Copper-catalyzed γ-selective and stereospecific allyl-aryl coupling between (Z)-acyclic and cyclic allylic phosphates and arylboronates

Article abstract of DOI:10.1021/ol100841y

A Cu-catalyzed allyl-aryl coupling reaction between (Z)-acyclic or cyclic allylic phosphates and arylboronates proceeds with excellent γ-E-selectivity and 1,3-anti chirality transfer, which gives the corresponding coupling products with benzylic and allylic stereogenic centers. The wide availability and easy-to-handle nature of arylboronates, the inexpensiveness of the Cu catalyst system, and the high regio- and stereoselectivities are attractive features of this protocol.

Full text of DOI:10.1021/ol100841y

ORGANIC
LETTERS
2010
Vol. 12, No. 10
2438-2440
Copper-Catalyzed γ-Selective and
Stereospecific Allyl-Aryl Coupling
between (Z)-Acyclic and Cyclic Allylic
Phosphates and Arylboronates
Hirohisa Ohmiya,* Natsumi Yokokawa, and Masaya Sawamura*
Department of Chemistry, Faculty of Science, Hokkaido UniVersity,
Sapporo 060-0810, Japan
ohmiya@sci.hokudai.ac.jp; sawamura@sci.hokudai.ac.jp
Received April 13, 2010
ABSTRACT
A Cu-catalyzed allyl-aryl coupling reaction between (Z)-acyclic or cyclic allylic phosphates and arylboronates proceeds with excellent γ-E-
selectivity and 1,3-anti chirality transfer, which gives the corresponding coupling products with benzylic and allylic stereogenic centers. The
wide availability and easy-to-handle nature of arylboronates, the inexpensiveness of the Cu catalyst system, and the high regio- and
stereoselectivities are attractive features of this protocol.
Arylboronic acid derivatives have found wide application in
modern organic synthesis because of their broad availability,
air stability, and easy-to-handle nature.1 They are especially
useful for C-C bond formations. In fact, various types of
electrophilic reagents are now capable of being efficiently
coupled with arylboronic acid derivatives under the influence
of transition-metal catalysts. However, the utilities of coupling
reactions between arylboronic acid derivatives with allylic
electrophiles are still limited because of the difficulty in
controlling regioselectivity and stereoselectivity, especially in
cases where an allylic system is not present at the terminus of
the molecule but in an internal position.2 The recently reported
Pd-catalyzed allyl-aryl coupling between acyclic (E)-allylic
esters and arylboronic acids is a rare case that occurs generally
with excellent regio- and stereoselectivities (γ-E-selectivity and
1,3-syn stereospecificity).3
arylboronates, which proceeds with excellent γ-E-selectivity
and 1,3-anti chirality transfer. The Cu system is comple-
mentary to the reported Pd system in that it is applicable
toward (Z)-acyclic and cyclic allylic systems.3 Notably, the
use of the arylmetal reagents as nucleophilic coupling
partners has not been well exploited previously in Cu
chemistry due to the poor nucleophilicity of the arylcopper
species relative to that of the alkylcopper species.4–7
(2) Pd-catalyzed allyl-aryl couplings between allyl alcohol derivatives
and arylboron compounds have been reported. These studies, however, have
focused on cases in which an allylic system is located at the terminus of a
molecule or is highly asymmetrized by electronic and/or steric substituent
effects. See: (a) Miyaura, N.; Yamada, K.; Suginome, H.; Suzuki, A. J. Am.
Chem. Soc. 1985, 107, 972–980. (b) Legros, J.-Y.; Fiaud, J.-C. Tetrahedron
Lett. 1990, 31, 7453–7456. (c) Uozumi, Y.; Danjo, H.; Hayashi, T. J. Org.
Chem. 1999, 64, 3384–3388. (d) Bouyssi, D.; Gerusz, V.; Balme, G. Eur.
