A general synthesis of alkenyl-substituted benzofurans, indoles, and isoquinolones by cascade palladium-catalyzed heterocyclization/oxidative heck coupling

Article abstract of DOI:10.1002/chem.201001535

Structurally diverse C3-alkenylbenzofurans, C3-alkenylindoles, and C4-alkenylisoquinolones are efficiently prepared by using consecutive Sonogashira and cascade Pd-catalyzed heterocyclization/oxidative Heck couplings from readily available ortho-iodosubst

Full text of DOI:10.1002/chem.201001535

DOI: 10.1002/chem.201001535
A General Synthesis of Alkenyl-Substituted Benzofurans, Indoles, and
Isoquinolones by Cascade Palladium-Catalyzed Heterocyclization/Oxidative
Heck Coupling
Rosana ꢀlvarez,*[a] Claudio Martꢁnez,[a] Youssef Madich,[b] J. Gabriel Denis,[a]
Josꢂ M. Aurrecoechea,*[b] and ꢀngel R. de Lera*[a]
Dedicated to Professor Josꢀ Barluenga on the occasion of his 70th birthday
Abstract: Structurally diverse C3-alke-
nylbenzofurans, C3-alkenylindoles, and
C4-alkenylisoquinolones are efficiently
prepared by using consecutive Sonoga-
shira and cascade Pd-catalyzed hetero-
cyclization/oxidative Heck couplings
from readily available ortho-iodosubsti-
tuted phenol, aniline, and benzamide
substrates, alkynes, and functionalized
olefins. The cyclization of O- and N-
heteronucleophiles follows regioselec-
tive 5-endo-dig- or 6-endo-dig-cycliza-
tion modes, whereas the subsequent
Heck-type coupling with both mono-
and disubstituted olefins takes place
stereoselectively with exclusive forma-
tion of the E isomers in most cases.
Keywords:
Heck reaction
cyclization
heterocycles
·
·
·
homogeneous catalysis · palladium
Introduction
based[1b,c,g,k] have been documented as nucleophilic partners,
usually of the NuH type,[3] whereas, in most cases, the partic-
ipating p bond has been the constituent of a C=C double
bond (either alkene or cumulene). In contrast, the use of
Palladium-catalyzed cascade heterocyclization/Heck-type
coupling reactions have found application in recent years in
the synthesis of heterocyclic derivatives.[1] In these reactions
ꢁ
C C triple bonds, as exemplified in Scheme 1, has been re-
PdII usually activates a C C p bond toward nucleopallada-
stricted to a limited use of carbamate or sulfonamide nitro-
gen nucleophiles.[1b,c]
ꢀ
tion with a tethered nucleophile, and the resulting organo-
palladium intermediate then reacts with an alkene through
carbopalladation and b-H elimination steps. External oxi-
dants are required to render the reaction catalytic in PdII by
oxidation of the Pd0 species produced in the last step.[2] Dif-
ferent functionalities, both oxygen[1a,e,f,h–j] and nitrogen
We realized that the tactical combination of a Sonoga-
shira reaction of 2-haloaryl heteroatom derivatives 1–3, fol-
lowed by an oxidative cyclization/Heck cascade, had the po-
tential to become a general and practical approach to the
preparation of alkenyl-substituted heterocyclic derivatives
[a] Dr. R. ꢀlvarez, C. Martꢁnez, J. G. Denis, Prof. Dr. ꢀ. R. de Lera
Departamento de Quꢁmica Orgꢂnica
Facultad de Quꢁmica, Universidade de Vigo
Lagoas Marcosende, 36310 Vigo (Spain)
Fax : (+34)986-811940
E-mail: rar@uvigo.es
qolera@uvigo.es
[b] Y. Madich, Dr. J. M. Aurrecoechea
Departamento de Quꢁmica Orgꢂnica II
Facultad de Ciencia y Tecnologꢁa
Universidad del Paꢁs Vasco
Apartado 644, 48080 Bilbao (Spain)
Fax : (+34)946-013500
E-mail: jm.aurrecoechea@ehu.es
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201001535.
Scheme 1. A cyclization/Heck cascade.
12746
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 12746 – 12753

FULL PAPER
9–12 (Scheme 2). An added benefit of this strategy would be
structural diversity in products 9–12 emanating from varia-
tions in the starting halide- and alkyne-derivatives 1–3 and
was needed for sulfonamide substrate preparation. Other
important shortcomings included a limited scope, particular-
ly for the alkene partner involved in the Heck coupling.[13–15]
Results and Discussion
Representative substrates 5–7 were prepared straightfor-
wardly, as indicated in Scheme 2, from the appropriate
halide and alkyne precursors (1–3 and 4, respectively) under
typical Sonogashira conditions (see the Supporting Informa-
tion).
A survey of reaction conditions was then performed with
representative 2-alkynylphenol substrates in couplings with
n-butyl acrylate (Table 1). Initially, compound 5a was used
under a range of reaction conditions with a common base,
which was thought to be useful as a trap for the acid re-
leased in the cyclization and Heck reactions (Table 1, en-
tries 1–5).
Included here were conditions closely related to those
previously found to be successful with buta-1,2,3-triene car-
binols (Table 1, entry 1)[1i] and a-allenols (Table 1,
entry 2).[1e] However, the yields of product 9a with these
and other combinations of solvent, catalyst, and base
(Table 1, entries 3–5) were not satisfactory. In the last case
Scheme 2. Strategy for the preparation of benzofurans 9, indoles 10, and
isoquinolones 11.
