Conversion of hexose into 5-hydroxymethylfurfural in imidazolium ionic liquids with and without a catalyst

Article abstract of DOI:10.1016/j.carres.2011.03.015

Conversion of fructose and glucose into 5-hydroxymethylfurfural (HMF) was investigated in various imidazolium ionic liquids, including 1-butyl-3-methylimidazolium chloride (BmimCl), 1-hexyl-3-methylimidazolium chloride (HmimCl), 1-octyl-3-methylimidazolium chloride (OmimCl), 1-benzyl-3-methylimidazolium chloride (BemimCl), 1-Butyl-2,3-dimethylimidazolium chloride (BdmimCl), and 1-butyl-3-methylimidazolium p-toluenesulfonate (BmimPS). The acidic C-2 hydrogen of imidazolium cations was shown to play a major role in the dehydration of fructose in the absence of a catalyst, such as sulfuric acid or CrCl₃. Both the alkyl groups of imidazolium cations and the type of anions affected the reactivity of the carbohydrates. Although, except BmimCl and BemimCl, other four ionic liquids could only achieve not more than 25% HMF yields without an additional catalyst, 60-80% HMF yields were achieved in HmimCl, BdmimCl, and BmimPS in the presence of sulfuric acid or CrCl₃ in sufficient quantities.

Full text of DOI:10.1016/j.carres.2011.03.015

Carbohydrate Research 346 (2011) 956–959
Contents lists available at ScienceDirect
Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres
Conversion of hexose into 5-hydroxymethylfurfural in imidazolium ionic
liquids with and without a catalyst
Quan Cao a, Xingcui Guo a, Shengxi Yao a, Jing Guan a, Xiaoyan Wang a, Xindong Mu a, , Dongke Zhang a,b,
⇑
⇑
a Key Laboratory of Biofuels, Qingdao Institute of Bioenergy and Bioprocess Technology, Chinese Academy of Sciences, 189 Songling Road, Qingdao 266101, People’s Republic of China
b Centre for Energy (M473), The University of Western Australia, 35 Stirling Highway, Crawley, WA 6009, Australia
a r t i c l e i n f o
a b s t r a c t
Article history:
Conversion of fructose and glucose into 5-hydroxymethylfurfural (HMF) was investigated in various
imidazolium ionic liquids, including 1-butyl-3-methylimidazolium chloride (BmimCl), 1-hexyl-3-meth-
ylimidazolium chloride (HmimCl), 1-octyl-3-methylimidazolium chloride (OmimCl), 1-benzyl-3-
methylimidazolium chloride (BemimCl), 1-Butyl-2,3-dimethylimidazolium chloride (BdmimCl), and
1-butyl-3-methylimidazolium p-toluenesulfonate (BmimPS). The acidic C-2 hydrogen of imidazolium
cations was shown to play a major role in the dehydration of fructose in the absence of a catalyst, such
as sulfuric acid or CrCl3. Both the alkyl groups of imidazolium cations and the type of anions affected the
reactivity of the carbohydrates. Although, except BmimCl and BemimCl, other four ionic liquids could
only achieve not more than 25% HMF yields without an additional catalyst, 60–80% HMF yields were
achieved in HmimCl, BdmimCl, and BmimPS in the presence of sulfuric acid or CrCl3 in sufficient
quantities.
Received 12 January 2011
Received in revised form 6 March 2011
Accepted 8 March 2011
Available online 12 March 2011
Keywords:
Fructose
Glucose
HMF
Imidazolium ionic liquids
Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction
1-butyl-3-methylimidazolium chloride (BmimCl).7 However, the
reason why the imidazolium chlorides themselves can catalyze
Biomass derived carbohydrate, as an abundant, relatively inex-
pensive and renewable chemical feedstock can be converted into
many useful chemicals.1–6 Among those chemicals, 5-hydroxym-
ethylfurfural (HMF), which can be easily produced by the dehydra-
tion of fructose as shown in Scheme 1,7 is a potential intermediate.
