Diniobium inverted sandwich complexes with μ-η6:η6-arene ligands: Synthesis, kinetics of formation, and electronic structure

Article abstract of DOI:10.1021/ja311966h

Monometallic niobium arene complexes [Nb(BDI)(N^tBu)(R-C₆H₅)] (2a: R = H and 2b: R = Me, BDI = N,N′-diisopropylbenzene-β-diketiminate) were synthesized and found to undergo slow conversion into the diniobium inverted arene sandwich complexes [[(BDI)Nb(N^tBu)]₂(μ-RC₆H₅)] (7a: R = H and 7b: R = Me) in solution. The kinetics of this reaction were followed by ¹H NMR spectroscopy and are in agreement with a dissociative mechanism. Compounds 7a-b showed a lack of reactivity toward small molecules, even at elevated temperatures, which is unusual in the chemistry of inverted sandwich complexes. However, protonation of the BDI ligands occurred readily on treatment with [H(OEt₂)][B(C₆F₅)₄], resulting in the monoprotonated cationic inverted sandwich complex 8 [[(BDI^#)Nb(N^tBu)][(BDI)Nb(N^tBu)](μ-C₆H₅)][B(C₆F₅)₄] and the dicationic complex 9 [[(BDI^#)Nb(N^tBu)]₂(μ-RC₆H₅)][B(C₆F₅)₄]₂ (BDI^# = (ArNC(Me))₂CH₂). NMR, UV-vis, and X-ray absorption near-edge structure (XANES) spectroscopies were used to characterize this unique series of diamagnetic molecules as a means of determining how best to describe the Nb-arene interactions. The X-ray crystal structures, UV-vis spectra, arene ¹H NMR chemical shifts, and large J_CH coupling constants provide evidence for donation of electron density from the Nb d-orbitals into the antibonding π system of the arene ligands. However, Nb L_3,2-edge XANES spectra and the lack of sp³ hybridization of the arene carbons indicate that the Nb → arene donation is not accompanied by an increase in Nb formal oxidation state and suggests that 4d² electronic configurations are appropriate to describe the Nb atoms in all four complexes.

Full text of DOI:10.1021/ja311966h

Article
pubs.acs.org/JACS
Diniobium Inverted Sandwich Complexes with μ‑η6:η6‑Arene Ligands:
Synthesis, Kinetics of Formation, and Electronic Structure
Thomas L. Gianetti,† Gregory Nocton,†,‡ Stefan G. Minasian,§,∥ Neil C. Tomson,† A. L. David Kilcoyne,⊥
́
Stosh A. Kozimor,∥ David K. Shuh,§ Tolek Tyliszczak,⊥ Robert G. Bergman,*,† and John Arnold*,†
†Department of Chemistry, University of California, Berkeley, California 94720, United States
‡Laboratoire Het
́
er
́
oel
́
em
́
ents et Coordination, UMR CNRS 7653, Ecole Polytechnique, 91128 Palaiseau, France
⊥
§Chemical Sciences Division and Advanced Light Source, Lawrence Berkeley National Laboratory, Berkeley, California 94720,
United States
∥Chemistry Division, Los Alamos National Laboratory, Los Alamos, New Mexico 87545, United States
S
* Supporting Information
ABSTRACT: Monometallic niobium arene complexes [Nb(BDI)(NtBu)(R-C6H5)] (2a: R = H and 2b: R = Me, BDI = N,N′-
diisopropylbenzene-β-diketiminate) were synthesized and found to undergo slow conversion into the diniobium inverted arene
sandwich complexes [[(BDI)Nb(NtBu)]2(μ-RC6H5)] (7a: R = H and 7b: R = Me) in solution. The kinetics of this reaction were
1
followed by H NMR spectroscopy and are in agreement with a dissociative mechanism. Compounds 7a-b showed a lack of
reactivity toward small molecules, even at elevated temperatures, which is unusual in the chemistry of inverted sandwich
complexes. However, protonation of the BDI ligands occurred readily on treatment with [H(OEt2)][B(C6F5)4], resulting in the
monoprotonated cationic inverted sandwich complex 8 [[(BDI#)Nb(NtBu)][(BDI)Nb(NtBu)](μ-C6H5)][B(C6F5)4] and the
dicationic complex 9 [[(BDI#)Nb(NtBu)]2(μ-RC6H5)][B(C6F5)4]2 (BDI# = (ArNC(Me))2CH2). NMR, UV−vis, and X-ray
absorption near-edge structure (XANES) spectroscopies were used to characterize this unique series of diamagnetic molecules as
a means of determining how best to describe the Nb−arene interactions. The X-ray crystal structures, UV−vis spectra, arene 1H
NMR chemical shifts, and large JCH coupling constants provide evidence for donation of electron density from the Nb d-orbitals
into the antibonding π system of the arene ligands. However, Nb L3,2-edge XANES spectra and the lack of sp3 hybridization of
the arene carbons indicate that the Nb → arene donation is not accompanied by an increase in Nb formal oxidation state and
suggests that 4d2 electronic configurations are appropriate to describe the Nb atoms in all four complexes.
shared between the metal and the arene. Support for reduction
of the arene to form an anionic ligand can be derived from
detailed structural investigations, which have revealed elonga-
tion4−24 of the average C−C distances and disruption of arene
planarity.6,7,9,10,13,16,17,19,20 However, evaluating arene charge
based on these structural distortions can be challenging because
both neutral4,6−11 and anionic5,12−24 configurations have been
assigned previously.25 In addition, questions regarding the
mechanism of formation of these bimetallic complexes have not
been resolved fully. In a report on the hafnium species
Hf2I4(PMe2Ph)2(μ-η6:η6-arene), Cotton and co-workers pro-
posed the intermediacy of the monometallic
HfI2(PMe2Ph)2(η6-arene).6 More recently, P. Arnold et al.
provided computational and experimental evidence for
INTRODUCTION
■
Arene complexes of low-valent metal ions are an intriguing class
of molecules because of the high variability in metal−arene
interactions.1−3 Facially bound arene ligands are known to bind
to a wide range of metals in unusual oxidation states using a
variety of bonding modes. In 1983, Kruger et al. reported the
̈
first dinuclear C6H6 complex, in which a benzene ligand was
bound by two CpV moieties in a μ-η6:η6 fashion.4 Although the
inverted arene sandwich motif remains very rare, similar
molecules have been found with s-,5 d-,4,6−12 and f-block13−24
metal ions. Typically, these molecules are synthesized by the
reduction of metal−halide precursors in aromatic solvents.
Upon formation of electron-rich metal centers, coordination of
an aromatic solvent molecule provides chemical stability by
mixing of the occupied metal d-manifold with the vacant π
system of the arene ring. Important questions remain, however,
regarding the bonding in these compounds, such as how to
assign formal oxidation states and how electron density is
Received: December 10, 2012
Published: January 23, 2013
© 2013 American Chemical Society
3224
dx.doi.org/10.1021/ja311966h | J. Am. Chem. Soc. 2013, 135, 3224−3236

Journal of the American Chemical Society
Article
Scheme 1. Synthesis of Complexes 2a−b
formation of U3+ inverted sandwich complexes via a
monometallic arene intermediate.23
Recently, in an effort to explore the catalytic potential of low-
valent niobium, we reported the catalytic, Z-selective hydro-
genation of internal alkynes by the Nb(III) complex [Nb-
(BDI)(NtBu)(CO)2] (1) (BDI = N,N′-2,6-diisopropylphenyl-
β-diketiminate).26 While performing these catalytic studies in
aromatic solvents, we became intrigued by the observation of
arene complexes. These complexes formed in the absence of
CO, presumably by trapping a low-coordinate trivalent niobium
complex “Nb(BDI)(NtBu)”, which was proposed to be an
active species in the catalytic cycle. The study of low-valent,
group V transition metals, such as Nb, could lead to new
chemical transformations and catalytic processes. In earlier
work, remarkable reactivity and redox behavior has been
reported for monomeric low-valent Nb(III) complexes that
were isolated or generated in situ, spanning activation of small
molecules27−32 hydrogenation,26,33 C−C coupling,34−38 and
C−H bond activation reactions.39−44 The study presented here
describes the synthesis and characterization of several new Nb
arene complexes, including the kinetics of formation of the first
bimetallic μ-η6:η6-arene inverted sandwich complexes45 from
the parent monometallic species. Protonation of the BDI
ligands afforded the mono- and dicationic μ-η6:η6-benzene
complexes. All four inverted sandwich complexes were
subjected to a suite of characterization methods, including
single-crystal X-ray diffraction as well as UV−vis and Nb L3,2-
edge X-ray absorption near-edge structure (XANES) spectros-
copies. Each of the four inverted sandwich complexes also bears
a diamagnetic electronic configuration, which allowed an
Figure 1. Thermal ellipsoid plot of the complex 2a. Diisopropyl aryl
groups of the BDI ligand have been removed, and Bu moieties have
been truncated for clarity. Full structure and structural parameters are
presented in the SI.
t
Nb−NBDI distances are 2.246(4) Å, which are similar to those
observed in related BDI complexes of Nb(III) and Nb-
(V).32,44,46 The Nb−Nimido−CMe3 bond angle (175.3(4)°) is
close to linear, and the Nb−Nimido distance is 1.775(3) Å, which
is also typical of values reported in the literature.32,44,46 The
puckered C6H6 ligand of 2a is characteristic of niobium−arene
complexes.47−50,52,53,57,62 The fold angle at C3···C6 is 22.8°,
such that the Nb(1)−C(3) (2.280(5) Å) and Nb(1)−C(6)
(2.284(5) Å) distances are shorter than observed for the other
four Nb−C bonds (average 2.471(4) Å). These Nb−C
distances compare well with Nb−C distances observed
previously for arene compounds (from 2.198(5) Å to
2.513(4) Å).47−62 The C−C bond lengths of the benzene
ring vary, with C(1)−C(2) (1.337(6) Å) and C(4)−C(5)
(1.384(6) Å) being significantly shorter than bonds involving
the C(3) and C(6) atoms (average of 1.450(5) Å).
1
unusual opportunity to undertake an in-depth H and 13C
(1D and 2D) NMR spectroscopic study of the metal−arene
interactions. The combined experimental approach provides
new insight into the electronic structure of d-block metal−
arene complexes.
The 1H NMR spectrum of the η6-benzene complex
[Nb(BDI)(NtBu)(η6-C6H6)], 2a, shows 12 signals, correspond-
ing to a Cs-symmetric Nb(NtBu)(BDI) moiety, with a singlet
(6H) at high-field (3.60 ppm) for the six equivalent benzene
protons. The toluene complex [Nb(BDI)(NtBu)(η6-Me-
RESULTS AND DISCUSSION
■
1
Monometallic Arene Complexes. The Nb(V) complex
[Nb(BDI)(NtBu)Me2]46 (1) reacts with H2 in benzene or
toluene at room temperature to form the complexes [Nb-
(BDI)(NtBu)(R-C6H5)] (R = H, 2a or Me, 2b) as diamagnetic
red powders in high purities and very good yields (93% and
91%, respectively, Scheme 1). Initial efforts to obtain X-ray
quality crystals of 2a−b were thwarted by their propensity to
form mixtures with their bimetallic counterparts upon isolation
(see below). Nevertheless, rapid crystallization of 2a from a
mixture of toluene/hexanes formed material suitable for X-ray
crystallographic analysis (Figure 1, Tables 1 and S1). Although
several niobium arene compounds have been reported,47−62 to
the best of our knowledge 2a is the first niobium−benzene
complex that has been structurally characterized. In 2a the BDI
and imido ligands occupy facial positions in a pseudo-
octahedral geometry. The N−Nb−N angle of the BDI ligand
is 81.4°, and the average N−Nb−Nimido angle is 98.5°. The two
C6H5)], 2b, presents similar H NMR spectroscopic features,
with three sharp resonances at δ = 3.93, 3.71, 3.59 ppm that
were assigned to the aromatic toluene protons and one at δ =
2.0 ppm corresponding to the methyl group of the coordinated
toluene. The η6 aromatic rings of 2a−b exchange with C6D6 (in
C6D6 solution) within 120 and 20 min, respectively, at room
2
temperature. The H NMR spectrum of 2a-d6 also shows a
singlet at 3.60 ppm, in accordance with the proposed structure.
1
These H NMR data are consistent with the presence of an
averaged η6-arene adduct in solution (Scheme 1).
Consistent with this behavior, the coordinated arene rings of
2a−b were found to be readily displaced by carbon monoxide
or 2,6-xylyl-isocyanide, leading to the previously reported
[Nb(BDI)(NtBu)(CO)2] (3)44 and [Nb(BDI)(NtBu)-
(CNXyl)3] (4)32 complexes, respectively (Scheme 2). Reac-
tivity studies also confirm that the coordinated arene ligands in
2a−b serve to stabilize the Nb(BDI)(NtBu) moiety, which can
3225
dx.doi.org/10.1021/ja311966h | J. Am. Chem. Soc. 2013, 135, 3224−3236

