Catalytic conversion of carbohydrates into 5-hydroxymethylfurfural by germanium(IV) chloride in ionic liquids

Article abstract of DOI:10.1002/cssc.201000279

Direct conversion of carbohydrates into 5-hydroxymethylfurfural (HMF) catalyzed by germanium(IV) chloride in ionic liquids has been investigated in search of an efficient and environmentally friendly process. Monosaccharides D-fructose and D-glucose, disaccharides sucrose and maltose, and even the polysaccharide cellulose were successfully converted into HMF with good yields under mild conditions (yield up to 92 % in 5 min in the case of fructose). The structure of ionic liquids, catalyst loading, reaction temperature and water content had noticeable effects on this catalytic system. Addition of 5 ? molecular sieves during the dehydration of glucose resulted in an increase in HMF yield from 38.4 % to 48.4 %. A mechanism for glucose conversion to HMF catalyzed by germanium(IV) chloride was proposed according to ¹³C NMR spectra obtained in situ under different conditions using D-glucose-2- ¹³C as the substrate. Ge whiz: Germanium(IV) chloride in ionic liquids catalyzes the direct conversion of carbohydrates into 5- hydroxymethylfurfural (HMF). The catalyst shows excellent activity for fructose dehydration, and a high yield of HMF up to 90 % is obtained in 5 min. This nontoxic catalytic system is also efficient for other carbohydrates such as glucose, sucrose, and even cellulose.

Full text of DOI:10.1002/cssc.201000279

DOI: 10.1002/cssc.201000279
Catalytic Conversion of Carbohydrates into
5
-Hydroxymethylfurfural by Germanium(IV) Chloride in
Ionic Liquids
[a, c]
[a]
[a, b]
[a]
[a, b]
Zehui Zhang,
Qian Wang, Haibo Xie,*
Wujun Liu, and Zongbao (Kent) Zhao*
Direct conversion of carbohydrates into 5-hydroxymethyl-
furfural (HMF) catalyzed by germanium(IV) chloride in
ionic liquids has been investigated in search of an efficient and
environmentally friendly process. Monosaccharides d-fructose
and d-glucose, disaccharides sucrose and maltose, and even
the polysaccharide cellulose were successfully converted into
HMF with good yields under mild conditions (yield up to 92%
in 5 min in the case of fructose). The structure of ionic liquids,
catalyst loading, reaction temperature and water content had
noticeable effects on this catalytic system. Addition of 5 ꢁ mo-
lecular sieves during the dehydration of glucose resulted in an
increase in HMF yield from 38.4% to 48.4%. A mechanism for
glucose conversion to HMF catalyzed by germanium(IV) chlo-
13
ride was proposed according to C NMR spectra obtained
13
in situ under different conditions using d-glucose-2- C as the
substrate.
Introduction
[
11]
Concerns about global warming and energy security have led
to the exploration of alternatives for fossil resources to supply
for 3 h in 1-ethyl-3-methylimidazolium tetrafluoroborate.
Stahlberg et al. used the lanthanide salt-based system for the
[
1]
[12]
chemicals and energy. Biomass with an estimated global pro-
reaction at 1408C for 6 h, affording HMF in 24% yield. Re-
11
duction of around 1.0ꢀ10 tons per year has attracted consid-
erable attention as an alternative source for both fuels and
gardless of these developments, it remains a challenge to
obtain HMF from carbohydrates in terms of efficiency and
greenness. Herein, we report a germanium(IV) chloride-based
catalytic system that promotes efficient HMF production under
mild conditions with a variety of carbohydrates as the
substrates.
[
2]
chemicals. Carbohydrates make up the majority of biomass,
and it is estimated that up to 30% of raw materials for the
chemical industry will be produced from renewable sources by
[
3]
2
025. To achieve this goal, it is pivotal to develop more
efficient and environmentally friendly methods to convert
[
4]
carbohydrates into useful chemicals.
-Hydroxymethylfurfural (HMF) has been recognized as a
versatile and key precursor for the production of fine chemi- Results and Discussion
5
[
5]
cals, polymeric materials, and biofuels. In the past few years,
the preparation of HMF through the dehydration of biomass-
based sugars has received much attention. In the case of fruc-
Catalyst screening
We initiated our study using fructose as the substrate and
screened several metal salts that had rarely been considered as
catalysts for HMF production in ionic liquids. HMF yields were
largely dependent on the catalysts (Table 1). An excellent yield
[6]
tose, excellent yields were achieved using many methods.
