Synergistic effect of Fe and Ga incorporation into ZSM-5 to increase propylene production in the cracking ofn-hexane utilizing a microchannel reactor

Article abstract of DOI:10.1039/d1nj01866c

In the present study, the effect of various amounts of Fe and Ga in the catalytic cracking ofn-hexane in a microchannel reactor was investigated using experimental design by the D-optimal method. Nano zeolites incorporated with Fe and Ga metals were synthesized in a fluorine environment to investigate the synergistic effect of the metals on the textural and acidic properties of the catalysts, which ultimately improved the performance of the synthesized catalysts in the efficient production of light olefins, in particular propylene. Three synthesis parameters including the Si/Al, Si/Fe and Si/Ga ratios were considered as the main factors to determine the optimal conditions for obtaining the maximum conversion ofn-hexane, yield of light olefins, and P/E ratio and minimum yield of alkanes as the responses. In sample FeGa-1, the P/E ratio reached 3.97, indicating the significant effect of the substituted metals in improving the desirable routes for propylene production. According to the results of the acidic properties, Fe, Al and Ga increased the number of total acid sites and the strengths of strong and weak acid sites, respectively. In addition, according to the results obtained from sample FeGa-7, the synergistic effect of Fe and Ga increased the number of weak acid sites.

Full text of DOI:10.1039/d1nj01866c

NJC
View Article Online
View Journal | View Issue
PAPER
Synergistic effect of Fe and Ga incorporation into
ZSM-5 to increase propylene production in the
cracking of n-hexane utilizing a
Cite this: New J. Chem., 2021,
45, 13833
microchannel reactor
Mohsen Rostami Sakha,ab Saeed Soltanali, *b Darush Salari,a Mehdi Rashidzadehb
and Parya Halimitabriziac
In the present study, the effect of various amounts of Fe and Ga in the catalytic cracking of n-hexane in
a microchannel reactor was investigated using experimental design by the D-optimal method. Nano
zeolites incorporated with Fe and Ga metals were synthesized in a fluorine environment to investigate
the synergistic effect of the metals on the textural and acidic properties of the catalysts, which
ultimately improved the performance of the synthesized catalysts in the efficient production of light
olefins, in particular propylene. Three synthesis parameters including the Si/Al, Si/Fe and Si/Ga ratios
were considered as the main factors to determine the optimal conditions for obtaining the maximum
conversion of n-hexane, yield of light olefins, and P/E ratio and minimum yield of alkanes as the
responses. In sample FeGa-1, the P/E ratio reached 3.97, indicating the significant effect of the
substituted metals in improving the desirable routes for propylene production. According to the results
of the acidic properties, Fe, Al and Ga increased the number of total acid sites and the strengths of
strong and weak acid sites, respectively. In addition, according to the results obtained from sample
FeGa-7, the synergistic effect of Fe and Ga increased the number of weak acid sites.
Received 16th April 2021,
Accepted 27th June 2021
DOI: 10.1039/d1nj01866c
rsc.li/njc
propylene production process due to emission of greenhouse
gases such as CO2, high energy consumption at high temperatures
1. Introduction
As expected, light olefins with a trade of about $1 trillion by (4800 1C) and the low propylene/ethylene ratio.12 One of the
2025 are one of the main chemical products. Meanwhile, the approaches to overcome the limitations of this process is the
global demand for propylene surpasses that of ethylene as the utilization of stable and efficient catalysts.
U.S.A’s second largest produced chemical, doubling the importance
Shape selective catalysts as solid acids are suitable choices
of increasing the production efficiency of this raw material widely for this process because they increase the selectivity to the
used in the production of chemicals, including polypropylene, desired products by sieving the reactants, intermediates and
acrylonitrile, propylene epoxide and acetone.1–5 Conventional products based on their pore structure and size.13–16 The ZSM-5
methods for light olefin production include methanol to olefin zeolite, which consists of tetrahedral units causing negative
conversion (MTO), steam cracking of naphtha (SCN) and fluid charges in the lattice by bonding Si atoms to Al atoms via
catalytic cracking (FCC).6–9 As propylene is a byproduct of olefin oxygen bridges, is the most commonly used MFI topology in
production processes, the need for technologies which increase the isomerization, cracking, aromatization and alkylation
production efficiency of propylene is felt. In other words, processes.17–19 By neutralizing the negative charges by cations
on-purpose production of propylene should be achieved through (K+, Na+, Cs+, Li+ and H+), Brønsted acid sites (BAS) will
using novel reactors and catalysts.10,11 The SCN process cannot form.20,21 ZSM-5 has 12 orthorhombic distinct T-sites in which
meet the needs of the global market for propylene as the main sites T4, T10 and T8, T11 are accessible on the sinusoidal and
straight channels, respectively.14,22
The framework Al position and distribution in high-Si
a Reactor and Catalysis Research Lab., Department of Chemistry,
zeolites are not random or controlled by statistical rules.23,24
University of Tabriz, Tabriz, Iran
b Catalysis Technologies Development Division, Research Institute of Petroleum
Since most of the acid sites are Al-dependent, their positioning
at different T-sites affects the catalytic performance of the
Industry (RIPI), Tehran, Iran. E-mail: sssoltan@gmail.com, soltanalis@ripi.ir
c Department of Chemical and Petroleum Engineering, University of Tabriz, Tabriz,
zeolites.25 Al forms Lewis acid sites (LAS) by being located
outside the zeolite structure (extra-framework Al).13,26,27
Iran
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem., 2021, 45, 13833–13846 | 13833

