Gold catalysis: Evidence for the in-situ reduction of gold(III) during the cyclization of allenyl carbinols

Article abstract of DOI:10.1002/ejoc.200600009

Products of an oxidative coupling were obtained in the gold(III)-catalyzed cycloisomerization of tertiary allenyl carbinols. The absence of reduced organic products and an increase of these coupling products with the amount of gold(III) catalyst suggests

Full text of DOI:10.1002/ejoc.200600009

SHORT COMMUNICATION
DOI: 10.1002/ejoc.200600009
Gold Catalysis: Evidence for the In-situ Reduction of Gold(III) During the
Cyclization of Allenyl Carbinols
A. Stephen K. Hashmi,*[a,b] M. Carmen Blanco,[a] Dirk Fischer,[b] and Jan W. Bats[b]
Keywords: Alkynes / Allenes / Dihydrofurans / Gold / Homogeneous catalysis
Products of an oxidative coupling were obtained in the gol-
d(III)-catalyzed cycloisomerization of tertiary allenyl carbi-
nols. The absence of reduced organic products and an in-
gold(I) catalysts which were also shown to be active for these
transformations.
crease of these coupling products with the amount of gold(III
catalyst suggests that gold(III) is reduced in situ, possibly to
)
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2006)
Introduction
Results and Discussion
We started with the terminal allene 1, which, upon treat-
ment with 5 mol-% AuCl3 in acetonitrile, delivered 2, 3 and
4 as the products (Scheme 1). The structure of 3 was de-
duced by the spectroscopic relationship to 2, and the mol-
ecular weight obtained from the mass spectrum and finally
proven by an X-ray crystal structure analysis (Figure 1).[13]
Compound 4 is the product of a SNЈ-like substitution of the
hydroxy group by chloride.
One central question of gold catalysis is the oxidation
state of the catalytically active species.[1] This is true for
homogeneous[2] as well as heterogeneous catalysis.[3] For ex-
ample the addition of oxygen nucleophiles like water or
alcohol was first achieved by Utimoto et al.[4] with AuIII
catalysts, later Teles et al.[5] prove AuI catalysts to be very
efficient, a similar behavior was observed by Utimoto et
al.[6] and Lok et al.[7] for the hydroamination reaction. So
far, neither an in situ detection of a catalytically active gold
species was reported nor did comperative theoretical studies
provide clear insight. Recent theoretical work of Straub[8]
on the gold-catalyzed naphthalene synthesis of Yamamoto
et al.[9] and theoretical studies accompanying the work of
Periana et al.[10] on the C,H-activation of methane also
showed that for both AuI and AuIII the energies of acti-
vation calculated would allow an efficient conversion and
suggested that even the ratio of both species might deter-
mine, which of the two species is responsible for the major
amount of the product formed.
In work following our gold-catalyzed cycloisomerization/
dimerization of allenyl ketones,[11] we now investigated the
cycloisomerization of tertiary allenyl carbinols. Similar re-
actions of allenyl carbinols have been reported by Krause
et al. earlier,[12] but here for the first time we observe side-
products, which provide evidence for a mechanism of an in-
situ reduction of gold().
Scheme 1. Formation of 3 by a gold() catalyst.
[a] Institut für Organische Chemie, Universität Stuttgart,
Pfaffenwaldring 55, 70569 Stuttgart, Germany
Fax: +49-711-685-4321
E-mail: hashmi@hashmi.de
[b] Institut für Organische Chemie und Chemische Biologie,
Johann Wolfgang Goethe-Universität,
Marie-Curie-Str. 11, 60439 Frankfurt am Main, Germany
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
Figure 1. Crystal structure of 3.
Eur. J. Org. Chem. 2006, 1387–1389
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1387

