
O2−Hg1---F11
O2−Hg1---F12
O1−Hg1---O2B
O2B---Hg1---F13B
O2B---Hg1---F11
O2B---Hg1---F12
F12---Hg1---F11
F13B---Hg1---F11
F13B---Hg1---F12
87.0(2)
86.6(3)
90.2(2)
91.8(2)
90.0(2)
87.5(2)
173.6(3)
173.5(2)
180.0
87.7(1)
87.1(2)
90.1(1)
91.3(2)
90.1(1)
88.0(1)
174.2(2)
174.2(1)
180.0
179.4(2)
158.3(3)
119.3(2)
79.7(2)
99.2(2)
77.2(2)
101.1(2)
84.4(2)
96.4(3)
103.3(2)
139.9(2)
77.2(2)
69.8(2)
144.7(2)
142.5(2)
77.2(2)
178.9(1)
150.6(2)
119.5(1)
78.5(1)
99.4(1)
75.9(1)
103.0(1)
83.0(2)
98.2(2)
103.4(1)
139.6(1)
74.2(1)
69.4(1)
140.8(1)
146.0(1)
72.2(1)
Dihedral Angle (°)
50.3(3)
Te1−O1−Hg1−O2−Te2
45.2(5)
a
Atom labeling schemes correspond to those used in Figures S4 and 2 for Ng = Xe and Kr, respectively.
Spectral assignments for Hg(OTeF5)2 were made by
(OTeF5)2 is also compared with that of Xe(OTeF5)2.32 The
vibrational assignments for Hg(OTeF5)2·1.5NgF2 were made
by comparison with the experimental frequencies of Hg-
(OTeF5)2 (Table 4) and NgF2 (Tables S5 and S6), and the
calculated frequencies and assignments of NgF2 (Tables S5 and
S6) and the [Hg(OTeF5)2]3·2NgF2 model compounds. These
models enabled the assignments of the intramolecular coupled
modes that occur among Hg(OTeF5)2 units and/or NgF2 units.
The following frequency assignments and related discussions
exclusively refer to the modes associated with the central
Hg(OTeF5)2 units of gas-phase [Hg(OTeF5)2]3 and
[Hg(OTeF5)2]3·2NgF2 and with the NgF2 ligands. The
calculated frequencies of νas(NgF2), νs(NgF2), and δ(KrF2) of
free XeF2 and KrF2 were overestimated, whereas that of
δ(XeF2) was close to the experimental value (Tables S5 and
S6). This pattern aided in the assignment of the corresponding
calculated frequencies of the [Hg(OTeF5)2]3·2NgF2 complexes.
comparison with the calculated frequencies and Raman
intensities (Tables 4, S1, and S2) obtained for the energy-
minimized, gas-phase geometries of Hg(OTeF5)2 (C2)
monomer and the presently unknown trimer, [Hg(OTeF5)2]3
(C1) (Figure 6). The central Hg(OTeF5)2 unit of the trimeric
model provides a good approximation of the repeat unit in the
chain structure (Figure 1). A similar approach has been
successfully used to assign the Raman spectra of the polymeric
34
open chain structures OsO3F2 and MoSF435. In another
related structure, XeOF4·XeF2, the model compounds, 2XeOF4·
XeF2 and XeOF4·4XeF2, have provided good approximations
for the local environments of XeF2 and XeOF4 and their
vibrational assignments.36
The −OTeF5 ligands of the central Hg(OTeF5)2 unit of
[Hg(OTeF5)2]3 have a gauche-conformation and two
cis-Hg---O secondary contacts to the Hg(II) atoms from the
terminal Hg(OTeF5)2 units. The Raman spectrum of Hg-
3892
dx.doi.org/10.1021/ja412193z | J. Am. Chem. Soc. 2014, 136, 3888−3903

Journal of the American Chemical Society
Article
Figure 3. The structural unit in the X-ray crystal structure of
Hg(OTeF5)2·1.5KrF2 (see Figure 2) with thermal ellipsoids drawn at
the 50% probability level. Secondary bonding interactions are
indicated by dashed lines drawn from the F and O atoms of adjacent
KrF2 and −OTeF5 groups to Hg(II) and show the pseudo-octahedral
coordination around Hg(II).
Figure 1. The X-ray crystal structure of Hg(OTeF5)2 showing (a) its
chain structure viewed along the b-axis and running parallel to the c-
axis and (b) the pseudo-octahedral coordination around Hg(II)
resulting from the secondary bonding interactions (indicated by
dashed lines) between Hg(II) and the F and O atoms of −OTeF5
groups in adjacent Hg(OTeF5)2 units; thermal ellipsoids are shown at
the 50% probability level.
Figure 4. The Raman spectrum of Hg(OTeF5)2·1.5XeF2 recorded in a
quartz tube at −155 °C using 1064-nm excitation. Symbols denote
unreacted XeF2 at 496 cm−1 (†) and an instrumental artifact (‡).
Figure 2. The chain structure in the X-ray crystal structure of
Hg(OTeF5)2·1.5KrF2; thermal ellipsoids are shown at the 50%
probability level. Secondary bonding interactions from the F and O
atoms of adjacent KrF2 and −OTeF5 groups to Hg(II) are indicated by
dashed lines.
Figure 5. The Raman spectrum of Hg(OTeF5)2·1.5KrF2 recorded in a
quartz tube at −155 °C using 1064-nm excitation. Symbols denote
unreacted KrF2 at 464 cm−1 (†) and an instrumental artifact (‡).
Hg(OTeF5)2. The vibrational assignments of Hg(OTeF5)2
were initially based on a gas-phase monomeric model (C2
symmetry, Table S1, and Figure 6a). The 39 vibrations of
monomeric Hg(OTeF5)2 span the irreducible representations
Γvib = 20A + 19B, which are Raman and infrared active.
The calculated vibrational frequencies and intensities at the
B3LYP and PBE1PBE levels of theory using the def2-TZPP
and aug-cc-PTVZ basis sets reproduce the experimental trends,
but the Hg(OTeF5)2 frequencies obtained at the B3LYP level
(Table S1) were significantly underestimated when compared
with the frequencies calculated at the PBE1PBE level.
Consequently, calculations for [Hg(OTeF5)2]3 and
[Hg(OTeF5)2]3·2NgF2 were carried out at the PBE1PBE
3893
dx.doi.org/10.1021/ja412193z | J. Am. Chem. Soc. 2014, 136, 3888−3903

Journal of the American Chemical Society
Article
Table 4. Experimental Raman Frequencies and Intensities
for Hg(OTeF5)2 and Calculated Vibrational Frequencies and
Intensities for [Hg(OTeF5)2]3
Figure 6. The gas-phase, energy-minimized geometries of (a)
monomeric Hg(OTeF5)2 (C2) and (b) trimeric [Hg(OTeF5)2]3 (C1)
calculated at the PBE1PBE/def2-TZVPP level of theory. The long
contacts (dashed lines) between the Hg(II) atom of the central
Hg(OTeF5)2 unit and two oxygen atoms of two adjacent terminal
Hg(OTeF5)2 units are shown in (b).
not accounted for by the monomeric model suggested that this
model was too limited. Moreover, the anti-conformation of the
−OTeF5 groups in the gas-phase monomer differs from that of
the solid-state gauche-conformation.
The use of the trimeric model, [Hg(OTeF5)2]3, (Figure 6b)
addresses these differences by reproducing the gauche-
conformation and revealing that the additional bands that
were not accounted for in the monomeric model arise from
intermolecular couplings among its Hg(OTeF5)2 units (see
above and Tables 4 and S2).
The ν(Hg−O) and ν(Te−O) stretches couple, giving rise to
eight vibrational modes. Four of these modes are derived from
“symmetric” [ν(Hg−O) + ν(Te−O)]-type stretching modes.
