Copper-Catalyzed Enantioselective Conjugate Addition to α,β-Unsaturated Aldehydes with Various Organometallic Reagents

Article abstract of DOI:10.1055/s-0035-1562487

β-Substituted aldehydes constitute a very important class of compounds found in nature. Synthesis of this motif can be envisioned by C-C bond formation on enals. For this purpose, we report herein the development of enantioselective copper-catalyzed conjugate addition of various organometallic reagents to α,β-unsaturated aldehydes with (R)-H₈BINAP, (R)-TolBINAP, and (R)-SEGPHOS as chiral ligands. Three sets of conditions were successfully developed and several enals were used. Reactivity and regio- and enantioselectivities were strongly dependent on reaction conditions and substrates. Good to excellent regio- and enantioselectivities were obtained with zinc reagents R₂Zn and aluminum reagents R₃Al. However, the asymmetric conjugate addition of Grignard reagents afforded only moderate to good regio- and enantioselectivities.

Full text of DOI:10.1055/s-0035-1562487

SYNTHESIS
0
0
3
9
-
7
8
8
1
1
4
3
7
-
2
1
0
X
© Georg Thieme Verlag Stuttgart · New York
2016, 48, A–H
paper
A
S. Goncalves-Contal et al.
Paper
Synthesis
Copper-Catalyzed Enantioselective Conjugate Addition to α,β-Un-
saturated Aldehydes with Various Organometallic Reagents
R
CuX cat. / L*
Sylvie Goncalves-Contal
Ludovic Gremaud
Laëtitia Palais
R1
+
R–M
O
R1
asymmetric conjugate addition
(ACA)
O
Three sets of conditions:
RMgBr (2 equiv)
TMSCl (1.3 equiv)
CuTC (5 mol%)
L2 (5.25 mol%)
Et2O, –78 °C
R2Zn (2 equiv)
CuTC (5 mol%)
L3 (5.25 mol%)
Et2O, –20 or 0 °C
R3Al (2 equiv)
CuTC (5 mol%)
L11 (5.25 mol%)
THF, –78 °C
Lucille Babel
Alexandre Alexakis*
O
Department of Organic Chemistry, University of Geneva,
Quai Ernest Ansermet 30, 1211 Geneva 1211, Switzerland
alexandre.alexakis@unige.ch
O
O
PTol2
PTol2
PPh2
PPh2
PPh2
L2 =
L3 =
L11 =
PPh2
Dedicated to the memory of Professor Jean F. Normant
(1936–2016)
O
ee's up to 90%
1,4/1,2 up to 85:15
ee's up to 95%
1,4/1,2 up to 100:0
ee's up to 96%
1,4/1,2 up to 77:23
Received: 19.04.2016
Accepted after revision: 20.05.2016
Published online: 28.06.2016
direct addition to the carbonyl could occur and lead to the
1,2-adduct as byproduct (Scheme 1). The intermediate eno-
late could also react with the highly reactive enal to give the
aldol product as the second possible byproduct. In 2005,
Bräse highlighted the problem when he reported the enan-
tioselective copper-free 1,4-addition of diorganozinc to
α,β-unsaturated aldehydes (Scheme 2 a).2 By using [2.2]pa-
racyclophaneketimine ligands, excellent enantioselectivi-
ties were attained but these were counterbalanced by low
regioselectivities (with1,4/1,2 ratios between 4:1 and 1:1).
DOI: 10.1055/s-0035-1562487; Art ID: ss-2016-c0260-op
Abstract β-Substituted aldehydes constitute a very important class of
compounds found in nature. Synthesis of this motif can be envisioned
by C–C bond formation on enals. For this purpose, we report herein the
development of enantioselective copper-catalyzed conjugate addition
of various organometallic reagents to α,β-unsaturated aldehydes with
(R)-H8BINAP, (R)-TolBINAP, and (R)-SEGPHOS as chiral ligands. Three
sets of conditions were successfully developed and several enals were
used. Reactivity and regio- and enantioselectivities were strongly de-
pendent on reaction conditions and substrates. Good to excellent re-
gio- and enantioselectivities were obtained with zinc reagents R2Zn and
aluminum reagents R3Al. However, the asymmetric conjugate addition
of Grignard reagents afforded only moderate to good regio- and enan-
tioselectivities.
R1
O
1,4-adduct
R
hydrolysis
1)
R1
O
R1
O
R1
OM
1,4-addition
"RM"
HO
R1
R1
O
Key words α,β-unsaturated aldehydes, conjugate addition, organo-
metallic reagents, enantioselectivity, natural products
2) hydrolysis
R
1,2-addition
R
R1
OH
hydrolysis
R1
OM
aldol product
Among the different methodologies reported to create
carbon–carbon bonds in an enantioselective manner, asym-
metric copper-catalyzed conjugate addition (ACA) of or-
ganometallic reagents to Michael acceptors is one of the
most powerful.1 As a consequence, several α,β-unsaturated
compounds were developed successfully, such as carbonyl
derivatives, sulfones, or nitroalkenes. Among all the classes
of Michael acceptors, α,β-unsaturated aldehydes appear to
be the most interesting ones, as the 1,4-addition to such
compounds results in the formation of important β-alkyl-
substituted chiral molecules having interesting properties
or being key synthons for the preparation of natural prod-
ucts. However, nowadays, the direct enantioselective conju-
gate addition to α,β-unsaturated aldehydes still remains a
challenge in terms of regio- and enantioselectivity, with
few examples reported so far. Due to their high reactivity,
R
R
1,2-adduct
Scheme 1 Regioselectivity in the ACA of organometallics to enals
One year later, Marshall described the silyl-promoted
copper-catalyzed 1,4-addition of organozinc reagents to
enals with a cyanocuprate as catalyst.3 Here, 1,4-adducts
were obtained as the sole product in racemic and diastereo-
selective manners. Recently, Córdova described the enan-
tioselective 1,4-addition of a dimethylsilyl group to α,β-un-
saturated aldehydes.4 Combining copper activation of the
nucleophile and chiral aminocatalysis allowed exclusive
1,4-selectivity with moderate to good enantioselectivities.
