Catalytic Generation of Indium Hydride in a Highly Diastereoselective Reductive Aldol Reaction

Article abstract of DOI:10.1002/anie.200352738

A one-pot synthesis of β-hydroxyketones has been achieved by the reductive aldol reaction of enones and aldehydes in the presence of Et ₃SiH and a catalytic amount of InBr₃ (see scheme). The reaction is highly diastereoselective and no reduction of the starting aldehydes occurs. This method is superior to conventional reductive aldol reactions in terms of applicability to a wide range of enones and aldehydes.

Full text of DOI:10.1002/anie.200352738

Angewandte
Chemie
Reductive Aldol Reaction
Catalytic Generation of Indium Hydride in a
Highly Diastereoselective Reductive Aldol
Reaction**
Ikuya Shibata,* Hirofumi Kato, Tatsuya Ishida,
Makoto Yasuda, and Akio Baba*
The reductive aldol reaction of metal hydrides, enones, and
aldehydes is a valuable route to b-hydroxyketones because it
is a convenient one-pot method without the need to
synthesize metal enolates. However, a difficulty of this
reaction is that aldehydes are more sensitive than enones to
[*] Dr. I. Shibata, H. Kato, T. Ishida, Dr. M. Yasuda, Prof. Dr. A. Baba
Department of Molecular Chemistry
Science and Technology Center for Atom, Molecules and Ions
Control
Graduate Schoolof Engineering, Osaka University
2-1 Yamadaoka, Suita, Osaka 565-0871 (Japan)
Fax: (+81)66879-7387
E-mail: shibata@chem.eng.osaka-u.ac.jp
baba@chem.eng.osaka-u.ac.jp
[**] This research was carried out at the Strategic Research Base
“Handai Frontier Research Center” supported by the Japanese
Government's SpecialCoordination Fund for Promoting Science
and Technology, and was partially supported by a Grant-in-Aid for
Scientific Research from the Ministry of Education, Science, Sports,
and Culture.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2004, 43, 711 –714
DOI: 10.1002/anie.200352738
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
711

Communications
conventional metal hydrides.[1] Furthermore, reagents that
promoted reaction in Scheme 2 has opposite diastereoselec-
tivity and gives syn-aldolate 3a even under aprotic conditions.
Moreover, the result given in Scheme 2 is noteworthy because
the reduction of aldehyde scarcely occurred.
achieve high diastereoselectivity have scarcely been reported.
We have already reported the initiation of reductive aldol
reactions by dihaloindium hydrides (Cl2InH and Br2InH),
generated by transmetalation of nBu3SnH and InIII halides
InX3.[2] However, since indium hydrides should be generated
prior to the addition of enones or aldehydes, equimolar
amounts of InX3 must be treated with nBu3SnH before the
reaction.[3,4] Here we report a superior method in which no
presynthesis of indium hydrides is required and where
nBu3SnH is not used. In particular, the catalytic use of
InBr3 was acieved (Scheme 1).
Next, we investigated the generation of indium hydride by
transmetalation. Et3SiH/InCl3 was chosen for comparison
with our previously reported generation of Cl2InH from
nBu3SnH/InCl3.[6] When InCl3 and Et3SiH were mixed in
CD3CN at À108C for 5 min, 1H NMR analysis showed a new
peak at d = 6.6 ppm besides the Et3SiH peak at d =
3.6 ppm.[6a] The peak at d = 6.6 ppm is consistent with our
previously reported value for Cl2InH generated from the
nBu3SnH/InCl3 system, for which transmetalation occurred
smoothly at À788C in THF. After stirring the solution of
Et3SiH/InCl3 at À108C for 20 min, the peak of Cl2InH
gradually decreased because of its instability, while the peak
1
of Et3SiH still remained. When the H NMR analysis was
performed in [D8]THF, no peaks other than that of Et SiH
3
were observed. Although the transmetalation was slow
compared with the nBu3SnH/InCl3 system, it was found that
Cl2InH was generated from Et3SiH/InCl3 in nitrile solvents.
Thus, the effective reaction shown in Scheme 2 is explain-
able: Indium hydride generated in situ promotes 1,4-reduc-
tion of enone 1a to the indium enolate, which gives aldolate 3
by reaction with aldehyde 2. The formation of syn-aldolates
3a indicates that immediate trapping of kinetically controlled
syn-indium aldolate would occur by Et3SiH. This result
suggests the possibility of using indium halides as catalysts,
because the silicon trapping agent generates indium hydride.
Next we investigated the catalytic use of indium halides.
As shown in Table 1, the use of InCl3 (10 mol%) resulted in
an unsatisfacory yield of silyl aldolate 3a (entry 1). However,
3a was obtained in 75% yield when InBr3 was used as the
catalyst (entry 2). p-Nitrobenzaldehyde and benzaldehyde
gave 3b and 3c, respectively (entries 3 and 4). Aliphatic
aldehydes were also applicable to give 3d–f (entries 5–7). It is
noteworthy that no reduction of aldehydes 2 took place in
these cases. Thus, the present system exhibits high chemo- and
regioselectivity for enones. As mentioned previously, these
results represent an advantage over the nBu3SnH/InBr3
system, which is seriously limited to the reaction with p-
anisaldehyde because the system could not prevent the
reduction of electrophilic aldehydes such as p-nitrobenzalde-
hyde. Enones bearing aromatic and aliphatic substituents
were also reactive (entries 8–12).
Scheme 1. Reductive aldol reaction.
We focused on hydrosilanes as hydride sources instead of
nBu3SnH. Active metal hydrides such as NaBH4 and LiAlH4
are not appropriate as they readily reduce aldehydes in the
absence of InX3. Trialkyl silanes are stable liquids that are
easy to handle and have low toxicity.[5] They have no reactivity
towards carbonyl compounds in the absence of additives.[5b]
When transmetalation with InX3 occurs, only indium hydrides
could act as reactive species in reactions with electrophiles.
Initially, we applied the Et3SiH/InCl3 system to the
reductive aldol reaction. Thus, 1-phenyl-2-buten-1-one (1a,
1.2 mmol) and p-methoxybenzaldehyde (2, 1 mmol) were
added in one portion to a solution of InCl3 (1 mmol) and
Et3SiH (1.2 mmol) in THF, but only a trace of silyl aldolate3a
was obtained. However, when the reaction was performed in
EtCN, the yield of 3a was increased to 59% based on 2
(Scheme 2).
A plausible catalytic cycle for InBr3 is shown in Scheme 3.
Initially, Br2InH is generated by the slow transmetalation of
InBr3 with Et3SiH. The generated Br2InH next undergoes 1,4-
addition with enone 1 to give indium enolate A. In this step,
Br2InH does not reduce the coexistent aldehydes 2. The
absence of aldehyde reduction is due to the low concentration
of Br2InH from the slow transmetalation. In contrast, in the
case of nBu3SnH/InBr3, Br2InH is formed in high concen-
tration because of the easy transmetalation of nBu3SnH with
InBr3 and the equimolar reaction, which was accompanied by
partial reduction of the aldehydes. (Z)-Indium enolate A can
be considered to be generated initially because of the
preferred 1,4-addition of indium hydride to the s-cis form of
enone 1.[7] Indium enolate A reacts with 2 via a Zimmerman–
Scheme 2. Reductive aldol reaction with an equimolar amount of InCl3.
For the nBu3SnH/InX3 system, we earlier reported the
formation of thermodynamically stable anti-aldol adducts
under aprotic conditions.[2] Furthermore, a serious problem
was the reduction of the starting aldehydes, which gave 4 as an
unavoidable side product. In contrast, the hydrosilane-
712
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 711 –714

