Conversion of carbohydrates into 5-hydroxymethylfurfural in an advanced single-phase reaction system consisting of water and 1,2-dimethoxyethane

Article abstract of DOI:10.1039/c5ra18824e

5-Hydroxymethylfurfural (HMF) is a bio-based platform chemical that may be converted into various chemicals and fuels. In the present study, we developed an advanced low-boiling single-phase reaction system for producing HMF from glucose. It consists of water and 1,2-dimethoxyethane (DMOE) and uses AlCl₃ as catalyst. Our results show that introduction of DMOE can substantially enhance HMF production because of the polar aprotic solvent effect provided by DMOE. Under optimal conditions, a high HMF yield (58.56%) was obtained. GC-MS of the liquid-phase products revealed that HMF and furans comprised 80% and ～90% of the detected products. Formation of liquid-phase products, including furans, oxygenated aliphatics, cyclopenten-1-ones, and pyrans is discussed. Further study of the humins formed during glucose conversion showed the effective inhibition of humin formation by DMOE. The structure of humins was characterized by FTIR spectroscopy. Finally, HMF production from disaccharides (sucrose, maltose and cellobiose) and polysaccharide (cellulose) using the water-DMOE system resulted in good yields, demonstrating that our single-phase water-DMOE solvent system has good potential use in HMF production from glucose and complex carbohydrates.

Full text of DOI:10.1039/c5ra18824e

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Conversion of carbohydrates into 5-
hydroxymethylfurfural in an advanced single-phase
reaction system consisting of water and 1,2-
dimethoxyethane
Cite this: RSC Adv., 2015, 5, 84014
*
*
Shurong Wang, Haizhou Lin, Jingping Chen, Yuan Zhao, Bin Ru, Kunzan Qiu
and Jinsong Zhou
5-Hydroxymethylfurfural (HMF) is a bio-based platform chemical that may be converted into various
chemicals and fuels. In the present study, we developed an advanced low-boiling single-phase reaction
system for producing HMF from glucose. It consists of water and 1,2-dimethoxyethane (DMOE) and uses
AlCl3 as catalyst. Our results show that introduction of DMOE can substantially enhance HMF production
because of the polar aprotic solvent effect provided by DMOE. Under optimal conditions, a high HMF
yield (58.56%) was obtained. GC-MS of the liquid-phase products revealed that HMF and furans
comprised 80% and ꢀ90% of the detected products. Formation of liquid-phase products, including
furans, oxygenated aliphatics, cyclopenten-1-ones, and pyrans is discussed. Further study of the humins
formed during glucose conversion showed the effective inhibition of humin formation by DMOE. The
structure of humins was characterized by FTIR spectroscopy. Finally, HMF production from disaccharides
(sucrose, maltose and cellobiose) and polysaccharide (cellulose) using the water–DMOE system resulted
in good yields, demonstrating that our single-phase water–DMOE solvent system has good potential use
in HMF production from glucose and complex carbohydrates.
Received 14th September 2015
Accepted 24th September 2015
DOI: 10.1039/c5ra18824e
www.rsc.org/advances
have been proven to be high-performance catalysts for HMF
production from glucose.8,9 In aqueous medium, they dissociate
Introduction
to form metal-complex ions and hydrogen ions. Metal-complex
ions behave as Lewis acids and promote the isomerization
reaction, while hydrogen ions behave as a Brønsted acid, which
catalyzes the dehydration reaction.10 Among the metal chlo-
rides, AlCl3 holds promise in industrial applications because of
its low cost and low toxicity.
Biomass, the only renewable carbon resource that can be con-
verted to chemicals and liquid fuels, is a highly signicant
substitute for fossil resources.1,2 5-Hydroxymethylfurfural
(HMF), a dehydration product of C6 carbohydrates, is one of the
10 most value-added bio-based platform chemicals, as it can be
used in the synthesis of chemicals and liquid fuels.3,4 Therefore,
the efficient and economical conversion of carbohydrates to
HMF is crucial to biomass utilization. Production of HMF using
fructose as feedstock has high yield, but the limited sources and
high price of fructose restrict the large-scale application of this
process.5 Compared with fructose, glucose is the more abun-
dant monosaccharide in nature and is inexpensive.6
HMF production from glucose mainly comprises two steps,
namely, isomerization of glucose to fructose and dehydration of
fructose to HMF.7 Isomerization may be catalyzed by a Lewis
acid, and dehydration is usually catalyzed by a Brønsted acid.8
Thus, combining a Lewis acid and a Brønsted acid is an
important strategy in the catalytic conversion of glucose to
HMF.8 Metal chlorides such as CrCl3, SnCl4, CrCl2, and AlCl3
The reaction medium is one of the main factors that affect
HMF production. A good solvent favors not only a high HMF
yield, but also facilitates product separation and solvent circu-
lation.11 Water, a green solvent, is cheap and readily available.
