Efficient process for the conversion of xylose to furfural with acidic ionic liquid

Article abstract of DOI:10.1139/V10-153

An efficient and simple process for the dehydration of xylose into furfural was studied, which was catalyzed by 1-(4-sulfonic acid)butyl-3- methylimidazolium hydrogen sulfate (IL-1) under mild conditions. A xylose conversion of 95.3% with a furfural yield of 91.45% was achieved in a 25 min reaction time at 150 °C; the xylose conversion and furfural yield greatly depended on the operation conditions, such as reaction time and temperature. IL-1 could be recycled and exhibited constant activity for five successive trials. This system was an important improvement, relative to the reactions using mineral acids at high temperatures, for further development on a pilot scale.

Full text of DOI:10.1139/V10-153

83
Efficient process for the conversion of xylose to
furfural with acidic ionic liquid
Furong Tao, Huanling Song, and Lingjun Chou
Abstract: An efficient and simple process for the dehydration of xylose into furfural was studied, which was catalyzed by
1-(4-sulfonic acid)butyl-3-methylimidazolium hydrogen sulfate (IL-1) under mild conditions. A xylose conversion of
95.3% with a furfural yield of 91.45% was achieved in a 25 min reaction time at 150 8C; the xylose conversion and furfu-
ral yield greatly depended on the operation conditions, such as reaction time and temperature. IL-1 could be recycled and
exhibited constant activity for five successive trials. This system was an important improvement, relative to the reactions
using mineral acids at high temperatures, for further development on a pilot scale.
Key words: xylose, dehydrate, furfural, ionic liquids.
´
´
´
´
´
´
´
Resume : On a etudie une methode simple et efficace de deshydratation du xylose en furfural qui est catalysee par le sul-
´
´
fate acide de l’acide 1-(4-sulfonique)butyl-3-methylimidazolium (IL-1) dans des conditions douces. Avec un temps de re-
`
´
´
action de 25 min a une temperature de 150 8C, on a obtenu un degre de conversion de 95,3 % du xylose en furfural et un
´
rendement en furfural de 91,45 %; la conversion du xylose et le rendement en furfural dependent fortement des conditions
´
`
´
´
ˆ
´
´
de la reaction, particulierement du temps de reaction et de la temperature. Le IL-1 peut etre recycle et il presente encore
´
`
´
`
`
´
une activite constante apres cinq essais successifs. Dans l’optique d’un developpement d’un systeme a l’echelle pilote, ce
`
`
´
´
`
´
`
systeme correspond a une amelioration importante par rapport aux reactions faisant appel a des acides mineraux et a des
´
´
´
temperatures plus elevees.
´
´
Mots-cles : xylose, deshydratation, furfural, liquides ioniques.
´
[Traduit par la Redaction]
into furans in ionic liquids; the yield of furfural was about
30%, and it was difficult to separate furfural from the reac-
tion system.
Introduction
Diminishing fossil fuel reserves and growing concerns
about global warming indicate that sustainable sources of
energy are needed in the near future.1–5 The development of
the chemistry of furanic compounds from carbohydrates for
the preparation of nonpetroleum polymeric materials was re-
cently updated.6–13 Within this framework, furfural is one of
the key compounds that is produced on an industrial scale,
and improving the chemical technology for its production
still remains of great interest.
At present, furfural is usually obtained through dehydra-
tion of pentoses, particularly xylose or hemicelluloses, at
high temperature (200–250 8C) or in the presence of mineral
acids as catalysts.14 Under these conditions, the selectivity
of furfural does not exceed 70%. Sako et al.15 reported that
in the case of continuous extraction with supercritical CO2,
the yield of furfural reached 80% in the hydrolysis of xy-
lose. Moreau et al.16 demonstrated that the selectivity of fur-
fural was up to 90% over microporous solid acid catalysts at
170 8C from xylose. Recently, Zhang and Zhao17 studied
microwave-assisted conversion of lignocellulosic biomass
In this work, we developed a simple and efficient cata-
lytic process for the conversion of xylose into furfural, using
water as a pertinent solvent, methylisobutylketone (MIBK)
as a phase modifier,18 and the acidic ionic liquid, 1-(4-
sulfonic acid)butyl-3-methylimidazolium hydrogen sulfate
(IL-1), was used as catalyst. With the given catalytic process
and with xylose as the starting material, a furfural yield of
91.4% could be achieved at 150 8C for a reaction time of
25 min. The IL-1 could be recycled, and the catalyst exhib-
ited constant activity over five cycles of evaluation.