J. Org. Chem. 2002, 2445–2448. (e) Mino, T.; Kajiwara, K.; Shirae, Y.;
Sakamoto, M.; Fujita, T. Synlett 2008, 2711–2715. (f) Nishikata, T.;
Lipshutz, B. H. J. Am. Chem. Soc. 2009, 131, 12103–12105, and references
therein.
Here, we report a new Cu-catalyzed allyl-aryl coupling
reaction with (Z)-acyclic or cyclic allylic phosphates and
(3) (a) Ohmiya, H.; Makida, Y.; Tanaka, T.; Sawamura, M. J. Am. Chem.
Soc. 2008, 130, 17276–17277. (b) Ohmiya, H.; Makida, Y.; Li, D.; Tanabe,
M.; Sawamura, M. J. Am. Chem. Soc. 2010, 132, 879–889.
(1) Boronic Acids; Hall, D. G., Ed.; Wiley-VCH: Weinheim, 2005.
10.1021/ol100841y  2010 American Chemical Society
Published on Web 04/29/2010

Scheme 1
Table 1. Cu-Catalyzed Allyl-Aryl Coupling of (Z)-Acyclic
Allylic Phosphatesa
In numerous studies in this laboratory, we have found that
the reaction of (Z)-allylic phosphate 1a with 5,5-dimethyl-
2-phenyl-1,3,2-dioxaborinane (2a) (2 equiv) in the presence
of CuCl (5 mol %), acetylacetone (acac-H, 10 mol %),
KOtBu (3 equiv), and H2O (3 equiv) in CH3CN at 80 °C for
12 h afforded allyl-aryl coupling product 3aa in 90%
isolated yield with excellent regio- (γ/R >20:1) and E/Z (>20:
1) selectivities (Scheme 1). The reaction of (E)-allylic
phosphate resulted in moderate E/Z selectivity (87:13) with
the excellent γ-regioselectivity retained. The Cu loading
could be reduced to 1 mol % with only a slight decrease in
the yield (84%; acac-H, 10 mol %).
Several observations concerning the optimum reaction
conditions are to be noted. The use of a small amount of
H2O (3 equiv) is critical: the reaction without H2O resulted
in a complex mixture. The amount of KOtBu is also critical:
reducing it from 3 to 2 equiv decreased the yield from 90 to
34% under otherwise identical conditions, and no reaction
occurred in the absence of KOtBu. The reaction proceeded
even without acac-H but with a significantly reduced yield
(68%, 5 mol % Cu). Reducing the amount of boronate 2a to
1 equiv caused considerable hydrolysis of 1a. Phenylboronic
acid instead of 2a was also useful, although the yield of 3aa
was slightly decreased (62%).
a Conditions: 1 (0.3 mmol), 2 (0.6 mmol), CuCl (5 mol %), acac-H (10
mol %), H2O (0.9 mmol), KOtBu (0.9 mmol), CH3CN (entries 1-3 and
6-9, 0.3 mL; entry 4, 0.6 mL; entry 5, 1 mL), 80 °C, 12 h. b Isolated
c
1
yield. Isomeric ratios (γ/R >20:1, E/Z > 20:1). Determined by H NMR
or GC analysis.
was coupled with 1a in a reasonable yield (entry 1). The
allylic phosphate 1c with a Bu instead of a Me group was
phenylated at the γ-position effectively (entry 8). A sterically
more demanding γ-substituent such as an isobutyl group was
also tolerated (entry 9). However, the reaction was nearly
inhibited by the steric demand of an isobutyl substituent at
the R-position (in place of R-Me for 1c) (5% yield, not
shown).
The coupling reaction took place with excellent R-to-γ
chirality transfer with 1,3-anti stereochemistry (Scheme 2).
The reaction of (S)-(Z)-1c (96% ee), which has R-Me and
γ-Bu substituents, with the phenylboronate (2a) in the
presence of CuCl, acetylacetone, KOtBu, and H2O gave (R)-
(E)-3c (94% ee) with only a slight decrease of enantiomeric
purity. The reaction of (S)-(Z)-1c′ (96% ee), which has R-Bu
and γ-Me substituents, with 2a gave (R)-(E)-3c′ (93% ee),
an isomer of 3c with regard to the R/γ-regioselectivity.
