High performance mesoporous zirconium phosphate for dehydration of xylose to furfural in aqueous-phase

Article abstract of DOI:10.1039/c3ra43413c

The conversion of sugars to chemicals in aqueous-phase is especially important for the utilization of biomass. In current work, zirconium phosphate obtained by hydrothermal methods using organic amines as templates has been examined as a solid catalyst for the dehydration reaction of xylose to furfural in aqueous-phase. The use of dodecylamine and hexadecylamine in the synthesis process results in mesoporous zirconium phosphate with uniform pore width of ～2 nm and in morphology of nanoaggregates, which is characterized by powder X-ray diffraction, N₂ isothermal sorption, NH₃ temperature-programmed desorption, FT-IR, and ³¹P MAS NMR spectroscopy. When used as a catalyst for xylose dehydration to furfural in aqueous-phase, the mesoporous zirconium phosphate presents excellent catalytic performance with high conversions up to 96% and high furfural yields up to 52% in a short time of reaction. Moreover, the catalyst is easily regenerated by thermal treatment in air and shows quite stable activity. The open structure with numerous active sites of the Bronsted/Lewis acid sites is responsible for the high catalytic efficiency of mesoporous zirconium phosphate.

Full text of DOI:10.1039/c3ra43413c

RSC Advances
View Article Online
View Journal | View Issue
PAPER
High performance mesoporous zirconium phosphate for
dehydration of xylose to furfural in aqueous-phase
Cite this: RSC Adv., 2013, 3, 23228
Liyuan Cheng,† Xiangke Guo,† Chenhai Song, Guiyun Yu, Yuming Cui, Nianhua Xue,*
Luming Peng, Xuefeng Guo and Weiping Ding*
The conversion of sugars to chemicals in aqueous-phase is especially important for the utilization of
biomass. In current work, zirconium phosphate obtained by hydrothermal methods using organic
amines as templates has been examined as a solid catalyst for the dehydration reaction of xylose to
furfural in aqueous-phase. The use of dodecylamine and hexadecylamine in the synthesis process results
in mesoporous zirconium phosphate with uniform pore width of ꢀ2 nm and in morphology of
nanoaggregates, which is characterized by powder X-ray diffraction, N
2 3
isothermal sorption, NH
31
temperature-programmed desorption, FT-IR, and P MAS NMR spectroscopy. When used as a catalyst
for xylose dehydration to furfural in aqueous-phase, the mesoporous zirconium phosphate presents
excellent catalytic performance with high conversions up to 96% and high furfural yields up to 52% in a
short time of reaction. Moreover, the catalyst is easily regenerated by thermal treatment in air and
shows quite stable activity. The open structure with numerous active sites of the Brønsted/Lewis acid
sites is responsible for the high catalytic efficiency of mesoporous zirconium phosphate.
Received 4th July 2013
Accepted 1st October 2013
DOI: 10.1039/c3ra43413c
www.rsc.org/advances
3
1
32
titanoniobate nanosheets, vanadium phosphates, and other
acidic oxides.
1. Introduction
33–37
Lewis acid is an efficient catalyst for xylose
Furfural is an important heterocyclic compound and organic
intermediate. Lots of derivatives which have been widely used in
synthetic plastics, pharmaceuticals, pesticides industries could
isomerization to xylulose which is more rapidly dehydrated to
38
furfural on a Brønsted acid. It has been reported the furfural
selectivity in xylose dehydration is a direct function of the density
33
be obtained through hydrogenation, oxidation, and condensa- of Brønsted acid sites. Therefore, a suitable ratio between the
tion of furfural. The production of furfural is rstly realized by densities of Brønsted acids and the Lewis acids and the cooper-
utilizing mineral acids (mainly sulfuric acid) as catalysts from ation among them over the solid acid catalysts are much more
1
renewable biomass-derived carbohydrates. Chloride salts are important to produce furfural in high efficiency. Due to the
2
oen used to increase the furfural production. The biomass relative strong Brønsted acidity of hydroxyl groups together with
rich in pentosans such as corn cob, sugar cane bagasse, oat
hulls could be used to produce furfural. The strong corrosive
effect on experiment apparatus, difficult recycle and reuse,
the relative strong Lewis acidity of the metal centers simulta-
neously, metal phosphates have attracted considerable attention
as solid acid catalysts. In an earlier report, zirconium phosphate
serious environmental pollution of sulfuric acid encourage the has a layered structure consisting of zirconium atom planes and
exploring of novel catalytic processes green to environment as planes of phosphate species, and it exhibits both Lewis and
39
well as with economic properties.
Brønsted acidity. Zirconium phosphate (ZrP) catalyst has shown
A variety of solid catalysts have been employed in xylose interesting performance for xylose conversion to furfural among
3
–11
dehydration reaction, e.g. zeolites,
mesoporous molecular the solid acid catalysts such as SiO –Al O , WO /ZrO , g-Al O
2 2 3 X 2 2 3
13
12
33
sieve MCM-41, MCM-41-supported niobium-oxide, hetero-
and HY zeolite.
