Biopolymer templated porous TiO2: An efficient catalyst for the conversion of unutilized sugars derived from hemicellulose

Article abstract of DOI:10.1016/j.apcata.2012.06.002

An efficient procedure for the conversion of unutilized abundant sugar derivatives, such as d-mannose, d-galactose and lactose, into platform molecule 5-hydroxymethylfurfural in aqueous and organic medium is reported. This process involves biopolymer sodi

Full text of DOI:10.1016/j.apcata.2012.06.002

Applied Catalysis A: General 435–436 (2012) 197–203
Contents lists available at SciVerse ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Biopolymer templated porous TiO : An efficient catalyst for the conversion of
2
unutilized sugars derived from hemicellulose
a
a
b
c
c
a,∗
Sudipta De , Saikat Dutta , Astam K. Patra , Bharat S. Rana , Anil K. Sinha , Basudeb Saha
,
b,∗∗
Asim Bhaumik
a
Laboratory of Catalysis, Department of Chemistry, University of Delhi, North Campus, Delhi 110 007, India
Department of Materials Science, Indian Association for the Cultivation of Science, Jadavpur, Kolkata 700 032, India
Catalytic Conversion Process Division, Indian Institute of Petroleum, Dehradun 248 005, India
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
An efficient procedure for the conversion of unutilized abundant sugar derivatives, such as d-mannose, d-
galactose and lactose, into platform molecule 5-hydroxymethylfurfural in aqueous and organic medium
is reported. This process involves biopolymer sodium alginate templated porous TiO2 nanocatalysts con-
taining strong Lewis acidic sites as determined by pyridine-IR and temperature programmed desorption
Received 3 February 2012
Received in revised form 7 May 2012
Accepted 2 June 2012
Available online 12 June 2012
(TPD) of ammonia studies. Porous TiO2 materials were characterized by XRD, SEM, HR-TEM and N2
sorption techniques. Biopolymer templating pathway provided effective synthetic route for the TiO2
nanomaterials, which contain considerable mesoscopic void space as revealed from the HR-TEM and N2
sorption studies. Using this TiO2 catalyst, improved HMF yields were obtained without using expensive
ionic liquids as solvents. Hydrothermally obtained TiO2 nanocatalysts showed maximum activity in d-
mannose conversion to HMF, whereas nonporous TiO2 is inactive in this reaction. Porosity (surface area),
small particle size and strong acidic sites generated through biopolymer templating pathway are crucial
for high catalytic activity.
Keywords:
Biopolymer templating
Mesoporosity
Nanostructure
Biomass
Platform chemical
Mannose
©
2012 Elsevier B.V. All rights reserved.
1
. Introduction
cellulose using ionic liquids–metal halide and other active cata-
lysts [9–12]. Although HMF can be easily prepared from fructose
by using a wide variety of catalysts [13,14], advances in the syn-
thesis of HMF from glucose [15,16], cellulose [17,18] and hexoses
have not been studied extensively. Transformation of C-2 epimers
of glucose, such as d-mannose and d-galactose (Scheme 1), into
HMF would be interesting since d-mannose and d-galactose are
comparatively unutilized among the sugar derivatives and it should
be noted that mannose-containing polymers can account for 10% of
the dry weight of pine wood [19], while galactose can be as much
as 2% of the carbohydrate in corn stover [20]. Even mannose and
galactose are present in wood-biomass in small quantities. The con-
version of these glucose epimers to platform chemical would be
interesting study as the isomerization mechanism of glucose is a
difficult transformation. In addition to these, lactose can also be
considered as inexpensive renewable feedstock since it is an under-
utilized sugar derivative and several million tons of it can be obtain
annually as a by-product from the dairy industry [21].
The consumption rate of fossil fuels compared to the available
resources and the calamity of global warming resulting increased
CO2 emissions are two major problems faced by humanity. These
crises of entire planet require the right solution as a top priority.
The effects of global warming and energy crisis have spurred many
research projects focused on utilization of renewable biomass as
carbon-neutral feedstock and their efficient conversion to fuels
and chemicals by using active chemical catalysts [1]. Toward
sustainability, the efficient utilization of cellulose required the
development of new technologies for the efficient and selective
conversion of cellulose [2] into platform molecule such as 5-
hydroxymethylfurfural (HMF) for a wide range of downstream
applications [3–5]. In the recent years, considerable progress has
been made to obtain fuels [6] and polyester building block chem-
icals [7,8] from HMF platform. Significant progresses have also
been made in the synthesis of HMF from sugar derivatives and
Recently, biopolymers have been shown to exert a remarkable
degree of control over the nucleation and growth of crystalline
metal oxides [22]. Alginate, for example, is the structural biopoly-
mer found in seaweed and contains blocks of guluronate monomers
∗
Corresponding author.