View Article Online
Paper
NJC
Table 1 Investigated parameters and their levels
Framework Al changes the accessibility of steric molecules as
well as the acidic strength by incorporating in different
T-sites.28 BAS (tetrahedral framework Al) alter the enthalpic
and entropic effects and selectivity of products through incor-
poration in straight and sinusoidal channels or intersections.21
Investigations have proved that the incorporation of BAS in the
channels encourages an alkene-based cycle and propylene
selectivity in the MTO process. In addition, the propylene
selectivity was also increased in the cracking of alkanes and
alkenes by incorporation of Al in the channels.19,29–33 The
combination of using TPAOH, colloidal silica,30,31,34
NH4F11,35,36 and Na+ as the organic structure directing agent
(OSDA), Si source, mineralizer and inorganic compensating
cation, respectively, results in more incorporation of Al and
possibly other metals in the channels.10,17,30,37
Conventional zeolites are predominantly microporous, limiting
the transport of reactants, intermediates and products. Various
methods including the preparation of nano-sized, mesoporous and
hollow zeolites have been developed to overcome the limitations of
mass transfer.38 Mesoporous structures increase the activity and
selectivity to the desired products by facilitating access to acid sites
and rapid exit of the main products produced, and avoiding
side reactions and formation of aromatics as precursors of coke
deposition. On the other hand, the size of the cavities formed in
hierarchical structures should not be too large because larger
cavities will form larger molecules such as BTX and decrease the
propylene selectivity.2,3,8,31,39–44 Hard-templating, post-treatment
and soft-templating are common methods to form hierarchical
structures. Various groups have recently reported novel methods
for the synthesis of hierarchical ZSM-5 by optimization of the
gelation and crystallization conditions as well as synthesis in
fluoride medium as the nucleation and/or mineralization
agent.31,34,45
Factor
Name
Type
Low level (ꢀ1)
High level (+1)
A
B
C
Si/Al
Si/Fe
Si/Ga
Numerical
Numerical
Numerical
100
100
100
400
400
400
achieving the optimal catalyst composition is significantly
increased. This in turn leads to an increase in the use of raw
materials, and, due to the possibility of consumables being
toxic in the production of catalysts, increasing the amount
of consumption of these materials can lead to irreparable
damage to the environment. Experimental design methods
define an experimental area (the range of parameter values)
in which a mathematical model is defined to estimate the effect
of the survey parameters. The D-optimal experimental design
method, as the most up-to-date method, is used in cases where
the test areas are irregular, pre-designed tests should be added
to the design tables, the quantitative factors have more than 2
levels, the number of tests should be reduced and special
regression models should be adapted, and process and mixing
factors must be used simultaneously in a design.58–60
In this investigation, the synergistic effect of Fe and Ga
incorporation into the zeolite structure in the catalytic cracking
of n-hexane as well as the acidic and structural properties have
been investigated for the first time. Microchannels were used as
novel reactors to increase the activity and stability of the
obtained catalysts by enhancing mass and heat transfer. XRD,
FE-SEM, EDX elemental mapping, BET, NH3-TPD and H2-TPR
analyses were applied to determine the intrinsic properties of
the synthesized catalysts. The D-optimal method was also used
in the experimental design of 12 experiments to demonstrate
the effect of the Si/Al, Si/Fe and Si/Ga factors on the four
responses. The valid interactions of the variable factors in the
conversion, yield of light olefins, P/E ratio and yield of alkanes
were determined as the responses. Moreover, the optimum
conditions for maximum production of propylene via n-hexane
cracking were estimated.
One of the main approaches to increase the yield of light
olefins and the lifetime of zeolites is the modification of the
acidity using various elements. Conventional methods for
metal modification are impregnation (wet or dry) and ion
exchange. Another method to modify the structural and acidic
properties of zeolites is incorporation. Meanwhile, the ion
exchange and impregnation methods often cause the formation
of extra-crystalline metal oxides.46 The incorporation of Fe and
Ga improves the selectivity of light olefins and the stability of
ZSM-5 by reducing the acidic strength. On the other hand, the
higher distribution of extra-framework Ga species, which exist
Table 2 The responses and experimental design for three independent
variables
Factors
Responses
Alkane
(C=2 + C=3)
in various forms of Ga+, GaO+, Ga2O22+, GaH2+, GaH2+
,
Run
A
B
C
Conversion (%) yield (%) yield (%) P/E
Ga(OH)2 and GaH(OH)+, improves the selectivity of arenes by
enhancing dehydrogenation reactions near protonic sites.47–57
Experimental design methods have been used in catalytic
fields for many years, including the design of catalytic reactors,
the formulation and preparation of catalysts, and the kinetic
modeling of catalysts. The purpose of experimental design can
be achieved by repeating the experiments several times so that
each of the parameters which affect the synthesis can be
examined alone along with constant consideration of the other
parameters. This means that without the use of experimental
design, the required time and labor during the process of
+
1
2
3
4
5
6
7
8
400 220 279 71.4
400 400 100 89.6
221 400 279 93.6
213 213 287 85.3
310 317 175 87.3
291 100 400 73.6
400 100 100 79.8
222 222 100 96.0
100 100 203 84.4
100 400 100 97.3
100 100 395 84.4
100 294 400 95.3
400 400 400 79.2
18.9
22.4
27.2
25.1
20.5
19.1
18.1
25.8
22.9
27.9
23.3
30.4
23.5
37.3
51.2
51.1
44.7
50.1
39.4
44.4
52.0
45.4
52.9
45.1
49.0
40.0
3.97
2.83
2.52
2.97
3.62
3.48
3.52
2.13
3.48
1.92
2.74
1.95
3.20
9
10
11
12
13
13834
|
New J. Chem., 2021, 45, 13833–13846
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021

View Article Online
NJC
Paper
2. Experimental
2.1. Catalyst preparation
Zeolites were hydrothermally synthesized in fluoride medium.
Aluminum sulphate (Al2(SO4)3ꢁ18H2O, Merck Chemical Co.)
and colloidal silica (40 wt% SiO2, Aldrich Chemical Co.) were used
as Al and Si sources, respectively, to increase the incorporation of Al
in the framework. The main molar composition of the solutions
was
SiO2:xAl2O3:yFe2O3:zGa2O3:0.2TPAOH:0.19Na2O:0.205NH4F:
32H2O (x, y and z were calculated from Tables 1 and 2). Firstly, a
solution containing NH4F, aluminum(III) sulfate, iron(III) nitrate
and GaCl3 (their amounts were carefully calculated according to the
different ratios in Table 2 and added to the solution following
NH4F at the same time) was prepared and then the solution was
stirred at ambient temperature for 20 minutes. After obtaining a
clear solution, colloidal silica, NaOH, and TPAOH were added to
form a white gel. The gel was then aged at room temperature for
24 h at 300 rpm. The zeolites were crystallized dynamically at
180 1C for 72 h at 300 rpm followed by centrifuging, drying and
calcination at 550 1C for 6 h. H-ZSM-5 was obtained by ion-
exchanging the powder with NH4NO3 (1 M) at 80 1C two times,
followed by calcination at 550 1C for 6 h.
Fig. 1 XRD patterns of HZSM-5 catalysts synthesized using various
amounts of Al, Fe and Ga.
to 100 1C. Afterwards, the N2 flow was changed to a mixture of
H2/N2. The TPR profiles were then obtained by reducing the
samples under the mixture of H2 and N2 pulses (H2/N2 = 10%)
with temperature raised to 700 1C at a ramping rate of
10 1C minꢀ1
.
2.3. Catalytic performance
2.2. Characterization
Catalytic tests for the catalytic cracking of n-hexane were
performed in a reactor with 5 microchannels with dimensions
of 50 ꢂ 1 ꢂ 1 mm at T = 600 1C and atmospheric pressure under
an N2 flow as the carrier gas. Each time the test was performed,
0.125 g of meshed zeolite (sieve fraction 150–250 mm) was
packed in the microchannels. The catalyst was activated
in situ by heating at 550 1C for 2 h in a 30 cc per min N2 flow.
The outlet gas was analyzed after an hour’s reaction. The
products were analyzed using an online gas chromatograph
(Shimadzu 2010 Plus GC) equipped with a flame ionization
detector (FID) with an HP-PLOT Al2O3 capillary column, every
hour during the 6 h of reaction.
XRD (X-ray diffraction) patterns were acquired on an Inel
EQUINOX 3000 X-ray diffractometer with Cu Ka radiation at
40 kV and 100 mA. All spectra were collected in the 2y = 7–501
range with a step size of 0.021.
Scanning electron microscopy (SEM) and energy dispersive
X-ray analysis (EDX) were performed to determine the presence
of various elements and the morphology of the zeolites. SEM and
EDX maps were obtained using a Zeiss Sigma VP field emission
scanning electron microscope (FE-SEM).
The textural properties of the synthesized catalysts were
characterized using a Micromeritics instrument from the BET
model by N2 adsorption–desorption at 77 K. The samples were
first outgassed under a vacuum (o0.5 Pa) for 7 h before N2
physisorption. The external surface areas were determined by
2.4. D-Optimal experimental design
the Brunauer–Emmett–Teller (BET) method using the In order to improve the properties of ZSM-5 and optimize the
desorption data in the relative pressure (p/p0) range of 0.01– amount of substituted metals in the structure of the synthesized
0.99. The pore size distribution and the contribution from both catalysts, experimental design was used with three numerical
micro- and mesoporosity were derived using the t-plot and the factors due to advantages such as reduced costs, a decreased
Barrett–Joyner–Halenda (BJH) model.
number of tests, a comprehensive review of the system under
NH3 temperature-programmed desorption (NH3-TPD) and study and the capability to optimize the desired values of the
H2 temperature-programmed reduction (H2-TPR) were carried answers with the input variables. In this research, the D-optimal
out using a Micromeritics TPR/TPD 2900 chemisorption analyzer. design method was applied to optimize the quantitative value of
NH3-TPD and H2-TPR were used to determine the number and the Si ratio to each of the substituted metals (Al, Fe and Ga) and
strength of acid sites and evaluate the reducibility of the estimate the performance of the prepared catalysts in the
synthesized catalysts, respectively. For NH3-TPD evaluation about catalytic cracking of n-hexane. The main and interaction effects
0.2 g of zeolite was pre-treated under a flow of He at 550 1C for 1 h along with the analysis of variance (ANOVA) on the conversion,
and then cooled down to 150 1C. Afterwards, ammonium was yield of alkanes, yield of propylene and ethylene and P/E
introduced into the flow system. Finally, the TPD curves were ratio were evaluated as quantitative responses. As shown in
acquired using the NH3 saturated sample in the temperature Tables 1 and 2, Design-Expert software (Version 7.0.0) was used
range of 150–600 1C at a ramping speed of 10 1C minꢀ1. As for to evaluate the responses of the Si/Al (A), Si/Fe (B) and Si/Ga
the H2-TPR investigation, 0.2 g of zeolite was pre-treated at (C) factors with a sum of 12 experiments. The D-optimal
500 1C under an N2 flow (30 mL minꢀ1) and then cooled down design of experiment and experimental results are displayed in
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem., 2021, 45, 13833–13846 | 13835

