Copper-catalyzed borylative allyl-allyl coupling reaction

Article abstract of DOI:10.1002/anie.201404173

Borylative allyl-allyl coupling using allenes, bis(pinacolato)diboron, and allyl phosphates has been developed in the presence of a copper catalyst bearing an N-heterocyclic carbene ligand. The reaction affords boryl-substituted 1,5-diene derivatives in g

Full text of DOI:10.1002/anie.201404173

Angewandte
Chemie
DOI: 10.1002/anie.201404173
À
C C Coupling
Copper-Catalyzed Borylative Allyl–Allyl Coupling Reaction**
Kazuhiko Semba, Naoto Bessho, Tetsuaki Fujihara, Jun Terao, and Yasushi Tsuji*
Abstract: Borylative allyl–allyl coupling using allenes, bis(pi-
nacolato)diboron, and allyl phosphates has been developed in
the presence of a copper catalyst bearing an N-heterocyclic
carbene ligand. The reaction affords boryl-substituted 1,5-
diene derivatives in good to high yields with high regioselec-
tivity and Z selectivity.
between allyl boronates and allyl electrophiles in the presence
of a chiral palladium/bis(phosphane) catalyst.[1m]
Herein, we report the first borylative allyl–allyl coupling
reaction using allenes (1),[4] bis(pinacolato)diboron [B2-
(pin)2], and allyl phosphates (2)[1e,f] in the presence of
a copper catalyst (Scheme 1). In the reaction, the allyl
copper species bearing a boryl functionality at the bN-position
is generated catalytically,[4,5] and reacts with 2. The reaction
provides a variety of boryl-substituted 1,5-dienes (3) with
excellent stereo- and regioselectivities (aN–gE).[6] The boryl
functionality is very useful for further derivatization.[7]
As shown in Table 1, the reaction conditions were
optimized with 1a, (Z)-2a, and B2(pin)2 as the substrates in
A
llyl–allyl coupling[1] between allyl nucleophiles and allyl
electrophiles is a powerful tool for providing direct access to
1,5-dienes, which are abundant in naturally occurring ter-
penes,[2] and are versatile building blocks in organic syn-
thesis.[3] Stoichiometric amounts of allyl lithium,[1a,b] magne-
sium,[1c–h] tin,[1i,j] boron,[1k–n] silicon,[1o] and indium[1p] nucleo-
philes were reacted with allyl electrophiles. However, with
unsymmetrical allyl reagents, the coupling would occur
between either the a- or g-position of the nucleophile (aN
or gN) and that of the electrophile (aE or gE). Therefore, there
are four possible regioisomers, and more isomers may appear
if stereoisomers are also considered [Eq. (1)]. In fact, allyl–
allyl couplings often suffer from low regio- and stereoselec-
Scheme 1. Borylative allyl–allyl coupling.
Table 1: Reaction optimization.[a]
tivities, even if transition-metal catalysts such as copper,[1e–g]
nickel,[1k] and palladium[1k,p] are employed. Thus, achieving
both regio- and stereoselective allyl–allyl coupling is a chal-
lenging task. Recently, Morken and co-workers successfully
developed a selective (gN–gE) allyl–allyl coupling reaction
Entry
Changes from the
standard conditions
(Z)-3a
Yield [%][b]
Isomeric
Purity [%][c]
[*] Prof. Dr. K. Semba,[+] N. Bessho, Prof. Dr. T. Fujihara,
Prof. Dr. J. Terao, Prof. Dr. Y. Tsuji
1
2
3
4
5
6
7
8
none
84 (77)[d]
95 (98)[e]
SIMes·HBF4 instead of ICy·HBF4
IMes·HCl instead of ICy·HBF4
MeIMes·HCl instead of ICy·HBF4
IPr·HCl instead of ICy·HBF4
PCy3 instead of ICy·HBF4
dppb instead of ICy·HBF4
(E)-2a instead of (Z)-2a
LG=Br: (Z)-2aa
90
88
85
8
84
71
58
16
1
94
91
94
78
91
90
91
97
–
Department of Energy and Hydrocarbon Chemistry
Graduate School of Engineering
Kyoto University, Kyoto 615-8510 (Japan)
E-mail: ytsuji@scl.kyoto-u.ac.jp
Homepage: http://twww.ehcc.kyoto-u.ac.jp/
[+] Present Address: Department of Material Chemistry
9
10
11
Graduate School of Engineering
Kyoto University, Kyoto 615-8510 (Japan)
LG=OC(O)OMe: (Z)-2ab
LG=OAc: (Z)-2ag
0
–
[**] This work was supported by a Grant-in-Aid for Scientific Research on
Innovative Areas (“Organic synthesis based on reaction integra-
tion” and “Molecular activation directed toward straightforward
synthesis”) from MEXT (Japan). K.S. is grateful to a Research
Fellowship of JSPS for Young Scientists.
