Catalytic conversion of glucose to 5-hydroxymethyl-furfural with a phosphated TiO2 catalyst

Article abstract of DOI:10.1002/cctc.201402794

Nanosized phosphated TiO₂ catalysts with different phosphate contents were synthesized and tested for the conversion of glucose to 5-hydroxymethylfurfural. The resulting materials were characterized by using N₂-adsorption, XRD, inductively coupled plasma atomic emission spectroscopy, X-ray spectroscopy, TEM, temperature-programmed desorption of ammonia, and FTIR spectroscopy of pyridine adsorption techniques to determine their structural, bulk, surface, and acid properties. We found that TiO₂ nanoparticles catalyzed this reaction under mild conditions in a water-butanol biphasic system. The incorporation of phosphorus into the TiO₂ framework remarkably enhances the target product selectivity, which is ascribed to increased surface area, enhanced acidity, as well as thermal stability resulting from the Ti-O-P bond formation. Under optimal reaction conditions, phosphated TiO₂ was found to exhibit excellent catalytic performance, which resulted in 97-% glucose conversion and 81-% HMF yield after 3 h of reaction at 175-°C. More importantly, the catalyst showed good stability and could be reused for several reaction cycles. A moment of truth: The as-synthesized nanosized phosphated TiO₂ material serves as a bifunctional catalyst in the production of 5-hydroxymethylfurfural (HMF) from glucose. The relationship between catalyst synthesis, properties, and performance indicates a high potential of the TiO₂-based catalyst for HMF production. This provides a step further toward a sustainable route to generate HMF from renewable carbohydrate feedstock.

Full text of DOI:10.1002/cctc.201402794

DOI: 10.1002/cctc.201402794
Full Papers
Catalytic Conversion of Glucose to
5-Hydroxymethyl-furfural with a Phosphated TiO2 Catalyst
Luqman Atanda,[a] Swathi Mukundan,[a] Abhijit Shrotri,[a, b] Qing Ma,[a] and Jorge Beltramini*[a]
Nanosized phosphated TiO2 catalysts with different phosphate
contents were synthesized and tested for the conversion of
glucose to 5-hydroxymethylfurfural. The resulting materials
were characterized by using N2-adsorption, XRD, inductively
coupled plasma atomic emission spectroscopy, X-ray spectros-
copy, TEM, temperature-programmed desorption of ammonia,
and FTIR spectroscopy of pyridine adsorption techniques to
determine their structural, bulk, surface, and acid properties.
We found that TiO2 nanoparticles catalyzed this reaction under
mild conditions in a water–butanol biphasic system. The incor-
poration of phosphorus into the TiO2 framework remarkably
enhances the target product selectivity, which is ascribed to in-
creased surface area, enhanced acidity, as well as thermal sta-
bility resulting from the TiꢀOꢀP bond formation. Under opti-
mal reaction conditions, phosphated TiO2 was found to exhibit
excellent catalytic performance, which resulted in 97% glucose
conversion and 81% HMF yield after 3 h of reaction at 1758C.
More importantly, the catalyst showed good stability and
could be reused for several reaction cycles.
Introduction
Conversion of sugars to platform chemicals is a promising pro-
cess for sustainable production of value-added chemicals. The
sugars obtained from nonedible biomass resources have at-
tracted considerable attention owing to their potential to
offset the dependence on fossil fuels.[1] Cellulose is an inexpen-
sive and readily available raw material that can undergo hy-
drolysis to produce C6 sugars (hexoses). The subsequent dehy-
dration of these sugars produces 5-hydroxymethylfurfural
(HMF), which has been identified as a key biorefining inter-
mediate for chemicals and fuels.[2]
mers to HMF is desirable for industrial production. Although
the use of homogeneous metal halides with ionic liquids re-
sults in high selectivities and yields of HMF from glucose, cata-
lyst handling, separation, and reuse remain a concern in scal-
ing up for industrial application.[9] Therefore, it is essential to
develop a heterogeneously catalyzed process that enables
high HMF selectivity along with high glucose conversion.
Titania (TiO2) serves as a cheap, sustainable, and ecofriendly
metal oxide with various applications such as photocatalysis,[10]
sensors,[11] and electrodes.[12] Owing to its chemical stability
and rich surface chemistry, the acid–base property of TiO2 can
be modified and used in heterogeneous catalysis.[13] Several
authors have reportedly used TiO2 to catalyze the transforma-
tion of carbohydrates to HMF. Watanabe et al.[14] studied the
dehydration of glucose to HMF in hot compressed water cata-
lyzed with anatase TiO2. They suggested that the presence of
high-density acid and basic sites influenced the catalytic activi-
ty of TiO2. Basic sites were responsible for glucose isomeriza-
tion to fructose, whereas acid sites facilitated the successive
dehydration to HMF. De et al.[15] described the catalytic dehy-
dration of glucose with use of self-assembled mesoporous TiO2
nanospheres. They also identified that the surface acidity of
mesoporous TiO2 was responsible for catalyzing the dehydra-
tion reaction. Likewise, mesoporous TiO2 nanospheres templat-
ed with sodium salicylate were tested for the dehydration of
various carbohydrate substrates.[16] The high surface area and
Lewis acidity of the nanoparticles (NPs) played a significant
role in the microwave-assisted conversion of carbohydrates to
HMF. A later study by the same group also achieved efficient
conversion of biomass and carbohydrates to HMF with titani-
um phosphate NPs under microwave-assisted heating condi-
tions.[17] Kuo et al.[18] recently explored the synthesis of acidic
TiO2 NPs and their application for the catalytic conversion of
Hexose conversion to HMF is primarily an acid-catalyzed re-
action. Mineral acids such as H2SO4, H3PO4, and HCl[3] were con-
ventionally used as catalysts for this reaction, which can cause
significant corrosion and environmental problems. In recent
times, various solid acids such as zeolites,[4] acidic ion-exchange
resins,[4c,5] oxides, sulfates, and phosphates,[6] and heteropoly
acid salts[4c,7] have been investigated in search of a suitable
and efficient catalyst for the synthesis of HMF. The production
of HMF from fructose is widely studied, and significant prog-
ress has been made. However, glucose is cost-effective and
readily available in comparison to fructose.[8] Therefore, an effi-
cient process for the direct conversion of glucose and its poly-
[a] L. Atanda, S. Mukundan, Dr. A. Shrotri, Dr. Q. Ma, Dr. J. Beltramini
Nanomaterials Center—AIBN and School of Chemical Engineering
The University of Queensland
Brisbane, QLD 4072 (Australia)
Fax: (+61)7-3346-3973
E-mail: j.beltramini@uq.edu.au
[b] Dr. A. Shrotri
Catalysis Research Center
Hokkaido University
Kita 21 Nishi 10, Kita-Ku, Sapporo 001-0021 (Japan)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cctc.201402794.