1
4–16
17–19
polyacids,
sulfated TiO , ZrO and SnO ,
modied silicas
2
0–24
2
2
In terms of solvent used in the dehydration of xylose, organic
and polar solvent such as dimethylsulfoxide (DMSO) or mixed
2
25
with porous structure,
sulfonated graphene oxide, micro-
porous and mesoporous niobium silicates (AM-11), micropo- solvent such as water–toluene (W/T) are adopted by many
rous silicoaluminophosphates, solid acid and base in one-pot,
26
27
28
research groups to promote the yield of furfural, due to the
29,30
ion-exchange resins,
exfoliated titanate, niobate and timely extraction and separation of the product and the reac-
tant. The decomposition of DMSO at high temperatures in the
presence of air and the use of large quantities of toluene,
however, would be questions to meet the actual demand.
Considering the demands on green routes for the furfural
production in the future, water should be the most suitable and
Key Lab of Mesoscopic Chemistry, the School of Chemistry and Chemical Engineering,
Nanjing University, Nanjing 210093, China. E-mail: nianhua@nju.edu.cn; dingwp@
nju.edu.cn; Fax: +86 25 83317761; Tel: +86 25 83595077
†
The rst two authors contribute equally to this investigation.
2
3228 | RSC Adv., 2013, 3, 23228–23235
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
RSC Advances
convenient solvent for the practical utilization of sustainable Table 1 Structural and surface properties of the zirconium phosphate samples
biomass. Some reports have achieved interesting results in the
aqueous media for xylose conversion. The catalysts include
a
b
c
c
S
BET
V
p
Brønsted acid
Lewis acid
(mmol g
2
ꢂ1
3
ꢂ1
ꢂ1
ꢂ1
Sample
(m g
)
(cm g
)
(mmol g
)
)
3
7
TiO /CB (carbon black), mesoporous (zirconium, aluminium)-
2
35
9
tungsten, and aluminosilicates. Considering both Lewis and ZrP-HF-C
8
20
316
—
0.02
0.05
0.55
—
—
—
229
—
—
—
307
—
39
Brønsted acidity in the zirconium phosphate and based on our ZrP-HT-C
ZrP-HT-Am-C
previous work on morphology controlled synthesis of metal
H
3 4
PO (1.3%)
40
phosphates, a convenient route to mesoporous zirconium
phosphate has been developed in this work. The zirconium
phosphate obtained through hydrothermal method with mixed
a
b
BET specic surface arcea.
Pore volume calculated from N
2
adsorption at p/p
0
ꢀ 0.97. Calculated based on published extinction
ꢂ1
41
coefficients for the bands at 1545 and 1454 cm
.
ꢁ
amines as templates at high temperature (180 C) has large
surface area and mesoporous structure. It has offered excellent
catalytic performance for dehydration of xylose to furfural in
isotherm, indicative of typical mesoporosity upon the addition
of organic amines in the preparation. Both ZrP-HT-C and
water with high yield of furfural up to 52% in 2 h of reaction at
ꢁ
1
70 C. The catalyst is also easily regenerated by thermal
2
ꢂ1
ZrP-HF-C show low BET specic surface area (20 and 8 m g
,
treatment and exhibits good performance in reuses.
respectively), but the ZrP-HT-Am-C sample exhibits much
2
ꢂ1
higher surface area (316 m g ). The pore size distributions
shown in Fig. 2b reveals a broad one in ZrP-HF-C and a relatively
narrow one centered at 6 nm in ZrP-HT-C. A narrow pore size
distribution centered at 2 nm is observed in ZrP-HT-Am-C with
much larger pore volume.
2. Results and discussion
2.1. Characterization
XRD patterns of a series of prepared zirconium phosphate (ZrP)
samples are shown in Fig. 1. The as-prepared ZrP-HT, ZrP-HF
can be indexed as Zr(HPO ) $H O (JCPDS PDF no. 34-0127). The
SEM images of ZrP-HT-C and ZrP-HF-C, shown in Fig. 3A and
B, reveal the samples as general exfoliation of layered materials
without the existence of mesopores, in agreement with their low
surface area. The TEM images of ZrP-HT-Am-C, depicted in
Fig. 3C and D, indicate the sample consisting of small particles
connected each other with rich mesoporous structure to
resulting in high surface area as measured by nitrogen sorption.
