Solvent-Free Production of Isosorbide from Sorbitol Catalyzed by a Polymeric Solid Acid

Article abstract of DOI:10.1002/cssc.201901922

A series of polymeric solid acid catalysts (PDSF-x) is prepared by grafting strong electron-withdrawing groups (?SO₂CF₃) on a sulfonic acid-modified polydivinylbenzene (PDS) precursor synthesized hydrothermally. The effect of acid strength on sorbitol dehydration is investigated. The textural properties, acidity, and hydrophobicity are characterized by using Brunauer–Emmett–Teller analysis, elemental analysis, and contact angle tests. The results of FTIR spectroscopy and X-ray photoelectron spectroscopy show that both ?SO₃H and ?SO₂CF₃ are grafted onto the polymer network. We used solid-state ³¹P NMR spectroscopy to show that the acid strength of PDSF-x is enhanced significantly compared with that of PDS, especially for PDSF-0.05. As a result, PDSF-0.05 exhibits the highest isosorbide yield up to 80 %, a good turnover frequency of 231.5 h^?1 (compared to other catalysts), and excellent cyclic stability, which is attributed to its large specific surface area, appropriate acid strength, hydrophobicity, and stable framework structure. In addition, a plausible reaction pathway and kinetic analysis are proposed.

Full text of DOI:10.1002/cssc.201901922

DOI: 10.1002/cssc.201901922
Full Papers
Solvent-Free Production of Isosorbide from Sorbitol
Catalyzed by a Polymeric Solid Acid
Danping Yuan,[a, b] Lei Li,[a] Feng Li,[a] Yanxia Wang,[a, b] Feng Wang,[a] Ning Zhao,*[a] and
Fukui Xiao*[a]
A series of polymeric solid acid catalysts (PDSF-x) is prepared
by grafting strong electron-withdrawing groups (ꢀSO2CF3) on a
sulfonic acid-modified polydivinylbenzene (PDS) precursor syn-
thesized hydrothermally. The effect of acid strength on sorbitol
dehydration is investigated. The textural properties, acidity,
and hydrophobicity are characterized by using Brunauer–
Emmett–Teller analysis, elemental analysis, and contact angle
tests. The results of FTIR spectroscopy and X-ray photoelectron
spectroscopy show that both ꢀSO3H and ꢀSO2CF3 are grafted
onto the polymer network. We used solid-state 31P NMR spec-
troscopy to show that the acid strength of PDSF-x is enhanced
significantly compared with that of PDS, especially for PDSF-
0.05. As a result, PDSF-0.05 exhibits the highest isosorbide
yield up to 80%, a good turnover frequency of 231.5 hꢀ1 (com-
pared to other catalysts), and excellent cyclic stability, which is
attributed to its large specific surface area, appropriate acid
strength, hydrophobicity, and stable framework structure. In
addition, a plausible reaction pathway and kinetic analysis are
proposed.
Introduction
Biomass resources, as an abundant renewable carbon resource
in nature, are considered as a promising alternative to non-re-
newable fossil fuels in the production of value-added chemical
products and fuels.[1] Sorbitol, a “top 10” valued-added bio-
based building block, is obtained from cellulose by the hydro-
genation of glucose.[2] Since the 1940s,[3] the use of sorbitol as
a feedstock in biorefinery has attracted much attention. Isosor-
bide, the doubly dehydrated product of d-sorbitol, and its
derivatives have been used widely in the fields of medicine,[4]
food, cosmetics,[5] and polymers.[6]
the key to the production of isosorbide is a suitable acid cata-
lyst to promote the complete conversion and a high selectivity
to sorbitol.
Homo- and heterogeneous catalysts have been used in the
sorbitol dehydration reaction. Traditional homogeneous cata-
lysts, such as mineral acids (H2SO4, HCl, H3PO4, etc.),[10] Lewis
acids,[10b,11] and ionic liquids,[12] favor isosorbide formation.
However, corrosion, environmental harm, and poor reusability
limit their industrial applications. Consequently, the develop-
ment of an economic and environmentally friendly solid cata-
lyst is critical to obtain a high yield of isosorbide. To date,
different heterogeneous catalysts, such as acid resins,[13] zeo-
lites,[14] metal phosphates,[15] supported metal oxides,[16] sulfat-
ed and phosphate metal oxides,[17] supported heteropolya-
cids,[18] and carbon-based acids,[19] have been employed for the
sorbitol dehydration reaction. Generally, the catalytic activity of
solid acid catalysts is worse than that of homogeneous
acids.[20] An important reason for this is that the acid sites on
the solid acid catalyst cannot generate cooperative effects as
in homogeneous catalysts.[21] Therefore, the design of efficient
sorbitol-dehydration catalysts, such as the regulation of acid
sites, acid strength, thermal stability, and hydrophilic/hydro-
phobic properties, still poses many challenges.
For the sorbitol dehydration reaction, a mechanism accepted
widely is the intramolecular SN2 substitution reaction.[7] For ex-
ample, isosorbide is produced by the formation of 1,4-sorbitan
(or 3,6-sorbitan) followed by dehydration. The second step re-
quires a higher reaction temperature and a longer reaction
time than the first step under the same conditions.[7,8] How-
ever, as a result of the effects of chemo- and regioselectivity,
byproducts such as isomers and humins are likely to be
formed during the double dehydration reaction.[9] Therefore,
[a] Dr. D. Yuan, Prof. L. Li, Prof. F. Li, Dr. Y. Wang, Prof. F. Wang, Prof. N. Zhao,
Prof. F. Xiao
State Key Laboratory of Coal Conversion
Institute of Coal Chemistry
In this sense, sulfonate resins have been evaluated as solid
acid catalysts for sorbitol dehydration. Goodwin et al. studied
sorbitol dehydration over Amberlite IR-120 type resin, over
which an isosorbide yield of 39% was obtained.[22] Bock et al.
found that the isosorbide yield reached 57% under harsh reac-
tion conditions (1708C, 2 h, 10 mmHg) over the same cata-
lyst.[8] Since then, resin-type solid acid catalysts have gained
considerable attention because of their large specific surface
area, controllable acidity, and hydrophobicity. Commercial ion-
Chinese Academy of Sciences
Taiyuan 030001 (P.R. China)
E-mail: zhaoning@sxicc.ac.cn
xiaofk@sxicc.ac.cn
[b] Dr. D. Yuan, Dr. Y. Wang
University of Chinese Academy of Sciences
Beijing 100049 (P.R. China)
Supporting information and the ORCID identification number(s) for the
author(s) of this article can be found under:
https://doi.org/10.1002/cssc.201901922.
ChemSusChem 2019, 12, 1 – 11
1
ꢀ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
exchange resins such as Amberlyst, Purlite, and Nafion have
been used.[13c,d,h] Although some commercial ion-exchange
resins are able to achieve a good sorbitol yield (>70%), their
thermal stability needed to be improved. The SA-SiO2-x cata-
lysts synthesized by Shi et al. were able to achieve an isosor-
bide yield of 84% under mild reaction conditions.[13g] Despite
the good performance of propylsulfonic acid functionalized
SBA-15, the stability of the catalyst was poor.[13e] Zhang et al.
found that hydrophobic PꢀSO3H can catalyze the dehydration
efficiently, and 87.9% isosorbide yield was obtained.[13b]
Previously, we found that the polymer network and acidity of
sulfonic acid-modified polydivinylbenzene (PDS) solid acid cat-
alysts have an important effect on the dehydration reaction of
sorbitol, and a yield of 81.7% can be obtained.[13a]
obtain the PDS catalyst. Finally, the ꢀSO2CF3 group was grafted
onto the PDS catalyst by acidification with CF3SO3H.
