Glycerol hydrogenolysis to propanediols over supported Pd-Re catalysts

Article abstract of DOI:10.1039/c3ra46134c

Supported Pd-Re bi-component catalysts were prepared by an impregnation method and employed in a glycerol hydrogenolysis reaction. The addition of Re into a Pd catalyst increased the conversion of glycerol and the selectivities to propanediols. The catalysts were characterized by N₂ adsorption-desorption, XRD, CO chemisorption, TPR and NH₃-TPD techniques. The results showed that the addition of Re changed the reduction behaviors of the Pd and Re components, indicating that Pd and Re might have an interaction. Re could increase the ability of the catalysts for the activation of a C-O bond and further promote the activity of Pd in glycerol hydrogenolysis. At the same time, the addition of Re also increased the acidity of the Pd-Re catalysts. Pd-Re supported on acidic oxides showed a tendency of decreasing selectivities to 1,2-propanediol, while Pd-Re catalysts with basic oxides as supports could increase the conversion of glycerol, but maintain the selectivities to 1,2-propanediol.

Full text of DOI:10.1039/c3ra46134c

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Glycerol hydrogenolysis to propanediols over
supported Pd–Re catalysts
Cite this: RSC Adv., 2014, 4, 5503
a
Yuming Li,a Huimin Liu,a Lan Maab and Dehua He
*
Supported Pd–Re bi-component catalysts were prepared by an impregnation method and employed in a
glycerol hydrogenolysis reaction. The addition of Re into a Pd catalyst increased the conversion of
glycerol and the selectivities to propanediols. The catalysts were characterized by N2 adsorption–
desorption, XRD, CO chemisorption, TPR and NH3-TPD techniques. The results showed that the addition
of Re changed the reduction behaviors of the Pd and Re components, indicating that Pd and Re might
have an interaction. Re could increase the ability of the catalysts for the activation of a C–O bond and
further promote the activity of Pd in glycerol hydrogenolysis. At the same time, the addition of Re also
increased the acidity of the Pd–Re catalysts. Pd–Re supported on acidic oxides showed a tendency of
decreasing selectivities to 1,2-propanediol, while Pd–Re catalysts with basic oxides as supports could
increase the conversion of glycerol, but maintain the selectivities to 1,2-propanediol.
Received 5th June 2013
Accepted 8th November 2013
DOI: 10.1039/c3ra46134c
www.rsc.org/advances
metals, Cu and Ni have been widely investigated. Cu and Ni
catalysts can obtain 70–85% of glycerol conversion and 50–99%
of selectivity to 1,2-PD.9,10 Pd, Ru and Pt are active components
Introduction
In recent years, the increased consumption of fossil fuels and
the serious pollution of the environment has encouraged people
to nd new ways to solve the problem of energy supply. There
are forecasts that a 50% growth in energy supply will be needed
by 2025.1,2 Biodiesel is a new kind of diesel fuel which is envi-
ronment friendly, cheap and reproducible. It has gained much
attention from the petrochemical industry for new resource
production and its market is still growing. Biodiesel is produced
by the transesterication of fatty acid esters or vegetable oils,
with methanol or ethanol. In this process, glycerol, that
accounts for nearly 10 wt% of production, is cumulated as a
by-product.3,4 For this reason, glycerol can not be considered as
a high value-added chemical, but an over-produced feedstock.
Traditionally, glycerol is used in cosmetics and to produce
glycerin trinitrate. How to change this abundant glycerol into
more valuable downstream products is a problem for chemists
and manufacturers.5,6
From glycerol hydrogenolysis, 1,2-propanediol (1,2-PD) and
1,3-propanediol (1,3-PD) can be formed and they are high value-
added products. 1,2-PD can be used to produce polyester resins,
cosmetics, antifreeze and so on. 1,3-PD can be used in polyester
bers, medicines and so on.7,8
The main catalysts for this reaction can be divided into two
types. One type are non-noble metal based catalysts and the
other one is noble metal based catalysts. For the non-noble
used as noble metal catalysts. Musolino et al.11 used Pd/Fe2O3 at
200 ꢀC and 2-propanol as a H2 donator and obtained results of
100% of glycerol conversion and 94% of selectivity to 1,2-PD
aer an 18 h reaction. Jiang et al.12 used Ru–Cu bimetallic
catalysts and found 100% of glycerol conversion and 86.4% of
selectivity to 1,2-PD aer 10 h reaction at 230 ꢀC. Hamzah et al.13
prepared Ru/bentonite–TiO2 as a catalyst, over which the
reaction was carried out at 150 ꢀC, 2 MPa of H2 for 7 h; 69.6% of
glycerol conversion and 80.6% of selectivity to 1,2-PD were
obtained. Additionally, Re has also been used in glycerol
hydrogenolysis recently. In our previous study,14 Re was added
into a Ru catalyst and the Ru–Re catalyst was used in glycerol
hydrogenolysis at 160 ꢀC and 8 MPa H2 for 8 h. A yield of 51.7%
of glycerol conversion and 44.8% of selectivity to 1,2-PD were
obtained for Ru–Re/SiO2. These results were much higher than
those of the Ru catalyst itself. In other papers, Pt–Re,15 Rh–Re16
and Ir–Re17 were also investigated. In this literature, the
researchers reported that the addition of Re into other metals
could increase the activities of the active components in glycerol
hydrogenolysis.
In recent research on glycerol hydrogenolysis, Pd has not
been investigated signicantly.11 Compared with Ru and Pt, the
activity of Pd is not as high as the Ru and Pt components. In our
previous study, the addition of Re could increase the activity of
Ru catalysts.14 However, Ru based catalysts also tended to
promote the over-hydrogenolysis or degradation of glycerol,
which resulted in the formation of a large amount of CH4 as a
by-product. Compared with the Ru component, Pd, as an active
component, produces little CH4 in glycerol hydrogenolysis. In
aInnovative Catalysis Program, Key Lab of Organic Optoelectronics & Molecular
Engineering of Ministry of Education, Department of Chemistry, Tsinghua
University, Beijing 100084, China. E-mail: hedeh@mail.tsinghua.edu.cn; Fax: +86-
10-62773346; Tel: +86-10-62773346
bInstitute of Chemical Defence, Beijing 102205, China
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 5503–5512 | 5503

View Article Online
RSC Advances
Paper
this study, we report the catalytic behaviors of Pd catalysts and contents of the Pd, Ag, Fe, and Mn were 5 wt%. The precursors
the feature of the products distribution. Furthermore, the for Fe and Mn were Fe(NO3)3 and Mn(NO3)2. For the preparation
bi-component Pd–Re catalysts were also investigated in glycerol of Pd–Ag/SBA-15, Pd(NO3)2 and AgNO3 were used as the
hydrogenolysis. The addition of Re could increase the catalytic precursors.
activity of Pd catalysts. For Pd–Re/SBA-15, as the content of Re
increased, the conversion of glycerol also increased; at the same
time, the selectivity to 1,2-PD decreased. The relationship
between the catalytic performance and the physico–chemical
properties of the Pd–Re catalysts were also examined.
