Complexes: Synthesis, characterization and catalytic activities in reduction reactions and Click Chemistry. on the advantage of using well-defined catalytic systems

Article abstract of DOI:10.1039/c0dt00218f

The preparation of three series of [(NHC)CuX] complexes (NHC = N-heterocyclic carbene, X = Cl, Br, or I) is reported. These syntheses are high yielding and only use readily available starting materials. The prepared complexes were spectroscopically and st

Full text of DOI:10.1039/c0dt00218f

View Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
[(NHC)CuX] complexes: Synthesis, characterization and catalytic activities in
reduction reactions and Click Chemistry. On the advantage of using
well-defined catalytic systems†
Silvia D´ıez-Gonza´lez,*‡a Eduardo C. Escudero-Ada´n,a Jordi Benet-Buchholz,a Edwin D. Stevens,b
Alexandra M. Z. Slawinc and Steven P. Nolan*a,c
Received 28th March 2010, Accepted 19th May 2010
First published as an Advance Article on the web 13th July 2010
DOI: 10.1039/c0dt00218f
The preparation of three series of [(NHC)CuX] complexes (NHC = N-heterocyclic carbene, X = Cl, Br,
or I) is reported. These syntheses are high yielding and only use readily available starting materials. The
prepared complexes were spectroscopically and structurally characterized. Notably, two of them
present a bridging NHC ligand between two copper centers in the solid state, an extremely rare
coordination mode for these ligands. These complexes were then applied to two distinct organic
reactions: the hydrosilylation of ketones and the 1,3-dipolar cycloaddition of azides and alkynes. In
both transformations, outstanding catalytic systems were found for preparing the corresponding
products in excellent yields and short reaction times. Most remarkably, the screening of well-defined
systems in the hydrosilylation reaction allowed for the identification of a pre-catalyst previously
overlooked since, originally, catalytic species were in situ generated. Under such conditions, major
formation of [(NHC)2Cu]+ species, inactive in this reduction reaction, occurred instead of the expected
copper hydride. These results highlight one of the most important advantages of employing
well-defined complexes in catalysis: gaining an improved control of the nature of the catalytically
relevant species in the reaction media.
have naturally led to their application in a plethora of organic
1. Introduction
transformations.2
Transition metal catalysis is one of the most powerful tools
The first [(NHC)Cu] complex was reported in 1993 by
Arduengo et al.,3 but the first application in catalysis only
available to chemists for the development of cleaner and more
sustainable processes. The use of different ligand families in order
appeared in 2001, with Woodward’s work on conjugate addition
reactions.4 The increasing number of [(NHC)Cu]-based catalytic
systems reported in the literature since, clearly demonstrates the
versatility of these species in organic chemistry.5 In particular,
we have studied their application in reduction and cycloaddition
reactions. Hence, we reported the ability of [(IPr)CuCl] to act
as pre-catalyst in the hydrosilylation of simple ketones,6 whereas
hindered or functionalized substrates reacted better in the
presence of [(ICy)CuCl], or [(SIMes)CuCl] in the particular case
of heteroaromatic ketones.7,8 On the other hand, we have shown
that [(NHC)CuI]-based systems are remarkable catalysts for the
[3+2] cycloaddition of alkynes and azides, allowing notably for
the use of an internal alkyne,9 extremely low catalyst loadings,10
and the development of a latent system.11,12
to protect the metal center and/or modulate its activity has been
essential for the advancement of this field. The catalytically active
species can either be generated in situ upon reaction of the proper
metal and ligand precursors or, alternatively, from a well-defined
complex. Whereas the in situ strategy is usually regarded as more
user friendly (no need to pre-isolate any metallic species), the latter
approach allows for a better control of the nature of the species
present in the reaction media and generally avoids the need for
excess of ligand and/or reagents for achieving optimal catalytic
performance. In this context, N-heterocyclic carbenes (NHCs)
have recently gained a prominent place in the chemist’s toolbox
due to their outstanding binding affinities to metal centers,1 which
Herein, we report the preparation of three complete se-
ries of NHC-containing copper complexes of general formulae
[(NHC)CuX], where X = Cl, Br, or I.13 The structures of the
NHC ligands employed in this work are shown in Fig. 1. The
preparation of these series has notably allowed us to uncover
rare coordination modes of the NHC ligands and to identify
significantly improved catalysts for the hydrosilylation of ketones
and the 1,3-dipolar cycloaddition of alkynes and azides. Besides
the remarkable catalytic systems identified in both cases, this work
clearly shows the neat advantages that using well-defined systems
in catalysis can represent.
aInstitute of Chemical Research of Catalonia (ICIQ), Av. Pa¨ısos Catalans,
16, 43007, Tarragona, Spain
bDepartment of Chemistry, University of New Orleans, 2000 Lakeshore
Drive, New Orleans, LA, 70148, USA
cSchool of Chemistry, University of St. Andrews, North Haugh, St. Andrews,
KY, 16 9ST, UK. E-mail: snolan@st-andrews.ac.uk
† Electronic supplementary information (ESI) available: Experimental
procedures, product characterization and crystallographic data. CCDC
reference numbers 750314–750323. For ESI and crystallographic data in
CIF or other electronic format see DOI: 10.1039/c0dt00218f
‡ Present address: Department of Chemisty, Imperial College London,
Exhibition Road, South Kensington, London SW7 2AZ, UK; E-mail:
s.diez-gonzalez@imperial.ac.uk
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 7595–7606 | 7595
©

View Online
ficient, due to the significant formation of unidentified products.
In this case, the best results were obtained using CuBr·SMe2 as
metal source, although heating the reaction mixtures was required
to ensure consumption of the starting material. Surprisingly, the
reactions with CuI proceeded smoothly and no additives were
required to ensure good synthetic outcomes.
The counterion of the imidazolium/imidazolinium salt had a
significant influence on the success of these syntheses. Whereas
IPr·HBF4 and SIPr·HBF4 led systematically to the major for-
mation of [(NHC)2Cu]+ species, IAd·HCl was not efficiently
deprotonated under these reaction conditions and around 25%
of the starting salt remained unreacted after overnight stirring.
These observations might be correlated with the more important
steric hindrance of IAd ligands when compared with those of IPr
and SIPr. It is important to note that a large excess of base should
be avoided in these reactions since it eases the undesired formation
of bis-NHC copper complexes.
In contrast to these results, [(IMes)CuCl] 4a,15 [(IMes)CuBr]
4b,16 [(SIMes)CuCl] 5a,7 and [(SIMes)CuBr] 5b9 were best pre-
pared from the corresponding NHC·HCl and copper salts in the
presence of NaOt-Bu in THF. Unfortunately, all attempts to pre-
pare the corresponding iodide-containing compounds only led to
the generation of complex mixtures in which [(NHC)2Cu]+ species
were the major products. Capitalizing on the previously observed
convenience of CuBr·SMe2 as copper source, we tested SMe2 and
SPh2 as additives17 in the preparation of these iodo-containing
complexes. Unfortunately, even if the formation of mono-NHC
species was greatly improved under these conditions, our best
results (mono : bis = 77 : 23) were still far from synthetically useful.
Fig. 1 NHC ligands employed.
2. Results and discussion
Synthesis of [(NHC)CuX] complexes
(A) In situ-generated NHCs: IPr, SIPr, IAd, IMes and SIMes
The first complex of this series, [(IPr)CuCl] 1a, was originally
prepared in 2003 from an imidazolium salt in the presence of
1 equivalent of a strong base.6,8a Following the same experimental
procedure, [(IPr)CuBr] 1b could be prepared in 80% yield (eqn (1)).
(1)
However, when these reaction conditions were used with CuI
as copper source, a significant amount of [(IPr)2Cu]+ species was
formed, even when using a large excess of metal salt.14 In this case,
we found that the combination of KOMe and toluene was more
efficient for the preparation of [(IPr)CuI] 1c. We first reported
these conditions for the preparation of [(SIPr)CuCl] 2a,11 and
eventually found that they were optimal for the preparation of
different [(NHC)CuX] complexes, as shown in Table 1. For all
these compounds, the combination KOMe/toluene was found
higher yieldingthantheNaOt-Bu/THF oneand also led tocleaner
reactions.
For complexes 1c and 2b–c, the generation of [(NHC)2Cu]+
species could be minimized under these reaction conditions, and
for complexes 3a–c, the formation of complex mixtures of prod-
ucts, observed with NaOt-Bu/THF, was avoided. Nevertheless,
reactions with copper bromide were found to be particularly inef-
(B) Previously isolated carbenes: ICy and ItBu
Copper complexes bearing ICy and ItBu ligands were best pre-
pared from the corresponding free carbene, although for different
reasons. ICy-containing complexes were found particularly sensi-
tive to the excess of base, probably due to the readily formation of
the corresponding copper alkoxides.18 Therefore, the use of a free
carbene ensured the straightforward isolation of complexes 6a–c in
high yields (Table 2).19 In the preparation of 7a from ItBu·HBF4,
large quantities of the starting imidazolium salt were recovered
when the reactions were conducted at room temperature (either
with NaOt-Bu/THF or KOMe/toluene). However, raising the
reaction temperature led to the formation of [(ItBu)2Cu]+ species
as the major products (73% of the crude material according to the
1H NMR) in addition to 7a (15%) and the starting ItBu salt (12%).
Satisfyingly, reactions with free ItBu cleanly produced complexes
7a–c in good yields. These ItBu-bearing complexes are the most
sensitive towards oxygen and moisture within these series. Hence,
these should be kept under an inert atmosphere to avoid partial
decomposition after only a few days.
Table 1 Synthesis of [(NHC)CuX] complexes
NHC·HY
CuX
[(NHC)CuX]
Yield
Table 2 Preparation of ICy and ItBu-containing complexes
IPr·HCl
CuI
[(IPr)CuI]
1c
2b
2c
3a
3b
3c
81%
70%a
77%
81%
81%a
75%
SIPr·HCl
SIPr·HCl
IAd·HBF4
IAd·HBF4
IAd·HBF4
CuBr·SMe2
[(SIPr)CuBr]
[(SIPr)CuI]
[(IAd)CuCl]
[(IAd)CuBr]
[(IAd)CuI]
CuI
NHC = ICy
Yield
NHC = ItBu
Yield
CuCl
CuBr·SMe2
CuI
[(ICy)CuCl]
[(ICy)CuBr]
[(ICy)CuI]
6a
6b
6c
71%
82%
84%
[(ItBu)CuCl]
[(ItBu)CuBr]
[(ItBu)CuI]
7a
7b
7c
73%
85%
80%
a Reaction carried out in refluxing toluene.
7596 | Dalton Trans., 2010, 39, 7595–7606
This journal is
The Royal Society of Chemistry 2010
©