The FT-IR spectra of the samples are shown in Fig. 4. For the
4
2
2
two samples present phase transformation aer the loss of
water in crystal upon heating and the calcined samples, named
as ZrP-HT-C, ZrP-HF-C, can be indexed as Zr
2 4 3
H(PO ) (JCPDS
PDF no. 38-0004) and ZrP (JCPDS PDF no. 36-0188),
2 7
O
respectively. The ZrP-HT-Am-C, with weak diffraction peaks,
could be rather dened as amorphous. The BET (Brunauer–
Emmett–Teller) specic surface areas and pore volumes
obtained by the nitrogen physisorption are presented in Table
ꢂ1
ZrP-HT-Am sample, the peak of 1626 cm is attributed to
bending vibrations of water molecules, while the bands in the
1. As shown in Fig. 2a, the sorption isotherm of ZrP-HT-Am-C
ꢂ1
region 1000–1200 cm are due to –PO
4
stretching vibrations.
presents a signicant hysteresis loop at p/p > 0.45 with type IV
ꢂ1
0
The bands at 2921 and 2846 cm are attributed to stretching
vibrations of –CH in the organic amines. The bands at
2
ꢂ1
1464 and 720 cm are from the distorted vibration absorptions
of –CH . The IR spectrum of ZrP-HT-Am-C aer calcination
2
shows a relatively simple structure along with disappearances of
2
characteristic peaks of –CH . For ZrP-HT and ZrP-HF, the peaks
ꢂ1
of 3590, 3507 and 3135 cm correspond to the symmetric and
4
2
asymmetric stretching vibrations of lattice water, respectively.
ꢂ1
The peak of 1615 cm is attributed to bending vibrations of
ꢂ1
water molecules. The absorbance at 1247 cm
is usually
attributed to a hydrogen phosphate characteristic peak. Also the
Fig. 1 XRD patterns of prepared zirconium phosphates. (ZrP-HT: zirconium
phosphate synthesized without amines; ZrP-HT-C: ZrP-HT calcined; ZrP-HF: zirco-
nium phosphate synthesized with HF; ZrP-HF-C: ZrP-HF calcined; ZrP-HT-Am-C:
zirconium phosphate as synthesized with organic amine after calcination).
Fig. 2 (a) N
2
physisorption isotherms at ꢂ196 ꢁC and (b) pore size distribution
curves of the samples.
This journal is ª The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 23228–23235 | 23229

View Article Online
RSC Advances
Paper
Fig. 3 FESEM images of (A) ZrP-HT-C and (B) ZrP-HF-C. (C) and (D) TEM images of ZrP-HT-Am-C (inset shows the magnified part as pointed).
3
Fig. 5 (a) NH -TPD profiles and (b) pyridien-IR spectra of ZrP-HF-C, ZrP-HT-C and
ZrP-HT-Am-C samples.
the P–OH groups and/or some strong Lewis acid sites. The
desorption of the ammonia at high temperature is absent in the
other two samples ZrP-HF-C and ZrP-HT-C, implying the lack of
large quantity of strong acid sites in these two samples. The IR
spectrum of chemically adsorbed pyridine on mesoporous ZrP-
HT-Am-C sample reveals the coexistence of Lewis and Brønsted
Fig. 4 FT-IR spectra of samples before and after calcination.
ꢂ1
ꢂ1
acid sites, by bands of absorption at 1454 cm and 1545 cm
,
as shown in Fig. 5b. The quantitative results on density of
Brønsted and Lewis acid sites are obtained from the pyridine
adsorption IR spectrum and listed in Table 1. In contrast, no
obvious absorption of the adsorbed pyridine is detected from
the IR spectra on both ZrP-HF-C and ZrP-HF-C samples aer
ꢂ1
set of peaks in range of 1000–1200 cm are belonged to –PO
stretching vibrations.
4
The NH
tion peak appeared around 120 C for all three samples is
attributed to desorption of NH weakly bound to the surface of
3
-TPD results are shown in Fig. 5a. A broad desorp-
ꢁ
ꢁ
evacuation at 150 C (Fig. 5b).
3
the zirconium phosphate samples. Only small amount of des-
orbed ammonia is detected from ZrP-HF-C and ZrP-HT-C
2.2. Xylose dehydration
sample due to their low density of acid sites. For ZrP-HT-Am-C, a
ꢁ
desorption peak centered at ꢀ260 C, besides the peak around Scheme 1 shows the reaction of xylose dehydration to furfural.
ꢁ
1
20 C, is also observed, indicative of the ammonia adsorbed on Considering the demand on green processes, the xylose
2
3230 | RSC Adv., 2013, 3, 23228–23235
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
RSC Advances
of H
3
PO
4
(1.3 wt%, equal to the amount of P species in the
ZrP-HT-Am-C used) is also performed, in which 33% of furfural
selectivity and 14% of xylose conversion are detected. For the
dehydration of xylose to furfural, it has been reported that
the xylose isomerizes to xylulose on a Lewis acid and then the
38
xylulose dehydrates to furfural on a Brønsted acid. According
to the structure of the current ZrP-HT-Am-C, it is neither a
strongly acidic material nor a strongly basic material but
possesses large number of Lewis acid and Brønsted acid
simultaneously. Its good performance for the conversion reac-
tion of xylose would be assigned to the effective cooperation
between the Lewis and Brønsted acid sites.
