Dehydration of sorbitol to isosorbide over sulfonic acid resins under solvent-free conditions

Article abstract of DOI:10.1016/j.apcata.2017.03.006

Different commercial sulfonic acid resins (Purolite and Amberlyst type) have been evaluated as acid catalysts for the dehydration of sorbitol to isosorbide. These acidic resins differ in their acid properties, as well as in their thermal and mechanical stabilities. Dehydration of sorbitol has been carried out under solvent-free conditions, by melting sorbitol. At the beginning of reaction, different sorbitans (monodehydration products) were detected, whose dehydration and subsequent cyclization lead to the formation of isosorbide. A maximum yield of 75% is reached after 12?h at 413?K, by using a 5?wt% of Purolite CT269, at atmospheric pressure. This resin maintains its catalytic activity after four runs, and the stability is corroborated by the negligible presence of sulphur species (coming from sulfonic acid leaching) in the reaction medium. This suitable catalytic performance can be explained by its high acid capacity (5.2?meq?g^?1) and mechanical and thermal stabilities associated to the macroreticular structure.

Full text of DOI:10.1016/j.apcata.2017.03.006

Accepted Manuscript
Title: Dehydration of sorbitol to isosorbide over sulfonic acid
resins under solvent-free conditions
´
´
´
´
´
Authors: Marıa Jose Gines Molina, Ramon Moreno-Tost, Jose
´
´
Santamarıa-Gonzalez, Pedro Maireles-Torres
PII:
S0926-860X(17)30095-9
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.apcata.2017.03.006
APCATA 16165
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
23-12-2016
24-2-2017
4-3-2017
´
´
´
´
´
Please cite this article as: Marıa Jose Gines Molina, Ramon Moreno-Tost, Jose
´
´
Santamarıa-Gonzalez, Pedro Maireles-Torres, Dehydration of sorbitol to isosorbide
over sulfonic acid resins under solvent-free conditions, Applied Catalysis A, General
http://dx.doi.org/10.1016/j.apcata.2017.03.006
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.

Dehydration of sorbitol to isosorbide over sulfonic acid resins under
solvent-free conditions
María José Ginés Molina, Ramón Moreno-Tost, José Santamaría-González, Pedro
Maireles-Torres*
Universidad de Málaga, Departamento de Química Inorgánica, Cristalografía y
Mineralogía (Unidad Asociada al ICP-CSIC), Facultad de Ciencias, Campus de
Teatinos, 29071 Málaga, Spain
* Corresponding author: Tel.: +34 952137534; fax: +34 952131870
E-mail: maireles@uma.es (P. Maireles-Torres)
Graphical abstract
1

Highlights
 Purolite CT 269DR is very active in sorbitol dehydration to isosorbide.
 This resin possesses suitable acidity, mechanical and thermal stabilities.
 Purolite CT 269DR maintains its activity during 4 cycles, at 413 K for 90 min.
 An isosorbide yield of 75% is attained, after 12 h, at atmospheric pressure.
Abstract
Different commercial sulfonic acid resins (Purolite and Amberlyst type) have been
evaluated as acid catalysts for the dehydration of sorbitol to isosorbide. These acidic
resins differ in their acid properties, as well as in their thermal and mechanical
stabilities. Dehydration of sorbitol has been carried out under solvent-free conditions,
by melting sorbitol. At the beginning of reaction, different sorbitans (monodehydration
products) were detected, whose dehydration and subsequent cyclization lead to the
formation of isosorbide. A maximum yield of 75 % is reached after 12 h at 413 K, by
using a 5 wt% of Purolite CT269, at atmospheric pressure. This resin maintains its
catalytic activity after four runs, and the stability is corroborated by the negligible
presence of sulphur species (coming from sulfonic acid leaching) in the reaction
medium. This suitable catalytic performance can be explained by its high acid capacity
(5.2 meq g-1) and mechanical and thermal stabilities associated to the macroreticular
structure.
Keywords: Biomass; Lignocellulose; Sorbitol; Isosorbide; sulfonic acids resins
2

