Pt nanoparticles over TiO2-ZrO2 mixed oxide as multifunctional catalysts for an integrated conversion of furfural to 1,4-butanediol

Article abstract of DOI:10.1016/j.apcata.2014.04.012

1,4-butanediol (BDO) is an important commodity chemical for manufacturing many basic chemicals and valuable polymers. Its current manufacturing processes are exclusively based on feedstocks derived from oil and natural gas. The biomass-to-BDO chemical transformation is via furfural, a key platform molecule from glucose and xylose. The integrated conversion involves two sequent reaction steps: selective oxidation of furfural to furanones and hydrogenation of the mixture of furanones to BDO. Platinum nanoparticles supported over TiO ₂-ZrO₂ perform well for both oxidation and hydrogenation steps and the total yield of BDO reaches 85.2%. The chemical composition and crystallinity of the mixed oxide support significantly affect the catalytic performance. The best catalyst is platinum supported over TiO ₂-ZrO₂ mixed oxide (Ti/Zr, 1:1) calcined at 823 K, which also exhibits good recoverability and recyclability in the five-run test.

Full text of DOI:10.1016/j.apcata.2014.04.012

Applied Catalysis A: General 478 (2014) 252–258
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Pt nanoparticles over TiO2–ZrO2 mixed oxide as multifunctional
catalysts for an integrated conversion of furfural to 1,4-butanediol
Fengbo Li∗, Tao Lu, Bingfeng Chen, Zhijun Huang, Guoqing Yuan∗
Beijing National Laboratory of Molecular Science, Key Laboratory of Green Printing, Institute of Chemistry, Chinese Academy of Sciences, Zhongguancun
North 1st Street, Haidian, Beijing 100190, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
1,4-butanediol (BDO) is an important commodity chemical for manufacturing many basic chemicals
and valuable polymers. Its current manufacturing processes are exclusively based on feedstocks derived
from oil and natural gas. The biomass-to-BDO chemical transformation is via furfural, a key platform
molecule from glucose and xylose. The integrated conversion involves two sequent reaction steps: selec-
tive oxidation of furfural to furanones and hydrogenation of the mixture of furanones to BDO. Platinum
nanoparticles supported over TiO2–ZrO2 perform well for both oxidation and hydrogenation steps and
the total yield of BDO reaches 85.2%. The chemical composition and crystallinity of the mixed oxide sup-
port significantly affect the catalytic performance. The best catalyst is platinum supported over TiO2–ZrO2
mixed oxide (Ti/Zr, 1:1) calcined at 823 K, which also exhibits good recoverability and recyclability in the
five-run test.
Received 17 November 2013
Received in revised form 30 March 2014
Accepted 5 April 2014
Available online 13 April 2014
Keywords:
Biomass conversion
Heterogeneous catalysis
Integrated process
Furfural
BDO
© 2014 Elsevier B.V. All rights reserved.
1. Introduction
the scientists from Genomatica. BDO from glucose, xylose, sucrose
and biomass-derived mixed sugar streams were produced by the
1,4-Butanediol (BDO) is an industrially important feedstock
for manufacturing many basic chemicals such as tetrahydrofuran
(THF), ␥-butyrolactone, N-vinylpyrrolidone, polybutene terephtha-
late (PBT), and so on [1]. Because of the increasing worldwide
interest of this C4 feedstock, several manufacturing processes have
been practiced by many great chemical companies. All these pro-
cesses are based on starting molecules derived from fossil feedstock
(petroleum, coal, and natural gas) (Fig. 1a). The Reppe process (1),
as practiced by, e. g., BASF, Du Pont or GAF/Hüls, uses acetylene
as a base for the manufacture of 2-butyne-1,4-diol, a precursor of
BDO. Approximately half of world production capacity is based on
the Reppe process. The hydroformylation of ally alcohol (2) was
harnessed by Lyondell in a 55,000 t/a unit. The butadiene-based
process (3) was built up in Japan by Mitsubishi Chemical and Toyo
Soda. The hydrogenation of Maleic anhydride (4) is another manu-
facturing process which has been used by Japanese companies and
by ICI and UCC/Davy McKee for a long time.
