Synthesis of furfural from xylose, xylan, and biomass using AlCl 3·6H2O in biphasic media via xylose isomerization to xylulose

Article abstract of DOI:10.1002/cssc.201100688

Furfural was prepared in high yields (75%) from the reaction of xylose in a water-tetrahydrofuran biphasic medium containing AlCl₃· 6H₂O and NaCl under microwave heating at 140°C. The reaction profile revealed the formation of xylulose as an intermediate en route to the dehydration product (furfural). The reaction under these conditions reached completion in 45 min. The aqueous phase containing AlCl₃· 6H₂O and NaCl could be recycled multiple times (>5) without any loss of activity or selectivity for furfural. Extension of this biphasic reaction system to include xylan as the starting material afforded furfural in 64% yield. The use of corn stover, pinewood, switchgrass, and poplar gave furfural in 55, 38, 56, and 64% yield, respectively, at 160°C. Even though AlCl₃·6H₂O did not affect the conversion of crystalline cellulose, moderate yields of the by-product 5-hydroxymethylfurfural (HMF) were noted. The highest HMF yield of 42% was obtained from pinewood. The coproduction of HMF and furfural from biomass was attributed to the weakening of the cellulose network in the biomass, as a result of hemicellulose hydrolysis. The multifunctional capacity of AlCl₃·6H₂O (hemicellulose hydrolysis, xylose isomerization, and xylulose dehydration) in combination with its ease of recyclability make it an attractive candidate/catalyst for the selective synthesis of furfural from various biomass feedstocks. Aluminum in microwaves: AlCl₃·6H₂O successively hydrolyzes biomass hemicellulose to xylose, isomerizes xylose to xylulose, and dehydrates xylulose to furfural under microwave heating at 160°C with yields in the range of 38-64%, depending on the biomass feedstock. The use of a biphasic medium allows the recycling of AlCl ₃·6H₂O (which remains in the aqueous phase) for multiple cycles without loss of activity or selectivity (see figure). Copyright

Full text of DOI:10.1002/cssc.201100688

DOI: 10.1002/cssc.201100688
Synthesis of Furfural from Xylose, Xylan, and Biomass
Using AlCl3·6H2O in Biphasic Media via Xylose
Isomerization to Xylulose
Yu Yang,[a, b] Chang-Wei Hu,*[a] and Mahdi M. Abu-Omar*[b]
Furfural was prepared in high yields (75%) from the reaction
of xylose in a water–tetrahydrofuran biphasic medium contain-
ing AlCl3·6H2O and NaCl under microwave heating at 1408C.
The reaction profile revealed the formation of xylulose as an
intermediate en route to the dehydration product (furfural).
The reaction under these conditions reached completion in
45 min. The aqueous phase containing AlCl3·6H2O and NaCl
could be recycled multiple times (>5) without any loss of ac-
tivity or selectivity for furfural. Extension of this biphasic reac-
tion system to include xylan as the starting material afforded
furfural in 64% yield. The use of corn stover, pinewood, switch-
grass, and poplar gave furfural in 55, 38, 56, and 64% yield, re-
spectively, at 1608C. Even though AlCl3·6H2O did not affect the
conversion of crystalline cellulose, moderate yields of the by-
product 5-hydroxymethylfurfural (HMF) were noted. The high-
est HMF yield of 42% was obtained from pinewood. The cop-
roduction of HMF and furfural from biomass was attributed to
the weakening of the cellulose network in the biomass, as a
result of hemicellulose hydrolysis. The multifunctional capacity
of AlCl3·6H2O (hemicellulose hydrolysis, xylose isomerization,
and xylulose dehydration) in combination with its ease of recy-
clability make it an attractive candidate/catalyst for the selec-
tive synthesis of furfural from various biomass feedstocks.
Introduction
With the increasing prices for fossil energy and environmental
concerns associated with its use, the utilization of renewable
biomass as a source of energy and chemicals has received con-
siderable attention.[1] Furfural is an important chemical that
can be obtained from biomass. It can be used as a precursor
to many important chemicals, such as furfuryl alcohol, furan,
and THF.[2] Furfural can also serve as a platform molecule for
the production of fuel, such as liquid hydrocarbons[3] and 2-
methyltetrahydrofuran.[4]
yield of 56% was obtained from xylose at 1008C after 4 h.
