Mesoporous Al-promoted sulfated zirconia as an efficient heterogeneous catalyst to synthesize isosorbide from sorbitol

Article abstract of DOI:10.1016/j.apcata.2018.06.024

Mesoporous aluminum-promoted sulfated zirconia (named as mAl-SZ) was directly prepared by grind method, and firstly used to catalyze the solvent-free dehydration of sorbitol to isosorbide. The physicochemical properties of as-prepared catalysts were characterized by FT-IR, TGA, XRD, N₂ sorption, NH₃-TPD and pyridine-infrared (IR) spectroscopy techniques in order to elucidate the relevance of the catalyst properties and the catalytic performance. It was found that the catalysts prepared by grind method possessed mesoporous structures with high surface area, pure tetragonal phase as well as high sulfur contents, which were advantage to eliminate diffusion limitation and generate abundant acidic sites, especially Br?nsted acidic sites. Furthermore, aluminum promoters could contribute to the considerable increase of the strong acidic sites and ratio of Br?nsted to Lewis acidic sites. Consequently, the mesoporous sulfated zirconia with 6 mol% Al-promoter (6Al-SZ) attained porous structure with improved acidic properties, thus showing the optimal catalytic behavior. The full sorbitol conversion with 73% isosorbide selectivity was achieved under milder conditions (175 °C, 2 h) than those of similar type of catalysts. In addition, the 6Al-SZ exhibited favorable reusability with insignificant drop in isosorbide yield during five reaction cycles.

Full text of DOI:10.1016/j.apcata.2018.06.024

Applied Catalysis A, General 562 (2018) 258–266
Contents lists available at ScienceDirect
Applied Catalysis A, General
journal homepage: www.elsevier.com/locate/apcata
Mesoporous Al-promoted sulfated zirconia as an efficient heterogeneous
catalyst to synthesize isosorbide from sorbitol
a,b
a,⁎
a,b
a
a
Yi Zhang , Tong Chen , Gang Zhang , Gongying Wang , Hua Zhang
a
Chengdu Institute of Organic Chemistry, Chinese Academy of Sciences, Chengdu 610041, China
University of Chinese Academy of Sciences, Beijing 100049, China
b
A R T I C L E I N F O
A B S T R A C T
Keywords:
Mesoporous aluminum-promoted sulfated zirconia (named as mAl-SZ) was directly prepared by grind method,
Biomass
and firstly used to catalyze the solvent-free dehydration of sorbitol to isosorbide. The physicochemical properties
Mesoporous sulfated zirconia
Aluminum promotor
Sorbitol dehydration
Isosorbide
of as-prepared catalysts were characterized by FT-IR, TGA, XRD, N
2 3
sorption, NH -TPD and pyridine-infrared
(IR) spectroscopy techniques in order to elucidate the relevance of the catalyst properties and the catalytic
performance. It was found that the catalysts prepared by grind method possessed mesoporous structures with
high surface area, pure tetragonal phase as well as high sulfur contents, which were advantage to eliminate
diffusion limitation and generate abundant acidic sites, especially Brönsted acidic sites. Furthermore, aluminum
promoters could contribute to the considerable increase of the strong acidic sites and ratio of Brönsted to Lewis
acidic sites. Consequently, the mesoporous sulfated zirconia with 6 mol% Al-promoter (6Al-SZ) attained porous
structure with improved acidic properties, thus showing the optimal catalytic behavior. The full sorbitol con-
version with 73% isosorbide selectivity was achieved under milder conditions (175 °C, 2 h) than those of similar
type of catalysts. In addition, the 6Al-SZ exhibited favorable reusability with insignificant drop in isosorbide
yield during five reaction cycles.
1. Introduction
and easy synthetic route, which requires strong acidity, especially
strong Brönsted acidic sites to enhance selectivity of isosorbide [7,9].
Currently, continuous efforts have been paid to the utilization of
biomass as renewable and sustainable feedstocks for the production of
This process is industrially carried out under homogeneous catalysis in
the presence of strong mineral acids such as H PO , H SO and HCl
3
4
2
4
high-value chemicals or liquid fuels to fulfill sustainable development
[10,11]. Although a high value about 75% of isosorbide yield is
achieved under mild condition, these liquid acids bring obstacles from
the standpoint of environment, safety, corrosiveness, separation and
operating cost. Therefore, the development of green heterogeneous
solid acid catalyst to replace conventional mineral acids have prompted
an extensive demand from an economic and environmental perspective.
The various heterogeneous solid acid catalysts, such as metal (IV)
[
1]. Sorbitol is considered to be one of the most promising platform
chemicals for biomass conversion, which is easily obtained from cel-
lulose via glucose hydrogenation [2]. Among the various sorbitol-de-
rived chemicals, isosorbide is regarded as one of the most potential
derivates in view of its typical application in the medicine and surfac-
tant fields due to its special molecular structure [3]. Especially, in-
corporation of isosorbide monomer into the poly (ethyleneter-
ephthalate) polymer (PET) might greatly enhance its glass transition
temperature and impact resistance [4]. In recent years, isosorbide is
also regarded as a promising alternative of bisphenol A in the synthesis
process of epoxy resins or polycarbonate to eliminate environmental
pollution [5,6].
2 5
phosphates [12–14], phosphoric acid modified Ta O [15], supported
heteropolyacid [16] and cellulose-derived solid acid catalyst [17] had
been studied for the catalytic dehydration of sorbitol. Most of those
catalysts usually required high reaction temperature but obtained un-
satisfactory isosorbide yield. An acceptable isosorbide yield of 70–80%
was obtained under long reaction time in the presence of Hβ zeolite
[18–20], but a large amounts of catalyst loading was required. Fur-
thermore, the Hβ zeolites have the disadvantages of being expensive
catalysts. On the other hand, sulfonic acid resins and mesostructured
SBA-15 silica functionalized with organic sulfonic acid groups were
Generally, isosorbide is produced by double dehydration of sorbitol
as displayed in Scheme S1 [7]. In the view of green chemistry, the
process for dehydration of sorbitol under solvent-free and milder con-
ditions would be highly promising [8]. However, this is not a simple
⁎
Corresponding author.
E-mail address: chentongw@sina.com.cn (T. Chen).
https://doi.org/10.1016/j.apcata.2018.06.024
Received 10 May 2018; Received in revised form 12 June 2018; Accepted 13 June 2018
0926-860X/ © 2018 Published by Elsevier B.V.

