Rational design of bifunctional hierarchical Pd/SAPO-5 for the synthesis of tetrahydrofuran derivatives from furfural

Article abstract of DOI:10.1016/j.jcat.2021.03.003

The one-pot aldol condensation/crotonization reaction between furfural and methyl isobutyl ketone (MIBK), followed by hydrogenation with molecular H₂, was implemented for preparing tetrahydrofuran derivatives. To this aim, we developed a robust Pd/HPSAPO-5 catalyst based on the crystalline silico-aluminophosphate SAPO-5 with hierarchical porosity and optimized silicon content. The hierarchical HPSAPO-5 catalyst was synthesized using a bottom-up method starting from pre-synthesized MCM-41, with the surfactant (CTAB) inside the mesopores, serving both as Si source and mesoporogen. NH₃-TPD and FT-IR spectroscopy of adsorbed probe molecules combined with solid-state ¹H MAS NMR were used to assess the nature, strength and accessibility of the acid sites. The structural and textural properties of the catalysts were investigated using X-ray diffraction (XRD) and N₂ adsorption. HR-TEM was used to assess the dispersion and location of Pd nanoparticles on HPSAPO-5. The spent catalyst could be restored and reused after calcination.

Full text of DOI:10.1016/j.jcat.2021.03.003

Journal of Catalysis 397 (2021) 75–89
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Rational design of bifunctional hierarchical Pd/SAPO-5 for the synthesis
of tetrahydrofuran derivatives from furfural
L. Gao a,1, I. Miletto , C. Ivaldi , G. Paul , L. Marchese , S. Coluccia , F. Jiang , E. Gianotti
⇑
,
M. Pera-Titus
a
b,1
b
b
b
c
a
b,
a,
⇑
Eco-Efficient Products and Processes Laboratory (E2P2L), UMI 3464 CNRS-Solvay, 3966 Jin Du Road, Xin Zhuang Ind. Zone, 201108 Shanghai, China
Department of Science and Technological Innovation, Università del Piemonte Orientale, Viale T. Michel 11, 15121 Alessandria, Italy
Department of Chemistry, Università degli Studi di Torino, Via P. Giuria 7, 10125, Torino, Italy
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
The one-pot aldol condensation/crotonization reaction between furfural and methyl isobutyl ketone
MIBK), followed by hydrogenation with molecular H , was implemented for preparing tetrahydrofuran
derivatives. To this aim, we developed a robust Pd/HPSAPO-5 catalyst based on the crystalline silico-
aluminophosphate SAPO-5 with hierarchical porosity and optimized silicon content. The hierarchical
HPSAPO-5 catalyst was synthesized using a bottom-up method starting from pre-synthesized MCM-
Received 15 September 2020
Revised 31 January 2021
Accepted 2 March 2021
Available online 11 March 2021
(
2
4
1, with the surfactant (CTAB) inside the mesopores, serving both as Si source and mesoporogen.
3
-TPD and FT-IR spectroscopy of adsorbed probe molecules combined with solid-state H MAS NMR
Keywords:
Furfural
Methyl isobutyl ketone
Acid catalyst
SAPO-5
1
NH
were used to assess the nature, strength and accessibility of the acid sites. The structural and textural
properties of the catalysts were investigated using X-ray diffraction (XRD) and N adsorption. HR-TEM
was used to assess the dispersion and location of Pd nanoparticles on HPSAPO-5. The spent catalyst could
be restored and reused after calcination.
2
Cascade reaction
Ó 2021 Elsevier Inc. All rights reserved.
1
. Introduction
Lignocellulose is the most abundant and inexpensive source of
devise simple, easily accessible and robust catalytic systems
affording multi-step reactions to engineer intensified processes
avoiding the separation of intermediates. Sequential [21,22,24,
29] and one-pot [87,88] reactions have been developed targeting
the synthesis of branched alkanes as jet fuels and monomers for
polyester and epoxy resins starting from the aldol condensation
reaction of FF with ketones (e.g., acetone, methyl isobutyl ketone,
cyclopentanone), followed by hydrodeoxygenation (HDO) at
solvent-free conditions. Likewise, one-pot reactions have also been
reported for the synthesis of furan- and THF-derived alcohols and
amines by the one-pot (one-reactor tandem) aldol condensation
nonedible biomass that can be used for the synthesis of fuels,
chemicals and materials [1-5]. Among the different intermediates
that can be produced from carbohydrates, furanic compounds
hold
a
strategic place. In particular, furfural (FF) and
5
-hydroxymethylfurfural (HMF), which can be accessed by the
acid-catalyzed dehydration of pentoses and hexoses [6-8], respec-
tively, are considered by the U.S. Department of Energy (DoE) as
two of the most value-added biomass building blocks or platform
chemicals [9]. Both bio-based furanic derivatives can be further
oxidized, hydrolyzed or reduced to generate valuable downstream
chemicals, such as methyl- and dimethylfuran or tetrahydrofuran
reaction of FF with ketones, followed by hydrogenation with H
2
3 2
[20,30,89,90], or reductive amination with NH and H [31,32],
respectively.
[
10], levulinic acid [86],
c
-valerolactone [11], furandicarboxylic
Considerable research has been carried out in the last years to
engineer basic catalysts for conducting aldol condensation reac-
tions of FF and ketones, including metal and mixed oxides (e.g.,
acid [12], maleic acid [13], and hydrocarbons [14-19].
Functionalizing bio-based furanic derivatives in a minimum
number of synthetic steps is of huge interest to increase the molec-
ular diversity and complexity. In this view, it would be desirable to
ZrO
substituted zeotypes [40]. However, fewer examples have been
reported on acid catalysts, relying on acid oxides (e.g., ZrO and
ZnO) [41], acid zeolites [42-44] and metal–organic frameworks
45]. Zeolites are particularly attractive due to their tunable acid
properties, but can exhibit poor activity due to mass transport
2 2
, ZnO, MgO-ZrO ) [33-37], zeolites [38,39] and metal-
2
⇑
Corresponding authors.
E-mail addresses: enrica.gianotti@uniupo.it (E. Gianotti), marc.pera-titus-ex-
[
t@solvay.com (M. Pera-Titus).
1
These authors contributed equally to this work.
https://doi.org/10.1016/j.jcat.2021.03.003
0
021-9517/Ó 2021 Elsevier Inc. All rights reserved.

L. Gao, I. Miletto, C. Ivaldi et al.
Journal of Catalysis 397 (2021) 75–89
limitations within micropores, affecting the accessibility of the
reagents to the acid sites, as well as low hydrothermal stability
and deactivation due to coke deposition [46-50]. To overcome
these shortcomings, hierarchical acid zeolites have been developed
using different methodologies, some of which have recently shown
promising results in aldol condensation reactions [51,52]. In partic-
ular, hierarchical silicoaluminophosphates (SAPOs) acid zeo-type
catalysts represent a viable alternative. Indeed, microporous SAPOs
have been extensively used in acid catalyzed reactions for hydro-
carbon conversion [53,54]. In particular, SAPO-5, with an AFI
framework, a pore diameter of 7.3 Å and moderate acidity, has
been used as heterogeneous catalyst in methanol-to-olefin conver-
sion, toluene alkylation [55] and xylene isomerization [56].
Recently, hierarchical SAPO catalysts combining the micropores
of the zeolite framework and an additional mesoporous network,
have been synthesized using both top-down [57,58] and bottom-
up strategies [59-64].
solved in 38 mL of anhydrous ethanol. Then, 4.16 g (20 mmol) of
tetraethylorthosilicate (TEOS) and 0.054 g (0.26 mmol) of alu-
minium isopropoxide [(CH ) CHO] Al and deionized water were
3 2 3
added under vigorous stirring. After stirring for 24 h at room tem-
perature, the mixture was poured into a Petri dish to initiate the
EISA process. The as-obtained gels were calcined at 550 °C for
ꢁ1
6 h under static air with a heating rate of 1 °C min
.
2.3. Synthesis of microporous SAPO-5 with variable Si loading
Aluminium isopropoxide (7.0 g) was added slowly to deionized
water (10 mL) and triethylamine (TEA, 2.77 mL) under stirring, and
the mixture was further stirred for 1 h. Tetraethylorthosilicate
(TEOS, 2.31 mL, 4.62 mL and 6.92 mL for Si = 0.30, Si = 0.60 and
Si = 0.90, respectively) was added dropwise under stirring, along
with 7.29 mL of deionized water, and the mixture was further stir-
2
red for 2 h. Phosphoric acid (2.18 mL, 85 wt% in H O) was added
Herein, a novel bottom-up approach was used to synthesize hier-
archical SAPO-5 catalysts (HPSAPO-5). This method avoids post-
synthetic demetallation under acid or basic conditions as commonly
found in top-down methods, resulting in a localized collapse of the
zeolite framework. A series of HPSAPO-5 catalysts with variable Si
loadings were prepared using cetyl trimethylammonium bromide
dropwise under stirring. The gel was vigorously stirred for
30 min to produce a white gel with the following composition:
2
1 Al:0.93P:x Si:0.58 TEA:28 H O where x = 0.3; 0.6; 0.9.
The gel was transferred to a Teflon-lined stainless-steel auto-
clave and subjected to hydrothermal synthesis at 473 K for 60 h
under autogenous pressure. The solid product was then recovered
from the autoclave by filtration, washed with water, dried in air at
373 K and finally calcined in a tube furnace at 873 K for 16 h under
airflow to generate a white crystalline solid.
For the sake of clarity, the nomenclature SAPO-5/xSi is here-
inafter used, where ‘x’ refers to the molar Si content in the synthetic
gels.
(
CTAB) encapsulated within MCM-41, serving both as Si source
and mesoporogen [62]. This particular synthetic methodology miti-
gates the use of expensive and sophisticated surfactants to induce
mesoporosity in the microporous framework, and benefits from a
superior retention of the acid properties of microporous SAPOs,
whilst simultaneously enhancing mass transfer. The Si loading
allowsa modulationof the aciddensity and strength by a partialsub-
stitution of framework P with Si [65], resulting in the formation of
Brønsted acid sites of medium acid strength [66,67] through a
type-2 substitution mechanism (SM2) [68]. An increase of the Si
loadingcan also induce theformation of silicaislands in SAPO frame-
works, altering the acid strength. Besides, by impregnating Pd over
HPSAPO-5, a bifunctional catalyst can be designed, affording the for-
mation of furan- and tetrahydrofuran-derivatives in an efficient
one-step process starting from FF and ketones. These products could
be applied as potential biofuels or biosolvents.
2.4. Synthesis of hierarchical HPSAPO-5 with variable Si loading
Aluminium isopropoxide (7.0 g) was added slowly to deionized
water (10 mL) and TEA (2.77 mL) under stirring, and the mixture
was further stirred for 1 h. CTAB-containing MCM-41 (1.03 g,
2.06 g and 3.09 g for Si = 0.30, Si = 0.60 and Si = 0.90, respectively)
was slowly added, and the mixture was further stirred for 2 h.
2
Phosphoric acid (2.18 mL, 85 wt% in H O) was added dropwise
under stirring. The mixtures with three variable Si contents were
vigorously stirred for 30 min to produce white gels with the fol-
lowing composition
2
. Experimental
2.1. Materials
1
2
Al: 0.93P: x Si: y CTAB:0.58 TEA: 28 H O where x = 0.3 and
Aluminum isopropoxide (ꢀ98%), triethylamine (ꢀ99.5%), tetra-
y = 0.033; x = 0.6 and y = 0.066; x = 0.9 and y = 0.099.
ethyl orthosilicate (ꢀ99%) and phosphoric acid (85 wt% in H
2
O), all
used for the synthesis of SAPO-5, were supplied by Sigma-Aldrich
Milano, Italy). Palladium(II) acetate (98%), also supplied by
Sigma-Aldrich, was used for the synthesis of Pd/SAPO-5 catalysts.
NH
The gels were transferred to a Teflon-lined stainless-steel auto-
clave and crystallized at 473 K for 60 h under autogenous pressure.
The solid product was then recovered from the autoclave by filtra-
tion and washed with water. The as-prepared product was dried in
air at 373 K and calcined in a tube furnace under airflow at 873 K
for 16 h to remove organic surfactant and micropore template, pro-
ducing a white crystalline solid.
For the sake of clarity, the nomenclature HPSAPO-5/xSi is here-
inafter used, where ‘HP’ refers to ‘hierarchical porous, and ‘x’ refers
to the molar Si content in the synthetic gels.
(
3
(ꢀ99.7%), purchased from SIAD (Bergamo, Italy), and 2,6-di-
tert-butylpyridine (ꢀ97%), purchased from Sigma Aldrich (Milano,
Italy), were used as probe molecules in the FTIR experiments. Fur-
fural (99%) and methyl isobutyl ketone (ꢀ99%), both procured form
Sigma-Aldrich, were used for the catalytic tests. ZSM-5 (Si/Al = 23),
ZSM-5 (Si/Al = 80), ZSM-5 (Si/Al = 200), ZSM-5 (Si/Al = 280) and Y
(
Si/Al = 30) were purchased from Zeolyst, while BEA (Si/Al = 150)
was supplied by Clariant and MOR (Si/Al = 6.1) was kindly supplied
by ITQ. ZnO and ZrO were procured from Sigma-Aldrich.
2
2
.5. Synthesis of Pd/HPSAPO-5Si = 0.9
2.2. Synthesis of Al-SBA-15 (Si/Al = 76)
4.9 g of HPSAPO-5, 0.9Si were dispersed in a solution of 0.54 g of
palladium acetate in 30 mL of acetone. The final suspension was
sonicated for 30 min, and acetone was evaporated at 313 K under
vigorous stirring. The sample was then dried in an oven overnight
at 353 K, calcined at 673 K for 4 h and reduced at 473 K for 4 h
The Al-doped SBA-15 catalyst with a nominal SiO
ratio of 76 was prepared by the evaporation-induced self-assembly
EISA) method. The preparation method can be found elsewhere
69]. Briefly, 2.0 g of Pluronic P-123 triblock co-polymer was dis-
2 2 3
/Al O molar
(
[
2
under H .
76

