Preparation of 2,3-disubstituted indoles by sequential Larock heteroannulation and silicon-based cross-coupling reactions

Article abstract of DOI:10.1016/j.tet.2008.10.043

A simple and convergent synthesis of 2,3-disubstituted indoles has been developed using a sequential Larock indole synthesis and silicon-based, cross-coupling reaction. Substituted 2-iodoanilines reacted with an alkynyldimethylsilyl tert-butyl ether to af

Full text of DOI:10.1016/j.tet.2008.10.043

Tetrahedron 65 (2009) 3120–3129
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Preparation of 2,3-disubstituted indoles by sequential Larock heteroannulation
and silicon-based cross-coupling reactions
*
Scott E. Denmark , John D. Baird
Roger Adams Laboratory, Department of Chemistry, University of Illinois, Urbana, IL 61801, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 26 August 2008
Received in revised form 11 October 2008
Accepted 16 October 2008
Available online 22 October 2008
A simple and convergent synthesis of 2,3-disubstituted indoles has been developed using a sequential
Larock indole synthesis and silicon-based, cross-coupling reaction. Substituted 2-iodoanilines reacted
with an alkynyldimethylsilyl tert-butyl ether to afford indole-2-silanols under the Larock hetero-
annulation conditions after hydrolysis. The corresponding sodium 2-indolylsilanolate salts successfully
engaged in cross-coupling with aryl bromides and chlorides to afford multi-substituted indoles. The
development of an alkynyldimethylsilyl tert-butyl ether as a masked silanol equivalent enabled a smooth
heteroannulation process and the identification of a suitable catalyst/ligand combination provided for
a facile cross-coupling reaction.
Keywords:
Cross-coupling
Indoles
Ó 2008 Elsevier Ltd. All rights reserved.
Silanols
Palladium catalysis
Alkynes
1. Introduction
N
Substituted indoles are among the most common and important
heterocycles in nature.1 Additionally, the indole motif is repre-
sented within a range of pharmaceutical agents and materials
(Chart 1). In nature, substituted indoles serve various purposes that
range from cell signaling agents and biological function (seratonin
1), to the structural building blocks of proteins (tryptophan 2).
Many indole-containing natural products, including such well-
known indole plant alkaloids such as strychnine 3 and yohimbine 4
have been identified for their biological uses. Of the many types of
substituted indoles, one particular sub-class of interest is those
indoles bearing substitution at the C(2) and C(3) positions, and
several promising therapeutic agents belong to this class. Gonad-
otropin releasing hormone antagonist 5 was identified for the
treatment of developmental disorders.2 Compound 6 is a glycine
receptor antagonist that was identified for the treatment of stroke,3
NH2
NH2
H
HO
CO2H
N
O
H
O
N
H
H
N
H
Serotonin (1)
Tryptophan (2)
Strychnine (3)
N
N
H
N
H
H
NH
H
H3C
H3C CH3
MeO
OH
CH3
N
O
Yohimbine (4)
O
N
H
CH3
and fluvastatin
7 is a drug for the treatment of primary
CH3
hypercholesterolemia.4
NH2
5
F
Given the importance of substituted indoles, it is not surprising
that many methods exist for their formation.5 Some classic
methods6 include the Fischer,7 Leimgruber-Batcho,8 Castro,9
Nenitzescu,10 indole syntheses. Among these, the venerable Fischer
indole synthesis, discovered in 1883, is certainly the most well-
CO2Et
Cl
CO2Et
OH
N
H
N
H
Cl
OH
6
7
CO2H
* Corresponding author. Tel.: þ1 217 333 0066; fax: þ1 217 333 3984.
E-mail address: sdenmark@illinois.edu (S.E. Denmark).
Chart 1.
0040-4020/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2008.10.043

S.E. Denmark, J.D. Baird / Tetrahedron 65 (2009) 3120–3129
3121
known and most commonly employed. However two major
drawbacks usually identified with the Fischer process are: (1) un-
methods offer several advantages over more conventional methods
in view of the low toxicity, ease of preparation and handling, and
stability of the silicon reagents.32
symmetrical ketones give
a mixture of isomeric products,
depending upon the structure of the hydrazine and (2) harsh re-
action conditions are required, typically refluxing in concentrated
acid.11 Because of the milder reaction conditions and defined
placement of substituents, newer methods have recently achieved
prominence such as the reductive cyclization reactions reported by
So¨derberg,12 Cadogan13 and Sundberg.14
Replacement of the simple trimethylsilyl group employed by
Larock with a masked silanol equivalent, such as a silyl ether, would
serve two purposes by: (1) directing the heteroannulation and,
after unmasking the silanol, (2) allowing for a silicon-based cross-
coupling reaction. Furthermore, this strategy could take advantage
of the commercial availability of numerous substituted anilines and
terminal alkynes, as well as the ease of installation and manipu-
lation of silicon-containing compounds.
The successful development of a sequential Larock hetero-
annulation and cross-coupling reaction involved several new
challenges. First, the development of an alkynyl silyl ether that is
stable under the heteroannulation conditions, but easily cleaved
without protodesilylation posed a difficult challenge. This challenge
is particularly important in light of studies that demonstrated that
silyl ethers can be cleaved under basic conditions33 and premature
release of the silanol could lead to undesired side products. Fur-
thermore, the cross-coupling of (2-indolyl)silanols bearing a sub-
stituent at the C(3) position has not been studied. The successful
cross-coupling of a 3-substituted-(2-indolyl)silanol may prove
challenging with some substrates because of the increased steric
congestion near the silicon moiety. Moreover, the steric bulk of
a protecting group on the nitrogen atom may also present a chal-
lenge in the cross-coupling step. Herein we describe in full our
In addition, more modern approaches to the synthesis of di-
substituted indoles have emerged that take advantage of the cur-
rent advances in transition metal chemistry.15 Some newer methods
to construct 2,3-disubstituted indoles include the radical cyclization
of 2-alkenylisocyanides,16 reductive cyclization of acylamido car-
bonyl compounds,17 cyclization of N-(2-halophenyl)allenamides,18
palladium-catalyzed cyclization of 2-alkynyltrifluoroanilines in the
presence of an alkenyl- or aryl halide,19 and palladium-catalyzed
cyclization of 2-iodoanilines with disubstituted alkynes. Among
these, the Larock indole synthesis has emerged as a versatile and
powerful method for the formation of substituted indoles.20
In 1991, Larock and co-workers reported the synthesis of 2,3-
disubstituted indoles by the Pd-catalyzed annulation of 2-iodo-
anilines and disubstituted alkynes (Scheme 1).21 In contrast to
other indole syntheses such as those from Fukuyama, Cacchi, or
Furstner, the Larock procedure does not require a substituent or
a protecting group on the aniline nitrogen. The reaction typically
tolerates a range of functional groups on either partner and the
heteroannulation is highly site selective. In general, the Larock
heteroannulation favors the product in which the bulkier alkyne
substituent resides at the C(2) position of the indole.22 One ex-
ample has appeared where reversed regioselectivity was observed,
but the origin of this process remains unknown.23 Thus, a limitation
to the Larock process arises with alkynes that bear terminal sub-
stituents with similar steric demands. A recent illustration of this
limitation to the Larock indole synthesis was reported where poor
selectivities were observed when alkynes bearing cycloalkyl and
aryl groups were employed.24 Nevertheless, the reaction has been
widely used in industrial (on a multi-kilogram scale) and academic
settings,25 and the transformation has achieved named reaction
status.26
studies on the development of
a sequential Larock hetero-
annulation/silicon-directed cross-coupling process for the synthe-
sis of polysubstituted indoles.
2. Results
2.1. Synthesis of alkynyl silyl ethers
To identify a suitable alkynylsilane for the heteroannulation/
cross-coupling process, we first selected (1-heptynyl)dimethylsilyl
ethers because 1-heptyne is an easily handled, inexpensive alkyne
and different silyl ethers can be readily installed. The silyl group
must withstand the conditions of the heteroannulation and then
undergo hydrolysis to the silanol without protodesilylation under
mild conditions.33 Thus, 1-heptyne was lithiated and the lith-
ioalkyne trapped with dimethyl chlorosilane to afford silyl hydride
9 in 70% yield (Scheme 2, Eq.1). Alcoholysis34 of 9 in the presence of
2 mol % of [RuCl2(p-cymene)]2 in acetonitrile/alcohol provided the
ethoxysilyl ether 10 in 77% and the isopropoxysilyl ether 11 in 86%
yield (Scheme 2, Eqs. 2 and 3). Unfortunately this route failed to
access the tert-butoxysilyl ether, most likely because of the steric
bulk of the alcohol.
RS
RS
RL
I
Pd cat., LiCl
base, DMF, 100 °C
RL
+
R
R
N
NH2
H
Scheme 1.
Larock modified the annulation process to access 3-substituted
indoles by employing silyl-substituted alkynes. In this case, the
bulky silyl group dominates the regioselectivity of the annulation
and thus serves as a phantom directing group in the hetero-
annulation step. Silylated alkynes provide 2-silyl-3-substituted
indoles with excellent regioselectivity. Subsequent desilylation af-
fords 3-substituted indoles in good yield.