Although fructose is the preferred feedstock for high HMF
yield, it is clear that large scale, sustainable production of
HMF requires cellulosic biomass as the feedstock. However,
dehydration of glucose is much more problematic, and
of 92% was achieved in 5 min when GeCl was employed
4
(Table 1, entry 2). Only moderate yields (Table 1, entries 5 and
6) or low yields (Table 1, entries 1, 3, 4, and 7) were obtained
with other choices of catalyst. Zhao et al. reported that HMF
could be produced from fructose without a catalyst in
[
7]
considerable efforts have been devoted to this area .
A recent study demonstrated that the conversion of glucose
into HMF may include consecutive steps, namely, mutarotation
and isomerization of glucose into fructose followed by dehy-
[
8]
[a] Z. Zhang, Q. Wang, Prof. H. Xie, Dr. W. Liu, Prof. Z. K. Zhao
dration of fructose. Both the catalyst and the reaction envi-
ronment had a major effect on the efficiency of the conversion
Dalian Institute of Chemical Physics, CAS
4
57 Zhongshan Road, Dalian (China)
[
9]
of glucose into HMF. In 2007, Zhao et al. reported that
chromium(II) chloride catalyzed the dehydration of glucose
into HMF in 1-ethyl-3-methylimidazolium chloride ([EMIM]Cl) in
Fax: (+86)411-84379211
E-mail: hbxie@dicp.ac.cn
zhaozb@dicp.ac.cn
[
10]
[b] Prof. H. Xie, Prof. Z. K. Zhao
a good yield.
Later on, conversion of carbohydrates into
Dalian National Laboratory for Clean Energy, Dalian 116023 (China)
HMF was realized with less toxic catalysts. For example, Hu
[
c] Z. Zhang
Graduate University of Chinese Academy of Sciences
Beijing 100039 (China)
et al. reported that the common Lewis acid SnCl could cata-
4
lyze glucose conversion into HMF in about 60% yield at 1008C
ChemSusChem 2011, 4, 131 – 138
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
131

H. Xie, Z. K. Zhao et al.
Table 1. The results of fructose dehydration to HMF catalyzed by
[
a]
different catalysts.
Entry
Catalyst
SnCl ·4H
Yields [%]
1
2
3
4
5
6
7
4
2
O
5.5
92.1
8.1
GeCl
BiCl
ScCl
ZrCl
HfCl
4
3
3
·6H
2
O
4.0
4
43.8
57.7
0.9
4
CeCl
–
3
[
b]
8
ND
[a] Reaction conditions: Fructose (100 mg) was added to [BMIM]Cl (2.0 g)
at 1008C, then catalyst (10 mol%) was added. The reaction was stopped
after 5 min. [b] The reaction was done under otherwise identical
conditions in the absence of a catalyst. ND=not detected.
[10]
[EMIM]Cl especially at a high temperature for a long time.
For a control experiment in which fructose was held at 1008C
in the absence of a catalyst, we did not detect any HMF after
5
min (Table 1, entry 8). All of the results clearly indicated that
GeCl was a superior catalyst for the dehydration of fructose.
4
Chan et al. recently reported that the dehydration of fruc-
tose catalyzed by WCl at 508C for 4 h gave HMF in 63%
6
[
13]
yield. Inspired by that study, we reduced the reaction tem-
perature. At 808C, the HMF yield reached 90.2% in 5 min
(Figure 1). The time course of HMF formation at 808C was
almost identical to that at 1008C. However, when the reaction
temperature was 508C, it took 20 min to reach a yield of
Figure 1. The results of fructose dehydration into HMF catalyzed by GeCl at
4
8
4.3%, and thereafter HMF yield increased only marginally
different temperatures. Reaction conditions: Fructose (100 mg) was added
to [BMIM]Cl (2.0 g), then GeCl (10 mol%) was added.
4
over time. It should be pointed out that fructose conversion
was 96.1% at 508C after 20 min, and reached over 99.1% after
30 min. To our knowledge, this result is the best reported to
date for the production of HMF from fructose in a Lewis
acid–ionic liquids system. The low reaction temperature should
facilitate an extraction of HMF using a low boiling point
solvent (e.g., ethyl acetate), providing much potential to
develop a more economical and sustainable process for
[
14]
HMF production
.
