Reactivity of scandium β-diketiminate alkyl complexes with carbon dioxide

Article abstract of DOI:10.1021/om2012002

The reactions of two highly air- and moisture-sensitive scandium bis(alkyls) supported by a bulky β-diketiminato (nacnac) ligand with carbon dioxide are described. [κ²-ArNC(^tBu)CHC( ^tBu)NAr]ScR₂ (Ar = 2,6-ⁱPr₂C ₆H₃; R = CH₃, 1a; R = CH₂SiMe ₃, 1b) react rapidly with CO₂ to give mixtures of mono- and bis(carboxylato) insertion products 2a/2b and 3a/3b depending on the stoichiometry and conditions of the reaction. Compound 2a (R = CH₃) is a dimeric complex with bridging acetato groups, as determined by X-ray crystallography. These compounds were characterized by NMR spectroscopy, and 3a could be isolated in pure form. Treatment of these compounds with excess CO ₂ resulted in addition to the central carbon of the Sc(nacnac) six-membered ring and displacement of the nitrogen donors to yield dimeric scandium carboxylates 4a/4b; compound 4b was characterized by X-ray crystallography. Reactions of the nacnac scandium cations formed upon abstraction of one or two methides from 1a using B(C₆F ₅)₃ with CO₂ were also explored. Although the products were qualitatively more thermally robust, eventually ligand displacement occurred in these cationic acetato complexes as well. Nevertheless, insertion products were characterizable in solution using NMR spectroscopy. Overall, this study shows the facility with which CO₂ is taken up by scandium alkyls but that the nacnac ligand framework is too reactive to support chemistry aimed at catalytic conversion of CO₂ into other products.

Full text of DOI:10.1021/om2012002

Article
pubs.acs.org/Organometallics
Reactivity of Scandium β-Diketiminate Alkyl Complexes with
Carbon Dioxide†
Francis A. LeBlanc, Andreas Berkefeld, Warren E. Piers,* and Masood Parvez
Department of Chemistry, University of Calgary, 2500 University Dr. NW, Calgary, Alberta, Canada, T2N 1N4
S
* Supporting Information
ABSTRACT: The reactions of two highly air- and moisture-sensitive scandium bis(alkyls) supported by a bulky β-diketiminato
(nacnac) ligand with carbon dioxide are described. [κ2-ArNC(tBu)CHC(tBu)NAr]ScR2 (Ar = 2,6-iPr2C6H3; R = CH3, 1a; R =
CH2SiMe3, 1b) react rapidly with CO2 to give mixtures of mono- and bis(carboxylato) insertion products 2a/2b and 3a/3b
depending on the stoichiometry and conditions of the reaction. Compound 2a (R = CH3) is a dimeric complex with bridging
acetato groups, as determined by X-ray crystallography. These compounds were characterized by NMR spectroscopy, and 3a
could be isolated in pure form. Treatment of these compounds with excess CO2 resulted in addition to the central carbon of the
Sc(nacnac) six-membered ring and displacement of the nitrogen donors to yield dimeric scandium carboxylates 4a/4b;
compound 4b was characterized by X-ray crystallography. Reactions of the nacnac scandium cations formed upon abstraction of
one or two methides from 1a using B(C6F5)3 with CO2 were also explored. Although the products were qualitatively more
thermally robust, eventually ligand displacement occurred in these cationic acetato complexes as well. Nevertheless, insertion
products were characterizable in solution using NMR spectroscopy. Overall, this study shows the facility with which CO2 is taken
up by scandium alkyls but that the nacnac ligand framework is too reactive to support chemistry aimed at catalytic conversion of
CO2 into other products.
compound supported by a tridentate bis(phenoxide)-aryl ether
ligand15 was found to catalyze the hydrosilylation of CO2 in the
presence of the Lewis acidic cocatalyst B(C6F5)3, is intriguing.
While turnovers and activities were moderate, the B(C6F5)3/
silane reagent system16,17 was able to provide a thermodynamic
driver for the conversion of the zirconium carboxylate inter-
mediates that presumably form upon reaction of the Zr−C
bond with CO2.18,19 Mechanistic details for this system remain
somewhat obscure, but it demonstrates that an early metal
approach to hydrosilylation of CO2 using B(C6F5)3 as a
cocatalyst is promising.
Inspired by this study, we have begun to examine organo-
scandium catalyst precursors for useful CO2 conversion
chemistry. A handful of CO2 insertions into Sc−C20,21 and
Sc−Si22 bonds have been reported, but subsequent chemistry
of the carboxylate products has not been explored in detail. We
have a long-standing interest in the chemistry of neutral23−26
and cationic27−30 organoscandium compounds supported by
the β-diketiminato, or “nacnac”, ligand framework31 and report
here some initial studies on the reactivity of compounds 1a−1d,
shown in Scheme 1, with CO2. These compounds incorporate the
INTRODUCTION
■
The activation and conversion of carbon dioxide (CO2) into
useful products has long been recognized by organometallic
chemists as a desirable goal because of its plentitude and attractive
economic attributes as a C1 synthon.1−4 More recently, activity in
this area has also been motivated by the growing effect of high
levels of atmospheric CO2 on global climate trends.5,6 New
catalytic reactions involving CO2 as a feedstock are, therefore, of
considerable interest, although, realistically, the benefits of such
reactions will (unfortunately) continue to be economic rather than
environmental since the scales at which they must be run in order
to mitigate atmospheric CO2 are daunting even if they use more
carbon than they generate.7
One of the most established reactivity modes of CO2 is
insertion into M−H or M−C bonds.8,9 This is especially facile
when M is an early transition metal or an alkaline earth metal;10
indeed, the seminal work of Grignard11 on the conversion of
organomagnesium reagents to carboxylic acids exemplifies this
chemistry. However, the high oxophilicity of these elements
means that turnover by conversion of the resulting M−O bond
is thermodynamically difficult,12 and so these tend to be stoichio-
metric reactions for such metals, unless subsequent reactions also
involve formation of M−O bonds.13 In this context, a recent
report by Kawaguchi et al.,14 wherein a cationic organozirconium
Received: December 1, 2011
Published: January 11, 2012
© 2012 American Chemical Society
810
dx.doi.org/10.1021/om2012002 | Organometallics 2012, 31, 810−818

Organometallics
Article
isolated as a pure compound via two methods. Treatment of 1a
with an excess of CO2 at −78 °C, followed by removal of the
CO2 atmosphere prior to warming, led to excellent yields of 3a;
this reaction is sufficiently rapid that no intermediates or
unreacted starting materials were detected by NMR spectros-
copy on a sample maintained at −78 °C for the CO2 addition
and immediately inserted into a probe precooled to the same
temperature. Compound 3a was also synthesized independ-