View Article Online
Paper
NJC
Fig. 2 FE-SEM images of the zeolites.
13836
|
New J. Chem., 2021, 45, 13833–13846
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021

View Article Online
NJC
Paper
Table 3 Particle size of the synthesized zeolites obtained from SEM
images
Catalyst sample
Particle size (mm)
FeGa-1
FeGa-2
FeGa-3
FeGa-4
FeGa-5
FeGa-6
FeGa-7
FeGa-8
FeGa-9
FeGa-10
FeGa-11
FeGa-12
FeGa-13
6.5
7.7
9.2
7.7
1.3
4.2
4.9
6.6
8.8
11.5
9.5
9.1
3.2
Tables 1 and 2. The experimental results were obtained by carrying
out the experiments randomly to avoid any systematic bias.
Fig. 4 N2 adsorption–desorption isotherms of the samples.
the amounts of Fe and Al in the zeolite structure reduces the
particle size by strengthening the nucleation and growth rate of
the crystals (FeGa-5).47 The atomic radii of Fe, Al, and Ga are
126, 143, and 135 pm, respectively. Considering the various
atomic radii of the metals, the particle size increased with the
increment of Al and Ga, but increasing Fe reduced the average
atomic radii of the formed particles and the particle size
decreased dramatically. In addition, careful observation of the
SEM images shows that no cloud-like layers were formed, which
generally indicates the presence of impurities and/or the
formation of metal oxides on the surface of the particles. As
observed in Fig. 3, the obtained maps of FeGa-13 synthesized
using equal amounts of metals demonstrated the symmetric
distribution of all metals in different parts of the particles.
According to the IUPAC classification, all synthesized zeolites
demonstrated a combination of type I and IV isotherms (Fig. 4)
accompanied by capillary condensation due to the sharp
uptake at very low pressure and hysteresis loop at higher
pressure (p/p0 4 0.4).45,47,62 The combination feature of the
isotherms indicates the coexistence of narrow micropores
(o1 nm) and mesopores (o2.5 nm) as well as wider mesopores
(44 nm), which is mainly due to the formation of homo-
3. Results and discussion
3.1. Textural properties
In the X-ray patterns (Fig. 1), the peaks corresponding to Fe and
Ga oxides are not visible, which can result from the incorporation
of metals into the zeolite framework or at least the excellent
distribution of metals in the cavities and surface. In addition,
the typical peaks of the MFI topology structure were observed,
confirming the successful synthesis of pure zeolites.61
According to the SEM images (Fig. 2), all the synthesized
zeolites with various Si/M ratios showed the same uniform
spherical morphology resulting from the aggregation of nano-
spheres. The comparison of the particle sizes in Table 3 showed
that reducing the Si/Al ratio simultaneously with decreasing the
Si/Ga ratio caused a significant increment in the particle size
(FeGa-10). In contrast, reducing the Si/Ga ratio individually led
to a much smaller increase in the particle size (FeGa-2). On the
other hand, the simultaneous decrease of the Si/Ga and Si/Fe
ratios did not have a considerable effect on the particle size,
which probably indicates the synergistic nature of Fe and Ga
metals in the reduction of the particle size (FeGa-7). It seems
that increasing the amount of Ga along with a slight growth in
Fig. 3 EDX maps of FeGa-13 for the Al, Fe and Ga elements.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem., 2021, 45, 13833–13846 | 13837

