A novel route towards high yield 5-hydroxymethylfurfural from fructose catalyzed by a mixture of Lewis and Br?nsted acids

Article abstract of DOI:10.1039/c4ra04906c

We have developed an effective route for obtaining 5-hydroxymethylfurfural with a yield of 92.6 mol% from the dehydration of fructose in N,N-dimethylformamide using a mixture of AlCl₃, H₂SO₄and H₃PO₄as catalyst. The NMR analysis showed the intermediate formed among fructose, AlCl₃and H₃PO₄plays an important role in the novel result. This journal is

Full text of DOI:10.1039/c4ra04906c

RSC Advances
View Article Online
View Journal | View Issue
COMMUNICATION
A novel route towards high yield
5-hydroxymethylfurfural from fructose catalyzed
Cite this: RSC Adv., 2014, 4, 42035
by a mixture of Lewis and Bronsted acids†
¨
Received 24th May 2014
Accepted 28th August 2014
a
a
b
b
b
Yingxin Liu, Zhenbin Li, Yao Yang, Yaxin Hou and Zuojun Wei*
DOI: 10.1039/c4ra04906c
www.rsc.org/advances
We have developed an effective route for obtaining 5-hydrox- the decomposition of 5-HMF to levulinic acid as occurred in
5
,7–9
ymethylfurfural with a yield of 92.6 mol% from the dehydration of the presence of water and other protic solvents.
fructose in N,N-dimethylformamide using a mixture of AlCl , H SO application of imidazolium-based ionic liquids and other
and H PO as catalyst. The NMR analysis showed the intermediate deep eutectic mixtures as dual acid catalyst and solvent may
The
3
2
4
3
4
formed among fructose, AlCl
3 3 4
and H PO plays an important role in be the most effective strategy for the dehydration of fructose
4
,10–12
the novel result.
as well as other carbohydrates.
However, their high
boiling points and excellent miscibility with 5-HMF bring a
huge challenge to the post-separation of 5-HMF from the
reaction bulk, which seems less feasible for industrialization
even though much progress has been made in the separation
In recent years, the hydrolysis and dehydration of carbohydrate-
based biomass to 5-hydroxymethylfurfural (5-HMF) is viewed as
a promising way to utilize renewable resources and therefore
deserves much attention. Among the tested carbohydrates,
fructose is conrmed to be the most direct precursor towards
the formation of 5-HMF. To achieve a high yield of 5-HMF from
fructose, the selection of a suitable acid catalyst and solvent is
primarily considered.
4,11,13
of 5-HMF from ionic liquids.
In the present article, we focus on developing a process for
the industrial manufacture of 5-HMF from fructose at lower
cost. To reach this goal, we limited the catalyst and the solvent
within the conventional chemicals. The preliminary study was
1
3
started by conducting the dehydration of fructose using AlCl ,
Almost all types of the acids, including inorganic/organic
liquid Br ¨o nsted acids, metal salts and their oxides, zeolites,
heteropoly acids and their salts, cation exchange resins, as
well as the newly emerged ionic liquids have been applied
and showed different reactivity in the dehydration of fruc-
one of Lewis acids, as the catalyst in various aprotic polar
solvents (Table 1). As expected, ionic liquid 1-methyl-3-butyl
imidazolium chloride ([BMIM]Cl) exhibited the best perfor-
mance, with the highest 5-HMF yield of 58.5 mol% within 30
min under the reaction conditions used in this work.
2
–6
tose. In addition to the catalyst, the selection of an ideal
solvent is of the same importance. At least three aspects
should be considered: (1) being benecial to the dissolution
of fructose; (2) being benecial to obtaining a high 5-HMF
yield; (3) being benecial to the post-separation of 5-HMF. So
far, it is well known that some polar aprotic solvents like
dimethyl sulfoxide (DMSO), N,N-dimethylformamide (DMF),
N,N-dimethylacetamide (DMA), and 1,4-dioxane are the most
favourite solvents for the dehydration of fructose because
they have considerable solubility to fructose and can inhibit
1
,4-Dioxane was found to be poorly active, with a 5-HMF yield
of only 30.7 mol% even though the reaction time was extended
to 60 min. While conducted in DMSO and DMF solvents, the
yield of 5-HMF were similar, with the value of 48.3 mol% and
ꢀ
4
4.3 mol% at 120 C (entries 4 and 7 in Table 1), respectively.
