Can simple enones be useful partners for the catalytic stereoselective alkylation of indoles?

Article abstract of DOI:10.1021/jo0487202

A new catalytic system for the first example of enantioselective Friedel-Crafts-type (FC) addition of indoles to simple enones is described. The use of an equimolar amount of chiral [Al(salen)Cl] and 2,6-lutidine (10 mol%) was found to be effective in pro

Full text of DOI:10.1021/jo0487202

Ca n Sim p le En on es Be Usefu l P a r tn er s for th e Ca ta lytic
Ster eoselective Alk yla tion of In d oles?
Marco Bandini,* Matteo Fagioli, Marco Garavelli, Alfonso Melloni, Valerio Trigari, and
Achille Umani-Ronchi*
Dipartimento di Chimica “G. Ciamician”, Via Selmi 2, 40126 Bologna, Italy
marco.bandini@unibo.it; achille.umanironchi@unibo.it
Received J uly 26, 2004
A new catalytic system for the first example of enantioselective Friedel-Crafts-type (FC) addition
of indoles to simple enones is described. The use of an equimolar amount of chiral [Al(salen)Cl]
and 2,6-lutidine (10 mol %) was found to be effective in promoting the conjugate addition of indoles
to (E)-arylcrotyl ketones, furnishing the corresponding â-indolyl ketones in excellent yield and high
enantioselectivity (ee up to 89%). The role of the base was investigated through spectroscopic as
well as computational analyses, which suggested that in situ formation of a new chiral
(base‚[Al(salen)]) complex was operating under our reaction conditions. In particular, a stable
cationic [Al(salen)] hexacoordinate trans complex with the additive base and the enone is suggested
as being responsible for the stereocontrolled reaction. Finally, detailed monitoring of the reaction
course was carried out showing that a conventional FC pathway induced by [Al(salen)Cl] acting as
a Lewis acid is operating.
In tr od u ction
phosphonates, R,â-unsaturated heteroaromatic thioesters)
were required in order to achieve high levels of stereoin-
duction. On the other hand, the use of simple nonchelat-
The synthesis of indolyl-containing compounds1 bear-
ing a stereocenter adjacent to the heteoaromatic ring is
of high chemical and biochemical importance. In fact,
many pharmacologically and agrochemically active com-
pounds are characterized by the presence of this frame-
work in their molecular skeletons.2 In this context, after
the pioneering study of Kerr et al. on the use of Yb(OTf)3
for the catalytic addition of indoles to electron-deficient
carbon-carbon double bonds,3 an ever-increasing number
of mild, catalytic, and environmentally friendly FC alky-
lations of indoles via Michael-type addition have been
described.4 Homogeneous as well as heterogeneous cata-
lysts showed their effectiveness in this protocol, and in
some cases, both microwave irradiations and ultrasound
activations were found to be beneficial by improving the
chemical yields.5 More recently, as a consequence of the
astonishing growth of interest of both academic and
industrial communities, the development of catalytic
asymmetric FC strategies has been the subject of inten-
sive studies in order to access enantiomerically enriched
aromatic compounds bearing benzylic stereocenters.6
(4) (a) Taylor, S. K.; May, S. A.; Stansby, E. S. J . Org. Chem. 1996,
61, 2075. (b) Manabe, K.; Aoyama, N.; Kobayashi, S. Adv. Synth. Catal.
2001, 343, 174. (c) Yadav, J . S.; Abraham, S.; Reddy, B. V. S.; Sabitha,
G. Synthesis 2001, 2165. (d) Yadav, J . S.; Subba Reddy, B. V.; Murthy,
Ch. V. S. R.; Mahesh Kumar, G.; Madan, Ch. Synthesis 2001, 783. (e)
Reddy, R.; J aquith, J . B.; Neelagiri, V. R.; Saleh-Hanna, S.; Durst, T.
Org. Lett. 2002, 4, 695. (f) Nagumo, S.; Miyoshi, I.; Akita, H.;
Kawahara, N. Tetrahedron Lett. 2002, 43, 2223. (g) Bandini, M.; Cozzi,
P. G.; Giacomini, M.; Melchiorre, P.; Selva, S.; Umani-Ronchi, A. J .
Org. Chem. 2002, 67, 3700. (h) Bandini, M.; Melchiorre, P.; Melloni,
A.; Umani-Ronchi, A. Synthesis 2002, 1110. (i) Bandini, M.; Fagioli,
M.; Melloni, A.; Umani-Ronchi, A. Synthesis 2003, 397. (j) Srivastava,
N.; Banik, B. K. J . Org. Chem. 2003, 68, 2109. (k) Agnusdei, M.;
Bandini, M.; Melloni, A.; Umani-Ronchi, A. J . Org. Chem. 2003, 68,
7126. (l) J i, S.-J .; Wang, S.-Y. Synlett 2003, 2074. (m) J i, S.-H.; Zhou,
M.-F.; Gu, D.-G.; Wang, S.-Y.; Loh, T.-P. Synlett 2003, 2077. (n) Bartoli,
G.; Bartolacci, M.; Bosco, M.; Foglia, G.; Giuliani, A.; Marcantoni, E.;
Sambri, L.; Torregiani, E. J . Org. Chem. 2003, 68, 4594. (o) Ramesh,
C.; Banerjee, J .; Pal, R.; Das, B. Adv. Synth. Catal. 2003, 345, 557. (p)
Mujahid Alam, M.; Varala, R.; Adapa, S. R. Tetrahedron Lett. 2003,
44, 5115. (q) Vijender Reddy, A.; Ravinder, K.; Venkateshwar Goud,
T.; Krishnaiah, P.; Raju, T. V.; Venkateswarlu, Y. Tetrahedron Lett.