Experimental
Materials
D-(+)-Xylose (98%), 1-methyl imidazole (CP, >99%), and
1,4-butanesultone (>99%) were purchased from Alfa Aesar
and used without further purification. 4-Methyl-2-pentanone
(AR, >90%), furfural (CP, >90%), and acetonitrile (HPLC)
were purchased from Tianjin Chemical Reagent Company
Received 28 May 2010. Accepted 25 August 2010. Published on the NRC Research Press Web site at canjchem.nrc.ca on 23 December
2010.
F. Tao. State Key Laboratory for Oxo Synthesis and Selective Oxidation, Lanzhou Institute of Chemical Physics, Chinese Academy of
Sciences, Lanzhou 730000, P. R. China; Graduate School of Chinese Academy of Sciences, Beijing 100049, P. R. China.
H. Song and L. Chou.1 State Key Laboratory for Oxo Synthesis and Selective Oxidation, Lanzhou Institute of Chemical Physics,
Chinese Academy of Sciences, Lanzhou 730000, P. R. China.
1Corresponding author (e-mail: ljchou@licp.cas.cn).
Can. J. Chem. 89: 83–87 (2011)
doi:10.1139/V10-153
Published by NRC Research Press

84
Can. J. Chem. Vol. 89, 2011
(Tianjin, China). All other reagents and solvents were re-
agent grade and were used as received.
Fig. 1. Conversion of xylose to furfural in the presence of different
catalysts (xylose = 1.0 g, catalyst = 0.5 g, H2O = 1.5 mL, methyl-
isobutylketone (MIBK) = 8 mL, T = 150 8C, t = 25 min, P = 1 atm;
1 atm = 101.325 kPa).
Preparation of IL-1
1-Methyl imidazole (16.4 g) and 1,4-butanesultone
(27.28 g) were mixed in a flask (250 mL), stirred for 17 h
at 42–45 8C, and a white solid was received. Then the white
solid was ground, washed repeatedly with a small portion of
ether, filtrated, and vacuumized for 4 h. Equimolar white
solid was added to H2SO4 (98%), and the mixture was
stirred at 80 8C for 6 h. The obtained viscous liquid was
washed with ether three times and dried in vacuum to form
the ionic liquid IL-1 (1-(4-sulfonic acid)butyl-3-methyl-
1
imidazolium hydrogen sulfate). H NMR (400 MHz, D2O)
d: 1.53 (m, 2H), 1.80 (m, 2H), 2.72 (t, 2H), 3.67 (s, 3H),
4.02 (t, 2H), 7.24 (d, 2H), 8.47 (s, 1H). ESI-MS m/z:
(+)218.6, (–)96.3.
General procedure for the conversion of xylose to
furfural
Experiments were carried out in a stainless steel autoclave
with a glass liner tube (100 mL); the reactor was loaded
with 1.0 g of xylose, 0.5 g of catalyst, 1.5 mL of H2O, and
8 mL of MIBK. The reaction mixture was heated by an
electrical heating jacket to 150 8C for 25 min, and then the
reactor was removed and quickly quenched in a cool water
bath. After filtration, extraction, and separation, the organic
and aqueous phases were collected to characterize the prod-
ucts. For the recycling of IL-1, furfural was extracted from
the water phase five times with 8 mL of ethyl acetate. After
extraction, the aqueous phase was heated at 60 8C for 24 h
in a vacuum oven to remove water and residual ethyl ace-
tate. The IL-1 was then used directly for the next run by
adding xylose and solvents. All results were replicated at
least three times.
