Ru/HZSM-5 as an efficient and recyclable catalyst for reductive amination of furfural to furfurylamine

Article abstract of DOI:10.1016/j.mcat.2019.110755

Furfurylamine converted from biomass-based platform molecules furfural was proven a significant intermediate in the synthesis of different valuable compounds. The combination of Ruthenium with HZSM-5 was acted as an excellent selective and reusable catalyst for the reduction amination of furfural with environmentally friendly ammonia and hydrogen. Incorporation of Ru species into HZSM-5 had a significant enhancement to the acid sites of Ru/HZSM-5. The Ru/HZSM-5(46) catalyst with optimized acid sites and interaction of the Ru-O-Al bond displayed an excellent catalytic performance, producing 76 % yield of furfurylamine at only 15 min, and could be recycled five times without loss of performance. Synergistic effect between RuO₂ and metallic Ru in the Ru/HZSM-5 catalyst facilitated the reduction amination of furfural.

Full text of DOI:10.1016/j.mcat.2019.110755

Molecular Catalysis 482 (2020) 110755
Contents lists available at ScienceDirect
Molecular Catalysis
journal homepage: www.elsevier.com/locate/mcat
Ru/HZSM-5 as an efficient and recyclable catalyst for reductive amination of
furfural to furfurylamine
a
a
a
a
a
a
Chenglong Dong , Hongtao Wang , Haochen Du , Jiebang Peng , Yang Cai , Shuai Guo ,
b
d
a,b,c,
Jianli Zhang , Chanatip Samart , Mingyue Ding
School of Power and Mechanical Engineering, Hubei International Scientific and Technological Cooperation Base of Sustainable Resource and Energy, Hubei Province Key
*
a
Laboratory of Accoutrement Technique in Fluid Machinery & Power Engineering, Wuhan University, Wuhan 430072, China
b
State Key Laboratory of High-efficiency Utilization of Coal and Green Chemical Engineering, Ningxia University, Yinchuan 750021, China
Shenzhen Research Institute of Wuhan University, Shenzhen 518108, China
Department of Chemistry, Faculty of Science and Technology, Thammasat University, Rangsit Campus, Klongluang, Pathumtani 12120, Thailand
c
d
A R T I C L E I N F O
A B S T R A C T
Keywords:
Furfural
Reduction amination
Furfurylamine
2
Interaction of RuO and Al
Furfurylamine converted from biomass-based platform molecules furfural was proven a significant intermediate
in the synthesis of different valuable compounds. The combination of Ruthenium with HZSM-5 was acted as an
excellent selective and reusable catalyst for the reduction amination of furfural with environmentally friendly
ammonia and hydrogen. Incorporation of Ru species into HZSM-5 had a significant enhancement to the acid sites
of Ru/HZSM-5. The Ru/HZSM-5(46) catalyst with optimized acid sites and interaction of the Ru-O-Al bond
displayed an excellent catalytic performance, producing 76 % yield of furfurylamine at only 15 min, and could
Synergistic effect
2
be recycled five times without loss of performance. Synergistic effect between RuO and metallic Ru in the Ru/
HZSM-5 catalyst facilitated the reduction amination of furfural.
1. Introduction
In recent years homogeneous and heterogeneous catalytic methods
for obtaining primary amines have been studied [11–17]. Senthamarai
N-containing compounds, especially for primary amines, are im-
2 3 3
et al. [18] used the RuCl (PPh ) catalyst to obtain primary benzylic
portant building blocks used widely for the preparation of surfactants,
dyes, polymers, and agrochemicals [1–4]. Compared to other alter-
native feedstock, there are broader implications when it comes to the
chemical synthesis of valuable N-containing compounds from biomass,
and aliphatic amines starting from carbonyl compounds. The complex
of homogeneous Ir and Rh was adopted to catalyze reductive amination
of aromatic ketones, but obtaining few primary amines [19]. In order to
improve the catalyst separation and reuse, heterogeneous catalysts have
been developed recently. Support modified Co-based catalysts were
prepared for the amination of 1,2-propanediol and the catalytic activity
was related to the catalyst with the special structure derived from the
preparation method [20]. Bódis et al. [21] reported carbon-based noble
which may be used as a promising strategy to decrease CO
2
emission
[
5–7]. Various processes for the synthesis of primary amines including
direct reaction of ammonia with alkyl halides, reductive amination of
carbonyl compounds or alcohols, hydrogenation of nitriles and hydro-
cyanation of alkenes followed by reduction have been developed [8].
Among them, reductive amination (RA) of Biomass-derived carbonyl
compounds with ammonia as a nitrogen source has received increasing
attention for producing primary amines, which is an effective route
compared to other synthetic processes using specific nitrogen sources
with low atomic efficiency. However, directly producing primary
amines are difficult because the initially generated amines are more
nucleophilic, resulting easily in the formation of by-products of sec-
ondary and tertiary amines [9,10]. Developing efficient catalysts for
reductive amination of carbonyl compounds remains a challenging
goal.
