Enyne chlorides: Substrates for copper-catalyzed asymmetric allylic alkylation

Article abstract of DOI:10.1002/anie.201107129

A select few: Several prochiral enyne chlorides were employed as substrates in the title reaction using Grignard reagents as the alkylation reagents (see scheme; CuTC=copper(I) thiophenecarboxylate). Excellent 1,3 substitution regioselectivities and good to excellent enantioselectivities were obtained. The substrate scope is additionally extended to diene chlorides. Copyright

Full text of DOI:10.1002/anie.201107129

Angewandte
Chemie
DOI: 10.1002/anie.201107129
Asymmetric Catalysis
Enyne Chlorides: Substrates for Copper-Catalyzed Asymmetric Allylic
Alkylation**
Hailing Li and Alexandre Alexakis*
Copper-catalyzed asymmetric allylic alkylation (AAA) is one
of the most useful and efficient carbon–carbon bond-forming
methods that leads to optically enriched molecules.[1] Owing
to the properties of copper, this transformation allows the use
of nonstabilized carbon nucleophiles, namely alkyl or aryl
groups in the form of organometallic species. After the
pioneering work of van Koten, Bꢀckvall, and co-workers on
copper-catalyzed AAA using Grignard reagents,[2] much
attention has focused on this field. Compared to other
organometallic reagents, such as organozinc or organoalumi-
nium reagents, Grignard reagents are advantageous under
several circumstances because of their diversity and avail-
ability.
the more vulnerable and synthetically interesting prochiral
substrates, which generate chiral products, 1,4-enynes in this
case, as a result of the AAA process (Scheme 1b)
As far as we know, the AAA on the extended multiple
bond system, especially the enyne substrates, has not been
studied much.[6] Trost et al. have performed an enantioselec-
tive molybdenum-catalyzed process on enyne carbonates,
with malonate salts as nucleophiles; good regioselectivities
were observed in favor of the 1,3-substitution together with
excellent ee values (Scheme 2).[7] One year later, Takeuchi
In the past few years, our group has reported highly regio-
and enantioselective copper/phosphoramidite catalytic sys-
tems for AAA of allylic chlorides with organomagnesium
reagents (Scheme 1a);[3] later, the range of substrates was
Scheme 2. Possible products by substitution reaction.
et al. described an iridium-catalyzed allylic substitution
system; however the regioselectivity for the linear enyne
acetate substrate was moderate.[8] The very first example to
introduce alkyl groups, was reported by Krause et al. By
performing the substitution reactions on enyne acetates and
employing the readily in situ generated cuprate, they
observed the exclusive 1,5-substitution, thus directly access-
ing vinylallenes, without the formation of products arising
from substitution at the a or g position (SN2 or SNꢁ) of the
allylic system.[9] Subsequent to this work a highly enantiose-
lective process, using enantiopure enyne acetates under
“remote stereocontrol” to introduce the chirality from the
starting materials to the adducts, was described.[10] Although
the formation of allenes was taken for granted, we were
intrigued by a report from Hoveyda et al. describing a copper-
catalyzed AAA on two examples of enyne phosphate
substrates using diethylzinc, which lead to the formation of
tertiary and quaternary stereogenic carbon centers.[11]
Herein, we report a series of enyne chloride substrates
that undergo the copper-catalyzed AAA by employing
Grignard reagents to provide excellent regioselectivities for
the 1,3-allylic substitution products without any trace of
vinylallene formation, and excellent enantiomeric excesses.
To our knowledge, these results represented the first catalytic
asymmetric 1,3-substitution on this type of enyne substrate by
using a variety of Grignard reagents as nucleophiles without
Scheme 1. a) Previous work. b) Proposed use of enyne chlorides as
new substrates for AAA.
extended to include b-disubstituted allylic chlorides, 1,4-
difunctionalized allylic type substrates, etc.[4] Additionally, the
dynamic kinetic asymmetric transformation process was also
studied, and involved using racemic cyclic or acyclic sub-
strates under copper-catalyzed AAA conditions with phos-
phoramidite ligands.[5] Subsequently, we focused on the use of
[*] H. Li, Prof. Dr. A. Alexakis
Department of Organic Chemistry, University of Geneva
Quai Ernest Ansermet 30, 1211 Geneva (Switzerland)
E-mail: alexandre.alexakis@unige.ch
[**] This work is supported by the Swiss National Research Foundation
(grant No. 200020-126663) and COSTaction D40 (SER contract No.
