Structural diversity of copper(I)-N-heterocyclic carbene complexes; Ligand tuning facilitates isolation of the first structurally characterised copper(I)-NHC containing a copper(I)-alkene interaction

Article abstract of DOI:10.1002/chem.201301896

The preparation of a series of imidazolium salts bearing N-allyl substituents, and a range of substituents on the second nitrogen atom that have varying electronic and steric properties, is reported. The ligands have been coordinated to a copper(I) centre and the resulting copper(I)-NHC (NHC=N-heterocyclic carbene) complexes have been thoroughly examined, both in solution and in the solid-state. The solid-state structures are highly diverse and exhibit a range of unusual geometries and cuprophilic interactions. The first structurally characterised copper(I)-NHC complex containing a copper(I)-alkene interaction is reported. An N-pyridyl substituent, which forms a dative bond with the copper(I) centre, stabilises an interaction between the metal centre and the allyl substituent of a neighbouring ligand, to form a 1D coordination polymer. The stabilisation is attributed to the pyridyl substituent increasing the electron density at the copper(I) centre, and thus enhancing the metal(d)-to-alkene(π*) back-bonding. In addition, components other than charge transfer appear to have a role in copper(I)-alkene stabilisation because further increases in the Lewis basicity of the ligand disfavours copper(I)-alkene binding.

Full text of DOI:10.1002/chem.201301896

DOI: 10.1002/chem.201301896
Structural Diversity of Copper(I)–N-Heterocyclic Carbene Complexes;
Ligand Tuning Facilitates Isolation of the First Structurally Characterised
Copper(I)–NHC Containing a Copper(I)–Alkene Interaction
Benjamin R. M. Lake and Charlotte E. Willans*[a]
Abstract: The preparation of a series
of imidazolium salts bearing N-allyl
substituents, and a range of substitu-
ents on the second nitrogen atom that
have varying electronic and steric prop-
erties, is reported. The ligands have
been coordinated to a copper(I) centre
and the resulting copper(I)–NHC
(NHC=N-heterocyclic carbene) com-
plexes have been thoroughly examined,
both in solution and in the solid-state.
The solid-state structures are highly di-
verse and exhibit a range of unusual
geometries and cuprophilic interac-
tions. The first structurally character-
ised copper(I)–NHC complex contain-
ing a copper(I)–alkene interaction is
reported. An N-pyridyl substituent,
which forms a dative bond with the
copper(I) centre, stabilises an interac-
tion between the metal centre and the
ligand, to form a 1D coordination poly-
mer. The stabilisation is attributed to
the pyridyl substituent increasing the
electron density at the copper(I)
centre, and thus enhancing the
metal(d)-to-alkene(p*) back-bonding.
In addition, components other than
charge transfer appear to have a role
in copper(I)–alkene stabilisation be-
cause further increases in the Lewis ba-
sicity of the ligand disfavours cop-
per(I)–alkene binding.
allyl substituent of
a neighbouring
Keywords: alkenes · catalysis · cop-
per · carbene ligands · pyridyl rings
Introduction
complexes, in which L is a chelating nitrogen-donor ancillary
ligand.[9–11] Several copper(I)–tris(pyrazolyl)borate–ethylene
complexes have been isolated and structurally character-
ised.[2,12,13] It has been found that decreasing the Lewis basic-
ity of the ancillary ligand, by introducing fluoroalkyl sub-
stituents, decreases the amount of copper!ethylene p back-
bonding. This decrease can be explained using the Dewar–
Chatt–Duncanson model. Surprisingly, the fluorinated com-
plex is an air-stable solid and does not lose ethylene under
reduced pressure, whereas its methylated analogue oxidises
upon exposure to air.[2] This finding is counter-intuitive
when considering the different levels of p back-bonding in
the complexes. Recently, York and co-workers published
a DFT study on the impact of ancillary ligand basicity on
copper(I)–ethylene binding interactions.[14] They report that
both charge-transfer and components other than charge-
transfer are affected by the basicity of the ligand. Although
copper!ethylene p back-bonding increases with increasing
ligand basicity, polarisation and frozen-density terms simul-
taneously become less favourable, resulting in little overall
binding-energy change with alteration of ligand basicity.
Some reported reactions, in which a copper(I)–alkene in-
teraction is postulated, use copper(I)–N-heterocyclic car-
bene (NHC) complexes.[15,16] To the best of our knowledge,
interaction of an alkene with a copper(I)–NHC complex has
not been previously observed in the solid state. The strong
s-donating property of NHC ligands may result in a low fa-
vourable polarisation term and a more repulsive frozen-den-
sity term. Understanding the nature of the copper(I)–NHC
interaction and its effect on copper(I)–alkene binding is im-
perative for the development of copper(I)–NHC complexes
The binding of alkenes to copper(I) metal centres has been
proposed as an important step in a variety of chemical pro-
cesses, for example, in alkene aziridination and cyclopropa-
nation, and in conjugate additions to a,b-unsaturated ke-
tones.[1–5] Furthermore, in nature, it is believed that the for-
mation of a copper(I)–ethylene adduct in the ethylene re-
sponse 1 (ETR 1) plant protein is a key step in the regula-
tion of several aspects of the plant life cycle.[6] As the
bonding is relatively weak, many systems take advantage of
the reversible binding nature of copper(I)-alkenes. Other
potential applications in which the strength of the cop-
per(I)-alkene bond may play an important role include sepa-
ration chemistry, in which alkenes are selectively separated
from crude organic feedstock,[7] and in molecular sensing to
detect the presence of alkenes.[8] As metal–carbonyl bonding
is similar to metal–alkene bonding, applications such as sep-
aration and sensing may also take advantage of the revers-
ACHTUNGTRENNUNG
studied and reported in several papers. Most of the com-
plexes previously examined involve copper(I)–L–ethylene
[a] B. R. M. Lake, Dr. C. E. Willans
School of Chemistry, University of Leeds
Woodhouse Lane, Leeds, LS2 9 JT (UK)
Fax : (+44)1133436565
E-mail: c.e.willans@leeds.ac.uk
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201301896.
16780
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 16780 – 16790

FULL PAPER
for use in catalytic processes, which rely on alkene coordina-
tion to the copper centre. Herein, we report the syntheses
and full characterisation of copper(I)–NHC complexes, in
which the NHC ligands contain an alkene substituent. The
first structurally characterised copper(I)–NHC complex con-
taining a copper(I)–alkene interaction is described.
Results and Discussion
Scheme 1. Synthesis of copper(I)–NHC complexes 2a–2e.
A potential strategy for the stabilisation and examination of
a copper(I)–alkene interaction in a copper(I)–NHC complex
is to use a hemilabile alkene substituent on the NHC ligand
itself. Thus, imidazolium ligand precursors 1a–1e were pre-
pared using established procedures, and fully characterised
(Figure 1) (see the Supporting Information).[17–19] All of the
ligands possess either one or two N-allyl substituents, and
have a variety of substituents with different steric and elec-
tronic properties on the second nitrogen atom.
bond lengths of 2.44 and 2.46 ꢁ. The (CuBr)2 core forms
a plane with a torsion angle of approximately 248 with re-
spect to the plane defined by the benzimidazol-2-ylidene
ꢀ
rings. The Cu C bond length, at 1.91 ꢁ, is normal for this
type of complex.[24,25,28]
Figure 1. Imidazolium ligand precursors 1a–1e.
Figure 2. Molecular structure of 2a. Ellipsoids are shown at 50% proba-
bility level. Hydrogen atoms have been omitted for clarity.
Copper(I)–NHC complexes can be synthesised using
a range of methods, including: 1) Deprotonation of an imi-
dazolium salt using a strong base, followed by coordination
to a metal centre.[20–22] 2) Transmetalation of the correspond-
ing silver(I)–NHC complex.[23] 3) Deprotonation of the imi-
dazolium and coordination in situ using a basic metal pre-
cursor.[24] 4) Electrochemical reduction of the imidazolium
and coordination to a metal centre.[25,26]
Complex 2b crystallises as a dimer with a weak, though
unusual, cuprophilic interaction (Cu—Cu 2.74 ꢁ) (Figure 3).
