Formation of contiguous quaternary and tertiary stereocenters by sequential asymmetric conjugate addition of grignard reagents to 2-substituted enones and mg-enolate trapping

Article abstract of DOI:10.1002/chem.201500292

Herein a comprehensive study is provided on the asymmetric conjugate addition (ACA) of Grignard reagents to α-substituted cyclic enones. After the elucidation of the optimal experimental conditions, the scope of Grignard reagents and Michael acceptors was examined. Whereas secondary Grignards gave better enantioselectivities with 2-cyclopentenones, both linear and branched Grignard reagents were tolerated for the ACA to 2-methylcyclohexenone. The sequential ACA-enolate trapping, which leads to quaternary stereocenters, was then studied. Thus, many electrophiles have been tested, thereby giving rise to highly functionalized cyclic ketones with contiguous α-quaternary and β-tertiary centers. The present technique is believed to bring a new approach to versatile terpenoid-like skeletons of bioactive natural products. Straightforward derivatizations of enantioenriched saturated cyclic ketones further support the potential of the present methodology in synthesis.

Full text of DOI:10.1002/chem.201500292

DOI: 10.1002/chem.201500292
Full Paper
&
Asymmetric Addition
Formation of Contiguous Quaternary and Tertiary Stereocenters
by Sequential Asymmetric Conjugate Addition of Grignard
Reagents to 2-Substituted Enones and Mg–Enolate Trapping
Nicolas Germain and Alexandre Alexakis*[a]
Abstract: Herein a comprehensive study is provided on the
asymmetric conjugate addition (ACA) of Grignard reagents
to a-substituted cyclic enones. After the elucidation of the
optimal experimental conditions, the scope of Grignard
reagents and Michael acceptors was examined. Whereas sec-
ondary Grignards gave better enantioselectivities with 2-cy-
clopentenones, both linear and branched Grignard reagents
were tolerated for the ACA to 2-methylcyclohexenone. The
sequential ACA–enolate trapping, which leads to quaternary
stereocenters, was then studied. Thus, many electrophiles
have been tested, thereby giving rise to highly functional-
ized cyclic ketones with contiguous a-quaternary and
b-tertiary centers. The present technique is believed to bring
a new approach to versatile terpenoid-like skeletons of bio-
active natural products. Straightforward derivatizations of
enantioenriched saturated cyclic ketones further support the
potential of the present methodology in synthesis.
Introduction
One-pot synthesis of multiple stereogenic centers is
a
tremendously desirable process in organic chemistry.[1] Indeed,
stereo- and regiocontrolled cascade reactions can be powerful
methods to install complex molecular scaffolds such as natural
products precursors. However, sequential one-pot trans-
formations are not always compatible with all chemical events,
especially when organometallic complexes are involved.
Nevertheless, transition-metal-catalyzed conjugate addition of
organometallic reagents is indeed compatible with tandem
and/or sequential reactions. Sequential conjugate addition fol-
lowed by metal–enolate trapping has been described for vari-
ous metal-enolates and electrophiles.[2] This transformation has
been studied for the formation of contiguous tertiary stereo-
centers (Scheme 1A). Owing to the emergence of the asym-
metric conjugate addition (ACA) to trisubstituted enones, the
trapping of the consecutive metal–enolates emerged soon
after (Scheme 1B). Unfortunately, quaternary centers poorly dis-
criminate between the enolate faces, thus leading to a low
level of diastereoselection. The case of a-substituted cyclic
enones offers a promising third route for the sequential ACA–
enolate trapping. Indeed, it allows for the construction of con-
tiguous quaternary and tertiary centers (Scheme 1C). The prev-
alence of quaternary centers in nature in conjunction with the
need for techniques based on cheap and readily available
Scheme 1. Sequential ACA-enolate trapping.
chemicals convinced us to focus on the development of the
ACA–enolate trapping sequence.[3,4] Preliminary results sug-
gested that the b substituent efficiently shields one face of the
enolate, thus leading to high diastereomeric ratios (d.r.).[5]
The ACA to a-substituted cyclic enones has already been
considered in our laboratory.[6]
A catalytic system was
optimized in 2007, and it was found that the complex
Cu/phosphoramide L1 was an active catalyst for the ACA to 2-
methylcyclohexenone too. However, even after optimizations,
the challenge of ACA to 2-methylcyclopentenone 4 and 5
remained.[6c] Furthermore, a sequential ACA–enolate trapping
event based on these prior results was developed by Cramer
and co-workers.[7] We thus decided to face three challenges at
the same time: ACA to a-substituted cyclic enones, the devel-
opment of this methodology to five-membered rings, and the
[a] Dr. N. Germain, Prof. Dr. A. Alexakis
Dꢀpartement de chimie Organique, Science II
Universitꢀ de Genꢁve
Quai Ernest Ansermet 30, 1211 Genꢁve 4 (Switzerland)
E-mail: alexandre.alexakis@unige.ch
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201500292.
