Copper-catalyzed enantioselective allyl-allyl cross-coupling

Article abstract of DOI:10.1021/ja312487r

A Cu(I)-phosphoramidite-based catalytic system that allows asymmetric allyl-allyl cross-coupling with high enantioselectivity is reported. This transformation tolerates a large variety of functionalized substrates. The versatility of this new reaction is illustrated in the catalytic asymmetric synthesis of the Martinelline alkaloids chromene derivative core.

Full text of DOI:10.1021/ja312487r

Communication
pubs.acs.org/JACS
Copper-Catalyzed Enantioselective Allyl−Allyl Cross-Coupling
Valentín Hornillos, Manuel Perez, Martín Fananas-Mastral, and Ben L. Feringa*
́
́
̃
Stratingh Institute for Chemistry, University of Groningen, Nijenborgh 4, 9747 AG Groningen, The Netherlands
S
* Supporting Information
stereogenic centers.6 This transformation offers the possibility
ABSTRACT: A Cu(I)-phosphoramidite-based catalytic
system that allows asymmetric allyl−allyl cross-coupling
with high enantioselectivity is reported. This trans-
formation tolerates a large variety of functionalized
substrates. The versatility of this new reaction is illustrated
in the catalytic asymmetric synthesis of the Martinelline
alkaloids chromene derivative core.
of using nonstabilized organometallic nucleophiles, allowing the
introduction of alkyl,7 aryl,8 vinyl,9 allenyl,10 or alkynyl11
fragments. However, the transfer of a simple allyl group (e.g.,
allylCuX), with control of enantioselectivity, remains a major
challenge. Compared with simple alkyl groups, allylcopper
species can exist in a delocalized contact η3 π-allyl ion-pair
structure.4a,12 This structural feature may have hampered the
discovery of an effective catalytic system to enable this
transformation with high stereoselectivity.
Herein, we report the first Cu-catalyzed enantioselective
allyl−allyl cross-coupling of an allyl Grignard reagent and allyl
bromides to afford chiral 1,5-dienes in good yields and high
enantioselectivity (Scheme 1b).
atalytic asymmetric C−C bond formation methodologies
C
with high efficiency and selectivity are among the most
important tools in organic synthesis.1 Of particular importance
is the metal-catalyzed allyl−allyl cross-coupling between
allylmetal species and allylic electrophiles, which has the
capacity to establish a stereogenic center bearing both an allyl
and a vinyl group. These chiral 1,5-diene structures are
abundant in naturally occurring terpenes and also serve as
highly versatile intermediates and building blocks in organic
synthesis.2,3 Several transition-metal catalysts, including Pd, Au,
Cu, and Ni, have been employed to enable catalytic allyl−allyl
cross-coupling.4 Nonetheless, high levels of enantioselectivity
have only been achieved recently by the use of Pd catalysis.5
Morken and co-workers5a described the Pd-catalyzed regio- and
enantioselective cross-coupling of allylic carbonates and
allylboronate (Scheme 1a) involving an inner-sphere 3,3′-
We started our study with a ferrocenyl-type chiral
diphosphine L1 and CuBr·SMe2 (Table 1), which has been
shown previously to be effective for the AAA of allyl bromides
with simple alkyl Grignard reagents.7b Allyl bromide 1a was
used for the reaction optimization. Unfortunately, reaction
between 1a and allylmagnesium bromide in CH2Cl2 at −80 °C,
with this catalytic system, led to less than 10% conversion and
predominantly linear product (b:l = 15:85) (Table 1, entry 1).
Other biphosphine ligands showed similar results, and so
monodentate phosphoramidite ligands13 were tested. We
examined the Cu-phosphoramidite ligand L2 using chloride
as a leaving group, which has been also shown previously to be
effective for Cu-AAA with alkyl Grignard reagents.7a Although
in this case the reaction led to full conversion, again very low
regioselectivity with negligible enantioselectivity (59:41, e.r.)
for the branched product was obtained (Table 1, entry 2). An
improved regioselectivity and promising enantioselectivity
(87:13 e.r.) were achieved by using the same catalytic system
but bromide as a leaving group (entry 3). The nature of the
allyl organometallic reagent proved also to be critical to obtain
good conversion and selectivity. Thus, the use of allylmagne-
sium chloride gave only 20% conversion with poor regio-
(41:59) and enantioselectivity (58:42 e.r., entry 4). We also
used allyllithium, encouraged by our recent discovery of an
enantioselective Cu-catalyzed method for the allylic alkylation
with organolithium reagents.7c Nonetheless, although the
conversion was complete, the regioselectivity was very low,
albeit high enantioselectivity was found (b:l = 22:78, 91:9 e.r.,
entry 5).