Moreover, the reusability of [BMIM]Cl/GeCl (BMIM=1-butyl-
4
3-methylimidazolium) system was estimated. In this experi-
ment, HMF was extracted with ethyl acetate prior to the
addition of fructose into the system for the next run under
essentially identical conditions. Our results demonstrated
that the [BMIM]Cl/GeCl4 system could be reused five times
without significant activity loss (Figure 2).
Figure 2. Reuse of the GeCl
4
/[BMIM]Cl system. Reaction conditions: Fructose
(10 mol%), T=1008C, t=5 min.
(
100 mg), [BMIM]Cl (2.0 g), GeCl
4
Effect of catalyst loading
Although fructose has been the preferred feedstock for HMF
production, its occurrence in nature is limited. We decided to
use more abundant carbohydrates, glucose and cellulosic bio-
mass, as the raw material in our system. When glucose was
treated at 1008C for 75 min in the presence of 10 mol% GeCl4,
the maximal yield was 38.4% with a glucose conversion of
prove the HMF yield. Firstly, the effect of catalyst loading was
studied. Increasing the catalyst loading from 5 mol% to
15 mol% resulted in an increase in glucose conversion
(Figure 3). For example, glucose conversion at 120 min reached
87.3% and 98.5% in the presence of 5 mol% and 15 mol%
9
2.5%. Because the yield was lower than those of other cata-
GeCl , respectively. It was clear that the maximal HMF yields
4
[
10,11]
lytic systems,
we changed the reaction conditions to im-
were almost identical regardless of whether the catalyst was
1
32
www.chemsuschem.org
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2011, 4, 131 – 138

Conversion of Carbohydrates by GeCl4
Figure 4. The results of glucose dehydration to HMF catalyzed by GeCl
different temperatures. Reaction conditions: Glucose (100 mg) was added to
BMIM]Cl (2.0 g), then GeCl (10 mol%) was added.
4
at
4
Figure 3. The results of glucose dehydration to HMF catalyzed by GeCl with
different catalyst loadings. Reaction conditions: Glucose (100 mg) was added
into [BMIM]Cl (2.0 g) at 1008C, then GeCl was added.
[
4
4
loaded at 10 mol% or 15 mol%. However, at the early stage
higher HMF yields were found for the reaction with 15 mol%
Effect of water content
GeCl . These results indicated that increasing the catalyst load-
We also investigated the effect of water content on the reac-
tion, because water could promote rehydration of HMF and
other side reactions. Both glucose conversion and HMF yield
decreased with increasing water content (Figure 5). For exam-
ple, when 2 equivalents of water were added, glucose conver-
sion and HMF yield after 120 min were 79.2% and 20.0%, re-
spectively, whereas those values were 92.5% and 38.4% with-
out water addition (see above). Levulinic acid and formic acid
are usually the main byproducts of the dehydration of hexoses
in an aqueous solution. However, levulinic acid was not detect-
ed in this work, indicating that rehydration of HMF was inhibit-
ed. In our experiment, we observed the formation of humins,
which were likely generated through the polymerization and
4
ing led to a faster reaction, whereas the reaction selectivity did
not change. Because the maximal HMF yield was achieved
using 10 mol% GeCl , we used this catalyst loading for the
4
subsequent experiments.
Effect of reaction temperature
Experiments were carried out at 80, 100, and 1208C. Higher re-
action temperatures clearly gave higher glucose conversions
and HMF yields (Figure 4). At 1208C, glucose conversion and
HMF yield increased rapidly in the early stage, and the maximal
HMF yield of 47.5% was received with 99.1% glucose conver-
sion at 30 min. At 1008C, the maximal HMF yield of 38.4% was
detected after 75 min. In both cases, HMF yields dropped over
time beyond the maximal point, indicating that degradation of
HMF was significant at higher temperatures. However, glucose
conversion and HMF yield were 71.8% and 28.6%, respectively,
at 808C for 120 min, indicating that the reaction was sup-
pressed at a lower temperature. Fructose was converted at a
lower temperature than glucose (Figure 1 versus Figure 4).
[14]
cross-polymerization of HMF and intermediates.