Angewandte
Chemie
Table 1: Diastereoselective reductive aldol reactions.[a]
Traxler six-membered cyclic transi-
tion state[8] to form syn-indium
aldolate B, which is immediately
trapped by Et3SiH. Thus, syn-silyl
aldolate 3 is obtained with regena-
tion of Br2InH. In this step, there is
the possibility of trapping of indium
aldolate B with Et3SiBr to regener-
ate InBr3, because Et3SiBr is pro-
duced by the initial transmetalation
step. However, we consider that the
trapping of indium aldolate A by
Et3SiH is plausible considering the
initial concentrations of Et3SiH and
Et3SiBr. Moreover, it was con-
firmed that the Et3SiBr-free
indium aldolate generated by
Entry
Enone 1
Aldehyde 2
Product
Yield [%]
syn:anti
1
2
3a
3a
33[b]
75
>99:1
90:10
3
3b
59
>99:1
4
5
6
7
PhCHO
3c
3d
3e
3 f
78
40
73
87
92:8
>99:1
>99:1
90:10
PhCH2CHO
tBuCHO
Et2CHO
8
3g
3h
3i
82
>99:1
78:22
[2]
nBu3SnH/InBr3 is easily trapped
9
PhCHO
PhCHO
65[c]
61
by Et3SiH (Scheme 4).
In summary, the Et3SiH-pro-
moted diastereoselective reductive
aldol reaction has been established
by using InBr3 as a catalyst. This
three-component reaction afforded
only silyl aldolates as products with-
out any side reactions. The silicon
compounds, including the Et3SiH
10
11
12
>99:1
>99:1
>99:1
3j
40
3k
46
[a] Conditions: Et3SiH (1.2 mmol), InBr3 (0.1 mmol), enone 1 (1 mmol), EtCN (1 mL). [b] InCl3 was used
instead of InBr3. [c] Product was isolated as desilylated aldol adduct.
starting material, could be easily removed by evaporation
after the reaction. The syn selectivity obtained here is higher
than that of any other reductive aldol reaction, including
those promoted by tin hydride.
Received: August 28, 2003 [Z52738]
Keywords: aldol reaction · diastereoselectivity · hydroxyketones ·
.
indium · silanes
[1] a) I. Matsuda, K. Takahashi, S. Sato, Tetrahedron Lett. 1990, 31,
5331 – 5334; b) E. J. Enholm, Y. Xie, K. A. Abboud, J. Org. Chem.
1995, 60, 1112 – 1113; c) T. Kawakami, M. Miyatake, I. Shibata, A.
Baba, J. Org. Chem. 1996, 61, 376 – 379; d) T. Ooi, K. Doda, D.
Sakai, K. Maruoka, Tetrahedron Lett. 1999, 40, 2133 – 2136; e) S. J.
Taylor, M. O. Duffey, J. P. Morken, J. Am. Chem. Soc. 2000, 122,
4528 – 4529.
[2] K. Inoue, T. Ishida, I. Shibata, A. Baba, Adv. Synth. Catal. 2002,
344, 283 – 287.
Scheme 3. Plausible catalytic cycle.
[3] The potential of indium compounds in the field of organic
synthesis has been intensively studied. Barbier-type allylation:
a) S. Araki, H. Ito, Y. Butsugan, J. Org. Chem. 1988, 53, 1831 –
1833; b) C. J. Li, T. H. Chan, Tetrahedron Lett. 1991, 32, 7017 –
7020; c) C. J. Li, T. H. Chan, Tetrahedron 1999, 55, 11149 – 11176;
d) C. J. Li, Tetrahedron Lett. 1995, 36, 517 – 518; e) M. B. Isaac,
T. H. Chan, Tetrahedron Lett. 1995, 36, 8957 – 8960; f) L. A.
Paquette, T. M. Mitzel, J. Org. Chem. 1996, 61, 8799 – 8804; Pd-
catalyzed allylation: g) S. Araki, T. Kamei, T. Hirashita, H.
Yamamura, M. Kawai, Org. Lett. 2000, 2, 847 – 849. Pinacol
coupling h) H. J. Lim, G. Keum, S. B. Kang, B. Y. Chung, Y. Kim,
Tetrahedron Lett. 1998, 39, 4367 – 4368; Transmetalation: i) J. A.
Marshall, K. W. Hinkle, J. Org. Chem. 1995, 60, 1920 – 1921;
j) J. A. Marshall, K. W. Hinkle, J. Org. Chem. 1996, 61, 105 – 108;
k) M. Yasuda, T. Miyai, I. Shibata, R. Nomura, H. Matsuda, A.
Scheme 4. Trapping of indium aldolate by a trialkyl silane.
Angew. Chem. Int. Ed. 2004, 43, 711 –714
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
713