1. Introduction
Currently, much attention is being paid to the use of biomass as a renewable and
sustainable feedstock for the production of biofuels, energy and chemicals [1-4]. In this
regard, the biorefinery concept emerges as an alternative to the traditional refinery
based on fossil raw materials (oil and natural gas), where different integrated and
sustainable mechano-physical, biochemical, chemical and thermochemical processes
allow biomass conversion into chemicals, biofuels and energy.
Sorbitol is one of the most important platform molecules, obtained from
hydrolysis/hydrogenation of lignocellulosic biomass (mainly, from cellulose and
hemicellulose) or by hydrogenation of glucose, requiring high temperature and pressure
for accomplishing hydrogenation, and after subsequent filtration and refined by ion
exchange, sorbitol can be finally concentrated, being employed for the synthesis of a
large spectrum of chemicals [5]. Industrially, the raw material for sorbitol manufacture
is glucose, usually obtained from starch [6]. Sorbitol was included as platform molecule
in the revised list of top chemicals derived from biomass, together with ethanol, furans,
glycerol, lactic acid, succinic acid, hydroxypropionic acid/aldehyde, levulinic acid and
xylitol [7,8].
The importance of isosorbide for the production of monomers and building
blocks for new polymers and functional materials, as well as for medical and
pharmaceutical uses, has been recently highlighted by Rose and Palkovits [8]. In this
sense, isosorbide exhibits excellent physicochemical properties applicable to various
fields of industry, since it is an important pharmaceutical intermediate (diuretic, and
mainly to treat hydrocephalus and glaucoma), used as additive to improve the strength
and rigidity of polymers as polyethylene terephthalate (PET), and as monomer for the
production of biodegradable polymers.
3

Among the compounds derived from isosorbide are dinitrate and mononitrate,
the latter being a compound widely used as vasodilator for angina and congestive heart
failure. Isosorbide derivatives can also find potential application as fuels or fuel
additives due to the high energy content, which can bring aliphatic substituents
(dimethyl isosorbide (DMI)).
The dehydration of sorbitol to isosorbide occurs through two consecutive steps:
a first cyclization with loss of a water molecule, which can lead to different chemical
intermediates such as 2,5- and 1,5-sorbitans, considered as byproducts since they don’t
evolve toward isosorbide, and 1,4- and 3,6-sorbitans, whose subsequent dehydration
and cyclization gives rise to isosorbide.
This process is industrially performed under homogeneous catalysis in the
presence of strong mineral acids [10,11]. Although these systems provide high sorbitol
conversion at low temperature, drawbacks associated to the high costs of separation and
corrosion problems in the equipment have prompted the search of alternative processes.
In this context, the development of solid acid catalysts represents a more sustainable
alternative from an economic and environmental point of view, and would allow in
some cases a modulation of selectivity.
In this sense, various reaction systems (aqueous solution of sorbitol in gas and
liquid phases, molten sorbitol) in the presence of a wide range of solid acid catalysts
have been essayed. Among them, it has been reported the use of zeolites [12-15],
tetravalent and trivalent metal phosphates [16-18], heteropolyacids supported on silica
[19], silicotungstic acid [20], sulfated metal (Ti, Cu, Sn, Zr) oxides [21-24], niobium
phosphate [25], phosphated tantalum oxide [26], and cellulose-derived solid acid
catalyst [27].
4

Sulfonic acid resins have been also evaluated as solid acid catalysts for the
conversion of sorbitol into isosorbide, and the best catalytic results have been obtained
with Amberlyst-35, under different experimental conditions (molten sorbitol, sorbitol
aqueous solution in gas phase) [28-31]. More recently, the catalytic performances of a
commercial Amberlyst-15 and a superhydrophobic mesoporous polymer-based acid
catalyst (P-SO3-H) have been compared [32]. After 10 h at 413 K, the conventional
resin showed a high conversion (94%) and selectivity to isosorbide (72%), but an
isosorbide yield close to 88% at full sorbitol conversion was attained with the P-SO3-H
polymer. In addition, reutilization studies have corroborated the excellent performance
and stability of the P-SO3-H polymer, which maintained a selectivity 78% after 5
recycles, whereas the Amberlyst-15 only gave a yield of 15% after 3 recycling essays.
This suitable catalytic activity of the P-SO3-H polymer was explained by considering
both its large porosity and strong hydrophobic character, which facilitate the removing
of formed water molecules from the catalyst surface.
In the present work, different commercial sulfonic resins (Amberlyst and
Purolite type) have been evaluated as acid catalysts for the dehydration of sorbitol to
isosorbide, by using molten sorbitol. Several experimental variables have been
optimized in order to achieve the highest isosorbide yield, and special emphasis was put
on the reutilization study of the most active resin.
2. Materials and Methods
2.1. Chemicals
D-Sorbitol (99%, Aldrich), isosorbide (Dianhydro-D-glucitol) (98%, Aldrich), 1,5-
anhydroglucitol (Aldrich) and 1,4-sorbitan (USP Reference Standard) were of HPLC
grade. Different commercial ion-exchange resins were studied: Purolite CT275 (P-
5

