Convenient synthesis of fluoride-alkoxides of Nb(v) and Ta(v): A spectroscopic, crystallographic and computational study

Article abstract of DOI:10.1039/c2dt31453c

The synthesis and the spectroscopic characterization of fluoride-alkoxides of niobium and tantalum in the highest oxidation state are reported. Suspensions of MF₅ (M = Nb, Ta) in a chlorinated solvent reacted with up to three equivalents of ROSiMe₃ (R = Me, Et, Ph) to afford polynuclear derivatives and variable amounts of FSiMe₃. Thus MF₄(OR) (R = Et, Ph) and MF₃(OR)₂ were obtained by selective 1:1 and 1:2 reactions almost exclusively as single isomeric products; otherwise mixtures of MF₄(OMe) species were afforded from the equimolar reactions of MF₅ with MeOSiMe₃. The 1:3 reaction of TaF₅ with MeOSiMe₃ led to different forms of TaF ₂(OMe)₃. The synthesis of TaF(OPh)₄ was forced by high temperature conditions or the use of a large excess of PhOSiMe ₃. DFT studies were carried out in order to predict, in the distinct cases, the most stable structures of the metal products. The molecular structures of [NbF₂(OPh)₂(μ-F)]₃ and [TaF(OPh)₃(μ-OPh)]₂ were ascertained by X-ray diffraction.

Full text of DOI:10.1039/c2dt31453c

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2012, 41, 12898
www.rsc.org/dalton
PAPER
Convenient synthesis of fluoride-alkoxides of Nb(V) and Ta(V):
a spectroscopic, crystallographic and computational study†
Marco Bortoluzzi,a Nicola Guazzelli,b Fabio Marchetti,*b Guido Pampalonib and Stefano Zacchinic
Received 11th April 2012, Accepted 28th August 2012
DOI: 10.1039/c2dt31453c
The synthesis and the spectroscopic characterization of fluoride-alkoxides of niobium and tantalum in the
highest oxidation state are reported. Suspensions of MF5 (M = Nb, Ta) in a chlorinated solvent reacted
with up to three equivalents of ROSiMe3 (R = Me, Et, Ph) to afford polynuclear derivatives and variable
amounts of FSiMe3. Thus MF4(OR) (R = Et, Ph) and MF3(OR)2 were obtained by selective 1 : 1 and 1 : 2
reactions almost exclusively as single isomeric products; otherwise mixtures of MF4(OMe) species
were afforded from the equimolar reactions of MF5 with MeOSiMe3. The 1 : 3 reaction of TaF5 with
MeOSiMe3 led to different forms of TaF2(OMe)3. The synthesis of TaF(OPh)4 was forced by high
temperature conditions or the use of a large excess of PhOSiMe3. DFT studies were carried out in order to
predict, in the distinct cases, the most stable structures of the metal products. The molecular structures of
[NbF2(OPh)2(μ-F)]3 and [TaF(OPh)3(μ-OPh)]2 were ascertained by X-ray diffraction.
complexes WF6−x(OEt)x (x = 1, 2) have been obtained from
Introduction
WF6 by treatment with ethanol.13,15
In the last few years there has been increasing development in
the chemistry of niobium and tantalum derivatives, encouraged
by the easy availability and the substantial non-toxicity of the
metals.1
In this context the alkoxides/aryloxides of Nb(V) and Ta(V)
have found interesting applications in catalysis1a,2 and as precur-
sors of the corresponding oxides,3 which in turn have remarkable
electrical and optical properties.4
Both the pentaalkoxide species M(OR)5 (M = Nb, Ta; R =
alkyl or aryl)2a,3a,5 and the mixed halide-alkoxides MXn(OR)5−n
(X = Cl or Br, n = 1–3)6 are well-known compounds, which
typically adopt a dinuclear structure with two ligands bridging
between the two metal centres. On the other hand, very little is
known on the corresponding fluoride-alkoxides. Mixed fluoride-
alkoxides of a variety of metals have been usually prepared by
Nevertheless, in the case of Nb(V) and Ta(V) this last pro-
cedure is not viable for the preparation of MF5−x(OR)x com-
plexes (M = Nb, Ta), due to the special strength of the M–F
bond in MF5, preventing the fluoride displacement.1a,16 Other-
wise the synthesis of some Nb(V) and Ta(V) fluoride alkoxides
(and related coordination adducts17) was claimed in the past
by metathesis reaction between MF5 and M(OR)5 (M = Nb, Ta;
R = Et, Ph),18 see point (c) above. Nevertheless the characteri-
zation of the products was limited and a clear structural determi-
nation has been still missing.
The successful competition for fluorine of Si towards Nb was
casually demonstrated, when high thermal treatment of NbF5 in
silicon-containing tubes gave SiF4 and niobium oxy-fluorides.19
Hence amido-fluorides and chloro-fluorides were prepared by
reactions of MF5 (M = Nb, Ta) with trimethylsilyl species.20
On considering these facts, we decided to investigate the reactiv-
ity of niobium and tantalum pentafluorides, MF5 (M = Nb, 1a;
M = Ta, 1b),21 with trimethylsilyl ethers, ROSiMe3.
three strategies: (a) alkoxide-fluoride exchange by reaction
10
of metal alkoxides with HF,7,8 KHF2,9 PF3
or MeCOF;11
(b) fluoride-alkoxide exchange by reaction of metal fluorides
with alcohols; (c) metathesis reaction between metal alkoxides
and metal fluorides.12–14 As an example (point b), mononuclear
The reactions, with different molar ratios, proceeded straight-
forwardly with both alkyl (R = Me, Et) and aryl (R = Ph)
groups, and led to the high-yield formation of fluoride-alkoxide
(aryloxide) derivatives. The determination of the structures of
the products obtained relied on combined spectroscopic and
DFT studies. The two first examples of X-ray characterization of
niobium and tantalum fluoride-phenolates will be presented.
aUniversity of Venezia, Dipartimento di Scienze Molecolari e
Nanosistemi, Dorsoduro 2137, I-30123 Venezia, Italy
bUniversity of Pisa, Dipartimento di Chimica e Chimica Industriale, Via
Risorgimento 35, I-56126 Pisa, Italy. E-mail: fabmar@dcci.unipi.it;
Fax: +39 050 2219246; Tel: +39 050 2219245
cUniversity of Bologna, Dipartimento di Chimica Fisica e Inorganica,
Viale Risorgimento 4, I-40136 Bologna, Italy
Results and discussion
†Electronic supplementary information (ESI) available. CCDC 876459
(3eB) and 876460 (5A). For ESI and crystallographic data in CIF or
other electronic format see DOI: 10.1039/c2dt31453c
When suspensions of MF5 (M = Nb, 1a; M = Ta, 1b) in
dichloromethane were treated with 1–3 molar equivalents of
12898 | Dalton Trans., 2012, 41, 12898–12906
This journal is © The Royal Society of Chemistry 2012

View Article Online
fluorines in the range 156–85 ppm, and the resonances related to
the bridging ones in the range −66 to −84 ppm. It should be
noted here that the low-temperature 19F NMR patterns (in
CD2Cl2) of the anions [M2F11]− (M = Nb, Ta) consist of three
resonances, e.g. at δ = 116 and 71 ppm (terminal-F) and
−74 ppm (bridging-F) in the case of [Ta2F11]−.16b
The IR spectra of the mixtures 2a–f (see Experimental) are
helpful in the identification of the structure of the related com-
pounds, indeed they suggest the almost absence of bridging alk-
oxides. For instance, the IR spectrum of 2c shows two clear,
strong absorptions in the 1000–1100 cm−1 region (i.e. at 1096
and 1063 cm−1). Since the IR spectrum of [Nb(OEt)5]2 com-
prises two C–O bands due to the terminal ligands at 1110 and
1066 cm−1 and one C–O band attributed to the bridging ligands
at 1030 cm−1 22
it may be concluded by comparison that 2c does
,
not contain bridging ethoxides.
Scheme 1 New MF5−n(OR)n compounds.
DFT calculations were carried out with the aim to predict the
molecular structures of [TaF4(OR)]n (R = Me, 2b; R = Ph, 2f).
Three nuclearities were considered, i.e. n = 4, 3, 2: this appears
reasonable since tantalum pentafluorides are tetranuclear21 while
tantalum pentaalkoxides are generally dinuclear in the solid state
(see Introduction). Hence a series of plausible isomeric struc-
tures, for each nuclearity, were optimized for the gas phase; an
implicit solvation model for CH2Cl2 was added subsequently.