Corresponding author.
E-mail addresses: bsaha@chemistry.du.ac.in (B. Saha), msab@iacs.res.in
∗
∗
(Scheme 2) that are readily cross-linked by strong electrostatic
(
A. Bhaumik).
bonds to multivalent metal cations. High thermal stability and
0
926-860X/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.apcata.2012.06.002

1
98
S. De et al. / Applied Catalysis A: General 435–436 (2012) 197–203
Perkin Elmer Spectrum 100 spectrophotometer. Temperature pro-
grammed desorption (TPD) of NH studies in the temperature range
3
◦
1
00–800 C were employed for the quantitative estimation of sur-
face acidity of the samples by using a Micromeritics ChemiSorb
720 attached with a thermal conductivity detector. Powder X-ray
2
diffraction of the TiO nanoparticles for wide-angle was carried out
2
in a Rigaku MiniFlex II XRD machine. Conversions of sugar deriva-
tives (d-mannose, d-galactose, lactose) into HMF were performed
in a CEM Matthews WC Discover microwave reactor, model no.
Scheme 1. Mono- and disaccharide precursors as HMF feedstock.
9
08010 DV9068, equipped with programmable pressure and tem-
1
perature controller. H NMR spectral analysis was performed on
a JEOL JNM ECX-400 P 400 MHz instrument and data were pro-
cessed using a JEOL DELTA program version 4.3.6. The yield of HMF
was determined by UV–Visible spectrometric technique using a
UV-SPECORD 250 analytikjena spectrophotometer.
extended gel network structure in aqueous media made biopoly-
mers capable of maintaining the pre-organized dispersion of metal
cations to relatively high temperature i.e. influence the nucleation
and growth of crystalline phases [23].
Moreover, metal oxides with desired mesoporosity can be pre-
pared by using various templates, such as block copolymers [24],
amines [25], and surfactants [26]. Additionally, mesoporous TiO2
nanoparticles prepared by salicylic acid and aspartic acid templat-
ing pathway were found to be effective catalysts for the conversion
of hexoses into HMF in aqueous and organic media [27]. Metal bind-
ing and structure directing capacity of the biopolymer alginate with
its ability to reduce metal salts to nanoparticles of desired meso-
porous structures is something that yet to be explored. Herein, we
Nitrogen sorption isotherms were obtained using a Beckmann
Coulter SA3100 surface area analyzer at 77 K. Prior to the measure-
ment, the samples were degassed at 393 K for 12 h.
2.1. Preparation of TiO2 from titanium isopropoxide and sodium
alginate
Sodium alginate (1.0 g) was dissolved in 300 mL water by stir-
◦
ring at 80 C; to this was added titanium isopropoxide (14 mL) at
report the synthesis method of porous TiO nanospheres using algi-
room temperature under continuous stirring. The pH of the reaction
mixture was adjusted to 10 by the dropwise addition of aque-
ous ammonia (25%) solution. Then the reaction mixture was then
2
nate biopolymer as the structure directing agent. In the present
synthesis, primary porous TiO2 nanospheres was produced from
titanium isopropoxide in aqueous media under alkali hydrothermal
treatment method (Scheme 1). Apart from the full characteriza-
tion of the porous TiO2 material using HR-TEM, SEM and powder
XRD, surface area and pore diameter analysis showed mesoporous
◦
divided into three parts in equal volume and stored in freeze (0 C),
◦
at room temperature and at 60 C in hot oven for aging for 24 h
in each case. The resultant solid was collected by repeated cen-
trifugation (8000 rpm, 10 min) and washing with distilled water.
◦
The solid was dried in oven at 80 C for overnight 8 h and then cal-
nature of the material. Pyridine-IR and NH -TPD analyses showed
3
◦
cined at 300 C in furnace for 4 h. TiO materials prepared at room
2
the presence of strong Lewis acidic sites in the pores of the
◦
◦
nanospheres of TiO . Consequently, the acidic TiO2 nanomaterials
temperature, 0 C, and under hydrothermal (60 C) condition were
designated as TiO -R, TiO -F, and TiO -H, respectively.
2
were used as catalysts for the conversion of d-mannose, d-galactose
and lactose into HMF under mild conditions in aqueous and organic
media.
2
2
2
2.2. Measurement of acid content of porous TiO2 by temperature
programmed FT IR and NH -TPD analyses
3
2
. Experimental
For the temperature programmed FT IR study, calcined sam-
ple was activated. Then 100 mg of this sample and 3 g of pyridine
were taken separately in a 2 mL glass bottle and 70 mL polypropy-
lene bottle, respectively. Then the glass bottle was kept inside
the polypropylene bottle very cautiously. Here TiO2 and pyridine
were separated by glass contact and the polypropylene bottle
d-mannose, d-galactose, lactose, sodium alginate, titanium iso-
propoxide were supplied by Sigma–Aldrich and were used without
further purification. Dimethylacetamide (DMA), lithium chloride,
2
5% ammonia solution were supplied by Spectrochem, India. Unless
otherwise stated, distilled water was used as aqueous phase. HR-
TEM images were recorded on a JEOL DATUM Model No. JEM1011.
