Unusually stable tungstenacyclobutadienes featuring an ONO trianionic pincer-type ligand

Article abstract of DOI:10.1039/c2dt32653a

This report presents the synthesis of the first neutral trianionic ONO pincer-type tungsten alkylidyne complex, [CF₃-ONO]WC( ^tBu)(OEt₂) (5) {where CF₃-ONO = (MeC ₆H₃[C(CF₃)₂

Full text of DOI:10.1039/c2dt32653a

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Unusually stable tungstenacyclobutadienes featuring
an ONO trianionic pincer-type ligand†
Cite this: Dalton Trans., 2013, 42, 3326
Matthew E. O’Reilly, Ion Ghiviriga, Khalil A. Abboud and Adam S. Veige*
This report presents the synthesis of the first neutral trianionic ONO pincer-type tungsten alkylidyne
complex, [CF3–ONO]WuC(tBu)(OEt2) (5) {where CF3–ONO = (MeC6H3[C(CF3)2O])2N3−}. Treating 5 with
1-phenylpropyne, 4,4-dimethyl-2-pentyne, and cyclooctyne yields the corresponding tungstenacyclobuta-
diene complexes [CF3–ONO]W[κ2-C(tBu)C(Me)C(Ph)] (6), [CF3–ONO]W[κ2-C(tBu)C(Me)C(tBu)] (7), and
[CF3–ONO]W[κ2-C(tBu)C(CH2)6C] (8). Complexes 6, 7, and 8 do not undergo retro-[2 + 2]-cycloaddition
even at 200 °C or in the presence of PMe3. DFT methods to elucidate the electronic structure of com-
plexes 5 and 6 reveal important electronic factors that contribute to the lack of reactivity for the tungsten-
acyclobutadienes. An important bonding combination between the pincer N-atom lone pair and the
WuC bond within 5, termed an inorganic enamine, provides an explanation for the lack of retro-[2 + 2]-
cycloaddition from 6, 7, and 8. 15N NMR spectroscopy was used to confirm the computational finding of
an inorganic enamine bonding combination. Single crystal X-ray analysis of 5, 6, 7, and 8 provides insight
into possible steric inadequacies within the CF3–ONO3− ligand to promote catalytic metathesis.
Received 6th November 2012,
Accepted 18th December 2012
DOI: 10.1039/c2dt32653a
www.rsc.org/dalton
Introduction
Transition metal-catalyzed alkyne cross metathesis (ACM),1–12
nitrile-alkyne cross metathesis (NACM),1,13–20 ring-closing
alkyne metathesis (RCAM),21–24 ring-opening alkyne meta-
thesis polymerization (ROAMP),25–28 and acyclic diyne meta-
thesis (ADIMET)29–31 continue to attract considerable interest
as useful catalytic methods for creating CuC triple bonds.
Work still remains to create highly active catalysts that tolerate
a broad scope of substrates comparable to their alkene meta-
thesis counterparts.
Jia and Lin32 examined alkyne metathesis computationally
and some striking results emerged regarding the use of amido
vs. alkoxide vs. Mo(VI) vs. W(VI) catalysts on the rates of metath-
esis. Dissecting the transition state for catalyst (MeO)3WuC
(Me), the authors determined that ligand deformation (trigo-
nal → T-shape; Fig. 1) accounted for ∼24 kcal mol−1 of the
relative energy difference during transition, while alkyne
Fig. 1 Ancillary ligand rearrangement during alkyne metathesis.
reorganization energy, making alkyne binding the rate-deter-
mining step.
In addition to mitigating the activation energy for the
forward and reverse [2 + 2]-cycloaddition, the relative energies
of the alkylidyne and metallacyclobutadiene intermediates
must be tailored to permit reversibility (Fig. 1). The stability of
a metallacyclobutadiene relative to the alkylidyne precursor is
binding was surprisingly exothermic by ∼16 kcal mol−1 32
.
Therefore, an ideal catalyst would remove the ligand
dependent on
contributors.33–35 Schrock and co-workers first demonstrated
that poorly basic ancillary ligands stabilize
a combination of steric and electronic
University of Florida, Department of Chemistry, Center for Catalysis, P.O. Box
117200, Gainesville, Florida 32611, USA. E-mail: veige@chem.ufl.edu;
Fax: +1 352-392-3255; Tel: +1 352-392-9844
†Electronic supplementary information (ESI) available: NMR spectra, X-ray crys-
metallacyclobutadienes.33–35 Some relevant examples are the
tungstenacyclobutadiene W(κ2-C3R3)(OR)3 with weakly basic
alkoxides –OC(CF3)2CH3,35 –OCH(CF3)2,35 and –OC6H3(iPr)2.34
In contrast, W(κ2-C3R3)(OtBu)3, bearing the strongly basic alk-
oxide –OC(CH3)3, has never been isolated. Poorly basic
tallographic data, and experimental procedures. CCDC 907724 (5), 907721 (6),
907723 (7), and 907722 (8). For ESI and crystallographic data in CIF or other elec-
tronic format see DOI: 10.1039/c2dt32653a
3326 | Dalton Trans., 2013, 42, 3326–3336
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
alkoxides also facilitate alkyne binding by creating a more elec-
trophilic metal center, thus lowering the [2 + 2]-cycloaddition
activation barrier.32 A computational study by Eisenstein et al.
provides additional support for these ligand effects in analo-
gous alkene metathesis.36
Using sterically cumbersome alkoxides arguably raises the
energy barrier for ligand rearrangement (trigonal → T-shape)
and hinders substrate access, but serves an important role in
destabilizing the corresponding metallacyclobutadiene. For
example, W(κ2-C3Me3)(OC(CF3)2CH3)3 will metathesize 20
equiv. of 3-heptyne with t1/2 < ∼2.5 min in pentane or diethyl
ether, but W(κ2-C3Me3)(OCH(CF3)2)3, with sterically smaller
Scheme 1 Isobutylene expulsion from [CF3–ONO]W(uCEt)(OtBu) upon
addition of Me3SiOTf.45
alkoxides, only slowly converts 3-heptyne in pentane (t1/2
=
21 h).35 The slower rate is attributable to a more stable WC3
intermediate caused by poor steric repulsion between the alk-
oxides and the WC3 ring. Adding donor solvents can compen-
sate and the rate increases to t1/2 = 10 min for W(C3Me3)(OCH-
(CF3)2)3 in Et2O. Interestingly, ΔS‡ for the retro-[2 + 2]-cyclo-
addition of W(κ2-C3Et3)(OCH(CF3)2)3 is consistent with an
associative transition state.35
derivative providing the first example of a bicyclic metallacyclo-
butadiene. Curiously, the tungstenacyclobutadienes are remark-
ably stable. Presented below are data and
a discussion
regarding the factors promoting their inherent stability.
