Copper-catalyzed asymmetric allylic substitution with aryl and ethyl Grignard reagents

Article abstract of DOI:10.1039/b809140d

Phenyl- and ethyl-magnesium bromides undergo regioselective asymmetric allylic substitution with high enantioselectivity under the catalysis of chiral amidophosphane-copper(i) complexes. The Royal Society of Chemistry.

Full text of DOI:10.1039/b809140d

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Copper-catalyzed asymmetric allylic substitution with aryl and ethyl
Grignard reagentsw
Khalid B. Selim, Ken-ichi Yamada and Kiyoshi Tomioka*
Received (in Cambridge, UK) 29th May 2008, Accepted 16th July 2008
First published as an Advance Article on the web 9th September 2008
DOI: 10.1039/b809140d
Phenyl- and ethyl-magnesium bromides undergo regioselective
asymmetric allylic substitution with high enantioselectivity
under the catalysis of chiral amidophosphane–copper(I) complexes.
Transition metal-catalyzed asymmetric allylic substitution has
proven to be an efficient method in C–C bond-forming reac-
tions to obtain enantiomerically enriched compounds.1 Exten-
sive accounts report a wide variety of metals (Pd, Ir, Mo, Rh
Fig. 1 Amidophosphane ligands.
magnesium bromides that affords the SN20 product as the
and Ru) that use soft stabilized nucleophiles for the reaction.2
major regioisomer with high enantioselectivity.
In contrast, copper enables the transfer of hard nonstabilized
We started our study with the reaction of cinnamyl bromide
nucleophiles such as alkyl groups in the form of organometallic
(2) (1 mmol) and EtMgBr. Using amidophosphane 1a9d
species to be delivered in the allylic position with high regios-
electivity.3 Rapid growth of Cu-catalyzed asymmetric allylic
alkylation (AAA) over the past decade has occurred. By using
Grignard or dialkylzinc reagents with unsymmetrically substi-
tuted allylic substrates, a branched chiral product (SN20 pro-
duct) could be obtained in high regio- and stereocontrol with
excellent enantioselectivity.4 However, the transfer of sp2 and
sp carbon have not been well exploited and most of the reports
were limited to symmetrically substituted allylic substrates
utilizing transition metals other than copper as catalysts.5
Recently, Hoveyda and co-workers have developed an
efficient Cu-catalyzed AAA with diarylzinc as well as vinyl-
aluminium reagents under the control of Cu–NHC com-
plexes.6 Moreover, Alexakis and co-workers have disclosed
the Ir-catalyzed regioselective allylic substitution with arylzinc
reagents in high enantioselectivity.7
(6 mol%) and CuBrꢀSMe2 (5 mol%) in CH2Cl2 (10 mL) at
ꢁ78 1C, SN20 product 3 with 78% ee was obtained along with
SN2 product 4 (100% conversion, 3 : 4 = 66 : 34) (Table 1,
entry 1). This result encouraged us to perform further optimi-
zation of the allylic alkylation.