As a typical aprotic polar solvent, DMSO was frequently
applied as a reaction medium for the dehydration of fructose
due to its high solubility to fructose (38.2 w% at 25 C as shown
in Table 1) and effectiveness in obtaining a high yield of
ꢀ
6,9,15
5-HMF.
Compared with DMSO, DMF has a boiling point
ꢀ
lower by 36.2 C, which means much energy may be saved if
vacuum distillation is used for the subsequent concentration of
5
a
College of Pharmaceutical Science, Zhejiang University of Technology, Hangzhou
-HMF when using DMF as the solvent. Moreover, it is generally
3
10014, P.R. China
b
known that even a trace amount of DMSO in the products will
Key Laboratory of Biomass Chemical Engineering of Ministry of Education,
Department of Chemical and Biological Engineering, Zhejiang University, Hangzhou introduce unpleasant smell which is hard to remove. Based on
3
10027, P.R. China. E-mail: weizuojun@zju.edu.cn; Tel: +86 13588810769
the above two reasons, we nally chose DMF as the dehydration
medium for fructose instead of DMSO, although a slightly lower
1
13
†
Electronic supplementary information (ESI) available: H and C-NMR, HPLC,
elemental analysis data, and other materials. See DOI: 10.1039/c4ra04906c
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 42035–42038 | 42035

View Article Online
RSC Advances
Communication
c
Table 1 Dehydration of fructose in various aprotic polar solvents
Solvent
Volume
(ml)
Boiling point
( C)
Solubility to
Fructose (w%)
Fructose
(g)
Temperature
( C)
Time
(min)
Fructose conversion
(mol%)
5-HMF yield
(mol%)
ꢀ
a
ꢀ
Entry
Name
b
1
2
3
4
5
6
7
8
9
[BMIM]Cl
1,4-Dioxane
2 g
100
100
100
50
275
101.1
56
50 mg
120
101
101
120
120
120
120
100
120
120
30
30
60
30
30
30
30
30
30
30
>99
88
92
>99
>99
>99
>99
97
58.5
26.6
30.7
48.3
40.4
42.0
44.3
36.3
39.1
38.7
0.60
1
1
5
5
5
5
5
5
5
DMSO
DMF
189.0
152.8
38.2
15.5
80
100
100
150
200
>99
>99
1
0
a
ꢀ
b
ꢀ
c
3
Detected at 25 C. Adapted from ref. 14, data detected at 110 C. Note: AlCl was applied as the catalyst, with the amount equal to 26.9 mol% of
fructose.
1
7
yield of 5-HMF might be achieved. Further optimization showed
Inspired by Harmer's work in which a mixture of
ꢁ
1
that an initial concentration of fructose as 50 g l (5 g of phosphoric-sulfuric acid showed surprising performance in the
fructose dissolved in 100 ml of DMF, entry 7 in Table 1) was the transformation of biomass to sugars, we further investigated
most preferred.
3 2 4 3 4
the synergistic effect of AlCl , H SO and H PO on the dehy-
In the next step, we tested the catalytic reactivity of the acids dration of fructose to 5-HMF in DMF. As shown in Fig. 2, the
AlCl , HCl, H SO and H PO for the dehydration of fructose to yield of 5-HMF was even higher when fructose was catalyzed by
3
2
4
3
4
5
-HMF in DMF solvent, and found that either a Lewis acid or a combination of AlCl
Br ¨o nsted acid alone failed to obtain a 5-HMF yield higher than mixtures with a range of S/P ratio (S/P denotes the molar ratio of
0 mol% (Fig. 1). We therefore further tried the mixture of SO /H PO ) from 0.5 to 2.0 as well as L/B ratio (L/B denotes
Br ¨o nsted and Lewis acids as a catalyst system, which has been the molar ratio of AlCl
/(H SO + H PO )) from 0 to 0.33. Over
reported to have positive synergistic effect on many catalytic this range, a 5-HMF yield as high as 92.6 mol% was achieved at
3 2 4 3 4
, H SO and H PO . We investigated the
6
H
2
4
3
4
3
2
4
3
4
16,17
reactions.