Y. Zhang et al.
Applied Catalysis A, General 562 (2018) 258–266
favorable for the conversion of sorbitol into isosorbide, while high-cost
and low thermal stability constrained their industrial applications
Thermogravimetric analysis (TGA) was carried out by means of a
−1
NETZSCH STA 449F3 instrument, a heating rate of 10 K min , an air
−1
[
21–24].
Among the various solid acids catalysts, the sulfated metal oxides
such as ZrO , SnO , TiO , and Nb have attracted considerable at-
gas flow of 100 ml min , over the 313–1273 K temperature range. The
sulfur content was calculated based on the weight loss from 873 to
2
2
2
2
O
5
1273 K. The weight loss corresponded to the content of SO
content could be 40% of this weight loss [36].
3
, and sulfur
tentions as the promising catalysts for industrial processes, which are
used in various hydrocarbon reactions due to their high thermal sta-
bility, eco-friendly, as well as strong acidic properties [25,26]. In recent
years, several groups reported the utilization of sulfated metal oxides
for the conversion of sorbitol to isosorbide [27–30]. Among these sul-
X-ray diffraction (XRD) patterns were recorded in the 2θ range of
20–70° on a Bruker D8 ADVANCE diffractometer with Cu Kα radiation
source (1.5418 Å). The crystalline size of crystal phase was determined
from the peak width of characteristic peak using Scherrer’s equation:
D = Kλ/βcosθ, where K = 0.9, D represented crystallite size, λ re-
presented the wavelength of Cu Kα radiation, and β represented the
corrected half-width of the diffraction peak.
2−
fated metal oxides, SO
4
2
/ZrO have been extensively concerned due
to its relatively high Brønsted acid strength and acidic sites, which are
preferable to sorbitol dehydration [31]. Also, sulfated zirconia is less
expensive and readily available at industrial scale. However, conven-
tional preparation of sulfated zirconia is generally accomplished by
post-sulfonation and calcination method, which contribute to a rela-
A Micromeritics ASAP 2010 was used to determine BET surface
area, pore volume and pore size at 77 K using N adsorption. The BET
2
surface areas were determined by the BET method. The total pore vo-
lume was calculated from the amount of vapor adsorbed at a relative
2−
tively weak interaction between SO
4
2
and ZrO framework with low
0
sulfur content, thereby the low ratio of Brønsted to Lewis acidic sites
and low activity. Some control on sulfur loading of sulfated zirconia can
be achieved by varying the calcination temperature, but the favorable
tetragonal phase and crystallinity are affected. Additionally, poor por-
osity such as small pore size and low pore volume inhibits their in-
herent catalytic activities for bulky biomass-derived molecules like
polyalcohol [32]. Therefore, some unconventional preparation methods
were conducted to improve its porosity and strengthen its interactions
pressure (p/p ) close to unity. Pore size distribution curves were es-
tablished from the desorption branches of isotherms using the
Barrett–Joyner–Halenda (BJH) model.
3
Temperature-programmed desorption of ammonia (NH -TPD)
measurement was performed on a Micro TP-5080 automatic adsorption
instrument equipped with a thermal conductivity detector (TCD) to
determine the acid strength and acid amount of catalysts. Before ad-
sorption, 100 mg catalyst loaded in a quartz reactor was pretreated at
−
1
between sulfate species and ZrO
2
framework, which could lead to an
300 °C for 2 h under N
2
flow (50 mL min ). After cooling down to
adsorption was performed by N flow to a
(50 mL min ) and maintaining the tem-
to remove physically adsorbed
-TPD measurements were taken up to 650 °C at a heating
enhanced catalytic activity and stability [33–35].
room temperature, NH
3 2
stream of 10 vol% NH /N
3
2
−
1
In the present study, a mesoporous aluminum metal promoted sul-
fated zirconia with the high sulfur contents and ratio of Brönsted to
Lewis acidic sites was synthesized by a simple grind method and used as
an efficient catalyst in solvent-free dehydration of sorbitol to iso-
sorbide. This study systematically evaluated the correlations between
catalyst structure and catalytic performance in sorbitol dehydration at
different reaction parameters such as catalyst amount, reaction tem-
perature and reaction time. In addition, its catalytic performance was
compared with that of conventional sulfated zirconia prepared by post-
sulfonation and calcinations method. It was found that grinding pre-
pared 6Al-SZ exhibited superior catalytic activity and reliable reusa-
bility under relatively mild reaction conditions.
perature for 1 h. After a purge with N
ammonia, NH
rate of 10 °C min in flowing N
2
3
−1
−1
2
at a flow rate of 50 mL min and the
process was monitored by a thermal conductivity detector (TCD).
The types of acid sites of the as-prepared catalysts were determined
by pyridine-infrared (IR) spectroscopy (Nicolet 5700), self-supporting
wafers of samples were placed into a quartz cell and evacuated at
−2
300 °C for 2 h (10 Pa) to remove the adsorbed impurities over the
catalyst surface. After cooling to 30 °C, the pretreated samples were
exposed to pyridine for 1 h and then outgassed at 150 °C or 300 °C for
1 h to remove the physically adsorbed pyridine, after which the FT-IR
spectra were recorded. The ratios of Brönsted to Lewis acid sites were
calculated from integrated areas of the peaks located about 1540 and
2. Experimental
−1
1
446 cm
.
2.1. Catalysts preparation
2.3. Activity measurements
Mesoporous Al-promoted sulfated zirconia was prepared by simple
grind method under solvent-free condition. Typically, a mixture of
.22 g ZrOCl ·8H O, 7.92 g (NH SO and desirable amounts of Al
NO ·9H O was ground for 20 min. After standing for 24 h, the sample
was calcined at 600 °C in air for 5 h, which was denoted as mAl-SZ, m
referred to molar percentage ratio of Al/Zr. Accordingly, mesoporous
sulfated zirconia was prepared by the same method with the exception
The solvent-free dehydration of sorbitol was carried out in a 50 mL
3
(
2
2
4
)
2
4
round bottom glass reactor attached to a vacuum pump to maintain the
reaction pressure of 40 kPa. Typically, 10 g sorbitol and a certain
amounts of catalyst were placed into the flask and allowed to melt at
3
)
3
2
1
30 °C under mechanical agitation. The reaction mixture was quickly
heated to designed temperature. After reaction, the reaction products
were cooled to room temperature and an appreciable amount of water
was added. The resulting mixture was centrifuged to separate the cat-
alyst. Liquid solution was analyzed by HPLC (Waters e2695) equipped
of a mixture only containing ZrOCl
pared catalyst was labelled SZ.
