Purification of biomass-derived 5-hydroxymethylfurfural and its catalytic conversion to 2,5-furandicarboxylic acid

Article abstract of DOI:10.1002/cssc.201402105

A simple and effective water extraction method is presented for the purification 5-hydroxylmethylfurfural (HMF) obtained from a biomass dehydration system. Up to 99 % of the HMF can be recovered and the HMF in aqueous solution is directly converted to 2,5-furandicarboxylic acid (FDCA) as the sole product. This purification technique allows an integrated process to produce FDCA from fructose via HMF prepared in an isopropanol monophasic system, with an overall FDCA yield of 83 % obtained. From Jerusalem raw artichoke biomass to FDCA via HMF prepared in a water/MIBK (methyl isobutyl ketone) biphasic system, an overall FDCA yield of 55 % is obtained.

Full text of DOI:10.1002/cssc.201402105

CHEMSUSCHEM
COMMUNICATIONS
DOI: 10.1002/cssc.201402105
Purification of Biomass-Derived 5-Hydroxymethylfurfural
and Its Catalytic Conversion to 2,5-Furandicarboxylic Acid
Guangshun Yi, Siew Ping Teong, Xiukai Li, and Yugen Zhang*[a]
A simple and effective water extraction method is presented
for the purification 5-hydroxylmethylfurfural (HMF) obtained
from a biomass dehydration system. Up to 99% of the HMF
can be recovered and the HMF in aqueous solution is directly
converted to 2,5-furandicarboxylic acid (FDCA) as the sole
product. This purification technique allows an integrated pro-
cess to produce FDCA from fructose via HMF prepared in an
isopropanol monophasic system, with an overall FDCA yield of
83% obtained. From Jerusalem raw artichoke biomass to FDCA
via HMF prepared in a water/MIBK (methyl isobutyl ketone) bi-
phasic system, an overall FDCA yield of 55% is obtained.
distillation or multiple-cycle extraction with cyclopentane. Sub-
sequently, the HMF was further oxidized to FDCA with Au/TiO2
catalyst in concentrated NaOH solution, with about 80% FDCA
yield of purified HMF by distillation, and 60% FDCA yield of
purified HMF with cyclopentane extraction. The overall yields
from glucose to FDCA were 50% and 35%.
During the past few decades, a variety of methods have
been developed to convert biomass to HMF.[11] HMF can be
produced from single-phase systems by using solvents such as
water, dimethyl sulfoxide,[21] ionic liquids,[22] isopropanol,[23]
THF, and g-valerolactone (GVL).[12] It can also be produced from
biphasic systems, using the aqueous layer for reaction, and sol-
vents like alcohols and ketones as extracting layers.[24,25]
As HMF is an intermediate molecule and the market for HMF
itself is small, it is important not only to optimize its synthetic
process, but also to develop an efficient isolation/purification
method that could be integrated into its subsequent upgrad-
ing reactions, in order to render the whole process economi-
cally viable for large-scale production. One of the problems
during HMF production is the presence of impurities like
humins, which are soluble in many organic solvents and co-
exist with HMF.[26] The impurities will not only affect the ap-
pearance of the final product, but also affect the downstream
conversion of HMF to other useful products.[12,14]
The depletion of fossil resources has generated world-wide in-
terest in the development of renewable and sustainable alter-
natives for fuels and chemicals.[1–9] Chemical industries are
shifting their focus to the development of sustainable manu-
facturing processes by utilizing abundant biomass in environ-
mentally benign solvents.[10–14] In this context, 2,5-furandicar-
boxylic acid (FDCA) has received significant attention as a possi-
ble replacement for terephthalic acid for the production of
polyamides, polyesters, and polyurethanes.[15] A furan-based
polymer poly(ethylene-2,5-furandicarboxylate) (PEF) has been
prepared from renewable sources, and it has demonstrated
comparable thermal stability to polyethylene terephthalate
(PET), a polymer commonly made into consumer goods for
thousands of applications.[16]
Various methods, including extraction, column/HPLC, and ac-
tivated carbon absorbent, have been used to isolate and purify
HMF.[11] The extraction method typically requires multiple
cycles that not only consume large amount of solvents, but
also introduce noticeable impurities to the organic layer.
