Role of acid sites and selectivity correlation in solvent free liquid phase dehydration of sorbitol to isosorbide

Article abstract of DOI:10.1016/j.apcata.2014.12.014

A number of Br?nsted acids (methanesulfonic acid, p-toluene sulfonic acid, triflic acid, sulfamic acid, citric acid, NaHSO₄, and boric acid) and Lewis acids (metal sulfate/triflates) were employed in solvent free dehydration of sorbitol and their influence on anhydroalcohols selectivity has been investigated. The outcome indicated that all the acid catalysts produced first mono-dehydrated product sorbitan followed by second dehydration of 1,4-sorbitan to isosorbide. However, the formation and yield isosorbide were found to depend on the nature of acid sites and their acidic strength. The Br?nsted acids are more efficient to convert sorbitol to isosorbide than Lewis acids. The Br?nsted acids having lower pKa value (i.e. strong acid) exhibited high catalytic activity as well as yield of isosorbide. In the case of Lewis acids, the catalytic activity and selectivity were radically depended on which metal used and their stability during the reaction. The water formed during reaction induced Br?nsted acidity on Lewis acid metal site. The Lewis-assisted Br?nsted acid site enabled high yield of isosorbide up to 70% at moderate temperature (160°C).

Full text of DOI:10.1016/j.apcata.2014.12.014

Accepted Manuscript
Title: Role of acid sites and selectivity correlation in solvent
free liquid phase dehydration of sorbitol to isosorbide
Author: Aasif A. Dabbawala Dinesh K. Mishra George W.
Huber Jin-Soo Hwang
PII:
S0926-860X(14)00765-0
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.apcata.2014.12.014
APCATA 15152
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
10-9-2014
13-11-2014
5-12-2014
Please cite this article as: <doi>http://dx.doi.org/10.1016/j.apcata.2014.12.014</doi>
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Role of acid sites and selectivity correlation in solvent free liquid phase dehydration of
sorbitol to isosorbide
Aasif A. Dabbawala a, Dinesh K. Mishra a, George W. Huber b, Jin-Soo Hwang a *
a
Bio-refinery Research Center, Korea Research Institute of Chemical Technology (KRICT),
Sinseoungno 19, Yuseong, Daejeon 305-600, Republic of Korea
b
Department of Chemical and Biological Engineering, University of Wisconsin-Madison,
Madison, WI 53706, USA
* Corresponding Author: Tel.: +82-42-860-7382; fax: +82-42-860-7676
E-mail address: jshwang@krict.re.kr (J.-S. Hwang)
Graphic Abstract
OH
HO
B
OH
O
S
HO
ONa
Bronsted acids
O
OH
O
S
H
H
Activity and Selectivity
depend on pKa value
O
HO
NH2
O
OH
O
O
S
O
HO
O
S
OH
OH
S
Isosorbide
-H2O
OH
O
S
O
O
F
O
F
OH
HO
HO
OH
O
F
O
OH
Sc
Yb
1,4-Sorbitan
Ti
Activity and Selectivity
depend on metal used &
its stability during reaction
-H2O
OH OH
Zr
OH
HO
Fe
Lewis acid
OH OH
Sorbitol
Metal sulfates/triflates
Al
Cu
1
Page 1 of 31

Research highlights
 Influence of acid sites investigated in dehydration of sorbitol to isosorbide
 Brönsted acids are efficient to convert sorbitol to isosorbide than Lewis acids
 Strong Brönsted acids exhibit high TOF as well as yield of isosorbide
 The water formed during reaction induced Brönsted acidity on Lewis acid catalysts
 The Lewis-assisted Brönsted acid site enables high yield of isosorbide at 160 °C
2
Page 2 of 31

Role of acid sites and selectivity correlation in solvent free liquid phase dehydration of
sorbitol to isosorbide
Aasif A. Dabbawala a, Dinesh K. Mishra a, George W. Huber b, Jin-Soo Hwang a *
a
Bio-refinery Research Center, Korea Research Institute of Chemical Technology (KRICT),
Sinseoungno 19, Yuseong, Daejeon 305-600, Republic of Korea
b
Department of Chemical and Biological Engineering, University of Wisconsin-Madison,
Madison, WI 53706, USA
* Corresponding Author: Tel.: +82-42-860-7382; fax: +82-42-860-7676
E-mail address: jshwang@krict.re.kr (J.-S. Hwang)
Abstract
A number of Brönsted acids (methanesulfonic acid, p-toluene sulfonic acid, triflic acid,
sulfamic acid, citric acid, NaHSO4, boric acid) and Lewis acids (metal sulfate/triflates) were
employed in solvent free dehydration of sorbitol and their influence on anhydroalcohols
selectivity has been investigated. The outcome indicated that all the acid catalysts produced
first mono-dehydrated product sorbitan followed by second dehydration of 1,4-sorbitan to
isosorbide. However, the formation and yield isosorbide were found to depend on the nature
of acid sites and their acidic strength. The Brönsted acids are more efficient to convert
sorbitol to isosorbide than Lewis acids. The Brönsted acids having lower pKa value (i.e.
strong acid) exhibited high catalytic activity as well as yield of isosorbide. In the case of
Lewis acids, the catalytic activity and selectivity were radically depended on which metal
used and their stability during the reaction. The water formed during reaction induced
Brönsted acidity on Lewis acid metal site. The Lewis-assisted Brönsted acid site enabled high
yield of isosorbide up to 70% at moderate temperature (160 °C).
Key words: Dehydration; sorbitol; selectivity; Brönsted acid; Lewis acid
3
Page 3 of 31

1. Introduction
Over the last few decades, the catalytic conversion of sugar and sugar alcohols into
fuels and value added chemicals have gained considerable interest due to their low cost and
availability from renewable resources [1−9]. Sorbitol, a sugar alcohol can be produced either
from hydrolysis-hydrogenation of cellulose/lignocellulosic biomass or by direct
hydrogenation of glucose [10−15]. Recently, sorbitol has been listed as a platform bio-based
molecule which can be used in the preparation of a variety of valuable non-petroleum derived
commodity chemicals [10−20]. Dehydration of sorbitol produces anhydrosugar alcohols,
including sorbitan (mono-anhydrosorbitol) and isosorbide (di-anhydrosorbitol). Both products
sorbitan and isosorbide have achieved commercial importance and can be used to synthesize
numerous intermediates applicable in the preparation of surfactants, pharmaceuticals,
cosmetics and synthetic polymer industries [20−22]. Isosorbide, a compound containing
bicyclic hetero atoms in its structure is obtained by sequential dehydration of sorbitol as
shown in Scheme 1. In intramolecular dehydration of sorbitol, sorbitol is first dehydrated to
give monocyclic products, such as 1,4-sorbitan (1,4-anhydro-d-sorbitol), 1,5-sorbitan and 2,5-
sorbitan. 1,4-sorbitan is further dehydrated to produce bicyclic product, isosorbide (1,4:3,6-
dianhydrosorbitol) whereas other sorbitans cannot produce isosorbide. Isosorbide can also be
formed via 3,6-sorbitan although it has been considered that first cyclization occurred
between C1 and C4 position. Moreover, the formation of ‘humins’ (colored polymeric side
products) is major impurity in the case of sorbitol dehydration. The high molecular weight
condensation/degradation undesired colored by products (‘humins’) are start to generate from
the beginning of the reaction indicating major formation of side products from sorbitol and
sorbitan [22].
4
Page 4 of 31