As expected, a mixture of AlCl
3
with either H
2
SO
4
S/P ¼ 2/3 and L/B ¼ 0.15 (i.e., AlCl
3 2 4 3 4
: H SO : H PO was
or H PO was more effective for the dehydration of fructose to 1 : 2.7 : 4). The yield could be repeated well while the reaction
3
4
5
1
-HMF, with an increase in the yield of 5-HMF by 10 mol% to was scaled up to 5–25 times in 1–5 l stirred reactors. To our best
5 mol% compared with using AlCl
3
alone.
knowledge, this is a very high yield of 5-HMF from fructose by
using conventional solvents and catalysts.
Owing to the application of the relatively lower boiling DMF
as the solvent, we are able to remove the solvent and obtain pure
Fig. 1 Effects of the type and amount of the acids on the yield of
5
-HMF from fructose. Reaction conditions: DMF 100 ml, fructose 5 g, Fig. 2 Synergistic effect of AlCl
3
, sulfuric acid and phosphoric acid on
ꢀ
1
20 C, 20 min. O Hydrochloric acid; , Sulfuric acid; B phosphoric the yield of 5-HMF from fructose. Reaction conditions: DMF 100 ml,
ꢀ
acid; V Sulfuric acid + AlCl
7.49 mmol).
3
(7.49 mmol); > phosphoric acid + AlCl
3
fructose 5 g, 120 C, 20 min. S/P denotes the molar ratio of H
2
SO
4
/
(
3 4 3 2 4 3 4
H PO ; L/B denotes the molar ratio of AlCl /(H SO + H PO ).
42036 | RSC Adv., 2014, 4, 42035–42038
This journal is © The Royal Society of Chemistry 2014

View Article Online
Communication
RSC Advances
5-HMF more easily by simple distillation. Aer reaction, a given
amount of Na CO was added to neutralize the reaction solu-
2
3
tion, and then DMF was completely removed by reduced pres-
ꢀ
sure distillation at 120 C to achieve brown viscous slurry.
Further vacuum distilling the slurry at 10–100 Pa at raised
ꢀ
temperature of 150–180 C, a light yellow crystal of 5-HMF with
the purity higher than 95% was obtained. Nevertheless, the
recovery of 5-HMF only reached 20–30%, which is due to the
strong interaction between 5-HMF and the byproduct of humic
acid, sulfuric acid or phosphoric acid that prevents 5-HMF from
evaporation from the reaction solution. However, by treating
13
the slurry with our previous developed EIVRD process, in
which nitrogen gas was introduced to intensify the vacuum
evaporation of 5-HMF, a much higher recovery of 85–90% of 5-
HMF with the purity around 95% was obtained. For compar- Fig. 3 C-1 chemical shifts of fructose isomers in DMF-d7 in the
presence of different combinations of acids. (1) fructose; (2) fructose +
ison, the slurry was also extracted with various solvent combi-
H
+
3
PO
AlCl
4
; (3) fructose + H
2
SO
4
; (4) fructose + AlCl
3 3 4
; (5) fructose + H PO
nations for optimization and 1 : 1 molar ratio of ethyl acetate to
hexane achieved the best 5-HMF recovery of 98.7% with the
purity less than 70% aer evaporation of the solvents. Consid-
ering that it is very difficult for 5-HMF to crystallize from
3
and (6) fructose + H
3
PO +H SO + AlCl .
4
2
4
3
ꢀ
conventional solvents due to its low melting point (ca. 33.6 C)
and excellent miscibility with most organic solvents, obtaining

nal 5-HMF with high purity by solvent extraction method is
challenging.
