Metal and acid sites instantaneously prepared over Ni/SAPO-11 bifunctional catalyst

Article abstract of DOI:10.1016/j.jcat.2019.04.031

The Ni/SAPO-11 bifunctional catalyst for hydroisomerization of n-hexane was prepared via a novel synthesis method. It involved grinding of nickel source with amorphous precursors used for SAPO-11 followed by crystallization at 473 K for 24 h, thus avoiding the use of extra solvents in the synthesis. The highly dispersed nickel species and acid sites in the Ni/SAPO-11 bifunctional catalyst were instantaneously formed. The Ni/SAPO-11 catalyst contains framework nickel, nickel monoxide (NiO) and nickel aluminate spinel. The nickel monoxide with a size of 2–4 nm provides (de)hydrogenation function after reduction, while the framework nickel supplies more acid sites leading to an enhanced isomerization activity. The Ni/SAPO-11 catalyst shows a great synergetic effect between the metallic nickel and acid sites with a high metal-to-acid sites ratio (C_Ni/C_A) and close proximity. A single metallic nickel site is able to balance ca. 5 acid sites (C_Ni/C_A ≈ 0.19) over the Ni/SAPO-11 catalyst in n-hexane hydroisomerization. The high dispersion of nickel over the catalyst provides relatively excessive metal sites (C_Ni/C_A > 0.19), leaving the rate limiting reaction to occur on the acid sites. The Ni/SAPO-11 catalyst exhibits comparable n-hexane conversion (71.2%) and iso-hexane yield (66.7%) to the classical Pt/SAPO-11 catalyst. With enhancing acidity, the Ni/SAPO-11 catalyst exhibits one of the highest iso-hexane yields reported in the n-hexane hydroisomerization, which render the new material as a promising candidate for the hydroisomerization catalysts.

Full text of DOI:10.1016/j.jcat.2019.04.031

Journal of Catalysis 374 (2019) 208–216
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Metal and acid sites instantaneously prepared over Ni/SAPO-11
bifunctional catalyst
a,c
Yuchao Lyu a,b, Zhumo Yu a, Ye Yang a, Yuxiang Liu a, Xinxin Zhao a, Xinmei Liu a, , Svetlana Mintova
,
⇑
Zifeng Yan a, Guofeng Zhao d
a State Key Laboratory of Heavy Oil Processing, China University of Petroleum, Qingdao 266555, China
b EMS Energy Institute, The Pennsylvania State University, University Park, PA 16802, USA
c Laboratoire Catalyse & Spectrochimie, ENSICAEN, Université de Caen, 14000 Caen, France
d Shanghai Key Laboratory of Green Chemistry and Chemical Processes, School of Chemistry and Molecular Engineering, East China Normal University, Shanghai 200062, China
a r t i c l e i n f o
a b s t r a c t
Article history:
The Ni/SAPO-11 bifunctional catalyst for hydroisomerization of n-hexane was prepared via a novel syn-
thesis method. It involved grinding of nickel source with amorphous precursors used for SAPO-11 fol-
lowed by crystallization at 473 K for 24 h, thus avoiding the use of extra solvents in the synthesis. The
highly dispersed nickel species and acid sites in the Ni/SAPO-11 bifunctional catalyst were instanta-
neously formed. The Ni/SAPO-11 catalyst contains framework nickel, nickel monoxide (NiO) and nickel
aluminate spinel. The nickel monoxide with a size of 2–4 nm provides (de)hydrogenation function after
reduction, while the framework nickel supplies more acid sites leading to an enhanced isomerization
activity. The Ni/SAPO-11 catalyst shows a great synergetic effect between the metallic nickel and acid
sites with a high metal-to-acid sites ratio (CNi/CA) and close proximity. A single metallic nickel site is able
to balance ca. 5 acid sites (CNi/CA ꢀ 0.19) over the Ni/SAPO-11 catalyst in n-hexane hydroisomerization.
The high dispersion of nickel over the catalyst provides relatively excessive metal sites (CNi/CA > 0.19),
leaving the rate limiting reaction to occur on the acid sites. The Ni/SAPO-11 catalyst exhibits comparable
n-hexane conversion (71.2%) and iso-hexane yield (66.7%) to the classical Pt/SAPO-11 catalyst. With
enhancing acidity, the Ni/SAPO-11 catalyst exhibits one of the highest iso-hexane yields reported in
the n-hexane hydroisomerization, which render the new material as a promising candidate for the
hydroisomerization catalysts.
Received 7 March 2019
Revised 14 April 2019
Accepted 18 April 2019
Keywords:
Nickel
SAPO-11
Synthesis
Bifunctional catalyst
Hydroisomerization
Ó 2019 Elsevier Inc. All rights reserved.
1. Introduction
hydroisomerization catalysts is an appealing task. Several attempts
have been made to prepare non-noble metal catalysts, and it has
Hydroisomerization of light n-alkanes is an important catalytic
process to obtain clean gasoline with high octane number [1–3].
Increasing proportion of the isomerized gasoline in the gasoline
pool contributes to the products upgrading and environmental
protection [2,4,5]. The current catalysts used in the hydroisomer-
ization process are bifunctional, namely containing acid sites for
skeletal isomerization and metal sites for (de)hydrogenation
[6,7]. Noble metals (platinum, palladium) are widely used in the
bifunctional catalysts [2,8–10], but suffer from limited availability
and high costs. The widely used Pt-promoted chlorinated alumina
catalysts cause serious corrosion problems and environmental pol-
lution due to the need of continuous chlorine addition [11]. There-
fore, the development of non-noble metal (nickel) based
been shown that large metal particles are formed caused by a sev-
ere sintering which is unavoidable especially at high metal loading
[12,13]. This problem restricts the optimization of the existing
non-noble catalysts. Therefore, it is vital to look for a novel strategy
for preparation of the non-noble metals with reduced particle sizes
for hydroisomerization catalysts.
In order to reduce the particle size of metal species in catalysts,
strengthening of the interactions between the metal species and
support to inhibit the metals aggregation is of critical importance
[14,15]. Using the impregnation method with solvents with weak
polarities is an effective way to achieve this goal [16–19]. In the
previous work [13], ethanol was used as solvent to support nickel
on SAPO-11 and to produce Ni/SAPO-11 catalyst. The particle size
of nickel monoxide (20 nm) was much smaller than that over the
catalyst prepared with water as solvent (50 nm). Compared with
the conventional impregnation method, a stronger metal-support
⇑
Corresponding author.
E-mail address: lxmei@upc.edu.cn (X. Liu).
https://doi.org/10.1016/j.jcat.2019.04.031
0021-9517/Ó 2019 Elsevier Inc. All rights reserved.

