Gold-catalyzed hydrative carbocyclization of 1,5- and 1,7-allenynes mediated by π-allene complex: Mechanistic evidence supported by the chirality transfer of allenyne substrates

Article abstract of DOI:10.1021/jo8004777

(Chemical Equation Presented) We report PPh₃AuCl/AgOTf-catalyzed hydrative carbocyclization of 1,5- and 1,7-allenynes to give cyclized ketones chemoselectively. In this transformation, hydration occurrs regioselectively at the C≡CPh carbon, acc

Full text of DOI:10.1021/jo8004777

Gold-Catalyzed Hydrative Carbocyclization of 1,5- and
1,7-Allenynes Mediated by π-Allene Complex: Mechanistic Evidence
Supported by the Chirality Transfer of Allenyne Substrates
Chun-Yao Yang,† Guan-You Lin,† Hsin-Yi Liao,*,‡ Swarup Datta,† and Rai-Shung Liu*,†
Department of Chemistry, National Tsing-Hua UniVersity, Hsinchu, Taiwan 300, ROC, Department of
Science Education, National Taipei UniVersity of Education, Taipei, Taiwan, ROC
rsliu@mx.nthu.edu.tw
ReceiVed February 29, 2008
We report PPh3AuCl/AgOTf-catalyzed hydrative carbocyclization of 1,5- and 1,7-allenynes to give cyclized
ketones chemoselectively. In this transformation, hydration occurrs regioselectively at the Ct CPh carbon,
accompanied by addition of the Ct CPh carbon to the two terminal allenyl carbons. This method is
effective for the construction of a quaternary carbon center. On the basis of the chirality transfer of
allenyne substrates, control experiments, and theoretic calculations, we propose that this hydrative
carbocyclization proceeds through an initial π-allene complex with a small energy barrier.
Introduction
tions6 of allenynes are mediated exclusively by π-alkyne species
that form stable allyl cation intermediates through an exo or
endo attack of the central allene carbon at the π-alkyne (eq 1).
As a notable exception, Malacria obtained several cyclized pro-
ducts from the π-allene-complexation of hydroxylated 1,5-
allenynes.4i
We report here gold-catalyzed hydrative carbocyclization of
allenynes to give R-ketonyl cyclopentanes or cyclopentenes,
which are generated through an attack of the Ct CR carbon at
the two terminal allenyl carbons7 (eq 2).The chirality transfer
The metal-catalyzed carbocyclization of an acyclic molecule
with an external nucleophilic addition is useful because new
functional groups are thereby introduced onto the cyclized
frameworks.1,2 The cycloisomerization of enynes3 and allenynes4
that provides rapid access to carbocyclic compounds is a recent
advance in gold and platinum catalysis. These reactions generate
reactive carbocations that are trapped with water, alcohols, or
electron-rich arenes to give functionalized cyclized products.5,6
Reported cycloisomerizations4 and nucleophilic carbocycliza-
† National Tsing-Hua University.
(3) Recent reviews: (a) Fu¨rstner, A.; Davies, P. W. Angew. Chem., Int. Ed.
2007, 46, 3410. (b) Gorin, D. J.; Toste, F. D. Nature 2007, 446, 395. (c) Jimenez-
Nunez, E.; Echavarren, A. M. Chem. Commun. 2007, 333. (d) Zhang, L.; Sun,
J.; Kozmin, S. A. AdV. Synth. Catal. 2006, 348, 2271. (e) Hashmi, A. S. H.
Angew. Chem., Int. Ed. 2005, 44, 6990.
‡ National Taipei University of Education.
(1) For metal-catalyzed cyclizations with addition of water or alcohols, see
selected examples: (a) Trost, B. M.; Rudd, M. T. J. Am. Chem. Soc. 2003, 125,
11516. (b) Trost, B. M.; Brown, R. E.; Toste, F. D. J. Am. Chem. Soc. 2000,
122, 5877. (c) Yao, T.; Zhang, X.; Larock, R. C. J. Am. Chem. Soc. 2004, 126,
11164. (d) Chen, Y.; Ho, D. M.; Lee, C. J. Am. Chem. Soc. 2005, 127, 12184.
(e) Odedra, A.; Wu, C.-J.; Pratap, T. B.; Huang, C.-W.; Ran, Y. F.; Liu, R.-S.
J. Am. Chem. Soc. 2005, 127, 3406. (f) Chang, H.-K.; Datta, S.; Das, A.; Odedra,
A.; Liu, R.-S. Angew. Chem., Int. Ed. 2007, 46, 4744.
