Aqueous-phase deoxygenation of glycerol to 1,3-propanediol over Pt/WO 3/ZrO2 catalysts in a fixed-bed reactor

Article abstract of DOI:10.1039/c0gc00005a

Deoxygenation of glycerol in aqueous medium catalyzed by Pt/WO ₃/ZrO₂ at relatively low temperatures (110-140°C) under hydrogen pressure range from 2 to 5 MPa in a fixed-bed continuous-flow reactor gives 1,3-propanediol (1,3-PDO) and n-propanol (n-PrOH) as the predominant products, indicating high selectivity for deoxygenation of the secondary hydroxyl group over the primary hydroxyl groups of the glycerol. The optimum catalyst was prepared by calcination of WO₃/ZrO₂ at 700°C and loading of 3.0 wt% Pt with W content of 10 wt%. The effect of reaction temperature, hydrogen pressure and initial water content were evaluated to find the optimum reaction conditions. The glycerol conversion and the yield of 1,3-PDO greatly depended on these factors. At 130°C, 4 MPa and 70.2% conversion, the yield of 1,3-PDO was up to 32.0% (1,3-PDO/1,2-PDO = 17.7). The proposed mechanism for glycerol deoxygenation in aqueous medium over Pt/WO ₃/ZrO₂ is an ionic pathway involving proton and hydride ion transfer steps.

Full text of DOI:10.1039/c0gc00005a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/greenchem | Green Chemistry
Aqueous-phase deoxygenation of glycerol to 1,3-propanediol over
Pt/WO3/ZrO2 catalysts in a fixed-bed reactor
Li-Zhen Qin, Min-Jie Song and Chang-Lin Chen*
Received 16th April 2010, Accepted 25th May 2010
First published as an Advance Article on the web 15th July 2010
DOI: 10.1039/c0gc00005a
Deoxygenation of glycerol in aqueous medium catalyzed by Pt/WO3/ZrO2 at relatively low
temperatures (110–140 ◦C) under hydrogen pressure range from 2 to 5 MPa in a fixed-bed
continuous-flow reactor gives 1,3-propanediol (1,3-PDO) and n-propanol (n-PrOH) as the
predominant products, indicating high selectivity for deoxygenation of the secondary hydroxyl
group over the primary hydroxyl groups of the glycerol. The optimum catalyst was prepared by
calcination of WO3/ZrO2 at 700 ◦C and loading of 3.0 wt% Pt with W content of 10 wt%. The
effect of reaction temperature, hydrogen pressure and initial water content were evaluated to find
the optimum reaction conditions. The glycerol conversion and the yield of 1,3-PDO greatly
depended on these factors. At 130 ◦C, 4 MPa and 70.2% conversion, the yield of 1,3-PDO was up
to 32.0% (1,3-PDO/1,2-PDO = 17.7). The proposed mechanism for glycerol deoxygenation in
aqueous medium over Pt/WO3/ZrO2 is an ionic pathway involving proton and hydride ion
transfer steps.
are low.18 Hawley and co-workers19 have carried out a selective
1. Introduction
dehydroxylation of glycerol to 1,3-PDO in three steps, but
despite a high selectivity for 1,3-PDO (>70%) the reaction
sequence was rather complicated and required the stoichiometric
use of tosyl chloride.
Production of chemicals by catalytic transformation of bio-
sustainable resources is a global challenge.1 Recent development
of biodiesel production by transesterification of plant oils and
animal fat has made a large amount of glycerol available as a
renewable feedstock for the synthesis of a wide range of value-
added chemicals.2–5 Significant research effort has been focused
on the conversion of glycerol by catalytic processes such as
reforming, oxidation, dehydration, esterification, etherification,
deoxygenation and so on.1–6 Synthesis of 1,2-propanediol (1,2-
PDO) and 1,3-propanediol (1,3-PDO) from glycerol via catalytic
deoxygenation has attracted significant interest.7–17 1,2-PDO is
an important commodity chemical, which finds use as antifreeze,
aircraft de-icer, and as a component of coating and lubricant.
1,3-PDO is more valuable than 1,2-PDO, in particular, as
a raw material for the production of polyesters. Significant
advances have been made into the deoxygenation of glycerol
to 1,2-PDO by using heterogeneous catalysts.8,16 In contrast, the
conversion of glycerol to 1,3-PDO is still a challenge. Relatively
well developed is the fermentation process using bacterial
strains in which glycerol is converted to 1,3-PDO. However,
the fermentation process has some drawbacks. One of the main
drawbacks is its low theoretical yield, owing to the byproducts,
such as ethanol, acetate, lactate, butyrate and CO2 formed in
this process. Another main drawback is that the process is
substrate inhibited. The bacteria used in the fermentation are
generally not able to stand a glycerol concentration above 17%.