A JEOL JEM 6700F field emission scanning electron microscope
FESEM) was used for the determination of the particle morphol-
ogy of the TiO2 nanoparticles. Fourier transform infrared (FT IR)
spectra of the pyridine adsorbed samples were recorded on a
◦
was sealed with a cap. Then it was kept in the oven at 80 C for
24 h. Then the temperature variation FT IR was studied with this
pyridine adsorbed sample [28]. For the temperature programmed
desorption (TPD) of ammonia studies, samples were activated at
(
◦
300 C and then the catalyst was cooled down to room tempera-
ture. Immediately ammonia was injected at room temperature in
the absence of carrier gas flow. The temperature of the catalyst
was then raised in a stepwise manner at a linear heating rate of
◦
1
1
0 C/min. The desorption of ammonia in the temperature range
◦
00–800 C was analyzed by using a Micromeritics ChemiSorb 2720
containing a thermal conductivity detector.
2.3. Procedure of conversion of sugar into HMF in DMA–LiCl
In a typical experiment, a microwave tube was charged with
d-mannose (25 mg), TiO2 catalyst (10 mg) and DMA–LiCl (0.5 g,
0 wt% LiCl). The loaded microwave tube was placed in the
1
microwave reactor. The reaction was continued for a desired time at
◦
1
40 C. After completion of reaction for the desired time, the reac-
tion mixture was allowed to cool to room temperature and HMF
was extracted with diethyl ether. The purity of HMF was checked
Scheme 2. Synthesis of HMF catalyzed by alginate derived TiO2 NPs.

S. De et al. / Applied Catalysis A: General 435–436 (2012) 197–203
199
◦
◦
◦
◦
◦
◦
at 2Â value of 25.3 , 37.8 , 48.0 , 53.7 , 54.9 and 62.5 , which
corresponding to anatase (1 0 1), (0 0 4), (2 0 0), (1 0 5), (2 1 1) and
(2 0 4) crystal planes (JCPDS 21-1272) [29]. The wide angle powder
XRD results revealed that alginate template method produced
highly stable and crystalline TiO2 nanoparticles. Representative
TEM images of porous TiO2 nanoparticles (TiO -F and TiO -H)
2
2
◦
calcined at 300 C are shown in Fig. 2. The HR TEM images are
shown in Fig. 2(b) and (e). As seen in these figures that spherical
tiny TiO2 nanoparticle are assembled by forming self-aggregated
(loose assembly) nanostructure for both the samples. Further,
from the HR-TEM images, the (1 0 1) plane is quite clear in Fig. 2(b)
and the planes (1 0 1) and (0 0 4) are also seen in Fig. 2(e). The
selected area electron diffraction (SAED) patterns for both samples
are shown in Fig. 2(c) and (f) suggested the diffraction spots for
crystalline anatase TiO2 phase. Thus XRD and TEM analyses show
the phase purity and crystallinity of the TiO2 nanoparticles.
FE SEM images (Fig. 3) of TiO2 samples are composed of
tiny spherical nanoparticles. The self-assembled structure of TiO2
nanoparticles for samples prepared hydrothermally and at room
temperature are observed from their respective FE SEM images
Fig. 1. XRD profile of the TiO2 samples (TiO2-R, TiO2-F and TiO2-H) with indexed
peaks.
1
by H NMR spectroscopic method. The yield of HMF was measured
by UV–Vis spectrophotometric technique.
(Fig. 3a and b). Both samples are composed of uniform tiny spheri-
cal nanoparticles and there are inter-particle voids of ca. 10 nm. A
2.4. Catalyst life-time study
closer look at these SEM images revealed that porous TiO -F and
2
TiO -R synthesized at lower temperatures have relatively smaller
particle size (ca. 8–10 nm), whereas high temperature synthesized
2
A microwave tube was charged with d-mannose (100 mg), TiO2-
H catalyst (40 mg) and DMA–LiCl (2 g, 10 wt% LiCl). The mixture was
TiO -H material composed of relatively larger particles of ca. 15 nm
2
◦
heated for 5 min under microwave irradiation at 140 C. After the
size.
reaction, the reaction mixture was cooled down to room tempera-
ture and filtered the catalyst. HMF component was extracted with
diethyl ether. Spent catalyst was recovered, washed with distilled
The mesoporosity of the TiO2 materials was investigated from
the N2 adsorption/desorption study. The isotherms of TiO2 mate-
rials obtained by using sodium alginate as template are shown
in Fig. 4. Isotherms for TiO -H (Fig. 4), TiO -F, and TiO -R (ESI
◦
water and dried in oven at 80 C for 12 h. The recovered catalyst
2
2
2
was reused for next catalytic cycle using fresh DMA–LiCl solvent.