These types of interaction are usually observed because of
the bidentate nature of a ligand bringing the copper centres
into close proximity. Cuprophilic interactions in complexes
The reaction of compound 1b with a base (tBuOK) in the
presence of CuBr resulted in desired product 2b, in addition
to two allyl rearrangement products (see the Supporting In-
formation).[27] Therefore, because the allyl substituents
appear to be sensitive to strong bases, the coordination of li-
gands 1a–1e to a copper(I) centre was achieved through de-
protonation and coordination in situ, by using copper oxide
(Scheme 1).
Complexes 2a–2e were fully characterised using NMR
spectroscopy, mass spectrometry and elemental analysis, all
of which support formation of complexes of the type [Cu-
ACHTUNGTRENNUNG(NHC)Br]. Complexes 2a, 2b, 2d and 2e were further ex-
amined using X-ray crystallography. Complex 2a crystallises
as a centrosymmetric (m-Br)2-bridged dimer (Figure 2). The
Figure 3. Molecular structure of 2b. Ellipsoids are shown at 50% proba-
ꢀ
central (CuBr)2 core is slightly asymmetric, with Cu Br
bility level. Hydrogen atoms have been omitted for clarity.
Chem. Eur. J. 2013, 19, 16780 – 16790
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16781

C. E. Willans and B. R. M. Lake
that lack a bidentate ligand are less common than their ar-
gentophilic and aurophilic analogues, although some exam-
ples have been described in the literature.[29–32] It is evident
from the solid-state structure that distortion of the C-Cu-Br
bond angles (to 169.88 and 167.78) away from linearity
occurs, presumably as a result of the cuprophilic interaction.
ꢀ
The Cu C bond lengths of 1.91 and 1.92 ꢁ are within the
ꢀ
usual range for copper(I)–NHC complexes, as are the Cu
Br bond lengths (both 2.28 ꢁ).[24,25,28] These findings indicate
that the cuprophilic interaction does not significantly modify
the electronic properties of the copper(I) centres.
Complex 2d crystallises as a [CuACHTUNRGTNEUNG(NHC)2Br]-type complex,
with a rather unusual distorted-trigonal coordination geome-
try (Figure 4). This type of complex has only been observed
Figure 5. Molecular structure of 2e. Ellipsoids are shown at 50% proba-
bility level. Hydrogen atoms have been omitted for clarity.
twisting would act to prohibit p-overlap between the meth-
AHCTUNGTREGoNNNU xyphenyl and NHC rings, resulting in the highly p-elec-
tron-donating OMe group having little electronic effect on
the NHC ligand.
The steric properties of the ligands of 2a–2e have been
varied and result in a range of interesting structures that
differ significantly around the copper(I) centres. However,
these variations did not facilitate any copper(I)–alkene coor-
dination. The ligand precursors described (see below) were
therefore appended with 2-pyridyl N-substituents. It was an-
ticipated that this would modulate the electronic properties
of the NHC ligand, in addition to providing the opportunity
for metal chelation. The complexes formed from these li-
gands are fascinating in their own right. Moreover, these li-
gands allowed the isolation of the first example of a structur-
ally characterised copper(I)–NHC complex containing a cop-
per(I)–alkene interaction.
Figure 4. Molecular structure of 2d. Ellipsoids are shown at 50% proba-
bility level. Hydrogen atoms have been omitted for clarity.
previously in two other examples.[21,33] This structure is
somewhat unexpected, especially given that elemental anal-
ysis gives a ratio of 1:1 NHC/CuBr. Due to the labile nature
of copper(I)–NHC complexes, it is possible that transmeta-
lation occurs during crystallisation, providing one mole
Ligand precursors 1 f–1i (Figure 6) were prepared, and
fully characterised using NMR spectroscopy, mass spectrom-
etry and elemental analysis.[34,35] These ligands offer the pos-
equivalent of [CuACHTUNGTRENNUNG(NHC)2Br] and one mole equivalent of
CuBr. The trigonal geometry is surprising, especially consid-
ering the steric bulk of the mesityl wingtip groups. The
bulky mesityl groups widen the C-Cu-C bond angle (151.18),
an effect that in turn decreases the C-Cu-Br bond angles
(103.38 and 105.68) and also forces the bromide ion further
ꢀ
from the copper(I) centre (Cu Br 2.62 ꢁ) in order to re-
ꢀ
lieve steric congestion. This Cu Br bond is rather long com-
pared with those of other complexes described in the litera-
ture, and to some extent, may also be caused by the strong
s-donor effect of the two NHC ligands coordinated to the
ꢀ
copper(I) centre. The Cu C bond lengths, at 1.93 and
1.94 ꢁ, are slightly elongated with respect to those of previ-
ous complexes; however, these values do compare favour-
Figure 6. Imidazolium ligand precursors 1 f–1i.
ACHTUNGTRENNUNG
sibility of chelation by binding to the metal centre through
both the carbenic carbon and pyridyl nitrogen atoms, with
ligand 1i having two potential NHC donors. In the solid-
state structures of imidazolium salts 1 f, 1g and 1i, the pyrid-
yl and imidazolium rings lie almost coplanar, which increas-
es p-overlap between the two rings (Figure 7). This signifi-
cant p-overlap between the two aromatic systems results in
a particularly low-field shift of the imidazolium C2 proton
ACHTUNGTRENNUNG
ꢀ
ꢀ
The Cu C and Cu Br bond lengths, at 1.92 and 2.28 ꢁ, re-
spectively, are within typical ranges for this type of complex.
The 2-methoxyphenyl ring is twisted, with respect to the
plane defined by the NHC ring, by approximately 478. This
16782
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 16780 – 16790

Copper(I)–N-Heterocyclic Carbene Complexes
FULL PAPER
plexes 2a–2e. However, reaction of 1 f with Cu2O at reflux
in dichloromethane, in the presence of 4 ꢁ molecular sieves,
led to the isolation of an intractable green oil. Therefore,
another approach for the synthesis of desired [Cu-
AHCTUNGTREG(NNNU NHC)Br]-type complex, 2 f, was followed. Complex 1 f was
treated with one equivalent of CuBr in the presence of an
excess of Cs2CO3 in anhydrous acetonitrile at 508C. Cs2CO3
is too weak a base to cause allylic-rearrangement reactions
to occur, but is strong enough so that deprotonation of 1 f
can occur. After workup, a yellow solid was obtained (2 f’),
1
which was fully characterised by H and 13C{1H} NMR spec-
troscopy, high-resolution mass spectrometry and elemental
1
analysis. H NMR spectroscopy revealed the complete loss
of the low-field imidazolium proton resonance, and the pres-
ence of
a broad low-field resonance at d=183.7 ppm
(126 MHz, CD3CN) in the 13C{1H} NMR spectrum strongly
suggested the formation of a copper(I)–NHC complex. Ele-
mental analysis shows that, instead of a [CuCAHTUNGTRENNNUG
plex, 2 f’ was a bis-NHC complex ([Cu
The formation of a bis-NHC
complex is also implied in the
high-resolution mass spectrum,
in which the dominant molecu-
lar ion corresponds to [Cu-
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
(NHC)2]+. However, great care
should be taken when interpret-
ing the ESI mass spectra of
copper(I)–NHC complexes be-
cause their labile nature can
Figure 8. Suggested structure
of complex 2 f’.
lead to ligand scrambling, resulting in the appearance of bis-
NHC species in samples of pure mono-NHC compounds.
This ligand-scrambling behaviour is well-documented for sil-
ver(I)–NHC complexes.[38] Complex 2 f’ is proposed to be an
ionic complex on the basis of its poor solubility in common
organic solvents (e.g. dichloroACTHNUGRTNENUGmethACHTUNGTRENNaUGN ne, chloroform and tet-
rahydrofuran), and slightly better solubility in more polar
acetonitrile.