The [ν(Hg16−O22) + ν(Te17−O22)] mode is in-phase coupled
to [ν(Hg16−O37) + ν(Te32−O37)] and is also coupled in-phase
(481 cm−1) and out-of-phase (472 cm−1) with the analogous
modes of the two outer Hg(OTeF5)2 units of the trimer. The
totally in-phase coupled mode at 481 cm−1 is the second most
intense band in the spectrum. The calculated frequencies (506
and 516 cm−1, respectively) are in good agreement with the
observed values. As in the experimental spectrum, the
calculated Raman intensity of the 506 cm−1 band is also very
intense. The [ν(Hg16−O22) + ν(Te17−O22)] mode is also out-
of-phase coupled to the [ν(Hg16−O37) + ν(Te32−O37)] mode
(511 cm−1). The [ν(Hg16−O22) + ν(Te17−O22)] mode is
further coupled in-phase and out-of-phase (530 cm−1) and out-
of-phase and in-phase (528 cm−1) to the analogous modes of
the two terminal Hg(OTeF5)2 units. The agreement between
the observed (511 cm−1) and calculated values (530 and 528
cm−1) is again very good. Four modes are predicted which
involve the “asymmetric” [ν(Hg16−O22) − ν(Te17−O22)] and
[ν(Hg16−O37) − ν(Te32−O37)] stretching modes and are
expected to be out-of-phase (793 cm−1) and in-phase (787,
806, and 824 cm−1) coupled. The latter three modes arise from
a
b
Frequencies are given in cm−1. Values in parentheses denote relative
Raman intensities. The Raman spectrum was recorded in an FEP
c
sample tube at −150 °C using 1064-nm excitation. The abbreviations
d
denote shoulder (sh) and not observed (n.o.). Values in parentheses
denote calculated Raman intensities (Å4 amu−1), whereas values in
square brackets denote calculated infrared intensities (km mol−1).
e
Assignments are for the energy-minimized geometry calculated at the
PBE1PBE/def2-TZVPP level; only simplified mode assignments
(separated by the symbol “/”) that involve the central Hg(OTeF5)2
unit are listed. See Table S2 for a complete listing of frequencies and
detailed descriptions of the assignments. The abbreviations denote
out-of-plane (o.o.p.) and in-plane (i.p.) where the planes contain the
(Te−O−Hg−O-Te) groups, umbrella (umb), equatorial (e), axial (a),
stretch (ν), bend (δ), twist (ρt), wag (ρw), and rock (ρr). The atom
labeling scheme is given in Figures 6b and S5.
level using only the def2-TZPP basis set due to the large sizes
of these molecules. Although most of the observed frequencies
and intensities could be accounted for using the gas-phase
monomeric model, the observation of additional bands that are
3894
dx.doi.org/10.1021/ja412193z | J. Am. Chem. Soc. 2014, 136, 3888−3903

Journal of the American Chemical Society
Article
Table 5. Experimental Raman Frequencies and Intensities for Hg(OTeF5)2·1.5XeF2 and Calculated Vibrational Frequencies and
Intensities for [Hg(OTeF5)2]3·2XeF2
a
b
Frequencies are given in cm−1. Values in parentheses denote relative Raman intensities. The Raman spectrum was recorded in a quartz sample
c
tube at −155 °C using 1064-nm excitation. A band at 496(27) cm−1 (not listed, see Figure 4) is assigned to excess XeF2. The abbreviations denote
d
shoulder (sh) and not observed (n.o.). Values in parentheses denote calculated Raman intensities (Å4 amu−1), whereas values in square brackets
e
denote calculated infrared intensities (km mol−1). Assignments are for the energy-minimized geometry calculated at the PBE1PBE/def2-TZVPP
level. Only simplified mode assignments (separated by the symbol “/”) that involve the central Hg(OTeF5)2 unit are listed; the modes involving
XeF2 are fully described. See Table S3 for a complete listing of frequencies and more detailed descriptions of the assignments. The abbreviations
denote out-of-plane (o.o.p.) and in-plane (i.p.) where the planes may contain the (Te−O−Hg−O−Te) groups or the two XeF2 molecules, umbrella
(umb), equatorial (e), axial (a), stretch (ν), bend (δ), twist (ρt), wag (ρw), and rock (ρr). The subscript “small” denotes that the bracketed
vibrational mode makes a small contribution relative to the other coupled vibrations. The atom labeling scheme is given in Figure S6a.
3895
dx.doi.org/10.1021/ja412193z | J. Am. Chem. Soc. 2014, 136, 3888−3903

Journal of the American Chemical Society
Article
Table 6. Experimental Raman Frequencies and Intensities for Hg(OTeF5)2·1.5KrF2 and Calculated Vibrational Frequencies and
Intensities for [Hg(OTeF5)2]3·2KrF2
a
b
Frequencies are given in cm−1. Values in parentheses denote relative Raman intensities. The Raman spectrum was recorded in a quartz sample
c
tube at −155 °C using 1064-nm excitation. The band at 464(21) cm−1 (not listed, see Figure 5) is assigned to excess KrF2. The abbreviations
d
denote shoulder (sh) and not observed (n.o.). Values in parentheses denote calculated Raman intensities (Å4 amu−1), whereas values in square
e
brackets denote calculated infrared intensities (km mol−1). Assignments are for the energy-minimized geometry calculated at the PBE1PBE/def2-
TZVPP level. Only simplified mode assignments (separated by the symbol “/”) that involve the central Hg(OTeF5)2 unit are listed; the modes
involving KrF2 are fully described. See Table S4 for a complete listing of frequencies and more detailed descriptions of the assignments. The
abbreviations denote out-of-plane (o.o.p.) and in-plane (i.p.) where the planes may contain the (Te−O−Hg−O−Te) groups or the two KrF2
molecules, umbrella (umb), equatorial (e), axial (a), stretch (ν), bend (δ), twist (ρt), wag (ρw) and rock (ρr) modes. The subscript “small” denotes
that the bracketed vibrational mode makes a small contribution relative to the other coupled vibrations. The atom labeling scheme is given in Figures
7 and S6b.
3896
dx.doi.org/10.1021/ja412193z | J. Am. Chem. Soc. 2014, 136, 3888−3903

Journal of the American Chemical Society
Article
additional coupling with the analogous modes of the terminal
Hg(OTeF5)2 units. The calculated band at 824 cm−1 is
expected to be relatively intense and is observed as a
medium-intensity band at 825 cm−1. The weak Raman band
at 801 cm−1 has been assigned to the weak modes calculated at
793 and 787 cm−1. The calculated mode at 806 cm−1 was not
observed and is predicted to be weak in the Raman spectrum.
The above frequencies and relative intensities are reminiscent
of, but are at higher frequency than those observed for the
coupled ν(Xe−O) and ν(Te−O) stretches in Xe(OTeF5)2
(440/445, 796/788, and 730 cm−1).32 The band analogous to
the low-intensity band of Hg(OTeF5)2 at 511 cm−1 is predicted
at 547 cm−1 in Xe(OTeF5)2, i.e., [ν(Xe−O) + ν(Te−O)] −
[ν(Xe−O′) − ν(Te′−O′)] + δ(OXeO′), but was not
observed.32 The O22−Hg16−O37 bending mode of Hg(OTeF5)2
is observed as a weak band at 331 cm−1 (calcd, 332 cm−1), and
its frequency is very similar to that of δ(O−Xe−O) (328 cm−1)
in Xe(OTeF5)2.32
significantly affected by complex formation (see X-ray
Crystallography); moreover, the Te−Fe bond lengths are
equal within 3σ among all crystal structures. In practice,
these bands show very little, if any, change among the
complexes and Hg(OTeF5)2 (Xe: 702−753, 623−640 cm−1;
Kr: 704−723, 624−645 cm−1; Hg(OTeF5)2: 735, 624−699
cm−1). The calculated frequencies also remain within the same
ranges (Xe: 695−721, 622−644 cm−1; Kr: 693−723, 623−646
cm−1; [Hg(OTeF5)2]3: 724−726, 640−719 cm−1). The
frequencies of the δ(TeF4e)umb umbrella modes also remain
essentially unchanged. The bands below 332 (Xe) and 329
cm−1 (Kr) are assigned to coupled deformation and torsional
modes and are well reproduced by the calculations (Tables 5
and 6). Modes that are exclusively of the “symmetric” [ν(Hg−
O) + ν(Te−O)]-type occur in a frequency range similar to that
of Hg(OTeF5)2. As observed for the “asymmetric” [ν(Hg−O)
− ν(Te−O)]-type modes, the “symmetric” modes occur at
higher frequencies for the XeF2 complex (exptl, 474/478 and
508 cm−1) than those of the KrF2 complex (exptl, 458 and 484
cm−1). In the case of the XeF2 complex, a third band occurs at
445 cm−1 which has an additional coupling with ν(Xe16−F23) in
the theoretical model. In the KrF2 complex, only two
[ν(Hg−O) + ν(Te−O)]-type modes are predicted where
[ν(Hg12−O28) + ν(Te14−O28)] is only coupled in-phase (exptl,
458 cm−1; calcd, 505 cm−1) or out-of-phase (exptl, 484 cm−1;
calcd, 527 cm−1) with [ν(Hg12−O31) + ν(Te15−O31)]. In the
XeF2 complex, only one mode is predicted in which [ν(Hg12−
O28) + ν(Te14−O28)] is out-of-phase coupled with [ν(Hg12−
O31) + ν(Te15−O31)] (exptl, 508 cm−1; calcd, 524 cm−1). Two
modes are predicted where [ν(Hg12−O28) + ν(Te14−O28)] is
in-phase coupled with [ν(Hg12−O31) + ν(Te15−O31)] (exptl,
478/474 and 445 cm−1; calcd, 502 and 499 cm−1) because, in
both cases, there is additional coupling with ν(Xe16−F23) and/
or analogous modes of the terminal Hg(OTeF5)2 units.