C07.0097). The authors warmly thank Dr. Jean-Baptiste Langlois for
his valuable advice. We also thank BASF for the generous gift of
chiral amines.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201107129.
Angew. Chem. Int. Ed. 2012, 51, 1055 –1058
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1055

.
Angewandte
Communications
reacting with the triple bonds. As Hoveyda and co-workers
recently published a reversed process of a copper-catalyzed
AAA using alkynyl aluminium reagents as nucleophiles to
directly introduce the triple bond in the allylic substitution to
provide the 1,4-enynes,[12] our methodology can be also
considered as an easily accessible alternative to achieving
the 1,4-enyne products bearing tertiary chiral carbon centers.
Our preliminary investigation started from the AAA
reaction of (E)-1-chloronon-2-en-4-yne (1) with ethylmagne-
sium bromide in dichloromethane at À788C. Under the
reaction conditions published by our group for the cinnamyl
chloride substrate,[3] a combination of 3 mol% copper(I)
thiophenecarboxylate (CuTC) and the phosphoramidite
ligand L1 served as the catalyst.[3c] After 4 hours full
conversion was observed together with good regioselectivity
(SN2’/SN2 = 98:2) and in 96% ee (Table 1, entry 1). An
increase in catalyst loading to 5 mol% led to a negligible
combination with CuBr·SMe2, as reported by Feringa and co-
workers,[13] provided both poorer regio- and enantioselectiv-
ity. Furthermore, full conversion was not observed after an
even longer reaction time. It should be added that the
substrate 13 (see Table 3 for structure) gave only the allene
and the SN2 adduct when reacted with Bu2CuLi under the
reaction conditions described by Krause et al.[9]
These observations indicate that this methodology should
be such a robust catalytic process, as it is not dependant on the
copper source or the catalyst loading. We have shown that:
1) decrease of catalyst loadings (from 5 mol% to 1 mol%)
has no significant results influence on regio- and enantiose-
lectivity, and this can be useful for scale-up or industrial
processes; and 2) a cheaper copper source and the more
simple and available ligand L4 provided similar results in
terms of reactivity and selectivity. As a result, 5 mol% CuTC
with 5.5 mol% L2 were chosen as the optimized reaction
conditions for further studies.
Table 1: Optimization of the methodology.[a]
A range of alkylmagnesium bromide reagents were
employed to target the corresponding SN2’ products, and
demonstrate the advantage of using Grignard reagents over
alkylzinc reagents (Table 2). The employment of simple
primary alkyl Grignard reagents provided (entries 1–3)
excellent regio- and enantioselectivities in all the cases.
Those primary alkyl nucleophiles bearing a double bond
(entries 4 and 5) showed no interference with the selectivity.
More-bulky groups, such as the secondary alkyl nucleophiles,
led to a reasonable drop in enantioselectivity, but maintained
the excellent SN2’/SN2 ratios (entries 6 and 7). However, in the
case of phenyl magnesium bromide, the SN2 product was
obtained nearly exclusively (entry 8), thus illustrating the
special character of the phenyl nucleophile in the copper-
catalyzed AAA. The more challenging methyl group was
Entry
L
Cu salt
Catalyst
loading [%] [%][b]
Conv. SN2’/SN2[b]
ee
[%][c]
1
2
3
4
5
6
L1 CuTC
L1 CuTC
L1 CuTC
L2 CuTC
L3 CuTC
L4 CuTC
L1 CuBr·SMe2
L5 CuBr·SMe2
3
5
1
5
5
5
5
5
>99
>99
99
95
99
>99
>99
72
98:2
98:2
94:6
83:17
88:12
95:5
97:3
96
97
95
63
75.5
95
96
94
Table 2: Scope of Grignard reagents.[a]
7
8[d]
90:10
Entry[a]
R
Prod. Yield [%] SN2’/SN2[b] ee [%][c]
[a] Reaction conditions: the substrate (0.25 mmol) was added to a
1
2
Et
nBu
2
3
4
5
6
7
8
9
72
84
92
92
90
76
96
–
98:2
98:2
98:2
95:5
96:4
97
96
>99
95
95
85
solution of CuTC and chiral ligand in dry CH2Cl2 at À788C. The ethereal
solution of the Grignard reagent (1.2 equiv) was added dropwise over a
30 min period and the reaction mixture was stirred at À788C for 4 h.
