Morphology-Tuned Activity of Ru/Nb2O5 Catalysts for Ketone Reductive Amination

Article abstract of DOI:10.1002/cctc.201900335

Amines are important compounds in natural products and medicines. Specifically, cyclopentylamine is one of the value-added chemicals widely used in the production of pesticides, cosmetics and medicines. In this work, three Ru/Nb₂O₅ catalysts with different Nb₂O₅ morphologies were used in the reductive amination of cyclopentanone under mild reaction conditions (90 °C, 2 MPa H₂), among which 1 %Ru/Nb₂O₅?L catalyst exhibits best performance with the yield of cyclopentylamine reaching 84 %. This catalytic system is stable and has not significant deactivation even after 5 runs in the durability test. In addition, this catalyst can be extended to a series of aldehydes/ketones. Further comprehensive characterizations (XPS analysis and CO-adsorption DRIFT) reveal that the electronic effect of Ru species can be ruled out; instead, the activity of the catalyst is strongly influenced by the geometric effect. Layered Nb₂O₅ material possesses the highest surface area, resulting in the highest Ru dispersion, and therefore shows the highest catalytic activity. The in-situ DRIFT-MS technique was also used to reveal and understand the reaction mechanism. It is found that Ru species play a key role in activating carbonyl groups. This study illustrates a promising application of Ru/Nb₂O₅-Layer catalyst in the synthesis of amine and provides an understanding to the reaction mechanism.

Full text of DOI:10.1002/cctc.201900335

Accepted Article
Title: Morphology-tuned activity of Ru/Nb2O5 catalysts for ketones
reductive amination
Authors: Wanjun Guo, Tao Tong, Xiaohui Liu, Yong Guo, and Yanqin
Wang
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.201900335
Link to VoR: http://dx.doi.org/10.1002/cctc.201900335
A Journal of
www.chemcatchem.org

10.1002/cctc.201900335
ChemCatChem
FULL PAPER
Morphology-tuned activity of Ru/Nb2O5 catalysts for ketones reductive
amination
⊥a
⊥a
Wanjun Guo , Tao Tong , Xiaohui Liua, Yong Guoa, Yanqin Wang*a
Abstract: Amines are important compounds in natural products and
medicines. Specifically, cyclopentylamine is one of the value-added
chemicals widely used in the production of pestisides, cosmetics and
medicines. In this work, three Ru/Nb2O5 catalysts with different
Nb2O5 morphologies were used in the reductive amination of
cyclopentanone under mild reaction conditions (90 oC, 2 MPa H2),
among which 1%Ru/Nb2O5-L catalyst exhibits best performance with
the yield of cyclopentylamine reaching 84%. This catalytic system is
stable and has not significant deactivation even after 5 runs in the
durability test. In addition, this catalyst can be extended to a series
of aldehydes/ketones. Further comprehensive characterizations
(XPS analysis and CO-adsorption DRIFT) reveal that the electronic
effect of Ru species can be ruled out; instead, the activity of the
catalyst is strongly influenced by the geometric effect. Layered
Nb2O5 material possesses the highest surface area, resulting in the
highest Ru dispersion, and therefore shows the highest catalytic
activity. The in-situ DRIFT-MS technique was also used to reveal
and understand the reaction mechanism. It is found that Ru species
play a key role in activating carbonyl groups. This study illustrates a
promising application of Ru/Nb2O5-Layer catalyst in the synthesis of
amine and provides an understanding to the reaction mechanism.
from the substrates.[7] As water is the only by-product during the
reaction, this process is usually considered as a green process,
but high reaction temperature (180~220 oC) is essential to
achieve a good performance when heterogeneous catalysts are
used in this strategy. The most practical and convenient
approach for the synthesis of primary amines under mild
conditions is the catalytic reductive amination of carbonyl
compounds with ammonia and hydrogen. Up to now, many
efforts have been made to the development of homogeneous
catalysts for the selective preparation of primary amines by
reductive amination.[8] Meanwhile, increasing attention has been
attracted to the exploration of novel and high active
heterogeneous catalysts in this area, as these kinds of catalysts
9]
show great commercial potential from an industrial view.[8a,
Very recently, Hara reported a high selective reductive amination
of various aromatic carbonyl compounds over Ru/Nb2O5
catalysts and attributed the excellent performance to the weak
electron-donating Ru on Nb2O5 surfaces.[10] They also suggested
that the morphology of Ru species and the properties of Nb-
based supports can strongly affect the reductive amination of
furfural in their latest reports.[11]Therefore, it is believed that
Ru/Nb2O5 catalytic system could be promising for the selective
synthesis of primary amines from carbonyl compounds.
Cyclopentylamine (CPA) is an important fine chemical widely
used as the ingredient in the manufacture of cosmetics,
pesticides and medicines. Benefitting from the rapid
development of biorefinery, the new route of CPA synthesis from
Introduction
Amines are important compounds in natural products and
medicines, and are irreplaceable intermediates in the synthesis
of fine chemicals, including dyes, pesticide and herbicide.[1]
Generally, there are three typical amination approaches:
amination of halides, [2]hydrogen-borrowing amination of alcohol
or phenol, [3]and the reductive amination of ketone or aldehyde.
biomass-derived
cyclopentanone
becomes
promising.