3
Although we lack a clear picture of why a mixture of AlCl ,
H SO and H PO4 is a distinctively effective catalyst for the
2
4
3
dehydration of fructose to 5-HMF in DMF solvent, we are able to
offer some insights into the mechanism. As we know, D-fructose
exists as a mixture of ve tautomeric forms in an aprotic solvent,
i.e. (a) open-chain; (b) a-D-fructofuranose (F5-a); (c) b-D-fructo-
furanose (F5-b); (d) a-D-fructopyranose (F6-a); (e) b-D-
fructopyranose (F6-b). Recent work from NMR and FT-IR tech-
Scheme
complex to 5-HMF.
1 Speculated pathway from fructose–phosphate–AlCl
3
niques as well as molecular dynamics calculations has The same phenomenon was also found on the chemical shis
1
conrmed that only fructofuranoses (F5-a and F5-b) can lead to of hydrogen atoms in F5-b isomer by H-NMR analysis (Table
the formation of 5-HMF, while pyranose isomers may lead to the S4†). We therefore deduce that some stronger interactions
3
,6,8,18
formation of humins.
In our research, we rst investigated the distribution of cyclic According to the research on the glycose–phosphate–metal
3 4 3
should have formed among fructose, H PO and AlCl .
19
tautomers of fructose in DMF-d
7
solvent with or without interactions, we propose a possible intermediate during the
1
3
different combinations of acids by C-NMR (Fig. 3). The results dehydration of fructose, as presented in Scheme 1. Owing to the
showed that the composition of the four tautomers in pure formation of the fructose-phosphate-AlCl complex, the fructo-
DMF-d was about 10% (F5-a): 27% (F5-b): 61% (F6-b): 3% (F6-a
furanose can stably exist in the reaction system, which shis the
Table S2†). The addition of H PO or AlCl shied the equi- balance to the formation of fructofuranose from fructopyr-
3
7
)
(
3
4
3
librium from F6-b to F5-b and F5-a isomers, whereas sulfuric acid anose. The formed complex further dehydrates along the
tended to transfer F5-a to F5-b isomer but little transformation of pathway to 5-HMF in the presence of high proton concentration
F
2 4
6-b to F5-b isomer occurred, indicating that too strong Br ¨o nsted provided by H SO to obtain a high yield of 5-HMF.
acidity is not preferred to the formation of fructofuranose. The
It should be pointed out that the discussion on the mecha-
result is also consistent with the reports that Lewis acids nism is rather preliminary due to the complexity of the reaction
possess better catalytic ability than Br ¨o nsted acids in the reac- system and the limitation of the analysis techniques. To further
4,11
tion. Although H
they caused different effects on the transformation of isomers, phosphoric acid and AlCl
which suggests that not only protons from phosphoric acid advanced experimental in situ C, H, P, Al-NMR, and in situ
2
SO
4
and H
3
PO
4
are typical Br ¨o nsted acids, reveal the mechanism of the synergistic effect of sulfuric acid,
on the dehydration of fructose, more
3
1
3
1
31 27
participate in the isomerization and dehydration of fructose.
By further comparing the chemical shis of carbon atoms in
fructose, we found that the chemical shis of all the carbon
atoms from F5-b isomer moved to higher elds by 0.6–1.4 ppm
FTIR may be needed.
Conclusions
in the presence of both H
3
4 3
PO and AlCl , whereas single addi- We have developed a simple and effective method to convert
tion of H SO , H PO or AlCl
2
4
3
4
3
changed them far less (Table S3†). fructose to 5-HMF using DMF as the solvent and a mixture of
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 42035–42038 | 42037

View Article Online
RSC Advances
Communication
AlCl
3
, H
2
SO
4
and H
3
PO
4
(optimized mole ratio of 1 : 2.7 : 4) as
5 F. Wang, A. W. Shi, X. X. Qin, C. L. Liu and W. S. Dong,
Carbohydr. Res., 2011, 346, 982–985.
6 A. S. Amarasekara, L. D. Williams and C. C. Ebede,
Carbohydr. Res., 2008, 343, 3021–3024.
