σ-organoniobium compounds with [NbR4]- and NbR4 stoichiometries

Article abstract of DOI:10.1002/anie.201002666

Securely nested: The first examples of σ-organoniobium compounds with [NbIIIR4]-and NbIVR4 stoichiometries are reported. The Nb centers within the [Nb(C6Cl5)4]q- units (q=0, 1) are located in triakis tetrahedral environments formed by the combination of i

Full text of DOI:10.1002/anie.201002666

Angewandte
Chemie
DOI: 10.1002/anie.201002666
Niobium Compounds
s-Organoniobium Compounds with [NbR4]ꢀ and NbR4
Stoichiometries**
Pablo J. Alonso, Irene Ara, Ana B. Arauzo, M. Angeles Garcꢀa-Monforte, Babil Menjꢁn,* and
Conrado Rillo
Dedicated to Prof. Dr. S. Alvarez on the occasion of his 60th birthday
Despite the well-developed organometallic chemistry of Nb,[1]
homoleptic s-organoniobium compounds remain particularly
scarce and confined to extreme oxidation states. On the one
hand, neutral high-valent (presumably square-pyramidal)
[NbvMe5] is unstable,[2] while the more-saturated trigonal-
prismatic species [NbvMe6]ꢀ is stable enough to allow its
structural characterization.[3] Little is known about the even
more saturated species [NbvMe7]2ꢀ, which appears to have
been detected only in solution.[2b] On the other hand, low-
Scheme 1. Succinct synthetic procedures to obtain compounds 1 and
2 with indication of their electrochemical relationship (E1/2 potential
given vs. SCE).
valent compounds [Nbi(CNXy)7]+ (Xy = 2,6-dimethyl-
phenyl),[4] [Nbꢀi(L)6]ꢀ (octahedral; L = CO,[5] CNXy[4]), and
[Nbꢀiii(CO)5]3ꢀ are also known,[6] all of them containing
typically p-acceptor ligands and complying with the effective
atomic number (18 electron) rule. Herein we report on what
we believe are the first homoleptic s-organoderivatives of Nb
in intermediate oxidation states III and IV. Their properties
are compared with those predicted for the homologous
hydrides [NbiiiH4]ꢀ and NbivH4.[7]
was initially formulated as Li4[NbPh6],[9] it was later found to
be a heteroleptic compound containing h2-coordinated ben-
zyne ligands.[10] In fact, no homoleptic aryl derivative of Nb in
any oxidation state has been isolated so far.
The cyclic voltammogram (CV) of a CH2Cl2 solution of 1
between ꢀ1.6 and + 1.6 V shows a quasi-reversible process
occurring at E1/2 = ꢀ0.44 V (vs. SCE) with DE = 95 mV and
ipa/ipc = 1.06. In view of the moderate potential at which the
electron-exchange process takes place, we treated 1 with Br2,
aiming to prepare and identify the oxidized species. As a
result, the neutral organoniobium(IV) compound [Nbiv-
(C6Cl5)4] (2) was obtained in good yield (Scheme 1). Since
the CV of 2 is similar to that of 1, it follows that both
compounds are electrochemically related by the referred
electron-exchange process. Oxidation of the metal center is
evidenced by an increase in the X-sensitive vibration fre-
quency in the IR spectra, which goes from 824 cmꢀ1 in 1 to
842 cmꢀ1 in 2.[11] The potential of this Nbiv/Nbiii semisystem is
much lower than that observed for the corresponding Viv/Viii
one (E1/2 = + 0.84 V vs. SCE).[12] This large difference (DE =
1.2 V!) evidences a general trend within the Group 5 triad:
intermediate oxidation states are relatively less stable for the
heavier elements Nb (4d) and Ta (5d) than for the lighter 3d
element V. No further features were observed in the CVof the
[Nbiv(C6Cl5)4]/[Nbiii(C6Cl5)4]ꢀ semisystem under the referred
experimental conditions. Tetraaryl derivatives [NbR’4] (R’ =
2-MeC6H4, 2,5-Me2C6H3) had been assumed to form by
reaction of NbCl4 and LiR’ or R’MgBr, but were neither
isolated nor identified.[8]
The low-temperature arylation of NbCl5 with LiC6Cl5
takes place under reduction of the metal center. As a result,
the homoleptic organoniobium(III) compound [NBu4][NbIII-
(C6Cl5)4] (1) is obtained as an air-sensitive violet solid in
moderate yield (Scheme 1). This reaction is quite remarkable,
since early attempts to obtain homoleptic aryl derivatives of
Nb had failed—the reaction of NbCl5 with Li(2-MeC6H4) in
Et2O gave 2,2’-dimethylbiphenyl as the only identifiable
product.[8] Although the reaction product of NbX5 with LiPh
[*] Prof. Dr. P. J. Alonso, Dr. I. Ara, Dr. M. A. Garcꢀa-Monforte,
Dr. B. Menjꢁn, Prof. Dr. C. Rillo
Instituto de Ciencia de Materiales de Aragꢁn (I.C.M.A.)
