Design and synthesis of new bidentate alkoxy-NHC ligands for enantioselective copper-catalyzed conjugate addition

Article abstract of DOI:10.1016/j.jorganchem.2005.04.027

A new family of chiral alkoxy-N-heterocyclic carbene (NHC) ligands has been designed for the enantioselective copper-catalyzed conjugate addition of dialkylzincs to enones. These new bidentate NHC ligands were synthesized in high overall yields using a five-step procedure starting from commercially available β-aminoalcohols. Influence of the temperature, base, solvent and copper source were studied in order to optimize the stereoselectivity of the addition. High reactivity and excellent enantioselectivity were obtained at ambient temperature with a range of cyclic enones and dialkylzinc. Addition to acyclic enones has also been studied.

Full text of DOI:10.1016/j.jorganchem.2005.04.027

Journal of Organometallic Chemistry 690 (2005) 5237–5254
www.elsevier.com/locate/jorganchem
Design and synthesis of new bidentate alkoxy-NHC ligands
for enantioselective copper-catalyzed conjugate addition
a
Herve Clavier , Ludovic Coutable , Lo¨ıc Toupet ,
a
b
´
Jean-Claude Guillemin , Marc Mauduit
a
a,
*
a
` ´ ´
UMR CNRS 6052, Laboratoire de Syntheses et Activations de Biomolecules Ecole Nationale Superieure de Chimie de Rennes,
Institut de Chimie de Rennes, 35700 Rennes, France
´
b
` ´ ´
UMR CNRS 6626, Groupe Matieres Condensees et Materiaux Universite de Rennes 1, 35700 Rennes, France
Received 27 January 2005; received in revised form 5 April 2005; accepted 8 April 2005
Available online 6 June 2005
Abstract
A new family of chiral alkoxy-N-heterocyclic carbene (NHC) ligands has been designed for the enantioselective copper-catalyzed
conjugate addition of dialkylzincs to enones. These new bidentate NHC ligands were synthesized in high overall yields using a five-
step procedure starting from commercially available b-aminoalcohols. Influence of the temperature, base, solvent and copper source
were studied in order to optimize the stereoselectivity of the addition. High reactivity and excellent enantioselectivity were obtained
at ambient temperature with a range of cyclic enones and dialkylzinc. Addition to acyclic enones has also been studied.
Ó 2005 Elsevier B.V. All rights reserved.
Keywords: Chiral N-heterocyclic carbene; Alkoxy-imidazolinium salts; Asymmetric catalysis
1. Introduction
better thermal stability than complexes containing phos-
phine units, allow many synthetic transformations
requiring harsher conditions. Use of NHCs in enantiose-
lective catalysis is a natural extension in this field and
many complexes containing chiral NHC unit have been
synthesized and used in stereoselective catalysis with
moderate to high enantiomeric excess [6] (see Fig. 2).
In the copper-catalyzed area, Arduengo has devel-
oped a few NHC–copper complexes in 1993 but the cat-
alytic properties of these complexes were not
investigated [7]. It was not until 1999 that Woodward
et al. could demonstrated that a copper complex con-
taining an Arduengo-type carbene 1 accelerates consid-
erably the copper-catalyzed addition of diethylzinc to
cyclohexenone [8]. In light of these results, Alexakis
and Mangeney [9] have reported the use of chiral NHCs
in the copper-catalyzed conjugate addition [10] in 2001.
Designed a C2-symmetry strategy, the chiral NHC pre-
cursors 2 (Fig. 2) gave low to good enantioselectivities
in the alkylation of diethylzinc to cyclic enones (up to
During the last decade, the chemistry of carbene has
gained importance in the catalysis area through the use
of carbene species as spectator ligands for transition-me-
tal centers. The spectacular advance in this area is due to
the remarkable efforts and strategies developed by chem-
ists to build stable and isolable carbene species [1]. N-
Heterocyclic carbenes (NHC) are the most well known
example of this phenomenon [2]. Indeed, since the first
isolation and characterization of stable NHC 1 by Ardu-
engo et al. [3] in 1991 (see Fig. 1), many complexes con-
taining NHC have been synthesized and used in
important organometallic transformations including
many C–C and C–N cross-coupling [4] and the popular
metathesis reactions [5]. These new NHC complexes
showing an exceptional catalytic activity associate to a
*
Corresponding author. Tel.: +33 223 238 112; fax: +33 223 238 108.
E-mail address: marc.mauduit@ensc-rennes.fr (M. Mauduit).
0022-328X/$ - see front matter Ó 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2005.04.027

5238
H. Clavier et al. / Journal of Organometallic Chemistry 690 (2005) 5237–5254
R
PF6
N
N
O
R
NH2
N
N
+
HO
NH2
Y
Y
HO
1
5
Fig. 1.
O
N
N
PF
6
O
3 mol%
R
N
R
5a
HO
BF4
N
(1)
Et2Zn
(1.5 equiv)
+
Ar
Ar
8 mol% nBuLi,
2 mol% Cu(OTf)2
Et
R1
R1
Et2O, -50˚C
>99%
66% ee
2
Alexakis/Mangeney (2001)
Scheme 1. Use of chiral alkoxy-imidazolinium salts 5 in enantioselec-
tive 1,4-addition.
Cl
N
N
N
N
ligands in the enantioselective copper-conjugate addi-
tion of diethylzinc to cyclohexenone (Eq. (1)). The de-
sired copper (II)-alkoxy-NHC catalyst is formed in
situ by adding two equivalents of n-butyllithium to the
alkoxy-imidazolinium salt 5 in the presence of the cop-
per salt Cu(OTf)2 at ꢀ50 °C in diethylether. Under these
conditions, the conjugate addition of diethylzinc to
2-cyclohexenone is finished within 4 h. However, low
to moderate enantioselectivities were obtained, reaching
66% ee with imidazolinium salt 5a derived from tert-leu-
cine [15].
Me
Cl
Mes
Cu
O
THF
HO
Hoveyda (2002/2004)
3
4
Arnold (2004)
Fig. 2.
93% ee). Recently, Okamoto et al. [11] have used the
same ligands 2 in copper-catalyzed allylic alkylations,
but moderate enantioselectivities were obtained (up to
70% ee).
The high reactivity but modest enantioselectivity ob-
tained with these salts 5 led us to place the chiral center
on the alkoxy side chain near the imidazolinium ring.
As reported in a preliminary study [16], the alkoxy-
imidazolinium salts 6 (Fig. 3) lead to efficient copper
(II)-alkoxy-NHC catalysts for the enantioselective
copper-catalyzed conjugate addition to 2-cyclohexenone.
Herein, we report an extension of this new class of
chiral chelating alkoxy-imidazolinium salts through the
synthesis of several other salts 6 and 7 (Fig. 3) easily
accessible from commercially available b-aminoalcohols
and various substituted anilines or alkylamines. A com-
plete study of these new ligands in the enantioselective
conjugate addition to various cyclic enones and also
other Michael acceptors involving acyclic enones and
nitroalkenes was performed. Influence of the aryl and al-
kyl groups of the ligand in the stereoselective induction
and the reaction conditions (effect of the temperature,
the base, the solvent and the copper source) were fully
studied for optimization of the enantioselective
addition.
In 2002, Hoveyda et al. [12] described a very efficient
chiral bidentate NHC precursor 3 based on an axial
symmetric aminohydroxybinaphtalene, which was first
used to form a chiral Ru-complex for asymmetric olefin
metathesis (up to 96% ee). Hoveyda has recently studied
its use in copper-catalyzed allylic alkylations and ob-
tained high enantioselectivities (up to 98% ee) showing
the power of this new class of chelating NHCs. At the
same time, Arnold et al. [13] isolated the first chiral che-
lating alkoxy-NHC–copper (II) complex 4 and use it in
the copper conjugate addition of diethylzinc on cycloh-
exenone. However, moderate enantioselectivity was ob-
tained (up to 51% ee).
Recently, we have reported the first synthesis of chiral
alkoxy-imidazolinium salts 5 [14] (Scheme 1) derived
from amino acid in a six-step procedure. These salts
have been first used as chiral derivatizing agents for ee
determination of chiral carboxylate. Based on the work
of Woodward and Alexakis and Mangeney, we decided
to evaluate their potential as new chelating alkoxy-NHC
PF6
PF6
R2
R3
R2
H2N
HO
R2
R3
R2
H2N
N
N
N
N
Mes
R1
H
H
HO
HO
HO
6
7
Fig. 3.

H. Clavier et al. / Journal of Organometallic Chemistry 690 (2005) 5237–5254
5239
R2
R3
OH
2. Preparation of chiral alkoxy-imidazolinium salts 6
and 7
O
O
R1-NH2
H2N
OEt
OEt
R1HN
Cl
The simple procedure for synthesizing alkoxy-imidaz-
pyridine, CH2Cl2
0˚C
CH2Cl2, reflux
O
O
olinium salts 6 and 7 from enantiopure amino-alcohols
is described in Scheme 2 [17]. Ethyloxalylchloride 8 is
condensed onto primary amine (aniline or alkylamine)
to give the corresponding oxanilic ethylester 9. Treat-
ment with the chiral aminoalcohol in refluxed dichloro-
methane provides the oxalamide 10 in pure form
without purification. After LiAlH4 reduction, the result-
ing diamine 11 was treated with anhydrous HCl fol-
lowed by condensation with trimethylorthoformate to
provide the desired imidazolinium chloride.
8
9
O
LiAlH4
THF, reflux
R2
R3
R2
R3
NH
HN HN
R1
R1HN
O
10
HO
HO
11
PF6
R2
R1 N
N
HO
1) HCl:MeOH/Et2O then
Finally, the chloride salt was easily purified by a sim-
ple anion exchange with KPF6 to give the corresponding
potassium hexaflurophosphate alkoxy-imidazolinium
salts 6 or 7 in pure form. This short and modular synthe-
sis of chiral alkoxy-NHCs provides an efficient route to
the development of new class of chiral catalysts. Two
different families of chelating NHC ligands have been
developed for the copper-catalyzed conjugate addition:
one derived from 2-substituted b-amino alcohols (imi-
dazolinium salts 6a–p, Scheme 3) and the other one de-
rived from several 1,2-disubstituted b-amino alcohols
(imidazolinium salts 7a–d, Scheme 4).
HC(OMe)3 toluene 90˚C
6a-p
or
2) KPF6/H2O
PF6
R2
R3
R1 N
N
HO
7a-d
Scheme 2. Synthesis of alkoxy-imidazolinium salts 6 and 7.
of this new type of NHC ligands (Fig. 4). Importantly,
all imidazolinium salts were isolated in pure form with
modest to good overall yields (33–76%). They are air
Two X-ray analyses were performed on salts 6a and
7b to confirm the structure and absolute configuration
PF6
PF6
PF6
PF6
N
N
N
N
N
N
N
N
HO
HO
HO
HO
6a (76%)
6b (66%)
6c (71%)
6d (74%)
Ph
PF6
PF6
PF6
PF6
iPr
N
N
N
N
N
N
N
N
iPr
HO
HO
HO
HO
6e (76%)
6f (73%)
6g (67%)
6h (59%)
PF6
PF6
PF6
PF6
N
N
N
N
N
N
N
N
tBu
SMe
HO
HO
HO
HO
6i (54%)
6k (53%)
6l (72%)
6j (33%)
PF6
PF6
N
PF6
PF6
N
N
N
N
N
N
N
HO
HO
HO
HO
6p(70%)
6o (83%)
6m (71%)
6n (80%)
Scheme 3. Alkoxy-imidazolinium salts 6a–p derived from 2-substituted b-amino alcohols.