genetically engineered Escherichia coli [12,13]. The biomass-to-BDO
chemical transformation (5) is via a platform molecule, furfural,
which is produced on a large-scale from biomass containing glucose
and xylose [14]. The hydrogenation of furfural is its most important
application [15–17]. Furfural is oxidized to maleic anhydride over
vanadium oxide catalysts at around 573 K with molecular oxygen
and to fumaric acid in aqueous solution by using excessive amounts
of sodium chlorate [18,19]. Maleic anhydride or fumaric acid is fur-
ther hydrogenated to BDO. However, high reaction temperature
leads to inevitable thermopolymerization of furfural and excessive
amounts of sodium chlorate results in high cost in the disposal
of reaction wastes. There are other routes to produce diols from
furfural or its derivatives by direct hydrogenolysis over supported
metal catalysts [20–22]. It is almost inevitable that the products
are multicomponent mixtures and have relatively low selectivity
to alpha–gamma diols.
In this work, a new integrated catalytic process is investigated
to promote the sustainability and efficiency of chemical transfor-
mation from furfural to BDO. Furanones are the intermediates in
the integrated catalytic conversion (Fig. 1b). The reaction mixtures
are incipiently treated with H2O2 and furfural is selectively oxi-
dized to 2(5H)-furanone and its isomer 2(3H)-furanone [23]. The
residual peroxide is destroyed by catalytic self-decomposition with
the only side-product of water. The reaction mixtures are then
hydrogenated to convert furanones into BDO. These reactions are
Biomass is the most abundant renewable resource available, and
it is currently viewed as a feedstock for the sustainable chemical
manufacturing process [2–11]. The first direct biocatalytic route
to BDO from renewable carbohydrate feedstocks was reported by
∗
Corresponding author. Tel.: +86 10 62634920; fax: +86 10 62559373.
E-mail addresses: lifb@iccas.ac.cn (F. Li), yuangq@iccas.ac.cn (G. Yuan).
http://dx.doi.org/10.1016/j.apcata.2014.04.012
0926-860X/© 2014 Elsevier B.V. All rights reserved.

F. Li et al. / Applied Catalysis A: General 478 (2014) 252–258
253
Fig. 1. (a) Industrial manufacturing processes of BDO (the biomass route 5 is not yet industrially applied, just for comparison). (b) Illustration of the proposed conversion of
furfural to BDO.
performed over a multifunctional catalyst (Pt/TiO2–ZrO2) under
mild conditions in a one-pot model (solvent: HCOOH/H2O/MeOH).
Green oxidant, hydrogen peroxide, was harnessed. Formic acid
was the additive for all steps. Formic acid activates hydrogen
peroxide [24] and acidifies the reaction mixture to facilitate the
hydrogenation of furanones [25]. After the conversion, solid cat-
alysts are separated by simple filtration and cleaned for the next
recycling batch. As promoted by the European Technology Platform
on Sustainable Chemistry (ETP SusChem) [26], the sustainability
of chemical process is relevant to process intensification. The
multifunctional catalyst is the core strategy to achieve process
intensification and abate process costs through longer catalyst life,
milder reaction condition, and reduction of separation and envi-
ronmental costs [27].
nanoparticles and introduced into the TiO2–ZrO2 support through
wet impregnation. The metal load of Pt is 1.0 wt%. The sample was
dried at 353 K for 10 h and reduced in a flow of Ar/H2 (v/v, 90/10)
at 623 K for 3 h to give the activated catalyst.
2.2. Catalytic activity test
To a 100-ml autoclave with Teflon inner-coat were added fur-
fural (4.8 g, 50 mmol), the catalyst (0.5 g, Pt load: 1.0 wt%), 30 ml of
HCOOH/H2O/MeOH (v/v/v, 10:10:80). 9.0 ml of hydrogen peroxide
(30% aqueous solution) was added at 298 K during 1.0 h while stir-
ring. The reaction mixture was stirred further for a required time
to remove residual peroxide, which was detected by a negative
peroxide test every 5 min. Then, the system was flushed with nitro-
gen for three times. Hydrogenation was performed at 393 K with
3.5 MPa of hydrogen for 6.0 h. Product/catalyst separation was car-
ried out through simple filtration in the five-run recyclability test.
In consecutive batch tests, continuous filter was applied to separate
the catalyst. The catalyst was recycled into the autoclave with the
feedstock stream.