The reusability of DMA–LiCl solvent was also tested by recycling
the spent reaction mixture containing solvent and catalyst after
Fig. S4) could be classified as type IV isotherm corresponding
to the mesoporous materials based on their behavior at high
P/P . Hydrothermally prepared TiO -H material shows high surface
0
2
extracting HMF with diethyl ether. Fresh solvent and TiO -H cata-
2
area among them. BET surface area of TiO -H, TiO -R and TiO -F
2
2
2
2
lyst were not added in the later reaction.
2
−1
2
−1
−1
samples were 124.0 m g , 74.0 m g and 50.2 m g . Pore vol-
umes of these samples were 0.44 ccg , 0.21 ccg and 0.44 ccg ,
−
1
−1
−1
2.5. Determination of HMF yield
respectively. In these isotherms, between P/P0 of 0.05–0.70 the N2
adsorption gradually increases for all the samples. The sharp uptake
in N2 adsorption observed at higher P/P0 (0.70–0.85) indicating
the presence of large mesopore in these samples [30]. The pore
size distributions of the samples, measured using the Non Local
Density Functional Theory (NLDFT) method (using N2 adsorption
The yield of HMF in the product solution was determined by
UV–Visible spectrophotometric technique.
2.6. UV–Visible spectrophotometric method
on silica as a reference), suggested that the TiO -H synthesized
2
◦
The UV–Visible spectrum of pure HMF solution has a distinct
at 60 C has a larger pore (ca. 11.68 nm, Fig. 4b) than TiO -R and
2
peak at 284 nm with corresponding molar extinction coefficient
TiO -F materials. TiO -F has an average pore dimension ca. 9.2 nm.
2
2
4
−1
−1
(
ε) value of 1.66 × 10 M cm . The percentage of HMF in each of
Similar decrease in dimension of pores was also observed for
our TiO2 nanoparticles synthesized by using sodium salicylate as
template [29].
the reaction product was calculated from the measured absorbance
values at 284 nm and the extinction coefficient value. First, a
standard HMF solution of 99% purity was analyzed for corre-
lating the percentage of actual and calculated amount of HMF.
Once a good correction was established, the extracted HMF prod-
uct samples were run and the percentage of HMF yield was
calculated. Repeated measurement of the same solution shows
the percentage of error associated with this measurement was
Temperature programmed FT IR spectroscopy is one of the most
important analytical tools for characterizing the Lewis acidic prop-
erty of the materials using a Lewis base like pyridine as probe
molecule. From Fig. 5 and ESI Fig. S5, it is clearly observed that all
the pyridine adsorbed mesoporous TiO2 materials show two char-
−
1
−1
acteristic bands at 1587 and 1443 cm . The band at 1443 cm
could be attributed to the adsorbed pyridine at the Lewis acid site
29] and with increase in desorption temperature this band showed
±
3
3
3%.
[
. Results and discussion
slow decrease in intensity due to removal of pyridine from the TiO2
surface. This result suggested the presence of considerably strong
Lewis acid site in our self-assembled porous TiO2 nanoparticles.
.1. Characterization of porous TiO2
From the NH -TPD study, we measured quantitatively the surface
3
The wide angle XRD patterns of the TiO2 nanoparticles are
acidity of the porous materials. The NH -TPD results of our porous
3
shown in Fig. 1. This XRD result reveals that TiO -H, TiO -F and
TiO2 samples are shown in Table 1. All three porous TiO2 materi-
als possess three types of acidic sites: weak, medium and strong
[31,32]. As seen from the table that surface acidity decreases in the
order TiO -R > TiO -H > TiO -F. Thus, although the hydrothermally
2
2
TiO -R samples are composed of crystalline anatase phase. The
2
crystalline planes corresponding to the peaks for anatase TiO2
have been indexed in Fig. 1. All calcined samples show major peak
2
2
2

2
00
S. De et al. / Applied Catalysis A: General 435–436 (2012) 197–203
Fig. 2. TEM image of typical nanostructure of TiO2-F (a), HR-TEM images of typical nanostructure of TiO2-F showing (1 0 1) crystalline planes (b), selected area electron
diffraction (SAED) pattern of the calcined mesoporous TiO2-F (c) and the nanostructure of hydrothermally synthesized materials TiO2-H (d), HR-TEM images of typical
nanostructure of TiO2-H showing (1 0 1) and (0 0 4) crystalline planes (e), selected area electron diffraction (SAED) pattern of the calcined mesoporous TiO2-H (f).
prepared porous TiO -H material has highest surface area, strong
acid sites of this sample are lower.
followed by almost a plateau up to 15 min. A control experiment
without catalyst resulted in only 4% HMF in 5 min as compared to
2
2
6% with catalyst, confirming the true effectiveness of the catalyst.