2 2 4 2 4
·8H O and (NH ) SO . The as-pre-
For comparison, conventional sulfated zirconia, denoted as SZ-C,
was prepared by post-sulfonation and calcinations method as pre-
viously described [30]. Briefly, 2.0 g of zirconium (IV) hydroxide was
with a 2414 refractive index (RI) detector and Asahipak column (NH
2
P-
5
0 4E, No. N1670026). Acetonitrile/water (80/20) mixture was used as
2 4
added into 40.0 mL of 0.05 M aqueous H SO , then vigorously stirred
an eluent for the analysis with the flow rate of 1.0 mL/min. The tem-
perature of RI detector was maintained at 35 °C throughout the ana-
lysis.
The recovered catalyst was washed with a large amounts of alcohol
and then regenerated by calcined at 500 °C for 2 h to recover its original
white. Thereafter, it was used again for the dehydration of sorbitol as
previously described process to evaluate its reusability.
for 5 h at room temperature. The suspension was centrifuged to obtain
the white solid powders. After dried at 100 °C overnight, the obtained
solids were calcined at 600 °C for 5 h.
2.2. Catalysts characterization
Fourier transform infrared (FT-IR) spectra was obtained with a
i
Conversion of sorbitol (Csorbitol) and product selectivity (S ) were
defined according to the following equation:
Nicolet Magna-IR 560 spectrometer in the wavelength range of
−1
−1
4000–400 cm
and the resolution was 4 cm , using KBr powders.
259

Y. Zhang et al.
Applied Catalysis A, General 562 (2018) 258–266
Fig. 2. TG curves of the catalysts: (a) SZ-C, (b) SZ, (c) 2Al-SZ, (d) 6Al-SZ, (e)
2
Fig. 1. FT-IR spectra of catalysts: (a) ZrO , (b) SZ-C, (c) SZ, (d) 2Al-SZ, (e) 6Al-
SZ, (f) 10Al-SZ.
10Al-SZ.
Table 1
C
sorbitol% = (moles of reacted sorbitol/moles of initial sorbitol) ∗ 100%
Sulfur content of catalysts based on TGA.
Catalysts
SZ-C
SZ
2Al-SZ
6Al-SZ
10Al-SZ
i
S % = (moles of carbon in a product defined/moles of carbon in sor-
bitol reacted) ∗ 100%
Weight loss above 600 °C (wt%)
S content (wt%)
4.5
1.8
22.8
9.1
25.2
10.1
29.6
12.0
30.7
12.3
Where i standed for products of 1,4-sorbitan, isosorbide. (trace of 3,6-
sorbitan was located at the same peak with 1,4-sorbitan in HPLC, and
calculated as 1,4-sorbitan). The non-identified soluble and insoluble
compounds including 1,5-sorbitan, 2,5-sorbitan, humins, etc. were
considered as “Unknown”, calculated by carbon balance.
order of the sulfur contents was SZ < 2Al-SZ < 6Al-SZ < 10Al-SZ. It
implied that addition of aluminum to sulfated zirconia could contribute
to the more sulfate groups bonded to ZrO
the FT-IR results.
2
, which was in accord with
Apparently, the temperature of partial sulfate decomposition for the
mAl-SZ became slightly higher than SZ. This meant that partial sulfates
3. Results and discussion
2
more tightly bonded to ZrO for aluminum promoted sulfated zirconia,
3
.1. Characterization of the catalyst
and the additional ca. 5 wt% loss between 740–900 °C for mAl-SZ was
probably associated with the formation of stronger acid sites caused by
the effect of Al-O-Zr [42].
3.1.1. IR-spectroscopy analysis
FT-IR spectra of all catalysts and zirconia were displayed in Fig. 1.
−
1
−1
The characteristic peaks of ZrO
2
at around 740 cm , 620 cm
and
−1
5
80 cm
around 3440 cm
hydroxyl groups and non-dissociated water (δHOH), respectively
39,40]. The IR spectra of all catalysts showed the broad peaks located
were observed for all samples [37,38]. The wide peaks at
3.1.3. X-ray diffraction analysis
−1
−1
and 1650 cm
were attributed to the vibration of
From the XRD patterns in Fig. 3, the ZrO
phases of tetragonal and monoclinic. After introducing sulfate species
for SZ-C, the contents of the monoclinic ZrO decreased, while the
2
displayed a mixture
[
2
−1
2−
at 900–1200 cm , assigned to bidentate SO
zirconia. However, SZ exhibited stronger characteristic peak intensity
as compared to SZ-C, confirming the more sulfate groups bonded to
ZrO
stronger peak intensity around 900–1200 cm
implied that addition of aluminum to sulfated zirconia possibly in-
creased the amount of sulfate groups onto ZrO
4
species coordinated to
tetragonal phase increased and became predominant. Interestingly, the
SZ by grind method showed characteristic peaks of pure tetragonal
phase at 2θ = 30.3, 34.7, 50.2, and 60.2, which was more readily to
2
[41]. Furthermore, 2Al-SZ, 6Al-SZ and 10Al-SZ showed slightly
−1
compared with SZ. It
2
.
3.1.2. Thermogravimetric analysis
The thermal stability of as-prepared catalysts was investigated by
TGA as shown in Fig. 2, and sulfur contents were measured based on the
decomposition of sulfates above 600 °C as displayed in Table 1. Ob-
viously, two weight loss stages for as-prepared samples were observed
in the TGA profiles with a gradual loss below 250 °C and a sharp loss
between 600–900 °C. The first one was in connection with the removal
of physic-adsorbed H
2
O and hydroxyl groups bonded to surface of ZrO
2
,
and the second stage of weight loss was attributed to the removal of SO
3
induced by sulfate species decomposition. All samples showed a neg-
ligible weight loss in the range of 250–600 °C, indicating the sulfated
species stably bonded to zirconia below 600 °C and as-prepared cata-
lysts were suitable for application at operating temperatures. From
Table 1, SZ possessed more sulfur contents than SZ-C, which confirmed
Fig. 3. XRD patterns of the catalyst: (a) ZrO
Al-SZ, (f) 10Al-SZ.
2
, (b) SZ-C, (c) SZ, (d) 2Al-SZ, (e)
6
2
that grind method was beneficial for sulfates bonding to ZrO . The
260

Y. Zhang et al.
Applied Catalysis A, General 562 (2018) 258–266
connect with sulfur containing groups and more advantageous to acid
catalysis reactions [43,44]. The diffraction peaks of pure tetragonal
phase were not obviously changed for 2Al-SZ and 6Al-SZ. However, a
clear peak at about 25.3° in the diffractogram of 10Al-SZ occurred,
which was attributed to the alumina crystal structure [45]. For sulfated
Table 2
Textural properties of as-prepared catalysts.
a
2
Pore volb (cm3/g)
Pore sizec (nm)
Catalysts
SBET (m /g)
SZ-C
SZ
67.9
0.13
0.20
0.25
0.27
0.24
–
3
+
4+
110.1
105.1
108.0
89.1
6.7
7.8
7.2
7.2
zirconia, Al
sites; however, excessive Al
was probably more tended to occupy the surface Zr
2
6
Al-SZ
Al-SZ
3
+
would embed in the bulk of the solid
framework or agglomeration over the surface to form crystalline alu-
mina [46,47].