However, HMF easily rehydrates into levulinic acid (LA) in
aqueous systems, leading to a decrease in the HMF yield.12 To
overcome this problem, a single-phase system consisting of an
aprotic polar organic solvent (such as N,N-dimethylacetamide
(DMA), N,N-dimethylformamide (DMF), and dimethylsulfoxide
(DMSO)) and a biphasic solvent system (such as water–methyl
isobutyl ketone (MIBK), water–tetrahydrofuran (THF), and
water–2-sec-butylphenol) was used as reaction media. Both
media can markedly increase the HMF yield.5 For example, Liu
et al.13 reported that the HMF yield from glucose in DMF and
DMA in the presence of CrCl2 catalyst reached 49.7% and
56.7%, respectively. When De et al.14 performed microwave-
assisted glucose dehydration in DMSO solvent and in
State Key Laboratory of Clean Energy Utilization, Zhejiang University, Hangzhou,
310027, China. E-mail: srwang@zju.edu.cn; qiukz@zju.edu.cn; Fax: +86-571-
87951616; Tel: +86-571-87952801; +86-571-87952066
84014 | RSC Adv., 2015, 5, 84014–84021
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
a water–MIBK biphasic system using AlCl3 catalyst, they ob-
tained HMF yields of 52% and 43%, respectively. Yang et al.15
obtained a HMF yield of 61% from glucose in a biphasic water–
THF system using AlCl3 as catalyst and NaCl as additive. Pagan-
Torres et al.9 achieved a HMF yield of 62% in glucose conversion
in a water–2-sec-butylphenol biphasic system using AlCl3 and
HCl as catalysts. However, use of the aforementioned reaction
media still has some disadvantages. DMA, DMF, and DMSO
have high boiling points, which make product separation
difficult.2 Biphasic solvent systems require a large amount of
organic solvent for extraction, and additional salt such as NaCl
is usually necessary to increase extraction efficiency, making the
reaction system complex and energy-intensive.5 In 2007, Zhao
et al.16 obtained ꢀ70% HMF yield in glucose conversion using
the ionic liquid 1-ethyl-3-methyl-imidazolium chloride as
solvent and CrCl2 as catalyst. This high yield has drawn interest
in the use of ionic liquids in HMF production and has led to its
further improvement.5 However, the high price of ionic liquids
and the difficulty of HMF separation from ionic liquids due to
their high viscosity and high boiling point prevent the extensive
application of ionic liquids.5 Recently, Liu et al.17 obtained
a HMF yield of 68% from fructose conversion in isopropanol
using NH4Cl as catalyst. When Gallo et al.18 adopted a single-
phase system consisting of water and THF (1 : 9 w/w) for
glucose conversion, they achieved 50.4% HMF yield in the
presence of AlCl3 and HCl. These important results suggest that
a single-phase reaction system based on a low-boiling-point
green solvent has strong potential use in large-scale HMF
production.11
In the present study, we developed a single-phase reaction
system for HMF production. It is composed of water and
1,2-dimethoxyethane (DMOE) solvent and uses AlCl3 as catalyst.
DMOE is a polar aprotic organic solvent with low boiling point
(84.6 ꢁC) and has good solubility in water. Therefore, the water–
DMOE solvent system can signicantly decrease the cost of
product separation. In addition, DMOE is recognized as
a renewable solvent because it can be synthesized from bio-
based ethylene glycol.19 We rst focused on the conversion of
glucose to HMF in this reaction system. Specically, we inves-
tigated the effects of the water/DMOE ratio, reaction tempera-
ture, reaction time, catalyst dosage and glucose load to optimize
the reaction conditions for HMF production. We also analyzed
the liquid-phase products and polymers formed under
optimum conditions by gas chromatography-mass spectrom-
etry (GC-MS) and Fourier Transform Infrared Spectroscopy
(FTIR), respectively. Finally, the conversion of disaccharides
and polysaccharide to HMF in the single-phase reaction system
was analyzed.
Procedure for conversion of carbohydrates into HMF
A Parr reactor (Parr Instrument, 100 mL) was charged with
a mixture of 10 mmol sugar (glucose, sucrose, maltose, cello-
biose, or cellulose) based on monosaccharide units, 50 mL
H2O–DMOE at a specied volume ratio (1 : 0, 1 : 1, 1 : 2, 1 : 4
and 0 : 1), and a specied amount of the catalyst AlCl3$6H2O
(0.25 mmol to 2.5 mmol). The reactions were carried out under
mechanical agitation in a N2 atmosphere, and the stirring rate
was maintained at 300 rpm. Aer the preset reaction tempera-
ture and time were reached, heating was stopped and the
reactor was cooled by air ow. Samples were ltered through
a 0.22 mm syringe lter prior to high performance liquid chro-
matography (HPLC) and GCMS analyses.
