Production of furfural from xylose at atmospheric pressure by dilute sulfuric acid and inorganic salts

Article abstract of DOI:10.1016/j.carres.2011.11.023

In this paper, the dehydration of xylose to furfural was carried out under atmospheric pressure and at the boiling temperature of a biphasic mixture of toluene and an aqueous solution of xylose, with sulfuric acid as catalyst plus an inorganic salt (NaCl or FeCl₃) as promoter. The best yield of furfural was 83% under the following conditions: 150 mL of toluene and 10 mL of aqueous solution of 10% xylose (w/w), 10% H₂SO₄ (w/w), 2.4 g NaCl, and heating for 5 h. FeCl₃ as promoter was found to be more efficient than NaCl. The addition of DMSO to the aqueous phase in the absence of an inorganic salt was shown to improve the yield of furfural.

Full text of DOI:10.1016/j.carres.2011.11.023

Carbohydrate Research 350 (2012) 77–80
Contents lists available at SciVerse ScienceDirect
Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres
Note
Production of furfural from xylose at atmospheric pressure by dilute sulfuric
acid and inorganic salts
⇑
Chunguang Rong, Xuefeng Ding , Yanchao Zhu, Ying Li, Lili Wang, Yuning Qu, Xiaoyu Ma, Zichen Wang
College of Chemistry, Jilin University, Changchun 130012, People’s Republic of China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 1 October 2011
Received in revised form 17 November 2011
Accepted 22 November 2011
Available online 3 December 2011
In this paper, the dehydration of xylose to furfural was carried out under atmospheric pressure and at the
boiling temperature of a biphasic mixture of toluene and an aqueous solution of xylose, with sulfuric acid
as catalyst plus an inorganic salt (NaCl or FeCl3) as promoter. The best yield of furfural was 83% under the
following conditions: 150 mL of toluene and 10 mL of aqueous solution of 10% xylose (w/w), 10% H2SO4
(w/w), 2.4 g NaCl , and heating for 5 h. FeCl3 as promoter was found to be more efficient than NaCl. The
addition of DMSO to the aqueous phase in the absence of an inorganic salt was shown to improve the
yield of furfural.
Keywords:
Furfural
Dilute sulfuric acid
Solvent extraction
Atmospheric pressure
Inorganic salts
Ó 2011 Elsevier Ltd. All rights reserved.
Rapidly growing worldwide energy demand has triggered a
renewed interest in producing fuels from biomass to add to
worldwide energy supplies. Renewable biomass resources have
the potential to serve as a sustainable supply of fuels and chemical
intermediates.1 The challenge for the effective utilization of these
sustainable resources is to develop cost-effective processing
methods to transform highly functionalized carbohydrate moieties
into value-added chemicals.2 Biomass is made up of four major
components: cellulose, hemicellulose, lignin, and starch. Hemicel-
luloses are the second most abundant plant material after cellu-
lose. The hemicellulose present in biomass undergoes hydrolysis
in acidic media to form xylose (pentose), and then xylose is dehy-
drated to form furfural. There is no synthetic route available for
furfural production; therefore furfural is exclusively produced
from renewable biomass resources by acid-catalyzed dehydration
of pentoses. Due to its unsaturated bonds and aldehyde group, fur-
fural is a highly versatile and key derivative used in the manufac-
ture of a wide range of important chemicals, and it is likely to be of
increasing demand in different fields, such as oil refining, plastics,
pharmaceutical, and agrochemical industries.3 In addition,
researchers have recently shown that HMF and furfural can serve
as precursors for production of liquid alkanes (C7–C15) that serve
as diesel fuel components.2 About 250,000 tons of furfural,
probably the only unsaturated, large volume, organic chemical,
are prepared from carbohydrate sources annually.4
The industrial technology for furfural production relies on batch
or continuous reactors where the pentosan fraction of the
lignocellulosic is converted into monosaccharides (pentoses) by
acid hydrolysis. Further dehydration reactions of the pentoses
yield furfural. The feedstock is loaded to the digester and mixed
with an aqueous solution of inorganic acids. The system is main-
tained at the desired reaction temperature by adding steam to
the digester. Furfural is continuously extracted from the reactor
by steam distillation to minimize its loss through secondary reac-
tions of degradation and condensation. The concentration of furfu-
ral in the reactor outlet stream is around 3% (w/w), and it is
recovered in a series of distillation columns. Depending on the con-
figuration of the separation process, methanol and acetic acid may
also be obtained as marketable byproducts. At the end of the reac-
tion period, the solid residue is separated from the liquid, which
may be processed to recover the acid catalyst. The residual solid
is made up of lignin and depolymerized cellulose, and this solid
may be dried and burned in a boiler to provide steam for the plant.
