Synthesis of indoles by intermolecular cyclization of unfunctionalized nitroarenes and alkynes: one-step synthesis of the skeleton of fluvastatin

Article abstract of DOI:10.1002/ejoc.200801274

The addition of Ru₃(CO)₁₂, dimethyl carbonate, or both to the reaction mixture improves the selectivity of the palladium/phenanthroline-catalyzed reaction of nitroarenes, arylalkynes, and CO to give 3-arylindoles. When 4-fluorophenyl

Full text of DOI:10.1002/ejoc.200801274

FULL PAPER
DOI: 10.1002/ejoc.200801274
Synthesis of Indoles by Intermolecular Cyclization of Unfunctionalized
Nitroarenes and Alkynes: One-Step Synthesis of the Skeleton of Fluvastatin
Fabio Ragaini,*[a] Flavia Ventriglia,[a] Mohamed Hagar,[a] Simone Fantauzzi,[a] and
Sergio Cenini[a]
Keywords: Alkynes / Arenes / Nitrogen heterocycles / Palladium / Carbonylation
The addition of Ru3(CO)12, dimethyl carbonate, or both to the
reaction mixture improves the selectivity of the palladium/
phenanthroline-catalyzed reaction of nitroarenes, aryl-
alkynes, and CO to give 3-arylindoles. When 4-fluorophen-
ylacetylene and nitrobenzene are used as substrates, the in-
dole skeleton of Fluvastatin and other pharmaceutically
active compounds is obtained in one step.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2009)
nitrosoarenes are used as substrates, the problem was ap-
parently solved by working in the presence of dimethyl sul-
fate, which methylates N-hydroxyindoles to afford more
stable N-methoxyindoles.[6] However, nitrosoarenes are
much less friendly starting materials than nitroarenes and
very few of them are commercially available. In this paper
we report our approaches to the improvement of the cata-
lytic reaction by employing nitroarenes as substrates. The
application of the optimized reaction conditions to the syn-
thesis of the indole skeleton of important pharmaceutically
active compounds is also discussed.
Introduction
The indole skeleton is central to many pharmaceutically
active compounds, and its synthesis continues to attract the
attention of many researchers.[1] We have recently reported
that palladium/phenanthroline complexes catalyze the reac-
tion of nitroarenes (1) with arylalkynes (2) under CO pres-
sure to give 3-arylindoles (3) (Scheme 1).[2,3]
Results and Discussion
Scheme 1.
When the reaction between 4-nitrotoluene (1b, R1 = Me,
R2 = H, X = CH; Scheme 1) and phenylacetylene (2a, R3
= Ph, R4 = H) was performed under the previously[2] opti-
The reaction is completely regioselective; no 2-arylindole
was detected, and the reaction does not require any pre-
functionalization of the ortho position of the nitroarene.
The palladium catalyst showed a much higher activity (ca.
500-fold) with respect to the ruthenium–cyclopentadienyl
catalyst originally reported by Nicholas for the same reac-
tion,[4] but the selectivity remained moderate. The reaction
likely proceeds by the intermediate formation of N-hy-
droxyindoles,[2,5] and this kind of product can be isolated if
the reaction is performed employing nitrosoarenes as sub-
strates, in the absence of CO and any catalyst.[5] More re-
cently, Nicholas and coworkers proposed that the accumu-
lation of hydroxyindoles is responsible for the moderate
yields because of their easy oxidative dimerization.[6] When
1
mized conditions (Table 1, Run 7), the H NMR spectrum
of the solid obtained after evaporation of the solvent (Sup-
porting Information) showed the presence of just one out-
standing signal in the aliphatic region due to 5-methyl-3-
phenylindole (3b). However its integration against an in-
ternal standard (2,4-dinitrotoluene) showed it to corre-
spond to only 44% of the starting nitrotoluene and at least
eight more small signals of comparable intensity were also
present, whose total integration precisely matched the 56%
missing mass balance. Only four of these signals could be