A different method was required to synthesise the [Cu-
AHCTUNGTREG(NNUN NHC)Br] complex of ligand precursor 1 f. Reaction of an
Figure 7. Molecular structure of imidazolium salts 1 f, 1g and 1i. Ellip-
soids are shown at 50% probability level. Hydrogen atoms and solvent
molecules (acetonitrile molecules in compound 1g, and a water molecule
and an acetonitrile molecule in compound 1i) have been omitted for
clarity.
excess of Ag2O with 1 f in anhydrous dichloromethane, in
the presence of 4 ꢁ molecular sieves, led to the formation
of a dark suspension that after filtration, removal of the sol-
vent and recrystallisation, yielded silver(I)–NHC complex
Ag-2 f (Scheme 2).
1
resonances in the H NMR spectra, with that of compound
1g being observed at d=12.22 ppm (300 MHz, CDCl3).
There is a strong anion–p interaction between the N C N
ꢀ ꢀ
1
p-cloud and a neighbouring bromide anion in these ligand
precursors, interatomic distances falling well below the sum
of the van der Waals radii (3.36 ꢁ in 1 f, 3.31 ꢁ and 3.32 ꢁ
in 1g and 3.29 ꢁ in 1i). Anion–p interactions are much less
common than cation–p interactions; this interaction is facili-
tated by the particularly electron-poor (and positively
Characterisation of Ag-2 f by H NMR spectroscopy indi-
cated the loss of the imidazolium proton resonance, and for-
mation of the silver(I)–NHC complex was confirmed by the
13C{1H} NMR spectrum, which showed a low-field resonance
at d=181.6 ppm (75 MHz, CDCl3). Ag-2 f crystallises as
+
ꢀ
ꢀ
[AgACTHNURTGNE(UNG NHC)2] AgBr2 , in which the C Ag bonds lie almost
ꢀ ꢀ
ꢀ
charged) N C N moiety of the imidazolium owing to the
perpendicular to the Br Ag bonds. The two silver centres
are linked by a weak argentophilic interaction (see the Sup-
porting Information). Despite the decomposition observed
during the reaction of 1 f with Cu2O, transmetalation from
Ag-2 f was performed successfully, under strictly inert condi-
tions (Scheme 2).
effect of the electron-withdrawing pyridyl substituents on
the imidazolium ring. The presence of anion–p interactions
in these structures may facilitate the use of imidazolium
salts as anion binders.[36,37]
The reaction of Cu2O with ligand precursors 1a–1e led to
the formation of analytically pure copper(I)–NHC com-
Chem. Eur. J. 2013, 19, 16780 – 16790
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16783

C. E. Willans and B. R. M. Lake
solid-state that could be formed selectively by varying the
crystallisation conditions.
Elemental analysis on a sample of 2 f strongly indicated
that the target complex, [CuACHTNUTRGNE(UNG NHC)Br], had been successful-
ly formed. To confirm this, single crystals suitable for X-ray
diffraction analysis were grown from a dilute solution of the
product in a mixture of dichloromethane and pentane (1:2).
Complex 2 f crystallised as a one-dimensional polymer,
propagated by coordination of the copper(I) centre of one
unit to the alkene of a neighbouring unit (Figure 10). This
result is the first example of alkene coordination to a cop-
per(I)–NHC complex. Interestingly, although alkene coordi-
nation to the copper(I) centre is observed in the solid state,
Scheme 2. Synthesis of silver(I)–NHC complex Ag-2 f and transmetala-
tion to give copper(I)–NHC complex 2 f.
1
noticeable shifts in the resonances of the alkenic H nuclei
1
were not observed in the H NMR spectrum at room tem-
1
Complex 2 f was characterised by H NMR spectroscopy,
perature, thus indicating that alkene coordination to the
metal centre is highly labile. The coordination geometry
around the copper centre is best described as distorted trigo-
nal pyramidal, with the alkene (considered as a monodentate
ligand with regards to geometry), carbenic carbon and bro-
mide atoms occupying the basal positions (bond angles of
130.748, 115.308 and 112.608) and the pyridyl nitrogen atom
occupying the apical position. Significant distortion of an
idealised trigonal-pyramidal geometry occurs, in part, be-
cause of the geometric restriction imposed by the 5-mem-
bered chelate ring, which has a bite angle of 74.98 as defined
by the C-Cu-N angle. The chelate ring itself is essentially
planar. Notably, long bonds are observed between the NHC
ligand and copper centre (1.97 ꢁ) and between the pyridyl
nitrogen atom and copper centre (2.50 ꢁ), possibly as a con-
sequence of the copper centre trying to optimise orbital
high-resolution mass spectrometry and elemental analysis.
1
The H NMR spectrum showed that the resonances of many
of the ligand protons were broadened considerably in com-
parison with Ag-2 f and 2 f’. The protons most affected by
the spectral broadening included all but one of the pyridyl
protons, the backbone NHC protons and the allylic methyl-
ene protons. One of the pyridyl protons and the alkenic pro-
tons retain reasonably sharp resonances (Figure 9).
ꢀ
overlap with both donors. The Cu Br bond, at 2.44 ꢁ, is rel-
ꢀ
atively long with respect to the terminal Cu Br bonds de-
scribed for complexes 2b and 2e (2.28/2.28 and 2.28 ꢁ) but
ꢀ
compares well with the bridging Cu Br bonds of 2a (2.44
and 2.46 ꢁ). This lengthening may result from a need to
reduce steric clashing around the four-coordinate copper(I)
centre, and also from the possible increase in electron densi-
ty at the copper(I) centre, caused by having three other
donor moieties bound to the metal.
Figure 9. 1H NMR spectrum (4.4–8.8 ppm region) of complex 2 f
(300 MHz, CDCl3).
Alkene coordination in complex 2 f results in a pro-
nounced lengthening of the alkenic C=C bond from 1.33 ꢁ
(1 f) to 1.37 ꢁ (2 f). The coordination of the alkene to the
metal centre appears to have been facilitated by the addi-
tion of the pyridyl group to the NHC ligand. The pyridyl
group acts as both an ancillary donor and also as a modifier
of the electronic properties of the NHC donor. With this in
mind, it is unsurprising that many of the structurally charac-
terised copper(I)–alkene complexes reported thus far con-
tain chelating N-donor ligands such as tris(pyrazolyl)bo-
rate.[2,39] However, the exact role of the pyridyl ring in ena-
bling alkene coordination to the metal centre in this case is
unclear; the pyridyl ring may be altering the properties of
the copper(I) centre directly, by coordination to it, or it may
be indirectly affecting the copper(I) centre by modifying the
donor ability of the NHC ligand (decreasing the energy of
the LUMO).
A similar observation was made previously for a related
complex by Danopoulos.[24] The spectral broadening is at-
tributed to non-rigidity in solution, although it is suggested
that an equilibrium involving mono- and bis-NHC com-
plexes could contribute to the broadening. It is notable that
the protons with the most broadened resonances are those
in closest proximity to the metal (that is, nearest the carben-
ic carbon and pyridyl nitrogen atoms). On this basis, it is
proposed that the spectral broadening is caused by coordi-
native lability at the copper(I) centre. This lability may lead
to the formation of mono- and bis-NHC complexes along
with mono-NHC dimers. Indeed, from a slightly different
ligand, Danopoulos and co-workers were able to character-
ise two different complexes (a dimer and a polymer) in the
16784
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 16780 – 16790

Copper(I)–N-Heterocyclic Carbene Complexes
FULL PAPER
Figure 10. Molecular structure of 2 f (top) and one-dimensional polymer propagated by coordination of the copper(I) centre of one unit to the alkene of
a neighbouring unit (bottom). Ellipsoids are shown at 50% probability level. Hydrogen atoms have been omitted for clarity.