As a result, both the product concentration and the productivity
There have been several reports on homogeneous and het-
erogeneous catalysts for the catalytic deoxygenation of glycerol
to 1,3-PDO. Che20 developed the catalytic deoxygenation of
glycerol in 1-methyl-2-pyrolidinone under 32 MPa syngas at
200 ◦C and in the presence of a homogeneous rhodium com-
plexes catalyst Rh(CO)2(acac) and H2WO4. Through the process,
1,3-PDO and 1,2-PDO were produced with 21% and 23% yield,
respectively. Bullock and co-workers21,22 described the deoxy-
genation of glycerol catalyzed by a homogeneous ruthenium
◦
complex in sulfolane under 5.2 MPa and 110 C. The yield of
1,3-PDO was less than 5%. Chaminand et al.7 reported a method
for the production of 1,3-PDO using a catalyst system containing
Rh/C and H2WO4 with discontinuous operation, achieving a 4%
yield at 180 ◦C and 8.0 MPa. The highest 1,3-PDO/1,2-PDO
ratio was 2.0 obtained in sulfolane. Kurosaka et al.12 reported
the deoxygenation of glycerol in 1,3-dimethyl-2-imidazolidinone
at 170 ◦C and 8.0 MPa with discontinuous operation, in which
the yields of 1,3-PDO was 24% over Pt/WO3/ZrO2 catalysts.
The 1,3-PDO/1,2-PDO ratio was about 2.0. Tomishige and co-
workers11,23,24 reported the deoxygenation of glycerol in water at
120 ◦C and 8.0 MPa with discontinuous operation, in which the
1,3-PDO/1,2-PDO ratio was less than 0.7 over Rh/SiO2 and
modified Rh/SiO2 catalysts.
Huang et al.25 reported conversion of glycerol into 1,3-
propanediol over Cu–H4SiW12O40/SiO2 in vapor phase without
employing organic solvent. Through the process, 1,3-PDO and
1,2-PDO were produced with 32.1% and 22.2% selectivity,
respectively. For high reaction temperature (210 ◦C), a lot of
State Key Laboratory of Materials-Oriented Chemical Engineering,
College of Chemistry and Chemical Engineering, Nanjing University of
Technology, Nanjing, 210009, China. E-mail: clchen@njut.edu.cn
1466 | Green Chem., 2010, 12, 1466–1472
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
unknown products were produced. Since the evaporation of
glycerol is energy-consuming, the liquid phase catalytic process
will be more suitable.
was obtained by hydrolysis of zirconium nitrate (99%, SCRC)
solution with the addition of aqueous ammonium hydroxide
(25–28%, SCRC) solution up to pH 9. The precipitate ob-
tained was filtered, washed with de-ionized water and dried at
110 ◦C overnight. The dried Zr(OH)4 was impregnated with
the pre-calculated amount aqueous solution of ammonium
metatungstate (99.999%, Acros Organics), the water was evap-
orated, and the samples were dried overnight at 110 ◦C. To
investigate the effect of calcination temperatures, the samples
were calcined at different temperatures in static air for 3 h. These
WO3/ZrO2 samples are denoted WZc(T), where T refers to the
calcination temperature/◦C and c the tungsten content in weight
percent (wt%). For example, WZ10(700) means the sample
containing 10 wt% tungsten and calcined at 700 ◦C. Pt/WZc(T)
catalysts were prepared by impregnation using an aqueous
solution of H2PtCl6 (40% Pt, Acros Organics). The loaded Pt
samples were dried at 110 ◦C for 12 h, and then calcined at
450 ◦C in static air for 3 h.
Hydrogen and CO chemisorption uptakes of catalyst were
measured using a BELCAT-A instrument equipped with a
thermal conductivity detector (TCD). The study was conducted
in a quartz tubular reactor. Prior to the H2 or CO chemisorption
the samples were pretreated in air at 450 ◦C for 3 h, reduced
at 250 ◦C in H2 for 1 h, and switched to argon for 0.5 h to
remove the adsorbed H2 at 250 ◦C and then cooling down
to the desired temperatures. The H2 pulse chemisorption was
measured at 100 ◦C under atmospheric pressure. The CO
pulse chemisorption was measured at 30 ◦C under atmospheric
pressure and the platinum dispersion was calculated by assuming
an adsorption of one CO molecule per surface platinum atom.
Temperature-programmed desorption (TPD) of pre-adsorbed
NH3 was carried out on a BELCAT-A instrument and the
evolved ammonia concentration was monitored by a Mass
Spectrometer and TCD detector. Prior to NH3 adsorption,
about 0.2 g sample was dehydrated under flowing air at 450 ◦C
for 1 h, and reduced by H2 at 250 ◦C, then cooled to 100 ◦C in
He. The NH3 adsorption was carried out at this temperature for
0.5 h, then He was allowed to flow over the sample for 1.0 h at
100 ◦C. The sample◦was then heated from 100 ◦C to 800 ◦C with
a heating rate of 10 C min-1. The area under the TCD curve was
integrated to determine the relative total acidity of the catalysts.
The deoxygenation of glycerol was carried out in a fixed-
bed down-flow reactor made of a stainless steel tube (length,
340 mm; internal diameter, 10.0 mm). The catalyst (2.0 mL,
20–40 mesh) was placed between two layers of quartz sands for
each test. A thermocouple was situated in the catalytic bed for
measurement of the temperature. Prior to reaction, the catalyst
was activated in flowing H2 (50 ml min-1) from ambient to 250 ◦C
at the rate of 0.5 ◦C min-1 and maintaining that temperature for
1 h at atmospheric pressure. After reduction, the system was
cooled to the desired reaction temperature. Reaction conditions
were changed to investigate the dependence of the conditions.
Details of the reaction conditions are described for each result.