Hydrogenolysis of glycerol
The hydrogenolysis of glycerol was carried out in a stainless steel
autoclave (100 mL) equipped with a magnetic stirrer. 10 mL of a
40 wt% of glycerol aqueous solution and 0.15 g Pd catalyst were
added to the autoclave. Aer the autoclave was sealed, H2 was
Experiments
Materials
used to replace the air inside it. The autoclave was tested for
leaks, and then heated to 200 ꢀC at 0.1 MPa H2. Finally the
pressure of the H2 in the autoclave was increased to 8 MPa H2
and the reaction started, with stirring. The reaction continued
for 18 h. Aer reaction, the autoclave was cooled below 5 ꢀC and
then decompressed. The gas phase products were collected and
analyzed with a gas chromatograph (Beijing Weisifu-GC 6890)
P123 (EO20PO70EO20, Mw ¼ 5800) was purchased from Sigma-
Aldrich. PdCl2 and Pd(NO3)2 were brought from Shenyang
Nonferrous Metal Research Institute. HReO4 (75–80% solution)
was purchased from Alfa Aesar.
Al2O3 was purchased from the Aluminum Corporation of
China. CNTs were obtained from the Department of Chemical
equipped with a TDX-01 (4 mm ꢂ 2 m) column and a thermal
conductivity detector (TCD). The gas phase products were
Engineering, Tsinghua University. TiO2, MgO, La2O3 and CeO2
quantied with an external standard method. The liquid phase
and the other materials (all samples were analytical reagents)
products were separated from the catalyst with centrifugation
were purchased from the Beijing Chemical Reagent Company.
and analyzed with a gas chromatograph (Lunan-SP 6890)
equipped with a PEG-20M (0.22 mm ꢂ 30 m) column and a
ame ionization detector (FID). An internal standard method
was used to quantify the liquid products.
Catalyst preparation
SBA-15 was prepared by using the method reported in the
literature.18–20 P123, as a template, was dissolved in a HCl
aqueous solution (2 mol Lꢁ1) with strong stirring, at 35 ꢀC. Aer
the solution became transparent, tetraethoxysilane (TEOS) was
added to the solution and stirring continued for another 20 h at
35 ꢀC. 0.017P123 : 40H2O : 4.78HCl : 1TEOS was the molar ratio
of the initial gel. Then, the solution was sealed and kept stati-
cally in an oven at 80 C for 24 h. Aer that, the mixture was
cooled to room temperature and washed with deionized water,
until the ltrate became neutral. Aer being dried at 40 ꢀC, the
The conversion of glycerol and the selectivity to the products
were calculated based on following equations.
Sum of C mol of all products
Conversionð%Þ¼
ꢂ100%
Added glycerol before reaction ðC molÞ
ꢀ
C mol of each product
Selectivity ð%Þ ¼
ꢂ 100%
Sum of C mol of all products
ꢀ
solid was calcined at 500 C for 6 h in the air.
The CNTs were pretreated with a 5 mol Lꢁ1 HNO3 solution at
Catalyst characterization
ꢀ
110 C for 5 h. Aer that, they were ltered and washed with
deionized water, until the ltrate became neutral. The resultant The texture properties of all the samples were characterized by
sample was dried in the air at 110 ꢀC for 12 h. The other N2 adsorption–desorption, with a Micromeritics ASAP 2010C
supports were used as received.
analyzer. The BET and BJH methods were used for the analysis.
The Pd/support catalysts were prepared with the impregna- Each sample was rstly treated at 300 ꢀC in N2 for 1.5 h and
ꢀ
tion method. PdCl2 was dissolved in an HCl aqueous solution degassed at 300 C for 4 h, then measured by N2 adsorption at
rst. Then, the support was added into a certain amount of this ꢁ197 ꢀC.
PdCl2 solution. Aer being impregnated, with stirring for 10 h
Powder X-ray diffraction (XRD), which was used to charac-
in room temperature, the solvent was removed by evaporation. terize the phase structures of the 5Pd–xRe/SBA-15 and others,
The catalyst was dried at 110 ꢀC for 10 h, and reduced with H2 at was carried out on a Rigaku D/max 2500. The instrument was
250 ꢀC for 2 h. The theoretical content of the Pd was 5 wt%. The powered at 40 kV and 200 mA. For wide angle XRD, the scan step
catalysts were denoted as 5Pd/support.
was 10ꢀ minꢁ1 with CuKa radiation (l ¼ 0.1542 nm). For small
5Pd–xRe/support catalysts were also prepared by the method angle XRD, the scan step was changed to 1ꢀ minꢁ1. The particle
of impregnation. A HReO4 aqueous solution was mixed with the sizes of the Pd were calculated by the Scherrer equation.
PdCl2 aqueous solution, rstly. The other procedures were the
same as the preparation of the Pd/support catalysts. (The 5Pd– measurements were carried out in a chemical adsorption
xRe/support means a 5 wt% of Pd and an x wt% of Re.) instrument (Quantachrome CHEM-300). A sample, of 0.1 g, was
The H2 temperature programmed reduction (H2-TPR)
Other bi-component catalysts, such as 5Pd–5Ag/SBA-15, 5Pd– rstly treated in pure Ar gas (99.999%) at 250 ꢀC for 2 h and then
5Fe/SBA-15 and 5Pd–5Mn/SBA-15 catalysts, were all prepared by cooled down to room temperature. The TPR was rstly per-
the method, the same as the Pd–Re/SBA-15 catalysts. The formed by maintaining the sample at 30 ꢀC in 5% of H2–Ar gas
5504 | RSC Adv., 2014, 4, 5503–5512
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
(20 mL minꢁ1) for 20 min. (Step I) Then, the sample was heated
The morphologies of the catalysts were characterized by
from 30 ꢀC to 500 ꢀC at a heating rate of 15 ꢀC minꢁ1. (Step II) A high-resolution transmission electron microscopy (HR-TEM). A
cool trap, prepared by liquid nitrogen and isopropanol, was JEM-2010 JEOL, equipped with an energy dispersive spectrom-
used before the TCD detector, to prevent the effect of water eter (EDS) was used. The catalysts were all reduced at 250 ꢀC
during the TPR measurements.
with H2 for 2 h before the characterization.
