Fabrication of Ni3N nanorods anchored on N-doped carbon for selective semi-hydrogenation of alkynes

Article abstract of DOI:10.1016/j.jcat.2019.12.005

Nickel is a highly active catalyst for the semi-hydrogenation of alkynes. However, the low selectivity of the alkene product caused by the over-hydrogenation reaction on Ni has hindered its practical applications. In this work, we report a new nickel nitride (Ni₃N)-catalyzed semi-hydrogenation of alkynes to the corresponding alkenes. The Ni₃N nanorods were facilely fabricated via a direct pyrolysis of the solid mixture of nickel acetate tetrahydrate and melamine (Mlm). The Ni₃N phase in the optimum catalyst (Ni₃N/NC-6/5-550) is shown to be effective and stable in the semi-hydrogenation of alkynes, with a high yield and good selectivity for alkenes (Z/E ratios up to >99/1). Both terminal and internal alkynes bearing a broad scope of functional groups are readily converted into alkenes with good chemo- and stereoselectivity. Notably, it was found that the over-hydrogenation can be markedly suppressed even at high conversion of alkyne. Density functional theory (DFT) calculations reveal that the low interaction between the alkene product and the Ni₃N might plays a critical role in the selectivity enhancement.

Full text of DOI:10.1016/j.jcat.2019.12.005

Journal of Catalysis 382 (2020) 22–30
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Fabrication of Ni N nanorods anchored on N-doped carbon for selective
3
semi-hydrogenation of alkynes
a
a,
⇑
a
a
b,
⇑
a
a
a,
⇑
Xiaozhen Shi , Xin Wen , Shilin Nie , Jie Dong , Jingde Li , Yongqing Shi , Huiling Zhang , Guoyi Bai
a
Key Laboratory of Chemical Biology of Hebei Province, College of Chemistry and Environmental Science, Hebei University, Baoding 071002, China
National-Local Joint Engineering Laboratory for Energy Conservation of Chemical Process Integration and Resources Utilization, School of Chemical Engineering and
b
Technology, Hebei University of Technology, Tianjin 300130, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Nickel is a highly active catalyst for the semi-hydrogenation of alkynes. However, the low selectivity of
the alkene product caused by the over-hydrogenation reaction on Ni has hindered its practical applica-
Received 13 August 2019
Revised 27 November 2019
Accepted 4 December 2019
tions. In this work, we report a new nickel nitride (Ni
the corresponding alkenes. The Ni N nanorods were facilely fabricated via a direct pyrolysis of the solid
mixture of nickel acetate tetrahydrate and melamine (Mlm). The Ni N phase in the optimum catalyst
Ni N/NC-6/5-550) is shown to be effective and stable in the semi-hydrogenation of alkynes, with a high
3
N)-catalyzed semi-hydrogenation of alkynes to
3
3
(
3
Keywords:
yield and good selectivity for alkenes (Z/E ratios up to >99/1). Both terminal and internal alkynes bearing
a broad scope of functional groups are readily converted into alkenes with good chemo- and stereoselec-
tivity. Notably, it was found that the over-hydrogenation can be markedly suppressed even at high con-
version of alkyne. Density functional theory (DFT) calculations reveal that the low interaction between
Nickel Nitride
Semi-Hydrogenation
Alkynes
Z-Alkenes
N-Doped Carbon
the alkene product and the Ni
3
N might plays a critical role in the selectivity enhancement.
Ó 2019 Elsevier Inc. All rights reserved.
1
. Introduction
Catalytic semi-hydrogenation of alkynes to alkenes is an impor-
Ni + P, Ni + Ga, Ni + Sn, Ni + B, Ni + Al, Ni + Zn, Ni + In, and
Ni + Tl [49–56]. However, the doping effect of the second element
on the catalytic performance was much less discussed. For exam-
ple, Corma and co-workers reported that the selectivity for cis-
alkene can be tuned by the P atom in the nickel phosphide
nanoparticles, which can be attributed to the blocking of the unsat-
urated Ni sites by P atoms [53]. By comparing two kinds of nickel
tant chemical process in fine chemical intermediates production
and alkenes purification industry [1–9]. For decades, Pd-based cat-
alysts have dominated this transformation [10–22], and the most
widely used catalyst is known as Lindlar catalyst (Pd/CaCO
3
poi-
soned with Pb(OAc) and quinoline). However, the high cost of
2
2 5 4
phosphide catalyst, namely Ni P and Ni P , Albani and co-workers
Pd, toxic lead additives, and over-hydrogenation have limited its
practical application for alkene production. Replacement of Pd-
based catalyst with the non-precious metals (e.g. Ni, Co, Cu, Fe)
has captured increasing attention due to the cost performance
and resource availability [23–37]. One promising class of non-
precious metal catalysts for the semi-hydrogenation of alkynes is
Ni-based catalysts [38–42], and recently graphitic shell encapsu-
lated Ni [42], single Ni atoms [43], ligand-free Ni nanoparticles
demonstrated that the formation of spatially-isolated nickel tri-
mers was responsible for the selectivity enhancement [50]. In a
work carried out by Laursen et al., some strategies for designing
the intermetallic compound catalyst composed of Ni and the boron
group elements were reported [52]. Zhou and coworkers have
demonstrated that the NiZn and NiGa alloys on AlSBA-15 showed
high selectivity for alkene due to the geometric and electronic
effects of the second element [56].
[
44], confined cationic nickel [45], have been reported to show
Nowadays, despite these achievements, the Ni-based catalysts
still suffer from the low alkene selectivity or over-hydrogenation
issues, which represent the key obstacle for alkynes semi-
hydrogenation. It has been reported that the over-hydrogenation
of alkene may be ascribed to two reasons: (i) the strong adsorption
energy of alkene prolongs its resident time on the catalyst surface,
increasing its likelihood for further hydrogenation reaction [57];
excellent catalytic activity towards alkynes semi-hydrogenation.
Particularly, doping mono-metal catalysts with second element
often leads to unique properties in the semi-hydrogenation of alky-
nes with respect to conventional catalyst [10–16,46–49], such as
⇑
Corresponding authors.
E-mail addresses: wenxin767@hotmail.com (X. Wen), jingdeli@hebut.edu.cn
(
ii) after the full conversion of alkyne, the accelerated over-
(
J. Li), baiguoyi@hotmail.com (G. Bai).
hydrogenation always takes place due to the lack of competitive
https://doi.org/10.1016/j.jcat.2019.12.005
021-9517/Ó 2019 Elsevier Inc. All rights reserved.