Our initial approach37,38 has focused on eliminating or sig-
nificantly reducing the kinetic barrier to [2 + 2]-cycloaddition
by employing trianionic pincer ligands39–42 that are rigid and
pre-organized to adopt a T-shape. Another advantage provided
by the trianionic pincer ligand is an open/labile43,44 coordi-
nation site to allow large substrates access to the metal center.
The [CF3–ONO]3− ligand45 depicted in Fig. 2 incorporates a
push–pull12,28,46–49 electronic environment created by pairing
an electron-rich amido with fluorinated alkoxides.35 In addition
to confining the anionic donors to a T-shape, the [CF3–ONO]3−
constrains the pincer N-atom lone pair to establish the inor-
ganic equivalent45 of an enamine50,51 (Fig. 2, N–WuCR ↔
N+vWvC−R), thus providing enhanced nucleophilicity, a prop-
erty that should facilitate [2 + 2]-cycloaddition. Unfortunately,
our previous attempts to isolate a neutral W-alkylidyne bearing
the [CF3–ONO]3− ligand from the anionic propylidyne complex
A were met with synthetic challenges (Scheme 1). Owing to the
high nucleophilic character of the WuC bond, the neutral
species B self-destructs via C–H bond activation and isobutylene
expulsion, to provide the corresponding alkylidene C.45
Results and discussion
ð1Þ
Our synthetic strategy to avoid isobutylene expulsion was to
replace the propylidyne within complex A with the larger 2,2-
dimethylpropylidyne in an effort to provide enough steric hin-
drance to prevent CH bond activation. Treating [CF3–ONO]H3
(1) with (tBuO)WuCtBu52 in benzene yields complex 2, a tung-
sten alkylidene supported by a [CF3–ONO]3− pincer-type ligand
(eqn (1)). Complex 2 crystallizes from a pentane solution at
−35 °C. The 1H NMR spectrum of 2 exhibits protons at 1.15
and 1.24 ppm attributable to the WvCHC(CH3)3 and OC-
(CH3)3 protons, respectively. The WvCHtBu proton resonates
2
at 6.44 ppm with JHW = 8.80 Hz, and the corresponding
Having overcome this obstacle, we now report the successful
isolation of the neutral ONO3− trianionic pincer-type tungsten
alkylidyne complex [CF3–ONO]WuCtBu(OEt2) (5). Moreover, as
expected, complex 5 reacts rapidly with alkyne substrates
carbon signal appears at 262.6 ppm in the 13C{1H} NMR spec-
trum. The aryl rings of [CF3–ONO]3− are unable to lie coplanar
rendering the complex C1-symmetric as exemplified by two
separate Ar–CH3 proton resonances at 1.99 and 1.94 ppm. As a
further consequence of the C1-symmetry, the 19F{1H} NMR
spectrum of 2 contains four quartets, one for each –CF3 group,
at −70.7, −71.5, −73.4, and −77.3 ppm.
t
PhCuCMe, BuCuCMe, and cyclooctyne; with the cyclooctyne
ð2Þ
Treating a pentane solution of 2 with Ph3PCH2 deproto-
Fig. 2 The push–pull electronic effect of the [CF3–ONO] pincer-type ligand
(left) and the inorganic enamine bonding structure (right).
nates the alkylidene and precipitates the anionic alkylidyne,
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 3326–3336 | 3327

View Article Online
Paper
Dalton Transactions
Scheme 2
complex 3, as an analytically pure pink powder (eqn (2)). The
1H NMR spectrum of 3 contains a doublet at 2.36 ppm (2JHP
=
+
13.31 Hz) corresponding to the H3CPPh3 counter cation, and
the 31P{1H} spectrum contains a single resonance at 21.6 ppm.
The WuCtBu carbon resonates downfield in the 13C{1H} NMR
spectrum at 286.0 ppm; and again the 19F{1H} NMR spectrum
reveals the characteristic four quartets indicative of C1-sym-
metry at −68.7, −71.2, −74.4, and −76.2 ppm.
Fig. 3 Molecular structure of [CF3–ONO]W(uCtBu)(OEt2) (5) with ellipsoids
drawn at 50% probability level, with hydrogens removed for clarity. Only one of
two disordered conformation at C28 are shown for clarity.
Upon treating a benzene solution of complex 3 with excess
methyl triflate, a color change from red to deep blue occurs
over 0.5 h. Evaporating the solvent under vacuum yields a deep
blue oil, which was further purified by dissolving in a minimal
amount of benzene and adding dropwise to a cold pentane
solution to precipitate a deep blue oil of 4. Unfortunately,
some decomposition occurs, and free [CH3PPh3][OTf] cannot
be removed (Scheme 2). Nonetheless, 1H, 19F{1H}, and 2D
NMR spectra unambiguously confirm the identity of 4. The 1H
NMR spectrum displays one tBu proton resonance at 1.07 ppm
and two Ar–CH3 proton resonances at 2.10 and 2.07 ppm. The
19F{1H} NMR spectrum of 4 contains four quartets at −69.0,
−73.2, −73.9, and −76.7 ppm, and two additional singlet reso-
nances at −76.7 and −78.2 ppm corresponding to coordinated
and free –OSO2CF3 anions. The WuCtBu carbon appears at
309.4 ppm in the 13C{1H} NMR spectrum. For practical
reasons, the isolation of 4 is not necessary, since dissolving
blue 4 in Et2O yields a light blue solution of 5 and a colorless
precipitate of [CH3PPh3][OTf] (Scheme 2). The majority of the
[CH3PPh3][OTf] can be removed initially by filtration, and an
additional small amount precipitates in Et2O at −35 °C.
However, variable amounts of [CH3PPh3][OTf] always remain
during bulk scale synthesis of 5, thus thwarting combustion
analysis.