(4) Selected examples for cycloisomerization of 1-allen-n-ynes: (a) Cadran,
N.; Cariou, K.; Herve´, G.; Aubert, C.; Fensterbank, L.; Malacria, M.; Marco-
Contelles, J. J. Am. Chem. Soc. 2004, 126, 3408. (b) Petit, M.; Aubert, C.;
Malacria, M. Tetrahedron 2006, 62, 10582. (c) Petit, M.; Aubert, C.; Malacria,
M. Org. Lett. 2004, 6, 3937. (d) Brummond, K. M.; Painter, T. O.; Probst, D. A.;
Mitasev, B. Org. Lett. 2007, 9, 347. (e) Bayden, A. S.; Brummond, K. M.; Jordan,
K. D. Organometallics 2006, 25, 5204. (f) Matsuda, T.; Kadowaki, S.; Goya,
T.; Murakami, M. Synlett 2006, 575. (g) Lee, S. I.; Sim, S. H.; Kim, S. M.;
Kim, K.; Chung, Y. K. J. Org. Chem. 2006, 71, 7120. (h) Lin, G.-Y.; Yang,
C.-Y.; Liu, R.-S. J. Org. Chem. 2007, 72, 6753. (i) Zriba, R.; Gandon, V.; Aubert,
C.; Fensterbank, L.; Malacria, M. Chem. Eur. J. 2008, 14, 1482.
(2) For metal-catalyzed cyclization with addition of carbon nucleophiles, see
selected examples: (a) Shintani, R.; Okamoto, K.; Otomaru, Y.; Ueyama, K.;
Hayashi, T. J. Am. Chem. Soc. 2005, 127, 54. (b) Miura, T.; Shimada, M.;
Murakami, M. J. Am. Chem. Soc. 2005, 127, 1094. (c) Cauble, D. F.; Gipson,
J. D.; Krische, M. J. J. Am. Chem. Soc. 2003, 125, 1110. (d) Lautens, M.;
Mancuso, J. J. Org. Chem. 2004, 69, 3478.
10.1021/jo8004777 CCC: $40.75
Published on Web 06/04/2008
2008 American Chemical Society
J. Org. Chem. 2008, 73, 4907–4914 4907

Yang et al.
SCHEME 1
of allenyne substrates to their resulting R-ketonyl cyclopentanes
or cyclopentenes provides evidence for a π-allene pathway.
Results/Discussion
gave [4 + 2]-cycloaddition adduct 39 in 43-55% yields.
Murakami’s catalyst, PtCl2/MeOH, notably led only to a 55%
recovery of starting allenyne 1. AgOTf that showed a distinct
chemoselectivity in wet dioxane (100 °C, 6 h) gave a 1:1
diastereomeric mixture of 1-(E-stryl)-2-carbonylcyclopentane 2
in a 75% yield. The yield of cyclized ketone 2 improved to
85% with AuClPPh3/AgOTf catalyst. In dry 1,4-dioxane,
PPh3AuCl/AgOTf catalyst gave ketone 2 (dr ) 1.0) in 12%
yield besides [4 + 2]-cycloadduct 3 (33%). Compound 3 is
formed from a π-alkyne9 intermediate, which is preferentially
mediated by PtCl2 catalyst.
We prepared various 1-allen-7-ynes 4-20 to assess the
generality of this new hydrative carbocyclization; the results
are depicted in Table 2. Similar to compound 2, the resulting
cyclized ketones 21-35 bear an E-olefin substituent. Entries
1-3 show the variation of the allenyl R3 substituent; upon
comparison of their reaction time and product yields, allenyne
5 bearing a 4-methoxyphenyl substituent appears to be more
efficient than its phenyl 4 and n-butyl analogues 6 in this
carbocyclization. Unreacted allenyne 6 was recovered in a 48%
yield (entry 3). Entries 1 and 4-7 show the variation of the
alkynyl R1 substituent of allenyne substrates; this cyclization
was less efficient for allenynes 7-9 bearing an electron-rich
3-thienyl, 4-methylphenyl, and 4-methoxyphenyl group, and
became particularly unsuitable for allenyne 10 bearing an
electron-deficient 4-trifluoromethylphenyl group. In these cases,
the resulting products 21-26 existed preferentially in an anti-
isomeric form (anti/syn ) 2.6-9.8). This carbocyclization works
effectively for allenynes 11-19 bearing an alkyl group at the
terminal C(3)-allene carbon (R2 ) Me, Et, nBu, entries 8-16),
which gave satisfactory yields (67-83%) of resulting products
27-35 despite generation of a quaternary carbon. Furthermore,
these cyclizations were complete within 1-10 h, much less
duration than for their unsubstituted analogues 4-10 (R2 ) H,
18-72 h). According to these substituent effects, we deduce
that a C(3)-substituted 1-allen-7-yne and a phenylalkynyl group
are the best combination for allenyne substrates.