Finally, rhodium5 and palladium6 were also reported as pos-
sible transition metal catalysts in the asymmetric conjugate
© Georg Thieme Verlag Stuttgart · New York — Synthesis 2016, 48, A–H

B
S. Goncalves-Contal et al.
Paper
Synthesis
jugate addition of organozinc and Grignard reagents to
enals, giving rise to the desired 1,4-adducts in moderate to
good regio- and enantioselectivities.13 Focusing our efforts
on this topic, herein, we present our plentiful experience in
this field as well as our recent progress to enhance regio-
and enantioselectivities in the copper-catalyzed 1,4-addi-
tion to enals with various organometallic reagents (R2Zn,
RMgBr, and R3Al) and using commercially available biden-
tate ligands and copper salts.
Previous work
a) Cu-free 1,4-addition (Bräse)
R
1) R2Zn, L* (2 mol%)
2) H2O
R1
*
O
R1
O
b) Indirect methodologies
R2
R
R2
R
RM, CuX, L*
*
*
R1
R1
O
R1
O
O
We first envisioned the ACA to enals by using dialkyl-
zinc reagents. Indeed, diorganozinc reagents are less reac-
tive than the corresponding organomagnesium and triorga-
noaluminum ones. Several functional groups on the sub-
strates as well as on the organozinc itself are well
tolerated.14 Therefore, we assumed that they may prevent
or slow down 1,2-addition compared to other organometal-
lic reagents. Ligands used in all the following studies are de-
picted in Figure 1: mainly di- and monophosphorylated
binaphthyl-type ligands (L1–L4), SYNPHOS and SEGPHOS
derivatives (L6–L14) and ferrocene-type ligands (L17 and
L18).15
O
OSiR3
R3 R3
R2 = SR3
N
COOEt
O
This work
c) Direct methodology
R
1) RM, CuX, L*
2) H2O
*
R1
R1
O
O
Scheme 2 Strategy developed in this article
addition to enals of arylboronic acid, arylzinc chloride, and
triarylbismuth. With all these chiral systems, only β-aryl-
substituted aldehydes could be obtained. Moreover, in cop-
per and copper-free systems, the scope of the nucleophiles
seemed to be one of the limitations (apart from 1,4-selec-
tivity) as only organozinc and silyl reagents were described.
With the aim of efficiently accessing β-substituted chi-
ral aldehydes, indirect pathways were developed where the
1,4-adduct could easily be converted into the corresponding
aldehydes. These strategies dealt with the copper-catalyzed
conjugate addition to N-acyloxazolidinones,7 thioesters,8 α′-
silyloxy enones,9 α-chloroallylic acetates,10 or, more recent-
ly, 2H-pyran-2-one11 and keto esters12 (Scheme 2 b). How-
ever, there was still a need for a direct asymmetric conju-
gate addition to α,β-unsaturated aldehydes for the sake of
ease, speed, and atom-economic processes. The lack of such
an efficient and general direct methodology prompted us to
report in 2010 the first asymmetric copper-catalyzed con-
Asymmetric Cu-Catalyzed Conjugate Addition of
R2Zn to Enals
Conjugate addition of diethylzinc to trans-2-decenal
(S1) was first studied in order to access 1,4-adduct A1 ex-
clusively (Table 1). Previously, we described that using (R)-
BINAP (L1) as ligand gave only 1,4-addition with a promis-
ing enantiomeric excess of 75% (entry 1).13 However, the re-
action could be further improved in terms of enantioselec-
tivity, and thus we focused on screening the chiral ligand.
Among binaphthyl derivatives, (R)-H8BINAP (L3) allowed
the exclusive conjugate addition to adduct A1 with a very
good ee of 95% (entry 3). No traces of 1,2-adduct or aldol
product were observed. When SYNPHOS derivatives were
tested, only (R)-SYNPHOS (L6) was efficient, providing A1
with a 1,4/1,2 ratio of 95:5 and a good ee of 92% (entry 6).
O
O
O
O
PAr2
PAr2
PAr2
PAr2
PPh2
PPh2
OMe
PPh2
MeO
MeO
PPh2
PPh2
PAr2
PAr2
O
O
L1: Ar = Ph (R)-BINAP
L2: Ar = Tol (R)-TolBINAP
L3: (R)-H8BINAP
L4: (R)-MOP
L5: (S)-MeOBPHEP
L6: Ar = Ph (R)-SYNPHOS L11: Ar = Ph (R)-SEGPHOS
L7: Ar = Tol
L12: Ar = Xylyl
L13: Ar = m-tBu-p-MeOC6H3
L8: Ar = Xylyl
L9: Ar = p-MeOC6H4
L10: Ar = m-tBu-p-MeOC6H3
O
F
i-Pr
i-Pr
P
F
O
PCy2
PCy2
O
i-Pr
i-Pr
PPh2
PPh2
Ph2P
Ar2P
Fe
Fe
PPh2
N
O
O
F
F
P
L14: (R)-DifluorPHOS
L15: (S,S)-BPE
L16: (S,S)-DUPHOS
L17: (S,S)
L18: (S,S)-4-MeO-3,5-Me2C6H2
L19: (R)-PHOX
Figure 1 Chiral ligands used in this study
© Georg Thieme Verlag Stuttgart · New York — Synthesis 2016, 48, A–H

C
S. Goncalves-Contal et al.
Paper
Synthesis
The enantiomeric excess also decreased when the steric
hindrance on the ligand increased (entries 6–10). Indeed,
the excess dropped from 92 to 22% going from L6, with
phenyl on the phosphorus atoms, to L10 with 3-tert-butyl-
4-methoxyphenyl substituents. Finally, no improvements
were observed by using L11 to L14 from the SEGPHOS fam-
ily or ferrocene-type ligands (entries 11–14, 17, and 18).
