Efficient and environmental-friendly dehydration of fructose to 5-hydroxymethyl-2-furfural in water under high pressure of CO2

Article abstract of DOI:10.1016/j.tetlet.2016.09.036

To develop reaction systems of chemical conversion of biomass, fructose was heated in an aqueous medium under pressurized CO₂, which caused in situ generation of carbonic acid. It gave 5-hydroxymethyl-2-furfural (HMF) as a dehydration product in good yields. The maximum yield of HMF was 92% in the reaction at 90?°C under 7.0?MPa of CO₂for 168?h. The reaction proceeded in the absence of any other reagents such as non-volatile acids, organic solvents, inorganic, and organic salts. It means green, environmental-friendly, and efficient process was achieved for production of HMF.

Full text of DOI:10.1016/j.tetlet.2016.09.036

Tetrahedron Letters 57 (2016) 4742–4745
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Efficient and environmental-friendly dehydration of fructose to
5-hydroxymethyl-2-furfural in water under high pressure of CO2
b
a
c
Suguru Motokucho a, , Hiroshi Morikawa , Hisayuki Nakatani , Bart A. J. Noordover
⇑
a Chemistry and Material Engineering Program, Nagasaki University, 1-14, Bunkyo-Machi, Nagasaki-shi 852-8521, Japan
b Department of Applied Chemistry, Kanagawa Institute of Technology, 1030, Shimo-ogino, Atsugi, Kanagawa 243-0292, Japan
c Department of Chemical Engineering and Chemistry, Eindhoven University of Technology, P.O. Box 513, Eindhoven 5600MB, The Netherlands
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 28 July 2016
Revised 5 September 2016
Accepted 8 September 2016
Available online 13 September 2016
To develop reaction systems of chemical conversion of biomass, fructose was heated in an aqueous med-
ium under pressurized CO2, which caused in situ generation of carbonic acid. It gave 5-hydroxymethyl-2-
furfural (HMF) as a dehydration product in good yields. The maximum yield of HMF was 92% in the reac-
tion at 90 °C under 7.0 MPa of CO2 for 168 h. The reaction proceeded in the absence of any other reagents
such as non-volatile acids, organic solvents, inorganic, and organic salts. It means green, environmental-
friendly, and efficient process was achieved for production of HMF.
Keywords:
Biomass
Ó 2016 Elsevier Ltd. All rights reserved.
Dehydration reaction
Carbonic acid
Fructose
CO2
Introduction
with ethylene glycol to produce a furan-based polyester, which
is potentially usable as an alternative to poly(ethylene
To realize sustainable society, utilization of natural sources
instead of fossil chemicals and petroleum has been required.
Therefore, efficient conversion of biomass to available chemicals
such as polymerizable monomers for functional polymers has
recently attracted considerable attention.1–3 Among them,
monosaccharide is one of the promising renewable resources,
and its base units are composed of 5 or 6 carbon atoms. Especially,
fructose is abundant in nature and an easily available monosaccha-
ride obtained from honey, fruit, and beet. Thus, fructose is an
inherently versatile resource for functional chemicals. A typical
chemical conversion of fructose is production of 5-hydrox-
ymethyl-2-furfural (HMF).4 HMF can be also obtained from glucose
through the isomerization to fructose.5
Recently, efficient synthetic methods of HMF production have
been widely studied6–8 because HMF is an important platform as
not only a precursor of biofuel dimethylfuran9 but also fine chem-
icals and bioactive ingredients.10–13 For the valuable chemicals,
further reactions such as oxidation or reduction of HMF afford
furan-derived functional monomers, i.e., 2,5-furandicarboxylic acid
(FDCA), 2,5-dihydroxylmethylfuran, 2,5-diformylfuran, levulinic
acid, and others.6–8,14–16 In particular, FDCA could be polymerized
terephthalate).17,18
The reaction from fructose to HMF proceeds via successive
dehydration steps in the presence of acid catalysts (Scheme 1).