Slight modification of 1 f, by the addition of a nitro group
to the pyridyl ring, gave 1g. The intermediate silver(I)–
NHC complex Ag-2g was found to be less soluble than Ag-
2 f; despite the poor solubility, analysis by 1H NMR spec-
troscopy indicated loss of the imidazolium proton resonance.
Complex Ag-2g was used to synthesise the corresponding
copper(I)–NHC complex without further purification or
1
analysis. Full characterisation of 2g by H and 13C{1H} NMR
spectroscopy, high-resolution mass spectrometry and ele-
mental analysis indicated the presence of analytically pure
[CuACHTUNGTRENNUNG(NHC)Br]. Broadening of certain resonances was ob-
1
Figure 11. Molecular structure of 2g. Ellipsoids are shown at 50% proba-
bility level. Hydrogen atoms have been omitted for clarity.
served in the H NMR spectrum of 2g. However, it should
be noted that the resonance broadening is not nearly as pro-
nounced as that for 2 f, possibly indicating that 2g is a much
less labile complex. Single crystals suitable for X-ray diffrac-
tion analysis were grown by the slow evaporation of a con-
centrated solution of 2g in acetonitrile (Figure 11).
Complex 2g crystallised as a monomer, with the coordina-
tion environment around the copper(I) centre best de-
scribed as distorted T-shaped. A similar coordination envi-
ronment has been reported previously by Danopoulos et
al.[24] The extended structure of 2g shows that alkene coordi-
nation is switched off by addition of the nitro group, with
the closest approach of an alkene to the copper(I) centre oc-
curring at a distance of 5.23 ꢁ. The alkenic C=C bond at
1.32 ꢁ is shorter than in complex 2 f (1.37 ꢁ), and is closer
ꢀ
the C-Cu-Br bond angle (176.88). The Cu N bond length on
the other hand is extremely long (2.56 ꢁ), implying the pres-
ence of a weak interaction between the pyridyl nitrogen and
copper(I) centre. The fact that this bond is lengthened with
ꢀ
respect to the Cu N bond of 2 f (2.50 ꢁ), despite the lower
coordination number of 2g, suggests that the primary effect
of adding the nitro group has been to inductively withdraw
electron density away from the pyridyl nitrogen atom,
making it a poorer donor to the metal centre.
Ligand 1h, which contains an electron-donating para-
methACTHUNRTGNEUNGoxy substituent (with respect to the pyridyl nitrogen
atom), was prepared and coordinated to the copper centre
using the same route as outlined in Scheme 2. The resulting
copper(I)–NHC complex (2h) was characterised by
1H NMR spectroscopy, high-resolution mass spectrometry
ꢀ
in length to the ligand precursor 1g (1.33 ꢁ). The Cu C and
ꢀ
Cu Br distances, at 1.91 and 2.29 ꢁ, respectively, are within
the range expected for a linear [Cu
(NHC)Br] complex, as is
Chem. Eur. J. 2013, 19, 16780 – 16790
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16785

C. E. Willans and B. R. M. Lake
1
and elemental analysis. The H NMR spectrum shows signif-
tempted using the procedure described for the synthesis of
1 f’, instead of the procedure involving the silver(I)–NHC
complex because bis-imidazolium salts are known to require
more forcing conditions when reacting with Ag2O.[40] The
1
icant resonance broadening, similar to the H NMR of com-
plex 2 f. Single crystals suitable for X-ray diffraction analysis
were obtained from a solution of the complex in CH2Cl2/n-
hexane (1:3) (Figure 12).
1
solubility of the product was reasonably poor, although H
and 13C{1H} NMR spectra with good signal-to-noise ratios
could be obtained at elevated temperatures (333.6 K) with
1
a large number of scans. The H NMR spectrum illustrates
the presence of only one ligand environment and also the
loss of the imidazolium proton. Unlike the spectra for com-
plexes 2 f–2h, the spectrum of the product is sharp, perhaps
indicating that this complex is less labile than complexes
2 f–2h. The 13C{1H} NMR spectrum shows a characteristic
low-field resonance at d=181.9 ppm (126 MHz, CD3CN,
333.6 K), indicating the formation of a copper(I)–NHC com-
plex. The high-resolution mass spectrum was dominated by
a peak at m/z 354.08, which was found to correspond to
a [Cu2L2]2+ fragment. Single crystals suitable for X-ray dif-
fraction analysis were obtained by cooling a saturated solu-
tion of the product from 333 K to ambient temperature in
anhydrous acetonitrile. The solid-state structure shows a for-
Figure 12. Molecular structure of 2h. Ellipsoids are shown at 50% proba-
bility level. Hydrogen atoms have been omitted for clarity.
mulation of [Cu3ACHTNUTRGNE(UGN NHC)2Br3] for 2i (Figure 13).
Complex 2i was found to crystallise as a ligand-supported
cuprophilic trimer. Cuprophilic trimers that have similarities
to 2i have been described previously in the literature,[41] al-
though many of these occur as C3-symmetric triangular cop-
per(I) clusters.[42–44] The geometries (excluding possible cu-
prophilic interactions) about all three of the copper(I) cen-
tres are best described as distorted linear, with bond angles
Complex 2h was found to crystallise as a monomer, anal-
ogous to complex 2g, with the coordination environment
around the copper(I) centre being distorted T-shaped. The
ꢀ
ꢀ
Cu C (1.91 ꢁ) and Cu Br (2.27 ꢁ) distances are compara-
ꢀ
ble to those found in complex 2g, whereas the Cu N dis-
tance is markedly shortened in complex 2h (2.31 ꢁ) com-
pared with 2g (2.56 ꢁ), and indeed, 2 f (2.50 ꢁ). This indi-
cates that the ability of the pyridyl moiety to act as a s-
donor is significantly enhanced as a result of the para-meth-
ꢀ
ꢀ
of 171.08, 164.68 and 156.98. The Cu C and Cu Br bond
lengths, in complex 2i, are within typical ranges for cop-
per(I)–NHC complexes. The distances between the three
copper(I) centres are 2.87 ꢁ and 2.78 ꢁ, and thus, possibly
represent weak cuprophilic interactions. It is notable that
pyridyl coordination does not occur; the closest copper(I)–
pyridyl interaction occurs between Cu(3) and N(8) at a dis-
tance of 2.64 ꢁ. This lack of coordination presumably results
from the conformational restriction imposed by the ligand
on the positions of the copper(I) centres (i.e. away from the
ACHTUNGTRENNUNGoxy group, leading to a much stronger interaction between
the pyridyl nitrogen atom and copper(I) centre. Surprisingly,
the extended structure of 2g does not show any interaction
between the copper(I) centre and an alkene, with the closest
ꢀ
approach occurring at a distance of 4.09 ꢁ. The short Cu N
distance is likely to destabilise the filled frontier orbital at
the copper(I) centre, thus, stabilisation of copper(I)–alkene
coordination through increased
copper(I)!ethylene
p
back-
bonding would be expected.
The absence of alkene coordi-
nation in this complex indicates
that components other than
charge transfer may be playing
a role, with polarisation and
frozen-density terms becoming
less favourable with increased
electron density on the metal
centre.
To examine the effect of an
extra NHC donor on copper(I)
coordination, a symmetrical bis-
NHC analogue of 1 f was syn-
thesised (1i). Complexation of
Figure 13. Molecular structure of 2i (left) and 2i’ (right). Ellipsoids are shown at 50% probability level. Hy-
1i to a copper(I) centre was at- drogen atoms and bromide anions in 2i’ have been omitted for clarity.
16786
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 16780 – 16790

Copper(I)–N-Heterocyclic Carbene Complexes
FULL PAPER
pyridyl nitrogen atoms), but may also result from the with-
drawal of electron density from the pyridyl nitrogen atoms
by the NHC rings, making the pyridyl rings poorer donors.