The bands between 624 and 735 cm−1 are assigned to
stretching modes of the TeF5 groups and are in good
agreement with the calculated values (640−726 cm−1) and
with those observed in Xe(OTeF5)2 (635−710 cm−1).32 In
both cases, the coupled axial vs(Te−Fa) modes occur as strong
bands in the Raman spectrum (Hg(OTeF5)2, 709 cm−1;
Xe(OTeF5)2, 690 cm−1). The coupled umbrella mode,
[δ(Te17F4e)umb + δ(Te32F4e′)umb], is observed as a weak band
at 349 cm−1 (calcd, 341 cm−1) that also in-phase couples with
the analogous modes of the two outer Hg(OTeF5)2 units. This
mode was not observed for Xe(OTeF5)2 (calcd, 360 cm−1).32
Hg(OTeF5)2·1.5NgF2. Spectral assignments were aided by
obtaining the energy-minimized geometries of the unknown
model complexes, [Hg(OTeF5)2]3·2NgF2, and their vibrational
frequencies at the PBE1PBE level using the def2-TZPP basis
set (Tables 5, 6, S3, and S4). These approximations reliably
reproduced the experimental trends. Overall, couplings among
the vibrational modes of the XeF2 complex are more extensive
than among the vibrational modes of the KrF2 complex.
As observed for [Hg(OTeF5)3], the two highest frequency
bands (Xe, 853 and 825 cm−1; Kr, 844 and 821 cm−1) involve
“asymmetric” [ν(Hg12−O28) − ν(Te14−O28)] and [ν(Hg12−
O31) − ν(Te15−O31)] stretches which are in-phase and out-of-
phase coupled. These bands are shifted to higher frequencies
relative to the analogous Raman bands of solid Hg(OTeF5)2
(825 and 801 cm−1), a trend that is also observed for the
calculated frequencies (Xe, 828 and 817/815 cm−1; Kr, 826 and
812 cm−1; [Hg(OTeF5)2]3, 824/806 and 787/793 cm−1).
These shifts are noteworthy because the experimental Hg−O
and Te−O bond lengths are equal within 3σ in the crystal
structures of Hg(OTeF5)2 and Hg(OTeF5)2·1.5NgF2. As
observed in earlier studies, the present studies also illustrate
that Raman spectroscopy can be a more sensitive probe than X-
ray crystallography for the detection of small bond strength/
bond length differences. The high-frequency shifts of the XeF2
adduct are anticipated because the Hg---F(NgF) contacts are
shorter in the XeF2 complex (2.606(5) and 2.623(4) Å) than in
the KrF2 complex (2.664(3) and 2.675(3) Å) (see X-ray
crystallography). The experimental stretching frequencies of
the axial fluorine atoms are also affected, shifting to lower
frequency (Xe, 685 cm−1; Kr, 683 and 688 cm−1) relative to
Hg(OTeF5)2 (709 cm−1). The corresponding calculated
frequencies also follow the same trend (Xe: 699, 705 cm−1;
Kr: 699, 700, 706 cm−1; [Hg(OTeF5)2]3: 707, 709 cm−1).
Bands involving ν(Te−Fe) stretches are not predicted to be
The calculated vibrational displacements show that the
stretching modes of the bridging KrF2 and XeF2 units are
extensively coupled and reveal differences between their inter-
and intraligand couplings in their respective complexes. The
four coupled modes of the KrF2 complex are each comprised of
inter- and intraligand coupling components, whereas there is
less intraligand coupling in the XeF2 complex. Instead, some
coupling with the terminal Hg(OTeF5)2 units occurs. Coupling
among the stretching modes of the NgF2 units in the
15
4
[BrOF2][AsF6]·2XeF2 and [BrOF2][AsF6]·2KrF2 complexes
has also been observed.
The modes derived from the symmetric (Raman-active and
infrared-inactive) stretches of the free NgF2 molecules, i.e.,
[ν(Ng13−F17) + ν(Ng13−F20)] and [ν(Ng16−F23) + ν(Ng16−
F40)], are expected to occur at lower frequency than those
derived from the asymmetric stretching mode of free NgF2, i.e.,
[ν(Ng13−F17) − ν(Ng13−F20)] and [ν(Ng16−F23) − ν(Ng16−
F40)]. The “symmetric” modes in the KrF2 complex are
expected to in-phase couple, [ν(Kr13−F17) + ν(Kr13−F20)] +
[ν(Kr16−F23) + ν(Kr16−F40)], and out-of-phase couple,
[ν(Kr13−F17) + ν(Kr13−F20)] − [ν(Kr16−F23) + ν(Kr16−
F40)]. The latter modes were calculated at 512 and 513 cm−1
and are assigned to a single band at 468 cm−1, the most intense
band in the Raman spectrum. In the XeF2 complex, the totally
in-phase analogue, [ν(Xe13−F17) + ν(Xe13−F20)] + [ν(Xe16−
F23) + ν(Xe16−F40)], is observed at 501 cm−1 and is also a
strong band. In addition, there are two bands corresponding to
the out-of-phase stretching mode, [ν(Xe13−F17)] − [ν(Xe16−
F23)] (489 cm−1) and [ν(Xe13−F20)] − [ν(Xe16−F40)] (508
3897
dx.doi.org/10.1021/ja412193z | J. Am. Chem. Soc. 2014, 136, 3888−3903

Journal of the American Chemical Society
Article
cm−1). These frequencies are comparable to the Raman-active
νs(Ng−F2) mode of free NgF2 (Xe, 494 cm−1;37 Kr, 464
cm−1)38 and are in accordance with the observed Ng−F bond
lengths (Xe, 1.981(4)−2.012(4) Å and Kr, 1.883(3)−1.897(3)
Å in the complexes; Xe, 1.999(4) Å 3 and Kr, 1.894(5) Å 2 in
free NgF2). The corresponding calculated NgF2 frequencies
and Ng−F bond lengths of the complexes and free NgF2 follow
the same trend (Xe: 516, 521, 510 cm−1, 1.991−2.003 Å and
Kr: 513, 512 cm−1, 1.868−1.878 Å; free XeF2: 530 cm−1, 1.980
Å and free KrF2: 519 cm−1, 1.865 Å). In both complexes, the
“symmetric” stretching mode appears at slightly higher
frequency than the symmetric stretching mode of free NgF2.
The “symmetric” stretch of the XeF2 complex appears at lower
frequency than other “symmetric” stretching modes of the
bridging XeF2 molecules in the Cd2+ coordination complexes,
Cd(XeF2)4(AsF6)2 (521 cm−1)10 and Cd(XeF2)5(PF6)2 (521
cm−1).11
254, and 266 cm−1 for NgF2 in the complexes; cf., Xe, 215
cm−1; Kr, 250 cm−1 in free NgF2).
COMPUTATIONAL RESULTS
■
Unless otherwise noted, the following discussion refers to the
central units of the gas-phase model compounds,
[Hg(OTeF5)2]3 and [Hg(OTeF5)2]3·2NgF2.
Calculated Geometries. Hg(OTeF5)2 and [Hg(OTeF5)2]3.
The gas-phase geometry of monomeric Hg(OTeF5)2 (C2)
(Figure 6a) was optimized at the B3LYP and PBE1PBE levels
of theory using the def2-TZPP and aug-cc-PTVZ basis sets,
resulting in stationary points with all frequencies real (Table
S1). The calculated bond lengths and angles are provided in
Table S7. Although both levels of theory well reproduced the
observed trends (see Raman Spectroscopy), better agreement
was obtained at the PBE1PBE level. All attempts to optimize
the monomeric Hg(OTeF5)2 unit, regardless of the starting
geometry, resulted in an anti-conformation (C2 symmetry)
resembling that observed (C2h in the crystal structure) and
calculated (C2) for Xe(OTeF5)2.32 The authors were unable to
reproduce the previously reported30 calculated C2h geometry
for Hg(OTeF5)2. However, the anti-conformation of the gas-
phase Hg(OTeF5)2 monomer contrasts with the gauche-
conformation observed in the crystal structure. The optimiza-
tion of the presently unknown trimer, [Hg(OTeF5)2]3 (Figures
6b and S5), using the PBE1PBE/def2-TZPP method also
resulted in a stationary point with all frequencies real (C1
symmetry). This model reproduced the observed gauche-
conformation of the central Hg(OTeF5)2 molecule showing
that crystal packing and accompanying Hg---O contacts (2.737
Å) with neighboring terminal Hg(OTeF5)2 molecules are likely
major factors in stabilizing the solid-state gauche-conformation.
This is supported by the fact that the two terminal
Hg(OTeF5)2 units also retain the gauche-conformation with
similar Hg---O contacts (2.775 Å). Because of its size, the
unknown trimer was only calculated with the smaller basis set,
def2-TZPP.
In the trimer, the largest discrepancies occur for the
Hg---O(1A,1C) secondary bonding interactions (calcd, 2.737
Å; exptl, 2.641(7) Å) and the O(1)−Hg---O(1A,1C) bond
angles (calcd, 73.0 and 103.8°; exptl, 78.7 and 93.8(3)°). These
differences are likely attributable to the model itself, where the
Hg---F secondary contacts are absent for the central
Hg(OTeF5)2 unit. Ideally, a total of five Hg(OTeF5)2
molecules would be required to reproduce all secondary
contacts observed in the crystal structure. The calculated
O−Hg−O bond angle (176.0°) is more open than the
observed angle in Hg(OTeF5)2 (170.5(4)°) (see X-ray
Crystallography), but is in close agreement with the
O−Hg−O angle calculated for Hg(OTeF5)2 monomer (calcd,
176.8°).