So far, there have been various kinds of mineral acids, organic
acids, ammonium salts, ionic liquids, and inorganic salts to pro-
mote the synthesis of HMF.6–8
Instead of Brǿnsted acid used in earlier studies,19–22 heteroge-
neous catalysts such as ion-exchange resins have been used.23–26
However, these reactions are usually carried out in polar organic
solvents such as dimethylsulfoxide (DMSO) and N,N-dimethylfor-
mamide (DMF) exhibiting high boiling points. Separation of HMF
from the reaction mixture sometimes requires a troublesome pro-
cess. In many of these works, yields of HMF were measured by
HPLC or GC analyses, and the yielded HMF was not isolated.6,7
Recently, ionic liquids and metal salts have been used for HMF
production.6,7 For the representative report, a chromium catalyst in
an ionic liquid has been also reported.27 However, Cr ion is toxic.
Large amounts of expensive ionic liquids are sometimes used.28
Moreover, they are non-volatile, resulting in complicated separa-
tion of HMF from the reaction mixture.
Various kinds of acids, organic and inorganic salts, ionic liquids,
as well as organic solvents in biphased systems were used, and the
reaction systems by using these additives often involve difficulty in
the separation, high cost, toxicity, and requirement of large
amounts.1–3,6–8 To date, HMF has been synthesized in relatively
⇑
Corresponding author. Tel./fax: +81 95 819 2653.
E-mail address: motoku@nagasaki-u.ac.jp (S. Motokucho).
http://dx.doi.org/10.1016/j.tetlet.2016.09.036
0040-4039/Ó 2016 Elsevier Ltd. All rights reserved.

S. Motokucho et al. / Tetrahedron Letters 57 (2016) 4742–4745
4743
of fructose in the aqueous solution gave HMF with consumption
of fructose even at 90 °C, which is a relatively mild condition.
The reaction temperature is lower than that reported by Wu et al.35
The product HMF in water could be obtained by only depressuriza-
tion of CO2 without filtration and neutralization. Therefore, this
process is very efficient. On the other hand, without CO2 at 90 °C
for 24 h, fructose did not give HMF. This indicates that fructose
was converted to HMF by pressurizing the reaction vessel with
CO2. Moreover, these results with and without CO2 differ from
the study with inulin under CO2-water system.35
OH
OH
O
acid
O
CHO
HO
HO
-3H2O
HO
OH
5-hydroxymethyl-2-furfural
(HMF)
fructose
Scheme 1. Dehydration reaction of fructose to HMF.
good yields in many previous studies. However, it is quite impor-
tant and required that HMF is prepared efficiently and environ-
mental-friendly free from organic solvents and additives in order
to obtain and separate a large scale of HMF easily from the reaction
mixture.6,7,29
It is well-known that CO2, especially under high pressure, effec-
tively gives carbonic acid by reaction with water.30,31 This carbonic
acid is expected to act as an acid catalyst. Therefore, saccharides in
aqueous solution under high pressure CO2 could be converted to
dehydration products, HMF. In this work, dehydration of fructose
to give HMF was investigated in an aqueous solution under super-
or sub-critical CO2 conditions (below and above 7.3 MPa). It is
noteworthy that our system is composed of three components, that
is, only CO2 gas, water, and fructose as starting material.
For industrial and large-scale production of HMF, this system is
quite beneficial in many aspects:
Figure 1 shows the reaction time dependence of the isolated
yields of HMF at 90 and 120 °C under 7.0 MPa CO2. At 90 °C, the
yield increased with increasing reaction time. After 24 and 48 h,
it reached 29% and 50%, respectively. Surprisingly, after a reaction
time of 168 h, the yield gave up to 92%. Although there remains a
requirement of a longer time of 168 h, this yield is of quite a higher
value in many reports.6–8 It suggests that the dehydration of fruc-
tose smoothly proceeded, and is hardly accompanied by side reac-
tions such as formation of levulinic acid, fumin, and polymers.35 On
the other hand, at 120 °C, yields of HMF increased with increasing
reaction time until 36 h. At that time, the yield reached 79%. How-
ever, when the reaction time was elongated to 50, 72 h, the yields
were decreased. Under these conditions, fumin precipitated from
the reaction mixture. It is reported that HMF was polymerized to
produce fumin.37 Indeed, after the reaction for 72 h, insoluble black
precipitates were observed in the reaction mixture. Judging from
these results at two reaction temperatures, it was concluded that
HMF could be synthesized most efficiently at 90 °C. In the previous
study, longer reaction times resulted in decrease of the yields.35
This behavior is similar to our results at 120 °C.