ently by reaction of 1a with 2 equiv of rigorously dried acetic
acid at −29 °C (ortho-xylene/liq. N2 bath). The precise co-
ordination mode of the acetato ligands could not be established
on the basis of a structural determination, but the high solu-
bility of 3a in aromatic solvents and its characteristic fluxional24
NMR behavior are indicative of a monomeric structure. In the
1H NMR spectrum at room temperature, there is a single
resonance for the acetate methyl groups that broadens and
decoalesces as the temperature is lowered, re-emerging as two
singlets for the diastereotopic endo and exo groups now distinct
on the NMR time scale. The barrier for this exchange was
estimated to be ΔG⧧ = 10.5 kcal/mol at the temperature of
coalescence (240 K), similar to values obtained for a variety of
bis(alkyl) scandium nacnac derivatives.24 A dimeric complex
would be expected to exhibit a quite different dynamic behavior
due to the presence of three diastereomeric forms (i.e., endo/
endo, exo/exo, and endo/exo) for dimeric scandium nacnac
compounds.39
Mixing equimolar amounts of pure 1a and 3a establishes the
equilibrium with the dimeric mono insertion product 2a
through ligand metathesis; unfortunately, attempts to study the
temperature dependence of this equilibrium were hampered by
the poor solubility of 2a in aromatic solvents at low tem-
peratures and the tendency of 1a to undergo metalation24 at
higher temperatures. Because dimer 2a comprises 70−80% of
the mixture at room temperature and is of lower solubility, it
can be preferentially crystallized from these solutions, or those
obtained by reaction of 1a with 1 equiv of CO2, as described
above. X-ray analysis of crystals obtained in this way confirmed
the dimeric structure of 2a, featuring bridging acetato groups
and terminal scandium methyl groups; redissolution of these
crystals reproduces the equilibrium mixture of compounds indi-
cated in Scheme 2. A thermal ellipsoid diagram of the molecular
structure of 2a is shown in Figure 1, along with selected
metrical parameters. The molecule has a crystallographic C2
Scheme 1
bulky nacnac ligand featuring 2,6-diisopropylphenyl groups on
the nitrogen and tert-butyl groups on the imine carbons of the
diketiminato backbone,32 affording well-defined, base-free neutral
dialkyl compounds, here the dimethyl (1a) and bis(trimethylsilyl)-
methyl (1b) derivatives. When the dimethyl compound is
treated with 1 or 2 equiv of the alkide abstracting33 Lewis acid
B(C6F5)3,34,35 the well-behaved mono- (1c) and dicationic (1d)29
ion pairs are formed; together, these four compounds provide a
well-understood platform for probing the reactivity of CO2 with
organoscandium compounds. Organoscandium compounds are
notoriously Lewis acidic,36 and exploration of their fascinating
chemistry would not be possible without the foundational
techniques for handling such compounds developed by
Wilhelm Johann Schlenk.37,38
RESULTS AND DISCUSSION
■
Initial experiments on the reaction of dimethyl complex 1a with
excess CO2 gas at room temperature gave complex product
mixtures, so more careful investigations using stoichiometric
quantities of CO2 at lower temperatures were undertaken and
the chemistry summarized in Scheme 2 was consequently revealed.
Treatment of 1a with 1 equiv of CO2 in toluene at room tem-
perature gave a mixture of compounds identified as the dimeric
mono insertion product 2a (70−80%) along with 10−15% each
of the starting dimethyl complex and the bis insertion product,
diacetate 3a. This latter compound could be prepared and
Scheme 2
811
dx.doi.org/10.1021/om2012002 | Organometallics 2012, 31, 810−818

Organometallics
Article
the room-temperature NMR spectra. A H−1H EXSY NMR
experiment showed crosspeaks between the singlet resonances
for the nacnac ligand CH protons, indicative of exchange
between the major isomer and both minor isomers (Figure S1,
Supporting Information). However, exchange peaks between
the two minor isomers were not observed, indicating that this
exchange is too slow to observe by NMR spectroscopy, or that
the crosspeaks are not of high enough intensity to resolve. The
lesser of the two minor isomers exhibited a higher degree of
asymmetry than the others, giving rise to two ligand methyne
resonances in a 1:1 ratio, but assignments and solution struc-
ture are difficult to comment on due to weak signal-to-noise
ratios and/or overlapping signals. It appears, however, on the
basis of 2D 1H−1H ROESY experiments, that the major isomer
corresponds to the structure observed in the solid state,
averaged into a C2v symmetric structure. Of the four diastereo-
topic methyl groups (a−d, Figure 1, inset), one correlates with
the acetate methyl groups (a) and one with the Sc-methyl
group (b), while the others, which are pointed away from the
molecular core, do not. The other two isomers are variations on
this structure wherein the nacnac ligand ring “flips” its con-
formation relative to the other ligand in the dimer.39
When an analogous set of reactions was performed with 1b,
incorporating the bulkier CH2SiMe3 alkyl ligands, a slightly
different chemistry was observed (Scheme 3). Reaction with 1
equivalent of CO2 led to mixtures of unreacted starting
material, mono insertion product 2b and the bis(carboxylate)
3b, along with the appearance of a further product, 4b (vide
infra), in a 65:10:20:5 ratio, indicating that the rate of insertion
of CO2 into the second Sc−C bond is competitive with that of
the first. Even when 2 equiv of CO2 were employed, the
product mixture remained complex, although all of the starting
material was consumed; a mixture of 2b, 3b, and 4b was
obtained in an 11:65:24 ratio.
1
Figure 1. Thermal ellipsoid diagram (50%) of 2a; isopropyl groups on
the Ar substituents have been removed for clarity. The inset shows a
stick depiction of the structure. Selected bond distances (Å): Sc−
N(1), 2.1879(12); Sc−N(2), 2.1925(12); Sc−O(1), 2.0635(12); Sc−
O(2), 2.0864(11); Sc−C(12), 2.2109(18). Selected bond angles
(deg): N(1)−Sc−(N2), 88.22(5); O(1)−Sc−O(2), 83.45(5); N(2)−
Sc−O(2), 167.18(5); N(1)−Sc−C(12), 104.84(6); O(1)−Sc−C(12),
128.58(7); N(1)−Sc−O(1), 126.57(5).
symmetry relating the two halves of the dimer, with a distorted
trigonal-bipyramidal geometry at the scandium centers; N(1),
O(1), and C(12) occupy the trigonal plane, while N(2) and
O(2) are in the apical positions (N(2)−Sc−O(2) = 167.18(5)°).
The axial atoms N(2) and O(2) exhibit slightly longer distances to
Sc as compared with their equatorial counterparts, but all distances
to scandium fall into typical ranges. As is common for scandium
nacnac compounds, the scandium center lies out of the plane
defined by the five essentially coplanar ligand atoms, here by
0.955 Å.