L. Gao, I. Miletto, C. Ivaldi et al.
Journal of Catalysis 397 (2021) 75–89
2
.6. Catalyst characterization
A typical test was carried out as follows: 70 mg of catalyst were
pre-treated at 250 °C for 2 h under a He flow [40 mL(STP)/min]
to remove adsorbed moisture and vapors, and then cooled down
Before undertaking structural, volumetric and spectroscopic
analysis, calcined samples were outgassed at 573 K to remove
adsorbed water.
The phases present in the different catalysts were analyzed by
powder X-ray diffraction (PXRD). The PXRD patterns were recorded
3
to room temperature. Subsequently, NH was chemisorbed at
100 °C using pulsed injections until saturation. The temperature
was then increased up to 1000 °C using a heating rate of 10 °C/
min and the TPD profiles were recorded from 100 to 1000 °C.
The quantification method used for measuring the number of acid
sites can be found elsewhere [74,75]. All the profiles were normal-
ized to a constant mass of wafer.
on a PANalytical X’Pert PRO diffractometer provided with Cu-K
a
radiation (k = 1.54062 Å) generated at 45 kV and 30 mA. The spec-
tra were collected in the range 5° ꢂ 2h ꢂ 40° with a step size of
0
.02°. The patterns were indexed using the Joint Committee on
2
The H -TPR profiles were collected for Pd-containing catalysts
Powder Diffraction (JCPDS) database.
on a Micromeritics AutoChem II2920 system equipped with a
quartz U-type tubular reactor and a TCD detector. A cold trap
was used before the detector to avoid any interference of water
in the TPR plot. In each test, 50 mg of the given catalyst was loaded
into the reactor and purged with 30 mL(STP)/min He at 300 °C for
1 h to remove adsorbed moisture and vapors, and then cooled
down to room temperature. The temperature was then increased
The textural properties of the catalysts were inspected by N
2
and Ar physisorption at 77 K in the relative pressure range from
ꢁ
6
1
ꢃ 10 to 1 P/P
0
using a Quantachrome ASiQ apparatus. Prior
to the analysis, the samples were outgassed at 573 K for 24 h
ꢁ6
(
residual pressure lower than 10 Torr). Specific surface areas
were determined by the Brunauer-Emmett-Teller (BET) method
in the relative pressure range from 0.01 to 0.1 P/P . The adsorption
0
up to 750 °C at a heating rate of 10 °C/min under a H
in Ar (10 v/v%). The system was calibrated using an Ag
(>99% purity) to measure the H consumption in the TPR profiles.
2
flow diluted
branch was analyzed by the NLDFT (non-local density functional
theory) method to obtain the pore size distributions of the differ-
ent samples [70-72].
FTIR spectra of self-supporting pellets were collected under vac-
uum conditions (residual pressure < 10 mbar) using a Bruker
Equinox 55 spectrometer equipped with a pyroelectric detector
2
O standard
2
All the profiles were normalized to a constant mass of wafer.
Inductively Coupled Plasma Atomic Emission Spectroscopy
(ICP-AES) was used to quantify the bulk Pd content of the catalysts
using a Thermo IRIS Intrepid II XSP atomic emission spectrometer.
Before the measurements, the samples were dissolved using a
ꢁ
5
ꢁ
1
(
DTGS type) with a resolution of 4 cm ꢄNH
3
and 2,6-di-tert-
butylpyridine (2,6-dTBP) were adsorbed at room temperature
using specially designed cells permanently connected to a vacuum
line to perform adsorption–desorption in situ measurements. FTIR
spectra were normalized with respect the pellet weight and, when-
ever specified, were reported in difference-mode by subtracting
the spectrum of the sample in vacuum from the spectrum of the
adsorbed molecules. The total number of accessible Brønsted acid
sites (N) was estimated using the Lambert-Beer law in the form
3 2 4
HNO /H SO /HF solution.
2.7. Catalytic tests
The aldol condensation/crotonization reaction was conducted
in a stainless steel autoclave reactor (30 mL) from Taiatsu. In a typ-
ical test, the autoclave was charged with 0.2 g (2 mmol) of distilled
FF, 3.8 g (38 mmol) of MIBK, and 40 mg of catalyst (optimized con-
ditions) with a magnetic stirring bar. The mixture was heated to
the desired temperature under a stirring speed of 1000 rpm, and
the kinetic profiles were measured out for 1–24 h. Preliminary
experiments were conducted ensuring the absence of external
mass transfer limitations on the reaction rate.
A =
eNq, where A is the integrated area of the bands of the proto-
ꢁ
1
2
nated species (cm ),
e
is the molar extinction coefficient (cm .
mol ), N is the concentration of the vibrating species ( mol
is the density of the disk (mass/area ratio of the pellet,
ꢁ
1
l
g
l
ꢁ
1
), and q
ꢁ2
mg cm ) [73].
Solid-state (SS) NMR spectra were acquired on a Bruker Avance
III 500 spectrometer and a wide bore 11.7 Tesla magnet with oper-
The one-pot reaction was conducted by performing first the
aldol condensation/crotonization reaction between FF and MIBK
in an autoclave (30 mL) at the conditions described above, followed
ational frequencies for 1H, Si, P and Al of 500.13, 99.35,
29
31
27
2
02.45 and 130.33 MHz, respectively. A 4-mm triple resonance
probe in double resonance mode with magic angle spinning
MAS) was employed in all the experiments, and the samples were
2
by the introduction of H at the desired pressure (10–20 bar), and
the hydrogenation reaction was conducted at 80–160 °C for 6–12 h
(
ensuring complete reduction of the Pd phase according to the H
TPR profiles of the catalysts.
2
-
packed on a Zirconia rotor and spun at a MAS rate of 15 kHz. The
magnitudes of radio frequency fields were 100, 83 and 42 kHz
In each experiment, the FF conversion, selectivity, yield and the
initial reaction rate of formation (ri,0) of each product, as well as
the carbon balance (CB), were defined using the following expres-
sions (Eqs. (1)–(6))
1
31
29
27
for H, P and Si respectively. The Al MAS spectra were
acquired on large sweep width with small pulse angle ( /12) to
p
29
31
ensure quantitative interpretation. In the case of Si, P and
2
7
Al MAS NMR, high-power proton decoupling was applied. The
relaxation delay, d1, between accumulations was 5, 1, 20, and
n
FF
1
27
31
29
FF conversion ¼ 1 ꢁ
ꢃ 100
ð1Þ
ð2Þ
ð3Þ
ꢅ
6
0 s for H, Al, P and Si MAS NMR spectra, respectively. All
nFF
chemical shifts were reported using d scale, and were externally
1
29
3+
referenced to TMS for H and Si NMR, Al(H
2
31
O)
6
ion in 1.0 m AlCl
3
n
i
2
7
Selectivity ¼
ꢃ 100
solution for Al NMR and H PO (85%) for P NMR. The chemical
shifts for Al NMR were not corrected by second-order quadrupole
effects. The H MAS NMR spectra were fitted (DMFIT) for quantita-
3 4
i
ꢅ
nFF ꢁ nFF
2
7
1
n
i
Yield
i
¼
ꢅ
ꢃ 100
tive deconvolution of overlapping peaks. The samples were packed
nFF
ꢁ4
into a NMR rotor and dehydrated at 573 K under vacuum (1 ꢃ 10
ꢀ
mbar) for 2 h prior to loading into the magnet and recording the
NMR spectra.
1
dn
i
ꢀ
ꢀ
ꢀ
r
i;0
¼
ð4Þ
ð5Þ
W dt
3
NH -TPD was used to measure the density and strength of acid
P
sites in the catalysts. The measurements were performed on a
Micromeritics AutoChem II 2920 system equipped with a quartz
U-type tubular reactor and a thermal conductivity (TCD) detector.
n
i
þ n
FF
CBFF
¼
ꢅ
ꢃ 100
nFF
77

