Reductive amination of furfural to furfurylamine using aqueous ammonia solution and molecular hydrogen: An environmentally friendly approach

Article abstract of DOI:10.1039/c5gc01352f

A simple and highly efficient method was developed for the transformation of furfural (a biomass derived aldehyde) to furfurylamine by reductive amination using an aqueous solution of ammonia and molecular hydrogen as an amine source and a reducing agent, respectively. By choosing a suitable catalyst, such as Rh/Al₂O₃, and reaction conditions, a very high selectivity of furfurylamine (～92%) can be achieved within the reaction time of 2 h at 80 °C. A detailed analysis of the reaction system sheds some light on the reaction pathway and provides an understanding about each elementary step. The reaction was believed to proceed via an imine pathway although no such intermediate was detected because of the highly reactive nature. Optimization of different reaction parameters such as hydrogen pressure, temperature and substrate/ammonia mole ratio is shown to be critical to achieve high selectivity of furfurylamine. Time-dependent reaction profiles suggested that a Schiff base type intermediate was in the detectable range, which offers indirect evidence of the formation of imine. Competitive hydrogenation and amination of an aldehyde group were strongly dictated by the nature of the metal used. The studied protocol represents an environmentally benign process for amine synthesis, which can be effectively extended to the other aldehydes also. The studied catalyst could be recycled successfully without any significant loss of catalytic activity.

Full text of DOI:10.1039/c5gc01352f

Green Chemistry
View Article Online
View Journal
PAPER
Reductive amination of furfural to furfurylamine
using aqueous ammonia solution and molecular
hydrogen: an environmentally friendly approach†
Cite this: DOI: 10.1039/c5gc01352f
Maya Chatterjee,*a Takayuki Ishizakaa and Hajime Kawanami*a,b
A simple and highly efficient method was developed for the transformation of furfural (a biomass derived
aldehyde) to furfurylamine by reductive amination using an aqueous solution of ammonia and molecular
hydrogen as an amine source and a reducing agent, respectively. By choosing a suitable catalyst, such as
Rh/Al2O3, and reaction conditions, a very high selectivity of furfurylamine (∼92%) can be achieved within
the reaction time of 2 h at 80 °C. A detailed analysis of the reaction system sheds some light on the reac-
tion pathway and provides an understanding about each elementary step. The reaction was believed to
proceed via an imine pathway although no such intermediate was detected because of the highly reactive
nature. Optimization of different reaction parameters such as hydrogen pressure, temperature and sub-
strate/ammonia mole ratio is shown to be critical to achieve high selectivity of furfurylamine. Time-
dependent reaction profiles suggested that a Schiff base type intermediate was in the detectable range,
which offers indirect evidence of the formation of imine. Competitive hydrogenation and amination of an
aldehyde group were strongly dictated by the nature of the metal used. The studied protocol represents
an environmentally benign process for amine synthesis, which can be effectively extended to the other
aldehydes also. The studied catalyst could be recycled successfully without any significant loss of catalytic
activity.
Received 17th June 2015,
Accepted 11th August 2015
DOI: 10.1039/c5gc01352f
www.rsc.org/greenchem
compelling advantages such as mild reaction conditions, inexpen-
sive reagents and a wide availability of substrates over the
Introduction
Amines are an important class of compounds, found in other classical amine synthesis methods.1 Generally, there are
natural products and pharmaceuticals, and act as intermedi- two different approaches to reductive amination: (i) indirect
ates in the synthesis of different valuable organic compounds methods involving pre-formed imines through the reaction
including dyes and agrochemicals.1,2 Selective synthesis of between amines and carbonyl compounds followed by hydro-
primary amines from carbonyl compounds is still a challen- genation using a suitable hydride source and (ii) direct
ging task because primary amines are more nucleophilic than methods in which carbonyl compounds are mixed with an
ammonia and easily react with aldehydes to form secondary amine source and a reducing agent that selectively reduces the
amines or a mixture of products. One of the best suited in situ generated imine. Thus, most of the literature reports are
methods for effective C–N bond formation is the reductive related to exploration of a suitable process with different redu-
amination of carbonyl compounds, which involves an amine cing agents, solvents, catalysts and amine sources, which has a
source and a reducing agent. The reductive amination is a profound effect on the reductive amination of aromatic carbo-
highly useful and robust method for various transformations nyl compounds.5–7 The choice of reducing agents plays an
involved in synthetic medicinal chemistry.3,4 It offers important role in the success of the reaction because it must
be able to reduce imine intermediates instead of the carbonyl
compounds. Hence, a stoichiometric amount of different types
of reducing agents such as borohydride,8 formate,9 silane,10
aMicroflow Chemistry Group, Research Institute for Chemical Process Technology,
etc. were used. However, these reducing agents generate large
amounts of waste including some toxic waste. In this context,
molecular hydrogen is considered as the most atom-economical
and environmentally friendly reducing agent for reductive
amination, particularly in the large-scale production.11,12
In the case of primary amine synthesis, it is worth using
AIST Tohoku, 4-2-1, Nigatake, Miyagino-ku, Sendai 983-8551, Japan.
E-mail: c-maya@aist.go.jp, h-kawanami@aist.go.jp; Fax: +81 22 237 5388;
Tel: +81 22 237 5213
bCREST, Japan Science and Technology (JST), 4-1-8, Honcho, Kawaguchi,
Saitama 332-0012, Japan
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
c5gc01352f
This journal is © The Royal Society of Chemistry 2015
Green Chem.