Before our work, Murakami reported hydrative cyclization
of 1-allen-7-ynes of special types [X ) C(CO2Me)2, NTs, O]
using PtCl2 and MeOH solvent;8 according to a deuterium-
labeling experiment. This reaction was proposed to proceed
through a π-alkyne complex; the mechanism is depicted in
Scheme 1. A disadvantage of this cyclization is the complete
loss of product chirality with use of chiral allenynes because
an allylic cation is formed as depicted in eq 2. The π-allene
moiety7 is also represented by a zwitter ionic form (eq 1), and
a highly substituted allene is generally more active toward
nucleophilic attack. As we envisage the connecting unit [X )
C(CO2Me)2, NTs, O] in Murakami’s system8 to be unsuitable
to activate the adjacent C(3)-allenyl carbon, we undertook to
investigate this cyclization by varying 1-allen-7-yne substrates
and catalysts to achieve a π-allene pathway. This approach
appears to be more efficient than a π-alkyne pathway in the
formation of a new quaternary carbon.
We selected 1-allen-7-yne 1 as the relevant species because
its cyclized ketone product 2 bears a quaternary carbon; the
methyl group also enhances the allene activity. Table 1 shows
the reactions of 1-allen-7-yne 1 over various π-acid catalysts
in wet 1,4-dioxane. AuCl, AuCl3, and AgSbF6 caused complete
decomposition of starting 1 at 100 °C, whereas AuClPPh3 led
exclusively to unreacted 1 under the same conditions. Highly
electrophilic PPh3AuCl/AgSbF6, PtCl2/CO, and PtCl4 preferably
(5) For cyclization of enynes with nucleophiles, see selected examples: (a)
Nieto-Oberhuber, C.; Munoz, M. P.; Lopez, S.; Jimenez-Nunez, E.; Nevado, C;
Herrero-Gomez, E.; Raducan, M.; Echavarren, A. M. Chem. Eur. J. 2006, 12,
1677. (b) Nieto-Oberhuber, C.; Munoz, P. M.; Bunuel, E.; Nevado, C.; Cardenas,
D. J.; Echavarren, A. M. Angew. Chem., Int. Ed. 2004, 43, 2402. (c) Toullec,
P. Y.; Genin, E.; Leseurre, L.; Genet, J.-P.; Michelet, V. Angew. Chem., Int.
Ed. 2006, 45, 7427. (d) Amijis, C. H. M.; Ferre, C.; Echavarren, A. M. Chem.
Commun. 2007, 698.
(6) For cyclization of allenynes with nucleophiles, see: Lemie´re, G.; Gandon,
V.; Agenet, N.; Goddard, J.; de.Kozak, A.; Aubert, C.; Fensterbank, L.; Malacria,
M. Angew. Chem., Int. Ed. 2006, 45, 7596.
(7) Nucleophilic attack on π-allenes exclusively occurs at the two terminal
carbons except for a few cases reported by Widenhoefer,7d and see selected
examples: (a) Hamilton, G. L.; Kang, E. J.; Mba, M.; Toste, D. F. Science 2007,
317, 496. (b) Staben, S. T.; Kennedy-Smith, J. J.; Huang, D.; Corkey, B. K.;
Lalonde, R. L.; Toste, F. D. Angew. Chem., Int. Ed. 2006, 45, 5991. (c) Nishina,
N.; Yamamoto, Y. Angew. Chem., Int. Ed. 2006, 45, 3314. (d) Zhang, Z.; Liu,
C.; Kinder, R. E.; Han, X.; Qian, H.; Widenhoefer, R. A. J. Am. Chem. Soc.
2006, 128, 9066. (e) Liu, Z.; Wasmuth, A. S.; Nelson, S. G. J. Am. Chem. Soc.
2006, 128, 10352. (f) Gockel, B.; Krause, N. Org. Lett. 2006, 8, 4485. (g) Wang,
S.; Zhang, L. Org. Lett. 2006, 8, 4585. (h) Zhang, G.; Catalano, V. J.; Zhang,
L. J. Am. Chem. Soc. 2007, 129, 11359. (i) Hashmi, A. S. K.; Blanco, M. C.;
Fischer, D.; Bats, J. W. Eur. J. Org. Chem. 2006, 1387. (j) Hashmi, A. S. K.;
Schwartz, L.; Choi, J.-H.; Frost, T. M. Angew. Chem., Int. Ed. Engl. 2000, 39,
2285.