2003, 44, 6257. (r) Yadav, J . Reddy, B. V. S.; Swamy, T. Tetrahedron
Lett. 2003, 44, 9121. (s) Bandini, M.; Fagioli, M.; Umani-Ronchi, A.
Adv. Synth. Catal. 2004, 346, 545.
(5) (a) Chakrabarty, M.; Basak, R.; Ghosh, N. Tetrahedron Lett.
2001, 42, 3913. (b) Chakrabarty, M.; Basak, R.; Ghosh, N.; Harigaya,
Y. Tetrahedron 2004, 60, 1941.
(6) (a) Bolm, C.; Hildebrand, J . P.; Muniz, K.; Hermanns, N. Angew
Chem., Int. Ed. 2001, 40, 3285. (b) Wang, Y.; Ding, K.; Dai, L.
Chemtracts: Org. Chem. 2001, 14, 610. (c) J ørgensen, K. A. Synthesis
2003, 1117. (d) Bandini, M.; Melloni, A.; Umani-Ronchi, A. Angew.
Chem., Int. Ed. 2004, 43, 550.
In particular, chiral Cu(I/II)-,7 Sc(III)-,8 and Pd(II)-
based9 Lewis acids proved to be highly effective promoters
for stereoselective FC alkylations of indoles. In these
cases, bidentate chelating carbonyls (â,γ-unsaturated
R-ketoesters, arylidene malonates, R,â-unsaturated acyl
(7) (a) [Cu(II)/Box]: Zhuang, W.; Hansen, T.; J ørgensen, K. A. Chem.
Commun. 2001, 347. Zhuang, W.; Gathergood, N.; Hazell, R. G.;
J ørgensen, K. A. J . Org. Chem. 2001, 66, 1009. J ensen, K. B.;
Thorhauge, J .; Hazell, R. G.; J ørgensen, K. A. Angew. Chem., Int. Ed.
2001, 40, 160. Zhou, J .; Tang, Y. J . Am. Chem. Soc. 2002, 124, 9030.
Zhou, J .; Tang, Y. Chem. Commun. 2004, 432. (b) [Cu (I)/BINAP]:
J ohannsen, M. Chem. Commun. 1999, 2233. Saaby, S.; Fang, X.;
Gathergood, N.; J ørgensen, K. A. Angew. Chem., Int. Ed. 2000, 39,
4114.
* Corresponding authors. Tel: +39-051-2099509. Fax: +39-051-
2099456.
(1) (a) Sundberg, R. J . Indoles; Academic Press: San Diego, 1996.
(b) Nicolaou, K. C.; Snyder, S. A. Classic in Total Synthesis II; Wiley-
VCH: Weinheim, 2003.
(2) (a) Kozokowsi, A. P.; Greco, M. N. J . Am. Chem. Soc. 1980, 102,
1165. (b) Shinohara, T.; Suzuki, K. Tetrahedron Lett. 2002, 43, 6937.
(3) Harrington, P. E.; Kerr, M. A. Synlett 1996, 1047.
10.1021/jo0487202 CCC: $27.50 © 2004 American Chemical Society
Published on Web 10/08/2004
J . Org. Chem. 2004, 69, 7511-7518
7511

Bandini et al.
SCHEME 1. Ca libr a tion An a lysis for th e
F u n ction a l a n d Ba sis Set Used : Com p a r ison
betw een Sim u la ted a n d X-r a y-Resolved Str u ctu r es
for th e Ca tion ic Com p lex (AlP y4Cl2)+
F IGURE 1. (R,R)-[Al(salen)Cl].
ing R,â-unsaturated ketones as electrophiles still repre-
sents a considerable synthetic challenge.10 In fact, the
steric similarity of the two carbonyl substituents renders
the stereodifferentiation of the two enantiotopic faces of
the unsaturated ketone a difficult task.
We recently reported on the effective enantioselective
conjugate addition of indoles to (E)-arylcrotyl ketones in
the presence of a chiral aluminum complex.11 In particu-
lar, the combined use of [Al(salen)Cl] (1a , Figure 1)12 and
catalytic amounts of coordinating bases (10 mol %) were
found to be beneficial to the stereochemical outcome of
the reaction, leading the enantiomeric excess to be
increased up to 89% when 2,6-lutidine was employed.