An analogous study was carried out on TaF4(OEt) (2d), by con-
sidering a restricted series of possibilities. The structures calcu-
lated for 2b,d,f and the related energies are represented
respectively in Fig. S1–S4 (2b), S5–S8 (2f) and S9 (2d), within
the ESI.†
trimethylsilyl ethers ROSiMe3 (R = Me, Et, Ph), exothermic dis-
solution of the solid took place. The metal containing products
MF4(OR) (2a–f), MF3(OR)2 (3a–f) and TaF2(OMe)3 (4) could
be isolated in pure form upon removal of the volatile materials,
see Scheme 1. The compounds 2–4 are solids at 20 °C, apart
from MF3(OEt)2 which appear as viscous liquids. The colours
range from colourless to yellow, with the exception of the pale
blue NbF3(OMe)2, probably due to some Nb(IV) impurity. NMR
experiments pointed out that the reactions in CD2Cl2 proceeded
with complete consumption of the organic reagent and clean for-
mation of FSiMe3, according to eqn (1).
The most probable structures of 2b,d,f in dichloromethane are
tetranuclear and homologous (respectively 2bN, 2dN and 2fN).
However the DFT calculations have indicated that three possible
structures of 2b (i.e. 2bJ, trinuclear; 2bO and 2bR, tetranuclear)
have energies [referred to the TaF4(OMe) unit] higher than that
of 2bN by less than 1.5 kcal mol−1, after addition of the sol-
vation model (Fig. S4†). Similar considerations are valid for 2f
(Fig. S8†). The ground-state energy values of dinuclear species
are significantly higher in the case of both 2b and 2f (Fig. S4
and S8†). The same stability scale (dinuclear ≪ trinuclear <
tetranuclear) has been found for the ethoxy-derivative TaF4(OEt)
(2d), see Fig. S9.†
MF5 þ nROSiMe3 ! MF5ꢀnðORÞn þ nFSiMe3
ð1Þ
NMR samples were prepared respectively by addition of 1, 2
and 3 molar equivalents of ROSiMe3 (R = Me, Et, Ph) to
CD2Cl2 suspensions of MF5, at room temperature. Thus the
NMR spectra were recorded at low temperature in order to
prevent ligand-exchange processes. Furthermore IR analyses
were carried out, when possible, on the solid mixtures obtained
by the reactions conducted in CH2Cl2, after removal of the vola-
tile materials.
The four lowest-energy structures calculated for 2b are shown
in Fig. 1.
Compounds MF4(OR) (2a–f)
The 13C spectra of MF4(OPh) at 188 K display one resonance
accounting for ipso-carbon nuclei (e.g. at 159.7 ppm in the case
of M = Ta, 2f). This fact suggests the presence of one single
species bearing symmetrically-arranged phenolato ligands. Ana-
logous conclusion may be traced for MF4(OEt) and TaF4(OMe),
The DFT outcomes regarding the tantalum complexes 2b,d,f
may be extended to the corresponding niobium ones (2a,c,e). In
fact the NMR patterns of 2a,c,e resemble those of 2b,d,f,
respectively. In addition, we have recently demonstrated
thoroughly that the identity of the metal does not play a percepti-
ble role in the chemistry of MF5 (M = Nb, Ta) and their
derivatives.1a
The presumed prevalent formation of tetranuclear species in
the course of the 1 : 1 reactions of MF5 with ROSiMe3 is in
accord with previous hypothesis on the nuclearity of NbF4(OEt),
based on cryoscopy in benzene.18a
on account of the fact that the H and 13C spectra contain one
1
largely-prevalent set of resonances [e.g. for 2d: δ(1H) = 5.22,
(q, 2 H, CH2), 1.60 (t, 3 H, CH3) ppm; δ(13C) = 81.6 (CH2),
16.7 (CH3) ppm]. On the other hand, NbF4(OMe) is obtained in
several forms (see Experimental).
The 19F NMR spectra of 2a–f at room temperature display
very broad signals; on lowering the temperature, the signals
become progressively more resolved and the integrated ratio
between resonances due to terminally-bound fluorines and reso-
nances due to bridging fluorines increases. The 19F spectrum of
2f at 188 K shows the resonances related to the terminal
However the theoretical existence of different forms of com-
parable energies, bearing tetra- or trinuclear structure, may
explain the formation of more than one product. According to
this, the NMR spectra of 2a,b,d at low temperature display more
1
than one set of resonances; for instance, the H spectrum of 2a
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 12898–12906 | 12899

View Article Online
Fig. 2 Molecular structure of [NbF2(μ-F)(OPh)2]3, 3eB, with key
atoms labelled. The H-atoms have been omitted for clarity. Thermal
ellipsoids are at the 50% probability level.
Table 1 Selected bond distances (Å) and angles (°) for 3eB
Nb(1)–O(1)
Nb(1)–F(1)
Nb(1)–F(4)
Nb(2)–O(3)
Nb(2)–F(1)
Nb(2)–F(6)
Nb(3)–O(5)
Nb(3)–F(2)
1.808(6)
2.073(5)
1.873(5)
1.829(5)
2.083(4)
1.898(4)
1.815(6)
2.070(5)
1.882(5)
81.69(18)
81.35(18)
153.3(2)
160.8(2)
160.8(2)
99.5(3)
Nb(1)–O(2)
Nb(1)–F(3)
Nb(1)–F(5)
Nb(2)–O(4)
Nb(2)–F(2)
Nb(2)–F(7)
Nb(3)–O(6)
Nb(3)–F(3)
1.835(5)
2.071(4)
1.879(5)
1.822(5)
2.091(4)
1.878(5)
1.827(6)
2.075(5)
1.870(5)
80.65(17)
155.9(2)
157.7(3)
159.6(2)
99.4(3)
Fig. 1 Lowest-energy DFT-optimized structures of TaF4(OMe).
shows singlets at δ 4.69, 4.59, 4.49, 3.93, 3.84 ppm, in an
8 : 6 : 4 : 1 : 2 ratio (see Experimental). Otherwise the NMR
spectra of 2c,e,f show single sets of resonances, thus suggesting
Nb(3)–F(8)
Nb(3)–F(9)
F(3)–Nb(1)–F(1)
F(2)–Nb(3)–F(3)
Nb(3)–F(2)–Nb(2)
F(4)–Nb(1)–F(5)
F(8)–Nb(3)–F(9)
O(3)–Nb(2)–O(4)
F(1)–Nb(2)–F(2)
Nb(1)–F(1)–Nb(2)
Nb(1)–F(3)–Nb(3)
F(6)–Nb(2)–F(7)
O(1)–Nb(1)–O(2)
O(5)–Nb(3)–O(6)
1
the formation, in each case, of a unique product form. The H
NMR spectrum of a CDCl3 solution obtained by 1 : 1 reaction of
NbF5 with EtOSiMe3 did not show variations after prolonged
heating at high temperature.
98.0(3)
An important point of agreement between the DFT outcomes
and the spectroscopic features (NMR and IR, see above) is given
by the fact that the lowest energy structures calculated for 2
show symmetrically-distributed terminal –OR ligands [see Fig. 1
for TaF4(OMe)].
In view of the tetranuclear structure of TaF5, the reactions
leading to 2a–f might proceed with initial formation of inter-
mediates of formula M4F19(OR). According to DFT calculations,
the most stable Ta4F19(OMe) molecule holds the methoxy group
in the equatorial-terminal position (see Fig. S10†). The further
multiple substitution of fluorides by [OMe] to give 2bN, in
CH2Cl2, is a strongly exothermic process (see eqn (2)).
in Fig. 2, whereas relevant bond lengths and angles are reported
in Table 1.