View Article Online
Paper
NJC
Table 4 Textural properties of the zeolite samples
Table 5 Concentration of weak and strong acid sites according to NH3-
TPD
Catalyst
SBET
Sext
Sext/SBET
Vtc
Vmic
Vmes
df
a
b
d
e
a
Total acid
Weak acid Strong acid TW
FeGa-2
FeGa-7
FeGa-9
FeGa-10
FeGa-13
392
391
413
400
378
282
280
296
289
255
0.71
0.71
0.71
0.72
0.67
0.180
0.180
0.197
0.180
0.213
0.059
0.059
0.066
0.062
0.056
0.121
0.121
0.131
0.118
0.157
1.88
1.90
1.90
1.86
2.25
Catalyst (mmol gꢀ1
)
(mmol gꢀ1
)
(mmol gꢀ1
) (1C) TS (1C) W/S
b
FeGa-2 0.140
FeGa-5 0.043
FeGa-7 0.178
FeGa-9 0.187
FeGa-10 0.122
FeGa-13 0.082
0.090
0.027
0.117
0.127
0.087
0.053
0.050
0.016
0.061
0.060
0.040
0.029
210 414
194 386
202 409
203 424
201 417
197 393
1.80
1.68
1.91
2.11
2.05
1.83
a
b
The specific surface area (m2 gꢀ1). The external surface area (m2 gꢀ1).
c
e
d
The total pore volume (cm3
g
ꢀ1). The micropore volume (cm3
g
ꢀ1).
The mesopore volume (cm3 gꢀ1). The average pore diameter (nm).
f
a
b
The temperature related to the weak acidity zone. The temperature
related to the strong acidity zone.
geneous nanocrystals. Moreover, according to Table 4, the
distribution of the micropore and mesopore volumes of the
samples confirms the pore size distribution and also the
simultaneous formation of type I and IV isotherms. This is in
agreement with the SEM results. Moreover, all synthesized
zeolites using different amounts of metals displayed a sharp
desorption peak in the range of p/p0 4 0/4, which is one of the
main features of H2 type hysteresis loops. The combination of
H2 and H4 type hysteresis loops (due to more pronounced
uptake at low p/p0) indicates the formation of smaller mesopores
and more interconnections between the crystals. In addition, the
combination of hysteresis loops represents neck obstruction of
the micropores and mesopores, most likely due to the formation
of non-framework species with the presence of Fe and Ga, which
have different inherent properties from Al. In addition, the H4
type hysteresis loop confirms the accumulation of ultra-uniform
nanocrystals, which is verified by the SEM results. Smaller
mesopores were also evident in the pore size distribution profiles
because the samples had a smaller number and smaller pore
size distribution, with the exception of sample FeGa-13. In FeGa-
13, Fe, Ga, and Al were probably more evenly distributed than
the other samples due to the equal amount of metals.
amounts of Fe, Ga and Al. According to Tables 2 and 5, with
increasing the total amounts of substituted metals, the total
acidity increased remarkably (except for sample FeGa-5).
Samples FeGa-13, FeGa-7 and FeGa-10 clearly illustrated an
increase in the number of total acid sites with increasing the
amount of metals. FeGa-7 and FeGa-10 had a constant Si/Ga
ratio, but as the amount of the other metals changed, the total
acid sites increased. The comparison of the total acid sites of
the two samples (0.178 vs. 0.122 mmol gꢀ1) demonstrated that,
as the amount of Fe increased, the total acidity increased to a
much greater extent than with the increment of Al. The
comparison of the corresponding temperature to the weak acid
site (WAS) zone (150–250 1C) also showed that, with increasing
the amount of metals, the strength of WAS increased due to the
desorption of ammonium at higher temperature. The comparison
of samples FeGa-7 and FeGa-10 with equal Si/Ga ratios showed
that the Fe and Al metals alone do not play a role in enhancing the
strength of WAS since the alteration in the amount of Fe and Al
did not cause a significant difference in the ammonium
desorption temperature corresponding to the WAS. However,
according to FeGa-2, increasing the amount of Ga increased the
strength of the WAS. Therefore, the amount of Ga is an influential
factor to enhance the strength of WAS, which is likely to improve
olefin production processes. The comparison of the ammonium
desorption temperature corresponding to the strong acid site
(SAS) zone (375–500 1C) showed that the strength of SAS, which
are predominantly related to the strong Brønsted acid sites
(SBAS),30,53 increased with increasing the amount of metals due
to the temperature shift toward higher temperatures. As observed
in the NH3-TPD plot of samples FeGa-7 and FeGa-10, the
temperatures corresponding to the SAS zone were equal to 409
and 417 1C, respectively, which indicates increasing the strength
of SAS in the sample with an increased amount of Al (FeGa-10).
The comparison of the number of WAS and SAS of the synthesized
zeolites using different ratios of Fe, Ga and Al showed attractive
consequences from the synergistic effect of Fe and Ga atoms.
According to Tables 2 and 5, increasing Ga alone led to a
significant increase in the number of WAS in sample FeGa-2.
On the other hand, increasing the amount of Al simultaneously
with Ga in equal amounts reduced the number of weak, strong
and total acid sites in the FeGa-10 sample. It was expected that in
this sample, due to the higher amount of metals incorporated into
the zeolite, the formation of more acid sites would be observed,
but, contrary to expectations, not only did the amount of total acid
3.2. Acidic properties
Fig. 5 and Table 5 display the results obtained from NH3-TPD
analysis of the synthetic zeolites with different incorporated
Fig. 5 NH3-TPD plot of HZSM-5 catalysts synthesized using various
amounts of Al, Fe, and Ga.
13838
|
New J. Chem., 2021, 45, 13833–13846
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021

View Article Online
NJC
Paper
these areas indicated the major incorporation of metals into
the structure of the synthesized zeolites and/or a lack of
reducible extra-framework gallium species. This phenomenon
has occurred due to the presence of NH4F as a mineralizer
agent as well as low amounts of Fe and Ga metals. The use of
this compound results in the formation of (NH4)3GaF6 solution, the
hydrolysis of which forms Ga(OH)3. The complete incorporation
of gallium hydroxide in the hydroxyl cavities and/or tetrahedral
structure of the zeolites has reduced or eliminated gallium oxides
or extra-structural species. Reducible extra-framework gallium
species, the absence of which is one of the intrinsic reasons for
the low selectivity of aromatics, are the active acid sites for the
production of aromatics via dehydrogenation processes. Therefore,
it can be claimed that the maximum modifications have been
made to reduce the production of aromatics. It should be noted
that the results of this analysis confirm those of the XRD and SEM
analyses.47
Fig. 6 H2-TPR profiles of FeGa-5, FeZSM-5 and GaZSM-5.
3.3. Analysis of variance
According to Table 6, a total of 12 experiments were performed
using the D-optimal method. Significant and insignificant
parameters can be determined using the Fischer variance ratios
of the parameter effect dispersion to the error dispersion. The
F-test is used to compare the statistical models with the
experimental data. The higher the fit between the model and
the data, the higher the exact value of F. In this research, a 95%
confidence interval has been used to evaluate the significance
of the parameters. Therefore, a probability value (p-value) of
0.05% would be a significant level in the F-tests for the inter-
pretation of the effects. In the present study, analysis of
variance was used to quantitatively investigate the effects of
the input variables on the responses. The results are presented
in the ANOVA table (Table 6).
The suitable R2 and high F-value of the model imply the
ability of the models for prediction of responses. F shows the
probability of model error; therefore, a small value of F
indicates that the model is correct. The polynomial predictive
models for the responses developed by the model are presented
in Table 7, based on the coded value.
sites decrease compared to sample FeGa-2, but the numbers of
SAS and WAS were also reduced. This phenomenon indicates the
undeniable effect of the Ga metal on the increasing active sites,
especially WAS, in the synthesis of ZSM-5 zeolites. In sample
FeGa-7, decreasing the amount of Al and increasing the amount
of Fe in equal proportions increased the number of total acid sites
as well as significantly increasing the number of WAS as the main
active centers to direct the reaction path toward the production of
light olefins, especially propylene.
Temperature programmed reduction via hydrogen analysis
was used to determine the distribution of Ga and Fe atoms. The
comparison of the H2-TPR curves of sample FeGa-5 and zeolites
incorporated with Fe and Ga metals alone is shown in Fig. 6.
According to the diagram, no peaks corresponding to Fe and Ga
species were observed. The reduction of extra-framework GaO+
to Ga+ at temperatures above 710 1C has clear peaks.
In addition, the reduction of Ga2O3 at a lower temperature in
the range of 500–570 1C possesses an obvious peak. The
reduction of iron oxides also shows peaks in the temperature
ranges of 300–400 1C and 400–700 1C. The absence of peaks in
Table 6 ANOVA result for the model
Conversion (R1)
F value
Yield (C1–C4) (R2)
F value
Yield (Cꢀ2 –Cꢀ3 ) (R3)
P/E (R4)
F value
Source
p-value
p-value
F value
p-value
p-value
Model
A
B
C
82.15
316.31
205.33
87.62
—
o0.0001
o0.0001
o0.0001
0.0002
—
23.59
33.52
20.76
—
0.0002
0.0002
0.0010
—
47.52
102.10
81.72
90.33
—
o0.0001
o0.0001
0.0001
o0.0001
—
13.44
23.45
12.63
—
0.0011
0.0009
0.0062
—
AB
—
—
—
—
AC
40.86
20.17
36.79
—
0.0014
0.0065
0.0018
—
—
—
—
—
—
—
—
—
31.60
—
21.48
—
0.0014
—
0.0036
—
—
—
—
—
—
—
—
—
BC
A2
B2
C2
27.63
0.0033
—
—
6.71
0.0412
9.14
0.0144
Model statistics
R2
0.9914
0.9793
0.9103
0.8251
0.7901
0.7238
0.9794
0.9588
0.8855
0.8175
0.7567
0.6587
R2 adj.
R2 pred.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem., 2021, 45, 13833–13846
|
13839