Scheme 1. Catalytic Asymmetric Allyl−Allyl Cross-Coupling
Methodologies
reductive elimination mechanism. This catalytic allylation
represents a highly valuable synthetic tool but leaves ample
opportunities to develop non-Pd-based alternatives. To the best
of our knowledge, Cu-catalyzed enantioselective allyl−allyl
cross-coupling is unprecedented.
Encouraged by the results obtained with L2, we studied
variations in ligand structure (see L2−L11, entries 6−14) using
CuBr·SMe2. It was found that ligand (R,S,S)-L413a led to a
major increase in regioselectivity toward the branched product
In recent years, Cu-catalyzed asymmetric allylic alkylation
(Cu-AAA) with organometallic reagents has become a very
efficient transformation for the enantioselective construction of
Received: December 21, 2012
Published: January 27, 2013
© 2013 American Chemical Society
2140
dx.doi.org/10.1021/ja312487r | J. Am. Chem. Soc. 2013, 135, 2140−2143

Journal of the American Chemical Society
Communication
and enantioselectivity (entry 15), but longer addition times did
not lead to further improvement. Variations in solvents and
temperatures did not improve the selectivity of the reaction.
(See Table S2 for details.)
Table 1. Screening of Different Conditions and
Representative Ligands
We next investigated the role of the Cu salt. It has been
shown that the use of Cu(I) cyanide as catalyst provides high
SN2′ selectivity in the reaction between allylic substrates and
Grignard reagents.4b,14 Indeed, the highest regioselectivity for
the reaction between allyl bromide 1a and allylmagnesium
bromide was obtained for CuCN; however, it gave a negligible
enantiomeric excess (entry 17). All the other Cu(I) salts tested
provided similar high enantioselectivity in C−C bond
formation as observed for CuBr·SMe2 in the allylation of allyl
bromide 1 (Tables 1 and S2). In general, the use of Cu salts
with noncoordinating counteranions favors the branched allyl−
15
allyl cross-coupling product, and (CuOTf)2·C6H6 was found
to be the most suitable Cu source for this reaction (b:l = 77:23,
97:3 e.r., entry 18). A more electron-deficient Cu species
presumably accelerates the reductive elimination step, prevent-
ing the formation of a linear allyl−allyl cross-coupling
product.14a
ab
,
c
d
entry
L
M
[Cu]
2a:3a
e.r., 2a
Having established the optimal copper−ligand combination
(Table 1, entry 18), we next explored the scope and limitation
of this new reaction. A variety of synthetically useful allylic
substrates bearing different functionalities, including protected
alcohols, amines, alkenes, and acetals, were tolerated, showing
excellent enantioselectivities in nearly all cases (Table 2).
Allylation of compound 1c, bearing a benzyloxy group, was
accomplished with good regioselectivity and excellent enantio-
selectivity, providing 2c, an advanced intermediate in the
synthesis of d-sabinene,16 a monoterpene widely distributed in
essential oils from plants. Allylic bromides bearing a protected
amine were also suitable substrates for this reaction, delivering
chiral building blocks for natural product synthesis (2d−2k).
The presence of a methoxy group or bromide substituent
(which can be sensitive to oxidative addition in a Pd-based
catalytic system) in the aromatic part of the substrate was also
well tolerated, providing synthetically useful functionalities for
further transformations (2a, 2e−2g). An interesting behavior
can be observed with amines bearing ortho-substituents at the
aryl ring. Both o-bromo and o-methoxy substituents led to an
increase in regioselectivity when compared to the unsubstiti-
tuted analogues (2d vs 2f and 2g). Amines 2i and 2j can be
readily transformed to the corresponding unsaturated piper-
idines, azepanes, or azocanes by using olefin ring-closing
metathesis.17 A decrease in the regio- and enantioselectivity was
observed for enyne 2k, where a possible intramolecular
coordination between the Cu catalyst and the triple bond
might interfere in the enantiodiscriminating step.