HO
OH
3
6
OH
Humins
HO
2
5
OH
H O
1
4
HO
D-Sorbitol
O
OH
HO
OH
O
-H2O
OH
OH
HO
OH
OH
2,5-anhydro-L-iditol
O
HO
OH
HO
HO
O
1,5-Sorbitan
HO
OH
2,5-anhydro-L-mannitol
OH
HO
O
HO
OH
OH
3,6-Sorbitan
1,4-Sorbitan
Humins
-H2O
Humins
HO
H
H
O
O
OH
Isosorbide
Scheme 1 Acid catalyzed dehydration of sorbitol to isosorbide via sorbitan
The dehydration of sorbitol was first studied under homogeneous condition using
traditional strong mineral acids such as H2SO4, H3PO4, HCl and high yield of isosorbide was
achieved with H2SO4 [23]. Later on, various processes have been developed for the
production of isosorbide by sorbitol dehydration using range of acid catalysts to enhance
conversion of sorbitol and yield of isosorbide under homogeneous as well as heterogeneous
condition [24−32]. Many researchers employed the solid acid heterogeneous catalysts to
avoid the drawbacks of mineral acids i.e. required tedious product separation, mineral acid
handling risk, equipment corrosion and less stability of product [33−42]. The various solid
acids such as zeolites, Amberlyst-15 resin, Metal (IV) phosphates, phosphated Nb2O5,
sulfated copper oxides, sulfated zirconia/titania, and supported heteropolyacid have been
applied for the synthesis of isosorbide under different reaction conditions [33−42]. Gu et al
studied aqueous phase dehydration of sorbitol to isosorbide using metal (IV) phosphates of
5
Page 5 of 31

tin, zirconium and titanium and achieved 97% conversion of sorbitol and isosorbide yield in
the range of 45−47% at the reaction temperature of 300 °C [35]. Khan et al investigated
dehydration of sorbitol under solvent free condition and reported 61% isosorbide selectivity
with 100% conversion of sorbitol at 210 °C using sulfated zirconia catalyst [38]. A number of
advantages have been realized by using solid acid catalysts compared to the homogeneous
mineral acid catalysts as they are safe in handling, easy to separate from products and can be
reuse. However, most of the solid acid catalysts showed a low catalytic activity and required
comparatively high temperature (200−300 C) and longer reaction time to achieve

satisfactory yield of isosorbide [33−42]. On the other hand, Li et al studied the dehydration of
sorbitol using molten salt hydrate as reaction medium (ZnCl2, 70 wt% in water) and achieved
high selectivity to isosorbide at 200 C [43,44]. Yamaguchi et al showed sorbitol dehydration

without using any acid catalyst in high temperature liquid water [45]. The maximum yield of
isosorbide reached up to 57% but requires high temperature (317 °C) and pressure. Based on
the reported literature and our previous studies on dehydration of sorbitol [38−40], we
observed that (i) the acid catalyst is key requirement for efficient conversion of sorbitol at
moderate reaction temperature, (ii) the selectivity of anhydroalcohols is affected by type of
selected acid catalyst and its properties. However, there is no systematic study available in the
literature outlining the influence of acid catalyst properties such as Brönsted acid (proton
donors) and Lewis acid (electron pair acceptor) on conversion of sorbitol and the
anhydrosugar alcohols selectivity. Since the activity of acid catalysts depends on the catalyst
acid strength and nature of acid sites. More importantly, such systematic data would be useful
in the designing of rational catalyst system for the efficient dehydration of sorbitol. With this
view, we intended to explore the dehydration of sorbitol using a series of acid catalysts having
different acidic strength and nature of acid sites (Brönsted and Lewis sites). Organic acids
having Brönsted acid sites and metal sulfate/triflates as Lewis acids were selected to examine
6
Page 6 of 31

the role of Brönsted acid and Lewis acid in the dehydration of sorbitol. The influence of
Brönsted acid and Lewis acid are emphasized by linking the catalyst activity and
anhydroalcohols selectivity. Under optimum reaction condition, the complete conversion of
sorbitol with high yield of isosorbide up to 70% was achieved at moderate reaction
temperature (160 °C). In addition, the dehydration of sorbitol was investigated under solvent
free condition instead of aqueous phase and without using complicated pressure reactor setup.
The water formed during reaction was continuously removed by vacuum. In fact, the water
present in the reaction medium directly affects the rate of dehydration and requires longer
time to complete the reaction. Moreover, the additional distillation step is required to separate
water from the product.
2. Results and discussion
2.1. Brönsted acid catalyzed dehydration of sorbitol
The dehydration reaction of sorbitol was performed in a liquid phase batch type
dehydration reactor under solvent free condition. The main dehydrated products observed
were 1,4−sorbitan and isosorbide along with small amount of 1,5−sorbitan. Non-indentified
soluble and insoluble polymeric compounds (‘humins’) are reported here as ‘unknown. The
catalytic dehydration of sorbitol was investigated under homogeneous condition using distinct
acid catalysts. Two types of acid catalyst were selected in the present study: (i) Organic acid
catalysts with ‘Brönsted acid sites’ and (ii) transition/non transition metal salt acting as
‘Lewis acid’. Compared to strong mineral acids, selected acid catalysts are less corrosive and
ease in handling. Firstly, we studied the effect of Brönsted acid catalysts on activity and
selectivity of sorbitol dehydration by picking number of acid catalysts having different pKa
value such as Triflic acid (TFA, pKa = −12), p-toluenesulfonic acid (PTSA, pKa = −2.8)
methanesulfonic acid (MSA, pKa = −1.9), Sulfamic acid (SFA, pKa = 1.0), sodium hydrogen
sulfate, NaHSO4 (SHS, pKa = 1.99), citric acid (CTA, pKa = 3.13), boric acid (BA, pKa =
7
Page 7 of 31

