Deprotonation of Al2Me6 by Sterically Bulky NHCs: Scope, Rationale through DFT Studies, and Application in the Methylenation of Carbonyl Substrates

Article abstract of DOI:10.1021/acs.organomet.6b00159

The sterically bulky NHCs 1,3-di-tert-butylimidazol-2-ylidene (I^tBu), 1,3-di-tert-butylimidazolin-2-ylidene (S^tBu), and 1,3-di-tert-butyl-3,4,5,6-tetrahydripyrimidin-2-ylidene (C6-^tBu) were found to readily react with excess Al₂Me₆ at room temperature to form salts 2, 3, and 4, respectively, consisting of the polynuclear Me₃Al(μ³-CH₂)(AlMe₂)₂(μ²-CH₃)^- anion associated with either the cation I^tBu-H⁺, S^tBu-H⁺, or C6-^tBu-H⁺. Such a reaction involving the deprotonation of an Al₂Me₆ moiety by a NHC does not proceed with less sterically hindered NHCs such as 1,3-bis(2,6-diisopropylphenyl)imidazol-2-ylidene (IDipp) and 1,3-bis(2,4,6-trimethylphenyl)imidazol-2-ylidene (IMes), for which only the classical Lewis pair adducts (NHC)AlMe₃ were isolated. In line with experimental and density functional theory (DFT) calculations data, such reactivity thus appears to be driven by steric frustration, resulting in the destabilization of the corresponding (NHC)AlMe₃ adducts, then more prone to dissociate and hence allowing Al₂Me₆ activation/deprotonation. The DFT-estimated profile of the reaction of model adduct (I^tBu)AlMe₃ (I) with Al₂Me₆ agrees with a ready adduct dissociation at low energy cost, with a subsequent deprotonation of Al₂Me₆ by the NHC fragment to afford model salt II (isostructural to salt 2). Anion Me₃Al(μ³-CH₂)(AlMe₂)₂(μ²-CH₃)^- behaves as an efficient CH₂^2- group transfer agent with the methylenation of aldehydes and ketones to afford the corresponding methylene organics in good conversions. Bonding analysis of the latter anion agrees with an enhanced nucleophilicity of the Al-CH₂ moiety (compared to the Al-Me groups), in line with the observed reactivity. DFT calculations also allowed a detailed bonding description of the pentacoordinate methylene carbon in the anion Me₃Al(μ³-CH₂)(AlMe₂)₂(μ²-CH₃)^-.

Full text of DOI:10.1021/acs.organomet.6b00159

Article
pubs.acs.org/Organometallics
Deprotonation of Al2Me6 by Sterically Bulky NHCs: Scope, Rationale
through DFT Studies, and Application in the Methylenation of
Carbonyl Substrates
Gilles Schnee,† David Specklin,† Jean-Pierre Djukic,⧧ and Samuel Dagorne*,†
⧧
†SRCO Group and Laboratoire de Chimie et System
́
ique Organo-met
́
alliques, Institut de Chimie de Strasbourg, CNRS-Universite
́
de Strasbourg, UMR 7177, 1 Rue Blaise Pascal, F-67000 Strasbourg, France
S
* Supporting Information
ABSTRACT: The sterically bulky NHCs 1,3-di-tert-butylimidazol-2-ylidene (ItBu), 1,3-di-tert-butylimidazolin-2-ylidene (StBu),
and 1,3-di-tert-butyl-3,4,5,6-tetrahydripyrimidin-2-ylidene (C6-tBu) were found to readily react with excess Al2Me6 at room
temperature to form salts 2, 3, and 4, respectively, consisting of the polynuclear Me3Al(μ3-CH2)(AlMe2)2(μ2-CH3)− anion
associated with either the cation ItBu-H+, StBu-H+, or C6-tBu-H+. Such a reaction involving the deprotonation of an Al2Me6
moiety by a NHC does not proceed with less sterically hindered NHCs such as 1,3-bis(2,6-diisopropylphenyl)imidazol-2-ylidene
(IDipp) and 1,3-bis(2,4,6-trimethylphenyl)imidazol-2-ylidene (IMes), for which only the classical Lewis pair adducts
(NHC)AlMe3 were isolated. In line with experimental and density functional theory (DFT) calculations data, such reactivity
thus appears to be driven by steric frustration, resulting in the destabilization of the corresponding (NHC)AlMe3 adducts, then
more prone to dissociate and hence allowing Al2Me6 activation/deprotonation. The DFT-estimated profile of the reaction of
model adduct (ItBu)AlMe3 (I) with Al2Me6 agrees with a ready adduct dissociation at low energy cost, with a subsequent
deprotonation of Al2Me6 by the NHC fragment to afford model salt II (isostructural to salt 2). Anion Me3Al(μ3-
CH2)(AlMe2)2(μ2-CH3)− behaves as an efficient CH22− group transfer agent with the methylenation of aldehydes and ketones to
afford the corresponding methylene organics in good conversions. Bonding analysis of the latter anion agrees with an enhanced
nucleophilicity of the Al-CH2 moiety (compared to the Al-Me groups), in line with the observed reactivity. DFT calculations also
allowed a detailed bonding description of the pentacoordinate methylene carbon in the anion Me3Al(μ3-CH2)(AlMe2)2(μ2-
CH3)−.
formation of unstable and reactive NHC-M adducts due to
severe steric congestion or even prevent adduct formation in
the case of frustrated Lewis pairs (FLPs). Such compounds are
of particular interest for small-molecule activation.3 In this
regard and most notably, Stephan and Tamm showed that no
adduct forms upon combination of the bulky NHC ItBu with
B(C6F5)3 and that the ItBu/B(C6F5)3 FLP readily cleaves H2 at
room temperature to afford the corresponding imidazolium salt
[ItBu-H][HB(C6F5)3].4 Also, several sterically bulky imidazol-
2-ylidene discrete metal adducts [based on Al(III), Ga(III),
Mn(II), and Fe(II)] were recently reported to be quite reactive
and to isomerize to the corresponding abnormal/mesoionic
INTRODUCTION
■
Due to their exceptional σ-donating properties and steric
tunability, N-heterocyclic carbenes (NHCs) are a privileged
class of ligands for the stabilization of a wide range of metal/
heteroatom centers.1 Consequently, NHC-bearing metal
complexes have found numerous applications in various areas
ranging from fundamental reactivity to their widespread use in
homogeneous catalysis.1 The use of NHCs across transition
metal and main group chemistry for the stabilization of various
unusual structural motifs and study of the derived reactivity is
also currently attracting attention.1,2 Although the steric
properties of NHCs (through the nature of the N-substituents)
are typically exploited to protect the coordinated metal center
(kinetic stabilization), the coordination of sterically hindered
NHC ligands to metal centers/heteroatoms may lead to the
Received: February 26, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.6b00159
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
NHC (aNHC, C4/C5-bonded NHC) metal/heteroatom
species for steric relief.5 In the case of sterically bulky Al(III)
derivatives, the formation of the aNHC metal derivatives was
shown to be thermodynamically favored for steric reasons
despite the electronic preference of Al(III) for NHC
coordination at the C2 carbene carbon.5a
Such a hydrodynamic volume value is similar to that estimated
for salt 2 from its solid-state structure (VX‑ray = 433 Å3, Figure
S11),9 which is in line with salt 2 consisting of discrete ItBu-
H+/Me3Al(μ3-CH2)(AlMe2)2(μ2-CH3)− ion pairs (and not as
fully dissociated cations/anions) in CD2Cl2 at room temper-
ature. For comparison, the DOSY NMR data for carbene ItBu
(nearly identical in size to the ItBu-H+ cation) agree with a
much smaller molecule volume (hydrodynamic radius = 3.2(7)
Å and VDOSY = 146(3) Å3, Figure S10).