assigned to known compounds. Because the final indole
was shown to be quite stable under the reaction conditions,
it is clear that the other byproducts are derived from a side
reaction of a reaction intermediate, and we agree with Nich-
olas that the most likely species responsible for alternative
reaction pathways are N-hydroxyindoles, although under
our conditions the byproduct identified in his work can at
best be one of several possibilities. To speed up the conver-
[a] Dipartimento di Chimica Inorganica, Metallorganica e Analit-
ica “Lamberto Malatesta”, University of Milano and ISTM-
CNR,
Via Venezian 21, 20133 Milano, Italy
Fax: +39-0250314405
E-mail: fabio.ragaini@unimi.it
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
Eur. J. Org. Chem. 2009, 2185–2189
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2185

F. Ragaini, F. Ventriglia, M. Hagar, S. Fantauzzi, S. Cenini
Table 1. Synthesis of 3-arylindoles from nitroarenes, phenylacetylene, and CO catalyzed by [Pd(Phen)2][BF4]2 and Ru3(CO)12.[a]
FULL PAPER
Run
ArNO2
t
[h]
[Pd(Phen)2][BF4]2/
Ru3(CO)12 [mol ratio]
Conversion of ArNO2
[%][b]
Selectivity of Indole
[%][c]
1
2
PhNO2
PhNO2
PhNO2
PhNO2
PhNO2
PhNO2
TolNO2
TolNO2
TolNO2
TolNO2
TolNO2
TolNO2
6
6
6
6
6
3
3
3
3
1:0
1:1
0:1
1:1
1:2
1:1
1:0
1:1
100
100
5.0
45
53
–
3
4[d]
5
100
100
100
100
100
100
100
88
43
46
54
44
55
45
56
57
–
6
7
8
9
10
11
12[d]
1:0.5
1:1.5
1:1
3
1.5
3
0:1
9.0
[a] Experimental conditions: [Pd(Phen)2][BF4]2 = 4.7ϫ10–3 mmol [except for Runs 3 and 12, where no palladium was present and Ru3-
(CO)12 (4.7ϫ10–3 mmol) was employed as the only catalyst]; mol ratio Pd/Phen/ArNO2/PhCϵCH, 1:20:300:1800; T = 170 °C; PCO
=
60 bar; 1,2-dimethoxyethane (10 mL). [b] Calculated with respect to the initial ArNO2. [c] Calculated with respect to the reacted ArNO2.
[d] Ph-BIAN [mol ratio Ph-BIAN/Ru3(CO)12 = 4.5] was also added.
sion of N-hydroxyindoles into other products, we investi- activating it towards both the reduction of nitroarenes to
gated two different strategies, which are described in the anilines[10] and the synthesis of allylic amines by reaction
following.
of nitroarenes with olefins.[11–13] However, the addition of
During the reaction, N-hydroxyindoles are reduced to Ru3(CO)12 to the palladium catalyst increases the indole
the corresponding indole by carbon monoxide. To the best selectivity by 8–11% (Table 1, compare Runs 1, 2 vs. 7, 10).
of our knowledge, very little precedent exists for this reac- The best Ru3(CO)12/[Pd(Phen)2][BF4]2 ratio is between 1
tion.[5,7] Several years ago, work in our group showed that and 1.5. Surprisingly, the addition of Ph-BIAN in the pres-
N-hydroxyindoles were formed during an intramolecular ence of both metals caused a decrease in the selectivity
cyclization reaction catalyzed by Pd(Phen)(TMB)2 (TMB = (Table 1, Runs 2 and 4). Apparently some interference oc-
2,4,6-trimethylbenzoate, Phen = 1,10-phenanthroline) and curred. The reaction time has a very minor influence
that prolonging the reaction time led to reduction of these (Table 1, Runs 2, 6 vs. 8, 11), indicating that the produced
compounds to the corresponding indoles.[7] It was noted in indole is stable under the reaction conditions. The fact that
the literature[5] that N-hydroxyindoles can be reduced by the selectivity is unchanged even when the nitroarene con-
CO with the use of [Cp*Ru(CO)2]2 as catalyst, but no ex- version is not complete (Table 1, Run 11) and the fact that
1
perimental results were reported and no comparison with no new peak was observed in the H NMR spectrum at
other catalysts was apparently attempted. Thus, no study the end of this reaction indicates that reduction of the N-
has been reported in the literature on the optimization of hydroxyindoles is relatively fast and no detectable accumu-
the reduction of hydroxyindoles by CO, but the related hy- lation of it was observed.