View Article Online
Paper
Green Chemistry
ammonia for the transformation of carbonyls to amines trial scale, amination of furfural was generally conducted on a
because it is an inexpensive commodity chemical, produced in hydrogenation catalyst, mainly Ni or Co, along with additives
bulk, easily available and reduced waste.13 However, reductive and gaseous ammonia in dioxane or alcohol as the solvent.
amination with ammonia has not been commonly investigated One of the drawbacks of this process is the deactivation of the
because typically it requires high pressure and high tempera- catalysts. On the laboratory scale, Au, Pt and Ir colloids sup-
ture, interaction with transition metal catalysts to form inactive ported on sulfonic acid functionalized silica were developed
complexes and additionally most of the times a mixture of pro- for reductive amination of furfural27 using aniline. In addition,
ducts was obtained.14 Gross et al. first accomplished the reduc- electrochemical reduction of furfuryl oxime has been reported
tive amination of benzaldehyde using aqueous ammonia in for the preparation of furfurylamine, however, the yield was
the presence of a homogeneous Rh catalyst, which resulted in relatively low.28 Therefore, a more selective method for the
86% yield of benzylamine at 135 °C and 6.5 MPa of hydrogen transformation of furfural to furfurylamine under mild reac-
pressure.15 Moreover, to improve the catalyst separation and tion conditions would be valuable.
reuse, heterogeneous catalysts were developed.16 As hydrogen-
Here, we targeted the reductive amination of furfural as a
ation of carbonyl compounds is a major side reaction, to model reaction to achieve high selectivity of furfurylamine
hinder the highest yield of primary amines, different types of using a 28% aqueous solution of ammonia and molecular
catalysts were tested. Among the heterogeneous catalysts, hydrogen in the presence of a suitable hydrogenation catalyst.
RANEY® Ni was the most frequently used catalyst to obtain The commercially available 5% Rh/Al2O3 was used as the cata-
primary amines from aldehydes with fairly good yield.17 lyst and the reaction was conducted under mild reaction con-
However, it has a number of disadvantages such as being ditions without any additional solvents or additives.
difficult to handle, environmentally harmful and requiring
harsh reaction conditions. Thus, along with the modification
including addition of other metals or promoters to improve
Results and discussion
the performance and longevity, Ni and Co based catalysts were
introduced,17 which also cause problems of waste due to the
Scheme 1 represents a general reaction path of the conversion
presence of additives. In a very recent publication, cobalt oxide
of aldehydes to amines in the presence of ammonia. Reductive
supported on carbon was used for the synthesis of secondary
amination of aldehydes follows some elementary steps.
and tertiary amines.18 Supported noble metal catalysts are con-
According to Mignonac, who first demonstrated the reductive
sidered as an excellent choice to obtain primary amines with
amination of carbonyl compounds using a heterogeneous Ni
high purity under mild reaction conditions. Generally, acti-
based catalyst proposed that the reaction between ammonia
vated carbon was used as a support material because of its
and the carbonyl compounds results in imines. Hydrogenation
stability as well as cost-effectiveness. Pd is one of the most
of imines affords primary amines and consequently secondary
common metal catalysts supported on carbon used for the
amines were formed by the interaction of primary amines with
conversion of benzaldehyde to benzylamine19 whereas Rh was
the carbonyl compounds. Therefore, the possibility of finding
used occasionally.20 Moreover, Ru/C pre-treated with
secondary amines along with the primary amines is inevita-
(NH4)2S2O8 was also used for the reductive amination of benz-
ble.29 In addition, the selectivity of primary amines was also
aldehyde using a methanol solution of ammonia at 90 °C and
hampered by the formation of by-products mainly originating
4 MPa of hydrogen pressure to obtain benzylamine with 77%
from hydrogenation and/or aldol condensation. To overcome
the possibility of the formation of hydrogenated products like
alcohol, optimization of different reaction parameters such as
selectivity.21 Bodis et al. compared the activity of carbon sup-
ported Pt, Pd, Ru and Rh to explore the reductive amination of
butyraldehyde.22 However, a metal supported on activated
carbon is difficult to separate from the aqueous matrix and as
it originates from natural material, and has different pro-
perties, which could affect the catalytic performance.19a
Pyrolysis of lignin primarily produced aldehydes and
phenols with diverse structures.23 Furanic aldehydes such as
furfural are some of the important chemicals readily accessible
from biomass. Furfural is a key compound for the production
of non-petroleum derived materials. Furfurylamine is one of
the most important derivatives used in the manufacture of
pharmaceuticals, pesticides, etc. In addition, ionic liquids
were also successfully synthesized by the reductive amination
of furanic aldehydes.24 However, those aldehydes, which could
be an ideal candidate for a sustainable extension of the amine
chemistry are rarely explored for amination because of the easy
hydrogenation of –CHO groups as well as the ring opening
Scheme 1 A general pathway of reductive amination of aldehydes with
reaction and reported in the patent literature.25,26 On an indus- ammonia.
Green Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Green Chemistry
Paper
temperature, substrate/ammonia ratio, hydrogen pressure, Rh/C produced mainly a Schiff base type intermediate (57.9%)
reaction time as well as pre-equilibration time was required.
(Table 1; entry 4) and the selectivity of furfurylamine was com-
paratively low (27.3%) under similar reaction conditions. Con-
sidering the Pt catalysts, independent of the support,
hydrogenation was the main route of transformation of the
aldehyde group. Thus, instead of furfurylamine, furfuryl
alcohol was obtained as a major product and the selectivity
varies from ∼47 to 72% (Table 1; entries 7–9). When the reac-
tion was conducted on Ru/C catalysts, 97.9% of secondary
amines were produced, but Ru/Al2O3 was active towards the
hydrogenation rather than amination (Table 1: entries 10 and
11). Hence, the selectivity of furfurylamine followed the
decreasing order of Rh/Al2O3 > Rh/MCM-41 ≫ Rh/C ≈ Pt/C >
Pd/Al2O3 ≈ Pt/MCM-41 ≈ Pd/C. Comparison of the results
revealed that the selectivity of furfurylamine can be modulated
by the nature of the metal ion instead of the support material.