In this paper we report a full account of our investiga-
tion focusing on the generality of the procedure. To shed
some light on the role of the additive on the stereoselec-
tion of the protocol, spectroscopic as well as theoretical
investigations were performed. Both novel penta- and
hexacoordinate complexes between [Al(salen)], the base,
and a model R,â-unsaturated carbonyl have been fully
optimized and characterized here, showing that an
interaction between the coordinating base and the alu-
minum complex is involved. In particular, a stable
cationic hexacoordinate chiral trans complex with the
electrophile and the base is suggested to be responsible
for the observed higher stereocontrol. Finally, both
kinetic evidence and NLE (nonlinear effect) investiga-
tions call for a first-order dependence of the reaction rate
on the [Al(salen)Cl] concentration, thereby ruling out a
hypothetical pathway involving the simultaneous double
interaction/activation of the Lewis acid with the electron-
rich aromatic nucleophile and the carbonyl substrate.13
processor system, using the DFT/B3LYP functional (i.e.,
Becke’s three-parameter hybrid functional with the Lee-
Yang-Parr correlation functional)15 in combination with a
mixed basis set: a split-valence type plus polarization for all
C and H atoms (i.e., 6-31G*) and a triple-ú valence type plus
polarization for Al(III) (i.e., 6-311G*), including diffuse func-
tions for N and O atoms (i.e., 6-311+G*). Indeed, since the N
and O lone pairs strongly interact with Al(III), the use of
diffuse functions is fundamental for a correct description of
these interactions.16 According to the literature, the functional
and basis set used have been shown to yield reasonable
energetic and structural results for metal-complexes of this
kind, either in a penta- or hexacoordination status, including
salen complexes with Al(III) and transition metals such as
Mn(III) and Cu(II).16-18 Anyway, to further validate the
computational level employed, we calibrated DFT results on
a known (i.e., X-ray resolved) structure: the cationic hexaco-
ordinate complex (AlPy4Cl2)+.19 Remarkably, computational
results agree very well with the experimental data (the error
bar in bond distances being lower than 0.05 Å, Scheme 1) and
Com p u ta tion a l Deta ils
(14) Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J . R.; Zakrzewski, V. G.; Montgomery, J . A.,
J r.; Stratmann, R. E.; Burant, J . C.; Dapprich, S.; Millam, J . M.;
Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J .;
Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo,
C.; Clifford, S.; Ochterski, J .; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J . B.; Cioslowski, J .; Ortiz, J . V.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.;
Fox, D. J .; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; J ohnson, B. G.; Chen,
W.; Wong, M. W.; Andres, J . L.; Head-Gordon, M.; Replogle, E. S.;
Pople, J . A. Gaussian 98, revision A.9; Gaussian, Inc.: Pittsburgh, PA,
1998.
All calculations were carried out using the tools available
in the Gaussian 9814 package on an SGI Origin 3800 multi-
(8) Evans, D. A.; Scheidt, K. A.; Fandrick, K. R.; Lam, H. W.; Wu,
J . J . Am. Chem. Soc. 2003, 125, 10780.
(9) Bandini, M.; Melloni, A.; Tommasi, A.; Umani-Ronchi, A. Helv.
Chim. Acta 2003, 86, 3753.
(10) Austin, J . F.; MacMillan, D. W. C. J . Am. Chem. Soc. 2002,
124, 1172.
(11) Bandini, M.; Fagioli, M.; Melchiorre, P.; Melloni, A.; Umani-
Ronchi, A. Tetrahedron Lett. 2003, 44, 5843.
(12) Selected examples of [Al(salen)Cl]-catalyzed asymmetric organic
transformations: (a) Myers, J . K.; J acobsen, E. N. J . Am. Chem. Soc.
1999, 121, 8959. (b) Evans, D. A.; J aney, J . M.; Magomedov, N.;
Tedrow, J . S. Angew. Chem., Int. Ed. 2001, 40, 1884. (c) J ha, S. C.;
J oshi, N. N. Tetrahedron: Asymmetry 2001, 12, 2463. (d) Sammins,
G. M.; J acobsen, E. N. J . Am. Chem. Soc. 2003, 125, 4442. (e) Taylor,
M. S.; J acobsen, E. N. J . Am. Chem. Soc. 2003, 125, 11204. (f) Sammis,
G. M.; Danjo, H.; J acobsen, E. N. J . Am. Chem. Soc. 2004, 126, 9928.
Chiral aluminum-Schiff base complexes have been fully characterized
spectroscopically as well as crystallograpically: Atwood, D. A.; Harvey,
M. J . Chem. Rev. 2001, 101, 37, and refs cited therein.
(13) Tarakeshwar, P.; Lee, J . Y.; Kim, K. S. J . Phys. Chem. A 1998,
102, 2253.
(15) Becke, D. J . Chem. Phys. A 1993, 98, 5648.
(16) (a) Tunega, D.; Haberhauer, G.; Gerzabek, M.; Lischka, H. J .
Phys. Chem. A 2000, 104, 6824. (b) Aquino, A. J . A.; Tunega, D.;
Haberhauer, G.; Gerzabek, M.; Lischka, H. Phys. Chem. Chem. Phys.
2000, 2, 2845. (c) Aquino, A. J . A.; Tunega, D.; Haberhauer, G.;
Gerzabek, M. H.; Lischka, H. Phys. Chem. Chem. Phys. 2001, 3, 1979.
(17) Cavallo, L.; J acobsen, H. Angew. Chem., Int. Ed. 2000, 39, 589.
(b) El-Bahraoui, J .; Wiest, O.; Feichtinger, D.; Plattner, D. A. Angew.
Chem., Int. Ed. 2001, 40, 2073. (c) J acobsen, H.; Cavallo, L. Chem.s
Eur. J . 2001, 7, 800.
(18) Thorhauge, J .; Roberson, M.; Hazell, R. G.; J ørgensen, K. A.
Chem.sEur. J . 2002, 8, 1888.