In the H2-TPR experiments, the sample was pre-treated in
The X-ray photoelectron spectroscopy (XPS) was recorded on
pure Ar gas before starting the TPR. Therefore, when switching a PHI Quantera SXM of ULVAC-PHI Inc. The binding energy of
the sample from pure Ar gas to the H2–Ar gas to start the TPR, the sample was calibrated with the C 1s peak (284.6 eV). The
some residual Ar gas remained in the sample tube. So, the 5Pd–5Re/SBA-15 catalyst was reduced at 250 ꢀC with H2 for 2 h,
residual Ar gas in the sample tube would cause a response in before the characterization.
the baseline of the TCD detector, which showed as a peak at the
ꢀ
beginning of the TPR prole at 30 C. On the other hand, the
PdO could be reduced, even at 30 C. This also brought a H2
consumption peak at 30 ꢀC at the beginning of the TPR prole,
ꢀ
Results and discussion
The catalytic performance of bi-component Pd–M/SBA-15
aer changing the gas from Ar into H2–Ar. For these reasons, a
blank TPR (with SBA-15 only in the tube) was measured as a
comparative test, to conrm the disturbance of changing the
gas from Ar into H2–Ar.
catalysts and the effect of Re addition on glycerol
hydrogenolysis
The catalytic performances of the different bi-component
catalysts with Pd and another metal components (Ag or Fe, Mn,
Re) in the glycerol hydrogenolysis are shown in Table 1. From
the results, the addition of Ag, Fe or Mn had little effect on the
performance of the Pd/SBA-15 in this reaction. The conversions
of glycerol for these three catalysts were all lower than 2%, but
the selectivities to 1,2-PD were all higher than that of the
Pd/SBA-15 itself. The addition of Re had the most obvious effect
among these catalysts. The conversion of glycerol increased
from 1.5% to 40.7% when Re was added. Moreover, the addition
of Re to Pd/SBA-15 also showed an effect on the selectivity to
1,3-PD. The selectivities to 1,3-PD and 1-propanol (1-PO)
increased from 4.4% and 14.6% to 8.2% and 24.6%, respec-
tively. However, the selectivity to 1,2-PD decreased from 72.2%
to 59.9%. In addition, for these mono-component Pd and
bi-component Pd–M catalysts, CH4 was nearly not detected. It is
suggested that the addition of Re into the Pd catalyst might
bring some interaction between the Pd and Re, and could affect
the glycerol hydrogenolysis signicantly. In order to investigate
the performance of the addition of Re into the Pd catalysts, a
further investigation was undertaken.
The dispersion of Pd was measured with a CO pulsed
chemical adsorption method. The experiment was carried out
in a chemical adsorption instrument Quantachrome CHEM-
300. The catalyst sample (0.1 g) was rstly reduced in pure H2
ꢀ
gas for 2 h at 250 C. Then, the sample was cooled in 99.999%
He gas (the trace amount of oxygen in He gas was removed by a
ꢀ
deoxidizer) to 30 C. The CO chemisorption was performedꢀby
pulsing a certain amount of CO into the sample tube at 30 C.
The dispersion of Pd was obtained by calculating the adsorption
amount of the CO in the Pd sample.
The NH3-TPD proles of the catalysts were measured in a
chemical adsorption instrument (Quantachrome CHEM-300)
with a quadrupole mass-spectrometer detector (DM200M Gas
Analyzer). In aꢀ typical operation, 0.1 g of the sample was pre-
treated at 250 C for 2 h in owing He gas (99.999%) and then
ꢀ
cooled. Aer the temperature reached 100 C, the owing gas
was changed into 1% NH3–He gas and swept through the sample
for 0.5 h, for the adsorption. Aer sweeping again with pure He
gas at 100 ꢀC for 0.5 h to remove the physical adsorption of NH3,
the sample was heated in owing He gas (20 mL minꢁ1) from
100 ꢀC to 500 ꢀC at the heating rate of 15 ꢀC minꢁ1 and the mass
signals were recorded. The NH3 desorption signal was identied
with m/z ¼ 16 (NH2). The desorption amount of NH3 was
calculated by integrating the NH3-TPD peak areas, which were
calibrated by the mass signal of the standard NH3–He gas.
The texture properties of 5Pd–xRe/SBA-15 catalysts and effect
of Re amount on catalytic activity
The pore structures and surface areas of all the 5Pd–xRe/SBA-15
catalysts were analyzed by
a low temperature nitrogen
Table 1 Performance of 5Pd–5M/SBA-15 in glycerol hydrogenolysisa
Select. (%)
Cat.
Conv. (%)
CH4
MeOH
EtOH
EG
1-PO
2-PO
1,2-PD
1,3-PD
5Pd/SBA-15
1.5
0.9
1.7
1.8
40.7
0.0
0.0
0.0
0.0
2.0
0.7
0.7
1.1
0.9
0.1
2.1
1.2
1.2
1.0
1.3
5.3
8.6
5.1
9.7
1.8
14.6
8.2
8.3
2.9
24.6
0.7
1.1
0.4
0.2
2.1
72.2
79.2
83.2
84.6
59.9
4.4
1.1
0.7
0.7
8.2
5Pd–5Ag/SBA-15
5Pd–5Fe/SBA-15
5Pd–5Mn/SBA-15
5Pd–5Re/SBA-15
a
Reaction conditions: 10 mL 40 wt% glycerol solution, 0.15 g 5% Pd catalyst, 200 ꢀC, 8 MPa H2, 18 h, 700 rpm. MeOH: methanol, EtOH: ethanol, EG:
ethylene glycol, PO: propanol.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 5503–5512 | 5505

View Article Online
RSC Advances
Paper
Fig. 1 N2 adsorption–desorption isotherms (A) and pore size distributions (B) of 5Pd–xRe/SBA-15 (a): SBA-15, (b): 5Pd/SBA-15, (c): 5Pd–1Re/
SBA-15, (d): 5Pd–3Re/SBA-15, (e): 5Pd–5Re/SBA-15, (f): 5Pd–10Re/SBA-15, and (g): 5Re/SBA-15.
15 had the largest pore volume (0.71 cm3 gꢁ1) and average pore
Table 2 Texture properties of 5Pd–xRe/SBA-15
size (5.70 nm); 5Pd–xRe/SBA-15 catalysts all had similar pore
SBET
V
DBJH
(nm)
volume (0.5–0.6 cm3 gꢁ1) and average pore size (5.3–5.4 nm).
