Catalytic behaviour of TiO2-ZrO2 binary oxide synthesized by sol-gel process for glucose conversion to 5-hydroxymethylfurfural

Article abstract of DOI:10.1039/c5ra15739k

The catalytic application of as-synthesized TiO₂-ZrO₂ binary oxides was examined for the conversion of glucose to produce 5-hydroxymethylfurfural (HMF). Highest HMF yield (74%) at glucose concentration of 5 wt% was obtained with TiO₂-ZrO₂ (1/1) and Amberlyst 70 catalyst system in a water-THF biphasic reaction system. Notably, a much higher HMF yield (86%) was achieved when the organic phase of the biphasic system was replaced with dioxane. The increased product yield may be ascribed to the role of dioxane as an aqueous phase modifier that stabilizes HMF in the reactive phase as well as promotes partitioning of HMF into the extractive layer. Furthermore, the combined catalyst and biphasic solvent systems were also effective for the conversion of glucose polymers to HMF.

Full text of DOI:10.1039/c5ra15739k

View Article Online
View Journal
RSC Advances
This article can be cited before page numbers have been issued, to do this please use: L. Atanda, A.
Silahua, S. Mukundan, A. Shrotri, G. Torres-Torres and J. Beltramini, RSC Adv., 2015, DOI:
10.1039/C5RA15739K.
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. This Accepted Manuscript will be replaced by the edited,
formatted and paginated article as soon as this is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the Ethical guidelines still
apply. In no event shall the Royal Society of Chemistry be held
responsible for any errors or omissions in this Accepted Manuscript
or any consequences arising from the use of any information it
contains.
www.rsc.org/advances

Please do not adjust margins
RSC Advances
Page 1 of 8
View Article Online
DOI: 10.1039/C5RA15739K
RSC Advances
ARTICLE
Catalytic behaviour of TiO2–ZrO2 binary oxide synthesized by sol-
gel process for glucose conversion to 5-hydroxymethylfurfural
Luqman Atanda,a Adib Silahua,b Swathi Mukundan,a Abhijit Shrotri,c Gilberto Torres-Torresb and
Jorge Beltraminia*
Received 00th January 20xx,
Accepted 00th January 20xx
DOI: 10.1039/x0xx00000x
The catalytic application of as-synthesized TiO2–ZrO2 binary oxides was examined for the conversion of glucose to produce
5-hydroxymethylfurfural (HMF). Highest HMF yield (74%) at glucose concentration of 5 wt% was obtained with TiO2–ZrO2
(1/1) and Amberlyst 70 catalyst system in a water-THF biphasic reaction system. Notably, a much higher HMF yield (86%)
was achieved when the organic phase of the biphasic system was replaced with dioxane. The increased product yield may
be ascribed to the role of dioxane as an aqueous phase modifier that stabilizes HMF in the reactive phase as well as
promotes partitioning of HMF into the extractive layer. Furthermore, the combined catalyst and biphasic solvent systems
were also effective for the conversion of glucose polymers to HMF.
www.rsc.org/
(ZrO2) have been reported to possess exceptional redox and
acid-base properties, which makes them a good choice as
catalysts and catalyst supports.21 Moreover, catalytic property
1. Introduction
Diversification of energy sources to reduce dependence on
fossil fuels has motivated a strong research interest to attain
sustainability in biomass conversion to transportation fuels
and chemical building blocks.1-6 For example, 5-
hydroxymethylfurfural (HMF), which is a dehydration product
of biomass-derived carbohydrates, has been identified as a
of these oxides can be improved by mixing them together. This
improvement is attributable to the generation of new catalytic
sites due to strong interaction between the individual oxides,
giving rise to a mixed metal oxide of profound surface acid-
base properties and high thermal stability.21 Thus, the superior
properties of TiO2–ZrO2 binary oxide make it more suitable for
catalytic applications than individual component oxide.22-25
However, limited reports exist on the catalytic potential of
TiO2–ZrO2 binary oxide to transform biomass-derived
carbohydrates to HMF. Chareonlimkun and co workers26
explored simultaneous hydrolysis/dehydration of a variety of
lignocellulosic biomass under hot compressed water. They
found that superior product yield was achieved with TiO2–ZrO2
binary oxide in comparison with TiO2 and ZrO2, which was
ascribed to the synergy between the base sites of ZrO2 and
acid sites of TiO2. Furthermore, catalyst preparation technique
and calcination temperature were two parameters observed to
influence the acid-base properties of the oxides, hence their
catalytic reactivity.
versatile intermediate for value-added chemicals and
biofuels.7,
8
Conventionally, HMF is produced by triple
dehydration of hexoses in an acidic medium.9 Throughout
these studies, it is generally accepted that glucose is less
reactive than fructose because of its stable glucopyranose
structure.10,
11
However, glucose is a preferred starting
material since it is cheaper and relatively available in
13
Significant improvement on glucose
abundance.12,
conversion to HMF can be attained by performing the reaction
in the presence of a metal halide, which catalyzes the
isomerization of glucose to fructose, in tandem with mineral
15
acids14,
or acidic ionic liquid.16-19 In spite of these recent
achievements, efficient isolation and purification of HMF from
the reaction medium remains a challenge.20 On the other
hand, heterogeneous catalysis is a promising cost-effective
route for large scale HMF synthesis due to ease of catalyst
handling, simple separation and recovery steps.
