Activation of N-heterocyclic carbenes by {BeH2} and {Be(H)(Me)} fragments

Article abstract of DOI:10.1021/om501314g

A stable three-coordinate dimethylberyllium species coordinated by the 1,3-bis(2,4,6-trimethylphenyl)imidazol-2-ylidene (IMes) ligand is readily converted to the corresponding methylhydrido derivative through metathetical reaction with phenylsilane. Attempts to synthesize the corresponding molecular dihydrides are, however, unsuccessful and result in ring opening of an IMes ligand through hydride transfer to the donor carbon atom and the consequent formation of a heterocyclic beryllium organoamide. In agreement with previous calculations, we suggest that this process occurs via a Schlenk-type equilibration process and formation of a four-coordinate bis-NHC beryllium dihydride. These species are not observed, however, as the steric pressure exerted by coordination of the two sterically demanding IMes ligands is sufficient to induce hydride transfer. The latter deduction is supported by the observation that a similar ring-opened product, but derived from methyl and hydride transfer, is available through the introduction of a further equivalent of IMes to the isolated beryllium methyl hydride species. In the latter case the ring-opening process is more facile, which we ascribe to the increased steric pressure achieved upon the formation of four-coordinate beryllium. In a further striking reaction under more forcing thermal conditions, the carbene carbon center of an IMes ligand is observed to be completely eliminated with selective formation of a three-coordinate diamidoberyllium species.

Full text of DOI:10.1021/om501314g

Article
pubs.acs.org/Organometallics
Activation of N‑Heterocyclic Carbenes by {BeH2} and {Be(H)(Me)}
Fragments
Merle Arrowsmith, Michael S. Hill,* and Gabriele Kociok-Kohn
̈
Department of Chemistry, University of Bath, Claverton Down, Bath, BA2 7AY, U.K.
S
* Supporting Information
ABSTRACT: A stable three-coordinate dimethylberyllium
species coordinated by the 1,3-bis(2,4,6-trimethylphenyl)-
imidazol-2-ylidene (IMes) ligand is readily converted to the
corresponding methylhydrido derivative through metathetical
reaction with phenylsilane. Attempts to synthesize the
corresponding molecular dihydrides are, however, unsuccessful
and result in ring opening of an IMes ligand through hydride transfer to the donor carbon atom and the consequent formation of
a heterocyclic beryllium organoamide. In agreement with previous calculations, we suggest that this process occurs via a Schlenk-
type equilibration process and formation of a four-coordinate bis-NHC beryllium dihydride. These species are not observed,
however, as the steric pressure exerted by coordination of the two sterically demanding IMes ligands is sufficient to induce
hydride transfer. The latter deduction is supported by the observation that a similar ring-opened product, but derived from
methyl and hydride transfer, is available through the introduction of a further equivalent of IMes to the isolated beryllium methyl
hydride species. In the latter case the ring-opening process is more facile, which we ascribe to the increased steric pressure
achieved upon the formation of four-coordinate beryllium. In a further striking reaction under more forcing thermal conditions,
the carbene carbon center of an IMes ligand is observed to be completely eliminated with selective formation of a three-
coordinate diamidoberyllium species.
action with one stoichiometric equivalent of PhSiH3 was
observed to proceed with the smooth formation of the desired
methyl monohydride derivative [Me(μ-H)Be·(IPr)]2 (II),
further attempts to form the putative dihydride species were
thwarted by a previously unobserved NHC degradation
pathway in d8-THF at 80 °C over 6 h involving ring opening
by C−N bond cleavage of the imidazolylidene framework to
form the heterocyclic organoamido species III (Scheme 1).21
Subsequent to this initial report the groups of Rivard and
Radius described analogous reactions between IPr and the
aminoborane [H2B(NHDipp)] (IV)22 and 1,3-dialkyl- and 1,3-
diarylimidazolin-2-ylidenes and PhnSiH4‑n (V, n = 1, 2, 3),23
respectively, although these reactions required higher temper-
atures (100 °C) and longer reaction times (up to 3 days) to
proceed. Computational analyses by Dutton,24 Su,25 Fang,26
and Brown27 of the silane- and borane-mediated processes have
highlighted a common mechanism involving initial hydrogen
atom migration to the carbene carbon followed by ring
expansion through insertion of the main group hydride into the
ring prior to the hydrogen transfer process. Dutton and co-
workers deduced a similar mechanism for the beryllium-
centered reaction based on calculations performed on a
simplified model featuring IMes ligands (IMes = 1,3-
bis(2,4,6-trimethylphenyl)imidazol-2-ylidene) and from a start-
ing point of our initial postulate that the ring-opening reactivity
occurs via the in situ formation of a [H2Be·NHC]
INTRODUCTION
■
While the past decade has witnessed significant advances in the
chemistry of the heavier alkaline earth elements (Mg, Ca, Sr,
and Ba),1 their lightest congener, beryllium, maintains its role
as the black sheep of the family. Central to recent progress in
heavier group 2 element reactivity has been the exploitation of
well-defined molecular hydride species2 such as the μ-hydrido
magnesium and calcium β-diketiminate derivatives [HC{(Me)-
CN(2,6-iPr2C6H3}2MH(THF)n]2 (M = Mg, n = 0;3 M = Ca, n
= 14) and several magnesium hydride cluster molecules
supported by either polynucleating β-diketiminate or neutral
N-heterocyclic carbene (NHC) ligands.5−7 Although the
synthesis and structural characterization of several molecular
beryllium monohydrides has been described during the past
four decades,2,8−16 impeded by its reputation as among the
most toxic of the nonradioactive elements,17 concerted study of
the reaction chemistry of even these simplest beryllium
derivatives has been unsurprisingly sporadic. Derived from
the smallest group 2 element, however, the Be2+ cation may be
expected to display unique characteristics as a consequence of
its high charge density and resultant polarizing capacity,
variations of which have been implicated in the ability of its
heavier congeners to effect a variety of cross metathetical18 and
heterofunctionalization catalyses.1,19
Motivated by these considerations, we have previously
described attempts to synthesize an NHC-supported molecular
beryllium dihydride by silane metathesis with the three-
coordinate dimethyl species [Me2Be·(IPr)] (I, IPr = 1,3-
bis(2,6-di-isopropylphenyl)imidazol-2-ylidene).20 Although re-
Received: December 22, 2014
© XXXX American Chemical Society
A
DOI: 10.1021/om501314g
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 1
Scheme 2
Figure 1. ORTEP representations of compounds 1 (left) and 2 (right). Ellipsoids are drawn at 20% probability. Hydrogen atoms are omitted for
clarity, except for the bridging hydride atom H1 in 2. Symmetry transformations used to generate equivalent atoms in 2: #1 −x, −y, −z+1; #2 −x+1,
−y+1, −z+1.
compound.24b Their calculations suggested that this mono-
meric species would preferentially bind a second NHC prior to
the first hydrogen atom transfer step. While this latter
observation is consistent with the products derived from the
reaction stoichiometry depicted in Scheme 1 and deuterium
labeling studies described in our initial communication, further
synthetic mechanistic interrogation was hampered by the virtual
insolubility of the intermediate methylhydrido beryllium
species, II, in all but the donor solvent THF at elevated (80
°C) temperatures. In this contribution we describe the
synthesis of highly soluble analogues of compounds I−III,
which are more amenable to mechanistic analysis.