Increasing the concentration of the reaction medium to
0.4 M of 2 from 0.1 M gave higher 88% enantioselectivity
with slight decrease in regioselectivity (entries 2 and 3). How-
ever, choice of the copper source, which is usually a crucial
factor for the selectivity in Cu-catalyzed allylic alkylation, did
not influence the outcome of the process significantly (entries
5–7), except for copper cyanide giving 3 with 9% ee (entry 4),
Table 1 Optimization of Cu–1-catalyzed asymmetric allylic alkyla-
tion with EtMgBra
We have been involved in the development of catalytic
asymmetric reactions applying chiral amidophosphanes, such
as 1a–e (Fig. 1), as chiral ligands for copper- or rhodium-
catalyzed reactions. These chiral amidophosphane ligands
have enabled catalytic asymmetric conjugate arylation and
alkylation of cycloalkenones with arylboronic acids,8
Grignard reagents9 and diorganozincs.10
Conv.b
T/1C (%)
eec
(%)
Entry Cu salt
1
3 : 4b
1d
2e
3
CuBrꢀSMe2
1a ꢁ78 100
66 : 34 78
Since the p-allylcopper(III) species is a common intermediate
in conjugate addition as well as in allylic alkylation chemis-
try,11 we expected that amidophosphane ligands could provide
regio- and enantioselective transfer of an aryl moiety in
Cu-catalyzed allylic alkylation. Herein, we report that chiral
copper-amidophosphane complexes can perform, for the first
time, regio- and enantioselective allylic arylation with aryl-
CuBrꢀSMe2
CuBrꢀSMe2
CuCN
1a ꢁ78 100 (100) 63 : 37 85
1a ꢁ78 100
1a ꢁ78 100
1a ꢁ78 100
61 : 39 88
62 : 38
4
9
5
Cu(OTf)2
63 : 37 90
65 : 35 90
62 : 38 91
69 : 31 86
73 : 27 78
57 : 43 60
58 : 42 40
60 : 40 44
63 : 37 88
6
(CuOTf)2ꢀC6H5Me 1a ꢁ78 100 (91)
7
8
9
10
11
12
13
a
Cu(MeCN)4BF4
Cu(MeCN)4BF4
Cu(MeCN)4BF4
Cu(MeCN)4BF4
Cu(MeCN)4BF4
Cu(MeCN)4BF4
Cu(MeCN)4BF4
1a ꢁ78 100 (95)
1a ꢁ60 100
1a ꢁ40 100
1b ꢁ78 100
1c
ꢁ78 100
1d ꢁ78 100
1e
ꢁ78 100
Graduate School of Pharmaceutical Sciences, Kyoto University,
Yoshida, Sakyo-ku, Kyoto, 606-8501, Japan.
E-mail: tomioka@pharm.kyoto-u.ac.jp; Fax: +81 75 753 4604;
Tel: +81 75 753 4553
w Electronic supplementary information (ESI) available: Experimental
details and spectroscopic data of the products. See DOI: 10.1039/
b809140d
0.4 M concentration of 2. EtMgBr (1.3 equiv., 3.0 M Et2O solution)
b
diluted with 0.5 mL CH2Cl2 was added over 30 min. Conversion
(conv.) and regioselectivity (3 : 4) were determined by achiral GC.
c
Isolated yields are given in parentheses. Determined by chiral GC.
d
e
0.1 M concentration of 2. 0.25 M concentration of 2.
ꢂc
This journal is The Royal Society of Chemistry 2008
5140 | Chem. Commun., 2008, 5140–5142

View Article Online
and slightly better enantioselectivity (91% ee) was obtained
with Cu(MeCN)4BF4 (entry 7).
Table 2 Optimization of Cu–1-catalyzed asymmetric allylic arylation
with PhMgBra
The reaction using allylic substrates having chloride, acetate
and phosphate as a leaving group, failed to improve the
enantioselectivity. The reaction at higher temperature
(ꢁ60 1C, ꢁ40 1C) improved the regioselectivity12 at the
expense of the enantioselectivity (entries 8 and 9).
Conv.b
1/mol% (%)
eec
Although a number of amidophosphane ligands were also
screened, 1a was found to be the best. The use of 1b9d having a
dimethylcarbamoyl group resulted in lower enantio- and regio-
selectivity (60% ee, 3 : 4 = 57 : 43) (entry 10). Using amino-acid-
connected amidophosphanes 1c and 1d,13 the products were
obtained with lower enantioselectivities and slightly lower regio-
selectivities (entries 11 and 12). Amidophosphanes 1e14 having two
mesitylmethyl groups on the pyrrolidine ring gave comparable
results (entry 13). Thus, the best conditions were determined to be
those in entry 7, giving 3 with 91% ee as a 62 : 38 mixture with 4.