Scheme 3 Synthesis of tungstenacyclobutadienes 6, 7, and 8.
alkylidyne bond sits in the apical position and is nearly linear
(∠W1–C21–C22 is 171.2(2)°) with a W1–C21 bond length of
1.754(3) Å, consistent with other WuC bond lengths ranging
The 1H NMR spectroscopic characterization of 5 reveals one
coordinated Et2O molecule. The –CH2– protons of the coordi-
nated ether are diastereotopic, appearing as two multiplets at
3.83 and 3.64 ppm. The tBu protons resonate at 0.84 ppm, and
the two Ar–CH3 protons appear at 2.04 and 2.06 ppm. The 19F
{1H} spectrum displays the prototypical quartets at −69.2,
−71.8, −75.4, and −77.2 ppm. The signal from the WuCtBu
carbon appears downfield at 311.5 ppm in the 13C{1H} NMR
spectrum.
Blue crystals of 5 co-crystallize with residual [CH3PPh3]-
[OTf] by slowly evaporating a concentrated Et2O solution of the
mixture. Depicted in Fig. 3 is the solid state structure of 5. The
alkylidyne 5 contains a tungsten(VI) ion in a square pyramidal
geometry (τ = 0.266),53 with the [CF3–ONO]3− ligand and the
coordinated Et2O occupying the basal positions (Fig. 3). The
between
1.745–1.838
Å
for
neutral
W(VI)
alkylidynes.28,33,38,47,48,54–67 The C1-symmetry of 5 is apparent
in Fig. 3 by the underlying twist of the [CF3–ONO]3− ligand.
The nitrogen atom, N1, adopts a planar sp2 hybridized geome-
try, evidenced by a 358.8(3)° sum of angles around N1. The
vector perpendicular to the C3–N1–C13 plane representing the
nitrogen lone pair forms a ∼45° angle from parallel with the
WuC bond.
Complex 5 (or 4 generated in situ) readily reacts with excess
PhCuCCH3 to yield the tungstenacyclobutadiene complex 6
(Scheme 3). The H NMR spectrum of 6 exhibits the WC3–CH3
1
t
protons at 2.79 ppm. Protons for the Bu group (1.21 ppm)
shift downfield relative to 5 (0.84 ppm) and the WC3–C6H5
protons appear between 7.02 and 7.13 ppm. Complex 6 is
C1-symmetric yielding four quartets in the 19F{1H} NMR at
3328 | Dalton Trans., 2013, 42, 3326–3336
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
−71.5, −72.1, −76.1, and −76.5 ppm, and the 13C{1H} NMR spec-
trum contains two W–Cα resonances at 245.4 and 243.0 ppm.
Crystals were initially grown by slowly evaporating a diethyl
ether solution of 6. However, crystals more suitable for an
X-ray diffraction experiment were obtained from a slowly eva-
porating pentane solution of 6. Table 1 lists pertinent bond
length and angle data. The solid state structure of 6, presented
in Fig. 4, contains a tungsten atom in a distorted-square pyra-
midal geometry (τ = 0.097)53 with the CtBu (C21) occupying the
apical position. The W1–C21 and W1–C23 bond distances are
1.9046(16) Å and 1.9106(18) Å, respectively, which are consist-
ent with other reported WC3 rings (Fig. 5). Quite interesting is
that the two W–C bond lengths are nearly equal, with a differ-
ence of only 0.006(3) Å, whereas other reported structures
display a larger difference between the W–Cα bonds, ranging
from 0.023 to 0.113 Å (Fig. 5). The W1–N1 bond distance is
2.0158(14) Å and the vector perpendicular to the C3–N1–C13
plane representing the nitrogen lone pair forms an angle of
∼42.3° to the WC3 plane.
Table 1 Selected metric parameters for the WC3 rings of 6, 7, and 8
Bond length (Å) and angles (°)
6
7
8
W1–C21
W1–C23
C21–C22
C22–C23
1.9046(16) 1.882(3)
1.9106(18) 1.908(3)
1.911(3)
1.897(3)
1.443(4)
1.473(4)
2.023(2)
1.450(3)
1.473(2)
1.456(4)
1.453(4)
W1–N1
2.0158(14) 2.022(2)
∠N1–W1–C21
∠N1–W1–C23
∠O1–W1–O2
122.99(7)
153.60(6)
147.78(5)
122.65(10) 123.66(10)
154.02(10) 153.22(10)
147.39(8)
147.80(7)
The phenyl ring (C29–C34) attached to the WC3 ring is
∼44.7° from collinearity with the WC3 plane. This may be
attributed to steric interactions from the nearby WC3–CH3
group (C28). An interesting structural feature arising from the
constrained pincer-type geometry is the O1–W1–O2 angle of
complex 6 is tied-back resulting in an angle of 147.78(5)°,
which is significantly more acute (10°–18°) than other crystal-
lographically characterized tungstenacyclobutadienes (see
Fig. 5).34,35,48,68 Overall, the WC3 ring of complex 6 contains
similar structural features to the other reported WC3 ring moi-
eties depicted in Fig. 5.
To our surprise, complex 6 does not react further with
PhCuCCH3 to yield any cross-metathesis products, even after
heating at 200 °C, with Et2O as a free donor ligand. Despite
the seemingly similar electronic features of the [CF3–ONO]3−
ligand and structural components of the WC3 ring to reported
alkyne metathesis catalysts, complex 6 does not undergo retro-
Fig. 4 Molecular structure of [CF3–ONO]W[κ2-C(tBu)C(Me)C(Ph)] (6) with ellip-
soids drawn at 50% probability level, with hydrogens removed for clarity.
Fig. 5 Reported X-ray crystallographic bond length and angles of W(κ2-C3Et3)(ImN){OCMe(CF3)2}2 (A),48 W[κ2-C(tBu)C(Me)C(Me)]Cl3 (B),68 W(κ2-C3Et3)[O-2,6-
C6H3(iPr)2]3 (C),34 W(κ2-C3Et3)[OCH(CF3)2]3 (D),35 6, 7, and 8.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 3326–3336 | 3329

View Article Online
Paper
Dalton Transactions
t
[2 + 2]-cycloaddition. Something unique to the [CF3–ONO]3− exhibits a single resonance for both Bu groups at 1.19 ppm,
pincer-type ligand must render the tungstenacyclobutadiene affirming a fluxional process between the two conformers. The
fragment exceptionally stable. One possibility is the pincer- rapid exchange in solution produces an overall C2-symmetry as
type ligand, being overly rigid, may prevent the fluxional evidenced in the 19F{1H} NMR spectrum of 7, which exhibits
exchange between the apical and equatorial positions within only two quartets at −71.9 and −76.5 pm. Similarly, the 13C
the WC3 ring as illustrated in Fig. 6. The solid state structure {1H} NMR spectrum contains only a single resonance for both
t
of 6 contains only a single conformer, where Bu occupies the Cα in the WC3 ring at 252.8 ppm. Despite the increased steric
apical position, but does the [CF3–ONO]3− ligand allow a con- bulk within the WC3 ring compared to 6, complex 7 again does
former change in solution? Since complex 6 contains different not react with additional PhCuCMe, even at 200 °C in the
groups in position R1 and R3, it is impossible to determine if a presence of Et2O.
fluxional process occurs in solution. However, if the process is
If thermolysis does not provide the energy necessary to
fluxional and R1 = R3 = Bu, a single resonance should be cleave the WC3 ring, an internal driving force that weakens the
present in the 1H NMR spectrum.