In the area of Al(III)-NHC chemistry,6 we earlier
communicated that the bulky adduct (ItBu)AlMe3 (1, Scheme
1) is unstable in the presence of excess Al2Me6 to yield the
To better understand the formation of salt 2 from adduct 1
and Al2Me6, the stability of adduct 1 toward dissociation in the
presence/absence of Al2Me6 was studied. Species 1 on its own
shows no observable sign of dissociation (CD2Cl2 or C6D6,
room temperature) according to 1H NMR data. In contrast, the
1H NMR spectrum (CD2Cl2, room temperature) of a 1/4
mixture of (ItBu)AlMe3 and Al2Me6 contains two broad AlMe3
singlet signals (Figure S12), indicative of a dynamic process.10
Scheme 1
1
For comparison, the H NMR resonances of the less sterically
11
bulky adduct (IDipp)AlMe3
(IDipp = (1,3-bis(2,6-
diisopropylphenyl)imidazol-2-ylidene) are not affected at all
by the presence of 4 equiv of Al2Me6 (CD2Cl2, room
temperature; C6D6, from 25 to 75 °C; Figures S13 and S14).
The observed dynamic behavior for the (ItBu)AlMe3/Al2Me6
mixture is thus clearly related to the severe steric congestion in
adduct 1 when not observed with the less sterically hindered
adduct (IDipp)AlMe3. It seems reasonable to assume a
decoordination/coordination of adduct (ItBu)AlMe3 in the
presence of AlMe3 under the studied conditions, although a
complete VT NMR analysis (along with the determination of
the associated parameters) could not be performed.10 Note that
the ready dissociation of adduct (ItBu)AlMe3 in the presence of
quantitative formation of the imidazolium salt [ItBu-H]]-
[Me3Al(μ3-CH2)(AlMe2)2(μ2-CH3)] (2, Scheme 1), a reaction
that may be formally described as a deprotonation of Al2Me6
(i.e., dimeric AlMe3) by ItBu, although the mechanism of such a
process remained unclear at the time.7 Besides its peculiarity,
the interest in this deprotonation reaction also arises from the
−
formation of an unusual polynuclear AlMe3-stabilized Al-CH2
anion, the Brønstedt conjugate base of dimer Al2Me6 stabilized
by an AlMe3 group. In addition to its structural interest
(including its bonding features), such a readily accessible anion
may act as a methylenating agent of carbonyl substrates by
analogy with the reactivity of the Tebbe reagent and related
metal methylene/methylidene species.8 In the present paper,
we provide a full account of our studies in the area including (i)
the scope of such a deprotonation reactivity, (ii) DFT studies
to establish the mechanism converting 1 (in the presence of
excess Al2Me6) to salt 2, (iii) the electronic and bonding
characteristics of adduct 1 and salt 2, and (iv) the use of such
1
Al2Me6 is suggested by DFT studies (vide infra). A H NMR
monitoring of the reaction of adduct 1 in the presence of 4
equiv of Al2Me6 (CD2Cl2, room temperature) led to no
observable intermediate prior to the generation of the final
product 2.
In the present studies, we also found that salt 2 may be
produced in a straightforward manner by direct reaction of
carbene ItBu in neat Al2Me6, the latter thus acting both as
solvent and as reagent (quantitative reaction after 2 h at room
temperature). Such an improved procedure (i.e., in neat AlMe3)
was also tested with other carbene sources to probe/extend the
scope of such a deprotonation reactivity. In this regard, unlike
what is observed with carbene ItBu, the reaction of the less
sterically demanding NHCs IDipp and IMes (1,3-bis(2,4,6-
trimethylphenyl)imidazol-2-ylidene) in neat Al2Me6 (room
temperature, 20 h) led only to the isolation of the
corresponding Al-NHC adducts (IDipp)AlMe3 and (IMes)-
AlMe3, as deduced from NMR data and comparison with
literature data.6a,11 Also, species (IDipp)AlMe3 and (IMes)-
AlMe3 are stable in the presence of an excess Al2Me6 (4 equiv,
CD2Cl2 or C6D6) either at room temperature or upon heating
(40 and 75 °C). These experiments further highlight and
confirm the key role of steric bulk for Al2Me6 deprotonation by
NHCs to occur. In contrast, the more sterically demanding
carbenes such as StBu and C6-tBu are also readily protonated in
neat Al2Me6 to afford the corresponding salt species [StBu-
H][Me3Al(μ3-CH2)(AlMe2)2(μ2-CH3)] (3) and [C6-tBu-H]-
[Me3Al(μ3-CH2)(AlMe2)2(μ2-CH3)] (4), which were isolated
in nearly quantitative yields as analytically pure colorless oils
−
Al-CH2 -based anions for the methylenation of carbonyl
substrates.
RESULTS AND DISCUSSION
■
Access to the Aluminate Salts 2−4 by Deprotonation
of Al2Me6: Scope and Stability. Preliminary studies showed
that the Al-NHC adduct (ItBu)AlMe3 (1, Scheme 1) slowly
reacts with excess Al2Me6 (2.5 equiv of Al2Me6, CH2Cl2, room
temperature, 20 h) to quantitatively afford the corresponding
salt [ItBu-H][Me3Al(μ3-CH2)(AlMe2)2(μ2-CH3)] (2), which
was isolated in a pure form as a colorless oil. The formation of
2 formally arises from the deprotonation of dimer Al2Me6 by
ItBu (in adduct 1) to yield trinuclear aluminate anion
Me3Al(μ3-CH2)(AlMe2)2(μ2-CH3)− along with the corre-
sponding imidazolium cation ItBu-H+, as rationalized by DFT
calculations (vide infra). According to X-ray crystallographic
analysis, salt 2 crystallizes as discrete ItBu-H+ and Me3Al(μ3-
CH2)(AlMe2)2(μ2-CH3)− ions with no apparent anion−cation
interactions. In solution, a DOSY (diffusion ordered spectros-
copy) measurement for 2 (CD2Cl2, room temperature, Figure
S9), allowing an estimation of its hydrodynamic radius and
volume, suggests associated ion pairs in solution under the
studied conditions. Thus, a hydrodynamic radius and volume of
4.6(5) Å and 420(5) Å3, respectively, were measured for salt 2.
