[3,3]-Sigmatropic rearrangement step in the gold-catalyzed cyclization of allyl-(ortho-alkinylphenyl)methyl ethers

Article abstract of DOI:10.1002/asia.201300324

The gold-catalyzed conversion of allyl-(ortho-alkynylphenyl)methyl ethers was investigated, and allylated isochromenes were obtained. An optimization of the catalysis conditions with respect to different phosphane and carbene ligands on gold, different counterions, and different solvents was conducted. Subsequently, the scope and limitations of this reaction were investigated with 21 substrates. The mechanistic studies show an allylic inversion, as supported by NMR data and an X-ray crystal structure analysis, as well as an intermolecular reaction, as determined by crossover experiments. There is no competition of protodeauration even in the presence of water. All these observations differ from other related conversions and clearly indicate product formation by a [3,3]sigmatropic rearrangement in the step forming the new C-C bond. The Golden Key: Allylic inversion and crossover experiments reveal a sigmatropic rearrangement as a key step in the gold-catalyzed intramolecular allyl transfer in ortho-alkinyl benzyl allyl ethers. Copyright

Full text of DOI:10.1002/asia.201300324

FULL PAPER
FULL PAPER
Sigmatropic Rearrangement
Martin Ackermann, Janina Bucher,
Melissa Rappold, Katharina Graf,
Frank Rominger,
The Golden Key: Allylic inversion and
crossover experiments reveal a sigma-
tropic rearrangement as a key step in
the gold-catalyzed intramolecular allyl
transfer in ortho-alkinyl benzyl allyl
ethers.
A. Stephen K. Hashmi*
&&&& — &&&&
AHCTUNGTRENN[GUN 3,3]-Sigmatropic Rearrangement Step
in the Gold-Catalyzed Cyclization of
Allyl-(ortho-alkinylphenyl)methyl
Ethers
1
&
&
Chem. Asian J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
These are not the final page numbers! ÞÞ

www.chemasianj.org
A. Stephen K. Hashmi et al.
DOI: 10.1002/asia.201300324
ACHTUNGTRENNUNG[3,3]-Sigmatropic Rearrangement Step in the Gold-Catalyzed Cyclization of
Allyl-(ortho-alkinylphenyl)methyl Ethers
Martin Ackermann, Janina Bucher, Melissa Rappold, Katharina Graf,
Frank Rominger, and A. Stephen K. Hashmi*[a]
Abstract: The gold-catalyzed conver-
sion of allyl-(ortho-alkynylphenyl)-
methyl ethers was investigated, and al-
lylated isochromenes were obtained.
An optimization of the catalysis condi-
tions with respect to different phos-
phane and carbene ligands on gold, dif-
ferent counterions, and different sol-
vents was conducted. Subsequently, the
scope and limitations of this reaction
were investigated with 21 substrates.
The mechanistic studies show an allylic
inversion, as supported by NMR data
and an X-ray crystal structure analysis,
as well as an intermolecular reaction,
as determined by crossover experi-
ments. There is no competition of pro-
todeauration even in the presence of
water. All these observations differ
from other related conversions and
clearly indicate product formation by
a [3,3]sigmatropic rearrangement in the
À
step forming the new C C bond.
Keywords: alkenes
·
alkynes
·
ethers · gold · heterocycles
Introduction
up using this strategy.[2] Amongst these, palladium-[2a–f] and
platinum-catalyzed[2g–o] reactions played a dominating role
for years until Nakamura et al. established a gold-catalyzed
synthesis of benzothiophene using the same methodology.[2p]
Since the field of homogeneous gold catalysis was (and still
is) growing exponentially,[3] it is not surprising that the
number of examples in which gold catalysts were used in
In 1988, Tsuda and Saegusa published the first example of
a transition-metal-catalyzed heterocyclization with an allylic
migration in a domino reaction.[1] Starting with allylic esters
of homopropargylic acid 1, they accomplished the lactoniza-
tion of the ester, the fragmentation of the allylic moiety, and
À
the subsequent C C-bond formation that yielded the rear-
ranged products 4 (Scheme 1).
a
cyclization–rearrangement methodology for forming
highly substituted heterocycles has risen tremendously.[2p–ae]
Among these examples, benzannulation reactions play an
important role; from easily available starting materials,
product scaffolds of high importance like indoles,[2b,f,i,m,n,q,ag]
benzofurans,[2c,d,e,j,k,l] or isocoumarines[2k,y,z,af] could be synthe-
sized under exceptionally mild conditions. Using the cycliza-
tion–rearrangement strategy, we recently contributed a gold-
catalyzed synthesis of 4-substituted isocoumarine derivatives
8.[4] The detailed mechanistic investigations revealed an in-
termolecular transfer of the allyl fragments. Furthermore,
a palladium co-catalyst[2y] is not needed in most cases
(Scheme 2).
