Hydrogenation of Furfural to Cyclopentanone under Mild Conditions by a Structure-Optimized Ni?NiO/TiO2 Heterojunction Catalyst

Article abstract of DOI:10.1002/cssc.202001424

The catalytic conversion of biomass-derived furfural (FFA) into cyclopentanone (CPO) in aqueous solution is an important pathway to obtain sustainable resources. However, the conversion and selectivity under mild conditions are still unsatisfactory. In this study, a catalyst consisting of Ni?NiO heterojunction supported on TiO₂ with optimized composition of anatase and rutile (Ni?NiO/TiO₂-Re450) is prepared by pyrolysis at 450 °C. With Ni?NiO/TiO₂-Re450, as catalyst, complete conversion of FFA and 87.4 % yield of CPO are achieved under mild reaction conditions (1 MPa, 140 °C, 6 h). 95.4 % FFA conversion is retained up to the fifth run, indicating the high stability of the catalyst. Multiple characterizations, control experiments, and theoretical calculations demonstrate that the good catalytic performance of Ni?NiO/TiO₂-Re450 can be attributed to a synergistic effect of the Ni?NiO heterojunction and the TiO₂ support. This low-cost catalyst may expedite the catalytic upgrading and practical application of biomass-derived chemicals.

Full text of DOI:10.1002/cssc.202001424

Chemistry–Sustainability–Energy–Materials
Accepted Article
Title: High-performance hydrogenation of furfural to cyclopentanone
under mild conditions by structure-optimized Ni-NiO/TiO2
heterojunction catalyst
Authors: Hong Jiang, Shuo Chen, Ting-Ting Qian, Li-Li Ling, Wenhua
Zhang, and Bing-Bing Gong
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: ChemSusChem 10.1002/cssc.202001424
Link to VoR: https://doi.org/10.1002/cssc.202001424
01/2020

10.1002/cssc.202001424
ChemSusChem
High-performance hydrogenation of furfural to cyclopentanone under mild
conditions by structure-optimized Ni-NiO/TiO2 heterojunction catalyst
Shuo Chen†1, Ting-Ting Qian‡1, Li-Li Ling†, Wenhua Zhang§, Bing-Bing Gong§,
Hong Jiang†, *
† CAS Key Laboratory of Urban Pollutant Conversion, Department of Applied
Chemistry, University of Science and Technology of China, Hefei 230026, China.
‡ Advanced Environmental Biotechnology Centre, Nanyang Environment and Water
Research Institute, Nanyang Technological University, 1 Cleantech Loop, Singapore
637141, Singapore
§Hefei National Laboratory for Physical Sciences at the Microscale, CAS Key
Laboratory of Materials for Energy Conversion and Synergetic Innovation Centre of
Quantum Information & Quantum Physics, University of Science and Technology of
China, Hefei, Anhui 230026, China
1These authors contributed equally to this work.
*Corresponding author: jhong@ustc.edu.cn (H. J.), Tel/Fax: +86-551-63607482
1
This article is protected by copyright. All rights reserved.

10.1002/cssc.202001424
ChemSusChem
TOC Art
2
This article is protected by copyright. All rights reserved.

10.1002/cssc.202001424
ChemSusChem
Abstract: The catalytic conversion of biomass-derived furfural (FFA) to
cyclopentanone (CPO) in aqueous solution is an important pathway to obtain
sustainable resources, however, the conversion and selectivity under mild conditions
are still unsatisfactory. Herein, a catalyst that is made of Ni-NiO heterojunction
supported by TiO2 with optimized composition of anatase and rutile (Ni-NiO/TiO2-
Re450) was facilely prepared by pyrolysis at 450 °C. By using Ni-NiO/TiO2-Re450,
complete conversion of FFA and 87.4% of CPO yield can be achieved under mild
reaction conditions (1 MPa, 140 °C, 6 h). The FFA conversion was kept at 95.4% in
the fifth run, indicating the high stability of the catalyst. Multiple characterizations,
control experiments, and theoretical calculation demonstrate that the high catalytic
performance of Ni-NiO/TiO2-Re450 mainly attributes to the synergistic effect of Ni-
NiO heterojunction and the TiO2 support. This low-cost catalyst may expedite the
catalytic upgrading and practical application of biomass-derived chemicals.
3
This article is protected by copyright. All rights reserved.

10.1002/cssc.202001424
ChemSusChem
Introduction
Biomass can be converted into renewable fuels and chemicals, such as gases, liquids,
[1]
solids, and other valuable chemicals.
Among these biomass-deserved products,
furan derivatives, e.g., furfural (FFA), are readily acquired.[2] FFA is an extremely
useful chemical intermediate of important nonpetroleum-derived chemicals, including
cyclopentanone (CPO), cyclopentanol (CPL), furfuralcohol (FA), tetrahydrofurfuryl
alcohol (THFA), and 2-methylfuran.[3] For instance, FFA can be converted to CPO (a
raw material for the pharmaceutical and fragrance industries) by hydrogenation
[4]
reaction.
In previous studies, some precious metals (e.g., Pt, Pd, and Ru), which
was loaded on diversified stable supports, were used to catalyze the hydrogenation
[5]
reaction. However, the highly efficient and selective conversion of FFA to CPO is
still challenging, including production of a large number of by-products, high cost of
precious metals, and high reaction temperature and H2 pressure. Therefore, it is
imperative to design and synthesis of low-cost catalysts with high catalytic
performance under mild conditions.
Nickel-based catalysts are widely used in various hydrogenation reactions due to
[6]
their high catalytic activity and low cost.
Particularly, the Ni-NiO heterojunction
formed by the partial reduction of nickel oxide can improve the performance of the
Ni-based catalysts.[7] On the other hand, TiO2 is a favorable support for Ni-based
catalysts used in FFA hydrogenation. Oxygen vacancies on the surface of TiO2 can
promote the formation of high active furfuryl-oxy compound by electron transfer and
facilitate FFA hydrogenation. These oxygen vacancies were reported to increase when
[8]
TiO2 is calcined at H2 atmosphere.
In addition, TiO2 has different crystalline
phases: anatase, rutile, and brookite. [9] Rutile may facilitate the formation of Ni-NiO
[10]
heterojunction at a low temperature.
Therefore, an optimum catalyst for FFA
4
This article is protected by copyright. All rights reserved.