4, respectively, as well as from the use of a variety of alkene
coupling partners 8. To explore these possibilities, we target-
ed the corresponding reactions of 2-alkynylphenols (5;
YH=OH), -anilines (6; YH=NHR’’, R’’=H or alkyl) and
-carboxyamides (7; YH=CONHR’’), whereby formation of
products 9–11 would involve either 5- or 6-endo-dig cycliza-
tions, whereas the alternative 5-exo-dig-process (in the case
of 7) could lead to 12. Because both 5-exo- and 6-endo-cycli-
(Table 1, entry 5), the use of [Pd2ACHTNUGTRENUNG(dba)3] resulted in efficient
formation of benzofuran 13a, the product of cycloisomeriza-
tion of the starting alkynylphenol 5a, and no coupling prod-
uct 9a was detected. On the other hand, inspired by condi-
tions reported for allenoic acids in other oxidative coupling
processes,[16] the use of PdCl2/KI in dimethylformamide
(DMF) proved successful, and benzofuran 9a was obtained
in good yield (Table 1, entry 6). The same reaction condi-
tions worked well for alkynylphenol 5b, which was convert-
ed cleanly into benzofuran 9b in very high yield (Table 1,
entry 7). This product was a more convenient model than
9a, which decomposed partially upon purification, and fur-
ther tests were conducted with 5b. The use of KI was not
mandatory for the cyclization/coupling reaction (Table 1,
entry 9) but the yield was found to improve considerably in
its presence (compare Table 1, entries 7 and 9). The alterna-
tive use of nBu4NI as a nonmetallic iodide salt (Table 1,
entry 8) also benefited the yield, albeit to a lesser extent,
relative to yields obtained in the absence of iodide (Table 1,
entry 9). The possible effect of iodide anions in these reac-
tions is discussed below. Other palladium catalysts, such as
zations have been reported for 2-alkynylbenzamides
7
(YH=CONHR’’) under Pd-catalyzed conditions,[4–6] the
control of the regiochemistry of cyclization became an addi-
tional challenge with these particular substrates.
As a result of those studies,[1j] we now report a cascade
nucleopalladation/Heck process, encompassing 5-endo- and
6-endo-dig cyclizations, leading to structurally diverse 3-al-
kenylbenzofurans (9), 3-alkenylindoles (10), and 4-alkenyli-
soquinolones (11) from Sonogashira products 5–7 and sub-
stituted alkenes 8. Products 9–11 are related to compounds
with interesting biological profiles[7] and, therefore, repre-
sent valuable targets. The synthesis of 3-(3-oxoprop-2-enyl)-
benzofurans and corresponding 4-alkenylisoquinolones has
often been realized by using either Wittig reactions from 3-
formyl-derivatives[8,9] or Heck couplings from the appropri-
ate halo-derivatives.[9,10] In comparison, the one-pot cascade
palladium-catalyzed cyclization/Heck-coupling reported
here (Scheme 2) is a more direct and potentially versatile
route, and this strategy would also benefit from the ready
availability of substrates 5–7, which can be obtained from
Sonogashira-type reactions.[11,12] This type of strategy has al-
ready been employed in the synthesis of 3-alkenylindoles,
but the reported methodology was found to have important
limitations.[1b,c] For example, the amino group had to be suit-
ably protected as a carbamate (YH=NHCOR) or sulfona-
mide (YH=NHSO2R) for successful cyclization and subse-
quent coupling reaction, and an elaborate synthetic route
[PdCl2ACHTNUTRGENN(UG PPh3)2] and PdACHTUNGTREN(NUGN OAc)2 (Table 1, entries 10 and 11)
also proved useful, but yields were somewhat inferior to
those obtained with PdCl2. Under the new sets of conditions,
the use of a base was again found to have deleterious effects
on the coupling, as shown by the result summarized in
Table 1, entry 12. On the other hand, Pd0 catalysts were
much less efficient (Table 1, entries 13 and 14). As expected,
the use of oxidizing conditions was required for efficient cat-
alysis. As a result, when the reaction was carried out under
an Ar atmosphere, the yield of coupling product was consid-
erably reduced (Table 1, entry 15). Excess acrylate was also
Chem. Eur. J. 2010, 16, 12746 – 12753
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
12747

R. ꢀlvarez, J. M. Aurrecoechea, ꢀ. R. de Lera et al.
Table 1. Survey of reaction conditions for the preparation of benzofurans 9a or 9b.[a,b]
the standard 0.5 equivalents to
just 0.2 equivalents was clearly
detrimental to the yield of 11b
(Table 2, entry 15), whereas
doubling the amount was incon-
sequential (Table 2, entry 16).
Similar to the benzofuran case,
control experiments performed
with either variable amounts of
alkene (Table 2, entries 3–6 and
17) or in an Ar atmosphere
(Table 2, entries 8 and 18) con-
firmed the need for excess
alkene and oxygen, respectively,
whereas bases were again found
to inhibit the reaction (Table 2,
entries 9 and 10).
