G. Zi et al. / Carbohydrate Polymers 115 (2015) 146–151
147
catalyst (Tian et al., 2010). Although considerable efforts have been
made, it is still a major challenge to have a heterogeneous catalyst
with high yield of 5-HMF by direct conversion of cellulose in water.
Metal–organic frameworks (MOFs) possess several advantages
such as high surface areas, well-defined structures, the ease of
processability, and structural diversity, which make them very
attractive for catalysis (Li, Eddaoudi, O’Keeffe, & Yaghi, 1999; Phan,
Nguyen, Luu, & Nguyen, 2012). Recently, phosphotungstic acid
were measured by the nitrogen adsorption/desorption measure-
ments using a Beishide 3H-2000PM2 Surface area and porosity
analyzer. Scanning electron microscopy (SEM) images were taken
on a FEI Quanta200 FEG microscope at an accelerating voltage of
15 kV with the pressure in the sample chamber of 1 Torr. High-
resolution transmission electron microscopy (HRTEM) imagines
were taken on a TEM (JEM-2100). To investigate the acidic prop-
erties of the samples, NH3 Temperature-Programmed Desorption
(
PTA) encapsulated in MIL-101(Cr) was used as a solid acid cat-
(NH -TPD) was carried out using a chemisorption analyzer system
3
alyst for dehydration of carbohydrates to 5-HMF in ionic liquids
or DMSO, an HMF yield of 63% was obtained in DMSO (Zhang,
Degirmenci, Li, & Hensen, 2011). However, a major effort has been
required to functionalize these porous frameworks with specific
sites or acid groups for carrying out the catalytic reactions. And
catalytic studies on MOFs without any functionalization as solid
acid catalysts for conversion of CMC to 5-HMF in single aqueous
solvent system has not been reported so far as we know.
MIL-53(Al) with 1D lozenge-shaped tunnels built up from chains
of octahedra sharing OH vertices is appropriate for a Brønsted-
type acid catalyst (Farrusseng, Aguado, & Pinel, 2009; Ferey, 2008).
And many researchers also reported that acid catalysts are helpful
for accelerating the production of HMF under hydrothermal con-
dition (Kuo et al., 2013; Takeuchi et al., 2007). Therefore, these
properties have inspired potential applications of MIL-53(Al) in
catalysis for hydrolysis of biomass. Moreover, water can still be
considered as a preferred solvent for biomass conversion since the
actual biomass can contain a substantial amount of water, and its
properties are tunable with the variation of its temperature and
pressure (Qi, Watanabe, Aida, & Smith, 2008a). Herein, MIL-53(Al)
without any functionalization was employed as an acid catalyst
for the hydrothermal conversion of CMC to 5-HMF using water as
single solvent for the first time.
(ChemBET 3000).
2.3. Carboxymethyl cellulose hydrolysis
A typical catalytic reaction procedure was as follows: 100 mg
of CMC and 75 mg catalyst were dispersed into 80 mL of water
under stirring. Then the solution was transferred and sealed in a
Teflon-lined steel autoclave and heated to a given reaction tem-
perature at autogenous pressure. After cooling the autoclave in
an ice-water bath to room temperature, the reaction mixture was
filtered through a Millipore filter (pore size, 0.45 m) and ana-
lyzed by UV–vis spectrometer (Shimadzu UV-2401PC photometer)
over 200–400 nm. The determined wavelength is 282 nm (Fig. S4),
which is the maximum absorption wavelength of 5-HMF detected
by UV–vis spectrometer. The concentration of 5-HMF was calcu-
lated based on a standard curve obtained with 5-HMF. In addition to
Ultra-performance liquid chromatography analysis (UPLC, Waters)
1
(Fig. S3), the structure of final product was also analyzed by H NMR
spectroscopy (Bruker Avance 300 spectrometer, 300 MHz, CDCl3) in
order to confirm it is HMF. 1H NMR peaks at d (ppm): 4.660 (s, 2H),
6.447–6.459 (d, J = 3.6 Hz, 1H), 7.140–7.152 (d, J = 3.6 Hz, 1H), 9.546
(s, 1H).