Universidad de Zaragoza–C.S.I.C.
C/Pedro Cerbuna 12, 50009 Zaragoza (Spain)
Fax: (+34)976-761-187
E-mail: menjon@unizar.es
Dr. A. B. Arauzo
Servicio de Instrumentaciꢁn Cientꢀfica-Medidas Fꢀsicas
Universidad de Zaragoza
C/Pedro Cerbuna 12, 50009 Zaragoza (Spain)
[**] This work was supported by the Spanish MICINN (DGPTC)/FEDER
(projects CTQ2008-06669-C02-01/BQU and MAT2008-03461/MAT)
and the Gobierno de Aragꢁn (Grupo de Excelencia: Quꢀmica
Inorgꢁnica y de los Compuestos Organometꢁlicos). We are indebted to
Prof. Dr. S. Alvarez (Universitat de Barcelona) and Prof. J. Vicente
(Universidad de Murcia) for kindly providing values of CShM and
analytical data, respectively.
The molecular structures of species [Nbiii(C6Cl5)4]ꢀ
(Figure 1) and [Nbiv(C6Cl5)4] (Figure 2) were determined by
X-ray diffraction methods on single crystals of 1 and 2·2C6H14,
respectively.[13] In both cases, the Nb centers are located in
slightly elongated tetrahedral (T-4) environments defined by
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201002666.
Angew. Chem. Int. Ed. 2010, 49, 6143 –6146
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6143

Communications
Figure 3. Local coordination environment of the Nbiv center in
2·2C6H14 (thermal ellipsoid diagram with 50% probability) with
indication of the inner NbC4 and outer NbCl4 tetrahedra. Secondary
ortho-Cl···Nb interactions are indicated with broken lines.
Figure 1. Thermal ellipsoid diagram (50% probability) of the
[Nbiii(C6Cl5)4]ꢀ anion in 1. Secondary ortho-Cl···Nb interactions with
average Nb···Cl=304.2(1) pm are not drawn for clarity. Selected bond
length [pm] and angles [8] with estimated standard deviations: average
Nb–C 222.8(4); C(1)-Nb-C(7) 111.45(14), C(1)-Nb-C(13) 98.89(13),
C(1)-Nb-C(19) 117.63(14), C(7)-Nb-C(13) 116.03(13), C(7)-Nb-C(19)
101.66(12), C(13)-Nb-C(19) 111.98(13).
ꢀ
much longer than those found in standard Nb Cl covalent
bonds (239 pm average)[17] and denote just a loose interac-
tion.[18] The resulting global structure can be described as
triakis tetrahedral (TT-8) with the following CShM values:
S(TT-8) = 1.54 for 1 and 1.38 for 2.[15,19] The whole [Nbiv-
(C6Cl5)4] molecule has exact (crystallographically imposed) S4
symmetry. Essentially the same symmetry was also found in
the [Nbiii(C6Cl5)4]ꢀ unit. It is interesting to note that the
symmetry of the NbC4 core in compounds 1 and 2 is similar to
that theoretically predicted for the binary hydrides [NbiiiH4]ꢀ
and NbivH4.[7]
In this type of compound, the interplay between structural
and magnetic properties is of fundamental importance.[20]
Thus, the spin states of compounds 1 and 2 were determined
by isothermal magnetization measurements, M(m0H), at low
temperature. The dependence of the value of M/mBNA (where
M is the molar magnetization) on the reduced magnetic field
mBH/kBTat 1.8 K (Figure 4) can be described in each case by a
Brillouin function with S = 1 and g = 1.78(5) for compound 1
and S = 1/2 and g = 1.94(3) for compound 2. The thermal
dependence of the magnetic susceptibility c(T) was also
measured for 1 and 2 in the range 1.8–273 K. The observed
evolutions (Figure S1 in the Supporting Information) follow
Figure 2. Thermal ellipsoid diagram (50% probability) of the neutral
species [Nbiv(C6Cl5)4] in 2·2C6H14. Secondary ortho-Cl···Nb interactions
with Nb···Cl=299.8(1) pm are not drawn for clarity (cf. Figure 3).