Chem. Eur. J. 2015, 21, 1 – 11
1
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
lates. No improvement was ob-
served when a phosphoramidite
or phosphinamine or BINAP
ligand was used in combination
with copper thiophene carboxyl-
ate (CuTC) for the ACA of ethyl-
magnesium bromide to 4. It was
then decided to study the influ-
ence of N-heterocyclic carbene
(NHC) ligands for this transfor-
mation. The ACA of EtMgBr car-
ried out with carbene ligands L3
and C2-symmetric L4 furnished
a racemic mixture of the desired
saturated ketone 8 (Table 1, en-
tries 1 and 2). The same results
were obtained with L5 (Table 1,
entry 3). The breakthrough came
[8]
with Mauduit-type NHC L6 for
the addition of EtMgBr in the
presence of copper triflate: full
conversion and a promising 43%
ee was obtained (Table 1,
entry 4). When diethylzinc was
Scheme 2. ACA to a-substituted enones.
tested under the same condi-
tions, no reactivity was ob-
served. Neither L7 (Table 1,
optimization of a tandem ACA–enolate trapping for a broad
scope of electrophiles (Scheme 2).
entry 5, 16% ee) nor L8 (Table 1, entry 6, rac) showed satisfying
results. A complex mixture of compounds was detected with
TaniaPhos-type ligand L8 (Table 1, entry 7), thus narrowing the
path towards an extensive screening of ferrocene-based
ligands.[9]
In this article, the scope of the ACA of Grignard reagents to
a-substituted cyclic enones will be overviewed. After all the
optimization experiments, the scope of Grignard reagents and
Michael acceptors will be shown. The sequential ACA–enolate
trapping will then be detailed through a comprehensive
It should be noted that adduct 8 was obtained as cis/trans
mixture in variable amounts. As the enantioselectivity is deter-
mined during the conjugate addition step, both diastereomers
have the same ee. This was checked in many instances.
Isomerization with 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU)
provided mixtures that contained >90% of the trans isomer.
Various solvents were surveyed for the ACA of EtMgBr to 2-
methylcyclopentenone 4 in the presence of a catalytic amount
of Cu(OTf)2/L6 complex (formed in situ, Table 2). Acyclic ethers
gave significantly higher results than the cyclic ones (Table 2,
entries 1–3 versus entries 8 and 9). Although diethyl ether was
the best choice for this transformation, the use of diisopropyl
ether was not detrimental, and 8 was obtained in 37% ee
(Table 2, entry 2). Comparable results were obtained with cy-
clopentyl methyl ether (Table 2, entry 3, 26% ee), toluene
(Table 2, entry 4, 25% ee), and dichloromethane (Table 2,
entry 5, 21% ee), probably indicating that the nature of the sol-
vent (e.g., polarity, coordination abilities) is not a determining
factor for the species involved in the enantio-determining step.
Moderate results were obtained with 2-methyltetrahydrofuran
(Table 2, entry 7, 14% ee) and methyl tert-butyl ether (Table 2,
entry 8, 13% ee). A racemic mixture was recovered with tetra-
hydrofuran and dimethoxyethane (Table 2, entries 9 and 10).
With the best solvent in hand, and with L6 as reference, an
extensive screening of various NHCs, including Mauduit-type
survey of electrophiles, and
a few derivatizations will
eventually demonstrate the synthetic potential of our method
towards natural product synthesis.[5]
Results and Discussion
ACA to a-substituted enones
The ACA to 2-methylcyclohexenone 4 was first studied with
classical ACA catalysts. The previous results were reproduced
using L2, as well as with other phosphoramidites, phosphin-
amines,
and
2,2’-bis(diphenylphosphino)-1,1’-binaphthyl
(BINAP).[6c] In addition to the disappointingly low enantiomeric
excess (ee) values observed, this method is essentially limited
by itself. Indeed, organoaluminum reagents are mostly useful
for the introduction of a methyl moiety. Also, trimethylalumi-
num is a relatively cheap reagent. Numerous research groups
have successfully demonstrated the versatility of trimethyl-
aluminum in ACA throughout highly selective processes.
However, the number of organoaluminum derivatives readily
available is rather limited, so we focused on Grignard reagents
instead. In addition, it was predicted that the reactivity of the
resulting Mg–enolate would be higher than Zn– or Al–eno-
&
&
Chem. Eur. J. 2015, 21, 1 – 11
www.chemeurj.org
2
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Table 1. Screened catalytic systems for the ACA to 2-methylcyclo-
Table 2. Solvent optimization for the ACA of ethylmagnesium bromide
pentenone.
to 2-methylcyclopentenone.
Entry[a]
Solvent
Et2O
ee [%][b]
1
2
3
4
5
6
7
8
9
10
43
37
26
25
21
14
13
rac
rac
rac
iPr2O
CPME[c]
toluene
CH2Cl2
2-MeTHF
MTBE[c]
dioxane
THF
DME[c]
[a] The reactions were all stopped after total consumption of the starting
material. [b] Determined by chiral GC. [c] CPME=cyclopentyl methyl
ether; MTBE=methyl tert-butyl ether.