Column or HPLC methods can generate a high purity of
HMF.[23] However, these methods are more suitable for small-
scale production rather than mass production. Activated
carbon can efficiently remove impurities from HMF solution,[27]
but up to 50% of HMF was also lost and absorbed onto the
absorbent in our experiments (Supporting Information, Fig-
ure S1). All these methods are costly and not suitable for mass
production.
The biomass-derived FDCA can be produced by aerobic oxi-
dation of 5-hydroxymethylfurfural (HMF) with various metal
catalysts, while HMF can be prepared by acid-catalyzed dehy-
dration of sugars or cellulose. Although the process for HMF
oxidization to FDCA has been reported in literature, most of
the investigations used pure/commercial HMF for the transfor-
mation with almost quantitative FDCA yield.[17,18] Only a few
examples considered the direct conversion of fresh biomass-
derived HMF into FDCA or combined an acid-catalyzed dehy-
dration step with a metal-catalyzed oxidization step.[19,20] When
adopting this strategy, one critical challenge is the deactivation
of oxidization catalysts as the impurities in the biomass-de-
Our group recently reported an isopropanol system for the
production of HMF from fructose.[23] HMF can be easily separat-
ed from the reaction system by distillation of solvent for recy-
cling. In our study, during the further oxidization of the crude
HMF to FDCA using literature reported method with Au/HT[18b]
or Pt/C[28] catalysts, it was found that the impurities significant-
ly affected the catalytic oxidization reaction. In both cases, im-
purities deposited on the surface of the metal catalyst, result-
ing in the deactivation of the catalyst. As a result, a mixture of
FDCA and 5-hydroxymethyl-2-furancarboxylic acid (HFCA) was
observed (Supporting Information, Figure S2).
rived HMF will seriously affect the subsequent reaction.[12]
A
recent report described a two-step process converting glucose
to FDCA.[12] In this process, HMF was produced and purified by
[a] Dr. G. Yi, S. P. Teong, Dr. X. Li, Dr. Y. Zhang
Institute of Bioengineering and Nanotechnology
31 Biopolis Way, The Nanos
Singapore 138669 (Singapore)
E-mail: ygzhang@ibn.a-star.edu.sg
Supporting Information for this article is available on the WWW under
http://dx.doi.org/10.1002/cssc.201402105.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 5
&
1
&
ÞÞ
These are not the final page numbers!

CHEMSUSCHEM
COMMUNICATIONS
www.chemsuschem.org
Further investigation found that the dark humin impurities
are highly soluble in isopropanol or other organic solvents
(e.g., methanol, methyl isobutyl ketone (MIBK), THF, and
DMSO), but insoluble in water. Since HMF is soluble in both
water and organic solvent, we have devised a simple yet effec-
tive method to remove these impurities from HMF. As shown
in Figure 1a, the original HMF solution after the process of de-
solution, which dissolved more impurities into the water
phase.
After that, oxidation reactions using Au/HT[18b] or Pt/C[28] cat-
alyst were carried out with HMF before and after purification,
following the reported methods. As described in the litera-
tures, Au/HT[18b] was used in a base-free aqueous solution, re-
sulting in 99% FDCA yield, and Pt/C[28] was used in NaOH
aqueous solution, resulting in FDCA yield of >95%. Both litera-
tures use high purity HMF as a starting material. It was report-
ed that FDCA as an acid can dissolve HT (hydrotalcite) and de-
stroy Au/HT catalyst.[29] In our experiment, a molar equivalent
of Na2CO3 was added to the reaction solution in order to neu-
tralize FDCA and protect HT from dissolving. As a control,
nearly quantitative yield of FDCA could also be achieved with-
out the addition of any base. As shown in Table 1, all reactions
Table 1. Oxidation of HMF to FDCA.
Figure 1. a) As-synthesized HMF in isopropanol solution (left), crude HMF
product after evaporation of solvent (middle), and HMF re-dissolved in
water (right). b) Extraction times versus HMF recovery rate. Each time 5 mL
of water was used for extraction.
Entry
HMF
Catalyst
Time
FDCA Yield
1[a]
2[b]
3[a]
4[b]
no extraction
no extraction
water extraction
water extraction
Au/HT
Pt/C
Au/HT
Pt/C
20 h
20 h
7 h
39%+HFCA
51%+HFCA
99%
4.25 h
98%
[a] Reaction conditions: 1 mmol HMF in 10 mL H2O, 0.25 g Au/HT catalyst,
HMF/Au=40 (mol/mol), 1 mmol Na2CO3, O2 bubbling, 508C, 2 h, 958C.
