Effective reductive amination of carbonyl compounds with hydrogen catalyzed by iridium complex in organic solvent and in ionic liquid

Article abstract of DOI:10.1016/j.tet.2005.05.024

The direct reductive amination (DRA) of carbonyl compounds with amines has been achieved using homogenous iridium catalyst and gaseous hydrogen. It appeared that the cationic iridium catalyst, [Ir(cod)₂]BF ₄, without any other ligands was sufficient for the reaction. For the DRA of the ketone substrates, an ionic liquid, [Bmim]BF₄, was found to be superior to the other organic solvent used. Especially, the counter anion of the ionic liquid has a significant influence on the selectivity, and at the same time, a high reaction temperature was found to be crucial for the excellent selectivity.

Full text of DOI:10.1016/j.tet.2005.05.024

Tetrahedron 61 (2005) 6988–6992
Effective reductive amination of carbonyl compounds with
hydrogen catalyzed by iridium complex in organic
solvent and in ionic liquid
*
Daisuke Imao, Shoichiro Fujihara, Takeshi Yamamoto, Tetsuo Ohta and Yoshihiko Ito
Department of Molecular Science and Technology, Faculty of Engineering, Doshisha University, Kyotanabe, Kyoto 610-0394, Japan
Received 30 November 2004; revised 11 May 2005; accepted 11 May 2005
Available online 6 June 2005
Abstract—The direct reductive amination (DRA) of carbonyl compounds with amines has been achieved using homogenous iridium catalyst
and gaseous hydrogen. It appeared that the cationic iridium catalyst, [Ir(cod)2]BF4, without any other ligands was sufficient for the reaction.
For the DRA of the ketone substrates, an ionic liquid, [Bmim]BF4, was found to be superior to the other organic solvent used. Especially, the
counter anion of the ionic liquid has a significant influence on the selectivity, and at the same time, a high reaction temperature was found to
be crucial for the excellent selectivity.
q 2005 Elsevier Ltd. All rights reserved.
1. Introduction
reductive amination.7 However, this system requires the
addition of a catalytic amount of acid and phosphine ligand,
and complete deoxygenation of the reaction solution in
order to prevent oxidation of the phosphine ligand. In this
paper, we introduced another possibility for the highly
selective direct reductive amination of carbonyl compounds
with gaseous hydrogen catalyzed by an iridium complex
without any additives or any acid and even without any
effort to complete the deoxygenation.
The reductive amination of carbonyl compounds is
attractive in organic synthesis because ketones and
aldehydes can be transformed, in one vessel, directly to
the corresponding secondary or primary alkylamines with-
out isolation of the intermediary imines or hydroxy amines.1
Sodium cyanoborohydride has been used to carry out this
reaction, but the reagent needs to be stoichiometrically
used.2 To utilize gaseous hydrogen as a reducing reagent,
palladium carbon (Pd/C) is usually used, but a hetero-
geneous catalyst is difficult to modify in order to control its
reactivity.3 Many enantioselective reductions of an isolated
imine with a homogeneous catalyst have been reported.
Those catalysts are effective though the imine is very
sensitive to hydrolysis that sometimes the difficulty of
completing the condensation to obtain the imine has been
considered as a big problem for reduction.4 To overcome
this problem, some groups have recently proposed some
ideas to realize the direct reductive amination with a
homogeneous iridium catalyst and gaseous hydrogen.5
Zhang et al. achieved the reaction by means of adding
Ti(OiPr)4 and I2 to the reaction system to promote the
condensation reaction and almost quantitatively produced a
secondary amine.6 Chaloner et al. have also proclaimed that
the cationic iridium–phosphine system is effective for
2. Results and discussion
First of all, we examined several hydrogenation catalysts for
the direct reductive amination (DRA). To achieve an
efficient catalytic DRA method, we paid close attention
not only to its reactivity but also to its selectivity, because,
under the DRA reaction condition, ketone 1 can also be
Keywords: Direct reductive amination; Iridium; Ionic liquid; Secondary
amine.