Nucleophilic attack on a metal-π-allene occurs regioselec-
tively at two terminal carbons.7 We prepared 1-allen-5-ynes
36-39 to explore the cyclization via attack of the Ct CPh
carbon at the remote allenyl carbon (Table 3). Treatment of these
species with AuPPh3OTf catalyst (5 mol %) delivered cyclo-
pentenyl ketones 41-44 in 65-80% yields, in favor of syn-
(9) For mechanistic formation of [4 + 2]-cycloadducts 3, see: Nieto-
Oberhuber, C.; Lopez, S.; Echavarren, A. M. J. Am. Chem. Soc. 2005, 127,
6178.
(8) Matsuda, T.; Kadowaki, S.; Murakami, M. HelV. Chim. Acta 2006, 89,
1672.
4908 J. Org. Chem. Vol. 73, No. 13, 2008

Gold-Catalyzed HydratiVe Carbocyclization of Allenynes
TABLE 1. Catalytic Cyclization over Various Catalysts
yieldc (%)
catalysta
(1) AuClb
solvent (conditions)
1
2
3
1,4-dioxane (100 °C, 24 h)
1,4-dioxane (100 °C, 24 h)
1,4-dioxane (100 °C, 24 h)
1,4-dioxane (100 °C, 24 h)
CH2Cl2 (23 °C, 30 min)
1,4-dioxane (100 °C, 4 h)
1,4-dioxane (100 °C, 24 h)
MeOH (70 °C, 48 h)
b
(2) AuCl3
(3) AgSbF6
(4) AuClPPh3
b
b
85
55
(5) AuClPPh3/AgSbF6
45
55
43
(6) PtCl2/COb
b
(7) PtCl4
b
(8) PtCl2 (10%)
(9) AgOTfb
1,4-dioxane (100 °C, 6 h)
1,4-dioxane (100 °C, 4 h)
dry 1,4-dioxane (100 °C, 24 h)
75
85
12
(10) AuClPPh3/AgOTfb
(11) AuClPPh3/AgOTfb
33
a 5 mol % catalyst, [allenyne] ) 0.20 M. b At 23 °C for 12 h, starting species 1 was recovered exclusively for entries 1-3 and 5-7. c Products were
separated on a silica column.
TABLE 2. Gold-Catalyzed Hydrative Carbocyclizations of 1,7-Allenynes
entry
allenynesa
R1
R2
R3
time (h)
productsb (yields, anti/syn)
1
2
3
4
5
6
7
8
4
5
6
7
8
Ph
Ph
Ph
H
H
H
H
H
H
H
Ph
36
18
48
28
30
72
72
10
8
6
6
8
8
21 (73%, 5.1)
22 (82%, 2.6)
23 (35%, 11.4)c
24 (55%, 9.1)
25 (46%, 7.1)
26 (16%, 9.8)
4-MeOC6H4
n-Bu
Ph
Ph
Ph
Ph
Ph
Ph
methyl
Ph
n-butyl
2-thienyl
4-MeOC6H4
4-CF3C6H4
n-butyl
Ph
3-thienyl
4-MeC6H4
4-MeOC6H4
4-CF3C6H4
Ph
Ph
Ph
3-thienyl
Ph
Ph
Ph
Ph
Ph
9
10
11
12
13
14
15
16
17
18
19
20
ethyl
27 (78%, 1.2)
28 (75%, 1.0)
29 (70%, 1.4)
30 (67%, 1.2)
31 (75%, 1.7)
32 (74%, 1.7)
33 (83%, 1.3)
34 (75%, 1.2)
35 (75%, 1.2)
20 (65%)
9
n-butyl
methyl
methyl
methyl
methyl
methyl
methyl
ethyl
10
11
12
13
14
15
16
17
1
10
8
n-propyl
methyl
12
a 1,7-allenyne 1 (1 equiv, 0.20 M), PPh3AuCl/AgOTf (5 mol %), 1,4-dioxane (100 °C). b Product yields are given after separation from a silica
column. c Starting allenyne 6 was recovered in a 48% yield.
isomers (syn/anti ) 1.5-6.7),10 But the cyclization of 1-allen-
6-ynes 40 led to its recovery (56%) in wet 1,4-dioxane (100
°C, 12 h); we isolated no side product from an alkyne hydration
of species 40.