Analytical methods
The ltered samples were analyzed on a Dionex HPLC system
equipped with a Bio-Rad Aminex HPX-87H column and an RI
2000 refractive index detector. A H2SO4 solution (pH 2.5) was
used as the mobile phase. The ow rate and column tempera-
ture were kept at 0.6 mL minꢂ1 and 60 ꢁC, respectively. The
concentrations of glucose, HMF, and LA were determined by
comparison against standard calibration curves. The conver-
sion of glucose, as well as the yields of HMF and LA were
calculated as follows:
moles of glucose reacted
Glucose conversion ¼
ꢃ 100%
ꢃ 100%
ꢃ 100%
moles of starting glucose
moles of fructose produced
moles of starting glucose
Fructose yield ¼
HMF yield ¼
moles of HMF produced
moles of starting glucose or hexoses
moles of LA produced
LA yield ¼
ꢃ 100%
moles of starting glucose or hexoses
In order to understand the process of glucose decomposi-
tion, the ltered samples were also analyzed by GC-MS (Thermo
Scientic, Trace DSQII) equipped with a DB-wax capillary
column (30 m ꢃ 0.25 mm ꢃ 0.25 mm). Helium (99.999%) at
a ow rate of 1 mL minꢂ1 was used as carrier gas. The GC oven
temperature was programmedꢂt1o increase from 40 ꢁC (1 min) to
240 ꢁC (20 min) at 8 ꢁC min heating rate. The mass spec-
trometer was operated in electron ionization mode at 70 eV ion-
source electron energy. The scanning m/z ratio ranged from 35
to 450. All detected chemicals were identied by comparison
with the NIST (National Institute of Standards) MS library.
Experimental
Materials
Glucose, sucrose, maltose, cellobiose, microcrystalline cellu-
lose, DMOE, HMF, and LA were purchased from Aladdin
Humins isolation and analyses
Industrial Corporation. AlCl3$6H2O was purchased from Sino- Insoluble and soluble polymeric humins of the glucose
pharm Chemical Reagent Co., Ltd. All chemicals were of conversion process were obtained as follows. Aer reaction, the
analytical grade and were used without further purication.
samples were ltered through a 0.22 mm organic lter
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 84014–84021 | 84015

View Article Online
RSC Advances
Paper
membrane to obtain the insoluble residue. The remaining solid reacting with intermediates to form polymers, or from reacting
residue aer washing with water and DMOE was considered as with water to form LA.8,21 While in pure DMOE solvent, the HMF
humin-insoluble. Filtrates were further evaporated under yield decreased to 9.13% sharply with a glucose conversion of
vacuum to remove DMOE, and the precipitate was collected and 92.99% without fructose formation, indicating that the absence
then washed with deionized water. The obtained solid was of water might alter glucose decomposition route. Similarly,
considered as humin-soluble. Both humins were dried at 105 ꢁC Cao et al.22 performed the glucose dehydration in THF and also
for 2 h prior to further analyses. The elemental composition of found that the HMF yields were 4% and 19% in pure THF and
the humins was determined on a Vario Micro elemental THF with 2.7 wt% water, respectively, proposing that glucose
analyzer (Elenemtar Analysensysteme GmbH, Germany). FTIR was mainly dehydrated to produce anhydrosugars in pure polar
spectra of the humins were recorded from 400 to 4000 cmꢂ1 at aprotic solvent. Therefore, DMOE with a small amount of water
a resolution of 4 cmꢂ1 on a Nicolet 5700 FTIR spectrometer by may be more benecial for selective conversion of glucose to
averaging 36 scans.
HMF. LA, an important byproducts of hexose decomposition,
was easily formed via rehydration of HMF catalyzed by Brønsted
acid. Zhang et al. reported that the methyl carbon of LA was
derived from the hydroxymethyl carbon of HMF.23 As shown in
Fig. 1, the LA yield increased slightly with addition of DMOE
into water, staying at a low level with a maximum yield of 6.5%
at the water–DMOE ratio of 1 : 4.
Results and discussion
Conversion of glucose into HMF
Effect of solvent system composition. The effect of solvent
system composition on glucose conversion at the reaction
temperature of 150 C in the presence of AlCl3 is depicted in
Effect of reaction temperature. Fig. 2 shows the effect of
ꢁ
ꢁ
reaction temperatures of 130 to 170 C on the dehydration of
Fig. 1. At a water–DMOE ratio of 1 : 0, the HMF yield was only
16.35% at a glucose conversion of 78.73%. And the yield of
fructose was up to 27%, indicating the good efficiency in
glucose isomerization catalyzed by AlCl3 in aqueous phase.