3
metals (Pt, Pd, Rh, and Ru) catalysts for RA of butyraldehyde with NH ,
and found that the Ru and Rh catalysts were in favor of the formation of
primary amine, while the Pt and Pd catalysts were active for the pro-
duction of secondary amine. The Ru-NP based catalyst was reported for
reductive amination of various carbonyl compounds, and the specific
flat-shaped morphology and surface electronic structure of Ru NPs were
contributed to the catalytic performance [22]. On the other hand, Yan
et al. [23] developed an efficient chemical means to produce amino
acids from renewable feedstock, in which the designed Ru/CNT catalyst
converted glucose into alanine in 43 % yield. Significantly, methanol is
identified as an effective solvent for reductive amination of carbonyl
⁎
Corresponding author.
E-mail address: dingmy@whu.edu.cn (M. Ding).
https://doi.org/10.1016/j.mcat.2019.110755
Received 4 October 2019; Received in revised form 2 December 2019; Accepted 25 December 2019
2468-8231/ © 2019 Elsevier B.V. All rights reserved.

C. Dong, et al.
Molecular Catalysis 482 (2020) 110755
[
24].
Furfurylamine (FAM) known as one of primary amines is a useful
measurements.
-TPR and NH
H
2
3
-TPD were carried out using a multifunctional dy-
compound in preparing drugs, fibers and pesticides. It can be synthe-
sized through reductive amination of its corresponding aldehyde or
alcohol over metal catalysts. Recently, the synthesis of furfurylamine
from furfuryl alcohol over RANEY® nickel have been reported [25].
Furfuryl alcohols are synthesized by hydrogenation of furfural, it’s ap-
parently direct reductive amination of furfural would be a more de-
sirable process. But furfural amination is still rarely explored due to
easy hydrogenation of eCHO groups and the ring opening reaction. It
has been well known that the activity and selectivity of RA of carbonyl
compounds over metal-supported catalysts are influenced obviously by
acid sites [26]. The results of Nakamura et al. [27] indicated that Lewis
acid sites on the Pt-based catalyst played a critical role in promoting the
formation of primary amine for the amination of ketones. Liang et al.
namic adsorption instrument equipped with a TCD detector (DAS-7000
type of Hunan Huasi Instrument Co., Ltd.). 0.05 g of the catalyst was
loaded in a quartz tube reactor, and heated in N
with a ramp of 10 °C/min. The sample was cooled down to 50 °C and
switched to 5 % H /N mixture for reduction. After the baseline was
2
flow at 350 °C for 1 h
2
2
stabilized, the temperature was raised to 800 °C at a ramp of 10 °C/min
and the consumption of hydrogen was recorded by TCD. CuO powder
was used to calibrate the detector signal. For NH
catalyst was loaded and heated in N flow at 350 °C for 2 h, after which
the sample was cooled down to 100 °C. The sample was exposed to a 5
% NH /N flow for 1 h, and then temperature-programmed desorption
3
-TPD, 0.1 g of the
2
3
2
was performed up to 800 °C, with a heating ramp of 8 °C/min.
XRD patterns were obtained using a Bruker D8 Advanced X-ray
diffractometer with Cu Kα radiation. X-ray photoelectron spectra (XPS)
was performed on a Thermo Scientific Escalab 250-X-ray photoelectron
spectrometer with Al Kα radiation for the X-ray source. The binding
energies of Ru 3p were calibrated by referencing to the C1s line at
284.6 eV.
[
28] reported reductive amination of biomass-derived aldehydes/ke-
tones, and found that RuO provided the strong Lewis acid sites for
2
facilitating the activation of carbonyl groups, while metallic Ru worked
as active phase to convert imine to primary amines. Therefore, control
of acid sites on metal-based catalysts leads to an efficient role for re-
ductive amination of carbonyl compounds. However, systematic re-
search on the relationship between catalyst acidity and RA performance
is still scarce. Zeolite possessed with both the Lewis acid sites and
Brönsted acid sites may be used to optimize the catalytic performance
of reductive amination, while zeolite modified metal-based catalysts for
reductive amination of carbonyl compounds are rarely reported [29].
In this work, HZSM-5 modified Ru-based catalysts with different
2.4. General procedure for reductive amination of furfural
The experiment of the catalysts for reductive amination of furfural
was carried out in a stainless-steel 50 ml autoclave equipped with a
magnetic stirrer, temperature and pressure control system. In all reac-
tions, the solvent is methanol. Typically, the substrate (2 mmol), cata-
lyst (100 mg) and ammonia solution (6 mL, Aladdin, 7 M in methanol)
were charged into the autoclave. The autoclave was closed, flushed
SiO
tion of furfural. The synergistic effect between RuO
well as tuning acid sites of HZSM-5 by changing the SiO
2
/Al
2
O
3
ratios were prepared and applied to the reductive amina-
and metallic Ru as
/Al ratios
2
2
2
O
3
2 2
with H five times to remove air, and then pure H gas was charged
were further investigated to reveal the structure-performance re-
lationship. In addition, the reaction conditions such as substrate/am-
monia ratio, reaction temperature, hydrogen pressure and catalyst
amount were studied in detail to optimize the catalytic performance.
until the pressure of 2.0 MPa. The autoclave was carried out at 80 °C
under stirring for 2 h. After the reaction, the autoclave was cooled down
in ice-water and then depressurized slowly. The suspension of autoclave
was filtered to separate the products. All liquid products were identified
by GC–MS. Conversion and yield of products were determined by GC
using n-heptane as an internal standard.
