Br?nsted Acid-Catalyzed Cascade Reactions Involving 1,2-Indole Migration

Article abstract of DOI:10.1002/chem.201502174

A cascade reaction of indoles with propargylic diols involving an unprecedented metal-free 1,2-indole migration onto an alkyne was carried out. DFT calculations support a mechanism consisting of a concerted nucleophilic attack of the indole nucleus with loss of water, followed by the 1,2-migration and subsequent Nazarov cyclization. This Br?nsted acid-catalyzed protocol affords indole-functionalized benzofulvene derivatives in high yields.

Full text of DOI:10.1002/chem.201502174

DOI: 10.1002/chem.201502174
Communication
&
Cascade Reactions
Brønsted Acid-Catalyzed Cascade Reactions Involving 1,2-Indole
Migration
Estela lvarez,[a] Olalla Nieto Faza,[b] Carlos Silva López,[b] Manuel A. Fernµndez-Rodríguez,[a]
and Roberto Sanz*[a]
Dedicated to our inspiring mentor Professor JosØ Barluenga on the occasion of his 75th birthday
Abstract: A cascade reaction of indoles with propargylic
diols involving an unprecedented metal-free 1,2-indole mi-
gration onto an alkyne was carried out. DFT calculations
support a mechanism consisting of a concerted nucleo-
philic attack of the indole nucleus with loss of water,
followed by the 1,2-migration and subsequent Nazarov
cyclization. This Brønsted acid-catalyzed protocol affords
indole-functionalized benzofulvene derivatives in high
yields.
Metal-catalyzed rearrangements have become the most
powerful tool for the construction of cyclic scaffolds from
simple materials, usually under mild reaction conditions.[1] For
Scheme 1. Previous work and our proposal.
instance, propargylic esters are versatile precursors of a wide
range of carbo- and heterocyclic compounds, through tandem
processes initiated by 1,2-acyl migration reactions onto the
alkyne activated by gold or platinum catalysts.[2] In this regard,
we have reported that 3-propargylindoles evolve in the
presence of catalytic amounts of cationic gold(I) complexes by
previously unknown 1,2-indole migration (Scheme 1).[3] Howev-
er, to the best of our knowledge, this type of rearrangement in
propargylic derivatives, which implies the rupture and
formation of carbon–carbon bonds, has not been reported
under metal-free conditions.
gylic alcohols, and their derivatives, have been demonstrated
to be a useful synthetic tool[6] through their transformation
into allenic carbocations[7] and subsequent intramolecular trap-
ping with electron-rich arenes, alkenes, enols, and heteroatom
nucleophiles.[8]
We have considered the possibility that a Brønsted acid
could trigger an intramolecular nucleophilic attack of the
indole onto a propargylic cation, generated from the corre-
sponding alcohol, followed by the opening of the cyclopropyl
ring, thus mimicking the behavior of cationic gold(I) catalysts
(Scheme 1); it has been reported that some reactions can be
catalyzed by p-acids as well as by protons.[9] We describe
herein the first examples of metal-free promoted 1,2-indole
migration onto alkynes. This Brønsted acid-catalyzed
rearrangement is feasible in a cascade fashion, starting from
simple indoles and propargylic diols, delivering a new and
efficient synthetic access to 2-indol-3-ylbenzofulvenes with
water as the only byproduct.