Although Larock and co-workers demonstrated the usefulness
of silylated alkynes in this reaction, subsequent transformations of
the resulting 2-silyl indoles have been limited to desilylation,
bromination, or simple Heck-type cross-coupling with alkenes.22
The ability to exploit the silyl group both for site selectivity in the
annulation as well as a functional handle for further synthetic
elaboration would add significant value to this reaction.
Me
1) n-BuLi, hexane
Me
C5H11
H
(eq. 1)
(eq. 2)
C5H11
Si
2) Me2SiHCl
0 °C to rt
H
8
9
70%
Me
Me
Me
Me
[RuCl2(p-cymene)]2
C5H11
Si
C5H11
Si
OEt
EtOH, CH3CN, rt
H
9
10
77%
Me
Me
Me
Me
[RuCl2(p-cymene)]2
i-PrOH, CH3CN, rt
86%
C5H11
Si
(eq. 3)
C5H11
Si
Oi-Pr
We have recently reported the palladium-catalyzed, cross-cou-
pling of sodium 2-indolyldimethylsilanolates with aryl halides27 as
a promising alternative to the classical Stille,28 Negishi,29 and
Suzuki30 cross-coupling reactions. Silicon-based cross-coupling31
H
9
11
Scheme 2.

3122
S.E. Denmark, J.D. Baird / Tetrahedron 65 (2009) 3120–3129
Instead, (1-heptynyl)-tert-butoxysilyl ether 15 was prepared in
production of desired silanol 18 was encouraging, the formation of
undesired siloxane side products prompted the evaluation of other
bases for the heteroannulation process.
two steps from 1-heptyne via the intermediate chlorosilane 13
(Scheme 3). The synthesis of chlorosilane 13 proceeded in an
unoptimized 22% yield by metalation of 1-heptyne with n-BuLi and
trapping with dimethyldichlorosilane (Scheme 3, Eq.1). To examine
the effect of the 3-substituent, chlorosilane 14 bearing a cyclo-
pentyl group was prepared 34% yield (also unoptimized) by met-
alation of cyclopentylacetylene with n-BuLi followed by trapping
the lithioalkyne with dimethyldichlorosilane (Scheme 3, Eq. 1).
Combining chlorosilane 13 with anhydrous t-BuOH in the presence
of triethylamine and DMAP (0.05 equiv) produced the desired silyl
ether 15 in 88% yield whereas chlorosilane 14 provided tert-
butoxysilyl ether 16 in 88% yield (Scheme 3, Eq. 2).
C5H11
Me
C5H11
TBAF (2.0 equiv)
rt, 5 min
+
19
Si OH
Me
N
N
Me
Me
18
20
Scheme 5.
All of the other bases examined (including Na2CO3, Cs2CO3,
NaOAc, Na2HPO4, NaOt-Bu, KOt-Bu, CsOH, and KOTMS) proved in-
ferior to K2CO3 under the conditions in Scheme 4 and led to either
decomposition of 10 or complex reaction mixtures. In the course of
these studies, it became apparent that the irreproducibility of the
reactions could be related to varying moisture levels in the DMF.
To probe the effect of water on this reaction, a series of exper-
iments were conducted that varied the water content in the re-
action from 1.0 to 5.0 equiv relative to aniline (Table 1). DMF from
a Solvent Delivery System (SDS) was used in these experiments (KF
Me
Si
Cl
1) n-BuLi, hexane
2) Me2SICl2
-70 °C to rt
Me
R
H
(eq. 1)
R
R = n-pentyl: 8
R = n-pentyl: 13 (22%)
R = cyclopentyl: 14 (34%)
R = cyclopentyl: 12
Et3N (1.1 equiv)
t-BuOH (1.1 equiv)
DMAP (0.05 equiv)
THF, 0 °C to rt
Me
Me
Me
Me
R
Si
Cl
(eq. 2)
R
Si
Ot-Bu
approximately 40 mg H2O/mL of DMF). The presence of water af-
fected both the conversion of iodoaniline 17 as well as the ratio
between 18 and 19. The addition of 3.0 equiv of water proved op-
timal, providing a 69% conversion of aniline, but an approximate
61:39 ratio in favor of silanol 18 (entry 4). A loading of less than
3.0 equiv of water gave either poor conversion or a greater pro-
portion of 19 (entries 1–3) as did increasing the water loading
above 3.0 equiv (entries 5 and 6).
R = n-pentyl: 13
R = cyclopentyl: 14
R = n-pentyl: 15 (88%)
R = cyclopentyl: 16 (88%)
Scheme 3.
2.2. Optimization of the heteroannulation conditions
To investigate the feasibility of the Larock heteroannulation with
these silyl ethers, 2-iodo-N-methylaniline22 was chosen because
the nitrogen substituent will survive the heteroannulation condi-
tions and the resulting N-methyl(2-indolyl)silanol would be
a suitable starting point for implementing the cross-coupling re-
actions. To evaluate the direct applicability of Larock’s conditions,
a mixture of aniline 17, silyl ether 10 (1.2 equiv), K2CO3 (5.0 equiv),
LiCl (1.0 equiv), Pd(OAc)2 (0.05 equiv), and Ph3P (0.05 equiv) was
heated in DMF at 100 ꢀC for 1.5 h.22 The only products isolated (in
low yield) were 18 and polysiloxane silanol derivatives 19 as an
inseparable mixture in an approximate 76:24 ratio by 1H NMR
analysis (Scheme 4).
Table 1
The effect of water on heteroannulation of 17 with 10
Ph3P (5 mol %)
C5H11 Pd(OAc)2 (5 mol %)
C5H11
I
Me
Si
Me
water 'X' equiv.
+
+ 19
OH
LiCl (1.0 equiv)
K2CO3 (5.0 equiv)
DMF, 100 °C, time
N
Me
NH
Me
Si
Me
OEt
Me
18
17
10
Entry
Water loading, equiv
Conversion, %a (18/19)
1 h
3 h
1
2
3
4
5
6
0
0
0
87
36
27
0
d
0
d
d
C5H11
C5H11
PPh3 (5 mol %)
Pd(OAc)2 (5 mol %)
LiCl (1.0 equiv)
K2CO3 (5.0 equiv)
DMF, 100 °C, 1.5 h
1.0
2.0
3.0
4.0
5.0
d
Trace
86
69
Decomp.
0
I
Me
38:62
66:44
54:46
d
17:87
61:39
d
+
Si OH
Me
N
Me
NH
Me
Si
Me
OEt
d
Me
% Conversion measured by 1H NMR analysis of crude reaction mixtures. GC
a
17
10
18
analysis.
+
C5H11
Me
Si
Because the formation of 19 most likely resulted from the
cleavage of the ethoxysilyl ether we chose to test a more robust silyl
ether would be stable under the heteroannulation conditions. Thus,
the isopropyl silyl ether 11 was tested under the optimized reaction
conditions for the Larock annulation with 17 using K2CO3, LiCl,
Pd(OAc)2, PPh3, and 3.0 equiv of water under vigorous mechanical
stirring (2000 rpm) (Table 2). In contrast to reactions using the
ethoxysilyl ether 10, which gave silanol after silica gel filtration, the
reaction provided the isopropoxysilyl ether after filtration. An ex-
periment conducted at 80 ꢀC established that smooth conversion
could still be obtained within 2 h at lower temperatures. Moreover,
using tert-butoxysilyl ether 15 in conjunction with 17 provided the
tert-butoxysilyl ether after filtration.
Me
Si
Me
O
O
N
Me
n
Me
19
Scheme 4.
Treatment of this mixture with tetrabutylammonium fluoride
(TBAF) afforded indole 20 as judged by TLC and 1H NMR analyses
(Scheme 5). Interestingly, none of the undesired constitutional
isomer, N-methyl-3-(n-pentyl)indole was observed. Although the

S.E. Denmark, J.D. Baird / Tetrahedron 65 (2009) 3120–3129
3123
With a procedure for the clean formation of the silyl ethers,
attention focused on identifying conditions to hydrolyze the silyl
ether. A streamlined process was envisioned wherein the crude 2-
indolylsilyl ether would be hydrolyzed directly to the silanol for
final purification. Thus, a series of acidic and basic conditions were
surveyed in conjunction with isopropoxysilyl ether in DMF and the
reactions were monitored for completion by TLC analysis (Table 2).
filtration and treatment with 0.01 M HCl afforded silanol 18 in 62%
yield following silica gel column chromatography (Scheme 6, Eq. 1).
In our previous studies of cross-coupling of 2-indolyl-, 2-pyr-
rolyl-, 2-thienyl, and 2-furylsilanols, we found that the preformed
silanolate salts were the superior coupling partners.27c Thus, Naþ
18ꢁ was prepared by the dropwise addition of a toluene solution of
18 in to a stirred suspension of NaH in toluene (preformation
protocol).27b The removal of solvent under reduced pressure
afforded Naþ18ꢁ as a semi-solid in 79% yield (Scheme 6, Eq. 2).