the same procedure used to obtain I (Scheme 2).20 Extraction
into toluene and filtration away from the dense white LiCl
precipitate yielded a clear yellow solution. Recrystallization
from a saturated toluene solution at −18 °C afforded a crop of
large colorless crystals of compound 1 (58% yield). The yield
could be improved to 80% by reacting the IMes precursor with
a premade stock solution of BeMe2 in toluene obtained by salt
metathesis between BeCl2 and two equivalents of methyl-
1
lithium. The H NMR spectrum of compound 1 displayed a
characteristic upfield (6H) singlet at −0.73 ppm attributed to
the beryllium-coordinated methyl groups, 0.24 ppm downfield
of the analogous resonance observed for I. Comparison with
NMR data of the free NHC precursor also confirmed the
presence of a single beryllium-coordinated NHC ligand. The
13C{1H} NMR spectrum comprised a distinctive downfield
resonance at 184.9 ppm for the carbenoid carbon as well as an
upfield signal at −3.7 ppm assigned to the beryllium-bound
RESULTS AND DISCUSSION
■
The IMes-stabilized dimethylberyllium NHC adduct 1 (IMes =
N,N′-bis(2,4,6-trimethylphenyl)imidazol-1-ylidene) was syn-
thesized by a salt metathesis route in diethyl ether, following
B
DOI: 10.1021/om501314g
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
methyl carbons (confirmed by an HMQC experiment), which
was broadened by coupling with the adjacent quadrupolar
beryllium atom. As was the case for I, compound 1 also
structural data for [Me2Be·(IPr)] (I) for comparative purposes.
Although structurally very similar, compounds I and 1
crystallize in different space groups (I monoclinic P21/n; 1
orthorhombic Pcab) as three-coordinate monomers. In both
cases the beryllium centers adopt a trigonal planar geometry
with C−Be−C angles close to 120° [I C1−Be−C28
118.07(12)°; C1−Be−C29 113.95(11)°; C28−Be−C29
127.70(12)°; 1 C1−Be−C22 113.6(2)°; C1−Be−C23
118.6(3)°; C28−Be−C29 127.8(3)°]. The BeMe2 fragment
in compound 1 subtends an angle of ca. 78° with the planar
NHC framework, the beryllium atom lying ca. 0.27 Å above the
ligand plane. In compound 1 this distance is reduced to ca. 0.12
Å, inducing a rotation of the BeMe2 fragment such that the
angle with the ligand plane becomes more acute (ca. 62°). As a
result, the N−C1−Be angles around the carbene carbon (C1)
differ greatly between the two compounds [I 131.46(11)°,
123.91(10)°; 1 127.0(2)°, 129.1(2)°]. These variations may be
rationalized as a consequence of the differing steric
compression exerted on the BeMe2 fragment by the respective
isopropyl and methyl ortho-substituents of the NHCs’ pendant
aryl groups. All other angles and bond lengths in the two
compounds are otherwise very similar. The Be−C bond lengths
[I Be−C1 1.816(2); Be−C28 1.734(2); Be−C29 1.749(2) Å; 1
Be−C1 1.820(4); Be−C22 1.736(5); Be−C23 1.739(5) Å], in
particular, are comparable to those in the only other
crystallographically characterized NHC-supported diorganober-
yllium compound [Ph2Be·{C(N(iPr)CMe)2}] [Be−Ccarbene
1.807(4); Be−CPh 1.745(4), 1.757(4) Å]28 and Robinson’s
beryllium chloride and borohydride adducts of the IPr ligand
[1.773(5), 1.765(2) Å].29
9
provided a single broad Be NMR resonance around 25 ppm,
comparable to that of the previously reported NHC-stabilized
organoberyllium compound [Ph2Be·{C(N(iPr)CMe)2}] (δ9Be
21.1 ppm).28 A H−9Be HMQC experiment showed a two-
1
9
bond correlation between this Be NMR shift and the upfield
BeCH3 protons at −0.73 ppm.
Highly air-sensitive single crystals of compound 1 suitable for
X-ray diffraction analysis were obtained from saturated toluene
solutions stored at low temperature. The structure of
compound 1 is displayed in Figure 1, and selected bond
lengths and angles are provided in Table 1, together with
Table 1. Selected Bond Lengths and Angles for Compounds
I, 1, II, and 2
a
b
c
d
I
1
II
2
Be−C1
Be−X
Be−Y
Be−H1#
C1−N1
C1−N2
N1−C2
N2−C3
1.816(2)
1.734(2)
1.749(2)
1.820(4)
1.736(5)
1.739(5)
1.822(2)
1.8002(19)
1.777(2)
1.813(2)
1.552(18)
1.630(18)
1.3608(16)
1.3622(16)
1.3858(17)
1.3878(17)
1.344(2)
1.456(15)
1.496(15)
1.3579(16)
1.3558(15)
1.3897(16)
1.3923(16)
1.340(2)
1.3555(15)
1.3568(16)
1.3887(17)
1.3853(16)
1.3430(19)
118.07(12)
113.95(11)
1.359(3)
1.353(3)
1.387(3)
1.394(3)
1.341(4)
113.6(2)
118.6(3)
C2−C3
C1−Be−X
C1−Be−Y
C1−Be−H1#
X−Be−Y
115.47(11)
107.6(7)
114.49(10)
106.9(6)
103.5(7)
105.7(6)
As was the case for I, the reaction between 1 and two
equivalents of phenylsilane in toluene at room temperature did
not yield the desired dihydridoberyllium compound but the
dimeric NHC-stabilized methylhydridoberyllium compound, 2,
together with one equivalent of the metathesis byproduct
PhSiMeH2. In contrast to the analogous IPr compound, II,
which was virtually insoluble in both noncoordinating and
coordinating hydrocarbon solvents, precluding the acquisition
of meaningful 13C NMR data, compound 2 was highly soluble
127.70(12)
127.8(3)
113.9(7)
118.6(6)
X−Be−H1#
H1−Be−H1#
Be−H1−Be#
N1−C1−Be
N2−C1−Be
N1−C1−N2
C1−N1−C2
C1−N2−C3
N1−C2−C3
N2−C3−C2
117.9(7)
119.5(6)
96.3(7)
88.6(6)
83.7(7)
91.4(6)
131.46(11)
123.91(10)
103.87(10)
111.26(10)
111.65(10)
106.80(11)
127.0(2)
129.1(2)
103.7(2)
111.5(2)
111.8(2)
106.9(2)
106.2(2)
127.27(11)
129.21(11)
103.42(11)
111.45(11)
111.89(11)
107.09(12)
106.15(12)
129.14(10)
126.85(10)
103.98(10)
111.33(11)
111.34(10)
106.73(11)
1
in C6D6 and d8-toluene. Figure 2 shows annotated H NMR
1
and H−9Be HMQC spectra of the in situ reaction after full
106.40(12)
b
106.61(11)
d
1
a
c
consumption of 1. The IMes H NMR resonances were very
X = C28, Y = C29. X = C22, Y = C23. X = C28, Y = H1. X = C22,
Y = H1.
similar to those of compound 1, with the exception of the 6H
singlet of the mesityl ortho-substituents, which underwent a
Figure 2. 1H NMR (left) and 1H−9Be HMQC (right) spectra recorded in C6D6 after completion of the reaction between 1 and two equivalents of
⧫
PhSiH3. indicates the resonances of the methylhydrido dimer, 2.