9.24). For comparison, few experiments were also performed with H2SO4 (SA, pKa = −3).
The results of dehydration of sorbitol using different acid catalysts are summarized in Table
1.
Table 1 Dehydration of sorbitol using Brönsted acid catalystsa
Selectivity (%)
ST
Temp.
(C)
Catalyst
pKa
Time
(min)
Conv.
(%)
(H+ mmol)
(pH)b
IS
48
68
39
63
45
66
46
66
14
41
10
14
04
06
03
04
UN
13
16
15
19
12
15
11
14
15
16
07
07
07
08
07
07
H2SO4 (SA)
(0.52)
-3
(3.29)
30
60
30
60
30
60
30
60
30
60
30
60
60
120
60
120
100
100
100
100
100
100
100
100
95
39
16
46
18
43
19
43
20
71
43
81
79
89
86
90
89
160
160
160
160
180
180
180
180
TFA
(0.52)
-12
(3.21)
PTSA
(0.52)
-2.8
(3.27)
MSA
(0.52)
-1.9
(3.25)
SFA
(1.04)
1.0
(2.8)
100
92
SHS
(1.04)
1.99
(2.9)
98
CTA
(1.56)
3.13
(3.5)
48
64
BA
(1.56)
9.24
(6.7)
41
55
a Reaction condition: Sorbitol = 54.9 mmol, Stirring speed = 800 rpm
IS: Isosorbide, ST: Sorbitan (1,4-sorbitan + 1,5-sorbtian), UN: Unknown, non-detected
products (‘Humins’)
b pH of reaction mixture after reaction (2% aqueous solution)
The results pointed out that the catalyst activity and yield of isosorbide were changed
depending on pKa value of acid catalyst used. The formation of side products was also
influenced by the type of acid catalyst and its acidic strength. It was noticed that acid catalysts
8
Page 8 of 31

having pKa value < 1 were found to produce high yield of isosorbide at 160 °C however their
catalyst performance were almost similar in term of activity and selectivity. In another case of
acid catalysts having pKa value > 1, the catalyst activity and formation of isosorbide were
found to decrease gradually with an increase in pKa value of acid catalyst. In addition, such
acid catalysts required higher catalyst concentration, temperature (180 °C) and longer
reaction time to attain complete conversion of sorbitol (Table 1). For example, organic acids
MSA, PTSA, and TFA having pKa < 1 exhibited high selectivity to isosorbide (63−66%) in
60 min at 160 °C comparable to H2SO4. On other side, SFA (pKa = 1.0), SHS (pKa = 1.99),
CTA (pKa = 3.13) and BA ( pKa = 9.24) displayed isosorbide selectivity 41%, 14%, 4% and
3% respectively in 60 min at 180 °C and also required higher catalyst concentration (Table
1). Initially, we compared results of all Brönsted acid catalyst based on selective formation of
isosorbide at higher conversion level of sorbitol. Because the dehydration of sorbitol to
isosorbide is sequential reaction and the quantitative formation of isosorbide was observed
after 70% conversion of sorbitol. For better evaluation of studied acid catalysts as well as
their proper comparison, the catalyst activities in term of turn over frequency for sorbitol
(represented as TOFSB where SB = sorbitol) were determined. The initial TOFs for sorbitol
(TOFSB) for all Brönsted acid catalyst were calculated at sorbitol conversion in the range of
10−20% and presented in Fig. 1.
9
Page 9 of 31

900
800
700
600
500
400
300
200
100
0
Fig. 1 The catalytic activity (in term of TOF) of distinct Brönsted acids in dehydration of
sorbitol; Reaction conditions are the same as mentioned in Table 1.
The conversion of sorbitol and yield of isosorbide were also determined with different
Brönsted acid catalysts and the results are presented at different time of interval (Fig. 2). The
organic Brönsted acids MSA, PTSA and TFA exhibited almost the same TOFs (760−855 hr-1)
and yield of isosorbide (63−66%) as TOF (823 hr-1) and yield of H2SO4 (68%). The SFA (310
hr-1), SHS (240 hr-1), CTA (57 hr-1) and boric acid (42 hr-1) gave lower TOFs and isosorbide
yield than H2SO4. The H2SO4 and organic acids (MSA, PTSA and TFA having –SO3H) are
10
Page 10 of 31

considered as strong acids. The strong acids have a larger Ka (acid dissociation constant) i.e.
ionizes or dissociates completely in a solution. These acids are stronger in aqueous solution
than hydronium ion (H3O+) i.e. strong acids have pKa < −1.74 (pKa of H3O+) [46]. The above
achieved results with different Brönsted acids suggested that strong acidity is necessary to
achieve high activity and yield of isosorbide at moderate temperature. However, the
behaviors of all strong acids (having pKa < −1.74) are same exhibiting almost comparable
TOFs and yield of isosorbide (Fig. 1). On other hand, the catalytic activity and yield of
isosorbide were observed to decrease gradually with an increase in pKa value of catalyst (in
the case of pKa > 1) as the acid strength of catalyst is decreased depending on pKa of acid
(Fig. 1 and Fig. 2).
SA
SFA
TFA
SHS
PTSA
CTA
MSA
BA
SA
SFA
TFA
SHS
PTSA
CTA
MSA
BA
100
90
80
70
60
50
40
30
20
10
0
70
60
50
40
30
20
10
0
0
10 20 30 40 50 60 70 80 90
Time (min)
0
10 20 30 40 50 60 70 80 90
Time (min)
Fig. 2 Influence of distinct Brönsted acids on sorbitol conversion and isosorbide yield in
dehydration of sorbitol; Reaction conditions are the same as mentioned in Table 1.
G. Fleche and M. Huchette studied aqueous phase dehydration of sorbitol using different
mineral acids such as H2SO4 (pKa = −3), HCl (pKa = −6.3) and H3PO4 (pKa = 2.1) [23]. Under
similar reaction conditions (Temp. 135 °C, reaction time = 20 h), they observed high
conversion (100%) and yield of isosorbide (70−75%) with H2SO4 and HCl as both acids are
strong acids (pKa < −1.74). Whereas H3PO4 having pKa = 2.1 showed only 12% yield of
isosorbide with 87% sorbitol conversion. These results are consistent with our achieved
11
Page 11 of 31

results further indicating the requirement of strong acidity for efficient conversion of sorbitol
to isosorbide at moderate reaction temperature.
The effects of reaction time, temperature and H+ concentration were then examined to
understand if further improvements are possible with organic acid using MSA as a catalyst.
The reaction profile for conversion of sorbitol (at 160 °C) and formation of sorbitan as well as
isosorbide is presented with respect to time in Fig. 3. In 3 min, 39% conversion of sorbitol
was observed that was increased to 68% in 5 min. The conversion of sorbitol was reached
89% in 10 min then the reaction became sluggish. 99% of conversion of sorbitol was attained
in 15 min. The catalytic dehydration of sorbitol yielded sequentially sorbitan and isosorbide.
At the outset sorbitan selectivity was high and selectivity to isosorbide was low. As the
reaction proceeded, selectivity to sorbitan was decreased leading to increase isosorbide
selectivity up to 66% in 60 min (Fig. 3). On further increase of reaction time, the yield of
isosorbide was not changed significantly.
% Conv.
% Selc. (sorbitan)
% Selc. (isosorbide)
% Selc. (others)
100
90
80
70
60
50
40
30
20
10
0
0
10
20
30
Time (min)
40
50
60
Fig. 3 Reaction profile of sorbitol dehydration using MSA as catalyst; Reaction condition:
sorbitol = 54.9 mmol, catalyst (MSA) = 0.52 mmol, reaction temperature = 160 °C.
12
Page 12 of 31