The solution behavior of 2a is complex. In addition to the
aforementioned equilibrium with 1a and 3a, dimer 2a exists as a
mixture of three isomers in solution, in an approximately
30:15:1 ratio. As mentioned above, this behavior is anticipated
for dimeric scandium nacnac complexes.39 Typically, exchange
between the three possible diastereomers is slower on the
NMR time scale than the exchange of endo and exo positions
in monomeric compounds, and so all isomers are apparent in
Unlike the dimeric 2a, it appears that 2b, incorporating the
bulkier substituted carboxylate group, remains monomeric, based
on the following three observations. First, addition of 1b to a
mixture of 2b, 3b, and 4b generated as described above did not
perturb the ratio of the compounds as was observed in the “a”
series, suggesting that ligand redistribution processes (which
likely proceed via dimers) are suppressed in the CH2SiMe3
1
substituted compounds. Second, the H NMR spectra for 2b
are broad at room temperature and, when cooled to 235 K,
resolve into two distinct sets of resonances, consistent with the
formation of diastereomers in which the two different groups
Scheme 3
812
dx.doi.org/10.1021/om2012002 | Organometallics 2012, 31, 810−818

Organometallics
Article
Scheme 4
Figure 2. Thermal ellipsoid diagram (50%) of rac-4b; methyl groups attached to the silicon atoms have been removed for clarity. Selected bond
distances (Å): Sc−O(1), 2.158(3); Sc−O(2), 2.175(3); Sc−O(3), 2.180(3); Sc−O(4), 2.2.110(3); Sc′−O(4), 2.337(3); Sc−O(5), 2.091(3); Sc−
O(6), 2.116(3); C(3)−N(1), 1.273(5); C(5)−N(2), 1.260(5).
(alkyl and carboxylate) occupy the endo and exo coordination
sites (see Scheme 3). This is commonly observed for scandium
nacnac compounds with two different ligands R24 and contrasts
with the three different isomers typically observed for dimeric
structures.39 Finally, the chemical shifts for the carboxylate
carbon atoms in these two diastereomers are nearly coincident
(193.4 and 193.1 ppm) and are characteristic of terminal
carboxylates, which occur further upfield than those in bridging
carboxylates (cf. 178.2 ppm for 2a). The bis(carboxylate)
product 3b is, like 2b, monomeric and fluxional. At room
thermodynamically favored tris(carboxylate) complex, that by
virtue of the removal of steric bulk around the metal imposed
by the nacnac ligand is able to dimerize into a species with four
bridging carboxylate ligands. These reactions proceed at room
temperature over the course of a few hours to yield solutions of
4a/4b that give complex proton NMR spectra (Figures S2 and
S3, Tables S1−S4, Supporting Information).
The molecular structure of the CH2SiMe3-substituted
derivative 4b is shown in Figure 2 along with selected bond
distances. The dimer has a crystallographic center of inversion,
rendering the two halves of the dimer identical. Each scandium
center is coordinated by seven carboxylate oxygen atoms; the
ligand derived from the nacnac framework is a terminal
κ2O,O′carboxylato ligand, whereas those arising from the
CH2SiMe3 groups bridge the two scandium centers via two
distinct modes. In one mode, two carboxylato ligands bridge
the scandium centers via a symmetrical μ-carboxylato-κO:κO′
span. The other mode is less symmetrical in the sense that one
of the oxygen atoms, O(4), is tilted toward the second
scandium center such that it forms a weak interaction and
bridges the metal centers. The Sc−O(4) bond distance of
2.110(3)Å is similar to the others in the molecule, but the
distance to the second scandium is substantially longer (Sc′−
O(4), 2.337(3)Å).45 As a consequence of this weaker inter-
action, the scandium centers are seven coordinate, with a
distorted pentagonal-bipyramidal geometry in which the atoms
of the symmetrically bridging carboxylato ligands, O(5) and
O(6), occupy the apical sites.
1
temperature, H NMR spectra consistent with an averaged C2v
symmetrical structure are observed, and lowering the temper-
ature to 245 K freezes out the expected endo/exo fluxional
behavior. The exchange between isomers occurs with an esti-
mated barrier of ΔG⧧ = 11.6 kcal/mol, slightly higher than that
found in 2a, as anticipated due to the added bulk of the alkyl group.
The reactions of bis(alkyl) compounds 1a/1b with an excess
of CO2 eventually produce compounds 4a/4b, which derive
from addition of a third equivalent of CO2 to the ScN2C3
framework, followed by decoordination of the two nitrogen
donors; the result is the formation of scandium carboxylate
dimers, as shown in Scheme 4. This was assessed mainly via an
X-ray analysis of the product 4b (vide infra) but finds pre-
cedent in the addition of other unsaturated molecules to nacnac
Al,40 Ru,41 and Pt42 complexes in an analogous fashion. Here,
the electropositive, Lewis acidic scandium center, in coopera-
tion with the basic central carbon of the nacnac ligand, is able
to activate the CO2 ligand.43,44 Addition of the CO2 across
the nacnac-Sc ring in this way triggers the decoordination of
the two imine donors and formation of a (presumably)
In the decoordinated nacnac ligand, the C−N bonds are now
clearly double bonds (C−N distances are 1.275(5) and
813
dx.doi.org/10.1021/om2012002 | Organometallics 2012, 31, 810−818

Organometallics
Article
Scheme 5
1.260(5)Å). Additionally, the imine moieties differ in their
geometry in that one features trans-N-aryl and tBu groups, the
other cis, as drawn in Scheme 4. Consequently, the diimine
spectrum. In 2D 1H−1H ROESY experiments, the methyl
groups of the bridging acetates correlate to resonances for the
aryl isopropyl groups; the strong upfield shift to 1.01 ppm of
one set of acetate groups is likely due to the influence of a
proximal aryl ring. Conversely, the signal for the methyl group
of the bridging acetate group in the counteranion (2.10 ppm)
shows no correlations to nacnac ligand resonances. The averaged
Cs symmetry of the cation is also implied by the appearance of
four septets and eight separate doublets for the aryl isopropyl
groups, and two singlets for the tert-butyl groups on the ligand
backbone. Attempts to isolate ion pair 5 resulted in recovery of
oily, clathrate-like materials, a common phenomenon in early
metal cation chemistry utilizing these perfluoroarylborates as
weakly coordinating anions.40,49−52 This material is stable in
solution, but under an atmosphere of CO2 at room temper-
ature, conversion to a mixture of products is observed.
1
methyne carbons are chiral and the complexity of both the H
and the 13C NMR spectra can be explained by invoking the
presence of rac and meso diastereomers of 4a and 4b; the C2
symmetric S:R/R:S diastereomer preferentially crystallizes and
is the isomer depicted in Figure 2, assigning the cis-imine
moiety as the higher priority group.46 In solution, however,
both diastereomers are in evidence; this is particularly evident
in the 13C NMR spectra, where a doubling of most of the
resonances is observed for both compounds. Although it is not
possible to assign the sets of resonances to a particular diaste-
reomer, it is clear that they are both present in each case. A
1
complete listing of the observed H and 13C NMR resonances
for these compounds is given in Tables S1−S4 (Supporting
Information).