[b] Reaction conditions: 10 mL H2O, 0.4 g Pt/C catalyst, HMF/Pt=10 (mol/
mol), 0.5 g NaOH; 1 mmol HMF in 5 mL H2O was added dropwise, O2
bubbling, 248C.
hydration of fructose in isopropanol[23] was dark brown in
color, whereas pure HMF solution is pale yellow. Apparently,
the dark brown color came from the impurities, which are
mainly humins. After evaporation of isopropanol, a viscous
black liquid was obtained. Water was then added into the vial
to this viscous liquid. HMF was dissolved in water, and formed
a transparent yellow color solution. Most of the impurities
were insoluble and remained at the bottom as black solid (Fig-
ure 1a). Further study showed that the water-extraction pro-
cess is very efficient and 99% of HMF could be recovered
within two rounds of extraction (Figure 1b).
with unpurified HMF encountered the problem of catalyst de-
activation. A mixture of HFCA and FDCA was obtained as
a final product. Even with an extended reaction time to 20 h,
no improvement was observed. However, for the HMF after
purification, the reaction was completed in 7 h over Au/HT cat-
alyst and in 4.25 h over Pt/C catalyst, and most importantly,
only FDCA (>98% yield) was detected as the final product
(Scheme 1).
It was found that the complete removal of the solvent from
the original HMF solution was a key to achieving high quality
HMF. This was done by first evaporating isopropanol from the
original solution of HMF at 90 mbar, 408C, followed by evapo-
ration at lower vacuum conditions (using the continuous evap-
oration mode of the rotary evaporator) to completely remove
the solvent. Without the second evaporation step, the trace
amount of isopropanol that remains in the crude HMF will in-
troduce noticeable impurities to the subsequent water extrac-
tion solution, and as a result, a darker color aqueous solution
will get, and a slower catalytic reaction to FDCA will usually
occur. This was also achieved by first evaporating isopropanol
from the original solution of HMF at 90 mbar, 408C, and subse-
quently leaving the crude HMF drying in air overnight.
A kinetics study was conducted for both pure HMF and our
fructose-derived HMF with Au/HT catalyst under the same reac-
tion conditions. It was shown that the conversion of HMF to
FDCA with fructose-derived HMF was completed within 7 h.
This was slightly slower than the reaction with pure HMF in
which the reaction was finished in 4 h (Figure S3). After further
investigation, the optimized reaction conditions for fructose-
derived HMF were found to be as follows: Firstly, oxygen gas
was bubbled through the reaction mixture at 508C for 2 h,
during which HMF was nearly quantitatively converted to
HFCA.[30] The reaction temperature was then raised to 958C for
another 7 h, during which the HFCA was fully converted to
FDCA with yields up to 99%. Alternatively, a palladium-modi-
fied gold catalyst (Au8Pd2/HT) could convert purified HMF to
FDCA in 7 h with 99% yield, without the need for pretreat-
ment at lower temperature (Supporting Information, Fig-
ure S3).
Alternatively, an equal volume of water was mixed with the
HMF-containing isopropanol solution, and isopropanol was
evaporated from the solution. After removing the impurities
from the aqueous solution by either filtration or centrifugation,
>99% of HMF remained in the water solution. However, more
impurities were observed in the aqueous solution. This was
also due to the small amount of isopropanol remaining in the
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 5
&
2
&
ÞÞ
These are not the final page numbers!

CHEMSUSCHEM
COMMUNICATIONS
www.chemsuschem.org
Scheme 1. Process for conversion of fructose to FDCA.