Cyclopentanone (CPO) is an important biomass-based platform
compound and can be produced from furfural.[12] During the
reductive amination of CPO, some side-reactions, such as over-
[9c]
[10]
hydrogenation,
self-coupling of ketones and amines,
and
[4]
The reaction from halides to amines has been well
[12]
Aldol condensation,
would be accompanied, so the high
established.[1b] However, in this strategy some strict conditions
selective production of cyclopentylamine is a big challenge.
are often
required and the formation of the toxic and
environmentally unfriendly by-products is also inevitable.[5] To
overcome the drawbacks in this route, increasing attentions
have been paid to the so-called ‘hydrogen-borrowing
amination’[6] and the reductive amination processes.[4] In the
hydrogen-borrowing amination process, alcohols are used as
the substrates and hydrogen resource as well. Typically,
alcohols will directly react with ammonia after dehydrogenation,
and the intermediates are then hydrogenated by the H2 formed
[a]
W. J. Guo, T. Tao, Y. Guo, X.H. Liu, Prof. Y.Q. Wang
Shanghai Key Laboratory of Functional Materials Chemistry and
Research Institute of Industrial Catalysis
School of Chemistry and Molecular Engineering, East China
University of Science of Technology
No. 130 Meilong Road, Shanghai, 200237 (P.R. China)
E-mail: wangyanqin@ecust.edu.cn
⊥
These two authors contribute equally.
Scheme 1. Amination of cyclopentanone and accompanied by-reactions.
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900335
ChemCatChem
FULL PAPER
Herein, Ru species supported on three kinds of niobium oxide
(Nb2O5) with different morphology was studied for the reductive
amination of cyclopentanone to cyclopentanylamine. The
comprehensive
structure-properties
relationship
was
investigated through XPS, CO-titration and DRIFT-MS
techniques. It is found that layered niobium oxide (Nb2O5-L)
supported Ru catalyst has the best performance and it’s the size
of Ru, not the electronic properties influenced the activity.
Moreover, the Ru/Nb2O5-L catalytic system shows a good
durability and can be expanded to a series of aldehyde/ketone
compounds, which suggest this system has a bright potential in
industrial application.
Results and Discussion
(001)
(110)
(101)
(110)
(102)
(002)
(c)
(b)
(a)
(001)
(002)
10
20
30
40
50
2 ()
60
70
80
Figure 2. TEM images of (a) 1%Ru/Nb2O5-L, (b)1%Ru/ Nb2O5-H, and (c)
1%Ru/Nb2O5-F.
Figure 1. XRD patterns of Ru/Nb2O5 samples (a) 1%Ru/Nb2O5-L, (b)
1%Ru/Nb2O5-H, (c) 1%Ru/Nb2O5-F.
116.6 o
C
The XRD patterns present in Figure 1 show that both
Ru/Nb2O5-F and Ru/Nb2O5-H catalysts have the characters of T-
phase niobium oxide (JCPDS card no. 28-0317). The Ru/Nb2O5-
L shows only two diffraction peaks at 22.7° and 46.2° obviously,
indicating its layered structure and low degree of crystallinity.[13]
It is need mentioned that no Ru diffraction peaks observed in all
three catalyst, probably due to the low Ru loading (1
wt.%) or the good dispersion of Ru species over the niobium-
based supports.
92.7 o
C
(a)
122.2 o
C
139.3 o
C
(b)
(c)
117.5 o
C
The morphologies of these three Ru/Nb2O5 samples were
observed by TEM images and found that 1%Ru/Nb2O5-L has the
rod-shaped morphology (Figure 2a), the 1%Ru/Nb2O5-H catalyst
has spherical morphology, which may be accumulated by small
Nb2O5 nanoparticles (Figure 2b) and the 1%Ru/Nb2O5-F shows
a nanoslice-like morphology and assembled to nanoflower-like
morphology when we observed in a large scale (under FE-
SEM), which is in line with the previous report (Figure S1).[14]
However, it is hard to destinguish the Ru particles in TEM
images, due to the extremely small Ru particle size, which is in
agreement with XRD results.
The Nitrogen sorption was carried out to measure the surface
areas of all Ru/Nb2O5 catalysts, the data were summarized in
Table 1. It is found that Ru/Nb2O5-L had the highest surface area
(203 m2g-1), followed with Ru/ Nb2O5-H (42 m2g-1) and
Ru/Nb2O5-F had the lowest (16 m2g-1).
50
100
150
200
250
Temperature(C)
Figure 3. H2-TPR profile of Ru/Nb2O5 catalysts (a) 1%Ru/Nb2O5-L, (b)
1%Ru/Nb2O5-H, (c) 1%Ru/Nb2O5-F.
NH3-TPD analyses were also carried out to determine the
surface acidity over all Ru loaded and pristine Nb2O5 samples
(Figure.S5 and S6). Among all three Nb2O5 supports, Nb2O5-L
possesses the largest acidity and Nb2O5-H has the lowest. After
loading with Ru, the total acid amount changed, but the trend of
acidity remains the same as that over pristine Nb2O5 samples.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900335
ChemCatChem
FULL PAPER
Table 1. Yield of each product of CPO conversion over various catalysts[a]
Ru
d[b]
Yield (%)
TOF[d]
SBET
D[c]
(%)
Conv.
