Ruthenium-catalyzed direct reductive amination in HCOOH/NEt3 mixture

Article abstract of DOI:10.1007/s10562-014-1314-5

A one-pot direct reductive amination of aldehydes with primary and secondary amines using Noyori's catalyst (η ⁶-arene)Ru(H)TsDPEN in neat HCOOH/NEt₃ mixture has been developed. The catalyst formed in situ allows full conversion of aldehydes with high selectivities towards the desired amine products within hours or even minutes at room temperature. For aromatic aldehydes, both 5:2 and 5:3 HCOOH/NEt₃ mixtures prove to be suitable reaction media, with the former affording higher selectivities while the latter leading to much faster reaction rates. For aliphatic aldehydes, the 5:3 HCOOH/NEt₃ mixture gives much higher selectivities due to the minimal loss from aldol condensation. This method has great application potential for the synthesis of amine products given the mild conditions. Graphical Abstract: [Figure not available: see fulltext.]

Full text of DOI:10.1007/s10562-014-1314-5

Catal Lett (2014) 144:1568–1572
DOI 10.1007/s10562-014-1314-5
Ruthenium-Catalyzed Direct Reductive Amination in HCOOH/
NEt3 Mixture
Mengping Zhu
Received: 7 May 2014 / Accepted: 6 July 2014 / Published online: 25 July 2014
Ó Springer Science+Business Media New York 2014
Abstract A one-pot direct reductive amination of alde-
hydes with primary and secondary amines using Noyori’s
catalyst (g6-arene)Ru(H)TsDPEN in neat HCOOH/NEt3
mixture has been developed. The catalyst formed in situ
allows full conversion of aldehydes with high selectivities
towards the desired amine products within hours or even
minutes at room temperature. For aromatic aldehydes, both
5:2 and 5:3 HCOOH/NEt3 mixtures prove to be suitable
reaction media, with the former affording higher selectiv-
ities while the latter leading to much faster reaction rates.
For aliphatic aldehydes, the 5:3 HCOOH/NEt3 mixture
gives much higher selectivities due to the minimal loss
from aldol condensation. This method has great application
potential for the synthesis of amine products given the mild
conditions.
agrochemical industries. In 1996, Noyori et al. reported
imine reduction [13] by the Ru(II) catalyst in 5:2 formic
acid/triethylamine azeotrope with various aprotic solvents,
such as acetonitrile, DMF, acetone, DMSO and dichloro-
methane. This reaction is particularly useful for asym-
metric reduction of cyclic imines with high efficiency and
enantioselectivity, towards the synthesis of natural and
unnatural isoquinoline alkaloids. In 2003, the Wills group
[14] discovered a one-pot process for the enantioselective
synthesis of cyclic amines via a sequence of deprotection
followed by intramolecular reductive amination of t-Boc-
protected amino ketones under asymmetric transfer
hydrogenation conditions.
So far, Noyori’s catalyst has been mainly applied to the
reduction of cyclic imines [13–16] while its application in
acyclic imine reduction [13, 17] has been less developed,
especially direct reductive amination. Direct reductive
amination is a challenging reaction to be achieved and has
received great attention [18–21]. The reaction scheme is
depicted in Scheme 1 and the challenges of this reaction
can be summarized as follows: firstly, imine formation step
from aldehydes/ketones with amines is a reversible reac-
tion which is susceptible to water, pH and temperature.
Thus minimal decomposition of imine is a prerequisite for
a successful direct reductive amination. Secondly, the
aldehydes/ketones are also subject to the hydrogenation
under the catalytic conditions. To achieve a high selectivity
of the desired amine product, the reduction of imine must
be significantly faster than the reduction of aldehydes/
ketones. Thirdly, N-formylation of the substrate amine can
take place as a side reaction when HCOOH is used as the
hydrogen source [22], especially at high temperature.