ChemCatChem 0000, 00, 0 – 0
1
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
biomass-derived carbohydrates. They observed that the NPs
could promote the dehydration of carbohydrates to HMF as
well as other value-added products such as levulinic acid and
HMF-derived esters. Nakajima et al.[19] significantly improved
the performance of TiO2 to catalyze the conversion of glucose
to HMF by immobilizing TiO2 with H3PO4. As such, TiO2 NPs
have shown great potential in the catalytic transformation of
carbohydrates to HMF. A limitation of the TiO2 catalyst report-
ed so far is that it exhibits only Lewis acid sites. Convincing
evidence has shown that bifunctional catalytic systems with
both Lewis and Brønsted acid functionalities exhibit an im-
proved performance for HMF formation than catalysts with
either Lewis or Brønsted acid functionalities only.[7,20] Therefore,
we believe that the TiO2 catalyst with Lewis and Brønsted acid
functionalities would be more effective in HMF production
from glucose.
phosphate monobasic. After impregnation, the sample was
dried and then calcined at 6008C for 4 h to obtain the phos-
phated TiO2 NP. The phosphated TiO2 catalysts with different
phosphate contents were obtained by varying the amount of
the precursor loaded during the impregnation step. The sam-
ples obtained after calcination were denoted as xP-TiO2, in
which x represents the theoretical weight percentage of phos-
phorus in the catalyst based on the amount of the phosphate
precursor added.
Powder XRD
The XRD patterns of TiO2 NPs with 0–25 wt% phosphate con-
tent are shown in Figure 1. The diffraction patterns of all the
samples can be indexed as an anatase phase of TiO2 (JCPDS
No. 21-1272). The diffraction pattern of pure TiO2 appeared
The design of an optimum catalyst is one of the most crucial
roadblocks in HMF synthesis. However, the solvent environ-
ment can also significantly influence reaction pathways. There-
fore, the role of an appropriate reaction medium cannot be
overlooked.[21] The synthesis of HMF from hexoses can general-
ly involve isomerization, dehydration, fragmentation, reversion,
and condensation steps either in an aqueous or a nonaqueous
reaction medium.[22] The reaction in an aqueous medium suf-
fers a low HMF yield owing to its concomitant rehydration to
levulinic acid and its degradation to polymeric humins. Ionic
liquids[23] and polar organic solvents, such as DMSO,[6b]
DMF,[5a,24] and dimethylacetamide,[25] are found to be efficient
for high-yield HMF production owing to their capacity to inhib-
it undesirable reactions. Nevertheless, poor solubility of hex-
oses/cellulosic biomass and the associated high separation
cost of the target product are the major disadvantages of
these solvents. Various biphasic systems of the water–organic
reaction medium have been developed to facilitate selective
production of HMF, and these have proved to be effective in
overcoming the limitations of either aqueous or nonaqueous
reaction media.[3]
Figure 1. XRD patterns of a) TiO2, b) 5P-TiO2, c) 10P-TiO2, d) 15P-TiO2,
e) 20P-TiO2, and f) 25P-TiO2 NPs.
sharp and intense, which indicated high crystallinity of the
NPs. In contrast, the crystallinity of TiO2 particles reduced after
phosphate treatment, indicated by peak broadening and re-
duction in peak intensity. This result suggests the incorpora-
tion of phosphorus into the TiO2 framework.[26] All the samples
were composed of nanosized crystals, and their mean sizes ob-
tained by using the Debye–Scherrer equation ranged from 5.3
to 27.3 nm (Table 1). An increase in phosphate content resulted
in a decrease in the crystal size of the NP.
Herein, we report the synthesis of a series of bifunctional
phosphated TiO2 NPs with both Lewis and Brønsted acid func-
tionalities. The catalysts were extensively characterized by
using N2-adsorption, XRD, inductively coupled plasma atomic
emission spectroscopy (ICP-AES), TEM, FTIR spectroscopy of
pyridine adsorption, temperature-programmed desorption of
ammonia (NH3-TPD), and X-ray spectroscopy (XPS) techniques.
The catalytic activity of phosphated TiO2 for glucose conver-
sion was evaluated in a water–butanol biphasic system. n-Buta-
nol is a cheap, biorenewable, and environmentally friendly sol-
vent. In addition, it is a promising medium for upgrading HMF
to useful fuels and chemicals, which helps reduce overall pro-
cess costs. The effects of reaction time and temperature, cata-
lyst loading, and substrate concentration were also investigat-
ed to optimize the process.
N2 adsorption–desorption
As shown in Figure 2, the N2 adsorption–desorption isotherms
of all the samples are of type IV, with a capillary condensation
step characteristic of mesoporous materials having narrow
pore size distribution. The onset of the mesoporous filling step
for phosphated TiO2 shifts to a lower pressure in comparison
with pure TiO2. Likewise, a shift in pore diameter to lower
values indicates a distortion in the mesoporous structure of
TiO2 by the incorporation of phosphorus. The specific surface
area (SBET) of the NPs is given in Table 1. The surface area of
Results and Discussion
The phosphated TiO2 catalyst was prepared through the im-
pregnation of freshly prepared hydrated TiO2 with ammonium
&
ChemCatChem 0000, 00, 0 – 0
www.chemcatchem.org
2
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
to the oxygen from the hydroxyl
group.[29] After phosphate treat-
ment, an additional peak was
observed at 531.5 eV, which can
be assigned to oxygen species
of PꢀO bond linkages.[30]
Table 1. Physical, compositional, and acid properties of TiO2 NPs.