L. Gao, I. Miletto, C. Ivaldi et al.
Journal of Catalysis 397 (2021) 75–89
P
n
i
þ n þ nMIBK
especially for x = 0.6 and x = 0.9, with values of 49% and 40% (entries
12 and 14), respectively, while the FF conversion for SAPO-5/0.3Si is
only 23% (entry 10). In parallel, the ALD-1 selectivity is about 60% for
SAPO/0.6Si and SAPO-5/0.9Si, approaching the value obtained for
ZSM-5 (Si/Al = 80) and Y zeolites, whereas SAPO-5/0.3Si shows a
much lower ALD-1 selectivity of 47%. The selectivity of the aldoliza-
tion by-products ALD-2a and ALD 2b is similar for the different
SAPO-5/xSi samples with a value in the range 8–10%.
We further investigated the catalytic activity of the hierarchical
HPSAPO-5/xSi zeolites (x = 0.3, 0.6, 0.9). A neat increase of the FF
conversion is observed for HPSAPO-5/0.9Si (entry 15) and
HPSAPO-5/0.3Si (entry 11) compared to the parent SAPO-5/0.9Si
FF
CBFFþMIBK
¼
ꢅ
ꢅ
ꢃ 100
ð6Þ
nFF þ n
MIBK
i
where n is the number of moles of product i, and W is the catalyst
loading.
The initial reaction rate of formation of the different products
was measured from the kinetic plots using rational polynomial
functions (1st order numerator, 2nd order denominator) for fitting
the kinetic curves. The parameters in the rational polynomial func-
tions were fitted using a least-square non-linear optimization
method based on the Levenberg-Marquardt algorithm by compar-
ison of predicted and experimental yields as a function of time.
After the reaction, the mixture was centrifuged to separate the
solution from the catalyst. The concentration of the non-reacted FF
and MIBK, as well as the reaction products, were analyzed and
quantified by gas chromatography on a Thermo Scientific Trace
(
73% vs. 49%) and SAPO-5/0.3Si (45% vs. 23%), respectively, whereas
no apparent change is observed for HPSAPO-5/0.6Si (49%, entry
1
5
3). The ALD-1 selectivity for HPSAPO-5/0.9Si and HPSAPO-
/0.6Si is comparable to that obtained for the parent zeolites, but
1
100 GC equipped with a flame ionization detector (FID) and a
the selectivity is much higher for HPSAPO-5/0.3Si compared to
SAPO-5/0.3Si (72% vs. 47%). The selectivity of the aldolization by-
products ALD-2a and ALD-2b is similar for the different HPSAPO-
HP-5 capillary column with 5 wt% phenyl groups (length 30 m;
inner diameter 0.25 mm). The analytical methods were adjusted
for the different mixtures depending on the boiling point and
polarity of the compounds. In all the methods, the injector temper-
ature was set at 250 °C, the detector temperature was 300 °C and
the sample injection volume was 2 mL. The calibration was per-
formed using biphenyl as internal standard for the aldolization
products and dodecane for the hydrogenation products. Control
experiments in the absence of catalyst revealed that the reaction
did not proceed (<1% conversion).
Nuclear magnetic resonance spectroscopy (NMR) was used to
determine the structure of organic compounds. Each compound
can be characterized using several descriptors including the chem-
ical shift, spin multiplicity, coupling constants, and integration. In
this study, liquid H NMR analysis was used to identify the struc-
ture of the products generated in the different reactions. The
NMR spectra were recorded on a 400 MHz Bruker spectrometer.
5
/xSi samples with a value about 13%, which is slightly higher than
the value obtained on the parent SAPO-5/xSi samples. To assess
for the role of mesoporosity on the catalytic properties, we mea-
sured the catalytic activity for Al-SBA-15 (Si/Al = 76) (entry 16).
Unlike the HPSAPO/xSi catalysts, especially HPSAPO/0.9Si, the FF
conversion and ALD-1 selectivity are lower, with values in line to
those obtained for Y zeolite.
Overall, the results presented above point out an important
effect of the Si loading and the presence of a hierarchical architec-
ture on the catalytic properties of acid SAPO-5 catalysts in the aldol
condensation/crotonization reaction of FF with MIBK. The acid
properties and pore architecture of HPSAPO-5/xSi make these cat-
alysts attractive for the reaction unlike microporous zeolites and
other reference acid catalysts. These encouraging results prompted
us to study in detail the reaction kinetics of the HPSAPO-5/xSi at
different temperatures.
1
3
. Results and discussion
3.2. Kinetic profiles for FF + MIBK aldolization/crotonization over
3.1. Aldol condensation/crotonization of FF with MIBK over acid
HPSAPO-5/xSi
zeolites
The kinetic profiles were measured for the HPSAPO-5/xSi and
the parent SAPO-5/xSi catalysts in the temperature range 140–
180 °C. The initial FF:MIBK molar ratio was set to 1:18 (5.2 wt%
FF) while keeping the FF-to-catalyst weight ratio at a constant
value of 5.0 to promote the formation of ALD-1 and keep a high
carbon balance with respect to FF (see SI, Fig. S1). Fig. 1 plots the
curves obtained for the FF conversion and the ALD-1 and ALD-2a,
b selectivites, whereas Fig. 2 plots the corresponding selectivity-
conversion curves. Additional kinetic curves at lower temperature
(120 °C) can be found in the SI (Fig. S2).
Regardless of the temperature, HPSAPO-5/0.3Si exhibits a
monotonous increase of the FF conversion with the reaction time
until a plateau value lower than 100%, which increases with the
temperature from ca. 30% at 140 °C (Fig. 1C1) to 60% at 180 °C
(Fig. 1C3) after 24 h. An analogous trend is observed for the parent
SAPO-5/0.3Si, with an almost complete superposition of the kinetic
curves at longer reaction times for the different temperatures,
although a markedly lower conversion is observed for times below
12 h. The presence of a plateau in the FF conversion can be
explained by a potential catalyst deactivation during the reaction,
which appears to be unaffected by the porous texture of the cata-
lysts. As a matter of fact, the carbon balance with respect to FF
decreases along the reaction with a final down to 80% (see SI,
Fig. S3). However, the total carbon balance (including MIBK) keeps
at a value higher than 95%.
In a first step, we investigated the catalytic performance of a
library of reference acid catalysts in the aldol condensation/cro-
tonization reaction of FF and MIBK, including ZrO , ZnO and a ser-
ies of acid zeolites with variable topologies, i.e. ZSM-5 (Si/Al = 23–
80), BEA (Si/Al = 150), MOR (Si/Al = 6.1) and SAPO-5 with variable
2
2
Si loadings (SAPO-5/xSi, x = 0.3, 0.6, 0.9). In all cases, the main pro-
duct was (E)-1-(furan-2-yl)-5-methylhex-1-en-3-one (ALD-1)
(Scheme 1), whereas (E)- and (Z)-3-(furan-2-ylmethylene)-4-met
hylpentan-2-one (ALD-2a and ALD-2b, respectively) were obtained
as by-products. The alcohol intermediate issued from the aldol
condensation was only observed at trace levels for a FF conversion
higher than 10%. Besides, no products issued from the aldol self-c
ondensation/crotonization of MIBK were detected. Table 1 com-
piles the FF conversion, ALD-1 and ALD-2a / ALD-2b selectivities
and ALD-1 yield, as well as the carbon balance with respect to
FF, for the different catalysts.
2
ZnO and ZrO (entries 1 and 2) display only 18% and 25% FF con-
version, respectively, with an ALD-1 selectivity of 76% and 32%.
Regardless of the Si/Al molar ratio, the ZSM-5 zeolites (channel
size = 5.5 Å) exhibit poor FF conversion (range 10–20%) and an
ALD-1 selectivity up to 60% for ZSM-5 (Si/Al = 80) (entries 3–6). Sim-
ilar results are obtained for BEA (channel size = 5.5 Å) and MOR
(
channel size = 6.7 Å) (entries 7 and 8). In contrast, Y zeolite (cavity
size = 7.4 Å) displays a higher FF conversion (40%) and 52% selectivity
to ALD-1 (entry 9). The microporous SAPO-5/xSi zeolites with a com-
parable micropore size (7.3 Å), show similar FF conversion,
Unlike HPSAPO-5/0.3Si, HPSAPO-5/0.6Si and HPSAPO-5/0.9Si do
not display a plateau for the FF conversion at longer reaction times,
78