10Al-SZ
a
S
BET = BET total specific surface area obtained from adsorption data.
The average crystallite size calculated from the characteristic tet-
b
c
0
Pore vol = pore volume was obtained at p/p = 0.95.
Pore
ragonal peak of ZrO
using Debye–Scherrer’s formula, which suggested sulfate groups pre-
sumably prevented the over-growth of ZrO grain. Surprisingly, the SZ
2
and SZ-C were 22.2 nm and 11.3 nm, respectively
size = average
pore
diameters
calculated
using
Barrett–Joyner–Halenda (BJH) method.
2
showed broadened characteristic peaks of tetragonal phase, and the
crystallite size was only 4.9 nm. Small amounts of alumina promotor in
mAl-SZ only led to a bit bigger in crystallite size of 5–7 nm.
It was worthy to mention that grind method resulted in much
smaller crystallite size of 5–7 nm as compared to conventional post-
sulfonation and calcination method. In grind method, the interaction
of nano-crystal size, higher BET surface area as well as pore volume,
which would facilitate the mass-transport and the accessibility for
bulky biomass-derived molecules to more active sites.
3.1.5. Acidic properties
between zirconium species and a large number of (NH
4 2 4
) SO re-
The process of sorbitol dehydration to isosorbide depends much on
markably inhibited the over-growth of ZrO grain so as to form a nano-
2
the acidic properties of catalysts [51]. Herein the NH
ization was carried out to evaluate the acidity of the catalysts. The
desorption signal of NH at 100–250 °C, 250–400 °C and above 400 °C
3
-TPD character-
crystal catalyst. Consequently, the smaller size of nano-crystal was
probably favor to stabilize the tetragonal phases in view of the fact that
crystal phase composition of zirconia was affected by crystallite size
3
suggested a broad distribution of acid strength from weak, medium
strong and very strong acid sites respectively, and the amount of acid
[
48].
sites were associated with the peak areas of the NH
[12,14,27]. As shown in Fig. 5, the SZ-C exhibited a clear NH
rption signal in the temperature range from 100 to 550 °C, suggesting
strong solid acid property. The SZ sample showed a much stronger NH
desorption signal, especially in the high temperature NH desorption
region compared with SZ, indicating more strong acidic sites existed on
catalyst surface. The NH desorption signals of high temperature region
3
-TPD curves
3
.1.4. N
The N
Fig. 4(A). Apparently, all the isotherms exhibited a typical type-IV
isotherm of mesoporous materials with type-H hysteresis loop. The
2
adsorption-desorption analysis
3
deso-
2
adsorption-desorption isotherms of catalysts were shown in
3
2
3
mesopores were probably produced by formation of inter-crystallite
voids because of nano-crystal accumulation [49,50]. The textural
parameters such as BET surface areas, pore volumes and pore diameters
of various samples including SZ-C, SZ, 2Al-SZ, 6Al-SZ and 10Al-SZ were
summarized in Table 2. The pore size distribution of SZ was uniform,
and the average size was about 6.7 nm by BJH method as shown in
Fig. 4(B). The BET surface area and total pore volume of SZ were
3
increased with increasing the amounts of aluminum promotor in mAl-
SZ, and the 6Al-SZ showed the strongest signal [47,52]. From Table 3,
the 6Al-SZ possessed highest acidity with total acidic sites of
3.45 mmol/g. However, the acidity of 10Al-SZ began to slightly de-
2−
crease, probably due to the formation of a mixture of SO
4
/Al
, which was proved by appearance of crystalline Al
Although the total amounts of NH desorption of the mAl-SZ followed
2 3
O and
2
3
2−
1
10.1 m /g and 0.20 cm /g, which was higher than those of conven-
SO
4
/ZrO
2
2 3
O .
2
3
tional SZ-C of 67.9 m /g and 0.13 cm /g, respectively. Generally, high
specific surface area was beneficial to the formation of more acid sites,
and large pore volume as well as uniform mesopores diameter would be
beneficial to the diffusion of reactants and products. The BET surface
area and BJH pore size of 2Al-SZ and 6Al-SZ were similar to SZ.
3
the order SZ ＜ 2Al-SZ ＜ 6Al-SZ > 10Al-SZ, the total desorption
amounts of medium-strong plus very strong acid sites followed the
order of 6Al-SZ > 10Al-SZ > 2Al-SZ > SZ according to Table 3.
Moreover, the pyridine-IR measurements were conducted to eval-
uate the Lewis and Brønsted acid properties of catalysts. As shown in
However, the 10Al-SZ exhibited an apparent drop of BET surface area
2
−1
(
89.1 m /g).
Fig. 6(A), the peaks located at ca. 1540 cm− was assigned to the ex-
1
The results above distinctly proved the effects of grind method on
istence of Brönsted acid sites, ca. 1446 cm
istence of Lewis acid sites; the 1488 cm
was assigned to the ex-
peak was observed from
−1
the structural and textural properties of sulfated zirconia in the matter
Fig. 4. (A) Adsorption-desorption isotherm and (B) pore size distribution of catalysts: (a) SZ-C, (b) SZ, (c) 2Al-SZ, (d) 6Al-SZ, (e) 10Al-SZ.
261

Y. Zhang et al.
Applied Catalysis A, General 562 (2018) 258–266
slightly decreased as shown in Fig. 6(B). However, Lewis acid sites
−
1
(
1446 cm ) was obviously reduced after evacuation at high tem-
peratures. The results demonstrated the presence of stronger Brönsted
acidity than Lewis acidity in grinding prepared catalysts.
Acidic properties of sulfated zirconia were verified to be affected by
both preparation method and aluminum addition based on above re-
sults. An improvement of acid sites were closely related to the presence
of adequate sulfate species for sulfates metal oxides [43,47]. In this
work, the presence of much more bonded sulfate species for SZ was
related to the pure tetragonal phase and small grain size, which con-
versely affected the specific surface area and total acid sites of catalysts.
