Gold-catalyzed cascade cyclization-oxidative alkynylation of allenoates

Article abstract of DOI:10.1021/ol102061k

A gold(I)-catalyzed cascade cyclization-oxidative cross-coupling process has been applied to prepare β-alkynyl-γ-butenolides directly from allenoates and various terminal alkynes. Following an initial gold-catalyzed C-O bond forming allenoate cyclization, a mechanism based on a Au ^I/Au^III redox cycle has been proposed with Selectfluor acting as the external oxidant.

Full text of DOI:10.1021/ol102061k

ORGANIC
LETTERS
2010
Vol. 12, No. 21
4904-4907
Gold-Catalyzed Cascade
Cyclization-Oxidative Alkynylation of
Allenoates
Matthew N. Hopkinson,† Jonathan E. Ross,† Guy T. Giuffredi,† Antony D. Gee,‡
and Ve´ronique Gouverneur*,†
Department of Chemistry, UniVersity of Oxford, Chemistry Research Laboratory,
Mansfield Road, Oxford OX13TA, U.K., and GSK Clinical Imaging Centre, Imperial
College London, Hammersmith Hospital, Du Cane Road, London W120NN, U.K.
Veronique.gouVerneur@chem.ox.ac.uk
Received August 30, 2010
ABSTRACT
A gold(I)-catalyzed cascade cyclization-oxidative cross-coupling process has been applied to prepare ꢀ-alkynyl-γ-butenolides directly from
allenoates and various terminal alkynes. Following an initial gold-catalyzed C-O bond forming allenoate cyclization, a mechanism based on
a AuI/AuIII redox cycle has been proposed with Selectfluor acting as the external oxidant.
The homogeneous catalysis of organic reactions by gold has
received significant attention in recent years.1 In contrast to other
late transition metals, gold rarely changes oxidation state during
the course of a reaction and instead most commonly acts as a
tunable soft π-acid, activating multiple bonds toward nucleo-
philic attack. In the past few years, however, several gold-
catalyzed homo- and cross-coupling reactions proceeding via
proposed AuI/AuIII redox cycles have been reported.2 In many
cases, the key oxidation of gold(I) to gold(III) is performed by
an external oxidant precluding the need for preactivation of the
starting materials.3 In 2009 Zhang et al. reported the cascade
rearrangement-oxidative homocoupling of propargyl acetates
catalyzed by gold(I) complexes in the presence of the electro-
philic fluorinating reagent Selectfluor.4,5 This methodology has
(2) For a recent review, see: (a) Garcia, P.; Malacria, M.; Aubert, C.;
Gandon, V.; Fensterbank, L. ChemCatChem 2010, 2, 493. For selected
examples of gold-catalyzed cross-coupling with substrates as oxidants, see:
(b) Gonza´lez-Arellano, C.; Corma, A.; Iglesias, M.; Sa´nchez, F. J. Catal.
2006, 238, 497. (c) Gonza´lez-Arellano, C.; Abad, A.; Corma, A.; Garc´ıa,
H.; Iglesias, M.; Sa´nchez, F. Angew. Chem., Int. Ed. 2007, 46, 1536. (d)
Li, P.; Wang, L.; Wang, M.; You, F. Eur. J. Org. Chem. 2008, 5946. (e)
Brand, J. P.; Charpentier, J.; Waser, J. Angew. Chem., Int. Ed. 2009, 48,
9346. For a mechanistic study on the unlikeliness of gold-catalyzed
Sonogashira reactions, see: (f) Lauterbach, T.; Livendahl, M.; Rosello´n,
A.; Espinet, P.; Echavarren, A. M. Org. Lett. 2010, 12, 3006.
† University of Oxford.
‡ GSK Clinical Imaging Centre.
(1) For selected reviews on gold catalysis, see: (a) Hoffmann-Ro¨der,
A.; Krause, N. Org. Biomol. Chem. 2005, 3, 387. (b) Hashmi, A. S. K.
Angew. Chem., Int. Ed. 2005, 44, 6990. (c) Hashmi, A. S. K.; Hutchings,
G. Angew. Chem., Int. Ed. 2006, 45, 7896. (d) Hashmi, A. S. K. Chem.
ReV. 2007, 107, 3180. (e) Fu¨rstner, A.; Davies, P. W. Angew. Chem., Int.
Ed. 2007, 46, 3410. (f) Gorin, D. J.; Toste, F. D. Nature 2007, 446, 395.
(g) Muzart, J. Tetrahedron 2008, 64, 5815. (h) Li, Z.; Brouwer, C.; He, C.