Twisting of the NHC rings, with respect to the pyridyl rings,
occurs to varying degrees (torsion angles between 10.38 and
31.88); this twisting acts to decrease p-overlap between the
NHC and pyridyl rings. Possibly as a result of the lack of co-
ordination of the pyridyl rings to the copper(I) centre in 2i,
no alkene coordination is observed; it appears that all three
copper(I) centres are coordinatively saturated in complex
2i.
back-bonding is the dominant effect in these complexes, and
that increasing the electron density of the metal centre
serves to stabilise copper(I)–alkene coordination. However,
increasing the electron density of the metal centre even fur-
ther, through the addition of a methoxy group to the pyridyl
moiety (complex 2h), does not result in a copper(I)–alkene
interaction. This observation is interesting considering the
ꢀ
relatively short Cu N bond. This indicates that effects other
than charge transfer cannot be ruled out, and that low fa-
vourable polarisation and repulsive frozen-density terms
have come into effect.
An explanation for the dominant ion in the high-resolu-
tion mass spectrum being a [Cu2L2]2+ fragment, and an ex-
planation for the highly symmetric NMR data, was found by
serendipitous crystallisation. Complex 2i’ crystallised during
the slow evaporation of a sample of 2i exposed to air and
moisture (Figure 13). It is evident that exposure of samples
of 2i to air and moisture results in the formation of helicate
2i’. The highly symmetric NMR data indicates that helicate
2i’ is the dominant species in the NMR solvent (with con-
comitant release of CuBr from 2i). Elemental analytical
data is indicative of a 3:1 mixture of 2i/2i’ in the final prod-
uct. The stability of 2i’ is a somewhat surprising result, espe-
cially given the tendency of copper(I)–NHC complexes
bearing non-bulky N-substituents (i.e. allyl) to decompose,
reforming imidazolium salts and unidentifiable copper(II)-
containing detritus. The formation of a copper(I)–NHC
complex over any other species suggests that the 16-mem-
bered metallacycle contained within 2i’ is highly stable,
even after prolonged exposure to air and moisture. The re-
markable stability of 2i’ makes it an interesting candidate
for use in catalysis.
The coordination of pyridyl, NHC and alkene moieties of
one ligand to the same copper(I) centre would result in high
angular strain, hence the metal centre instead bonds to
a nearby alkene to form a 1D coordination polymer. Com-
plex 2 f is an extremely interesting candidate for catalytic
processes, in which binding of alkenes to the copper(I)
centre is proposed. Furthermore, applications such as sepa-
ration chemistry, gas purification or sensing, which all rely
on the reversible nature of a copper(I)–alkene interaction
(or a copper(I)–carbonyl interaction), could make use of
this ligand system. These binding interactions are clearly
more complex than simply increasing the basicity of the
ligand to stabilise alkene coordination, with both charge-
transfer and non-charge-transfer components requiring con-
sideration. This is in agreement with recent studies of York
et al. on copper(I)–alkene binding using DFT calcula-
tions.[14]
Experimental Section
General: Where stated, manipulations were performed under an atmos-
phere of dry argon by means of standard Schlenk line or glovebox tech-
niques. The gas was dried by it passing through a twin-column drying ap-
paratus containing molecular sieves (4 ꢁ) and P2O5. Anhydrous solvents
were prepared by passing the solvent over activated alumina to remove
water, copper catalyst to remove oxygen and molecular sieves to remove
any remaining water by using the Dow–Grubbs solvent system. Deuterat-
ed chloroform, acetonitrile and DMSO were dried over CaH2, cannula
filtered or distilled, and then degassed using freeze-pump-thaw cycles
prior to use. All other reagents and solvents were used as supplied.
1H and 13C{1H} NMR spectra were recorded on either a Bruker DPX300
or a Bruker AV500 spectrometer. The values of chemical shifts are given
in ppm and values for coupling constants (J) in Hz. Assignment of some
1H NMR spectra was aided by the use of 2D 1H–1H COSY experiments
and the assignment of some 13C{1H} NMR spectra was aided by 13C{1H}
dept 135 experiments. Mass spectra were collected on a Bruker Daltonics
(micro TOF) instrument operating in the electrospray mode. Microanaly-
ses were performed using a Carlo Erba Elemental Analyser MOD 1106
spectrometer. X-ray diffraction data were collected on either a Bruker
Nonius X8 diffractometer fitted with an Apex II detector with Mo-Ka ra-
diation (l=0.71073 ꢁ), or an Agilent SuperNova diffractometer fitted
with an Atlas CCD detector with Mo-Ka radiation (l=0.71073 ꢁ). Crys-
tals were mounted under oil on glass or nylon fibres. Data sets were cor-
rected for absorption using a multiscan method, and the structures were
solved by direct methods using SHELXS-97 and refined by full-matrix
least squares on F2 using SHELXL-97, interfaced through the program
X-Seed. Molecular graphics for all structures were generated using POV-
RAY in the X-Seed program.
Conclusion
A series of copper(I)–NHC complexes with N-allyl substitu-
ents and varying steric and electronic properties have been
described. The complexes exhibit a range of fascinating
structures; copper(I)–alkene coordination in the solid state
being observed in one of the complexes (complex 2 f).
There are two possible explanations for this observation:
1) The co-planar pyridyl group withdraws electron density
from the carbenic carbon atom, thus reducing electron den-
sity on the copper(I) centre and increasing polarisation ef-
fects between the metal and the alkene. 2) The co-planar
structure naturally enforces the pyridyl-N atom towards the
copper(I) centre to form a dative bond. This dative bond, al-
though weak, increases the electron density of the copper(I)
centre, thus increasing the affinity of the copper(I) centre
for the alkene through its enhanced metal(d)-to-alkene(p*)
back-bonding ability.
The addition of a nitro group to the pyridyl substituent
ꢀ
significantly increases the pyridyl Cu N bond length in com-
plex 2g (compared with 2 f). This increase in bond length
switches off the interaction between the copper(I) centre
and the alkene, suggesting that metal(d)-to-alkene(p*)
1,3-Diallylbenzimidazol-2-ylidene copper(I) bromide (2a):
A Schlenk
flask was charged with 1,3-diallylbenzimidazolium bromide (0.42 g,
Chem. Eur. J. 2013, 19, 16780 – 16790
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16787

C. E. Willans and B. R. M. Lake
1.5 mmol), Cu2O (0.43 g, 3.0 mmol) and 4 ꢁ molecular sieves. The con-
tents of the flask were dried and degassed thoroughly in vacuo. Anhy-
drous dichloromethane (30 mL) was added through a cannula and the
mixture was stirred at reflux for 24 h. After this time, the mixture was al-
lowed to cool and was filtered under inert conditions. The solvent was re-
moved from the filtrate in vacuo to give the product as a white solid.
Yield: 0.30 g, 0.88 mmol, 59%. Single crystals suitable for X-ray diffrac-
tion analysis were grown by the slow evaporation of a concentrated solu-
C15H18BrCuN2.2= CH2Cl2: C 44.13, H 4.57, N 6.57; found: C 44.60, H 5.00,
N 6.10.