C
DOI: 10.1021/om501314g
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
downfield shift of 0.24 ppm. The characteristic 6H Be(CH3)2
singlet at −0.73 ppm was replaced with a triplet further upfield
at −1.26 ppm (3J = 1.9 Hz), integrating to three protons
relative to the IMes ligand and suggesting the retention of a
dimeric hydride-bridged compound in solution. A new broad
resonance corresponding to 1H by integration was observed to
appear concomitantly at 0.61 ppm, which was correlated with
the protons of the beryllium-bound methyl group in the
derived environment characterized by a distinctive AB spin
system at δ1H 5.02 and 4.79 ppm and a set of unsymmetrical
mesityl methyl resonances. Comparison with NMR data
recorded for the previously reported ring-expanded IPr
derivative, III, confirmed the product as the analogous IMes-
supported six-membered alkylamidoberyllium ring system
[{MesNCHCHN(Mes)CH2}Be·(IMes)] (3). Similarly to
compound III, HMQC data for compound 3 revealed coupling
between an otherwise unobserved 13C NMR resonance at 40.2
ppm, broadened by 9Be−13C coupling, and a 1H NMR
resonance at 2.32 ppm, overlapping with a mesityl methyl
singlet, which was attributed to the beryllium-bound methylene
fragment (Figure 3).
9
corresponding COSY experiment. The Be NMR spectrum
displayed a single new resonance at 5.6 ppm, significantly
upfield from that of the starting material 1 (25.4 ppm),
displaying a 1H−9Be HMQC interaction with the 1H resonance
at δ1H 0.61 ppm.
An unstirred preparative-scale reaction between 1 and one
equivalent of PhSiH3 in benzene yielded a crop of large
colorless crystals of compound 2 (85% yield), suitable for X-ray
diffraction analysis. The resulting crystal structure, displayed in
Figure 1, revealed a dimeric IMes-supported methylhydridober-
yllium compound with bridging hydride atoms, which were
located and freely refined. Table 1 presents selected bond
lengths and angles for 2 and its previously reported IPr
analogue, II. In both compounds the beryllium centers display a
pseudotetrahedral geometry [II C1−Be−C28 115.47(11)°;
C1−Be−H1 107.6(7)°; C1−Be−H1# 103.5(7)°; 2 C1−Be−
C22 114.49(10)°; C1−Be−H1 106.9(6)°; C1−Be−H1#
105.7(6)°]. While the Be−Ccarbene bond length [1.822(2) Å]
of II is similar to that in the lower coordinate dimethylber-
yllium precursor, I [1.816(2) Å], the corresponding bond in
compound 2 is somewhat shortened [2 1.8002(19); 1 1.820(4)
Å]. The Be−CMe bond lengths, however, increase by ca. 0.08 Å
in II and 0.04 Å in compound 2 relative to the respective
dimethylberyllium NHC precursors. The central planar Be2H2
units of both dimers form quasi-square arrays [II Be−H−Be
96.3(7)°, H−Be−H 83.7(7)°; 2 Be−H−Be 88.6(6)°, H−Be−
H 91.4(6)°]. While the Be−H bond lengths in compounds II
and 2 are strongly dependent on the steric demands of the
NHC coligand [II 1.552(18) and 1.630(18) Å; 2 1.456(15) and
1.496(15) Å], all these distances are somewhat longer than
those in the similarly four-coordinate hydridoberyllium tris-
(pyrazolyl)borate reported by Parkin and co-workers
[1.228(70) Å]16 and Shearer’s dimeric N,N,N′-trimethylethy-
lenediamido beryllium hydride compound [1.39(2) Å],30 both
of which contain a terminal hydride ligand. The Be−H
distances in II and 2 are, however, similar to the bridging
interactions reported for the dimeric ate-compound
[Na2{Be2Et4H2}·2Et2O] [1.53(3) Å].14
X-ray quality crystals of compound 3 were obtained through
recrystallization of the product of the NMR-scale reaction in d8-
toluene at room temperature. The results of the X-ray
diffraction experiment are displayed in Figure 4. Selected
bond length and angle data are provided in Table 2, along with
the equivalent parameters for compound III. The apparent
asymmetry of the ligand backbone observed in the NMR
spectra is caused by what can be effectively considered as the
insertion of a BeH2 unit into a Ccarbene−N bond with transfer of
the hydrogens onto the carbenoid carbon atom. The result is a
six-membered BeC3N2 ring with a flattened boat conformation,
in which the three-coordinate beryllium center is bound to a
bidentate amidoalkyl ligand and a single unchanged IMes
ligand. The beryllium centers in both compounds are trigonal
planar with little variation in the bond lengths and angles,
except for the Be−N1 distance, which is shorter in 3 [1.560(5)
Å] than the corresponding measurement in one of the two
independent molecules observed in structure III [Be(2)−N(8)
1.592(5) Å]. This may be ascribed to the differences in steric
bulk of the respective N1-aryl groups. As expected, the newly
formed Be−C3 bond to the dianionic alkylamido ligand [III
1.741(5), 1.744(5); 3 1.756(5) Å] is significantly shorter than
the Be−Ccarbene bond to the neutral NHC ligand [III 1.803(5),
1.805(5); 3 1.807(5) Å]. The geometry around C3 clearly
indicates sp3 hybridization and a C3−N2 single bond [Be−
C3−N2: III 112.5(2)°, 113.0(2)°; 3 111.1(2)°; C3−N2: III
1.471(4), 1.473(4); 3 1.462(4) Å]. It is also notable that the
N1−C1 [III 1.422(4), 1.414(4); 3 1.406(3) Å] and N2−C2
bond lengths [III 1.389(4), 1.396(4); 3 1.379(4) Å] are no
longer identical and are significantly longer than in the parent
NHC ligands (ca. 1.35 Å), suggesting some loss of
delocalization. While NHC compounds have been shown to
be generally stable under catalytic conditions, they are also
susceptible to various decomposition processes involving C−H,
C−C, and C−N bond-activation reactions at peripheral and
exocyclic positions.21 The reactions leading to compounds III
and 3, however, are particularly unusual, as they involve loss of
possible aromatic character, reduction of the carbene carbon
atom, and ring expansion. The C−C and C−N bond lengths
and angles within the alkylamido ligands in the borane and
silane ring-opened products (IV and V) are closely comparable
to those in compounds III and 3.22,23
In a similar manner to that observed previously for
compound II in THF solution, a d8-toluene reaction mixture
of compound 2 and a further equivalent of PhSiH3 heated at 80
1
°C for 6 h yielded a clear dark red solution. Analysis by H
NMR spectroscopy evidenced full conversion of PhSiH3 to the
expected metathesis byproduct PhMeSiH2 (Scheme 3) as well
as the formation of a single well-defined product, compound 3,
containing one coordinated NHC ligand and another NHC-
Scheme 3
It is notable that the amidoalkylberyllium fragments of
compounds III and 3 are isomeric to the target [H2Be·(NHC)]
compounds. The presence of a second equivalent of unreacted
IPr or IMes ligand to complete the beryllium coordination
sphere therefore requires the formal loss of “BeH2” during the
9
reaction. Indeed, examination of the Be NMR spectra for the
in situ formation of both III in d8-THF and 3 in d8-toluene
revealed additional resonances at 0.7 and −3.9 ppm,
D
DOI: 10.1021/om501314g
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
1
Figure 3. Annotated H NMR spectrum of isolated crystals of compound 3.