The de-
creased glucose conversion in the presence of water may be
ascribed to the reaction of GeCl with water. A further experi-
4
ment was carried out in the presence of 5 ꢁ molecular sieves
to remove water from the system. The HMF yield was found to
increase moderately, from 38.4% to 48.4%, in the presence of
0.5 g of 5 ꢁ molecular sieves.
ChemSusChem 2011, 4, 131 – 138
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
133

H. Xie, Z. K. Zhao et al.
Figure 5. The results of glucose dehydration to HMF catalyzed by GeCl
4
with
Figure 6. The results of glucose dehydration to HMF catalyzed by different
different amounts of water. Reaction conditions: Glucose (100 mg) was
germanium salts. Reaction conditions: Glucose (100 mg) was added to
4
added to [BMIM]Cl (2.0 g) at 1008C, then water and GeCl (10 mol%) were
added.
[
BMIM]Cl (2.0 g) at 1008C, then germanium salt (10 mol%) was added.
Conversion of glucose catalyzed by different germanium
salts
reaction was performed with 10 mol% GeBr2 under a N2
atmosphere.
We then tested two other commercially available germanium
Conversion of glucose in different ionic liquids
salts, Ge(OEt)4 and GeBr , for the dehydration of glucose.
2
Ge(OEt)4 was clearly less competent for HMF production
Ionic liquids are regarded as tunable solvents, and their struc-
[16]
(Figure 6). Glucose conversion and HMF yield were 63.0% and
tures have a remarkable effect on their properties. The inter-
actions among cation, anion and carbohydrates play a key role
during the process. Thus we tested a variety of ionic liquids as
25.7%, respectively, after 120 min in the presence of 10 mol%
Ge(OEt) . The difference may be explained by the anion effect.
4
À
When using Ge(OEt) as the catalyst, the EtO group is much
solvents for GeCl -promoted HMF production from glucose
4
4
À
larger than Cl , presenting more steric hindrance for the coor-
(Figure 7). Besides [BMIM]Cl, 1-hexyl-3-methylimidazolium
chloride ([HMIM]Cl), 1-octyl-3-methylimidazolium chloride
dination between glucose and the catalytic center. Further-
À
À
more, EtO is more nucleophilic than Cl , leading to a stronger
([OMIM]Cl),
and
1-decyl-3-methylimidazolium
chloride
IV
interaction with Ge , which inhibited the interaction between
([DMIM]Cl) were also proved as reasonably good solvents for
the reaction. These ionic liquids all incorporated chloride as
the anion, in which HMF yields ranged from 23.3% in
[DMIM]Cl to 38.4% in [BMIM]Cl. It was interesting to scrutinize
the effect of the chain length of the alkyl group of these liq-
uids on the reaction. Although glucose conversions were close,
HMF yields decreased from 38.4% to 23.3% when the alkyl
group changed from butyl to decyl. The opposite phenomena
were recently published for the reaction at 1608C using lantha-
IV
glucose and Ge . When GeBr was used, it was interesting to
2
note that both glucose conversion and HMF yield were low by
the first 30 min. However, HMF yield increased linearly after
3
0 min, and the result was slightly better than that with
II
IV [15]
Ge(OEt) . Ge is liable to be oxidized aerobically to Ge .
4
Therefore, apparent catalytic activity when GeBr2 was used
IV
might have been the result of the formation of Ge species
during the reaction process. The fact that HMF yield was
reduced to 2.1% was in line with our speculation when the
[12]
nide chlorides as catalysts in ionic liquids. Therefore, a gen-
1
34
www.chemsuschem.org
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2011, 4, 131 – 138

Conversion of Carbohydrates by GeCl4
4
Figure 7. The results of glucose dehydration to HMF catalyzed by GeCl in
Figure 8. The results of HMF production from various sugars. Reaction con-
ditions: Sugar (100 mg) was added to [BMIM]Cl (2.0 g) at 1008C followed by
different solvents for 75 min. Reaction condition: Glucose (100 mg) was
added to the solvent (2.0 g) at 1008C, then GeCl (10 mol%) was added.
4
GeCl
0 min. Maltose monohydrate was used because anhydrous maltose was
unavailable.