Compound 3eB is a cyclic trimer composed of three (dis-
torted) octahedral Nb(V) units bridged by three F-atoms. Each
niobium is coordinated to two terminal OPh (in relative cis-equa-
torial positions), two terminal fluorides (in trans-axial positions)
and two bridging fluorides (lying on the same plane of the
phenolato ligands). As far as we are aware, no other Nb3(μ-F)3
species has been known hitherto. For the sake of comparison, it
should be cited that NbF5 in the solid state adopts a tetrameric
structure, composed of octahedral units.23 In the case of 3eB, the
Nb–Fterminal interactions [average 1.880(12) Å] are shorter than
the bridging contacts [average 2.077(11) Å], as found also in
[NbF5]4 [Nb–Fterminal 1.77 Å; Nb–Fbridging 2.06 Å]. Nonetheless,
it must be remarked that whereas the Nb–Fbridging distances in
both 3eB and [NbF5]4 are almost identical, Nb–Fterminal inter-
actions in 3eB are 0.11 Å longer than in [NbF5]4. This difference
likely originates from the different electronic and steric proper-
ties of the ligands present in the two complexes herein described,
i.e. [NbF5]4 and 3eB. The Nb–F–Nb angle [average 155.6(4)°] is
considerably bended in view of the trimeric nature of 3eB, com-
pared to the almost linear Nb–F–Nb interactions [182.5°] found
in the tetrameric [NbF5]4. The Nb–O lengths are in the range of
typical niobium–terminal alkoxide interactions.2a,24
Ta4F19ðOMeÞ þ 3 MeOSiMe3 ! ½TaF4ðOMeÞꢁ4 þ 3 FSiMe3
ΔHcalc ¼ ꢀ44 kcal molꢀ1
ð2Þ
This outcome suggests that partially substituted products are
probably not obtained when the reaction of MF5 with ROSiMe3
is performed with a M/Si = 1 ratio.
Compounds MF3(OR)2 (3a–f)
X-ray quality crystals of [NbF2(μ-F)(OPh)2]3 (3eB) were col-
lected from a solution obtained by 1 : 2 molar reaction of NbF5
with PhOSiMe3. Hence the molecular structure could be eluci-
dated by X-ray diffraction: the ORTEP representation is shown
With the aim to carry out a DFT study on the 1 : 2 reaction of
NbF5 with PhOSiMe3, we considered a series of possible
12900 | Dalton Trans., 2012, 41, 12898–12906
This journal is © The Royal Society of Chemistry 2012

View Article Online
structures of NbF3(OPh)2. The optimized structures are given in
Fig. S11 (ESI†) with the related energies. A solvation model
having the dielectric constant of CH2Cl2 was included in the
calculations.
According to the DFT results, the 1 : 2 reaction of NbF5 with
PhOSiMe3 in CH2Cl2 should afford a mixture of products, with
slight prevalence of [NbF2(μ-F)(OPh)2]3 (3eB). In fact two
additional molecules bearing high symmetry and terminal phe-
nolato ligands (see Fig. S11:† 3eC, trinuclear; 3eD, tetranuclear)
present energies [referred to the NbF3(OPh)2 unit] comparable to
that of 3eB.
A selection of calculated geometric parameters for [NbF2(μ-F)-
(OPh)2]3 (3eB) is supplied within the ESI (Table S1†). A com-
pared reading of Table 1 and Table S1† outlines that the DFT
parameters are in satisfying agreement with the experimental
data (X-ray). A slight overestimation of bond lengths, typical of
most of DFT functionals, is observable.25
Fig. 3 Lowest-energy DFT-optimized dinuclear structures of
TaF2(OMe)3 (4).
TaF3(OMe)2 did not form, according to NMR. Therefore the
resulting mixture probably corresponded to different forms of
TaF2(OMe)3, 4.
The 13C NMR spectrum (at 188 K) of the CD2Cl2 mixture
obtained by 2 : 1 reaction of PhOSiMe3 with NbF5 (3e) exhibits
a pattern including three resonances accounting for ipso carbons
[δ = 162.6, 162.0, 161.4 ppm], indicating the presence of more
than one species. We think that the latter are the trinuclear
isomers 3eB and 3eC, which differ in the orientation of the phe-
nolato ligands (all equatorial in 3eB, half equatorial and half
axial in 3eC), see Fig. S11.† This is suggested by the fact that
the IR spectrum (solid state) of the mixture matches well the
pattern of the superimposed IR spectra computed for 3eB and
3eC. The 19F spectrum of 3e is coherent with the presence of
both terminal (120–91 ppm) and bridging (−53 to −65 ppm)
fluorines.
The low-temperature NMR spectra of 3a–f have evidenced,
for each case, the formation of a largely prevalent product,
bearing symmetrically-distributed alkoxide ligands. Minor
isomers have been detected for 3a–d, while compound 3f has
been obtained as single species.
The DFT study aimed for the identification of the struc-
tures adopted by the compounds of general formula MF3(OR)2
has been extended to 3b. Hence a group of plausible struc-
tures of 3b have been optimized: they are represented in
Fig. S12–S14.†
We calculated several possible structures of 4 by a DFT
method: the computed structures are shown in Fig. S16–S18
(ESI†), together with the corresponding relative energies.
On introduction of the solvation model for CH2Cl2, four low-
energy structures have energy values [referred to the TaF2-
(OMe)3 unit] included in a range of ca. 2.5 kcal mol−1
.
These structures are respectively dinuclear (4C, 4D), trinuclear
(4F) and tetranuclear (4H), see Fig. 3 and Fig. S16–S19 in the
ESI.†
The DFT outcome indicates that the dinuclear frame, that
appears unfavourable in the cases of MF4(OR) and MF3(OR)2
compounds, on theoretical grounds becomes competitive for
TaF2(OMe)3. Interestingly, the 19F NMR spectrum of
4
(in CD2Cl2 at 188 K) displays a multitude of resonances due to
terminal fluorines (in the range 38–21 and at 16 and 9 ppm), and
only very low-intensity signals at negative values of chemical
shifts (δ = −101, −104, −110 ppm), see Experimental. This
fact suggests that dinuclear species lacking bridging fluo-
rines (presumably 4C and 4D) largely prevail in the mixture.26
The preference of alkoxide ligands for bridging positions
is a typical feature of dinuclear Nb(V) and Ta(V) halide-
alkoxides.6b,c,f
In a number of cases, NMR analyses (see the Experimental
section, NMR studies) on the mixtures obtained by the reaction
of MF5 with 3 equiv. of ROSiMe3 pointed out the presence of
variable amounts of unreacted organic material, and/or the pro-
duction of some MF3(OR)2 (2). This evidence implies that the
clean synthesis of MF2(OR)3 compounds is not generally viable
with the present method, at least at room temperature. In other
words, the stepwise substitution of [F] ligands by [OR] groups
in dichloromethane, starting from MF5, seems to become
increasingly less favourable on decreasing the fluorine content.
This trend has been detailed for M = Ta and R = Me, see
eqn (3)–(5) (referring to the most stable structures).
According to the calculations related to the solvated phase,
two trinuclear structures (3bF, 3bG) and two tetranuclear ones
(3bK, 3bL) have energies [referred to the TaF3(OMe)2 unit]
whose differences fall within 1 kcal mol−1 after addition of
the solvation model (Fig. S15†). Conversely dinuclear struc-
tures appear not accessible. The trinuclear molecules 3bF
and 3bG are homologous to 3eB and 3eC, respectively, that
probably arise in prevalence from NbF5 + 2PhOSiMe3 (see
above).
Compounds TaF2(OMe)3 (4a) and TaF(OPh)4 (5)
TaF5 þ MeOSiMe3 ! TaF4ðOMeÞ þ FSiMe3
The reactions of 1a,b with three equivalents of ROSiMe3, in
CD2Cl2 at room temperature, gave complicated mixtures of
products.
ð3Þ
ΔHcalc ¼ ꢀ61 kcal molꢀ1
In the case of TaF5/MeOSiMe3, the reaction took place clearly
with complete consumption of the organic reactant and pro-
duction of three equivalents of FSiMe3; TaF4(OMe) and
TaF4ðOMeÞ þ MeOSiMe3 ! TaF3ðOMeÞ2 þ FSiMe3
ð4Þ
ΔHcalc ¼ ꢀ30 kcal molꢀ1
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 12898–12906 | 12901

View Article Online
The structures 5A and 5B bear two bridging phenoxides and
differ by the orientation of the F-ligands (i.e. axial in 5A,
equatorial in 5B). They present very close energies in CH2Cl2
(ΔEcalc ≈ 0.1 kcal mol−1). Conversely structure 5C, bearing two
bridging fluorides, is less stable than both 5A and 5B by ca.
6 kcal mol−1. A list of the main calculated geometric parameters
of 5A is provided in Table S2,† showing a good agreement with
the X-ray data.