Commercial anatase TiO2 was ineffective to catalyze the dehydra-
tion of mannose (yield less than 2%). This anatase TiO sample
3
.2. Catalytic activity of porous TiO2
2
2
−1
and 0.01 mmol/g
has BET surface area and acidity of 48 m g
33], which are much lower than the biopolymer template porous
The catalytic activity of the alginate templated porous TiO2
[
nanoparticulate materials was investigated for the conversion of
mannose, lactose and galactose into HMF at varying reaction condi-
tions. As shown in Table 2, dehydration reaction of 25 mg mannose
with 10 mg TiO -H catalyst produced 29% HMF under microwave
irradiation for 0.5 min in DMA–LiCl medium at 140 C. The yield of
HMF increased from 29% to 42% upon continuing the same reaction
for 2 min. A further increase of reaction time from 2 min to 5 min
showed only 2% improvement in HMF yield. At lower temperature
TiO samples. This result suggests that mesoporosity together with
2
strong surface acidic sites generated through biopolymer templat-
ing pathway are crucial for the catalytic efficiency. The higher
yield of HMF in DMA–LiCl solvent than that in water–MIBK can be
interpreted by facile hydrolysis of lactose in DMA–LiCl due to the
presence of highly ion-paired chloride ions in DMA–LiCl. Previous
literature report has shown moderate HMF yields from mannose
dehydration using 6 mol% CrCl2 catalysts in organic or ionic liq-
uid solvent solvents [34]. In the present reaction, we avoided using
chromium halide catalyst due to their toxicity and environmental
pollution. Nevertheless, the investigated solid acid catalyst showed
promising results in environmentally benign solvent.
2
◦
◦
(
120 C), the yield of HMF decreased from 42% to 35%. The variation
of TiO -H catalyst loading from 2 mg to 20 mg for the dehydration
2
of fixed amount of mannose (25 mg) showed an improvement in
HMF yields from 27% to 45%. A blank experiment without catalyst
resulted in only 9% HMF in 2 min as compared to 36% HMF with 5 mg
catalyst for mannose dehydration reaction. Once the catalytic activ-
ity of TiO -H material in DMA–LiCl solvent was realized, we have
tested environmentally benign water–MIBK biphasic solvent. The
catalyst was also effective in water–MIBK solvent, giving 26% HMF
Lactose is another potential substrate for HMF synthesis. How-
ever, due to the presence of a large number of hydroxyl groups,
lactose has lower solubility than mannose. It is reported that
only 20% HMF yield was obtained from the lactose dehydration
2
◦
reaction with chromium chloride/lithium halide/6 mol% H SO cat-
from mannose in 5 min at 120 C. The yield of HMF increased from
2% to 26% upon increasing the reaction time from 0.5 min to 5 min
2 4
alytic systems [34]. Given the toxicity issue of chromium salt and
1
Table 1
Surface acidity different porous TiO2 samples estimated from NH3-TPD studies.
Sample name
Total acidity (mmol/g)
Weak acidity (mmol/g)
Medium acidity (mmol/g)
Strong acidity (mmol/g)
TiO2-R
TiO2-F
TiO2-H
1.216
0.95
1.11
0.480
0.353
0.427
0.218
0.156
0.267
0.518
0.441
0.416

S. De et al. / Applied Catalysis A: General 435–436 (2012) 197–203
201
Table 2
HMF synthesis from d-mannose, d-galactose and lactose with TiO2 obtain via alginate templating route.
◦
Entry
Substrate (mg)
Catalyst (mg)
Solvent (0.5 g)
T ( C)
t (min)
HMF yield (%)
1
2
3
4
5
6
7
8
9
Mannose, 25
Mannose, 25
Mannose, 25
Mannose, 25
Mannose, 25
Mannose, 25
Mannose, 25
Mannose, 25
Mannose, 25
Mannose, 25
Mannose, 25
Mannose, 25
Mannose, 25
Mannose, 25
Lactose, 25
TiO2-H, 10
TiO2-H, 10
TiO2-H, 10
TiO2-H, 10
TiO2-H, 2
TiO2-H, 5
TiO2-H, 20
–
DMA–LiCl
DMA–LiCl
DMA–LiCl
DMA–LiCl
DMA–LiCl
DMA–LiCl
DMA–LiCl
DMA–LiCl
Water–MIBK
Water–MIBK
Water–MIBK
Water–MIBK
Water–MIBK
Water–MIBK
DMA–LiCl
Water–MIBK
Water–MIBK
DMA–LiCl
140
140
140
120
140
140
140
140
120
120
120
120
120
120
140
120
120
140
120
120
0.5
2
5
5
2
2
2
2
5
0.5
2
5
10
15
5
5
10
5
5
10
29
42
44
35
27
36
45
9
–
4
10
11
12
13
14
15
16
17
18
19
20
TiO2-H, 10
TiO2-H, 10
TiO2-H, 10
TiO2-H, 10
TiO2-H, 10
TiO2-H, 10
TiO2-H, 10
TiO2-H, 10
TiO2-H, 10
TiO2-H, 10
TiO2-H, 10
12
18
26
29
27
35
11
14
27
6
Lactose, 25
Lactose, 25
Galactose, 25
Galactose, 25
Galactose, 25
Water–MIBK
Water–MIBK
10
DMA–LiCl contains 10 wt% LiCl; Water:MIBK = 1:1 (v/v).
