Facile catalytic dehydration of fructose to 5-hydroxymethylfurfural by Niobium pentachloride

Article abstract of DOI:10.1016/j.tetlet.2012.04.045

Different metal chlorides in ionic liquids were examined as catalysts for fructose dehydration to 5-hydroxymethylfurfural (5-HMF). Niobium pentachloride (NbCl₅) exhibited the highest 5-HMF yield. Reaction of 1.0 mmol fructose in 10.0 mmol 1-butyl-3-methylimidozolium chloride ([bmim]Cl) with 0.20 mmol NbCl₅ afforded 79% of 5-HMF at 80 °C after 30 min. No degradation products were formed under these conditions. FTIR analysis indicates the moderate Lewis acidity of NbCl₅, which was found suitable for the selective formation of 5-HMF. Other catalysts with higher Lewis acidities promoted 5-HMF side product formation which ultimately resulted in lower yields.

Full text of DOI:10.1016/j.tetlet.2012.04.045

Tetrahedron Letters 53 (2012) 3149–3155
Contents lists available at SciVerse ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Facile catalytic dehydration of fructose to 5-hydroxymethylfurfural
by Niobium pentachloride
⇑
Neha Mittal , Grace M. Nisola , Wook-Jin Chung
Energy and Environment Fusion Technology Center, Department of Energy Science and Technology, Myongji University, San 38-2, Nam-dong, Cheoin-gu, Yongin-City, Gyeonggi-Do
449-728, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 14 January 2012
Revised 6 April 2012
Accepted 11 April 2012
Available online 17 April 2012
Different metal chlorides in ionic liquids were examined as catalysts for fructose dehydration to 5-
hydroxymethylfurfural (5-HMF). Niobium pentachloride (NbCl
Reaction of 1.0 mmol fructose in 10.0 mmol 1-butyl-3-methylimidozolium chloride ([bmim]Cl) with
.20 mmol NbCl afforded 79% of 5-HMF at 80 °C after 30 min. No degradation products were formed
under these conditions. FTIR analysis indicates the moderate Lewis acidity of NbCl , which was found
5
) exhibited the highest 5-HMF yield.
0
5
5
suitable for the selective formation of 5-HMF. Other catalysts with higher Lewis acidities promoted 5-
HMF side product formation which ultimately resulted in lower yields.
Ó 2012 Elsevier Ltd. All rights reserved.
Keywords:
5
-Hydroxymethylfurfural
Fructose
Niobium pentachloride
1
-Butyl-3-methylimidozolium chloride
Introduction
reaction time.9 But the high acidities of mineral acids also create
1
5
problems as they promote 5-HMF decomposition. Alternatively,
mild Lewis acid catalysts could minimize the problem and there-
With a bi-functional moiety, 5-hydroxymethylfurfural (5-HMF)
is a recognized novel base material for the synthesis of various di-
substituted furanic compounds and an important precursor of
transportation fuels, pharmaceuticals, and other petroleum-de-
1
6
fore improve the reaction selectivity for 5-HMF. Particularly, me-
tal chlorides are environmentally benign Lewis acid catalysts
which can be used instead of mineral acids. High conversions of
1
–5
rived chemicals.
The synthesis of 5-HMF involves acid-catalyzed
mono- and polysaccharides to 5-HMF using metal chlorides have
6
16–19
triple dehydration of hexoses (Fig. 1), with initial works being per-
been reported.
So far, certain metal chlorides have success-
formed in aqueous systems.7 But the inevitable conversion of 5-
fully demonstrated high 5-HMF yield but the potential of others
has not yet been fully elucidated.
HMF to levulinic and formic acids as degradation products in the
presence of H
strategies have been developed to minimize the decomposition
product formations wherein careful selection of catalysts and
solvent systems as well as identification of optimal operating
conditions have been proven to be equally important.9
8
2
O decreases the selectivity of the reaction. Several
For solvent selection, 5-HMF production from hexoses is typi-
cally performed under sub- or supercritical conditions using polar
1
2,20–24
aprotic solvents.
However, their high water solubility, high
boiling points, and necessity of high reaction temperatures could
interfere with reaction selectivity, promote side product formation,
2
0,22,25
Organic and mineral acids were conventionally employed as
catalysts in which 5-HMF yields of 40–80% have been
and create difficulty in 5-HMF recovery.
Alternatively, the
high solvency power of ionic liquids (ILs) indicates their suitability
as solvents for metal chlorides even at low reaction temperatures.