This communication introduces a mild, highly efficient,
one-pot direct reductive amination by using Noyori’s cat-
alyst in HCOOH/NEt3 mixture. This direct reductive
Keywords Direct reductive amination ꢀ Noyori’s
catalyst ꢀ Transfer hydrogenation ꢀ Formic acid/
triethylamine
1 Introduction
Transfer hydrogenation using Noyori’s catalyst (g6-
arene)Ru(H)TsDPEN (TsDPEN = (1S,2S)- or (1R,2R)-N-
tosyl-1,2-diphenylethylenediamine) has achieved great
success in the reduction of various substrates [1–8]. Its
application in the asymmetric reductions of imines makes
great contribution to the preparation of chiral amines [9–
12], which are important compounds in pharmaceutical and
M. Zhu (&)
Department of Chemistry, Georgetown University, Washington,
DC 20057, USA
e-mail: mengpingzhu@gmail.com; mz63@georgetown.edu
123

Amination in HCOOH/NEt3 Mixture
1569
catalyst formed in situ (entry 3), N-benzylideneaniline was
almost completely transformed to the N-benzylaniline with
only 8 % selectivity loss to benzyl alcohol after 24 h. From
the one hour reaction (entry 4), it was found that 51 % of
N-benzylideneaniline was converted to N-benzylaniline
(36 %) and benzaldehyde (64 %), and no benzyl alcohol
was detected. This suggests that reduction of benzaldehyde
under these conditions is much slower than the reduction of
N-benzylideneaniline. It also indicates that reduction of N-
benzylideneaniline is thermodynamically controlled and
decomposition of imine is reversible.
Scheme 1 Reaction scheme of direct reductive amination in
HCOOH/NEt3 mixture
Benzaldehyde, acetophenone and benzophenone (entry
5–7) reacted with aniline to study the substrate scope of
direct reductive amination. Delightedly, Benzaldehyde was
converted to N-benzylaniline with 90 % selectivity which
was only slightly lower than imine reduction using N-
benzylideneaniline as the substrate. Compared to benzal-
dehyde, both acetophenone and benzophenone failed to
react with aniline due to their lower reactivities. It was
found that 26 % acetophenone was converted to 1-phen-
ylethanol with no reductive amination product detected,
while benzophenone was even more inert since neither
alcohol product nor amine product was observed. This test
suggests direct reductive amination using Noyori’s catalyst
in the 5:2 HCOOH/NEt3 azeotrope might be limited to
aldehydes under these conditions.
amination applies to aromatic and aliphatic aldehydes,
primary and secondary amines. The pH was found to be a
key factor affecting the efficiency and selectivity of the
direct reductive amination. A cheaper ligand TsEN (N-
tosylethylenediamine) was also tested and proved to be
slightly inferior to TsDPEN. These findings have expanded
the application scope of Noyori’s catalyst.
2 Results and Discussion
Further study on the direct reduction amination between
benzaldehyde and aniline showed that the reaction can be
completed after 6 h, with 94 % selectivity towards the
desired product (entry 8). This suggests that longer reaction
time could deteriorate the selectivity, which might be due
to the possibility of the dehydrogenation of the amine
product, driving the reaction equilibrium shifted towards
the formation of benzyl alcohol. Same reaction was also
carried out at 60 °C to study the temperature effect (entry
9). The result shows that a higher temperature can increase
Preliminary study was carried out to study the feasibility
and reaction conditions of the imine reduction and direct
reductive amination, shown in Table 1. N-benzylideneani-
line was used as the substrate for imine reduction in 5:2
HCOOH/NEt3 azeotrope. Surprisingly, N-benzylideneani-
line was found reduced to N-benzylaniline with 63 %
selectivity in the absence of Ru(II) catalyst (entry 1). In the
presence of [RuCl2(p-cymene)]2 (entry 2), both conversion
and selectivity were slightly enhanced but the reaction still
could not reach complete conversion. With Noyori’s
Table 1 Preliminary study of imine reduction in 5:2 HCOOH/NEt3 azeotrope
Entry
Substrate
R
–
Catalyst
Time (h)
Temperature (°C)
Conversion (%)
Selectivity (%)
1
2
3
4
5
6
7
8
9
–
24
24
24
1
25
25
25
25
25
25
25
25
60
83
87
63a
76a
92
36a
90
0
[RuCl2(p-cymene)]2
[RuCl2(p-cymene)]2/TsDPEN
[RuCl2(p-cymene)]2/TsDPEN
[RuCl2(p-cymene)]2/TsDPEN
[RuCl2(p-cymene)]2/TsDPEN
[RuCl2(p-cymene)]2/TsDPEN
[RuCl2(p-cymene)]2/TsDPEN
[RuCl2(p-cymene)]2/TsDPEN
100
51
H
24
24
24
6
100
26
Me
Ph
H
0
0
100
100
94
87
H
1
Conditions: substrate (1 mmol), catalyst (0.01 mmol), 5:2 HCOOH/NEt3 azeotrope (1.0 mL), conversions and selectivities were determined by
1H NMR spectroscopy
a
The byproduct is benzaldehyde
123

1570
M. Zhu
the reaction rate dramatically but slightly decrease the
selectivity. This indicates that increasing temperature can
accelerate the hydrogenation of both imine and aldehyde,
with the latter being more influenced.