In contrast to many excellent reports on Cu-catalyzed AAA
with alkyl Grignard reagents, examples on those with aryl
Grignard reagents are rare and suffer from low reactivity in
addition to poor regio- and/or enantioselectivity (up to 21%
ee).15 We then applied our catalyst to this more challenging
arylation reaction. As depicted in Table 2, entry 1, we first
examined the reaction of (E)-1-bromooct-2-ene (5a) with PhMgBr
under our aforementioned conditions, 1a (6 mol%) and
Cu(MeCN)4BF4 (5 mol%), to give the corresponding SN20 pro-
duct 6a with 56% ee. SN2 product 7a was also obtained (6a : 7a =
62 : 38) as a mixture of E : Z isomers (88 : 12). The sense of
enantiofacial selectivity was opposite to that in the reactions in
Table 1.16
Entry Cu salt/mol%
6a : 7ab (%)
1
2
3
4
Cu(MeCN)4BF4/5
Cu(MeCN)4BF4/5
Cu(MeCN)4BF4/2
Cu(MeCN)4BF4/1
CuTCd/2
1a/6
100 (100) 62 : 38
100 (94) 64 : 36
100 (90) 71 : 29
100
100 (93) 73 : 27
100
100
100
100
56
66
72
60
69
12
57
59
60
73
17
16
38
71
74
76
79
81
81
1a/10
1a/4.4
1a/2.2
1a/4.4
1a/4.4
1a/4.4
70 : 30
5
6
7
8
CuCN/2
CuBrꢀSMe2/2
87 : 13
64 : 36
66 : 34
64 : 36
(CuOTf)2ꢀC6H5Me/1 1a/4.4
9
Cu(OTf)2/2
1a/4.4
1b/4.4
1c/4.4
1d/4.4
1e/4.4
1a/4.4
1a/4.4
1a/4.4
1a/4.4
1a/4.4
1b/4.4
10
11
12
13
14e
15f
16g
17h
18h
19h
a
Cu(MeCN)4BF4/2
Cu(MeCN)4BF4/2
Cu(MeCN)4BF4/2
Cu(MeCN)4BF4/2
Cu(MeCN)4BF4/2
Cu(MeCN)4BF4/2
Cu(MeCN)4BF4/2
Cu(MeCN)4BF4/2
CuTC/2
100 (89) 69 : 31
100
100
100
55 : 45
55 : 45
63 : 37
100 (96) 72 : 28
100
100
100 (91) 75 : 25
100 (99) 76 : 24
100 (87) 73 : 27
70 : 30
74 : 26
CuTC/2
PhMgBr (1.3 equiv., 3.0 M Et2O solution) diluted with 0.5 mL
b
CH2Cl2 was added over 30 min. Conversion (conv.) and regioselec-
tivity (6a : 7a) were determined by achiral GC. Isolated yields are given
c
d
in parentheses. Determined by chiral GC. CuTC = copper(I)
e
thiophencarboxylate. PhMgBr (1.3 equiv.) was added over 1 h.
f
Both 5a and PhMgBr (1 equiv.) were independently added over 1
g
h at the same time. PhMgBr (1.3 equiv.) was added over 2 h.
The enantioselectivity was improved when Cu(I) and the
ligand were used in 1 : 2 ratio (entry 2) and was further
improved when just 2 mol% Cu salt and 4.4 mol% 1a were
employed to afford 6a with 72% ee and a 71 : 29 ratio of 6a : 7a
(entry 3). Further decrease of the catalyst loading led to lower
enantioselectivity (entry 4).
h
PhMgBr (1.3 equiv.) was added over 4 h.
