Formation of quaternary chiral centers by N-Heterocyclic carbene-CuCatalyzed asymmetric conjugate addition reactions with grignard reagents on trisubstituted cyclic enones

Article abstract of DOI:10.1002/chem.201000471

The copper-catalyzed conjugate addition of Grignard reagents to 3-substituted cyclic enones allows the formation of all-carbon chiral quaternary centers. We demonstrate in this article that N-heterocyclic carbenes act as efficient chiral ligands for this transformation. High enantioselectivities (up to 96% ee) could be obtained for a variety of substrates.

Full text of DOI:10.1002/chem.201000471

DOI: 10.1002/chem.201000471
Formation of Quaternary Chiral Centers by N-Heterocyclic Carbene–Cu-
Catalyzed Asymmetric Conjugate Addition Reactions with Grignard
Reagents on Trisubstituted Cyclic Enones
Stefan Kehrli,[a] David Martin,[b] Diane Rix,[c] Marc Mauduit,[c] and
Alexandre Alexakis*[a]
Dedicated to Professor Josꢀ Barluenga on the occasion of his 70th birthday
Abstract: The copper-catalyzed conjugate addition of Grignard reagents to 3-sub-
stituted cyclic enones allows the formation of all-carbon chiral quaternary centers.
We demonstrate in this article that N-heterocyclic carbenes act as efficient chiral
ligands for this transformation. High enantioselectivities (up to 96% ee) could be
obtained for a variety of substrates.
Keywords: asymmetry · conjugate
addition · copper · Grignard reac-
tion · N-heterocyclic carbenes
Introduction
posed an alternative by using organoaluminum reagents
with phosphorus-based ligands on nonactivated[11–14] cyclic
enones. The main drawback of these approaches is the limi-
tation in terms of easily available nucleophiles.
The asymmetric copper-catalyzed conjugate addition
(A.C.A.) is a reaction of choice for forming C–C bonds.[1]
However, it remains a challenging reaction when applied to
trisubstituted enones.[2–5] Some possibilities were already
proposed, for example, organozinc nucleophiles associated
with phosphoramidite ligands give excellent enantioselectivi-
ties on doubly activated Meldrum acid derivatives.[6–8] The
zinc species were also associated with peptidic ligands on ni-
troolefin[9] or doubly activated cyclic enones[10] in high yields
and enantiometric excesses (ee). However, the use of acti-
vated Michael acceptors is almost always needed to obtain
good conversions, which is the fault of zinc nucleophiles that
often react slowly, due to their low nucleophilicity. We pro-
A new family of promising ligands called N-heterocylic
carbenes (NHCs),[15–21] which are believed to possess better
electrodonating properties[22–24] and steric factors[23,25,26] than
their phosphine counterparts, emerged and were tested in
1,4-addition reactions. Since the first publications in 2001
about the use of NHC–Cu complexes in the conjugate addi-
tion of organometallic nucleophiles on disubstituted enones,
first by Woodward[27] in a racemic way, then by Alexakis[28a]
and Mangeney[28b] in the chiral version, many papers were
published about the subject.[29–34] The majority of these
A.C.A. reactions were carried out by using organozinc nu-
cleophiles on linear and cyclic enones with good ee values
lying between 50 to 93%. Only Hoveyda reached 97% ee
on the A.C.A. of silicon-based nucleophiles to cyclic disub-
stituted enones.
The use of trisubstituted enones to create quaternary
chiral centers is more difficult due the steric hindrance of
the b-position. As a result, only a few documents report on
the subject. In 2006, Alexakis[35] and Hoveyda[36] published
separately the first papers about the formation of quaterna-
ry chiral centers catalyzed by Cu–NHC complexes. Hoveyda
used a large excess of organozinc nucleophiles with an Ag–
[a] S. Kehrli, Prof. A. Alexakis
Dꢀpartement de Chimie Organique
Universitꢀ de Genꢁve
30 quai Ernest-Ansermet, 1211 Genꢁve-4 (Switzerland)
Fax : (+41)223793215
E-mail: alexandre.alexakis@unige.ch
[b] Dr. D. Martin
UPCAM iSm2 Service A62
Campus Scientifique de St Jꢀrꢂme
F-3397 Marseille cedex 20 (France)
[c] D. Rix, Dr. M. Mauduit
NHC pre-catalyst. The use of
a 1:1 mixture of Cu-
Ecole Nationale Supꢀrieure de Chimie de Rennes
Av. Gꢀnꢀral Leclerc, F-35700 Rennes (France)
AHCTUNGRTEN(NGUN OTf)·C6H6 and NHC allowed ee values of up to 93% to be
obtained for the addition of Et2Zn on 3-methylcyclohex-2-
enone, and up to 97% to be obtained for the addition of
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201000471.
9890
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 9890 – 9904

FULL PAPER
Ph2Zn on the same substrate. The variation of the chelating
group, from the phenol to sulfonyl allowed Hoveyda[37] to
add zinc nucleophiles to activated five-membered trisubsti-
tuted cyclic enones. The difference of reactivity seems to
come from the more strained metallacycle (seven-membered
ring) and the lower basicity of the sulfonyl group relative to
the phenol. To accelerate the reaction rate, Hoveyda also
applied his methodology to the nucleophilic addition of or-
ganoaluminum reagents to five, six, and seven-membered
cyclic enones.[38,39] The alkyl nucleophiles, namely methyl,
ethyl, and isobutyl are commercially available. The aryl ones
were synthesized by mixing the corresponding aryl lithium
reagent with AlEt2Cl. This methodology allowed ee values
of up to 97% to be obtained for the addition of Et3Al on
methylcyclopentenone. After our group had carried out the
first A.C.A. reactions with Grignard reagents and NHCs in
high yields and ee values,[35] Tomioka proposed a C2-sym-
metric NHC in 2008,[40] with the chiral back-induction of
two phenyls on the methoxyphenol nitrogen substituent.
With a 1:1.3 Cu/NHC ratio, he was able to obtain up to
80% ee for the addition of EtMgBr to the 3-methylcyclo-
hex-2-enone. The purpose of the present paper is to find a
simple methodology to functionalize trisubstituted cyclic
enones by combining high reaction rates and ee values with
a large scope of alkyl and aryl chains. We therefore propose
to compare our C2-symmetric and nonsymmetric NHCs to
investigate the behavior of each family on trisubstituted
cyclic enones.
Scheme 1. C2-symmetric NHC synthesis. BINAP=2,2’-bis(diphenylphos-
phino)-1,1’-binaphthyl; dba=(E,E)-dibenzyllideneacetone.
With a panel of C2-symmetric NHCs in hand, we tested
them in the copper-catalyzed 1,4-addition on trisubstituted
cyclic enones. We started our study by a screening of nucleo-
philes. The first observation we made was the total regiose-
lectivity of the reactions in favor of the conjugate addition.
As expected, the simplest NHC 1b, with no chelating func-
tionality or fixed stereoinducing groups, gave the worst re-
sults in terms of enantioselectivities and conversion. Howev-
er, it allowed us to point out that organozinc nucleophiles
were less reactive than the other ones, as there is no conver-
sion after 16 h (Table 1, entry 1). Et3Al and EtMgBr gave
NHCs may be separated into three families of ligands, de-
pending on their chiral inducing ability. Herrmannꢄs type
C2-symmetric NHC A bearing chiral substituents on the ni-
Table 1. Organometallics selection.
Entry ImH+ RM
t [h] Prod. Conv. [%][a] ee [%], config.[b]
trogen atoms were the first to be synthesized.[41] The second
C2-symmetric family B possesses chiral bulky groups on the
back of the ring, those inducing the chirality to the N- sub-
stituents and lowering their flexibility. Finally, the third
family C is composed of NHCs bearing N-chiral substituents
containing a strong chelating functionality, which binds to
the TM (transition metal) and prevents any movement of
the steric bulk around the reacting center.
1
2
3
4
5
1b
1b
1b
2a
2a
Et2Zn
Et3Al
EtMgBr 0.5
Et3Al 16
EtMgBr 0.5
16
16
1
85
86
94
92
–
9, (À) S
3
2
3
3
17, (+) R
54, (À) S
68, (À) S
[a] Determined by GCMS analysis after 16 h. [b] Determined by chiral
GC analysis (Lipodex E).
the best conversions but the enantioselectivities remained
low (entries 2 and 3). For the ImH+ 2a, the rotation around
the C–N axis was constrained, due to the stereoinduction
from the back of the heterocycle. As a result, the stereoin-
formation was better transmitted to the reacting center and
it was not surprising to see an increase of the ee values with
those ligands (Scheme 2). As the chiral information is not
fixed on the TM by a heteroatom chelation, the main ad-
vantage of that kind of C2-symmetric NHC is some possible
adaptability of the reactive volume, depending on the nucle-
C2-Symmetric NHCs: We started our investigations by syn-
thesizing different C2-symmetric NHCs through a Buch-
wald–Hartwig
coupling
of
diphenylethanediamine
(DPEDA)[42] and different bromo-aromatic reagents
(Scheme 1). The obtained substituted diamines were then
cyclized with triethylorthoformate to give the corresponding
ImH+. The fixed phenyl moieties on the back of the imida-
zolinylidene ring induced the stereoinformation on the ni-
trogen substituents, and by the way, to the reacting center.
Chem. Eur. J. 2010, 16, 9890 – 9904
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9891

A. Alexakis et al.
Table 3. C2-symmetric NHC screening.
ophile size. This is not possible
when the substituent is bound
to the TM.
For both families, the alumi-
num species gave worse results
in terms of enantioselectivity
than their organomagnesium
Scheme 2. C2-symmetric NHC
perspective view.
counterparts
(Table 1,
en-
tries 2–3 and 4–5). Moreover,
the Grignard reagents gave the shortest reaction times and
we will see later that those nucleophiles were also basic
enough to deprotonate the ImH+, facilitating the procedure.
A screening of different copper sources, solvents, and
temperatures was then carried out to find the best reaction
conditions on the 1,4-addition of EtMgBr to 3,3-methylcy-
clohex-2-enone (Table 2).
Entry
ImH+
Prod.
Conv. [%][a]
ee [%], config.[b]
1
2
3
4
1a
1b
1c
2a
2a
2a
2a
2b
2c
2d
2
2
2
3
3
3
3
3
3
3
81
86
75
92
89
94
96
82
87
85
9, (+) R
17, (+) R
42, (+) R
68, (À) S
2, (À) S
5[c]
6[d]
7[e]
8
73, (À) S
70, (À) S
17, (À) S
63, (À) S
10, (À) S
Table 2. Copper salt and solvent screening.
9
10
[a] Conversion determined by GCMS analysis after 30 min or 3 h (T<
08C). [b] Determined by chiral GC analysis (Lipodex E). [c] The sub-
strate was added first, and then the Grignard reagent was added drop-
(OTf)2 (3 mol%), ImH+ (4 mol%). [e] Cu
ACHUTGTNRNENUG HCATUTGNREN(NUGN OTf)2 (3 mol%),
Entry CuX
Solvent
T [8C] Conv. [%][a] ee [%],
wise. [d] Cu
config.[b]
ImH+ (6 mol%).
1
2
3
4
5
6
7
8
Cu
Cu
Cu
Cu
Cu
Cu
Cu
Cu
N
THF
0
0
0
0
0
>99
77
80
87
32
95
92
85
82
73
98
98
0
57, (À)
40, (À)
38, (À)
41, (À)
60, (À)
68, (À)
37, (À)
38, (À)
8, (À)
MtBE
CH2Cl2
toluene
dioxane
Et2O
Et2O
Et2O
Et2O
Et2O
gave only 9% ee (entry 1). With ligand 1b containing a 2-
ethylnaphthyl group, the ee drops to 17% (entry 2).
For ImH+ salts 2a–d, the results were more tricky: the
enantioselectivity decreased in the following order: 2a
(68%)>2c (63%)@2b (17%)>2d (10%) (Table 3, en-
tries 4 and 8–10). The ligand 2d, which gave up to 90% ee
for the desymmetrization of trienes by Grubbs metathesis,[43]
led to a disappointing result under our experimental condi-
tions. The difference between 2a and 2b was certainly due
to the remoteness of the naphthyl group from the reacting
center going from 1-naphthyl to 2-naphthyl. When the envi-
ronment surrounding the copper was less hindered the ee
decreased dramatically.
The order of addition was also highly important. Indeed,
if the Grignard reagent was added on the substrate, the ee
dropped to 2% (Table 3, entry 5) when using ImH+ 2a.
This observation could be explained if the active asymmetric
species was an “ate-complex” or a higher-order cuprate,
RT
0
À40
9
CuBr
0
10
11
12
CuTC[c]
0
0
0
[Cu
[Cu
(MeCN)4]PF6
E
Et2O
68, (À)
(MeCN)4]BF4 Et2O
40, (À)
[a] Conversion determined by GCMS analysis after 30 min or 3 h. (T<
08C). [b] Determined by chiral GC analysis (Lipodex E). [c] CuTC=
copper thiophene carboxylate.
The most important parameter seemed to be the solvent.
With the same copper source, the variation of solvent
changed the enantioselectivity of the reaction dramatically.
Indeed, as Et2O gave the best ee (68%; Table 2, entry 7),
the use of THF gave a racemate (entry 1). Methyl tert-buty-
lether (MtBE), CH2Cl2, and toluene gave no improvement
to our reaction (entries 2–4). Interestingly, dioxane in which
the Schlenk equilibrium is shifted in favor of R2Mg, since
magnesium salt MgBr2 precipitates in that solvent, still gave
41% ee with low conversion (entry 5). The best temperature
such as [CuEt2ACHTUNTRGNEUNG(NHC)]. Indeed, when the Grignard reagent
was added after the substrate, the Mg–Cu transmetallation
time was shorter and only organocopper reagents were
formed. When the Grignard reagent was added first, the
transmetallation had time to occur twice to form the higher-
order cuprate, which could be the asymmetric catalytic spe-
cies that brings high ee values.
for this reaction was 08C and CuACHTUNRGTENNGU(OTf)2 or [CuACHTUNTGERN(NUGN MeCN)4]PF6
(entries 1 and 12) appeared to be appropriate copper sour-
ces for the reaction.
By using the previously optimized reaction conditions, the
addition of EtMgBr on 3-methylcyclohex-2-enone was
tested with the different imidazolium salts (ImH+)
(Table 3).
This contrasts with the copper-catalyzed asymmetric allyl-
ic substitution for which the Grignard reagent is added very
slowly to the substrate to avoid the formation of cuprate
species.[44,45] Finally, a Cu/NHC ratio of 1:1.3 gave the best
ee values (Table 3, entry 6) and a 1:2 ratio decreased the ee
values (entry 7).
For Hermannꢄs type ImH+, the best result was obtained
by using ligand 1c to give 42% ee (Table 3, entry 3) as 1a
9892
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 9890 – 9904