In this study, we prepared a series of TiO2–ZrO2 binary
oxides by a neutral amine sol-gel technique. The effect of
TiO2/ZrO2 weight ratio on the acid-base properties of the
binary oxides was examined by temperature programmed
desorption of NH3 and CO2. Furthermore, the pyhsico-chemical
properties of the oxide materials were investigated with N2
adsorption-desorption, XRD, Raman and XPS techniques. We
evaluated their catalytic performance in transforming glucose
to HMF using water as the reaction solvent. In addition, the
cooperative role of a co-solvent and a solid Bronsted acid co-
catalyst was examined. Thereafter, the robustness of the
Several metal oxides, such as titania (TiO2) and zirconia
a.Nanomaterials Centre, Australian Institute for Bioengineering & Nanotechnology
and School of Chemical Engineering, The University of Queensland, Brisbane, St.
Lucia 4072, Australia. E-mail: j.beltramini@uq.edu.au
b.Heterogeneous Catalysis Laboratory, Universidad Juárez Autónoma de Tabasco,
Cunduacán, Tabasco, México.
c. Catalysis Research Center, Hokkaido University, Saporro, Japan.
† Electronic Supplementary Information (ESI) available: supplemented data. See
DOI: 10.1039/x0xx00000x
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 1
Please do not adjust margins

Please do not adjust margins
RSC Advances
Page 2 of 8
View Article Online
DOI: 10.1039/C5RA15739K
ARTICLE
RSC Advances
reaction system was extended to other sugar substrates such mg) at 150 °C. Excess gaseous and physisorbed pyridine were
as cellobiose, sucrose and cellulose.
removed by holding the temperature at 150 °C for 30 min
under N2 flow. FTIR spectra were collected using a Nicolet
6700 (Smart Orbit Accessory) at 128 scans and 4 cm–1
resolution. X-ray photoelectron spectra (XPS) of the catalysts
was acquired using Kratos Axis ULTRA X-ray photoelectron
spectrometer equipped with a 165 mm hemispherical electron
energy analyser using a monochromatic Al Kα (1486.6 eV) x-
ray source. The binding energies were referenced to C 1s peak
at 284.8 eV. Raman spectroscopy was performed on a
Reinshaw In-Via Raman microscope equipped with a Leica DM
LM Microscope using a 633 nm HPNIR diode laser as an
excitation source. The raman spectra were collected with a
CCD array detector in the region of 100–2000 cm–1 at 4 cm–1
resolution and acquisition time of 1 s.
2. Experimental
2.1. Chemicals and materials
Glucose, fructose, sucrose, cellobiose, cellulose, 5-
hydroxymethylfurfural (HMF), levoglucosan, titanium IV
butoxide, zirconium IV butoxide, ammonium hydroxide
solution (28 wt%), phosphotungstic acid (H3PW12O4),
silicotungstic acid (H4SiW12O4), propan-1-ol, butan-1-ol, Methyl
isobutyl ketone (MIBK) and Nafion NR50 were obtained from
Sigma-Aldrich. Tetrahydrofuran and 1,4-dioxane were
obtained from Merck Millipore. Amberlyst 70 was supplied by
Roms and Haas. All the chemicals were used without further
purification. Ultrapure water (18.2 MΩ cm-1) from Elga
distillation system was used for all the experiment.
2.5. Catalytic activity procedure and product analysis
Glucose conversion to HMF was carried out in a Parr reactor.
In a typical experimental run, required amount of glucose,
solvent and catalysts were charged into a 300 mL reactor,
purged with high purity Ar and then pressurized to 30 bar. The
reactor was ramped to its set point temperature which was
monitored by a thermocouple inside a thermowell immersed
2.2. Preparation of TiO2–ZrO2 binary oxides
TiO2, ZrO2 and a series of TiO2–ZrO2 with varying weight
percentage of TiO2 to ZrO2 were prepared by sol-gel method.
The sol was prepared by the dropwise addition of the alkoxide
precursors into an aqueous solution containing butan-1-ol
under stirring. The pH of the solution was adjusted to 7 by in the reaction mixture. Time zero of the reaction was defined
adding ammonium hydroxide solution and the resulting
solution was maintained at 70 °C under reflux for 24 h. After
gelation, excess solvent was removed using rotary evaporator
and followed by vacuum drying at 120 °C for 12 h. The dried
samples were then calcined at 500 °C for 12 h to obtain the
corresponding pure or mixed oxides.
as the time when the reactor reached its set point
temperature usually after approx. 9 mins of ramping. Catalytic
runs were repeated for data reproducibility. Experimental
errors were in the range of ±2%. Liquid samples were taken
after reaction and the concentration of the product species
were quantified using Shimadzu Prominence HPLC with a Bio-
Rad Aminex HPX-87H as the analytical column and both RID–
10 (refractive index) and SPD–M20A (UV–Vis) as detectors. The
HPLC was operated under the following conditions: oven
temperature, 50 °C, mobile phase, 5 mM H2SO4; flow rate, 0.6
ml/min; injection volume, 20 µL. The concentrations of
glucose, fructose, HMF and other identifiable products were
quantified by HPLC analysis through the external standard
method and calibration curves of commercially available
standard substrates. Glucose conversion (Conv. mol %) and
products yield (mol %) were calculated according to:
2.3. Preparation of heteropolyacid salts
Cs3.5SiW
- Cs2.5PW and
Cesium salt of H3PW12O4 and H4SiW12O4 were obtained by
titrating an aqueous solution of H3PW12O4 or H4SiW12O4 (0.08
M) with an aqueous solution of Cs2CO3 (0.25 M) while stirring.