ILs have high thermal and chemical stabilities while their negligi-
ble vapor pressure makes them environment-friendly as they do
1
0–13
achieved.
While yields were satisfactory, product separation
was difficult to perform. To resolve this drawback, solid acids were
alternatively used as heterogeneous catalysts. But homogenous
catalyst reactions (i.e., mineral acids) remained superior over het-
erogeneous systems in terms of 5-HMF yield.8 Both types of cat-
alysts (i.e., liquid and solid acids) are reacted at high reaction
temperatures to promote a faster 5-HMF production and to avoid
the side products which are eventually generated at prolonged
2
6,27
not evolve toxic gases.
Moreover, 5-HMF recovery and catalyst
,14
⇑
Corresponding author. Tel.: +82 31 336 8471; fax: +82 31 337 2902.
E-mail address: wjc0828@gmail.com (W.-J. Chung).
These authors contributed equally to this work.
Figure 1. Dehydration of D-fructose to 5-HMF.
0
040-4039/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tetlet.2012.04.045

3
150
N. Mittal et al. / Tetrahedron Letters 53 (2012) 3149–3155
recyclability could be easier in IL system.28 Fructose dehydration
has been performed in ILs with excellent yields and selectivity.
r
bond component of C„N. In FTIR, this occurrence is indicated
2
9
by the blue shift in C„N band or formation of new peak at higher
wave numbers. It was pointed out previously that the higher the
blue shift, the higher the Lewis acidity strength of the metal chlo-
ride. Other works also indicated that the higher peak intensities
of the blue-shifted bands indicate a higher number of Lewis acidic
sites.
C„N group in ACN with a Lewis acid could result to charge density
repulsion of the bond component of the methyl (CH ) group in
ACN, resulting to C–H bond weakening. In FTIR, this is typically
indicated by the red shift of CH
in IR spectra were monitored between 2500 and 2000 cm
In Figure 3, pure ACN features two characteristic peaks. The
stronger peak at 2252 cm
stretching while the weaker band at 2292 cm is due to the CH
3
3
Particularly, imidazolium based ILs like 1-butyl-3-methylimidazo-
lium chloride ([bmim]Cl) were found to effectively dehydrate fruc-
tose to 5-HMF with satisfactory yield of 63% after 50 min reaction
3
2
3
0
at 120 °C.
3
3,34
Several studies have already investigated certain IL/metal chlo-
ride systems for this application but the basis for their selection re-
mains vaguely explored.31 Thus in this work, several metal
chlorides were screened for their catalytic abilities to dehydrate
fructose to 5-HMF. The appropriate solvent was selected from
two imidazolium-based ILs namely [bmim]Cl and 1-ethyl-3-meth-
ylimidazolium chloride ([emim]Cl). To elucidate the reaction effi-
ciency of each IL/metal chloride pair, Lewis acidities of metal
chlorides in IL were examined using Fourier transform infrared
Moreover, the dipole-dipole interaction between the
r
3
3
4
3
frequency band. The changes
ꢁ1
.
ꢁ
1
is attributable to the C„N bond
ꢁ
1
3
3
2
32,35
spectroscopy (FTIR). The effects of different reaction parameters
were tested to determine the most efficient and mildest condition
for 5-HMF production.
bending and C–C stretching.
with ACN was similar except for the slight red shift from 2252 to
2247 cm . Likewise, no new band formation at a higher wave
number was observed from SnCl O and only slight blue shift
was observed at 2267 cm for LiCl. The absence or slight blue shift
The spectrum of [bmim]Cl mixed
ꢁ
1
2
ꢀ2H
2
ꢁ1
Results and discussion
of C„N stretching indicates that pure [bmim]Cl, [bmim]Cl/LiCl,
and [bmim]Cl/SnCl
character. While evidence of 5-HMF production was observed from
CrCl (Fig. 2), no new peak was observed in Figure 3. This is due to
the difficulty encountered in dissolving the catalyst in ACN/
2
ꢀ2H
2
O reaction systems have no Lewis acid
Fructose dehydration by various metal chlorides in [bmim]Cl
Seven metal chlorides in [bmim]Cl were initially screened for
the production of 5-HMF at 80 °C as illustrated in Figure 2.
Among the tested catalysts, five metal chlorides exhibited activi-
ties for fructose dehydration. The yields of 5-HMF can be arranged
3
16
[
bmim]Cl during FTIR analysis.