For instance, the selectivity of furfurylamine was 91.5% on Rh/
Al2O3, whereas Pd/Al2O3 and Pt/Al2O3 exhibit a very low selecti-
vity of 18.4 and 12.4%, respectively. In addition, no furfuryl-
amine was detected when a Ru/Al2O3 catalyst was used. Hence,
from the present results it could be suggested that the selecti-
vity of furfurylamine mainly depends on the metal rather than
the support used. Evaluating the catalytic performance in
terms of activity and selectivity to the desired furfurylamine,
we observed that Rh/Al2O3 displayed best results. Thus, further
studies were continued with the Rh/Al2O3 catalyst.
Catalyst screening
Before the optimization of reaction conditions, it is necessary
to determine a suitable catalyst, which could promote amin-
ation over the hydrogenation of carbonyl compounds. Hence,
we have studied different potential catalysts for reductive
amination of furfural and the results are shown in Table 1.
Targeting the high selectivity of furfurylamine, here, we con-
sidered different noble metal catalysts such as Pt, Pd, Rh and
Ru supported on activated carbon (C), mesoporous silica
(MCM-41) and alumina (Al2O3) to investigate the influence of
support materials (Table 1) operating under similar reaction
conditions. Among the catalysts screened, complete conver-
sion was detected only on Pd and Rh independent of the
support used (Table 1; entries 1–6). The conversion of furfural
varies from 42 to 100% on Pt (Table 1; entries 7–9) depending
on which support was used. On the other hand, a compara-
tively low conversion of 25 and 33% was obtained on Ru/C and
Ru/Al2O3, respectively, whereas Ru/MCM-41 was completely
inactive (Table 1; entries 10–12). Considering the product dis-
tribution, the reaction catalyzed by Pd/C, proceeded with the
formation of a ring hydrogenated secondary amine as a major
product (selectivity = 87.3%) along with furfurylamine (10.6%)
and a very small amount of hydrogenated product (furfuryl
alcohol; 2.1%) (Table 1; entry 1). Other Pd catalysts such as Pd/
Al2O3 and Pd/MCM-41 generate secondary amines with very
high selectivity of 77.2 and 87.0%, respectively (Table 1;
Optimization of different reaction parameters
Effect of temperature. Based on the importance to obtain
entries 2 and 3). However, furfurylamine was detected with furfurylamine with very high selectivity, different reaction para-
very poor selectivity of 18.4% only on Pd/Al2O3 (Table 1; entry meters were optimized. The influence of reaction temperature
2). Focusing on the Rh catalysts, a very high selectivity of fur- was explored in the range of 35 to 130 °C and the results are
furylamine was obtained on Rh/MCM-41 (82.8%) and Rh/Al2O3 shown in Fig. 1. It can be seen that there was no effect of temp-
(91.5%) catalysts (Table 1; entries 5 and 6). On the contrary, erature on the conversion of furfural, but the selectivity of fur-
Table 1 Catalyst screening for reductive amination of furfural
Select.
Run
Catal.
Conv.
Othersa
1
2
3
4
5
6
7
8
Pd/C
100.0
100.0
100.0
100.0
100.0
100.0
100.0
42.2
55.6
25.2
33.5
—
10.6
18.4
—
27.3
82.8
91.5
26.9
10.6
12.4
—
—
—
—
—
—
—
1.5
42.3
—
—
—
—
—
2.1
—
—
—
3.8
—
—
—
—
—
—
<0.1
—
—
—
—
13.0
57.9
—
—
—
—
—
—
87.3
4.4
—
—
—
—
—
—
—
—
—
—
—
—
—
—
<0.5
—
—
—
—
<0.1
—
Pd/Al2O3
Pd/MCM
Rh/C
Rh/MCM
Rh/Al2O3
Pt/C
Pt/MCM
Pt/Al2O3
Ru/C
Ru/Al2O3
Ru/MCM
Rh/Al2O3
77.2
87.0
14.3
13.4
8.5
—
—
18.8
97.9
—
—
71.5
47.1
68.7
—
62.2
—
9
10
11
12
13
—
37.7
—
2.1
—
—
—
—
85.5
<0.1
—
<4.0
—
11.2
100.0
—
—
—
Reaction conditions: catalyst = 0.002 g, substrate = 0.2 g, substrate/ammonia (mole ratio) = 0.03, temperature = 80 °C, reaction time =
2 h, PH = 2 MPa. a Others = unknown.
2
This journal is © The Royal Society of Chemistry 2015
Green Chem.

View Article Online
Paper
Green Chemistry
Fig. 2 Effect of substrate/ammonia ratios on product distribution.
Fig. 1 Effect of temperature on the transformation of furfural to fur-
Reaction conditions: substrate = 0.2 g, catalyst = 0.002 g, PH = 2 MPa,
furylamine. Reaction conditions: substrate = 0.2 g, catalyst = 0.002 g,
2
temperature = 80 °C and reaction time = 2 h.
substrate/ammonia = 0.03, PH = 2 MPa and reaction time = 2 h.
2
furylamine changed. A low reaction temperature of 35 °C furfurylamine and displayed the highest selectivity of 91.5%
resulted in a product mixture composed of mainly a Schiff when the substrate to ammonia ratio reached 0.03. Moreover,
base type intermediate (selectivity = ∼76%), furfurylamine with the mole ratio of furfurylamine and the secondary amine was
low selectivity of 18.0% and a very small amount of secondary also changed from 57 : 43 to 95 : 5 with a change in the sub-
amines (5.7%). An increase in temperature significantly strate to ammonia ratio from 0.12 to 0.03 and then remains
increased the selectivity of furfurylamine and reached a almost constant. A further increase in the concentration of
maximum of 91.5% at 80 °C. Notably, a Schiff base type inter- ammonia did not favor the exclusive formation of furfuryl-
mediate was obtained by the interaction between ammonia amine. Thus, under the applied reaction conditions
a
and aldehydes in the reductive amination of aldehydes, which maximum selectivity of 91.5% of furfurylamine could be
might be transformed to primary and secondary amines via achieved using an optimum ratio of substrate/ammonia (0.03).