Entry Sample Ti/P molar ratio Ti/P molar ratio ICP[a] SBET
Pore volume Crystal size Total acidity
[m2 gꢀ1
]
[cm3 gꢀ1
]
[nm][c]
[mmolgꢀ1
]
[b]
[d]
1
2
3
4
5
6
TiO2
5P-TiO2
10P-TiO2 14.7
15P-TiO2
20P-TiO2
25P-TiO2
–
31.1
–
54.7
124.8
146.8
151.0
145.7
79.4
0.22
0.31
0.36
0.43
0.34
0.25
27.3
7.34
6.36
6.12
5.55
5.30
0.76
1.92
1.96
2.36
1.99
1.40
33.5
15.8
10.1
7.1
The oxygen peaks represent-
ing TiꢀO and hydroxyl group
shifted to higher binding ener-
gies of 530.6 and 533.6 eV, re-
spectively. As shown in Fig-
ure 3b, the Ti2p3/2 peaks of pure
and phosphated TiO2 centered
9.3
6.5
4.9
5.1
[a] Determined from ICP-AES analysis; [b] BET surface area; [c] Measured by XRD using the Debye–Scherrer
equation for the (101) plane; [d] Determined from NH3-TPD analysis.
at 458.8 and 459.4 eV, respectively, can be fitted as a single
peak, which confirms the presence of Ti ions in an octahedral
environment. Notably, there is a shift of lattice oxygen in the
O1s region and Ti2p3/2 of the phosphated sample to higher
binding energies. This shift is due to partial electron transfer
from Ti to P owing to the incorporation of phosphorus into
the TiO2 lattice, which thus decreases the electron density on
Ti. This result is consistent with the report of Guo et al.[31] The
high-resolution spectra of P2p shown in Figure 3c reveals
a single peak at 134.1 eV, which indicates the existence of
phosphorus in the pentavalent oxidation state.[31] This peak
was not observed in the pure TiO2 sample (Figure 3d). Hence,
the XPS results confirmed the incorporation of phosphorus
into the TiO2 framework and the XRD and BET results suggest-
ed the existence of a TiꢀOꢀP linkage.
Figure 2. A) BET isotherms and B) pore size distributions of a) TiO2, b) 5P-
TiO2, c) 10P-TiO2, d) 15P-TiO2, e) 20P-TiO2, and f) 25P-TiO2 NPs.
the samples increased from 54.7 to 151.8 m2 gꢀ1 with an in-
crease in phosphate content up to 15 wt%. A further increase
in phosphate content led to a reduction in the surface area to
79.4 m2 gꢀ1, which can be attributed to partial pore blockage
by excess phosphate.[27] All phosphated TiO2 samples have
higher surface area in comparison with pure TiO2, which can
be related to the stabilization effect of phosphate anions
against sintering. This stabilization effect is assumed to be due
to phosphate anions replacing some of the hydroxyl bridges
originally present in the dried uncalcined TiO2 during impreg-
nation. After calcination, these impregnated anions result in
the formation of TiꢀOꢀP bond linkages. Dalai et al.[28] also re-
ported a similar beneficial role of incorporated sulfate ions in
TiO2 in retarding sintering through bond strengthening.
Compositional and surface analysis
The elemental composition of the catalyst samples was deter-
mined from ICP-AES analysis. The results summarized in
Table 1 indicate that the theoretical values of the Ti/P molar
ratio estimated from the loaded phosphate precursor are in
good agreement with the values determined from ICP analysis.
XPS measurements were used to identify the oxidation states
and bonding characteristics of the elements. The deconvoluted
XPS spectrum of the O1s region of TiO2 (Figure 3a) revealed
two peaks indicating the presence of two oxygen species. The
oxygen peak observed at 530.1 eV is ascribed to the lattice
oxygen of TiO2, and the peak centered at 531.7 eV corresponds
Figure 3. High-resolution XPS spectra of a) O1s, b) Ti2p, c) P2p, and d) wide-
scan survey.
ChemCatChem 0000, 00, 0 – 0
www.chemcatchem.org
3
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
TEM
The TEM image of TiO2 (Figure 4a) shows a plate-like morphol-
ogy. The crystal size estimated from the particle size distribu-
tion curve shown in Figure S1 ranges from 23 to 38 nm, which
is similar to the particle size estimated from XRD analysis. The
high-resolution TEM (HRTEM) image of TiO2 is shown in Fig-
ure 4b, which shows clear crystalline lattice fringes with a mea-
Figure 5. IR spectra of pyridine adsorbed on a) TiO2 and b) 15P-TiO2 after
evacuation at 1508C.
copy of pyridine adsorption. The characteristic absorption
bands at 1443 and 1600 cmꢀ1 observed in pure TiO2 are typical
of surface coordinated pyridine molecules with Lewis acid
sites[6b,34] (Figure 5a). As shown in Figure 5b, phosphate treat-
ment results in the appearance of a peak at 1540 cmꢀ1 as-
signed to the pyridinium ion coordinated with Brønsted acid
sites, as well as in the disappearance of the peaks at 1565 and
1600 cmꢀ1. However, the peak at 1443 cmꢀ1 was retained but
has a weak intensity. In addition, an intense peak at 1490 cmꢀ1
is assigned to the superimposed signals of pyridine adsorbed
on both Lewis and Brønsted acid sites.[35] Therefore, in contrast
to pure TiO2, phosphated TiO2 is characterized with two types
of acid functionalities, that is, Lewis and Brønsted acidity.
Figure 4. TEM images of a and b) TiO2 and c and d) 15P-TiO2. Images (b) and
(d) represent HRTEM images of TiO2 and 15P-TiO2, showing lattice fringes of
the (101) plane.
sured spacing of 0.351 nm, assigned to the (101) plane.[32] In
contrast to pure TiO2, phosphated TiO2 shows a significant de-
crease in the particle size, forming clusters of small nanosized
crystals (Figure 4c). The HRTEM image of phosphated TiO2 (Fig-
ure 4d) also shows lattice fringes, which suggests that the TiO2
structure is retained after phosphate treatment.
Catalytic activity
The dehydration of glucose to HMF was investigated with pure
and phosphated TiO2 catalysts. The initial tests were performed
with 2 wt% glucose concentration in the water–butanol bipha-
sic system. The conversion of glucose was only 53.8% with
27.6% HMF yield if the reaction was performed with pure TiO2.
Under similar reaction conditions, phosphated TiO2 was sub-
stantially more active than pure TiO2, with HMF yield ranging
from 60 to 70% depending on the phosphate content as
shown in Figure 6.
Acidity measurement
The acidity measurement of TiO2 NPs was performed by using
NH3-TPD. The total acid amount evaluated by the quantity of
desorbed ammonia is summarized in Table 1. All phosphated
samples exhibited high acidity than pure TiO2. A progressive
increase in total acidity was observed with an increase in phos-
phate loading up to 15 wt%; thereafter, acidity decreased in
a trend similar to that of surface area. Surface area could be
relatively responsible for influencing the total acidity in terms
of accessibility of the probe gas to available acid sites. On the
basis of the temperature of the desorption peak, acid sites can
be categorized into low (<2008C), medium (200–4008C), and
high (>4008C) strength.[33] The desorption spectra of ammonia
shown in Figure S2 indicate that the bulk of acid sites present
is of low to medium strength even after phosphate treatment.