Journal of the American Chemical Society
Article
Table 1. Selected Bond Distances (in Å) for Complexes 2a, 7a−b, 8, and 9
2a
7a
7b
8
9
C(1)−C(2)
C(2)−C(3)
C(3)−C(4)
C(4)−C(5)
C(5)−C(6)
C(6)−C(1)
1.337(6)
1.477(7)
1.434(7)
1.384(6)
1.430(8)
1.460(6)
1.420
2.543(5)
2.556(5)
2.280(5)
2.385(6)
2.400(6)
2.284(5)
−
−
−
−
−
−
−
1.956
1.452(3)
1.451(3)
1.453(3)
1.447(3)
1.473(3)
1.450(3)
1.454
1.462(4)
1.476(4)
1.422(5)
1.451(5)
1.443(4)
1.475(4)
1.455
1.474(3)
1.464(3)
1.424(4)
1.432(4)
1.426(3)
1.463(3)
1.447
1.444(3)
1.461(3)
1.452(4)
1.453(3)
1.443(3)
1.465(3)
1.453
average C−C
Nb(1)−C(1)
Nb(1)−C(2)
Nb(1)−C(3)
Nb(1)−C(4)
Nb(1)−C(5)
Nb(1)−C(6)
Nb(2)−C(1)
Nb(2)−C(2)
Nb(2)−C(3)
Nb(2)−C(4)
Nb(2)−C(5)
Nb(2)−C(6)
Nb(1)−Nb(2)
Nb(1)−Cent
Nb(2)−Cent
average Nb(1)−C
average Nb(2)−C
average Nb−C
Nb(1)−N(1)
Nb(1)−N(2)
Nb(1)−N(3)
Nb(2)−N(4)
Nb(2)−N(5)
Nb(2)−N(6)
Nb(1)−N(3)−C(36)
Nb(2)−N(6)−C(70)
2.2599(16)
2.478(2)
2.3273(19)
2.5526(17)
2.6291(19)
2.3324(19)
2.5533(18)
2.3302(19)
2.4785(18)
2.2673(17)
2.324(2)
2.6197(19)
3.914
2.304(4)
2.296(4)
2.630(5)
2.518(4)
2.324(4)
2.446(3)
2.286(4)
2.449(3)
2.328(4)
2.527(5)
2.636(4)
2.289(3)
3.868
2.607(2)
2.504(2)
2.256(2)
2.371(2)
2.325(2)
2.305(2)
2.345(2)
2.202(2)
2.501(2)
2.528(2)
2.748(2)
2.835(2)
3.989
2.301(2)
2.620(2)
2.511(2)
2.325(2)
2.482(2)
2.277(2)
2.605(2)
2.308(2)
2.297(2)
2.460(2)
2.294(2)
2.493(2)
3.876
1.957
1.944
1.919
1.947
−
−
−
1.959
1.945
2.089
1.932
2.430
2.420
2.395
2.419
2.429
2.419
2.527
2.409
2.495
2.245(4)
2.246(4)
1.775(3)
−
2.429
2.419
2.461
2.414
2.2764(16)
2.2136(17)
1.7770(15)
2.2787(16)
2.2266(15)
1.7815(15)
171.64(15)
174.83(15)
2.249(3)
2.271(3)
1.776(3)
2.241(3)
2.271(3)
1.782(3)
175.6(2)
176.5(3)
2.3236(18)
2.3109(17)
1.7860(18)
2.2209(18)
2.2055(18)
1.7718(19)
168.65(14)
174.73(15)
2.289(2)
2.3226(19)
1.774(2)
2.2983(19)
2.3287(19)
1.7819(19)
169.43(19)
166.62(17)
−
−
175.3(4)
−
Scheme 2. Reactivity of Complexes 2a−b
be oxidized by two electrons with suitable substrates. For
example, exposure of a benzene solution of 2a or a toluene
solution of 2b to N2O (1 atm) quickly leads to the formation of
the previously reported μ-oxo dinuclear complex [Nb(BDI)-
(NtBu)(μ-O)]2 (5, Scheme 2).32 In the absence of an external
oxidant, loss of the arene via thermolysis of either 2a−b in THF
leads to formation of the bis-imido Nb(V) complex 6 (Scheme
2), where the BDI ligand has been reductively cleaved. This
latter compound has been observed previously following
hydrogenolysis of 1 in THF, via cleavage of the BDI ligand
in the related Nb(III) species [Nb(BDI)(NtBu)(THF)].32
Bimetallic Arene Complexes. Synthesis. In an attempt to
1
hinder arene exchange with C6D6, H NMR spectra of 2a−b
were recorded in C6D12. Unexpectedly, collecting additional 1H
3226
dx.doi.org/10.1021/ja311966h | J. Am. Chem. Soc. 2013, 135, 3224−3236

Journal of the American Chemical Society
Article
Scheme 3. Formation of Complexes 7a−b
NMR spectra of these solutions after 3 days revealed the clean
formation of new diamagnetic arene species that were
subsequently formulated as μ-η6:η6-arene diniobium inverted
sandwich complexes [[(BDI)Nb(NtBu)]2(μη6:η6-RC6H5)] (R
= H, 7a or Me, 7b; Scheme 3). Red crystals were obtained on a
preparative scale in good yields from saturated, dark-red,
hexanes solutions of either 2a or 2b.
kinetics of this process by 1H NMR spectroscopy. The
experiment was performed in C6D12 with 0.07 M 2a and 4.90
mM internal standard at 334 K. In the absence of benzene, the
kinetic data cannot be fitted to a rate law that is either first or
second order in 2a because the concentration of free benzene is
changing during the experiments. Introducing an excess of
benzene (from 4 to 30 equiv) yields data that are consistent
with a dissociative mechanism, and a rate law that is second
order in [2a] and inverse first order in benzene (Figure 2 and
Unlike the monometallic complexes 2a−b discussed above,
and relative to other inverted sandwich molecules reported in
the literature,4,6−8,10−12,14−23 the bridged arenes of 7a−b did
not exchange with deuterated benzene, toluene, or strong π-
acidic ligands, such as carbon monoxide or isocyanide, even at
temperatures up to 110 °C. To our knowledge, only two other
inverted sandwich complexes did not show exchange with
solvent molecules at reasonable temperatures.9,23 Complexes
7a−b also did not react with small molecules, such as Ph2N2,
Ph2S2, N2O, organic azides, alkynes, CO2, and H2, within a
temperature range of 20 to 110 °C, unlike many other inverted
sandwich complexes.4,8,10,14−17,19,23 In contrast, treatment of
complex 7a with either 1 or 2 equiv of [H(OEt2][B(C6F5)4]
produced two products that were formulated as new inverted
sandwich complexes, [[(BDI#)Nb(NtBu)][(BDI)Nb(NtBu)]
(μ-η6:η6-C6H6)][B(C6F5)4] (8) and [[(BDI#)Nb(NtBu)]2(μ-
η6:η6-C6H6)][B(C6F5)4]2 (9) in which one (8) or two (9) BDI
ligands have been protonated (Scheme 4, BDI# = (ArNC-
Figure 2. Benzene dependence on the observed second-order rate
constant for the formation of 7a..
Scheme 5. Reactions of the Formation of Complex 7a and
Kinetic Law Derivation
Scheme 4. Formation of Complexes 8 and 9
Figure S1). The rate law for the formation of 7a predicted for
the dissociative mechanism illustrated in Scheme 5 is included
at the bottom of the graphic. When benzene is in a large excess,
the condition k−1[Benz] ≫ k2[2a] is satisfied. Therefore, the
formation of complex 7a follows second-order kinetics,
consistent with earlier work,23 with kobs being inversely
proportional to the initial benzene concentration, where k1k2/
k−1 is 0.0723 min−1 at 334 K (Figure 2).
Knowing the rate law, an Eyring analysis was performed in
the presence of a large excess of benzene (0.593 M, 8 equiv)
over the temperature range 313−344 K, at 5 K intervals
(Figures 3 and S2). The large enthalpy of activation (28.8(4)
kcal·mol−1) is consistent with the slow conversion of 2a−7a at
room temperature. The large positive entropy of activation
(Me))2CH2). Compounds 8 and 9 have been synthesized in
diethyl ether and chlorobenzene, respectively, and can both be
isolated on a preparative scale in good yields. The lack of
reactivity at the coordinated benzene ligand may be evidence of
a neutral C6H6 configuration; however, it is difficult to rule out
the possibility that the ancillary ligands are providing steric
protection (see Figure S10).
Kinetic Studies. To the best of our knowledge, experimental
evidence for the formation of an arene bridged dinuclear
complex from a parent, mononuclear species has only been
reported once,23 although related mechanisms have been
suggested since the early 1990s.6 The relatively slow rate of
conversion from 2a to 7a provides an opportunity to follow the
3227
dx.doi.org/10.1021/ja311966h | J. Am. Chem. Soc. 2013, 135, 3224−3236