2. Experimental section
2.1. Materials
3. Results and discussion
HZSM-5(x) zeolites and S-1(silicalite-1 with MFI zeotype) zeolite
3.1. Catalytic performance on reductive amination of furfural
Different potential catalysts were studied for reductive amination of
were purchased from Nankai University Catalyst Co., Ltd (x represents
the ratio of SiO /Al ). Ruthenium trichloride hydrate (RuCl ·xH O)
35.0–42.0 wt%) was purchased from Aladdin Chemical. 5.0 wt% Ru/C
2
2
O
3
3
2
(
2
furfural with ammonia solution and H selected as a model reaction,
was purchased from TCI Chemical. Furfural (Sinopharm Chemical),
furfurylamine (Aladdin Chemical), furfuryl alcohol (Aladdin Chemical),
tetrahydrofurfurylamine (Aladdin Chemical), n-heptane (Sinopharm
Chemical) were purchased and used as received.
and the results were shown in Table 1. In order to obtain the high yield
of furfurylamine, different noble metal catalysts such as Ru, Pt, Pd and
Rh supported on HZSM-5 (46, SiO /Al O = 46) were carried out
2 2 3
firstly. During the reaction process, furfural is converted to furfur-
ylamine (2a) and several byproducts, such as furfuryl alcohol (2b),
Schiff base N-furfurylidenefurfurylamine (2c), tetrahydrofurfurylamine
2.2. Catalyst preparation
(
2d), difurfurylamine (2e) and 4,5-tris(2-furyl)imidazoline (2f). Sup-
Ru-supported zeolite catalysts (Ru loading = 5 wt%) were prepared
ported Ru, Pd, Pt and Rh catalysts are tested under similar conditions
and obtain 2a in good to moderate yields (44–76 %). Ru/HZSM-5(46)
exhibits the best catalytic performance of furfurylamine (76 % yield of
2a, entry 1) compared to the Pt, Pd and Rh based catalysts, demon-
strating that Ru metal is in favor for the formation of primary amine in
contrast to Pt, Pd and Rh metals. In addition, only 4 % of 2b (entry 4) is
detected at the Pt-based catalyst (entry 4), indicating that direct hy-
drogenation of carbonyl is not the main pathway. Hydrogenation of
furan ring is more likely to occur on metal Rh giving a moderate yield of
2d (34 %, entry 5). Hydrogenation of dimeric imine 2c to secondary
amine 2e is notable for the Pd-based catalyst (20 %, entry 3), suggesting
that the relative rate of hydrogenolysis and hydrogenation for 2c is
mainly depended on active metal species. Especially, compared to Ru/
HZSM-5(46), a commercial Ru/C catalyst presents only half the yield of
furfurylamine (40 %, entry 2), implying that the combination of Ru and
HZSM-5 displays an optimized catalytic performance on the reductive
via a wet impregnation method. Typically, the zeolite support was
mixed with the solution of RuCl , and the mixture was slowly evapo-
3
rated at 60 ℃ with stirring and dried at 80 °C for 12 h. Then, the as-
prepared solid was calcined at 400 °C for 4 h and was reduced at 300 °C
for 2 h under H atmosphere, passivated under 2 % O /N for 4 h at
2 2 2
room temperature. Other supported metal catalysts were also prepared
under the same conditions.
2
.3. Catalyst characterization
Scanning electron microscopy (SEM; TESCAN MIRA 3 LMH) was
used to examine the morphological characteristics of the catalysts.
Nitrogen adsorption isotherms were measured at −196 °C using an
ASAP 2010 apparatus (Micromeritics). All samples were degassed for
6
h at 200 °C in a nitrogen flow prior to the physisorption
2

C. Dong, et al.
Molecular Catalysis 482 (2020) 110755
Table 1
Reductive amination of furfural over supported metal catalysts.a
Entry
Catalysts
Con.(%)
Yield (%)
2b
2a
2c
2d
2e
2f
1
2
3
4
5
Ru/HZSM-5(46)
Ru/C
> 99
> 99
> 99
> 99
> 99
> 99
> 99
> 99
> 99
> 99
98
76
40
44
47
49
14
22
31
25
17
< 1
< 1
< 1
< 1
< 1
4
7
8
4
1
< 1
26
17
45
16
24
< 1
< 1
< 1
6
2
< 1
34
< 1
< 1
< 1
< 1
< 1
< 1
< 1
< 1
< 1
20
5
1
< 1
< 1
< 1
< 1
< 1
< 1
< 1
8
2
< 1
3
2
8
3
9
4
Pd/HZSM-5(46)
Pt/HZSM-5(46)
Rh/HZSM-5(46)
Ru/HZSM-5(27)
Ru/HZSM-5(36)
Ru/HZSM-5(46)
Ru/HZSM-5(81)
Ru/S-1
1
b
6
< 1
< 1
< 1
< 1
< 1
< 1
< 1
b
7
b
8
b
9
b
10
11
12
9
8
12
b
b
HZSM-5(46)
Blank
> 99
a
Reaction conditions: furfural (2 mmol), catalyst (0.15 g), 3 MPa H
2
, NH
3
(7 M in MeOH 6 mL), 100 °C,15 min.
/Al
b
Catalyst (0.1 g), 2 MPa H
2
, 80 °C, 2 h. The catalysts were labeled as Ru/HZSM-5(x) (x represent the SiO
2
2
O
3
ratio).
amination of furfural.