Meanwhile, the amounts of Brönsted acid sites largely depended on the
surface concentration of the sulfur. The sulfur species for grinding
prepared catalysts exceeded a monolayer coverage, thereby more
Brönsted acid sites rather than Lewis acid sites were generated [42]. In
addition, the promotion effect caused by the formation of AleOeZr not
only increased acidic strength and generated additional acidic sites, but
also affected the distribution of Brönsted and Lewis acidic sites [47].
3
Fig. 5. NH -TPD profiles of the catalysts: (a) SZ-C, (b) SZ, (c) 2Al-SZ, (d) 6Al-
SZ, (e) 10Al-SZ.
3.2. Activity measurements
Table 3
Acidic properties of as-prepared catalysts.
3.2.1. Effect of preparing methods
As shown in Table 4, the complete sorbitol conversion was observed
for both SZ-C and SZ at every tested reaction temperature. By increasing
the reaction temperature, dehydration rate was raised and intermediate
Catalysts Acid amounts (mmol/g)
M + S (mmol/ Total (mmol/
B L
A /A
g)
g)
W
M
S
1
,4-sorbitan was quickly converted to isosorbide. Simultaneously,
SZ-C
SZ
0.55
1.52
1.35
0.78
0.84
0.42
1.22
1.24
1.70
1.30
0.26
0.33
0.63
0.97
0.79
0.68
1.55
1.87
2.67
2.09
1.23
3.07
3.22
3.45
2.93
0.75
1.41
1.47
2.88
2.79
Table 4 obviously showed that the preparation method of sulfated zir-
conia significantly influenced the catalytic performance. The SZ
achieved higher isosorbide selectivity over the reaction temperature
range of 165–205 ℃ as compared to the conventional SZ-C. The better
catalytic performance of SZ was resulted from its mesoporous structure
and nano-crystal size favoring the diffusion for bulky reactant and
product molecules to active sites [21,22], which was beneficial for less
by-products production like “humins”. More importantly, the isosorbide
selectivity was greatly affected by the catalyst acidity and type of acid
sites [13,51]. The larger surface area and small crystallite size for SZ
allowed it containing much more sulfur species than SZ, which was
significant to produce more acidic sites, particularly Brønsted acidic
sites. The abundant accessible active sites allowed 1,4-sorbitan quickly
converted to isosorbide products, thus decreased the formation of hu-
mins [24].
2
6
1
Al-SZ
Al-SZ
0Al-SZ
Note: W, M, and S refer to weak (373–523 K), medium-strong (523–673 K), and
very strong (673–873 K) acid sites.
pyridine adsorbed over both Brönsted and Lewis acid sites [53]. The
ratios of Brönsted to Lewis acid sites (A /A ) were calculated by semi-
B L
quantitative analysis from the relative areas of peaks corresponding to
Brønsted and Lewis acidity, and the results were presented in Table 3
B L
54]. Apparently, the integral areas on the 1540 cm peak and A /A
were much higher for SZ compared with SZ-C, which confirmed that
grinding prepared samples contained much more Brønsted acidic sites.
B L
Furthermore, the A /A exhibited a dependence on aluminum addition,
a maximum of 2.88 was obtained over 6Al-SZ, demonstrated the trace
amounts of aluminum were beneficial to generate more Brønsted acid
sites for mAl-SZ. It was worthy to note that the Brönsted acid sites were
extremely strong in all grinding prepared catalysts even if they were
existed after heating at 300 °C, despite the intensity of the peaks was
−
1
[
Based on these results, it implied that the special structural and
textural properties of SZ, along with the improved acidic properties
resulted into better catalytic behavior in the dehydration of sorbitol as
compared with conventional SZ.
Fig. 6. IR spectra of (a) SZ-C, (b) SZ, (c) 2Al-SZ, (d) 6Al-SZ, (e) 10Al-SZ after pyridine adsorption at room temperature and desorption at (A) 150 °C; and (B) 300 °C.
262

Y. Zhang et al.
Applied Catalysis A, General 562 (2018) 258–266
sorbitol dehydration to isosorbide with different reaction conditions
[7]. Thus the effect of reaction parameters such as catalyst dosage,
reaction temperature and reaction time on isosorbide production were
optimized over the most active 6Al-SZ to achieve high yield.
Table 4
Comparison of catalytic activity over SZ and SZ-C in dehydration of sorbitol to
isosorbide.
Conv.a (%)
Selec. (1,4-sor)b (%)
Selec. (IS)c (%)
Catalysts
Temp. (°C)
Catalyst dosage played an important role in reaction kinetics and
was essential to be optimized. In the absence of catalyst, 1,4-sorbitan
and isosorbide were absolutely not detected with trace sorbitol con-
version at 175 °C and 2 h. Fig .8(A) showed that only 0.05 g 6Al-SZ
greatly improved both sorbitol conversion and isosorbide selectivity to
ca.100% and 39.7% respectively under same conditions. But the se-
lectivity of intermediate 1,4-sorbitan was still as high as 33.3%. With
increasing the dosage of 6Al-SZ to 0.2 g, the isosorbide selectivity
gradually increased to the highest value of 73% with the gradual de-
cline of intermediate 1,4-sorbitan. However, a slight decrease of iso-
sorbide selectivity occurred when the catalyst amount was further in-
creased. It was probably due to the excessive amounts of acid sites
would also favored yield-loss reactions by the accompanying generation
of humins by-products. According to Yamaguchi et al. [56], the self-
polymerization or cross-polymerization of reactant and product mole-
cules resulted in the decline of desired isosorbide selectivity. Therefore,
SZ-C
SZ-C
SZ-C
SZ-C
SZ-C
SZ
SZ
SZ
SZ
SZ
165
175
185
195
205
165
175
185
195
205
100
100
100
100
100
100
100
100
100
100
37.8
24.2
12.2
8.3
29.8
39.3
51.3
54.0
59.0
45.5
56.1
62.4
67.3
66.4
2.1
25.7
10.9
3.9
0.2
0
Reaction conditions: sorbitol 10 g; catalyst 0.2 g; time 2 h.
a
Conversion of sorbitol.
Selectivity to 1,4-sorbitan.
Selectivity to isosorbide.
b
c
3
.2.2. Promotion effect of aluminum
Apart from the textural properties, the acidic properties could also
0
.2 g of the 6Al-SZ was considered to be best amount for later re-
searches.
The effect of reaction temperature on the isosorbide production over
Al-SZ was investigated in the temperature varying from 155 °C to
20 °C under fixed reaction parameters of 0.2 g 6Al-SZ and 2 h. As
play significant role in isosorbide production from sorbitol [14,55].