The tolerance of Au/HT catalyst for this biomass-based HMF
was also studied. The Au/HT catalyst showed high activity for
the first 2 runs with an FDCA yield of 98–99% (Supporting In-
formation, Figure S4). However from the 3rd run, a significant
slowdown of reaction speed was observed (ꢀ56%). In a paral-
lel experiment with pure HMF (from Aldrich) as the starting
material, a lower reaction speed of 78% was also observed for
the 3rd run (Supporting Information, Figure S4). To increase the
durability of the catalyst, we have prepared palladium-modi-
fied Au/HT (Au8Pd2/HT) as a new catalyst, which showed excel-
lent recyclability. The catalyst was kept at high activity for at
least 5 runs with FDCA yield of 98–99% (Supporting Informa-
tion, Figure S5), using the purified HMF from fructose. A
carbon-supported Au–Pd (Au–Pd/C) catalyst was also reported
to have good recyclability over Au/C in a recent literature.[18d]
We have successfully demonstrated the conversion of fruc-
tose to HMF in isopropanol and the purification of HMF, fol-
lowed by oxidation of HMF to FDCA. The two-step reactions
were then integrated for the direct conversion of fructose to
FDCA. In this process, fructose was converted to HMF in iso-
propanol with 5 mol% of HCl as catalyst. After the reaction,
isopropanol was separated by evaporation and collected for
the next run reaction. Then, HMF was purified with water-ex-
traction and this aqueous solution was directly used for the ox-
idization reaction. As shown in Table 2, an overall FDCA yield
of 83% was achieved. During this integrated process, both the
solvent (isopropanol) and cata-
for the next step catalytic reaction. This purification process
could be simply incorporated into the mass production of
FDCA or other processes.
A biphasic system is another widely used method for HMF
production from fructose, as well as glucose or cellulose.[24,25]
Herein, we tested our integrated process for the conversion of
fructose to FDCA by using a biphasic fructose dehydration
method. In a water/MIBK biphasic system,[25] HMF product
(ꢀ55% yield) in the organic layer was dried, extracted with
water, and then directly used for oxidization reaction (Support-
ing Information, Figure S6). Under our standard conditions,
>97% FDCA yield with 100% HMF conversion was achieved.
We also tested HMF prepared from glucose in a biphasic
Table 2. Integrated Process from Fructose to FDCA.
Entry
Fructose [mmol]
HMF[a] Yield [%][c]
FDCA[b] Yield [%][c]
1
2
1
5
85.3 (84.0)
80.2 (79.4)
83.0
78.0
[a] Reaction conditions for step one: fructose 1 mmol, isopropanol
1.94 mL, H2O 0.06 mL, HCl 0.05 mmol, 1208C, 3 h. Entry 2: 5ꢁ scale up.
[b] Reaction conditions for step two: HMF in step 1 was extracted in
10 mL H2O, Au/HT 0.25 g, Na2CO3 1 mmol, O2 bubbling, 508C 2 h, 958C
7 h. [c] HMF yield is based on HPLC analysis. Isolated yield is given in pa-
renthesis. FDCA overall yield is based on fructose.
lyst were recycled. The whole
process did not produce any ad-
ditional waste and only water
was consumed during the HMF
purification process (Scheme 2).
This method allows FDCA to be
produced from fructose in
a highly efficient, costly-effective,
and environmentally benign
way.
The key step for this process
is to use water to extract and
purify HMF. This method not
only removes impurities in
a simple and effective way, but
also generates HMF aqueous so-
Scheme 2. Integrated system for the conversion of fructose to FDCA: apart from the reactants only water is con-
lution that can be used directly sumed during the reaction.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 5
&
3
&
ÞÞ
These are not the final page numbers!

CHEMSUSCHEM
COMMUNICATIONS
www.chemsuschem.org
chemicals were used directly without any pretreatment. Pt/C
(5 wt%) was purchased from Aldrich.
system of water/THF using HCl/AlCl3 as catalyst with a HMF
yield of 52%.[12] For as-prepared HMF in THF solution, with our
standard process, >99% of HMF was recovered after purifica-
tion, and the final overall FDCA yield was 50%. We have there-
fore demonstrated that this HMF purification method can also
be applied in other reaction systems with different solvents
and different starting materials.
General reaction procedure to produce HMF from fructose: The
reaction procedure was based on our previous paper[23] with minor
adaptations. Briefly, in a 8 mL sealed glass tube fructose (2.5 mmol,
0.45 g), anhydrous isopropanol (4.85 mL), water (0.15 mL, 3 vol%),
and 37% HCl (10 mL, 5 mol%) were added. The solution was
purged with argon gas 3 times until all the air was replaced. Under
magnetic stirring at 700 rpm, the reaction was heated to 1208C in
an oil bath for 3 h. After the reaction, the solution was cooled
down in an ice bath. The solution was diluted in water for HPLC
analysis. For an amount of starting materials of 1 mmol and
5 mmol, the experiments were conducted in 4 mL and 15 mL
sealed glass tubes, respectively.