Figure 2 shows typical HPLC profiles of soluble component of
the reaction mixture (Reaction conditions: 90 °C, 7.0 MPa of CO2).
Two peaks were observed at 7.8 and 14.3 min, respectively. These
peaks were assigned to HMF and fructose, respectively, with the
help of reference samples. It is obvious that, when reaction time
passed from 48 h to 168 h, the peak for HMF was increased and
accompanied by a decrease of the peak for fructose. No other peaks
for by-product such as levulinic acid were observed in the HPLC
profiles. These results mean that the dehydration of fructose pro-
ceeded with high selectivity, that is, efficient conversion to HMF
was achieved. They also agreed well with the reaction behavior
in Figure 1, in which it gave high yields of HMF.
(1) CO2 gas can be removed after depressurization.
(2) Water is removable below atmospheric pressure.
(3) Cheap, abundant, and readily available chemicals.
(4) Nontoxic and non-flammable.
(5) Without any organic solvents such as DMSO and DMF.
(6) Without any additives such as metal salts and ion-exchange
resins.
(7) Without any acids such as HCl and H2SO4, meaning that neu-
tralization of the reaction mixture is not required after the
reaction.
The CO2–water system can enhance some organic reactions
such as reduction32 and hydrolysis.33,34 In a previous study, Wu
et al. reported that inulin was decomposed under pressured
CO2–water to produce HMF at relatively high temperature of
160–200 °C.35 However, the yield was up to 53% (reaction
condition: 180 °C for 2 h), and difference between under 0 MPa
CO2 and 4–9 MPa CO2 was very small, i.e., 5–15% in the yields. In
the literature, HMF was not isolated, and the yield was measured
by HPLC. Similarly, Liu et al. have reported that 1-hydroxypen-
tane-2,5-dione (HHD) from fructose through the formation of
HMF was synthesized with Pd/C catalyst in CO2-water.36 In the
literature, HMF in situ was produced in a 50% yield. However,
the study focused on the synthesis of HHD, resulting in that HMF
was not isolated and the details such as optimized reaction
conditions and relationships between the yield of HMF and
reaction temperature/CO2 pressure were not clear.
Therefore, the studies of (A) HMF isolation directly from
monosaccharide fructose and (B) the optimized conditions giving
a high yield of HMF production are required. In the previous stud-
ies,35,36 these two points (A) and (B) were not fully achieved. On
the other hand, it is novel that we here clarified both (A) and (B),
and obtained HMF from fructose under more optimized conditions
(especially, at 90 °C) giving much higher yields. Our reaction con-
ditions and starting material are different from the previous
report.35
100
80
60
40
20
0
0
50
100
150
200
Time / h
In this study, we carried out dehydration of fructose to HMF in
relatively high concentration of 21 wt% fructose in water (see the
section of ‘General procedure’). Under pressured CO2, the heating
Figure 1. Time dependence of yields of HMF at 90 (d) and 120 °C (s) under
7.0 MPa CO2.

4744
S. Motokucho et al. / Tetrahedron Letters 57 (2016) 4742–4745
in aqueous solution at pressures lower than 7.0 MPa without side
HMF
reactions. However, when performing the reaction at 120 °C, the
formation of fumin as by-product was observed. Although further
investigation is in progress to precisely settle the mechanism, by
using CO2-water system, we have provided a green, environmen-
tal-friendly, and easily separable method of HMF production from
fructose in a high yield of 92%.
Fructose
Reaction time
48 h
72 h
General procedure
168 h
In an autoclave of 200 mL, fructose 5.0 g (27.7 mmol) was dis-
solved in water (18.0 g, 1.0 mol) with stirring by using a magnetic
stirring bar. Then, predetermined weights of liquid CO2 was intro-
duced into the autoclave. The mixture in the autoclave was stirred
and heated at 90 °C under 7.0 MPa CO2. After the reaction for 168 h,
the autoclave was cooled, depressured, and the aqueous reaction
mixture was evaporated to give the crude products. By using the
aliquot, HPLC was measured. The crude product was purified by
short column chromatography on silica gel with CHCl3 eluent to
obtain HMF as a slightly yellowish liquid (92% yield). 1H NMR
(400 MHz, CDCl3) d 9.59 (s, 1H), 7.22 (d, J = 8.7 Hz, 1H), 6.52 (d,
J = 8.7 Hz, 1H), 4.72 (s, 2H). The same reactions were individually
carried out for predetermined times at 90 or 120 °C under
0–9 MPa CO2. After the reaction, in the same procedure aforemen-
tioned, conversion of fructose was measured, and HMF was
isolated to estimate the yield.