Scheme 1 The dehydration reaction of xylose to furfural.
The effect of xylose/catalyst ratio on the xylose dehydration
reaction is investigated with ZrP-HT-Am-C as catalyst and the
results are shown in Fig. 7a. The conversion of xylose mono-
tonically increases with the catalyst amount. But the best
selectivity to furfural is observed at the ratio of xylose/catalyst
around 4. With the increase in reaction temperature, as shown
in Fig. 7b, the xylose conversion increases monotonically and
the selectivity to furfural basically keeps at constant in the range
ꢁ
ꢁ
of 140–170 C. At 170 C, 96% conversion and 52% yield to
furfural are measured. It is very interesting that the selectivity to
furfural over the ZrP-HT-Am-C is almost constant versus the
conversion of xylose (Fig. 7c), implying the continuous reaction
of furfural on the ZrP-HT-Am-C catalyst could not happen. This
should be very important phenomenon offering a research clew
Fig. 6 Conversion of xylose and yield of furfural in the presence of different
ꢁ
catalysts. Reaction conditions: xylose (0.4 g), catalyst (0.1 g), water (40 mL), 140 C,
2
h.
dehydration is performed in an aqueous system and the for the future development of new catalyst based on zirconium
ꢁ
primary results of reaction at 140 C with reaction time of 2 h phosphate.
are summarized in Fig. 6. In a blank test, 13% conversion of
The reaction results of current work are summarized in
xylose and low selectivity of ꢀ13% to furfural are detected due Table 2, and some results of xylose conversion to furfural with
to the action of high temperature water. For the catalysts of ZrP- water as solvent reported in literature are also listed. The
HF, ZrP-HT, ZrP-HF-C and ZrP-HT-C, the selectivity to furfural is performance of current ZrP-HT-Am-C is amongst high
signicantly improved, but the conversion of xylose is still very performance catalysts documented. Considering the catalyst
low (<12%). The as-prepared ZrP-HT-Am with presence of concentration used in batch reactor, the present catalyst ZrP-
ꢂ1
amines shows improved conversion (20%) and selectivity (34%). HT-Am-C with load of 2.5 gcat
L
gives fairly high conversion
Aer calcination, ZrP-HT-Am-C exhibits inspiring catalytic of xylose and selectivity to furfural at the same time in
performance with 52% conversion and 50% selectivity. For aqueous-phase reaction. The solvent effect on the furfural
comparison, the catalytic reaction in the presence of liquid acid production from the xylose dehydration over ZrP-HT-Am-C is
ꢁ
Fig. 7 (a) Effect of amount of ZrP-HT-Am-C on furfural production from the xylose dehydration reaction. Xylose: 0.4 g, water: 40 mL, temp.: 140 C, reaction time: 2 h.
(
b) Variation of catalytic performance with temperature over ZrP-HT-Am-C. Xylose: 0.4 g, cat.: 0.1 g, water: 40 mL, reaction time: 2 h. (c) Furfural selectivity versus xylose
ꢁ
ꢁ
ꢁ
conversion at 140–180 C in aqueous reaction for 2 h. xylose: 0.4 g, cat.: 0.1 g, water: 40 mL, 100 C; xylose: 0.4 g, cat.: 0.1 g, water: 40 mL, 120 C; xylose 0.4 g,
ꢁ
ꢁ
ꢁ
catalyst 0.05 g, water: 40 mL, 140 C; xylose 0.4 g, catalyst 0.2 g, water: 40 mL, 140 C; xylose 0.4 g, catalyst 0.3 g, water: 40 mL, 140 C; xylose: 0.4 g, cat.: 0.1 g,
ꢁ
ꢁ
water: 40 mL, 140 C; xylose: 0.4 g, cat.: 0.1 g, water: 40 mL, reuse after 2 h reaction at 140 C; xylose: 0.4 g, cat.: 0.1 g, water: 40 mL, second reuse after 2 h reaction
ꢁ
ꢁ
ꢁ
ꢁ
at 140 C; first regeneration by calcination in air at 500 C after 2 h reaction at 140 C; second regeneration by calcination at 500 C in air after 2 h reaction at
ꢁ ꢁ ꢁ
40 C; xylose: 0.4 g, cat.: 0.1 g, water: 40 mL, 160 C; xylose: 0.4 g, cat.: 0.2 g, water: 40 mL, 180 C.
1
This journal is ª The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 23228–23235 | 23231

View Article Online
RSC Advances
Paper
a
Table 2 Performance of different catalysts in the dehydration of the xylose to furfural in a water system
Xylose
conv. (%)
Furfural
yield (%)
ꢁ
ꢂ1
Catalysts
Solvent
DMSO
T( C)
Time (h)
Xylose/catalyst
Catalyst (gcat
L
)
Ref.