We tried to force the selective formation of 5 by high tempera-
ture reaction of TaF5 with PhOSiMe3, in a 4 : 1 ratio. The reac-
tion was carried out in CDCl3 in a sealed tube at ca. 90 °C for
several hours, and afforded a yellow mixture which was analyzed
by NMR. The NMR data indicate the formation of four equiva-
lents of FSiMe3 and different metal species (see Experimental).
The 19F pattern may be attributed to the presence in solution of
5A and 5B, and of a third product containing bridging fluorines.
The latter is likely not to coincide with 5C (see above), and may
hold nuclearity higher than two.
Fig. 4 Molecular structure of [TaF(OPh)3(μ-OPh)]2, 5A, with key
atoms labelled. The H-atoms have been omitted for clarity. Thermal
ellipsoids are at the 50% probability level.
Table 2 Selected bond distances (Å) and angles (°) for 5Aa
Ta(1)–F(1)
Ta(1)–O(3)
Ta(1)–O(1)
Ta(1)–O(1)–Ta(1_2)
1.901(6)
1.875(6)
2.113(7)
111.2(3)
Ta(1)–O(5)
Ta(1)–O(4)
Ta(1)–O(1_2)
O(1)–Ta(1)–O(1_2)
1.891(7)
1.879(7)
2.100(6)
68.8(3)
Conclusions
A series of polynuclear fluoride-alkoxides of niobium and tanta-
lum in the highest oxidation state, MFx(OR)5−x (x = 2–4), have
been prepared and spectroscopically characterized. The reactions
of MF5 with different amounts of trimethylsilyl ethers, in a
chlorinated solvent, take place to afford metal products of
given empirical formulas. However mixtures of isomers and/or
species with different nuclearity may form in some cases.
Although DFT calculations have been not really conclusive in
the determination of the nuclearities of the new compounds, the
formation of tetra- and trinuclear products appears favourable for
both MF4(OMe) and MF3(OMe)2. Dinuclear frames become
energetically competitive on further decreasing the fluorine
content.
Spectroscopic data and DFT outcomes agree in that highly-
symmetric molecules are produced prevalently, and the –OR
ligands preferentially occupy terminal sites in MF4(OR) and
MF3(OR)2, but bridging sites in TaF2(OMe)3 and TaF(OPh)4.
The X-ray structures of [NbF2(μ-F)(OPh)2]3 and [TaF(OPh)3-
(μ-OPh)]2, which represent the first examples of crystallographic
characterizations of niobium/tantalum fluoride-alkoxides(aryl-
oxides), show agreement with both the spectroscopic features
and the computer results.
a Symmetry transformations used to generate equivalent atoms: −x,
−y − 1, −z.
TaF3ðOMeÞ2 þ MeOSiMe3 ! TaF2ðOMeÞ3 þ FSiMe3
ð5Þ
ΔHcalc ¼ ꢀ23 kcal molꢀ1
On an average, a maximum of three equivalents of fluoride
ligands in MF5 undergo substitution with –OR groups by reac-
tion with ROSiMe3, in CD2Cl2 at room temperature. According
to NMR analyses, the 1 : 4 and 1 : 5 reactions produced only
three equivalents of FSiMe3 per mole of the metal (see the
Experimental section, NMR studies).
In spite of this fact, crystals of the tetrasubstituted product
[TaF(OPh)3(μ-OPh)]2 (5A) could be collected from a CH2Cl2
mixture obtained by a prolonged reaction of TaF5 with a large
excess of PhOSiMe3. X-ray analysis was carried out on 5A: the
ORTEP representation is shown in Fig. 4, whereas relevant bond
lengths and angles are reported in Table 2. The unit cell contains
discrete dinuclear [TaF(OPh)3(μ-OPh)]2 molecules, which reside
on a crystallographic inversion centre. The dinuclear species
adopts an edge sharing bioctahedral geometry, with one terminal
F-ligand (in the axial position), three terminal and two bridging
OPh ligands. The two F-ligands on the different Ta-atoms are in
relative trans position with respect to the equatorial plane com-
prising the Ta-atoms and the two bridging phenoxides. The Ta–
O and O–C distances are comparable with those previously
reported for Ta(V) aryloxides.5g,h,27 In particular, the bridging
OPh ligand is almost symmetric [O(1)–Ta(1) 2.113(7) Å;
O(1_2)–Ta(1) 2.100(6) Å]. The Ta–Obridging contacts are con-
siderably elongated compared to the Ta–Oterminal ones [average:
1.882(12) Å], in analogy with what was reported for [Ta(OTol)3-
(μ-OTol)]2 (Tol = 4-C6H4Me).5h
Experimental
General
All the metal compounds cited herein are air/moisture sensitive
and were manipulated under an atmosphere of pre-purified argon
using standard Schlenk techniques. The reaction vessels were
oven dried at 140 °C prior to use, evacuated (10−2 mm Hg) and
then filled with argon. NbF5, 1a, and TaF5, 1b, were purchased
from Strem (99.5% purity), stored as received in sealed tubes
under argon, and used without further purification. Solvents
(Sigma-Aldrich) were distilled from P4O10 before use. ROSiMe3
(R = Me, Et, Ph; Sigma-Aldrich) and CD2Cl2 (Cortecnet) were
stored under an argon atmosphere as received. Infrared
spectra were recorded at 298 K on a FT IR-Perkin Elmer
Different, possible isomers of [TaF(OPh)4]2 were the object of
DFT investigation: the respective computed molecular structures
(5A–C) are represented in Fig. S20.†
12902 | Dalton Trans., 2012, 41, 12898–12906
This journal is © The Royal Society of Chemistry 2012

View Article Online
1
Spectrometer, equipped with a UATR sampling accessory. H,
901m, 864m, 804m cm−1
.
1H NMR (CD2Cl2, 298 K): δ =
13C and 19F NMR spectra were recorded on a Varian Gemini
200BB instrument; 93Nb NMR spectra were recorded on a
Bruker Avance DRX400 instrument equipped with a BBFO
broadband probe. The chemical shifts for 1H and 13C were
referenced to the non-deuterated aliquot of the solvent; the
chemical shifts for 19F were referenced to external CFCl3;
the chemical shifts for 93Nb were referenced to external
[NEt4][NbCl6].
Conductivity measurements were carried out on CH2Cl2 solu-
tions with an Eutech Con 700 Instrument (cell constant =
1.0 cm−1).28 Carbon and hydrogen analyses were performed on
a Carlo Erba mod. 1106 instrument. The fluoride content was
determined by a fluoride ion selective electrode, after boiling the
sample in an alkaline solution. The metal was analyzed as M2O5
obtained by high temperature treatment of the solid sample with
a HNO3 solution, followed by calcination in a platinum crucible.
The metal analyses were repeated twice in order to check for
reproducibility.
4.96 (br, 2 H, CH2), 1.57 (br, 3 H, CH3) ppm. 1H NMR
(CD2Cl2, 183 K): δ = 4.90 (br, 2 H, CH2), 1.45 (br, 3 H, CH3)
ppm. 13C NMR{1H} (CD2Cl2, 188 K): δ = 85.5 (major, CH2),
84.5 (minor, CH2), 16.5 (CH3) ppm. 19F NMR (CD2Cl2, 188 K):
δ = 172–162, 146–128, 94–85 (m, 3 F, Fterminal), −50 to −62
(s-br, 1 F, Fbridging) ppm. ΛM (293 K) = 0.2 S cm2 mol−1
.
TaF4(OEt), 2d. Colourless rubbery solid, yield 75%. Anal.
calcd for C2H5F4OTa: C, 7.95; H, 1.67; Ta, 59.92; F, 25.16.
Found: C, 8.28; H, 1.55; Ta, 60.12; F, 25.42. IR (solid state):
2991w, 1450w, 1389w, 1270w, 1110vs and 1084s (νCO),
1007w-m, 946w-m, 896m, 853m cm−1
.
1H NMR (CD2Cl2,
3
183 K): δ = 5.22, 4.73 (q, JHH = 7 Hz, 2 H, CH2, ratio 20 : 1),
3
1.60 (t, JHH = 7 Hz, 3 H, CH3) ppm. 13C NMR{1H} (CD2Cl2,
188 K): δ = 81.6 (CH2), 16.7 (CH3) ppm. 19F NMR (CD2Cl2,
188 K): δ = 153–141, 115–99, 76–66 (m, 2.8 F, Fterminal), −67,
−81 (s-br, 1 F, Fbridging) ppm.
