Mesoporous PdxPt alloys: High-performance catalysts for methane combustion

Article abstract of DOI:10.1016/j.mcat.2017.09.002

Mesoporous cubic Pd_xPt (x = 0.43–8.52) alloys with surface areas of 26–32 m²/g were synthesized using the KIT-6-templating method. Physicochemical properties of the materials were characterized by means of various techniques, and their catalytic activities were evaluated for methane combustion. It is found that the Pd and Pt were uniformly distributed in the Pd_xPt alloys. The addition of Pt to Pd exerted a significant effect on the redox property of Pd. The Pd_xPt alloys possessed a higher methane activation ability than the monometallic Pd. The oxidized Pd–Pt (i.e., PdO–PtO₂) were more active than the metallic Pd⁰–Pt⁰. The Pd_2.41Pt sample performed the best for methane combustion (T_10%, T_50%, and T_90% were 272, 303, and 322 °C at SV = 100,000 mL/(g h); TOF_Pd, TOF_Pt, TOF_{Pd + Pt}, and specific reaction rate at 280 °C were 0.85 × 10^?3 s^?1, 1.98 × 10^?3 s^?1, 0.59 × 10^?3 s^?1, and 4.46 μmol/(g_cat s), respectively). The deactivation of the Pd_2.41Pt sample induced by 2.5–5.0 vol% CO₂ or H₂O addition was reversible, but its deactivation due to 100 ppm SO₂ introduction was irreversible. It is concluded that the excellent catalytic performance of the Pd_2.41Pt sample was associated with its good mesoporous structure, Pd–Pt alloy and PdO–PtO₂ coexistence, and good methane and oxygen activation ability.

Full text of DOI:10.1016/j.mcat.2017.09.002

Molecular Catalysis 442 (2017) 191–201
Contents lists available at ScienceDirect
Molecular Catalysis
journal homepage: www.elsevier.com/locate/mcat
Editor’s choice paper
Mesoporous Pd Pt alloys: High-performance catalysts for methane
x
combustion
∗
∗
Xingtian Zhao, Yuxi Liu , Jiguang Deng , Peng Xu, Jun Yang, Kunfeng Zhang, Zhuo Han,
Hongxing Dai
∗
Beijing Key Laboratory for Green Catalysis and Separation, Key Laboratory of Beijing on Regional Air Pollution Control, Laboratory of Catalysis Chemistry
and Nanoscience, Department of Chemistry and Chemical Engineering, College of Environmental and Energy Engineering, Beijing University of Technology,
Beijing 100124, China
a r t i c l e i n f o
a b s t r a c t
2
Article history:
Received 23 June 2017
Received in revised form 18 August 2017
Accepted 2 September 2017
Mesoporous cubic Pd Pt (x = 0.43–8.52) alloys with surface areas of 26–32 m /g were synthesized using
x
the KIT-6-templating method. Physicochemical properties of the materials were characterized by means
of various techniques, and their catalytic activities were evaluated for methane combustion. It is found
that the Pd and Pt were uniformly distributed in the PdxPt alloys. The addition of Pt to Pd exerted a
significant effect on the redox property of Pd. The PdxPt alloys possessed a higher methane activation
ability than the monometallic Pd. The oxidized Pd–Pt (i.e., PdO–PtO2) were more active than the metallic
Keywords:
Mesoporous Pd–Pt alloy
PdO–PtO2 active site
Redox property
Methane combustion
Methane activation ability
0
0
Pd –Pt . The Pd2.41Pt sample performed the best for methane combustion (T10%, T50%, and T90% were 272,
◦ ◦
03, and 322 C at SV = 100,000 mL/(g h); TOFPd, TOFPt, TOFPd + Pt, and specific reaction rate at 280 C were
3
0
−3
−1
−3 −1
−3 −1
.85 × 10
s
, 1.98 × 10
s
, 0.59 × 10
s , and 4.46 ␮mol/(gcat s), respectively). The deactivation of
the Pd2.41Pt sample induced by 2.5–5.0 vol% CO2 or H2O addition was reversible, but its deactivation due
to 100 ppm SO2 introduction was irreversible. It is concluded that the excellent catalytic performance
of the Pd2.41Pt sample was associated with its good mesoporous structure, Pd–Pt alloy and PdO–PtO2
coexistence, and good methane and oxygen activation ability.
©
2017 Elsevier B.V. All rights reserved.
1
. Introduction
In the past years, the Pd-based catalysts have been investigated
intensively and extensively due to their good catalytic perfor-
mance for methane combustion [2–5]. For example, PdO or Pd–PdO
mixture were recognized as the active phase of the supported
Pd catalysts at low temperatures, but the metallic Pd0 was the
active phase at high temperatures [1–3,6,7]. The performance of the
Pd-based catalysts for methane lean-burn combustion at low tem-
peratures could be improved if the Pd was alloyed with other metals
(e.g., Pt [8]). The Pt was found to be more active under the rich-burn
Methane is widely used in industrial and transportation activ-
ities as it is the main component of natural gas. Due to its tough
combustion, however, unburned methane at low temperatures
always exists in the exhaust. It is generally accepted that methane
induces a strong greenhouse effect with a global warming poten-
tial 20 times higher than that of CO [1]. Therefore, much attention
2
has been paid on reducing the emission of methane. Flame combus-
tion of methane at high temperatures often results in the emissions
of nitrogen oxides (NOx) and CO that are harmful to the environ-
ment. Compared to the conventional flame combustion, catalytic
combustion of methane is regarded as an efficient and green path-
way, in which the key issue is the availability of high-performance
catalysts at low temperatures. Therefore, it is highly desirable to
design and prepare novel catalysts with excellent low-temperature
activity and stability.
conditions (O /CH4 molar ratio <2) [9]. Although monometallic Pt
2
catalysts were less active than the monometallic Pd catalysts under
lean-burn conditions [9], some researchers [10–14] have pointed
out that substituting Pt for Pd could generate a catalyst that showed
enhanced catalytic activity and stability. The influence of Pt on cat-
alytic activity of a bimetallic catalyst, however, is still a debated
issue. For instance, Strobel et al. [15] investigated the combus-
tion of methane over the supported Pd–Pt catalysts with different
amounts of Pt and Pd, and observed a positive effect on catalytic
performance over the supported Pd–Pt catalysts with smaller Pt
amounts, but a negative effect on catalytic activity was observed
over the supported Pd–Pt catalysts with higher Pt loadings. Such a
phenomenon was also reported by Persson et al. [16], who claimed
∗ Corresponding authors.
E-mail addresses: yxliu@bjut.edu.cn (Y. Liu), jgdeng@bjut.edu.cn (J. Deng),
hxdai@bjut.edu.cn (H. Dai).
http://dx.doi.org/10.1016/j.mcat.2017.09.002
2
468-8231/© 2017 Elsevier B.V. All rights reserved.