On the other hand, three peaks were evident from ACN with
[
bmim]Cl/MCl
x
ꢀnH
2
O (M = Nb, Ru, Cu, Fe; x = 2, 3, 5; n = 0, 2). Slight
in the following order: NbCl
Highest 5-HMF yield of 79% was attained from NbCl
shortest reaction time of 30 min. No 5-HMF was produced in
bmim]Cl with LiCl and SnCl O even after 70 min reaction.
ꢀ2H
As an acid catalyzed reaction, the relationship between the
5
> RuCl
3
> CuCl
2
ꢀ2H
2
O > FeCl
3 3
> CrCl .
ꢁ
1
ꢁ1
CH
for CuCl
served for the CH
chlorides also showed the original band of C„N stretching at
3
red shifts were observed at 2291 cm for FeCl
3
, at 2290 cm
. No red shift was ob-
band in RuCl (2293 cm ). These four metal
3
5
within the
ꢁ1
2 2
ꢀ2H O, and at 2289 cm for NbCl
5
ꢁ1
[
2
2
3
ꢁ
1
2
251 cm . For the same group of metal chlorides (MCl
x 2
ꢀnH O,
acidity of the reaction systems and their catalytic activities for 5-
HMF production was examined. Earlier works used a weak Lewis
base, acetonitrile (ACN), as a probe molecule to observe the Lewis
M = Nb, Ru, Cu, Fe; x = 2, 3, 5; n = 0, 2) new bands were formed at
higher wave numbers: at 2316 cm
ꢁ
ꢁ1
1
for NbCl
5
and FeCl , at
3
ꢁ
1
3
2
2
328 cm for RuCl
appearance at 2360 cm for CuCl
frequencies of CH (red shift) and C„N bonds (blue shift) in the four
3
, and at 2325 cm with an additional peak
acidities in various IL-metal chloride pairs via FTIR analysis. It is
known that the interaction between the lone pair of nitrogen in
ACN and the Lewis acid sites of the metal chlorides in [bmim]Cl
ꢁ1
2
ꢀ2H
2
O. The changes in vibrational
3
3
3
metal chlorides are indicative of Lewis acid interaction with ACN.
Considering that the same group of metal chlorides exhibited Lewis
acidities according to the FTIR results, it can be surmised that the
would affect the vibrational bands of C„N stretching.
The
nitrogen orbitals would undergo re-hybridization, enhancing the
catalytic fructose dehydration activities of NbCl
CuCl O in [bmim]Cl could be associated to this property.
ꢀ2H
According to FTIR data, RuCl and CuCl O exhibited the
ꢀ2H
widest blue shifts and therefore, must have the strongest Lewis
5 3 3
, RuCl , FeCl , and
2
2
3
2
2
3
2
acidities. However, both catalysts showed inferior 5-HMF yields
than NbCl . Earlier works have pointed out the promotion of 5-
HMF by product formation by catalysts with strong Lewis acidi-
5
3
6–38
ties.
This is consistent with the remarkable formation of
and
Thus, the lower reaction
and [bmim]Cl/CuCl O than in
ꢀ2H
ultimately resulted to lower 5-HMF yields.
The FTIR results suggest that NbCl has moderate Lewis acidity rel-
ative to RuCl and CuCl O. On the other hand, the blue shift of
dark-brown precipitates known as humins in RuCl
3
9
,15
CuCl
selectivities in [bmim]Cl/RuCl
bmim]Cl/NbCl
2
2
ꢀ2H O in [bmim]Cl systems.
3
2
2
9
,15,36,37
[
5
5
3
2
ꢀ2H
2
ꢁ1
C„N 2316 cm has the greatest intensity among all the tested
metal chlorides. This indicates the high number of Lewis acidic
3,34
sites in NbCl
Thus herein, NbCl
acid catalyst among the tested metal chlorides for the dehydration
of fructose. The moderate Lewis acidity of NbCl resulted to a reac-
5
and possibly explains its high 5-HMF yield.3
5
is determined as the most suitable Lewis
5
tion with high selectivity toward 5-HMF; the major product was 5-
HMF and interestingly, no degradation products (i.e., levulinic acid
8
or formic acid) were detected. The dehydration of fructose with
5
NbCl in [bmim]Cl was facile and rapid with remarkably high 5-
Figure 2. 5-HMF yield of various metal chlorides using 1.0 mmol fructose,
0
.2 mmol catalyst, and 10.0 mmol [bmim]Cl at 80 °C.
HMF yield of 79% and fructose conversion of P95% at 80 °C.