0

X. Shi et al. / Journal of Catalysis 382 (2020) 22–30
23
adsorption of alkyne species on the Ni active sites [58]. Such cat-
alytic systems always require a strict-time monitoring especially
at the high alkyne conversion. Accordingly, the optimum semi-
hydrogenation catalysts for alkenes production should have the
following features: (i) active for alkynes adsorption and hydro-
genation; (ii) weak interaction with alkenes and higher activation
energy barrier for alkenes hydrogenation. Thus, the searching for
new efficient and practical catalytic system with these features is
essential for synthesizing high-purity alkenes.
tetrahydrate and Mlm (Scheme 1). The effect of mass ratios of
nickel acetate tetrahydrate and Mlm (Ni:Mlm = a/b), as well as
the pyrolysis temperatures (T), were investigated. The resulting
Ni catalysts prepared under different conditions are denoted as
‘‘Ni/NC-a/b-T” or ‘‘Ni
Detailed information on the preparation of reference catalysts
(Ni/SiO , Ni/AC and Ni N) are displayed in the Supplementary
Material.
3
N/NC-a/b-T”, accordingly.
2
3
Nickel nitride (Ni
has been widely studied as electrocatalyst [59,60]. However, the
application of Ni N in the heterogeneous catalysis is much less dis-
3
N) is an interstitial metallic compound, which
2
.3. Catalyst characterization
3
The Ni content of the catalyst was measured by inductively cou-
cussed [61,62]. The Ni atoms with the low plus valent (0.23 elec-
trons) are isolated by N in the interstices. The strong interaction
between Ni and N alters the electronic structure and the contrac-
tion of d-bands of Ni results in higher electron density near Fermi
level [63]. On the one hand, the isolation of Ni active sites by the N
atoms leads to a high barrier for the segregation of Ni atoms and
the subsequent hydride formation, which may inhibit the over-
hydrogenation. On the other hand, the absorption energy of alkyne
and alkene can also be tuned by the second elements. In this con-
pled plasma optical emission spectrometry (ICP-OES) using a Vista
MPX spectrometer. CO-chemisorption was performed on
Micromeritics AutoChem II 2920 chemisorber. X-ray diffraction
XRD) patterns were obtained using a Bruker D8-ADVANCE X-ray
a
(
diffractometer using Cu Ka radiation at a scan step of 0.02°. Scan-
ning electron microscopy (SEM) images were performed on a FEI
Inspect F50 electron microscope. Transmission electron micro-
scopy (TEM) images were acquired on a FEI Tecnai G2 F20 S-
TWIN instrument at a voltage of 200 kV. The samples were first
dispersed in ethanol and then deposited on a carbon-coated double
copper grid, dried at 80 °C for 4 h. X-ray photoelectron spec-
troscopy (XPS) was recorded on a PHI 1600 spectrometer using
Mg Ka X-ray source for excitation, in which the hydrocarbon C 1s
line at 284.8 eV from adventitious carbon was used for energy ref-
erencing. The NMR spectra were measured with a Bruker DRX
text, we proposed that Ni
alkyne semi-hydrogenation.
Herein, we report a straightforward protocol for the facile
preparation of Ni N nanorods supported on N-doped carbon (NC)
via the direct pyrolysis of a mixture of nickel salt and melamine
Mlm). Notably, Ni N nanorods can be in situ formed during the
pyrolysis process under optimal conditions [Ni(OAc) /Mlm, 6/5;
3
N might be an efficient catalyst for
3
(
3
2
6
00 MHz spectrometer using tetramethylsilane as the reference
pyrolysed at 550 °C], constituting a novel metallic nitride catalyst
for the semi-hydrogenation of a variety of alkynes to their corre-
sponding (Z)-alkenes. To the best of our knowledge, this is the first
time that Ni N is introduced as an efficient catalyst for alkyne
3
semi-hydrogenation. Moreover, density functional theory (DFT)
1
13
compound for H and C NMR. The reaction mixtures were anal-
ysed using Agilent 7820A gas chromatography with a 30 m HP-5
capillary column. Gas chromatography–mass spectrometry (GC–
MS) was used to identify the products on a Thermo TRACE-
1
300GC-ISQ-LT instrument.
calculations confirmed that the improved selectivity for Z-alkene
(
3
1,2-diphenylethyne) over Ni N/NC can be ascribed to its
2
.4. Semi-hydrogenation of alkynes
decreased adsorption energy and increased hydrogenation energy
barriers when compared with those of Ni, providing a fundamental
understanding of the hydrogenation mechanism at atomic level.
Catalytic tests of Ni
3
N/NC or Ni/NC were performed as follows:
N/NC catalyst
1
,2-diphenylethyne (5.38 mmol), Ni/NC or Ni
3
(
7.8 mol%, based on ICP results) and solvent (70 mL) were mixed
2
. Experimental and computational details
in a stainless-steel autoclave (100 mL) equipped with an electric
heating system and a magnetically driven mechanical stirrer. The
reaction system was flushed with hydrogen three times and pres-
surized with hydrogen (2.0 MPa), and then heated to 100 °C for
hydrogenation. After the completion of reaction, it was allowed
to cool to room temperature and the residual hydrogen was
released. The solvent was removed under reduced pressure, and
the crude products were purified by column chromatography on
silica gel to withhold the catalyst.
2.1. Reagents and materials
Unless otherwise noted, all chemicals were of analytical reagent
grade and used without any pre-treatment. Nickel(II) acetate
tetrahydrate [Ni(AcO) O], nickel nitrate hexahydrate [Ni
ꢀ4H
NO O] and Mlm were purchased from Aladdin Reagent Co.
ꢀ6H
Ltd.. 1-(Phenylethynyl)-4-(trifluoromethyl) benzene, 1-fluoro-4-
phenylethynyl) benzene, 1-methyl-4-(phenylethynyl)benzene,
2
2
(
3
)
2
2
(
For the substrate scope, the semi-hydrogenation of alkynes was
conducted using a similar method. For a typical run: the alkyne
(0.3 mmol), 7.8 mol% Ni N/NC-6/5-550 and ethanol (3 mL) were
and 1-(4-(phenylethynyl)phenyl)ethan-1-one were synthesized
according to a reported method [64].