hydrogenolysis or hydrogenation, respectively.30 From the The role of ammonia is critical as it could help the Schiff base
results it could be inferred that the temperature below 80 °C formation as well as improve the selectivity of furfurylamine.
was insufficient for the generation of furfurylamine with good As mentioned before, the use of excess ammonia improves the
selectivity within 2 h, which required higher temperatures or selectivity of primary amines by suppressing the generation of
longer reaction times. For instance, when the reaction time secondary amines. According to Scheme 1, primary amines
was extended to 17 h at 35 °C, the selectivity of furfurylamine were formed through the reaction between aldehydes and
increased to 46.5%. These results might be attributed to the ammonia via an imine intermediate. Primary amines are more
shifting of adsorption–desorption equilibrium of the inter- active than ammonia and react preferentially to form second-
mediate, which dominated the reaction path to generate the ary or tertiary amines. A large excess of ammonia prevents
furfurylamine and consequently enhanced the selectivity.19 further reaction of primary amines by shifting the equilibrium
Unexpectedly, further increase of temperature above 100 °C and enhanced the selectivity.31
turns the product mixture black and the selectivity of furfuryl-
Effect of pre-equilibration time on the product distribution.
amine abruptly dropped to 17.6% because of the formation of As described in the Experimental section, before the introduc-
some unidentified products. Hence, we selected a threshold tion of hydrogen, the reaction mixture containing the catalyst,
temperature of 80 °C to achieve the highest selectivity of the substrate and ammonia was pre-equilibrated for 1 h without
desired compound.
any stirring to ensure a constant temperature. We speculated
Influence of substrate/ammonia mole ratios. Generation of the probability of the transformation of furfural and ammonia
primary amines with good yield strongly depends on the con- to the Schiff base type intermediate in the pre-equilibration
centration of ammonia used. Fig. 2 shows the effect of the sub- stage.
strate/ammonia ratio on product distribution at 80 °C and the
To check this possibility, a series of experiments were con-
reaction time of 2 h. The substrate/ammonia ratio varied from ducted with different pre-equilibration times (0, 15, 30, 60,
0.12 to 0.015. In the absence of ammonia, furfural was trans- and 120 min) and we observed a time dependent color change
formed to furfuryl alcohol by hydrogenation. Addition of of the reaction mixture (ESI Fig. S1†). All mixtures were ana-
ammonia causes reductive amination of the –CHO group and lyzed using GC-MS and the results are shown in Fig. 3a. Inter-
based on the substrate/ammonia ratio the selectivity of furfuryl- estingly, furfural and ammonia remain unreacted until
amine changed. For instance, at higher ratio (0.12) (low con- 15 min, but after 30 min (the minimum time required for
centration of ammonia), a significant amount of secondary stabilization of the required temperature of 80 °C) a detectable
amines (51.9%) and hydrogenated products (10.9%) were amount of Schiff base type of product was found as the
formed. A decrease in the substrate/ammonia ratio (increase only product. Applying the pre-equilibration time of 30 min,
in the concentration of ammonia) improved the selectivity of 9% furfural was converted to the intermediate, which was
Green Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Green Chemistry
Paper
genation. This result could be attributed to the strong binding
of the Schiff base type intermediate with the catalyst through
the nitrogen atom, which may stimulate fast hydrogenation
rather than hydrogenolysis.32 Hence, an optimum pre-equili-
bration time of 1 h was chosen before the introduction of
hydrogen.
Reaction time profile vs. product distribution. As men-
tioned in the previous section a Schiff base type intermediate
was the only detectable intermediate in the product mixture
after 1 h of pre-equilibration, which prompted us to explain
the reaction sequences of the formation of furfurylamine
under the studied protocol in detail. Thus, a time course of
the reaction was investigated by varying the reaction time from
5 min to 18 h after maintaining the pre-equilibration time of
1 h and the results are shown in Fig. 4. The displayed results
evidenced a complete transformation of the substrate to the
product mixture from the beginning. We were unable to detect
any primary imines even in the shortest reaction time of 5 min
probably because of the highly reactive nature of the imines. It
is worth mentioning that after 1 h of pre-equilibration, the
resultant mixture contains ∼89% of the substrate, and 11% of
the intermediate. After the introduction of hydrogen, all sub-
strates were consumed within the reaction time of 5 min. The
generated product mixture contains mainly a Schiff base type
intermediate with a very high selectivity of 83.8%. In addition,
furfurylamine and secondary amines were also detected with a
very low selectivity of 12.2 and 3.9%, respectively. Note that a
relatively short time was required for complete conversion of
furfural, which suggested a very fast reaction between furfural
and ammonia in the presence of hydrogen. Furthermore, it
Fig. 3 (a) Pre-equilibration time dependent yield of a Schiff base type
intermediate in the absence of hydrogen. (b) Influence of pre-equili-
bration time on product distribution. Reaction conditions: substrate =
0.2 g, catalyst = 0.002 g, substrate/ammonia = 0.03, PH = 2 MPa, temp-
2
erature = 80 °C and reaction time = 2 h.
increased to 11.4% when the time was extended to 1 h and the appears that an already formed Schiff base intermediate was
conversion further improved to 26% within 2 h, however, com- converted to primary and secondary amines. As the reaction
plete consumption of the substrate was not observed before time increased, the selectivity of furfurylamine was enhanced
the introduction of hydrogen despite further extending the significantly and reaches a maximum of 91.5% within 2 h.
reaction time. Hence, from the obtained results it could be Apparently, the time vs. selectivity profile as depicted in Fig. 4
inferred that the complete transformation of furfural did not revealed that the furfurylamine was formed with the expense
occurred in the pre-equilibration stage, which requires the of the Schiff base type intermediate because the selectivity of
presence of hydrogen; one of the key components of this the intermediate was considerably high in the beginning
reaction.
