Dry reaction of indoles with carbonyl compounds on montmorillonite K10 clay: A mild, expedient synthesis of diindolylalkanes and vibrindole A

Article abstract of DOI:10.1016/S0040-4039(02)00682-2

Dry reaction of indoles 1a-c with aldehydes 2a-i, ketones 2j-o and an acetal 2p on M. K10 clay at room temperature furnished 3,3′-diindolylalkanes (DIAs: 3a-h, j-r). The novel formation of tris(3-indolyl)methane 4 from 2i through tandem reactions and of the DIAs 6 and 8 from 1d through Plancher rearrangements were also observed.

Full text of DOI:10.1016/S0040-4039(02)00682-2

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 4075–4078
Dry reaction of indoles with carbonyl compounds on
montmorillonite K10 clay: a mild, expedient synthesis of
diindolylalkanes and vibrindole A
Manas Chakrabarty,a,* Nandita Ghosh,a Ramkrishna Basaka and Yoshihiro Harigayab
aDepartment of Chemistry, Bose Institute, 93/1, APC Road, Kolkata 700009, India
bSchool of Pharmaceutical Sciences, Kitasato University, Minato-ku, Tokyo 108, Japan
Received 12 February 2002; revised 27 March 2002; accepted 5 April 2002
Abstract—Dry reaction of indoles 1a–c with aldehydes 2a–i, ketones 2j–o and an acetal 2p on M. K10 clay at room temperature
furnished 3,3%-diindolylalkanes (DIAs: 3a–h, j–r). The novel formation of tris(3-indolyl)methane 4 from 2i through tandem
reactions and of the DIAs 6 and 8 from 1d through Plancher rearrangements were also observed. © 2002 Elsevier Science Ltd.
All rights reserved.
The diindolylalkanes (DIAs) constitute an old1a,b but
important class of compounds, reviewed recently by
us.2 The syntheses of the DIAs reported up to the
sixties are well reviewed.3a,b The first isolation of a
hallucinogenic DIA from a fungus in 19774 and the
subsequent isolation of other DIAs, including some
bioactive members, from natural sources5a–d have made
this field a rapidly growing area. The principal synthetic
avenues to the DIAs comprise the protic or Lewis
acid-catalysed reaction of indoles or indolyl Grignard
reagents with aldehydes (or their acetals), ketones, a-
ketoacids, imines, iminium salts or nitrones.2 Most of
these methods suffer from various disadvantages such
as long reaction periods (e.g. 10 days6), use of expensive
Lewis acids (e.g. dysprosium triflate7) or preformed
In continuation of our ongoing interest10a,b in green
chemistry, we have developed a mild and expedient
synthesis of symmetrical DIAs using montmorillonite
K10 clay as the catalyst. The method is highly efficient
and free from the aforesaid drawbacks. A brief account
of our work and its main findings as well as the
advantages of this method over the extant synthetic
routes are discussed in this communication.
Montmorillonite K10 clay is known to behave as both
a protic and a Bronsted acid (Hammett acidity func-
tion, H0: −5.5 to 5.9) and has a large specific surface
area (500–760 m2/g), compared to other solid acidic
catalysts like silica gel (400–600 m2/g).11 We, therefore,
used this clay in a solvent-free reaction of indoles with
mainly aldehydes and ketones. Initially, indole 1a was
treated with 0.5 equiv. of acetaldehyde 2a, both
adsorbed on M. K10 clay (2 g/1 mM of 1a). A simple
work-up furnished vibrindole A 3a, a bacterial metabo-
lite,5c as the sole product in 10 min and in very good
yield (Scheme 1).
reagents
(e.g.
oxazolidines/tetrahydrooxazines,8a,b
nitrones9a,b), etc., yet furnishing the DIAs in extremely
low (e.g. 2%5a) or unspecified8b yields in some cases.
Moreover, none of these methods is environmentally
friendly, since always an excess of solvent and often
toxic or hazardous chemicals are employed.
Scheme 1.
Keywords: diindolylalkanes; vibrindole A; green synthesis; clay.