Fig. 2. Effect of catalyst dosage on xylose conversion and furfural
yield (xylose = 1.0 g, H2O = 1.5 mL, methylisobutylketone
(MIBK) = 8 mL, T = 150 8C, t = 25 min, P = 1 atm; 1 atm =
101.325 kPa).
Analyses
The xylose concentration in the products was analyzed by
the dinitrosalicylic acid (DNS) method.19 Analysis of furfu-
ral was performed by HPLC on an Waters Alliance 2695
series equipped with a Waters 2996 PDA detector and an in-
tersil ODS-EP C18 reversed-phase column (250 mm Â
4.6 mm, 5 mm). During this process, the column temperature
remained constant at 30 8C, and the mobile phase applied was
water–acetonitrile (15:85, v/v) at a flow rate of 0.5 mL/min
with UV detection at 280 nm.
other hand, IL-1, SnCl2, and H2SO4 showed very high cata-
lytic activity for the conversion of xylose to furfural. Under
the reaction conditions, IL-1 gave a 95.3% of xylose conver-
sion and 91.4% of furfural yield at 150 8C for 25 min. Since
the results were most favorable for IL-1, this catalyst was
chosen to be studied in detail.
Results and discussion
Xylose conversion with various catalysts
Also, compared with the results that other authors have
achieved with ionic liquids for xylose conversion to furfural,
our work showed that the efficiency and simplicity of xylose
dehydration mainly depended on the reaction media and cat-
alyst employed. It has been reported that there was an ob-
vious relationship between activity and catalyst acidity in
the reaction, and that high catalyst acidity leads to high re-
action activity.9 So we considered that the higher acidity of
IL-1 over that of other catalysts, such as AlCl3 and
To obtain efficient acidic catalysts for the hydrolysis of
xylose, the catalytic activities of different catalysts were
tested. As shown in Fig. 1, the present method was effective
for various acidic catalysts, which included mineral acids
(HCl, H2SO4), Lewis acids (AlCl3, SnCl2), heteropolyacid
(H3PW12O40), and acidic ionic liquids (IL-1, C4mimH2PO4).
It can be seen that H3PW12O40 exhibited weaker catalytic
activity than others under the same experimental conditions,
whereas HCl and C4mimH2PO4 had obvious activity. On the
Published by NRC Research Press

Tao et al.
85
Table 1. Effect of reaction conditions on the conversion of xylose
to furfural (xylose = 1.0 g, H2O = 1.5 mL, methylisobutylketone
(MIBK) = 8 mL, P = 1 atm; 1 atm = 101.325 kPa).
was performed at 150 8C for 25 min, which was a little
higher than the results for 35 min at the same temperature.
Indeed, it was reported that further degradation of furfural
led to fragmentation reactions of furfural to compounds
identified as pyruvaldehyde, formaldehyde, glyceraldehyde,
or glycolaldehyde.14 Other side reactions, such as condensa-
tion or resinification of furfural, were easier to perform than
fragmentation reactions. We considered that maybe it was
these side reactions that led to the decrease of furfural selec-
tivity. Nevertheless, it should be mentioned in the case of
xylose dehydration that the formation of heavy compounds
is less probable compared with what occurs in fructose hy-
drolysis to 5-hydroxymethylfurfural (5-HMF), which can re-
act through its hydroxymethyl function, thus yielding higher
molecular weight compounds.
Reaction
temperature time
(8C)
60
Reaction Xylose
Furfural
yield
(wt%)
Furfural
selectivity
(wt%)
conversion
(wt%)
(min)
20
40
60
15
35
55
10
25
40
5
20.5
45.5
59.7
27.3
52.9
67.4
54.1
60.7
71.3
82.3
85.7
89.0
92.1
88.6
92.7
95.3
97.5
13.9
31.0
43.8
18.4
37.5
52.7
40.9
49.0
62.1
74.4
79.2
83.5
87.6
80.3
85.1
91.4
88.7
67.8
68.1
73.4
67.4
70.9
78.2
75.6
81.7
87.1
90.4
92.4
93.8
95.1
90.6
91.8
95.9
91.0
80
100
120
Effect of initial xylose concentration
15
25
35
5
15
25
35
From a practical point of view, if higher concentrations of
xylose could be used as feedstock, the technique could be
economical. In the acid catalytic dehydration of xylose, the
initial xylose concentration has a large effect on furfural
yield and selectivity. Different initial concentrations of xy-
lose were studied, which were 0.3, 0.7, 1.0, 1.5, and 2.0 g.