3
placed in a glass vial (4 mL) with cap, septum, and needle, and then
two glass vials were sealed in a stainless steel autoclave (100 mL)
equipped with an electrical heating system, where a certain
amount of ethanol was added to keep the same horizontal level
of ethanol inside and outside the glass vials. Then, the reaction sys-
tem was treated following the same steps as described above.
The turnover frequency (TOF), which was defined as alkyne
consumption per total active site per second, was calculated as
follows:
2
.2. Catalyst preparation
The Ni N/NC or Ni/NC catalysts were prepared using a basic
3
pyrolysis process. In a typical run, the desired ratio of nickel acet-
ate tetrahydrate and Mlm was first mixed thoroughly using an
electric pulverizer. Then, the mixture (4.0 g) was pyrolysed in a
furnace using a temperature-programmed method under a high
ꢁ1
purity N
2 2
atmosphere (total N flow of 300 mL min ) at room
temperature for 10 min. Afterwards, the temperature was raised
A
WNt
a
ꢁ1
(
(
2.3 °C min ) to the desired temperature at low N
2
flow rates
31 mL min ), held for 4 h, and then cooled to room temperature.
N/NC or Ni/NC catalysts were prepared by a solvent-free
method via the direct pyrolysis of a solid mixture of nickel acetate
TOF ¼
ꢁ1
The Ni
3
where A is the amount of alkyne in the feedstock, mole;
conversion of alkyne, %; W is the catalyst weight, g; and N is the
a
is the

2
4
X. Shi et al. / Journal of Catalysis 382 (2020) 22–30
3
Scheme 1. Fabrication of Ni N/NC nanohybrids for the selective semi-hydrogenation of alkynes.
number of active sites determined by CO chemisorption, mol gꢁ1; t
(1 1 3) crystal planes of the Ni N phase (JCPDS No. 10-0280 and
3
is the reaction time, s.
Table S1) [62,69]. The results suggested that, at 550 °C, the Ni N
3
phase can only be produced at a low Ni:Mlm ratio. Fig. 1b shows
the XRD patterns of the catalysts developed at different pyrolysis
temperatures but with a constant mass ratio (Ni:Mlm) of 6/5.
2.5. Computational method
One can see that the Ni
50 °C, presumably due to the decomposition of the Ni
higher pyrolysis temperatures [61].
Fig. 2a and 2b show the SEM and TEM images of the optimum
N/NC-6/5-550 catalyst, respectively. One can clearly see that
the Ni N nanorods were supported on the platelike g-C . As pre-
3
N phase can only be obtained at or below
The DFT calculations were performed using the VASP package
5
3
N phase at
[
65,66]. The Perdew-Burke-Ernzerhof (PBE) functional was used
for the exchange and correlation energy terms. The Ni and Ni
catalysts were modeled using their (1 1 1) surfaces. The Ni
1 1 1) and Ni
N (1 1 1) surfaces were modeled using a (7 ꢂ 5)
3
N
Ni
3
(
3
and (2 ꢂ 2) unit cell with four Ni-atomic layer slabs, respectively.
The vacuum height was set to 15 Å. The bottom Ni layer was fixed
in their bulk positions, whereas the remaining layers and the
adsorbates were allowed to relax. A planewave cutoff energy was
set to be 400 eV. The k-space was sampled using a 2 ꢂ 2 ꢂ 1 Mon-
khorst–Pack grid. Structures were fully relaxed until the forces act-
ing on the atoms were smaller than 0.03 eV/Å. On the basis of fully
relaxed structures, adsorption energy (Eads) of the reactant and two
intermediate surface species were calculated by the following
equation:
3
3 4
N
sented in the high-resolution TEM (HRTEM) image (Fig. 2c), the
nanorods exhibited a fringe space of 0.203 nm which can be
indexed to the (1 1 1) lattice of Ni
formation of Ni N. The elemental mapping from energy-
dispersive X-ray spectrometry (EDX) revealed that the Ni, N, C
and O elements were uniformly distributed throughout the Ni N/
3
N [70], further confirming the
3
3
NC-6/5-550 (Fig. 2d). Notably, by increasing the Ni:Mlm to 7/5 or
the pyrolysis temperatures to 650 °C, the obtained Ni catalysts
showed a clear core-shell structure (Figs. S1 and S2). The Ni cores
have a rather broad size distribution and were embedded within
E
ads ¼ ꢁðEX=slab ꢁ E
X
ꢁ EslabÞ
the graphite-like C
tune the Ni N or Ni electronically and hence are crucial for the
selectivity enhancement in the semi-hydrogenation of 1a.
To further verify the formation of Ni N phase in Ni N/NC-6/5-
50, the chemical state of the catalyst was investigated by XPS
Fig. 3a and b). In the Ni 2p3/2 spectrum (Fig. 3a), the characteristic
peak at the binding energy (BE) of 852.9 eV can be attributed to
3 4 3 4
N shells. The formed C N shells might also
3
where EX=slab represents the total energy of the adsorbate X bound to
the Ni (1 1 1) and Ni N (1 1 1) slabs, E is the energy of adsorbate
alone, Eslab is the energy of the optimized Ni (1 1 1) or Ni
1 1 1) slab before adsorption. A more positive Eads corresponds to
a more stable adsorbate/slab system. The reaction transition states
TS) are determined using the climbing image nudged elastic band
3
X
3
3
3
N
5
(
(
(
+
3
monovalent Ni [71], indicative for the presence of Ni N. The fitted
method (CI-NEB) and the TS was confirmed by its associated one
imagery frequency [67].
peak located at 855.0 eV can be ascribed to the Ni 2p3/2 of NiO,
which was probably generated due to the surface atmospheric oxi-
dation of Ni
3
N [63]. Moreover, the peaks at 859.5 and 862.2 eV
3
. Results and discussion
(
denoted as ‘‘Sat”) can be assigned to the corresponding satellite
+
2+
peaks of Ni and oxygen-binding Ni
Fig. 3b, the N 1s spectrum can be deconvoluted into four N-
, respectively [70]. In
3.1. Catalyst characterizations
bonding configurations with binding energies of 397.2, 398.3,
To understand the catalytic performance of these novel materi-
3
99.3 and 400.9 eV. According to the previous report [15], the peak
at 397.2 eV can be assigned to the N in Ni N. The other peaks at
98.3, 399.3 and 400.9 eV can be attributed to the N of pyridinic
als, detailed characterizations of selected catalysts were done.