*
Corresponding author. Tel.: C81 774 65 6548; fax: C81 774 65 6789;
e-mail: tota@mail.doshisha.ac.jp
Scheme 1.
0040–4020/$ - see front matter q 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2005.05.024

D. Imao et al. / Tetrahedron 61 (2005) 6988–6992
Table 1. DRA of benzaldehyde (6) with benzylamine (7)a
6989
Entry
Cat.b
Time (h)
Yield of 8 (%)c
Yield of 9 (%)c
Yield of 10 (%)c
1
2
3
4
A
A
B
B
0.5
3
0.5
3
51
0
68
0
45
93
31
94
4
7
1
6
a H2 pressure (50 kg/m2) and EtOH as solvent were used at room temperature.
b A, [IrCl(cod)]2; B, [Ir(cod)2]BF4.
c Yields were determined by 1H NMR.
reduced to an alcohol 5, which is recognized as a by-product
(Scheme 1). Using ruthenium and rhodium complexes, the
alcohol 5 was produced as the main product, while the
iridium complexes showed reverse and desirable selectivi-
ties. Representative results obtained from the reaction of
benzaldehyde (6) with benzylamine (7) using [IrCl(cod)]2
(A) and [Ir(cod)2]BF4 (B) are listed in Table 1. The DRA
was performed using 2 mol% of a catalyst (based on iridium
atom) in EtOH at rt under a H2 atmosphere (50 kg/cm3)
without any purging procedure. Both catalysts did work
with no problem, but [Ir(cod)2]BF4 showed a slightly lower
catalytic activity, but seemed more effective in terms of the
DRA selectivity.
complete even within 0.5 h with excellent selectivity. In
contrast, the reaction with the 1- and 2-naphthaldehydes
produced considerable amounts of alcohols. On the other
hand, no difference was observed in terms of the position of
the substitutent attached to the phenyl ring of the substrate.
Now that an effective DRA with aldehydes was achieved,
the reaction with ketones should be next investigated.
Acetophenone (14) was chosen as the ketone, and the
reaction became sluggish. In order to obtain a high
conversion and/or yield, heating and prolonging the reaction
time were necessary. Again, the solvent effect was studied
(Table 4). As shown in the Table, a higher temperature gave
a better conversion, but the selectivity was not different.
[IrCl(cod)]2 was tested instead of [Ir(cod)2]BF4 in toluene
(entry 5), and the result showed a clearer difference in
selectivity.
The solvent effect is summarized in Table 2. To observe
their differences in detail, the reactions were quenched in
0.5 h. These results indicated no crucial difference except
the one with CH2Cl2. As CH2Cl2 seems not to have
significant effect for the deteriorating formation of an imine,
CH2Cl2 may have some influence on the catalytic cycle, but
the mechanism is still unclear.
Various ketones were subjected to this RDA reaction
(Table 5). The results appeared to be good to excellent
yields and selectivities with some exceptions. First, in
contrast to the reductive amination of aldehydes, the
reaction with aniline gave low alcohol yields and no
secondary amines (entries 2 and 5), in which considerable
amount of unidentifiable by-products were formed. This
may be caused by the lower nucleophilicity of aniline
compared to benzylamine. No distinguishable effect was
observed with the difference in the substituent position
attached to the aromatic ring of the substrate ketone.
Interestingly, 2-acetonaphthone gave the worst selectivity,
Now that the optimized condition has been determined,
other aldehydes were employed for this reaction to create a
variety of secondary amines (Table 3). All these results
showed good to excellent yields and selectivities in 3 h.
Aniline was also a good amine substrate instead of
benzylamine (entry 2 and 4), and there seemed to be no
difference in reactivity. When propanal was used with
benzylamine or aniline (entries 3 and 4), the reaction was
Table 2. Solvent effect on DRA of 6 with 7a
Entry
Solvent
Conv. of 6 (%)b
Yield of 8 (%)b
Yield of 9 (%)b
Yield of 10 (%)b
1
2
3
4
5
6
MeOH
EtOH
Benzene
Toluene
CH2Cl2
CH3CN
100
100
100
100
93
58
51
46
47
79
46
40
45
50
48
13
40
2
4
4
5
1
4
100
a H2 pressure (50 kg/m2) and [Ir(cod)2]BF4 as catalyst were used at room temperature, 0.5 h.
b Yields were determined by 1H NMR using internal standard (bibenzyl) method.