Scheme 1. Lactonization combined with an allylic shift according to
Tsuda and Saegusa.[1]
In order to expand this synthetic protocol to other sub-
strates, we envisioned the possibility to form isochromene
derivatives by using ortho-alkynyl benzylic ethers instead of
ortho-alkinyl benzoic esters as starting materials. The iso-
chromene motif belongs to the most important pyrane deriv-
atives and can be found in a variety of natural compounds
with biological or physiological activity.[5] However, the clas-
sical synthetic approaches to isochromenes suffer from
harsh reaction conditions and are incompatible with more
complex structures.[6,7] The transition-metal-catalyzed cycli-
zations of ortho-alkynyl benzylic alcohols 9 overcome these
problems. Yet, another disadvantage persists: After p-philic
activation of the multiple bond, the cyclization forms an or-
ganometallic intermediate 11 that is rapidly proto-demeta-
During the last 20 years, related transition-metal-catalyzed
cyclization–rearrangement sequences have been reported,
and an array of different heterocyclic motifs have been built
[a] Dr. M. Ackermann, J. Bucher, M. Rappold, K. Graf,
Dr. F. Rominger, Prof. Dr. A. S. K. Hashmi
Organisch-Chemisches Instiut
Universitꢁt Heidelberg
Im Neuenheimer Feld 270, 69120 Heidelberg (Germany)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/asia.201300324.
&
&
2
Chem. Asian J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

www.chemasianj.org
A. Stephen K. Hashmi et al.
while they used terminal al-
kynes exclusively without any
aromatic backbone (13), we an-
ticipated that our substrates 15
may follow a different cycliza-
tion mode leading to the six-
membered heterocycles 16 in-
stead of isobenzofuran deriva-
tives (Scheme 4).
Results and Discussion
Scheme 2. Synthesis of 4-substituted isocoumarine derivatives 8 using a gold-catalyzed cyclization–rearrange-
ment strategy.
Prior to the catalytic investiga-
tions, two test systems were
easily prepared in three steps
lated by the proton released during the attack of the alcohol
(Scheme 3).[8–10] Thus, no substituent other than H can be
tethered in the 4-position of 12.
(Scheme 5). First, ortho-bromobenzaldehyde 17 was con-
verted into the corresponding alkyne derivatives 18a and
18b by Sonogashira coupling, reduction with NaBH4, and
subsequent etherification to fur-
nish the desired test substrates
20a and 20b in an overall yield
of 35% and 39%, respectively.
Our first test reactions with
20a and 20b were carried out
Scheme 3. Transition-metal-catalyzed synthesis of isochromene derivatives 12.
under the optimized conditions
of our isocoumarine synthesis.[4]
The substrates were added to
Herein, we report on our efforts to circumvent this draw-
back by a cyclization–rearrangement sequence using ortho-
alkynyl benzylethers (15) as starting material. To the best of
a stirred solution of 5 mol% catalyst K1 and 5 mol%
AgOTf in technical grade 1,4-dioxane at room temperature.
After 16 hours, full conversion was achieved for both test
substrates. However, the isolated yield for phenyl-substitut-
ed isochromene 21a was unsatisfactory (18%), while prod-
uct 21b containing the alkyl substituent was isolated in an
acceptable yield of 66% (Scheme 6). To our delight, no
traces of the simple cyclization product 22, which could
have been formed by a competing protodeauration pathway,
were detected.
The 1H NMR spectra clearly
À
proved that the new C C bond
of the rearranged cyclization
products had been formed ex-
clusively with inversion of the
cinnamyl fragment (Figure 1).
Scheme 4. Comparison of the furan synthesis by Gagosz et al.[2ac] with our
current studies.
our knowledge, this strategy has
never been used for the synthe-
sis of isochromene derivatives
with any transition metal cata-
lyst before.[11] However, the
structural skeleton of the start-
ing material was similar to sub-
strates used by Gagosz et al. for
the synthesis of highly substitut-
ed furans (Scheme 4).[2ac] But, Scheme 5. Synthetic approach to the test systems.