Figure 2. HPLC profiles of soluble components of reaction mixture using
a
Cosmosil-sugar-D column (Nakarai tesque) (Experimental conditions: eluent;
CH3CN/H2O = 3:1, RI detector; RI-2031(JASCO), column temperature; 30 °C, flow
rate; 0.5 mL/min).
Figure 3 shows CO2 pressure dependence of conversion of fruc-
tose and yields of HMF at 90 °C for 24 h. The conversion of fructose
was calculated based on the integral ratio of chemical shifts in 1H
NMR spectrum of the reaction mixture by the internal standard
(anisole) method. Both the conversion of fructose and the yield
of HMF showed maximum values at 7.0 MPa. These results indicate
that the applied CO2 pressure accelerates the dehydration of fruc-
tose to form HMF below 7.0 MPa, but suppressed the reaction over
this pressure. At CO2 pressures below 7.0 MPa, the formation of
carbonic acid possibly increased with increasing pressure to accel-
erate the dehydration. However, high pressures in excess of
7.0 MPa suppressed conversion of fructose. Since the phase change
of CO2 to a supercritical fluid occur around 7.3 MPa, the decrease of
the conversion under high CO2 pressures is probably due to the
reduced solubility of fructose in the solution, and/or reduction of
polarity and permittivity of water with increasing pressure of
hydrophobic CO2 to suppress dissociation of carbonic acid.31
In summary, we developed an efficient conversion of fructose
into HMF in aqueous solution under applied CO2 pressure. Dehy-
dration of fructose at 90 °C was accelerated by CO2 pressurization
Acknowledgments
This work was supported by the Ministry of Education, Culture,
Sports, Science, and Technology in Japan KAKENHI Grant-in-Aid for
Young Scientists (B) Grant Number 25810078 and Japan Society for
the Promotion of Science is funding agency of Japanese Ministry of
Science Grant-in-Aid for ‘‘Program for Advancing Strategic Interna-
tional Networks to Accelerate the Circulation of Talented
Researchers”. And the author is sincerely grateful to Dr. Toshio
Inoue, JX Nippon Oil & Energy Corporation.
References and notes
1. Delidovich, I.; Hausoul, P. J. J.; Deng, L.; Pfützenreuter, R.; Rose, M.; Palkovits, R.
Chem. Rev. 2016, 116, 1540–1599.
2. Isikgor, F. H.; Becer, C. R. Polym. Chem. 2015, 6, 4497–4559.
3. Yao, K.; Tang, C. Macromolecules 2013, 46, 1689–1712.
4. Mittal, N.; Nisola, G. M.; Cheng, W.-J. Tetrahedron Lett. 2012, 53, 3149–3155.
5. Jadhav, H.; Taarrning, E.; Pedersen, C. M.; Bols, M. Tetrahedron Lett. 2012, 53,
983–985.
6. Putten, R. J.; Waal, J. C.; Jong, E.; Rasrendra, C. B.; Heeres, H. J.; Vries, G. Chem.
Rev. 2013, 113, 1499–1597.
7. Rosatella, A. A.; Simeonov, S. P.; Frade, R. F. M.; Afonso, C. A. M. Green Chem.
2011, 13, 754–793.
8. Gandini, A.; Belgacem, M. N. In Monomers, Polymers and Composites from
Renewable Sources; Belgacem, M. N., Gandini, A., Eds.; Elsevier: Amsterdam,
2008; pp 115–152. Chapter 6.
9. Roman-Leshkov, Y.; Barrett, C. J.; Liu, Z. Y.; Dumesic, J. A. Nature 2007, 447,
982–985.
10. Lukevics, E.; Ignatovich, L.; Shestakova, I. Appl. Organomet. Chem. 2003, 17,
898–905.
11. Lewis, T. A.; Bayless, L.; Eckman, J. B.; Ellis, J. L.; Grewal, G.; Libertine, L.;
Nicolas, J. M.; Scannell, R. T.; Wels, B. F.; Wenberg, K.; Wypij, D. M. Bioorg. Med.