Journal of the American Chemical Society
Article
Figure 3. Eyring plot for the formation of complex 7a (top).
Activation parameters derived from the Eyring plot (bottom).
(56(2) cal·mol−1·K−1) must be interpreted carefully due to the
relatively small temperature range of the study. However, the
large positive value further supports a dissociative mechanism
in which the first step corresponds to the formation of a three-
coordinate species and free benzene (Scheme 5).
X-ray Crystal Structure Descriptions. Thermal ellipsoid
plots of complexes 7a−b, 8, and 9 are shown in Figure 4, and
bond distances and angles are summarized in Table 1 (see also
the Table S1 for crystallographic parameters). In each of the
four inverted sandwich complexes 7a−b, 8, and 9, the BDI and
imido ligands occupy facial positions in a pseudo-octahedral
geometry. Complexes 7a, 8, and 9 adopt geometries with
dihedral angles between the two imido groups of 114.3°,
132.3°, 117.9° respectively, which minimizes steric repulsion
between the aryl rings of the BDI ligands. The BDI-aryls are
distorted away from the central toluene ligand in 7b, resulting
in a smaller dihedral angle of 40.6° (Figure 4). Average Nb−
NBDI distances are 2.249(2) and 2.258(3) Å in 7a and 7b,
respectively, and 2.311(2) Å in 9. Because 8 is dissymmetric,
two different sets of average Nb−NBDI distances are observed:
2.317(2) Å on the side with the protonated BDI ligand and
2.203(2) Å on the other.
The C−C distances of the arene rings in 7a−b, 8, and 9 are
all very similar and provide an average C−C distance of 1.45 Å
for all four complexes with a standard deviation of 0.02 Å (see
Supporting Information, SI). These distances are slightly longer
than those reported for free benzene and toluene (1.397(1)63
and 1.39(1)64 Å) but in the range of the reported values found
for μ-η6:η6-arene ligands in the literature.4−24
In the four complexes 7a−b, 8 and 9, the bridging arenes are
nonplanar. Observed Nb−C bond distances vary widely for 7a
(2.2673(17)−2.6291(19) Å), 7b (2.286(4)−2.636(4) Å), and
9 (2.277(2) to 2.620(2) Å). The benzene centroid is nearly
equidistant from both niobium atoms in 7a, 7b, and 9, with
average Nb-centroid bond distances of 1.958 Å (7a), 1.944 Å
(7b), and 1.939 Å (9). For complex 8, in which one of the BDI
ligands is protonated, the benzene ligand lies closer to the Nb
bearing the protonated ligand (Nb(1)−C range 2.256(2)−
2.607(2) Å; Nb(1)−centroid 1.919 Å) than it does to Nb atom
with the unprotonated ligand (Nb(2)−C range 2.202(2)−
2.835(2) Å; Nb(2)−centroid 2.089 Å). The Nb(1)−Nb(2)
distances decrease by 0.046(5) Å from the C6H6 complex 7a
(3.914(2) Å) to C7H8 complex 7b (3.868(4) Å), which could
be a consequence of changes in bonding or an increase in steric
pressure in 7b. However, the Nb(1)−Nb(2) distances also
Figure 4. Thermal ellipsoid plots of the complexes 7a (first), 7b
(second), 8 (third) and 9 (fourth). Diisopropyl aryl groups of the BDI
ligand and B(C6F5)4 have been removed, and Bu moieties have been
truncated for clarity. Full structures and structural parameters are
presented in the SI.
t
decrease by 0.038(2) Å from neutral 7a (3.914(2) Å) to twice-
protonated 9 (3.876(2) Å), which is clearer evidence that
protonation disrupts the BDI−Nb interaction and enhances
bonding between the arene ligand and Nb atoms.
The mean deviation of the arene rings from planarity is
0.051, 0.052, 0.050, and 0.053 Å respectively, while the largest
distortions from the mean plane are 0.170 Å (7a), 0.147 Å
(7b), 0.159 Å (8), and 0.163 Å (9). The deviation from
planarity was also measured from the different dihedral angles
of the ring, with average distortions of 14.4° (7a), 15.1° (7b),
15.2° (8), and 15.2° (9) (see the SI). Such deformations of the
3228
dx.doi.org/10.1021/ja311966h | J. Am. Chem. Soc. 2013, 135, 3224−3236

Journal of the American Chemical Society
Article
1
η6,η6-arene ring have been described in the litera-
ture6,7,9,10,13,16,17,19,20 and support three possible conclusions:
(i) the arene ring is significantly reduced, such that there is a
disruption of aromaticity and the carbon atoms are best
regarded as sp3 hybridized; (ii) the distortions are sterically
induced; or (iii) both these electronic and steric factors are at
work. Therefore, additional measurements were performed to
understand whether the origin of these structural deformations
is inherent to the electronic structure of these complexes.
the H NMR time scale at room temperature. Although 9
undergoes slow decomposition at temperatures higher than 335
K, a variable-temperature experiment (295−330 K) provided a
coalescence temperature (Tc = 320(2) K, Figure S5) which was
higher than for complex 7b (Tc = 300(2) K). A SIR experiment
was not possible due to the small chemical shift difference of
the exchanging peaks. The coalescence temperature trend
measured for the geometrical rearrangement of 7a−b, 8, and 9
(Tc < 213 K, = 300 K, < 230 K, and = 320 K, respectively) is
not strictly consistent with what would be expected from steric
considerations. Rather, the solution data suggest that the
interaction between the arene and the metal centers is stronger
in 9 relative to both 7a and 7b, resulting in a higher activation
barrier of the geometric rearrangement.
1
NMR Spectroscopy. The H NMR spectrum of 7a exhibits
one singlet for the methyl groups of both the BDI ligands, one
t
singlet for the equivalent Bu groups of the imido ligands, and
one singlet (δ = 2.21 ppm) for the benzene ring. These data are
consistent with free rotation of the (BDI)Nb(NtBu) moieties in
solution around the Nb···Nb axis, resulting in an average C2v
symmetry on the 1H NMR time scale at room temperature (see
Scheme 6). Although the spectrum of the benzene complex 7a
The 1H NMR resonances of the bridging arene rings of 7a−
b, 8, and 9 are all found in the 2−3 ppm range, which is
consistent with data provided for a diamagnetic, Hf inverted
sandwich of toluene (3.12 to 3.34 ppm).6 The chemical shifts of
the benzene rings observed in 7a (2.21 ppm) were slightly
more shielded than those observed for 9 (2.53 ppm). Shifts to
higher field from the free arene could result from either sp3
hybridization of the carbon due to covalent bonding with the
metal centers or from a strong shielding effect from the flanking
rings on the BDI ligand.
Scheme 6. Proposed Dynamic Processes Observed in
Solution for Complex 7b
Proton-coupled 13C NMR spectroscopy was performed to
determine arene JCH values and further refine the bonding
1
models proposed above. The coupling constants were found to
be JCH = 178.2(6) Hz for 7a, JCH = 178.9(7) Hz for complex 8,
and JCH = 184.5(6) Hz for 9. A similar experiment performed
on the monometallic complex 2a gave a JCH value of 170.5(6)
Hz. These differ from the corresponding parameters for
aromatic sp2 carbons (160(1) Hz) by approximately +20 Hz
for bimetallic complexes 7a, 8, and 9 and by +10 Hz for
complex 2a.65 Such increases in coupling strength are
inconsistent with a decrease in s-character relative to an sp2-
hybridized aromatic carbon and suggest that coordination of
the Nb atoms does not result in a formal sp3 hybridization of
the bridging arene-ring atoms. Taken together, the NMR data
lead us to conclude that the upfield chemical shifts of the
benzene ring are due to shielding induced by the ligand set.
This effect has been previously observed in the Nb(BDI)-
(NtBu) system.44
remains unchanged down to 213 K in C7D8, lowering the
temperature to 213 K for the toluene derivative 7b results in a
1H NMR spectrum that reflects the same C2-symmetric
coordination environment observed in the solid state, thus
1
implying that the dynamic motion has been frozen on the H
1
NMR time scale. Variable-temperature H NMR data collected
between 213 and 383 K provided a coalescence temperature of
Tc = 300(2) K (Figure S4). Thermodynamic parameters for
this geometrical rearrangement were also measured via a
selective inversion recovery (SIR) experiment, yielding ΔH‡=
13.9(6) kcal mol−1 and ΔS‡= −4(2) cal K−1 mol−1 (Figures 5,
S6, and S7).
As an additional note, the amplitude of the observed
coupling constant JCH is also proportional to the electron
density located on the atom.66 Hence, the magnitude of the
increase in JCH is due to electron density donated from the
niobium metal centers to the arene ligands in 2a, 7a, 8, and 9.
The relative amount of donation may be evaluated by
consideration of the JCH coupling constants in a manner
related to the accepted practice of using shifts in νCO as a
measure of the metal-based electron density in carbonyl
Figure 5. Eyring plots for the dynamic processes observed by 1H
NMR in tol-d8 for complex 7b.
complexes.67 In the present case, the H NMR data indicate
1
that Nb → benzene donation increases in the order 2a < 7a ≈
8 < 9. These solution data are also correlated with the slight
decrease observed in Nb(1)−Nb(2) distances in the solid state
for 7a (3.914(2) Å) relative to 9 (3.876(4) Å).
Optical Spectroscopy. UV−vis spectra were recorded to
further investigate the electronic structure of monometallic 2a−
b and bimetallic 7a−b, 8, and 9 (Figure 6 top). Each of the
bimetallic complexes has one band at high energy (340−360
nm, 27 700−29 410 cm−1) that gradually decreases in intensity
in the order 7a = 7b > 8 > 9. These bands are assigned to
intraligand transitions and show a decrease in intensity upon
1
As expected, the room-temperature H NMR spectrum of
complex 8 reflects a lower symmetry coordination environment
than that in 7a−b, involving two singlets (6H) for the methyl
groups of the two different BDI ligands, two singlets (9H) for
each of the Bu substituents on the two imido groups, and a
singlet for the C6H6 ligand (δ = 2.40 ppm). Complex 9, which
has two protonated BDI ligands, exhibits two singlets for the
methyl groups of the BDI ligands, one singlet for the Bu
groups and a singlet for the benzene ring (δ = 2.55 ppm). This
indicates that the dynamic motion in this complex is slow on
t
t
3229
dx.doi.org/10.1021/ja311966h | J. Am. Chem. Soc. 2013, 135, 3224−3236

Journal of the American Chemical Society
Article
Figure 7. Representative background-subtracted Nb L3,2-edge XANES
spectra for compounds 7a−b, 8, and 9. Also included are reference
materials 10, Nb metal, and Nb2O5. All data was collected using a
STXM.
Figure 6. Electronic absorbance spectrum of 87.8 μM solution of 2a in
benzene, 76.0 μM solution of 2b in toluene, 47.8 μM solution of 7a in
benzene, 56.2 μM solution of 7b in toluene, 61.3 μM solution of 8,
and 36.4 μM solution of 9 in α,α,α-trifluorotoluene. Full spectrum
(top), zoom on the visible area (bottom).
dividing by the continuum intensity, as described previously
and in the Experimental Section (Figure 7).69,70 Integer values
for the occupancy of the Nb 4d orbitals were determined
assuming a 4d45s1 valence electron configuration for Nb metal;
however, hybridization may account for some variation in the
4d-orbital occupancies from strictly integer values.70 The
normalized Nb L3,2-edge intensities of 0.48(2) and 0.91(4)
for Nb metal and Nb2O5 reference materials, respectively, agree
well with those previously reported by Bach and co-workers
and are consistent with an increase in Nb L3,2-edge intensity
with decreasing 4d-orbital occupancy (Figure 8).70 As expected,
a +5 formal oxidation state for the Nb atom in 10 is determined
from the normalized edges intensity of 0.89(5). An increase in
Nb 4d-orbital occupancy is reflected by a decrease in the edges
intensity to 0.65(3) and 0.68(4) for 7a and 7b, respectively.
These results suggest that 7a and 7b are best described as
having neutral arene ligands and Nb atoms with 4d2 electronic
configurations (+3 formal oxidation states). The same
electronic configurations for 8 and 9 were also derived from
white-line intensities of 0.69(2) and 0.67(2), respectively.
Using these assignments, a plot of the normalized white-line
intensity versus 4d-orbital occupancy for 7a−b, 8, 9, 10, Nb
powder, and Nb2O5 was developed (Figure 8). A linear fit of
the data in Figure 8 provides a reliable equation (I4d = (0.889
0.012) − (0.107 0.006)n4d) and confirms a 4d2 configuration
for the Nb atoms in 7a−b, 8, and 9. Future efforts will be
focused on comparisons with related d1, d2, and d3 systems in
order to fully assess the extent of mixing between the arene
ligand and Nb atoms.
protonation of the BDI ligands in 8 and 9, which disrupts the
conjugation of the BDI π-system. The spectra also exhibit broad
d−d bands at around 460 nm for 2a−b and 7a−b and 425 nm
for 9 (Figure 6 bottom). Complex 8 shows two bands at 460
and 580 nm, which is in agreement with the dissymmetric
structure. Spectral intensities (ε) in this region are unusually
large for d−d transitions (ε ≈ 2000 cm−1·M−1 for the
monometallic species and 4000 < ε < 8000 cm−1·M−1 for the
bimetallic species). Large values of ε can be explained by a
mixture of d- and π-character in the final state orbital that is
involved in the transition.68 These transitions are, therefore,
partially Laporte allowed, which results in unusually high
intensities for d−d transitions. This concept of metal−arene
mixed transitions has been described previously and supports
Nb−arene mixing in the inverted sandwich complexes.21 The
absorption spectroscopy data reinforce the bonding picture
developed above with the JCH analysis, which suggests that
coordination of a second Nb atom to form a bimetallic Nb−
arene−Nb interaction enhances Nb−C orbital mixing.
XANES spectroscopy. XANES spectra were obtained using a
scanning transmission X-ray microscope (STXM) to evaluate
the 4d-electron count of Nb atoms in 7a−b, 8, and 9 through
comparison with reference materials [Nb(BDI)(NtBu)Cl2py],
10; Nb2O5 and Nb powder (Figure 7). The 4d-orbital
occupancies were determined from normalized Nb L3,2-edge
intensities, in analogy to measurements made using electron
energy-loss spectroscopy (EELS) at the Nb L69−72 and M-
edges.73−75
CONCLUSIONS
In our experiments, the L-edge absorptions arise from dipole
allowed 2p64dn → 2p54dn+1 transitions that are split by roughly
94 eV into L3 (2p3,2) and L2 (2p1/2) edges due to spin−orbit
coupling of the core hole.76 The intensities of the L3 and L2
edges were corrected for the background and normalized by
■
Described here are high-purity syntheses of four diniobium,
inverted sandwich compounds, [[(BDI)Nb(NtBu)]2(μ-
RC6H5)] (7a: R = H and 7b: R = Me), monocationic 8
[[(BDI#)Nb(NtBu)][(BDI)Nb(NtBu)](μ-C6H5)]1+, and dica-
3230
dx.doi.org/10.1021/ja311966h | J. Am. Chem. Soc. 2013, 135, 3224−3236