Fig. 1 shows the pore size distributions of the 5Pd–xRe/SBA-15
catalysts. From the results, it revealed that aer impregnation
and reduction, the 5Pd–xRe/SBA-15 catalysts all had a wider
pore size distribution than SBA-15. This may be because the
addition of Pd and Re to the SBA-15 would decrease the steric
regularity of the SBA-15.
(m2 gꢁ1
)
(cm3 gꢁ1
)
SBA-15
5Pd/SBA-15
856.1
670.5
648.9
662.0
618.2
589.6
704.9
0.71
0.53
0.52
0.52
0.53
0.59
0.59
5.70
5.39
5.37
5.31
5.36
5.36
5.38
5Pd–1Re/SBA-15
5Pd–3Re/SBA-15
5Pd–5Re/SBA-15
5Pd–10Re/SBA-15
5Re/SBA-15
The catalytic performances of the 5Pd–xRe/SBA-15 catalysts
with the addition of different Re content in the glycerol hydro-
genolysis are shown in Table 3. From the results, a mono-
component Pd or Re catalyst showed very low conversions of
glycerol (1.5% and 3.1%, respectively). Aer adding Re into the
5Pd/SBA-15, the conversions of glycerol increased. As the
content of Re increased from 1% to 10%, the conversions of
glycerol increased from 10.2% to 45.1%, and the selectivities to
1,3-PD and 1-PO also increased from 5.2% and 5.8% to about
8% and 20%, respectively. But the selectivity to 1,2-PD
decreased from 77.5% to 60%. The total selectivities to CH4,
methanol and ethanol were lower than 6% for each catalyst. It is
suggested that in the mono-component Pd or bi-component
Pd–Re catalysts, the tendency of degradation of glycerol was very
weak. From the above results, it indicated that the addition of
Re can inuence the Pd catalysts performance in this reaction.
To compare the activities of the Pd–Re and Pd catalysts more
clearly, the TOF (turnover frequency) of the 5Pd/SBA-15 and
5Pd–5Re/SBA-15 were calculated, with the conversion of glycerol
being near 10%. For the 5Pd/SBA-15, the TOF was 1.7 hꢁ1. Aer
adsorption technique. The N2 adsorption–desorption isotherms
are shown in Fig. 1. These results indicated that the SBA-15 and
all 5Pd–xRe/SBA-15 catalysts had hysteresis loops and showed
type IV isotherms, revealing that all the catalysts maintained a
mesoporous structure aer the reduction. In the meantime, it
also showed that aer the reduction, the N2 adsorption–
desorption isotherms were quite similar, except that the
hysteresis loops moved to lower P/P0, compared with the SBA-15
itself. Table 2 shows the BET surface area, the pore volume and
the average pore size of the 5Pd–xRe/SBA-15 samples. From the
results, SBA-15 had the largest surface area (856.1 m2 gꢁ1). Aer
the reduction, the surface areas of all the 5Pd–xRe/SBA-15
catalysts became smaller. The 5Pd–10Re/SBA-15 had the
smallest surface area (589.6 m2 gꢁ1). It can be seen from the
results that the surface area of 5Pd/SBA-15 was smaller than that
of 5Re/SBA-15. For the pore volume and average pore size, SBA-
Table 3 Performance of 5Pd–xRe/SBA-15 in glycerol hydrogenolysisa
Select. (%)
Conv.
Cat.
(%)
CH4
MeOH
EtOH
EG
1-PO
2-PO
1,2-PD
1,3-PD
5Pd/SBA-15
1.5
10.2
19.9
40.7
45.1
3.1
0.0
5.0
2.6
2.0
3.0
0.0
0.7
0.3
0.2
0.1
0.1
0.4
2.1
0.7
0.8
1.3
1.6
1.4
5.3
5.0
3.7
1.8
3.0
1.7
14.6
5.8
9.3
24.6
21.7
38.7
0.7
0.6
0.8
2.1
1.6
1.5
72.2
77.5
74.4
59.9
60.4
49.6
4.4
5.2
8.3
8.2
8.6
6.7
5Pd–1Re/SBA-15
5Pd–3Re/SBA-15
5Pd–5Re/SBA-15
5Pd–10Re/SBA-15
5Re/SBA-15
a
Reaction conditions: 10 mL 40 wt% glycerol solution, 0.15 g 5% Pd catalyst, 200 ꢀC, 8 MPa H2, 18 h, 700 rpm.
5506 | RSC Adv., 2014, 4, 5503–5512
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
the addition of Re, the TOF of 5Pd–5Re/SBA-15 increased to the Pd/SBA-15 catalyst made the dispersion of Pd on 5Pd–xRe/
22.7 hꢁ1. The addition of Re therefore increased the activity of SBA-15 decrease somewhat, but with no great difference. The Pd
Pd catalysts in glycerol hydrogenolysis.
dispersion of 5Pd–1Re/SBA-15 was 14.3%. As the added amount
of Re increased from 1% to 10%, the dispersion of Pd on the
5Pd–xRe/SBA-15 (x ¼ 3–10) caused no evident change.
The structure of 5Pd–xRe/SBA-15 catalysts and the interaction
Fig. 3 and 4 show the TEM images of the 5Pd/SBA-15 and
5Pd–5Re/SBA-15 catalysts. From Fig. 3, the Pd particle size
distribution of the 5Pd/SBA-15 catalyst had a broad range, of
3–21 nm. With EDX analysis, the Pd element was conrmed on
the surface of the 5Pd/SBA-15. The mean particle size of the Pd
in the 5Pd/SBA-15 was 11.1 nm. For the 5Pd–5Re/SBA-15 (Fig. 4),
it was clearly shown that the noble metal particles were
homogeneously distributed on the support. With the EDX
analysis of the metal particles, both the Pd and Re elements
were observed, indicating that the two metals might be
combined together on the surface of the support. For these
particles, the average particle size was 5.1 nm, with the size
distribution between 3–7 nm. At the same time, it could be seen
that the particles grew along the channels of the SBA-15, which
made the particles grow with a rod-like shape.
between Pd and Re
The XRD patterns of 5Pd–xRe/SBA-15 are shown in Fig. 2. For
the small angle XRD, all the 5Pd–xRe/SBA-15 catalysts had a
diffraction peak at 2q ¼ 1ꢀ, which meant that they all main-
tained the well-ordered mesoporous structure of SBA-15. From
the wide angle XRD patterns of 5Pd/SBA-15 and 5Pd–xRe/SBA-
15, the diffraction peaks of Pd(111) at 2q ¼ 40ꢀ, Pd(200) at
2q ¼ 46ꢀ and Pd(220) at 2q ¼ 68ꢀ were observed. These results
are consistent with references.21–23 For the wide angle XRD
pattern of the 5Re/SBA-15, a very weak peak at 2q ¼ 43ꢀ, which
belonged to Re, could be found. However, for 5Pd–xRe/SBA-15,
the peaks belonging to Re could not be observed. This might be
due tꢁo the dispersion of the Re component on the support16 or
ReO4 formation from the interaction of Re and the support.24
Some of the Re might interact with the Pd, and an alloy or
bimetallic cluster might be formed.25,26 This perhaps also led to
the unobserved Re peaks.