Formation of Quaternary Chiral Centers
FULL PAPER
With our optimized conditions in hand, we applied our
methodology to different alkyl organomagnesium reagents.
It is interesting to observe the same facial selectivity of the
nucleophilic approach. Indeed, the nucleophile always en-
tered from the Re side of the cyclic enone when using ImH+
2a, which was proven by the fact that the opposite major
enantiomer was obtained when inverting the nucleophile
and the substrateꢄs substituent (Table 4, entries 1 and 6).
Table 4. Substrate variation.
Scheme 3. Synthesis of 2e.
Table 5. Grignard variation with ligand 2e.
Entry R1 R2
R3
Prod. Conv. [%][a] ee [%], config.[b]
1
2
3
4
5
6
7
8
H
H
H
H
H
H
H
Me
Me
Me
Me
Me
Et
Et
nBu
but-3-en
c-pentyl
iPr
3
4
5
6
7
92
99
81
97
99
99
88
93
73, (À) S
73, (À) S
73, (+) R
60, (À) S
39, (À) S
21, (+) R
50, (+) S
71, (À) S
Entry
R
Prod.
Conv. [%][a]
ee [%], config.[b]
1
2
3
4
5
Et
3
4
6
7
10
>99
>99
>99
>99
>99
67, (À) S
72, (À) S
56, (À) S
70, (À) S
0
Me
ACHTUNGTRENNUNG
nBu
c-pentyl
iPr
but-3-en Et
Et
8
9
Me Me
tBu
[a] Conversion determined by GCMS analysis after 30 min or 3 h (T<
08C). [b] Determined by chiral GC analysis (Lipodex E).
[a] Conversion determined by GCMS analysis after 30 min or 3 h (T<
08C). [b] Determined by chiral GC analysis (Lipodex E).
In accordance with Tomiokaꢄs work,[40] the variation of
À
the counter-ion from BF4 to ClÀ has almost no effect on
pocket and the result was an increased ee. Moreover, the re-
action pocket stayed big enough with ligand 2e to allow the
addition of a tert-butyl group on the 3-methylcyclohex-2-
enone (S1) (Table 5, entry 5), but in a racemic way. The
group of Tomioka[40] recently published a study in which
they used a C2-symmetric NHC substituted by two methoxy-
phenyls.
Their ee values were slightly better for the addition of iso-
propyl and ethyl organomagnesium nucleophile, but the
presence of two naphthyls in-
the ee of the conjugate addition. The linear nucleophiles,
except methyl, which reacts with poor enantioselectivity,
gave good results with ee values of around 73% (Table 4,
entries 1–3). The addition of EtMgBr to the poorly reactive
isophorone gave a promising 71% ee (entry 8). Unfortunate-
ly, when the Grignard reagent was a-branched, the ee de-
creased (entries 4–5).
As the NHC naphthyl substituents were not coordinated
to the copper, the steric bulk of the nucleophile may “push”
those naphthyls away from the reacting center during the
transmetallation step, and by doing so, leave the copper
with weaker chiral induction as observed with the 2-naph-
thyl substituents. To prove this hypothesis, we used the C2-
symmetric NHC 2e, containing the same structure as 2a,
but with two extra chelating methoxy groups (Scheme 3).
A copper salt and solvent screening showed that the best
conditions for 2a remained the best for 2e. With those con-
ditions in hand, we applied ImH+ 2e to a series of linear
and branched Grignard reagents.
The linear nucleophiles gave the same results as with
ImH+ 2a. In contrast, for the a-branched Grignard reagents,
the results were totally different. With c-pentyl, the ee de-
creased slightly (60 to 56%; Table 5, entry 4), but with iso-
propyl, the ee increased dramatically from 39 to 70%
(entry 6).
These results were in accordance with our previous hy-
pothesis. Indeed, the presence of chelating substituents on
the naphthyl group prevents the “opening” of the chiral
stead of two phenyls seems to
bring
Indeed, in Et2O we do not ob-
serve 1,2-addition product
a
better regiocontrol.
a
with 2e, unlike Tomiokaꢄs
ligand, which gives around a
95:5 1,4/1,2 (1,4/1,2-addition)
ratio.
With these encouraging results in hand, we wanted to
extend the scope of our methodology by successfully adding
aromatic nucleophiles on trisubstituted cyclic enones. As the
two C2-symmetric analogues react almost in the same way
with linear Grignard reagents but differently with branched
nucleophiles, we tested them in the addition of aromatic nu-
cleophiles. Therefore, we started as usual by finding the best
reaction conditions for the addition of PhMgBr to the 3-
methylcyclohex-2-enone.
As for the alkyl Grignard reagents, the Herrmannꢄs
family gave the worst results (Table 6, entry 1). The ImH+
Chem. Eur. J. 2010, 16, 9890 – 9904
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9893

A. Alexakis et al.
Table 6. C2-symmetric ligand screening for PhMgBr addition.
copper complex seemed to become unstable at 08C, because
the reaction was messy and the complex became black after
10 min.
In conclusion, the C2-symmetric NHCs were good ligands
to catalyze regioselectively and enantioselectively the asym-
metric conjugate addition of alkyl Grignard reagents, with
ee values of up to 73% (Table 4). The presence of two me-
thoxy chelating substituents on the NHC allowed the main-
tenance of the chiral information around the reactive center,
and by doing so, enhanced the ee for the addition of a-
branched nucleophiles (isopropyl, 39% ee with 2a to
70% ee with 2e). The addition of an aromatic organomag-
nesium nucleophile gave an impressive 88% ee with 2a, but
the regioselectivity dropped down. The use of the methoxy
ImH+ 2e allowed the regioselectivity in favor of the 1,4-ad-
dition to increase by keeping a good ee value (70%).
Entry ImH+
Add. t [min] Ratio 1,2/1,4 Conv. [%][a] ee [%][b]
1
2
3
4
1b
2a
2b
2e
30
30
30
30
97:3
97
99
97
99
4, (+)
72:28
49:51
16:84
88, (+)
10, (+)
48, (+)
[a] Conversion determined by GCMS analysis after 60 min. [b] Deter-
mined by chiral GC analysis (Hydrodex-B-3P).
2a gave very good ee values (88%), but the 1,4 regioselec-
tivity was very low (entry 2). As for the addition of aliphatic
Grignard reagents, ligand 2b gave poor ee values, because
of the remoteness of the naphthyl substituents from the re-
acting center. The main drawback of this reaction was the
regioselectivity, largely in favor of the 1,2-addition. The use
of ImH+ 2e allowed a better regiocontrol, although the ee
was disappointingly low (48%; entry 4). Therefore, we made
a short screening of different parameters to see if the ee
could be increased.
Alkoxy-substituted NHCs: As mentioned previously, the
NHC containing a chelating group on the chiral substituent
should induce better enantioselectivity, due to their fixed
steric hindrance. We therefore followed a procedure devel-
oped by Mauduit[31] to synthesize a third family of NHCs
containing an alkoxy group, whereas some were directly
provided by Mauduitꢄs group. The obtained ImH+ com-
pounds were then tested in the A.C.A. reaction. To begin
the study with the bidentate ligands 3b, a screening of sol-
vents and temperatures was first carried out on the conju-
gate addition of EtMgBr to 3-methylcyclohex-2-enone.
Surprisingly, in this case, unlike the addition of an alkyl
nucleophile, the best solvent was CH2Cl2 with high regiose-
lectivity and ee values (Table 7, entry 6). The presence of the
Table 7. Reaction condition variation to increase regioselectivity.
Entry
CuX
Solvent
Conv.
[%][a]
Ratio
1,2/1,4
ee [%],
config.[b]
1
2
3
4
5
6
7
8
Cu
CuBr·Me2S
[Cu
[Cu
Cu
CuBr·Me2S
[Cu(OTf)]2·C6H6
Cu(OAc)·H2O
G
Et2O
Et2O
Et2O
toluene
toluene
CH2Cl2
CH2Cl2
CH2Cl2
>99
>99
>99
>99
>99
>99
>99
>99
16:84
11:89
11:89
17:83
95:5
13:87
28:72
20:80
48, (+)
46, (+)
46, (+)
64, (+)
68, (+)
70, (+)
72, (+)
70, (+)
The best results were obtained in Et2O at 08C (Table 8,
entry 1). MtBE, which often showed better enantioselectivi-
ties than Et2O[46] gave worse results in our case (entry 9).
The temperature is also an important parameter. Indeed, in
all solvents tested (except THF), the ee decreased with the
temperature. In THF, the opposite was observed (entries 5
and 6); however, the ee was by far the worst of all the sol-
vents tested.
[a] Conversion determined by GCMS analysis after 60 min. [b] Deter-
mined by chiral GC analysis (Lipodex E).
À
OMe functionalities were also shown to be highly impor-
Another advantage of using Grignard reagents, in addi-
tion to the ease of synthesizing a broad range of different
alkyl or aryl magnesium nucleophiles, is the strong basicity
of the compounds, which can deprotonate the ImH+ in situ.
Indeed, unlike aluminum and zinc nucleophiles, strong
bases, such as nBuLi, were not required to activate the
ImH+. The absence of nBuLi had no influence on conver-
sion and ee (Table 8, entries 3 and 4).
tant, not for the enantioselectivity but for the regioselectivi-
ty of the reaction. In fact, almost all the reactions gave a
much better 1,4/1,2 ratio with 2e than with 2a. The temper-
ature parameter was also investigated, but À308C remained
the optimal temperature in terms of ee and regioselectivity.
By lowering the temperature to À458C or warming it to
08C, the 1,2-addition was favored. Moreover, the phenyl
9894
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 9890 – 9904