Hydrogen from a gas cylinder was depressurized to 4.0 MPa
and its flow rate was set by a mass flow controller giving a
constant flow of 10 ml min-1. Aqueous solutions of glycerol
were continuously pumped into the fixed-bed reactor with a
high-pressure pump. Hydrogen and glycerol solution flows were
mixed at the inlet of the reactor. Gas and liquid were flowed in
However, there are several drawbacks in the previous attempts.
The greatest drawback is the reaction medium. Regardless
of catalyst type, 1,2-PDO is the predominant product and
the selectivity to 1,3-PDO is lower in aqueous phase.7–17,23,24
Chaminand et al.7 found that the deoxygenation of glycerol
to 1,3-PDO is favorable in polar aprotic solvent, thus most of
the previous studies have been performed in high boiling-point
organic solvent, e.g., sulfolane ,11,20,21 1-methyl-2-pyrolidinone ,20
1,3-dimethyl-2-imidazolidinone.12 It is well understood that this
drawback will greatly reduce the environmental and economical
viability. Another drawback is the slurry reactor, which presents
difficulties in terms of catalyst separation from the reaction
mixture.
The deoxygenation of glycerol to 1,3-PDO is a substitution
reaction, in which a H atom substitutes for the secondary
hydroxyl group in glycerol. As shown in eqn (1), one approach
to selective deoxygenation of the secondary hydroxyl group of
glycerol is sequential treatmentwith H+ and H- sources, releasing
water and formally resulting in the conversion of OH to H.20,21
Achieving catalytic deoxygenation of glycerol would require that
proton and hydride transfer steps occur, followed by catalyst
regeneration through heterolytic activation of H2 to H+ and H-.
A catalyst capable of heterolytic cleavage of H2 could potentially
enable use of H2 as the source of both H+ and H-. Bullock and
co-workers20,21 described the deoxygenation of glycerol catalyzed
by a homogeneous ruthenium complex in sulfolane through an
ionic pathway involving proton and hydride transfer steps. But
the yield of 1,3-PDO was less than 5%.
(1)
The Pt/WO3/ZrO2 and its modified version are efficient cata-
lysts for skeletal isomerization of alkanes.26–29 On Pt/WO3/ZrO2
catalyst, hydrogen molecules can be activated into protons and
hydride ions via heterolytic cleavage at lower temperature.30,31
Kurosaka et al.12 reported the deoxygenation of glycerol in
1,3-dimethyl-2-imidazolidinone at 170 ◦C and 8.0 MPa over
Pt/WO3/ZrO2 catalysts with discontinuous operation.
Considering water is a by-product of the glycerol deoxygena-
tion, safe, inexpensive water is an ideal solvent. In order to
avoid the evaporation of water and aqueous phase reforming of
glycerol, we investigated Pt/WO3/ZrO2 as catalyst for glycerol
deoxygenation in the aqueous phase in a fixed-bed continuous-
◦
flow reactor at low temperatures of 110–140 C and hydrogen
pressures of 2–5 MPa. Details of Pt/WO3/ZrO2 catalysts and the
optimal reaction parameters were also investigated to improve
the selectivity of the reaction towards 1,3-PDO. We will also
discuss the formation routes of deoxygenation and consecutive
deoxygenation products based on the catalytic performance
when the products in the glycerol reaction were used as reactants.
2. Experimental
A series of WO3/ZrO2 samples with different tungsten contents
were prepared by methods previously reported.27,28 Zr(OH)4
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 1466–1472 | 1467
©

View Article Online
co-current down-flow mode through the reactor. At the reactor
outlet, the liquid and gas flows were cooled and recovered in
the gas-liquid separator. The gas flow was depressurized to
atmospheric pressure by a back-pressure regulator.
In addition, we also carried out the reaction test of 1,3-
PDO and 1,2-PDO in order to discuss the reaction routes of
glycerol deoxygenation. The concentration of these reactants in
the aqueous solution was 60 mass%.
The liquid products were analyzed using a SP-6800A GC
(Shandong Ruihong chromatogram analysis Co., Ltd, China)
equipped with a flame ionization detector (FID) and a capillary
column (Alltech ECTM-1 capillary column, 30 m ¥ 0.53 mm ¥ 1.2
mm). The quantitative analysis was based on internal standard
method using n-butanol and 1,4-butanediol as double internal
standard. The gas-phase products in the tail gas were analyzed
with another GC (Shimadzu GC-2014) equipped with a thermal
conductivity detector (TCD). Molar yields were given in all cases
at total conversion of glycerol. Product selectivity was calculated
as the molecular percentage of all products obtained.
increase in the glycerol conversion from 14.2% to 84.5%. The
yield of 1,3-PDO increased with platinum content and reached
a maximum at 3.0 wt% Pt. Further increase of platinum content
caused a decrease in the yield of 1,3-PDO. The increase of
platinum content from 1.0 to 4.0 wt% led to a uniform increase
of the yield of n-PrOH from 7.1% to 56.2%, and yield of i-
PrOH from 1.4% to 5.3%. Along with platinum content from
1.0 to 4.0 wt%, the 1,3-PDO/1,2-PDO ratio increased distinctly
from 8.5 to 37.2. These results indicate that the excess platinum
content further promotes excessive consecutive deoxygenation
of propanediols to n-PrOH. Hence, to get a good conversion of
glycerol with high selectivity to 1,3-PDO an optimal amount of
platinum should be loaded on WZ10(700).