10.1002/cssc.202001424
ChemSusChem
hydrogenation may be prepared by pyrolyzing the nickel oxide and TiO2 under
controlled conditions, under which Ni-NiO heterojunction loaded on specific TiO2
crystals will form.
To our delight, the optimum TiO2-supported Ni-NiO heterojunction catalyst for
the hydrogenation of FFA to CPO was facilely synthesized at 10% H2/Ar atmosphere
by tuning the heat treatment temperature to 450 °C (Ni-NiO/TiO2-Re450). The
complete conversion of FFA and the CPO yield of 87.4% were achieved under mild
reaction conditions (1 MPa, 140 °C, 6 h) by the catalyst. Notably, the conversion
maintained at 95.4% even in the fifth run, indicating the high stability of the catalyst.
To our knowledge, achieving both the high conversion and high selectivity of FFA to
CPO using transition metal catalyst has not been reported.
Results and Discussion
Reduction temperature, an easily controlled parameter during catalyst preparation, can
greatly impact the catalytic performance of the catalyst. To reveal the effect of
reduction temperature on catalyst properties, the SEM images and the pore size
distribution of the catalysts prepared under H2/Ar atmosphere (10% H2/Ar) at
different temperatures (350 °C to 750 °C) are shown in Figure 1, Figure S1, and Table
S1. As shown in Figure 1a-1e, the aggregation increased with the reduction
temperature. As the reduction temperature raised from 350 °C to 750 °C, the pore size
gradually increased. When the reduction temperature was up to 750 °C, no peak is
observed due to the collapse of the porous columnar structure (Figure S1), which is
consistent with previous reports. [11] In addition, the drastic decrease of surface area
with the increase of the reduction temperature (Table S1) may be caused by the
[12]
gradual conversion of small anatase crystals to relatively large rutile ones.
5
This article is protected by copyright. All rights reserved.

10.1002/cssc.202001424
ChemSusChem
Meanwhile, thermal sintering also leads to the continuous crystallite aggregation with
[13]
the increasing temperature.
The small particle size of catalyst generally
corresponds to large surface area. Therefore, the catalyst obtained at low temperature
generally has higher activity in the catalytic reaction.
Figure 1f shows the XRD spectra of the Ni-NiO/TiO2 catalysts prepared at
different reduction temperature. The results show that the reduction temperature has a
drastic influence on the crystal composition of the catalyst. The XRD spectrum of
NiO/TiO2 only shows the characteristic peaks of NiO at 37.2°, 62.9° and 75.2°,
whereas the XRD spectrum of Ni-NiO/TiO2-Re450 shows the characteristic peaks of
metallic Ni at 44.5°, 51.8°, and 76.4°. When the reduction temperature increased from
350 °C to 750 °C, NiO was gradually reduced to elemental Ni. From the spectrum of
Ni-NiO/TiO2-Re750, only a weak NiO peak is observed, which indicates that most
NiO on the catalyst surface was converted to Ni. The XRD spectra demonstrate the
successful formation of Ni-NiO heterojunction at relatively low reduction
temperature.
The XRD spectrum of Ni-NiO/TiO2-Re450 shows four strong diffraction peaks
of rutile phase at 27.5°, 36°, 55.2°, and 70.7°, and four strong diffraction peaks of
anatase phase at 25.3°, 48.2°, 54°, and 68.9°. The proportions of rutile and anatase of
different catalysts were calculated (the calculation method can be found in Text S1 of
SI) according to the XRD spectra. The proportion of rutile increases from 0.206 to 1.0
with the reduction temperature increased from 450 °C to 750 °C (Figure S2), and no
diffraction peak of anatase is observed on the spectrum of Ni-NiO/TiO2-Re750. The
reduction temperature can change the crystal structure of TiO2 and tune the ratio of
anatase to rutile, which is consistent with the results in the literature.[14]
Figure 1g shows the H2-TPR profiles of the catalysts at different reduction
6
This article is protected by copyright. All rights reserved.

10.1002/cssc.202001424
ChemSusChem
temperature. For NiO/TiO2-Re450, the reduction peak of NiO is mainly centered at
450 °C, which is attributed to the NiO species in NiO-TiO2 or the NiO particles
dispersed on the surface of TiO2. [15] With the increase on the reduction temperature,
the area of the NiO reduction peak gradually decreased, indicating NiO was gradually
reduced to elemental Ni. As shown in Table S2, the NiO content in the catalyst
decreased from 19.2% to 0.5% when the reduction temperature increased from 350 °C
to 750 °C. Figure 1g also shows that the position of the reduction peak gradually
shifts to low temperature. This phenomenon may be caused by the gradual conversion
[16,10b]
of the anatase to rutile with the increase of reduction temperature.
Rutile
possesses a stable crystal structure, high crystallization state, few defects, and low
activity, which can weaken the interaction between NiO and TiO2 and make the
reduction peak of NiO shift to low temperature. [9]
Figure 1. SEM images of catalysts reduced at different temperature under 10% H2/Ar
atmosphere (a) Ni-NiO/TiO2-Re350, (b) Ni-NiO/TiO2-Re450, (c) Ni-NiO/TiO2-
7
This article is protected by copyright. All rights reserved.

10.1002/cssc.202001424
ChemSusChem
Re550, (d) Ni-NiO/TiO2-Re650, (e) Ni-NiO/TiO2-Re750, (f) XRD diffraction patterns
for the Ni-NiO/TiO2 catalyst under different reduction temperatures. (g) H2-TPR
profiles of the Ni-NiO/TiO2 catalyst under different reduction temperatures.
A series of hydrogenation experiments were conducted to test the performance of
Ni-NiO/TiO2 catalysts on FFA hydrogenation. Figure 2a shows the product
distributions of FFA hydrogenation over the Ni-NiO/TiO2 catalysts obtained at
different reduction temperature. It is observed that Ni-NiO/TiO2-Re450 achieved the
highest FFA conversion (100%) and the CPO yield of 87.4%. The carbon equilibrium
of FFA hydrogenation by Ni-NiO/TiO2-Re450 was 95.3% on the basis of the external
standard method (Table S3), indicating that there was only a small amount of carbon
loss during the reaction. The total nickel content in the Ni-NiO/TiO2-Re450 was only
8.2%, including 1.2% of NiO and 7% of Ni (Table S2), which indicates that the
catalyst exhibited high catalytic activity with a low metal loading.[6, 17] Although Ni-
NiO/TiO2-Re350 also exhibited the high FFA conversion, the yield of CPO over
NiO/TiO2-Re350 was lower than that over Ni-NiO/TiO2-Re450 due to the presence of
other by-products such as CPL and THFA. Ni-NiO/TiO2-Re550 exhibited comparable
performances to Ni-NiO/TiO2-Re450 on FFA conversion and CPO yield, however, it
needed higher energy consumption during catalyst preparation. When the reduction
temperature increased from 650 °C to 750 °C, FFA conversion significantly
decreased, and the total conversion of FFA by Ni-NiO/TiO2-Re750 was only 10.7%
and no other intermediates were detected, which may be caused by resinification of
FFA or production of other trace intermediates. [5a, 18]
To reveal the effects of the reaction parameters on the catalytic reaction, we
performed a series of FFA hydrogenation experiments with Ni-NiO/TiO2-Re450 under
different reaction conditions, including reaction temperature, pressure, time, and
8
This article is protected by copyright. All rights reserved.

