Design of chiral hydroxyalkyl- and hydroxyarylazolinium salts as new chelating diaminocarbene ligand precursors devoted to asymmetric copper-catalyzed conjugate addition

Article abstract of DOI:10.1002/ejic.200801243

The design and the synthesis of a set of new chiral hy- droxyalkyl- and hydroxyaryl-chelating diaminocarbene ligands is reported. Comparative catalytic studies show the importance of the scaffold design around the NHC unit to obtain a high enantiocontrol in Cu-catalyzed asymmetric conjugate addition (ACA). Whereas low selectivities are observed when the stereogenic centre is placed within the N-heterocyclic ring, an interesting match effect can be observed when central chirality is located within both of the two side chains, which enables up to 92 % ee in the catalysis reaction. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2009.

Full text of DOI:10.1002/ejic.200801243

FULL PAPER
DOI: 10.1002/ejic.200801243
Design of Chiral Hydroxyalkyl- and Hydroxyarylazolinium Salts as New
Chelating Diaminocarbene Ligand Precursors Devoted to Asymmetric Copper-
Catalyzed Conjugate Addition
Diane Rix,[a,b] Stéphane Labat,[a,b] Loïc Toupet,[c] Christophe Crévisy,*[a,b] and
Marc Mauduit*[a,b]
Keywords: Asymmetric catalysis / Diaminocarbene / Carbene ligands / Ligand design / Conjugate addition / Atropisomerism
The design and the synthesis of a set of new chiral hy-
droxyalkyl- and hydroxyaryl-chelating diaminocarbene li-
gands is reported. Comparative catalytic studies show the
importance of the scaffold design around the NHC unit to
obtain a high enantiocontrol in Cu-catalyzed asymmetric
heterocyclic ring, an interesting match effect can be ob-
served when central chirality is located within both of the
two side chains, which enables up to 92% ee in the catalysis
reaction.
conjugate addition (ACA). Whereas low selectivities are ob- (© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
served when the stereogenic centre is placed within the N-
Germany, 2009)
Introduction
Discovered by Öfele[1] and Wanzlick[2] in 1968 and iso-
lated in the free state by Arduengo[3] in 1991, N-heterocyclic
carbene (NHC) ligands have been widely and successfully
employed in organometallic catalysis.[4] Because of their ex-
ceptional capacity to establish strong bonds with metal cen-
ters, these ligands will likely supplant phosphorus-based li-
gands, which will allow a remarkable acceleration in a large
number of chemical transformations[4] such as olefin me-
tathesis, carbon–carbon and carbon–nitrogen cross-coup-
ling reactions, as well as hydrogenation and hydrosilylation.
Logically, the exploration of these valuable ligands for Figure 1. Hydroxyl-chelating azolium salts used in asymmetric
asymmetric catalysis has been reported.[5] Among the grow-
transformations as diaminocarbene ligand precusors.
ing number of chiral NHC ligands based on a wide range
of structures, a new promising class has recently emerged:
Pioneering in this field are Hoveyda and coworkers with
the hydroxy-chelating diaminocarbenes (Figure 1).
the chiral binaphthol-NHC precursor 1,[6] and more re-
cently with the readily available biphenol-NHC precursor
2[7] based on the 1,2-diphenylethylenediamine skeleton as a
[a] Ecole Nationale Supérieure de Chimie de Rennes, CNRS, source of chirality.
UMR 6226,
These NHC precursor ligands were successfully used in
Avenue du Général Leclerc, CS 50837, 35708 Rennes cedex 7,
asymmetric metathesis,[7,8] in enantioselective allylic substi-
France
tution[9] and also in asymmetric conjugate addition
Fax: +33-2-23238108
E-mail: marc.mauduit@esnc-rennes.fr
(ACA).[10] Another class such as the alkyloxy-NHCs de-
christophe.crevisy@esnc-rennes.fr
rived from the (tert-butyl-hydroxy-ethyl)azolium salts 3 and
4 have independently been reported by Arnold[11] and our
group.[12] Whereas the Arnold’s imidazolium salt 3 gives a
moderate selectivity in the copper-catalyzed conjugate ad-
[b] Université Européenne de Bretagne
[c] UMR CNRS 6626 GMCM, Université de Rennes 1,
Campus de Beaulieu-bât 11A, 35042 Rennes cedex, France
Supporting information for this article is available on the
WWW under http://www.eurjic.org or from the author.
Eur. J. Inorg. Chem. 2009, 1989–1999
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1989

D. Rix, S. Labat, L. Toupet, C. Crévisy, M. Mauduit
FULL PAPER
dition of diethylzinc to 2-cyclohexenone (51%ee), our hy-
droxyalkyl salt 4, containing the stereogenic centre in the
α-position, afforded higher selectivities (89% ee). Moreover,
4 is extremely suitable for the formation of all-chiral quater-
nary centres using Grignard reagents toward various 3-sub-
stituted-2-cyclohexenones (up to 96% ee)[13] or 3-allyl-sub-
stituted 2-cyclohexenones where a remarkable regiodiverg-
ent addition (1,4 vs. 1,6) occurred (up to 98% ee).[14] To
improve the potential of this promising class of chelating
NHCs, we have kept on making efforts to improve the scaf-
fold’s design, and we have attempted to compare two dif-
ferent types of NHC precursors: on one hand, the imid-
azolinium salts 5 and 8 exhibit a stereogenic centre within
the N-heterocycle, and on the other hand, the azolium salts
6–7 and 9 bear a stereogenic centre within the chelating side
chain (Figure 2). The activity and the selectivity of these
new ligands, which are readily available from chiral amino
acids or α-methylarylamines, have been evaluated in the
asymmetric copper-catalyzed conjugate addition of organo-
metallic reagents to several cyclic enones.
Results and Discussion
Synthesis of Hydroxyalkyl-Imidazolinium Salts 5
The synthetic routes to enantiopure imidazolinium salts
5 are shown in Scheme 1.[15] In the first approach, the
mixed-anhydride-mediated coupling between commercially
available Boc-protected amino acids 10 and substituted ani-
lines led to the formation of the corresponding amides. Af-
ter the removal of the carbamate by anhydrous hydrochloric
acid in methanolic solution, the amide function was re-
duced with LiAlH4 to afford, after chromatography, the di-
amines 11 in excellent yields. At this point, formation of the
diamine chlorhydrate followed by condensation of trimethyl
orthoformate in toluene resulted in the formation of
enantiopure imidazolines 12. Finally, the desired chiral hy-
droxyalkyl-imidazolinium salts 5 were obtained by an alky-
lation with 2-bromoethanol followed by an anion exchange
with KPF6 or LiNTf2. Through this methodology, six op-
tically pure hydroxyalkyl-imidazolinium salts, 5a–5f, were
isolated after purification on silica gel with moderate to
good overall yields.
Synthesis of Hydroxyalkyl-Imidazolinium Salts 6
The five-step synthetic procedure for the family of hy-
droxyalkyl salts 6 is shown in Scheme 2.[12] Primary amines
such as 1-naphthylmethylamine, (R)- or (S)-α-methyl-1-
naphthylmethylamine or (R)-indolamine were condensed
onto commercially available ethyloxalylchloride 13 to give
the corresponding oxanilic diethylesters 14. The reaction of
enantiopure (R)-leucinol or (R)-tert-leucinol with 14 in re-
fluxing dichloromethane provided the oxalamides 15 in
pure form without any purification. After reduction of the
diamide function by lithium and aluminium hydride, the
resulting diamines 16 were successively treated with a 2 
anhydrous HCl solution in methanol and then with tri-
methyl orthoformate in refluxing toluene to provide the de-
sired imidazolinium chloride salts. Finally, the chloride salts
Figure 2. Proposed design study of chelating diaminocarbene li-
gands for asymmetric conjugate addition (ACA).
Scheme 1. Synthesis of hydroxyalkyl-imidazolinium salts 5a–5f.
1990
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 1989–1999

Chiral Chelating Diaminocarbene Ligands
were treated with KPF6 to give the corresponding hexafluo- verted into their hydrochlorides before the cyclization step,
rophosphate hydroxyalkyl-imidazolinium salts 6a–6h in which was performed in the presence of trimethyl orthofor-
pure form and good overall yields ranging from 34 to 62%. mate in refluxing toluene. Finally, anionic metathesis with
potassium hexafluorophosphate led to the desired hexafluo-
rophosphate benzimidazolium salts 7a–7e in good overall
yields ranging from 17 to 38% over five steps.
