Catalysis Science & Technology
Communication
butylamine (1c). The two catalysts gave comparable conversion in Scheme 1, where any benzylamine (1e) formed as a by-product
of 1a into products, however, higher selectivity of the desired from over reduction of benzonitrile (1a) is less able to compete
product, 1d was observed for Pt/C catalyst (Table 1, ESI†). with the excess of butylamine (1c) present for amination of the
Therefore, Pt/C was selected as a catalyst for further experi- initially formed imine (1b).
ments. Choice of a suitable solvent is necessary in the flow
We were pleased to find that a range of nitriles and amines
reactor not only to avoid precipitation of reactants/products underwent reductive coupling to give di-substituted amines
mixture which leads to blockage in the reactor but also to with very good conversions and selectivities (Table 1). Benzonitrile
achieve higher conversion and selectivity. Initial experiments (1a) was converted into a range of N-benzylamines (Table 1,
were carried out using methanol, 2-propanol, dichloromethane entries 1–11), where only cyclohexylamine gave less than 90%
and toluene as solvents (Table 2, ESI†) for reductive amination conversion under these reaction conditions, possibly for steric
of (1a) with (1c). Toluene and methanol gave similar results in reasons. The use of phenylacetonitrile (2a) and phenylpropio-
terms of conversion and selectivity to product (1d). Lower nitrile (3a) led to similar results, with the formation of the
conversion of 1a was observed in the case of 2-propanol and reductively coupled products with good conversions and selectiv-
dichloromethane solvents. No by-products were seen other ities (Table 1, entries 12–25). Although good selectivities could be
than benzylamine (1e) and dibenzylamine (1f). Therefore, obtained using butyronitrile (4a) when an excess of amine was
toluene was selected as a solvent for further study.
present, the conversions were lower.
Initially, the one pot reductive amination of benzonitrile (1a)
The Pt/C catalyst is quite stable. No drop in the conversion
was achieved with almost complete conversion (99%) in the of 1a and selectivity of 1d was observed up to 480 min time on
reaction with butylamine (1c) using a molar ratio of 1 : 1. The stream data for reductive amination of 1a with 1c (Fig. 1). The
selectivity was good, although 12% of dibenzylamine was main advantages of the continuous flow microreactor in the
observed as a by-product. However, by increasing the amount present study are high catalyst loading and low residence time
8
of butylamine present to three equivalents, the amount of self- (less than 10 seconds) as compared with 2–12 h in batch reactors.
coupled product was reduced to 4%. In all cases, using an
For several cases, we chose to purify the amine products by
excess of the amine led to a higher selectivity for product column chromatography and were pleased to find that the
formation. This is consistent with the intermediates identified products could be obtained in good isolated yield. Benzonitrile
was converted into a range of amines in up to 93% isolated yield
(Table 2, entries 1–7) and phenylacetonitrile also underwent
reductive amination with a satisfactory yield (Table 2, entry 8).
Table 2 Isolation of representative amines
Conclusions
In summary, various substituted secondary amines were
synthesised by a Pt/C catalysed one pot selective reductive
amination of aromatic and aliphatic nitriles with primary
amines in a continuous flow multichannel microreactor using
molecular hydrogen as a reducing agent.
a
Entry
Amine product
Isolated yield (%)
1
2
81
93
We thank the EPSRC for funding through grant EP/G028141/
1
and GlaxoSmithKline for additional support.
3
4
72
76
Notes and references
‡
Brief experimental procedure: in a typical experimental run, nitrile
and primary amine were premixed at the desired concentration in
toluene and charged to the feed vessel. Pt/C (3 wt% Pt) as a catalyst
was filled in the 3 Â 3 mm channel. The reaction mixture from the feed
vessel was fed into the reactor using an HPLC pump (Kontron) at
À1
0
.1 mL min flowrate. Hydrogen gas was supplied to the reactor using
À1
mass flow controller (Brooks) at 17 mL min
flowrate and 6 bar
5
67
70
pressure. Constant temperature (105 1C) of the reactor was maintained
by circulating heat transfer fluid through the micro-heat exchangers
using a recirculating bath (Haake). The pressure of the reactor was
controlled by a back pressure regulator (Brooks) and pressure drop
across the reactor was monitored using a differential pressure transducer
6
7
(
Bronkhorst). After completion of the reaction, the reaction mixture was
collected by a high pressure valve. Analysis of reaction mixture was
carried out by gas chromatography (Varian 3900) equipped with CP-Sil
81
74
5CB capillary column (15 m length and 0.25 mm dia.) and GC-MS.
8
1 S. A. Lawrence, Amines: Synthesis, Properties, and Application,
Cambridge University Press, Cambridge, 2004.
(a) S. L. Buchwald, C. Mauger, G. Mignani and U. Scholz, Adv. Synth.
Catal., 2006, 348, 23; (b) J. F. Hartwig, Synlett, 2006, 1283.
2
a
Isolated yield after column chromatography.
This journal is c The Royal Society of Chemistry 2013
Catal. Sci. Technol., 2013, 3, 85--88 87
Sharma, Sumeet K.