N. Mittal et al. / Tetrahedron Letters 53 (2012) 3149–3155
3151
Figure 3. FTIR spectra of various metal chlorides in ionic liquid using ACN as IR probe at room temperature (1:5 mass ratio of ACN/sample; 0.5:1 mol ratio of IL/catalyst as
sample).
5
Effect of IL, reaction time, and NbCl dosage on the 5-HMF yield
The production of 5-HMF was determined at different reaction
times and NbCl catalyst dosages using ILs [bmim]Cl or [emim]Cl,
5
reacted at 80 °C. The obtained 5-HMF yields are shown in Figure 4,
wherein the highest conversions were achieved after 30- and
6
0 min in [bmim]Cl/NbCl
5
and [emim]Cl/NbCl
5
,
respectively,
regardless of the catalyst dosage.
In Figure 3, the spectrum of NbCl
5
in [emim]Cl (with ACN)
showed the appearance of the new peak at the same region with
that of NbCl in [bmim]Cl. But the peak intensity of NbCl in
emim]Cl is lower than that of NbCl
in [bmim]Cl.34 Thus, the lower
in [emim]Cl can be associated to its lower Le-
in [bmim]Cl. Moreover, previous work
5
5
[
5
5
-HMF yield of NbCl
5
wis acidic sites than NbCl
indicated the slightly lower pK
5
3
9
a
value of [bmim]Cl than [emim]Cl
and the acidity differences between ILs could have also influenced
the production of 5-HMF.
The 5-HMF yields declined after achieving the maximum values
in both IL systems, concomitant with the observed humin forma-
9
,15
tion.
The self-polymerization of 5-HMF or the cross-polymeri-
zation between 5-HMF and fructose, are promoted at a longer
reaction period, which resulted in the eventual decrease of 5-
9
,15
HMF yields.
At optimum reaction periods, both ILs (i.e., without NbCl
added), were unable to produce 5-HMF at 80 °C. But addition of
NbCl effectively promoted 5-HMF formation, which gradually im-
proved when the catalyst dosage was increased to some extent.
Particularly, when NbCl loading was increased from 0.05 to
.20 mmol (for 1 mmol fructose), the maximum 5-HMF yields in-
5
5
5
0
creased from 67% to 79% in [bmim]Cl (at 30 min) and from 59%
to 70% in [emim]Cl (at 60 min). However, 5-HMF yields declined
with further increase in catalyst dosage (>0.20 mmol). It is possible
that the high catalyst dosage resulted to high acidity in the system
which might have accelerated humin formation.9
5
Figure 4. Effect of reaction time and NbCl dosage on 5-HMF yield using 1.0 mmol
fructose and 10.0 mmol of (a) [bmim]Cl and (b) [emim]Cl at 80 °C.
,15,36–38
From these results, [bmim]Cl exhibited better 5-HMF yield than
emim]Cl. Regardless of the IL used, the optimum NbCl dosage of
5
0.20 mmol for 1 mmol fructose was determined and subsequently
used for the rest of the experiments, if not otherwise stated.
[

3
152
N. Mittal et al. / Tetrahedron Letters 53 (2012) 3149–3155
Effect of reaction temperature on the 5-HMF yield
when NbCl
5
attacks the hydroxyl (OH) at the hemiketal carbon.6
From here, the reaction may proceed in two possible directions.
First, the ‘nucleophile’ pathway reported by Binder and Raines con-
siders the formation of 2-deoxy-2-halo intermediate from the
The 5-HMF yields at different reaction temperatures from 50 °C
to 180 °C were also investigated using 0.2:1 mole ratio of NbCl
5
4
4
with respect to fructose, at optimum reaction periods. In Table 1,
it is revealed that the reaction temperature is a critical parameter
in obtaining satisfactory yields of 5-HMF. For example, low 5-HMF
nucleophilic attack on the oxocarbenium ion. The IL [bmim]Cl
ꢁ
is a good source of Cl anions which could participate as nucleo-
4
5
philes in the reaction. The subsequent deprotonation at C
1
leads
5
yields were obtained from [bmim]Cl/NbCl (24%) at 50 °C and
to the formation of an enol which could rearrange into the corre-
4
4,45
[
emim]Cl/NbCl
5
(59%) at 55 °C. Increasing the temperature to
sponding aldehyde.