4
(10 mol%). The reaction mixture was then stirred at 1208C for
3
eral relationship between the chain length of the alkyl group
and HMF yields was not clear, and the catalyst and operational
conditions should also be taken into consideration. However,
ionic liquids with different anions, including 1-butyl-3-methyli-
midazolium acetate ([BMIM]OAc), 1-butyl-3-methylimidazolium
[
19]
[BMIM]Cl–IBMK/CrCl3 ; IBMK=isobutyl methyl ketone). When
cellobiose was utilized, HMF yield was 41.0%, which was com-
[19]
parable to that in a previous study. When maltose monohy-
drate was used, HMF yield was clearly lower than with the
other two disaccharides, cellobiose and sucrose. We reasoned
that the water introduced with maltose spoiled the reaction. It
should be noted that water also inhibited the conversion of
glucose into HMF (see above). More interestingly, when
cellulose was used, HMF yield was 35.0%. Although the yield
was a little lower than that for glucose conversion, the use of
cellulose avoided tedious hydrolysis and glucose separation
tetrafluoroborate ([BMIM]BF ), and 1-butyl-3-methylimidazoli-
4
um bistriflate imide ([BMIM]NTf ) were found to form catalyti-
2
cally inactive systems. Although the glucose conversion in
[BMIM]OAc was high, the selectivity of HMF was rather low,
which may be due to possible side reactions between glucose
and the imidazolium ring induced by the strongly basic acetate
[
17]
group. [BMIM]BF was found to be a poor solvent for the re-
4
action, even though it was a good solvent for the SnCl -based
4
system. Both glucose conversion and HMF selectivity were low
processes. Clearly, the GeCl -based system deserves further
4
in the hydrophobic ionic liquid [BMIM]NTf . Our results indicat-
investigation to facilitate sustainable biorefining of cellulosic
materials.
2
ed that anions in the ionic liquids had a major effect on the
HMF yield, and were consistent with the fact that a moderate
basicity of the anion was required to obtain a satisfactory glu-
Possible mechanism for glucose conversion with
[
18]
cose conversion and HMF selectivity. Other polar aprotic sol-
vents, dimethylsulfoxide (DMSO) and dimethylformamide
[
BMIM]Cl/GeCl4
(
DMF), were also effective, giving HMF yields of 37.1% and
4.3%, respectively, under otherwise identical reaction condi-
tions (Figure 7). DMSO is known to give similar yields of HMF if
Much effort has been devoted to revealing the mechanism of
the conversion of glucose into HMF catalyzed by Lewis acids.
However, in-depth understanding of this process remained
scarce. To attain more insights into this transformation, we re-
3
[
11]
monosaccharides are used as the substrates. It is likely that
DMSO and DMF provided similar environments to ionic liquids
to dissolve the catalyst and sugar. However, using ionic liquids
such as [BMIM]Cl had two obvious advantages in terms of
HMF recovery by extraction with ethyl acetate and direct
conversion of cellulose (see below).
13
corded a series of in situ C NMR spectra at different tempera-
13
tures using d-glucose-2- C as the substrate (Figure 9). Based
on these data, we have formulated a putative mechanism for
the GeCl -promoted HMF production from glucose (Scheme 1).
4
13
The C NMR spectra indicated that glucose was in the forms
of a-pyranose (2) and b-pyranose (1; Scheme 1) in [BMIM]Cl at
room temperature in the absence of GeCl (Figure 9a). Howev-
4
Conversion of various substrates into HMF with
er, when 1 equivalent of GeCl was added into the mixture, a
4
[BMIM]Cl/GeCl4
new peak at d=76.4 ppm appeared (Figure 9b). The ability of
heavier main-group elements to accommodate more than
eight electrons in their valence shells, thus forming so-called
hypervalent or hypercoordinate compounds, is well estab-
Next, we sought to explore the substrate scope of this catalytic
system. HMF was obtained in 55.4% yield when sucrose was
used (Figure 8). This result was slightly lower than previous
[
4
11]
[20]
results (64% HMF yield for [BMIM]BF /SnCl
and 73% for
lished. The cations of Group 14 elements can be stabilized
4
ChemSusChem 2011, 4, 131 – 138
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
135