[b] Determined by 1H NMR spectroscopy. [c] Determined by GC analysis
using a chiral stationary phase. [d] Overnight (18 h).
3[d]
4
PhCH2CH2
CH2 CHCH2CH2
CH2 C(CH3)CH2CH2
iPr
Cy
=
=
5
6
7
8
99:1
98:2
87
–
Ph
Me
<1:99
47:53[e]
93
improvement of the enantioselectivity (entry 2). Even with a
1 mol% catalyst loading, the reaction kept rendering an
excellent enantioselectivity with a small decrease in the SN2’/
SN2 ratio (entry 3). The use of ligand L2 led to lowering of
both the regio- and enantioselectivity, and can be reasoned to
arise from the mismatch effect of the prochiral center with the
copper/phosphoramidite complex (entry 4). The diastereo-
isomeric phosphoramidite ligands L3 and L4 were tested, and
provided no further improvement (entries 5 and 6). Replacing
the CuTC by CuBr·SMe2 showed a slight decrease in
selectivity. And finally, the use of (R,R)-Taniaphos (L5) in
9
10
–
[a] Reaction conditions: the substrate (0.25 or 0.5 mmol) was added to a
solution of CuTC and chiral ligand in dry CH2Cl2 at À788C. The ethereal
solution of Grignard reagent (1.2 equiv) was added dropwise over a
30 min period and the reaction mixture was stirred at À788C for 4 h.
[b] Determined by 1H NMR spectroscopy. [c] Determined by GC analysis
using a chiral stationary phase. [d] 1.05 equiv of phenethyl Grignard
reagent was used; the desired product 4 is mixed with 5.7 mol%
homocoupling product of phenethyl Grignard reagent, namely, 1,4-
diphenylbutane. Yield calculated from the percentage in the mixture.
[e] Besides the a and g adducts, the Z isomer of the SN2 adduct (ꢀ10%)
was detected in the resulting mixture. Cy=cyclohexyl.
1056
www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 1055 –1058

Angewandte
Chemie
introduced with good enantioselectivity but with an SN2’/SN2
ratio of nearly 1:1, thus showing that the conventional
limitation of AAA reactions can still not be overcome in
this system (entry 9).[14] Finally, a control experiment with the
Z isomer of 1 gave the opposite enantiomer, albeit with a
poor ee value of 32% and a low SN2’/SN2 ratio (79:21).
The generality of the reaction was exploited by employing
different E-configured enyne chloride substrates (Table 3). A
longer carbon chain (R) has the same reactivity (entry 1).
Replacing it by more-bulky alkyl groups such as cyclohexyl or
tert-butyl groups, maintained the good selectivity (entries 2
Scheme 3. Examples of copper-catalyzed AAA using diene chlorides as
substrates.
are very encouraging and additional investigations on this
reaction are underway in the laboratory.
In conclusion, we have developed a method that employs
prochiral E-configured enyne chlorides as substrates in the
copper-catalyzed AAA, wherein different alkyl magnesieum
bromide reagents can be introduced as nucleophiles, thus
leading to interesting chiral 1,4-enyne building blocks. In most
cases excellent regio- and enantioselectivities (SN2’/SN2 ratio
up to 98:2; ee values up to > 99%) were obtained. Addition-
ally, two examples of E,E-diene chlorides were shown to
behave similarly in the catalytic process with very good
selectivities to provide the chiral 1,4-diene products in high
optical purity.