(%)
content
Catalysts
(min-1)
others
(m2g-1)
(nm)
1
2
3
4
5
(wt.%)
58
1%Ru/Nb2O5-L
1% Ru/Nb2O5-H
1% Ru/Nb2O5-F
203
42
0.98
0.97
0.96
1.8
2.3
8.4
76
58
16
100
94.6
73.5
84.3
23.2
0
0
0
4.1
5.0
0
1.0
9.1
2.4
2.0
1.0
48.6
5.9
56
57
16
3.3
35.7
16.7
[a] Condition: 2 mmol CPO,5 ml NH3-MeOH(2M),20 mg catalyst, 90 oC，2 MPa H2, 4 h; [b] Ru size(d) calculated by the dispersion of Ru ; [11] [c] Ru dispersion (D)
determined by CO chemisorption; [d] The TOF value is calculated at ca. 20% conversion
three 1%Ru/Nb2O5 catalysts, we then turned to compare their
catalytic performance in the reductive amination of CPO, and the
results are summarized in Table 1. It is obvious that
100
1%Ru/Nb2O5-L shows the best activity under the investigated
conditions (90 oC，2 MPa H2, 4 h) and the CPA yield reaches
80
84% with CPO completely converted. While, both 1%Ru/Nb2O5-
Conv.
H and 1%Ru/Nb2O5-F catalysts cannot convert all CPO within 4
hours, and a large amount of secondary imine (4, intermediate)
remained in the catalytic system. What’s worse is that no
targeted product (CPA) was obtained over 1%Ru/Nb2O5-F during
the test.
Reaction time courses were collected to investigate the
reaction pathways (Figure 4) over 1%Ru/Nb2O5-L catalyst. It is
found that imine (2) is generated at the very beginning of the
reaction, and then rapidly hydrogenated to CPA. However, due
60
40
20
0
1
2
3
4
5
others
0
1
2
3
4
to the higher nucleophilicity of CPA (compared with that of
[15]
molecular ammonia),
part of CPA can react with CPO and
Time (h)
forms the intermediate 4. When the reaction time was further
prolonged, the amount of the secondary imine (4) decreased
gradually with the yield of CPA increased. This suggests that
intermediate 4 can be ammonolyzed to CPA under the co-
existence of H2 and NH3. However, there is competitive
hydrogenation of 4 to the secondary amine (3) and this process
is irreversible. Therefore, the secondary amine (3) is the major
by-product. In addition, a small amount of cyclopentanol was
also detected, which comes from the direct hydrogenation of
CPO. Besides these reactions, the aldol condensation of CPO
also happens during this process, and the structures of these
by-products (Figure 4, others) are verified by MS spectra (Figure
S13-14). From above analysis, the whole reaction network can
be clearly identified, which is summarized in Scheme 2.
Figure 4．Effect of reaction time on the yield of products of cyclopentanone
conversion
H2 temperature-programmed reduction analysis (H2-TPR) was
introduced to compare the interaction between Ru species and
Nb2O5 supports (Figure 3). All the samples show broad peaks in
the range of 75-175 oC and Ru/Nb2O5-L, Ru/Nb2O5-H and
Ru/Nb2O5-F have a strong reduction peak at 116 °C, 122 oC, and
117 oC, respectively. In addition, a peak at 139 oC is observed in
the case of 1%Ru/Nb2O5-H, which may be caused by the
stronger interaction between Ru species and Nb2O5-H support.
Interestingly, a weak reduction peak appears at 92.7 oC over
1%Ru/Nb2O5-L, which hints that Ru species may have a better
dispersion on Nb2O5-L, this can be further confirmed by CO-
titration results (Table 1) and will be discussed later.
The effect of reaction temperature and initial H2 pressure was
also investigated to find out the optimum conditions for this
reaction, and the results are shown in Figure 5. Increasing either
reaction temperature or H2 pressure can accelerate the reaction
rate. Clearly, the CPA yield rises steadily in the investigated
.
After gaining the morphology and surface properties of these
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900335
ChemCatChem
FULL PAPER
period, while the yield of intermediate 4 continuously declines
with the rising of reaction temperature or initial H2 pressure. To
balance the cost and productivity, the optimized operation
conditions can be set at 90 oC and 2 MPa H2 pressure.
effect in the reaction. The H2-TPR results show that the Ru
dispersion over Nb2O5-L support are much higher than those of
the rest two, so CO-pulse chemisorption was carried out to
determine the dispersion of Ru particles and the data were
added in Table 1.
Conv.
A
100
80
60
40
20
0
1
3
4
otehrs
120
60
90
45
Temperature(C)
Conv.
1
B
100
80
60
40
20
0
3
4
otehrs
Scheme 2. Reaction pathways according to the time courses.
Recently, the chemical states of supported metals have been
found to play an important role on the catalytic activity and
15]
selectivity in a wide range of reactions.[10,
Considering the
previous report that the suitable chemical state of Ru species
over Nb-based materials may have the promotional effect in
reductive amination, [10] we measured the Ru states first in these
three catalysts.
Pressur2e (MPa)
4
1
The XPS analyses of Ru 3d5/2 orbit of all three reduced
1%Ru/Nb2O5 catalysts were conducted (Figure 6). Ru species
can be divided into two categories in all three catalysts, which
can be associated to Ru0 species at ca. 280.4 eV and Ruδ+
species at around 281.0 eV, respectively.[16] It can be found that
the differences of Ru states over these three catalysts were not
obvious.