[a] Standard conditions: 1a (0.38 mmol, 1.5 equiv), B2(pin)2 (0.40 mmol,
1.6 equiv), (Z)-2a (LG=OP(O)(OEt)2, 0.25 mmol), CuCl (0.025 mmol,
10 mol%), ICy·HBF4 (0.030 mmol, 12 mol%), KOtBu (0.38 mmol,
1.5 equiv), THF (2.0 mL), 258C, 24 h. [b] Yield of (Z)-3a as determined
by GC. [c] A ratio of (Z)-3a/other isomers. [d] Yield of isolated (Z)-3a.
[e] Purity of the isolated product. THF=tetrahydrofuran.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201404173.
Angew. Chem. Int. Ed. 2014, 53, 9007 –9011
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
9007

.
Angewandte
Communications
Table 2: Borylative allyl–allyl coupling.[a]
Entry
Allene
Allyl
phosphate
(Z)-3
Yield
[%][b]
Figure 1. Structure of ligands.
1
2
76
82
the presence of a catalytic amount of CuCl and a ligand in
THF at 258C. The ligand ICy (Figure 1) was found to be the
most effective, thus affording (Z)-3a in 84% yield with 95%
isomeric purity (standard reaction conditions; Table 1,
entry 1). Without the ligand, (Z)-3a was afforded in 2%
yield. From the reaction mixture in entry 1, (Z)-3a was
isolated in 77% yield with 98% isomeric purity. The
Z configuration of the product was determined by X-ray
crystal structure analysis.[8] Furthermore, the present proce-
dure is easily amenable to a gram-scale reaction: by employ-
ing 1.0 gram of 1a (8.2 mmol), 1.4 grams (75% yield with
98% isomeric purity) of (Z)-3a were obtained. Upon
reducing the amount of KOtBu to 0.30 equivalents, the yield
of (Z)-3a decreased to 17%. Other carbene ligands such as
SIMes, IMes, and MeIMes (Figure 1) also afforded the
products in high yields, but the selectivity was somewhat
decreased (Table 1, entries 2–4). With the bulky IPr as the
ligand the yield was considerably reduced to 8% (entry 5). As
for the phosphane ligands,[9] PCy3 and dppb were efficient,
thus affording (Z)-3a in good yields with greater than 90%
selectivity (entries 6 and 7). In contrast, phosphanes such as
dppp, dppe, dppbz, Xantphos, and PPh3 provided the product
in much lower yields and with lower selectivities (see Table S1
in the Supporting Information).[8] Even when employing (E)-
2a instead of (Z)-2a as the allyl phosphate, the same (Z)-3a
was obtained as the product in 58% yield with 91% isomeric
purity (entry 8). With respect to the leaving group of the allyl
electrophile, the corresponding allyl bromide (Z)-2aa,
instead of the phosphate, afforded (Z)-3a in 16% yield
(entry 9). Only a trace amount of (Z)-3a, if any, was provided
from the corresponding allyl carbonate (Z)-2ab and acetate
(Z)-2ag (entries 10 and 11).
Other allenes (1b–h) and allyl phosphates (2a–d) were
reacted under the standard reaction conditions (Table 2).