The above observations imply that, for acetato ligands, there
is a strong tendency for the cations to dimerize, despite the
considerable bulk of the nacnac ligand employed. This trend is
also apparent in the products of the reactions of CO2 with the
cationic compound 1c. When 1c is generated in situ from
(nacnac)ScMe2 and B(C6F5)3 and reacted with 1 equiv of
CO2 at temperatures near the freezing point of C6D5Br (−20 to
−30 °C), clean formation of a new product formulated as the
dimer 6 is observed (Figure 3). The precise structure of this
compound is unknown, but a four coordinate, acetate bridged
structure of high symmetry is indicated by the following spectro-
scopic observations. One septet and two doublets are observed for
The CO2-induced decoordination process described above
represents a potential catalyst degradation pathway in any reac-
tions mediated by neutral compounds 1, for example, hydro-
silylation. Indeed, while turnover is observed in the hydro-
silylation of CO2 using B(C6F5)3 as a cocatalyst in a similar way
to that reported by Kawaguchi et al., the reactions terminate
quite quickly. Since B(C6F5)3 is known to form the mono-
cationic 1c and dicationic 1d (Scheme 1), we postulated that
the preformation of these cations might dissuade the ligand
from undergoing displacement due to stronger Sc−N bonding.
We therefore tested this postulate by preparing cations via
reactions of 3a with B(C6F5)3 and treatment of cations 1c and
1d with CO2.
1
the N-aryl isopropyl groups in the H NMR spectra, consistent
with averaged C2v symmetry; this pattern is observed to the
freezing point of the C6D5Br solvent (245 K). 19F NMR spectro-
scopy shows that the methylborate counteranions do not interact
significantly with the scandium center by virtue of the small Δm,p
Reaction of the well-defined neutral bis(acetate) compound
3a with 1 equiv of B(C6F5)3 indicated that the borane was
capable of abstracting an acetate ligand from scandium. A single
product was rapidly formed, which we assign as the dimeric
cation 5 (Scheme 5).
value of 2.7 ppm.35,53 Finally, the observed H and 13C NMR
1
chemical shifts of 1.65 and 183.6 ppm, respectively, are consistent
with bridging, rather than terminal, acetato groups.
Here, a putative monomeric ion pair of the formula
[(nacnac)Sc(O2CCH3)]+[H3CC(O)OB(C6F5)3]− probably
forms, but the carbonyl group of the counteranion coordinates
the remaining half equivalent of borane,47,48 while the cationic
scandium acetate dimerizes with the unreacted bis(acetate) 3a.
The use of an excess of B(C6F5)3 (up to 5 equiv relative to
scandium) had no effect on the outcome of the reaction. In ion
pair 5, the dimeric cation is overall Cs symmetric, with two five
coordinate scandium centers bridged by two distinct acetate
groups found in a 1:2 ratio. Signals for the acetato methyl groups
The assignment of 6 as a dimer is also convincingly demon-
strated by its reactivity with varying equivalencies of 13CO2
(Figure 3; note that, when generated in situ in this way, the
irreversibly formed briding acetato groups are also labeled with
13C). The inset of Figure 3 shows the evolution of the region of
the 1H NMR spectrum associated with the ligand backbone C-
H resonances, and the chemistry depicted outlines our inter-
pretation of these observations. Upon introduction of 1 equiv
of 13CO2 to solutions of 6, the singlet for the CH group in 6
(red dot) transforms into a singlet at 6.17 ppm (dark blue) and
a doublet at 5.79 ppm (2JCH = 6.8 Hz, light blue) in a 1:1 ratio,
which we assign as the unsymmetrical dimer 7 in which one of
1
are found at 2.03 and 1.03 ppm in the H NMR spectrum,
whereas acetate carbon nuclei appear in the region associated with
bridging acetates at 181.3 and 181.0 ppm in the 13C NMR
814
dx.doi.org/10.1021/om2012002 | Organometallics 2012, 31, 810−818

Organometallics
Article
Figure 3. Reactions of nacnac scandium cations 1c and 1d with CO2. Inset: a portion of the 400 MHz NMR spectrum of the reaction of 1c with
varying equivalencies of 13CO2, showing the dimeric nature of the ionic complexes 6, 7, and presumably 8.
the nacnac ligands has been modified through transannular
bonding of CO2, in a similar fashion to that proposed above in
the formation of compounds 4a and 4b. Minor amounts of a
second product are already in evidence, and as more 13CO2 is
added, the doublet for this species at 5.72 ppm grows in until it
is the only compound present (yellow dot). In this product, 8,
13CO2 has added to both Sc(nacnac) rings of the dimer and is a
workup procedures. Since the 19F NMR data for each of these
species indicate that the [H3CB(C6F5)3]− anions do not
interact significantly with the scandium centers via the methide
group (Δm,p = 2.5 ppm for both 7 and 8), the poor crystallinity
of these ionic compounds is not surprising.
Overall, these results suggest that cationic nacnac scandium
methyl ions undergo rapid insertion of CO2 to give acetates and
subsequent addition of further CO2 equivalents to the nacnac-
Sc ring. Ligand displacement, although qualitatively slower in
the cationic systems, is still the ultimate fate of these com-
pounds. Nonetheless, it is clear that the oxophilic nature of
scandium renders it highly reactive toward CO2. For example,
when the reaction of 1b with 3 atm of CO2 at 235 K was
monitored by 1H NMR spectroscopy, kinetic modeling of
speciation plots indicated that the first insertion (to form 2b)
occurred with a pseudo-first-order rate constant of ≈2.3 × 10−3
s−1, and the second insertion (to form 3b) took place at only
slightly slower rates (kobs ≈ 0.4 × 10−3 s−1). Unfortunately, the
1
highly symmetrical species on the basis of its H NMR spectrum
(see Figure S7, Supporting Information). It may also be formed
directly from either 1c or 1d when treated with an excess of CO2
and slowly decomposes, presumably via ligand displacement
processes, upon prolonged residence in solution under an
atmosphere of CO2. The capture of CO2 by the nacnac ligand
framework is reversible; upon exposure of the 13C-labeled 8 to
an atmosphere of unlabeled CO2, the doublet at 5.72 ppm
disappears in favor of a singlet as the coupling to the 13C
nucleus is lost. None of the compounds 6−8 were isolable as
solids, instead emerging as oily materials upon attempted
815
dx.doi.org/10.1021/om2012002 | Organometallics 2012, 31, 810−818

Organometallics
Article
warmed to room temperature, and the solvent was evacuated in
vaccuo. The resulting oily yellow solid was triturated with pentane to
give a fine yellow solid (123 mg, 85%). 1H NMR (400 MHz, C72D8): δ
ligand displacement processes leading to compound 4b also
ensued to a significant degree under these conditions, again
suggesting that more robust ancillary ligands are required for
the development of scandium-based catalysts for reduction of
CO2 using, for example, silanes as the sacrificial reductant.
Investigations along these lines are currently underway.
