Gold catalyzed diastereoselective cascade allylation/enyne cycloisomerization to construct densely functionalized oxygen hetereocycles

Article abstract of DOI:10.1021/ol1012923

(Equation Presented). A new tandem allylation/enyne cycloisomerization reaction was developed to construct densely functionalized oxygen hetereocycles with high diastereoselectivities from the intermolecular reaction of allylic acetates with propargylic a

Full text of DOI:10.1021/ol1012923

ORGANIC
LETTERS
2010
Vol. 12, No. 15
3468-3471
Gold Catalyzed Diastereoselective
Cascade Allylation/Enyne
Cycloisomerization to Construct Densely
Functionalized Oxygen Hetereocycles
Zili Chen,*,† Yu-Xin Zhang,‡ Ya-Hui Wang,† Li-Li Zhu,† Heng Liu,† Xiao-Xiao Li,†
and Lin Guo*,‡
Department of Chemistry, Renmin UniVersity of China, Beijing 100872, China, and
School of Chemistry and EnVironment, Beijing UniVersity of Aeronautics and
Astronautics, Beijing 100083, China
zilichen@ruc.edu.cn; guolin@buaa.edu.cn
Received June 4, 2010
ABSTRACT
A new tandem allylation/enyne cycloisomerization reaction was developed to construct densely functionalized oxygen hetereocycles with high
diastereoselectivities from the intermolecular reaction of allylic acetates with propargylic alcohols via gold catalysis. Terminal and nonterminal
propargylic alcohols take different reaction routes either to provide 3-oxa-bicyclo[4.1.0]hept-4-ene derivatives 5 or to give endocyclic rearrangement
products 7 and alkoxycyclization adducts 8. Cyclopropane’s stereochemistry was mainly determined by allylic substituents.
Transition metal-catalyzed cycloisomerization of 1,n-enynes
is one of the most attractive strategies for the synthesis of
the complicated cyclic molecules.1,2 Almost all reported
results to date in this area are intramolecular reactions, which
require additional synthetic steps to prepare the enyne
substrates. An intermolecular-type reaction, if possible, would
not only minimize the use of chemicals and the waste
production, but also improve the reaction efficiency and
broaden substrate diversity. However, extending these in-
tramolecular rearrangements to intermolecular processes is
very difficult, due to the unfavored entropic binding penalties,
and inefficient regio- and stereocontrol. A partial approach
to this transformation has just been reported in the research
of the Pauson-Khand reaction by using a pseudointermo-
lecular process, which combined the synthesis of enyne and
its further cyclization into a one-pot reaction, induced by
either one or two metal complexes.3,4 Despite these achieve-
(2) Selected reviews on Pt- or Au-catalyzed reactions involving enyne
cycloisomerizations: (a) Gorin, D. J.; Sherry, B. D.; Toste, F. D. Chem.
ReV. 2008, 108, 3351. (b) Jime´nez-Nu´n˜ez, E.; Echavarren, A. M. Chem.
ReV. 2008, 108, 3326. (c) Shen, H. C. Tetrahedron 2008, 64, 7847. (d)
Hashmi, A. S. K. Chem. ReV. 2007, 107, 3180. (e) Gorin, D. J.; Toste,
F. D. Nature 2007, 446, 395. (f) Jime´nez-Nu´n˜ez, E.; Echavarren, A. M.
Chem. Commun. 2007, 333. (g) Ma, S.-M.; Yu, S.; Gu, Z. Angew. Chem.,
Int. Ed. 2006, 45, 200. (h) Hashmi, A. S. K.; Hutchings, G. J. Angew. Chem.,
Int. Ed. 2006, 45, 7896. (i) Zhang, L.; Sun, J.; Kozmin, S. A. AdV. Synth.
Catal. 2006, 348, 2271. (j) Bruneau, C. Angew. Chem., Int. Ed. 2005, 44,
2328. (k) Diver, S. T.; Giessert, A. J. Chem. ReV. 2004, 104, 1317. (l)
An˜orbe, L.; Dominguez, G.; Pe´rez-Castells, J. Chem.sEur. J. 2004, 10,
4938. (m) Mendez, M.; Mamane, V.; Fu¨rstner, A. Chemtracts 2003, 16,
397. (n) Aubert, C.; Buisine, O.; Malacria, M. Chem. ReV. 2002, 102, 813.