WC3 ring may be necessary. Treating complex 5 with cyclo-
t
t
Heating complex 5 in the presence of BuCuCMe in C6H6 octyne in C6H6 provides complex 8, which contains a WC3 ring
at 60 °C for 3 h yields complex 7, where R1 and R3 contain the fused to a cyclooctene (Scheme 3). The internal ring strain
t
same Bu appendage (Scheme 3). Single crystals suitable for within cyclooctene contributes 7.4 kcal mol−1 69 directed
1
X-ray diffraction experiments grow by slowly evaporating an towards destabilizing the WC3 unit. The H NMR spectrum of
Et2O solution of 7. Table 1 lists pertinent bond lengths and 8 is indicative of a C1-symmetric complex with six aromatic
angles and Fig. 7 depicts the molecular structure. Complex 7 is resonances at 7.69 (s), 7.61 (s), 7.12 (d), 7.10 (d), 6.88 (d), and
C1-symmetric with the tungsten ion again in a distorted square 6.85 (d) ppm. The Ar–CH3 resonances appear at 2.05 and
t
pyramidal geometry (τ = 0.111).53 The refined X-ray data con- 2.01 ppm and the Bu protons resonate at 1.18 ppm. Within
tains a slight disorder (7%) in the tungsten position. A salient the fused cyclooctene structure, the two –CH2– protons adja-
t
feature in the solid state structure is that the Bu groups are cent to the WC3 ring are diastereotopic appearing as four mul-
not equivalent. Evidence includes different W–C bond lengths tiplets at 3.82, 3.66, 3.36, and 3.18 ppm. The remaining –CH2–
for W1–C21 = 1.882(3) Å and W1–C23 = 1.908(3) Å, and protons (8H) appear as several broad resonances between 0.90
different N–W–C bond angles (∠N1–W1–C21 = 122.65(10)° and and 1.55 ppm. The 19F{1H} NMR spectrum contains four quar-
∠N1–W1–C23 = 154.02(10)°). However, the asymmetry only tets at −70.9, −72.2, −76.1, and −76.6 ppm, and the 13C{1H}
exists in the solid state, because the 1H NMR spectrum of 7 spectrum exhibits two unique resonances for W–Cα atoms at
252.8 and 238.6 ppm.
Single crystals grow from slow evaporation of a concen-
trated Et2O solution of 8. Table 1 lists pertinent bond lengths
and angles and Fig. 8 depicts the molecular structure of 8. The
solid state structure consists of a distorted square pyramidal
tungsten ion. Attached to the WC3 ring is a tBu group (C24–27)
and a fused cyclooctene ring (C28–C32). The cyclooctene ring
is disordered, containing two different conformations at C30
t
Fig. 6 Fluxional WC3 ring conformations.
and C31. The Cα Bu again occupies the apical position and
Fig. 7 Molecular structure of [CF3–ONO]W[C(tBu)C(Me)C(tBu)] (7) with ellip-
soids drawn at 50% probability level, with hydrogens removed for clarity. A dis-
ordered W ion position (7%) is removed for clarity.
Fig. 8 Molecular structure of [CF3–ONO]W[C(tBu)C(CH2)6C] (8) with ellipsoids
drawn at 50% probability level, with hydrogens removed for clarity. Only one of
two disordered conformation at C30 and C31 are shown for clarity.
3330 | Dalton Trans., 2013, 42, 3326–3336
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
forms a 123.66(10)° bond angle with N1 of the pincer ligand broadening of proton resonances in the 1H NMR, but eventually
1
(N1–W1–C21) and the adjacent N1–W1–C23 bond angle is slow decomposition occurs in solution. The H NMR spectra of
153.22(10)°. The apical W1–C21 bond length of 1.911(3) Å is complexes 7 and 8 displayed no signs of reaction with PMe3.
slightly longer by 0.014(4) Å than the W–C23 bond length of The unusual stability of complexes 6–8 prompted us to investi-
1.897(3) Å. Overall the bond lengths and angles for the WC3 gate a possible electronic component to the apparent energetic
ring within complexes 6, 7, and 8 are comparable (see Table 1). disparity between 6, 7, and 8 and a putative alkylidyne.
Again, despite the strain imposed by the fused cyclooctene
Computational studies
ring, treating complex 8 with excess PhCuCMe and diethyl
ether at 200 °C in toluene-d8 does not yield metathesis pro-
ducts. One salient feature within the molecular structures of 6,
7, and 8 is the acute O1–W1–O1 bond angles of 147.78(5)°,
147.39(8)°, and 147.80(7)°, respectively. As a result of this
acute angle and the overall pincer-type architecture, the –CF3
groups are tied-back and prevented from exerting steric
pressure on the tungstenacyclobutadiene ring. In contrast are
the crystallographically characterized tungstenacyclobutadiene
of active alkyne metathesis catalysts (Fig. 5, structures A, C,
and D), where the large ancillary ligands reside over the
WC3Et3 ring. As mentioned earlier, the stability of the metalla-
cyclobutadiene complexes depends on the steric repulsion
between the ancillary ligand and the substituents attached to
the WC3 ring. Consequently, these isolated structures feature
small ethyl substituents attached to the tungstenacyclobuta-
diene ring to minimize the destabilizing steric repulsion.
Fig. 9 depicts the geometry optimized structure of both 5 and
6, as calculated by DFT methods. For complex 5, the calculated
bond lengths and angles of the tungsten ion core are in good
agreement with the crystallographically determined values. For
example, the W1–C21–C22 bond angle of 172.35° matches the
experimental value of 171.2(2)°. Similarly, the O1–W1–O2
angle of 143.64° agrees with the experimental value of 144.97(8)°.
The vector perpendicular to the C3–N1–C13 plane representing
the nitrogen lone pair is ∼42.4° from parallel with the tung-
sten alkylidyne bond (W1–C21). This angle is slightly more
acute than the experimental angle of ∼45°.