Normally, when batch reactors are used, several digesters are oper-
ated in coordinated rotation so that the distillation plant can be
operated continuously. Typical reaction conditions are 3% (Wacid
/
Wsolution+lignocellulosic) sulfuric acid, an acid solution to lignocellu-
losic mass ratios of between 2:1 and 3:1, reaction temperatures
of around 170–185 °C, and reaction times of 3 h. With this technol-
ogy, maximum furfural yields are between 45% and 50% of the
potential.5
In the early work, researchers used corncob and corn stover for
furfural production.6,7 Recently, a few investigations involving
cyclodehydration of xylose in liquid phase using various
⇑
Corresponding author. Tel./fax: +86 431 85155358.
E-mail address: dingxf@jlu.edu.cn (X. Ding).
0008-6215/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2011.11.023

78
C. Rong et al. / Carbohydrate Research 350 (2012) 77–80
homogeneous or solid acid catalysts have been reported. Scientists
have studied the dehydrocyclization of xylose into furfural using
acid catalysts, including mineral acids, zeolites, acid-functionalized
MCM materials, and heteropolyacids, sulfated zirconia, exfoliated
titanate, niobate, and titanoniobate nanosheets, mesoporous
silica-supported 12-tungstophosphoric.3,8–13 Morin et al. also re-
ported the dehydration of xylose over (H+) mordenite, using a con-
tinuous two-liquid phase (aqueous-toluene) flux in a plug-flow
reactor at 260 °C and 55 atm, with a 98% conversion rate, a 98%
furfural molar yield, and a 98% furfural selectivity.14 Solid acid
has a lot of advantages, but its synthesis is very complicated. Also
it is costly and prone to inactivation. The experiments which use
heterogenous catalysts (solid acids) usually need special condi-
tions, so the application field of solid acids is constricted.
The present study deals with production of furfural from xylose.
To our knowledge, there are few studies about the dehydration of
xylose to furfural at atmospheric pressure. Only early scientists
studied the dehydration of xylose with hydrochloric acid.15,16 No
reports about vast scale organic solvent extracting furfural
production from xylose with mineral acid and inorganic salts at
atmospheric pressure exist. Our experimental condition was very
mild. Therefore, optimization of dilute sulfuric acid catalyst for
the production of furfural from xylose was performed in a biphasic
system with quantity of organic extraction solvent, initial xylose
concentration, and acid concentration as the major factors to
consider. And we chose toluene as an organic extraction solvent
because furfural had good solubility in it.
The ratio of extracting solvent (toluene) to aqueous system
affected the yield of furfural. The results of mineral acid dehydra-
tion of xylose in biphasic systems are presented in Figure 1. When
the ratio of toluene/water was 100:77 (V/V), the yield of furfural
was only 14%. With the ratio of toluene/water increased to
100:10 (V/V) and 150:10 (V/V), the yields became 21% and 33%,
respectively. When the ratio reached to 175:10 (V/V), the yield in-
creased to 43%.
The experiment was performed by stirring the biphasic system
with high speed. Water was dispersed in toluene. Each aqueous
drop could be regarded as a small reactor. The surface of aqueous
contacted with the organic phase completely. As the ratio of
toluene/water (100:77) became lower, aqueous drop was bigger
comparatively which caused low yield of furfural. When the ratio
was enhanced to 150:10 (V/V) or 175:10 (V/V), aqueous drop
became much smaller, which showed great character and resulted
in high yield of furfural. Furfural was produced in aqueous drop,
and then it was immediately extracted to the organic phase. Some
side reactions were avoided, which could decrease the yield of
furfural.