hazardousness of H SO , the present work of lactose dehydra-
comparable reaction conditions, lactose dehydration with TiO -H
2
4
2
tion reaction was carried out using mesoporous TiO -H catalyst in
catalyst produced 11% less HMF than that obtained from man-
nose. The involvement of an additional hydrolysis step in lactose
dehydration process accounted for lower HMF yield from lactose.
When lactose dehydration was carried out in water–MIBK sol-
vent, a significantly less HMF yield was recorded; 11% from lactose
versus 26% from mannose. The yield slightly improved to 14% upon
2
water–MIBK and DMA–LiCl solvents. DMA–LiCl solvent is known
+
to form DMA•Li macrocations, resulting in highly ion-paired chlo-
ride ions [26]. When lactose (25 mg) dehydration reaction was
carried out using 10 mg TiO -H catalyst in DMA–LiCl (10 wt% LiCl)
2
◦
solvent at 140 C for 5 min, 35% HMF yield was obtained. Under
Fig. 4. N2 adsorption (᭹)–desorption (o) isotherm of the calcined TiO2-H (a) mea-
sured at 77 K and the representative pore size distribution using NLDFT method is
shown (b).
◦
Fig. 3. FE-SEM image of calcined (300 C) mesoporous TiO2 samples prepared under
hydrothermal condition and at low temperature: (a) TiO2-H and (b) TiO2-F.

2
02
S. De et al. / Applied Catalysis A: General 435–436 (2012) 197–203
Fig. 5. (a) FT IR spectra of mesoporous TiO2-H (298 K) and pyridine desorbed TiO2
sample at (298 K, 323 K, 423 K, and 523 K).
Fig. 7. A comparison of the catalytic activity of TiO2 NPs prepared under three syn-
thesis conditions for the synthesis of HMF from d-mannose. (For interpretation of
the references to color in the text, the reader is referred to the web version of the
article.)
conversion of mannose to HMF in both DMA–LiCl and water–MIBK
under microwave assisted heating. All reactions in DMA–LiCl and
◦
◦
water–MIBK solvents were carried out at 140 C and 120 C, respec-
tively. The effectiveness of three catalysts (TiO -H, TiO -R, TiO -F),
2
2
2
in terms of HMF yields, is compared in Fig. 7. As seen, the effec-
tiveness of all three catalysts was comparable in both solvents. In
the case of DMA–LiCl solvent mediated reactions, the yields were
in the range of 42–44% (red bars, Fig. 7). Further, TiO2-H sample
possess lower concentration of strong Lewis acid sites although it
possess highest surface area among these three samples. If we see
the HMF yields under similar reaction conditions this hydrother-
mally prepared porous TiO2 showed marginally better catalytic
activity than other two samples. Although, this TiO2 sample has
higher surface area but surface acidity, specially strong acid sites,
which could be actually responsible for catalytic conversion of the
utilized sugars to HMF is lower than the materials synthesized at
Scheme 3. Isomerization of galactose to tagatose and further dehydration to HMF.
continuing the reaction for 10 min. The scope of present investi-
gation was further extended for galactose substrate. In the case of
galactose, its isomerization to ketose results in the formation of
tagatose intermediate [35], the C-4 epimer of fructose (Scheme 3).
Inefficient dehydration of tagatose intermediate to HMF is the
reason of low HMF yields from galactose [35]. The dehydration reac-
tions of galactose with TiO -H catalyst were carried out in both
2
DMA–LiCl and water–MIBK solvents. As shown in Table 1, HMF
yields from galactose was about 4–8% lower than that obtained
from lactose dehydration under comparable reaction conditions. A
comparison of HMF yields from all substrates in both solvents is
pictorially shown in Fig. 6.