With the optimized conditions in hands, other substrates were
applied to the reaction (Table 3). Aliphatic allylic bromide 5b
(entry 2) afforded the product 6b with 67% ee and high regios-
electivity (83 : 17). Gratifyingly, difunctionalized allylic substrate,
trans-1,4-dibromo-2-butene (5c), was also a promising substrate in
the reactions using PhMgBr and 4-FC6H4MgBr, which exclusively
afforded the SN20 products 6c and 6d in 80 and 72% ee,
respectively (entries 3 and 4).17 In these reactions, (E)-1,4-diphe-
nylbut-2-ene and 3,4-diphenybut-1-ene (entry 3), and (E)-1,4-bis-
(4-fluorophenyl)but-2-ene and 3,4-bis(4-fluoropheny)but-1-ene
(entry 4) were also obtained in 5, 7, 9 and 10% yield, respectively.
These results clearly indicate that the apparently exclusive SN20
selectivity in entries 3 and 4 should be due to further reactions of
the once-formed SN2 products with the Grignard reagents.
Interestingly, inversion of the sense of enantiofacial selec-
tivity was observed in the reaction of substrate 5d having a
benzyloxy group to give product 6e with 34% ee and 66 : 34
regioselectivity (entry 5). To the best of our knowledge, these
regio- and enantioselectivities are unprecedentedly high in
Cu-catalyzed allylic arylation using aryl Grignard reagents.
In the reactions of aryl-type substrates 5e and 5f, 6f and 6g
were formed in 71 and 77% ee, respectively, with low
SN20-regiocontrol probably due to the bulky phenyl group
causing severe steric and/or electronic repulsion with the
incoming nucleophilic species (entries 6 and 7).18
Other copper salts were tested. Copper(I) thiophen-
carboxylate (CuTC) was found to be as efficient as
Cu(MeCN)4BF4 (entry 5), while CuCN expectedly gave higher
regioselectivity (87 : 13) albeit with 12% ee (entry 6). The use
of CuBrꢀSMe2, CuOTf and Cu(OTf)2 had deleterious effects
on regio- and enantioselectivity (entries 7–9).
Different amidophosphane ligands were screened in the
reaction of 5a with PhMgBr. In contrast to the reaction with
cinnamyl bromide (2) and EtMgBr (Table 1, entry 10), the
carbamoyl amidophosphane ligand 1b was proven to be an
alternative ligand to 1a, producing the desired chiral product
6a with 73% ee in 69 : 31 regioisomeric ratio (entry 10). The
amino-acid-connected ligands 1c and 1d showed lower regio-
selectivity and poor enantioselectivity (entries 11 and 12). In
addition, ligand 1e having more bulky substituents did not
improve the selectivity of the reaction (entry 13).
Finally, the addition rate of PhMgBr was found to be of
significance. The slower was the addition, the higher were the
regio- and enantioselectivity (entries 14, 16 and 17). Simulta-
neous slow addition of 5a and PhMgBr made only slight
improvement (entry 15). When CuTC–1a and CuTC–1b com-
plexes were used, 6a with 81% ee was obtained in 76 : 24 and
73 : 27 regioselectivity, respectively (entries 18 and 19).
ꢂc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 5140–5142 | 5141

View Article Online
Table 3 CuTC–1a-catalyzed asymmetric allylic arylationa
2 For reviews of AAA reactions with various metals, see: (a) Z. Lu
and S. Ma, Angew. Chem., Int. Ed., 2008, 47, 258; (b) H.
Miyabe and Y. Takemoto, Synlett, 2005, 1641; (c) B. M. Trost,
J. Org. Chem., 2004, 69, 5813; (d) B. M. Trost and M. L. Crawley,
Chem. Rev., 2003, 103, 2921; (e) R. Takeuchi, Synlett, 2002
1954.
3 A. S. E. Karlstrom and J.-E. Backvall, in Modern Organocopper
¨
¨
Chemistry, ed. N. Krause, Wiley-VCH, Weinheim, Germany, 2001,
p. 259.