droxyamines are likely intermediates in the reduction of ni-
As a second strategy to improve selectivity, we employed
troarenes by CO/H2O, a reaction for which ruthenium car- the same approach reported by Nicholas. Since the begin-
bonyl complexes show a much higher activity than palla- ning it was obvious to us that the use of dimethyl sulfate
dium compounds.[8] Because the latter are more active for may not be compatible with our catalytic system, because
the activation of nitroarenes in most cases,[9] we attempted it could methylate phenanthroline. This was indeed found
to use [Pd(Phen)2][BF4]2/Ru3(CO)12 mixtures in different to be true and dimethyl sulfate almost completely inhibited
proportions as catalysts for the process in Scheme 1 with any catalytic activity (Table 2, Run 1). We thus resorted to
the idea that palladium may catalyze the formation of N- a milder methylating agent, dimethyl carbonate. The latter
hydroxyindoles and ruthenium their reduction. Results are is surely compatible with our catalytic system, because pal-
reported in Table 1. Both nitrobenzene (affording 3-phen- ladium–phenanthroline complexes or their bipyridyl ana-
ylindole, 3a) and 4-nitrotoluene [affording 3-(4-tolyl)indole, logues are among the catalysts that have been reported to
3b] were employed as substrates, with phenylacetylene as be active for the synthesis or organic carbonates by oxidat-
the alkyne, to better identify possible byproducts by gas ive carbonylation of alcohols.[14–17] With respect to dimethyl
chromatography (PhNO2) and 1H NMR spectroscopy sulfate, dimethyl carbonate has the additional advantage of
(TolNO2). The results were quite comparable with the two being much less toxic. Moreover, as the monomethyl ester
different substrates.
of carbonic acid is unstable and spontaneously decomposes
Data clearly show that Ru3(CO)12 alone does not cata- to CO2 and methanol, no base should be necessary to neu-
lyze indole formation, at least when added in such a low tralize the formed acid. Results are shown in Table 2.
(0.33 mol-%) amount (Table 1, Run 3), even in the presence
First of all it is clear that, as expected, dimethyl sulfate
of Ph-BIAN [Table 1, Run 12; Ph-BIAN = 1,2-bis(diphenyl- almost completely inhibits the reaction (Table 2, Run 1). In
imino)acenaphthene], the ligand which is more effective in contrast, the addition of dimethyl carbonate in the proper
2186
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 2185–2189

Synthesis of Indoles by Cyclization of Unfunctionalized Nitroarenes and Alkynes
Table 2. Synthesis of 3-arylindoles from nitroarenes, phenylacetyl- the reaction with 4-aminophenylacetylene, where the free
ene, and CO catalyzed by [Pd(Phen)2][BF4]2 in the presence of di-
amino group would probably be methylated or methoxycar-
bonylated by dimethyl carbonate under the reaction condi-
methyl carbonate.[a]
Run ArNO2
t
[h]
Me2CO3/Pd [mol ratio]
or Me2CO3 volume
Base
(base/Pd)
Selectivity of
Indole [%][b]
tions. We rather tested the efficiency of our reaction in the
reaction of nitrobenzene with 4-fluorophenylacetylene (2b)
to give 3-(4-fluorophenyl)indole (3c). This indole consti-
tutes the skeleton of many pharmaceutically active com-
pounds, among which Fluvastatin (Lescol®, a cholesterol-
lowering agent, A in Scheme 2) is that produced on the
largest scale, although other compounds of type B in
Scheme 2 (which are active 5-HT2 antagonists)[21] are also
important.