Formation of Quaternary Chiral Centers
Table 8. Solvent and temperature screening.
FULL PAPER
tries 7, 10, and 11). The active species seemed to become
too hindered to favor the 1,4-addition. Compound 3d gave
the best enantioselectivity and led to almost total conversion
(entry 4).
With our optimized conditions in hand, we extended the
scope of the reaction with other Grignard reagents on 3-
methylcyclohex-2-enone.
The bidentate ligand 3d proved to efficiently catalyze the
regio- and enantioselective A.C.A reactions of linear and
branched Grignard reagents. Indeed, the addition of linear
alkyl chains (ethyl, butyl, and butenyl) led to ee values of
around 80 to 90% (Table 9, entry 4 and Table 10, entries 1
Entry Solvent ImH+ Base
T [8C] Conv. [%][a] ee [%], config.[b]
1
2
3
4
5
6
7
8
Et2O
Et2O
Et2O
Et2O
THF
THF
CH2Cl2 3b
CH2Cl2 3b
MtBE
MtBE
3b
3b
3d
3d
3b
3b
nBuLi
0
>99
85
69, (+)
67, (+)
80, (+)
80, (+)
26, (+)
35, (+)
46, (+)
44, (+)
49, (+)
13, (+)
nBuLi À30
nBuLi
–
0
0
0
>99
>99
92
nBuLi
nBuLi À78
71
DBU[c]
0
63
DBU[c] À30
66
9
3b
3b
nBuLi
0
47
10
nBuLi À78
66
Table 10. Variation of the Grignard reagent.
[a] Conversion determined by GCMS analysis after 30 min. [b] Deter-
mined by chiral GC (Lipodex E). [c] DBU=1,8-diazabicyclo[5.4.0]undec-
7-ene.
AHCTUNGTRENNUNG
Entry
R
T
Substrate
Prod. Conv.
[%][a]
Yield ee, con-
With these defined parameters in hand, we optimized the
Cu/NHC ratio. The minimum ratio to obtain the highest ee
(80%) was 1:1.3. Below that ratio, the ee decreased (70% ee
for 1:1). Surprisingly, by increasing the ratio to 1:2, the ee
stayed almost unchanged (78% ee for 1:2).
[8C] add. t [min]
[%][b] fig.[c]
1
2
3
4
5
6
7
8
nBu
0
60
12
13
14
15
15
16
16
16
17
10
18
100
91
100
70[d]
100
86
100
100
100
40
–
80
–
–
77
–
78, R
90, S
96, S
n.d.[e]
78, R
84, R
85, R
86, R
79, R
–
Butenyl À30 15
iBu
iPr
iPr
À30 60
0
10
À18 50
À15 10
À30 15
À30 60
À30 10
À30 10
The results with different alkoxy NHCs showed clearly
that the enantioselectivity of the reaction decreases with the
steric hindrance on the nitrogen substituent. Indeed, the
best ee values were obtained with tBuꢀiBu>iPr>Me>
Bn@Ph. As expected, the two ImH+ compounds 3e and 3 f,
which contain an aromatic group, did not bring a strong
chiral induction on the reacting center. In fact, as there is no
p interaction between the aliphatic Grignard reagent and
the phenyl moiety, only the steric hindrance influenced the
reaction. By increasing the steric bulk around the reacting
center, modifying the chiral or achiral substituent, the regio-
selectivity of the reaction decreased. Indeed, with exotic li-
gands 3g, 3k, and 3j, the regioselectivity of the reaction de-
creased dramatically to almost 50:50 1,2/1,4 (Table 9, en-
c-Pent
c-Pent
c-Pent
c-Hex
tBu
–
80
77
0
9
10
11
Me-
À30
5
99
–
0
TMS[f]
[a] Conversion determined by GCMS analysis after 15–30 min. [d] Isolat-
ed yield. [c] Determined by chiral GC analysis (Lipodex E). [d] For-
mation of many byproducts, observed per GCMS. [e] n.d. =not deter-
mined. [e] TMS=trimethylsilyl
and 2). For a-branched nucleophiles (isopropyl, c-pentyl,
and c-hexyl), the enantioselectivities stayed close to the
range obtained for linear ones with 78 to 86% ee (Table 10,
entries 5, 8, and 9). The b-branched nucleophiles gave con-
trasting results. Isobutyl gave the best result, with 96% ee
(entry 3), but the trimethylsilyl-substituted nucleophile gave
a poor 6% ee (entry 11). Knowing that this Grignard re-
agent was biphasic in Et2O and the reaction gave almost
only racemates in THF, we tried to make the same reaction
with the Grignard reagent prepared in 2-methyltetrathydro-
furan (Me-THF). Schmalz[47] and co-workers have reported
that this solvent gave impressively high ee values in the 1,4-
addition of Grignard reagents to cyclohexenone, whereas
the THF gave racemates. Unfortunately, in our case, the ob-
tained ee was smaller than 5% by using a Grignard reagent
prepared in Me-THF and Et2O as the reaction solvent and
0% when using only Me-THF.
Table 9. Alkoxyde ImH+ screening.
Entry
ImH+
Ratio 1,2/1,4
Conv. [%][a]
ee [%][b]
1
2
3
4
5
6
7
8
3a
3b
3c
3d
3e
3 f
3g
3h
3i
0:100
0:100
0:100
0:100
0:100
0:100
10:90
0:100
0:100
45:55
43:57
87
91
85
98
42
78
99
99
99
99
99
68, (+) R
73, (+) R
74, (+) R
80, (+) R
37, (+) R
62, (+) R
0
13, (+) R
0
25, (+) R
6, (À) S
9
10
11
Finally, in contrast to the C2-symmetric NHC 2e, for
which a tert-butyl group could be added in a racemic way
(Table 5, entry 8), the ImH+ 3d gave 40% conversion to
only the 1,2-addition product and degradation products
3j
3k
[a] Conversion determined by GCMS analysis after 30 min. [b] Deter-
mined by chiral GC analysis (Lipodex E).
Chem. Eur. J. 2010, 16, 9890 – 9904
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9895

A. Alexakis et al.
after 2 h (Table 10, entry 10). The lack of reactivity of the
tert-butyl organomagnesium nucleophile was certainly due
to the tiny reactive pocket, which prevented the approach of
such a bulky group. Indeed, in the racemic way, without
NHC, the 1,4-addition product was obtained with more than
99% conversion, but with the NHC, many degradation
products and 30% of the 1,2-addition product were ob-
tained.
none was treated with methyl lithium, to give the corre-
sponding tertiary allylic alcohol 44 in good yield. This last
product was mixed with PCC to form the desired methyl cy-
cloheptenone in poor yield.
The substrates containing a linear substituent (S3–S4)
gave good ee values of around 70% (Table 11, entries 1–2).
This result was not surprising for the addition of MeMgBr.
It is well known that methyl nucleophiles are poorly reactive
The reaction temperature was an important factor. In all
cases, the chiral R2Cu species seemed to be active between
a range of temperatures. Above or below that temperature,
we observed competition in regioselectivity and byproduct
formation (Table 10, entries 4–5). Surprisingly, we could
expect that when slowing down the reaction rate by decreas-
ing the temperature (À15 to À308C), the ee should increase
but no variation was observed (entries 6–7). As the 1,4- and
1,2-additions were competitive, it seemed that when the
temperature was below À408C, the 1,4-addition rate de-
creased faster than the 1,2-addition, and the direct addition
was consequently promoted.
Now that the methodology was well established for the
A.C.A. reaction with 3-methylcyclohex-2-enone, we wanted
to extend the scope of our research to different aliphatic
substrates. We started therefore by synthesizing different
linear and branched trisubstituted cyclic enones (Scheme 4).
Table 11. Substrate variation.
Entry Substrate R3
T
Prod. Conv.
[%][a]
Yield
[%][b]
ee [%], con-
[8C]
fig.[c]
1
2
3
4
5
6
7
8
S3
S4
S5
S6
S7
S8
S8
S9
Me
Et
Et
Et
Et
Et
0
0
0
0
0
0
3
19
20
21
22
23
98
99
100
98
98
98
67
84
85
69
87
90
90
76
68, (À) S
69, (+) R
82, (+) R
81, (+) R
72, (+) S
38, (+) R
46, (+) R
82, (+) R
Et À10 23
98
99
Et 24
0
[a] Conversion determined by GCMS analysis after 30 min. [b] Isolated
yield. [c] Determined by chiral GC analysis (Lipodex E).
and difficult to add with good stereocontrol. A motivating
point was the good ee obtained with the less reactive iso-
phorone S5 (entry 3). By going to a bulkier isobutyl sub-
stituent, the enantio-discrimination worked better and the
ee values rose up to 81% (entry 4). Another promising
result was obtained with the addition of EtMgBr on the
phenyl substituted substrate S7 (72% ee; entry 5). A general
trend, as for C2-symmetric NHCs, is the observation of the
same facial selectivity as obtained with the NHC 3d.
Indeed, we always obtained the opposite major enantiomer
by inverting the nucleophile and the substituent on the sub-
strate. In contrast to the C2-symmetric 2a ImH+, the stereo-
selectivity for the ligand 3d is on the Si face.
The experiments realized upon varying the ring size gave
interesting results. In the case of S8, contrary to the six-
membered ring, the temperature played an important role.
Indeed, the ee rose up from 38% at 08C to 46% at À108C
(Table 11, entries 6 and 7). On the other hand, the seven-
membered substrate S9 gave ee values comparable to those
obtained with the six-membered rings (entry 8). It seems,
therefore, that the five-membered ring is too flat to allow a
good approach of the catalyst, which led to low enantiocon-
trol.
Scheme 4. Substrates synthesis.
The six-membered rings were synthesized by treating cyclo-
hexanedione with iodine, to form the cyclic ketoenol ether
S2 in good yield on a 60 g scale.[48] This intermediate was
then treated with the corresponding Grignard reagent to
give the desired substrate in good yield.[11] In this way, we
obtained the two substrates S3–S4, one b-branched S6 and
one aromatic substituted substrate S7. To test the activity of
our complex on a desactivated cyclic enone, we used the
commercially available isophorone S5.
The seven-membered ring was synthesized by applying a
procedure published by Dauben (Scheme 5).[49] Cyclohepte-
As the addition of EtMgBr on the phenylcyclohexenone
S7 gave a promising 72% ee (Table 11, entry 5), new aro-
matic substrates containing electro-attracting and withdraw-
ing aryl groups were synthesized. We used the same meth-
odology as described previously to observe the electronic ef-
fects on the conjugate addition reactions in the copper-cata-
Scheme 5. Seven-membered ring synthesis. PCC=pyridinium chlorochro-
mate.
9896
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 9890 – 9904

Formation of Quaternary Chiral Centers
Table 12. Aromatic substituted cyclohexenones.
FULL PAPER
As described previously, the conjugate addition on trisub-
stitued enones is a tricky reaction. But with the promising
results obtained with aliphatic Grignard reagents, we in-
creased the challenge and tried to insert aromatic Grignard
reagents. These reactions were not only difficult in term of
enantioselectivity, but also in term of regioselectivity. There
were only few papers concerning A.C.A. reactions describ-
ing the formation of quaternary carbon centers containing
an aromatic substituent. Most of the time, the aromatic
group was already on the substrate[13,50] or only phenyl and
p-anisyl[35,36,40] were inserted. It was only recently, that our
group proposed a viable method to insert a large panel of
aromatic aluminum species.[51] Therefore, we tried to apply
our methodology to the addition of aromatic organomagne-
sium reagents on aliphatic cyclohexenones.
Entry R- Ar- Cu
NHC
Prod. Conv. Yield Ratio ee
[mol%] [mol%]
[%][a] [%] 1,2/
1,4
[%][b]
1
Et S10
Et S10
Et S10 10
Et S11
Me S11
Et S12
Et S13
Me S13
3
5
0
7.5
13
4
7
4
4
7
25
25
25
26
27
28
29
30
97
99
40
84
90
99
99
99
–
–
–
72
–
72
86
51
23:77
98:2
99:1
0
2[c]
3
n.d.[d]
n.d.[d]
4
5
6
7
3
5
3
3
5
52:48 78
55:45 15
33:67 80
20:80 80
51:49 23
To begin the study, we first made a screening of solvents
and copper salts to find out the best conditions for the addi-
tion of the simplest aromatic, PhMgBr, on the 3-methylcy-
clohex-2-enone.
8
Relative to the methodology with alkyl Grignard reagents,
Et2O remained the best solvent for the reaction but Cu-
AHCTUNGTERNGUN(N acetate)2·H2O gave the best results in terms of regio- and
[a] Conversion determined by GCMS analysis after 60 min. [b] Deter-
mined by chiral GC analysis (Hydrodex-B-3P). [c] Reaction carried out
at À308C. [d] n.d.=not determined.
enantioselectivity (Table 13, entry 2). Interestingly, as for
lyzed 1,4-addition with different alkyl magnesium nucleo-
philes (Table 12).
Table 13. Solvent and copper salt screening for aromatic addition.
The obtained results were very interesting in terms of re-
gioselectivity. Indeed, the racemic addition of EtMgBr on
the ortho-substituted aromatic substrate S10 gave a race-
mate with a 1,2/1,4 ratio of 23:77. However, by adding the
chiral complex, the regioselectivity was inversed and the 1,2-
addition product became favored (Table 12, entries 1–3). As
the addition of more catalyst did not change the outcome of
the reaction, we concluded that the o-OMe-substituted sub-
strate S10 is too stericaly hindered to allow the approach of
the larger NHC–Cu complex, in contrast to the free copper
complex.
The p-OMe electron-donating substrate S11 gave 78% ee
for the addition of EtMgBr, but MeMgBr gave a poor
15% ee (Table 12, entries 4 and 5). The regioselectivity
around 50:50 in the two cases was mediocre. When going to
electron-withdrawing substrates, the outcomes of the reac-
tion change drastically. Indeed, the p-CF3 substrate S12 gave
a very good 80% ee with a regioselectivity of 33:67 in favor
of the 1,4-addition product (entry 6). The regioselectivity
was even better when starting from the chloro-substituted
adduct S13 (20:80) while keeping a high enantiocontrol
(entry 7).
Entry Solv.
1
CuX
Add. t Conv.
[min]
Ratio
ee
[%][a]
1,2/1,4 [%][b]
[c]
Cu
Cu
Cu
Cu
Cu
Cu
Cu
Cu
G
>99
99
99
5:95
12:88
18:82
30:70
22:78
65
70
70
66
50
2
ACHTUNGTRENNUNG
Et2O
3[d]
20
T
4
A
72
[c]
[c]
5
6
A
98
98
98
88
MtBE
20
20
(acetate)·H2O
(acac)2
99:1 n.d[e]
16:84
30:70
7
58
44
CH2Cl2
8
A
[a] Conversion determined by GCMS analysis after 30 min. [b] Deter-
mined by chiral GC (Hydrodex-B-3P). [c] acac=acetylacetonate.
À
[d] Carried out with ClÀ instead of PF6 as the counter-ion. [e] n.d.=not
determined.
the C2-symmetric NHC 2e, the variation of the counter-ion
had no effect on the enantioselectivity of the reaction (en-
tries 2 and 3). In this case, contrary to the work of Loh,[46]
MtBE gave worse results than Et2O in both regio- and enan-
tioselectivities (entries 1 vs. 5 and 2 vs. 6). CH2Cl2 gave a
lower regioselectivity, and the ee values were about 20%
lower than with their Et2O counterparts (entries 7 and 8). In
THF and tetrahydropyran (THP), which coordinates strong-
ly to the magnesium, we tested seven copper salts, but we
obtained almost no enantioselectivity for the reaction. Final-
ly, with toluene as the solvent, which was expected to bring
some p–p interactions with the Grignard reagent, the enan-
tioselectivity was poor and moderate regioselectivity was ob-
served at around 50:50 for the seven tested copper salts.
For these reactions, the regioselectivity depends on the
electronic effects of the aromatic substituents. Indeed, by
comparing S11, S12, and S13, one may observe that the elec-
À
trodonating OMe group (Table 12, entry 5), by decreasing
the reactivity of the b-position, gave worse regioselectivity
À
À
than the electron-withdrawing CF3 (entry 6) and Cl sub-
stituents (entry 7). These groups activate the b-position of
the cyclic substrate by decreasing its electronic density, and
in doing so, increased the electrophilicity of that position.
Therefore, the 1,4-addition was favored.
Chem. Eur. J. 2010, 16, 9890 – 9904
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9897