The TPR proles of 5Pd–xRe/SBA-15 are shown in Fig. 5. In
Step I of the TPR experiments (the sample was kept in owing
ꢀ
5% H2/Ar at 30 C for 20 min), all the TPR proles revealed a
reduction peak. For the blank test (with SBA-15 only), this peak
(blank peak) in the TPR prole belonged to the TCD signal of
the residual Ar, due to changing the carrier gas (Ar) into 5%
H2–Ar, while for the other samples (5Pd/SBA-15, 5Pd–xRe/SBA-
15 and 5Re/SBA-15), the reduction peaks at 30 C belonged to
the sum of the blank and the H2 consumption. For Step II of the
From the XRD characterization results of 5Pd/SBA-15, 5Pd–
1Re/SBA-15 and 5Pd–3Re/SBA-15, it can be seen that the
intensities of the Pd peaks of those samples were similar. As the
content of Re increased to 5% or 10%, the Pd peak intensity of
5Pd–5Re/SBA-15 or 5Pd–10Re/SBA-15 became a little higher.
The Pd dispersions of 5Pd–xRe/SBA-15 were measured by the
CO chemisorption method, as shown in Table 4. For the 5Pd/
SBA-15, the dispersion of Pd was 17.3%. The addition of Re into
ꢀ
TPR experiments, the proles of 5Pd/SBA-15 and 5Pd–xRe/SBA-
ꢀ
15 showed negative peaks at 100 C, due to the decomposition
Fig. 2 Small angle (A) and wide angle (B) XRD patterns of 5Pd–xRe/SBA-15 (a): SBA-15, (b): 5Pd/SBA-15, (c): 5Pd–1Re/SBA-15, (d): 5Pd–3Re/SBA-
15, (e): 5Pd–5Re/SBA-15, (f): 5Pd–10Re/SBA-15, and (g): 5Re/SBA-15.
Table 4 Dispersions of Pd on 5Pd–xRe/SBA-15 and H2 consumption in TPRa
H2 consumption
of Step I (mmol)
H2 consumption
of Step II (mmol)
Total H2 consumption
(mmol)
Dispersiona (%)
5Pd/SBA-15
17.3
14.3
14.1
13.7
13.3
—
0.025
0.035
0.054
0.076
0.092
0.013
0.016
0.030
0.055
0.126
0.097
0.037
0.051
0.085
0.132
0.217
0.096
5Pd–1Re/SBA-15
5Pd–3Re/SBA-15
5Pd–5Re/SBA-15
5Pd–10Re/SBA-15
5Re/SBA-15
ꢁ0.001
a
Obtained from CO chemisorption.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 5503–5512 | 5507

View Article Online
RSC Advances
Paper
Fig. 3 The TEM image of the Pd–Re/SBA-15 catalyst after reduction at 250 ꢀC.
Fig. 4 The TEM image of the Pd/SBA-15 catalyst after reduction at 250 ꢀC.
of PdH. For the 5Pd/SBA-15, Pd oxide could be reduced at nearly
30 ꢀC (Fig. 5b). During the reduction process of PdO, the
reduced Pd could form a PdH species.27,28 At about 100 C, the
ꢀ
PdH species would decompose and release H2, forming a
negative peak.27,28 Re oxide on the 5Re/SBA-15 could be reduced
at 330 ꢀC (Fig. 5f). It has been reported that Re oxides alone were
hardly reduced at low temperatures, and easily sublimated at
high temperatures. For 5Pd–xRe/SBA-15, the TPR proles were
different from the 5Pd/SBA-15 and 5Re/SBA-15. For the 5Pd–
3Re/SBA-15 and 5Pd–5Re/SBA-15 (Fig. 5d and e), the intensities
of the peaks at 30 ꢀC in the TPR proles were higher than that of
the 5Pd/SBA-15. For the TPR prole of the 5Pd–5Re/SBA-15,
ꢀ
besides the reduction peak at 30 C, a new reduction peak at
ꢀ
130 C appeared. These results suggested that some of the Re
oxide on the 5Pd–xRe/SBA-15 might be reduced at lower
temperatures, compared with the 5Re/SBA-15. As the amount of
Re increased to 10%, the reduction peaks in the TPR prole of
the 5Pd–10Re/SBA-15 were observed at about 30 ꢀC, 130 ꢀC and
220 ꢀC. For all the 5Pd–xRe/SBA-15, no reduction peak was
found when the temperature was higher than 250 ꢀC. This was
quite similar to another report.28 The consumption amounts of
H2 during the TPR were calculated and the results are provided
in Table 4. From the results, it can be seen that as the content of
Re increased, the consumption amounts of H2 in both Step I
and Step II all increased for 5Pd–xRe/SBA-15, compared with
5Pd/SBA-15. Combined with the TPR proles, it is suggested
that the addition of Re could have an interaction with Pd,
making the reduction temperature of Re lower.
Fig. 5 H2-TPR profiles (maintained at 30 ꢀC) of 5Pd–xRe/SBA-15 and
blank tests (a): SBA-15, (b): 5Pd/SBA-15, (c): 5Pd–1Re/SBA-15, (d):
5Pd–3Re/SBA-15, (e): 5Pd–5Re/SBA-15, (f): 5Pd–10Re/SBA-15, and
(g): 5Re/SBA-15. A, Step I: the samples were maintained at 30 ꢀC for 20
min in flowing 5% H2/Ar B, Step II: then the samples were heated in
flowing 5% H2/Ar from 30 ꢀC to 500 ꢀC at 15ꢀC minꢁ1
.
5508 | RSC Adv., 2014, 4, 5503–5512
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
Fig. 6 XPS spectrum of Pd 3d and Re 4f after reduction at 250 ꢀC A: Pd 3d in 5Pd–5Re/SBA-15 B: Re 4f in 5Pd–5Re/SBA-15.