Further evolution was focused on the influence of the pre-
(83.8%) and then completely diminished within 2 h followed
equilibration time on the product distribution (Fig. 3b). From
the results it is evident that selectivity towards the desired
product can be achieved by an appropriate choice of pre-
equilibration time. The lowest pre-equilibration time (0 min)
resulted in furfurylamine (69.9%), secondary amines (20.3%)
and furfuryl alcohol (7.7%). Interestingly, no hydrogenation
product was identified after the application of pre-equili-
bration time before hydrogen introduction. The highest selecti-
vity of furfurylamine was achieved when the time was extended
to 1 h, but a prolonged initiation period decreased the selecti-
vity. When we used the pre-equilibration time of less than
30 min, the system temperature was insufficient to initialize
the reaction. On the other hand, as the initiation period was
increased to 2 h, the yield of the Schiff base type intermediate
increased. As a result when the hydrogen was introduced the
Fig. 4 Reaction time vs. product distribution profile. Reaction con-
ditions: substrate = 0.2 g, catalyst = 0.002 g, substrate/ammonia = 0.03,
reaction favored the formation of secondary amines via hydro- PH = 2 MPa and temperature = 80 °C.
2
This journal is © The Royal Society of Chemistry 2015
Green Chem.

View Article Online
Paper
Green Chemistry
by the increased selectivity of furfurylamine. On the other sation product of aldehydes and imines in which the imine
hand, no definite trend was observed between the disappear- function is stabilized by the mesomerism with the two furyl
ance of the intermediate and the selectivity of the secondary groups.31a From the time course of the reaction, we suggested
amine which was comparatively low and remains almost con- the formation of furfurylamine from the Schiff base type inter-
stant with time. From these results it might be suggested that mediate, but the hydrogen pressure dependent selectivity of
the Schiff base type intermediate was more inclined to the furfurylamine encouraged us to consider the route of imine
generation of furfurylamine rather than hydrogenation to hydrogenation to primary amines. Here, we can speculate two
produce secondary amines. Further extending the reaction reactions using imines; (i) condensation with aldehydes and
time to 18 h, the ring hydrogenated product tetrahydrofurfuryl- (ii) hydrogenation to furfurylamine. At low hydrogen pressure,
amine (73.9%), along with methylfurfural and secondary the reaction between imines and aldehydes was evident
amines were formed. It is imperative to note that longer reac- through the formation of the intermediate with a compara-
tion times preferred hydrogenation.
tively high selectivity (39.6%), which correlates well with the
Effect of hydrogen pressure. In the reductive amination of result of this reaction in the absence of hydrogen. When the
aldehydes, hydrogen as the reducing agent plays a significant pressure increased, hydrogenation of imines to primary
role in the success of the reaction as the selectivity of the amines was favored and the selectivity of furfurylamine
desired primary amine is generally hampered by the hydrogen- increased. Combining the time profile of product distribution
ation of the aldehyde group. Thus, we investigated the effect of one could infer that during the initial phase of the reaction
hydrogen pressure on the conversion and selectivity of furfuryl- the formation of the Schiff base type intermediate was faster
amine keeping other parameters constant (Fig. 5). In the (time = 5 min, selectivity = 83.8%) compared to the conversion
absence of hydrogen, the conversion of furfural was low at of imines to amines (time = 5 min, selectivity = 12.2%). Most
32%, but the introduction of hydrogen results in better per- likely in the initial stage, hydrogen coverage was not sufficient
formance. The displayed results confirmed that independent enough to activate the route of imine hydrogenation within
of the hydrogen pressure, complete conversion was achieved the reaction time of 5 min and in this situation the Schiff base
within the reaction time of 2 h at 80 °C. However, the selecti- type intermediate was converted to furfurylamine by hydroge-
vity of furfurylamine was influenced by the change in hydro- nolysis, which required a very low hydrogen content.33 As the
gen pressure. In the absence of hydrogen, only the Schiff base time progressed, the formation of the Schiff base type inter-
type intermediate was detected in the mixture (pre-equili- mediate dropped as imines were successfully hydrogenated to
bration time = 1 h + reaction time = 2 h). When hydrogen was amines. Moreover, it is very critical to explain the effect of
introduced into the system we observed a pressure dependent hydrogen pressure on the conversion of imines to amines as it
product distribution. With the change in hydrogen pressure involves coordination, oxidative addition of dihydrogen34,35
from 0.5 to 2 MPa, the selectivity of furfurylamine also requires further investigations. Thus, upon considering these
increased from 29.7 to 91.5% which reflects a substantial role results, Scheme 2 was proposed in which hydrogenation of
of hydrogen. Thus, the influence of hydrogen pressure on the imines to furfurylamine emerged as a major pathway and the
selectivity of furfurylamine predicted the formation of amines conversion of the Schiff base type intermediate to furfuryl-
via the imine pathway. Although we were unable to detect any amine and secondary amines as the minor one. As the pressure
imine intermediate, indirect evidence of the presence of
imines in the reaction mixture could be suggested from the
formation of the Schiff base type of intermediate; a conden-
Fig. 5 Effect of hydrogen pressure on the transformation of furfural to
furfurylamine. Reaction conditions: substrate = 0.2 g, catalyst = 0.002 g,
substrate/ammonia = 0.03, reaction time = 2 h and temperature =
80 °C.
Scheme 2 A possible reaction path of the formation of furfurylamine
under the studied reaction conditions.