So far, we have successfully demonstrated the conversion of
fructose and glucose to FDCA in an integrated process. It
would be more interesting and challenging if we were able to
convert raw biomass rather than pure sugar to FDCA. Jerusa-
lem artichoke tuber (JAT) is an abundant, easy and fast grow-
ing biomass with very high inulin/fructose component (ꢀ68–
83% fructans).[31] As an example, the conversion of JAT bio-
mass to FDCA was tested by first converting JAT to HMF in
a water/MIBK biphasic system. Compared to the monophasic
system, the biphasic system works better for the conversion of
JAT to HMF/FDCA, as impurities in JAT, such as biomolecules
(proteins, DNA, RNA, vitamins), ions (Na+, K+, Mg2+, Ca2+, Fe3+
), fibers and gels, will remain in the water layer, and the HMF
will be extracted to MIBK, making the HMF purer. After the re-
action, the crude HMF in MIBK was evaporated to remove
MIBK for reuse, and the raw HMF was purified with our water
extraction method to achieving a light yellow aqueous solu-
tion. The HMF aqueous solution was then used as feedstock
for the Au/HT-catalyzed oxidization reaction. In this process,
HMF was produced in 57% yield in the first step and the over-
all yield for FDCA was 55% (based on the fructose component
in JAT) (Scheme 3).
Reaction procedure to produce HMF from Jerusalem artichoke
tuber (JAT): In a 8 mL sealed glass tube dried JAT powder (0.3 g,
equivalent to 1.25 mmol fructose), HCl saturated with NaCl (1.2 mL,
0.25m), and MIBK (4 mL) were added. The solution was purged
thoroughly with argon gas 3 times until all the air was replaced.
Under magnetic stirring at 700 rpm, the reaction was heated to
1808C in a heating block for 30 min. After the reaction, the solu-
tion was cooled down in an ice bath and centrifuged. The MIBK
layer was taken out for further usage.
General procedure to purify HMF: 1 mmol of HMF original solu-
tion in isopropanol was evaporated at 90 mbar and 408C for
15 min, followed by another 15 min of a continuous evaporation
process to completely remove the solvent. A dark brown viscous
liquid of crude HMF product was obtained. After that, 5 mL of
water was added and rotated at room temperature until the dark
impurities were separated from the water solution. The transparent
yellowish upper solution was taken out and the black impurities
remained on the surface of the bottle wall. Another 5 mL of water
was added and rotated for another 15 min, before the
upper solution was collected. The collected solution was
mixed and centrifuged to remove any remaining residue,
and a transparent yellowish solution was obtained. This
solution was used for further reaction.
Preparation of Au/HT and Au8Pd2/HT catalysts: Au/HT
was prepared according to the method reported in the
literature.[18b] Au8Pd2/HT was prepared following the
same method with slight modifications. Briefly, HAuCl4
(0.1 mmol) and NaPdCl4 (0.025 mmol) were dissolved in
water (40 mL). To this solution, hydrotalcite (1 g) was
added, followed by addition of NH3 aqueous solution (29.5%,
0.425 mL) until pH 10 was reached. The solution was vigorously
stirred for 6 h and refluxed for 30 min at 373 K. The resulting solid
was filtered, washed thoroughly with water, and heated at 473 K
overnight.
Scheme 3. Integrated process for the conversion of JAT biomass to FDCA.
In summary, we have reported a simple and effective water
extraction method to purify HMF obtained from a biomass de-
hydration system. Up to 99% of the HMF could be recovered
and the HMF in aqueous solution could be directly used for
further catalytic oxidization reaction to FDCA, the sole product.
With this purification technique, an integrated process from
fructose to FDCA via HMF prepared in an isopropanol mono-
phasic system, we achieved an overall FDCA yield of 83%. We
also used this process in a successfully demonstration of the
direct conversion of JAT biomass to FDCA via HMF prepared in
a water/MIBK biphasic system, with an overall FDCA yield of
55%.
Catalytic reaction from HMF to FDCA: The reaction was conduct-
ed by using the method reported in the literature[18b] with Na2CO3
as base. With oxygen gas bubbling, the solution was first heated
to 508C for 2 h, and HMF was fully converted to HFCA. After that,
the reaction was heated to 958C and kept for 7 h. The solution
was diluted for HPLC analysis. For product isolation, the aqueous
solution was adjusted to pH 1, and FDCA was precipitated from
the solution. The precipitate was filtered and washed with ethanol.