L. Gao, I. Miletto, C. Ivaldi et al.
Journal of Catalysis 397 (2021) 75–89
Scheme 1. Main products obtained in the aldol condensation/crotonization reaction of FF with MIBK over acid zeolites.
Table 1
Summary of catalytic properties for the different zeolites tested in this study.*
Entry
Catalyst
ZnO
Conv FF (%)
Sel ALD-1 (%)
Sel by-products (%)
Yield ALD-1 (%)
CBFF (%)
1
2
3
4
5
6
7
8
9
25
18
14
20
10
13
26
16
40
23
45
49
49
40
73
38
76
32
25
60
46
28
42
64
52
47
72
57
61
60
62
45
14
5
5
6
4
6
4
4
5
19
6
4
12
5
4
11
10
20
11
32
28
30
24
46
17
87
83
80
83
85
82
80
82
84
83
81
79
81
87
76
84
ZrO
2
ZSM-5 (Si/Al = 23)
ZSM-5 (Si/Al = 80)
ZSM-5 (Si/Al = 200)
ZSM-5 (Si/Al = 280)
BEA (Si/Al = 150)
MOR (Si/Al = 6.1)
Y (Si/Al = 30)
SAPO-5/0.3Si
HPSAPO-5/0.3Si
SAPO-5/0.6Si
1
1
1
1
1
1
1
0
1
2
3
4
5
6
9
13
10
13
8
14
7
HPSAPO-5/0.6Si
SAPO-5/0.9Si
HPSAPO-5/0.9Si
Al-SBA-15 (Si/Al = 76)
*
Reaction conditions: 0.2 g FF, 3.8 MIBK, 40 mg catalyst, 180 °C, 3 h.
even if the reaction proceeds slowly in the former case after 12 h,
and reach higher FF conversions. The FF conversion evolves from
catalysts, the presence of mesopores enriched with Si-OH groups
on the external surface is expected to favor mass transfer between
SAPO-5 particles and promote MIBK adsorption (vide infra),
enhancing the reaction at lower FF conversion. Besides, the pres-
ence of a plateau for the ALD-1 selectivity (<100%) at high FF con-
version can be attributed to catalyst deactivation, blocking the
accessibility to the acid centers in the micropores.
56% at 140 °C (Fig. 1B1) to 65% at 180 °C (Fig. 1B3) for HPSAPO-
5/0.6Si, and from 60% at 140 °C (Fig. 1A1) to full conversion at
180 °C (Fig. 1A3) for HPSAPO-5/0.9Si, after 24 h. These results sug-
gest that, whereas both samples show partial deactivation during
the reaction, this is much less prominent on HPSAPO-5/0.9Si.
A similar conclusion can be drawn from the kinetic plots on the
parent SAPO-5/0.6Si and SAPO-5/0.9Si, but with two divergent
behaviors compared to the hierarchical counterparts: SAPO-
3.3. Reaction rates, reaction order and activation energies for ALD-1
synthesis
5
5
/0.6Si exhibits higher FF conversions compared to HPSAPO-
/0.6Si, especially at 160 °C, with a FF conversion reaching 75%
The initial reaction rates of ALD-1 formation were measured
after 24 h, whereas SAPO-5/0.9Si exhibits much lower FF conver-
sions compared to HPSAPO-5/0.9Si, even at the highest tempera-
ture (180 °C) and long reaction times. This different behavior
points out a positive effect of the hierarchical architecture on the
catalytic activity for HPSAPO-5/0.9Si, whereas the effect is low
and even detrimental for HPSAPO-5/0.6Si.
Turning now our attention into the selectivity, irrespective of
the temperature, the microporous SAPO-5/xSi catalysts exhibit a
sharp increasing trend for the ALD-1 selectivity until a plateau
value at about 60%, especially for SAPO-5/0.3Si. This becomes
clearly visible in the selectivity-conversion plots, especially at
lower temperature (140 °C) (Fig. 2A2–C2). In contrast, the hierar-
chical HPSAPO-5/xSi catalysts show a less prominent increase of
the ALD-1 selectivity with the FF (Fig. 2A1–C1), being almost con-
stant for HPSAPO-5/0.9Si. In all cases, the aldol by-products ALD-2a
and ALD-2b show similar selectivities for the hierarchical and par-
ent SAPO-5 catalysts with a selectivity up to 15% at 180 °C after
from the time-evolution of the ALD-1 yield (see SI, Figs. S4, S5)
using Eq (4) and rational polynomial functions for fitting the
kinetic curves. The parity plots are collected in Fig. S6. Fig. 3 plots
the rates on HPSAPO-5/xSi and SAPO-5/xSi as a function of the
temperature and the Si loading (x). Among the SAPO-5/xSi cata-
lysts (Fig. 3B), the initial rate of ALD-1 formation is sensibly higher
ꢁ
1
on SAPO-5/0.6Si, especially at 140 °C, evolving from 2.6 mmol.g
h
.
ꢁ1
ꢁ1 ꢁ1
at 140 °C to 12 mmol.g .h at 180 °C. In contrast, among the
HPSAPO-5/xSi catalysts (Fig. 3A), the highest rates are observed on
HPSAPO-5/0.9Si for all temperatures, reaching a value as high as
ꢁ
1
ꢁ1
1
10 mmol.g .h at 180 °C. When comparing the initial reaction
rates of ALD-1 formation on HPSAPO-5/xSi and SAPO-5/xSi cata-
lysts (Table 2), the highest enhancement is observed for HPSAPO-
5
/0.9Si with ratios of 30, 16 and 17 at 140 °C, 160 °C and 180 °C,
respectively. This observation points out a remarkable effect of
the hierarchical architecture on the reaction rate of ALD-1 forma-
tion. In contrast, these ratios are much lower for HPSAPO-5/0.3Si
2
4 h.
(
range 5.6–9.8), and especially for HPSAPO-5/0.6Si, showing ratios
The increasing trend of the ALD-1 selectivity for SAPO-5/xSi
lower than 2 and accordingly an almost negligible effect of the
hierarchical architecture on the catalytic properties.
With these results in hand, we measured the activation energies
of ALD-1 formation by representing the Arrhenius plots for the
HPSAPO-5/xSi and SAPO-5/xSi catalysts (Fig. 4). Within the limits
of the experimental error, all the catalysts except SAPO-5/0.3Si,
suggests a hindering effect of internal mass transfer on the reaction
rate due to a poor accessibility of FF and MIBK to the acid centers in
the zeolite micropores. This accessibility is promoted at higher
temperature, enhancing the diffusivity of the reactants and ALD-
1
within the micropores. In the case of hierarchical HPSAPO-5/xSi
79

L. Gao, I. Miletto, C. Ivaldi et al.
Journal of Catalysis 397 (2021) 75–89
Fig. 1. Time-evolution of the FF conversion, ALD-1 and ALD-2a,b selectivity in the aldol condensation/crotonization reaction of FF with MIBK over HPSAPO-5/xSi (filled
symbols, straight lines) and SAPO-5/xSi (empty symbols, dotted lines) at different temperature: (A) x = 0.9, (B) x = 0.6 and (C) x = 0.3. Reaction conditions: 0.2 g FF, 3.8 g MIBK,
4
0 mg catalyst, 3 h, (A1-C1) 140 °C, (A2-C2), 160 °C and (A3-C3) 180 °C.
irrespective of their porous architecture, exhibit comparable
apparent activation energies at about 100 kJ/mol. This observation
points out an identical catalytic mechanism for the different
samples with the absence of internal mass transfer effects on the
measured reaction rates. In this view, the higher activity observed
on HPSAPO-5/0.9Si should be attributed, at first sight, to a higher
density of accessible acid sites with the right strength (see
chapter 3.6).
run, the catalyst was separated, calcined at 400 °C under airflow
and reused again. The catalytic activity is recovered, with a FF con-
version of 58% and ALD-1 and AlD-2a,b selectivities of 58% and 8%,
respectively.
3.5. Single-reactor tandem aldol condensation/crotonization of FF with
MIBK, followed by hydrogenation
Given the high catalytic activity obtained over HPSAPO-5/0.9Si,
we further investigated the single-reactor tandem aldol condensa-
tion/crotonization reaction of FF with MIBK, followed by hydro-
genation (Scheme 2). To this aim, we impregnated HSAPO-5/0.9Si
with Pd at three different loadings (0.3 wt%, 0.8 wt%, 4.8 wt%). First,
we conducted the aldol condensation/crotonization reaction over
0.3%Pd/HSAPO-5/0.9Si and 4.8%Pd/HSAPO-5/0.9Si in the tempera-
ture range 140–180 °C. Fig. 5 plots the kinetic profiles obtained
for the FF conversion and selectivity of ALD-1 and ALD-2a,b by-
products. For comparison, the kinetic profiles for the parent
HPSAPO-5/0.9Si are also represented. Unlike HPSAPO-5/0.9Si,
0.3%Pd/HSAPO-5/0.9Si and 4.8%Pd/HSAPO-5/0.9Si exhibit a sus-
tained increase of the FF conversion until 100% after 12 h. The
ALD-1 selectivity also exhibits a plateau, but at a higher value than
in the case of HPSAPO-5/0.9Si (75% vs. 62%). The initial reaction
rate of ALD-1 formation is lower in the case of 0.3%Pd/HSAPO-
3
.4. Catalyst recyclability and reuse
The catalyst recyclability and reuse was assessed for HPSAPO-
/0.9Si in four consecutive catalytic runs for the aldol condensa-
5
tion/crotonization reaction of FF with MIBK at 160 °C for 3 h (see
SI, Fig. S7). The first run results in a FF conversion of 65% and
ALD-1 and ALD-2a,b selectivities of 61% and 13%, respectively.
After the reaction, the catalyst was separated, washed with ethanol
and reused for a further catalytic test. The FF conversion decreases
to 36%, whereas the ALD-1 selectivity increases to 72% at the
expense of the ALD-2a,b selectivity with a value of 8%. A slight
decline of the FF conversion together with an increase of the
ALD-1 selectivity is observed between the 2nd-4th runs. These
results reinforce the idea pointed out above of partial catalyst
deactivation during the reaction, most likely by the formation of
oligomers or coke in the micropores. In this view, after the fourth
ꢁ1
ꢁ1
5/0.9Si and 4.8%Pd/HSAPO-5/0.9Si (9 mmol.g .h
at 160 °C),
80

L. Gao, I. Miletto, C. Ivaldi et al.
Journal of Catalysis 397 (2021) 75–89
Fig. 2. Selectivity-conversion plots in the aldol condensation/crotonization reaction of FF with MIBK over (A1-C1) HPSAPO-5/xSi and (A2-C2) SAPO-5/xSi: (A) x = 0.9, (B)
x = 0.6 and (C) x = 0.3. Reaction conditions as in Fig. 1.
whereas the rate measured on HSAPO-5/0.9Si is 60 mmol.g 1.hꢁ1
ꢁ
3.6. Understanding the structure and acid properties of HPSAPO-5/xSi
catalysts
(
see fittings in SI, Figs. S8, S9).
With these results in hand, we performed the single-reactor
tandem aldol condensation/crotonization reaction of FF with MIBK,
followed by hydrogenation, over the Pd-loaded HPSAPO-5/0.9Si
catalysts. Two reaction products were obtained, that is the furan-
derivative issued from the hydrogenation of the conjugated C@C
bond (F-ALD-1), and the THF-derivative issued also from the
hydrogenation of the furan ring (THF-ALD-1). However, the relative
selectivity to both products, as well as the total yield, depends on
the hydrogenation conditions (Table 3). Noteworthy, neither furan-
nor THF-derivatives issued from the hydrogenation of the C@O
bond are detected. The best result is obtained over 4.8%Pd/
HSAPO-5/0.9Si, affording the synthesis of THF-ALD-1 with 45%
yield with respect to FF. F-ALD-1 is also obtained with 5% yield.
Additional hydrogenated products, including products issued from
ALD-2a,b, are obtained with an overall yield of 14%.
The catalytic results presented above point out an enhanced
activity and selectivity of HPSAPO-5/0.9Si in the aldol condensa-
tion/crotonization reaction of FF with MIBK, as well as the high
efficiency of 4.8%Pd/HPSAPO-5/0.9Si for the synthesis of THF-
ALD-1. In light of these results, we explored the framework struc-
ture, porous architecture, acidity and acid strength of this catalyst
compared to HPSAPO-5/0.6Si and HPSAPO-5/0.3Si, as well as their
microporous counterparts. We also characterized the nature of the
Pd phase over HPSAPO-5/0.9Si.
3.6.1. Structural and textural properties of HPSAPO-5/xSi catalysts
The XRD patterns of HPSAPO-5/xSi and the microporous SAPO-5/
xSi counterparts reveal the typical reflections belonging to the AFI
framework (Fig. S10). This suggests a high phase purity and
81