The syn-diastereomer of 46 was obtained in 55% ee, whereas
the anti-diastereomer was obtained in 67% ee. Some loss of
the enantiopurity of products 46 is likely caused by gold-
catalyzed epimerization12 of allenynes in hot 1,4-dioxane. In
the case of (R)-1-allen-5-yne 47 (59% ee), the mild conditions
(25 °C) effected complete chirality transfer to resulting products
48-syn (58% ee) and 48-anti (58% ee). The chirality transfer
of substrates tentatively excludes a π-alkyne mechanism pro-
posed by Murakami, which is expected to show a complete loss
of product chirality. To assess the role of allenyl ketone 49 in
this AgOTf/AuClPPh3 system, Scheme 2 shows a control
experiment, which gave only unreacted 49 in the absence or
presence of Et3N; this observation eliminates the intermediacy
We prepared chiral allenynes 45 and 4711 to elucidated the
reaction mechanism (Scheme 2). We propose the absolute
configurations of resulting ketones 46 and 48 on the basis of
our reaction model depicted in Scheme 3. Treatment of (S)-1-
allen-7-yne 45 (ee >98%) with 5 mol % AgOTf/AuClPPh3 in
wet 1,4-dioxane at 90 °C for 2 h produced chiral R-ketonyl
cyclopentene product 46 (80% yield) in syn/anti ) 2:1 ratio.
(10) We have treated cyclopentenyl ketone 42 (syn/anti ) 5.1) with
PPh3AuOTf (5 mol %) in hot 1,4-dioxane (100 °C, 3 h); we recovered ketone
42 present in a 1:1 mixture of syn/anti isomers. This information reveals that
PPh3AuOTf promotes the syn/anti isomerization of ketone 42 in hot 1,4-dioxane.
(11) Preparation of chiral allenes and determination of the ee values of
products are provided in the Supporting Information.
(12) (a) Zhang, Z.; Bender, C. F.; Widenhoefer, R. A. J. Am. Chem. Soc.
2007, 129, 14148. (b) Sherry, B. D.; Toste, F. D. J. Am. Chem. Soc. 2007, 126,
15978.
J. Org. Chem. Vol. 73, No. 13, 2008 4909

Yang et al.
SCHEME 2
TABLE 3. Catalytic Cyclization of 1,5- and 1,6- Allenyens
SCHEME 3
a 1,7-Allenyne (1 equiv, 0.20 M). b Product yields are given after
separation from a sillica column. c syn/anti ratios.
behavior of π-allene, due to gold complexation. Scheme 3 shows
a plausible mechanism to rationalize the accelerating effect of the
allene groups of allenynes 36-39 on the alkyne hydration. This
mechanistic pathway is proposed with reference to a paper by
Widenhoefer.7d Among π-alkyne A9 or π-allene B intermediates,
PPh3AuCl/AgOTf preferably triggers the π-allene pathway via
5-exo-attack of the Ct CPh at the proximate allenyl terminal carbon
to form species C, of which the adjacent phenyl group stabilizes
the vinyl cation. To rationalize the observed E-styryl geometry of
cyclized ketones E, we envisage that species B adopts a cis-
conformation with its R2 substituent opposite the entering alkyne
group to minimize the steric hindrance because the C(2)-C(6) bond
is formed parallel to the C(8)-H bond. In the case of 1-allen-5-
ynes 36-37, we propose that PPh3Au+ polarizes the π-allene B′
of ketone 49. As shown in Table 3 (entry 5), we obtained no
hydration product from allenyne 49 in wet 1,4-dioxane under
catalytic conditions (25-100 °C, 12 h). To understand the role
of alkyne hydration for substrates 36-39 (entry 1-4) which
were readily converted to cyclized ketoens 41-44 at 25 °C,
we prepared alkyne 50, but we observed no alkyne hydration
only its exclusive recovery under similar conditions. This
information indicates that the allene functionality of allenynes
36-39 facilitates their alkyne hydration.
The carbocyclization proceeds through a regioselective hydration
at the Ct CPh carbon, accompanied by attack of the Ct CPh
carbon at the two terminal allenyl carbons. Furthermore, the use
of chiral substrates 45 and 47 led to chirality transfer to their
products 46 and 48. These observations show the electrophilic
4910 J. Org. Chem. Vol. 73, No. 13, 2008

Gold-Catalyzed HydratiVe Carbocyclization of Allenynes
SCHEME 4
to form an angular geometry as represented by B′′, which facilitates
the 5-endo-cyclization.13
production of products E, substrates 8-9 bearing an electron-rich
aryl group are less efficient than their phenyl analogue 4 because
the former greatly favors π-alkyne complex A in the alkyne/allene
equilibrium. In contrast, the electron-deficient arylalkyne as in
allenyne 10 (Table 2, entry 7) is incapable of stabilizing the vinyl
cation C to give desired product E.