Y. Lyu et al. / Journal of Catalysis 374 (2019) 208–216
209
interaction can be obtained through equilibrium adsorption
method. The metal species were deposited through adsorption
and reaction with receptor sites developed on the support [20].
Goula et al. [21] prepared Ni/Al2O3 by the equilibrium adsorption
method for the biogas dry reforming reaction. They found that a
smaller metal particle size of 8 nm was obtained through this
method compared with metal particles of 13 nm of the catalysts
prepared by the impregnation method; the former catalyst showed
enhanced catalytic performance. Controllable absorption method
could also be used to strengthen the metal-support interaction in
Ni/Al2O3 catalysts. Optimal absorption conditions were identified
using point of zero charge, pH precipitation and adsorption exper-
iments [22]. The Ni/Al2O3 catalyst prepared under such conditions
presented a stronger metal-support interaction and higher metal
dispersion than the catalyst prepared by impregnation. Many other
trials including introduction of additives [23–25] or changing
metal precursors [10,26,27] to inhibit the aggregation of metal spe-
cies have been performed.
a muffle oven for 6 h. Varying the amount of nickel addition, cata-
lysts with different nickel loading were prepared. When H-Beta
zeolite was added into the initial synthesis gel of the Ni/SAPO-11
(s), the final product was abbreviated as Ni/SAPO-11-b(s). The mass
ratio of H-Beta relative to the alumina source used for the SAPO-11
(pseudo-boehmite) was 0.1. As a reference, SAPO-11(s) without
nickel was also prepared by the same procedure.
For comparison, two samples via incipient impregnation
method abbreviated as Ni/SAPO-11(im) and Pt/SAPO-11(im) were
prepared as follow: SAPO-11 was immersed in aqueous solutions
of Ni(NO3)2ꢂ6 H2O or H2PtCl6. After the impregnation, the material
was dried at 383 K for 2 h and then calcined in air at 673 K for 4 h.
The loading of nickel in the Ni/SAPO-11(im) sample was 4.0 wt%
that was the same as for the Ni/SAPO-11(s) sample; the loading
of Pt was 0.3 wt%. The SAPO-11 used for the impregnation was syn-
thesized according to the procedure reported before [13].
2.3. Characterization
However, most of these methods lead to more complex synthe-
sis procedures, violating the green synthesis of the catalyst.
Besides, the main focus is on metal dispersion without paying
due attention to acidity. Even more the acid sites on supports dur-
ing the impregnation or adsorption of metal precursor are modi-
fied. This is not a favorable situation for bifunctional catalysts
whose performance depends not only on the metal sites, but also
on the acidity and metal-acid balance.
In the present work, direct synthesis method was used for
instantaneous synthesis of highly-dispersed nickel species and acid
sites over the Ni/SAPO-11 hydroisomerization catalyst. The synthe-
sis method involved initial grinding of SAPO-11 precursors with
nickel source without extra solvents followed by crystallization.
This method leads to significant reduction of waste and solvent
use as well as the synthesis procedure is shortened significantly.
It differs from the solvent-free method that requires pre-reaction
of the raw materials and washing of intermediates with ethanol
[28,29]. The features of metal and acid sites over the Ni/SAPO-11
catalyst were investigated systematically. The catalytic perfor-
mance was assessed by hydroisomerization of n-hexane, and
active site for (de)hydrogenation reaction was disclosed. The
metal-acid balance and reaction pathway were also discussed.
X-ray diffraction (XRD) patterns of samples were collected
with X’ Pert PRO MPD diffractometer (PANalytical B.V. Nether-
lands) with Cu Ka radiation (k = 0.15418 nm) operated at 40 kV
and 40 mA. Transmission electron microscopy (TEM) and elemen-
tal mapping of nickel were performed using an FEI Titan G2 80-
200 TEM/STEM. The elemental mapping was obtained by energy
dispersive X-ray spectroscopy using the super-X detector. X-ray
photoelectron spectroscopy (XPS) was carried out using a PHI
500 spectrometer with Al K
a radiation. The binding energy was
calibrated by measuring the C 1s peak at 284.8 eV. Practical metal
loading and chemical compositions of the samples were mea-
sured by X-ray fluorescence (XRF, ZSX-100e using Rh and Au exci-
tation tubes). X-ray absorption fine structure (XAFS) spectra were
obtained at the 1W1B station in the BSRF (Beijing Synchrotron
Radiation Facility, People’s Republic of China) operated at
2.5 GeV with a maximum current of 250 mA. X-ray absorption
spectroscopy (XAS) measurements at the Ni K-edge were per-
formed in fluorescence mode using a Lytle detector. All samples
were pelletized as disks of 13 mm diameter with 1 mm thickness
using graphite powder as a binder. Temperature programmed
reduction of hydrogen (H2-TPR) and temperature programmed
desorption of ammonia (NH3-TPD) were carried out on a dynamic
chemisorption analyzer (Micromeritics AutoChem 2920). The IR
spectra of catalysts were recorded by a Nicolet-6700 FTIR spec-
trometer. The acidity of the catalysts was evaluated using pyri-
dine as a probe molecule followed by FTIR. All samples were
evacuated at 573 K for 3 h, followed by adsorption of pyridine
at room temperature and desorption of pyridine in a vacuum at
423 K. Finally the spectra were recorded at room temperature.
2. Experimental
2.1. Materials
Pseudo-boehmite (70.0 wt% Al2O3) was purchased from Yantai
Henghui Petrochemical Co., Ltd. Phosphoric acid (85.0 wt%
H3PO4) was purchased from XiRong Petrochemical Co., Ltd. Di-
propylamine (DPA, ꢁ99.0 wt% C6H15N) and Nickel (II) nitrate hex-
ahydrate (Ni(NO3)2ꢂ6H2O) were purchased from Sinopharm Chem-
ical Reagent Co., Ltd. Silica sol (30.0 wt% SiO2) was purchased from
Qingdao Haiyang Petrochemical Co., Ltd. All chemicals were used
without purification.