In the past few years, our group has developed methodolo-
gies for the direct nucleophilic substitution reaction of p-acti-
vated alcohols under metal-free Brønsted acid-catalysis.[4] In
this context, we have described a robust method for the reac-
tion of indoles with propargylic alcohols that provides suitable
access to a wide variety of 3-propargylindole derivatives
(Scheme 1).[5] In addition, cascade rearrangements of propar-
[a] E. lvarez, Dr. M. A. Fernµndez-Rodríguez, Prof. Dr. R. Sanz
rea de Química Orgµnica, Departamento de Química
Facultad de Ciencias, Universidad de Burgos
Pza. Misael BaÇuelos s/n, 09001-Burgos (Spain)
E-mail: rsd@ubu.es
Our study began by assessing the viability of a cascade
reaction in alkyne 1a, bearing an indole at one propargylic po-
sition and an activated hydroxyl group at the other propargylic
position, promoted by p-toluenesulfonic acid (PTSA) as
a simple and easily available Brønsted acid catalyst. Gratifying-
ly, we found that under our standard conditions for Brønsted
acid-catalyzed direct nucleophilic substitution reactions (MeCN,
5 mol% PTSA), a new product with a benzofulvene core 2a
was formed (Scheme 2). It is worth noting that benzofulvenes
[b] Dr. O. Nieto Faza, Dr. C. Silva López
Departamento de Química Orgµnica
Universidade de Vigo, Lagoas Marcosende
36310-Vigo, Galicia (Spain)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201502174.
Chem. Eur. J. 2015, 21, 12889 – 12893
12889
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
found that the tandem 1,2-indole migration/cyclization is gen-
eral for a wide variety of alkynols possessing tertiary hydroxyl
groups (Table 1, entries 1–11) as well as secondary ones
(Table 1, entries 12–16). For the latter, activating groups such
as aryl, alkenyl, or cyclopropyl are required as R1 substitu-
ents.[14] When the two groups, R1 and R2, are different, variable
mixtures of geometrical isomers are obtained, typically
favoring the E·isomer. To ensure homogeneity of the reaction,
some experiments were conducted in MeCN instead of TFE. In
addition, for less activated alkynols higher loads of catalyst
and/or the stronger 2,4-dinitrobenzensulfonic acid (DNBSA)
were used. In most cases, excellent yields for the benzofulvene
derivatives 2 were obtained in short reaction times (1–6 h).[15]
We decided to use DFT to study the mechanism of this
novel reaction, in order to understand the steps involved in
such a dramatic one-pot transformation (see Scheme 3 and
Figure 1). For the calculations, we have chosen a model system
bearing methyl groups at all the propargylic positions apart
from the required phenyl-substituted one. Starting from the
protonated alkynol, IH, a cationic cascade is triggered. The de-
parture of water as a leaving group occurs without barriers
and proceeds in concert with the SN2’ nucleophilic attack of
the indole C3’ position to the incipient carbocation in C2. The
resultant product, 11.7 kcalmolÀ1 lower in energy than IH, is an
ethenylidenecyclopropane intermediate (IIH), with a C1ÀC3’
bond significantly larger (by 0.07 Š) than the newly formed
C2ÀC3’ bond. This slightly lower bond order for C1ÀC3’ (0.76
vs. 0.78) facilitates the cleavage of the C1ÀC3’ bond through
TSIIÀIII(H), resulting in a low energy step requiring only 3.7 kcal
molÀ1 to complete the 1,2-indole migration. This migration
leads to vinylallenyl cation IIIH, which undergoes a relatively
facile (8.5 kcalmolÀ1 barrier) con-
Scheme 2. Preliminary results and proof of concept.
are interesting compounds with relevance in applied chemis-
try.[10] Although different synthetic strategies for accessing the
benzofulvene core have been reported,[11] the development of
straightforward and efficient methods for the synthesis of
functionalized benzofulvenes is still highly desirable.
Full conversion was easily achieved in the model reaction by
increasing the catalyst load and the reaction time. Moreover,
by using trifluoroethanol (TFE) as solvent,[12] the reaction time
was shortened and the final benzofulvene 2a was isolated as
a mixture of geometrical isomers in pure form and excellent
yield by simple filtration. Interestingly, its structure, which was
confirmed by X-ray diffraction of the Z isomer,[13] indicates that
the intended metal free-promoted indole migration has
occurred along with the formation of a new five-membered
ring involving the phenyl group initially attached to the same
progargylic carbon as the indole.