Scheme 3. Synthesis of hydroxyalkyl-benzimidazolium salts 7a–7e.
Synthesis of Benzyloxy-Imidazolinium Salt 8 and (R)-α-
Methylbenzyloxy-Imidazolinium Salts 9
In comparison with the hydroxyalkyl-imidazolinium salts
5, the synthetic route to the benzyloxy-imidazolinium salts
8 was slightly adapted in order to accommodate the intro-
duction of the chelating hydroxyaryl fragment (Scheme 4).
Reductive amination performed on amine 20 in the pres-
ence of 2-hydroxy-3,5-tert-butyl-benzaldehyde followed by
reduction of the amide function gave the corresponding di-
amines 21 in good yields (84%). The cyclization process
followed by the aforementioned anionic metathesis step
yielded the expected hexafluorophosphate salts 8 in 72%
Scheme 2. Synthesis of hydroxyalkyl-imidazolinium salts 6a–6h.
yield. For the synthesis of the (R)-α-methylbenzyloxy-che-
lating imidazolinium salts 9, the synthetic route developed
for hydroxyalkyl-imidazolinium salt 6 was not suitable (see
Scheme 2). Indeed, the steps for both the reduction of oxal-
amide 15 and the methyl phenyl ether cleavage gave moder-
Synthetic Route to Hydroxyalkyl-Benzimidazolium Salts 7
As shown in Scheme 3, the synthesis of compounds 7 ate yields. Consequently, the expected salts 9 were isolated
started by reacting enantiopure β-amino alcohols 18 with in lower overall yields (20%). We therefore decided to de-
2-fluoronitrobenzene 17 in dichloromethane at reflux. After velop a more suitable synthetic route, as shown in
the reduction of the nitro group with tin in refluxing 3  Scheme 4. The condensation of various substituted anilines
hydrochloric aqueous solution, the expected monoalkylated or 1-naphthylmethylamine onto the commercial 2-bromo-
diamines 19 were isolated in moderate to good yields. Alky- ethanoyl chloride 22 afforded the 2-bromoacetamides 23.
lations of the free amine were performed by a reductive am- Alkylation of the enantiopure α-methyl-2-methoxybenzyl-
ination step using cyclohexanone, acetone or 1-naphth- amine 24 by 23 in acetonitrile over 12 h at reflux, followed
aldehyde. The resulting dialkylated diamines were then con- by methyl phenyl ether cleavage in the presence of BBr3 in
Eur. J. Inorg. Chem. 2009, 1989–1999
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1991

D. Rix, S. Labat, L. Toupet, C. Crévisy, M. Mauduit
FULL PAPER
dichloromethane at room temperature, afforded the corre- diethylzinc to 2-cyclohexenone as the model reaction
sponding 2-aminoacetamides 25 in good isolated yields af- (Scheme 5). The hydroxyalkyl-NHC–copper catalytic spe-
ter column chromatography. Reduction of amides 25 to the cies was prepared in situ by a double deprotonation of the
corresponding diamines 26 was achieved by the use of lith- hydroxyalkyl-azolium salt (3 mol-%) with n-BuLi in the
ium and aluminium hydride in refluxing tetrahydrofuran. presence of copper(II) triflate (2 mol-%) in anhydrous di-
Finally, the usual cyclization procedure led to the isolation ethyl ether. The addition of the diethylzinc leads to the for-
of the expected hexafluorophosphate imidazolinium salts 9 mation of the catalytic copper(I) species[16] that can react
in good overall yields ranging from 28 to 45% over six with the substrate.
steps.
Scheme 5. Model of asymmetric conjugate addition studied with
azolium salt NHC precursors.
Evaluation of Azolium Salt NHC Precursors 5a–5f Bearing
a Stereogenic Center on the NHC Ring
As shown in Table 1, a complete conversion occurred af-
ter 4 h at –50 °C in diethyl ether when the azolium salts 5a–
5h were used, which shows the efficiency of the NHC–Cu
catalytic entities. As expected, a very low enantioselectivity
was observed with the salt 5a bearing a methyl substituent
at the stereogenic center (entry 1, 29%), while the use of a
more sterically hindered moiety such as isopropyl 5b or
phenyl 5c resulted in an improvement of the enantiocontrol
(entries 2 and 3, 50 and 45% ee, respectively). The best
selectivity was obtained with the salt 5d containing the
more bulky tert-butyl group (entry 4, 60% ee). Interestingly,
a higher ee (entry 5, 66%) could be reached when the reac-
tion was performed at room temperature.
Table 1. Evaluation of the hydroxyalkyl-imidazolinium salts 5a–5f
bearing a stereogenic center within the N-heterocyclic ring for the
conjugate addition of diethylzinc to 2-cyclohexenone.[a]
Entry
Ligand
Temp. [°C]
Time [h][b]
ee [%][c]
1
2
3
4
5
6
7
5a
5b
5c
5d
5d
5e
5f
–50
–50
–50
–50
+20
–50
–50
4
4
4
4
1
4
4
29 (R)
50 (R)
45 (R)
60 (R)
66 (R)
41 (R)
40 (R)
[a] Conditions: 5 (3 mol-%), nBuLi (8 mol-%), Cu(OTf)2 (2 mol-
%), 2-cyclohexenone (1 mmol), Et2Zn (1.5 mmol), Et2O. [b] Time
necessary to reach complete conversion. [c] Determined by chiral
GC analysis (Lipodex E).
Scheme 4. Synthetic routes to benzyloxy-imidazolinium salts 8 and
9a–9b.
To explain these moderate enantioselectivities, we sup-
posed that the three-dimensional unit was too far from the
metallic center and was therefore unable to correctly trans-
fer the chiral information to the substrate during the ad-
dition. One alternative to improve the chiral induction
Activity and Selectivity of Hydroxy-Chelating Azolium
Salts 5–9 in Copper-Catalyzed Asymmetric Conjugate
Addition (ACA)
With these different hydroxy-chelating azolium salts 5– within this class of NHC precursors may lie in the possibil-
9 in our hands, we then evaluated both their activity and ity of locking the N substituents in fixed conformations, as
selectivity in copper-catalyzed asymmetric conjugate ad- recently reported by Grubbs and coworkers[19] (Figure 3).
dition (ACA). For this purpose, we chose the addition of Indeed, the stereogenic centers on the N heterocycle were
1992
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 1989–1999

Chiral Chelating Diaminocarbene Ligands
able to efficiently transfer the chiral information directly
from the α-position of the nitrogen atom through steric re-
pulsions with the ortho substituents in the aromatic unit.
Under these constrained structural conditions, only one
atropisomer could be formed as a result of anti relation-
ships. We therefore investigated the synthesis of some che-
lating homologues of these NHC precursors, bearing only
one chiral center at the β-position of the N-aryl substituent
instead of the usual α-position (Figure 3). However, the cru-
cial point was to prove that a similar transfer of chirality
could be reached with the azolium salt 5e bearing a 2-tert-
butylphenyl group.
Figure 4. Preliminary studies of the transfer of chiral information
from the β-position of the nitrogen through steric repulsions.
Figure 3. Proposed transfer of chirality for hydroxyalkyl-imid-
azolinium 5e.
A preliminary study of the transfer of the chiral infor-
mation from the stereogenic center at the β-position to the
aryl substituent of the nitrogen was realized on the -pro-
line scaffold.[17] As shown in Figure 4, when a small aryl
substituent such as a methyl was used, the steric hindrance
was too small to generate a preferentially locked conforma-
tion.
Therefore, the cyclization of diamine 27 gave both con-
formers 28a (syn) and 28b (anti) as a result of the rapid
internal rotation of the o-tolyl unit around the C–N axis.
We were surprised to observe these two conformers in the
same unit cell, as shown in a single-crystal diffraction
study.[18] On the other hand, in the case of diamine 29, bear-
ing the more bulky tert-butyl group, the rate of the C–N
internal rotation is significantly reduced by steric repul-
sions, which promotes the unique formation of anti-atropi-
somer 30, as expected. Once this concept was validated, we
then applied it through the formation of the hydroxyalkyl-
Figure 5. Solid-state structures of chiral imidazoline 12e and chiral
imidazolinium salt 5g obtained by methylation of imidazoline 12e
–
(PF6 counter anions have been omitted for clarity).