H2 chemisorption measurements were carried out on
a
Micromeritics AutoChem 2920; the accessible metal sites (CNi
)
of the catalysts after reduction were studied. Prior to the
chemisorption at ambient temperature, the samples were pre-
treated in hydrogen atmosphere at 823 K for 2 h and purged at
this temperature in an argon flow for 2 h to avoid the presence
of residual absorbed hydrogen. The ratio (CNi/CA) of metal sites
(CNi) relative to the acid sites (CA) was calculated according to
the method disclosed before [30–32]. Ultraviolet–visible (UV–
vis) diffuse reflectance spectra were recorded on a UV-2600
UV–vis spectrophotometer using BaSO4 as a reference. The turn-
over frequency (TOF) per acid site, labeled as TOFn-C6, was calcu-
lated using the following equation:
2.2. Sample synthesis
Direct synthesis of Ni/SAPO-11(s) was carried out using a mix-
ture with the following molar ratio: 1.0 P2O5: 1.9 DPA: 0.4 SiO2:
0.19 Ni (NO3)2ꢂ6 H2O: 1.0 Al2O3. In a typical run, 4.38 g of
pseudo-boehmite, 6.92 g of H3PO4, 5.76 g DPA, 2.4 g of silica sol
and 1.65 g of Ni(NO3)2ꢂ6H2O were mixed in a mortar. After grind-
ing for 10 min, the mixture was transformed into an autoclave
and heated at 473 K for 24 h. The solid product was washed with
deionized water, dried at 373 K for 12 h and calcined at 773 K in
ꢃ1
ꢃ1
Rate of converted n - hexane ðmol gcat
h
Þ
TOFnꢃC6
¼
ꢃ1
Number of acid sites ðmol gcat
Þ

210
Y. Lyu et al. / Journal of Catalysis 374 (2019) 208–216
2.4. Catalytic tests
3.2. State of nickel species
The hydroisomerization of n-hexane (n-C6) was selected as a
model reaction to assess the catalytic performance of the catalysts.
The catalytic tests were carried out at a pressure of 2.0 MPa, a
weight hourly space velocity (WHSV) of 1.0 hꢃ1, and a H2/n-C6
molar ratio of 4.0. Prior to the reaction, the catalyst was reduced
in situ under a hydrogen flow at 823 K for 2 h and then cooled to
reaction temperature. The reaction products were analyzed on-
line by a gas chromatograph (Agilent 7820) equipped with a flame
ionization detector. The n-hexane conversion might be increased
due to the thermal behavior of the reaction at high temperature.
The thermal reaction of the current feed was carried out at the
highest temperature (653 K) in the present work. The n-hexane
conversion was less than 1.0%, indicating that in all cases investi-
gated, the thermal reaction did not bias the hydroisomerization
results.
UV–vis diffuse reflectance spectra of the calcined catalysts are
shown in Fig. 2; the spectrum of pure NiO is added as a reference.
The UV–vis spectra of the Ni/SAPO-11(im) catalyst and the bulk
NiO are very similar in bands positions, consisting of absorption
bands at 280, 380, 420, 640 and 720 nm. This implies that the
nickel in the Ni/SAPO-11(im) exists as NiO [38]. For the Ni/SAPO-
11(s), apart from the absorption bands assigned to the NiO, the
new one at 257 nm is attributed to the electron transfer from
O2ꢃ (2p) to tetrahedral Ni2+ (3d) in framework of the SAPO-11
[39–41]; the peak at 400 nm (Fig. 2(b)) is assigned to the octahe-
drally coordinated nickel species in nickel aluminate spinel [42,43].
XPS was also carried out to further elucidate the chemical envi-
ronment and nature of nickel species in the Ni/SAPO-11(s) and Ni/
SAPO-11(im) catalysts (Fig. 3); the spectrum of bulk NiO is added
as a reference. The Ni 2p region of the bulk NiO comprises four
easily discernible features: the Ni 2p3/2 main peak and its satellite
at ꢄ854.6 and ꢄ861.2 eV, and the Ni 2p1/2 main peak and its satel-
lite at ꢄ871.8 and ꢄ879.1 eV. This result is consistent with the XPS
spectrum of bulk NiO reported in literature [44]. However, the
electronic structure is more complex. The two peaks at 853.8 and
855.4 eV in the Ni 2p3/2 region after deconvolution can be
explained in terms of a simplified model of final state effects in
which photoemission is potentially accompanied by rearrange-
ment of electrons in the valence band [45]. They shift to higher
binding energy (854.3 and 856.1 eV) for the Ni/SAPO-11(im) due
to the interaction between NiO and the support. In contrast, for
the Ni/SAPO-11(s), the peak at 856.3 eV of Ni 2p3/2, together with
the one at 873.6 eV of Ni 2p1/2 indicate the existence of octahedral
Ni2+ in the NiO that strongly interacts with the SAPO-11 [46,47].
The peaks at 857.6 and 875.3 eV, which are in the range of Ni
2p3/2 and Ni 2p1/2, respectively, are assigned to nickel aluminate
spinel [48]. The small peaks at 859.0 (Ni 2p3/2) and 876.9 eV (Ni
2p1/2) can be assigned to tetrahedral Ni2+ in the framework of
SAPO-11 [46], which is also supported by the UV–vis results.
X-ray absorption near-edge structure (XANES) was carried out
to provide further insights into the electronic structure of nickel
in the Ni/SAPO-11(s) (Fig. 4). Compared with the NiO reference,
white line of the Ni/SAPO-11(s) shifts towards lower energy, prob-
ably due to the formation of tetrahedral Ni2+ [49] in the SAPO-11
framework, consistent with the XRD, XPS and UV–vis results. The
experimental and fitted EXAFS curves for the Ni/SAPO-11(s) cata-
lyst is shown in Fig. S1, and Table S1 shows the average shell dis-
tances around the nickel atoms. The Ni-O distance of 2.09 Å and Ni-
3. Results and discussion
3.1. Crystalline phases
Fig. 1 shows XRD patterns of the calcined Ni/SAPO-11(s), Ni/
SAPO-11(im) and SAPO-11(s) samples. All the XRD patterns
(Fig. 1(a)) correspond to well-crystalline samples with the AEL type
framework structure [33,34]. The relative crystallinity of Ni/SAPO-
11(s) increases compared with SAPO-11(s) and the growth of
(0 0 2) crystal plane is enhanced mostly. Only a slight decrease in
the peak width is observed for the Ni/SAPO-11(s) compared with
the SAPO-11(s), which implies a small difference in grain size
between the two samples. Thus, the increased crystallinity of the
Ni/SAPO-11(s) might be due to the effective structure directing
effect of amine complex of nickel formed by nickel and di-
propylamine in the synthesis gel of Ni/SAPO-11(s). Similar phe-
nomenon was observed in the synthesis of aluminophosphates
molecular sieves with nickel or cobalt complex of diethylenetri-
amine as a structure directing agent [35,36]. In this case, part of
nickel ultimately resided in tetrahedral sites in the framework of
SAPO-11 [36], which is confirmed by the slight shifting of the
diffraction peak to a lower angle for the Ni/SAPO-11(s) (0.03°)
compared with the SAPO-11(s) (Fig. 1(b)). The substitution of
smaller tetrahedrally coordinated skeletal atoms (Al 0.51 Å or P
0.38 Å in crystal form) with heteroatoms of nickel that has a larger
size (0.69 Å) as ions in crystalline environment, yields an increased
lattice parameter value [37].