The nature of acid sites was examined by using FTIR spectros-
The enhanced catalytic activity of phosphated TiO2 can be
ascribed to the presence of Brønsted acid sites. This view is
clarified by the activity results summarized in Table S1, which
indicate the beneficial role of a catalytic system consisting of
both Brønsted and Lewis acidity in HMF formation. With
a pure solid Brønsted acid catalyst (Amberlyst 70), a low HMF
yield of 8.3% at 70% glucose conversion could be achieved,
whereas a pure Lewis acid catalyst in the form of TiO2 gave
a 27.6% HMF yield at 53.8% conversion. Conversion and HMF
yield reached 76.6 and 31.2%, respectively, for a physical mix-
ture of Amberlyst 70 and pure TiO2, which thus demonstrated
the beneficial role of the two acid functionalities in promoting
higher glucose conversion and HMF yield.
&
ChemCatChem 0000, 00, 0 – 0
www.chemcatchem.org
4
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Figure 6. Effect of phosphate content on the catalytic activity of TiO2 NPs.
Reaction conditions: glucose (2 g), catalyst weight (0.4 g), water (30 mL),
n-butanol (70 mL), T=1608C, t=5 h. *, Glucose conversion; *, HMF yield.
Notably, both conversion and yield on phosphated TiO2 in-
creased as the phosphate loading increased to 15 wt%. A fur-
ther increase in phosphorus content led to a reduction in cata-
lytic activity. The optimal catalytic performance of TiO2 was
achieved at 15 wt% phosphate loading corresponding to the
NP with highest surface area and acidity, which indicated that
catalytic activity can be correlated with surface acidity. More-
over, the formation of small nanosized particles contributed to
the overall higher catalytic performance of TiO2 after phos-
phate treatment.
In an attempt to investigate the effect of crystal size and
acidity on the catalytic activity of the NPs, TiO2 was calcined at
varying temperatures (400–7008C) before and after phosphate
treatment. Calcination temperature is believed to influence the
crystal growth of TiO2 NPs,[36] and this is evidenced by the XRD
results presented in Figure S3a. The variations reflected in the
peak intensities represent a range of crystal sizes. The estimat-
ed crystal size of the NPs increased considerably from 6.8 to
34.1 nm with increasing temperature. Crystallinity also in-
creased with increasing temperature. The transformation of
the anatase phase to the rutile phase was observed for TiO2
calcined at 7008C. The catalytic activity of TiO2 NPs calcined at
different temperatures is presented in Figure 7a. We observed
that an increase in temperature corresponds to an increase in
crystal size, which led to a significant decrease in activity. For
instance, at 4008C, 81.8% glucose conversion and 41.4% HMF
yield were attained but decreased to 31.8 and 19.0%, respec-
tively, after calcination at 7008C. Small nanosized crystals have
been reported to contribute to a large surface area, which fa-
cilitate the adsorption of reactants.[37] Dutta et al.[17] also report-
ed that smaller nanosized particles facilitated the mass trans-
port of substrate within the catalytic material. This affords
better access to the internal mesoscopic void spaces and ulti-
mately exposes more active sites for surface interaction with
reactants for enhanced reactivity. Meanwhile, if TiO2 was treat-
ed with phosphate (15 wt%) before calcination at various tem-
peratures, no distinct structural changes were observed from
Figure 7. Effect of calcination temperature on the crystal size and catalytic
activity of TiO2 NPs: a) pure TiO2 and b) phosphated TiO2 (15 wt%). Reaction
conditions: glucose (2 g), catalyst weight (0.4 g), water (30 mL), n-butanol
(70 mL), T=1608C, t=5 h. *, Glucose conversion; ~, HMF yield; &, crystal
size.
the XRD patterns (Figure S3b) and the calculated crystal size
ranges from 5.61 nm at 4008C to 7.82 nm at 7008C. Moreover,
the anatase to rutile transformation was inhibited at 7008C,
which was present in pure TiO2 calcined at the same tempera-
ture. Hence, we may conclude that phosphate treatment sig-
nificantly inhibited the crystal growth of anatase TiO2 and its
transformation to rutile TiO2 at temperatures up to 7008C. This
result was supported by the report of Kçrçsi etal.,[38] who
showed that phosphate treatment enhanced the thermal sta-
bility of TiO2 NPs.
All the phosphated NPs calcined at various temperatures ex-
hibited identical performance for glucose conversion. In all
cases, approximately 80% glucose conversion with 70% HMF
yield was recorded (Figure 7b). This result indicates that
a slight change in crystal size has a negligible effect on the cat-
alytic performance. Hence, we may conclude that similar activi-
ty can be achieved with nanosized TiO2 crystals of similar di-
mensions and acid contents.
ChemCatChem 0000, 00, 0 – 0
www.chemcatchem.org
5
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
To explore the effect of acid content on glucose dehydration
to HMF, we examined TiO2 NPs of similar crystal sizes but dif-
ferent acid contents, that is, pure TiO2 at 4008C (crystal size:
6.82 nm, SBET: 136.9 m2 gꢀ1, and acidity: 1.28 mmolgꢀ1) and
15P-TiO2 at 6008C (crystal size: 6.12 nm, SBET: 151 m2 gꢀ1, and
acidity: 2.36 mmolgꢀ1). As shown in Figure 7, approximately
80% glucose conversion was achieved on both samples. How-
ever, a remarkable increase in HMF yield was attained on 15P-
TiO2, which may be associated with increased acid sites. This
result suggests that both TiO2 and 15P-TiO2 of similar crystal
sizes had comparable activity for glucose conversion. However,
modification of the catalyst surface through phosphate treat-
ment was essential to enhance the selectivity toward HMF.
Hence, we conclude that the crystal size reduction achieved by
only thermal treatment is insufficient in enhancing the activity
of the TiO2 catalyst for HMF synthesis.