cyclohexenone catalyzed by complex Cu/ent-L6 gave ent-8
with À41% ee. The influence of the blocking part of the
chelating arm was further optimized. The relative effect of the
mesitylamine (ent-L6, À41% ee) and the mesitylmethanamine
(L22 and L23, 22% and 15% ee, respectively) group negatively
affected the enantioinduction. It was eventually found that
bulkier aromatic groups considerably increased the ee values
measured for 3-ethyl-2-methylcyclopentanone. This study
ended with the test of L27 (60% ee), which remained as our
best ligand for the next experiments. The corresponding
ligand that bore a pendant isobutyl group—that is, leucinol-
based
Entry
Ligand
Copper salt
RnM
EtMgBr
EtMgBr
EtMgBr
EtMgBr
EtMgBr
EtMgBr
EtMgBr
ee [%][a]
[b]
1
2
3
4
5
6
7
L3
L4
L5
L6
L7
L8
L9
CuTC
Cu(OTf)2
CuTC
Cu(OTf)2
Cu(OTf)2
Cu(OTf)2
Cu(OTf)2
–
–
–
[b]
[b]
43
16
[b]
–
[c]
[a] Determined by chiral GC. [b] A mixture of racemic 3-ethyl-2-methyl-
cyclohexanone, 3-ethyl-1-methyl-cyclohex-1,2-diene, and 1-ethyl-3-
methylcyclohex-2-en-1-ol was recovered. [c] Not determined.
NHCs, was achieved for the ACA
of ethylmagnesium bromide to
a-methylcyclopentenone
(Figure 1). We first examined the
influence of a chiral backbone
(L10), but a racemic mixture of
the desired saturated ketone 8
was detected.
Non-natural amino acid based
NHCs such as L10 and L11 were
also unsuccessful. Compound L12
was synthesized from (1R,2R)-2-
amino-1,2-diphenylethanol, but
no enantioselectivity was mea-
sured with 8. The first Mauduit-
type NHCs tested were L13 and
L14, but only poor ee values were
observed. Various scaffolds were
then tried, although modest ee
values or racemic mixtures were
obtained (L15–L21). As expected,
the ACA of EtMgBr to 2-methyl- Figure 1. NHC ligands surveyed for the ACA of ethylmagnesium bromide to 2-methylcyclopentenone.
Chem. Eur. J. 2015, 21, 1 – 11 www.chemeurj.org ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3
&
&
These are not the final page numbers! ÞÞ

Full Paper
NHC—led to lower ee values (42%, not shown). Although this
study highlighted the inefficiency of valine-based Mauduit-
type NHCs as well as other natural amino acid based NHCs,
nowadays tert-leucinol can also be considered to be a relatively
cheap starting material for this methodology. Further studies
of catalyst recycling might support this hypothesis and encour-
age prospective large-scale processes.
of the reaction media was also tested but was not significant;
the quality of the Grignard reagent rather than the overall dilu-
tion of the system led to enhanced results. Additional tests
were conducted with 3-pentyl- and 5-heptylmagnesium bro-
mides, thus leading to the corresponding ketones in 71 and
74% ee. ACA of cyclic Grignard reagents allowed for the prepa-
ration of ketones 17–19 in 80 to 83% ee and high yields.
The methodology was extended to various cyclic enones
(Scheme 3). When a-pentylcyclopentenone 5 was tested under
The influence of temperature on the reaction outcome was
also studied. Optimal results were found for temperatures that
ranged from À20 to À408C. Slightly higher yields and regio-
selectivity were obtained at À308C.
The scope of Grignard reagents was then examined for the
b-alkylation of a-methylcyclopentenone. No stereocontrol was
obtained for the transfer of the methyl moiety using MeMgBr.
Concerning linear Grignard reagents (Table 3, entries 2–6), the
Table 3. Scope of Grignard reagents.
Entry[a]
Product
RMgBr
Conv. [%]
d.r.[b]
ee [%][b]
[c]
1
2
3
4
5
6
7
8[d]
9
10
11
12
13
6
8
9
Me
Et
nPr
–
50:50
82:18
93:7
82:18
94:6
–
75
65
90
91
60
52
48
24
34
24
90
71
74
83
80
83
10
11
12
13
14
15
16
17
18
19
nBu
butenyl
phenethyl
iBu
[c]
–
94:6
88
89
71
60
85
91
65
87:13
60:40
82:18
70:30
52:48
60:40
52:48
iPr
3-pentyl
5-heptyl
cPent
cHex
Scheme 3. ACA to a-substituted enones: Scope of substrates.
cHept
our optimized conditions, an interesting 80% ee was obtained.
Comparable results were observed for the ACA of ethyl- and
isopropylmagnesium bromide to six-membered rings (80 and
81% ee, respectively). It is worth noting that the addition of
both linear and branched moieties was tolerated when
applying our strategy to 2-methylcyclohexenone.
[a] The reactions were carried out in dry diethyl ether at À308C with
1 mol% of L27, 0.75 mol% of copper triflate, and 1.2 equiv of Grignard re-
agent. [b] Determined by gas chromatography. The absolute configura-
tion was determined in our preliminary communication.[5] [c] A mixture of
starting material, desired product, and 1,2-addition products. [d] 2 mol%
of L27 and 1.67 mol% of copper triflate.
Analogously to five-membered rings, we next wanted to
study the effect of an a-alkyl chain for cyclohexenones
(Scheme 4). The synthesis of 24 was achieved after a-iodina-
tion of cyclohexenone followed by a palladium(0)-catalyzed
Negishi-type coupling to afford the desired 2-n-butylcyclo-
hexenone in 35% yield over two steps.[10] The ACA of EtMgBr
to 24 yielded 25 in 58% yield and 66% ee.