Besides, acid sites of HZSM-5 have an obvious effect on the RA
performance of furfural. From Table 1 it is seen that Ru/S-1 with no
acid sites displays only 17 % yield of 2a. As aluminum element is added
into the catalyst, the furfurylamine yield increases obviously. A de-
crease in the SiO
the maximum value is obtained as the SiO
As the SiO /Al ratio is below 46, the yield of 2a begins to decrease,
2
/Al
2
O
3
ratio of HZSM-5 improves the yield of 2a, and
2
/Al ratio decreases to 46.
2 3
O
2
2 3
O
indicating that the addition of excess acid sites on HZSM-5 is not ben-
eficial to the formation of furfurylamine (vide infra). In contrast, the
blank or single zeolite has no activity to produce furfurylamine (entry
11 and 12). This demonstrates that combining Ru with acidic HZSM-5
together plays a critical role in promoting the formation of furfur-
ylamine.
3.2. Catalytic characterization of the Ru/HZSM-5 catalysts
In order to reveal the structure-performance relationship of RA of
furfural, numerous techniques were conducted together to characterize
2
Fig. 1. H -TPR profile of the different catalysts.
2 2 3
microstructures of the Ru/HZSM-5 catalysts with different SiO /Al O
ratios.
HZSM-5 is strengthened, resulting in the formation of the Ru-O-Al
bond. Besides, the reduction peak of interface RuO shifts gradually
towards higher temperature with the decrease of SiO /Al ratio,
suggesting that the Ru-O-Al bond in the Ru/HZSM-5 catalyst is en-
hanced in the lower SiO /Al ratio. In addition, the reduction degree
of Ru species over all of the Ru/HZSM-5 catalysts calculated from H
TPR is 73.8–81.7 % (Table S1). In contrast, the Ru/S-1 catalyst displays
a lower reduction degree of 64.5 % for Ru species, which may be one of
the reasons for the poor catalytic performance of Ru/S-1.
The H
2
2
-TPR profile of the Ru/HZSM-5 catalysts with different SiO /
2
2
2 3
O
Al O ratios was shown in Fig. 1. There is one reduction peak for Ru/S-
2 3
1
5
and Ru/HZSM-5(81), while two reduction peaks for other Ru/HZSM-
catalysts, respectively. The lower reduction peak at about 170 °C is
2
2 3
O
2
-
ascribed to the reduction of surface RuO
peak at about 180 °C is attributed to the reduction of interface RuO
which RuO is probably combined with aluminum in HZSM-5, forming
a strong Ru-O-Al bond by sharing the common oxygen atoms [30,31].
Only surface RuO appears for the Ru/S-1 catalyst without the addition
of aluminum species. As few aluminum is added into the zeolite (HZSM-
(81)), no interface RuO is observed for the Ru/HZSM-5(81) catalyst,
which may be attributed to the weakened combination between RuO
and aluminum, restraining the formation of interface RuO . As the
SiO /Al ratio decreases continually to 46, the reduction peak of
interface RuO appears, implying that the interaction of RuO and Al in
2
, and the higher reduction
2
, in
2
Such interaction between Ru metal and HZSM-5 was also found in
2
NH
3
-TPD, as shown in Fig. 2. No peaks of acid sites is observed for S-1.
/Al ratios display two
All of the HZSM-5 zeolites with different SiO
2
2 3
O
5
2
peaks at about 280 and 460 ℃, which are attributed to the weak and
strong acid sites with framework Al, respectively [32]. The intensity of
both the weak and strong acid sites presents an increasing trend with
2
2
2
2 3
O
2 2 3
the decrease of SiO /Al O ratio from 81 to 27. The value of total
2
2
3

C. Dong, et al.
Molecular Catalysis 482 (2020) 110755
Fig. 2. (a) NH
3
-TPD profile of catalysts without Ru loading (b) NH
3
-TPD for catalysts after Ru loading.
Table 2
Amount and type of acid sites as determined by NH
3
-TPD.
Catalyst
Low Temperature Peaks (℃) Amount of weak acid sites
High Temperature Peaks (℃) Amount of strong acid sites
Total acidity (mmol/
g)
(
mmol/g)
(mmol/g)
1
2
3
4
5
6
7
8
9
1
HZSM-5(27)
HZSM-5(36)
HZSM-5(46)
HZSM-5(81)
S-1
Ru/HZSM-5(27) 258
Ru/HZSM-5(36) 266
Ru/HZSM-5(46) 243
Ru/HZSM-5(81) 256
280
274
288
291
–
0.092
0.077
0.079
0.012
–
0.089
0.065
0.077
0.029
–
465
444
483
540
–
382
395
382
384
–
0.219
0.207
0.170
0.030
–
0.335
0.292
0.182
0.005
–
0.311
0.284
0.249
0.042
–
0.424
0.357
0.259
0.034
–
0
Ru/S-1
–
acidity increases from 0.042 mmol/g of HZSM-5(81) to 0.311 mmol/g
of HZSM-5(27), as listed in Table 2. After the introduction of Ru species
into the zeolites, the peak of strong acid sites approximately reduces to
while Ru/S-1 displays the largest dispersion (29.6 %) and smallest
particle size (3.4 nm), further confirming the excellent dispersion
ability of S-1. Synchronously, Ru/S-1 has no acid sites, which displays
the poor catalytic performance in comparison to the Ru/HZSM-5 cat-
alysts with acid sites, suggesting that the dispersion of Ru nanoparticles
has a negligible role whereas acid sites of the Ru/HZSM-5 catalysts play
a critical role in promoting the reductive amination of furfural.
