See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/244551421

Catalytic dehydration of fructose into 5-hydroxymethylfurfural by ion-

exchange resin in mixed-aqueous system by microwave heating

Article in Green Chemistry · July 2008

DOI: 10.1039/b801641k

CITATIONS READS

349 887

3 authors:

Xinhua Qi Masaru Watanabe

Nankai University Tohoku University
125 PUBLICATIONS 6,691 CITATIONS 113 PUBLICATIONS 8,272 CITATIONS

SEE PROFILE SEE PROFILE

Taku Michael Aida

Fukuoka University
67 PUBLICATIONS 4,066 CITATIONS

SEE PROFILE

All content following this page was uploaded by Xinhua Qi on 15 June 2014.

The user has requested enhancement of the downloaded file.

PAPER www.rsc.org/greenchem | Green Chemistry

Catalytic dehydration of fructose into 5-hydroxymethylfurfural by

ion-exchange resin in mixed-aqueous system by microwave heating†
Xinhua Qi,a,b Masaru Watanabe,*a Taku M. Aidaa and Richard Lee Smith, Jr.*a
Received 30th January 2008, Accepted 24th April 2008
First published as an Advance Article on the web 13th June 2008
DOI: 10.1039/b801641k

Catalytic dehydration of fructose into 5-hydroxymethylfurfural by microwave heating was studied

in acetone-water mixtures in the presence of a cation exchange resin catalyst. The use of
acetone–water reaction media resulted in yields of 5-HMF as high as 73.4% for 94% conversion at
150 ◦ C. It was confirmed that there was no decrease of catalytic activity and selectivity for five
reuses of the resin, which was in accordance with the elemental analysis results that showed that
sulfonic acid groups attached on the resin were stable at the experimental conditions. A
comparison between conventional sand bath heating and microwave heating revealed that the
latter had a remarkable accelerating effect not only on fructose conversion, but also on 5-HMF
yield. Under the conditions (5 ml of 2 wt% fructose solution, 0.1 g of resin, 150 ◦ C and 10 min),
fructose conversion and HMF yields by microwave heating (91.7% and 70.3%, respectively) were
higher than those by sand bath heating (22.1% and 13.9% respectively). Therefore, the process that
we developed in this study showed that high 5-HMF yields from fructose could be achieved under
mild conditions.

1. Introduction
Diminishing fossil fuel reserves and growing concerns about
global warming indicate that renewable sources of energy and
chemicals are needed in the near future. Abundant biomass
resources are a promising alternative for the sustainable supply
of fuel and valuable chemicals.1 Among the many possible
biomass-derived chemicals, 5-hydroxymethylfurfural (5-HMF)
is a valuable intermediate for fine chemicals, pharmaceuticals
and furane-based polymers,2 and has been called a “sleeping
giant” as an intermediate.3 The study of acid-catalyzed de-
hydration of fructose or fructose-precursors into 5-HMF has
received considerable attention,1,4–11 and its chemistry is shown in
Scheme 1. The detailed mechanism of the reaction was proposed
by Antal et al.12
This process has been widely studied by the other authors,
where many types of acid catalysts have been used, such as
Scheme 1 Possible reaction products of the acid catalyzed dehydration
mineral acids (such as H2 SO4 , HCl, H3 PO4 ),1,3–5,13,14 organic
and subsequent rehydration of fructose and related compounds.12,13
acids (such as oxalic acid and levulinic acid),4 H-form zeolites,15
transition metal ions,11,16–18 solid metal phosphates19–22 and but it has serious drawbacks in terms of separation and recycling
strong acid cation exchange resins.1,9,13,23,24 Homogenous acid as well as equipment corrosion. Solid acid catalysts, such as H-
catalyzed processes are effective and can achieve medium 5- form zeolites and metal phosphates can be recycled and have
HMF yields (40–60%) with high fructose conversion (70–90%),4 high selectivity (60–90%), but only low fructose conversion (30–
60%) even at reaction times as long as 2 h.20,21 An ion exchange
a
resin could be a candidate as a green catalyst for 5-HMF
Research Center of Supercritical Fluid Technology, Tohoku University,
6-6-11 Aoba, Aramaki, Aoba-ku, Sendai, 980-8579, Japan. formation, however, only limited studies have been performed
E-mail: [email protected], [email protected]; since resins are considered to be useful below 100 ◦ C, which can
Fax: (+81) 022-795-5864 greatly limit the reaction rate.25
b
College of Environmental Science and Engineering, Nankai University, Solvents used as reaction media can be divided into four
Tianjin, 300071, China
main groups: reactions in water,4,15,17–21,26,27 reactions in or-
† Electronic supplementary information (ESI) available: Microwave
heating experimental setup (Fig. S1), typical work-up procedure and ganic solvents or organic–water mixtures,1,5–7,16 reactions in
a typical heating profile (Fig. S2). See DOI: 10.1039/b801641k ionic liquids,9–11,28,29 and more recently, biphasic water/organic