H. Xie, Z. K. Zhao et al.
103.6 ppm appeared. The intensity of the signal at d=
152.1 ppm increased when the reaction was held at higher
temperatures (Figure 9d ande) and also increased over time
data not shown). Therefore, this signal was assigned to the
(
1
3
C-labeled carbon of HMF at the 2-position. Assignment of the
signal at d=103.6 ppm was more tricky. Because this signal
was found only at 508C, it should be associated to a relatively
stable intermediate with a short lifetime along the reaction
pathway to HMF. In other words, at higher temperature, the
turnover of this intermediate was greatly accelerated. There-
[22]
fore, we suppose it was from fructose, which indicated that
the intermediate fructose was formed through glucose isomer-
ization. Moreover, the signal at d=76.4 ppm was missing at
808C, indicating that the conversion of 3 was fast. As proposed
by zhao et al., fructose was formed from glucose through an
[10]
enediol intermediate catalyzed by GeCl4. Fructose then lost
one water molecular to form the fructofuranosyl oxocarbenium
[18]
ion. Due to the basicity and nucleophilicity of the halide ion
À
(
X ), the oxocarbenium ion lost HX to form the enolic species
through the nucleophile or the base pathway, followed by loss
of two water molecules to form HMF.
Our results also suggested that increasing the reaction tem-
1
3
13
perature would enhance the interaction of glucose and GeCl4,
Figure 9. C NMR spectra of 5 wt% d-glucose-2- C in [D
different conditions: a) Without GeCl at 258C; b) in the presence of
equivalent GeCl at 258C; c) in the presence of 1 equivalent GeCl
d) in the presence of 1 equivalent GeCl at 808C; e) in the presence of
equivalent GeCl at 1008C.
6
]DMSO under
4
and thus promote a ring-opening step of the intermediate 3
to form the straight-chain intermediate 4 through hydrogen
transfer, which could be enolated to form 5, followed by the
1
4
4
at 508C;
4
1
4
release of GeCl to give fructose
4
(
Scheme 1). At 808C, three new
signals at d=56.1, 110.9 and
24.3 ppm appeared, which
1
were attributed to C6, C4 and
C3, respectively, of HMF. A new
peak at d=80.1 ppm was also
observed at 808C and 1008C,
which was attributed to a by-
product formed during the reac-
tion.
Additional mechanistic investi-
gation was done using the “ger-
manium reagent” phenylfluor-
one (Figure 10). When the reac-
tion was run with glucose at
1
1
008C for 2 h in the presence of
0 mol% GeCl4 and 30 mol%
phenylfluorone, glucose conver-
sion and HMF yield were 33.3%
and 3.5%, respectively. Appa-
rently, phenylfluorone signifi-
cantly inhibited glucose conver-
4
Scheme 1. Proposed mechanism of HMF formation from glucose catalyzed by GeCl .
thermodynamically by introduction of auxiliary NR2 or OR
groups, the lone electron pairs of which could, in principle, be
capable of bonding to the empty p orbital of the Group 14 ele-
sion and HMF production. As phenylfluorone could form stable
IV
complexes with Ge , there would be much less catalytic
species available for the turnover of glucose.
[
21]
IV
ment. Therefore, we speculated that Ge interacted with the
two O atoms of 1,2-dihydroxy group of a-glucose to form a
3
. Conclusion
13
stable intermediate 3 (Scheme 1). When the C NMR spectrum
was recorded at 508C, the relative intensity of the peak at d=
In this study, a new catalytic system based on germanium(IV)
chloride has been established for the conversion of carbohy-
76.4 ppm decreased, and two new peaks at d=152.1 and
1
36
www.chemsuschem.org
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2011, 4, 131 – 138

Conversion of Carbohydrates by GeCl4
acetic acid aqueous solution with the volume ratio 15:85. The
À1
flow rate was set at 1.0 mLmin .
Determination of glucose
The ion chromatography system was from Dionex. The hard-
ware consisted of an ICS-2500 system equipped with a GP50
gradient pump, an ED50 A integrated amperometry detector, a
CarbonPac PA10 guard column (4 mmꢀ50 mm), a high capaci-
ty CarbonPac PA20 analytical column (3 mmꢀ150 mm), a
Figure 10. The structure of phenylfluorone.
2
5 mL sample loop. Samples were eluted with 30 mm NaOH at
À1
drates into HMF. This system was excellent for fructose and
moderate for other carbohydrates, such as glucose, cellobiose,
a rate of 0.5 mLmin .