The micro-structures of Ru/HZSM-5(46) were further analyzed.
Fig. 3 showed the SEM micrograph and EDS mapping of the Ru/HZSM-
5(46) catalyst. It is apparent that a normal zeolite morphology is dis-
played for the HZSM-5 support [33], and Ru species are well dispersed
on the surface of HZSM-5. The microstructures of Ru/HZSM-5(46) were
characterized by XRD and XPS, respectively. From Fig. 4a it is found
that HZSM-5 is composed of the MFI structure, and adding Ru metal
into zeolite does not change the MFI structure of HZSM-5. Ru species
3
85 ℃ accompanied by a significant enhancement of intensity. The
enhancement of strong acid sites may be attributed to the interface
RuO forming by the strong interaction between Ru and aluminum,
2
which is more obvious as the aluminum content increases. Taking Ru/
HZSM-5(27) as a sample, the amount of strong acid sites increases by 53
%
after the loading of Ru into HZSM-5(27). Additionally, the total acid
sites content of Ru/HZSM-5(81) is 0.034 mmol/g, which increases
gradually to 0.424 mmol/g of Ru/HZSM-5(27) as the SiO /Al ratio
decreases from 81 to 27. The experiment results mentioned above in-
dicate that the yield of 2a increases firstly with the decreasing SiO
Al ratio, and then reaches the maximum value at the SiO /Al
ratio of 46. Further decrease in SiO /Al ratio results in the decrease
of the 2a yield. Meanwhile, the amount of acid sites and the Ru-O-Al
bond force present an increasing trend with the decrease in the SiO
Al ratio. Therefore, it is considered that the Ru/HZSM-5(46) catalyst
2
2 3
O
2
/
2
O
3
2
2 3
O
2
2 3
O
exists in the RuO
After reduction and reaction, new diffraction peaks of Ru metal appear
at 38.4° (100) and 44.0° (101), indicating that part of RuO species have
2
form on the surface of fresh Ru/HZSM-5(46) catalyst.
2
/
2
O
3
2
with modified acid sites and interaction of metal-support has the opti-
mized catalytic performance for RA of furfural.
The textural properties of the Ru/HZSM-5 catalysts were summar-
ized in Table 3. It is found that the Ru/HZSM-5 catalysts with different
been converted to metallic Ru during the reduction and reaction.
Especially, the MFI structure of HZSM-5 remains unchanged in this
process. The results of XPS (Fig. 4b) show that the diffraction peaks at
462.2 eV and 463.5 eV are observed on the surface of reduced Ru/
SiO
2
/Al
2
O
3
ratios show comparable surface areas of approximately
HZSM-5(46), which are ascribed to Ru (0) and RuO
2
, respectively,
2
273−315 m /g with micropores of approximately 0.37 nm and meso-
confirming that RuO species on the surface of HZSM-5 are converted
2
pores of 2.6–4.0 nm. In contrast, Ru/S-1 presents the maximum value of
BET surface area (315 m /g) and the minimum value of mesopores
partially to metallic Ru species during reduction. The reduction degree
of Ru species was calculated to be 77.3 % from XPS, which is consistent
2
(
2.2 nm), indicating that the S-1 support is more favorable to the dis-
with the value from the H
catalyst will not be further oxidized in the atmosphere. The results of
Liang et al. indicated that RuO provided the Lewis acid sites for pro-
moting the activation of carbonyl groups, whereas metallic Ru fa-
cilitated the hydrogenation of imine [28]. In the present study, both the
2
‐TPR (Table S1, 78.4 %) and ensured that the
persion of Ru species on the surface layers. Especially, the results of H
2
-
TPD shown in Table S2 indicate that the dispersion degrees of Ru
species are in the range of 8.7–24.7 %, and the average Ru nano-
particles sizes are between 4.1–11.7 nm for the Ru/HZSM-5 catalysts,
2
4

C. Dong, et al.
Molecular Catalysis 482 (2020) 110755
Table 3
Physicochemical properties of the Ru/HZSM-5 catalysts.
a
2
b
3
b
c
3
d
Entry
Catalyst
S
BET (m /g)
V
mic (cm /g)
D
mic (nm)
V
meso (cm /g)
D
meso (nm)
1
2
3
4
5
Ru/HZSM-5(27)
Ru/HZSM-5(36)
Ru/HZSM-5(46)
Ru/HZSM-5(81)
Ru/S-1
316
315
273
295
350
0.14
0.14
0.12
0.12
0.15
0.37
0.38
0.37
0.34
0.36
0.08
0.07
0.08
0.14
0.11
3.78
3.85
3.99
2.60
2.23
a
Determined by BET method.
Determined by Horvath-Kawazoe method.
b
c
Vmeso = Vtotal − Vmic.
d
Determined by BJH method from the desorption branch of the isotherm curves.