The Cu-catalyzed allyl-aryl coupling showed a range of
substrate scope of allylic phosphates (1) and arylboronates
(2) (Table 1). The reactions proceeded with excellent γ- (>20:
1) and E- (>20:1) selectivities. Functional groups such as
MeO, CF3, Cl, ester, and silyl ether in 1 or 2 were compatible
with the Cu system (entries 2-5 and 7). 3-Thiopheneboronic
acid ester 2g also participated in the coupling (entry 6).
The efficiency of the reaction toward steric demand in both
(Z)-acyclic allylic phosphates (1) and arylboronates (2) is
shown in Table 1 (entries 1, 8, and 9). o-Tolylboronate (2b)
(4) For γ-selective allylic substitution reactions with stoichiometric
arylcopper(I) reagents with excellent 1,3-chirality transfer, see: (a) Har-
rington-Frost, N.; Leuser, H.; Calaza, M. I.; Kneisel, F. F.; Knochel, P.
Org. Lett. 2003, 5, 2111–2114. (b) Kiyotsuka, Y.; Acharya, H. P.; Katayama,
Y.; Hyodo, T.; Kobayashi, Y. Org. Lett. 2008, 10, 1719–1722.
(5) For Cu-catalyzed γ-selective and enantioselective reaction of cin-
namyl bromides with aryl Grignard reagents, see: Selim, K. B.; Matsumoto,
Y.; Yamada, K.; Tomioka, K. Angew. Chem., Int. Ed. 2009, 48, 8733–
8735.
Scheme 2. Chirality Transfer
(6) For Cu-catalyzed C-C bond formations with arylboron reagents,
see: (a) Takaya, J.; Tadami, S.; Ukai, K.; Iwasawa, N. Org. Lett. 2008, 10,
2697–2700. (b) Ohishi, T.; Nishiura, M.; Hou, Z. Angew. Chem., Int. Ed.
2008, 47, 5792–5795. (c) Yamamoto, Y.; Kirai, N.; Harada, Y. Chem.
Commun. 2008, 2010–2012. (d) Tomita, D.; Kanai, M.; Shibasaki, M. Chem.
Asian, J. 2006, 1, 161–166. (e) Tomita, D.; Yamatsugu, K.; Kanai, M.;
Shibasaki, M. J. Am. Chem. Soc. 2009, 131, 6946–6948.
(7) For Cu-catalyzed (30 mol %), γ-selective allylic substitution reactions
of 4-cyclopentene-1,3-diol monoesters with aryl- and alkenylzinc reagents,
see: Nakata, K.; Kiyotsuka, Y.; Kitazume, T.; Kobayashi, Y. Org. Lett.
2008, 10, 1345–1348.
The Cu catalyst system was also applicable to cyclic allylic
substrates (Table 2).7 The coupling reactions between cis-
Org. Lett., Vol. 12, No. 10, 2010
2439

alkylboranes.8 This suggests interesting mechanistic differ-
ence between the two Cu-catalyzed coupling reactions. As
one of the possible mechanisms for the latter, we proposed
1,2-addition-ꢀ-elimination of organocopper species (R-Cu
rather than a cuprate). On the other hand, considering that
the present allyl-aryl coupling reaction must be conducted
in the presence of KOtBu/H2O in excess relative to the
organoboronate, an active copper intermediate of the allyl-aryl
coupling reaction is likely to be a monoorganoheterocuprate
Table 2. Cu-Catalyzed Allyl-Aryl Coupling of Cyclic Allylic
Phosphatesa
t
[ArCu(OR)-Ln]- species (R ) Bu or H).