As shown in Table 1, the hydrogen uptakes on Pt/WZ10(700)
catalysts increased with platinum content. However, the degree
of increase reduced after the platinum reached 3 wt%. This
indicated platinum aggregation on the surface of WZ10(700)
with too high Pt loading. The average size of the platinum
particles on the surface of WZ10(700) support was deter-
mined using the measured amounts of chemisorbed CO and
is presented in Table 1, where the particle size increased with
increasing platinum content. The concentration of protons and
hydride ions formed on Pt/WO3/ZrO2 from hydrogen molecules
increased with hydrogen uptakes.29,30 The protons and hydride
ions formed from molecular hydrogen act as active sites for
glycerol deoxygenation and glycerol conversion increased with
the concentration of protons and hydride ions. However, the
excess amount of protons and hydride ions may enhance the
consecutive deoxygenation of propanediols to propanols, which
suppress the selectivity towards 1,3-PDO and 1,2-PDO.
The effect of tungsten content on the catalytic activity and the
product distributions over 2.0Pt/WZ(700) at 130 ◦C under 4.0
MPa hydrogen pressure is shown in Table 2. Both the activity and
the product distributions are dependent on tungsten content.
Glycerol conversion starting from 27.0% over 2.0Pt/WZ5(700)
to 41.6% with 2.0Pt/WZ10(700). Further increasing tungsten
content to 15 wt% leads to a decrease in glycerol conversion
to 29.9%. The highest glycerol conversion of 41.6% and 1,3-
PDO yield of 18.3% are achieved on 2.0Pt/WZ10(700). Further
increasing tungsten content to 15 wt% leads to the decrease
of 1,3-PDO yield to 10.7% and glycerol conversion to 29.9%.
The maximum value was found at 10 wt% tungsten which is
slightly higher than the theoretical monolayer of WZ supports.
It was reported that the acidity present on WZ sample increased
with loading of tungsten oxide up to monolayer coverage.32 The
surface concentration of acid sites, i.e. acidity (NH3/nm2) of WZ
3. Results and discussion
The deoxygenation of glycerol in aqueous phase was carried
out over Pt/WO3/ZrO2 catalysts in a fixed-bed reactor under
constant pressure and temperature in the following domain, 2.0–
5.0 MPa and 110–140 ◦C, respectively. Under the given reaction
conditions, only the deoxygenation products of 1,3-PDO, 1,2-
PDO, n-propanol (n-PrOH), and i-propanol (i-PrOH) were
formed. Degradation products, e.g. 1,2-ethanediol, ethanol,
methanol, ethane, methane, CO and CO2 were not detected
in the obtained liquid phase or gas phase. This observation
indicates that aqueous phase reforming of glycerol is avoided
and the catalyst is not able to cleave C–C bond at the reaction
temperature used. 1,3-PDO and n-PrOH, which were formed
from deoxygenation of internal OH of glycerol, were the major
products. 1,2-PDO and i-PrOH, which were formed from deoxy-
genation of terminal OH of glycerol, were the minor products.
The yield of the products varied with the catalyst composition,
catalysts pretreatment, and the reaction parameters.
The performance of the Pt/WZ10(700) catalysts containing
between 1.0 and 4.0 wt% platinum for glycerol deoxygenation
has been compared at 130 ◦C under 4.0 MPa hydrogen pressure.
The results shown in Table 1 represent the activity and selectivity
after 24 h on-stream; at this time there was a pseudo-steady state,
and the catalyst activity was relatively stable. As the platinum
content increased from 1.0 to 4.0 wt% there was a uniform
Table 1 Effect of Pt content on glycerol deoxygenation over Pt/WZ10(700) catalysta
Yield (mol%) c
H2 uptakes (mmol g-1.cat) Particle size/nmb Conversion (%) i-PrOH n-PrOH 1,2-PDO 1,3-PDOd 1,3-PDO/1,2-PDO
Catalysts
1.0Pt/WZ10 30.6
2.0Pt/WZ10 59.6
3.0Pt/WZ10 68.5
4.0Pt/WZ10 73.7
1.8
—
2.4
3.3
14.2
41.6
70.2
84.5
1.4
3.4
5.3
5.3
7.1
0.6
1.5
1.8
0.6
5.1 (35.9) 8.5
18.3 (44.0) 12.2
32.0 (45.6) 17.7
22.3 (26.4) 37.2
18.4
31.0
56.2
a Reaction conditions: 130 ◦C; 4.0 MPa; 2.0 ml of catalyst was used; The aqueous glycerol (60% wt/wt) was fed at the speed of 0.5 ml h-1; Hydrogen
flow rate was 10 ml min-1. Data obtained after 24 h on-stream. b Data obtained from CO chemisorption. c Molar yields are given in all cases at total
conversion of glycerol. d Selectivity in brackets.