RuO
catalyst during reaction. It is possible that synergistic effect between
RuO and metallic Ru strengthens the reductive amination of furfural,
2
and metallic Ru species commonly exist in the Ru/HZSM-5(46)
promoting the formation of Schiff base type intermediates. Therefore,
tuning the ratio of substrate/ammonia to 0.05 achieves the maximum
yield of furfurylamine under the existed reaction conditions.
2
promoting the formation of furfurylamine compared to Pt, Pd and Rh
supported on the HZSM-5 catalysts.
The effect of reaction temperature on the performance of reductive
amination of furfural was investigated in the temperature range of
40–120 °C (Fig. 5b). A low temperature of 40 °C results in the formation
of Schiff base type intermediates composed mainly of 2c and 2f (∼52 %
of yield), whereas no formation of furfurylamine is observed. A slight
increase of temperature to 60 °C increases distinctly the yield of Schiff
base type intermediates. Further increasing reaction temperature re-
sults in the formation of furfurylamine, the amount of which reaches a
maximum of 49 % at 100 °C accompanied by the decrease in Schiff base
type intermediates. It is possible that the reductive amination of fur-
fural promotes firstly the formation of Schiff base type intermediates,
which is converted continually to furfurylamine via hydrogenolysis or
hydrogenation [34]. The temperature below 60 °C is insufficient for
producing furfurylamine. Further increase in the temperature above
3
.3. Optimization of the reaction conditions
Ru/HZSM-5(46) was proved as an efficient catalyst for producing
selectively furfurylamine in the reductive amination of furfural.
Subsequently, the effect of reaction conditions including substrate/
ammonia ratio, reaction temperature, hydrogen pressure and catalyst
amount were further investigated over Ru/HZSM-5(46). Fig. 5a showed
the effect of the substrate/ammonia ratio on product distribution. Ad-
justment of substrate/ammonia ratio may change the selectivity of
furfurylamine because reductive amination of the −CHO group could
be changed via adding ammonia. From Fig. 5a it can be seen that a
decrease in the substrate/ammonia ratio increases the yield of furfur-
ylamine and displays the maximum yield of 32 % as the substrate/
ammonia ratio reaches 0.05. A further increase in the ammonia content
inhibits the formation of furfurylamine, whereas increases the yield of
Schiff base N-furfurylidenefurfurylamine, indicating that excess am-
monia may facilitate the conversion of furfurylamine and furfural,
100 °C leads to a decrease in the yield of furfurylamine, which may be
attributed to the formation of unidentified products.
Fig. 5c showed the catalytic performance of Ru/HZSM-5(46) with
various hydrogen pressure. A low hydrogen pressure of 0.5 MPa results
in the formation of 16 % yield of 2a, and 40 % yield of Schiff base type
intermediates (2c and 2f). No formation of imine is observed in this
Fig. 3. SEM micrograph and EDS mapping of Ru/HZSM-5(46) catalyst showing the elemental distributions of Si, Al and Ru.
5

C. Dong, et al.
Molecular Catalysis 482 (2020) 110755
Fig. 4. (a) XRD pattern of the (1) ZSM-5(46), (2) calcined Ru/ZSM-5(46), (3) Ru/ZSM-5(46) after reduction, and (4) Ru/ZSM-5(46) after reaction (b) Ru 3p XPS
spectra of reduced Ru/HZSM-5(46).
process, which may be attributed to rapid conversion of imine to fur-
furylamine in the initial stage under H atmosphere [35]. With the
increase of H pressure, the yield of furfurylamine increases gradually,
accompanied by a decrease of the yield to 2c. The highest yield to 2a is
7 % when hydrogen pressure is 3 MPa and 3 % yield of tetra-
hydrofurfurylamine 2d is observed in this condition. Further increase in
the hydrogen pressure to 4 MPa does not change the yield of 2a, while
the yield of 2d increases to 5 %, suggesting higher hydrogen pressure is
conducive to the hydrogenation of furan ring.
The effect of the catalyst amount on the performance of reductive
amination of furfural was shown in Fig. 5d. The furfurylamine yield
increases gradually with an increasing Ru/HZSM-5(46) amount and
2
2
6
Fig. 5. (a) Effect of substrate/ammonia ratios on product distribution for the reductive amination of furfural. Reaction conditions: furfural (2 mmol), catalyst (0.1 g),
MPa H , 80 °C, 2 h. (b) Effect of temperature on product distribution for the reductive amination of furfural. (c) Effect of hydrogen pressure on product distribution
for the reductive amination of furfural. Reaction conditions: 100℃. (d) Effect of substrate/catalyst ratio on product distribution for the reductive amination of
furfural. Reaction conditions: 3 MPa H
2
2
2
.
6

C. Dong, et al.
Molecular Catalysis 482 (2020) 110755
Fig. 7. Reuse experiments of Ru/ZSM-5(46) catalyst for the reductive amina-
tion of furfural. Reaction conditions: furfural (2 mmol), catalyst (0.15 g), 3 MPa
H , NH (7 M in MeOH 6 mL), 100 °C, 15 min.
2 3
Fig. 6. Time courses for the reductive amination of furfural. Reaction condi-
tions: furfural (2 mmol), catalyst (0.15 g), 3 MPa H
00 ℃.