However, the furfural yield was only 25% from xylan, and 22%
from corn stover even at the higher temperature of 1408C and
with HCl as a cocatalyst. Zhao et al.[9] obtained furfural from
xylan in 63% yield with CrCl3 in ionic liquids as a catalyst
under microwave-assisted heating at roughly 2008C and later
extended this method to include the use of corn stalk, rice
straw, and pinewood as raw materials. The furfural yields from
these biomass variants were only 23–31%, and humins were
formed in significant amounts. Thus, a bifunctional catalyst
that can effectively convert xylose, xylan, and biomass to furfu-
ral is still needed.
Furfural can be produced from the dehydration of xylose.
Brønsted acids, such as HCl and H2SO4, are typically used.[5]
However, these mineral acids are limited by the fact that they
cause corrosion, safety problems, and require critical reaction
conditions. FeCl3·6H2O was recently used successfully in the
dehydration of xylose.[6] Additionally, the hydrolysis of
FeCl3·6H2O (0.08m) in aqueous environments resulted in a
strong acidic medium (pH 1.4). As an isomer of xylose, the con-
version of xylulose to furfural is more facile than the conver-
sion of xylose to furfural.[7] Efficient production of furfural from
xylose could be achieved by using a one-pot combination of
xylose-to-xylulose isomerization and xylulose dehydration.[8]
Ebitani et al.[8a] produced furfural from xylose via xylulose by
using a combination of solid base and acid. Hydrotalcite was
used for the isomerization and Amberlyst-15 for the dehydra-
tion of xylulose at 1008C. After 3 h, furfural was produced with
a selectivity of 51%, whereas the conversion of xylose was
72%. However, this one-pot reaction did not work in water,
and the authors did not report on the use of xylan or biomass
as starting materials. Binder et al.[8b] used CrCl3 as a bifunctional
catalyst for both, the isomerization and dehydration. A furfural
We efficiently produced 5-hydroxymethylfurfural (HMF) from
glucose, glucose-based disaccharides (maltose and cellobiose),
and polysaccharides (starch and cellulose) with AlCl3·6H2O in a
water–THF biphasic medium.[10] This process involved the one-
pot, sequential aldose–ketose (glucose–fructose) isomerization
and ketose (fructose) dehydration. We reasoned that because
[a] Y. Yang, Prof. C.-W. Hu
Key Laboratory of Green Chemistry and Technology
Ministry of Education, College of Chemistry
Sichuan University, Chengdu, Sichuan 610064 (PR China)
Fax: (+11)86 85411105
E-mail: gchem@scu.edu.cn
[b] Y. Yang, Prof. M. M. Abu-Omar
Brown Laboratory, Department of Chemistry
Center for direct Catalytic Conversion of
Biomass to Biofuels (C3Bio), Purdue University
560 Oval Drive, West Lafayette, Indiana 47907 (USA)
Fax: (+1)765 4940239
E-mail: mabuomar@purdue.edu
ChemSusChem 2012, 5, 405 – 410
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
405

C.-W. Hu, M. M. Abu-Omar, and Y. Yang
occur by cross-polymerization of
furfural with xylose and the in-
termediates, as previously sug-
gested for water–organic bipha-
sic systems[5b] and ionic liquid
solvent.[11] Relative to HCl,
AlCl3·6H2O was more effective.
With HCl, the conversion of
xylose was 6%, and the yield in
furfural was 3% after 1 h at
pH2.22, which was identical to
the pH value of an aqueous so-
lution of AlCl3·6H2O. This pH
Scheme 1. Route for the synthesis of furfural from xylose, xylan, and lignocellulosic hemicellulose via xylulose.
xylose was an aldopentose, this one-pot system would be ap-
plicable for the formation of furfural from xylose, xylan, and
lignocellulosic hemicellulose (Scheme 1).
value was higher than that of other homogeneous catalysts in
aqueous media.[5,6] Hence, corrosion caused by AlCl3·6H2O was
less severe.
