Construction of indole skeletons by sequential actions of sunlight and rhodium on α-aminoacetophenones

Article abstract of DOI:10.1246/cl.130463

Indole skeletons were constructed from 2-(N-aryl-N-methylamino) acetophenones by the sequential actions of sunlight and a rhodium catalyst. This method presents an example of the direct use of sunlight in organic synthesis.

Full text of DOI:10.1246/cl.130463

doi:10.1246/cl.130463
1076
Published on the web June 11, 2013
Construction of Indole Skeletons by Sequential Actions
of Sunlight and Rhodium on ¡-Aminoacetophenones
³
Naoki Ishida, David Nečas, Yasuhiro Shimamoto, and Masahiro Murakami*
Department of Synthetic Chemistry and Biological Chemistry, Kyoto University, Katsura, Kyoto 615-8510
(Received May 17, 2013; CL-130463; E-mail: murakami@sbchem.kyoto-u.ac.jp)
Indole skeletons were constructed from 2-(N-aryl-N-methyl-
amino)acetophenones by the sequential actions of sunlight and a
rhodium catalyst. This method presents an example of the direct
use of sunlight in organic synthesis.
An indole is a privileged structural motif found in a wide
variety of natural products and bioactive molecules. Conse-
quently, the construction of the indole skeleton has been a long-
standing subject of extensive research in organic synthesis.1
On the other hand, sunlight is an ideal energy source. The
development of organic transformation driven by sunlight would
contribute to the promotion of environmentally friendly organic
synthesis.2 We recently reported the solar-driven incorporation
of CO2 into ¡-amino ketones, taking photosynthesis as the role
model.3 N-Arenesulfonylamino ketones are initially irradiated
with sunlight to afford azetidinols, harvesting the solar energy
and storing it in the form of structural strain.4 Subsequent
incorporation of CO2 takes place with relief of the structural
strain as the driving force. The solar energy is stored as
structural strain and ultimately used to chemically transform
CO2. Herein, we report another example of a solar-driven
transformation, which constructs indole skeletons from 2-(N-
aryl-N-methylamino)acetophenones (Scheme 1).5-7 The sequen-
tial actions of sunlight, a rhodium catalyst, and an acid prompt
cyclization, which mechanistically involves cleavage of C-H
and C-C bonds. Neither halogen nor metallic atoms, but only
water molecules, evolves from the starting substance in addition
to the desired product.8 An application to enantioselective
synthesis of acid-labile indolinols is also demonstrated.
Scheme 1. Formally depicted construction of indole skeleton.
Figure 1. A photograph of a photoreaction with a sunlight
collector. A Pyrexμ flask is put in the center of a semicircular
column covered with an aluminum foil. See Supporting
Information for details.
Amino ketone 1a was easily prepared by reaction of
phenacyl bromide with N-methylaniline.9 A solution of 1a in
THF was put in a Pyrexμ flask equipped with a sunlight
collector, shown in Figure 1, which was exposed to sunlight at
ambient temperature for 8 h on a sunny day (Scheme 2).10
Subsequently, [Rh(OH)(cod)]2 (5 mol %) and rac-BINAP
(12 mol %) were added to the flask. The resulting mixture was
stirred at room temperature for 20 h. The mixture was then
treated with aqueous HCl (1 M) for 10 min, and subjected to
extractive work-up followed by column chromatography on
silica gel. The indole 2a was isolated in 60% yield based on
1a.11,12 Formally, the indole skeleton was constructed from the
¡-amino ketone along with elimination of water as the only
waste.
A mechanistic scenario for the production of 2a from 1a is
shown in Scheme 3. It consists of three separate steps. The
initial step is a photocyclization reaction of ketone 1a (first step),
which was originally reported by Yang,13a and later applied to
N-arylamino ketones.13c The carbonyl group is excited and the
carbonyl oxygen abstracts the £-hydrogen in a homolytic
manner.14 The resulting 1,4-biradical intermediate I intramolec-
Scheme 2. Construction of indole skeleton.
ularly couples to construct a four-membered ring skeleton II.