Table 3: Scope of enyne chloride substrates.[a]
Entry[a]
R
Sub. Prod. Yield [%] SN2’/SN2[b] ee [%][c]
1
2
nPent
Cy
tBu
cPropyl
Ph
TMS
11
12
13
14
15
16
18
19
20
21
22
23
24
83
93
80
77
94
74
85
97:3
96:4
99:1
96:4
94:6
96:4
96:4
97
97
97
95
97
97.5
94
3[d]
4
5
6
7
CH2OTBS 17
Experimental Section
[a] Reaction conditions: the substrate (0.25 mmol) was added to a
solution of CuTC and chiral ligand in dry CH2Cl2 at À788C. The ethereal
solution of Grignard reagent was added dropwise over a 30 min period
and the reaction mixture was stirred at À788C for 4 h. [b] Determined by
1H NMR spectroscopy. [c] Determined by GC analysis using a chiral
stationary phase. [d] For convenient GC separation, nBuMgBr was
employed instead of EtMgBr. TBS=tert-butyldimethylsilyl, TMS=tri-
methylsilyl.
A dried Schlenk tube was charged with a copper salt (5 mol%) and
the chiral ligand (5.5 mol%). Dichloromethane (1.5 mL) was added
and the mixture was stirred at room temperature for 10 min. The
allylic chloride (0.25 mmol) was introduced dropwise and the reaction
mixture was stirred at room temperature for an additional 5 min
before cooling the reaction mixture to À788C using an ethanol/dry ice
cold bath. The Grignard reagent (3m in diethyl ether, 1.2 equiv) was
added manually over a 30 min period. Once the addition was
complete the reaction mixture was left at À788C for an additional
4 h. The reaction was quenched by addition of aqueous hydrochloric
acid (1n, 2 mL). Diethyl ether (5 mL) was added and the aqueous
phase was separated and extracted with diethyl ether (3 ꢂ 2 mL). The
combined organic fractions were washed with brine (3 mL), dried
over anhydrous sodium sulfate, filtered, and concentrated in vacuo.
The crude reaction mixture was purified by chromatography on silica
gel using pentane as the eluant. The desired product was recovered as
a colorless oil. For additional details, see the Supporting Information.
and 3); this was also the case for the cyclopropyl group
(entry 4), which is an important moiety present in many
natural products. As shown in entry 5, the introduction of a
phenyl group leads to a more conjugated and rigid structure,
but this does not interfere with the selectivity. A more
versatile substituent, the trimethylsilyl group, also had no
significant influence on the reactivity (entry 6). Finally, a
protected alcohol successfully provided a valuable SN2’
product with high optical purity (entry 7). The scale-up
reaction (2.5 mmol substrate 15) lead to no change in the
regio- and stereoselectivity, and the product yield was 87%.[15]
To additionally test the generality of this method, we
investigated the conjugated E,E-diene chloride substrates,
which maybe a priori more challenging because there is a
reduced electronic discrimination between the 1,3-substitu-
tion versus the 1,5-substitution. We studied two examples
(Scheme 3), and to our delight these E,E-diene chloride
substrates (25 and 26) behaved similarly in the catalytic
reactions as their aforementioned enyne chloride counter-
parts, thus affording almost exclusively the 1,3-substitution
adduct with good regio- and enantioselectivity; note that this
outcome is not common for alkyl nucleophiles,[7,16,17] namely
alkyl Grignard reagents in our case. These preliminary results
Received: October 8, 2011
Published online: December 8, 2011
Keywords: alkylation · asymmetric catalysis · copper · enynes ·
.
Grignard reagents
[1] For recent reviews of copper-catalyzed AAA reactions, see:
a) A. Alexakis, C. Malan, L. Lea, K. Tissot-Croset, D. Polet, C.
Falciola, Chimia 2006, 60, 124; b) H. Yorimitsu, K. Oshima,
Angew. Chem. 2005, 117, 4509; Angew. Chem. Int. Ed. 2005, 44,
4435; c) A. Kar, N. P. Argade, Synthesis 2005, 2995; d) S. R.
Harutyunyan, T. den Hartog, K. Geurts, A. J. Minnaard, B. L.
Feringa, Chem. Rev. 2008, 108, 2824; e) Z. Lu, S. M. Ma, Angew.
Chem. 2008, 120, 264; Angew. Chem. Int. Ed. 2008, 47, 258; f) A.
Alexakis, J. E. Bꢀckvall, N. Krause, O. Pamies, M. Dieguez,
Chem. Rev. 2008, 108, 2796; g) C. A. Falciola, A. Alexakis, Eur. J.
Org. Chem. 2008, 3765.
Angew. Chem. Int. Ed. 2012, 51, 1055 –1058
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1057

.