Fig. 1a shows the XRD patterns of the obtained catalysts prepared
at 550 °C under different mass ratios (Ni:Mlm), 7/5, 6/5, and 5/5,
respectively. For all samples, a broad diffraction peak at about
3
3
3 4
N, pyrrolic N and graphitic N in the g-C N phase, respectively [72].
2
C
6.4° was observed, corresponding to the (0 0 2) crystalline g-
formed during the thermal condensation process of Mlm
3
N
4
3.2. Optimization of reaction conditions
[
68]. In the cases of the pyrolysis temperature at 550 °C (Fig. 1a),
the peaks in Ni/NC-7/5-550 at 44.5°, 51.8° and 76.4° can be
assigned to the (1 1 1), (2 0 0) and (2 2 0) planes of the crystalline
Ni (JCPDS No. 87-0712), suggesting that the nickel acetate decom-
posed to form metallic Ni during the pyrolysis process. In contrast,
no peak related to crystalline Ni was observed in the samples of
Ni N/NC-6/5-550, and Ni N/NC-5/5-550. Instead, six new peaks
3 3
appeared at 39.2°, 41.6°, 44.6°, 58.5°, 71.1° and 78.0°, which were
Table 1 summarized the catalytic performance of the as-
prepared Ni N/NC or Ni/NC catalysts in the semi-hydrogenation
of 1,2-diphenylethyne 1a. The results showed that Ni N/NC-6/5-
550 (Table 1, entry 2) exhibited the highest conversion (99%) and
excellent selectivity (98%) for (Z)-stilbene 2a in ethanol using
molecular hydrogen (2.0 MPa) as hydrogen source at 100 °C. In
case of higher mass ratio 7/5 (Table 1, entry 1), both conversion
and selectivity were lower than those of 6/5. In contrast, in case
3
3
in agreement with the (1 1 0), (0 0 2), (1 1 1), (1 1 2), (3 0 0) and

X. Shi et al. / Journal of Catalysis 382 (2020) 22–30
25
Fig. 1. XRD patterns of the as-developed catalysts prepared under different conditions (a) prepared at 550 °C under different mass ratios (Ni:Mlm): 7/5, 6/5, and 5/5; (b)
prepared under the mass ratio (Ni:Mlm) of 6/5 at different pyrolysis temperatures: 850, 750, 650, 550 and 450 °C.
3
Fig. 2. (a) SEM, (b) TEM, (c) HRTEM images and (d) elemental maps of the Ni N/NC-6/5-550 catalyst.
of lower mass ratios 5/5 (Table 1, entry 3), the selectivity slightly
increased to 99%, but the conversion of 1a decreased to 81%.
Although the catalysts pyrolysed at 650 °C (Table 1, entry 5) and
the precursor (Fig. S1), or the increasing of pyrolysis temperature
(Fig. S2), which can increase the exposure of the activity sites,
and thus enhance the activity. However, Ni/NC-6/5-850 showed a
low activity due to the severe aggregation of Ni nanoparticles
(Fig. S2).
7
50 °C (Table 1, entry 6) were active, the selectivities for stilbene
and the Z/E ratio were much lower than those of Ni N/NC-6/5–
50. Furthermore, those catalysts pyrolysed at even lower
450 °C, Table 1, entry 4) and higher (850 °C, Table 1, entry 7) tem-
3
5
(
In order to compare the selectivity for alkene, control experi-
ments were carried out using NC support, SiO
(Ni/SiO ), activated carbon supported Ni (Ni/AC) and unsupported
Ni N as reference catalysts (Table 1, entries 8-11). In the absence of
the Ni N or Ni phase, the NC support gave the lowest conversion of
only 4% in 5 h (Table 1, entry 8). Although Ni/SiO and Ni/AC
2
supported Ni
peratures gave significantly lower conversion in this transforma-
tion. With respect to the TOF value, a negative correlation was
observed between the activity and the feed ratio of Mlm in the
mixture of nickel acetate tetrahydrate and Mlm (Table 1, entry 1-
2
3
3
2
3
). In contrast, the activity showed a positive correlation with the
showed high activities in the semi-hydrogenation of 1a with the
temperature (Table 1, entry 2, 4-7). These phenomena can be
attributed to the fact that the carbon layer on the catalyst surface
gradually thinned with the decreasing of the feed ratio of Mlm in
conversion up to 99%, the selectivity for stilbene was much lower
than that of Ni
3
N/NC-6/5-550 (Table 1, entries 9 and 10). Using
pure Ni N, the selectivity for stilbene was increased to 71% with
3

2
6
X. Shi et al. / Journal of Catalysis 382 (2020) 22–30
3
Fig. 3. XPS spectra of (a) Ni 2p3/2, and (b) N 1s of the Ni N/NC-6/5-550 catalyst.
Table 1
Semi-hydrogenation of 1,2-diphenylethyne over Ni-based catalysts.
Entry
Catalyst
Ni loading of catalyst (wt %)
Conv.a (%)
Sel. For stilbeneb (%)
Z/Ec
TOFd (sꢁ1
)
1
2
3
4
5
6
7
8
Ni/NC-7/5–550
51.0
36.3
20.7
11.4
55.2
63.8
71.3
–
89
99
81
71
99
97
8
90
98
99
96
89
97
96
98
37
22
71
89/11
98/2
98/2
97/3
91/9
95/5
94/6
95/5
91/9
95/5
93/7
0.293
0.110
0.067
0.034
0.308
0.975
0.059
–
Ni
3
N/NC-6/5–550
N/NC-5/5–550
N/NC-6/5–450
Ni
Ni
3
3
Ni/NC-6/5–650
Ni/NC-6/5–750
Ni/NC-6/5–850
NC
4
e
9
Ni/SiO
Ni/AC
2
14.3
14.3
–
97
99
99
–
–
–
1
1
0f
1
Ni
3
N
Reaction conditions: 1a (5.38 mmol), 7.8 mol% Ni (based on ICP results), ethanol (70 mL), 5 h, 100 °C, initial P(H
2
) = 2.0 MPa.
a
Determined by GC.
Selectivity for (Z)- and (E)-stilbene.
Relative percent ratio of (Z)-stilbene 2a and (E)-stilbene 3a.