Characterization
N2 adsorption–desorption isotherms and pore size distribution
curves of all samples are presented in Figure 1, and the textur-
al information is summarized in Table 1. A type-IV isotherm
with an H4 hysteresis loop at a high relative pressure confirms
the presence of mesopores.[24] The precursor (PDVB) has large
specific surface area (1008 m2 gꢀ1) and pore volume
(1.88 cm3 gꢀ1) and an average pore diameter (9.14 nm; Table 1),
which favor the grafting of acid groups. If ꢀSO3H and ꢀSO2CF3
groups were grafted alone, the specific surface area of the cat-
alysts decreased slightly (Table 1, Entries 2 and 3). However, as
the grafting amount of ꢀSO2CF3 increased, the specific surface
area, pore volume, and pore diameter of the catalysts de-
creased gradually (Table 1, Entries 3–8) as a result of the graft-
ing of acid groups on the inner and outer surfaces of the cata-
lysts, which is consistent with our previous work.[13a] The de-
crease in the specific surface area of PDSF-x samples may be
attributed to the acid groups that occupy the surface of the
polymer network that cannot contribute to the specific surface
area.[25] In addition, the grafted acid groups may block the
pores and cause a decrease in the pore size and pore volume
in accordance with the results of Liu et al.[26]
Dabbawala et al. found that Brønsted acids can catalyze the
dehydration of sorbitol efficiently and that the acid strength
has a significant effect on the reaction.[10a] Recently, Cubo et al.
claimed that although the increased acid strength limited the
first step to some extent, the second step could be accelerated
effectively.[13e] In addition, the hydrophobicity of the catalyst
surface also has a crucial effect on sorbitol dehydration, that is,
the hydrophobic surface is able to remove the water produced
by the reaction from the surface rapidly, which prevents the
coverage of the acid sites with water and leads to decreased
activity.[13a,b,14a,b,23] According to previous reports and our re-
search, the acid strength of the catalyst plays an important
role to improve the sorbitol conversion and the selective trans-
formation to isosorbide. However, there is no systematic study
on the influence of the acid strength of solid acid catalysts on
sorbitol conversion.
The elemental analysis and acid–base titration results for all
of the catalysts are summarized in Table 1. Compared with that
of PDS (0.3) (1.38 mmolgꢀ1), the S content increased gradually
upon the grafting of ꢀSO2CF3 groups (1.37–3.73 mmolgꢀ1). In-
In this study, a series of heterogeneous acid cata-
lysts (PDSF-x, polydivinylbenzene solid acid catalyst
modified by sulfonic acid group and trifluorometha-
nesulfonic acid group) with different acid strengths
was prepared by grafting different amounts of
strongly electron-withdrawing trifluoromethanesul-
fonic acid groups (ꢀSO2CF3). The textural properties,
morphologies, acid properties, coordination environ-
ment, and thermal stability of the samples were stud-
ied. The acid sites that are actually exposed to the re-
actants play a crucial role in the catalytic effect of the
sorbitol dehydration reaction. The relationship be-
tween the properties and the catalytic performance
of the catalysts was discussed. The results showed
that a sorbitol conversion of 100% and isosorbide
yield of 80% can be obtained at 1408C without sol-
vent.
Results and Discussion
The preparation of SO2CF3-modified polydivinylben-
zene solid acid catalyst (PDVBF), PDS, and PDSF-x is
shown in Scheme S1. The PDSF-x catalysts were syn-
thesized in three steps. Firstly, the precursor PDSNa
(divinylbenzene and sodium p-styrene sulfonate poly-
mer) was synthesized. Then, the dried white solid
Figure 1. N2 adsorption–desorption isotherms (left) and pore size distribution curves
was ion-exchanged in 1m H2SO4 solution for 24 h to (right) of the catalysts.
&
ChemSusChem 2019, 12, 1 – 11
www.chemsuschem.org
2
ꢀ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Table 1. Textural properties, elemental analysis, and acid–base titration results.
[a]
[b]
[c]
[d]
Entry
Catalyst
S content [mmolgꢀ1
]
Acid exchange capacity [mmolgꢀ1
]
SBET [m2 gꢀ1
]
Vp [cm3 gꢀ1
]
Dp [nm][e]
1
2
3
4
5
6
7
8
9
PDVB
PDS(0.3)
0.00
1.38
0.29
1.37
1.75
1.97
2.34
3.73
1.52
0.00
1.53
0.25
1.43
1.12
1.01
1.03
1.17
1.08
1008
715
806
658
485
323
247
243
427
1.88
1.52
1.74
1.16
0.63
0.30
0.25
0.27
0.49
9.14
9.49
9.61
8.02
6.08
4.30
4.93
5.46
5.24
PDVBF-0.05
PDSF-0.01
PDSF-0.05
PDSF-0.1
PDSF-0.15
PDSF-0.2
PDSF-0.05[f]
[a] Measured by using elemental analysis. [b] Measured by using acid–base titration. [c] BET surface area. [d] Total pore volume. [e] Average pore size. [f] Re-
cycled five times.
terestingly, the total S content of PDS (0.3) (1.38 mmolgꢀ1) and
PDVBF-0.05 (0.29 mmolgꢀ1) was approximately equal to that of
PDSF-0.05 (1.75 mmolgꢀ1), which indicates that both acid
groups (ꢀSO3H and ꢀSO2CF3) were grafted successfully onto
the catalysts. However, according to the results of acid–base ti-
tration, as the grafting amount of the ꢀSO2CF3 group increases,
the amount of accessible H+ sites of the catalyst sample was
somehow lower than the S content obtained by elemental
analysis. This phenomenon is probably caused by the poor
swelling ability of the resin particles in the reaction system,[27]
and Zhang et al. obtained similar results.[21]
its larger specific surface area (715 vs. 485 m2 gꢀ1). These results
are consistent with those of the water droplet contact angles.
FTIR spectra of all samples are shown in Figure S5. The band
located at n˜ =1030–1040 cmꢀ1 is associated with the CꢀS
stretch on the benzene rings. Characteristic absorptions of ꢀ
SO3H at n˜ =1008, 1125, and 1178 cmꢀ1 are caused by the sym-
metrical and asymmetrical stretching vibrations of O=S=O.[28]
In addition, for the PDSF-x catalysts, the absorption of CꢀF ap-
peared at n˜ =1290 cmꢀ1, the intensity of which increased grad-
ually with the increase of the amount of grafted ꢀSO2CF3.[29]
These results suggest that ꢀSO3H and ꢀSO2CF3 are grafted on
the PDVB, which is in accordance with the results obtained by
using elemental analysis.
Powder XRD patterns of all catalysts are shown in Figure S1.