ZrP-HT-Am-C
ZrP-HT-Am-C
ZrP-HT-Am-C
ZrP-HT-Am-C
140
140
140
170
170
170
170
170
170
2
2
2
2
2.5
8
32
1
4
60
16
52
96
94
93
91–97
32
80
15
12
26
52
47
47
54–56
18
24
4
4
4
4
2.5
2.5
2.5
2.5
10
20
20
3.3
20
This work
This work
This work
This work
37
35
9
21
32
b
H
H
H
H
H
H
H
H
2
2
2
2
2
2
2
2
O/toulene
O
O
O
O
O
O
O
TiO
2
/CB
3
Zr(W,Al) mixed oxides
H-MCM-22, ITQ-2
MSHS
1.5
1.5
9
VPO
1.5
a
b
Reaction conditions of this work, xylose: 0.4 g; cat.: 0.1 g; solvent: 40 mL.
2
H O–toluene in 1 : 1 volume.
also tested, as shown in Table 2. In general, for current catalyst
ZrP-HT-Am-C, both the DMSO and mixed H O–toluene
2
systems give inferior performance for the reaction, indicating
the zirconium phosphate material prepared in present work is
an effective catalyst for xylose dehydration to furfural in the
water system.
The reusability of the ZrP-HT-Am-C catalyst is tested with
the catalyst separated by ltration from the reaction suspen-
sion. The catalyst ltrated for reuse is thermally treated in air
ꢁ
at 500 C for 4 h before the next run. The results are shown in
Fig. 8 and it can be concluded that the thermal treatment
between reuse intervals basically regenerates the catalyst
performance. The catalyst shows fairly good property and
stability for the reaction of xylose dehydration to furfural.
Simultaneously, only trace amount of zirconium is detected by
ICP analysis in the solution with catalyst separated aer reaction.
The variation of the catalytic property of ZrP-HT-Am-C with the
reaction time is also tested and the results are shown in Fig. 9.
Fig. 9 The catalytic performance of ZrP-HT-Am-C sample upon the reaction time
ꢁ
at 170 C.
The excellent performance of ZrP-HT-Am-C catalyst should
ꢁ
At 170 C, the conversion of xylose reaches 96% in 2 h of
be ascribed to its open mesoporous structure with numerous
and accessible active sites. The Zr/P ratio in the sample was
determined as 0.8 by XRF analysis. And the state of phosphorus
reaction with 52% selectivity to furfural. In prolonged time of
reaction, the slight decrease in the selectivity to furfural
indicates the minor transformation of furfural to by-products,
implying the fairly good stability of furfural under the reaction
conditions.
31
species in ZrP-HT-Am-C is detected by solid sate P MAS NMR
spectroscopy. The spectrum is presented in Fig. 10. The broad
line between ꢂ15 and 35 ppm is attributed to the P species
neighboring to Zr species. There are no signals of pyrophos-
43
phates which appeared in the range of ꢂ35 to ꢂ45 ppm. Subtle
acid structure in this special materials, such as the quantity
and relative position of the Brønsted and Lewis acid sites, is
ꢁ
Fig. 8 Reuse study for the ZrP-HT-Am-C catalyst at 170 C after 2 h reaction.
Reaction conditions: xylose (0.4 g), catalyst (0.1 g), H
2
O (40 mL). For the second
ꢁ
31
and third run, the catalyst is regenerated at 500 C for 4 h in air.
Fig. 10 Solid state P MAS NMR spectrum of ZrP-HT-Am-C.
2
3232 | RSC Adv., 2013, 3, 23228–23235
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
RSC Advances
responsible for its excellent performance for the reaction. The
The specic surface areas and pore volumes of samples were
delicate cooperation among the Brønsted and Lewis acid sites is obtained by means of nitrogen adsorption measurements at
ꢁ
full of charm of research, which causes the adsorbed xylose ꢂ196 C using an automated gas adsorption analyzer (Micro-
ꢁ
transformed effectively and selectively. Related investigation meritics ASAP 3020). Samples were pretreated at 300 C for 2 h
using NMR method with designed probe molecules is under under vacuum. Total surface area was calculated using the
way in our lab. The current results are important for the future classical BET (Brunauer–Emmett–Teller) method and the pore
research on the conversion of sugars to useful chemicals.
volume was evaluated with the t-plot method using the equation
of Harkins and Jura.
Field emission scanning electron microscopy (FESEM)
analysis was carried out on a Hitachi-S4800. Transmission
electron microscopy (TEM) images were taken with a JEM-2100
using an accelerating voltage of 200 kV. FT-IR spectra of the
prepared zirconium phosphates were recorded on Nicolet iS10
3
. Experimental
3.1. Catalyst preparation
The zirconium phosphate was synthesized by hydrothermal
method. In a typical synthesis, 24 mL of ethanol solution
containing 4.0 g dodecylamine (J & K Technology Co., Ltd.) and
40
ꢂ1
spectrometer using ATR accessory with a resolution of 4 cm
and accumulation of 64 scans.