The TOF was calculated based on conversion lower than 30% and the number of active sites (Table S2).
b
c
d
e
2
2
mol% Ni, 80 °C, 10 min.
mol% Ni, 80 °C, 3 min.
f
a 99% conversion (Table 1, entry 11), verifying the enhanced selec-
tivity for stilbene over the Ni
port in the Ni N catalyst (Ni
3
N catalyst. After introducing NC sup-
N/NC-6/5-550, Table 1, entry 2), the
3
3
selectivity was further increased to 98% with a 99% conversion,
indicating that the NC support is also beneficial for the selectivity
enhancement in alkyne semi-hydrogenation [27,48,73]. In compar-
ison with reference catalysts, the selectivity for alkene at high con-
version follows the order of Ni
of solvents was also examined with the optimum catalyst Ni
3
N/NC > Ni
3
N > Ni. In addition, effect
N/
3
NC-6/5-550. It was found that the semi-hydrogenation of 1a pro-
ceeded sluggishly in a range of solvents, including N,N-dimethyl
formamide, acetonitrile, tetrahydrofuran, toluene, ethyl acetate
and dioxane, compared to ethanol (Table S3). Combined, these
experimental results demonstrated that the Ni
3
N phase is the most
active species in the selective semi-hydrogenation of 1a.
3.3. Kinetic profile and DFT study
3
Fig. 4. Time course of the semi-hydrogenation of 1a over Ni N/NC-6/5-550.
The kinetic profile for the semi-hydrogenation of 1a over
N/NC-6/5-550 is shown in Fig. 4. It can be seen that Ni N/NC-
3 3
/5-550 afforded a 99% conversion and 96% selectivity for 2a
within 5 h. Notably, high chemo- and stereoselectivity was
maintained even during prolonging reaction times up to 12 h.
evaluated by a control experiment (Scheme 2), in which a mixture
of 1a (4%) and a large excess of 2a (96%) was hydrogenated at
Ni
6
1
2 3
00 °C under 2.0 MPa of H in the presence of Ni N/NC-6/5-550.
The results showed that only trace amounts of 3a isomer and
over-hydrogenated product 4a were detected after full conversion
3
The excellent catalytic performance of Ni N/NC-6/5-550 was also

X. Shi et al. / Journal of Catalysis 382 (2020) 22–30
27
3
Scheme 2. Control experiment: the hydrogenation of a mixture of 1a and 2a over Ni N/NC-6/5-550.
of 1a, demonstrating that the isomerization and over-
hydrogenation of 2a were inhibited in this catalytic system involv-
3
for which the energy barrier on Ni N (1 1 1) was 1.44 eV higher
than that on Ni (1 1 1). The high adsorption energy and the rela-
tively low energy barrier on Ni (1 1 1) surface indicated that Ni
3
ing Ni N/NC-6/5-550. The superior chemo- and stereoselectivity of
Ni N/NC-6/5-550 makes it an excellent candidate for the purifica-
tion of alkenes by the selective semi-hydrogenation of any alkyne
impurity, without the need for strict reaction monitoring.
3
was more active for the hydrogenation reaction than Ni
ever, this would also lead to over-hydrogenation during reaction.
On the other hand, the reaction kinetics on Ni N (1 1 1) might be
3
N. How-
3
In order to clarify the role of Ni
ment in alkyne semi-hydrogenation, DFT calculations were per-
formed. Specifically, using 1,2-diphenylethyne (1a)
3
N in the selectivity enhance-
slightly slow. However, the low adsorption energy of stilbene indi-
cated that, upon the formation of stilbene product, it would release
from the Ni N surface, making it a promising catalyst towards the
3
hydrogenation as the model reaction, the adsorption energy of
intermediate species produced during its semi-hydrogenation on
Ni (1 1 1) and Ni N (1 1 1) surfaces were calculated (Fig. 5). These
3
intermediate species include the first (Z-int and E-int) and second
hydrogenation products (Z-stilbene and E-stilbene). The results
showed that, for the 1,2-diphenylethyne reactant, its adsorption
on Ni (1 1 1) surface (1.81 eV) was much stronger than that
semi-hydrogenation of alkynes. These calculation results agree
well with the experimental observations. As shown in Table 1,
the hydrogenation reaction proceeded rapidly on Ni (10 min).
However, the selectivity for stilbene production is higher on
3
Ni N-based catalyst (71-98%) than Ni (37%).
3.4. Substrate scope
(
0.23 eV) on Ni
for the hydrogenation reaction, and in accordance with the result
of control experiment (Ni/SiO ). Moreover, Ni N (1 1 1) exhibited
3
N (1 1 1), suggesting that Ni might be more active
With the most active catalyst and optimal conditions in hand,
2
3
the semi-hydrogenation of various alkynes was also studied to
evaluate the versatility of Ni N/NC-6/5 catalyst. The results sum-
lower adsorption towards the intermediate Z-int, E-int, Z-stilbene
and E-stilbene species as well. For example, on Ni (1 1 1), the
semi-hydrogenation product Z-stilbene and E-stilbene has adsorp-
3
marized in Table 2 confirm that a broad range of functional groups,
such as fluoro, chloro, bromo, cyano, trifluoromethyl, methyl, ethyl,
propyl, alkyl, thienyl, ester and hydroxy, in both terminal and
internal alkynes are tolerated in this semi-hydrogenation protocol.
Terminal alkynes with either phenylic or aliphatic substituents can
be smoothly hydrogenated to afford the corresponding alkenes
exclusively with up to > 99% conversion (Table 2, entries 1-8). A
substrate bearing a thienyl substituent displayed low reactivity
under standard conditions, probably due to the coordination with
the metal catalyst [74], but still afforded a 91% conversion with
tion energies of 0.99 and 2.50 eV, respectively. However, on Ni
1 1 1), their corresponding Eads was only 0.10 and 0.31 eV, respec-
tively. The lower adsorption energies of stilbene products observed
on Ni N (1 1 1) indicated that desorption of stilbene into gas phase
3
N
(
3
was preferred, suppressing its further hydrogenation and therefore
improve the selectivity of stilbene. Moreover, the hydrogenation
kinetics of 1,2-diphenylethyne towards stilbene production were
also investigated. Fig. 6 presents the transition states (T.S) and cor-
responding activation energies for the forward/backwards of the
hydrogenation reaction on Ni (1 1 1) and Ni N (1 1 1) surfaces.