CT275), Purolite CT269 (P-CT269) and Purolite PD206 (P-PD206), kindly provided by
Purolite®, and Amberlyst 35 (A-35) and Amberlyst 70 (A-70).
2.2. Characterization of catalysts
The textural parameters were provided by the commercial supplier. However,
the nitrogen adsorption–desorption isotherms at 77 K of selected samples were obtained
by using an automatic ASAP 2020 model of gas adsorption analyser from
Micromeritics. Prior to N2 adsorption, the samples were outgassed at 373 K and 10-4
mbar for 10 h. Surface areas were determined by using the Brunauer–Emmett–Teller
(BET) equation and a nitrogen molecule cross section of 16.2 Å2, and the Barrer-
Joyner-Halenda method was used to determine average pore volume values.
X-ray photoelectron spectroscopy (XPS) studies were carried out by using a
Physical Electronics PHI 5700 spectrometer equipped with a hemispherical electron
analyzer (model 80-365B) and a Mg Kα (1253.6 eV) X-ray source. High-resolution
spectra were obtained at 45º take-off-angle by a concentric hemispherical analyzer,
operating in the constant pass energy mode at 29.35 eV, using a 720 m diameter
analysis area. Charge referencing was done against adventitious carbon (C 1s at 284.8
eV). The pressure in the analysis chamber was kept lower than 5 . 10-6 Pa. PHI
ACCESS ESCA-V6.0 F software package was used for data acquisition and analysis. A
Shirley-type background was subtracted from the signals. Recorded spectra were
always fitted using Gauss-Lorentz curves in order to determinate more accurately the
binding energy of the different element core levels.
Elemental analysis (C, H, N and S) was performed by using a LECO analyzer,
CHNS-932 model, coupled to a pyrolysis oven LECO VTF. An inductively coupled
plasma emission and mass spectrometry Perkin Elmer Optima 7300 DV, with a
6

nebulizer gas flow of 0.6 L/min, a plasma gas flow of 15 L/min and potencial RF ICP
1300W, were used for elemental analysis of liquid solutions.
2.3. Catalytic reaction
Catalytic experiments were performed in batch, by using a glass pressure tube
with thread bushing (Ace, 15 mL, pressure rated to 10 bars) and magnetic stirring. In a
typical procedure, 2 g of D-sorbitol (SigmaUltra, >99%) and 0.100 g of catalyst were
loaded into the reactor. Prior the experiments, reactors were always purged with
nitrogen. The reaction mixture was heated with a thermostatically controlled oil bath
and magnetically stirred at 600 rpm. After reaction time, the reaction was quenched by
submerging the reactor in a water bath cooled with ice. Remaining sorbitol and reaction
products were separated from the solid resin by adding distilled water. Liquid phase was
filtered and the analysis of products was performed by high performance liquid
chromatography (HPLC). A JASCO instrument, equipped with quaternary gradient
pump (PU-2089), multiwavelength detector (MD-2015), refractive index detector (RI-
2031-PLUS), autosampler (AS-2055), column oven (CO-2065) and a Monosaccharide
Phenomenex REZEX Ca2+ Monosaccharide (300mm x 7.8 mm) column for sugars
detection, was employed. The remaining sorbitol and reaction products were monitored
using the refractive index detector. The column is set at 80 °C and distilled water was
used as mobile phase (flow rate 0.6 ml/min).
3. Results and Discussion
The main physico-chemical characteristics of resins are compiled in Table 1. They
possess large pore diameters which provide easy accessibility to sulfonic acid sites, thus
preventing diffusional limitations. Two types of commercial resins have been used: i)
macroreticular structures, with low (A-70) and high (A-35, P-CT275, P-CT269)
crosslinking degree, and ii) gel-type resins (P-PD206) with non-permanent porosity,
7

unlike macroreticular resins whose high crosslinking degree maintains a permanent
porosity, with suitable textural properties. Concerning the sulfonation degree, A-70 is
conventionally sulfonated, whereas the rest of resins are oversulfonated, as can be
inferred from their high acid capacity.
Chemical analysis (C and S) of Purolites has confirmed their different degree of
sulfonation, as could be inferred from the S/C molar ratio (Table 2). The acid capacity
has been deduced from the percentage of sulphur, and values differ from those provided
by the supplier. However, the highest value is obtained for P-CT275, according to the
acid capacity values provided by the supplier, but, in general, elemental analyses gives
lower acidity values. For this reason, the concentration of acid sites of P-CT269 has
been measured, after treatment with a NaCl aqueous solution, by titration with a NaOH
aqueous solution, in the presence of phenolphthalein as acid-base indicator. Thus, an
acid capacity of 4.6 eq/kg was obtained, which is an intermediate value between those
deduced from elemental analysis (S content) and provided by the supplier.
The evaluation of the catalytic behavior in the dehydration of sorbitol was firstly
carried out at 413 K, with pure sorbitol, and a reaction time of 90 min. This temperature
is above the melting points of sorbitol (368 K), isosorbide (336 K) and sorbitans (384
K), which assures that all of them will be in liquid phase at 413 K. Under these
experimental conditions, sorbitol conversion is higher than 90 % for dry Amberlyst and
Purolite resins, whereas isosorbide formation is mainly favored on Purolites and dry A-
35 (Fig. 1). These results are in agreement with the high acid capacity of these resins,
with values of 4.1-4.8 eq kg-1for Purolites (Table 2), and 5.3 eq kg-1 for A-35.
By using the acid capacity determined from elemental analysis (Table 2), the
productivity of Purolite resins has been calculated after the normalization with the
available active sites. After 90 min at 413 K, isosorbide yields of P-CT269 and P-
8