NbF4(OPh), 2e. Yellow solid, yield 84%. Anal. calcd for
C6H5F4NbO: C, 27.51; H, 1.92; Nb, 35.46; F, 29.00. Found: C,
27.33; H, 1.80; Nb, 35.21; F, 28.75. IR (solid state): 1587m,
1479m-s, 1224m, 1162m and 1069w (νCO), 1021w, 898vs,
Synthesis isolation and characterization of MF4(OR), MF3(OR)2
and MF2(OR)3
1
750vs, 684vs cm−1. H NMR (CD2Cl2, 298 K): δ = 7.38–7.17
(Ph) ppm. 13C NMR{1H} (CD2Cl2, 298 K): δ = 162.5 (ipso-Ph),
General procedure: ROSiMe3 was added to a suspension of MF5
(0.80 mmol) in CH2Cl2 (10 mL), in the appropriate molar ratio,
and the mixture was stirred at room temperature for 5–10 h. Pro-
gressive dissolution of the solid was observed. The volatile
materials were removed in vacuo; the product was isolated in the
solid state by the treatment of the residue with CHCl3 (2 mL)
and pentane (15 mL). NMR analyses were carried out on
CD2Cl2 solutions (ca. 0.7 mL) prepared by addition of the
appropriate amount of ROSiMe3 to MF5.
1
129.9, 128.0, 119.4 (Ph) ppm. H NMR (CD2Cl2, 188 K): δ =
7.42–6.93 (Ph) ppm. 13C NMR{1H} (CD2Cl2, 188 K): δ = 162.7
(ipso-Ph), 130.2, 128.8, 120.1 (Ph) ppm. 19F NMR (CD2Cl2,
188 K): δ = 198–189, 156–140 (m, 2.2 F, Fterminal), −48 to −64
(m, 1 F, Fbridging) ppm. 93Nb NMR (CD2Cl2, 223 K): δ = 1083
(Δν1/2 = 4 × 104 Hz) ppm.
TaF4(OPh), 2f. Pale yellow solid, yield 85%. Anal. calcd
for C6H5F4OTa: C, 20.59; H, 1.44; Ta, 51.69; F, 21.71. Found:
C, 20.86; H, 1.29; Ta, 51.35; F, 21.84. IR (solid state):
3069w, 1587m, 1480s, 1455w, 1249s-sh, 1223s, 1163m (νCO),
NbF4(OMe), 2a. White solid, yield 80%. Anal. calcd for
CH3F4NbO: C, 6.01; H, 1.51; Nb, 46.47; F, 38.01. Found: C,
6.18; H, 1.32; Nb, 45.80; F, 37.82. IR (solid state): 2946w,
1069w, 1023w, 1001w, 916m, 842m, 750vs, 684s cm−1
.
1444w, 1393w, 1112s (νCO), 983m, 737vs cm−1 1H NMR
.
1H NMR (CD2Cl2, 188 K): δ = 7.40–7.05 (Ph) ppm. 13C NMR
{1H} (CD2Cl2, 188 K): δ = 159.7 (ipso-Ph), 129.9, 127.7,
119.8 (Ph) ppm. 19F NMR (CD2Cl2, 188 K): δ = 156–149,
128–107, 88–85 (m, 2.6 F, Fterminal), −66 to −84 (m, 1 F,
(CD2Cl2, 298 K): δ = 4.60 (s, Me) ppm. 13C NMR{1H}
(CD2Cl2, 298 K): δ = 76.6 (Me) ppm. 1H NMR (CD2Cl2,
188 K): δ = 4.69, 4.59, 4.49, 3.93, 3.84 (s, Me, ratio
8 : 6 : 4 : 1 : 2) ppm. 13C NMR{1H} (CD2Cl2, 188 K): δ =
75.9–73.5 (Me) ppm. 19F NMR (CD2Cl2, 188 K): δ = 177–161,
148–127, 96–85 (m, 3 F, Fterminal), −60 (m, 1 F, Fbridging) ppm.
93Nb NMR (CD2Cl2, 223 K): δ = 1050 (Δν1/2 = 8 × 104 Hz)
ppm.
F
bridging) ppm.
NbF3(OMe)2, 3a. Very light blue solid, yield 73%. Anal.
calcd for C2H6F3NbO2: C, 11.33; H, 2.85; Nb, 43.83; F, 26.89.
Found: C, 11.20; H, 3.01; Nb, 43.42; F, 26.60. IR (solid state):
2939w, 2839w, 1447w, 1260w, 1117s-sh and 1070vs (νCO),
1
992m, 852w-m, 798m cm−1. H NMR (CD2Cl2, 298 K): δ =
TaF4(OMe), 2b. Colourless solid, yield 76%. Anal. calcd
for CH3F4OTa: C, 4.17; H, 1.05; Ta, 62.83; F, 26.39. Found:
C, 4.32; H, 0.97; Ta, 61.68; F, 26.60. IR (solid state):
2957w, 1448w, 1395w, 1138vs (νCO), 972m, 871m, 826m,
4.49 (s, Me) ppm. 13C NMR{1H} (CD2Cl2, 298 K): δ = 68.4
1
(Me) ppm. H NMR (CD2Cl2, 223 K): δ = 4.50, 4.40, 3.89 (s,
Me, ratio 6 : 1 : 0.2) ppm. 13C NMR{1H} (CD2Cl2, 223 K): δ =
68.9 (br, Me) ppm. 19F NMR (CD2Cl2, 211 K): δ = 114–106,
93, 88–80, 73–68 (m, 1.5 F, Fterminal), −71.0 to −77.3 (m, 1 F,
Fbridging) ppm. 93Nb NMR (CD2Cl2, 223 K): δ = 1080 (Δν1/2 =
2.5 × 104 Hz) ppm.
1
679m cm−1. H NMR (CD2Cl2, 188 K): δ = 4.79, 4.04, 3.94
(s, Me, ratio 20 : 1 : 1.5) ppm. 19F NMR (CD2Cl2, 188 K):
δ = 120–93, 76–65, 32 (m, 4 F, Fterminal), −82.0 (s-br, 1 F,
F
bridging) ppm.
NbF4(OEt), 2c. Colourless rubbery solid, yield 77%. Anal.
TaF3(OMe)2, 3b. Colourless rubbery solid, yield 78%. Anal.
calcd for C2H6F3O2Ta: C, 8.01; H, 2.02; Ta, 60.31; F, 19.00.
Found: C, 7.83; H, 2.11; Ta, 60.13; F, 19.10. IR (solid state):
calcd for C2H5F4NbO: C, 11.23; H, 2.36; Nb, 43.42; F, 35.52.
Found: C, 11.03; H, 2.54; Nb, 42.85; F, 35.79. IR (solid state):
2992w, 2942w, 1467w, 1385w, 1096s and 1063vs (νCO), 944m,
1
2990w, 1386w, 1111s (νCO), 986m, 879s, 735s cm−1. H NMR
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 12898–12906 | 12903

View Article Online
(CD2Cl2, 188 K): δ = 4.50, 4.40 (s, Me, ratio 13 : 4) ppm. 13C
NMR{1H} (CD2Cl2, 188 K): δ = 66.8–66.4 (br, Me) ppm.
the formation of FSiMe3 (FSiMe3: 1,2-dichloroethane ≈ 4).
1
Resonances related to metal products appeared as follows. H
NMR (CDCl3, 213 K): δ = 7.40–6.11 (Ph). 13C NMR{1H}
(CDCl3, 213 K): δ = 161.4, 161.0 (major, ipso-Ph), 160.3,
159.4, 159.0 (minor, ipso-Ph); 129.5, 129.1, 128.8, 124.5,
123.2, 122.6, 121.7, 121.4, 120.1, 119.2 ppm. 19F NMR
(CDCl3, 213 K): δ = 47.8 (s, 0.1 F), 35.7 (s, 0.2 F), 33.4 (s,
1.1 F), 32.5 (s, 1.5 F), −89.8 (s, 1.0 F), −91.3 (s, 0.1 F) ppm. IR
and elemental analyses were carried out on the yellow powder
obtained upon removal of the volatile materials. Anal. calcd for
C12H10F3O2Ta: C, 50.36; H, 3.52; Ta, 31.61; F, 13.44. Found: C,
50.16; H, 3.39; Ta, 31.15; F, 13.22. IR (solid state): 3063w,
1586m, 1477vs, 1452w-m, 1284m, 1226vs, 1182s, 1161s,
1067m, 1022m, 1000w-m, 891vs, 861m, 811s, 765m, 747vs,
684s cm−1. Few crystals of [TaF(OPh)3(μ-OPh)]2 (5A) suitable
for X-ray analysis were collected from a CH2Cl2 mixture
obtained by treatment of TaF5 (ca. 1.2 mmol) with excess
PhOSiMe3 (reaction time = 72 h), layered with pentane and
stored at −30 °C for one week.