1
(Scheme 2). The H and 13C NMR data for 3 and 4 (CD2Cl2,
room temperature) are identical to those for salt 2, except the
resonances for cations StBu-H+ and C6-tBu-H+, respectively. In
B
DOI: 10.1021/acs.organomet.6b00159
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
presence of excess InMe3 (20 equiv of InMe3, CH2Cl2, room
temperature, overnight).13
Scheme 2
DFT Studies of Model Species (ItBu)AlMe3 (I). The
peculiar reactivity of adduct (ItBu)AlMe3 (1) [relative to
adducts such as (IDipp)AlMe3 and (IMes)AlMe3] in the
presence of Al2Me6 prompted us to investigate its bonding and
thermodynamic properties through DFT calculations to gain
insight into the observed reactivity. The singlet ground-state
geometry of model species I was faithfully reproduced at the
meta-GGA and GGA levels of theory, i.e., respectively with
ZORA-TPSS-D3(BJ)/all-electron TZP and ZORA-PBE-
D3(BJ)/all-electron TZP methods, with a rather better fit
with experimental values (most importantly with the Al−
Ccarbene distance) being obtained with the meta-GGA method
(Table S1). Natural bond order (NBO)14 analysis of the Al−
Ccarbene bond indicated a Wiberg bond index of 0.51, indicative
of a dative bond, and a strongly polarized bond with a charge at
Ccarbene and Al(III) of +0.07 and +1.37, respectively. The
particular, the 13C NMR spectra for 3 and 4 (CD2Cl2, room
temperature) both contain two upfield signals at δ −5.8 and
−4.7 ppm assigned to the Al-Me and Al-CH2 groups,
respectively. On the basis of NMR data, ionic liquids 3 and 4
are most probably isostructural to salt 2 in solution and in the
solid state.
Although stable for days in CD2Cl2, salts 2−4 decompose
within a few hours at room temperature in C6D6 to unknown
species, with a reaction apparently involving both the anion and
cation according to 1H NMR data. Such a difference in stability
possibly reflects a somewhat better cation−anion separation in
CD2Cl2 vs C6D6, hence an improved stability of these salts in
CD2Cl2.
natural bonding orbital for the Al−Ccarbene bond, i.e., ψC1−Al
=
0.915(sp1.4)C1 + 0.403(sp4.8)Al (population 1.93 electrons),
indicates a major contribution of the sp1.4 hybrid at Ccarbene, as
illustrated by the isosurface plot in Figure 1. These data agree
Interestingly, the observed deprotonation reaction of Al2Me6
by sterically bulky carbenes may be reversible, as demonstrated
by the reactivity of salt 3. Thus, the reaction of [StBu-
H][Me3Al(μ3-CH2)(AlMe2)2(μ2-CH3)] (3) with an excess of
THF (10 equiv, C6D6, 15 min, room temperature) led to the
immediate and quantitative formation of free carbene StBu
along with 3 equiv of adduct THF-AlMe3 (Scheme 3; Figure
Scheme 3
Figure 1. Natural bonding orbital ψAl−C(NHC) and associated
parameters for the Al−Ccarbene bond in I.
with a covalent Al−Ccarbene bond with a strong ionic character, a
bond only slightly fragilized by the much longer Al−Ccarbene
bond (2.161 Å) compared to that in less sterically cluttered
NHC-AlMe3 Lewis adducts (2.11 Å on average based on
literature).6a
The nature of the Al−Ccarbene bond in model I was also
investigated by the quantum theory of atoms in molecule
(QTAIM), that is, the analysis of the electron density topology.
The bond critical point (3,−1) located in the Al−Ccarbene
segment bears a residual electron density ρ of 0.052 au that
is slightly lower than that expected for a typical shared
interaction (covalent bond). The sign of the Laplacian of the
density −1/4∇ρ = −0.04 suggests some degree of ionicity in
the bonding interaction. In other terms the ρ and −1/4∇ρ
values point at an intermediate situation between a closed-shell
interaction and a shared interaction in that the density at the
bond critical point remains relatively high with a negative value
of the −1/4∇ρ Laplacian. This further supports the stance on
the role of electrostatic interactions in the overall cohesion of I.
A NOCV-EDA (natural orbitals for chemical valence−energy
decomposition analysis) analysis was performed for model
species I to get a more precise picture of the type of
interactions (and their extent) between the Me3Al and ItBu
fragments.15,17 As depicted in the Supporting Information
(Figure S20), such an analysis reveals, as expected, the major
contribution due to the Ccarbene→Al σ-bond formation,
representing about 65% of the orbital interaction energy. A
Rauk−Ziegler energy decomposition analysis16 was carried out
S14). In the presence of excess THF, the deprotonation of the
imidazolinium StBu-H+ in salt 3 presumably results from the
cleavage of the aggregate anion [Me3Al(μ3-CH2)(AlMe2)2(μ2-
CH3)] by THF coordination to Al(III) center(s), which in turn
enhances the basicity of the Al-CH2− moiety so that it may then
deprotonate the StBu-H+ cation to yield carbene StBu and
adduct THF-AlMe3. Despite the much stronger σ-donating
character of StBu vs THF, the formation of free StBu + THF-
AlMe3 is preferred over that of adduct (StBu)AlMe3 + free
THF, reflecting severe steric congestion in adduct (StBu)-
AlMe3.12 A similar preference for THF coordination was earlier
observed with carbene ItBu.5a
Higher Al alkyls such as Al2Et6 (dimeric AlEt3) and Al(iBu)3
as well as lower group 13 metal analogues GaMe3 and InMe3
were also studied. The reaction of StBu or ItBu in neat Al2Et6 or
Al(iBu)3 (room temperature, 2 to 18 h) produced a
complicated mixture of compounds that could not be identified,
thus in sharp contrast with the clean formation of salts 2 and 3
observed with AlMe3. On the other hand, the reaction of ItBu in
neat GaMe3 (room temperature, overnight) led only to the
isolation of the corresponding adduct (ItBu)GaMe3. Similarly,
adduct (ItBu)InMe3 was isolated upon reaction of ItBu in the
C
DOI: 10.1021/acs.organomet.6b00159
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
to extract the orbital interaction energy term, ΔEorbital, from the
total interaction energy, ΔEint (−53.1 kcal/mol), defined as
0.06, qAl = 1.39). The only major and significant difference
between I and I′ is the Gibbs enthalpy of formation (computed
at the gas-phase PBE-D3(BJ) level), which is nearly two times
lower for I with respect to I′ (ΔGf(298.15 K)= −17 and −28
kcal/mol for I and I′, respectively). The higher thermodynamic
stability of I′ vs I purportedly arises from the lower steric
hindrance in I′ compared to I.
ΔEint = ΔEPauli + ΔEelectrostat + ΔEorbital + ΔEdispersion
the terms standing respectively for Pauli repulsion (+99.5 kcal/
mol), electrostatic (−91.2 kcal/mol), orbital (−50.4 kcal/mol),
and dispersion (−11.6 kcal/mol) interaction energy compo-
nents. It must be pointed out here that the EDA suggests that
the orbital interaction alone is not exclusively responsible for
the stabilization of the bond. Very much in line with Frenking’s
previous statements,17 the Al−Ccarbene bond in I can be
described as an electrostatically supported donor−acceptor
bond. Indeed, among all attractive energy components obtained
by EDA, the electrostatic is the strongest: ΔEelectrostat > ΔEorbital
> ΔEdispersion. In the case of model I′ wherein the tBu fragments
have been replaced by Me, EDA leads to a similar picture of the
attractive terms ΔEorbital (−44.4 kcal/mol), ΔEdisp (−6.8 kcal/
mol), and ΔEelectrostat (−90.1 kcal/mol), thus dominated by
electrostatics. The value of ΔEint (−55.3 kcal/mol) in I′ differs
only by 2 kcal/mol with that for I.