The bands at 558 and 553 (Kr) cm−1 and 518 (Xe) cm−1 are
assigned to NgF2 stretching modes that are derived from the
asymmetric (infrared-active and Raman-inactive) stretches of
the free NgF2 molecules. In order to understand why the
formally Raman inactive bands in free NgF2 are observed in the
Raman spectra of both NgF2 complexes, the positioning of the
two crystallographically inequivalent NgF2 molecules in the
crystal structures must be taken into account. One NgF2
molecule is positioned on an inversion center, so that the
vibrational activities of the corresponding stretching modes will
be the same as those of free NgF2, i.e., the symmetric stretch
will be Raman active and the asymmetric stretch will be infrared
active. The second NgF2 molecule is on a general position,
resulting in two crystallographically inequivalent Ng−F bonds.
As a result, both coupled modes derived from the asymmetric
stretch of free NgF2 will be Raman and infrared active. The
[Hg(OTeF5)2]3·2NgF2 models (C1 symmetry) display this
behavior.
The “asymmetric” NgF2 stretches are coupled in-phase,
[ν(Ng13−F17) − ν(Ng13−F20)] + [ν(Ng16−F23) − ν(Ng16−
F40)] (Xe, 518 cm−1; Kr, 558 cm−1), and out-of-phase,
[ν(Ng13−F17) − ν(Ng13−F20)] − [ν(Ng16−F23) − ν(Ng16−
F40)] (Xe, 518 cm−1; Kr, 553 cm−1). These “asymmetric”
modes occur at lower frequencies than their infrared-active
asymmetric counterparts in free XeF2 (555 cm−1)37 and KrF2
(580 cm−1).39 This trend is reproduced by the calculations (Xe,
521/524 and 528/534 cm−1; Kr, 574 and 560 cm−1 for NgF2 in
the complexes; Xe, 568 cm−1; Kr, 607 cm−1 in free NgF2). It is
noteworthy that, although the XeF2 bridging molecules in
Cd(XeF2)4(AsF6)2,10 Cd(XeF2)5(PF6)2,11 Ca(XeF2)n(AsF6)2
(n = 4, 2.5),40 Ca2(XeF2)9(AsF6)4,41 Ca(XeF2)5(PF6)2,11
8
Sr3(XeF2)10(PF6)6,8 and Pb3(XeF2)11(PF6)6 also have two
crystallographically inequivalent Xe−F bonds, their “asymmet-
ric” XeF2 stretches were not identified.
The Hg−O bond lengths of the calculated gas-phase
[Hg(OTeF5)2]3 molecule (2.008 Å) are in better agreement
with the experimental bond length (2.016(6) Å) than the
calculated Hg−O bond lengths of Hg(OTeF5)2 (1.976 Å). The
Te−F and Te−O bond lengths are overall slightly shorter for
the calculated structure of [Hg(OTeF5)2]3 (1.832−1.855 and
1.865 Å, respectively) when compared with those calculated for
monomeric Hg(OTeF5)2 (1.835−1.866 and 1.856 Å, respec-
tively), but are also in better agreement with those observed for
solid Hg(OTeF5)2 (Table 2).
As previously observed for [BrOF2][AsF6]·2NgF2,4,15 the
double degeneracy of the NgF2 bending modes of free NgF2
(ν2, IIu) is removed when NgF2 is asymmetrically fluorine
bridged to mercury, resulting in splitting into out-of-plane,
δ(NgF2)o.o.p., and in-plane, δ(NgF2)i.p., modes with respect to
the plane containing both NgF2 ligands. The bending modes
are observed at 223 and 241 cm−1 (Xe) and at 237 and 260
cm−1 (Kr) and are slightly shifted to higher frequencies relative
to those of free XeF2 (213 cm−1)37 and free KrF2 (236 cm−1).39
The calculated and experimental frequencies are also in good
agreement (Xe: 218, 219, 221, and 236 cm−1; Kr: 252, 252,
[Hg(OTeF5)2]3·2NgF2 (Ng = Xe, Kr). The calculated gas-phase
geometries of the model complexes, [Hg(OTeF5)2]3·2NgF2
3898
dx.doi.org/10.1021/ja412193z | J. Am. Chem. Soc. 2014, 136, 3888−3903

Journal of the American Chemical Society
Article
a
(Ng = Xe, Kr) (Figures 7 and S6, Table 7), were optimized at
the PBE1PBE level of theory using the def2-TZPP basis set,
Table 7. Calculated Geometrical Parameters for
Hg(OTeF5)2·1.5NgF2 (Ng = Xe, Kr)
Xe
Kr
Xe
Kr
Bond Lengths (Å)
Hg12−O31
Hg12−O28
O31−Te15
O28−Te14
Te15−F27
Te15−F29
Te15−F26
Te15−F25
Te15−F30
Te14−F19
1.991
2.002
1.847
1.847
1.852
1.843
1.870
1.836
1.838
1.855
1.988
2.000
1.849
1.849
1.851
1.843
1.869
1.835
1.838
1.855
Te14−F18
Te14−F24
Te14−F22
Te14−F21
Hg12---F23
Hg12---F17
Ng16−F23
Ng16−F40
Ng13−F17
Ng13−F20
1.854
1.834
1.854
1.839
2.723
2.724
2.003
1.991
1.997
1.993
1.855
1.833
1.851
1.840
2.754
2.762
1.878
1.869
1.877
1.868
Bond Angles (°)
O31−Hg12−O28
Hg12−O31−Te15
Hg12−O28−Te14
O31−Te15−F27
O31−Te15−F29
O31−Te15−F26
O31−Te15−F25
O31−Te15−F30
O28−Te14−F19
O28−Te14−F18
O28−Te14−F24
O28−Te14−F22
O28−Te14−F21
F30−Te15−F29
F30−Te15−F26
F30−Te15−F25
F30−Te15−F27
F29−Te15−F25
F25−Te15−F27
F27−Te15−F26
F26−Te15−F29
173.6
122.4
123.6
93.1
92.9
92.5
91.0
179.8
90.3
93.0
178.3
92.9
92.1
87.3
87.6
88.9
86.8
91.0
90.7
88.3
89.6
173.7
122.4
122.4
93.0
92.8
92.5
90.9
179.8
90.4
92.9
178.4
92.7
92.1
87.3
87.7
88.9
86.9
91.0
90.7
88.3
89.6
F26−Te15−F25
176.4
173.8
87.3
176.6
173.9
87.4
F27−Te15−F29
F24−Te14−F22
F24−Te14−F19
F24−Te14−F18
F24−Te14−F21
F22−Te14−F18
F18−Te14−F21
F21−Te14−F19
F19−Te14−F22
F22−Te14−F21
F19−Te14−F18
F40−Ng16−F23
F17−Ng13−F20
Ng16−F23---Hg12
Ng13−F17---Hg12
O31−Hg12---F23
O31−Hg12---F17
O28−Hg12---F23
O28−Hg12---F17
F17---Hg12---F23
Figure 7. The gas-phase, energy-minimized geometry of [Hg-
(OTeF5)2]3·1.5KrF2 calculated at the PBE1PBE/def2-TZVPP level
of theory. The dashed lines show the contacts between the Hg(II)
atom of the central Hg(OTeF5)2 unit and two fluorine atoms of two
adjacent KrF2 molecules.
88.0
88.0
88.6
88.7
87.7
87.7
89.2
89.3
90.8
90.6
89.8
89.7
resulting in stationary points with all frequencies real (Tables
S3 and S4). These systems were too large and demanding of
CPU time to be optimized using the larger aug-cc-PTVZ basis
set. The calculated structures of [Hg(OTeF5)2]3·2NgF2 mimic
the local environments of both Hg(OTeF5)2 and NgF2 in the
crystal structures of Hg(OTeF5)2·1.5NgF2. The central
Hg(OTeF5)2 unit of the starting models takes into account
the two shorter Hg---F(NgF) contacts that are trans to one
another. In the optimized geometries, the central Hg(OTeF5)2
units retained the gauche-conformation observed in the crystal
structures, as was calculated for the central Hg(OTeF5)2 unit of
[Hg(OTeF5)2]3. The calculated models retained the two
Hg---F(NgF) contacts to the central Hg atom (Xe, 2.724 and
2.723 Å; Kr, 2.762 and 2.754 Å) but optimized so that the
Hg---F(NgF) secondary bonds are cis to one another with
contact distances that better reproduce the longer Hg---F(NgF)
contacts observed in the crystal structures (Xe, 2.701(5) Å; Kr,
2.741(3) Å). In both the Kr and Xe models, one of the outer
Hg(OTeF5)2 units optimized to a syn-conformation (Te−O−
Hg−O−Te dihedral angles of 126.5° for Xe and 122.8° for Kr),
when only a single, long secondary Hg---F(NgF) bond is
present (Xe, 2.746 Å; Kr, 2.816 Å). In contrast, the other
terminal Hg(OTeF5)2 unit optimized to a gauche-conformation
presumably because the secondary Hg---F(NgF) contacts (Xe,
2.694 Å; Kr, 2.739 Å) are somewhat shorter and more covalent.