The conditions given in
Table 2, entries 2, 13, and 14
were then taken as standard for
aniline and benzamide sub-
strates, respectively. It is inter-
esting that, in line with the in-
hibitory effect of bases, the re-
action of the aniline substrate
6a required higher temperature
than that of the less basic phe-
nols 5, and for those required
for the benzamide 7 series. Re-
action times for N-phenylbenz-
amides were consistently short-
5
Catalyst [equiv]
5a [Pd(PPh3)4]
5a Pd(OAc)2 (0.1)
Additive/Conditions[b]
Et3N (5), LiCl (4), air[d]
Solvent T [8C] t [h] 9/13
Yield [%][c]
1
2
G
THF
60
RT
20
5
9a
9a
11
20
G
PPh3 (0.2), LiCl (5), K2CO3 (7), DMF
air, Cu
(OAc)2 (2.1)[d]
(PPh3)4] (0.1) Et3N (10), air
AHCTUNGTRENNUNG
3
4
5
6
7
8
9
5a [Pd
5a [Pd
5a
5a PdCl2
5b PdCl2
5b PdCl2
5b PdCl2
G
DMF
toluene RT
RT
8
24
11
20
20
20
20
20
20
20
20
20
20
20
20
20
9a
9a
13a
9a
9b
9b
9b
9b
9b
–
9b
9b
9b
9b/13b
9b/13b
9b/13b
10
A
34
60
62
94
82
76
81
A
N
PPh3 (0.2), Et3N (10), air[d]
KI (0.5), air
KI (0.5), air
Bu4NI (0.5), air
air
KI (0.5), air
KI (0.5), air
Et3N (5), air
KI (0.5), air
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
RT
80
80
80
80
80
80
80
80
80
80
80
80
80
10 5b
11 5b Pd
12 5b Pd
13 5b [Pd
14 5b [Pd
15 5b PdCl2
16 5b PdCl2
17 5b PdCl2
18 5b PdCl2
A
E
72
[e]
–
<50[f]
41[g]
U
PPh3 (0.2), KI (0.5), air
KI (0.5), Ar
20
KI (0.5), air[h]
KI (0.5), air[i]
KI (0.5), air[j]
45/20
42/14
47/20
[a] Unless otherwise stated, 5 mol% of Pd complex and 6 equiv of n-butyl acrylate were used relative to 5.
dba=dibenzylideneacetone. [b] Number of equivalents relative to 5 are given in parentheses. [c] Isolated
yield. [d] n-Butyl acrylate (4 equiv) was used. [e] Degradation. [f] Product was not pure. [g] Starting material
was recovered (56%). [h] n-Butyl acrylate (1 equiv) was used. [i] n-Butyl acrylate (2 equiv) was used. [j] n-
Butyl acrylate (3 equiv) was used.
found to be important, as shown by the effect of lowering
the amount of olefin from 6 to 1–3 equivalents (Table 1, en-
tries 16–18), which had the effect of reducing the yield of
coupling product substantially while also leading to forma-
tion of the cycloisomerization product 13b in moderate
amounts.
er than for the more basic n-butyl analogues (see also
Table 3 below).
The standard set of reaction conditions in each case was
then applied to different alkynyl-substituted phenol, aniline,
and benzamide substrates 5–7 using an acrylate derivative as
standard alkene partner. The results, which are collected in
Table 3, show variations at the terminal alkynyl position
(R1), the alkynyl-substituted aryl ring (R2–R4) and the
amino group (R5), the latter only in the case of anilines and
benzamides.
The reaction conditions summarized in Table 1, entries 6
and 7 were applied to representative 2-alkynylaniline and 2-
alkynylbenzamide substrates to assess the generality of the
reaction (Table 2). Aniline 6a reacted at 808C with compet-
ing formation of cycloisomerization product 14a (Table 2,
entry 1). However, raising the temperature to 1008C pro-
duced a much better yield of coupling product 10a, without
any observable cycloisomerization, under otherwise identical
reaction conditions (Table 2, entry 2). On the other hand,
under the standard conditions, formation of isoquinolones
11 from benzamides 7 took place without apparent interfer-
ence from cycloisomerization, however, the yields were only
moderate (Table 2, entries 11 and 12). In this case, the
simple expedient of changing the catalyst was tried; among
several palladium complexes that were tested, [PdCl2-
As shown in Table 3, substrates with different alkynyl R1
substituents, such as aryl (both electron-poor and electron-
rich) and alkyl groups, provided the expected products in
high yields in most cases. A trimethylsilyl (TMS) group was
also tolerated in the case of phenol and aniline substrates
(Table 3, entries 6 and 12), but the corresponding benzamide
failed to give the desired product (Table 3, entry 16). In the
last case, the substrate was recovered largely unchanged
under the standard conditions at 808C, and underwent de-
composition at higher temperatures with loss of the TMS
group and formation of unidentified products. It is likely
that, in comparison to the phenol and aniline cases, the ad-
ditional amide substituent R5 provided a very crowded envi-
ronment that contributed to the failure of the reaction. On
the other hand, the coupling cascade was tolerant of addi-
tional substituents at the alkynyl-bearing aryl moiety (R2–
ACHTUNGTRENNUNG(PPh3)2] produced the desired improvement, and products
11 were obtained in much higher yields (Table 2, entries 13
and 14). The effect of KI was assessed with benzamide sub-
strate 7a, and found to be much more pronounced than in
the benzofuran case. Thus, lowering the amount of KI from
12748
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 12746 – 12753

Alkenyl-Substituted Ring Systems
FULL PAPER
alkyl, and benzyl amino substituents (R5) were all found to
be successful (Table 3, entries 14–20). In one case (Table 3,
entry 19), a potential protecting group (p-methoxyphenyl)
was introduced at the amino group and the reaction pro-
ceeded with similar efficiency.
Table 2. Test of the standard reaction conditions for aniline and benza-
mide substrates.[a]
Trimethylsilyl-substituted products 9 f and 10 f are precur-
sors of the corresponding 2-unsubstituted benzofuran 15 and
indole 16, which can be generated by protodesilylation
(Scheme 4), offering a convenient alternative to the hypo-
6/7
L
T [8C] Variables[b]
t [h] Product Yield [%][c]
1
2
3
4
5
6
8
9
6a
6a
6a
6a
6a
6a
6a
6a
–
–
–
–
–
–
–
–
–
–
–
80
100
100
100
100
100
100
100
100
80
80
80
80
80
80
80
80
–
–
20
20
20
20
20
20
20
20
4
15
23
16
23
14
14
10a/14a 56/39
10a
10a
85
60
8a (5)
8a (4)
8a (2)
8a (1)
Ar
NaOAc (3)
Et3N (3)
–
10a/14a 58/30
10a/14a 43/43
10a/14a 20/80
14a
14a
–
60
40[d]
Scheme 4. Preparation of 2-unsubstituted benzofuran 15 and indole 16.