DFT-Estimated Reaction Profile for the Formation of
Salt II/II′ from the Corresponding Model Adduct I/I′ and
Al2Me6. The formation of model salt II/II′ was considered to
proceed through the direct reaction of dimeric AlMe3, i.e.,
Al2Me6, with the corresponding Al-NHC adduct I/I′ [I′, model
of adduct (IMe)AlMe3]. The two energy profiles are depicted
in Figure 3, while the corresponding model structures are
Yang’s noncovalent interaction visualization method18 based
on the discrimination of the reduced density gradient offers a
good intuitive way to visualize the importance of noncovalent
interactions (NCIs) because they can be readily sorted into two
classes, attractive NCIs and either repulsive or van der Waals
NCIs. The reduced gradient method applied to the electron
density associated with I produced the isosurface plot depicted
in Figure 2. Regarding the Al−Ccarbene bond, the large red
Figure 3. Gibbs energy profiles (in kcal/mol, T = 298.15 K) for the
formation of salt II (a) and fictitious salt II′ (b).
shown in Figure 4 and Figures S21−23. The reaction was DFT-
investigated by standard methods (PBE-D3(BJ)) for model I
and the less sterically bulky I′. The systems were considered in
the gas phase in order to evaluate the neat thermodynamics
that rule the process. Although the deprotonation process I′ in
the presence of Al2Me6 is fictitious, it served as a reference case
to highlight the peculiar reactivity of I. The reaction of I with
Al2Me6 first proceeds with the formation of reactant complex
RC-I at little energetic cost (+1.7 kcal/mol), in which adduct I
is already dissociated into AlMe3 and ItBu fragments that are
stabilized by the incoming Al2Me6 through noncovalent
interactions. Such a ready formation of RC-I, and thus
dissociation of adduct I in the presence of Al2Me6, is consistent
with the experimental data. From RC-I, a C−H bond cleavage
of the Al(μ-CH3)Al methyl group (that interacts with the
AlMe3 and ItBu fragments from the dissociation of 1) occurs via
a low barrier transition state TS-I (7 kcal/mol) to afford salt II,
consisting of Me3Al(μ3-CH2)(AlMe2)2(μ2-CH3)− and ItBu-H+
anions, with an overall exoergonic process (ΔG° = −6.7 kcal/
mol at T = 298.15 K). In TS-I, the activated H atom, located
midway between the two carbon centers (at ca. 1.45 Å), is
clearly the one abstracted by the basic carbenic carbon atom via
a typical proton exchange reaction. The overall reaction profile
features an activation free Gibbs energy of roughly 9 kcal/mol,
which is certainly consistent with a reaction occurring at room
temperature, especially in neat AlMe3. Alternatively, a
Figure 2. ADFview2013 plots of noncovalent interaction (NCI)
regions18 materialized by reduced density gradient isosurfaces (cutoff
value s = 0.02 au, ρ = 0.05 au) colored according to the sign of the
signed density λ2ρ (red and blue colors are associated with negatively
and positively signed terms) for gas-phase relaxed singlet ground state
models of I (a) and I′ (b). The “through” within the red-colored
isosurface is associated with the covalent component of the C(NHC)−Al
bond.
isosurface of attractive NCIs located in between the two atoms
contains a “covalent through”, which provides an intuitive
illustration of how noncovalent interactions support the dative
bond in I.
For comparison, the model compound of the much less
sterically hindered (IMe)AlMe3 (I′, IMe = 1,3-dimethylimida-
zol-2-ylidene) was also analyzed within the NBO Lewis-
structural framework. It features an Al−Ccarbene bond (2.105 Å)
in line with other reported NHC-AlMe3 Lewis adducts and thus
0.06 Å shorter than that for adduct I. However, the bonding
features of the Al−Ccarbene in I′ are quite similar to those of I, as
reflected by an identical or nearly identical Wiberg bond index
(0.51) and natural population analysis (NPA) charges (qC =
D
DOI: 10.1021/acs.organomet.6b00159
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 4. Optimized geometries for the main structures involved in the reaction profile (depicted in Figure 3a) resulting in the formation of salt II
from adduct I and Al2Me6. Distances are in Å.
mechanism involving the complete dissociation of adduct 1 and
a subsequent interaction of the incoming Al2Me6 with the
formed ItBu and AlMe3 fragments seems less favored
energetically given the Gibbs enthalpy for the dissociation of
adduct 1 (ΔG(298.15 K) = 17 kcal/mol, computed at the gas-
phase PBE-D3(BJ) level). Although the latter would also be
compatible with experimental data (which does not rule out the
dissociation of adduct 1 on its own), an Al2Me6-promoted
dissociation of adduct (ItBu)AlMe3 is energetically preferred on
the basis of DFT calculations.
species have been structurally characterized and typically
incorporate at least an M-(μ-CH2)-Al motif, where M is, for
the most part, a high oxidation state early transition/rare-earth
metal center.21 In particular, akin to Schrock-type carbene
complexes, this class of heterometallic species is well known to
readily methylenate carbonyls, including esters in some
instances. To our knowledge, the presently studied AlMe3-
stabilized Al-CH2 anion (in salts 2−4) is the first well-
characterized homometallic Al(III) methylene anion, and its
reactivity toward carbonyl species was thus studied.
−
In contrast, for the reaction involving the less sterically
hindered model I′ no reactant complex akin to RC-I could be
isolated on the potential map. Instead, calculations systemati-
cally led to the formation of Ad-I′ (Figure S21), which bears an
intact Al−Ccarbene bond (2.101 Å). Even though transition state
TS-I′ could be isolated on the energy surface map (Figure
S22), its direct affiliation with Ad-I′ was not established.
Anyhow, the high barrier of activation necessary to perform the
transformation of Ad-I′ into anion II′ (ΔG0≠ = +22.8 kcal/
mol) combined with the endergonic character of the reaction
(ΔG0 = +9 kcal/mol) strongly disfavors the formation of salt
II′ (Figure S22) from adduct I′ and Al2Me6.
Thus, in agreement with experimental observations, the
present calculations rationalize that the deprotonation reactivity
of model adduct I toward Al2Me6 can be attributed to the steric
cluttering around the carbene carbon, which destabilizes the
formed (NHC)AlMe3 Lewis pair adduct. Steric hindrance
kinetically and thermodynamically favors a deprotonation
reaction with the carbene moiety acting as a Brønstedt base.