6990
D. Imao et al. / Tetrahedron 61 (2005) 6988–6992
Table 3. DRA for a range of aldehydes with aminesa
Entry
Sub.
Yield of product (%)b
R1
R2
Conv. (%)
11
12
13
1
2
3
4
5
6
7
8
9
10
Ph
Ph
Bn
Ph
Bn
Ph
Bn
Bn
Bn
Bn
Bn
Bn
100
100
100
100
100
100
100
100
100
94
0
0
0
0
0
0
0
2
0
11
94
91
O99
O99
83
85
94
90
92
6
9
0
Etc
Etc
0
1-Naph
2-Naph
o-MeOPh
m-MeOPh
p-MeOPh
p-ClPh
17
15
6
8
8
79
4
a H2 pressure (50 kg/m2) and [Ir(cod)2]BF4 as catalyst and benzene as solvent were used at room temperature, 3 h.
b Yields were determined by 1H NMR using internal standard (bibenzyl) method.
c Reaction was completed in 0.5 h.
while 1-acetonaphthone was converted to the secondary
amine with a similar selectivity for the reaction of the
acetophenone derivatives.
characteristics8 but for its unique effect on catalytic
cycles.9,10 The reaction of acetophenone with benzylamine
was carried out in commonly used several kind of ILs
(Fig. 1, Table 6).
The DRA of an aldehyde with an amine proceeded with a
good selectivity better than 10 in the ratio of secondary
amine 12/by-product alcohol 13, except for 1- and
2-naphthalenecarbaldehyde, while that of ketone with the
amine was less than 10 in the ratio of 19/20. For achieving
better selectivity, an ionic liquid was investigated for this
DRA of ketones with amine. The ionic liquid has recently
been used as a solvent for many catalyses and has shed light
on not only due to its environmentally friendly
In the ionic liquid, [Bmim]Cl or [Emim]Cl, in which the
chloride is a counter anion (entry 1 and 2), the reaction
selectivity was significantly affected and was worse than the
one in toluene. In contrast to these results, the selectivity in
K
IL, [Bmim]BF4, in which BF4 is the counter anion, was
the best of all results obtained from the DRA of ketones in
this study. The reaction with [IrCl(cod)]2 instead of
[Ir(cod)2]BF4 in [Bmim]BF4 was carried out and gave a
Table 4. Solvent effect on DRA of acetophenone (14) with benzylamine (7)a
Entry
Solvent
Temp. (8C)
Conv. of 14 (%)b
Yield of 15 (%)b
Yield of 16 (%)b
Yield of 17 (%)b
1
2
3
4
5c
6
7
MeOH
EtOH
60
70
70
100
100
rt
97
3
0
0
0
0
84
88
87
89
78
8
10
12
13
11
22
4
100
100
100
100
60
Benzene
Toluene
Toluene
CH2Cl2
CH3CN
48
0
70
100
88
12
a H2 pressure (50 kg/m2) and [Ir(cod)2]BF4 as catalyst were used for 24 h.
b Yields were determined by 1H NMR using internal standard (bibenzyl) method.
c [IrCl(cod)]2 was used instead of [Ir(cod)2]BF4.

D. Imao et al. / Tetrahedron 61 (2005) 6988–6992
6991
Table 5. DRA of a range of ketones with aminesa
Entry
Sub.