10.1002/cssc.202001424
ChemSusChem
solvents.
The effect of reaction temperature ranging from 100 °C to 160 °C on the
conversion of FFA is presented in Figure 2b. It can be seen that the temperature
exhibited a considerable effect on the product distribution. At low temperature
(100 °C), the yield of FA, a key intermediate compound in FFA conversion, was
87.1%. As the reaction temperature increased, FA decreased significantly, and no FA
was detected at 140 °C. At the same time, the CPO yield reached the maximum
(87.4%) at 140 °C. When the temperature increased to 160 °C, the CPO yield
decreased to 77.3%, and the yields of the byproducts, i.e., CPL and THFA, increased
to 8.4% and 14.3%, respectively. It can be seen from Figure S3 that the conversion of
FFA increased as the reaction temperature increased, but the yield of CPO showed a
downward trend after 4 h of reaction when the temperature was 160 °C. Therefore, the
optimal temperature and time for FFA conversion and CPO production were 140 °C
and 6 h, respectively.
The effect of reaction pressure on the FFA conversion is presented in Figure 2c.
Under 0.5 MPa, FFA conversion and the CPO yield were 86.1% and 39.8%,
respectively. When the pressure increased to 1 MPa, FFA conversion and CPO yield
increased to the maximum of 100% and 87.4%, respectively. With further increase of
the pressure, FFA conversion leveled off, whereas the CPO yield decreased. Under 3
MPa, the CPO yield reduced to 78%, and that of the byproduct-CPL increased to
8.2%. It can be seen from Figure S4 that the conversion of FFA increased with the
increasing of the reaction pressure when the reaction time was constant, but the yield
of CPO showed a decreasing trend after 4 h of the reaction when the pressure was
above 2 MPa. The CPO yield of 89.2% under the reaction pressure of 3 MPa and
reaction time of 4 h was slightly greater than that (87.4%) under the reaction pressure
9
This article is protected by copyright. All rights reserved.

10.1002/cssc.202001424
ChemSusChem
of 1 MPa and reaction time of 6 h, however the reaction pressure of 1 MPa was milder
and safer. These results indicate that 1 MPa of H2 is optimum for CPO production.
The effect of reaction time on FFA conversion under optimal reaction conditions
(140 °C, 1 MPa and 6 h) is presented in Figure 2d. The results show that the FA yield
was 80% within 1 h, and it gradually decreased with the increase of reaction time. The
CPO yield increased with the reaction time and reached the maximum (87.4%) at 6 h.
The CPL yield increased with the reaction time from 0% (1 h) to 2.8% (6 h), while the
THFA yield increased from 0.6% (1 h) to 9.8% (6 h). According to FFA conversion
and product distribution, the intermediate FA was completely converted to other
compounds within 6 h, and the selectivity of CPO achieved the highest value (87.4%).
Figure 2. (a) Effect of reduction temperature. Reaction conditions: 140 °C, 1 MPa
and 6 h. (b) Effect of temperature on FFA hydrogenation. Reaction conditions: 1 MPa
and 6 h. (c) Effect of pressure on FFA hydrogenation. Reaction conditions: 140 °C
and 6 h. (d) Effect of reaction time on FFA hydrogenation. Reaction conditions:
10
This article is protected by copyright. All rights reserved.

10.1002/cssc.202001424
ChemSusChem
140 °C and 1 MPa. (e) Performance of the Ni-NiO/TiO2-Re450 in the recycling
experiment. Reaction conditions: 140 °C, 1 MPa and 6 h (All reactions conditions: 1
mmol of FFA, 50 mg of catalyst, 10 mL of H2O).
Kinetics of FFA conversion at four different reaction temperatures (i.e., 100, 120,
140, and 160 °C) were studied, and the Eyring-Polanyi equation (Eq.1) was used to
[19]
study the thermodynamic characteristics of this reaction
According to Table S4,
the FFA conversion conforms to the first-order reaction, and the reaction rate constant
k increases from 0.4982 h-1 to 1.0038 h-1 with the increase of the reaction temperature.
k
∆
RT
H
k
b 
h

∆S
R
 


ln
= −
+ ln
+

 
T
 

(1)
(2)
∆G = ∆H −T∆S
k is the reaction rate constant at different temperatures, ΔH and ΔS are the changes in
enthalpy and entropy in the reaction system, respectively. R, h and kb are the gas
constant, planck constant and boltzmann constant, respectively.
ΔH and ΔS are calculated from the slope and intercept of the Eyring-Polanyi
curve of the reaction from Figure S5. The change of Gibbs free energy (ΔG) can be
calculated using Eq.2 , and the values of ΔH, ΔS and ΔG for the reaction are shown in
Table S5. The positive value of ΔH indicates that Ni-NiO/TiO2-Re450 involved FFA
hydrogenation is an endothermic reaction. The negative value of ΔS indicates that the
entropy of the reaction decreased, and the positive value of ΔG indicates that the
reaction is not spontaneous in nature.
The performance of Ni-NiO/TiO2-Re450 on FFA conversion was compared with
those of other non-metal catalysts (Table 1). Among the non-precious metal catalysts,
Ni-NiO/TiO2-Re450 exhibits a relatively high reaction rate and CPO yield under the
mild reaction conditions (140 °C, 1 MPa). Both the reaction pressure and temperature
are relatively low, indicating a safe and economical hydrogenation process.
Table 1. Comparison of Some Critical Parameters during the Hydrogenation of FFA
11
This article is protected by copyright. All rights reserved.

10.1002/cssc.202001424
ChemSusChem
to CPO by Non-precious Metals Catalysts in Different Studies
Reaction conditions Con.(%) Yield(%) TOF(h-1) Refs.
No. Catalyst
17
20
21
22
23
24
1
2
3
4
5
6
7
CuZnAl Catalysts
150 °C, 4 MPa
150 °C, 4 MPa
140 °C, 4 MPa
97.9
96.4
100
62
0.83
0.94
0.25
10.08
1.56
0.68
2.09
Ni/HY
86.5
95.8
90
Cu-Ni-Al-HT
NiFe/SBA-15
NiCu-5/ SBA-15
CuZn / CNT
160 °C, 3.4 MPa 99.8
160 °C, 4 MPa
140 °C, 4 MPa
100
95
62
85.3
87.4
Ni-NiO/TiO2-Re450 140 °C, 1 MPa
100
This study
The effect of the solvent on the catalytic activity of Ni-NiO/TiO2-Re450 on FFA
hydrogenation is shown in Table 2. The results show that the reaction solvent
exhibited a considerable influence on the conversion and the yield of FFA
hydrogenation. The highest FFA conversion (100%) and yield of CPO plus CPL
(90.2%) could be obtained using H2O as the reaction solvent. Whereas, low yields of
CPO plus CPL (no more than 35.7%) were present when dichloromethane (CH2Cl2),
methanol (CH3OH), acetone (CH3COCH3) and glacial acetic acid (CH3COOH) were
used as the reaction solvent. These solvents all contain a small amount of H2O, for
example, CH3COOH contains 0.2% water and this part of water may be involved in
the conversion of CPO.
To further prove that water is the most suitable solvent, we prepared several
solvents with different proportion of ethanol and water for FFA hydrogenation
experiments. The GC spectra of the solution after the reaction in different solvent
(different ethanol and water mixtures) are shown in Figure S6. It can be seen from
Figure S6 that when the proportion of ethanol increased from 40% to 100%
12
This article is protected by copyright. All rights reserved.