1468 | Green Chem., 2010, 12, 1466–1472
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Table 2 Influence of tungsten content on glycerol deoxygenation over 2.0Pt/WZ(700) catalysts a
Yield (mol%) b
Catalysts
Conversion (%)
i-PrOH
n-PrOH
1,2-PDO
1,3-PDO c
1,3-PDO/1,2-PDO
2.0Pt/WZ5(700)
2.0Pt/WZ10(700)
2.0Pt/WZ15(700)
27.0
41.6
29.9
2.1
3.4
2.9
10.2
18.4
14.4
2.5
1.5
1.9
12.0 (44.4)
18.3 (44.0)
10.7 (35.8)
4.8
12.2
5.6
a Reaction conditions: see note in Table 1. b Molar yields are given in all cases at total conversion of glycerol. c Selectivity in brackets.
Table 3 Surface concentration of acid sites of WZ catalysts with
protonic acid site. A second spiltover hydrogen atom reacts with
the electron trapped by the Lewis acid site to form a hydride
ion (H-), which stabilized on the Lewis acid site. Increasing
the WZ calcination temperature is favorable as it increases the
concentration of proton and hydride ions on Pt/WO3/ZrO2.
The influence of reaction temperature on the deoxygenation
of glycerol in aqueous-phase was investigated over 2.0 wt%
Pt/WZ10(700) catalyst, using conditions described in Table 5.
Increasing the reaction temperature has a positive effect on the
conversion, as shown in Table 5. As the temperature increases
from 110 C to 140 C, the glycerol conversion increases from
20.9% to 60.5%, with 1,3-PDO selectivity decreasing from
50.9% to 35.7%. The lower reaction temperature is favorable
in terms of the selectivity to 1,3-PDO. When the temperature
was above 130 ◦C, the amount of n-PrOH, which is the product
of consecutive deoxygenation of PDOs, greatly increased. The
ratio of 1,3-PDO to 1,2-PDO exhibited little dependence on
the reaction temperature. So the optimal reaction temperature
to perform aqueous-phase glycerol deoxygenation over 2.0 wt%
Pt/WZ10(700) catalyst is 130 ◦C.
Reactions were carried out at 2.0, 3.0, 4.0 and 5.0 MPa
pressure of H2 at a constant temperature of 130 ◦C over the
2.0Pt/WZ10(700) catalyst to determine the effect of hydrogen
pressure on the reaction. Table 6 provides the summary of
the conversions of 60% glycerol solution at 130 ◦C under
different hydrogen pressures. As shown in Table 6, both glycerol
conversion and the yield of 1,3-PDO increased when hydrogen
pressure increased from 2.0 MPa to 5.0 MPa. The yields of 1,2-
PDO and i-PrOH were exhibited little dependence on hydrogen
pressure. The 1,3-PDO/1,2-PDO ratio increased with hydrogen
pressure. The increase of hydrogen pressure provided a positive
effect on the glycerol deoxygenation, which might be due to the
increase in the concentration of protons and hydride ions formed
from hydrogen on Pt/WO3/ZrO2.
The effect of initial water content on the performance of
glycerol deoxygenation was examined over 2.0Pt/WZ10(700)
catalyst at 130 ◦C and 4 MPa pressure of H2. The results are
shown in Table 7. The conversion of glycerol increased with
initial water content. It is important that the increase of initial
water content from 20% to 80% led to a increase of the selectivity
to 1,3-PDO from 39.9% to 49.8%. Interestingly, the selectivity
to 1,2-PDO exhibited a reverse trend to that of 1,3-PDO. The
1,3-PDO/1,2-PDO ratio increased distinctly along with initial
water content from 20% to 80% and the maximum was as
high as 16.5. It was obvious that water is a good solvent for
the glycerol deoxygenation over 2.0Pt/WZ10(700). Considering
water is a by-product of the glycerol deoxygenation, this property
is promising. On the basis of the above observation, we can
different tungsten content and WZ10 calcined at different temperatures
Acidity
Acidity
Catalyst
(NH3/nm2) a
Catalyst
(NH3/nm2) a
WZ10(500)
WZ10(600)
WZ10(700)
2.85
3.09
3.15
WZ10(800)
WZ5(700)
WZ15(700)
3.35
2.78
2.28
a Total number of NH3 molecules desorbed per nm2 of catalyst surface
area from 120 to 750 ◦C.
◦
◦
samples with different tungsten content, was measured using
TPD of NH3 and is presented in Table 3. TPD of NH3 shows that
the surface concentration of acid sites increases with tungsten
content and reaches its maximum for a WZ10 sample calcined at
750 ◦C. The surface concentration of acid sites decreases when
the WO3 loading exceeds monolayer coverage.
Table 4 shows the effect of calcination temperature of WZ10
support (10 wt% W loading) on the catalytic performance of
2.0 wt% Pt/WZ10 catalyst in deoxygenation of glycerol at 130 ◦C
under 4.0 MPa hydrogen pressure. The collected data show
clearly that the increasing WZ10 calcination temperature led
to an increasing glycerol conversion starting from 12.7% over
a 2.0 wt% Pt/WZ10(500) sample to 47.5% with a 2.0 wt%
Pt/WZ10(800). The WZ10 calcinations temperature also has
an influence on the product distribution. The yields of 1,3-
PDO increases with WZ10 calcination temperature, reaching a
plateau at about 18%, when the calcination temperature reached
700 ◦C. The yields of 1,2-PDO and i-PrOH depended weakly
on WZ10 calcination temperature. Increasing WZ10 calcination
temperature led to an increase in the yield of n-PrOH, which
formed in consecutive deoxygenation of propanediols, starting
from 4.0% over 2.0 wt% Pt/WZ10(500) to 24.9% with 2.0 wt%
Pt/WZ10(800). The surface concentration of acid sites, i.e. acid-
ity (NH3/nm2) of WZ10 with different calcination temperatures,
was measured by TPD of ammonia and is presented in Table 3.