† Renmin University of China.
‡ Beijing University of Aeronautics and Astronautics.
(1) Selected general reviews: (a) Michelet, V.; Toullec, P. Y.; Geneˆt,
J.-P. Angew. Chem., Int. Ed. 2008, 47, 4268. (b) Fu¨rstner, A.; Davies, P. W.
Angew. Chem., Int. Ed. 2007, 46, 3410. (c) Aubert, C.; Fensterbank, L.;
Gandon, V.; Malacria, M. Top. Organomet. Chem. 2006, 19, 259. (d)
Echavarren, A. M.; Nevado, C. Chem. Soc. ReV. 2004, 33, 431
.
10.1021/ol1012923  2010 American Chemical Society
Published on Web 07/07/2010

The structure of 5a, as shown in Figure 1, was identified
to be a polysubstituted 3-oxa-bicyclo[4.1.0]hept-4-ene
Table 1. Gold-Catalyzed Intermolecular Reaction of
(E)-1,5-Diphenyl Pent-2-enyl Acetate 1a and
3-Phenylprop-2-yn-1-ol 2a To Give
3-Oxa-bicyclo[4.1.0]hept-4-ene Derivatives 5aa
yield of yield of
1a/2a time/temp 4a (%)b 5a (%)b
1
2
3
5% Au(pph3CI/AgOTf 1/1.5
5% Au(PPh3)NTf2
5% Au(PPh3)NTf2
a Unless noted, all reactions were carried out at 0.1 mmol scale in 3
1 h/rt
1/1.5 0.5 h/rt
1/6 1 h/30 °C
15
<5
45
86
8
trace
mL of CH2Cl2 at 25 °C. b Isolated yields.
Figure 1. X-ray chromatograph of compound 5a.
ments, there are few reports of employing a similar method
in the study of the enyne cycloisomerization reaction.4
In this paper, we will report a convenient new method to
construct densely functionalized 5- or 6-membered oxygen
hetereocycles with high diastereoselectivities from the in-
termolecular reaction of allylic acetates with propargylic
alcohols via gold catalysis. This is the first report, to the
best of our knowledge, of a two-step tandem process
involving intermolecular allylic substitution5 and intramo-
lecular 1,6-enyne cycloisomerization6 in which the Au(I)
catalyst played a mechanistically distinctive dual role.
The reaction of (E)-1, 5-diphenylpent-2-enyl acetate 1a with
3-phenylprop-2-yn-1-ol 2a was chosen as the model system for
our initial investigation. When 1 equiv of 1a and 1.5 equiv of
2a were treated with 5% equiv of Au(PPh3)Cl/AgOTf in DCM
for 1 h, 1a’s rearrangement isomer 3a was obtained in 70%
yield,5c,d along with enyne ether 4a (15% yield), and a trace
amount of cyclic product 5a (Table 1, entry 1).
derivative,6c,d,7 in which the 2-phenylethyl group and the
cyclopropyl group are located on the same side of the
dihyropyran ring (the reaction in Table 1). It was notable
that 5a was obtained as the single diastereomer.
Further experiments proved that Au(PPh3)NTf2 (5% equiv)
performed better than other catalysts (Table 1, entry 2).