7.04 (6H, C6H3), 6.08 (s, 1H, CH), 3.26 (sp, 2H, CH(CH3)2, JHH
=
2
6.7 Hz), 1.60 (s, 3H, OCOCH3), 1.36, 1.33 (d, 12H, JHH = 6.7 Hz,
CH(CH3)2), 1.16 (s, 18H, C(CH3)3). 13C NMR (101 MHz, C7D8):
191.2 (OCOCH3), 173.4 (NCC(CH3)3), 145.4 (Cipso), 142.0 (C6H3),
124.7, 123.3 (C6H3), 96.2 (CH), 44.5 (C(CH3)3), 32.1 (C(CH3)3),
28.7 (CH(CH3)2), 26.2, 25.7 (CH(CH3)2), 22.6 (OCOCH3). Anal.
Calcd for C39H59N2O4Sc2: C, 70.45; H, 8.94; N, 4.21. Found: C,
70.05; H, 8.72; N, 4.20.
EXPERIMENTAL SECTION
■
General Procedures and Equipment. An argon atmosphere
MBraun glovebox was employed for manipulation and storage of all
oxygen and moisture-sensitive compounds. Reactions were performed
on a double manifold high-vacuum line using standard techniques.
Toluene and hexane were dried using the Grubbs/Dow purification
system54 and stored over sodium/benzophenone in evacuated glass
vessels. Benzene and pentane were predried and distilled from 3 Å
molecular sieves, then stored over sodium/benzophenone in evacuated
glass vessels. Bromobenzene and d5-bromobenzene were predried and
distilled from 3 Å molecular sieves, then stored over fresh sieves in the
glovebox. d6-Benzene and d8-toluene were predried and distilled from
sodium/benzophenone and stored in a glass vessel in the glovebox.
NMR spectra were obtained on Bruker DRX400, AVANCE 400 MHz,
Generation of 2b. A resealable NMR tube was charged with 1b
(10 mg, 0.014 mmol) and C7D8 and degassed at −78 °C. CO2 was
expanded into a 19.1 mL transfer bulb (14 mmHg, 0.014 mmol) and
condensed into the tube at −196 °C. The tube was warmed to room
temperature and allowed to stand for 1 h. The resulting solution
contained a mixture of 2b, 3b, and 4b. NMR spectra were recorded
at 245 K, and assignments were made with reference to speciation
plots and 2D NMR experiments. Endo/exo isomers were detected in
solution; only one isomer is reported here. 1H NMR (600 MHz, C7D8,
245 K): 7.06 (m, 6H, C6H3), 6.04 (s, 1H, CH), 3.87, 2.96 (sp, 4H,
3
CH(CH3)2, JHH = 6.6 Hz), 1.90 (s, 2H, OCOCH2), 1.77, 1.53, 1.23,
1.15 (d, 6H, CH(CH3)2, 3JHH = 6.6 Hz), 1.15 (s, C(CH3)3), 0.37, 0.10
(s, 9H, SiC(CH3)3), 0.24 (s, 2H, ScCH2). 13C NMR (151 MHz, C7D8,
245 K): 193.4 (OCOCH2), 175.4 (NCC(CH3)3), 145.8 (Cipso), 142.8,
141.5, 126.2, 124.9, 124.6 (C6H3), 94.8 (CH), 45.1 (C(CH3)3), 32.9
(C(CH3)3), 29.2 (OCOCH2), 29.9, 29.0, 28.7, 27.03, 25.9, 25.2
(CH(CH3)2), 4.6 (Si(CH3)3), −0.5 (OCOCH2Si(CH3)3). The Sc-CH2
carbon nucleus was not detected.
1
AVANCE III 400 MHz, or AVANCE III 600 MHz spectrometers. H
and 13C{1H} chemical shifts were referenced to residual proton, and
naturally abundant 13C resonances of the deuterated solvent,
respectively, relative to tetramethylsilane. 11B NMR spectra were
referenced to an external standard of BF3·OEt2, and 19F NMR to an
external standard of CFCl3. NMR spectra were processed and analyzed
with MestReNova software (v6.2.1-7569). Glacial acetic acid (EMD
chemicals) was purified by refluxing at 110 °C over acetic anhydride
and KMnO4 under an argon atmosphere for 1.5 h, followed by
fractional distillation into a Schlenk flask. Tris(pentafluorophenyl)-
borane (B(C6F5)3) was doubly sublimed at 65 °C under static vacuum
and stored in the glovebox. Carbon dioxide, bone-dry grade 3.0, and
13CO2 (99% 13C) were obtained from Praxair and Cambridge Isotopes,
respectively, and used as received. Compounds LSc(CH3)2 (1a),
LSc(CH2Si(CH3)3)2 (1b), [LScMe][H3CB(C6F5)3], and [LSc][H3CB-
(C6F5)3]2 (L = 2,6-(i-Pr)2C6H3)NC(t-Bu)}2CH) were prepared accord-
ing to reported literature procedures.29 Elemental analyses were obtained
by the Instrumentation Facility of the Department of Chemistry,
University of Calgary.
Generation of 3b. A procedure identical to that for the synthesis
of 2b was used, but 2 stoichiometric equiv of CO2 were employed
1
(27 mmHg, 0.028 mmol). H NMR (600 MHz, C7D8): 7.05 (m, 6H,
C6H3), 6.06 (s, 1H, CH), 3.32 (br, 4H, CH(CH3)2), 1.62 (d, 4H,
ScCH2, 2JHH = 6.9 Hz), 1.40, 1.36 (d, 12H, CH(CH3)2, 3JHH = 6.6 Hz),
1.19 (s, 18H, C(CH3)3), 0.05 (s, 18H, Si(CH3)3). 13C NMR (151
MHz, C7D8): 192.8 (OCOCH2), 173.5 (NCC(CH3)3), 145.8 (Cipso),
142.25, 125.61, 124.60 (C6H3), 95.82 (CH), 44.60 (C(CH3)3), 32.33
(C(CH3)3), 29.10 (OCOCH2), 28.60, 26.72, 25.81 (CH(CH3)2),
−0.62 (Si(CH3)3).
Generation of 4a. A resealable NMR tube was charged with 1a
(16 mg, 0.028 mmol) and C7D8. The tube was degassed at −78 °C,
and CO2 was expanded into the tube (1 atm), which was allowed to
stand at room temperature overnight. Isomerism is evident in the
NMR spectra by nearly coincident or overlapping sets of peaks; only
the major isomer is reported here. See Figure S2 and Tables S1 and S3
Generation of 2a. A 10 mm diameter resealable tube was charged
with 1a (125 mg, 0.035 mmol) and 0.4 mL of C6H6. The resulting
yellow solution was degassed by freeze−pump−thaw cycles. CO2 was
expanded into a 19.1 mL transfer bulb (34 mmHg, 0.035 mmol) and
deposited into the tube at −196 °C. The tube was warmed to room
temperature and allowed to stand for 4 h. The resulting yellow
crystalline solid was isolated by filtration in the glovebox, then washed
1
in the Supporting Information for a complete list of peaks. 4a: H
NMR (400 MHz, C7D8): δ 7.21, 7.04 (m, C6H3), 5.49 (s, 1H, CH),
3
3.98, 3.78, 2.94, 2.85 (app p, 1H, CH(CH3)2, JHH = 6.7 Hz), 2.24,
1
2.02 (s, 3H, OCOCH3), 1.78, 0.78 (s, 9H, C(CH3)3), 1.55, 1.42, 1.41,
with cold benzene and hexanes (79 mg, 59%). H NMR (400 MHz,
3
1.36, 1.35, 1.34, 1.28, 1.22 (d, 3H, CH(CH3)2, JHH = 6.7 Hz).