HMF can be transformed into many other useful products, such as
2,5-furandicarboxylic acid which may replace terephthalic acid as
a monomer in the production of plastics, and 2,5-dimethylfuran8,9
which is a potential liquid transportation fuel. The synthesis of
HMF can be performed in water10–15 and other highly polarized or-
ganic solvents, such as dimethyl sulfoxide (DMSO),14,16,17 dimeth-
ylformamide (DMF)18, and dimethylacetamide (DMA)19 in the
presence of an acidic catalyst. Different catalysts, such as mineral
acids,19 strongly acidic ion-exchange resin Amberlyst-15,16 hetero-
polyacids16, and acidic ionic liquids17 have all been shown to be
effective catalysts in this reaction.
the reaction is still a puzzle. How the structure of the imidazolium
ionic liquid may influence the reaction requires further
investigation.
In the work present in this contribution, we report an investiga-
tion into the dehydration of fructose and glucose into HMF in
Ionic liquid, as a powerful solvent for carbohydrate conversion,
has already been employed in HMF production.7,20–27 It has
also been found that some non-acidic imidazolium ionic liquids
can act as both solvents and catalysts in fructose dehydration,
such as 1-ethyl-3-methylimidazolium chloride (EmimCl)23 and
⇑
Corresponding author. Tel.: +86 532 8066 2723; fax: +86 532 8066 2724 (X.M.);
tel.: +61 8 6488 7600; fax: +61 8 6488 7235 (D.Z.).
E-mail addresses: muxd@qibebt.ac.cn (X. Mu), Dongke.Zhang@uwa.edu.au
(D. Zhang).
Scheme 1. Possible reaction pathways for the dehydration of fructose and glucose.
0008-6215/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2011.03.015

Q. Cao et al. / Carbohydrate Research 346 (2011) 956–959
957
3. Results and discussion
3.1. Dehydration of fructose in imidazolium ionic liquids
without a catalyst
The dehydration of fructose in various ionic liquids was first
studied at 120 °C without an additional catalyst. Visual observation
indicated that the fructose was fully soluble in all of the ionic liq-
uids under the experimental conditions. It can be seen from Table 1
that the alkyl groups of the imidazolium cations can greatly affect
the reactivity of fructose. Except in BmimCl and BemimCl, no or al-
most no HMF was produced in the other four ionic liquids. Com-
pared to BdmimCl, BmimCl could achieve a HMF yield of 63.1%,
while no HMF product was detected in BdmimCl. As the structural
difference between BmimCl and BdmimCl is the hydrogen/methyl
group on the C-2 position of the imidazolium cation, it is obvious
that the C-2 hydrogen plays a major role in the reaction. The reason
can be ascribed to the acidity of the C-2 hydrogen, which is already
proven by the formation of carbene for imidazolium cations under
basic conditions.28 Hence, the fructose dehydration in previously
reported EmimCl23 and BmimCl7 without a catalytic additive was
catalyzed by an acid.
Scheme 2. Chemical structures of the ionic liquids used in this work.
Chu et al29 have investigated the acidity scale of 1,3-dialkylim-
idazolium salts and found that the acidity of Bemim+ was a little
stronger than that of Bmim+, Hmim+, and Omim+, while the acidity
of Bmim+, Hmim+, and Omim+ was of the same level. However, for
HmimCl and OmimCl, only low HMF yields of 7.3% and 0% were ob-
tained, respectively, which could not be explained by just the acid-
ity of the C-2 hydrogen. A similar trend has also been observed that
increasing the chain length of alkyl group from ethyl to octyl led to
a decrease in the HMF yield.23 BemimCl, with its bulky benzyl
group, achieved only 17.7% HMF yield. The substantial differences
in the HMF yields synthesized in the ionic liquids among BmimCl,
BemimCl, HmimCl and OmimCl are thought to be due to the steric
effect of the alkyl groups on the imidazolium ring, which may hin-
der the contact between the acidic hydrogen and the fructose mol-
ecule. As the lengths of the alkyl groups are longer in BmimCl,
HmimCl, and OmimCl, their catalytic activities are correspondingly
lower. For BemimCl, its steric effect is between those of BmimCl
and HmimCl and its HMF yield is also between them. Accordingly,
an imidazolium chloride with a stronger acidic C-2 hydrogen and
smaller alkyl groups may be more capable of catalyzing the fruc-
tose dehydration. This explanation warrants further studies in
the future.