A. S. K. Hashmi, M. C. Blanco, D. Fischer, J. W. Bats
SHORT COMMUNICATION
The diaryl-substituted compound 5 led to the products and 7 were in agreement with the fact that the maximum
6–10 (Scheme 2). Again the structure of 7 was unambigu- yield of these dimers is twice the percentage of the gold()
ously proven by a crystal structure analysis (Figure 2), the catalyst. Triplicating the amount of AuCl3 (to 15 mol-%)
same is true for 6 and 9.[13] Compound 8 is the analogue of also increased the amount of 3 or 7, but only by a factor
the substitution product 4, 9 the product of a 1,4-elimi- of 2.3 or 2.5, respectively; these reactions were even less
nation instead of a substitution. Compound 10 would be selective, especially the amount of the chloro compounds 4
the product of a dehydrative coupling, not a chloride ion or 8 went up significantly (9% or 20%, respectively). A pos-
but a dihydrofuryl moiety substitutes the hydroxy group. sible mechanism for the formation of 3 and 7 is shown in
Selective routes to products of type 4, 8 and 10 have been Scheme 3.
reported before.[14]
Scheme 2. Formation of 7 by a gold() catalyst.
Scheme 3. Possible mechanism for the formation of 3 and 7 by
gold().
The normal catalytic cycle is assumed to proceed via
AuIII and consist of a coordination of the remote double
bond in A, cyclization to B and deprotonation to C to af-
ford the dihydrofurans after proto-demetallation of C. Cal-
culations of Krause et al.[15] for allenyl carbinols back up
such a mechanism, and a similar mode of reaction has been
suggested for the cyclizations of the related allenyl
ketones[16] and the ring-enlargement of alkynyl oxiranes.[17]
Figure 2. Crystal structure of 7.
However, a ligand exchange between two molecules of C
The crucial reactions are the oxidative coupling reactions can also lead to D, which then delivers the dimer and AuI
to 3 and 7, which go along with the reduction of one com- by reductive elimination. The latter step would be related
ponent. Since neither reduced allene is isolated or detected to the oxidative dimerization of arylboronic acids reported
nor any other reduced organic species (as shown with com- by Corma et al.[18] The route through D sets free two pro-
pound 4, our analysis includes species formed in 1% tons (two equivalents of HCl in the case of AuCl3), thus
amounts) and no other oxidant like molecular oxygen is this process is ultimately combined with the substitution
present, it is only reasonable to assume that the strongest and elimination products 4, 8 and 9. Compound 10 could
oxidant present, AuIII, is reduced to AuI. The yields of 3 be formed by the reaction of C with the starting material
1388
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 1387–1389