5240
H. Clavier et al. / Journal of Organometallic Chemistry 690 (2005) 5237–5254
PF6
PF6
PF6
PF6
N
N
N
N
N
N
N
N
HO
HO
HO
HO
7d (83%)
7c (51%)
7a (52%)
7b (43%)
Scheme 4. Alkoxy-imidazolinium salts 7a–d derived from 1,2-disubstituted b-amino alcohols.
Fig. 4. X-ray structures of alkoxy-imidazolinium salts 6a and 6b (PF6 counter anions were omitted for clarity).
Table 1
Optimization of reaction conditions: importance of the temperature
stable and can be synthesized in multigram scale without
further purification.
PF6
N
N
O
O
3 mol%
HO
6a
R
3. Enantioselective 1,4-addition of diethylzinc to
cyclohexen-2-one: optimization of reaction conditions
+
Et2Zn
8 mol% nBuLi
(1.5 equiv)
2 mol% Cu(OTf)2 or Cu(eaa)2
Et2O, T˚C, time
conv. >99%
The potential of this new class of chiral alkoxy-NHCs
in the enantioselective copper-conjugate addition was
then evaluated. We chose the alkylation of cyclohexen-
2-one with diethylzinc as the reference reaction. To
optimize the reactivity and the enantiocontrol of the
conjugate addition, the effect of temperature, of copper
source, of ligand:precatalyst ratio, of base and finally of
solvent were examined. This study has been achieved
with the imidazolinium salt 6a.
Entry
T (°C)
Cu(OTf)2
Time (h)
Cu(eaa)2
Time (h)
ee (%)a
ee (%)a
1
2
3
4
5
6
a
ꢀ50
ꢀ78
ꢀ25
0
4
10
1
75
15b
78
83
86
84
1
76
70
68
56
48
45
4
1
0.5
0.5
0.5
0.5
0.5
0.5
20
45
Determined by chiral GC analysis (Lipodex E).
Conv. = 42%.
b
3.1. Influence of the temperature
Results of the study of the effect of temperature on
the reactivity and enantioselectivity of the conjugate
addition are summarized in Table 1.
(Cu(eaa)2) accelerates the reaction allowing the addition
to be carried out in only 1 h without loss of enantioselec-
tivity (entry 2).
Complete conversion was obtained in diethylether
after 8 h at ꢀ50 °C with 75% ee using 3 mol% of imidaz-
olinium salt 6a and 8 mol% of nBuLi in the presence of
2 mol% of copper (II) triflate (Cu(OTf)2) as precatalyst
(entry 1). Replacement of precatalyst by a more soluble
copper complex such as copper (II) ethylacetoacetate
Lowering the reaction temperature (ꢀ78 °C) led to a
slight decrease of enantioselectivity to 70% ee with
Cu(eaa)2 (entry 2). The phenomenon was dramatically
accentuated with Cu(OTf)2 where the stereoselectivity
reached only 15% ee. We are unable to bring an
adequate explanation for this important difference of

H. Clavier et al. / Journal of Organometallic Chemistry 690 (2005) 5237–5254
5241
Table 2
Evidence of temperature effect in the formation of the alkoxy-NHC
copper (II) catalyst species
stereoselectivity between these two pre-catalysts. We then
decided to perform the reaction at higher temperature. In
the case of Cu(eaa)2, significant decrease of the enantiose-
lectivity was observed with 48% ee at ambient tempera-
ture. However, in the case of Cu(OTf)2, as clearly shown
in the graphic (Fig. 5), the effect was curiously inverted
to obtain higher enantioselectivities at higher tempera-
ture. Thus, the best result (86% ee, entry 5) was obtained
by performing the reaction at room temperature.
PF6
N
N
O
O
3 mol%
HO
6a
R
+
Et2Zn
8 mol% nBuLi
(1.5 equiv)
2 mol% Cu(OTf)2
Et2O
Hoveyda [18] and Leighton [19] have recently ob-
served similar unusual relationship between temperature
and enantioselectivity in the copper-catalyzed enantiose-
lective conjugate addition using chiral phosphine li-
gands. To explain this surprising phenomenon, we
postulated that the catalytic copper species formed at
lower temperature are different than species formed at
ambient temperature, and performed two experiments
with Cu(OTf)2 to test this (Table 2). First, we preformed
the alkoxy-NHC copper (II) catalyst at 0 °C and per-
formed the addition at ꢀ50 °C (entry 1). In this case, a
very poor enantioselectivity was observed (49%). Inver-
sely, when we preformed the copper catalyst at ꢀ50 °C
and added the 2-cyclohexenone and the diethylzinc at
ambient temperature, good enantioselectivity up to
85% was obtained. In the first case, the copper clusters
formed at ambient temperature conserve their form at
lower temperature and are inactive. In the second case,
we suspect that the copper cluster formed at ꢀ50 °C is
immediately transformed to the active cluster when the
reaction takes place at 20 °C.
Entry
1
Conditions
Time
(h)
Conv.
(%)a
ee
(%)b
nBuLi was added at O °C
Et2Zn and 2-cyclohexenone
were added at ꢀ50 °C
2
20
49
85
2
nBuLi was added at ꢀ50 °C
Et2Zn and 2-cyclohexenone
were added at 20 °C
1
>99
a
Determined by GC analysis.
Determined by chiral GC analysis (Lipodex E).
b
3.2. Importance of the hydroxymethylene side chain in the
enantiocontrol of the conjugate addition
The hydroxymethylene side chain in the NHC ligand
is probably crucial for the enantiocontrol of the conju-
gate addition. To prove this, the hydroxyl function
was protected by a tert-butyldimethylsilyl group to
avoid the covalent O–metal bond. Fortunately, it was
possible to realize several synthetic transformations di-
rectly onto the imidazolinium salts by using a weak
base. Thus, the protected imidazolinium 12 was isolated
in excellent yield from imidazolinium salt 6a using a
small excess of Et3N as proton sponge (Scheme 5). Its
use in conjugate addition of diethylzinc to cyclohexe-
none at room temperature led to a significant decrease
in stereoselectivity (24% ee).
PF6
PF6
TBDMSCl
N
N
N
N
Mes
Mes
Et3N, DMAP
CH2Cl2, 12h, RT
TBDMSO
HO
6a
94%
12
PF6
N
N
O
O
3 mol%
TBDMSO
12
R
+
Et2Zn
(1.5 equiv)
8 mol% nBuLi
2 mol% Cu(OTf)2
Et2O, 20˚C, 2h
Conv.> 99%
24% ee
Fig. 5. Unusual relationship between the copper source, the temper-
ature and the enantioselectivity.
Scheme 5.

5242
H. Clavier et al. / Journal of Organometallic Chemistry 690 (2005) 5237–5254
PF6
N
N
O
3 mol%
O
MeO
13
R
+
Et2Zn
8 mol% nBuLi
2 mol% Cu(OTf)2
Et2O, 20˚C, 2h
(1.5 equiv)
Conv.> 99%
42 % ee
Scheme 6.
Table 3
PF6
Study of the copper source in the enantioselective addition
N
N
O
3 mol%
O
PF6
HO
N
N
O
O
ent-6a
3 mol%
S
+
Et2Zn
HO
6a
8 mol% nBuLi
2 mol% Cu(OTf)2
Et2O, 20˚C, 1h
(1.5 equiv)
R
+
Et2Zn
8 mol% nBuLi
2 mol% [Cu]
(1.5 equiv)
Conv.> 99%
conv. >99%
ee (%)a
83 % ee
Et2O, 20˚C
Entry
Copper precatalyst
Time (h)
Scheme 7.
1
2
3
4
5
6
7
8
a
Cu(OTf)2
CuOTfÆC6H6
Cu(OAc)2 ÆH2O
Cu(eaa)2
0.5
1
86
80
80
48
54
60
59
65
Moreover, we tested the methyl ether imidazolium
salt 13 to observe if the dative MeO–metal bond allowed
to conserve similar enantioselectivity in the conjugate
addition. However, 42% ee has been obtained indicating
that covalent O–metal bond is critical for the enantio-
control of the reaction (Scheme 6). The requirement of
the hydroxyl function for high enantioselectivity
strongly suggests that these alkoxy-NHC ligands act as
a LX bidentate ligand.
1
1
Cu(acac)2 ÆH2O
CuCl2
CuCl
1
1
2
CuTC
1
Determined by chiral GC analysis (Lipodex E).
At this stage of the study, it was necessary to verify
whether the absolute configuration of the stereogenic
center in the alkoxy-NHC ligand showed a significant
influence on the enantiocontrol of the conjugate addi-
tion. Thus, using the b-amino alcohol derived from
unnatural D-leucine, the enantiomer of imidazolinium
salt 6a was synthesized. A slight variation of enantiose-
lectivity (83% ee instead of 86% ee) was observed with
the (ent)-6a salt in the alkylation of Et2Zn on 2-cycloh-
exenone (Scheme 7).
At last, we tested the copper (I) salts such as CuCl or the
well-known Cu–TC developed by Alexakis [10], to finally
reach 59–65% ee. In conclusion, the X2 type copper
salts are the most efficient pre-catalysts for the conjugate
addition using the alkoxy-NHC ligands.
Because the alkoxy-NHC copper catalyst is formed in
situ, we then evaluated the ratio between the chiral li-
gand and the copper salt. As shown in Table 4, the best
enantioselectivity is reached with a small excess of alk-
oxy-NHC ligand (ratio 3/2, entry 2). A large excess of
chiral ligand (entries 3 and 4) leads to a dramatic effect
on the enantioselectivity, which fell to 29% ee.
3.3. Influence of the copper pre-catalyst source and the
ligand/copper salt ratio
Moreover, we optimized the catalytic process where
the Cu(OTf)2 loading could be decreased to 0.10 mol%
and the alkoxy-NHC ligand 6a loading to 0.15 mol%.
Under these conditions, complete conversion was ob-
served after 1 h without significant loss of enantioselec-
tivity (83% ee, Scheme 8).
As we can see above, the copper salt presents an
important effect of the enantioselectivity. Thus, we
tested several other copper pre-catalysts including cop-
per (I) salts in the enantioselective addition at ambient
temperature. Compared to the initial copper triflate salt,
the benzene–Cu(OTf) complex or the hydrated copper
acetate complex gave lower ee (Table 3, entries 2 and
3). Like Cu(eaa)2, the LX type copper complex Cu(aca-
c)2 ÆH2O gave poor results at ambient temperature.
We then tested the less expensive copper (II) chloride.
In this case, the enantioselectivity reached only 60% ee.
3.4. Influence of the base and the solvent
We then studied the effect of the base used for the for-
mation of the alkoxy-NHC copper cluster on the enan-
tiocontrol of the conjugate addition. Three different

H. Clavier et al. / Journal of Organometallic Chemistry 690 (2005) 5237–5254
5243
Table 4
Importance of the ligand/copper precatalyst ratio
is basic enough to generate the N-heterocyclic carbene
but not the alkoxide.
PF6
In the case of metallic base such as n-butyllithium,
potassium or sodium tert-butoxide, the enantioselectiv-
ity was very similar, around 86% ee (entries 1, 3 and
4). However, when carbonate bases were used, the
enantioselectivity was strongly affected and reached
56% ee in the best case (obtained with BaCO3, entry
8). These results could be explained by the lower basicity
of these inorganic bases. Indeed, the carbonates were
able to form the carbene but were not basic enough to
generate the alkoxylate and form the effective covalent
O–metal bond. More interestingly, the use of organic
base, which are often less expensive and easier to handle
than metallic bases, gave similarly high enantioselectiv-
ity (87% ee, entry 9). However, this was only observed
with strong organic base, such as DBU.
N
N
O
O
X mol%
HO
6a
R
+
Et2Zn
2.6X mol% nBuLi
Y mol% Cu(OTf)2
(1.5 equiv)
conv. >99%
ee (%)a
Et2O, 20˚C
Entry
6a/Cu(OTf)2
Time (h)
1
2
3
4
a
2/2
3/2
5/2
10/2
1
82
86
52
29
0.5
1
1
Determined by chiral GC analysis (Lipodex E).
We have therefore studied the effect of the solvent
and the results are summarized in Table 6. By perform-
ing the addition in toluene or THF in the presence of
DBU instead of nBuLi, the best enantiocontrol was ob-
tained and reached 89% ee (entries 4 and 6). One other
advantage in the use of organic base is the possibility
to perform the conjugate addition in organic solvents
such as dichloromethane, acetonitrile or ethylacetate,
which are known to be incompatible with metallic bases.
However, the enantioselectivity was not really improved
in these solvents (around 84%, entries 7–9).
PF6
N
N
O
O
HO
6a
0.15 mol%
Et2Zn
+
0.20 mol% nBuLi
0.10 mol% Cu(OTf)2
Et2O, 20˚C, 1h
(1.5 equiv)
>99%
83% ee
Scheme 8. Optimization of catalytic process.
To conclude this section, copper (II) triflate is the
most efficient complex to perform the conjugate addi-
tion of diethylzinc to cyclohexen-2-one at room temper-
ature in quantitative yield and high enantioselectivity
with these new bidentate alkoxy-NHC ligands. A small
excess of chiral ligand (ratio 3:2) gives the best ees.
bases were successively evaluated: metallic base, carbon-
ate base and organic base. In a first time, we performed
the reaction without base and were surprised to obtain
28% ee (Table 5, entry 2). We presume that diethylzinc
Table 5
Variation of the base
PF6
N
N
Table 6
Variation of the solvent
O
O
3 mol%
HO
6a
PF6
R
+
Et2Zn
N
N
O
8 mol% Base
2 mol% Cu(OTf)2
Et2O, 20˚C
O
(1.5 equiv)
3 mol%
HO
6a
R
+
Et2Zn
8 mol% Base
2 mol% Cu(OTf)2
Solvent, 20˚C
(1.5 equiv)
Entry
Base
Time (h)
Conv. (%)a
ee (%)b
Conv.> 99%
Time (h)
ee (%)a
1
2
nBuLi
Neat
0.5
1
>99
>99
>99
>99
>99
15
86
28
87
85
43
45
35
56
87
72
25
Entry
Solvent
Base
3
tBuOK
NaOtBu
Cs2CO3
Ag2CO3
K2CO3
BaCO3
DBU
1
4
2
1
2
3
4
5
6
7
8
9
a
Et2O
Et2O
THF
THF
nBuLi
DBU
nBuLi
DBU
nBuLi
DBU
DBU
DBU
DBU
0.5
1
86
87
85
89
75
89
82
85
84
5
1
6
4
1
7
1
>99
>99
>99
>99
>99
1
8
1
Toluene
Toluene
CH2Cl2
AcOEt
1
9
1
1
10
11
a
Pyridine
Et3N
1
1
1
1
CH3CN
1
Determined by GC analysis.
b
Determined by chiral GC analysis (Lipodex E).
Determined by chiral GC analysis (Lipodex E).