View Article Online
Paper
NJC
Table 7 Final models in terms of actual factors for responses
Responses
Final model
Conversion (R1)
Yield (C1–C4) (R2)
Yield (C=2–C=3) (R3)
P/E (R4)
R-square 0.9914
R1 = + 86.04 ꢀ 7.94 ꢂ A + 6.38 ꢂ B ꢀ3.99 ꢂ C ꢀ 3.34 ꢂ A ꢂ C
+ 2.35 ꢂ B ꢂ C ꢀ 4.83 ꢂ A2+ 4.35 ꢂ C2
Adjusted R-square 0.9793
Predicted R-square 0.9103
R-square 0.8271
Adjusted R-square 0.7901
Predicted R-square 0.7238
R-square 0.9794
Adjusted R-square 0.9588
Predicted R-square 0.8855
R-square 0.8175
Adjusted R-square 0.7567
Predicted R-square 0.6581
R2 = + 23.47 ꢀ 3.49 ꢂ A + 2.74 ꢂ B
R3 = + 46.87 ꢀ 3.77 ꢂ A + 3.50 ꢂ B ꢀ 3.53 ꢂ C ꢀ 2.55 ꢂ A ꢂ C
ꢀ 3.18 ꢂ A2 + 1.84 ꢂ C2
R4 = + 3.36 + 0.56 ꢂ A ꢀ 0.41 ꢂ B ꢀ 0.63 ꢂ C2
3.3.1. Regression model. To implement the D-optimal estimated coefficients. A combination of factors (such as XiXj)
method, a mathematical model must first be selected to represents the interactions between the individual factors in
correlate the independent variables with the dependent ones. that term.
P
In this research, the most common method, namely the
k
k
X
X
X
quadratic model of the equation, was used. In D-optimal
design, it is possible to modify the quadratic model in any
way. Thus, in the case of insignificant effects, some parameters
can be removed without negatively affecting the results.
Y ¼ b0 þ
biXi þ
biiXi2
þ
bijXiXj þ e
(1)
i o j
i¼1
i¼2
The experimental results and the predicted coefficients of the
The general form of the quadratic model is presented as equation for the system responses are shown in Fig. 7.
eqn (1). In this equation, Y is the studied response, Xi and Xj are As observed, the predicted coefficients are in good agreement
the variables considered in the study and b0, bi and bji are the with the experimental values obtained from the experiments
Fig. 7 Comparison of the experimental and predicted values using the equation of the models.
13840
|
New J. Chem., 2021, 45, 13833–13846
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021

View Article Online
NJC
Paper
Fig. 8 Effect of the Si/Al main parameter on the responses.
with respect to the appropriate values obtained for the ratio increased the selectivity of ethylene, but decreased the
R-square, and adjusted and predicted R-square parameters selectivity of propylene, while decreasing the W/S ratio led to
(Table 7). Therefore, these models are the basis for interpreting increased selectivity of propylene and decreased selectivity of
the obtained relations Y = f (A, B, C).
ethylene. On the other hand, the decrease of Al in ZSM-5
decreases the W/S ratio, which is probably due to the reduction
of non-framework Al.67 Extra-structural Al increases the W/S
ratio by increasing the formation of weak LAS. Therefore, in
the synthesized zeolites, with decreasing the Al amount, the W/S
ratio decreased, leading to the increase of the P/E ratio. As shown
in Fig. 8, the reduction of the Al amount reduced the yield of
light olefins by reducing the total number of acid sites, especially
the weak Brønsted and Lewis acid sites, which are the most
considerable active centers in the formation of light olefins by
partial dehydrogenation reactions.10
(a) Si/Al effect. According to Table 6, the amount of Al has
the strongest impact on all responses because the relevant
corresponding F-value is higher than all other factors. In fact,
SAS, which are mainly composed of skeleton Si–O–Al, are the
most important factors in the study of a catalytic process.26,63
According to Fig. 8, the conversion of hexane was diminished
due to the increase of the Si/Al ratio from the approximate
range of 200, which can be mainly related to the substantial
reduction in the number and strength of SBAS by decreasing Al
in the structure.32,64 The yield of light alkanes was also reduced
by increasing the Si/Al ratio due to the reduction in the strength
(b) Si/Fe effect. Table 6 shows that the amount of Fe affects
and amount of acid sites, especially SBAS, which are the main all responses. However, it is less effective in all responses
centers of hydrogenation/protonation reactions and cracking of compared with Al and even less effective in the production
heavy alkanes to lighter alkanes.65,66 According to Fig. 8, the efficiency of light olefins compared with Ga, mainly due to the
production efficiency of propylene and ethylene decreased with lower dissociation ability of protons in Si–OH–Fe.
reducing the amount of Al. On the other hand, the P/E ratio
As shown in Fig. 9, decreasing the amount of Fe in the
increased. According to recent studies, the selectivity of ethylene synthesized zeolites led to the linear enhancement of the
and that of propylene act in the opposite direction, mainly due to catalytic activity. In fact, this result showed that increasing
the difference in the weak/strong acid sites. Increasing the W/S the amount of Fe reduced the number and/or strength of SBAS
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem., 2021, 45, 13833–13846 | 13841