CT275 were 33.4 and 39.4 mol%, respectively, (Fig. 1). Nevertheless, in spite of the P-
CT275 catalyst exhibits a better yield, the lower acid capacity of P-CT269 leads to a
similar productivity (0.16 mol isosorbide formed s-1 eqacid
-1). This similar
sites
productivity and its higher mechanical stability led us to choose this P-CT269 resin for
the optimization of different experimental variables.
Sorbitol dehydration follows the reaction pathways displayed in Scheme 1. The
first dehydration and cyclization process can lead to the 1,4-anhydro-d-sorbitol (1,4-
AHSO) intermediate, which, after subsequent dehydration and new cyclization, gives
rise to isosorbide, or two by-products resulting from the former dehydration: 2,5- and
1,5-anhydro-d-sorbitol (2,5- and 1,5-AHSO). In all cases, degradation and/or
degradation products could be generated from reactant and reaction products.
Yamaguchi et al. found, by studying sorbitol dehydration in high temperature liquid
water, that the rates of formation of 1,4- and 2,5-sorbitans were much faster than that of
1,5-sorbitan, thus explaining the prevalent formation of 1,4-sorbitan at the beginning of
the process, whereas its dehydration to isosorbide is slower [33]. The present catalytic
data point out that the rate of sorbitol dehydration to sorbitan, mainly 1,4-sorbitan, is
faster compared with the other catalytic steps. Moreover, dry Amberlyst resins attain
high sorbitol conversion and isosorbide yields than the parent resins, but, in the case of
Purolites, it is not necessary any pretreatment. This latter agrees well with data
published on the esterification of oleic acid with trimethylolpropane, in the presence of
different cation exchange resins, which demonstrated that macro-reticular resins don’t
need any pretreatment to exhibit catalytic activity [34]. By considering that P-PD206
has a lower maximum operation temperature (Table 1), the catalytic study was also
carried out at 393 K (Fig. 2) by comparing only Purolites. As expected, sorbitol
conversion is lower than that obtained at 413 K, being the best isosorbide yield (8.3 %)
9

reached with P-CT275. However, P-CT269 and P-PD206 display a similar catalytic
behavior, giving rise to isosorbide yields of 4.6-4.7 %.
Textural data reveal that PD206 possesses the worst textural properties, with
almost null porosity (Table 1), whereas the rest of resins exhibit similar specific surface
areas (20-40 m2 g-1) and pore volumes (0.3-0.6 cm3 g-1), according with data reported in
the literature. Regarding the average pore diameter, the P-CT275 resin has the highest
value (46.4 nm). Textural properties of most active samples (P-CT269 and P-CT275)
has been also determined to confirm data provided by the commercial supplier. In this
sense, SBET values of 39.5 and 22.2 m2 g-1 (relative pressure range of application of the
BET method: 0.06-0.20) were found for P-CT269 and P-CT275, respectively, which
agree well with values included in Table 1. Concerning average pore diameters (as
determined by the BJH method), they were 21.4 and 41.3 nm, respectively, also in
agreement with provided values, although low pore volumes (0.20 and 0.16 cm3 g-1)
were observed, due to the shrinkage of porous structure as a consequence of the pre-
treatment of samples before N2 adsorption. In any case, the textural properties, mainly
average pore sizes, are suitable to facilitate the accessibility of sorbitol molecules to
acid sites, minimizing diffusional limitations.
The comparison of the TG-DTA curves of P-CT269 and P-PD206 also confirm
the low thermal stability of P-PD206 (Fig. 3). Both resins present two well defined
regions in the TG and DTA curves. At lower temperatures (< 573 K), endothermic
effects are associated to the loss of hydration water, although P-CT269 only shows a
thermal event close to 380 K and a plateau in the TG curve extending from 410 until
505 K, whereas P-PD206 presents endothermic effects at 380, 450 and 570 K and an
almost continuous weight loss from room temperature until 1040 K. The second region
10