10.1002/cssc.202001424
ChemSusChem
(anhydrous ethanol), the yield of CPO decreased from 75.2% to 0.5% indicating that
H2O mixed in the solvent is very critical to the production of CPO. This could be
caused by that H3O+ in water can attack the 5-position of FA, leading to the opening
and closing of the furan ring to form 4-hydroxy-2-cyclopentenone (4-HO-2-CPEO),
[5a]
which is a crucial intermediate product of CPO.
pathway will be discussed later.
The detailed FFA conversion
Table 2. Effect of Different Solvents on FFA Hydrogenation
Product yield (%)
Solvents
Conversion (%)
CPO+CPL
3.5
FA
2.7
15.5
1.3
1.4
0
THFA
0
Others
CH2Cl2
6.2
0
CH3OH
CH3COCH3
CH3COOH
H2O
97.3
83.9
37.1
100
12.9
2.9
1.3
0
66
79.6
0
1.7
35.7
90.2
9.8
0
Reaction conditions: 1 mmol of FFA, 50 mg of catalyst, 10 mL of H2O, the
temperature of 140 oC, H2 pressure of 1 MPa, and the reaction time of 6 h.
Stability is a pivotal property for a catalyst, thus the stability of the as-prepared
Ni-NiO/TiO2-Re450 in the hydrogenation reaction was examined. A batch of Ni-
NiO/TiO2-Re450 was cycled for five times and FFA conversion and product yield
were tested. As shown in Figure 2e, the conversion of FFA was above 95.4% in five
cycles. The yield of CPO remained essentially unchanged, with a CPO selectivity of
84.6% in the fifth cycle. FA yield remained stable during the first and second runs and
increased to 5.4% in the fifth cycle. The THFA yield slightly decreased from 9.8% to
3.8% in the five cycles. Figure S7 shows the TGA data of fresh Ni-NiO/TiO2-Re450
and five times-used Ni-NiO/TiO2-Re450. The weight loss of fresh catalyst was 1%
13
This article is protected by copyright. All rights reserved.

10.1002/cssc.202001424
ChemSusChem
and the weight loss of catalyst after five cycles was about 9%, which suggests that the
decrease on catalyst activity after five cycles may be due to the enrichment of
oligomers on the catalyst surface.[26] The XRD analysis of the catalyst before and after
the five cycles was carried out. As shown in Figure S8 the difference between the
XRD pattern of the unused Ni-NiO/TiO2-Re450 and that of reused catalyst is
insignificant, indicating that the morphology of Ni-NiO/TiO2-Re450 remained
basically unchanged after five cycles.
In order to further confirm the advantages of Ni-NiO heterojunction structure
and the specific TiO2 crystal phase composition of the catalyst on FFA hydrogenation,
a series of TiO2-based catalysts were prepared, and the performance of these catalysts
on FFA hydrogenation were examined (Table 3). It can be observed from Table 3 that
the performances on FFA conversion and CPO yield of different catalyst in the tested
system are as follows: Ni-NiO/TiO2-Re450 > Ni/TiO2-450 > NiO/TiO2-450 > TiO2-
Cal. It can be found that FFA hydrogenation reaction catalyzed by TiO2-Cal were not
occurred under the same reaction conditions, indicating that TiO2 has no catalytic
activity during the reaction, and Ni and NiO are the active sites in the hydrogenation
process. However, TiO2 can act as a carrier, and O-vacancies on the surface can
accelerate the electron transfer.[ 8a]
The formation of Ni-NiO heterojunction structure remarkably improved the
catalytic activity of the catalyst and played a vital role in the hydrogenation of FFA to
CPO. In addition, pure anatase and rutile were chosen as the supports, and the order of
the performances of different catalysts on FFA conversion and the yield of CPO in the
test system decreased as follows: Ni-NiO/TiO2-Re450 > Ni-NiO/rutile-Re450 > Ni-
NiO/anatase-Re450. Strangely, as we all know, anatase is more active than rutile and
contains more oxygen vacancies which could easily transport electrons, but we found
14
This article is protected by copyright. All rights reserved.

10.1002/cssc.202001424
ChemSusChem
that that both FFA conversion and the yield of CPO by Ni-NiO/anatase-Re450 were
lower than those of Ni-NiO/rutile-Re450. Therefore, the composition of active
components of Ni was further studied by H2-TPR (Figure S9). The proportions of Ni
and NiO on different TiO2 substrates are given in Table S2. It can be seen that in Ni-
NiO/rutile-Re450, Ni0 accounted for 97.8% of the total nickel, which is much higher
than that in Ni-NiO/anatase-Re450 (78.4%). This result could explain that Ni-
NiO/rutile-Re450 had a higher catalytic activity than Ni-NiO/anatase-Re450. By
comparing the performances of these TiO2-based catalysts on FFA hydrogenation, it
can be found that the Ni-NiO heterojunction structure and the TiO2 crystal
composition indeed played crucial roles on FFA conversion and CPO yield.
Table 3. Comparison of different catalysts for FFA hydrogenation
Product yield (%)
catalyst
Conversion(%)
CPO
/
CPL
/
FA
/
THFA
/
TiO2-Cal
0
NiO/TiO2-450
18.7
74.7
48.6
53.9
10.0
65.6
41.1
44.9
0
8.7
0.4
3.2
0.4
0
Ni/TiO2-450
1.3
1.4
2.1
7.4
2.9
6.5
Ni-NiO/anatase-Re450
Ni-NiO/rutile-Re450
Reaction conditions: 1 mmol of FFA, 50 mg of catalyst, 10 mL of H2O, the
temperature 140 °C, H2 pressure 1 MPa, and the reaction time of 6 h.
The measurement of Photoluminescence (PL) emission was used to reveal the
efficiency of electron migration, transfer and separation. Figure S10 shows the PL
spectra of P25 and Ni-NiO/TiO2-Re450 materials excited at an excitation wavelength
of 300 nm. In the scanning range of 250–650 nm, two different emission bands were
observed. The peak appearing at about 412 nm is attributed to free exciton emission
and the other emission band is attributed to the bound exciton at 475 nm. The PL
emission intensity of Ni-NiO/TiO2-Re450 at 412 and 475 nm are only one fifth of the
15
This article is protected by copyright. All rights reserved.

