Asymmetric reduction of ketimines with trichlorosilane employing an imidazole derived organocatalyst

Article abstract of DOI:10.1039/b816051a

Organocatalysts for the asymmetric reduction of ketimines are presented that function well at low catalyst loadings providing chiral amines in good yield and enantioselectivity, the latter appearing to be independent of the ketimine substrate geometry.

Full text of DOI:10.1039/b816051a

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/obc | Organic & Biomolecular Chemistry
Asymmetric reduction of ketimines with trichlorosilane employing an
imidazole derived organocatalyst†
Franc¸ois-Moana Gautier,a Simon Jones*a and Stephen J. Martinb
Received 17th September 2008, Accepted 30th October 2008
First published as an Advance Article on the web 19th November 2008
DOI: 10.1039/b816051a
Organocatalysts for the asymmetric reduction of ketimines
are presented that function well at low catalyst loadings
providing chiral amines in good yield and enantioselectivity,
the latter appearing to be independent of the ketimine
substrate geometry.
Catalysts 5–7 were prepared in acceptable yield for evaluation
purposes by condensation of the appropriate prolinol derivative
with ethyl 1-methylimidazole-2-carboxylate (Scheme 1). Catalyst
8 was prepared by nucleophilic displacement of 2-chloromethyl-1-
methylimidazole hydrochloride, followed by reaction with excess
Grignard reagent.
Trichlorosilane is a cheap and versatile reducing agent that has
been used in applications as diverse as the reduction of chiral phos-
phine oxides to phosphines with conservation of stereochemical
information, the hydrometallation of alkynes and the synthesis of
disulfide compounds.1 Kobayashi demonstrated that it could also
function as a reducing agent for aldehydes, ketones and imines if
a catalytic amount of DMF was present.2 This spurred an effort
to build suitable chiral catalysts which were often derived from
natural amino acids or compounds from the chiral pool. These
were, from a structural point of view, chiral DMF analogues.3
Even though the formamide paradigm is versatile, asymmet-
ric imine reductions using trichlorosilane have evolved further.
Matsumura et al., for instance, used diphenylprolinol instead of
proline as a core scaffold and replaced the formyl moiety by a
picolinoyl fragment to give catalyst 1 (Fig. 1).4 Kocˇovsky´ et al.
have progressed this by preparing several catalysts containing an
oxazoline attached to a quinoline 2 or pyridine 3, the latter being
particularly versatile and can be used to reduce imines and ketones
with equal ease.5
Scheme 1 Preparation of catalysts.
These five catalysts were then assessed in the reduction of
acetophenone imine 9 to N-phenyl-1-phenylethanamine 10 using
trichlorosilane as the reductant. Catalysts 4–7 all accelerated the
rate of the reaction, with catalyst 8 showing essentially no catalytic
activity and selectivity, stressing the importance of the role of
the amide functionality in this reaction (Scheme 2, Table 1). The
oxazoline 4 only gave modest enantioselectivity, in contrast to the
gem-diphenyl and gem-b-dinaphthyl prolinol derivatives 5 and 7
which gave the best all round results. Removal of the geminal
groups led to a substantial drop in ee (catalyst 6, entry 3).
Fig. 1 Examples of chiral catalysts previously used for the asymmetric
hydrosilyation of ketones and ketimines.
Our group has a long standing interest in bifunctional catalysts
featuring an imidazole ring,6 and there is a striking similarity
between Kocˇovsky´’s oxazoline 3 and catalyst 4 (Scheme 1), which
had been previously prepared in our group but never assessed in
this reaction.7 This prompted us to investigate the use of this and
related compounds 5–8 in this type of transformation.
aDepartment of Chemistry, University of Sheffield, Dainton Building, Brook
Hill, Sheffield, UK S3 7HF. E-mail: simon.jones@sheffield.ac.uk; Fax: +44
(0)114 2229346; Tel: +44 (0)114 2229483
bAesica Pharmaceuticals Ltd., Windmill Industrial Estate, Shotton Lane,
Cramlington, Northumberland, UK NE23 3JL
† Electronic supplementary information (ESI) available: Experimental. See
DOI: 10.1039/b816051a
Scheme 2
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 229–231 | 229
©