L. Gao, I. Miletto, C. Ivaldi et al.
Journal of Catalysis 397 (2021) 75–89
Fig. 4. Arrhenius plots for the reaction rate of ALD-1 formation in the aldol
condensation/ crotonization reaction of FF with MIBK over HPSAPO-5/xSi (filled
symbols) and SAPO-5/xSi (empty symbol), with x = 0.3, 0.6, 0.9. Reaction conditions
as in Fig. 1.
Fig. 3. Evolution of the initial reaction rate of ALD-1 formation as a function of the
temperature and the Si loading in the aldol condensation/crotonization reaction of
FF with MIBK over (A) HPSAPO-5/xSi and (B) SAPO-5/xSi, with x = 0.3, 0.6, 0.9.
Reaction conditions as in Fig. 1. The curves are a guide to the eye.
3
.6.2. Chemical environment of HPSAPO-5/xSi catalysts
The chemical environment of the framework atoms in the
HPSAPO-5/xSi and the microporous SAPO-5/xSi catalysts was
probed by Al, P, and Si MAS NMR (Figs. S12–S14). In all cases,
a unique resonance band at ca. 37 ppm is observed in the Al NMR
spectra (Fig. S12), which can be unambiguously ascribed to
tetrahedrally-coordinated Al atoms in Al(OP)
suggests in turn the absence of Lewis acid sites in all the catalysts.
2
7
31
29
2
7
crystallinity of all the samples regardless of their Si loading and por-
ous architecture, both for the fresh and spent catalyst after reaction.
The presence of the hierarchical architectures, with micro- and
4
. This observation
mesoporosity, in the HPSAPO-5/xSi catalysts was evidenced by N
2
3
1
A single band at about ꢁ30 ppm is also observed in the P NMR
adsorption/desorption analysis at 77 K (Fig. S11A). All the hierar-
chical catalysts exhibit Type IV isotherms with a hysteresis loop
in the P/P range 0.7–1.0, which is indicative of the presence of
0
mesopores with an average size of ca. 30 Å as inferred using the
NLDFT method (Fig. S11B). Besides, a peak centered at ca. 10 Å
can be visualized, which is typical of tensile strength effect due
spectra (Fig. S13), which can be assigned to tetrahedrally-
2
7
31
coordinated P atoms. Since both the Al and P NMR spectra exhi-
bit a single resonance, it is possible to confirm a strict alternation
of Al and P at the T-positions of the aluminophosphate framework
[
77]. In the case of the samples with the highest Si loading (i.e.
3
1
x = 0.9), the P NMR spectra exhibit a broad foot beneath the main
to N
was obtained from the Ar adsorption/desorption isotherm at
7 K on HP SAPO-5/0.6Si to highlight the microporous range typi-
2
adsorption [76]. The green distribution in panel Fig. S11B
band at ca. ꢁ24 ppm that can be assigned to P-OH units. Finally,
2
9
the Si MAS NMR spectra reveal multiple signals in the range
7
ꢁ
90 to ꢁ110 ppm due to tetrahedrally-coordinated framework Si
cal of SAPO-5, revealing a micropore size at around 7.3 Å as
expected for the AFI structure.
atoms bonded to four, three, two, one and zero Al atoms (Fig. S14).
Table 4 lists the BET surface area, as well as the total, microp-
orous and mesoporous specific surface areas and pore volumes
measured by the NLDFT method. Both the BET and NLDFT surface
areas increase with the Si loading. The mesoporous surface area
3
3
.6.3. Acidity and acid strength of HPSAPO-5/xSi catalysts
.6.3.1. H MAS NMR. H MAS NMR is an ideal tool to investigate
1
1
the nature and strength of the acid sites present in SAPO-5/xSi
and HPSAPO-5/xSi catalysts with different Si loadings and comple-
ments the studies performed with FTIR spectroscopy of adsorbed
(Smeso) and mesopore volume (Vmeso), together with total pore vol-
ume (Vtot), also increase with the Si loadings, while the contribu-
tion of the microporous surface area (Smicro) and the micropore
volume (Vmicro) are very similar in all samples. Volumetric data
strongly supports the successful preparation of hierarchical archi-
tectures, and indicates the coexistence of multiple levels of poros-
ity within HPSAPO-5.
1
probe molecules (see below). H MAS NMR spectra render direct
information on the nature and density of protonic species present
in the samples (Fig. 6, Table 5). In all cases, two well separated
bands centered at around 3.5 and 4.6 ppm are visible in the spec-
tra, which can be ascribed to two different Al(OH)Si Brønsted acid
Table 2
Ratio between the initial reaction rate of ALD-1 formation on HPSAPO-5/xSi and on SAPO-5/xSi as a function of the temperature and the Si loading (x).
o
x
T ( C)
1
20
140
160
180
0
0
0
.30
.60
.90
2.4
2.8
13
9.8
1.8
30
5.6
1.1
16
6.4
1.9
17
82

L. Gao, I. Miletto, C. Ivaldi et al.
Journal of Catalysis 397 (2021) 75–89
Scheme 2. Main products obtained in the single-reactor tandem aldol condensation/crotonization reaction of FF with MIBK, followed by hydrogenation.
sites (BAS1 and BAS2) located in the 12- and 6-membered ring
channels inside the AFI framework, respectively [77]. The intensity
of both bands decreases with the Si loading for HPSAPO-5/xSi cat-
alysts, while no definite trend is observed for their microporous
counterparts. Whereas the intensity of both bands is similar for
HPSAPO-5/0.6Si and SAPO-5/0.6Si, and for HPSAPO-5/0.3Si and
SAPO-5/0.3Si, HPSAPO-5/0.9Si exhibits much lower intensity for
both species compared to SAPO-5/0.9Si (21 vs. 12 for the species
at 3.5 ppm, and 14 vs. 5 for the species at 4.6 ppm).
intensity of the former band exhibits a neat increase with the Si
loading for HPSAPO-5/xSi catalysts from 4 (x = 0.3) to 19
(x = 0.9), whereas the increase is more moderate for SAPO-5/xSi
catalysts (from 11 for x = 0.3 to 15 for x = 0.9). When comparing
the SAPO-5/xSi and HPSAPO-5/xSi catalysts at the same Si loading,
the intensity is similar for x = 0.6 and x = 0.9, while a much lower
intensity is observed for HPSAPO-5/0.3Si compared to SAPO-
5/0.3Si (11 vs. 4). In contrast, the latter bands are systematically
enhanced for HPSAPO-5/xSi catalysts compared to the SAPO-5/
xSi counterparts (20 vs. 15 for x = 0.3, 27 vs. 19 for x = 0.6 and
33 vs. 12 for x = 0.9). Besides, for HPSAPO-5/xSi catalysts, the inten-
sity increases monotonously with the Si loading from 20 for x = 0.3
to 33 for x = 0.9. Overall, all these results point out that, by increas-
ing the Si loading, more defects such as Si-OH, Al-OH and/or P-OH
are produced in HPSAPO-5/xSi catalysts at the expense of Brønsted
acid sites.
Apart from Brønsted acid sites, a signal at ca. 1.7 ppm is visible
1
in all H MAS NMR spectra, which can be attributed to isolated Si-
OH groups, together with a band centered at around 2.6 ppm due
to Al-OH and/or P-OH defects with very weak acidity [78-80]. The
3
3.6.3.2. In situ vibrational FTIR spectroscopy using NH and 2,6-dTBP
as basic probes. The FTIR signature of O-H species in SAPO-5/xSi
catalysts was reported in a previous study [54]. Briefly, the FTIR
ꢁ1
spectra are constituted by two main bands, one at 3680 cm that
can be assigned to the O-H stretching mode of free P-OH defects,
ꢁ1
and a pronounced band at 3745 cm that can be attributed to
Si-OH groups, pointing out the presence of defects in the AFI
framework upon incorporation of Si.
The FTIR spectra in the OH stretching region of HPSAPO-5/xSi
ꢁ1
display two distinctive bands centered at 3630 and 3513 cm
Fig. S15A), which can be associated to Al(OH)Si Brønsted acid sites
located in the 12-ring and 6-ring channels of the AFI framework
(
[
(
54,78-80]. Additional signals arising from isolated SiOH
3745 cm ) and isolated POH (3678 cm ) sites are also visible,
ꢁ1
ꢁ1
pointing out the presence of defects in the AFI framework upon
incorporation of Si. In the FTIR spectra of microporous SAPO-5
(
Fig. S15B), the bands due to the stretching of BAS are not visible
due to the high scattering profile of the microporous SAPO-5 and
ꢁ1
only the contribution of SiOH (3745 cm ) can be appreciated [54].
The nature and accessibility of Brønsted acid sites in the
HPSAPO-5/xSi catalysts can be elucidated using basic probe mole-
cules with variable steric hindrance, such as NH
butyl-pyridine (2,6-dTBP), adsorbed at room temperature. The
total amount of Brønsted acid sites was measured using NH
3
and 2,6-di-tert-
3
,
which, due to its small kinetic diameter (0.26 nm), is usually con-
sidered as a suitable probe molecule for titrating acid hydroxyl
groups accessible through pores, channels or windows in zeolites
with a size < 4 Å [81]. In contrast, 2,6-di-tert-butyl-pyridine (2,6-
dTBP), with a kinetic diameter of 1.05 nm, is not expected to enter
the micropores, and can therefore only probe the more accessible
acid sites sitting on the mesopores or in the pore mouth of micro-
pores [82].
Fig. 5. Time-evolution of (A) FF conversion, and (B) ALD-1 and ALD-2a,b selectivity
at 160 °C in the aldol condensation/crotonization reaction of FF with MIBK over
HPSAPO-5/0.9Si, 0.3%Pd/HPSAPO-5/0.9Si and 4.8%Pd/HPSAPO-5/0.9Si. Reaction
conditions as in Fig. 1. The curves are a guide to the eye.
3
Fig. 7 plots the differential FTIR spectra of NH and 2,6-dTBP
adsorbed on HPSAPO-5/xSi catalysts in the low frequency zone,
where the signals of protonated species are observed. For
83