With this postulated atropisomer salt in our hands, we
chelating imidazolinium salt homologue 5e. Regarding the then decided to evaluate it in ACA while hoping to observe
structure and atom connectivity from a single-crystal dif- an increase in the asymmetric induction. As shown in
fraction study of the chiral imidazoline 12e, the direct pre- Table 1 (entries 6), salt 5e gave lower enantioselectivity
cursor of salt 5e, we were enthusiastic to observe the ex- (41% ee) than the homologous mesityl-N-substituted salt
pected anti relationship between the ortho-positioned tert- 5b (50% ee). Additionally, a similar selectivity (40% ee) was
butyl group and the isopropyl stereogenic center (Figure 5). observed with the phenyl analog 5f (entry 7). These low
Therefore, we anticipated that the alkylation of imidazoline selectivities might have arisen from the inability of the ste-
12e by bromoethanol (see Scheme 1) should give the locked reogenic centre to transfer the chiral information from the
conformer 5e with the aforementioned anti relationship. β-position of the N atom through steric repulsion. As we
Unfortunately, attempts to crystallize the isolated salt 5e in were not able to obtain the crystal structure of 5e, a struc-
order to confirm its solid-state structure failed.
tural elucidation of the structurally similar salt 5g was at-
Eur. J. Inorg. Chem. 2009, 1989–1999
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1993

D. Rix, S. Labat, L. Toupet, C. Crévisy, M. Mauduit
FULL PAPER
tempted. The latter was obtained in an excellent yield as could be minimized. Consequently, the rate of transfer of
white crystals through the alkylation of the imidazoline 12e the ethyl group to the enone would increase in comparison
by methyl iodide and anionic metathesis exchange. This with the case of the initial ligand 4a bearing a mesityl
time, 5g could be crystallized and its solid-state structure group, to lead to a better ee (Figure 6). Inversely, in the
was then confirmed by a single-crystal diffraction study case of the ligand bearing the (S)-methylnaphthyl entity, a
(Figure 5). We were astonished to observe in salt 5g a syn conformation is adopted which brings some unfavorable
relationship between the tert-butyl and the isopropyl groups steric interactions, which leads to a significant decrease of
although they exhibited an anti relation in the imidazoline the selectivity.
precursor 12e (Figure 5). This could arise from an inversion
of conformation during the nitrogen quaternarisation pro-
cess. An alternate explanation could be that both conform-
Table 2. Evaluation of the hydroxyalkyl-azolinium salts 6 and 7
bearing a stereogenic center within the chelating side chain for the
conjugate addition of diethylzinc to 2-cyclohexenone.[a]
ers (anti and syn) exist in solution but that only one (not
the same for 5e and 5g) crystallizes. This last hypothesis
might explain the low selectivities observed within this fam-
ily of ligands. Several ab initio calculations are currently
underway in order to find a clear-cut explanation.
After the failure to reach an efficient enantiocontrol in
the ACA reaction using the hydroxyalkyl-chelating NHC 5
bearing the stereogenic center within the heterocyclic unit,
we focused our attention on the azolium salts 6 and 7 bear-
ing the stereogenic center within the chelating side chain.
We anticipated that an increase in the steric hindrance
around the metal–NHC environment would lead to an im-
provement in the enantioselectivity in relation to that pre-
viously obtained with SIMes-leucinol salts 4a (86% ee).[12]
Entry
Ligand
Time [h][b]
ee [%][c]
1
2
3
4
5
6
7
8
6a
6b
6c
6d
6e
6f
6g
6h
7a
7b
7c
7e
1
1
1
1
1
1
1
1
1
6
1
2
80 (R)
90 (R)
82 (R)
92 (R)
80 (R)
68 (R)
63 (R)
89 (R)
19 (R)
26 (S)
25 (R)
0
9
10
11
12
[a] Conditions: 6 or 7 (3 mol-%), nBuLi (8 mol-%), Cu(OTf)2
(2 mol-%), 2-cyclohexenone (1 mmol), Et2Zn (1.5 mmol), Et2O. [b]
Time necessary to reach complete conversion. [c] Determined by
chiral GC analysis (Lipodex E).
In addition to these successful ligands, the two other sub-
stituted benzyl-based ligands 6g and 6h were evaluated (en-
tries 7–8). Whereas the mesitylenemethyl salt 6g gave a sig-
nificantly lower enantioselectivity (63%), a good ee value
Two approaches were attempted. Firstly, with hy- was reached with salt 6h, bearing a phenyl substituent in
droxyalkyl salts 6, we envisioned that a better enantiodiscri- the ortho position to the benzyl unit (89%, entry 7). This
mination could be reached by replacing the nonchelating result is similar to the ee value obtained with the
mesityl N substituent group by a chiral bulky fragment. The naphthylmethyl salt 6d. Unexpectedly, when the isobutyl
salt 6a, bearing an (R)-indane moiety, was first evaluated, stereogenic group within the chelating hydroxyalkyl side
but only a decrease of selectivity was observed (Table 2, en- chain of 6d was replaced by the bulkier tert-butyl group
try 1, 80% ee), which was probably due to a less favorable (salts 6e and 6f), the selectivity was not increased, but on
steric hindrance. Salts 6b and 6c, bearing a chiral α-methyl- the contrary was diminished to 80% and 68% respectively
naphthyl group, were therefore evaluated. A significant (entries 5–6). It arises from these results that the presence
match/mismatch effect between the two chiral centers was of a bulky group at the stereogenic center of both the che-
observed (entries 2–3). Whereas the (S)-α-methylnaphthyl lating and the nonchelating side chains is unfortunately in-
6c gave an ee value similar to the one obtained with the compatible with the goal of improving the selectivity.[19]
indane salt 6a (entry 2, 82%), a doping of enantioselectivity
Secondly, we tested the bicyclic NHC-based ligands 7,
was observed with the matched ligand 6b (entry 3, 90%). where the presence of a bicyclic NHC system in which the
With respect to ligand design, we clearly demonstrated that NHC ring is fused together with an aromatic group, was
the orthogonally oriented mesityl group cannot be regarded expected to freeze the conformation and also engender ste-
as the ideal bulky unit for the application of NHC ligands ric repulsions, which in turn were expected to improve the
in ACA of 2-cyclohexenone with Et2Zn. In addition, we selectivity. Unfortunately, very poor ee values were obtained
were astonished to obtain a similar doping effect in the case with ligands 7a–7e, regardless of the nature of both the
of salt 6d in which the methyl stereogenic center is absent stereogenic center of the chelating arm and the nonchelat-
(entry 4, 92% ee). This important result suggests that an ing N substituent (entries 9–12). The absence of selectivity
optimal conformation of the naphthyl unit should be observed in the case of 7e could arise from the formation
reached in the case of the ligands bearing either no α- of π-stacking interactions between the naphthyl and the
methyl or the (R)-α-methyl group so that the steric repul- phenyl groups, which would lead to a conformation where
sions between the enone entity and the naphthyl fragment the naphthyl group is far from the reacting center.[20]
1994
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 1989–1999

Chiral Chelating Diaminocarbene Ligands
Figure 6. Proposed intermediates for the match/mismatch effect observed with NHCs derived from 6b and 6c.
In the concept of architectural design of NHCs dedicated tained when 9a and tBuOK were employed. As a similar ee
to improving the selectivity of ACA, a different approach value (78%) was obtained with SIMes-Ala 4b derived from
was then attempted with imidazolinium salts 8 and 9, bear- alaninol,[12] which bears a methyl group on the chiral car-
ing a hydroxyphenyl-chelating function. Based on Hovey- bon, it can be anticipated that a high ee should be reached
da’s results with salt 2, where the effects of axial and atrop- if an analog of 9 bearing a more bulky group (e.g. iBu or
isomeric chiralities were combined, we tried to check if the tBu) on the chiral carbon is used.[21]
chirality transfer from the isopropyl group in 8 could also
Table 3. Evaluation of the benzyloxy-imidazolinium salts 8 and 9
lead to interesting selectivities (Figure 7).
in the conjugate addition of diethylzinc to 2-cyclohexenone.[a]
Entry
Ligand
Base
Time [h][b]
ee [%][c]
1
2
3
4
5
6
8
nBuLi
tBuOK
nBuLi
tBuOK
nBuLi
tBuOK
1
2
1
1
1
2
20 (S)
8 (S)
64 (S)
78 (S)
11 (S)
4 (S)
8
9a
9a
9b
9b
[a] Conditions: 8 or 9 (3 mol-%), base (8 mol-%), Cu(OTf)2 (2 mol-
%), 2-cyclohexenone (1 mmol), Et2Zn (1.5 mmol), Et2O. [b] Time
necessary to reach complete conversion. [c] Determined by chiral
GC analysis (Lipodex E).