exhibits strong exothermic effect centered at 775 and 960 K for P-CT269 and P-PD206,
respectively.
Regarding these data, the use of P-PD206 is limited at reaction temperatures
lower than 393 K, according with the maximum operation temperature provided by the
supplier. This low thermal stability has been corroborated by Scanning Electron
Microscopy, since micrographs evidence the deterioration of P-PD206 beads after only
90 min of reaction time, at 413 K (Fig. 4). However, the other two Purolites (P-CT269
and P-CT275) resist these experimental conditions. These micrographs also allow to
distinguish the different range of particle sizes, according to data shown in Table 1.
However, although the catalytic behavior of P-CT269 and P-CT275 resins is similar, the
larger particle sizes of the latter resin are responsible of its lower mechanical stability,
as could be inferred from the existence of more important break lines in the P-CT275
particles.
This lower mechanical stability leads us to choose P-CT269 to study the influence
of the reaction temperature was carried out by using the P-CT269 resin, with a reaction
time of 90 minutes and varying the reaction temperature between 373 and 433 K (Fig.
5). A moderate increase of isosorbide yield is observed as temperature rises from 393 to
433 K, although it could be mostly attributed to the important increment of sorbitol
conversion, which goes from 25 until 93 %, while in contrast isosorbide selectivity only
varies from 18 to 38 %. However, by taking into account the maximum operation
temperature of this resin (453 K), and to ensure the thermal stability of the catalyst, a
reaction temperature of 413 K was chosen to optimize the isosorbide yield.
The kinetic study was performed by extending the reaction time until 44 h (Fig.
6). Almost full sorbitol conversion is already attained after 3 h, although isosorbide and
1,4-sorbitan yields are 43 and 37 %, respectively. From 3 until 10 h, a total (isosorbide
11

+ 1,4-sorbitan) yield close to 81 % is maintained, which would be explained by the
progressive dehydration and cyclization of 1,4-sorbitan to give rise to isosorbide, since
isosorbide selectivity increases at the expenses of 1,4-sorbitan. The decrease of sorbitan,
concomitantly with the increase of isosorbide, is in agreement with results previously
reported [19,33,35]. Moreover, a similar selectivity pattern is found in a kinetic study of
pure sorbitol dehydration, from which it was possible to infer that the kinetics of the
reaction can be described by a Langmuir-Hinshelwood type model, or first order
reactions [31]. At 12 h, 1,4-sorbitan is no longer detected, whereas isosorbide yield still
rises until 75%. That means that degradation processes have started, leading to the
transformation of isosorbide into insoluble and soluble polymers. From this moment,
isosorbide yield decays until a value of 42 % after 44 h of reaction time.
The kinetic parameters have been deduced by considering the reaction pathways
shown in Scheme 1, where the rate constants k1, k2 and k3 correspond to the conversion
of sorbitol to 1,4-AHSO, 1,4-AHSO to isosorbide and sorbitol to 1,5- and 2,5-AHSO
(and humins and other unknown compounds), respectively. It is assumed that the
formation rate of these by-products is unfavoured in comparison with 1,4- and 3-6-
AHSO, and isosorbide. Data displayed in Figure 6 reveal that maximum formation of
1,4-AHSO and 3,6-AHSO coincides with the drastic increase of sorbitol conversion in
the first 45 min. After this time, the concentration of 1,4-AHSO and 3,6-AHSO
decreases, while the amount of isosorbide raises. However, the amount of by-products
remains unchanged after 75 min of reaction, with a value close to 2 mmol.
Kinetic data confirm a first order kinetic model, as previously pointed out by
Polaert et al [31]. In this sense, the dehydration of sorbitol to 1,4-AHSO (and 3,6-
AHSO) at 413 K with P-CT269 has a rate constant of 393.1 kg-1 min-1 (k1) with an
activation energy of 56.38 kJ mol-1, which is in the same order of magnitude of a SBA-
12

Pr-SO3H catalyst (k1 = 332.3 kg-1 min-1, Ea = 87.30 KJ mol-1), at 443 K, reported by
Cubo et al. [36], although the use of this Purolite resin decreases the activation energy
of this monodehydration step. Previous studies [36, 37] have demonstrated that the
formation of isosorbide is kinetically less favorable and exists a high energy barrier.
This fact is corroborated by the value of the rate constant k2 (87.2 kg-1 min-1) (k2 << k1).
However, P-CT269 decreases the Ea associated to this second dehydration and
cyclization step (59.93 KJ mol-1), in comparison with data reported for other
sulphonated catalysts, such as SBA-Pr-SO3H and Amberlyst-70. In the case of the
formation of byproducts, its rate constant of K3 is only 21.1 kg-1 min-1.
By using a sulfonic acid-functionalized micro-bead silica, 84 % yield of
isosorbide was attained at 393 K for 10 h, but under vacuum, which decreases until 75
% after 10 catalytic cycles [38]. These authors claimed the strong hydrophilic nature
and the suitable large pore diameter as responsible of the excellent catalytic
performance. However, a superhydrophobic mesoporous polymer-based acid catalyst
allowed to reach an isosorbide yield close to 88% at full sorbitol conversion, and this
catalytic activity was contrarily associated to its hydrophobic nature [32]. Nevertheless,
in both cases, a pressure lower than 0.1 bar was used, whereas, in the present work, the
catalytic performance of P-CT269 is maintained at atmospheric pressure, with a positive
impact on overall cost of the dehydration process. The favorable role of hydrophobicity
was also inferred from the study of a high-silica aluminosilicate zeolite for the
dehydration of sorbitol in aqueous solution [15]. Similarly, by using H-beta with
different Si/Al ratio, it was concluded that the strong adsorption of water molecules in
micropores is the main cause of low catalytic activity, and increasing the hydrophobic
character of zeolites (high Si/Al ratio) ameliorates the catalytic performance in sorbitol
13