addition of Et2Zn (entries 1–4 and 6). Due to its lower reac-
tivity, addition of Me2Zn must be run at 0 °C to get complete
conversion and 1,4-adducts bearing primary alkyl chains
B1, B2, B3 and B6 were obtained with lower ee between 59
and 78% (entries 8–10 and 13). Conjugate addition was
more difficult with enals bearing a phenyl or secondary al-
kyl group in β-position (entries 4, 5, 7, 11, 12, and 14). In-
deed, unsatisfactory 1,4/1,2/aldol ratios were obtained even
though the ee’s were good. The only exception is the conju-
gate addition of Et2Zn to isopropyl enal A4 with an excellent
1,4/1,2 ratio and a very good 91% ee (entry 4).
Table 1 Screening of Ligands for the ACA of Diethylzinc to trans-Dece-
nal
Et2Zn (2 equiv)
CuTC (5 mol%)
L* (5.25 mol%)
Table 2 ACA of Dimethylzinc and Diethylzinc to Various Enals
O
Et2O, –20 °C, 6 h
S1
O
A1
CuTC (5 mol%)
(R)-H8BINAP L3 (5.25 mol%)
R
R1
+
R2Zn
O
Entry
Ligand L*
Conv. (%)a
Ratio 1,4/1,2/aldol ee (%)b
R1
Et2O, –20 °C, 6 h
O
(2 equiv)
S1–S7
A1–A7 : R = Et
B1–B7 : R = Me
1
2
L1
100
84
100:0:0
79:21:0
100:0:0
71:29:0
82:18:0
95:5:0
75
67
95
10
65
92
86
80
78
22
84
75
29
58
11
9
L2
Entry
R1
Adduct
Conv. (%)a Ratio
1,4/1,2
ee (%)b
3
L3
100
95
4
L4
1
2
(CH2)6Me
n-Bu
i-Bu
A1
100 (77)c
100 (58)c
100
100:0
100:0
100:0
100:0
31:69
100:0
95 (75)d
92 (64)d
89 (70)d
91e (70)d
78 (52)d
81
5
L5
88
A2
A3
A4
A5
A6
A7
B1
B2
B3
B4
B5
B6
B7
6
L6
100
90
3
7
L7
76:18:6
69:24:7
61:32:7
71:18:11
100:0:0
86:14:0
100:0:0
90:10:0
84:0:16
66:0:34
100:0:0
69:0:31
85:0:15
4
i-Pr
100 (46)c
100 (23)c
100 (74)c
100
8
L8
95
5
Cy
9
L9
85
6
(CH2)2Ph
Ph
10
11
12
13
14
15
16
17
18
19
L10
L11
L12
L13
L14
L15
L16
L17
L18
L19
95
7
62:0:38g
100:0
71 (44)d
78 (68)d
76 (76)d
59
98
8f
(CH2)6Me
n-Bu
i-Bu
100
91
9f
100
100:0
85
10f
11f
12f
13f
14f
100
100:0
98
i-Pr
100
44:0:56g
31:69
81 (70)d
58 (60)d
78
80
Cy
100
87
(CH2)2Ph
Ph
100
65:0:35g
32:68
87
78
57
23
100
62 (64)d
87
a Determined by 1H NMR spectroscopy.
b Determined by chiral GC.
100
c Value in parentheses is the isolated yield determined after reduction to
the corresponding alcohol. Due to the volatility of some alcohols, some iso-
lated yields are lower.
a Determined by 1H NMR spectroscopy.
b Determined by chiral GC.
d Value in parentheses is the previous ee reported by our group.13
e The ee of the corresponding alcohol after reduction.
f Reaction carried out at 0 °C.
Having identified (R)-H8BINAP as the best ligand for the
conjugate addition of diethylzinc to trans-2-decenal (S1),
these optimal conditions were next applied to various α,β-
unsaturated aldehydes (Table 2). In almost all cases, the en-
antiomeric excesses obtained with (R)-H8BINAP were high-
er than the ones previously described13 using (R)-BINAP.
Concerning the conjugate addition of Et2Zn, small and long
primary alkyl chains gave excellent results with perfect re-
gioselectivity and very high ee between 81 and 95% for the
g Ratio 1,4/1,2/aldol.
Finally, we determined the stereochemistry of the inter-
mediate enolate by trapping experiments providing the si-
lyl enol ether16 or the enol acetate17 (Scheme 3). In both
cases, E compounds were obtained. Thus, conjugate addi-
tion to enals proceeded through the s-trans conformation,
giving the E-enolate.