Green Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Green Chemistry
Paper
was increased to more than 2 MPa, the selectivity of furfuryl-
Reductive amination of other aldehydes. Finally, we were
amine dropped substantially to 26% because of the formation interested to extend the applicability of this method to the
of tetrahydrofurfurylamine (ring hydrogenation), furfuryl other aldehydes especially to those that are related to biomass.
alcohol (hydrogenation of –CHO) and methylfurfural (hydroge- To explore the scope of this strategy, we investigated a series of
nolysis). Hence, to achieve the highest yield of 91.5%, an aldehydes under the studied reaction conditions and the
optimum hydrogen pressure of 2 MPa was required under the results are shown in Table 2. Most of the aldehydes were
studied reaction conditions.
smoothly converted to their corresponding primary amines.
Table 2 Reductive amination of different aldehydes
Entry
1
Substrate
Conv. (%)
100.0
Product (selectivity %)
—
—
2
100.0
100.0
100.0
3a
4b
—
—
—
5
6
90.0
—
100.0
7
8
100.0
100.0
—
—
9
42.1
10
100.0
—
—
—
Reaction conditions: catalyst = 0.002 g, substrate = 0.2 g, substrate/ammonia (mole ratio) = 0.03, temperature = 80 °C, reaction time = 2 h,
PH = 2 MPa. a Reaction time = 4 h. b Reaction time = 5 h.
2
This journal is © The Royal Society of Chemistry 2015
Green Chem.

View Article Online
Paper
Green Chemistry
For example, furfural containing methyl and dimethyl substi- amine was generated from the Schiff base type intermediate
tuents result in primary amines via reductive amination and via hydrogenolysis, whereas hydrogen pressure dependent
the yield varies from 60 to >90% (Table 2; entries 1 and 2). The selectivity of furfurylamine revealed that hydrogenation of
aldehyde group of 5-hydroxymethylfurfural was successfully imines to primary amines was the major route for the studied
converted to the corresponding amine with 86.3% selectivity synthesis strategy. The developed process was devoid of any
under the studied reaction conditions (Table 2; entry 3). Fur- additives, expensive reducing agents, additional solvents and
thermore, benzaldehyde with and without substitution also harsh reaction conditions. The obvious advantage of this pro-
delivered better performance in terms of the formation of tocol is to employ atom-economical and environmentally
primary amines. In the present strategy, benzaldehyde was friendly hydrogen as the reducing agent albeit with simplicity.
converted to benzylamine and dibenzylamine with the selecti- On the other hand, the use of ammonia as an amine source
vity of 77.1 and 22%, respectively (Table 2; entry 4). Benz- makes the process more challenging as mild reaction con-
aldehyde with electron donating groups was easily converted ditions were used. Further studies are required to exploit the
to its corresponding primary amine and depending on the current strategy for the reductive amination of different other
nature of the substituent, the yield was different. For instance, carbonyl compounds obtained from biomass, aliphatic alde-
p-anisaldehyde (4-methoxybenzaldehyde) furnished 86.6% hydes and dialdehydes. Based on the mechanistic consider-
selectivity, whereas 4-hydroxybenzaldehyde resulted in 98.2% ations deeper understanding of the reaction is necessary and
selectivity of the corresponding amine (Table 2; entries 5 and to improve the reusability of the catalyst as ammonia tolerance
6). As with the two electron withdrawing groups (–OMe and of the metal catalyst is a challenging issue.
–OH), vanillin was also transformed to the corresponding
amine with very high selectivity of 90.8% (Table 2; entry 7). On
the other hand, the scenario was different when 3- and 4-nitro-
benzaldehyde were used as substrates. In the presence of the
Experimental
Materials
electron withdrawing NO2 group, hydrogenation was the major
pathway of transformation (Table 2; entries 8 and 9). Among
the nitrobenzaldehydes, a steric effect was also prominent as
evidenced from the results. Furthermore, 3-phenylpropanal
performed extremely well in the conversion of aldehydes to
amines; complete conversion and 100% yield of the amine was
detected (Table 2; entry 10). Reductive amination of aliphatic
aldehyde using aqueous ammonia is also another interesting
example of the extension of the studied protocol, but it needs
further research.
Furfural (used as received) and 28% aqueous ammonia solu-
tion were obtained from Wako Pure Chemicals. Carbon
dioxide (>99.99%) was supplied by Nippon Sanso Co. Ltd.
Pd/Al2O3 Pt/C, Rh/C, Ru/C and Ru/Al2O3 all containing 5 wt%
of metal were purchased from Sigma-Aldrich. 5% Pd/C and 5%
Rh/Al2O3 were from Wako Pure Chemicals. Pd, Pt, Rh and
Ru/MCM-41 were synthesized in our lab.
Catalytic activity
Recycling of catalysts. As heterogeneous catalysts have been
claimed to offer the advantage of recycling, and to make a
process economically beneficial, we have separated the catalyst
from the product mixture simply by filtration for reuse. The
catalyst was highly reactive and continued its performance, but
after the fifth recycle the reactivity dropped (Table 1) (ESI
Fig. S2†). The catalyst also maintained its structural character-
istics as confirmed by the TEM images before and after the
reaction (ESI Fig. S3†). Hence, we expected that due to the
presence of an adsorbed species (e.g. amine or ammonia)
active sites become blocked after certain runs, deactivating the
catalyst, which is not confirmed yet.
The reductive amination of furfural was conducted in a stain-
less steel 50 ml batch reactor. In a typical experiment, 0.2 g of
substrate, 5 wt% Rh/Al2O3 (0.002 g) and 28% aqueous
ammonia solution were charged into the reactor placed in an
oven equipped with a magnetic stirrer and a temperature con-
troller. The reactor was subjected to temperature stabilization
for 1 h (pre-equilibration time) without any stirring. After
attaining the desired temperature of 80 °C, hydrogen was
introduced and pressurized to 2 MPa. The reaction mixture
was stirred continuously with a magnetic stirrer for a desired
period of time. After the reaction, the reactor was quenched in
ice-water and then depressurized slowly. The contents of the
reactor were filtered to separate the product from the catalyst
simply by filtration. The detailed analytical methods are given
in the ESI.†
Conclusion
In conclusion, we have demonstrated the selective synthesis of
furfurylamine from furfural in the presence of aqueous
ammonia solution and hydrogen. The developed method
offers very high yield and selectivity to primary amines under
Acknowledgements
an optimized reaction condition, where hydrogen pressure H. Kawanami is funded by the Japan–U.S. cooperation project
played a critical role in determining the formation of a Schiff for research and standardization of Clean Energy Techno-
base type intermediate and furfurylamine. Interestingly, from logies, The Ministry of Economy, Trade and Industry (METI),
the time profile of the reaction it can be suggested that furfuryl- Japan and CREST, Japan Science and Technology (JST).