A. Alexakis et al.
Table 14. Aromatic Grignard reagent variation.
As for the alkyl Grignard reagents, the ee was not influ-
enced by the addition time, but the regioselectivity was
largely better when the addition was done slowly. By in-
creasing the addition time of the substrate, the regioselectiv-
ity improved, going from 22:78 to 12:88. This would imply
that the catalytic cycle of the aryl Cu–NHC complexes was
slower than for the alkyl counterpart. Indeed, by using alkyl
Cu–NHC complexes in the same range of addition time,
only the 1,4-addition product was observed. For aryl Cu–
NHC complexes, if the substrate was added too quickly, the
catalytic cycle was too slow to absorb it completely, and a
part of the substrate was, therefore, consumed by the free
Grignard reagent in excess in the reaction, giving the 1,2-ad-
dition product. This hypothesis was supported by the addi-
tion of an increased amount of copper to 10 mol%. Indeed,
the reaction with three times more catalyst gave an im-
provement in the regioselectivity going from around 37:63
to 22:78. In contrast, the ee decreased dramatically from 70
to 53% with formation of a brown aggregate. In fact, as the
Cu–NHC complex was poorly soluble in Et2O, the presence
of a too high concentration of complex favors the aggrega-
tion of an insoluble NHC/Cu heap of unknown ratio. This
could leave some free copper in solution, which by giving
only a racemic product, lowers the total enantioselectivity of
the reaction, but increases the regioselectivity.
The Cu/NHC ratio was also important. There was a
maxima at around 1:1.5. Below and above this ratio, the ee
diminished. Surprisingly, the regioselectivity in favor of the
conjugate addition was better when the formed complex
contains more than one NHC. Indeed, with a 1:1 ratio, the
regioselectivity was worse than for a 1:1.5 or 1:2 ratio. This
effect could be explained by the electron-donating effect of
the NHC. Although the steric hindrance increases in the
presence of more NHC, which should disadvantage the con-
jugate addition, the electron density brought by the extra
NHC on the reacting center, may accelerate the catalytic
cycle, and by the way, favor the 1,4-addition.
Entry R- Ar-
Cu
NHC
Prod. Conv. Ratio
ee [%],
[mol%] [mol%]
[%][a]
1,2/1,4 config.[b]
1
2
3
4
Me N1
Me N2
Me N3
Me N4
3
3
5
5
4
4
7.5
6.5
31
32
33
34
99
>99
67
12:88
100:0
70:30
77:23
70, (À)
n.d.
90
>99
70
[a] Conversion determined by GCMS analysis after 60 min. [b] Deter-
mined by SFC.
meta- counterpart N3, it gave even worse enantio- and re-
gioselectivity (entry 4). Perhaps, the electron-donating me-
thoxy group in the para-position brings more electron densi-
ty to the nucleophile, thus making it harder and more reac-
tive. This means that the catalytic cycle becomes even less
competitive against the free nucleophile, entering in a 1,2-
addition way. The same electron-donating group in the
meta-position brings no extra electronic density onto the
carbon nucleophile and the steric bulk is barely bigger than
its para-counterpart. Therefore, as the steric hindrance is
almost the same for N3 and N4, but N4 is more nucleophilic,
this could explain why there was more 1,2 product when
using para-methoxy-substituted phenyl Grignard reagents
then meta- ones.
To see the limitations of our methodology, we go one step
further and try the formation of all-carbon quaternary cen-
ters containing two aromatic rings. Different copper sources,
NHCs, and temperatures were applied. Also activations
with TMSCl was attempted, but for now, only the 1,2-addi-
tion product was obtained (Scheme 6).
These attempts have not brought the expected improve-
ments, because the 1,2/1,4 ratio could not be enhanced. We
could lower the Cu/NHC ratio to 1:1,3 by keeping a high ee
and good regioselectivity, which means that the best ratio
for alkyl additions was also the best for aryl ones.
New aromatic Grignard reagents N1–N4 were freshly syn-
thesized by adding dropwise an ethereal solution of aryl bro-
mide onto magnesium turnings. We then used these aromat-
ic Grignard regents on the two alkylcyclohexenones S1 and
S3.
To conclude this chapter, the alkoxy-ImH+ 3d gave the
best results in terms of conversions and enantiomeric excess-
es. This bidentate NHC family shows good to excellent apti-
tude in the enantioselective catalysis of the addition of alkyl
(up to 96%) and aryl nucleophiles (up to 90%). The regio-
selectivity for the aryl addition is the drawback of our meth-
odology, but it will certainly be improved in following re-
search.
The o-anisol derivative N2 was the worst one. The reac-
tion worked almost in a complete 1,2-addition way. No ee
was obtained, even if the amount of copper was increased to
5 mol% (Table 14, entry 2). The steric bulk issued from the
ortho-methoxy group certainly prevented the approach on
the stericaly hindered b-position. The m-anisole N3 gave an
excellent 90% ee, but the regioselectivity was strongly in
favor of the 1,2-addition product, even if 5 mol% of catalyst
was added (entry 5). Surprisingly, under the same conditions,
although the p-anisole N4 should be less bulky than its
Scheme 6. Quaternary centers containing two aromatic rings.
9898
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 9890 – 9904

Formation of Quaternary Chiral Centers
FULL PAPER
Applications in synthesis: With these different ImH+ com-
pounds in hand, we now possess the tools needed to catalyze
the A.C.A. reactions of Grignard reagents to different ali-
phatic and aromatic trisubstituted cyclic enones. This pro-
cess allows the isolation of valuable synthons with high
yields and ee values. This is why we focused on developing
synthetic applications to demonstrate the value of our meth-
odology.
ment of the synthetic possibilities of our methodology. To
improve the value of the addition/allylic trapping sequence,
we thought to go one step further and try to synthesize a bi-
cyclic compound by using our methodology (Scheme 7). We
As the 1,4-addition reaction goes through a magnesium
enolate step, it should be possible to trap it with an electro-
phile in a one-pot procedure, to open the door to easy a-
functionalization. The problem lies in the fact that magnesi-
um enolates, relative to lithium enolates,[52] are much less
nucleophilic. They have generally to be trapped by an oxo-
philic electrophile, such as Ac2O[53] or TMSOTf,[50] in a first
step, and then be treated by MeLi·LiBr to form the lithium
enolate, which becomes then highly reactive.[54]
We first tried to make the allylic trapping reaction in one
pot, by adding allyl iodide after the conjugate addition. Un-
fortunately, the magnesium enolate was not reactive enough
to attack it directly. Also by activating it with a Pd source
and warming the reaction to 458C for 5 h, no conversion
was obtained and only the protonated product was obtained.
After some optimization, we observed that adding a combi-
nation of HMPA/THF 1:1, as the cosolvent, permits the a-
allylation of the substrate with total retention of configura-
tion and good diastereoselectivity (Table 15, entry 3). The
same methodology was then used to a-alkylate the sub-
strate, by using methyl iodide (entry 4). In this case, the dia-
stereoselectivity of the reaction was not good, because of
the too small steric inducting difference between the methyl
and ethyl at the b-position for the small methyl iodide.
For the trapping of benzaldehyde (Table 15, entry 1) and
bromine (entry 2), it was not necessary to add HMPA be-
cause of the high electrophilicity of the reactant. The reac-
tions work well by keeping the high ee of the ethyl addition.
For benzaldehyde, the alcohol obtained was oxidized to
eliminate one asymmetric center, and by the way, make the
analysis easier.
Scheme 7. Addition/trapping/ring-closing metathesis sequence. HMPA=
hexamethylphosphoramide.
started, therefore, by adding a butenyl Grignard reagent
onto the 3-methylcyclohex-2-enone and trapping the formed
magnesium enolate with allyl iodide. The obtained dialkene
substrate was then cyclized by metathesis with Grubbs II
catalyst to form bicyclic six–seven membered rings with high
ee values and moderate cis/trans diastereoselectivity.
Recently, a report by Williams described the synthesis of
the 5,14-bis-epi-spirovibsanin A and other natural products
of the vibsanin family.[55] The first step was the racemic con-
jugate addition of 2-methylpent-2-enyl Grignard reagent to
3-methylcyclohex-2-enone. As it was exactly the field we
had investigated, we tried to reproduce the first step of this
synthesis but with asymmetric induction.
We applied our best conditions to complete the first step
enantioselectively and were pleased to see that our method-
ology allowed to us reach 86% ee for the first step of the
vibsane synthesis (Scheme 8).
The opportunity to insert different functionalities as
halide, alkyl, allyl, or benzylic alcohol allowed the enhance-
Table 15. Enolate trapping.
Entry E+
Reaction Prod. Conv. Yield ee [%][c] 1st/ d.r.[d]
time
[%][a] [%][b] 2nd diaste-
reomer
Scheme 8. Total synthesis of bis-epi-spirovibsanin A.
1
2
benzaldehyde[e] 30 min
37
38
35
36
100
100
>99
99
75
78
81
79
77/68
76/78
76[g]
41:59
32:67
70:30
59:41
Br2
allyl iodide
MeI
30 min
25 h
3[f]
4[f]
Conclusion
12 h
76[g]
[a] Conversion determined by GCMS analysis after 30 min. [b] Isolated yield.
[c] Determined by chiral GC analysis. [d] d.r.=diastereomeric ratio. [e] The
obtained alcohol is oxidized to eliminate diastereoisomers. [f] HMPA/THF
1:1, 408C. [g] ee of the first step.
We have showed that the two different families of NHC li-
gands we tested are viable ligands for the A.C.A. reactions
of Grignard reagents on cyclic trisubstituted enones.
Chem. Eur. J. 2010, 16, 9890 – 9904
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9899