Besides the imidazolium cations, the anions of the ionic liquids
can also affect the fructose dehydration as indicated by the exper-
imental results on BmimCl and BmimPS, whose structural differ-
ence lies only in their anions. The fact that p-toluenesulfonate
anion can make Bmim+ completely inert suggests that ClÀ is more
favorable for HMF production than the p-toluenesulfonate anion.
The result is in agreement with the mechanism proposed by Binder
and Raines,19 where the chloride acts as a nucleophile and pro-
motes the dehydration of fructose. The p-toluenesulfonate anion
various imidazolium ionic liquids, as shown in Scheme 2. The ef-
fects of both the ionic liquid structure and the amount of a catalyst
on the reaction are also discussed.
2. Experimental
2.1. Materials and experimental procedures
Fructose (P99%) was purchased from Solarbio Science
&
Technology Co., Ltd and used as a standard. Anhydrous glucose
(analytical grade) was obtained from Guangdong Guanghua
Chemical Reagent Co., Ltd. 1-Butyl-3-methylimidazolium chloride
(BmimCl) (99%), 1-hexyl-3-methylimidazolium chloride (HmimCl)
(99%), 1-octyl-3-methylimidazolium chloride (OmimCl) (99%),
1-benzyl-3-methylimidazolium chloride (BemimCl) (99%), 1-bu-
tyl-2,3-dimethylimidazolium chloride (BdmimCl) (99%), and
1-butyl-3-methylimidazolium p-toluenesulfonate (BmimPS) (99%)
were purchased from Shanghai Chengjie Chemical Co., Ltd. Concen-
trated sulfuric acid (98%) was supplied by Laiyang Chemical Co.,
Ltd. CrCl3Á6H2O (P99%) was obtained from Sinopharm Chemical
Reagent Co., Ltd. 5-Hydroxymethylfurfural (P99%) was obtained
from Wujiang Yingchuang Chemical Reagent Co., Ltd and used as
a standard.
In a typical experiment, 5 g of ionic liquid was weighed in a
glove box and heated in an oil bath at 100 °C. 0.5 g fructose was
added after the ionic liquid was completely molten. 0.027 g of con-
centrated sulfuric acid was added when the fructose was dissolved,
0.027 g of concentrated sulfuric acid was added and the vial was
then sealed. Zero time started after the addition of sulfuric acid.
2.2. Analytical methods
HMF was quantified using a Waters HPLC with an UV detector
at 284 nm and a SunFire™ C18 column (4.8 Â 150 mm) at 35 °C.
The eluent was a mixture of methanol and water with a methanol
to water volumetric ratio of 1:4. The flow rate was 0.6 mL/min. The
amount of HMF was calculated by using the external standard as
described in the experimental section above.
Table 1
Conversion of fructose in ionic liquida
Ionic liquid
HMF yieldb (%)
Selectivity (%)
Conversionb (%)
BmimCl
HmimCl
OmimCl
BemimCl
BdmimCl
BmimPS
63.1
7.3
0
17.7
0
67.6
11.1
0
30.8
0
93.4
65.7
42.5
57.5
44.3
46.7
The sugar concentration was determined using
performance ion-exchange chromatograph with a pulsed ampero-
metric detector (Dionex, ICS-3000) and CarboPac™ PA10
a high-
a
0
0
carbohydrate column (4 Â 250 mm) at 30 °C. The eluent was a
7.5 mM NaOH aqueous solution with a flow rate of 1.0 mL/min.
The amount of fructose was calculated using the external standard.
a
Reaction conditions: ionic liquid 5 g; fructose 0.5 g; T = 120 °C and t = 50 min.