These differences are reminiscent of those observed for the
[Hg(OTeF5)2]3 model (vide supra) and may also result from
the model’s inability to take into account two additional Hg---F
and Hg---O secondary contacts that are also present in the
crystal structures (see X-ray Crystallography).
90.0
90.1
175.0
176.6
179.2
179.1
138.4
118.3
97.3
175.2
176.7
179.3
179.5
136.9
117.7
96.5
99.7
98.7
87.1
87.3
76.8
77.1
75.8
77.6
Dihedral Angle (°)
Te15−O31−Hg12−O28−
31.0
34.0
Te14
a
The atom labeling scheme corresponds to that used in Figures 7 and
S6 for Xe and Kr, respectively. All bond lengths and angles refer to the
central Hg(OTeF5)2 unit and to the coordinated NgF2 molecules of
the unknown [Hg(OTeF5)2]3·2NgF2 molecules calculated at the
PBE1PBE/def2-TZVPP level of theory.
The Hg−O (Xe: 1.991, 2.002 Å; Kr: 1.998, 2.000 Å) and
Te−O (Xe: 1.847 Å; Kr: 1.849 Å) bond lengths are slightly
under- and overestimated, respectively, compared to the Hg−O
(Xe: 2.015(5), 2.037(5) Å; Kr: 2.017(3), 2.029(3) Å) and
Te−O (Xe: 1.815(5), 1.811(6) Å; Kr: 1.819(3), 1.836(3) Å)
bond lengths in the crystal structures of Hg(OTeF5)2·1.5NgF2.
The calculated O−Hg−O bond angles (Xe, 173.6°; Kr, 173.7°)
accurately reproduce the observed O−Hg−O bond angles (Xe,
173.0(2)°; Kr, 173.3(1)°) of the complexes, whereas the
Hg−O−Te bond angles (Xe: 122.4, 123.6°; Kr: 122.4°) are
smaller than observed (Xe: 132.8(3), 127.2(3)°; Kr: 129.6(2),
126.5(1)°). This may reflect the elongation of the secondary
bonding interactions and the absence of the two additional
Hg---F and Hg---O contacts that are present in the crystal
structure. This limitation in the model may contribute to the
calculated Te−O−Hg−O−Te dihedral angles (Xe, 31.0°; Kr,
34.0°) which give a central Hg(OTeF5)2 unit that more closely
approximates a syn-conformation than those observed in the
The complexed NgF2 molecules are essentially linear with
F−Ng−F angles (Xe, 179.2 and 179.1°; Kr, 179.5 and 179.3°)
and Ng−F bond lengths (Xe, 1.991−2.003 Å; Kr, 1.868−1.878
Å) that well reproduce those observed in the crystal structures
(Xe: 180 and 179.4(2)°, 1.981(4)−2.012(4) Å; Kr: 180 and
178.9(1)°, 1.883(3)−1.897(3) Å). The calculated Ng−F bond
lengths are slightly underestimated for free NgF2 (calcd: Xe,
3
1.980 Å and Kr, 1.865 Å; exptl: Xe, 1.999(4) Å and Kr,
2
1.894(5) Å ).
3899
dx.doi.org/10.1021/ja412193z | J. Am. Chem. Soc. 2014, 136, 3888−3903

Journal of the American Chemical Society
Article
solid state (Xe, 45.2(5)°; Kr, 50.3(3)°). All Te−F bond lengths
(Xe, 1.834−1.870 Å; Kr, 1.833−1.869 Å) are in the same
ranges as those observed in the crystal structures (Xe,
1.814(5)−1.845(4) Å; Kr, 1.824(3)−1.848(3) Å) with the
exception of the Te15−F26 bond lengths (Xe, 1.870 Å; Kr, 1.869
Å), which are slightly longer.
intensities of [Hg(OTeF5)2]3·2NgF2 aided in the assignment of
the experimental Raman spectra of Hg(OTeF5)2·1.5NgF2. The
Hg(OTeF5)2·1.5NgF2 complexes are currently the only
examples of coordination complexes in which KrF2 and XeF2
are coordinated to mercury in a neutral compound and provide
the only example of a bridging KrF2 ligand.
Natural Bond Orbital (NBO) Analyses. The NBO
analyses reported in this section (Table S8) refer to the central
Hg(OTeF5)2 units of [Hg(OTeF5)2]3 and [Hg(OTeF5)2]3·
2NgF2 (Figures 6, 7, and S6) unless noted otherwise.
EXPERIMENTAL SECTION
■
Caution. Anhydrous HF must be handled using appropriate
protective gear with immediate access to proper treatment
procedures42−44 in the event of contact with liquid HF, HF vapor,
or HF-containing solutions. Krypton difluoride and its coordination
complex, Hg(OTeF5)2·1.5KrF2, are highly energetic materials and
strong oxidizers that are only stable under the rigorously anhydrous
handling conditions employed in the experimental procedures outlined
below. Both compounds are likely to detonate upon contact with
organic materials. Thus, adequate protective apparel and working
behind adequate shielding are crucial for the safe manipulation of these
materials. It is therefore recommended that the syntheses of the
aforementioned compounds be carried out on small scales (<200 mg).
Apparatus and Materials. General. Manipulations involving air-
sensitive materials were carried out under anhydrous conditions on
glass and metal high-vacuum lines and inside an inert atmosphere
drybox as previously described.45 Preparative work was carried out in
When compared with gas-phase Hg(OTeF5)2, the charge on
Hg is little affected by contacts with adjacent units within
[Hg(OTeF5)2]3. The situation is essentially the same when
NgF2 coordinates to [Hg(OTeF5)2]3, providing the model
complexes, [Hg(OTeF5)2]3·2NgF2. The highest negative
charges reside on the O atoms of [Hg(OTeF5)2]3 (−1.208),
with their charges becoming more positive upon NgF2
coordination (Kr and Xe, −1.137). This is reflected by small
increases in the Hg−O bond orders from 0.320 to 0.419/0.439
for KrF2 and to 0.413/0.434 for XeF2 and in the oxygen atom
valencies from 0.948 to 1.025/1.015 for KrF2 and to 1.023/
1.011 for XeF2. Little change in the Te−O bond orders and Te
valencies occurs upon NgF2 coordination.
In each NgF2 complex, there are small negative charge
transfers from both NgF2 ligands (Kr, 0.033/0.045; Xe, 0.040/
0.052) to the central Hg(OTeF5)2 units (Kr, −0.038; Xe,
−0.046) and combined charge transfers to the two terminal
Hg(OTeF5)2 units (Kr, −0.040 ; Xe, −0.046). The small Hg---
F(Ng) bridge bond orders (∼0.06) and small degrees of NgF2
polarization by the central Hg(OTeF5)2 unit of [Hg(OTeF5)2]3
are consistent with weak covalent interactions between the
Hg(II) acceptor sites and the σ-donor fluorine ligands of NgF2.
1
reaction vessels constructed from /4-in. o.d. (1/16-in. wall thickness)
lengths of FEP (tetrafluoroethylene-hexafluoropropylene block co-
polymer) tubing. The tubing was heat-sealed at one end, heat flared at
the other end, and connected through a 45° SAE flare nut to the
conical end of a Kel-F (chlorotrifluoroethylene polymer) valve to form
a compression seal. Reaction vessels and sample tubes were rigorously
dried under dynamic vacuum prior to passivation for at least 8 h with 1
1
atm of F2 gas. Vacuum line connections were made using /4-in. 316
stainless steel Swagelok Ultratorr unions fitted with Viton O-rings.
2,47,48
Xenon difluoride46 and KrF2
were prepared and purified as
previously described. High-purity HgF2 and HOTeF5 were prepared
using synthetic procedures outlined in the Supporting Information.
Anhydrous HF (Harshaw Chemicals Co.) was purified as previously
described.49 Methylene chloride (Caledon, reagent grade) was dried
over previously vacuum-dried (250 °C) Davison type 3 Å molecular
sieves (Fisher Scientific) for 3 days followed by vacuum distillation
into a dry glass bulb equipped with a 4-mm J. Young (glass/Teflon)
valve. Methylene chloride-d2 (D, 99.5%; BDH Chemicals) was dried
over CaH2 powder (99.5%, BDH Chemicals). Sulfuryl chloride
fluoride (Allied Chemical, Baker Adamson Division) was purified as
previously described.50 High-purity Ar (99.998%, Air Liquide) or N2
(obtained from liquid N2 boil-off and dried by passage through a
column of dry 3 Å molecular sieves) gases were used for backfilling
reaction and sample vessels.
CONCLUSION
■
The −OTeF5 analogue of HgF2, Hg(OTeF5)2, was structurally
characterized by single-crystal X-ray diffraction and the low-
temperature Raman spectrum was assigned using quantum-
chemical calculations. The crystal structure of Hg(OTeF5)2
showed that the Hg(OTeF5)2 molecules are not isolated but
participate in a chain structure that results from long Hg---O
and Hg---F secondary bonding interactions with Hg(II) centers
of adjacent Hg(OTeF5)2 molecules. The Raman spectrum was
assigned using the calculated vibrational frequencies and
intensities of the hypothetical trimer, [Hg(OTeF5)2]3, which
reproduced the solid-state gauche-conformation and the
shortest Hg---O contacts observed in the crystal structure.