Journal of the American Chemical Society
Article
may be a kinetic effect due to the steric bulk of the BDI ligands.
Future work will be focused on efforts to further refine our
understanding of electronic structure with computational
modeling and will include extensions of this chemistry to a
wider range of metal systems and ligand sets.
EXPERIMENTAL SECTION
■
General Considerations. Unless otherwise noted, all reactions
were performed either using standard Schlenk line techniques or in an
MBraun inert atmosphere glovebox under an atmosphere of purified
nitrogen (<1 ppm O2/H2O). Glassware, cannulae, and Celite were
stored in an oven at ∼160 °C for at least 12 h prior to use. n-Pentane,
hexanes, Et2O, THF, toluene, and benzene were purified by passage
through a column of activated alumina, stored over 3 or 4 Å molecular
sieves, and degassed prior to use.77 α,α,α-Trifluorotoluene, dichloro-
ethane, and chlorobenzene were dried over P2O5, distilled under
reduced pressure, degassed, and stored over 4 Å molecular sieves.
Deuterated solvents (C6D6, C7D8, and C6D12) were dried over
sodium/benzophenone, and C6D5Cl was dried over CaH2. The
deuterated solvents were then vacuum transferred to a storage flask
and degassed before being stored in the drybox. C6D6, C7D8, and
C6D5Cl were stored over activated molecular sieves. BDI,78
Li(BDI)·Et2O,79 (BDI)pyCl2Nb(NtBu),44 and (BDI)(Me)2Nb-
(NtBu)44 were prepared using literature procedures. All other reagents
were acquired from commercial sources and used as received. NMR
spectra were recorded on Bruker AV-300, AVQ-400, AVB-400, DRX-
500, AV-500, and AV-600 spectrometers. Chemical shifts were
measured relative to residual solvent peaks, which were assigned
Figure 8. A plot of normalized L3,2-edge intensity vs 4d occupancy of
7a−b, 8, 9, 10, Nb2O5, and Nb metal. Error bars are taken from the
standard deviation in the intensities determined over repeated
measurements. A linear fit to all data (dashed line) gives the equation
shown.
1
relative to an external TMS standard set at 0.00 ppm. H and 13C
NMR assignments were routinely confirmed by 1H−1H (COSY,
NOESY) and 1H−13C (HSQC and HMBC) experiments. Samples for
UV−vis NIR spectroscopy were prepared in a Schlenk-adapted quartz
cuvette and analyzed on a Varian Cary 50 scanning spectropho-
tometer. The uncorrected melting points were determined using
sealed capillaries prepared under nitrogen on an Optmelt SRS.
Elemental analyses were performed at the College of Chemistry
Microanalytical Laboratory, University of California, Berkeley. The X-
ray structural determinations were performed at CHEXRAY,
University of California, Berkeley on Bruker SMART 1000 or
SMART APEX diffractometers.
tionic 9 [[(BDI#)Nb(NtBu)]2(μ-RC6H5)]2+ (BDI# = (ArNC-
(Me))2CH2). Both of the neutral complexes 7a and 7b were
synthesized directly from their monometallic precursors 2a and
2b, which is a new approach to access the chemistry of inverted
sandwich compounds. The kinetics of formation of 7a from 2a
were determined using 1H NMR spectroscopy and suggest that
the reaction follows a dissociative mechanism that may involve
a transient, three-coordinate “(BDI)NbNtBu” intermediate.
The combined structural, reactivity, and spectroscopic data
provide new insight into the nuanced bonding interactions that
are involved in bimetallic complexes with bridging arene
ligands. X-ray crystal structures reveal bridging arene ligands
that are distorted from planarity, with carbon atoms deviating
from the average plane by as much as 0.16 Å. However, these
distortions cannot be ascribed to a disruption of arene
aromaticity resulting from Nb→arene donation because of
the undefined role of steric strain, which may also be impacting
(BDI)Nb(NtBu)(η6-C6H6) (2a). Benzene (30 mL) was added to a
100-mL Schlenk flask containing (BDI)(Me)2Nb(NtBu) (0.548 g, 0.90
mmol, 1.0 equiv) at room temperature. The clear yellow solution was
degassed with two freeze−pump−thaw cycles. While warming the
solution during the second cycle, the headspace of the flask was filled
with 1 atm of H2 (70 mL, 2.8 mmol, 3.0 equiv), and the solution was
stirred vigorously. The color of the solution turned to dark red within
5 min. After 12 h, the volatile materials were removed under reduced
pressure to yield a red powder. Yield: 530 mg, 93%. X-ray suitable
crystals were obtained by rapid crystallization from a mixture of
toluene/hexanes (30 mg of 2a in 0.1 mL of solvent at −40 °C for 6 h).
1H NMR (500 MHz, C6D6, 293 K): δ(ppm) 7.10 (t, 2 H, p-Ar, 3JHH
=
3
7.6 Hz), 6.97 (d, 2 H, m-Ar, JHH = 7.2 Hz), 5.13 (s, 1 H,
1
arene reactivity. The H NMR spectra also show increases in
HC(C(Me)NAr)2), 4.53 (sept, 2 H, CHMe2, 3JHH = 6.6 Hz), 3.71 (s, 6
1
the arene JCH coupling constants relative to those for the free
3
H, C6H6), 2.77 (sept, 2 H, CHMe2, JHH = 6.9 Hz), 1.73 (s, 6 H,
arene, which is inconsistent with an sp3 hybridization of the
bridging arene C atoms. Taken together with the results from
the Nb L3,2-edge XANES study, there is little support for a
model based on a reduced arene or oxidized Nb atoms.
t
3
HC(C(Me)NAr)2), 1.45 (s, 9 H, Bu), 1.37 (d, 6 H, CHMe2, JHH
=
3
6.9 Hz), 1.05 (d, 6 H, CHMe2, JHH = 6.6 Hz), 1.04 (d, 6 H, CHMe2,
3JHH = 6.6 Hz), 1.00 (m, 6 H, CHMe2, 3JHH = 6.9 Hz). 13C{1H} NMR
(125.8 MHz, C6D6, 293 K) δ(ppm) 168.56 (C, HC(C(Me)NAr)2),
153.15 (C, Ar), 143.08 (C, Ar), 141.31 (C, Ar), 125.71 (CH, Ar),
125.63 (CH, Ar), 124.06 (CH, Ar), 105.21 (broad, CH, C6H6), 103.56
(CH, HC(C(Me)NAr)2), 33.04 (CH3, NbNtBu, Cβ), 28.46 (CH,
CHMe2 of CNAr), 27.39 (CH, CHMe2 of CNAr), 25.92 (CH3,
CHMe2 of CNAr), 25.71 (CH3, CHMe2 of CNAr), 25.07 (CH3,
CHMe2 of CNAr), 24.79 (CH3, CHMe2 of CNAr), 24.70 (CH3,
HC(C(Me)NAr)2). 13C NMR (100 MHz, C6D6, 293 K) δ(ppm)
1
However, the magnitudes of the increase in JCH also indicate
that some Nb→arene donation is present and increases in the
order 2a < 7a < 9, a conclusion that is also reached upon
examination of the variable temperature 1H NMR data, UV−vis
spectra, and the bridged Nb−Nb distances. The Nb→arene
donation is not accompanied by a change in formal oxidation
state, which is consistent with earlier reports of inverted
sandwich complexes based on d-block metals.4,6−11 Finally, we
conclude that the lack of reactivity at the arene ligands in 7a−b
1
1
105.2 (d, CH, C6H6, JCH = 170 Hz). H NMR (500 MHz, C6D12,
3
4
293 K): δ(ppm) 7.12 (dd, 2 H, m-Ar, JHH = 7.6 Hz, JHH = 1.6 Hz),
7.00 (t, 2 H, p-Ar, 3JHH = 7.6 Hz), 6.92 (dd, 2 H, m′-Ar, 3JHH = 7.6 Hz,
3231
dx.doi.org/10.1021/ja311966h | J. Am. Chem. Soc. 2013, 135, 3224−3236