The mixture is tested for peroxide as follows. Prepare an approx-
imately 1% solution of ferrous ammonium sulfate. Transfer 5 ml of
each of two test tubes and add 0.5 ml of 0.5 M sulfuric acid and
0.5 ml of 0.1 M potassium thiocyanate solution to each tube. Add
5 ml of the methylene chloride solution to one of the test tubes
and shake well. The aqueous phase in the methylene chloride tube
should not develop a brown-red color when examined parallel to
the blank.
2. Experimental
2.1. Preparation of catalytic materials
Porous TiO2–ZrO2 mixed oxide was prepared as follows. To
absolute ethanol (40 ml) were added Ti(OBu)4 (0.01 mol) and
Zr(OBu)4 (0.01 mol). The mixture was further stirred for 30 min
under a nitrogen atmosphere and dodecylamine (0.01 mol) in
absolute ethanol (40 ml) was gradually added. White precipites
were formed by the addition of a mixture of ethanol/water (v/v,
90/10) dropwize. After aging for 12 h, the white solids were col-
lected, washed with ethanol for three times, and dried in air.
Approximately 2.1 g TiO2–ZrO2 mixed oxides were obtained after
calcination at 823 K for 3 h. H2PtCl6 was the precursor of Pt

254
F. Li et al. / Applied Catalysis A: General 478 (2014) 252–258
Fig. 3. (a) XRD patterns of TiO2–ZrO2 mixed oxides with various titanium propor-
tions (calcined at 823 K in air for 3 h). (b) XRD patterns of TiO2–ZrO2 mixed oxides
(Ti/Zr, 1:1) calcined at different temperature.
Fig. 2. (a) The oxidation of furfural to furanones in various reaction systems. Reac-
tion conditions: to a 100-ml autoclave with Teflon inner-coat were added furfural
(4.8 g, 50 mmol), the catalyst (0.5 g, Pt load: 1.0 wt%), 30 ml of reaction solvent. 9.0 ml
of hydrogen peroxide (30% aqueous solution) was added at 298 K during 1.0 h while
stirring. The reaction mixture was stirred further for a required time to remove resid-
ual peroxide, which was detected by a negative peroxide test every five minutes. The
products were measured by GC. (b) Screening experimental results about the trends
of BDO yield change with reaction temperature and hydrogen pressure. The oxida-
tion products were further hydrogenation under selected reaction temperature and
hydrogen pressure.
ESCA-Lab200I-XL instrument manufacturer was used to manip-
ulate the acquired spectra. Nitrogen-adsorption isothermal and
textural properties of the as-synthesized materials were deter-
mined by using liquid nitrogen over a Quantachrome Autosorb
Automated Gas Sorption System (Quantachrome Corporation).
Transmission electron microscopy (TEM) was performed on a JEOL
2010 TEM with an accelerating voltage of 200 kV.
The products were monitored and analyzed by GC and GC–MS.
GC was carried out over GC-2014 (SHIMADZU) with high tempera-
ture capillary column (MXT-1, 30 m, 0.25 mm ID) and FID detector.
GC–MS was carried out over GCT Premier/Waters with capillary
column (DB-5MS/J&W Scientific, 30 m, 0.25 mm ID). The platinum
concentrations in the reaction mixture were determined by induc-
tively coupled plasma atomic emission spectroscopy (ICP–OES)
using Perkin Elmer Optima instrument.
2.4. Calculations
ꢀ
ꢁ
(F − F1)
Furfural conversion
C
=
× 100%
(
)
1
F
where F is the number of moles of incipient furfural in the reac-
tion mixture; F1 is the number of moles of furfural in the reaction
mixture after oxidation step.
Oxidation yield (OY) = C1 × S1
2.3. Materials characterization
where S1 is the selectivity to furanone in the oxidation step (deter-
mined by GC).