Figure 7. New architectural design related to the hydroxyphenyl-
chelating function in the imidazolinium salts 8 and 9.
Finally, various enones were tested towards the addition,
Unfortunately, very low ee values could be reached, inde- at room temperature, of Et2Zn in the presence of the Cu–
pendent of the conditions used (Table 3 entries 1–2). There- 6b catalyst (Scheme 6). Although better ee values were ob-
fore, we tried to evaluate the efficiency of salts 9, where tained with ligands 6c or 6d in the reaction of cyclohex-
the atropisomer is replaced by a central chiral center in the enone, 6b was selected because it was more chemically
hydroxyaryl-chelating arm. Thus salts 9a and 9b were tested stable than 6c and 6d, which thus led to more reproducible
(entries 3–6), and an encouraging result (78% ee) was ob- results. The results were compared with those obtained
Eur. J. Inorg. Chem. 2009, 1989–1999
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1995

D. Rix, S. Labat, L. Toupet, C. Crévisy, M. Mauduit
FULL PAPER
1
using salt 4a.[12] Except for the case of 2-cycloheptenone,
where similar and good ee values were obtained (89%), it
is worth noting that lower enantioselectivities and in some
cases a lower reactivity were observed with salt 6b.
than H. Chemical shifts are reported in ppm from tetramethylsil-
ane with the solvent resonance as the internal standard (CDCl3,
1H: δ = 7.27 ppm, 13C: δ = 77.0 ppm and (CD3)2CO: 1H: δ =
2.05 ppm, 13C: δ = 205.1 ppm). Data are reported as follows: chem-
ical shift (δ in ppm), multiplicity (s = singlet, d = doublet, t =
triplet, q = quadruplet, sept = septuplet, m = multiplet), coupling
constants [Hz], integration and attribution. High-resolution mass
spectra (HRMS) were recorded at the Centre Régional de Mesures
Physiques de l’Ouest (CRMPO), Université de Rennes 1 on a
Micromass ZABSpecTOF instrument. Melting points were mea-
sured on a heating Reichert microscope and were uncorrected. Op-
tical rotations were recorded using a Perkin–Elmer 341 polarimeter.
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 additions were measured using gas chromatography
¯
(capillary column – HP1, 0.25 µm, 30 m, 0.32 mm) with cyclodo-
decane as internal standard, and enantiomeric excesses were calcu-
lated using chiral gas chromatography (capillary column – Lipodex
E, 0.2 µm, 25 m, 0.25 mm). The products of conjugate additions
were identified according to the literature. All nonaqueous reac-
tions were performed under an argon atmosphere using oven-dried
glassware. Toluene and trimethyl orthoformate were distilled from
sodium metal under nitrogen. Tetrahydrofuran and diethyl ether
were distilled from sodium metal/benzophenone ketyl under nitro-
gen. Dichloromethane, methanol, triethylamine and pyridine were
distilled from calcium hydride under nitrogen. The 2  anhydrous
HCl/MeOH 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 without further
purification. Analytical TLC was performed on Merck silica gel
60F254 plates, and the plates were visualized under UV light. Chro-
matographic purifications were performed on a column with 230–
400 mesh silica gel (Merck 9385) using the indicated solvent sys-
tem.
Scheme 6. Scope of ACA substrates with the new hydroxyalkyl-
NHC precursor 6b derived from leucinol and (S)-α-methylnaph-
thylamine.[a] Results obtained with SIMes-leu 4a (see ref.[12]).
Conclusions
In conclusion, we have reported the development of sev-
eral classes of chiral hydroxyalkyl- and hydroxyphenyl-che-
lating NHC ligands using amino acids as a chiral pool. The
employed modular syntheses allowed the isolation of a few
small libraries of chiral hydroxyalkyl- or hydroxyaryl-azol-
ium hexafluorophosphate salts. Structural parameters were
also studied through the determination of structure–activity
relationships in the enantioselective copper-catalyzed conju-
gate addition of diethylzinc to 2-cyclohexenone. Whereas all
the NHC salt precursors bearing the stereogenic center
within the N-heterocyclic ring gave disappointing results
(salts 5 and 8), good to high enantioselectivities (up to 92%
ee) were obtained at room temperature with hydroxyalkyl-
and hydroxyphenyl-imidazolinium salt 6 and 9 where the
chiral centre is located on the chelating side chain. The
presence of an additional phenyl unit, fused together with
the N-heterocycle, led only to a decrease of the selectivity
General Procedure for the Synthesis of Hydroxyalkyl-Imidazolinium
Salts 5: To a solution of imidazole 12 (1 mmol) in acetone (1 mL)
was added 2-bromoethanol (5 mmol, 2 equiv.). The reaction mix-
ture was stirred at reflux for 48 h. After being cooled to room tem-
perature, LiNTf2 (1.2 mmol, 1.2 equiv.) or KPF6 (1.2 mmol,
1.2 equiv.) was added at room temperature. After 30 min of stirring,
the solvent was removed under reduced pressure. The residue was
dissolved in dichloromethane, and the organic layer was washed
with brine, dried with MgSO4 and concentrated in vacuo. Purifica-
(benzymidazolium salts 7). However, except for 2-cyclohep- tion by flash chromatography (CH2Cl2/acetone, 3:1) afforded the
corresponding imidazolium salt 5.
tenone (89%ee), we were disappointed to observe lower
enantioselectivities when different cyclic enones were scre-
ened as substrates with the best-identified hydroxyalkyl-
NHC precursor 6b (ranging from 36 to 79%). Further work
is currently in progress in our laboratory aiming at the in-
troduction of bulkier chiral groups into the hydroxyphenyl-
chelating side chain in salt 9a. We expect that a significant
improvement of the enantiomeric excesses in ACA will be
General Procedure for the Synthesis of Hydroxyalkyl-Imidazolinium
6 and -Benzimidazolium Salts 7: To a solution of diamine (1 equiv.)
in dry diethyl ether (5 mL/mmol) was added dropwise at 0 °C a
2  HCl/MeOH solution (1 equiv.). A white precipitate was rapidly
formed. After 10 min of stirring, the solvent was removed in vacuo
to afford the corresponding chlorhydrate of diamine. This salt was
dissolved in dry toluene (3 mL/mmol), and trimethyl orthoformate
reached. Extending the use of this family of ligands to other (10 equiv.) was added. The reaction mixture was stirred at 100 °C
overnight. After being cooling to room temperature, the solvents
were evaporated in vacuo, and the imidazolinium chloride was dis-
solved in distilled water. The aqueous layer was washed with ethyl
acetate before the addition of potassium hexafluorophosphate
(1.3 equiv.). After 20 min of stirring at room temperature, the imid-
azolinium or benzimidazolium salt was extracted with dichloro-
enantioselective metal-catalyzed transformations dealing
with palladium or copper will be intensively studied.
Experimental Section
General: 1H (400 MHz), 13C (100 MHz), 31P (162 MHz) and 19F methane. The organic layer was washed with brine, dried with mag-
(376.5 MHz) NMR spectra were recorded on a Bruker ARX400
spectrometer with complete proton decoupling for nuclei other
nesium sulfate and concentrated under reduced pressure. The re-
sulting crude product was purified by flash chromatography on sil-
1996
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 1989–1999

Chiral Chelating Diaminocarbene Ligands
ica gel (dichloromethane/acetone, 9:1) to afford the expected azol-
ium salt.
3,4-Dihydro-5-[(1S)-1-(hydroxymethyl)-3-methylbutyl]-2-(naphth-1-
ylmethyl)-1H-imidazolin-1-ium Hexafluorophosphate (6d): White
solid (270 mg, 55% yield). [α]2D0 = 2 (c = 0.1, acetone); m.p. 62 °C.
1H NMR (400 MHz, CD2Cl2): δ = 7.86 (m, 3 H, Har), 7.81 (s, 1
General Procedure for the Synthesis of Hydroxyaryl-Imidazolinium
Salts 8, 9: To a solution of diamine (1 equiv.) in dry diethyl ether
(5 mL/mmol) was added dropwise, at 0 °C, a 2  anhydrous HCl/
MeOH solution (2 equiv.). A white precipitate was rapidly formed.
After 10 min of strirring, the solvent was evaporated to afford the
corresponding chlorhydrate of diamine. This salt was dissolved in
dry toluene (3 mL/mmol), and trimethyl orthoformate (10 equiv.)
was added. The reaction mixture was stirred for 90 min at 100 °C.