0.50 g hexadecylamine (J & K Technology Co., Ltd.) was pumped
ꢂ1
3
Ammonia temperature-programmed desorption (NH -
slowly (0.25 mL min ) into 80 mL of aqueous solution con-
taining zirconium nitrate pentahydrate (Sinopharm Chemical
Reagent Co. Ltd, 3.0 g) and ammonium dihydrogen-phosphate
TPD) was carried out in a PX200 apparatus (Tianjin Golden
Eagle Technology Limited Corporation). Prior to the
measurements, the sample (0.1 g) charged into the quartz
ꢁ
(Nanjing Chemical Reagent Co., Ltd., 1.03 g) at 40 C. The
ꢁ
U-tube was rstly treated in a He stream at 500 C for 1 h at a
mixture was transferred into a Teon-lined autoclave and sealed
before hydrothermally heating in an oven at 180 C for 8 h, and
then cooled down and centrifuged to obtain the solid product.
The solid was further washed by water and anhydrous ethanol
repeatedly and dried under vacuum at 60 C for 24 h. The
ꢁ
ꢂ1
ꢁ
rate of 10 C min . Then the temperature was decreased to
ꢁ
5
0
C and NH
3
was switched into the system under a ow
ꢂ1
of He (40 mL min ) for 0.5 h adsorption. The desorption of
ꢁ ꢁ
ammonia was performed from 50 C to 500 C at a rate of
ꢁ
ꢁ
ꢂ1
ꢁ
10 C min . The desorbed NH was detected using a thermal
resulted sample was further subjected to calcination at 500 C
3
conductivity detector (TCD).
for 4 h in an air atmosphere. The samples before and
aer calcination treatment were noted as ZrP-HT-Am and
ZrP-HT-Am-C, respectively.
The other zirconium phosphate using hydrothermal method
without organic amine was prepared as follows. Zirconium
oxychloride octahydrate (Sinopharm Chemical Reagent Co.,
Ltd, 0.68 g) was dissolved in 25 mL water at room temperature,
the concentrated phosphoric acid (1 mL) was dropwise into the
zirconium oxychloride aqueous solution at continuous stirring.
Then the mixture was transferred to the sealed Teon-lined
In terms of pyridine IR, the sample was pressed into self-
ꢁ
supporting wafer and activated at 200 C for 1 h in vacuum
ꢂ3
(
pressure < 10 mbar). Aer cooling to room temperature, a
spectrum was recorded. Pyridine (1 mbar) was then adsorbed at
ꢁ
1
50 C for 0.5 h. Aer outgassing for 0.5 h to remove phys-
isorbed pyridine, a spectrum was recorded using the spectrum
of activated wafer as a background.
3
1
Solid-state P MAS NMR measurement was performed on a
.4 T Bruker Avance spectrometer at a resonance frequency of
9
ꢁ
161.97 MHz. Powdered sample was spun at 14 kHz in a 4 mm
rotor under ambient temperature. Spectrum was acquired
using a single pulse excitation pulse sequence (zg) with a 2.5 ms
autoclave and put into an oven at 180 C for 6 h. Aer hydro-
thermal treatment, the mixture was cooled down, centrifuged
and washed. The resulted product was dried under vacuum at
3
1
ꢁ
ꢁ
pulse length and 600 s recycle delay. P chemical shis were
60 C for 24 h and then subjected to the oven at 500 C for 4 h in
referenced to external ammonium dihydrogen phosphate
air. The samples before and aer calcination treatment were
marked as ZrP-HT and ZrP-HT-C, respectively.
(
0.81 ppm).
The zirconium phosphate sample using hydrouoric acid
(
HF) method was obtained according to the synthesis procedure
3
.3. Catalysis tests
44
by Alberti. Simply, a certain amount of zirconium oxychloride
octahydrate, concentrated phosphoric acid, hydrochloric acid
and hydrouoric acid were mixed uniformly in a baker under
continuous stirring at room temperature for 96 h. The white
precipitate was then centrifuged and washed. The drying and
calcination process were the same with the two samples above.
The samples were named as ZrP-HF and ZrP-HF-C, respectively.
The batch catalytic experiments were performed in the Amar
autoclave (total volume, 100 mL). Typically, xylose (0.4 g),
powdered catalyst (0.05–0.3 g) and H O (40 mL) were poured
2
into the autoclave, and then the reaction mixtures were heated
to the target temperature under continuous stirring at 500 rpm.