3
&
&
Chem. Asian J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
These are not the final page numbers! ÞÞ

www.chemasianj.org
A. Stephen K. Hashmi et al.
out to be completely inactive in
this reaction (entry 8). In the
absence of any gold complex,
no conversion of the starting
material was observed; neither
the silver salt itself nor p-TsOH
as a Brønsted acid showed cata-
lytic activity (entries 9 and 10).
Notably, the expected proto-
deauration by-product 22 was
not detectable in any of the ex-
periments, although technical
grade solvent and open-flask
conditions were applied.
Scheme 6. Gold-catalyzed conversion of substrates 20a and 20b.
With K7 as the best catalyst
in hand, we then determined
the influence of different coun-
terions
on
the
reaction
(Table 2). The control experi-
ment without any activation of
the gold catalyst showed no
conversion of the starting mate-
rial (entry 1). AgOAc (entry 2),
Ag
(entry 4), and the silver salt of
binolphosphate (entry 5,
ACHTUGNTRENUN(GN CO2CF3) (entry 3), AgOTs
5 mol% each) were not effi-
cient for the activation. AgOTf
Figure 1. 1H NMR spectrum of 21b with a branched cinnamyl side chain.
(entry 6,
65%),
AgONf
(entry 7, 67%), and AgNTf2
(entry 8, 78%) gave good re-
This is noteworthy, as the corresponding ester derivatives
always retained the linear structure of the transformed
moiety.[4]
sults after 16 hours. Even higher yields were achieved when
AgPF6 (entry 9, 85%) or AgSbF6 (entry 10, 100%) were
used for the activation of the gold catalyst, most probably
due to their spherical symmetry shape, which results in their
lowest coordination tendency.
Finally, different solvents were investigated (Table 3). Op-
timized conditions from the earlier screenings were applied
(5 mol% of [SPhosAuCl] and 5 mol% of AgNTf2, Table 1,
entry 7). In order to enable milder conditions, all of the re-
actions were performed at room temperature.
1,4-Dioxane (entry 1, 71%) delivered only moderate
yields, but the other tested solvents significantly improved
the yields, no matter if protic (MeOH, entry 2, 88%), polar
nonprotic (acetonitrile, entry 3, 85%), or non-polar solvents
were applied (benzene, entry 4, 90%). Excellent results
were obtained in DMF (entry 5, 97%), diethyl ether
(entry 6, 98%), or CH2Cl2 (entry 7, 100%). In all of these
solvents, no traces of by-products were detected.
Under these optimized conditions, we next explored the
scope and limitations of this domino reaction with a small li-
brary of differently substituted (ortho-alkynyl benzylic)
ethers. Two possible approaches to the starting materials
were investigated. Firstly, a three-step protocol using ortho-
bromo benzaldehyde 17 as a starting material and utilizing
Sonogashira coupling, reduction with NaBH4, and subse-
quent etherification of the benzylic alcohol was examined
Both of these facts gave us first hints about the possible
reaction mechanism that seemed to differ from the mecha-
nism of the gold-catalyzed isocoumarine synthesis. Before
exploring the scope of this reaction, we first wanted to opti-
mize the reaction conditions with regard to different gold
catalysts, counterions, solvents, and temperatures. First, we
screened different ligands at the gold center. The test sub-
strate 20a that has a huge potential for optimization was
used. The results are shown in Table 1.
With AgNTf2 as a catalyst-activating additive, 1,4-dioxane
as the solvent, and the reaction being carried out at 708C,
the gold catalyst K1 with the pentylisonitrile ligand deliv-
ered only poor yields of the desired product (entry 1; 36%).
The same is true for the acyclic aminocarbene ligand
(entry 2; 20%), the phosphite ligand (entry 3; 33%), and
the bulky (bisadamantyl)-n-butyl-phosphane ligand (entry 4;
31%). Slightly better results were obtained with the Ph3PAu
catalyst K5 (entry 5, 54%). The commercially available IPr
gold catalyst K6 provided product 21a with 81% yield,
probably due to its higher thermal stability under the indi-
cated conditions (entry 6). The SPhos ligand containing
a biaryl-phosphane motif delivered isochromene 21a in an
excellent yield of 98% (entry 7). GoldACTHNUTRGNEUNG(III) chloride turned
&
&
4
Chem. Asian J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

www.chemasianj.org
A. Stephen K. Hashmi et al.
Table 1. Optimization with 20a and different gold catalysts.