respectively, alongside the broad singlet of the alkylamidober-
yllium compounds at 16.9 ppm (compound III) and 11.0 ppm
(compound 3). In the case of the formation of compound 3
this additional resonance appeared as a poorly defined
multiplet, indicating likely 1H−9Be coupling. Although it
seems likely that a solvated mononuclear constitution for
BeH2 is readily attributed in a coordinating solvent such as
THF, solutions in less basic hydrocarbon solvents do not
appear to have been studied. Crystalline beryllium hydride has
been shown to comprise a body-centered orthorhombic unit
cell, containing a network of corner-sharing {BeH4} tetrahe-
dra,31 while studies of the amorphous form also find that it
consists of a network of corner-shared tetrahedra.32 It seems
likely, therefore, that smaller oligomers built around similar
interactions are likely to persist under the conditions of the
formation of compounds III and 3 in nonbasic media.
Repetition of the reaction with two equivalents of PhSiD3 in
place of PhSiH3 led to the exclusive formation of [D2]-3,
comprising a fully deuterated beryllium-bound methylene
Figure 4. ORTEP representation of compound 3. Ellipsoids are drawn
at 20% probability. Hydrogen atoms are omitted for clarity, except for
those on the methylene carbon C3.
Table 2. Selected Bond Lengths and Angles for Compounds
III, 3, and 4
9
fragment. In this case the Be NMR resonance of the “BeD2”
byproduct appeared as an ill-defined multiplet at −1.4 ppm.
III
3
4
9
The slight downfield shift of this Be nucleus compared to
Be−Ccarbene
Be−N1
Be−C3
N1−C1
C1−C2
1.803(5), 1.805(5)
1.587(5), 1.592(5)
1.741(5), 1.744(5)
1.422(4), 1.414(4)
1.343(4), 1.339(4)
1.389(4), 1.396(4)
1.471(4), 1.473(4)
1.807(5)
1.560(5)
1.756(5)
1.406(3)
1.344(4)
1.379(4)
1.462(4)
1.808(3)
“BeH2” is consistent with its interaction with the marginally
reduced effective nuclear charge of the deuteride ligands.
1.577(3)
2
1.766(3)
Furthermore, H NMR data revealed, alongside the deuterated
1.419(2)
methylene resonance of 3 (2.11 ppm) and the resonance of the
PhSiMeD2 byproduct (4.38 ppm), a third resonance at 1.20
ppm, which may be attributed to “BeD2” (see Figure S1 in the
Supporting Information). Unfortunately 1H−9Be HMQC
experiments failed to confirm the location of the corresponding,
1.342(3)
C2−N2
N2−C3
1.391(2)
1.481(2)
C3−C4
1.535(3)
1
Ccarbene−Be−N1
Ccarbene−Be−C3
N1−Be−C3
Be−N1−C1
Be−C3−N2
Be−C3−C4
N2−C3−C4
N1−C1−C2
C1−C2−N2
C2−N2−C3
130.4(3), 129.1(3)
113.7(2), 114.8(3)
115.7(3), 115.6(3)
112.8(2), 113.4(2)
112.5(2), 113.0(2)
128.9(2)
115.1(3)
116.0(2)
116.4(2)
111.1(2)
128.08(16)
115.52(15)
116.19(15)
114.32(15)
111.81(14)
109.57(15)
110.13(15)
127.00(17)
127.55(17)
120.66(14)
presumably highly broadened, hydride resonance in the H
NMR spectrum.
In an analogous fashion to compound III, compound 3 was
also formed through heating a C6D6 solution of 2 in the
absence of PhSiH3 under forcing thermal conditions (110 °C, 5
days) as depicted in Scheme 4. In this case the formation of
compound 3 was accompanied by the appearance of two molar
equivalents of the dimethylberyllium compound 1 as identified
by 1H NMR spectroscopy. This latter process may be a
consequence of thermally induced Schlenk-type ligand
redistribution to the unreactive dimethylberyllium compound
1 and an unobserved dihydridoberyllium species, which then
126.5(3), 126.6(3)
125.5(3), 126.5(3)
117.9(2), 118.3(2)
125.5(3)
126.4(3)
121.9(2)
E
DOI: 10.1021/om501314g
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 4
Scheme 5
reacts further to yield the ring-opened product 3, releasing
“BeH2”.
resonances through a COSY experiment. Moreover, the methyl
and meta-aryl proton resonances of both the neutral bound
IMes and the ring-opened alkylamido ligands were observed to
become magnetically inequivalent, suggesting the presence of a
chiral center within the reaction product. 13C{1H} NMR
analysis and HMQC and HMBC experiments led to the
identification of this new product as compound 4, in which
both methyl and hydride ligands had been transferred to the
carbenoid carbon atom during the ring-expansion process
(Scheme 5, pathway B). In addition, it appears that the
formation of the methylated compound 4 is kinetically favored
over that of 3, a hypothesis that was confirmed through
monitoring of a further reaction between dimer 2, two
equivalents of IMes, and a single equivalent of PhSiH3 at 80
°C. In this case the rapid formation of one molar equivalent of
compound 4 was observed over a period of 2 h, while
compound 3 was formed much more slowly over a period of 48
h (Scheme 5, pathway C). We suggest that these observations
provide corroborative evidence for the deductions of Dutton
and co-workers that the ring-opening process necessitates the
formation of a four-coordinate and bis-NHC ligated Be center.
Furthermore, the kinetic facility of the methyl/hydride transfer
resulting in the formation of compound 4 is enhanced by the
greater steric compression induced in the putative [(IM-
es)2BeMeH] intermediate compound versus the corresponding
dihydrido species invoked during the synthesis of compound 3.
Large crimson single crystals of compound 4 were obtained
from a saturated toluene solution at room temperature over
several days. The structure of compound 4 is displayed in
Figure 5, while selected bond lengths and angles are provided
in Table 2. The Be−Ccarbene bond length [1.808(3) Å] is very
similar to that in the other two ring-expanded compounds, III
and 3, and the bond lengths and angles around C3 [Be−C3
1.766(3); C3−N2 1.481(2); C3−C4 1.535(3) Å; Be−C3−N2
111.81(14)°; Be−C3−C4 109.57(15)°; N2−C3−C4
These observations confirm that the presence of phenylsilane
is unnecessary for the formation of either 3 or III. It is also
apparent from the reaction stoichiometries depicted in Schemes
3 and 4 that the addition of two further equivalents of IMes
during these reactions should allow the remaining “BeH2” to
generate a further equivalent of the ring-opened alkylamidober-
yllium compound, 3. Accordingly the reaction of 2 with two
equivalents of both PhSiH3 and IMes at 80 °C over 24 h led to
the consumption of all reactants and the exclusive formation of
3 and two equivalents of PhSiMeH2 without any other
byproducts (Scheme 5, pathway A). From this observation
we postulate that this process, involving the formation of a
species of the form [H2Be(IMes)2] similar to the [H2Be-
(NHC)2] intermediates invoked in the computational DFT
study reported by Dutton and co-workers,24 is a prerequisite to
allow the ring-opening process for the formation of 3.