The chemical species of Pd and Re in the Pd–Re catalyst,
aer reduction, were evaluated by XPS. Fig. 6 shows the results
of the XPS of the 5Pd–5Re/SBA-15. All the XPS spectra of Pd and
Re were calibrated in energy by the C 1s signal at 284.6 eV. For
the Pd XPS spectrum (Fig. 6A), the binding energies (BE values)
of the Pd 3d3/2 and 3d5/2 core levels were 340.78 eV and 335.47
eV, respectively. These results all had a shi to higher BE values,
compared with other references.29,30 For the XPS spectrum of Re
(Fig. 6B), values of 46.02 eV and 43.11 eV for the Re4+ and Re6+
oxidation states were observed, respectively. Compared with
other references,31,32 the BE values of Re became a little lower for
the 5Pd–5Re/SBA-15 catalyst. Combined with the XPS analysis of
Pd, it indicated that the there should be an electron transfer,
Fig. 7 NH3-TPD profiles of 5Pd–xRe/SBA-15 (a): 5Pd/SBA-15, (b):
5Pd–1Re/SBA-15, (c): 5Pd–3Re/SBA-15, (d): 5Pd–5Re/SBA-15, (e):
5Pd–10Re/SBA-15, and (f): 5Re/SBA-15.
from Pd to Re in this catalyst, which suggested that Pd and Re
had an interaction.33,34
For the Pd–Re catalysts, from the TPR and XPS results, it could
be supposed that the addition of Re into Pd might cause the
interaction between the Pd and Re components, and could amount was 108.2 mmol gꢁ1, and the acid amount of Re/SBA-15
change the reduction behaviors and the electronic states of the was 245.2 mmol gꢁ1, which was twice as much as the 5Pd/SBA-
Pd and Re oxide. It was possible that part of the Pd and Re 15. From the NH3-TPD prole of the 5Pd–1Re/SBA-15, it clearly
interacted to form a Pd–Re alloy25 or a bi-component cluster.26 showed that an NH3 desorption peak at 160 ꢀC appeared, which
Pallassana et al.35 used non-local density functional theory (DFT) meant the addition of Re into 5Pd/SBA-15 could affect the
calculations and found that in the hydrogenolysis of acetic acid,
acidity of the catalysts. As shown in Fig. 7, as the contents of Re
the C–OH bond may be in favor of activating on the Re surface
increased, the NH3-TPD proles of the 5Pd–xRe/SBA-15 catalysts
rather than on the Pd surface. The reason may be that the barrier became increasingly similar to that of the 5Re/SBA-15. When
of C–OH bond activation on the Re surface was lower than on the the content of Re increased to 3% and 5%, the NH3 desorption
Pd surface. But for the hydrogenation process, Pd was much peaks at both 160 ꢀC and 240 ꢀC appeared. The NH3-TPD prole
better than Re.35,36 This may also occur in glycerol hydrogenolysis of the 5Pd–10Re/SBA-15 showed a strong NH3 desorption peak
with Pd–Re catalysts. Re could activate the C–O bond and Pd at 240 ꢀC, and the strength of the peak was stronger than that of
could help the intermediate react with the H2. The catalytic the other 5Pd–xRe/SBA-15 catalysts. From Table 5, it can also
efficiency of the Pd–Re catalysts may, therefore, become higher.
Fig. 7 shows the NH3-TPD proles of the 5Pd–xRe/SBA-15
catalysts. For the SBA-15, no NH3 desorption could be detected
in the TPD. This indicated that SBA-15 had no acidity. For 5Pd/
SBA-15, the TPD prole showed an NH3 desorption peak at
200 ꢀC. For 5Re/SBA-15, NH3 desorption peaks were detected at
160 ꢀC and 240 ꢀC, which meant it had two kinds of acidic sites
with different acid strengths. The acid strength of 5Re/SBA-15
was stronger than that of the 5Pd/SBA-15. Additionally, the
amount of acid for the 5Re/SBA-15 was also larger than that for
the 5Pd/SBA-15. Table 5 shows the acid amounts of the 5Pd–
xRe/SBA-15 catalysts, which were calculated from the desorp-
tion peaks of the NH3-TPD. From the 5Pd/SBA-15, the acid
been seen that the addition of Re from 1% to 10% would
Table 5 Acid amounts of 5Pd–xRe/SBA-15a
NH3 mmol gꢁ1
a
5Pd/SBA-15
108.2
129.0
135.2
174.6
210.6
245.2
5Pd–1Re/SBA-15
5Pd–3Re/SBA-15
5Pd–5Re/SBA-15
5Pd–10Re/SBA-15
5Re/SBA-15
a
Obtained from NH3-TPD.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 5503–5512 | 5509

View Article Online
RSC Advances
Paper
increase the acid amount of the catalysts from 108.2 mmol gꢁ1 to
210.6 mmol gꢁ1
.
The acidity of the catalysts may inuence the catalytic
activity and the distribution of the products. Fig. 8 shows the
relationship of the acid amounts and the glycerol conversion.
From the results, it clearly shows that increasing the Re content
led to the increase of acidity of the 5Pd–xRe/SBA-15. Also, as the
acidity of the 5Pd–xRe/SBA-15 increased, the conversion of
glycerol increased. For the mono-component Re catalyst (5Re/
SBA-15), the conversion of glycerol was very low, because of the
absence of Pd, which was the active site for hydrogenation. For
the 5Pd–xRe/SBA-15, when the addition of Re was excessive
(5Pd–5Re/SBA-15 and 5Pd–10Re/SBA-15), the acid amount
signicantly increased. However, from the reaction results
(Table 3), it can be seen that the selectivity to 1,2-PD decreased
and the selectivity to 1-PO increased over the 5Pd–5Re/SBA-15
and 5Pd–10Re/SBA-15 catalysts. The selectivity to 1,3-PD
increased somewhat as the acid amount increased. From the
results mentioned above, the change in the acid amount might
cause the change of the conversion of glycerol and the distri-
bution of the products.