13
sucrose, and cellulose, in terms of HMF yield. C NMR spectros-
copy suggested that there was a dedicated interaction be-
tween glucose and the catalyst, and that fructose was formed
during the reaction. Furthermore, our system is more attractive
than many other Lewis acid-catalyzed systems, as inorganic
Acknowledgements
Financial support provided by the National High-Tech Research
and Development Program of China (2007AA05Z403) and the
Knowledge Innovation Program of CAS (KGCXZ-YW-336) is
greatly acknowledged.
[
23]
germanium species are of low toxicity. Our system should
provide a new opportunity for the transformation of abundant
and inexpensive cellulosic biomass through a nonfermentative
process into biofuels and bio-based products.
Keywords: carbohydrates
· germanium · ionic liquids ·
oligosaccharides · sustainable chemistry
Experimental Section
[1] a) Y. C. Lin, G. W. Huber, Energy Environ. Sci. 2009, 2, 68–80; b) N. U.
Soriano, R. Venditti, D. S. Argyropoulos, Fuel 2009, 88, 560–565.
Materials
[2] a) A. J. Ragauskas, C. K. Williams, B. H. Davison, G. Britovsek, J. Cairney,
C. A. Eckert, W. J. Frederick, J. P. Hallett, D. J. Leak, C. L. Liotta, J. R. Mie-
lenz, R. Murphy, R. Templer, T. Tschaplinski, Science 2006, 311, 484–489;
b) R. H. He, P. Ye, B. C. English, J. A. Satrio, Bioresour. Technol. 2009, 100,
All of the ionic liquids except [BMIM]OAc were synthesized ac-
cording to the known procedures.[24] [BMIM]OAc was supplied
by Lanzhou Greenchem ILS, LICP, CAS (Lanzhou, China). Glu-
cose was purchased from ABCR GmbH & Co. (Karlsruhe, Ger-
many) and fructose was from Sanland-Chem International Inc.
5305; c) Y. Y. Wu, Z. H. Fu, D. L. Yin, Q. Xu, F. L. Liu, C. L. Lu, L. Q. Mao,
Green Chem. 2010, 12, 696–700.
[
3] B. Kamm, Angew. Chem. 2007, 119, 5146–5149; Angew. Chem. Int. Ed.
2
007, 46, 5056–5058.
(
Xiamen, China). Avicel PH-101 (Cat. No. 11365) was purchased
from Sigma (St. Louis, USA), and was dried under vacuum at
008C for 24 h before use. Acetonitrile (HPLC grade) was pur-
[4] U. Wongsiriwan, Y. Noda, C. S. Song, P. Prasassarakich, Y. Yeboah, Energy
Fuels 2010, 24, 3232–3258.
[
5] a) Y. Y. Gorbanev, S. K. Klitgaard, J. M. Woodley, C. H. Christensen, A.
Riisager, ChemSusChem 2009, 2, 672–675; b) M. Mascal, E. B. Nikitin,
Green Chem. 2010, 12, 370–373.
1
chased from Merck & Co. (Darmstadt, Germany). All of the
metal salts were supplied by J & K Co. Ltd. (Beijing, China). All
other chemicals were supplied by local suppliers and used
without further purification.
[6] a) T. M. Aida, K. Tajima, M. Watanabe, Y. Saito, K. Kuroda, T. Nonaka, H.
Hattori, R. L. Smith, K. Arai, J. Supercrit. Fluids 2007, 42, 110–119; b) A.
Amarasekara, L. T. D. Williams, C. C. Ebede, Carbohydr. Res. 2008, 343,
3
2
021–3024; c) M. Bicker, J. Hirth, H. Vogel, Green Chem. 2008, 10, 280–
84; d) X. H. Qi, M. Watanabe, T. M. Aida, R. L. Smith, Green Chem. 2009,
1
2
1, 1327–1331; e) T. S. Hansen, J. M. Woodley, A. Riisager, Carbohydr. Res
009, 344, 2568–2572; f) C. Sievers, I. Musin, T. Marzialetti, M. B. V.
HMF production
Olarte, P. K. Agrawal, C. W. Jones, ChemSusChem 2009, 2, 665–671.
7] a) B. L. Scallet, Gardner, J. Am. Chem. Soc. 1945, 67, 1934–1935; b) H. E.
van Dam, A. P. G. Kieboom, H. van Bekkum, Starch/Staerke 1986, 38, 95–
101; c) H. Ishida, K. Seri, J. Mol. Catal. A: Chem. 1996, 112, 163–165;
d) H. Ishida, K. Seri, Bull. Chem. Soc. Jpn. 2001, 74, 1145–1150; e) T. M.