The calculated metric parameters for complex 6 are also
reasonable, although some bond angle differences deserve
mentioning. The N1–W1–C21 angle of 129.77° is 6.78° larger
than the crystallographically determined angle of 122.99(7).
Correspondingly, the adjacent angle, N1–W1–C23, of 148.39°
is smaller by 5.21° than the experimental value of 153.60(6)°.
These deviations are not uncommon and are observed for
other reported WC3 structures.47,48 A more significant devi-
ation is the O1–W1–O2 angle of 155.13° from the experimental
angle of 147.78(5)°. In agreement though, the nitrogen lone
pair represented by a vector perpendicular to the C3–N1–C13
plane forms a ∼42.8° to the WC3 ring, matching the experi-
mentally value of ∼42.3°.
t
Despite complexes 6–8 containing larger Bu substituents, the
WC3 ring is exceptionally inert. For complexes 6, 7, and 8, the
closest distances between the –CF3 of the pincer-type ligand
t
and the WC3–R groups (R = Bu, Ph, –C6H12–) are 3.54, 3.52,
and 3.53 Å, respectively. For complex A (Fig. 5), the closest
contact between the –CF3 and the –Et group is 3.11 Å.48
Considering the poor steric pressure from the [CF3–ONO]3−
ligand on the WC3R3 rings of 6, 7, and 8, perhaps adding a
stronger donor ligand (e.g. PMe3) will promote retro-[2 + 2]-cyclo-
addition, similar to adding Et2O to W(C3Me3)(OCH-(CF3)2)3.35
Treating complexes 6–8 with excess PhCuCMe and PMe3, as a
potential strong σ-donor ligand, at 100 °C did not generate any
metathesis products. Interestingly, a C6D6 solution of complex 6
Fig. 2 above relates the electronic consequence of having an
amido lone pair aligned with the WuC bond axis (inorganic
enamine). To assess this interaction in 5 and its fate upon [2 + 2]-
cycloaddition and possible role in preventing retro-[2 + 2]-
cycloaddition, Gaussian single point calculations were per-
formed on both complexes 5 and 6. Fig. 10 depicts a truncated
turns violet upon addition of PMe3, accompanied by
a
Fig. 9 DFT geometry optimized structures of 5 and 6 with calculated bond lengths (red) and crystallographic determined lengths (black).
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 3326–3336 | 3331

View Article Online
Paper
Dalton Transactions
Fig. 10 Truncated MO diagram of 5 (left) and 6 (right) (isovalues = 0.051687).
molecular orbital diagram of 5. The inorganic enamine inter- is consistent with other theoretical models.70–75 The HOMO−1
action is clearly observed in the HOMO and HOMO−2 orbitals. also shows a small amount of bonding overlap with the nitro-
The HOMO−2 contains a bonding interaction between the gen p-orbital, though it is unclear whether this interaction
WuC π-bond and the aligned N-atom p-orbital, thereby stabi- contributes much to the stability of 6. The most important
lizing the HOMO−2 relative to the unperturbed HOMO−1 by feature is the restricted orientation caused by chelation; the
0.01822 AU. The HOMO is correspondingly destabilized in the nitrogen lone pair does not π-donate into the LUMO, which
anti-bonding combination. This bonding combination is ana- would help destabilize the WC3 ring by populating a π* orbital.
logous to highly nucleophilic enamines.50,51
The drastic changes in the nitrogen bonding environment
Analysis of the electronic structure of 6 reveals that the in- from the alkylidyne complexes of 3–5 and the tungstenacyclo-
organic enamine bonding combination within 5 is lost. From butadiene complexes 6–8 are evident by 15N NMR spec-
the single point calculation of 6, the nitrogen lone pair troscopy. Table 2 lists the 15N resonances for complexes 2–8
(HOMO) is lower in energy than the HOMO of 5 by 0.0155 AU, (C6D6). The N-atom lone pair in the alkylidene complex 2 does
while the W–C π-bonds of 5 and 6 retain similar energy levels not have the appropriate orientation to overlap with the WvC
(−0.22214 and −0.22268 AU, respectively). The N-atom lone pair double bond, and as a consequence the lone pair is essentially
is essentially non-bonding. The LUMO consists of the π* inter- non-bonding.45 The 15N resonance for 2 appears downfield at
action between the dxy orbital and p-orbitals of the WC3 ring. 225.7 ppm. In contrast, the 15N resonance shifts dramatically
The HOMO−1 contains a dyz orbital in π-overlap with the two upfield to 149.3, 165.5, and 178.3 ppm upon forming the
Cα p-orbitals of the WC3 ring. This WC3 π-bonding interaction WuC triple bonds in 3, 4 and 5, respectively. Clearly, the
3332 | Dalton Trans., 2013, 42, 3326–3336
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
Table 2 15N NMR chemical shifts of 2–8
2
3
4
5
6
7
8
ppm 225.7
149.3 165.5 178.3
208.6 204.4 202.1
N-atom lone pair experiences a new environment involving
overlap with the WuC bond. This dramatic change in the 15N
NMR chemical shift provides experimental evidence for an
inorganic enamine orbital interaction. The interaction is lost
upon [2 + 2]-cycloaddition to form 6, 7, and 8. From the calcu-
lated electronic structure of 6, the N-atom lone pair is non-
bonding, and consequently the 15N resonance reverts back
downfield to 208.6 ppm. Similarly, tungstenacyclobutadienes 7
and 8 also exhibit downfield signals at 204.4 and 202.1 ppm,
respectively.
The thermodynamic irreversibility of the [2 + 2]-cyclo-
addition products 6, 7, and 8 is astonishing. As mentioned
earlier, complexes 6–8 lack significant steric repulsion between
the ancillary ligand and the WC3R3 ring. Unlike the example of
W(κ2-C3Me3)(OCH(CF3)2)3, where poor sterics were compen-
sated for by adding Et2O,35 adding a strong σ-donating PMe3
ligand was not sufficient to break apart the WC3 ring. More-
over, these results are intriguing since the [CF3–ONO]3− ligand
incorporates similar electronic features to the fluorinated alk-
oxides paired with an amido donor, utilized in Tamm’s highly
active catalysts.12,28,46–48,67 This leads us to think another
factor plays a significant part in the overly stable tungsten-
acyclobutadienes 6, 7, and 8. Fig. 11 depicts the [CF3–ONO]3−
pincer-type ligand where the amido lone pair is 45° from parallel
to the tungsten alkylidyne bond (I). For comparison, also
depicted are Tamm’s active catalysts, in which the amido
ligand either freely rotates (Fig. 11; II)47,48 or is restricted
(Fig. 11; III).47 In both II and III, the nitrogen lone pair orients
perpendicular to the tungsten–alkylidyne bond, thus avoiding
the inorganic enamine interaction. In case II where the imidazo-
lin-2-iminato ligand can freely rotate, the nitrogen lone pair
reorients 90° upon [2 + 2]-cycloaddition to lie coplanar to the
WC3 ring.47,48 In the case of III, the computational models
predict that the bulkier tert-butyl-3,5-(dimethylphenyl)-amido
ligand is unable to rotate, remaining perpendicular to the WC3
plane.47 Both II and III undergo retro-[2 + 2]-cycloaddition.