All samples showed broad Bragg reflection peaks at approxi-
mately 2q=15–258 attributed to the amorphous structure of
the catalysts. The overall structure of the catalysts did not
change significantly before and after the grafting of ꢀSO2CF3.
The morphologies of PDVB, PDS (0.3), and PDSF-0.05 were
examined by using SEM and TEM (Figure S2). The polymer
matrix of PDVB shows an irregular coralline villus structure
(Figure S2a), whereas PDS (0.3) grafted with ꢀSO3H has a loose
spongy structure (Figure S2b). Notably, the fluffy structure of
the catalyst disappeared after the grafting of ꢀSO2CF3 (Fig-
ure S2c) to form a rough surface composed of many small par-
ticles. According to the TEM images (Figure S2d–f), the cata-
lysts have an abundant wormhole-like mesoporous structure
(5–50 nm) after the grafting of acid groups. Compared with
that in the TEM image of PDS (0.3), the black portion of PDSF-
0.05 increased, which suggests that ꢀSO2CF3 is grafted success-
fully into the pores.
X-ray photoelectron spectra (XPS) of PDS (0.3), PDVBF-0.05,
and PDSF-0.05 are presented in Figure 2. The spectra of all
samples show characteristic peaks of C, S, and O elements (Fig-
ure 2a). Peaks with binding energies (BE) of approximately 533
and 285 eV are attributed to C1s and O1s. In addition, a new
peak of F1s (ca. BE=690 eV) appeared in the spectra of
PDVBF-0.05 and PDSF-0.05 (Figure 2a–c). Notably, the S2p
peak in the spectra of PDS (0.3) and PDVBF-0.05 appears at
BE=164.3 and 169.1 eV, respectively, whereas that of PDSF-
0.05 appears at BE=168.8 eV (Figure 2b). This might be be-
cause of the introduction of the strong electron-withdrawing
group ꢀSO2CF3, which enhances the acid strength of the cata-
lysts.[30] This conclusion is consistent with the results obtained
by using EPR spectroscopy (Figure S6). Peaks of CꢀC (BE=
284.7 eV), CꢀS (BE=286.2 eV), and CꢀF (BE=291.4 eV) are ob-
served in the C1s high-resolution XPS spectrum of PDSF-0.05
(Figure 2c), which indicate that the ꢀSO2CF3 group was grafted
onto PDS (0.3).[29]
Water droplet contact angles of the catalyst are presented in
Figure S3. The hydrophobic angles of PDVB, PDS (0.3), and
PDSF-0.05 are 153, 147, and 1408, respectively (Figure S3a–c).
However, if the grafting amount of ꢀSO2CF3 was increased (x>
0.05), the catalysts lost their hydrophobic properties. This may
be because excessive super acid (CF3SO3H) destroys the surface
structure of the catalysts, which is consistent with the results
obtained by using SEM.
According to previous reports, the introduction of F increas-
es the acid strength and thermal stability of Nafion.[31] To gain
a deep structural insight into the samples, 31P magic-angle
spinning (MAS) NMR spectra of the probe trimethylphosphine
oxide (TMPO) adsorbed on PDS (0.3), PDVBF-0.05, and PDSF-x
were collected (Figure 3). This method can be used to identify
multiple acid sites on the catalyst efficiently according to the
strength of the interaction between TMPO and a Brønsted
Adsorption isotherms of water on PDVB, PDS (0.3), PDVBF-
0.05, PDSF-0.05, and Amberlyst-15 are shown in Figure S4. No-
tably, all synthesized catalysts exhibit a much lower water ab-
sorption capacity than Amberlyst-15. The water absorption of
PDS (0.3) is larger than that of PDSF-0.05, probably because of
31P
acid. The relationship between proton affinity (PA) and d is
31P
d
=182.866ꢀ0.3902*PA, and a PA value of 250 kcalmolꢀ1
(85.3 ppm) is the threshold for superacidity.[32] Therefore, the
ChemSusChem 2019, 12, 1 – 11
www.chemsuschem.org
3
ꢀ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
Figure 2. a) Survey, b) S2p, and c) C1s XPS spectra of PDS (0.3), DVBF-0.05, and PDSF-0.05.
(PDSF-0.01, 32%), 82 (PDSF-0.05, 83%), 83 (PDSF-0.1, 22%),
and 81.5 ppm (PDSF-0.2, 12%), close to the threshold of super-
31P
acidity (d =85.3 ppm), which indicates that the interaction
between F and S species has a crucial impact on the acidity.[33]
If we compare PDSF-0.05 and PDS (0.3), the ratio of strong acid
sites increased significantly after modification with ꢀSO2CF3
(from 60 to 83%). However, upon the introduction of excess ꢀ
SO2CF3 (Figure 3e and f), the spectra of the samples exhibited
a broad overlapped 31P resonance peak that spanned from ap-
proximately 54 to 85 ppm, and the proportion of superacidity
was reduced significantly because of the inhomogeneous dis-
tribution of the acid sites.[34] Combined with the results ob-
tained by using acid–base titration and N2 adsorption–desorp-
tion (Table 1), we observe that the acidic ion-exchange capaci-
ty, specific surface area, and pore volume of PDSF-0.1 and
PDSF-0.2 all decreased compared to those of PDS (0.3). This
may be because excess ꢀSO2CF3 results in pore blockage and
the acid sites are not accessible to the probe molecules.
Thermogravimetric analysis results of various catalysts are
shown in Figure 4. There are three weight-loss steps that can
be attributed to the release of water (<2508C), the decompo-
sition of functional groups (290–4108C), and the decomposi-
tion of the polymer matrix (410–6008C).[26] Clearly, PDSF-0.05
has a better thermal stability than the commercial resins Am-
berlyst-15 and Nafion (one of the most stable commercial
resins). Moreover, from the differential thermogravimetric anal-
ysis (DTG) results, the decomposition temperature of PDSF-
0.05 (5208C) was higher than that of PDVB (4758C) and
PDS (0.3) (5008C). This may be because of the introduction of
a strong electron-withdrawing group (ꢀSO2CF3) that forms hy-
drogen bonds between H and F to stabilize the polymer tex-
ture.[35] Liu et al.[30] and Sun et al.[33] drew similar conclusions.
Figure 3. 31P MAS NMR spectra of TMPO adsorbed on a) PDS (0.3), b) PDVBF-
0.05, c) PDSF-0.01, d) PDSF-0.05, e) PDSF-0.1, and f) PDSF-0.2.
lower the PA value, the stronger the acid strength. Clearly, the
amount of grafted ꢀSO2CF3 has a significant effect on the acid
properties. If the ꢀSO2CF3 group was grafted (PDVBF-0.05; Fig-
ure 3b), only weak acid sites were found. The 31P chemical
shift for the PDS (0.3) catalyst was 83.5 (60%) and 79 ppm
(40%; Figure 3a). PDSF-x catalysts functionalized with different
Catalytic performance in the dehydration of sorbitol to
isosorbide
The catalytic activity of the PDSF-x catalysts was evaluated in
the conversion of sorbitol to isosorbide (Table 2). As water will
have a negative effect on solid acid catalysts,[36] vacuum condi-
tions were used to remove the generated water quickly.