AuIII-Catalyzed Cycloisomerization of Tertiary Allenyl Carbinols
SHORT COMMUNICATION
activated by protonation of the hydroxy group.[19] Applying
the AgI-catalysts we did not observe these side-products at
all.
Acknowledgments
This work was supported by the Fonds der Chemischen Industrie
and by the AURICAT EU-RTN (HPRN-CT-2002-00174).
We finally tested Gagosz[20] new (Ph3P)AuNTf2 catalyst
as a gold() catalyst for this reaction. Only 3 mol-% of this
catalyst in dichloromethane at room temperature converted
1 to the volatile 2 in an essentially quantitative yield (in situ
1H NMR) and 5 to 6 in 87% isolated yield.
[1]
[2]
A. S. K. Hashmi, Angew. Chem. 2005, 117, 7150–7154; Angew.
Chem. Int. Ed. 2005, 44, 6990–6993.
A. S. K. Hashmi, Gold Bull. 2004, 37, 51–65, see especially the
conclusion on page 63.
[3]
[4]
[5]
G. J. Hutchings, Gold Bull. 2004, 37, 3–11.
Conclusions
Y. Fukuda, K. Utimoto, J. Org. Chem. 1991, 56, 3729–3731.
J. H. Teles, S. Brode, M. Chabanas, Angew. Chem. 1998, 110,
1475–1478; Angew. Chem. Int. Ed. 1998, 37, 1415–1418.
Y. Fukuda, K. Utimoto, H. Nozaki, Heterocycles 1987, 25,
297–300.
Similar oxidative dimerizations of a substrate could re-
duce AuIII to AuI whenever an organogold() intermediate
is accessible. If this process is general, the question still re-
[6]
mains: in the individual case, does this reduction convert [7]
one catalytically active species into another, is it a way of
R. Lok, R. E. Leone, A. J. Williams, J. Org. Chem. 1996, 61,
3289–3297.
B. F. Straub, Chem. Commun. 2004, 1726–1728.
N. Asao, K. Takahashi, S. Lee, T. Kasahara, Y. Yamamoto, J.
Am. Chem. Soc. 2002, 124, 12650–12651.
[8]
forming a catalytically active species from a precatalyst or
[9]
is it even a method of catalyst deactivation?
[10]
C. J. Jones, D. Taube, R. A. Periana, Angew. Chem. 2004, 116,
4726–4729; Angew. Chem. Int. Ed. 2004, 43, 4626–4629.
A. S. K. Hashmi, L. Schwarz, J.-H. Choi, T. M. Frost, Angew.
Chem. 2000, 112, 2382–2385; Angew. Chem. Int. Ed. 2000, 39,
2285–2288.
Experimental Section
[11]
Reaction of 1 with AuCl3: Compound 1 (250 mg, 1.81 mmol) was
dissolved in absolute acetonitrile (5 mL) under dinitrogen and
AuCl3 (26.5 mg, 87.4 µol) was added. The reaction was monitored
by TLC, after consumption of the starting material the solvent was
removed and the crude material was purified by column
chromatography on silica gel (hexane/ethyl acetate, 40:1) to yield
118 mg (47%) of the known[21] compound 2, 25.4 mg (10%) of the
hitherto unknown 3 and 2.9 mg (1%) of the known[22] compound
4.
[12] A. Hoffmann-Röder, N. Krause, Org. Lett. 2001, 3, 2537–2538.
[13] CCDC-292999 (for 3), -293000 (for 6), -293001 (for 7) and -
293002 (for 9) contain the supplementary crystallographic data
for this paper. These data can be obtained free of charge from
The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
[14] a) S. Ma, G. Wang, Tetrahedron Lett. 2002, 43, 5723–5726; b)
S. Ma, W. Gao, J. Org. Chem. 2002, 67, 6104–6112.
[15] A. Hoffmann-Röder, N. Krause, Org. Biomol. Chem. 2005, 3,
387–391.
3: M.p. 98 °C. R = 0.46 (hexane/ethyl acetate, 10:1). IR (neat): ν
˜
f
= 3069, 2923, 2856, 1742, 1450, 1312, 1091, 931 cm–1. 1H NMR
(250 MHz, CDCl3): δ = 5.66 (t, J = 1.8 Hz, 2 H), 4.74 (d, J =
1.8 Hz, 4 H), 1.70–1.30 (m, 20 H) ppm. 13C NMR (62.9 MHz,
CDCl3): δ = 131.0 (s, 2 C), 130.1 (d, 2 C), 90.3 (s, 2 C), 73.0 (t, 2
C), 36.7 (t, 4 C), 25.1 (t, 2 C), 23.2 (t, 4 C) ppm. MS (70 eV):
m/z (%) = 274 (100) [M+], 203 (43), 231 (87). C18H26O2 (274.4):
calcd. C 78.79, H 9.55; found C 78.57, H 9.67.
4: Rf = 0.78 (hexane/ethyl acetate, 5:1). 1H NMR (250 MHz,
CDCl3): δ = 5.64 (q, J = 1.2 Hz, 1 H), 5.36 (d, J = 1.0 Hz, 1 H),
5.12 (t, J = 1.2 Hz, 1 H), 2.42–2.35 (m, 2 H), 2.20–2.10 (m, 2 H),
1.70–1.40 (m, 6 H) ppm. 13C NMR (62.9 MHz, CDCl3): δ = 146.8
(s), 120.0 (d), 114.2 (t), 77.0 (s), 36.9 (t), 29.5 (t), 28.1 (t), 27.6 (t),
26.3 (t) ppm. MS (70 eV): m/z (%) = 156 (2) [M+], 18 (100), 91 (44),
92 (34).
[16] a) J. Marshall, G. S. Bartley, J. Org. Chem. 1994, 59, 7169–
7171; b) A. S. K. Hashmi, T. L. Ruppert, T. Knöfel, J. W. Bats,
J. Org. Chem. 1997, 62, 7295–7304; c) A. S. K. Hashmi, L.
Schwarz, J. W. Bats, J. Prakt. Chem. 2000, 342, 40–51.
[17] A. S. K. Hashmi, P. Sinha, Adv. Synth. Catal. 2004, 346, 432–
438.
[18] a) S. Carretin, J. Guzman, A. Corma, Angew. Chem. 2005, 117,
2282–2285; Angew. Chem. Int. Ed. 2005, 44, 2242–2245; b) C.
González-Arellano, A. Corma, M. Iglesias, F. Sánchez, Chem.
Commun. 2005, 1990–1992.
[19] It is conceivable that the elimination of water is induced by
AuI, leading to a AuIII intermediate which then leads to 10 by
a ligand exchange with C and a subsequent reductive elimi-
nation.
[20] N. Mézailles, L. Ricard, F. Gagosz, Org. Lett. 2005, 7, 4133–
4136.
[21] B. Schmidt, H. Wildemann, Eur. J. Org. Chem. 2000, 3145–
3164.
[22] A. Horvath, J.-E. Baeckvall, J. Org. Chem. 2001, 66, 8120–
Supporting Information (for details see the footnote on the first
page of this article): Procedures and identification data for the
starting materials 1 and 5 and procedure for the gold-catalyzed
reaction that leads to 6–10 including characterization data for these
compounds.
8126.
Received: January 4, 2006
Published Online: January 30, 2006
Eur. J. Org. Chem. 2006, 1387–1389
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1389