Fig. 1. XRD patterns of (a) as synthesized SAPO-11(s), calcined Ni/SAPO-11(s) and Ni/SAPO-11(im) in the range of 5–80° 2 Theta, and (b) SAPO-11(s) and calcined Ni/SAPO-11
(s) in the range of 20.9–23.9° 2 Theta.

Y. Lyu et al. / Journal of Catalysis 374 (2019) 208–216
211
Fig. 2. UV–vis diffuse reflectance spectra of the calcined Ni/SAPO-11(s), Ni/SAPO-11(im) and bulk NiO samples (a) and peak simulation for the spectrum of sample Ni/SAPO-
11(s) in the range of 350–520 nm (b).
Fig. 4. XANES spectra at the Ni K-edge of (a) NiO reference, (b) Ni/SAPO-11(s) and
(c) Ni foil reference.
a uniform distribution of nickel (Fig. 6(c)). The higher dispersion of
nickel species could be ascribed to the spatial effect of the raw
material used for the preparation of the SAPO-11 support. All the
raw materials including the nickel source (nickel nitrate) were
Fig. 3. XPS spectra of the Ni 2p core level for calcined Ni/SAPO-11(s), Ni/SAPO-11
mixed with the precursor of SAPO-11 before the crystallization
was performed in order to inhibit the contact between the nickel
species.
(im) and bulk NiO samples recorded at room temperature.
Ni distance of 2.96 Å are assigned to the nickel in octahedral sites
of NiO [50]. The average Ni-O distance of 1.97 Å is mainly assigned
to the nickel aluminate spinel [51] that has predominant content in
the nickel species as indicated by the strongest peak at 857.6 eV in
XPS spectrum (Fig. 3) for the Ni/SAPO-11(s) catalyst. Meanwhile,
Ni-P shell is also found by fitting confirming the incorporation of
nickel into the framework of SAPO-11. In brief, three types of nickel
species are identified in the calcined Ni/SAPO-11(s) catalyst: NiO,
nickel aluminate spinel and framework nickel.
Diffraction peaks of NiO are not observable for the Ni/SAPO-11
(s) sample, but clearly, the NiO peaks are seen in the Ni/SAPO-11
(im) sample (see partly enlarged XRD patterns in Fig. S2). This indi-
cates a higher dispersion of NiO in the Ni/SAPO-11(s) sample. The
NiO spherical particles with a homogenous size (2–4 nm in diam-
eter) and high dispersion in the Ni/SAPO-11(s) catalyst are
observed (Fig. 5). Heavily agglomerated NiO particles with a size
of 30–100 nm and sharp edge on the surface of the Ni/SAPO-11
(im) catalyst are formed. Additionally, the elemental maps of
nickel (Fig. 6(f)) also show that nickel species aggregate in some
regions, showing an uneven distribution on the surface of the Ni/
SAPO-11(im). The catalyst prepared by the direct method presents
3.3. Reducibility
The H2-TPR profiles of samples Ni/SAPO-11(s), Ni/SAPO-11(im)
and SAPO-11(s) are shown in Fig. 7. The Ni/SAPO-11(im) contains
two peaks centering at ca. 643 and 715 K, which are assigned to
the reduction of NiO with weak and strong interactions with
SAPO-11, respectively [52]. Compared with Ni/SAPO-11(im), the
Ni/SAPO-11(s) catalyst exhibits higher reduction temperature sug-
gesting the stronger interaction between the nickel species and the
SAPO-11 support [53]. The maximum peak at 990 K is assigned to
the reduction of nickel aluminate spinel [54]. The small shoulder at
1050 K is ascribed to the reduction of framework nickel species,
supporting the incorporation of nickel into the framework of
SAPO-11 as previously shown by UV–vis and XPS. Another distinct
peak at ca. 823 K corresponds to the reduction of NiO strongly
interacting with SAPO-11. The hydrogen consumption at ca.
1250 K is attributed to the reduction of PO34- species in SAPO-11.
Such a high temperature is due to the high stability of P-O bond
[55]. Notably, only NiO can be reduced during the reduction step
according to the H2-TPR profile of the Ni/SAPO-11(s) catalyst.

212
Y. Lyu et al. / Journal of Catalysis 374 (2019) 208–216
Fig. 5. TEM images of the calcined Ni/SAPO-11(s) (left) and Ni/SAPO-11(im) (right) catalysts.
Fig. 6. STEM images and the corresponding EDS elemental maps of nickel of calcined Ni/SAPO-11(s) (a, b and c) and Ni/SAPO-11(im) (d, e and f) catalysts.
3.4. Acidity
The FT-IR spectra of pyridine (Py-IR) absorbed on reduced Ni/
SAPO-11(s), Ni/SAPO-11(im) and SAPO-11(s) samples are depicted
in Fig. 8. The IR bands at 1540 and 1450 cmꢃ1 are assigned to pyr-
idine adsorbed on Brønsted and Lewis acid sites, respectively
[56,57]. Obviously, the impregnated catalyst has less acid sites
than the Ni/SAPO-11(s) catalyst. It should be noticed that the Ni/
SAPO-11(s) has more Brønsted acid sites than the SAPO-11(s).
The Ni/SAPO-11(s) and SAPO-11(s) exhibit similar compositions
in terms of Si, P and Al (Table S2). The increased Brønsted acid sites
over the Ni/SAPO-11(s) catalyst are mainly contributed by the
incorporation of divalent nickel ions into the SAPO-11 framework
that introduces more negative charges, thus generating more
Brønsted acid sites. The Ni2+ was most likely to substitute for
Al3+ to form bridged NiP-OH structures in the Ni/SAPO-11(s) sam-
ple [58]. Similar phenomenon was observed in the metal-
Fig. 7. H2-TPR profiles of samples Ni/SAPO-11(s), Ni/SAPO-11(im) and SAPO-11(s).
substituted AlPO4 molecular sieves [59]. The Brønsted acid sites

Y. Lyu et al. / Journal of Catalysis 374 (2019) 208–216
213
The nickel content and concentration of metallic nickel sites
(CNi) in the reduced catalysts are shown in Table 1. Obviously,
the reduced Ni/SAPO-11(s) contains more metal sites on the sur-
face in comparison to the reduced Ni/SAPO-11(im). This is attribu-
ted to the smaller Ni particles formed over the reduced Ni/SAPO-11
(s) catalyst, which can be confirmed by the TEM images and Ni par-
ticle size distributions of the reduced Ni/SAPO-11(s) and Ni/SAPO-
11(im) catalysts (Fig. S5). The Ni particles with a relatively
homogenous size (8–12 nm) are observed over the reduced Ni/
SAPO-11(s) catalyst. In contrast, much bigger Ni particles (30–
90 nm) are formed over the reduced Ni/SAPO-11(im) catalyst.