31P
amounts of CF3SO3H revealed strong acid sites with d =79
&
ChemSusChem 2019, 12, 1 – 11
www.chemsuschem.org
4
ꢀ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
alytic performance is achieved over PDSF-0.05 as a
result of its large specific surface area, good hydro-
phobicity, and uniform acid distribution. Meanwhile,
despite the higher amount of F species on PDSF-0.15
and PDSF-0.2, their catalytic performance is poor,
which may result from their low specific surface area,
small pore size, and low hydrophobicity. Our results
show that an appropriate pore structure and surface
properties are beneficial for the free access of the re-
actants to the active sites on the inner and outer sur-
faces. This conclusion is consistent with those report-
ed previously.[13a,30]
Then, the reaction conditions for the PDSF-0.05
catalyst were optimized. A high amount of catalyst
(3 wt%) leads to fast and complete sorbitol conver-
sion (Figure 5b). However, the subsequent increase
in byproducts and humins leads to a low yield of iso-
sorbitol. Finally, 2 wt% was selected as the optimum
amount of catalyst.
Next, the influence of the temperature on the reac-
tion of sorbitol was explored (Figure 5c). The initial
slope of the time–yield curve increases gradually as
the reaction temperature increases, which indicates
that temperature has an important influence on the
reaction rate. A low temperature is not conducive to
the rapid production of isosorbide, for example, if
the temperature is between 110 and 1308C, the yield
Figure 4. a) Thermogravimetric and b) DTG profiles of Nafion, Amberlyst-15, PDSF-0.05,
PDS (0.3), and PDVB.
of isosorbide increases very slowly. Additionally, a high reaction
temperature favors the polymerization reaction. Therefore,
1408C was chosen as the optimum evaluation temperature.
The time–conversion/selectivity plot for sorbitol dehydration
to isosorbide over PDSF-0.05 is shown in Figure 5d. Basically,
the reaction equilibrium of sorbitol dehydration is reached at a
reaction time of 8 h. The conversion of sorbitol was 100%, and
the selectivity of isosorbide was 80%.
Table 2. Performance of various catalysts in sorbitol dehydration to iso-
sorbide.[a]
Entry Catalyst
Conversion[b] [%] Product yield[b] [%] TOF[c] [hꢀ1
]
Y1
Y2
Y3
[d]
[d]
1
2
3
4
5
6
7
8
Blank
<1
95.3
96.1
100
100
100
100
100
100
–
–
<1
3.1 396.4
–
Catalytic performance of PDSF-x
PDVBF-0.05
PDS (0.3)
PDSF-0.01
PDSF-0.05
PDSF-0.1
PDSF-0.15
PDSF-0.2
H2SO4
68.3 23.9
25.7 57.8
16.0 74.5
10.1 80.0
17.1 73.9
26.9 61.4
19.7 64.3
20.5 69.7
41.3 40.8
64.1 28.9
In the absence of the catalyst (Table 2, Entry 1), isosorbide was
not detected after 8 h. With PDVBF-0.05 (Table 2, Entry 2), al-
though the conversion was up to 95.3%, the yield of isosor-
bide was still as low as 23.9%, similar to that of PDS (0.3) (con-
version 96.1%, yield 57.8%; Table 2, Entry 3). The two commer-
cial resins, Amberlyst-15 and Nafion (Table 2, Entries 10 and
11), exhibited a very poor selectivity for the target product. In
contrast, the PDSF-x catalysts show a good catalytic per-
formance (Table 2, Entries 4–8), and with an increase of the
amount of ꢀSO2CF3 added, the isosorbide yield increased first
and then decreased to reach a maximum at x=0.05. The main
products were isosorbide and 1,4-dehydrated sorbitol, trace
1,5-dehydrated sorbitol, and 2,5-mannitan. The undetermined
products are defined as “Others”. Among the catalysts, PDSF-
0.05 has the highest turnover frequency (TOF) of 231.5 hꢀ1,
higher than that of H2SO4 (219.4 hꢀ1), Amberlyst-15 (22.2 hꢀ1),
and Nafion (39.2 hꢀ1). Additionally, compared with various re-
ported catalysts, PDSF-0.05 gives a better catalytic activity and
higher TOF value (detailed comparisons are listed in Table S1).
12.6
80.2
9.5 137.6
9.9 231.5
9.0 170.1
11.7 206.3
16.0 230.6
9.8 219.4
9
10
11
Nafion
Amberlyst-15 93.8
84.2
2.1
0.8
39.2
22.2
[a] Reaction conditions: sorbitol 25 g, catalyst 2 wt%, 1408C, 0.3 bar, 8 h.
[b] Determined by using HPLC with an Aminex HPX-87H column at 608C
(0.6 mLminꢀ1, degassed 0.005m H2SO4 solution as eluent). [c] The TOF
value is calculated from the results of acid–base titration. [d] Undetecta-
ble.
Optimization of the reaction conditions
The reaction conditions were optimized, and the results are
listed in Figure 5. The yield of isosorbide increases gradually
with an increase of the reaction time (Figure 5a). The best cat-
ChemSusChem 2019, 12, 1 – 11
www.chemsuschem.org
5
ꢀ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
Figure 5. a) Time–yield plot for sorbitol dehydration to isosorbide over PSDF-x. Reaction conditions: sorbitol (25 g), catalyst (0.5 g, 2 wt%), 1408C, 0.3 bar,
1200 rpm. b) Time–conversion/yield plot for sorbitol dehydration to isosorbide over different amounts of PSDF-0.05. Reaction conditions: sorbitol (25 g), cata-
lyst (1, 2, and 3 wt%), 1408C, 0.3 bar, 1200 rpm. c) Time–yield plot for sorbitol dehydration to isosorbide over 2 wt% PSDF-0.05 at different temperatures. Re-
action conditions: sorbitol (25 g), catalyst (0.5 g, 2 wt%), 110–1608C, 0.3 bar, 1200 rpm. d) Time–conversion/selectivity plot for sorbitol dehydration to isosor-
bide over 2 wt% PSDF-0.05. Reaction conditions: sorbitol (25 g), catalyst (0.5 g, 2 wt%), 1408C, 0.3 bar, 1200 rpm.
Compared to PDSF-0.05, less F species grafting (PDSF-0.01) led
to weakened acidity, which may be the reason for poor catalyt-
ic activity. Notably, for PDSF-0.05, the amount of humins (9–
16%) was significantly lower than that over PDSF-0.1/0.15/0.2,
which may be because of the hydrophobic microenvironment
provided by the polymer substrate. Moreover, the large specif-
ic surface area, pore volume, and pore diameter facilitate the
transport of reactants and products. Additionally, excessive
grafting of strong electron-absorbing groups may lead to a
large amount of byproducts, therefore, the acid strength
should be in an appropriate range.
No significant deactivation was found even after five cycles
(from 80 to 75%) under the investigated conditions, which in-
dicates that PDSF-0.05 is stable for the reaction (Figure 6).
Elemental analysis, N2 adsorption–desorption, and FTIR spec-
troscopy were employed to study the properties of the spent
catalyst. The values of the textural properties of PDSF-0.05 re-
cycled five times decreased slightly (485 vs. 427 m2 gꢀ1, 0.63 vs.