7512 J . Org. Chem., Vol. 69, No. 22, 2004

Catalytic Stereoselective Alkylation of Indoles
TABLE 1. Scr een in g of Solven ts for th e
En a n tioselective Ad d ition of 3a to 2c Ca ta lyzed by
[Al(Sa len )Cl]a
entrya
solvent
yield of 4ca (%)b ee of 4ca (%)c
1
2
3
4
5
6
7
toluene
Et2O
THF
CH3CN
n-Hex/toluene (85:15)
CH2Cl2
80
67
20
60
95
95
88
55
31
5
12
22
31
31d
toluene
F IGURE 2. Pictorial representation of the dynamic coordina-
tion of chiral Lewis acids with carbonyls.
a
All reactions were carried out at room temperature under a
nitrogen atmosphere, employing 10 mol % [Al(salen)Cl] for a
b
reaction time of 24 h. Chemical yields are given on the isolated
SCHEME 2. En a n tioselectivity a n d Str u ctu r e of
th e r,â-Un sa tu r a ted Keton e: A Na r r ow
Dep en d en ce
product after chromatographic purification. c Enantiomeric ex-
cesses were determined by HPLC analysis with chiral column
d
(Chiralcel OD). Reagent-grade toluene was utilized as the sol-
vent.
pose, enones 2a -c21 were reacted at room temperature
with 2-methylindole (3a ) and 2d 22 with indole 3c in the
presence of 10 mol % commercial [Al(salen)Cl] (1a ) as
the catalyst. The data collected in Scheme 2 clearly show
that the highest chemical as well as optical yields were
obtained with enone 2c in which a sterically demanding
aromatic group is bound to the carbonyl and a small
aliphatic group (Me) is linked to the C-C double bond.
In this case, the desired â-indolyl ketone 4ca was
isolated in 80% yield and 55% enantiomeric excess. On
the other hand, 4-phenyl-3-buten-2-one (2a ), chalcone
(2b), and 3-penten-2-one (2d ) reacted with 3a and 3c (in
the presence of [Al(salen)Cl], 10 mol %, toluene), giving
rise to indolyl ketones 4 in moderate yields (43-69%) but
with poor stereocontrol (up to 32%).
A subsequent screening of solvents led us to establish
toluene as the solvent of choice. As a matter of fact, when
more polar media, namely, Et2O, THF, and CH3CN, were
examined, 4ca was isolated in lower enantioselectivity
(Table 1, entries 2-4). Moreover, the decreased stereo-
induction recorded with reagent-grade toluene (ee, 31%,
entry 7) indicates that anhydrous conditions are neces-
sary in order to achieve high enantiocontrols.
Our efforts to optimize the reaction conditions were
also addressed to explore the effectiveness of other chiral
Al-Schiff base complexes 1b-e in promoting the addition
of 3a to 2c (Figure 3). However, the commercially
available [Al(salen)Cl] complex furnished the highest
enantioselectivity.
For instance, while the di-tert-butyl-salen aluminum
complex 1b provided 4ca in 90% and 39% ee, both Al-
Schiff base complexes characterized by different chiral
backbones (1c, 1e) and a chiral ligand bearing sterically
demanding adamantyl substituents (1d ) promoted the
reaction to some extent with no enantiocontrol.
with the X-ray crystallographic structures reported for similar
[Al(salen)] complexes,12b,20 thus proving the accuracy of these
computations.
Both the two external salen tBu-groups and the phenyl
substituent of the ketone have been removed in the model
system due to the computational cost. Finally, several con-
formers were optimized according to different orientations of
the coordinating ligands (bases, carbonyls), but only the lowest
energy minima have been shown here (a scan has been
performed to unveil the stable conformations of the attacking
base and enone).
Resu lts a n d Discu ssion
In the field of our current studies focused on the
development of new catalytic conjugate additions of
indoles to enones,4g-i,k,s we recently described the stereo-
selective first version of this transformation as a means
of [Al(salen)Cl] complex.11 Stereoselective Lewis acid-
catalyzed Michael additions to simple monodentate R,â-
unsaturated carbonyl compounds are widely recognized
to be challenging transformations, and a careful design
of the structural characteristics of the electrophile must
be done in order to guarantee high levels of stereoselec-
tion. Usually, suitable stereodiscrimination between the
two enantiotopic faces of the ketone is obtained by
adopting tailored structural motifs bearing markedly
different sterically demanding substituents (R, R′) at the
carbonyl moiety (Figure 2).
In keeping with our mechanistic hypothesis in which
the incoming indole reacts with the Lewis acid-activated/
coordinated enone to give the aluminum enolate 5 as an
intermediate,23 we considered plausible the presence in
Taking these guidelines into account, our initial at-
tempts toward the optimization of the reaction conditions
were addressed into the comprehension/optimization of
the above-mentioned sterical requirements. To this pur-
(21) Ketone 2c was synthesized following the known procedure:
Brown, C. D.; Chong, J . M.; Shen, L. Tetrahedron 1999, 55, 14233.
(22) Arisawa, M.; Torisawa, Y.; Kawahara, M.; Yamanaka, M.;
Nishida, A.; Nakagawa, M. J . Org. Chem. 1997, 62, 4327.
(19) Pullmann, P.; Hensen, K.; Bats, J . W. Z. Naturforsch. B 1982,
37, 1312.
(20) (a) Duxbury, J . P.; Warne, J . N. D.; Mushtaq, R.; Ward, C.;
Thornton-Pett, M.; J iang, M.; Greatrex, R.; Kee, T. P. Organometallics
2000, 19, 4445. (b) Atwood, D. A.; J egier, J . A.; Rutherford, D. Inorg.