Angewandte
Communications
[2] M. Van Klaveren, E. S. M. Persson, A. del Villar, D. M. Grove,
J. E. Bꢀckvall, G. van Koten, Tetrahedron Lett. 1995, 36, 3059.
[3] a) A. Alexakis, C. Malan, L. Lea, C. Benhaim, X. Fournioux,
Synlett 2001, 927; b) K. Croset, A. Alexakis, Org. Lett. 2002, 4,
4147; c) K. Tissot-Croset, D. Polet, A. Alexakis, Angew. Chem.
2004, 116, 2480; Angew. Chem. Int. Ed. 2004, 43, 2426; d) K.
Tissot-Croset, D. Polet, S. Gille, C. Hawner, A. Alexakis,
Synthesis 2004, 2586; e) K. Tissot-Croset, A. Alexakis, Tetrahe-
dron Lett. 2004, 45, 7375.
[8] a) R. Takeuchi, K. Tanabe, Angew. Chem. 2000, 112, 2051;
Angew. Chem. Int. Ed. 2000, 39, 1975.
[9] a) N. Krause, A. Gerold, Angew. Chem. 1997, 109, 194; Angew.
Chem. Int. Ed. Engl. 1997, 36, 186; b) M. Purpura, N. Krause,
Eur. J. Org. Chem. 1999, 267.
[10] N. Krause, M. Purpura, Angew. Chem. 2000, 112, 4512; Angew.
Chem. Int. Ed. 2000, 39, 4355.
[11] M. A. Kacprzynski, A. H. Hoveyda, J. Am. Chem. Soc. 2004, 126,
10676.
[4] a) C. A. Falciola, K. Tissot-Croset, A. Alexakis, Angew. Chem.
2006, 118, 6141; Angew. Chem. Int. Ed. 2006, 45, 5995; b) C. A.
Falciola, A. Alexakis, Angew. Chem. 2007, 119, 2673; Angew.
Chem. Int. Ed. 2007, 46, 2619; c) C. A. Falciola, K. Tissot-Croset,
H. Reyneri, A. Alexakis, Adv. Synth. Catal. 2008, 350, 1090;
d) C. A. Falciola, A. Alexakis, Chem. Eur. J. 2008, 14, 10615.
[5] a) J.-B. Langlois, A. Alexakis, Chem. Commun. 2009, 3868; b) J.-
B. Langlois, A. Alexakis, Adv. Synth. Catal. 2010, 352, 447; c) J.-
B. Langlois, A. Alexakis, Angew. Chem. 2011, 123, 1917; Angew.
Chem. Int. Ed. 2011, 50, 1877.
[6] For examples of catalytic carbon–carbon bond formation
processes on enyne substrates, see: a) M. Tissot, D. Mꢃller, S.
Belot, A. Alexakis, Org. Lett. 2010, 12, 2770; b) M. Tissot, A.
Pꢄrez Hernꢅndez, D. Mꢃller, M. Mauduit, A. Alexakis, Org.
Lett. 2011, 13, 1524.
[12] J. A. Dabrowski, F. Gao, A. H. Hoveyda, J. Am. Chem. Soc.
2011, 133, 4778.
[13] a) F. Lꢆpez, A. W. van Zijl, A. J. Minnaard, B. L. Feringa, Chem.
Commun. 2006, 409; For corrected configuration of the Tania-
phos ligand, see: b) A. W. van Zijl, Szymanski, F. Lopez, A. J.
Minnaard, B. L. Feringa, J. Org. Chem. 2008, 73, 6994.
[14] In the case of MeMgBr, reactions were performed under several
different reaction conditions without affording significant
improvement.
[15] For details, see the Supporting Information.
[16] For selected examples using stabilized nucleophiles, see: a) G.
Lipowsky, N. Miller, G. Helmchen, Angew. Chem. 2004, 116,
4695; Angew. Chem. Int. Ed. 2004, 43, 4595; b) W.-H. Zheng, N.
Sun, X.-L. Hou, Org. Lett. 2005, 7, 5151.
[17] For an example of copper-catalyzed process using nonstabilized
alkyl nucleophiles, see Ref. [11].
[7] B. M. Trost, S. Hildbrand, K. Dogra, J. Am. Chem. Soc. 1999, 121,
10416.
1058
www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 1055 –1058