0.49 cm3 gꢀ1, 6.08 vs. 5.24 nm; Table 1, Entry 9 and Figure S7).
Catalyst reusability
The stability of PDSF-0.05 was investigated. The used PDSF-
0.05 catalyst was activated as follows. Briefly, the solid catalyst
was separated from the reaction mixture by filtration. Then,
the recovered catalyst was washed thoroughly with a large
amount of ethanol and deionized water to remove the ad-
sorbed substances on the surface followed by drying under
vacuum at 508C for 3 h. Thereafter, the catalyst was activated
with 0.1m H2SO4 for 4 h, washed with deionized water until
neutral, and dried at 508C for 3 h before the next use.
Figure 6. Reusability of PDSF-0.05 in sorbitol dehydration to isosorbide. Re-
action conditions: sorbitol (25 g), catalyst (2 wt%), 1408C, 0.3 bar, 8 h.
&
ChemSusChem 2019, 12, 1 – 11
www.chemsuschem.org
6
ꢀ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
The FTIR spectrum of the spent catalyst showed the same
characteristic peaks as that of the fresh one, but the absorp-
tion peak of CꢀS was weakened slightly, which is consistent
with the results of elemental analysis (Table 1, Entry 9). The
slow decrease in activity may be caused by the gradual loss of
the acid groups.
which indicates that the PDSF-0.05 catalyst could promote the
1,4-sorbitan dehydration reaction more efficiently. The lower
selectivity over PDSF-0.1/0.15/0.2 may result from the excessive
amount of grafted F species that led to the low specific surface
area and pore size of the catalysts, that is, reactant and prod-
uct transport were inhibited to result in an increase of the for-
mation of byproducts during the rapid conversion of sorbi-
tol.[13e] In addition, the poor catalytic activity of PDSF-0.01 may
be caused by the lower grafting amount of F species and its
low acid strength. Moreover, the hydrophobicity of the catalyst
can influence the dehydration reaction of sorbitol (a polar sub-
stance) because of the weak affinity between the reactant and
the catalyst surface, that is, the low selectivity of 1,4-sorbitan
over PDSF-0.01 (Figure 7a) may also be affected by the hydro-
phobicity.
Possible reaction mechanism for the conversion of sorbitol
over PDSF-x
To understand the reaction pathway of sorbitol dehydration
over the PDSF-x catalysts, the selectivity to sorbitol was stud-
ied (Figure 7). For all catalysts, 1,4-sorbitan is formed as the
main product, the selectivity of which reached 95% at low
conversion (<20%; Figure 7a). With the increase of conversion,
the selectivity to 1,4-sorbitan decreases gradually, and the se-
lectivity to isosorbide increases. At a high conversion (>90%;
Figure 7b), isosorbide becomes the main product. These re-
sults indicate that isosorbide formation is divided mainly into
two steps: the formation of 1,4-sorbitan (step I) followed by
further dehydration to obtain isosorbide (step II). The selectivi-
ty to isosorbide over PDSF-0.05 was higher than that over the
other catalysts at the same conversion (>90%; Figure 7b),
To compare the relative kinetics of the sorbitol dehydration
reaction of three catalysts (PDSF-0.05, PDS(0.3), and H2SO4), the
reaction was performed at different temperatures (130–1508C)
with the same amount of catalyst active sites. All the calcula-
tions and data (k1, k2, kall, and Ea) are presented in the Support-
ing Information. The rate constant of PDSF-0.05 (kall =0.78/
1.45/1.71 hꢀ1) is higher than that of PDS (0.3) (kall =0.54/0.58/
0.80 hꢀ1) and H2SO4 (kall =0.70/1.15/1.27 hꢀ1; Table S2). In addi-
tion, the order of the activation energy for the three catalysts
was all Ea2 >Ea1, which indicates that the second step of the
sorbitol dehydration reaction is more difficult. The activation
energy of PDSF-0.05 (Ea1 =72 kJmolꢀ1, Ea2 =87 kJmolꢀ1) is
higher than that of H2SO4 (Ea1 =64 kJmolꢀ1, Ea2 =66 kJmolꢀ1),
which indicates the advantages of homogeneous over hetero-
geneous catalysts. The activation energy of PDSF-0.05 (Ea1 =
72 kJmolꢀ1, Ea2 =87 kJmolꢀ1) is lower than that of PDS (0.3)
(Ea1 =80 kJmolꢀ1, Ea2 =89 kJmolꢀ1), which indicates that a high
acid strength is beneficial for the sorbitol dehydration reaction.
Conclusion
A series of polydivinylbenzene solid acid catalysts modified by
sulfonic acid and trifluoromethanesulfonic acid groups (PDSF-
x) catalysts with different acid strengths were synthesized and
showed a good catalytic performance in sorbitol dehydration
to isosorbide. A complete conversion of sorbitol, 80% isosor-
bide yield, and high turnover frequency (231.5 hꢀ1) were ob-
tained under mild reaction conditions (1408C without solvent)
over PDSF-0.05. Meanwhile, the recyclability of PDSF-0.05 is ex-
cellent (75% isosorbitol yield after five recycles). Moreover, an
excessive grafting of F species would lead to the blockage of
the catalyst channels and decrease the amount of exposed
active sites. Moreover, the catalytic performance is related
closely to the amount of acid sites accessible to the reactants.
Therefore, an appropriate grafting amount of F species is nec-
essary. In addition, the sorbitol dehydration reaction was con-
firmed to occur in two key steps (sorbitol!1,4-sorbitan!iso-
sorbide). Kinetic analysis showed that an increase in acid
strength accelerated the rate of sorbitol dehydration. The as-
prepared PDSF-0.05 catalyst has the potential to substitute tra-
ditional mineral acids to catalyze the dehydration of sorbitol to
isosorbide.
Figure 7. Selectivity to a) 1,4-sorbitan and b) isosorbide versus sorbitol con-
version over various catalysts. Reaction conditions: sorbitol (25 g), catalyst
(0.5 g, 2 wt%), 1408C, 6 h, 0.3 bar, 1200 rpm.
ChemSusChem 2019, 12, 1 – 11
www.chemsuschem.org
7
ꢀ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
distribution were calculated by using the Brunauer–Emmett–Teller
(BET) and Barrett–Joyner–Halenda (BJH) methods, respectively.
Before adsorption, all samples were outgassed for 10 h at 1508C
under vacuum.
Experimental Section
Materials and methods
Materials
The amount of grafted acid groups on the polymer network was
assessed by calculation of the S content by using a Vario EL CUBE
elemental analyzer (CHNS). For all catalysts, S only exists in the ꢀ
SO3H and ꢀSO2CF3 groups, so the S content represents the content
of acid groups.