Reactivity of the Aluminum Methylene Complexes 2
and 3 with Carbonyl Substrates. Apart from their structural
interest, metal methylene/methylidene (CH22−) complexes
may be useful entities for the functionalization of unsaturated
organics and have most notably been exploited in olefin
metathesis and methylenation of carbonyl substrates. Since
Tebbe’s initial studies on Ti(IV)/Al(III) methylidene sys-
tems,8,19,20 various Al-containing heterometallic methylidene
Salts 2 and 3 were found to readily react with aldehyde and
ketone substrates to produce the corresponding methylene
products with good to quantitative conversions, thus showing
the Me3Al(μ3-CH2)(AlMe2)2(μ2-CH3)− anion to be an effective
CH22− transfer agent under the studied conditions. In all cases,
salts 2 and 3 behaved identically and 1H NMR data agree with
the spectator role of the ItBu-H+/StBu-H+ cation, as expected
for these methylation reactions. For some substrates, the
addition of an external Lewis base such as tetrahydrofuran
(THF) was required to achieve effective methylenation. The
results are compiled in Table 1. Thus, the reaction of
benzaldehyde/benzophenone with 1 equiv of 2 or 3
(CD2Cl2, room temperature, 15 min) led to the immediate
and quantitative formation of styrene and 1,1′-diphenyl-
1
ethylene, respectively, as deduced from H NMR data. Such a
functionalization may also be performed with enolizable and/or
sterically bulky ketones such as acetophenone and 1-phenyl-2-
methyl-1-propanone, although a lower conversion to the
desired products was observed in these cases (66% and 70%
conversion). Interestingly, despite the presence of eight Al-Me
groups in anion Me3Al(μ3-CH2)(AlMe2)2(μ2-CH3)−, which are
well-known moieties for the methylation of carbonyls, all
methylenation reactions proceeded with little to no formation
of methylation products. Unlike Tebbe’s reagent, anion
Me3Al(μ3-CH2)(AlMe2)2(μ2-CH3)− was found to be much
less reactive toward ester methylenation, as reflected by the low
conversion of benzyl benzoate to the corresponding methyl-
E
DOI: 10.1021/acs.organomet.6b00159
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Table 1. Methylenation of Various Carbonyl Substrates with
the Al Methylene Species 2 and 3
substrate
time (h)
T (°C)
RT
conversion (%)
b
R = Ph, R′ = H
0.25
0.25
1
>95
>95
50
ac
,
R = R′ = Ph
RT
b
R = Me, R′ = Ph
R = Pr, R′ = Ph
R = Ph, R′ = OPh
−30 to RT
−30 to RT
RT
bc
,
i
1
70
b
12
18
a
b
Salt 2 was used for the methylenation reaction. Salt 3 was used for
the methylenation reaction. Reaction carried out in the absence of
THF.
c
Figure 6. ETS-NOCV22 deformation densities Δρ and their
stabilization energy ΔEorb (in kcal/mol) for the interaction of prepared
fragments of anion Me3Al(μ3-CH2)(AlMe2)2(μ2-CH3)−. Red- and
blue-colored isosurfaces materialize regions where charge density
depletion and buildup occur, respectively: electron density transfer
operates from red- to blue-colored areas upon bonding. Deformation
density isosurface contour was set to 0.005 e/bohr3.
ation product (1-(benzyloxy)vinyl)benzene (CD2Cl2, room
temperature, 12 h, 18% conversion).
Bonding Characteristics of Anion Me3Al(μ3-CH2)-
(AlMe2)2(μ2-CH3)−. As experimentally shown above, anion
Me3Al(μ3-CH2)(AlMe2)2(μ2-CH3)− may preferentially methyl-
enate (rather than methylate) carbonyls, indicating that the
reaction of the Al-CH2-Al methylene moiety with such
unsaturated organics is kinetically favored over that of the Al-
Me and Al-Me-Al groups. DFT analysis of the natural charges
indicates that the Al-CH2 carbon atom (C9, Figure 5) bears an
Using the A3B− formula and decomposing it into an interaction
between A3 and B2− so-called prepared moieties, where A3
+
+
stands for the Al2Me5 and AlMe3 dyad, and B2− for CH22−, it
can readily be concluded that the shortest Al2−C9 bond results
from the buildup of a σ bond, whereas the longer Al1−C9 and
Al3−C9 bonds arise from interactions of the two Al(III)
centers Al1 and Al3 with electrons of the shared p orbital at C9:
each such individual Al−p interaction represents about a fourth
of the orbital interaction energy responsible for the Al2−C9
bond. NBO analysis confirms this bonding feature for Al1−C9
and Al2−C9 bonds (Figure S23). The electron population for
Al3−C9 lies under the standard occupation population
threshold of 1.5 electron to be accounted for in the inferred
Lewis structure: ψAl1−C9 (1.82 e) = 0.36(sp3.5)Al + 0.93(sp25)C9,
ψAl2−C9 = 0.40(sp2.7)Al2 + 0.92(sp1.2)C9. Further, NCI analysis of
Me3Al(μ3-CH2)(AlMe2)2(μ2-CH3)− was also performed (Fig-
ure S24), indicating that noncovalent interactions also play a
role in the stability of such a polynuclear anionic aggregate.
+
Figure 5. Gas-phase singlet ground state geometry of anion Me3Al(μ3-
CH2)(AlMe2)2(μ2-CH3)− (left) and its associated electrostatic
potential map drawn on a density isosurface (0.04 e/Å3) computed
at the ZORA-PBE-D3(BJ)/all-electron TZP level. Distances are in Å.
SUMMARY AND CONCLUSION
NPA charge (−1.58) that is 0.4 unit more negative than those
of the Al-Me carbons (average value −1.17), consistent with an
enhanced nucleophilicity of the Al-CH2 group toward carbonyl
functions.
■
Sterically bulky NHCs such as ItBu, StBu, and C6-tBu may
behave as Brønstedt bases in the presence of excess Al2Me6
(dimer of AlMe3) to afford an unusual polynuclear AlMe3-
stabilized Al-CH2− anion through a formal deprotonation of an
Al2Me6 unit. In contrast, no such reactivity is observed with the
less sterically bulky IDipp and IMes carbenes, for which only
classical Lewis base/acid chemistry was observed, with the
isolation of robust (NHC)AlMe3 adducts found to be stable in
AlMe3. According to experimental and computational data, the
observed deprotonation or C−H activation reactivity at AlMe3
thus arises from severe steric congestion in the corresponding
(NHC)AlMe3 Lewis adducts. Such destabilization favors
dissociation with the decoordinated fragments that may
subsequently deprotonate Al2Me6: in that sense, such reactivity
is akin to small-molecule activation by FLP-type systems. This
is best substantiated by the DFT-estimated reaction profile for
the ItBu/AlMe3 system, with the reaction between (ItBu)AlMe3
and Al2Me6 proceeding via an initial (and low energy cost)
dissociation of the sterically bulky (ItBu)AlMe3 adduct. The
Although several discrete metal-supported species containing
a pentacoordinate methylene carbon are known, this bonding
scheme has rarely been thoroughly assessed through
computations. The bonding of the methylene moiety in
anion Me3Al(μ3-CH2)(AlMe2)2(μ2-CH3)− was studied using a
fragment analysis point of view. In this regard, such an anion
2−
may be seen as an sp2-hybridized CH2 dianion stabilized by
the Lewis acidic moieties Al2Me5+ and AlMe3, formally resulting
in a pentacoordinate methylene carbon (Figure 6, top).
Interatomic Al−C distances range from 2.007 to 2.125 Å (in
agreement with experimental data), with the Al2−C9 bond
being shortest. The Wiberg indices, which follow the order
Al2−C9 (0.51) > Al1−C9 (0.44) > Al3−C9 (0.37), indicate
unequal organometallic bonding to C9. Again, the NOCV-EDA
method4 provides a clear picture of the bonding situation in
anion Me3Al(μ3-CH2)(AlMe2)2(μ2-CH3)− (Figure 6, bottom).
F
DOI: 10.1021/acs.organomet.6b00159
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
species 2 in a pure form (as a colorless liquid) on the basis of 1H and
13C NMR data, X-ray crystallographic data, and elemental analysis.