dehydration [14]. This drawback can be overcome by utilizing sulfonic acid resins with
high pore sizes, as inferred from the results obtained in this work.
The stability of the P-CT269 catalyst has been monitored by chemical analysis,
determining the S content in the reaction medium, after different reaction times (Table
3). In this regard, it is important to preserve the concentration of sulfonic groups in the
resin to maintain its activity in further catalytic cycles, which is an important advantage
of heterogeneous catalysis. Sulfur concentration values lower than 7 ppm were found in
the liquid phase after 12 h of reaction time by inductively coupled plasma optical
emission spectrometry (ICP-OES), which points out that almost negligible loss of
sulfonic groups occurs, since it corresponds to 5.38 % of the maximum leachable
amount (130 ppm).
On the other hand, the catalyst loading has been also optimized between 25 and
200 mg (Fig. 8). The maximum sorbitol conversion is attained with 150 mg of catalyst,
but the isosorbide yield increases by 14% when the amount of catalyst is doubled (200
mg). Therefore, it was estimated 100 mg as the optimal loading in the reaction. Other
important factor that must be taken into consideration in a catalytic study is the particle
size of solid catalysts, which, in the case of P-CT269, varies between 425 and 1200 m
(Table 1). For this reason, it has been considered of interest to evaluate its influence on
the catalytic performance (Fig. 9). For accomplishing this study, this resin was sieved to
obtain particle sizes in different ranges. Catalytic data have demonstrated a performance
improvement with use of catalysts with smaller particle size, reaching the highest
isosorbide yield for particle sizes ranging between 400 and 500 m.
Finally, the reutilization of the solid catalyst is a key issue in the development of
heterogeneous processes, as green and sustainable alternative to the use of liquid acids
and bases under homogeneous conditions. For meeting this requirement, and minimize
14

the weight loss associated to the catalyst recovery, five reactors were initially used
under similar reaction conditions (413 K, 90 min, 2 g sorbitol and 100 mg P-CT269).
Reaction mixture was allowed to cool down until room temperature, and solid was
separated, after adding deionized water to dissolve remaining sorbitol and reaction
products (sorbitans and isosorbide). Used catalysts were recovered, washed with
deionized water and dried at 373 K for 1 h. In successive catalytic cycles, the number of
identical reactors was lower. A slight decrease in conversion is observed after the first
cycle, but an average isosorbide yield of 24-29% was maintained during four cycles
(Fig. 10). The experimental error has been indicated, as deduced after repeating three
times each measurement. In order to check the leaching of sulfonic groups, chemical
analysis of used catalysts was carried out after the 2nd and 4th reaction cycles (Table 4).
The data reveal an increase of the C/S molar ratio, which could be attributed mainly to
the presence of insoluble high molecular weight product in used catalysts. This fact has
been confirmed by XPS. In this sense, the X-ray photoelectron spectrum in the C 1s
region of the fresh P-CT269 shows a main signal 284.8 eV, with a small tail at the
higher binding energy side, which is associated to C=C, C-C and C-H bonding of an
organic skeleton (Fig. 11) [39]. After 7 h of reaction time, the tail is more evident, and
its BE value of 286.5 eV is typical of alcohol/ether (COX), thus pointing out the
existence of organic deposits derived from carbohydrates [40]. However, a small
contribution of leaching of sulfonic groups cannot be completely ruled out.
4. Conclusions
This study, focused on the dehydration of sorbitol to isosorbide, under
heterogeneous conditions, has demonstrated that by using a Purolite CT269, a
maximum yield of 75 % can be attained after 12 h at 413 K, by using melt sorbitol. This
resin possesses a high acid capacity (4.1 eq kg-1), which, together with its mechanical
15