With the increase in the proportion of DMOE, the glucose
conversion and the HMF yield increased markedly, while the
yield of fructose decreased drastically. At a water–DMOE ratio of
1 : 4, the glucose conversion was 94.30%, and the HMF and
fructose yield reached 48.55% and 0.28%, respectively, sug-
gesting that DMOE favored the isomerization of glucose and
subsequent dehydration of fructose to HMF. This increase may
be due to the polar aprotic solvent effect provided by DMOE. As
the presence of polar aprotic organic solvent can change the
sugar tautomerism equilibrium to form more reactive tautomer,
thus enhancing the conversion of sugar;8,20 moreover, dilution
effect from DMOE solvent, together with hydrogen-bond inter-
action between DMOE and HMF, can also prevent HMF from
glucose to HMF. At 130 ꢁC, both the HMF and fructose yield
were low (24.29% and 2.88%, respectively), indicating that low
temperature hindered the isomerization of glucose to fructose.
With the increase in temperature to 150 ꢁC, glucose conversion
increased gradually to 93.97%, while the HMF yield increased to
a maximum of 48.55%, which then decreased to 39.96% at
ꢁ
170 C. LA production had the same trend as glucose conver-
sion, reaching 8.34% at 170 ꢁC. These observations indicate that
high temperature promoted the conversion of HMF to LA. In
their kinetic analysis of LA formation from glucose decompo-
sition in aqueous medium, Weingarten et al.24 found that low
temperature favored the rehydration of HMF to LA, in contrast
to the results of our study. This difference may be ascribed to
the different reaction medium used in glucose decomposition.
In the water–DMOE solvent system, high temperature can
accelerate water to contact more easily with the formed HMF to
Fig. 1 Effect of water/DMOE ratio on glucose conversion and prod- Fig. 2 Effect of reaction temperature on glucose conversion and
ucts formation (reaction condition: 10 mmol glucose, 2.5 mmol products formation (reaction condition: 10 mmol glucose, 2.5 mmol
AlCl3$6H2O, 50 mL water–DMOE, reaction temperature 150 ꢁC, AlCl3$6H2O, 50 mL water–DMOE (volume ratio ¼ 1 : 4), reaction time
reaction time 15 min).
15 min).
84016 | RSC Adv., 2015, 5, 84014–84021
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
form LA. Therefore, mild reaction conditions are more suitable the opposing trends in HMF and LA formation at high catalyst
for HMF production.
dosage is the increase in H+ concentration due to AlCl3 hydro-
Effect of catalyst dosage. In the water–DMOE solvent system, lysis, which promoted the condensation of HMF and interme-
AlCl3 hydrolysis forms the metal-complex ion [Al(OH)(H2O)5]2+, diates to polymers and the rehydration of HMF to LA.25
which behaves as a Lewis acid, and the Brønsted acid hydrogen
Effect of glucose load. As shown in Table 1, with the increase
ion.14 The AlCl3 dosage may affect hydrolysis equilibrium and of glucose load, the HMF yields decreased gradually. The HMF
may alter levels of active species, thereby inuencing product yields corresponding to glucose load of 10, 20 and 50 mmol
formation.10 As shown in Fig. 3(a), glucose conversion appar- were 58.55%, 50.55% and 39.57%, respectively. This decrease
ently increased when the AlCl3 dosage increased from 0.25 to may be due to the side reactions at high glucose concentration.
1 mmol; further addition of AlCl3 had negligible impact on the Nevertheless, the HMF yield maintained at a medium level
increase in glucose conversion. Notably, extending the reaction when the glucose load was up 50 mmol, indicating the potential
time improved glucose conversion to 95% at a reaction time of use of this single-phase water–DMOE solvent system for HMF
60 min. The fructose yield decreased obviously with the adding production from glucose.