Although an electric mercury lamp was used in the original
reports,13 our study proved that the photocyclization reaction
proceeded at a reasonable rate on exposure to sunlight.15 It is
noteworthy that the reaction is energetically uphill; the thermally
stable C(sp3)-H bond located £ to the carbonyl group is cleaved
and adds across the carbonyl group. A high-energy four-
membered ring arises, thereby harvesting solar energy in the
form of structural strain.4 The second step is a rhodium-
catalyzed skeletal rearrangement reaction, whose mechanism is
similar to that of cyclobutanols (second step);16 the hydroxy
group of II is deprotonated by rhodium hydroxide to generate
the rhodium azetidinolate III. Then, the four-membered ring is
opened through ¢-carbon elimination to afford the amino-
methylrhodium intermediate IV.17 A 1,4-rhodium shift follows,
furnishing the arylrhodium species V,18 which intramolecularly
adds to the carbonyl group.19 The resulting tertiary alkoxide VI
deprotonates another molecule of the azetidinol II to yield the
Chem. Lett. 2013, 42, 1076-1078
© 2013 The Chemical Society of Japan
www.csj.jp/journals/chem-lett/

1077
Table 1. Synthesis of indoles 2 from aminoacetophenones 1a
Yield
/%b
Entry
1
R
Ar
1
2
3
4
5
6
7
8
9
1b
1c
1d
1e
1f
1g
1h
1i
H
H
H
H
Me
CF3
Cl
4-CH3C6H4
4-MeOC6H4
4-ClC6H4
2-Naphthyl
Ph
Ph
Ph
Ph
Ph
66
70
66
69
64
60
65
65
66
F
1j
CO2Me
aReaction conditions: ¡-amino ketone 1 (0.10 mmol), THF
(2.0 mL), sunlight, 8 h; then [Rh(OH)(cod)]2 (5 mol %), rac-
b
BINAP (12 mol %), 50 °C, 20 h. Isolated yield.
Scheme 3. Plausible mechanism.
Scheme 4. Regioselective construction of indole skeletons.
hydroxyindoline VII, together with the rhodium azetidinolate
III, which re-enters the next catalytic cycle. This energetically
downhill reaction is followed by the third step, which is an acid-
mediated dehydration reaction of the tertiary alcohol VII.20
We further examined the present one-pot sequential process
from the synthetic viewpoint. Table 1 demonstrates its func-
tional group tolerance. Various ¡-aminoacetophenones 1 were
successfully transformed into indoles 2 in yields ranging from
60 to 70%. It is noteworthy that various electron-withdrawing
substituents on the aniline moiety (Entries 6-9), which would
have detrimental effects on acid-mediated cyclizations,6 are
allowed.
The unsymmetrically substituted ¡-amino ketones 1k and 1l
were subjected to the standard reaction conditions (Scheme 4).
The ¡-amino ketone 1k selectively produced 6-acetylindole 2k
in 65% isolated yield. The amount of regioisomer (4-acetyl-
indole) detected in the crude reaction mixture was less than 5%.
This product selectivity is attributed to the 1,4-rhodium shift
occurring with intermediate IV (Scheme 3, second step). The
rhodium atom migrates to the sterically less hindered 6-
position.21 A similar site selectivity was also observed in the
reaction of ester-substituted ¡-amino ketone 1l. Indole 2l, which
is a key intermediate in the synthesis of a monoacylgycerol
lipase inhibitor,22 was obtained in 62% isolated yield. The site
selectivity of the present sequential process stands in marked
contrast to that of the previously reported iridium-catalyzed
cyclization,7 which selectively affords 4-acetylindoles via
acetyl-group-directed C-H activation.
It was possible to produce chiral indolinols 3, which were
labile to acid-promoted dehydration, when the rhodium-cata-
lyzed reaction was carried out in the presence of a base (e.g.,
Cs2CO3). Even enantioselective synthesis was achieved using a
chiral ligand for rhodium (Scheme 5). Thus, after completion of
the sunlight-promoted photocyclization of 1a, [Rh(OH)(cod)]2,
(R)-DIFLUORPHOS
(5,5¤-bis(diphenylphosphino)-2,2,2¤,2¤-
tetrafluoro-4,4¤-bi-1,3-benzodioxole),23 and Cs2CO3 were added
to the reaction vessel, which was then heated at 40 °C for 16 h.
Indolinol 3a was obtained in 62% NMR yield,24 with an
enantiomeric ratio of 97:3. Mechanistically, the arylrhodium
intermediate V adds to the carbonyl group in an enantioselective
manner (Scheme 3, second step).25 Chiral indolinol 3j was also
produced in 72% yield with an enantiomeric ratio of 96:4.
Chem. Lett. 2013, 42, 1076-1078
© 2013 The Chemical Society of Japan
www.csj.jp/journals/chem-lett/

1078
Dtsch. Chem. Ges. 1892, 25, 2860.