Yield of product (%)b
R1
R2
Conv. (%)
18
19
20
1
2
3
4
5
6
7
8
Ph
Ph
Bn
Ph
Bn
Bn
Ph
Bn
Bn
Bn
Bn
Bn
Bn
100
100
100
100
100
100
100
100
100
100
100
0
29
0
0
0
0
0
0
2
89
0
86
91
0
87
59
89
91
92
89
11
15
14
2
12
13
41
11
9
Phc
Et
Et
1-Naph
2-Naph
o-MeOPh
m-MeOPh
p-MeOPh
p-ClPh
9
10
11
0
0
8
11
a H2 pressure (50 kg/m2) and [Ir(cod)2]BF4 as catalyst and toluene as solvent were adopted at 100 8C, 24 h.
b Yields were determined by 1H NMR using internal standard (bibenzyl) method.
c Propiophenone was used.
alcohol 17) increased 54 times from 20 to 100 8C. The
reaction selectivity also became better as the H2 pressure
increased (16/17: 7 times increased from 1 to 40 kg/cm2).
These phenomena have not yet been explained, but a high
temperature (100 8C) and a high H2 pressure (40 kg/cm2)
were the choices for obtaining an excellent selectivity.
Figure 1. Ionic liquid used in this study.
3. Conclusion
selectivity comparable to that of [Ir(cod)2]BF4. In toluene,
[IrCl(cod)]2 was less effective than [Ir(cod)2]BF4 (Table 4,
entry 5) in selectivity. This would be attributed to the effect
of the anion exchange on the rhodium catalyst with ClK and
BF4K, and the cationic iridium catalyst was assumed be
generated in situ.
In summary, we have found that [Ir(cod)2]BF4 is sufficiently
effective for the direct reductive amination of carbonyl
compounds with amines without any deoxygenation pro-
cedure. Yields up to 99% with an aldehyde, and up to 92%
with a ketone were obtained in benzene and toluene,
respectively. In [Bmim]BF4, acetophenone was effectively
converted to the corresponding secondary amine in up to
97% yield with excellent selectivity. To the best of our
knowledge, this would be the first example of the direct
reductive amination with a homogeneous catalyst was
carried out in an ionic liquid. It appeared that the reaction
conditions, especially temperature, played a critical role in
exclusively obtaining secondary amines.
The combination of [Bmim]BF4 as the solvent and
[Ir(cod)2]BF4 as the catalyst gave the best selectivity for
the DRA of ketone. The effect of temperature and hydrogen
pressure on the DRA were investigated for determining
milder reaction conditions (Table 7).
Surprisingly, the higher temperature gave not only a better
yield, but also a better selectivity. The ratio (product 16/
Table 6. Ionic liquid effect on DRAa
Entry
Solvent
Cat.b
Conv. of 14 (%)c
Yield of 15 (%)c
Yield of 16 (%)c
Yield of 17 (%)c
1
2
3
4
[Bmim]Cl
[Emim]Cl
[Bmim]BF4
[Bmim]BF4
B
B
B
A
93
92
98
98
8
0
0
2
57
69
97
95
28
17
1
1
a H2 pressure (40 kg/m2) was adopted at 100 8C, 24 h.
b A, [IrCl(cod)]2; B, [Ir(cod)2]BF4.
c Yields were determined by 1H NMR using internal standard (bibenzyl) method.

6992
D. Imao et al. / Tetrahedron 61 (2005) 6988–6992
a
Table 7. Temperature and pressure effect on DRA in [Bmim]BF4
Entry
Temp.
p(H2) (kg/cm2)
Conv. of 14
(%)b
Yield of 15 (%)b Yield of 16 (%)b Yield of 17 (%)b
Ratio of 16/17
1
2
3
4
5
6
7
8
9
20
50
80
100
100
100
100
100
100
40
40
40
40
1
56
77
99
98
65
88
91
93
97
39
11
0
0
22
8
6
1
—
11
54
86
97
40
75
82
88
95
6
11
13
1
3
5
3
4
2
1.8
4.9
6.6
97
13
15
27
22
48
5
10
20
30
a [Ir(cod)2]BF4 as catalyst was adopted for 24 h.
b Yields were determined by 1H NMR using internal standard (bibenzyl) method.