Following the same strategy, the influence of a-aryl groups
has also been evaluated (Scheme 5). Ketone 26 was
synthesized from 23 by Suzuki-type coupling in 60%,[11] and
ACA of EtMgBr gave a disappointing 15% ee. The present
methodology is restricted to a-alkyl substituted cyclic enones.
The ACA to a-substituted cyclic enones is a versatile
approach to introduce two stereodefined tertiary centers.
Whereas the absolute configuration of the b-stereogenic
carbon is determined by the Cu/NHC chiral complex, the rela-
tive configuration of these carbon atoms—that is, the ratio
enantiomeric excesses did not exceed 60% ee despite excel-
lent chemicals yields and good d.r. (>82:18). The reaction
proceeded smoothly to afford the desired 1,4-adducts so that
the crude material was simply purified by a path of silica gel
to separate the product from the catalyst. Branched Grignard
reagents led to much more interesting results. Indeed, for five-
membered rings, the introduction of an isopropyl moiety to
a cyclopentane motif is of critical importance in natural prod-
uct synthesis of terpinoids. To our delight, iPrMgBr was treated
with 2-methylcyclopentenone in the presence of copper to
give the targeted saturated ketone in 90% ee and 89%
isolated yield. Having access to this improved result required
an elevated amount of catalyst loading (2 mol% in copper
triflate instead of 0.75) and a 4m solution of iPrMgBr in
diethyl ether (compared to 3m previously). The concentration
&
&
Chem. Eur. J. 2015, 21, 1 – 11
www.chemeurj.org
4
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
scribed in the literature and has formed the scaffold of many
total syntheses of bioactive compounds.[2] Whereas the reactiv-
ity of Zn– and Al–enolates is relatively low, Mg-enolates are
the next best after Li–enolates. Owing to our interest in natural
products and our experimental results, the sequential ACA of
Grignard reagents and Mg–enolate trapping starting from 2-
substituted cyclic enones was extended. This technique would
give access to contiguous quaternary and tertiary stereo-
defined carbon atoms.
To achieve this sequence, many challenges must be over-
come: i) Tetrasubstituted metal-enolates might be too sterically
congested and thus annihilate the trapping event. ii) Typically,
sequential conjugate additions and enolate trapping might re-
quire a large quantity of electrophile together with a cosolvent
(THF, hexamethylphosphoramide, and so forth). iii) Additional
transition metals are sometimes needed, most notably for
allylations.
Scheme 4. ACA to a-phenylcyclohexenone.
Alkylations
We first investigated the behavior of Mg–enolates in the pres-
ence of benzyl bromides (Figure 2). To our delight, an excellent
level of diastereoselection was measured for all the examples
Scheme 5. ACA to a-butylcyclohexenone.
between the cis and trans diastereomers—is directly correlated
to the enolate-trapping event, which can be improved by an
equilibration using DBU in methanol. For all the results ex-
posed above, the sequence ends with non-diastereoselective
Mg-enolate protonation using a diluted hydrochloric acid solu-
tion. Different options to circumvent this problem have drawn
our interest. All are based on the simple concept of increasing
the steric hindrance of the protic source. Nonchiral stoichio-
metric carboxylic esters have been tested but no significant
improvement was noted.[12] Fehr and co-workers proved that
an ephedrine-based reagent allowed for the stereoselective
protonation of Li-enolates.[13] The tests carried out with their
system did not lead to significantly higher d.r.. We eventually
considered using (+)-BINOL and (À)-BINOL (BINOL=1,1’-bi-2-
naphthol) with a potential match/mismatch effect, but these
experiments were not conclusive. The solution to introduce
larger electrophiles that also react at lower temperature was
the most successful approach.
Figure 2. Benzylation of Mg–enolates.
(98:2 to 99:1 d.r.), and ketones 28–31 were isolated in practical
yields. As expected, the electrophile approached from the
upper face of the enolate, that is, the isopropyl and the benzyl
groups are in a trans relationship. This, as well as the absolute
configuration, was previously confirmed by X-ray analysis of
30.[5]
We next aimed to provide more data on the trapping of
tetrasubstituted enolate using propargyl halides (Figure 3). This
has been previously discussed by Cramer and co-workers with
Al-enolates,[7] even though rather large amounts of reagents
and reactant were required and the reaction was limited to
less-accessible propargyl iodides. In our example, only two
equivalents of commercially available propargyl bromide were
enough to deliver the desired compounds in high d.r. (up to
99:1).
Sequential ACA-enolate trapping
Our interest in developing the ACA of Grignard reagents to a-
cyclopentenone lies in the possibility of using the resulting
Mg–enolate. The sequential conjugate addition of lithium
cuprate reagents followed by enolate trapping is well de-
Chem. Eur. J. 2015, 21, 1 – 11
www.chemeurj.org
5
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Allylation reactions were studied as well. Even though many
previous works have studied these reactions, their application
in tandem ACA–enolate trapping has often encountered
issues. Whereas Zn–enolates can be trapped with allyl halides,
a catalytic amount of [Pd(PPh3)4] is required for the reaction to
proceed.[16] In the case of tetrasubstituted Mg–enolates, the se-
quence smoothly delivered the expected compounds 39–41 in
62–65% isolated yields (Figure 5) without added Pd. In
Figure 3. Propargylation of Mg–enolates.