Journal of the American Chemical Society
Article
4JHH = 1.6 Hz), 5.16 (s, 1 H, HC(C(Me)NAr)2), 4.45 (sept, 2 H,
CHMe2, 3JHH = 6.8 Hz), 3.61 (s, 6 H, C6H6), 2.73 (sept, 2 H, CHMe2,
= 1.6 Hz), 4.92 (s, 2 H, HC(C(Me)NAr)2), 4.13 (sept, 4 H, CHMe2,
3JHH = 6.8 Hz), 3.05 (sept, 4 H, CHMe2, 3JHH = 6.8 Hz), 2.41 (s, 6 H,
C6H6), 1.68 (s, 12 H, HC(C(Me)NAr)2), 1.27 (d, 12 H, CHMe2, 3JHH
= 8.0 Hz), 1.26 (s, 18 H, tBu), 1.23 (m, 6H, pentane), 1.153 (d, 12 H,
CHMe2, 3JHH = 6.4 Hz), 1.145 (d, 12 H, CHMe2, 3JHH = 6.8 Hz), 1.08
t
3JHH = 6.8 Hz), 1.77 (s, 6 H, HC(C(Me)NAr)2), 1.35 (s, 9 H, Bu),
1.26 (d, 6 H, CHMe2, 3JHH = 6.4 Hz), 1.03 (d, 6 H, CHMe2, 3JHH = 6.8
Hz), 1.00 (d, 6 H, CHMe2, 3JHH = 6.4 Hz), 0.98 (m, 6 H, CHMe2, 3JHH
= 6.8 Hz). Anal. calcd for C39H59N3Nb1 powders: C, 70.99; H, 8.56;
N, 6.37. Found: C, 69.61; H, 8.45; N, 6.33. mp: 85−86 °C (decomp).
(BDI)Nb(NtBu)(η6-C6D6) (2a-d6). Benzene-d6 (10 mL) was added
to a 50 mL Schlenk flask containing (BDI)(Me)2Nb(NtBu) (0.219 g,
0.36 mmol, 1.0 equiv) at room temperature. The clear yellow solution
was degassed with two freeze−pump−thaw cycles. While warming the
solution during the second cycle, the headspace of the flask was filled
with 1 atm of H2 (40 mL, 1.6 mmol, 4.5 equiv), and the solution was
stirred vigorously overnight. The volatile materials were removed
under reduced pressure to yield a bright red powder (210 mg, 86%
yield). The 1H and 13C NMR spectra were similar to those of complex
2a with the exception of the resonance assigned to the coordinated
benzene. 2H NMR (92 MHz, C6D6, 293 K): δ(ppm) 3.60 (br, s, 6D);
13C NMR (125 MHz, C6D6, 293 K): δ(ppm) 79.96 (t, CD, C6D6, 1JCD
= 27.5 Hz). mp: (decomp) 84−85 °C.
3
(d, 12 H, CHMe2, JHH = 6.8 Hz), 0.83 (t, 6H, pentane). 13C{1H}
NMR (125.8 MHz, C6D6, 293 K) δ(ppm) 168.37 (C, HC(C(Me)-
NAr)2), 151.77 (C, Ar), 142.53 (C, Ar), 142.05 (C, Ar), 125.96 (CH,
Ar), 125.47 (CH, Ar), 123.93 (CH, Ar), 101.59 (CH, HC(C(Me)-
NAr)2), 79.96 (CH, C6H6), 34.45 (pentane), 32.82 (CH3, NbNtBu,
Cβ), 28.94 (CH, CHMe2 of CNAr), 27.21 (CH, CHMe2 of C
NAr), 25.95 (CH3, CHMe2 of CNAr), 25.86 (CH3, CHMe2 of C
NAr), 25.73 (CH3, CHMe2 of CNAr), 25.70 (CH3, CHMe2 of C
NAr), 25.06 (CH3, HC(C(Me)NAr)2), 22.72 (pentane), 14.25
(pentane). 13C NMR (125.8 MHz, C6D6, 293 K) δ(ppm) 79.96 (d,
1
1
CH, C6H6, JCH = 178.2 Hz). H NMR (500 MHz, C6D12, 293 K):
δ(ppm) 7.04 (dd, 4 H, m-Ar, 3JHH = 7.6 Hz, 4JHH3 = 1.6 Hz), 6.97 (t, 4
3
4
H, p-Ar, JHH = 7.6 Hz), 6.84 (dd, 4 H, m′-Ar, JHH = 7.6 Hz, JHH
=
1.6 Hz), 4.93 (s, 2 H, HC(C(Me)NAr)2), 3.98 (sept, 4 H, CHMe2,
3JHH = 6.8 Hz), 2.91 (sept, 4 H, CHMe2, 3JHH = 6.8 Hz), 2.27 (s, 6 H,
C6H6), 1.69 (s, 12 H, HC(C(Me)NAr)2), 1.13 (d, 12 H, CHMe2, 3JHH
= 6.8 Hz), 1.06 (s, 18 H, tBu), 0.995 (d, 12 H, CHMe2, 3JHH = 7.2 Hz),
(BDI)Nb(NtBu)(η6-C7H8) (2b). Toluene (30 mL) was added to a
100 mL Schlenk flask containing (BDI)(Me)2Nb(NtBu) (0.509 g, 0.83
mmol, 1 equiv) at room temperature. The clear yellow solution was
degassed with two freeze−pump−thaw cycles. While warming the
solution during the second cycle, the headspace of the flask was filled
with 1 atm of H2 (70 mL, 2.8 mmol, 3 equiv), and the solution was
vigorously stirred. The solution rapidly turned dark red within 5 min
and was stirred for 12 h at room temperature. The volatile materials
were removed under reduced pressure to yield a red powder (620 mg,
91%). 1H NMR (500 MHz, C7D8, 293 K): δ(ppm): 7.19 (dd, 2 H, m-
0.96 (d, 12 H, CHMe2, 3JHH = 7.4 Hz), 0.94 (d, 12 H, CHMe2, 3JHH
=
7.5 Hz). Anal. calcd for C77H118N6Nb2: C, 70.41; H, 9.05; N, 6.40.
Found: C, 70.33; H, 9.11; N, 6.51. mp: 117.9−119.4 °C.
[(BDI)Nb(NtBu)]2(μ-η6:η6-C6D6) (7a-d6). The complex 2a-d6 (57.1
mg, 0.086 mmol) was stirred in hexanes (20 mL) for 12 h. The dark-
red solution was then concentrated to ca. 10 mL and stored at −40 °C
to form bright red crystals which were filtered and residual solvent was
1
removed under reduced pressure (43.2 mg, 76% yield). The H and
2
13C NMR spectra were similar to those of complex 2. H NMR (92
3
3
3
Ar, JHH = 8.0 Hz, JHH = 2.0 Hz), 7.09 (t, 2 H, p-Ar, JHH = 7.6 Hz),
MHz, C6D6, 293 K): δ(ppm) 2.2 (br, s, 6D); 13C NMR (125 MHz,
3
4
6.97 (dd, 2 H, m′-Ar, JHH = 7.6 Hz, JHH = 1.2 Hz), 5.07 (s, 1 H,
HC(C(Me)NAr)2), 4.61 (sept, 2 H, CHMe2, 3JHH = 6.5 Hz), 3.93 (t, 1
H, p-C7H8, 3JHH = 7.5 Hz), 3.71 (t, 2 H, m-C7H8, 3JHH = 7.2 Hz), 3.59
1
C6D6, 293 K): δ(ppm) 79.96 (t, CD, C6D6, JCD = 27.45 Hz).
[(BDI)Nb(NtBu)]2(μ-η6:η6-C7H8) (7b). Hexanes (20 mL) was
added to a 50 mL flask containing the complex (2) (0.104 g, 0.11
mmol). The red solution was stirred overnight, and the volatile
materials were removed under reduced pressure to yield a dark-red
powder, which was recrystallized from hexanes at −40 °C (10 mL),
yielding red crystals that were collected by filtration, and residual
solvent was removed under reduced pressure (0.057 mg, 72%). One
molecule of pentane remained present in the lattice of the crystal, as
confirmed by elemental analysis and 1H NMR spectroscopy.
Recrystallization from toluene at −40 °C yielded X-ray suitable
3
3
(d, 2 H, o-C7H8, JHH = 5.0 Hz), 2.74 (sept, 2 H, CHMe2, JHH = 7.0
Hz), 1.70 (s, 6 H, HC(C(Me)NAr)2), 1.45 (s, 9 H, tBu), 1.37 (d, 6 H,
CHMe2, 3JHH = 6.6 Hz), 1.15 (d, 6 H, CHMe2, 3JHH = 6.0 Hz), 1.40 (s,
3 H, CH3 of C7H8), 1.02 (d, 12 H, CHMe2, JHH = 6.6 Hz). 13C{1H}
3
NMR (125.8 MHz, C7D8, 293 K) δ(ppm) 166.6 (C, HC(C(Me)-
NAr)2), 152.3 (C, Ar), 143.9 (C, Ar), 141.1 (C, Ar), 124.7 (CH, Ar),
124.6 (CH, Ar), 124.0 (CH, Ar), 102 (broad, CH, C7H8), 101.5 (CH,
HC(C(Me)NAr)2), 33.4 (CH3, NbNtBu, Cβ), 29.4 (CH, CHMe2 of
CNAr), 27.2 (CH, CHMe2 of CNAr), 26.4 (CH3, CHMe2 of C
NAr), 25.9 (CH3, CHMe2 of CNAr), 25.4 (CH3, CHMe2 of C
NAr), 24.8 (CH3, CHMe2 of CNAr), 24.50 (CH3, HC(C(Me)-
1
crystals within 2 days. H NMR (500 MHz, C7D8, 293 K): δ(ppm)
3
4
7.10 (broad, Ar + C7D8), 6.90 (dd, 4 H, m-Ar, JHH = 6.4 Hz, JHH
=
2.4 Hz), 4.84 (s, 2 H, HC(C(Me)NAr)2), 4.5 (broad, 4 H, CHMe2,
3JHH = 7.0 Hz), 2.88 (broad, 2 H, o-C7H8), 2.83 (broad, 1 H, p-C7H8,),
2.76 (broad, 6 H, m-C7H8 and CHMe2), 1.50 (broad, 12 H,
1
NAr)2), 23.2 (broad, CH3, C7H8). H NMR (500 MHz, C6D12, 293
3
4
K): δ(ppm) 7.14 (dd, 2 H, m-Ar, JHH = 7.5 Hz, JHH = 1.8 Hz), 7.01
(t, 2 H, p-Ar, 3JHH = 7.5 Hz), 6.93 (dd, 2 H, m′-Ar, 3JHH = 7.8 Hz, 4JHH
= 1.5 Hz), 5.12 (s, 1 H, HC(C(Me)NAr)2), 4.54 (sept, 2 H, CHMe2,
t
3
HC(C(Me)NAr)2), 1.41 (s, 18 H, Bu), 1.31 (d, 12 H, CHMe2, JHH
3
= 6.4 Hz), 1.14 (d, 6 H, CHMe2, JHH = 6.8 Hz), 1.04 (d, 18 H,
3
3JHH = 6.6 Hz), 3.81 (q, 1 H, p-C7H8, JHH = 4.5 Hz), 3.59 (d, 4 H,
CHMe2, 3JHH = 7.0 Hz), 0.61 (broad, 6 H, CHMe2), 0.40 (broad, 6 H,
CHMe2). (500 MHz, C7D8, 353 K): δ(ppm) 7.10 (broad, Ar + C7D8),
3
3
o,m-C7H8, JHH = 4.5 Hz), 2.70 (sept, 2 H, CHMe2, JHH = 6.9 Hz),
t
3
1.88 (s, 6 H, HC(C(Me)NAr)2), 1.37 (s, 9 H, Bu), 1.29 (d, 6 H,
CHMe2, JHH = 6.9 Hz), 1.26 (s, 3 H, CH3 of C7H8), 1.12 (d, 6 H,
CHMe2, JHH = 6.6 Hz), 1.04 (d, 6 H, CHMe2, JHH = 6.9 Hz), 1.00
(m, 6 H, CHMe2, JHH = 6.6 Hz). Anal. calcd for C40H61N3Nb1
6.90 (d, 4 H, m-Ar, JHH = 7.6 Hz), 4.86 (s, 2 H, HC(C(Me)NAr)2),
4.39 (sept, 4 H, CHMe2, JHH = 7.0 Hz), 2.88 (broad, 2 H, o-C7H8),
3
3
3
3
2.83 (broad, 1 H, p-C7H8,), 2.76 (broad, 6 H, o-C7H8 and CHMe2),
3
2.04 (s, 3 H, CH3 of C7H8), 1.53 (s, 12 H, HC(C(Me)NAr)2), 1.38 (s,
t
3
powders: C, 71.30; H, 8.68; N, 6.24. Found: C, 69.59; H, 8.33; N,
6.18. mp: (decomp) 92−93 °C.
18 H, Bu), 1.20 (d, 12 H, CHMe2, JHH = 6.5 Hz), 1.07 (d, 12 H,
CHMe2, 3JHH = 7.0 Hz), 1.05 (broad, 12 H, CHMe2), 0.77 (broad, 12
H, CHMe2). 13C{1H} NMR (125.8 MHz, C7D8, 353 K) δ(ppm)
167.95 (C, HC(C(Me)NAr)2), 151.66 (C, Ar), 143.23 (C, Ar), 126.25
(CH, Ar), 124.42 (CH, Ar), 102.98 (CH, HC(C(Me)NAr)2), 87.53
(CH, C7H8), 43.06 (CH3, C7H8), 32.90 (CH3, NbNtBu, Cβ), 28.29
(CH, CHMe2 of CNAr), 27.43 (CH, CHMe2 of CNAr), 25.95
(CH3, CHMe2 of CNAr), 26.52 (CH3, CHMe2 of CNAr), 25.73
(CH3, CHMe2 of CNAr), 25.54 (CH3, CHMe2 of CNAr), 25.29
(CH3, HC(C(Me)NAr)2). Anal. calcd for C78H120N6Nb2: C, 70.57; H,
9.11; N, 6.33. Found: C, 70.24; H, 9.85; N, 6.20. mp: 132.1−133.7 °C.
Formation of [(BDI)Nb(NtBu)(L)x] (3: L = CO, x = 2 and 4: L =
XylNC, x = 3). CO (1 atm) was added to an evacuated J-Young tube
containing 2a (12.0 mg, 18.2 mmol) or 2b (17.2 mg, 25.7 mmol) in
[(BDI)Nb(NtBu)]2(μ-η6:η6-C6H6) (7a). Hexanes (50 mL) was added
to a 100 mL Schlenk flask containing 2a (0.378 g, 0.575 mmol). The
dark-red solution was stirred for 12 h. The volatile materials were
removed under reduced pressure to yield a red powder, which was
dissolved in pentane and concentrated to ∼10 mL. The solution was
stored at −40 °C, and the bright-red crystals that formed within 1 day
were filtered, and residual solvent was removed under reduced
pressure. One molecule of pentane remained present in the crystal
lattice, as confirmed by elemental analysis and 1H NMR spectroscopy.
X-ray suitable crystals were obtained by recrystallization from Et2O.
Yield: 0.305 g, 80%. 1H NMR (500 MHz, C6D6, 293 K): δ(ppm) 7.10
(t, 4 H, p-Ar, 3JHH = 7.6 Hz), 6.97 (dd, 4 H, m-Ar, 3JHH = 7.6 Hz, 4JHH
3232
dx.doi.org/10.1021/ja311966h | J. Am. Chem. Soc. 2013, 135, 3224−3236