p-Bromocinnamyl bromide 1l underwent facile reaction to
give the coupling product 2l with good yield albeit moderate
regio- and enantioselectivity. A major improvement in the
regioselectivity was again observed when a coordinating
functionality in the ortho position, such as an o-methoxy
substituent, was introduced at the phenyl ring, providing diene
2m with high enantioselectivity. This result and those for 2a, 2f,
and 2g suggest a possible coordination between the methoxy or
bromide substituent and the Cu complex, which could also be a
key factor to increase the enantioselectivity in cinnamyl-type
substrates. The use of a dioxolane-containing allylic bromide18
1n led to the diastereoselective formation of valuable 1,2-
hydroxyallyl moiety 2n with good stereocontrol of the anti
isomer. It should be mentioned that simple linear substrate 1p
ef
,
1
2
3
4
5
6
7
8
9
L1
L2
L2
L2
L2
L3
L4
L5
L6
L7
L8
L9
L10
L11
L4
L4
L4
L4
MgBr
MgBr
MgBr
MgCl
CuBr·SMe2
CuBr·SMe2
CuBr·SMe2
CuBr·SMe2
CuBr·SMe2
CuBr·SMe2
CuBr·SMe2
CuBr·SMe2
CuBr·SMe2
CuBr·SMe2
CuBr·SMe2
CuBr·SMe2
CuBr·SMe2
CuBr·SMe2
CuBr·SMe2
CuTC
15:85
37:63
47:53
41:59
22:78
51:49
65:35
45:55
59:41
65:35
56:44
50:50
60:40
17:83
68:32
70:30
82:18
77:23
n.d.
g
59:41
87:13
58:42
91:9
h
i
Li
MgBr
MgBr
MgBr
MgBr
MgBr
MgBr
MgBr
MgBr
MgBr
MgBr
MgBr
MgBr
MgBr
85:15
94:6
61:39
79:21
92:8
10
11
12
13
14
15
16
17
18
86:14
94:6
80:20
72:28
97:3
j
j
j
j
97:3
CuCN
56:44
97:3
(CuOTf)2·C6H6
a
Conditions: 0.2 mmol of allyl bromide, 1.5 equiv of allylmagnesium
bromide diluted in CH2Cl2, 0.05 M. AllylMgBr was added over 2 h
unless otherwise noted. Conversion by GC-MS. All reactions gave full
conversion unless noted. 2a/3a ratios determined by GC-MS or H
NMR spectroscopy. Determined by chiral HPLC analysis. Normal
addition. <10% conversion. Leaving group = Cl. 20% conversion.
AllylLi was added as a solution in tert-butyl methyl ether. AllylMgBr
was added over 5 h.
b
c
1
d
e
f
g
h
i
j
1b (b:l = 65:35) with high enantioselectivity (94:6 e.r., entry 7).
The use of other phosphoramidite or BINAP-derived ligands
(Table 1, entries 6, 8−14, and Supporting Information, Table
S1) did not improve these results. In all cases, full conversion
was obtained overnight at −80 °C. It is remarkable that
introduction of methyl groups at the 3,3′ positions in the
binaphthol moiety (entry 10) or use of a spiro-biphenol
structure in the ligands (entry 12) led to similar e.r. values.
However, variations of the amine moiety caused a drastic
decrease of e.r. (entry 14 and Table S1). It should also be noted
that this reaction does not proceed in the absence of the Cu
catalyst, leading to complete recovery of the substrate.
The effects of different conditions, solvents, and temper-
atures were also studied. Slow addition (5 h) of the allyl
Grignard reagent to the reaction mixture increased both regio-
2141
dx.doi.org/10.1021/ja312487r | J. Am. Chem. Soc. 2013, 135, 2140−2143

Journal of the American Chemical Society
Communication
membered-ring Heck product 4 with no isomerization to the
endocyclic double bond. Ozonolysis of both double bonds
followed by reductive amination gave compound 6 in good
overall yield without lowering the e.r.