The other plausible route regards the
8
0 °C showed the maximum 5-HMF yields of 79% and 67% from
interaction of the primary OH with the oxocarbenium ion which
leads to the formation of an epoxide intermediate, followed by
[
bmim]Cl/NbCl
5
(30 min) and [emim]Cl (60 min), respectively. Fur-
ther increase in temperatures (100 °C) resulted to drastic declines
in 5-HMF yields which may be attributed to the increased inci-
dence of humin formation. Nonetheless, no formation of levulinic
acid or formic acid as side products was observed even at high
its rearrangement via NbCl
5
-promoted C–O bond cleavage.46 The
2
subsequent release of two moles of H O in both pathways affords
5-HMF.
temperatures. Thus, reaction at 80 °C was found optimal for both
Potential use of [bmim]Cl/NbCl for other hexoses
5
IL systems. Compared to previous works reported,3
0,40
the present
system features a facile and more energy-efficient method of pro-
ducing 5-HMF from fructose dehydration.
As for the substrate scope of [bmim]Cl/NbCl5 system, glucose
dehydration was also performed using the optimum conditions
for fructose. 5-HMF yield of 42% was obtained after 50 min reac-
tion, at 80 °C. This result clearly indicates that the [bmim]Cl/NbCl5
system is worthy of further investigation on its applicability for the
dehydration of other sugars aside from fructose.
Comparison among efficiency of group V metal chlorides on the
5
-HMF yield
Using [bmim]Cl as the appropriate solvent, the catalytic poten-
tial of other metal chlorides belonging to the same group with Nio-
bium was also explored. The 5-HMF yields of other naturally
Conclusion
occurring group V metal chlorides, vanadium trichloride (VCl
3
),
A new reaction system for fructose dehydration to 5-HMF has
and tantalum pentachloride (TaCl ), are shown in Figure 5a. Com-
5
been successfully demonstrated. The system, [bmim]Cl/NbCl
volves a rapid and energy-efficient reaction wherein the maximum
-HMF yield was achieved only after 30 min of reaction, at 80 °C.
Likewise, the other group V metal chlorides like VCl and TaCl
showed promising yields but the minor formation of humins re-
sulted in their slightly lower yields than NbCl . Reaction tempera-
ture, NbCl dosage, and reaction time were found critical for the
minimization of humin formation and for attaining high yields of
-HMF. From these results, it is considered that NbCl in [bmim]Cl
5
, in-
pared to other metal chlorides tested (see Fructose dehydration by
various metal chlorides in [bmim]Cl), group V catalysts showed the
highest 5-HMF yields. But within the group, highest 5-HMF yield
5
3
5
5
was still attained by NbCl . Group V metal chlorides have similar
ꢁ1
FTIR spectra as shown in Figure 5b. The band at 2316 cm indi-
cates their comparable Lewis acidities but the peak intensities de-
5
5
creased in the order of NbCl
with the trend of their catalytic activities. Moreover, minor forma-
tions of humins were evident in VCl and TaCl , which could also
explain their lower 5-HMF yields than NbCl
5 5 3
> TaCl > VCl , which is consistent
5
5
3
5
is a cheap and competent reaction system for 5-HMF production.
5
.
Proposed fructose dehydration mechanism in [bmim]Cl/NbCl
5
Experimental
Niobium (V) chloride is a renowned Lewis acid catalyst which
has been used in various organic transformations such as tetrahy-
dropyranylation, allylation, Diels–Alder reaction, ring-opening of
epoxides, Mukaiyama aldol reaction, Biginelli reaction, dealkyla-
tion of alkyl aryl ethers, and C–H insertion reaction.4 But so far,
no study has been reported regarding its use in fructose dehydra-
tion for 5-HMF production.
Materials
Ionic liquids [bmim]Cl (98%) and [emim]Cl (97%), metal chlo-
rides CrCl (99%, anhydrous), and RuCl (35–40% Ru, hydrate) as
3 3
well as 5-HMF (98%, reagent grade) were purchased from Acros
Organics (USA). Acetonitrile (anhydrous, 99.8%), fructose (>99%),
1
NbCl
5
(99.9%), TaCl
5
(99.9%), and VCl
3
(97%) were purchased from
.2H O (>95%) were
As the first report on [bmim]Cl/NbCl
5
system for fructose dehy-
Sigma–Aldrich (USA). LiCl (98.2%), and SnCl
2
2
dration, tentative mechanisms are proposed in Figure 6. Figure 6A
procured from Samchun Chemical Co., Ltd (South Korea), and Dae-
jung Chemicals & Metals Co., Ltd (South Korea), respectively. Other
4
2
5
shows the possible interaction between NbCl and [bmim]Cl.