Table 2
Survey of conditions for the hydrolysis of 2-indolylsilyl ethers
2.3. Cross-coupling of NaD18L with substituted aryl bromides
Ph3P (5 mol %)
C5H11
Me
1. Pd(OAc)2 (5 mol %)
C5H11
I
The next challenge for this process was to evaluate the effect of
the alkyl substituent at C(3) in the key cross-coupling reaction. The
cross-coupling of Naþ18ꢁ with a range of substituted aryl bromides
was surveyed using palladacycle catalyst 2335 (Table 3). Gratify-
ingly, the presence of the 3-pentylsubstituent did not appear to
hinder the cross-coupling reaction in comparison to the simple
N-methyl-2-indolylsilanol.27c In general, smooth cross-coupling
was observed for electron-deficient as well as electron-neutral aryl
bromides (entries 1 and 2). Sterically encumbered 1-bromonaph-
thalene and electron-rich 4-bromoanisole also coupled successfully
(entries 3 and 5). However, 2-bromotoluene proved sluggish and
provided the desired product in slightly attenuated yield (entry 4).
H2O (3.0 equiv)
LiCl (1.0 equiv)
K2CO3 (5.0 equiv)
+
Si OH
Me
N
Me
NH
Me
OR2
Si
Me
Me
DMF, 100 °C, 2 h
2000 rpm
R = i-Pr: 11
R = t-Bu: 15
17
18
2. hydrolysis conditions
Entry
R
Conditions
Time, h
Resulta
Decomposition
Decomposition
No reaction
1
2
3
i-Pr
i-Pr
i-Pr
10 M acetic acid
2 M acetic acid
0.2 M acetate
buffer pH 5
0.2 M acetate
buffer pH 3
0.8 M HCl
0.5
0.5
6
4
i-Pr
24
10% Hydrolysis
Table 3
5
6
7
8
9
i-Pr
i-Pr
i-Pr
i-Pr
t-Bu
6
6
0.2
6
0.3
Decomposition
Cross-coupling of Naþ18ꢁ with aryl bromides
0.08 M HCl
60% Silanol, 40% decomposition
Complete hydrolysis
No reaction
C5H11
Me
C5H11
0.009 M HCl
0.2 M NaOH
0.001 M HCl
R
23 (2.5 mol %)
Si O-Na+
Me
Complete hydrolysis
toluene, 50 °C, time
a
N
Me
N
Me
Reactions monitored by TLC analysis.
R
Br
-
Na+18
22
These experiments revealed a delicate balance between hy-
drolysis of 21 to silanol 18 and desilylation to N-Me-3-pentylindole
(20). Strongly acidic conditions gave decomposition products ex-
clusively (entries 1, 2, and 5) whereas weak acids gave poor con-
version (entries 3 and 4). However, very dilute HCl provided the
silanol cleanly and quickly (entries 5–7). Basic conditions failed to
provide any cleavage to silanol (entry 8). We were delighted to find
that the tert-butoxysilyl ether could also be converted to the de-
sired silanol with dilute HCl in acetonitrile (entry 9). Thus, both the
isopropoxy- and the tert-butoxysilyl ethers could be used for the
heteroannulation/hydrolysis sequence. In the studies described
below, the choice was dictated by the conditions of the hetero-
annulation; the less reactive anilines that needed more vigorous
conditions mandated the use of the more robust the tert-butoxy-
silyl ether. Conducting the Larock heteroannulation with 11 at 80 ꢀC
under the conditions described above followed by silica gel plug
Cl
(t-Bu)3PPd
23
PdP(t-Bu)3
Me
Entry
R
Time, h
Product
Yield,a
%
1
2
3
4
5
4-CO2Et
H
1
6
3
12
2
22a
22b
22c
22d
22e
73
86
81
65
75
b
2-Me
4-OMe
a
Yield of isolated, analytically pure product.
1-Bromonaphthalene.
b
2.4. Expansion of substrate scope
Ph3P (5 mol %)
Pd(OAc)2 (5 mol %)
C5H11
Me
C5H11
I
The first step in the generalization of the sequential process
involved the identification of a substituent on the nitrogen atom
that is compatible with both steps, but is more easily cleaved than
a methyl group.36 Second, the ability to use 4-substituted anilines
in the reaction would enable the preparation of 2,3,5-trisubstituted
indoles in a highly convergent manner. Third, an expanded scope of
the substituent at C(3) would further demonstrate the generality of
this method and provide another point of diversification in the
preparation of multi-substituted indoles. With these consider-
ations in mind, studies were conducted on a series of N-benzyl-4-
substituted-2-iodoanilines.
1. H2O (3.0 equiv)
+
(eq. 1)
Si OH
Me
LiCl (1.0 equiv)
K2CO3 (5.0 equiv)
DMF, 80 °C, 2 h
N
Me
NH
Me
Si
Me
11
Oi-Pr
Me
17
18
2. 10 mM HCl, 10 min
(62% overall)
C5H11
Me
C5H11
Me
NaH
toluene, 10 min
(eq. 2)
Si OH
Me
Si O-Na+
Me
N
Me
N
Me
N-Benzyl indole derivatives are generally more easily depro-
tected than analogous N-Me derivatives,37 and the electron-rich
nature of the benzyl group should enable smooth cross-coupling
with aryl bromides and chlorides. Toward this end, a series of N-
79%
18
Na+18-
Scheme 6.

3124
S.E. Denmark, J.D. Baird / Tetrahedron 65 (2009) 3120–3129
benzyl-4-substituted-2-iodoanilines bearing -H,38 -Cl,39 and -OMe40
substituents at the 4-position were prepared by following literature
procedures.
Gratifyingly, N-benzyl-4-methoxy-2-iodoaniline engaged in
successful annulation to provide silanol 27 in 62% yield following
hydrolysis (Scheme 8, Eq. 1). In addition, N-benzyl-4-chloro-2-
iodoaniline reacted cleanly with silyl ether 15 and provided silanol
29 in 71% yield (Scheme 8, Eq. 2).
2.4.1. Larock heteroannulation with 4-substituted-N-benzyl-2-
iodoanilines
The next step in the expansion of substrate scope was to ex-
amine the ability to introduce a different alkyl substituent on the
alkyne in the heteroannulation process. The sequential Larock
heteroannulation and silicon-based cross-coupling reaction could
provide a way to selectively access disubstituted indoles bearing
sterically similar groups at C(2) and C(3). For this purpose a bulky
cyclopentyl substituent was introduced on the alkyne. A series of
experiments were conducted that combined silyl ether 16 with
a benzylaniline under the optimized heteroannulation conditions
(Table 4). Whereas electron-rich N-benzyl-4-methoxy-2-iodoani-
line provided the silanol in moderate 59% overall yield (entry 1), the
annulation proceeded smoothly for anilines bearing a chloride or
hydrogen atom (entries 2 and 3). Efforts were next focused on the
crucial cross-coupling reaction.
Orienting studies with N-benzyl-2-iodoaniline established that
this substrate would react under conditions described above for the
N-methyl variant. However, attempts to use either electron-de-
ficient N-benzyl-4-chloro-2-iodoaniline (28) or electron-rich N-
benzyl-4-methoxy-2-iodoaniline (26) under these conditions gave
complex mixtures of decomposition products. It was speculated
that the resulting N-benzyl-5-substituted-3-pentyl-(2-indolyl)iso-
propoxysilyl ethers were decomposing under the reaction condi-
tions. To overcome this problem, recourse was made to the more
robust tert-butoxydimethylsilyl ether.
Thus, a solution of N-benzyl-2-iodoaniline (24), K2CO3, LiCl,
Pd(OAc)2, Ph3P, and 3.0 equiv of water was stirred for 1 h at 100 ꢀC
(Scheme 7). Following treatment with 100 mM HCl (0.04 equiv),
silanol 25 was isolated in 72% yield.
Table 4
Larock heteroannulation of N-benzyl-4-substituted-2-iodoanilines with 16
Ph3P (5 mol %)
Pd(OAc)2 (5 mol %)
C5H11
Me
C5H11
Si
Ph3P (5 mol %)
I
1. H2O (3.0 equiv)
Pd(OAc)2 (5 mol %)
+
Si OH
Me
R
I
R
Me
Si OH
Me
LiCl (1.0 equiv)
K2CO3 (5.0 equiv)
DMF, 100 °C, 2 h
1. H2O (3.0 equiv)
N
Bn
NH
Bn
+
16
Ot-Bu
LiCl (1.0 equiv)
K2CO3 (5.0 equiv)
DMF, 100 °C, 2 h
Me
N
Bn
NH
Bn
Me
24
15
25
2. 100 mM HCl, 2 h
(72% overall)
R = OMe: 30
R = H: 31
2.
10 mM HCl
R = Cl: 32
Scheme 7.
Entry
R
Product
Yield,a
%
The effectiveness of 15 in the heteroannulation reaction was
evaluated with the N-benzyl-4-substituted-2-iodoanilines men-
tioned above under the optimized conditions at 100 ꢀC in DMF.
Although small-scale reactions using magnetic stirring proceeded
to completion cleanly, problems were encountered when these
reactions were executed on greater than 1.5 mmol scale. Upon re-
action scale-up, either incomplete conversion or the formation of
decomposition products was observed. These problems were
solved through the use of a mechanical stirrer that provided strong,
uniform agitation throughout the reaction. Filtration of the crude
reaction mixtures through silica gel or Florisil, and treatment with
HCl afforded the desired silanols (Scheme 8).
1
2
3
OMe
H
Cl
30
31
32
59
71
71
a
Yield of isolated, chromatographed product.
2.4.2. Cross-coupling of substituted sodium N-benzyl-(2-indolyl)-
dimethylsilanolates with aryl bromides and chlorides
To establish the scope of the cross-coupling reaction, several
representative acceptors were selected including electron-rich,
heterocyclic, and electron-deficient aryl chlorides and bromides.