Figure 9. Effect of reaction time and temperature on glucose dehydration
on the 15P-TiO2 catalyst. Reaction conditions: glucose (2 g), catalyst weight
(0.6 g), water (30 mL), n-butanol (70 mL). &, 1608C; *, 1708C; ~, 1758C;
!, 1808C.
yield increases gradually with time. Temperature had a signifi-
cant effect on both glucose conversion and HMF yield. At
1808C glucose conversion was 97% after 4 h of reaction,
whereas at 1608C glucose conversion was 93% after 8 h of re-
action. This result indicates that longer time is needed at lower
temperature to reach similar glucose conversion at higher tem-
perature. However, because HMF is more reactive at high tem-
perature,[3c] a longer reaction time resulted in the decrease in
HMF yield at 1808C after 6 h. According to these results, the
optimal condition for this reaction was 3 h of reaction time at
1758C. Under these optimal conditions, we investigated the
effect of initial glucose concentration. The effect of glucose
concentration on HMF yield is presented in Figure 10. Glucose
Effect of reaction parameters
As shown in the previous section, 15P-TiO2 was most active for
glucose to HMF conversion. This catalyst was used to evaluate
the effect of critical reaction parameters. The effect of catalyst
loading on conversion and yield is presented in Figure 8. We
observed a progressive increase in both conversion and yield
with an increase in catalyst loading up to 0.6 g. The yield of
HMF appeared to plateau after 0.6 g with a slight increase in
the conversion. Hence, the optimum catalyst loading was
0.6 g, which corresponds to a glucose/catalyst weight ratio of
10:3.
Furthermore, we examined the effect of both reaction tem-
perature and time at this optimum catalyst loading. The effect
of both reaction temperature and time on HMF yield is pre-
sented in Figure 9.
The time analysis shows that the reaction proceeded rapidly
in the first 3 h to produce HMF and thereafter the product
Figure 10. Effect of initial glucose concentration on HMF yield on the 15P-
TiO2 catalyst. Reaction conditions: glucose/catalyst weight ratio=10:3, water
(30 mL), n-butanol (70 mL), T=1758C, t=3 h. &, Glucose conversion; &,
HMF yield.
conversion was similar but HMF yield decreased with an in-
crease in the initial glucose concentration. The decrease in
HMF yield with increasing glucose concentration may be due
to the propensity to form humins via condensation polymeri-
zation of HMF with glucose and other reactive intermediates in
the aqueous phase.[39]
Figure 8. Glucose dehydration to HMF as a function of catalyst loading in
a water–butanol biphasic solvent on the 15P-TiO2 catalyst. Reaction condi-
tions: glucose (2 g), water (30 mL), n-butanol (70 mL), T=1608C, t=5 h. *,
Glucose conversion; *, HMF yield.
&
ChemCatChem 0000, 00, 0 – 0
www.chemcatchem.org
6
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Plausible reaction pathway
runs (Figure 11). However, HMF yield decreased continuously
over the first three runs from 81 to 70%. We observed
a change in color of the catalyst from white to dark brown,
which is likely due to the deposition of humin compounds on
the catalyst surface, causing deactivation of the active sites.
The Lewis acid-catalyzed isomerization of glucose to fructose is
unaffected because of the minor change in glucose conver-
sion. We hypothesized that the humins deposited on the cata-
lyst surface most likely affect the
The scope of this study does not entail an in-depth elucidation
of the reaction mechanism. This comprises our future work,
and the results will be the focus of a subsequent publication.
However, on the basis of the composition of the products, we
proposed a plausible reaction pathway shown in Scheme 1.
Major intermediates and byproducts formed during the reac-
dehydration step catalyzed by
Brønsted acid. HMF is reported
to have stronger adsorption af-
finity for the Brønsted acid
site.[35,41] HMF adsorbed on the
surface of the recycled catalyst
undergoes a condensation reac-
tion to form humins; this finding
is consistent with the catalyst
color change. The strongly ad-
sorbed HMF on the Brønsted
acid sites can also cause block-
age of active sites as well as un-
dergo further rehydration to
form organic acids.
This finding is evidenced by
Scheme 1. Glucose conversion to HMF in a water–butanol (30:70 v/v) biphasic system on 15P-TiO2 NPs.
FA =formic acid; LA=levulinic acid.
the slightly increased yields of
fructose and levulinic acid (Fig-
ure S6). To test the aforemen-
tioned hypothesis, we decided
tion were identified by using HPLC and GC–MS (Figure S4).
Small amounts of levoglucosan, furfural, formaldehyde, formic
acid, and levulinic acid were detected along with HMF. The
time course of the products is shown in Figure S5. The pres-
ence of levoglucosan is indicative of the loss of one molecule
of water from glucose.[24] In addition to the anticipated prod-
ucts of HMF rehydration (levulinic acid and formic acid), the
formation of furfural indicates that HMF decomposed via loss
of formaldehyde.[40] In addition, formic acid may be produced
as a result of hydrolytic fission of furfural.[40b] On the basis of
these results, we propose that the isomerization of glucose to
fructose is accompanied by the concurrent dehydration reac-
tion of glucose to form levoglucosan. After the formation of
HMF via dehydration of fructose, HMF undergoes rehydration
to produce levulinic acid and formic acid. Furthermore, the de-
composition of HMF can occur via loss of formaldehyde to pro-
duce furfural.
to regenerate the catalyst before subsequent reaction runs. A
post-treatment of the catalyst was performed through calcina-
tion at 6008C for 4 h. The recovered catalyst after post-treat-
ment was approximately 0.51 g. The catalyst loss was during
the decantation and filtration process. The fresh catalyst (15%
Catalyst reusability
One of the most important factors contributing to catalytic
performance is the catalyst stability and reusability during the
reaction. The reusability of the 15P-TiO2 catalyst was investigat-
ed by performing the glucose dehydration reaction up to six
cycles. After the completion of each reaction cycle, the catalyst
was recovered through filtration, washed with acetone, dried,
and then used for the next reaction run. There was no signifi-
cant change in the conversion of glucose for all the reaction
Figure 11. Reusability test of the 15P-TiO2 catalyst for the conversion of glu-
cose to HMF. Reaction conditions: glucose (2 g), catalyst weight (0.6 g),
water (30 mL), n-butanol (70 mL), T=1758C, t=3 h. *, Glucose conversion;
&, HMF yield.