Niobium is highly electrophilic, oxophilic, and capable of forming
hexacoordinates with halides, oxygen, and other nucleophiles.
metal chlorides like CuCl
Pure Chemicals Co., Ltd (South Korea) whereas FeCl
2
ꢀ2H
2
O (>95%) was supplied by Duksan
(97%) was
4
3
3
In Figure 6B, fructofuranosyl oxocarbenium ion is initially formed
from Showa Chemical Co., Ltd (Japan). All chemicals were directly
used without further purification. Ionic liquids were dried at 60–
7
0 °C before reactions in a vacuum oven. The reactions were per-
Table 1
formed in glass vials heated in a temperature-controlled oil bath
with magnetic stir bars.
Effect of reaction temperature on 5-HMF yield using 1.0 mmol fructose, 0.20 mmol
NbCl
5
, and 10.0 mmol ionic liquid
o
Temperature ( C)
5-HMF yield (%)
bmim]Cl, 30 min [emim]Cl, 60 min
FTIR characterization
[
o
The FTIR analysis was performed using Varian 2000 (Scimitar
Series) FTIR spectrophotometer at room temperature. Prior to anal-
ysis, samples were prepared by mixing ACN with IL or IL/metal
chloride by 1:5 weight ratio. The prepared IL/metal chloride sam-
ples contained 67 mol % of metal chloride according to the method
5
8
00
50
80
0
0
24
79
60
47
15
59 (55 C)
70
57
20
1
1
1
1

N. Mittal et al. / Tetrahedron Letters 53 (2012) 3149–3155
3153
Figure 5. (a) 5-HMF yield of group V metal chlorides using 1.0 mmol fructose, 0.2 mmol catalyst, 10.0 mmol [bmim]Cl at 80 °C in 30 min. (b) FTIR spectra of group V metal
chlorides in [bmim]Cl using ACN as IR probe at room temperature (1:5 mass ratio of ACN/sample; 0.5:1 mol ratio of IL/catalyst as sample).
of Yang and Kou.32 Samples were then spread into liquid films on
CaF
windows (25 ꢂ 4 mm, PIKE Technologies, USA) in a dry,
anaerobic environment.
the IL was cooled to room temperature. Fructose (1.0 mmol) and
catalyst (amount as indicated in the discussion) were added to
the molten IL. The samples were then reacted at a constant tem-
perature (as indicated in the discussion) for a given period of time.
Aliquots of reaction mixture were collected at different time inter-
vals and diluted with deionized water (5 mL). The solution was
shaken well and after the settling of insoluble precipitates (hu-
2
5
-HMF analysis
All reactions were monitored by Waters HPLC system equipped
with binary 1525 pumps, a 2707 auto sampler, and 2414 refractive
index detector, maintained at 40 °C. Analytes were separated in
Bio-Rad Aminex HPX-87H (300 ꢂ 7.8 mm) ion-exclusion column,
mins), the upper liquid was filtered through 0.2 lm Nylon syringe
filters. The filtered sample was analyzed using HPLC.
maintained at 60 °C using 5 mM H
2
SO
4
as mobile phase with
5-HMF quantification method
0
.6 mL/min flow rate. The HPLC system was controlled and pro-
cessed by Empower software. Standard 5-HMF, levulinic, and for-
mic acid solutions were prepared for calibration. Each product
sample was diluted with a known volume of ultra pure deionized
water before analysis to prevent the overloading of the column.
All experiments were done in triplicates, and the average values
were reported; standard deviations of triplicates were <2.0%.
An aliquot from reaction mixture was diluted with 5 mL deion-
ized water. After shaking the samples and settling of precipitates,
the upper clear supernatant was collected for analysis. The 5-
HMF concentration was measured using HPLC (Water Series) with
RI detector. The 5-HMF yield was calculated using Eq. 1 wherein
M
HMF (mg) is the total mass of 5-HMF formed in the reaction mix-
ture, M (mg) is the mass of fructose as substrate while MW (g/mol)
is the molecular weight of fructose or 5-HMF. On the other hand,
HMF was estimated using (Eq. (2)) where CHMF is the 5-HMF con-
F
General procedure for fructose dehydration
M
Approximately 10 mmol of IL was placed in a 10 mL thick
walled glass vial. The vial was capped and heated at 150 °C to melt
the ionic liquid with continuous stirring for 20 min. After heating,
centration (mg/mL) from HPLC analysis, V is the volume of water
added (mL) while WRM and WAL are the masses of total reaction
mixture and aliquot, respectively.