Figure 2 shows the effect of concentration of sulfuric acid on the
selectivity for dehydration of xylose and on the yield of furfural
while using NaCl as the promoter. The experiments were
conducted in water using toluene as an extracting solvent. The
yield of furfural was higher in high concentration of sulfuric acid
systems compared to that in low concentration. The best furfural
yield was 75% achieved with 10% (w/w) acid after a 5-h reaction
time. For xylose dehydration to furfural, the yield under 10% acid
was nearly a ninefold improvement than that under 2.5% (w/w)
acid. But when the concentration of sulfuric acid became 12.5%
(w/w), the yield reduced to 51%. The rate of xylose dehydration in-
creased with increasing the concentration of H+. The low yields of
furfural with 12.5% (w/w) acid could be attributed to side reac-
tions, such as xylose dehydration to other aldehydes (e.g.,
formaldehyde), degradation of furfural, or polymerizations of
intermediate, and furfural. In addition, xylose forms oligosaccha-
rides, which contain reactive hydroxyl groups leading to higher
rates of cross polymerizations with reactive intermediates and
furfural.
Different concentrations of xylose were researched at 150:10
(V/V) toluene/aqueous, with 10% acid for 5 h. The amount of NaCl
as promoter was 1.2 g. The best furfural yield achieved was 75%
at 10% (w/w) xylose (Fig. 3), which had been detected with change
of acid conditions. When the concentration of xylose was higher
than 10%, the yield of furfural reduced to 68% and 66%. The
decrease of furfural yield was not very rapid.
Previous studies have shown that the inorganic salts enhanced
xylose monomer and xylotriose degradation;17 also chloride ions
enhanced furfural formation from
D
-xylose.18 But their studies
were performed in hermetic systems and under certain temper-
atures. In our study, dehydration experiments were conducted at
atmospheric pressure with NaCl and FeCl3 as addition agents.
The adding of NaCl increased the yield of furfural from 20% to
43% with 175:10 (V/V) toluene/aqueous at 10% xylose, 5% acid
for 5 h (Fig. 4a). Then the amount of NaCl was tested to show the
effect on the yield of furfural. Furfural yield increased along with
the amount of NaCl increase. The yield reached nearly 83% when
80
70
60
50
40
30
20
10
0
40
30
20
10
0
100:77
100:10
125:10
150:10
175:10
2
4
6
8
10
12
14
Toluene:aqueous phase(V/V)
Acid concentration (%)
Figure 1. Effect of the toluene-to-aqueous phase ratio (V/V) on the furfural yield
under the following conditions: 10% xylose (w/w), 5% H2SO4 (w/w), 1.2 g NaCl,
heating for 5 h.
Figure 2. Yield of furfural at different sulfuric acid concentrations under the
following conditions: 150 mL of toluene, 10 mL of an aqueous solution of 10%
xylose (w/w), 1.2 g NaCl, heating for 5 h.

C. Rong et al. / Carbohydrate Research 350 (2012) 77–80
79
80
75
70
65
60
55
50
45
90
a
80
70
60
50
40
30
20
0.0
0.4
0.8
1.2
1.6
2.0
2.4
4
6
8
10
12
14
16
NaCl (g)
Concertration of xylose (%)
Figure 3. Effect of changing initial concentration of xylose on furfural yield under
the following conditions: 150 mL of toluene, 10 mL of an aqueous solution of 10%
H2SO4 (w/w), 1.2 g NaCl, heating for 5 h.
b
40
30
20
10
0
the amount of NaCl was 2.4 g at the toluene/aqueous ratio of
150:10 (V/V) with 10% xylose and 10% sulfuric acid (Fig. 4a). We
further studied the effect of changing the inorganic salt from NaCl
to FeCl3. As seen in Figure 4b, the yield of furfural is 45% with 0.6 g
FeCl3, higher than that with 0.6 g NaCl. At atmospheric pressure,
FeCl3 also promoted furfural formation from D-xylose better than
NaCl. When FeCl3 was used only as catalyst without sulfuric acid
added, the yield of furfural was very low (5%) with 10% xylose
for 5 h. Single FeCl3 without acid had poor advancement in xylose
dehydration to furfural. Although FeCl3 alone could serve as
catalyst accelerator at atmospheric pressure, it could not substitute
sulfuric acid in these experiments.
The addition of electrolytes to the aqueous could generally
induce significant changes by attracting or orienting its molecules,
or by changing its ordered short range structure and in general by
its interaction with the solute. In the same way, these effects could
be regarded as responsible factors for the reported results in terms
of xylose reaction rates and furfural yield. Furthermore, when com-
pared the results for FeCl3 with those for NaCl, it can be noticed
that the nature of the ions is very important. Thus a direct
contribution of the ions to the chemistry of the reaction could be
considered to play a major role.