Chem. Lett. 2004, 14, 2265–2268.
12. Villain-Guillot, P.; Gualtieri, M.; Bastide, L.; Roquet, F.; Martinez, J.; Amblard,
M.; Pugniere, M.; Leonetti, J.-P. J. Med. Chem. 2007, 50, 4195–4204.
13. Schinzer, D.; Bourguet, E.; Ducki, S. Chem. Eur. J. 2004, 10, 3217–3224.
14. Lv, G.; Wang, H.; Yang, Y.; Deng, T.; Chen, C.; Zhu, Y.; Hou, X. Green Chem. 2016,
18, 2302–2307.
15. Yi, G.; Teng, S. P.; Zhang, Y. Green Chem. 2016, 18, 979–983.
16. Nakagawa, Y.; Tomishige, K. Catal. Commun. 2010, 154–156.
17. Burgess, S. K.; Leisen, J. E.; Kraftschik, B. E.; Mubarak, C. R.; Kriegel, R. M.; Koros,
W. J. Macromolecules 2014, 47, 1383–1391.
50
40
30
20
10
0
50
40
30
20
10
0
0
2
4
6
8
10
Pressure / MPa
18. Eerhart, A. J. J. E.; Faaij, A. P. C.; Patel, M. K. Energy Environ. Sci. 2012, 5, 6407–
6422.
19. Harris, D. W.; Feather, M. S. J. Org. Chem. 1974, 39, 724–725.
Figure 3. CO2 Pressure dependence of conversion of fructose (d) and yields of HMF
(s) at 90 °C for 24 h.

S. Motokucho et al. / Tetrahedron Letters 57 (2016) 4742–4745
30. Toews, K. L.; Shroll, R. M.; Wal, C. M. Anal. Chem. 1995, 67, 4040–4043.
4745
20. Kuster, B. F. M. Carbohydr. Res. 1977, 54, 177–183.
21. Moye, C. J.; Goldsack, R. J. J. Appl. Chem. 1966, 16, 206–208.
22. Liu, Y.; Li, Z.; Yang, Y.; Hou, Y.; Wei, Z. RSC Adv. 2014, 4, 42035–42038.
23. Lima, S.; Pillinger, M.; Valente, A. A. Catal. Commun. 2008, 9, 2144–2148.
24. Nakamura, Y.; Morikawa, S. Bull. Chem. Soc. Jpn. 1980, 53, 3705–3706.
25. Takagaki, A.; Ohara, M.; Nishimura, S.; Ebitani, K. Chem. Commun. 2009, 6276–
6278.
26. Shimizu, K.; Uozumi, R.; Satsuma, A. Catal. Commun. 2009, 10, 1849–1853.
27. Zhao, H.; Holladay, J. E.; Brown, H.; Zhang, Z. C. Science 2007, 316, 1597–1600.
28. Zhang, J.; Yu, X.; Zhou, F.; Zhing, Y.; Du, N.; Huang, X. ACS Sustain. Chem. Eng.
2015, 3, 3338–3345.
31. West, K. V.; Wheeler, C.; McCarney, J. P.; Griffith, K. N.; Bush, D.; Liotta, C. L.;
Eckert, C. A. J. Phys. Chem. A 2001, 105, 3947–3948.
32. Li, G.; Jiang, H.; Li, J. Green Chem. 2001, 3, 250–251.
33. Miyazawa, T.; Funazukuri, T. Biotechnol. Prog. 2005, 21, 1782–1785.
34. Motokucho, S.; Matsumoto, T.; Nakayama, Y.; Horiuchi, R.; Morikawa, H.;
Nakatani, H. Polym. Bull. in press. Online published doi:http://dx.doi.org/10.
1007/s00289-016-1733-0.
35. Wu, S.; Fan, H.; Xie, Y.; Cheng, Y.; Wang, Q.; Zhang, Z.; Han, B. Green Chem.
2010, 12, 1215–1219.
36. Liu, F.; barrault, J.; Vigier, K. O.; Jérôme, F. ChemSusChem 2014, 7, 2089–2095.
37. Chuntanapum, A.; Matsumura, Y. Ind. Eng. Chem. Res. 2009, 48, 9837–9846.
29. Simeonov, S. P.; Coelho, J. A.; Afonso, C. A. M. ChemSusChem 2012, 5, 1388–
1391.