View Online
(C) Transmetalation from [(NHC)AgCl] complexes
characterized complexes, the aryl groups are almost perpendicular
with respect to the imidazole backbone plane, resulting in a
favorable steric arrangement with the metal center. Surprisingly,
the aryl rings in [(SIMes)CuBr] 5b presented different torsion
angles with respect to the NHC plane (Fig. 3).
Finally, we also explored the preparation of these NHC-containing
copper(I) complexes by carbene transfer from the corresponding
NHC–silver(I) reagents, a frequent approach for the preparation
of NHC complexes of late transition metals.20 Some complexes,
such as SIPr-bearing 2a–c were accordingly isolated upon reaction
of [(SIPr)AgCl] with the corresponding copper salt at room
temperature (eqn (2)). However, mixtures of mono- and bis-
NHC complexes were systematically obtained from [(IMes)AgCl],
[(SIMes)AgCl] and [(ICy)AgCl].
All these complexes, along with [(IAd)CuCl] 3a and
[(IAd)CuBr] 3b, have a two-coordinated copper(I) center in
a near linear environment with a C–Cu–X angle very close
or equal to 180◦. Both IAd-bearing complexes belong to the
orthorhombic crystalline system and 3a notably showed two
independent molecules in the unit cell, whereas the latter refined
as a racemic crystal twinning in a ratio 78 : 22. For all the
monomeric complexes, the C–Cu bond distances were in the
24
˚
(2)
range of 1.87–1.96 A, within the standard range of Cu–C single
bonds, and comparable to the precedent Cu–C bonds in NHC
complexes.6,7,11,16,25 No trends were observed on this Cu–C bond
distance as a function of the NHC ligands. Similarly, the copper–
halogen bond distances were not significantly different across these
series.
Silver transmetalation is a well-established method for the
preparation of [(NHC)M] complexes. Still, evidence of halogen
exchange during this reaction was recently reported in the prepa-
ration of a [(NHC)Cu] derivative.21 It is important to note that
batches of complexes 2a–c prepared either from the corresponding
azolium salt or by transmetalation, showed identical catalytic
activities, which indicates that no halogen exchange occurred in
these syntheses.
In the view of these results, it is evident that the synthesis
of [(NHC)CuX] complexes is strongly dependent on both the
chosen NHC ligand and the copper source. Our comprehensive
screening has allowed us to optimize different approaches to these
compounds,22 which should make them accessible from virtually
any combination of NHC and CuX species. In this context,
avoiding the formation of [(NHC)2Cu]+ species is crucial. Even
if both families of complexes are easily differentiated by 1H NMR
spectroscopy,23 their separation through selective crystallization
or chromatography is extremely tedious, and in most cases,
ineffective.
Even if most complexes in these series are monomeric in the solid
state, three exceptions were found. [(IAd)CuI] 3c, crystallized with
a half molecule in the unit cell in an apparent dimeric manner, with
two adjacent symmetrical NHC–Cu–I units associated to each
other (Fig. 4).26 This complex presents a Ci symmetry, the same
one as its crystal structure. The copper atoms present a trigonal
planar geometry and are held together by two m-I atoms forming
a planar [Cu2I2] core. The two NHC rings are significantly twisted
relative to the [Cu2I2] plane (angle close to 90◦), but they are both
in the same plane.27 The Cu–Cu distance in [(IAd)CuI] 3c, 2.95 A,
˚
suggest a weak metal–metal interaction, as previously proposed.28
Furthermore, [(ICy)CuBr] 6b and [(ICy)CuI] 6c presented very
interesting structures in which the NHC ligands either bind a
single copper center, or act as bridging ligands between two
metal atoms (Fig. 4). This bridging coordination mode remains
extremely rare with N-heterocyclic carbenes,29 and, to the best of
our knowledge, only two examples have previously been reported
in the literature for a copper complex.30 In these species 6b and 6c,
the plane of the imidazole ring is twisted approximately by 65◦ with
respect to the Cu∧C∧Cu plane, and the carbenic carbons bound
to two copper centers present a distorted tetrahedral geometry.
Two main possibilities have been postulated to rationalize this
bonding mode:30 formation of a s-bond with one copper atom
and a not so documented p-interaction with the second metal
center,29 or alternatively, a three-center-two-electron bond seems
also plausible.31 Moreover, in both complexes, the [Cu3X3] cores
are appreciably distorted, as shown by the different copper–
Crystallographic studies
Crystallographic analyses were carried out to unambiguously
confirm the structures of the prepared compounds. Suitable
crystals for X-ray diffraction were grown for complexes 1b–c,
2c, 3b–c, 4a, 5b and 6b–c. Structures for 1a,6 2a,11 4b,16 5a7
and 6a7 have already been reported in the literature, and all
attempts to crystallize the other complexes were unsuccessful,
generally leading to decomposition products and the formation
of the corresponding azolium salt. Ball-and-stick representations
are given in Fig. 2 (1b, 1c, 2c and 4a), Fig. 3 (3a, 3b, 5b) and
Fig. 4 (3c, 6b and 6c). A comparison of selected bond distances
and angles for representative complexes is provided in Table 3 and
Table 4. More relevant crystallographic data for all complexes can
be found in the ESI.†
[(IPr)CuBr] 1b and [(IPr)CuI] 1c are isomorphous with the
known [(IPr)CuCl] 1a (orthorhombic in the Pccn space group),
but this is not the case for the SIPr-containing complexes –
monoclinic in P21/m for [(SIPr)CuCl] 2a and orthorhombic in
Pccn for [(SIPr)CuI] 2c, but in the latter case, solvents (acetonitrile
and octane) cocrystallized with 2c. These complexes, along with
[(IMes)CuCl] 4a and [(SIMes)CuBr] 5b, contain NHC ligands
with N-aryl substituents and, as in other crystallographically
˚
˚
halogen bonds: 2.2–3.8 A for [(ICy)CuBr] 6b and 2.6–3.8 A for
[(ICy)CuI] 6c. On the other hand, the shortest Cu–Cu distance is
˚
2.506 A in 6b, and 2.453 in 6c. Since both values are significantly
shorter than the sum of the van der Waals radii (2.80 A),
˚
they strongly support an important metal–metal interaction in
complexes 6b and 6c. It is important to note that in contrast with
the heavier group 11 elements, bonding interactions between two
copper atoms remain scarce in the literature.32
Finally, it is important to note that both ICy-containing com-
plexes with a bridging NHC ligand showed different solubilities
in organic solvents when compared to the other compounds in
these series. In particular, their solubility in chlorinated solvents
was appreciably lower, whereas 6b and 6c were readily soluble
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 7595–7606 | 7597
©