other hand, the amine scope was much wider including ten
para-substituted aniline derivatives and six aliphatic
amines. A cheaper ligand TsEN (N-tosylethylenediamine)
was also applied. Overall, aldehydes in all the catalytic sets
were completely converted to the desired amine products
as well as alcohol side products unless stated otherwise. As
expected, compared to the reactions in the 5:2 HCOOH/
NEt3 azeotrope, reactions involving benzaldehyde in the
5:3 HCOOH/NEt3 mixture were much faster, however, all
the selectivities towards to the desired products reduced to
different extents. TsDPEN is superior to TsEN in terms of
efficiency and selectivity in most cases.
It is known that pH plays an important role in the
transfer hydrogenation using formic acid as hydrogen
source [23]. Therefore, transfer hydrogenation of benzal-
dehyde, acetophenone and N-benzylideneaniline were
conducted in a series of formic acid and triethylamine
mixtures with different ratios, and the results are summa-
rized in Table 2. It shows that higher pH dramatically
enhances the reaction rate for the reduction of all three
substrates, with only a slight loss on the selectivity of imine
reduction. The 5:1 HCOOH/NEt3 mixture with a pH of
1.80 cannot serve as a hydrogen source because no cata-
lytic activity was observed for benzaldehyde and aceto-
phenone, while complete decomposition was observed for
N-benzylideneaniline. In the 5:2 HCOOH/NEt3 azeotrope,
the typical hydrogen source for Noyori’s catalyst, reaction
exhibited poor catalytic performance. Changing HCOOH/
NEt3 from 5:2 to 5:3 leads to much faster reaction rate for
all three substrates, e.g., the hydrogenation of benzalde-
hyde can be completed with 30 min. A tradeoff of using
5:3 HCOOH/NEt3 mixture is the selectivity loss of imine
reduction. Further increasing the ratio of NEt3 changes the
media from acidic to basic, leading to inferior catalytic
performance. It was noted that the 5:4 HCOOH/NEt3
mixture formed two layers which can no longer be used.
Based on the above observations, 5:2 and 5:3 HCOOH/
NEt3 mixtures can be used as the reaction media for the
direct reductive amination, depending on the demand of
selectivity or efficiency.
Reactions of benzaldehyde with aniline derivatives
(entry 1–10) follow a distinct trend showing that time
needed for a complete conversion is directly proportional
to the electron withdrawing properties of the substituents,
and selectivities diminish gradually as well. On the other
hand, reactions of benzaldehyde and aliphatic amines
(entry 11–15) shows that primary amines give better
selectivity than secondary amines and steric hindrance is
also an important factor to determine the selectivity.
Reactions involving aliphatic aldehydes (entry 16–19)
can be dramatically improved by changing from 5:2 to 5:3
HCOOH/NEt3 mixture. In the former media, only around
10 % desired product was formed because the aliphatic
aldehydes are subject to aldol condensation under this
acidic condition, while this side reaction can be minimized
in the latter media, leading to much higher selectivities.