Furthermore, the CO-adsorption DRIFT spectra were
recorded to confirm the chemical states of Ru again over the
three catalysts (Figure 7). Three bands can be clearly
distinguished in all three spectra: the first band at the centre of
2014–2017 cm-1 can be assigned to the linearly adsorbed CO on
Ru0 species [Ru0-CO]; the peak of the linearly adsorbed CO on
partially oxidized Ruδ+ appears at 2077-2080 cm-1 [Run+-CO];
and a tiny band centred at 2133-2141 cm-1 , which is usually
considered to be the signal of tricarbonyl species on partially
oxidized Ru [Run+-(CO)3].[16] Again, we didnot find significant
difference of the location of CO over Ru species in all three
catalysts, in consistence with XPS results.
Figure 5. Effect of temperature and pressure on the conversion of CPO、
Condition: (A) 2 mmol CPO,5 ml NH3-MeOH (2 M),20mg 1%Ru/Nb2O5, 4 h, 2
MPa H2. (B) 2 mmol CPO, 5 ml NH3-MeOH (2 M), 20 mg 1% Ru/Nb2O5, 90 oC,
4 h.
Ru 280.9 eV
Ru0 280.3 eV
(a)
Ru 281.0 eV
Ru0 280.5 eV
(b)
(c)
Ru 281.1 eV
Ru0 280.4 eV
Based on the results and discussions above, we can now rule
out the possibility that the differences of catalytic performances
for CPO reductive amination over the three catalysts are caused
by the electronic effect of Ru species.
283
282
281
280
279
278
Binding Energy (eV)
Since the electronic states of Ru species are identical in all
these Ru/Nb2O5 catalysts, we turn to consider the geometric
Figure 6. XPS spectra of Ru 3d5/2 for (a) 1%Ru/Nb2O5-L, (b) 1%Ru/Nb2O5-H,
(c) 1%Ru/Nb2O5-F.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900335
ChemCatChem
FULL PAPER
1592 cm-1
1558 cm-1
1386cm-1
1299cm-1
2080 cm-1
1635cm-1
1169 cm-1
1095 cm-1
2017 cm-1
2141 cm-1
2134 cm-1
2133 cm-1
CPO
(a)
a
1457 cm-1
1589 cm-1
1442 cm-1
2078 cm-1
1635 cm-1
1346 cm-1
1236 cm-1
1675 cm-1
930 cm-1
966 cm-1
2012 cm-1
CPO+NH3
b
(b)
(c)
2077 cm-1
CPO+NH3+H2
2014 cm-1
2000
c
2200
2100
1900
1800
2200
2000
1800
1600
1400
1200
1000
800
Wavenumbers (cm-1)
Wavenumbers(cm-1)
Figure 7. CO-adsorption DRIFT spectra at 25 oC over (a) 1%Ru/Nb2O5-L, (b)
Figure 9. In situ DRIFT spectra for adsorption of (a) CPO, (b) CPO and NH3,
1%Ru/Nb2O5-H, (c) 1%Ru/Nb2O5-F.
(c) CPO, NH3 and H2 on Ru/Nb2O5-L
Obviously, Ru dispersion increases with the increase of the
support’s surface areas, and 1%Ru/Nb2O5-L possesses the
largest dispersion of 76%. This result points out that under the
premise of the identical Ru chemical states, a higher Ru
dispersion is beneficial to the CPO reductive amination process.
Moreover, the TOF values were calculated at lower conversion
(ca. 20%) and added in Table1. It is found that the three data
are similar, further confirming the key role of Ru partlcle size.
In addition, when we further analyzed the reaction result over
1%Ru/Nb2O5-F, it can be seen that besides its low activity, a
large amount of condensed by-products is formed. This is
basically caused by the low hydrogenation activity of
1%Ru/Nb2O5-F, whose dispersion is the lowest among all three
catalysts. Namely, the competative side-reactions (mainly the
Aldol condensation reaction) dominate the reaction, which are
taking place over the acidic sites. This result again illustrates the
importance of the high Ru dispersion in CPO reductive
amination process.
To investigate the mechanism of the reaction and identify the
active sites in the Ru-Nb systems, we conducted a series of
DRIFT-MS experiments. The CPO-adsorption DRIFT spectra
(Figure 8) over Nb2O5-L and 1%Ru/Nb2O5-L were collected,
respectively. It can be seen only one broad band at ca. 1635 cm-
1
appeared in the spectrum of Nb2O5-L, which can be assigned
to the coordination of carbonyl group of CPO with L-acid sites.[17]
In the case of 1%Ru/Nb2O5-L, the broad peak at around 1635
cm-1 still exists, suggesting that CPO is indeed adsorbed on the
support. However, two new peaks at 1592 cm-1 and 1558 cm-1
appear in the spectrum of 1%Ru/Nb2O5-L. The former one can
be assigned to the carbonyl group in CPO adsorbed on the
surface of Ru particles, and the latter one can be considered as
the carbonyl group of the aldol condensed by-products (Figure
[17]
5a, others) coordinated to L-acid sites.
The wavenumber of
C=O groups shows an approximately 40 cm-1 red-shift over Ru
species than that over the pristine Nb2O5 support, demonstrating
that Ru species show a stronger ability to activate the C=O
group. Thus, besides the function of H2 activation, the Ru
species can also act as active sites for the activation of carbonyl
compounds during the reaction.
Ru colloid (1.2 nm) was also prepared and loaded on various
Nb2O5 to further clarify the role of Ru and support. It is found that
Ru-colloid loaded Nb2O5 catalysts exhibit a slight difference in
the catalytic performances (Table S2), compared to the
remarkable differences over Ru-impregnated catalysts. This
again highlights the decisive role of high Ru dispersion in CPO
reductive amination.