Chem. 1996, 35, 63.
(23) For examples of enolates of Al-porphyrin and Al-Schiff base
complexes, see: (a) Kuroki, M.; Watanabe, T.; Aida, T.; Inoue, S. J .
Am. Chem. Soc. 1991, 113, 5903. (b) Cameron, P. A.; Gibson, V. C.;
Irvine, D. J . Angew. Chem., Int. Ed. 2000, 39, 2141.
J . Org. Chem, Vol. 69, No. 22, 2004 7513

Bandini et al.
TABLE 2. Scr een in g of Ba ses a s P ossible Ad d itives for
th e [Al(Sa len )Cl]-Ca ta lyzed Con ju ga te Ad d ition of In d ole
3a to 2c
entrya
base (%)
yield of 4ca (%)b ee of 4ca (%)c
1
2
3
4
5
6
7
8
9
none
80
70
54
65
65
87e
67f
59
87
18
25
55 (R)d
71 (R)
76 (R)
77 (R)
79 (R)
77 (R)
77 (R)
29 (R)
25 (R)
66 (R)
86 (R)
aniline (10)
pyridine (10)
Et3N (10)
2,6-lut (10)
2,6-lut (10)
2,6-lut (5)
2,6-ditBu-py (10)
2-CN pyridine (10)
Et3N (100)
10
11
2,6-lut (50)
a
All reactions were carried out in anhydrous toluene at room
temperature for a reaction time of 48 h unless otherwise specified.
b Isolated yield after flash chromatography. c Determined by chiral
d
HPLC analysis with Chiralcel OD column. Reaction time ) 24
h. e Reaction was carried out using an excess of 2c (1.5 equiv) with
respect to 3a (1 equiv). f Performed with 5 mol % [Al(salen)Cl] as
the catalyst. Reaction time ) 72 h.
F IGURE 3. Screening of Al-Schiff base complexes in the FC
alkylation of 3a .
the coordinating ability of the additive and the observed
stereoselectivity prompted us to examine a plausible
coordinating action of the base on the aluminum cata-
lyst.24 As a partial support of the latter hypothesis, we
observed that the chiral Lewis acid-Brønsted base ratio
significantly affects the outcome of the reaction. In fact,
using an excess of base with respect to the organometallic
catalyst (TEA 100%, lut 50%, entries 10-11) dramatically
affected the turnover of the catalytic cycle (coordination
poisoning of the catalyst), but high ees were still observed
(66-86%).
SCHEME 3. Hyp oth esis of Mich a el-Typ e
Mech a n ism for th e [Al(sa len )Cl]-Ca ta lyzed
1,4-Ad d ition
The optimization of the reaction conditions was then
followed by the study of the scope and generality of the
protocol. To this aim, we investigated the reaction of a
series of acyclic enones with 1a /lut (10 mol %) in toluene
at room temperature.25
The synthesis of the R,â-unsaturated aryl ketones was
readily accomplished following known procedures. In
particular, while enones 2e-j were obtained by classic
Friedel-Crafts acylation of the corresponding substituted
aromatic compounds (Scheme 4a),26a ketones bearing
strong electron-withdrawing substituents (2l-m ) and
ortho-substituted aryl groups (2k ) were synthesized in
three steps starting from the corresponding arylaldehyde
in moderate overall yields (13-25%) under nonoptimized
conditions (Scheme 4b).26b
Then, data collected in Table 3 show that R,â-unsatur-
ated substrates bearing electron-withdrawing substitu-
ents on the aromatic ring (2h -j, 2l, and 2m ) generally
allowed the indolyl derivatives to be isolated in excellent
chemical yields (90-98%) and good enantioselectivity.
In particular, pentafluoro derivative 2m smoothly
reacted with 3a , providing 4m a in 90% yield and 88%
enantiomeric excess (entry 9, Table 3). Ketones 2e and
2f bearing electron-rich aromatic rings bound to the
solution of non-negligible amounts of hydrochloric acid
that could promote the reaction through a nonstereo-
selective pathway (Scheme 3).
In light of these considerations, the effect of some base
scavengers on the chemical and optical outcome of the
1,4-addition of 3a to 2c was taken into account. The
results obtained, employing 10 mol % 1a as the catalyst,
are summarized in Table 2.
Compared to the base-free protocol (entry 1, Table 2),
the use of 10 mol % base generally reduced the rate of
the process but increased the enantioselectivity on the
desired â-indolyl ketone 4ca . Among all the additives
screened, 2,6-lutidine (lut) was the base of choice, afford-
ing 4ca in 65% yield and 79% ee (entry 5, Table 2).
Interestingly, lowering the loading of catalyst to 5 mol
% did not significantly affect either the chemical or
optical outcomes of the Michael addition, but longer
reaction times (72 h) were needed in order to reach
comparable conversion (entry 7). In contrast, the highly
sterically hindered 2,6-di-tert-butylpyridine (entry 8) and
the heterofunctionalized 2-cyano-pyridine (entry 9) led
to a surprising decrease in stereoselectivity. Such unex-
pected results and the remarkable correlation between
(24) (a) Miller, J . A.; J in, W.; Nguyen, S. T. Angew. Chem., Int. Ed.
2002, 41, 2953. (b) Cozzi, P. G.; Dolci, L. S.; Garelli, A.; Montalti, M.;
Prodi, L.; Zaccheroni, N. New J . Chem. 2003, 27, 692.