The HMF yield and fructose conversion were determined by HPLC.
b

958
Q. Cao et al. / Carbohydrate Research 346 (2011) 956–959
Table 2
Table 4
Conversion of fructose in ionic liquida
Conversion of fructose with CrCl3Á6H2O in ionic liquida
Ionic liquid
HMF yieldb (%)
Selectivity (%)
Conversionb (%)
Ionic
liquid
Catalyst loading
(mol %)
HMF yieldb
(%)
Selectivity
(%)
Conversionb
(%)
BmimCl
HmimCl
OmimCl
BemimCl
BdmimCl
BmimPS
60.4
21.9
8.6
53.2
0
60.4
25.7
10.8
53.2
0
100
85.1
79.6
100
67.5
49.9
BmimCl
HmimCl
10
10
25
10
25
75.3
43.7
62.0
0.39
14.2
71.0
48.7
60.2
66.2
75.3
47.3
62.0
0.45
15.5
71.0
54.1
60.2
66.2
100
92.4
100
87.0
91.6
100
90.0
100
100
OmimCl
0
0
BemimCl 10
BdmimCl 10
25
a
Reaction conditions: ionic liquid 5 g; fructose 0.5 g; T = 140 °C and t = 50 min.
The HMF yield and fructose conversion were determined by HPLC.
b
BmimPS
10
a
Reaction conditions: ionic liquid 5 g; fructose 0.5 g; T = 100 °C and t = 120 min.
The HMF yield and fructose conversion were determined by HPLC.
is larger than the chloride anion in size, which makes it more
difficult for the acidic hydrogen in BmimPS to contact the fructose
molecules than those in BmimCl.
b
Besides, it should be mentioned that at 120 °C, though HMF
was not produced in some of the ionic liquids, fructose could
be converted into other by-products such as soluble polymers
as described in Scheme 1 since the stability of fructose is sensitive
not only to the catalytic additives26,30 but also to the high
temperature.23,31
Compared to the reaction data at 120 °C, at 140 °C, the HMF
yields increased for most of the ionic liquids tested, as shown in
Table 2. One possible reason for this is that the higher temperature
may impair the strength of the C–H bond, leading to easier ioniza-
tion of the C-2 hydrogen and enhancing the acidity of the hydro-
gen.32 For BmimCl, the slight decrease in the HMF yield may be
due to the accelerated side reactions at higher temperatures.23,24
The inertness of BdmimCl can also be ascribed to its absence of
acidity, while for BmimPS, the side effect of the p-toluenesulfonate
anion on the reaction may be more dominant than the increase of
temperature.
literature.7 Therefore, the fact that the added sulfuric acid had no
obvious catalytic activity in HmimCl and OmimCl was not unex-
pected considering the steric effect of their alkyl groups, that is,
the bulky hexyl and octyl groups of HmimCl and OmimCl might
prevent the added H+ from contacting the fructose molecules effec-
tively. The steric effect may also explain why the use of BemimCl,
BdmimCl, and BmimPS gave lower HMF yields than BmimCl. When
the amount of sulfuric acid was increased to 25 mol %, as high as
81.9% HMF yield could be achieved in HmimCl; while OmimCl
was not absolutely inert. The results imply that the higher concen-
tration of H+ can lessen the side effects of the ionic liquids on the
reaction by facilitating the contact of H+ ions with the fructose
molecules.
The dehydration of fructose into HMF has been reported in the
literature at low temperatures.24,26 In this work, the catalytic activ-
ity of H2SO4 in BmimCl was further investigated at 40 °C. The
experimental procedure was identical to that reported in the liter-
ature.26 When the weight ratio of fructose to BmimCl was 1 to 10
with 24 mol % H2SO4 and the reaction proceeded for 50 min, the
HMF yield could reach as high as 83.3% at 100% fructose conver-
sion. Compared to the results in Table 3, the HMF yield at 40 °C
is close to that at 100 °C. A substantial amount of insoluble solids
was generated at 100 °C. The formation of insoluble solids, so-
called humins, at 120 °C has already been reported.7 While at
40 °C, only a very small amount of insoluble solids were observed.