The effect of temperature was studied by varying reaction temperature in the range of 140 to
170 °C at constant amount of sorbitol (54.9 mmol) and MSA catalyst (0.52 mmol). From the
Fig. 4, it can be seen that the temperature had significant effect on catalytic activity and
isosorbide selectivity. With an increase in temperature the reaction proceeded faster and led to
increase isosorbide selectivity via sorbitan. At lower reaction temperature (140 °C), initial
TOF of sorbitol was 401 hr-1. With an increase in temperature from 140 to 170 °C, the initial
TOF of sorbitol was increased considerably from 401 to 1266 hr-1 indicating catalyst activity
increased by factor of ~3. Similarly, selectivity of isosorbide was found only 18% in 30 min
at 140 °C which was considerably increased up to 62% at 170 °C. However, the formations of
side products were also increased with an increase of reaction temperature (Fig. 4).
100
1400
1200
1000
800
% Selc. (Isosorbide)
% Selc. (Sorbitan)
% Selc. (Others)
(b)
(a)
90
80
70
60
50
40
30
20
10
0
600
400
200
140
150
160
170
140
150
160
C)
170
Temp. (C)
Temp. (

Fig. 4 Effect of temperature on TOF (a) and product selectivity (b) in MSA catalyzed
dehydration of sorbitol; Reaction condition: sorbitol = 54.9 mmol, catalyst (MSA) = 0.52
mmol, reaction time = 30 min.
The effect of H+ concentration was also studied by varying MSA concentration 0.26, 0.52,
0.78 and 1.04 mmol (Fig. 5). The conversion of sorbitol was increased with an increase of H+
concentration and complete conversion of sorbitol was attained relatively in shorter reaction
time. Similarly, the isosorbide yield was increased with an increase H+ concentration and high
yield was obtained comparatively in shorter reaction time. These results indicate that the
13
Page 13 of 31

sorbitol conversion and yield of isosorbide are influenced by number of bronsted acid sites
and augmented up to certain limit with an increase of relative number of Brönsted sites. It is
also noticed that higher H+ concentration and longer reaction time led to the formation of side
products.
+
1.04 H mmol
+
70
60
50
40
30
20
10
0
(b)
100
90
80
70
60
50
40
30
20
0.78 H mmol
(a)
0.26 H+mmol
0.52 H+mmol
0.78 H+mmol
1.04 H+mmol
+
0.52 H mmol
+
0.26 H mmol
0
5
10 15 20 25 30 35 40 45 50 55 60
Time (min)
0
5
10
15
20
25
30
Time (min)
Fig. 5 Effect of H+ concentration on sorbitol conversion (a) and isosorbide yield (b) in MSA
catalyzed dehydration of sorbitol; Reaction condition: sorbitol = 54.9 mmol, reaction
temperature = 160 °C.
2.2. Lewis acid catalyzed dehydration of sorbitol
Next, we studied the effect of Lewis acid catalysts on dehydration of sorbitol.
Transition/non transition metal salt (MS, where M = metal cation, S= anion) of different
-2
metal M = Ti, Fe, Co, Ni, Cu, Zn, Zr, and Al with specific anion S= Cl-1 and SO4 were
selected to elucidate the role of Lewis acid site. The metal salts of Fe, Cu, Al, Zr, and Ti in
sulfate form were found to give appreciable yields of isosorbide compared to their chloride
and the conversion of sorbitol and yield of isosorbide were varied depending on which metal
used (Table 2). For example, CuSO4 gave very low initial TOF (18 hr-1) with only 35%
conversion of sorbitol at 170 °C and also formation of isosorbide was not observed. At 180
°C, the initial TOF was increased up to 249 hr-1 and exhibited complete conversion of sorbitol
with 61% isosorbide selectivity in 2 h (Table 2). The strong Lewis acids, Ti(SO4)2 and
Zr(SO4)2 displayed high initial TOFs and isosorbide selectivity (65−68%) at 160 °C in 60 min
14
Page 14 of 31

resemblance to strong Brönsted acids. The other metal sulfates gave lower initial TOF and
isosorbide selectivity (<60%). Additionally, they required relatively higher reaction
temperature to enhance yield of isosorbide.
Table 2 Dehydration of sorbitol using Lewis acid catalysts a
d
TOFSB
(hr-1)
Selectivity (%)
IS UN
Temp.
(°C)
Time
(min)
Conv.
(%)
pHb Χic
Catalyst
ST
82
96
47
23
88
80
57
29
38
18
38
16
96
70
33
38
18
5.79 ---
5.51 9.5
126
18
12
0
06
04
09
16
08
11
12
15
14
17
12
16
02
09
12
13
15
CuCl2
180
170
180
120
120
60
100
35
CuSO4
249
44
61
04
09
31
56
48
65
50
68
02
21
55
49
67
99
120
60
100
56
3.04 12.6
3.97 10.5
2.94 12.6
3.01 13.5
86
Fe2(SO4)3
Al2(SO4)3
Zr(SO4)2
Ti(SO4)2
Yb(OTf)3
170
170
160
160
120
60
77
240
798
886
99
120
30
100
100
100
100
100
21
60
30
60
5.6
---
22
160
180
60
133
60
96
120
30
100
100
100
3.8
---
675
Sc(OTf)3
160
60
a Reaction condition: Sorbitol = 54.9 mmol, catalyst = 0.52 mmol, stirring speed = 800 rpm
b pH of reaction mixture after reaction (2% aqueous solution)
c Χi = (1+ 2Z) Xo; where Xo = electronegativity of the neutral atom, Z = charge of the ion [47]
d TOFSB = initial turnover frequency for sorbitol calculated at 10−20% sorbitol conversion
15
Page 15 of 31