After being cooling to room temperature, the solvents were evapo-
rated in vacuo, and the imidazolinium chloride was dissolved in
distilled water. The aqueous layer was washed with ethyl acetate
before with the addition of potassium hexafluorophosphate
(1.3 equiv.). After 20 min of stirring at room temperature, the imid-
azolinium salt was extracted with dichloromethane. The organic
layer was washed with brine, dried with magnesium sulfate and
concentrated under reduced pressure. The resulting crude product
was purified by flash chromatography on silica gel (dichlorometh-
ane/acetone, 9:1) to afford the expected azolium salt.
3
4
H, H-19), 7.56 (ddd, 3J = 6.9 Hz, J = 8.4 Hz, J = 1.5 Hz, 1 H,
3
3
4
Har), 7.50 (ddd, J = 6.9 Hz, J = 8.0 Hz, J = 1.2 Hz, 1 H, Har),
7.45 (m, 2 H, Har), 4.98 (s, 2 H, H-1), 3.82 (m, 2 H, H-12), 3.76
2
3
(m, 2 H, H-13), 3.67 (m, 1 H, H-14), 3.59 (dd, J = 12.0 Hz, J =
2
3
3.5 Hz, 1 H, H-15), 3.44 (dd, J = 12.0 Hz, J = 8.7 Hz, 1 H, H-
15), 2.68 (s broad, 1 H, OH), 1.40 (m, 2 H, H-16), 1.21 (m, 1 H,
3
3
H-17), 0.83 (d, J = 6.5 Hz, 3 H, H-18), 0.80 (d, J = 6.5 Hz, 3 H,
H-18) ppm. 13C NMR (100 MHz, CD2Cl2): δ = 157.9 (1 CH, C-
19), 134.8 (1 C, Car), 131.8 (1 C, Car), 131.2 (1 C, Car), 130.0 (1
CH, Car), 129.5 (1 CH, Car), 128.3 (1 CH, Car), 128.1 (1 CH, Car),
127.4 (1 CH, Car), 126.3 (1 CH, Car), 123.1 (1 CH, Car), 61.8 (1
CH2, C-15), 60.0 (1 CH, C-14), 50.8 (1 CH2, C-1), 48.9 (1 CH2, C-
12), 46.0 (1 CH2, C-13), 37.7 (1 CH2, C-16), 25.4 (1 CH, C-17),
23.1 (1 CH3, C-18), 22.3 (1 CH3, C-18) ppm. 31P NMR (162 MHz,
CD2Cl2): δ = –144.3 (sept, 1J = 712.4 Hz, 1 P) ppm. 19F NMR
1
(376 MHz, CD2Cl2): δ = –71.4 (d, J = 712.4 Hz, 6 F) ppm.
3,4-Dihydro-5-[(1S)-1-(hydroxymethyl)-2-dimethylpropyl]-2-[(1R)-1-
(naphth-1-yl)ethyl]-1H-imidazolin-1-ium Hexafluorophosphate (6f):
White solid (667 mg, 42% yield). [α]2D0 = –41.0 (c = 1, acetone);
m.p. 65 °C. 1H NMR (400 MHz, CD2Cl2): δ = 7.87 (m, 4 H, H-20,
Selected Salts
(4S)-1-(2-tert-Butylphenyl)-4,5-dihydro-3-(2-hydroxyethyl)-4-isopro-
pyl-1H-imidazol-3-ium Bis(trifluoromethylsulfonyl)azanide (5e):
Slightly yellow oil (352 mg, 62%). [α]2D0 = +16.8 (c = 1, chloroform).
3
Har), 7.52 (m, 3 H, Har), 7.40 (d, J = 6.4 Hz, 1 H, Har), 5.55 (q, J
= 6.8 Hz, 1 H, H-1), 3.82 (m, 6 H, H-13, H-14, H-15, H-16), 3.44
(dd, 2J = 10.4 Hz, 3J = 3.6 Hz, 1 H, H-16), 2.45 (s broad, 1 H,
OH), 1.82 (d, 3J = 6.8 Hz, 3 H, H-12), 0.90 (s, 9 H, H-18) ppm.
13C NMR (100 MHz, CD2Cl2): δ = 157.87 (CH, C-20), 134.56 (C,
Car), 132.51 (C, Car), 130.75 (C, Car), 130.41 (CH, Car), 129.78 (CH,
Car), 127.91 (CH, Car), 126.86 (CH, Car), 125.85 (CH, Car), 124.50
(CH, Car), 122.05 (CH, Car), 70.75 (CH, C-15), 57.91 (CH2, C-16),
54.82 (CH, C-1), 48.13 (CH2, C-13), 47.56 (CH2, C-14), 33.82 (C,
C-17), 27.44 (CH3, C-18), 19.36 (CH3, C-12) ppm. 31P NMR
3
1H NMR (400 MHz, CDCl3): δ = 8.08 (s, 1 H, H-13), 7.55 (d, J
3
3
= 7.4 Hz, 1 H, H-9), 7.43 (t, J = 7.4 Hz, 1 H, H-7), 7.32 (t, J =
3
3
7.4 Hz, 1 H, H-8), 7.18 (d, J = 7.4 Hz, 1 H, H-6), 4.62 (dt, J =
4.8, 11.8 Hz, 1 H, H-2), 4.30 (t, J = 11.8 Hz, 1 H, H-1), 3.95 (t,
3
3J = 11.8 Hz, 1 H, H-1Ј), 3.85 (m, 3 H, H-14, H-15), 3.64 (m, 1 H,
H-14Ј), 2.39 (m, 1 H, H-3), 1.39 (s, 9 H, H-12), 1.07 (d, 3J = 7.0 Hz,
3 H, H-4), 1.01 (d, 3J = 7.0 Hz, 3 H, H-4Ј) ppm. 13C NMR
(100 MHz, CDCl3): δ = 159.2 (1 C, C-13), 147.4 (1 C, C-5), 134.1
(1 C, C-10), 130.7 (1 C, C-9), 129.5 (1 C, C-7), 128.8 (1 C, C-8),
1
(162 MHz, CD2Cl2): δ = –144 (sept, J = 710.3 Hz, 1 P) ppm. 19F
1
128.1 (1 C, C-6), 119.1 (q, JC,F = 321.2 Hz, 1 C), 65.0 (1 C, C-2),
1
NMR (376 MHz, CD2Cl2): δ = –72.5 (d, J = 710.3 Hz, 6 F) ppm.
57.3 (1 C, C-1), 54.8 (1 C, C-15), 47.7 (1 C, C-14), 35.6 (1 C, C-
HRMS (Maldi/TOF): calcd. for C21H29N2O [M]+ 325.2280; found
325.228. C21H29F6N2OP (470.43): calcd. C 53.62, H 6.21, N 5.95;
found C 53.06, H 6.24, N 5.67.
11), 31.7 (3 C, C-12), 26.7 (1 C, C-3), 17.7 (1 C, C-4), 14.4 (1 C,
1
C-4Ј) ppm. 19F NMR (376 MHz, CDCl3): δ = –73.0 (d, JF,C
=
321.2 Hz, 6 F) ppm. HRMS calcd. for C18H29N2O [M]+ 289.2280;
found 289.2280.