5244
H. Clavier et al. / Journal of Organometallic Chemistry 690 (2005) 5237–5254
The alkoxymethylene side chain is necessary in order to
obtain good enantiocontrol of the reaction. Thus, to
form the alkoxylate-NHC bidentate ligand, the use of
metallic bases is not really necessary and can be replaced
by a strong organic base. At last, any common organic
solvent can be used without significant variation of the
enantioselectivity.
the mesityl unit by a less hindered aromatic group such
as phenyl (imidazolinium 6g) resulted in a dramatic ef-
fect on the enantioselectivity reaching 54% ee. This effect
becomes more pronounced when the mesityl is replaced
by a more hindered aromatic moiety such as 2,6-(iPr)-
C6H3 group (imidazolinium 6h), where the ee decreases
to 29%. Hoveyda has recently observed this phenome-
non during the enantioselective allylic substitution with
the bidentate NHC ligand 3 [12]. Gade and Bellemin-
Laponaz [6i] have also reported the effect of the NHC
aromatic substituents size in the enantioselectivity dur-
ing the asymmetric hydrosilylation reaction catalyzed
by an oxazoline–NHC rhodium complex. On the other
hand, desymmetrization of the aromatic unit with a
bulky substituent such as tert-butyl group (imidazolin-
ium 6i) or a chelating thioalkyl ether group (imidazolin-
ium 6j) in ortho position do not improve the ee (77% and
75% ee, respectively).
The aromatic unit in the alkoxy-NHC ligand appears
to be crucial in order to obtain higher enantioselectivity
in the copper conjugate addition. We therefore investi-
gated the potential of the other alkoxy-NHC ligands
6k–p bearing an alkyl group instead of the aromatic unit
(Table 7). Moderate to poor enantioselectivities were
observed with these salts despite the presence of the
bulky tert-butyl group (imidazolinium 6k, entry 1).
Similarly moderate enantioselectivities were obtained
with the imidazolinium 6m–p bearing a benzylic group
4. Enantioselective 1,4-addition of diethylzinc to
cyclohexen-2-one: design of the structure ligand
Having established the optimal conditions for the use
of these new chelating NHCs in the conjugate addition,
we have then studied the relationships between the chiral
backbone of the imidazolinium salts and the stereoselec-
tivity. We initially examined the influence of the alkyl
group of the stereogenic center in the alkoxy-side chain
derived from the b-amino alcohols (Fig. 6). Except for
the methyl group which gave lower ee (imidazolinium
salt 6c, 79% ee), we have observed a slight difference
in the enantioselectivity between the more hindered
tert-butyl group (imidazolinium salt 6b, 87% ee) and
the other alkyl groups such as phenyl (86% ee, 6d), ben-
zyl (83% ee, 6e) and isopropyl (85% ee, 6f).
Most important in the structural relationship, the
aromatic unit derived from the anilines leads to a signif-
icant variation of the enantioselectivity. Replacement of
O
O
3 mol% Alkoxy-NHC 6a-j
8 mol% nBuLi
R
Et2Zn
(1.5 equiv)
2 mol% Cu(OTf)2
Et2O, 20˚C, 1h
Conv.> 99%
PF6
PF6
PF6
PF6
N
N
N
N
N
N
N
N
HO
HO
HO
HO
6a
6b
6c
6d
86% ee
87% ee
79% ee
86% ee
PF6
PF6
PF6
PF6
iPr
N
N
N
N
N
N
N
N
iPr
HO
HO
HO
HO
6g
6e
6f
6h
83% ee
85% ee
54% ee
29% ee
PF6
PF6
N
N
N
N
SMe
tBu
HO
HO
6j
6i
77% ee
Fig. 6. Evaluation of the alkoxy-NHC ligands 6a–j.
75% ee

H. Clavier et al. / Journal of Organometallic Chemistry 690 (2005) 5237–5254
5245
Table 7
Evaluation of the alkoxy-NHC ligands 6k–p
Table 8
Evaluation of the 1,2-disubstituted alkoxy-NHCs 7a–d
O
O
O
O
3 mol%
Alkoxy-NHC 6k-p
3 mol% Alkoxy-NHC 7a-d
R
Et2Zn
Et2Zn
8 mol% nBuLi
2 mol% Cu(OTf)2
Et2O, 20˚C
8 mol% nBuLi
(1.5 equiv)
(1.5 equiv)
2 mol% Cu(OTf)2
Et2O, 20˚C
Conv.> 99%
Conv.> 99%
ee (%)a
Entry
Alkoxy-NHC
Time (h)
ee (%)a
Entry
Alkoxy-NHC
Time (h)
1
2
3
4
5
6
a
6k
6l
2
1
1
1
1
4
73
71
63
75
80
2
1
2
3
4
a
7a
7b
7c
7d
1
42 (R)
42 (S)
7 (S)
0.5
1
6m
6n
6o
6p
1
17 (R)
Determined by chiral GC analysis (Lipodex E).
PF6
PF6
Determined by chiral GC analysis (Lipodex E).
N
N
N
N
PF6
PF6
PF6
N
N
N
N
N
N
HO
HO
7a
7b
HO
HO
HO
6m
6k
6l
PF6
PF6
N
N
N
N
PF6
PF6
PF6
HO
HO
N
N
N
N
N
N
7d
7c
HO
HO
HO
6p
6o
6n
(Table 7, entries 3–6). Moreover, the presence of a sup-
plementary stereogenic center in the benzyl unit (salts 6n
and 6o, derived from D or L enantiopure a-methyl ben-
zylamine) does not allow an increase of the enantiose-
lectivity. To verify if the match/mismatch effect was
responsible for this, we performed the conjugate addi-
tion with the alkoxy-NHC 6p, which possess only one
stereogenic center derived from the benzyl unit. In this
case, no enantioselectivity was observed.
To complete the study on the backbone imidazolin-
ium salt, we tested the alkoxy-NHC ligands 7 bearing
a second stereogenic center on the alkoxy side chain (Ta-
ble 8). The significantly lower enantioselectivity of the
NHCs 7a–c (entries 1–3) derived respectively, from nor-
ephedrine, 1,2-diphenylethanol and amino-isoborneol,
indicates that the presence of a substituent near the che-
lating function induces a dramatic steric hindrance for
the enantiocontrol of the addition. To improve this,
we synthesized the mono-alkyl substituted 1-phenyl-1-
hydroxylmethylene imidazolinium salt 7d and obtained
a lower enantioselectivity (17% ee, entry 4) than with
salts 7a and 7b. This last result is completely in accor-
dance with the one reported by Arnold et al. [13] with
the alkoxy-NHC copper (II) complex 7 (Fig. 2) bearing
a 1-tert-butyl substituent in alkoxy side chain (51% ee
have been observed at ꢀ20 °C).
In summary, in the area design of the alkoxy-NHCs,
the nature of the chiral center is not truly significant for
the enantioselectivity of the copper-catalyzed conjugate
addition. Nevertheless, two points are crucial that lead
to high stereoselectivities: firstly, because the design of
these ligands is not based on C2-symmetry, a mesityl
unit is recommended, hindering one side of the ligand
and thereby disfavoring approach of the substrate from
3
PF6
PF6
PF6
N
N
N
N
N
N
2
1
HO
HO
HO
5b
47% ee
6d
7d
17% ee
86% ee
Fig. 7.

5246
H. Clavier et al. / Journal of Organometallic Chemistry 690 (2005) 5237–5254
this direction. Secondly, as shown in Fig. 7, the
stereogenic center in the alkoxymethylene side chain
must be placed absolutely in the C2 position, near the
NHC backbone.
the more steric bulky diisopropylzinc with complete con-
version after 1 h and lower stereoselectivity up to 79% ee
(entry 2). With the more hindered cyclic enone, such as
4,4-dimethylcyclohexenone (entry 3), 12 h of reaction
are necessary to obtain a total conversion with diethyl-
zinc at room temperature. However, the best enantiose-
lectivity is thus observed in this case, reaching 93% ee.
The use of Cu(eaa)2 allowed to perform the addition
within 4 h, but the stereoselectivity was truly affected
(55% ee, entry 3). Curiously, the reaction time can be
diminished to 2 h with ligand 6e using similar conditions
without loss of enantioselectivity (94%, entry 7). More-
over, when ligand 6f is used, the reactivity is preserved
but the enantioselectivity decreased to 85% ee (entry
8). An adequate explanation for this ambiguous rela-
tionship between reactivity and enantioselectivity is
not obvious. As recently emphasized by Hoveyda [18],
it is impossible to generate a general catalyst for all
5. Enantioselective Cu-catalyzed conjugate addition of
dialkylzinc to cyclic enones at ambient temperature
With the imidazolinium salts 6a, 6f and 6e identified
as the most efficient alkoxy-NHCs ligands for the enan-
tioselective copper conjugate addition, we undertook a
survey of the reaction scope involving several cyclic en-
ones and dialkylzinc (Table 9). Remarkably, alkylation
with the dimethylzinc, which proved to give poor reactive
performance in conjugate addition, lead to the desired 3-
methylcyclohexanone in quantitative yield after 3 h and
88% ee (entry 1). 2-Cyclohexenone may be alkylated with
Table 9
Enantioselective Copper-catalyzed conjugate addition of dialkylzinc to a series of cyclic enones
Product
Alkoxy-NHC
Alkylzinc
Copper salt
Time(h) Conv.(%)a
ee (%)b
88c
Substrat
Entry
1
O
Me2Zn
Cu(OTf)2
3
1
>99
>99
O
Me
O
Cu(OTf)2
2
79
iPr2Zn
iPr
O
O
O
O
Cu(OTf)2
Cu(eaa)2
12
4
>99
>99
93
55
3
4
Et2Zn
Et2Zn
PF6
N
N
Et
HO
6a
Cu(OTf)2
0.5
>99
90
Et
O
O
O
O
Cu(OTf)2
Cu(OTf)2
O
Et2Zn
1
1
>99
>99
72
53
5
6
Et
O
O
iPr2Zn
iPr
O
PF6
N
N
Et2Zn
Cu(OTf)2
7
2
>99
94
Et
Et
HO
6e
O
O
O
O
>99
>99
85
86
2
2
Cu(OTf)2
Cu(OTf)2
Et2Zn
Et2Zn
8
9
PF6
N
N
HO
6f
Et
a Determined by GC analysis. b Determined by chiral GC analysis (Lipodex E). c Determined by chiral GC analysis (Chiraldex BTA).