3
Overall, the energy barriers of these hydrogenation steps were very
similar on both surfaces except for the Z-stilbene production step,
9
3% selectivity while prolonging the reaction time to 11 h (Table 2,
entry 8). On the other hand, a series of internal alkynes was
readily hydrogenated to the desired (Z)-alkene products.
Aromatic substrates containing both electron-donating and
3
Fig. 5. The most stable adsorption configurations of the surface species invloved in 1,2-diphenylethyne hydrogenation reaction on Ni (1 1 1) and Ni N (1 1 1) surface models.
Orange (Ni), gray (C), blue (N), white (H).

2
8
X. Shi et al. / Journal of Catalysis 382 (2020) 22–30
Fig. 6. Geometric structures of the transition state and corresponding forward/backward activation energies of the species and elementary reactions in 1,2-diphenylethyne
semi-hydrogenation reaction on Ni (1 1 1) and Ni N (1 1 1) surface models. Orange (Ni), gray (C), blue (N), white (H).
3
Table 2
Nitride nickel-catalysed semi-hydrogenation of various alkynes.
Entry
1
Substrate
Time (h)
1
Conv.a (%)
>99
Sel. For alkene (%)b
87
Z/Ec
–
2
3
4
5
1.5
4
>99
92
92
94
85
87
–
–
–
–
4
>99
96
2
6
7
1.3
>99
>99
83
80
–
–
0.83
8
9
11
5
91
93
98
95
95
97
–
99
98/2
95/5
96/4
91/9
1
1
1
0
4.7
2
>99
96
1
2d
11
>99
1
1
1
3
4
5
9.5
14.5
7
44
>99
>99
97
96
95
96/4
95/5
96/4
d
e
1
1
6
7
6
3
>99
>99
94
89
97/3
95/5
1 f
8
3.7
2
>99
95
82
83
93/7
93/7
1
9
2
2
0
1
10
11
76
>99
>99
99
>99/1
>99/1
e
2
2
2
>99
96
>99/1
3 2
Reaction conditions: substrate (0.3 mmol), 7.8 mol% Ni N/NC-6/5–550, ethanol (3 mL), 100 °C, initial P(H ) = 2.0 MPa.
a
The conversion and selectivity were determined by GC.
b
c
d
e
f
Selectivity for (Z)- and (E)-stilbene.
Relative percent ratio of (Z)-alkene and (E)-alkene.
o
1
1
20 C, initial P(H
2
) = 3.0 MPa.
) = 4.0 MPa.
o
20 C, initial P(H
2
Ethyl (Z)-3-phenylacrylate was detected due to the transesterification.

X. Shi et al. / Journal of Catalysis 382 (2020) 22–30
29
Fig. 7. (a) Influence of cycle number on the catalytic performance of Ni
3
N/NC-6/5-550; (b) TEM image and XRD pattern (inset) of recycled Ni
3
N/NC-6/5-550.
electron-withdrawing groups (Table 2, entries 9-18) provided the
corresponding (Z)-alkene products in good conversion and with
satisfactory selectivity. In particular, ketones were compatible with
the reaction conditions (Table 2, entry 13), and no homologous
product was detected even under harsh conditions (Table 2, entry
Acknowledgements
The authors thank Professor Laurence M. Harwood for his kind
help in preparing this manuscript. Financial support by the
National Natural Science Foundation of China (21676068,
21706049 and 21376060), the Natural Science Foundation of Hebei
Province (B2016418005, B2019201341) and the Hebei University
construction project for comprehensive strength promotion of
Midwest colleges and universities is gratefully acknowledged.
1
4). In contrast, substrates bearing an ester group afforded rela-
tively lower alkene selectivities (Table 2, entries 17–19). Notably,
the reaction of aliphatic alkynes, either dialkylacetylenes or alky-
nols, proceeded smoothly to provide the corresponding alkenes
in excellent yield with Z/E ratios above 99:1 (Table 2, entries 20
and 21), probably due to their difficulty in accessing the active site
Appendix A. Supplementary material
of Ni
demonstrate the general applicability of the Ni
toward the semi-hydrogenation of a range of alkynes.
3
N on the polar support [75]. In summary, the above results
3
N/NC-6/5 catalyst
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.jcat.2019.12.005.
3.5. Stability test
References
Finally, the recoverability and reusability of the Ni
50 catalyst was evaluated in the semi-hydrogenation reaction of
a. The used catalyst can be readily isolated by precipitation, and
then be successfully cycled for 5 successive runs without loss of
its initial catalytic activity (Fig. 7a). In addition, the lattice param-
3
N/NC-6/5-
[1] K. Chernichenko, A. Madarász, I. Pápai, M. Nieger, M. Leskelä, T. Repo, Nat.
Chem. 5 (2013) 718–723.
5
1
[
[
2] R. Chinchilla, C. Nájera, Chem. Rev. 114 (2014) 1783–1826.
3] M. Crespoquesada, F. Cárdenaslizana, A.L. Dessimoz, L. Kiwiminsker, ACS Catal.
2 (2012) 1773–1786.
[
4] J.A. Delgado, O. Benkirane, C. Claver, D. Curulla-Ferré, C. Godard, Dalton Trans.
46 (2017) 12381–12403.
eter and morphology of the recycled Ni
unchanged after the recycling process (Fig. 7b), indicating the good
stability of Ni N catalyst.
3
N/NC-6/5-550 remain
[
5] C. Oger, L. Balas, T. Durand, J.-M. Galano, Chem. Rev. 113 (2013) 1313–1350.
[6] Y. Zhang, X. Wen, Y. Shi, R. Yue, L. Bai, Q. Liu, X. Ba, Ind. Eng. Chem. Res. 58
2019) 1142–1149.
(
3
[
[
7] S. Zhou, L. Gao, F. Wei, S. Lin, H. Guo, J. Catal. 375 (2019) 410–418.
8] C. Riley, S. Zhou, D. Kunwar, A. De La Riva, E. Peterson, R. Payne, L. Gao, S. Lin, H.
Guo, A. Datye, J. Am. Chem. Soc. 140 (2018) 12964–12973.
4
. Conclusion
[
9] R. Lin, D. Albani, E. Fako, S.K. Kaiser, O.V. Safonova, N. López, J. Pérez-Ramírez,
Angew. Chem. 131 (2019) 514–519.
