Conversion of xylose, xylan and rice husk into furfural via betaine and formic acid mixture as novel homogeneous catalyst in biphasic system by microwave-assisted dehydration

Article abstract of DOI:10.1016/j.molcata.2016.07.003

Dehydration of D-xylose and direct transformation of xylan into furfural were achieved by means of betaine-formic acid (HCOOH) catalytic system. All reactions were microwave-assisted and carried out in a CPME-water biphasic system. At 170?°C, in a pH range between 1.9 and 2.3, highest yields of 80% and 76% were obtained respectively for the pentose and the polysaccharide. Time dependence of the dehydration and influence of the temperature on the reaction kinetics were studied. Besides, at 190?°C, using the optimized condition of the reaction, rice husk was also employed as a source of furfural with a single stage reaction.

Full text of DOI:10.1016/j.molcata.2016.07.003

Accepted Manuscript
Title: Conversion of xylose, xylan and rice husk into furfural
via betaine and formic acid mixture as novel homogeneous
catalyst in biphasic system by microwave-assisted dehydration
Author: Frederic Delbecq Yantao Wang Christophe Len
PII:
S1381-1169(16)30253-9
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.molcata.2016.07.003
MOLCAA 9937
To appear in:
Journal of Molecular Catalysis A: Chemical
Received date:
Revised date:
Accepted date:
23-5-2016
1-7-2016
2-7-2016
Please cite this article as: Frederic Delbecq, Yantao Wang, Christophe Len, Conversion
of xylose, xylan and rice husk into furfural via betaine and formic acid mixture as novel
homogeneous catalyst in biphasic system by microwave-assisted dehydration, Journal
of Molecular Catalysis A: Chemical http://dx.doi.org/10.1016/j.molcata.2016.07.003
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.

Highlights
-xylose to furfural.
Water-CPME as biphasic system to limit the production of humin.
Betaine-formic acid mixture is an efficient acid catalyst system for the synthesis of
furfural.
reusability of the catalyst was efficient during ten cycles.
Application to xylan and rice husk gave the furfural in good yields.

O
CPME
O
water
furfural (80%)
OH
O
HO
HO
betaine, formic acid
OH
xylose
OH
170°C, 20 bars
60min, MW

Conversion of xylose, xylan and rice husk into furfural via betaine and formic acid
mixture as novel homogeneous catalyst in biphasic system by microwave-assisted
dehydration
Frederic Delbecqa, Yantao Wangb, Christophe Lenb,*
[a] Ecole Supérieure de Chimie Organique et Minérale (ESCOM), 1 allée du Réseau
Jean-Marie Buckmaster ; 60200 Compiègne, France
[b] Sorbonne Universités, Université de Technologie de Compiègne (UTC), Centre de
Recherches Royallieu, CS 60319, 60203 Compiègne Cedex, France
Corresponding author. +33 344 238 828
E-mail address: christophe.len@utc.fr (C. Len).
Abstract: Dehydration of D-xylose and direct transformation of xylan into furfural were
achieved by means of betaine-formic acid (HCOOH) catalytic system. All reactions were
microwave-assisted and carried out in a CPME-water biphasic system. At 170°C, in a pH
range between 1.9 and 2.3, highest yields of 80 % and 76 % were obtained respectively
for the pentose and the polysaccharide. Time dependence of the dehydration and
influence of the temperature on the reaction kinetics were studied. Besides, at 190°C,
using the optimized condition of the reaction, rice husk was also employed as a source of
furfural with a single stage reaction.