7 G. Sampath and S. Kannan, Catal. Commun., 2013, 37, 41–44;
J. N. Chheda, Y. Rom ´a n-Leshkov and J. A. Dumesic, Green
Chem., 2007, 9, 342–350; J. Wang, J. Ren, X. Liu, J. Xi,
Q. Xia, Y. Zu, G. Lu and Y. Wang, Green Chem., 2012, 14,
2506; K. Beckerle and J. Okuda, J. Mol. Catal. A: Chem.,
2012, 356, 158–164.
the catalyst, by which a 92.6 mol% yield of 5-HMF with a nearly
ꢀ
100% conversion of fructose could be achieved at 120 C in 20
min. By vacuum distillation, a nal light yellow crystal of 5-HMF
1
3
1
with purity around 95% was obtained. C-NMR and H-NMR
analysis showed that the existence of PO or AlCl
promotes the transformation of F6-b isomer of fructose to F5-b
and F5-a isomers, while H SO is favourable to the trans-
formation of F5-a to F5-b isomer. The novel catalytic reactivity
may be explained that the complex formed by AlCl , H PO and
H
3
4
3
2
4
3
3
4
fructose accelerates the dehydration of fructose along the
pathway to 5-HMF in the presence of high proton concentration
8 H. Kimura, K. Yoshida, Y. Uosaki and M. Nakahara, J. Phys.
Chem. A, 2013, 117, 10987–10996.
provided by H
2
SO
4
.
9 C. B. Rasrendra, J. N. M. Soetedjo, I. G. B. N. Makertihartha,
S. Adisasmito and H. J. Heeres, Top. Catal., 2012, 55, 543–
549.
Financial support from the National Natural Science Foun-
dation of China (nos 21106134 and 21276230), Zhejiang
Provincial Natural Science Foundation of China (no. 10 S. Dutta, S. De and B. Saha, Biomass Bioenergy, 2013, 55, 355–
LY14B060003) and Zhejiang Leading Team of S&T Innovation
no. 2011R50002) are greatly appreciated.
369.
(
11 R. J. van Putten, J. C. van der Waal, E. de Jong,
C. B. Rasrendra, H. J. Heeres and J. G. de Vries, Chem.
Rev., 2013, 113, 1499–1597.
Notes and references
1
2 S. Eminov, J. Wilton-Ely and J. P. Hallett, ACS Sustainable
Chem. Eng., 2014, 2, 978–981; W. Liu and J. Holladay,
Catal. Today, 2013, 200, 106–116; C. Shi, Y. Zhao, J. Xin,
J. Wang, X. Lu, X. Zhang and S. Zhang, Chem. Commun.,
2012, 48, 4103–4105; X.-L. Shi, M. Zhang, Y. Li and
W. Zhang, Green Chem., 2013, 15, 3438–3445;
S. P. Simeonov, J. A. Coelho and C. A. Afonso,
ChemSusChem, 2012, 5, 1388–1391.
1
2
Y. Li, Z. Wei, C. Chen and Y. Liu, Prog. Chem., 2010, 22, 1603–
609; J. Lewkowski, ARKIVOC, 2001, 1, 17–54; A. A. Rosatella,
S. P. Simeonov, R. F. Frade and C. A. Afonso, Green Chem.,
011, 13, 754–793; B. Saha and M. M. Abu-Omar, Green
Chem., 2014, 16, 24–38; S. P. Teong, G. Yi and Y. Zhang,
Green Chem., 2014, 16, 2015–2026.
1
2
B. Zheng, Z. Fang, J. Cheng and Y. Jiang, Chem. Sci., 2010, 65,
1
68–172; P. Carniti, A. Gervasini, S. Biella and A. Auroux, 13 Z. J. Wei, Y. X. Liu, D. Thushara and Q. L. Ren, Green Chem.,
Catal. Today, 2006, 118, 373–378; C. Fan, H. Guan,
2012, 14, 1220–1226.