Previously, we and Xia et al studied dehydration of sorbitol to isosorbide using sulfated metal
oxides of different metals such as Ni, Cu, Al, Zr and Ti under heterogeneous condition
[37−39]. It was observed that catalyst activity and isosorbide selectivity were varied with
metal used and high catalyst isosorbide selectivity was achieved with sulfated TiO2 (70%).
The above achieved results with homogeneous Lewis acid catalysts are in accordance with
previous results and realized high yield at relatively low reaction temperature (160 °C against
200−220 °C) was due to homogenous condition. The reason for high activity of Ti(SO4)2 and
Zr(SO4)2 may be due to their higher electronegativity (Table 2, for Ti(SO4)2, Χi= 13.5 and for
Zr(SO4)2, Χi= 12.6). The electronegativity is an important parameter for metal sulfates to
determine their catalytic activity and selectivity in acid-catalyzed reactions [47–49].
However, Fe2(SO4)3 having high electronegativity (Χi = 12.6) produced less isosorbide
because of less solubility of Fe2(SO4)3 in molten sorbitol as reaction was performed in solvent
free condition. The acidic properties of metal sulfates also depend upon their degree of
rehydration [50]. In presence of water, Lewis acids induce Brönsted acidity by co-ordination
of water molecules to the metal centre, thereby polarizing the O–H bond and rendering them
substantially more Brönsted acidic (water derived induce Brönsted acidity) than the parent
compound [51]. On the other side, It has also been reported that metal chloride react with the
hydroxyl groups of sugar to form metal alkoxides, M(OR)n and HCl [52]. Similarly, it can be
assume the formation of H2SO4 or the generation of bronsted acid sites on metal sulfates due
to the formation of water during reaction. The degree of metal sulfates rehydration and
coordination of sugar to metal are vary from one to another during dehydration; hence the
activity/selectivity of metal sulfates varied from one metal to another metal. In view of above,
metal sulfates could not act as ‘true Lewis acid’ due to the generation of water during reaction
and their activities are water dependent. The metal sulfates Ti(SO4)2 and Zr(SO4)2 displayed
high activity but their the reaction mixture (2% aqueous solution) are quite acidic i.e. pH
16
Page 16 of 31

around ~3 which is similar to the pH of organic acids catalyzed reaction mixture (Table 2).
Also, the dehydration of sorbitol catalyzed by such metal sulfates Ti(SO4)2 and Zr(SO4)2
exhibited similar isosorbide yield to Brönsted acid under identical reaction condition with
major intermediate product 1,4-sorbitan (see supporting information). It is suggested that
observed higher activity with such metal sulfates may not raised solely by Lewis acids sites
but generation of Brönsted acid sites on the surface of metal sulfates or formation of H2SO4.
Later, we selected metal triflates such as Yb(OTf)3, and Sc(OTf)3 as Lewis acid catalysts to
study dehydration of sorbitol. Such metal triflates are considered as stable water-soluble
Lewis acids. At 160 °C, The catalytic activity of Yb(OTf)3 was considerably less than
Sc(OTf)3 exhibiting only 21% conversion with 2% isosorbide selectivity. Sc(OTf)3 showed
initial TOF (675 hr-1) somewhat lower than Ti(SO4)2 and Zr(SO4)2 but produced isosorbide
with high selectivity (67%) in 60 min at 160 °C. Initially, catalyst activity and formation of
isosorbide to some extent were low due to the slow mixing of this triflate. Once it dissolved
completely in molten sorbitol ensuing high yield of isosorbide comparable to Ti(SO4)2,
Zr(SO4)2 and strong organic acid. However, pH of Sc(OTf)3 reaction mixture is relatively less
around 3.8 than the Yb(OTf)3 (pH = 5.6). The generation of triflic acid from metal triflate has
also been observed in some cases termed as ‘Hidden Brönsted Acidity’ [53]. Wang et al
studied dehydration of fructose to 5-hydroxymethylfurfural (5-HMF) in DMSO using various
metal triflates. The high catalytic activity and yield of 5-HMF were achieved with Sc(OTf)3
and Yb(OTf)3 due to their smaller ionic radius [54]. In the case of sorbitol dehydration by
such metal triflates displayed incongruous catalyst performance. The high catalytic activity
only achieved with Sc(OTf)3 with less pH of reaction mixture. This indicates the generation
of hidden Brönsted acid site in the case of Sc(OTf)3 which considerably enhance its catalytic
activity. From the results of Brönsted and Lewis acid catalysts, it can be seen that Brönsted
acid site and Lewis-assisted Brönsted acidity (water-derived Brönsted acidity) played
17
Page 17 of 31

significant role for efficient dehydration of sorbitol to isosorbide. The above results are in
connection with results of xylose (pentose sugar) dehydration to furfural reported by
Weingarten et al [55]. They investigated dehydration of xylose using homogeneous Brönsted
and Lewis acid catalysts. They observed that the catalyst selectivity is a function of the
Brönsted to Lewis acid site ratio for both the heterogeneous and homogeneous catalyst. The
Brönsted acid catalyst or catalyst with higher Brönsted to Lewis acid ratio produced high
yield of furfural while Lewis acid sites decrease furfural selectivity by catalyzing a side
reaction between xylose and furfural [55]. In the case of xylose dehydration, Lewis acid site
first isomerizes the xylose to xylulose or lyxose. The further conversion of these products
strongly depends on nature of acid sites whereas isomerization does not take place in case of
sorbitol (a sugar alcohol). Therefore, influence of Lewis acid site in sorbitol dehydration is
less than Brönsted acid site. Moreover, G. Fleche and M. Huchette reported that the
performance of Lewis acid catalysts (such as AlCl3 and SnCl4) in aqueous phase dehydration
of sorbitol is lower than Brönsted acid catalyst and are less efficient for production of
isosorbide [23].
Based on the literature and characterization data [34,42,43], possible mechanism is
outline for sorbitol dehydration using Brönsted and Lewis acid catalyst (Scheme 2). The
mechanism for dehydration of sorbitol by Brönsted acid (H+) can explained by preferential
protonation of the hydroxyl group at C1 or C6 position in the sorbitol (Scheme 2) and then
cyclic dehydration of these carbons with C4 or C3 produces 1,4-sorbitan or 3,6-sorbitan [34].
However, most of authors suggest first cyclization occur between C1 and C4 positions (1,4-
cyclization, B→C) which produces 1,4-sorbitan [34,40,42−45]. The further protonation of C6
hydroxyl group of 1,4-sorbitan and cyclic dehydration with C3 (3,6 cyclization, C1→D)
forms a bicyclic product, isosorbide (1,4:3,6-dianhydrosorbitol). The first cyclic dehydration
can also be take placed between C2 and C5 or C1 and C5 which leads to the formation of 2,5-
18
Page 18 of 31