2-Cyclohexyl-5-[(1S)-1-(hydroxymethyl)-3-methylpropyl]-1H-benz-
imidazol-1-ium Hexafluorophosphate (7a): White solid (500 mg,
39% yield). [α]2D0 = –7.0 (c = 1, acetone); m.p. 132 °C. 1H NMR
(400 MHz, CD2Cl2): δ = 8.94 (s, 1 H, H-17), 7.76 (m, 2 H, Har),
3,4-Dihydro-5-[(1S)-1-(hydroxymethyl)-3-methylbutyl]-2-[(1R)-1-
(naphth-1-yl)ethyl]-1H-imidazolin-1-ium Hexafluorophosphate (6b):
White solid (160 mg, 65% yield). [α]2D0 = –44.0 (c = 1, acetone);
3
3
7.64 (m, 2 H, Har), 4.47 (tt, J = 3.7 Hz, J = 12.0 Hz, 1 H, H-1),
1
2
3
3
m.p. 101 °C. H NMR (400 MHz, CD2Cl2): δ = 7.99 (m, 4 H, H-
4.34 (ddd, J = 6.3 Hz, J = 9.7 Hz, J = 3.2 Hz, 1 H, H-14), 4.16
3
4
20, Har), 7.68 (td, J = 7.6 Hz, J = 1.6 Hz, 1 H, Har), 7.61 (m, 2
(ddd, 2J = 6.3 Hz, 3J = 11.6 Hz, 3J = 4.9 Hz, 1 H, H-14), 3.99 (ddd,
3
3
3
H, Har), 7.53 (m, 1 H, Har), 5.63 (q, J = 6.8 Hz, 1 H, H-1), 3.89
3J = 12.4 Hz, J = 4.9 Hz, J = 3.2 Hz, 1 H, H-13), 2.77 (m, 1 H,
(m, 4 H, H-13, H-14), 3.75 (dd, 2J = 12.0 Hz, J = 3.6 Hz, 1 H, H-
OH), 2.45 (m, 1 H, H-15), 2.23 (d, J = 12.2 Hz, 2 H, Hcy), 1.88
3
3
16), 3.56 (dd, 2J = 12.0, 3J = 8.8 Hz, 1 H, H-16), 1.93 (d, 3J = (m, 4 H, Hcy), 1.73 (d, J = 12.8 Hz, 1 H, Hcy), 1.56 (m, 2 H, Hcy),
3
6.8 Hz, 3 H, H-12), 1.53 (m, 3 H, H-15, H-17), 1.36 (m, 1 H, H- 1.34 (dt, 2J = 13.2 Hz, 3J = 3.6 Hz, 1 H, Hcy), 1.27 (dt, 2J =
3
3
3
3
18), 0.98 (d, J = 6.8 Hz, 3 H, H-19), 0.92 (d, J = 6.4 Hz, 3 H, H-
19) ppm. 13C NMR (100 MHz, CD2Cl2): δ = 156.7 (1 CH, C-20),
134.6 (1 C, Car), 132.6 (1 C, Car), 130.8 (1 C, Car), 130.4 (1 CH,
Car), 129.8 (1 CH, Car), 127.9 (1 CH, Car), 126.8 (1 CH, Car), 125.9
13.2 Hz, J = 3.6 Hz, 1 H, Hcy), 1.10 (d, J = 6.6 Hz, 3 H, H-16),
0.74 (d, 3J = 6.7 Hz, 3 H, H-16) ppm. 13C NMR (100 MHz,
CD2Cl2): δ = 137.0 (1 CH, C-17), 131.4 (1 C, Car), 130.0 (1 C, Car),
126.7 (1 CH, Car), 126.5 (1 CH, Car), 113.1 (1 CH, Car), 112.8 (1
(1 CH, Car), 124.6 (1 CH, Car), 122.1 (1 CH, Car), 61.4 (1 CH2, C- CH, Car), 66.5 (1 CH, C-13), 60.1 (1 CH2, C-14), 58.2 (1 CH, C-
16), 59.8 (1 CH, C-15), 54.8 (1 CH, C-1), 47.4 (1 CH2, C-13), 45.1 1), 31.7 (1 CH2, Ccy), 31.6 (1 CH2, Ccy), 28.9 (1 CH, C-15), 24.6
(1 CH2, C-14), 36.9 (1 CH2, C-17), 25.1 (1 CH, C-18), 22.7 (1 CH3, (1CH2, Ccy), 24.5 (1 CH2, Ccy), 24.0, (1 CH2, Ccy), 18.6 (1 CH3, C-
C-19), 21.9 (1 CH3, C-19), 19.4 (1 CH3, C-12) ppm. 31P NMR 16), 18.5 (1 CH3, C-16) ppm. 31P NMR (162 MHz, CD2Cl2): δ =
(162 MHz, CD2Cl2): δ = –143.1 (sept, 1 P, 1J = 711.5 Hz) ppm. 19
F
–144.4 (sept, 1J = 710.6 Hz, 1 P) ppm. 19F NMR (376 MHz,
1
1
NMR (376 MHz, CD2Cl2): δ = –73.1 (d, 6 F, J = 711.5 Hz) ppm. CD2Cl2): δ = –72.5 (d, 6 F, J = 710.6 Hz). HRMS (Maldi/TOF):
HRMS (Maldi/TOF): calcd. for C21H29N2O [M]+ 325.2280; found
calcd. for C1 8 H2 7 N2 O [M]+ 287.2123, found 287.2122.
325.2287. C21H29F6N2OP (470.43): calcd. C 53.62, H 6.21, N 5.95; C18H27F6N2OP (432.38): calcd. C 50.00, H 6.29, N 6.48; found C
found C 53.15, H 6.15, N 6.17.
49.72, H 6.11, N 6.43.
Eur. J. Inorg. Chem. 2009, 1989–1999
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1997

D. Rix, S. Labat, L. Toupet, C. Crévisy, M. Mauduit
FULL PAPER
5-[(1S)-1-(Hydroxymethyl)-3-dimethylpropyl]-2-(naphth-1-yl)-1H-
benzimidazol-1-ium Hexafluorophosphate (7e): White solid (55 mg,
81% yield). [α]2D0 = +1.2 (c = 1, acetone); m.p. 123 °C. 1H NMR
dX = 1.061 Mgm–3, λ(Mo-Kα) = 0.71073 Å, µ = 0.64 cm–1, F(000)
= 556, T = 293 K. The sample (0.55ϫ0.45ϫ0.40 mm) was studied
with an automatic CAD4 NONIUS diffractometer with graphite-
(400 MHz, CD2Cl2): δ = 8.83 (s, 1 H, H-22), 7.91 (m, 2 H, Har), monochromatized Mo-Kα radiation. The cell parameters were ob-
7.75 (m, 2 H, Har), 7.60 (m, 3 H, Har), 7.49 (m, 2 H, Har), 7.44 (d, tained by fitting a set of 25 high-θ reflections. The data collection
3J = 8.2 Hz, 1 H, Har), 7.38 (dd, 3J = 7.0 Hz, 4J = 1.0 Hz, 1 H,
(2θmax = 54°, scan ω/2θ = 1, tmax = 60 s, hkl range: h 0,30, k 0,7, l
Har), 6.07 (s, 2 H, H-1), 4.50 (dd, J = 8.4 Hz, J = 4.2 Hz, 1 H, –16,14) gave 1935 unique reflections from which 1494 had
H-18), 4.15 (m, 2 H, H-19), 2.48 (s broad, 1 H, OH), 0.86 (s, 9 H, IϾ2.0σ(I). After Lorenz and polarization corrections, the structure
H-21) ppm. 13C NMR (100 MHz, CD2Cl2): δ = 140.5 (1 CH, C- was solved with SIR-97, which revealed the non-hydrogen atoms
3
3
22), 138.4 (1 C, Car), 134.4 (1 C, Car), 132.8 (1 C, Car), 132.2 (1 C,
Car), 131.3 (1 C, Car), 131.0 (1 CH, Car), 130.8 (1 CH, Car), 129.8
(1 CH, Car), 128.2 (1 CH, Car), 128.1 (1 CH, Car), 128.0 (1 CH,
Car), 127.6 (1 CH, Car), 127.1 (1 CH, Car), 125.9 (1 CH, Car), 122.1
of the compound. After anisotropic refinement, a Fourier differ-
ence map revealed many hydrogen atoms. The whole structure was
refined with SHELXL97 by the full-matrix least-squares techniques
(use of F square magnitude); x, y, z, βij for C, O and N atoms, x,
(1 CH, Car), 113.8 (1 CH, Car), 60.5 (1 CH2, C-19), 50.1 (1 CH2, y, z in riding mode for H atoms; 168 variables and 1494 observa-
2
C-1), 35.3 (1 CH, C-18), 30.3 (1 C, C-20), 27.3 (3 CH3, C-21) ppm.
tions; calcd. w = 1/[σ2(Fo2) + (0.14P)2 + 0.35P] where P = (Fo
+
31P NMR (162 MHz, CD2Cl2): δ = –144.6 (sept, 1J = 710.6 Hz, 1 P) 2Fc )/3 with the resulting R = 0.064, Rw = 0.178 and Sw = 1.034
2
ppm. 19F NMR (376 MHz, CD2Cl2): δ = –72.7 (d, 1J = 710.6 Hz, 6
F) ppm. HRMS (Maldi/TOF): calcd. for C24H27N2O [M]+
359.2123, found 359.2119. C24H27F6N2OP (504.45): calcd. C 57.14,
H 5.39, N 5.55; found C 57.22, H 5.44, N 5.13.
(residual ∆ρ Յ 0.18 eÅ–3).