© Georg Thieme Verlag Stuttgart · New York — Synthesis 2016, 48, A–H

D
S. Goncalves-Contal et al.
Paper
Synthesis
Et2Zn (2 equiv)
CuTC (5 mol%)
L3 (5.25 mol%)
TMSOTf
O
OZnEt
Et2O, –20 °C, 6 h
OSiMe3
Ac2O 74%
OAc
Scheme 3 Enolate trapping experiment
Asymmetric Cu-Catalyzed Conjugate Addition of
RMgBr to Enals
Table 3 Optimization of the ACA of Ethylmagnesium Bromide to
trans-Decenal
EtMgBr (2 equiv)
CuTC (5 mol%)
L* (5.25 mol%)
Grignard reagents seem more interesting than dior-
ganozincs if we consider commercial diversity. This is why
we envisioned them in the conjugate addition to enals, and
first to trans-decenal S1 (Table 3). Due to their higher reac-
tivity, regioselectivity may be more difficult to control lead-
ing to complex mixtures of 1,4- and 1,2-adducts as well as
aldol products. For this reason, the reactions were carried
out at –78 °C and gave excellent conversions in almost all
cases. (R)-BINAP (L1) and (R)-TolBINAP (L2) were first test-
ed using CuTC [copper(I) 2-thiophenecarboxylate] as cata-
lyst and EtMgBr (entries 1 and 2). However, good enantio-
selectivities (around 90%) were counterbalanced by low re-
gioselectivities. By adding TMSCl (trimethylchlorosilane), it
was possible to reach a good 1,4/1,2 ratio of 85:15 along
with 90% ee, with L2 as ligand (entry 3). Several bidentate
ligands were then considered to enhance the regio- and en-
antioselectivities. (R)-SYNPHOS (L6) gave a quite similar
1,4/1,2 ratio but with a lower ee of 67% (entry 5). Finally, al-
most only the 1,2-adduct was obtained when (R)-SEGPHOS
derivatives or ferrocene-type ligands were used (entries 6–
11).
O
Et2O, –78 °C, 8 h
S1
O
A1
Entry
Ligand L* Additive
Conv. (%)a Ratio 1,4/1,2/aldol ee (%)b,c
1
2
L1
–
98
99
100
96
84
95
98
95
94
87
86
62:38:0
32:42:26
85:15:0
38:56:6
79:21:0
4:96:0
89
L2
–
92
3
L2
TMSCl
TMSCl
TMSCl
TMSCl
TMSCl
TMSCl
TMSCl
TMSCl
TMSCl
90
4
L3
86
5
L6
67
6
L11
L12
L13
L14
L17
L18
n.d.
n.d.
70
7
3:97:0
8
13:82:5
18:82:0
8:86:6
9
86
10
11
n.d.
87
24:76:0
a Determined by 1H NMR spectroscopy.
b Determined by chiral GC.
c n.d. = not determined.
As shown above, adding TMSCl was crucial to access sat-
isfactory regioselectivity. Indeed, since the pioneering work
of Normant,18 TMSCl has been known to promote 1,4-addi-
tion to enals. It also increases the rate of conjugate addition
of the cuprate.19 In an effort to further improve these re-
sults, various halosilanes were investigated in the conjugate
addition (Table 4). We hypothesized that a more reactive si-
lane would trap the enolate faster, increasing the rate of re-
ductive elimination and thus favor the 1,4-addition. Unfor-
tunately, this was not the case, as raising reactivity favored
the 1,2-process (entries 2–4). TMSOTf was even too reac-
tive, leading to complete degradation of the enal (entry 5).
The last attempts to improve regioselectivity consisted of
testing other solvents with TMSCl as additive. Neither tolu-
ene, nor CH2Cl2, nor MeTHF (2-methyltetrahydrofuran)
gave better results than Et2O (entries 6–8). In all the cases,
1,2-addition was preferred.
in Et2O at –78 °C. We next applied them to various enals in
the reaction with EtMgBr (Table 5, entries 1–7). Enals bear-
ing a primary alkyl group in the β-position (S1, S2, S3, and
S6) provided the best ee’s (up to 90%) with the 1,4-regiose-
lectivity from 71 to 85% (entries 1–3 and 6). As a general
trend, the regioselectivity remains moderate despite the
benefit provided by TMSCl. A decrease of the chiral induc-
tion was observed for substrates having a secondary alkyl
moiety (entries 4 and 5). Finally, the use of cinnamaldehyde
S7 resulted in an important drop both in terms of regio-
and enantioselectivity (entry 7). Then the addition of a
methyl group in β-position was investigated by employing
MeMgBr as Grignard reagent (entries 8–13). In all cases,
both 1,4-regio- and enantioselectivity were inferior, except
for the phenethyl-substituted S6. The latter gave 1,4-adduct
B6 with a slightly better ee of 88% (entry 12). It has to be
noted that B6 is a floral fragrance called Citralis®.
The best conditions for the conjugate addition of
Grignard reagents to enals were thus identified: CuTC as
catalyst, (R)-TolBINAP (L2) as ligand, and TMSCl as additive
© Georg Thieme Verlag Stuttgart · New York — Synthesis 2016, 48, A–H

E
S. Goncalves-Contal et al.
Paper
Synthesis
Table 4 Influence of the Halosilane Additive and the Solvent
Lastly, various Grignard reagents were added to trans-2-
pentenal (S8) and crotonaldehyde (S9), with the aim of get-
ting the enantiomers of the 1,4-adducts obtained previous-
ly by addition of EtMgBr and MeMgBr (Table 6). Concerning
the conjugate addition to S8, three 1,4-adducts have a bet-
ter ee than their corresponding enantiomers obtained pre-
viously by addition of EtMgBr. Cyclohexylmagnesium bro-
mide gave access to the adduct ent-A5 with 86% ee, however
with lower regioselectivity compared to A5, obtained in
80% ee and a 1,4/1,2 ratio of 63:37 (Table 5, entry 5 vs Table
6, entry 3). Secondly, the addition of phenethylmagnesium
bromide to S8 took place in 88% ee and a 1,4/1,2 ratio of
67:33, whereas the addition of EtMgBr to S6 proceeded in
83% ee and a 1,4-regioselectivity of 77% (Table 6, entry 4 vs
Table 5, entry 6). A great improvement was finally obtained
by addition of phenylmagnesium bromide. The ee increased
from 53 to 80% but, still, with a low 1,4-regioselectivity (Ta-
ble 6, entry 5 vs Table 5, entry 7). For the conjugate addition
to an enal bearing a methyl in β-position S9, cyclohexyl-
magnesium bromide gave the best result: 1,4-adduct ent-
B5 was obtained in 89% ee and a 1,4-regioselectivity of 71%
(Table 6, entry 8). Both the regio- and enantioselectivity
were better than the ones observed for the addition of
MeMgBr to S5 (Table 5, entry 11).