A larger-scale reaction led to a slower conversion of xylose
because some amount of xylose and xylulose remained after
45 min of reaction (Table 1, entry 2 versus 1). This may be at-
Results and Discussion
We initially investigated the reaction of xylose in
water–THF with AlCl3·6H2O at 1408C (Figure 1), the
Table 1. Comparison of microwave and conventional heating.
conversion of which was >99% within 45 min. In
the early stages of the reaction, xylulose was formed,
reaching a maximum yield of 30% in 5 min, followed
by a decline to nearly zero at 45 min. This trend indi-
cated that xylulose was an intermediate in the pro-
duction of furfural. At the end of the reaction
(45 min), xylose and xylulose were consumed, and
the yield in furfural was 75%. A further increase of re-
action time had little effect on the furfural yield. The
introduction of an organic solvent (an extracting sol-
vent for furfural) enhanced the stability of furfural by
preventing further degradation.[6b] After heating pure
furfural in this biphasic system at 1408C for 45 min,
98% of the starting furfural was recovered. The for-
mation of soluble humins (no formation of insoluble
humin was observed) appeared to predominantly
Entry
Heating
method
Time[a]
[min]
xylose+xylu-
Furfural selec-
tivity [%]
Insoluble
humins
lose[b] yield [%]
1[c]
2[d]
3[e]
microwave
microwave
conventional
<2
<2
ꢀ20
1
19
2
76
80
64
no
no
formed
[a] From RT to 1408C. [b] Unreacted xylulose. [c] Reaction conditions: xylose
(0.25 mmol), AlCl3·6H2O (0.1 mmol), NaCl (6.0 mmol), water (1 mL), THF (3 mL) in a
10 mL reaction tube at 1408C under microwave heating for 45 min. [d] Reaction con-
ditions: xylose (1.0 mmol), AlCl3·6H2O (0.4 mmol), NaCl (24.0 mmol), water (4 mL), THF
(12 mL) in a 35 mL reaction tube at 1408C under microwave heating for 45 min.
[e] Reaction conditions: xylose (1.0 mmol), AlCl3·6H2O (0.4 mmol), NaCl (24.0 mmol),
water (4 mL), THF (12 mL) in a 35 mL reaction tube at 1408C under conventional heat-
ing for 45 min.
tributed to the different mass transfer in larger-scale reactions.
However, a high selectivity in furfural formation was observed
in the larger-scale reaction. Conventional heating required
more time to reach the desired temperature than microwave
heating (20 versus 2 min). Although a high conversion of
xylose could be obtained by conventional heating, the selectiv-
ity in furfural production was significantly lower. The formation
of insoluble humins was observed for conventional heating,
but not in microwave batch reactions. Microwave heating re-
sulted in improved heat transfer and, in turn, in better selectiv-
ity.[12]
Xylose conversion required a temperature of 1408C or
higher. Below 1408C, the furfural yield was low. Between 140
and 1608C, there was essentially no effect on yield and selec-
tivity (Figure 2).
Given that AlCl3·6H2O remained dissolved in the aqueous
phase, it could be easily recycled and reused by simply remov-
ing the THF phase. The aqueous phase was extracted with THF
&
*
Figure 1. Time profiles for xylose conversion with AlCl3·6H2O ( xylose, xy-
(3ꢁ2 mL), xylose and THF (3 mL) were added to the recycled
AlCl3·6H2O aqueous phase, and the reaction mixture was
heated to 1408C in a microwave reactor. The aqueous solution
~
lulose, furfural). Reaction conditions: xylose (0.25 mmol), AlCl3 ·6H2O
(0.1 mmol), NaCl (6.0 mmol) in water–THF (1:3 mL) at 1408C under micro-
wave heating.
406
www.chemsuschem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2012, 5, 405 – 410

Synthesis of Furfural from Xylose, Xylan, and Biomass
Figure 4. Effect of xylose concentration on its conversion to furfural. Reac-
tion conditions: AlCl3·6H2O (0.1 mmol), NaCl (6.0 mmol) in water–THF (1:3
mL) at 1408C under microwave heating for 45 min.
Figure 2. The effect of reaction temperature on xylose conversion to furfural.
Reaction conditions: xylose (0.25 mmol), AlCl3·6H2O (0.1 mmol), NaCl
(6.0 mmol) in water–THF (1:3 mL) under microwave heating for 45 min.
containing AlCl3·6H2O was recycled and used for multiple con-
secutive runs without loss of activity or selectivity (Figure 3).
These results showed that there was no loss in the furfural
Figure 5. Effect of AlCl3·6H2O catalyst loading on xylose conversion to furfu-
ral. Reaction conditions: xylose (0.25 mmol), NaCl (6.0 mmol) in water–THF
(1:3 mL) at 1408C under microwave heating for 45 min.
Figure 3. Recycling of the aqueous phase containing AlCl3·6H2O and NaCl.