◦
room temperature and at 0 C (Table 1). Apart from this, TiO2-F and
TiO2-R samples synthesized at lower temperature have relatively
smaller particle size (∼8–10 nm), whereas that for porous TiO2-H is
∼
15 nm. Thus the advantage of high surface area of porous TiO -H
2
3.3. Effectiveness of TiO2 NPs prepared under different synthesis
could have been nullified by the bigger particle size and relatively
smaller number of strong acid sites, and thus it did not show much
improvement in HMF yield over other porous TiO2 samples.
method
The effectiveness of the TiO2 NPs prepared under different
synthesis conditions was examined as acid catalysts for the
3.4. Catalyst recyclability
The reusability of the mesoporous TiO -H catalyst was exam-
2
ined for d-mannose dehydration reaction in DMA–LiCl by recycling
the spent catalyst. Prior to recycle the catalyst for the next run, the
reaction mixture was filtered and HMF component was extracted
from the reaction mixture with diethyl ether. The solid TiO -H cat-
2
alyst was recovered by filtration, washed with distilled water and
dried. The catalyst was reused for four catalytic cycles. After fourth
cycle, the loss of catalyst activity in HMF yields was about 6% (Fig. 8,
Table S1).
From the perspective of renewable chemistry, recyclability of
both catalyst and solvent is highly desirable. In order to address
the practical application of the current method of HMF synthe-
sis, both catalyst and DMA–LiCl solvent were recycled for mannose
(
500 mg) dehydration reaction with TiO -H catalyst (50 mg) under
2
◦
conventional oil-bath heating conditions at 140 C. The reaction
produced 39% HMF in 45 min in the first cycle (Table S2). In the
consecutive second and third cycles, the reaction produced 34%
and 32% HMF, respectively, which are much lower than those
under microwave heating conditions. Although DMA–LiCl solvent
Fig. 6. A comparison of HMF yields obtained from dehydration reactions of man-
nose, lactose and galactose using TiO2-H catalyst.

S. De et al. / Applied Catalysis A: General 435–436 (2012) 197–203
203
Delhi for providing the instrument facility through Nano Mission
Initiative.
Appendix A. Supplementary data
Supplementary data associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/
j.apcata.2012.06.002.
References
[
[
1] R.A. van Santen, in: G. Centi, R.A. van Santen (Eds.), Catalysis for Renewables,
Wiley-VCH, Weinheim, 2007, pp. 9–19.
2] M.E. Zakrzewska, E. Bogel-Lukasik, R. Bogel-Lukasik, Energy Fuels 24 (2010)
737–745.
[
[
[
3] G.W. Huber, S. Iborra, A. Corma, Chem. Rev. 106 (2006) 4044–4098.
4] A. Corma, S. Iborra, A. Velty, Chem. Rev. 107 (2007) 2411–2502.
5] J.N. Chheda, G.W. Huber, J.A. Dumesic, Angew. Chem. Int. Ed. 46 (2007)
Fig. 8. Recyclability study of TiO2-H catalyst for d-mannose dehydration reaction.
Reaction conditions: d-mannose = 100 mg, TiO2-H = 40 mg, T = 140 C, and t = 5 min.
7
164–7183.
6] Y. Román-Leshkov, C.J. Barrett, Z.Y. Liu, J.A. Dumesic, Nature 447 (2007)
82–986.
7] P. Venkitasubramanian, E.C. Hagberg, P.D. Bloom, Polym. Prepr. (Am. Chem.
Soc. Div.) Polym. Chem. 49 (2008) 914–915.
◦
[
[
9
perhaps mixed with resulting water from the dehydration reac-
tion, the loss in activity of the used catalyst in spent solvent was
not significant, in terms of HMF yield.
[
[
8] A. Gandini, M.N. Belgacem, Prog. Polym. Sci. 22 (1997) 1203–1379.
9] M.E. Zakrzewska, E. Bogel-Lukasik, R. Bogel-Lukasik, Chem. Rev. 111 (2011)
397–417, and references therein.
[
[
10] X.L. Tong, Y. Ma, Y.D. Li, Appl. Catal. A: Gen. 385 (2010) 1–13.
11] M.J. Climent, A. Corma, S. Iborra, Green Chem. 13 (2011) 520–540, and refer-
ences therein.
4
. Conclusion
[
12] V. Degirmenci, E.A. Pidko, P.C.M.M. Magusin, E.J.M. Hensen, ChemCatChem 3
(2011) 969–972.
13] H. Zhao, J.E. Holladay, H. Brown, Z.C. Zhang, Science 316 (2007) 1597–1600.
In conclusion, we have successfully prepared porous TiO2
[
[
nanoparticulate catalysts via biopolymer alginate templating path-
way under hydrothermal conditions and compared the structural
characteristics of the TiO2 materials obtained from the processes.
Three different TiO2 catalysts have been employed to catalyze
the transformation of unutilized sugar derivatives, namely d-
mannose, d-galactose, and lactose to useful platform chemical HMF
14] J.N. Chheda, Y. Roman-Leshkov, J.A. Dumesic, Green Chem.