of AlCl3, inferring that higher catalyst dosage accelerated the
conversion of fructose. Fig. 3(b) presents the effect of AlCl3
dosage on HMF and LA production. The HMF yield increased
Glucose decomposition products analyses
Liquid-phase products analysis. To further our under-
standing of glucose decomposition in the water–DMOE solvent
system, we performed GC-MS to analyze the evolution of typical
products under optimal conditions for HMF production. As
shown in Fig. 4, the decomposition products may be classied
into four categories, namely, furans, oxygenated aliphatics,
cyclopenten-1-ones, and pyrans. HMF (1) was the dominant
furan product, comprising >80%, which increased gradually
with extended reaction time. Other furans mainly included
furfural (2), 2-furanmethanol (3), 2-(2-hydroxyacetyl)furan (4),
5-methylfurfural (5), 2-methylfuran (6), 5-methyl-2(3H)-fur-
anone (7), and 2,5-dicarboxaldehydefuran (8). Together with
HMF, the furans comprised ꢀ90% of the products detected by
GC-MS, indicating that the water–DMOE solvent system could
promote the formation of furans. In addition to LA (9) and
formic acid (10) formed from the rehydration of HMF, other
oxygenated aliphatics composed of glycolaldehyde (11), dihy-
droxyacetone (12), acetic acid (13), 1,3-propanediol (14), and
1,2-propanediol-3-methoxy (15) were mainly formed by retro-
aldolization of the intermediates.26–28 Levels of oxygenated
aliphatics had low values varied in range of 3–4.5% with pro-
longed reaction time. Cyclopenten-1-ones mainly consisted of
2-cyclopenten-1-one (16), 2-hydroxy-2-cyclopenten-1-one (17),
and 2-methyl-2-cyclopenten-1-one (18). These ve-carbon-ring
products might have been produced by cyclization of chain
intermediates containing multifunctional groups. Their
proportion increased from 0.76% to 1.16% with extended
reaction time. GC-MS detected two pyrans, namely, levogluco-
san (19) and 2,3-dihydro-3,5-dihydroxy-6-methyl-4H-pyran-4-
rst and then decreased with the addition of AlCl3; maximum
yields of HMF corresponding to AlCl3 dosages of 0.25, 0.5, 1,
and 2.5 mmol were 48.96%, 58.56%, 57.07%, and 48.55%,
respectively. The reaction time needed to reach the maximum
yield shortened from 45 to 15 min. Higher AlCl3 dosage
(2.5 mmol) enhanced the yield of LA to >6.5%. The reason for
Table 1 Effect of glucose load on HMF yield (reaction conditions:
50 mL water–DMOE (volume ratio ¼ 1 : 4), reaction temperature
150 ꢁC, reaction time 45 min)
Glucose
load
(mmol)
Catalyst
dosage
(mmol)
Glucose
conversion
(%)
HMF yield
(%)
LA yield
(%)
Entry
1
2
3
10
20
50
0.5
0.5
0.5
97.80
96.60
96.67
58.56
50.59
39.57
2.99
2.23
2.07
Fig. 3 Effect of AlCl3$6H2O catalyst dosage on glucose conversion (a)
and products formation (b) (reaction condition: 10 mmol glucose, 50 mL
water–DMOE (volume ratio ¼ 1 : 4), reaction temperature 150 ꢁC).
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 84014–84021 | 84017

View Article Online
RSC Advances
Paper
have a furanic polymer structure formed by condensation of
sugars, HMF, and intermediates.30 As they are carbonaceous
materials, humins have to be structurally characterized for their
value-added utilization. For example, humins can be used as
feedstock for production of synthesis gas or liquid-fuels
through thermal–chemical conversion, which may further
improve the market competitiveness of HMF production from
carbohydrates.31,32
In aqueous medium, acid-catalyzed conversion of glucose
can produce >30 wt% insoluble humins;31 whereas pure organic
solvent or a biphasic system may suppress formation of humins
or may cause them to dissolve, thereby reducing the effect of
insoluble solids on reactor operation.5 In our water–DMOE
solvent system, the amounts of humin-insoluble and humin-
soluble byproducts formed under optimal conditions for HMF
production were 0.67 and 4.17 wt% (Table 2), suggesting that
the water–DMOE solvent system was capable of inhibiting
humin formation. Moreover, most of the formed humins
(86.16%) could dissolve, implying that our water–DMOE system
may be used as an alternative to large-scale production of HMF
from glucose.
Fig. 4 Products distribution from glucose conversion in water–DMOE
solvent system based on GC-MS analysis (reaction condition: 10 mmol
glucose, 0.25 mmol AlCl3$6H2O, 50 mL water–DMOE (volume ratio ¼
1 : 4), reaction temperature 150 ꢁC).
Elemental compositions of the two humins are presented in
Table 2. The carbon content of humin-insoluble and humin-
soluble was 52.37 and 58.13 wt%, respectively, which are
lower than that of humin (65 wt%)33 formed in aqueous
medium via catalysis by H2SO4. These results reveal that the
dehydration and condensation catalyzed by AlCl3 was much
milder than mineral acids (H2SO4, HCl, etc.). In particular, the
carbon and hydrogen content of humin-insoluble was lower
than that of humin-soluble, suggesting that formation of
humin-insoluble might not have resulted from aggregation and
particle growth of humin-soluble. The van Krevelen diagram
(Fig. 6) reveals the reactions for the formation of two humins.
Humin-insoluble is located in the dehydration line and between
one (DDMP) (20). Levoglucosan is a dehydration product of
glucose, while DDMP forms from the dehydration and cycliza-
tion of enediol intermediates.27 The proportion of levoglucosan
and DDMP decreased to zero at 30 and 60 min, respectively,
implying that they could further convert to other products.
Levoglucosan can be dehydrated to HMF via acid catalysis.29
Fig. 5 depicts the simplied reaction pathways for products
formed from glucose. However, the mechanism related to
formation of various products is very complex; much work is
still required to reveal them.