C7D8) major isomer: δ 7.02, 6.91, 6.8 (m, C6H3), 6.07 (s, 1H, CH),
13C{1H} NMR (101 MHz, C7D8): δ 191.6 (OCOSc), 188.5, 184.6,
184.0, 181.5 (OCOCH3), 173.6, 168.9 (NCC(CH3)3), 145.7, 145.6
(Cipso), 137.1, 135.3, 134.9, 133.2 (CCH(CH3)2), 124.0, 123.9, 123.8,
123.5, 123,1, 122.7 (C6H3), 56.7 (CH), 45.1, 43.5 (C(CH3)3), 31.2,
28.9 (C(CH3)3), 29.7, 28.3, 26.9, 26.6, 25.9, 23.9, 23.2, 21.8
(CH(CH3)2), 22.9 (OCOCH3).
3
3.84, 2.96 (sp, 2H, CH(CH3)2, JHH = 6.7 Hz), 1.53, 1.38, 1.19, 0.90
3
(d, 6H, CH(CH3)2, JHH = 6.7 Hz), 1.17 (s, 18H, C(CH3)3), 0.26 (s,
3H, ScCH3), 0.14 (s, 3H, OCOCH3). 13C NMR (101 MHz, C6D6):
178.2 (OCOCH3), 177.1 (NCC(CH3)3), 144.9 (Cipso), 142.1, 128.9,
128.8 (C6H3), 98.5 (CH), 45.03 (C(CH3)3), 33.6, 32.0 (C(CH3)3),
32.6 (CH(CH3)2), 28.7, 28.4 (CH(CH3)2), 22.6 (OCOCH3). Anal.
Calcd for C82H123BrN4O4Sc2 (1a, containing 1 C6H5Br of solvation):
C, 70.41; H, 8.86; N, 4.01. Found: C, 69.58; H, 8.94; N, 4.04.
Synthesis of 3a. Method A: A 50 mL flask was charged with 1a
(200 mg, 0.347 mmol) and 5 mL of toluene. CO2 was expanded into
the flask (1 atm), and the mixture was stirred for 3 h at −78 °C to
ensure complete reaction. The flask was evacuated for 15 min at this
temperature to remove unreacted CO2, then warmed slowly to room
temperature to remove the solvent in vacuo. The resulting yellow oil
was triturated with pentane, and the solvent was removed to give a
yellow powder (210 mg, 97%). Method B: A 50 mL two-neck flask
was charged with 1a (126 mg, 0.218 mmol) and toluene (25 mL) and
cooled to −29 °C with an ortho-xylene/LN2 cold bath. Rigorously
dried acetic acid (25 μL, 0.437 mmol) was added to the solution via a
gastight microsyringe and stirred for 1 h. The solution was then
Generation of 4b. A procedure identical to that for the synthesis
of 4a was used, but beginning with 1b. Isomerism is evident in the
NMR spectra by nearly coincident or overlapping sets of peaks; only
the major isomer is reported here. See Figure S3 and Tables S2 and S4
1
in the Supporting Information for a complete list of peaks. H NMR
(400 MHz, C7D8): δ 7.32, 7.06 (m, C6H3), 5.50 (s, 1H, CH), 4.02,
3
3.89, 2.94, 2.86 (sp, 1H, CH(CH3)2, JHH = 6.7 Hz), 1.98 (s, 2H,
2
OCOCH2), 1.98 (d, 2H, CH2Si, JHH = 10.4 Hz), 1.79, 0.77 (s, 9H,
C(CH3)3), 1.54, 1.46, 1.41, 1.36, 1.35, 1.32, 1.27, 1.22 (d, 3H,
3
CH(CH3)2, JHH = 6.7 Hz), 0.19 (s, 18H, Si(CH3)3). 13C{1H} NMR
(101 MHz, C7D8): δ 192.6 (OCOSc), 185.2, 185.0 (OCOCH2), 173.7,
169.5 (NCC(CH3)3), 146.2 (Cipso), 137.3, 136.0, 135.8, 133.0
(CCH(CH3)2), 124.3, 124.1, 124.0, 123.8, 123.6, 123.5 (C6H3), 57.4
(CH), 45.7, 44.3 (C(CH3)3), 31.8, 29.3 (C(CH3)3), 30.1, 28.8, 27.4,
816
dx.doi.org/10.1021/om2012002 | Organometallics 2012, 31, 810−818

Organometallics
Article
27.3, 26.1, 24.1, 23.5, 22.2 (CH(CH3)2), 30.5 (OCOCH2), −0.9
(Si(CH3)3).
Biography
Generation of 5. A resealable NMR tube was charged with 3a (9
mg, 13 μmol), B(C6F5)3 (7 mg, 14 μmol), and C7D8, then shaken
1
briefly at room temperature to give a yellow solution. H NMR (400
MHz, C7D8): δ 6.97 (m, 12H, C6H3), 6.01 (s, 2H, CH), 2.95, 2.84,
2.67, 2.36 (sp, 2H, CH(CH3)2, 3JHH = 6.7 Hz), 2.12 (BCH3), 2.03 (s,
6H, OCOCH3), 1.18, 1.09 (s, 18H, C(CH3)3), 1.33, 1.29, 1.23, 1.08,
1.03, 0.94, 0.87, 0.48 (d, 6H, CH(CH3)2, 2JHH = 6.6 Hz), 1.03 (s, 12H,
OCOCH3). 13C NMR (101 MHz): δ 185.6 (OCOB), 181.3, 181.0
(OCOCH3), 179.2, 170.0 (C(CH3)3), 146.3, 143.0 (Cipso), 140.3,
139.9, 139.0, 137.4 (CCH(CH3)2), 127.5, 126.2, 125.5, 125.2, 124.9,
124.3 (C6H3), 97.6 (CH), 44.2, 44.0 (C(CH3)3), 31.3, 29.7 (C(CH3)3),
33.4, 28.7, 28.4, 27.1 (CH(CH3)2), 28.1, 27.9, 27.4, 26.8, 26.2, 24.6, 24.5,
24.4 (CH(CH3)2), 25.3, 23.3 (OCOCH3), 23.4 (br, BOCOCH3). 19F
3
NMR (376 MHz): δ −135.1 (app d, 6F, o-CF, JFF = 21 Hz), −160.0
(app t, 3F, p-CF, 3JFF = 20 Hz), −166.1 (br, 6F, m-CF, 3JFF = 20 Hz). 11
NMR (128 MHz): −14.5.