1
92
X. Zhao et al. / Molecular Catalysis 442 (2017) 191–201
that the Pt could act either as a promoter or as an inhibitor for
methane combustion. Castellazzi et al. [13] attributed this result
to the extent of PdO formation, the strength of the metal–support
of silica was 1.0. The hexane solution was sonicated for 1 h. After
removing the excessive solution, the obtained brown composite
was dried at 120 C for 24 h. Such an impregnation process was
◦
0
interaction, and the stabilization of metallic Pd .
repeated several times to gain a high noble metal incorporation
(30–45 wt%) into the mesopores of the KIT-6. After being dried at
Recently, the materials consisting of mesoporous networks have
been found to be highly active for some reactions. These porous
materials possess large specific surface areas, high structural stabil-
ity, and good catalytic durability due to their nanoscale skeletons
and inter-connected hollow channels that are favorable for easy
mass transfer and high electron conductivity [17,18]. In the past
years, porous noble metal nanostructures have gained enormous
attention [19–24]. For example, the 3D porous architecture could
not only maximize the availability of electron transfer within nano-
sized surface but also provide better mass transfer of reactants to
the catalyst [18]. Xu and coworkers [22,24] conducted a systematic
investigation on the catalytic properties of nanoporous gold, and
found that the porous Au catalysts performed well in the oxidation
of CO.
It has been generally accepted that the metal–support interac-
tion plays an important role in improving the catalytic activity of
a supported noble metal catalyst. In order to leave out the effect
of the metal-support interaction, only noble metals can be used as
catalyst to investigate their catalytic performance and hence clar-
ify the active sites. Up to now, however, there have been no reports
on the preparation and catalytic applications of mesoporous PdxPt
alloys for methane combustion. In the present work, we used the
mesoporous silica (KIT-6) as hard template to fabricate the meso-
porous Pd, Pt and PdxPt alloys, characterized their physicochemical
properties, and evaluated their catalytic performance for methane
combustion. It is found that the PdxPt alloys showed much better
catalytic activity than the monometallic Pd or Pt.
◦
120 C for 24 h, the resulting powders were placed into a three-neck
flask at the ambient temperature for 30 min under the protection
of N . Then, a NaBH4 aqueous solution (0.30 mol/L) was added to
2
reduce the H PdCl4 and H PtCl6 precursors (NaBH /nobel metal
2
2
4
molar ratio = 10.00:1.00). Until no H2 was released, the black pow-
ders were centrifuged and the KIT-6 template was removed using
a hydrofluoric acid aqueous solution (10 wt%). After being washed
five times with 250 mL of deionized water (50 mL for each time),
◦
the sample was dried at 120 C overnight, and calcined in air at a
◦
◦
ramp of 1 C/min from RT to 400 C and kept at this temperature
for 2 h. By adjusting the molar ratios of H PdCl4 and H PtCl , the
2
2
6
PdxPt samples with different Pd/Pt molar ratios (x) were obtained.
For comparison proposes, single metal Pd or Pt was also fabricated
using the above method. The nominal x value was 9.00, 2.33, 1.00,
and 0.43, respectively. The real noble metal contents of the sam-