An identical reaction in the presence of an additional
equivalent of IMes but without added phenylsilane induced a
rapid color change in the reaction mixture to orange at 60 °C,
again suggestive of N−C bond cleavage and beryllium insertion.
Notably, this process occurred under significantly less energetic
conditions than had been the case in the presence of PhSiH3,
implying an inhibitory effect of the silane. Thus, after 3 h at 60
1
°C, H NMR analysis indicated near-quantitative consumption
of all reactants with clean formation (90%) of a single ring-
expanded product reminiscent of compound 3. In this case the
AB spin system arising from the desymmetrized ligand
backbone methine centers was found to have undergone a
0.07 ppm downfield shift (δ1H 4.86 vs 4.79 ppm in 3), and the
spectrum did not exhibit the expected methine singlet at δ1H
2.32 ppm. Rather, the appearance of a distinctive 3H doublet at
1.36 ppm was observed, which was shown to couple to a
resonance at 2.07 ppm overlapping with the IMes methyl
F
DOI: 10.1021/om501314g
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
first followed by the mixed H/D species, D-3, and finally the
fully deuterated product, D2-3. These observations indicate
facile H/D exchange between silane and any beryllium hydrido
species present in solution, as well as a marked kinetic isotope
effect upon the overall ring expansion reaction.
The reaction of 10 molar equiv of PhSiH3 with 2 also
evidenced a marked suppression of the ring-opening reactivity.
Less than 10% conversion to compound 3 was observed after
18 h of heating at 110 °C despite over 90% of the beryllium
methyl groups having undergone σ-bond metathesis (Scheme
6). While 1H NMR resonances were too broad to
Scheme 6
Figure 5. ORTEP representation of compound 4. Ellipsoids are drawn
at 20% probability. Hydrogen atoms are omitted for clarity, except for
that on the methine carbon C3.
110.13(15)°] are clearly indicative of sp3 hybridization.
Furthermore, the higher steric demands of the C3-methyl
functionality of 4 compared to the hydrogens of the methylene
fragment of compound 3 result in a lengthening of all bonds
within the six-membered BeN2C3 ring by 0.010 to 0.019 Å,
except for the C1−C2 vinylene backbone. The ring itself is only
slightly distorted from planarity and subtends a 47° angle with
the planar framework of the neutral IMes ligand.
unambiguously identify the beryllium-containing product(s)
in solution, the near-quantitative conversion of beryllium
methyl groups suggests the formation of a potentially dimeric
NHC- and silane-stabilized dihydridoberyllium species. At-
tempts to isolate the latter species by removing volatiles were
unsuccessful, with the reaction mixture rapidly turning dark red,
whereupon NMR analysis of the redissolved solid residue
evidenced near-quantitative conversion to 3. The presence of
excess silane, while promoting σ-bond metathesis and the
formation of a dihydridoberyllium species, can, thus, be
deduced to inhibit hydride transfer and NHC ring opening.
In the absence of the stabilizing influence of the silane,
however, the dihydridoberyllium NHC adduct is unstable and
undergoes spontaneous decomposition to 3.
Careful monitoring of the formation of compounds 3 and 4
1
1
by H and H−9Be HMQC NMR experiments revealed the
formation and consumption of small amounts of the
dimethylberyllium compound 1 as well as a new compound
characterized by a distinctive upfield triplet at δ1H −1.32 ppm.
COSY analysis indicated a coupling of this latter signal to a
resonance overlapping with the methyl protons of PhSiMeH2 at
ca. 0.3 ppm, which, in turn, correlated with a very broad
Heating of the methylhydrido compound 2 in d8-toluene
under more forcing thermal conditions (110 °C) led to the
formation of one equivalent of compound 3 and half an
equivalent of a new species apparently displaying two distinct
1
resonance at ca. 2.8 ppm. Furthermore, a H−9Be HMQC
NMR experiment performed on the reaction mixture revealed a
new 9Be NMR resonance at ca. 2.0 ppm, which coupled to both
symmetrical IMes environments in a 1:1 ratio in the H NMR
1
1
the H NMR resonances at 0.3 and 2.8 ppm (see Figures S2
spectrum (see Figure 6). Heating of a sample of 2 with one
equivalent of PhSiH3 in d8-toluene at 110 °C also led to
exclusive and quantitative conversion to this same compound,
which a DOSY experiment confirmed to be a single species,
compound 6. 9Be NMR data indicated the presence of a single
major beryllium-containing product with a broad singlet at 4.4
ppm, as well as a series of low-intensity and ill-defined
overlapping resonances further upfield. An X-ray diffraction
experiment performed on single crystals of this orange
compound yielded the unexpected structure of an NHC-
supported beryllium compound of the diamido ligand
[{CHNMes}2]2−. Although the crystals diffracted too weakly
to collect data beyond θ = 24°, the structure provided
unambiguous connectivity (Figure 7) and confirmed the loss of
the formerly carbene carbon atom in one of the IMes ligands.
Monitoring of the formation of 6 from 2 and PhSiH3
revealed the appearance and disappearance of compound 3 as
a reaction intermediate. Heating of a d8-toluene solution of
isolated crystals of compound 3 to 150 °C for several hours,
however, did not result in any reactivity. It was therefore
inferred that the byproduct of the formation of 6 from 2 and
PhSiH3, formally [BeH2], had to be present for the reaction to
occur. Polymeric [H2Be(THF)n]m was, thus, obtained by
precipitation from the reaction of BeMe2 with excess PhSiH3
in boiling THF and added to the sample of 3 in d8-toluene.
Heating at 150 °C for a week indeed provided about 70%
and S3 in the Supporting Information). The integrity of this
new compound (5) was demonstrated by the observation of a
single diffusion coefficient for these resonances in the
corresponding 1H DOSY NMR spectrum, leading to its
tentative identification as the dimeric methyldihydrido
diberyllium silanate [(IMes)·BeMe(μ-H2)Be(SiH2MePh)·
(IMes)], compound 5 (see Scheme 5). Attempts to isolate
compound 5 failed due to its solution instability at the given
reaction temperatures. The appearance of both 1 and 5 during
both of these reactions emphasizes that Schlenk-type ligand
exchange is a key process during the formation of both 3 and 4
in the absence of PhSiH3.
The role of phenylsilane was further investigated with a
proton−deuterium crossover experiment. Addition of two
equivalents of PhSiD3 to a solution of 2 led to H/D exchange
to form PhSiD3−nHn (n = 1−3) and a 2:3 mixture of D2-2 and 2
after 30 min at room temperature as determined by NMR
spectroscopy. Five minutes of heating at 100 °C resulted in full
deuteration of 2 without any evidence for Be−C/Si−H σ-bond
metathesis. After a further 30 min at 100 °C, however, the
appearance of the σ-bond metathesis product PhSiMeD2−nHn
(n = 0−2) was observed, alongside compounds 1 and 5.
Notably, discernible evidence for the commencement of the
ring-opening process was observed only after several hours’
heating at 110 °C, with the diprotio product, 3, being formed
G
DOI: 10.1021/om501314g
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
1
Figure 6. Annotated H NMR spectrum of the reaction leading to the formation of compound 6.
is more facile, which we ascribe to the greater relief of steric
pressure achieved after the formation of four-coordinate
beryllium. Schlenk-type redistribution is evidently a vital
process throughout this chemistry, both between the individual
Be centers and between the beryllium hydride and silane
reaction partners. Although the full scope of this reactivity
requires further investigation, these unusual transformations
and the fragmentation process leading to compound 6 suggest
that beryllium hydrido and organometallic species are worthy of
more concerted study than that currently sanctioned by their
somewhat fearsome reputation.
