2-(2-Haloalkenyl)-aryl halides as substrates for palladium-catalysed tandem C-N bond formation: Efficient synthesis of 1-substituted indoles

Article abstract of DOI:10.1002/adsc.200505484

2-(2-Haloalkenyl)-aryl halides, conveniently prepared in a single step from the corresponding o-halobenzaldehydes, are combined with amines under Pd catalysis to provide 1-substituted indoles. All combinations of Br and Cl leaving groups can be employed,

Full text of DOI:10.1002/adsc.200505484

COMMUNICATIONS
DOI: 10.1002/adsc.200505484
2-(2-Haloalkenyl)-aryl Halides as Substrates for
ꢀ
Palladium-Catalysed Tandem C N Bond Formation: Efficient
Synthesis of 1-Substituted Indoles
Michael C. Willis,a,* Gareth N. Brace,a Thomas J. K. Findlay,a Ian P. Holmesb
a
Department of Chemistry, University of Bath, Bath, BA2 7AY, U.K
Fax: (þ44)-1225-386-231, e-mail: m.c.willis@bath.ac.uk
b
Discovery Research, GlaxoSmithKline, Medicines Research Centre, Gunnels Wood Road, Stevenage, Hertfordshire,
SG1 2NY, U.K.
Received: December 20, 2005; Accepted: February 28, 2006
Supporting Information for this article is available on the WWW under http://asc.wiley-vch.de/home/.
Abstract: 2-(2-Haloalkenyl)-aryl halides, convenient-
ly prepared in a single step from the corresponding
o-halobenzaldehydes, are combined with amines un-
der Pd catalysis to provide 1-substituted indoles. All
combinations of Br and Cl leaving groups can be em-
ployed, and a range of substituents on the arene, al-
kene and amine, can all be tolerated. The use of 1,3-
dichloro-substituted arenes allows a third amination
process to take place; these three-component proc-
esses deliver the corresponding 4-aminoindoles in
good yields.
variety of 1-functionalised indoles. Despite the success
of this method there remained limitations we wished
to address; although the preparation of 2,3-disubstituted
products proceeded efficiently it was difficult to access
the substrates required for the preparation of 2- or 3-
mono-substituted structures. In addition, the alkenyl trif-
late substrates required a two-step synthesis from com-
mercial compounds. In this communication we report
that both these shortcomings can be overcome by the
use of haloalkenylaryl halide substrates.
ꢀ
Keywords: C N bond formation; heterocycles; ho-
mogeneous catalysis; indoles; palladium; tandem re-
actions
ꢀ
Scheme 1. Tandem Pd-catalysed C N route in indoles.
The presence of the indole heterocycle as a recurring
motif in natural products and medicinal agents has re-
The difficulty in accessing the 2- and 3-mono- (or non-)
sulted in sustained interest in the development of new substituted triflates corresponding to 1 stemmed from
methods for the preparation of this valuable structural their poor stability, and we reasoned that a triflate to hal-
unit.[1] Palladium-mediated methods are particularly ide substitution should provide more robust substrates.
prominent amongst these,[2] with the coupling of o-halo- This proved to be the case with the required haloalkenyl-
anilines with alkenes and alkynes remaining as a bench- aryl halides being available in a single step from the cor-
mark system.[3] More recently a number of syntheses responding aldehydes using Wittig methodology.[7] The
based on the coupling of a nucleophilic N unit with large number of commercial o-halobenzaldehydes
aryl and alkenyl halides has been described.[4] In the ma- (>250)[8] was a further attractive feature of this route.
ꢀ
jority of these syntheses C N bond formation is used as To fully exploit the range of commercial starting materi-
the final ring-closing, indole-forming event and takes als we sought a catalyst that could tolerate all combina-
place on an acyclic substrate that incorporates the N tions of chloro and bromo leaving groups.[9]
atom. We recently reported an indole synthesis in which
We selected simple styrenes 2 as suitable test sub-
ꢀ
tandem Pd-catalysed C N bond forming reactions, one strates and evaluated a range of ligands in coupling reac-
at an alkenyl triflate,[5] the second at an aryl halide, are tions between 2 and aniline (Table 1). With the dibromo
used to preparethe indole core (Scheme 1).[6] The ability substrate (2a) the ligand and base combination em-
to introduce an external N unit in the final step of the ployed in our initial alkenyl triflate system (DPEphos
synthesis is particularly useful for the preparation of a 3, Cs2CO3, Scheme 1) was unsuccessful; however, the
Adv. Synth. Catal. 2006, 348, 851 – 856
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
851