A. Alexakis et al.
prior to use. Organic solutions were concentrated under reduced pressure
on a Bꢅchi rotary evaporator. 1H (400 MHz) and 13C (100 MHz) NMR
spectra were recorded in CDCl3, and chemical shifts (d) are given in ppm
relative to residual CHCl3. Evolution of the reactions was followed by a
GCMS Hewlett Packard (EI mode) HP6890–5973. An asterisk on GCMS
indicates the major peak. Optical rotations were measured at 208C in a
1 cm cell in the stated solvent; [a]D values are given in 10À1 8cm2 gÀ1 (con-
centration c given as g/100 mL). Enantiomeric excesses were determined
by chiral-GC (capillary column, 10 psi H2). Temperature programs are
described as follows: initial temperature (8C)—initial time (min)—tem-
perature gradient (8CminÀ1)—final temperature (8C); retention times
(tR) are given in min. All Grignard reagents except ethyl and methyl
magnesium bromide (Aldrich) were synthesized in Et2O by addition of
the corresponding bromide onto magnesium. Flash chromatography was
performed by using silica gel 32–63 mm, 60 ꢆ. The syntheses of starting
substrates and imidazolium catalysts are described in the Supporting In-
formation.
The C2-symmetric DPEDA-based family of NHC ligands
allowed to good results to be obtained in terms of conver-
sion and enantioselectivity. As these ligands had the possi-
bility to adapt somewhat the size of the reactive pocket to
the size of the nucleophile, they could catalyze the 1,4-addi-
tion of a bulky tert-butyl group on 3-methylcyclohex-2-
enone with total 1,4 regioselectivity but with no enantiose-
lectivity. Despite the fact that these kind of ligands were dif-
ficult to synthesize with good yield, they allowed the 1,4-ad-
dition of linear Grignard reagents to the 3-methylcyclohex-
2-enone with high yields and good ee values lying between
70–80% with perfect regioselectivity. By switching to
branched nucleophiles (c-pentyl and isopropyl), the ee
values dropped to between 40 to 60%, but with perfect re-
gioselectivity.
The alkoxy-substituted ligand family gave better results
for the 1,4-addition of linear Grignard reagents (ethyl, butyl,
and butenyl) to 3-methylcyclohex-2-enone with ee values be-
tween 80 and 90%. The use of branched nucleophiles (iso-
butyl, isopropyl, c-hexyl, and c-pentyl) also gave excellent
results in term of enantioselectivities lying between 78 and
96% with perfect regioselectivities. We were pleased to see
that our methodology was not substrate dependant. Indeed,
the variation of the alkyl group on the cyclic substrate
(ethyl, butenyl, isophorone, isobutyl, and phenyl) gave very
good results keeping the ee values between 68 and 90%. By
changing the ring size to a seven-membered ring, the ee
values were comparable to the six-membered ring ones, but
by decreasing the ring size to five carbon atoms, the ee
values dropped to 40%. We also used aromatic substituted
six-membered enones, which produced very good ee values
of around 80%. Unfortunately in these cases, the regioselec-
tivity of the reaction decreased dramatically. The presence
of an electron-withdrawing aromatic group gave better re-
giocontrol than an electron-donating substituent, but the re-
sults were between 50:50 and 20:80 for the 1,2/1,4-addition.
Finally, the addition of aromatic Grignard reagents on alkyl-
substituted cyclic enones gave very good ee values of up to
90% for the addition of m-OMe phenyl Grignard with often
poor 1,4 regioselectivity.
Typical procedure for 1,4-addition reactions: A flame-dried Schlenk tube
was charged with copper salt (3.0 mol%) and the chiral ImH+ salt (4.0
mol%). The system was flushed under N2 and dry Et2O (2.5 mL) was
added. The mixture was cooled down to the desired reaction temperature
in an ethanol cold bath. The Grignard reagent (1.2 equiv) in Et2O was
added dropwise to the solution for 5 min. A solution of the substituted
cyclohexenone in Et2O (8 mL) was then added dropwise to the solution
at the desired low reaction temperature for 15 min and the solution was
stirred for another 30 min. The reaction was hydrolyzed at the reaction
temperature by addition of HCl 1m (30 mL) (or MeOH if the reaction
temperature lies under À208C) and the aqueous layer was separated and
extracted further with diethyl ether (3ꢇ10 mL). The combined organic
layers were dried on MgSO4, filtrated, and concentrated in vacuo to give
an oily residue. This crude product was purified by flash chromatography
on a silica column with pentane/Et2O 10:1 to give the pure product.
(R)-3,3-Ethylmethylcyclohexanone (2):[11] 1H NMR (400 MHz, CDCl3):
d=2.26 (t, 2H, J=6.6 Hz), 2.16 (d, 1H, J=13.6 Hz), 2.08 (d, 1H, J=
13.6 Hz), 1.88–1.81 (m, 2H), 1.64–1.48 (2 m, 2H), 1.33 (q, 2H, J=
7.3 Hz), 0.88 (s, 3H), 0.83 ppm (t, 3H, J=7.6 Hz); 13C NMR (100 MHz,
CDCl3): d=212.8, 53.7, 41.4, 39.0, 35.7, 34.3, 24.7, 22.5, 8.1 ppm; [a]D20
=
+4.74 (c=1.64 in CHCl3); ee=80% R (the absolute configuration was
assigned in analogy with the literature;[11] enantiomeric excess was mea-
sured by chiral GC analysis (lipodex E, 75–23–20–170—5, v=40 cmsÀ1):
tR1 =15.5 (R), tR2 =19.9 min (S)).
(S)-3,3-Ethylmethylcyclohexanone (3):[11] 1H NMR (400 MHz, CDCl3):
see data for compound 2; [a]D20 =À4.50 (c=1.70 in CHCl3); ee=68% S
(absolute configuration was assigned in analogy with the literature;[11]
enantiomeric excess was measured by chiral GC analysis (lipodex E, iso-
therm 758C, v=40 cmsÀ1): tR1 =16.3 (R), tR2 =19.5 min (S)).
(S)-3,3-Butylmethylcyclohexanone (4):[56] 1H NMR (400 MHz, CDCl3):
d=2.19 (t, 2H), 2.11 (q, 2H), 1.78 (quint., 2H), 1.55–1.41 (m, 2H), 1.18
(m, 6H), 0.83 ppm (m, 6H); 13C NMR (100 MHz, CDCl3): d=212.2, 53.8,
41.3, 41.0, 38.5, 35.8, 25.5, 25.1, 23.4, 22.1, 14.0 ppm; [a]2D0 =À12.3 (c=
15.4 in CHCl3); ee=72% S (absolute configuration was assigned in anal-
ogy with the literature;[11] enantiomeric excess was measured by chiral
We finally developed synthetic applications, which in-
volved trapping the formed magnesium enolates with MeI,
benzaldehyde, bromine, and allyl iodide. These enolate trap-
pings allowed the functionalization of the cyclic substrates
at the a-position with many different functionalities. Our
methodology was also successfully applied to the first step
of the total synthesis of bis-epi-spirovibsanin A with 86% ee
for the addition of 2-methylpent-2-enyl Grignard reagent to
the 3-methylcyclohex-2-enone.
GC analysis (lipodex E, 80–0–1–170–5; v=30 cmsÀ1): tR1 =25.7 (R), tR2
27.4 min (S)).
=
(R)-3-(3-Butenyl)-3-methylcyclohexanone (5):[11] 1H NMR (400 MHz,
CDCl3): d=5.82–5.72 (m, 1H), 5.01 (d, 1H, J=16.9 Hz), 4.92 (d, 1H, J=
10.1 Hz), 2.28–2.24 (t, 2H, J=6.8 Hz), 2.20–2.08 (q, 2H, J=13.6 Hz),
2.03–1.97 (m, 2H), 1.88–1.82 (m, 2H), 1.66–1.51 (m, 2H), 1.36–1.32 (t,
2H, J=8.6 Hz), 0.92 ppm (s, 3H); 13C NMR (125 MHz, CDCl3): 212.2,
138.9, 114.5, 53.8, 41.1, 41.0, 38.6, 35.9, 27.9, 25.0, 22.2 ppm; [a]2D0 =d=
+1.9 (c=1.9 in CHCl3); ee=73% R (absolute configuration was assigned
in analogy with the literature;[11] enantiomeric excess was measured by
Experimental Section
chiral GC analysis (Hydrodex B-3P, isotherm 1408C): tR1 =10.6 (S), tR2
10.9 min (R)).
=
General procedures: All reactions were conducted under an inert atmos-
phere. Unless otherwise stated, all reagents were purchased from com-
mercial suppliers and used without further purification. All solvents em-
ployed in the reactions were dried on alumina columns and degassed
(S)-3,3-c-Pentylmethylcyclohexanone (6): 1H NMR (400 MHz, CDCl3):
d=2.23–2.17 (m, 3H), 2.06–2.02 (m, 1H), 1.85–1.69 (m, 3H), 1.61–1.50
(m, 8H), 1.17–1.13 (m, 2H), 0.79 ppm (s, 3H); 13C NMR (100 MHz,
CDCl3): d=212.7, 52.4, 49.6, 41.2, 40.6, 35.1, 26.4, 25.6, 22.1, 20.9 ppm;
9900
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 9890 – 9904