The alkylation of enolates using non-activated electrophiles
is particularly challenging. However, their application to the
synthesis of biologically active substances such as pheromones
led us to investigate this particular case.[14] We thus took ad-
vantage of the flexibility of our method to carry out an ACA to
a-pentylcyclopentenone followed by Mg–enolate trapping
using methyl iodide in 64% overall yield (Scheme 6). Indeed,
Figure 5. Allylation of Mg–enolates.
contrast to the previous example described by Piers et al.
with Li–enolate (see below for the example with methyl vinyl
ketone), the sequence using 3-bromo-2-ethylpropene reached
full conversion under Pd-free conditions.[17]
We have already described in our preliminary communica-
tion[5] how this methodology can be applied to the formal
synthesis of (+)-4,5-deoxyneodolabelline,[18] Crinipellin B,[17] and
Guanacastepene A.[19]
Scheme 6. Formation of trialkylated cyclopentanone.
other aliphatic iodides did not furnish the desired compounds.
Although a slightly lower d.r. of 87:13 was recovered, the se-
quence gives rise to a unique trialkylated cyclopentanone 34.
The installation of easily derivatizable functions is of the
utmost importance when developing new methodologies
(Figure 4). We thus decided to test a simple alkylation of tetra-
substituted Mg–enolates using ethyl iodoacetate under the
conditions shown in Scheme 6. Slightly lower d.r. values than
the aforementioned results were measured (83:17 and 89:11),
but the synthesis of 37 and 38 (99:1 d.r.) afforded highly
satisfying results.[15]
Michael additions
We then carried out the relevant sequential ACA–Michael addi-
tion, even if this sequence is certainly among the most chal-
lenging. Methyl vinyl ketone (MVK) was examined first, but
only oligomers were detected. This is due to the competitive
pathways between enolates that lead to an alternative path-
way, oligomerization. The problem could be circumvented by
choosing allyliodide derivative 45, which was formed after the
reaction of MVK with trimethylsilyliodine in dichloromethane
at low temperature.[20] The resulting 1,5-diketone could be iso-
lated in 45% yield after reaction of 44 and the Mg–enolate,
which were both formed in situ (Scheme 7). The synthesis of
42 and analogues addresses the need for enantioenriched
trisubstituted cyclopentanones in natural products syntheses,
notably in bioactive terpenoids. In this particular example,
Guanacastepene A was clearly identified as a candidate for the
direct application of our methodology.[19]
The reaction between sterically congested tetrasubstituted
Mg–enolates and phenyl vinyl disulfone was also attempted.[21]
To our delight, functionalized ketone 45 was isolated in 68%
yield and 95:5 d.r. (Scheme 8). Thanks to these versatile
sulfones, 45 could be derivatized into various targets.[22]
The reaction of Li–enolates with a,b-unsaturated electro-
philes was studied by Posner and co-workers. They notably
reported on a particular case of enolate trapping, namely, the
Michael–Michael ring-closure sequence (MIMIRC).[23] In 1981,
the tandem MIMIRC based on two consecutive Michael addi-
tions followed by a Wittig-type cyclization was disclosed for
Figure 4. Trapping of Mg–enolates with an activated electrophilic alkyl
moiety.
&
&
Chem. Eur. J. 2015, 21, 1 – 11
www.chemeurj.org
6
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Scheme 9. MIMIRC concept applied to the ACA to a-substituted enones.
Scheme 7. Mg–enolate trapping with MVK.
Cyanation
We optimized our strategy to form enantioenriched a-cyano-
ketones (Scheme 11). The cyanation of Li–enolates is well
known; however, to the best of our knowledge, the conjugate
addition–cyanation sequence has never been disclosed, not
even in the racemic version.[24] To our delight, the sequence
was carried out to successfully give ketone 50 and 51 in 72
and 69% yield, respectively. Cyano groups can be derivatized
into carboxylic acids or amines, thereby affording numerous
synthetic options.
Scheme 8. Michael–Michael addition.
O-trapping
the reaction between Li–enolates and triphenylvinylphosphoni-
um bromide, thereby allowing the formation of cyclopropanes
and steroids.[23b] The same principle was applied to our con-
temporary system (Scheme 9). Inspired by our previ-
ous results in sequential ACA–Michael additions,[21]
the tetrasubstituted Mg–enolate generated from
1 was trapped with triphenylvinylphosphonium bro-
mide to give a phosphonium ylide, which was again
trapped with another equivalent of triphenylvinyl-
phosphonium bromide. The second equivalent of tri-
phenylvinylphosphonium bromide led to the forma-
The O-trapping of tetrasubstituted enolates has been reported
but can often be problematic (low yield, low selectivity).[25] The
Scheme 10. Halogenation of tetrasubstituted Mg–enolates.
tion of another phosphonium ylide, which reacted
with the carbonyl to give the cyclized product 46 in
31% overall yield after four steps.
Halogenation
The halogenation of Mg-enolates was equally successful
(Scheme 10). The bromination of Mg–enolates has been al-
ready reported using molecular bromine. In our case, we have
developed a standard procedure for the halogenation of tet-
rasubstituted enolates that results from ACA. N-Iodosuccini-
mide (NIS), N-bromosuccinimide (NBS), and N-chlorosuccini-
mide (NCS) have been used to allow for the halogenation of
Mg–enolates. The halogenation of enolates offers a practical
platform for further derivatizations and with, for instance, or-
ganometallic reagents or transformation that involves radi-
cals.