sorbitan (through intramolecular hydrogen bonding, B→B2) and 1,5-sorbitan (B→B3). There
is also possibility of protonation at C5 in the case of 1,4-sorbitan through intramolecular
hydrogen bonding (C1→C2) [45]. Therefore, the second cyclic dehydration proceeds slowly
then first dehydration step. In addition, the formation of colored side products (‘humins’) can
be observed from the beginning of the reaction (Scheme 1). This indicates that the side
products may be generate by the dehydration of two sorbitol molecule or by the condensation
of sorbitol with one of the sorbitans and their degradation [23,34]. On the other hand, In the
case of metal sulfate, the reaction is either proceed with direct protonation of the C1 or C6
hydroxyl groups of sorbitol by metal cation or metal sulfate first react with the hydroxyl
groups of sorbitol to form metal alkoxides, M(OR)n [52] and H2SO4 ensuing isosorbide
similar to Brönsted acid.
H2SO
4 + M(OR)n
MSO4
HO
OH
3
HO
OH
3
6
OH
HO
6
OH
OH
2
5
OH
2
H O
4
5
OH
1
H O
OH
1
4
1,4-cyclization
OH
HO
-M+
HO
HO
O
-H+
OH
OH
1,4-Sorbitan
M+
(A)
D-Sorbitol
HO
OH
(C)
OH
+H+
-H2O
-H2O 3,6-cyclization
O
OH
HO
OH
OH
-H2O
OH
O
OH
H
M
HO
H
HO
OH
H O+
M
OH
O
O
H
OH
H O+
HO
HO
OH
H
O
(B)
1,5-cyclization
OH
H
OH
(D)
O
M
Isosorbide
OH
HO
O+
H
OH
H
O
H
OH
OH
+
H
OH
OH
HO
OH
H O
O
+
OH2
HO
OH
2,5-cyclization
1,5-Sorbitan
(B3)
OH
HO
(B1)
O
(C1)
OH
OH
O
(C2)
H+
1,4-Sorbitan
+
O
OH2
OH
HO
(C)
HO
OH
2,5-Sorbitan (B2)
Scheme 2 Proposed reaction mechanism for Brönsted and Lewis acid catalyzed dehydration
of sorbitol
19
Page 19 of 31

From the achieved results of metal sulfates, it has been observed that the catalyst TS
exhibited catalyst activity similar to MSA. The catalytic behavior of both catalysts (TS and
MSA) was also compared by considering kinetic aspect in order to get more insight about
formation products with their activation barriers using both catalysts. Based on reaction
mechanism of sorbitol dehydration and experimental data, the simplified reaction scheme is
proposed with rate constants (k1, k2 and k3) and corresponding rate equations (1 to 4) are
derived from the reaction scheme 3. The sorbitol undergoes to produce sorbitan or other by
products (parallel reactions) and sorbitan further dehydrates to produce isosorbide
(consecutive reaction). The constant k1 represents the dehydration of sorbitol (SB) to sorbitan
(ST), the constant k2 represents dehydration of sorbitan (ST) to isosorbide (IS) and the
constant k3 correspond to the formation of other dehydration and degradation by products
(OT) from Sorbitol or sorbitan. In order to determine order of reaction, the tests for zero, first
and second order were carried out. The plot of –log[SB]/[SB]0 against time is observed to be
linear indicating that the dehydration of sorbitol using both catalysts follows the first order
with respect to sorbitol (Fig. S4 and S5).
20
Page 20 of 31

Sorbitan (ST)
k2
k1
Sorbitol (SB)
Isosorbide (IS)
k3
By-products (OT)
Sorbitan (ST) = 1,4-Sorbitan + 1,5-Sorbitan
or
3,6-Sorbitan
d[SB]
dt
= - k1 [SB] - k3 [SB]
= k1 [SB] - k2 [ST]
= k2 [ST]
(1)
d[ST]
dt
(2)
(3)
(4)
d[IS]
dt
d[OT]
dt
= k3 [SB]
Scheme 3 Simplified reaction scheme for sorbitol dehydration with rate constant and
corresponding rate equations.
Oltmanns et al [42], Li et al [44] and Yamaguchi et al [45] also proposed kinetic model for
dehydration of sorbitol assuming first order reaction for all the steps. The experiment data of
both catalysts MSA and TS at different temperatures (140,150,160 and 170 C) were fitted to

the above kinetic model for the estimation of the three rate constants. The POLYMATH 6.1
software was used for the numerical integration of above differential equations and for
parameter estimations. The good concord between experiment data and modeled data can be
seen in all cases (Figure S6). In the present model, the measure data of sorbitan is combined
concentration of 1,4- and 1,5-sorbitan as peak of both intermediate products are not well
separated by HPLC. For both catalysts of MSA and TS, the rate constants (k1, k2 and k3) of
the three reaction steps at various reaction temperatures have been determined and presented
in Table 3. The activation energies and pre-exponential factors corresponding to each rate
21
Page 21 of 31

constants were calculated for both catalysts by Arrhenius plot (Fig. 6) and presented also in
Table 3. The observed activation energy values of the steps k1 and k2 for both catalysts MSA
and TS are comparatively less than other reported silicotungsticacid [42] and ZnCl2 molten
salt catalyst [44] systems indicating efficiency of both catalyst systems under solvent free
conditions. Although, both catalysts exhibited the activation energy values comparable to
H2SO4 catalyst for the formation of sorbitan and isosorbide (Ea1 = 47.9 and Ea2 = 51 kJ/mol)
[56]. It is noted that activation energy barrier of step 2 for isosorbide formation from sorbitan
is higher than step 1 for sorbitan formation. This indicated that the second cyclic dehydration
proceeds slowly then first dehydration step and coincides with above proposed mechanism. In
individual comparison of TS with MSA catalyst, the achieved activation energy values using
TS catalyst for each step are found to be comparable with MSA catalyst. This confirm that the
behavior of TS catalyst is similar to Brönsted acid MSA and generation of bronsted acid site
on the surface of metal sulfates as pure lewis acid catalyst is less efficient for sorbitol
dehydration to isosorbide.
Table 3 Estimated rate constants and corresponding activation energies for the dehydration of
sorbitol using MSA and TS catalysts
Rate
Temp. (K)
Catalyst
MSA
Ea
(kj/mol)
A
Constant
(10-2 min-1)
413
423
433
443
k1
k2
k3
k1
k2
k3
12.4
16.85
1.45
2.08
15.5
1.57
1.92
26.1
30.8
48.2
57.5
96.4
52.7
62.1
101.7
1.57E+05
1.99E+05
1.69E+10
5.42E+05
7.44+E05
7.27E+10
1.10
1.10
12.0
1.0
2.58
4.20
26.15
2.86
4.3
3.20
7.20
32.0
3.18
7.1
TS
1.0
22
Page 22 of 31

