Gold(I)-catalyzed rearrangement of alkynloxiranes: A mild access to divinyl ketones

Article abstract of DOI:10.1021/ol800219k

Acyloxylated divinyl ketones are conveniently formed by a new gold(l)-catalyzed rearrangement of (3-acyloxyprop-1-ynyl)oxiranes.

Full text of DOI:10.1021/ol800219k

ORGANIC
LETTERS
2008
Vol. 10, No. 8
1569-1572
Gold(I)-Catalyzed Rearrangement of
Alkynyloxiranes: A Mild Access to
Divinyl Ketones
Marie-Caroline Cordonnier, Aure´lien Blanc, and Patrick Pale*
Laboratoire de Synthe`se et Re´actiVite´ Organique, associe´ au CNRS, Institut de Chimie,
UniVersite´ Louis Pasteur, 4 rue Blaise Pascal, 67000 Strasbourg, France
ppale@chimie.u-strasbg.fr
Received January 30, 2008
ABSTRACT
Acyloxylated divinyl ketones are conveniently formed by a new gold(I)-catalyzed rearrangement of (3-acyloxyprop-1-ynyl)oxiranes.
Gold catalysis is increasingly gaining interest in organic
chemistry1 due to the efficiency, the mildness, and the
peculiar properties associated with gold atoms.2 Among these
properties, alkynophilicity of gold salts or complexes3 already
led to numerous applications, usually based on nucleophilic
addition to an intermediate alkyne-gold complex. In con-
trast, oxophilicity of gold species was far less investigated,
and only a few examples were reported in the literature
mainly based on Lewis acid activation of carbonyl com-
pounds.1,4 We thus wondered if such properties could be
combined, leading to new chemistry promoted by gold salts
or complexes.
with R- and â-alkynyl epoxides. Such compounds are known
to rearrange into furans in the presence of gold trichloride
with modest to excellent yields (Scheme 1).5-7 With alkyno-
Scheme 1. Known Rearrangements of Alkynyloxiranes to
Furans and Proposed Intermediate for These Reactions
Having an acetylenic moiety and a poorly nucleophilic
oxygen atom within the same molecule could be achieved
(1) For recent reviews on gold chemistry, see: (a) Hashmi, A. S. K.
Chem. ReV. 2007, 107, 3180-3211. (b) Jime´nez-Nu´n˜ez, E.; Echavarren,
A. M. Chem. Commun. 2007, 333-346. (c) Hashmi, A. S. K.; Hutchings,
G. J. Angew. Chem., Int. Ed. 2006, 45, 7896-7936.
(2) (a) Gorin, D. J.; Toste, D. Nature 2007, 446, 395-403. (b) Pyyko¨,
P. Angew. Chem., Int. Ed. 2004, 43, 4412-4456. (c) Bagus, P. S.; Lee, Y.
S.; Pitzer, K. S. Chem. Phys. Lett. 1975, 33, 408-411.
(3) Hashmi, A. S. K. Gold Bull. 2003, 36, 3-9.
(4) (a) Ito, Y.; Sawamura, M.; Hayashi, T. J. Am. Chem. Soc. 1986, 108,
6405-6406. (b) Teles, J. H.; Brode, S.; Chabanas, M. Angew. Chem., Int.
Ed. 1998, 37, 1415-1418. (c) Georgy, M.; Boucard, V.; Campagne, J.-M.
J. Am. Chem. Soc. 2005, 127, 14180-14181. (d) Jung, H. H.; Floreancig,
P. E. Org. Lett. 2006, 8, 1949-1951. (e) Jung, H. H.; Floreancig, P. E. J.
Org. Chem. 2007, 72, 7359-7366. (f) Lin, C.-C.; Teng, T.-M.; Odedra,
A.; Liu, R.-S. J. Am. Chem. Soc. 2007, 129, 3798-3799. (g) Yamamoto,
Y. J. Org. Chem. 2007, 72, 7817-7831.
philicity in mind, oxonium ions resulting from internal
nucleophilic attack of the oxirane oxygen atom on acetylene-
(5) Hashmi, A. S. K.; Pradipta, S. AdV. Synth. Catal. 2004, 346, 432-
438.
(6) Shu, X.-Z.; Liu, X.-Y.; Xiao, H.-Q.; Ji, K.-G.; Guo, L.-N.; Qi,
C.-Z.; Liang, Y.-M. AdV. Synth. Catal. 2007, 349, 2493-2498.
10.1021/ol800219k CCC: $40.75
© 2008 American Chemical Society
Published on Web 03/22/2008

gold π-complex were proposed as intermediate in these furan
formations. However, a double activation of the acetylene
and the oxygen atom could also be responsible for this
transformation, with allenic species as intermediates.
On these bases, we thus investigated the behavior of
various alkynyl oxiranes, trying to reveal the presence or
not of allenic intermediates as a first objective and trying to
take benefit of them by trapping as a second objective. To
do so, we envisaged placing a nucleophile within the
molecule at the right position to react with any intermediate
allene, which could be formed by double activation of the
acetylene and the oxirane oxygen atom (Scheme 2).8 Due to
coupling reaction. The resulting acyloxylated enynes were
then epoxidized upon treatment with m-chloroperbenzoic acid
(m-CPBA). The overall yields of this two- to three-step
sequence were routinely higher than 50%.
In order to find the more appropriate conditions for the
reaction of such substrates, we applied various conditions
and gold catalysts to the simplest acetoxylated alkynyl
oxirane 1a (Table 1). A rapid catalyst screening revealed
Table 1. Screening of Reaction Conditions for the
Transformation of Acetoxylated Alkynyl Oxirane 1a
Scheme 2. Hypothesis for Au-Catalyzed Conversion of
Alkynyloxiranes to Heterocycles or Rearranged Systems
time yieldb
entry
catalyst (mol %)
AuCl (5)
PPh3AuCl (5)
AuCl3 (5)
PPh3AuCl/AgSbF6 (5) dry CH2Cl2, rt
AgSbF6 (5)
AgCl (5)
conditionsa
(h)
(%)
1
2
3
4
5
6
dry CH2Cl2, rt 16
dry CH2Cl2, rt 16
dry CH2Cl2, rt 16
c
c
14
88d
49
c
0.1
0.1
dry CH2Cl2, rt
dry CH2Cl2, rt 16
a Reactions run under argon, c ) 0.1 mol/L. b Yields of 2a was calculated
by 1H NMR relative to an internal standard (hexamethylbenzene). c No
conversion. d 70% of isolated yield.
precedents in the literature involving propargyl acetate and
related esters,9 we selected an acyloxy group at the pro-
pargylic position as the first nucleophile investigated. We
report here that such functionalized alkynyl oxiranes, i.e.,
(3-acyloxyprop-1-ynyl)oxiranes,10 indeed react in the pres-
ence of gold catalyst, leading to functionalized divinyl
ketones in high yields.
Various (3-hydroxyprop-1-ynyl)oxiranes 1a-l were con-
veniently produced by addition of the lithium acetylide
derived from various enynes to aldehydes or ketones. The
in situ formed propargyl alcoholates could be either directly
acylated by further addition of acyl anhydride or chloride
or simply hydrolyzed. In the latter, the isolated alcohol was
then acylated in classical conditions. For the preparation of
1d, the required alcohol was prepared by a Sonogashira
that in situ generated cationic gold(I) species were the most
effective.
Gold chloride, alone or as its triphenylphosphine complex,
did not promote any transformation, whatever the solvent
and the amount of catalyst, and the starting material 1a was
recovered even after prolonged contact time (entries 1 and
2). The more electrophilic gold trichloride proved to be
slightly more effective in dichloromethane, cleanly convert-
ing 1a into a new compound, although in low yield (entry
3). Physical data of the new compound so formed showed a
complete reorganization of the carbon skeleton, with the
concomitant presence of two vinyl motifs and a conjugated
ketone. Therefore, the single compound formed in these con-
ditions corresponded to a rearranged product, the acetoxy-
vinyl vinyl ketone 2a (see the proposed mechanism below).
Switching to the even more electrophilic cationic gold(I)
produced by premixing triphenylphosphinogold chloride and
a silver salt led to a complete conversion and provide the
same compound in high yield (entry 4). Interestingly enough,
control experiments with silver salts revealed that the highly
electrophilic silver hexafluoroantimonate also promoted the
same reaction but to a lesser extent (entry 5).11
(7) Gold(I)-catalyzed cascade reactions involving cyclization of δ-alkynyl
epoxides and further water or alcohol addition were also recently described.
See: Dai, L.-Z.; Qi, M.-J.; Shi, Y.-L.; Liu, X.-G.; Shi, M. Org. Lett. 2007,
9, 3191-3194.
(8) Hashmi has nevertheless shown that the presence of primary alcohol
on the substrate did not affect the formation of furans (Scheme 1, R1
)
-(CH2)4OH).5 Therefore, in this case, the nucleophilic alcohol seems to
be too far to interact with the alkynyloxirane part.
(9) (a) Marion, N.; Nolan, S. P. Angew. Chem., Int. Ed. 2007, 46, 2750-
2752. (b) Zhang, L.; Wang, S. J. Am. Chem. Soc. 2006, 128, 1442-1443.
(c) Buzas, A.; Gagosz, F. J. Am. Chem. Soc. 2006, 127, 12614-14615. (d)
Marion, N.; Nolan, S. P. Angew. Chem., Int. Ed. 2006, 46, 3647-3650. (e)
Shi, X.; Gorin, D. J.; Toste, F. D. J. Am. Chem. Soc. 2005, 127, 5802-
5803. (f) Fu¨rstner, A.; Hannen, P. Chem. Commun. 2004, 2546-2547.
(10) (3-acyloxyprop-1-ynyl)oxiranes are known to rearrange to furans
in the presence of SmI2/Pd(II) via allene intermediates: (a) Aurrecoechea,
J. M.; Pe´rez, E.; Solay, M. J. Org. Chem. 2001, 66, 564-569. (b)
Aurrecoechea, J. M.; Pe´rez, E. Tetrahedron Lett. 2001, 42, 3839-3841.
(c) Aurrecoechea, J. M.; Pe´rez, E. Tetrahedron Lett. 2003, 44, 3263-3266.
In order to study the scope of this new formation of divinyl
ketones, various representative acyloxyalkynyl oxiranes 1b-j
were then submitted to the above conditions (Table 2). The
role of the nucleophilic acyloxy group was examined with a
series of esters 1a-c. The bulky pivalate 1b gave a
significantly higher isolated yield of divinyl ketone 2b than
acetate 1a (entry 2 vs 1). As expected, the benzoate 1c,
1570
Org. Lett., Vol. 10, No. 8, 2008

(entries 7 and 8). Comparatively, the presence of a secondary
acetoxy and an aryl substituent at the oxiranyl end lowered
the stability of the rearranged products and thus the yields
(entry 8 and 1i below). Surprisingly, the derivatives 1i,j, also
obtained as a nonseparable 1:1 mixture of diastereoisomers,
gave the expected acetoxydivinyl ketones but as single
stereoisomers 2i and 2j, respectively.12 The stereochemistry
of these products was clearly established with NOESY
experiments. It is noteworthy that no trace of furans was
observed during theses reactions.13
Table 2. Scope of the Gold-Catalyzed Rearrangement
As the reaction was too fast to detect any intermediate,12
we look at stereoselectivity issue by studying the Au-cata-
lyzed rearrangement of (-)-menthone derivatives.
Nucleophilic additions to (-)-menthone provided two
diastereoisomers, which can usually be separated, the major
one corresponding to equatorial addition.14 Indeed, the
addition of the lithium acetylide derived from 3-methyl-
butenyne provided a 70/30 mixture of diastereoisomers, the
equatorial one being the major one. After separation, the
minor isomer was acetylated15 and then epoxidized, yielding
a nonseparable 1/1 mixture of epoxide diastereoisomers 1k.
Interestingly, this mixture placed in the rearrangement
conditions gave the expected acetoxydivinyl ketones 2k but
not with the same diastereoisomeric ratio (60/40; Scheme
3), suggesting some stereoselectivity in the rearrangement.
Scheme 3. Au-Catalyzed Rearrangement of Alkynyloxiranes
Derived from (-)-Menthone and Isomerization of Its Products
However, control experiments revealed that the mixture of
products slowly evolved toward a 70/30 mixture when
replaced in the reaction conditions. Therefore, the acetoxy-
divinyl ketones so obtained are equilibrating in the reaction
conditions, and in this case, the bulkiness of the menthone
moiety seemed to slow an easy equilibrium between E and
Z stereoisomers. These results clearly showed that the E/Z
ratio observed in the Au-catalyzed formation of acyloxydi-
a Isolated yields.
having a conjugated and thus less nucleophilic carbonyl
group, was clearly less effective since only 32% yield of 2c
were isolated after a longer reaction time (entry 3). Varying
the alkyl substituents at either the oxirane or the acyloxy
groups did not much influence the reaction course and yields
(entries 4-6). The nonsymmetrical derivatives 1g,h obtained
as a nonseparable 1/1 mixture of diastereoisomers gave a
mixture of stereoisomers 2g,h at the acetoxyvinyl moiety
with the same diastereomeric ratio as their progenitors
(12) Rearrangement reaction of 1j run in CD2Cl2 at -70 °C and
1
monitored by H NMR did not show the presence of intermediates of the
Z isomer 2j. However, the faster conversion of one of the two isomers of
1j was observed.
(13) Secondary less activated ester (R1 ) R2 ) -(CH2)4-, R3 ) H, R4
) C5H11) furnished poor yield of divinyl ketone (10 %) but again no trace
of furane was observed even after a long contact time.
(14) (a) Le Goanvic, D.; Holler, M.; Pale, P. Tetrahedron: Asymmetry
2002, 13, 119-121. (b) Xu, Q.; Wang, G.; Pan, X. Chan, A. S. C.
Tetrahedron: Asymmetry 2001, 12, 381-385. (c) Koning, B.; Buter, J.;
Hulst, R.; Stroetinga, R.; Kellogg, R. M. Eur. J. Org. Chem. 2000, 2735-
2743.
(11) Ag salts are known to promote rearrangement of propargyl acetoxy
derivatives: (a) Benn, W. R. J. Org. Chem. 1968, 33, 3113-3118. (b)
Saimoto, H.; Hiyama, T.; Nozaki, H. J. Am. Chem. Soc. 1981, 103, 4975-
4977. (c) Saimoto, H.; Hiyama, T.; Nozaki, H. Bull. Chem. Soc. Jpn. 1983,
56, 3078-3087. (d) Saimoto, H.; Yukari, K.; Hiyama, T. Tetrahedron Lett.
1986, 27, 1607-1610. (e) Saimoto, H.; Yasui, M.; Ohrai, S-i.; Oikawa,
H.; Yokoyama, K.; Shigemasa, Y. Bull. Chem. Soc. Jpn. 1999, 72, 279-
284. (f) Bluthe, N.; Gore, J.; Malacria, M. Tetrahedron 1986, 42, 1333-
1344.
(15) All of the classical methods failed to acetylate the major isomer,
and only n-BuLi/Ac2O gave 1k in good yield. See the Supporting
Information.
Org. Lett., Vol. 10, No. 8, 2008
1571

vinyl ketones only reflected the thermodynamic stability of
the compounds.
The Au-catalyzed rearrangement of (3-acyloxyprop-1-
ynyl)oxiranes to acyloxydivinyl ketones described here
obviously involved a cascade of events (Scheme 4). Gold
of alkynyl aldehydes and ketones,17 cycloreversion of oxete
would then occurred, furnishing the acyloxydivinyl ketones.
Although allenes are probable intermediates in this rear-
rangement, it is difficult to assess stereoselectivity since the
products formed isomerize in the reaction conditions.
In conclusion, we have reported for the first time an
original catalytic approach to functionalized divinyl ketones18
through an Au-catalyzed rearrangement of (3-acyloxyprop-
1-ynyl)oxiranes. The reaction proceeds efficiently under mild
conditions with commercially available catalysts. Further
work is now underway in our laboratory to broaden the scope
of this reaction and better understand its mechanism, as well
as to continue to explore gold catalysts in organic synthesis.
Scheme 4. Proposed Mechanism for Au-Catalyzed
Rearrangement of Acyloxypropargyloxiranes to Acyloxydivinyl
Ketones
Acknowledgment. We thank the CNRS and the
French Ministry of Research for financial support and the
DECOMET laboratory (Institut de Chimie, Strasbourg) for
the use of their glovebox.
Supporting Information Available: Selected experi-
mental procedures and spectral data. This material is available
free of charge via the Internet at http://pubs.acs.org.
OL800219K
(16) (a) Kim, J. T.; Kel’in A. V.; Gevorgyan, V. Angew. Chem., Int. Ed.
2003, 42, 98-101. (b) Schwier, T.; Sromek, A. W.; Yap, D. M. L.;
Chernyak, D.; Gevorgyan, V. J. Am. Chem. Soc. 2007, 129, 9868-9878.
(c) Dudnik, A. S.; Sromek, A. W.; Rubina, M.; Kim, J. T.; Kel’in A. V.;
Gevorgyan, V. J. Am. Chem. Soc. 2008, 130, 1440-1452.
interaction with both the alkynyl and oxirane moieties might
lead to an allenic alcoholate, inducing migration of the
adjacent acyloxy group.8,16 This transformation could be
either a step-by-step or a concerted process (route b or a,
respectively). The allenic alcoholate A or B could then
cyclized to an oxete C, probably through gold assistance.
As already proposed for Ag- or Au-catalyzed cyclizations
(17) (a) Rhee, J. U.; Krische, M. J. Org. Lett. 2005, 7, 2493-2495. (b)
Jin, T.; Yamamoto, Y. Org. Lett. 2007, 9, 5259-5262.
(18) Such type of substrates can potentially lead to the formation of the
corresponding useful cyclopentenones via the Nazarov reaction. For reviews,
see: (a) Pellissier, H. Tetrahedron 2005, 61, 6479-6510. (b) Habermas,
K. L.; Denmark, S. E. Org. React. 1994, 45, 1-158, 11971.
1572
Org. Lett., Vol. 10, No. 8, 2008