Selected bond length [pm] and angles [8] with estimated standard
deviations: Nb–C(1) 218.2(3); C(1)-Nb-C(1A) 115.70(8), C(1)-Nb-
C(1B) 97.62(15), C(1)-Nb-C(1C) 115.70(8), C(1A)-Nb-C(1B) 115.70(8),
C(1A)-Nb-C(1C) 97.62(15), C(1B)-Nb-C(1C) 115.70(8).
the C-donor atoms of the C6Cl5 groups according to the values
of continuous shape measure (CShM) obtained for that
geometry: S(T-4) = 0.73 for 1 and 1.07 for 2.[14,15] A slightly
ꢀ
shorter Nb C distance was observed in 2 (218.2(3) pm) than
in 1 (222.8(4) pm average), according to the different
oxidation state of the metal center. Additional secondary
bonding interactions were established between Nb and one of
the ortho-Cl atoms of every C6Cl5 group.[16] Each interacting
Cl atom caps one of the faces of the inner NbC4 tetrahedron.
This capping does not introduce any significant distortion in
the NbC4 central core geometry, thereby building a second
NbCl4 coordination sphere with complementary T-4 geometry
(Figure 3). The associated Nb…Cl distances (ca. 300 pm) are
*
Figure 4. Dependence of M/mBNA on mBH/kBT for compounds 1 ( )
*
and 2 ( ) measured at 1.8 K. The dashed lines correspond to the
Brillouin functions best matching the experimental data in each case.
Data obtained at 5 and 78 K (not shown for clarity) overlap with those
given here.
6144
www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 6143 –6146

Angewandte
Chemie
the Curie law with the following effective magnetic moments:
meff = 2.51(5) mB for compound 1 and meff = 1.68(5) mB for
compound 2. These values nicely approach those expected
for spin-only paramagnetic systems with triplet (meff ꢁ 2.83 mB)
and doublet (meff ꢁ 1.73 mB) spin states, respectively. In view of
the current results, we conclude that the secondary ortho-
Cl···Nb bonding interactions cannot involve donation of two
electrons per Cl atom because in that case, the compound
would be diamagnetic (18 electrons) as in the dodecahedral
pseudohalide complex K5[Nbiii(CN)8].[21] Furthermore, we
suggest that the singlet-spin state used for hypoelectronic
[NbiiiH4]ꢀ (d2) in the calculations[7a] is a nonrealistic approach.
Compound 1 was found to be EPR-silent under our
available experimental conditions.[22] Frozen solutions of
compound 2 gave, in turn, a well-resolved EPR spectrum
(Figure 5). This markedly anisotropic spectrum can be
are related to the principal g values as given in Equa-
tion (1),[12] where P = ge mB gN mN hrꢀ3i.
ꢀ
ꢁ
À
Á
3
7
4
7
Ajj ¼ ꢀP k þ ge ꢀ gjj
þ
ðge ꢀ g?Þ þ
ð1Þ
ꢀ
ꢁ
11
2
7
A? ¼ ꢀP k þ ðge ꢀ g?Þ ꢀ
14
The k parameter is related to the spin density in the metal
nucleus, c, through the expression: c = ꢀ(2/3)khrꢀ3i. By
introducing the experimental values of gk, g?, Ak, and A? in
Equation (1), we obtain that j P j = 352(10) MHz and k =
0.99(4); consequently, hrꢀ3i = 2.69(8) a.u. and j c j =
1.8(1) a.u. Moreover, since the free Nbiv ion has been assigned
a value of P0 = 576 MHz,[25] it follows that in compound 2, P/
P0 = 0.61. Since a similar reduction (0.63) was observed in the
Viv homologous compound,[12] it follows that the outer NbCl4
shell in 2 does not significantly affect the radial distribution of
the unpaired-electron ground-state function. By applying the
same formalism to the matrix-confined species NbivD4,[23]
however, a significantly higher reduction is obtained, P/P0 =
0.45, which points to more covalent metal–ligand bond
character in that case. According to the structural and
magnetic properties of compounds 1 and 2, the secondary
ortho-Cl···Nb interactions defining the outer NbCl4 shell exert
a shielding effect on the NbC4 core without significantly
altering its geometry or electronic state.