H. Zhang, J. Wang, S. Wang and X. Wang, Biomass 14 V. M. Egorov, S. V. Smirnova, A. A. Formanovsky, I. V. Pletnev
Bioenergy, 2011, 35, 2659–2665; H. B. Xie, Z. H. Zhang and
and Y. A. Zolotov, Anal. Bioanal. Chem., 2007, 387, 2263–
Q. A. Wang, ChemSusChem, 2011, 4, 131–138; Y. Tian,
2269.
J. Deng, T. Pan, Q. Guo and Y. Fu, Chin. J. Catal., 2011, 32, 15 N. Rajabbeigi, R. Ranjan and M. Tsapatsis, Microporous
97–1002; B. Girisuta, B. Danon, R. Manurung,
Mesoporous Mater., 2012, 158, 253–256.
L. P. Janssen and H. J. Heeres, Bioresour. Technol., 2008, 16 L. Bui, H. Luo, W. R. Gunther and Y. Roman-Leshkov, Angew.
9
9
9, 8367–8375; X. Qi, M. Watanabe, T. M. Aida and
Chem., Int. Ed., 2013, 52, 8022–8025; Y. J. Pag ´a n-Torres,
T. Wang, J. M. R. Gallo, B. H. Shanks and J. A. Dumesic,
ACS Catal., 2012, 2, 930–934; M. Schwiderski, A. Kruse,
R. Grandl and D. Dockendorf, Green Chem., 2014, 16,
1569–1578.
J. R. L. Smith, Green Chem., 2008, 10, 799–805; S. Hu,
Z. Zhang, Y. Zhou, B. Han, H. Fan, W. Li, J. Song and
Y. Xie, Green Chem., 2008, 10, 1280–1283; J. B. Binder and
R. T. Raines, J. Am. Chem. Soc., 2009, 131, 1979–1985;
Y. Roman-Leshkov, J. N. Chheda and J. A. Dumesic, 17 M. A. Harmer, A. Fan, A. Liauw and R. K. Kumar, Chem.
Science., 2006, 312, 1933–1937; Y. Takeuchi, F. Jin, K. Tohji
Commun., 2009, 6610–6612.
and H. Enomoto, J. Mater. Sci., 2007, 43, 2472–2475; 18 G. R. Akien, L. Qi and I. T. Horvath, Chem. Commun., 2012,
T. Armaroli, G. Busca, C. Carlini, A. M. R. Galletti,
48,
5850–5852;
T.
Barclay,
M.
Ginic-Markovic,
M. Giuttari and G. Sbrana, J. Mol. Catal. A: Chem., 2000,
M. R. Johnston, P. D. Cooper and N. Petrovsky, Carbohydr.
Res., 2012, 352, 117–125; Y. N. Li, J. Q. Wang, L. N. He,
Z. Z. Yang, A. H. Liu, B. Yu and C. R. Luan, Green Chem.,
2012, 14, 2752–2758.
151, 233–243; X. Qi, M. Watanabe, T. M. Aida and
R. L. Smith Jr, Catal. Commun., 2008, 9, 2244–2249;
Y. Takeuchi, F. Jin, K. Tohji and H. Enomoto, J. Mater. Sci.,
2
008, 43, 2472–2475.
H. Kimura, M. Nakahara and N. Matubayasi, J. Phys. Chem. A,
011, 115, 14013–14021; N. Nikbin, S. Caratzoulas and
19 K. Gilg, T. Mayer, N. Ghaschghaie and P. Klufers, Dalton
Trans., 2009, 7934–7945; P. Grimminger and P. Klufers,
Dalton Trans., 2010, 39, 715–719; M. Steinborn, M. Suhanji
and P. Klufers, Dalton Trans., 2013, 42, 5749–5754.
3
4
2
D. G. Vlachos, ChemCatChem, 2012, 4, 504–511.
M. E. Zakrzewska, E. Bogel-Lukasik and R. Bogel-Lukasik,
Chem. Rev., 2011, 111, 397–417.
42038 | RSC Adv., 2014, 4, 42035–42038
This journal is © The Royal Society of Chemistry 2014