Reaction conditions: xylose (0.25 mmol), AlCl3·6H2O (0.10 mmol), NaCl
(6.0 mmol) in water–THF (1:3 mL) at 1408C under microwave heating for
45 min.
yield even after five cycles, which represented >12 turnovers
(on the basis of moles of xylose converted per mole of
AlCl3·6H2O). Thus, at higher concentrations of xylose (Figure 4)
or lower catalyst loading (Figure 5), AlCl3·6H2O still showed
reasonable activity. Nevertheless, at higher concentrations of
xylose or lower catalyst loading, small amounts of xylose and
xylulose remained unreacted after 45 min, and, as a result, fur-
fural yields were lower. Longer reaction times were needed to
complete the dehydration reaction.
The conversion of xylan to furfural was investigated under
the same reaction conditions used for xylose (Figure 6). After
only 1 min, the yield in xylose was 45%, whereas the xylulose
yield was 12%, and the furfural yield was approximately 2%.
&
*
Figure 6. Time profiles for the reaction of xylan with AlCl3·6H2O ( xylose,
~
xylulose, furfural). Reaction conditions: xylan (0.25 mmol, based on mono-
saccharide units), AlCl3·6H2O (0.1 mmol), NaCl (6.0 mmol) in water–THF (1:3
mL) at 1408C under microwave heating.
ChemSusChem 2012, 5, 405 – 410
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
407

C.-W. Hu, M. M. Abu-Omar, and Y. Yang
The hydrolysis of xylan to monosaccharides was surprisingly
fast. The reaction profiles from this point onward were similar
to that of pure xylose: xylulose was observed as an intermedi-
ate, and after 60 min furfural yields plateaued at 64%. The fact
that yields were lower than those obtained with pure xylose
could be attributed to interference from other components in
xylan.
grass (56%, entry 9), and poplar (64%, entry 10). The furfural
yield from pinewood remained significantly lower than the
other biomass variants. However, further increase of the reac-
tion temperature to 1808C led to an improved furfural yield of
61% (entry 13) from pinewood, which was significantly higher
than that observed when using CrCl3 in N,N-dimethylacetamide
or ionic liquids.[8b,9]
Subsequently, we applied the described reaction conditions
to untreated lignocellulosic biomass. The furfural yields ob-
tained from corn stover, pinewood, switchgrass, and poplar
were 51, 29, 50, and 45%, respectively (Table 2). The yields of
Notably, even at 1408C small amounts of HMF and glucose
were observed, albeit in low yields (4–19%). We presumed
that cellulose must have been partly hydrolyzed to account for
the observed HMF. However, when pure cellulose was used
with or without added xylan, only trace amounts of HMF and
glucose were observed. Therefore, we concluded that breaking
the network of bonds between hemicellulose and cellulose
must be affecting the structure of cellulose in biomass sam-
ples, which made the hydrolysis of cellulose in biomass easier
than that of pure cellulose.[14] Additionally, and in support of
the previous hypothesis, pinewood gave the lowest furfural
yield, and the highest HMF yield at 1408C (19%). The increase
of the reaction temperature to 1608C also improved the yield
of HMF. An HMF yield of 15% could be obtained from pure
cellulose. The majority of pure cellulose remained unconverted
as a white solid suspension in the aqueous phase. The HMF
yields from corn stover (14%), switchgrass (13%), and poplar
(15%) were equivalent to that from pure cellulose. However,
the HMF yield from pinewood approached 42% at 1608C,
which was incidentally higher than that obtained from cellu-
lose (37%) at 1808C (Table 2, entry 14). These results implied
that cellulose in pinewood was nearly totally converted at
1608C. A further increase of temperature to 1808C in reactions
of pinewood led to a decrease in HMF yield (entry 13).