3
1-Allyl-3-(2-methoxybenzene)imidazol-2-ylidene copper(I) bromide (2e):
A Schlenk flask was charged with 1-allyl-3-(2-methoxybenzene)imidazoli-
um bromide (0.58 g, 2.0 mmol), Cu2O (0.57 g, 4.0 mmol) and 4 ꢁ molecu-
lar sieves. The contents of the flask were dried and degassed thoroughly
in vacuo. Anhydrous dichloromethane (30 mL) was added through a can-
nula and the mixture was stirred at reflux for 74 h. After this time, the
mixture was allowed to cool and was filtered under inert conditions. The
solvent was removed from the filtrate in vacuo to give the product as
1
tion of the product in chloroform. H NMR (500 MHz, CDCl3): d=7.46–
7.42 (m, 2H, bimH), 7.41–7.36 (m, 2H, bimH), 6.04 (ddt, J=17.0, 10.3,
5.7 Hz, 2H, CH=CH2), 5.36 (dd, J=10.3, 0.8 Hz, 2H, HC=CHH(cis)), 5.29
(dd, J=17.0, 0.8 Hz, 2H, HC=CHH(trans)), 5.06 ppm (d, J=5.7 Hz, 4H,
NCH2); 13C{1H} NMR (126 MHz, CDCl3): d=186.1, 133.8, 131.9, 124.1,
119.5, 111.82, 51.5 ppm; HRMS (ESI+): m/z calcd for [MꢀBr]+:
261.0448; found: 261.0440; elemental analysis calcd (%) for
C13H14BrCuN2: C 45.69, H 4.13, N 8.20; found: C 45.40, H 4.30, N 7.90.
a
yellow oil, which slowly crystallised on standing. Yield: 0.41 g,
1.2 mmol, 58%. Single crystals suitable for X-ray diffraction analysis
were grown by the slow evaporation of a concentrated solution of the
product in dichloromethane. 1H NMR (300 MHz, CDCl3): d=7.51 (dd,
J=8.0, 1.4 Hz, 1H, ArH), 7.41 (td, J=8.0, 1.4 Hz, 1H, ArH), 7.18 (d, J=
1.7 Hz, 1H, imH), 7.06 (t, J=8.0 Hz, 2H, ArH), 7.01 (d, J=1.7 Hz, 1H,
imH), 6.04 (ddt, J=16.5, 10.3, 6.0 Hz, 1H, CH=CH2), 5.42–5.30 (m, 2H,
HC=CH2), 4.85 (d, J=6.0 Hz, 2H, NCH2), 3.85 ppm (s, 3H, OCH3);
13C{1H} NMR (75 MHz, CDCl3): d=178.6, 153.1, 132.5, 130.4, 128.4,
127.5, 123.4, 121.2, 120.0, 119.8, 112.5, 56.0, 54.2 ppm; HRMS (ESI+):
m/z calcd for [MꢀBr]+: 277.0397; found: 277.0360; elemental analysis
calcd (%) for C13H14BrCuN2O: C 43.65, H 3.94, N 7.83; found: C 44.00,
H 4.40, N 7.35.
1-Allyl-3-methylimidazol-2-ylidene copper(I) bromide (2b): A Schlenk
flask was charged with 1-allyl-3-methylimidazolium bromide (0.60 g,
3.0 mmol), Cu2O (0.85 g, 5.9 mmol) and 4 ꢁ molecular sieves. The con-
tents of the flask were dried and degassed thoroughly in vacuo. Anhy-
drous dichloromethane (40 mL) was added through a cannula and the
mixture was stirred at reflux for 80 h. After this time, the mixture was al-
lowed to cool and was filtered under inert conditions. The solvent was re-
moved from the filtrate in vacuo to give the product as a white solid.
Yield: 0.39 g, 1.5 mmol, 50%. Single crystals suitable for X-ray diffrac-
tion analysis were grown by the slow evaporation of a concentrated solu-
1-Allyl-3-(2-pyridyl)imidazol-2-ylidene copper(I) bromide (2 f): 1-Allyl-3-
(2-pyridyl)imidazolium bromide (0.23 g, 0.85 mmol) and Ag2O (0.13 g,
0.55 mmol) were added to a Schlenk flask along with 4 ꢁ molecular
sieves. Anhydrous dichloromethane (30 mL) was added and the dark sus-
pension was stirred overnight in the absence of light. After this time, the
dark suspension was filtered and the solvent removed in vacuo to yield
Ag-2 f as a white oily solid, which was used without further purification.
In a glovebox, Ag-2 f was dissolved in dichloromethane (35 mL). With
stirring, solid CuBr was added (0.13 g, 0.93 mmol). Immediately, the solu-
tion began to turn yellow with the concomitant formation of a pale
yellow precipitate (AgBr). The suspension was stirred for 2 h and then
filtered to remove the insoluble AgBr/CuBr. The solid was washed with
dichloromethane until the filtrate ran clear. The solvent was removed
from the filtrate in vacuo to yield the product as a yellow crystalline
powder. Yield: 0.20 g, 0.62 mmol, 73%. Single crystals suitable for X-ray
diffraction analysis were grown on standing of a dilute solution of the
product in dichloromethane/pentane (1:2). 1H NMR (300 MHz, CDCl3):
d=8.55 (br s, 1H), 8.08 (br s, 1H), 7.91 (t, J=7.7 Hz, 1H), 7.80 (br s,
1H), 7.37 (br s, 1H), 7.10 (br s, 1H), 6.12–5.94 (m, 1H, CH=CH2), 5.39
(d, J=10.0 Hz, 1H, CH=CHH(cis)), 5.36 (d, J=17.0 Hz, 1H, CH=
CHH(trans)), 4.88 ppm (br s, 2H, NCH2), definite assignment of the aro-
matic resonances to either pyridyl or backbone imidazolylidene protons
is hampered because of spectral broadening; HRMS (ESI+): m/z calcd
for [MꢀBr+MeCN]+: 289.0509; found: 289.0517; elemental analysis
calcd (%) for C11H11BrCuN3: C 40.20, H 3.37, N 12.78; found: C 39.90, H
3.30, N 12.50.
1
tion of the product in dichloromethane. H NMR (300 MHz, CDCl3): d=
6.91 (s, 1H, imH), 6.91 (s, 1H, imH), 5.94 (ddt, J=16.3, 10.2, 6.0 Hz, 1H,
CH=CH2), 5.36–5.20 (m, 2H, HC=CH2), 4.72 (d, J=6.0 Hz, 2H, NCH2),
3.83 ppm (s, 3H, CH3); 13C{1H} NMR (75 MHz, CDCl3): d=178.1, 132.6,
122.0, 120.7, 119.8, 53.8, 38.4 ppm; HRMS (ESI+): m/z calcd for
[MꢀBr]+: 185.0135; found: 185.0145; elemental analysis calcd (%) for
C7H10BrCuN2: C 31.65, H 3.79, N 10.55; found: C 31.30, H 3.80, N 10.10.
1,3-Diallylimidazol-2-ylidene copper(I) bromide (2c): A Schlenk flask
was charged with 1,3-allylimidazolium bromide (0.54 g, 2.4 mmol), Cu2O
(0.67 g, 4.7 mmol) and 4 ꢁ molecular sieves. The contents of the flask
were dried and degassed thoroughly in vacuo. Anhydrous dichloro
ACHTUNGTNERmNUNG eth-
ACHTUNGTRENNUNG
reflux for 24 h. After this time, the mixture was allowed to cool and was
filtered under inert conditions. The solvent was removed from the filtrate
in vacuo to give the product as a pale yellow oil. Yield: 0.35 g, 1.2 mmol,
51%. 1H NMR (300 MHz, CDCl3): d=6.93 (s, 2H, imH), 5.95 (ddt, J=
16.6, 10.2, 6.0 Hz, 2H, CH=CH2), 5.31 (dd, J=10.2, 0.9 Hz, 2H, HC=
CHH(cis)), 5.26 (dd, J=16.6, 0.9 Hz, 2H, HC=CHH(trans)), 4.73 ppm (d, J=
6.0 Hz, 4H, NCH2); 13C{1H} NMR (75 MHz, CDCl3): d=177.8, 132.6,
120.8, 120.0, 54.0 ppm; HRMS (ESI+): m/z calcd for [MꢀBr]+: 211.0291;
found: 211.0280; elemental analysis calcd (%) for C9H12BrCuN2: C 37.35,
H 4.20, N 9.70; found: C 37.06, H 4.15, N 9.61.