Communications
Baba, Tetrahedron Lett. 1995, 36, 9497 – 9500; l) T. Miyai, K.
Inoue, M. Yasuda, A. Baba, Synlett 1997, 699 – 700.
[4] Although several indium hydrides have been reported, they have
scarcely been used in organic synthesis. LiPhnInH4Àn (n = 0–2):
a) M. Yamada, K. Tanaka, S. Araki, Y. Butsugan, Tetrahedron
Lett. 1995, 36, 3169 – 3172; b) M. Yamada, T. Horie, M. Kawai, H.
Yamamura, S. Araki, Tetrahedron 1997, 53, 15685 – 15690;
InH3[CN(Mes)C2H2N(Mes)] and InH3[P(C6H11)3]: c) C. D. Aber-
nethy, M. L. Cole, A. J. Davies, C. Jones, Tetrahedron Lett. 2000,
41, 7567 – 7570.
[5] a) N. Greeves in Comprehensive Organic Synthesis, Vol. 8 (Eds.:
B. M. Trost, I. Fleming), Pergamon, Oxford, 1991, p. 1; b) L. Fry,
M. Orfanopoulos, M. G. Adlington, W. R. Dittman, S. B. Silver-
man, J. Org. Chem. 1978, 43, 374 – 375.
[6] We reported the first generation of Cl2InH, which showed a
1H NMR peak at d = 6.5 ppm in THF. a) T. Miyai, K. Inoue, M.
Yasuda, I. Shibata, A. Baba, Tetrahedron Lett. 1998, 39, 1929 –
1932; b) K. Inoue, M. Yasuda, I. Shibata, A. Baba, Tetrahedron
Lett. 2000, 41, 113 – 116; c) K. Inoue, A. Sawada, I. Shibata, A.
Baba, Tetrahedron Lett. 2001, 42, 4661 – 4663.
[7] a) G. P. Boldrini, F. Mancini, E. Tagliavini, C. Trombini, A.
Umani-Ronchi, J. Chem. Soc. Chem. Commun. 1990, 1680 – 1681;
b) G. P. Boldrini, M. Bortolotti, F. Mancini, E. Tagliavini, C.
Trombini, A. Umani-Ronchi, J. Org. Chem. 1991, 56, 5820 – 5826.
[8] H. E. Zimmerman, M. D. Traxler, J. Am. Chem. Soc. 1957, 79,
1920 – 1923.
714
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 711 –714