Preliminary experiments using preformed sodium N-benzyl-3-
pentyl-(2-indolyl)dimethylsilanolate (Naþ25ꢁ) established that
smooth cross-coupling could be achieved with electron-deficient
aryl chlorides such as 4-chlorobenzotrifluoride and 4-chlor-
obenzophenone. In the presence of 2.5 mol % of allylpalladium
chloride (APC) dimer and 5 mol % of S-Phos.41 These reactions
completely consumed the aryl halide in 1 h at 70 ꢀC in toluene.
However, 2-chloroanisole provided only partial conversion to
product under these conditions. To address the poor reactivity of
Naþ25ꢁ with less reactive chlorides, a survey was conducted with
several biphenyl phosphine ligands (Chart 2). The cross-coupling of
Naþ25ꢁ with 2-chloroanisole was evaluated in the presence of
2.5 mol % of APC and 5 mol % of ligand (Table 5).
The nature of the ligand had a significant impact on the
efficiency of this cross-coupling reaction. Partial conversion of
2-chloroanisole was observed using 33 or X-Phos derivative 34
(entries 1 and 2). Whereas S-Phos afforded 44% conversion, S-Phos
derivative 36 provided none of the desired product (entries 3 and
4). However, RuPhos 37 proved optimal and provided complete
conversion of chloride in 12 h (entry 5).
Ph3P (5 mol %)
Pd(OAc)2 (5 mol %)
C5H11
Me
MeO
I
MeO
1. H2O (3.0 equiv)
+
15
Si OH
Me
(eq. 1)
LiCl (1.0 equiv)
K2CO3 (5.0 equiv)
DMF, 100 °C, 2 h
N
Bn
NH
Bn
26
27
2. 10 mM HCl (0.05 equiv)
(62% overall)
Ph3P (5 mol %)
C5H11
Pd(OAc)2 (5 mol %)
Cl
I
Cl
Me
1. H2O (3.0 equiv)
+
15
Si OH
Me
(eq. 2)
LiCl (1.0 equiv)
K2CO3 (5.0 equiv)
DMF, 100 °C, 2 h
N
Bn
NH
Bn
28
29
2. 10 mM HCl (0.04 equiv)
(71% overall)
With the discovery that RuPhos enables smooth cross-coupling
with electron-rich aryl chlorides, focus shifted to studying the cross-
Scheme 8.

S.E. Denmark, J.D. Baird / Tetrahedron 65 (2009) 3120–3129
3125
Table 6
CF3
Cross-coupling sodium N-benzyl-5-substituted-3-pentyl-(2-indolyl)dimethylsilanols
with aryl chlorides and aryl bromides
C5H11
C5H11
Me
R1
R2
R1
PCy2
OMe
PCy2
i-Pr
P
i-Pr
CF3
1. NaH / toluene
Si OH
Me
MeO
i-Pr
i-Pr
2. catalyst (2.5 mol %)
ligand (5.0 mol %)
N
Bn
N
Bn
toluene, temp., time
i-Pr
X-Phos (33)
i-Pr
R2
R1 = OMe: 27
R1 = OMe: 39
S-Phos (35)
34
X
R
R
1 = H: 25
1 = Cl: 29
R
R
1 = H: 38
1 = Cl: 40
Entry R1
X
R2
Cat. Ligand Temp, ꢀ
C
Time, h Product Yield,a
%
PCy2
Oi-Pr
Pt-Bu2
OMe
1
2
3
4
5
6
7
8
9
OMe Cl 4-CN
OMe Cl
OMe Cl 2-OMe APC 37
APC 35
APC 37
70
70
70
70
70
70
50
50
50
1
1
24
1
12
12
12
12
24
39g
39h
39f
38g
38h
38f
40i
87
81
71
82
87
82
88
88
72
b
i-PrO
MeO
H
H
H
Cl
Cl
Cl
Cl 4-CN
Cl
Cl 2-OMe APC 37
APC 35
APC 37
b
RuPhos (37)
36
Br 4-CF3
23
23
d
d
d
b
Br
40h
40e
Chart 2.
Br 4-OMe 23
a
Yield of isolated, analytically pure product.
3-Bromopyridine.
b
coupling of sodium N-benzyl-5-substituted-3-pentyl-(2-indolyl)-
dimethylsilanolates. The 3-pentyl-2-indolylsilanols prepared during
the course of this study were converted to their sodium salts with
NaH following the aforementioned general protocol.27c To evaluate
the reactivity of these silanolates, a variety of aryl chlorides and
bromides were chosen as cross-coupling partners (Table 6).
To test the reactivity of the corresponding sodium N-benzyl-5-
substituted-3-cyclopentyl-(2-indolyl)dimethylsilanolates, a series
of aryl chlorides and bromides were surveyed (Table 7).
Cross-coupling 5-methoxy derivative Naþ30ꢁ provided good
yields with electron-deficient, heteroaryl, and electron-rich aryl
chlorides (entries 1–3). The parent Naþ31ꢁ underwent smooth
cross-coupling with an electron-deficient chloride (entry 1), but
reacted more slowly with heteroaryl or electron-rich chlorides to
give products in moderate yields. The cross-coupling of 5-chloro
derivative Naþ32ꢁ proved more challenging and provided the de-
sired products in modest yields (entries 7 and 8). Unfortunately
attempts to engage Naþ32ꢁ with 3-bromopyridine under these
conditions gave only trace amounts of the desired product.
Table 5
Ligand optimization survey for the cross-coupling of preformed Naþ25ꢁ with 2-
chloroanisole
C5H11
Me
C5H11
1. NaH / toluene
Si OH
Me
2. APC (2.5 mol %)
ligand (5.0 mol %)
toluene, 70 °C, 12 h
N
Bn
N
Bn
O
Me
25
38f
Cl
Table 7
Cross-coupling sodium N-benzyl-5-substituted-3-cyclopentyl-(2-indolyl)-
dimethylsilanols with aryl chlorides and aryl bromides
MeO
Entry
Ligand
Conversion,a
%
R1
1
2
3
4
5
33
34
35
36
37
17
54
44
R2
R1
Me
Si OH
Me
1. NaH / toluene
2. catalyst (2.5 mol %)
ligand (5.0 mol %)
N
Bn
N
Bn
d
100
toluene, temp., time
% Conversion determined by 1H NMR analysis of crude reaction mixtures.
R2
a
R1 = OMe: 30
R1 = H: 31
R1 = OMe: 41
R1 = H: 42
X
R1 = Cl: 32
R1 = Cl: 43
Overall, smooth cross-coupling was achieved for a range of
substrates. The 5-methoxy derivative Naþ27ꢁ coupled readily with
electron-deficient, heteroaryl, as well as electron-rich aryl chlo-
rides (entries 1–3). However, reaction with 2-chloroanisole gave
the products in slightly lower yield and required longer reaction
time (entry 3). The parent Naþ25ꢁ afforded products in uniformly
high yields (entries 4–6) although 4-chlorobenzonitrile reached
completion in 1 h in the presence of S-Phos (entry 4). Because 2-
indolylsilanol 29 bears a chloro substituent, Naþ29ꢁ was coupled
with aryl bromides using palladacycle catalyst 23. Whereas cross-
coupling with electron-deficient substrates reached completion in
12 h (entries 7 and 8), electron-rich 4-chloroanisole required 24 h
to reach completion (entry 9).
Entry R1
X
R2
Cat. Ligand Temp, ꢀ
C
Time, h Product Yield,a
%
1
2
3
4
5
6
7
8
OMe Cl 4-CN
OMe Cl
OMe Cl 4-OMe APC 37
H
H
H
Cl
Cl
APC 35
APC 37
70
90
90
70
90
90
50
50
1
12
12
1
41g
41h
41e
42g
42h
42e
43i
88
86
74
89
69
65
64
62
b
Cl 4-CN
Cl
Cl 4-OMe APC 37
Br 4-CF3
APC 35
APC 37
b
12
12
12
12
23
d
Br 4-OMe 23
d
43e
a
Yield of isolated, analytically pure product.
3-Bromopyridine.
b
Although the facile cross-coupling observed with sodium N-
benzyl-5-substituted-3-pentyl-(2-indolyl)dimethylsilanolates was
encouraging, it by no means implied that the cross-coupling of the
more sterically demanding cyclopentyl variant would be assured.
3. Discussion
A successful sequential reaction protocol was developed by
optimization of a number of variables such as the silicon group on

3126
S.E. Denmark, J.D. Baird / Tetrahedron 65 (2009) 3120–3129
the alkyne component in the Larock process, and identification of
an appropriate ligand for the cross-coupling reaction.
Another surprising finding in this study was that the 4-sub-
stituent on the 2-iodoaniline had no noticeable effect on the rate of
the heteroannulation reaction. N-Benzyl-4-methoxy-2-iodoaniline,
N-benzyl-2-iodoaniline, and N-benzyl-4-chloro-2-iodoaniline all
reacted completely within 2 h.
3.1. Optimization of the Larock heteroannulation process
The heteroannulation reaction proceeded under the general
conditions described by Larock and Yum.21 Surprisingly, the re-
action displayed a marked dependence on the base used such that
K2CO3 proved optimal. Successful construction of the indole ring
required only minor optimization of the reaction conditions in-
cluding the addition of water and the use of a tert-butoxysilyl ether.