Considering the similar structure of mAl-SZ compared to SZ as illu-
strated in Section 3.1, hence the isosorbide selectivity was further in-
vestigated in this section to understand the promotion effect of alu-
minum on acidic properties. Observed from Table 5, an obvious
improvement of isosorbide selectivity was observed with increasing
amounts of aluminum promoter in mAl-SZ. As aluminum contents in-
creased up to 6 mol%, the isosorbide selectivity enhanced to 73%. It
was very likely that the small amounts of aluminum content doping in
sulfated zirconia led to a different distribution of acid strength and
concentration of Brönsted acid sites. However, the 10Al-SZ showed
slightly decreased in isosorbide selectivity, which was probably resulted
from a slight decrease of the scattered strong acid sites due to the
6
2
shown in Fig. 8(B), the full sorbitol conversion was achieved when the
reaction temperature was as low as 155 °C. With the increase of the
reaction temperature from 155 °C to 175 °C, the selectivity of 1,4-sor-
bitan decreased and the isosorbide steadily increased up until achieving
its maximum at 175 °C. Thereafter, a sustained decrease occurred and
only 61% isosorbide selectivity was obtained when elevating the re-
action temperature to 220 °C. It suggested that the unknown by-pro-
ducts like humins were observed to increase by increasing the reaction
temperature higher than 175 °C, which contributed to decline of iso-
sorbide selectivity.
2 3
presence of Al O crystal.
To note that a higher number of medium-strong plus very strong
acid sites was required to facilitate the production of isosorbide from
sorbitol as shown in Fig. 7(A). On the other hand, Rusu et al. [14] found
that the moderate and strong acid sites instead of excessively strong
acidity were beneficial to isosorbide production from sorbitol. Thus, it
could be inferred that sorbitol dehydration was favored by moderate
and strong acid sites rather than weak or excessively strong acid sites.
Furthermore, Fig. .7 (B) showed the variation of isosorbide yield and
the B/L ratio as a function of the Al content. The isosorbide yield was
steadily enhanced with the Al content up to 6 mol%, and then reduced
with further increasing. The B/L ratio followed a similar trend, in-
dicated that the most efficient catalyst was related to the highest B/L
ratio. In this sense, the production of isosorbide was favored by
Brönsted acid sites instead of Lewis acid sites. Based on above results,
the isosorbide production from sorbitol was probably supported by
moderate and strong Brönsted acid sites.
The reactions were also conducted at various reaction time intervals
under 175 °C using 0.2 g 6Al-SZ. As shown in Fig. 8(C), the complete
sorbitol conversion with only 37.2% isosorbide selectivity was obtained
at the initial 30 min. With increasing reaction time, the intermediate
1,4-sorbitan selectivity decreased and the isosorbide selectivity con-
tinued increased to 73% within 2 h. When reaction time was prolonged
to 3 h, a slight drop of the isosorbide selectivity from 73% to 69% oc-
curred due to the formation of humins by-product by self-polymeriza-
tion of isosorbide. It was important to note that the intermediate 1,4-
sorbitan was absolutely undetectable after 1.5 h, whereas isosorbide
selectivity still increased until 73% after 1.5 h. The reason might be that
the small amounts of sorbitol isomerized into galactitol, then dehy-
dration of galactitol resulted in the formation of isosorbide [7]. In ad-
dition, the gradual decrease of 1,4-sorbitan selectivity after obtaining
its highest value within 30 min confirmed that 1,4-sorbitan acted as an
intermediate, which was in good agreement with the previous result
[
57].
Therefore, the highest isosorbide yield ca.73% was achieved over
Al-SZ at the optimum operating conditions of 175 °C, 2 h and 0.2 g
3.2.3. Optimization of reaction conditions
6
A large amount of by-products might be formed in the process of
catalyst, which was almost twice that of SZ-C as displayed in Fig. 8(D).
It was also found that the optimum reaction temperature for this work
was obviously lower than that of other works using similar types of
catalyst, which were usually above 200 °C [27,29,30].
Table 5
Effect of aluminum contents on catalytic performance of mAl-SZ.
Al/Zr (mol.%)
Sorbitol Conversion (%)
Selectivity (%)
,4-sorbitan
3.3. Reusability and deactivation of 6Al-SZ
1
Isosorbide
0
2
6
100
100
100
100
11.2
3.9
0
56.1
62.1
73.0
65.3
Catalyst stability and reusability play a significant role in the in-
dustrial feasibility. Thus, the reusability of the 6Al-SZ was investigated
at optimized conditions. The 6Al-SZ catalyst after the first run was di-
rectly used for the next test without the calcination treatment, and the
results were showed in Fig. 9(A). Although sorbitol conversion was
1
0
1.6
Reaction conditions: sorbitol 10 g; catalyst 0.2 g; temp. 175 °C; time 2 h.
263

Y. Zhang et al.
Applied Catalysis A, General 562 (2018) 258–266
Fig. 7. (A) Correlation between total medium plus very strong acidic density and the isosorbide yield; (B) Dependence of isosorbide yield and B/L ratio as a function
of Al contents.
observed still 100% up to the third run, the selectivity of isosorbide was
drastically decreased from 73% to 20%. The used catalyst became pale
yellow in color, however it recovered white and preserved most of in-
itial activity after a calcination treatment. It indicated that the coke
deposition was formed and the deactivation was probably involved in
the deposition of carbonaceous matters on the surface acid sites of
catalysts [17,19]. Therefore, a simple calcination treatment was con-
ducted to regenerate the activity of the used 6Al-SZ after each run as
shown in Fig. 9(B). Obviously, the activity of used 6Al-SZ was recovered
by calcination. Sorbitol conversion maintained at 100% over
regenerated catalyst, the selectivity of isosorbide gradually decreased to
ca.61% after fifth run, suggesting the reliable reusability of the catalyst.
Although the deposition of coke on the surface of catalyst was the
main reason of the deactivation, there was still innegligible decline of
isosorbide selectivity from 73% to 61% over regenerated 6Al-SZ. It was
likely resulted from sulfate loss caused by the water, which was an
intrinsic defect of sulfated metallic oxides [58,59]. Take into account
the synthesis of isosorbide from sorbitol involved the production of
water, we carried out the sulfate leaching tests to evaluate the 6Al-SZ
leaching in water. Approximately 0.5 g of sample were washed with
Fig. 8. Conversion and selectivity affected by different reaction parameters: (A) Catalyst dosage (175 °C, 2 h), (B) Reaction temperature (0.2 g catalysts, 2 h), (C)
Reaction time (175 °C, 0.2 g catalysts), (D) Comparison of isosorbide yields between 6Al-SZ and SZ-C catalysts. (2 h, 0.2 g catalyst).