H. Clavier et al. / Journal of Organometallic Chemistry 690 (2005) 5237–5254
5247
Table 10
Enantioselective Cu-catalyzed conjugate addition of various Michael acceptors with alkoxy-NHC 6a and 7d (copper salt: Cu(OTf)2)
Entry
Alkoxy-NHC
Product
Time(h)
Conv.(%)a
ee (%)b
Substrat
Et
O
O
1
1
>99
34
Ph
Ph
Et
O
O
2
3
1
1
1
>99
>99
>99
11
5
C5H11
C5H11
PF6
O
Et
O
N
N
HO
6a
NO2
Et
NO2
4
21
Et
5
6
NO2
O
1
1
>99
>99
37
27
NO2
O
Ph
Ph
Ph
Ph
Et
PF6
N
N
O
Et
O
HO
1
>99
16
7
7d
a Determined by GC analysis. b Determined by chiral GC analysis (Lipodex E).
conjugate addition and more difficult to anticipate the
reactivity or the stereoselectivity.
of the addition. 34% ee is observed in the best case with
acyclic enones (entry 1). Similarly enantioselectivities
were observed for the nitroalkene family, with which
no value over 37% ee obtained (entry 5). We have also
verified if the 1-substituted alkoxy-NHC 7d bearing
the stereogenic center directly on the hydroxyl group al-
lowed to give higher ees. However, poor stereoselectivi-
ties were also observed in this case. These last results
show clearly the scope of use of the alkoxy-NHC family
in the copper-catalyzed conjugate addition. Neverthe-
less, at this time, no chiral NHC, including these re-
cently developed, has been reported as efficient ligands
for this class of enones. As enantioselectivities of over
95% were obtained with the most efficient Cu-phos-
phine-based systems reported by Hoveyda [21] or Leigh-
ton [19], development of new efficient chiral NHCs for
acyclic substrates continues to be a challenge.
Addition of Et2Zn on 2-cycloheptenone with alkoxy-
NHC 6a is achieved in 1 h at room temperature with
90% ee (entry 4). Under similar conditions, 86% ee is ob-
served with ligand 6f (entry 9). We have then tried to
alkylate the a,b-unsaturated lactone with Et2Zn (entry
5) and obtained total conversion after 1 h. However,
moderate enantioselectivity was observed (72% ee). Fi-
nally, in the case of 2-cyclopentenone, alkylation with
the diisopropylzinc gave a low stereoselectivity up to 53%.
6. Enantioselective Cu-catalyzed conjugate addition of
diethylzinc to other Michael acceptors
In order to generalize the use of our alkoxy-NHC 6a
in the enantioselective conjugate addition, we next
screened several other Michael acceptors with acyclic
enones [20] and nitroalkenes [21,22]. These substrates,
which are known to be problematic, were tested under
identical conditions at ambient temperature. The results
are summarized in Table 10.
Whereas the high reactivity of the catalytic system is
conserved for all substrates (complete conversion within
1 h), we have been disappointed in the stereoselectivity
7. Conclusion
We have reported the development of a new class of
chiral alkoxy-NHC ligands readily accessible in a five-
step procedure from b-aminoalcohols. A small library
of these chelating NHCs has been achieved by a modular
synthesis allowing the design of the structure backbone.
High reactivity and enantioselectivity were obtained at

5248
H. Clavier et al. / Journal of Organometallic Chemistry 690 (2005) 5237–5254
PF6
PF6
N
1) MsCl, Et3N
CH2Cl2, 1h, RT
N
N
N
Mes
Mes
2) MePhNa,
S
THF, 0.5h, RT
HO
6a
70%
14
Scheme 9.
room temperature in the copper-conjugate addition of
dialkylzinc to cyclic enones with low loading of catalyst
(up to 0.10 mol%). Future works will focus on the devel-
opment of new bidentate NHCs bearing other chelating
groups (such as S-Ar or PPh2), which could be synthe-
sized directly from the alkoxy-imidazolinium salts in
two steps (Scheme 9). These new ligands are currently
being examined in other asymmetric catalytic transfor-
mations, particularly in the allylic alkylation and the
Heck coupling reaction. These results will be reported
in due course.
8.2. Materials
All non-aqueous reactions were performed under an
argon atmosphere using oven-dried glassware. Toluene
and trimethylorthoformate were distilled from sodium
metal under nitrogen. Tetrahydrofuran and diethyl ether
were distilled from sodium metal/benzophenone ketyl
under nitrogen. Dichloromethane, methanol, triethyl-
amine and pyridine were distilled from calcium hydride
under nitrogen. The 2 N anhydrous HCl/solution was
prepared by addition, at 0 °C, of acetyl chloride to dry
methanol. All others chemical reagents and solvents
were obtained from commercial sources and used with-
out further purification. Analytical TLC were performed
on Merck silica gel 60F254 plates, and visualized under
UV-light. Chromatographic purifications were per-
formed on a column with 230–400 mesh silica gel
(Merck 9385) using the indicated solvent system.
8. Experimental
8.1. General
1H (400 MHz), 13C (100 MHz), 31P (162 MHz) and 19
F
(376.5 MHz) NMR spectra were recorded on a Bruker
ARX400 spectrometer with complete proton decoupling
1
for nucleus other than H. Chemical shifts are reported
8.3. Typical procedure for the synthesis of oxalamic acid
ethyl ester (9)
in ppm from tetramethylsilane with the solvent resonance
as the internal standard (CDCl3, 1H: d 7.27 ppm, 13C: d
77.0 ppm and (CD3)2CO: 1H: d 2.05 ppm, 13C: d
205.1 ppm). Data are reported as follows: chemical shift
d in ppm, multiplicity (s = singlet, d = doublet, t = triplet,
q = quadruplet, sept = septuplet, m = multiplet), cou-
pling constants (Hz), integration and attribution. High-
resolution mass spectra (HRMS) were recorded at the
To a solution of aniline or amine (10 mmol) and pyr-
idine (970 lL, 12 mmol, 1.2 equiv.) in dry DCM
(10 mL) was added dropwise ethyl oxalyl chloride 8
(1.33 mL, 12 mmol, 1.2 equiv.) at 0 °C. The resulting
solution was stirred at room temperature overnight.
The mixture was diluted with ethyl acetate, washed suc-
cessively with 1 N HCl (two times), saturated NaHCO3
solution and brine. The organic layer was dried over
MgSO4, filtered and concentrated in vacuo to afford
the desired product 9.
´
Centre Regional de Mesures Physiques de lꢀOuest
´
(CRMPO), Universite de Rennes 1 on a Micromass
ZABSpecTOF instrument. Melting points were measured
on a heating microscope Reichert and were uncorrected.
Optical rotations were recorded using a polarimeter
Perkin–Elmer 341. Elemental analysis was performed at
Service de microanalyse I.C.S.N. - C.N.R.S. 91198 Gif
sur Yvette, France. The conversions of conjugate addi-
tions were measured using gas chromatography (capillary
column – HP-1, 0.25 lm, 30 m, 0.32 mm) with cyclodode-
cane as internal standard and enantiomeric excesses were
calculated using chiral gas chromatography (capillary
column – Lipodex E, 0.2 lm, 25 m, 0.25 mm). The prod-
ucts of conjugate additions were identified according to
the literature.
8.4. Typical procedure for the synthesis of oxalamides
(10)
A
solution of 9 (2 mmol) and aminoalcohol
(2.5 mmol, 1.25 equiv.) in dry DCM (5 mL) was refluxed
overnight. The mixture was diluted with ethyl acetate,
washed with 1 N HCl and brine. The organic layer
was dried over MgSO4, filtered and concentrated in va-
cuo to afford the oxalamide 10.

H. Clavier et al. / Journal of Organometallic Chemistry 690 (2005) 5237–5254
5249
3
˚
8.5. Typical procedure for the synthesis of diamines (11)
V = 2090.96(14) A , Z = 4, Dx = 1.380 Mg mꢀ3, k(Mo
Ka) = 0.71073 A, l = 1.94 cmꢀ1
,
F(000) = 912, T =
˚
The oxalamide 10 (2 mmol) was added to a suspension
of LiAlH4 (300 mg, 8 mmol, 4 equiv.) in dry THF
(10 mL) at 0 °C. The reaction mixture was refluxed over-
night. After cooling to 0 °C, water and 15% NaOH solu-
tion were added dropwise and the resulting solid material
was removed by filtration under a celite bed. The filtrate
was concentrated in vacuo to afford the diamine 11.
120 K. The sample (0.42 0.32 0.30 mm) is studied on
* *
a NONIUS Kappa CCD with graphite monochroma-
tized Mo Ka radiation. The cell parameters are obtained
with Denzo and Scalepack [23] with 10 frames (psi rota-
tion: 1° per frame). The data collection [24] (2hmax = 54°,
97 frames via 2.0° omega rotation and 20 s per frame,
range HKL: H, ꢀ10,10; K, ꢀ17,17; L ꢀ23,24) gives
15,689 reflections. The data reduction with Denzo and
Scalepack [23] leads to 4787 independent reflections
from which 3877 with I > 2.0r(I). The structure was
solved with SIR-97 [25] which reveals the non hydrogen
atoms of the molecule. After anisotropic refinement,
many hydrogen atoms may be found with a Fourier
Difference. The whole structure was refined with
SHELXL-97 [26] by the full-matrix least-square techniques
(use of F2 magnitude; x, y, z, bij for P, F, N, O and
C atoms, x, y, z in riding mode for H atoms; 245 vari-
ables and 3877 observations with I > 2.0r(I); calc.
8.6. Typical procedure for the synthesis of imidazolinium
salts (6, 7 and 13)
To a solution of diamine 11 (2 mmol) in dry diethyl
ether (10 mL) was added dropwise, at 0 °C, a 2 N anhy-
drous HCl/methanol solution (1 mL, 2 mmol, 1 equiv.).
A white precipitate was immediately formed. After
10 min. stirring, the solvent was evaporated to afford
the corresponding chlorhydrate of diamine in quantita-
tive yield. The salt was dissolved in dry toluene
(6 mL), trimethylorthoformate (1.1 mL, 10 mmol, 5
equiv.) was added and the mixture was heated to
90 °C for 12 h. After cooling to room temperature, the
solvents were evaporated in vacuo, the imidazolinium
chloride was dissolved in distilled water (20 mL). The
aqueous layer was washed with ethyl acetate before add-
ing KPF6 (740 mg, 4 mmol, 2 equiv.) at room tempera-
ture. A white precipitate was rapidly formed. After 2 h
stirring, the imidazolinium salt was extracted with
dichloromethane (20 mL). The organic layer was
washed with brine, dried over MgSO4 and concentrated
in vacuo to afford the desired product.
2
w ¼ 1=½r2ðF o2Þ þ ð0.105PÞ þ 0.73Pꢁ where P ¼ ðF 2oþ
2F 2cÞ=3 with the resulting R = 0.057, Rw = 0.149 and
Sw = 1.010, Dq < 0.35 e Aꢀ3).
˚
8.6.2. 1-((1S)-1-Hydroxymethyl-2,2-dimethyl-propyl)-3-
(2,4,6-trimethyl-phenyl)-4,5-dihydro-3H-imidazol-1-ium
hexafluorophosphate (6b)
White solid (573 mg, 66%); m.p. = 155 °C (AcOEt/
20
pentane). ½aꢁ ¼ þ27.0 (c = 1 in chloroform). 1H
D
NMR (400 MHz, CD3COCD3): d = 8.69 (s, 1H, CH),
7.05 (s, 2H, 2CHar), 4.71 (s, 1H, OH), 4.47 (m, 4H,
2CH2), 4.02 (m, 1H, 1CH2), 3.95 (m, 1H, 1CH2), 3.81
(dd, J(H,H) = 4 and 10 Hz, 1H, 1CH), 2.33 (s, 6H,
2CH3), 2.30 (s, 3H, 1CH3), 1.13 (s, 9H, 3CH3). 13C
NMR (100 MHz, CD3COCD3): d = 160.1, 139.7,
135.3(2C), 130.8, 129.3 (2C), 70.1, 56.5, 50.3, 48.5,
33.0, 26.3 (3C), 19.7, 16.4 (2C). 19F NMR
(376.5 MHz, CD3COCD3): d = ꢀ73.0 (d, J(F,P) =
708.2 Hz). 31P NMR (162 MHz, CD3COCD3): d =
ꢀ143.0 (sept, J(P,F) = 708.2 Hz). HRMS (FAB),
calc. for C+: C18H29N2O: 289.2280, found: 289.2273.
Structure analysis C18H29N2O, PF6, Mr = 434.40,
8.6.1. 1-((1S)-1-Hydroxymethyl-3-methyl-butyl)-3-
(2,4,6-trimethyl-phenyl)-4,5-dihydro-3H-imidazol-1-ium
hexafluorophosphate (6a)
White solid (3.3 g, 76%). (reaction realized on
20
D
10 mmol); m.p. = 173 °C (AcOEt/pentane). ½aꢁ ¼ þ5.0
1
(c = 1 in acetone). H NMR (400 MHz, CD3COCD3):
d = 8.68 (s, 1H, 1CH), 7.05 (s, 2H, 2CHar), 4.54 (s,
1H, OH), 4.43 (m, 4H, 2CH2), 4.11 (m, 1H, CH), 3.83
(m, 1H, 1CH2), 3.79 (m, 1H, 1CH2), 2.32 (s, 6H,
2CH3), 2.29 (s, 3H, 1CH3), 1.81 (m, 2H, 1CH2), 1.52
(m, 1H, 1CH), 1.01 (d, J(H,H) = 7 Hz, 3H, 1CH3),
0.98 (d, J(H,H) = 7 Hz, 3H, 1CH3). 13C NMR
(100 MHz, CD3COCD3): d = 160.9, 141.6, 137.3, 132.8
(2C), 131.2 (2C), 62.0, 61.0, 52.0, 47.4, 38.0, 26.0, 23.8,
22.7, 21.6, 18.2 (2C). 19F NMR (376.5 MHz,
CD3COCD3): d = ꢀ73.0 (d, J(F,P) = 708.2 Hz). 31P
NMR (162 MHz, CD3COCD3): d = ꢀ143.0 (sept,
J(P,F) = 708.2 Hz). HRMS (FAB), calc. for C+:
C18H29N2O: 289.2280, found: 289.2283. Elemental
Analysis: calc. for C18H29F6N2OP: C, 49.77; H, 6.73;
N, 6.45; found: C, 49.81; H, 6.97; N, 6.31. Structure
analysis: C18H29N2O, PF6, Mr = 434.40, orthorhombic,
orthorhombic, P212121, a = 8.3706(3), b = 15.6046(6),
3
c =16.1125(6) A, V = 2104.6(1) A , Z = 4, Dx =
˚
˚
ꢀ1
1.371 Mg mꢀ3, k(Mo Ka) = 0.71073 A, l = 1.93 cm
F(000) = 912, T = 120 K. The sample (0.45 0.40
,
˚
*
*
0.40 mm) is studied on a NONIUS Kappa CCD with
graphite monochromatized Mo Ka radiation. The cell
parameters are obtained with Denzo and Scalepack
[23] with 10 frames (psi rotation: 1° per frame). The data
collection [24] (2hmax = 54°, 148 frames via 2.0° omega
rotation and 20 s per frame, range HKL: H, ꢀ10,10;
K, ꢀ20,20; L ꢀ20,20) gives 24,755 reflections. The data
reduction with Denzo and Scalepack [23] leads to 4827
independent reflections from which 4191 with
I > 2.0r(I). The structure was solved with SIR-97 [25]
˚
P212121, a = 8.2453(4), b = 13.6604(5), c = 18.5642(5) A,