Table 2. Product yields obtained from various sources of lignocellulosic
biomass.[a]
Entry
Sample
Yield [%]
furfural
xylose
HMF
glucose
1
2
3
4
5
corn stover
pinewood
switchgrass
poplar
cellulose[b]
51
29
50
45
-
2
12
8
6
-
6
19
11
4
2
5
2
1
2
4
6
7
8
9
10
11
12
13
14
cellulose/xylan[c]
corn stover[d]
pinewood[d]
switchgrass[d]
poplar[d]
cellulose[b,d]
cellulose/xylan[c,d]
pinewood[e]
cellulose[b,e]
64
55
38
56
64
–
66
61
–
2
3
1
1
1
<1
<1
1
1
1
<1
<1
1
<1
–
0
<1
–
14
42
13
15
15
14
35
37
<1
[a] Reaction conditions: biomass (0.05 g), AlCl3·6H2O (0.1 mmol), NaCl
(6.0 mmol), water (1 mL), THF (3 mL), reaction temperature 1408C, and re-
action time 60 min. Yields for furfural and xylose were based on a pen-
tose content of 27, 22, 23, and 19% (by weight), and yields of HMF and
glucose from corn stover, pinewood, switchgrass, and poplar were based
on a hexose content of 35, 26, 35, and 45% (by weight), respectively.
[b] 40.5 mg cellulose (0.25 mmol based on glucose units). [c] 20.2 mg cel-
lulose (0.125 mmol based on glucose units) and 18.3 mg xylan
(0.125 mmol based on xylose units). [d] Reaction temperature 1608C and
reaction time 60 min. [e] Reaction temperature 1808C and reaction time
30 min.
Other by-products, such as acetic acid (from the hydrolysis
of acetyl in hemicellulose), formic acid, and levulinic acid (from
the rehydration of HMF) were also observed for lignocellulosic
biomass samples. However, their yields were low, and the
major by-products were humins.
furfural were clearly dependent on the biomass feedstock. Sim-
ilar observations were reported in the literature. The research
groups of Mazza[5c] and Zhao[8] found disparate furfural yields
using different biomass, suggesting that hemicellulose recalci-
trance was species-dependent. Unlike pure xylan, hemicellu-
lose in biomass served as linkers of cellulose fibers into microfi-
brils, and cross-linkers of cellulose with lignin to create a com-
plex network of bonds that provided structural stability.[13]
Such a network in lignocellulosic biomass made the hydrolysis
of hemicellulose more difficult than that of pure xylan. For ex-
ample, the conversion of corn-stover hemicellulose after 1 min
was 25% (15% xylose and 10% xylulose), relative to a conver-
sion of 58% for pure xylan after 1 min. The fact that uncon-
verted xylose (3–12%) could be detected even after 60 min
provided support for these arguments. Higher temperatures
(1608C) slightly improved furfural yields in reactions with corn
stover (55%, Table 2, entry 7), pinewood (38%, entry 8), switch-
Conclusions
AlCl3·6H2O enabled the synthesis of furfural from xylose in a
water–THF biphasic medium system at moderate temperatures
(1408C). The reaction proceeded through the isomerization of
xylose to xylulose, followed by the dehydration of the latter to
give furfural. Moreover, this system was effective in the hydrol-
ysis of xylan and lignocellulosic hemicelluloses to xylose. Al-
though in each loading substoichiometric amounts of
AlCl3·6H2O were used (relative to the monosaccharide),
AlCl3·6H2O could not be categorized as a catalyst, because the
quantity (of AlCl3·6H2O) was not sufficiently low. However, the
aqueous solution containing AlCl3·6H2O could be recycled mul-
tiple times (>5 cycles) without any noticeable loss of activity.
In this context, therefore, AlCl3·6H2O could be referred to as a
catalyst for the conversion of xylan and lignocellulosic biomass
into furfural.
408
www.chemsuschem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2012, 5, 405 – 410

Synthesis of Furfural from Xylose, Xylan, and Biomass
maintained at 338 K. The concentration of furfural and HMF in the
organic phase was determined by using an HPLC equipped with a
Experimental Section
General methods
Waters 1525 pump,
a
XDB-C18 column, (Agilent), and
a
Waters 2487 UV detector. A solution of formic acid (0.5 wt%) was
used as the mobile phase at a flow rate of 0.4 mLminꢁ1, and the
column temperature was maintained at 303 K. The concentrations
of xylose, xylulose, glucose, furfural, and HMF were calculated
based on standard curves constructed by using authentic samples.