Powder X-ray diffraction (XRD) patterns were measured on a
Rigaku Rotaflex diffractometer equipped with a rotating anode
and a Cu-K␣ radiation source (40 kV, 200 mA; ꢀ = 1.54056 A). XPS
ꢀ
ꢁ
F × OY − F
(
)
2
Furanone conversion
C
=
× 100%
(
)
2
˚
F
data were obtained with an ESCALab220i-XL electron.spectrometer
from VG Scientific using 300 W Al-K␣ radiation. The base pres-
sure was approximately 3 × 10−9 mbar. The binding energies were
referenced to the C1s line at 284.8 eV from adventitious car-
bon. The Eclipse V2.1 data analysis software supplied by the VG
where F2 is the number of moles of furanone in the reaction mixture
after hydrogenation step.
Hydrogenation yield (Y) = C2 × S2

F. Li et al. / Applied Catalysis A: General 478 (2014) 252–258
255
Fig. 4. (a) Nitrogen adsorption-desorption isothermal curves of TiO2–ZrO2 (1:1;
823 K) (A), Pt(1.0 wt%)/TiO2–ZrO2 (1:1; 823 K) (B), and TiO2–ZrO2 (1:1; 1023 K) (C).
(b) Pore size distribution curves of TiO2–ZrO2 (1:1; 823 K) (A), Pt(1.0 wt%)/TiO2–ZrO2
(1:1; 823 K) (B), and TiO2–ZrO2 (1:1; 1023 K) (C).
where S2 is the selectivity to BDO in the hydrogenation step (deter-
mined by GC).
Fig. 5. (a) TEM image of the porous structure of TiO2–ZrO2 (1:1; 823 K) mixed oxide.
(b) TEM image of ZrTiO4 nanocrystals embedded in TiO2-ZrO2 (1:1; 823 K) mixed
oxide matrix.
Total yield = (OY × HY) × 100%
3. Results and discussion
furfural to furanones is performed by gradual addition of hydro-
gen peroxide into the reaction mixture during 1.0 h. The residual
peroxide is removed by self-decomposition over supported Pt cat-
alysts. Pt/TiO2–ZrO2 removes the residual peroxide within 25 min
(Table 1). No residual peroxide is detected by a negative per-
oxide test. The reaction mixture is further hydrogenated to give
BDO without any separation or purification. Pt/TiO2–ZrO2 (Pt load:
1.0 wt%) is the most active catalyst and the total yield of BDO
reaches 85.2%.
The oxidation of furfural to furanones is performed in various
reaction systems (Fig. 2a). Hydrogen peroxide cannot selectively
convert furfural to furanones and even with the aid of Pt/TiO2–ZrO2
no furanones are produced. Under the reaction conditions, formic
acid reacts with hydrogen peroxide to form performic acid, which
is the oxidant to produce furanones. The oxidation of furfural with
formic acid/hydrogen peroxide can proceed spontaneously. How-
ever, the conversion rate is very slow and the yield of furanones is
43.2% after nine-hour reaction. Pt/TiO2–ZrO2 exhibits very marked
promoting effect on the oxidation step. The conversion of furfural
3.1. Catalytic conversion of furfural into BDO
Furfural is oxidized with performic acid to give a mixture of
2(5H)-furanone and its isomer (2(3H)-furanone) [28]. The direct
oxidation of furfural with formic acid/hydrogen peroxide can
proceed without the catalyst and the yield of furanones is 43.2%
after nine-hour reaction (Table 1). This conversion of furfural to
furanones is too slow to be synthetically useful. Supported Pt and
Pd show marked catalytic activity for improving the conversion
(Fig. S1). The leaching of Pd in the acidic solution is serious and
gradual deactivation is observed. In the oxidation of furfural with
formic acid/hydrogen peroxide, supported Pt is the most suitable
catalysts. The residual peroxide after the oxidation is hazardous and
dangerous to the following hydrogenation. Supported platinum can
catalyze the self-decomposition of peroxide under mild conditions.
Table 1 lists experimental results about the conversion from fur-
fural to BDO over supported platinum catalysts. The oxidation of

256
F. Li et al. / Applied Catalysis A: General 478 (2014) 252–258
Fig. 6. (a) TEM image of Pt/TiO2–ZrO2 (Ti/Zr, 1:1; 823 K). (b) Supported platinum nanoparticles over the mixed oxide. (c) HRTEM of a single platinum nanoparticle over the
mixed oxide. (d) XPS survey spectrum of Pt/TiO2–ZrO2 (Ti/Zr, 1:1; 823 K) (the inset is the enlarged view of Pt 4f7/2 and Pt 4f5/2 peaks).
to furanones catalyzed by the supported platinum catalyst is per-
formed by gradual addition of hydrogen peroxide into the reaction
mixture to avoid a violent and uncontrolled reaction. The conver-
sion of furfural oxidation over Pt/TiO2–ZrO2 reaches 95.8% through
only one-hour reaction. The selectivity of furanones is 93.6%. The
mixed oxide support shows superiority over single metal oxide.