ChemCatChem 0000, 00, 0 – 0
www.chemcatchem.org
7
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
The pH of the solution was adjusted to 7 by adding ammonium
hydroxide solution, and the resulting solution was maintained
under reflux for 24 h. After gelation, excess solvent was removed
with a rotary evaporator followed by vacuum drying at 808C to
obtain titanium hydroxide hydrate. Phosphated TiO2 was prepared
through the impregnation of titanium hydroxide hydrate with an
aqueous solution of the required amount of ammonium phos-
phate monobasic to give TiO2 NPs with 5–25 wt% phosphate con-
tent. A typical synthesis of 5 wt% phosphate loading on TiO2 (5P-
TiO2) was as follows: The phosphate precursor (0.085 g) was dis-
solved in water. Then, dried hydrated TiO2 (2 g) was added, fol-
lowed by stirring at RT for 4 h. Excess water was evaporated, and
the resulting solid was dried at 808C for 12 h. Finally, pure and
phosphated TiO2 NPs were obtained after calcination at 6008C for
4 h.
of the total catalyst weight) was supplemented to achieve an
initial catalyst weight of 0.6 g for the next reaction. The reac-
tion was then repeated under similar conditions as described
previously for the first three reaction runs. A comparable activi-
ty was achieved both before and after catalyst regeneration,
which suggests that the catalytic activity can be restored. XRD
and HRTEM were used to analyze the structural stability of the
spent catalyst. The results of these analyses (Figures S7 and S8)
compared with those for the fresh sample indicate that no sig-
nificant changes occurred in the structural framework of the
NPs after the reaction. Overall, these findings from the reusa-
bility test suggest the stability and recyclability of the catalyst.
Moreover, catalyst deactivation due to humin deposition is re-
versible by regeneration.
Catalyst characterization
Conclusions
The wide-angle XRD patterns were recorded on a Rigaku Miniflex
using monochromatic CoKa radiation (30 kV, 15 mA). The data were
collected at 2q=10–908 (step size: 0.02; step time: 1 s). The N2 ad-
sorption–desorption isotherm measurement was done at ꢀ1968C
with a Micromeritics TriStar II 3020 surface area and porosity ana-
lyzer. Specific surface areas were calculated by using the BET
method and pore size distributions using the BJH model on the
desorption branch of the isotherm. Total pore volumes were esti-
mated from the amount of N2 adsorbed at P/P0 =0.99. Composi-
tional analysis was performed by using ICP. The procedure involved
sample digestion with a Milestone Ethos-1 microwave digester and
then analysis by using a Varian Vista-PRO ICP-OES instrument. The
XPS spectra were acquired with a Kratos AXIS Utlra X-ray photo-
electron spectrometer equipped with a 165 mm hemispherical
We demonstrated the conversion of glucose to 5-hydroxyme-
thylfurfural (HMF) in the water–butanol biphasic system cata-
lyzed by TiO2 nanoparticles, which were synthesized by using
the sol–gel method. TiO2 was modified via phosphate treat-
ment by impregnation. Phosphorus was incorporated into the
structural framework of TiO2, as confirmed by XPS analysis.
Phosphate treatment enhanced the thermal stability of TiO2
and prevented the phase transformation from anatase to rutile
at high temperature. The N2 adsorption–desorption analysis re-
vealed that phosphated TiO2 has a larger surface area than un-
modified TiO2. Furthermore, the presence of Lewis and Brønst-
ed acid functionalities was confirmed by using FTIR spectrosco-
py of pyridine adsorption. Phosphated TiO2 was effective for
the catalytic conversion of glucose to HMF owing to the for-
mation of small nanosized crystals with enhanced surface acid-
ity. We studied the effects of reaction temperature and time,
catalyst weight, and substrate concentration. In general, all
these parameters influence the catalytic performance and yield
of HMF. Under optimal reaction conditions, a remarkably high
HMF yield of approximately 81 and 97% glucose conversion
was achieved at a glucose concentration of 2 wt%. Under simi-
lar reaction conditions, an HMF yield of 60 and 45% was at-
tained at a glucose concentration of 5 and 10 wt%, respective-
ly. Finally, catalyst stability under the given reaction conditions
afforded its recycling for multiple reaction runs.
electron energy analyzer using
a monochromatic AlKa (hn=
1486.6 eV) X-ray source. A wide-scan survey spectrum was record-
ed at an analyzer pass energy of 160 eV and multiplex (narrow)
high-resolution scans at 20 eV. The binding energies were refer-
enced to the C1s peak of adventitious carbon at 284.8 eV to ac-
count for the charging effects. Data analysis was done by using
the CasaXPS software, version 2.3.12, in which a Shirley back-
ground subtraction was used before fitting the spectra using Gaus-
sian–Lorentzian curves. NH3-TPD was performed with a Micromerit-
ics AutoChem II chemisorption analyzer to determine total acid
sites on the catalyst samples. Each sample (ꢁ70 mg) was placed in
a quartz U tube and pretreated at 5008C in a flow of He (flow rate:
50 mLminꢀ1) for 1 h. Then, the sample was saturated with ammo-
nia (15 vol% in He) at 1008C for 30 min. Physisorbed ammonia on
the sample surface was removed by purging the system with He
stream at 1008C for 2 h. The sample was then heated linearly from
100 to 8008C (heating rate: 10 Kminꢀ1) in a flow of He (flow rate:
25 mLminꢀ1) while monitoring the ammonia desorption profile
with a thermal conductivity detector. Pyridine-IR spectroscopy was
used to examine the nature of acid sites. Pyridine was chemisorbed
on the catalyst surface (50 mg) at 1508C. Excess gaseous and phys-
isorbed pyridine were removed by holding the temperature for
30 min in N2 flow. The FTIR spectra were obtained at 128 scans
and 4 cmꢀ1 resolution by using a Nicolet 6700 spectrometer (Smart
Orbit Accessory). The TEM images were taken with a JEOL JSM-
2100 microscope operating at an acceleration voltage of 200 kV.
Experimental Section
Materials and catalyst synthesis
Titanium(IV) butoxide, n-butanol, ammonium phosphate monoba-
sic, ammonium hydroxide solution (28%), glucose, fructose, HMF,
cellobiose, levoglucosan, and furfural were all purchased from
Sigma–Aldrich. Levulinic acid was purchased from Merck Schu-
chardt, and formic acid was supplied by Ajax Finechem. All solu-
tions were prepared using water with a conductivity of 18 MWcmꢀ1
obtained with an ELGA ultrapure water distillation apparatus.
TiO2 NPs were prepared following the neutral amine sol–gel syn-
thesis technique, using titanium(IV) butoxide as the precursor. The
sol was prepared by the dropwise addition of the alkoxide precur-
sor to an aqueous solution containing n-butanol under stirring.