Due to the dynamic behavior of the aluminate anion Me3Al(μ3-
CH2)(AlMe2)2(μ2-CH3)− in solution at room temperature, 13C{1H}
and DEPT NMR were recorded at room temperature and at −30 °C;
these data are consistent with the aluminate anion in complex 2
essentially retaining its solid-state structure in solution. In neat AlMe3,
in a glovebox, AlMe3 (1 mL) was added at room temperature via a
pipet to the free carbene 1,3-di-tert-butylimidazol-2-ylidene (50.0 mg,
0.28 mmol). The resulting colorless solution was vigorously stirred at
room temperature. After 24 h the excess AlMe3 was evaporated under
vacuum and the residue was washed with pentane (2 × 5 mL) to yield
a colorless oil. The latter was dried in vacuo to afford the ionic liquid
compound 2 as a colorless oil (105 mg, 96% yield). Anal. Calcd for
C20H47Al3N2 (396.56): C 60.57; H 11.95; N 7.06. Found: C, 60.82; H
11.63. 1H NMR (400 MHz, CD2Cl2): δ −0.84 (s, 2H, AlCH2), −0.72
subsequent deprotonation of Al2Me6 (occurring at a bridging
Me group) by ItBu may then readily occur to yield salt 2 as the
kinetic and thermodynamic product. On the other hand, such a
deprotonation reaction was computed to be unfavorable both
kinetically and thermodynamically with the much less sterically
bulky and thus more robust model adduct (IMe)AlMe3, in line
with the importance of adduct dissociation for the reaction to
proceed.
Anion Me3Al(μ3-CH2)(AlMe2)2(μ2-CH3)− was shown to
efficiently methylenate various carbonyl substrates such as
aldehydes and ketones, a reactivity clearly preferred over
methylation. This agrees well with the DFT bonding analysis
concluding on a more nucleophilic Al-CH2-Al (vs Al-Me
moieties) group.
4
(br s, 24H, Al-CH3), 1.70 (s, 18H, CH3-tBu), 7.49 (d, JHH = 1.8 Hz,
EXPERIMENTAL SECTION
1H, NCHCHN), 8.20 (d, 4JHH = 1.8 Hz, 1H, NCHN). 1H NMR (400
MHz, CD2Cl2, 243 K): δ −1.04 (br s, 2H, AlCH2), −0.88 (br s, 24H,
■
General Procedures. All experiments were carried out under N2
using standard Schlenk techniques or in an MBraun Unilab glovebox.
Toluene, pentane, and dichloromethane were collected after going
through drying columns and stored over activated molecular sieves (4
Å) for 24 h in a glovebox prior to use. Tetrahydrofuran was distilled
over Na/benzophenone and stored over activated molecular sieves (4
Å) for 24 h in a glovebox prior to use. CD2Cl2, C6D6, d8-toluene, and
d8-THF were distilled from CaH2, degassed under a N2 flow, and
stored over activated molecular sieves (4 Å) in a glovebox prior to use.
All deuterated solvents were obtained from Eurisotop (CEA, Saclay,
France). All chemicals were purchased from Aldrich or Strem
Chemicals. Elemental analysis for all compounds was performed at
t
AlMe), 1.74 (s, 18H, Bu), 7.42 (s, 1H, CH), 8.15 (s, 1H, CH).
13C{1H} NMR (75 MHz, CD2Cl2): δ −5.9 (AlCH2), −4.7 (AlCH3),
30.1 (CH3-tBu), 59.3 (C(CH3)3), 61.5 (C(CH3)3), 121.1
(NCHCHN), 129.5 (NCHN). 13C{1H} NMR (100 MHz, CD2Cl2,
243 K): δ −6.4 (AlCH2), −5.3 (AlMe), −4.5 (AlMe), 30.2 (tBu), 61.4
(tBu), 121.2 (CH), 129.4 (CH).
[StBu-H][Me3Al(μ3-CH2)(AlMe2)2(μ2-CH3)] (3). In a glovebox,
AlMe3 (2 mL) was added at room temperature via a pipet to the
free carbene 1,3-di-tert-butylimidazolin-2-ylidene (367 mg, 2.01
mmol). The resulting colorless, cloudy solution was vigorously stirred
at room temperature. Over the course of 2 h, a massive precipitation of
a colorless solid occurred. The resulting suspension was then filtered
through a glass frit, and the residue was washed three times with
pentane (3 × 10 mL) to yield a colorless oil. The latter was dried in
vacuo to afford the ionic liquid compound 3 as an analytically pure
solid (606 mg, 79% yield). Anal. Calcd for C20H47Al3N2 (398.56): C
́
the Service de Microanalyse of the Universite de Strasbourg
(Strasbourg, France). 1,3-Di-tert-butylimidazol-2-ylidene (ItBu) was
purchased from TCI Europe and was used as received. 1,3-Bis(2,6-
diisopropylphenyl)imidazol-2-ylidene (IDipp), 1,3-di-tert-butylimida-
zolin-2-ylidene (StBu), and 1,3-di-tert-butyl-3,4,5,6-tetrahydripyrimi-
din-2-ylidene (C6-tBu) were synthesized according to literature
procedures.23−25 The Al-NHC adduct (ItBu)AlMe3 (1) was also
synthesized according to a literature procedure. For all compounds,
the combination of 13C and DEPT NMR data allowed the
unambiguous assignment of all resonances. Caution: AlMe3 is a
hazardous and highly pyrophoric colorless liquid that should be handled
with great care under a controlled atmosphere devoid of H2O and O2. All
manipulations involving AlMe3 (whether in solution or neat) were
carried out in a N2-filled (<1 ppm of O2, <1 ppm of H2O) drybox.
Theoretical Investigations. Theoretical studies were carried
within the SCM-ADF2013.00126 package using the methods of density
functional theory with two distinct functionals, that is, the meta-GGA
TPSS27-D3(BJ)28 and the GGA PBE29-D3(BJ),28 that include the
long-range D3 dispersion correction developed by Grimme et al. with
the Becke−Johnson damping function variant (noted BJ). In all cases
relativistic effects were accounted for by scalar ZORA using ad hoc all-
electron singly polarized triple-ζ basis sets for all atoms.30 Even though
both methods produced very good reproduction of the experimental
geometry, the above-mentioned meta-GGA method was exclusively
used for the investigation of electronic structures and properties,
whereas the GGA method was used for the computation of reactant
complexes, transition states, and reaction thermodynamics (vide
supra). The Rauk−Ziegler energy decomposition analysis16 (EDA)
framework implemented in ADF was used throughout this article and
coupled whenever necessary to the extended transition-state natural
orbitals for chemical valence22 analysis (EDA-NOCV). Noncovalent
interaction regions18 were computed and drawn using ADFview2013,
and natural bond orbital analysis14 was performed with the
GENNBOv06 program embedded within ADF.
1
60.27; H 12.39; N 7.03. Found: C 60.01; H 12.23; N 6.89. H NMR
(300 MHz, CD2Cl2): δ −0.83 (br s, 26H, AlCH2 and AlMe3), 1.45 (s,
18H, CH3-tBu), 4.01 (s, 4H, NCH2CH2N), 7.53 (s, 1H, NCHN).
13C{1H} NMR (100 MHz, CD2Cl2): δ −5.8 (AlCH2), −4.6 (AlMe3),
28.3 (CH3-tBu), 45.8 (NCH2CH2N), 57.9 (C(CH3)3), 151.0
(NCHN).
[C6-tBu-H][Me3Al(μ3-CH2)(AlMe2)2(μ2-CH3)] (4). The salt species
4 was prepared following an identical procedure to that used for the
synthesis of 3 but using 1,3-di-tert-butyltetrahydropyrimidin-2-ylidene
(C6-tBu, 72.7 mg, 0.37 mmol) as a carbene source. The ionic liquid 4
was isolated as an analytically pure pale yellow oil (152 mg, 99% yield).