This journal is © The Royal Society of Chemistry 2008 Green Chem., 2008, 10, 799–805 | 799
systems.1,13,14,24 Aqueous processes are favored in respect to 2. Experimental
ecological aspects but are unfortunately, inefficient, because 5-
HMF further rehydrates into levulinic acid and formic acid in 2.1 Material
water. Roman–Leshkov et al.14 studied acid-catalyzed fructose Fructose (purity: 99%) and acetone were purchased from Wako
dehydration in a two-phase reactor system, in which fructose Pure Chemical Company and used without further purification.
was dehydrated in the aqueous phase with dimethylsulfoxide Dowex 50wx8-100 ion-exchange resin was purchased from
(DMSO) and poly(1-vinyl-2-pyrrolidinone) (PVP) was added Sigma–Aldrich Corporation. Pure water (with a conductivity
to suppress undesired side reactions. The 5-HMF product was of 18 MX cm−1 ), which was distilled after deionization, was
continuously extracted into an organic phase (methylisobutylke- obtained by a water distillation apparatus (Yamato Co., Model
tone) modified with 2-butanol to enhance partitioning from the WG-220).
reactive aqueous solution. A maximum 5-HMF selectivity of
85% with 89% fructose conversion was obtained. Although a 2.2 Resin properties
high 5-HMF yield could be achieved in this system, the product
had to be separated from high boiling point solvents such as The strong acid cation exchange resin Dowex 50wx8-100 (50-
DMSO from both the reaction phase and extraction phase. 100 mesh beads, gel, water content: 40–70%) that is insoluble in
Zhao et al.11 studied the catalytic conversion of fructose into water, consists of a sulfonated copolymer of styrene and divinyl
5-HMF in an ionic liquid solvent (1-alkyl-3-methylimidazolium benzene in the hydrogen form. The matrix of the resin was
chloride) with metal halides such as chromium(II) chloride as styrene–DVB, and total exchange capacity was 1.7 meq ml−1
catalysts, in which a yield of 73% of 5-HMF was achieved at a (H+ ).
temperature of 120 ◦ C for a reaction time of 3 h. Ionic liquids are
advanced solvents in view of controllability of its properties, but 2.3 Typical work-up procedure
many of these are too expensive to allow their use on a practical
The microwave reaction apparatus and experimental procedure
scale.
are described in the ESI.† Briefly, a solution (5 g) of fructose
Within the past decade, green chemistry has attained the
and a given amount of the catalyst were loaded into a thick-wall
status of a major scientific discipline. Among the 12 principles
Pyrex glass tube (volume 10 ml, id: 11.6 mm, od: 18 mm, wall
of green chemistry, the need for using “safe solvents” and for
thickness: 3.2 mm, and maximum supporting pressure 10 MPa).
“energy efficiency” can be considered as two key principles.30 In
The glass tube was mounted into a polycarbonate (PC) tube that
this respect, it has been shown that microwave-assisted organic
was closed with PEEK screw caps. This assembly was placed
syntheses (MAOS) have much higher yield and selectivity for
into the microwave oven. N2 gas was used for purging air inside
a given reaction time for many types of reactions.30 A study
the reactor at a pressure of about 1.2 MPa, which would give
comparing the energy efficiency between a conventional oil
calculated gas pressures of 1.73 MPa and 1.86 MPa assuming
bath organic synthesis and a MAOS has indicated that a
ideal gas at the given temperatures of 150 and 180 ◦ C. The
significant energy savings (up to 85-fold) can be expected for
use of N2 gas pressure increase the boiling point of the solvent
most chemical transformations using microwaves as an energy
mixtures and reduces volatilization. According to calculation
source on a laboratory scale.31 If those data can be applied to
with UNIFAC, the solvent mixtures will not boil below 150 ◦ C
larger scale microwave applications on a pilot or production
under the given pressures, and the amount of acetone volatilizing
scale, the possibility of performing reactions in a very short
into the vapor would cause liquid phase composition changes of
time period by direct interaction of microwave energy with
5% maximum. Boiling of the solvent mixtures was not observed
the reaction mixture could certainly be considered “green” and
during all of the experiments. When the microwave irradiation
highly efficient, because of the reduced energy consumption and
was started, the reaction mixture could be heated up to 150 ◦ C
the associated time savings.
within 30 s. In the reaction analyses, zero time was taken to
Up to now, most studies on MAOS have been applied to or-
be when the temperature reached 150 ◦ C. After the desired
ganic synthesis,30,32–35 and only a limited number of applications
reaction time passed, the microwave irradiation was turned off
have been studied in carbohydrate conversion36–41 related to non-
and the reactor was cooled down by loading cooling water. The
food products. Orozco et al.37 studied dilute acid hydrolysis of
reactor was taken down and the reaction solution was collected
grass and cellulose with phosphoric acid in a microwave reactor
by washing the glass tube with an amount of distilled water.
system and showed that high yields (90%) of glucose could be
For the sand bath experiments, a SS316 tube bomb reactor
obtained in short reaction times. Cottier and Descotes described
(inner volume: 6 cm3 ) as reactor was used, in which the
a microwave process that produced 5-HMF with 28% yield in
procedures for loading and recovering the samples were similar
aqueous fructose (or sucrose) mixed with inorganic phosphates
to those used in the microwave heating experiments, Previous
for 3 min reaction time.8
work27 and the ESI† provides additional information.
In this work, we report on microwave-assisted catalytic
dehydration of fructose into 5-HMF with a strong acid cation
2.4 Analysis
exchange resin as catalyst in a mixed organic–aqueous solvent
system. The resin is stated as being potentially dangerous at high An HPLC (Jasco) with SH 1011 column (Shodex) and refractive
temperatures with possible reaction products being aromatics, index detector (ERC-7571A) were employed to analyze the
hydrocarbons, organic sulfonates, and sulfur oxides. We exam- liquid samples. Each sample was diluted with ultra pure water
ined the applicability of the resin at high temperatures ranging before analysis to prevent the overloading of the column with
between 100 ◦ C and 180 ◦ C under microwave irradiation. organic solvents.