To examine the scope of this novel Brønsted acid-catalyzed
cascade cyclization, we applied this reaction protocol to a set
of 3-propargylindoles 1 having varied substitution at the alco-
hol position. The results have been summarized in Table 1. We
rotatory four-electron electrocyc-
Table 1. Synthesis of 2-indol-3-ylbenzofulvenes 2. Scope of the reaction.
lization to the much more ther-
modynamically
18.0 kcalmolÀ1
favored
(by
)
structure IVH,
which results in the formation of
the benzofulvene product after
rearomatization. Though low,
the barrier of this electrocyclic
ring-closure
is
significantly
higher than the barrier corre-
sponding to the electrocycliza-
tion of the allylallene (4.6 kcal
Entry
1
R1
R2
H+ (x [mol%])
Solvent
2
Z/E[a]
Yield [%][b]
1
2
3
4
5
6
7
8
1a
1b
1c
1d
1e
1 f
1g
1h
1i
1j
1k
1l
1m
1n
1o
1p
Ph
Ph
Ph
Me
c-C3H5
iPr
Me
Me
PTSA (10)
PTSA (10)
PTSA (10)
PTSA (10)
PTSA (10)
PTSA (10)
PTSA (10)
DNBSA (30)
PTSA (30)
PTSA (30)
PTSA (30)
DNBSA (20)
DNBSA (50)
DNBSA (10)
PTSA (10)
DNBSA (20)
TFE
TFE
TFE
TFE
2a
2b
2c
2d
2e
2 f
2g
2h
2i
2j
2k
2l
2m
2n
2o
2p
1.2/1
1/3.5
1/3.4
1/2
1.3/1
–
–
–
–
–
98
96
88
95
93
98
88
93
67
69
93
95
75
75
85
94
molÀ1
) or pentadienyl cations
(4.3 kcalmolÀ1), due to the loss
of aromaticity of the phenyl
group along the reaction path.
This latter proposed step has
been reported in the catalysis of
a-allenyl benzyl alcohols using
Brønsted or Lewis acids.[16]
4-MeOC6H4
2-Th
Ph
4-MeOC6H4
4-ClC6H4
Me
TFE
Ph
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
TFE
4-MeOC6H4
4-ClC6H4
Me
9
10
11
12
13
14
15
16
Et
Et
À(CH2)4À
4-MeC6H4
4-BrC6H4
2-Th
CH=CH(4-MeOC6H4)
c-C3H5
–
H
H
H
H
H
1/1
1/1.5
1/1.7
1/2
1/2.4
Using [AuPH3]+-coordinated
1 as a starting point, we calculat-
ed the equivalent gold-catalyzed
mechanism (see the Supporting
Information for details) to ex-
TFE
TFE
MeCN
1
[a] Z/E ratio determined by H NMR analysis. [b] Isolated yield after filtration or column chromatography.
Chem. Eur. J. 2015, 21, 12889 – 12893
www.chemeurj.org
12890
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
stability of the water leaving
group results in a cationic cas-
cade of reactions downhill from
the reactant, whereas, for the
gold-catalyzed manifold, only
the final product is lower in
energy than the activated sub-
strate IAu (see the Supporting In-
formation).
At this point, and taking ad-
vantage from our developed
methodology for the direct
Scheme 3. Detailed mechanism based on our computational exploration. The free energies values (relative to
starting compound) are given in kcalmolÀ1
propargylation of indoles under
Brønsted acid catalysis,[5] we en-
.
visaged that this new process
plore the potential of Brønsted acids as a replacement for cat-
ionic gold in synthesis. The Brønsted acid-catalyzed and the
gold-catalyzed mechanisms display a similar sequence of steps
and reaction profile, with the main difference being the lower
barrier for the electrocyclization in the latter. The most remark-
able feature, however, is the downhill path followed by the
Brønsted acid-catalyzed system upon protonation of 1, which
is expected to be much more favorable than the alternative re-
action with gold. Protonation of 1 provides a substrate already
poised to react, resulting in a barrierless TSI-II(H). After this, the
could be performed in a straightforward manner from simple
indoles 3 and but-2-yne-1,4-diols 4 with the loss of water as
the only byproduct (Scheme 4). The required acetylenic 1,4-
diols 4 are easily prepared, though few, albeit highly interest-
ing, synthetic applications have been described for them in
recent years,[17] including a remarkable acid-catalyzed domino
reaction of highly activated tetraarylbut-2-yne-1,4-diols with 1-
naphthol.[18]
So, we evaluated the feasibility of our proposed intermolec-
ular approach by employing 1H-indole 3a and symmetric
Figure 1. Schematic M06/def2-SVP (PCM, CH2Cl2) free energy profile for a model 1–2 transformation under Brønsted acid (black) and gold catalysis (red).