The resulting precipitate was dried at 110 °C for 12 h in
vacuum and then calcined at 300 °C for 3 h.
2.4. Catalyst characterization
Powder XRD measurements were performed on a Rigaku
Miniflex with a monochromatic CoKα radiation (30 kV, 15 mA)
in the 2θ range 10–90 °. Nitrogen adsorption data were
collected using a Micromeritics TriStar II 3020 surface area and
porosity analyzer. Prior to physisorption measurements, all
samples were outgassed under vacuum at 200 °C overnight.
The specific surface area of the oxides was determined
applying the BET method. Total pore volumes were estimated
from the amount of adsorbed N2 at p/po value of 0.99.
Temperature programmed desorption (TPD) of ammonia and
CO2 were performed on a BEL Japan BELCAT-A instrument
equipped with a mass spectrometer and thermal conductivity
detector (TCD). This was used to determine the amount and
strength of the acid-base sites available on the catalyst
samples. About 70 mg of sample was saturated with either
NH3 or CO2 as the probe gas at 100 °C, flushed with He to
remove physisorbed gas and then ramped to 800 °C at a
heating rate of 10 °C/min under He flow. Pyridine infrared
spectroscopy was used to determine the nature of surface acid
sites. Pyridine was chemisorbed on the catalyst surface (50




nC
6
o
6
Conv.
(mol%
)
= 1 −
×100%






nC


n


i
Pr oduct yield
(mol%
)
=
×100%






o
nC
6
where nC6 and nC6o denote number of moles of
product and feed, respectively, and is the number of moles
C
6 sugar in the
n
i
of identified products (HMF, levulinic acid, levoglucosan etc.).
3. Results and discussion
3.1. Characterization of TiO2–ZrO2 binary oxides
2 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
RSC Advances
Page 3 of 8
View Article Online
DOI: 10.1039/C5RA15739K
RSC Advances
ARTICLE
The crystalline phases of TiO2, ZrO2, and TiO2–ZrO2 binary
oxides were investigated using XRD analysis. The XRD pattern
of the oxides is shown in Fig. 1.
Table 1 also reports the acid–base property of the oxides measured
by temperature programme desorption (TPD). From the table it can
be seen that as the ZrO2 content increases, acidity of the oxides
reduces with a corresponding increment in basicity. Furthermore,
the distribution and strength of the acid and base sites is shown in
Fig. S1. Acid28 and base29 strength is categorized as low, medium
and strong, depending on desorption temperature. The CO2–TPD
profile of ZrO2 (Fig. S1a) shows three peaks centered at 185 °C, 550
°C and 610 °C, which can be attributed to weak and strong basic
sites, respectively. Meanwhile, TiO2 is predominantly characterized
with a small broad peak centered at 600 °C corresponding to strong
basic site. TiO2–ZrO2 (1/1) has a CO2 desorption profile defined by
two well resolved peak signals centered at 170 °C and 550 °C,
similar to that of ZrO2. On the other hand, ammonia desorption
profile of the oxides is shown in Fig. S1b. TiO2 shows a large
concentration of weak-to-medium strength acid sites, whereas ZrO2
is characterized by strong acid sites albeit weak concentration.
TiO2–ZrO2 (1/1) shows a composite profile of high concentration of
weak-to-medium acid sites and low concentration of strong acid
sites. The nature of the acid sites available on the oxides was
identified through infrared spectra of adsorbed pyridine. As shown
in Fig. 2, the peak intensities observed indicates the oxides are
characterized by Lewis acid sites. Pure TiO2 displays intense bands
at 1446 cm–1 and 1608 cm–1 with moderate bands at 1489 cm–1 and
Fig. 1 XRD patterns of: a) TiO2, b) TiO2–ZrO2 (3/1), c) TiO2–ZrO2
(1/1), d) TiO2–ZrO2 (1/3) and e) ZrO2. A: anatase; M: monoclinic; T:
tetragonal
The indexed diffraction pattern of TiO2 revealed that it exists as
anatase phase, whereas for ZrO2, a combination of monoclinic (m-
ZrO2) and tetragonal (t-ZrO2) phases were identified, corresponding
to 2θ values of 28.1 ° and 35.1 °, respectively. However, in the case
of the binary oxides, regardless of the TiO2/ZrO2 ratio, Fig. 1b-d
show a composite diffraction pattern that corresponds to the
anatase-TiO2 (2θ = 29.5 °) and t-ZrO2 (2θ = 35.1 °) phases. The m-
ZrO2 phase ceases to exist even with the smallest TiO2/ZrO2 ratio i.e.