0.6gNaCl+H2SO
4 0.6gFeCl3+H2SO4
0.6gFeCl3
Figure 4. (a) Effect of amount of NaCl on furfural yield. The red points represent the
furfural yield for dehydration in water aqueous with 175:10 (V/V) toluene/aqueous
at 10% xylose (w/w), 5% H2SO4 (w/w) for 5 h. The black line represent the furfural
yield for dehydration in water aqueous with 150:10 (V/V) toluene/aqueous at 10%
xylose (w/w), 10% H2SO4 (w/w) for 5 h. (b) Effect of FeCl3 on furfural yield under the
following conditions: 150 mL of toluene, 10 mL of an aqueous solution of 10%
xylose (w/w), 10% H2SO4 (w/w), heating for 5 h. (For interpretation of the references
to colour in this figure legend, the reader is referred to the web version of this
article.).
Earlier work has suggested that DMSO improved selectivities
and dehydration rates in formation of 5-hydroxymethylfurfural
(HMF). There was similarity between the formation of furfural
from xylose and the formation of HMF from glucose, so we studied
the effects of varying the DMSO level in the aqueous phase. Xylose
dehydration was conducted at a constant acid concentration, equal
to 10%, and 10% xylose in the presence of toluene as an extracting
solvent after 2 h. As seen in Figure 5, DMSO was added to a level of
3:7 (V/V) DMSO/water, which improved the yield from 20% to 30%,
with a further increase in yield to 46% for 5:5 (V/V) DMSO/water.
Previous study has suggested that DMSO suppresses both the
formation of condensation byproducts and the HMF rehydration
by lowering the overall water concentration. So it could be
speculated that DMSO could play the same role in furfural produc-
tion. DMSO also changed the form of xylose in water, which may
profit the formation of furfural, as the furanose form of glucose
upon adding DMSO.2
azeotropy. The process variables that influence the yields of furfu-
ral are the ratio of toluene to water, the concentration of sulfuric
acid, the DMSO content in aqueous phase, the initial sugar
concentration, and the amount of inorganic salt. The yield of furfu-
ral reached nearly 83% when NaCl was 2.4 g at the toluene/aqueous
ratio of 150:10 (V/V) with 10% xylose and 10% sulfuric acid. The
concentration of acid was considered to play the major role on
the yield of furfural. The results showed that inorganic salts could
enhance the reaction of xylose to furfural in aqueous acidic solu-
tion at atmospheric pressure, especially FeCl3.
1. Experimental section
1.1. Materials and experimental methods
This work demonstrates that a biphasic system at which the
ratio of organic phase to aqueous phase was larger than 10:1 could
be tuned to process xylose to produce furfural. All the experiments
were operated at atmospheric pressure and biphasic system
1.1.1. Materials
All chemical reagents, such as xylose, sulfuric acid, toluene,
ferric trichloride and sodium chloride, etc. were of analytical grade
and obtained from Sinopharm Chemical Regent Co. Ltd in PRC.

80
C. Rong et al. / Carbohydrate Research 350 (2012) 77–80
and detection port were 230 °C and 250 °C, respectively. The
injection volume was 1 mL. The initial temperature of the oven
was 70 °C for 1 min, followed with a ramp of 20 °C/min to final
temperature of 180 °C and then held for 2 min. Under these
conditions furfural has a retention time of 5.9 min.
40
30
20
10
0
1.3. Calculations
Xylose concentration refers to the aqueous phase concentration
because no xylose was present in the organic phase. Furfural yield
was calculated as given below.
Mfurfural;obtain
Yield ¼
Mfurfural;potential
5:5
7:3
10:0
Acknowledgements
Water:DMSO(V/V)
This research was supported by the financial support of the
China National Key Technology R&D Programme under Contract
No. 2008BAE66B00 and Jilin Provincial Environmental Protection
Department under Contract No. 200917.
Figure 5. Effect of DMSO content in water/DMSO (V/V) mixture on yield of furfural
under the following conditions: 150 mL of toluene, 10 mL of an aqueous solution of
10% xylose (w/w), 10% H2SO4 (w/w), 1.2 g NaCl, heating for 2 h.