EtMgBr (2 equiv)
CuTC (5 mol%)
L2 (5.25 mol%)
TMSX (1.3 equiv)
solvent
–78 °C, 8 h
O
O
Entry
TMSX
Solvent
Conv. (%)a Ratio 1,4/1,2a ee (%)b,c
1
TMSCl
TMSBr
TMSI
Et2O
100
100
60
85:15
43:57
10:90
7:93
90
90
4
2
Et2O
3d
4d
5d
6
Et2O
TMSCN
TMSOTf
TMSCl
TMSCl
TMSCl
Et2O
94
n.d.
n.d.
77
70
56
Et2O
100e
100
100
95
n.d.
toluene
CH2Cl2
MeTHF
2:98
7
43:57
16:84
8
a Conversion (conv.) was determined by 1H NMR spectroscopy.
b The ee was determined by chiral GC.
c n.d. = not determined.
d TMSX was added at –78 °C to the copper complex, followed by trans-
decenal 20 min thereafter.
e Complete degradation of the enal.
Table 5 ACA of Methylmagnesium Bromide and Ethylmagnesium Bro-
mide to Various Enals
Table 6 ACA of Various Grignard Reagents RMgBr to Enals S8 and S9
CuTC (5 mol%)
(R)-TolBINAP L2 (5.25 mol%)
CuTC (5 mol%)
(R)-TolBINAP L2 (5.25 mol%)
R
TMSCl (1.3 equiv)
R
TMSCl (1.3 equiv)
R1
+
RMgBr
O
+
RMgBr
R1
R1
O
O
Et2O, –78 °C, 6 h
R1
(1.5 equiv)
Et2O, –78 °C, 6 h
O
(1.5 equiv)
S8 : R1 = Et
S9 : R1 = Me
ent-A1–ent-A7 : R1 = Et
ent-B1–ent-B7 : R1 = Me
S1–S7
A1–A7 : R = Et
B1–B7 : R = Me
R1
Adduct Conv. (%)a
Ratio 1,4/1,2 ee (%)b,c
Entry Enal
R
Adduct
Conv. (%)a Ratio 1,4/1,2 ee (%)b
Entry
1
2
S8
S8
S8
S8
S8
S9
S9
S9
S9
S9
i-Bu
i-Pr
ent-A3
ent-A4
ent-A5
ent-A6
ent-A7
ent-B2
ent-B4
ent-B5
ent-B6
ent-B7
100
36:64
67:33
53:47
67:33
33:67
36:74
63:37
71:29
45:55
26:74
80
74
86
88
80
88
65
89
74
48
1
2
(CH2)6Me
n-Bu
A1
A2
A3
A4
A5
A6
A7
B1
B2
B4
B5
B6
B7
100 (75)d
100 (46)d
100 (44)d
100
85:15
60:40
71:21
36:64
63:37
77:23
20:80
65:35
40:60
43:57
63:37
48:52
10:90
90
90
90
74
80
83
53
81
86
84
77
88
n.d.
100
3
Cy
100
3
i-Bu
4
(CH2)2Ph
Ph
100 (50)c
88
4
i-Pr
5
5
Cy
100 (51)d
100 (59)d
100
6
n-Bu
i-Pr
100
6
(CH2)2Ph
Ph
7
100
7
8
Cy
100 (60)c
100
8
(CH2)6Me
n-Bu
100 (40)d
100
9
(CH2)2Ph
Ph
9
10
96
10
11
12
13
i-Pr
86
a Conversion (conv.) was determined by 1H NMR spectroscopy.
b The ee was determined by chiral GC.
Cy
100 (49)d
97
(CH2)2Ph
Ph
c Isolated yield given in parentheses.
100
a Conversion (conv.) was determined by 1H NMR spectroscopy.
b The ee was determined by chiral GC.
c n.d. = not determined.
d Isolated yield given parentheses.
© Georg Thieme Verlag Stuttgart · New York — Synthesis 2016, 48, A–H

F
S. Goncalves-Contal et al.
Paper
Synthesis
Asymmetric Cu-Catalyzed Conjugate Addition of
Me3Al to Enals
Applications of the Asymmetric Cu-Catalyzed Conju-
gate Addition to Enals
To explore the limits of the methodology, we decided to
study again the addition of triorganoaluminum. Previously,
we reported that AlR3 gave 1,2-adducts and aldol products
in substantial amounts.13 At that time, we did not manage
to get any 1,4-adduct. To overcome this failure, the reaction
was studied again on cinnamaldehyde (S7) by using stan-
dard conditions for the addition of Me3Al.12 A promising re-
sult was observed with (R)-BINAP (L1) as ligand and CuTC
as catalyst in THF at –78 °C (Table 7, entry 1). The reaction
gave 1,4-adduct B7 in 66% ee, along with 1,2 and aldol prod-
ucts. (R)-SYNPHOS (L6) and (R)-SEGPHOS (L11) were then
considered. The latter gave B7 with incomplete conversion,
excellent 96% ee and improved 1,4/1,2 ratio of 77:23 (entry
3). With this ligand, no traces of the aldol product were ob-
served. Therefore, Me3Al appeared more efficient than
Me2Zn or MeMgBr for the conjugate addition on cinnamal-
dehyde (96% ee against 62%). Next, several solvents were
studied with the aim of enhancing the 1,4-regioselectivity
while maintaining the excellent enantiomeric excess. In all
cases, the ee remained attractive even if it dropped slightly
(entries 4–7). TMSCl was also tested as an additive, in the
hope that it would be as beneficial as before. Conversion
was better, but the 1,4-regioselectivity slightly decreased to
66% and the ee fell to 8% (entry 8). Unlike the addition of
Grignard reagents, TMSCl was deleterious to the addition of
Me3Al. Finally, the conversion was not improved by using
five equivalents of Me3Al (entry 9).