1-Allyl-3-mesitylimidazol-2-ylidene copper(I) bromide (2d): A Schlenk
flask was charged with 1-allyl-3-mesitylimidazolium bromide (0.32 g,
1.0 mmol), Cu2O (0.30 g, 2.1 mmol) and 4 ꢁ molecular sieves. The con-
tents of the flask were dried and degassed thoroughly in vacuo. Anhy-
drous dichloromethane (20 mL) was added through a cannula and the
mixture was stirred at reflux for 60 h. After this time, the mixture was al-
lowed to cool and was filtered under inert conditions. The solvent was re-
moved from the filtrate in vacuo to give the product as a clear oil, which
slowly crystallised on standing. Yield: 0.24 g, 0.64 mmol, 62%. Single
crystals suitable for X-ray diffraction analysis were grown by the vapour
diffusion of pentane in to a concentrated solution of the product in di-
chloromethane. The X-ray crystal structure showed a complex of the
1-Allyl-3-(2-(5-nitro)pyridyl)imidazol-2-ylidene copper(I) bromide (2g):
1-Allyl-3-(2-(5-nitropyridyl)imidazolium bromide (0.31 g, 1.0 mmol) and
Ag2O (0.19 g, 0.82 mmol) were added to a Schlenk flask along with 4 ꢁ
molecular sieves. Anhydrous dichloromethane (30 mL) was added and
the mixture was stirred in the absence of light at ambient temperature
for 72 h. After this time, the mixture was filtered through Celite and the
solid washed with copious dichloromethane. The combined filtrate was
collected and reduced in volume to 30 mL. Addition of diethyl ether
(100 mL) led to the precipitation of the Ag–NHC complex as white nee-
dles. The Ag–NHC complex (0.082 g, 0.20 mmol) and CuBr (0.056 g,
0.39 mmol) were added to a Schlenk flask in a glovebox. Anhydrous di-
chloromethane (25 mL) was added and the resulting suspension was
stirred at ambient temperature in the absence of light for 3.5 h. After this
time, the suspension was filtered to remove AgBr/CuBr and the pale
orange filtrate collected. Anhydrous pentane (40 mL) was added to the
stirring filtrate and a crystalline red solid formed, collected by filtration
and dried in vacuo. Yield: 0.046 g, 0.12 mmol, 63%. Single crystals suit-
type [CuACHTUNGTRENNUNG
(NHC)2Br]. 1H NMR (300 MHz, CDCl3): d=7.10 (d, J=1.7 Hz,
1H, imH), 6.95 (s, 2H, ArH), 6.87 (d, J=1.7 Hz, 1H, imH), 6.03 (ddt, J=
15.8, 10.2, 5.8 Hz, 1H, CH=CH2), 5.35 (dd, J=10.2, 0.7 Hz, 1H, HC=
CHH(cis)), 5.26 (d, J=15.8 Hz, 1H, HC=CHH(trans)), 4.85 (d, J=5.8 Hz,
2H, NCH2), 2.31 (s, 3H, p-CH3), 2.00 ppm (s, 6H, o-CH3); 13C{1H} NMR
(75 MHz, CDCl3): d=179.0, 139.5, 135.2, 134.7, 132.7, 129.5, 122.5, 120.7,
119.7, 53.9, 21.2, 17.9 ppm; HRMS (ESI+): m/z calcd for [MꢀBr+
MeCN]+: 330.1026; found: 330.1024; elemental analysis calcd (%) for
AHCTUNGTRENNUNG
16788
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 16780 – 16790

Copper(I)–N-Heterocyclic Carbene Complexes
FULL PAPER
CDCl3): d=9.38 (br s, 1H, pyH), 8.72 (dd, J=8.8, 2.6 Hz, 1H, pyH), 8.59
(br d, J=8.8 Hz, 1H, pyH), 7.97 (br s, 1H, imH), 7.16 (br s, 1H, imH),
6.05 (ddt, J=16.3, 10.2, 6.1 Hz, 1H, CH=CH2), 5.50–5.38 (m, 2H, CH=
CH2), 4.92 ppm (d, J=6.1 Hz, 2H, NCH2); 13C{1H} NMR (126 MHz,
CDCl3): d=145.2, 135.1, 131.5, 121.4, 119.8, 114.4, 55.4 ppm (some of the
13C resonances are not observed due to broadening); HRMS (ESI+):
m/z calcd for [MꢀBr+MeCN]+: 334.0360; found: 334.0371; elemental
analysis calcd (%) for C11H10BrCuN4O2: C 35.36, H 2.70, N 14.99; found:
C 35.45, H 2.65, N 14.75.
[5] P. J. Perez, M. M. Diaz-Requejo, Comprehensive Organometallic
Chemistry III, Vol. 2 (Eds.: R. H. Crabtree, D. M. P. Mingos),
Elsevier, Oxford, 2007, Chapter 2.03.
[6] F. I. Rodrꢂguez, J. J. Esch, A. E. Hall, B. M. Binder, G. E. Schaller,
A. B. Bleecker, Science 1999, 283, 996.
[7] T. Suzuki, R. D. Noble, C. A. Koval, Inorg. Chem. 1997, 36, 136.
[8] B. Esser, T. M. Swager, Angew. Chem. 2010, 122, 9056; Angew.
Chem. Int. Ed. 2010, 49, 8872.
[9] H. V. R. Dias, J. Wu, Eur. J. Inorg. Chem. 2008, 509.
[10] H. V. R. Dias, C. J. Lovely, Chem. Rev. 2008, 108, 3223.
[11] X. L. Dai, T. H. Warren, Chem. Commun. 2001, 1998.
[12] J. S. Thompson, R. L. Harlow, J. F. Whitney, J. Am. Chem. Soc. 1983,
105, 3522.
[13] V. A. K. Adiraju, J. A. Flores, M. Yousufuddin, H. V. R. Dias, Orga-
nometallics 2012, 31, 7926.
[14] N. C. Pernicone, J. B. Geri, J. T. York, Theor. Chem. Acc. 2012, 131,
1105.
[15] J. C. Y. Lin, R. T. W. Huang, C. S. Lee, A. Bhattacharyya, W. S.
Hwang, I. J. B. Lin, Chem. Rev. 2009, 109, 3561.
[16] L. Dang, H. Zhao, Z. Lin, T. B. Marder, Organometallics 2008, 27,
1178.
1-Allyl-3-(2-(4-methoxy)pyridyl)imidazol-2-ylidene copper(I) bromide
(2h): 1-Allyl-3-(2-(4-methoxy)imidazolium bromide (0.16 g, 0.54 mmol)
and Ag2O (0.081 g, 0.35 mmol) were added to a Schlenk flask along with
freshly-activated 4 ꢁ molecular sieves. Anhydrous dichloromethane
(25 mL) was added, and the mixture was stirred in the absence of light at
room temperature for 2.5 h. After this time, the mixture was filtered
through Celite and the solvent removed in vacuo to yield the crude prod-
uct as an off-white oily solid. Recrystallisation from acetone/pentane
gave Ag-2h as a white solid. Yield=0.17 g, 0.43 mmol, 80%. Ag-2h
(0.15 g, 0.37 mmol) and CuBr (0.059 g, 0.41 mmol) were added to a small
Schlenk flask in a glovebox. Anhydrous dichloromethane (25 mL) was
added and the suspension formed was stirred at room temperature in the
absence of light for 16 h. After this time, the suspension was filtered to
remove AgBr/CuBr and the yellow filtrate collected. The solvent was re-
moved in vacuo to give the product as a yellow crystalline solid. Yield:
[17] F. E. Hahn, C. Holtgrewe, T. Pape, Z. Naturforsch B: J. Chem. Sci
2004, 59, 1051.
[18] F. E. Hahn, B. Heidrich, T. Pape, A. Hepp, M. Martin, E. Sola, L. A.
Oro, Inorg. Chim. Acta 2006, 359, 4840.
[19] G. Occhipinti, V. R. Jensen, K. W. Tornroos, N. A. Froystein, H. R.
Bjorsvik, Tetrahedron 2009, 65, 7186.
[20] S. Dꢂez-Gonzꢃlez, E. D. Stevens, N. M. Scott, J. L. Petersen, S. P.