Michael C. Willis et al.
COMMUNICATIONS
Table 1. Ligand optimization.[a]
Entry
Substrate[b] X, Y
Ligand
Base
Temp. [8C]
Time [h]
Yield[c] [%]
1
2
3
4
5
6
7
8
9
2a Br, Br
2a Br, Br
2b Cl, Br
2b Cl, Br
2b Cl, Br
2c Br, Cl
2d Cl, Cl
2d Cl, Cl
2a Br, Br
2a Br, Br
2b Cl, Br
3
3
3
4
4
4
4
4
4
5
5
Cs2CO3
100
100
100
100
80
80
80
100
100
80
20
20
20
5
5
5
0
71
25
70
76
80
76
0
NaO-t-Bu
NaO-t-Bu
NaO-t-Bu
NaO-t-Bu
NaO-t-Bu
NaO-t-Bu
Cs2CO3
NaO-t-Bu
NaO-t-Bu
NaO-t-Bu
5
20
20
5
59
85
64
10
11
80
5
[a]
Reaction conditions: substrate (1.0 equiv.), aniline (1.2 equivs.), Pd2(dba)3 (2.5 mol %), ligand (7.5 mol %), base
(2.5 equivs.), toluene.
Z isomers are the major components. See Supporting Information for details.
[b]
[c]
Yields of isolated products.
same ligand used in combination with NaO-t-Bu deliv- the chloro-containing substrates (entry 11). The opti-
ered the required indole in 71% yield (entries 1 and mized conditions for any of the chloro-containing sub-
2). When this combination was applied to the chloroar- strates comprise the dimethylamino-susbtituted ligand
ene-bromoalkene substrate (2b) the indole was deliv- (4) in combination with NaO-t-Bu, while for the dibro-
ered in a modest 25% (entry 3). The development of mo substrate, the dimethoxy-substituted ligand (5) per-
more effective supporting ligands has had a significant forms best.
impact on the efficiency of Pd-catalysed aryl halide ami-
With reagent combinations available for all arrange-
nation reactions,[10] with Buchwaldꢁs biphenyl back- ments of chloro and bromo leaving groups on the simple
bone-based phosphines being particularly effective.[11] test substrate, we next explored the scope of the process
Application of the 2’-dimethylamino-2-dicyclohexyl- (Table 2).[7] We first evaluated the range of N units that
phosphine variant 4 using NaO-t-Bu as base delivered could be introduced; substituted anilines, benzyl- and
the indole in 70% yield (entry 4). This could be in- cyclohexylamine all delivered the expected indoles in
creased to 76% by lowering the reaction temperature good yields (entries 1–5). Access to 1-aminoindoles
to 808C, presumably due to slight product degradation was possible by the use of the protected hydrazine deriv-
occurring at the higher temperature (entry 5).
ative shown in entry 6. Ethyl and tert-butyl carbamates
The same reagent combination was also effective on could also be introduced, allowing potential access to
ꢀ
the bromoarene-chloroalkene, and the dichloride sub- the indole N H systems (entries 7 and 8). It was neces-
strate, delivering the indoles in excellent yields (en- sary to employ the triisopropyl-substituted ligand 6 in
tries 6 and 7). An attempt to use the weaker base order to achieve good conversions with the carbamate
Cs2CO3 was ineffective (entry 8). The use of ligand 4 substrates.
with the original dibromo substrate delivered the indole
The wide availability of variously substituted o-halo-
in only 59% yield (entry 9), however the use of the dime- benzaldehydes allowed a variety of substituents around
thoxy-substituted ligand (5) was more effective and pro- the aryl unit to be introduced (Table 3); using a variety
vided the indole in an excellent 85% yield (entry 10). of dibrominated substrates the 6-methyl-, 5-fluoro-
The dimethoxy-substituted ligand was less effective for and 5,6-dioxolane-substituted indoles were all available
852
asc.wiley-vch.de
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 851 – 856