Additionally, the stirring rate was found to impact the reaction
efficiency when the reaction was executed on greater than
1.5 mmol scale. In addition, optimization studies identified that
irreproducibility in the annulation reaction could be traced to the
variability of moisture in the DMF. A survey of water loading
demonstrated that ca. 3.0 equiv of water was optimal. Using less or
more than 3.0 equiv of water provided either poor conversion or
increased amounts of decomposition products. Although its exact
role remains unknown, the water may help solubilize the salts in
this heterogeneous reaction, or may serve to alter the dielectric
constant of the solution. There is some precedent for the use of
water in the Larock heteroannulation, although its role remains
unclear.25a
The size of the silyl ether group on the alkyne component had
a profound effect on the ability to expand the scope of the heter-
oannulation reaction. The ethoxysilyl ether 10 was the least stable
to the heteroannulation conditions. Potentially, a base-promoted
attack at the silicon center of alkyne 10, or the indole-substituted
silyl ether might induce a fragmentation that could give rise to the
polysiloxane-containing indole 19. Fortunately, switching to the
bulkier isopropoxysilyl ether 11 enabled the facile preparation of
21, and likely better shielded the silicon atom from attack. Further
supporting the increased stability of 21 was the observation that
the heteroannulation product 21 survived filtration through a plug
of silica gel whereas the indole derived from alkyne 10 provided
silanol upon exposure to mildly acidic silica gel.42
3.2. Site selectivity of the reaction
The proposed mechanism of this reaction involves several key
steps (Scheme 9). Larock has suggested that the reaction begins
with coordination of chloride to Pd0 to form an anionic palladium
species that undergoes oxidative addition to the aryl iodide.20 Next,
coordination of the alkyne to the Pd atom occurs (46) followed by
syn migratory insertion. Displacement of the iodide by nitrogen
affords palladacycle 47 that reductively eliminates to give the in-
dole product and regenerate the palladium catalyst. It is worth
noting that this mechanistic proposal is not supported by any ex-
periments and remains speculative.
Pd0
RS
I
Cl-
RL
N
R
NRH
L2PdCl-
48
44
RS
RL
L
L
I
Pd
L
Cl
Pd Cl
L
N
R
NRH
45
47
RL
I
RS
RL
Rs
Pd
Cl
HI
L
However, an unfortunate corollary of using the more stable
isopropoxysilyl ether was the need to develop conditions for its
removal following the heteroannulation. Whereas conditions had
been developed in these laboratories for the facile hydrolysis of
ethoxysilyl ethers,33 the hydrolysis of an isopropoxysilyl ether on
a sensitive indole substrate was more challenging. Gratifyingly,
dilute (0.01 M) HCl afforded the desired silanol cleanly. Crucial to
the success of this operation was the delicate balance of providing
sufficient acid to effect cleavage, but not so much as to induce
protiodesilylation. Generally, 4 mol % of HCl relative to the 2-
indolylsilyl ether delivered from a stock concentration of between
10 and 50 mM gave satisfactory results. Notably, the hydrolysis of
silyl ethers under acidic conditions has been reported to generate
undesired disiloxanes.33 However, none of the corresponding dis-
iloxanes were observed during this process, perhaps because the
bulk of the indole prevented dimerization.
The bulkier tert-butoxysilyl ether 15 was needed to enable the
smooth heteroannulation with less reactive N-benzyl-4-substituted-
2-iodoanilines. Particularly challenging were those anilines bearing
electron-deficient groups at the 4-position. When an isopropoxysilyl
ether was used in these cases, complex mixtures of decomposition
products were observed. Potentially, the electron-withdrawing
substituent inductively rendered the silicon atom more susceptible
to attack by base and subsequent desilylation. Fortunately, the
bulky tert-butoxysilyl ether 15 enabled smooth heteroannulation
with N-benzyl-4-chloro-2-iodoaniline. The heteroannulation of
N-benzyl-4-trifluoromethyl-2-iodoaniline remains an unsolved
challenge and suggests that anilines bearing strongly electron-
withdrawing groups may represent a limitation of this reaction.
NRH
L
46
Scheme 9.
Larock and co-workers have observed exquisite selectivity for
alkynyltrimethylsilanes in these reactions. The high selectivity is
rationalized on the basis of steric interations.22 The migratory in-
sertion is speculated to occur such that the bulkier tert-butoxysilyl
ether substituent on the alkyne is pointing away from the aniline in
the coordination step to form 49 (Scheme 10, Eq. 1). According to
Larock’s proposal, the migratory insertion should occur such that
steric strain near the shorter C–C bond is minimized (Scheme 10,
Eq. 2).20
A
similar hypothesis has been forwarded by Cacchi and
co-workers during their studies on the hydroarylation of silyl-
alkynes.43 Cacchi proposed that the carbopalladation of the alky-
nylsilanes is directed by the steric bulk of the silane to minimize
interaction between the silane and the arene.
3.3. The effect of the nitrogen substituent
The nitrogen substituent demonstrated no apparent effect on
the heteroannulation reaction. Both N-methyl-2-iodoaniline as
well as N-benzyl-2-iodoaniline afforded similar yields of silanols
within similar endpoints of reaction. A firm conclusion on the effect
of the nitrogen substituent on the cross-coupling reaction is diffi-
cult to draw because N-methyl indole variant 18 was cross-coupled
with aryl bromides while the analogous N-benzyl silanol 25 was

S.E. Denmark, J.D. Baird / Tetrahedron 65 (2009) 3120–3129
3127
Me
Me
Me
provide an open coordination site on the palladium catalyst. At this
juncture, it remains difficult to speculate why the RuPhos catalyst is
much better for the cross-coupling of bulky indolyl silanolates with
hindered or electron-rich aryl chlorides in this reaction, but one
possibility may be that the reaction requires an open coordination
site on palladium before either displacement or transmetalation
and this ligand is influencing this process.
Me
Pd
Me
Me
MeMe
O
Me
Me
O
Si
I
Si
Me
Me
Cl
(eq. 1)
PdLnICl
L
NH
R
NRH
49
50
favored
4. Conclusion and outlook
Me
Me
Me
Me
O
Me
Me
Me
O
Me
Me
Me
Si
Si
This study demonstrated the versatility and synthetic power of
silicon reagents. Herein, the silicon atom assumed a dual role; first
as a directing group in the Larock heteroannulation, and later as an
activating group for a silicon-based cross-coupling reaction. This
sequential Larock/cross-coupling strategy required only slight
modification of Larock’s original conditions, including the addition
of water and rapid stirring. Crucial to the success of the reaction
was the development and strategic use of a tert-butoxysilyl ether.
This sequence is complementary to the Cacchi reaction in that this
reaction places the alkyl group on the alkyne at the C(3) position of
the indole and employs a cross-coupling reaction to install a group
at the C(2) position.
Future studies involve expanding the scope of the substituent at
the C(3) position by evaluating other alkynyl silyl ethers in the
heteroannulation process and applying the sequence to 2-amino-3-
iodopyridines to provide access to 7-aza-indoles, a class of bi-
ologically important compounds.46 Application of this reaction to
the synthesis of indole therapeutic agents will be reported in due
course.
I
Me
Me
Pd Cl
(eq. 2)
PdLnICl
L
NH
R
NRH
51
disfavored
52
Scheme 10.
cross-coupled with aryl chlorides. A more concrete comparison will
have to await a full exploration of the scope of this reaction. Po-
tentially, the less sterically hindered N-Me could be expected to be
more reactive a priori.
3.4. Optimization of the cross-coupling process
The cross-coupling of Naþ18ꢁ with aryl bromides proceeded
smoothly using palladacycle catalyst 23 under previously described
conditions.27c The cross-coupling reaction did not reveal a strong
dependence on electronic effects as the electron-deficient ethyl 4-
bromobenzoate required 1 h to reach completion and afforded
a 73% yield of 22a while electron-rich 4-bromoanisole required 2 h
for complete conversion to provide a 75% yield of 22e. More re-
vealing was the apparent effect of increasing steric bulk on the
bromide as 2-bromotoluene required 12 h to reach completion and
gave the product in slightly lower yield. Hindered 1-bromonaph-
thalene required slightly longer time than 4-bromoanisole but less
than 2-bromotoluene. The presence of the n-pentyl group at C(3)
may slow the rate of coupling through steric congestion.
To effect the successful cross-coupling with aryl chlorides,
a careful selection of the ligand on the palladium catalyst was re-
quired. Cross-coupling of N-benzyl-3-n-pentyl-(2-indolyl)silano-
late Naþ25ꢁ proceeded smoothly with electron-deficient aryl
chlorides using the S-Phos ligand described by following the pre-
viously published conditions.27c However, cross-coupling of elec-
tron-rich or sterically encumbered aryl chlorides required the
diisopropyl derivative of S-Phos, RuPhos 37. Moreover, cross-cou-
pling the more sterically encumbered 3-cyclopentyl-2-indolyl
silanolate derivatives with electron-rich aryl chlorides required the
RuPhos ligand in conjunction with higher temperatures (90 ꢀC) to
reach completion.
5. Experimental section
5.1. General experimental methods
See Supplementary data.