Chem. Eur. J. 2015, 21, 12889 – 12893 www.chemeurj.org ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
12891

Communication
of the corresponding benzofulvenes 2 were obtained in short
reaction times,[15] with a small decrease in the yield observed
when 5-bromoindole 3c was used (Table 2, entries 1–3). Start-
ing from highly activated tetraphenylbut-2-yne-1,4-diol (4b),
we observed that its reaction with indole 3a under PTSA-catal-
ysis in MeCN gave rise to the expected benzofulvene 2ab in
moderate yield, along with furan derivative 5 that was isolated
in 25% yield (Table 2, entry 4). Its formation shows that in this
case an alternative mechanism should be at least partially op-
erative. Based on a recent report about the reaction of 1-naph-
thol with diol 4b,[18] we propose a competitive SN’ attack of the
indole to 4b leading to an allenyl alcohol that subsequently
undergoes furan formation[19] instead of a Nazarov electrocycli-
zation.[20] The formation of the 2,5-dihydrofuran derivative 5
was suppressed by using DNBSA as catalyst, wherein 2ab was
isolated in 90% yield (Table 2, entry 5). Under these conditions,
benzofulvene derivatives 2db, 2eb, and 2fb were also
obtained in excellent yields (Table 2, entries 6–8), even when
using less nucleophilic indoles 3e,f.[15]
Scheme 4. Proposed direct synthesis of benzofulvenes 2 from indoles 3 and
acetylenic 1,4-diols 4.
acetylenic 1,4-diol 4a as substrate under PTSA-catalysis in TFE.
Gratifyingly, a high yield of the benzofulvene derivative
2aa was obtained (Table 2, entry 1.) We further tested this
sequence with a variety of functionalized indoles 3 and symm-
etric diols 4a,b. With diol 4a mixtures of geometrical isomers
The diversity and applicability of this acid-catalyzed
cascade cyclization for the synthesis of 2-indol-3-yl-
benzofulvenes 2 starting from non-symmetric acety-
Table 2. Synthesis of 2-indol-3-ylbenzofulvenes 2 from indoles 3 and acetylenic 1,4-
diols 4.
lenic 1,4-diols 4c–j were also surveyed (Table 2, en-
tries 9–16). We reasoned that the different degree of
activation of both hydroxyl groups could bias the
cascade sequence to selectively give rise to one ben-
zofulvene derivative 2. Ditertiary diols 4c–e contain-
ing one aromatic and one alkyl group at each of the
propargylic positions are the more challenging sub-
strates. At these sites, selective reactions took place,
provided that one of the hydroxyl groups was more
activated than the other one, which was the case for
4c (cyclopropyl vs. methyl) and 4d (4-methoxyphenyl
vs. phenyl) (Table 2, entries 9 and 10). However, for
4e (4-chlorophenyl vs. phenyl) a mixture of two ben-
zofulvene derivatives 2ae and 2’ae was obtained, de-
rived from the initial attack of the indole at the two
different hydroxyl groups (Table 2, entry 11). Not un-
expectedly at this point, diols 4 f–j, which possess dif-
ferent activated alcohols (tertiary vs. secondary, or
tertiary benzylic vs. tertiary) selectively gave rise to
the corresponding benzofulvenes 2 in moderate to
good yields (Table 2, entries 12–16). The lower yields
obtained with diols 4 f,g are probably due to the
lower reactivity of the secondary hydroxyl group that
also leads to the need of a higher catalyst load (en-
tries 12 and 13). Interestingly, with these non-sym-
metric diols 4c–j, this cascade reaction formally in-
volves the regioselective SN’ addition of the indole to
the less activated alkynol, leading to a tertiary allenyl
alcohol that undergoes a Nazarov cyclization.