TBAF=tetrabutylammonium fluoride.
[e]
10 6a
11 7a
12 7b
13 7a PPh3
14 7b PPh3
15 7a PPh3
16 7a PPh3
17 7a PPh3
18 7a PPh3
–
11a
11b
11a
11b
11a
11a
11a
11a
63
57
78
82
37
79
61
21[f]
–
–
–
thetical use of acetylene as the alkyne component in the So-
nogashira coupling. In the case of the TMS-derivative 10 f,
the synthetic approach described above (Table 3, entry 12)
was, in fact, accompanied by competing formation of the de-
silylated product 16 upon reaction of 6 f with 8a. By intro-
ducing treatment with TBAF into the workup procedure, ex-
clusive formation of 16 was realized (Scheme 4).
Even more structural diversity could be gained with varia-
tions of the olefin substituents. A very extensive study has
been performed by using phenol substrates 5, whereas the
application to anilines 6 and benzamides 7 was more selec-
tive. As shown in Table 4 and Scheme 5, a wide variety of
functional groups were incorporated with the olefin compo-
nent, including carbonyls of different types, such as ester (8c
and 8d), ketone (8e–h), and amide (8i and 8j), as well as
cyano (8k and 8l), sulfonyl (8m), and phenyl groups (8n).
Application of the standard reaction conditions was straight-
forward for most monosubstituted alkenes (R2 =R3 =H;
Table 4, entries 6, 8, 10, 11, and 14–20). One exception was
the reaction of aniline 6a with methyl vinyl sulfone 8m for
which, under the standard conditions, competing formation
of cycloisomerization product 14 (see Table 2) was observed
(34% yield, in addition to 48% of the expected product
KI (0.2)
KI (1)
KI (1), 8b (1) 20
Ar 14
[a] Unless otherwise stated, 5 mol% of Pd complex, 0.5 equiv of KI and
6 equiv of 8a or 8b were used relative to 6 or 7 under an air atmosphere.
[b] Variations with respect to the standard conditions. Number of equiva-
lents relative to 6 or 7 are given in parentheses. [c] Isolated yield.
[d] Starting material (60%) was recovered. [e] Starting material (100%)
was recovered. [f] Starting material (77%) was recovered.
R4), with both electron-donating and electron-withdrawing
groups being well-represented (Table 3, entries 4, 5, 8, 9,
and 20). In one case, a heteroaromatic alkynyl derivative
was tested successfully (Scheme 3). Aniline and benzamide
substrates also allowed additional variations at the amino
group (R5). Thus, besides the unprotected primary anilines
displayed in Table 3, entries 7–12, a secondary amino group
in aniline 6g was also found to afford a similarly good yield
(Table 3, entry 13); a result that considerably expands the
scope of this reaction. As for the benzamide substrates, aryl,
10m). Interestingly, changing the catalyst to [PdCl2ACHTUNGTRENNUNG(PPh3)2]
(Table 4, entry 17) completely suppressed formation of 14,
and the desired product 10m was obtained in excellent
yield. In the case of a,b-unsaturated ketones (Table 4, en-
tries 2–5, 14, and 18), the alternative conjugate addition type
(hydroarylation) product[17] 18 (Scheme 5) was only ob-
served with the benzamide 7b (Table 4, entry 18), albeit in
minor amounts, and even in this case the yield of the Heck
product was high.
Scheme 3. Preparation of 6-azaisoquinolone (a 2,6-naphthyridine-1-one)
11h.
Chem. Eur. J. 2010, 16, 12746 – 12753
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
12749

R. ꢀlvarez, J. M. Aurrecoechea, ꢀ. R. de Lera et al.
Table 3. General preparation of 3-alkenylbenzofurans, 3-alkenylindoles, and 4-alkenylisoquinolones from al-
kynyl substrates 5–7 and n-butyl (8a) or ethyl acrylate (8b).[a]
(Table 4, entry 12) or to change
the catalyst (Table 4, entry 13).
In a few instances, some disub-
stituted alkenes required in-
creased temperatures, and this
had a negative impact on yields
(Table 4, entries 4, 5, and 7).
This was particularly noticeable
with the benzamide substrate
7b, which led mainly to cycloi-
somerization and degradation
products.
Interestingly, in two instan-
ces, formation of a regioisomer
of type 17 (Scheme 5) was ob-
served in reactions of phenol
5b with b-substituted a,b-unsa-
turated carbonyl derivatives
(Table 4, entries 1 and 7). Iso-
mers 17 are probably generated
as a result of double-bond mi-
gration from products 9g and
9m through hydropalladation/
dehydropalladation processes
promoted by HPdX. This palla-
dium species is released in the
Heck coupling and is normally
expected to undergo base-pro-
moted elimination of HX to
form Pd0, which then undergoes
oxidation back to the PdII
needed to initiate the cycliza-
tion/coupling process. However,
the reaction conditions involved
here are base-free and that
pathway is not functional, thus
making it more likely that
HPdX could be available. In
any case, isomerization is prob-
ably triggered by the congested
nature of the trisubstituted,
conjugated products.