View Online
Fig. 2 Ball-and-stick representations of [(IPr)CuBr] 1b, [(IPr)CuI] 1c, [(SIPr)CuI] 2c, and [(IMes)CuCl] 4a. Most hydrogen atoms have been omitted for
clarity.
◦
˚
Table 3 Selected bond lengths (A) and angles ( ) for complexes 1b, 2c, 3a, 4a, and 5b
[(IPr)CuBr] 1b
[(SIPr)CuI] 2c
[(IAd)CuCl] 3a
[(IMes)CuCl]4a
[(SIMes)CuBr] 5b
Cu(1)–C(1)
N(1)–C(1)
N(2)–C(1)
C(2)–C(3)
Cu(1)–X
N(1)–C(1)–N(2)
C(1)–Cu(1)–X
1.884(2)
1.352(2)
1.352(2)
1.352(2)
2.2090(4)
104.5(2)
180.0
1.8874(17)
1.3322(14)
1.3322(14)
1.533(3)
2.3804(2)
108.58(14)
180.0
1.894(2)
1.361(3)
1.365(2)
1.355(3)
2.1114(6)
104.56(17)
177.39(6)
1.956(10)
1.310(8)
1.310(8)
1.399(12)
2.091(2)
109.3(8)
180.000(1)
1.884(7)
1.323(10)
1.345(9)
1.510(11)
2.1903(14)
108.7(6)
177.7(2)
◦
˚
Table 4 Selected bond lengths (A) and angles ( ) for complexes 3c, and 6c
[(IAd)CuI] 3c
[(ICy)CuI] 6c
[(IAd)CuI]3c
[(ICy)CuI]6c
Cu(1)–C(1)
N(1)–C(1)
N(2)–C(1)
Cu(1)–X(1)
Cu(2)–C(2)
Cu(1)–C(2)
N(3)–C(2)
N(4)–C(2)
Cu(2)–X(2)
Cu(3)–C(1)
N(5)–C(3)
1.9463(12)
1.3653(14)
1.3625(15)
2.62654(19)
1.927(8)
Cu(3)–X(3)
N(6)–C(3)
Cu(1)–Cu(2)
Cu(2)–Cu(3)
N(1)–C(1)–N(2)
N(3)–C(2)–N(4)
N(5)–C(3)–N(6)
Cu(1)–C(2)–Cu(2)
C(1)–Cu(1)–X(1)
C(2)–Cu(2)–X(2)
C(3)–Cu(3)–X(3)
2.8509(13)
1.355(10)
2.4528(15)
2.5332(14)
104.2(7)
103.6(6)
107.1(7)
72.5(3)
1.349(10)
1.359(10)
2.5487(11)
1.973(8)
2.9510(3)
104.60(10)
2.165(8)
1.373(10)
1.369(10)
2.6206(12)
1.928(8)
131.91(3)
131.91(3)
128.2(2)
128.9(2)
110.5(3)
1.355(10)
7598 | Dalton Trans., 2010, 39, 7595–7606
This journal is
The Royal Society of Chemistry 2010
©

View Online
Table 5 Catalysts screening for the hydrosilylation of propiophenone
[(NHC)CuX]
Time/h
Conv/%a
[(NHC)CuX]
Time/h
Conv/%a
[(IPr)CuCl] 1a
[(IPr)CuBr] 1b
[(IPr)CuI] 1c
[(SIPr)CuCl] 2a
[(SIPr)CuBr] 2b
[(SIPr)CuI] 2c
[(IAd)CuCl] 3a
[(IAd)CuBr] 3b
[(IAd)CuI] 3c
2
3
4
5
24
24
4
4
24
>99
>99
>99
>99
50
[(IMes)CuCl] 4a
[(IMes)CuBr] 4b
[(SIMes)CuCl] 5a
[(SIMes)CuBr] 5b
[(ICy)CuCl] 6a
[(ICy)CuBr] 6b
[(ICy)CuI] 6c
[(ItBu)CuCl] 7a
[(ItBu)CuBr] 7b
[(ItBu)CuI] 7c
1
2
0.5
0.25
6
4
24
8
6
24
>99
>99
>99
>99
>99
>99
0
>99
>99
27
59
>99
>99
95
a GC conversions are the average of at least two independent runs.
Fig. 3 Ball-and-stick representations of [(IAd)CuCl] 3a, [(IAd)CuBr] 3b, and [(SIMes)CuBr] 5b. Most hydrogen atoms have been omitted for clarity.
in toluene and benzene. However, only one set of signals was
observed in the NMR spectra of these complexes, indicating that
the observed solid structures were not kept in solution.
unsaturated bonds generally leads to the formation of the reduced
product under mild and/or selective conditions.33 In the case of
carbonyl compounds,34 such as ketones, this approach has the
added advantage of providing the expected product in the form of
a protected alcohol, which can be of particular interest in synthesis.
We recently reported that [(IPr)CuCl] 1a is an efficient pre-
catalyst for the hydrosilylation of simple ketones under mild
conditions (3 mol% 1a, 12 mol% NaOt-Bu, 3 equiv Et3SiH,
Catalytic studies: Hydrosilylation of ketones
Oxidation and reduction reactions are among the most commonly
used transformations in organic chemistry. The hydrosilylation of
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 7595–7606 | 7599
©

View Online
Fig. 4 Ball-and-stick representations of [(IAd)CuI] 3c, [(ICy)CuBr] 6b, and[(ICy)CuI] 6c. Most hydrogen atoms have been omitted for clarity.
toluene, RT).6 More challenging substrates, either because of
their steric hindrance or the presence of functional groups were
found to best ◦react in the presence of [(ICy)CuCl] 6a, although
heating at 80 C was required.7 Despite the broad scope of 6a,
we found that in the particular case of heteroaromatic ketones,
[(SIMes)CuCl] 5a was the catalyst of choice, albeit under slightly
different reaction conditions (3 mol% 5a, 3 mol% NaOt-Bu, 2
equiv Et3SiH, 80 ◦C). Capitalizing on these studies on [(NHC)Cu]
systems in the reduction of ketones, we screened complexes 1–7 in
the hydrosilylation of propiophenone as a model substrate. In a
first step, we used the reaction conditions previously found optimal
with [(IPr)CuCl] 1a for this reaction.6 It is important to note here
that under such conditions no reaction was observed when CuCl,
CuBr or CuI salts were employed as pre-catalysts. As shown in
Table 5, most of these complexes led to the complete conversion of
the starting ketone in less than 8 h at room temperature. Exceptions
are complexes 2b–c and 7c, which only afforded low conversions
even after extended reaction times and complex 6c, inactive in
this transformation. These results evidenced also the important
influence of the halogen present in the pre-catalyst. As a general
trend, chloride-containing complexes were slightly more active
than their brominated analogues, while iodide-bearing complexes
were significantly the least active.
Moreover, we were certainly surprised by the high catalytic
activity of complexes 5a and 5b, especially considering the low con-
versions obtained in the original screening with the corresponding
azolium salts.6 As shown in Table 6, whereas a catalyst generated in
situ from SIMes·HBF4 led to a maximum conversion of only 35%
of the starting cyclohexanone, the corresponding silyl ether was
quantitatively formed in only one hour with the corresponding
well-defined complex 5a.
In situ systems are generally very convenient from a user point
of view. However, their main drawback is the lack of control
on the species actually generated under catalytic conditions.35
During our synthetic studies we observed that reactions with
SIMes·HBF4 for preparing 5a36 systematically led to the major
formation of [(SIMes)2Cu]+ derivatives, which are actually inactive
in the hydrosilylation of ketones (Scheme 1).23 This phenomenon
was also observed with IPr, for instance, although to a lesser
extent. It is important to note that in the case of IPr, no difference
7600 | Dalton Trans., 2010, 39, 7595–7606
This journal is
The Royal Society of Chemistry 2010
©