Various 1-monosubstituted allenes (1b–h) reacted with (Z)-
2a to provide the corresponding products (Z)-3b–h in high
yields upon isolation, regio- and stereoselectively (isomeric
purities > 95%; entries 1–7). Silyl ether[9] (entry 4), olefin
(entry 5), ester (entry 6), and bromo (entry 7) functionalities
were tolerated under these reaction conditions. In contrast, 1-
phenylallene, and 1,1-di- and 1,3-disubstituted allenes did not
give the desired products selectively. The g-cyclohexyl-
substituted allyl phosphate 2b and 1a afforded the corre-
sponding adduct (Z)-3i in 62% yield, with a slightly lower
isomeric purity of 92% (entry 8). In the case of b-methyl- (2c)
and b-cyclohexyl-substituted (2d) allyl phosphates, the cor-
responding products, (Z)-3j and (Z)-3k, were obtained in
good yields with high selectivities (entries 9 and 10). The
Z configurations of (Z)-3b, (Z)-3c, and (Z)-3h were deter-
mined through NOESY measurements of these products. The
(Z)-2a
(Z)-2a
3
75
4
5
(Z)-2a
(Z)-2a
57
84
6
7
(Z)-2a
(Z)-2a
79
72
8[c]
9
1a
1a
1a
62[d]
76
10
63
[a] Allene (0.75 mmol, 1.5 equiv), B2(pin)2 (0.80 mmol, 1.6 equiv), allyl
phosphate (0.50 mmol), CuCl (0.050 mmol, 10 mol%), ICy·HBF4
(0.060 mmol, 12 mol%), KOtBu (0.75 mmol, 1.5 equiv), THF (4.0 mL),
258C, 24 h. [b] Yield of the isolated product: isomeric purity >95%.
[c] CuCl (0.10 mmol, 20 mol%), ICy·HBF4 (0.12 mmol, 24 mol%),
KOtBu (1.0 mmol, 2.0 equiv) and B2(pin)2 (1.1 mmol, 2.1 equiv).
[d] Isomeric purity, 92%. TBS=tert-butyldimethylsilyl.
Z geometries of all the other products were confirmed
similarly after derivatization of (Z)-3 through a Suzuki–
Miyaura coupling reaction with 4-bromotoluene (see
Scheme 3c and the Supporting Information).
The reaction of a-substituted allyl phosphates (2e–g) may
afford more stereoisomers (5E and 5Z). Gratifyingly, with
ICy as the ligand under the standard reaction conditions
(Table 1, entry 1), 2e, 2 f, and 2g gave the 5E-configured
products (1Z,5E)-3l–n in good yields upon isolation with
good (84%) to excellent (98%) 5E selectivities (Scheme 2a).
Here, yields of the by-products other than (1Z,5E)-3 and
(1Z,5Z)-3 were less than 5%. Remarkably, with SIMes as the
ligand, the stereoselectivity was switched from 5E to 5Z. Thus,
(1Z,5Z)-3l and (1Z,5Z)-3m were isolated in good yields with
9008 www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 9007 –9011

Angewandte
Chemie
Scheme 4. A possible catalytic cycle.
Scheme 2. Reactions with a-substituted allyl phosphates (2e–h).
high 5Z selectivity (96% and 91%, respectively; Scheme 2b).
Furthermore, the a,g-di-substituted allyl phosphate (Z)-2h
reacted with 1a to afford (1Z,5E)-3o with 98% isomeric
purity (Scheme 2c).
A possible catalytic cycle is shown in Scheme 4. The tert-
butoxy copper species A[11] is generated from CuCl, the
carbene ligand (L), and KOtBu (step a). Then, A reacts with
B2(pin)2 to afford the borylcopper species[5a,12] B (step b). The
allene 1 reacts with B to generate the b-boryl (Z)-s-allyl
The boryl moieties of the adducts were removed easily
through the protodeborylation reaction.[10] Thus, (Z)-3e and
(Z)-3k provided high yields of the corresponding (Z)-1,5-
dienes [(Z)-3eH and (Z)-3kH] (Scheme 3a,b), which are not
easily prepared by the conventional allyl–allyl coupling
reaction.[1f] Suzuki–Miyaura coupling[7] of (Z)-3g with 4-
bromotoluene proceeded smoothly to afford (Z)-3gAr in 99%
yield (Scheme 3c). Similar coupling of (Z)-3d–f, (Z)-3i–k,
(1Z,5E)-3l–o, and (1Z,5Z)-3l–m with 4-bromotoluene
afforded the corresponding (Z)-3dAr–fAr, (Z)-3iAr–kAr,
(1Z,5E)-3lAr–oAr, and (1Z,5Z)-3lAr–mAr in good to quantita-
tive yields (see Table S2).[8] The coupling reaction with vinyl
bromide also provided the corresponding triene (Z)-3kV
quantitatively (Scheme 3d).