Polymeric humins analysis. In the acid-catalyzed conversion
of carbohydrates to HMF by liquid-phase reaction, solid poly-
mer humin is the main byproduct. Humin has been reported to
Fig. 5 Liquid-phase products formation pathways for glucose conversion in water–DMOE solvent reaction system.
84018 | RSC Adv., 2015, 5, 84014–84021
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
Table 2 Yields and elemental compositions of the humins formed under optimal conditions (10 mmol glucose, 0.5 mmol AlCl3$6H2O, 50 mL
water–DMOE (volume ratio ¼ 1 : 4), reaction temperature 150 ꢁC, reaction time 45 min)
Humin
Yield (wt%)
C (wt%)
H (wt%)
O (wt%)
O/C
H/C
Humin-insoluble
Humin-soluble
0.67
4.17
52.37
58.13
5.28
5.39
42.35
36.48
0.61
0.47
1.21
1.11
glucose and HMF, implying that the humin-insoluble formed by formation of humin-soluble may incorporate with HMF. This
condensation of glucose, HMF, and intermediates, along with may be conrmed by the elemental compositions of both
formation of water.30 Humin-soluble is located below the humins, which were close to those of glucose and HMF,
dehydration line and on the right side of HMF, indicating that respectively. In particular, the absence of absorption at
its formation might involve alkylation or alkyl-chain 1665 cmꢂ1 for the two humins, which corresponds to the
elongation.34
aldehyde group in HMF, indicates that humin formation
To further illuminate the structural properties of the two involves aldol addition or condensation reactions that destroy
humins, we carried out FTIR spectroscopy. As shown in Fig. 7, carbonyl groups.37
the two spectra are similar, differing only in absorbance
intensity. The broad peaks at 3600–3200 cmꢂ1 correspond to
O–H strong stretching, thus indicating the existence of hydroxyl
Conversion of disaccharides and polysaccharide into HMF
groups.35 Absorptions at 2960–2850 cmꢂ1 were due to stretching The conversion of disaccharides (sucrose, maltose, and cello-
of methyl and methylene groups,36 and humin-soluble showed biose) and polysaccharide (cellulose) to HMF in the single-
sharper absorptions, indicating the higher content of alkyl phase water–DMOE solvent system was examined (Table 3).
group in humin-soluble, which was in line with the above Under the optimal conditions for HMF production from
conclusion that formation of humin-soluble may involve alkyl- glucose, the disaccharides sucrose, maltose, and cellobiose had
ation or alkyl-chain elongation. The peak at 1703 cmꢂ1 may be modest HMF yields (36.81%, 35.24%, and 33.51%, respectively).
attributed to C]O stretching in acids, aldehydes, and Cellulose had a very low HMF yield (7.15%), which was much
ketones.33 Peaks derived for the substituted furanic rings could lower than that from glucose (58.56%). This difference may be
be observed in the spectra of the two humins. For example, due to the insufficient acidity of the Brønsted acid from AlCl3
peaks at 1619 and 1518 cmꢂ1 correspond to C]C stretching in leading to inefficient breakage of glucosidic bonds. Cellulose
furan rings. Peaks at 1395 cmꢂ1 and bands at 798 cmꢂ1 are has a stable crystalline structure with strong intramolecular and
respectively ascribed to C–O–C stretching and C–H out-of-plane intermolecular hydrogen bonds, which makes its depolymer-
deformation in furan rings.30,31 Absorptions at 1022 cmꢂ1 and at ization very difficult.38,39 This stability is conrmed by the large
1164–1072 cmꢂ1 correspond to primary and secondary alcohols, amount of unreacted cellulose remaining aer the reaction. To
respectively. Humin-soluble showed notable absorption at promote the cleavage of glucosidic bonds, the AlCl3 dosage was
1022 cmꢂ1, while humin-insoluble showed sharp absorptions at increased from 0.5 to 2.5 mmol. This increase improved the
1164–1072 cmꢂ1. Considering that HMF contains only primary HMF yields from sucrose, maltose, cellobiose, and cellulose to
alcohol while sugars contain both primary and secondary 44.86%, 40.74%, 41.98%, and 18.89%, respectively, suggesting
alcohols, we can conclude that the formation of humin- the positive effect of breaking glucosidic bonds efficiently on
insoluble may incorporate with HMF and glucose, while the the conversion of polysaccharides to HMF. Although the HMF
yield from cellulose distinctly increased, it was still low.
Therefore, we performed cellulose conversion under more
Fig. 6 van Krevelen diagram of humins, HMF, glucose and cellulose.
Fig. 7 FTIR spectra of humin-insoluble and humin-soluble.