The gauche-conformation was rationalized based on the
occurrence of secondary Hg---F and Hg---O bonding
interactions with the Hg(II) center. In contrast, the optimized
gas-phase structure of monomeric Hg(OTeF5)2 provided a
geometry having its −OTeF5 groups in an anti-conformation
similar to that observed in Xe(OTeF5)2. The coordination
complexes, Hg(OTeF5)2·1.5NgF2 (Ng = Xe, Kr), were also
synthesized and structurally characterized by single-crystal X-
ray diffraction and low-temperature Raman spectroscopy. In
both chain structures, the NgF2 molecules form bridges
between mercury centers by coordination to the metal through
their fluorine ligands. The contact distances between Hg(II)
and the F atoms of XeF2 are shorter than those of the KrF2
analogue, consistent with the greater ionic character of the
Xe−F bonds in XeF2. NBO analyses are consistent with weak
covalent interactions between the Hg(OTeF5)2 acceptor and
the NgF2 σ-donor ligands. The calculated frequencies and
Synthesis of Hg(OTeF5)2 and Crystal Growth. A passivated FEP
reaction vessel equipped with a Kel-F valve was loaded with HgF2
(0.4938 g, 2.069 mmol) inside a drybox. The reaction vessel was then
transferred to a metal vacuum line where HOTeF5 was distilled into it.
The contents of the reaction vessel were allowed to react at 50 °C for
several hours with periodic agitation. Residual HOTeF5, observed by
Raman spectroscopy, and HF formed in the reaction (eq 1) were
removed by pumping under dynamic vacuum for 3 h at room
temperature, resulting in a friable, white solid in essentially quantitative
yield (99.1%). The Raman spectrum of the product was recorded at
−150 °C (Figure S1).
Crystals of Hg(OTeF5)2 were grown by slow evaporation of a
1
CH2Cl2 solution. The solution was prepared in a /4-in. o.d. FEP T-
shaped reaction vessel by dissolving Hg(OTeF5)2 (0.0314 g, 0.0463
mmol) in ∼0.3 mL of CH2Cl2 at room temperature under anhydrous
conditions. The void above the solution was backfilled with 0.5 atm of
dry N2 at −78 °C. A temperature gradient was established by cooling
the empty side arm of the vessel to −78 °C in a dry ice/acetone bath
while maintaining the solution at 0 °C. This temperature gradient
resulted in slow evaporation of the CH2Cl2 solvent and growth of
colorless crystals over the course of 11 days. The side arm containing
3900
dx.doi.org/10.1021/ja412193z | J. Am. Chem. Soc. 2014, 136, 3888−3903

Journal of the American Chemical Society
Article
the evaporated supernatant was then cooled to −196 °C and heat-
sealed off under dynamic vacuum. The crystalline material was further
dried at −78 °C under dynamic vacuum. A Hg(OTeF5)2 crystal having
the dimensions of 0.22 × 0.10 × 0.04 mm3 was selected for a low-
temperature X-ray structure determination.
In the sections that follow, square brackets denote quantities/
conditions used for crystal growth and unbracketed quantities/
conditions denote Raman sample preparations.
1.5XeF2, whereas a Bruker Triumph curved crystal monochromator
with a Mo Kα source (λ = 0.71073 Å) was used for Hg(OTeF5)2·
1.5KrF2. The diffraction data sets consisted of a full ϕ-rotation (1010
frames collected at 0.36° intervals) at fixed χ = 54.74°, followed by a
series of short ω scans (250 frames) at various ϕ-settings to fill the
gaps for Hg(OTeF5)2 and Hg(OTeF5)2·1.5XeF2. The diffraction data
collections consisted of a full ϕ-rotation (3474 frames collected at
0.50° intervals) at fixed χ = 54.74°, followed by a series of short ω
scans (1935 frames) at various ϕ-settings to fill the gaps for
Hg(OTeF5)2·1.5KrF2. The crystal-to-detector distances were 4.9540
cm, and the data collection was carried out in a 512 × 512 pixel mode
using 2 × 2 pixel binning. All diffraction data were processed by use of
the APEX2 GUI software,52 which applied Lorentz and polarization
corrections to three-dimensionally integrated diffraction spots. The
program SADABS53 was used for scaling the diffraction data, the
application of a decay correction, and an empirical absorption
correction based on redundant reflections.
Synthesis of Hg(OTeF5)2·1.5XeF2 and Crystal Growth. In a typical
synthesis, 0.1816 g (0.2679 mmol) [0.0879 g (0.1298 mmol)] of
1
Hg(OTeF5)2 was weighed, inside a drybox, into a /4-in. o.d. quartz
[T-shaped FEP] reaction vessel equipped with a 4-mm J. Young [Kel-
F] valve. Xenon difluoride, 0.0750 g (0.443 mmol) [0.0267 g (0.1578
mmol)], was added to the reactor at −140 °C inside the drybox. The
reactor was removed from the drybox at −196 °C and attached to a
glass vacuum line while maintaining the reagents at −78 °C. Sulfuryl
chloride fluoride (∼0.2 mL) was condensed onto the reagents, and the
temperature was increased to 0 °C for 5 min and continuously agitated
to dissolve the reactants. The solvent was removed from the Raman
sample under dynamic vacuum at −78 °C leaving behind a white solid.
The Raman spectrum of the product was recorded at −155 °C. The
solution used for crystallization was pale yellow and was cooled to −78
°C. Over the course of 5 days, colorless crystals formed. The
supernatant was decanted into the side arm of the T-shaped FEP
vessel at −78 °C. Once the majority of the supernatant had been
transferred, the contents of the side arm were cooled to −196 °C, and
the supernatant was isolated and removed by heat sealing off this
portion of the reaction vessel under dynamic vacuum at −196 °C. This
was followed by removal of the residual solvent from the crystalline
sample under dynamic vacuum at −78 °C. A Hg(OTeF5)2·1.5XeF2
crystal having the dimensions 0.25 × 0.04 × 0.04 mm3 was selected for
a low-temperature X-ray structure determination.
Solution and Refinement of the Structures. The XPREP54
program was used to confirm the unit cell dimensions and the crystal
system and space group. The structures were solved in their respective
space groups by use of direct methods, and the solutions yielded the
positions of all the heavy atoms as well as some of the lighter atoms.
Successive difference Fourier syntheses revealed the positions of the
remaining light atoms. The final refinement was obtained by
introducing anisotropic parameters for all the atoms, an extinction
parameter, and the recommended weighting factor. The maximum
electron densities in the final difference Fourier maps were located
around the heavy atoms. The PLATON program55 could not suggest
additional or alternative symmetries.
Raman Spectroscopy. Raman spectra were recorded on a Bruker
RFS 100 FT-Raman spectrometer at −150 or −155 °C using 1064-nm
excitation at laser powers ranging from 300−400 mW and 1 cm−1
resolution as previously described.51 A total of 1200 scans were
acquired for each spectrum. Samples were contained in 1/4-in. o.d. FEP
Synthesis of Hg(OTeF5)2·1.5KrF2 and Crystal Growth. In a typical
synthesis, KrF2 was sublimed under static vacuum from a FEP storage
container at room temperature into a preweighed, fluorine-passivated
4-mm o.d FEP vessel cooled to −196 °C. Krypton difluoride, 0.0161 g
(0.132 mmol) [0.0538 g (0.4417 mmol)] was condensed under static
1
vessels or in /4-in. o.d. quartz tubes.
NMR Spectroscopy. Instrumentation. The 19F spectrum of
Hg(OTeF5)2 was recorded at 25 °C on a Bruker AVANCE DRX-500
spectrometer equipped with an 11.744-T cryomagnet. The 19F NMR
spectrum was acquired using a 5-mm combination 1H/19F probe
operating at 470.568 MHz. The 19F spectrum was recorded in a 32K
memory, with a spectral width setting of 24 kHz, yielding a data-point
resolution of 0.73 Hz/data point and an acquisition time of 0.68 s. The
pulse width, corresponding to bulk magnetization tip angles of ∼90°,
was 7.7 μs. A relaxation delay of 2 s was used, and 1200 transients were
accumulated. A line broadening of 0.3 Hz was used in the exponential
multiplication of the free induction decay prior to Fourier trans-
formation. The 19F spectrum was referenced externally at room
temperature (25 °C) to a sample of neat CFCl3. The chemical shift
convention used is that a positive (negative) sign indicates a chemical
shift to high (low) frequency of the reference compound.
1
vacuum through a FEP connection into a fluorine-passivated /4-in.
o.d. quartz [T-shaped FEP] reaction vessel cooled to −196 °C that
had been previously loaded with Hg(OTeF5)2, 0.0404 g (0.0596
mmol) [0.1152 g (0.1610 mmol)] inside a drybox. Sulfuryl chloride
fluoride was condensed onto the reagents (∼0.3 mL) [∼0.5 mL], and
upon warming the reaction vessel to −20 °C for 2 min, the solid
mixture partially dissolved to give a pale yellow solution plus a
suspension of white solid. The reaction mixture used for preparation of
the Raman sample was allowed to react at −78 °C for 3 h followed by
removal of SO2ClF under dynamic vacuum at −78 °C. The Raman
spectrum of the product was recorded at −155 °C. The FEP reaction
vessel and solution used for crystal growth were maintained at −78 °C
for 2 weeks. Crystals were isolated as described in the section above. A
Hg(OTeF5)2·1.5KrF2 crystal having the dimensions 0.18 × 0.08 × 0.05
mm3 was selected for a low-temperature X-ray structure determination.