In summary, we have developed a novel method for the facile
fabrication of N-doped carbon supported nickel nitride (Ni N/NC)
via a direct pyrolysis of a solid mixture of nickel acetate tetrahy-
drate and Mlm. Notably, Ni N has been demonstrated as the active
species in the semi-hydrogenation of alkynes. The Ni N/NC-6/
-550 catalyst thus obtained showed high activity, excellent
selectivity for (Z)-alkenes and good functional group tolerance in
the semi-hydrogenation of both internal and terminal alkynes.
We have experimentally demonstrated and theoretically con-
firmed that the selectivity for alkene in semi-hydrogenation of
[10] Y. Liu, A.J. McCue, C. Miao, J. Feng, D. Li, J.A. Anderson, J. Catal. 364 (2018) 406–
14.
4
3
[
11] D. Albani, M. Shahrokhi, Z. Chen, S. Mitchell, R. Hauert, N. López, J. Pérez-
Ramírez, Nat. Commun. 9 (2018) 2634.
3
[12] R. Guo, Q. Chen, X. Li, Y. Liu, C. Wang, W. Bi, C. Zhao, y. Guo, M. Jin, J. Mater.
Chem. A 7 (2019) 4714–4720.
3
[
13] A.J. McCue, A. Guerrero-Ruiz, I. Rodríguez-Ramos, J.A. Anderson, J. Catal. 340
2016) 10–16.
[14] X. Zhao, L. Zhou, W. Zhang, C. Hu, L. Dai, L. Ren, B. Wu, G. Fu, N. Zheng, Chem 4
2018) 1080–1091.
5
(
(
[
15] A.J. McCue, A. Guerrero-Ruiz, C. Ramirez-Barria, I. Rodríguez-Ramos, J.A.
Anderson, J. Catal. 355 (2017) 40–52.
[16] Y. Liu, A.J. McCue, J. Feng, S. Guan, D. Li, J.A. Anderson, J. Catal. 364 (2018) 204–
15.
2
3
alkynes can be markedly enhanced using the Ni N/NC-6/5-550 cat-
alyst. Furthermore, this non-precious metal catalyst showed good
stability, making it an excellent candidate for further applications.
[
17] X. Li, Y. Pan, H. Yi, J. Hu, D. Yang, F. Lv, W. Li, J. Zhou, X. Wu, A. Lei, L. Zhang, ACS
Catal. 9 (2019) 4632–4641.
[18] F. Liguori, P. Barbaro, J. Catal. 311 (2014) 212–220.
[
[
19] Y. Kuwahara, H. Kango, H. Yamashita, ACS Catal. 9 (2019) 1993–2006.
20] D. Teschner, E. Vass, M. Hävecker, S. Zafeiratos, P. Schnörch, H. Sauer, A. Knop-
Gericke, R. Schlögl, M. Chamam, A. Wootsch, A.S. Canning, J.J. Gamman, S.D.
Jackson, J. McGregor, L.F. Gladden, J. Catal. 242 (2006) 26–37.
Declaration of Competing Interest
[
21] S. Wang, Z.-J. Zhao, X. Chang, J. Zhao, H. Tian, C. Yang, M. Li, Q. Fu, R. Mu, J.
Gong, Angew. Chem., Int. Ed. 58 (2019) 7668–7672.
The authors declare that they have no known competing finan-
cial interests or personal relationships that could have appeared
to influence the work reported in this paper.
[
22] F. Huang, Y. Deng, Y. Chen, X. Cai, M. Peng, Z. Jia, P. Ren, D. Xiao, X. Wen, N.
Wang, H. Liu, D. Ma, J. Am. Chem. Soc. 140 (2018) 13142–13146.

3
0
X. Shi et al. / Journal of Catalysis 382 (2020) 22–30
[
[
23] K. Tokmic, A.R. Fout, J. Am. Chem. Soc. 138 (2016) 13700–13705.
24] K.K. Tanabe, M.S. Ferrandon, N.A. Siladke, S.J. Kraft, G. Zhang, J. Niklas, O.G.
Poluektov, S.J. Lopykinski, E.E. Bunel, T.R. Krause, J.T. Miller, A.S. Hock, S.T.
Nguyen, Angew. Chem., Int. Ed. 53 (2014) 12055–12058.
[49] H. Zhou, X. Yang, L. Li, X. Liu, Y. Huang, X. Pan, A. Wang, J. Li, T. Zhang, ACS
Catal. 6 (2016) 1054–1061.
[50] D. Albani, K. Karajovic, B. Tata, Q. Li, S. Mitchell, N. López, J. Pérez-Ramírez,
ChemCatChem 10 (2018) 1–9.
[
[
[
[
25] P.-H. Phua, L. Lefort, J.A.F. Boogers, M. Tristany, J.G. de Vries, Chem. Commun.
[51] Y. Liu, X. Liu, Q. Feng, D. He, L. Zhang, C. Lian, R. Shen, G. Zhao, Y. Ji, D. Wang, G.
Zhou, Y. Li, Adv. Mater. 28 (2016) 4747–4754.
[52] Y. Song, S. Laursen, J. Catal. 372 (2019) 151–162.
[53] S. Carenco, A. Leyva-Pérez, P. Concepción, C. Boissière, N. Mézailles, C. Sanchez,
A. Corma, Nano Today 7 (2012) 21–28.
(
2009) 3747–3749.
26] F. Studt, F. Abild-Pedersen, T. Bligaard, R.Z. Sørensen, C.H. Christensen, J.K.
Nørskov, Science 320 (2008) 1320–1322.
27] F. Chen, C. Kreyenschulte, J. Radnik, H. Lund, A.-E. Surkus, K. Junge, M. Beller,
ACS Catal. 7 (2017) 1526–1532.
[54] C. Li, Y. Chen, S. Zhang, J. Zhou, F. Wang, S. He, M. Wei, D.G. Evans, X. Duan,
ChemCatChem 6 (2014) 824–831.
28] C. Chen, Y. Huang, Z. Zhang, X.-Q. Dong, X. Zhang, Chem. Commun. 53 (2017)
4
612–4615.
[55] K. Schutte, A. Doddi, C. Kroll, H. Meyer, C. Wiktor, C. Gemel, G. van Tendeloo, R.
A. Fischer, C. Janiak, Nanoscale 6 (2014) 5532–5544.
[56] L. Yang, S. Yu, C. Peng, X. Fang, Z. Cheng, Z. Zhou, J. Catal. 370 (2019) 310–320.