800 | Green Chem., 2008, 10, 799–805 This journal is © The Royal Society of Chemistry 2008
 Elemental analysis (C, H, S) of the resin before and after being and supercritical conditions in which a maximum selectivity of
used was carried out using Elementar Analysensysteme GmbH 78% with 90% of fructose conversion was obtained at 180 ◦ C
VarioEL (Germany). and 20 MPa with H2 SO4 in an acetone–water mixture (90 :
The fructose conversion (mol%), the product yield (mol%) 10 v/v) for 2 min of reaction time.5 More mild conditions would
and selectivity (mol%) of each carbon compound was evaluated be preferable. Therefore, solvent composition was changed
from carbon base as shown below: with acetone to replace some part of the water to study the
Fructose conversion (mol%): dehydration of fructose in acetone–water mixture using resin as
catalyst under microwave irradiation heating at 150 ◦ C and at
 Fructose concentration in product  the saturation pressure. The composition of the reaction solvent
X = 1 −  ×100%
 Fructose n
concentration in the loaded sample  was varied from 0 to 85 wt% acetone in water.
(1) Fig. 1 shows the influence of the reaction solvent composition
on the fructose conversion and 5-HMF yield. As the percentage
Product yield (mol%):
of acetone in the mixture was increased, the conversion rate and
Moles of carbon in product 5-HMF yield increased, and the 5-HMF yield reached 73.4%
Y= ×100% (2)
Moles of carbon loaded as frucctose in 70 wt% acetone mixture for a reaction time of 15 min at
150 ◦ C. The 5-HMF selectivity was as high as 83.7% for 5 min
Product selectivity (mol%): for the 70 wt% acetone mixture, but it reduced with increasing
Yield of product reaction time due to the rehydration of 5-HMF into levulinic
S= × 100% (3) acid and formic acid for 30 wt% water. Theoretically, the 5-HMF
Fructose conversion
is stable in pure acetone reaction media and the rehydration will
The weight ratio of substrate to ion exchange resin was defined not occur, but, because the solubility of fructose in acetone–
as R: water mixture decreases with increasing acetone content, the
solubility of fructose in pure acetone is as low as 0.5 g L−1 .3 When
Weight of substrate
R= (4) acetone–water is 90 : 10 (v/v), the fructose solubility increases
Weight of ion exchange resin
to 14 g L−1 (25 ◦ C).3 Therefore, an acetone–water mixture of 70 :
30 (w/w) was employed as reaction media hereafter.
3. Results and discussion
3.1 Products