5–7
R1
R2
R3
R4
R5
Conditions[a]
9–11
Yield
[%][b]
1
2
3
4
5
6
7
8
5a
5b
5c
5d
5e
5 f
6a
6b
6c
6d
6e
6 f
6g
7a
7b
7c
7d
7e
7 f
7g
n-C6H13
p-Tolyl
n-C6H13
p-Tolyl
p-(CO2Et)C6H4
TMS
p-Tolyl
p-Tolyl
p-Tolyl
n-C6H13
H
H
H
H
H
H
H
H
tBu
H
H
H
H
H
H
H
H
–
–
–
–
–
–
–
–
–
–
–
–
–
H
H
H
H
H
H
Me
Ph
nBu
Ph
Ph
CH2Ph
A
A
A
A[e]
A[f]
A
B
B
B
B
B
9a
9b
9c
9d
9e
9 f
10a
10b
10c
10d
10e
10 f
10g
11a
11b
–
62[c]
91
[d]
–
57[c]
99
CO2Me
tBu
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
75
79
85
76
57
76
H
CO2Me
Me
H
H
H
H
H
H
H
H
H
H
MeO
9
10
11
12
13
14
15
16
17
18
19
20
A
32
TMS
p-Tolyl
Ph
B
B
C
C
72[g]
74
78
Ph
82
[i]
TMS
n-C6H13
Ph
Ph
Ph
C[h]
C
–
11d
11e
11 f
11g
77
85
72
83
C
C
C
A
MeO
[a] Unless otherwise indicated, reactions were conducted with PdII complex (5 mol%), KI (50 mol%), and ac-
rylate ester (6 equiv) in DMF under an air atmosphere for the indicated time. Conditions A: PdCl2, 808C,
20 h. Conditions B: PdCl2, 1008C, 20 h. Conditions C: [PdCl2ACHTNUTRGNEUNG
(PPh3)2], 808C, 21–25 h (R4 =nBu or Bn) or 15–
16 h (R4 =Ar). [b] Isolated yield. [c] Product decomposes partially upon purification. [d] R2, R3 =
-C(Me)2CH2CH2C(Me)2-. [e] Reaction time was 7 h. [f] Reaction conducted at 1008C. [g] Yield of a 1:1 mix-
ture of 10 f and 16 (see Scheme 4). [h] Reaction conducted at 1208C for 21 h. [i] Degradation.
The results shown in Tables 3
and 4 also attest to the high
regio- and stereoselectivity of
Besides monosubstituted alkenes, a range of disubstituted
a,b-unsaturated carbonyl-type derivatives (R2 or R3 ¼H)
was also found to participate effectively (Table 4, entries 1,
3–5, 7, 9, 12, and 13). This is remarkable, because the use of
disubstituted alkenes has often been problematic in Heck
reactions due to low reactivity,[18] and, in fact, remained
almost unreported in oxidative cyclization/Heck-type cou-
plings prior to our work.[1h] In the present case, these al-
kenes have often shown a tendency to react with extensive
degradation, and this has led to the need for some adjust-
ment of the reaction conditions. As a result, in some cases it
was found advantageous to increase the amount of olefin
formation of products 9–11. High isolated yields of isoquino-
lone products 11 were realized without apparent formation
of the alternative isoindolone-type products 12 (see
Scheme 2). Furthermore, exclusive formation of the E iso-
mers was observed in most cases. The exception was acrylo-
nitrile (Table 4, entries 8 and 16), which afforded the prod-
ucts as E/Z mixtures, a result that is in line with related
precedents.[18b,f] No stereochemical conclusions can be
drawn from the reactions of but-2-enenitrile (Table 4,
entry 9), because the starting alkene was already an E/Z
mixture.
12750
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 12746 – 12753

Alkenyl-Substituted Ring Systems
FULL PAPER
Table 4. General preparation of 3-alkenylbenzofurans, 3-alkenylindoles,
and 4-alkenylisoquinolones from alkynyl substrates 5–7 and alkenes 8c–n
(Scheme 5).[a]
Scheme 5. Olefins 8c–n used in the oxidative Heck coupling reactions
and side-products 17–19 obtained in the cyclization/coupling reactions
(Table 4).
8
5–7 Conditions[a] 9–11 R1
R2
R3
Yield [%][b]
70[c]
91
56
31
1
2
3
4
5
6
7
8
9
8c
8e
8 f
8g
5b
5b
5b
A
A
A
9g
9h
9i
CO2Me Me
COMe
COMe
H
H
H
H
H
H
H
H
H
H
H
H
H
Me
A plausible mechanism for this Pd-catalyzed cyclization/
Heck coupling is depicted in Scheme 6. After PdII-promoted
cyclization, intermediate 21 undergoes alkene insertion, fol-
lowed by b-H elimination with release of products 9–11 and
a palladium hydride. Oxidation of the latter then regener-
ates the catalytic PdX2 species.[19] Alternatively, protonation
of intermediates 21 or 22 affords cycloisomerization prod-
ucts 13 or 14, or hydroarylation adduct 18, respectively.