Titanium is the active component for the oxidation of furfural. The
catalysts involving titanium show higher selectivity of furanones
than that of the catalysts over inert support.
The hydrogenation of furanones has three critical points: (1)
the catalyst, (2) the solvent, and (3) the reaction conditions. The
catalysts in the oxidation step are directly used for the hydro-
genation step without any purification. The yield of hydrogenation
product is 95.0% over Pt/TiO2–ZrO2. Pt/C is a well-established com-
mercial hydrogenation catalyst and extensively suitable for general
substrates. It has stable and reproducible quality. Carbon is a chem-
ical inert support and the complex involvement of support in
the catalytic reaction can be avoided. Based on above-mentioned
features, Pt/C is selected as the benchmarked catalyst. The physi-
cochemical properties of the support show marked effects on the
hydrogenation activity. TiO2–ZrO2 mixed oxide supported plat-
inum is the most active catalyst and its activity is approximately
1.5 times of that of Pt/C (Table 1). The conversion of furanones
over Pt/TiO2–ZrO2 reaches 96.8% and the selectivity of BDO is 98.2%.
Metal oxide supports shows marked effects on the hydrogenation
activity. Screening experiments of hydrogenation activity of the
catalysts over various metal oxides is performed by using separated
furanones as the reactants in a separated operation (Fig. S2). The
experimental results exhibit that the mixed oxide support shows
superiority over single metal oxide due to its enhanced physico-
chemical properties.
Table 1
Detailed experimental results of the conversion of furfural to BDO over supported
platinum catalysts.
Entry
Oxidation
Hydrogenation
Catalyst
Total yield (%)
C1 (%)
S1 (%) C2 (%) S2 (%)
1
2
3
4
5
6
–
Pt/C
Pt/SiO2
Pt/TiO2
Pt/ZrO2
Pt/TiO2–ZrO2
58.6 (9 h)a 73.7
–
–
–
76.1
78.2
89.3
85.7
95.8
62.2
75.9
88.1
72.5
93.6
80.2
84.3
88.2
92.7
96.8
81.4
87.1
91.3
93.0
98.2
30.9
43.6
63.4
53.6
85.2
b
Reaction conditions: to
a 100-ml autoclave with Teflon inner-coat were
added furfural (4.8 g, 50 mmol), the catalyst (0.5 g, Pt load: 1.0 wt%), 30 ml of
HCOOH/H2O/MeOH (v/v/v, 10:10:80). 9.0 ml of hydrogen peroxide (30% aqueous
solution) was added at 298 K during 1.0 h while stirring. The reaction mixture was
stirred further for a required time to remove residual peroxide, which was detected
by a negative peroxide test every 5 min. Then, the system was flushed with nitrogen
for three times. Hydrogenation was performed at 393 K with 3.5 MPa of hydrogen
for 6.0 h.
a
The reaction proceeded without the catalyst and the reaction mixture was
stirred at 298 K for 9 h.
b
The catalyst support is TiO2–ZrO2 mixed oxide (Ti/Zr, 1:1) calcined at 823 K for
3 h in air.

F. Li et al. / Applied Catalysis A: General 478 (2014) 252–258
257
Table 2
Porosity properties of TiO2–ZrO2 mixed oxides.