Unless otherwise noted, all commercial chemicals were used with-
out further purification: tetrahydrofuran (THF, 99%, Aladdin), 2,2-
azobis(isobutyronitrile) (AIBN, 99%, Aladdin), trifluoromethanesul-
fonic acid (CF3SO3H, 98%, Aladdin), sodium p-styrenesulfonate hy-
drate (SPSS, C8H7NaO3S·xH2O, 90%, Aladdin), sodium chloride (Tian-
jin Tianli Chemical Reagent Co., Ltd.), sulfuric acid (98%, Tianjin
Tianli Chemical Reagent Co., Ltd.), ethanol absolute (Sinopharm
Chemical Reagent Co., Ltd.), divinylbenzene (DVB, 80%, Macklin
Biochemical Co., Ltd.), sodium hydroxide (Tianjin Hengxing Chemi-
cal Reagent Co., Ltd.), Amberlyst-15 (Acros Organics). For the cata-
lytic activity tests we used: d-sorbitol (98%, Aladdin), isosorbide
(98%, Aladdin), 3,6-anhydro-d-galactose (ꢁ95%, Aladdin), 1,5-an-
hydro-d-glucitol (Aladdin), 1,4-anhydro-sorbitol (TRC Canada), and
2,5-anhydro-d-glucitol (J&K Chemical Ltd.). The divinylbenzene
(DVB) used in the catalyst synthesis was washed with sodium hy-
droxide solution (5%) to remove the polymerization inhibitor.
The acid exchange capacity of the solid acid catalysts was deter-
mined by using acid–base titration. Typically, catalyst (0.05 g) was
dried at 1008C for 10 h and then stirred in NaCl (2m, 50 mL) solu-
tion for 12 h until equilibrium was reached. The solution was then
sonicated for 30 min. The sample was collected by filtration and
washed, and the free H+ in the filtrate was titrated by standard
sodium hydroxide solution (0.01m) with phenolphthalein as an in-
dicator.
XRD patterns were measured by using a DX-2700 with a mono-
chromator and CuKa radiation (l=1.54184 ꢁ) at 40 kV and 30 mA.
The scanning rate was 48minꢀ1 in the range 2q=5–858.
The morphology of the catalyst samples was investigated by using
SEM at 5 kV by using an Extreme-resolution Analytical Field Emis-
sion SEM (Tescan Mira 3).
Catalyst Preparation
PDVBꢀSO3H (PDS)
TEM images were recorded by using a JEM-2100F high-resolution
transmission electron microscope operated at an acceleration volt-
age of 200 kV.
The functional mesoporous material PDVBꢀSO3H was obtained by
the hydrothermal synthesis of DVB and SPSS followed by ion ex-
change with dilute sulfuric acid. The molar ratio of DVB, SPSS,
AIBN, THF, and H2O was 1/0.96/0.02/16.1/7.23. Typically, DVB (2 g)
and SPSS (0.96 g) were added to a solution that contained AIBN
(0.05 g), THF (20 mL), and H2O (2 mL). After it was stirred at RT for
3 h, the solution was transferred to an autoclave and treated hy-
drothermally at 1008C for 24 h. Then, the white solid was evaporat-
ed at RT for 48 h to obtain PDSNa. Furthermore, PDSNa (1 g) was
treated with 1m sulfuric acid (100 mL) in ethanol for 24 h to obtain
PDVBꢀSO3H (PDS).[13a,26] The synthesis of the superhydrophobic
mesoporous network PDVB was similar to that of PDS except that
SPSS was not added.
Contact angles (CA) were tested by using a SL200B, Kino, USA.
Adsorption isotherms of water were performed by using a BEL-
SORP-max instrument at 258C. The samples (ꢂ25 mg) were out-
gassed for 10 h at 1508C before the measurements.
FTIR spectra were recorded by using a Nicolet Nexus 470 FTIR
Spectrometer. The range and resolution of acquisition were 4000–
400 cmꢀ1 with 64 scans and 2 cmꢀ1, respectively. A self-supporting
wafer of each sample was diluted with KBr.
To analyze the chemical composition in all samples, XPS was per-
formed by using a Thermo Fisher ESCALAB 250 xi instrument with
AlKa radiation (1486.6 eV).
PDVBꢀSO3HꢀSO2CF3 (PDSF-x)
Thermogravimetric analysis was performed by using a Rigaku
Thermo plus Evo TG 8120 in a flow of dry air (30 mLminꢀ1). The
heating rate was 108Cminꢀ1 (RT to 8008C).
The polymer solid superacid PDSF-x was obtained by grafting dif-
ferent amounts of the strong electron-withdrawing group ꢀSO2CF3
on the PDS precursor. Typically for PDSF-0.05, PDS (3 g) was added
to a flask that contained toluene (100 mL), and the temperature
was increased rapidly to 1008C. Then, CF3SO3H (5 mL) was dropped
slowly into the above solution. After it was stirred for 24 h, the
PDSF-0.05 catalyst was washed with a large amount of CH2Cl2 fol-
lowed by Soxhlet extraction for 48 h. The sample was dried under
vacuum at 808C. The synthesis of the other PDSF-x (x=0.01, 0.1,
0.15, 0.2) catalysts was the same except for the amount of CF3SO3H
(1, 10, 15, 20 mL).[30]
Solid-state 31P NMR spectra were recorded by using a Bruker
Avance 600 MHz wide-bore spectrometer equipped with a 4 mm
double-resonance probe. Details of the TMPO adsorption process
have been reported previously.[32] Before the tests, the samples
were pretreated as follows. To remove the water adsorbed on the
surface of the sample completely, the test temperature was in-
creased gradually to 1308C for 12 h under vacuum (10ꢀ3 Pa) fol-
lowed by cooling. Subsequently, a completely dry sample was dis-
persed in a mixture of TMPO (known in content) and CH2Cl2 in a
N2-filled glovebox, followed by the removal of CH2Cl2 at RT. The
sample was then heated at 1708C for 8 h to ensure the uniform
adsorption of probe molecules in the sample channels. Before test-
ing, the sealed sample tube was opened and transferred to the
NMR rotor sealed by a gas-tight Kel-F cap in a N2-filled glovebox. A
Larmor frequency of 600 MHz and a typical p/2 pulse length of
4.6 ms were adopted. For the 31P MAS NMR spectroscopy experi-
ment, the single pulse sequence was equivalent to ca. p/4 (for
For comparison, a PDVBF-0.05 catalyst that contained only one
acid group (ꢀSO2CF3) was synthesized in which PDVB was used as
the precursor, and then the grafting method of the ꢀSO2CF3 group
was the same as that used for PDSF-0.05.
Characterization
N2 isotherms were measured at ꢀ1968C by using a Micromeritics
Tristar II (3020) instrument system. The surface area and pore sizes
&
ChemSusChem 2019, 12, 1 – 11
www.chemsuschem.org
8
ꢀ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
15 s). The chemical shifts were referenced to (NH4)2HPO4 (d=
1.0 ppm), and the MAS frequency was 10 kHz.
201701D221052), the National Natural Science Foundation of
China (No. 21776294), the National Natural Science Foundation
of China (No. 21802158), and the Independent Research Project
of the State Key Laboratory of Coal Conversion (No.
2018BWZ002).
EPR spectra were obtained by using an EMXPLUS10/12 working in
the X-band (9.84 GHz) at 258C.
Catalyst tests
Conflict of interest
Sorbitol dehydration reaction
The sorbitol dehydration reaction was conducted in a 250 mL
three-necked flask. Typically, sorbitol (25 g) was added to the
three-necked flask and heated to the desired temperature to
obtain a clear liquid (melted sorbitol without solvent), after which
a certain amount of the catalyst was added to the reaction system
(the reaction time was recorded as t=0). The reaction was per-
formed under vacuum (0.3 bar), and the stirring speed was
1200 rpm. Each reaction was repeated at least twice.