1:3. This indicates that t-ZrO2 phase was stabilized due to the
presence of Ti atom in ZrO2 lattice, which results in dissimilar grain
boundaries that prevents the formation and crystal growth of the
monoclinic phase.27 Moreover, no diffraction peaks related to
zirconium titanate phase could be detected. It may be that this
phase is composed of very small particles below detection limit or
could exist in an amorphous state. Surface area and pore volume as
determined by N2-adsorption measurement are summarized in
Table 1. The surface areas of TiO2 and ZrO2 are 79.4 m2/g and 40.8
m2/g, respectively. ZrO2 has the lowest surface area of all the
oxides. Surface areas of the binary oxides gradually increase as it
becomes enriched with TiO2.
1575 cm–1
.
Fig. 2 Infrared spectra of pyridine adsorbed on: a) TiO2, b) TiO2–ZrO2
(1/1) and c) ZrO2
Table 1 Textural and acid-base properties of the mixed oxides
Meanwhile, pure ZrO2 showed fewer and less intense bands,
which agrees well with NH3–TPD result that ZrO2 has fewer
acid sites than TiO2. TiO2–ZrO2 (1/1) mixed oxide has peak
reflections similar to TiO2, though with slightly reduced
intensity. Moreover, it has an extra peak at 1590 cm–1, which is
a reflection of ZrO2.
a
Catalyst
SBET
Pore
volb
Acid site
conc.c
(µmol/g)
476.6
469.8
452.3
Basic site
conc.d
(µmol/g)
12.1
16.7
23.7
(m2/g)
(cc/g)
0.33
0.27
0.18
0.13
0.12
TiO2
79.4
68.0
62.0
60.0
40.8
TiO2-ZrO2 (3/1)
TiO2-ZrO2 (1/1)
TiO2-ZrO2 (1/3)
ZrO2
Surface analysis of the oxides was examined by Raman and
XPS. Raman spectra of all the oxides are shown in Fig. 3. Fig. 3a
depicts Raman spectrum of TiO2 is dominated by vibrational
bands at 144, 400, 515 and 635 cm–1, which can be attributed
to ν1(Eg), ν2(B1g), ν3(A1g) or ν3(B1g) and ν4(Eg), respectively, of
anatase-TiO2 phase.30 ν1(Eg) and ν2(B1g) represent the O–Ti–O
372.2
202.1
27.5
34.2
a Specific surface area calculated by the BET method. b Pore
vol calculated at P/Po = 0.99. c Calculated from ammonia–TPD.
d Calculated from CO2–TPD
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

Please do not adjust margins
RSC Advances
Page 4 of 8
View Article Online
DOI: 10.1039/C5RA15739K
ARTICLE
RSC Advances
bending vibrational mode, whereas ν3(A1g) or ν3(B1g) and ν4(Eg) alkoxy groups originating from the sol-gel process. The binding
represent the Ti–O bond stretching vibrational mode.31
energies associated with Ti 2p3/2, Zr 3d5/2 and O 1s are also
reported in Table 2. Binding energies of 458.6 eV and 182.2 eV
corresponds to Ti 2p3/2 and Zr 3d5/2 of TiO2 and ZrO2 single
oxide, respectively. O 1s peak of TiO2 and ZrO2 are assigned
529.9 eV and 529.8 eV binding energies, respectively. These
values confirm that both Ti and Zr exist in the tetravalent
oxidation state.32 As for the binary oxides, Ti 2p3/2 has similar
binding energy as that of pure TiO2 while Zr 3d5/2 is about 0.1–
0.2 eV lower than that of pure ZrO2. This shift towards lower
binding energy may indicate the substitution of some Zr4+ ions
with Ti4+, thereby causing reduction of interatomic potentials
due to reduced overall atomic size.33
3.2. Activity tests
Catalytic application of the metal oxide system was examined
for the conversion of biomass-derived sugars into HMF, using
glucose as
a model compound. Initial reactivity study
conducted with water as the reaction medium was used to
evaluate catalytic performance of the oxides. Dependence of
glucose conversion capacity and products yield in relation to
the composition of TiO2/ZrO2 binary oxide is shown in Fig. 4. It
can be seen from the Figure that glucose conversion ranges
from 72–86%, with HMF as the main product. With increase in
ZrO2 content of the binary oxide, glucose conversion decreases
whereas yield of HMF increases and goes through a maximum
at TiO2/ZrO2 (1:1) followed by a minor decline.
Fig. 3 Raman spectra of: a) TiO2, b) TiO2–ZrO2 (3/1), c) TiO2–
ZrO2 (1/1), d) TiO2–ZrO2 (1/3) and e) ZrO2
Fig. 3b through 3d shows that the intensity of these bands
attenuates gradually as the binary oxide becomes enriched
with ZrO2. Elemental surface compositions and binding
energies from the XPS analysis are listed in Table 2. The
changes in the surface composition of Ti and Zr atoms are
consistent with the observed pattern in the Raman intensities
of the oxides, wherein elemental composition of Ti declines as
ZrO2 content is increased. This indicates mutual interaction
between the oxides as a result of Ti4+ substitution with Zr4+ and
vice versa, which is also accompanied by structural changes as
depicted by the variations in the Raman signals of the metal
oxides.