Aida, Y. Saito, M. Watanabe, K. Tajima, T. Nonaka, H. Hattori, K. Arai, J.
Supercrit. Fluids 2007, 40, 381–388.
In a typical run, glucose (100 mg) and GeCl (10 mol% based
4
[
on glucose) were added into [BMIM]Cl (2.0 g). The mixture was
heated at a specified temperature for the desired time under
atmospheric pressure with a magnetic stirrer. Samples were
withdrawn, diluted with water, centrifuged at 10000 rpm for
[
25]
5
min and analyzed by HPLC.
[
8] Z. Srokol, A. G. Bouche, A. van Estrik, R. C. J. Strik, T. Maschmeyer, J. A.
Peters, Carbohydr. Res. 2004, 339, 1717–1726.
[
9] B. H. Shanks, ACS Chem. Biol. 2007, 2, 533–535.
Determination of HMF
[10] H. B. Zhao, J. E. Holladay, H. Brown, Z. C. Zhang, Science 2007, 316,
597–1600.
1
HPLC analysis of HMF was done on a Dionex system (Dionex
Co., CA, USA) equipped with Dionex P680 four-unit pump and
PDA-100 photodiode array detector. The samples were separat-
ed using a reversed-phase C18 column (200ꢀ4.6 mm) at
[
11] S. Hu, Z. Zhang, J. Song, Y. Zhou, B. Han, Green Chem. 2009, 11, 1746–
1749.
[
12] T. Stahlberg, M. G Sorensen, A. Riisager, Green Chem. 2010, 12, 321–
325.
[
[
13] J. Y. G. Chan, Y. Zhang, ChemSusChem 2009, 2, 731–734.
14] Y. Roman-Leshkov, J. N. Chheda, J. A. Dumesic, Science 2006, 312, 1933–
1937.
280 nm. The column temperature was maintained at 308C. The
optimized mobile phase consisted of acetonitrile and 0.1 wt%
ChemSusChem 2011, 4, 131 – 138
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
137

H. Xie, Z. K. Zhao et al.
[
[
15] O. Johnson, Chem. Rev. 1952, 51, 431–469.
16] Y. Deng, S. Morrissey, N. Gathergood, A. M. Delort, P. Husson, M. F. C.
Gomes, ChemSusChem 2010, 3, 377–385.
[22] C. Yao, Y. Shin, L. Wang, C. F. Windisch, W. D. Samuels, B. W. Arey, C.
Wang, W. M. Risen, G. J. Exarhos, J. Phys. Chem. C 2007, 111, 15141–
15145.
[
17] G. Ebner, S. Schiehser, A. Potthast, T. Rosenau, Tetrahedron Lett. 2008,
[23] M. Kanisawa, H. Schroeder, Cancer Res. 1967, 27, 1192–1195.
[24] a) A.K. Burrell, R. E. Del Sesto, S. N. Baker, T. M. McCleskey, G. A. Baker,
Green Chem. 2007, 9, 449–454; b) P. Bonhꢄte, A. P. Dias, N. Papageor-
giou, K. Kalyanasundaram, M. Grꢅtzel, Inorg. Chem. 1996, 35, 1168–
49, 7322–7324.
[
18] J. B. Binder, R. T. Raines, J. Am. Chem. Soc. 2009, 131, 1979–1985.
19] S. Lima, P. Neves, M. M. Antunes, M. Pillinger, N. Ignatyev, A. A. Valente,
Appl . Catal. A: Gen. 2009, 363, 93–99.
[
1178.
[
25] Z. H. Zhang, Z. K. Zhao, Bioresour. Technol. 2010, 101, 1111–1114.
[
[
20] K. Jurkschat, N. Pieper, S. Seemeyer, M. Schꢃrmann, M. Biesemans, I. Ver-
bruggen, R. Willem, Organometallics 2001, 20, 868–880.
21] D. Kost, V. Kingston, B. Gostevskii, A. Ellern, D. Stalke, B. Walfort, I. Ka-
likhman, Organometallics 2002, 21, 2293–2305.
Received: August 28, 2010
Revised: October 15, 2010
Published online on December 23, 2010
1
38
www.chemsuschem.org
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2011, 4, 131 – 138