Nolan, Chem. Eur. J. 2008, 14, 158.
1
0.11 g, 0.31 mmol, 84%. H NMR (300 MHz, CDCl3): d=8.31 (br s, 1H),
7.82 (br s, 1H), 7.78 (br s, 1H), 7.07 (br s, 1H), 6.88 (br s, 1H), 6.14–5.94
(m, 1H, CH=CH2), 5.45–5.32 (m, 2H, CH=CH2), 4.86 (br s, 2H, NCH2),
4.02 ppm (s, 2H, OCH3); HRMS (ESI+): m/z calcd (%) for [MꢀBr+
MeCN]+: 319.0615; found: 319.0615.
2,6-Bis(3-allylimidazol-2-ylidene)pyridyl copper(I) bromide (2i): An am-
poule was charged with 1,1’-(2,6-pyridyl)-bis(3-allylimidazolium) bromide
(0.15 g, 0.33 mmol), CuBr (0.10 g, 0.66 mmol) and Cs2CO3 (1.1 g,
3.3 mmol). The contents of the ampoule were dried and degassed thor-
oughly in vacuo. Anhydrous acetonitrile (15 mL) was added and the sus-
pension was heated at 508C for 1.5 h. After this time, the mixture was al-
lowed to cool to ambient temperature, filtered and the solvent removed
in vacuo to give the crude product as a yellow solid. The product was re-
crystallised from acetonitrile/diethyl ether. Yield: 0.10 g, 0.10 mmol,
60%. Single crystals of 2i suitable for X-ray diffraction analysis were
grown by cooling a saturated solution of the product from 333 K to ambi-
ent temperature in anhydrous acetonitrile ([Cu3Br2L2]Br·MeCN). Single
crystals of 2i’ suitable for X-ray diffraction analysis were grown by the
slow evaporation of a concentrated solution of the product in wet aceto-
nitrile ([Cu2L2]2Br·2H2O). 1H NMR (500 MHz, CD3CN, 333.6 K): d=
8.28 (t, J=8.0 Hz, 2H, pyH), 7.81 (d, J=2.0 Hz, 4H, imH), 7.73 (d, J=
8.0 Hz, 4H, pyH), 7.23 (d, J=2.0 Hz, 4H, imH), 5.72–5.61 (m, 4H, CH=
CH2), 5.02 (d, J=10.3 Hz, 4H, HC=CHH(cis)), 4.93 (d, J=16.5 Hz, 4H,
HC=CHH(trans)), 4.40 ppm (d, J=4.5 Hz, 8H, NCH2); 13C{1H} NMR
(126 MHz, CD3CN, 333.6 K): d=181.9, 150.6, 145.1, 134.0, 124.5, 120.0,
119.9, 114.7, 55.4 ppm; HRMS (ESI+): m/z calcd for [Cu2L2]2+: 354.0774;
found: 354.0787; elemental analysis calcd (%) for C136H136Br11Cu11N40
(3:1 2i/2i’): C 41.79, H 3.51, N 14.33; found: C 41.60, H 3.45, N 13.85.
[21] G. Venkatachalam, M. Heekenroth, A. Neels, M. Albrecht, Helv.
Chim. Acta 2009, 92, 1034.
[22] S. Dꢂez-Gonzꢃlez, E. C. Escudero-Adan, J. Benet-Buchholz, E. D.
Stevens, A. M. Z. Slawin, S. P. Nolan, Dalton Trans. 2010, 39, 7595.
[23] X. L. Hu, I. Castro-Rodriguez, K. Meyer, J. Am. Chem. Soc. 2003,
125, 12237.
[24] A. A. D. Tulloch, A. A. Danopoulos, S. Kleinhenz, M. E. Light,
M. B. Hursthouse, G. Eastham, Organometallics 2001, 20, 2027.
[25] B. R. M. Lake, E. K. Bullough, T. J. Williams, A. C. Whitwood,
M. A. Little, C. E. Willans, Chem. Commun. 2012, 48, 4887.
[26] E. K. Bullough, M. A. Little, C. E. Willans, Organometallics 2013,
32, 570.
[27] C. Holtgrewe, C. Diedrich, T. Pape, S. Grimme, F. E. Hahn, Eur. J.
Org. Chem. 2006, 3116.
[28] P. L. Arnold, Heteroat. Chem. 2002, 13, 534.
[29] R. D. Kçhn, G. Seifert, Z. D. Pan, M. F. Mahon, G. Kociok-Kohn,
Angew. Chem. 2003, 115, 818; Angew. Chem. Int. Ed. 2003, 42, 793.
[30] A. Sundararaman, L. N. Zakharov, A. L. Rheingold, F. Jakle, Chem.
Commun. 2005, 1708.
[31] J. P. Zhang, Y. B. Wang, X. C. Huang, Y. Y. Lin, X. M. Chen, Chem.
Eur. J. 2005, 11, 552.
[32] G. M. Chiarella, D. Y. Melgarejo, A. Rozanski, P. Hempte, L. M.
Perez, C. Reber, J. P. Fackler, Chem. Commun. 2010, 46, 136.
[33] A. C. Badaj, S. Dastgir, A. J. Lough, G. G. Lavoie, Dalton Trans.
2010, 39, 3361.
[34] H. Y. Gong, B. M. Rambo, E. Karnas, V. M. Lynch, J. L. Sessler,
Nat. Chem. 2010, 2, 406.
Acknowledgements
[35] M. J. McPhillie, R. Trowbridge, K. R. Mariner, A. J. OꢄNeill, A. P.
Johnson, I. Chopra, C. W. G. Fishwick, ACS Med. Chem. Lett. 2011,
2, 729.
Financial support for this research has been provided by the Royal Soci-
ety of Chemistry, BP and University of Leeds.
[36] E. K. Bullough, C. A. Kilner, M. A. Little, C. E. Willans, Org.
Biomol. Chem. 2012, 10, 2824.
[37] C. E. Willans, K. M. Anderson, L. C. Potts, J. W. Steed, Org. Biomol.
Chem. 2009, 7, 2756.
[38] J. C. Garrison, W. J. Youngs, Chem. Rev. 2005, 105, 3978.
[39] C. Martꢂn, J. M. Munoz-Molina, A. Locati, E. Alvarez, F. Maseras,
T. R. Belderrain, P. J. Perez, Organometallics 2010, 29, 3481.
[40] A. A. D. Tulloch, A. A. Danopoulos, S. Winston, S. Kleinhenz, G.
Eastham, J. Chem. Soc. Dalton Trans. 2000, 4499.
[1] P. Brandt, M. J. Sodergren, P. G. Andersson, P. O. Norrby, J. Am.
Chem. Soc. 2000, 122, 8013.
[2] H. V. R. Dias, H. L. Lu, H. J. Kim, S. A. Polach, T. Goh, R. G.
Browning, C. J. Lovely, Organometallics 2002, 21, 1466.
[3] A. G. Jarvis, A. C. Whitwood, I. J. S. Fairlamb, Dalton Trans. 2011,
40, 3695.
[4] B. F. Straub, I. Gruber, F. Rominger, P. Hofmann, J. Organomet.
Chem. 2003, 684, 124.
Chem. Eur. J. 2013, 19, 16780 – 16790
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16789

C. E. Willans and B. R. M. Lake
[41] B. Liu, Q. Q. Xia, W. Z. Chen, Angew. Chem. 2009, 121, 5621;
Angew. Chem. Int. Ed. 2009, 48, 5513.
[44] C. Chen, H. Y. Qiu, W. Z. Chen, J. Organomet. Chem. 2012, 696,
4166.
[42] B. Liu, Y. Zhang, D. C. Xu, W. Z. Chen, Chem. Commun. 2011, 47,
2883.
[43] V. J. Catalano, L. B. Munro, C. E. Strasser, A. F. Samin, Inorg.
Chem. 2011, 50, 8465.
Received: May 16, 2013
Published online: November 6, 2013
16790
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 16780 – 16790