3.6. Catalytic test
The results from the catalytic tests of samples Ni/SAPO-11(s)
and Ni/SAPO-11(im) are shown in Fig. 10; the Pt/SAPO-11(im)
sample is used only for comparison (reference sample). The Ni/
SAPO-11(s) shows much higher n-hexane conversion than the Ni/
SAPO-11(im) in the whole reaction temperature range (Fig. 10
(a)). Compared with the reference Pt/SAPO-11(im), the Ni/SAPO-
11(s) shows a slightly lower n-hexane conversion at temperature
below 613 K. As the temperature increased, they exhibit almost
the same n-hexane conversion, suggesting the high activity of
the Ni/SAPO-11(s) catalyst.
The iso-hexane selectivity decreases with n-hexane conversion
over all the catalysts (Fig. 10(c)), indicating the increased rate of
side reaction at higher n-hexane conversion. The Ni/SAPO-11(s)
and Pt/SAPO-11(im) catalysts show similar iso-hexane selectivity.
Both samples have much higher iso-hexane selectivity than the
Ni/SAPO-11(im) which show a significant increase in cracking with
n-hexane conversion (Fig. 10(d)). In all cases, the iso-hexane selec-
tivity follows the order: Ni/SAPO-11(s) ꢀ Pt/SAPO-11(im) > Ni/
SAPO-11(im). The yield of iso-hexane over the Ni/SAPO-11(s) is
slightly lower than that over the Pt/SAPO-11(im) at temperature
below 613 K, while it exceeds the latter as the temperature
increases continuously (Fig. 10(b)). Most interestingly, the Ni/
SAPO-11(s) and Pt/SAPO-11(im) catalysts have nearly the same
maximum iso-hexane yield at ca. 66.7% at 613 K. In contrast, the
Ni/SAPO-11(im) has the lowest iso-hexane yield at all the temper-
atures tested. Therefore, the Ni/SAPO-11(s) catalyst prepared by
the novel direct synthesis method has comparable performance
to the noble metal catalyst Pt/SAPO-11(im) in terms of n-hexane
conversion and iso-hexane selectivity. This result suggests that
the Ni/SAPO-11(s) is a promising candidate for hydroisomerization
catalysts.
A detailed products distribution for the Ni/SAPO-11(s), Ni/
SAPO-11(im) and Pt/SAPO-11(im) at similar n-hexane conversion
is shown in Table 2. The selectivity to iso-hexane over the Ni/
SAPO-11(s), Ni/SAPO-11(im) and Pt/SAPO-11(im) is 95.1, 84.4
and 95.7%, respectively. The Ni/SAPO-11(s) catalyst and the Pt/
SAPO-11(im) reference catalyst exhibit similar selectivity to
methyl pentanes (MP), 92.8% and 94.6%, respectively. While the
selectivity to methyl pentanes over Ni/SAPO-11(im) is only
81.3%. Ni/SAPO-11(s) and Ni/SAPO-11(im) show slightly higher
selectivity to dimethyl butanes (DMB), 2.3 and 3.1%, respectively
than that for the Pt/SAPO-11(im) (1.1%). This result indicates that
nickel based hydroisomerization catalysts (Ni/SAPO-11(s) and Ni/
SAPO-11(im)) favor the formation of di-branched isomers that
have higher octane number than the mono-branched ones. It is
consistent with the previous report by Eswaramoorthi and Lingap-
pan, who found that the addition of nickel to Pt/H-b or Pt/H-
mordenite catalysts increased the n-hexane conversion and DMB
selectivity, and they attributed this behavior to the metal-acid bal-
ance [60]. The Ni/SAPO-11(im) shows the highest selectivity to
cracking products (15.6%) vs the other catalysts Ni/SAPO-11(s)
(4.9%) and Pt/SAPO-11(im) (4.3%). It might be attributed to the
Fig. 8. FT-IR spectra recorded under pyridine adsorption on SAPO-11(s), and
reduced Ni/SAPO-11(s) and Ni/SAPO-11(im) catalysts.
from bridged MeP-OH (Me2+) in the MeAPO-11 sample are pro-
posed to be strong [59] and this will be confirmed by the NH3-
TPD (see Fig. S3). The acid distribution of all catalysts is summa-
rized in Table S3. The Ni/SAPO-11(s) catalyst shows more acid sites
than the Ni/SAPO-11(im). The amount of weak acid sites for Ni/
SAPO-11(s) and SAPO-11(s) is similar, while the Ni/SAPO-11(s)
contains more medium and strong acid sites. The Py-IR spectra col-
lected after desorption at different temperatures (Fig. S4) imply
that the increased acid sites in the Ni/SAPO-11(s) are mainly com-
posed of strong Brønsted acid sites.
3.5. Metal sites for (de)hydrogenation reaction
Three types of nickel species in the Ni/SAPO-11(s) catalyst are
identified as described above. Given that only NiO could be
reduced into metallic nickel during the reduction step and nickel
aluminate spinel is inactive in n-hexane hydroisomerization, the
reactions were carried out over SAPO-11(s), unreduced and
reduced Ni/SAPO-11(s) samples at 593 K (Fig. 9). Compared with
the SAPO-11(s), there is only a slight increase in the n-hexane con-
version and iso-hexane (i-C6) yield over the unreduced catalyst.
This increase is attributed to the contribution of weak (de)hydro-
genation function of NiO and enhanced isomerization function by
framework nickel species. In contrast, the n-C6 conversion and i-
C6 yield over reduced Ni/SAPO-11(s) catalyst are much higher than
that over the unreduced Ni/SAPO-11(s). This indicates that the
active sites for (de)hydrogenation are provided predominantly by
metallic nickel rather than by the framework nickel species.
Fig. 9. Catalytic performance of SAPO-11(s), unreduced and reduced Ni/SAPO-11(s)
catalysts.

214
Y. Lyu et al. / Journal of Catalysis 374 (2019) 208–216
Table 1
Nickel content, nickel sites (CNi), acid sites (CA) and the CNi/CA ratio over the Ni/SAPO-11(s) and Ni/SAPO-11(im) catalysts.
Sample
Ni content (wt %)
CNi
(l
mol/g)
CA
(lmol/g)
CNi/CA
Ni/SAPO-11(s)
Ni/SAPO-11(im)
4.0
4.0
55.4
28.0
247.6
176.9
0.224
0.158
Fig. 10. Catalytic performance of samples Ni/SAPO-11(s), Ni/SAPO-11(im) and Pt/SAPO-11(im) (reference): n-C6 conversion (a) and i-C6 yield (b) versus temperature,
selectivity to i-C6 (c) and cracking products (<C6) (d) versus n-C6 conversion.