View Online
Table 6 Comparison of in situ and well-defined systemsa
Table 7 [(SIMes)CuCl]-catalyzed hydrosilylation of ketones
Entry
1
Product
T/◦C
Time/h
0.5
Yield/%a
93
In situb
Well-definedc
RT
NHC
SIMes
IPr
Time/h
Conv/%d
35
99
38
Time/h
Conv/%d
>99
>99
8
2
8
1
2
8
ItBu
>99
2
3
4
5
RT
RT
RT
0.5
0.25
2
95b
96
90
a Data for in situ systems taken from ref. 6. b NaOt-Bu (20 mol%), Et3SiH
(5 equiv). c NaOt-Bu (12 mol%), Et3SiH (3 equiv). d GC conversions are
the average of at least two independent runs.
of activity was observed between the in situ6 and the well-
defined system (Table 6), probably due to the fact that [(IPr)2Cu]+
species are actually efficient catalysts for this transformation!23
Moreover, the lack of activity when using ItBu·HBF4 can now
be rationalized as the consequence of an inefficient formation
of [(ItBu)CuCl], precursor of the active species, under catalytic
conditions (Scheme 1).
80
RT
0.5
2
98
98
6
7
80
80
0.66
1
92
91
Scheme 1 Reactions with NHC·HBF4 salts.
8
9
80
20
55c
Hence, the screening of well-defined complexes instead of in situ
generated systems, allowed for the identification of SIMes as the
most efficient ligand for this transformation. With propiophenone,
little difference was observed between 5a and 5b, and the bromido-
containing complex was even found superior to the chlorido one.
Nevertheless, this was not a general trend and [(SIMes)CuCl] 5a
was indeed selected as the most performing pre-catalyst for this
reaction. Further optimization of the reaction conditions with 5a
led to lower catalyst and base loadings (2 mol% respectively) along
with the use of only a slight excess of hydride source (1.2 equiv).
This is particularly important from a waste removal angle and
from an economical point of view since hydrosilanes can be pricy,
especially the ones of interest as protecting groups.
We next studied the scope of this catalytic system with different
ketones. It is important to note that under the conditions shown in
Table 7, not only did the reactions proceed smoothly, but also, they
were actually faster than with larger quantities of reagents and/or
catalyst. Thus, propiophenone was completely converted into the
corresponding silyl ether in half the time (15 min vs. 30 min)
when compared to the conditions in Table 5. Other unencumbered
substrates such as cyclohexanone or naphthalone (Table 7, entries
1 and 4) also yielded the expected products in high yields and short
reaction times. The presence of an extra methyl group next to the
carbonyl function did not alter the reaction outcome (Table 7,
entry 2) and more hindered benzophenone and cyclopropyl ketone
80
RT
0.5
23
94d
90d
a Isolated yields are average of at least two independent runs. b trans : cis =
78 : 22, determined by 1H NMR. c 1H NMR conversion. d [(SIMes)CuCl]
5a (3 mol%), NaOt-Bu (3 mol%), Et3SiH (2 equiv).
(Table 7, entries 5–6) also yielded the corresponding silyl ethers in
high yields.
For the reduction of benzophenone in particular, heating the
reaction mixture was no longer required and the reaction efficiently
proceeded at room temperature. We also tested the influence of
other substitutions to find that, whereas halogen atoms were well
tolerated (Table 7, entry 7), the presence of a CF3- group led to
disappointing conversions (Table 7, entry 8). Therefore, despite
the remarkable activity of [(SIMes)CuCl] in the hydrosilylation of
ketones, [(ICy)CuCl] might still be the catalyst of choice for par-
ticularly challenging substrates.7 Moreover, since heteroaromatic
ketones had already been identified as very peculiar substrates, we
repeated the catalyst screening with 2-acetylpyridine.37 This time
[(SIMes)CuCl] was still found to be the best catalyst as previously
reported.7 It is important to note that slightly different reaction
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 7595–7606 | 7601
©

View Online
conditions were required in this case to ensure a good reaction
outcome although heating at 80 ◦C was not compulsory if longer
reaction times were allowed (Table 8, entry 9).
In order to identify the best performing catalyst for this
transformation, we next compared the three most promising
candidates (3b, 3c and 6a) using only a 0.25 mol% catalyst loading.
Under these conditions, 4 h were required for the reaction to be
completed in the presence of [(ICy)CuCl] 6a, 1.5 h for [(IAd)CuBr]
3b and only 1 h with [(IAd)CuI] 3c. Furthermore, this reaction
still proceeded smoothly in three hours with only 0.05 mol% of 3c!
Hence, [(IAd)CuI] was selected as the best catalyst and applied to
the cycloaddition of different azides and alkynes.
Due to the high exothermicity of this cycloaddition reaction,
particular precautions should be taken when carrying them
out under neat conditions. While we did not experience any
problem, all reactions were performed on a small scale of 1 mmol.
Gratifyingly, we found that water—with no organic co-solvent—
was the best solvent when using [(IAd)CuI] as catalyst, and similar
results can be obtained using the same catalyst loading after
slightly longer reaction times.
As shown in Table 9, a number of triazoles could be prepared
in extremely high yields and short reaction times. These reac-
tions were run in the presence of 0.8 mol% of 3c in order to
facilitate a direct comparison with our former best Click catalyst,
[(SIMes)CuBr] 5b.9 Indeed, 3c is substantially superior to 5b in this
transformation and in all entries, the IAd-based catalyst required
a shorter reaction time to produce the corresponding triazoles in
similar or better yields. One of the drawbacks of [(SIMes)CuBr]
as catalyst is that oily or low melting-point triazoles required
higher reaction temperatures or catalyst loadings.9 In contrast,
such products were efficiently formed with 3c at room temperature
with only 0.8 mol% of copper complex (Table 9, entries 2, 4, 6 and
7). Furthermore, this catalyst was also better performing with
sterically hindered substrates (Table 9, entries 3, 4, 5 and 7) and its
higher functionality tolerance allowed the preparation of triazoles
previously inaccessible with [(SIMes)CuBr] (Table 9, entry 8). It
is important to note here, that even if no direct correlation can be
made at this point, [(IAd)CuI] 3c presented a dimeric structure in
the solid state, in contrast with most species in these series.
Catalytic studies: [3+2] cycloaddition reaction of azides and
alkynes
1,3-Dipolar cycloadditions, also known as Huisgen cycloaddi-
tions, are a convenient approach for the construction of a wide
variety of five-membered heterocycles.38 Among these reactions,
the [3+2] cycloaddition of organic azides and terminal alkynes has
recently attracted enormous interest due to its status of the best
‘Click’ reaction to date.39 Hence, the copper(I)-catalyzed version
of this transformation led to the extremely efficient formation
of the corresponding 1,4-disubstituted [1,2,3]-triazoles as a sole
regioisomer.40 Even if the copper(I) species are most usually
generated in situ from CuII/reductant or CuII/Cu0 combinations,
different families of ancillary ligands have already been shown
to be particularly appealing for the stabilization of CuI species
during the cycloaddition reaction.41 Capitalizing on our previous
work on the applicability of NHC ligands in this context9,10,11,12
we decided to screen the whole series of [(NHC)CuX] complexes
in the preparation of [1,2,3]-triazoles. The obtained results for the
cycloaddition of benzyl azide with phenylacetylene are shown in
Table 8. With the exception of IPr- and SIPr-based complexes
1 and 2, the model reaction was complete in 2 h or less in the
presence of only 1 mol% of copper complex under neat conditions.
This reaction followed an opposite trend to the one observed for
the hydrosilylation reactions, and in general iodine > bromine >
chlorine. This observation might be rationalized as a result of
an easier displacement of an iodo ligand from the copper center
to initiate the catalytic cycle with the generation of a copper
acetylide.42 The complexes containing ICy (6b–c), and ItBu (7c)
are the exception to this trend, which might be explained by their
higher instability under the reaction conditions. In the presence of
these complexes the reaction mixture actually turned green, which
indicates decomposition of the initial catalyst. These complexes
could still be highly active provided the presence of air and/or
water is avoided. However such precautions would go against the
simple reaction conditions imposed by Click laws39 and therefore
no further attempts were made in this direction.
Conclusion
We have prepared and fully characterized three series of
[(NHC)CuX] complexes with different NHC ligands and halogen
atoms. Two of these complexes were notably found to exhibit
bridging NHC ligands in the solid state, a scarcely encountered
coordination mode for this family of ligands. The straightforward
synthesis of these complexes, and the fact that most of them are
indefinitely bench-stable, make [(NHC)CuX] compounds excellent
candidates for virtually any copper-mediated process. Herein,
these complexes have proved active in two distinct transforma-
tions of prominent interest in organic chemistry: reduction and
cycloaddition reactions.
In particular, the screening of well-defined complexes has
allowed for the identification of [(SIMes)CuCl] 5a as a highly
efficient catalyst for the hydrosilylation of different families
of ketones. The potential of this catalyst had previously been
overlooked due to the use of in situ generated catalytic sys-
tems. Actually, under such conditions significant formation of
[(NHC)2Cu]+ derivatives was evidenced. This side reaction is
highly undesirable since [(SIMes)2Cu]+ species have been shown
catalytically inactive for this reaction. These observations remind
Table 8 Catalyst screening for the preparation of [1,2,3]-triazoles
[(NHC)CuX]
Time Conv/%a [(NHC)CuX]
Time Conv/%a
[(IPr)CuCl] 1a 24 h
[(IPr)CuBr] 1b 24 h
88
63
95
99
96
>99
[(IMes)CuCl] 4a 1 h
[(IMes)CuBr] 4b 0.5 h
[(SIMes)CuCl] 5a 20 min >99
[(SIMes)CuBr] 5b 20 min >99
>99
>99
[(IPr)CuI] 1c
4 h
[(SIPr)CuCl] 2a 15 h
[(SIPr)CuBr] 2b 10 h
[(SIPr)CuI] 2c 6 h
[(ICy)CuCl] 6a
[(ICy)CuBr] 6b
[(ICy)CuI] 6c
10 min >99
1 h
2 h
>99
87
>99
[(IAd)CuCl] 3a 20 min >99
[(IAd)CuBr] 3b 10 min >99
[(ItBu)CuCl] 7a 1 h
[(IAd)CuI] 3c
10 min >99
[(ItBu)CuBr] 7b 30 min >99
[(ItBu)CuI] 7c 2 h >99
a GC conversions are the average of at least two independent runs.
7602 | Dalton Trans., 2010, 39, 7595–7606
This journal is
The Royal Society of Chemistry 2010
©