Studies of the synthesis of HMF in EmimCl using metal chlo-
rides as catalysts demonstrated that Cr3+ possesses excellent cata-
lytic ability in fructose dehydration.23 In this work, the catalytic
activity of Cr3+ was further investigated in six imidazolium ionic
liquids. As seen from Table 4, the steric effect of the cations and an-
3.2. Effect of catalyst on fructose dehydration
To further evaluate how the ionic liquids affect the fructose
dehydration as solvents, the reactions were carried out by adding
sulfuric acid as a catalyst. A control experiment using an ionic li-
quid and fructose was conducted at 100 °C for 50 min. No HMF
was produced in all of the ionic liquids tested, which suggests that
the above ionic liquids acted only as solvents and not as catalysts
at 100 °C.
With the addition of 10 mol % sulfuric acid, it can be seen from
Table 3 that both imidazolium cations and anions of the ionic
liquids can affect the catalytic activity of sulfuric acid. BmimCl,
BemimCl, BdmimCl, and BmimPS achieved comparably higher
HMF yields of 82.9%, 76.7%, 77.6%, and 62.5%, respectively. The
reactivity of the previously reported inert BmimPS and BdmimCl
was dramatically enhanced. The excellent catalytic activity of
sulfuric acid in BmimCl has already been reported in the
ions of the ionic liquids can also affect the catalytic activity of Cr3+
.
BmimCl, BemimCl and BmimPS could achieve over 65% HMF yields.
More than 40% yield of HMF could be obtained in HmimCl and
BdmimCl. It was again OmimCl that was inert in the presence of
Cr3+. As described above, the higher H+ content can lead to an in-
crease in HMF yield for HmimCl and OmimCl. Analogously, the
amount of Cr3+ was further raised to 25 mol % and the HMF yields
increased by over ten percent for HmimCl, OmimCl, and BdmimCl,
respectively.
Table 3
Conversion of fructose with H2SO4 in ionic liquida
Ionic
liquid
Catalyst loading
(mol %)
HMF yieldb
(%)
Selectivity
(%)
Conversionb
(%)
BmimCl
HmimCl
10
10
25
10
25
82.9
7.3
81.9
0
82.9
13.4
81.9
0
100
54.6
100
55.1
85.8
100
100
92.0
3.3. Effect of catalyst on glucose dehydration
Glucose as the isomer of fructose is also a wonderful candidate
in HMF production due to its abundance and relatively low
price.15,23,33,34 Therefore, we also examined the dehydration of glu-
cose in the six ionic liquids with CrCl3Á6H2O as catalyst, which was
reported to have moderate catalytic ability in glucose dehydration
in EmimCl.23 To reduce the side effects of the solvents, 120 °C and
25 mol % catalyst loading was employed in the experiments. It can
OmimCl
3.4
4.0
BemimCl 10
BdmimCl 10
BmimPS
76.7
77.6
62.5
76.7
77.6
67.9
10
a
Reaction conditions: ionic liquid 5 g; fructose 0.5 g; T = 100 °C and t = 50 min.
The HMF yield and fructose conversion were determined by HPLC.
b

Q. Cao et al. / Carbohydrate Research 346 (2011) 956–959
959
the Key Science and Technology Program of Shandong Province
(No. 2008GG20002038), the Natural Science Foundation of
Shandong Province (No. ZR2010BQ014) and China’s World One
Thousand Chinese Talents Award.
References
1. Gandini, A.; Belgacem, M. N. Prog. Polym. Sci. 1997, 22, 1203–1379.
2. Bicker, M.; Hirth, J.; Vogel, H. Green Chem. 2003, 5, 280–284.
3. Moreau, C.; Belgacem, M. N.; Gandini, A. Top. Catal. 2004, 27, 11–30.
4. Kunkes, E. L.; Simonetti, D. A.; West, R. M.; Serrano-Ruiz, J. C.; Gartner, C. A.;
Dumesic, J. A. Science 2008, 322, 417–421.