View Article Online
Table 1 Asymmetric hydrosilylation of N-phenyl acetophenone imine 9
CHCl3 gave a marginal improvement in ee (entry 11) while toluene
led to a reduced yield of product (entry 12).
according to Scheme 2a
The applicability of the catalyst was next evaluated with
a selection of diverse ketimines (Fig. 2, Table 3). Significant
improvement in yield was observed when the parent substrate
9 was changed to the N-p-methoxyphenyl analogue 12 (entries 1
and 3), presumably since this more electron rich ketimine can
undergo more efficient binding to the Cl3SiH. This is also
highlighted with strongly electron withdrawing aryl groups giving
poor yields (entry 4). However, in this acetophenone series,
essentially no improvement in selectivity was noticeable. A similar
trend is also observed for propiophenone derived substrates 16 and
17 (entries 7 and 8), although a somewhat surprising low yield and
selectivity is obtained with tetralone derivatives (entries 9 and 10).
The ratio of the ketimine geometric isomer did not seem to have
great influence on the outcome of the reaction, most remarkably
demonstrated with the a-naphthyl derivative 21 (entry 12). This
latter observation is of particular interest, since the substrates need
not be geometrically pure, which is of significant advantage when
one considers the difficulty in preparing and handling the sensitive
ketimine substrates, let alone purifying them to a single geometric
isomer. This is in stark contrast to oxazaborolidine catalysed
borane reductions of oximes where stereospecific reduction of
Entry
Catalyst
Yield (%)b
ee (%)c/configurationd
1e
2
3
4
5
4
5
6
7
8
80
77
84
68
20
23 (R)
86 (S)
47 (S)
87 (S)
3 (S)
a The reaction was performed by addition of trichlorosilane (2.0 mmol)
to a stirred solution of ketimine 9 (1.0 mmol) and ligand (0.01 mmol)
in dry CH2Cl2 (1 mL) under an atmosphere of nitrogen at 0 ◦C. After
4 h, the reaction was quenched with 1 M HCl (1 mL) and subjected to
standard work-up procedures. b Based on isolated product. c Determined
by integration of the appropriate signals in the HPLC chromatogram
of the crude reaction mixture. d Confirmed by comparison of HPLC
retention times and specific rotations with those in the literature. e Reaction
conducted with 10 mol% catalyst.
Catalyst 5 was taken forward for further optimization (Table 2,
Scheme 3). The first and most remarkable discovery with catalyst
5 was the low loading that could be tolerated in the reaction
without detrimental effects on enantioselectivity (entries 1–3).
This compares well with other benchmark catalysts for N-aryl
substrates; catalyst 1 provides 67–90% yields in 71–80% ee at
20 mol%, while catalyst 2 provides 51–67% yields in 86–87% ee
at 20 mol%. At 1 mol% loading, temperature had little effect on
the outcome (entries 3–5), although the initial concentration of
the imine and the equivalents of Cl3SiH used did (entries 6–9),
and this could be used to offset the drop in yield when using
low catalyst loadings, while additionally helping to reduce reac-
tion times. Attempts to lower the catalyst loading of the optimized
reaction conditions (entry 9) further led to erosion in both yield
and selectivity of the reaction (entry 10). Switching the solvent to
Scheme 3
Fig. 2 Substrates used to evaluate the applicability of catalyst 5.
Table 2 Optimisation of the asymmetric hydrosilylation of N-phenyl acetophenone imine 9 using catalyst 5a
Entry
Time/h
T/◦C
Cl3SiH (eq)
Catalyst 5 (eq)
[imine]0/mol L-1
Yield (%)b
ee (%)c
1
2
3
4
5
6
7
8
9
13
13
13
13
13
13
13
13
4
0
0
0
25
-20
15
15
15
0
1.5
1.5
1.5
1.5
1.5
1.5
1.5
4
2
2
2
2
0.1
0.4
0.4
0.4
0.4
0.4
2.0
0.2
0.4
2.0
2.0
2.0
2.0
59
65
45
39
42
56
32
76