The authors declare no conflict of interest.
Keywords: acidity · biomass · carbohydrates · heterogeneous
catalysis · polymers
[1] a) T. W. Walker, A. H. Motagamwala, J. A. Dumesic, G. W. Huber, J. Catal.
2019, 369, 518–525; b) P. Sudarsanam, R. Zhong, S. Van den Bosch,
S. M. Coman, V. I. Parvulescu, B. F. Sels, Chem. Soc. Rev. 2018, 47, 8349–
8402.
[2] a) J. J. Bozell, G. R. Petersen, Green Chem. 2010, 12, 539–554; b) W. Wei,
C. Wang, Y. Zhao, S. Peng, H. Zhang, Y. Bian, H. Li, X. Zhou, H. Li, J.
Catal. 2015, 327, 78–85.
[3] R. C. Hockett, H. G. Fletcher, Jr., E. L. Sheffield, R. M. Goepp, Jr., J. Am.
Chem. Soc. 1946, 68, 927–930.
[4] J. D. Parker, J. O. Parker, N. Engl. J. Med. 1998, 338, 520–531.
[5] M. Durand, Y. Zhu, V. Molinier, T. Fꢂron, J. M. Aubry, J. Surfactants Deterg.
2009, 12, 371–378.
[6] a) D. J. Saxon, M. Nasiri, M. Mandal, S. Maduskar, P. J. Dauenhauer, C. J.
Cramer, A. M. LaPointe, T. M. Reineke, J. Am. Chem. Soc. 2019, 141,
5107–5111; b) H. Kobayashi, A. Fukuoka, Green Chem. 2013, 15, 1740;
c) M. Rose, R. Palkovits, ChemSusChem 2012, 5, 167–176.
[7] M. Yabushita, H. Kobayashi, A. Shrotri, K. Hara, S. Ito, A. Fukuoka, Bull.
Chem. Soc. Jpn. 2015, 88, 996–1002.
Product analysis
The reaction products were removed from the reactor at intervals,
cooled rapidly to RT, and diluted with deionized water. The solu-
tion was filtered through an organic filter (0.22 mm) before it was
analyzed by using HPLC (LC-10Avp) equipped with an Aminex
HPX-87H column (diameter 7.8 mm, 300 mm, 9.0 mm) and a differ-
ential refraction detector (RID-10A). The eluent was a 0.005m solu-
tion of H2SO4 (0.6 mLminꢀ1), and the column temperature was
608C. The mass balance for all the reactions was higher than 96%.
The specific product analysis method is consistent with our previ-
ous work.[13a] Equations (1) and (2) were used to calculate the sorbi-
tol conversion (CSorbitol) and product selectivity (Sx):
[8] K. Bock, C. Pedersen, H. Thogersen, Acta Chem. Scand. Ser. B 1981, 35,
441–449.
[9] C. Dussenne, T. Delaunay, V. Wiatz, H. Wyart, I. Suisse, M. Sauthier, Green
Chem. 2017, 19, 5332–5344.
moles of reacted sorbitol
moles of initial sorbitol
ð1Þ
ð2Þ
CSorbitolð%Þ¼
ꢃ100
moles of product x
moles of all products
Sxð%Þ¼
ꢃ100
[10] a) A. A. Dabbawala, D. K. Mishra, G. W. Huber, J. S. Hwang, Appl. Catal. A
2015, 492, 252–261; b) M. H. G. Flꢃche, Starch/Staerke 1986, 38, 26–30;
c) A. Duclos, C. Fayet, J. Gelas, Synthesis 1994, 1087–1090.
[11] a) F. Liu, K. D. O. Vigier, M. P. Titus, Y. Pouilloux, J. M. Clacens, F. De-
campo, F. Jꢂrꢄme, Green Chem. 2013, 15, 901–909; b) S. Kobayashi, M.
Sugiura, H. Kitagawa, W. W. L. Lam, Chem. Rev. 2002, 102, 2227–2302;
c) S. Kobayashi, K. Manabe, Acc. Chem. Res. 2002, 35, 209–217.
[12] a) J. Deng, B. H. Xu, Y. F. Wang, X. E. Mo, R. Zhang, Y. Li, S. J. Zhang,
Catal. Sci. Technol. 2017, 7, 2065–2073; b) J. Li, W. Buijs, R. J. Berger,
J. A. Moulijn, M. Makkee, Catal. Sci. Technol. 2014, 4, 152–163; c) A. Ka-
mimura, K. Murata, Y. Tanaka, T. Okagawa, H. Matsumoto, K. Kaiso, M.
Yoshimoto, ChemSusChem 2014, 7, 3257–3259; d) J. Li, A. Spina, J. A.
Moulijn, M. Makkee, Catal. Sci. Technol. 2013, 3, 1540–1546.
[13] a) D. Yuan, N. Zhao, Y. Wang, K. Xuan, F. Li, Y. Pu, F. Wang, L. Li, F. Xiao,
Appl. Catal. B 2019, 240, 182–192; b) J. Zhang, L. Wang, F. Liu, X. Meng,
J. Mao, F. S. Xiao, Catal. Today 2015, 242, 249–254; c) J. M. Fraile, C. J.
Saavedra, ChemistrySelect 2017, 2, 1013–1018; d) M. J. Ginꢂs-Molina,
R. M. Tost, J. S. Gonzꢅlez, P. M. Torres, Appl. Catal. A 2017, 537, 66–73;
e) A. Cubo, J. Iglesias, G. Morales, J. A. Melero, J. Moreno, R. Sꢅnchez-
Vꢅzquez, Appl. Catal. A 2017, 531, 151–160; f) R. S. Thombal, V. H.
Jadhav, Tetrahedron Lett. 2016, 57, 4398–4400; g) J. Shi, Y. Shan, Y. Tian,
Y. Wan, Y. Zheng, Y. Feng, RSC Adv. 2016, 6, 13514–13521; h) I. Polaert,
M. C. Felix, M. Fornasero, S. Marcotte, J. C. Buvat, L. Estel, Chem. Eng. J.
2013, 222, 228–239.
In which Sx stands for the selectivity of sorbitans or isosorbide. In
the product analysis, we used the respective response factors for
sorbitol and isosorbide. A trace amount of 3,6-sorbitan may be in-
cluded in the detected 1,4-sorbitan as they were epimers that
cannot be separated completely by using HPLC.[10a,13h,17a,b,19] In ad-
dition, the selectivity of undetected products and humins was con-
sidered as SOthers, and SOthers =100%ꢀSSorbitanꢀSIsosorbide
TOF values [hꢀ1] were calculated by Equation (3):
.
*
nsorbitol Csorbitol
TOF ¼
ð3Þ
mcatalyst
N
t
*
acid site
*
in which nsorbitol stands for the amount of the reactant sorbitol
[mol], Csorbitol stands for the conversion of sorbitol (10–20%), mcatalyst
stands for the mass of the catalyst [g], Nacid
number of the acid sites obtained by acid–base titration
[mmolgꢀ1], and the reaction time is represented by t [h].
stands for the
site
[14] a) H. Kobayashi, H. Yokoyama, B. Feng, A. Fukuoka, Green Chem. 2015,
17, 2732–2735; b) R. Otomo, T. Yokoi, T. Tatsumi, Appl. Catal. A 2015,
505, 28–35; c) G. Morales, J. Iglesias, J. A. Melero, J. Moreno, R. Sꢅnchez-
Vꢅzquez, ꢆ. Peral, A. Cubo, Top. Catal. 2017, 60, 1027–1039.