(25) Use of cyclic enones such as 2-cyclopenten-1-one and 2-cyclo-
hexen-1-one furnished the desired indolyl derivatives only in traces.
(26) (a) Oare, D. A.; Henderson, M. A.; Sanner, M. A.; Heathcock,
C. H. J . Org. Chem. 1990, 55, 132. (b) Toy, P. H.; Dhanabalasingam,
B.; Newcomb, M.; Hanna, I. H.; Hollenberg, P. F. J . Org. Chem. 1997,
62, 9114.
7514 J . Org. Chem., Vol. 69, No. 22, 2004

Catalytic Stereoselective Alkylation of Indoles
TABLE 4. Ster eoselective Ad d ition of In d oles to
SCHEME 4. Syn th esis of r,â-Un sa tu r a ted Ar yl
Keton es
(E)-En on es Ca ta lyzed by [Al(Sa len )Cl]/Lu t Com p lexa
entry T (°C) enone indole product yield (%)b ee (%)c
1
2
3
4
5
6
7
8
10
0
-20
-15
rt
rt
rt
rt
2c
2h
2i
3a
3a
3a
3a
3b
3c
3b
3c
4ca
4h a
4ia
48
25
68
78
35
41
72
67
84 (R)
82 (R)
89 (R)d
84 (R)
64
2j
4ja
2c
2c
2j
4cb
4cc
4jb
64
64
2m
4m c
80e (R)
a
All reactions were carried out employing 10 mol % [Al(salen)-
Cl] and 10 mol % lutidine for a reaction time of 48 h unless
b
otherwise specified. Chemical yields are given for the isolated
product after chromatographic purification. c Enantiomeric ex-
cesses were determined by HPLC analysis with chiral column
d
(Chiralcel OD). Reaction time ) 96 h. e Reaction was carried out
by using 20 mol % catalyst.
TABLE 3. Ster eoselective Ad d ition of 3a to (E)-En on es
Ca ta lyzed by [Al(Sa len )Cl]/Lu t Com p lex
reaction between 2m and 3c, the indolyl derivative 4m c
was isolated in 67% yield and 80% ee (entry 8).
entrya
enone
product
yield (%)b
ee (%)c
1
2
3
4
5
6
7
8
9
2e
2f
2g
2h
2i
2j
2k
2l
4ea
4fa
4ga
4h a
4ia
4ja
4k a
4la
80
58
nrd
96
98
92
27
95
90
73 (R)
49 (R)
Mech a n istic Con sid er a tion s. In an effort to ascer-
tain the precise role of the basic additive in the reaction
course, H NMR spectra of an equimolar solution of 1a
1
78 (R)
80 (R)
78 (R)
32 (R)
77 (R)
88 (R)
and base (i.e., Et3N, lut) were collected. Such an analysis
confirmed that a rapid and quantitative complexation
occurred when the additive and the aluminum complex
were dissolved in CD2Cl2 at room temperature. In par-
ticular, the signals of the free Al-Schiff base complex
disappeared to give new sets of signals representative
for the formation of a new Al-base species. For the case
of Et3N, a comparison of some diagnostic signals of 1a
(Figure 4a) and 1a ‚Et3N (in brackets, Figure 4b) is as
follows: δ 8.34[7.80] (CHdN), 7.58[7.41] (Ar), 7.18[7.03]
(Ar).
2m
4m a
a
All reactions were carried out at room temperature, employing
10 mol % [Al(salen)Cl] and 10 mol % lutidine for a reaction time
b
of 48 h. Chemical yields are given for the isolated product after
chromatographic purification. c Enantiomeric excesses were de-
termined by HPLC analysis with chiral column (Chiralcel OD).
Absolute configuration of adducts 4 was assigned as R by analogy
(elution times of the HPLC analysis and optical rotation) to 2ca .
d
No reaction.
1
To validate this hypothesis, we recorded the H NMR
spectra in the presence of sterically demanding and
noncoordinating 2,6-ditBu-py, and in this case, no sig-
nificant base-aluminum interaction was observed. On
the other hand, the spectra recorded when lutidine was
added to a solution of 1a in CD2Cl2 appeared to be
fluxional and more difficult to interpret, indicating that
some complex equilibria are taking place in solution.
Then, further insight into the postulated mechanism
and into the key structures involved in the reactive
process was obtained by carrying out accurate, uncon-
strained optimizations at a fully correlated level (DFT)
on a realistic model for the [Al(salen)]-catalyst/coordinat-
ing-additive/substrate-enone system. The optimized low-
est energy minima likely involved in the key steps of the
present asymmetric catalysis are shown in Figure 5 (the
higher energy conformers are not shown here).
Interestingly, while the model carbonyl substrate (3-
methyl-acrolein) and additive-base model used (trimeth-
ylamine, TMA) generate stable cationic pentacoordinate
[Al(salen)] complexes 6a ,b, the corresponding neutral
(i.e., chloride-bound) hexacoordinate systems are un-
stable, in agreement with the relative ease of displacing
chloride groups, as observed in similar [Al(salen)]
complexes.12b,20 On the other hand, when a coordinating
base (TMA) is used together with the enone, a very stable
cationic hexacoordinate complex 6c can be formed (where
the enone and the base are both bound to Al according
to a trans arrangement). This system could be considered
the hypothetical starting point for the reaction with
carbonyl moiety showed lower reactivity under the
present reaction conditions, affording 4ea and 4fa in 73
and 49% enantiomeric excess, respectively. On the other
hand, bulky mesitylcrotyl ketone 2g and oBr-phenyl
derivative 2k were not found to be suitable substrates
for the alkylation of 2-methylindole probably due to the
obstructed coordination of the carbonyl with the chiral
aluminum catalyst (entries 3, 7).