Thus, one electronic difference between I and cases II and III,
is II and III are able to avoid the unfavourable inorganic
enamine.
Fig. 11 Amido lone pair orientation for varying ligand systems of tungsten
alkylidyne and tungstenacyclobutadiene complexes (I, II,47,48 and III47).
Fig. 12 Reaction progress vs. free energy diagram for retro-[2
+ 2]-
cycloaddition.
As retro-[2 + 2]-cycloaddition occurs, the inorganic enamine sterics remain the same if not somewhat reduced, since the
interaction turns on according to Fig. 12 and is endergonic. nitrile-N-atom does not bear a substituent.
Removing the amido-alkylidyne interaction should permit
Indeed, treating complex 5 with excess acetonitrile in C6D6
facile retro-[2 + 2]-cycloaddition. One way to do this is to yields a 57 : 43 mixture of free 4,4-dimethyl-2-pentyne and 7
attempt Nitrile-Alkyne Cross Metathesis (NACM). Upon for- along with unidentifiable species as evidenced by 1H NMR
mation of an azametallacyclobutadiene, retro-[2 + 2]-cyclo- spectroscopy (eqn (3)). 4,4-Dimethyl-2-pentyne is removed by
1
addition would provide a WuN bond instead of a WuC, vacuum and the H and 19F{1H} NMR spectra unambiguously
which would not have the correct energy match to overlap with confirm the identity of 7. The unidentifiable products are pre-
the N-atom lone pair, and moreover, WuN bond formation sumably tungsten nitrido species, though their identity is not
1
should make the overall reaction exergonic. Importantly, the clear from the spectroscopic data. Broad resonances in the H
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 3326–3336 | 3333

View Article Online
Paper
Dalton Transactions
and 19F{1H} NMR spectra of the reaction mixture may rep- 5 to generate an azatungstenacyclobutadiene intermediate.
resent several multi-nuclear species; it is well-known for W- Retro-[2 + 2]-cycloaddition from the azatungstenacyclobuta-
nitrido complexes to exist as dimers,20,76–79 trimers,77–80 and diene forms a stable WuN bond, which is unable to partici-
larger oligomers.77,78 Most importantly, complex 7 and the free pate in an inorganic enamine bonding combination. The
4,4-dimethyl-2-pentyne indicate retro-[2
+ 2]-cycloaddition reaction of 5 with CH3CuN yields the metathesis product
occurred. In fact, retro-[2 + 2]-cycloaddition must occur swiftly MeCuCtBu and unidentifiable tungsten–nitride products.
because 7 is the product of alkylidyne 5 and free 4,4-dimethyl- Additional support comes from Tamm’s catalysts.28,46–48,67
t
2-pentyne. Adding excess BuCuCMe to the reaction mixture Both complexes II and III (Fig. 12) are able to avoid the inor-
does not result in any further nitrile–alkyne cross metathesis ganic enamine bonding combination, and are consequently
from the presumed tungsten–nitrido species.
active catalysts.
There is merit in employing a trianionic ONO pincer ligand
for alkyne metathesis catalysts; [2 + 2]-cycloaddition occurs
swiftly at room temperature, and the confined geometry allows
sterically cumbersome substrates access to the metal center as
evidenced by the formation of 7. However, the design of the
[CF3–ONO]3− pincer ligand underestimated the steric and elec-
tronic factors contributing to the stability of the metallacyclo-
butadiene. Current work involves designing a new ONO ligand
to destabilize the metallacyclobutadiene by increasing the peri-
pheral steric bulk and mitigating/avoiding any inorganic
enamine interactions.
ð3Þ
Conclusions
Presented is the first synthesis of a neutral trianionic pincer-
type tungsten alkylidyne (5). Complex 5 reacts with PhCuCMe,
tBuCuCMe, and cyclooctyne to yield tungstenacyclobutadiene
complexes 6, 7, and 8, respectively. Complex 8 is the first
example of a bicyclic metallacyclobutadiene, a remarkable find
considering cyclooctyne is typically polymerized as a conse-
quence of its internal ring strain.27,81,82 Tungstenacyclobuta-
dienes 6–8 contain bond angles and lengths that are similar to
known crystallographically characterized metallacyclobuta-
dienes of active metathesis catalysts. Despite the similar
metric parameters, the WC3 rings of 6, 7, and 8 do not cleave
to regenerate the alkylidyne, even at 200 °C.
Acknowledgements
This work was financially supported by UF, the Alfred P. Sloan
Foundation, and an NSF CAREER award to A.S.V.
(CHE-0748408). MEO was awarded a 2012 Eastman Chemical
Company summer fellowship. K.A.A. thanks UF and the NSF
for funds to purchase X-ray equipment (CHE-0821346).
DFT calculations elucidate the key electronic features of 5
and 6. Complex 5 exhibits a bonding combination between the
N-atom lone pair on the pincer and the WuC bond that is
best described as an inorganic enamine. Evidence for substan-
tial overlap between the N-atom lone pair and the alkylidyne
π-bond comes from 15N NMR spectroscopy. A dramatic 47 ppm
shift upfield occurs upon formation of the alkylidyne complex
5 (178.3 ppm) from the alkylidene complex 2 (225.7 ppm). In
contrast, the inorganic enamine interaction is lost upon for-
mation of the tungstenacyclobutadiene 6, and the 15N NMR
resonance shifts back downfield to 208.6 ppm.
Notes and references
1 M. Bindl, R. Stade, E. K. Heilmann, A. Picot, R. Goddard
and A. Furstner, J. Am. Chem. Soc., 2009, 131, 9468–9470.
2 L. Kloppenburg, D. Song and U. H. F. Bunz, J. Am. Chem.
Soc., 1998, 120, 7973–7974.