Increasing the WZ10 calcination temperature led to an increase
in the surface concentration of acid sites starting from 2.85
NH3/nm2 on a WZ10(500) sample to 3.35 NH3/nm2 on a
WZ10(800). The heterolytic activation of H2 into protons and
hydride ions formed on Pt/WO3/ZrO2 involved in the Lewis
acid site.30,31 Hydrogen molecules were dissociatively adsorbed
on Pt sites to form hydrogen atoms which undergo spillover onto
the surface of WO3/ZrO2 support. The spiltover hydrogen atoms
diffuse on the WO3/ZrO2 surface to reach Lewis acid sites where
each hydrogen atom donates an electron to the Lewis acid and
becomes a proton (H+). The proton is stabilized on the O atom
(or platinum atom) adjacent to the Lewis acid site and acts as a
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 1466–1472 | 1469
©

View Article Online
Table 4 Influence of WZ10 calcination temperatures on the performance of 2.0Pt/WZ10 catalysts
Yield (mol%) b
Catalysts
Conversion (%)
i-PrOH
n-PrOH
1,2-PDO
1,3-PDO c
1,3-PDO/1,2-PDO
2.0Pt/WZ10(500)
2.0Pt/WZ10(600)
2.0Pt/WZ10(700)
2.0Pt/WZ10(800)
12.7
35.7
41.6
47.5
1.0
3.1
3.4
3.3
4.0
1.7
1.6
1.5
0.95
6.0 (47.3)
15.7 (44.0)
18.3 (44.0)
18.3 (38.6)
3.5
9.8
12.2
19.3
15.3
18.4
24.9
a Reaction conditions: see note in Table 1. b Molar yields are given in all cases at total conversion of glycerol. c Selectivity in brackets.
Table 5 Influence of reaction temperature on glycerol deoxygenation over the 2.0Pt/WZ10(700) catalyst a
Yield (mol%) b
Reaction T/◦C
Conversion (%)
i-PrOH
n-PrOH
1,2-PDO
1,3-PDO c
1,3-PDO/1,2-PDO
110
120
130
140
20.9
33.0
41.6
60.5
1.8
3.0
3.4
5.4
7.4
1.0
1.45
1.5
1.9
10.6 (50.9)
14.9 (45.2)
18.3 (44.0)
21.6 (35.7)
10.6
10.2
12.2
11.6
13.7
18.4
31.6
a Reaction conditions: see note in Table 1 except the reaction temperature. b Molar yields are given in all cases at total conversion of glycerol.
c Selectivity in brackets.
Table 6 Influence of hydrogen pressure on glycerol deoxygenation over the 2.0Pt/WZ10(700) catalyst a
Yield (mol%) b
Pressure/MPa
Conversion (%)
i-PrOH
n-PrOH
1,2-PDO
1,3-PDO c
1,3-PDO/1,2-PDO
2.0
3.0
4.0
5.0
15.6
33.8
41.6
45.4
2.3
3.3
3.4
3.7
6.4
1.8
1.7
1.5
1.8
5.1 (32.5)
12.6 (37.3)
18.3 (44.0)
20.8 (45.8)
2.8
7.5
12.2
11.5
16.2
18.4
19.1
a Reaction conditions: see note in Table 1 except the hydrogen pressure. b Molar yields are given in all cases at total conversion of glycerol. c Selectivity
in brackets.
suggest that water as a protic solvent facilitated the transfer
of the formed proton that acts as a Brønsted acid in the
liquid reaction. The freely transferable protons can overcome
the steric hindrance to the internal OH in glycerol such that
the internal OH has almost accessibility for activation as the
terminal OH. Nimlos et al.33 reported that the proton affinity
for the internal hydroxyl group in glycerol was 195.4 kcal mol-1,
while that for the terminal hydroxyl group was 194.8 kcal mol-1,
which means that the internal hydroxyl group in glycerol is easier
to activate when it has the same exposure to protons. This would
explain the higher selectivity for 1,3-PDO when the reaction was
performed in dilute aqueous solution.
As
a
representative example,
a
time-on-stream be-
havior of aqueous-phase deoxygenation of glycerol over
2.0Pt/WZ10(700) catalyst was examined for a period of 50 h
at 130 ◦C and 4 MPa, and the results are shown in Fig. 1. The
catalyst shows a large increase in glycerol conversion and in the
yields of 1,3-PDO and n-PrOH with increase in the reaction
time up to 8 h. Upon further increasing the reaction time,
the yields of 1,3-PDO, 1,2-PDO and i-PrOH are stable, even
though there is a slight decrease in glycerol conversion and in
the yield of n-PrOH. This suggests that glycerol is converted
to n-PrOH by a different pathway. The gradual decrease in
glycerol conversion might be due to some loss of activity in
the consecutive deoxygenation of propanediols to n-PrOH.