ples were determined by the inductively coupled plasma atomic
emission spectroscopic (ICP–AES) technique.
2.2. Catalyst characterization
Physicochemical Properties of the Pd, PdxPt, and Pt samples
were characterized by means of techniques, such as ICP–AES,
X-ray diffraction (XRD), transmission electron microscopy
(TEM), high-angle annular dark-field-scanning transmission
electron microscopic (HAADF–STEM) and element mapping,
N2 adsorption-desorption (BET), X-ray photoelectron spec-
troscopy, CO temperature-programmed reduction (CO-TPR), O2
temperature-programmed desorption (O -TPD), CH temperature-
2
. Experimental
2
4
programmed desorption (CH -TPD), and Fourier transform infrared
4
2.1. Catalyst preparation
(FT–IR) spectroscopy. The detailed characterization procedures
can be seen in the Supplementary material.
Ordered mesoporous silica (KIT-6) was synthesized under the
◦
hydrothermal conditions at 100 C for 24 h according to the proce-
dure of the previous literature [25]. As confirmed by the small-angle
X-ray diffraction pattern (Fig. S1) and transmission electron micro-
scopic image (Fig. S2), the as-synthesized KIT-6 template possessed
an ordered mesoporous architecture with an average pore size
of 6.8 nm. The mesoporous bimetallic PdxPt alloys were prepared
using the nanocasting method with KIT-6 as hard template. The
typical preparation procedures are as follows: 1.00 g of KIT-6 was
added to 40 mL of hexane at room temperature (RT). After being
vigorously stirred for 30 min and sonication for another 30 min,
a mixture of H PdCl (0.56 mol/L) and H PtCl6 (0.56 mol/L) aque-
2.3. Catalytic activity evaluation
Catalytic activities of the samples for methane combustion were
evaluated in a continuous flow fixed-bed quartz tube microreac-
tor. The reactant mixture was composed of 2.5 vol% CH + 20 vol%
4
O + 77.5 vol% N2 (balance), and the space velocity (SV) was ca.
2
1
00,000 mL/(g h). All of the activity data were collected when the
combustion of methane reached a steady state at a given temper-
ature. Catalytic activities of the samples were evaluated using the
temperatures (T10%, T50%, and T90%) required for achieving methane
conversions of 10, 50, and 90%, respectively. The detailed activity
evaluation procedures are put in the Supplementary material.
2
4
2
ous solution was then added dropwise under stirring. The ratio
of the volume of precursor aqueous solution to the pore volume
Table 1
BET surface areas, average pore sizes, pore volumes, average noble metal particle sizes (D), real Pd or Pt contents, and real Pd/Pt molar ratios of the samples.
Nominal
sample
Real sample
BET surface
area (m /g)
Average pore
size (nm)
Pore volume
(cm /g)
XRD result
ICP–AES result
2
3
Crystal phase
Da (nm)
Real Pd
content
Real Pt
content
(wt%)
Real Pd/Pt
molar ratiob
(mol/mol)
b
b
(
wt%)
Pd
Pd
30.2
25.5
31.9
29.2
29.6
31.1
19.8
18.1
15.6
14.9
14.4
13.7
0.137
0.131
0.142
0.149
0.147
0.123
Cubic
Cubic
Cubic
Cubic
Cubic
Cubic
9.4
7.9
7.1
7.2
7.4
6.8
100
82.3
56.8
35.1
18.9
–
–
–
Pd9Pt1
Pd7Pt3
Pd5Pt5
Pd3Pt7
Pt
Pd8.52Pt
Pd2.41Pt
Pd0.99Pt
Pd0.43Pt
Pt
17.7
43.2
64.9
81.1
100
8.52
2.41
0.99
0.43
–
a
Data were calculated according to the Scherrer equation using the FWHM of the (111) line of cubic Pd or Pt.
Data were determined by the ICP − AES technique.
b