EXPERIMENTAL SECTION
■
Figure 7. ORTEP representation of compound 6. Ellipsoids are drawn
at 20% probability. Hydrogen atoms are omitted for clarity.
All reactions dealing with air- and moisture-sensitive compounds were
carried out under an argon atmosphere using standard Schlenk line or
glovebox techniques in an MBraun Labmaster glovebox at O2, H2O <
0.1 ppm. NMR experiments using air-sensitive compounds were
conducted in Youngs tap NMR tubes prepared and sealed in a
glovebox under argon. All NMR data were acquired on Bruker 300 or
400 Ultrashield spectrometers for 1H (300/400 MHz), COSY,
13C{1H} (75.48 MHz), and 9Be (56.23 MHz) spectra at room
temperature. 1H/13C NMR spectra were referenced using residual
solvent resonances, and 9Be NMR spectra were referenced to an
external standard of aqueous beryllium chloride. Elemental analyses of
all moisture- and air-sensitive compounds were performed by Stephen
Boyer of London Metropolitan Enterprises. Solvents for air- and
moisture-sensitive reactions were provided by an Innovative
Technology Solvent Purification System. C6D6, d8-toluene, and d8-
THF were purchased from Fluorochem and dried over molten
potassium prior to vacuum transfer into a sealed ampule and storage in
the glovebox under argon. BeCl2 was purchased from Sigma-Aldrich.
1,3-Bis(2,4,6-trimethylphenyl)imidazol-2-ylidene (IMes) was synthe-
sized by a literature procedure.33
CAUTION: Beryllium and its compounds are extremely toxic. Suitable
precautions (e.g., use of protective clothing, breathing apparatus, and a
well-ventilated fume cupboard) should be taken for all manipulations
involving them.
Synthesis of [Me2Be·IMes], 1. IMes (0.5 g, 1.64 mmol) was
combined with BeCl2 (131 mg, 1.64 mmol) and 1.6 M MeLi in Et2O
(2.1 mL, 3.36 mmol) at −78 °C. The reaction mixture was allowed to
warm to room temperature and stirred overnight. Removal of the
solvent in vacuo, followed by extraction into hot toluene and filtration,
yielded a clear, pale yellow solution. Compound 1 was isolated as
colorless crystals from a concentrated toluene solution stored at −18
conversion to compound 6 as well as a few minor unidentified
decomposition products, indicating the involvement of [BeH2]
in the reaction. Although the fate of the remaining elements
required to provide mass balance during the formation of
compound 6 could not been definitively resolved, this
observation of activation of both N−C bonds within an N-
heterocyclic carbene is, to the best of our knowledge, unique.
CONCLUSION
■
We have observed that dimethylberyllium NHC adducts are
stable species and readily form the methylhydrido derivatives
upon silane-mediated methyl for hydride exchange. Attempts to
synthesize the corresponding molecular dihydrides were
unsuccessful, however, and resulted in the formation of
heterocyclic beryllium organoamides as a consequence of
hydride transfer to the donor carbon atom of a coordinated
NHC ligand. In agreement with previous calculations, we
suggest that this process is likely to occur via the formation of a
four-coordinate bis-NHC beryllium dihydride, whereupon the
steric congestion of the Be center is sufficient to induce hydride
transfer. The latter deduction is supported by the observation
that similar ring-opened products, but derived from methyl and
hydride transfer, are available through the introduction of a
further equivalent of NHC to the isolated beryllium methyl
hydride species. In this case, however, the ring-opening process
H
DOI: 10.1021/om501314g
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
1
°C (0.33 g, 0.96 mmol, 58% yield). H NMR (300 MHz, 298 K, d8-
tol): 6.67 (s, 4H, m-Ar-H), 6.09 (s, 2H, MesNCH), 2.06 (s, 12H, o-
CH3), 2.05 (s, 6H, p-CH3), −0.73 (s, 6H, BeCH3). 13C{1H} NMR (75
MHz, 298 K, d8-tol): 184.9 (Ccarbene), 139.5, 135.6, 135.5, 129.7, 121.8,
21.3, 18.0, −3.7 (br, BeCH3). 9Be (56 MHz, 298 K, d8-tol): 25.4 ppm.
Anal. Calcd for C23H30BeN2 (MW = 353.5): C, 80.42; H, 8.80; N,
8.16. Found: C, 80.30; H, 8.75; N, 8.13.
123.6 (MesNCHDAB), 120.8 (MesNCHNHC), 21.6, 21.5, 20.3, 18.3. 9Be
(56 MHz, 298 K, d8-tol): 4.4 ppm. Acceptable CHN microanalysis
could not be obtained for this compound. Best values obtained: Anal.
Calcd: C, 81.28; H, 7.99; N, 9.25. Found: C, 78.10; H, 8.26; N, 8.71.
X-ray Crystallography. Data for compounds 1−4 and 6 were
collected at 150 K on a Nonius KappaCCD diffractometer equipped
with an Oxford Cryosystem, using graphite-monochromated Mo Kα
radiation (λ = 0.710 73 Å). Data were processed using the Nonius
Software.34 Crystal parameters and details of data collection, solution,
and refinement for the compounds are provided in Table S1 in the
Supporting Information. Structure solution, followed by full-matrix
least-squares refinement, was performed using the WINGX-1.70 suite
of programs throughout.35 Due to a very weakly diffracting crystal,
data of 6 were truncated to 48° in 2θ.
Synthesis of [MeHBe·IMes]2, 2. Compound 1 (30 mg, 87.3
μmol) and PhSiH3 (10.8 μL, μmol) were combined in C6D6.
Quantitative conversion to 2 was reached after 5 h at room
temperature. Colorless crystals of 2 were obtained after storing the
reaction mixture at 4 °C overnight. 1H NMR (300 MHz, 298 K,
C6D6): 6.76 (s, 4H, m-Ar-H), 6.09 (s, 2H, MesNCH), 2.29 (s, 6H, p-
CH3), 2.07 (s, 12H, o-CH3), 0.61 (br s, 1H, BeH), −1.26 (t, 3H,
BeCH3, J = 1.9 Hz). 13C{1H} NMR (75 MHz, 298 K, C6D6): 179.6
3
(Ccarbene), 137.5 (i-Ar-C), 136.9, 136.4, 129.0, 121.3 (MesNCH), 21.7,
ASSOCIATED CONTENT
* Supporting Information
Crystallographic information files (CIF) for 1−4 and 6, details
of the X-ray analyses (Table S1), and additional figures of
selected NMR spectra. This material is available free of charge
via the Internet at http://pubs.acs.org.
■
18.9, −8.2 (br, BeCH3). 9Be (56 MHz, 298 K, C6D6): 5.6 ppm (cross-
S
9
peak in Be−1H HMQC spectrum at δ1H 0.61 ppm). Anal. Calcd for
C22H28BeN2 (MW = 329.5): C, 80.20; H, 8.57; N, 8.50. Found: C,
80.13; H, 8.48; N, 8.37.