(R)-Citronellal was first synthesized by distillation of
geraniol in the presence of copper by Treibs,20 with the pro-
cess further improved by Kapabas and Kogami.21 In 1983,
Mangeney et al.22 reported its preparation by conjugate ad-
dition of organocuprates on chiral oxazolidines with 85% ee.
Other method consisted of the asymmetric catalytic isom-
erization of N,N′-diethylgeranylamine23 or geraniol,24 giving
rise to (R)-citronellal in 98% and 82% ee, respectively. As no
other conjugate addition has been tried since 1983, we de-
cided to evaluate the potential of our methodology for the
preparation of this natural compound. The asymmetric
conjugate addition of a Grignard reagent to S9 was elected
for this purpose (Scheme 4). Applying the conditions previ-
ously developed on S9 gave the desired (R)-citronellal in
87% ee and a 1,4/1,2 ratio of 28:71. Satisfactorily, the ee ob-
tained is in the range of the reported ones. However, the
1,4-regioselectivity dropped significantly compared to pre-
vious results. This may be the result of the dilution of the
reaction medium due to the lower concentration of the pre-
pared Grignard reagent (1 M) compared to commercially
available solutions (usually 3 M).
CuTC (5 mol%)
(R)-TolBINAP L2 (5.25 mol%)
MgBr
TMSCl (1.3 equiv)
+
O
Et2O, –78 °C, 8 h
conv. >99%
O
(R)-citronellal
1,4:1,2 ratio = 28:71
87% ee
S9
Scheme 4 Synthesis of (R)-citronellal by ACA to S9
Table 7 Evaluation of the ACA of Trimethylaluminum on Cinnamalde-
hyde (S7)
We were also interested in (S)-florhydral, an olfactory
compound discovered in 1989 by Chalk.25 The first synthe-
sis was reported by Brenna et al.26 and was based on enzy-
matic catalysis. Seven steps were required to access (S)-flor-
hydral in a low yield of 3% and excellent 97% ee. In 2006,
Paganelli27 reported the hydroformylation of 3-isoprenyl-
benzene derivatives to get the compound but with an ex-
tremely low ee of 5%. Two years later, Stadler28 described a
new pathway based on an asymmetric organocatalytic hy-
drogenation that provided (S)-florhydral in 98% ee and 39%
yield. We also reported a total synthesis13 of this natural
compound in eight steps with an excellent 99.5% ee. The key
step was the asymmetric copper-catalyzed conjugate addi-
tion of Me3Al to a β,γ-unsaturated α-keto ester. However,
we thought that (S)-florhydral could also be obtained by
conjugate addition to the appropriate α,β-unsaturated alde-
hyde. Therefore, we envisioned applying one of the previ-
ously developed methodologies to access (S)-florhydral
more quickly (Scheme 5).
CuTC (5 mol%)
L* (5.25 mol%)
Me
O
+
Me3Al
O
solvent
–78 °C, 17 h
(2 equiv)
S7
B7
Entry Ligand Solvent
Conv. (%)a Ratio 1,4/1,2/aldola ee (%)b
L*
1
L1
THF
83
56
66
51
32
34
65
83
63
65:19:16
64:36:0
77:23:0
28:72:0
39:61:0
38:62:0
47:53:0
66:34:0
60:0:40
66
87
96
84
85
90
93
8
2
L6
THF
3
L11
L11
L11
L11
L11
L11
L11
THF
4
EtOAc
5
dioxane
THF/dioxanec
THF/EtOAcc
THF
6
7
8d
9e
THF
91
a Conversion (conv.) was determined by 1H NMR spectroscopy.
b The ee was determined by chiral GC.
c Solvent ratio 9:1.
The synthesis began with the preparation of 1-bromo-
3-isopropylbenzene. This was done in two steps according
to a procedure described by Voskoboynikov,29 starting from
d Reaction done in the presence of TMSCl (1.3 equiv).
e Reaction done with Me3Al (5 equiv).
© Georg Thieme Verlag Stuttgart · New York — Synthesis 2016, 48, A–H

G
S. Goncalves-Contal et al.
Paper
Synthesis
NH2
NH2
Br
1. Ac2O, AcOH
2. Br2, AcOH
1. NaNO2, H2SO4
2. Cu reflux
Br
1.
CH(OEt)2
3. KOH
Bu4NOAc, K2CO3
KCl, Pd(OAc)2, DMF
2. 2 N HCl
75%
77%
Me3Al (2 equiv)
CuTC (5 mol%)
(R)-SEGPHOS L11 (5.25 mol%)
conv. >99%
1,4:1,2 ratio = 37:63
94% ee
THF, –78 °C, 17 h
O
O
(S)-florhydral
71%
Me2Zn (2 equiv)
CuTC (5 mol%)
(R)-SYNPHOS L6 (5.25 mol%)
conv. 89%
1,4:1,2 ratio = 57:43
96% ee
Et2O, –78 °C, 17 h
Scheme 5 Synthesis of (S)-florhydral by ACA
p-isopropylaniline. A Heck reaction followed by hydrolysis
was then carried out under classic conditions, using acro-
lein diethyl acetal, affording the desired α,β-unsaturated al-
dehyde in 71% yield. The enal was first engaged in conjugate
addition with Me3Al. By applying the developed conditions,
(S)-florhydral was obtained in a very good 94% ee but a low
1,4-regioselectivity. The conjugate addition of Me2Zn ap-
peared better, as it provided the desired compound in 96%
ee and better 1,4-regioselectivity (57% against 37%)
(Scheme 5).