In summary, s-organoderivatives with unprecedented
[NbiiiR4]ꢀ and NbivR4 stoichiometries were isolated and
characterized. The Nb centers within the [Nb(C6Cl5)4]qꢀ
units (q = 0, 1) are in S4 sites formed by an inner NbC4 and
an outer NbCl4 coordination spheres (Figure 3) with a global
triakis tetrahedral structure. The geometry of the NbC4 core is
in keeping with that theoretically predicted for the homolo-
gous hydrides [NbiiiH4]ꢀ and NbivH4.[7] This parallelism
underlines the formal relationship between the singly charged
hydride and s-organyl ligands, regardless of the obvious
difference in size. The synthesis of compounds 1 and 2 fills a
gap in the exclusive family of homoleptic s-organoniobium
compounds.
Figure 5. X-band EPR spectrum of 2 measured at 77.4 K on a chilled
CH2Cl2/CH2ClCH2Cl (1:1) solution. EPR spectra of polycrystalline
samples are given in Figure S2 in the Supporting Information.
described by an axial spin-Hamiltonian including a hyperfine
coupling contribution with the 93Nb nucleus (I = 9/2, 100%
natural abundance). By using this approach with values of
gk = 1.883(2), g? = 1.969(2), Ak = 598 MHz, and A? =
258(5) MHz, excellent agreement was obtained between
calculated and experimental spectra (Figure S3 in the Sup-
porting Information). The spectroscopically obtained value of
hgi = 1.940(2) for compound 2 perfectly matches that derived
from bulk magnetization measurements (Figure 4). The solid-
state spectra of polycrystalline samples of 2 (Figure S2 in the
Supporting Information) are also satisfactorily reproduced
with the same set of parameters by applying a suitable line
width in the calculation (Figures S4 and S5 in the Supporting
Information). This coincidence suggests very similar (if not
identical) electronic and structural environments for the Nbiv
center in 2 both in solution and in the solid state. An
elongated T-4 structure was also assigned to the matrix-
confined deuteride NbivD4 on the basis of its EPR spectro-
scopic properties.[23] Since gk < g? < ge (ge = 2.0023 for the free
electron), it can be concluded that the ground state of
compound 2 in an S4 environment is 2B, that is, the Nbiv center
has a (dxy)1 electron configuration.[24] In that case, the
principal values of the hyperfine coupling tensor Ak and A?
Received: May 3, 2010
Published online: July 22, 2010
Keywords: homoleptic compounds · magnetic properties ·
.
niobium · noncovalent interactions · subvalent compounds
[1] A. Otero, A. Antiꢀolo, A. Lara in Comprehensive Organome-
tallic Chemistry III, Vol. 5 (Eds.: R. H. Crabtree, D. M. P.
Mingos, K. H. Theopold), Elsevier, Amsterdam, 2007,
chap. 5.02, pp. 61 – 100; L. G. Hubert-Pfalzgraf in Encyclopedia
of Inorganic Chemistry, 2nd ed. (Ed.: R. B. King), Wiley,
Chichester, UK, 2005, pp. 3462 – 3487; D. E. Wigley, S. D. Gray
in Comprehensive Organometallic Chemistry II, Vol. 2 (Eds.:
E. W. Abel, F. G. A. Stone, G. Wilkinson), Elsevier, Oxford, UK,
1995, chap. 2, pp. 57 – 153; J. A. Labinger in Comprehensive
Organometallic Chemistry I, Vol. 3 (Eds.: G. Wilkinson, F. G. A.
Stone, E. W. Abel), Pergamon, Oxford, UK, 1982, chap. 25,
pp. 705 – 782.
Angew. Chem. Int. Ed. 2010, 49, 6143 –6146
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6145

Communications
[2] a) R. R. Schrock, J. Organomet. Chem. 1976, 122, 209; b) R. R.
[16] The ability of the C6Cl5 group to establish secondary interactions
between the ortho-Cl atoms and an acidic metal center is well
documented: M. A. Garcꢃa-Monforte, P. J. Alonso, J. Forniꢂs, B.
Menjꢁn, Dalton Trans. 2007, 3347.
[17] A. G. Orpen, L. Brammer, F. H. Allen, O. Kennard, D. G.
Watson, R. Taylor, J. Chem. Soc. Dalton Trans. 1989, S1.