View Online
Table 9 [3+2] Cycloaddition reaction of azides and alkynes
a glovebox containing less than 1 ppm oxygen. Imidazolium
and imidazolinium salts and free carbenes were synthesized
according to literature procedures.43 [(NHC)AgCl] complexes were
prepared following the reported conditions.44 1H NMR and 13C
NMR spectra were recorded on a 400 MHz spectrometer at
room temperature. Chemical shifts (d) are reported with respect
to tetramethylsilane as internal standard in ppm. Elemental
analyses were performed by Robertson Microlit Laboratories, Inc.,
Madison, NJ (USA) and at the Centro de Microana´lisis Elemental
of the Universidad Complutense de Madrid (Spain). All reported
yields are isolated yields and in the catalytic studies are the average
of at least two runs.
[(IAd)CuI] 3c
[(SIMes)CuBr] 5b
Yield/%a Time Yield/%a
20 min 98b
Entry Product
1
Time
10 min 99
2
3
1.5 h
2 h
95
92
20 h
3.5 h
87c
94d
Synthesis of [(NHC)CuX] complexes
18 h
76
[1,3-Bis(2,6-diisopropylphenyl)imidazol-2-ylidene]copper(I) bro-
mide [(IPr)CuBr] (1b). In a vial fitted with a screw cap, copper
bromide (0.250 g, 1.74 mmol), IPr·HCl (0.740 g, 1.74 mmol) and
NaOt-Bu (0.167 g, 1.74 mmol), were loaded inside a glovebox and
stirred in dry THF (10 mL) outside the glovebox for 15 h. After
filtering the reaction mixture through a plug of Celite (DCM), the
filtrate was mixed with pentane to form a precipitate. A second
filtration led to the isolation of the title compound as a white solid
(0.738 g, 80%). Suitable crystals for X-ray diffraction were grown
by slow diffusion of hexane from a saturated solution of 1b in
4
5
11 h
95
18 h
20c
45 min 98b,e
20 min 92
45 min 98
18 h
4 h
80c
94b,d
1
DCM. CCDC-750314. H NMR (400 MHz, CDCl3): d = 7.50
(t, J = 7.8 Hz, 2H, HAr), 7.30 (d, J = 7.8 Hz, 4H, HAr), 7.14 (s,
2H, NCH=), 2.57 (septet, J = 6.8 Hz, 4H, ArCHCH3), 1.31 (d,
J = 6.8 Hz, 12H, ArCHCH3), 1.24 ppm (d, J = 6.8 Hz, 12H,
ArCHCH3); 13C NMR (100 MHz, CDCl3): d = 181.4 (C, NCN),
145.5 (CH, CAr), 134.3 (C, CAr), 130.5 (C, CAr), 124.1 (CH, CAr),
123.1 (CH, NCH=), 28.6 (CH, ArCHCH3), 24.8 (CH3), 23.8 ppm
(CH3); elemental analysis calcd (%) for C27H36BrCuN2 (532.04):
C, 60.95; H, 6.82; N, 5.27; found: C, 61.20; H, 6.82; N, 4.99.
6
7
9 h
4 h
92
96d
1 h
4 h
96
93
18 h
5 h
80c
95b,d
8
4 h
0f
[1,3-Bis(2,6-diisopropylphenyl)imidazol-2-ylidene]copper(I)
iodide [(IPr)CuI] (1c). In a vial fitted with a screw cap, copper
iodide (0.314 g, 1.1 equiv), IPr·HCl (0.637 g, 1.5 mmol) and KOMe
(0.126 g, 1.2 equiv), were loaded inside a glovebox and stirred in
dry toluene (5 mL) outside the glovebox for 15 h. After filtering the
reaction mixture through a plug of Celite (DCM), the filtrate was
mixed with pentane to form a precipitate. A second filtration led to
the isolation of the title compound as a white solid (0.701 g, 81%).
Suitable crystals for X-ray diffraction were grown by slow diffusion
of hexane from a saturated solution of 1c in DCM. CCDC-750315.
a Isolated yields are the average of two independent runs. b Data from ref.
9 c Conversion determined by 1H NMR. d 2 mol% of 5b. e Reaction run at
45 ◦C. f Decomposition products.
us that the nature of in situ-generated species needs to be carefully
verified before drawing any conclusions on the efficiency of a given
system.
On the other hand, the wider availability of [(NHC)CuX] com-
plexes allowed us to discover a better performing catalyst for the
preparation of [1,2,3]-triazoles under Click conditions. [(IAd)CuI]
was systematically superior to our former best catalyst within
this family of complexes, [(SIMes)CuBr], and even previously
unsuccessful reactions took place at room temperature in the
presence of a low copper loading.
1H NMR (400 MHz, CDCl3): d = 7.50 (t, J = 7.8 Hz, 2H, HAr),
Ar
=
7.31 (d, J = 7.8 Hz, 4H, H ), 7.15 (s, 2H, NCH ), 2.58 (septet,
J = 6.8 Hz, 4H, ArCHCH3), 1.31 (d, J = 6.8 Hz, 12H, ArCHCH3),
1.24 ppm (d, J = 6.8 Hz, 12H, ArCHCH3); 13C NMR (100 MHz,
CDCl3): d = 183.1 (C, NCN), 145.6 (CH, CAr), 134.2 (C, CAr),
Ar
Ar
=
130.6 (C, C ), 124.2 (CH, C ), 123.0 (CH, NCH ), 28.7 (CH,
ArCHCH3), 24.9 (CH3), 23.9 ppm (CH3); elemental analysis calcd
(%) for C27H36CuIN2 (579.04): C, 56.00; H, 6.27; N, 4.84; found:
C, 56.38; H, 6.19; N, 4.79.
Experimental section
[1,3 - Bis(2,6 - diisopropylphenyl)imidazolin - 2 - ylidene]copper(I)
chloride [(SIPr)CuCl] (2a). In a vial fitted with a screw cap,
copper chloride (23 mg, 5 equiv) and [(SIPr)AgCl] (25 mg,
0.046 mmol), were loaded inside a glovebox and stirred in
dry acetonitrile (2.5 mL) outside the glovebox for 15 h. After
General considerations
All reagents were used as received. Solid reagents for the prepa-
ration of [(NHC)CuX] complexes were stored under argon in
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 7595–7606 | 7603
©