Green Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Green Chemistry
Paper
(c) T. Bunlaksananusorn and F. Rampf, Synlett, 2005, 2682–
2684.
References
1 S. A. Lawrence, Amines: Synthesis, Properties and Appli- 12 (a) L. Rubio-Perez, F. J. Perez-Flores, P. Sharma, L. Velasco
cations, Cambridge University Press, Cambridge, 2004.
2 R. N. Salvatore, C. H. Yoon and K. W. Jung, Tetrahedron,
2001, 57, 7785–7811.
and A. Cabrera, Org. Lett., 2009, 11, 265–268;
(b) V. I. Tararov, R. Kadyrov, T. H. Riermeier and A. Borner,
Adv. Synth. Catal., 2002, 344, 200–208.
3 (a) R. O. Hutchins and M. K. Hutchins, Reduction of 13 S. Enthaler, ChemSusChem, 2010, 3, 1024–1029.
C-N to CHNH by metal hydrides, in Comprehensive Organic 14 J. L. Klinkenberg and J. F. Hartwig, Angew. Chem., Int. Ed.,
Synthesis, ed. B. N. Trost and I. Fleming, Pergamon Press,
New York, 1991, vol. 8; (b) V. A. Tarasevich and 15 T. Gross, A. M. Seayad, M. Ahmad and M. Beller, Org. Lett.,
N. G. Kozlov, Russ. Chem. Rev. (Engl. Transl.), 1999, 68, 2002, 4, 2055–2058.
55–72; (c) E. W. Baxter and A. B. Reitz, Org. React., 2002, 59, 16 (a) F. B. Siclari, P. P. Rossi, P. Garlasco and M. S. Gaetano,
2011, 50, 86–95.
1–714.
DT 2647317A1, 1977; (b) D. Kampmann, J. Weber and
C. Kniep, EP 0400426A3, 1991; (c) P. A. M. Grotenhuis,
FR 2656864A1, 1991; (d) C. Kos, E. Artner, E. Kloimstein,
F. Hebesberger, R. Haar and E. Lust, DE 4322065A1, 1995;
(e) F. Merger, C. U. Priester, T. Witzel, G. Koppenhoefer
and L. Schuster, DE 4010252A1, 1991; (f) J. J. Birtill,
M. Chamberlain, J. Hall, R. Wilson and I. Costello, Cata-
lysis of Organic Reactions, 17th Conference on Catalysis of
Organic Reactions, ed. F. E. Herkes, Dekker, New York,
1998, pp. 255–271; (g) R. E. Malz Jr., E. H. Jancis,
M. P. Reynolds and S. T. OLeary, Catalysis of Organic Reac-
tions, 15th Conference on Catalysis of Organic Reactions,
ed. M. G. Scaros and M. L. Prunier, Dekker, New York,
1994, pp. 263–271.
4 (a) A. F. Abdel-Magid and S. J. Mehrman, Org. Process Res.
Dev., 2006, 10, 971–1031; (b) T. C. Nugent and M. El-Shazly,
Adv. Synth. Catal., 2010, 352, 753–815; (c) C. Wang,
A. Pettman, J. Basca and J. Xiao, Angew. Chem., Int. Ed.,
2010, 49, 7548–7552; (d) C. Wang, C. Li, X. Wu and
J. Xiao, Angew. Chem., Int. Ed., 2009, 48, 6524–6526;
(e) T. W. J. Cooper, I. B. Campbell and S. J. F. MacDonald,
Angew. Chem., Int. Ed., 2010, 49, 8082–8091; (f) S. D. Roughley
and A. M. Jordan, J. Med. Chem., 2011, 54, 3451–
3479.
5 (a) R. Apodaca and W. Xiao, Org. Lett., 2001, 3, 1745–1748;
(b) R. Hiroi, N. Miyoshi and M. Wada, Chem. Lett., 2002,
31, 274–275; (c) B. C. Ranu, A. Majee and A. Sarkar,
J. Org. Chem., 1998, 63, 370–373; (d) S. Bhattacharyya, 17 M. Freifelder, Practical Catalytic Hydrogenation, Wiley,
A. Chatterjee and S. K. Dutta Chowdhury, J. Chem. Soc., New York, 1971, pp. 333–345.
Perkin Trans. 1, 1994, 689–693; (e) S. Sato, T. Sakamoto, 18 S. Pisiewicz, T. Stemmler, A.-E. Surkus, K. Junge and
E. Miyazawa and Y. Kikugawa, Tetrahedron, 2004, 60, 7899–
7906.
M. Beller, ChemCatChem, 2015, 7, 62–64.
19 (a) A. W. Heinen, J. A. Peters and H. van Bekkum, J. Org.
Chem., 2000, 65, 2501–2506; (b) M. Allegretti, V. Berdini,
M. C. Cesta, R. Curti, L. Nicolini and A. Topai, Tetrahedron
Lett., 2001, 42, 4257–4260.
20 R. P. Tripathi, S. S. Verma, J. Pandey and V. K. Tiwari, Curr.
Org. Chem., 2008, 12, 1093–1115.