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 84014–84021 | 84019

View Article Online
RSC Advances
Paper
Table
3 HMF yield from various carbohydrates in water–DMOE
results suggest that our simple single-phase water–DMOE
solvent system could be used efficiently and economically in
HMF production from carbohydrates.
solvent reaction systema
Entry
Feed
Catalyst/mol%
HMF yield/%
1
2
3
4
5
6
7
8
9
Glucose
5
5
25
5
25
5
25
5
25
25
58.56
36.81
44.86
35.24
40.74
33.51
41.98
7.15
Acknowledgements
Sucrose
Sucroseb
Maltose
The authors are grateful for the nancial support from the
National Natural Science Foundation of China (51476142).
Maltoseb
Cellobiose
Cellobioseb
Cellulose
Celluloseb
Cellulosec
Notes and references
18.89
37.73
10
1 G. W. Huber, S. Iborra and A. Corma, Chem. Rev., 2006, 106,
4044–4098.
2 M. Dashtban, A. Gilbert and P. Fatehi, RSC Adv., 2014, 4,
2037–2050.
a
Reaction conditions: feed load 10 mmol, 50 mL water–DMOE (volume
ratio ¼ 1 : 4), reaction temperature 150 ꢁC, reaction time 45 min.
b
c
Reaction time 15 min. Reaction temperature 170 ꢁC, reaction time
80 min.
3 S. M. Kang, J. Ye, Y. Zhang and J. Chang, RSC Adv., 2013, 3,
7360–7366.
4 H. P. Gao, Y. X. Peng, J. M. Pan, J. Zeng, C. H. Song,
Y. L. Zhang, Y. S. Yan and W. D. Shi, RSC Adv., 2014, 4,
43029–43038.
5 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.
6 Y. Yang, W. T. Liu, N. N. Wang, H. J. Wang, Z. X. Song and
W. Li, RSC Adv., 2015, 5, 27805–27813.
7 N. Shi, Q. Liu, L. Ma, T. Wang, Q. Zhang, Q. Zhang and
Y. Liao, RSC Adv., 2014, 4, 4978–4984.
severe conditions of increased reaction temperature and
extended reaction time (170 ꢁC and 80 min) using 2.5 mmol
AlCl3. We thus obtained a satisfactory HMF yield (37.73%). This
is comparable to the results for the biphasic reaction system.
Yang et al.15 reported a HMF yield of 37% from cellulose in
a biphasic water–THF medium using AlCl3 as catalyst and NaCl
as additive. Similarly, Shen et al.40 observed a HMF yield of
39.7% when they applied a biphasic water–THF system for
cellulose conversion using InCl3 as catalyst and NaCl as addi-
tive. Therefore, our single-phase water–DMOE system also has
potential use in production of HMF from complex
polysaccharides.
8 T. F. Wang, M. W. Nolte and B. H. Shanks, Green Chem.,
2014, 16, 548–572.
9 Y. J. Pagan-Torres, T. F. Wang, J. M. R. Gallo, B. H. Shanks
and J. A. Dumesic, ACS Catal., 2012, 2, 930–934.
10 V. Choudhary, S. H. Mushrif, C. Ho, A. Anderko,
V. Nikolakis, N. S. Marinkovic, A. I. Frenkel, S. I. Sandler
and D. G. Vlachos, J. Am. Chem. Soc., 2013, 135, 3997–4006.
Conclusions
We investigated the application of DMOE, a low-boiling polar
aprotic organic solvent, in HMF production from carbohy- 11 S. P. Teong, G. Yi and Y. Zhang, Green Chem., 2014, 16, 2015–
drates. The results show that the HMF yield of a single-phase 2026.
reaction system of water and DMOE using AlCl3 catalyst 12 T. D. Swi, C. Bagia, V. Choudhary, G. Peklaris, V. Nikolakis
increased substantially relative to that of pure water reaction and D. G. Vlachos, ACS Catal., 2014, 4, 1320.
medium with the increase in DMOE proportion because of the 13 D. J. Liu and E. Y. X. Chen, Biomass Bioenergy, 2013, 48, 181–
polar aprotic solvent effect of DMOE. Under optimal conditions 190.
(1 : 4 water–DMOE volume ratio, 5 mol% AlCl3$6H2O dosage, 14 S. De, S. Dutta and B. Saha, Green Chem., 2011, 13, 2859–
ꢁ
150 C reaction temperature, 45 min reaction time), a signi-
2868.
cant HMF yield (58.56%) from glucose was obtained. GC-MS of 15 Y. Yang, C. W. Hu and M. M. Abu-Omar, Green Chem., 2012,
the liquid-phase products showed that the total proportion of 14, 509–513.
HMF and furans respectively comprised 80% and up to 90% of 16 H. B. Zhao, J. E. Holladay, H. Brown and Z. C. Zhang, Science,
the products detected. In addition to furans, oxygenated 2007, 316, 1597–1600.
aliphatics, cyclopenten-1-ones, and pyrans, were present. 17 J. T. Liu, Y. Tang, K. G. Wu, C. F. Bi and Q. Cui, Carbohydr.