View Article Online
Paper
NJC
Fig. 9 Effect of the Si/Fe main parameter on the responses.
as the main sites of alkane cracking.54 Recent studies showed which means the increase of ethylene production and decrease of
that increasing Fe reduced the WAS and SAS and increased the propylene production with the reduction of Fe. Reduction of Fe
moderate acid sites in ZSM-5, which ultimately increased the with increasing the partial dehydrogenation and methylation
total acid sites. Therefore, the main increase in the number of increased the yield of propylene and ethylene. However, lighter
acid sites is related to the moderate and WAS, which have a products such as ethylene were produced more due to the higher
limited effect on the process of cracking of alkanes. The strength of SBAS.
principal presence of Fe in the structure of the zeolites was
confirmed by H2-TPR analysis. Therefore, the only reason for
(c) Si/Ga effect. According to Table 6, the amount of Ga had
decreasing the catalyst activity by increasing Fe was the the lowest effect on the responses and probably had no effect
decreased strength of SBAS, which was related to the increase on the yield of alkanes and the P/E ratio, mainly due to the
of the H+ bond strength in Si–OH–Fe.53,68 In fact, it seems that unique intrinsic properties of Ga, such as the low cation
with increasing Fe the proton bond in the tetrahedral structure of electronegativity.70
MFI becomes stronger and its separation requires more energy for
The amount of Ga in the catalytic cracking of hexane
hexane cracking by hydrogenation processes, ultimately leading to demonstrated a much more complex effect than other metals.
the reduction of hexane cracking.69 Moreover, the reduction of Fe According to Fig. 10, increasing the Si/Ga ratio to the approximate
increased the yield of alkanes probably due to the increased range of 300 reduced the catalyst activity. However, upon a further
hydrogenation reactions caused by SBAS. Indeed, the reduction decrease of Ga, the hexane conversion increased. In synthetic
of Fe by increasing the strength of acid sites due to structural zeolites, Ga is placed in three forms of tetrahedral substituted in
proton cleavage increased the hydrogen transfer reactions and the structure, and gallium oxides and extra-framework Ga species
consequently the cracking of hexane to lighter alkanes. According in the cavities, the last being the main acid site for dehydro-
to Fig. 9, the reduction of Fe increased the production efficiency of cyclization reactions in the olefin and aromatic production
light olefins. In addition, according to the results of hexane processes.71–73 Investigations revealed that decreasing the amount
conversion and the yield of alkanes, it seems that, with the of Ga in synthesized ZSM-5 increased the amount of gallium
reduction of Fe, the strength of SBAS increased, which showed oxides and extra-framework Ga species.74 Therefore, the decrease
its effect in increasing the selectivity of ethylene. As shown in of Ga to the ratio of Si/Ga = 300 increases the formation of extra-
Fig. 9, the P/E ratio decreased with increasing the Si/Fe ratio, framework species. However, by creating extra-framework species,
13842
|
New J. Chem., 2021, 45, 13833–13846
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021

View Article Online
NJC
Paper
Fig. 10 Effect of the Si/Ga main parameter on the responses.
hexane cracking may be improved at the surface centers before ultimately increasing the catalyst activity even though the
entering the cavities, which ultimately enhances the catalyst amount of Ga was reduced in the upper limit, which confirms
activity. According to Fig. 10, with increasing the Si/Ga ratio to the effect of Fe.
approximately 260, the P/E ratio is increased, but then decreased,
According to the interactions of Al and Ga in the catalytic
which means a decrease in propylene and an increase in ethylene activity (Fig. 11), the reduction of Al decreased the catalytic
production. The Al cation is more electronegative than Ga. Thus activity in both the upper and lower limits of the Si/Ga ratio,
the resulting acid strength of the Ga cation is much weaker.74,75 and increased the difference of hexane conversion between
Therefore, with increasing Ga, it seems that the WAS increase the the upper and lower limits. At Si/Al ratios below 180, due to the
propylene production by methylation reactions. However, from a high catalytic activity at a low Si/Ga ratio, less Ga increased the
certain proportion onwards, where extra-framework sites are conversion, but more Ga showed more activity. In addition,
increased, far more propylene is formed by methylation of increasing the Si/Al ratio at both the upper and lower limits of
ethylene in the cavities. However, the production efficiency of the Si/Ga ratio reduced the total yield of light olefins and
light olefins decreases linearly with decreasing Ga, indicating the increased the distance between the yields of light olefins at
formation of weak and moderate acid sites with increasing Si/Ga the upper and lower limits of the Si/Ga ratio. At the lower limit
ratio. In fact, minor changes from a certain ratio onwards could of the Si/Ga ratio up the Si/Al = 300, decreasing the amount of
not affect the overall yield of light olefins, but only enhance partial Ga increased the yield of light olefins until Si/Al = 300, and then
methylation and dehydrogenation by creating WAS outside the decreased the yield of light olefins with a slower slope in
structure.
comparison with the upper limit. This phenomenon confirms
3.3.2. Interactions. According to Table 6, the amounts of the results obtained from the effect of Ga on the yield of light
Fe and Ga interact only in the catalytic activity. Fig. 11 shows olefins, because the increase of Ga, a cation with weaker
the effect of variations in Fe in the upper and lower limits of the electronegativity, creates more WAS, which are the main
Si/Ga ratio on the conversion. According to Fig. 11, the decrease centers for the formation of light olefins.
in the amount of Fe in both limits increased the activity.
3.3.3. Optimization of the Si/metal ratios. Due to the
Moreover, the conversion of hexane increased with increasing importance of the responses, the data were optimized using
the Si/Fe ratio in the upper and lower limits. In fact, as the Design Expert software. The optimization was performed in
amount of Fe decreased, the strength of SBAS increased, four modes: (1) optimization with the aim of maximum
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem., 2021, 45, 13833–13846 | 13843

View Article Online
Paper
NJC
Fig. 11 Valid interactions.
from n-hexane was completely investigated. Nano zeolites
were synthesized in fluorine medium using colloidal
silica and TPAOH to maximize the incorporation of metals in
the structure. Experimental design by the D-optimal method
was utilized to investigate the effect of three factors on four
responses. The predicted values of the D-optimal experimental
design showed acceptable agreement with the experimental
results along with appropriate R2, R2 adjusted and R2 predicted.
The study of the structural properties showed the
synergistic effect of Fe and Ga on the reduction of the
particle size by increasing the nucleation rate, which was
probably due to the variation in atomic radii. In addition,
the absence of extra-framework species of Ga and Fe was
confirmed by studying the acidic properties, and Fe, Al and
Ga increased the number of total acid sites, the strength of SAS
and the strength of WAS, respectively. The simultaneous
presence of Fe and Ga also improved the number of
WAS. The optimal conditions for obtaining the maximum
conversion, light olefins, and P/E ratio and minimum
alkanes as the responses were Si/Al = 110; Si/Fe = 377; Si/
Ga = 123.
Table 8 Optimization conditions
Si/ Si/ Si/
Response Quantity Al Fe Ga Conversion YAlkanes YOlefins
P/
E
Conversion Maximum 110 377 123 96.8
29.0
18.5
29.0
19.6
52.8 2.0
44.6 3.7
52.8 2.0
41.2 4.0
YAlkanes
YOlefins
P/E
Minimum 370 130 130 81.1
Maximum 110 377 123 96.8
Maximum 317 123 263 75.5
conversion, (2) optimization with the aim of the minimum yield
of alkanes, (3) optimization with the aim of the maximum
yield of light olefins and (4) optimization with the aim of the
maximum P/E ratio. As observed in Table 8, the optimal
conditions in the two cases of maximum conversion and
maximum yield of light olefins were equal. On the other hand,
the significance of the responses corresponding to the
maximum conversion and maximum yield of light olefins is
higher than that of the other two responses. Therefore,
considering the significance of light olefins and the subsequent
catalytic activity, the optimal catalyst conditions can be defined
as follows: Si/Al = 110; Si/Fe = 377; Si/Ga = 123.
4. Conclusions
Conflicts of interest
In this work, the synergistic effect of incorporating Fe and Ga
into the ZSM-5 structure in the production of light olefins There are no conflicts to declare.
13844
|
New J. Chem., 2021, 45, 13833–13846
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021