It was also noted that an open air catalysis with the use
of TsEN ligand exhibit comparable catalytic performance
with that under nitrogen, indicating that the reaction in the
HCOOH/NEt3 mixture does not require a strict oxygen free
environment. A 59 scale-up reaction was conducted with
benzaldehyde and 40-aminoacetophenone (entry 8) in 5:2
HCOOH/NEt3 azeotrope and an isolated yield of 41 %
desired product was obtained.
A variety of substrates were chosen to investigate the
applicability of this direct reductive amination method in
the 5:2 and 5:3 HCOOH/NEt3 mixtures, and the results are
shown in Table 3. The aldehyde scope involved one aro-
matic aldehyde (benzaldehyde) and two aliphatic alde-
hydes (phenylacetaldehyde and valeraldehyde). On the
Reaction profile between benzaldehyde and 4-fluoroan-
iline was chosen as an example to illustrate the direct
Table 2 pH dependence on the hydrogenation of aldehyde, ketone and imine
HCOOH:NEt3
pH
Conversion (%)/Time (h)
Conversion (%)/Time (h)
Conversion (%)/Time (h)
Selectivity (%)
5:1
1.80
4.60
5.85
6.90
8.08
0 % (1 h)
23 % (1 h)
0 % (1 h)
0 % (1 h)
100 % (1 h)
29 % (1 h)
100 % (1 h)
100 % (1 h)
100 % (1 h)
0 %a
94 %
83 %
76 %
73 %
5:2
5:2.5
5:3
100 % (1 h)
100 % (0.5 h)
100 % (0.5 h)
19 % (1 h)
29 % (1 h)
19 % (1 h)
5:3.5
Conditions: substrate (1 mmol), catalyst[RuCl2(p-cymene)]2/TsDPEN (0.01 mmol), HCOOH/NEt3 mixture (1.0 mL)
a
N-benzylideneaniline fully decomposed to benzaldehyde and aniline
123

Amination in HCOOH/NEt3 Mixture
1571
Table 3 Direct reductive
amination in HCOOH/NEt3
mixture
Entry
H2N–R2
Completion time (h)/Selectivity (%)
5:2 HCOOH/NEt3
Ru(II)/TsDPEN
5:3 HCOOH/NEt3
Ru(II)/TsDPEN
Ru(II)/TsEN
1
Ph–
–C6H4(p-OCH3)
–C6H4(p-OC2H5)
–C6H4(p-CH3)
–C6H5
5 h (100 %)
5 h (100 %)
6 h (97 %)
6 h (94 %)
6 h (95 %)
6 h (89 %)
9 h (52 %)
9 h (51 %)
18 h (32 %)
35 h (17 %)
5 h (93 %)
3 h (84 %)
6 h (13 %)
2 h (33 %)
16 h (13 %)
1 h (9 %)
0.2 h (77 %)
0.2 h (89 %)
0.2 h (77 %)
0.5 h (66 %)
0.5 h (55 %)
0.5 h (42 %)
0.5 h (11 %)
0.5 h (11 %)
0.5 h (4 %)
0.5 h (0 %)
0.5 h (80 %)
0.5 h (68 %)
0.5 h (2 %)
0.5 h (14 %)
0.5 h (0 %)
1 h (62 %)
0.7 h (77 %)
0.7 h (82 %)
1 h (64 %)
1 h (43 %)
2 h (42 %)
1.5 h (22 %)
1.5 h (4 %)
1.5 h (4 %)
1.5 h (0 %)
1.5 h (0 %)
1.5 h (74 %)
1 h (66 %)
1 h (3 %)
2
3
4
5
–C6H4(p-F)
–C6H4(p-Cl)
–C6H4(p-CF3)
–C6H4(p-COCH3)
–C6H4(p-CN)
–C6H4(p-NO2)
–CH2Ph
6
7
8
9
10
11
12
13
14
15
16a
17a
18a
19a
–(CH2)3CH3
–C(CH3)3
–(CH2)5–
1 h (19 %)
1 h (0 %)
–(CH3)Ph
Conditions: substrate (1 mmol),
catalyst [RuCl2(p-cymene)]2/
TsDPEN(0.01 mmol), HCOOH/
NEt3 mixture (1.0 mL)
PhCH2–
–Ph
1 h (60 %)
1 h (47 %)
1 h (30 %)
1 h (27 %)
–CH2Ph
1 h (8 %)
1 h (61 %)
CH3(CH2)3–
–Ph
1 h (11 %)
1 h (7 %)
1 h (43 %)
a
Conversions and selectivities
were determined by GC–MS
–CH2Ph
1 h (25 %)
alcohol product, the intermediate imine under current
conditions is more susceptible to the decomposition,
resulting in small amount of benzyl alcohol side product.