We have reported herein our complete studies on the
asymmetric copper-catalyzed conjugate addition of or-
ganometallic reagents to α,β-unsaturated aldehydes. Three
sets of conditions were successfully developed. The first
one consisted of the addition of organozinc reagents. The
use of (R)-H8BINAP as ligand and CuTC as catalyst provided
1,4-adducts in good to excellent ee’s (up to 95%) and mod-
erate to excellent 1,4/1,2 ratios (up to 100:0). The second
one concerned the addition of Grignard reagents in the
presence of (R)-TolBINAP as ligand and TMSCl as additive.
Under these conditions, several enals successfully gave rise
to β-substituted aldehydes in ee’s up to 90%. However, the
regioselectivity remained moderate to good. In a last set of
experiments, we developed the conjugate addition of
trimethylaluminum. Here, the ligand of choice was (R)-SEG-
PHOS, which allowed the formation of the desired 1,4-ad-
duct in excellent 96% ee and a good 1,4/1,2 ratio of 77:23.
Finally, we applied our methodologies to the synthesis of
natural compounds. We were able to access (R)-citronellal
as well as (S)-florhydral in a few steps with very good ee’s
(87% and 96% respectively).
the corresponding bromide to magnesium or were commercially
available (EtMgBr, MeMgBr, and PhMgBr; 3 M in Et2O); commercially
available Me2Zn (1.2 M in toluene), Et2Zn (1 M in hexane), and Me3Al
(2 M in heptane) were also used. Phosphoramidite ligand studies as
well as GC chromatograms are reported in the Supporting Informa-
tion. All compounds in this paper have already been reported in the
literature (see Supporting Information).
ACA with Dialkylzinc Reagents; General Procedure
A solution of CuTC (5 mol%) and (R)-H8BINAP (L3; 5.25 mol%) in Et2O
(1 mL) was stirred for 20 min at r.t., followed by 20 min at 0 °C or –20
°C. R2Zn (2 equiv) was then added, and after 15 min, a solution of the
enal (0.25 mmol) in Et2O (0.5 mL) was slowly added over 1 h. The re-
action mixture was stirred at 0 °C or –20 °C for 6–17 h before being
quenched by 1 M aq HCl. The aqueous phase was extracted by Et2O (3
× 5 mL). The combined organic extracts were dried over MgSO4, fil-
tered, and concentrated to give the crude α,β-unsaturated aldehydes.
The products were isolated either directly by chromatography (silica
gel, pentane–Et2O) or after reduction to the corresponding alcohol
(using NaBH4 in a THF–MeOH mixture).
ACA with Grignard Reagents; General Procedure
A solution of CuTC (5 mol%) and (R)-TolBINAP (L2; 5.25 mol%) in Et2O
(1 mL) was stirred for 20 min at r.t., followed by 20 min at –78 °C.
RMgBr (2 equiv) was then added, and after 15 min, a solution of the
enal (0.25 mmol) and TMSCl (1.3 equiv) in Et2O (0.5 mL) was slowly
added over 1 h. The reaction mixture was stirred at –78 °C for 8 h be-
fore being quenched by 1 M aq HCl–MeOH. The aqueous phase was
extracted by Et2O (3 × 5 mL). The combined organic extracts were
dried over MgSO4, filtered, and concentrated to give the crude α,β-un-
saturated aldehydes. The products were isolated either directly by
chromatography (silica gel, pentane–Et2O) or after reduction to the
corresponding alcohol (using NaBH4 in a mixture of THF and MeOH).
ACA with Me3Al; General Procedure
A solution of CuTC (5 mol%) and (R)-SEGPHOS (L11; 5.25 mol%) in THF
(1 mL) was stirred for 20 min at r.t., followed by 20 min at –78 °C.
Me3Al (2 equiv) was then added, and after 30 min, a solution of cin-
namaldehyde (S7; 0.25 mmol) in THF (0.5 mL) was slowly added over
1 h. The reaction mixture was stirred at –78 °C for 17 h before being
All reagents were purchased from Sigma-Aldrich and Acros or Fron-
tierScientific and were used as received. Solvents were dried by filtra-
tion over alumina previously activated at 350 °C for 12 h under N2 be-
fore use. Grignard reagents were synthesized in Et2O by addition of
© Georg Thieme Verlag Stuttgart · New York — Synthesis 2016, 48, A–H

H
S. Goncalves-Contal et al.
Paper
Synthesis
quenched by 1 M aq HCl. The aqueous phase was extracted by Et2O (3
× 5 mL). The combined organic extracts were dried over MgSO4, fil-
tered, and concentrated to give β-methyl-substituted aldehyde B7.
(10) Fañanas-Mastral, M.; Feringa, B. L. J. Am. Chem. Soc. 2010, 132,
13152.
(11) Mao, B.; Fañanas-Mastral, M.; Feringa, B. L. Org. Lett. 2013, 15,
286.
(12) Gremaud, L.; Alexakis, A. Angew. Chem. Int. Ed. 2012, 51, 794.
(13) Palais, L.; Babel, L.; Quintard, A.; Belot, S.; Alexakis, A. Org. Lett.
2010, 12, 1988.
Acknowledgment
(14) Knochel, P.; Singer, R. D. Chem. Rev. 1993, 93, 2117.