Journal of the American Chemical Society
Article
C6D12, resulting in a slow color change from dark red to green within 6
or 4 h, respectively. Solution analysis by 1H and 13C NMR
spectroscopies agreed with the formation of the previously reported
complex 3 (BDI)Nb(NtBu)(CO)2.44 Similarly, 5 equiv of XylNC
(10.9 mg, 83.1 mmol) were added to a J-Young tube containing 2a
(11.2 mg, 17.8 mmol) or 2b (13.5 mg, 20.0 mmol) in C6D12, resulting
in a fast color change from dark red to purple within 2 h or 40 min,
residual solvent was removed under reduced pressure (100 mg, 42%).
One molecule of α,α,α-trifluorotoluene remained present in the lattice
of the crystal, as confirmed by elemental analysis, X-ray crystallog-
1
raphy, and 1H NMR spectroscopy. H NMR (500 MHz, C6D5Cl, 293
K): δ (ppm) 7.38 (broad, 2 H, Ar), 7.20 (broad, 4 H, Ar), 7.06 (broad,
2 H, Ar), 6.90 (broad, 2 H, Ar), 6.84 (broad, 2 H, Ar), 4.78 (d, 1 H,
2
CHH′(C(Me)NAr)2), JHH = 16.5 Hz), 3.98 (d, 1 H, CHH′(C(Me)-
1
respectively. The product formed was assigned by H and 13C NMR
NAr)2), 2JHH = 16.2 Hz), 3.30 (broad, 2 H, CHMe2), 3.14 (broad, 2 H,
CHMe2), 2.55 (s, 6 H, C6H6), 1.90 (s, 6 H, CHH′(C(Me)NAr)2), 1.83
(broad, 10 H, CH(C(Me)NAr)2 + 2 CHMe2), 1.08 (broad, 12 H,
CHMe2), 0.94 (broad, 12 H, CHMe2), 0.84 (s, 30 H, tBu), 0.73 (broad,
12 H, CHMe2). 13C{1H} NMR (125.8 MHz, C6D5Cl, 293 K) δ(ppm)
167.95 (C, H2C(C(Me)NAr)2), 150.5 (C, Ar), 147.3 (C, Ar), 140.03
(CH, Ar), 138.5 (CH, Ar), 84.16 (CH, C6H6), 68.8 (CH2,
H2C(C(Me)NAr)2), 31.4 (CH3, NbNtBu, Cβ), 29−27 (CH,
CHMe2 of CNAr), 25−22 (CH3, CHMe2 of CNAr). 13C NMR
(125.8 MHz, C6D5Cl, 293 K) δ(ppm) 84.2 (d, CH, C6H6, 1JCH = 184.5
Hz). 19F NMR (470.4 MHz, C6D5Cl, 293 K) δ(ppm) −62.7 (PhCF3),
−133.8, −166.1, and −169.6 (B(C6F5)4). Anal. calcd for
C131H120B2F43N6Nb2: C, 56.50; H, 3.69; N, 3.02. Found: C, 56.63;
H, 3.81; N, 2.78. mp: 204.9−206.2 °C.
spectroscopies to the previously reported complex 4 (BDI)Nb(NtBu)-
(CNXyl)3.32
Reaction with N2O to Form [(BDI)Nb(NtBu)(μ2-O)]2 (5).
Addition of 1 atm of N2O to a J-Young tube containing 2a (11.8
mg, 17.2 mmol) in C6D6 or 2b (13.2 mg, 19.6 mmol) in C7D8 resulted
in a color change from dark red to bright orange within 20 or 5 min,
respectively with yellow microcrystalline material rapidly crystallizing
1
out of the solution. This yellow material has been assigned by H
NMR spectroscopy in both cases to the previously synthesized
[(BDI)Nb(NtBu)(μ2-O)]2 complex 4.32
Formation of (N(Ar)C(CH3)CHC(CH3))(THF)Nb(NtBu)(NAr) (6).
Heating a solution of 2a (11.2 mg, 17.0 mmol) or 2b (11.9 mg, 17.8
mmol) in THF-d8 to 70 °C for 1 h resulted in the formation of 6
(confirmed by 1H NMR spectroscopy), which was previously
synthesized via a different route.32
Kinetic Studies. The kinetics of the reaction that transformed 2a
to 7a were followed by 1H NMR spectroscopy in a J-Young tube with
a 0.07 M solution of 2a in C6D12 at different temperatures and in
presence of various concentrations of benzene. The rate constants
determined by the disappearance of starting material and appearance
of product were identical. The mass balance was conserved through
the course of each experiment. When the C6D12 was stored over
molecular sieves, the kinetics data were inconsistent and not
reproducible. This was attributed to the acidity of the sieves,80
which may lead to the formation of side products. Reproducible kinetic
data were obtained either by using NMR solvents stored in the
absence of molecular sieves or by stirring the solvent over basic
alumina and filtering prior to use. Complex 2a was added to a J-Young
tube (22.1−23.3 mg, 0.032−0.033 mmol), followed by 0.4 mL of
C6H6/C6D12 stock solution containing 4.92 mM internal standard
(tetrakis(trimethylsilyl)silane). The concentration of each species was
calculated by integrating the methyl peaks of the BDI backbone (1.77
ppm for complex 2a and 1.69 ppm for complex 7a) relative to the
internal standard. All data used for the Eyring plot were obtained using
the same C6D12-C6H6-internal standard stock solution to minimize
experimental error (4.92 mM internal standard and 0.593 M C6H6).
Temperatures inside the probe were determined as a function of the
frequency difference (in Hz) between the two signals observed for a
sample containing 100% ethylene glycol.81
[[(BDI#)Nb(NtBu)][(BDI)Nb(NtBu)](μ-η6:η6C6H6)][B(C6F5)4] (8).
Et2O (20 mL) was added to a 50 mL flask containing the complex
[(BDI)Nb(NtBu)]2(μ-η6:η6-C6H6) (7a) (0.180 g, 0.136 mmol, 1.0
equiv), and the red solution was cooled to −80 °C. Ten mL of a
diethylether solution of [(H(OEt2)][B(C6F5)4] (97 mg, 0.143 mmol,
1.05 equiv) was added to the red solution, and the mixture was stirred
at −80 °C for 20 min, warmed to room temperature and stirred
overnight. The purple microcrystalline material that formed was
separated by filtration, washed with Et2O (2 × 20 mL) and extracted
with DCE (10 mL). The resulting purple solution was cooled at −15
°C to yield dark purple crystals (150 mg, 61%). X-ray suitable crystals
1
were obtained by recrystallization from DCE. H NMR (500 MHz,
C6D5Cl, 293 K): δ (ppm) 7.16−6.87 (multiple peaks, broad, Ar +
C6D5Cl), 5.05 (s, 1 H, HC(C(Me)NAr)2), 4.99 (d, 1 H, CHH′(C-
2
(Me)NAr)2), JHH = 16.3 Hz), 3.79 (d, 1 H, CHH′(C(Me)NAr)2),
2JHH = 16.2 Hz), 3.70 (sept, 2 H, CHMe2, 3JHH = 7.0 Hz), 3.47 (sept, 2
3
3
H, CHMe2, JHH = 7.0 Hz), 2.75 (sept, 2 H, CHMe2, JHH = 7.0 Hz),
2.40 (s, 6 H, C6H6), 1.90 (sept, 2 H, CHMe2, 3JHH = 7.0 Hz), 1.80 (s, 6
H, CHH′(C(Me)NAr)2), 1.63 (s, 6 H, CH(C(Me)NAr)2), 1.12 (d, 6
3
t
H, CHMe2, JHH = 6.8 Hz), 1.08 (s, 9 H, Bu), 1.01 (d, 6 H, CHMe2,
3JHH = 6.4 Hz), 1.00 (d, 6 H, CHMe2, JHH = 6.8 Hz), 0.95 (d, 6 H,
3
CHMe2, 3JHH = 7.0 Hz), 0.92 (d, 6 H, CHMe2, 3JHH = 6.4 Hz), 0.88 (d,
SIR Experiments.82,83 The kinetics of the intramolecular
rearrangement reactions was studied at six different temperatures
within the range of 235−265 K. Temperatures were calculated using a
100% methanol sample by measuring the separation between two
peaks in Hz.81 For complex 7b, the SIR experiment was performed by
saturating the resonance at 2.55 ppm, then following the exchange
with the resonance at 2.95 ppm, corresponding to the CH group of the
isopropyl moiety. These resonances were unambiguously assigned
using a combination of COSY, TOCSY, NOESY, HSQC, and HMBC
experiments. Unfortunately, the latter resonance overlaps with the
peaks of the meta and para protons of the bridging toluene. This was
corrected by subtracting 3/5 of the value for each data point, since the
signal of interest integrates for 2 H atoms and the signals for the meta
and para hydrogens of toluene integrate to 3 H atoms.
3
3
6 H, CHMe2, JHH = 6.4 Hz), 0.85 (d, 6 H, CHMe2, JHH = 6.4 Hz),
0.84 (s, 9 H, Bu), 0.74 (d, 6 H, CHMe2, JHH = 6.4 Hz). 13C{1H}
NMR (125.8 MHz, C6D5Cl, 293 K) δ(ppm) 167.5 (C, HC(C(Me)-
NAr)2), 164.9 (C, HC(C(Me)NAr)2), 151.66 (C, Ar), 150.5 (C, Ar),
147.3 (C, Ar), 143.23 (C, Ar), 140.03 (CH, Ar), 138.5 (CH, Ar),
126.25 (CH, Ar), 124.42 (CH, Ar), 102.98 (CH, HC(C(Me)NAr)2),
80.6 (CH, C6H6), 68.8 (CH2, H2C(C(Me)NAr)2), 32.9 (CH3, Nb
NtBu, Cβ), 31.4 (CH3, NbNtBu, Cβ), 29−27 (CH, CHMe2 of C
NAr), 25−22 (CH3, CHMe2 of CNAr). 13C NMR (125.8 MHz,
t
3
C6D5Cl, 293 K) δ(ppm) 80.6 (d, CH, C6H6, JCH = 178.9 Hz). 19F
1
NMR (470.4 MHz, C6D5Cl, 293 K) δ(ppm) −133.8, −166.1, and
−169.6 (B(C6F5)4). Anal. calcd for C96H107N6Nb2B1F20: C, 60.01; H,
5.61; N, 4.37. Found: C, 59.76; H, 5.56; N, 4.30. mp: 215.4−217.5 °C.
[[(BDI#)Nb(NtBu)]2(μ-η6:η6-C6H6)][B(C6F5)4]2 (9). Chlorobenzene
(20 mL) was added to a 50 mL flask containing the complex
[[(BDI#)Nb(NtBu)][(BDI)Nb(NtBu)](μ-η6:η6-C6H6)][B(C6F5)4]
(8) (0.180 g, 0.09 mmol, 1 equiv). To the dark-purple solution was
added 10 mL a chlorobenzene solution of [(H(OEt2)][B(C6F5)4]
(168 mg, 0.20 mmol, 2.1 equiv) at room temperature, and the mixture
was stirred overnight, resulting in a clear orange solution. The volatile
materials were removed under reduced pressure to yield an orange
powder, which was washed with Et2O (2 × 20 mL) and extracted with
10 mL α,α,α-trifluorotoluene. Et2O (20 mL) was layered over the
α,α,α-trifluorotoluene solution at room temperature, yielding bright-
orange X-ray suitable crystals within 2 days, which were filtered, and
Crystallographic Analysis. X-ray structural determinations were
performed on a Bruker SMART 1000 or SMART APEX
diffractometer. Both are 3-circle diffractometers that couple a CCD
84
detector with a sealed-tube source of monochromated Mo Ka
radiation (λ = 0.71073 Å). A crystal of appropriate size was coated in
Paratone-N oil and mounted on a Kapton loop. The loop was
transferred to the diffractometer, centered in the beam, and cooled by
a nitrogen flow low-temperature apparatus that had been previously
calibrated by a thermocouple placed at the same position as the crystal.
Preliminary orientation matrices and cell constants were determined
by collection of 60 10 s frames, followed by spot integration and least-
3233
dx.doi.org/10.1021/ja311966h | J. Am. Chem. Soc. 2013, 135, 3224−3236