copper intermediate C[4,5] regio- and stereoselectively
[13]
=
(step c). Addition of C to the C C bond of 2 occurs, thus
giving D (step d). Subsequently, stereoselective b-elimina-
tion[13] releases the copper phosphate E and provides 3 as the
product (step e). Finally, the reaction of E with KOtBu
regenerates A and the catalytic cycle is completed (step f).
These catalytic steps in Scheme 4 were confirmed by
stoichiometric reactions[8] employing MeIMes as the ligand,
which is an efficient ligand in the catalytic reaction (entry 4 in
Table 1). As a model reaction for step a, the stoichiometric
reaction of [(MeIMes)CuCl] with NaOtBu afforded
[(MeIMes)Cu(OtBu)] (A’) in 72% yield after recrystallization.
As in step b, A’ reacted with B2(pin)2 to give [(MeIMes)Cu-
B(pin)] (B’; Scheme 5a).[5a,12] Moreover, reaction of B’ with
1a provided the b-boryl Z-s-allyl copper C’[14] (Scheme 5b;
see step c). Finally, C’ reacted with (Z)-2a to afford (Z)-3a in
60% yield (Scheme 5b; see steps d and e).[15] In contrast,
there might be some possibility that allenes (1) and B2(pin)2
react first to give diboration adducts,[16] after which the
adducts react with allyl phosphates (2) to afford 3. Hence, the
corresponding diboration adduct 4a was prepared by a liter-
Scheme 3. Derivatization of the products.
Scheme 5. Reactions relevant to the reaction mechanism.
Angew. Chem. Int. Ed. 2014, 53, 9007 –9011
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 9009

.
Angewandte
Communications
ature method.[17] However, the reaction between 4a and (Z)-
2a was very sluggish[18] and the desired product (Z)-3a was
afforded in only low yield (Scheme 5c). Therefore, these
observations in Scheme 5 indicate that the catalytic reaction
proceeds via the b-boryl (Z)-s-allyl copper species C as shown
in Scheme 4, rather than by the diboration of the allenes. As
for the 5E/5Z switch with ICy and SIMes (Scheme 2), both
the ligands might afford the same intermediate such as D’[19]
(Scheme 6) in step d (Scheme 4). From D’, (5E)-3 could be
[2] a) Terpenes-Flavors, Fragrances, Pharmaca, Pheromones (Ed.:
E. Breitmaier), Wiley-VCH, Weinheim, 2006; b) K. C. Nicolaou,
T. Montagnon, Molecules that Changed the World, Wiley-VCH,
Weinheim, 2008.
[3] Selected examples for the use of 1,5-dienes in organic synthesis:
a) R. J. Felix, D. Weber, O. Gutierrez, D. J. Tantillo, M. R.
Gagnꢁ, Nat. Chem. 2012, 4, 405 – 409; b) J. A. Feducia, M. R.
Gagnꢁ, J. Am. Chem. Soc. 2008, 130, 592 – 599; c) Y.-J. Zhao, S.-
S. Chng, T.-P. Loh, J. Am. Chem. Soc. 2007, 129, 492 – 493; d) T. J.
Donohoe, S. Butterworth, Angew. Chem. 2003, 115, 978 – 981;
Angew. Chem. Int. Ed. 2003, 42, 948 – 951; e) R. C. D. Brown,
J. F. Keily, Angew. Chem. 2001, 113, 4628 – 4630; Angew. Chem.
Int. Ed. 2001, 40, 4496 – 4498; f) L. E. Overman, F. M. Knoll, J.
Am. Chem. Soc. 1980, 102, 865 – 867.
[4] For copper-catalyzed borylative transformation of allenes, see:
a) F. Meng, F. Jung, A. H. Hoveyda, Org. Lett. 2013, 15, 1414 –
1417; b) B. Jung, A. H. Hoveyda, J. Am. Chem. Soc. 2012, 134,
1490 – 1493; c) W. Yuan, S. Ma, Adv. Synth. Catal. 2012, 354,
1867 – 1872; d) W. Yuan, X. Zhang, Y. Yu, S. Ma, Chem. Eur. J.