View Article Online
NJC
Paper
25 S. Wang, P. Wang, Z. Qin, Y. Chen, M. Dong, J. Li, K. Zhang,
P. Liu, J. Wang and W. Fan, ACS Catal., 2018, 8, 5485–5505.
26 Z. Wang, L. A. O’Dell, X. Zeng, C. Liu, S. Zhao, W. Zhang,
M. Gaborieau, Y. Jiang and J. Huang, Angew. Chem., Int. Ed.,
2019, 58, 18061–18068.
27 S. Bailleul, I. Yarulina, A. E. Hoffman, A. Dokania, E. Abou-
Hamad, A. D. Chowdhury, G. Pieters, J. Hajek, K. De Wis-
pelaere and M. Waroquier, J. Am. Chem. Soc., 2019, 141,
14823–14842.
References
1 V. Blay, B. Louis, R. Miravalles, T. Yokoi, K. A. Peccatiello,
M. Clough and B. Yilmaz, ACS Catal., 2017, 7, 6542–6566.
2 X. Dong, S. Shaikh, J. R. Vittenet, J. Wang, Z. Liu,
K. D. Bhatte, O. Ali, W. Xu, I. Osorio and Y. Saih, ACS
Sustainable Chem. Eng., 2018, 6, 15832–15840.
3 K. Wang, J. Zhang, S. Fan, X. Peng, N. Tsubaki and T. Zhao,
New J. Chem., 2021, 45, 4860–4866.
4 P. Cnudde, K. De Wispelaere, L. Vanduyfhuys,
R. Demuynck, J. Van der Mynsbrugge, M. Waroquier and
V. Van Speybroeck, ACS Catal., 2018, 8, 9579–9595.
5 X. Hou, Y. Qiu, E. Yuan, F. Li, Z. Li, S. Ji, Z. Yang, G. Liu and
X. Zhang, Appl. Catal., A, 2017, 543, 51–60.
6 M. Arsalanfar, New J. Chem., 2020, 44, 18457–18468.
7 U. Olsbye, S. Svelle, M. Bjørgen, P. Beato, T. V. Janssens,
F. Joensen, S. Bordiga and K. P. Lillerud, Angew. Chem., Int.
Ed., 2012, 51, 5810–5831.
28 A. Vjunov, J. L. Fulton, T. Huthwelker, S. Pin, D. Mei,
G. K. Schenter, N. Govind, D. M. Camaioni, J. Z. Hu and
J. A. Lercher, J. Am. Chem. Soc., 2014, 136, 8296–8306.
29 C. Li, A. Vidal-Moya, P. J. Miguel, J. Dedecek, M. Boronat
and A. Corma, ACS Catal., 2018, 8, 7688–7697.
30 T. Liang, J. Chen, Z. Qin, J. Li, P. Wang, S. Wang, G. Wang,
M. Dong, W. Fan and J. Wang, ACS Catal., 2016, 6,
7311–7325.
31 J. Li, H. Ma, Y. Chen, Z. Xu, C. Li and W. Ying, Chem.
Commun., 2018, 54, 6032–6035.
32 A. Janda and A. T. Bell, J. Am. Chem. Soc., 2013, 135,
19193–19207.
8 M. Rostami Sakha, S. Soltanali, D. Salari, M. Rashidzadeh
and P. Halimi Tabrizi, J. Solid State Chem., 2021,
301, 122342.
9 B. K. Yalcin and B. Ipek, Appl. Catal., A, 2021, 610, 117915.
10 T. Ma, L. Zhang, Y. Song, Y. Shang, Y. Zhai and Y. Gong,
Catal. Sci. Technol., 2018, 8, 1923–1935.
33 S. Mu¨ller, Y. Liu, F. M. Kirchberger, M. Tonigold,
M. Sanchez-Sanchez and J. A. Lercher, J. Am. Chem. Soc.,
2016, 138, 15994–16003.
´
´
11 S. Mitchell, M. Boltz, J. Liu and J. Perez-Ramırez, Catal. Sci.
34 J. Li, M. Liu, X. Guo, S. Zeng, S. Xu, Y. Wei, Z. Liu and
C. Song, Ind. Eng. Chem. Res., 2018, 57, 15375–15384.
35 Y. Ji, H. Yang and W. Yan, Mol. Catal., 2018, 448, 91–99.
36 Q. Shang, G. Xu, N. Tang, C. Wu, S. Chen and Y. Cong,
Microporous Mesoporous Mater., 2019, 288, 109616.
37 P. Losch, A. B. Pinar, M. G. Willinger, K. Soukup, S. Chavan,
B. Vincent, P. Pale and B. Louis, J. Catal., 2017, 345, 11–23.
38 Y. Ji, H. Yang and W. Yan, Catalysts, 2017, 7, 367.
39 Y. Nakasaka, J.-i. Nishimura, T. Tago and T. Masuda, Chem.
Eng. J., 2015, 278, 159–165.
40 S. Kim, G. Park, M. H. Woo, G. Kwak and S. K. Kim, ACS
Catal., 2019, 9, 2880–2892.
41 H. Konno, R. Ohnaka, J.-I. Nishimura, T. Tago, Y. Nakasaka
and T. Masuda, Catal. Sci. Technol., 2014, 4, 4265–4273.
42 G. Song, W. Chen, P. Dang, Y. Wang and F. Li, R. Soc. Open
Sci., 2018, 5, 181691.
Technol., 2017, 7, 64–74.
12 X. Hou, Y. Qiu, X. Zhang and G. Liu, Chem. Eng. J., 2017, 307,
372–381.
13 S. Xin, Q. Wang, J. Xu, Y. Chu, P. Wang, N. Feng, G. Qi,
´
J. Trebosc, O. Lafon and W. Fan, Chem. Sci., 2019, 10,
10159–10169.
14 B. C. Knott, C. T. Nimlos, D. J. Robichaud, M. R. Nimlos,
S. Kim and R. Gounder, ACS Catal., 2018, 8, 770–784.
15 C. Wang, L. Zhang, X. Huang, Y. Zhu, G. K. Li, Q. Gu,
J. Chen, L. Ma, X. Li and Q. He, Nat. Commun., 2019, 10, 1–8.
16 I. Yarulina, A. D. Chowdhury, F. Meirer, B. M. Weckhuysen
and J. Gascon, Nat. Catal., 2018, 1, 398–411.
17 T. Yokoi, H. Mochizuki, S. Namba, J. N. Kondo and
T. Tatsumi, J. Phys. Chem. C, 2015, 119, 15303–15315.
ˇ
ˇ
18 J. Dedecek, E. Tabor and S. Sklenak, ChemSusChem, 2019,
12, 556–576.
43 Y. Zhang, X. Han and S. Che, Chem. Commun., 2019, 55,
810–813.
44 S. Zhao, W. D. Wang, L. Wang, W. Schwieger, W. Wang and
J. Huang, ACS Catal., 2020, 10, 1185–1194.
19 S. Svelle, F. Joensen, J. Nerlov, U. Olsbye, K.-P. Lillerud,
S. Kolboe and M. Bjørgen, J. Am. Chem. Soc., 2006, 128,
14770–14771.
´
20 E. Dib, T. Mineva, P. Gaveau, E. Veron, V. Sarou-Kanian,
45 H. Li, Y. Wang, C. Fan, C. Sun, X. Wang, C. Wang, X. Zhang
and S. Wang, Appl. Catal., A, 2018, 551, 34–48.
46 N. Rahimi and R. Karimzadeh, Appl. Catal., A, 2011, 398,
1–17.
47 X. Su, Y. Fang, X. Bai and W. Wu, Ind. Eng. Chem. Res., 2019,
58, 20543–20552.
F. Fayon and B. Alonso, J. Phys. Chem. C, 2017, 121,
15831–15841.
21 K. Chen, M. Abdolrahmani, S. Horstmeier, T. N. Pham,
V. T. Nguyen, M. Zeets, B. Wang, S. Crossley and J. L. White,
ACS Catal., 2019, 9, 6124–6136.
22 A. Ghorbanpour, J. D. Rimer and L. C. Grabow, Catal.
Commun., 2014, 52, 98–102.
48 K. Iisa, Y. Kim, K. A. Orton, D. J. Robichaud, R. Katahira,
M. J. Watson, E. C. Wegener, M. R. Nimlos, J. A. Schaidle,
C. Mukarakate and S. Kim, Green Chem., Green Chem., 2020,
22, 2403–2418.
ˇ
ˇ
´
´
23 J. Dedecek, Z. Sobalık and B. Wichterlova, Catal. Rev.: Sci.
Eng., 2012, 54, 135–223.
´
24 E. Tabor, M. Bernauer, B. Wichterlova and J. Dedecek, Catal.
49 X. Li, A.-A. Alwakwak, F. Rezaei, A. A. Rownaghi and
A. C. S. Appl, Energy Mater., 2018, 1, 2740–2748.
Sci. Technol., 2019, 9, 4262–4275.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021
New J. Chem., 2021, 45, 13833–13846 | 13845