Several observations lend support to this mechanistic in-
terpretation. As discussed earlier, formation of regioisomers
17 in reactions of b-substituted a,b-unsaturated carbonyl de-
rivatives (Table 4, entries 1 and 7) suggests that HPdX is in-
volved in hydropalladation/dehydropalladation processes
with products 9g or 9m and, presumably, also with excess
5b A[d]
9j
G
8h 5b A[d]
9k
9l
N
36
50
8i
8j
5b A[d]
5b A[d]
H
9m
9n
9o
9p
9q
43[e]
60[g]
57[h]
57
8k 5b A[f]
H
Me
H
8l
5b
A
10 8m 5b A[d]
11 8n 5b
12 8c 6a
13 8d 6a
14 8e
15 8i
A
H
60
51
B[i,j]
B[k,l]
B
10h CO2Me Me
10i
10j
10k CONH2
10l CN
10m SO2Me
11i
11j
H
H
H
H
H
H
H
H
Me 67
6a
6a
H
H
H
H
H
H
H
79
B
95
16 8k 6a
17 8m 6a
B[f,i]
B[k]
77[m]
87
18 8e
19 8i
7b C[j]
71[n]
78
7b
C
20 8n 7b C[d]
11k Ph
39[o]
[a] Unless otherwise indicated, reactions were conducted with PdII com-
plex (5 mol%), KI (0.5 equiv), and alkene (6 equiv) in DMF under an air
atmosphere for the indicated time. Conditions A: PdCl2, 808C, 20 h. Con-
ditions B: PdCl2, 1008C, 20 h. Conditions C: [PdCl2ACTHNUGTRNEUG(N PPh3)2], 808C, 21–
23 h. [b] Isolated yield. [c] A 1:1 mixture of 9g and 17a (Scheme 5) was
obtained. [d] Reaction conducted at 1008C. [e] A 1:1.5 mixture of 9m
and 17b (Scheme 5) was obtained. [f] Reaction conducted in a sealed
tube. [g] A 3:1 E/Z mixture was obtained. [h] A 1.2:1 E/Z mixture was
obtained. [i] Reaction conducted with 12 equiv of olefin. [j] Reaction
time: 8 h. [k] [PdCl2ACHTNUTRGNEUG(N PPh3)2] was used as catalyst. [l] Reaction conducted
at 608C. [m] A 13:1 E/Z mixture was obtained. [n] The conjugate addi-
tion product 18 (20%) was also obtained (Scheme 5). [o] A 5:1 mixture
of regioisomers 11k and 19 (Scheme 5) was obtained.
alkene reagent. In fact, addition of HPdX to n-butyl acrylate
has been reported to be involved in regeneration of a cata-
lytically active PdII species in the absence of oxygen, and n-
butyl propionate was found as a byproduct in such “oxida-
tive Heck” reactions.[18f,20] In our reactions, however, oxygen
is required for efficient catalysis and in no case was n-butyl
propionate (or related products) formation observed, there-
Scheme 6. A plausible mechanism for the Pd-catalyzed cyclization/Heck
coupling.
Chem. Eur. J. 2010, 16, 12746 – 12753
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
12751

R. ꢀlvarez, J. M. Aurrecoechea, ꢀ. R. de Lera et al.
fore, ruling out the involvement of excess alkene as a practi-
cal “oxidizing agent”. Nevertheless, excess alkene is impor-
tant in our reactions and could provide, through the same
hydropalladation/dehydropalladation equilibria, control of
the effective concentration of HPdX, as suggested by the re-
sults presented in Tables 1 (entries 16–18) and 2 (entries 4–
6) with reduced amounts of alkene, whereupon cycloisome-
rization products were formed, presumably by either
[HPdX] or HX promoted cyclization.[21]
haloaryl or b-halo-a,b-unsaturated carbonyl derivatives, and
obviates the need for protection–deprotection steps required
with alternative procedures. The cyclization step preceding
oxidative coupling is shown to be effective for 5-endo-dig-
and 6-endo-dig-cyclization modes, is compatible with O- and
N-heteronucleophilic partners, and proceeds with excellent
regioselectivity. The subsequent Heck-type coupling takes
place stereoselectively with exclusive formation of E iso-
mers in most cases, and allows the use of both mono- and
disubstituted alkene partners. Finally, a very high degree of
structural diversity is attainable through simple variations of
the Sonogashira and Heck coupling partners, which also
allows the introduction of useful functionality in the final
products.
Regeneration of an active PdII species is thus likely to
proceed by oxidation of Pd0 released from HPdX, in a pro-
cess that is reported to require the presence of acid.[19,22,23]
In fact, acid is released in both the initial PdII-promoted nu-
cleopalladation and the final product-forming b-hydride
elimination steps. Therefore, the inhibitory effect displayed
by bases in these reactions (evidenced in Tables 1 and 2) is
probably associated with a reduction in the concentration of
acid available for catalyst regeneration; indeed, some indi-
rect evidence indicates that acid is actually consumed in our
reactions. For example, isoquinolones 11 are acid-sensitive
materials, which tend to decompose upon silica gel chroma-
tography (unless this is deactivated with Et3N). In one case,
the decomposition product has been identified as 23e[24]
(Scheme 7), and the conversion of amides 11 into lactones
Experimental Section
General procedure for the cascade cyclization/Heck coupling: The Pd
complex (0.012 mmol), KI (0.02 g, 0.120 mmol), and alkene
8
(1.44 mmol), were added to a solution of 5–7 (0.24 mmol) in DMF
(2 mL), and the mixture was stirred in an air atmosphere under the con-
ditions given in Tables 3 and 4. The mixture was allowed to cool to 258C
and water (in the case of 5–6) or a saturated aqueous solution of
NaHCO3 (in the case of 7) was added. The mixture was extracted with
EtOAc (ꢄ3), the combined organic layers were dried (Na2SO4), and the
solvent was removed. The residue was purified by flash column chroma-
tography under the conditions indicated in the Supporting Information
for the individual cases.
Acknowledgements
We thank the Spanish Ministerio de Educaciꢅn
y Ciencia (Grants
Scheme 7. Formation of lactone 23e from amide 11e.
CTQ2008–06647 and CTQ2009–14186, FEDER), Xunta de Galicia (IN-
BIOMED, contract to C.M.), Universidad del Paꢁs Vasco (fellowship to
M.Y.) and the Universidade de Vigo (fellowship to J.G.D.) for financial
support, and Dr. Efrꢆn Pꢆrez-Santꢁn for substrates 5c and 5e.