Keywords: Furfural, biphasic system, microwave-assisted dehydration, xylan, rice husk.
1. Introduction
Furfural or furan-2-carbaldehyde is a compound easily produced from xylose or
directly from the hemicellulosic biomass dehydration. Due to its composition, rich in
D-xylose, xylan should also become a good model for the direct conversion into furfural
before starting the experiments with xylose-rich feedstock. Furfural is now recognized as
a top-value bio-based chemical and expresses good potential as a selective solvent for
diesel fuels, resins and lubricating oils. It could be also employed as precursor in the
chemical industry for the preparation of various solvents such as tetrahydrofuran (THF),
methyltetrahydrofuran (MeTHF), but also for the transformation into furfuryl alcohol, a
well-known commercial available monomer. Its industrial production is currently carried
out by using strong mineral acid such as HCl [1,2] or H2SO4 [3] in batch or continuous
reactions against pentosan containing lignocellulosic biomass. However, these mineral
acids are responsible for serious corrosion, safety problems, difficulty for separation and
excessive waste, especially in monophasic systems, when the reactions are carried out in
water only. In this case the product requires being distillated progressively during the
process in order to avoid the generation of black insoluble carbonaceous material called
humin. Furfural recovery is also clearly a cause of high energy consumption. To remedy
to this problem and for improving the yield of the formed furfural, the target compound
has to be extracted as soon as possible with a non-miscible hydrophobic solvent such as
methylisobutylketone (MIBK), toluene or nitrotoluene. Thus, biphasic system

demonstrated potential for improving the selectivity of furfural. Recently, De Oliveira
Vigier et al. developped a strategy to produce 5-hydroxymethylfurfural (HMF) from
D-fructose using betaine hydrochloride catalyst [4]. HMF was obtained under
conventional heating with a maximum yield of 84% using an atypical biphasic system
consisting of choline chloride, a deep eutectic solvent and MIBK as the active extraction
solvent. On the other hand, other betaine based acid ionic liquids (ILs) obtained by
equimolar addition of inorganic acid to the N-base material, have already proved their
efficiency to promote the esterification of organic materials [5]. Currently, zwitterionic
glycine betaine or N, N, N-trimethylglycine is widely produced in the carbohydrate
industry and, in our opinion, it appeared to be interesting to determine both impact on the
sugar dehydration and synergistic effect of an IL immerged in a Bronsted acid. Besides,
in the literature, several works reported the employ of formic acid to accomplish
dehydration of the pentose [6-8]. Moreover formic acid is an organic acid and now it is
easily recovered from reaction by modern extraction methods. Unfortunately, if the
reaction is realized in water without addition of extraction solvent, the
cross-polymerization currently observed between D-xylose and furfural during the
process limits the furfural selectivity. For our study, an eco-friendly solvent, the
cyclopentylmethyl ether (CPME), was our first choice in regards its lower miscibility
with water, its stability in presence of strong acids and it is not a peroxide-forming liquid.
One of our recent work reports the successful employ of CPME as the furfural extractive
solvent in a biphasic system for D-xylose and xylan dehydration in presence of FeCl3 and
NaCl under microwave irradiation [9]. To the best of our knowledge, there is no study
about zwitterionic betaine-formic acid combined catalyst employed in a biphasic system

for the conversion of D-xylose into furfural. More recently, microwave-assisted
dehydration gave promising results for increasing yield of dehydration products from
D-xylose [2, 9, 10], xylan and feedstocks. The residence time of the substrate could be
shortened and only a few humin are formed. The main purpose of this work is to
determine the effect of reaction temperature, the reaction time, pH and betaine-formic
acid ratio and the CPME volume percentage on the yield and selectivity of formed
furfural for each sample of interest.
2. Materials and methods
2.1. Materials
Furfural (99% minimum), D(+)-xylose L-arabinose and anhydrous betaine 98% were
purchased from Acros Organic. Cyclopentylmethyl ether was supplied by Sigma Aldrich.
Xylan (from benchwood >90 % (HPLC)) was obtained from Sigma Life Sciences). Rice
husk was furnished from the Fukui city in Japan. The water used in all experiments was a
Millipore Milli-Q grade.
2.2. General procedure for the synthesis of furfural in water-CPME as biphasic media
from D-xylose, xylan and rice husk
In a typical experiment, 0.15 g (1.0 mmol) of D(+)-xylose, 0.07 g (0.59 mmol) of
betaine and 0.25 g of HCOOH were introduced in a 10 mL vial closed with a septum,
followed by the addition of water (1 mL) and CPME (3 mL). Biphasic batch reactions
were carried out by way of microwave heating apparatus (AntonPaar Monowave 300)
and stirred under magnetic stirring (600 rpm) for the desired time. Temperature in the