The total reducing sugar (TRS) content was determined using the
3
,5-dinitrosalicylic acid (DNS) method. A mixture contained 1 mL
of 3,5-dinitrosalicylic acid (DNS) regent, 3 mL of reaction sample,
2
. Experimental
and 3 mL of H O, was heated for 15 min in a boiling water bath,
2
then cooled to room temperature, and mixed with 5 mL of deion-
ized water. The color intensity of the mixture was measured in a
UV-2401 Model spectrophotometer. The concentration of TRS was
calculated based on a standard curve obtained with glucose.
2
.1. Synthesis of MIL-53(Al)
MIL-53(Al) was synthesized via
a
hydrothermal method
according to Loiseau et al.’s (2004) previous report. 1,4-
benzenedicarboxylic acid (H BDC, 2.88 g) Al(NO ) ·9H O (13.0 g),
2
3
3
2
2.4. Fast hot catalyst filtration experiments
and deionized water (50 mL) were added into a 100 mL Teflon-
lined steel autoclave and heated at 473 K for 72 h. The product
was filtered and washed with water to ensure the removal of
nitric acid formed during the synthesis. The powder obtained was
dried overnight at 353 K in air. The as-synthesized dry powder
was denoted as MIL-53(Al)as. As-synthesized MIL-53(Al)as was acti-
vated according to the previous literature procedure (Rallapalli
et al., 2009). The product was filtered and washed with N,N-
dimethylformamide (DMF) to remove the unreacted BDC, and dried
under vacuum for 2 h. Further it was treated with methanol in order
to replace the DMF molecules trapped inside the cavities of the
product. Finally, it was filtered, washed with methanol, and dried
in an air oven at 353 K for 2 h (Rallapalli et al., 2009). Then, the
sample was further activated at 423 K under vacuum overnight to
ensure complete removal of solvent molecules. The obtained prod-
uct was calcined in conventional manner at 603 K for 72 h and the
calcined product was denoted as MIL-53(Al).
1
00 mg of CMC and 75 mg catalyst were dispersed into 80 mL
of water and loaded into a stainless steel autoclave with a 100 mL
Teflon container. Then, the autoclave and its contents were heated
to 473 K at autogenous pressure for 30 min, then quickly filtrated
the catalyst. The filtrates obtained from the fast hot catalyst fil-
tration was detected by ICP-AES analyzer (Shimadzu SEQUE-NTIAL
PLASMA SPETROMETER ICPS-1000II).
3
. Results and discussion
3
.1. The characterization of catalyst
The diffraction patterns of MIL-53(Al)as and MIL-53(Al) sam-
ples clearly indicates that the materials are well crystalline, and the
obtained peaks of MIL-53(Al)as and MIL-53(Al) are in good agree-
ment with the literature patterns reported earlier by Ferey et al.
(Fig. 1) (Loiseau et al., 2004). The diffraction patterns are obviously
2.2. Characterizations
different, since the guest removal by heating leads to an expansion
of the cell for MIL-53(Al).
Wide angle X-ray powder diffraction (XRD) experiments were
The FT-IR spectra of MIL-53(Al)as and MIL-53(Al) (Fig. 2) exhibit
−
1
conducted on a Rigaku TTRAX III spectrometer with Cu K␣ radia-
vibration bands in the region 1400–1700 cm for the carboxylic
functional groups (Loiseau et al., 2004). The FT-IR of MIL-53(Al)as
◦
◦
tion from 5 to 40 . The Fourier transform infrared spectra (FI-IR)
measurements were performed on a Thermo Nicolet 8700 instru-
ment. Pore size distributions, BET surface areas and pore volumes
−
1
−1
samples exhibits bands at 1601 cm and 1512 cm corresponds
−
1
to COO asymmetric stretching whereas the bands at 1413 cm