Metal-free catalytic reduction of aldehydes, ketones, aldimines, and ketimines

Article abstract of DOI:10.1016/j.cclet.2010.05.029

The metal-free combination of catalytic amounts of PPh₃, B(C₆F₅)₃, and PhSiH₃ can efficiently hydrosilylate aldehydes, ketones, aldimines and ketimines to afford the corresponding reduction products in good yields.

Full text of DOI:10.1016/j.cclet.2010.05.029

Available online at www.sciencedirect.com
Chinese Chemical Letters 21 (2010) 1314–1317
www.elsevier.com/locate/cclet
Metal-free catalytic reduction of aldehydes, ketones,
aldimines, and ketimines
Hiroaki Matsuoka a, Kazuhiro Kondo a,b,
*
a Graduate School of Medical Sciences, Nagoya City University, Nagoya 467-8601, Japan
b Graduate School of Pharmaceutical Sciences, Nagoya City University, Nagoya 467-8603, Japan
Received 18 March 2010
Abstract
The metal-free combination of catalytic amounts of PPh3, B(C6F5)3, and PhSiH3 can efficiently hydrosilylate aldehydes,
ketones, aldimines and ketimines to afford the corresponding reduction products in good yields.
# 2010 Kazuhiro Kondo. Published by Elsevier B.V. on behalf of Chinese Chemical Society. All rights reserved.
Keywords: Metal-free hydrosilylation; Reduction; Ketone; Ketimine
Reduction of C O and C N bonds to alcohols and amines, respectively, is an important reaction in organic
synthesis [1]. Among the available reduction methods, hydrosilylation is very useful because of its less excessive
reaction, as compared with other reduction methods such as the one that uses LiAlH4. Thus far, considerable research
has focused on the development of hydrosilylation using metals [1]. Herein, we report the metal-free catalytic
reduction (hydrosilylation) [2–4] of aldehydes, ketones, aldimines and ketimines.
1. Results and discussion
We investigated the combinations of non-metal Lewis acids and bases in the activation of a substrate and a silane,
respectively [5]. After intensive screening, the combination of B(C6F5)3 as a Lewis acid, PPh3 as a Lewis base [6],
PhSiH3 as a hydride source and toluene as a solvent was found to afford the best yield: the reaction conversion within
2 mol equiv of PhSiH3 was low. Representative screening results are shown in entries 1–8 of Table 1. Absence of PPh3
or B(C6F5)3 resulted in no reaction (entries 7 and 8). Although bis-phosphines were also found to be effective as shown
in entries 9 and 10, PPh3 was chosen from the cost perspective.
As shown in Table 2, the reaction under the optimized conditions proved to be general and may be applied to a broad
range of ketones, aldehydes, aldimines and ketimines. Specifically, the reaction proceeds smoothly under mild
conditions with a synthetically acceptable catalyst loading (5 mol.%). Various acetophenone derivatives 1b–1e,
aliphatic acyclic and cyclic ketones 1f–1 h, aromatic aldehydes, aliphatic and a,b-unsaturated aldehydes 1i–1o,
* Corresponding author.
E-mail address: kazu@gakushikai.jp (K. Kondo).
1001-8417/$ – see front matter # 2010 Kazuhiro Kondo. Published by Elsevier B.V. on behalf of Chinese Chemical Society. All rights reserved.
doi:10.1016/j.cclet.2010.05.029

H. Matsuoka, K. Kondo / Chinese Chemical Letters 21 (2010) 1314–1317
1315
Table 1
Effects of differing silane, phosphine and solvent on reaction yield.
[DT$INLE]
.
Entry
Silane
Phosphine
Solvent
Yield (%)
1
2
(EtO)3SiH
Ph2SiH2
PhSiH3
PhSiH3
PhSiH3
PhSiH3
PhSiH3
PhSiH3
PhSiH3
PhSiH3
PPh3
Toluene
Toluene
Toluene
DMF
Trace
74
PPh3
3
PPh3
90
4
PPh3
Trace
25a
5
PPh3
Dioxane
Toluene
Toluene
Toluene
Toluene
Toluene
6
PCy3
None
PPh3
Trace
Trace
Trace
83
7
8b
9
dppe
10
(Æ)-BINAP
87
a
TLC revealed the major component to be the starting ketone.
No B(C6F5)3 was used.
b
Table 2
Substrate generality.
[TD$INLINE]
.
Entry
Substrate
Time (h)
Yield (%)
1
2
1
1
90
79
[DT$INLE]
3
1
92
[DT$INLE]
4
5
1
1
86
93
[DT$INLE]
[DT$INLE]
6
7
1
1
93
87
[DT$INLE]
[DT$INLE]