Entry
3
R1
H
4
Ar
R2
R3
R4 Met.[a] 2[b] Yield [%][c]
1
2
3
4
5
6
7
8
3a
3b 2-Me
3c 5-Br
3a
3a
3d 1-Me
4a Ph
4a Ph
4a Ph
4b Ph
4b Ph
4b Ph
Me
Me
Me
Ph
Ph
Ph
Ph
Ph
c-C3H5 Ph
Ph
Me
c-C3H5 Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Me
Me
Me
A
A
A
2aa 87
2ba 75
2ca 66
2ab 53[e]
2ab 90
2 db 85
2eb 86
2 fb 75
2ac 90
2ad 95
2ae 65[f]
2af 44
2ag 42
2ah 63
2ai 70
2aj 58
H
H
Ph A[d]
Ph
Ph
Ph
Ph
Me
Me
B
B
B
B
A
A
3e 5-CO2Me 4b Ph
3 f 6-NO2
4b Ph
4c Ph
4d 4-MeOC6H4 Et
4e Ph
4 f Ph
9
3a
3a
3a
3a
3a
3a
3a
3a
H
H
H
H
H
H
H
H
10
11
12
13
14
15
16
4-ClC6H4 Me
A
H
H
B[g]
B[g]
4g 4-MeOC6H4 Et
Ph
4h Ph
4i Ph
4j Ph
c-C3H5 Me
Me B[g]
B[g]
c-C3H5 À(CH2)4À
Me
À(CH2)5À
B[g]
[a] Method A: PTSA, TFE; Method B: DNBSA, MeCN. [b] When R3 ¼ R4, the resultant Z/E
ratio was approximately 1/1, as determined by 1H NMR analysis, except for 2ba with
a Z/E ratio of approximately 1/2.4. [c] Isolated yield after filtration or column chroma-
tography. [d] Carried out in MeCN. [e] 22% of furan derivative 5 (see below) was also
isolated. [f] Two benzofulvene derivatives 2ae and 2’ae were obtained in a ca. 1.2:1
ratio. The major one corresponds to Ar=Ph and the minor one to Ar=ClC6H4. Using
MeCN as solvent, an approximately 2.8:1 mixture of 2ae and 2’ae was obtained.
[g] Carried out with one (entries 12 and 13) or two (entries 14–16) subsequent addi-
tions of PTSA (10 mol%).
In conclusion, simple Brønsted acids have been
demonstrated as useful catalysts for emulating the
previously reported gold-catalyzed 1,2-indole migra-
tion of 3-propargylindoles, being the first examples
of a metal-free promoted migration of a carbon-cen-
tered moiety in such propargylic substrates. In addi-
Chem. Eur. J. 2015, 21, 12889 – 12893
www.chemeurj.org
12892
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
[7] For examples of intramolecular trapping of allenic carbocations formed
from propargylic alcohols in the presence of I2, see: F. Yang, K. G. Ji,
H. T. Zhu, A. Shaukat, X. Y. Liu, Y. M. Liang, Chem. Eur. J. 2011, 17, 4986–
4990.
tion, DFT calculations reveal that the exothermicity of the
water elimination step and the moderate energy requirements
of the Nazarov-like electrocyclization are key to the occurrence
of this reaction cascade under Brønsted acid catalysis. Al-
though the mechanism is very similar to that experienced by
the substrate under gold activation, replacement of gold by
a Brønsted acid leads to a more favorable catalytic process.
From a synthetic point of view, this methodology, one that im-
plies the rupture and formation of carbon-carbon bonds, rep-
resents a significant addition to the armory of transformations
catalyzed by Brønsted acids and provides an operationally
simple and efficient manner to construct highly interesting
indole-functionalized benzofulvene derivatives under metal-
free conditions from easily accessible starting materials.
[8] Selected examples: Electron-rich arenes: a) W. Huang, P. Zheng, Z.