and thermal stability, explain its excellent catalytic performance. The leaching of
sulfonic groups is almost negligible, and, most importantly, this resin can be reused for
four catalytic runs. The kinetic study has evidenced the sequential dehydration process
and degradation of isosorbide at longer reaction times.
Acknowledgements
The authors are grateful to financial support from the Spanish Ministry of
Economy and Competitiveness (CTQ2015-64226-C3-3-R project), Junta de Andalucía
(RNM-1565) and FEDER (European Union) funds. We also acknowledge the company
Purolite® for Purolite resins.
References
[1] M.J. Climent, A. Corma, S. Iborra, Green Chem.16 (2014) 516-547.
[2] D. Martin-Alonso, J.Q. Bond, J.A. Dumesic, Green Chem. 212 (2010) 1493–1513.
[3] P. Gallezot, Chem. Soc. Rev. 41 (2012) 1538–1558.
[4] B. Kamm, P. R. Gruber, M. Kamm, Biorefineries - Industrial processes and
products: Status Quo and Future Directions, Wiley-VCH Verlag GmbH & Co.,
Weinheim, 2006.
[5] R. Rinaldi, F. Schuth, ChemSusChem 2 (2009) 1096-1100.
[6] Ullmann's Encyclopedia of Industrial Chemistry, Wiley-VCH Verlag GmbH & Co,
KGaA, Weinheim (2012).
[7] J.J. Bozell, G.R. Petersen, Green Chem. 12 (2010) 539.
[8] T. Werpy, G. Petersen, Top Value Added Chemicals from Biomass, Volume I,
Results of Screening for Potential Candidates from Sugar and Synthesis Gas, US
Department of Energy DOE/ GO-102004-1992, August 2004.
[9] M. Rose, R. Palkovits, ChemSusChem 5 (2012) 167-176.
[10] K.K. Bhatia, US6407266 B2 (2002).
16

[11] L.A. Hartmann, US3484459 (1969).
[12] M.A. Andrews, K.K. Bhatia, P.J. Fagan, WO2001092266 A2 (2001).
[13] A. Liu, C. Luckett, EP2146998 A1 (2010).
[14] H. Kobayashi, H. Yokoyama, B. Feng, A. Fukuoka, Green Chem. 17 (2015) 2732-
2735.
[15] R. Otomo, T. Yokoi, T. Tatsumi, Appl. Catal. A 505 (2015) 28-35.
[16] M. Gu, D. Yu, H. Zhang, P. Sun, H. Huang, Catal. Lett. 133 (2009) 214-220.
[17] M. Ibert, H. Wyart, W. Hölderich, O.A. Rusu, WO2014023789 A1 (2014).
[18] O.A. Rusu, W.F. Hoelderich, H. Wyart, M. Ibert, Appl. Catal. B 176-177 (2015)
139-149.
[19] P. Sun, D.H. Yu, Y. Hu, Z.C. Tang, J.J. Xia, H. Li, H. Huang, Korean J. Chem.
Eng. 28 (2011) 99-105.
[20] J.U. Oltmanns, S. Palkovits, R. Palkovits, Appl. Catal. A 456 (2013) 168-173.
[21] I. Ahmed, N.A. Khan, D.K. Mishra, J.S. Lee, J.-S. Hwang, S.H. Jhung, Chem. Eng.
Sc., 2013, 93, 91-95.
[22] J. Xia, D. Yu, Y. Hu, B. Zou, P. Sun, H. Li, H. Huang, Catal. Commun. 12 (2011)
544-547.
[23] A.A. Dabbawala, D.K. Mishra, J.-S. Hwang, Catal. Commun. 42 (2013) 1-5.
[24] N.A. Khan, D.K. Mishra, I. Ahmed, J.W. Yoon, J.-S. Hwang, S.H. Jhung, Appl.
Catal. A 452 (2013) 34-48.
[25] J. Xi, Y. Zhang, D. Ding, Q. Xia, J. Wang, X. Liu, G. Lu, Y. Wang, Appl. Catal. A
469 (2014) 108-115.
[26] X. Zhang, D. Yu, J. Zhao, W. Zhang, Y. Dong, H. Huang, Catal. Commun. 43
(2014) 29-33.
17