Efficient Synthesis of 1-Substituted Indoles
COMMUNICATIONS
Table 2. Scope of the N substituent.[a]
Several of the examples in Table 3 suggested that the
geometry of the starting alkenyl halide is unimportant
to the success of the transformation and that an isomer-
isation process was taking place. To investigate this the
two isomers of the vinyl chloride substrates employed
in entry 5 were independently subjected to the reaction
conditions; the separate isomers performed almost iden-
tically (Scheme 2). No isomerisation of the vinyl chlor-
ide substrates was observed, rather isomerisation of
the initially formed enamine intermediates is thought
to account of the funnelling of the isomer mixtures to
a single indole product. A similar isomerisation has
been reported in the preparation of enamides from vinyl
triflates.[12] From a synthetic perspective the ability to
use either geometrical isomer (or mixtures of the two)
of a substrate is attractive as it allows the complications
of producing geometrically pure starting materials to be
avoided.
Entry N Unit
1
Ligand Time [h] Yield [%][b]
5
5
5
6
85
75
2
3
5
3
3
5
20
20
77
70
63
4[c]
5[c]
6
5
6
6
6
10
24
62
68
54
7[d]
8[d]
[a]
Reaction conditions; substrate (1.0 equiv.), N-unit (1.2
equivs.), Pd2(dba)3 (2.5 mol %), ligand (7.5 mol %),
NaO-t-Bu (2.5 equivs.), toluene, 808C. Alkene used as a
9: 2 mixture of Z :E isomers.
Yields of isolated products.
At 1008C.
Scheme 2. Utility of E and Z alkenes.
[b]
[c]
[d]
As well as allowing mixtures of alkene isomers to be
employed as substrates, the greater reactivity of the al-
kenyl halide site to the amination conditions allowed se-
quential amination processes to be explored. For exam-
ple, the use of an extra equivalent of aniline in the cou-
At 1108C, Cs2CO3 as base, dioxane as solvent.
in good yields (entries 1–3). The 4-chloroindole origi- pling employing the 1,3-dichloro substrate allowed in-
ꢀ
nated from the corresponding 1,3-dichloro-arene-bro- dole formation to be followed by a third C N bond for-
moalkene substrate (entry 4). Aryl chloride-substituted mation to introduce an aniline substituent at the indole
ketones (as opposed to aldehydes) also served well as 4-position (Scheme 3). More usefully, provided thirty
substrates, with the methyl and phenyl ketones deliver- minutes had elapsed from the start of the reaction, a sec-
ing the corresponding 3-substituted indoles as expected ond amine could be successfully introduced. For exam-
(entries 5 and 6).
ple, the 5-morpholino- and 5-N-methyl-aniline substi-
2-Substituted indoles could be accessed using two tuted N-phenylindoles were both obtained in good
methods; the first involved the use of a substituted Wit- yields (Scheme 3).
tig reagent to prepare the methyl-substituted cyclisation
In conclusion, we have demonstrated that the haloal-
substrate (entry 7). The second employed an initial Su- kenylaryl halides are efficient substrates for tandem Pd-
ꢀ
zuki coupling on the corresponding 1,1-dibromoalkene catalysed C N bond formation and deliver a variety of
to introduce a 2-aryl-substituent on the indole precursor N-functionalised indoles in good yields. The majority
(entries 8 and 9).[7] The substrates prepared using either of substrates used in these reactions are available in a
route delivered the indoles in an efficient manner. The single step from commercial compounds. The use of
lower yield obtained in entry 9 is compensated for by readily available starting materials allows access to 1-,
the presenceof the synthetically useful2-(4-chlorophen- 2-, 3-, 4-, 5- and 6-substituted indoles. By careful choice
yl)-substituent in the product.[9]
of the halide leaving group it is possible to construct se-
Adv. Synth. Catal. 2006, 348, 851 – 856
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
853

Michael C. Willis et al.
COMMUNICATIONS
Table 3. Substrate scope.[a]
Entry
1
Substrate
Z :E[b]
Ligand
Product
Yield [%][c]
73
9: 1
5
2
3
9: 1
4: 1
5
5
59
67
4[d]
1: 20
4
80
5[d]
1: 1
7: 2
4
4
78
94
6[d]
7
1: 1
5
5
79
81
8[d]
20: 1
9[d]
20: 1
5
51
[a]
Reaction conditions; substrate (1.0 equiv.), aniline (1.2 equivs.), Pd2(dba)3 (2.5 mol %), ligand (7.5 mol %), NaO-t-Bu
(2.5 equivs.), toluene, 808C.
Determined by H NMR spectroscopy.
Yield of isolated products.
At 1008C.
[b]
[c]
[d]
1
854
asc.wiley-vch.de
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 851 – 856