It can be seen from Fig. 3 that the rate of xylose conversion
had little effect with different initial concentrations of xy-
lose, but the furfural yield reduced gradually with increasing
initial xylose concentration. As shown in Fig. 3, the highest
furfural yield was 91.4% with 1.0 g of initial xylose load-
ings, but the yield was reduced to 78% when the amount of
xylose was 2.0 g. The loss in furfural yield was thought to
be due to the higher xylose concentration, which would in-
crease the probability that reactive compounds such as xy-
lose and furfural would collide with each other or cross
polymerize.20
150
H3PW12O40, should probably be responsible for its higher
catalytic activity in this work.
Effect of catalyst dosage on xylose conversion and
furfural yield
Figure 2 shows the effect of the catalyst dosage on xylose
conversion and furfural yield. The amount of IL-1 used was
0.1, 0.3, 0.5, and 0.7 g. In the absence of catalyst, at a
150 8C reaction temperature, no xylose conversion was ob-
served for a reaction time of 25 min (not shown in Fig. 2).
In the presence of IL-1, the formation of furfural occurred.
When the catalyst dosage increased from 0.3 to 0.5 g, the
furfural yield increased from 88% to 92%. However, when
the amount of IL-1 was increased from 0.5 to 0.7 g, there
was little change in xylose conversion and furfural yield.
As the xylose conversion and furfural yield had no further
change with further IL-1 dosage over 0.5 g, this implied
that there were sufficient catalytic sites available for the
substrate xylose (1.0 g) in the system at this experimental
condition.
Influence of water content on the conversion of xylose
Since IL-1 is a hydrophilic ionic liquid, the influence of
water content on xylose dehydration was studied, which is
shown in Fig. 4. Previous works already indicated that water
had a negative effect on the dehydration of fructose to 5-
HMF;8,21–22 this point was also suitable for the hydrolysis of
xylose. As shown in Fig. 4, when the water content in the
system was <1.5 g, it had little effect on the xylose conver-
sion and furfural yield. However, as the amount of water in-
creased from 1.5 to 3.5 g, the conversion of xylose and
furfural yield decreased significantly, only 86% and 79%,
respectively. Zhao et al.23 also reported on adding water or
organic solvents such as glycerol into ionic liquids and pro-
posed that lowering the dielectric constant of the reaction
media would result in the loss of catalytic activity.
Effect of reaction temperature and time
The effects of reaction temperature and time on xylose
conversion are shown in Table 1. It can be seen that the re-
action temperature and time had a large effect on both the
xylose conversion and furfural yield. When the reaction
temperature was 60 8C, with a 40 min reaction time, the xy-
lose conversion was 45.5% and a furfural yield of 31% was
obtained. The furfural yield increased from 31% to 83.5%
when conditions changed from 60 to 120 8C in a 25 min re-
action time. A furfural yield of 91.45% could be obtained
for 25 min at a reaction temperature of 150 8C.
Recycling of IL-1
In the principles of green engineering,24 the recycling of a
catalyst is essential, so this walso also examined in our
work. Experiments were conducted at 150 8C for a reaction
time of 25 min. As the catalyst IL-1 was only in an aqueous
phase after the reaction, the product furfural was separated
from the water phase after reaction by extracting five times
with 8 mL ethyl acetate, which is similar to the procedure in
the literature.25 IL-1 and xylose were found to be insoluble
in ethyl acetate and furfural was the sole product in the
As listed in Table 1, by prolonging the reaction time, the
reaction proceeded more thoroughly, and the color of the or-
ganic phase changed from almost colorless to a deep yellow.