[27] A.A. Dabbawala, D.K. Mishra, G.W. Huber, J.-S. Hwang, Appl. Catal. A 492
(2015) 252-261.
[28] A. Chakrabarti, M.M. Sharma, Reactive Polymers 20 (1993) 1-45.
[29] K.M. Moore, A.J. Sanborn, US6848748 B2 (2005).
[30] J. Hu, J.E. Holladay, X. Zhang, Y. Wang, US20070173653 A1 (2007).
[31] I. Polaert, M. Felix, M. Fornasero, S. Marcotte, J. Buvat, L. Estel, Chem. Eng. J.
222 (2013) 228-239.
[32] J. Zhang, L. Wang, F. Liu, X. Meng, J. Mao, F. Xiao, Catal. Today 242 (2015)
249-254.
[33] A. Yamaguchi, N. Hiyoshi, O. Sato, M. Shirai, Green Chem. 13 (2011) 873-881.
[34] M. Kuzminska, R. Backov, E. Gaigneaux, Appl. Catal. A 504 (2015) 11-16.
[35] N.A. Khan, D.K. Mishra, J.-S. Hwang, Y.-W. Kwak, S.H. Jhung, Res. Chem.
Intermed. 37 (2011) 1231-1238.
[36] A. Cubo, J. Iglesias, G. Morales, J.A. Melero, J. Moreno, R. Sánchez-Vázquez,
Appl. Catal. A 531 (2017) 151-160.
[37] J. Li, W. Buijs, R. J. Berger, J. A. Moulijn, M. Makkee, Catal. Sci. Technol. 4
(2014) 152-163.
[38] J. Shi, Y. Shan, Y. Tian, Y. Wan, Y. Zheng, Y. Feng, RSC Advances 6 (2016)
13514-13521.
[39] C.D. Wagner, W.M. Riggs, L.E. Davies, J.F. Moulder, G.E. Muilenberg,
Handbook of X-ray Photoelectron Spectroscopy, Perkin-Elmer (1979).
[40] M. Dhayal, D.M. Ratner, Langmuir 25 (2009) 2181-2187.
18

List of captions.
Figure 1. Comparison of the catalytic performance of commercial sulfonic resins in the
dehydration of sorbitol at 413 K for 90 min (Experimental conditions: 2 g sorbitol and
100 mg catalyst).
Figure 2. Catalytic behavior in the dehydration of sorbitol at 393 K for 90 min of
Purolites (Experimental conditions: 2 g sorbitol and 100 mg catalyst).
Figure 3. DTA-TG curves of P-PD206 and P-CT269 resins.
Figure 4. SEM micrographs of Purolite resins, before and after the catalytic test.
Figure 5. Influence of the reaction temperature on the sorbitol dehydration over P-
CT269 (Experimental conditions: 2 g sorbitol, 100 mg catalyst, time= 90 min).
Figure 6. Catalytic data as a function of the reaction time over P-CT269 (Experimental
conditions: 2 g sorbitol, 100 mg catalyst, T= 413 K).
Figure 7. Kinetic study of the sorbitol dehydration over P-CT269 (Experimental
conditions: 2 g sorbitol, 100 mg catalyst, T= 413 K).
Figure 8. Study of the catalyst loading of P-CT269 in sorbitol dehydration
(Experimental conditions: 2 g sorbitol, T= 413 K, time= 90 min).
Figure 9. Study of the particle size of P-CT269 in sorbitol dehydration (Experimental
conditions: 2 g sorbitol, T= 413 K, time= 90 min).
Figure 10. Reutilization of P-CT269 (Experimental conditions: 2 g sorbitol, 100 mg
catalyst, T= 413 K, time= 90 min).
Figure 11. XPS spectra in the C 1s region of P-CT269, fresh and after 7 h of reaction.
19

C
Siso
Yiso
100
90
80
70
60
50
40
30
20
10
0
A-35
A-35-d
A-70
A-70-d
P-CT269
P-CT275
Catalyst
Figure 1
20

100
80
60
40
20
0
C
Siso
Yiso
P-CT269
P-CT275
P-PD206
Figure 2
21

100
75
100
75
50
25
0
PD206
CT269DR
50
25
0
-25
-50
-75
300
450
600
750
900
1050
1200
K
Figure 3
22

Fresh P-PD206
Fresh P-CT269
Fresh P-CT275
Used P-CT275 (90 min)
Used P-CT269 (90 min)
Used P-PD206 (90 min)
Used P-CT275 (10 h)
Used P-PD206 (10 h)
Used P-CT269 (10 h)
Figure 4
23

100
90
80
70
60
50
40
30
20
10
0
C
Siso
Yiso
373
393
413
433
Temperature [K]
Figure 5
24

100
90
80
70
60
50
40
30
20
10
0
CSORB
SISO
YISO
S1,4-AHSO
Y1,4-AHSO
0
10
20
30
40
50
Time [h]
Figure 6
25

Figure 7
26

100
90
80
70
60
50
40
30
20
10
0
C
Siso
Yiso
0
25
50
75
100
125
150
175
200
225
catalyst loading (mg)
Figure 8
27

100
75
50
25
0
C
Siso
Yiso
[0,4-0,5]
[0,5-0,7]
[0,7-1,0]
[1,0-1,2]
Particle size [mm]
Figure 9
28

100
90
80
70
60
50
40
30
20
10
0
C
Siso
Yiso
Cycle 1
Cycle 2
Cycle 3
Cycle 4
Figure 10
29