2013, 19, 7193 – 7202; e) F. Meng, H. Jang, B. Jung, A. H.
Hoveyda, Angew. Chem. 2013, 125, 5150 – 5155; Angew. Chem.
Int. Ed. 2013, 52, 5046 – 5051.
[5] a) K. Semba, M. Shinomiya, T. Fujihara, J. Terao, Y. Tsuji, Chem.
Eur. J. 2013, 19, 7125 – 7132; b) K. Semba, T. Fujihara, J. Terao,
Y. Tsuji, Angew. Chem. 2013, 125, 12626 – 12629; Angew. Chem.
Int. Ed. 2013, 52, 12400 – 12403.
[6] For copper-catalyzed carboborations, see: a) P. Liu, Y. Fukui, P.
Tian, Z.-T. He, C.-Y. Sun, N.-Y. Wu, G.-Q. Lin, J. Am. Chem. Soc.
2013, 135, 11700 – 11703; b) K. Kubota, E. Yamamoto, H. Ito, J.
Am. Chem. Soc. 2013, 135, 2635 – 2640; c) H. Yoshida, I.
Kageyuki, K. Takaki, Org. Lett. 2013, 15, 952 – 955; d) R.
Alfaro, A. Parra, J. Alemꢂn, J. L. G. Ruano, M. Tortosa, J. Am.
Chem. Soc. 2012, 134, 15165 – 15168; e) L. Zhang, J. Cheng, B.
Carry, Z. Hou, J. Am. Chem. Soc. 2012, 134, 14314 – 14317.
[7] a) N. Miyaura, A. Suzuki, Chem. Rev. 1995, 95, 2457 – 2483; b) A.
Suzuki, H. C. Brown, Organic Syntheses Via Boranes, Vol. 3,
Aldrich, Milwaukee, 2003.
Scheme 6. Stereochemistry of step e in Scheme 4.
obtained with ICy by b-elimination (step e) through the
antiperiplanar conformation[13] (D’anti; Scheme 6). In contrast,
the less-electron-donating[20] SIMes might facilitate coordi-
nation of the phosphate moiety to the copper and the
resulting synperiplanar conformation[13a] (D’syn) could afford
(5Z)-3.
In conclusion, a highly stereo- and regioselective copper-
catalyzed borylative allyl–allyl coupling has been developed.
The reaction affords a wide variety of boryl-substituted 1,5-
dienes in good to high yields. The reaction proceeds via the b-
boryl (Z)-s-allyl copper species as a key catalytic species.
Further studies on the reaction, using optically active
substrates and the reaction mechanism are now in progress.
[8] See the Supporting Information for details.
Received: April 10, 2014
Revised: May 8, 2014
Published online: July 1, 2014
[9] Abbreviations: PCy3, tricyclohexylphosphine; dppb, 1,4-bis(di-
phenylphosphino)butane; dppp, 1,3-bis(diphenylphosphino)-
propane; dppe, 1,2-bis(diphenylphosphino)ethane; dppbz, 1,2-
diphenylphosphinobenzene; Xantphos, 4,5-bis(diphenylphos-
phino)-9,9-dimethylxanthene; TBS, tert-butyldimethylsilyl.
[10] a) T. G. Elford, S. Nave, R. P. Sonawane, V. K. Aggarwal, J. Am.
Chem. Soc. 2011, 133, 16798 – 16801; b) H. C. Brown, G. A.
Molander, J. Org. Chem. 1986, 51, 4512 – 4514; c) H. C. Brown,
K. J. Murray, Tetrahedron 1986, 42, 5497 – 5500.
À
Keywords: allenes · allylic compounds · boron · C C coupling ·
.
copper
[1] a) Y. Yamamoto, K. Maruyama, J. Am. Chem. Soc. 1978, 100,
6282 – 6284; b) Y. Yamamoto, H. Yatagai, K. Maruyama, J. Am.
Chem. Soc. 1981, 103, 1969 – 1975; c) A. Goliaszewski, J.