Sample
SBET (m2/g)
Pore volume
(cm3/g)
Pore size
(nm)
TiO2–ZrO2 (1:1; 823 K)
TiO2–ZrO2 (1:1; 1023 K)
Pt(1.0 wt%)/TiO2–ZrO2 (1:1; 823 K)
243.0
24.0
212.2
0.159
0.073
0.159
1.66
3.85
2.17
The hydrogenation is carried out in
a
mixture of
HCOOH/H2O/MeOH (v/v/v, 10:10:80). Formic acid is used to
acidify the reaction mixture to facilitate the ring-opening hydro-
genation of furanones. The use of mineral acids, such as HCl,
H3PO4, and H2SO4, leads to many side reactions and complicates
the products. Model tests with these typical mineral acids show
very low yield to BDO (Fig. S3). In general, water is considered as
a green solvent. But the distillation of the BDO/water mixture is
the energy-costing separation. The use of methanol as the solvent
would reduce the energy-cost by approximately 50%. Increasing
reaction temperature and hydrogen pressure lead to yield more
BDO (Fig. 2b). To balance the process productivity with the energy
cost, the optimal reaction conditions are selected as 393 K, 3.5 MPa
of hydrogen and reaction time of 6.0 h.
3.2. The correlation between physicochemical properties of the
catalyst and its catalytic performance
TiO2–ZrO2 mixed oxides not only have the advantages of both
TiO2 as active catalyst or support and ZrO2 as acid-base sites, but
also exhibit novel catalytic properties imparted by the formation of
active intermediate states [29,30]. The composite and calciantion
temperature affect the crystallinity of the TiO2–ZrO2 mixed oxide.
An amorphous phase is observed for Ti/Zr (1:1) calcined at 823 K
(Fig. 3). Increasing the portion of titanium or zirconium in the
mixed oxide leads to the emergence of anatase-type TiO2 or ZrO2
crystals. The crystallinity of the mixed oxide (Ti/Zr, 1:1) increases
with the calcination temperature up to 1023 K and the crystalline
ZrTiO4 is noted at 923 K. The porosity properties are measured by N2
adsorption-desorption isotherm (Fig. 4a). The introducing of plat-
inum into the mixed oxide exhibits little influence of the porosity
of the support. The TiO2–ZrO2 mixed oxide calcined at 1023 K has a
sharp decrease in the adsorption volume of the plateau domain,
which indicates the loss of porosity. The hysteresis loop of the
isotherm is markedly broadened. This is attributed to the increase
of pore size. The pore size distribution curves confirm this change.
The amorphous TiO2–ZrO2 mixed oxide (Ti/Zr, 1:1) has a high sur-
face area of 243 m2/g and the average pore size is 1.66 nm (Fig. 4b
and Table 2). The materials are mixtures of ZrTiO4 nanocrystals and
amorphous mixed oxides. The size of ZrTiO4 nanocrystals is in the
range from 5 nm to 10 nm, which are embedded in the mixed oxide
matrix and give novel catalytic properties of TiO2–ZrO2 binary
oxide (Fig. 5).
Pt/TiO2–ZrO2 (Ti/Zr, 1:1; 823 K; Pt: 1.0 wt%) is clearly char-
acterized by TEM (Fig. 6a). Platinum nanoparticles were kept in
a mono-dispersed state and their size was approximately 5 nm
(Fig. 6b). Fig. 6c shows a close view of a single platinum nanopar-
ticle and reveals the typical metallic crystal lattice. All chemical
elements of Pt/TiO2–ZrO2 were detected by XPS survey spectrum
(Fig. 6d) and the binding energy of Pt 4f7/2 is 70.6 eV (the inset of
Fig. 6d), which is indexed to zero-valent platinum species.
TiO2–ZrO2 mixed oxides supporting platinum catalysts are
screened for the integrated catalytic conversion from furfural
to BDO (Fig. 7a). Pt/TiO2–ZrO2 (Ti/Zr, 1:1; 823 K) is the most
efficient catalyst. The optimal mole ratio of titanium to zirconium
is 1:1 for both the oxidation and hydrogenation process. As
revealed by above data-interpretation, the amorphous TiO2–ZrO2
Fig. 7. (a) Screening experiments on platinum catalysts supported over various
TiO2–ZrO2 mixed oxides. (b) The five-run recycling test and corresponding leached
platinum concentration in the reaction solution after each recycling run. All testing
operations were carried out according to the typical procedure described in Section
2.2.
mixed oxide (Ti/Zr, 1:1) has a high surface area of 243 m2/g
and the average pore size is 1.66 nm. High porosity is the key
property of facilitating mass and heat tranfer during catalytic
reaction. The amorphous TiO2–ZrO2 mixed oxide (Ti/Zr, 1:1) is
a mixture of ZrTO4 nanocrystals and amorphous mixed oxides,
which are more efficient catalytic materials than highly crys-
talline mixed oxides. ZrTO4 nanocrystals are active intermediate
states with novel catalytic properties. High porosity and special
micro-structure make TiO2–ZrO2 mixed oxide (Ti/Zr, 1:1) superior
to the mixed oxides with other Ti/Zr ratios. The catalysts over
the mixed oxides calcined above 923 K exhibit marked loss of
catalytic activity for both the oxidation and the hydrogenation
process.