[57] Y. Liu, J. Zhao, J. Feng, Y. He, Y. Du, D. Li, J. Catal. 359 (2018) 251–260.
[58] W. Long, N.A. Brunelli, S.A. Didas, E.W. Ping, C.W. Jones, ACS Catal. 3 (2013)
1700–1708.
[
[
29] E. Richmond, J. Moran, J. Org. Chem. 80 (2015) 6922–6929.
30] E. Korytiakova, N.O. Thiel, F. Pape, J.F. Teichert, Chem. Commun. 53 (2017)
7
32–735.
[
[
[
[
31] O.G. Salnikov, H.-J. Liu, A. Fedorov, D.B. Burueva, K.V. Kovtunov, C. Coperet, I.V.
Koptyug, Chem. Sci. 8 (2017) 2426–2430.
32] A. Fedorov, H.-J. Liu, H.-K. Lo, C. Copéret, J. Am. Chem. Soc. 138 (2016) 16502–
[59] M. Shalom, D. Ressnig, X. Yang, G. Clavel, T.P. Fellinger, M. Antonietti, J. Mater.
Chem. A 3 (2015) 8171–8177.
[60] F. Gillot, J. Oro-Sole, M.R. Palacin, J. Mater. Chem. 21 (2011) 9997–10002.
[61] G. Clavel, V. Molinari, A. Kraupner, C. Giordano, Chem.-Eur. J. 20 (2014) 9018–
9023.
1
6507.
33] T.N. Gieshoff, A. Welther, M.T. Kessler, M.H.G. Prechtl, A. Jacobi von Wangelin,
Chem. Commun. 50 (2014) 2261–2264.
34] M. Tejeda-Serrano, J.R. Cabrero-Antonino, V. Mainar-Ruiz, M. López-Haro, J.C.
Hernández-Garrido, J.J. Calvino, A. Leyva-Pérez, A. Corma, ACS Catal. 7 (2017)
[62] M. Shalom, V. Molinari, D. Esposito, G. Clavel, D. Ressnig, C. Giordano, M.
Antonietti, Adv. Mater. 26 (2014) 1272–1276.
3
721–3729.
[
[
[
35] X. Qi, X. Liu, L.-B. Qu, Q. Liu, Y. Lan, J. Catal. 362 (2018) 25–34.
36] N. Kaeffer, K. Larmier, A. Fedorov, C. Copéret, J. Catal. 364 (2018) 437–445.
37] N. Kaeffer, H.-J. Liu, H.-K. Lo, A. Fedorov, C. Copéret, Chem. Sci. 9 (2018) 5366–
[63] K. Xu, P. Chen, X. Li, Y. Tong, H. Ding, X. Wu, W. Chu, Z. Peng, C. Wu, Y. Xie, J.
Am. Chem. Soc. 137 (2015) 4119–4125.
[64] G. Chen, X. Zhu, J. Cai, Y. Wan, Synth. Commun. 37 (2007) 1355–1361.
[65] G. Kresse, J. Furthmüller, Comp. Mater. Sci. 6 (1996) 15–50.
[66] G. Kresse, J. Furthmüller, Phys. Rev. B 54 (1996) 11169–11186.
[67] G. Henkelman, B.P. Uberuaga, H. Jónsson, J. Chem. Phys. 113 (2000) 9901–
9904.
[68] S. Kumar, T. Surendar, A. Baruah, V. Shanker, J. Mater. Chem. A 1 (2013) 5333–
5340.
[69] M. Chen, J. Qi, D. Guo, H. Lei, W. Zhang, R. Cao, Chem. Commun. 53 (2017)
9566–9569.
[70] F. Xie, H. Wu, J. Mou, D. Lin, C. Xu, C. Wu, X. Sun, J. Catal. 356 (2017) 165–172.
[71] Y. Wang, L. Chen, X. Yu, Y. Wang, G. Zheng, Adv. Energy Mater. 7 (2017)
1601390.
[72] D. Guo, R. Shibuya, C. Akiba, S. Saji, T. Kondo, J. Nakamura, Science 351 (2016)
361–365.
[73] X. Li, W. Zhang, Y. Liu, R. Li, ChemCatChem 8 (2016) 1111–1118.
[74] S. Fu, N.Y. Chen, X. Liu, Z. Shao, S.P. Luo, Q. Liu, J. Am. Chem. Soc. 138 (2016)
8588–8594.
[75] E. Vasilikogiannaki, I. Titilas, G. Vassilikogiannakis, M. Stratakis, Chem.
Commun. 51 (2015) 2384–2387.
5
371.
[
[
38] M.D. de los Bernardos, S. Pérez-Rodríguez, A. Gual, C. Claver, C. Godard, Chem.
Commun. 53 (2017) 7894-7897
39] Y. Chen, C. Li, J. Zhou, S. Zhang, D. Rao, S. He, M. Wei, D.G. Evans, X. Duan, ACS
Catal. 5 (2015) 5756–5765.
[
[
40] H. Konnerth, M.H.G. Prechtl, Chem. Commun. 52 (2016) 9129–9132.
41] K. Murugesan, C.B. Bheeter, P.R. Linnebank, A. Spannenberg, J.N.H. Reek, R.V.
Jagadeesh, M. Beller, ChemSusChem 12 (2019) 3363–3369.
42] K. Murugesan, A.S. Alshammari, M. Sohail, M. Beller, R.V. Jagadeesh, J. Catal.
[
[
3
70 (2019) 372–377.
43] X. Dai, Z. Chen, T. Yao, L. Zheng, Y. Lin, W. Liu, H. Ju, J. Zhu, X. Hong, S. Wei, Y.
Wu, Y. Li, Chem. Commun. 53 (2017) 11568–11571.
[
[
44] X. Wen, X. Shi, X. Qiao, Z. Wu, G. Bai, Chem. Commun. 53 (2017) 5372–5375.
45] Y. Chai, G. Wu, X. Liu, Y. Ren, W. Dai, C. Wang, Z. Xie, N. Guan, L. Li, J. Am. Chem.
Soc. 141 (2019) 9920–9927.
[
[
[
46] S.C. Smith, D.F. Rodrigues, Carbon 91 (2015) 122–143.
47] M. Zhao, Chem.-Asian J. 11 (2016) 461–464.
48] L. Shen, S. Mao, J. Li, M. Li, P. Chen, H. Li, Z. Chen, Y. Wang, J. Catal. 350 (2017)
1
3–20.