Efficient Synthesis of 1-Substituted Indoles
COMMUNICATIONS
126.4, 127.9, 129.3, 129.6, 135.8, 139.8. Data in accordance
with those reported in the literature.[13]
quential reactions to introduce further amine function-
ality via a third C N bond formation. Studies applying
similar strategies to the synthesis of alternative hetero-
cycles are under way.
ꢀ
Acknowledgements
This work was supported by the EPSRC and GlaxoSmithKline.
The EPSRC Mass Spectrometry Service at the University of
Wales Swansea is also thanked for their assistance.
References and Notes
[1] Indole reviews: a) R. J. Sundberg, Indoles, Academic
Press, London, 1996; b) G. W. Gribble, J. Chem. Soc. Per-
kin Trans. 1 2000, 1045.
[2] a) J. J. Li, G. W. Gribble, Palladium in Heterocyclic
Chemistry, Pergamon, Amsterdam, 2000; b) I. Naka-
mura, Y. Yamamoto, Chem. Rev. 2004, 104, 2127; c) S.
Cacchi, G. Fabrizi, Chem. Rev. 2005, 105, 2873.
[3] For reviews, see: a) S. Cacchi, F. Marinelli, in: Handbook
of Organopalladium Chemistry for Organic Synthesis,
(Ed.: E.-I. Negishi), Wiley, New York, 2002, Vol. 2,
p. 2227; b) G. Zeni, R. C. Larock, Chem. Rev. 2004,
104, 2285.
[4] Selected recent examples: a) K. Aoki, A. J. Peat, S. L.
Buchwald, J. Am. Chem. Soc. 1998, 120, 3068; b) S. Wa-
gaw, B. H. Yang, S. L. Buchwald, J. Am. Chem. Soc.
1999, 121, 10251; c) J. A. Brown, Tetrahedron Lett.
2000, 41, 1623; d) K. Yamazaki, Y. Nakamura, Y. Kondo,
J. Org. Chem. 2003, 68, 6011; e) H. Siebeneicher, I. By-
tschkov, S. Doye, Angew. Chem. Int. Ed. 2003, 42, 3042;
f) S. Thielges, E. Meddah, P. Bisseret, J. Eustache, Tetra-
hedron Lett. 2004, 45, 907; g) L. Ackermann, Org. Lett.
2005, 7, 439; h) Y.-Q. Fang, M. Lautens, Org. Lett.
Scheme 3. Sequential amination reactions.
Experimental Section
ꢀ
Tandem Palladium Catalysed C N Bond Formation to
´
2005, 7, 3549; i) J. Barluenga, M. A. Fernandez, F. Aznar,
form the Indole Nucleus. General Procedure,
Exemplified by 1-Phenylindole from 1-Bromo-2-(2-
bromovinyl)benzene (Table 2, Entry 1)
´
C. Valdes, Chem. Eur. J. 2005, 11, 2276.
[5] Examples of Pd-catalyzed C-N formation using alkenyl
halides or triflates as substrates: a) Y. Kozawa, M.
Mori, Tetrahedron Lett. 2002, 43, 111; b) M. C. Willis,
G. N. Brace, Tetrahedron Lett. 2002, 43, 9085; c) J. Bar-
Sodium tert-butoxide (125 mg, 1.298Â10ꢀ3 mol) was added
to an oven-dried flask charged with Pd2(dba)3 (12 mg, 1.3Â
10ꢀ5 mol) and ligand 5 (16 mg, 3.968Â10ꢀ5 mol) under nitro-
gen. The flask was flushed with nitrogen and the reagents sus-
pended in anhydrous toluene (1.04 mL). To this, 1-bromo-2-(2-
bromovinyl)benzene (136 mg, 5.192Â10ꢀ4 mol) and aniline
(58 mg, 57 mL, 6.230Â10ꢀ4 mol) were added and the reaction
mixture was heated at 808C for 5 hours under nitrogen. After
cooling, the reaction mixture was diluted with diethyl ether
(5 mL) and filtered through a celite pad, washing with diethyl
ether (30 mL). The filtrate was reduced under vacuum. The
product was separated via flash chromatography (eluant:
0.5% diethyl ether-hexane) to afford the indole as a pale amber
oil; yield: 88 mg (85%): IR (liquid film): nmax ¼3055, 2924, 2854,
1597, 1515, 1497, 1456, 1331, 1233, 1213, 1135, 1014, 953, 774,
´
´
luenga, M. A. Fernandez, F. Aznar, C. Valdes, Chem.
Commun. 2002, 2362; d) D. J. Wallace, D. J. Klauber,
C.-y. Chen, R. P. Volante, Org. Lett. 2003, 5, 4759; e) J.
´
´
Barluenga, M. A. Fernandez, F. Aznar, C. Valdes,
Chem. Eur. J. 2004, 10, 494; f) J. Barluenga, M. A. Fer-
´
´
nandez, F. Aznar, C. Valdes, Chem. Commun. 2004,
1400 ; g) A. Klapers, K. R. Campos, C.-y. Chen, R. P. Vol-
ante, Org. Lett. 2005, 7, 1185; h) M. C. Willis, J. Chauhan,
W. G. Whittingham, Org. Biomol. Chem. 2005, 3094;
i) M. C. Willis, G. N. Brace, I. P. Holmes, Synthesis
2005, 3229.
[6] M. C. Willis, G. N. Brace, I. P. Holmes, Angew. Chem. Int.
Ed. 2005, 44, 403.
[7] See Supporting Information for details.
[8] ACDsearch November 2005.
1
740, 695 cmꢀ1; H NMR (CDCl3): d¼6.71 (1H, dd, J¼3.3
and 0.8 Hz, ArH), 7.16–7.28 (2H, m, ArH), 7.34–7.42 (1H,
[9] For a review on the use of aryl chlorides in palladium-
catalysed coupling reactions, see: A. F. Littke, G. C. Fu,
Angew. Chem. Int. Ed. 2002, 41, 4176.
m, ArH), 7.37 (1H, d, J¼3.3 Hz, ArH), 7.52–7.55 (4H, m,
ArH), 7.57–7.62 (1H, m, ArH), 7.67–7.74 (1H, m, ArH); 13
C
NMR (CDCl3): d¼103.5, 110.5, 120.3, 121.1, 122.3, 124.4,
Adv. Synth. Catal. 2006, 348, 851 – 856
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
855