Table 2. Optimization with 20a and different counterions.
Entry
1
Catalyst
AgNTf2
5 mol%
Yield of 21a [%][a,b]
Entry
Silver salt
AgX
Yield of 21a [%][a,b]
K1
K2
36
1
2
3
4
–
0
0
0
2
ÀOAc
À
CF3CO2
2
3
5 mol%
5 mol%
20
33
ÀOTs
5
1
K3
K4
6
7
8
9
ÀOTf
ÀONf
65
67
78
85
100
À
NTf2
PF6
SbF6
À
À
4
5 mol%
31
10
[a] Yields were determined by GC/MS in presence of n-dodecane as an
internal standard. [b] The cinnamylic fragment was incorporated in
a branched fashion in all cases.
5
6
K5
K6
5 mol%
5 mol%
54
81
Table 3. Optimization with 20a and different solvents.
Yield of 21a [%][a,b]
7
K7
–
98
Entry
Solvent
1
7
5
6
4
2
3
1,4-Dioxane
MeOH
CH3CN
Benzene
DMF
71
88
85
90
97
98
100
8
9
10
AuCl3
–
p-TsOH
K8
–
–
–
22
0
0
5 mol%
–
[a] Yields were determined by GC/MS in the presence of n-dodecane as
an internal standard. [b] The cinnamylic fragment was incorporated in
a branched fashion in all cases.
Et2O
CH2Cl2
[a] Yields were determined by GC/MS in presence of n-dodecane as an
internal standard. [b] The cinnamylic fragment was incorporated in
a branched fashion in all cases.
(Scheme 7). Secondly, a two-step protocol was applied start-
ing from ortho-iodobenzylic alcohol 23. Etherification, fol-
lowed by a Sonogashira coupling (Scheme 7) delivered the
desired starting materials. The preferred route for varying
the ether substituent was the one according to Scheme 4,
while the alternative path according to Scheme 6 was pref-
erably chosen when the alkyne motif needed to be diversi-
fied.
our optimized catalytic conditions are summarized in
Table 4. Our special attention was focused on the cyclization
manner (5-exo-dig vs. 6-endo-dig) as well as on the binding
properties of the rearranged fragments that could be either
linear (without allylic inversion) or branched (with allylic in-
version).
Based on the easy accessibility of the starting materials,
an array of substrates was prepared within a short time in
overall yields ranging from 27 to 80%. The results of the
successive cyclization–rearrangement sequence induced by
The substrate 20c bearing a simple allylic group was iso-
merized successfully to the expected allylic isochromene de-
rivative 21c with good yield (entry 1, 66%). Excellent yields
were obtained with crotyl- (entry 2, 99%) and prenyl-ethers
(entry 3, 99%). The latter one
caused the formation of a quar-
ternary carbon center after the
completion of the rearrange-
ment. Even with the cyclic allyl-
Scheme 7. Alternative access to a variety of substrates 20.
ic system in compound 20 f, the
5
&
&
Chem. Asian J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
These are not the final page numbers! ÞÞ

www.chemasianj.org
A. Stephen K. Hashmi et al.
Table 4. Scope and limitations of the gold-catalyzed isochromene synthesis with allyl transfer.
result of the cyclization–rear-
rangement sequence was stun-
ning (entry 4, 99%). We fur-
ther tested a wide range of dif-
ferently substituted cinnamylic
ether substrates with varying
electronic and steric properties
at the alkyne terminus. While
the compounds with a phenyl
Entry
1[b,c,f]
Starting material
Catalyst
Producct
Yield [%][a]
66
20c
20d
5 mol%[d]
21c
21d
substituent
such
as
20a
2[e,f]
2 mol%
2 mol%
99
99
(entry 5, 99%) or with the
more electron-rich para-me-
thoxyphenyl substituent (20g,
entry 6, 99%) afforded the cor-
responding isochromenes with
excellent yields, compound 20h
3[e,f]
20e
21e
with
substituent decomposed under
these conditions (entry 7).