Scheme 11. Cyanation of Mg–enolates.
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2015, 21, 1 – 11
www.chemeurj.org
7
&
&
These are not the final page numbers! ÞÞ

Full Paper
biological properties. Triazoles
have also been associated with
the concept of bioorthogonal
chemistry.[29] By introducing
a propargylic moiety, we now
have access to the Huisgen cy-
cloaddition procedure and thus
to triazoles.[30] We have estab-
Figure 6. O-trapping of enolates.
lished that tosylazide was
a
suitable coupling partner for this
transformation (Scheme 13). Ke-
synthetic potential of this technique led us to consider their
reactivity towards acetic anhydrides and allyl chloroformates tones 32 and 33 were converted into the corresponding tria-
(Figure 6). Acetates 53 and 54 have been prepared in 30 and zoles 62 and 63 in 71 and 68% yield, respectively.
76% yield, respectively. Allyl enol carbonates 55–57 could be
synthetized in good yields too (65–73%). We have already
shown in our preliminary communication[5] how these allyl-
carbonates can be used under Stoltz and Hong’s conditions[26]
for the diastereoselective Pd-catalyzed allylation.
Derivatization and sequential ACA–enolate trapping
In addition to the formal synthesis of terpenes described previ-
ously,[5] we attempted to derivatize a few representative ad-
ducts shown below. Starting from a mixture of diastereomers
of the saturated ketone 14, lactam 59 was formed in 35%
overall yield using Beckmann conditions (Scheme 12).[27] The
same strategy was envisaged for the synthesis of Baeyer–Villig-
er adducts 60 and 61 in 52 and 42% yield, respectively.[28] Both
transformations offer a new approach to enantioenriched
Scheme 13. Application to the formation of triazoles.
4,5-disubstituted d-lactam, 4,5-disubstituted d-lactone, and 5,6-
disubstituted e-lactones. In all the cases, the initial ee values
were recovered in the crude mixtures.
Conclusion
Triazoles have been widely explored by the chemical indus-
try for the production of a new generation of fungicide. The The ACA to a-substituted cyclic enones is a tremendously im-
development of antifungal drugs has been based on the same portant challenge in organic chemistry with a high synthetic
potential. Our methodology, which is based on an
easily prepared Mauduit-type NHC and a cheap
copper salt, was shown to give access to contiguous
tertiary centers in good ee values for a large scope of
branched Grignard reagents. Our conditions were ap-
plied to six-membered rings as well. We then capital-
ized
on
the
use
of
Grignard
reagents and the intrinsically high reactivity of
Mg–enolates to develop sequential ACA–enolate
trapping. Hence, a large scope of highly functional-
ized substrates that bear contiguous quaternary and
tertiary centers could be isolated in excellent d.r..[31]
We eventually gained some insight into the synthetic
potential of our substrates through several derivatiza-
tions. Our recent report that highlights the applica-
tion of these findings in the formal total syntheses of
bioactive diterpenes supports the versatility of the
exposed methodology.
Scheme 12. Baeyer–Villiger and Beckmann reactions.
Chem. Eur. J. 2015, 21, 1 – 11 www.chemeurj.org
&
&
8
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
[2] a) R. J. K. Taylor, Synthesis 1985, 364–392; b) H. C. Guo, J. A. Ma, Angew.
Chem. Int. Ed. 2006, 45, 354–366; Angew. Chem. 2006, 118, 362–375;
Experimental Section
ˇ
ˇ
c) Z. Galestokovꢄ, R. Sebesta, Eur. J. Org. Chem. 2012, 2012, 6688–6695.
[3] a) E. J. Corey, A. Guzman-Perez, Angew. Chem. 1998, 110, 402–415;
b) C. J. Douglas, L. E. Overman, Proc. Natl. Acad. Sci. USA 2004, 101,
5363–5367; c) J. Christoffers, A. Baro, Adv. Synth. Catal. 2005, 347,
1473–1482; d) B. M. Trost, C. Jiang, Synthesis 2006, 369–396; e) C.
Hawner, A. Alexakis, Chem. Commun. 2010, 46, 7295–7306; f) J. P. Das, I.
Marek, Chem. Commun. 2011, 47, 4593–4623.
All NMR spectroscopic measurements were recorded on a Bruker
500FT, Bruker 400FT, or Bruker 300FT instrument. The chemical
shifts (d) are indicated in parts per million relative to residual CHCl3
(d=7.26 ppm for 1H, d=77.16 ppm for 13C in CDCl3). Coupling
constants are given in hertz. Mass spectra (MS) were measured by
means of electrospray ionization (ESI) and high-resolution mass
spectra (HRMS) by means of ESI using the SMS service at the Uni-
versitꢁ de Genꢂve. Optical rotations were obtained using a Perkin–
Elmer 241 or 341 polarimeter at 208C, d=10 cm. Enantiomeric ex-
cesses were measured by means of chiral GC (HP6890GC System,
10 psi H2). Retention times (RT) are indicated in minutes. All com-
mercial materials were used without further purification. All sol-
vents were desiccated on a drying column of aluminum oxide
under nitrogen. All the experiments were conducted using Schlenk
techniques, and no glovebox was required to employ these
synthetic methods.