4. Experimental
References and notes
The 1H NMR spectra were measured using a Varian
MARCURYplus300-4N (300 MHz) spectrometer with
tetramethylsilane as the internal standard. The solvents
were purified by conventional methods and stored under an
1. (a) Trost, B. M.; Verhoeven, T. R. In Willkinson, G., Stone,
F. G. A., Abel, E. W., Eds.; Comprehensive Organometallic
Chemistry; Pergamon: Oxford, 1982; Vol. 8, p 84. (b) March,
J.; Advanced Organic Chemistry; Wiley: New York, 1992;
pp 889–890. (c) Hutchins, R. O.; Hutchins, M. K. In Trost,
B. M., Fleming, I., Eds.; Comprehensive Organic Synthesis;
Pergamon: Oxford, 1991; Vol. 8; p 25.
11
12
argon atmosphere. [IrCl(cod)]2 and [Ir(cod)]BF4 were
prepared according to the literature. All other materials were
purchased and used without further purification. The
1
reaction products were identified by H NMR analysis and
compared to authentic commercial products.
2. (a) Lane, C. F. Synthesis 1975, 135. (b) Borch, R. F.;
Bernstein, M. D.; Durst, H. D. J. Am. Chem. Soc. 1971, 93,
2897.
4.1. General procedure for direct reductive amination
3. Heinen, A. W.; Peters, J. A.; van Bekkum, H. Eur. J. Org.
Chem. 2000, 2501.
The catalyst (2.0!10K2 mmol), carbonyl compound
(1.0 mmol), amine (1.0 mmol), solvent (0.2 mL) and a
magnetic stir bar were added to a stainless steel autoclave
under hydrogen atmosphere at the pressure and temperature
cited in each table. After the reaction, bibenzyl as an internal
standard was added, and the mixture was directly analyzed
by 1H NMR. For isolation of the secondary amine, the
reaction mixture was poured into a mixture of ether and 2 N
HCl aq. The water layer was washed with ether several
times, and became basic with 2 N NaOH aq. Addition. The
ether extract of the water layer was concentrated and then
isolated by column chromatography (Alumina). When an
ionic liquid was used as the solvent, the reaction mixture
was extracted with hexane until no organic materials were
detected in hexane layer. The hexane layer was concen-
trated, and then the similar procedure described above was
performed.
4. Blaser, H.-U.; Spindler, F. In Jacobsen, E. N., Pfaltz, A.,
Yamamoto, H., Eds.; Comprehensive Asymmetric Catalysis;
Springer: Berlin, 1999; Vol. 1, pp 247–265.
5. (a) Tararov, V. I.; Kadyrov, R.; Riermeier, T. H.; Dingerdis-
¨
sen, U.; Borner, A. Org. Prep. Proc. Int. 2004, 36, 99. (b)
¨
Tararov, V. I.; Kadyrov, R.; Riermeier, T. H.; Borner, A. Adv.
Synth. Catal. 2002, 344, 200. (c) Gross, T.; Seayad, A. M.;
Ahmad, M.; Beller, M. Org. Lett. 2002, 4, 2055.
6. Chi, Y. W.; Zhou, Y.-G.; Zhang, X. J. Org. Chem. 2003, 68,
4120.
7. Chaloner, P. A.; Collard, S.; Ellis, R. D.; Keep, A. K. EP
1078915 A1, 2001.
8. Wasserscheid, P.; Keim, W. Angew. Chem., Int. Ed. 2000, 39,
3772.
9. Dupont, J.; de Souza, R. F.; Suarez, P. A. Z. Chem. Rev. 2002,
102, 3667.
10. Welton, T. Chem. Rev. 1999, 99, 2071.
11. Herde, J. L.; Lambert, J. C.; Senoff, C. V. Inorg. Synth. 1974,
15, 18.
Acknowledgements
12. Shenck, T. G.; Downs, J. M.; Milne, C. R. C.; Mackenie, P. B.;
Boucher, H.; Whealan, J.; Bosnich, B. Inorg. Synth. 1985, 24,
2334.
This work was partially supported by a grant to RCAST at
Doshisha University from the Ministry of Education, Japan.