Formation of Quaternary Chiral Centers
FULL PAPER
ee=56% S (absolute configuration was assigned in analogy with the liter-
ature;[11] enantiomeric excess was measured by chiral GC analysis (Lipo-
dex E, 120–1–20–170–5, v=40 cmsÀ1): tR1 =13.1 (R), tR2 =14.3 min (S)).
(S)-3,3-Isopropylmethylcyclohexanone (7):[10] 1H NMR (400 MHz,
CDCl3): d=2.32–2.22 (m, 3H), 2.11 (m 1H), 1.98–1.77 (m, 2H), 1.64 (m,
2H), 1.51 (sept., 1H), 0.85–0.86 (d, 6H, J=1.7 Hz), 0.80 ppm (s, 3H);
13C NMR (100 MHz, CDCl3): d=212.9, 52.0, 41.2, 36.4, 34.2, 22.1, 19.9,
17.1, 16.9 ppm; ee=70% S (absolute configuration was assigned in analo-
gy with the literature;[11] enantiomeric excess was measured by chiral GC
analysis (Lipodex E, 60–0–1–110–20–170–5, v=45 cmsÀ1): tR1 =21.4 (R),
(EI-MS): m/z: calcd for C12H20O: 180.151415 [M]+; found: 180.151440;
[a]2D0 =+7.11 (c=1.70 in CHCl3); ee=86% R (absolute configuration was
assigned in analogy with the literature;[11] enantiomeric excess was mea-
sured by chiral GC analysis (Lipodex E, isotherm 1258C, v=60 cmsÀ1):
tR1 =6.9 (R), tR2 =7.4 min (S)).
(R)-3,3-c-Hexylmethylcyclohexanone (17): 1H NMR (400 MHz, CDCl3):
d=2.31–2.12 (m, 3H), 2.02–1.99 (m, 1H), 1.85–1.78 (m, 1H), 1.77–1.62
(m, 5H), 1.60–1.56 (m, 2H), 1.52–1.46 (m, 1H), 1.17–1.00 (m, 4H), 0.93–
0.83 (m, 2H), 0.75 ppm (s, 3H); 13C NMR (100 MHz, CDCl3): d=212.8,
51.9, 46.6, 41.1, 41.0, 34.1, 27.0, 27.0, 26.8, 26.6, 26.6, 21.9, 21.0 ppm; MS
(EI): m/z: 194 (<1), 112 (12), 111 (90), 110 (17), 97 (39), 83 (20), 82 (17),
69 (17), 67 (14), 55 (100); HRMS (EI-MS): m/z: calcd for C13H22O:
194.167066 [M]+; found: 194.166980; [a]D20 =+4.55 (c=1.70 in CHCl3);
ee=79% R (absolute configuration was assigned in analogy with the lit-
erature;[11] enantiomeric excess was measured by chiral GC analysis (Lip-
odex E, isotherm 1258C, v=60 cmsÀ1): tR1 =10.9 (R), tR2 =11.2 min (S)).
tR2 =24.6 min (S)).
(S)-3-Ethyl-3,5,5-trimethylcyclohexan-1-one (9):[57] 1H NMR (400 MHz,
CDCl3): d=2.19–2.06 (m, 4H), 1.60 (d, 1H, J=14.2 Hz), 1.49 (d, 1H, J=
14.2 Hz), 1.43–1.24 (m, 2H), 1.02 (m, 6H), 0.96 (s, 3H), 0.85–0.82 ppm (t,
3H, J=7.6 Hz); 13C NMR (100 MHz, CDCl3): 212.9, 54.5, 53.0, 48.8, 39.0,
37.3, 36.3, 32.5, 30.9, 27.0, 8.4 ppm; [a]2D0 =À1.89 (c=1.70 in CHCl3); ee=
71% S (absolute configuration was assigned in analogy with the litera-
ture;[11] enantiomeric excess was measured by chiral GC analysis (Chirasil
DEX-CB, 60–110–2–170–5, v=40 cmsÀ1): tR1 =133.7 (R), tR2 =134.0 min
(S)).
3,3-Methyl-[(trimethylsilyl)methyl]cyclohexanone
(18):
1H NMR
(400 MHz, CDCl3):d=2.22–2.09 (m, 4H), 1.83–1.77 (m, 2H), 1.61–1.56
(m, 2H), 0.95 (s, 3H), 0.68 (s, 2H), 0.00 ppm (s, 9H); 13C NMR
(100 MHz, CDCl3): 212.5, 57.3, 41.2, 39.6, 32.3, 29.1, 28.4, 22.8, 1.26; MS-
EI: m/z: 197, 183, 170, 155, 130, 115, 75, 73*, 55; ee=0% (enantiomeric
excess was measured by chiral GC analysis (Lipodex E, isotherm 1008C,
v=45 cmsÀ1): tR1 =7.0, tR2 =7.3 min).
(R)-3,3-Butenyl-ethylcyclohexanone (19): 1H NMR (400 MHz, CDCl3):
d=5.82–5.72 (ddt, 1H, J=6.6, 10.1, 13.1 Hz), 4.97 (m, 2H), 2.28–2.25 (m,
2H), 2.15 (s, 2H), 1.97–1.90 (m, 2H), 1.86–1.80 (m, 2H), 1.60–1.57 (m,
2H), 1.34–1.28 (m, 4H), 0.81–0.77 ppm (t, 3H, J=7.3 Hz); 13C NMR
(100 MHz, CDCl3): d=212.6, 139.0, 114.6, 52.0, 41.3, 41.1, 36.1, 33.6, 29.5,
27.5, 21.8, 7.5 ppm; [a]2D0 =+7.6 (c=1.70 in CHCl3); ee=69% R (abso-
lute configuration was assigned in analogy with the literature;[11] enantio-
meric excess was measured by chiral GC (Hydrodex-B-3P isotherm
1358C, v=40 cmsÀ1): tR1 =13.1 (S), tR2 =13.3 min (R)).
3,3-tert-Butylmethylcyclohexenone (10):[58] 1H NMR (300 MHz, CDCl3):
d=2.35–2.31 (m, 2H), 2.15–2.11 (m, 2H), 2.0–1.74 (m, 4H), 0.92 (s, 9H),
0.86 ppm (s, 3H); MS (EI): m/z: 168, 112, 97*, 83, 69, 55; ee=0% (enan-
tiomeric excess was measured by chiral GC analysis (Hydrodex B6 tert-
butyldimethylsilyl (TBDMS), 80–0–1–95–0–170–5, v=45 cmsÀ1): tR1
10.3, tR2 =10.7 min).
=
(S)-3,3-Phenylmethylcyclohexanone (11):[59] 1H NMR (400 MHz, CDCl3):
d=7.33–7.32 (m, 4H), 7.22–7.19 (m, 1H), 2.89 (d, 1H, J=14.2), 2.45 (d,
1H, J=14.2), 2.33–2.30 (t, 2H), 2.22–2.16 (m, 1H), 1.96–1.84 (m, 2H),
1.71–1.63 (m, 1H), 1.33 ppm (s, 3H); 13C NMR (100 MHz, CDCl3): d=
211.5, 147.5, 128.6, 126.3, 125.6, 53.2, 42.9, 40.9, 38.0, 29.9, 22.1 ppm; ee=
70% S (absolute configuration was assigned in analogy with the litera-
ture;[11] enantiomeric excess was measured by chiral GC analysis (Hydro-
(R)-3-Ethyl-3,5,5-trimethylcyclohexan-1-one (20):[57] 1H NMR (400 MHz,
CDCl3): see data for compound (9); [a]2D0 =+8.52 (c=1.70 in CHCl3);
ee=82% R (absolute configuration was assigned in analogy with the lit-
erature;[11] enantiomeric excess was measured by chiral GC analysis
dex-B-3P, isotherm 1408C, v=48 cmsÀ1):
(S)).
tR1 =23.1 (R), tR2 =23.8 min
(R)-3,3-Butylmethylcyclohexanone (12):[56] 1H NMR (400 MHz, CDCl3):
see data for compound 4; [a]D20 =+4.74 (c=1.64 in CHCl3); ee=78% R
(absolute configuration was assigned in analogy with the literature;[11]
enantiomeric excess was measured by chiral GC analysis (lipodex E, 80–
0–1–170–5, v=40 cmsÀ1): tR1 =26.5 (R), tR2 =27.2 min (S)).
(S)-3-(3-Butenyl)-3-methylcyclohexanone (13):[11] 1H NMR (400 MHz,
CDCl3): see data for compound (5); [a]2D0 =À2.45 (c=1.71 in CHCl3),
ee=90% S (absolute configuration was assigned in analogy with the liter-
ature;[11] enantiomeric excess was measured by chiral GC analysis (Hy-
drodex B-3P, isotherm 1358C): tR1 =6.5 (S), tR2 =6.7 min (R)).
(S)-3,3-Isobutylmethylcyclohexanone (14):[11] 1H NMR (400 MHz,
CDCl3): d=2.29–2.26 (m, 2H), 2.20 (d, 1H, J=13.4 Hz), 2.10 (d, 1H, J=
13.4 Hz), 1.92–1.83 (m, 2H), 1.74–1.64 (m, 2H), 1.60–1.56 (m, 1H), 1.26–
1.17 (m, 2H), 0.95 (s, 3H), 0.93 ppm (m, 6H); 13C NMR (100 MHz,
CDCl3): d=212.7, 54.6, 51.2, 41.3, 39.7, 36.8, 25.8, 25.7, 25.6, 24.2,
22.5 ppm; IR (neat): n˜ =2956, 1716, 1467 cmÀ1; MS (EI): m/z: 153 (2),
125 (9), 112 (9), 111 (100), 110 (3), 107 (2), 98 (2), 97 (9), 95 (7), 93 (2),
85 (1), 84 (2), 83 (22), 82 (4), 81 (2), 79 (2), 77 (1), 71 (2), 70 (6), 69 (25),
68 (3), 67 (6), 65 (1), 58 (1), 57 (5), 56 (15), 55 (98), 54 (2), 53 (5);
HRMS (ESI-MS): m/z: calcd for C11H20ONa: 191.1406364 [M+Na]+;
found: 191.1408390; [a]2D0 =+2.19 (c=1.73 in CHCl3); ee=96% S (abso-
lute configuration was assigned in analogy with the literature;[11] enantio-
meric excess was measured by chiral GC analysis (lipodex E, isotherm
808C, v=60 cmsÀ1): tR1 =12.9 (S), tR2 =15.8 min (R)).
(Chirasil DEX-CB, 60–110–2–170–5, v=40 cmsÀ1): tR1 =133.8 (R), tR2
134.5 min (S)).
=
(R)-3,3-Ethyl-isobutylcyclohexanone (21): 1H NMR (400 MHz, CDCl3):
d=2.26 (t, 2H, J=6.8 Hz), 2.18 (d, 1H, J=13.7), 2.14 (d, 1H, J=13.7),
1.91–1.75 (m, 2H), 1.70–1.63 (m, 1H), 1.60 (t, 2H, J=12.4 Hz), 1.37 (d,
1H, J=7.6 Hz), 1.34 (d, 1H, J=7.3 Hz), 1.19–1.15 (m, 2H), 0.92–0.85 (m,
6H), 0.79 ppm (t, 3H, J=7.3 Hz); 13C NMR (100 MHz, CDCl3): d=
212.9, 52.7, 46.0, 42.0, 41.3, 34.0, 29.8, 25.6, 25.6, 23.7, 21.9, 7.8 ppm; MS
(EI): m/z: 182 (<1), 153 (32), 125 (66), 97 (69), 95 (11), 83 (22), 70 (10),
69 (28), 67 (11), 55 (100); HRMS (EI-MS): m/z: calcd for C12H22O:
182.167066 [M]+; found: 182.167160; [a]D20 =+4.43 (c=1.70 in CHCl3);
ee=81% R (absolute configuration was assigned in analogy with the lit-
erature;[11] enantiomeric excess was measured by chiral GC analysis (lipo-
dex E, isotherm 878C, v=40 cmsÀ1): tR1 =25.8 (R), tR2 =27.9 min (S)).
(S)-3,3-Ethyl-phenylcyclohexanone (22): 1H NMR (400 MHz, CDCl3):
d=7.33–7.29 (m, 2H), 7.27–7.24 (m, 2H), 7.21–7.17 (m, 1H), 2.94 (d, 1H,
J=14.4 Hz), 2.39 (d, 1H, J=14.4 Hz), 2.31–2.27 (m, 2H), 2.20–1.95 (m,
2H), 1.87–1.53 (m, 4H), 0.62–0.58 ppm (t, 3H, J=7.3 Hz); 13C NMR
(100 MHz, CDCl3): d=211.8, 146.0, 128.6, 126.7, 126.3, 50.8, 46.7, 41.3,
36.5, 35.9, 21.8, 8.2 ppm; IR (neat): nꢈ=>=2993, 2961, 2873, 1710,
1444, 1227 cmÀ1; MS (EI): m/z: 202 (12), 174 (11), 173 (72), 145 (13), 131
(13), 117 (15), 115 (12), 103 (10), 91 (28), 69 (13), 55 (100); [a]2D0 =+58.3
(c=1.70 in CHCl3); ee=72% S (absolute configuration was assigned in
analogy with the literature;[11] enantiomeric excess was measured by
(R)-3,3-Isopropylmethylcyclohexanone (15):[10] 1H NMR (400 MHz,
CDCl3): see data for compound 7; [a]2D0 =+12.2 (c=1.70 in CHCl3); ee=
77% R (absolute configuration was assigned in analogy with the litera-
ture;[11] enantiomeric excess was measured by chiral GC analysis (Lipo-
dex E 60–1–130, v=49 cmsÀ1): tR1 =24.3 (R), tR2 =26.3 min (S)).
(R)-3,3-c-Pentylmethylcyclohexanone (16): 1H NMR (400 MHz, CDCl3):
see data for compound (6); MS (EI): m/z: 180 (2), 122 (14), 112 (10), 111
(77), 110 (18), 97 (18), 83 (14), 82 (13), 69 (25), 67 (17), 55 (100); HRMS
chiral GC analysis (Chirasil DEX-CB 135–0–1–160, v=40 cmsÀ1): tR1
18.5 (S), tR2 =19.0 min (R)).
=
(R)-3,3-Ethylmethylcyclopentanone (23):[60] 1H NMR (400 MHz, CDCl3):
d=2.30–2.26 (m, 2H), 2.07 (d, 1H, J=6.00 Hz), 2.00 (d, 1H, J=6.00 Hz),
1.83–1.70 (m, 2H), 1.43 (q, 2H, J=7.6 Hz), 1.03 (s, 3H), 0.90 ppm (t, 3H,
J=7.3 Hz); 13C NMR (100 MHz, CDCl3): d=220.4, 52.0, 39.9, 37.0, 34.9,
34.1, 24.7, 9.2 ppm; [a]2D0 =+22.59 (c=1.70 in CHCl3); ee=46% R (abso-
lute configuration was assigned in analogy with the literature;[11] enantio-
Chem. Eur. J. 2010, 16, 9890 – 9904
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9901