L. Gao, I. Miletto, C. Ivaldi et al.
Journal of Catalysis 397 (2021) 75–89
Table 3
Main products obtained in the single-reactor tandem aldol condensation/crotonization reaction of FF with MIBK, followed by hydrogenation, over Pd-loaded HPSAPO-5/xSi.
Reaction conditions
1
st step *
2nd step
Pd (wt%)
0.3%
Yield F-ALD-1
Yield THF-ALD-1
Yield Other products
1
60 °C, 12 h
160 °C, 12 h, 20 bar H
2
2
22%
9%
11%
12%
5%
11%
26%
22%
40%
45%
42%
8%
18%
13%
15%
11%
14%
14%
20%
0
4
.8%
.8%
1
80 °C, 3 h
160 °C, 12 h, 20 bar H
0.3%
.8%
4
1
1
80 °C, 3 h
80 °C, 3 h
120 °C, 3 h, 20 bar H
80 °C, 3 h, 20 bar H
2
0.3%
4.8%
7%
12%
2
*
Other reaction conditions: 0.2 g FF, 3.8 MIBK, 40 mg catalyst
Table 4
Textural properties of HPSAPO-5/xSi catalysts with variable Si loading.
2
ꢁ1
3
ꢁ1
x
Specific surface areas (m .g
)
Pore volumes (cm .g
)
y
y
SSABET
SSANLDFT
S
micro
S
meso
*
V
tot NLDFT
V
micro
V
meso
0
0
0
.30
.60
.90
267
340
464
375
450
552
315
327
333
60
123
219
0.304
0.323
0.448
0.079
0.077
0.075
0.225
0.246
0.373
*
Smeso = SDFT – Smicro.
Fig. 6. 1H MAS NMR spectra of (A) HPSAPO-5/xSi and (B) microporous SAPO-5/xSi with x = 0.30 (black curvess), x = 0.6 (red curves) and x = 0.9 (blue curves). (For
interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Table 5
Intensity of protonic species measured from single-pulse 1H MAS NMR spectroscopy in HPSAPO-5/xSi and microporous SAPO-5/xSi, with x = 0.3, 0.6, 0.9.
1H Chemical shift d (ppm)
Assignment
HPSAPO-5/xSi
x = 0.3
Micro SAPO-5/xSi
x = 0.3
x = 0.6
x = 0.9
x = 0.6
x = 0.9
1
2
.7–2.1
.2–2.4
SiOHiso
AlOH, POH
H-bonded
BAS1
4
20
16
27
19
33
11
15
16
19
15
12
3
4
.5–3.9
.6–4.9
22
19
16
11
12
5
24
13
17
9
21
14
BAS2
84

L. Gao, I. Miletto, C. Ivaldi et al.
Journal of Catalysis 397 (2021) 75–89
Fig. 7. Differential FTIR spectra of (A) NH
3
and (B) 2,6-dTBP adsorption on calcined HPSAPO-5/xSi with x = 0.3 (black curves), x = 0.6 (red curves) and x = 0.9 (blue curves) after
outgassing the probe molecules at room temperature. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this
article.)
comparison, NH
compare the amount of Brønsted acid sites in microporous and
hierarchical catalysts (Table 6, Fig. S16). NH adsorption induces
proton transfer from the Brønsted acid sites to NH , resulting in
the formation of NH+
species that can be visualized in the band
3
was also probed on microporous SAPO-5/xSi to
can be clearly observed. Table 6 lists the density of accessible
Brønsted acid sites (N) in HPSAPO-5/xSi catalysts titrated using
2,6-dTBP compared to the density of accessible sites titrated by
3
3
3
NH . These data evidence that only a small fraction of Brønsted
4
acid sites are accessible to the bulkier probe molecules. The acces-
sibility factor (AF) [83], defined as the density of sites titrated by
2,6-dTBP divided by the density of Brønsted acid sites titrated by
ꢁ1
appearing at 1450 cm (bending mode) (Fig. 7A). The density of
accessible Brønsted acid sites (N) titrated by NH is similar for
mol.g ) and SAPO-5/0.6Si
mol.g ), whereas it shows a higher value for SAPO-
3
ꢁ1
microporous SAPO-5/0.3Si (46.80
l
3
NH , reveals a higher value for HPSAPO-5/0.3Si (AF = 0.049), while
ꢁ1
(
44.08
l
HPSAPO-5/0.6Si and HPSAPO-5/0.9Si show
(AF = 0.025).
a similar value
ꢁ1
5
/0.9Si (52.30
lmol.g ) (Table S1). This observation points out a
higher degree of Si incorporation in the SAPO-5 framework at the
highest Si loading (x = 0.9). In the case of hierarchical HPSAPO-5/
xSi catalysts, HPSAPO-5/0.6Si and HPSAPO-5/0.9Si contain a higher
By assuming that the Brønsted acid sites titrated by 2,6-dTBP
are preferentially located on the mesoporous surface, we can esti-
mate the density of Brønsted acid sites located in the micropores
by subtracting the density of acid sites titrated by 2,6-dTBP from
ꢁ1
1
density of Brønsted acid sites with 46.25 and 45.32
respectively, compared to HPSAPO-5/0.3Si, with 42.79
l
l
mol.g
mol.g
,
.
ꢁ
3
that titrated by NH . Interestingly, regardless of the Si loading,
However, HPSAPO-5/0.9Si displays a lower density of Brønsted
the density of Brønsted acid sites is similar in the different
acid sites than SAPO-5/0.9Si, as also evidenced by 1H MAS NMR
HPSAPO-5/xSi catalysts, with a value of 0.13
l
mol.m or 0.08
ꢁ2
ꢁ
2
(
Table 5).
Proton transfer with more accessible Brønsted acid sites in hier-
archical catalysts also occurs when 2,6-dTBP is adsorbed and a
sites.nm of microporous surface area. In line with the results
1
from H MAS NMR, this observation points out that the density
of Brønsted acid sites is not affected by the Si loading in the
HPSAPO-5/xSi catalysts, and that a higher Si loading contributes
ꢁ1
(m
8a), due to protonated 2,6-dTBPH+ species,
band at 1618 cm
Table 6
Integrated area (A) of the dasym NH+
(1450 cm ) and
3
and 2,6-dTBP probe molecules for HPSAPO-5/xSi catalysts with x = 0.3, 0.6 and 0.9.
ꢁ1
m
8a 2,6-dTBPH+ (1618 cm ) bands, density of accessible Brønsted sites (N), accessibility factor (AF) and density of acid sites
ꢁ1
4
in micropores measured using NH
Acid sites in micropores (l
mol g 1) c
ꢁ
SiOH density (sites/nm2)d
x
NH
3
adsorption
2,6-dTBP adsorption
AF
Aa (cmꢁ1
)
Nb
(lmol.g
ꢁ1
)
Aa (cmꢁ1
)
Nb
(lmol g )
ꢁ1
0
0
0
.3
.6
.9
6.29
6.80
6.66
42.8
46.2
45.3
1.04
0.58
0.58
2.1
1.2
1.2
0.049
0.025
0.025
40.7 (0.078)
45.1 (0.083)
44.2 (0.080)
0.04
0.13
0.14
a
ꢁ1
Integrated area A of the 1450 and 1618 cm bands were obtained by using the spectra of the catalysts outgassed at 298 K to eliminate the contribution of physisorbed
molecules.
b
= 0.147 cm2
ꢁ1
+
= 0.50 cm2
ꢁ1
e
l
mol [84] for NH
4
band and
e
l
mol [82].
c
2
In parentheses, acid sites per nm of microporous surface area (Table 4).
d
SiOH sites (estimated on the basis of H ss NMR abundances presented in Table 5) per nm2 of mesoporous surface area (Table 4).
1
85

L. Gao, I. Miletto, C. Ivaldi et al.
Journal of Catalysis 397 (2021) 75–89
to the generation of Si-OH moieties at the expense of acid Si-O(H)-
Al moieties for microporous SAPO-5/xSi, sitting mostly on the
mesopores. Despite their different location on the catalyst, the
interconnectivity of micro- and mesopores in the HPSAPO-5/xSi
catalysts favors the cooperativity between the Brønsted acid sites
and Si-OH groups during acid-catalyzed reactions, as recently
unveiled by combining spectroscopic (inelastic neutron scattering,
FTIR and MAS NMR) and computational analyses on HPSAPO-5/xSi
catalysts [60]. This property is expected to facilitate the activation
of MIBK and further attack to FF, catalyzing the aldol condensation/
crotonization reaction (Scheme 3), considering also that the SiOH
density is similar for HPSAPO-5/xSi catalysts with 0.6 and 0.9 Si
content and it is low for HPSAPO-5/0.3Si (Table 6). In light of these
considerations, the highest catalytic activity observed for HPSAPO-
5
/0.9Si can be attributed to the synergistic effect between the
amount of OH sites and its hierarchical structure (highest meso-
porous area and pore volume, Table 4).
The interaction of MIBK via H-bond with the Si-OH was con-
firmed by MIBK adsorption on the hierarchical catalysts, followed
by FTIR spectroscopy. Upon MIBK adsorption (Fig. S17), the band
ꢁ
1
ascribed to the O-H stretching of Si-OH groups (3745 cm ) van-
ishes, whereas characteristic bands attributed to the vibrational
features of MIBK appear at lower wavenumbers. On the contrary,
ꢁ1
the bands ascribed to Brønsted acid sites (3630 and 3513 cm
)
are visible, suggesting that Brønsted sites are still available. The
shift of the C-O stretching mode of adsorbed MIBK with respect
to that in liquid phase confirms that MIBK and SiOH interact by
H-bonding (Table S2). We also measured FF adsorption after MIBK
outgassing. Interestingly, the bands due to the O-H stretching of
Brønsted sites completely disappear. However, the bands ascribed
to Si-OH groups are visible, pointing out a preferential adsorption
of FF in the micropores (Fig. S18). Overall, these results confirm
the mechanism proposed in Scheme 3, encompassing the preferen-
tial adsorption of MIBK on the external SiOH groups.
Fig. 8. NH
3
-TPD profiles of (A) fresh and (B) spent HPSAPO-5/0.9Si catalysts. The
peak deconvolution can be found in the SI (Table S2).
3
3 3
.6.3.3. NH -TPD profiles. NH -TPD measurements were also per-
formed on the fresh and spent HPSAPO-5/0.9Si after the aldol con-
densation/crotonization reaction to gain a better insight into the
density distribution and strength of acid sites (Fig. 8, Table S3 –
NH
integrating the NH
belonging to bands I-IV is 775 mmol.g , whereas the density of
3
as molecular probe, can be ascribed to Brønsted acid sites. By
3
-TPD profile, the density of weak acid sites
ꢁ1
ꢁ1
entries 1 and 2). The NH
two main regions (Fig. 8A): (1) a low-temperature region (100–
50 °C), and (2) a high-temperature region (450–650 °C). The first
3
-TPD profile of the fresh catalysts exhibits
strong acid sites is 32.7 mmol.g . The latter value compares well
with the density of sites measured by FTIR using NH as probe
3
ꢁ1
3
molecule (45.3 mmol.g ), confirming that band VI can be assigned
to strong Brønsted acid sites.
region can be deconvoluted into four bands (I-IV) that can be
attributed to weakly acid sites in P-OH and Si-OH groups in line
with the Al, P and Si NMR MAS findings. Besides, NH
associations promoted by H-bonding cannot be excluded. The sec-
ond region is constituted by a broad band (VI) that is indicative of
strong acid sites and that, on the guidance of the FTIR spectra using
To further rationalize the higher activity of HPSAPO-5/0.9Si
2
7
31
29
+
.nNH
3
compared to reference acid catalysts, we measured the NH
profiles for the series of catalysts listed in Table 1 (Figs. S21,
S22A-B). ZrO and ZnO exhibit bands mainly located in the low-
temperature region (in blue), whereas the three acid zeolites (i.e.
3
-TPD
4
2
Scheme 3. Possible catalytic mechanism in the aldol condensation/crotonization reaction of FF with MIBK catalyzed by HPSAPO-5/0.9Si.
86