10.1002/cssc.202001424
ChemSusChem
original P25. These PL signals are attributed to surface O-vacancies and defects in the
TiO2 sample, indicating that the hydrogenation process of TiO2 in 10% H2/Ar
atmosphere and the doping of nickel metal are beneficial to the formation of O-
vacancies, which are consistent with the results reported by Jiang et al. and Chen et
al.[8c,27] Therefore, the TiO2 substrate in the Ni-NiO/TiO2-Re450 material considerably
inhibits the electron recombination rate and accelerates the electron transfer, which
are beneficial to the hydrogenation reaction.
The synergistic effect of the metal/oxide heterojunction can effectively promote
the hydrogenation reaction, and the formation of the Ni-NiO heterojunction can play a
key role in the activity of the Ni-based catalyst. The anatase lattice exhibits higher
activity than the rutile lattice due to the abundant defects and vacancies.[28] Therefore,
the changes in the Ni and NiO composition and TiO2 crystal phase structure of the
catalyst could result in changes in catalyst activity and product yield. The correlation
between the physical and chemical properties of the catalyst and the catalytic
performance will be elaborated below.
To reveal the chemical structure of Ni-NiO/TiO2-Re450 from the molecular
level, we carried out an XPS survey. As shown in Figure 3a, the Ni 2p region in the
XPS spectrum contains four discernible features, i.e., the Ni 2p3/2 satellite peak
located at ∼854 and ∼862 eV and the Ni 2p1/2 satellite peak located at ∼872 and
∼879 eV [29]. The high-resolution XPS spectrum of Ni 2p of Ni-NiO/TiO2-Re450 is
shown in Figure 3a. The peaks at 855.6 and 861.5 eV are the typical peaks for Ni2+
2p3/2, and the satellite peaks of 873.8 and 879.4 eV correspond to Ni2+ 2p1/2, thereby
indicating the existence of NiO. The peak at 852.1 and 869.1eV correspond to
metallic Ni0 2p3/2 and metallic Ni0 2p1/2, respectively. Figure 3b shows the high
resolution XPS spectra of O1s, two distinct peaks with the binding energies of 529.3
16
This article is protected by copyright. All rights reserved.

10.1002/cssc.202001424
ChemSusChem
and 530.2 eV are observed. The O1s region is assigned to the O2− from the lattice of
TiO2 (530.3 ± 0.2 eV) and O produced by NiO (529.3 eV), which further indicates the
presence of NiO in the catalyst. [30] Figure 3c shows the XPS spectrum of Ti 2p, with
the spin orbits (2p3/2 and 2p1/2) of the peaks well deconvoluted by two curves
(approximately 458.7 and 464.3 eV, respectively). This result indicates that Ti4+ was
the main Ti species in the catalyst, which is in line with previous study. [31]
The TEM images of Ni-NiO/TiO2-Re450 (Figure S11) show that the TiO2
particles are consisted of rod-shaped and spherical particles, indicating the presence of
both rutile and anatase crystal in the catalyst. HRTEM can be used to distinguish the
lattice fringes and identify the crystallites by accurately measuring the lattice fringe
spacing. The stripe spacing of 0.352 nm shown in Figure S12(a) is in good agreement
with the distance between the planes of anatase TiO2 (101). The lattice fringe spacing
of 0.21 nm exactly equals to the lattice spacing of the NiO plane (002). [32] The stripe
pitches of 0.32 and 0.172 nm shown in Figure S12(b) correspond to the rutile TiO2
[33]
(110) crystal plane and the Ni (200) crystal plane, respectively.
In addition, from
Figure 3d, two interplanar distances, i.e. 0.242 and 0.204 nm, can be observed on the
Ni-containing particles, which were ascribed to NiO (111) and Ni (111), respectively,
indicating the formation of Ni-NiO heterojunction. According to the HRTEM image,
Ni and NiO are evenly distributed in the mixture of rutile and anatase, which could
expedite the electron transfer between radicals.
17
This article is protected by copyright. All rights reserved.

10.1002/cssc.202001424
ChemSusChem
Figure 3. XPS high-resolution spectra of Ni-NiO/TiO2-Re450; (a) the spectrum of Ni
2p and (b) the spectrum of O1s (c) the spectrum of Ti 2p; (d) HRTEM images of Ni-
NiO/TiO2-Re450
To further understand the synergistic effect of Ni-NiO in the FFA hydrogenation,
the adsorption of atomic hydrogen, H2, and FFA on the pure metal Ni and NiO are
investigated by density functional theory (DFT) calculations. Two surface models of
Ni (111) and NiO (111) are constructed according to HRTEM observations, which
represent Ni and NiO in the catalyst structure, respectively. The lowest energy levels
of H atom, H2, and FFA adsorbed on Ni (111) and NiO (111) are shown in Figure 4
and the corresponding data of adsorption energy and bond length are summarized in
Table 4. The adsorption energies of H2 on Ni (111) is -0.48 eV and that on NiO (111)
is -0.21 eV, which indicates that H2 prefers to adsorb on Ni (111) to dissociate to
atomic hydrogen. While the adsorption of H on NiO (111) surface is stronger than that
on Ni (111) surface (-1.23 eV vs. -0.89 eV), which suggests that dissociated atomic
hydrogen tends to be adsorbed on NiO (111). FFA tends to be adsorbed on NiO (111)
18
This article is protected by copyright. All rights reserved.

10.1002/cssc.202001424
ChemSusChem
with the adsorption energy of -2.60 eV, which is stronger than that on Ni (111) surface
with adsorption energy of -1.77 eV.
Table 4.The absorbed energy (eV) of H atom, H2, and FFA on Ni (111) and NiO (111)
and the corresponding bond length (Å).
Substrate
Surface
Eads(eV)
-0.48
Bond length (Å)
H-Ni:1.59
H-O:2.68
(2x2)Ni(111)
(2x2)NiO(111)
(2x2)Ni(111)
(2x2)NiO(111)
(4x4)Ni(111)
(4x4)NiO(111)
H2
H
-0.21
-0.89
H-Ni:1.69
H-O:0.97
-1.23
FFA
-1.77
O-Ni:1.92
O-Ni:2.14
-2.60
Thus, when both Ni (111) and NiO (111) are presented in the catalyst, FFA
hydrogenation may undergo the following synergistic reaction pathways. H2
preferentially adsorbs on Ni (111) and dissociates into atomic hydrogens
exothermically. Due to the thermodynamic driving force, the atomic hydrogen will be
adsorbed on NiO (111) and react with adjacent FFA adsorbed on NiO (111). In this
process, NiO (111) not only acted as a support to enhance the surface area of the
catalyst, but also provide the adsorption sites for FFA due to the affinity of Ni2+ to
[7, 34]
C=O bond of FFA, both of which could benefit FFA hydrogenation.
The
experimental and theoretical results well prove and explain the positive effect of Ni-
NiO heterojunction on the FFA hydrogenation.
19
This article is protected by copyright. All rights reserved.