Optimization of the 1a/2a ratio, catalyst loading, and reaction
time/temperature identified a set of best conditions to give
5a in 86% yield (Table 1, entry 3).8
Under the conditions from entry 3 in Table 1, the scope
and limitations for this reaction were then explored.9 Some
representative examples for the preparation of 3-oxa-
bicyclo[4.1.0]hept-4-ene derivatives 5 were summarized in
Table 2.6c,d The reactions of a number of disubstituted allylic
acetates with mono- or disubstituted propargylic alcohols
were investigated to determine the influence of various
substitution patterns (R1-R5). Both aromatic and aliphatic
substituents worked well at the R4 position, in which
substrates with electron-rich aryl groups worked better than
that with electron-deficient aryl groups and alkyl groups
(2a-f, Table 2, entries 1-6). At the R2, R3 position, both
aryl (1a, 1b,c) and dialkyl (1d) substrates gave the desired
products in moderate to good yields (Table 2, entries 1 and
7-9). The low reaction yield of 5h is due to the low
reactivity of the in situ generated enyne ether intermediate.9
At the R1 position, replacement of the phenylethyl group
with the bulky phenyl group (1e, Table 2, entries 10 and
11) lowered the reaction yield. Disubstituted propargylic
alcohols were also examined in this reaction. The reaction
of 1a with 2g or 2i gave the desired product 5m and 5o in
moderate yields (Table 2, entries 12 and 14), while the
reaction of 1e with 2h gave 5n in low yield (Table 2, entry
13). As shown in Table 2, when “R5” is a hydrogen atom,
(3) Selected reviews on intermolecular cascade reactions: (a) Malacria,
M. Chem. ReV. 1996, 96, 289. (b) Padwa, A.; Weingarten, M. D. Chem.
ReV. 1996, 96, 223. (c) Aubert, C.; Fensterbank, L.; Gandon, V.; Malacria,
M. Top. Organomet. Chem. 2006, 19, 259. (d) Nicolaou, K. C.; Edmonds,
D. J.; Bulger, P. G. Angew. Chem., Int. Ed. 2006, 45, 7134. (e) Fogg, D. E.;
dos Santos, E. N. Coord. Chem. ReV. 2004, 248, 2365
.
(4) Examples of intermolecular allylation/Pauson-Khand cascade reac-
tions: (a) Malacria, B. L.; Miller, K. A.; Smith, A. J.; Tran, K.; Martin,
S. F. Org. Lett. 2005, 7, 1661. (b) Evans, P. A.; Robinson, J. E. J. Am.
Chem. Soc. 2001, 123, 4609. (c) Jeong, N.; Seo, S. D.; Shin, J. Y. J. Am.
Chem. Soc. 2000, 122, 10220.
(5) Gold-catalyzed cyclization of allylic acetate: (a) Wang, Y.-H.; Zhu,
L.-L.; Zhang, Y.-X.; Chen, Z. Chem. Commun. 2010, 46, 577. (b) Porcel,
S.; Lo´pez-Carrillo, V.; Garcssa-Yebra, C.; Echavarren, A. M. Angew. Chem.,
Int. Ed. 2008, 47, 1883. (c) Marion, N.; Gealageas, R.; Nolan, S. P. Org.
Lett. 2007, 9, 2653. (d) Gourlaouen, C.; Marion, N.; Nolan, S. P.; Maseras,
F. Org. Lett. 2009, 11, 81.
(6) Selected examples of Au-catalyzed 1,6-enyne cyclization: (a) Cabello,
N.; Jime´nez-Nu´ez, E.; Bun˜uel, E.; Ca´rdenas, D. J.; Echavarren, A. M. Eur.
J. Org. Chem. 2007, 4217. (b) Nieto-Oberhuber, C.; Mun˜oz, M. P.; Lp´ez,
S.; Jime´nez-Nu´n˜ez, E.; Nevado, C.; Herrero-Go´mez, E.; Raducan, M.;
Echavarren, A. M. Chem.sEur. J. 2006, 12, 1677. (c) Lee, S. I.; Kim,
S. M.; Choi, M. R.; Kim, S. Y.; Chung, Y. K.; Han, W.-S.; Kang, S. W. J.
Org. Chem. 2006, 71, 9366. (d) Nieto-Oberhuber, C.; Mun˜oz, M. P.; Bun˜uel,
E.; Nevado, C.; Ca´rdenas, D. J.; Echavarren, A. M. Angew. Chem., Int. Ed.