23 under GC–MS conditions has also been inferred in other
cases. Nevertheless, high yields of products 11 are obtained
under preparative conditions, after prolonged heating, with-
out detectable formation of 23, indicating a relative lack of
acid. In this context, the additional beneficial effect of
iodide anions in these reactions could also be ascribed to its
participation in the Pd0 oxidation step. Thus, in the presence
of iodide salts under an oxygen atmosphere, iodine would
be formed,[22] and this has been documented as being an oxi-
dizing agent for Pd0.[16]
[1] a) M. F. Semmelhack, W. R. Epa, Tetrahedron Lett. 1993, 34, 7205–
7208; b) A. Yasuhara, M. Kaneko, T. Sakamoto, Heterocycles 1998,
48, 1793–1799; c) A. Yasuhara, Y. Takeda, N. Suzuki, T. Sakamoto,
Chem. Pharm. Bull. 2002, 50, 235–238; d) Q. H. Huang, R. C.
Larock, J. Org. Chem. 2003, 68, 980–988; e) B. Alcaide, P. Almen-
dros, R. Rodrꢁguez-Acebes, Chem. Eur. J. 2005, 11, 5708–5712;
f) L. F. Tietze, F. Stecker, J. Zinngrebe, K. M. Sommer, Chem. Eur.
J. 2006, 12, 8770–8776; g) K.-T. Yip, M. Yang, K.-L. Law, N.-Y. Zhu,
D. Yang, J. Am. Chem. Soc. 2006, 128, 3130–3131; h) F. Yu, X. D.
Lian, S. M. Ma, Org. Lett. 2007, 9, 1703–1706; i) J. M. Aurrecoechea,
A. Durana, E. Pꢆrez, J. Org. Chem. 2008, 73, 3650–3653; j) For pre-
liminary communication of part of this work, see: C. Martinez, R.
Alvarez, J. M. Aurrecoechea, Org. Lett. 2009, 11, 1083–1086;
k) K. T. Yip, N. Y. Zhu, D. Yang, Org. Lett. 2009, 11, 1911–1914.
[2] For reviews of Pd-catalyzed oxidative transformations, see: a) M. S.
Sigman, M. J. Schultz, Org. Biomol. Chem. 2004, 2, 2551–2554;
b) E. M. Beccalli, G. Broggini, M. Martinelli, S. Sottocornola, Chem.
Rev. 2007, 107, 5318–5365; c) K. H. Jensen, M. S. Sigman, Org.
Biomol. Chem. 2008, 6, 4083–4088; d) B. Karimi, H. Behzadnia, D.
Elhamifar, P. F. Akhavan, F. K. Esfahani, A. Zamani, Synthesis 2010,
1399–1427.
Conclusion
The cascade, intramolecular alkyne nucleopalladation/inter-
molecular Heck-type coupling under oxidative conditions is
an effective methodology for the preparation of benzofuran-,
indole-, and isoquinolone-type derivatives, starting from
readily available phenol, aniline, or benzamide substrates
and functionalized alkenes. Furthermore, a combination of
consecutive Sonogashira (in substrate preparation) and oxi-
dative couplings conveniently sidesteps the preparation of
[3] For a report using imine nucleophiles, see ref. [1d].
[4] H. Sashida, A. Kawamukai, Synthesis 1999, 1145–1148.
[5] N. G. Kundu, M. W. Khan, Tetrahedron 2000, 56, 4777–4792.
12752
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 12746 – 12753

Alkenyl-Substituted Ring Systems
FULL PAPER
[6] H. Cao, L. McNamee, H. Alper, Org. Lett. 2008, 10, 5281–5284.
[7] Benzofurans: a) X. L. Hou, Z. Yang, K. S. Yeung, H. N. C. Wong.
Five-Membered Ring Systems: Furans and Benzofurans, in Progress
in Heterocyclic Chemistry, Vol. 21 (Eds.: G. W. Gribble, J. A. Joule),
Elsevier, Oxford, 2009, pp. 179–223, and previous issues in the
series. Indoles: b) J. S. Russel, E. T. Pelkey, S. J. P. Yoon Miller.
Five-Membered Ring Systems: Pyrroles and Benzo Analogs, in Prog-
ress in Heterocyclic Chemistry, Vol. 21 (Eds.: G. W. Gribble, J. A.
Joule), Elsevier, Oxford, 2009, pp. 145–178, and previous issues in
the series. Isoquinolones: c) F. Itoh, S. Kitamura, H. Kimura, H.
Igata (Takeda Pharm. Co.), JP 2002–366257, 2004 [Chem. Abstr.
2004, 141, 71461]; d) R. P. Korivi, C. H. Cheng, Chem. Eur. J. 2010,
16, 282–287, and references therein.
[8] B. Marabout, M. Sevrin, G. Estenne Bouhtou, J. Froissant, P. Magat
(Synthelabo S. A.), FR 2772758, 1999 [Chem. Abstr. 1999, 132,
3325].
[9] K. Ando, Y. Kawamura, Y. Akai, J.-I. Kunitomo, T. Yokomizo, M.
Yamashita, S. Ohta, T. Ohishi, Y. Ohishi, Org. Biomol. Chem. 2008,
6, 296–307.
ugi, T. Satoh, M. Miura, Org. Lett. 2008, 10, 1159–1162; g) L. Dja-
kovitch, P. Rouge, Catal. Today 2009, 140, 90–99.
[14] For metal-catalyzed synthesis of 2,3-disubstituted indoles from 2-al-
kynylanilines and electron-deficient alkenes, see: a) M. Alfonsi, A.
Arcadi, M. Aschi, G. Bianchi, F. Marinelli, J. Org. Chem. 2005, 70,
2265–2273; b) Z. M. Shen, X. Y. Lu, Tetrahedron 2006, 62, 10896–
10899; c) N. Isono, M. Lautens, Org. Lett. 2009, 11, 1329–1331;
d) A. Boyer, N. Isono, S. Lackner, M. Lautens, Tetrahedron 2010, 66,
6468–6482.