10.1002/cssc.202001424
ChemSusChem
Figure 4. Adsorption structure models of H, H2 and FFA on Ni (111) and NiO (111).
According to the intermediates and final products, a possible reaction pathway is
proposed. The reaction mechanism depicted in Figure 5 indicates that FFA is firstly
[35]
converted to FA by hydrogenation at C=O under the H2 atmosphere.
After 1 h of
reaction, the yield of the key intermediate FA achieved 80% (Figure 2c). The FA yield
decreased drastically with the increase of reaction time, suggesting that one part of FA
[25, 36]
is hydrogenated on C=C bonds to produce THFA,
and the other part of FA is
subjected to rearrangement reaction to produce 4-HO-2-CPEO, a key intermediate
product of CPO. H2O is very critical during the formation of 4-HO-2-CPEO, as H3O+
can attack the 5-position of FA, leading to the opening and closing of the furan ring to
form 4-hydroxy-2-cyclopentenone (4-HO-2-CPEO), which is a crucial intermediate
product of CPO. [5a, 23, 25, 37] Then, the 4-HO-2CPEO converts to 2-cyclopentenone (2-
CPEO) by the immediate hydrogenation and dehydration reactions. As both reaction
steps are extremely fast, the highly reactive intermediates (4-HO-2-CPEO and 2-
[25, 38]
CPEO) cannot be detected in the system.
2-CPEO is further hydrogenated to
form CPO, and finally CPO is hydrogenated to form CPL.
20
This article is protected by copyright. All rights reserved.

10.1002/cssc.202001424
ChemSusChem
Figure 5. Possible Reaction Pathway of FFA Hydrogenation in Water over Ni-
NiO/TiO2-Re450.
Conclusions
In summary, the Ni-NiO/TiO2-Re450 exhibited a high catalytic performance in the
hydrogenation of FFA to CPO. FFA conversion of 100% and CPO yield of 87.4%
were achieved under mild reaction conditions (1 MPa, 140 °C, 6h), and 95.4% of FFA
conversion remained in the fifth run. Detailed characterizations, such as XRD, H2-
TPR, SEM, and HRTEM indicate that the Ni-NiO heterojunction structure and the
suitable composition of anatase and rutile of the catalyst could significantly enhance
the catalytic performance of Ni-NiO/TiO2-Re450 on FFA hydrogenation. Appropriate
reaction conditions including reaction temperature, hydrogen pressure, and reaction
time are critical to the yield of CPO. A series of characterizations, control
experiments, and theoretical calculation illustrate that the synergistic effect of Ni-NiO
heterojunction structure and the specific phase composition of the TiO2 crystal played
critical role on the high catalytic performance of Ni-NiO/TiO2-Re450 on the FFA
hydrogenation.
Experimental Section
21
This article is protected by copyright. All rights reserved.

10.1002/cssc.202001424
ChemSusChem
Materials
P25 TiO2 (TiO2 in short), which constitutes approximately 80% anatase and 20%
rutile, was purchased from Degussa Corporation (Essen, Germany). Nano anatase,
nano rutile, Ni(NO3)2·6H2O, FFA, sodium borohydride, ethyl acetate, and other
reagents were all of analytical grade. All gases (Ar and 10% H2/Ar), deionized water,
and chemicals used for catalyst preparation were not further purified.
Catalyst Preparation
The monometallic catalysts supported on TiO2 were prepared by a standard wet
impregnation method reported previously. [39] Briefly, 0.4 g TiO2 was added to 20 mL
of deionized water under stirring at room temperature. After the TiO2 was dispersed, a
solution of Ni(NO3 ) ·6H2O (0.72 mmol) was added dropwise with stirring for 3 h.
2
The mixture was evaporated in an oil bath at 100 °C and the solid residue was dried in
vacuum at 60 °C overnight. Then the solid residue was subjected to two sequential
treatment, i.e., calcination and reduction. For calcination, the precursor was calcined
at 450 °C under Ar atmosphere for 3 h with a heating rate of 2 °C min−1 to obtain
NiO/TiO2. Then, NiO/TiO2 was reduced in a 10% H2/Ar atmosphere at different
temperature, i.e., 350, 450, 550, 650, and 750 °C for 3 h with the heating rate of 2 °C
min−1, and the resultant catalysts were denoted as Ni-NiO/TiO2-Re350, Ni-NiO/TiO2-
Re450, Ni-NiO/TiO2-Re550, Ni-NiO/TiO2-Re650, and Ni-NiO/TiO2-Re750,
respectively.
To find out the active component during the catalytic process, a series of TiO2-
based catalysts were prepared. To investigate the activity of the TiO2 support, TiO2-
Cal was prepared by the calcination of TiO2 under Ar atmosphere and then reduced in
a 10% H2/Ar atmosphere at 450 °C for 3 h. To study the role of Ni-NiO
heterojunction of the catalyst, NiO/TiO2 was obtained by the abovementioned
22
This article is protected by copyright. All rights reserved.

10.1002/cssc.202001424
ChemSusChem
method, and Ni/TiO2 was obtained by reducing NiO/TiO2 with borohydride. This
reduction method was applied is because high reduction temperature under 10%
H2/Ar atmosphere may change the crystal phase of TiO2. To look into the effect of the
crystal pattern of TiO2 on the catalyst activity, Ni-NiO/anatase-Re450 and Ni-
NiO/rutile-Re450 were prepared by the same method of preparing Ni-NiO/TiO2-
Re450, except for that pure anatase or rutile nanoparticles were used as the substrate,
respectively.
Material Characterization
The X-ray diffraction (XRD) analysis of the materials was performed on an 18 kW
rotating anode X-ray diffractometer (MXPAHF, Rigaku, Japan) with a Ni-filtered Cu
Kα radiation source (30 kV/160 mǺ, λ = 0.154056 nm). During the measurement, the
sample was scanned from 20° to 80°, with a scan rate (2θ) of 0.02° s−1. Temperature-
programmed reduction (TPR) experiments were carried out on an automatic chemical
instrument (FINESORB-3010, FINETEC INSTRUMENTS, China). After being
pretreated at 120 °C for 1 h in an Ar atmosphere, the sample was subjected to the
reduction process under 10 % H2/Ar atmosphere from 100 °C to 700 °C with a
heating rate of 10 °C/min. The area under the H2-TPR curve is integrated, and the
NiO content in the catalyst is quantitatively calculated by the consumption of H2 (Text
S2). Ni loading of the catalyst was obtained by determining the Ni concentration in
the acid (concentrated nitric acid) extraction solution of the sample using inductively
coupled plasma-optical emission spectrometry (ICP-MS, Optima 7300 DV, Perkin
Elmer Corporation, USA). The X-ray photoelectron spectroscopy (XPS) spectra of the
catalysts were recorded through an X-ray photoelectron spectrometer (ESCALAB250,
Thermo-VG Scientific, UK) by using monochromatized Al Kα radiation (1486.92
eV). The specific surface areas of the samples were determined by the Brunauer,
23
This article is protected by copyright. All rights reserved.