X-ray Crystallography. Crystal Mounting Procedure. All
crystalline products were stored at −78 °C until a suitable crystal
could be selected and mounted on the X-ray diffractometer using a
previously described low-temperature crystal mounting technique.51
Dried crystalline samples that had been maintained at −78 °C were
dumped, without warming, into an aluminum trough cooled to −110
5 °C by means of a cold stream of dry N2 gas, allowing selection of
individual crystals under a stereomicroscope. Single crystals were
mounted at the tip of a glass fiber at −110 5 °C using a Fomblin oil
as the adhesive and transferred to a goniometer head using cryotongs
(Hampton Research) which had been cooled to −196 °C in liquid
N2.51
NMR Sample Preparation. A Hg(OTeF5)2 sample was prepared in
a precision thin-wall Pyrex glass NMR sample tube (Wilmad) as
previously described.25 The NMR sample tube was fused to a /4-in.
1
Pyrex glass tube which was connected to a grease-free 6-mm J. Young
1
glass stopcock outfitted with a Teflon barrel using a /4-in. stainless
steel Swagelok Ultratorr union fitted with Viton elastomer O-rings and
was rigorously dried under dynamic vacuum. The CD2Cl2 solvent was
distilled into the vessel at −78 °C before the sample was transferred
into a drybox where Hg(OTeF5)2 was added to the frozen solvent at
−140 °C. The union and valve assembly were replaced, and the reactor
was attached to a vacuum manifold where the NMR sample tube was
cooled to −196 °C, heat-sealed under dynamic vacuum, and stored at
−78 °C until the 19F NMR spectrum could be obtained. The sample
was dissolved at 25 °C just prior to data acquisition.
Collection and Reduction of X-ray Data. Single crystals were
centered on a SMART APEX II diffractometer, equipped with an
APEX II 4K charge-coupled device (CCD) and a triple-axis
goniometer, controlled by the APEX2 Graphical User Interface
(GUI) software.52 Graphite-monochromated Mo Kα radiation (λ =
0.71073 Å) was used in the case of Hg(OTeF5)2 and Hg(OTeF5)2·
Computational Details. The optimized gas-phase geometry and
vibrational frequencies of Hg(OTeF5)2 were calculated at the B3LYP
and PBE1PBE levels of theory using two different basis sets. The aug-
cc-pVTZ basis sets where used for H, O, and F, whereas aug-cc-pVTZ-
(PP) basis sets having pseudopotentials were used for Hg and Te and
the def2-TZVPP basis sets for H, O, F, Hg, and Te. The optimized
3901
dx.doi.org/10.1021/ja412193z | J. Am. Chem. Soc. 2014, 136, 3888−3903

Journal of the American Chemical Society
Article
gas-phase geometries and vibrational frequencies of [Hg(OTeF5)2]3
and [Hg(OTeF5)2]3·2NgF2 (Ng = Xe, Kr) were only calculated using
the PBE1PBE/def2-TZVPP (H, O, F, Kr, Te, Xe, and Hg) method
due to the large sizes of these molecules. The noble-gas difluorides,
NgF2 (Ng = Xe, Kr), were also calculated for comparison using the
PBE1PBE/def2-TZVPP method. All basis sets were obtained online
from the EMSL Basis Set Exchange (https://bse.pnl.gov/bse/
portal).56−61 The NBO analyses62−65 were performed for the
PBE1PBE optimized local minima. Quantum-chemical calculations
were carried out using the program Gaussian 0966 for geometry
optimizations, vibrational frequencies, and their intensities. All
geometries were fully optimized using analytical gradient methods.
The program GaussView67 was used to visualize the vibrational
displacements that form the basis for the vibrational mode descriptions
given in Tables 4−6 and S1−S5.
REFERENCES
■
̌
(1) Brock, D. S.; Schrobilgen, G. J.; Zemva, B. Noble-Gas Chemistry.
In Comprehensive Inorganic Chemistry; Reedijk, J., Poeppelmeier, K.,
Eds.; Elsevier: Oxford, 2013; Vol 1, Chapter 1.25, pp 755−822.
(2) Lehmann, J. F.; Dixon, D. A.; Schrobilgen, G. J. Inorg. Chem.
2001, 40, 3002−3017.
(3) Elliott, H. St. A.; Lehmann, J.; Mercier, H. P. A.; Jenkins, H. D.;
Schrobilgen, G. J. Inorg. Chem. 2010, 49, 8504−8523.
(4) Brock, D. S.; Casalis de Pury, J. J.; Mercier, H. P. A.; Schrobilgen,
G. J.; Silvi, B. Inorg. Chem. 2010, 49, 6673−6689.
̌
(5) Tavcar, G.; Tramsek, M.; Bunic, T.; Benkic, P.; Zemva, B. J.
̌
̌
̌
̌
Fluorine Chem. 2004, 125, 1579−1584.
̌
(6) Tramse
(7) Tavca
(8) Bunic,
Inorg. Chem. 2007, 46, 5276−5282.
(9) Bunic, T.; Tramsek, M.; Goreshnik, E.; Zemva, B. Solid State Sci.
2008, 10, 1511−1516.
(10) Tavcar, G.; Benkic,
1457.
(11) Bunic,
2006, 45, 1038−1042.
̌
k, M.; Zemva, B. J. Fluorine Chem. 2006, 127, 1275−1284.
̌
̌
r, G.; Zemva, B. Inorg. Chem. 2005, 44, 1525−1529.
̌
̌
T.; Tramse
̌
k, M.; Goreshnik, E.; Tavca
̌
r, G.; Zemva, B.
ASSOCIATED CONTENT
■
̌
̌
̌
S
* Supporting Information
̌
̌
P.; Zemva, B. Inorg. Chem. 2004, 43, 1452−
̌
Complementary discussion for the preparation of high-purity
HgF2; Raman spectrum of pure Hg(OTeF5)2 (Figure S1);
Raman spectrum of sublimed Hg(OTeF5)2 (Figure S2); crystal
packing for Hg(OTeF5)2 along the c- and a-axes (Figure S3);
crystal structure of Hg(OTeF5)2·1.5XeF2 (Figure S4); exper-
imental Raman frequencies and intensities, calculated vibra-
tional frequencies, and infrared and Raman intensities and
detailed assignments of the Raman spectra for Hg(OTeF5)2
(exptl and calcd) (Table S1), Hg(OTeF5)2 (exptl) and
[Hg(OTeF5)2]3 (calcd) (Table S2), Hg(OTeF5)2·1.5XeF2
(exptl) and [Hg(OTeF5)2]3·2XeF2 (calcd) (Table S3),
Hg(OTeF5)2·1.5KrF2 (exptl) and [Hg(OTeF5)2]3·2KrF2
(calcd) (Table S4); experimental and calculated Raman
frequencies for XeF2 (Table S5) and KrF2 (Table S6);
experimental (Hg(OTeF5)2) and calculated (monomer
Hg(OTeF5)2) bond lengths and bond angles (Table S7);
fully labeled calculated geometry of [Hg(OTeF5)2]3 (Figure
S5); fully labeled calc ulated geometries of
[Hg(OTeF5)2]3·2NgF2 (Figure S6); NBO valencies, bond
orders, and NPA charges for Hg(OTeF5)2, [Hg(OTeF5)2]3,
and [Hg(OTeF5)2]3·2NgF2 (Table S8); complementary
discussion for the preparation of HOTeF5; X-ray crystallo-
graphic files in CIF format for the structure determinations of
Hg(OTeF5)2 and Hg(OTeF5)2·1.5NgF2 (Ng = Xe, Kr). This
information is available free of charge via the Internet at http://
pubs.acs.org/.
̌
̌ ̌
T.; Tavcar, G.; Tramsek, M.; Zemva, B. Inorg. Chem.
̌
(12) Tavca
1368−1373.
(13) Tavca
̌
r, G.; Goreshnik, E.; Mazej, Z. J. Fluorine Chem. 2006, 127,
̌
r, G.; Goreshnik, E.; Mazej, Z. Abstracts of Papers. Group
12 Coordination Chemistry with Fluoride Ligands. Presented at the
2011 National Meeting of the Slovenian Chemical Society; Slovenian
Chemistry Days (Slovenski Kemijski Dnevi), Portoroz, Slovenia,
September 14−16, 2011; Paper, Anorganska Kemija in Structurna
Kemija ter Gradiva, p 69.
̌
̌
(14) Tavcar, G.; Goreshnik, E. Abstracts of Papers. Mercury Ion as a
Center for Coordination of the XeF2 Ligand. Presented at the 17th
European Symposium on Fluorine Chemistry, Paris, France, July 21−
25, 2013; Poster P2.46.