A. Alexakis et al.
meric excess was measured by chiral GC analysis (Lipodex E isotherm
(R)-3,3-Ethyl-phenylcyclohexanone (32):[36] 1H NMR (400 MHz, CDCl3):
see data for compound 22; [a]2D0 =À22.2 (c=1.69 in CHCl3); ee=34% R
(absolute configuration was assigned in analogy with the literature;[11,36]
enantiomeric excess was measured by chiral GC analysis (Chirasil DEX-
CB, 135–0–160–20–170–5; v=40 cmsÀ1): tR1 =18.2 (S), tR2 =18.6 min (R)).
608C, v=40 cmsÀ1): tR1 =15.2 (R), tR2 =18.0 min (S)).
(R)-3,3-Ethylmethylcycloheptanone (24):[36] 1H NMR (400 MHz, CDCl3):
d=2.53 (d, 1H, J=12.2 Hz), 2.46–2.42 (m, 2H), 2.36 (d, 1H, J=12.2 Hz),
1.83–1.51 (m, 6H), 1.41–1.23 (m, 2H), 0.91 (s, 3H), 0.87 ppm (t, 3H, J=
7.6 Hz); 13C NMR (100 MHz, CDCl3): d=214.4, 54.4, 44.0, 42.1, 35.3,
34.9, 25.4, 24.7, 24.2, 8.0 ppm; IR (neat): n˜2966, 2932, 1736, 1797,
1457 cmÀ1; MS (EI): m/z: 154 (2), 125 (23), 112 (10), 97 (40), 96 (50), 86
(47), 84 m (19), 84 (74), 83 (18), 81 (15), 70 (17), 69 (21), 56 (15), 55
(100), 49 (16), 47 (20); HRMS (ESI-MS): m/z: calcd for C10H19O:
155.1439; found: 155.1436 [M+H]+; [a]D20 =+10.33 (c=1.41 in CHCl3);
ee=82% R (absolute configuration was assigned in analogy with the lit-
erature;[36] enantiomeric excess was measured by chiral GC (Lipodex E
isotherm 608C, v=50 cmsÀ1): tR1 =24.3 (R), tR2 =25.7 min (S)).
(R)-3-(3-Methoxyphenyl)-3-methylcyclohexanone
(33):[51]
1H NMR
(300 MHz, CDCl3): d=7.27–7.22 (m, 1H), 7.62–6.86 (m, 2H), 6.77–6.73
(dd, 1H, J=8.2, 2.4 Hz), 3.80 (s, 3H), 2.86 (d, 1H, J=14.1 Hz), 2.42 (d,
1H, J=14.2 Hz), 2.31 (t, 2H, J=6.7 Hz), 2.21–2.13 (m, 1H), 1.94–1.81
(m, 2H), 1.74–1.63 (m, 1H), 1.31 ppm (s, 3H); [a]2D0 =À55.3 (c=0.8 in
CHCl3); ee=90% R (absolute configuration was assigned in analogy
with the literature;[11] enantiomeric excess was measured by chiral SFC
(S5-AD, 5%-2–1–15%, v=2 mLminÀ1): tR1 =4.7 (R), tR2 =7.1 min (S)).
(R)-3-(4-Methoxyphenyl)-3-methylcyclohexanone
(34):[36]
1H NMR
(S)-3-Ethyl-3-(4-methoxyphenyl)cyclohexanone
(26):
1H NMR
(400 MHz, CDCl3): see data for compound (27); [a]2D0 =À42.5 (c=1.5 in
CHCl3): ee=70% R (absolute configuration was assigned in analogy with
the literature;[11,36] enantiomeric excess was measured by chiral SFC (S2-
OD, 2%-2–1–15%, v=0 2 mLminÀ1): tR1 =5.8 (R), tR2 =6.3 min (S)).
(400 MHz, CDCl3): d=7.17–7.15 (m, 2H), 6.86–6.84 (2H, m), 3.79 (s,
3H), 2.88 (d, 2H, J=14.2 Hz), 2.37 (d, 2H, J=14.3 Hz), 2.30–2.26 (m,
2H), 2.16–1.91 (m, 2H), 1.86–1.78 (m, 2H), 1.76–1.52 (m, 2H), 0.59 ppm
(t, 3H, J=7.6 Hz); 13C NMR (100 MHz, CDCl3): d=211.9, 157.8, 136.9,
127.7, 113.8, 55.3, 50.9, 46.1, 41.2, 36.5, 36.1, 21.7, 8.1 ppm; [a]2D0d=+92.7
(c=1.9 in CHCl3); ee=78% S (absolute configuration was assigned in
analogy with the literature;[11] enantiomeric excess was measured by
chiral SFC (S2-AS 5%-2–1–15%, v=2 mLsÀ1): tR1 =4.3 (R), tR2 =4.7 min
(S)).
2-Allyl-3,3-ethylmethylcyclohexanone (35): Mixture of two diastereomers
70:30 (175 mg, 97%); 13C NMR (100 MHz, CDCl3): d=213.2, 212.9,
137.9, 137.7, 115.4, 115.3, 61.5, 58.3, 42.2, 41.7, 41.4, 40.9, 34.7, 34.4, 33.4,
28.8, 28.2, 27.0, 24.8, 22.9, 22.6, 20.9, 7.9, 7.4 ppm; ee=76% R; (absolute
configuration was assigned in analogy with the literature;[11] enantiomeric
excess was measured by chiral GC analysis on the ethyl addition (Lipo-
dex E, iso 758C, v=45 cmsÀ1): tR1 =11.0 (R), tR2 =13.8 min (S)); GCMS
(80–1–20–270–6; v=45 cmsÀ1): m/z: 6.91 (major) (180, 151, 109, 96, 81,
67, 55*), 6.95 (180, 151, 109, 96, 81, 67, 55*); HRMS (EI-MS): m/z:
calcd: 180.1512; found: 180.1514 (accuracy=À0.2).
3-Ethyl-2,3-dimethylcyclohexanone (36): Mixture of two diastereomers
59:41 (122 mg, 79%); 1H NMR (400 MHz, CDCl3): d=2.41–2.32 (m,
2.8H), 2.31–2.22 (m, 2.1H), 1.97–1.69 (m, 5.5H), 1.52–1.45 (m, 1.6H),
1.43–1.35 (m, 2.1H), 1.31–1.21 (m, 1H), 1.19–1.15 (m, 1H), 0.98–0.96 (m,
4.1H), 0.93–0.92 (m, 2.9H), 0.90–0.86 (m, 3.1H), 0.79–0.75 (m, 2.2H),
0.73 ppm (s, 3H); 13C NMR (100 MHz, CDCl3): d=214.7, 214.4, 55.6,
52.1, 41.6, 41.1, 41.0, 40.7, 35.0, 34.4, 33.7, 25.5, 25.1, 22.5, 22.2, 20.0, 8.8,
8.3, 7.9, 7.4 ppm; ee=76% R (absolute configuration was assigned in
analogy with the literature;[11] enantiomeric excess was measured by
chiral GC analysis on the ethyl addition (Lipodex E, iso 758C, v=
45 cmsÀ1): tR1 =11.5 (R), tR2 =14.4 min (S)).
(S)-3-Methyl-3-(4-methoxyphenyl)cyclohexanone
(27):[36]
1H NMR
(400 MHz, CDCl3): d=7.24–7.22 (m, 2H), 6.87–84 (m, 2H), 3.79 (s, 3H),
2.85 (d, 1H, J=14.2 Hz), 2.41 (d, 1H, J=14.2 Hz), 2.20–2.11 (m, 2H),
1.92–1.82 (m, 2H), 1.71–1.61 (m, 2H), 1.30 ppm (s, 3H); 13C NMR
(75 MHz, CDCl3): d=126.8, 116.2, 115.0, 113.9, 55.4, 53.4, 42.5, 40.9, 38.2,
30.2, 22.2 ppm; [a]2D0 =+63.6 (c=1.3 in CHCl3); ee=15% S (absolute
configuration was assigned in analogy with the literature;[11, 36] enantio-
meric excess was measured by chiral SFC (S2-OD 2%-2–1–15% v=
2 mLminÀ1): tR1 =5.4 (R), tR2 =6.0 min (S)).
(S)-3-Ethyl-3-[4-(trifluoromethyl)phenyl]cyclohexanone (28): 1H NMR
(400 MHz, CDCl3): d=7.58–7.56 (m, 2H), 7.39–7.36 (m, 2H), 2.91 (d,
1H, J=14.3 Hz), 2.44 (d, 1H, J=14.4 Hz), 2.34–2.28 (m, 2H), 2.22–2.16
(m, 1H), 2.06–1.97 (m, 1H), 1.92–1.73 (m, 1H), 1.72–1.64 (m, 2H), 1.62–
1.47 (m, 1H), 0.60 ppm (t, 3H, J=7.4 Hz); 13C NMR (100 MHz, CDCl3):
d=210.9, 149.3, 127.1, 125.6, 125.6, 125.5, 125.5, 50.5, 46.9, 41.1, 36.6,
35.9, 21.6, 8.0 ppm; [a]2D0 =+41.2 (c=2.1 in CHCl3); ee=80% S (absolute
configuration was assigned in analogy with the literature;[11] enantiomeric
excess was measured by chiral GC analysis (Chirasil-Dex-CB 100–0–1–
170—10, v=40 cmsÀ1): tR1 =50.7 (S), tR2 =53.2 min (R)).
2-Benzoyl-3,3-ethylmethylcyclohexanone (37): Mixture of two diastereo-
mers 41:59 (132 mg, 72%); 1H NMR (400 MHz, CDCl3): d=8.00–7.94
(m, 2H), 7.58–7.54 (m, 1H), 7.47–7.43 (m, 2H), 4.39 (m, 1H), 2.83–2.72
(m, 1H), 2.52–2.44 (m, 1H), 2.40–2.19 (m, 2H), 2.07–1.95 (m, 1H), 1.92–
1.76 (m, 1H), 1.56–1.31 (m, 3H), 1.25 (s, 1H), 0.97 (s, 1H), 0.92 (s, 1H),
0.86 (t, 2H, J=7.3 Hz), 0.70–0.67 ppm (t, 1H, J=7.6 Hz); 13C NMR
(100 MHz, CDCl3): d=208.6, 208.3, 197.0, 197.0, 138.7, 138.4, 133.4,
133.4, 128.8, 128.8, 128.7, 128.7, 69.1, 68.0, 43.5, 42.6, 39.3, 38.9, 32.4, 31.9,
31.9, 31.1, 23.6, 23.2, 22.3, 21.7, 7.9, 7.6 ppm; ee=77% R (1st diastereo-
mer), ee 68% R (2nd diastereomer) (absolute configuration was assigned
in analogy with the literature;[11] enantiomeric excess was measured by
chiral GC analysis (Chirasil Dex-CB, 110–0–1–170–5; v=40 cmsÀ1): 1st
diastereomer: tR1 =50.7 (R), tR2 =51.1 (S); 2nd diastereomer: tR1 =53.2,
(S)-3-(3,4-Dichlorophenyl)-3-ethylcyclohexanone
(29):
1H NMR
(400 MHz, CDCl3): d=7.43–7.38 (m, 2H), 7.15–7.11 (m, 1H), 2.88 (d,
1H, J=14.2 Hz), 2.47 (d, 1H, J=14.2 Hz), 2.39–2.32 (m, 2H), 2.21–1.85
(m, 3H), 1.80–1.61 (m, 3H), 0.66 ppm (t, 3H, J=7.4 Hz); 13C NMR
(100 MHz, CDCl3): d=211.5, 145.5, 132.8, 130.9, 130.5, 128.8, 126.3, 50.5,
46.6, 41.0, 36.1, 35.7, 21.6, 8.0 ppm; GCMS (80–1–20–270—6, v=
45 cmsÀ1): m/z: 11.68 (272, 270, 241, 173, 159, 149, 136, 128, 115, 89, 69,
55*); ee=80% S (absolute configuration was assigned in analogy with
the literature;[11] enantiomeric excess was measured by chiral SFC (S2-
AS, 2%-2–1–15%, v=2 mLsÀ1): tR1 =5.07 (R), tR2 =5.46 min (S)).
tR2 =53.6 min).
(S)-3-(3,4-Dichlorophenyl)-3-methylcyclohexanone
(30):
1H NMR
2-Bromo-3,3-ethylmethylcyclohexanone (38): Mixture of two diastereo-
mers 32:67 (171 mg, 78%); H NMR (400 MHz, CDCl3): d=4.23 (s, 1H),
1
(400 MHz, CDCl3): d=7.39–7.37 (m, 2H), 7.16–7.13 (m, 1H), 2.81 (d,
1H, J=14.1 Hz), 2.45 (d, 1H, J=14.2 Hz), 2.36–2.31 (m, 2H), 2.17–2.10
(m, 1H), 1.96–1.87 (m, 2H), 1.73–1.64 (m, 1H), 1.30 ppm (s, 3H);
13C NMR (100 MHz, CDCl3): d=211.3, 147.9, 130.6, 128.1, 125.4, 117.7,
115.4, 52.9, 42.9, 40.8, 37.8, 29.7, 22.1 ppm; [a]2D0 =+76.0 (c=1.0 in
CHCl3); ee=23% S (absolute configuration was assigned in analogy with
the literature;[11] enantiomeric excess was measured by chiral SFC (S2-
AS, 2%-2–1–15%, v=2 mLsÀ1): tR1 =7.0 (R), tR2 =7.6 min (S)).
4.05 (s, 1H), 3.11–3.03 (m, 1H), 2.95–2.88 (m, 1H), 2.30–2.23 (m, 1.5H),
1.87–1.80 (m, 4H), 1.68–1.55 (m, 1.5H), 1.50–1.43 (m, 3.2H), 1.42–1.32
(m, 2.2H), 1.02 (s, 3H), 0.99 (s, 3H), 0.92–0.82 ppm (m, 4.5H); 13C NMR
(100 MHz, CDCl3): d=204.7, 204.0, 65.5, 63.7, 42.3, 41.2, 37.1, 36.1, 31.9,
31.4, 31.2, 28.0, 22.9, 21.4, 21.3, 21.1, 8.0, 7.1 ppm; ee=76% R (1st diaste-
reomer), ee=78% R (2nd diastereomer) (absolute configuration was as-
signed in analogy with the literature;[11] enantiomeric excess was mea-
sured by chiral GC analysis (Lipodex E, 80–60–1–120–80–170–3, v=
40 cmsÀ1): 1st diastereomer: tR1 =68.8 (R), tR2 =74.0 (S); 2nd diastereo-
mer: tR1 =88.0, tR2 =88.9 min).
(R)-3,3-Phenylmethylcyclohexanone
(31):[59]
1H NMR
(400 MHz,
CDCl3): see data for compound 11; [a]2D0: À26.3 (c=1.70 in CHCl3); ee=
66% R (absolute configuration was assigned in analogy with the litera-
ture;[11] enantiomeric excess was measured by chiral GC analysis (Hydro-
2-Allyl-3-(but-3-enyl)-3-methylcyclohexanone (39): Mixture of two dia-
stereomers 94:6 (148 mg, 72%); 13C NMR (100 MHz, CDCl3): d=212.8,
212.4, 138.8, 137.8, 137.5, 136.7, 136.5, 116.4, 116.3, 115.6, 115.4, 114.6,
dex-B-3P, isotherm 1408C, v=38 cmsÀ1):
(S)).
tR1 =31.7 (R), tR2 =32.7 min
9902
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 9890 – 9904