[18] The even longer ortho-Cl···M distances found in the isoleptic
species [Viii(C6Cl5)4]ꢀ were not considered as denoting any
bonding interactions (Ref. [12]), given the smaller size of the
metal center: rcov(V) = 134 pm vs. rcov(Nb) = 147 pm, in P.
Pyykkꢄ, M. Atsumi, Chem. Eur. J. 2009, 15, 186.
Schrock, P. Meakin, J. Am. Chem. Soc. 1974, 96, 5288.
[3] S. Kleinhenz, V. Pfennig, K. Seppelt, Chem. Eur. J. 1998, 4, 1687.
[4] M. V. Barybin, W. W. Brennessel, B. E. Kucera, M. E. Minyaev,
V. J. Sussman, V. G. Young, Jr., J. E. Ellis, J. Am. Chem. Soc.
2007, 129, 1141.
[5] F. Calderazzo, U. Englert, G. Pampaloni, G. Pelizzi, R. Zamboni,
Inorg. Chem. 1983, 22, 1865.
[6] M. V. Barybin, J. E. Ellis, M. K. Pomije, M. L. Tinkham, G. F.
Warnock, Inorg. Chem. 1998, 37, 6518.
[7] a) C. A. Bayse, M. B. Hall, J. Am. Chem. Soc. 1999, 121, 1348;
b) P. E. M. Siegbahn, M. R. A. Blomberg, M. Svensson, J. Am.
Chem. Soc. 1993, 115, 4191.
[19] D. Casanova, M. Llunell, P. Alemany, S. Alvarez, Chem. Eur. J.
2005, 11, 1479.
[8] S. U. Koschmieder, B. Hussain-Bates, M. B. Hursthouse, G.
Wilkinson, J. Chem. Soc. Dalton Trans. 1991, 2785.
[9] B. Sarry, V. Dobrusskin, H. Singh, J. Organomet. Chem. 1968, 13,
1.
[10] R. A. Bartlett, P. P. Power, S. C. Shoner, J. Am. Chem. Soc. 1988,
110, 1966; B. Sarry, R. Schaffernicht, Z. Naturforsch. B 1981, 36,
1238.
[11] R. Usꢁn, J. Forniꢂs, Adv. Organomet. Chem. 1988, 28, 219.
[12] P. J. Alonso, J. Forniꢂs, M. A. Garcꢃa-Monforte, A. Martꢃn, B.
Menjꢁn, Chem. Eur. J. 2005, 11, 4713.
[13] CCDC 774466 (1) and 774465 (2·2C6H14) contain the supple-
mentary crystallographic data for this paper. These data can be
obtained free of charge from The Cambridge Crystallographic
Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
[14] J. Cirera, P. Alemany, S. Alvarez, Chem. Eur. J. 2004, 10, 190.
[15] M. Llunell, D. Casanova, J. Cirera, J. M. Bofill, P. Alemany, S.
Alvarez, M. Pinsky, D. Avnir, SHAPE (version 1.1), Universitat
de Barcelona (Spain) and The Hebrew University of Jerusalem
(Israel), 2003.
[20] S. Alvarez, J. Cirera, Angew. Chem. 2006, 118, 3078; Angew.
Chem. Int. Ed. 2006, 45, 3012.
[21] M. B. Hursthouse, A. M. Galas, A. M. Soares, W. P. Griffith, J.
Chem. Soc. Chem. Commun. 1980, 1167; P. M. Kiernan, W. P.
Griffith, J. Chem. Soc. Dalton Trans. 1975, 2489.
[22] From the analysis of the isothermal magnetization results
(Figure 4) as well as from the linear dependence of 1/c(T) over
the whole temperature range (Figure S1a), an upper limit for the
zero-field splitting of approximately 50 GHz can be obtained for
the non-Kramers system 1. Following this estimate, EPR spectra
would be expected to be observable as in the Viii case (see
Ref. [12]), at least in Q band. Our failure to observe any EPR
spectrum for 1 would thus not be related to the energy of the
transitions involved, but rather to experimental broadening of
the signals, possibly arising from manifold distribution of
paramagnetic centers in the bulk sample.
[23] R. J. van Zee, S. Li, W. Weltner, J. Chem. Phys. 1995, 102, 4367.
[24] For the appropriate election of frame of reference, see Ref. [12].
[25] B. R. McGarvey, J. Chem. Phys. 1967, 71, 51.
6146 www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 6143 –6146