264

Y. Zhang et al.
Applied Catalysis A, General 562 (2018) 258–266
Fig. 9. Reusability of 6Al-SZ (A) without calcination and (B) with calcination at 500 °C after each run. Reaction temperature 175 °C; catalyst dosage 0.2 g; time 2 h.
39% obtained with fresh conventional SZ-C catalyst.
4. Conclusions
The catalytic dehydration of sorbitol to isosorbide over mesoporous
aluminum-promoted sulfated zirconia was investigated. It was ascer-
tained that the isosorbide yield was greatly affected by structural and
acidic properties of the catalyst. Sulfated zirconia prepared by grind
method led to the formation of mesoporous structure with high surface
area and pore volume, which eliminated the diffusion restriction.
Meanwhile, the increased total acidic sites and Brönsted acidic sites in
SZ provided abundant accessible active sites for the reactants.
Furthermore, aluminum promoters changed the sulfur contents re-
tained in ZrO
Brönsted to Lewis acid sites. Under the optimized reaction conditions of
75 °C, 2 h and 0.2 g catalysts, the isosorbide yield of 73% over 6Al-SZ
2
so as to increase the strong acidic sites and ratio of
1
was attained. Importantly, this catalyst was easy to prepare/regenerate
and showed insignificant deactivation during five consecutive tests. The
satisfactory results over 6Al-SZ in present research can also provide
clues to develop high-performance heterogeneous catalysts for renew-
able biomass conversion to other high-valued chemicals.
Fig. 10. FT-IR spectra of (a) fresh 6Al-SZ; (b) regenerated 6Al-SZ after the fifth
run and (c) 6Al-SZ obtained after 10 h of leaching in water.
H
2
O (50 mL) for 10 h, then the suspensions were separated by cen-
trifugation and filtrate was checked with a BaCl solution for leached
2
Appendix A. Supplementary data
sulfate. The FT-IR spectra of the catalyst obtained after the sulfate
leaching test was recorded in Fig. 10 and showed the characteristic
2
−
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.apcata.2018.06.024.
peaks for SO
4
coordinated to the zirconium cation. However, the
solution confirmed
occurrence of the white precipitates with a BaCl
2
the existence of sulfate ions in the water. It verified that partial sulfates
were gradually leaching into water. Considering only trace of water
existed in this reaction system because of immediate removal of pro-
duced water by vacuum-pressure, there would be just trace of weakly
bonded sulfate species leaching, which was confirmed by FT-IR spectra
of the regenerated 6Al-SZ in Fig. 10 [60].
In order to further confirm the heterogeneity of 6Al-SZ, a hot
leaching test was conducted, and the results were displayed in Fig. S1.
The 6Al-SZ was removed from reaction mixture by a hot-filtration after
the reaction proceeded for 30 min, and then the filtrate was continued
to react under the same condition. It was found that the yield of iso-
sorbide increased until achieving its highest of 73% at 2 h, while it just
showed slight increase from 37.2% to 43.6% in the filtrate without
catalyst even prolonging the reaction time to 3 h. It suggested that 6Al-
SZ was a relatively stable catalyst, though a little homogeneous con-
tribution was introduced to reaction system by leaching.
References
[1] S.I. Avelino Corma, Alexandra Velty, Chem. Rev. 107 (2007) 2411–2502.
[2] J. Zhang, J.-b. Li, S.-B. Wu, Y. Liu, Ind. Eng. Chem. Res. 52 (2013) 11799–11815.
[3] M.M. Redfield, K.J. Anstrom, J.A. Levine, G.A. Koepp, B.A. Borlaug, H.H. Chen,
M.M. LeWinter, S.M. Joseph, S.J. Shah, M.J. Semigran, G.M. Felker, R.T. Cole,
G.R. Reeves, R.J. Tedford, W.H.W. Tang, S.E. McNulty, E.J. Velazquez, M.R. Shah,
E. Braunwald, J. Med. 373 (2015) 2314–2324.
[
4] B.A.J. Noordover, D. Haveman, R. Duchateau, R.A.T.M. van Benthem, C.E. Koning,
J. Appl. Polym. Sci. 121 (2011) 1450–1463.
[5] J.M. Sadler, F.R. Toulan, A.-P.T. Nguyen, R.V. Kayea, S. Ziaee, G.R. Palmese, J.J. La
Scala, Carbohydr. Polym. 100 (2014) 97–106.
[
[
6] Y.S. Eo, H.-W. Rhee, S. Shin, J. Ind. Eng. Chem. 37 (2016) 42–46.
7] J. Li, A. Spina, J.A. Moulijn, M. Makkee, Catal. Sci. Technol. 3 (2013) 1540–1546.
[8] W. Ni, D. Li, X. Zhao, W. Ma, K. Kong, Q. Gu, M. Chen, Z. Hou, Catal. Today (2018),
http://dx.doi.org/10.1016/j.cattod.2018.03.034 (Accessed 18 March 2018).
[
9] C. Dussenne, T. Delaunay, V. Wiatz, H. Wyart, I. Suisse, M. Sauthier, Green Chem.
9 (2017) 5332–5344.
[10] G. Fleche, M. Huchette, Starch—Starke 38 (1986) 26–30.
1
[11] L. Došen-Mićović, Ž. Čeković, J. Phys. Org. Chem. 11 (1998) 887–894.
[12] M. Gu, D. Yu, H. Zhang, P. Sun, H. Huang, Catal. Lett. 133 (2009) 214–220.
[13] D. Cao, B. Yu, S. Zhang, L. Cui, J. Zhang, W. Cai, Appl. Catal. A: Gen. 528 (2016)
59–66.
The above results revealed that both coke deposition and small
amounts of sulfate leaching were causes for the deactivation of the 6Al-
SZ. However, most of the catalytic activity for used catalysts could be
recovered by simple calcination. The value of isosorbide yield slightly
decreased to 61% after five cycles due to sulfur leaching from re-
generated 6Al-SZ, even this result was much better than the yield of
[
[
14] O.A. Rusu, W.F. Hoelderich, H. Wyart, M. Ibert, Appl. Catal. B: Environ. 176 (2015)
1
39–149.
15] Z.-C. Tang, D.-H. Yu, P. Sun, H. Li, H. Huang, Bull. Korean Chem. Soc. 31 (2010)
3679–3683.
265

Y. Zhang et al.
Applied Catalysis A, General 562 (2018) 258–266
[
[
[
16] P. Sun, D.H. Yu, Y. Hu, Z.C. Tang, J.J. Xia, H. Li, H. Huang, Korean J. Chem. Eng. 28
2010) 99–105.