5250
H. Clavier et al. / Journal of Organometallic Chemistry 690 (2005) 5237–5254
which reveals the non hydrogen atoms of the molecule.
After anisotropic refinement, many hydrogen atoms
may be found with a Fourier Difference. The whole
structure was refined with SHELXL-97 [26] by the full-ma-
trix least-square techniques (use of F2 magnitude; x, y, z,
bij for P, F, N, O and C atoms, x, y, z in riding mode for
H atoms; 254 variables and 4191 observations with I >
(m, 4H, 2CH2), 3.95 (m, 2H, 1CH2), 3.17 (t,
J(H,H) = 7 Hz, 2H, 1CH2), 2.26 (s, 6H, 1CH3), 2.23
(s, 3H, 1CH3). 13C NMR (100 MHz, CD3COCD3):
d = 158.5, 139.6, 136.5, 130.5, 129.1 (2C), 128.7 (2C),
128.4 (2C), 127.7, 126.6 (2C), 62.1, 59.9, 49.9, 45.9,
33.7, 19.6, 16.1 (2C). 19F NMR (376.5 MHz,
CD3COCD3): d = ꢀ73.0 (d, J(F,P) = 708.4 Hz). 31P
NMR (162 MHz, CD3COCD3): d = ꢀ143.0 (sept,
J(P,F) = 708.4 Hz). HRMS (FAB), calc. for C+:
C21H27N2O: 323.2123, found: 323.2119.
2
2.0r(I); calc w ¼ 1=½r2ðF 2oÞ þ ð0.12PÞ þ 1.2Pꢁ where
P ¼ ðF 2o þ 2F c2Þ=3 with the resulting R = 0.068, Rw =
0.185 and Sw = 1.046, Dq < 0.57 e Aꢀ3).
˚
8.6.3. 1-((1S)-2-Hydroxy-1-methyl-ethyl)-3-(2,4,6-
trimethyl-phenyl)-4,5-dihydro-3H-imidazol-1-ium
hexafluorophosphate (6c)
8.6.6. 1-((1S)-1-Hydroxymethyl-2-methyl-propyl)-3-
(2,4,6-trimethyl-phenyl)-4,5-dihydro-3H-imidazol-1-ium
hexafluorophosphate (6f)
White solid (554 mg, 71%); m.p. = 169 °C (AcOEt/
White solid (3.07 g, 73%). (reaction realized on
20
pentane). ½aꢁ ¼ þ18.2 (c = 1 in acetone). 1H NMR
20
D
D
10 mmol); m.p. = 205 °C (AcOEt/pentane). ½aꢁ ¼ ꢀ1.2
1
(c = 1 in acetone). H NMR (400 MHz, CD3COCD3):
(400 MHz, CD3COCD3): d = 8.45 (s, 1H, 1CH), 6.91
(s, 2H, 2CHar), 4.43 (t, J(H,H) = 5 Hz, 1H, OH), 4.28
(m, 4H, 2CH2), 4.01 (m, 1H, 1CH), 3.71 (m, 1H,
1CH2), 3.61 (m, 1H, 1CH2), 2.19 (s, 3H, 1CH3), 2.16
(s, 6H, 2CH3), 1.31 (d, J(H,H) = 7 Hz, 3H, 1CH3). 13C
NMR (100 MHz, CD3COCD3): d = 158.3, 139.6,
135.5, 130.9 (2C), 129.2 (2C), 61.3, 56.3, 50.1, 45.8,
38.0, 19.7, 16.3 (2C). 19F NMR (376.5 MHz,
CD3COCD3): d = ꢀ73.0 (d, J(F,P) = 707.7 Hz). 31P
NMR (162 MHz, CD3COCD3): d = ꢀ143.0 (sept,
J(P,F) = 707.7 Hz). HRMS (FAB), calc. for C+:
C15H23N2O: 247.1810, found: 247.1808.
d = 8.64 (s, 1H, 1CH), 7.06 (s, 2H, 2CHar), 4.53 (t,
J(H,H) = 5 Hz, 1H, OH), 4.45 (m, 4H, 2CH2), 4.00 (m,
1H, 1CH2), 3.89 (m, 1H, 1CH2), 3.67 (m, 1H, 1CH),
2.33 (s, 6H, 2CH3), 2.30 (s, 3H, 1CH3), 2.19 (m, 1H,
1CH), 1.13 (d, J(H,H) = 7 Hz, 3H, 1CH3), 1.06 (d,
J(H,H) = 7 Hz, 3H, 1CH3). 13C NMR (100 MHz,
CD3COCD3): d = 159.3, 139.9, 136.5, 131.0 (2C), 129.4
(2C), 67.1, 58.5, 50.3, 46.6, 26.9, 19.9, 19.1, 18.3, 16.5
(2C). 19F NMR (376.5 MHz, CD3COCD3): d = ꢀ73.0
(d,
J(F,P) = 708.1 Hz).
31P
NMR
(162 MHz,
CD3COCD3): d = ꢀ143.0 (sept, J(P,F) = 708.1 Hz).
HRMS (FAB), calc. for C+: C17H27N2O: 275.2123,
found: 275.2120.
8.6.4. 1-((1S)-2-Hydroxy-1-phenyl-ethyl)-3-(2,4,6-
trimethyl-phenyl)-4,5-dihydro-3H-imidazol-1-ium
hexafluorophosphate (6d)
8.6.7. 1-((1S)-1-Hydroxymethyl-2-methyl-propyl)-3-
phenyl-4,5-dihydro-3H-imidazol-1-ium
hexafluorophosphate (6g)
White solid (670 mg, 74%); m.p. = 143 °C (AcOEt/
20
pentane). ½aꢁ ¼ þ22.5 (c = 1 in acetone). 1H NMR
D
(400 MHz, CD3COCD3): d = 8.83 (s, 1H, 1CH), 7.51
(m, 5H, 5CHar), 7.01 (s, 2H, 2CHar), 5.14 (dd,
J(H,H) = 4 and 9 Hz, 1H, 1CH2), 4.82 (t,
J(H,H) = 5 Hz, 1H, OH), 4.44 (m, 2H, 1CH2), 4.32
(m, 2H, 1CH2), 4.24 (m, 1H, 1CH2), 4.18 (m, 1H,
1CH2), 2.35 (s, 6H, 2CH3), 2.30 (s, 3H, 1CH3). 13C
NMR (100 MHz, CD3COCD3): d = 158.7, 139.8,
135.5, 134.0 (2C), 131.0, 129.4 (2C), 128.9 (2C), 128.8,
127.5 (2C), 64.1, 60.7, 50.3, 47.4, 19.8, 16.5 (2C).
19F NMR (376.5 MHz, CD3COCD3): d = ꢀ73.0 (d,
J(F,P) = 708.2 Hz). 31P NMR (162 MHz, CD3COCD3):
d = ꢀ143.0 (sept, J(P,F) = 708.2 Hz). HRMS (FAB),
calc. for C+: C20H25N2O: 309.1967, found: 309.1971.
White solid (506 mg, 67%); m.p. = 109 °C (AcOEt/
20
pentane). ½aꢁ ¼ ꢀ18.7 (c = 1 in acetone). 1H NMR
D
(400 MHz, CD3COCD3): d = 9.26 (s, 1H, 1CH), 7.47
(m, 4H, 4CHar), 7.36 (m, 1H, 1CHar), 4.68 (m, 2H,
1CH2), 4.49 (t, J(H,H) = 5 Hz, 1H, OH), 4.42 (m,
2H, 1CH2), 3.99 (m, 1H, 1CH2), 3.92 (m, 1H,
1CH2), 3.71 (m, 1H, 1CH), 2.22 (m, 1H, 1CH), 1.10
(d, J(H,H) = 7 Hz, 3H, 1CH3), 1.07 (d, J(H,H) = 7 Hz,
3H, 1CH3). 13C NMR (100 MHz, CD3COCD3):
d = 154.5, 136.3, 129.6 (2C), 126.4, 117.5 (2C), 67.8,
58.9, 47.9, 46.7, 26.8, 19.1, 18.3. 19F NMR
(376.5 MHz, CD3COCD3): d = ꢀ73.0 (d, J(F,P) =
708.2 Hz). 31P NMR (162 MHz, CD3COCD3):
d = ꢀ143.0 (sept, J(P,F) = 708.2 Hz). HRMS (FAB),
calc. for C+: C14H21N2O: 233.1654, found: 233.1654.
8.6.5. 1-((1S)-1-Hydroxymethyl-2-phenyl-ethyl)-3-
(2,4,6-trimethyl-phenyl)-4,5-dihydro-3H-imidazol-1-ium
hexafluorophosphate (6e)
White solid (712 mg, 76%); m.p. = 51 °C (AcOEt/
8.6.8. 3-(2,6-Diisopropyl-phenyl)-1-((1S)-1-
hydroxymethyl-2-methyl-propyl)-4,5-dihydro-3H-
imidazol-1-ium hexafluorophosphate (6h)
20
pentane). ½aꢁ ¼ ꢀ66.3 (c = 1 in acetone). 1H NMR
D
(400 MHz, CD3COCD3): d = 8.42 (s, 1H, CH), 7.40
(m, 3H, 3CHar), 7.23 (m, 2H, 2CHar), 7.00 (s, 2H,
2CHar), 4.59 (s, 1H, OH), 4.46 (m, 1H, CH), 4.44
White solid (545 mg, 59%); m.p. = 141 °C (AcOEt/
20
pentane). ½aꢁ ¼ þ15.8 (c = 1 in acetone). 1H NMR
D