From the view of mechanism, the lone paired electrons in NH3
molecule firstly undergoes nucleophilic attack with the C=O
group of CPO to get the imine 2 during the CPO reductive
[18]
amination reaction,
and then the intermediate 2 is further
hydrogenated to the CPA. Since the C=O bond can be activated
over the Ru/Nb2O5 catalyst, the behaviour of NH3 over the
catalysts should also be concerned. As the activated C=O bond
can directly react with NH3 through nucleophilic attack, the
reaction may undergo via an Eley-Rideal (E-R) mechanism. We
then used the in-situ DRIFT-MS technique to follow the surface
reactions and explore the surface reaction mode during the CPO
reductive amination over 1%Ru/Nb2O5-L catalyst.
1635 cm-1
(a)
1592 cm-1
1558 cm-1
1635 cm-1
After the adsorption of CPO (Figure 9 a), 30 mL NH3 was
pulsed into the chamber (Figure 9 b), and then changed to 5%
H2/Ar after 5-minute sweeping in Ar (Figure 9 c). The spectra of
the latter two steps are of high similarity, and both show the
typical characters of CPA and the reductive amination products
of Aldol condensed by-product (Figure S4b and Table S1). This
phenomenon seems so strange that since no molecular H2 was
introduced in the second step of our DRIFT-MS experiment,
(b)
1900
1800
1700
1600
1500
Wavenumber (cm-1)
Figure 8. CPO-adsorption DRIFT spectra over (a) Nb2O5-L, (b) 1%Ru/Nb2O5-
L.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900335
ChemCatChem
FULL PAPER
Table 2 Conversion of different aldehydes and ketones over the 1%Ru/Nb2O5-
composition nor Ru content of the spent catalyst (0.9 wt.%)
show remarkable changes, compared with the fresh one. Then,
the 1%Ru/Nb2O5-L catalyst was further applied to a wide range
of aldehydes and ketones, including long-chain ketones, cyclic
ketone and aromatic aldehyde (Table 2). The results show that
this system is suitable to all investigated substrates and a high
yield of primary amines can be achieved.The good stability of
1%Ru/Nb2O5-L and its excellent compatibility in the conversion
of different substrates highlight this system’s bright future in the
amine-production.
L catalyst
Time
(h)
Yield
(%)
Entry
1
Substrate
Product
6
6
78
76
2
3
4
5
90
78
Conv.
1
4
100
80
60
40
20
0
Others
4
5
6
8
8
60
77
1
2
3
4
5
Condition: 2 mmol CPO, 5 mL NH3-MeOH (2 M), 20 mg 1%Ru/Nb2O5-L, 90 oC,
2 MPa H2
Runs
Figgure 10. The durability test of 1%Ru/Nb2O5-L catalysts.
the production of CPA in this stage is totally unexpected.
Surprisingly, when we analyzed the MS signal (Figure S2) during
the reaction, it obviously showed that the ammonia molecules
decompose to H2 and N2 over 1%Ru/Nb2O5-L, which is believed
Condition: 2 mmol CPO, 5 mL NH3-MeOH (2 M), 20 mg Catalysis, 90 oC, 2
MPa H2, 1 h.
[19]
Conclusions
to be easy to occur over the Ru contented catalysts.
Consequently, it can now be expected that the formed H2
immediately hydrogenates the intermediate 2 to CPA (1); as a
result, no signal of 2 was detected. This finding suggests that the
activation of NH3 over Ru species is unnecessary for the
reductive amination and the adsorption of ammonia over Ru
sites can cause the NH3-deposition side-reaction, declining the
ultilisation of ammonia. Also, a large amount of NH3 was found to
be adsorbed on the B-acid and L-acid sites (1457 cm-1 and 1675
cm-1 respectively，see Figure S4a and Table S1) over
In summary, we successfully synthesis three 1%Ru/Nb2O5
catalysts with different morphologies for
a
comparative
investigation in the reductive amination of cyclopentanone, and
the best catalytic performance can be achieved over
1%Ru/Nb2O5-L (a CPA yield of 84%). Furthermore,
1%Ru/Nb2O5-L catalyst shows good stability during CPO
reductive amination process, and can successfully be expanded
to a series of substrates to produce the corresponding primary
amines. Our results manifest that the activity of the reaction
systems is mainly controlled by the geometric effect of Ru
particles when the chemical states of Ru are identical. Nb2O5
supports (like Nb2O5-L
material) with higher specific area are able to promote the
dispersion of active Ru species over the surface of catalysts.
Importantly, good Ru dispersion of catalysts can enhance the
hydrogenation capability during this reaction, and thus can
suppress the acid-related side-reaction over Nb-based supports.
Our DRIFT-MS data illustrate that Ru species activate the
carbonyl groups and H2 during the reaction, and the reaction
between aldehyde/ketone species and ammonia is considered
to undergo via an E-R mechanism. This study provides an in-
depth understanding of surface processes during the reaction
and firmly points out that our 1%Ru/Nb2O5-L catalyst can be one
of the promising candidates for green aldehyde/ketone reductive
amination process.