3 J. Sancho and R. R. Schrock, J. Mol. Catal., 1982, 15, 75–79.
4 J. H. Wengrovius, J. Sancho and R. R. Schrock, J. Am. Chem.
Soc., 1981, 103, 3932–3934.
5 F. Pennella, R. L. Banks and G. C. Bailey, Chem. Commun.,
1968, 1548–1549.
Both steric and electronic features influence the thermo-
dynamic stability of 6, 7, and 8 relative to the corresponding
alkylidyne. The chelating nature of the pincer ties back the
pendant perfluoroalkoxides, and as a consequence, the CF3–
ONO3− pincer-type ligand exerts minimal steric pressure on
the WC3 core. Surprisingly, adding excess donor ligands such
as Et2O and PMe3 and heating does not induce cycloreversion.
Evidence points to an inorganic enamine interaction as a sub-
6 R. R. Schrock, Angew. Chem., Int. Ed., 2006, 45, 3748–3759.
7 J. Heppekausen, R. Stade, A. Kondoh, G. Seidel,
R. Goddard and A. Furstner, Chem.–Eur. J., 2012, 18,
10281–10299.
8 J. Heppekausen, R. Stade, R. Goddard and A. Furstner,
J. Am. Chem. Soc., 2010, 132, 11045–11057.
9 R. R. Schrock and C. Czekelius, Adv. Synth. Catal., 2007,
349, 55–77.
stantial electronic impediment to cycloreversion. The retro- 10 W. Zhang and J. S. Moore, Adv. Synth. Catal., 2007, 349,
[2 + 2]-cycloaddition is prevented thermodynamically by re- 93–120.
establishing the inorganic enamine orbital interaction. Proof 11 A. Furstner and P. W. Davies, Chem. Commun., 2005, 2307–
to support this claim comes from the addition of CH3CuN to
2320.
3334 | Dalton Trans., 2013, 42, 3326–3336
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
12 X. A. Wu and M. Tamm, Beilstein J. Org. Chem., 2011, 7, 82– 37 S. Kuppuswamy, A. J. Peloquin, I. Ghiviriga, K. A. Abboud
93. and A. S. Veige, Organometallics, 2010, 29, 4227–4233.
13 A. M. Geyer, E. S. Wiedner, J. B. Gary, R. L. Gdula, 38 S. Sarkar, K. P. McGowan, S. Kuppuswamy, I. Ghiviriga,
N. C. Kuhlmann, M. J. A. Johnson, B. D. Dunietz and
J. W. Kampf, J. Am. Chem. Soc., 2008, 130, 8984–8999.
K. A. Abboud and A. S. Veige, J. Am. Chem. Soc., 2012, 134,
4509–4512.
14 J. P. Boyd, M. Schlangen, A. Grohmann and H. Schwarz, 39 S. Sarkar, K. P. McGowan, J. A. Culver, A. R. Carlson,
Helv. Chim. Acta, 2008, 91, 1430–1434.
J. Koller, A. J. Peloquin, M. K. Veige, K. A. Abboud and
A. S. Veige, Inorg. Chem., 2010, 49, 5143–5156.
40 T. Agapie and J. E. Bercaw, Organometallics, 2007, 26, 2957–
2959.
15 A. M. Geyer, R. L. Gdula, E. S. Wiedner and
M. J. A. Johnson, J. Am. Chem. Soc., 2007, 129, 3800–3801.
16 B. C. Bailey, A. R. Fout, H. J. Fan, J. Tomaszewski,
J. C. Huffman, J. B. Gary, M. J. A. Johnson and 41 A. Sattler and G. Parkin, J. Am. Chem. Soc., 2012, 134, 2355–
D. J. Mindiola, J. Am. Chem. Soc., 2007, 129, 2234–2235. 2366.
17 R. L. Gdula and M. J. A. Johnson, J. Am. Chem. Soc., 2006, 42 R. A. Zarkesh and A. F. Heyduk, Organometallics, 2011, 30,
128, 9614–9615. 4890–4898.
18 R. L. Gdula, M. J. A. Johnson and N. W. Ockwig, Inorg. 43 M. O’Reilly, J. M. Falkowski, V. Ramachandran, M. Pati,
Chem., 2005, 44, 9140–9142.
19 R. R. Schrock, M. L. Listemann and L. G. Sturgeoff, J. Am.
Chem. Soc., 1982, 104, 4291–4293.
20 Z. J. Tonzetich, R. R. Schrock, K. M. Wampler, B. C. Bailey,
C. C. Cummins and P. Muller, Inorg. Chem., 2008, 47,
1560–1567.
K. A. Abboud, N. S. Dalal, T. G. Gray and A. S. Veige, Inorg.
Chem., 2009, 48, 10901–10903.
44 M. E. O’Reilly, T. J. Del Castillo, J. M. Falkowski,
V. Ramachandran, M. Pati, M. C. Correia, K. A. Abboud,
N. S. Dalal, D. E. Richardson and A. S. Veige, J. Am. Chem.
Soc., 2011, 133, 13661–13673.
21 K. C. Majumdar, H. Rahaman and B. Roy, Curr. Org. Chem., 45 M. E. O’Reilly, I. Ghiviriga, K. A. Abboud and A. S. Veige,
2007, 11, 1339–1365. J. Am. Chem. Soc., 2012, 134, 11185–11195.
22 P. W. Davies, in Metathesis in Natural Product Synthesis, ed. 46 X. Wu, C. G. Daniliuc, C. G. Hrib and M. Tamm, J. Organo-
J. Cossy, S. Arseniyadis and C. Meyer, Wiley-VCH Verlag
GmbH & Co. KGaA, 2010, pp. 205–223.
23 A. Fürstner, O. Guth, A. Rumbo and G. Seidel, J. Am. Chem.
Soc., 1999, 121, 11108–11113.
met. Chem., 2011, 696, 4147–4151.
47 S. Beer, K. Brandhorst, C. G. Hrib, X. Wu, B. Haberlag,
J. Grunenberg, P. G. Jones and M. Tamm, Organometallics,
2009, 28, 1534–1545.
24 A. Fürstner and G. Seidel, Angew. Chem., Int. Ed., 1998, 37, 48 S. Beer, C. G. Hrib, P. G. Jones, K. Brandhorst, J. Grunenberg
1734–1736. and M. Tamm, Angew. Chem., Int. Ed., 2007, 46, 8890–8891.
25 D. F. Sedbrook, D. W. Paley, M. L. Steigerwald, C. Nuckolls 49 W. Zhang, S. Kraft and J. S. Moore, Chem. Commun., 2003,
and F. R. Fischer, Macromolecules, 2012, 45, 5040–5044. 832–833.