Table 2
3.7. Synergetic effect of metal and acid sites
Products distribution of n-hexane hydroisomerization over Ni/SAPO-11(s), Ni/SAPO-
11(im) and Pt/SAPO-11(im) catalysts at the n-hexane conversion of 51.5 ( 2.0)%.
The classical mechanism describing the alkanes hydroisomer-
Sample
Ni/SAPO-11(s)
Ni/SAPO-11(im)
Pt/SAPO-11(im)
ization involves the following successive chemical steps: (i) dehy-
drogenation of alkanes into alkenes on metal sites, (ii) skeletal
rearrangement of the alkenes on acid sites and (iii) hydrogenation
of the branched alkenes to corresponding branched alkanes [32].
Thus, the catalytic behavior of the hydroisomerization catalyst is
closely related to features of (de)hydrogenation sites and acid sites
as well as the balance between them. The Ni/SAPO-11(s) in com-
parison to the Ni/SAPO-11(im) exhibits enhanced activity, which
is attributed to more accessible metal and acid sites. The Ni/
SAPO-11(s) catalyst also demonstrates higher iso-hexane selectiv-
ity than the Ni/SAPO-11(im). The better metal-acid balance for the
Ni/SAPO-11(s), which is confirmed by its higher CNi/CA ratio (0.224)
than that over the Ni/SAPO-11(im) (0.158, Table 1), is the first
explanation for this phenomenon. Under such circumstance, each
alkene intermediate comes into fewer acid sites during its diffusion
from one Ni site to another [61]. Thus, the subsequent cracking
reaction on acid sites is suppressed, resulting in high iso-hexane
selectivity. In contrast, owing to the small amount of metal sites
in the Ni/SAPO-11(im) catalyst, the alkene intermediates will con-
front acid sites that are relatively excessive and then crack into
smaller molecules, thus reducing the iso-hexane selectivity.
Another explanation could be the proximity between metal and
acid sites. For an ideal isomerization catalyst, metal and acid sites
should be close enough so that the number of acid sites between
two metal sites is low enough to catalyze only one step of skeletal
Conversion (%)
Sta (%)
49.5
95.1
92.8
57.5
35.3
2.3
0.1
2.2
4.9
53.5
84.4
81.3
49.2
32.1
3.1
0.5
2.6
15.6
53.4
95.7
94.6
58.3
36.3
1.1
–
1.1
4.3
a
SMP (%)
2-MP
3-MP
a
SDMB (%)
2,2-DMB
2,3-DMB
a
SC (%)
a
SMP, SDMB, St, and SC correspond to the selectivity to methyl pentanes, dimethyl
butanes, iso-hexanes and cracking products, respectively.
poor balance between (de)hydrogenation and acid functions which
will be discussed later.
High catalytic stability is also indispensable for practical appli-
cation of the hydroisomerization catalyst. The stability of the Ni/
SAPO-11(s) was studied preliminarily at 613 K and the result is
shown in Fig. S6. The n-hexane conversion and iso-hexane yield
exhibit a slight decrease during the initial 2 h and then remain
almost constant. The iso-hexane selectivity experiences no signifi-
cant change in spite of a slight increase in the initial 2 h. The results
indicate that the Ni/SAPO-11(s) catalyst prepared via the novel
method has a promising stability in the hydroisomerization of n-
hexane.

Y. Lyu et al. / Journal of Catalysis 374 (2019) 208–216
215
rearrangement [62]. It helps to reduce the probability of cracking
the alkene intermediates during the diffusion from acid sites to
the nearest metal sites. The Ni/SAPO-11(s) catalyst has better
metal-acid balance and closer metal-acid proximity which is con-
tributed by the small metal particles. It seems to have higher level
of ideality in terms of synergetic effect between metal and acid
sites, which can be evidenced by the reaction pathway and prod-
ucts distribution as discussed later.
The cracking/isomer (C/I) and di-branched/mono-branched
(DMB/MP) ratios vs. n-hexane conversions over the Ni/SAPO-11
(s), Ni/SAPO-11(im) and Pt/SAPO-11(im) catalysts are shown in
Fig. 11. When the n-hexane conversion is extrapolated to zero,
the high DMB/MP value over the Ni/SAPO-11(im) indicates that
both the di-branched and mono-branched isomers are primary
products. The DMB/MP and C/I ratios increase significantly with
n-hexane conversion, indicating the large number of transforma-
tion of mono-branched isomers to di-branched or even cracking
products. It seems that the n-hexane hydroisomerization over
the Ni/SAPO-11(im) follows the route shown in Fig. 12(a), which
is attributed to the poor metal-acid balance. However, for Ni/
SAPO-11(s) or Pt/SAPO-11(im), the DMB/MP and C/I ratios are
approximately zero when the n-hexane conversion is extrapolated
to zero, and they increase slightly until the n-hexane conversion
reaches 70.0%. This implies that n-hexane hydroisomerization over
the Ni/SAPO-11(s) follows a successive route (Fig. 12(b)). Through
this route, the alkene intermediates prefer to be hydrogenated on
the metal sites, instead of further transformed on acid sites, which
is evidenced by the predominant content of mono-branched iso-
mers in the isomer products (Fig. S7). Therefore, the catalytic
behavior of the Ni/SAPO-11(s) is superior to the Ni/SAPO-11(im)
prepared by impregnation.
The skeletal rearrangements of alkene intermediates on the acid
sites are suggested to be the rate limiting step over the Ni/SAPO-11
(s), according to the reaction pathway and products distribution. In
order to verify this point, different ratios of metal sites to acid sites
(CNi/CA) were attained in the Ni/SAPO-11(s) catalysts by varying
nickel loading. The n-hexane conversion and turn over frequency
(TOFn-C6) per acid site as a function of CNi/CA are shown in
Fig. S8. They increase at CNi/CA < 0.193, and then remain constant
at higher CNi/CA values. Similar phenomenon has been observed
for noble metal based bifunctional catalyst [62]. This result is in
agreement with the change in rate limiting reaction: from (de)hy-
drogenation on the nickel sites at low CNi/CA values, to skeletal
arrangements on the acid sites at high values. Therefore, the
hypothesis proposed earlier is confirmed by the results presented
for the Ni/SAPO-11(s) catalyst with a CNi/CA of 0.224.
Fig. 12. Reaction pathways of n-hexane hydroisomerization over catalysts with (a)
poor and (b) ideal metal-acid balance.