Journal of the American Chemical Society
Article
squares refinement. The reported cell dimensions were calculated from
all reflections with I > 10 σ. The data were corrected for Lorentz and
polarization effects; no correction for crystal decay was applied. An
empirical absorption correction based on comparison of redundant
and equivalent reflections was applied using SADABS.85 All software
used for diffraction data processing and crystal structure solution and
refinement are contained in the APEX2 program suite (Bruker AXS,
Madison, WI).86 Thermal parameters for all nonhydrogen atoms were
refined anisotropically. For all structures, R1 = ∑(|Fo| − |Fc|)/∑(|Fo|);
AUTHOR INFORMATION
Corresponding Author
john.arnold@berkeley.edu
Notes
■
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the AFOSR (grant no. FA9550-11-1-0008) for
financial support, and Prof. Richard Andersen, Dr. Odile
Eisenstein, Dr. Nikolas Kaltsozannis, Dr. Henry S. La Pierre,
Benjamin M. Kriegel, and Mark Abubekerov for helpful
discussions. T.L.G. is grateful to the UCB Department of
Chemistry for the Howard W. Crandall Fellowship. The X-ray
absorption work was supported by the Division of Chemical
Sciences, Geosciences, and Biosciences, Office of Basic Energy
Sciences, U.S. Department of Energy at LBNL (contract DE-
AC02-05CH11231) and under the Heavy Element Chemistry
Program at LANL (operated by Los Alamos National Security,
LLC, for the National Nuclear Security Administration;
contract DE-AC52-06NA25396). The X-ray absorption work
was also supported at LANL by Glenn T. Seaborg Institute
Postdoctoral Fellowships (S.G.M.), and at the ALS by the
Director, Office of Science, Office of Basic Energy Sciences, of
the U.S. Department of Energy under contract no. DE-AC02-
05CH11231.
wR2 = [∑{w(Fo − Fc2)2}/∑{w(Fo )2}]1/2. Thermal ellipsoid plots
were created using the ORTEP-3 software package and POV ray.87
Nb L3,2-Edge XANES Measurements. Sample preparation
methodology was similar to that utilized previously.88,89 In an argon-
filled glovebox, each sample was ground in a mortar and pestle, and the
particles were transferred to a Si3N4 window (100 nm, Silson). A
second window was placed over the sample to sandwich the particles,
and the windows were sealed together using epoxy. Single-energy
images and Nb L3,2-edge XANES spectra (Figure S21) were acquired
using the STXM instrument at the Advanced Light Source beamline
5.3.2.1, which is operated in top-off mode at 500 mA, in a ∼0.75 atm
He-filled chamber.90 This bending magnet beamline performs well
with X-ray photons up to 2500 eV, which includes the Nb L3- and L2-
edges (approximately 2370 and 2465 eV, respectively). Prior to each
measurement, an energy calibration was performed at the Si K-edge for
Si (1838.90 eV). For these measurements, the X-ray beam was focused
with a zone plate onto the sample, and the transmitted light was
detected. All measurements were performed using a 25 nm zone plate,
and dwell times were ∼10 ms per pixel. To quantify the absorbance
signal, the measured transmitted intensity (I) was converted to optical
density using Beer−Lambert’s law: OD = ln(I/I0) = μρd, where I0 is
the incident photon flux intensity, d is the sample thickness, and μ and
ρ are the mass absorption coefficients and density of the sample
material, respectively. Incident beam intensity was measured through
the sample-free region of the Si3N4 windows. Regions of particles with
an absorption of >1.5 OD were omitted to ensure the spectra were in
the linear regime of the Beer−Lambert law. The energy resolution
(fwhm) was estimated at 1.0 eV.90 During the STXM experiment,
particles showed no signs of radiation damage, and each spectrum was
reproduced from multiple independent particles. Positions of the L3-
and L2-edge peak maxima were determined from the first derivative of
the experimental spectra.
2
2
REFERENCES
■
(1) Muertterties, E. L.; Bleek, J. R.; Wucherer, E. J.; Albright, T. A.
Chem. Rev. 1982, 82, 499.
(2) Cassani, M. C.; Gun’ko, Y. K.; Hitchcock, P. B.; Lappert, M. F.;
Laschi, F. Organometallics 1999, 18, 5539.
(3) Bochkarev, M. N. Chem. Rev. 2002, 102, 2089.
(4) Duff, A. W.; Jonas, K.; Goddard, R.; Kraus, H. J.; Krueger, C. J.
Am. Chem. Soc. 1983, 105, 5479.
(5) Krieck, S.; Gorls, H.; Yu, L.; Reiher, M.; Westerhausen, M. J. Am.
̈
Chem. Soc. 2009, 131, 2977.
(6) Cotton, F. A.; Kibala, P. A; Wojtczak, W. A. J. Am. Chem. Soc.
1991, 113, 1462.
(7) Troyanov, S. I.; Meetsma, A.; Teuben, J. H. Inorg. Chim. Acta
1998, 271, 180.
In order to evaluate d orbital occupancies from the Nb L3,2 edge
spectra, white line intensities must be isolated from the postedge
regions beyond the absorption thresholds and normalized. This was
achieved using IGOR 6.0 through a modification to an empirical
procedure that was described in detail by Pearson and co-workers.91
First, a straight line was fit to the pre-edge region (<2350 eV) and
subtracted from the experimental data to eliminate the background of
the spectrum. Postedge spectral intensity that was not a component of
the L3,2 white lines was modeled by fitting two additional lines to
approximately 50 eV regions after the L3 and L2 absorption thresholds.
The lines were extrapolated to span the full spectrum, and the
integrated area under the lines at energies above the L3 or L2 edge
maxima was subtracted from the total L3 and L2 edge intensities.
Continuum intensity was determined by integrating over a 55 eV
region of the spectrum beginning 30 eV above the energy of the L3-
edge maximum.92 Normalized Nb L3,2-edge intensities were calculated
by dividing the total L3,2-edge intensity by the continuum intensity.
(8) Tsai, Y.-C.; Wang, P.-Y.; Chen, S.-A.; Chen, J.-M. J. Am. Chem.
Soc. 2007, 129, 8066.
(9) Nikiforov, G. B.; Crewdson, P.; Gambarotta, S.; Korobkov, I.;
Budzelaar, P. H. M. Organometallics 2007, 26, 48.
(10) Tsai, Y.-C.; Wang, P.-Y.; Lin, K.-M.; Chen, S.-A.; Chen, J.-M.
Chem. Commun. 2008, 205.
(11) Ni, C.; Ellis, B. D.; Fettinger, J. C.; Long, G. J.; Power, P. P.
Chem. Commun. 2008, 1014.
(12) Ben
Chem., Int. Ed. 2008, 47, 5597.
́ ́
ard, M.; Rohmer, M. M.; Lopez, X.; Theopold, K. H. Angew.
(13) Fryzuk, M. D.; Love, J. B.; Rettig, S. J. J. Am. Chem. Soc. 1997,
119, 9071.
(14) Diaconescu, P. L.; Arnold, P. L.; Baker, T. A.; Mindiola, D. J.;
Cummins, C. C. J. Am. Chem. Soc. 2000, 122, 6108.
(15) Diaconescu, P. L.; Cummins, C. C. J. Am. Chem. Soc. 2002, 124,
7660.
(16) Evans, W. J.; Kozimor, S. A.; Ziller, J. W.; Kaltsoyannis, N. J. Am.
Chem. Soc. 2004, 126, 14533.
(17) Evans, W. J.; Traina, C. A.; Ziller, J. W. J. Am. Chem. Soc. 2009,
131, 17473.
(18) Mills, D. P.; Moro, F.; McMaster, J.; van Slageren, J.; Lewis, W.;
Blake, A. J.; Liddle, S. T. Nat. Chem 2011, 3, 454.
(19) Huang, W.; Khan, S. I.; Diaconescu, P. L. J. Am. Chem. Soc.
2011, 133, 10410.
ASSOCIATED CONTENT
■
S
* Supporting Information
Details concerning NMR spectroscopy, X-ray structures,
electronic spectroscopy, XANES and DFT calculations. This
information is available free of charge via the Internet at http://
pubs.acs.org.
(20) Monreal, M. J.; Khan, S. I.; Kiplinger, J. L.; Diaconescu, P. L.
Chem. Commun. 2011, 47, 9119.
3234
dx.doi.org/10.1021/ja311966h | J. Am. Chem. Soc. 2013, 135, 3224−3236