Anal. Calcd for C21H51Al3N2 (412.59): C 61.13; H 12.46; N 6.79.
Found: C 61.31; H 12.54. 1H NMR (300 MHz, CD2Cl2): δ −0.83 (br
3
s, 17H, AlMe3), 1.46 (s, 18H, CH3-tBu), 2.10 (q, JHH = 5.7 Hz, 2H,
NCH2CH2CH2N), 3.44 (t, 3JHH = 5.7 Hz, 4H, NCH2CH2CH2N), 7.83
(s, 1H, NCHN). 13C{1H} NMR (100 MHz, CD2Cl2): δ −5.8
(AlCH2), −4.7 (AlMe3), 20.1 (NCH2CH2CH2N), 27.9 (CH3-tBu),
40.3 (NCH2CH2CH2N), 62.1 (C(CH3)3), 146.2 (NCHN).
Typical Procedure for the Methylenation of Carbonyl
Substrates with Salts 2 and 3. All runs were carried out on an
NMR scale. To a solution of the appropriate carbonyl substrate (0.1
mmol) in CD2Cl2 (0.2 mL) was added a CD2Cl2 solution (0.5 mL) of
the salt species 2 and 3 (0.1 mmol). For the methylenation of
acetophenone, 3-methyl-2-phenylbutanone, and benzyl benzoate, THF
(2 equiv) was also added along with the carbonyl substrate. The
mixture was stirred for 1 h at room temperature to yield the
1
corresponding methylenation products as deduced by H NMR and
comparison with literature data.31 Thus, the reaction of benzaldehyde,
acetophenone, benzophenone, 3-methyl-2-phenyl-1-propanone, and
benzyl benzoate with salt 2 or 3 led to the respective formation of
styrene (>95% conversion), 2-phenylpropene (66% conversion), 1,1′-
diphenylethylene (>95% conversion), 3-methyl-2-phenyl-1-butene
(70% conversion), and (1-(benzyloxy)vinyl)benzene (18% conver-
sion). In the case of 3-methyl-2-phenylbutanone, the identity of the
[ItBu-H][Me3Al(μ3-CH2)(AlMe2)2(μ2-CH3)] (2). In a glovebox, a
dichloromethane solution (1 mL) of AlMe3 (5 equiv, 2.00 mmol,
158.0 mg) was added at room temperature via a pipet to a stirring
dichloromethane solution of (ItBu)AlMe3 (1, 100 mg, 0.40 mmol).
The resulting colorless solution was stirred for 20 h at room
temperature, and the volatiles were removed in vacuo to afford salt
G
DOI: 10.1021/acs.organomet.6b00159
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
methylenation product 3-methyl-2-phenyl-1-butene was also con-
firmed by GC-MS analysis of the crude reaction mixture.
(7) Schmitt, A.-L.; Schnee, G.; Welter, R.; Dagorne, S. Chem.
Commun. 2010, 46, 2480−2482.
(8) (a) Scott, J.; Mindiola, D. J. Dalton Trans. 2009, 8463−8472.
(b) Hartley, R. C.; Li, J.; Main, C. A.; McKiernan, G. J. Tetrahedron
2007, 63, 4825−4864. (c) Kratsch, J.; Roesky, P. W. Angew. Chem., Int.
Ed. 2014, 53, 376−383. (d) Sada, M.; Komagawa, S.; Uchiyama, M.;
Kobata, M.; Mizuno, T.; Utimoto, K.; Oshima, K.; Matsubara, S. J. Am.
Chem. Soc. 2010, 132, 17452−17458.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.organo-
met.6b00159.
■
S
(9) Giuseppone, N.; Schmitt, J.-L.; Allouche, L.; Lehn, J.-M. Angew.
Chem., Int. Ed. 2008, 47, 2235−2237.
NMR data for all compounds, monitoring reactions, and
methylenation of carbonyl substrates (Figures S1−S19);
computational details including ETS-NOCV deformation
densities for adduct 1; molecular structures of models I′,
Ad-I′, TS-I′, and II′ (Figures S20−S25); atom
coordinates and list of frequencies and energies for all
modeled species (PDF)
(10) A complete VT NMR analysis along with the determination of
the associated parameters was precluded due to the poor stability of
the (ItBu)AlMe3/Al2Me6 mixture above room temperature.
(11) (a) Shih, W.-C.; Wang, C.-H.; Chang, Y.-T.; Yap, G. P. A.; Ong,
T.-G. Organometallics 2009, 28, 1060−1067. (b) Wu, M.; Gill, A. M.;
Yunpeng, L.; Falivene, L.; Yongxin, L.; Ganguly, R.; Cavallo, L.; Garcia,
F. Dalton Trans. 2015, 44, 15166−15174.
(12) Adduct (StBu)AlMe3, synthesized by reaction of StBu with
AlMe3, was independently NMR characterized: Schnee, G.; Dagorne,
S., unpublished results.
Optimized Cartesian coordinates (XYZ)
(13) For characterization data of adducts (ItBu)GaMe3 and (ItBu)
InMe3, see ref 5a.
AUTHOR INFORMATION
Corresponding Author
*E-mail: dagorne@unistra.fr.
Notes
■
(14) (a) Weinhold, F. In Encyclopedia of Computational Chemistry;
von Rague-Schleyer, P., Allinger, N. L., Clark, T., Gasteiger, J.,
́
Kollman, P. A., Schaefer, H. F., Schreiner, P. R., Eds.; John Wiley &
Sons: Chichester, UK, 1998; Vol. 3, pp 1792−1811. (b) Weinhold, F.;
Landis, C. R. Valency and Bonding, a Natural Bond Orbital Donor-
Acceptor Perspective; University Press: Cambridge, UK, 2005.
(15) For a representative example of NOCV-EDA analysis of metal−
NHC bonding interactions, see: Tonner, R.; Heydenrych, G.;
Frenking, G. Chem. - Asian J. 2007, 2, 1555−1567.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors acknowledge the CNRS and the University of
Strasbourg for financial support. G.S. is grateful to the French
Ministry of Higher Education and Research for a Ph.D.
fellowship.
(16) Ziegler, T.; Rauk, A. Inorg. Chem. 1979, 18, 1755−1759.
(17) For examples of NOCV-EDA analysis of Al−NHC bonding
interactions, see: (a) Holtzmann, N.; Stasch, A.; Jones, C.; Frenking,
G. Chem. - Eur. J. 2013, 19, 6467−6479. (b) Holtzmann, N.; Stasch,
A.; Jones, C.; Frenking, G. Chem. - Eur. J. 2011, 17, 13517−13525.
(18) (a) Johnson, E.; Keinan, S.; Mori-Sanchez, P.; Contreras-Garcia,
J.; Cohen, A. J.; Yang, W. J. Am. Chem. Soc. 2010, 132, 6498−6506.
(b) Contreras-Garcia, J.; Johnson, E. R.; Keinan, S.; Chaudret, R.;
Piquemal, J. P.; Beratan, D. N.; Yang, W. J. Chem. Theory Comput.
2011, 7, 625−632.