The selectivity of furfural was up to 96% when the reaction
Published by NRC Research Press

86
Can. J. Chem. Vol. 89, 2011
Fig. 3. Effect of xylose initial loading on the reaction (IL-1 = 0.5 g,
H2O = 1.5 mL, methylisobutylketone (MIBK) = 8 mL, T = 150 8C,
t = 25 min, P = 1 atm; 1 atm = 101.325 kPa).
Fig. 5. Recycling of IL-1 (xylose = 1.0 g, IL-1 = 0.5 g, H2O =
1.5 g, methylisobutylketone (MIBK) = 8 mL, T = 150 8C, t =
25 min, P = 1 atm; 1 atm = 101.325 kPa).
biomass as a promising substitute for petroleum-based
chemicals. It was demonstrated herein that this reaction
could be smoothly carried out by using the acidic ionic
liquid IL-1 as catalyst in a biphase separation system. The
process we developed was shown to be efficient, and a xy-
lose conversion of 95.3% with a 91.45% furfural yield was
obtained when the reaction was carried out at 150 8C for a
reaction time of 25 min. As the reaction time increased to
35 min at 150 8C, the selectivity of furfural decreased from
95.9% to 91%. Initial loadings of up to 2.0 g of xylose in
the system led to an obvious decrease in furfural yield,
which was about 14%, but the conversion of xylose seldom
decreased (~2%). A water content <1.5 g had little effect on
the xylose conversion and furfural yield, but a significant
decrease was detected both in the xylose conversion and fur-
fural yield when water content was >1.5 g. The catalyst IL-1
could be recycled almost without loss of activity for five
times after the product furfural was separated with ethyl
acetate. The results strongly suggested that the ionic liquid
IL-1 could play an important role in exploring new, effi-
cient, and easy-to-use processes for the preparation of furfu-
ral from xylose. Although the process developed efficiently
for the conversion of xylose to furfural, further work is re-
quired to improve the furfural yield under milder conditions.
Fig. 4. Influence of water content on xylose conversion (xylose =
1.0 g, methylisobutylketone (MIBK) = 8 mL, T = 150 8C, t =
25 min, P = 1 atm; 1 atm = 101.325 kPa).
ethyl acetate phase, similar to that noted by Hu et al.21 After
extraction, the aqueous phase was heated at 60 8C for 24 h
in a vacuum oven to remove the water and residual ethyl
acetate. Such an operation is rather important as three water
molecules are produced during the course of the reaction. It
was then used directly for the next run by adding an equal
amount of xylose and solvents. It can be seen in Fig. 5 that
in the recycled reaction, IL-1 gave almost a comparable
amount of furfural, which indicated that the catalyst IL-1 re-
tained a very high activity for the conversion of xylose into
furfural. The yield of furfural and xylose conversion were
not obviously decreased over five runs with the recycled cat-
alyst.
Acknowledgement
We gratefully acknowledge the financial support of the
State Key Laboratory for Oxo Synthesis and Selective Oxi-
dation of China.
References
(1) Kunkes, E. L.; Simonetti, D. A.; West, R. M.; Serrano-Ruiz,
¨
J. C.; Gartner, C. A.; Dumesic, J. A. Science 2008, 322
(5900), 417. doi:10.1126/science.1159210.
(2) Dodds, D. R.; Gross, R. A. Science 2007, 318 (5854), 1250.
doi:10.1126/science.1146356.
(3) Huber, G. W.; Corma, A. Angew. Chem. Int. Ed. 2007, 46
(38), 7184. doi:10.1002/anie.200604504.
(4) Chheda, J. N.; Huber, G. W.; Dumesic, J. A. Angew. Chem.
Int. Ed. 2007, 46 (38), 7164. doi:10.1002/anie.200604274.
Conclusions
Preparation of furfural through the dehydration reaction of
xylose is one of the most important approaches to utilize
Published by NRC Research Press

Tao et al.
87
(5) Huber, G. W.; Iborra, S.; Corma, A. Chem. Rev. 2006, 106
Hiaki, T.; Sato, M. J. Chem. Eng. of Jpn 1992, 25 (4), 372.
doi:10.1252/jcej.25.372.