10.1002/cssc.202001424
ChemSusChem
Emmett, and Teller (BET) method, while using the amount of nitrogen adsorbed at
temperature of −195.8 °C to calculate their pore volumes. The surface morphology of
the as-synthesized catalysts was characterized by transmission electron microscopy
(TEM; JEM-2100F, JEOL, Japan) and scanning electron microscopy (SEM; Sirion
200, FEI Electron Optics Company, USA).
FFA Hydrogenation Reaction
FFA hydrogenation was carried out using a 50 mL stainless steel-stirred autoclave. In
the experiment, 10 mL of aqueous FFA solution (1 mmol) and catalyst (50 mg) were
loaded onto the reactor. The reactor was sealed and purged with H2 five times to
exclude air, and then the reaction was conducted with 1 MPa H2 at 140 °C for 6 h,
with a stirring speed of 1000 rpm. After the reaction, the reactor was cooled to room
temperature, and the organic products were extracted by ethyl acetate from the liquid.
The products were analyzed by gas chromatography (GC, GC-1690) with a DM-5
column (30 m × 0.32 mm × 0.25 μm) and identified by a gas chromatography-mass
spectrum (GC-MS, Agilent 7890A). The reaction product was quantitatively
determined by an area normalization method and an external standard method. Text
S3 provides the specific calculation method, and the conversions and product yield
are evaluated on the basis of the amounts of FFA. The FFA hydrogenated carbon (C)
equilibrium calculation method is also given in Text S3. The stability of the catalyst in
the hydrogenation reaction was examined by the method as follows. After each
reaction, the reaction mixture was centrifuged at 6000 rpm/min, then the catalyst was
collected, dried, and subjected to the next experiment cycle.
Theoretical simulations
Spin-polarized DFT calculations are performed by Vienna ab initio simulation
package (VASP) and the plane wave basis sets are set as 400 eV.[40] Generalized
24
This article is protected by copyright. All rights reserved.

10.1002/cssc.202001424
ChemSusChem
gradient approximation (GGA-PBE) of Eenzerh, Burke, and Perdew is adopted and
the projector augmented wave (PAW) method is used to describe electron-ion
interaction. DFT-D3 method[41] is used to better describe the van de Waals interaction
between molecules and substrates. To overcome the shortcoming of the insufficient
consideration of the on-site Columbic repulsion between d electrons, the Hubbard-U
method (PBE + U) with Ueff = 5.3 eV is adopted. [42, 43]
Ni (111) surface is simulated by a four layer slab model with the bottom two
layers fixed. While the NiO (111) surface is modeled by a slab with four reconstructed
NiO double layers with bottom two NiO double layers fixed, where 3/4ML O missing
at the top layer and 1/4 ML Ni missing at the second layer in a p(2 × 2) supercell
according to previous work [44]. To calculate the adsorption energy of H2 and atomic
hydrogen, small p (2×2) supercells are adopted for Ni (111) and NiO (111) surface.
5×5×1 k-points grid is used for calculation and the geometric optimizations are
performed until the residue forces on each atom are less than 0.01 eV/Å. While to
investigate the adsorption of FFA molecule, a larger p (4× 4) supercells for Ni (111)
and NiO (111) surfaces are used. 3×3×1 k-points grid is used for calculation and the
geometric optimizations are performed until the residue forces on each atom are less
than
0.03
eV/Å.
The
adsorption
energy
is
defined
as
represent
Eads = Esurf +adsorbate − Esurf − Eadsorbate
Esurf +adsorbate Esurf
Eadsorbate
，where
,
and
total energy of the surface with adsorbate, the surface, and the adsorbate species in the
gas phase, respectively.
Acknowledgements
The authors gratefully acknowledge financial support from National Natural Science
Foundation of China (21677138, 21876166), the Key Special Program on the S&T for
25
This article is protected by copyright. All rights reserved.

10.1002/cssc.202001424
ChemSusChem
the Pollution Control, and Treatment of Water Bodies (No.2017ZX07603-003), and
the National Key Research and Development Program (Grants No.
2018YFA0208600, 2016YFA0200600). The calculations were performed on the
super-computing system in USTC-SCC, Tianjin-SCC and Guangzhou-SCC.
Keywords: furfural · cyclopentanone · Ni-NiO heterojunction · rutile/anatase
ratio·high selectivity
26
This article is protected by copyright. All rights reserved.

10.1002/cssc.202001424
ChemSusChem
[1] a) E. L. Kunkes, D. A . Simonetti, R. M. West, J. C. Serrano-Ruiz, C. A. Gärtner, J.
A. Dumesic, Sci. 2008, 322, 417-421; b) b) F.X. Zeng, W.J. Liu, H. Jiang, H.Q. Yu,
R.J. Zeng, Q.X. Guo, Bioresour. Technol. 2011, 102, 1982-1987; c) G. Chieffi, C.
Giordano, M. Antonietti, D. Esposito, J. Mater. Chem. A 2014, 2, 11591-11596.
[2] J. J. Bozell, G. R. Petersen, Green Chem. 2010, 12, 539-554; b) R. Lu, G.P.
Sheng, Y.Y. Hu, P. Zheng, H. Jiang, Y. Tang, H.Q. Yu, Biomass Bioenerg. 2011,
35, 71-678; c) Y.Y. Wang, L.L. Ling, H. Jiang, Green Chem. 2016, 18. 4032-4041.
[3] a) H. Zhang et al., Angew. Chem. Int. Ed. 2014, 53, 12132-12136; Angew. Chem.
2014, 126, 12328 –12332; b) G.-H. Wang et al., Angew. Chem. Int. Ed. 2016, 55,
11101-11105; Angew. Chem. 2016, 128, 11267 –11271.
[4] a) M. Audemar, C. Ciotonea, K. De Oliveira Vigier, S. Royer, A. Ungureanu, B.
Dragoi, E. Dumitriu, F. Jérôme, ChemSusChem 2015, 8, 1885-1891; b) X. Chang, A.-
F. Liu, B. Cai, J.-Y. Luo, H. Pan, Y.-B. Huang, ChemSusChem 2016, 9, 3330-3337; c)
H. Yu, W. Tang, K. Li, S. Zhao, H. Yin, S. Zhou, ACS Appl. Mat. Interfaces. 2019, 11,
6958-6969.
[5] a) M. Hronec, K. Fulajtárová, I. Vávra, T Soták, E. Dobročka, M. Mičušík, Appl.
Catal. B 2016, 181, 210-219; b) P. Panagiotopoulou, N. Martin, D. G. Vlachos,
ChemSusChem 2015, 8, 2046-2054; c) T. W. Goh, C.-K.Tsung, W. Huang, ACS Appl.
Mat. Interfaces. 2019, 11, 23254-23260.
[6] a) N. Perret, A. Grigoropoulos, M. Zanella, T. D. Manning, J. B. Claridge, M. J.
Rosseinsky, ChemSusChem 2016, 9, 521-531; b) S. Nandi, A. Saha, P. Patel, N.-u. H.
Khan, R.I. Kureshy, A. B. Panda, ACS Appl. Mat. Interfaces. 2018, 10, 24480-24490.
[7] S. Song, S. Yao, J. Cao, L. Di, G. Wu, N. Guan, L. Li, Appl. Catal. B 2017, 217,
115-124.
[8] a) P. Chen et al., ACS Catal. 2017, 7, 1197-1206; b) L. R. Baker, G. Kennedy, M.
27
This article is protected by copyright. All rights reserved.