H. Clavier et al. / Journal of Organometallic Chemistry 690 (2005) 5237–5254
5251
(400 MHz, CD3COCD3): d = 8.76 (s, 1H, 1CH), 7.52
(t, J(H,H) = 8 Hz, 1H, 1CHar), 7.38 (d, J(H,H) = 8 Hz,
2H, 2CHar), 4.58 (t, J(H,H) = 5 Hz, 1H, OH), 4.47 (m,
4H, 2CH2), 3.98 (m, 1H, 1CH2), 3.88 (m, 1H, 1CH2),
3.68 (m, 1H, 1CH), 3.15 (m, 2H, 2CH), 2.19 (m, 1H,
1CH), 1.31 (d, J(H,H) = 7 Hz, 6H, 2CH3), 1.22 (d,
J(H,H) = 7 Hz, 6H, 2CH3), 1.12 (d, J(H,H) = 7 Hz,
3H, 1CH3), 1.08 (d, J(H,H) = 7 Hz, 3H, 1CH3). 13C
NMR (100 MHz, CD3COCD3): d = 159.1, 146.5
(2C), 130.6, 130.2, 124.4 (2C), 67.0, 58.0, 52.7, 46.4,
26.7, 25.0 (2C), 23.8 (2C), 22.8 (2C), 18.8, 18.1. 19F
NMR (376.5 MHz, CD3COCD3): d = ꢀ73.0 (d,
J(F,P) = 708.1 Hz). 31P NMR (162 MHz, CD3COCD3):
d = ꢀ143.0 (sept, J(P,F) = 708.1 Hz). HRMS (FAB),
calc. for C+: C20H33N2O: 317.2593, found: 317.2594.
8.6.11. 3-tert-Butyl-1-((1S)-1-hydroxymethyl-2-methyl-
propyl)-4,5-dihydro-3H-imidazol-1-ium
hexafluorophosphate (6k)
White solid (480 mg, 67%); m.p. = 93 °C (AcOEt/
20
pentane). ½aꢁ ¼ ꢀ6.8 (c = 1 in acetone). 1H NMR
D
(400 MHz, CD3COCD3): d = 8.46 (s, 1H, 1CH), 4.35
(s, 1H, OH), 4.20 (m, 2H, 1CH2), 4.12 (m, 2H, 1CH2),
3.87 (m, 1H, 1CH2), 3.75 (m, 1H, 1CH2), 3.50 (m, 1H,
1CH), 2.04 (m, 1H, CH), 1.48 (s, 9H, 3CH3), 1.00 (d,
J(H,H) = 7 Hz, 3H, 1CH3), 0.98 (d, J(H,H) = 7 Hz,
3H, 1CH3). 13C NMR (100 MHz, CD3COCD3):
d = 155.8, 66.6, 58.7, 56.0, 45.3, 44.8, 26.8 (3C), 26.6,
19.0, 18.2. 19F NMR (376.5 MHz, CD3COCD3):
d = ꢀ73.0 (d, J(F,P) = 708.1 Hz). 31P NMR (162 MHz,
CD3COCD3): d = ꢀ143.0 (sept, J(P,F) = 708.1 Hz).
HRMS (FAB), calc. for C+: C12H25N2O: 213.1967,
found: 213.1961.
8.6.9. 3-(2-tert-Butyl-phenyl)-1-((1S)-1-hydroxymethyl-
2-methyl-propyl)-4,5-dihydro-3H-imidazol-1-ium
hexafluorophosphate (6i)
8.6.12. 3-Butyl-1-((1S)-1-hydroxymethyl-2-methyl-
propyl)-4,5-dihydro-3H-imidazol-1-ium
hexafluorophosphate (6l)
White solid (469 mg, 54%); m.p. = 165 °C (AcOEt/
20
pentane). ½aꢁ ¼ ꢀ4.9 (c = 1 in acetone). 1H NMR
20
D
Limpid oil (514 mg, 72%). ½aꢁ ¼ ꢀ3.7 (c = 1 in
D
(400 MHz, CD3COCD3): d = 8.71 (s, 1H, 1CH), 7.68
(dd, J(H,H) = 1 and 8 Hz, 1H, 1CHar), 7.50 (m, 2H,
2CHar), 7.41 (td, J(H,H) = 1 and 8 Hz, 1H, 1CHar),
4.56 (m, 3H, OH and 1CH2), 4.37 (m, 2H, 1CH2),
3.99 (m, 1H, 1CH2), 3.89 (m, 1H, 1CH2), 3.70 (m,
1H, 1CH), 2.18 (m, 1H, 1CH), 1.46 (s, 9H, 3CH3),
1.13 (d, J(H,H) = 7 Hz, 3H, 1CH3), 1.05 (d,
J(H,H) = 7 Hz, 3H, 1CH3). 13C NMR (100 MHz,
CD3COCD3): d = 159.7, 147.3, 134.6, 130.2, 129.6,
128.6, 127.6, 67.2, 58.2, 54.4, 46.4, 35.3, 31.0 (3C),
25.1, 19.0, 18.1. 19F NMR (376.5 MHz, CD3COCD3):
d = ꢀ73.0 (d, J(F,P) = 708.6 Hz). 31P NMR (162 MHz,
CD3COCD3): d = ꢀ143.0 (sept, J(P,F) = 708.6 Hz).
HRMS (FAB), calc. for C+: C18H29N2O: 298.2280,
found: 298.2283.
1
acetone). H NMR (400 MHz, CD3COCD3): d = 8.42
(s, 1H, 1CH), 4.37 (t, J(H,H) = 5 Hz, 1H, OH), 4.13
(m, 4H, 2CH2), 3.87 (m, 1H, 1CH2), 3.78 (m, 1H,
1CH2), 3.63 (t, J(H,H) = 7 Hz, 2H, 1CH2), 3.46 (m,
1H, 1CH), 2.04 (m, 1H, CH), 1.71 (m, 2H, 1CH2),
1.37 (m, 2H, 1CH2), 1.01 (d, J(H,H) = 7 Hz, 3H,
1CH3), 0.99 (d, J(H,H) = 7 Hz, 3H, 1CH3), 0.93 (t,
J(H,H) = 7 Hz, 3H, 1CH3). 13C NMR (100 MHz,
CD3COCD3): d = 157.7, 66.4, 58.6, 47.5, 47.3, 45.9,
29.1, 28.5, 26.8, 18.8, 18.1, 12.5. 19F NMR
(376.5 MHz, CD3COCD3): d = ꢀ73.0 (d, J(F,P) =
708.2 Hz). 31P NMR (162 MHz, CD3COCD3):
d = ꢀ143.0 (sept, J(P,F) = 708.2 Hz). HRMS (FAB),
calc. for C+: C12H25N2O: 213.1967, found:
213.1964.
8.6.10. 1-((1S)-1-Hydroxymethyl-2-methyl-propyl)-3-
(2-methylsulfanyl-phenyl)-4,5-dihydro-3H-imidazol-1-
ium hexafluorophosphate (6j)
8.6.13. 3-Benzyl-1-((1S)-1-hydroxymethyl-2-methyl-
propyl)-4,5-dihydro-3H-imidazol-1-ium
hexafluorophosphate (6m)
20
Limpid oil (280 mg, 33%). ½aꢁ ¼ þ4.7 (c = 1 in ace-
White solid (555 mg, 71%); m.p. = 49 °C (AcOEt/
D
20
1
tone). H NMR (400 MHz, CD3COCD3): d = 8.84 (s,
pentane). ½aꢁ ¼ ꢀ6.7 (c = 1 in acetone). 1H NMR
D
1H, 1CH), 7.54 (m, 3H, 3CHar), 7.35 (m, 1H, 1CHar),
4.68 (m, 1H, OH), 4.57 (m, 2H, 1CH2), 4.41 (m, 2H,
1CH2), 4.14 (m, 1H, 1CH), 3.81 (m, 2H, 1CH2), 2.58
(s, 3H, 1CH3), 2.82 (m, 2H, 1CH2), 1.53 (m, 1H,
1CH), 1.02 (d, J(H,H) = 7 Hz, 3H, 1CH3), 0.99 (d,
J(H,H) = 7 Hz, 3H, 1CH3). 13C NMR (100 MHz,
CD3COCD3): d = 158.6, 136.0, 133.9, 130.2, 127.6,
126.7, 126.2, 60.6, 59.6, 50.8, 46.1, 36.3, 24.1, 22.1,
20.9, 14.5. 19F NMR (376.5 MHz, CD3COCD3): d =
ꢀ73.0 (d, J(F,P) = 708.0 Hz). 31P NMR (162 MHz,
CD3COCD3): d = ꢀ143.0 (sept, J(P,F) = 708.0 Hz).
HRMS (FAB), calc. for C+: C16H25N2OS: 293.1688,
found: 293.1689.
(400 MHz, CD3COCD3): d = 8.62 (s, 1H, 1CH), 7.44
(m, 5H, 5CHar), 4.84 (d, J(H,H) = 7 Hz, 2H, 1CH2),
4.44 (t, J(H,H) = 5 Hz, 1H, OH), 4.15 (m, 2H, 1CH2),
4.00 (m, 2H, 1CH2), 3.89 (m, 1H, 1CH2), 3.78 (m, 1H,
1CH2), 3.52 (m, 1H, 1CH), 2.05 (m, 1H, CH), 1.03 (d,
J(H,H) = 7 Hz, 3H, 1CH3), 1.01 (d, J(H,H) = 7 Hz,
3H, 1CH3). 13C NMR (100 MHz, CD3COCD3):
d = 157.7, 66.4, 58.6, 47.5, 47.3, 45.9, 29.1, 28.5, 26.8,
26.6, 18.8, 18.1, 12.5. 19F NMR (376.5 MHz,
CD3COCD3): d = ꢀ73.0 (d, J(F,P) = 708.2 Hz). 31P
NMR (162 MHz, CD3COCD3): d = ꢀ143.0 (sept,
J(P,F) = 708.2 Hz). HRMS (FAB), calc. for C+:
C15H23N2O: 247.1810, found: 247.1810.