Zhang, R. Liu, Z. Chen, X. Zhou, J. Org. Chem. 2008, 73, 6845–6848;
b) T. Yokosaka, N. Shiga, T. Nemoto, Y. Hamada, J. Org. Chem. 2014, 79,
3866–3875; Alkenes: c) K.-G. Ji, X.-Z. Shu, S.-C. Zhao, H.-T. Zhu, Y.-N. Niu,
X.-Y. Liu, Y.-M. Liang, Org. Lett. 2009, 11, 3206–3209; d) L.-F. Yao, Y. Wei,
M. Shi, J. Org. Chem. 2009, 74, 9466–9469; e) T. Haven, G. Kubik, S. Hau-
benreisser, M. Niggemann, Angew. Chem. Int. Ed. 2013, 52, 4016–4019;
Angew. Chem. 2013, 125, 4108–4111; Enols: f) W. T. Teo, W. Rao, C. J. H.
Ng, S. W. Y. Koh, P. W. H. Chan, Org. Lett. 2014, 16, 1248–1251; Heteroa-
tom nucleophiles: g) P. Kothandaraman, W. Rao, S. J. Foo, P. W. H. Chan,
Angew. Chem. Int. Ed. 2010, 49, 4619–4623; Angew. Chem. 2010, 122,
4723–4727.
[9] a) A. S. K. Hashmi, L. Schwarz, P. Rubenbauer, M. C. Blanco, Adv. Synth.
Catal. 2006, 348, 705–708; b) J. G. Taylor, L. A. Adrio, K. K. Hii, Dalton
Trans. 2010, 39, 1171–1175.
[10] a) Y. Kosaka, K. Kitazawa, S. Inomata, T. Ishizone, ACS Macro Lett. 2013,
2, 164–167; b) Y. Zhang, H. Ma, J. Huang, J. Mol. Catal. A 2013, 373, 85–
95.
Acknowledgements
We gratefully acknowledge the Junta de Castilla y León
(BU237U13) and Ministerio de Economía y Competitividad
(MINECO) and FEDER (CTQ2013-48937-C2-1P and 2-P) for finan-
cial suport. E. A. thanks Ministerio de Educación y Ciencia
(MEC) for a predoctoral FPU contract.
[11] For recent reports, see: a) D.-Y. Li, Y. Wei, M. Shi, Chem. Eur. J. 2013, 19,
15682–15688; b) Q. Xiao, H. Zhu, G. Li, Z. Chen, Adv. Synth. Catal. 2014,
356, 3809–3815; c) C. Martinelli, A. Cardone, V. Pinto, M. M. Talamo,
M. L. D’arienzo, E. Mesto, E. Schingaro, F. Scordari, F. Naso, R. Musio,
G. M. Farinola, Org. Lett. 2014, 16, 3424–3427, and references therein.
[12] Fluorinated alcohols have been shown as effective reaction media for
activating alcohols through hydrogen bonding and promoting direct
nucleophilic substitution reactions. See, for instance: P. Trillo, A. Baeza,
C. Nµjera, J. Org. Chem. 2012, 77, 7344–7354.
[13] CCDC 1012587 (2a), 1012589 (2g), 1012588 (2 db) contain the supple-
mentary crystallographic data for this paper. These data can be ob-
tained free of charge from The Cambridge Crystallographic Data
Centre.
[14] No reaction took place with a secondary alkynol bearing R1 =nBu.
[15] See Supporting Information for experimental details. Structural assign-
ments of all products 2 were based on a series of NMR experiments
and confirmed by single-crystal X-ray diffraction analysis of 2a, 2g, and
2 db (see ref. [13]).
[16] P. Cordier, C. Aubert, M. Malacria, E. Lacôte, V. Gandon, Angew. Chem.
Int. Ed. 2009, 48, 8757–8760; Angew. Chem. 2009, 121, 8913–8916.
[17] a) K.-G. Ji, H.-T. Zhu, F. Yang, X.-Z. Shu, S.-C. Zhao, X.-Y. Liu, A. Shaukat,
Y.-M. Liang, Chem. Eur. J. 2010, 16, 6151–6154; b) S. R. Mothe, P. Ko-
thandaraman, S. J. L. Lauw, S. M. W. Chin, P. W. H. Chan, Chem. Eur. J.