+
1%Ru/Nb2O5 catalyst, and the adsorbed NH4 and NH3→L
species are usually regarded to be inactive during the reductive
amination of ketones/aldehydes, as they have no ability to share
the paired electrons. In brief, the in situ DRIFT-MS data infer
that the ammonia molecule is more likely to directly react with
the Ru-adsorbed CPO over the surface of catalysts, which
follows an Eley-Rideal (E-R) mechanism.[20]
Durability test and substrate scope over 1%Ru/Nb2O5-L
catalyst
The durability of the catalytic systems is one of the most
important properties concerned in the industry, so the recycling
tests were conducted to see the stability of the 1%Ru/Nb2O5-L
catalyst. Figure 10 clearly shows that no significant decrease in
activity or selectivity was observed during the initial three
successive runs and only a tiny decline of the catalytic
performance was seen in the 4th and 5th runs, which illustrates
the good stability of the catalytic system. Also, the XRD (Figure
S3) and ICP-AES results of the spent catalyst were also
recorded. These results show that neither the phase
Experimental Section
Catalyst preparation
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900335
ChemCatChem
FULL PAPER
The Layer Nb2O5 (labeled as Nb2O5-L) was prepared by a
The liquid phase was separated from the catalyst by
centrifugation, and two individual GC/GC-MS systems were
used for product analyses. The qualitative analysis of products
was carried out on a GC-MS system (Agilent 7890A-5975C),
and the quantitative analysis was executed on a GC system
(Agilent 7890B) equipped with an HP-5 column and an FID
detector.
modified hydrothermal method.[13] Typically, 19.2
g
self-
synthesized niobium oxalate solution (0.5 mol/L) and 0.71 g
ammonium oxalate were well mixed in 40 mL deionized water,
and then transferred to a 100 mL Teflon-lined stainless-steel
autoclave. After the autoclave was steeled, the mixed solution
was kept under 180 C for 3 days. Then, the white powder was
collected by filtration, followed by drying at 100 C for 12 h, and
calcined at 400 C for 4 h in air. The hollow Nb2O5 (labeled as
The TOF value of CPO reductive amination is calculated by
the following equation:
Nb2O5-H) was fabricated through
a
modified one-pot
hydrothermal method according to the literature.[21] In a typical
process, 1.368 g of resorcinol (R) 2.016 g of formaldehyde and
9.6 g of niobium oxalate aqueous solutions were mixed in 50 mL
Conv. % ( nCPO/ nRu )
 )
TOF=
o
t i me mi n  D
DI water at room temperature. After stirring for 4 h at 40 C, the

)
Ru
solution was transferred to a 100 mL Teflon-lined autoclave, and
heated at 180 oC for 24 h. The resulting precipitates were
collected, filtered, washed with DI-water several times, then
dried at 60 oC for 24 h and calcined at 600 C (2 oC·min-1 ) for 2
h in air. Flower-like Nb2O5 (labeled as Nb2O5-F) were also
synthesized via a modified hydrothermal process according to
literature.[14] In a typical synthesis process, 0.4 g NbCl5 was
dissolved in 4 ml ethanol. Then, 17 ml deionized water and 3 ml
ammonium hydroxide solution were added into the NbCl5
ethanol solution under vigorous stirring, and the mixture was
stirred for another 2 h at room temperature. Then the white
precipitates were separated by centrifugation and ultrasonically
re-dispersed in 56 ml DI water. The suspension was transferred
and sealed into a 100 mL Teflon-lined autoclave, and heated at
200 oC for 24 h. The resulting precipitates were collected,
filtered, washed with DI-water three times, then dried at 60 oC for
24 h and calcined at 400 C (2 oC·min-1) for 2 h in air to obtain
Nb2O5-F.
Catalyst characterization
All synthesized Ru/Nb2O5 catalysts were extensively
characterized to understand the structure-property relationships
of the catalytic system.
Powder X-ray diffraction (XRD) patterns were recorded in the
θ-2θ mode on a D8 Focus diffractometer (CuKα1 radiation, k =
1.5406 Å), operated at 40 kV and 40 mA within scattering angels
of 10-80. Nitrogen sorption measurements were performed on a
Micromeritics ASAP 2020 sorption analyzer. The samples were
degassed at 200 C for 6 h, and measured at -196 oC. The
Brunauer-Emmett-Teller (BET) method was adopted to calculate
the specific surface area. Scanning electron microscopy (SEM)
was recorded on a Nova NanoSEM 450 microscopy at 3 kV.
Transmission electron microscopy (TEM) was obtained on a
JEOL Model 2100 electron microscopy at 200 kV. Hydrogen
temperature-programmed-reduction (H2-TPD) was carried out in
a PX200 apparatus with a thermal conductivity detector (TCD).
All Ru/Nb2O5 catalysts were prepared by incipient wetness
impregnation method. Typically, a certain amount of 10 wt.%
RuCl3 aqueous solution was added to the Nb2O5 supports and
stirred to be a slurry. Then the slurry was dried at 60 C for 12 h,
and 110 C for another 12 h. Before used, the catalysts were
reduced at 400 C for 2 h under 10% H2-Ar mixed gas. The
catalysts prepared were labeled as Ru/Nb2O5-n (n = L, H, F),
respectively, depending on the Nb2O5 support used. The initial
content of Ru is fixed at 1 wt.% and the final content was
measured by an Agilent 725ES inductively coupled plasma
atomic emission spectrometry (ICP-AES), which is very close to
the initial one (Table 1).