5. Simonetti, D. A.; Dumesic, J. A. Chemsuschem 2008, 1, 725–733.
6. West, R. M.; Liu, Z. Y.; Peter, M.; Dumesic, J. A. Chemsuschem 2008, 1, 417–424.
7. Sievers, C.; Musin, I.; Marzialetti, T. ChemSusChem 2009, 2, 665–671.
8. Jarosław, L. ARKIVOC 2001, 2, 17–54.
9. Lichtenthaler, F. W. Acc. Chem. Res. 2002, 35, 728–737.
10. Hansen, T. S.; Woodley, J. M.; Riisager, A. Carbohydr. Res. 2009, 344, 2568–2572.
11. Asghari, F. S.; Yoshida, H. Ind. Eng. Chem. Res. 2006, 45, 2163–2173.
12. Asghari, F. S.; Yoshida, H. Carbohydr. Res. 2006, 341, 2379–2387.
13. Asghari, F. S.; Yoshida, H. Ind. Eng. Chem. Res. 2007, 46, 7703–7710.
14. Qi, X.; Watanabe, M. Catal. Commun. 2009, 10, 1771–1775.
15. Qi, X.; Watanabe, M. Catal. Commun. 2008, 9, 2244–2249.
16. Shimizu, K.-I.; Uozumi, R.; Satsuma, A. Catal. Commun. 2009, 10, 1849–1853.
17. Tong, X.; Li, Y. ChemSusChem 2010, 3, 350–355.
Figure 1. Conversion of glucose in ionic liquids. Conditions: ionic liquid 5 g;
glucose 0.5 g; CrCl3Á6H2O loading 25 mol % based on sugar; 120 °C and 60 min
reaction time. The number of x-coordinate represents: (1) BmimCl; (2) HmimCl; (3)
OmimCl; (4) BdmimCl; (5) BemimCl; (6) BmimPS.
18. Tong, X.; Ma, Y.; Li, Y. Carbohydr. Res. 2010, 345, 1698–1701.
19. Binder, J. B.; Raines, R. T. J. Am. Chem. Soc. 2009, 131, 1979–1985.
20. Li, C.; Zhao, Z. K.; Wang, A.; Zheng, M.; Zhang, T. Carbohydr. Res. 2010, 345,
1846–1850.
21. Clément, L.-M.; Moreau, C. Catal. Commun. 2003, 4, 517–520.
22. Moreau, C.; Finiels, A.; Vanoye, L. J. Mol. Catal. A: Chem. 2006, 253, 165–169.
23. Zhao, H.; Holladay, J. E.; Brown, H.; Zhang, Z. C. Science 2007, 316, 1597–1600.
24. Chan, J. Y. G.; Zhang, Y. ChemSusChem 2009, 2, 731–734.
25. Qi, X.; Watanabe, M.; Aida, T. M. Green Chem. 2009, 11, 1327–1331.
26. Qi, X.; Watanabe, M.; Aida, T. M. ChemSusChem 2009, 2, 944–946.
27. Bao, Q.; Qiao, K.; Tomida, D.; Yokoyama, C. Catal. Commun. 2008, 9, 1383–1388.
28. Jafarpour, L.; Stevens, E. D. J. Organomet. Chem. 2000, 606, 49–54.
29. Chu, Y.; Deng, H. J. Org. Chem. 2007, 72, 7790–7793.
be seen from Figure 1 that approximately 65% HMF yield could be
achieved in BmimCl, BemimCl, and BmimPS, suggesting that com-
parable reactivity of BemimCl and BmimPS with BmimCl, HmimCl,
and BdmimCl could give more than 55% HMF yields. However, less
than 20% HMF yield was achieved in OmimCl. The ionic liquids
influence the synthesis of HMF from glucose in a manner that is
similar to that from fructose.
30. Hu, S.; Zhang, Z.; Zhou, Y.; Han, B. Green Chem. 2008, 10, 1280–1283.
31. Orsi, F. J. Therm. Anal. 1973, 5, 329–335.
Acknowledgments
32. Reijenga, J. C.; Gagliardi, L. G. J. Chromatogr., A 2007, 1155, 142–145.
33. Yan, H.; Yang, Y.; Tong, D.; Xiang, X.; Hu, C. Catal. Commun. 2009, 10, 1558–
1563.
Thisworkis supportedbytheNationalHighTechnologyResearch
and Development Program of China (No. 2009AA05Z410), the
Natural Science Foundation of China (No. 20803038, 21003146),
34. Chidambaram, M.; Bell, A. T. Green Chem. 2010, 12, 1253–1262.