17] J. Zou, D. Cao, W. Tao, S. Zhang, L. Cui, F. Zeng, W. Cai, RSC Adv. 6 (2016)
9528–49536.
18] H. Kobayashi, H. Yokoyama, B. Feng, A. Fukuoka, Green Chem. 17 (2015)
732–2735.
4065–4071.
(
[37] N. Kamoun, M.K. Younes, A. Ghorbel, A.S. Mamede, A. Rives, Ionics 21 (2014)
221–229.
4
[38] C. García-Sancho, J.A. Cecilia, A. Moreno-Ruiz, J.M. Mérida-Robles, J. Santamaría-
González, R. Moreno-Tost, P. Maireles-Torres, Appl. Catal. B: Environ. 179 (2015)
139–149.
2
[
[
19] R. Otomo, T. Yokoi, T. Tatsumi, Appl. Catal. A: Gen. 505 (2015) 28–35.
20] G. Morales, J. Iglesias, J.A. Melero, J. Moreno, R. Sánchez-Vázquez, Á. Peral,
A. Cubo, Top. Catal. 60 (2017) 1027–1039.
[39] M. Signoretto, F. Pinna, G. Strukul, P. Chies, G. Cerrato, S. Di Ciero, C. Morterra, J.
Catal. 167 (1997) 522–532.
[40] F. Babou, G. Coudurier, J.C. Vedrine, J. Catal. 152 (1995) 341–349.
[
[
[
[
21] J. Zhang, L. Wang, F. Liu, X. Meng, J. Mao, F.-S. Xiao, Catal. Today 242 (2015)
[41] T. Yamaguchi, Appl. Catal. 61 (1990) 1–25.
2
49–254.
[42] K. Föttinger, K. Zorn, H. Vinek, Appl. Catal. A: Gen. 284 (2005) 69–75.
[43] G.D. Yadav, J.J. Nair, Microporous Mesoporous Mater. 33 (1999) 1–48.
[44] J.S. Lee, J.W. Yoon, S.B. Halligudi, J.-S. Chang, S.H. Jhung, Appl. Catal. A: Gen. 366
(2009) 299–303.
[45] X. Yang, R.E. Jentoft, F.C. Jentoft, Catal. Lett. 106 (2006) 195–203.
[46] C.-C. Hwang, C.-Y. Mou, J. Phys. Chem. C 113 (2009) 5212–5221.
[47] W. Hua, Y. Xia, Y. Yue, Z. Gao, J. Catal. 196 (2000) 104–114.
[48] W. Stichert, F. Schüth, Chem. Mater. 10 (1998) 2020–2026.
[49] P. Canton, R. Olindo, F. Pinna, G. Strukul, P. Riello, M. Meneghetti, G. Cerrato,
C. Morterra, A. Benedetti, Chem. Mater. 13 (2001) 1634–1641.
[50] R. Beerthuis, L. Huang, N.R. Shiju, G. Rothenberg, W. Shen, H. Xu, ChemCatChem
10 (2018) 211–221.
22] M.J. Ginés-Molina, R. Moreno-Tost, J. Santamaría-González, P. Maireles-Torres,
Appl. Catal. A: Gen. 537 (2017) 66–73.
23] A.A. Dabbawala, J.J. Park, A.H. Valekar, D.K. Mishra, J.-S. Hwang, Catal. Commun.
6
9 (2015) 207–211.
24] A. Cubo, J. Iglesias, G. Morales, J.A. Melero, J. Moreno, R. Sánchez-Vázquez, Appl.
Catal. A: Gen. 531 (2017) 151–160.
[25] B.M. Reddy, M.K. Patil, Chem. Rev. 109 (2009) 2185–2208.
[26] M.S. Tiwari, A.B. Gawade, G.D. Yadav, Green Chem. 19 (2017) 963–976.
[27] J. Xia, D. Yu, Y. Hu, B. Zou, P. Sun, H. Li, H. Huang, Catal. Commun. 12 (2011)
5
44–547.
[
[
28] A.A. Dabbawala, D.K. Mishra, J.-S. Hwang, Catal. Commun. 42 (2013) 1–5.
29] I. Ahmed, N.A. Khan, D.K. Mishra, J.S. Lee, J.-S. Hwang, S.H. Jhung, Chem. Eng.
Sci. 93 (2013) 91–95.
[51] A.A. Dabbawala, D.K. Mishra, G.W. Huber, J.-S. Hwang, Appl. Catal. A: Gen. 492
(2015) 252–261.
[
30] N.A. Khan, D.K. Mishra, I. Ahmed, J.W. Yoon, J.-S. Hwang, S.H. Jhung, Appl. Catal.
[52] J.-H. Wang, C.-Y. Mou, Appl. Catal. A: Gen. 286 (2005) 128–136.
[53] W. Ciptonugroho, M.G. Al-Shaal, J.B. Mensah, R. Palkovits, J. Catal. 340 (2016)
17–29.
A: Gen. 452 (2013) 34–38.
[
[
[
31] F. Su, Y. Guo, Green Chem. 16 (2014) 2934–2957.
32] L. Wang, F.-S. Xiao, Green Chem. 17 (2015) 24–39.
33] Y. Sun, S. Ma, Y. Du, L. Yuan, S. Wang, J. Yang, F. Deng, F.-S. Xiao, J. Phys. Chem. B
[54] C.A. Emeis, J. Catal. 141 (1993) 347–354.
[55] M. He, M. Wang, G. Tang, Y. Fang, T. Tan, Appl. Catal. A: Gen. 550 (2018) 160–167.
[56] A. Yamaguchi, N. Hiyoshi, O. Sato, M. Shirai, Green Chem. 13 (2011) 873–881.
[57] G. Flèche, M. Huchette, Starch—Stärke 38 (1986) 26–30.
[58] A. Corma, Curr. Opin. Solid Stae Mater. 2 (1997) 63–75.
[59] W. Hua, A. Goeppert, J. Sommer, Appl. Catal. A: Gen. 219 (2001) 201–207.
[60] A. Kiss, A. Dimian, G. Rothenberg, Adv. Synth. Catal. 348 (2006) 75–81.
1
09 (2005) 2567–2572.
[
[
[
34] I. Jiménez-Morales, J. Santamaría-González, P. Maireles-Torres, A. Jiménez-López,
Appl. Catal. B: Environ. 103 (2011) 91–98.
35] Z. Miao, J. Zhou, J. Zhao, D. Liu, X. Bi, L. Chou, S. Zhuo, Appl. Surf. Sci. 411 (2017)
4
19–430.
36] S. Tominaka, T. Momma, B. Scrosati, T. Osaka, J. Power Sources 195 (2010)
266