References
1.1.2. Methods
All dehydration reactions were carried out in a two-phase batch
1. Ragauskas, A. J.; Williams, C. K.; Davison, B. H.; Britovsek, G.; Cairney, J.; Eckert,
C. A.; Williams, J. F., Jr.; Hallett, J. P.; Leak, D. J.; Liotta, C. L.; Mielenz, J. R.;
Murphy, R.; Templer, R.; Tschaplinski, T. Science 2006, 311, 484–489.
2. Chheda, J. N.; Román-Leshkov, Y.; Dumesic, J. A. Green Chem. 2007, 9, 342–350.
3. Dias, A.; Lima, S.; Pillinger, M.; Valente, A. Carbohydr. Res. 2006, 341, 2946–
2953.
4. Mamman, A. S.; Lee, J.-M.; Kim, Y.-C.; Hwang, I. T.; Park, N.-J.; Hwang, Y. K.;
Chang, J.-S.; Hwang, J.-S. Biofuels, Bioprod. Biorefin. 2008, 2, 438–454.
5. Montané, D.; Salvadó, J.; Torras, C.; Farriol, X. Biomass Bioenergy 2002, 22, 295–
304.
reactor system containing
a reactive aqueous layer and an
extracting organic layer. All experiments were conducted using a
glass flask, which was heated in a heating jacket as a batch reactor
with a working volume of 250 mL. In a typical experiment, 10 mL
of aqueous phase solution to which NaCl (or FeCl3), xylose, and
H2SO4 as acid catalyst were added and 100 mL of organic phase
solution were poured into the reactor. The reactor was placed in
a heating jacket until boiling for the time specified. After reaction,
the reactor was cooled to room temperature by flowing air.
6. Slngh, A.; Das, K.; Sharma, D. K. Ind. Eng. Chem. Res. 1984, 23, 257–262.
7. Sproull, R. D.; Bienkowski, P. R.; Tsao, G. T. Biotechnology and bioengineering
symposium; Wiley, 1986. pp. 561–577.
8. Dias, A. S.; Lima, S.; Carriazo, D.; Rives, V.; Pillinger, M.; Valente, A. A. J. Catal.
2006, 244, 230–237.
1.2. Analytical procedures
9. Dias, A. S.; Pillinger, M.; Valente, A. A. J. Catal. 2005, 229, 414–423.
10. Dias, A. S.; Lima, S.; Brandão, P.; Pillinger, M.; Rocha, J.; Valente, A. A. Catal. Lett.
2006, 108, 179–186.
11. Dias, A. S.; Lima, S.; Pillinger, M.; Valente, A. A. Catal. Lett. 2007, 114, 151–160.
12. Dias, A. S.; Pillinger, M.; Valente, A. A. Appl. Catal., A 2005, 285, 126–131.
13. Dias, A. S.; Pillinger, M.; Valente, A. A. Microporous Mesoporous Mater. 2006, 94,
214–225.
14. Lessard, J.; Morin, J.-F.; Wehrung, J.-F.; Magnin, D.; Chornet, E. Top. Catal. 2010,
53, 1231–1234.
15. Fulmer, E. I.; Christensen, L. M.; Hixon, R. M.; Foster, R. L. J. Phys. Chem. 1936,
40, 133–141.
The sugars were analyzed using HPLC (Hewlett–Packard,
Agilent 1100, 23 USA) with a Zorbax-NH2 column (4.6 ꢀ 250 mm,
5-Micron, Agilent, USA) combined with an RI detector. The mobile
phase was acetonitrile and water in a ratio of 75:25 (v/v), the
operating temperature was controlled at 25 °C, and the flow rate
was set at 0.72 mL/min. Standard sugars were run under the same
conditions.
Furfural (or furaldehyde) was analyzed by GC–FID (HP 6850,
Hewlett–Packard) using a StabilwaxÒ-DA column. The flow rate
of helium was 1.2 mL/min. The temperatures of injection port
16. Hockett, R. C.; Guttag, A.; Smith, M. E. J. Am. Chem. Soc. 1943, 65, 1–3.
17. Liu, C.; Wyman, C. Carbohydr. Res. 2006, 341, 2550–2556.
18. Marcotullio, G.; De Jong, W. Green Chem. 2010, 12, 1739–1746.