Catalytic evaluation
Glucose dehydration to HMF was performed in a stainless steel
Parr reactor. In a typical experimental run, the required amount of
&
ChemCatChem 0000, 00, 0 – 0
www.chemcatchem.org
8
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
A. M. Raspolli Galletti, G. Sbrana, M. A. Massucci, P. Galli, Appl. Catal. A
2000, 193, 147–153; d) C. Carlini, P. Patrono, A. M. R. Galletti, G. Sbrana,
Appl. Catal. A 2004, 275, 111–118; e) F. S. Asghari, H. Yoshida, Carbohydr.
Res. 2006, 341, 2379–2387.
substrate, solvent, and catalyst was charged into a 300 mL reactor,
purged, and then pressurized to 30 bar (1 bar=100 kPa) with high-
purity Ar. The water–n-butanol (30:70 v/v) biphasic mixture was
used as the reaction medium. The reactor was heated to its set
point temperature, which was measured by a thermocouple
placed inside the reaction mixture. Time zero in the reaction was
defined as the time when the reactor reached its set point temper-
ature. Catalytic experiments were repeated to evaluate the repro-
ducibility of data. Typical errors observed were in the range of
ꢂ2%. Product analysis was done with a Shimadzu Prominence
HPLC system equipped with both refractive index (RID-10) and UV/
Vis (SPD-M20A) detectors and a Bio-Rad Aminex HPX-87H analytical
column. The column was operated at 508C with H2SO4 (5 mm) as
the mobile phase at a flow rate of 0.6 mLminꢀ1 for the analysis of
both aqueous and organic phases. Product identification was per-
formed with a Shimadzu GCMS-QP2010 Ultra equipped with an
Rxi-5ms column. Glucose conversion and product yield were calcu-
lated by using Equations (1) and (2):
[7] C. Y. Fan, H. Y. Guan, H. Zhang, J. H. Wang, S. T. Wang, X. H. Wang, Bio-
mass Bioenergy 2011, 35, 2659–2665.
[8] A. I. Torres, P. Daoutidis, M. Tsapatsis, Energy Environ. Sci. 2010, 3, 1560–
1572.
[9] a) H. Zhao, J. E. Holladay, H. Brown, Z. C. Zhang, Science 2007, 316,
1597–1600; b) X. Qi, M. Watanabe, T. M. Aida, R. L. Smith, ChemSusChem
2010, 3, 1071–1077; c) C. Li, Z. Zhang, Z. K. Zhao, Tetrahedron Lett.
2009, 50, 5403–5405.
[10] a) T. Peng, D. Zhao, K. Dai, W. Shi, K. Hirao, J. Phys. Chem. B 2005, 109,
4947–4952; b) D. Chen, D. Yang, Q. Wang, Z. Jiang, Ind. Eng. Chem. Res.
2006, 45, 4110–4116.
[11] a) A. Teleki, S. E. Pratsinis, K. Kalyanasundaram, P. I. Gouma, Sens. Actua-
tors B 2006, 119, 683–690; b) S.-J. Bao, C. M. Li, J.-F. Zang, X.-Q. Cui, Y.
Qiao, J. Guo, Adv. Funct. Mater. 2008, 18, 591–599.
[12] E. S. Kwak, W. Lee, N.-G. Park, J. Kim, H. Lee, Adv. Funct. Mater. 2009, 19,
1093–1099.
[13] a) P. V. Kamat, J. Phys. Chem. C 2007, 111, 2834–2860; b) D. Chandra, N.
Mukherjee, A. Mondal, A. Bhaumik, J. Phys. Chem. C 2008, 112, 8668–
8674.
[14] a) M. Watanabe, Y. Aizawa, T. Iida, R. Nishimura, H. Inomata, Appl. Catal.
A 2005, 295, 150–156; b) X. Qi, M. Watanabe, T. M. Aida, R. L. Smith, Jr.,
Catal. Commun. 2008, 9, 2244–2249.
[15] S. De, S. Dutta, A. K. Patra, A. Bhaumik, B. Saha, J. Mater. Chem. 2011,
21, 17505–17510.
[16] S. Dutta, S. De, A. K. Patra, M. Sasidharan, A. Bhaumik, B. Saha, Appl.
Catal. A 2011, 409–410, 133–139.
ꢀ
ꢁ
n
no
Conversionð%Þ ¼ 1 ꢀ
ꢃ 100%
ð1Þ
ð2Þ
ꢀ
ꢁ
ni
no
Yieldð%Þ ¼
ꢃ 100%
in which no and n are the number of moles of glucose in the feed
and product, respectively, and ni is the number of moles of prod-
uct i (e.g., HMF, fructose, levulinic acid, and levoglucosan).
[17] A. Dutta, A. K. Patra, S. Dutta, B. Saha, A. Bhaumik, J. Mater. Chem. 2012,
22, 14094–14100.
[18] C.-H. Kuo, A. S. Poyraz, L. Jin, Y. Meng, L. Pahalagedara, S.-Y. Chen, D. A.
Kriz, C. Guild, A. Gudz, S. L. Suib, Green Chem. 2014, 16, 785–791.
[19] K. Nakajima, R. Noma, M. Kitano, M. Hara, J. Mol. Catal. A 2014, 388–
389, 100–105.
[20] E. Nikolla, Y. Romꢁn-Leshkov, M. Moliner, M. E. Davis, ACS Catal. 2011, 1,
408–410.
Acknowledgements
We gratefully acknowledge financial support from the Nanoma-
terials Center (NANOMAC) at the University of Queensland, Aus-
tralia. We also acknowledge the facilities and the scientific and
technical assistance of the Australian Microscopy & Microanalysis
Research Facility at the University of Queensland. L.A. acknowl-
edges sponsorship from the International Postgraduate Research
Scholarship and UQ Centennial Scholarship).
[21] H. Kimura, M. Nakahara, N. Matubayasi, J. Phys. Chem. A 2013, 117,
2102–2113.
[22] J. Zhang, E. Weitz, ACS Catal. 2012, 2, 1211–1218.
[23] a) G. Yong, Y. Zhang, J. Y. Ying, Angew. Chem. Int. Ed. 2008, 47, 9345–
9348; Angew. Chem. 2008, 120, 9485–9488; b) T. Stꢂhlberg, W. Fu, J. M.
Woodley, A. Riisager, ChemSusChem 2011, 4, 451–458.
[24] M. Ohara, A. Takagaki, S. Nishimura, K. Ebitani, Appl. Catal. A 2010, 383,
149–155.
Keywords: carbohydrates · heterogeneous catalysis · HMF ·
mesoporous materials · titanium
[25] J. B. Binder, R. T. Raines, J. Am. Chem. Soc. 2009, 131, 1979–1985.