5.2. General procedure I. Preparation of dimethyl(1-methyl-
3-pentyl-1H-indol-2-yl)silanol (18) from 15
To a flame-dried, 500-mL, 3-necked (thermal-couple, mechan-
ical stirrer, septum) round-bottomed flask equipped with a me-
chanical stirrer were added 2.073 g (15.0 mmol, 5.0 equiv) of
potassium carbonate, 127 mg (3.0 mmol, 1.0 equiv) of lithium
chloride, 34 mg (0.15 mmol, 0.05 equiv) of palladium acetate, and
39 mg (0.15 mmol, 0.05 equiv) of triphenylphosphine under an at-
mosphere of argon. To this mixture was added 10.0 mL of DMF
(KF¼530
mg/mL). To an oven-dried 8-mL DRAM vial were added
699 mg (3.0 mmol, 1.0 equiv) of N-methyl-2-iodoaniline and
679 mg (3.0 mmol, 1.0 equiv) of tert-butoxy(hept-1-ynyl)dime-
thylsilane under an atmosphere of argon. To this vial was added
2.5 mL of DMF and the resulting solution was transferred to the
500-mL, round-bottomed flask above. The vial was washed with
2.5 mL of DMF and this rinse was added to the round-bottomed
The RuPhos ligand was developed for the cross-coupling of
sterically hindered arylzinc reagents with hindered aryl chlo-
rides,44 and subsequent studies have established its use in C–N
coupling reactions.45 While the exact origin of the success of the
RuPhos catalyst remains unknown, Buchwald and co-workers have
attributed the high reactivities of biphenyl-based ligands to the
formation of a monoligated LPd(0) that is stabilized by a very weak
interaction with the ipso carbon of the arene ring as demonstrated
by X-ray crystallographic and computational studies.41 It has been
suggested that the steric bulk of these ligands facilitates ligand
dissociation, while the weak Pd-ipso carbon is able to stabilize the
Pd catalyst.41 Potentially, the increased bulk provided by the
RuPhos catalyst is better able to facilitate ligand dissociation, and
flask above. To the stirred solution was added 154 mL (8.5 mmol,
2.9 equiv) of water. The flask was submerged in a 105 ꢀC oil bath
(internal temperature 95–100 ꢀC) and stirred (2000 rpm) for 2 h.
After cooling, the crude reaction mixture was eluted through 5.0 g
of SiO2 on a glass frit filter with 150 mL of ethyl acetate into a 500-
mL, round-bottomed flask. The resulting red solution was concen-
trated under reduced pressure to provide a dark-red oil that was
diluted with 100 mL of CH3CN. To this round-bottomed flask were
added a magnetic stir bar and 1.2 mL (0.12 mmol, 0.04 equiv) of
100 mM HCl dropwise. After stirring at rt for 20 min, 90 mL of
benzene was added and the solution was extracted with 90 mL
of satd NH4Cl and washed twice with 90 mL of H2O, and the organic
layer was dried over MgSO4, filtered through #4 Whatman filter

3128
S.E. Denmark, J.D. Baird / Tetrahedron 65 (2009) 3120–3129
paper (15 cm diameter), and concentrated under reduced pressure
to a volume of approx. 20 mL whereupon it was immediately
transferred to a glass frit filter containing 2.5 g of SiO2. The solution
was eluted with 200 mL of EtOAc and concentrated under reduced
pressure. The resulting dark-red oil was purified by silica gel col-
umn chromatography (30ꢂ80 mm) with hexane/EtOAc, 4:1
(15ꢂ40-mL fractions). The combined fractions were further puri-
fied by silica gel column chromatography (30ꢂ 80 mm) by first
eluting with 200 mL of hexane followed by a gradient to hexane/
EtOAc, 9:1 (30ꢂ 8 mL fractions) to afford 477 mg (58%) of 18 as
a light red oil. Attempts to obtain analytically pure material were
unsuccessful, and the silanol was found to contain a small amount
(ca. 10%) of polysiloxane contaminants. This material was found to
provide satisfactory results in subsequent cross-coupling experi-
5.3.1. Preparation 1-methyl-2-(4-methoxyphenyl)-3-pentyl-1H-
indole (22a) (Table 3, entry 1)
Following general procedure II was prepared a solution of
Naþ18ꢁ (357 mg, 1.2 mmol) and toluene (2.0-mL) in a 5-mL round-
bottomed flask under dry argon atmosphere inside a dry box. To this
conical flask were added 23 (17.7 mg, 0.025 mmol) and ethyl 4-
bromobenzoate (163 m
L, 1.0 mmol). After being stirred in a 50 ꢀC oil
bath for 1 h, the crude reaction mixture was filtered through a plug
of silica gel (3 g) and eluted with EtOAc (200 mL) to give a red so-
lution that was concentrated under reduced pressure. The resulting
red oil was purified by silica gel column chromatography
(20ꢂ100 mm) and eluted with hexane/EtOAc, 4:1 (10ꢂ10-mL frac-
tions) gradient to hexane/EtOAc 1:1 (20ꢂ10-mL fractions). The
combined fractions were purified by C18 reverse phase column
chromatography (20ꢂ100 mm) and eluted with MeOH/H2O, 9:1
(30ꢂ10-mL fractions) to afford 257 mg (73%) of 22a as a light yellow
ments. Data for 18: 1H NMR (500 MHz, CDCl3)
d
7.60 (d, J¼7.9, 1H,
HC(4)), 7.30 (d, J¼8.2, 1H, HC(7)), 7.24 (t, J¼7.6, 1H, HC(6)), 7.08 (t,
J¼7.4, 1H, CH(5)), 3.90 (s, 3H, HC(10)), 2.86 (m, 2H, CH(10)), 1.93 (s,
1H, HO(200)), 1.61 (m, 2H, CH(20)), 1.38 (m, 4H, CH(30), CH(40)), 0.91
(t, J¼6.6, 3H, HC(50)), 0.55 (s, 6H, HC(100)); 13C NMR (125.6 MHz,
semi-solid. Data for 22a: 1H NMR (500 MHz, CDCl3)
d
8.22 (d, J¼8.6,
2H, HC(C30)), 7.70 (d, J¼7.9, 1H, HC(C4)), 7.51 (d, J¼8.4, 2H, HC(C20)),
7.38 (d, J¼8.4, 1H, HC(C7)), 7.32 (t, J¼8.1, 1H, HC(C6)), 7.20 (t, J¼8.0,
1H, HC(C5)), 4.47 (q, J¼7.3, 2H, HC(C60)), 3.63 (s, 3H, HC(C10)), 2.75
(t, J¼7.9, 2H, HC(C100)), 1.67 (m, 2H, HC(C200)), 1.48 (t, J¼7.3, 3H,
HC(C70)), 1.30 (m, 4H, HC(C300), HC(C400)), 0.88 (t, J¼7.0, 3H,
CDCl3)
d 139.8 (C(8)), 133.9 (C(9)), 128.1 (C(2)), 126.8 (C(3)), 122.6
(C(6)), 119.3 (C(4)), 118.5 (C(5)), 109.1 (C(7)), 32.9 (C(10)), 32.8
(C(20)), 32.3 (C(30)), 25.7 (C(10)), 22.7 (C(40)), 14.1 (C(50)), 2.8 (C(100));
IR (film) cmꢁ1 3335 (br), 3053 (w), 2956 (s), 2927 (s), 2857 (m),
1653 (w), 1615 (w), 1573 (w), 1503 (m), 1484 (m), 1467 (m), 1423
(m), 1377 (m), 1354 (m), 1316 (m), 1259 (s), 1168 (m), 1141 (w), 1131
(w), 1044 (s), 1013 (m), 832 (s), 803 (s), 738 (s); MS (EI, 70 eV) m/z
275 (4), 218 (14), 207 (18), 201 (23), 158 (3), 144 (100), 115 (4), 77
(4); TLC Rf 0.31 (hexane/EtOAc, 4:1) [silica gel, KMnO4]; HRMS m/z
calcd for C16H25NOSi (Mþ): 275.1705; found: 275.1703.
HC(C500)); 13C NMR (125.6 MHz, CDCl3) 166.4 (C(50)), 137.5 (C(40)),
d
136.9 (C(8)), 136.4 (C(9)), 130.4 (C(20)), 129.6 (C(10)), 129.5 (C(30)),
127.6 (C(2)),122.0 (C(5)),119.3 (C(6)),119.2 (C(4)),114.9 (C(3)),109.4
(C(7)), 61.0 (C(60)), 31.8 (C(300)), 31.8 (C(200)), 30.9 (C(10)), 24.5
(C(100)), 22.4 (C(400)), 14.3 (C(70)), 14.1 (C(500)); IR (Nujol) cmꢁ1 3054
(w), 2956 (s), 2930 (s), 2871 (m), 2857 (m), 1716 (s), 1611 (s), 1468
(s),1366 (s),1273 (s),1177 (s),1107 (s),1019 (s), 870 (m), 770 (m); MS
(EI, 70 eV) 349 (Mþ, 50), 292 (100), 264 (5), 219 (35), 124 (5); TLC Rf
0.32 (hexane/EtOAc, 9:1) [silica gel, UV]. Anal. Calcd for C23H27NO2
(349.47): C, 79.05; H, 7.79; N, 4.01. Found: C, 78.65; H, 7.78; N, 4.17.