Fig. 1 Time-on-stream behavior of aqueous-phase deoxygenation of
glycerol over 2.0Pt/WZ10(700) catalyst. Reaction conditions: 130 ◦C;
4.0 MPa; 4 mL (5 g) of catalyst was used; The aqueous glycerol (60%
wt/wt) was fed at the speed of 1.0 mL h-1; Hydrogen flow rate was
40 mL min-1.
In order to explain the reaction scheme and the product
distribution of glycerol deoxygenation, the deoxygenation of
1470 | Green Chem., 2010, 12, 1466–1472
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Table 7 Influence of initial water content on the performance of glycerol deoxygenation over the 2.0Pt/WZ10(700) catalysta
Yield (mol%) b
Initial water content (wt%)
Conversion (%)
i-PrOH
n-PrOH
1,2-PDOc
1,3-PDOc
1,3-PDO/1,2-PDO
80
60
40
20
59.9
43.6
41.6
20.8
4.1
3.5
3.4
2.0
24.1
17.4
18.4
9.4
1.8 (3.1)
1.5 (3.5)
1.5 (3.6)
1.0 (4.8)
29.8 (49.8)
21.2 (48.6)
18.3 (44.0)
8.3 (39.9)
16.5
14.1
12.2
8.3
a Reaction conditions are the same as in Table 1 except initial water content. b Molar yields are given in all cases at total conversion of glycerol.
c Selectivity in brackets.
Table 8 Results of deoxygenation of 1,2-PDO, 1,3-PDO and glycerol over 2.0Pt/WZ10(700) catalyst a
Selectivity (mol%)
Reactant
1,2-PDO
Reaction T/◦C
Conversion (%)
i-PrOH
n-PrOH
1,2-PDO
1,3-PDO
120
130
140
120
130
140
120
130
140
60.8
78.3
91.4
12.0
22.0
36.8
33.0
41.6
60.5
9.6
9.2
10.2
0.0
0.0
0.0
9.0
8.2
9.0
90.3
90.7
89.7
99.9
99.9
99.9
41.4
44.2
52.2
—
—
—
0.0
0.0
0.0
4.4
3.6
3.1
0.0
0.0
0.0
—
—
—
45.2
44.0
35.7
1,3-PDO
Glycerol
a Reaction conditions: 130 ◦C; 4.0 MPa; 2.0 ml of catalyst was used; The aqueous reactant (60% wt/wt) was fed at the speed of 0.5 ml h-1; Hydrogen
flow rate was 10 mL min-1. TOS = 24 h.
Scheme 1 Proposed reaction mechanism of glycerol deoxygenation over Pt/WZ
1,2-PDO and 1,3-PDO over 2.0Pt/WZ10(700) were also tested
under similar conditions as used for glycerol. The results are
listed in Table 8. Regarding the conversion of reactants, 1,3-PDO
was much lower than 1,2-PDO and glycerol. The reactivity of
1,2-PDO can be much higher than that of 1,3-PDO, and this
can explain that the selectivity of 1,3-PDO was much higher
than 1,2-PDO in deoxygenation of glycerol. In addition, the
formation of n-PrOH and i-PrOH was observed on Pt/WZ10
in the glycerol reaction. The selectivity to n-PrOH was much
higher than that to i-PrOH in the deoxygenation of 1,2-PDO,
the ratio of n-PrOH to i-PrOH is 7.0, which can explain that
the selectivity of n-PrOH was much higher than i-PrOH in
deoxygenation of glycerol. The n-PrOH/i-PrOH ratio in the
glycerol deoxygenaton was similar to that in the 1,2-PDO
deoxygenation, and this suggests that n-PrOH and i-PrOH are
mainly formed via 1,2-PDO in the glycerol deoxygenation. It
shows high regioselectivity for removing the internal hydroxyl
group over Pt/WZ10 catalyst. In the deoxygenation of glycerol,
the main product was 1,3-PDO and n-PrOH, and this means
that the formation rate of 1,3-PDO is high and its conversion
rate is relatively low, which is supported by the result in
Table 8.
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 1466–1472 | 1471
©

View Article Online
The proposed reaction mechanism of glycerol deoxygenation
over Pt/WO3/ZrO2 is shown in Scheme 1. The mechanisms
involved heterolytic cleavage of hydrogen molecules on catalyst
to form protons and hydride ions, and then sequentially
delivering H+ followed by H- to glycerol and diols.
6 G. W. Huber, J. W. Shabaker and J. A. Dumesic, Science, 2003, 300,
2075–2077.
7 J. Chaminand, L. Djakovitch, P. Gallezot, P. Marion, C. Pinel and
C. Rosier, Green Chem., 2004, 6, 359–361.
8 M. A. Dasari, P. P. Kiatsimkul, W. R. Sutterlin and G. J. Suppes,
Appl. Catal., A, 2005, 281, 225–231.
9 T. Miyazawa, Y. Kusunoki, K. Kunimori and K. Tomishige, J. Catal.,
2006, 240, 213–321.
4. Conclusions
10 S. Wang and H. Liu, Catal. Lett., 2007, 117, 62–67.
11 I. Furikado, T. Miyazawa, S. Koso, A. Shimao, K. Kunimori and K.
Tomishige, Green Chem., 2007, 9, 582–588.