In this study, the main product of the reaction of fructose was

always 5-HMF regardless of the experimental conditions. The
main by-products were levulinic acid (1 to 20% yield), formic
acid (0.2 to 2% yield), furfural (0.5 to 2% yield) and glucose
(0.2 to 0.6% yield). When the fructose initial concentration was
higher than 10%, there were also some polymer products that
were not characterized.

3.2 Influence of the reaction media composition on fructose

conversion and 5-HMF yield

The dehydration of fructose in water in the presence of the

resin was initially studied. For this case, the fructose conversion
reached 82.6% at a reaction time of 60 min, temperature 150 ◦ C
for R = 1, but the 5-HMF yield was only 34%. The low yield can
be attributed to the rehydration of 5-HMF into levulinic acid
and formic acid that occurs in aqueous and acidic solvents.2,19
The corresponding levulinic acid yield was 24%. To depress the
formation of levulinic acid, changing the reaction solvent was
considered.
DMSO has been proven to be an effective solvent that can
depress side-reactions in the fructose dehydration because the
furanoid form is preferred in DMSO. The dehydration of
fructose to form 5-HMF is thought to be the most selective if the
carbohydrate molecule is in its furanoid form,1,12,13 but DMSO
has some disadvantages due to its high boiling point (189 ◦ C).
On the other hand, acetone, has a low boiling point (56 ◦ C,
1 atm) and fructose has been shown to rearrange to the furanoid
form in acetone–water mixtures.3 Bicker et al. examined the Fig. 1 Influence of the reaction media composition on the fructose
fructose dehydration in a water–acetone mixture at subcritical conversion (a) and 5-HMF yield (b) (2 wt% fructose, 150 ◦ C, R = 1).

This journal is © The Royal Society of Chemistry 2008 Green Chem., 2008, 10, 799–805 | 801
3.3 Influence of reaction temperature on fructose conversion of the dehydration of fructose in acetone–water (70 : 30 w/w)
and 5-HMF yield was performed by plotting ln (1 − X) versus reaction time (t) to
obtained the rate constants at different temperatures (T). With
Fig. 2 shows the influence of the temperature with respect to the
those constants, an Arrhenius plot was generated, as shown
fructose conversion and 5-HMF yield. Reaction temperature
in Fig. 3. Activation energy determined was 103.4 kJ mol−1 .
had a large effect on the fructose conversion and 5-HMF
Bicker et al. reported an activation energy of 99 kJ mol−1 for the
yield. When reaction temperature was 100 ◦ C, the fructose
fructose dehydration with 3 mmol L−1 sulfuric acid as catalyst
conversion was only 14.5% for 60 min reaction time and when
at a pressure of 20 MPa,5 Antal et al. obtained 100 kJ mol−1 for
the temperature increased to 120 ◦ C, the fructose conversion
the dehydration of glucose under sulfuric acid concentration of
and 5-HMF yield increased to 59.1% and 51.6%, respectively,
5 mmol L−1 .12 The value of this work (E a = 103 kJ mol−1 ) was
for 60 min reaction time. Conversion of fructose and yield of 5-
comparable with previous values reported in the literature.
HMF reached 94.2% and 76%, respectively, for 180 min reaction
time. Levulinic acid yield was always below 5%. The fructose
conversion and 5-HMF yield was 95.1% and 73.4%, respectively,
at 150 ◦ C for 15 min, and the corresponding levulinic acid yield
was 5.7%. At 180 ◦ C, the rate of fructose conversion rate was
much higher, and fructose conversion reached 99.2% for 5 min,
but the 5-HMF yield was only 65.3% due to the further reaction
of 5-HMF and a higher yield of levulinic acid formed which
reached 11.3% and 17.7% at a reaction time of 5 and 10 min,
respectively. Thus, a reaction temperature of 150 ◦ C was chosen
as the best temperature since fructose conversions and 5-HMF
yields were high and levulinic acid yields were low.