Table 2 Binding energies and surface composition of TiO2–ZrO2
mixed oxides from XPS
Sample Surface atoms (%)
BE (eV)
O/(Ti+Zr) O 1s
Ti
Zr
C
Ti
Zr
3d5/2
-
2p3/2
TiO2
20.3
17.8
-
33.4
2.28
2.21
529.9 458.6
Fig. 4 Glucose conversion to HMF over TiO2–ZrO2 binary oxides
in an aqueous medium. Reaction conditions: 2 g glucose, 0.8 g
catalyst wt., 3 h reaction time, 175 °C reaction temperature,
TiO2
ZrO2
(3/1)
TiO2
ZrO2
(1/1)
TiO2
ZrO2
(1/3)
ZrO2
-
-
-
3.3 32.2
7.5 34.5
15.5 29.6
22.3 29.0
529.9 458.6 182.0
529.9 458.6 182.1
529.9 458.6 182.2
100 ml water. ( ), glucose conversion and product yields: (
fructose, ( ) HMF, ( ) levulinic acid.
)
13.0
7.1
-
2.20
2.12
2.18
This occurrence may be interpreted in relation to the acidity of
the binary oxides. TiO2 possesses large concentration of acid
sites that can dehydrate fructose, an initial product of glucose
isomerization, into HMF. Hence, least concentration of
fructose is observed on TiO2 but the value rises gradually with
declining acidity of the binary oxide due to increase in ZrO2
content. More so, in an aqueous acidic medium, HMF
rehydrates to give levulinic acid and may also react with
529.8
-
182.2
Significant carbon atom was also identified on all the oxides.
This carbon impurity may be understood to come from the
4 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
RSC Advances
Page 5 of 8
View Article Online
DOI: 10.1039/C5RA15739K
RSC Advances
ARTICLE
route for glucose conversion to HMF is via intermediate
fructose formation, which is otherwise promoted by Lewis acid
glucose or other reactive intermediates to form humins.
Therefore, as TiO2 content of the binary oxide increases, which
correlates to increasing acidity, there is an upward trend of or base isomerisation of glucose.
both the yield of levulinic acid and glucose conversion as
shown in Fig. 5. To support this hypothesis, another reaction
was conducted at 160 °C for 5 h with fructose as feed but
keeping other reaction conditions same. As shown in Table S1,
fructose conversion of 95.3% and 85.6% was attained on TiO2
and ZrO2, respectively. Even though fructose conversion on
TiO2 was superior, a better yield of HMF was achieved on ZrO2
(29.6 vs 24.1%). Regardless of the pure metal oxide, it is
noteworthy that fructose conversion to HMF occurred
relatively faster in comparison to glucose. This indicates that
glucose proceeds via fructose to produce HMF. From Fig. 4, an
optimum HMF yield of 23.6% is obtained with TiO2–ZrO2 (1/1),
which can be ascribed to modified acid–base properties of the
binary oxide. The mutual interaction of the pure oxides
provided
a moderate acid–base concentration (Table 1)
suitable for the glucose-to-HMF reaction. This cooperative
interaction between the acid–base sites on TiO2–ZrO2 has been
reported in literature, wherein glucose is isomerized to
fructose preferentially on the basic sites of ZrO2 and
subsequent dehydration of fructose to HMF on the acid sites
of TiO2.26, 34, 35
Fig. 5 Catalytic conversion of glucose over TiO2–ZrO2 with solid
acid co-catalysts. Reaction conditions: 2 g glucose, 0.8 g
catalyst wt. (TiO2–ZrO2(1/1)/co-catalyst ratio = 1/1 w/w), 100
ml solvent (THF/water = 4/1 v/v), 4 g NaCl, 3 h reaction time,
175 °C reaction temperature.
Next, we examined the role of reaction medium on the
formation rate of HMF. A water-lean medium was considered
since the yield of HMF in a single aqueous medium is relatively
low. Tetrahydrofuran (THF) was chosen as a co-solvent. It lacks
hydroxyl groups and has the potential to suppress side
HMF yield varies with the co-catalysts as follows: Amberlyst 70
(85.6%) > Nafion NR50 (73.4%) > Cs3.5SiW (35.2%) > Cs2.5PW
(31.8%). The observed variation can be related to the acid
strength of the co-catalysts. H3PW12O40 and H4SiW12O40 are
reactions 36
In spite of adding THF to water in a ratio of 4:1
.
superacids of very strong acid strength
(H0 =
–13.6).39
(v/v) to make up the reaction medium, glucose conversion and
yield of HMF were not significantly enhanced as shown in Fig.