100
80
60
40
20
0
Benzaldehyde
3 Conclusions
Imine
Amine product
Benzyl alcohol
In summary, direct reductive amination of aldehydes by
using Noyori’s catalyst was successfully implemented. For
the reactions involving aromatic aldehydes, high selectiv-
ities can be obtained from the commonly used 5:2
HCOOH/NEt3 azeotrope, while a much faster reaction can
be achieved in the 5:3 HCOOH/NEt3 mixture. For the
reactions involving aliphatic aldehydes, 5:3 HCOOH/NEt3
mixture proves to be a better media due to the neutral pH.
TsEN can be used as an alternative ligand to TsDPEN for a
more cost-effective approach. No N-formylated side pro-
ducts were observed in any of the reactions. The next stage
of this project will focus on the asymmetric catalysis based
on this direct reductive amination method.
0
1
2
3
4
5
6
Time (h)
Fig. 1 Reaction profile for direct reductive amination of benzalde-
hyde and 4-fluoroaniline in 5:2 HCOOH/NEt3 azeotrope
reductive amination process. The concentration profile of
reactant, intermediate and products versus time is shown in
Fig. 1. The two substrates immediately react with each
other to give the intermediate imine, which is then slowly
transformed to the product amine. After around 20 min, the
distribution of intermediate, benzaldehyde and amine
product is about 80:10:10, and only trace amount of benzyl
alcohol is observed. As reaction progresses, amine product
is steadfastly formed and benzyl alcohol is formed at a
rather slow rate. Compared to the reaction involving
4-methoxyaniline, which does not lead to any benzyl
4 Experimental
Catalysis was carried out in a glove box under a nitrogen
atmosphere. All reagents and solvents were degassed
before being transferred into glove box. Conversions and
123

1572
M. Zhu
1
References
selectivities were monitored by H NMR spectroscopy or
GC–MS. The 1H NMR spectra were recorded with TMS as
an internal standard on a Varian 400 spectrometer. GC–MS
was performed on a Varian Saturn 2100 instrument in the
electron ionization (EI) mode. N,N-dimethylbenzylamine
was used as a GC–MS internal standard if needed.
1. Noyori R, Hashiguchi S (1997) Acc Chem Res 30:97–102
2. Murata K, Okano K, Miyagi M, Iwane H, Noyori R, Ikariya T
(1999) Org Lett 1:1119–1121
3. Vedejs E, Trapencieris P, Suna
64:6724–6729
E (1999) J Org Chem
4. Koike T, Murata K, Ikariya T (2000) Org Lett 2:3833–3836
5. Okana K, Muratsugu S, Ikariya T (2000) Tetrahedron Lett
41:9277–9280
Typical procedure of direct reductive amination or
hydrogenation by using Noyori’s catalyst in formic acid/
triethylamine mixture: [RuCl2(p-cymene)]2 (3.1 mg,
0.005 mmol) and (S,S)-TsDPEN (4.4 mg, 0.012 mmol)
were stirred in 1 mL formic acid-triethylamine mixture in a
7 mL glass vial (OD 9 H: 17 mm 9 54 mm) for 1 h, and
a homogeneous yellow solution was formed. To the solu-
tion, aldehyde (1 mmol) and amine substrate (1 mmol)
were added to start the reaction. After an appropriate per-
iod, an aliquot (5 lL) was withdrawn and mixed with
0.6 mL CDCl3 to prepare the NMR sample. The conver-
sion and selectivity were determined by taking the ratio of
the integration of the signal for the hydrogen atoms directly
attached to C=O, C=N, C–O and C–N. A more dilute
sample was prepared for GC–MS.