Fig. 3 Determination of activation parameters from Arrhenius equa-

tion for the fructose dehydration.

3.5 Effect of the catalyst dosage on the fructose conversion and

5-HMF yield

Table 1 shows the effect of the catalyst dosage with respect

to fructose conversion, 5-HMF, levulinic acid and formic acid
yield. In the absence of catalyst, at a reaction temperature of
150 ◦ C, no fructose conversion was observed for a reaction time
of 20 min. On the other hand, when catalyst was used with
R-values of 0.5, 1.0 and 2.0, fructose conversion remarkably
increased, whereas the 5-HMF yield did not change much
for R-values below 1 for a reaction time of 10 min. The
reason of the small effect of R on the 5-HMF yield might
be the decomposition of 5-HMF being accelerated at high R-
values due to the availability of acidic sites that favour the
rehydration of 5-HMF into levulinic acid, which offsets increases
in the 5-HMF yield. Therefore, an R of 1 was chosen as an
appropriate catalyst concentration in subsequent experiments if
not otherwise indicated.

3.6 Influence of the fructose initial concentration on fructose

conversion and 5-HMF yield
Fig. 2 Influence of temperature on the (a) fructose conversion and (b) From a practical point of view, if higher concentrations of
HMF yield by microwave irradiation (acetone–water 70 : 30 w/w, 2 wt% fructose can be used as feedstock, the technique can be
fructose, R = 1).
economical. Different initial concentrations of fructose were
studied. The fructose concentrations that were studied were
2 wt%, 5 wt%, 10 wt% and 20 wt%. The reaction media was
3.4 Reaction kinetics analysis
acetone–water (70 : 30 w/w) and the reaction temperature was
Kuster25 and Bicker et al.5 reported that the dehydration of fruc- 150 ◦ C with the resin concentrations being given in Table 2. It can
tose was first-order on fructose concentration. Kinetic analysis be seen that the rate of fructose conversion had little effect for

802 | Green Chem., 2008, 10, 799–805 This journal is © The Royal Society of Chemistry 2008
Table 1 Influence of the weight ration of fructose to resin (R) on the fructose conversion and primary product yield under microwave irradiation
(2 wt% fructose, acetone-water 70 : 30 w/w, 150 ◦ C)

Product yield (%)

R (w/w) Reaction time/min Fructose conversion (%) Formic acid Levulinic acid 5-HMF

0.5 3 70.5 1.6 2.7 56.5

5 86.8 1.6 3.9 66.6
10 97.5 2.8 7.0 72.3
15 99.2 3.5 10.2 70.8
1.0 5 64.5 1.2 2.2 54.0
10 91.7 2.0 4.6 70.3
15 95.1 2.0 5.7 73.4
20 99.0 2.8 7.6 71.5
2.0 5 44.7 1.6 1.6 37.0
10 63.4 1.7 2.2 51.4
15 76.4 1.5 3.0 61.3
20 84.8 1.6 3.7 66.5

Table 2 Different initial concentration fructose dehydration to 5-HMF carried out in acetone-water medium (70 : 30 w/w) by Dowex 50wx8
ion-exchange resin a

Fructose concentration Rb (w/w) Reaction time/min Conversion (%) 5-HMF yield (%) TONc

2 wt% 1 5 64.5 54.0 36.0

10 91.7 70.3 23.4
15 95.1 73.4 16.3
20 99.0 71.5 11.9
5 wt% 1 3 82.7 60.0 66.7
5 93.6 68.5 45.7
7 95.5 66.9 31.9
10 98.6 66.6 22.2
10 wt% 1 3 91.8 61.1 67.9
5 98.2 61.5 41.0
10 99.6 52.7 17.6
15 99.8 44.7 9.9
20 wt% 2 3 81.9 49.2 109.3
5 89.3 54.3 72.4
10 98.1 51.5 34.4
15 99.3 44.2 19.6
a
Reaction conditions: temperature: 150 ◦ C. b Substrate to catalyst weight ratio. c Turnover number expressed as (mmol of 5-HMF per g of
catalyst × h).