The insufficient acid sites limit the hydroisomerization process
over Ni/SAPO-11(s), and therefore H-Beta zeolite was added in the
synthesis system of Ni/SAPO-11(s) to increase the acidity and
improve the catalytic performance. The final crystalline product
was named as Ni/SAPO-11-b(s). The total acid sites increase to
374.1
lmol/g for sample Ni/SAPO-11-b(s). The number of accessi-
ble metal sites over the Ni/SAPO-11-b(s) (54.6
lmol/g) is similar
to that over the Ni/SAPO-11(s). The catalytic performance of the
Ni/SAPO-11-b(s) catalyst at 613 K is shown in Fig. 13. The n-
hexane conversion and iso-hexane yield over the Ni/SAPO-11(s)
are 71.2% and 66.7%, respectively while they increase to 80.9%
and 73.3% over the Ni/SAPO-11-b(s), respectively. The n-hexane
conversion and iso-hexane yield are also higher than the corre-
sponding values over the classical catalyst Pt/SAPO-11(im) (72.8%
and 67.1%). Besides, the Ni/SAPO-11-b(s) has approximately 9
and 10 times higher selectivity to di-branched isomers compared
to catalysts Ni/SAPO-11(s) and Pt/SAPO-11(im), respectively. Most
interestingly, with reducing the reaction temperature to 593 K, no
significant change in the n-hexane conversion (79.3%) is observed.
While the yield of iso-hexane increases to 76.0%, which is among
the highest values reported for n-hexane hydroisomerization to
our knowledge up to now. Therefore, the nickel based catalyst pre-
pared via the direct synthesis method shows promising catalytic
properties.
Fig. 11. The product distribution of n-hexane hydroisomerization over Ni/SAPO-11(s), Ni/SAPO-11(im) and Pt/SAPO-11(im) catalysts: (a) di-branched/mono-branched and
(b) cracking/isomer ratios vs. n-C6 conversions.

216
Y. Lyu et al. / Journal of Catalysis 374 (2019) 208–216
[11] P. Wang, J. Zhang, G. Wang, C. Li, C. Yang, J. Catal. 338 (2016) 124–134.
[12] C. Erkey, J. Supercrit. Fluids 47 (2009) 517–522.
[13] Y. Lyu, Y. Liu, L. Xu, X. Zhao, Z. Liu, X. Liu, Z. Yan, Appl. Surf. Sci. 401 (2017) 57–
64.
[14] S.-S. Liu, Y.-Y. Jin, Y. Han, J. Zhao, J. Ren, Fuel Process. Technol. 177 (2018) 266–
274.
[15] S. Wang, Y. Wang, C. Hu, Int. J. Hydrogen Energy 43 (2018) 13921–13930.
[16] S.-W. Ho, Y.-S. Su, J. Catal. 168 (1997) 51–59.
[17] M.F. Variava, T.L. Church, N. Noorbehesht, A.T. Harris, A.I. Minett, Catal Sci.
Technol. 5 (2015) 515–524.
[18] M. Tao, X. Meng, Y. Lv, Z. Bian, Z. Xin, Fuel 165 (2016) 289–297.
[19] F.T. Zangeneh, S. Mehrazma, S. Sahebdelfar, Fuel Process. Technol. 109 (2013)
118–123.
[20] C. Papadopoulou, J. Vakros, H.K. Matralis, C. Kordulis, A. Lycourghiotis, J.
Colloid Interf. Sci. 261 (2003) 146–153.
[21] M.A. Goula, N.D. Charisiou, K.N. Papageridis, A. Delimitis, E. Pachatouridou, E.F.
Iliopoulou, Int. J. Hydrogen Energy 40 (2015) 9183–9200.
[22] J.L. Ewbank, L. Kovarik, F.Z. Diallo, C. Sievers, Appl. Catal. A: Gen. 494 (2015)
57–67.
[23] G. Wang, H. Wang, H. Zhang, Q. Zhu, C. Li, H. Shan, ChemCatChem 8 (2016)
3137–3145.
Fig. 13. Catalytic performance of the Ni/SAPO-11(s), Ni/SAPO-11-b(s) and Pt/SAPO-
11(im) samples at 613 K, 2.0 MPa, H2/n-C6 = 4.0 and WHSV = 1.0 hꢃ1.
[24] J. Wu, Z. Peng, A.T. Bell, J. Catal. 311 (2014) 161–168.
[25] S. Wang, G.Q. Lu, Appl. Catal. B: Environ. 19 (1998) 267–277.
[26] K. Fang, J. Ren, Y. Sun, J. Mol. Catal. A: Chem. 229 (2005) 51–58.
[27] R.B. Duarte, O.V. Safonova, F. Krumeich, J.A. van Bokhoven, Phys. Chem. Chem.
Phys. 16 (2014) 26553–26560.
4. Conclusions
[28] Y. Jin, Q. Sun, G. Qi, C. Yang, J. Xu, F. Chen, X. Meng, F. Deng, F.-S. Xiao, Angew.
Chem. 125 (2013) 9342–9345.
[29] Y. Du, B. Feng, Y. Jiang, L. Yuan, K. Huang, J. Li, Eur. J. Inorg. Chem. 2018 (2018)
2599–2606.
[30] G.E. Giannetto, G.R. Perot, M.R. Guisnet, Ind. Eng. Chem. Prod. Res. Develop. 25
(1986) 481–490.
[31] F. Alvarez, F.R. Ribeiro, G. Perot, C. Thomazeau, M. Guisnet, J. Catal. 162 (1996)
179–189.
[32] J.E. Samad, J. Blanchard, C. Sayag, C. Louis, J.R. Regalbuto, J. Catal. 342 (2016)
203–212.
[33] S. Zhang, S.-L. Chen, P. Dong, G. Yuan, K. Xu, Appl. Catal. A: Gen. 332 (2007) 46–
55.
[34] M.Y. Kim, K. Lee, M. Choi, J. Catal. 319 (2014) 232–238.
[35] R. Garcia, I.J. Shannon, A.M.Z. Slawin, W. Zhou, P.A. Cox, P.A. Wright, Micropor.
Mesopor. Mat. 58 (2003) 91–104.
[36] Y.-H. Xu, Z. Yu, X.-F. Chen, S.-H. Liu, X.-Z. You, J. Solid State Chem. 146 (1999)
157–162.
The Ni/SAPO-11(s) hydroisomerization catalyst with high dis-
persion of nickel species was prepared via the novel direct synthe-
sis method; the nickel species and acid sites were instantaneously
formed in the catalyst. Three types of nickel species in the catalyst
were identified. The metallic nickel is the main active site for (de)
hydrogenation and framework nickel supplies more acid sites for
the n-hexane hydroisomerization. A great balance between the
metal and the acid sites was achieved via the direct synthesis
method. The isomerization of alkene intermediates on the acid
sites is the rate limiting step. The Ni/SAPO-11(s) catalyst exhibits
one of the highest iso-hexane yields reported in the hydroisomer-
ization of n-hexane. The novel direct synthesis method opens a
new door for preparation of metal supported catalysts.