2 3
, NH (7 M in MeOH 6 mL),
1
1
2), imines are only converted to 2f, suggesting that the formation of 2f
reaches a maximum of 68 % at 0.15 g catalyst loading. Further increase
in the catalyst loading results in the decrease of the furfurylamine yield,
which implies that a moderate amount of catalyst is necessary to pro-
mote the reductive amination of furfural.
may not be a catalytic process. As Ru is combined with HZSM-5, imines
are reduced to 2a. Due to more nucleophilic of 2a than ammonia, 2a
can react preferentially with 1a to form Schiff base 2c. The Schiff base
may further reduce to 2e by hydrogenation or hydrogenolysis to form
2
a. The selectivity between 2a and 2e is depended on the relative rate
3
.4. Probable reaction pathways
of hydrogenolysis and hydrogenation. On the other hand, the over-
hydrogenation of furan ring on 2a causes easily the formation of 2d. Liu
et al. [25] reported the reductive amination of furfuryl alcohol 2b to
furfurylamine 2a, and dehydrogenation of 2b to 1a. Inspired by 2b
detected in our products, we believe that there may be a reversible
transformation between 1a and 2b in the reaction.
In order to explain the reaction pathway of the formation of fur-
furylamine, the distribution of furfural reductive amination products by
varying the reaction time using Ru/HZSM-5(46) was shown in Fig. 6.
All substrates are converted within the reaction time of 5 min, which
means the reaction between furfural and ammonia happens rapidly.
The highest yield of 2a is quickly achieved as the yield of 2c decreases
rapidly at the reaction time of 15 min, suggesting that 2c is the key
intermediate. There is a pathway for 2c converting to 2a, which is
consistent with the previous reports [10,35,36]. As the reaction time
continues, both the yields of 2a and 2c present a decreasing trend while
that of 2d increases slightly, implying that the hydrogenation of furan
ring takes place with time on stream. The yield of 2f increases gradually
with time on stream, and remains at 12 % (Table S3) after 15 min
without converting to other products.
3.5. Recycle study of the catalyst
The possibility of recovery and reusability of Ru/HZSM-5(46) were
studied through the reuse experiments. After the reaction, the mixture
is centrifuged and filtered to separate the catalyst. The catalyst is wa-
shed with water and ethanol for several times, dried at 60 °C for 10 h.
The Ru/HZSM-5(46) catalyst is used for five times and noticeable de-
crease in 2a yield is not observed (Fig. 7). Furthermore, no apparent
difference is found in XRD patterns (Fig. 4a), indicating that Ru/HZSM-
Based on the above results probable reaction pathways were shown
3
in Scheme 1. Initially, the reaction of 1a with NH results in the for-
mation of imines, which are unable to be detected because of the poor
stability of imines. In the blank or single HZSM-5 (Table 1, entry 11 and
5(46) is stable and used as a reusable catalyst for the reductive ami-
nation of furfural.
Scheme 1. A possible reaction path of the formation of furfurylamine under the studied reaction conditions.
7

C. Dong, et al.
Molecular Catalysis 482 (2020) 110755
4. Conclusions
References
In summary, we performed the reductive amination of furfural over
the Ru/HZSM-5 catalysts with different SiO /Al O ratios. The strong
2 2 3
interaction between ruthenium and aluminum increased the acid sites
[1] Y. Park, Y. Kim, S. Chang, Chem. Rev. 117 (2017) 9247–9301.
[
2] S. Bähn, S. Imm, L. Neubert, M. Zhang, H. Neumann, M. Beller, ChemCatChem 3
(2011) 1853–1864.
[
3] Y. Liu, A. Afanasenko, S. Elangovan, Z. Sun, K. Barta, ACS Sustain. Chem. Eng. 7
of Ru/HZSM-5, affecting obviously the product distribution. Both the
RuO and metallic Ru species were existed commonly for the Ru/
2
(2019) 11267–11274.
[
[
[
4] M. Pera-Titus, F. Shi, ChemSusChem 7 (2014) 720–722.
5] N. Yan, Y. Wang, Chemistry 5 (2019) 739–743.
6] W. Chen, H. Yang, Y. Chen, X. Chen, Y. Fang, H. Chen, J. Anal. Appl. Pyrolysis 120
(2016) 186–193.
[7] J. Zhang, N. Yan, ChemCatChem 9 (2017) 2790–2796.
8] T. Gross, A.M. Seayad, M. Ahmad, M. Beller, Org. Lett. 4 (2002) 2055–2058.
HZSM-5 catalyst, synergistic effect of which enhanced the reductive
amination of furfural compared to other noble metals such as Pt, Pd and
Rh. The Ru/HZSM-5(46) catalyst with modified acid sites and interac-
tion of the Ru-O-Al bond displayed the optimized catalytic performance
for producing furfurylamine. Under the optimized conditions, the
highest yield of 76 % to furfurylamine was achieved after the reaction
only 15 min and the catalyst could be recycled five times without loss of
performance. These findings suggest that Ru/HZSM-5 is a potentially
very attractive catalyst for the conversion of furfural to furfurylamine,
which displays a potential pathway for further industrial applications.
[
[
9] J. Gallardo-Donaire, M. Ernst, O. Trapp, T. Schaub, Adv. Synth. Catal. 358 (2016)
3
58–363.
[10] M. Chatterjee, T. Ishizaka, H. Kawanami, Green Chem. 18 (2016) 487–496.