Formation of these various products involved many complex Res., 2012, 350, 20–24.
reactions, including dehydration, retroaldolization, cyclization, 18 J. M. R. Gallo, D. M. Alonso, M. A. Mellmer and J. A. Dumesic,
and fragmentation. Analyses of the polymer humins revealed Green Chem., 2013, 15, 85–90.
that DMOE inhibited the formation of humins, with yields of 19 B. Kim, C. A. Antonyraj, Y. J. Kim, B. Kim, S. Shin, S. Kim,
0.67 and 4.17 wt% for humin-insoluble and -soluble byprod-
ucts, respectively. On the basis of results for glucose conversion,
K.-Y. Lee and J. K. Cho, Ind. Eng. Chem. Res., 2014, 53,
4633–4641.
the reaction system was further optimized for the conversion of 20 M. H. Tucker, R. Alamillo, A. J. Crisci, G. M. Gonzalez,
disaccharides (sucrose, maltose, and cellobiose) and poly-
saccharides (cellulose) to achieve a satisfactory HMF yield. The
S. L. Scott and J. A. Dumesic, ACS Sustainable Chem. Eng.,
2013, 1, 554–560.
84020 | RSC Adv., 2015, 5, 84014–84021
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
21 X. Hu, R. J. M. Westerhof, D. H. Dong, L. P. Wu and C. Z. Li, 31 T. M. C. Hoang, E. R. H. van Eck, W. P. Bula,
ACS Sustainable Chem. Eng., 2014, 2, 2562–2575.
22 F. Cao, T. J. Schwartz, D. J. McClelland, S. H. Krishna,
J. G. E. Gardeniers, L. Lefferts and K. Seshan, Green Chem.,
2015, 17, 959–972.
J. A. Dumesic and G. W. Huber, Energy Environ. Sci., 2015, 32 S. R. Wang, B. Ru, H. Z. Lin, W. X. Sun and Z. Y. Luo,
8, 1808–1815.
Bioresour. Technol., 2015, 182, 120–127.
23 J. Zhang and E. Weitz, ACS Catal., 2012, 2, 1211–1218.
24 R. Weingarten, J. Cho, R. Xing, W. C. Conner and
G. W. Huber, ChemSusChem, 2012, 5, 1280–1290.
33 C. B. Rasrendra, M. Windt, Y. Wang, S. Adisasmito,
I. G. B. N. Makertihartha, E. R. H. van Eck, D. Meier and
H. J. Heeres, J. Anal. Appl. Pyrolysis, 2013, 104, 299–307.
25 B. Girisuta, L. P. B. M. Janssen and H. J. Heeres, Green Chem., 34 S. Kim, R. W. Kramer and P. G. Hatcher, Anal. Chem., 2003,
2006, 8, 701–709. 75, 5336–5344.
26 S. Dutta, S. De and B. Saha, Biomass Bioenergy, 2013, 55, 355– 35 S. R. Wang, B. Ru, H. Z. Lin and W. X. Sun, Fuel, 2015, 150,
369. 243–251.
27 M. B. Fusaro, V. Chagnault and D. Postel, Carbohydr. Res., 36 S. Wang, H. Lin, B. Ru, W. Sun, Y. Wang and Z. Luo, J. Anal.
2015, 409, 9–19. Appl. Pyrolysis, 2014, 108, 78–85.
28 Y. Yu and H. W. Wu, Ind. Eng. Chem. Res., 2011, 50, 10500– 37 S. K. R. Patil and C. R. F. Lund, Energy Fuels, 2011, 25, 4745–
10508. 4755.
29 R. Weingarten, A. Rodriguez-Beuerman, F. Cao, 38 S. R. Wang, X. J. Guo, T. Liang, Y. Zhou and Z. Y. Luo,
J. S. Luterbacher, D. M. Alonso, J. A. Dumesic and
G. W. Huber, ChemCatChem, 2014, 6, 2229–2234.
30 I. van Zandvoort, Y. Wang, C. B. Rasrendra, E. R. van Eck,
Bioresour. Technol., 2012, 104, 722–728.
39 Y. Qian, L. F. Zhu, Y. Wang and X. C. Lu, Renewable
Sustainable Energy Rev., 2015, 41, 633–646.
P. C. Bruijnincx, H. J. Heeres and B. M. Weckhuysen, 40 Y. Shen, J. K. Sun, Y. X. Yi, M. F. Li, B. Wang, F. Xu and
ChemSusChem, 2013, 6, 1745–1758.
R. C. Sun, Bioresour. Technol., 2014, 172, 457–460.
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 84014–84021 | 84021