different initial concentration of fructose, but the 5-HMF yield were obtained under milder conditions, recyclable catalyst and
reduced gradually with increasing initial fructose concentration. moderate temperatures with the present method.
The highest 5-HMF yield was 73.4% at 2 wt% of initial
fructose concentration, but the value reduced to 54.3% at 20 wt%
3.7 Resin catalyst recycle
initial concentration of fructose. The losses in 5-HMF yield are
thought to be due to the higher fructose concentrations, which Generally, the strong acid sulfonated copolymer resin is thought
would increase the probability that reactive compounds such to be unusable above 130 ◦ C,25 but our experimental results
as fructose and 5-HMF would collide with each other, cross- showed that the resin could work well above 130 ◦ C or even
polymerize and form humins.25 In aqueous systems, including 150 ◦ C. In green engineering, the recycling of catalyst is very
aqueous mixtures, losses due to humin formation can amount important in practice. Thus, the recycling of the resin was tried
up to 35% for 1 M fructose solution, and this value reduces for 5 times to study its change in activity (Fig. 4). From Fig. 4, the
to 20% for 0.25 M fructose solutions.25 These polymers were catalytic activity and selectivity of the resin for the dehydration
not analyzed in this work. In certain non-aqueous systems of fructose into 5-HMF did not show systematic decrease after 5
or biphasic water/organic systems, few problems occur with uses at 150 ◦ C for 10 min reaction time. The fructose conversions
polymerizations at high concentrations. Roman-Leshkov et al. and 5-HMF yields were always about 98% and 72%, respectively.
used a two-phase reactor system (7 : 3 water : PVP + 7 : 3 MIBK : If the resin had been decomposed from exposure to the
2-butanol) and obtained a fructose conversion of 90% with 77% experimental conditions, the sulfonic acid groups attached on
of 5-HMF selectivity for 50 wt% initial fructose concentration.1 the resin would probably be released and dissolved into solution,
Some other investigations for fructose conversion and 5-HMF and the sulfur content of the resin would decrease. To examine
yield with different initial fructose concentrations using different the decomposition of the resin after being reused 5 times, the
methods by other authors are listed in Table 3. Comparing these used resin was collected, rinsed with pure water and acetone, and
methods, high 5-HMF yields with high fructose conversions dried for 24 h in vacuum drying oven, and in the same manner

This journal is © The Royal Society of Chemistry 2008 Green Chem., 2008, 10, 799–805 | 803
Table 3 Fructose conversion (X) and 5-HMF yield (Y ) with different initial fructose concentrations using different methods by the other researchers

Solvent system Catalyst Fructose concentration (wt%) Reaction temperature/◦ C Reaction time/min X (%) Y (%) Ref.

H2 O H3 PO4 0.9 240 2 98.0 65.3 4

H2 O–MIBK Zeolite 1.7 165 60 76.0 69.2 15
H2 O NbOPO4 6.0 100 120 61.4 21.6 19
H2 O–DMSO/PVP IERa 10 90 480 76.0 58.5 1
Ionic Liquid PtCl2 10 80 180 98.0 83.0 11
H2 O–salt/butanol HCl 30 180 3 74.0 65.9 14
Acetone–water IER 2.0 150 15 95.1 73.4 This work
a
Ion exchange resin PK-216.

Fig. 4 Catalyst recycle (2 wt% fructose, reaction media: acetone–water

70 : 30 w/w, reaction time: 10 min; temperature: 150 ◦ C; R = 0.5).

that was used for new resin. Then, elemental analysis (C, H, S)
of both of these resins was carried out as shown in Table 4. It can
be seen that the C and H content in the reused resin increased by
about 4.6% and 0.94%, respectively over that of new resin, and
S content reduced by about 1.3%. The increasing in C and H is
probably due to adsorption of some reaction products such as
humins on the resin, and humins did not seem to be able to be
washed from the resin after reaction. It seems that the S content
in the reused resin reduced by about 1.3%, but if we take into
account that the humins adsorbed on the resin and increased the
C and H content, the S content decreasing in the resin matrix
would be lower. Therefore, it can be concluded that the resin was
stable in this system and the nearly constant fructose conversions
and 5-HMF yields verified this conclusion. Fig. 5 Comparison of (a) fructose conversion and (b) HMF yield by
sand bath and microwave heating (reaction media: acetone–water 70 :
3.8 Comparison of D-fructose dehydration by sand bath and 30 w/w, 2 wt% fructose solution, 150 ◦ C, R = 1).
microwave heating
was remarkably more efficient not only for fructose conversion,
The dehydration of D-fructose (2 wt%) in an acetone–water but also in achieving high 5-HMF yields. When the reaction
mixture (70 : 30, w/w) with the resin as catalyst by convective mixture was heated by sand bath, the fructose conversion
heating (sand bath) and microwave irradiation heating was and 5-HMF yield was 22.1% and 13.7%, respectively, at a
studied (Fig. 5). As shown in Fig. 5, microwave irradiation reaction time of 10 min, while the corresponding value for
microwave heating was 91.7% and 70.3%, respectively. These
Table 4 Elemental analysis results (C, H, S) of resin before and after
being reused for 5 times kind of difficult-to-rationalize effects have been referred to as
“specific” or “nonthermal” microwave effects.35 Generally, this
Element C (%) H (%) S (%) specific microwave effect has been proposed to be the result
from a direct interaction of the electric field with specific
New resin 42.5 5.38 13.7
molecules in the reaction medium. It has been argued that
Used resin 47.2 6.32 12.4
there is a decrease in activation energy or an increase in the