The effect of the temperature on the efficiency of 1a /
lut to catalyze this stereoselective FC-type protocol was
further examined. To this purpose, more reactive R,â-
unsaturated ketones were reacted with 3a at different
temperatures, and the results obtained are reported in
Table 4 (entries 1-4).
In all cases, the enones 2 underwent the FC process
in good yield and moderate to good enantioselectivity. For
instance, ketone 2i smoothly reacted with 3a at -20 °C,
furnishing the corresponding 1,4-adduct 4ia in 89% ee
and 68% yield after 96 h (entry 3). Analogous results were
obtained with 2c, 2h , and 2j at temperatures ranging
from -15 to 0 °C. The capacity of complex 1a /lut to
effectively promote the conjugate addition of various
indoles to aryl enones was further evaluated by reacting
5-methoxyindole 3b and indole 3c at room temperature
with carbonyls 2c, 2j, and 2m (entries 5-7, Table 4). In
these cases, the desired 1,4-adducts were isolated in
satisfactory yields and slightly lower enantiomeric excess
(64%); however, by using 20 mol % catalyst in the
J . Org. Chem, Vol. 69, No. 22, 2004 7515

Bandini et al.
F IGURE 4. (a) 1H NMR spectra of commercial [Al(salen)Cl] (CD2Cl2, rt). (b) 1H NMR spectra recorded after the addition of TEA
(1:1 ratio) (CD2Cl2, rt).
F IGURE 5. Structure of the calculated cationic pentacoordinate (6a and 6b) and hexacoordinate (6c) [Al(salen)] complexes with
the enone alone (6a ), the coordinating base alone (6b), and the enone plus the bases (TMA: 6c). Hydrogen atoms have been
removed for clarity. The dissociated chloride counteranion is not illustrated here. Distances and angles are in Å and degrees,
respectively.
indole. Remarkably, a significant thermodynamic driving
force exists for nonsterically congested (i.e., coordinating)
bases, which pushes the [Al(salen)]-catalyst/coordinating-
base/substrate-enone system toward the cationic trans
hexacoordinate complex. Indeed, this is the lowest energy
minimum optimized so far, the relative energies for the
three optimized structures 6c, 6a , 6b being 0.0, 7.8, and
12.0 kcal/mol, respectively. On the contrary, with a
sterically congested base (i.e., noncoordinating, such as
2,6-ditBu-py), no stable hexacoordinate complexes are
Although the mechanism for the reaction is not unveiled
here, these results still agree with the available observa-
tions (i.e., NMR analysis and base-coordinating-activity/
enantioselectivity correlation). In particular, the overall
conformation change of the hexacoordinate catalytic
species led the R,â-unsaturated compound to closer
vicinity at the tBu-groups, favoring better stereodiscrimi-
nation during the indole attack at the less shielded
enantiotopic face of the activated electrophile.
In the FC reactions, the role of Lewis acids is generally
limited to the formation of highly reactive electrophiles
that subsequently react with the aromatic compound.
However, recent experimental27 as well as theoretical13
studies proved that the Lewis acid can play an additional
role in the reaction course by interacting also with the
aromatic compound, significantly increasing its nucleo-
philicity. In this context, of particular relevance is the
colorimetric observation reported by Ottoni and co-
workers in the Lewis acid (AlCl3, TiCl4, SnCl4)-mediated
FC acylation of indoles. In this case, the formation of a
highly reactive indolyl-metal species was supposed to be
the real nuclophile of the FC process. To shed some light
on the operative role of [Al(salen)Cl] (1a ) in the catalytic
cycle of the present FC-type alkylation of indoles, we
carried out a kinetic study investigating the dependence
of the reaction rate on the concentration of 1a . The initial
1
formed in solution (consistent with H NMR spectra, no
stable hexacoordinate complexes could be located using
2,6-ditBu-py). Moreover, by means of lutidine as a
coordinating base (which is sterically more congested
than Me3N, although not as much as 2,6-ditBu-py), a
significantly longer Al-N bond is computed, revealing a
slightly bound complex and a possible equilibrium be-
tween bound and unbound forms, in agreement with the
1
recorded H NMR spectra (fluxional set of signals were
recorded at room temperature in CD2Cl2).
As previously described for Mn-salen complexes,17 and
also in the case of Al-Schiff systems, the presence of an
axial ligand profoundly affects the overall conformation
of the salen ligand. In fact, while the minimized struc-
tures of the pentacoordinate Al-complexes (6a ,b) showed
the typical steplike conformation, the simultaneous pres-
ence of a carbonyl compound as well as a base trans-
coordinated to the aluminum center (cationic hexacoor-
dinate species) evidenced a cup-shaped conformation (6c).
(27) Ottoni, O.; Neder, A. de V. F.; Dias, A. K. B.; Cruz, R. P. A.;
Aquino, L. B. Org. Lett. 2001, 3, 1005.