reaction vessel was measured by means of an IR sensor and the vial was pressurized due
to the normal vapor pressure of the solution at defined reaction temperature. At the end of
the reaction, each sample was diluted with acetonitrile to obtain 100 mL of diluted
solution and filtered prior to analysis through a syringe filter (PTFE, 0.45 m, VWR).
Identical procedure was done starting from 0.15 g of xylan and 0.15 g of rice husk.
2.3. Product analysis
Each sample of the reaction mixture was analyzed separately by means of a
Shimadzu Prominence HPLC. D-Xylose was detected with a low temperature evaporative
light scattering detector (ELSD-LTII) and products were detected with a UV-Vis detector
(SPD-M20A) at a wavelength of 275 nm. The column used was a Grace C18 column
-H2O (90:10) solution flowing at rate
of 0.5 mL.min-1. The column oven was set at 40°C. Regularly, the calibration was
checked out in order to avoid eventual experimental errors associated with all
measurements reported below.
D-Xylose conversion (X), furfural yield (Yi) and furfural selectivity (Si) were
calculated by the following equations:
(
)
(
)
(퐼푛푖푡푖푎푙 푥푦푙표푠푒 푎푚표푢푛푡 푚표푙 − 퐹푖푛푎푙 푥푦푙표푠푒 푎푚표푢푛푡 푚표푙 )
푋 =
푥 100
퐼푛푖푡푖푎푙 푥푦푙표푠푒 푎푚표푢푛푡 (푚표푙)
퐹푖푛푎푙 푓푢푟푓푢푟푎푙 푎푚표푢푛푡 (푚표푙)
푌푖 =
푥 100
퐼푛푖푡푖푎푙 푥푦푙표푠푒 푎푚표푢푛푡 (푚표푙)

퐹푢푟푓푢푟푎푙 푦푖푒푙푑
푆푖 =
푥 100
퐶표푛푣푒푟푠푖표푛 표푓푥푦푙표푠푒
A second and fast dual-wavelength spectroscopic method found in the literature was
employed to confirm the D-xylose concentration in the prehydrozylate from xylan
hydrolysis [19,20]. The procedure involves the preparation of a color agent as follow: 2
grams (15.8 mmol) of phloroglucinol were dissolved in 110 mL of glacial acetic acid,
followed by the addition of 10 mL of anhydrous alcohol and 2 mL of 36 % HCl aqueous
solution. For each sample 1 mL of the aqueous layer containing the sugar were added to
10 mL of the previous color agent and heated up at 100°C for 10 minutes. After returned
to the room temperature, 50 L of each solution was introduced in 3.0 mL of deionized
water before the spectral scanning conducted in a wavelength found between 380 and 700
nm. UV-Vis absorption spectra were recorded using a Jasco V530 UV-vis absorption
spectrophotometer. In acidic media, the furfural formed is allowed to react with
phloroglucinol to generate a derivative showing absorbance at 553 nm.
For rice husk, the furfural yields have been calculated from the measured HPLC
value according to the following equation:
퐹푖푛푎푙 푓푢푟푓푢푟푎푙 푐표푛푐푒푛푡푟푎푡푖표푛 (푚푔/푚퐿)
푌푖 =
푥 100
퐼푛푖푡푖푎푙 푠푢푏푠푡푟푎푡 푐표푛푐푒푛푡푟푎푡푖표푛 (푚푔/푚퐿)
3. Results and discussion
Currently, betaine hydrochloride (BHC) is recognized for its catalytic activity, but as

the best of our knowledge, its natural zwitterionic form was never employed as a catalyst
for dehydrating C5 or C6 sugars. Before starting our experiments, betaine (0.07 g, 0.59 eq)
was introduced in an aqueous solution of D-xylose (0.15 M). The resulted transparent
solution was stirred for 1 hour at 170°C in presence of 3 mL of CPME upon microwave
irradiation. In the case of the previous water-CPME ratio, for 1.0 mmol of furfural the
partition coefficient of furfural between water and CPME was determinated using this
formula Kp = [furfural]water/[furfural]CPME and gave a value of 4. Interestingly, furfural
was produced with a yield of 37 %. However even multiplied by two the amount of
betaine, the yield never exceeded 41 %. In the literature, some ILs holding an acid moiety
are reported to express potential for the dehydration of hexoses or pentoses in water using
conventional heating in batch reactors.[11, 12] Inspired by these results, and also by some
recent studies of formic acid catalytic activity for converting D-xylose into furfural, a
combination of both catalysts appeared for us to be a good opportunity to develop a new
homogeneous catalytic system. Addition of a small volume of formic acid to a betaine
aqueous solution was expected to improve the fundamental activity of the betaine. Herein,
IL [HBet][HCOO] is expected to bind to the released water during the dehydration
mechanism.
3.1. Influence of measured pH on the furfural yield
For dehydrating D-xylose, the best betaine-formic acid ratio was investigated at
170°C using the best ratio of water-CPME (1:3; v/v) (Table 1) [8].
Table 1

Various ratio of betaine-formic acid, their relative pH and furfural yields.a
Entry
Formic acid
(g.mL-1)
0.5
Betaine
(g)
pH
Furfural
(%)
53
1
2
3
4
0
1.72
1.88
2.08
2.30
0.5
0.07
0.07
0.07
75
0.25
81
0.125
80
a Reaction conditions: D-xylose (150 mg), formic acid, betaine, water-CPME, 1:3, v/v,
MW, 170°C, 1 h.
Without addition of the zwitterionic betaine (Table 1, entry 1), the reaction mixture
was too acidic and only 53 % of furfural was produced during the procedure with a
complete conversion of the D-xylose. In our opinion, betaine should be responsible of
intermediate isomerization of xylose into xylulose, a species recognized to be more
sensitive to dehydration in acidic condition, but not detected by HPLC. This result was
compatible with the current conclusions about the relation existing between acid
concentration and furfural selectivity [13, 14]. When added to 1 mL of aqueous formic
acid solution (0.125-0.5M), betaine (0.07 g) is able to reduce the pH of each solution
(Table 1, entries 2-4). For aqueous solutions of betaine (0.125 and 0.25 M), pH values
were 2.30 and 2.08 respectively. These values were compatible with D-xylose
dehydration and the yields of formed furfural reached 80 % in both cases (Table 1, entries
3 and 4). By increasing the concentration of formic acid using the same amount of betaine,
the value of pH fell to 1.88 and only a small drop of yield was recorded. In many cases, in

the literature the dehydration of the pentose is carried out at pH inferior or equal to 1.
Without betaine, using a formic acid diluted aqueous solution of pH = 2.1, 48 % of
furfural yield was obtained in the same above condition. At the lower pH, using mineral
inorganic acids such as HCl, the acidity is a negative factor for getting high yield of
furfural. Due to the formation of by-products generated by the resinification of the
furfural contents, the dehydration shows low selectivity of furfural especially at the
higher temperatures and promoted by the strong acidity of the catalyst.
3.2. Influence of the temperature on the furfural yield
Next, keeping the best condition of pH (Table 1, entry 3), influence of the
temperature was studied (Fig 1). Maximum of furfural yield was obtained at 170°C with
no trace of D-xylose contents remaining in the hydrophilic layer. Other higher yields were
provided at 160 and 180°C with values of 72 and 76 % respectively. Unfortunately, the
amount of produced humine was most significant at 180°C, one more time it became
evidence of possible furfural degradation even in presence of a potent extraction solvent.
However at 150°C, the selectivity of furfural showed a lowest value of 57 %. This result
points out the thermal control of the pentose dehydration also confirmed by the mild
furfural yield of 60 % obtained from L-arabinose when heated at190°C in the same
condition [15].

100
90
80
70
60
50
40
30
20
10
0
130
140
150
160
170
180
190
200
Temperature (°C)
Fig. 1. Reaction profile for the D-xylose dehydration in presence of betaine-HCOOH.
Xylose contents (blue dots) and furfural yield (red dots). Reaction conditions: D-xylose
(150 mg), betaine (0.07 g), HCOOH (0.25 g), water-CPME, 1:3, v/v, MW, 1 h.
3.3. Influence of the xylose loadings on the furfural yield
In the third part, a screening of different xylose loadings was also operated (Table 2).
Except the variation of the C5 sugar concentration, all parameters remained unchanged.
Table 2. Various loadings of xylose versus furfural yields.a
Entry
D-Xylose (mg.mL-1)
Furfural (%)
1
2
3
4
5
100
150
200
250
300
81
80
76
73
72

6
400
600
71
65
7
a
Reaction conditions: D-xylose, betaine (0.07 g), HCOOH (0.25 g), water-CPME, 1:3,
v/v, MW, 170°C, 1 h.
Usually, by increasing the contents of the pentose, yield of furfural became inversely
proportional to the pentose loading. Once the D-xylose concentration reached a value of
600 g.L-1, yield began to decline because of the enhanced viscosity of the aqueous layer
(Table 2, entry 7). Furthermore, the viscous state prevent of the efficient extraction of
furfural by CPME, the product remaining entrapped in the sugar excess. More exactly,
higher viscosity is reported to be the main cause of high rate of cross-polymerization and
by the way the formation of large amount of black humine, mostly when D-xylose
conversion is not quantitative. Despite our efforts, in each case, it was difficult to isolate
the sugar from the humine even tritured and no trace of D-xylose was detected from the
aqueous layer by the spectroscopic method
3.4. Influence of the residence time on the furfural yield
For this following purpose the study was realized at 3 defined temperatures (160, 170
and 180°C) (Fig 2). Each sample was heated up and the reaction was stopped at a defined
period of time before determine their furfural concentrations by HPLC.

Fig. 2. Reaction profile for dehydration of D-xylose in presence of betaine-HCOOH at

varying temperatures a) xylose contents; b) furfural yields. Reaction conditions: D-xylose
(150 mg), betaine (0.07 g), HCOOH (0.25 g), water-CPME, 1:3, v/v, MW.
At 160°C, the yield increase appeared to be quite linear and both yield and selectivity
of furfural increased progressively until they finally reached a maximum value of 66 %
after 80 minutes of stirring. Even the conversion of D-xylose was complete after 60
minutes, the reaction afforded a few humin in the vessel when heated for longer periods.
If a similar sample is heated at 170°C, the maximum of yield of 80% was observed just
after 60 minutes and remained constant beyond this time. On the above graphic, the
conversion of xylose at 170°C followed a typical pattern, complete from 60 minutes, then
the value of selectivity was definitively limited to 80 % for this condition. At this
temperature, with no surprise, CPME confirmed its good ability to extract furfural
product and this hydrophobic furfural content was not subjected to undesired second-type
of side-reaction usually observed between furfural itself. In our opinion, at 170°C the
cross-polymerization could be the main cause of the humin formation. This theory was
reinforced by means of a simple test carried out in a biphasic system consisting of 1 mL
of water containing 0.07 g of betaine and 0.25 g of formic acid brought in contact with 32
g.L-1 furfural solution in CPME. Herein, warmed up at 170°C in microwave oven, even
for period longer than 80 minutes, the concentration of furfural remained almost
unchanged. By opposite, for temperature superior to 170°C, furfural is currently degraded
via second type of side-reaction catalyzed by acid despite the moderate pH of 2.3. Thus in
the normal way, at 180°C, after 20 minutes of stirring 63 % of yield was higher than the
value obtained at 170°C for the same period. Unfortunately, the furfural selectivity was

clearly inferior with a value of 87 %. After 20 minutes of stirring, the selectively was
closed to 92 % at 170°C. Furthermore, if the reaction is carried out at 180°C after 20
minutes, both yields and selectively decreased jointly and the final value of selectivity
was 69 % From 180°C, in a pressurized vessel, the dielectric constant of water decreases
in parallel to its polarity, the solvent becoming more compatible with the less polar
furfural. This phenomenon could also explain the limit of selectivity expressed at this
temperature.
3.5. Influence of CPME volume percentage on the furfural yield
Starting from the initial CPME-water ratio, different CPME volume percentages
were tested for dehydrating 150 mg of D-xylose (Fig 3). Each sample was heated up at
170°C for one hour. Except with 4 mL of CPME, the conversions of D-xylose were
quantitative and almost near to 100 % also confirmed by the spectroscopic method.
Fig. 3. Effect of the water to CPME phase ratio (v/v) on the furfural production. Reaction

conditions: D-xylose (150 mg), betaine (0.07 g), HCOOH (0.25 g), water-CPME (4 mL),
MW, 170°C, 1 h.
As expected, the yield of furfural increased with the volume of CPME. On the other
hand, when reaction was performed only in water a large amount of humin was formed.
According to these results, maximum of yield was certainly achieved with the
CPME-H2O (3/1) ratio. In a biphasic system, a minimum volume of polar solvent such as
pure water or aqueous DMSO [16] is required to dehydrate effectively pentose
derivatives. Curiously, the present betaine-HCOOH catalytic system is able to convert
D-xylose into its furfural counterparts without any trace of water. Despite the lack of
D-xylose solubility in the active extraction solvent, 30 % of furfural was produced in the
sole organic layer.
3.6. Transformation of xylan and risk husk using optimized conditions
Direct transformation of amorphous xylan was operated using various ratio of
betaine-HCOOH, the depolymerization of 1->4--xylopyranosyl linkages to monomers
units is reported to be more active for pH near to 1. Furfural formation from xylan
involves a two-steps mechanism. Basically, the polysaccharide backbone is
depolymerized into single pentose monomer units before their own dehydration. However
the furfural yield is known to be dependent of the real concentration of D-xylose released
during the process. According the type of method employed and the pH of the mixture,
xylo-oligomers of various length and numbers found in the hydrozylate liquor could
influence the final yield of furfural. Using the optimized method described for the

D-xylose dehydration, 150 mg of the polymer introduced in 1 mL of water solution
loaded in presence of betaine and formic acid furnished the target furfural in 58% yield.
Later, with 0.5 g of HCOOH, at a lower pH of 1.9, the yield reached a value of 76 %. In
the other hand, if the aqueous phase is loaded with 300 mg of xylan, at 170°C the
resulting viscous hydrozylate afforded only 48 % of the desired furfural after one hour of
microwave heating. In order to determine the best condition of temperature in parallel to
the residence time study, all related experiments were successively realized at pH 1.9 (Fig
4).

Fig. 4. a) Temperature reaction vs furfural yield and percentage of released D-xylose; b)
Time residence vs furfural yield measured at 160 and 170°C. Reaction conditions: xylan
(150 mg), betaine (0.07 g), HCOOH (0.25 g), water-CPME, 1:3, v/v, MW.
According the graphic 4a, for temperatures found between 150 and 170°C, the
furfural yield increased linearly, until it began to fall from 180°C, characteristic feature of
furfural degradation encountered at the higher temperatures. In parallel to the
phenomenon, for temperatures comprised between 140 and 160°C, the released D-xylose
was not completely converted into furfural. In the Figure 4b, the influence of the reaction
time was carefully studied at 160°C and 170°C. At 160°C, the furfural yield enhancement
is more progressive and linear compared to the results recorded at 170°C. After 60
minutes, in both cases the degradation of furfural seems to be slow, but curiously at

170°C under microwave irradiation it became out of control. In our opinion, at the lower
pH of 1.9, the concentration of the released xylose could be accelerated promoting the
non desired cross-polymerization between xylose and furfural. Besides this result is also
in agreement with the yield of 53 % obtained from the dehydration of the same mass of
xylose in the same pH condition but this time without betaine.
The final purpose of our work involve the transfer of the optimized procedure to rice
husk, a renewable, cheap and largely available feedstock rich in cellulosic and
hemicellulosic biomass rich in complex polysaccharide backbone [17]. For this reason,
rice husk could be employed directly as a furfural source, but the process to access to the
top-value product requires various pretreatments such as sonication or boiling. At 190°C,
compared to the two-stage process developed by Suxia et al. [18] leading to furfural
yields of 9 %, our optimized procedure using dried rice husk (200 mg) in presence of
betaine (0.08 g) and 1 mL of aqueous formic acid solution (50 g.mL-1) in a typical
biphasic microwave-assisted reaction yielded the target furfural in 6%.
3.7. Resusability of the catalytic system
Several consecutive reactions were performed with the optimized procedure. After
one cycle the aqueous layer was recovered from the current reaction and loaded with
D-xylose (150 mg), fresh CPME (3 mL) were also added each time as the extraction
solvent. Finally, each cycle was carried out for 60 minutes at 170°C and starting from the
fifth cycle the yield of formed furfural began to decrease and gave only a value of 75 %.
The phenomenon could be explained by the pH measurement realized after each cycle.
Indeed, the pH value moved from 2.08 to 2.75 but remained enough acidic to produce

furfural. Interestingly, the system still remained active after ten cycle, however it
appeared difficult to control the yields found between 69 and 75 %
Fig. 5. Reusability of the aqueous phase containing betaine and HCOOH for the D-xylose
dehydration to furfural. Reaction conditions: D-xylose (150 mg), betaine (0.07 g),
HCOOH (0.25 g), water-CPME, 1:3, v/v, MW, 170°C, 1 h.
4. Conclusion
In summary, we have demonstrated the potential of betaine coupled with an excess of
formic acid for the furfural production in a water-CPME biphasic system heated in a
range of temperature between 160 and 180°C. The best yields were obtained at 170°C for
1 hour of stirring. This methodology was also suitable for producing furfural from xylan
and rice husk. In the next stage of our study, this procedure should be transposed to
hexoses and cellulose in order to produce 5-hydroxymethyl furfural (HMF) as the main
target compound.

Acknowledgement
YW would like to thank the China Scholarship Council (CSC) for the financial support.
References
[1] G. Marcotullio, W. De Jong, Green Chem. 12 (2010) 1739-1746.
[2] O. Yemis, G. Mazza, Bioresour. Technol. 102 (2011) 7371-7378.
[3] C. Rong, X. Ding, Y. Zhu, Y. Li, L. Wang, Y. Qu, X. Ma, Z. Wang, Carbohydr. Res.
350 (2012) 350, 77-80.
[4] K. De Oliveira Vigier, A. Benguerba, J. Barrault, F. Jerome, Green Chem. 4 (2012)
285-289.
[5] C. Chiappe, S. Rajamani, F. D’Andrea, Green Chem. 15 (2013) 137-143.
[6] K. Lamminpaa, J. Ahola, J. Tanskanen, Ind. Ing. Chem. Res. 51 (2012) 6297-6303.
[7] W. Yang, P. Li, D. Bo, H. Chang, X. Wang, T. Zhu, Bioresour. Technol. 133 (2013)
361-369.
[8] W. Yang, P. Li, D. Bo, H. Chang, Carbohydr. Res. 357 (2012) 53-61.
[9] S. Le Guenic, F. Delbecq, C. Ceballos, C. Len, J. Mol. Catal. A: Chem. 410 (2015)
1-7.
[10] R. Weingarten, J. Cho, W. C. Conner Jr, G. W. Huber, Green Chem. 12 (2010)
1423-1429.
[11] S. E. Sim, S. Kwon, S. Koo, Molecules 17 (2012) 12804-12811.
[12] J. C. Serrano-Ruiz, J. H. Campelo, M. Francavilla, A. A. Romero, R. Luque, C.
Menendez-Vazquez, A. B. Garcia, E. J. Garcia-Suarez, Catal. Sci. Technol. 2 (2012)
1828-1832.

[13] K. Dassan, B. Girissuta, M. Lopez, J. J. Leahy, H. M. B. Hayes, ChemSusChem 8
(2015) 1411-1428.
[14] K. Lamminpaa, J. Ahola, J. Tanskanen, RSC Adv. 4 (2014) 60243-60248.
[15] E. R. Garrett, B. H. Dvorchik, J. Pharm Sci. 58 (1969) 813-820.
[16] W. Wang, H. Li, J. Ren, R. Sun, J. Zheng, G. Sun, S. Liu, Chin. J. Catal. 35 (2014)
741-747.
[17] V. Da Silva Lacerda, J. B. Lopez-Sotelo, A. Correra-Guimanaes, S.
Hernandez-Naaro, H. Sanchez-Bascones, L. M. Navas-Gracia, P. Martin-Ramos, E.
Perez-Lebena, J. Martin-Gil, Bioresour. Technol. 180 (2015) 88-96.
[18] R. Suxia, X. Haiyan, Z. JinLing, L. Shunqing, H. Xiaofeng, L. Tinghou, Carbohydr.
Res. 359 (2012) 2012, 1-6.
[19] T. J. Eberts, R. H. B. Sample, H. R. Glick, G. H. Ellis, Clin. Chem. 25 (1979)
1440-1443.
[20] C. Chi, H. M. Chang, Z. Li, H. Jameel, Z. Zhang, BioRes. 8 (2013) 172-181.