(9), 4044. doi:10.1021/cr068360d.
´
(6) Roman-Leshkov, Y.; Barrett, C. J.; Liu, Z. Y.; Dumesic, J.
(16) Moreau, C.; Durand, R.; Peyron, D.; Duhamet, J.; Rivalier,
P. Ind. Crops Prod. 1998, 7 (2–3), 95. doi:10.1016/S0926-
6690(97)00037-X.
A. Nature 2007, 447 (7147), 982. doi:10.1038/nature05923.
(7) Ilgen, F.; Ott, D.; Kralisch, D.; Reil, C.; Palmberger, A.; Ko-
¨
nig, B. Green Chem. 2009, 11 (12), 1948. doi:10.1039/
b917548m.
(17) Zhang, Z.; Zhao, Z. K. Bioresour. Technol. 2010, 101 (3),
1111. doi:10.1016/j.biortech.2009.09.010.
´
(8) Qi, X. H.; Watanabe, M.; Aida, T. M.; Smith, R. L., Jr.
Green Chem. 2008, 10 (7), 799. doi:10.1039/b801641k.
(9) Bao, Q. X.; Qiao, K.; Tomida, D.; Yokoyama, C. Catal.
Commun. 2008, 9 (6), 1383. doi:10.1016/j.catcom.2007.12.
002.
(10) Lansalot-Matras, C.; Moreau, C. Catal. Commun. 2003, 4
(10), 517. doi:10.1016/S1566-7367(03)00133-X.
(11) Qi, X.; Watanabe, M.; Aida, T. M.; Smith, R. L., Jr. Catal.
Commun. 2009, 10 (13), 1771. doi:10.1016/j.catcom.2009.
05.029.
(12) Yan, H.; Yang, Y.; Tong, D.; Xiang, X.; Hu, C. Catal. Com-
mun. 2009, 10 (11), 1558. doi:10.1016/j.catcom.2009.04.020.
(13) Moreau, C.; Finiels, A.; Vanoye, L. J. Mol. Catal. Chem.
2006, 253 (1–2), 165. doi:10.1016/j.molcata.2006.03.046.
(14) Antal, M. J., Jr.; Leesomboon, T.; Mok, W. S.; Richards, G.
N. Carbohydr. Res. 1991, 217 (1), 71. doi:10.1016/0008-
6215(91)84118-X.
(18) Roman-Leshkov, Y.; Chheda, J. N.; Dumesic, J. A. Science
2006, 312 (5782), 1933. doi:10.1126/science.1126337.
(19) Wang, W. L.; Huang, X. S. Food Sci. (Beijing, China) 2006,
4, 196.
(20) Kuster, B. F. M. Starch 1990, 42 (8), 314. doi:10.1002/star.
19900420808.
(21) Hu, S. Q.; Zhang, Z. F.; Zhou, Y. X.; Han, B. X.; Fan, H. L.;
Li, W. J.; Song, J. L.; Xie, Y. Green Chem. 2008, 10 (12),
1280. doi:10.1039/b810392e.
(22) Qi, X. H.; Watanabe, M.; Aida, T. M.; Smith, R. L., Jr.
Green Chem. 2009, 11 (9), 1327. doi:10.1039/b905975j.
(23) Zhao, H. B.; Holladay, J. E.; Brown, H.; Zhang, Z. C.
Science 2007, 316 (5831), 1597. doi:10.1126/science.
1141199.
(24) Tang, S. Y.; Bourne, R. A.; Smith, R. L.; Poliakoff, M.
Green Chem. 2008, 10 (3), 268. doi:10.1039/b719469m.
(25) Yong, G.; Zhang, Y. G.; Ying, J. Y. Angew. Chem. Int. Ed.
2008, 47 (48), 9345. doi:10.1002/anie.200803207.
(15) Sako, T.; Taguchi, T.; Sugeta, T.; Nakazawa, N.; Okubo, T.;
Published by NRC Research Press