(15) Brock, D. S.; Casalis de Pury, J. J.; Mercier, H. P. A.;
Schrobilgen, G. J.; Silvi, B. J. Am. Chem. Soc. 2010, 132, 3533−3542.
(16) Holloway, J. H.; Schrobilgen, G. J.; Taylor, P. J. Chem. Soc.,
Chem. Commun. 1975, 2, 40−41.
(17) Holloway, J. H.; Schrobilgen, G. J. Inorg. Chem. 1980, 19, 2632−
2640.
(18) Holloway, J. H.; Schrobilgen, G. J. Inorg. Chem. 1981, 20, 3363−
3368.
(19) Christe, K. O.; Wilson, W. W.; Bougon, R. A. Inorg. Chem. 1986,
13, 2163−2169.
(20) Tucker, P. A.; Taylor, P. A.; Holloway, J. H.; Russell, D. R. Acta
Crystallogr. 1975, B31, 906−908.
(21) Sladky, F.; Kropshofer, H. Inorg. Nucl. Chem. Lett. 1972, 8, 195−
197.
AUTHOR INFORMATION
■
(22) Birchall, T.; Myers, R. D.; de Waard, H.; Schrobilgen, G. J. Inorg.
Chem. 1982, 21, 1068−1073.
Corresponding Author
schrobil@mcmaster.ca
(23) Riedel, S.; Kaupp, M. Coord. Chem. Rev. 2009, 253, 606−624.
(24) Mercier, H. P. A.; Moran, M. D.; Schrobilgen, G. J.; Steinberg,
C.; Suontamo, R. J. J. Am. Chem. Soc. 1994, 116, 2921−2937.
(25) Mercier, H. P. A.; Moran, M. D.; Schrobilgen, G. J.; Steinberg,
C.; Suontamo, R. J. J. Am. Chem. Soc. 2004, 126, 5533−5548.
(26) Mercier, H. P. A.; Moran, M. D.; Sanders, J. C. P.; Schrobilgen,
G. J.; Suontamo, R. J. Inorg. Chem. 2005, 44, 49−60.
(27) Ebert, F.; Woitinek, H. Z. Anorg. Allg. Chem. 1933, 210, 269−
272.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the Natural Sciences and Engineering Research
Council of Canada for support in the form of a Discovery Grant
(G.J.S.), and for the award of Undergraduate Summer Research
Awards (2011-13) (J.R.D.), the Ontario Graduate Scholarship
in Science and Technology for support in the form of a
scholarship (J.R.D.), and the computational resources provided
by SHARCNet (Shared Hierarchical Academic Research
Computing Network; www.sharcnet.ca). We also thank J. P.
Paxon for preliminary work on some of the systems described
in this work.
(28) Seppelt, K.; Nothe, D. Inorg. Chem. 1973, 11, 2727−2730.
(29) Sladky, F.; Kropshofer, H.; Leitzke, O.; Peringer, P. J. Inorg.
Nucl. Chem., H. H. Hyman Memorial Volume 1976, 69−71.
(30) Riedel, S.; Straka, M.; Kaupp, M. Chem.Eur. J. 2005, 11,
2743−2755.
(31) Bondi, A. J. Phys. Chem. 1964, 68, 441−451.
(32) Fir, B. A.; Mercier, H. P. A.; Sanders, J. C. P.; Dixon, D. A.;
Schrobilgen, G. J. J. Fluorine Chem. 2001, 110, 89−107.
3902
dx.doi.org/10.1021/ja412193z | J. Am. Chem. Soc. 2014, 136, 3888−3903

Journal of the American Chemical Society
Article
(33) Cordero, B.; Gom
Echeverría, J.; Cremades, E.; Barragan
2008, 2832−2838.
́
ez, V.; Platero-Prats, A. E.; Reves
́
, M.;
(66) 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.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,
N.; Millam, N. J.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
́
, F.; Alvarez, S. Dalton Trans.
(34) Hughes, M. J.; Mercier, H. P. A.; Schrobilgen, G. J. Inorg. Chem.
2009, 48, 4478−4490.
(35) Nieboer, J.; Mack, J. P.; Mercier, H. P. A.; Gerken, M. Inorg.
Chem. 2010, 49, 6153−6159.
(36) Hughes, M. J.; Brock, D. S.; Mercier, H. P. A.; Schrobilgen, G. J.
J. Fluorine Chem. 2011, 132, 660−668.
(37) Agron, P. A.; Begun, G. M.; Levy, H. A.; Mason, A. A.; Jones, C.
G.; Smith, D. F. Science 1963, 139, 842−844.
(38) Al-Mukhtar, M.; Holloway, J. H.; Hope, E. G.; Schrobilgen, G. J.
J. Chem. Soc., Dalton Trans. 1991, 2831−2834.
̈
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09,
revision D.01; Gaussian, Inc.: Wallingford, CT, 2009.
(39) Turner, J. J.; Pimentel, G. C. Science 1963, 140, 974−975.
̌
(40) Benkic,
1425−1434.
(41) Tramse
43, 3456−3458.
̌
P.; Tramse
̌
k, M.; Zemva, B. Solid State Sci. 2002, 4,
̌
(67) GaussView, release 3.0; Gaussian Inc.: Wallingford, CT, 2003.
̌
k, M.; Benkic,
̌
P.; Zemva, B. Angew. Chem., Int. Ed. 2004,
(42) Bertolini, J. C. J. Emerg. Med. 1992, 10, 163−168.
(43) Peters, D.; Mietchen, R. J. Fluorine Chem. 1996, 79, 161−165.
(44) Segal, E. B. Chem. Health Saf. 2000, 18−23.
(45) Casteel, W. J., Jr.; Dixon, D. A.; Mercier, H. P. A.; Schrobilgen,
G. J. Inorg. Chem. 1996, 35, 4310−4322.
(46) Mercier, H. P. A.; Sanders, J. C. P.; Schrobilgen, G. J.; Tsai, S. S.
Inorg. Chem. 1993, 32, 386−393.
(47) Kinkead, S. A.; FitzPatrick, J. R.; Foropoulos, J., Jr.; Kissane, R.
J.; Purson, J. D. In Fluorine Chemistry Toward the 21st Century;
Thrasher, J. S., Strauss, S. H., Eds.; ACS Symposium Series 555;
American Chemical Society: Washington, DC, 1994; Chapter 3, pp
40−55.
(48) Lehmann, J. F.; Mercier, H. P. A.; Schrobilgen, G. J. Coord.
Chem. Rev. 2002, 233−234, 1−39.
(49) Emara, A. A. A.; Schrobilgen, G. J. Inorg. Chem. 1992, 31, 1323−
1332.
(50) Schack, C.; Wilson, R. D. Inorg. Chem. 1970, 9, 311−314.
(51) Gerken, M.; Dixon, D. A.; Schrobilgen, G. J. Inorg. Chem. 2000,
39, 4244−4255.
(52) APEX2, release v2011.6-1; Bruker AXS Inc.: Madison, WI, 1995.
(53) Sheldrick, G. M. SADABS (Siemens Area Detector Absorption
Corrections), version 2.03; Siemens Analytical X-ray Instruments, Inc.:
Madison, WI, 1999.
(54) Sheldrick, G. M. SHELXTL-Plus, release 5.1; Siemens Analytical
X-ray Instruments, Inc.: Madison, WI, 1998.
(55) Spek, A. L. J. Appl. Crystallogr. 2003, 36, 7−13.
(56) Basis sets and pseudopotentials were obtained from the
Extensible Computational Chemistry Environment Basis set Database,
version 2/25/04, as developed and distributed by the Molecular
Science Computing Facility, Environmental and Molecular Science
Laboratory, which is part of the Pacific Northwest Laboratory, P.O.
Box 999, Richland, WA 99352.
(57) Dunning, T. H., Jr. J. Chem. Phys. 1989, 90, 1007−1023.
(58) Kendall, R. A.; Dunning, T. H., Jr.; Harrison, R. J. J. Chem. Phys.
1992, 96, 6796−6806.
(59) Peterson, K. A.; Figgen, D.; Goll, E.; Stoll, H.; Dolg, M. J. Chem.
Phys. 2003, 119, 11113−11123.
(60) Peterson, K. A.; Puzzarini, C. Theor. Chem. Acc. 2005, 114, 283−
296.
(61) Weigend, F.; Ahlrichs, R. Phys. Chem. Chem. Phys. 2005, 7,
3297−3305.
(62) Reed, A. E.; Weinstock, R. B.; Weinhold, F. J. Chem. Phys. 1985,
83, 735−746.
(63) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1998, 88,
899−926.
(64) Glendening, E. D.; Reed, A. E.; Carpenter, J. E.; Weinhold, F.
NBO Version 3.1; Gaussian Inc.: Pittsburgh, PA, 1990.
(65) Glendening, E. D.; Badenhoop, J. K.; Reed, A. E.; Carpenter, J.
E.; Bohmann, C. M.; Morales, C. M.; Weinhold, F. NBO Version 5.0;
Theoretical Chemistry Institute, University of Wisconsin: Madison,
WI, 2001.
3903
dx.doi.org/10.1021/ja412193z | J. Am. Chem. Soc. 2014, 136, 3888−3903