6
For acid-mediated cyclization of ¡-aminoacetophenones affording
3-arylindoles, see: a) A. F. Crowther, F. G. Mann, D. Purdie,
J. Chem. Soc. 1943, 58. b) K. E. Bashford, A. L. Cooper, P. D.
Kane, C. J. Moody, S. Muthusamy, E. Swann, J. Chem. Soc., Perkin
Trans. 1 2002, 1672.
7
8
For an iridium-catalyzed reaction of ¡-aminoacetophenones having
a 3-acetylphenyl group on nitrogen forming 4-acetylindoles, see: K.
Tsuchikama, Y.-k. Hashimoto, K. Endo, T. Shibata, Adv. Synth.
Catal. 2009, 351, 2850.
For cyclization of N-(o-iodophenyl)amino ketones, see: a) M.
Kihara, Y. Iwai, Y. Nagao, Heterocycles 1995, 41, 2279. b) D. Solé,
L. Vallverdú, X. Solans, M. Font-Bardía, J. Bonjoch, J. Am. Chem.
Soc. 2003, 125, 1587. c) Y. Kobayashi, J. Igarashi, C. Feng, T. Tojo,
Tetrahedron Lett. 2012, 53, 3742.
9
Supporting Information is available electronically on the CSJ-
Journal Web site, http://www.csj.jp/journals/chem-lett/index.html.
10 We defined it as “sunny” when the hourly solar radiation over
0.4 kW m¹2 was observed. The reaction proceeded even on a cloudy
day, although the rate decreased.
Scheme 5. Enantioselective synthesis of indolinols 3.
11 No reaction took place when 1a was directly treated with a rhodium
catalyst without irradiation with light.
12 The photoreaction was significantly retarded in the presence of a
rhodium catalyst probably because the rhodium complex quenched
the photoexcited ketone.
13 a) N. C. Yang, D.-D. H. Yang, J. Am. Chem. Soc. 1958, 80, 2913. b)
E. H. Gold, J. Am. Chem. Soc. 1971, 93, 2793. c) K. L. Allworth,
A. A. El-Hamamy, M. M. Hesabi, J. Hill, J. Chem. Soc., Perkin
Trans. 1 1980, 1671.
In conclusion, we have developed a solar-driven one-pot
procedure for the construction of indole skeletons from acyclic
N-aryl-N-methylaminoacetophenones, which are readily avail-
able. The present process consists of three steps. The first step is
an energetically uphill photocyclization promoted by sunlight.
Solar energy is harvested and stored as strain energy in the
resulting azetidinol. If compared with photosynthesis occurring
in green plants, the first step plays the role of the “light
reaction.” The second step is downhill in energy, corresponding
to the “dark reaction” in photosynthesis. The rhodium-catalyzed
skeletal rearrangement requires no sunlight and instead uses
the strain energy, which originates from sunlight, in order to
transform thermally stable ·-bonds such as C-H and C-C
bonds. The final step is an acid-mediated dehydration step. The
sequential process presents an example of the direct use of solar
energy in organic synthesis.
14 P. J. Wagner, Acc. Chem. Res. 1971, 4, 168.
15 Exposure to sunlight for 8 h on a sunny day formed 2a as the major
product along with unidentified by-products. Chromatographic
isolation furnished 2a in 67% yield. We also carried out the
photoreaction using a mercury lamp (250 W) for comparison.
Although the reaction was somewhat faster, the chemical yield of
azetidinol II was comparable, ranging 65-70%.
16 a) T. Seiser, O. A. Roth, N. Cramer, Angew. Chem., Int. Ed. 2009,
48, 6320. b) M. Shigeno, T. Yamamoto, M. Murakami, Chem.®Eur.
J. 2009, 15, 12929.
17 a) T. Matsuda, M. Makino, M. Murakami, Org. Lett. 2004, 6, 1257.
b) T. Seiser, N. Cramer, J. Am. Chem. Soc. 2010, 132, 5340.
18 a) K. Oguma, M. Miura, T. Satoh, M. Nomura, J. Am. Chem. Soc.
2000, 122, 10464. b) T. Hayashi, K. Inoue, N. Taniguchi, M.
Ogasawara, J. Am. Chem. Soc. 2001, 123, 9918. c) K. Sasaki, T.
Nishimura, R. Shintani, E. A. B. Kantchev, T. Hayashi, Chem. Sci.