The Ru dispersion was measured by CO chemisorption using
the dynamic adsorption method with an automatic chemisorbent
(ChemiSorb 2720). Before adsorption, the sample (0.1 g) was
reduced at 200 ° C for 30 min (10 °C/min) under pure H2 (60
mL/min), cooled to room temperature, and flushed in He for 30
minutes. Then, a pulse of 5% CO in He was injected through a
thermal conductivity detector (TCD) until saturation was
reached. Ru dispersion was calculated by assuming that the
stoichiometric ratio of CO adsorbed by Ru is 1:1.
The Ru-colloid loaded Nb-based catalysts with the same Ru
particles were prepared by adsorption of colloidal Ru
nanoparticles (NPs) onto the Nb2O5-L, Nb2O5-H and Nb2O5-F
supports.[22] Colloidal Ru NPs with size of 1.2 nm was prepared
through the reduction of RuCl3 with ascorbic acid in the aqueous
Ammonia temperature-programmed-desorption (NH3-TPD)
was carried out in a PX200 apparatus (Tianjin Golden Eagle
Technology Limited Corporation) with a thermal conductivity
detector (TCD). The sample (50 mg) was charged into a quartz
reactor, and the temperature was raised from room temperature
to 600 oC at a rate of 10 oC min -1 under a flow of N2 (50 ml min -
1), and then the temperature was cooled to 90 oC. Then, the NH3
(5 ml min-1) was injected into the reactor at 90 oC under a flow of
N2 (45 ml min-1). When the adsorption saturation is reached, the
N2 (25 ml min-1) is flowed at 90 oC for 1 hour, then the
temperature is raised from 90 oC to 600 oC （10 ° C min-1） and
o
phase 80 C for 1 min. Then, add the support into the colloidal
Ru particle solution and stirred for 24 h. Finally, it was washed
by DI water and dried at 50 oC vacuum oven. Before used, it was
o
solidified at 300 C for 4 h under a N2 flow. The content of Ru
NPs was 1.0 wt %.
Catalysts activity tests
o
the amount of ammonia was detected at 110 C by using a
thermal conductivity detector(TCD).
The reductive amination of CPO was conducted in a Teflon-
lined stainless-steel autoclave (50 mL). After sealing the desired
dosage of CPO, catalyst, and NH3-CH3OH solution (2M) in the
reactor, the autoclave was purged with H2 three times and
pressurized to the desired pressure. Then, the autoclave was
heated to the predetermined temperature in a short time. After
reaction, the reactor was quenched in an ice-water bath
immediately.
X-ray photoelectron spectroscopy (XPS) spectra were
recorded on a Thermo Scientific Escalab 250 Xi system with
monochromatic Al Kα radiation, and the results were calibrated
using the C 1s peak at 284.6 eV. Diffuse reflectance infrared
Fourier transform spectra (DRIFTS) were recorded on a Nicolet
Model iS-50 FT-IR spectrometer equipped with an MCT/A
detector, and the sample cell was fitted with ZnSe windows. The
DRIFT spectra of CO-adsorption (1% CO/Ar) were recorded with
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900335
ChemCatChem
FULL PAPER
a resolution of 4 cm-1 and 32 scans at 30 oC. DRIFT-MS spectra
was also recorded on a Nicolet Model iS-50 FT-IR spectrometer
which was same as DRIFT spectra. The samples were loaded
onto DRIFT IR cell, then reduced at 400 C for 1 h under 5% H2-
Technol. 2014, 4, 4250-4257.
[14] W. Dong, F. Pan, Y. Wang, S. Xiao, K. Wu, G. Q. Xu, W. Chen, Appl.
Surf. Sci. 2017, 392, 514-522.
[15] H. K. Hall, R. B. Bates, Tetrahedron Lett. 2012, 53, 1830-1832.
[16] a) S. Zhang, Z. Xia, T. Ni, Z. Zhang, Y. Ma, Y. Qu, J. Catal. 2018, 359,
101-111; b) T. Tong, X. Liu, Y. Guo, M. Norouzi Banis, Y. Hu, Y. Wang, J.
Catal. 2018, 365, 420-428; c) S. Li, Y. Xu, Y. Chen, W. Li, L. Lin, M. Li,
Y. Deng, X. Wang, B. Ge, C. Yang, S. Yao, J. Xie, Y. Li, X. Liu, D. Ma,
Angew. Chem., Int. Ed. 2017, 56, 10761-10765.
o
Ar mixed gas. After cooled to the 30 C, Ar (99.99% pure) was
passed through a bubbler containing CPO via a volume flow
controller before entering the DRIFT cell in 30 min，then flushed
in Ar without CPO for 30 minutes. After that the NH3 was pulsed
into the DRIFT cell, and then changed to 5%H2/Ar after 5-minute
sweeping in Ar. The flow rate remained constant at 30 ml/min for
all experiments. Collect infrared spectra after each gas
exchange. The gaseous effluent from the reactor was monitored
by a Mass Spectrometer (MS). The following m/z ratios were
monitored on the LingLu Instruments QAS 100 mass
spectrometer: 2 (H2), 17(NH3), 28 (N2) and 40 (Ar).
[17] J. An, Y. Wang, J. Lu, J. Zhang, Z. Zhang, S. Xu, X. Liu, T. Zhang, M.
Gocyla, M. Heggen, R. E. Dunin-Borkowski, P. Fornasiero, F. Wang, J.
Am. Chem. Soc. 2018, 140, 4172-4181.