(15) Phosphoramidites were also envisioned as ligands in these
studies (see ref. 12) but appeared less attractive, as they are not
commercially available. Nevertheless, results are presented in
the Supporting Information.
We thank the Swiss National Research Foundation (grant N° 200020-
126663) and COST action D40 (SER contract N° C07.0097) for finan-
cial support.
Supporting Information
(16) Knopf, O.; Alexakis, A. Org. Lett. 2002, 4, 3835.
(17) Vuagnoux-d’Augustin, M.; Alexakis, A. Tetrahedron Lett. 2007,
48, 7408.
(18) Chuit, C.; Foulon, J.-P.; Normant, J. F. Tetrahedron 1980, 36,
2305.
Supporting information for this article is available online at
http://dx.doi.org/10.1055/s-0035-1562487.
S
u
p
p
ortioInfgrmoaitn
S
u
p
p
ortiInfogrmoaitn
(19) (a) Corey, E. J.; Boaz, N. W. Tetrahedron Lett. 1985, 26, 6015.
(b) Corey, E. J.; Boaz, N. W. Tetrahedron Lett. 1985, 26, 6019.
(c) Alexakis, A.; Berlan, J.; Besace, Y. Tetrahedron Lett. 1986, 27,
1047. (d) Nakamura, E.; Matsuzawa, S.; Horiguchi, Y.; Kuwajima,
I. Tetrahedron Lett. 1986, 27, 4029.
(20) Treibs, W.; Schmidt, H. Ber. Dtsch. Chem. Ges. 1927, 60, 2335.
(21) (a) Kapabas, B. H. US Patent 118498, 1958. (b) Kogami, K.;
Kumanota, J. Bull. Chem. Soc. Jpn. 1968, 41, 2508.
(22) (a) Mangeney, P.; Alexakis, A.; Normant, J. F. Tetrahedron Lett.
1983, 24, 373. (b) Mangeney, P.; Alexakis, A.; Normant, J. Tetra-
hedron 1984, 40, 1803.
(23) Akutagawa, S. In Comprehensive Asymmetric Catalysis; Vol. 3;
Jacobsen, E. N.; Pfaltz, A.; Yamamoto, H., Eds.; Springer: Berlin,
1999, Chap. 41.4.
(24) (a) Mantilli, L.; Gerard, D.; Torche, S.; Besnard, C.; Mazet, C.
Chem. Eur. J. 2010, 16, 12736. (b) Mantilli, L.; Mazet, C. Chem.
Commun. 2010, 46, 445.
(25) Chalk, A. J. EP 0368156, 1989.
(26) Abate, A.; Brenna, E.; Negri, C. D.; Fuganti, C.; Serra, S. Tetrahe-
dron: Asymmetry 2002, 13, 899.
(27) Paganelli, S.; Ciappa, A.; Marchetti, M.; Scrivanti, A.; Matteoli, U.
J. Mol. Catal. A 2006, 247, 138.
(28) Stadler, M.; List, B. Synlett 2008, 597.
(29) Izmer, V. V.; Lebedev, A. Y.; Nikulin, M. V.; Ryabov, A. N.;
Asachenko, A. F.; Lygin, A. V.; Sorokin, D. A.; Voskoboynikov, A.
Z. Organometallics 2006, 25, 1217.
References
(1) (a) Alexakis, A.; Benhaim, C. Eur. J. Org. Chem. 2002, 3221.
(b) Alexakis, A.; Bäckvall, J. E.; Krause, N.; Pamies, O.; Dieguez,
M. Chem. Rev. 2008, 108, 2796. (c) Harutyunyan, S. R.; Den
Hartog, T.; Geurts, K.; Minnaard, A. J.; Feringa, B. Chem. Rev.
2008, 108, 2824.
(2) (a) Bräse, S.; Höfener, S. Angew. Chem. Int. Ed. 2005, 44, 7879.
(b) Ay, S.; Nieger, M.; Bräse, S. Chem. Eur. J. 2008, 14, 11539.
(c) For a racemic example without copper, see: Jones, P.; Reddy,
C. K.; Knochel, P. Tetrahedron 1998, 54, 1471.
(3) Marshall, J. A.; Herold, M.; Eidam, H. S.; Eidam, P. Org. Lett.
2006, 8, 5505.
(4) Ibrahem, I.; Santoro, S.; Himo, F.; Còrdova, A. Adv. Synth. Catal.
2011, 353, 245.
(5) As a reference for Rh-catalyzed ACA to enals, see: (a) Itooka, R.;
Iguchi, Y.; Miyaura, N. J. Org. Chem. 2003, 68, 6000. (b) Paquin,
J.-F.; Defieber, C.; Stephenson, C. R. J.; Carreira, E. M. J. Am.
Chem. Soc. 2005, 127, 10850. (c) Tokunaga, N.; Hayashi, T. Tetra-
hedron: Asymmetry 2006, 17, 607.
(6) As a reference for Pd-catalyzed ACA to enals, see: Nishikata, T.;
Yamamoto, Y.; Miyaura, N. Chem. Commun. 2004, 1822.
(7) Hird, A. W.; Hoveyda, A. H. Angew. Chem. Int. Ed. 2003, 42, 1276.
(8) Des Mazery, R.; Pullez, M.; Lòpez, F.; Harutyunyan, S. R.;
Minnaard, A. J.; Feringa, B. L. J. Am. Chem. Soc. 2005, 127, 9966.
(9) Garcìa, J. M.; Gonzàlez, A.; Kardak, B. G.; Odriozola, J. M.;
Oiarbide, M.; Razkin, J.; Palomo, C. Chem. Eur. J. 2008, 14, 8768.
© Georg Thieme Verlag Stuttgart · New York — Synthesis 2016, 48, A–H