REFERENCES
■
(1) Selected reviews/books on NHCs and derived metal complexes:
(a) Arduengo, A. J., III Acc. Chem. Res. 1999, 32, 913−921.
(b) Bourissou, D.; Guerret, O.; Gabbaï, F.; Bertrand, G. Chem. Rev.
2000, 100, 39−92. (c) Herrmann, W. A. Angew. Chem., Int. Ed. 2002,
́
41, 1290−1309. (d) Cesar, V.; Bellemin-Laponnaz, S.; Gade, L. H.
Chem. Soc. Rev. 2005, 33, 619−636. (e) N-Heterocyclic Carbenes in
Synthesis; Nolan, S. P., Ed.; Wiley-VCH: Weinheim, 2006. (f) Pugh,
D.; Danopoulos, A. A. Coord. Chem. Rev. 2007, 251, 610−641. (g) de
(19) For a review on Tebbe’s reagent, its legacy, and related metal
complexes, see ref 8a.
(20) Thompson, R.; Nakamaru-Ogiso, E.; Chen, C.-H.; Pink, M.;
Mindiola, D. J. Organometallics 2014, 33, 429−432.
́
Fremont, P.; Marion, N.; Nolan, S. P. Coord. Chem. Rev. 2009, 253,
862−892. (h) Arnold, P. L.; Casely, I. J. Chem. Rev. 2009, 109, 3599−
3611. (i) Bellemin-Laponnaz, S.; Dagorne, S. Chem. Rev. 2014, 114,
8747−8774. (j) N-Heterocyclic carbenes: Effective Tools in Organo-
metallic Synthesis; Nolan, S. P., Ed.; Wiley-VCH: Weinheim, 2014.
(2) (a) Wang, Y.; Robinson, G. H. Dalton Trans. 2012, 41, 337−345.
(b) Roesky, H. W. J. Organomet. Chem. 2013, 730, 57−62.
(21) Selected examples: (a) Dietrich, H. M.; Grove, H.; Tornroos, K.
̈
W.; Anwander, R. J. Am. Chem. Soc. 2006, 128, 1458−1459.
(b) Zimmermann, M.; Rauschmaier, D.; Eichele, K.; Tornroos, K.
̈
W.; Anwander, R. Chem. Commun. 2010, 46, 5346−5348. (c) Bojer,
D.; Neumann, B.; Stammler, H.-G.; Mitzel, N. W. Chem. - Eur. J. 2011,
17, 6239−6247. (d) Gerber, L. C. H.; Le Roux, E.; Tornroos, K. W.;
̈
(3) Leading reviews on FLPs and their reactivity: (a) Stephan, D. W.;
Erker, G. Angew. Chem., Int. Ed. 2015, 54, 6400−6441. (b) Stephan, D.
W.; Erker, G. Angew. Chem., Int. Ed. 2010, 49, 46−76.
(4) (a) Holschumacher, D.; Bannenberg, T.; Hrib, C. G.; Jones, P.
G.; Tamm, M. Angew. Chem., Int. Ed. 2008, 47, 7428−7432. (b) Chase,
P. A.; Stephan, D. W. Angew. Chem., Int. Ed. 2008, 47, 7433−7437.
(5) (a) Schnee, G.; Nieto Faza, O.; Specklin, D.; Jacques, B.;
Anwander, R. Chem. - Eur. J. 2008, 14, 9555−9564. (e) Scott, J.; Fan,
H.; Wicker, B. F.; Fout, A. R.; Baik, M.-H.; Mindiola, D. J. J. Am. Chem.
Soc. 2008, 130, 14438−14439. (f) Wei, P.; Stephan, D. W.
Organometallics 2003, 22, 1992−1994.
(22) (a) Mitoraj, M. P.; Michalak, A.; Ziegler, T. J. Chem. Theory
Comput. 2009, 5, 962−975. (b) Mitoraj, M. P.; Michalak, A.; Ziegler,
T. Organometallics 2009, 28, 3727−3733.
Karmazin, L.; Welter, R.; Silva Lop
2015, 21, 17959−17972. (b) Hernan
́
ez, O.; Dagorne, S. Chem. - Eur. J.
(23) Jafapour, L.; Stevens, E. D.; Nolan, S. P. J. Organomet. Chem.
2000, 606, 49−54.
́
-Gomez, A.; Armstrong, D. R.;
́
Kennedy, A. R.; Hevia, E. Chem. Sci. 2015, 6, 5719−5728. (c) Day, B.
M.; Pugh, T.; Hendriks, D.; Fonseca Guerra, C.; Evans, D. J.;
Bickelhaupt, F. M.; Layfield, R. A. J. Am. Chem. Soc. 2013, 135,
13338−13341. (d) Musgrave, R. A.; Turbervill, R. S. P.; Irwin, M.;
Goicoechea, J. M. Angew. Chem., Int. Ed. 2012, 51, 10832−10835.
(6) For reviews including NHC group 13 metal complexes:
(24) Denk, M. K.; Gupta, S.; Brownie, J.; Tajammul, S.; Lough, A. J.
Chem. - Eur. J. 2001, 7, 4477−4486.
(25) Arentsen, K.; Caddick, S.; Cloke, F. G. N. Tetrahedron 2005, 61,
9710−9715.
(26) (a) te Velde, G.; Bickelhaupt, F. M.; Baerends, E. J.; Fonseca
Guerra, C.; van Gisbergen, S. J. A.; Snijders, J. G.; Ziegler, T. J.
Comput. Chem. 2001, 22, 931−967. (b) In Amsterdam Density
Functional ADF2013; SCM: Theoretical Chemistry, Vrije Universiteit:
Amsterdam, The Netherlands, 2013.
́
(a) Fliedel, C.; Schnee, G.; Aviles, T.; Dagorne, S. Coord. Chem. Rev.
2014, 275, 63−86. (b) Willans, C. E. Organomet. Chem. 2010, 36, 1−
28. (c) Kuhn, N.; Al-Sheikh, A. Coord. Chem. Rev. 2005, 249, 829−
857.
H
DOI: 10.1021/acs.organomet.6b00159
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(27) Tao, J.; Perdew, J. P.; Staroverov, V. N.; Scuseria, G. E. Phys.
Rev. Lett. 2003, 91, 146401-1−141406-4.
(28) Grimme, S.; Ehrlich, S.; Goerigk, L. J. Comput. Chem. 2011, 32,
1456−1465.
(29) (a) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1996,
77, 3865−3868. (b) Adamo, C.; Barone, V. J. Chem. Phys. 1999, 110,
6158−6170.
(30) (a) van Lenthe, E.; van Leeuwen, R.; Baerends, E. J.; Snijders, J.
G. Int. J. Quantum Chem. 1996, 57, 281−293. (b) Van Lenthe, E.;
Baerends, E. J. J. Comput. Chem. 2003, 24, 1142−1156.
́
(31) (a) Lebel, H.; Davi, M.; Díez-Gonzalez, S.; Nolan, S. P. J. Org.
Chem. 2007, 72, 144−149. (b) Adam, W.; Bosio, S. G.; Turro, N. J.;
Wolff, B. T. J. Org. Chem. 2004, 69, 1704−1715. (c) de Nanteuil, F.;
Serrano, E.; Perrotta, D.; Waser, J. J. Am. Chem. Soc. 2014, 136, 6239−
6242.
I
DOI: 10.1021/acs.organomet.6b00159
Organometallics XXXX, XXX, XXX−XXX