7516 J . Org. Chem., Vol. 69, No. 22, 2004

Catalytic Stereoselective Alkylation of Indoles
SCHEME 5. Absolu te Con figu r a tion Assign m en t
for 4cc
F IGURE 6. NLE curve of the [Al(salen)Cl] catalyzed Michael
addition of indoles to arylenones.
reaction rates of the addition of 2-methylindole to the
model ketone 3c, calculated at different concentrations
[C]0 of [Al(salen)Cl] complex, were measured following
each run from 8% completion to 18-25% conversion, and
all the conjugate additions were monitored up to at least
75% conversion. From the trends of the curves recorded
at different catalyst concentrations (7.5-22.5 mM, see
Supporting Information), it appears that the rate of FC
reaction was strongly affected by the total [Al(salen)Cl]
concentration. The reaction order with respect to the
catalyst was obtained by plotting log[rate] versus
log[C]0.
The first-order dependence of the reaction rate on the
concentration of 1a clearly suggested that, in the rate-
determining step of the catalytic cycle (nucleophilic
addition of the indole to the â-position of the enone), a
single molecule of [Al(salen)Cl] complex is involved,
acting as the activator of the carbonyl substrate. This
finding ruled out the possible double role played by the
catalyst, that could operate as a concomitant electro-
philic/nucleophilic activator. More evidence on the real
role of the aluminum catalyst in the reaction course came
from studies concerning the NLE of the enantioselective
conjugate addition of 3a with 2c in the presence of 1a -
lut (10 mol %) as the catalyst system.28 In fact, the
dependence of the enantiomeric excess (ee) of the product
(4ca ) on the optical purity of the catalyst (1a ) was
examined and reported in Figure 6.
To this purpose, by the asymmetric organo-catalyzed
FC alkylation reaction of indoles described by MacMillan
and co-workers,10 we first synthesized the (R)-â-indolyl
aldehyde 4n d in 70% yield and 55% enantiomeric excess
by reacting N-Me-indole (3d ) with (E)-crotonaldehyde 2n
in the presence of imidazolidinone 6 and TFA (20 mol %,
Scheme 5).
Then, 4n d was easily converted into the corresponding
indolyl alcohol (4n d , yield 82%) as a diasteoisomeric
mixture (60:40) by reaction with PhMgBr in Et2O at
0 °C. Finally, the unresolved diastereomeric mixture was
converted to (R)-4cd by oxidation with MnO2, in CH2Cl2
(yield 60%, ee 55%). Comparison of both the optical
rotation value and chiral HPLC retention times with 4cd
(obtained by NH-methylation of 4cc) proved R to be the
absolute configuration of the 1,4-adduct 4cc. The absolute
configuration of the indolyl ketones 4 was assigned by
analogy.
Con clu sion
The absence of NLE in our process allows us to
speculate that a single molecule of catalyst is involved
in the stereochemistry-determining step of the reaction
pathway, supporting the single activation effect previ-
ously proposed.
In summary, we have developed a novel, effective
catalytic Friedel-Crafts-type alkylation of indoles through
1,4-addition to aryl ketones in the presence of [Al(salen)-
Cl] (10 mol %). The optical outcome of the process can be
significantly increased by adding a catalytic amount of
coordinating base. Theoretical as well as spectroscopic
evidence suggests that an interaction between the addi-
tive and the aluminum center is involved, leading to new
highly stereoselective chiral catalysts. Particular atten-
tion has been devoted to the generality of the process that
guarantees the synthesis of a range of â-indolyl ketones
in excellent yields and good enantiomeric excesses.
Computational evidence allows formulation of a mecha-
nistic hypothesis that is in agreement with the experi-
mental observations and NMR data. In particular, a
stable cationic hexacoordinate complex, formed by [Al-
(salen)] with the substrate and the additive base (in a
trans configuration), has been computed and suggested
as being responsible for the enhanced enantioselectivity.
Finally, the stability of the active catalytic species
during the entire reaction course of the stereoselective
addition of 3a to 2c was confirmed by monitoring the
enantiomeric excess of periodic aliquots of the reaction
mixture by HPLC analysis with chiral OD column. In
fact, the level of stereoinduction was found to be nearly
constant at about 76-79% (see Supporting Information).
Absolu te Con figu r a tion Assign m en t. In efforts to
establish the absolute configuration of the 1,4-adducts
4, we undertook a chemical correlation between the
1-phenyl-3-(1H-indol-3-yl)-butane-1-one 4 cc3 (ee ) 64%,
[R]D ) +7.4, c 1 CHCl3, Table 4, entry 6) and the known
(R)-3-(N-methyl-indol-3-yl)-butyraldehyde (4n d ).
(28) Girard, C.; Kagan, H. B. Angew. Chem., Int. Ed. 1998, 37, 2922.
J . Org. Chem, Vol. 69, No. 22, 2004 7517

Bandini et al.
Studies concerning the effectiveness of this [Al]‚base
system in new Lewis acid-catalyzed transformations are
currently under way.
potenziali farmaci innovativi) and from University of
Bologna (funds for selected research topics).
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
cedures and analytical and spectral characterization data for
all the compounds. This material is available free of charge
via the Internet at http://pubs.acs.org.
Ack n ow led gm en t. This work was supported by
M.I.U.R. (Rome), National project “Stereoselezione in
Sintesi Organica: Metodologie and Applicazioni”, FIRB
Project (Progettazione, preparazione e valutazione bio-
logica e farmacologica di nuove molecole organiche quali
J O0487202
7518 J . Org. Chem., Vol. 69, No. 22, 2004