Michael C. Willis et al.
COMMUNICATIONS
ꢀ
[10] Reviews on palladium-catalysed C N bond formation,
see: a) J. F. Hartwig, in: Handbook of Organopalladium
Chemistry for Organic Synthesis, (Ed.: E. Negishi), Wi-
ley-Interscience, New York, 2002, Vol. 1, p. 1051;
b) A. R. Muci, S. L. Buchwald, Top. Curr. Chem. 2002,
A. Klapars, S. L. Buchwald, J. Am. Chem. Soc. 2003,
125, 6653; d) S. D. Walker, T. E. Barder, J. R. Martinelli,
S. L. Buchwald, Angew. Chem. Int. Ed. 2004, 43, 1871;
e) E. R. Strieter, D. G. Blackmond, S. L. Buchwald, J.
Am. Chem. Soc. 2003, 125, 13978.
219, 131; c) D. Prim, J.-M. Campagne, D. Joseph, B. An- [12] See ref.[5d]
drioletti, Tetrahedron 2002, 58, 2041.
[13] a) L. Shi, M. Wang, C. A. Fan, F. M. Zhang, Y. Q. Tu,
[11] a) D. W. Old, J. P. Wolfe, S. L. Buchwald, J. Am. Chem.
Soc. 1998, 120, 9722; b) J. P. Wolfe, H. Tomori, J. P. Sa-
dighi, J. Yin, S. L. Buchwald, J. Org. Chem. 2000, 65,
1158; c) X. Huang, K. W. Anderson, D. Zim, L. Jiang,
Org. Lett. 2003, 5, 3515; b) F. Rataboul, A. Zapf, R. Jack-
stell, S. Harkal, T. Riermeier, A. Monsees, U. Dingerdis-
sen, M. Beller, Chem. Eur. J. 2004, 10, 2983.
856
asc.wiley-vch.de
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 851 – 856