1[c]
PhNO2
6
1200 (Me2SO4)
K2CO3
(600)
–
–
–
–
–
traces
2[d]
3[d]
4[d]
5
6
7
PhNO2
PhNO2
PhNO2
PhNO2
PhNO2
PhNO2
PhNO2
6
6
6
6
6
3
6
300
1200
1800
48
54
41
42
14
47
44
10 mL (neat)
5 mL (+5 mL DMF)
1200
–
8
1200
K2CO3
(600)
–
NEt3
(100)
–
9
10
TolNO2
TolNO2
6
6
1200
1200
51
42
11[e] TolNO2
3
1200
47
[a] Experimental conditions: [Pd(Phen)2][BF4]2 = 4.7ϫ10–3 mmol;
mol ratio Pd/Phen/ArNO2/PhCϵCH, 1:20:300:1800; T = 170 °C;
PCO = 60 bar; 1,2-dimethoxyethane (10 mL). ArNO2 conversion
was always complete except for Runs 1 and 11. [b] Calculated with
respect to the reacted ArNO2. [c] PhNO2 conversion = 5.0%.
[d] Compare with Run 1 in Table 1, where no DMC is present.
[e] TolNO2 conversion = 95.0%. Compare with Run 7 in Table 1,
where no DMC is present.
Scheme 2. Some pharmaceutically relevant compounds based on
the 3-(4-fluorophenyl)indole (3c) skeleton.
amount increases the indole selectivity (compare Table 1,
Run 1 vs. Table 2, Runs 2–5; Table 1, Run 7 vs. Table 2,
The modified conditions described in this paper
Run 9), although the increase is limited (9% at best). The {[Pd(Phen)2](BF4)2 = 4.7ϫ10–3 mmol; mol ratios Pd/
ideal ratio of Me2CO3/Pd is 1200. The presence of dimethyl Ru3(CO)12/Phen/ArNO2/PhCϵCH/dimethyl
carbonate,
carbonate slows down the reaction somewhat. Indeed, while 1:1:20:300:1800:1200; T = 170 °C; PCO = 60 bar; 1,2-di-
in its absence the conversion was always complete even in methoxyethane (10 mL); 6 h} were found to be especially
3 h, in its presence the conversion of nitrotoluene was not effective for this reaction and 3c was obtained in a 71%
complete in this time (Table 2, Run 11) and the indole selec- spectroscopic yield (only 37% if ruthenium and dimethyl
tivity was lower in the case of nitrobenzene after 3 h then carbonate were not present). Although a compound com-
after 6 h (Table 2, Runs 7 and 3).
pletely free from any impurity could be obtained by column
Because dimethyl carbonate is less polar than dimethoxy- chromatography, we also identified a nonchromatographic
ethane used as solvent, we also tested a mixture of dimethyl separation technique suitable for larger-scale applications
carbonate and the more polar DMF, but the presence of (see Experimental Section).
the latter solvent gave much lower selectivity of indole and
Compared to its synthetic importance as an intermediate
in the synthesis of several important drugs, very little has
azo- and azoxybenzene predominated.[18,19]
The addition of either inorganic or organic bases is not been reported in the free or patent literature on the prepara-
necessary, and it even lowers the selectivity (Table 2, Runs 8 tion of 3c.[22] Only two papers (and apparently no patent)
and 10).
describe its preparation from compounds that are either
Most importantly, the effect of the two strategies de- commercially available or whose synthesis has been fully
scribed above can be summed up. When the reaction of ni- described.[23,24] Walkup and Linder reported that 3c can be
trotoluene with phenylacetylene was performed under the obtained in 44% yield starting from α-chloro-p-fluoroace-
conditions of Run 9 in Table 2, but with the addition of tophenone and aniline,[24] whereas Cacchi and coworkers
Ru3(CO)12 (4.7ϫ10–3 mmol), complete conversion was ob- reported a higher 71% yield for the same product by palla-
served, with a 60.0% selectivity in indole (3b). It is worth dium-catalyzed cyclization of o-ethynyltrifluoroacetanilide
noting that although the total selectivity increase is only in the presence of 4-fluoroiodobenzene,[23] but the starting
about 16% with respect to the maximum possible yield, it material needs three synthetic steps to be prepared.[25] An
corresponds to a more sensible 36% increase in the amount alternative approach was also mentioned, in which 3c is ob-
of obtained indole.