2004, 43, 2402. Pt-catalyzed reaction: (e) Me´ndez, M.; Mun˜oz, M. P.;
Nevado, C.; Ca´rdenas, D. J.; Echavarren, A. M. J. Am. Chem. Soc. 2001,
123, 10511. (f) Fu¨rstner, A.; Stelzer, F.; Szillat, H. J. Am. Chem. Soc. 2001,
123, 11863. (g) Fu¨rstner, A.; Szillat, H.; Stelzer, F. J. Am. Chem. Soc. 2000,
122, 6785. (h) Blum, J.; Berr-Kraft, H.; Badrieh, Y. J. Org. Chem. 1995,
60, 5567. (i) Chao, C.-M.; Beltrami, D.; Toullec, P. Y.; Michelet, V. Chem.
Commun. 2009, 6988. (j) Nevado, C.; Ferrer, C.; Echavarren, A. M. Org.
Lett. 2004, 6, 3191.
(7) CCDC 771930 contains the supplementary crystallographic data for
compound 5a. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.Uk/ data_
request/cif.
(8) See the Supporting Information for the detailed reaction optimiza-
tions.
(9) The enyne ether intermediates can be monitored by the TLC method
in the reactions in Table 2.
Org. Lett., Vol. 12, No. 15, 2010
3469

Table 2. Gold-Catalyzed Intermolecular Reaction of Allylic
Acetates 1 with Nonterminal Propargylic Alcohols 2a,b
Table 3. Gold-Catalyzed Intermolecular Reaction of
(E)-1,5-Diphenyl Pent-2-enyl Acetate 1a and Prop-2-yn-1-ol 2ka
yield of yield of
ratio (1a/2k) catalysts/equiv time (min) 7a (%)b 8a (%)b
1
1/1.5
1/10
Au(PPh3)NTf2/5%
Au(PPh3)NTf2/5%
10
10
49
7
2c
trace
88
a Unless noted, all reactions were carried out at 0.1 mmol scale in 3
mL of CH2Cl2 at 25 °C. b Isolated yields. c Reaction temperature is 30 °C.
rearrangement product 7a in 49% yield,6a,d along with a trace
amount of 8a (Table 3, entry 2). A “concentration effect”
was observed in the reaction of 1a with 2k.8 As shown in
Table 3, optimization of the reaction temperature and the
1a/2k ratio provide a set of best conditions to give 8a
exclusively in 88% yield (Table 3, entry 8).
As shown in Table 4, a series of alkoxycyclization adducts
8a-e were then prepared in good yields by using the
Table 4. Gold-Catalyzed Intermolecular Reaction of Allylic
Acetates 1 and Prop-2-yn-1-ol 2ka
a Unless noted, all reactions were carried out at 0.1 mmol scale in 3
mL of CH2Cl2 at 25 °C with 5 mol % equiv of Ph3PAuNTf2 as catalyst
(1/2 ) 1/6). b Unless noted, all products were obtained as the single
diastereomers. c Isolated yields. d The dr value was determined by separated
yields.
a All reactions were carried out at 0.1 mmol scale in 3 mL of CH2Cl2
at 30 °C with 5 mol % equiv of Ph3PAuNTf2 as the catalyst (1/2k ) 1/10).
b All products were obtained as the single diastereomers. c Isolated yields.
all products were obtained as the single diastereomers
(5a-k). When “R5” is an alkyl group, the reaction yields
(5m-o) were somewhat lower than that of their primary
alcohol analogues. But their dr values were still very high.
Only 5o was obtained as a mixture of two diastereomers (dr
∼ 30/1). No diastereomers were detected in the reaction of
1a with 2g and 2h. Moreover, the reaction of 1a with but-
2-yne-1,4-diol 2j was also tested, which afforded a ring-
opening product 6 in 71% yield (Table 2, entry 15).10
We then turn to investigating the reaction of terminal
alkynyl alcohols (Table 3). The reaction of 1a (1 equiv) with
3-propynyl alcohol 2k (1.5 equiv) afforded endocyclic
condition from entry 8 in Table 3.6b,d,11 The reaction of 1c
with 2k give the highest reaction yield (Table 4, entry 3).
The reaction yields for substrates 1d and 1e were relatively
low (Table 4, entries 4 and 5).