Table 2. Scope of Cu-Catalyzed Enantioselective Allyl−Allyl
a
Cross-Coupling
Although detailed mechanistic studies are currently being
pursued, a catalytic cycle can be proposed for this novel Cu-
catalyzed allyl−allyl cross-coupling. In accordance with the
proposed mechanism for the Cu-catalyzed AAA,14 the
allylcopper species generated by transmetalation between the
Cu(I) salt and allylmagnesium bromide would undergo
oxidative addition with the allylic substrate. Taking into
account the behavior of allylcopper complexes,4a,12 the resulting
CuIII σ−σ complex could probably equilibrate to a 16-electron
penta-coordinated-CuIII σ−π intermediate,22 where the allyl
group is coordinated to Cu via an η3 coordinative mode
(Scheme 3). This CuIII σ−π intermediate would lead directly to
Scheme 3. Proposed Mechanism for the Cu-Catalyzed Allyl−
Allyl Cross-Coupling
a
Conditions: 0.2 mmol of allyl bromide (1 equiv), 1.5 equiv of
allylmagnesium bromide diluted in CH2Cl2, 0.05 M in CH2Cl2, 5 h
addition time. All reactions gave full conversion (GC-MS). Branched/
linear ratios determined by 1H NMR spectroscopy or GC-MS.
Enantiomer ratios determined by chiral HPLC or chiral GC analysis.
the branched product via reductive elimination. Nonetheless, in
this case, the extra stabilization caused by the η3 bonding mode
of the allyl group would retard the reductive elimination and
could slightly favor the isomerization from the σ−π complex to
the π−σ complex (it also maintains the penta-coordination),
which gives rise to the linear product. This behavior would
explain the variation in SN2′ selectivity. Similarly, the aryl ring
(especially with electron-withdrawing groups present) of a
cinnamyl-type substrate further might reduce the rate of
reductive elimination, favoring the formation of the π−σ
complex, since the transition state for the reductive elimination
involves a decrease of electron density at the benzylic carbon.14a
The results obtained with 2l and 2m (Table 2) support this
explanation.23
In summary, the first Cu-catalyzed asymmetric allyl−allyl
cross-coupling has been described. The use of commercially
available allylmagnesium bromide and easily accessible allylic
bromides with cheap Cu-phosphoramidite catalyst is key for the
success of this reaction. The only waste produced in this novel
transformation is MgBr2. The use of Cu complexes with
noncoordinating counteranions as present in (CuOTf)2·C6H6
was found beneficial to increase the regioselectivity of the
reaction. The potential applications in total synthesis were
demonstrated in a three-step synthesis of the biologically active
Martinelline alkaloid chromene derivative core structure.
provides branched product 2p with high regio- and
enantioselectivity.
To illustrate the synthetic utility of the method, allyl−allyl
cross-coupling product 2a was easily converted into the
Martinelline alkaloids chromene derivative core (Scheme
2).19,20 This class of natural occurring chromene derivatives
finds utility in the treatment of impulsive disorders, Parkinson’s
disease, psychoses, memory disorders, and anxiety.20
Synthesis of 2a was executed on a larger scale (5 mmol, 1.5
g) and still furnished the product with high enantioselectivity
(96:4 e.r.) without erosion of yield. The use of catalytic
amounts of Pd(OAc)2 under ligand-free conditions21 resulted
in selective transformation of this bromide into the six-
Scheme 2. Transformation of Product 2a into the
Martinelline Alkaloids Chromene Derivative Core
2142
dx.doi.org/10.1021/ja312487r | J. Am. Chem. Soc. 2013, 135, 2140−2143

Journal of the American Chemical Society
Communication
(7) For selected examples, see: (a) Tissot-Croset, K.; Polet, D.;
ASSOCIATED CONTENT
■
Alexakis, A. Angew. Chem., Int. Ed. 2004, 43, 2426. (b) Lop
Zijl, A. W.; Minnaard, A. J.; Feringa, B. L. Chem. Commun. 2006, 409−
411. (c) Perez, M.; Fananas-Mastral, M.; Bos, P. H.; Rudolph, A.;
́
ez, F.; van
S
* Supporting Information
Optimization tables, experimental procedures, and character-
ization data. This material is available free of charge via the
Internet at http://pubs.acs.org.
́
́
̃
Harutyunyan, S. R.; Feringa, B. L. Nature Chem. 2011, 3, 377.
(d) Langlois, J.-B.; Alexakis, A. Angew. Chem., Int. Ed. 2011, 50, 1877.
(e) Shido, Y.; Yoshida, M.; Tanabe, M.; Ohmiya, H.; Sawamura, M. J.
Am. Chem. Soc. 2012, 134, 18573.