-1.0
-1.5
-2.0
-2.5
-3.0
-3.5
-4.0
-4.5
-1.0
-1.5
-2.0
-2.5
-3.0
-3.5
-4.0
-4.5
k
k
1
1
(a)
(b)
k
k
k
k
3
3
2
2
2.24 2.26 2.28 2.30 2.32 2.34 2.36 2.38 2.40 2.42
2.24 2.26 2.28 2.30 2.32 2.34 2.36 2.38 2.40 2.42
-3
-3
1/T
(10
K)
1/T
(10
K)
Fig. 6 Arrhenius plots of three reaction steps for both catalysts MSA (a) and TS (b).
In order to confirm further role of acid sites and their influence on isosorbide
selectivity, the few experiments were also performed by combining two catalyst systems. It
has been reported that the combination of Brönsted and Lewis acid catalyst or the catalyst
having Brönsted as well as Lewis acid sites can greatly influence the catalyst activity and
selectivity [38−40, 55,57]. Subsequently, we combined Brönsted and Lewis acid catalyst in
equal ratio of 1:1 without varying final concentration of catalyst (0.52 mmol) and studied
their influence on activity/selectivity. Two sets of experiments were performed one with
combining MSA and Ti(SO4)2 (TS) and another with MSA and Sc(OTf)3 (SCT). Figure 7
shows yield of isosorbide at different time of interval for individual and combined catalysts.
In the case of combined catalyst (MSA+TS), the initial reaction rate and isosorbide yield were
same as individual catalyst system and exhibited a maximum isosorbide yield (70%) at final
stage. On another hand, the initial reaction rate and isosorbide yield were slightly improved at
initial stage in the case of MSA+SCT than individual one. The addition of MST to SCT
increased amount of Brönsted acid site which led to enhance catalyst performance. However,
this combined catalyst showed almost comparable isosorbide yield (68%) after 1 h. The
above results confirm the key role of Brönsted acid site and catalyst with Lewis-assisted
Brönsted acid site (water-derived Brönsted acidity) performs similar to Brönsted acid
23
Page 23 of 31

catalyst. Moreover, the catalyst with higher amount of Brönsted acid sites led to enhance
catalyst activity and isosorbide selectivity. In this view, present findings would be useful in
designing of efficient heterogeneous catalyst system for dehydration of sorbitol.
TS+MSA (0.52 mmol)
TS (0.52 mmol)
MSA (0.52 mmol)
MSA (0.26 mmol)
MSA+SCT (0.52 mmol)
SCT (0.52 mmol)
MSA (0.52 mmol)
80
70
60
50
40
30
20
10
0
70
60
50
40
30
20
10
0
(b)
(a)
0
10
20
30
40
50
60
0
10
20
30
40
50
60
Time (min)
Time (min)
Fig. 7 Dehydration of sorbitol using combination of Brönsted and Lewis acid catalyst, (a)
MSA and TS, (b) MSA and SCT; Reaction condition: Sorbitol = 54.9 mmol, temperature =
160 °C.
3. Conclusions
Dehydration of sorbitol to isosorbide has been investigated under solvent free condition
using a range of acid catalysts such as organic acids, metal sulfates and metal triflates. The
strength and nature of acid sites played an important role in improving the sorbitol conversion
and isosorbide selectivity. Based on the experiment results of distinct Brönsted acids, it is
stated that the strong acidity (pKa < −1.74) is necessary to achieve high activity and yield of
isosorbide at moderate temperature. All strong organic acids such as MSA, PTSA, TFA
having pKa < −1.74 exhibit almost similar catalyst activity and isosorbide selectivity. The
catalysts with modest or weak acidity produce high amount of sorbitan (a mono-dehydrated
product). Beyond acidic strength of catalyst, the catalyst activity and anhydroalcohols
selectivity are also affected by the nature of Brönsted and Lewis acid sites. The Brönsted acid
24
Page 24 of 31

catalyst or catalyst with high amount of Brönsted acidity is effective for dehydration of
sorbitol to isosorbide than Lewis acid. The catalyst activity of Lewis acid catalyst is depended
on which metal used and the generation of Brönsted acid site on the metal centre. The Lewis-
assisted Brönsted acid site (water-derived Brönsted acidity) exhibits similar performance as
Brönsted acid catalyst and enables high yield of isosorbide (70%) at moderate temperature
(160 °C).
4. Experimental
4.1. Materials
D-Sorbitol, 1,4-sorbitan, 1,5-sorbitan, isosorbide, p-toluenesulfonic acid (PTSA), methane
sulfonic acid (MSA), trifloromethane sulfonic acid (TFA), sulfamic acid (SFA), Citric acid
.
.
.
(CTA), NaHSO4 (SHS), Boric acid (BA), CuCl2 2H2O, CuSO4 5H2O, Fe2(SO4)3 xH2O, NiSO4,
.
Al2(SO4)3 xH2O, Yb(OTf)3, Sc(OTf)3, acetonitrile (99.5%) were purchased from Sigma-
Aldrich company, Inc, (USA). H2SO4 (95%), Ti(SO4)2 and Zr(SO4)2 were purchased from
Junsei chemical Co. Ltd, Japan. All the chemicals used in this study were used without
further purification. De-ionized water was used as solvent for making all solutions.
4.2. General procedure for the dehydration of sorbitol
The dehydration of sorbitol was performed in simple reactor setup under solvent free and
reduce pressure condition. This process involved the following steps: (i) Melting of sorbitol at
110 °C through conventional heating with heating rate of 5 °C/min in air. (ii) Addition of
catalyst in order to start dehydration reaction at desire reaction temperature. (iii)
Neutralization of acidic catalyst and then filtration of reaction mixture to separate solid
impurities from the product. The dehydration of sorbitol was performed under vacuum (100
mm Hg) and was maintained constantly throughout reaction to remove water from the
reaction mixture. The dehydration of sorbitol was carried out in liquid phase using a round
25
Page 25 of 31

bottomed 100 mL three-neck flask equipped with temperature indicator and a water-cooled
condenser attached to a vacuum pump. Sorbitol (54.9 mmol) was first charged into the
dehydration reactor. The temperature was heated to 110 °C. After melting of sorbitol, the
reactor was heated to desire reaction temperature. An appropriate amount of acid catalyst (the
number of acid site was kept constant at 0.52 mmole in most of cases) was then added to the
liquid sorbitol to start the dehydration reaction. The reaction mixture was stirred at 800 rpm
and vacuum was applied (100 mm Hg). The vacuum was maintained continuously to remove
water from the reaction mixture. Samples were withdrawn at selected times, diluted by DI
water and filtered with a 0.2 µm syringe filter prior to analysis. The composition of the
dehydrated products was analyzed using HPLC (Younglin Instrument, Acme 9000) equipped
with a refractive index (RI) detector and Asahipak column (NH2P-50 4E, No. N712004). An
acetonitrile/water (80/20) mixture was used as an eluent for the analysis with a flow rate of
1.0 mL/min. The temperature of the RI detector was maintained at 35 °C throughout the
analysis. The products and sorbitol were well separated by HPLC and indentified by their
pure samples (see supporting information). The main dehydration products detected were
isosorbide and sorbitan (mostly 1,4-sorbitan with small amount of 1,5-sorbitan). The sorbitol
and isosorbide were quantified using their own response factors whereas response factor of
1,4-sorbitan was used to quantify both sorbitans. Based on molar compositions, the
conversion of sorbitol (SB) and selectivity of isosorbide (IS) and sorbitan (ST) were
calculated according to the reported method [34,35] by using following equations. The non-
indentified soluble and insoluble compounds were reported as unknown or others (humins).
The yield of humins or unknowns was deduced from the difference between initial mole of
sorbitol and sum of the moles of isosorbide and sorbitan analyzed by HPLC (i.e. it gives mole
of remaining unknown compounds from initial mole of feed Sorbitol). In order to calculate
carbon balance, the total mass of carbon in the liquid phase is obtained by TOC analysis
26
Page 26 of 31