NbF3(OEt)2, 3c. Colourless viscous liquid, yield 74%. Anal.
calcd for C4H10F3NbO2: C, 20.02; H, 4.20; Nb, 38.71; F, 23.75.
Found: C, 19.96; H, 4.14; Nb, 38.12; F, 23.94. 1H NMR
(CD2Cl2, 188 K): δ = 4.63 (br, 2 H, CH2), 1.34 (br, 3 H, CH3)
ppm. 13C NMR{1H} (CD2Cl2, 188 K): δ = 78.6 (minor, CH2),
78.0 (major, CH2), 17.2 (CH3) ppm. 19F NMR (CD2Cl2, 188 K):
δ = 102–77 (m, 1.8 F, Fterminal), −69 to −76 (m, 1 F, Fbridging
ppm.
)
TaF3(OEt)2, 3d. Colourless viscous liquid, yield 73%. Anal.
calcd for C4H10F3O2Ta: C, 14.64; H, 3.07; Ta, 55.16; F, 17.37.
Found: C, 14.32; H, 3.19; Ta, 54.86; F, 17.12. 1H NMR
(CD2Cl2, 183 K): δ = 4.93 (br, 2 H, CH2), 1.34 (br, 3 H, CH3)
ppm. 13C NMR{1H} (CD2Cl2, 188 K): δ = 80.2 (major, CH2),
79.2 (minor, CH2), 16.8 (CH3) ppm. 19F NMR (CD2Cl2, 188 K):
δ = 106–92, 72–63, 38–30 (m, 3.3 F, Fterminal), −80 to −86
(m, 1 F, Fbridging) ppm.
NbF3(OPh)2, 3e. Yellow powder, yield 86%. Anal. calcd for
C12H10F3NbO2: C, 42.88; H, 3.00; Nb, 27.64; F, 16.96. Found:
C, 42.53; H, 3.11; Nb, 27.21; F, 16.49. IR (solid state): 1633w,
1587m, 1479vs, 1231vs, 1162m (νCO), 1069w-m, 863vs, 804vs,
NMR studies
The reactions of MF5 (0.40–0.80 mmol) with 1–5 equiv. of
ROSiMe3 were performed in CD2Cl2 solutions in sealed NMR
tubes in the presence of CHCl3 as standard (CHCl3/M ratio = 1).
1H, 13C and 19F NMR spectra were recorded after 48 h. Com-
plete consumption of ROSiMe3 was observed in the reactions of
MF5 with 1–3 equiv. of ROSiMe3. Unreacted ROSiMe3 was
found in the mixtures obtained from MF5 and 4–5 equiv. of
ROSiMe3. MeOSiMe3: δ(1H, CDCl3) = 3.44 (s, 3 H, OMe),
0.13 (s, 9 H, SiMe3) ppm. EtOSiMe3: δ(1H, CDCl3) = 3.67 (q,
1
749vs cm−1. H NMR (CD2Cl2, 298 K): δ = 7.19, 7.08 (Ph)
ppm. 13C NMR{1H} (CD2Cl2, 298 K): δ = 162.7 (ipso-Ph),
1
129.8, 125.9, 119.3 (Ph) ppm. H NMR (CD2Cl2, 211 K): δ =
7.36–6.95 (Ph) ppm. 13C NMR{1H} (CD2Cl2, 188 K): δ =
162.6, 161.4 (minor, ipso-Ph), 162.0 (major, ipso-Ph), 129.4,
126.1, 125.1, 120.5, 119.2 (Ph) ppm. 19F NMR (CD2Cl2,
188 K): δ = 120–91 (m, 1.5 F, Fterminal), −53 to −65 (m, 1 F,
F
bridging) ppm. 93Nb NMR (CD2Cl2, 223 K): δ = 1172 (Δν1/2 =
3
3
2.4 × 103 Hz) ppm. Few crystals of [NbF2(μ-F)(OPh)2]3 (3eB)
suitable for X-ray analysis were collected from a CH2Cl2 solu-
tion layered with pentane and stored at −30 °C for a few days.
2 H, JHH = 7.33 Hz, CH2), 1.21 (t, 3 H, JHH = 7.33 Hz,
CH2CH3), 0.13 (s, 9 H, SiMe3) ppm. PhOSiMe3: δ(1H,
CDCl3) = 7.31–6.86 (m, 5 H, Ph), 0.30 (s, 9 H, SiMe3) ppm;
δ(13C, CDCl3) = 155.3 (ipso-Ph), 129.5, 121.5, 120.1 (Ph), 0.23
(SiMe3) ppm. Some unreacted ROSiMe3 (ca. 0.4 equiv.) was
detected from NbF5 + 3 PhOSiMe3 and TaF5 + 3 EtOSiMe3,
respectively. FSiMe3 was detected in the final mixtures [δ(1H,
TaF3(OPh)2, 3f. Yellow powder, yield 81%. Anal. calcd for
C12H10F3O2Ta: C, 33.98; H, 2.38; Ta, 42.66; F, 13.44. Found: C,
1
34.21; H, 2.44; Ta, 41.76; F, 13.28. H NMR (CD2Cl2, 188 K):
δ = 7.40–7.00 (Ph) ppm. 13C NMR{1H} (CD2Cl2, 188 K): δ =
158.1 (ipso-Ph), 128.2, 123.4, 122.1, 117.6 (Ph) ppm. 19F NMR
(CD2Cl2, 188 K): δ = 66–58, 38–22 (m, Fterminal), −86 to −105
(m-br, Fbridging) ppm.
CD2Cl2) = 0.25 ppm, JHF = 7.33 Hz; δ(13C, CD2Cl2) =
3
2
0.34 ppm, JCF = 14.5 Hz; δ(19F, CD2Cl2) = −158.3 ppm] in
variable amounts: FSiMe3/CHCl3 = 1 from MF5 + 1 ROSiMe3;
FSiMe3/CHCl3 = 2 from MF5 + 2 Si(OR)Me3; FSiMe3/CHCl3 ≈
3 from MF5 + 3 ROSiMe3; FSiMe3/CHCl3 ≈ 3 from MF5 + 4
ROSiMe3; FSiMe3/CHCl3 ≈ 3 from MF5 + 5 ROSiMe3.
TaF2(OMe)3, 4. Colourless solid, yield 79%. Anal. calcd for
C3H9F2O3Ta: C, 11.55; H, 2.91; Ta, 57.99; F, 12.18. Found: C,
11.84; H, 3.06; Ta, 57.53; F, 12.37. 1H NMR (CD2Cl2, 188 K): δ
=
4.47, 4.40, 4.31, 4.22, 4.17, 4.12 (s, Me, ratio
X-ray crystallographic study
4 : 5 : 4 : 1 : 1.2 : 1) ppm. 13C NMR{1H} (CD2Cl2, 188 K): δ =
64.3 (major), 63.8, 63.3, 62.4, 62.0, 61.1 (br, Me) ppm. 19F
NMR (CD2Cl2, 188 K): δ = 38–21, 16, 9 (m, 27 F, Fterminal),
−101, −104, −110 (m, 1 F, Fbridging) ppm.