S2. Contrarily, a remarkable increment of HMF yield by almost
3 fold from 23.6% to 71% was observed after the introduction
of NaCl (20 wt% of the aqueous phase) into the THF/water
solvent mix. The addition of NaCl promoted solvent
partitioning into two phases: a reactive aqueous phase and an
extractive organic phase. This allowed in situ extraction of
HMF from the aqueous phase into the organic phase, thereby
preventing undesired side reactions. Moreover, the
continuous extraction shifts the reaction equilibrium to
produce more HMF. Furthermore, we investigate the role of
solid Bronsted acid supports such as Amberlyst 70, Nafion
NR50, Cs2.5PW and Cs3.5SiW as co-catalysts, as it has been
observed that co-existence of Lewis and Bronsted bifunctional
Nevertheless, H3PW12O40 is slightly more acidic than
H4SiW12O40.40 Cs2.5PW has acid strength similar to the parent
phosphotungstic acid41 and correspondingly, Cs3.5SiW is
anticipated to possesses weaker acid strength compared to
Cs2.5PW. Hammett acidity function, H0, of Nafion NR50 is –12,42
whereas Amberlyst 70 has a similar measure of acidity as
Amberlyst 35 with a H0 value of –5.6.43 From the H0 values,
Amberlyst 70 possesses the lowest Bronsted acidity. We may
infer that there is a linear correlation between Bronsted acid
strength and yield of HMF. The stronger the Bronsted acidity,
the greater the propensity of HMF to degrade to huminic
compounds because of the strong affinity of HMF onto the
Bronsted acid sites.44, 45 For this reason, the pair of TiO2–ZrO2
(1/1) and Amberlyst 70 appears to be the optimum catalytic
system for the transformation of glucose to HMF. Effectiveness
of the catalytic system was further evaluated for HMF
synthesis at a higher initial glucose concentration. As shown in
Fig. S6, we observed a reduction in HMF yield as the initial
glucose concentration was increased from 2 to 5 wt%, which
can be ascribed to polymerization and cross-polymerization of
HMF to humins.46-48 Visually, we observed fine dark-brown
powder deposit on the reactor wall, which was more
prominent at high glucose concentration. We speculate that
the choice of organic solvent of the biphasic system may help
reduce the severity of humin compounds formation. Hence,
we decided to replace THF with other organic solvents such as
38
acidity promote glucose conversion to HMF.37, Herein, co-
catalysts refer to physical mixture of TiO2–ZrO2 (1/1) and the
solid Bronsted acid supports. Under comparable reaction
conditions, Fig. 5 clearly demonstrates that the choice of
Amberlyst 70 as co-catalyst is the most beneficial, resulting in
HMF yield of about 85.6 % at a near-complete glucose
conversion. Meanwhile, using the solid Bronsted acid supports
without the binary oxide gave lesser HMF yields. For instance,
approximately 10% yield HMF was achieved with Cs3.5SiW at
complete glucose conversion. This indicates that the dominant
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 5
Please do not adjust margins

Please do not adjust margins
RSC Advances
Page 6 of 8
View Article Online
DOI: 10.1039/C5RA15739K
ARTICLE
RSC Advances
1-butanol, 1-propanol, methyl-isobutyl-ketone (MIBK) and 1,4- cellobiose, sucrose and cellulose. The result of HMF synthesis
dioxane. As shown in Fig. 6, dioxane was the most effective, from these various sugar substrates is given in Table 3.
selectively producing approximately 86% yield of HMF.
Table 3 Catalytic transformation of sugars to HMFa
Entry
Substrate
Temp.
°C
Conversion
(%)
HMF yield
(%)
1
2
3
4
Glucose
Sucrose
Cellobiose
Cellulose
175
180
180
180
99.9
>99
>99
42.1
85.9
86.5
80.8
25.5
a
Reaction conditions: 2 g substrate, solvent volume 100 mL
(THF/water = 4/1), 0.8 g catalyst wt. (TiO2–ZrO2/Amberlyst 70 = 1), 3
h reaction time, 30 bar Ar pressure.
As observed in Table 3, HMF yield from cellobiose and sucrose
is above 80%, nevertheless, sucrose gave a better yield. This is
because sucrose is a dimer of glucose-fructose molecules,
which makes it more reactive than cellobiose, a dimer of
glucose-glucose molecules. As anticipated, cellulose was found
to be the least reactive with HMF yield of 25.5%. This is
attributable to the strong intermolecular hydrogen bonding of
cellulose. A higher temperature such as 200 °C or greater is
required for reactivity of cellulose to improve due to increased
hydrothermolysis.52 However, Amberlyst 70 has relatively poor
thermal stability at such high temperature, thereby restricting
its usage to typically 190 °C.53
Fig. 6 Influence of organic solvent in the biphase system on
selective HMF yield with TiO2–ZrO2 (1/1) and Amberlyst 70
catalysts. Reaction conditions: 5 g glucose, 2 g catalyst weight
(TiO2-ZrO2/Amberlyst = 1/1 w/w), 100 ml solvent (THF/water
= 4/1 v/v), 4 g NaCl, 3 h reaction time, 175 °C reaction
temperature. (●) Glucose conversion, ( ) HMForg and (
HMFaq.