2012, 18, 6133–6137; c) S. R. Mothe, S. J. L. Lauw, P. Kothandaraman,
P. W. H. Chan, J. Org. Chem. 2012, 77, 6937–6947; d) J.-H. An, H. Yun, S.
Shin, S. Shin, Adv. Synth. Catal. 2014, 356, 3749–3754.
Keywords: benzofulvenes · Brønsted acid · catalysis · indole
migration · Nazarov cyclization
[1] a) Y. Yamamoto, Chem. Rev. 2012, 112, 4736–4769; b) C. Aubert, L. Fen-
sterbank, P. García, M. Malacria, A. Simonneau, Chem. Rev. 2011, 111,
1954–1993; c) V. Michelet, P. Y. Toullec, J.-P. GenÞt, Angew. Chem. Int. Ed.
2008, 47, 4268–4315; Angew. Chem. 2008, 120, 4338–4386.
[2] A. S. K. Hashmi, F. D. Toste, Modern Gold Catalyzed Synthesis Wiley-VCH,
Weinheim, 2012. For related 1,2-rearrangements, see: X. Zhao, Z.
Zhong, L. Penga, W. Zhang, J. Wang, Chem. Commun. 2009, 2535–
2537; T. Sato, T. Onuma, I. Nakamura, M. Terada, Org. Lett. 2011, 13,
4992–4995.
[3] a) R. Sanz, D. Miguel, F. Rodríguez, Angew. Chem. Int. Ed. 2008, 47,
7354–7357; Angew. Chem. 2008, 120, 7464–7467; b) R. Sanz, D. Miguel,
M. Gohain, P. García-García, M. A. Fernµndez-Rodríguez, A. Gonzµlez-
PØrez, O. Nieto-Faza, A. R. De Lera, F. Rodríguez, Chem. Eur. J. 2010, 16,
9818–9828; c) E. lvarez, D. Miguel, P. García-García, M. A. Fernµndez-
Rodríguez, F. Rodríguez, R. Sanz, Beilstein J. Org. Chem. 2011, 7, 786–
793; d) E. lvarez, D. Miguel, P. García-García, M. A. Fernµndez-Rodrí-
guez, F. Rodríguez, R. Sanz, Synthesis 2012, 44, 1874–1884.
[4] a) R. Sanz, A. Martínez, J. M. lvarez-GutiØrrez, F. Rodríguez, Eur. J. Org.
Chem. 2006, 1383–1386; b) R. Sanz, D. Miguel, A. Martínez, J. M. lvar-
ez-GutiØrrez, F. Rodríguez, Adv. Synth. Catal. 2006, 348, 1841–1845.
[5] a) R. Sanz, D. Miguel, J. M. lvarez-GutiØrrez, F. Rodríguez, Synlett 2008,
975–978; b) R. Sanz, D. Miguel, A. Martínez, M. Gohain, P. García-García,
M. A. Fernµndez-Rodríguez, E. lvarez, F. Rodríguez, Eur. J. Org. Chem.
2010, 7027–7039.
[18] C. M. Sousa, J. Berthet, S. Delbaere, P. J. Coelho, J. Org. Chem. 2014, 79,
5781–5786.
[19] For the synthesis of 2,5-dihydrofurans by cycloisomerization of a-hy-
droxyallenes, see: M. Poonoth, N. Krause, Adv. Synth. Catal. 2009, 351,
117–122, and references cited therein. See also ref. [16].
[20] See Supporting Information for further discussion about these mecha-
nistic proposals.
[6] For recent reviews, see: a) Y. Zhu, L. Sun, P. Lu, Y. Wang, ACS Catal.
2014, 4, 1911–1925; b) B. J. Ayers, P. W. H. Chan, Synlett 2015, 26,
1305–1339.
Received: June 4, 2015
Published online on July 24, 2015
Chem. Eur. J. 2015, 21, 12889 – 12893
www.chemeurj.org
12893
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim