Synergy effect between solid acid catalysts and concentrated carboxylic acids solutions for efficient furfural production from xylose

Article abstract of DOI:10.1016/j.cattod.2013.10.034

An efficient furfural formation from xylose was demonstrated combining a concentrated aqueous solution of acetic acid and solid acid catalysts. Higher furfural yields and selectivities were obtained by comparison to the catalytic performances obtained in pure water. The evident synergy effect observed at 150 °C between the aqueous carboxylic acid solution and the solid acid catalysts is tentatively explained by the occurrence of two phenomena: 1) the contribution of Lewis acid sites which would operate in cooperation with the homogeneous weak Br?nsted acidity brought by the aqueous acetic acid solution. According to the literature, the two steps mechanism involving the xylose-xylulose isomerization over Lewis acid sites and the successive Br?nsted acid catalyzed cyclodehydration to furfural would be the prevailing reaction pathway in the heterogeneous-homogenous catalytic system at 150 °C. 2) an enhancement of the surface solid acid coverage by the carbohydrate and furfural owing to the presence of carboxylic acid in the aqueous solution as shown by comparative liquid phase adsorption experiments done in pure water and in aqueous acetic acid solutions. Among a series of solid acid catalysts, ZrW, Cs₂HPW₁₂O₄₀, HY (Si/Al = 15), K10 and NbOH, the latter one, NbOH used non-calcinated was shown to be active, selective and stable in the aqueous acetic acid media. HY and K10 are as active and selective for furfural formation but suffer for a strong Al leaching which precludes their utilization as true solid acid catalyst in acetic acid media.

Full text of DOI:10.1016/j.cattod.2013.10.034

GModel
CATTOD-8660; No. of Pages9
ARTICLE IN PRESS
Catalysis Today xxx (2013) xxx–xxx
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Synergy effect between solid acid catalysts and concentrated
carboxylic acids solutions for efficient furfural production from xylose
∗
Aude-Claire Doiseau, Franck Rataboul, Laurence Burel, Nadine Essayem
Université de Lyon 1, IRCELYON, Institut de Recherches sur la Catalyse et l’Environnement de Lyon, CNRS, UMR 5256, 2 avenue Albert Einstein, 69626
Villeurbanne, France
a r t i c l e i n f o
a b s t r a c t
Article history:
An efficient furfural formation from xylose was demonstrated combining a concentrated aqueous solution
of acetic acid and solid acid catalysts. Higher furfural yields and selectivities were obtained by comparison
to the catalytic performances obtained in pure water. The evident synergy effect observed at 150 C
Received 29 June 2013
Received in revised form 7 October 2013
Accepted 8 October 2013
Available online xxx
◦
between the aqueous carboxylic acid solution and the solid acid catalysts is tentatively explained by the
occurrence of two phenomena: 1) the contribution of Lewis acid sites which would operate in cooperation
with the homogeneous weak Brønsted acidity brought by the aqueous acetic acid solution. According to
the literature, the two steps mechanism involving the xylose-xylulose isomerization over Lewis acid
sites and the successive Brønsted acid catalyzed cyclodehydration to furfural would be the prevailing
Keywords:
Furfural
Xylose cyclodehydration
Nobium hydroxide
Catalytic dehydration in water
◦
reaction pathway in the heterogeneous–homogenous catalytic system at 150 C. 2) an enhancement of
the surface solid acid coverage by the carbohydrate and furfural owing to the presence of carboxylic
acid in the aqueous solution as shown by comparative liquid phase adsorption experiments done in pure
water and in aqueous acetic acid solutions. Among a series of solid acid catalysts, ZrW, Cs2HPW12O40, HY
(Si/Al = 15), K10 and NbOH, the latter one, NbOH used non-calcinated was shown to be active, selective
and stable in the aqueous acetic acid media. HY and K10 are as active and selective for furfural formation
but suffer for a strong Al leaching which precludes their utilization as true solid acid catalyst in acetic
acid media.
©
2013 Elsevier B.V. All rights reserved.
1
. Introduction
furfural steam distillation. This process is characterized by a limited
yield in furfural, 35% due to side reactions and by an excessive water
vapour consumption [3]. Other new processes have been devel-
oped but are not yet commercialized such as the SupraYield process
which claims a furfural production (up to 50%) thanks to a discon-
Upgrading the carbohydrates biomass into high value-added
chemicals is a current research topic of high interest [1]. Among
the principal sources of carbohydrates, cellulose, a hexose biopoly-
mer, has already well established applications, mainly in the paper
industry. By contrast, hemicelluloses have limited identified appli-
cations and are still considered as an agriculture waste for example
in bagasse sugarcane. In this xylan rich biomass, hemicelluloses
are composed mainly of pentoses and the principal commercial
product derived from them is furfural, obtained from the dehydra-
tion of the C5 units forming the polysaccharide (Scheme 1). This is
an industrial process known since the earlier 1900, and the world
production is about 250,000 tons annually. The potential of the fur-
fural yield depend on the composition of the biomass and may
ranges from 150 to 220 kg of furfural produced per metric ton of
dry biomass [2].
◦
tinuous reactor working at high temperatures (200–225 C) and
high pressures (15–30 bars) but at a low contact time [4]. Nowa-
days, there is a need to develop a sustainable process which might
avoid the use of strong mineral acids, such as sulphuric acid, and
their drawbacks: low selectivity, equipment corrosion, health and
environmental issues, impossibility of catalyst recycling. This jus-
tifies the growing interest in the search for heterogeneous acid
catalysts for conducting in an efficient and sustainable way the
furfural production from xylose dehydration.
One of the first reports on the use of a heterogeneous catalyst
for the transformation of xylose in furfural is from Moreau et al.
in 1998 [5]. They reported the use of H-mordenite and H-faujasite
zeolites in a biphasic medium composed of water and toluene or
MIBK (methyl isobutyl ketone) as furfural extractive solvent. This
methodology allows the furfural extraction from the water phase,
in order to avoid its degradation. The furfural selectivity achieved
Today, the most widely used process in the industry is the
Quaker Oats process, which uses sulphuric acid as catalyst and a
8
2% for a xylose conversion of 51%. Since this pioneering work, oth-
∗ Corresponding author. Tel.: +33 4 72445315.
E-mail address: nadine.essayem@ircelyon.univ-lyon1.fr (N. Essayem).
ers zeolites such as ZSM-5 [6], HUSY, H-Beta [7–9], H-mordenite
0
920-5861/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.cattod.2013.10.034
Please cite this article in press as: A.-C. Doiseau, et al., Synergy effect between solid acid catalysts and concentrated carboxylic acids solutions
for efficient furfural production from xylose, Catal. Today (2013), http://dx.doi.org/10.1016/j.cattod.2013.10.034

GModel
CATTOD-8660; No. of Pages9
ARTICLE IN PRESS
2
A.-C. Doiseau et al. / Catalysis Today xxx (2013) xxx–xxx
OH
O
OAc
O
O
O
O
HO
7
O
HO
HO
O
O
HO
O
OH
n
O
xylan hemicellulose
OH
HOOC
OH
acid
OMe
O
HO
HO
acid
OH
O
OH
O
xylose
furfural
Scheme 1. Furfural formation from xylan hemicelluloses.
[
7,10], SAPO-11 [8,9] and solid acids like niobic acid [11] or even
deactivation via cationic exchange as evocated recently by Dumesic
et al. [23].
sulphonated graphenes [12] have been introduced for this reac-
tion. It can be mainly concluded from all these studies that zeolites
with small pore size and a low mesoporous volume, are more effec-
tive for the dehydration of xylose and more selective when used
in the presence of an extractive solvent, leading to high furfural
yields.
Indeed, as evocated above, in the more general field of catalytic
sugars cyclodehydration, most of the studies have been carried
out in the presence of an extractive solvent to increase the furanic
derivatives yields. Cabezas et al. and Corker et al. studied the influ-
ence of the solvent employed during the extractive reaction. They
reported that furfural is easily degradable due to side reactions
forming humins when the reaction occurs in water, at high tem-
perature [13,14]. Thus, various systems have been evaluated in
order to determinate the benefit of using either MIBK, toluene,
isobutyl acetate or alcohols such as methyl-2-butanol and 2-ethyl-
In order to go further in the development of an environment-
friendly process, we decided to study sugars cyclodehydration in
aqueous media without resort to an organic extractive solvent with
potential toxicity. Previous works from our group have shown the
benefit to perform hexoses dehydration in concentrated aqueous
solutions of carboxylic acids. From fructose, 5-HMF yields of 60%
were achieved in concentrated aqueous solutions of lactic acid. In
pure water, from the screening of a large panel of solid acid catalyst,
we have shown that the most efficient one was niobium hydroxide
acid (NbOH) which led to a maximum 5-HMF yield of 30% [24].
Notably, this 5-HMF yield was obtained only when a high catalyst to
sugar ratio was used. When NbOH was used in catalytic proportion,
5 wt%, the 5-HMF yield was limited to 10%.
To go further in the development of an efficient and green way
to dehydrate monosaccharides, the present work aims to evaluate a
large panel of different solid acid catalysts for xylose cyclodehydra-
tion in pure water and to perform an comparative evaluation of a
concentrated aqueous solution of acetic acid as reactive medium, in
the absence or in the presence of added solid acids. It was expected
from this methodology to evidence or not, the occurrence of a syn-
ergy effect between the homogeneous weak Brønsted acid, acetic
acid and the solid acid catalysts.
1
-hexanol. Alkyl-phenols are also applied with success to get high
furfural yields [15]. Toluene appeared to be the most efficient sol-
vent, leading to the highest furfural yields up to 80%. Despite its
toxicity and unlikely industrial large scale application, this solvent
has been further studied with different catalytic systems. Dias et al.
used this solvent, in proportion of 70 wt%/water to explore the
activity of several solid catalysts. Whatever their catalytic activ-
ity and selectivity in furfural in pure water, heteropolyacids (PW,
SiW and PMo) [16,17], cesium-based heteropolyacids [18], tita-
nium and niobium nanosheets [19], niobium-based silicates [20]
and silica supported sulphated zirconia [21] have greater catalytic
performances when used with toluene as extractive solvent. Note
that DMSO, when used as co-solvent added to the aqueous phase,
has also a positive influence to the furanic derivatives yields. Its role
is less documented, but it was proposed in the similar reaction of
glucose dehydration into 5-hydroxymethyl furfural (5-HMF), that
adding DMSO to water might favor the furanose form of hexoses
which exists in negligible proportion in pure water. It was also pro-
posed that DMSO addition to water could reduce the condensation
reactions and the successive rehydration of 5-HMF associated by
the reduced proportion of water [22]. However, beside the effi-
ciency of the extractive reaction methodology applied to sugar
dehydration, it is scarcely underlined that due to the low partition
coefficient of furfural to the organic solvent, its efficient extrac-
tion depends on salts addition in high amounts leading to complex
reaction media [15]. If this methodology can suit well to the use of
homogeneous mineral acids, it will not be compatible with the use
of a heterogeneous solid acid, due to the possible solid acid catalyst
2. Experimental
2.1. Catalysts
Tungstated zirconia (ZrW) was prepared by anionic exchange
according to an optimized method [25]. The acidic cesium salt
of phosphotungstic acid Cs HPW12O40 (Cs HP) was prepared by
2
2
precipitation through a previously reported procedure [26]. Nio-
bium hydroxide (NbOH) was obtained from CBMM (Brazil) and
used as received without calcination. Zeolites HY (Si/Al = 15), ZSM-
5 (Si/Al = 40) were provided by IFPEN. K10 was purchased from
Aldrich.
2.2. Catalysts characterization
The specific surface areas were determined by nitrogen adsorp-
◦
tion at −196 C using a Micrometrics ASAP 2020 equipment.
◦
Samples were previously degassed at 150 C under vacuum
(10 Pa) for 5 h. Surface areas were calculated using the
Brunauer–Emmet–Teller (BET) methods.
−
3
Please cite this article in press as: A.-C. Doiseau, et al., Synergy effect between solid acid catalysts and concentrated carboxylic acids solutions
for efficient furfural production from xylose, Catal. Today (2013), http://dx.doi.org/10.1016/j.cattod.2013.10.034

GModel
CATTOD-8660; No. of Pages9
ARTICLE IN PRESS
A.-C. Doiseau et al. / Catalysis Today xxx (2013) xxx–xxx
3
Ammonia adsorption was measured by gravimetry using a
Setaram DTA-DTG 92-12 apparatus. 25 mg of the sample was
weighed in a platinum holder and in situ pre-treated under helium
100
conversion
yield
8
6
4
0
0
0
3
−1
◦
flow (40 cm min ) at 150 C for 3 h. After cooling the solid down
◦
to 80 C under helium flow, the solid was saturated with ammo-
nia using an ammonia stream diluted in helium (1%), and then a
◦
pure helium flow was introduced at 80 C to remove the reversible
(%)
ammonia uptake. The total amount of acid sites was deduced from
◦
the irreversible ammonia adsorption at 80 C.
FTIR spectra of self-supported pellets (ZrW, NbOH, K10, HY) or
20
0
of Cs HPW12O40 spread on a silicon plate were recorded with a
2
Brucker Vector 22 spectrometer in the absorption mode with a res-
−1
olution of 2 cm . The samples were placed in an IR cell equipped
with CaF2 windows and treated in situ. The wafers of ZrW, K10, HY
and NbOH samples were pretreated under vacuum at respectively
0
2
4
6
8
10
12
14
16
Time (h)
◦
◦
at 400 C or 150 C for the later one. The heteropolyacid dispersed
Fig. 1. Kinetic of xylose dehydration into furfural performed in 20 wt% acetic acid
solution in the presence of NbOH. Conditions: Xylose 1%wt, solvent 60 mL (20 wt%
of acetic acid in water), 150 C, NbOH: 5 wt%/xylose.
◦
on the silicon plate was vacuum treated at 200 C. Pyridine was
◦
adsorbed under saturation vapor pressure at ambient temperature
◦
then desorbed at 150 C for 1 h in order to remove the physisorbed
species.
Calorimetry of NH adsorption: The acid properties were mea-
sured by NH3 adsorption at 80 C, using a TianCalvet calorimeter
coupled with a volumetric equipment. The catalysts samples of
3
Yield (%)
◦
Selectivity (%) =
× 100
Conversion (%)
◦
NbOH and Cs HPW12O
(0.1 g) were evacuated at 150 C and
2
40
Few preliminary experiments were performed in order to set
some conditions, and especially the reaction time. For that we stud-
ied the influence of the presence of one of the reported most active
catalyst, the Nb based catalyst, NbOH, in 20 wt% aqueous acetic acid
◦
2
00 C respectively for 1 h under secondary vacuum. The oth-
◦
ers samples, ZrW and HY were pretreated at 400 C in the same
conditions. Then, the pre-treated samples were placed into the
calorimeter up to the stabilization of the temperature (one night),
then contacted with small doses of gas up to equilibrium and the
differential enthalpy of adsorption was recorded together with the
◦
solution. The temperature was first set up at 150 C as reference to
literature data. Due to the volatility of furfural/water azeotrop as
regard xylose, the progress of the reaction was not followed by
taking liquid samples at regular times, but we performed distinct
reactions at various reaction times for kinetic studies. The xylose
conversion and the furfural yield are shown in Fig. 1.
amount of adsorbed NH3.
2
.3. Xylose dehydration reactions
We note that, while the xylose conversion increased regularly
with time, the furfural yield started to decrease in the second part
of the reaction. As already evocated above, this is most likely due to
furfural degradation after prolonged reaction times in the absence
of extraction solvents. From these preliminary experiments done
on one of the most active solid catalyst, we decided to fix the reac-
tion times at 6 h or 15 h for the following studies. The data obtained
at 6 h will be used to calculate the rate of xylose conversion and the
furfural production although the slight decrease of the reaction rate
after 6 h of the reaction observed for one of the most active cata-
lyst (Fig. 1). Reaction performed for 15 h will be especially useful to
compare the less active catalysts.
The catalytic conversion of xylose (Sigma–Aldrich, >99%) into
furfural was studied in a magnetically stirred 100 mL autoclave,
heated with a controlled heating envelop. In a typical procedure,
xylose (0.6 g), powdered catalyst (30 mg), water (60 mL), or in the
case of 20 wt% of acetic acid solution, water (48 mL) and acetic
acid (12 mL), were introduced into the autoclave. The reactor
was flushed and pressurized with argon, and then heated to the
desired temperature. Zero time was the instant where the internal
temperature of the vessel reached the target temperature. At the
end of a run the autoclave was cooled down to room temperature
using ice bath, the solid catalyst and possibly solid humins were
filtered off (0.45 ␮m Teflon filter) and the liquid phase were
recovered for analysis.
3. Results and discussion
The identification and quantification of the reaction products in
the aqueous phase were conducted by Shimadzu LC-20AD HPLC
3
.1. Catalysts characterization
◦
(
COREGEL 87C column, 80 C) equipped with a refractive index
−1
detector using water as eluent (0.6 mL min ). Authentic samples
of xylose, furfural and levulinic acid (Sigma–Aldrich, 99%) were
used for calibration. Levulinic acid is the main by-product identi-
fied when the reaction is performed in water; the presence of acetic
acid prevents its identification by HPLC due to overlapping peaks.
Xylose conversion, furfural yield and selectivity were calculated
as follows:
The main acidic features of these solid acids, such as the nature
of acidity (Lewis/Brønsted) determined by IR of Py adsorption and
their strength, measured by calorimetry of ammonia adsorption
(
[
Qdiff NH ) were already reported in our previous publications
24,27–29]. These data are summarized in Table 1, together with the
3
total number of acid sites determined by irreversible NH3 adsorp-
tion monitored by gravimetry and the BET surface areas.
The catalyst selection contains two strong solid acids: a
tungstated zirconia, dominated by the Lewis acidity and the acidic
cesium salt of 12-tungstophosphoric acid which contains exclu-
sively Brønsted acid sites. The three others solids, K10, HY and
NbOH are weaker acids. While K10 and HY are dominated by
the Brønsted acidity, NbOH presents more Lewis acid sites than
Brønsted ones (Fig. 2). Notably, NbOH presents the highest acid
sites density.
Conversion (%)
mol of initial xylose − mol of unreacted xylose
=
× 100
mol of initial xylose
mol of furfural
mol of initial xylose
Yield (%) =
× 100
Please cite this article in press as: A.-C. Doiseau, et al., Synergy effect between solid acid catalysts and concentrated carboxylic acids solutions
for efficient furfural production from xylose, Catal. Today (2013), http://dx.doi.org/10.1016/j.cattod.2013.10.034

GModel
CATTOD-8660; No. of Pages9
ARTICLE IN PRESS
4
A.-C. Doiseau et al. / Catalysis Today xxx (2013) xxx–xxx
Table 1
BET surface area and acidic features of the solid catalysts used in this study.
Aciditya
BET surface area (m2
g
−1
)
Total acid sitesb (␮mol g−1
)
Acid sites density (␮mol m
−2
)
Qdiffc NH3 adsorption (kJ mol−1
)
Catalyst
ZrW
Cs2HP
K10
HY(Si/Al = 14)
NbOH
Lewis (Brønsted)
Brønsted
Brønsted (Lewis)
Brønsted (Lewis)
Lewis (Brønsted)
118
105
233
718
144
189
189
129
187
469
1.6
1.8
0.6
0.3
3.3
160
200
–
140
150
a
Determined by in situ IR spectroscopy of pyridine adsorption. The dominant nature of acidity is written in bold letters.
Measured by NH3 adsorption monitored by gravimetry as described in the experimental part.
Measured by calorimetry of NH3 adsorption
b
c
1
00
conversion
yield
8
0
0
0
0
6
4
2
(%)
0
none
ZrW
K10
HY
NbOH
Cs HP
2
Si/Al = 15
Fig. 4. Xylose dehydration into furfural performed in 20 wt% acetic acid. Influence
◦
of solid acid catalysts. Xylose 0.6 g, catalyst 30 mg, solvent 60 mL, 150 C, 6 h.
Fig. 2. Infrared spectrum of pyridine adsorption on NbOH then vacuum treated at
1
◦
50 C.
In pure water, we can observe from Fig. 3 that, ZrW,
Cs HPW12O40 and the niobium hydroxide are active catalysts lead-
2
ing respectively to xylose conversions of 30, 33 and 53%. The
niobium hydroxide, which was used without preliminary calci-
nation step, is by far the catalyst leading to the highest furfural
production in pure water, reaching up to 22% yield in furfural. On
3
.2. Catalytic studies
3.2.1. Comparison of solid acid catalysts for xylose
cyclodehydration in furfural in pure water or in concentrated
acetic acid solution (20 wt%)
The catalytic performances of the selected solid acid catalysts
were evaluated in pure water first (Fig. 3) and then in a concentrated
aqueous solution of acetic acid (Fig. 4).
the contrary, ZrW and Cs HPW12O40 have a very low selectivity
2
in furfural, leading mostly to the formation of levulinic acid and
humins. The HY zeolite and the K10 clay exhibit a reduced activity
equivalent to that measured in the blank experiment. To go fur-
ther in the understanding of the catalysts reactivity, let us consider
the catalysts acidic features presented in Table 1. It seems that the
acid sites density might be correlated with the catalysts activity for
xylose conversion in pure water. The most active catalysts, ZrW,
In the absence of solid catalyst, the xylose conversion is limited
◦
to 10% after 6 h of reaction in pure water at 150 C with a furfural
yield of 6% (Fig. 3).
Higher xylose conversion and furfural yield are observed in
Cs HP, NbOH are all of them characterized by a significant highest
2
acetic acid solution, respectively 17% and 11% (Fig. 4). Thus, at
acid sites density compared to K10 and HY which are, compar-
atively, not efficient catalysts in these conditions. The nature of
the acidity Lewis/Brønsted and the acid strength do not find clear
correlations with the xylose conversion nor the furfural selectivity
in pure water. We can just underline that the most efficient solid
catalyst to produce furfural in pure water is, within our catalysts
selection, NbOH and this catalysts is characterized by the highest
density of acid sites of medium strength with a low Brønsted/Lewis
acid sites ratio. The present study does not confirm a previous one
which reported the required high Brønsted/Lewis acid sites ratio to
get selectively furfural from xylose in water [30].
◦
1
50 C, a 20 wt% aqueous solution of acetic acid is only slightly a
more reactive media than pure water as regards to xylose dehy-
dration. One can consider that these conditions are convenient to
evaluate the promotion effects of solid acids addition since the
xylose conversion is limited in these both homogeneous liquid
media under these conditions of reaction.
100
conversion
yield
8
0
0
0
0
The solid acids series has been evaluated in concentrated aque-
◦
ous acetic acid media, in the same conditions, for 6 h at 150 C
6
4
2
(Fig. 4). In the absence of solid acids, by changing the reaction
(
%)
medium from water to the aqueous solution of acetic acid (20 wt%),
the xylose conversion was raised from 10 to 17% and furfural yield
◦
from 6 to 11% only. Thus, these conditions, 6 h of reaction at 150 C,
are convenient to evidence the occurrence of a beneficial effect of
the addition of a heterogeneous acid function to the homogeneous
weak Brønsted acidity brought by the concentrated aqueous acetic
acid medium. Starting from glucose, we reported recently that glu-
cose is not dehydrated in 20 wt% acetic acid solution after 2 h of
0
none
ZrW
Cs2 HP
K10
HY
NbOH
Si/Al = 15
Fig. 3. Xylose dehydration into furfural performed in pure water. Influence of solid
acid catalysts. Xylose 0.6 g, catalyst 30 mg, water 60 mL, 150 C, 6 h.
◦
◦
reaction at 150 C, but upon addition of a strong solid Lewis acids
Please cite this article in press as: A.-C. Doiseau, et al., Synergy effect between solid acid catalysts and concentrated carboxylic acids solutions
for efficient furfural production from xylose, Catal. Today (2013), http://dx.doi.org/10.1016/j.cattod.2013.10.034

GModel
CATTOD-8660; No. of Pages9
ARTICLE IN PRESS
A.-C. Doiseau et al. / Catalysis Today xxx (2013) xxx–xxx
5
1
00
100
a)
b)
Xylose conversion (%)
Furfural yield (%)
Xylose conversion (%)
Furfural yield (%)
8
6
4
2
0
0
0
0
0
80
60
4
0
0
0
2
0
1
20
100
0
1
20
100
acetic acid %
acetic acid %
Fig. 5. Xylose cyclodehydration into furfural- Influence of acetic acid concentration in the absence (a) or presence of solid acid catalyst (b). Conditions: Xylose 0.6 g, water
◦
6
0 mL, 150 C, 6 h. Catalyst: H-ZSM-5, (Si/Al = 40) 30 mg (a).
Table 2
100
Rate of xylose conversion in pure water or in aqueous solution of acetic acid. Influ-
ence of the presence of solid acids.
conversion
yield
8
6
0
0
Rate of xylose conversiona (10 mol h
−4
−1
)
Catalyst
Water 100%
Acetic acid 20%
(%)
None
ZrW
Cs2HP
NbOH
K10
0.7
2.0
2.2
3.6
0.8
0.9
1.1
3.3
1.8
5.0
3.1
4.8
40
2
0
0
HY Si/Al = 15
none
ZrW
Cs HP
K10
HY
NbOH
2
a
Si/Al = 15
Calculated after 6 h of reaction. Xylose 0.6 g, catalyst 30 mg, solvent 60 mL,
◦
1
50 C.
Fig. 6. Xylose dehydration into furfural performed in pure water. Xylose 0.6 g, cat-
alyst 30 mg, water 60 mL, 150 C, 15 h.
◦
such as tungstated alumina (AlW), a promoting effect was observed,
leading to a glucose conversion of 25% in the case of AlW but unfor-
tunately to a low selectivity into 5-HMF [24]. These solids were
assumed to isomerize glucose into fructose, the latter being more
easily converted in the reactive aqueous acetic acid medium, shown
to be very efficient to dehydrate selectively fructose into 5-HMF. In
the present work, we observe from Fig. 4 that when the homo-
geneous acetic acid medium is combined to a heterogeneous acid
catalyst, the xylose conversion and the furfural yields are promoted
From Fig. 5a, one can deduce that the aqueous solutions of acetic
acid are not efficient to catalyze the cyclodehydration of xylose in
furfural at 150 C, nor the anhydride acetic acid which leads to a full
xylose conversion, but a limited amount of furfural. Other products
were detected by HPLC but not identified. From the experiments
done in the presence of the acidic zeolite, H-ZSM-5, it is seen in
Fig. 5b that an improved yield in furfural is obtained only when the
zeolite is used together with a concentrated solution of acetic acid,
of 20 wt%.
◦
in a general rule, but this is not true for Cs HP. For NbOH, HY and K10
2
the xylose conversion and the furfural yields overreached the sum
of the contribution of the solid acids in water in one part and the
homogeneous acetic acid solution in another part. This is indicative
of a synergy effect between the homogenous reactive phase, 20 wt%
aqueous solution of acetic acid, and the heterogeneous solid acid
sites. In order to quantify the synergy effect observed between the
acetic acid solution and the different heterogeneous acid catalysts,
the rates of xylose conversion have been calculated in both media,
in the presence and absence of added solid catalysts (Table 2). It is
shown that this synergy effect is marked for ZrW, NbOH, K10 and
HY catalysts.
3.2.3. Impact of more severe conditions: increase of the reaction
time and temperature
A longer reaction time of 15 h has been tested on all the catalysts,
in both media, pure water and 20 wt% aqueous acetic acid.
In pure water (Fig. 6), in the absence of solid acid, the xylose
conversion is not favored at longer reaction time. It is the opposite
in the presence of most of the solid acid catalysts except for K10.
For ZrW and Cs HP, the xylose conversion increased, particularly
2
for Cs HP, but the furfural yields did not. This may be explained by
2
It is noteworthy that the catalysts which were the most active
their too strong acidity which might promote successive transfor-
mation of adsorbed intermediates and favor some kind of polymer
and/or humins formation more or less soluble. This was supported
by the fact that carbon balances generally higher than 80% were
obtained from Total Organic Carbon (TOC) analysis for the reaction
performed in water. Note also that TGA-DTA analysis done on the
used catalysts recovered by filtration, evidenced the presence of
carboneous compounds and/or deposit. The yield of solid humins
deduced from the TGA experiment is between 3 and 5 wt%, with-
out significant differences between the two liquid media nor the
nature of the catalyst. When NbOH is used, the furfural yield grows
up to 31%, without noticeable decrease of the selectivity. The HY
zeolites presents a peculiar behavior with a marked enhancement
of its catalytic activity when the reaction time increased from 6
to 15 h: after 15 h of reaction, the furfural yield achieved 35% with
a furfural selectivity of 46%, which represents even a better per-
formance than NbOH for xylose dehydration in pure water. As a
in water, ZrW and Cs HP, respectively strong Lewis and Brønsted
2
acids, are still active in acetic acid media but more selective into fur-
fural, 41 and 65% respectively (Fig. 4). Niobium hydroxide becomes
also more selective into furfural in the acetic acid medium and then
becomes one of the most efficient catalyst for furfural production
from xylose giving a furfural yield up to 46% (60% selectivity) after
◦
6
h of reaction at 150 C. The same remark holds true for HY and
K10.
3
.2.2. Influence of the acetic acid concentration
To investigate how the concentration of acetic acid solutions
affect the furfural yield, two series of experiments were performed:
the first one using different concentration of acetic acid in the
absence of added solid catalyst (Fig. 5a), the second one in the pres-
ence of an added solid catalyst, H-ZSM-5 (Si/Al = 40) and different
concentration of acetic acid (Fig. 5b).
Please cite this article in press as: A.-C. Doiseau, et al., Synergy effect between solid acid catalysts and concentrated carboxylic acids solutions
for efficient furfural production from xylose, Catal. Today (2013), http://dx.doi.org/10.1016/j.cattod.2013.10.034

GModel
CATTOD-8660; No. of Pages9
ARTICLE IN PRESS
6
A.-C. Doiseau et al. / Catalysis Today xxx (2013) xxx–xxx
100
100
conversion
yield
conversion
8
0
0
0
0
80
yield
6
4
2
6
4
2
0
0
0
(
%)
0
0
none
ZrW
Cs HP
K10
HY
NbOH
2
Si/Al = 15
none
ZrW
Cs2HP
K10
HY
NbOH
Si/Al = 15
Fig. 7. Xylose dehydration into furfural performed in 20 wt% acetic acid. Xylose 0.6 g,
catalyst 30 mg, solvent 60 mL, 150 C, 15 h.
◦
Fig. 9. Xylose dehydration into furfural performed in 20 wt% acetic acid. Xylose 0.6 g,
catalyst 30 mg, solvent 60 mL, 180 C, 6 h.
◦
general remark, if for a category of catalysts, such as strong Lewis or
Brønsted acids, longer reaction times have increased the formation
of by-products such as levulinic acid and humins, for weaker solid
acids, presenting Lewis/Brønsted acid sites, such as HY and NbOH,
prolonged reaction times in water at 150 C is not detrimental for
the furfural selectivity.
In 20 wt% acetic acid solution (Fig. 7), a longer reaction time in
the absence of solid catalyst enhanced the xylose conversion and
the furfural yield up to 27% and 19%, respectively. Again, addition
Table 3
Furfural stability in different media and conditions.
Medium
Temperature ( C) Time (h) Furfural loss (%)a
◦
◦
Water
Water + 30mgNbOH
Water
150
150
150
6
6
15
5
15
15
20% Acetic acid
150
6
6
15
15
10
22
2
0% Acetic acid + 30mgNbOH 150
of ZrW and Cs HP, raised the xylose conversion as regards to the
2
20% Acetic acid
150
pure water medium together with the furfural yield and selectivity.
This tendency is not observed in pure water, the general rule in
Water
Water
180
180
6
15
22
39
water medium over ZrW and Cs HP is to observe a decrease of
2
2
2
0% Acetic acid
0% Acetic acid
180
180
6
15
12
64
the furfural selectivity when the conversion increases. Again, as
regard to the pure water medium, for longer reaction time over
K10, HY and NbOH, the improved activity is not accompanied by
an enhancement of the furfural selectivity.
a
Calculated introducing 100 mg of furfural in 60 mL of water or 20 wt% acetic acid
in water.
The xylose dehydration was also investigated at a higher tem-
◦
perature, 180 C, in the two media. At this temperature, the blank
experiments, done without solid catalyst, shows that the two media
catalyze xylose dehydration into furfural, in water a xylose con-
version higher than 40% was obtained with a furfural yield of 27%
of furfural yields accounting for the likely furfural degradation in
these severe conditions. However, higher furfural yields are still
observed in the acetic acid medium, i.e. the furfural yield is still
close 50% after 15 h of reaction at 180 C, in the aqueous acetic acid
solution over K10.
(Fig. 8), the acetic acid medium led to a complete xylose conver-
◦
sion with a furfural yield of 44% (Fig. 9). As a general comment, one
can observe that in acetic acid solution high xylose conversions are
observed, >90%, with furfural yield equals or higher than 40%. Note
that in the presence of the following catalysts, Cs HP, K10 and HY,
2
3
.2.4. Evaluation of furfural stability as a function of the medium,
furfural yields close to 55% are achieved. In water, lower conversion
levels are observed together with lower furfural yields, usually less
temperature, time and NbOH addition
Most of the data provided here correspond to the composi-
tion of the reaction medium at a given time since we did not
perform specific kinetic studies. Only the data obtained at the low-
◦
than 40% (Fig. 8). Therefore, at 180 C, we can still underline the
positive impact of the aqueous acetic acid medium to limit by-
products formation at the expense of furfural. Experiments were
◦
est temperature, 150 C, and short reaction time, 6 h, where the
◦
also conducted for longer reaction times at 180 C. In both liquid
xylose conversion is limited, can be considered as “initial data”
with a limited influence of the products adsorption/transformation.
Indeed, in the preliminary experiments, we have evidenced over
NbOH, in acetic acid medium, that the maximum furfural yield is
reached between 6 h and 15 h (Fig. 1). Thus, to clarify the impact of
the furfural instability on the data presented above, this was inves-
tigated in the different media and in the various used conditions.
Results are presented in Table 3.
From these data, it is clear that furfural is less stable in the
aqueous acetic acid medium than in pure water. However, to our
surprise, this difference of furfural stability in these two media does
not seem to be enhanced at higher temperature and prolonged
media, 100% water and the aqueous acetic acid solution, almost
total xylose conversions were obtained with in general a decrease
100
conversion
yield
8
0
0
0
0
6
4
2
(
%)
◦
reaction time. If after 6 h of reaction at 150 C, a maximum of 15% of
furfural has reacted in aqueous acetic acid medium and only 5% in
water, at 180 C after 15 h, the loss in furfural achieved 64% in aque-
ous acetic acid against 39% in water. Notably, upon NbOH addition,
the furfural stability was increased in acetic acid media and the
contrary is true in pure water.
0
◦
none
ZrW
Cs HP
K10
HY
NbOH
2
Si/Al = 15
Fig. 8. Xylose dehydration into furfural performed in pure water. Xylose 0.6 g, cat-
alyst 30 mg, water 60 mL, 180 C, 6 h.
◦
Please cite this article in press as: A.-C. Doiseau, et al., Synergy effect between solid acid catalysts and concentrated carboxylic acids solutions
for efficient furfural production from xylose, Catal. Today (2013), http://dx.doi.org/10.1016/j.cattod.2013.10.034

GModel
CATTOD-8660; No. of Pages9
ARTICLE IN PRESS
A.-C. Doiseau et al. / Catalysis Today xxx (2013) xxx–xxx
7
Fig. 10. (a) SEM image of fresh NbOH. (b) SEM image of used NbOH.
Table 4
1
00
100% water
Catalyst leaching given in terms of % of the leached elements.
9
0
0
20wt% acetic acid
Catalyst
100% water (metal)
20% acetic acid (metal)
8
Cs HP
ZrW
2.3 (W)
–
<0.2 (Nb)
3 (Al)
15 (Al)
2.7 (W)
–
0.2 (Nb)
25 (Al)
61 (Al)
70
2
K10
Cs2HPa
K10
NbOH
HY
ZrW
NbOH
6
0
HY
NbOH
K10
HY Si/Al = 15
K10
50
ZrW
NbOH
K10
NbOH
4
3
0
0
HY
Data obtained from the catalyst composition and the chemical analysis of the reac-
Cs HP
2
tion liquid media recovered after filtration of the used catalyst. Reaction conditions:
◦
20
10
0
1
50 C for 15 h.
ZrW
Cs HP
a
ZrW
Cs2HPW12O40 catalyst cannot be separated by filtration.
2
Cs HP
2
0
20
40
60
80
100
120
3.2.5. Catalysts stability
Xylose conversion (%)
To assess the stability of the catalysts, we have studied the
Fig. 11. Furfural selectivities as a function of xylose conversion. Dependence on the
leaching of the catalysts using elemental analysis of the recovered
reaction liquid after filtration. Results are given in Table 4. Eval-
uations of the leaching were based on the analysis of the main
elements: W, Nb and Al for ZrW, NbOH, K10 and HY, respectively.
This study shows that the K10 clay and the HY zeolite are not
stable in pure hot water, nor in the aqueous acetic acid medium;
the worst conditions being the hot aqueous acetic acid medium.
Fortunately, ZrW and NbOH present a relative stability in these
media and this is particularly true for NbOH, a selective catalyst
into furfural from xylose cyclodehydration.
liquid medium: 100% water versus 20 wt% aqueous acetic acid solution. Xylose 0.6 g,
catalyst 30 mg, solvent 60 mL, 150 C, 6 or 15 h.
◦
provide as well Lewis and Brønsted sites of moderate strength.
Indeed, NbOH provides heterogeneous superficial acid sites with
a dominant Lewis acidity since NbOH is stable is the acetic acid
media. On the contrary, HY and K10, known as mainly Brønsted
solid acids, are not stable in the acetic acid medium, they release
important amount of Al species which might act as soluble Lewis
species.
The morphology of fresh and used NbOH was observed SEM.
Fig. 10 shows the presence of additional spherical particles, char-
acteristic of the formation of solid humins.
It was recently proposed that xylose dehydration into furfural
might proceed efficiently via an intermediate aldose-ketose isom-
erization (xylose to xylulose) promoted by a solid base, and further
cyclodehydration of xylulose, a ketopentose, into fufural [31,32].
Such a picture is now well accepted for glucose dehydration
into 5-HMF which requires the intermediate isomerization of glu-
cose into fructose, the aldohexose which is easily dehydrated into
5-HMF. Also, it is known for a long time that the glucose isomeriza-
tion step is catalyzed by bases [33,34], but recently, the efficiency
of Lewis solid acids has been demonstrated even in water medium
[35]. In a similar way, the participation of Lewis acid sites to the
efficient furfural production from xylose was recently proposed by
Stein et al. [36] and Yang et al. [37] Thus, in the present work,
by combining different solid acid catalysts to the concentrated
aqueous carboxylic acid solution, resulting in high selectivities and
furfural yields, a reasonable explanation is to ascribe the synergic
effect to the Lewis acidity of moderate strength of the solid acid
catalysts in the case of NbOH or of the leached Al species for K10
and YH which have lost 25% and 61% of their Al content in the acetic
acid solution. Indeed, these solids could provide the Lewis acid sites
of adequate strength to assist the aldose-ketose isomerization and
therefore accelerate the rate of furfural formation by cyclodehy-
dration, a Brønsted acid catalyzed step (route 2, Scheme 2). Over
NbOH, a stable catalyst in the acetic acid medium, the aldose-ketose
3.3. Discussion
In an attempt to rationalize the impact of the two liquid media,
we plotted in Fig. 11 the furfural selectivities as a function of xylose
◦
conversions, in the presence of solid acid catalysts, at 150 C and
for 6 h of reaction time (conditions which do not lead to significant
xylose conversion without solid catalysts).
This figure evidences that the aqueous acetic acid solution is a
more adequate medium than pure water to produce selectively fur-
fural from xylose. For each medium, the promoting effect depends
on the nature of the solid acids.
In pure water, Fig. 11 shows the negative impact of strong solid
acid addition, such as ZrW and Cs HP. On the opposite, weaker
2
acids such as NbOH and HY favor the furfural formation from xylose
in pure water, with a furfural selectivity higher than 40% at high
conversion, 80%.
In concentrated aqueous solutions of acetic acid, it is also
clearly evidenced that the use of solid acids of moderate strength
such as HY, K10 or NbOH enhances the furfural selectivity at
high xylose conversion level (a maximum selectivity of 54% at
full conversion for HY). We can consider that these three solids
Please cite this article in press as: A.-C. Doiseau, et al., Synergy effect between solid acid catalysts and concentrated carboxylic acids solutions
for efficient furfural production from xylose, Catal. Today (2013), http://dx.doi.org/10.1016/j.cattod.2013.10.034

GModel
CATTOD-8660; No. of Pages9
ARTICLE IN PRESS
8
A.-C. Doiseau et al. / Catalysis Today xxx (2013) xxx–xxx
O
H
C
OH
HO
H
HO
H
OH
H+
H
HO
-
3 H O
CH OH
2
O
O
2
OH
O
furfural
CH OH
2
route 1
xylose
Lewis acid sites
CH OH
2
O
OH
HO
H+
HO
H
H
OH
CH OH
-
3 H O
O
O
OH
2
O
2
CH OH
furfural
2
route 2
xylulose
Scheme 2. Possible pathways for catalytic furfural synthesis from xylose.
Table 5
isomerization would occur on the solid catalyst surface, while over
HY and K10, leached Al species might also catalyze this isomeriza-
tion step.
Adsorption of xylose and furfural over NbOH from solutions of xylose and furfural
in 100% water or 20 wt% acetic acid.
◦
Components
Liquid media
Total adsorption (␮mol g−1)a
At high temperatures, 180 C, it is seen that high rates of furfural
formation are obtained even without out added solid catalysts. This
indicates that a direct pathway, involving a unique step, the direct
xylose cyclodehydration into fufural is a pure Brønsted acid cat-
alyzed step but a more energy demanding pathway (route 1, in
Scheme 2).
Xylose
Xylose
Furfural
Furfural
Pure water
20 wt% acid acetic in water
Pure water
28
37
129
2430
20 wt% Acid acetic in water
1
g of NbOH in suspension in 25 mL of xylose and furfural solutions (10 g L−1). Agita-
◦
tion of the suspension for 4 h, under isothermal conditions at 30 C. Determination
of xylose and furfural adsorption by HPLC analysis.
A last but not least phenomenon might also contribute to this
synergic effect between the concentrated aqueous acetic acid solu-
a
Adsorbed amount per gram of catalyst.
◦
tion and the solid acid catalysts at 150 C. Indeed, from the literature
data it is generally reported that over-catalytic amounts of solid
catalysts were systematic used to observe an effect of the solid cata-
lyst. It became more and more evident that the limited sugar affinity
with the catalyst surface might be the key issue which infer the con-
trol of the sugar transformation on the catalyst surface when the
reaction is done in water. One can anticipate that the carbohydrates
molecules, which are highly hydroxylated organic molecules with
high affinity with the liquid water molecules, must be adsorbed
in a limited way from an aqueous solution to a solid surface. Thus
the peculiar activity of NbOH (not calcinated) in water could find
an explanation in its high hydroxylated form and its high acidic
density which might favor the carbohydrate adsorption. Moreover,
the question raised by the synergy effect observed by combining
the aqueous concentrated acetic acid solution and solid acids could
account for an enhanced solid surface coverage by the carbohy-
drates, a consequence of the high concentration of the carboxylic
acid which would lower the bonding interaction between the car-
bohydrates and the water molecules.
This idea has prompted us to compare the adsorption of xylose,
and furfural on NbOH, in water or in 20 wt% acetic acid solution.
The adsorption data are reported in Table 5.
The adsorption of furfural is by far higher than that of xylose.
Interestingly, the addition of acetic acid significantly enhanced the
adsorption of furfural on NbOH while the effect is rather limited on
the xylose adsorption. This might be due to distinct intermolecu-
lar interaction between carbohydrates or furfural molecules with
the liquid media ascribed to distinct polarities and hydrogen bond-
ing network. Note that the total amount of furfural adsorbed on
NbOH in aqueous solution of acetic acid over-reaches the total num-
ber of acid sites of NbOH, the adsorption measured in liquid phase
involves the irreversible and reversible adsorption.
These data suggest that a higher catalyst surface coverage with
xylose and furfural exists and may contribute to the synergy effect
between the carboxylic acid solution and the solid acids, at the ori-
gin of the enhanced furfural formation from xylose. However, if
Please cite this article in press as: A.-C. Doiseau, et al., Synergy effect between solid acid catalysts and concentrated carboxylic acids solutions
for efficient furfural production from xylose, Catal. Today (2013), http://dx.doi.org/10.1016/j.cattod.2013.10.034

GModel
CATTOD-8660; No. of Pages9
ARTICLE IN PRESS
A.-C. Doiseau et al. / Catalysis Today xxx (2013) xxx–xxx
9
the favored xylose adsorption account for the conversion increase,
the enhanced furfural adsorption on NbOH would have a negative
impact on the furfural production considering that further forma-
tion of levulinic acid and humins might be favored. Interestingly,
the results do not support this scheme. Indeed, in Figs. 3 and 4,
we can see that upon acetic acid addition, the xylose conversion
was increased together with the furfural selectivity. Let us recall
that, unfortunately, levulinic acid cannot be quantified due to over-
lapping peaks with acetic acid in the HPLC chromatogram. Then,
to clarify the impact of acetic acid on the selectivity, we have also
quantify by TGA, the weight amount of solid humins recovered with
the aged catalysts by filtration: the yields in solid humins are similar
in both liquid media in presence of NbOH, close to 4 wt%. These data
preclude the favored successive formation of by-products caused
by the improved adsorption of furfural on the catalyst surface. To
strengthen this proposition, the furfural stability was also evalu-
ated in the presence of NbOH in water and in acetic acid solution
the uncontrolled homogenous ones since it was shown that upon
NbOH addition, the furfural stability was improved in acetic acid
medium.
We think that the use of water-acetic acid reactive media in the
presence of solid acid catalysts offer a new sustainable possibility
for furfural production.
Acknowledgements
This work was supported by FUI (Fonds unique intermin-
istériel), FEDER (Fonds européen de développement régional),
Conseil Régional de Picardie and Conseil Régional de la Réunion
within the PENTOVAL project.
References
[
[
1] P. Gallezot, Chem. Soc. Rev. 41 (2012) 1538–1558.
2] A.S. Mamman, J.M. Lee, Y.C. Kim, I.T. Hwang, N.J. Park, Y.K. Hwang, J.S. Chang,
J.S. Hwang, Biofuels Bioprod. Biorefining 2 (2008) 438–454.
[3] H.J. Brownlee, C.S. Miner, Ind. Eng. Chem. 40 (1948) 201–204.
[4] D.R. Arnold, A novel Process for Furfural Production in School of Chemical
Engineering, University of Natal, South Africa, 2003.
(
see Table 3). We show that, upon NbOH addition in water, the
furfural stability was decreased while in acetic acid aqueous solu-
tion, in presence of NbOH, the furfural stability was improved by
◦
3
0% after 6 h of reaction at 150 C: 15% of furfural were lost in
water against 10% in acetic acid solution. Therefore, according to
the unchanged solid humins yield upon acetic acid addition, to the
enhanced furfural adsorption on NbOH in the presence of acetic acid
and to the proved enhanced furfural stability, it seems reasonable
to propose that the degradation reactions of furfural on the NbOH
surface are slow down compared to the furfural degradation occur-
ring in the homogeneous aqueous acetic acid liquid medium. In
other words, the observed phenomenon seems to be caused by the
furfural and xylose partition between the homogeneous phase and
the catalyst surface, which was changed upon reactive co-solvent
addition, acetic acid, in favor of the catalyst coverage. Accordingly,
the superficial reactions would be favored at the expense of the
un-selective homogeneous reactions.
[
5] C. Moreau, R. Durand, D. Peyron, J. Duhamet, P. Rivalier, Ind. Crop. Prod. 7 (1998)
5–99.
[6] R. O’Neill, M.N. Ahmad, L. Vanoye, F. Aiouache, Ind. Eng. Chem. Res. 48 (2009)
300–4306.
7] P.L. Dhepe, R. Sahu, Green Chem. 12 (2010) 2153–2156.
8] S. Lima, M.M. Antunes, A. Fernandes, M. Pillinger, M.F. Ribeiro, A.A. Valente,
Appl. Catal. A 388 (2010) 141–148.
9
4
[
[
[
9] S. Lima, A. Fernandes, M.M. Antunes, M. Pillinger, F. Ribeiro, A.A. Valente, Catal.
Lett. 135 (2010) 41–47.
[
10] J. Lessard, J.F. Morin, J.F. Wehrung, D. Magnin, E. Chornet, Top. Catal. 53 (2010)
1231–1234.
11] S. Lima, M. Pillinger, A.A. Valente, Catal. Commun. 9 (2008) 2144–2148.
12] E. Lam, J.H. Chong, E. Majid, Y.L. Liu, S. Hrapovic, A.C.W. Leung, J.H.T. Luong,
Carbon 50 (2012) 1033–1043.
[13] J.L. Cabezas, L.A. Barcena, J. Coca, M. Cockrem, J. Chem. Eng. Data 33 (1988)
35–437.
14] J.R. Croker, R.G. Bowrey, Ind. Eng. Chem. Fundam 23 (1984) 480–484.
15] E.I. Gurbuz, S.G. Wettstein, J.A. Dumesic, ChemSusChem 5 (2012) 383–387.
[
[
4
[
[
4
. Conclusion
[16] A.S. Dias, M. Pillinger, A.A. Valente, Appl. Catal. A 285 (2005) 126–131.
17] A.S. Dias, M. Pillinger, A.A. Valente, Micro. Meso. Mater. 94 (2006) 214–225.
18] A.S. Dias, S. Lima, M. Pillinger, A.A. Valente, Carbohydr. Res. 341 (2006)
[
[
In this work, we have studied the influence of the presence of
2946–2953.
solid acid catalysts in aqueous acetic acid solution for the trans-
formation of xylose into furfural. Depending on the acidic features
of the solid catalyst, especially the presence of Brønsted and Lewis
acid sites, eventually provided by the leached Al element in some
cases, we observed a synergy effect between the reaction medium
and the added solid acid catalyst. This synergy effect is tentatively
explained by two different phenomenons:
[19] A.S. Dias, S. Lima, D. Carriazo, V. Rives, M. Pillinger, A.A. Valente, J. Catal. 244
(2006) 230–237.
[
20] A.S. Dias, S. Lima, P. Brandao, M. Pillinger, J. Rocha, A.A. Valente, Catal. Lett. 108
2006) 179–186.
21] A.S. Dias, S. Lima, M. Pillinger, A.A. Valente, Catal. Lett. 114 (2007) 151–160.
(
[
[22] J.N. Chheda, Y. Roman-Leshkov, J.A. Dumesic, Green Chem. 9 (2007) 342–350.
[
23] E.I. Gurbuz, J.M.R. Gallo, D.M. Alonso, S.G. Wettstein, W.Y. Lim, J.A. Dumesic,
Angew. Chem. Int. Ed. 52 (2013) 1270–1274.
[
[
24] R. Lopes de Souza, H. Yu, F. Rataboul, N. Essayem, Challenges 3 (2012) 212–232.
25] F. Figueras, N. Essayem, C. Feche, S. Loridant, J. Palomeque, G. Gelbard, WO
2
004/004893.
26] N. Essayem, G. Coudurier, M. Fournier, J.C. Vedrine, Catal. Lett. 34 (1995)
23–235.
-
The participation of the Lewis acid sites provided by the solid
catalysts itself or by the leached Al species in the overall reaction
pathway. These sites might assist the aldose-ketose isomerization
and makes the successive Brønsted acid catalyzed dehydration
[
2
[27] F. Chambon, F. Rataboul, C. Pinel, A. Cabiac, E. Guillon, N. Essayem, Appl. Catal.
B 105 (2011) 171–181.
28] B. Hamad, R.O.L. de Souza, G. Sapaly, M.G.C. Rocha, P.G.P. de Oliveira, W.A.
Gonzalez, E.A. Sales, N. Essayem, Catal. Commun. 10 (2008) 92–97.
[29] A.L. Mota Salinas, G. Sapaly, Y. Ben Taarit, J.C. Vedrine, N. Essayem, Appl. Catal.
A 336 (2008) 61–71.
[
◦
step to occur faster at the moderate temperature of 150 C.
-
The presence of high concentration of acetic acid in water was
proposed to enhance the surface catalyst coverage with the
carbohydrate and furfural. We think that upon acetic acid addi-
tion a simple solvent effect may intervene, producing weaker
intermolecular interactions between carbohydrates or furfural
molecules with the acetic acid/water media due to distinct polar-
ities and hydrogen bonding network. Therefore the partition of
xylose and furfural between the liquid phase and the surface can
be changed in favor of an improved surface coverage, as shown by
the adsorption measurements. Consequently, the more selective
heterogeneous catalyzed steps could occurred at the expense of
[
30] R. Weingarten, G.A. Tompsett, W.C. Conner, G.W. Huber, J. Catal. 279 (2011)
74–182.
1
[31] A. Takagaki, M. Ohara, S. Nishimura, K. Ebitani, Chem. Lett. 39 (2010) 838–840.
[32] V. Choudhary, S.I. Sandler, D.G. Vlachos, ACS Catalysis 2 (2012) 2022–2028.
33] C. Moreau, R. Durand, A. Roux, D. Tichit, Appl. Catal. A 193 (2000) 257–264.
34] R.O.L. Souza, D.P. Fabiano, C. Feche, F. Rataboul, D. Cardoso, N. Essayem, Catal.
Today 195 (2012) 114–119.
[
[
[35] Y. Roman-Leshkov, M. Moliner, J.A. Labinger, M.E. Davis, Angew. Chem. Int. Ed.
9 (2010) 8954–8957.
36] T. vom Stein, P.M. Grande, W. Leitner, P.D. de Maria, ChemSusChem 4 (2011)
592–1594.
[37] Y. Yang, C.W. Hu, M.M. Abu-Omar, ChemSusChem 5 (2012) 405–410.
4
[
1
Please cite this article in press as: A.-C. Doiseau, et al., Synergy effect between solid acid catalysts and concentrated carboxylic acids solutions
for efficient furfural production from xylose, Catal. Today (2013), http://dx.doi.org/10.1016/j.cattod.2013.10.034