5252
H. Clavier et al. / Journal of Organometallic Chemistry 690 (2005) 5237–5254
8.6.14. 1-((1S)-1-Hydroxymethyl-2-methyl-propyl)-3-
((1S)-1-phenyl-ethyl)-4,5-dihydro-3H-imidazol-1-ium
hexafluorophosphate (6n)
J(P,F) = 708.2 Hz). HRMS (FAB), calc. for C+:
C130H19N2O: 219.1497, found: 219.1492.
White solid (650 mg, 80%); m.p. = 93 °C (AcOEt/
8.6.17. 1-((1R,2S)-2-Hydroxy-1-methyl-2-phenyl-
ethyl)-3-(2,4,6-trimethyl-phenyl)-4,5-dihydro-3H-
imidazol-1-ium hexafluorophosphate (7a)
20
pentane). ½aꢁ ¼ ꢀ7.5 (c = 1 in acetone). 1H NMR
D
(400 MHz, CD3COCD3): d = 8.67 (s, 1H, 1CH), 7.42
(m, 5H, 5CHar), 5.02 (q, J(H,H) = 7 Hz, 1H, 1CH),
4.57 (t, J(H,H) = 5 Hz, 1H, OH), 4.12 (m, 2H, 1CH2),
3.97 (m, 2H, 1CH2), 3.89 (m, 1H, 1CH2), 3.78 (m, 1H,
1CH2), 3.56 (m, 1H, 1CH), 2.07 (m, 1H, CH), 1.78 (d,
J(H,H) = 7 Hz, 3H, 1CH3), 1.02 (d, J(H,H) = 7 Hz,
3H, 1CH3), 1.00 (d, J(H,H) = 7 Hz, 3H, 1CH3). 13C
NMR (100 MHz, CD3COCD3): d = 156.6, 138.2, 128.7
(2C), 128.2, 126.3 (2 C), 66.6, 58.4, 57.4, 46.4, 45.4,
26.6, 18.8, 18.6, 18.2. 19F NMR (376.5 MHz,
CD3COCD3): d = ꢀ73.0 (d, J(F,P) = 708.1 Hz). 31P
NMR (162 MHz, CD3COCD3): d = ꢀ143.0 (sept,
J(P,F) = 708.1 Hz). HRMS (FAB), calc. for C+:
C16H25N2O: 261.1967, found: 261.1966.
White solid (487 mg, 52%); m.p. = 165 °C (AcOEt/
20
pentane). ½aꢁ ¼ ꢀ21.0 (c = 1 in acetone). 1H NMR
D
(400 MHz, CD3COCD3): d = 8.50 (s, 1H, 1CH), 7.55
(m, 2H, 2CHar), 7.42 (m, 2H, 2CHar), 7.34 (m, 1H,
1CHar), 7.02 (s, 2H, 2CHar), 5.25 (d, J(H,H) = 4 Hz,
1H, OH), 5.18 (t, J(H,H) = 4 Hz, 1H, 1CH), 4.60 (m,
1H, 1CH), 4.32 (m, 4H, 2CH2), 2.28 (s, 6H, 2CH3),
2.20 (s, 3H, 1CH3), 1.44 (d, J(H,H) = 7 Hz, 3H,
1CH3). 13C NMR (100 MHz, CD3COCD3): d = 158.0,
140.6, 139.7, 135.4, 130.9 (2C), 129.3 (2C) 128.2 (2C),
127.7, 126.1 (2C), 73.1, 59.7, 50.2, 48.2, 19.8, 16.4
(2C), 12.4. 19F NMR (376.5 MHz, CD3COCD3):
d = ꢀ73.0 (d, J(F,P) = 708.0 Hz). 31P NMR (162 MHz,
CD3COCD3): d = ꢀ143.0 (sept, J(P,F) = 708.0 Hz).
HRMS (FAB), calc. for C+: C21H27N2O: 321.2123,
found: 321.2119.
8.6.15. 1-((1S)-1-Hydroxymethyl-2-methyl-propyl)-3-
((1R)-1-phenyl-ethyl)-4,5-dihydro-3H-imidazol-1-ium
hexafluorophosphate (6o)
White solid (674 mg, 83%); m.p. = 81 °C (AcOEt/
8.6.18. 1-((1S,2R)-2-Hydroxy-1,2-diphenyl-ethyl)-3-
(2,4,6-trimethyl-phenyl)-4,5-dihydro-3H-imidazol-1-ium
hexafluorophosphate (7b)
20
pentane). ½aꢁ ¼ þ0.6 (c = 1 in acetone). 1H NMR
D
(400 MHz, CD3COCD3): d = 8.64 (s, 1H, 1CH), 7.44
(m, 5H, 5CHar), 5.06 (q, J(H,H) = 7 Hz, 1H, 1CH),
4.41 (t, J(H,H) = 5 Hz, 1H, OH), 4.12 (m, 2H, 1CH2),
3.98 (m, 2H, 1CH2), 3.89 (m, 1H, 1CH2), 3.78 (m, 1H,
1CH2), 3.54 (m, 1H, 1CH), 2.06 (m, 1H, CH), 1.78 (d,
J(H,H) = 7 Hz, 3H, 1CH3), 1.02 (d, J(H,H) = 7 Hz,
3H, 1CH3), 0.99 (d, J(H,H) = 7 Hz, 3H, 1CH3). 13C
NMR (100 MHz, CD3COCD3): d = 156.6, 138.1, 128.8
(2C), 128.3, 126.4 (2C), 66.7, 58.6, 57.3, 46.3, 45.5,
26.6, 18.9, 18.6, 18.2. 19F NMR (376.5 MHz,
CD3COCD3): d = ꢀ73.0 (d, J(F,P) = 708.2 Hz). 31P
NMR (162 MHz, CD3COCD3): d = ꢀ143.0 (sept,
J(P,F) = 708.2 Hz). HRMS (FAB), calc. for C+:
C16H25N2O: 261.1967, found: 261.1970.
White solid (487 mg, 52%); m.p. = 148 °C (AcOEt/
20
pentane). ½aꢁ ¼ ꢀ10.0 (c = 1 in acetone). 1H NMR
D
(400 MHz, CD3COCD3): d = 8.81 (s, 1H, 1CH), 7.46
(m, 10H, 10CHar), 7.04 (s, 2H, 2CHar), 5.78 (d,
J(H,H) = 6 Hz, 1H, CH), 5.27 (d, J(H,H) = 6 Hz, 1H,
1CH), 4.99 (s, 1H, OH), 4.43 (m, 1H, 1CH2), 4.31 (m,
3H, 2CH2), 2.35 (s, 6H, 2CH3), 2.28 (s, 3H, 1CH3).
13C NMR (100 MHz, CD3COCD3): d = 158.3, 140.8,
139.8, 135.4, 133.2, 133.0 (2C), 130.8 (2C) 129.3 (2C),
129.2 128.9, 128.6 (2C), 128.2 (2C), 126.4 (2C), 72.2,
67.3, 50.1, 48.0, 19.7, 16.4 (2). 19F NMR (376.5 MHz,
CD3COCD3): d = ꢀ73.0 (d, J(F,P) = 708.0 Hz). 31P
NMR (162 MHz, CD3COCD3): d = ꢀ143.0 (sept,
J(P,F) = 708.0 Hz). HRMS (FAB), calc. for C+:
C26H29N2O: 385.2280, found: 385.2282.
8.6.16. 1-(2-Hydroxy-ethyl)-3-((1R)-1-phenyl-ethyl)-
4,5-dihydro-3H-imidazol-1-ium hexafluorophosphate
(6p)
8.6.19. 1-((2S,3R)-3-Hydroxy-4,7,7-trimethyl-
bicyclo[2.2.1]hept-2-yl)-3-(2,4,6-trimethyl-phenyl)-4,5-
dihydro-3H-imidazol-1-ium hexafluorophosphate (7c)
White solid (517 mg, 71%); m.p. = 47 °C (AcOEt/
20
pentane). ½aꢁ ¼ þ10.0 (c = 1 in acetone). 1H NMR
D
(400 MHz, CD3COCD3): d = 8.50 (s, 1H, 1CH), 7.43
(m, 5H, 5CHar), 5.02 (q, J(H,H) = 7 Hz, 1H, 1CH),
4.25 (t, J(H,H) = 5 Hz, 1H, OH), 4.16 (t, J(H,H) = 10
Hz, 2H, 1CH2), 3.93 (t, J(H,H) = 10 Hz, 2H, 1CH2),
3.82 (q, J(H,H) = 5 Hz, 1H, 1CH2), 3.78 (t,
J(H,H) = 5 Hz, 2H, 1CH2), 1.76 (d, J(H,H) = 7 Hz,
3H, 1CH3). 13C NMR (100 MHz, CD3COCD3):
d = 156.4, 138.1, 128.7 (2C), 128.2, 126.4 (2C), 57.5,
57.3, 50.0, 48.2, 46.8, 18.7. 19F NMR (376.5 MHz,
CD3COCD3): d = ꢀ73.0 (d, J(F,P) = 708.2 Hz). 31P
NMR (162 MHz, CD3COCD3): d = ꢀ143.0 (sept,
White solid (496 mg, 51%); m.p. = 217 °C (AcOEt/
20
pentane). ½aꢁ ¼ þ32.0 (c = 1 in acetone). 1H NMR
D
(400 MHz, CD3COCD3): d = 8.79 (s, 1H, 1CH), 7.04
(s, 2H, 2CHar), 5.04 (d, J(H,H) = 6 Hz, 1H, 1CH),
4.66 (m, 1H, OH), 4.36 (m, 4H, 2CH2), 4.07 (dd,
J(H,H) = 6 and 7 Hz, 1H, 1CH), 3.86 (d,
J(H,H) = 7 Hz, 1H, 1CH), 2.32 (s, 6H, 2CH3), 2.29 (s,
3H, 1CH3), 1.85 (m, 1H, 1CH2), 1.25 (s, 3H, 1CH3),
1.19 (m, 2H, 1CH2), 0.97 (s, 3H, 1CH3), 0.88 (s, 3H,
1CH3). 13C NMR (100 MHz, CD3COCD3): d = 158.3,
139.6, 135.1, 131.1 (2C), 129.2 (2C), 79.2, 65.3, 51.1,