* Corresponding author. Tel.: 91 33 350 2402/2403/6619; fax: 91 33 350 6790; e-mail: manas@boseinst.ac.in
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)00682-2

4076
M. Chakrabarty et al. / Tetrahedron Letters 43 (2002) 4075–4078
The potential of this method became apparent when the
conditions and the yield of 3a were compared with
those of its five previous syntheses, all from indole
(Table 1).
the aldehydes 2b and 2c, respectively forming the
respective DIAs 3q and 3r in excellent yields (Scheme
2).
Our findings (Table 2) reflect the wide applicability and
usefulness of our method. For most aldehydes 2a–e, the
reactions were exceedingly fast (5–10 min), whereas for
salicylaldehyde 2f, the reaction required 2 h perhaps
because of its hydrogen bonding. The yields were also
excellent, viz. 92–97%, except for acetaldehyde 2a
where the somewhat lower yield (82%) can be
Prompted by this success, we extended this reaction of
indole with a range of other aldehydes 2a–i, ketones
2j–o and an acetal 2p under similar conditions, furnish-
ing (except for 2i) the respective symmetrical DIAs
3a–p (except 3i) in good to excellent yields. 5-Bromoin-
dole 1b and 2-methylindole 1c behaved similarly with
Table 1. Comparative evaluation of the syntheses of vibrindole A (3a) from indole (1a)
Sl. no.
Reagent
Temp.
Time
Yielda (%)
Lit. ref.
1
2
3
4
5
6
MeCHO, AcOH
Rt
Reflux
Rt
Rt
Rt
10 days
5 h
15 min
Not stated
17 h
58
45
25
86
83
82
6
12
8a,b
13
9a,b
(i) Propiolic acid, MeOH; (ii) -CO2
An oxazolidine,b MeCN, CF3CO2H
EtOH, DMCD,c aq. NaClO4, Pt-electrode
MeCHꢀN+(Bn)O−, Me3SiCl, CH2Cl2
Present method: MeCHO, M. K10 clay
Rt
10 min
a Refer to isolated yields.
b 2-Methyl-3-phenyloxazolidine.
c 2,6-Di-O-methyl-b-cyclodextrin.
Scheme 2.
Table 2. Dry reaction of indoles (1 mM) with aldehydes and ketonesa (0.5 mM) using M. K10 clay (2 g)
Sl. no.
1a+Aldehydes/ketones
Time
Prod.
Yieldb (%)
1
2
3
4
5
6
7
8
1a+MeCHO (2a)
1a+PhCHO (2b)
10 min
5 min
5 min
5 min
10 min
2 h
3a
82
97
96
95
92
95
80
81
88
83
69
71
76
77
73
97
3b
1a+o-O2NC6H4CHO (2c)
1a+m-O2NC6H4CHO (2d)
1a+p-MeOC6H4CHO (2e)
1a+o-OHC6H4CHO (2f)
1a+2-CHO-thiophene (2g)
1a+2-CHO-furan (2h)
1a+2-CHO-pyrrole (2i)
1a+MeCOMe (2j)
3cc
3d
3e
3f
10 min
1 h
3gc
3h
9
4 h
4 (not 3i)
3j
10
11
12
13
14
15
16
17
18
19
20
4.5d (8) h
4 h
1a+MeCOEt (2k)
1a+MeCOPh (2l)
3kc
3l
6d (16) h
6 h
1a+Cyclopentanone (2m)
1a+Cyclohexanone (2n)
1a+4-Me-cyclohexanone (2o)
1a+Phthalimido-CH2CH(OEt)2 (2p)
1b+2b
1c+2c
1d+2c
1d+2d
3m
3n
2 h
2 h
3oc
3pc
3qc
3rc
4d (8) h
5 min
5 min
2 h
96
93
63; 27
52; 37
5c; 6c
7c; 8c
2 h
a Benzophenone failed to react with 1a.
b Refers to isolated yields.
c New compounds.
d Using 6 g clay; time periods within parentheses were needed using 2 g clay.