Chem. ReV. 2008, 108, 3239. (i) Arcadi, A. Chem. ReV. 2008, 108, 3266.
(j) Jime´nez-Nu´n˜ez, E.; Echavarren, A. M. Chem. ReV. 2008, 108, 3326. (k)
Gorin, D. J.; Sherry, B. D.; Toste, F. D. Chem. ReV. 2008, 108, 3351. (l)
Skouta, R.; Li, C.-J. Tetrahedron 2008, 64, 4917. (m) Widenhoefer, R.
Chem.sEur. J. 2008, 14, 5382. (n) Fu¨rstner, A. Chem. Soc. ReV. 2009, 38,
3208. (o) Shapiro, N. D.; Toste, F. D. Synlett 2010, 675. (p) Sengupta, S.;
Shi, X. ChemCatChem 2010, 2, 609.
(3) (a) Kar, A.; Mangu, N.; Kaiser, H. M.; Beller, M.; Tse., M. K. Chem.
Commun. 2008, 386. (b) Kar, A.; Mangu, N.; Kaiser, H. M.; Tse, M. K. J.
Organomet. Chem. 2009, 694, 524. (c) Wegner, H. A.; Ahles, S.; Neuburger,
M. Chem.sEur. J. 2008, 14, 11310. (d) Iglesias, A.; Mun˜iz, K. Chem.sEur.
J. 2009, 15, 10563. (e) de Haro, T.; Nevado, C. J. Am. Chem. Soc. 2010,
132, 1512. For a recent review, see: (f) Wegner, H. A. Chimia 2009, 63,
44.
10.1021/ol102061k  2010 American Chemical Society
Published on Web 10/11/2010

also been successfully applied to oxidative cross-coupling
with benzoates6 and preactivated arylboronic acids.7 In 2008,
we disclosed that oxidative fluorination of an organogold
intermediate is feasible using Selectfluor.8 More recently,
we showed that benzyl-substituted tert-butyl allenoates are
amenable to cascade cyclization-oxidative intramolecular
arylation using the same oxidant.9,10 This process, involving
direct aryl C-H functionalization, led to the facile synthesis
of indenofuranones from substrates not requiring preactiva-
tion. We sought to investigate whether this methodology
could be applied to effect a cascade cyclization-intermolecular
alkynylation delivering ꢀ-alkynyl-γ-butenolides directly from
allenoates and unfunctionalized alkynes. Currently, these
compounds are accessed via a two-step protocol where the
allene cyclization and subsequent alkynylation are performed
separately.11,12 This novel transformation, relying on gold
catalysis only, combines the well-established reactivity of
gold with an oxidative cross-coupling event (Scheme 1).
Table 1. Optimization Studies for Oxidative Alkynylation of 1a
entry
catalyst
oxidant
base time
4 d
yielda
44%
1
2
3
4
5
6
7
8
Ph3PAuNTf2
Ph3PAuNTf2
no catalyst
AuClb
AgOTf
PtCl2
Selectfluor none
Selectfluor K3PO4 4 h
94%
Selectfluor K3PO4 10 d NR
Selectfluor K3PO4 48 h NR
Selectfluor K3PO4 10 d NR
Selectfluor K3PO4 10 d NR
Selectfluor K3PO4 10 d NR
Selectfluor K3PO4 10 d NR
CuOAc
H2SO4
9
SIPrAuCl/AgOTfc Selectfluor K3PO4 10 d NR
Pd(OAc)2/ CuOAc Selectfluor K3PO4 24 h decomp.
10
11
12
13
14
15
16
17
AuCl3
Selectfluor K3PO4 6 d
Selectfluor K3PO4 24 h 72%
no oxidant K3PO4 5 d
PhI(OAc)2 K3PO4 7 d
22%
b
Ph3PAuNTf2
Ph3PAuNTf2
Ph3PAuNTf2
Ph3PAuNTf2
Ph3PAuNTf2
Ph3PAuNTf2
NR
NR
NR
NR
54%f
Scheme 1. Cascade Cyclization-Oxidative Alkynylation
tBuOOH
Oxoned
NFSIe
K3PO4 7 d
K3PO4 7 d
K3PO4 7 d
a Isolated yield. b 5 mol %. c SIPr ) 1,3-Bis(2,6-diisopropylphenyl)
imidazolin-2-ylidene. d Oxone ) KHSO5·1/2KHSO4·1/2K2SO4. e NFSI )
f
1
N-Fluorobenzenesulfonimide. Conversion estimated by H NMR.