)
Not only does dioxane enables good partitioning of HMF into
the organic phase, but HMF concentration in the reactive
aqueous phase is also highest, suggesting HMF is more stable
in the water-dioxane biphase system. Although the cause of
this phenomenon is unclear, we consider that the reactive
phase was modified by dioxane thereby minimizing
degradation of HMF into humins. The outcome of HMF
production using water-dioxane biphasic reaction system is
promising and provides a step further towards achieving an
efficient reaction medium for scaling-up. But dioxane poses a
degree of health risk due to its carcinogenic activity,49 which
must be taken into consideration. More so, replacing
homogeneous catalysts with solid ones facilitate product
separation and solvent recovery, which further improves the
attractiveness of biphasic system for industrial application. For
instance, we contrasted the catalytic performance of our
catalyst system with those reported in literature. Gallo et al.36
4. Conclusions
We demonstrated that TiO2–ZrO2 (1/1) binary oxide is an
effective catalyst for the production of HMF from glucose. A
water–THF biphasic system remarkably enhanced HMF
formation rate. Furthermore, co-addition of Amberlyst 70 to
the reaction system selectively improved the formation of
HMF. We also show that at relatively high glucose loading (5
wt%), good yield of HMF is achievable. This was further
improved by replacing the organic solvent of the biphase
system with dioxane, due to its added advantage of modifying
the aqueous phase to minimize degradation of HMF to
humins. HMF was also obtained in high yields via the
transformation of cellobiose and sucrose. Meanwhile, direct
conversion of cellulose to HMF was more difficult due to its
poor reactivity. Lastly, we conclude that a single multi-
functional solid catalytic system and an organic solvent of high
partitioning coefficient that can also serve as a phase modifier
advance the realization of biphasic reaction system for large
scale HMF synthesis.
reported HMF yields of 59%, 55% and 63% using
combination of Amberlyst 70 and Sn-Beta with γ-
valerolactone, γ-hexalactone and tetrahydrofuran,
a
respectively, as organic solvents of a water-organic biphasic
system. Yang et al.50 demonstrated the conversion of glucose
with AlCl3 and HCl using THF as a co-solvent with water to give
up to 62% HMF yield. Yang et al.51 reported 53% yield of HMF
using a combination of Sn-Beta and HCl to catalyze glucose
dehydration in water with THF as extracting solvent.
Acknowledgements
The authors gratefully acknowledge the funding from the
Sugar Research Australia (SRA) and ARC Centre of Excellence
for Functional Nanomaterials at The University of Queensland,
Australia. We also acknowledge the facilities and the scientific
and technical assistance of the Australian Microscopy &
Lastly, the catalytic system of TiO2–ZrO2 mixed oxide and
Amberlyst 70 was examined for the synthesis of HMF using
6 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
RSC Advances
Page 7 of 8
View Article Online
DOI: 10.1039/C5RA15739K
RSC Advances
ARTICLE
28 M. Paul, N. Pal, P. R. Rajamohanan, B. S. Rana, A. K. Sinha
and A. Bhaumik, Phys. Chem. Chem. Phys., 2010, 12, 9389-
9394.
Microanalysis Research Facility at The University of
Queensland.
29 G. V. Sagar, P. V. R. Rao, C. S. Srikanth and K. V. Chary, J.
Phys. Chem. B, 2006, 110, 13881-13888.
Notes and references
30 W. F. Zhang, Y. L. He, M. S. Zhang, Z. Yin and Q. Chen, J Phys
D Appl Phys, 2000, 33, 912-916.
1
2
3
G. W. Huber, S. Iborra and A. Corma, Chem. Rev., 2006, 106,
31 T. Ohsaka, F. Izumi and Y. Fujiki, J. Raman Spectrosc., 1978, 7,
4044-4098.
G. W. Huber and A. Corma, Angew. Chem. Int. Ed., 2007, 46
7184-7201.
J. N. Chheda, G. W. Huber and J. A. Dumesic, Angew. Chem.
,
321-324.
32 S. Polisetti, P. A. Deshpande and G. Madras, Ind. Eng. Chem.
Res., 2011, 50, 12915-12924.
Int. Ed., 2007, 46, 7164-7183.
F. Cherubini, Energy Convers. Manage., 2010, 51, 1412-1421.
J. C. Serrano-Ruiz, R. Luque and A. Sepulveda-Escribano,
33 H.-R. Chen, J.-L. Shi, W.-H. Zhang, M.-L. Ruan and D.-S. Yan,
Chem. Mater., 2001, 13, 1035-1040.
4
5
34 M. Watanabe, Y. Aizawa, T. Iida, R. Nishimura and H.
Inomata, Appl. Catal. A, 2005, 295, 150-156.
Chem. Soc. Rev., 2011, 40, 5266-5281.
6
7
R. A. Sheldon, Green Chem., 2014, 16, 950-963.
J. J. Bozell and G. R. Petersen, Green Chem., 2010, 12, 539-
554.
35 M. Watanabe, Y. Aizawa, T. Iida, T. M. Aida, C. Levy, K. Sue
and H. Inomata, Carbohydr. Res., 2005, 340, 1925-1930.
36 J. M. R. Gallo, D. M. Alonso, M. A. Mellmer and J. A. Dumesic,
Green Chem., 2013, 15, 85-90
8
9
Y. Román-Leshkov, C. J. Barrett, Z. Y. Liu and J. A. Dumesic,
Nature, 2007, 447, 982-985.
37 E. Nikolla, Y. Román-Leshkov, M. Moliner and M. E. Davis,
R.-J. van Putten, J. C. van der Waal, E. de Jong, C. B.