M. Chakrabarty et al. / Tetrahedron Letters 43 (2002) 4075–4078
4077
and 8, respectively. The structures of 5–8 and the
distinction of 5/7 from 6/8 were proved by analysing
their spectra and by comparing the same with those of
3r. Thus, both 5 and 6 possessed the same molecular
accounted for by our observation that 2a decomposes
rapidly on clay. For ketones 2j–o, the reactions were
slower (2–6 h) and the yields were also lower (69–83%),
presumably because of steric crowding in the resulting
DIAs. That acetone and acetophenone required
increasingly more time, whereas benzophenone did not
react at all with 1a, lends support to our view. For 2p,
the reaction was also slow (4 h), but the yield of 3p was
excellent (97%).
1
formulae (HRMS) and similar structural units (IR, H,
13C). But, for 6, two different signals appeared for the
1
two NH’s (IR, H) as well as for the two methyls (1H,
13C), which disclosed its unsymmetrical nature. Further,
the 13C chemical shifts of the 2-methyl-3-substituted
indole nucleus of 3r (a 3,3%-DIA) and those of the
3-methyl-2-substituted indole ring of 5 (a 2,2%-DIA)
were more or less reproduced in the 13C NMR spec-
trum of 6, which demonstrated it to be a 2,3%-DIA. The
situation was analogous for 7 and 8.
Of the 2-formylheteroarenes 2g–i, the reaction was
inexplicably very fast (10 min) only with 2g. More
surprisingly, whereas 2g and 2h furnished the expected
DIAs 3g and 3h, respectively, 2i provided tris(3-
indolyl)methane 4,10b,14a,b instead of the expected DIA
3i (X=Y=H; R,R%=2-pyrrolyl). In light of our
recently reported clay-mediated, general synthesis of
triindolylmethanes,10b the formation of 4 may be con-
sidered to proceed through the successive formations of
the expected DIA 3i and the indoleninium species 3%i
(Scheme 3). Pyrrole was found to decompose rapidly on
M. K10 clay, which accounts for the fact that, contrary
to expectation, no pyrrole could be isolated from the
reaction mixture. However, it remains obscure as to
why 2i behaved differently from 2g and 2h.
The reactions no doubt proceeded through the two
successive intermediates A and B, followed by separate
Plancher rearrangements.17a–c One involved the usual
migration of an arylalkyl group and led to the 2,2%-
DIAs 5 and 7. The other involved the hitherto unre-
ported migration of a methyl group, in preference to an
arylalkyl group, and led to the 2,3%-DIAs 6 and 8
(Scheme 4). Pertinently, this is the first record of the
formation of 2,3%-DIAs from 3-alkylindoles, since only
2,2%-DIAs were reported to have been formed in all
previous cases.2
The time required to complete the reactions of 1a with
2i, 2l and 2p could be considerably reduced from 8, 16
and 8 h to 4.5, 6 and 4 h, respectively using thrice the
amount of clay (6 g/1 mM). These results suggested
that the greater the surface area of clay available, the
faster is the rate of the reactions.
The clay, recovered from the reaction of 1a with 2j was
washed and dried at 110–120°C for 12 h and used in a
second reaction between 1a and 2b. The product 3b was
obtained in similar yield (96%) and time (5 min) as in
the experiment using fresh clay. The recyclability of M.
K10 clay in these reactions was thus demonstrated.
When skatole 1d was separately treated with 2c and 2d
in a similar manner, both the reactions were consider-
ably slower and furnished, besides the expected respec-
tive 2,2%-DIAs 515 and 7, the unexpected 2,3%-DIAs 616
In conclusion, we have developed an expedient, clay-
mediated green synthesis of symmetrical DIAs.18 It is
superior to the extant synthetic routes for a number of
Scheme 3.
Scheme 4.

4078
M. Chakrabarty et al. / Tetrahedron Letters 43 (2002) 4075–4078
reasons, viz. (i) it is fast and efficient, employs a cheap,
non-toxic and reusable clay as the catalyst and involves
a simple work-up; (ii) it furnishes DIAs 3c–d from
nitrobenzaldehydes 2c–d exceedingly fast (5 min) and in
excellent yields (]95%) in contrast to a recent slow (10
h) synthesis (using lithium perchlorate as catalyst) with
moderate yield (67%);19 (iii) on hydrazinolysis, the
phthalimido-DIA 3p produced the corresponding diin-
dolylethylamine (3: X=Y=H; R,R%=H, CH2NH2)
much faster (6.5 h) and in a better yield (92%) than its
earlier three-step synthesis from indole and glyoxylic
acid (>80 h; 32%).20 Our method has enormous poten-
tial in the synthesis of naturally occurring DIAs, upon
which we have recently embarked.