The calcination above 923 K leads to degradation in porosity
properties. The TiO2–ZrO2 mixed oxide calcined at 1023 K has a
surface area of 24 m2/g, which is only 10% of the value of the sam-
ple calcined at 823 K. Another arising problem is the increase in the
crystallinity. Sintering of ZrTO4 nanocrystals results in the loss of
catalytic active species. Loss of surface area and sintering of ZrTO4
nanocrystals are attributed to the relatively low catalytic activity
of the catalyst prepared from the mixed oxide support calcined at
high temperature (above 923 K).

258
F. Li et al. / Applied Catalysis A: General 478 (2014) 252–258
4. Conclusion
In summary, the direct conversion of furfural to BDO was devel-
oped through an integrated catalytic process. The process involves
two sequent reaction steps: selective oxidation of furfural to fura-
nones and hydrogenation of the mixture of furanones to BDO over
a multifunctional Pt/TiO2–ZrO2 catalyst in one-pot model. The best
catalyst is Pt supported over TiO2–ZrO2 mixed oxide (Ti/Zr, 1:1)
calcined at 823 K and the total yield of BDO reaches 85.2%. The
integrated catalytic process exhibits good recoverability and recy-
clability in the five-run test.
Acknowledgments
This work was financially supported by the National Natural Sci-
ences Foundation of China (nos. 21101161, 21174148, 21304101)
and YMB New Energy Co.
Appendix A. Supplementary data
Supplementary material related to this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.apcata.
2014.04.012.
References
[1] H.J. Arpe, Industrial Organic Chemistry, fifth ed., Wiley-VCH, Weinheim, 2010,
pp. 98–105.
[2] G.W. Huber, S. Iborra, A. Corma, Chem. Rev. 106 (2006) 4044–4098.
[3] A. Corma, S. Iborra, A. Velty, Chem. Rev. 107 (2007) 2411–2502.
[4] P. Mäki-Arvelaa, B. Holmboma, T. Salmia, D.Y. Murzin, Catal. Rev. Sci. Eng. 49
(2007) 197–340.
[5] J.N. Chheda, G.W. Huber, J.A. Dumesic, Angew. Chem. Int. Ed. 46 (2007)
7164–7183.
[6] D.A. Simonettia, J.A. Dumesic, Catal. Rev. Sci. Eng. 51 (2009) 441–484.
[7] J. Zakzeski, P.C.A. Bruijnincx, A.L. Jongerius, B.M. Weckhuysen, Chem. Rev. 110
(2010) 3552–3599.
[8] P. Gallezot, Chem. Soc. Rev. 41 (2012) 1538–1558.
[9] M.E. Zakrzewska, E.B. Lukasik, R.B. Lukasik, Chem. Rev. 111 (2011)
397–417.
[10] A.M. Ruppert, K. Weinberg, R. Palkovits, Angew. Chem. Int. Ed. 51 (2012)
2564–2602.
ˇ
[11] D. Kubicˇka, I. Kubicˇková, J. Cejka, Catal. Rev. Sci. Eng. 55 (2013) 1–78.
[12] H. Yim, R. Haselbeck, W. Niu, C. Pujol-Baxley, A. Burgard, J. Boldt, J. Khandurina,
J.D. Trawick, R.E. Osterhout, R. Stephen, J. Estadilla, S. Teisan, H.B. Schreyer,
S. Andrae, T.H. Yang, S.Y. Lee, M.J. Burk, S.V. Dien, Nat. Chem. Biol. 7 (2011)
445–452.
[13] A.P. Zeng, W. Sabra, Curr. Opin. Biotech. 22 (2011) 749–757.
[14] X. Tong, Y. Ma, Y. Li, Appl. Catal., A: Gen. 38 (2010) 1–13.