Crystal data and collection details for [NbF2(μ-F)(OPh)2]3·
0.5C6H14 (3eB·C6H14) and [TaF(OPh)3(μ-OPh)]2 (5A) are
reported in Table 3. The diffraction experiments were carried out
on a Bruker Apex II diffractometer equipped with a CCD detec-
tor using Mo-Kα radiation. Data were corrected for Lorentz
polarization and absorption effects (empirical absorption correc-
tion SADABS).29 Structures were solved by direct methods and
refined by full-matrix least-squares based on all data using F2.30
All non-hydrogen atoms were refined with anisotropic displace-
ment parameters, apart from the C6H14 molecule in 3eB·C6H14
which was treated isotropically. H-atoms were placed in
Synthesis and isolation of TaF(OPh)4, 5
TaF5 (0.250 mmol), CDCl3 (0.60 mL), 1,2-dichloroethane
(0.250 mmol) and PhOSiMe3 (1.00 mmol) were introduced into
a NMR tube in the order given. The tube was sealed, then the
mixture was heated at ca. 90 °C for 12 h. A yellow mixture was
obtained. NMR analyses showed the absence of PhOSiMe3 and
12904 | Dalton Trans., 2012, 41, 12898–12906
This journal is © The Royal Society of Chemistry 2012

View Article Online
Table 3 Crystal data and experimental details for [NbF2(μ-F)(OPh)2]3·
C6H14 (3eB·C6H14) and [TaF(OPh)3(μ-OPh)]2 (5A)
Acknowledgements
The authors thank the Ministero dell’Istruzione, dell’Università e
della Ricerca (PRIN 2009 project: ‘New strategies for the control
of reactions: interactions of molecular fragments with metallic
sites in unconventional species’) for financial support and the
CINECA Award N. HP10CRPVUO 2011 for the availability of
high performance computing resources and support.
Complex
3eB·C6H14
5A
Formula
Fw
T (K)
C39H37F9Nb3O6
1051.42
100(2)
0.71073
Triclinic
ˉ
P1
9.5980(18)
11.661(2)
18.367(3)
83.930(2)
75.313(2)
86.150(2)
1975.6(6)
2
C48H40F2O8Ta2
1144.70
100(2)
0.71073
Triclinic
ˉ
P1
9.940(4)
10.003(4)
12.384(5)
68.033(4)
86.135(4))
68.477(4)
1058.8(7)
1
1.795
5.227
556
0.16 × 0.14 × 0.11
1.78–25.03
9079
λ (Å)
Crystal system
Space group
a (Å)
b (Å)
c (Å)
Notes and references
α (°)
β (°)
γ (°)
1 (a) F. Marchetti and G. Pampaloni, Chem. Commun., 2012, 48, 635 and
references therein; (b) G. Pampaloni and A. M. Raspolli Galletti, Coord.
Chem. Rev., 2010, 254, 525; (c) M. O. Guerrero-Pérez and
M. A. Bañares, Catal. Today, 2009, 142, 245; (d) R. Arteaga-Müller,
H. Tsurugi, T. Saito, M. Yanagawa, S. Oda and K. Mashima, J. Am.
Chem. Soc., 2009, 131, 5370; (e) A. Spannenberg, H. Fuhrmann,
P. Arndt, W. Baumann and R. Kempe, Angew. Chem., Int. Ed., 1998, 37,
3363.
Cell volume (Å3)
Z
Dc (g cm−3
)
1.768
0.944
1046
μ (mm−1
F(000)
)
Crystal size (mm)
θ limits (°)
0.18 × 0.15 × 0.11
1.76–25.03
18 800
6969 [Rint = 0.1081] 3583 [Rint = 0.0573]
6969/492/573
0.977
0.0585
0.1550
0.895/−0.885
2 (a) F. Marchetti, G. Pampaloni, Y. Patil, A. M. Raspolli Galletti and
S. Zacchini, J. Polym. Sci., Part A: Polym. Chem., 2011, 49, 1664;
(b) A. Kanazawa, S. Kanaoka and S. Aoshima, J. Polym. Sci., Part A:
Polym. Chem., 2010, 48, 2509; (c) T. B. Do, M. Cai and
M. S. Ruthkosky, Patent WO 2009/152003 A2, GM Global Technology
Operations), Inc., 2009; (d) R. Petroff Saint-Arroman, B. Didillon, A. de
Mallmann, J. M. Basset and F. Lefebvre, Appl. Catal., A, 2008, 337, 78;
(e) J. Choi, J. Kang, J. Song, B. Park, C. Kang and S. Noh, Patent WO
2008/039015 A1, LG Chem, LTD. and LG Petrochemical Co., LTD.,
2008; (f) A. Di Benedetto and M. Aresta, Catal. Today, 2006, 115, 88.
3 (a) L. Appel, R. Fiz, W. Tyrra and S. Mathur, Dalton Trans., 2012, 41,
1981; (b) Y. Nakashima, W. Sakamoto and T. Yogo, J. Eur. Ceram. Soc.,
2011, 31, 2497; (c) A. Chowdhury, J. Bould, M. G. S. Londesborough
and S. J. Milne, J. Solid State Chem., 2011, 184, 317; (d) S. Mishra,
E. Jeanneau, M.-H. Berger, J.-F. Hochepied and S. Daniele, Inorg.
Chem., 2010, 49, 11184; (e) V. Boffa, H. L. Castricum, R. Garcia,
R. Schmuhl, A. V. Petukhov, D. H. A. Blank and J. E. ten Elshof, Chem.
Mater., 2009, 21, 1822; (f) G. A. Seisenbaeva, V. G. Kessler, R. Pazik
and W. Strek, Dalton Trans., 2008, 3412; (g) Z. Xu, S. Wang, J. Ren,
Y. Liu and Z. Fang, Dyes Pigm., 2008, 79, 140; (h) E. Rauwel, G. Clavel,
M.-G. Willinger, P. Rauwel and N. Pinna, Angew. Chem., Int. Ed., 2008,
47, 3592; (i) G. H. Thomas, E. D. Specht, J. Z. Larese, Z. B. Xue and
D. B. Beach, J. Mater. Res., 2008, 23, 3281; ( j) U. Schubert, Acc. Chem.
Res., 2007, 40, 730.
Reflections collected
Independent reflections
Data/restraints/parameters
Goodness of fit on F2
R1 (I > 2σ(I))
3583/144/271
1.011
0.0536
0.1354
5.382/−1.616
wR2 (all data)
Largest diff. peak/hole (e Å−3
)
calculated positions and treated isotropically using the 1.2 fold
Uiso value of the parent atom except for methyl protons, which
were assigned the 1.5 fold Uiso value of the parent C-atom.
Similar U restraints were applied to C-atoms (s.u. 0.01). Two
phenyl rings in 3eB·C6H14 are disordered. Disordered atomic
positions were split and refined using one occupancy parameter
per disordered group. The C6H14 molecule in 3eB·C6H14 is dis-
ordered over two positions related by an inversion centre: the
independent image has been, therefore, refined with a 0.5 occu-
pancy factor and restraining the C–C bond distances (s.u. 0.01)
to 1.47 Å. Large residual electron density is present in 5A close
to the tantalum atoms, due to the strong absorption of the heavy
atoms.
4 Some recent references are: (a) B. Knabe, D. Schütze, T. Jungk,
M. Svete, W. Assenmacher, W. Mader and K. Buse, Phys. Status Solidi A,
2011, 208, 857; (b) Y. Suyama, T. Yamada, Y. Hirano, K. Takamura and
K. Takahashi, Adv. Sci. Technol., 2010, 63, 7; (c) S. Furiukawa,
H. Amino, S. Iwamoto and M. Inoue, J. Phys.: Conf. Ser., 2008, 121,
082006; (d) Z. Cheng, K. Ozawa, M. Osada, A. Miyazaki and
A. Kimura, J. Am. Ceram. Soc., 2006, 89, 1188.
CCDC reference numbers 876459 (3eB) and 876460 (5A)
contain the supplementary crystallographic data for this paper.
5 (a) G. G. Nunes, G. A. Seisenbaeva and V. G. Kessler, J. Sol-Gel Sci.
Technol., 2007, 43, 105; (b) R. N. Kapoor, F. Cervantes-Lee,
C. F. Campana, C. Haltiwanger, K. Abney and K. H. Pannell, Inorg.
Chem., 2006, 45, 2203; (c) T. J. Boyle, J. J. Gallegos III, D. M. Pedrotty,
E. R. Mechenbier and B. L. Scott, J. Coord. Chem., 1999, 47, 155;
(d) S. C. Goel, J. A. Hollingsworth, A. M. Beatty, K. D. Robinson and
W. E. Buhro, Polyhedron, 1998, 17, 781; (e) T. J. Boyle, T. M. Alam,
D. Dimos, G. J. Moore, C. D. Buchheit, H. N. Al-shareef,
E. R. Mechenbier and B. R. Bear, Chem. Mater., 1997, 9, 3187;
(f) T. W. Coffindaffer, B. D. Steffy, I. P. Rothwell, K. Folting,
J. C. Huffman and W. E. Streib, J. Am. Chem. Soc., 1989, 111, 4742;
(g) R. Wang, K. Folting, J. C. Huffman, L. R. Chamberlain and
I. P. Rothwell, Inorg. Chim. Acta, 1986, 120, 81; (h) L. N. Lewis and
M. F. Garbauskas, Inorg. Chem., 1985, 24, 363; (i) A. A. Pinkerton,
D. Schwarzenbach, L. G. Hubert-Pfalzgraf and J. G. Riess, Inorg. Chem.,
1976, 15, 1196.