Schwartz, J. Am. Chem. Soc. 1984, 106, 5028 – 5030; d) A.
Goliaszewski, J. Schwartz, Tetrahedron 1985, 41, 5779 – 5789;
e) A. Yanagisawa, N. Nomura, H. Yamamoto, Synlett 1993, 689 –
690; f) A. Yanagisawa, N. Nomura, H. Yamamoto, Tetrahedron
1994, 50, 6017 – 6028; g) A. S. E. Karlstrçm, J.-E. Bꢀckvall,
Chem. Eur. J. 2001, 7, 1981 – 1989; h) V. Hornillos, M. Pꢁrez,
M. FaÇanꢂs-Mastral, B. L. Feringa, J. Am. Chem. Soc. 2013, 135,
2140 – 2143; i) B. M. Trost, E. Keinan, Tetrahedron Lett. 1980, 21,
2595 – 2598; j) J. Godschalx, J. K. Stille, Tetrahedron Lett. 1980,
21, 2599 – 2602; k) E. F. Flegeau, U. Schneider, S. Kobayashi,
Chem. Eur. J. 2009, 15, 12247 – 12254; l) A. Jimꢁnez-Aquino,
E. F. Flegeau, U. Schneider, S. Kobayashi, Chem. Commun.
2011, 47, 9456 – 9458; m) P. Zhang, L. A. Brozek, J. P. Morken, J.
Am. Chem. Soc. 2010, 132, 10686 – 10688; n) L. A. Brozek, M. J.
Ardolino, J. P. Morken, J. Am. Chem. Soc. 2011, 133, 16778 –
16781; o) M. Murakami, T. Kato, T. Mukaiyama, Chem. Lett.
1987, 1167 – 1170; p) P. H. Lee, S.-Y. Sung, K. Lee, S. Chang,
Synlett 2002, 146 – 148.
[11] N. P. Mankad, D. S. Laitar, J. P. Sadighi, Organometallics 2004,
23, 3369 – 3371.
[12] D. S. Laitar, J. P. Mꢃller, J. P. Sadighi, J. Am. Chem. Soc. 2005,
127, 17196 – 17197.
=
[13] The addition of alkyl copper species to the C C bond of allyl
phosphates and the b-elimination releasing copper phosphates
have been proposed: a) K. Nagao, U. Yokobori, Y. Makida, H.
Ohmiya, M. Sawamura, J. Am. Chem. Soc. 2012, 134, 8982 –
8987; b) H. Ohmiya, U. Yokobori, Y. Makida, M. Sawamura, J.
Am. Chem. Soc. 2010, 132, 2895 – 2897.
[14] We have isolated the first b-boryl s-allyl copper species,[5a] which
were postulated in several catalytic reactions,[4] and fully
characterized by X-ray crystal structure analysis.
[15] d-Silyl s-allyl copper species react with allyl phosphate to afford
1,5-dienes with low selectivity. See: V. Liepins, J.-E. Bꢀckvall,
Eur. J. Org. Chem. 2002, 3527 – 3535.
[16] For copper-catalyzed diborylation of alkynes and arynes using
B2(pin)2, see: H. Yoshida, S. Kawashima, Y. Takemoto, K.
Okada, J. Ohshita, K. Takaki, Angew. Chem. 2012, 124, 239 –
242; Angew. Chem. Int. Ed. 2012, 51, 235 – 238.
9010 www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 9007 –9011

Angewandte
Chemie
[17] F.-Y. Yang, C.-H. Cheng, J. Am. Chem. Soc. 2001, 123, 761 – 762.
[19] In the addition of alkyl copper species to allyl phosphates,[13]
=
similar adducts bearing the same configuration as D’ have been
proposed.
[18] Even at 1508C, reactions of (CH2 CHCH2)3B with allenes are
slow. See: B. M. Mikhailov, Pure Appl. Chem. 1974, 39, 505 – 523,
[20] For electron-donating ability of NHC ligands, see: T. Drçge, F.
Glorius, Angew. Chem. 2010, 122, 7094 – 7107; Angew. Chem. Int.
Ed. 2010, 49, 6940 – 6952.
and references therein.
Angew. Chem. Int. Ed. 2014, 53, 9007 –9011
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
9011