[18] M. I. Zaki, M. A. Hasan, L. Pasupulety, Langmuir 2001, 17, 768-774.
[19] S. F. Yin, B. Q. Xu, X. P. Zhou, C. T. Au, Appl. Catal., A 2004, 277, 1-9.
[20] J. Mai, W. von Niessen, Chem. Phys. 1991, 156, 63-69.
[21] L. Kong, C. Zhang, J. Wang, W. Qiao, L.Ling, D. Long, Sci. Rep. 2016,
6, 21177.
[22] L. Dong, L.-L. Yin, Q. Xia, X. Liu, X.-Q. Gong, Y. Wang, Catal. Sci.
Technol. 2018, 8, 735-745.
Acknowledgements
This work was supported financially by the NSFC of China (No.
21832002, 21872050, 21808063), the Fundamental Research
Funds for the Central Universities (222201718003) and the
Science and Technology Commission of Shanghai Municipality
(18ZR1408500, 10dz2220500).
Keywords: reduction amination• biomass• DRIFT-MS • niobium
oxide• geometric effect
[1]
a) V. Froidevaux, C. Negrell, S. Caillol, J.-P. Pascault, B. Boutevin,
Chem. Rev. 2016, 116, 14181-14224; b) M. Pelckmans, T. Renders, S.
Van de Vyver, B. F. Sels, Green Chem. 2017, 19, 5303-5331; c) K. S.
Hayes, Appl. Catal., A. 2001, 221, 187-195; d) K.-i. Shimizu, K. Kon, W.
Onodera, H. Yamazaki, J. N. Kondo, ACS Catal. 2013, 3, 112-117; e) J.
P. Wolfe, H. Tomori, J. P. Sadighi, J. Yin, S. L. Buchwald, J. Org. Chem.
2000, 65, 1158-1174; f) M. S. Driver, J. F. Hartwig, J. Am. Chem. Soc.
1996, 118, 7217-7218.
[2]
[3]
Z. Dong, G. Dong, J. Am. Chem. Soc. 2013, 135, 18350-18353.
A. Said Stålsmeden, J. L. Belmonte Vázquez, K. van Weerdenburg, R.
Rae, P.-O. Norrby, N. Kann, ACS Sustainable Chem. Eng. 2016, 4,
5730-5736.
[4]
A. F. Abdel-Magid, K. G. Carson, B. D. Harris, C. A. Maryanoff, R. D.
Shah, J. Org. Chem. 1996, 61, 3849-3862.
[5]
[6]
H. Kim, S. Chang, ACS Catal. 2016, 6, 2341-2351.
a) M. R. Ball, T. S. Wesley, K. R. Rivera-Dones, G. W. Huber, J. A.
Dumesic, Green Chem. 2018, 20, 4695-4709; b) T. Cuypers, P.
Tomkins, D. E. De Vos, Catal. Sci. Technol. 2018, 8, 2519-2523.
M. H. S. A. Hamid, P. A. Slatford, J. M. J. Williams, Adv. Synth. Catal.
2007, 349, 1555-1575.
[7]
[8]
a) T. Senthamarai, K. Murugesan, J. Schneidewind, N. V. Kalevaru, W.
Baumann, H. Neumann, P. C. J. Kamer, M. Beller, R. V. Jagadeesh,
Nat. Commun. 2018, 9, 4123; b) J. Gallardo-Donaire, M. Ernst, O.
Trapp, T. Schaub, Adv. Synth. Catal. 2016, 358, 358-363; c) H. Huang,
X. Liu, L. Zhou, M. Chang, X. Zhang, Angew. Chem., Int. Ed. 2016, 55,
5309-5312.
[9]
a) G. Liang, A. Wang, L. Li, G. Xu, N. Yan, T. Zhang, Angew. Chem., Int.
Ed. 2017, 56, 3050-3054; b) R. V. Jagadeesh, K. Murugesan, A. S.
Alshammari, H. Neumann, M. M. Pohl, J. Radnik, M. Beller, Science
2017, 358, 326-+;M. Chatterjee, T. Ishizaka, H. Kawanami, Green
Chem. 2016, 18, 487-496.
[10] T. Komanoya, T. Kinemura, Y. kita, K. Kamata, M. Hara, J. Am. Chem.
Soc. 2017, 139, 11493-11499.
[11] a) D. Chandra, Y. Inoue, M. Sasase, M. Kitano, A. Bhaumik, K. Kamata,
H. Hosono, M. Hara, Chem. Sci. 2018, 9, 5949-5956; b) D. Deng, Y.
Kita, K. Kamata, M. Hara, ACS Sustainable Chem. Eng. 2018.
[12] Y. L. Yang, Z. T. Du, Y. Z. Huang, F. Lu, F. Wang, J. Gao, J. Xu, Green
Chem. 2013, 15, 1932-1940.
[13] T. Murayama, J. Chen, J. Hirata, K. Matsumoto, W. Ueda, Catal. Sci.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900335
ChemCatChem
FULL PAPER
Text for Table of Contents
Wanjun Guo⊥, Tao Tong⊥, Xiaohui Liu,
Yong Guo, Yanqin Wang*
Cyclopentylamine is high selectively
obtained over layered Nb2O5
supported Ru catalyst by the
reductive amination of
Page No. – Page No.
Morphology-tuned activity of
RuNb2O5 catalysts for ketones
reductive amination
cyclopentanone.
This article is protected by copyright. All rights reserved.