tained by copper-catalyzed decarboxylation of the corre-
At this stage, we did not test the effect of the modified sponding 2-indole-carboxylic acid.[21,26] Although a 77%
procedure on all the substrates reported in our previous pa- yield was reported for the decarboxylation step, several syn-
per. However, on the basis of the general reactivity of ni- thetic steps were needed to prepare the starting material
troarenes in reductive carbonylation reactions[8] and on that and details and yields for this substrate were not given. Our
of dimethyl carbonate,[20] no problem is expected except for procedure compares favorably to the alternative synthetic
Eur. J. Org. Chem. 2009, 2185–2189
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2187

F. Ragaini, F. Ventriglia, M. Hagar, S. Fantauzzi, S. Cenini
FULL PAPER
conditions are reported as captions to the tables. At the end of the
reaction, the autoclave was cooled with an ice bath and vented.
Reactions involving nitrobenzene were analyzed by gas chromatog-
raphy (naphthalene as an internal standard, Equity 5 column). Ani-
line, azobenzene, and azoxybenzene were always detected in small
amounts as byproducts. In the case of reactions involving 4-nitro-
toluene, 2,4-dinitrotoluene was added as an internal standard, and
the solution evaporated in vacuo. The residue was dissolved in
CDCl3 and analyzed by 1H NMR spectroscopy by using a delay
of 10 s. In the case of 3c, 19F NMR spectroscopy was used to mea-
strategies in terms of number of steps, global yield, and,
where relevant, amounts of precious metal required.[27–38]
The published characterization of 3c is also surprisingly
1
incomplete, as only its H NMR spectrum (in [D6]DMSO)
has been reported in the literature. We have run several
NMR spectra of 3c in CDCl3 and found that the 1H NMR
spectrum is strongly dependent on indole concentration. In
particular, the signals due to the benzo-fused ring of the
indole skeleton are shifted to lower fields upon an increase
in concentration. The 13C NMR, 19F NMR and several 2D sure the reported spectroscopic yield. To this aim, CF3CH2OH was
employed as an internal standard (in this case the standard was
added after elimination of the reaction solvent).
spectra are reported in the Supporting Information. Al-
though it is outside the scope of this paper to investigate in
detail the origin of the shift, it is clear that some concentra-
tion-dependent interaction is occurring between different
indole molecules. It is noteworthy that we did not observed
such a phenomenon for any of the other 3-arylindoles we
previously prepared,[2] although we must acknowledge that
we did not intentionally perform a variable concentration
study for all of them. It is quite stimulating to note that
although the 3-arylindole skeleton is mentioned as a generic
scaffold in several patents concerning pharmaceutically
active compounds, when it comes to specific examples the
4-fluorophenyl group is invariably present (e.g., see ref.[26]).
That the most (or perhaps only) active compounds are
those containing the moiety for which an unusual intermo-
lecular interaction is observed is a coincidence that may be
worth deeper investigation, although it is obvious that the
aggregation in chloroform can be only a reference point for
that occurring under physiological conditions.
Nonchromatographic Purification of 3c: The solution after the end
of the reaction was evaporated in vacuo (the excess amount of the
alkyne was evaporated with the solvent, and this mixture could be
reused for further syntheses), the solid residue was then charged
onto a short pad of silica gel and washed with hexane. This re-
moved other nonpolar and nonvolatile byproducts, such as alkyne
dimers. Finally, the pad was washed with dichloromethane/hexane
(4:6) [polar byproducts such as bis(4-fluorophenyl)urea, the cata-
lyst, and the excess amount of phenanthroline ligand were not
eluted under these conditions], and the filtrate was evaporated in
vacuo to give a product (62% yield), whose 1H NMR spectrum
(Supporting Information) is almost indistinguishable from that of
a sample purified by a lengthier and less-scalable chromatography.
Supporting Information (see footnote on the first page of this arti-
cle): NMR and mass spectroscopic data for 3-(4-fluorophenyl)-
indole (3c).
Acknowledgments
We thank Dr. A. Caselli for help in the interpretation of the NMR
spectra, the MiUR (PRIN2007HMTJWP_004) for financial sup-
port, and Metalor Technologies SA for a generous gift of ruthe-
nium trichloride.