a trimethylsilyl (TMS)
4
5
20 f
20a
2 mol%
5 mol%
21 f
21a
99
99
After all, compound 20i with
a terminal acetylene substitu-
ent gave the isochromene 21i
in moderate yields (entry 8,
50%). Despite the fact that the
activation of terminal aryl acet-
ylene moieties by gold catalysts
does not normally favor an en-
docyclic ring closure, the 6-
membered heterocyclic sub-
structure was confirmed unam-
6
7
8
20g
20h
20i
5 mol%
2 mol%
5 mol%
21g
–
99
–
decomposition
1
biguously by H NMR coupling
constant analysis (Figure 2) and
2D-NMR experiments, namely
HSQC(me) and HBMC. Only
when the polarization of the
triple bond was inverted by
a CO2Me substituent, was the
6-endo-dig cyclization mode
suppressed and a complex reac-
21i
50
9
20j
5 mol%
5 mol%
21j
14
67
10[g]
20k
21k
tion
mixture
observed
(entry 9). The only product iso-
lated, with a low yield of 14%,
was 21j, wherein the ring clo-
sure had followed a 5-exo-dig
mode and the rearranged cin-
namylic fragment was rebound-
ed in a linear fashion without
previous allylic inversion. How-
ever, a further olefinic func-
tionality in conjugation with
the alkyne was tolerated and
delivered the corresponding
11[e]
20l
2 mol%
2 mol%
2 mol%
5 mol%
21l
96
68
77
12
20m
20n
20o
21m
21n
21o
13
<5
isochromene 21k in
a good
14
(92)[h]
yield (entry 10; 67%). Com-
pound 20l with an oxygen
atom in the tether reacted se-
lectively to the isochromene
&
&
6
Chem. Asian J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

www.chemasianj.org
A. Stephen K. Hashmi et al.
tives, we turned to benzylic
starting materials (entries 18–
20). Despite the fact that suc-
cessful similar rearrangements
of benzylic ethers have been
described with other substrates
before,[2p,t,v,w,z,ae,af,4] no conver-
sion of any of the tested benzyl-
ic ethers was observed in our
reaction, even at elevated tem-
peratures. The starting material
was recovered in each individu-
al case. A plausible reason for
that behavior will be presented
below in connection with a dis-
cussion of the reaction mecha-
nism.
Even though the connectivity,
the scope, and the limitations
gave us valuable hints concern-
ing a plausible reaction mecha-
nism, we additionally carried
out several experiments to
verify our assumptions. The
spectroscopic data of the puri-
fied products from Table 4
Table 4. (Continued)
Entry
Starting material
Catalyst
5 mol%
Producct
Yield [%][a]
30
15
20p
20q
21p
21q
(99)[h]
57
16
5 mol%
(99)[h]
17
18
20r
20s
5 mol%
5 mol%
21r
28
0
–
–
19[b,c]
20
20t
5 mol%[c]
2 mol%
–
–
–
–
0
0
20u
[a] Isolated yields. [b] The gold catalyst was activated with AgOTf; 1,4-dioxane was used as the solvent. [c] Al-
ternative reaction conditions did not improve the yield. [d] K1 was used as the catalyst. [e] 1,2-Dichloroethane
was used as the solvent. [f] The reaction was carried out at 708C. [g] K6 was used as the catalyst and activation
was done with AgNTf2. [h] The yields in parentheses were determined by integration of 1H NMR signals of
the raw product using hexamethylbenzene as an internal standard.
À
clearly proved the branched C
C bond coupling manner, which
was obviously caused by an al-
lylic inversion rather than by an
entire dissociative fragmenta-
tion. This result is exemplified
by the 1H NMR spectrum of
compound 21b in Figure 1. The
cyclization mode that we as-
sumed before to be 6-endo-dig
rather than 5-endo-dig was also
confirmed representatively by
4
Figure 2. 1H NMR of compound 21i. The specific J-coupling confirms its structure.
derivative 21l with 96% yield (entry 11). Inspired by this
result, we further went on testing other substrates with ether
functionalities. The TBS-protected phenyl ether substituent
(entry 12, 68%) was as well tolerated as the TBS-protected
benzylether substituent (entry 13, 77%). Moreover, a series
of alkyl substituents with terminal TBS ether functionalities
were tested. Different chain lengths were present in these
substrates and once more the isochromenes were formed ex-
clusively (entries 14–17). Since the isolated yields were poor
to moderate, we followed the reaction by continuous NMR
studies. The excellent yields determined by NMR spectro-
copy with hexamethyl benzene as an internal standard (en-
tries 14–16, 92–99%) indicated that the aforementioned
poor results may have originated from problems during the
work-up and purification procedures. All of these com-
pounds accomplished the ring closure selectively over the
cinnamylic ether oxygen—a competing nucleophilic attack
of the silylether oxygen on the triple bond did not occur.