28: C13H17N2·2PF6, Mr = 346.26, orthorhombic, P2121 21, a =
12.8406(4), b = 13.1442(4), c = 17.3827(4) Å, V = 2933.8(1) Å–3, Z
= 8, dX = 1.568 Mgm–3, λ(Mo-Kα) = 0.71073 Å, µ = 2.5 cm–1,
F(000) = 1424, T = 120 K. The sample (0.22ϫ0.12ϫ0.08 mm) was
studied on a NONIUS Kappa CCD with graphite-monochro-
matized Mo-Kα radiation. The cell parameters were obtained with
Denzo and Scalepack with 10 frames (ψ rotation: 1° per frame).
The data collection (2θmax = 54°, 173 frames via 1.8° ω rotation
and 30 s per frame, hkl range: h –16,16, k –17,17, l –22,22) gave
37066 reflections. The data reduction with Denzo and Scalepack
led to 6697 independent reflections from which 5029 had
IϾ2.0σ(I). The structure was solved with SIR-97, which revealed
the non-hydrogen atoms of the molecule. After anisotropic refine-
ment, many hydrogen atoms were found with a Fourier difference
map. The whole structure was refined with SHELXL97 by the full-
matrix least-squares techniques (use of F square magnitude); x, y,
z, βij for P, F, C and N atoms, x, y, z in riding mode for H atoms;
404 variables and 5029 observations with IϾ2.0σ(I); calcd. w = 1/
5-[3,5-Di(tert-butyl)methyl-2-hydroxyphenyl]-3,4-dihydro-4-isopro-
pyl-2-mesityl-1H-imidazolin-1-ium Hexafluorophosphate (8): White
solid (816 mg, 67% yield). [α]2D0 = –2.5 (c = 1, acetone); m.p. 95 °C.
1H NMR (400 MHz, CD2Cl2): δ = 7.79 (s, 1 H, H-7), 7.32 (d, 4J
4
= 2.4 Hz, 1 H, H-18), 7.08 (d, J = 2.4 Hz, 1 H, H-14), 6.91 (s, 2
2
2
H, H-3), 4.88 (d, J = 14.4 Hz, 1 H, H-12), 4.40 (d, J = 14.4 Hz,
1 H, H-12), 4.25 (m, 1 H, H-9), 3.92 (dd, 2J = 11.9 Hz, 3J = 4.0 Hz,
1 H, H-8), 3.77 (dd, 2J = 11.9 Hz, J = 9.5 Hz, 1 H, H-8), 2.40 (m,
3
1 H, H-10), 2.22 (s, 3 H, H-5), 2.15 (s, 6 H, H-6), 1.07 (s broad, 1
H, OH), 1.36 (s, 9 H, H-17), 1.22 (s, 9 H, H-21), 0.97 (d, 3J =
3
6.8 Hz, 3 H, H-11), 0.87 (d, J = 7.0, 3 H, H-11) ppm. 13C NMR
(100 MHz, CD2Cl2): δ = 158.6 (1 CH, C-7), 151.3 (1 C, Car), 144.7
(1 C, Car), 141.2 (1 C, Car), 136.9 (1 C, Car), 130.7 (1 C, Car), 130.4
(2 CH, Car), 129.8 (2 C, Car), 126.6 (1 CH, Car), 126.1 (1 CH, Car),
120.2 (1 C, Car), 65.4 (1 CH, C-9), 51.4 (1 CH2, C-8), 48.0 (1 CH2,
C-12), 34.5 (1 C, C-20), 31.6 (3 CH3, C-17), 30.4 (3 CH3, C-21),
27.3 (1 CH, C-10), 21.1 (2 CH3, C-11), 18.1 (1 CH3, C-5), 14.5 (2
CH3, C-6) ppm. 31P NMR (162 MHz, CD2Cl2): δ = –710.2 Hz)
ppm. 19F NMR (376 MHz, CD2Cl2): δ = –72.7 (d, 1J = –710.2 Hz,
6 F) ppm. HRMS (Maldi/TOF): calcd. for C30H45N2O [M]+
449.3532, found 449.3509. C30H45F6N2OP (594.66): calcd. C 60.59,
H 7.63, N 4.71; found C 60.97, H 7.69, N 4.75.
[σ2(Fo2) + (0.066P)2] where P = (Fo + 2Fc )/3 with the resulting R
2
2
= 0.045, Rw = 0.103 and Sw = 1.034, ∆ρ Ͻ 0.27 eÅ–3.
30: C16H23N2·PF6, Mr = 388.33, orthorhombic, P212121, a =
11.2502(2), b = 11.8523(2), c = 13.3138(3) Å, V = 1775.27(6) Å–3,
Z = 4, dX = 1.453 Mgm–3, λ(Mo-Kα) = 0.71073 Å, µ = 2.16 cm–1,
F(000) = 808, T = 120 K. The sample (0.45ϫ0.32ϫ0.30 mm) was
studied with a NONIUS Kappa CCD with graphite-monochro-
matized Mo-Kα radiation. The cell parameters were obtained with
Denzo and Scalepack with 10 frames (ψ rotation: 1° per frame).
The data collection (2θmax = 54°, 160 frames via 2.0° ω rotation
and 12 s per frame, hkl range: h –14,14, k –15,15, l –17,17) gave
19823 reflections. The data reduction with Denzo and Scalepack
led to 4075 independent reflections from which 3556 had
IϾ2.0σ(I). The structure was solved with SIR-97, which revealed
the non-hydrogen atoms of the structure. After anisotropic refine-
ment, many hydrogen atoms were found with a Fourier difference
map. The whole structure was refined with SHELXL97 by the full-
matrix least-square techniques (use of F square magnitude); x, y,
z, βij for P, C, O and F atoms, x, y, z in riding mode for H atoms;
227 variables and 3556 observations with IϾ2.0σ(I); calcd. w = 1/
4,5-Dihydro-3-[1-(2-hydroxyphenyl)ethyl]-1-mesityl-1H-imidazolin-
1-ium Hexafluorophosphate (9a): White solid (472 mg, 60% yield).
[α]2D0 = –32.2 (c = 1, acetone); m.p. 183 °C. 1H NMR [400 MHz,
(CD3)2CO]: δ = 8.78 (s, 1 H, H-7), 7.52 (dd, 3J = 7.8 Hz, 4J =
1.8 Hz, 1 H, Har), 7.47 (m, 2 H, Har), 7.04 (s, 2 H, H-3), 7.18–7.08
3
4
3
(td, J = 7.2 Hz, J = 1.8 Hz, 1 H, Har), 5.37 (q, J = 7.1 Hz, 1 H,
H-10), 4.30 (m, 3 H, H-8, H-9), 4.09 (m, 1 H, H-9), 2.33 (s, 6 H,
3
H-5), 2.30 (s, 3 H, H-1), 1.87 (d, J = 7.1 Hz, 3 H, H-11) ppm. 13C
NMR [100 MHz, (CD3)2CO]: δ = 159.4 (1 CH, C-7), 157.2 (1 C,
C-17), 141.2 (1 C, Car), 137.2 (2 C, Car), 132.7 (1 C, Car), 131.3 (1
CH, Car), 130.8 (2 CH, Car), 129.0 (1 CH, Car), 124.2 (1 C, Car),
120.8 (1 CH, Car), 117.7 (1 CH, Car), 55.3 (1 CH, C-10), 51.8 (1
CH2, C-8), 47.9 (1 CH2, C-9), 21.4 (1 CH3, C-11), 18.0 (2 CH3, C-
5), 15.1 (1 CH3, C-1) ppm. 31P NMR [162 MHz, (CD3)2CO]: δ =
–143.0 (sept, 1J = –706.3 Hz, 1 P) ppm. 19F NMR [376 MHz,
(CD3)2CO]: δ = –75.8 (d, 1J = –707.3 Hz, 6 F) ppm. HRMS (Maldi/
TOF): calcd. for C20H25N2O [C]+ 309.1967, found 309.1973.
[σ2(Fo2) + (0.084P)2 + 1.17P] where P = (Fo + 2Fc )/3 with the
resulting R = 0.047, Rw = 0.124 and Sw = 1.061, ∆ρ Ͻ 0.73 eÅ–3.
The absolute configuration was unambiguously confirmed [Flack
parameter: –0.02(9)].