Conv.b
(%)
eec
Entry R
Ar
Ph
Product
6 : 7b (%)
4 For recent reviews in Cu-catalyzed asymmetric allylic substitution:
(a) C. Falciola and A. Alexakis, Eur. J. Org. Chem., 2008, 3765; (b)
A. Alexakis, C. Malan, L. Lea, K. Tissot-Croset, D. Polet and C.
Falciola, Chimia, 2006, 60, 124; (c) H. Yorimitsu and K. Oshima,
Angew. Chem., Int. Ed., 2005, 44, 4435; (d) A. Kar and N. P.
Argade, Synthesis, 2005, 2995; (e) A. H. Hoveyda, A. W. Hird and
M. A. Kacprzynski, Chem. Commun., 200416), 1779.
5 (a) T. Hiyama and N. Wakasa, Tetrahedron Lett., 1985, 26, 3259;
(b) J.-C. Fiaud and L. Aribi-Zouioueche, J. Organomet. Chem.,
1985, 295, 383; (c) G. Consiglio, O. Piccolo, L. Roncetti and F.
Morandini, Tetrahedron, 1986, 42, 2043; (d) F. Fotiadu, P. Cros, B.
Faure and G. Buono, Tetrahedron Lett., 1990, 31, 77; (e) G.
Consiglio and A. Indolese, Organometallics, 1991, 10, 3425; (f)
U. Nagel and H. G. Nedden, Inorg. Chim. Acta, 1998, 269, 34; (g)
E. Gomez-Bengoa, N. M. Heron, M. T. Didiuk, C. A. Luchaco
and A. H. Hoveyda, J. Am. Chem. Soc., 1998, 120, 7649; (h) K.-G.
Chung, Y. Miyake and S. Uemura, J. Chem. Soc., Perkin Trans. 1,
2000, 2725; (i) H. Yasui, K. Mizutani, H. Yorimitsu and K.
Oshima, Tetrahedron, 2006, 62, 1410.
1d
C5H11
5a
100 (99) 76 : 24 81
100 (100) 83 : 17 67
100 (89) 100 : 0 80
100 (81) 100 : 0 72
100 (100) 66 : 34 34e
100 (100) 18 : 82 71
2
Et
5b
Ph
3
BrCH2
5c
Ph
4
BrCH2
5c
4-FC6H4
5
BnOCH2 Ph
5d
6
4-ClC6H4 Ph
5e
6 (a) M. A. Kacprzynski, T. L. May, S. A. Kazane and A. H.
Hoveyda, Angew. Chem., Int. Ed., 2007, 46, 4554; (b) Y. Lee, K.
Akiyama, D. G. Gillingham, M. K. Brown and A. H. Hoveyda,
J. Am. Chem. Soc., 2008, 130, 446.
7 A. Alexakis, S. E. Hajjaji, D. Polet and X. Rathgeb, Org. Lett.,
2007, 9, 3393.
7f
4-CF3C6H4 Ph
5f
(93)
16 : 84 77
a
8 (a) M. Kuriyama, K. Nagai, K. Yamada, Y. Miwa, T. Taga and K.
Tomioka, J. Am. Chem. Soc., 2002, 124, 8932; (b) M. Kuriyama
and K. Tomioka, Tetrahedron Lett., 2001, 42, 921; (c) Q. Chen, M.
Kuriyama, T. Soeta, X. Hao, K. Yamada and K. Tomioka, Org.
Lett., 2005, 7, 4439.
9 (a) M. Kanai and K. Tomioka, Tetrahedron Lett., 1995, 36, 4275;
(b) Y. Nakagawa, M. Kanai, Y. Nagaoka and K. Tomioka,
Tetrahedron Lett., 1996, 37, 7805; (c) Y. Nakagawa, M. Kanai,
Y. Nagaoka and K. Tomioka, Tetrahedron, 1998, 54, 10295; (d) M.
Kanai, Y. Nakagawa and K. Tomioka, Tetrahedron, 1999, 55,
3843.
10 (a) T. Soeta, K. Selim, M. Kuriyama and K. Tomioka, Adv. Synth.
Catal., 2007, 349, 629; (b) K. Selim, T. Soeta, K. Yamada and K.