After the extensive study on allylic and cinnamylic deriva-
NMR spectroscopy of compound 21i (Figure 2) and X-ray
crystal structure analysis of compound 21k (Figure 3).[12]
The branched fashion of the cinnamylic fragment is obvious
as well.
Another striking observation was the lack of a C4-unsub-
stituted by-product by protodeauration even if significant
traces of water were present or p-TsOH (5 mol%) was
added to the reaction mixture. This is all the more surprising
as our related isocoumarine synthesis always afforded by-
products from a competing protodeauration.[4] On the other
hand, no effect on the reaction rate was observed by exclu-
sion of moisture using Schlenk-flask techniques, addition of
4 ꢂ molecular sieves, and dry solvents. This means that pro-
tons unlikely take part in the reaction mechanism. Most
likely, the gold-mediated cyclization leads to an organogold
oxonium species B that undergoes an intramolecular [3,3]-
sigmatropic rearrangement to C instead of heterolytic cleav-
age with formation of an ion pair and subsequent migration
of the carbocationic fragment. The catalytic cycle closes
7
&
&
Chem. Asian J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
These are not the final page numbers! ÞÞ

www.chemasianj.org
A. Stephen K. Hashmi et al.
The mechanistic proposal is in accordance with the previ-
ous work of Gagosz et al. who also observed a [3,3]-sigma-
tropic key step for the rearrangement of the allyl moieties
after a gold-catalyzed cyclization step.[2ac]
To provide further evidence of the mechanism, we further
performed a crossover experiment which clearly ruled out
an intermolecular process (Scheme 9). Under the denoted
conditions, compounds 20c and 20v at a ratio of 1:1 were
converted straight into the products 21c and 21v without
any traces of crossover products 21w and 21x as confirmed
by gas chromatography–mass spectrometry (GC–MS).
Furthermore, the proposed mechanism is in accordance
with the observation that benzylic substituents did not lead
to the desired isochromene derivatives at all (Table 4, en-
tries 17–19). For this reaction path, the ability to undergo
a sigmatropic rearrangement rather than the ability to stabi-
lize a positive charge seems to be crucial. In the case of ben-
zylic ether substituents, this kind of sigmatropic rearrange-
ment would be accompanied by a temporary interruption of
the aromatic system (in an intermediate 27) and thus would
be energetically unfavored (Scheme 10; the situation re-
minds of the ortho-Claisen rearrangement, but seemingly
the thermal window of stability of the gold catalyst is not
compatible with the temperatures needed for the subse-
quent [3,3]sigmatropic rearrangement).
Figure 3. X-ray crystal structure of compound 21k confirming the 6-endo-
dig cyclization mode.Thermal ellipsoids are drawn at the 50% probability
level.
with the heterolytic cleavage of the gold–carbon bond,
thereby regenerating the active gold catalyst and forming
the product 21 (Scheme 8).
Conclusions
This mild and efficient approach to isochromene derivatives
by a cyclization–rearrangement strategy using homogeneous
gold catalysis allows the synthesis of specific substitution
patterns. Under optimized conditions, a variety of substrates
with different steric and electronic properties could be con-
verted with excellent yields and selectivity, even if air or
moisture was not excluded. Mechanistic investigations based
on spectroscopic data, control experiments, crossover ex-
periments, and the reaction scope clearly indicated an intra-
molecular mechanism composed of a gold-catalyzed cycliza-
tion step and a [3,3]-sigmatropic rearrangement with formal
inversion of the allylic moiety. To the best of our knowledge,
this is the first example where such a domino reaction is
used for the formation of isochromene derivatives and
where the mechanism is clearly different from the intermo-
lecular allyl transfer in the cor-
Scheme 8. Mechanism proposed for the gold-catalyzed isochromene syn-
thesis.
responding allyl ortho-alkynyl
benzoates.
Experimental Section
Procedures for the synthesis of com-
pounds and characterization data of
compounds are given in the Support-
ing Information.
Scheme 9. Crossover experiment confirming the intramolecular nature of the allyl transfer.
&
&
8
Chem. Asian J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

www.chemasianj.org
A. Stephen K. Hashmi et al.
878–885; ad) J. Y. Cheong, D. Im,
M. Lee, W. Lim, Y. H. Rhee, J.
Org. Chem. 2011, 76, 324–327;
ae) J. Renault, Z. Qian, P. Uriac,
N. Gouault, Tetrahedron Lett.
2011, 52, 2476–2479; af) K. Ko-
meyama, K. Takahashi, K.