10. (a) Chakrabarty, M.; Basak, R.; Ghosh, N. Tetrahedron
Lett. 2001, 42, 3913; (b) Chakrabarty, M.; Sarkar, S.
Tetrahedron Lett. 2002, 43, 1351.
11. Laszlo, P. Science 1987, 235, 1473.
12. Zee, S. H.; Chen, C. S. J. Chin. Chem. Soc. 1974, 21, 229;
Chem. Abstr. 1975, 82, 125215y.
13. Suda, K.; Takanami, T. Chem. Lett. 1994, 1915.
14. (a) Balmer, C. T.; Kinder, H.; Gutman, L. J. Med. Chem.
1965, 8, 397; (b) Akgun, E.; Pindur, U.; Muller, J. J.
Heterocycl. Chem. 1983, 20, 1303.
15. 5: Orange prisms, mp 218–220°C (pet. ether–CH2Cl2);
yield: 124 mg (63%); IR: 3396, 1520, 1348, 744 cm−1; MS:
m/z 395 (M+), 265 (100%), 235, 219–217, 204, 131, 130;
1H NMR (CDCl3; 500 MHz): l 2.11 (6H, s, 2×CH3), 6.75
(1H, s, Ar3CH), 7.12 and 7.15 (2H, t each, J 7.5 Hz), 7.21
(2H, d, J 8 Hz), 7.23 (1H, d, J 8.5 Hz), 7.45 and 7.51 (1H,
t, each J 7.5 Hz), 7.54 (2H, d, J 7.5 Hz), 7.59 (2H, br s,
2×NH), 7.94 (1H, d, J 8 Hz); 13C NMR (CDCl3; 125
MHz): l 8.8 (2×CH3), 37.7 (Ar3CH), 111.3, 119.1, 120.1,
122.5, 125.8, 128.9, 131.1, 133.8 (all Ar-CH), 109.9, 129.9,
131.8, 135.6, 135.8, 149.4 (all Ar-C).
Acknowledgements
The authors express their sincere thanks to the Direc-
tor, Bose Institute for providing laboratory facilities,
Mr. B. Majumdar (NMR Laboratory, B.I.), Professor
D. E. Games and Mrs. B. K. Stein (University of
Swansea Wales, UK) for recording the Mass spectra
and the CSIR, Govt. of India for financial support (a
project to M.C., N.G. and an ad hoc fellowship to
R.B.).
16. 6: Yellow prisms, mp 222–224°C (pet. ether–CH2Cl2);
yield: 54 mg (27%); IR: 3449, 3396, 1520, 1354, 738 cm−1
;
MS: m/z 395 (M+), 265 (100%), 235, 219–217, 204, 131,
1
130; H NMR (CDCl3; 500 MHz): l 2.18 and 2.26 (3H s
each, 2×CH3), 6.58 (1H, s, Ar3CH), 6.41–6.89 (1H, m),
7.09–7.22 (6H, m), 7.44–7.50 (2H, m), 7.51 and 7.82 (1H,
br s each, 2×NH), 7.56 (1H, d, J 8 Hz), 7.58 (1H, d, J 8.5
Hz), 8.01–8.06 (1H, m); 13C NMR (CDCl3; 125 MHz): l
8.8 and 10.1 (2×CH3), 53.1 (Ar3CH), 110.1, 111.4, 119.6,
119.7, 120.1, 120.2, 122.7, 123.2, 124.4, 126.1, 129.3,
129.7 (all Ar-CH), 112.0, 112.2, 129.46, 129.49, 129.8,
134.3, 134.9, 135.9, 136.7, 148.5 (all Ar-C).
References
1. (a) Fischer, E. Liebigs Ann. Chem. 1887, 242, 372; (b)
Noland, W. E.; Robinson, D. N. Tetrahedron 1958, 14,
68.