6. Yamada I, Noyori R (2000) Org Lett 2:3425–3427
7. Watanabe M, Murata K, Ikariya
67:1712–1715
T (2002) J Org Chem
8. Xue D, Chen Y-C, Cui X, Wang Q-W, Zhu J, Deng J-G (2005) J
Org Chem 70:3584–3591
9. Palmer MJ, Wills
10:2045–2061
M
(1999) Tetrahedron Asymmetry
10. Wang C, Wu X, Xiao J (2008) Chem Asian J 3:1750–1770
11. Nugent TC, El-Shazly M (2010) Adv Synth Catal 352:753–819
´
´
12. Fleury-Bregeot N, de la Fuente V, Castillon S, Claver C (2010)
ChemCatChem 2:1346–1371
13. Uematsu N, Fujii A, Hashiguchi S, Ikariya T, Noyori R (1996) J
Am Chem Soc 118:4916–4917
14. Willams GD, Pike RA, Wade CE, Wills M (2003) Org Lett
5:4227–4230
15. Strotman NA, Bazter CA, Brands KMJ, Cleator E, Krska SW,
Reamer RA, Wallace DJ, Wright TJ (2011) J Am Chem Soc
133:8362–8371
16. Wang T, Zhuo L-G, Li Z, Chen F, Ding Z, He Y, Fan Q-H, Xiang
J, Yu Z-X, Chan ASC (2011) J Am Chem Soc 133:9878–9891
An example of the scale-up direct reductive amination
between benzaldehyde and 40-aminoacetophenone:
[RuCl2(p-cymene)]2 (16 mg, 0.025 mmol) and (S,S)-TsD-
PEN (22 mg, 0.060 mmol) were stirred in 5 mL 5:2 formic
acid-triethylamine azeotrope for 1 h, and a homogeneous
yellow solution was formed. To the solution, benzaldehyde
(510 lL, 5 mmol) and 40-aminoacetophenone (675 mg,
5 mmol) were added, the resulting mixture were stirred for
9 h. The reaction mixture was mixed with 10 mL distilled
water and extracted with CH2Cl2 (3 9 10 mL). The com-
bined extract was dried over anhydrous MgSO4 and con-
centrated. The crude product obtained was further purified
by a flash chromatography on silica-gel using 1:1 EtOAc/
hexane solvent mixture as eluent. Light yellow powder
(460 mg) was obtained as the pure product which was
verified by 1H NMR spectra. Yield: 41 %. 1H NMR
(CDCl3): d 2.49 (s, 3H), 4.41 (s, 2H), 6.60 (d, J = 8.0 Hz,
2H), 7.26–7.35 (m, 5H), 7.82 (d, J = 8.0 Hz, 2H).
˚
17. Aberg JB, Samec JSM, Backvall J-E (2006) Chem Commun,
¨
2771–2773
18. Hu L, Cao X, Ge D, Hong H, Guo Z, Chen L, Sun X, Tang J,
Zheng J, Lu J, Gu H (2011) Chem Eur J 17:14283–14287
19. O’Connor D, Lauria A, Bondi SP, Saba S (2011) Tetrahedron
Lett 52:129–132
20. Byun E, Hong B, De Castro KA, Lim M, Rhee H (2007) J Org
Chem 72:9815–9817
21. Sato S, Sakanoto T, Miyazawa E, Kikugawa Y (2004) Tetrahe-
dron 60:7899–7906
22. Smith MB, March
J (2001) Advanced organic chemistry,
Chap. 16, 5th edn. Wiley, New York, p 1187
23. Wu X, Li X, King F, Xiao J (2005) Angew Chem Int Ed
44:3407–3411
123