H. Clavier et al. / Journal of Organometallic Chemistry 690 (2005) 5237–5254
5253
49.7, 49.4, 46.9, 46.7, 31.4, 26.3, 20.6, 20.1, 19.7, 16.4
(2C), 10.5. 19F NMR (376.5 MHz, CD3COCD3):
d = ꢀ73.0 (d, J(F,P) = 708.2 Hz). 31P NMR (162 MHz,
CD3COCD3): d = ꢀ143.0 (sept, J(P,F) = 708.2 Hz).
calc. for C+: C22H33N2O: 341.2593, found: 341.2602.
(100 MHz, CD3COCD3): d = 159.1, 139.8, 135.4, 130.9
(2C), 129.4 (2C), 70.3, 57.9, 56.9, 50.3, 45.7, 36.4, 24.2,
21.9, 20.9, 19.8, 16.3 (2C). 19F NMR (376.5 MHz,
CD3COCD3): d = ꢀ73.0 (d, J(F,P) = 708.0 Hz). 31P
NMR (162 MHz, CD3COCD3): d = ꢀ143.0 (sept,
J(P,F) = 708.0 Hz). HRMS (FAB), calc. for C+:
C19H31N2O: 303.2436, found: 303.2432.
8.6.20. 1-((2R)-2-Hydroxy-2-phenyl-ethyl)-3-(2,4,6-
trimethyl-phenyl)-4,5-dihydro-3H-imidazol-1-ium
hexafluorophosphate (7d)
8.6.23. 1-[(1S)-1-(tert-Butyl-dimethyl-
White solid (754 mg, 83%); m.p. = 192 °C (AcOEt/
silanyloxymethyl)-3-methyl-butyl]-3-(2,4,6-trimethyl-
phenyl)-4,5-dihydro-3H-imidazol-1-ium
hexafluorophosphate (12)
20
pentane). ½aꢁ ¼ þ65.2 (c = 1 in acetone). 1H NMR
D
(400 MHz, CD3COCD3): d = 8.50 (s, 1H, 1CH), 7.52
(d, 2H, J(H,H) = 7 Hz, 2CHar), 7.42 (dt, 2H,
J(H,H) = 2 and 7 Hz, 2CHar), 7.34 (m, 1H, 1CHar),
7.04 (s, 2H, 2CHar), 5.29 (m, 2H, 1CH2 EtOH), 4.60
(m, 1H, 1CH2), 4.39 (m, 3H, 2CH2), 3.97 (t,
J(H,H) = 4 Hz, 2H, 1CH2), 2.29 (s, 6H, 2CH3), 2.28 (s,
3H, 1CH3). 13C NMR (100 MHz, CD3COCD3):
d = 159.4, 141.1, 139.8, 135.5, 131.2 (2C), 129.4 (2C)
128.4, 127.7 (2C), 125.7 (2C), 69.6, 54.8, 50.7, 49.9,
19.8, 16.4 (2C). 19F NMR (376.5 MHz, CD3COCD3):
d = ꢀ73.0 (d, J(F,P) = 708.2 Hz). 31P NMR (162 MHz,
CD3COCD3): d = ꢀ143.0 (sept, J(P,F) = 708.2 Hz).
HRMS (FAB), calc. for C+: C20H25N2O: 309.1967,
found: 309.1963.
To a solution of imidazolinium salt 6a (217 mg,
0.5 mmol), triethylamine (85 lL, 0.6 mmol, 1.2 equiv.)
and DMAP (3 mg, 0.025 mmol, 0.05 equiv.) in dry
DCM (10 mL) was added tert-butyldimethylsilyl chlo-
ride (90 mg, 0.6 mmol, 1.2 equiv.). The reaction mixture
was stirred overnight at room temperature. The reaction
was then quenched with saturated NaHCO3 solution.
The organic layer was washed with distilled water
(10 mL) then brine, dried over MgSO4, filtered and con-
centrated. A purification by silica gel chromatography
dichloromethane/acetone (9/1) as eluent afford the de-
sired product as a white solid (254 mg, 0.47 mmol,
20
D
93%); m.p. = 119 °C (AcOEt/pentane). ½aꢁ ¼ þ18.8
1
(c = 1 in acetone). H NMR (400 MHz, CD3COCD3):
8.6.21. 1-((1S)-1-Methoxymethyl-3-methyl-butyl)-3-
(2,4,6-trimethyl-phenyl)-4,5-dihydro-3H-imidazol-1-ium
hexafluorophosphate (ent-6a)
d = 8.79 (s, 1H, 1CH), 7.07 (s, 2H, 2CHar), 4.46 (m,
4H, 2CH2), 4.17 (m, 1H, CH), 4.01 (m, 1H, 1CH2),
3.92 (m, 1H, 1CH2), 2.33 (s, 6H, 2CH3), 2.30 (s, 3H,
1CH3), 1.84 (m, 2H, 1CH2), 1.60 (m, 1H, 1CH), 1.02
(d, J(H,H) = 6 Hz, 3H, 1CH3), 1.00 (d, J(H,H) = 6 Hz,
3H, 1CH3), 0.92 (s, 9H, 3CH3), 0.14 (s, 6H, 2CH3). 13C
NMR (100 MHz, CD3COCD3): d = 158.8, 139.9, 135.4,
131.0 (2C), 129.5 (2C), 61.9, 59.1, 50.3, 46.2, 36.2, 25.1
(3C), 24.2, 22.0, 20.9, 19.8, 17.7 (2C), ꢀ4.3, ꢀ6.4 (2C).
19F NMR (376.5 MHz, CD3COCD3): d = ꢀ73.0 (d,
J(F,P) = 708.2 Hz). 31P NMR (162 MHz, CD3COCD3):
d = ꢀ143.0 (sept, J(P,F) = 708.2 Hz). HRMS (FAB),
calc. for C+: C24H43N2OSi: 403.3145, found: 403.3151.
White solid (599 mg, 69%); m.p. = 171 °C (AcOEt/
20
pentane). ½aꢁ ¼ ꢀ5.4 (c = 1 in acetone). 1H NMR
D
(400 MHz, CD3COCD3): d = 8.67 (s, 1H, 1CH), 7.05
(s, 2H, 2CHar), 4.54 (s, 1H, OH), 4.42 (m, 4H, 2CH2),
4.11 (m, 1H, CH), 3.82 (m, 1H, 1CH2), 3.80 (m, 1H,
1CH2), 2.32 (s, 6H, 2CH3), 2.29 (s, 3H, 1CH3), 1.82
(m, 2H, 1CH2), 1.53 (m, 1H, 1CH), 1.00 (d,
J(H,H) = 7 Hz, 3H, 1CH3), 0.97 (d, J(H,H) = 7 Hz,
3H, 1CH3). 13C NMR (100 MHz, CD3COCD3):
d = 160.9, 141.6, 137.3, 132.8 (2C), 131.2 (2C), 62.0,
61.0, 52.0, 47.4, 38.0, 26.0, 23.8, 22.7, 21.6, 18.2 (2C).
19F NMR (376.5 MHz, CD3COCD3): d = ꢀ73.0 (d,
J(F,P) = 708.2 Hz). 31P NMR (162 MHz, CD3COCD3):
d = ꢀ143.0 (sept, J(P,F) = 708.2 Hz). HRMS (FAB),
calc. for C+: C18H29N2O: 289.2280, found: 289.2282.
8.7. Typical procedure for conjugate addition to enones at
room temperature
A dry Schlenk tube was charged with copper
source (0.02 mmol, 0.02 equiv.), imidazolinium salt
(0.03 mmol, 0.03 equiv.) and cyclododecane (internal
standard) and dry solvent (3 mL) was added. The
resulting solution was cooled to 0 °C, upon which nBu-
Li (50 lL, 0.08 mmol, 0.08 equiv. as 1.6 M solution in
hexanes) was added. After stirring for 5 min, diethyl
zinc (1.5 mL, 1.5 mmol, 1.5 equiv. as 1 M solution in
hexanes) was added and the solution was warmed up
to room temperature. The mixture was stirred for
5 min before adding the substrate (1 mmol). The
advancement of the reaction was followed by gas chro-
matography or by TLC (hexane/diethyl ether 8/2). The
8.6.22. 1-((1R)-1-Hydroxymethyl-3-methyl-butyl)-3-
(2,4,6-trimethyl-phenyl)-4,5-dihydro-3H-imidazol-1-ium
hexafluorophosphate (13)
White solid (627 mg, 70%); m.p. = 139 °C (AcOEt/
20
pentane). ½aꢁ ¼ þ4.7 (c = 1 in acetone). 1H NMR
D
(400 MHz, CD3COCD3): d = 8.70 (s, 1H, 1CH), 7.08
(s, 2H, 2CHar), 4.47 (m, 4H, 2CH2), 4.26 (m, 1H, CH),
3.69 (d, J(H,H) = 6 Hz, 2H, 1CH2), 3.45 (s, 3H, 1CH3),
2.34 (s, 6H, 2CH3), 2.32 (s, 3H, 1CH3), 1.85 (m, 2H,
1CH2), 1.57 (m, 1H, 1CH), 1.02 (d, J(H,H) = 6 Hz, 3H,
1CH3), 1.00 (d, J(H,H) = 6 Hz, 3H, 1CH3). 13C NMR

5254
H. Clavier et al. / Journal of Organometallic Chemistry 690 (2005) 5237–5254
(f) M.C. Perry, X. Cui, M.T. Powell, D.-R. Hou, H.J. Reibenspies,
K. Burgess, J. Am. Chem. Soc. 125 (2003) 113;
(g) Y. Ma, C. Song, C. Ma, Q. Chai, M.B. Andrus, Angew.
Chem., Int. Ed. Engl. 42 (2003) 5871;
reaction was quenched by the addition of HCl (5 mL
of a 1 M aqueous solution). The resulting mixture
was stirred until clear and enantiomeric excess could
be measured directly using chiral gas chromatography.
The product could be extracted with diethyl ether,
washed with brine, dried over MgSO4 and the solvent
concentrated in vacuo to yield the crude product,
which could be purified by flash column chromatogra-
phy (hexane/diethyl ether 8/2) to provide the pure
product.
(h) L.G. Bonnet, R.E. Douthwaite, B.M. Kariuki, Organometal-
lics 22 (2003) 4187;
(i) L.H. Gade, V. Cesar, S. Bellemin-Laponnaz, Angew. Chem.,
Int. Ed. Engl. 43 (2004) 1014;
(j) E. Bappert, G. Helmchen, Synlett 10 (2004) 1789.
[7] (a) A.J. Arduengo III, H.V.R. Dias, J.C. Calbrese, F. Davidson,
Organometallics 12 (1993) 3405;
(b) H.G. Raubenheimer, S. Cronje, P.J. Olivier, J. Chem. Soc.
Dalton Trans. (1995) 313.
[8] P.K. Fraser, S. Woodward, Tetrahedron Lett. 42 (2001) 2747.
[9] (a) F. Guillen, C.L. Winn, A. Alexakis, Tetrahedron: Asymmetry
12 (2001) 2083;
9. Supplementary materials
(b) J. Pytkowics, S. Roland, P. Mangeney, Tetrahedron: Asym-
metry 12 (2001) 2087;
Crystallographic data for the structural have been
deposited with the Cambridge Crystallographic Data
Center, CCDC Nos. 250113 and 249960 for compounds
6a and 6b, respectively. Copies of this information may
be obtained free of charge from The Director, CCDC,
12 Union Road, Cambridge CB2 1EZ, UK (fax: +44
1223 336033; e-mail: deposit@ccdc.cam.ac.uk or http://
www.ccdc.cam.ac.uk).
(c) A. Alexakis, C.L. Winn, F. Guillen, J. Pytkowics, S. Roland,
P. Mangeney, Adv. Synth. Catal. 3 (2003) 345.
[10] For a review on enantioselective Cu-catalyzed conjugate addi-
tion, see: A. Alexakis, C. Benhaim, Eur. J. Org. Chem. (2002)
3221.
[11] S. Tominiga, Y. Oi, T. Kato, D.K. An, S. Okamoto, Tetrahedron
Lett. 45 (2004) 5585.
[12] A.O. Larsen, W. Leu, C.N. Oberhuber, J.E. Campbell, A.H.
Hoveyda, J. Am. Chem. Soc. 126 (2004) 11130.
[13] (a) P.L. Arnold, A.C. Scarisbrick, A.J. Blake, C. Wilson, Chem.
Commun. (2001) 2340;
Acknowledgments
(b) P.L. Arnold, M. Rodden, K.M. Davis, A.C. Scarisbrick, A.J.
Blake, C. Wilson, Chem. Commun. (2004) 1612.
[14] H. Clavier, L. Boulanger, N. Audic, L. Toupet, M. Mauduit,
J.-C. Guillemin, Chem. Commun. (2004) 1224.
[15] To be published separately.
This work was supported by the CNRS and the Min-
`
istere de la Recherche et de la Technologie (Grant to
H.C.). We thank Pr. A. Alexakis for helpful discussion,
for the generous gift of CuTC complex and also for the
CPG analysis of 3(R)-methylcyclohexanone.
[16] H. Clavier, L. Coutable, J.-C. Guillemin, M. Mauduit, Tetrahe-
dron: Asymmetry 16 (2005) 921.
[17] During the preparation of this paper, a similar synthesis using
aminophenol was reported: A.W. Waltman, R.H. Grubbs,
Organometallics 23 (2004) 3105.
References
[18] (a) H. Mizutani, S.J. Degrado, A.H. Hoveyda, J. Am. Chem. Soc.
124 (2002) 779;
(b) A.H. Hoveyda, A.W. Hird, M. Kacprzynski, Chem. Commun.
(2004) 1779.
[1] D. Bourissou, O. Guerret, F.P. Gabbai, G. Bertrand, Chem. Rev.
100 (2000) 39.
[19] I.J. Krauss, J.L. Leighton, Organic Lett. 5 (2003) 3201.
[20] A.P. Duncan, J.L. Leighton, Organic Lett. 6 (2004) 4117.
[21] D. Polet, A. Alexakis, Tetrahedron Lett. 46 (2005) 1529.
[2] W.A. Herrmann, Angew. Chem. Int. Ed. Engl. 41 (2002) 1291.
[3] A.J. Arduengo III, R.L. Harlow, M.A. Kline, J. Am. Chem. Soc.
113 (1991) 361.
[22] D.M. Mampreian, A.H. Hoveyda, Organic Lett.
2829.
6 (2004)
[4] For recent reviews, see: A.C. Hillier, S.M. Nolan, Platinum Met.
Rev. 46 (2002) 50.
[23] Z. Otwinowski, W. Minor, Processing of X-ray diffraction data
collected in oscillation mode, in: C.W. Carter, R.M. Sweet (Eds.),
Methods in Enzymology, Macromolecular Crystallography, Part
A, Academic Press, London, 1997, p. 307.
[5] T.M. Trnka, R.H. Grubbs, Acc. Chem. Res. 34 (2001) 18.
[6] (a) For recent reviews, see: M.C. Perry, K. Burgess, Tetrahedron:
Asymmetry 14 (2003) 951;
´
(b) V. Cesar, S. Bellemin-Laponnaz, L.H. Gade, Chem. Soc. Rev.
33 (2004) 619;
[24] Nonius, Kappa CCD Software, Nonius BV, Delft, The Nether-
lands, 1999.
(c) For the first synthesis of chiral-NHC ligands, see: A.W.
Coleman, P.B. Hitchcock, M.F. Lappert, R.K. Maskell, J.H.
Muller, J. Organomet. Chem. 296 (1985) 173;
¨
[25] A. Altomare, M.C. Burla, M. Camalli, G. Cascarano, C.
Giacovazzo, A. Guagliardi, A.G.G. Moliterni, G. Polidori, R.
Spagna, J. Appl. Crystallogr. 31 (1998) 74.
(d) M.F. Lappert, J. Organomet. Chem. 358 (1988) 185;
(e) For recent examples, see: M.T. Powell, D.-R. Hou, M.C.
Perry, X. Cui, K. Burgess, J. Am. Chem. Soc. 123 (2001) 8878;
[26] G.M. Sheldrick, SHELX-97. Program for the Refinement of Crystal
Structures, University of Go¨ttingen, Germany, 1997.