26 X. P. Zhang and G. C. Bazan, Macromolecules, 1994, 27, 50 The Chemistry of Enamines, ed. Z. Rapport, John Wiley &
4627–4628. Sons, Inc., 1994.
27 S. A. Krouse and R. R. Schrock, Macromolecules, 1989, 22, 51 Enamines: Synthesis, Structure, and Reactivity, ed.
2569–2576. A. G. Cook, Marcel Dekker, INC., New York, 1988.
28 B. Haberlag, X. A. Wu, K. Brandhorst, J. Grunenberg, 52 Z. J. Tonzetich, R. R. Schrock and P. Muller, Organometal-
C. G. Daniliuc, P. G. Jones and M. Tamm, Chem.–Eur. J.,
2010, 16, 8868–8877.
29 W. Zhang and J. S. Moore, Macromolecules, 2004, 37, 3973–
3975.
30 U. H. F. Bunz, ed. D. Astruc, Modern Arene Chemistry, Wiley-
VCH, Weinheim, 2002, pp. 217–249.
31 K. Weiss, A. Michel, E.-M. Auth, U. H. F. Bunz, T. Mangel
and K. Müllen, Angew. Chem., Int. Ed., 1997, 36, 506–509.
32 J. Zhu, G. Jia and Z. Lin, Organometallics, 2006, 25, 1812–
1819.
33 M. R. Churchill, Y. J. Li, L. Blum and R. R. Schrock, Organo-
metallics, 1984, 3, 109–113.
34 M. R. Churchill, J. W. Ziller, J. H. Freudenberger and
R. R. Schrock, Organometallics, 1984, 3, 1554–1562.
35 J. H. Freudenberger, R. R. Schrock, M. R. Churchill,
lics, 1985, 4, 74–83.
53 A. W. Addison, T. N. Rao, J. Reedijk, J. Vanrijn and
G. C. Verschoor, Dalton Trans., 1984, 1349–1356.
54 M. R. Churchill and W. J. Youngs, Inorg. Chem., 1979, 18,
2454–2458.
55 S. M. Rocklage, R. R. Schrock, M. R. Churchill and
H. J. Wasserman, Organometallics, 1982, 1, 1332–1338.
56 M. H. Chisholm, D. M. Hoffman and J. C. Huffman, Inorg.
Chem., 1983, 22, 2903–2906.
57 F. A. Cotton and W. Schwotzer, Inorg. Chem., 1983, 22, 387–
390.
58 F. A. Cotton, W. Schwotzer and E. S. Shamshoum, Organo-
metallics, 1984, 3, 1770–1771.
59 M. H. Chisholm, B. K. Conroy and J. C. Huffman, Organo-
metallics, 1986, 5, 2384–2386.
A. L. Rheingold and J. W. Ziller, Organometallics, 1984, 3, 60 M. H. Chisholm, B. W. Eichhorn, K. Folting and
1563–1573. J. C. Huffman, Organometallics, 1989, 8, 49–66.
36 X. Solans-Monfort, E. Clot, C. Coperet and O. Eisenstein, 61 M. H. Chisholm, J. C. Huffman and J. A. Klang, Polyhedron,
J. Am. Chem. Soc., 2005, 127, 14015–14025.
1990, 9, 1271–1276.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 3326–3336 | 3335

View Article Online
Paper
Dalton Transactions
62 T. P. Pollagi, S. J. Geib and M. D. Hopkins, J. Am. Chem. 73 Z. Y. Lin and M. B. Hall, Organometallics, 1994, 13, 2878–
Soc., 1994, 116, 6051–6052. 2879.
63 H. A. Brison, T. P. Pollagi, T. C. Stoner, S. J. Geib and 74 Y. Z. Huang, S. Y. Yang and X. Y. Li, J. Organomet. Chem.,
M. D. Hopkins, Chem. Commun., 1997, 1263–1264. 2004, 689, 1050–1056.
64 T. H. Warren, R. R. Schrock and W. M. Davis, J. Organomet. 75 F. Erdman and D. B. Lawson, J. Organomet. Chem., 2005,
Chem., 1998, 569, 125–137. 690, 4939–4944.
65 Z. J. Tonzetich, Y. C. Lam, P. Muller and R. R. Schrock, 76 T. P. Pollagi, J. Manna, S. J. Geib and M. D. Hopkins, Inorg.
Organometallics, 2007, 26, 475–477. Chim. Acta, 1996, 243, 177–183.
66 T. M. Gilbert and R. D. Rogers, Acta Crystallogr., Sect. C: 77 W. W. Schoeller and A. Sundermann, Inorg. Chem., 1998,
Cryst. Struct. Commun., 1993, 49, 677–680. 37, 3034–3039.
67 S. Lysenko, B. Haberlag, C. G. Daniliuc, P. G. Jones and 78 K. Dehnicke and J. Strahle, Angew. Chem., Int. Ed., 1992, 31,
M. Tamm, ChemCatChem, 2011, 3, 115–118. 955–978.
68 S. F. Pedersen, R. R. Schrock, M. R. Churchill and 79 M. H. Chisholm, K. Folting, M. L. Lynn, D. B. Tiedtke,
H. J. Wasserman, J. Am. Chem. Soc., 1982, 104, 6808–
6809.
F. Lemoigno and O. Eisenstein, Chem.–Eur. J., 1999, 5,
2318–2326.
69 P. V. Schleyer, J. E. Williams and K. R. Blanchard, J. Am. 80 M. H. Chisholm, K. Foltingstreib, D. B. Tiedtke,
Chem. Soc., 1970, 92, 2377–2386.
70 B. E. Bursten, J. Am. Chem. Soc., 1983, 105, 121–122.
F. Lemoigno and O. Eisenstein, Angew. Chem., Int. Ed.
Engl., 1995, 34, 110–112.
71 E. V. Anslyn, M. J. Brusich and W. A. Goddard, Organo- 81 S. A. Krouse, R. R. Schrock and R. E. Cohen, Macro-
metallics, 1988, 7, 98–105. molecules, 1987, 20, 903–904.
72 T. Woo, E. Folga and T. Ziegler, Organometallics, 1993, 12, 82 S. Lysenko, B. Haberlag, X. Wu and M. Tamm, Macromol.
1289–1298.
Symp., 2010, 293, 20–23.
3336 | Dalton Trans., 2013, 42, 3326–3336
This journal is © The Royal Society of Chemistry 2013