Formation of Quaternary Chiral Centers
FULL PAPER
114.5, 114.5, 61.7, 58.7, 53.9, 53.5, 49.6, 49.3, 43.8, 42.1, 41.6, 41.4, 40.9,
40.4, 37.2, 36.3, 35.5, 35.4, 34.8, 34.0, 33.8, 33.7, 28.9, 28.8, 28.3, 28.2, 28.0,
27.9, 27.7, 27.5, 25.5, 22.9, 22.8, 22.6, 21.1 ppm; GCMS (80–1–20–270–6,
v=45 cmsÀ1): m/z: 7.9 (10%) (206, 191, 162, 151, 109, 95, 81*, 67, 55),
8.0 (76%) (206, 191, 162, 151, 109, 95, 81*, 67, 55), 8.1 (14%) (206, 191,
162, 151, 109, 95, 81*, 67, 55); ee=91% S (absolute configuration was as-
signed in analogy with the literature;[11] enantiomeric excess was mea-
sured by chiral GC analysis on the butenyl addition (Hydrodex B3P, iso
1308C, v=43 cmsÀ1): tR1 =8.7 (S), tR2 =9.0 min (R)).
[14] M. Vuagnoux-d’Augustin, S. Kehrli, A. Alexakis, Synlett 2007, 2057–
2060.
[15] D. Bourissou, O. Guerret, F. P. Gabbaie, G. Bertrand, Chem. Rev.
2000, 100, 39–91.
[16] V. Cꢀsar, S. Bellemin-Laponnaz, L. H. Gade, Chem. Soc. Rev. 2004,
33, 619–636.
[17] O. Kuehl, Chem. Soc. Rev. 2007, 36, 592–607.
[18] S. T. Liddle, I. S. Edworthy, P. L. Arnold, Chem. Soc. Rev. 2007, 36,
1732–1744.
[19] H. Clavier, S. P. Nolan, Annu. Rep. Prog. Chem. Sect. B: Org. Chem.
2007, 103, 193–222.
4a-Methyl-2,3,4,4a,5,6,9,9a-octahydro-1H-benzo[7]annulen-1-one (40):
Two diastereomers 58:42 (62%); H NMR (400 MHz, CDCl3): 1st diaste-
1
[20] C. M. Crudden, D. P. Allen, Coord. Chem. Rev. 2004, 248, 2247–
2273.
[21] N. Marion, S. P. Nolan, Chem. Soc. Rev. 2008, 37, 1776–1782.
[22] R. A. Kelly III, H. Clavier, S. Giudice, N. M. Scott, E. D. Stevens, J.
Bordner, I. Samardjiev, C. D. Hoff, L. Cavallo, S. P. Nolan, Organo-
metallics 2008, 27, 202–210.
[23] B. R. Dible, M. S. Sigman, Inorg. Chem. 2006, 45, 8430–8441.
[24] S. Diez-Gonzalez, S. P. Nolan, Coord. Chem. Rev. 2007, 251, 874–
883.
[25] L. Cavallo, A. Correa, C. Costabile, H. Jacobsen, J. Organomet.
Chem. 2005, 690, 5407–5413.
reomer: d=5.81–5.75 (m, 1H), 5.74–5.69 (m, 1H), 2.54–2.51 (m, 1H),
2.43–2.29 (m, 3H), 2.19–2.14 (m, 3H), 2.01–1.82 (m, 3H), 1.63–1.57 (m,
1H), 1.55–1.45 (m, 2H), 0.87 (s, 3H); 2nd diastereomer: 5.70–5.68 (m,
2H), 2.59–2.52 (m, 1H), 2.48–2.45 (m.1H), 2.36–2.20 (m, 3H), 2.07–2.00
(m, 1H), 1.97–1.92 (m, 2H), 1.87–1.74 (m, 3H), 1.52–1.48 (m, 1H), 1.45–
1.39 (m, 1H), 1.02 ppm (s, 3H); 13C NMR (100 MHz, CDCl3): 1st diaste-
reomers: d=213.5, 132.3, 130.4, 57.5, 43.0, 41.9, 41.3, 40.0, 23.6, 23.3,
22.9, 18.6; 2nd diastereomer: 214.4, 131.6, 129.0, 59.4, 40.1, 39.8, 39.1,
35.9, 27.4, 25.5, 23.3, 22.0 ppm; MS-EI: m/z: 178, 163, 150, 145, 135, 117,
107, 91, 79*, 67, 55.
3-Methyl-3-(4-methylpent-3-enyl)cyclohexanone
(41):[55]
1H NMR
[26] R. H. Crabtree, J. Organomet. Chem. 2005, 690, 5451–5457.
[27] P. K. Fraser, S. Woodward, Tetrahedron Lett. 2001, 42, 2747–2750.
[28] a) F. Guillen, C. L. Winn, A. Alexakis, Tetrahedron: Asymmetry
2001, 12, 2083–2086; b) J. Pytkowicz, S. Roland, P. Mangeney, Tetra-
hedron: Asymmetry 2001, 12, 2087–2088.
[29] H. Clavier, J.-C. Guillemin, M. Mauduit, Chirality 2007, 19, 471–
476.
[30] A. Alexakis, C. L. Winn, F. Guillen, J. Pytkowicz, S. Roland, P. Man-
geney, Adv. Synth. Catal. 2003, 345, 345–348.
(400 MHz, CDCl3): d=5.08 (t, 1H, J=5.9 Hz), 2.31–2.26 (m, 2H), 2.23–
2.10 (m, 2H), 1.98–1.85 (m, 4H), 1.68 (s, 3H), 1.67–1.60 (m, 2H), 1.60 (s,
3H), 1.31–1.26 (m, 2H), 0.94 ppm (s, 3H); 13C NMR (100 MHz, CDCl3) :
212.4, 131.7, 124.4, 53.8, 41.8, 41.2, 38.7, 36.0, 25.8, 25.0, 22.3, 22.2,
17.7 ppm; [a]2D0 =À6.8 (c=1.17 in CHCl3); ee=86% S (absolute configu-
ration was assigned in analogy with the literature;[11] enantiomeric excess
was measured by chiral GC analysis (lipodex E, 80–0–1–120, v=
45 cmsÀ1): tR1 =28.1 (S), tR2 =29.5 min (R)).
[31] a) H. Clavier, L. Coutable, L. Toupet, J.-C. Guillemin, M. Mauduit,
J. Organomet. Chem. 2005, 690, 5237–5254; b) J. Wencel, M. Maud-
uit, H. Hꢀnon, S. Kehrli, A. Alexakis, Aldrichim. Chim. Acta 2009,
42, 43–50; c) D. Rix, S. Labat, L. Toupet, C. Crꢀvisy, M. Mauduit,
Eur. J. Inorg. Chem. 2009, 1989–1999.
[32] T. Moore, M. Merzouk, N. Williams, Synlett 2008, 21–24.
[33] T. Uchida, T. Katsuki, Tetrahedron Lett. 2009, 50, 4741–4743.
[34] K.-s. Lee, A. H. Hoveyda, J. Org. Chem. 2009, 74, 4455–4462.
[35] D. Martin, S. Kehrli, M. D’Augustin, H. Clavier, M. Mauduit, A.
Alexakis, J. Am. Chem. Soc. 2006, 128, 8416–8417.
[36] K.-s. Lee, M. K. Brown, A. W. Hird, A. H. Hoveyda, J. Am. Chem.
Soc. 2006, 128, 7182–7184.
[37] M. K. Brown, T. L. May, C. A. Baxter, A. H. Hoveyda, Angew.
Chem. 2007, 119, 1115–1118; Angew. Chem. Int. Ed. 2007, 46, 1097–
1100.
[38] T. L. May, M. K. Brown, A. H. Hoveyda, Angew. Chem. 2008, 120,
7468–7472; Angew. Chem. Int. Ed. 2008, 47, 7358–7362.
[39] M. K. Brown, A. H. Hoveyda, J. Am. Chem. Soc. 2008, 130, 12904–
12906.
Acknowledgements
The authors thank the Swiss National Research Foundation (grant no.
200020-126663) and COST action D40 (SER contract no C07.0097) for fi-
nancial support, and BASF for chiral amines.
[1] For reviews, see: a) A. Alexakis, C. Benhaim, Eur. J. Org. Chem.
2002, 3221–3223; b) A. Alexakis, J. E. Bꢉckvall, N. Krause, O.
Pamies, M. Dieguez, Chem. Rev. 2008, 108, 2796–2823; c) S. R. Har-
utyunyan, T. den Hartog , K. Geurts, A. J. Minnard, B. L. Feringa,
Chem. Rev. 2008, 108, 2824–2852.
[2] B. M. Trost, C. Jiang, Synthesis 2006, 369–396.
[3] A. Alexakis, C. Malan, L. Lea, K. Tissot-Croset, D. Polet, C. Falcio-
la, Chimia 2006, 60, 124–130.
[4] Quaternary Stereocenters Challenge And Solution in Organic Synthe-
sis (Eds.: J. Christoffers, A. Baro), Wiley, New York, 2006.
[5] T. Thaler, P. Knochel, Angew. Chem. 2009, 121, 655–658; Angew.
Chem. Int. Ed. 2009, 48, 645–648.
[40] Y. Matsumoto, K.-i. Yamada, K. Tomioka, J. Org. Chem. 2008, 73,
4578–4581.
[41] W. A. Herrmann, L. J. Goossen, C. Koecher, G. R. J. Artus, Angew.
Chem. 1996, 108, 2980–2982; Angew. Chem. Int. Ed. Engl. 1996, 35,
2805–2807.
[6] a) E. Fillion, A. Wilsily, J. Am. Chem. Soc. 2006, 128, 2774–2775;
b) E. Fillion, A. Wilsily, E. T. Liao, Tetrahedron: Asymmetry 2006,
17, 2957–2959.
[42] S. Pikul, E. J. Corey, Org. Synth. 1993, 71, 22–29.
[43] T. J. Seiders, D. W. Ward, R. H. Grubbs, Org. Lett. 2001, 3, 3225–
3228.
[44] K. Tissot-Croset, D. Polet, A. Alexakis, Angew. Chem. 2004, 116,
2480–2482; Angew. Chem. Int. Ed. 2004, 43, 2426–2428.
[45] K. Tissot-Croset, D. Polet, S. Gille, C. Hawner, A. Alexakis, Synthe-
sis 2004, 2586–2590.
[46] a) B. L. Feringa, R. Badorrey, D. Pena, S. R. Harutyunyan, A. J.
Minnaard, Proc. Nat. Acad. Sci. USA 2004, 101, 5834–5838; b) S.-Y.
Wang, S.-J. Ji, T.-P. Loh, J. Am. Chem. Soc. 2007, 129, 276–277.
[47] T. Robert, J. Velder, H.-G. Schmalz, Angew. Chem. 2008, 120, 7832–
7835; Angew. Chem. Int. Ed. 2008, 47, 7718–7721.
[48] R. S. Bhosale, S. V. Bhosale, S. V. Bhosale, T. Wang, P. K. Zubaidha,
Tetrahedron Lett. 2004, 45, 7187–7188.
[7] A. Wilsily, E. Fillion, Org. Lett. 2008, 10, 2801–2804.
[8] A. Wilsily, E. Fillion, J. Org. Chem. 2009, 74, 8583–8594.
[9] J. Wu, D. M. Mampreian, A. H. Hoveyda, J. Am. Chem. Soc. 2005,
127, 4584–4585.
[10] A. W. Hird, A. H. Hoveyda, J. Am. Chem. Soc. 2005, 127, 14988–
14989.
[11] M. d’Augustin, L. Palais, A. Alexakis, Angew. Chem. 2005, 117,
1400–1402; Angew. Chem. Int. Ed. 2005, 44, 1376–1378.
[12] M. Vuagnoux-d’Augustin, A. Alexakis, Eur. J. Org. Chem. 2007,
5852–5860.
[13] L. Palais, I. S. Mikhel, C. Bournaud, L. Micouin, C. A. Falciola, M.
Vuagnoux-d’Augustin, S. Rosset, G. Bernardinelli, A. Alexakis,
Angew. Chem. 2007, 119, 7606–7609; Angew. Chem. Int. Ed. 2007,
46, 7462–7465.
Chem. Eur. J. 2010, 16, 9890 – 9904
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9903

A. Alexakis et al.
[49] W. G. Dauben, D. M. Michno, J. Org. Chem. 1977, 42, 682–685.
[50] M. Vuagnoux-d’Augustin, A. Alexakis, Chem. Eur. J. 2007, 13,
9647–9162.
[51] C. Hawner, K. Li, V. Cirriez, A. Alexakis, Angew. Chem. 2008, 120,
8334–8337; Angew. Chem. Int. Ed. 2008, 47, 8211–8214.
[52] E. Reyes, J. L. Vicario, L. Carrillo, D. Badia, U. Uria, A. Iza, J. Org.
Chem. 2006, 71, 7763–7772.
c) B. D. Schwartz, D. P. Tilly, R. Heim, S. Wiedemann, C. M. Wil-
liams, P. V. Berhardt, Eur. J. Org. Chem. 2006, 3181–3192.
[56] G. H. Posner, C. M. Lentz, J. Am. Chem. Soc. 1979, 101, 934–946.
[57] J. Kabbara, S. Flemming, K. Nickisch, H. Neh, J. Westermann, Tetra-
hedron 1995, 51, 743–754.
[58] B. H. Lipshutz, R. S. Wilhelm, J. A. Kozlowski, J. Org. Chem. 1984,
49, 3938–3942.
[53] H. Hꢀnon, M. Mauduit, A. Alexakis, Angew. Chem. 2008, 120,
9262–9264; Angew. Chem. Int. Ed. 2008, 47, 9122–9124.
[54] M. Vuagnoux-d’Augustin, A. Alexakis, Tetrahedron Lett. 2007, 48,
7408–7412.
[59] H. O. House, J. M. Wilkins, J. Org. Chem. 1978, 43, 2443–2454.
[60] J. Cason, A. I. A. Khodair, J. Org. Chem. 1967, 32, 575–581.
[55] a) M. J. Gallen, C. M. Williams, Org. Lett. 2008, 10, 713–715;
b) A. P.-J. Chen, C. M. Williams, Org. Lett. 2008, 10, 3441–3443;
Received: February 22, 2010
Published online: June 10, 2010
9904
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 9890 – 9904