Journal of the American Chemical Society
Article
(21) Patel, D.; Moro, F.; McMaster, J.; Lewis, W.; Blake, A. J.; Liddle,
S. T. Angew. Chem., Int. Ed. 2011, 50, 10388.
(22) Diaconescu, P. L.; Cummins, C. C. Inorg. Chem. 2012, 51, 2902.
(23) Arnold, P. L.; Mansell, S. M.; Maron, L.; McKay, D. Nat. Chem.
2012, 4, 668.
(24) Mougel, V.; Camp, C.; Pecaut, J.; Coperet, C.; Laurent Maron,
́ ́
L.; Kefalidis, C. E; Mazzanti, M. Angew. Chem., Int. Ed. 2012, 51,
12280.
(56) Calderazzo, F.; De Benedetto, G. E; Pampaloni, G.; Rocchi, L.;
Englert, U. J. Organomet. Chem. 1993, 462, C10.
(57) Calderazzo, F.; Pampaloni, G.; Rocchi, L.; Englert, U.
Organometallics 1994, 13, 252.
(58) Clark, D. L.; Gordon, J. C.; McFarlan, J. T.; Vincent-Hollis, R.
L.; Watkin, J. G.; Zwick, B. D. Ingor. Chem. Acta. 1996, 244, 269.
(59) Tayebani, M.; Feghali, K.; Gambarotta, S.; Yap, G. Organo-
metallics 1998, 17, 4282.
(25) Minasian, S. G.; Krinsky, J. L.; Arnold, J. Chem.Eur. J. 2011,
(60) Kesanli, B.; Fettinger, J.; Eichhorn, B. Angew. Chem., Int. Ed.
2001, 40, 2300.
17, 12234.
(26) Gianetti, T. L.; Tomson, N. C.; Arnold, J.; Bergman, R. G. J. Am.
Chem. Soc. 2011, 133, 14904.
(27) Fei, P.; Khan, M. A.; Nicholas, L. M. J. Am. Chem. Soc. 1992,
(61) Fryzuk, M. D.; Kozak, C. M.; Mehrkhodavandi, P.; Morello, L.;
Patrick, B. O.; Rettig, S. J. J. Am. Chem. Soc. 2002, 124, 516.
(62) Sussman, V. J.; Ellis, J. E. Angew. Chem., Int. Ed. 2008, 47, 484.
(63) Plíva, J.; Johns, J. W. C.; Goodman, L. J. Mol. Spectrosc 1990,
140, 214.
(64) Landolt-Bornstein: Numerical Data and Functional Relationships
in Science and Technology; Madelung, O., Ed.; Springer: Berlin, 1987;
Vol. 15.
114, 6579.
(28) Fryzuk, M. D.; Kozak, C. M.; Bowdridge, M. R.; Patrick, B. O.;
Rettig, S. J. J. Am. Chem. Soc. 2002, 124, 8389.
(29) Kilgore, U. J.; Yang, X.; Tomaszewski, J.; Huffman, J. C.; Daniel
J. Mindiola, D. J. Inorg. Chem. 2006, 45, 10712.
(30) Figueroa, J. S.; Cummins, C. C. Dalton Trans. 2006, 2161.
(31) Hiromasa Tanaka, H.; Shiota, Y.; Matsuo, T.; Kawaguchi, H.;
Yoshizawa, K. Inorg. Chem. 2009, 48, 3875.
́
(65) Maksic, J. B. Int. J. Quantum Chem. 1971, 5, 301.
(66) Bodner, G. M.; Todd, L. J. Inorg Chem 1974, 13, 360.
(67) The Organometallic Chemistry of the Transition Metals, 4th ed.;
Crabtree, R. H., Ed.; John Wiley & Sons, Inc.: Hoboken NJ, 2005, p
87.
(32) Tomson, N. C.; Arnold, J.; Bergman, R. G. Organometallics
2010, 29, 5010.
(33) Steffey, B. D.; Chesnut, R. W.; Kerschner, J. L.; Pellechia, P. J.;
Fanwick, P. E.; Rothwell, I. P. J. Am. Chem. Soc. 1989, 111, 378.
(34) Labinger, J. A.; Schwartz, J. J. Am. Chem. Soc. 1975, 97, 1596.
(35) Roskamp, E. J.; Pedersen, S. F. J. Am. Chem. Soc. 1987, 109,
6551.
(68) Da Re, R. E; Kuehl, C. J.; Brown, M. G; Rocha, R. C.; Bauer, E.
D.; John, K. D.; Morris, D. E; Shreve, A. P.; Sarrao, J. L. Inorg. Chem.
2003, 42, 5551.
(69) Pearson, D. H.; Ahn, C. C.; Fultz, B. Phys. Rev. B 1993, 47, 8471.
(70) Bach, D.; Schneider, R.; Gerthsen, D.; Verbeeck, J.; Sigle, W.
Microsc. Microanal. 2009, 15, 505.
(36) Etienne, M.; Biasotto, F.; Mathieu, R.; Templeton, J. L.
(71) Pearson, D. H.; Fultz, B.; Ahn, C. C. Appl. Phys. Lett. 1988, 53,
Organometallics 1996, 15, 1106.
(37) Antinolo, A.; Otero, A. Organometallics 1997, 16, 2601.
(38) Obora, Y.; Kimura, M.; Ohtake, T.; Tokunaga, M.; Tsuji, Y.
Organometallics 2006, 25, 2097.
1405.
(72) Olszta, M. J.; Wang, J.; Dickey, E. C. J. Microsc. (Oxford, U. K.)
2006, 224, 233.
(39) Rothwell, I. P. Chem. Commun. 1997, 1331.
(40) Yu, J. S.; Fanwick, P. E.; Rothwell, I. P. J. Am. Chem. Soc. 1990,
112, 8171.
(41) Yu, J. S.; Felter, L.; Potyen, M. C.; Clark, J. R.; Visciglio, V. M.;
Fanwick, P. E.; Rothwell, I. P. Organometallics 1996, 15, 4443.
(42) Veige, A. S.; Slaughter, L. M.; Lobkovsky, E. B.; Wolczanski, P.
T.; Matsunaga, N.; Decker, S. A.; Cundari, T. R. Inorg. Chem. 2003, 42,
6204.
(43) Figueroa, J. S.; Piro, N. A.; Mindiola, D. J.; Fickes, M. G.;
Cummins, C. C. Organometallics 2010, 29, 5215.
(44) Tomson, N. C.; Yan, A.; Arnold, J.; Bergman, R. G. J. Am. Chem.
Soc. 2008, 130, 11262.
(45) A diniobium bridged benzene was reported that exhibits a μ-
η2:η2 structure: Veige, A. S.; Kleckley, T. S.; Chamberlin, R. M;
Neithamer, D. R; Lee, C. E.; Wolczanski, P. T.; Lobkovsky, E. B.;
Glassey, W. V. J. Organomet. Chem. 1999, 591, 194.
(46) Tomson, N. C.; Arnold, J.; Bergman, R. G. Organometallics
2010, 29, 2926.
(73) Bach, D.; Stoermer, H.; Schneider, R.; Gerthsen, D.; Verbeeck,
J. Microsc. Microanal. 2006, 12, 416.
(74) Bach, D.; Schneider, R.; Gerthsen, D. Microsc. Microanal. 2009,
15, 524.
(75) Tao, R.; Todorovic, R.; Liu, J.; Meyer, R. J.; Arnold, A.; Walkosz,
W.; Zapol, P.; Romanenko, A.; Cooley, L. D.; Klie, R. F. J. Appl. Phys.
2011, 110, 124313.
(76) Solomon, E. I.; Lever, A. B. P. Inorganic Electronic Structure and
Spectroscopy; John Wiley & Sons: Hoboken, NJ, 2006; Vol. 2.
(77) Alaimo, P. J.; Peters, D. W.; Arnold, J.; Bergman, R. G. J. Chem.
Educ. 2001, 78, 64.
(78) Feldman, J.; McLain, S. J.; Parthasarathy, A.; Marshall, W. J.;
Calabrese, J. C.; Arthur, S. D. Organometallics 1997, 16, 1514.
(79) Budzelaar, P. H. M.; van Oort, A. B.; Orpen, A. G. Eur. J. Inorg.
Chem. 1998, 10, 1485.
(80) Anderson, L. L.; Arnold, J.; Bergman, R. G. J. Am. Chem. Soc.
2005, 127, 14541.
(81) Amman, C.; Meier, P.; Merbach, A. E. J. Magn. Reson. 1982, 46,
319.
(47) Churchill, M. R.; Chang, S. W. -Y J. Chem. Soc., Chem. Commun.
(82) Bain, A. D.; Cramer, J. A. J. Magn. Reson. A 1996, 118, 21.
(83) Mugridge, J. S.; Szigethy, G.; Bergman, R. G; Raymond, K. N J.
Am. Chem. Soc. 2010, 132, 16256.
1974, 248.
(48) Goldberg, S. Z.; Spivack, B. D.; Stanley, G.; Eisenberg, R.;
Braitsch, D. M.; Miller, J. S.; Abkowitz, M. J. Am. Chem. Soc. 1977, 99,
110.
(84) SMART: Area-Detector Software Package; Brucker Analytic X-ray
Systems: Madison, WI, 2001−2003.
(49) Stollmaier, F.; Thewalt, U. J. Organomet. Chem. 1981, 222, 227.
(50) Thewalt, U.; Stollmaier, F. Angew. Chem., Int. Ed. 1982, 21, 133.
(51) Bandy, J. A.; Prout, K.; Cloke, F. G. N.; de Lemos, H. C.; Wallis,
J. M. J. Chem. Soc., Dalton Trans. 1988, 1475.
(52) Green, M. L. H.; O’Hare, D.; Watkin, J. G. J. Chem. Soc., Chem.
Commun. 1989, 698.
(53) Calderazzo, F.; Pampaloni, G.; Rocchi, L.; Strahle, J.; Wurst, K.
J. Organomet. Chem. 1991, 413, 91.
(54) Calderazzo, F.; Englert, U.; Pampaloni, G.; Rocchi, L. Angew.
Chem., Int. Ed. 1992, 31, 1235.
(55) Calderazzo, F.; Gingl, F.; Pampaloni, G.; Rocchi, L.; Strahle, J.
Chem. Ber. 1992, 125, 1005.
(85) SADABS: Bruker-Nonius Area Detector Scaling and Absorption,
V2.05; Brucker Analytic X-ray Systems: Madison, WI, 2003.
(86) Sheldrick, G. M. Acta. Crystallogr. A 2008, 64, 112.
(87) Farrugia, L. J. J. Appl. Crystallogr. 1997, 30, 565 Ed..
(88) Bradley, J. A.; Yang, P.; Batista, E. R.; Boland, K. S.; Burns, C. J.;
Clark, D. L.; Conradson, S. D.; Kozimor, S. A.; Martin, R. L.; Seidler,
G. T.; Scott, B. L.; Shuh, D. K.; Tyliszczak, T.; Wilkerson, M. P.;
Wolfsberg, L. E. J. Am. Chem. Soc. 2010, 132, 13914.
(89) Minasian, S. G.; Krinsky, J. L.; Rinehart, J. D.; Copping, R.;
Tyliszczak, T.; Janousch, M.; Shuh, D. K.; Arnold, J. J. Am. Chem. Soc.
2009, 131, 13767.
3235
dx.doi.org/10.1021/ja311966h | J. Am. Chem. Soc. 2013, 135, 3224−3236

Journal of the American Chemical Society
Article
(90) Kilcoyne, A. L. D.; Ade, H.; Attwood, D.; Hitchcock, A. P.;
McKean, P.; Mitchell, G. E.; Monteiro, P.; Tyliszczak, T.; Warwick, T.
AIP Conf. Proc. 2010, 1234, 459.
(91) Pearson, D. H.; Ahn, C. C.; Fultz, B. Phys. Rev. B 1993, 47, 8471.
(92) Bach, D.; Schneider, R.; Gerthsen, D.; Verbeeck, J.; Sigle, W.
Microsc. Microanal. 2009, 15, 505.
3236
dx.doi.org/10.1021/ja311966h | J. Am. Chem. Soc. 2013, 135, 3224−3236