Xylose, xylulose, glucose, beechwood xylan (95% xylose residues,
ca. 95% dry solids), cellulose, AlCl3·6H2O, NaCl, HCl (37 wt%), THF,
and HMF were purchased from Sigma–Aldrich. The above reagents
were all analytical grade and used without further purification
unless noted otherwise. Furfural obtained from Sigma–Aldrich was
distilled prior to use. The Purdue University Department of Agricul-
ture provided corn stover. Dr. Keith Johnson (Purdue University)
supplied the switchgrass, Mr. Jerry Warner (Defence LifeSciences)
supplied the pinewood, and the USDA Forest Product Lab (CAFI
Consortium) supplied the poplar. The biomass samples were dried
(<5% moisture) and milled to pass through a 1 mm mesh screen
for further use. The approximate composition of lignocellulosic bio-
mass was based on published values.[13]
Calculation of xylose, xylulose, and furfural percentages (xylose re-
action):
Xyloseð%Þ ¼ xyloseunreactedðmolÞ=xylosestartðmolÞ ꢂ 100ð%Þ
Xyluloseð%Þ ¼ xyluloseprod:ðmolÞ=xylosestartðmolÞ ꢂ 100ð%Þ
Furfuralð%Þ ¼ furfuralprod:ðmolÞ=xylosestartðmolÞ ꢂ 100ð%Þ
ðXyloseþxyluloseÞunreacted ð%Þ ¼ xyloseð%Þþxyluloseð%Þ
Selectivityfurfural ¼ furfuralð%Þ=½ðxyloseþxyluloseÞunreactedð%Þꢃ
ð1Þ
ð2Þ
ð3Þ
ð4Þ
ð5Þ
Synthesis of furfural from xylose, beech xylan, and lignocel-
lulosic biomass under microwave heating
Calculation of product yields (beechwood xylan, cellulose, and
lignocellulosic biomass feedstock reactions:
Reactions were performed in a Discover TM microwave batch reac-
tor (CEM Corp.). In a typical experiment, a reaction tube (10 mL)
was charged with xylose (37.5 mg, 0.25 mmol), xylan (36.6 mg,
0.25 mmol based on xylose units) or lignocellulosic biomass
(50 mg), AlCl3·6H2O (24.1 mg, 0.10 mmol), NaCl (0.35 g, 24.0 mmol),
deionized water (1 mL), and THF (3 mL), and heated to 1408C for
the desired time. All solutions were mixed at a maximum constant
rate using a magnetic stirrer. Temperatures in the reactor were
measured by means of a fiber optic sensor. The reaction vessel was
pressurized because of the vapor pressure of the solution at the re-
action temperatures used. After the desired reaction time, the reac-
tion was stopped by performing nitrogen-flow cooling, and the
phases were separated. When using xylose as the starting material,
the aqueous phase was diluted and then filtered with a syringe
filter (0.2 mm) prior to analysis. When using beech xylan or lignocel-
lulosic biomass as the raw material, the aqueous phase was first fil-
tered using filter paper to remove the unconverted solid samples,
and then the filtrate was diluted and filtered using a 0.2 mm sy-
ringe filter prior to analysis. The organic phase was diluted and fil-
tered using a 0.2 mm syringe filter prior to analysis.
Yieldxyloseð%Þ ¼ xyloseprod:ðmolÞ=pentosesstartðmolÞ ꢂ 100ð%Þ
Yieldxyluloseð%Þ ¼ xyluloseprod:ðmolÞ=pentosesstartðmolÞ ꢂ 100ð%Þ
Yieldfurfuralð%Þ ¼ furfuralprod:ðmolÞ=pentosesstartðmolÞ ꢂ 100ð%Þ
Yieldglucoseð%Þ ¼ glucoseprod:ðmolÞ=hexosesstartðmolÞ ꢂ 100ð%Þ
YieldHMFð%Þ ¼ HMFprod:ðmolÞ=hexosesstartðmolÞ ꢂ 100ð%Þ
ð6Þ
ð7Þ
ð8Þ
ð9Þ
ð10Þ
Acknowledgements
This work was financially supported by the National Basic Re-
search Program of China (973 program, 2007CB210203), NNSFC
21072136, PCSIRT (IRT0846), the Center for direct Catalytic Con-
version of Biomass to Biofuels (C3Bio), an Energy Frontier Re-
search Center funded by the U.S. Department of Energy, Office of
Science, and Office of Basic Energy Sciences under Award
Number DE-SC0000997. C3Bio provided funding for laboratory
space, instrumentation, and chemical supplies at Purdue Univer-
Synthesis of furfural from xylose under conventional heat-
ing
Reactions were performed in a Parr reactor. A reaction tube
(35 mL) was charged with xylose (1 mmol), AlCl3·6H2O (96.4 mg,
0.40 mmol), NaCl (1.4 g, 24.0 mol), deionized water (4 mL), and THF
(12 mL), and heated to 1408C for 45 min. Zero time is taken at the
moment the temperature reached 1408C. The solution was mixed
at a constant rate of 800 rpm using a magnetic stirrer. Tempera-
tures in the reactor were measured by using a thermocouple. The
reactor was removed from the electric furnace and cooled to ambi-
ent temperature. The aqueous and organic phases were separated,
diluted, and filtered using a 0.2 mm syringe filter prior to analysis.