Takaki, Org. Lett. 2008, 10, 5119–
5122; ag) P.-Y. Chiang, Y.-C. Lin,
Y. Wang, Y.-H. Liu, Organome-
tallics 2010, 29, 5776–5782.
Scheme 10. A rearrangement of the benzylic substituents with temporary interruption of the aromatic system
does not occur.
[3] For
selected
reviews,
see:
Acknowledgements
a) A. S. K. Hashmi, Gold Bull.
2004, 37, 51–65; b) A. S. K.
This work was supported by the Deutsche Forschungsgemeinschaft (SFB
Hashmi, G. J. Hutchings, Angew. Chem. 2006, 118, 8064–8105;
Angew. Chem. Int. Ed. 2006, 45, 7896–7936; c) R. A. Widenhoefer,
X. Han, Eur. J. Org. Chem. 2006, 4555–4563; d) A. S. K. Hashmi,
Chem. Rev. 2007, 107, 3180–3211; e) A. Fꢃrstner, P. W. Davies,
Angew. Chem. 2007, 119, 3478–3519; Angew. Chem. Int. Ed. 2007,
46, 3410–3449; f) E. Jimꢄnez-NfflÇez, A. M. Echavarren, Chem. Rev.
2008, 108, 3326–3350; g) Z. Li, C. Brouwer, C. He, Chem. Rev.
2008, 108, 3239–3265; h) A. Arcadi, Chem. Rev. 2008, 108, 3266–
3325; i) D. J. Gorin, B. D. Sherry, F. D. Toste, Chem. Rev. 2008, 108,
3351–3378; j) N. T. Patil, Y. Yamamoto, Chem. Rev. 2008, 108,
3395–3442; k) A. S. K. Hashmi, M. Bꢃhrle, Aldrichimica Acta 2010,
43, 27–33; l) A. Corma, A. Leyva-Pꢄrez, M. J. Sabater, Chem. Rev.
2011, 111, 1657–1712.
623); we thank Umicore AG&Co. KG for the generous donation of
noble metal salts.
[1] T. Tsuda, Y. Ohashi, N. Nagahama, R. Sumiya, T. Saegusa, J. Org.
Chem. 1988, 53, 2650–2653.
[2] a) T. Mandai, K. Ohta, N. Baba, M. Kawada, J. Tsuji, Synlett 1992,
671–672; b) S. Cacchi, G. Fabrizi, P. Pace, J. Org. Chem. 1998, 63,
1001–1011; c) S. Cacchi, G. Fabrizi, L. Moro, Synlett 1998, 741–746;
d) N. Monteiro, G. Balme, Synlett 1998, 746–747; e) S. Cacchi, G.
Fabrizi, L. Moro, Tetrahedron Lett. 1998, 39, 5101–5104; f) S.
Kamijo, Y. Yamamoto, J. Am. Chem. Soc. 2002, 124, 11940–11945;
g) A. Fꢃrstner, H. Szillat, F. Stelzer, J. Am. Chem. Soc. 2000, 122,
6785–6786; h) A. Fꢃrstner, F. Stelzer, H. Szillat, J. Am. Chem. Soc.
2001, 123, 11863–11869; i) T. Shimada, I. Nakamura, Y. Yamamoto,
J. Am. Chem. Soc. 2004, 126, 10546–10547; j) I. Nakamura, Y. Miz-
ushima, Y. Yamamoto, J. Am. Chem. Soc. 2005, 127, 15022–15024;
k) A. Fꢃrstner, P. W. Davies, J. Am. Chem. Soc. 2005, 127, 15024–
15025; l) A. Fꢃrstner, E. K. Heilmann, P. W. Davies, Angew. Chem.
2007, 119, 4844–4847; Angew. Chem. Int. Ed. 2007, 46, 4760–4763;
m) I. Nakamura, Y. Mizushima, U. Yamagishi, Y. Yamamoto, Tetra-
hedron 2007, 63, 8670–8676; n) K. Cariou, B. Ronan, S. Mignani, L.
Fensterbank, M. Malacria, Angew. Chem. 2007, 119, 1913–1916;
Angew. Chem. Int. Ed. 2007, 46, 1881–1884; o) I. Nakamura, C. S.
Chan, T. Araki, M. Terada, Y. Yamamoto, Org. Lett. 2008, 10, 309–
312; p) I. Nakamura, T. Sato, Y. Yamamoto, Angew. Chem. 2006,
118, 4585–4587; Angew. Chem. Int. Ed. 2006, 45, 4473–4475; q) I.