B
Warren Piers obtained his B.Sc. degree at the UBC in 1984 and
continued there as a NSERC Postgraduate Scholar under the tutelage of
Prof. Michael Fryzuk, graduating with a Ph.D. in 1988. He then spent
two years at the California Institute of Technology as an NSERC and
Killam Postdoctoral Fellow with Prof. John Bercaw. From 1990 to 1995,
he was an Assistant Professor at the University of Guelph. He then
moved back to western Canada to join the Chemistry Department at
the University of Calgary as an Associate Professor. In July 2000, he was
appointed to the S. Robert Blair Chair in Polymerization Catalysis and
Polymer Synthesis, an endowed research chair sponsored by Nova
Chemicals, and promoted to Full Professor. His research interests
include the development of synthetic applications of perfluoroaryl
boranes, mechanistic organometallic chemistry in catalysis, and the
development of novel boron-based organometallic materials. Piers has
more than 180 independent scholarly publications and several patents;
in addition to the inaugural Schlenk Lecture Award, Piers' honors
include the Province of Ontario's John C. Polanyi Prize in Chemistry
(1991), an Alfred P. Sloan Foundation Research Fellowship (1996−
2000), an NSERC E. W. R. Steacie Memorial Fellowship (2000−2002),
the Royal Society of Canada Rutherford Medal in Chemistry (2000),
the Merck Frosst Center for Therapeutic Research Award (2003), and
the CSC Alcan Lecture Award (2005). In 2006, he became an Elected
Fellow of the Royal Society of Canada and in January 2012 will take up
a Canada Council of the Arts Killam Research Fellowship for two
years. Other interests include music, wine, and mountaineering (not
necessarily in that order).
Generation of 6. A d5-bromobenzene solution of [LScCH3]-
[CH3B(C6F5)3] was prepared in situ by mixing precooled solutions of
1a (11 mg, 19 μmol) and B(C6F5)3 (10 mg, 19 μmol) in a vial and
storing in the freezer of the glovebox for 2 h. The resulting yellow/
orange solution was transferred to a resealable NMR tube and quickly
immersed in a cold bath (−20 to −30 °C). The tube was degassed
with freeze−pump−thaw cycles; then 13CO2 was expanded into a 19.1
mL transfer bulb (19 mmHg, 20 μmol) and deposited into the tube at
−196 °C. The tube was warmed to room temperature and allowed to
1
stand for 30 min. H NMR (400 MHz, C6D5Br): δ 7.16, 7.02 (m,
3
C6H3), 6.20 (s, 1H, CH), 2.14, (sp, 4H, CH(CH3)2, JHH = 6.7 Hz),
1.65 (d, 3H, OCOCH3, 2JHC = 6.1 Hz), 1.13, 0.79 (d, 6H, CH(CH3)2,
3JHH = 6.7 Hz), 1.03 (s, 18H, C(CH3)3), 0.99 (s, BCH3). 13C NMR
(101 MHz): 183.8 (OCOCH3), 139.6, 138.44, 128.39, 125.03 (C6H3),
91.02 (CH), 44.46 (C(CH3)3), 31.18 (C(CH3)3), 30.14 (CH(CH3)2),
29.81, 25.24 (CH(CH3)2), 23.47 (OCOCH3). Poor signal-to-noise
resulted from the complex oiling out of solution. Cipso carbon nuclei
were not detected. 19F NMR (376 MHz): δ −135.1 (br, 6F, o-CF),
−160.0 (br, 3F, p-CF), −166.1 (br, 6F, m-CF). 11B NMR (128 MHz):
δ −14.5.
Generation of 8. A procedure similar to that for the synthesis of
6 was used, but 10 stoichiometric equiv of 13CO2 were deposited
(186 mmHg, 0.19 mmol), and NMR data were collected after 16 h. 1H
NMR (400 MHz, C6D5Br): δ 7.09, 6.98 (m, 6H, C6H3), 5.72 (d, 1H,
3
2
CH, JHC = 6.8 Hz), 2.39, 2.30 (sp, 4H, CH(CH3)2, JHH = 6.6 Hz),
2
1.15, 1.05, 1.02, 0.95, (d, 6H, CH(CH3)2, JHH = 6.6 Hz), 1.09
2
(s, CH3B), 0.95 (s, 18H, C(CH3)3), 0.33 (d, 3H, JHC = 6.2 Hz,
OCOCH3). 13C{1H} NMR (101 MHz): 182.8 (nondecoupled q, 2JHC
=
3
ACKNOWLEDGMENTS
6.2 Hz, OCOCH3), 164.4 (nondecoupled d, JHC = 6.7 Hz, OCOSc),
129.7, 125.1 (C6H3), 29.1 (C(CH3)3), 30.25, 28.3 (CH(CH3)2), 26.8,
23.9, 23.4, 22.3 (CH(CH3)2), 20.6 (OCOCH3). Poor signal-to-noise
resulted from the complex oiling out of solution. Quaternary carbon
nuclei were not detected. 19F NMR (376 MHz): −131.8 (app d, 6H, o-
CF, 3JFF = 21 Hz), 164.1 (app t, 3H, p-CF, 3JFF = 21 Hz), 166.6 (app t,
■
This work was funded by the Natural Science and Engineering
Research Council (NSERC) of Canada in the form of a
Discovery Grant to W.E.P. F.A.L. would like to thank NSERC
and the Alberta Ingenuity Fund for Fellowship support. A.B.
thanks the German Research Foundation (DFG) for a post-
doctoral fellowship. Nova Chemicals Research and Technology
Centre, Calgary, Alberta, is acknowledged for a generous dona-
tion of B(C6F5)3. This manuscript was drafted during a stay at
3
6H, m-CF, JFF = 21 Hz). 11B NMR (128 MHz): δ −14.5.
ASSOCIATED CONTENT
■
the University of Tubingen, and W.E.P. would like to thank
̈
S
* Supporting Information
Prof. Reiner Anwander (Tubingen) and his colleagues for their
̈
Crystallographic data files for compounds 2a and 4b and
Figures S1−S7 and Tables S1−S4. This material is available free
of charge via the Internet at http://pubs.acs.org.
generous hospitality during this visit. W.E.P. would also like to
thank Dr. Thomas Groesser of BASF and his colleagues for
their generous support of the Schlenk Award.
AUTHOR INFORMATION
DEDICATION
■
■
†Dedicated to my friends and colleagues at the University of
Corresponding Author
*E-mail: wpiers@ucalgary.ca.
Tubingen for their hospitality during a three-week stay as the
̈
817
dx.doi.org/10.1021/om2012002 | Organometallics 2012, 31, 810−818

Organometallics
Article
(39) Knight, L. K.; Piers, W. E.; McDonald, R. Chem.Eur. J. 2000,
6, 4322−4326.
inaugural Schlenk Lecturer and Guest Professor in October
2011.
(40) Radzewich, C. E.; Coles, M. P.; Jordan, R. F. J. Am. Chem. Soc.
