Md.I. Alam et al. / Applied Catalysis A: General 486 (2014) 42–48
43
program version 4.3.6. The yield of HMF was determined from 1
H
NMR signals with reference to known amount of internal standard.
Conversions were calculated by measuring unconverted reduced
sugars by phenol sulfuric acid analysis method [32].
face area (740 m2 g−1) was realized when the same material
phoric acid solution and titanium propoxide or titanium chloride
in the presence of cationic surfactant [19]. Apart from cationic
surfactant, high surface area mesoporous titanium phosphate has
also been synthesized using anionic surfactant [31]. Most recently,
Dutta et al. [22] developed hierarchical macro/meso-porous tita-
nium phosphate through a slow evaporation method by using
titanium isopropoxide and orthophosphoric acid as inorganic
sources and Pluronic P123 as a structure directing agent. However,
the use of expensive precursors and template molecules, difficul-
ties in removal of template/surfactant, low yield and long reaction
time are some drawbacks of these methods. Therefore, a simple
method for the synthesis of TiP materials is more desirable.
This paper describes a simple approach for the synthesis of
acidic titanium hydrogenphosphate (TiHP), and its catalytic effec-
tiveness for the dehydration of carbohydrate substrate in bi-phasic
medium. The said TiHP material is prepared by refluxing commer-
cial TiO2 and H3PO4 at 160 ◦C without any template molecule, and
characterized by FTIR, XRD, HRTEM, CHNS elemental analyzer, TGA
and NH3-TPD methods. As-synthesized TiHP exhibits good catalytic
activity for HMF preparation in water/THF biphasic solvent under
conventional heating. A comparison of small scale reactions carried
out in a 50 mL pressure reactor and a larger scale reaction in 2 L
Parr reactor under mechanical stirring reveals that the later reac-
tion gives higher yields because of efficient stirring. Because of high
stability and water-tolerant nature of the TiHP, recovered catalyst
retained activity upon recycling. Material characterization data of
the recovered catalyst compared well with that of as-synthesized
material.
2.2. Preparation of TiHP
TiO2 (1.60 g) and H3PO4 (2.45 g) were mixed with distilled water
(5 mL) in a round bottom flask. The reaction mixture was heated
to reflux with continuous stirring at 160 ◦C for 8 h using chilled
water fitted with a condenser. The mixture was then cooled to room
temperature, filtered and washed with hot water until pH of filtrate
is neutral. The resultant white solid was oven dried at 65 ◦C for
12 h. A schematic representation for catalyst preparation is shown
in supporting information (Fig. S1).
2.3. General reaction procedure
Small scale experiments were performed in a 50 mL stain-
less steel reactor equipped with temperature controller, pressure
gauge, stirrer and rupture disk. The larger scale reactions (10–30 g
substrate) were carried out in a 2 L Parr reactor equipped with
mechanical stirrer, pressure and temperature controller, a chillier
and rupture disk. In typical small scale experiments, the reactor
was charged with NaCl saturated water as a reactive phase, THF as
an extracting phase, substrate and catalyst. The reactor head was
sealed tightly before heating the reaction mass at the desired tem-
perature for a set time. After completion of reaction, the mixture
was allowed to cool down to room temperature. The organic phase
was separated in a round bottom flask. The remaining HMF in the
aqueous phase was further extracted by ethyl acetate and collected
in the round bottom flask. After removing solvent by rotary evap-
orator under vacuum, crude HMF was analyzed by 1H NMR. The
aqueous layer was also analyzed by UV–vis spectrophotometric
method for measuring any remaining HMF in the aqueous phase.
2. Experimental
2.1. Materials and instrumentation
2.4. Determination of HMF yield
TiO2, d-glucose, d-fructose and galactose were purchased from
Thomas Baker (India). H3PO4 and tetrahydrofuran were supplied by
Fischer Scientific and S D Fine Chemical (India), respectively. Cel-
lobiose and maltose were purchased from Sigma Aldrich. Unless
otherwise stated, distilled water was used as an aqueous phase.
Fourier transform infrared (FTIR) spectra of TiHP were recorded on a
Perkin-Elmer spectrometer (model 1752X FTIR) by using KBr pellet.
Powder X-ray diffraction of the TiHP was carried out using a Brucker
Advance D-8 machine. Thermo gravimetric analyses (TGA) were
performed on a Shimadzu (model DTG-60) thermal analyzer under
air at a heating rate of 10 ◦C min−1. HR-TEM (transmission elec-
tron microscopy) images were recorded on a JEOL DATUM Model
No. JEM1011. SEM image was analyzed on Hitachi, S-4100 at an
accelerating voltage of 20 kV. N2 adsorption–desorption isotherms
were obtained by using a Beckman Coulter SA 3100 Surface Area
Analyzer at 77 K. Elementar Analysensysteme GmbH VarioEL V3
elemental analyzer was used for measuring hydrogen content of
the TiHP. Temperature programmed desorption (TPD) of ammonia
was performed on ALTAMIRA Q-800 instrument equipped with an
analyzer and a thermal conductivity detector. In this method, the
sample was activated at 450 ◦C for 2 h in a U-tube glass cell under
He flow and cooled to 150 ◦C. The sample was then saturated with
ammonia for 30 min at a rate of 20 mL min−1. After exposure to
ammonia, the material was subsequently purged with He to mea-
sure desorption of ammonia in the range of 100–900 ◦C at a heating
rate of 10 ◦C min−1. The catalytic conversions of carbohydrate sub-
strates into HMF were performed in stainless steel Parr reactors.
1H NMR spectral analysis was performed on a JEOL JNM ECX-400 P
400 MHz instrument and data were processed using a JEOL DELTA
The yields of HMF in the organic and aqueous phases were
measured separately by 1H NMR and UV–Vis spectrophotometric
techniques, respectively. Since NaCl saturated aqueous solution
was used to increase partition coefficient of the organic products
and the reactive phase was washed four times with ethyl acetate,
the amount of HMF remaining in the aqueous phase was neg-
ligible as compared to the organic phase. Crude HMF, obtained
after removal of extracting phase, was analyzed by 1H NMR by
using a known amount of mesitylene as internal standard. The
yield of HMF was calculated from integrated values of the CHO
proton (ı = 9.58 ppm) of HMF and the three aromatic ring pro-
tons of mesitylene (ı = 6.78 ppm) (Fig. S2). The aqueous phase
was analyzed directly by UV-Vis spectroscopic technique as HMF
has a distinct peak at 284 nm with the corresponding molar
extinction coefficient () of 1.66 × 104 M−1 cm−1. Unless otherwise
mentioned, the reported HMF yields (mol%) are the sum of HMF
obtained from both organic and aqueous phases.
3. Results and discussion
3.1. Catalyst characterization
The FTIR spectrum of as-synthesized TiHP material (Fig. 1(a))
shows characteristic band of phosphate group at about 1000 cm−1
and a band at 1260 cm−1 for P
O H deformation mode, which
are absent in the precursor TiO2. This comparison suggests that
acidic hydrogen phosphate group is successfully incorporated on