6 (a) S. Bhattacharyya, J. Org. Chem., 1995, 60, 4928–4929;
(b) S. Raoufmoghaddam, Org. Biomol. Chem., 2014, 12,
7179–7193; (c) Q. Lei, Y. Wei, D. Talwar, C. Wang, D. Xue
and J. Xiao, Chem. – Eur. J., 2013, 19, 4021–4029.
7 (a) F. I. McGonagle, D. S. MacMillan, J. Murray,
H. F. Sneddon, C. Jamieson and A. J. B. Watson, Green 21 S. Gomez, J. A. Peters, J. C. van der Waal, W. Zhou and
Chem., 2013, 15, 1159–1165; (b) R. Cartarius, T. Krause and
H. Vogel, Chem. Ing. Tech., 2001, 73, 118–123.
8 (a) R. J. Mattson, K. M. Pham, D. J. Leuck and K. A. Cowen,
T. Maschmeyer, Catal. Lett., 2002, 84, 1–5.
22 J. Bodis, L. Lefferts, T. E. Muller, R. Pestman and
J. A. Lercher, Catal. Lett., 2005, 104, 23–28.
J. Org. Chem., 1990, 55, 2552–2554; (b) A. F. Abdel-Magid 23 M. P. Pandey and C. S. Kim, Chem. Eng. Technol., 2011, 34,
and S. J. Mehrman, Org. Process Res. Dev., 2006, 10, 971–
1031.
9 (a) R. Kadyrov and T. H. Riermeier, Angew. Chem., Int. Ed.,
2003, 42, 5472–5474; (b) M. Kitamura, D. Lee, S. Hayashi,
S. Tanaka and M. Yoshimura, J. Org. Chem., 2002, 67, 8685–
8687; (c) D. O’Connor, A. Lauria, S. P. Bondi and S. Saba,
29–41.
24 A. M. Sochaa, R. Parthasarathia, J. Shia, S. Pattathile,
D. Whyte, M. Bergerona, A. Georgea, K. Trana, V. Stavilad,
S. Venkatachalame, M. G. Hahne, B. A. Simmons and
S. Singh, Proc. Natl. Acad. Sci. U. S. A., 2014, 111, E3587–
E3595.
Tetrahedron Lett., 2011, 52, 129–132; (d) S. Ogo, K. Uehara, 25 (a) K. Adam and E. Haarer, FR Patent 1468354, 1967;
T. Abura and S. Fukuzumi, J. Am. Chem. Soc., 2004, 126,
3020–3021.
10 (a) O.-Y. Lee, K.-L. Law, C.-Y. Ho and D. Yang, J. Org. Chem.,
2008, 73, 8829–8837; (b) T. Mizuta, S. Sakaguchi and
Y. Ishii, J. Org. Chem., 2005, 70, 2195–2199.
11 (a) B. Sreedhar, P. S. Reddy and D. K. Devi, J. Org.
Chem., 2009, 74, 8806–8809; (b) C. Li, B. Villa-Marcos and
J. Xiao, J. Am. Chem. Soc., 2009, 131, 6967–6969;
(b) L. F. Kuntschik and O. W. Rigdon, US Patent 3,976,697,
1976; (c) I. D. Dobson, P. S. Williams and W. A. Lidy,
EP Patent 0284398, 1988; (d) R. L. Zimmerman and
P. S. E. Dai, EP Patent 0524717, 1993.
26 (a) T. Ayusawa, S. Mori, T. Aoki and R. Hamana, JP Patent
60146885, 1984; (b) S. R. Deshmukh, IN Patent 185230,
1999; (c) S. Mori, R. Hamana, T. Aoki and T. Ayusawa,
JP Patent 61134384, 1984.
This journal is © The Royal Society of Chemistry 2015
Green Chem.

View Article Online
Paper
Green Chemistry
27 J. J. Martinez, E. Nope, H. Rojas, M. H. Brijaldo, F. Passos 32 A. Chojecki, H. Jobic, A. Jentys, T. E. Müller and
and G. Romanelli, J. Mol. Catal. A: Chem., 2014, 392, 235–
240.
28 (a) M. Aguilar, F. J. Alfan and M. Albores Velasco, Rev. Soc.
Quim. Mex., 1996, 40, 81–87; (b) E. V. Kryukova and
A. P. Tomilov, Russ. J. Electrochem., 1997, 33, 453–463.
29 G. Mignonac, Compt. Rend., 1921, 172, 223.
J. A. Lercher, Catal. Lett., 2004, 7, 155–158.
33 (a) M. Chatterjee, A. Chatterjee, T. Ishizaka and
H. Kawanami, Catal. Sci. Technol., 2015, 5, 1532–1539;
(b) H. Bernas, A. Taskinen, J. Warna and D. Yu. Murzin,
J. Mol. Catal. A: Chem., 2009, 306, 33–39.
34 G. D. Liu, K. X. Huang, C. X. Cai, B. N. Cao, M. X. Chang,
W. J. Wu and X. M. Zhang, Chem. – Eur. J., 2011, 17, 14559–
14563.
30 (a) S. Nishimura, Handbook of Heterogeneous Catalytic
Hydrogenation for Organic Synthesis, John Wiley & Sons,
2001; (b) R. L. Augustine, Catal. Today, 1997, 37, 419– 35 (a) A. Fabrello, A. Bachelier, M. Urrutigoity and P. Kalck,
440.
Coord. Chem. Rev., 2010, 254, 273–287; (b) D. Crozet,
A. Gual, D. McKay, C. Dinoi, C. Godard, M. Urrutigoity,
J. C. Daran, L. Maron, C. Claver and P. Kalck, Chem. – Eur.
J., 2012, 18, 7128–7140.
31 (a) S. Gomez, J. A. Peters and T. Maschmeyer, Adv. Synth.
Catal., 2002, 344, 1037–1057; (b) E. J. Schwoegler and
H. Adkins, J. Am. Chem. Soc., 1939, 61, 3499–3502.
Green Chem.
This journal is © The Royal Society of Chemistry 2015