5.2.1. Preparation of sodium dimethyl(1-methyl-3-pentyl-1H-indol-
2-yl)silanolate (Naþ18ꢁ)
To a flame-dried, 50-mL, round-bottomed flask equipped with
a magnetic stir bar was added 42.0 mg (1.7 mmol, 1.0 equiv) of NaH
and 2.0 mL of hexane inside a dry box. To this flask was added
dropwise a solution containing 479 mg (1.7 mmol, 1.0 equiv) of 18
in 3.0-mL of hexane in a 25 mL conical flask. The conical flask was
washed with 1.0-mL of hexane and this rinse was added to the
mixture above. After stirring for 20 min, the resulting solution was
concentrated under high vacuum (0.05 mmHg) to afford 411 mg
(79%) of Naþ18ꢁ as a light yellow semi-solid. Data for Naþ18ꢁ: 1H
5.4. General procedure IV: cross-coupling pre-formed sodium
N-benzyl-3,5-disubstituted-(2-indolyl)dimethylsilanolates
with aryl chlorides and aryl bromides (Tables 6 and 7)
To a flame-dried, 5-mL, round-bottomed, flask equipped with
a stir bar were added NaH (29 mg, 1.2 mmol, 1.2 equiv) and toluene
(0.4 mL) under dry argon atmosphere inside a dry box. In a separate
flame-dried, 5-mL, conical flask was added N-benzyl-3,5-di-
substituted-(2-indolyl)dimethylsilanol (1.2 mmol, 1.2 equiv), which
was dissolved in toluene (1.2 mL) and this silanol solution was
added dropwise to the above suspension by glass pipette. The
conical-bottomed flask containing the silanol was washed with
toluene (0.4 mL) and that rinse was added to the reaction mixture.
The resulting mixture was stirred for 10 min before aryl chloride or
bromide (1.0 mmol, 1.0 equiv), Pd catalyst (0.025 mmol,
0.025 equiv), and ligand were added. The flask was sealed with
a rubber septum and removed from the dry box. After being stirred
at the temperature for the time specified, the crude reaction mix-
ture was filtered through a plug of silica gel (3 g) and eluted with
EtOAc (200-mL) to give a red solution that was concentrated under
reduced pressure. The resulting red oil was purified by silica gel
column chromatography (20ꢂ100 mm) and eluted with hexane/
EtOAc, 9:1 (30ꢂ10-mL fractions). The combined fractions were
further purified by C18 reverse phase column chromatography
(20ꢂ100 mm) and eluted with MeOH/H2O, 9:1 (30ꢂ10-mL frac-
tions). The resulting clear, faint yellow oil was placed under re-
duced pressure (0.05 mmHg) for a period of 12 h, or further
purification by recrystallization provided the products.
NMR (500 MHz, CDCl3)
d
7.46 (d, J¼7.8,1H, HC(4)), 7.14 (d, J¼8.3,1H,
HC(7)), 7.03 (t, J¼7.1, 1H, HC(6)), 6.91 (t, J¼7.1, 1H, CH(5)), 4.00 (s,
3H, HC(10)), 2.84 (t, J¼8.3, 2H, CH(10)), 1.61 (m, 2H, CH(20)), 1.42 (m,
4H, CH(30), CH(40)), 0.94 (t, J¼7.1, 3H, HC(50)), 0.34 (s, 6H, HC(100)).
5.3. General procedure II: cross-coupling isolated NaD18L
aryl bromides (Table 3)
To an oven-dried, 5-mL, conical flask equipped with a stir bar
were added sodium N-methyl-(3-pentyl)-2-indolyldimethylsila-
nolate (Naþ18ꢁ) (357 mg,1.2 mmol) and toluene (2.0 mL) under dry
argon atmosphere inside a dry box. To this solution were added 23
(17.7 mg, 0.025 mmol) and 1.0 mmol of aryl bromide. After being
stirred at the temperature for the time specified, the crude reaction
mixture was filtered through a plug of silica gel (3 g) and eluted
with EtOAc (200 mL) to give a red solution that was concentrated
under reduced pressure. The resulting red oil was purified by silica
gel column chromatography (20ꢂ100 mm) and eluted with hex-
ane/EtOAc, 9:1 (30ꢂ10-mL fractions). The combined fractions were
further purified by C18 reverse phase column chromatography
(20ꢂ100 mm) and eluted with MeOH/H2O, 9:1 (30ꢂ10-mL frac-
tions). The resulting clear, faint yellow oil was placed under re-
duced pressure (0.05 mmHg) for a period of 12 h, or further
purification by recrystallization provided the products.
5.4.1. Preparation of 4-(1-benzyl-5-methoxy-3-pentyl-1H-indol-2-
yl)benzonitrile (39g) (Table 6, entry 1)
Following general procedure IV, a solution of Naþ27ꢁ was pre-
pared by combining NaH (29 mg, 1.2 mmol, 1.2 equiv) and 27

S.E. Denmark, J.D. Baird / Tetrahedron 65 (2009) 3120–3129
3129
11. Kurti, L.; Czako, B. Strategic Applications of Named Reactions in Organic Synthesis;
Elsevier: London, 2005; pp 172–173.
12. So¨derberg, B. C.; Chisnell, A. C.; O’Neil, S. N.; Shriver, J. A. J. Org. Chem. 1999, 64,
9731–9734.
13. (a) Cadogan, J. I. G.; Cameron-Wood, M. Proc. Chem. Soc. (London) 1962, 361; (b)
Cadogan, J. I. G. Synthesis 1969, 11–17.
14. (a) Sundberg, R. J. J. Org. Chem. 1965, 30, 3604–3610; (b) Sundberg, R. J.; Tam-
azaki, T. J. Org. Chem. 1967, 32, 290–294.
(458 mg, 1.2 mmol, 1.2 equiv) in toluene (2.0 mL) in a 5-mL round-
bottomed flask under an atmosphere of argon. To this mixture 4-
chlorobenzonitrile (138 mg, 1.0 mmol, 1.0 equiv), S-Phos (20.5 mg,
0.05 mmol, 0.05 equiv), and APC (9 mg, 0.025 mmol, 0.025 equiv)
were added. After being stirred at 70 ꢀC for 1 h, the black, crude
reaction mixture was filtered through a plug of silica gel (3 g) and
eluted with EtOAc (200 mL) to give a light yellow solution that was
concentrated under reduced pressure. The resulting yellow semi-
solid was purified by silica gel column chromatography
(20ꢂ100 mm) and eluted with hexane/EtOAc, 9:1 (30ꢂ10-mL
fractions). The combined fractions were further purified by con-
centration and recrystallization in boiling MeOH (2 mL) to afford
357 mg (87%) of 39g as a white solid (needles). Data for 39g: mp
15. Cacchi, S.; Fabrizi, G. Chem. Rev. 2005, 10
5, 2873–2920.
16. Fukuyama, T.; Chen, X.; Peng, G. J. Am. Chem. Soc. 1994, 116, 3127–3128.
17. Fu¨ rstner, A.; Hupperts, A. J. Am. Chem. Soc. 1995, 117, 4468–4475.
18. Fuwa, H.; Sasaki, M. Org. Biomol. Chem. 2007, 5, 2214–2218.
19. (a) Arcadi, A.; Cacchi, S.; Marinelli, F. Tetrahedron Lett. 1992, 33, 3915–3918; (b)
Lu, B. Z.; Zhao, W.; Wei, H.; Dufour, M.; Farina, V.; Senanayake, C. H. Org. Lett.
2006, 8, 3271–3274.
20. (a) Reviews: Zeni, G.; Larock, R. C. Chem. Rev. 2004, 104, 2285–2309; (b) Larock,
R. C. Top. Organomet. Chem. 2005, 14, 147–182.
21. Larock, R. C.; Yum, E. K. J. Am. Chem. Soc. 1991, 113, 6689–6690.
22. Larock, R. C.; Yum, E. K.; Refvik, M. D. J. Org. Chem. 1998, 63, 7652–7662.
23. Nishikawa, T.; Wada, K.; Isobe, M. Biosci. Biotechnol. Biochem. 2002, 66, 2273–
2278.
24. Roschanger, F.; Liu, J.; Estanove, E.; Dufour, M.; Rodriguez, S.; Farina, V.; Hickey,
E.; Hossain, A.; Jones, P. J.; Lee, H.; Lu, B. Z.; Varsolona, R.; Schroder, J.; Beaulieu,
P.; Gillard, J.; Senanayake, C. H. Tetrahedron Lett. 2008, 49, 363–366.
25. (a) Brodfuehrer, P. R.; Chen, B. C.; Sattelberg, T. R.; Smith, P. R.; Reddy, J. P.; Stark,
D. R.; Quinlan, S. L.; Reid, J. G.; Thottathil, J. K.; Wang, S. J. J. Org. Chem. 1997, 62,
9192–9202; (b) Chen, C.; Lieberman, D. R.; Larsen, R. D.; Reamer, R. A.; Ver-
hoeven, T. R.; Reider, P. J. Tetrahedron Lett. 1994, 35, 6981–6984.
26. Kurti, L.; Czako, B. Strategic Applications of Named Reactions in Organic Synthesis;
Elsevier: London, 2005; pp 260–261.
27. (a) Denmark, S. E.; Baird, J. D. Org. Lett. 2004, 6, 3649–3652; (b) Denmark, S. E.;
Baird, J. D. Org. Lett. 2006, 8, 793–795; (c) Denmark, S. E.; Baird, J. D.; Regens, C.