17. (a) Ciamician, G.; Plancher, G. Chem. Ber. 1896, 29,
2475; (b) Jackson, A. H.; Smith, P. Tetrahedron 1968, 24,
2227; (c) Jackson, A. H.; Naidoo, B.; Smith, P. Tetra-
hedron 1968, 24, 6119.
2. Chakrabarty, M.; Basak, R.; Harigaya, Y. Heterocycles
2001, 55, 2431.
18. In a typical experiment, a solution of indole (1a; 1 mM)
and thiophene-2-carboxaldehyde (2g; 0.5 mM) in CH2Cl2
(2 ml) was adsorbed on Montmorillonite K10 clay (2 g),
and the solvent was allowed to evaporate at rt. After
completion of the reaction (TLC), the organic matter was
leached with CH2Cl2 (3×15 ml), filtered, the solvent
removed from the filtrate and the residue crystallised
from pet. ether–CH2Cl2 to furnish 3g as red prisms, mp
188–190°C; 132 mg (80%); IR: 3403, 745 cm−1; MS: m/z
328 (M+; 100%), 295, 243, 210, 117; 1H NMR (CDCl3;
500 MHz): l 6.15 (1H, s, Ar3CH), 6.81 (2H, s), 6.88–6.93
(2H, m), 7.02 (2H, t, J 7.5 Hz), 7.14 (1H, dd, J 5, 1 Hz),
7.16 (2H, t, J 7.5 Hz), 7.33 (2H, t, J 8 Hz), 7.45 (2H, d,
J 8 Hz), 7.88 (2H, s, 2×NH); 13C NMR (CDCl3; 125
MHz): l 35.7 (Ar3CH), 111.5, 119.7, 120.18, 122.4, 123.5,
124.0, 125.5, 126.8 (all Ar-CH), 120.10, 127.1, 136.9,
149.0 (all Ar-C). All products were identified by IR
3. (a) Sundberg, R. J. The Chemistry of Indoles; Academic
Press: New York, 1970; p. 39; (b) Remers, W. A. In The
Chemistry of Heterocyclic Compounds; Houlihan, W. J.,
Ed.; Wiley-Interscience: New York, 1972; p. 1.
4. Porter, J. K.; Bacon, C. W.; Robbins, J. D.; Himmels-
bach, D. S.; Higman, H. C. J. Agric. Food Chem. 1977,
25, 88.
5. (a) Osawa, T.; Namiki, M. Tetrahedron Lett. 1983, 24,
4719; (b) Fahy, E.; Potts, B. C. M.; Faulkner, D. J.;
Smith, K. J. Nat. Prod. 1991, 54, 564; (c) Bell, R.;
Carmeli, S.; Sar, N. J. Nat. Prod. 1994, 57, 1587; (d)
Bifulco, G.; Bruno, I.; Riccio, R.; Lavayre, J.; Bourdy, G.
J. Nat. Prod. 2000, 63, 596.
6. Kamal, A.; Qureshi, A. A. Tetrahedron 1963, 19, 512.
7. Chen, D.; Yu, L.; Wang, P. G. Tetrahedron Lett. 1996,
37, 4467.
8. (a) Singh, H.; Sarin, R.; Singh, K. Heterocycles 1986, 24,
3039; (b) Singh, H.; Singh, K. Tetrahedron 1988, 44,
5897.
1
(nujol), MS (EI/CI), H (CDCl3; 500 MHz) and 13C (125
MHz) NMR, DEPT 135 spectra and, for new com-
pounds, additionally by elemental analysis and/or HR
EI/CI-MS.
9. (a) Denis, J.-N.; Mauger, H.; Vallee, Y. Tetrahedron Lett.
1997, 38, 8515; (b) Chalaye-Mauger, H.; Denis, J.-N.;
Averbuch-Pouchot, M.-T.; Vallee, Y. Tetrahedron 2000,
56, 791.
19. Yadav, J. S.; Reddy, B. V. S.; Murthy, Ch. V. S. R.;
Kumar, G. M.; Madan, Ch. Synthesis 2001, 787.
20. Julia, M.; Tilly, G. Bull. Soc. Chim. Fr. 1965, 2175.