Conclusions
Two independent approaches, the addition of a ruthe-
nium compound and that of a methylating agent, were
tested in the attempt to improve the selectivity of the palla-
dium/phenanthroline catalytic system for the synthesis of 3-
[1] G. R. Humphrey, J. T. Kuethe, Chem. Rev. 2006, 106, 2875–
2911.
arylindoles from nitroarenes, alkynes, and carbon mon- [2] F. Ragaini, A. Rapetti, E. Visentin, M. Monzani, A. Caselli, S.
Cenini, J. Org. Chem. 2006, 71, 3748–3753.
oxide. Both were successful, and importantly, they can be
used together showing additivity of the improvements. The
new approach is especially important in the synthesis of 3-
(4-fluorophenyl)indole, the scaffold of several pharmaceuti-
cally important drugs, for which it allowed an almost doub-
ling of the yield.
[3] For reviews on the applications of reduction reactions of ni-
troarenes by CO in fine chemistry, see: a) F. Ragaini, S. Cenini,
E. Gallo, A. Caselli, S. Fantauzzi, Curr. Org. Chem. 2006, 10,
1479–1510; b) B. C. G. Soderberg, Curr. Org. Chem. 2000, 4,
727–776; c) For the literature before 1996, see ch. 5 in ref.[8]
.
[4] A. Penoni, K. M. Nicholas, Chem. Commun. 2002, 484–485.
[5] A. Penoni, J. Volkmann, K. M. Nicholas, Org. Lett. 2002, 4,
699–701.
[6] A. Penoni, G. Palmisano, G. Broggini, A. Kadowaki, K. M.
Nicholas, J. Org. Chem. 2006, 71, 823–825.
[7] S. Tollari, A. Penoni, S. Cenini, J. Mol. Catal. A 2000, 152, 47–
54.
[8] S. Cenini, F. Ragaini, Catalytic Reductive Carbonylation of Or-
ganic Nitro Compounds, Kluwer Academic Publishers, Dord-
recht, The Netherlands, 1996.
[9] For a comparison in a related reaction, see: a) F. Ragaini, S.
Cenini, D. Brignoli, M. Gasperini, E. Gallo, J. Org. Chem.
2003, 68, 460–466; b) F. Ragaini, S. Cenini, E. Borsani, M.
Dompe, E. Gallo, M. Moret, Organometallics 2001, 20, 3390–
3398.
Experimental Section
Catalytic Reactions: In a typical reaction, the reagents (see Tables 1
and 2) were quickly weighed in a glass liner. The liner was placed
inside a Schlenk tube with a wide mouth under dinitrogen and
frozen at –78 °C with dry ice, evacuated and filled with dinitrogen,
after which the solvent was added. After the solvent was also fro-
zen, the liner was closed with a screw cap having a glass wool-filled
open mouth, which allowed exchange of gaseous reagents, and then
rapidly transferred to a 200-mL stainless steel autoclave with mag-
netic stirring. The autoclave was then evacuated and filled with
dinitrogen three times. CO was then charged at room temperature
at the required pressure, and the autoclave was immersed in an
oil bath preheated at the required temperature. Other experimental
[10] F. Ragaini, S. Cenini, M. Gasperini, J. Mol. Catal. A 2001,
174, 51–57.
[11] S. Cenini, F. Ragaini, S. Tollari, D. Paone, J. Am. Chem. Soc.
1996, 118, 11964–11965.
2188
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 2185–2189

Synthesis of Indoles by Cyclization of Unfunctionalized Nitroarenes and Alkynes
[12]
[13]
[27] The formation of 3c can be skipped in the preparation of Flu-
vastatin if the isopropyl group is introduced before the cycliza-
tion reaction described in ref.[24,28] However, this is not possible
in other cases. Starting from 3c, Fluvastatin can be obtained
by N-alkylation with an isopropyl group by well-established
procedures,[29–31] followed by introduction of the chiral substit-
uent as reported in the literature.[28,32]
F. Ragaini, S. Cenini, S. Tollari, G. Tummolillo, R. Beltrami,
Organometallics 1999, 18, 928–942.