S. J. Org. Chem. 2008, 73, 1440–1455.
28. (a) Kosugi, M.; Shimizu, Y.; Migita, T. J. Organomet. Chem. 1977, 129, C36–C38;
(b) Farina, V.; Krishnamurthy, V.; Scott, W. Org. React. 1997, 50, 1–652; (c)
Mitchell, T. N. In Metal-catalyzed Cross-coupling Reactions, 2nd ed.; de Meijere, A.,
Diederich, F., Eds.; Wiley-VCH: Weinheim, 2004; Chapter 3.
73–74 ꢀC (MeOH); 1H NMR (500 MHz, CDCl3)
d
7.67 (d, J¼8.0, 2H,
HC(C200)), 7.39 (d, J¼8.0, 2H, HC(C300)), 7.22 (m, 3H, HC(C40), HC(C4)),
7.10 (m, 2H, HC(C50), HC(C7)), 6.87 (m, 3H, HC(C30), HC(C6)), 5.15 (s,
2H, HC(C10)), 3.89 (s, 3H, HC(C10)), 2.67 (t, J¼7.8, 2H, HC(C1%)), 1.62
(t, 2H, HC(C2%)), 1.26 (m, 4H, HC(C300), HC(C400)), 0.84 (t, J¼7.0, 3H,
HC(C5%)); 13C NMR (125.6 MHz, CDCl3) 154.1 (C(5)), 138.0 (C(20)),
d
137.1 (C(100)), 136.2 (C(8)), 132.3 (C(9)), 132.0 (C(200)), 130.9 (C(300)),
128.6 (C(30)), 128.2 (C(2)), 127.2 (C(7)), 125.8 (C(40)), 118.6 (C(500)),
115.7 (C(3)), 112.6 (C(50)), 111.4 (C(400)), 111.1 (C(6)), 101.4 (C(4)), 55.9
(C(10)), 47.7 (C(10)), 31.7 (C(3%)), 30.5 (C(2%)), 24.5 (C(1%)), 22.3
(C(4%)), 14.0 (C(5%)); IR (Nujol)
d 2936 (s), 2860 (s), 1456 (s), 1377
(m), 1224 (w), 1170 (w), 799 (w), 738 (w), 728 (w); MS (EI, 70 eV) m/
z 408 (Mþ, 46), 351 (74), 91 (100), 219 (9), 83 (80); TLC: Rf 0.19
(hexane/EtOAc, 9:1) [silica gel, UV]. Anal. Calcd for C28H28N2O
(408.53): C, 82.32; H, 6.91; N, 6.86. Found: C, 82.18; H, 6.98; N, 6.95.
29. (a) Negishi, E.-i.; King, A. O.; Okukado, N. J. Organomet. Chem. 1977, 42, 1821–
1823; (b) Knochel, P.; Millot, N.; Rodriguez, A. L.; Tucker, C. E. Org. React. 2001,
58, 417–731; (c) Knochel, P. Science of Synthesis; Thieme Verlag: Stuttgart, 2004;
Vol. 3, 5–90; (d) Knochel, P.; Calaza, M. I.; Hupe, E. In Metal-catalyzed Cross-
coupling Reactions, 2nd ed.; de Meijere, A., Diederich, F., Eds.; Wiley-VCH:
Weinheim, 2004; Chapter 11.
30. (a) Miyaura, N.; Yamada, K.; Suginome, A.; Suzuki, A. J. Am. Chem. Soc. 1985, 107,
972–980; (b) Suzuki, A. Pure Appl. Chem. 1994, 66, 213–222; (c) Miyaura, N.;
Suzuki, A. Chem. Rev. 1995, 95, 2457–2483; (d) Hall, D. G. Boronic Acids; Wiley-
VCH: Weinheim, 2005; pp 3–14; (e) Miyaura, N. In Metal-catalyzed Cross-
Coupling Reactions, 2nd ed.; de Meijere, A., Diederich, F., Eds.; Wiley-VCH:
Weinheim, 2004; Chapter 2.
Acknowledgements
We are grateful to the National Institutes of Health for generous
financial support (R01 GM63167). J.D.B. thanks Pfizer, Inc. for
a graduate fellowship.
Supplementary data
Complete experimental details for all preparative procedures
along with full characterization of all starting materials and prod-
ucts as well as optimization experiments (66 pages) are provided.
Supplementary data associated with this article can be found in the
online version, at doi:10.1016/j.tet.2008.10.043.
31. For reviews on silicon-based cross-coupling, see: (a) Hiyama, T. In Metal-cata-
lyzed Cross-coupling Reactions; Diederich, F., Stang, P. J., Eds.; Wiley-VCH:
Weinheim, 1998; Chapter 10; (b) Hiyama, T.; Shirakawa, E. Top. Curr. Chem.
2002, 219, 62–85; (c) Denmark, S. E.; Sweis, R. F. Chem. Pharm. Bull. 2002, 50,
1531–1541; (d) Denmark, S. E.; Sweis, R. F. Acc. Chem. Res. 2002, 35, 835–846;
(e) Denmark, S. E.; Ober, M. H. Aldrichimica Acta 2003, 36, 75–85; (f) Denmark,
S. E.; Sweis, R. F. In Metal-catalyzed Cross-coupling Reactions, 2nd ed.; de Mei-
jere, A., Diederich, F., Eds.; Wiley-VCH: Weinheim, 2004; Chapter 4.
32. Denmark, S. E.; Baird, J. D. Chem.dEur. J. 2006, 12, 4954–4963.
33. Denmark, S. E.; Kallemeyn, J. M. Org. Lett. 2003, 5, 3483–3486.
34. Lee, M.; Ko, S.; Chang, S. J. Am. Chem. Soc. 2000, 122, 12011–12012.
35. Werner, H.; Kuhn, A. J. Organomet. Chem. 1979, 179, 439–445.
36. Kasiotis, K. M.; Haroutounian, S. A. Bioorg. Chem. 2006, 34, 1–14.
37. Greene, T. W.; Wuts, P. G. M. Protective Groups in Organic Synthesis; Wiley-In-
terscience: New York, NY, 1999; pp 615–631.
References and notes
1. Kawasaki, T.; Higuchi, K. Nat. Prod. Rep. 2005, 22, 761–793.
2. Walsh, T. F.; Toupence, R. B.; Ujjainwalla, F.; Young, J. R.; Goulet, M. T. Tetra-
hedron 2001, 57, 5233–5241.
3. Watson, T. J. N.; Horgan, S. W.; Shah, R. S.; Farr, R. A.; Schnettler, R. A.; Nevill, C.
R.; Weiberth, F. J.; Huber, E. W.; Baron, B. M.; Webster, M. E.; Mishra, R. J.;
Harrison, B. L.; Nyce, P. L.; Rand, C. L.; Gorlaski, C. T. Org. Process Res. Dev. 2000,
4, 477–487.
38. Cropper, E. L.; White, A. J. P.; Ford, A.; Hii, K. K. J. Org. Chem. 2006, 71, 1732–1735.
39. McLaughlin, M.; Palucki, M.; Davies, I. W. Org. Lett. 2006, 8, 3307–3310.
40. Lizos, D. E.; Murphy, J. A. Org. Biomol. Chem. 2003, 1, 117–122.
41. Barder, T. E.; Walker, S. D.; Martinelli, J. R.; Buchwald, S. L. J. Am. Chem. Soc. 2005,
127, 4685–4696.
4. Jacotot, B.; Banga, J. D.; Pfister, P.; Mehra, M. Br. J. Clin. Pharmacol. 1994, 38, 257–
263.
5. (a) Brown, R. K. In Indoles Part One; Houlihan, W. J., Ed.; Wiley: New York, NY,
1972; pp 227–559; (b) Sundberg, R. Indoles; Academic: San Diego, 1996; pp
7–53; (c) Humphrey, G. R.; Kuethe, J. T. Chem. Rev. 2006, 106, 2875–2911.
6. Gribble, G. W. J. Chem. Soc., Perkin Trans. 1 2000, 1045–1075.
7. Robinson, B. The Fischer Indole Synthesis; Wiley-Interscience: New York, NY,
1982.
8. Clark, R. D.; Repke, D. H. Heterocycles 1984, 63, 214–221.
9. Castro, C. E.; Stephens, R. D. J. Org. Chem. 1963, 28, 2163.
10. Nenitzescu, C. D. Bull. Soc. Chim. Romania 1929, 11, 37–43.
42. The pH of the silica gel used in these laboratories was found to be slightly acidic
(pH 6.6–7.0).
43. Arcadi, A.; Cacchi, S.; Marinelli, F. Tetrahedron Lett. 1986, 27, 6397–6400.
44. Milne, J. E.; Buchwald, S. L. J. Am. Chem. Soc. 2004, 126, 13028–13032.
45. (a) Zheng, N.; Anderson, K. W.; Huang, X.; Nguyen, H. N.; Buchwald, S. L. Angew.
Chem., Int. Ed. 2007, 46, 7509–7512; (b) Charles, M. D.; Schultz, P.; Buchwald, S.
L. Org. Lett. 2005, 7, 3965–3968.
46. Merour, J.; Benoit, J. Curr. Org. Chem. 2001, 5, 471–506.