Nakamura, U. Yamagishi, D. Song, S. Konta, Y. Yamamoto, Angew.
Chem. 2007, 119, 2334–2337; Angew. Chem. Int. Ed. 2007, 46, 2284–
2287; r) I. Nakamura, T. Sato, M. Terada, Y. Yamamoto, Org. Lett.
2007, 9, 4081–4083; s) F. M. Istrate, F. Gagosz, Org. Lett. 2007, 9,
3181–3184; t) I. Nakamura, T. Sato, M. Terada, Y. Yamamoto, Org.
Lett. 2008, 10, 2649–2651; u) G. Li, X. Huang, L. Zhang, Angew.
Chem. 2008, 120, 352–355; Angew. Chem. Int. Ed. 2008, 47, 346–
349; v) T. Sato, I. Nakamura, M. Terada, Eur. J. Org. Chem. 2009,
5509–5512; w) M. Uemura, I. G. D. Watson, M. Katsukawa, F. D.
Toste, J. Am. Chem. Soc. 2009, 131, 3464–3465; x) T. S. A. Heuge-
baert, C. V. Stevens, Org. Lett. 2009, 11, 5018–5021; y) Y. Shi, K. E.
Roth, S. D. Ramgren, S. Blum, J. Am. Chem. Soc. 2009, 131, 18022–
18023; z) H. Aikawa, S. Tago, K. Umetsu, N. Haginiwa, N. Asao,
Tetrahedron 2009, 65, 1774–1784; aa) M. Ueda, A. Sato, T. Miyoshi,
T. Naito, O. Miyata, Org. Lett. 2010, 12, 2594–2597; ab) I. Naka-
mura, M. Okamoto, T. Sato, M. Terada, Heterocycles 2010, 82, 689–
697; ac) F. M. Istrate, F. Gagosz, Beilstein J. Org. Chem. 2011, 7,
[4] A. S. K. Hashmi, C. Lothschꢃtz, R. Dçpp, M. Ackermann, J.
De Buck, M. Rudolph, C. Scholz, F. Rominger, Adv. Synth. Catal.
2012, 354, 133–147.
[5] For examples, see: a) M. Shamma, D.-Y. Hwang, Tetrahedron 1974,
30, 2279–2282; b) P. L. Majumder, E. Sabzabadi, Phytochemistry
1988, 27, 1899–1901; c) P. L. Majumder, S. Sen, S. Benerjee, Tetrahe-
dron 1999, 55, 6691–6702; d) J. Gripenberg, L. Hiltunen, T. Pakka-
nen, T. Pakkanen, Acta Chem. Scand. 1979, 33b, 6–10; e) J. Gripen-
berg, L. Hiltunen, L. Niinistç, T. Pakkanen, T. Pakkanen, Acta
Chem. Scand. 1979, 33b, 1–5; f) C. P. Ndi, S. J. Semple, H. J. Griess-
er, S. M. Pyke, M. D. Barton, J. Nat. Prod. 2007, 70, 1439–1443.
[6] A. S. Bailey, C. R. Worthing, J. Chem. Soc. 1956, 4535–4543.
[7] J. N. Chatterjea, Chem. Ber. 1958, 91, 2636–2638.
[8] B. Gabriele, G. Salerno, A. Fazio, R. Pittelli, Tetrahedron 2003, 59,
6251–6259.
[9] X. Li, A. R. Chianese, T. Vogel, R. H. Crabtree, Org. Lett. 2005, 7,
5437–5440.
[10] A. S. K. Hashmi, S. Schꢁfer, M. Wçlfle, C. D. Gil, P. Fischer, A.
Laguna, M. C. Blanco, M. C. Gimeno, Angew. Chem. 2007, 119,
6297–6300; Angew. Chem. Int. Ed. 2007, 46, 6184–6187.
[11] For an iodine initiated cyclization, see: a) R. Mancuso, S. Mehta, B.
Gabriele, G. Salerno, W. S. Jenks, R. C. Larock, J. Org. Chem. 2010,
75, 897–901; b) L. Su, C.-Y. Lei, W.-Y. Fan, L.-X. Liu, Synth.
Commun. 2011, 41, 1200–1207.
[12] CCDC 930859 (21k) contains 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.a-
c.uk/data_request/cif.
Received: March 12, 2013
Published online: && &&, 0000
9
&
&
Chem. Asian J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
These are not the final page numbers! ÞÞ