As a preliminary experiment, the alkyl-substituted tert-butyl
allenoate 1a was treated with phenylacetylene 2a (1.5 equiv),
Selectfluor (2.5 equiv), and Ph3PAuNTf2 (10 mol %)13 in
acetonitrile (0.15 M) and water (10 equiv) at room tempera-
ture.14 Pleasingly, the desired ꢀ-alkynyl-γ-butenolide 3aa was
isolated in 44% yield after 4 days (Table 1, entry 1). Notably,
no products resulting from cyclization-protodeauration of the
allenoate or homocoupling of either 1a or 2a were observed.15
The yield of 3aa was increased to 94%, and the reaction time
shortened to 4 h upon addition of potassium phosphate tribasic
(2 equiv, entry 2). After extensive optimization studies,16
Ph3PAuNTf2 was identified as the catalyst of choice for this
transformation, while AuCl, AgOTf, PtCl2, CuOAc, H2SO4, and
SIPrAuCl/AgOTf all led to recovered allenoate (entries 3-9).
The combination of Pd(OAc)2/CuOAc, a common catalytic
system for Sonogashira coupling, led to decomposition of the
starting materials (entry 10). AuCl3 was a suitable catalyst for
the cascade cyclization-oxidative alkynylation process but was
significantly less efficient, delivering 3aa in only 22% yield
after 6 days of reaction (entry 11). Alternative oxidants such
as PhI(OAc)2, tBuOOH, and Oxone led to no reaction with
complete recovery of 1a (entries 13-16). N-Fluorobenzene-
sulfonimide (NFSI), a mild electrophilic fluorinating reagent,
did lead to 3aa but with low conversion after an extended
reaction time (54% conversion after 7 days, entry 17).
With optimized reaction conditions in hand, the effect of the
alkyne substitution on the reaction efficiency was investigated
(Scheme 2). The transformation was compatible with a wide
range of arylacetylenes including para-, meta-, and ortho-
substituted derivatives. Alkyne 2e, bearing an electron-donating
para-OMe group on the benzene ring, reacted readily, affording
3ae in 88% yield. Electron-neutral and electron-poor arylacetyl-
enes were also tolerated, delivering the cross-coupled butenolide
products 3af-ah bearing para-F, para-CF3, and even para-
(4) Cui, L.; Zhang, G.; Zhang, L. Bioorg. Med. Chem. Lett. 2009, 19, 3884.
(5) For a report on oxidative homocoupling of stoichiometric gold(I)
complexes with N-fluorobenzenesulfonimide (NFSI), see: Hashmi, A. S. K.;
Ramamurthi, T. D.; Rominger, F. J. Organomet. Chem. 2009, 694, 592.
(6) Peng, Y.; Cui, L.; Zhang, G.; Zhang, L. J. Am. Chem. Soc. 2009, 131,
5062.
(7) (a) Zhang, G.; Peng, Y.; Cui, L.; Zhang, L. Angew. Chem., Int. Ed.
2009, 48, 3112. (b) Zhang, G.; Cui, L.; Wang, Y.; Zhang, L. J. Am. Chem.
Soc. 2010, 132, 1474. (c) Melhado, A. D.; Brenzovich, W. E., Jr.; Lackner,
A. D.; Toste, F. D. J. Am. Chem. Soc. 2010, 132, 8885. (d) Brenzovich,
W. E., Jr.; Benitez, D.; Lackner, A. D.; Shunatona, H. P.; Tkatchouk, E.;
Goddard, W. A., III.; Toste, F. D. Angew. Chem., Int. Ed. 2010, 49, 5519.
(8) (a) Schuler, M.; Silva, F.; Bobbio, C.; Tessier, A.; Gouverneur, V.
Angew. Chem., Int. Ed. 2008, 47, 7927. For more recent examples of gold-
catalyzed fluorination reactions, see: (b) de Haro, T.; Nevado, C. Chem.
Commun. 2010, doi: 10.1039/C002679D. (c) Hopkinson, M. N.; Giuffredi,
G. T.; Gee, A. D.; Gouverneur, V. Synlett 2010, doi: 10.1055/s-0030-
1258992.
(9) Hopkinson, M. N.; Tessier, A.; Salisbury, A.; Giuffredi, G. T.;
Combettes, L. E.; Gee, A. D.; Gouverneur, V. Chem.sEur. J. 2010, 16, 4739
.
(10) For examples of palladium-catalyzed cross-coupling reactions of
stoichiometric (butenolide)gold(I) complexes, see: (a) Shi, Y.; Ramgren,
S. D.; Blum, S. A. Organometallics 2009, 28, 1275. (b) Shi, Y.; Roth, K. E.;
Ramgren, S. D.; Blum, S. A. J. Am. Chem. Soc. 2009, 131, 18022
(11) Ma, S.; Shi, Z.; Yu, Z. Tetrahedron Lett. 1999, 40, 2393
(12) Hashmi, A. S. K.; Do¨pp, R.; Lothschu¨tz, C.; Rudolph, M.; Riedel,
.
(15) By contrast, treating 1a with Ph3PAuNTf2 (10 mol %) in dichlo-
romethane afforded 5-butyl-3-methylfuran-2(5H)-one (8a) resulting from
cyclization-protodeauration as the only product in 76% yield after 24 h. In
acetonitrile and water (10 equiv), 8a was produced in 39% yield after 6 days.
See also ref 9.
.
D.; Rominger, F. AdV. Synth. Catal. 2010, 352, 1307
.
(13) Me´zailles, N.; Ricard, L.; Gagosz, F. Org. Lett. 2005, 7, 4133.
(14) The addition of water (10 equiv) aids the solubility of the reagents.
(16) For details, see the Supporting Information.
Org. Lett., Vol. 12, No. 21, 2010
4905

affording cross-coupled products 3ba and 3ca in moderate
yields (entries 2 and 3). In these reactions, diyne 4a resulting
from oxidative homocoupling of the alkyne was also isolated
in 29% and 23% yield, respectively, relative to 2a. Phenyl-
substituted allenoate 1d failed to react and led to complete
recovery of the allenoate after 48 h (entry 4).
Scheme 2
.
Oxidative Alkynylation of 1a with 2a-j Isolated
Yields
The transformation was successful with ethyl allenoate 1e,
delivering 3aa in 29% yield (entry 5). The drop in efficiency
with this substrate in comparison with 1a can be attributed to
a less favorable cyclization step involving the loss of an ethyl
cation. In this reaction, competitive alkyne homocoupling was
the major reaction pathway. The benzyl-substituted allenoates
1f and 1g led to mixtures of products resulting from both
intermolecular oxidative alkynylation (3fa and 3ga) and in-
tramolecular oxidative arylation (5 and 6) with the alkynylation
process slightly favored (Scheme 3).9 Butenolides 3fa and 5 were
Scheme 3. Competitive Alkynylation and Arylation of 1f and 1g
a Reaction performed with NaOH (2 equiv) as base. b Reaction performed
with CuOAc (10 mol %) cocatalyst at 80 °C.
NO2 substituents in moderate to good yields. While the
trimethylsilyl-protected alkyne 2i was amenable to the cascade
cyclization-cross-coupling process, the isolated yield of the
corresponding butenolide 3ai, resulting from subsequent depro-
tection of the silyl group, was low (16%). The reaction with
1-pentyne 2j was similarly low yielding, delivering 3aj in only
14% yield. Following screening of conditions for this substrate,
the isolated yield could be improved to 28% upon heating to
80 °C in the presence of added CuOAc (10 mol %) cocatalyst.
To probe the effect of the allenoate substitution, substrates
1a-g were synthesized according to literature procedures9,17
and reacted with phenylacetylene 2a (Table 2). The allyl-
substituted allenoates 1b and 1c cyclized successfully,
crystalline solids and allowed for the unambiguous assignment of
the structures by X-ray crystallography (Figure 1).16,18
Table 2. Oxidative Alkynylation of 1a-e with 2a
Figure 1. X-ray crystal structure of 3fa.
Mechanistically, we envisage two plausible pathways for the
cascade allenoate cyclization-oxidative alkynylation process, both
involving a AuI/AuIII redox cycle.19 In the presence of base, initial
deprotonation of the alkyne and subsequent coordination to the
cationic gold(I) catalyst could lead to the alkynylgold(I) intermedi-
(17) Kang, J.-E.; Lee, E.-S.; Park, S.-I.; Shin, S. Tetrahedron Lett. 2005,
46, 7431.
(18) CCDC-795566 and CCDC-795567 contain the supplementary
crystallographic data for 3fa and 5, respectively. These data can be obtained
free of charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
a Isolated yield. b Isolated yield relative to 2a.
4906
Org. Lett., Vol. 12, No. 21, 2010

ate A. Oxidation at the gold center by Selectfluor would afford
the square planar gold(III) cationic intermediate B. A similar
gold(III) fluoride complex was recently observed by Toste et al.
upon treating gold(I) methyl complexes with the electrophilic
fluorinating reagent XeF2.20 Coordination of this species to the
allene followed by nucleophilic attack of the pendant tert-butyl
ester would lead to the gold(III) complex C bearing both butenolide
and alkyne substituents. After reductive elimination, this species
delivers the cross-coupled product and regenerates the cationic
gold(I) catalyst (Scheme 4, path I).3e Alternatively, the allenoate
Scheme 5. Oxidative Homocoupling of 1a and 2a
Scheme 4. Plausible Reaction Mechanism
explain the observed reactivity. The long reaction time and low
yield of this reaction in comparison with that above suggests that
path I could be more favored. This observation is consistent with
the scope and limitation studies where diynes were observed as
minor side products of the reactions. To support a mechanism
involving the intermediates A and D, these complexes were
synthesized according to literature procedures22,23 and subjected
to the reaction conditions with 1a and 2a, respectively. After
1 h at rt, complete consumption of each complex was
observed with the formation of 3aa, 4a, and 7a in both
reactions, suggesting that both path I and path II are feasible.
Additionally, if ligand exchange between the gold complexes is
sufficiently fast under the reaction conditions, each product 3, 4,
or 7 could be formed regardless of the initial coordination and
oxidation steps.24 The observed product distribution, in this case,
would be a function of the rate of ligand exchange and reductive
elimination at each diorganogold(III) complex.25
In conclusion, a novel, gold-catalyzed cascade cyclization-
oxidative cross-coupling process has been used to synthesize
ꢀ-alkynyl-γ-butenolides directly from starting materials not
requiring preactivation. This process provides a basis for the
development of novel cascade reactions combining traditional
gold catalysis and intermolecular oxidative alkynylation.
cyclization could occur prior to the alkyne coordination step. In
this case, oxidation of the organogold(I) intermediate D by
Selectfluor would lead to the gold(III) complex E. Reductive
elimination from intermediate C′, formed after alkynylation, would
afford the product and regenerate the catalyst (Scheme 4, path II).21
To investigate the proposed mechanisms, several control reac-
tions were performed. Treatment of 2a under the optimized reaction
conditions but in the absence of 1a afforded the product of oxidative
homocoupling 4a, in quantitative yield after 4 h (Scheme 5). The
formation of this product is consistent with path I, invoking a
second alkynylation to intermediate B followed by reductive
elimination. Allenoate 1a afforded the diastereomeric bibutenolides
7a, resulting from oxidative homocoupling, in 35% yield after 5
days when reacted in the absence of 2a (Scheme 5). Again,
coordination and cyclization of a second molecule of the allenoate
to intermediate E (path II) followed by reductive elimination can
Acknowledgment. We thank GlaxoSmithKline and the
EPSRC for financial support, Dr. B. Odell (University of
Oxford) for NMR studies, and Dr. A. L. Thompson (Oxford
Chemical Crystallography Service) for crystallographic services.
Supporting Information Available: Experimental pro-
cedures and characterization data for all compounds, opti-
mization table, and X-ray crystal data for 3fa and 5. This
material is available free of charge via the Internet at
http://pubs.acs.org.
OL102061K
(21) 5-Butyl-3-methylfuran-2(5H)-one (8a) was unreactive under the
optimized reaction conditions ruling out a mechanism involving the
protodeaurated product.
(22) Hashmi, A. S. K.; Lothschu¨tz, C.; Do¨pp, R.; Rudolph, M.;
Ramamurthi, T. D.; Rominger, F. Angew. Chem., Int. Ed. 2009, 48, 8243
(23) Liu, L.-P.; Xu, B.; Mashuta, M. S.; Hammond, G. B. J. Am. Chem.
Soc. 2008, 130, 17642
(24) The formation of diyne 4a from A in the absence of alkyne 2a and
the formation of bibutenolides 7a from D in the absence of allenoate 1a
are best explained by such intermolecular ligand exchange processes.
(25) Tamaki, A.; Kochi, J. K. J. Organomet. Chem. 1974, 64, 411.
.
(19) While a mechanism involving a ꢀ-fluoro-γ-butenolide intermediate
cannot be unequivocally ruled out for this transformation, such a pathway
was ruled out in the related cascade allenoate cyclization-intramolecular
arylation reaction (see ref 9). In addition, no fluorinated organic species
were isolated from any reaction mixture in the course of our studies.
(20) Mankad, N. P.; Toste, F. D. J. Am. Chem. Soc. 2010, 132, 12859.
.
Org. Lett., Vol. 12, No. 21, 2010
4907