View Online
filtering the reaction mixture through a plug of Celite (acetone),
solvents were evaporated under reduced pressure to isolate the title
compound as a white solid (19 mg, 83%). Spectroscopic data for
2a were in good accordance with literature data.11
dry toluene (5 mL) outside the glovebox for 15 h. After filtering
the reaction mixture through a plug of Celite (DCM), the filtrate
was mixed with pentane to form a precipitate. A second filtration
led to the isolation of the title compound as a white solid (0.464 g,
81%). Suitable crystals for X-ray diffraction were grown from a
saturated solution of 3a in acetone. CCDC-750317. 1H NMR
[1,3 - Bis(2,6 - diisopropylphenyl)imidazolin - 2 - ylidene]copper(I)
bromide [(SIPr)CuBr] (2b). (A) In a vial fitted with a screw cap,
copper bromide dimethyl sulfide complex (0.339 g, 1.1 equiv),
SIPr·HCl (0.640 g, 1.5 mmol) and KOMe (0.132 g, 1.25 equiv),
were loaded inside a glovebox and stirred in refluxing dry toluene
(5 mL) outside the glovebox for 15 h. After filtering the reaction
mixture at room temperature through a plug of Celite (DCM), the
filtrate was mixed with pentane to form a precipitate. A second
filtration led to the isolation of the title compound as a white solid
(0.552 g, 69%). (B) In a vial fitted with a screw cap, copper bromide
(68 mg, 5 equiv) and [(SIPr)AgCl] (50 mg, 0.09 mmol), were loaded
inside a glovebox and stirred in dry acetonitrile (5 mL) outside the
glovebox for 15 h. After filtering the reaction mixture through a
plug of Celite (DCM), solvents were evaporated under reduced
pressure to isolate the title compound as a white solid (42 mg,
84%). 1H NMR (400 MHz, CDCl3): d = 7.41 (t, J = 7.1 Hz, 2H,
HAr), 7.26 (d, J = 7.1 Hz, 4H, HAr), 4.03 (s, 4H, NCH2), 3.08
(septet, J = 6.9 Hz, 4H, ArCHCH3), 1.38 (d, J = 6.9 Hz, 12H,
ArCHCH3), 1.36 ppm (d, J = 6.9 Hz, 12H, ArCHCH3); 13C NMR
(125 MHz, CDCl3): d = 203.1 (C, NCN), 146.6 (C, CAr), 134.3 (C,
CAr), 129.8 (CH, CAr), 124.5 (CH, CAr), 53.7 (CH2, NCH2), 28.9
(CH, ArCHCH3), 25.4 (CH3), 23.9 ppm (CH3); elemental analysis
calcd (%) for C27H38BrCuN2 (532.15): C, 60.72; H, 7.17; N, 5.25;
found: C, 60.53; H, 7.04; N, 5.25.
=
(400 MHz, CDCl3): d = 7.04 (s, 2H, NCH ), 2.43–2.38 (m,
12H, HAd), 2.31–2.24 (m, 6H, HAd), 1.82–1.76 ppm (m, 12H, HAd);
13C NMR (125 MHz, CDCl3): d = 171.2 (C, NCN), 115.3 (CH,
NCH ), 57.9 (C, CAd), 44.9 (CH2Ad), 35.7 (CH2Ad), 29.7 ppm (CH,
=
CHAd); elemental analysis calcd (%) for C23H32ClCuN2 (434.16):
C, 63.43; H, 7.41; N, 6.43; found: C, 63.81; H, 7.43; N, 6.26.
[1,3-Bis(adamantyl)imidazol-2-ylidene]copper(I) bromide [(IAd)
CuBr] (3b). In a vial fitted with a screw cap, copper bromide
dimethyl sulfide complex (0.370 g, 1.2 equiv), IAd·HBF4 (0.637 g,
1.5 mmol) and KOMe (0.148 g, 1.4 equiv), were loaded inside
a glovebox and stirred in refluxing dry toluene (5 mL) outside
the glovebox for 15 h. After filtering the reaction mixture at room
temperature through a plug of Celite (DCM), the filtrate was mixed
with pentane to form a precipitate. A second filtration led to the
isolation of the title compound as a white solid (0.611 g, 81%).
Suitable crystals for X-ray diffraction were grown by slow diffusion
of octane from a saturated solution of 3b in DCM. CCDC-
1
=
750318. H NMR (400 MHz, CDCl3): d = 7.04 (s, 2H, NCH ),
2.42–2.38 (m, 12H, HAd), 2.31–2.25 (m, 6H, HAd), 1.82–1.76 ppm
(m, 12H, HAd); 13C NMR (125 MHz, CDCl3): d = 172.1 (C,
NCN), 115.2 (CH, NCH ), 57.8 (C, CAd), 44.8 (CH2Ad), 35.6
=
(CH2Ad), 29.7 ppm (CH, CHAd); elemental analysis calcd (%) for
C23H32BrCuN2 (478.10): C, 57.56; H, 6.72; N, 5.84; found: C, 57.86;
H, 6.49; N, 5.71.
[1,3-Bis(2,6-diisopropylphenyl)imidazolin-2-ylidene]copper(I) io-
dide [(SIPr)CuI] (2c). (A) In a vial fitted with a screw cap, copper
iodide (0.314 g, 1.1 equiv), SIPr·HCl (0.640 g, 1.5 mmol) and
KOMe (0.126 g, 1.2 equiv), were loaded inside a glovebox and
stirred in dry toluene (5 mL) outside the glovebox for 15 h. After
filtering the reaction mixture through a plug of Celite (DCM), the
filtrate was mixed with pentane to form a precipitate. A second
filtration led to the isolation of the title compound as a white solid
(0.675 g, 77%). (B) In a vial fitted with a screw cap, copper iodide
(35 mg, 4 equiv) and [(SIPr)AgCl] (25 mg, 0.046 mmol), were
loaded inside a glovebox and stirred in dry acetonitrile (2.5 mL)
outside the glovebox for 15 h. After filtering the reaction mixture
through a plug of Celite (acetone), solvents were evaporated under
reduced pressure to isolate the title compound as a white solid
(24 mg, 90%). Suitable crystals for X-ray diffraction were grown
by slow diffusion of octane from a saturated solution of 2c in
[1,3-Bis(adamantyl)imidazol-2-ylidene]copper(I) iodide [(IAd)
CuI] (3c). In a vial fitted with a screw cap, copper iodide (0.428 g,
1.5 equiv), IAd·HBF4 (0.637 g, 1.5 mmol) and KOMe (0.157 g,
1.5 equiv), were loaded inside a glovebox and stirred in dry toluene
(5 mL) outside the glovebox for 15 h. After filtering the reaction
mixture through a plug of Celite (DCM), the filtrate was mixed
with pentane to form a precipitate. A second filtration led to the
isolation of the title compound as a white solid (0.614 g, 75%).
Suitable crystals for X-ray diffraction were grown by slow diffusion
of octane from a saturated solution of 3c in DCM. CCDC-
1
=
750319. H NMR (400 MHz, CDCl3): d = 7.04 (s, 2H, NCH ),
2.44–2.39 (m, 12H, HAd), 2.31–2.26 (m, 6H, HAd), 1.82–1.74 ppm
(m, 12H, HAd); 13C NMR (125 MHz, CDCl3): d = 174.4 (C,
NCN), 115.2 (CH, NCH ), 58.0 (C, CAd), 45.0 (CH2Ad), 35.7
=
1
MeCN. CCDC-750316. H NMR (400 MHz, CDCl3): d = 7.42
(CH2Ad), 29.8 ppm (CH, CHAd); elemental analysis calcd (%) for
C23H32CuIN2 (526.09): C, 52.42; H, 6.12; N, 5.32; found: C, 52.29;
H, 6.30; N, 5.13.
(t, J = 7.7 Hz, 2H, HAr), 7.26 (d, J = 7.7 Hz, 4H, HAr), 4.03 (s,
4H, NCH2), 3.08 (septet, J = 6.9 Hz, 4H, ArCHCH3), 1.38 (d,
J = 6.9 Hz, 12H, ArCHCH3), 1.36 ppm (d, J = 6.9 Hz, 12H,
ArCHCH3); 13C NMR (125 MHz, CDCl3): d = 204.5 (C, NCN),
146.6 (C, CAr), 134.1 (C, CAr), 129.9 (CH, CAr), 124.5 (CH, CAr),
53.7 (CH2, NCH2), 28.9 (CH, ArCHCH3), 25.5 (CH3), 23.9 ppm
(CH3); elemental analysis calcd (%) for C27H38CuIN2 (580.14): C,
55.81; H, 6.59; N, 10.94; found: C, 55.78; H, 6.23; N, 11.23.
[1,3-Bis(cyclohexyl)imidazol-2-ylidene]copper(I) cloride [(ICy)
CuCl] (6a). In a vial fitted with a screw cap, copper chloride
(0.248 g, 2.5 mmol) and ICy (0.580 g, 2.5 mmol), were loaded inside
a glovebox and stirred in dry THF (10 mL) outside the glovebox
for 15 h. After filtering the reaction mixture through a plug of
Celite (DCM), solvents were evaporated under reduced pressure
to isolate the title compound as an off-white solid (0.554 g, 71%).
Spectroscopic data for 6a was in good accordance with literature
data.7
[1,3-Bis(adamantyl)imidazol-2-ylidene]copper(I) chloride [(IAd)
CuCl] (3a). In a vial fitted with a screw cap, copper chloride
(0.171 g, 1.2 equiv), IAd·HBF4 (0.533 g, 1.26 mmol) and KOMe
(0.132 g, 1.5 equiv), were loaded inside a glovebox and stirred in
7604 | Dalton Trans., 2010, 39, 7595–7606
This journal is
The Royal Society of Chemistry 2010
©