Product analysis: HMF and FDCA were analyzed by HPLC (Agilent
Technologies, 1200 series) and confirmed with isolation yield. HPLC
working conditions: column (Agilent Hi-Plex H, 7.7ꢁ300 mm,
8 mm), solvent 10 mm H2SO4, flow rate 0.7 mLminÀ1, 258C, UV de-
tector: 280 nm for HMF and 254 nm for FDCA. The retention times
for detected compounds were 20.7 min, 24.4 min, 29.4 min, and
36.5 min for FDCA, HFCA, FFCA, and HMF, respectively. Fructose
Experimental Section
Materials: d-Fructose was purchased from Alfa Aesar. HMF and
FDCA were purchased from Sigma–Aldrich. Anhydrous isopropanol
and hydrogen chloride (37%) were purchased from Merck. All the
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 5
&
4
&
ÞÞ
These are not the final page numbers!

CHEMSUSCHEM
COMMUNICATIONS
www.chemsuschem.org
[13] a) T. F. Wang, Y. J. Pagan-Torres, E. J. Combs, J. A. Dumesic, B. H. Shanks,
Top. Catal. 2012, 55, 657; b) A. Rosatella, A. Dumesic, S. P. Simeonov,
R. F. M. Frade, C. A. M. Afonso, Green Chem. 2011, 13, 754–793.
[14] S. G. Wettstein, D. M. Alonso, Y. Chong, J. A. Dumesic, Energy Environ.
Sci. 2012, 5, 8199–8203.
[15] C. Moreaua, M. N. Belgacemb, A. Gandini, Top. Catal. 2004, 27, 11–30.
[16] A. Gandini, A. J. D. Silvestre, C. P. Neto, A. F. Sousa, M. Gomes, J. Polym.
Sci. Part A 2009, 47, 295–298.
was measured using a sugar analyzer (DKK-TOA Corporation,
Japan. Model: SU-300).
Characterization: The product was characterized by 1H and
13C NMR (Brucker AV-400). Au/HT catalyst was characterized by TEM
(FEI Tecnai F20) and XRD (PANalytical X-ray diffractometer, X’pert
PRO, with CuKa radiation at 1.5406 ꢂ). TEM, XRD, and NMR spectra
are showing in the Supporting Information, Figure S7 and S8.
[17] S. E. Davis, L. R. Houk, E. C. Tamargo, A. K. Datye, R. J. Davis, Catal. Today
2011, 160, 55–60.
[18] a) E. Taarning, I. S. Nielsen, K. Egeblad, R. Madsen, C. H. Christensen,
ChemSusChem 2008, 1, 75–78; b) N. K. Gupta, S. Nishimura, A. Takagaki,
K. Ebitani, Green Chem. 2011, 13, 824–827; c) S. E. Davis, M. S. Ide, R. J.
Davis, Green Chem. 2013, 15, 17–45; d) A. Villa, M. Schiavoni, S. Campisi,
G. M. Veith, L. Prati, ChemSusChem 2013, 6, 609–612; e) M. Krystof, M.
Perez-Sanchez, P. D. de Maria, ChemSusChem 2013, 6, 826–830.
[19] M. Krçger, U. Prusse, K. D. Vorlop, Top. Catal. 2000, 13, 237–242.
[20] M. L. Ribeiro, U. Schuchardt, Catal. Commun. 2003, 4, 83–86.
[21] K. Shimizu, R. Uozumi, A. Satsuma, Catal. Commun. 2009, 10, 1849–
1853.
Acknowledgements
This work was supported by the Institute of Bioengineering and
Nanotechnology (Biomedical Research Council), Biomass-to-
Chemicals program (Science and Engineering Research Council),
Agency for Science, Technology and Research, Singapore.
Keywords: biomass · catalysis · 2,5-furandicarboxylic acid · 5-
hydroxymethylfurfural
[22] a) H. Zhao, J. E. Holladay, H. Brown, Z. C. Zhang, Science 2007, 316,
1597–1600; b) G. Yong, Y. G. Zhang, J. Y. Ying, Angew. Chem. Int. Ed.
2008, 47, 9345–9348; Angew. Chem. 2008, 120, 9485–9488; c) M. X.
Tan, L. Zhao, Y. G. Zhang, Biomass Bioenergy 2011, 35, 1367–1370;
d) J. Y. G. Chan, Y. G. Zhang, ChemSusChem 2009, 2, 731–734; e) L. K.