1316
H. Matsuoka, K. Kondo / Chinese Chemical Letters 21 (2010) 1314–1317
Table 2 (Continued )
Entry
Substrate
Time (h)
1
Yield (%)
91
8
[DT$INLE]
9
1
1
93
90
[DT$INLE]
10
[DT$INLE]
11
12
1
3
90
90
[DT$INLE]
[DT$INLE]
13
14
1
1
93
90
[DT$INLE]
[DT$INLE]
15
16
17
1
1
2
91
95
90
[DT$INLE]
18
6
87
[DT$INLE]
19
20
1
4
94
90
[DT$INLE]
[DT$INLE]
aldimines 1p–1s, and ketimines 1t and 1u were employable, affording the corresponding reduction products in good
yields. The a,b-unsaturated aldehydes 1n and 1o and the aldimine 1s were reduced in only the 1,2-mode.
Although Piers’ group has previously reported the hydrosilylation of carbonyl compounds with the combination of
Ph3SiH and a catalytic amount of B(C6F5)3 [3], our protocol has the following advantages as follows: (1) Unlike Piers’
protocol, slow addition over 1 h of silane is not required. (2) From the viewpoint of atom economy, the use of PhSiH3 is
more desirable than the use of Ph3SiH. (3) The process may be applied to aldimines, ketimines and easily enolizable b-
teralone (1h). Considering Piers’ and Oestreich’s results [3,4] and those reported herein [7], we assume that the
mechanism for the reduction is as follows (Scheme 1). Initially, both B(C6F5)3 and PPh3 activate the Si–H bond to form
A. The carbonyl- and imino-substrates are then activated by the Si group and the activated substrate C is attacked by
H–SB(C6F5)3 B, affording the corresponding product.

[(Scheme_1)TD$FIG]
H. Matsuoka, K. Kondo / Chinese Chemical Letters 21 (2010) 1314–1317
1317
Scheme 1. Plausible mechanism.
In summary, the combination of B(C6F5)3 and PPh3 was found to function as an effective catalyst in the reduction
(hydrosilylation) of aldehydes, ketones, aldimines and ketimines with PhSiH3 [8]. Development of an asymmetric
version is now in progress [9–11].
References
[1] P.G. Andersson, I.J. Munslow, Modern Reduction Methods, Wiley-VCH, New York, 2008.
[2] M. Tan, Y. Zhang, J.Y. Ying, Adv. Synth. Catal. 351 (2009) 1390.
[3] D.J. Parks, J.M. Blackwell, W.E. Piers, J. Org. Chem. 65 (2000) 3090.
[4] S. Rendller, M. Oestreich, Angew. Chem. Int. Ed. 47 (2008) 5997.
[5] M. Shibasaki, Enantiomer 4 (2000) 513.
[6] V. Sumerin, F. Schultz, M. Nieger, et al. Angew. Chem. Int. Ed. 6001 (2008) 47.
[7] When B(C6F5)3 (1 mol. equiv. to 1a) was added to the solution of PPh3 (1 mol. equiv. to 1a) in toluene-d8, the signal in the 11B NMR changed
from 42.91 ppm to À0.52 ppm. Further, when ketone 1a was added to this solution, no change of the signal change was observed. This result
implies that the complex of B(C6F5)3 and PPh3, even in the presence of ketone 1a, is formed preferentially.
[8] Representative procedure for the hydrosilylation (entry 3, Table 1). To a stirred solution of PPh3 (2.60 mg, 0.0100 mmol) in toluene (0.50 mL),
B(C6F5)3 (5.40 mg, 0.0100 mmol), 1a (34.0 mg, 0.200 mmol) and PhSiH3 (49 mL, 0.400 mmol) were added at rt. The reaction mixture was
stirred for 1 h at 30 8C and diluted with 1 mol /L NaOH. After usual work-up, purification by SiO2 column (hexane:EtOAc = 4:1) afforded 1-
(naphthalen-2-yl)ethanol (31.0 mg, 90%) as a white solid.
[9] C.T. Lee, B.H. Lipshutz, Org. Lett. 10 (2008) 4187.
[10] N.S. Shaikh, S. Enthaler, K. Junge, et al. Angew. Chem. Int. Ed. 47 (2008) 2497.
[11] T. Inagaki, Y. Yamada, L.T. Phong, et al. Synlett (2009) 253.