sity. Y.Y. was supported by
(2010624099).
a China Scholarship grant
Keywords: aluminum · biomass · biphasic catalysis · furfural ·
microwave chemistry
[1] D. Dodds, R. Gross, Science 2007, 318, 1250–1251.
[2] A. Corma, S. Iborra, A. Velty, Chem. Rev. 2007, 107, 2411–2502.
[3] a) G. W. Huber, J. N. Chheda, C. J. Barrett, J. A. Dumesic, Science 2005,
308, 1446–1450; b) R. M. West, Z. Y. Liu, M. Peter, J. A. Dumesic, Chem-
SusChem 2008, 1, 417–424.
Analytical methods
[4] W. Yang, A. Sen, ChemSusChem 2010, 3, 597–603.
[5] a) C. Sievers, I. Musin, T. Marzialetti, M. B. Valenzuela Olarte, P. K. Agraw-
al, C. W. Jones, ChemSusChem 2009, 2, 665–671; b) R. Weingarten, J.
Cho, W. C. Conner Jr, G. W. Huber, Green Chem. 2010, 12, 1423–1429;
c) O. Yemis¸, G. Mazza, Bioresour. Technol. 2011, 102, 7371–7378.
[6] T. vom Stein, P. M Grande, W. Leitner, P. Domꢂnguez de Marꢂa, ChemSu-
sChem 2011, 4, 1592–1594.
Quantitative analyses of the aqueous phases were performed by
using an HPLC equipped with a Waters 1525 pump, an aminex
HPX-87 column (Agilent), and a Waters 2412 refractive index detec-
tor. A solution of H2SO4 (0.005m) was used as the mobile phase at
a flow rate of 0.6 mLminꢁ1, and the column temperature was
ChemSusChem 2012, 5, 405 – 410
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
409

C.-W. Hu, M. M. Abu-Omar, and Y. Yang
[7] T. Ahmad, L. Kenne, K. Olsson, O. Theander, Carbohydr. Res. 1995, 276,
309–320.
[8] a) A. Takagaki, M. Ohara, S. Nishimura, K. Ebitani, Chem. Lett. 2010, 39,
838–840; b) J. B. Binder, J. J. Blank, A. V. Cefali, R. T. Raines, ChemSu-
sChem 2010, 3, 1268–1272.
[9] Z. Zhang, Z. K. Zhao, Bioresour. Technol. 2010, 101, 1111–1114.
[10] Y. Yang, C. W. Hu, M. M. Abu-Omar, Green Chem 2012, DOI: 10.1039/
c1gc15972k.<
[13] E. M. Rubin, Nature 2008, 454, 841–845.
[14] M. E. Himmel, S. Y. Ding, D. K. Johnson, W. S. Adney, M. R. Nimlos, J. W.
Brady, T. D. Foust, Science 2007, 315, 804–807.
[15] a) Y. L. Lu, N. S. Mosier, Biotechnol. Prog. 2007, 23, 116–123; b) Y. L. Lu,
N. S. Mosier, Biotechnol. Bioeng. 2008, 101, 1170–1181; c) C. E. Wyman,
B. E. Dale, R. T. Elander, M. Holtzapple, M. R. Ladisch, Y. Y. Lee, C. Mitchin-
son, J. N. Saddler, Biotechnol. Prog. 2009, 25, 333–339.
[11] S. Lima, P. Neves, M. M. Antunes, M. Pillinger, N. Ignatyev, A. A. Valente,
Appl. Catal. A: Gen. 2009, 363, 93–99.
Received: November 2, 2011
[12] J. Robinson, S. Kingman, D. Irvine, P. Licence, A. Smith, G. Dimitrakis, D.
Obermayer, C. O. Kappe, Phys. Chem. Chem. Phys. 2010, 12, 4750–4758.
Published online on February 7, 2012
410
www.chemsuschem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2012, 5, 405 – 410