To assess our proposed mechanism, Scheme 4 shows the
results of theoretical calculations14 (B3LYP/LANL2DZ). Two
possible π-allene conformers are B and B′. We take the sum of
energies of the free substrate and metal as the energy zero. In
the case of Ag+ and trans-PtCl2(H2O), cis-π-allenes B are likely
to be the active intermediates because their corresponding
transition structures TS have lower energies [TS: -40.28 kcal/
mol for Ag+, -32.53 kcal/mol for trans-PtCl2(H2O)] than those
of trans-π-allene complex B′ [TS′: -32.54 kcal/mol for Ag+,
-24.74 kcal/mol for trans-PtCl2(H2O)]. The preference for cis-
π-allene intermediates B is also shown by their resulting vinyl
cations C, which are more exothermic (ca. 6.2-7.3 kcal/mol)
than species C′ generated from trans-π-allene B′. For M )
AuPH3+, although the trans-π-allene B′ is more stable than the
The proposed mechanism rationalizes also our observed acti-
vity-structure relationship as shown in Table 2. Consistent with
our observations (entries 8-17 versus entries 1-7), π-allene
intermediate B bearing a R1 ) alkyl substituent is expected to be
more active than its unsubstituted analogue (R1 ) H) in this
catalysis because the former tends to form a more polarized
structure B′′′ stabilized by the tertiary cation. To account for the
effects of the arylalkynes of 1-allen-7-ynes, we envisage that, for
(13) The proposed mechanism is also supported by deuterium labeling
experiments. Hydrative carbocyclization with allenyne 4 in the presence of D2O
(1.0 equiv) gave cyclized ketone 21 with 40% and 43% deuterium content at its
CHdCHPh and CHCOPh carbons, respectively.
J. Org. Chem. Vol. 73, No. 13, 2008 4911

Yang et al.
SCHEME 5
cis-complex B by 6.35 kcal/mol, the barrier height for B f C
(+2.08 kcal/mol) is still much smaller than that for B′f C′
(+15.50 kcal/mol). Here, vinyl cation C has an energy of 5.1
kcal/mol less than that of complex C′. On the basis of the energy
a π-alkyne mechanism is excluded because its transition state
TS-1 has an extraordinary large energy (-25.65 kcal/mol). For
Ag+ and AuPH3+, the activation energies for transformation F
f G are up to 17.66 and 17.50 kcal/mol, respectively. This
information indicates that the π-alkyne mechanism is less likely
to occur here. Furthermore, cis-π-allene intermediate B is more
exothermic than π-alkyne species A in the complexation of
allenyne with metal species. The energy differences are 4.64,
profiles in Scheme 4, we envisage that Ag+ andAuPH3 are
+
more efficient than trans-PtCl2(H2O) in activating cis-π-allene
because the platinum system has a large barrier of +17.5 kcal/
mol for the transformation B f C.
5.02, and 3.76 kcal/mol for Ag+, trans-PtCl2(H2O), and AuPH3
+
Our calculations support the preference for allenyne substrates
bearing an arylalkyne group. For allenyne 20 bearing a
propylalkyne group, with AuPH3 catalyst, we found a high
energy of the corresponding vinyl cation C (-45.2 kcal/mol)
with a large activation barrier (+5.12 kcal/mol) in the B f C
conversion (see the Supporting Information).
,respectively, in favor of cis-π-allene B. The thermodynamic
perference for cis-π-allenes B and the small energy barrier in
the formation of vinyl cations C imply that the hydrative
+
+
cyclization of allenynes are performed better by AuPH3 and
Ag+ through cis-π-allene B (Scheme 5).