42–350.
[15] H. Zhao, J.E. Holladay, Z.C. Zhang, US Pat. Appl. (2008) 0,033,187.
[16] G. Yong, Y. Zhang, J.Y. Ying, Angew. Chem. Int. Ed. 47 (2008) 9345–9348.
9 (2007)
3
[
[
17] J.B. Binder, R.T. Raines, J. Am. Chem. Soc. 131 (2009) 1979–1985.
18] Y. Su, H.M. Brown, X. Huang, X.-D. Zhou, J.E. Amonette, Z.C. Zhang, Appl. Catal.
A: Gen. 361 (2009) 117–122.
◦
in DMA–LiCl under microwave irradiation at 140 C, which pro-
[19] S.P.S. Chundawat, B. Venkatesh, B.E. Dale, Biotechnol. Bioeng. 96 (2007)
19–231.
[
2
duced maximum 44% yield. The hydrothermally prepared TiO2
showed considerable mesoporosity. Substrate screening in the cat-
alytic studies revealed that d-mannose is the best candidate to
20] T. Marzialetti, M.B. Valenzuela Olarte, C. Sievers, T.J.C. Hoskins, P.K. Agrawal,
C.W. Jones, Ind. Eng. Chem. Res. 47 (2008) 7131–7140.
[21] P.G. Hobman, J. Dairy Sci. 67 (1984) 2630–2653.
22] Z. Schnepp, S.C. Wimbush, S. Mann, S.R. Hall, CrystEngComm 12 (2010)
410–1415.
[
produce HMF in maximum yield. TiO -H retains its catalytic activ-
2
1
ity for four cycles suggesting potential of our TiO2 nanomaterials
in the biomass conversion.
[23] A.A. Said, R.M. Hassan, Polym. Degrad. Stab. 39 (1993) 393–397.
[24] S. Jagtap, M.K.N. Yenkie, N. Labhsetwar, S. Rayalu, Micropor. Mesopor. Mater.
1
42 (2011) 454–463.
25] D. Chandra, N. Mukherjee, A. Mondal, A. Bhaumik, J. Phys. Chem. C 112 (2008)
668–8674.
26] K.M. Parida, N. Sahu, P. Mohapatra, M.S. Scurrell, J. Mol. Catal. A: Chem. 319
2010) 92–97.
[
[
[
Supporting information
8
(
Recycling results of the conversion of carbohydrates to HMF
under microwave and oil bath heating conditions (Tables S1–S2),
27] S. Dutta, S. De, A.K. Patra, M. Sasidharan, A. Bhaumik, B. Saha, Appl. Catal. A:
Gen. 409–410 (2011) 133–139.
1
13
representative H and C NMR spectra, UV–Vis spectrum and N2
sorption isotherms, pyridine adsorbed FT IR spectra (Figs. S1–S5)
have been provided.
[28] M. Paul, N. Pal, P.R. Rajamohanan, B.S. Rana, A.K. Sinha, A. Bhaumik, Phys. Chem.
Chem. Phys. 12 (2010) 9389–9394.
[
[
29] A.K. Patra, S.K. Das, A. Bhaumik, J. Mater. Chem. 21 (2011) 3925–3930.
30] I.T. Ghampson, C. Newman, L. Kong, E. Pier, K.D. Hurley, R.A. Pollock, B.R. Walsh,
B. Goundie, J. Wright, M.C. Wheeler, R.W. Meulenberg, W.J. DeSisto, B.G. Fred-
erick, R.N. Austin, Appl. Catal. A: Gen. 388 (2010) 57–67.
Acknowledgments
[
31] M.J. Meziani, J. Zajac, D.J. Jones, S. Partyka, J. Roziere, A. Auroux, Langmuir 16
(
2000) 2262–2268.
BS is grateful to the University Grant Commission (UGC), India
for financial support and University of Delhi. SD thanks UGC, India
for a DS Kothari Postdoctoral Fellowship. SD thanks UGC, India
for a Junior Research Fellowship. AKP thanks to CSIR, New Delhi
for Senior Research Fellowship and AB wishes to thank DST, New
[32] S.K. Das, M.K. Bhunia, A.K. Sinha, A. Bhaumik, J. Phys. Chem. C 113 (2009)
8918–8923.
33] J. Papp, S. Soled, K. Dwight, A. Wold, Chem. Mater. 6 (1994) 496–500.
34] H. Yan, Y. Yang, D. Tong, X. Xiang, C. Hu, Catal. Commun. 10 (2009) 1558–1563.
35] J.P. Binder, A.V. Cefali, J.J. Blank, R.T. Raines, Energy Environ. Sci. 3 (2010)
765–771.
[
[
[