View Online
[1,3-Bis(cyclohexyl)imidazol-2-ylidene]copper(I) bromide [(ICy)
CuBr] (6b). In a vial fitted with a screw cap, copper bromide
(0.236 g, 1.1 equiv) and ICy (0.348 g, 1.5 mmol), were loaded
inside a glovebox and stirred in dry THF (6 mL) outside the
glovebox for 15 h. After filtering the reaction mixture through
a plug of Celite (DCM), solvents were evaporated under reduced
pressure to isolate the title compound as an off-white solid (0.460 g,
82%). Suitable crystals for X-ray diffraction were grown by slow
(CH3); elemental analysis calcd (%) for C11H20BrCuN2 (323.74):
C, 40.81; H, 6.23; N, 8.65; found: C, 41.03; H, 6.39; N, 8.45.
[1,3-Bis(tert-butyl)imidazol-2-ylidene]copper(I) iodide [(ItBu)
CuI] (7c). In a vial fitted with a screw cap, copper iodide (0.286 g,
1.2 equiv) and ItBu (0.270 g, 1.5 mmol), were loaded inside a
glovebox and stirred in dry THF (15 mL) for 15 h. After filtering
the reaction mixture through a plug of Celite (DCM) inside the
glovebox, solvents were evaporated under reduced pressure to
isolate the title compound as a white solid (0.443 g, 80%). 1H NMR
diffusion of octane from a saturated solution of 6b in THF. CCDC-
1
=
750322. H NMR (400 MHz, CDCl3): d = 6.93 (s, 2H, NCH ),
4.36–4.22 (m, 2H, CHCy), 2.14–2.03 (m, 4H, CH2Cy), 1.98–1.82 (m,
4H, CH2Cy), 1.80–1.58 (m, 6H, CH2Cy), 1.56–1.37 (m, 4H, CH2Cy),
=
(400 MHz, C6D6): d = 6.26 (s, 2H, NCH ), 1.37 ppm (s, 18H,
CH3); 13C NMR (125 MHz, CDCl3): d = 175.6 (C, NCN), 116.5
1.31–1.15 ppm (m, 2H, CH2Cy); 13C NMR (125 MHz, C6D6): d =
=
(CH, NCH ), 57.8 (CH, CH(CH3)3), 32.1 ppm (CH3); elemental
Cy
176.6 (C, NCN), 117.4 (CH, NCH ), 61.3 (CH ), 35.1 (CH2Cy),
=
analysis calcd (%) for C11H20CuIN2 (370.74): C, 35.64; H, 5.44; N,
7.56; found: C, 35.81; H, 5.66; N, 7.33.
26.1 (CH2Cy), 25.6 ppm (CH2Cy); elemental analysis calcd (%) for
C15H24BrCuN2 (375.81): C, 47.94; H, 6.44; N, 7.45; found: C, 48.06;
H, 6.51; N, 7.56.
General procedure for the [(SIMes)CuCl]-catalyzed
hydrosilylation of ketones
[1,3-Bis(cyclohexyl)imidazol-2-ylidene]copper(I) iodide [(ICy)
CuI] (6c). In a vial fitted with a screw cap, copper iodide (0.314 g,
1.1 equiv) and ICy (0.348 g, 1.5 mmol), were loaded inside a
glovebox and stirred in dry THF (6 mL) outside the glovebox
for 15 h. After filtering the reaction mixture through a plug of
Celite (DCM), solvents were evaporated under reduced pressure to
isolate the title compound as a white solid (0.531 g, 84%). Suitable
crystals for X-ray diffraction were grown by slow diffusion of
octane from a saturated solution of 6c in THF. CCDC-750323.
1H NMR (400 MHz, C6D6): d = 6.29 (s, 2H, NCH=), 4.94–
4.77 (m, 2H, CHCy), 2.09–1.93 (m, 4H, CH2Cy), 1.68–1.51 (m, 4H,
CH2Cy), 1.50–1.38 (m, 6H, CH2Cy), 1.37–1.22 (m, 4H, CH2Cy), 1.03–
0.87 ppm (m, 2H, CH2Cy); 13C NMR (125 MHz, acetone-d6): d =
177.9 (C, NCN), 119.3 (CH, NCH=), 61.7 (CHCy), 35.7 (CH2Cy),
26.8 (CH2Cy), 26.4 ppm (CH2Cy); elemental analysis calcd (%) for
C15H24CuIN2 (422.81): C, 42.61; H, 5.72; N, 6.63; found: C, 42.75;
H, 5.81; N, 7.46.
In a vial fitted with a septum screw cap, [(SIMes)CuCl] 5a (8 mg,
0.02 mmol, 2 mol%) and sodium tert-butoxide (2.5 mg, 2 mol%)
were charged inside a glove box and stirred in dry THF (1 mL)
outside of the glove box for 5 min before adding triethylsilane
(0.2 mL, 1.2 mmol, 1.2 equiv) through the septum using a syringe.
After 5 more minutes of stirring, the ketone (1 mmol) was added.
When the starting material was a solid, it was added as a solution
in THF. The reaction was monitored by GC, after consumption
of the starting material or no further conversion, the reaction
mixture was opened to air and filtered through a plug of active
charcoal and Celite using DCM as solvent. The organic phase was
concentrated in vacuo and the purity of the residue established
1
by GC and H NMR analyses. Flash chromatography was then
performed unless crude product was estimated to be greater than
95% pure.
[1,3-Bis(tert-butyl)imidazol-2-ylidene]copper(I) chloride [(ItBu)
CuCl] (7a). In a vial fitted with a screw cap, copper chloride
(0.180 g, 1.2 equiv) and ItBu (0.270 g, 1.5 mmol), were loaded
inside a glovebox and stirred in dry THF (15 mL) for 15 h.
After filtering the reaction mixture through a plug of Celite
(DCM) inside the glovebox, the filtrate was mixed with pentane
to form a precipitate. A second filtration led to the isolation of
General procedure for the [(NHC)CuX]-catalyzed [3+2]
cycloaddition of azides and alkynes
In a vial fitted with a screw cap, azide (1.0 mmol), alkyne
(1.1 mmol) and [(NHC)CuX] (4.4 mg if 3c, or 3.6 mg if 5b,
0.8 mol%) were loaded. The reaction was allowed to proceed at
1
1
room temperature and monitored by H NMR or GC analysis
the title compound as a white solid (0.305 g, 73%). H NMR
of aliquots. After total consumption of the starting azide, the
solid product was simply dissolved in EtOAc and concentrated or
alternatively collected by filtration and washed with pentane. In
all examples, the crude products were estimated to be greater than
95% pure by 1H NMR. Reported yields are isolated yields and are
the average of at least two independent runs.
=
(400 MHz, CDCl3): d = 7.05 (s, 2H, NCH ), 1.79 ppm (s, 18H,
CH3); 13C NMR (125 MHz, CDCl3): d = 172.9 (C, NCN), 116.5
=
(CH, NCH ), 57.7 (CH, CH(CH3)3), 32.0 ppm (CH3); elemental
analysis calcd (%) for C11H20ClCuN2 (279.29): C, 47.31; H, 7.22;
N, 10.03; found: C, 47.24; H, 7.34; N, 10.30.
[1,3-Bis(tert-butyl)imidazol-2-ylidene]copper(I) bromide [(ItBu)
CuBr] (7b). In a vial fitted with a screw cap, copper bromide
(0.258 g, 1.2 equiv) and ItBu (0.270 g, 1.5 mmol), were loaded
inside a glovebox and stirred in dry THF (15 mL) for 15 h. After
filtering the reaction mixture through a plug of Celite (DCM)
inside the glovebox, solvents were evaporated under reduced
Acknowledgements
The ICIQ Foundation, ICREA, the ERC (Advanced Researcher
Grant to SPN, FUNCAT), and the Spanish Ministerio de Ciencia
e Innovacio´n are gratefully acknowledged for financial support.
SDG thanks the Ministerio de Ciencia e Innovacio´n for a Torres
Quevedo fellowship and Dr Nicolas Marion for insightful reading
of the manuscript.
pressure to isolate the title compound as a white solid (0.410 g,
1
=
85%). H NMR (400 MHz, C6D6): d = 6.27 (s, 2H, NCH ),
1.35 ppm (s, 18H, CH3); 13C NMR (125 MHz, CDCl3): d = 173.7
=
(C, NCN), 116.5 (CH, NCH ), 57.6 (CH, CH(CH3)3), 32.0 ppm
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 7595–7606 | 7605
©