10.1002/cssc.202001424
ChemSusChem
Van Spronsen, A. Hervier, X. Cai, S. Chen, L.-W. Wang, G. A. Somorjai, J. Am.
Chem. Soc. 2012, 134, 14208-14216; c) X. Jiang, Y. Zhang, J. Jiang, Y. Rong, Y.
Wang, Y. Wu, C. Pan, J. Phys. Chem. C 2012, 116, 22619-22624; d) J. Zhang, B.
Wang, E. Nikolla, J. W. Medlin, Angew. Chem. Int. Ed. 2017, 56, 6594-6598; Angew.
Chem. 2017, 129, 6694 –6698.
[9] U. Diebold. Surf. Sci. Rep. 2003, 48, 53-229.
[10] a) A. M. Ruppert, J. Grams, M. Jędrzejczyk, J. Matras-Michalska, N. Keller, K.
Ostojska, P. Saute. ChemSusChem 2015, 8, 1538-1547; b) B. Jongsomjit, T.
Wongsalee, P. Praserthdam, Catal. Commun. 2005, 6, 705-710.
[11] a) K. Lv, Q. Xiang, J. Yu, Appl. Catal. B. 2011, 104, 275-281; b) D. Chen, Q.
Zhu, F. Zhou, X. Deng, F. Li, J. Hazard. Mater. 2012, 235-236, 186-193.
[12] a) H. Zhang, J. F Banfield, J. Phys. Chem. B 2000, 104, 3481-3487; b)D. A. H.
Hanaor, C. C., Sorrell, J. Mater. Sci. 2011, 46, 855-874.
[13] a) R. Moliner, Y. Echegoyen, I. Suelves, M. J. Lázaro, J. M. Palacios, Int. J.
Hydrogen Energy. 2008, 33, 1719-1728; b) J. A. Montoya, P. del Angel, T. Viveros, J.
Mater. Chem. 2001, 11, 944-950.
[14] a) J. Zhang, Q. Xu, Z. Feng, M. Li, C. Li, Angew. Chem. 2008, 120, 1790 –1793;
Angew. Chem. Int. Ed. 2008, 47, 1766-1769; b) N. Wetchakun, B. Incessungvorn, K.
Wetchakun, S. Phanichphant, Mater. Lett. 2012, 82, 195-198.
[15] V. M. Shinde, G. Madras, AlChE J. 2014, 60, 1027-1035.
[16] B. Mile, D. Stirling, M. A. Zammitt, A. Lovell, M. Webb, J. Catal. 1988, 114,
217-229.
[17] J. Guo, G. Xu, Z. Han, Y. Zhang, Y. Fu, Q. Guo, Sustain. Chem. Eng. 2014, 2,
2259-2266.
[18] M. Hronec, K. Fulajtarová, T. Liptaj, Appl. Catal. A 2012, 437-438, 104-111.
28
This article is protected by copyright. All rights reserved.

10.1002/cssc.202001424
ChemSusChem
[19] a) H. Pan, H. Li, H. Zhang, A. Wang, D. Jin, S. Yang, Energy Convers. Manage.
2018, 166, 534-544; b) J. Long, W. Zhao, Y. Xu, W. Wu, C. Fang, H. Li, S. Yang, RSC
Adv. 2019, 9, 3063-3071.
[20] C.-Y. Liu, R.-P. Wei, G.-L. Geng, M.-H. Zhou, L.-J. Gao, G.-M. Xiao, Fuel
Process. Technol. 2015, 134, 168-174.
[21] H. Zhu, M. Zhou, Z. Zeng, G. Xiao, R. Xiao, Korean J. Chem. Eng. 2014, 31,
593-597.
[22] P. Jia, X. Lan, X. Li, T. Wang, ACS Sustainable Chem. Eng. 2019, 7, 15221-
15229.
[23] Y. Yang, Du, Z. Y. Huang, F. Lu, F. Wang, J. Gao, J. Xu, Green Chem. 2013, 15,
1932-1940.
[24] M. Zhou, J. Li, K. Wang, H. Xia, J. Xu, J. Jiang, Fuel 2017, 202, 1-11.
[25] M. Hronec, K. Fulajtárova, T. Soták, Appl. Catal. B 2014, 154-155, 294-300.
[26] X.-L Li, J. Deng, J. Shi, T. Pan, C.-G. Yu, H.-J Xu, Y. Fu, Green Chem. 2015, 17,
1038-1046.
[27] W.-T. Chen, A. Chan, D. Sun-Waterhouse, T. Moriga, H. Idriss, G. I. N.
Waterhouse, J. Catal. 2015, 326, 43-53.
[28] a) D. Reyes-Coronado, G. Rodríguez-Gattorno, M. E. Espinosa-Pesqueira, C.
Cab, R. d. Coss, G. Oskam, Nanotechnol. 2008, 19, 145605; b) A. Bouzoubaa, A.
Markovits, M. Calatayud, C. Minot, Surf. Sci. 2005, 583, 107-117.
[29] M. A. Peck, M. A. Langell, Chem. Mater. 2012, 24, 4483-4490.
[30] a) Y. J. He, J. F. Peng, W. Chu, Y. Z. Li, D. G Tong, J. Mater. Chem. A 2014, 2,
1721-1731; b) J. L Rodríguez, T. Poznyak, M. A. Valenzuela, H. Tiznado, I. Chairez
Chem. Eng. J. 2013, 222, 426-434; c) B. P. Payne, M. C Biesinger, N. S McIntyre, J.
Electron. Spectrosc. Relat. Phenom. 2009, 175, 55-65.
29
This article is protected by copyright. All rights reserved.