F. Ragaini, S. Cenini, F. Turra, A. Caselli, Tetrahedron 2004,
60, 4989–4994.
A. Vavasori, L. Toniolo, J. Mol. Catal. A 2000, 151, 37–45.
H. Yasuda, K. Watarai, J. C. Choi, T. Sakakura, J. Mol. Catal.
A 2005, 236, 149–155.
H. Ishii, M. Goyal, M. Ueda, K. Takeuchi, M. Asai, Appl.
Catal. A 2000, 201, 101–105.
H. Ishii, M. Goyal, M. Ueda, K. Takeuchi, M. Asai, Catal.
Lett. 2000, 65, 57–60.
The use of DMF as solvent has been reported to be instrumen-
tal for allowing indole synthesis by cyclization of o-nitrosty-
renes by palladium–phenanthroline catalysts under mild condi-
tions (100 °C and ca. 2 bar CO pressure).[19] We tested these
mild conditions in the present system, but only a 6.3% conver-
sion was observed.
[14]
[15]
[28] O. Repic, K. Prasad, G. T. Lee, Org. Process Res. Dev. 2001, 5,
[16]
[17]
[18]
519–527.
[29] S. Singh, K. M. Nicholas, Synth. Commun. 2001, 31, 3087–
3097.
[30] G. Vavilina, A. Zicmanis, S. Drozdova, P. Mekss, M. Klavins,
Chem. Heterocycl. Compd. 2008, 44, 530–541.
[31] Y. R. Jorapur, J. M. Jeong, D. Y. Chi, Tetrahedron Lett. 2006,
47, 2435–2438.
[32] P. C. Fuenfschilling, P. Hoehn, J.-P. Mutz, Org. Process Res.
Dev. 2007, 11, 13–18.
[33] Alkynes are also central to some alternative synthetic strategies
for the synthesis of indoles. These have been very recently re-
viewed by Beller and coworkers.[34] Indoles bearing an aryl
group in the 3-position have been obtained in only a few cases
by these reactions and the 2-position was also substituted in
all cases.[35–38] So these approaches cannot yield 3c.
[34] K. Kruger, A. Tillack, M. Beller, Adv. Synth. Catal. 2008, 350,
2153–2167.
[19] I. W. Davies, J. H. Smitrovich, R. Sidler, C. Qu, V. Gresham,
C. Bazaral, Tetrahedron 2005, 61, 6425–6437.
[20] P. Tundo, L. Rossi, A. Loris, J. Org. Chem. 2005, 70, 2219–
2224.
[21] K. Andersen, J. Perregaard, J. Arn, J. B. Nielsen, M. Begtrup,
J. Med. Chem. 1992, 35, 4823–4831.
[22] The 13C NMR and 19F NMR spectra for 3c have never been
described in the literature and are here reported for the first
time.
[35] L. Ackermann, R. Sandmann, A. Villar, L. T. Kaspar, Tetrahe-
dron 2008, 64, 769–777.
[23] S. Cacchi, G. Fabrizi, F. Marinelli, L. Moro, P. Pace, Synlett
1997, 1363–1366.
[24] R. E. Walkup, J. Linder, Tetrahedron Lett. 1985, 26, 2155–2158.
[25] A. Arcadi, S. Cacchi, F. Marinelli, Tetrahedron Lett. 1992, 33,
3915–3918.
[36] L. Ackermann, A. Althammer, Synlett 2006, 3125–3129.
[37] B. Z. Lu, W. Zhao, H.-X. Wei, M. Dufour, V. Farina, C. H.
Senanayake, Org. Lett. 2006, 8, 3271–3274.
[38] S. Cacchi, G. Fabrizi, A. Goggiamani, Adv. Synth. Catal. 2006,
348, 1301–1305.
[26] J. K. Perregaard, K. Andersen, Eur. Pat. EP 470039 A2
19920205, 1992.
Received: December 22, 2008
Published Online: March 10, 2009
Eur. J. Org. Chem. 2009, 2185–2189
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2189