[4] G. P. Howell, Org. Process Res. Dev. 2012, 16, 1258–1272.
[5] N. Germain, L. Guenee, M. Mauduit, A. Alexakis, Org. Lett. 2014, 16,
118–121. See the X-ray structures for the absolute configuration of all
the substrates. For selected recent examples of ACA of Grignard re-
agents, see: a) B. L. Feringa, R. Badorrey, D. PeÇa, S. R. Harutyunyan, A. J.
Minnaard, Proc. Natl. Acad. Sci. USA 2004, 101, 5834–5838; b) S. Kehrli,
D. Martin, D. Rix, M. Mauduit, A. Alexakis, Chem. Eur. J. 2010, 16, 9890–
9904; c) N. Germain, M. Magrez, S. Kehrli, M. Mauduit, A. Alexakis, Eur. J.
ˇ
´
Org. Chem. 2012, 5301–5306; d) B. Mao, M. Fanana-Mastral, B. L. Ferin-
ga, Org. Lett. 2013, 15, 286–289.
[6] a) M. d’Augustin, L. Palais, A. Alexakis, Angew. Chem. Int. Ed. 2005, 44,
1376–1378; Angew. Chem. 2005, 117, 1400–1402; b) M. Vuagnoux-
d’Augustin, A. Alexakis, Chem. Eur. J. 2007, 13, 9647; c) M. Vuagnoux-
d’Augustin, S. Kehrli, A. Alexakis, Synlett 2007, 2057–2060.
[7] D. T. Ngoc, M. Albicker, L. Schmeider, N. Cramer, Org. Biomol. Chem.
2010, 8, 1781–1784.
General procedure for the ACA of Grignard reagents to
a-substituted cyclic enones
A flame-dried flask was charged with NHC ligand and copper tri-
flate. Dry diethyl ether (2 mL) was added and the suspension was
stirred for 10 min at room temperature and 10 min at À308C. The
Grignard reagent (solution in diethyl ether, 1.2 equiv) was then
added and allowed to react over 30 min. The cyclic enone (1 equiv)
was added with a syringe pump (15 min). The reaction was stirred
for 4 h before acidic quenching with a solution of hydrochloric
acid (1m). The crude mixture was extracted with diethyl ether
(three times). The organic phase was then washed with brine,
dried over sodium sulfate, filtered, and concentrated under
reduced pressure. The pure saturated ketones were isolated as
colorless oils after silica gel flash chromatography.
[8] J. Wencel, M. Mauduit, H. Hꢁnon, S. Kehrli, A. Alexakis, Aldrichimica Acta
2009, 42, 43–50.
[9] B. C. Calvo, A. V. R. Madduri, S. R. Harutyunyan, A. J. Minnaard, Adv.
Synth. Catal. 2014, 356, 2061–2069.
[10] F. Nahra, Y. Macꢁ, D. Lambin, O. Riant, Angew. Chem. Int. Ed. 2013, 52,
3208–3212; Angew. Chem. 2013, 125, 3290–3294.
[11] E.-i. Negishi, Z. Tan, S.-Y. Liou, B. Liao, Tetrahedron 2000, 56, 10197–
10207.
[12] N. Krause, Angew. Chem. Int. Ed. Engl. 1994, 33, 1764–1765; Angew.
Chem. 1994, 106, 1845–1847.
[13] a) C. Fehr, Angew. Chem. Int. Ed. Engl. 1996, 35, 2566–2587; Angew.
Chem. 1996, 108, 2726–2748; b) C. Fehr, J. Galindo, Angew. Chem. Int.
Ed. 2006, 45, 2901–2904; Angew. Chem. 2006, 118, 2967–2970.
[14] J. U. Scher, M. H. Pillinger, Clin. Immunol. 2005, 114, 100–109.
[15] A. Alexakis, S. March, J. Org. Chem. 2002, 67, 8753–8757.
[16] M. Kitamura, T. Miki, K. Nakano, R. Noyori, Bull. Chem. Soc. Jpn. 2000, 73,
999–1014.
General procedure to the ACA-enolate trapping sequence
Following the typical procedure for 1,4-addition, after 4 h, the
crude mixture was diluted in dry tetrahydrofuran (1 mL) at À308C.
Except for all the Michael reactions, the cyanations, the acetyla-
tions, and the carbonations, an excess amount of hexamethylphos-
phoramide (HMPA; 1 mL) was also required for improved yields.
The resulting solution was allowed to stir for 30 min at À308C
before the addition of the electrophile. After addition, the reaction
mixture was allowed to warm overnight. The reaction was
quenched with water, and the desired materials were isolated by
following the same workup as that described above.
[17] E. Piers, J. Renaud, J. Org. Chem. 1993, 58, 11–13.
[18] D. R. Williams, R. W. Heidebrecht, Jr., J. Am. Chem. Soc. 2003, 125, 1843.
[19] M. Mandal, H. Yun, G. B. Dudley, S. Lin, D. S. Tran, S. J. Danishefsky, J.
Org. Chem. 2005, 70, 10619.
[20] R. J. Hamilton, L. N. Mander, S. P. Sethi, Tetrahedron 1986, 42, 2881–
2892.