We performed calculations on π-alkyne pathways, proposed
by Murakami, to assess their possible participation. The
activation energies to form allylic cation F are +9.11, +19.36,
and +8.42 kcal/mol for M ) Ag+, trans-PtCl2(H2O), and
AuPH3+, respectively. These barrier heights are larger than those
for the π-allene pathways that we propose. The role of PtCl2 in
In summary, we report a new hydrative carbocyclization of
1,5- and 1,7-allenynes catalyzed by PPh3AuCl/AgOTf, giving
cyclized ketones chemoselectively. In such cyclizations, water
attacks regioselectively at the Ct CPh carbon, accompanied by
attack of the tethered Ct CPh carbon at the two terminal allenyl
carbons. The use of chiral allenynes provides desired ketones
4912 J. Org. Chem. Vol. 73, No. 13, 2008

Gold-Catalyzed HydratiVe Carbocyclization of Allenynes
acetate, washed with water, dried over MgSO4, and concentrated
in vacuo. The residues were chromatographed on a silica column
(hexane:ethyl acetate 1:9, Rf 0.5) to afford compound 1d (697
mg, 2.74 mmol, 80%) as a colorless oil.
with small loss of chirality. In our control experiments, the allene
group of substrates facilitates the hydration of their tethered
alkynes. On the basis of theoretic calculations, we conclude that
the hydrative cyclization is likely mediated by π-allene inter-
mediates with a low energy barrier. This new method is efficient
to provide cyclized ketones bearing a new quaternary carbon;
the reaction is usefully complementary to Murakami’s work.8
The application of this method to the synthesis of bioactive
molecules is under current investigation.
(d) Synthesis of 3-Methyl-1,8-diphenylocta-1,2-dien-7-yne (1).
To a THF solution of phenylmagnesium bromide (181 mg, 1.0
mmol) was added ZnCl2 (1 M in THF, 1.0 mL, 1.0 mmol)
dropwise at 0 °C; the mixture was stirred for 1.0 h before adding
Pd(PPh3)4 (4.6 mg, 0.004 mmol) and compound 1d (203 mg,
0.8 mmol). The resulting mixture was stirred for 1.0 h before
warming near 23 °C; the solvent was removed in vacuo before
treatment with a saturated NaHCO3 solution. The organic layer
was extracted with diethyl ether, washed with a saturated NaCl
solution, and dried over anhydrous MgSO4. The residues were
chromatographed through a silica gel column (hexane) to afford
compound 1 (141 mg, 0.52 mmol, 65%) as a yellow oil. IR (neat,
Experimental Section
I. Procedures for Theoretic Calculations. All geometries
(reactants, intermediates, transition states, and products) had been
fully optimized by using the B3LYP method14,15 with the LANL2DZ
basis set16 (B3LYP/LANL2DZ). Density Functional Theory (DFT)
is recognized as a valuable tool for the study of transition sates
and reaction intermediates. The vibrational frequencies, at the same
level of theory, were computed to characterize the stationary points
as true minima or saddle points on the potential energy hypersur-
faces. All of the calculations were performed with the GAUSSIAN
03 package.17 The unscaled zero-point vibrational energies (ZPE)
are included in the reported energies.
1
cm-1): 3080 (m), 2210 (w), 1961 (s), 1598 (m), 1495 (s); H
NMR (400 MHz, CDCl3) δ 7.37-7.35 (m, 2 H), 7.28-7.24 (m,
7 H), 7.17-7.14 (m, 1 H), 6.08 (q, J ) 2.8 Hz, 1 H), 2.44 (t, J
)7.2 Hz, 2 H), 2.27-2.21 (m, 2 H), 1.83 (d, J ) 2.8 Hz, 3 H),
1.78 (td, J ) 7.4, 2.4 Hz, 2 H); 13C NMR (100 MHz, CDCl3) δ
202.5, 135.7, 131.5, 128.5, 128.1, 127.4, 126.5, 126.4, 123.9,
103.0, 94.3, 89.8, 80.9, 33.0, 26.5, 19.0, 18.9; HRMS calcd for
C21H20 272.1565, found 272.1568.
II. Experimental Procedures for Synthesis of the Substrates.
Synthesis of 3-Methyl-1,8-diphenylocta-1,2-dien-7-yne (1).
(a) Synthesis of 3-Methyl-1-(trimethylsilyl)-8-phenylocta-1,7-
diyn-3-ol (1b). To a mixture of trimethylsilylacetylene (0.80 g, 8.2
mmol) and THF (20 mL) was added n-BuLi (2.5 M in hexanes,
2.1 mL, 5.3 mmol) dropwise at -78 °C, and the solution was stirred
for 30 min before addition of compound 1a (762 mg, 4.1 mmol).
The solution was stirred for 1 h before treatment with an aqueous
NH4Cl solution, then extracted with ethyl acetate, dried over
anhydrous MgSO4, and concentrated under reduced pressure to
afford compound 1b (908 mg, 3.2 mmol, 78%) as a yellow oil.