This study has demonstrated that the aqueous-phase deoxy-
genation of glycerol to 1,3-PDO is catalyzed by Pt/WO3/ZrO2
catalyst in a fixed-bed continuous flow reactor at relatively
lower temperatures (110–140 ◦C). Aqueous-phase reforming
and degradation of glycerol over Pt/WO3/ZrO2 catalyst can be
avoided at the lower reaction temperature. The predominant
products are 1,3-PDO and n-propanol. The high selectivity,
giving 1,3-PDO and n-propanol rather than 1,2-PDO and i-
propanol, is due to the protonic acid-catalyzed dehydration
preferring dehydration at the secondary over the primary OH
in the glycerol and diol. Both the conversion of glycerol
and the yield of the products significantly varied with the
catalyst composition, catalyst pretreatment, and the reaction
parameters. A 45.6% selectivity of 1,3-PDO, with 1,3-PDO/1,2-
PDO ratio of 17.7 was achieved at 70.2% conversion, 130 ◦C and
4 MPa over 3.0 wt% Pt/WO3/ZrO2 catalyst. The deoxygenation
of glycerol is proposed to occur by an ionic mechanism, involving
proton transfer and hydride transfer steps.
12 T. Kurosaka, H. Maruyama, I. Naribayashi and Y. Sasaki, Catal.
Commun., 2008, 9, 1360–1363.
13 L. Huang, Y. Zhu, H. Zheng, Y. Li and Z. Zeng, J. Chem. Technol.
Biotechnol., 2008, 83, 1670–1675.
14 J. Feng, H. Fu, J. Wang, R. Li, H. Chen and X. Li, Catal. Commun.,
2008, 9, 1458–1464.
15 L. Ma, D. He and Z. Li, Catal. Commun., 2008, 9, 2489–2495.
16 Z. Huang, F. Cui, H. Kang, J. Chen, X. Zhang and C. Xia, Chem.
Mater., 2008, 20, 5090–5099.
17 A. Alhanash, E. F. Kozhevnikova and I. V. Kozhevnikov, Catal. Lett.,
2008, 120, 307–311.
18 A. Zeng and H. Biebl, Adv. Biochem. Eng. Biotechnol., 2002, 74,
237–257.
19 K. Wang, M. C. Hawley and S. J. DeAthos, Ind. Eng. Chem. Res.,
2003, 42, 2913–2923.
20 T. M. Che, U. S. Patent, 4642394, 1987.
21 M. Schlaf, P. Ghosh, P. J. Fagan, E. Hauptman and R. M. Bullock,
Angew. Chem., Int. Ed., 2001, 40, 3887–3890.
22 M. Schlaf, P. Ghosh, P. J. Fagan, E. Hauptman and R. M. Bullock,
Adv. Synth. Catal., 2009, 351, 789–800.
23 A. Shimao, S. Koso, N. Ueda, Y. Shinmi, I. Furikado and K.
Tomishige, Chem. Lett., 2009, 38, 540–541.
24 Y. Shinmi, S. Koso, T. Kubota, Y. Nakagawa and K. Tomishige, Appl.
Catal., B, 2010, 94, 318–326.
Acknowledgements
25 L. Huang, Y. Zhu, H. Zheng, G. Ding and Y. Li, Catal. Lett., 2009,
131, 312–320.
Financial support by the Research Fund for the Doctoral
Program of Higher Education of China (No. 20070291007) and
the Natural Science Foundation of Jiangsu Province, China (No.
BK2008375) are gratefully acknowledged.
26 M. Hino and K. Arata, J. Chem. Soc., Chem. Commun., 1988, 1259–
1260.
27 X. Chen, C. Chen, N. Xu, S. Han and C. Mou, Catal. Lett., 2003,
85, 177–182.
28 X. Chen, C. Chen, N. Xu and C. Mou, Catal. Today, 2004, 93–95,
129–134.
29 X. Chen, Y. Du, C. Chen, N. Xu and C. Mou, Catal. Lett., 2006, 111,
187–193.
References
30 S. Triwahyono, T. Yamada and H. Hattori, Appl. Catal., A, 2003,
1 A. Corma, S. Iborra and A. Velty, Chem. Rev., 2007, 107, 2411–2502.
2 M. Pagliaro, R. Ciriminna, H. Kimura, M. Rossi and C. D. Pina,
Angew. Chem., Int. Ed., 2007, 46, 4434–4440.
3 Y. Zheng, X. Chen and Y. Shen, Chem. Rev., 2008, 108, 5253–5277.
4 A. Behr, J. Eilting, K. Irawadi, J. Leschinski and F. Lindner, Green
Chem., 2008, 10, 13–30.
242, 101–109.
31 S. Triwahyono, A. A. Jalil and H. Hattori, J. Nat. Gas Chem., 2007,
16, 252–257.
32 C. D. Baertsch, S. L. Soled and E. Iglesia, J. Phys. Chem. B, 2001,
105, 1320–1330.
33 M. R. Nimlos, S. J. Blanksby, X. Qian, M. E. Himmel and D. K.
Johnson, J. Phys. Chem. A, 2006, 110, 6145–6156.
5 C. Zhou, J. N. Beltramini, Y. Fan and G. Q. Lu, Chem. Soc. Rev.,
2008, 37, 527–549.
1472 | Green Chem., 2010, 12, 1466–1472
This journal is
The Royal Society of Chemistry 2010
©