11] F.G. Mutti, W. Kroutil, Adv. Synth. Catal. 354 (2012) 3409–3413.
12] B. Dong, X. Guo, B. Zhang, X. Chen, J. Guan, Y. Qi, S. Han, X. Mu, Catalysts 5 (2015)
[
[
2
258–2270.
[13] D. Deng, Y. Kita, K. Kamata, M. Hara, ACS Sustain. Chem. Eng. 7 (2019)
692–4698.
4
[
[
14] W. Guo, T. Tong, X. Liu, Y. Guo, Y. Wang, ChemCatChem 11 (2019) 4130–4138.
15] M.E. Domine, M.C. Hernández-Soto, M.T. Navarro, Y. Pérez, Catal. Today 172
(
2011) 13–20.
CRediT authorship contribution statement
[
16] P. Veeraraghavan Ramachandran, P.D. Gagare, K. Sakavuyi, P. Clark, Tetrahedron
Lett. 51 (2010) 3167–3169.
Chenglong Dong: Conceptualization, Methodology, Writing - ori-
ginal draft, Investigation. Hongtao Wang: Methodology. Haochen Du:
Investigation. Jiebang Peng: Investigation. Yang Cai: Investigation.
Shuai Guo: Data curation. Jianli Zhang: Resources. Chanatip Samart:
Resources. Mingyue Ding: Supervision, Writing - review & editing.
[17] J. Kim, H.J. Kim, S. Chang, Eur. J. Org. Chem. 2013 (2013) 3201–3213.
[
18] T. Senthamarai, K. Murugesan, J. Schneidewind, N.V. Kalevaru, W. Baumann,
H. Neumann, P.C.J. Kamer, M. Beller, R.V. Jagadeesh, Nat. Commun. 9 (2018)
4
123.
[19] T. Riermeier, K. Haack, U. Dingerdissen, A. Boerner, V. Tararov, R. Kadyrov,
Method for producing amines by homogeneously catalyzed reductive amination of
carbonyl compounds, US6884887B1, 2005.
[
20] C. Yue, K. Di, L. Gu, Z. Zhang, L. Ding, Mol. Catal. 477 (2019) 110539.
[
21] J. Bódis, L. Lefferts, T.E. Müller, R. Pestman, J.A. Lercher, Catal. Lett. 104 (2005)
Declaration of Competing Interest
2
3–28.
[
[
22] D. Chandra, Y. Inoue, M. Sasase, M. Kitano, A. Bhaumik, K. Kamata, H. Hosono,
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper.
M. Hara, Chem. Sci. 9 (2018) 5949–5956.
23] W. Deng, Y. Wang, S. Zhang, K.M. Gupta, M.J. Hülsey, H. Asakura, L. Liu, Y. Han,
E.M. Karp, G.T. Beckham, P.J. Dyson, J. Jiang, T. Tanaka, Y. Wang, N. Yan, Proc.
Natl. Acad. Sci. U. S. A. 115 (2018) 5093–5098.
[
[
24] S. Song, Y. Wang, N. Yan, Mol. Catal. 454 (2018) 87–93.
25] Y. Liu, K. Zhou, H. Shu, H. Liu, J. Lou, D. Guo, Z. Wei, X. Li, Catal. Sci. Technol. 7
Acknowledgment
(2017) 4129–4135.
[
[
[
[
[
26] K.V.R. Chary, K.K. Seela, D. Naresh, P. Ramakanth, Catal. Commun. 9 (2008)
The authors gratefully acknowledge the financial support from the
National Natural Science Foundation of China (51861145102,
1978225), Science and Technology program of Shenzhen
JCYJ20180302153928437), Foundation of State Key Laboratory of
High-efficiency Utilization of Coal and Green Chemical Engineering
2019-KF-06) and Fundamental Research Fund for the Central
75–81.
27] Y. Nakamura, K. Kon, A.S. Touchy, K. Shimizu, W. Ueda, ChemCatChem 7 (2015)
9
21–924.
2
(
28] G. Liang, A. Wang, L. Li, G. Xu, N. Yan, T. Zhang, Angew. Chem. Int. Ed. 56 (2017)
3050–3054.
29] S.R. Kirumakki, M. Papadaki, K.V.R. Chary, N. Nagaraju, J. Mol. Catal. A: Chem.
3
21 (2010) 15–21.
(
30] D. Li, N. Ichikuni, S. Shimazu, T. Uematsu, Appl. Catal. A 172 (1998) 351–358.
Universities (2042019kf0221).
[31] K. Asakura, Y. Iwasawa, J. Chem. Soc. Faraday Trans. 86 (1990) 2657.
32] G.L. Woolery, G.H. Kuehl, H.C. Timken, A.W. Chester, J.C. Vartuli, Zeolites 19
1997) 288–296.
[33] K. Wang, X. Wang, Microporous Mesoporous Mater. 112 (2008) 187–192.
[
(
Appendix A. Supplementary data
[
[
34] R.L. Augustine, Catal. Today 37 (1997) 419–440.
35] T. Komanoya, T. Kinemura, Y. Kita, K. Kamata, M. Hara, J. Am. Chem. Soc. 139
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.mcat.2019.110755.
(2017) 11493–11499.
[36] S. Nishimura, K. Mizuhori, K. Ebitani, Res. Chem. Intermed. 42 (2016) 19–30.
8