[15] For recent reviews on indole synthesis, see: a) S. Cacchi, G. Fabrizi,
Chem. Rev. 2005, 105, 2873–2920; b) M. Bandini, A. Eichholzer,
Angew. Chem. 2009, 121, 9697–9701; Angew. Chem. Int. Ed. 2009,
48, 9533–9537; c) J. Barluenga, F. Rodrꢁguez, F. J. FaÇanꢂs, Chem.
Asian J. 2009, 4, 1036–1048.
[16] S. M. Ma, Z. Q. Yu, Z. H. Gu, Chem. Eur. J. 2005, 11, 2351–2356.
[17] For examples of competing Pd-catalyzed Heck and conjugate addi-
tion-type (hydroarylation) processes with a,b-unsaturated ketones,
see ref. [1i], and references therein.
[18] The use of boronic acids or siloxanes under oxidative conditions
offers an alternative to organic halides in reactions with disubstitut-
ed olefins, see: a) S. E. Denmark, N. Amishiro, J. Org. Chem. 2003,
68, 6997–7003; b) C. H. Yoon, K. S. Yoo, S. W. Yi, R. K. Mishra,
K. W. Jung, Org. Lett. 2004, 6, 4037–4039; c) K. Akiyama, K. Waka-
bayashi, K. Mikami, Adv. Synth. Catal. 2005, 347, 1569–1575;
d) K. S. Yoo, C. H. Yoon, K. W. Jung, J. Am. Chem. Soc. 2006, 128,
16384–16393; e) K. S. Yoo, C. P. Park, C. H. Yoon, S. Sakaguchi, J.
OꢇNeill, K. W. Jung, Org. Lett. 2007, 9, 3933–3935; f) J. Ruan, X. Li,
O. Saidi, J. Xiao, J. Am. Chem. Soc. 2008, 130, 2424–2425.
[19] M. M. Rogers, J. E. Wendlandt, I. A. Guzei, S. S. Stahl, Org. Lett.
2006, 8, 2257–2260.
[10] A. Arcadi, S. Cacchi, G. Fabrizi, F. Marinelli, L. Moro, Synlett 1999,
1432–1434.
[11] Compounds 5–7 have also been employed as starting materials in an
alternative Pd-catalyzed cyclization/coupling with b-halocyclohexe-
nones, but this procedure is limited by the availability of the re-
quired halo-derivatives, see: a) J. H. Chaplin, B. L. Flynn, Chem.
Commun. 2001, 1594–1595; b) D. J. Kerr, A. C. Willis, B. L. Flynn,
Org. Lett. 2004, 6, 457–460; c) M. Nakamura, L. Ilies, S. Otsubo, E.
Nakamura, Angew. Chem. 2006, 118, 958–961; Angew. Chem. Int.
Ed. 2006, 45, 944–947; d) M. Nakamura, L. Ilies, S. Otsubo, E. Na-
kamura, Org. Lett. 2006, 8, 2803–2805; e) A. Arcadi, R. Cianci, G.
Ferrara, F. Marinelli, Tetrahedron 2010, 66, 2378–2383.
[12] For other Pd-catalyzed cyclization/coupling cascade reactions be-
tween 5–7 and organic halides or pseudohalides, see: a) A. Arcadi,
S. Cacchi, M. Del Rosario, G. Fabrizi, F. Marinelli, J. Org. Chem.
1996, 61, 9280–9288; b) S. Cacchi, G. Fabrizi, L. Moro, Synlett 1998,
741–745; c) B. Z. Lu, W. Zhao, H.-X. Wei, M. Dufour, V. Farina,
C. H. Senanayake, Org. Lett. 2006, 8, 3271–3274.
[20] For other examples of trapping of [HPdX] with acrylates, see also:
D. Tanaka, S. P. Romeril, A. G. Myers, J. Am. Chem. Soc. 2005, 127,
10323–10333.
[21] Excess alkene could also be beneficial by acting as a Pd0-stabilizing
ligand and by preventing catalyst decomposition, as previously sug-
gested, see: C. C. Scarborough, S. S. Stahl, Org. Lett. 2006, 8, 3251–
3254.
[13] For alternative Pd-catalyzed syntheses of 3-alkenylindoles, see: a) Y.
Murakami, Y. Yokoyama, T. Aoki, Heterocycles 1984, 22, 1493–
1496; b) A. Takeda, S. Kamijo, Y. Yamamoto, J. Am. Chem. Soc.
2000, 122, 5662–5663; c) Y. Yokoyama, K. Tsuruta, Y. Murakami,
Heterocycles 2002, 56, 525–529; d) N. P. Grimster, C. Gauntlett,
C. R. A. Godfrey, M. J. Gaunt, Angew. Chem. 2005, 117, 3185–3189;
Angew. Chem. Int. Ed. 2005, 44, 3125–3129; e) L. Djakovitch, P.
Rouge, J. Mol. Catal. A 2007, 273, 230–239; f) A. Maehara, H. Tsur-
[22] B. Gabriele, G. Salerno, M. Costa, G. P. Chiusoli, J. Organomet.
Chem. 2003, 687, 219–228.
[23] J. A. Mueller, C. P. Goller, M. S. Sigman, J. Am. Chem. Soc. 2004,
126, 9724–9734.
[24] T. Yao, R. C. Larock, J. Org. Chem. 2003, 68, 5936–5942.
Received: June 2, 2010
Published online: September 21, 2010
Chem. Eur. J. 2010, 16, 12746 – 12753
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
12753