The mixture aer the reaction was cooled with recycled water at
once. Xylose conversion and furfural selectivity are calculated
using the formula as follows:
3
.2. Catalyst characterization
0
N
ꢁ
XRD patterns of samples were recorded in the 2q region of 5 –
9
Conversion% ¼ N ꢃ 100
(1)
0
ꢁ
0 using a diffractometer (Philips X'pert-Pro) with CuKa radi-
˚
ation (l ¼ 1.5418 A). The identication of crystalline phases was
N
0
0
: moles of xylose initially added to the autoclave.
made by matching the patters with the JCPDS les.
N : moles of xylose consumed.
This journal is ª The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 23228–23235 | 23233

View Article Online
RSC Advances
Paper
N
1
9 M. M. Antunes, S. Lima, A. Fernandes, M. Pillinger,
M. F. Ribeiro and A. A. Valente, Appl. Catal., A, 2012, 417–
Selectivity% ¼
ꢃ 100
(2)
0
N
4
18, 243–252.
1
N : moles of furfural formed during the reaction.
1
0 L. R. Ferreira, S. Lima, P. Neves, M. M. Antunes, S. M. Rocha,
M. Pillinger, I. Portugal and A. A. Valente, Chem. Eng. J.,
2
1 R. Weingarten, J. Cho, W. C. Conner, Jr and G. W. Huber,
Green Chem., 2010, 12, 1423–1429.
2 J. Zhang, J. Zhuang, L. Lin, S. Liu and Z. Zhang, Biomass
Bioenergy, 2012, 39, 73–77.
3 C. Garcia-Sancho, I. Sadaba, R. Moreno-Tost, J. Merida-
Robles, J. Santamaria-Gonzalez, M. Lopez-Granados and
P. Maireles-Torres, ChemSusChem, 2013, 6, 635–642.
4 A. S. Dias, M. Pillinger and A. A. Valente, Appl. Catal., A, 2005,
285, 126–131.
5 A. S. Dias, S. Lima, M. Pillinger and A. A. Valente, Carbohydr.
Res., 2006, 341, 2946–2953.
6 A. S. Dias, M. Pillinger and A. A. Valente, Microporous
Mesoporous Mater., 2006, 94, 214–225.
7 A. Chareonlimkun, V. Champreda, A. Shotipruk and
N. Laosiripojana, Fuel, 2010, 89, 2873–2880.
8 X. Shi, Y. Wu, P. Li, H. Yi, M. Yang and G. Wang, Carbohydr.
Res., 2011, 346, 480–487.
9 T. Suzuki, T. Yokoi, R. Otomo, J. N. Kondo and T. Tatsumi,
Appl. Catal., A, 2011, 408, 117–124.
0 A. S. Dias, M. Pillinger and A. A. Valente, J. Catal., 2005, 229,
The products were quantitatively determined by Agilent
HPLC 1260 equipped with an ion-exchange column (Bio-Rad
Aminex HPX-87H column 300 mm ꢃ 7.8 mm), coupled to a
013, 215–216, 772–783.
1
1
1
differential refractive index detector (for xylose) and an Agilent
ꢂ1
1
260 UV detector (280 nm, for furfural). A 5 mmol L H SO
2
4
aqueous solution was used as the mobile phase at a ow rate of
ꢂ1
ꢁ
0
.6 mL min and the column temperature was kept at 65 C.
Pure chemicals, i.e. D-xylose and furfural were used for cali-
bration without further treatment.
1
1
1
1
1
1
2
2
2
2
2
4
. Conclusions
Mesostructural zirconium phosphate is demonstrated to be an
effective catalyst for xylose dehydration to furfural in aqueous
media. By using dodecylamine and hexadecylamine as
templates in the hydrothermal synthesis process, the obtained
zirconium phosphate aer calcination is poor crystallized but
similar to zirconium oxide phosphate and possesses high
surface area and rich mesoporosity. For xylose conversion in
aqueous media, a furfural yield up to 52% has been achieved at
ꢁ
1
70 C for 2 h reaction. The adequate Lewis/Brønsted acid sites
which responsible for the isomerization and dehydration in the
catalyst would be responsible for its excellent performance of
catalysis. In addition, the reuse property of the catalyst seems
fairly good by calcination in air at the reuse intervals. The result
is important for the future research on the conversion of sugars
to useful chemicals.
4
14–423.
1 G. H. Jeong, E. G. Kim, S. B. Kim, E. D. Park and S. W. Kim,
Microporous Mesoporous Mater., 2011, 144, 134–139.
2 X. Shi, Y. Wu, H. Yi, G. Rui, P. Li, M. Yang and G. Wang,
Energies, 2011, 4, 669–684.
3 D. Hua, P. P. Li, Y. Wu, Y. Chen, M. Yang, J. Dang, Q. Xie,
J. Liu and X.-y. Sun, J. Ind. Eng. Chem., 2013, 19, 1395–1399.
4 I. Agirrezabal-Telleria, J. Requies, M. B. Guemez and
P. L. Arias, Appl. Catal., B, 2013, DOI: 10.1016/
j.apcatb.2012.11.010.