3
154
N. Mittal et al. / Tetrahedron Letters 53 (2012) 3149–3155
5 5 5
Figure 6. Tentative fructose dehydration mechanism in [bmim]Cl/NbCl system. (A) NbCl complexation with [bmim]Cl, (B) fructose dehydration in [bmim]Cl/NbCl .
ꢀM
ꢁꢀ
MW
F
MWHMF
ꢁ
8. Qi, X.; Watanabe, M.; Aida, T. M.; Smith, R. L., Jr. Green Chem. 2008, 10, 799.
Tong, X.; Ma, Y.; Li, Y. Appl. Catal., A: Gen. 2010, 385, 1.
HMF
5
-HMF yield ð%Þ ¼
ꢂ 100%
ð1Þ
ð2Þ
9.
M
F
1
1
0. Leshkov, Y. R.; Chheda, J. N.; Dumesic, J. A. Science 2006, 312, 1933.
1. Leshkov, Y. R.; Barrett, C. J.; Liu, Z. Y.; Dumesic, J. A. Nature 2007, 447, 982.
ꢀ
ꢁ
W
W
RM
12. Bicker, M.; Kaiser, D.; Ott, L.; Vogel, H. J. Supercrit. Fluids 2005, 36, 118.
M
HMF ¼ CHMF ꢂ V
1
1
1
1
1
3. Chheda, J. N.; Leshkov, Y. R.; Dumesic, J. A. Green Chem. 2007, 9, 342.
4. Qi, X.; Watanabe, M.; Aida, T. M.; Smith, R. L., Jr. Catal. Commun. 2009, 10, 1771.
5. Corma, A.; Iborra, S.; Velty, A. Chem. Rev. 2007, 107, 2411.
6. Zhao, H.; Holladay, J. E.; Brown, H.; Zhang, C. Science 2007, 316, 1597.
7. Kim, B.; Jeong, J.; Lee, D.; Kim, S.; Yoon, H.; Lee, Y.; Cho, J. K. Green Chem. 2011,
AL
Acknowledgments
1
3, 1503.
1
1
2
8. Hu, S.; Zhang, Z.; Song, J.; Zhou, Y.; Han, B. Green Chem. 2009, 11, 1746.
9. Yong, G.; Zhang, Y.; Ying, J. Y. Angew. Chem., Int. Ed. 2008, 47, 9345.
0. Chheda, J. N.; Dumesic, J. A. Catal. Today 2007, 123, 59.
This work is the outcome of the Priority Research Centers Pro-
gram through the National Research Foundation of Korea (NRF)
funded by the Ministry of Education, Science and Technology
21. Kuster, B. F. M. Carbohydr. Res. 1977, 54, 177.
2
2
2
2. Matras, C. L.; Moreau, C. Catal. Commun. 2003, 4, 517.
3. Szmant, H. H.; Chundury, D. D. J. Chem. Technol. Biotechnol. 1981, 31, 135.
4. Asghari, F. S.; Yoshida, H. Carbohydr. Res. 2006, 341, 2379.
(
2011-0022968). The authors would like to thank Professor Sangho
Koo of the Department of Chemistry at Myongji University, South
Korea for his suggestions and comments.
25. Kim, D. W.; Song, C. E.; Chi, D. Y. J. Org. Chem. 2003, 68, 4281.
26. Brennecke, J. F.; Maginn, E. J. AIChE J. 2001, 47, 2384.
27. Weingartner, H. Angew. Chem., Int. Ed. 2008, 47, 654.
28. Ray, D.; Mittal, N.; Chung, W.-J. Carbohydr. Res. 2011, 346, 2145.
29. Zhao, D.; Wu, M.; Kou, Y.; Min, E. Catal. Today 2002, 74, 157.
References and notes
1
2
3
.
.
.
Chheda, J. N.; Huber, G. W.; Dumesic, J. A. Angew. Chem., Int. Ed. 2007, 46, 7164.
Gupta, P.; Singh, S. K.; Pathak, A.; Kundu, B. Tetrahedron 2002, 58, 10469.
Carlini, C.; Patrono, P.; Galletti, A. M. R.; Sbrana, G.; Zima, V. Appl. Catal., A: Gen.
30. Cao, Q.; Guo, X.; Yao, S.; Guan, J.; Wang, X.; Mu, X.; Zhang, D. Carbohydr. Res.
2011, 346, 956.