Rasrendra, H. J. Heeres and J. G. de Vries, Chem. Rev., 2013,
113, 1499-1597.
ACS Catal., 2011,
38 L. Atanda, S. Mukundan, A. Shrotri, Q. Ma and J. Beltramini,
ChemCatChem, 2015, , 781-790.
1, 408-410.
7
10 J. N. Chheda, Y. Roman-Leshkov and J. A. Dumesic, Green
Chem., 2007, , 342-350.
11 A. Osatiashtiani, A. F. Lee, D. R. Brown, J. A. Melero, G.
Morales and K. Wilson, Catal. Sci. Technol. 2014, , 333-342.
39 R. Palkovits, K. Tajvidi, A. M. Ruppert and J. Procelewska,
Chem. Commun., 2011, 47, 576-578.
9
40 F. Lefebvre, F. X. Liu-Cai and A. Auroux, J. Mater. Chem.,
4
1994, 4, 125-131.
12 F. W. Lichtenthaler, Acc. Chem. Res., 2002, 35, 728-737.
41 T. Okuhara, T. Nishimura, H. Watanabe and M. Misono, J.
Mol. Catal., 1992, 74, 247-256.
13 A. I. Torres, P. Daoutidis and M. Tsapatsis, Energy Environ.
Sci., 2010,
14 Y. J. Pagán-Torres, T. Wang, J. M. R. Gallo, B. H. Shanks and J.
A. Dumesic, ACS Catal., 2012, , 930-934.
3, 1560-1572.
42 M. A. Harmer and Q. Sun, Appl. Catal. A, 2001, 221, 45-62.
43 E. Medina, R. Bringué, J. Tejero, M. Iborra and C. Fité, Appl.
Catal. A, 2010, 374, 41-47.
2
15 V. Choudhary, S. H. Mushrif, C. Ho, A. Anderko, V. Nikolakis,
N. S. Marinkovic, A. I. Frenkel, S. I. Sandler and D. G. Vlachos,
J. Am. Chem. Soc., 2013, 135, 3997-4006.
44 V. V. Ordomsky, J. van der Schaaf, J. C. Schouten and T. A.
Nijhuis, J. Catal., 2012, 287, 68-75.
45 J. S. Kruger, V. Choudhary, V. Nikolakis and D. G. Vlachos, ACS
16 H. Zhao, J. E. Holladay, H. Brown and Z. C. Zhang, Science,
2007, 316, 1597-1600.
Catal., 2013, 3, 1279-1291.
46 B. F. M. Kuster, Starch - Stärke, 1990, 42, 314-321.
47 S. De, S. Dutta and B. Saha, Green Chem., 2011, 13, 2859-
2868.
17 G. Yong, Y. Zhang and J. Y. Ying, Angew. Chem. Int. Ed., 2008,
47, 9345-9348.
18 J. B. Binder and R. T. Raines, J. Am. Chem. Soc., 2009, 131
,
48 S. J. Dee and A. T. Bell, ChemSusChem, 2011, 4, 1166-1173.
1979-1985.
49 M. Dourson, J. Reichard, P. Nance, H. Burleigh-Flayer, A.
Parker, M. Vincent and E. E. McConnell, Regul. Toxicol.
Pharm., 2014, 68, 387-401.
19 M. E. Zakrzewska, E. Bogel-Łukasik and R. Bogel-Łukasik,
Chem. Rev., 2011, 111, 397-417.
20 S. P. Teong, G. Yi and Y. Zhang, Green Chem., 2014, 16, 2015-
2026.
50 Y. Yang, C. Hu and M. M. Abu-Omar, Journal of Molecular
Catalysis A: Chemical, 2013, 376, 98-102.
21 B. M. Reddy and A. Khan, Catal. Rev., 2005, 47, 257-296.
22 A. Kitiyanan, S. Ngamsinlapasathian, S. Pavasupree and S.
Yoshikawa, J. Solid State Chem., 2005, 178, 1044-1048.
23 W. Zhou, K. Liu, H. Fu, K. Pan, L. Zhang, L. Wang and C.-c. Sun,
Nanotechnology, 2008, 19, 035610.
51 G. Yang, C. Wang, G. Lyu, L. A. Lucia and J. Chen,
BioResources, 2015, 10, 5863-5875.
52 W. Schwald and O. Bobleter, J. Carbohydr. Chem., 1989, 8,
565-578.
53 P. F. Siril, H. E. Cross and D. R. Brown, J. Mol. Catal. A: Chem.,
2008, 279, 63-68.
24 J.-Y. Kim, C.-S. Kim, H.-K. Chang and T.-O. Kim, Adv. Powder
Technol., 2010, 21, 141-144.
25 K.-T. Li, I. Wang and J.-C. Wu, Catal. Surv. Asia, 2012, 16, 240-
248.
26 A. Chareonlimkun, V. Champreda, A. Shotipruk and N.
Laosiripojana, Bioresour. Technol., 2010, 101, 4179-4186.
27 X. Fu, L. A. Clark, Q. Yang and M. A. Anderson, Environ. Sci.
Technol., 1996, 30, 647-653.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 7
Please do not adjust margins

RSC Advances
Page 8 of 8
View Article Online
DOI: 10.1039/C5RA15739K