[26] K. J. A. Raj, R. Shanmugam, R. Mahalakshmi, B. Viswanathan, Indian J.
Chem. Sect. A 2010, 49, 9–17.
[27] K. M. Parida, M. Acharya, S. K. Samantaray, T. Mishra, J. Colloid Interface
Sci. 1999, 217, 388–394.
[28] A. K. Dalai, R. Sethuraman, S. P. R. Katikaneni, R. O. Idem, Ind. Eng. Chem.
Res. 1998, 37, 3869–3878.
[29] D. Sethi, A. Pal, R. Sakthivel, S. Pandey, T. Dash, T. Das, R. Kumar, J. Pho-
tochem. Photobiol. B 2014, 130, 310–317.
[1] a) L. T. Fan, M. M. Gharpuray, Y. H. Lee, Cellulose hydrolysis, Springer,
Berlin, 1987; b) G. Centi, R. A. van Santen, Catalysis for Renewables,
Wiley-VCH, Weinheim, 2007; c) T. Wang, M. W. Nolte, B. H. Shanks, Green
Chem. 2014, 16, 548–572.
[2] 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.
[30] J. C. Yu, L. Zhang, Z. Zheng, J. Zhao, Chem. Mater. 2003, 15, 2280–2286.
[31] S. Guo, S. Han, M. Haifeng, C. Zeng, Y. Sun, B. Chi, J. Pu, J. Li, Mater. Res.
Bull. 2013, 48, 3032–3036.
[32] a) Y. Mao, S. S. Wong, J. Am. Chem. Soc. 2006, 128, 8217–8226; b) S. P.
Albu, A. Ghicov, S. Aldabergenova, P. Drechsel, D. LeClere, G. E. Thomp-
son, J. M. Macak, P. Schmuki, Adv. Mater. 2008, 20, 4135–4139.
[33] M. Paul, N. Pal, P. R. Rajamohanan, B. S. Rana, A. K. Sinha, A. Bhaumik,
Phys. Chem. Chem. Phys. 2010, 12, 9389–9394.
[3] a) Y. Romꢁn-Leshkov, J. N. Chheda, J. A. Dumesic, Science 2006, 312,
1933–1937; b) Y. Romꢁn-Leshkov, C. J. Barrett, Z. Y. Liu, J. A. Dumesic,
Nature 2007, 447, 982–985; c) J. N. Chheda, Y. Roman-Leshkov, J. A. Du-
mesic, Green Chem. 2007, 9, 342–350.
[4] a) K. Lourvanij, G. L. Rorrer, Ind. Eng. Chem. Res. 1993, 32, 11–19; b) Y. C.
Moreau, R. Durand, C. Pourcheron, S. Razigade, Ind. Crops Prod. 1994, 3,
85–90; c) K.-i. Shimizu, R. Uozumi, A. Satsuma, Catal. Commun. 2009,
10, 1849–1853.
[34] a) M. M. Mohamed, W. A. Bayoumy, M. Khairy, M. A. Mousa, Microporous
Mesoporous Mater. 2007, 103, 174–183; b) S. K. Das, M. K. Bhunia, A. K.
Sinha, A. Bhaumik, ACS Catal. 2011, 1, 493–501.
[35] V. V. Ordomsky, J. van der Schaaf, J. C. Schouten, T. A. Nijhuis, J. Catal.
2012, 287, 68–75.
[5] a) A. Takagaki, M. Ohara, S. Nishimura, K. Ebitani, Chem. Commun. 2009,
6276–6278; b) X. Qi, M. Watanabe, T. M. Aida, R. L. Smith, Ind. Eng.
Chem. Res. 2008, 47, 9234–9239; c) X. Qi, M. Watanabe, T. M. Aida,
J. R. L. Smith, Green Chem. 2009, 11, 1327–1331.
[6] a) X. H. Qi, M. Watanabe, T. M. Aida, R. L. Smith, Catal. Commun. 2009,
10, 1771–1775; b) H. Yan, Y. Yang, D. Tong, X. Xiang, C. Hu, Catal.
Commun. 2009, 10, 1558–1563; c) F. Benvenuti, C. Carlini, P. Patrono,
[36] V. Puddu, H. Choi, D. D. Dionysiou, G. L. Puma, Appl. Catal. B 2010, 94,
211–218.
ChemCatChem 0000, 00, 0 – 0
www.chemcatchem.org
9
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
[37] M. Xie, L. Jing, J. Zhou, J. Lin, H. Fu, J. Hazard. Mater. 2010, 176, 139–
145.
[40] a) F. Tao, H. Song, L. Chou, ChemSusChem 2010, 3, 1298–1303; b) R.
Weingarten, W. C. Conner, G. W. Huber, Energy Environ. Sci. 2012, 5,
7559–7574.
[41] J. S. Kruger, V. Choudhary, V. Nikolakis, D. G. Vlachos, ACS Catal. 2013, 3,
1279–1291.
[38] L. Kçrçsi, S. Papp, I. Bertꢃti, I. Dꢄkꢁny, Chem. Mater. 2007, 19, 4811–
4819.
[39] a) B. F. M. Kuster, Starch/Staerke 1990, 42, 314–321; b) S. De, S. Dutta, B.
Saha, Green Chem. 2011, 13, 2859–2868; c) S. J. Dee, A. T. Bell, Chem-
SusChem 2011, 4, 1166–1173; d) J. Wang, J. Ren, X. Liu, J. Xi, Q. Xia, Y.
Zu, G. Lu, Y. Wang, Green Chem. 2012, 14, 2506–2512.
Received: October 3, 2014
Revised: November 11, 2014
Published online on && &&, 0000
&
ChemCatChem 0000, 00, 0 – 0
www.chemcatchem.org
10
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

FULL PAPERS
A moment of truth: The as-synthesized
nanosized phosphated TiO2 material
serves as a bifunctional catalyst in the
production of 5-hydroxymethylfurfural
(HMF) from glucose. The relationship
between catalyst synthesis, properties,
and performance indicates a high
potential of the TiO2-based catalyst for
HMF production. This provides a step
further toward a sustainable route to
generate HMF from renewable carbohy-
drate feedstock.
L. Atanda, S. Mukundan, A. Shrotri,
Q. Ma, J. Beltramini*
&& – &&
Catalytic Conversion of Glucose to
5-Hydroxymethyl-furfural with
a Phosphated TiO2 Catalyst
ChemCatChem 0000, 00, 0 – 0
www.chemcatchem.org
11
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