From other references regarding the addition of Re,37–39 some
mechanistic paths were also discussed. Chia et al.37 used NH3-
TPD characterization and DFT calculations for Rh–Re catalysts
in the hydrogenolysis of polyols and cyclic ethers. They found
that the bi-component Rh–Re catalysts were acidic. The DFT
results suggested that the hydroxyl group in the Re would make
the intermediates more stable. Tomishige et al.38,39 found that
hydride and proton were easily formed and a C–O bond
neighbouring a CH2OH group could be catalyzed by Ir–Re, or
Rh–Re catalysts. For Pd–Re catalysts in the present study,
Scheme 1 Working mechanism of the Pd-Re/SBA-15 catalyst.
for the dehydration of glycerol into acetol. The fꢁo1rmation rate of
acetol, over 5Pd–5Re/SBA-15 (4.2 mmol gcatꢁ1 h ), was nearly 4
ꢁ1
times that over 5Pd/SBA-15 (1.1 mmol gcat hꢁ1). From these
results, the addition of Re could surely increase the dehydration
process in glycerol hydrogenolysis. Aer dehydration, these
intermediates would undergo a further hydrogenation reaction
and 1,2-PD, 1,3-PD and other produces would formed.
Combined with the reaction and characterization results
mentioned above, we propose a working mechanism of the Pd–
Re/SBA-15 catalyst in the reaction of glycerol hydrogenolysis, as
shown in Scheme 1. That is, Pd and Re might interact, to form
bimetallic clusters. With the help of ReOx, the processes of the
dehydration of glycerol into intermediates and the hydrogena-
tion of the intermediates into propanediols becomes easier,
compared with the mono-component Pd catalysts.
The support effect on catalytic performance of Pd–Re catalysts
in glycerol hydrogenolysis
combined with the references mentioned above, ReOx might Pd and Pd–Re catalysts with different supports (CNTs, Al2O3,
supply the hydroxyl group that could change the acid properties TiO2, MgO, La2O3 and CeO2) were also prepared by an impreg-
of the catalysts. This kind of hydroxyl group and the interaction nation method and employed in glycerol hydrogenolysis. The
of Pd–Re might be favourable for catalyzing the C–O bond of reaction results are shown in Table 6. The activities of the
glycerol and accelerating the dehydration of glycerol to inter- mono-component Pd catalysts with different supports were not
mediates. Acetol or 3-hydroxypropanal might be formed as high. The conversion of glycerol over 5Pd/CNTs was the highest
intermediates of the dehydration of glycerol. In the present among all the mono-component Pd catalysts (12.7%), and the
study, the 5Pd–5Re/SBA-15 and 5Pd/SBA-15 catalysts were used others were all lower than 10%. For 5Pd/Al2O3, the conversion
was the lowest, at only 3.3%. The selectivities to 1,2-PD over the
mono-component Pd catalysts were all higher than 70%. At the
same time, the selectivities to 1-PO over the Pd catalysts with
acidic oxide supports (such as Al2O3) were higher than those
over Pd supported on basic oxides (MgO, La2O3 and CeO2). Aer
the addition of Re to the Pd catalysts, the performance of the
Pd–Re catalysts with different supports was changed substan-
tially compared with the mono-component Pd catalysts. The
addition of Re increased the activities of the Pd catalysts sup-
ported on different supports. These results are in accordance
with Pd–Re/SBA-15.
On the other hand, from Table 6, the performance of the Pd–
Re catalysts supported on acid oxides (or neutral oxides) and
basic oxides still had some differences. The addition of Re into
5Pd/CNTs, for example, the conversion of glycerol over 5Pd–
5Re/CNTs increased from 12.7% to 49.6%. However, the selec-
tivities to 1,2-PD and ethylene glycol (EG) decreased from 71.0%
and 4.8% to 52.5% and 1.7%, respectively. Meanwhile, the
Fig. 8 Relationship of glycerol conversion, acid amount and Re
content of the 5Pd–xRe/SBA-15 catalysts.
5510 | RSC Adv., 2014, 4, 5503–5512
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
Table 6 Performance of Pd–Re catalysts supported on different supports in glycerol hydrogenolysisa
Selec. (%)
Cat.
Conv. (%)
CH4
MeOH
EtOH
EG
1.7
1-PO
2-PO
1,2-PD
1,3-PD
5Pd–5Re/CNTs
5Pd/CNTs
5Pd–5Re/Al2O3
5Pd/Al2O3
5Pd–5Re/TiO2
5Pd/TiO2
5Pd–5Re/MgO
5Pd/MgO
5Pd–5Re/La2O3
5Pd/La2O3
5Pd–5Re/CeO2
5Pd/CeO2
49.6
12.7
23.6
3.3
20.9
4.9
35.1
9.4
52.9
8.0
1.9
0.0
1.5
0.0
3.5
4.5
2.4
4.5
0.4
0.0
1.1
0.0
0.1
1.4
0.1
0.3
0.1
0.5
0.8
0.5
1.2
2.1
1.4
1.4
1.9
0.3
1.0
0.9
1.2
1.0
1.0
1.6
0.6
2.3
0.3
0.6
30.0
12.9
26.6
19.3
14.0
2.4
6.0
1.7
2.7
1.7
3.7
1.8
1.5
0.8
0.6
0.2
0.9
0.2
1.0
0.6
0.3
0.6
52.5
71.0
54.5
63.3
70.1
80.8
81.7
81.3
89.3
83.0
89.2
88.1
8.2
7.9
13.6
9.9
5.7
1.8
3.3
1.5
2.1
0.0
1.1
1.7
4.8
1.2
5.6
4.7
8.8
3.9
8.8
2.7
10.4
4.8
5.4
15.9
6.4
1.7
2.3
a
Reaction conditions: 10 mL 40 wt% glycerol solution, 0.15 g 5% Pd catalyst, 200 ꢀC, 8 MPa H2, 18 h, 700 rpm.
selectivity to 1-PO increased from 12.9% to 30.0%, and the
selectivity to 1,3-PD increased a little (from 7.9% to 8.2%). The
Notes and references
same phenomena were also observed over 5Pd–5Re/SBA-15
(in Table 3), 5Pd–5Re/Al2O3 and 5Pd–5Re/TiO2. In addition, over
5Pd–5Re/MgO, the conversion of glycerol increased from 9.4%
to 35.1% compared with 5Pd/MgO. Unlike the 5Pd–5Re/CNTs,
the selectivity to 1,2-PD over 5Pd–5Re/MgO was 81.7%, which
did not decrease compared with 5Pd/MgO. The distribution of
products changed little over the 5Pd/MgO and 5Pd–5Re/MgO
catalysts. The results of 5Pd–5Re/La2O3 and 5Pd–5Re/CeO2 also
showed similar behaviours, like the 5Pd–5Re/MgO. These
results suggested that the acidic oxides could inuence the
acidity of the Pd–Re catalysts and result in the further hydro-
genolysis of the products, while the basic oxides could limit
further reaction of the products and maintain the selectivity to
1,2-PD.