L. Gao, I. Miletto, C. Ivaldi et al.
Journal of Catalysis 397 (2021) 75–89
BEA, Y and MOR) also display bands in the high-temperature
region (in red), but centered at a temperature lower than 550 °C.
sion is observed with Pd nanoparticles smaller than 2.5 nm for
0.3%/HPSAPO-5/0.9Si, while two particle distributions centered at
The deconvoluted NH
in the high-temperature region centered at 500 °C, 605 °C and
30 °C (Fig. S22C, Table S3 entry 4), being mainly attributed to
Lewis acid centers [75]. These observations point out that strong
Lewis acid centers are detrimental to the catalytic activity.
3
-TPD profile of Al-SBA-15 shows three bands
2
5 nm and 14 nm are observed for 4.8%/HPSAPO-5/0.9Si. The H -
TPR profiles revealed in both cases the presence of a reduction
band centered at a temperature up to 60 °C, confirming the fast
and complete reduction of the palladium phase in the one-pot
7
2
reaction just upon introduction of H to the reactor after ALD-1
3
The key features of the NH -TPD profile on HPSAPO-5/0.9Si are
formation.
essentially kept after the aldol condensation/crotonization reaction
at 180 °C (Fig. 8B). Nonetheless, the low-temperature region
expands to 450 °C due to the apparition of a new broad band (V)
centered at 363 °C and a drastic decline of band I. The density of
3
We also measured the NH -TPD profile of the unreduced 4.8%
Pd/HPSAPO-5/0.9Si catalyst (Fig. S20, Table S2 – entry 3). The key
bands on the parent HPSAPO-5/0.9Si are essentially kept after Pd
impregnation. However, bands I and II become more intense with
ꢁ1
ꢁ1
weak acid sites belonging to bands I-V is 1076 mmol.g , which is
larger than that of the fresh catalyst, pointing out the generation
of additional weakly acid centers (band V) during the reaction.
The generation of new acid centers can be attributed to partial
hydroxylation of the external surface of HPSAPO-5/0.9Si during
the reaction [85], resulting most likely in a preferential location
near the micropore mouths not far from Al atoms. The presence
of new weak acid centers is not expected to impact the catalytic
activity and catalyst deactivation. In contrast, band VI ascribed to
strong acid sites exhibits a slight decline from 32.7 to 12.7 mmol.
a density of acid sites belonging to bands I-IV of 1163 mmol.g
ꢁ1
compared to 775 mmol.g for HPSAPO-5/0.9Si. This observation
points out the generation of additional weakly acid centers upon
palladium impregnation. In contrast, band VI ascribed to strong
acid sites exhibits a slight shift to lower temperature (from
540 °C to 400 °C) together with an increase of the density of acid
sites from 32.7 to 56.1 mmol.g , which can be attributed to addi-
tional Lewis acidity ascribed to PdO. This additional acid site den-
sity vanishes upon reduction (not shown).
ꢁ1
ꢁ1
g
, which can be attributed to partial blockage due to oligomers
or coke species during the reaction as inferred by thermogravimet-
ric analysys (TGA) (Fig. S19). Indeed, a characteristic weight loss is
observed between 200 and 600 °C on the spent HPSAPO-5/0.9Si
that can be attributed to the formation of aliphatic coke at the level
of 160 mgC/g. The presence of partial catalyst deactivation can
explain the presence of a plateau in the kinetic curves (Fig. 1)
and selectivity-conversion plots (Fig. 2) for the ALD-1 selectivity
at high conversion.
4. Conclusion
A series of crystalline silicoaluminophosphate SAPO-5 catalysts
with hierarchical porosity and variable silicon content were syn-
thesized using
synthesized MCM-41, with the surfactant (CTAB) inside the meso-
pores. In particular, HPSAPO-5/0.9Si catalyst exhibited a remark-
able increase of the catalytic activity compared to the
microporous counterpart in the aldol condensation/crotonization
reaction between furfural and methyl isobutyl ketone.
a
bottom-up method starting from pre-
3.6.4. Nature of the Pd phase over HPSAPO-5/xSi
To gain insight into the Pd dispersion on HPSAPO-5/0.9Si, we
HPSAPO-5/0.9Si exhibited a lower density of Brønsted acid cen-
visualized the 0.3%/HPSAPO-5/0.9Si and 4.8%/HPSAPO-5/0.9Si sam-
ples by HR-TEM (Fig. 9). In both cases, a high degree of Pd disper-
ters compared to microporous SAPO-5/0.9Si, as inferred by com-
1
bined H MAS NMR and in situ FTIR spectroscopy using NH
3
as
Fig. 9. HR-TEM micrographs of (A1,A2) 0.3%Pd/HPSAPO-5/0.9Si and (B1,B2) 4.8%Pd/HPSAPO-5/0.9Si.
87