5 E. Lam, J. H. Chong, E. Majid, Y. Liu, S. Hrapovic,
A. C. W. Leung and J. H. T. Luong, Carbon, 2012, 50, 1033–
1
6 A. S. Dias, S. Lima, P. Brandao, M. Pillinger, J. Rocha and
A. A. Valente, Catal. Lett., 2006, 108, 179–186.
Acknowledgements
The project is supported by the Fundamental Research Funds
for the Central Universities (Grant 1124020512, 1107020524),
NSF of China (21103087, 21273107) and the Natural Science
Foundation of Jiangsu Province (SBK201340143).
2
2
043.
References
1
2
3
4
5
6
7
8
J. N. Chheda, Y. Roman-Leshkov and J. A. Dumesic, Green 27 S. Lima, A. Fernandes, M. M. Antunes, M. Pillinger,
Chem., 2007, 9, 342–350.
F. Ribeiro and A. A. Valente, Catal. Lett., 2010, 135, 41–47.
G. Marcotullio and J. W. De, Green Chem., 2010, 12, 1739– 28 A. Takagaki, M. Ohara, S. Nishimura and K. Ebitani, Chem.
746.
Lett., 2010, 39, 838–840.
C. Moreau, R. Durand, D. Peyron, J. Duhamet and P. Rivalier, 29 I. Agirrezabal-Telleria, A. Larreategui, J. Requies,
1
Ind. Crops Prod., 1998, 7, 95–99.
S. Lima, M. Pillinger and A. A. Valente, Catal. Commun., 2008,
M. B. Gueemez and P. L. Arias, Bioresour. Technol., 2011,
102, 7478–7485.
9
, 2144–2148.
30 E. Lam, E. Majid, A. C. W. Leung, J. H. Chong,
K. A. Mahmoud and J. H. T. Luong, ChemSusChem, 2011,
4, 535–541.
R. O'Neill, M. N. Ahmad, L. Vanoye and F. Aiouache, Ind. Eng.
Chem. Res., 2009, 48, 4300–4306.
J. Lessard, J.-F. Morin, J.-F. Wehrung, D. Magnin and 31 A. S. Dias, S. Lima, D. Carriazo, V. Rives, M. Pillinger and
E. Chornet, Top. Catal., 2010, 53, 1231–1234.
A. A. Valente, J. Catal., 2006, 244, 230–237.
V. Choudhary, A. B. Pinar, S. I. Sandler, D. G. Vlachos and 32 I. Sadaba, S. Lima, A. A. Valente and G. M. Lopez, Carbohydr.
R. F. Lobo, ACS Catal., 2011, 1, 1724–1728. Res., 2011, 346, 2785–2791.
S. B. Kim, S. J. You, Y. T. Kim, S. Lee, H. Lee, K. Park and 33 R. Weingarten, G. A. Tompsett, W. C. Conner, Jr and
E. D. Park, Korean J. Chem. Eng., 2011, 28, 710–716. G. W. Huber, J. Catal., 2011, 279, 174–182.
2
3234 | RSC Adv., 2013, 3, 23228–23235
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
RSC Advances
3
3
4 J. Zhang, L. Lin, Z. Zhang, J. Zhang and S. Liu, J. Biobased 39 F. Benvenuti, C. Carlini, P. Patrono, A. M. Raspolli Galletti,
Mater. Bioenergy, 2011, 5, 539–545.
G. Sbrana, M. A. Massucci and P. Galli, Appl. Catal., A,
2000, 193, 147–153.
5 M. M. Antunes, S. Lima, A. Fernandes, J. Candeias,
M. Pillinger, S. M. Rocha, M. F. Ribeiro and A. A. Valente, 40 X. Guo, X. Guo, W. Tao, L. Chen, L. Peng and W. Ding, Chem.
Catal. Today, 2012, 195, 127–135. Commun., 2011, 47, 10061–10063.
6 J. Zhang, L. Lin and S. Liu, Energy Fuels, 2012, 26, 4560– 41 C. A. Emeis, J. Catal., 1993, 141, 347–354.
3
3
4
567.
42 D. J. MacLachlan and K. R. Morgan, J. Phys. Chem., 1992, 96,
3458–3464.
7 P. A. Russo, S. Lima, V. Rebuttini, M. Pillinger,
M.-G. Willinger, N. Pinna and A. A. Valente, RSC Adv., 43 K.-i. Segawa, Y. Nakajima, S.-i. Nakata, S. Asaoka and
013, 3, 2595–2603. H. Takahashi, J. Catal., 1986, 101, 81–89.
8 V. Choudhary, S. I. Sandler and D. G. Vlachos, ACS Catal., 44 G. Alberti and E. Torracca, J. Inorg. Nucl. Chem., 1968, 30,
012, 2, 2022–2028. 317–318.
2
3
2
This journal is ª The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 23228–23235 | 23235