AUTHOR INFORMATION
■
(8) See for example: (a) Selim, K. B.; Matsumoto, Y.; Yamada, K.;
Tomioka, K. Angew. Chem., Int. Ed. 2009, 48, 8733. (b) Shintani, R.;
Takatsu, K.; Takeda, M.; Hayashi, T. Angew. Chem., Int. Ed. 2011, 50,
8656. (c) Gao, F.; Lee, Y.; Mandai, K.; Hoveyda, A. H. Angew. Chem.,
Int. Ed. 2010, 49, 8370.
(9) (a) Gao, F.; Carr, J. L.; Hoveyda, A. H. Angew. Chem., Int. Ed.
2012, 51, 6613. (b) Gao, F.; McGrath, K. P.; Lee, Y.; Hoveyda, A. H. J.
Am. Chem. Soc. 2010, 132, 14315.
(10) Jung, B.; Hoveyda, A. H. J. Am. Chem. Soc. 2012, 134, 1490.
(11) Dabrowski, J. A.; Gao, F.; Hoveyda, A. H. J. Am. Chem. Soc.
2011, 133, 4778.
(12) (a) Yamanaka, M.; Kato, S.; Nakamura, E. J. Am. Chem. Soc.
2004, 126, 6287−6293. (b) Bartholomew, E. R.; Bertz, S. H.; Cope, S.;
Murphy, M. C.; Ogle, A. J. Am. Chem. Soc. 2008, 130, 11244.
(13) (a) De Vries, A. H. M.; Meetsma, A.; Feringa, B. L. Angew.
Chem., Int. Ed. 1997, 35, 2374−2376. For a recent review on the use
of phosphoramidites as ligands see: (b) Teichert, J. F.; Feringa, B. L.
Angew. Chem., Int. Ed. 2010, 49, 2486.
Corresponding Author
b.l.feringa@rug.nl
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The Netherlands Organization for Scientific Research (NWO-
CW) and the National Research School Catalysis (NRSC-C)
are acknowledged for financial support. V.H. thanks the Spanish
Ministry of Science and Innovation (MICINN) for a
postdoctoral fellowship. M.P. thanks the Xunta de Galicia for
an Angeles Alvarino contract.
̃
REFERENCES
■
(1) (a) Jacobsen, E. N., Pfaltz, A., Yamamoto, H., Eds. Comprehensive
Asymmetric Catalysis: Suppl. 2; Springer-Verlag: Berlin, 2004. (b) Lu,
Z.; Ma, S. Angew. Chem., Int. Ed. 2008, 47, 258. (c) Helmchem, G.;
(14) (a) Yoshikai, N.; Nakamura, E. Chem. Rev. 2011, 112, 2339.
(b) Tseng, C. C.; Paisley, S. D.; Goering, H. L. J. Org. Chem. 1986, 51,
́
2884. (c) Backvall, J.-E.; Sellen, M.; Grant, B. J. Am. Chem. Soc. 1990,
̈
Kazmaier, U.; Forster, S. In Catalytic Asymmetric Synthesis; Ojima, I.,
̈
112, 6615.
Ed.; John Wiley & Sons: Hoboken, NJ, 2010. (d) Negishi, E. Angew.
(15) Notably, the use of the Cu(II) salt Cu(OTf)2 resulted in a lower
regioselectivity although the e.r. remained high (b:l 56:44, 97:3 e.r.).
(16) Urabe, H.; Suzuki, K.; Sato, F. J. Am. Chem. Soc. 1997, 119,
10014. TBS protected alcohol was obtained in this case by substrate-
controlled chirality transfer:
Chem., Int. Ed. 2011, 50, 6738.
(2) (a) Breitmaier, E. Terpenes, Flavors, Fragrances, Pharmaca,
Pheromones; Wiley-VCH: Weinheim, 2006. (b) Medicinal Natural
Products: A Biosynthetic Approach; Dewick, P. M., Ed.; Wiley:
Chichester, 2002.
(3) For selected examples of organic synthesis by using 1,5-dienes,
see: (a) Brown, R. C. D.; Keily, J. F. Angew. Chem., Int. Ed. 2001, 40,
4496. (b) Nakamura, H.; Yamamoto, Y. Handbook of Organopalladium
Chemistry for Organic Synthesis; Wiley-Interscience: West Lafayette,
2002; Vol. 2. (c) Zhao, Y. J.; Chang, S. S.; Loh, T. P. J. Am. Chem. Soc.