(b) Synthesis of 3-Methyl-1-(trimethylsilyl)-8-phenylocta-1,7-
diyn-3-yl Acetate (1c). To a dichloromethane solution of 1b (0.50
g, 1.76 mmol) was added DMAP (12 mg, 0.1 mmol), acetic
anhydride (0.61 mg, 6.0 mmol), and triethylamine (0.92 g, 9.0
mmol) at 28 °C; the resulting solution was stirred for 12 h before
being quenched with a saturated NaHCO3 solution. The solution
was extracted with dichloromethane, washed with water, dried over
MgSO4, and concentrated under reduced pressure. The residues were
purified by filtration on a short silica pad to afford compound 1c
(401 mg, 1.23 mmol, 70%) as a yellow oil.
III. A Typical Procedure for Catalytic Cyclization of 3-Methyl-
1,8-diphenylocta-1,2-dien-7-yne (1).
Synthesis of (E)-(2-Methyl-2-styrylcyclopentyl)(phenyl)meth-
anone (2). A solution of PPh3AuOTf (5 mol %) was prepared
by mixing PPh3AuCl (9.1 mg, 0.018 mmol) and AgOTf (4.6
mg, 0.018 mmol) in 1,4-dioxane (1.9 mL). To this solution was
added compound 1 (100 mg, 0.37 mmol) at 100 °C, then the
mixture was stirred for 4 h. The resulting solution was filtered
through a celite bed and eluted through a silica gel column (ethyl
acetate/hexane 1/15) to give compound 2 (91 mg, 0.28 mmol,
85%) as a yellow oil. IR (neat, cm-1) 3050 (m), 2945 (s), 1708
1
(c) Synthesis of 3-Methyl-1-(trimethylsilyl)-8-phenylocta-1,7-
diyn-3-yl Acetate (1d). To a THF solution (20.0 mL) of
compound 1c (1.12 g, 3.43 mmol) was added Bu4NF (1.0 M,
3.5 mL, THF) at 25 °C. The resulting mixture was stirred for
30 min, the solvent was removed in vacuo, and a saturated NH4Cl
solution was added. The organic layer was extracted with ethyl
(s), 1513 (s); H NMR (400 MHz, CDCl3) (anti/syn ) 1.0) δ
7.90 (d, J ) 7.2 Hz, 2 H), 7.85 (d, J ) 8.8 Hz, 2 H), 7.51-7.45
(m, J ) 8.0 Hz, 2 H), 7.44-7.36 (m, 4 H), 7.30-7.24 (m, 4 H),
7.22-7.13 (m, 4 H), 7.08 (d, J ) 8.0 Hz, 2 H), 6.29, 6.22 (AB
quartet, J ) 16.0 Hz, 2 H), 6.20, 6.13 (AB quartet, J ) 16.0
Hz, 2 H), 3.82 (t, J ) 7.6 Hz, 1 H), 3.76 (t, J ) 7.2 Hz, 1 H),
2.40-2.29 (m, 2 H), 2.05-1.62 (m, 10 H), 1.08 (s, 3 H), 1.37 (s, 3
H); 13C NMR (100 MHz, CDCl3) δ 202.9, 202.2, 139.3, 138.7, 138.3,
137.6, 137.4, 136.1, 132.5, 132.3, 128.4, 128.3, 128.3, 128.2, 128.2,
128.1, 126.9, 126.7, 126.6, 126.2, 125.9, 125.9, 56.5, 54.9, 48.9, 48.4,
41.3, 39.6, 28.2, 27.8, 26.3, 23.0, 22.8, 20.9; HRMS calcd for C21H22O
290.1671, found 290.1673.
(14) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(15) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. 1988, B37, 785.
(16) (a) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270. (b) Wadt,
W. R.; Hay, P. J. J. Chem. Phys. 1985, 82, 284. (c) Hay, P. J.; Wadt, W. R.
J. Chem. Phys. 1985, 82, 299.
(17) Frisch, M. J. Gaussian 03, revision A1; Gaussian, Inc.: Pittsburgh, PA, 2003.
J. Org. Chem. Vol. 73, No. 13, 2008 4913

Yang et al.
1
Acknowledgment. The authors wish to thank the National
Science Council, Taiwan for supporting this work.
compounds 1-49, and copies of H and 13C NMR spectra for
compounds 1-49. This material is available free of charge via
the Internet at http://pubs.acs.org.
Supporting Information Available: Experimental proce-
dures including synthesis of allenyne 45, spectra data for
JO8004777
4914 J. Org. Chem. Vol. 73, No. 13, 2008