Lai, Y. G. Zhang, ChemSusChem 2010, 3, 1257–1259; f) S. P. Teong, G. Yi,
X. Cao, Y. G. Zhang, ChemSusChem 2014, DOI: 10.1002/cssc.201402131.
[23] L. K. Lai, Y. G. Zhang, ChemSusChem 2011, 4, 1745–1748.
[24] Y. Romꢃn-Leshkov, J. N. Chheda, J. A. Dumesic, Science 2006, 312, 1933–
1937.
[25] Y. Romꢃn-Leshkov, J. A. Dumesic, Top. Catal. 2009, 52, 297–303.
[26] K. D. O. Vigier, A. Benguerba, J. Barrault, F. Jerome, Green Chem. 2012,
14, 285–289.
[27] P. Vinke, H. Van Bekkum, Starch/Staerke 1992, 44, 90–96.
[28] W. L. Baak, US Patent 3326944, 1967, 1–3.
[1] J. P. Holdren, Science 2007, 315, 737–737.
[2] A. M. Ruppert, K. Weinberg, R. Palkovits, Angew. Chem. Int. Ed. 2012, 51,
2564–2601; Angew. Chem. 2012, 124, 2614–2654.
[3] J. N. Chheda, G. W. Huber, J. A. Dumesic, Angew. Chem. Int. Ed. 2007, 46,
7164–7183; Angew. Chem. 2007, 119, 7298–7318.
[4] G. W. Huber, A. Corma, Angew. Chem. Int. Ed. 2007, 46, 7184–7201;
Angew. Chem. 2007, 119, 7320–7338.
[5] O. R. Inderwildi, D. A. King, Energy Environ. Sci. 2009, 2, 343–346.
[6] Y. G. Zhang, J. Y. G. Chan, Energy Environ. Sci. 2010, 3, 408–417.
[7] J. J. Bozell, Science 2010, 329, 522–523.
[8] J. J. Bozell, G. R. Petersen, Green Chem. 2010, 12, 539–554.
[9] E. L. Kunkes, D. A. Simonetti, R. M. West, J. C. Serrano-Ruiz, C. A. Gartner,
J. A. Dumesic, Science 2008, 322, 417–421.
[10] S. Dutta, S. De, B. Saha, ChemPlusChem 2012, 77, 259–272.
[11] a) R. J. van Putten, J. C. van der Waal, E. de Jong, C. B. Rasrendra, H. J.
Heeres, J. G. de Vries, Chem. Rev. 2013, 113, 1499–1597; b) S. P. Teong,
G. Yi, Y. G. Zhang, Green Chem. 2014, 16, 2015–2026; c) J. Wang, J. Ren,
X. Liu, G. Lu, Y. Wang, AIChE J. 2013, 59, 2558–2566.
[29] B. N. Zope, S. E. Davis, R. J. Davis, Top. Catal. 2012, 55, 24–32.
[30] O. Casanova, S. Iborra, A. Corma, ChemSusChem 2009, 2, 1138–1144.
[31] a) L. K. Zhou, A. Q. Wang, C. Z. Li, M. Y. Zheng, T. Zhang, ChemSusChem
2012, 5, 932–938; b) R. N. Razmovski, M. B. Sciban, V. M. Vucurovic,
Rom. Biotechnol. Lett. 2011, 5, 6497–6503.
Received: February 25, 2014
Revised: April 9, 2014
Published online on && &&, 0000
[12] J. M. R. Gallo, D. M. Alonso, M. A. Mellmer, J. A. Dumesic, Green Chem.
2013, 15, 85–90.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 5
&
5
&
ÞÞ
These are not the final page numbers!

COMMUNICATIONS
G. Yi, S. P. Teong, X. Li, Y. Zhang*
Hell hath no furan: A simple and effec-
tive water extraction method to purify
5-hydroxylmethylfurfural (HMF) ob-
tained from biomass dehydration
&& – &&
Purification of Biomass-Derived 5-
Hydroxymethylfurfural and Its
Catalytic Conversion to 2,5-
Furandicarboxylic Acid
system is reported. Up to 99% of the
HMF can be recovered and the aqueous
solution of HMF can be directly used for
further catalytic oxidization reaction to
2,5-furandicarboxylic acid (FDCA) as the
sole product. This purification technique
allows an integrated process to produce
FDCA from fructose via HMF
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 5
&
6
&
ÞÞ
These are not the final page numbers!