Computational studies
The computational geometry optimisation of the complexes was
carried out using the hybrid DFT M06 functional31 without sym-
metry constraints, in combination with a polarized triple-ζ
quality basis set composed of the 6-311G(d,p) basis set on the
light atoms and the LANL2TZ(f) basis set on the metal
centres.32 Implicit solvation was added by using the C-PCM
model for dichloromethane.33 The “restricted” formalism was
applied in all the calculations. The stationary points were charac-
terized as true minima by IR simulations, from which zero-point
vibrational energies were obtained.25 The software used was
Gaussian 09.34 The simulations were performed at CINECA
(Centro Italiano di Supercalcolo, Bologna, Italy) using a IBM
P6-575 workstation equipped with 64-bit IBM Power6
processors.
6 (a) F. Akagi, Y. Ishida, T. Matsuo and H. Kawaguchi, Dalton Trans.,
2011, 40, 2375; (b) F. Marchetti, G. Pampaloni and S. Zacchini, Polyhe-
dron, 2009, 28, 1235; (c) F. Marchetti, G. Pampaloni and S. Zacchini,
Dalton Trans., 2009, 8096; (d) F. Marchetti, G. Pampaloni and
S. Zacchini, Dalton Trans., 2008, 7026; (e) F. Preuss, G. Lambing and
S. Müller-Becker, Z. Anorg. Allg. Chem., 1994, 620, 1812;
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 12898–12906 | 12905

View Article Online
(f) A. Antiñolo, A. Otero, F. Urbanos, S. Garcia-Blanco, S. Martinez-
Carrera and J. Sanz-Aparicio, J. Organomet. Chem., 1988, 350, 25;
(g) L. R. Chamberlain, I. P. Rothwell and J. C. Huffman, Inorg. Chem.,
1984, 23, 2575; (h) K. C. Malhotra, U. K. Banerjee and S. H. Chaudhry,
J. Ind. Chem., 1980, 57, 868.
22 C. G. Barrachlough, D. C. Bradley, J. Lewis and J. M. Thomas, J. Chem.
Soc., 1961, 2601.
23 A. J. Edwards, J. Chem. Soc., 1964, 3714.
24 F. Marchetti, G. Pampaloni and S. Zacchini, Inorg. Chem., 2008, 47, 365.
25 (a) C. J. Cramer, Essentials of Computational Chemistry, 2nd edn, Wiley,
Chichester, 2004; (b) F. Jensen, Introduction to Computational Chemistry,
2nd edn, Wiley, Chichester, 2007.
7 A. Dimitrov, S. Wuttke, S. Troyanov and E. Kemnitz, Angew. Chem., Int.
Ed., 2008, 47, 190.
8 S. K. Ruediger, U. Groβ, M. Feist, H. A. Prescott, S. C. Shekar,
S. I. Troyanov and E. Kemnitz, J. Mater. Chem., 2005, 15, 588.
9 M. Schormann, S. P. Varkey, H. W. Roesky and M. Noltemeyer, J. Orga-
nomet. Chem., 2001, 621, 310.
26 Cryoscopy measurements on presumed [MF2(OEt)3]n compounds in
benzene gave n = 2.5 for M = Nb and n = 2.4 for M = Ta, see ref. 17.
27 M. G. B. Drew and J. D. Wilkins, Inorg. Nucl. Chem. Lett., 1974, 10,
549.
10 (a) M. H. Chisholm, J. C. Huffman and R. L. Kelly, J. Am. Chem. Soc.,
1979, 101, 7100; (b) M. H. Crisholm, D. L. Clark and J. C. Huffman,
Polyhedron, 1985, 7, 1203.
11 M. H. Chisholm, D. L. Clark, R. J. Errington, K. Folting and
J. C. Huffman, Inorg. Chem., 1988, 27, 2071.
12 C. Rennekamp, H. Wessel, H. W. Roesky, P. Müller, H.-G. Schmidt,
M. Noltemeyer, I. Usón and A. R. Barron, Inorg. Chem., 1999, 38, 5235.
13 (a) Yu. A. Buslaev, Yu. V. Kokunov and V. A. Bochkareva,
Russ. J. Inorg. Chem., 1971, 16, 2611; (b) A. Dimitrov, S. Seidel and
K. Seppelt, Eur. J. Inorg. Chem., 1999, 95.
28 (a) A. Jutand, Eur. J. Inorg. Chem., 2003, 2017; (b) W. J. Geary, Coord.
Chem. Rev., 1971, 7, 81.
29 G. M. Sheldrick, SADABS, Program for empirical absorption correction,
University of Göttingen, Göttingen, Germany, 1996.
30 G. M. Sheldrick, SHELX97, Program for crystal structure determination,
University of Göttingen, Göttingen, Germany, 1997.
31 Y. Zhao and D. G. Truhlar, Theor. Chem. Acc., 2008, 120, 215.
32 (a) A. D. McLean and G. S. Chandler, J. Chem. Phys., 1980, 72, 5639;
(b) L. E. Roy, P. J. Hay and R. L. Martin, J. Chem. Theory Comput.,
2008, 4, 1029.
14 R. Latham and R. S. Drago, Inorg. Chem., 1964, 3, 291.
15 A. M. Noble and J. M. Winfield, J. Chem. Soc. A, 1970, 501.
16 (a) R. Bini, C. Chiappe, F. Marchetti, G. Pampaloni and S. Zacchini,
Inorg. Chem., 2010, 49, 339; (b) F. Marchetti, G. Pampaloni and
S. Zacchini, J. Fluorine Chem., 2010, 131, 21; (c) F. Marchetti,
G. Pampaloni and S. Zacchini, Dalton Trans., 2009, 6759.
33 (a) V. Barone and M. Cossi, J. Phys. Chem. A, 1998, 102, 1995;
(b) M. Cossi, N. Rega, G. Scalmani and V. Barone, J. Comput. Chem.,
2003, 669.
34 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, J. A. Montgomery Jr., T. Vreven, K. N. Kudin,
J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone,
B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A. Petersson,
H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li,
J. E. Knox, H. P. Hratchian, J. B. Cross, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi,
C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth,
P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich,
A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck,
K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul,
S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko,
P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-
Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill,
B. Johnson, W. Chen, M. W. Wong, C. Gonzalez and J. A. Pople,
GAUSSIAN03 (Revision A.1), Gaussian, Inc., Pittsburgh, 2003.
17 Yu. V. Kukunov, M. P. Gustyakova, V. A. Bochkareva, Yu. D. Chubar and
Yu. A. Buslaev, Sov. J. Coord. Chem., 1981, 7, 354.
18 (a) Yu. A. Buslaev, Yu. V. Kokunov, S. M. Kremer and
V. A. Shcherbakov, Bull. Acad. Sci. USSR, Inorg. Mater., 1967, 3, 1242;
(b) Yu. A. Buslaev, Yu. V. Kokunov, S. M. Kremer and
V. A. Shcherbakov, Bull. Acad. Sci. USSR, Inorg. Mater., 1967, 3, 1314;
(c) M. Schönherr, Z. Chem., 1976, 16, 374.
19 F. P. Gortsema and R. Didchenko, Inorg. Chem., 1965, 4, 182.
20 J. C. Fuggle, D. W. A. Sharp and J. M. Winfield, J. Chem. Soc., 1972, 1766.
21 The pentafluorides of niobium(V) and tantalum(V) (1a,b) have tetra-
nuclear structure in the solid state. A. F. Wells, Structural Inorganic
Chemistry, Clarendon Press, Oxford, 5th edn, 1993 and ref. 22 For sake
of simplicity, the empirical formulas may be used throughout the present
paper for 1a,b and for their alkoxide derivatives.
12906 | Dalton Trans., 2012, 41, 12898–12906
This journal is © The Royal Society of Chemistry 2012