1 A. Ragauskas, C. Williams and B. Davison, Science, 2006, 311,
484.
2 D. Liang, J. Gao, H. Sun, P. Chen, Z. Hou and X. Zheng, Appl.
Catal., B, 2011, 106, 423.
3 M. Musolino, L. Scarpino, F. Mauriello and R. Pietropaolo,
ChemSusChem, 2011, 4, 1143.
´
4 L. Li, T. Koranyi, B. Sels and P. Pescarmona, Green Chem.,
2012, 14, 1611.
5 D. Alonso, S. Wettstein and J. Dumesic, Chem. Soc. Rev.,
2012, 41, 8075.
6 J. Ma, J. Song, H. Liu, J. Liu, Z. Zhang, T. Jiang, H. Fan and
B. Han, Green Chem., 2012, 14, 1743.
7 T. Miyazawa, S. Koso, K. Kunimori and K. Tomishige, Appl.
Catal., A, 2007, 329, 30.
8 K. Wang, M. Hawley and S. DeAthos, Ind. Eng. Chem. Res.,
2003, 42, 2913.
9 S. Xia, R. Nie, X. Lu, L. Wang, P. Chen and Z. Hou, J. Catal.,
2012, 296, 1.
10 J. Hu, X. Liu, B. Wang, Y. Pei, M. Qiao and K. Fan, Chin.
J. Catal., 2012, 33, 1266.
11 M. Musolino, L. Scarpino, F. Mauriello and R. Pietropaolo,
Green Chem., 2009, 11, 1511.
Conclusions
Pd mono-component catalysts showed activities in glycerol
hydrogenolysis, but the activities were not high. Aer the
addition of Re, the activities of the Pd catalysts increased
signicantly. From the TPR of 5Pd–xRe/SBA-15, it was shown
that the reduction temperature of Re was decreased and that Re
and Pd might interact. The interaction between Pd and Re was
also found from the XPS of 5Pd–5Re/SBA-15. Moreover, the
addition amount of Re also inuenced the catalytic activities
and the distribution of the products. The acid amounts of 5Pd–
xRe/SBA-15 increased with increasing Re loading, and this
could result in the further reaction of propanediols and increase
the formation of 1-propanol.
12 T. Jiang, Y. Zhou, S. Liang, H. Liu and B. Han, Green Chem.,
2009, 11, 1000.
13 N. Hamzah, N. Nordin, A. Nadzri, Y. Nik, M. Kassim and
M. Yarmo, Appl. Catal., A, 2012, 419–420, 133.
14 L. Ma and D. He, Top. Catal., 2009, 52, 834.
15 O. Daniel, A. DeLaRiva, E. Kunkes, A. Datye, J. Dumesic and
R. Davis, ChemCatChem, 2010, 2, 1107.
16 Y. Shinmi, S. Koso, T. Kubota, Y. Nakagawa and
K. Tomishige, Appl. Catal., B, 2010, 94, 318.
17 Y. Nakagawa, X. Ning, Y. Amada and K. Tomishige, Appl.
Catal., A, 2012, 433–434, 128.
Acknowledgements
This work is supported by the National Natural Science Foun-
dation of China (no. 21033004, 20973098) and the Analytical 18 W. Zhou and D. He, Chem. Commun., 2008, 5839.
Foundation of Tsinghua University of China.
19 W. Zhou and D. He, Green Chem., 2009, 11, 1146.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 5503–5512 | 5511

View Article Online
RSC Advances
Paper
20 C. Zhang, W. Zhou and S. Liu, J. Phys. Chem. B, 2005, 109, 30 M. V. Castegnaro, A. S. Kilian, I. M. Baibich, M. Alves and
24319. J. Morais, Langmuir, 2013, 29, 7125.
21 B. Xue, P. Chen, Q. Hong, J. Lin and K. Tan, J. Mater. Chem., 31 F. John, F. William, E. Peter and D. Kenneth, Handbook of
2001, 11, 2378. X-ray Photoelectron Spectroscopy, Minnesota, USA, 1995.
22 L. Shao, B. Zhang, W. Zhang, D. Teschner, F. Schlol and 32 T. Tsoncheva, S. Vankova, O. Bozhkov and D. Mehandjiev,
D. Su, Chem.–Eur. J., 2012, 18, 14962.
Can. J. Chem., 2007, 85, 118.
23 X. Liang, C. Liu and P. Kuai, Green Chem., 2008, 10, 1318.
24 W. Phongsawat, B. Netiworaruksa, K. Suriye, S. Dokjampa,
33 R. Campbell, J. Rodriguez and D. Goodman, Phys. Rev. B:
Condens. Matter, 1992, 46, 7077.
P. Praserthdam and J. Panpranot, J. Nat. Gas Chem., 2012, 34 H. Kima, H. Parka, T. Kima, K. Jeonga, H. Chaea, S. Jeonga,
21, 158. C. Leeb and C. Kima, Catal. Today, 2012, 185, 73.
25 Y. Sato, K. Terada, S. Hasegawa, T. Miyao and S. Naito, Appl. 35 V. Pallassana and M. Neurock, J. Catal., 2002, 209, 289.
Catal., A, 2005, 296, 80.
26 W. Juszczyk and Z. Karpinski, Appl. Catal., A, 2001, 206, 67.
36 Y. Takeda, Y. Nakagawa and K. Tomishige, Catal. Sci.
Technol., 2012, 2, 2221.
27 N. Babu, N. Lingaiah, R. Gopinath, P. Reddy and P. Prasad, J. 37 M. Chia, Y. Torres, D. Hibbitts, Q. Tan, H. Pham, A. Datye,
Phys. Chem. C, 2007, 111, 6447.
M. Neurock, R. Davis and J. Dumesic, J. Am. Chem. Soc.,
´
´
˜
28 R. Perez-Hernandez, A. Avendano, E. Rubio and
2011, 133, 12675.
´
V. Rodrıguez-Lugo, Top. Catal., 2011, 54, 572.
38 S. Koso, Y. Nakagawa and K. Tomishige, J. Catal., 2011, 280,
29 B. Mao, R. Chang, S. Lee, S. Axnanda, E. Crumlin,
221.
M. E. Grass, S. Wang, S. Vajda and Z. Liu, J. Chem. Phys., 39 Y. Amadaa, Y. Shinmi, S. Kosoa, T. Kubot, Y. Nakagawaa and
2013, 138, 214304.
K. Tomishige, Appl. Catal., B, 2011, 105, 117.
5512 | RSC Adv., 2014, 4, 5503–5512
This journal is © The Royal Society of Chemistry 2014