View Online
26 (a) H. G. Raubenheimer, S. Cronje, P. H. van Rooyen, P. J. Olivier
and J. G. Toerien, Angew. Chem., Int. Ed. Engl., 1994, 33, 672–673;
(b) I. V. Shishkov, F. Rominger and P. Hofmann, Dalton Trans., 2009,
1428–1435.
27 For a similar X-ray structure of a [(NHC)AgX] complex, see: A. A. D.
Tulloch, A. A. Danopoulos, S. Winston, S. Kleinhenz and G. Eastham,
J. Chem. Soc., Dalton Trans., 2000, 2299–4506.
References
1 S. D´ıez-Gonza´lez and S. P. Nolan, Coord. Chem. Rev., 2007, 241, 874–
883.
2 (a) S. D´ıez-Gonza´lez, N. Marion and S. P. Nolan, Chem. Rev., 2009,
109, 3612–3676; (b) F. Glorius, N-Heterocyclic Carbenes in Transition
Metal Catalysis, Springer, Berlin, 2007.
3 A. J. Arduengo, III, H. V. R. Dias, J. C. Calabrese and F. Davidson,
Organometallics, 1993, 12, 3405–3409.
4 P. K. Fraser and S. Woodward, Tetrahedron Lett., 2001, 42, 2747–
2749.
5 (a) S. D´ıez-Gonza´lez and S. P. Nolan, Aldrichimica Acta, 2008, 41,
43–51; (b) S. D´ıez-Gonza´lez and S. P. Nolan, Synlett, 2007, 2158–
2167.
28 M. R. Churchill and F. J. Rotella, Inorg. Chem., 1979, 18, 166–171.
29 (a) J. C. Garrison, R. S. Simons, W. G. Koforn, C. A. Tessier and W. J.
Youngs, Chem. Commun., 2001, 1780–1781; (b) J. C. Garrison, R. S.
Simons, C. A. Tessier and W. J. Youngs, J. Organomet. Chem., 2003,
673, 1–4; (c) V. J. Catalano and M. A. Malwitz, Inorg. Chem., 2003, 42,
5483–5485.
30 (a) S. Gischig and A. Togni, Organometallics, 2005, 24, 203–205; (b) X.
Han, L.-L. Koh, A.-P. Liu, Z. Weng and T. S. A. Hor, Organometallics,
2010, 29, 2403–2405.
6 H. Kaur, F. K. Zinn, E. D. Stevens and S. P. Nolan, Organometallics,
2004, 23, 1157–1160.
7 S. D´ıez-Gonza´lez, H. Kaur, F. K. Zinn, E. D. Stevens and S. P. Nolan,
J. Org. Chem., 2005, 70, 4784–4796.
31 M. Hakansson, H. Eriksson, A. B. Ahman and S. Jagner, J. Organomet.
Chem., 2000, 595, 102–108.
32 G. M. Chiarella, D. Y. Melgarejo, A. Rozanski, P. Hempte, L. M. Perez,
C. Reber and J. P. Fackler, Jr., Chem. Commun., 2010, 136–138, and
references therein.
8 For important contributions from other research groups, see: (a) V.
Jurkauskas, J. P. Sadighi and S. L. Buchwald, Org. Lett., 2003, 5,
2417–2420; (b) J. Yun, D. Kim and H. Yun, Chem. Commun., 2005,
5181–5183; (c) B. Bantu, D. Wang, K. Wurst and M. R. Buchmeiser,
Tetrahedron, 2005, 61, 12145–12152; (d) C. Deutsch, B. H. Lipshutz and
N. Krause, Angew. Chem., Int. Ed., 2007, 46, 1650–1653; (e) C. Deutsch,
B. H. Lipshutz and N. Krause, Org. Lett., 2009, 11, 5010–5012; (f) M.
Nonnenmacher, D. Kunz and F. Rominger, Organometallics, 2008, 27,
1561–1568.
9 S. D´ıez-Gonza´lez, A. Correa, L. Cavallo and S. P. Nolan, Chem.–Eur. J.,
2006, 12, 7558–7564.
10 S. D´ıez-Gonzalez and S. P. Nolan, Angew. Chem., Int. Ed., 2008, 47,
8881–8884.
33 B. Marciniec, Hydrosilylation, A Comprehensive Review on Recent
Advances, Springer, Berlin, 2009.
34 For reviews on the metal-catalyzed hydrosilylation of carbonyl com-
pounds, see: (a) C. Deutsch and N. Krause, Chem. Rev., 2008, 108,
2916–2927; (b) S. D´ıez-Gonza´lez and S. P. Nolan, Acc. Chem. Res.,
2008, 41, 349–358; (c) S. Rendler and M. Oestreich, Angew. Chem.,
Int. Ed., 2007, 46, 498–504; (d) S. D´ıez-Gonza´lez and S. P. Nolan, Org.
Prep. Proced. Int., 2007, 39, 523–559; (e) O. Riant, N. Mostefa¨ı and J.
Courmarcel, Synthesis, 2004, 2943–2958.
35 For a very illustrative example with NHC ligands, see: (a) H. Lebel,
M. K. Janes, A. B. Charette and S. P. Nolan, J. Am. Chem. Soc., 2004,
126, 5046–5047; (b) For a relevant example with bis(oxazoline) ligans,
see: R. Rasappan, M. Hager, A. Gissibi and O. Reiser, Org. Lett., 2006,
8, 6099–6102.
36 To avoid the undesired formation of bis-NHC species, 5a must indeed
be prepared from SIMes·HCl instead, as described in ref. 7.
37 See the Supporting Information for details.
38 (a) A. Padwa, In Comprehensive Organic Synthesis, ed. B. M. Trost,
Pergamon, Oxford, 1991, vol. 4, pp. 1069-1109; (b) R. Huisgen, In
1,3-Dipolar Cycloaddition Chemistry, ed. A. Padwa, Wiley, New York,
1984, pp. 1-176.
11 S. D´ıez-Gonza´lez, E. D. Stevens and S. P. Nolan, Chem. Commun.,
2008, 4747–4749.
12 For contributions from other research groups see: (a) P. H. Li, L. Wang
and Y. C. Zhang, Tetrahedron, 2008, 64, 10825–10830; (b) G. M. Pawar,
B. Bantu, J. Weckesser, S. Blechert, K. Wurst and M. R. Buchmeiser,
Dalton Trans., 2009, 9043–9051; (c) M. L. Teyssot, A. Chevry, M.
Traikia, M. El-Ghozzi, D. Avignant and A. Gautier, Chem.–Eur. J.,
2009, 15, 6322–6326.
13 For references concerning [(NHC)CuF] complexes, see: (a) J. R. Herron
and Z. T. Ball, J. Am. Chem. Soc., 2008, 130, 16486–16487; (b) J. R.
Herron, V. Russo, E. J. Valente and Z. T. Ball, Chem.–Eur. J., 2009, 15,
8713–8716.
39 H. C. Kolb, M. G. Finn and K. B. Sharpless, Angew. Chem., Int. Ed.,
14 Between 5 and 10% of [(IPr)2Cu]+ was formed under these conditions
using up to 2.5 equivalents of CuI. This is in contrast with the previously
reported preparation of 1c, see: S. Zheng, F. Li, J. Liu and C. Xia,
Tetrahedron Lett., 2007, 48, 5883–5886.
2001, 40, 2004–2021.
40 For the first groundbreaking reports, see: (a) C. W. Tornøe, C.
Christensen and M. Meldal, J. Org. Chem., 2002, 67, 3057–3064;
(b) V. V. Rostovtsev, L. G. Green, V. V. Fokin and K. B. Sharpless,
Angew. Chem., Int. Ed., 2002, 41, 2596–2599; (c) For a comprehensive
review, see: M. Meldal and C. W. Tornøe, Chem. Rev., 2008, 108, 2952–
3015.
15 S. Okamoto, S. Tominaga, N. Saino, K. Kase and K. Shimoda,
J. Organomet. Chem., 2005, 690, 6001–6007.
16 J. Broggi, S. D´ıez-Gonza´lez, J. L. Petersen, S. Berteina-Raboin, S. P.
Nolan and L. A. Agrofoglio, Synthesis, 2008, 141–148.
17 M. Eriksson, T. Iliefski, M. Nilsson and T. Olsson, J. Org. Chem., 1997,
62, 182–187.
18 For representative examples, see: (a) N. P. Mankad, D. S. Laitar and
J. P. Sadighi, Organometallics, 2004, 23, 3369–3371; (b) L. A. Goj,
E. D. Blue, S. A. Delp, B. Gunnoe, T. R. Cundari, A. W. Pierpont, J. L.
Petersen and P. D. Boyle, Inorg. Chem., 2006, 45, 9032–9045.
19 For the preparation of [(ICy)CuCl] 6a from ICy·HBF4, see ref. 7.
20 I. J. B. Lin and C. S. Vasam, Coord. Chem. Rev., 2007, 251, 642–670,
and references therein.
41 For selected examples, see: (a) F. Pe´rez-Balderas, M. Ortega-Mun˜oz,
J. Morales-Sanfrutos, F. Herna´ndez-Mateo, F. G. Calvo-Flores, J. A.
Calvo-As´ın, J. Isac-Garc´ıa and F. Santoyo-Gonza´lez, Org. Lett., 2003,
5, 1951–1954; (b) T. R. Chan, R. Hilgraf, K. B. Sharpless and V. V.
Fokin, Org. Lett., 2004, 6, 2853–2855; (c) S. S. Gupta, J. Kuzelka, P.
Singh, W. G. Lewis, M. Manchester and M. G. Finn, Bioconjugate
Chem., 2005, 16, 1572–1579; (d) B. Gerard, J. Ryan, A. B. Beeler and
J. A. Porco, Jr., Tetrahedron, 2006, 62, 6405–6411; (e) V. O. Rodionov,
S. I. Presolsky, S. Gardinier, Y.-H. Lim and M. G. Finn, J. Am. Chem.
Soc., 2007, 129, 12696–12704; (f) N. Candelon, D. Laste´coue`res, A. K.
Diallo, J. Ruiz Aranzaes, D. Astruc and J.-M. Vincent, Chem. Commun.,
2008, 741–743.
21 A. C. Badaj, S. Dastgir, A. J. Lough and D. G. Lavoie, Dalton Trans.,
2010, 39, 3361–3365.
22 For a compilation of the [(NHC)CuX] complexes preparation, see the
Supporting Information.
42 V. D. Bock, H. Hiemstra and J. H. van Maarseveen, Eur. J. Org. Chem.,
2006, 51–68.
23 S. D´ıez-Gonza´lez, E. D. Stevens, N. M. Scott, J. L. Petersen and S. P.
Nolan, Chem.–Eur. J., 2008, 14, 158–168.
43 A. J. Arduengo, III, R. Krafczyk, R. Schmutzler, H. A. Craig, J. R.
Goerlich, W. J. Marshall and M. Unverzagt, Tetrahedron, 1999, 55,
14523–14534.
44 P. de Fre´mont, N. M. Scott, E. D. Stevens, T. Ramnial, O. C. Lightbody,
C. L. B. Macdonald, J. A. C. Clyburne, C. D. Abernethy and S. P. Nolan,
Organometallics, 2005, 24, 6301–6309.
24 The previously reported [(ICy)CuCl] 6a presents a longer Cu–C of
˚
2.11 A, see ref. 7.
25 N. P. Mankad, T. G. Gray, D. S. Laitar and J. P. Sadighi,
Organometallics, 2004, 23, 1191–1193.
7606 | Dalton Trans., 2010, 39, 7595–7606
This journal is
The Royal Society of Chemistry 2010
©