1998, 120, 9384−9385.
REFERENCES
(41) Moreno, A.; Pregosin, P. S.; Laurenczy, G.; Phillips, A. D.;
Dyson, P. J. Organometallics 2009, 28, 6432−6441.
(42) Scheuermann, M. L.; Fekl, U.; Kaminsky, W.; Goldberg, K. I.
Organometallics 2010, 29, 4749−4751.
■
(1) Aresta, M.; Dibenedetto, A. Dalton Trans. 2007, 2975−2992.
(2) Behr, A. Angew. Chem., Int. Ed. Engl. 1988, 27, 661−678.
(3) Leitner, W. Angew. Chem., Int. Ed. Engl. 1995, 34, 2207−2221.
(4) Olah, G. A.; Prakash, G. K. S.; Goeppert, A. J. Am. Chem. Soc.
2011, 133, 12881−12898.
(43) Gambarotta, S.; Arena, F.; Floriani, C.; Zanazzi, P. F. J. Am.
Chem. Soc. 1982, 104, 5082−5092.
(44) Audett, J. D.; Collins, T. J.; Santarsiero, B. D.; Spies, G. H. J. Am.
Chem. Soc. 1982, 104, 7352−7353.
(5) Solomon, S.; Plattner, G.-K.; Knutti, R.; Friedlingstein, P. Proc.
Natl. Acad. Sci. U.S.A. 2009, 106, 1704−1709.
(6) Keith, D. W. Science 2009, 325, 1654−1655.
(7) Sakakura, T.; Choi, J.-C.; Yasuda, H. Chem. Rev. 2007, 107,
2365−2387.
(8) Jessop, P. G.; Joo, F.; Tai, C.-C. Coord. Chem. Rev. 2004, 248,
2425−2442.
(9) Jessop, P. G.; Ikariya, T.; Noyori, R. Chem. Rev. 1995, 95, 259−
272.
(10) Zhou, X.; Zhu, M. J. Organomet. Chem. 2002, 647, 28−49.
(11) Grignard, V. Ann. Univ. Lyon 1901, 6, 1.
(12) Williams, V. A.; Manke, D. R.; Wolczanski, P. T.; Cundari, T. R.
Inorg. Chim. Acta 2011, 369, 203−214.
(13) Cui, D.; Nishiura, M.; Hou, Z. Macromolecules 2005, 38, 4089−
4095.
(45) Antonelli, D. M.; Cowie, M. Organometallics 1991, 10, 2173−
2177.
(46) Cahn, R. S.; Ingold, C.; Prelog, V. Angew. Chem., Int. Ed. Engl.
1966, 5, 385−415.
(47) Parks, D. J.; Piers, W. E.; Parvez, M.; Atencio, R.; Zaworotko, M.
J. Organometallics 1998, 17, 1369−1377.
(48) Berkefeld, A.; Piers, W. E.; Parvez, M. J. Am. Chem. Soc. 2010,
132, 10660−10661.
(49) Bochmann, M.; Lancaster, S. J. Organometallics 1993, 12, 633−
640.
(50) Jia, L.; Yang, X.; Stern, C. L.; Marks, T. J. Organometallics 1997,
16, 842−857.
(51) Horton, A. D.; Orpen, A. G. Organometallics 1991, 10, 3910−
3918.
(14) Matsuo, T.; Kawaguchi, H. J. Am. Chem. Soc. 2006, 128, 12362−
12363.
(15) Matsuo, T.; Kawaguchi, H. Inorg. Chem. 2007, 46, 8426−8434.
(16) Parks, D. J.; Piers, W. E. J. Am. Chem. Soc. 1996, 118, 9440−
9441.
(52) Spence, R.; Piers, W. E. Organometallics 1995, 14, 4617−4624.
(53) Horton, A. D.; de With, J. Organometallics 1997, 16, 5424−
5436.
(54) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J. Organometallics 1996, 15, 1518−1520.
(17) Parks, D. J.; Blackwell, J. M.; Piers, W. E. J. Org. Chem. 2000, 65,
3090−3098.
(18) Rosenthal, U.; Ohff, A.; Michalik, M.; Goerls, H.; Burlakov, V.
V.; Shur, V. B. Organometallics 1993, 12, 5016−5019.
(19) Hill, M.; Wendt, O. F. Organometallics 2005, 24, 5772−5775.
(20) Fryzuk, M. D.; Giesbrecht, G.; Rettig, S. J. Organometallics 1996,
15, 3329−3336.
(21) St. Clair, M. A.; Santarsiero, B. D. Acta Crystallogr., Sect. C 1989,
45, 850−852.
(22) Campion, B. K.; Heyn, R. H.; Tilley, T. D. Inorg. Chem. 1990,
29, 4355−4356.
(23) Lee, L. W. M.; Piers, W. E.; Elsegood, M. R. J.; Clegg, W.;
Parvez, M. Organometallics 1999, 18, 2947−2949.
(24) Hayes, P. G.; Piers, W. E.; Lee, L. W. M.; Knight, L. K.; Parvez,
M.; Elsegood, M. R. J.; Clegg, W. Organometallics 2001, 20, 2533−
2544.
(25) Conroy, K. D.; Hayes, P. G.; Piers, W. E.; Parvez, M.
Organometallics 2007, 26, 4464−4470.
(26) Conroy, K. D.; Piers, W. E.; Parvez, M. Organometallics 2009,
28, 6228−6233.
(27) Hayes, P. G.; Piers, W. E.; McDonald, R. J. Am. Chem. Soc. 2002,
124, 2132−2133.
(28) Hayes, P. G.; Piers, W. E.; Parvez, M. J. Am. Chem. Soc. 2003,
125, 5622−5623.
(29) Hayes, P. G.; Piers, W. E.; Parvez, M. Organometallics 2005, 24,
1173−1183.
(30) Hayes, P. G.; Piers, W. E.; Parvez, M. Chem.Eur. J. 2007, 13,
2632−2640.
(31) Bourget-Merle, L.; Lappert, M. F.; Severn, J. R. Chem. Rev. 2002,
102, 3031−3066.
(32) Budzelaar, P. H. M.; van Oort, A. B.; Orpen, A. G. Eur. J. Inorg.
Chem. 1998, 1485−1494.
(33) Chen, E. Y.-X.; Marks, T. J. Chem. Rev. 2000, 100, 1391−1434.
(34) Piers, W. E.; Chivers, T. Chem. Soc. Rev. 1997, 26, 345−354.
(35) Piers, W. E. Adv. Organomet. Chem. 2005, 52, 1−76.
(36) Piers, W. E.; Shapiro, P. J.; Bunel, E. E.; Bercaw, J. E. Synlett
1990, 74−84.
(37) Tidwell, T. T. Angew. Chem., Int. Ed. 2001, 40, 331−337.
(38) Seyferth, D. Organometallics 2008, 28, 2−33.
818
dx.doi.org/10.1021/om2012002 | Organometallics 2012, 31, 810−818