Synthesis of [{MesN(CH)2NMesCH2}Be·IMes], 3. Compound 2
(29 mg, 87.3 μmol) was heated with one equivalent of PhSiH3 (10.8
μL, μmol) in d8-toluene for 5 h at 80 °C. NMR data showed full
conversion to the ring-opened product. Orange-red crystals of
compound 3 were obtained from a saturated toluene solution at
AUTHOR INFORMATION
Corresponding Author
*E-mail (M. S. Hill): msh27@bath.ac.uk.
Notes
■
1
room temperature. H NMR (300 MHz, 298 K, C6D6): 6.93 (s, 2H,
H6/6′), 6.74 (s, 2H, H6/6′), 6.67 (s, 4H, H14), 5.82 (s, 2H, H11),
3
3
5.02 (d, 1H, H2, J = 6.3 Hz), 4.79 (d, 1H, H3, J = 6.3 Hz), 2.51 (s,
6H, H8/8′), 2.32 (s, 2H, H1), 2.31 (s, 3H, H9/9′), 2.23 (s, 3H, H9/
9′), 2.15 (s, 6H, H17), 2.12 (s, 6H, H8/8′), 1.79 (s, 12H, H16).
13C{1H} NMR (75 MHz, 298 K, C6D6): 176.2 (C10), 152.4, 150.8
(C4/4′), 139.2 (C15), 136.2 (C5/5′), 135.2 (C12), 134.9 (C13), 131.9
(C7/7′), 129.5 (C14), 128.4, 128.2 (C6/6′), 122.8 (C11), 116.0 (C2),
115.7 (C3), 40.2 (C1), 21.2, 21.0 (C9/9′), 20.0 (C17), 18.7 (C8/8′),
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We acknowledge financial support from the EPSRC and thank
Prof. Stuart Macgregor and Dr. David McKay for stimulating, if
ultimately frustrating, discussions.
9
17.7 (C16). Be NMR (56 MHz, 298 K, C6D6): 16.9 ppm (br). Anal.
Calcd for C42H50BeN4 (MW = 619.9): C, 81.38; H, 8.13; N, 9.04.
Found: C, 81.37; H, 8.38; N, 8.85.
REFERENCES
■
(1) For reviews, see: (a) Barrett, A. G. M.; Crimmin, M. R.; Hill, M.
S.; Procopiou, P. A. Proc. R. Soc. A 2010, 466, 927−963. (b) Harder, S.
Chem. Rev. 2010, 110, 3852−3876. (c) Arrowsmith, M.; Hill, M. S.
Alkaline Earth Chemistry: Applications in Catalysis. In Comprehensive
Inorganic Chemistry II; Reedijk, J., Poeppelmeier, K. Eds.; Elsevier:
Oxford, 2013; Vol. 1, pp 1189−1216. (d) Crimmin, M. R.; Hill, M. S.
Top. Organomet. Chem. 2013, 45, 191−242.
Synthesis of [{MesN(CH)2NMesCHMe}Be·IMes], 4. Compound
4 was stirred in toluene with one equivalent of PhSiH3 at 50 °C for 1
week. The solution slowly turned red. Analysis of the crude showed
about 50% conversion to compound 7. Removal of volatiles and
fractional crystallization in hexanes allowed recovery of the unreacted
methylhydrido compound 4. Orange-red crystals of the ring-opened
product were obtained from a saturated 9:1 hexanes/toluene solution
at room temperature in 34% yield. 1H NMR (300 MHz, 298 K, C6D6):
6.95 (s, 1H, H6/6′), 6.89 (s, 1H, H6/6′), 6.79 (s, 1H, H6/6′), 6.67 (s,
2H, H6/6′), 6.61 (s, 1H, H14), 6.54 (s, 2H, H14), 5.83 (s, 2H, H11),
(2) For reviews of main group hydride chemistry, see: (a) Aldridge,
S.; Downs, A. J. Chem. Rev. 2001, 101, 3305−3365. (b) Harder, S.
Chem. Commun. 2012, 48, 11165−11177.
3
3
(3) Green, S. P.; Jones, C.; Stasch, A. Angew. Chem., Int. Ed. 2008, 47,
9079−9083.
5.02 (d, 1H, H2, J = 6.3 Hz), 4.86 (d, 1H, H3, J = 6.3 Hz), 2.64 (s,
3H, H8/8′), 2.32 (s, 3H), 2.29 (s, 3H), 2.27 (s, 3H), 2.16 (s, 3H), 2.15
(s, 6H), 2.07 (m, 4H, H1/Mes-CH3), 1.98 (s, 6H), 1.66 (s, 6H), 1.36
(4) Harder, S.; Brettar, J. Angew. Chem., Int. Ed. 2006, 45, 3474−
3
(d, 3H, H18, J = 6.3 Hz). 13C{1H} NMR (75 MHz, 298 K, C6D6):
3478.
(5) Arrowsmith, M.; Hill, M. S.; MacDougall, D. J.; Mahon, M. F.
Angew. Chem., Int. Ed. 2009, 48, 4013−4016.
176.4 (C10), 152.7, 147.8 (C4/4′), 139.2 (C15), 136.5, 136.3, 136.1
(C5/5′), 135.1 (C12), 134.8 (C13), 134.4, 131.1, 130.0 (C7/7′), 129.8
(C14), 129.4, 129.3, 129.0, 128.4 (C6/6′), 123.0 (C11), 117.4 (C2),
114.2 (C3), 45.0 (C1), 21.2, 21.0 (C9/9′), 20.7, 20.2 (C17), 20.0, 19.9,
(6) Harder, S.; Spielmann, J.; Intemann, J.; Bandmann, H. Angew.
Chem., Int. Ed. 2011, 50, 4156−4160.
9
(7) Intemann, J.; Spielmann, J.; Sirsch, P.; Harder, S. Chem.Eur. J.
19.2 (C8/8′), 18.0, 17.7 (C16). Be NMR (56 MHz, 298 K, C6D6):
2013, 19, 8478−8489.
14.7 ppm (br). Anal. Calcd for C43H52BeN4 (MW = 633.9): C, 81.47;
H, 8.27; N, 8.84. Found: C, 81.36; H, 8.38; N, 8.79.
(8) Blindheim, U.; Coates, G. E.; Srivastava, R. C. J. Chem. Soc.,
Dalton Trans. 1972, 2302−2305.
Synthesis of [{(CH)2(NMes)2}Be·IMes], 6. Compound 2 (29 mg,
87.3 μmol) was heated overnight with one equivalent of PhSiH3 (10.8
μL, 88 μmol) in d8-toluene at 110 °C. Analysis by NMR spectroscopy
showed full conversion to the diamidoberyllium compound, 6. Flaky
orange crystals of 6 were obtained from slow vapor diffusion of
hexanes into a saturated toluene solution at room temperature over a
period of 2 weeks. 1H NMR (300 MHz, 298 K, d8-tol): 6.63 (s, 4H, m-
Ar-HNHC), 6.43 (s, 4H, m-Ar-HDAB), 5.90 (s, 2H, MesNCHNHC), 5.71
(s, 2H, MesNCHDAB), 2.32 (s, 6H, p-CH3‑NHC), 2.16 (s, 6H, p-
CH3‑DAB), 2.11 (s, 12H, o-CH3‑NHC), 1.82 (s, 12H, o-CH3‑DAB).