L. Gao, I. Miletto, C. Ivaldi et al.
Journal of Catalysis 397 (2021) 75–89
probe molecule. However, HPSAPO-5/0.9Si displayed a higher
mesoporous specific surface area combined with the presence of
Si-OH moieties on the external surface. The interconnectivity of
micro- and mesopores in HPSAPO-5/xSi catalysts is expected to
favor the cooperativity between strong Brønsted acid sites located
in the SAPO-5 channels, which are characterized by a band cen-
[24] B. Pholjaroen, N. Li, J. Yang, G. Li, W. Wang, A. Wang, Y. Cong, X. Wang, T.
Zhang, Ind. Eng. Chem. Res. 53 (2014) 13618–13625.
[
29] H. Chang, A. Hussain Motagamwala, G.W. Huber, J.A. Dumesic, Green Chem. 21
2019) 5532–5540.
[30] H.J. Cho, D. Kim, J. Li, D. Su, B. Xu, J. Am. Chem. Soc. 140 (2018) 13514–13520.
(
[
[
[
[
[
[
31] S. Jiang, E. Muller, M. Pera-Titus, F. Jerôme, K. De Oliveira Vigier,
ChemSusChem. 13 (2020) 1699–1704.
32] S. Jiang, C. Ma, E. Muller, M. Pera-Titus, F. Jerome, K. De Oliveira Vigier, ACS
Catal. 10 (2019) 8893–8902.
33] W. Shen, G.A. Tompsett, K.D. Hammond, R. Xing, F. Dogan, C.P. Grey, W. Curtis
Conner Jr, S.M. Auerbach, G.W. Huber, Appl. Catal. A: Gen. 392 (2011) 57–68.
34] I. Sádaba, M. Ojeda, R. Mariscal, R. Richards, M. López Granados, Catal. Today
167 (2011) 77–83.
35] I. Sádaba, M. Ojeda, R. Mariscal, J.L.G. Fierro, M. López Granados, Appl. Catal. B:
Environ. 101 (2011) 638–648.
3
tered at 540 °C in the NH -TPD profile, and external Si-OH groups,
facilitating the activation of MIBK and further attack to FF.
By impregnating Pd, 4.8%Pd/HPSAPO-5/0.9Si afforded the syn-
thesis of hydrogenated THF-derivatives with 45% yield in the
one-reactor tandem aldol condensation/crotonization reaction
between furfural and methyl isobutyl ketone, followed by hydro-
36] L. Faba, E. Díaz, S. Ordónez, Appl. Catal. B: Environ. 113–114 (2012) 201–211.
genation with molecular H
2
. The catalytic activity declined after
[37] D. Nguyen Thanh, O. Kikhtyanin, R. Ramos, M. Kothari, P. Ulbrich, T. Munshi, D.
Kubicka, Catal. Today 277 (2016) 97–107.
the reaction mostly due formation of carbon deposits, but could
be recovered after calcination.
[
38] O. Kikhtyanin, R. Bulanek, K. Frohlich, J. Cejka, D. Kubicka, J. Mol. Catal. A: Gen.
24 (2016) 358–368.
39] O. Kikhtyanin, Y. Ganjkhanlou, D. Kubicka, R. Bulanek, J. Cejka, Appl. Catal. Gen.
49 (2018) 8–18.
4
[
5
Declaration of Competing Interest
[
40] A. Philippou, M.W. Anderson, J. Catal 189 (2000) 395–400.
[
41] G. Liang, A. Wang, X. Zhao, N. Lei, T. Zhang, Green Chem. 18 (2016) 3430–3438.
The authors declare that they have no known competing finan-
cial interests or personal relationships that could have appeared
to influence the work reported in this paper.
[42] O. Kikhtyanin, V. Kelbichova, D. Vitvarova, M. Kubu, D. Kubicka, Catal. Today
27 (2014) 154–162.
43] O. Kikhtyanin, P. Chlubna, T. Jindrova, D. Kubicka, Dalton Trans. 43 (2014)
0628–10641.
44] J. Cueto, V. Korobka, L. Faba, E. Diaz, S. Ordonez, A.C.S. Sust, Chem. Eng. 8
2020) 4371–4383.
2
[
[
1
(
Acknowledgements
[
45] O. Kikhtyanin, D. Kubicka, J. Cejka, Catal. Today 243 (2015) 158–162.
46] D. Verboekend, J. Perez-Ramírez, Catal, Sci. Technol. 1 (2011) 879–890.
[
This project has received funding from the European Union’s
Horizon 2020 research and innovation program under grant agree-
ment N. 720783-MULTI2HYCAT. The authors would like to provide
their gratitude to Dr Urbano Diaz from ITQ (Valencia, Spain) for
providing a MOR (Si/Al = 6.1) sample.
[47] J. Aguado, J. Sotelo, D. Serrano, J. Calles, J. Escola, Energy Fuels 11 (1997) 1225–
231.
48] V.L. Zholobenko, A.Y. Khodakov, D. Durand, Micropor. Mesopor. Mater. 66
2003) 297–302.
[49] X. Zhu, J.P. Hofmann, B. Mezari, N. Kosinov, L. Wu, Q. Qian, B.M. Weckhuysen,
S. Asahina, J. Ruiz-Martinez, E.J.M. Hensen, ACS Catal. 6 (2016) 2163–2177.
1
[
(
[
50] R. Ramos, J.M. Hidalgo, M. Göpel, Z. Tisler, F. Bertella, A. Martinez, O.
Kikhtyanin, D. Kubicka, Catal. Commun. 114 (2018) 42–45.
[
[
51] M. Su, W. Li, Q. Ma, S. Li, T. Yang, X. Dou, Energy Fuels 33 (2019) 12518–12526.
52] A. Al-Ani, C. Freitas, V. Zholobenko, Micropor. Mesopor. Mater. 293 (2020)
Appendix A. Supplementary material
109805.
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.jcat.2021.03.003.
[53] M. Hartmann, L. Kevan, Chem. Rev. 99 (1999) 635–664.
[
54] E. Gianotti, M. Manzoli, M.E. Potter, V.N. Shetti, D. Sun, J. Paterson, T.M. Mezza,
A. Levy, R. Raja, Chem. Sci. 5 (2014) 1810–1819.
[
[
[
55] R.J. Pellet, G.N. Long, J.L. Rabo, P.K. Coughlin, US 4,751,340, 1988.
56] R.J. Pellet, G.N. Long, J.L. Rabo, P.K. Coughlin, US 4,740,650, 1988.
57] D. Verboekend, M. Milina, J. Perez-Ramirez, Chem. Mater. 26 (2014) 4552–
References
4
562.
58] D. Jadav, R. Bandyopadhyay, M. Bandyopadhyay, Eur. J. Inorg. Chem. (2020)
47–853.
[
[
[
[
1] M. Dusselier, M. Mascal, B.F. Sels, Top. Curr. Chem. 353 (2014) 1.
2] D.M. Alonso, J.Q. Bond, J.A. Dumesic, Green Chem. 12 (2010) 1493.
3] P. Anastas, N. Eghbali, Chem. Soc. Rev. 39 (2010) 301.
4] A.A. Koutinas, C. Du, R.H. Wang, C. Webb, Introduction to Chemicals from
Biomass, Wiley, Hoboken, 2008, p. 77.
[
[
8
59] C. Wang, M. Yang, P. Tian, S. Xu, Y. Yang, D. Wang, Y. Yuan, Z. Liu, Dual
template-directed synthesis of SAPO-34 nanosheet assemblies with improved
stability in the methanol to olefins reaction, J. Mater. Chem. A 3 (2015) 5608–
[
[
5] Y. Ogiwara, T. Uchiyama, N. Sakai, Angew. Chem. Int. Ed. 55 (2016) 1864.
6] R. Mariscal, P. Maireles-Torres, M. Ojeda, I. Sádaba, M. López Granados, Energy
Environ. Sci. 9 (2016) 1144.
5616.
[
[
[
60] S.H. Newland, W. Sinkler, T. Mezza, S.R. Bare, M. Caravetta, I.M. Haies, A. Levy,
S. Keenan, R. Raja, ACS Catal. 5 (2015) 6587–6593.
[
7] R.J. van Putten, J.C. van der Waal, E. de Jong, C.B. Rasrendra, H.J. Heeres, J.G. de
Vries, Chem. Rev. 113 (2013) 1499.
61] S. Chapman, A.J. O’Malley, I. Miletto, M. Carravetta, P. Cox, E. Gianotti, L.
Marchese, S.F. Parker, R. Raja, Chem. Eur. J. 25 (2019) 9938–9947.
62] I. Miletto, G. Paul, S. Chapman, G. Gatti, L. Marchese, R. Raja, E. Gianotti, Chem.
Eur. J. 23 (2017) 9952–9961.
63] N. Danilina, F. Krumeich, J.A. van Bokhoven, J. Catal. 272 (2010) 37–44.
64] S. Chapman, M. Carravetta, I. Miletto, C.M. Doherty, H. Dixon, J.D. Taylor, E.
Gianotti, J. Yu, R. Raja, Angew. Chem. Int Ed. 59 (2020) 2–11.
65] J.A. Rabo, G.A. Gajda, Catal. Rev. Sci. Eng. 31 (1989) 385–430.
66] B.M. Lok, C.A. Messina, R.L. Patton, R.T. Gajek, T.R. Cannan, E.M. Flanigen, J. Am.
Chem. Soc. 106 (1984) 6092–6093.
67] J.A. Rabo, In E.G. Derouane, F. Lemos, C. Naccache, F. R. Ribeiro, (Eds.) Zeolite
Microporous Solids: Synthesis, Structure and Reactivity, NATO ASI Ser. C 352,
[
8] X. Li, P. Jia, T. Wang, ACS Catal. 6 (2016) 7621–7640.
[
9] T. Werpy, G. Petersen, (Eds.), Top Value Added Chemicals from Biomass, U.S.
Department of Energy (DOE) report: DOE/GO-102004-1992, 2004.
[
[
[
[
10] L. Hu, L. Lin, S. Liu, Ind. Eng. Chem. Res. 53 (2014) 9969.
11] L. Bui, H. Luo, W.R. Gunther, Y. Roman-Leshkov, Angew. Chem. Int. Ed. 52
(
2013) 8022.
12] M. Kim, Y. Su, A. Fukuoka, E.J.M. Hensen, K. Nakajima, Angew. Chem. Int. Ed. 57
2018) 8235.
13] N. Alonso-Fagúndez, M.L. Granados, R. Mariscal, M. Ojeda, ChemSusChem. 5
2012) 1984.
14] G. Li, N. Li, Z. Wang, C. Li, A. Wang, X. Wang, Y. Cong, T. Zhang, ChemSusChem.
(2012) 1958.
15] A. Corma, O. de la Torre, M. Renz, N. Villandier, Angew. Chem. Int. Ed. 50 (2011)
375.
16] A.D. Sutton, F.D. Waldie, R. Wu, M. Schlaf, L.A. ‘Pete’ Silks III, J.C. Gordon, Nat.
Chem. 5 (2013) 428–432.
17] G.W. Huber, S. Iborra, A. Corma, Chem. Rev. 106 (2006) 4044.
18] G.W. Huber, J.N. Chheda, C.J. Barrett, J.A. Dumesic, Science 308 (2005) 1446–
450.
19] J. Xu, L. Li, G. Li, A. Wang, Y. Cong, X. Wang, N. Li, A.C.S. Sust, Chem. Eng. 6
2018) 6126–6134.
20] C.J. Barrett, J.N. Chheda, G.W. Huber, J.A. Dumesic, Appl. Catal. B: Envion. 66
2006) 111–118.
21] R. Xing, A.V. Subrahmanyam, H. Olcay, W. Qi, G.P. van Walsum, H. Pendse, G.
W. Huber, Green Chem. 12 (2010) 1933–1946.
22] J. Yang, N. Li, G. Li, W. Wang, A. Wang, X. Wang, Y. Cong, T. Zhang,
ChemSusChem 6 (2013) 1149–1152.
[
[
[
[
[
[
[
(
[
(
1
992, p. 531.
5
[
[
[
[
68] G. Sastre, D.W. Lewis, C.R.A. Catlow, J. Phys. Chem. B 101 (1997) 5249–5262.
69] C.J. Brinker, Y. Lu, A. Sellinger, H. Fan, Adv. Mater. 11 (1999) 579–585.
70] M. Thommes, K.A. Cychosz, Adsorption 20 (2014) 233–250.
71] M. Errahali, G. Gatti, L. Tei, G. Paul, G.A. Rolla, L. Canti, A. Fraccarollo, M. Cossi,
A. Comotti, P. Sozzani, L. Marchese, J. Phys. Chem. C 118 (2014) 28699–28710.
72] K.A. Cychosz, R. Guillet-Nicolas, J. Garcia, M. Thommes, Chem. Soc. Rev. 46
2
[
[
[
[
[
(
2017) 389–414.
1
73] G.V.A. Martins, G. Berlier, C. Bisio, S. Coluccia, H.O. Pastore, L. Marchese, J. Phys.
Chem. C 112 (2008) 7193–7200.
74] F. Jing, B. Katryniok, M. Araque, R. Wojcieszak, M. Capron, S. Paul, M. Daturi, J.-
M. Clacens, F. De Campo, A. Liebens, F. Dumeignil, M. Pera-Titus, Catal, Sci.
Technol. 6 (2016) 5830–5840.
[
[
[
[
(
(
[
75] F. Jing, B. Katryniok, S. Paul, L. Fang, A. Liebens, B. Hu, F. Dumeignil, M. Pera-
Titus, ChemCatChem. 9 (2017) 258–262.
88

L. Gao, I. Miletto, C. Ivaldi et al.
Journal of Catalysis 397 (2021) 75–89
[
76] W. Lai, S. Yang, Y. Jiang, F. Zhao, Z. Li, B. Zaman, M. Fayaz, X. Li, Y. Chen,
Adsorption 26 (2020) 633–644.
77] B. Zibrowius, E. Loffler, M. Hunger, Zeolites 12 (1992) 167–174.
78] J. Chen, P.A. Wright, S. Natarajan, J.M. Thomas, Stud. Surf. Sci. Catal. 843 (1994)
[83] K. Sadowska, K. Gorà-Marek, J. Datka, J. Phys. Chem. C 117 (2013) 9237–9244.
[84] J. Datka, B. Gil, A. Kubacka, Zeolites 15 (1995) 501–506.
[85] B.M. Lok, B.K. Marcus, C.L. Angell, Zeolites 6 (1986) 185–194.
[86] S.G. Wettstein, D.M. Alonso, Y. Chong, J.A. Dumesic, Energy Environ. Sci. 5
(2012) 8199–8203.
[
[
1
731–1738.
[
[
79] S.I. Lee, H.J. Chon, Chem. Soc., Faraday Trans. 93 (1997) 1855–1860.
80] G. Müller, J. Bodis, G. Eder-Mirth, J. Kornatowski, J.A. Lercher, J. Mol. Struct.
[87] L. Faba, E. Diaz, S. Ordonez, ChemSusChem 7 (2014) 2816–2820.
[88] C. Ki, D. Ding, Q. Xia, Y. Wang, ChemSusChem 9 (2016) 1712–1718.
[89] L. Faba, E. Diaz, S. Ordonez, Catal. Today 164 (2011) 451–456.
[90] M. Li, X. Xu, Y. Gong, Z. Wei, Z. Hou, H. Li, Y. Wang, Green Chem. 16 (2014)
4371–4377.
4
10–411 (1997) 173–178.
81] C.V. Hidalgo, H. Itoh, T. Hattori, N. Niwa, Y. Murakami, J. Catal. 85 (1984) 362–
69.
82] K. Gòra-Merek, K. Tarach, M. Choi, J. Phys. Chem. C 118 (2014) 12266–12274.
[
[
3
89