Although the roles of acac-H and H2O are not clear at
present, according to DFT studies by Nakamura et al. on
the mechanism of the reaction of monoorganoheterocuprate
with allyl acetate,9 the mechanistic model for the Cu-
catalyzed reaction of (S)-(Z)-1c,c′ (Scheme 2), and cyclic
allylic phosphates 1e,f (Table 2, entries 1 and 2) with
phenylboronate 2a can be postulated as shown in Scheme
3, path a. The catalytic cycle is initiated by transmetalation
between CuOR and 2a in the presence of KOR to produce
the nucleophilic heterocuprate A (Ph-Cu-OR-). Subse-
quently, A forms π-complex B with an allylic phosphate (1).
Then, oxidative addition through the transition state C(TS)
with anti-stereochemistry with respect to the phosphate group
leads to (γ,σ-enyl)copper(III) species D (enyl[σ + π]
complex). Finally, reductive elimination results in C-C bond
formation at the γ-position and regenerates CuOR. The
regioselectivity is determined at the oxidative addition step
as a consequence of the asymmetric nature of B. The
incomplete γ-selectivity in the reactions of 1f,h suggests the
minor involvement of path b, in which the diastereomeric
π-complex B′ leads to enyl[σ + π] complex D′, which forms
the R-substitution product (3′).
a Conditions: 1 (0.3 mmol), 2a (0.6 mmol), CuCl (5 mol %), acac-H
(10 mol %), H2O (0.9 mmol), KOtBu (0.9 mmol), CH3CN (0.3 mL), 80
c
°C, 12 h. b Isolated yield. Determined by H NMR.
1
4-cyclopentene-1,3-diol derivative 1e and boronate 2a pro-
ceeded with excellent γ-selectivity and 1,3-anti selectivity,
giving trans-1,2-isomer 3e (entry 1). The reaction with 1f
also proceeded with excellent 1,3-anti selectivity, but the
product was contaminated with a significant amount of
R-substitution product (3e′, not shown) (γ/R 76:24) (entry
2). The reaction with 2-cyclohexenyl phosphate (1g) pro-
ceeded smoothly (entry 3). Even when six-membered cyclic
allylic phosphate (1h) with considerable steric congestion
at the γ-position was used, the coupling occurred preferen-
tially at the more hindered γ-position albeit with moderate
regioselectivity (γ/R 81:19) (entry 4). Seven-membered
cyclic substrate 1i also underwent efficient coupling (entry
5).
In conclusion, we developed a Cu-catalyzed γ-selective
and stereospecific allyl-aryl coupling reaction between (Z)-
acyclic or cyclic allylic phosphates and arylboronates, which
gives allyl-aryl coupling products with benzylic and allylic
stereogenic centers. The wide availability and easy-to-handle
nature of arylboronates, the inexpensiveness of the Cu
catalyst system, and the high regio- and stereoselectivities
are attractive features of this protocol.
Scheme 3. Possible Mechanism
Acknowledgment. This work was supported by Grants-
in-Aid for Scientific Research (B) (No. 21350020) and for
Young Scientists (B) (No. 21750086), JSPS.
Supporting Information Available: Experimental details
and characterization data for new compounds. This material
is available free of charge via the Internet at http://pubs.acs.org.
OL100841Y
The incompleteness of the γ-selectivity in the reaction of
1f and 1h (Table 2, entries 2 and 4) is in sharp contrast to
the solid γ-selectivity of the recently reported Cu-cata-
lyzed allyl-alkyl coupling between allylic phosphates and
(8) Ohmiya, H.; Yokobori, U.; Makida, Y.; Sawamura, M. J. Am. Chem.
Soc. 2010, 132, 2895–2897.
(9) (a) Yoshikai, N.; Zhang, S.-L.; Nakamura, E. J. Am. Chem. Soc.
2008, 130, 12862–12863. (b) Yamanaka, M.; Kato, S.; Nakamura, E. J. Am.
Chem. Soc. 2004, 126, 6287–6293.
2440
Org. Lett., Vol. 12, No. 10, 2010