A similar “concentration effect” could not be observed in
the reaction of dialkyl allylic acetate 1g with 2k (Scheme 1,
eq 1) or in the reaction of 1a with substituted termial alkynyl
alcohol 2m (Scheme 1, eq 2).11 The former reaction gave
tetrahydropyran derivative 7b exclusively in 83% yield, while
the reaction of 1a with 2m (Scheme 1, eq 2) give 7c in 45%
(10) The stereochemistry of compounds 6, 7c, and 10 were determined
by their NOESY NMR spectral data.
(11) The enyne ether intermediates cannot be detected by TLC monitor-
ing in these reactions.
3470
Org. Lett., Vol. 12, No. 15, 2010

Scheme 1
.
Gold-Catalyzed Intermolecular Reaction of Allylic
Scheme 3
.
A Plausible Mechanism for the Synthesis of
Acetate 1 with Terminal Propargylic Alcohol 2
Compounds 5, 7, and 8
yield,10,12 along with a mixture of some inseparatable
products.
The derivation of compound 8a was then performed to
determine the relative stereochemistry of products 8a-e. As
shown in Scheme 2, hydrogenation of 8a over Lindlar’s
intermolecular allylation,5a underwent subsequent cycloi-
somerization to yield a series of oxygen heterocycles. In the
second step, trapping the gold activated triple bond by the
alkene group was favored to proceed though transition state
A, in which allylic R1 group pointed away from the alkyne
bond to avoid the steric hindrance between R1 and gold
complexes (Scheme 3). The stereochemistry of bicyclo[4.1.0]-
heptylidene gold(I) carbene C (route I) and bicyclo[3.1.0]-
hexylidene gold carbene D (route II) was then determined
by the favored transition state A. Product 5 was obtained
from intermediate C by ꢀ-hydrogen elimination. Trapping
intermediate D1 with 3-propynyl alcohol 2k afforded 8, while
intermediate D rearranged with ring-opening to provide E,
which then gave 7 via ꢀ-elimination.6a
Scheme 2. Derivation of 8a and Investigating Substitution
Effect on Cyclopropane’s Stereochemistry
In summary, we have developed an efficient new method
to construct the densely functionalized oxygen hetereocycles
with high diastereoselectivities from the intermolecular
reaction of allylic acetates with propargylic alcohols via gold
catalysis, in which gold promoted two mechanistically
distinct processes, including intermolecular allylation and
intramolecular enyne cycloisomerization. Different substitu-
tion patterns affected the reaction’s regioselectivity and
cyclopropane unit’s stereochemistry. It was found that
cyclopropane stereochemistry was mainly determined by the
allylic substituents. The effort to extend this reaction and
broaden its application is still underway in this laboratory.
catalyst followed by alkene metathesis under the second
generation Grubbs catalyst provided a bicyclo product 10 in
82% yield (Scheme 2, eq 1).10,13
To examine the effect of different substitution patterns on
stereochemistry, chiral substrate (R)-11 (93% ee) was
prepared and treated with Au(PPh3)NTf2 in DCM, which
gave 5a in 93% ee (Scheme 2, eq 2). However, when
compound 12 (dr ) 0%, ee ) 92% for one diastereomer)
was examined in the same condition,14 only racemic 5o was
obtained in 74% yield (Scheme 2, eq 3). These results
indicated that the stereochemistry of cyclopropane was
mainly determined by the allylic substituents (R1).
Acknowledgment. Support of this work by the grant from
National Sciences Foundation of China (Nos. 20872176) is
gratefully acknowledged.
A plausible mechanism was then proposed. As shown in
Scheme 3, enyne ether intermediates 4, in situ generated from
(12) The reason for the formation of one diastereomer in the reaction
of 1a with 2m is still unclear.
Supporting Information Available: Experimental pro-
cedures and data for all new compounds. This material is
available free of charge via the Internet at http://pubs.acs.org.
(13) (a) Vougioukalakis, G. C.; Grubbs, R. H. Chem. ReV. 2010, 110,
1746. (b) Samojblowicz, G. C.; Bieniek, M.; Grela, K. Chem. ReV. 2009,
109, 3708
.
(14) 12 was obtained as a mixture of two diastereomers. The ee value
of one diastereomer was determined to be 92%.
OL1012923
Org. Lett., Vol. 12, No. 15, 2010
3471