(Table S1). The initial TOF for sorbitol (TOFSB) were calculated at sorbitol conversion in the
range of 10−20%.
moles of sorbitol converted
_______________________
(in the case of Bronsted acid)
TOFSB
TOFSB
=
moles of H+ x hr
moles of sorbitol converted
_______________________
(in the case of Lewis acid)
=
moles of acid x hr
mole of sorbitol at particular time
_______________________
Initial mole of sorbitol
x 100
1
Conversion SB (%) =
mole of sorbitan in product
_______________________
Initial mole of sorbitol
x 100
Yield ST (%) =
Yield of sorbitan
_______________________
x 100
Selectivity ST (%) =
Sorbitol conversion
mole of isosorbide in product
_______________________
Initial mole of sorbitol
x 100
Yield IS (%) =
Yield of isosorbide
_______________________
x 100
Selectivity IS (%) =
Sorbitol conversion
Acknowledgements
We would like to acknowledge the financial support from the R&D Convergence Program of
MSIP (Ministry of Science, ICT and Future Planning) and ISTK (Korea Research Council for
Industrial Science and Technology), Republic of Korea by a Grant (B551179-10-03-00).
27
Page 27 of 31

References
[1]
J. N. Chheda, G. W. Huber, J.A. Dumesic, Angew. Chem. Int. Ed. 46 (2007) 7164–
7183.
[2]
[3]
[4]
M. J. Climent, A. Corma, S. Iborra, Green Chem. 16 (2014) 516–547.
A. Corma, O. de la Torre, M. Renz, Energy Environ. Sci. 5 (2012) 6328–6344.
H. Kobayashi, Y. Ito, T. Komanoya, Y. Hosaka, P. L. Dhepe, K. Kasai, K. Hara, A.
Fukuoka, Green Chem. 13 (2011) 326–333.
[5]
[6]
[7]
[8]
[9]
J. Chen, S. Wang, J. Huang, L. Chen, L. Ma, X. Huang, ChemSusChem 6 (2013)
1545–1555.
P. Khemthong, P. Daorattanachai, N. Laosiripojana, K. Faungnawakij, Catal.
Commun. 29 (2012) 96–100.
B.O. Beeck, J. Geboers, S.V. Vyver, J.V. Lishout, J. Snelders, W.J.J. Huijgen, C.M.
Courtin, P.A. Jacobs, B.F. Sels, ChemSusChem 6 (2013) 199–208.
P. Daorattanachai, P. Khemthong, N. Viriya-empikul, N. Laosiripojana, K.
Faungnawakij, Carbohydr. Res. 363 (2012) 58–61.
B. Saha, M. M. Abu-Omar, Green Chem. 16 (2014) 24–38.
[10] J. Geboers, S. Van de Vyver, K. Carpentier, P. Jacobs, B. Sels, Chem. Commun. 47
(2011) 5590–5592.
[11] S. Dutta, S. De, M. I. Alam, M. M. Abu-Omar, B. Saha, J. Catal. 288 (2012) 8–15.
[12] B. Kamm, Angew. Chem. Int. Ed. 46 (2007) 5056–5058.
[13] A. Corma, S. Iborra, A. Velty, Chem. Rev. 107 (2007) 2411–2502.
[14] R. Rinaldi, F. Shcuth, ChemSusChem 2 (2009) 1096–1100.
[15] D. K. Mishra, A. A. Dabbawala, J. J. Park, S. H. Jhung, J. S. Hwang, Catal. Today 232
(2014) 99–107.
28
Page 28 of 31

[16] A. Shrotri, A. Tanksale, J. N. Beltramini, H. Gurav, S. V. Chilukuri, Catal. Sci.
Technol. 2 (2012) 1852–1858.
[17] M. Banu, P. Venuvanalingam, R. Shanmugam, B. Viswanathan, S. Sivasanker, Top.
Catal. 55 (2012) 897–907.
[18] M. Yabushita, H. Kobayashia, A. Fukuokaa, Appl. Catal. B: Env. 145 (2014) 1–9.
[19] Y. T. Kim, J. A. Dumesic, G. W. Huber, J. Catal. 304 (2013) 72–85.
[20] R. M. de Almeida, J. Li, C. Nederlof, P. O’Connor, M. Makkee , J.A. Moulijn
ChemSusChem 3 (2010) 325–328.
[21] M. Rose, R. Palkovits, ChemSusChem 5 (2012) 167–176.
[22] C.-H. Lee, H. Takagi, H. Okamoto, M. Kato, J. Appl. Polym. Sci. (2013) 530–534.
[23] G. Fleche, M. Huchette, Starch 38 (1986) 26–30.
[24] M. A. Andrews, K. K. Bhatia, P. J. Fagan, WO Patent 2001092266, 2001.
[25] G. Fleche, P. Fuertes, T. Rodolphe, H. Wyart, U.S. Patent 7122661, 2006.
[26] J. E. Holladay, J. Hu, Y. Wang, T. A. Werpy, X. Zhang, U.S. Patent 7649099, 2010.
[27] T. W. Abraham, D. M. Ference, W. Zhang, WO Patent 2012083149, 2012.
[28] H. Ryu, Y. J. Jung, Y. S. Kim, WO Patent 2012165676, 2012.
[29] S. J. Howard, A. J. Sanborn, WO Patent 2012015616, 2012.
[30] J. E. Holladay, J. Hu, Y. Wang, T. A. Werpy, X. Zhang, U.S. Patent 7615652, 2009.
[31] J. E. Holladay, J. Hu, X. Zhang, Y. Wang, U.S. Patent 7772412, 2010.
[32] J. Hu, J. E. Holladay, X. Zhang, Y. Wang, U.S. Patent 7728156, 2010.
[33] N. A. Khan, D. K.Mishra, J.S. Hwang, Y. W. Kwak, S. H. Jhung, Res. Chem.
Intermed. 37 (2011) 1231–1238.
[34] I. Polaert, M.C. Felix, M. Fornasero, S. Marcotte, J.-C. Buvat, L. Estel, Chem. Eng. J.
222 (2013) 228–239.
29
Page 29 of 31