View Article Online
Paper
NJC
50 W. Dai, L. Yang, C. Wang, X. Wang, G. Wu, N. Guan, U. Obenaus, 63 T. Biligetu, Y. Wang, T. Nishitoba, R. Otomo, S. Park,
M. Hunger and L. Li, ACS Catal., 2018, 8, 1352–1362.
51 K. Van der Borght, V. V. Galvita and G. B. Marin, Appl.
Catal., A, 2015, 492, 117–126.
52 M. Hu, C. Wang, X. Gao, Y. Chu, G. Qi, Q. Wang, G. Xu, J. Xu
and F. Deng, ACS Catal., 2020, 10, 4299–4305.
H. Mochizuki, J. N. Kondo, T. Tatsumi and T. Yokoi,
J. Catal., 2017, 353, 1–10.
64 Y.-H. Yeh, R. J. Gorte, S. Rangarajan and M. Mavrikakis,
J. Phys. Chem. C, 2016, 120, 12132–12138.
65 S. Park, T. Biligetu, Y. Wang, T. Nishitoba, J. N. Kondo and
T. Yokoi, Catal. Today, 2018, 303, 64–70.
53 X. Jiang, X. Su, X. Bai, Y. Li, L. Yang, K. Zhang, Y. Zhang,
Y. Liu and W. Wu, Microporous Mesoporous Mater., 2018, 66 M. Bernauer, E. Tabor, V. Pashkova, D. Kauck´y,
´
´
263, 243–250.
Z. Sobalık, B. Wichterlova and J. Dedecek, J. Catal.,
54 K. Bian, A. Zhang, H. Yang, B. Fan, S. Xu, X. Guo and
C. Song, Ind. Eng. Chem. Res., 2020, 59, 22413–22421.
55 Z.-Y. Du, B.-B. Zhang, T.-S. Chen, Y. Betancur and W.-Y. Li,
Energy Fuels, 2019, 33, 10176–10184.
2016, 344, 157–172.
67 I. Yarulina, K. De Wispelaere, S. Bailleul, J. Goetze,
M. Radersma, E. Abou-Hamad, I. Vollmer, M. Goesten,
B. Mezari and E. J. Hensen, Nat. Chem., 2018, 10, 804–812.
56 Y. T. Cheng, J. Jae, J. Shi, W. Fan and G. W. Huber, Angew. 68 D. S. Lima and O. W. Perez-Lopez, Renewable Energy, 2019,
Chem., Int. Ed., 2012, 124, 1416–1419. 136, 828–836.
57 Z. Li, A. W. Lepore, M. F. Salazar, G. S. Foo, B. H. Davison, 69 M. Rostamizadeh and F. Yaripour, Fuel, 2016, 181, 537–546.
Z. Wu and C. K. Narula, Green Chem., 2017, 19, 4344–4352. 70 M. Xin, E. Xing, X. Gao, Y. Wang, Y. Ouyang, G. Xu, Y. Luo
58 Z. Hajjar, M. Kazemeini, A. Rashidi and S. Soltanali,
and X. Shu, Ind. Eng. Chem. Res., 2019, 58, 6970–6981.
71 P. Gao, Q. Wang, J. Xu, G. Qi, C. Wang, X. Zhou, X. Zhao,
N. Feng, X. Liu and F. Deng, ACS Catal., 2018, 8, 69–74.
72 Y.-H. Yeh and R. J. Gorte, Ind. Eng. Chem. Res., 2016, 55,
12795–12805.
73 X. Su, Y. Fang, P. Gao, Y. Liu, G. Hou, X. Bai and W. Wu,
Microporous Mesoporous Mater., 2020, 306, 110388.
74 Z. Han, F. Zhou, Y. Liu, K. Qiao, H. Ma, L. Yu and G. Wu,
J. Taiwan Inst. Chem. Eng., 2019, 103, 149–159.
J. Taiwan Inst. Chem. Eng., 2017, 78, 566–575.
¨
59 J. Sjoblom, K. Papadakis, D. Creaser and C. I. Odenbrand,
Catal. Today, 2005, 100, 243–248.
60 G. B. Ferraris, P. Forzatti, G. Emig and H. Hofmann, Chem.
Eng. Sci., 1984, 39, 81–85.
61 R. Feng, X. Yan, X. Hu, Y. Zhang, J. Wu and Z. Yan, Appl.
Catal., A, 2020, 594, 117464.
62 M. Thommes, K. Kaneko, A. V. Neimark, J. P. Olivier,
F. Rodriguez-Reinoso, J. Rouquerol and K. S. Sing, Pure 75 G. Krishnamurthy, A. Bhan and W. N. Delgass, J. Catal.,
Appl. Chem., 2015, 87, 1051–1069.
2010, 271, 370–385.
13846
|
New J. Chem., 2021, 45, 13833–13846
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2021