Tomioka, Chem.–Asian J., 2008, 3, 342; (c) T. Soeta, K. Selim, M.
Kuriyama and K. Tomioka, Tetrahedron, 2007, 63, 6573.
11 (a) E. Nakamura and S. Mori, Angew. Chem., Int. Ed., 2000, 39,
3750; (b) M. Yamanaka, S. Kato and E. Nakamura, J. Am. Chem.
Soc., 2004, 126, 6287; (c) S. Woodward, Angew. Chem., Int. Ed.,
2005, 44, 5560.
ArMgBr (1.3 equiv.) diluted with 1 mL CH2Cl2 was added over
b
4 h. Conversion (conv.) and regioselectivity (6 : 7) were determined by
1H NMR or GC. Isolated yields are given in parentheses. Determined
c
d
e
by chiral GC. Table 2, entry 18. Determined after conversion to the
f
corresponding alcohol by debenzylation using BCl3. Both 5f and
PhMgBr (1 equiv.) were independently added over 4 h at the same time.
In conclusion, we have developed regio- and enantioselective
allylic substitution with arylmagnesium bromide as well as ethyl-
magnesium bromide, using chiral amidophosphane–copper(I)
complexes as catalysts. It is noteworthy that unprecedentedly
high regio- and enantioselectivities were achieved in the asym-
metric arylation of aliphatic and difunctionalized brominated
substrates with aryl Grignard reagents by the present protocol.
This research was partially supported by the 21st Century COE
(Center of Excellence) Program ‘‘Knowledge Information Infra-
structure for Genome Science’’, a Grant-in-Aid for Scientific
Research in Priority Areas ‘‘Advanced Molecular Transformations
of Carbon Resources’’, a Grant-in-Aid for Scientific Research
(A) and the Targeted Proteins Research Program of the Ministry
of Education, Culture, Sports, Science, and Technology, Japan. K.
B. S. thanks the Egyptian Government for a predoctoral
fellowship.
12 J.-E. Backvall, M. Selle
´
n and B. Grant, J. Am. Chem. Soc., 1990,
¨
112, 6615.
13 M. Kuriyama, T. Soeta, X. Hao, Q. Chen and K. Tomioka, J. Am.
Chem. Soc., 2004, 126, 8128.
14 T. Soeta, K. Nagai, H. Fujihara, M. Kuriyama and K. Tomioka,
J. Org. Chem., 2003, 68, 9723.
15 No allylic substitution was observed with PhMgI catalyzed by
Cu(I)–arenethiolate complex: (a) G. J. Meuzelaar, A. S. E.
Karlstrom, M. van Klaveren, E. S. M. Persson, A. del Villar, G.
¨
¨
van Koten and J.-E. Backvall, Tetrahedron, 2000, 56, 2895; 21% ee
was observed in reaction catalyzed by CuCN and TADDOL
derived ligand using 2-MeOC6H4MgBr; (b) A. Alexakis, C. Malan,
L. Lea, C. Benhaim and X. Fournioux, Synlett, 2001, 927.
16 For determination of the absolute configuration, see ESIw.
17 A similar complete regioselectivity was also reported in ref. 4a and C.
A. Falciola and A. Alexakis, Angew. Chem., Int. Ed., 2007, 46
2619.
Notes and references
1 (a) B. M. Trost and C. Lee, in Catalytic Asymmetric Synthesis, ed.
I. Ojima, Wiley, New York, 2nd edn, 2000, p. 593; (b) G.
Helmchen, in Asymmetric Synthesis—The Essentials, ed. M.
Christmann and S. Brase, Wiley-VCH, Weinheim, Germany,
¨
2007, pp. 95–99.
18 M. Kimura, T. Yamazaki, T. Kitazume and T. Kubota, Org. Lett.,
2004, 6, 4651 and references therein.
ꢂc
This journal is The Royal Society of Chemistry 2008
5142 | Chem. Commun., 2008, 5140–5142