13C{1H} NMR (75 MHz, 298 K, d8-tol): 173.6 (Ccarbene), 151.1 (i-Ar-
CDAB), 138.8, 135.5, 135.1 (i-Ar-CNHC), 134.2, 129.8, 129.7, 128.6,
(9) Coates, G. E.; Smith, D. L.; Srivastava, R. C. J. Chem. Soc., Dalton
Trans. 1973, 618−622.
(10) Bell, N. A.; Coates, G. E. J. Chem. Soc. A 1966, 1069−1073.
(11) Bell, N. A.; Coates, G. E. J. Chem. Soc. 1965, 692−699.
(12) Coates, G. E.; Tranah, M. J. Chem. Soc. A 1967, 236−239.
(13) Coates, G. E.; Glockling, F. J. Chem. Soc. 1954, 22−27.
(14) Adamson, G. W.; Bell, N. A.; Shearer, H. M. M. Acta Crystallogr.
1981, B37, 68−71.
(15) Bell, N. A.; Coates, G. E.; Shearer, M. L.; Shearer, H. M. M. J.
Chem. Soc., Chem. Commun. 1983, 828−829.
(16) Han, R.; Parkin, G. Inorg. Chem. 1992, 31, 983−988.
I
DOI: 10.1021/om501314g
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(17) Boffetta, P.; Fryzek, J. P.; Mandel, J. S. Crit. Rev. Toxicol. 2012,
42, 107−118.
(18) For example, see: (a) Liptrot, D. J.; Hill, M. S.; Mahon, M. F.;
MacDougall, D. J. Chem.Eur. J. 2010, 16, 8508−8515. (b) Hill, M.
S.; Hodgson, M.; Liptrot, D. J.; Mahon, M. F. Dalton Trans. 2011, 40,
7783−7790. (c) Bellham, P.; Hill, M. S.; Liptrot, D. J.; MacDougall, D.
J.; Mahon, M. F. Chem. Commun. 2011, 47, 9060−9062. (d) Bellham,
P.; Hill, M. S.; Kociok-Kohn, G.; Liptrot, D. J. Dalton Trans. 2013, 42,
̈
737−745. (e) Bellham, P.; Hill, M. S.; Kociok-Kohn, G.; Liptrot, D. J.
̈
Chem. Commun. 2013, 49, 1960−1962. (f) Dunne, J. F.; Neal, S. R.;
Engelkemier, J.; Ellern, A.; Sadow, A. D. J. Am. Chem. Soc. 2011, 133,
16782−16785. (g) Hill, M. S.; Liptrot, D. J.; MacDougall, D. J.;
Mahon, M. F.; Robinson, T. P. Chem. Sci. 2013, 4, 4212−4222.
(h) Arrowsmith, M.; Hill, M. S.; Kociok-Kohn, G. Organometallics
̈
2010, 29, 4203−4206.
(19) For example, see: (a) Crimmin, M. R.; Casely, I. J.; Hill, M. S. J.
Am. Chem. Soc. 2005, 127, 2042−2043. (b) Datta, S.; Gamer, M. T.;
Roesky, P. W. Organometallics 2008, 27, 1207−1213. (c) Arrowsmith,
M.; Crimmin, M. R.; Barrett, A. G. M.; Hill, M. S.; Kociok-Kohn, G.;
̈
Procopiou, P. A. Organometallics 2011, 30, 1493−1506. (d) Liu, B.;
Roisnel, T.; Carpentier, J.-F.; Sarazin, Y. Chem.Eur. J. 2013, 19,
2784. (e) Arrowsmith, M.; Hill, M. S.; Kociok-Kohn, G. Organo-
̈
metallics 2011, 30, 1291−1294. (f) Brinkmann, C.; Barrett, A. G. M.;
Hill, M. S.; Procopiou, P. A.; Reid, S. Organometallics 2012, 31, 7287−
7297.
(20) Arrowsmith, M.; Hill, M. S.; Kociok-Kohn, G.; MacDougall, D.
̈
J.; Mahon, M. F. Angew. Chem., Int. Ed. 2012, 51, 2098−2100.
(21) For reviews of NHC degradation, see: (a) Crudden, C. M.;
Allen, D. P. Coord. Chem. Rev. 2004, 248, 2247−2273. (b) Cavell, K. J.
Dalton Trans. 2008, 6676−6685. (c) Albrecht, M. Chem. Rev. 2010,
110, 576−623.
(22) Al-Rafia, S. M. I.; McDonald, R.; Ferguson, M. J.; Rivard, E.
Chem.Eur. J. 2012, 18, 13810−13820.
(23) Schmidt, D.; Berthel, J. H. J.; Pietsch, S.; Radius, U. Angew.
Chem., Int. Ed. 2012, 51, 8881−8885.
(24) (a) Iversen, K. J.; Wilson, D. J. D.; Dutton, J. L. Dalton Trans.
2013, 42, 11035−11038. (b) Iversen, K. J.; Wilson, D. J. D.; Dutton, J.
L. Organometallics 2013, 32, 6209−6217. For reviews of progress in
this area, see: (c) Iversen, K. J.; Wilson, D. J. D.; Dutton, J. L. Dalton
Trans. 2014, 43, 12820−12823. (d) Iversen, K. J.; Couchman, S. A.;
Wilson, D. J. D.; Dutton, J. L. Coord. Chem. Rev. 2015, in press (doi:
10.1016/j.ccr.2014.11.008).
(25) Su, M.-D. Inorg. Chem. 2014, 53, 5080−5087.
(26) Fang, R.; Yang, L.-Z.; Wang, Q. Organometallics 2014, 33, 53−
60.
(27) Momeni, M. R.; Rivard, E.; Brown, A. Organometallics 2013, 32,
6201−6208.
(28) Gottfriedsen, J.; Blaurock, S. Organometallics 2006, 25, 3784−
3786.
(29) Gilliard, R. J., Jr.; Abraham, M. Y.; Wang, Y.; Wei, P.; Xie, Y.;
Quillian, B.; Schaefer, H. F., III; Schleyer, P. v. R.; Robinson, G. H. J.
Am. Chem. Soc. 2012, 134, 9953−9955.
(30) Bell, N. A.; Coates, G. E.; Schneider, M. L.; Shearer, H. M. M.
Acta Crystallogr. 1984, C40, 608−610.
(31) Smith, G. S.; Johnson, Q. C.; Smith, D. K.; Cox, D. E.; Snyder,
R. L.; Zhou, R.-S.; Zalkin, A. Solid State Commun. 1988, 67, 491−494.
(32) Sampath, S.; Lantzky, K. M.; Benmore, C. J.; Neuefeind, J.;
Siewenie, J. E. J. Chem. Phys. 2003, 119, 12499−12502.
(33) Bantreil, X.; Nolan, S. P. Nat. Protoc. 2011, 6, 69−77.
(34) DENZO-SCALEPACK: Otwinowski, Z.; Minor, W.Processing
of X-ray Diffraction Data Collected in Oscillation Mode. In Methods in
Enzymology, Vol. 276: Macromolecular Crystallography, part A; Carter,
C. W., Jr., Sweet, R. M., Eds.; Academic Press, 1997; pp 307−326.
(35) Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837−838.
J
DOI: 10.1021/om501314g
Organometallics XXXX, XXX, XXX−XXX