[37] Y. Meng, H.C. Genuino, C.-H. Kuo, H. Huang, S.-Y. Chen, L. Zhang, A. Rossi, S.L.
Suib, J. Am. Chem. Soc. 135 (2013) 8594–8605.
[38] G. Garbarino, S. Campodonico, A.R. Perez, M.M. Carnasciali, P. Riani, E.
Finocchio, G. Busca, Appl. Catal. A: General 452 (2013) 163–173.
[39] Y. Zhao, B. Liu, R. Amin, Ind. Eng. Chem. Res. 55 (2016) 6931–6942.
[40] X. Gao, H. Mao, M. Lu, J. Yang, B. Li, Micropor. Mesopor. Mater. 148 (2012) 25–
33.
[41] M.H.M. Ahmed, O. Muraza, A.K. Jamil, E.N. Shafei, Z.H. Yamani, K.-H. Choi,
Energy Fuels 31 (2017) 5482–5490.
[42] E. Heracleous, A.F. Lee, K. Wilson, A.A. Lemonidou, J. Catal. 231 (2005) 159–
171.
Acknowledgments
This work was supported by the ‘‘Fundamental Research Funds
for the Central Universities” (Grant No. 16CX06012A), the ‘‘Shan-
dong Province Natural Science Foundation” (Grant No.
ZR2019MB029), and the ‘‘National Natural Science Foundation of
China” (Grant No. 21376267).
[43] C. Ragupathi, J.J. Vijaya, P. Surendhar, L.J. Kennedy, Polyhedron 72 (2014) 1–7.
[44] M.A. Peck, M.A. Langell, Chem. Mater. 24 (2012) 4483–4490.
[45] R.J.O. Mossanek, I. Preda, M. Abbate, J. Rubio-Zuazo, G.R. Castro, A. Vollmer, A.
Gutiérrez, L. Soriano, Chem. Phys. Lett. 501 (2011) 437–441.
Appendix A. Supplementary material
´
[46] A. Sre˛bowata, R. Baran, D. Łomot, D. Lisovytskiy, T. Onfroy, S. Dzwigaj, Appl.
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.jcat.2019.04.031.
Catal. B: Environ. 147 (2014) 208–220.
[47] L. Zhao, Y. Wang, H. An, X. Zhao, Y. Wang, Catal. Commun. 103 (2018) 74–77.
[48] A.N. Kharat, P. Pendleton, A. Badalyan, M. Abedini, M.M. Amini, J. Catal. 205
(2002) 7–15.
[49] A. Dugué, O. Dymshits, L. Cormier, B. Cochain, G. Lelong, S. Belin, A. Zhilin, J.
Non-Crystal, Solids 413 (2015) 24–33.
References
[50] P. Lin, Y. Fu, Spectrosc. Lett. 34 (2001) 83–92.
[51] S.D. Kelly, N. Yang, G.E. Mickelson, N. Greenlay, E. Karapetrova, W. Sinkler, S.R.
Bare, J. Catal. 263 (2009) 16–33.
[1] J. Pastvova, D. Kaucky, J. Moravkova, J. Rathousky, S. Sklenak, M. Vorokhta, L.
Brabec, R. Pilar, I. Jakubec, E. Tabor, P. Klein, P. Sazama, ACS Catal. (2017) 5781–
5795.
[52] Y. Li, B. Zhang, X. Xie, J. Liu, Y. Xu, W. Shen, J. Catal. 238 (2006) 412–424.
[53] S. Das, S. Thakur, A. Bag, M.S. Gupta, P. Mondal, A. Bordoloi, J. Catal. 330 (2015)
46–60.
[54] A. Masalska, Appl. Catal. A: Gen. 294 (2005) 260–272.
[55] S. Tian, J. Chen, Fuel Process. Technol. 122 (2014) 120–128.
[56] L. Guo, Y. Fan, X. Bao, G. Shi, H. Liu, J. Catal. 301 (2013) 162–173.
[57] S. Inagaki, K. Sato, S. Hayashi, J. Tatami, Y. Kubota, T. Wakihara, A.C.S. Appl,
Mat. Interf. 7 (2015) 4488–4493.
[2] K. An, S. Alayoglu, N. Musselwhite, K. Na, G.A. Somorjai, J. Am. Chem. Soc. 136
(2014) 6830–6833.
[3] V.M. Akhmedov, S.H. Al-Khowaiter, Catal. Rev. 49 (2007) 33–139.
[4] N.V. Chekantsev, M.S. Gyngazova, E.D. Ivanchina, Chem. Eng. J. 238 (2014)
120–128.
[5] H. Song, L. Zhao, N. Wang, F. Li, Appl. Catal. A: Gen. 526 (2016) 37–44.
[6] N.A.A. Fatah, S. Triwahyono, A.A. Jalil, A. Ahmad, T.A.T. Abdullah, Appl. Catal. A:
Gen. 516 (2016) 135–143.
[58] D. Arias, l. Campos, D. Escalante, J. Goldwasser, C.M. López, F.J. Machado, B.
Méndez, D. Moronta, M. Pinto, V. Sazo, M.M.R. de Agudelo, J. Mol. Catal. A:
Chem. 122 (1997) 175–186.
[59] X. Yang, H. Ma, Z. Xu, Y. Xu, Z. Tian, L. Lin, Catal. Commun. 8 (2007) 1232–1238.
[60] I. Eswaramoorthi, N. Lingappan, Korean J. Chem. Eng. 20 (2003) 207–216.
[61] M. Guisnet, Catal. Today 218–219 (2013) 123–134.
[62] N. Batalha, L. Pinard, C. Bouchy, E. Guillon, M. Guisnet, J. Catal. 307 (2013) 122–
131.
[7] G. Ye, Y. Sun, Z. Guo, K. Zhu, H. Liu, X. Zhou, M.-O. Coppens, J. Catal. 360 (2018)
152–159.
[8] C.-H. Geng, F. Zhang, Z.-X. Gao, L.-F. Zhao, J.-L. Zhou, Catal. Today 93–95 (2004)
485–491.
[9] S. Parmar, K.K. Pant, M. John, K. Kumar, S.M. Pai, B.L. Newalkar, J. Mol. Catal. A:
Chem. 404–405 (2015) 47–56.
[10] Y. Wang, Z. Tao, B. Wu, J. Xu, C. Huo, K. Li, H. Chen, Y. Yang, Y. Li, J. Catal. 322
(2015) 1–13.