31. Guan, J.; Cao, Q.; Guo, X.; Mu, X. Comput. Theor. Chem. 2011, 963, 453.
32. Yang, Y.; Kou, Y. Chem. Commun. 2004, 226.
33. Platero, E. E.; Mentruit, M. P.; Morterra, C. Langmuir 1999, 15, 5079.
34. Cao, F.; Tian, L.; Fang, D.; Ying, W.; Wang, J.-A. Catal. Commun. 2009, 10, 1310.
35. Elabd, Y. A.; Sloan, J. M.; Barbari, T. A. Polymer 2000, 41, 2203.
2
005, 289, 197.
4
5
6
7
.
.
.
.
Casanova, O.; Iborra, S.; Corma, A. J. Catal. 2009, 265, 109.
Amarasekara, A. S.; Green, D.; McMillan, E. Catal. Commun. 2008, 9, 286.
Antal, M. J.; Mok, W. S. L.; Richards, G. N. Carbohydr. Res. 1990, 199, 91.
Asghari, F. S.; Yoshida, H. Ind. Eng. Chem. Res. 2006, 45, 2163.

N. Mittal et al. / Tetrahedron Letters 53 (2012) 3149–3155
3155
3
3
6. Sievers, C.; Valenzuela-Olarte, M. B.; Marzialetti, T.; Musin, I.; Agrawal, P. K.;
Jones, C. W. Ind. Eng. Chem. Res. 2009, 48, 1277.
7. Sievers, C.; Musin, I.; Marzialetti, T.; Valenzuela-Olarte, M. B.; Agrawal, P. K.;
Jones, C. W. ChemSusChem 2009, 2, 665.
Matos, R. A. F. Synlett 2003, 8, 1189; (i) Sudo, Y.; Arai, S.; Nishida, A. Eur. J. Org.
Chem. 2006, 3, 752; (j) Hernández, H.; Bernes, S.; Quintero, L.; Sansinenea, E.;
Ortiz, A. Tetrahedron Lett. 2006, 47, 1153; (k) Yadav, J. S.; Reddy, B. V. S.;
Eeshwaraiah, B.; Reddy, P. N. Tetrahedron 2005, 61, 875.
3
3
4
4
8. Takagaki, A.; Ohara, M.; Nishimura, S.; Ebitani, K. Chem. Commun. 2009, 6276.
9. Chu, Y.; Deng, H.; Cheng, J. P. J. Org. Chem. 2007, 72, 7790.
0. Tong, X.; Ma, Y.; Li, Y. Carbohydr. Res. 2010, 345, 1698.
42. Alves, M. B.; Santos, V. O., Jr.; Soares, V. C. D.; Suarez, P. A. Z.; Rubim, J. C. J.
Raman Spectrosc. 2008, 39, 1388.
43. Hubert-Pfalzgraf, L. G. Niobium
& Tantalum Inorganic & Coordination
1. (a) Nagaiah, K.; Reddy, B. V. S.; Sreenu, D.; Narsaiah, A. V. ARKIVOC 2005, iii,
Chemistry. In Encyclopedia of Inorganic Chemistry; John Wiley & Sons Ltd, 2006.
44. (a) Binder, J. B.; Raines, R. T. J. Am. Chem. Soc. 2009, 131, 1979; (b) Defaye, J.;
Gadelle, A. Carbohydr. Res. 1985, 136, 53; (c) Stahlberg, T.; Sorensen, M. G.;
Riisager, A. Green Chem. 2010, 12, 321.
45. Boovanahalli, S. K.; Kim, D. W.; Chi, D. Y. J. Org. Chem. 2004, 69, 3340.
46. Marouka, K.; Murase, N.; Bureau, R.; Ooi, T.; Yamamoto, H. Tetrahedron 1994,
50, 3663.
1
92; (b) Andrade, C. K. Z.; Azevedo, N. R. Tetrahedron Lett. 2001, 42, 6473; (c)
Guo, Q.; Miyaji, T.; Hara, R.; Shen, B.; Takahashi, T. Tetrahedron 2002, 58, 7327;
(
(
d) Yadav, J. S.; Reddy, B. V. S.; Naidu, J. J.; Sadashiv, K. Chem. Lett. 2004, 33, 926;
e) Arai, S.; Sudo, Y.; Nishida, A. Synlett 2004, 6, 1104; (f) Kobayashi, S.;
Busujima, T.; Nagayama, S. Chem. Eur. J. 2000, 6, 3491; (g) Andrade, C. K. Z.;
Azevedo, N. R.; Oliveira, G. R. Synthesis 2002, 7, 928; (h) Andrade, C. K. Z.;