2
2
Crystal Structure Data of Imidazoline 12e and Imidazolinium Salts
28, 30 and 5g
5g: C17H27N2·PF6, Mr = 404.37, monoclinic, P21, a = 9.1683(2), b
= 20.2115(7), c = 10.8752(3) Å, β = 101.599(2)°, V = 1974.1(1) Å–3,
12e: C16H24N2·1/2H2O, Mr
= 506.76, monoclinic, C2, a =
23.516(5), b = 6.124(9), c = 12.726(2) Å, V = 1587(2) Å–3, Z = 4, Z = 4, dX = 1.361 Mgm–3, λ(Mo-Kα) = 0.71073 Å, µ = 1.97 cm–1,
1998
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 1989–1999

Chiral Chelating Diaminocarbene Ligands
Laponnaz, L. Gade, in: N-Heterocyclic Carbenes in Transition
Metal Catalysis (Ed.: F. Glorius), Topics in Organometallic
Chemistry, Springer Verlag, 2007, 21, p. 117–157.
F(000) = 848, T = 120 K. The sample (0.35ϫ0.32ϫ0.12 mm) was
studied on a NONIUS Kappa CCD with graphite-monochro-
matized Mo-Kα radiation. The cell parameters were obtained with
Denzo and Scalepack with 10 frames (ψ rotation: 1° per frame).
The data collection (2θmax = 54°, 278 frames via 1.5° ω rotation
and 10 s per frame, hkl range: h –11,11, k –26,26, l –14,14) gave
35228 reflections. The data reduction with Denzo and Scalepack
led to 8929 independent reflections from which 5857 had
IϾ2.0σ(I). The structure was solved with SIR-97, which revealed
the non-hydrogen atoms of the molecule. After anisotropic refine-
ment, many hydrogen atoms were found with a Fourier difference
map. The whole structure was refined with SHELXL97 by the full-
matrix least-squares techniques (use of F square magnitude); x, y,
z, βij for P, C, F and N atoms, x, y, z in riding mode for H atoms;
470 variables and 5857 observations with IϾ2.0σ(I); calcd. w = 1/
[6] a) J. J. Van Veldhuizen, S. B. Garber, J. S. Kingsbury, A. H. Ho-
veyda, J. Am. Chem. Soc. 2002, 124, 4954–4955; b) J. J.
Van Veldhuizen, D. G. Gillingham, S. B. Garber, O. Kataoka,
A. H. Hoveyda, J. Am. Chem. Soc. 2003, 125, 12502–12508;
c) A. H. Hoveyda, D. G. Gillingham, J. J. Van Veldhuizen, O.
Kataoka, S. B. Garber, J. S. Kingsbury, J. P. A. Harrity, Org.
Biomol. Chem. 2004, 2, 8–23; d) D. G. Gillingham, O. Kataoka,
S. B. Garber, A. H. Hoveyda, J. Am. Chem. Soc. 2004, 126,
12288–12290; e) A. L. Larsen, W. Leu, N. C. Oberhuber, J. E.
Campbell, A. H. Hoveyda, J. Am. Chem. Soc. 2004, 126, 11130.
[7] J. J. Van Veldhuizen, J. E. Campbell, R. E. Guidici, A. H. Hov-
eyda, J. Am. Chem. Soc. 2005, 127, 6877–6882.
[8] For a recent review on enantioselective metathesis reactions,
see: S. J. Malcolmson, S. J. Meek, E. S. Sattely, R. R. Schrock,
A. H. Hoveyda, Nature 2008, 456, 933–937.
[9] For a recent review on enantioselective substitution reactions,
see: B. S. R. Harutyunyan, T. den Hartog, K. Geurts, A. J.
Minnaard, B. L. Feringa, Chem. Rev. 2008, 108, 2824–2852.
[10] For a recent review on enantioselective Cu-catalyzed conjugate
addition, see: A. Alexakis, J. E. Backvall, N. Krause, O. Pamies,
M. Dieguez, Chem. Rev. 2008, 108, 2796–2823.
[σ2(Fo2) + (0.15P)2 + 0.48P] where P = (Fo + 2Fc )/3 with the
2
2
resulting R = 0.078, Rw = 0.204 and Sw = 1.044, ∆ρ Ͻ 0.81 eÅ–3.
CCDC-227053 (for 12e), -714010 (for 28), -714009 (for 30) and
-227054 (for 5g) contain the crystallographic data for this article.
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 www:
http://www.ccdc.cam.ac.uk).
[11] P. L. Arnold, M. Rodden, K. M. Davis, A. C. Scarisbrick, A. J.
Blake, C. Wilson, Chem. Commun. 2004, 1612–1613.
[12] a) H. Clavier, L. Coutable, J.-C. Guillemin, M. Mauduit, Tetra-
hedron: Asymmetry 2005, 16, 921–924; b) H. Clavier, L. Cout-
able, L. Toupet, J.-C. Guillemin, M. Mauduit, J. Organomet.
Chem. 2005, 690, 5237–5254.
Supporting Information (see also the footnote on the first page of
this article): Experimental procedures for the synthesis of ligands
5, 6, 7, 8 and 9 and their full characterization data.
[13] D. Martin, S. Kehli, M. d’Augustin, H. Clavier, M. Mauduit,
A. Alexakis, J. Am. Chem. Soc. 2006, 128, 8416–8417.
[14] H. Henon, M. Mauduit, A. Alexakis, Angew. Chem. Int. Ed.
2008, 47, 9122–9124.
Acknowledgments
[15] H. Clavier, L. Boulanger, N. Audic, L. Toupet, M. Mauduit,
J.-C. Guillemin, Chem. Commun. 2004, 1224–1225.
[16] A. Alexakis, C. Benhaim, S. Rosset, M. Human, J. Am. Chem.
Soc. 2002, 124, 5262–5263.
[17] The diamines 27 and 29 were synthesized following the synthe-
sis described in Scheme 1 using the Boc--proline and the 2-
methylaniline or the 2-tert-butylaniline respectively.
This work was supported by the Centre National de la Recherche
Scientifique (CNRS), the Région Bretagne and the Ministère de la
Recherche et de la Technologie (grants to H. C. and D. R.). M. M.
and D. R. thank the Agence Nationale de la Recherche (ANR)
(CP2D/RDR2 program) for his financial support. Both, Rennes
Métropole and Bretagne Valorisation are gratefully acknowledged
by M. M. (grant V33-2-DiF-2007). The authors thank Prof. A. [18] This phenomenon is extremely rare and only a few examples
have been reported in the literature.
Alexakis for the generous gift of chiral 2-methoxy-α-methylben-
zylamine 24.
[19] The remarkable benefit of using a bulkier tert-butyl group on
the chelating side chain for the ligand 6 has been clearly dem-
onstrated during the enantioselective formation of the all-qua-
ternary carbon center derived from the addition of 3-substi-
tuted cyclohexenone, see ref.[13]
[20] The π-stacking effect has been postulated by Roland and
Alexakis, see: C. L. Winn, F. Guillen, J. Pytkowicz, S. Roland,
P. Mangeney, A. Alexakis, J. Organomet. Chem. 2005, 690,
5672–5695.
[21] The homologue of the chiral (2-methoxybenzyl)amine 24 bear-
ing a tert-butyl group instead of the methyl has been recently
synthesized by Kündig and coworkers, see: E. P. Kündig, T. M.
Seidel, Y.-X. Jia, G. Bernardinelli, Angew. Chem. Int. Ed. 2007,
46, 8484–8487.
[1] K. Öfele, J. Organomet. Chem. 1968, 12, 42–43.
[2] H. W. Wanzlick, H. J. Schönherr, Angew. Chem. Int. Ed. Engl.
1968, 7, 141–142.
[3] A. J. Arduengo, R. L. Harlow, M. J. Kline, J. Am. Chem. Soc.
1991, 113, 361–363.
[4] For excellent books dealing with NHCs in synthesis and cataly-
sis, see: a) N-Heterocyclic Carbenes in Synthesis (Ed.: S. P. No-
lan), Wiley-VCH, Weinheim, 2006; b) N-Heterocyclic Carbenes
in Transition Metal Catalysis (Ed.: F. Glorius), Topics in Orga-
nometallic Chemistry, Springer Verlag, 2006.
[5] For recent reviews on chiral NHC, see: a) H. Clavier, M. Mau-
duit, in: N-Heterocyclic Carbenes in Synthesis (Ed.: S. P. No-
lan), Wiley-VCH, Weinheim, 2006, p. 183–222; b) S. Bellemin-
Received: December 23, 2008
Published Online: April 2, 2009
Eur. J. Inorg. Chem. 2009, 1989–1999
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1999