[15] X. Chen, H. Li, H. Luo, M. Qiao, Appl. Catal., A: Gen. 233 (2002) 13–20.
[16] S. Sitthisa, T. Pham, T. Prasomsri, T. Sooknoi, R.G. Mallinson, D.E. Resasco, J.
Catal. 280 (2011) 17–27.
Fig. 8. (a) XPS survey spectra of the fresh catalyst (A), the catalyst after the twice
recycling run (B), and the catalysts after the fifth recycling run (C). (b) XRD patterns
of the fresh catalyst (A), the catalyst after the twice recycling run (B), and the catalyst
after the fifth recycling run (C).
3.3. The recyclability of the catalyst in the furfural conversion
The recoverability and recyclability of Pt/TiO2–ZrO2 (Ti/Zr, 1:1;
823 K) is further investigated through a five-run recycling test
(Fig. 7b). A whole testing run includes oxidation and hydrogenation.
The catalyst is recycled to the next batch after simple purification.
There is approximately 8.4% loss of the total yield of BDO after the
fifth recycling test for the as-synthesized catalyst. In the practi-
cal operation, such a gradual decrease in the catalytic activity is
inevitable. The manual operation cannot guarantee 100% recycle of
the catalyst. The loss of catalyst amount is inevitable during transfer
operation among containers in cleaning, separation and purifica-
tion. The catalytic performance is dependent on reaction time and
process. Leached Pt species are detected in the reaction solution
after each run. There is a relatively high concentration of leached
Pt during the first run, which may be attributed to loosely anchored
Pt nanoparticles. During the next testing runs, this value fluctuates
around 10 ppb. Such amount of Pt leaching is negligible and shows
little influence upon the catalytic activity. XPS and XRD measure-
ments were performed over the catalyst samples after the twice
run and the fifth run (Fig. 8). The chemical and physical properties
are kept constant during the recycling test.
[17] L.R. Baker, G. Kennedy, M. Van Spronsen, A. Hervier, X. Cai, S. Chen, L. Wang,
G.A. Somorjai, J. Am. Chem. Soc. 134 (2012) 14208–14216.
[18] N.A. Milas, Org. Synth. Coll. 2 (1943) 302–303.
[19] T. Tachibana, T. Masuda, M. Funabashi, M. Kunioka, Biomacromolecules 11
(2010) 2760–2765.
[20] M. Chia, Y.J. Pagan-Torres, D. Hibbitts, Q. Tan, H.N. Pham, A.K. Datye, M. Neurock,
R.J. Davis, J.A. Dumesic, J. Am. Chem. Soc. 133 (2011) 12675–12689.
[21] S. Koso, I. Furikado, A. Shimao, T. Miyazawa, K. Kunimoria, K. Tomishige, Chem.
Commun. 203 (2009) 5–2037.
[22] Y. Amada1, H. Watanabe, Y. Hirai, Y. Kajikawa, Y. Nakagawa, K. Tomishige1,
ChemSusChem 5 (2012) 1991–1999.
[23] C.W. Jones, Applications of Hydrogen Peroxide and Derivatives, The Royal Soci-
ety of Chemistry, Cambridge, 1999, pp. 14–32.
[24] C.W. Jones, Applications of Hydrogen Peroxide and Derivatives, The Royal Soci-
ety of Chemistry, Cambridge, 1999, pp. 61–72.
[25] L.E. Schniepp, H.H. Geller, R.W. von Korff, J. Am. Chem. Soc. 69 (1947)
672–674.
[26] The
European
Technology
Platform
for
Sustainable
Chemistry.
ꢀhttp://www.suschem.org/about-suschem/vision-and-mission-sustainable-
chemistry.aspxꢁ (accessed in 2009).
[27] G. Centi, S. Perathoner, in: F. Cavani, G. Centi, S. Perathoner, F. Trifiró (Eds.),
Sustainable Industrial Processes, Wiley-VCH, Weinheim, 2009, pp. 1–72.
[28] J.H. Näsman, Org. Synth. Coll. 8 (1993) 396–403.
[29] I. Salem, Catal. Rev. Sci. Eng. 45 (2003) 205–296.
[30] B.M. Reddy, A. Khan, Catal. Rev. Sci. Eng. 47 (2005) 257–296.