2012, 3, 1278.
This work was supported in part by a Grant-in-Aid for
Scientific Research from MEXT, the ACT-C program from JST,
and The Asahi Glass Foundation. D.N. and Y.S. acknowledge
JSPS Research Fellowship for Young Scientists.
19 a) M. Sakai, M. Ueda, N. Miyaura, Angew. Chem., Int. Ed. 1998, 37,
3279. b) A. Takezawa, K. Yamaguchi, T. Ohmura, Y. Yamamoto, N.
Miyaura, Synlett 2002, 1733.
20 When the isolated azetidinol II was subjected to the rhodium-
catalyzed skeletal rearrangement reaction followed by acid treat-
ment, the indole 2a was isolated in 93% yield.
This paper is dedicated to Professor Teruaki
Mukaiyama in celebration of the 40th anniversary of the
Mukaiyama aldol reaction.
References and Notes
³
Present address: Department of Organic Chemistry, Charles Uni-
versity in Prague, HLavova 8, 128 42 Praha 2, Czech
21 A sequential reaction of N-methyl-N-(3-tolyl)aminoacetophenone
afforded 6-methylindole selectively, suggesting the site-selectivity
responds to the steric hindrance of the ortho C-H bond.
22 H. Bian, K. M. Chevalier, P. J. Connolly, C. M. Flores, S.-C. Lin, L.
Liu, J. Mabus, M. J. Macielag, M. E. McDonnell, P. M. Pitis, Y.-M.
Zhang, S.-P. Zhang, J. Clemente, B. Zhu, PCT Int. Appl. WO 2010/
124114 A1.
23 S. Jeulin, S. D. de Paule, V. Ratovelomanana-Vidal, J.-P. Genêt, N.
Champion, P. Dellis, Angew. Chem., Int. Ed. 2004, 43, 320.
24 Partial dehydration of indolinol 3a occurred when the mixture was
subjected to purification by column chromatography on silica gel.
25 For catalytic reactions involving enantioselective addition of aryl-
and alkenylrhodium intermediates onto ketones, see: a) R. Shintani,
K. Okamoto, Y. Otomaru, K. Ueyama, T. Hayashi, J. Am. Chem.
Soc. 2005, 127, 54. b) T. Miura, M. Shimada, M. Murakami, Synlett
2005, 667. c) T. Korenaga, A. Ko, K. Uotani, Y. Tanaka, T. Sakai,
Angew. Chem., Int. Ed. 2011, 50, 10703. d) D. W. Low, G. Pattison,
M. D. Wieczysty, G. H. Churchill, H. W. Lam, Org. Lett. 2012, 14,
2548.
1
Reviews on indole synthesis: a) G. W. Gribble, Contemp. Org.
Synth. 1994, 1, 145. b) G. W. Gribble, J. Chem. Soc., Perkin Trans.
1 2000, 1045. c) G. R. Humphrey, J. T. Kuethe, Chem. Rev. 2006,
106, 2875. d) D. F. Taber, P. K. Tirunahari, Tetrahedron 2011, 67,
7195. e) S. Cacchi, G. Fabrizi, Chem. Rev. 2011, 111, PR215. f) M.
Inman, C. J. Moody, Chem. Sci. 2013, 4, 29.
Reviews on organic synthesis utilizing sunlight: a) P. Esser, B.
Pohlmann, H.-D. Scharf, Angew. Chem., Int. Ed. Engl. 1994, 33, 2009.
b) S. Protti, M. Fagnoni, Photochem. Photobiol. Sci. 2009, 8, 1499.
N. Ishida, Y. Shimamoto, M. Murakami, Angew. Chem., Int. Ed.
2012, 51, 11750.
For a review on solar energy storage by photoisomerizaiton: G.
Jones, II, S.-H. Chiang, P. T. Xuan, J. Photochem. 1979, 10, 1.
For an acid-mediated reaction of anilines with 2-bromoacetophe-
nones forming 2-arylindoles (Bischler-Möhlau reaction), see: a) R.
Möhlau, Ber. Dtsch. Chem. Ges. 1881, 14, 171. b) E. Fischer, T.
Schmitt, Ber. Dtsch. Chem. Ges. 1888, 21, 1071. c) A. Bischler, Ber.
2
3
4
5
Chem. Lett. 2013, 42, 1076-1078
© 2013 The Chemical Society of Japan
www.csj.jp/journals/chem-lett/