[21] N. Germain, D. Schlaefli, M. Chellat, S. Rosset, A. Alexakis, Org. Lett.
2014, 16, 2006–2009.
[22] a) B. M. Trost, M. R. Ghadiri, J. Am. Chem. Soc. 1984, 106, 7260–7261;
b) E. Kꢅndig, A. Cunningham, Tetrahedron 1988, 44, 6855–6860; c) J. Yu,
H. Cho, J. Falck, J. Org. Chem. 1993, 58, 5892–5894; d) J. Yu, H.-S. Cho,
S. Chandrasekhar, J. R. Falck, C. Mioskowski, Tetrahedron Lett. 1994, 35,
5437–5440; e) S. Sulzer-Mossꢁ, A. Alexakis, Chem. Commun. 2007,
3123–3135; f) S. B. Tsogoeva, Eur. J. Org. Chem. 2007, 1701–1716; g) A.
Quintard, A. Alexakis, Adv. Synth. Catal. 2010, 352, 1856–1860; h) K. Hal-
skov, T. Johansen, R. Davis, M. Steurer, F. Jensen, K. Jørgensen, J. Am.
Chem. Soc. 2012, 134, 12943–12946; i) Y. m. Chen, P.-H. Lee, J. Lin, K.
Chen, Eur. J. Org. Chem. 2013, 2699–2707.
Acknowledgements
This work was supported by the Swiss National Research
Foundation (grant no. 200020-126663).
[23] a) R. D. Little, J. R. Dawson, Tetrahedron Lett. 1980, 21, 2609–2612;
b) G. H. Posner, J. P. Mallamo, A. Y. Black, Tetrahedron 1981, 37, 3921–
3926; c) G. H. Posner, J. P. Mallamo, M. Hulce, L. L. Frye, J. Am. Chem.
Soc. 1982, 104, 4180–4185; d) G. H. Posner, S. B. Lu, J. Am. Chem. Soc.
1985, 107, 1424–1426; e) G. H. Posner, S. B. Lu, E. Asirvatham, E. F. Sil-
versmith, E. M. Shulman, J. Am. Chem. Soc. 1986, 108, 511–512; f) G. H.
Posner, K. S. Webb, E. Asirvatham, S. S. Jew, A. Degl’Innocenti, J. Am.
Chem. Soc. 1988, 110, 4754–4762.
Keywords: addition · copper · enantioselectivity · Grignard
reagents · natural products
[1] a) L. F. Tietze, U. Beifuss, Angew. Chem. Int. Ed. Engl. 1993, 32, 131–163;
Angew. Chem. 1993, 105, 137–170; b) D. J. Ramꢃn, M. Yus, Angew.
Chem. Int. Ed. 2005, 44, 1602–1634; Angew. Chem. 2005, 117, 1628–
1661; c) K. C. Nicolaou, D. J. Edmonds, P. G. Bulger, Angew. Chem. Int. Ed.
2006, 45, 7134–7186; Angew. Chem. 2006, 118, 7292–7344.
[24] D. Kahne, D. B. Collum, Tetrahedron Lett. 1981, 22, 5011–5014.
[25] M. Vuagnoux-d’Augustin, PhD thesis N83851, University of Geneva,
2007.
Chem. Eur. J. 2015, 21, 1 – 11
www.chemeurj.org
9
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
[26] a) A. Y. Hong, B. M. Stoltz, Eur. J. Org. Chem. 2013, 2745–2759; b) W.-B.
Liu, C. M. Reeves, S. C. Virgil, B. M. Stoltz, J. Am. Chem. Soc. 2013, 135,
10626.
[27] B. M. Reyes, M. a. T. Ramꢆrez-Apan, R. n. A. Toscano, G. Delgado, J. Nat.
Prod. 2010, 73, 1839–1845.
[30] H. C. Kolb, M. G. Finn, K. B. Sharpless, Angew. Chem. Int. Ed. 2001, 40,
2004–2021; Angew. Chem. 2001, 113, 2056–2075.
[31] K. Tomioka, H. Kawasaki, K. Yasuda, K. Koga, J. Am. Chem. Soc. 1988,
110, 3597–3601.
[28] F. Kazmierczak, P. Helquist, J. Org. Chem. 1989, 54, 3988–3992.
[29] H. C. Hang, C. Yu, D. L. Kato, C. R. Bertozzi, Proc. Natl. Acad. Sci. USA
2003, 100, 14846–14851.
Received: January 22, 2015
Published online on && &&, 0000
&
&
Chem. Eur. J. 2015, 21, 1 – 11
www.chemeurj.org
10
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
FULL PAPER
&
Asymmetric Addition
N. Germain, A. Alexakis*
&& – &&
Formation of Contiguous Quaternary
and Tertiary Stereocenters by
Sequential Asymmetric Conjugate
Addition of Grignard Reagents to 2-
Substituted Enones and Mg–Enolate
Trapping
Adding machine: The asymmetric con-
jugate addition of Grignard reagents to
2-methylcyclopentone and -hexenone
was achieved for the construction of
multiple stereocenters up to 90% ee. In
addition, the Mg–enolate can be
trapped by various electrophiles. The
present methods give access to
synthetically versatile enantioenriched
small molecules for natural product
synthesis (see scheme).
Chem. Eur. J. 2015, 21, 1 – 11
www.chemeurj.org
11
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