804 | Green Chem., 2008, 10, 799–805 This journal is © The Royal Society of Chemistry 2008
pre-exponential factor in the Arrhenius law due to orientation 10 C. Moreau, R. Durand, A. Roux and D. Tichit, Appl. Catal., A, 2000,
effect of polar species in an electromagnetic field.34,42 Some 193, 257–264.
11 H. B. Zhao, J. E. Holladay, H. Brown and Z. C. Zhang, Science,
researchers have proposed that the effective collision among 2007, 316, 1597–1600.
reactant molecules under microwave irradiation is enhanced 12 M. J. Antal, W. S. L. Mok and G. N. Richards, Carbohydr. Res., 1990,
and result in the pre-exponential factor in the Arrhenius law 199, 91–109.
increases.34 Furthermore, a similar effect should be observed 13 J. N. Chheda, Y. Roman-Leshkov and J. A. Dumesic, Green Chem.,
2007, 9, 342–350.
for reaction mechanisms where reactants contain polar groups, 14 Y. Roman-Leshkov, C. J. Barrett, Z. Y. Liu and J. A. Dumesic, Nature,
where the rotation of polar species increases and leads to 2007, 447, 982–U985.
molecules going from the ground state to the transition state 15 C. Moreau, R. Durand, S. Razigade, J. Duhamet, P. Faugeras, P.
Rivalier, P. Ros and G. Avignon, Appl. Catal., A, 1996, 145, 211–224.
under the microwave irradiation, thus enhance the reactivity by 16 K. Seri, Y. Inoue and H. Ishida, Chem. Lett., 2000, 22–23.
lowering the activation energy.35 Currently microwave effects are 17 K. Seri, Y. Inoue and H. Ishida, Bull. Chem. Soc. Jpn., 2001, 74,
still the subjects of investigation in many fields of research. 1145–1150.
18 H. Ishida and K. Seri, J. Mol. Catal. A: Chem., 1996, 112, L163–
L165.
4. Conclusions 19 T. Armaroli, G. Busca, C. Carlini, M. Giuttari, A. M. R. Galletti and
G. Sbrana, J. Mol. Catal. A: Chem., 2000, 151, 233–243.
Microwave-assisted dehydration of fructose into 5-HMF using 20 C. Carlini, M. Giuttari, A. M. R. Galletti, G. Sbrana, T. Armaroli
a strong acid cation exchange resin as catalyst in aqueous- and G. Busca, Appl. Catal., A, 1999, 183, 295–302.
21 C. Carlini, P. Patrono, A. M. R. Galletti and G. Sbrana, Appl. Catal.,
acetone mixtures was investigated. The fructose conversion
A, 2004, 275, 111–118.
rate improved with an increase in acetone concentration in the 22 F. Benvenuti, C. Carlini, P. Patrono, A. M. R. Galletti, G. Sbrana,
acetone–water mixtures, and an increase in the 5-HMF yield was M. A. Massucci and P. Galli, Appl. Catal., A, 2000, 193, 147–
also obtained due to the suppression of 5-HMF dehydration. 153.
23 Y. Nakamura and S. Morikawa, Bull. Chem. Soc. Jpn., 1980, 53,
At 20 wt% initial concentration of fructose, the 5-HMF yield 3705–3706.
was 54.3% with 89.3% fructose conversion due to the soluble 24 L. Rigal, A. Gaset and J. P. Gorrichon, Ind. Eng. Chem. Prod. Res.
polymer formation. The resin could work well at temperatures Dev., 1981, 20, 719–721.
up to 150 ◦ C, and there was no catalytic activity decrease after 25 B. F. M. Kuster, Starch, 1990, 42, 314–321.