[15] a) W. Ni, D. Li, X. Zhao, W. Ma, K. Kong, Q. Gu, M. Chen, Z. Hou, Catal.
Today 2019, 319, 66–75; b) H. Li, D. Yu, Y. Hu, P. Sun, J. Xia, H. Huang,
Carbon 2010, 48, 4547–4555.
Acknowledgements
This work was supported by the Natural Science Foundation of
Shanxi Province, China (No. 201801D121070), the Science Foun-
dation for Young Scientists of Shanxi Province, China (No.
ChemSusChem 2019, 12, 1 – 11
www.chemsuschem.org
9
ꢀ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
[16] M. Gu, D. Yu, H. Zhang, P. Sun, H. Huang, Catal. Lett. 2009, 133, 214–
220.
[26] F. Liu, X. Meng, Y. Zhang, L. Ren, F. Nawaz, F. S. Xiao, J. Catal. 2010, 271,
52–58.
[27] a) X. Zhang, Y. Zhao, S. Xu, Y. Yang, J. Liu, Y. Wei, Q. Yang, Nat. Commun.
2014, 5, 3170; b) J. M. Fraile, E. Garcꢇa-Bordejꢂ, E. Pires, L. Roldꢅn, J.
Catal. 2015, 324, 107–118.
[28] J. Scaranto, A. P. Charmet, S. Giorgianni, J. Phys. Chem. C 2008, 112,
9443–9447.
[29] C. Kolbeck, M. Killian, F. Maier, N. Paape, P. Wasserscheid, H. P. Steinrꢈck,
Langmuir 2008, 24, 9500.
[30] F. Liu, A. Zheng, I. Noshadi, F. S. Xiao, Appl. Catal. B 2013, 136, 193–201.
[31] M. Alvaro, A. Corma, D. Das, V. Fornes, H. Garcia, Chem. Commun. 2004,
956–957.
[17] a) Y. Zhang, T. Chen, G. Zhang, G. Wang, H. Zhang, Appl. Catal. A 2018,
562, 258–266; b) X. Zhang, D. Yu, J. Zhao, W. Zhang, Y. Dong, H. Huang,
Catal. Commun. 2014, 43, 29–33; c) N. A. Khan, D. K. Mishra, I. Ahmed,
J. W. Yoon, J. S. Hwang, S. H. Jhung, Appl. Catal. A 2013, 452, 34–38;
d) I. Ahmed, N. A. Khan, D. K. Mishra, J. S. Lee, J. S. Hwang, S. H. Jhung,
Chem. Eng. Sci. 2013, 93, 91–95; e) A. A. Dabbawala, D. K. Mishra, J. S.
Hwang, Catal. Commun. 2013, 42, 1–5; f) J. Xia, D. Yu, Y. Hu, B. Zou, P.
Sun, H. Li, H. Huang, Catal. Commun. 2011, 12, 544–547; g) Z. C. Tang,
D. H. Yu, P. Sun, H. Li, H. Huang, Bull. Korean Chem. Soc. 2010, 31, 3679–
3683.
[32] a) A. Zheng, H. Zhang, X. Lu, S. B. Liu, F. Deng, J. Phys. Chem. B 2008,
112, 4496–4505; b) A. Zheng, S. B. Liu, F. Deng, Chem. Rev. 2017, 117,
12475–12531.
[18] a) J. Zou, D. Cao, W. Tao, S. Zhang, L. Cui, F. Zeng, W. Cai, RSC Adv. 2016,
6, 49528–49536; b) P. Sun, D. H. Yu, Y. Hu, Z. C. Tang, J. J. Xia, H. Li, H.
Huang, Korean J. Chem. Eng. 2011, 28, 99–105.
[33] Q. Sun, K. Hu, K. Leng, X. Yi, B. Aguila, Y. Sun, A. Zheng, X. Meng, S. Ma,
F. S. Xiao, J. Mater. Chem. A 2018, 6, 18712–18719.
[34] C. Bispo, K. D. O. Vigier, M. Sardo, N. Bion, L. Mafra, P. Ferreira, F. Jꢂrꢄme,
Catal. Sci. Technol. 2014, 4, 2235–2240.
[35] H. Lv, X. Zhao, Z. Li, D. Yang, Z. Wang, X. Yang, Polym. Chem. 2014, 5,
6279–6286.
[36] a) H. Tang, N. Li, G. Li, W. Wang, A. Wang, Y. Cong, X. Wang, ACS Sustain-
able Chem. Eng. 2018, 6, 5645–5652; b) Z. Zhang, J. Song, B. Han,
Chem. Rev. 2017, 117, 6834–6880.
[19] Y. Zhang, T. Chen, G. Zhang, G. Wang, H. Zhang, Appl. Catal. A 2019,
575, 38–47.
ˇ
[20] a) D. Farcas¸iu, A. Ghenciu, G. Marino, R. V. Kastrup, J. Mol. Catal. A 1997,
ˇ
126, 141–150; b) D. Farcas¸iu, A. Ghenciu, G. Marino, K. D. Rose, J. Am.
Chem. Soc. 1997, 119, 11826–11831.
[21] X. Zhang, Y. Zhao, Q. Yang, J. Catal. 2014, 320, 180–188.
[22] J. C. Goodwin, J. E. Hodge, D. Weisleder, Carbohydr. Res. 1980, 79, 133–
141.
[23] Y. Xiu, A. Chen, X. Liu, C. Chen, J. Chen, L. Guo, R. Zhang, Z. Hou, RSC
Adv. 2015, 5, 28233–28241.
Manuscript received: July 15, 2019
Revised manuscript received: August 19, 2019
[24] D. Y. Zhao, Q. S. Huo, J. L. Feng, B. F. Chmelka, G. D. Stucky, J. Am. Chem.
Soc. 1998, 120, 6024–6036.
[25] J. A. Melero, R. Van Grieken, G. Morales, Chem. Rev. 2006, 106, 3790–
3812.
Accepted manuscript online: September 1, 2019
Version of record online: && &&, 0000
&
ChemSusChem 2019, 12, 1 – 11
www.chemsuschem.org
10
ꢀ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

FULL PAPERS
Hard graft: A series of polymeric solid
acid catalysts (PDSF-x) is prepared by
grafting strong electron-withdrawing
groups (ꢀSO2CF3) on a sulfonic acid-
modified polydivinylbenzene (PDS) pre-
cursor. The effect of acid strength on
sorbitol dehydration is investigated.
D. Yuan, L. Li, F. Li, Y. Wang, F. Wang,
N. Zhao,* F. Xiao*
&& – &&
Solvent-Free Production of Isosorbide
from Sorbitol Catalyzed by a
Polymeric Solid Acid
ChemSusChem 2019, 12, 1 – 11
www.chemsuschem.org
11
ꢀ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