2007, 129, 492. (d) Feducia, A. J.; Gagne, M. R. J. Am. Chem. Soc.
2008, 130, 592. (e) Felix, R. J.; Weber, D.; Gutierrez, O.; Tantillo, D.
J.; Gagne, M. R. Nature Chem. 2012, 4, 405.
(17) (a) Deiters, A.; Martin, S. F. Chem. Rev. 2004, 104, 2199.
(b) Teichert, J. F.; Zhang, S.; van Zijl, A. W.; Slaa, J. W.; Minnaard, A.
J.; Feringa, B. L. Org. Lett. 2010, 12, 4658.
́
(18) Fananas-Mastral, M.; ter Horst, B.; Minnaard, A. J.; Feringa, B.
̃
(4) Cu-catalyzed allylation: (a) Karlstrom, A. S. E.; Backvall, J.-E.
̈
̈
L. Chem. Commun. 2011, 47, 5843.
Chem.Eur. J. 2001, 7, 1981 and references cited therein.
(19) Ikeda, S.; Shibuya, M.; Iwabuchi, Y. Chem. Commun. 2007, 504.
(20) Arumugam, N.; Raghunathan, R.; Almansour, A. I.; Karama, U.
Bioorg. Med. Chem. Lett. 2012, 22, 1375 and references cited therein.
(21) Hornillos, V.; van Zijl, A. W.; Feringa, B. L. Chem. Commun.
2012, 48, 3712.
(22) The fact that the reaction between 1a and crotylmagnesium
bromide resulted in a nonselective mode of substitution at the two
different allylic termini supports the formation of this intermediate
(see Scheme S1).
(b) Yanagisawa, A.; Nomura, N.; Yamamoto, H. Tetrahedron 1994,
́
50, 6017. Au-catalyzed allylation: (c) Porcel, S.; Lopez-Carrillo, V.;
García-Yebra, C.; Echavarren, A. M. Angew. Chem., Int. Ed. 2008, 47,
1883. Ni-catalyzed allylation: (d) Matsubara, R.; Jamison, T. F. J. Am.
Chem. Soc. 2010, 132, 6880. (e) Sumida, Y.; Hayashi, S.; Hirano, K.;
́
Yorimitsu, H.; Oshima, K. Org. Lett. 2008, 10, 1629. (f) Jimenez-
Aquino, A.; Flegeau, E. F.; Schneider, U.; Kobayashi, S. Chem.
Commun. 2011, 47, 9456. Pd-catalyzed allylation: (g) Nakamura, H.;
Bao, M.; Yamamoto, Y. Angew. Chem., Int. Ed. 2001, 40, 3208.
(h) Flegeau, E. F.; Schneider, U.; Kobayashi, S. Chem.Eur. J. 2009,
15, 12247.
(23) For a similar behavior in the Cu-catalyzed allylic alkylation with
allenyl boron reagents, see ref 10.
(5) Pd-catalyzed enantioselective allylation: (a) Zhang, P.; Brozek, L.
A.; Morken, J. P. J. Am. Chem. Soc. 2010, 132, 10686. (b) Zhang, P. A.;
Le, H.; Kyne, R. E.; Morken, J. P. J. Am. Chem. Soc. 2011, 133, 9716.
(c) Brozek, L. A.; Ardolino, M. J.; Morken, J. P. J. Am. Chem. Soc.
2011, 133, 16778.
(6) For general reviews: (a) Geurts, K.; Fletcher, S. P.; van Zijl, A.
W.; Minnaard, A. J.; Feringa, B. L. Pure Appl. Chem. 2008, 80, 1025.
(b) Harutyunyan, S. R.; den Hartog, T.; Geurts, K.; Minnaard, A. J.;
Feringa, B. L. Chem. Rev. 2008, 108, 2824. (c) Alexakis, A.; Backvall, J.
̈
́ ́
E.; Krause, N.; Pamies, O.; Dieguez, M. Chem. Rev. 2008, 108, 2796.
(d) Langlois, J.-B.; Alexakis, A. In Transition Metal Catalyzed Allylic
Substitution in Organic Synthesis; Kazmaier, U., Ed.; Springer-Verlag:
Berlin, 2012; pp 235−268.
2143
dx.doi.org/10.1021/ja312487r | J. Am. Chem. Soc. 2013, 135, 2140−2143