26 M. Watanabe, Y. Aizawa, T. Iida, T. M. Aida, C. Levy, K. Sue and
5 times reuse. Compared with conventional sand bath heating, H. Inomata, Carbohydr. Res., 2005, 340, 1925–1930.
microwave irradiation heating had a remarkable accelerating 27 M. Watanabe, Y. Aizawa, T. Iida, R. Nishimura and H. Inomata,
effect, not only on the fructose conversion, but also on the Appl. Catal., A, 2005, 295, 150–156.
28 C. Moreau, A. Finiels and L. Vanoye, J. Mol. Catal. A: Chem., 2006,
5-HMF yield. In conclusion, fructose can be converted to 5- 253, 165–169.
HMF in high yields at mild conditions with microwave heating 29 S. K. Tyrlik, D. Szerszen, M. Olejnik and W. Danikiewicz, Carbohydr.
in acetone–water mixtures in the presence of a strong acid Res., 1999, 315, 268–272.
ion exchange resin. An examination on how to avoid polymer 30 D. Dallinger and C. O. Kappe, Chem. Rev., 2007, 107, 2563–2591.
31 M. J. Gronnow, R. J. White, J. H. Clark and D. J. Macquarrie, Org.
formation and how to improve 5-HMF yield for high initial Process Res. Dev., 2005, 9, 516–518.
concentrations (>10 wt%) of fructose is being carried out. 32 E. Comer and M. G. Organ, J. Am. Chem. Soc., 2005, 127, 8160–
8167.
33 T. N. Glasnov and C. O. Kappe, Macromol. Rapid Commun., 2007,
References 28, 395–410.
34 M. Hosseini, N. Stiasni, V. Barbieri and C. O. Kappe, J. Org. Chem.,
1 Y. Roman-Leshkov, J. N. Chheda and J. A. Dumesic, Science, 2006, 2007, 72, 1417–1424.
312, 1933–1937. 35 C. O. Kappe, Angew. Chem., Int. Ed., 2004, 43, 6250–6284.
2 J. Lewkowski, Arkivoc, 2001, 2, 17–54. 36 C. Limousin, J. Cleophax, A. Petit, A. Loupy and G. Lukacs,
3 M. Bicker, D. Kaiser, L. Ott and H. Vogel, J. Supercrit. Fluids, 2005, J. Carbohydr. Chem., 1997, 16, 327–342.
36, 118–126. 37 A. Orozco, M. Ahmad, D. Rooney and G. Walker, Process Safety
4 F. S. Asghari and H. Yoshida, Ind. Eng. Chem. Res., 2006, 45, 2163– Environ., 2007, 85, 446–449.
2173. 38 A. M. Sarotti, R. A. Spanevello and A. G. Suarez, Green Chem.,
5 M. Bicker, J. Hirth and H. Vogel, Green Chem., 2003, 5, 280–284. 2007, 9, 1137–1140.
6 M. Bicker, S. Endres, L. Ott and H. Vogel, J. Mol. Catal. A: Chem., 39 A. Corsaro, U. Chiacchio, V. Pistara and G. Romeo, Curr. Org.
2005, 239, 151–157. Chem., 2004, 8, 511–538.
7 G. W. Huber, J. N. Chheda, C. J. Barrett and J. A. Dumesic, Science, 40 M. R. Couri, I. Luduvico, L. Santos, R. Alves, M. A. Prado and
2005, 308, 1446–1450. R. F. Gil, Carbohydr. Res., 2007, 342, 1096–1100.
8 L. Cottier and G. Descotes, Trends Heterocyclic Chem., 1991, 233– 41 M. M. Andrade, M. T. Barros and P. Rodrigues, Eur. J. Org. Chem.,
248. 2007, 3655–3668.
9 C. Lansalot-Matras and C. Moreau, Catal. Commun., 2003, 4, 517– 42 C. O. Kappe and A. Stadler, Microwaves in Organic and Medicinal
520. Chemistry, Wiley-VCH, Weinheim, Germany, 2005.

This journal is © The Royal Society of Chemistry 2008 Green Chem., 2008, 10, 799–805 | 805
View publication stats