82
57
81
76
87
86
85
83
88
83
84
84
85
77
86
86
0.05
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.001
0.01
0.01
10
11d
12e
4
4
4
0
0
0
a The reaction was performed by addition of trichlorosilane to a stirred solution of ketimine 9 and catalyst in dry CH2Cl2 under an atmosphere of nitrogen.
After the required time, the reaction was quenched with 1 M HCl (2 mL) and subjected to standard work-up procedures. b Based on isolated product.
c Determined by integration of the appropriate signals in the HPLC chromatogram of the crude reaction mixture. In all cases the (S) enantiomer was
formed as the major product. d Reaction performed using CHCl3 as solvent. e Reaction performed using toluene as solvent.
230 | Org. Biomol. Chem., 2009, 7, 229–231
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
Table 3 Evaluation of substrate specificity and reactivity in the hydrosily-
ation catalysed by catalyst 5a
In conclusion, we have developed a catalyst for the asymmetric
reduction of a wide variety of ketimines at low catalyst loadings,
the selectivity of which appears to be independent of the geometric
purity of the substrate. Further studies which capitalize upon this
reactivity and selectivity are underway.
Entry
Ketimine
Ketimine ratiob
Yield (%)c
ee (%)d
1
2
3
4
5
6
7
8
9
10
11
12
13
9
100 : 0
100 : 0
100 : 0
100 : 0
100 : 0
100 : 0
91 : 9
88 : 12
100 : 0
100 : 0
80 : 20
60 : 40
100 : 0
82
77
96
72
81
85
59
95
42
41
85
71
86
85
85
87
82
85
86
79
83
19
22
73
74
86
11
12
13
14
15
16
17
18
19
20
21
22
We thank Aesica Pharmaceuticals Ltd. and The University of
Sheffield for funding (FMG).
Notes and references
1 W. P. Weber, Silicon Reagents for Organic Synthesis, Springer-Verlag,
Berlin, 1983, pp. 31, 168, 325, 333, 335.
2 (a) R. A. Benkeser and D. C. Snyder, J. Organomet. Chem., 1982, 225,
107; (b) S. Kobayashi, M. Yasuda and I. Hachiya, Chem. Lett., 1996,
407.
3 (a) F. Iwasaki, O. Onomura, K. Mishima, T. Maki and Y. Matsumura,
Tetrahedron Lett., 1999, 40, 7507; (b) F. Iwasaki, O. Onomura, K.
Mishima, T. Kanematsu, T. Maki and Y. Matsumura, Tetrahedron Lett.,
2001, 42, 2525.
4 O. Onomura, Y. Kouchi, F. Iwasaki and Y. Matsumura, Tetrahedron
Lett., 2006, 47, 3751.
a The reaction was performed using optimized conditions as described in
1
Table 2, entry 9. b Relative ratio of E/Z isomers determined by H nmr
spectroscopy. c Based on isolated product. d Determined by integration of
the appropriate signals in the HPLC chromatogram of the crude reaction
mixture.
5 (a) A. V. Malkov, S. Stoncˇius, K. N. MacDougall, A. Mariani,
G. D. McGeoch and P. Kocˇovsky´, Tetrahedron, 2006, 62, 264;
(b) A. V. Malkov, A. J. P. S. Liddon, P. Ram´ırez-Lo´pez, L. Bendova´,
D. Haigh and P. Kocˇovsky´, Angew. Chem., Int. Ed., 2006, 45,
1432.
6 S. Jones, J. Northen and A. Rolfe, Chem. Commun., 2005,
3832.
7 S. Jones and H. C. Norton, Synlett, 2004, 338.
8 Y. Sakito, Y. Yoneyoshi and G. Suzukamo, Tetrahedron Lett., 1988, 29,
223.
9 See C. Wang, X. Wu, L. Zhou and J. Sun, Chem.–Eur. J., 2008, 14,
8789, and references therein.
each geometric isomer is observed.8 Similar results have been
remarked upon previously, but insufficient experimental data
has been accrued to facilitate interpretation of this result.9 For
ketimines 9, 12–15, 20, and 22, the absolute stereochemistry of
the product was confirmed to be (S) by comparison of the specific
rotation with compounds of known configuration.10 However, in
the case of the other amine products, no unambiguous report
of the absolute configuration have been reported and these are
therefore assumed to be (S) by analogy.
10 See supporting information for details.
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 229–231 | 231
©