X. Zhao et al. / Molecular Catalysis 442 (2017) 191–201
193
nominal Pd/Pt molar ratios in the PdxPt samples were rather close,
indicating that the preparation method was quite effective.
Fig. 1 shows the XRD patterns of the Pd, Pt, and PdxPt samples.
Compared to the XRD patterns of the standard Pd (JCPDS PDF# 46-
1
043) and Pt (JCPDS PDF# 04-0802) samples, one can realize that
◦
◦
◦
◦
◦
the diffraction peaks 2ꢀ = 40 , 46 , 68 , 82 , and 86 were assigned
to the (111), (200), (220), (311) and (222) crystal planes, and could
be indexed to the face-centered cubic crystal structure. It can be
also observed that the diffraction peak due to the (111) crystal plane
gradually shifted to a lower 2ꢀ with increasing the Pt content in the
PdxPt samples, and it located at the 2ꢀ between those of the Pd(111)
and Pt(111) crystal planes, indicating that an alloy was formed in
the PdxPt samples. Since the Pd and Pt samples were of a face-
centered cubic structure, the diffraction patterns of the PdxPt alloys
were similar to those of the pure Pd or Pt sample [26]. The average
grain sizes (Table 1) could be estimated according to the Scherrer
equation (D = 0.89ꢁ/(ˇcos ꢀ), where ꢁ is the X-ray wavelength, ˇ is
the full width at half maximum (FWHM) of the (111) plane of the
sample, and ꢀ is the corresponding diffraction angle). It is seen that
the calculated grain sizes of the Pd, Pt, and PdxPt samples were in
the range of 6.8–9.4 nm.
Fig. 2 shows the TEM images of the as-obtained noble metal
samples. The distinct color contrast between the dark skeletons
and bright background region suggests the formation of a three-
dimensional continuous network structure with interconnected
hollow voids in each of the noble metal sample. In other words,
a mesoporous architecture was generated in these samples. From
the SAED patterns (insets of Fig. 2) of the samples, one can see
several bright diffraction rings, which was indicative of formation
of polycrystalline noble metals. The wall thickness of the meso-
pores in the samples was 7–10 nm. Furthermore, the HAADF–STEM
and element analysis images (Fig. 3) of the typical Pd2.41Pt sample
clearly show that the distributions of Pd and Pt in the Pd2.41Pt sam-
ple throughout the entire domain were highly uniform. That is to
say, the Pd and Pt was present in the form of an alloy in the Pd2.41Pt
sample, in good agreement with the XRD results.
Fig. 1. XRD patterns of (a) Pd, (b) Pd8.52Pt, (c) Pd2.41Pt, (d) Pd0.99Pt, (e) Pd0.43Pt, and
f) Pt.
(
3
. Results and discussion
3.1. Crystal phase composition, pore structure, and surface area
The results (Table 1) of ICP–AES investigations reveal that the
real Pd/Pt molar ratios (x) in the PdxPt samples were 8.52, 2.41,
.99, and 0.43, respectively, corresponding to the nominal Pd/Pt
molar ratios of 9.00, 2.33, 1.00, and 0.43. Obviously, the real and
0
Fig. 2. TEM images of (a) Pd, (b) Pd8.52Pt, (c) Pd2.41Pt, (d) Pd0.99Pt, (e) Pd0.43Pt, and (f) Pt.

1
94
X. Zhao et al. / Molecular Catalysis 442 (2017) 191–201
Fig. 3. (a and e) HAADF-STEM images and (b–d, f–h) elemental mapping of the Pd2.41Pt sample in different regions.
Fig. 4. (A) Nitrogen adsorption–desorption isotherms and (B) pore-size distributions of (a) Pd, (b) Pd8.52Pt, (c) Pd2.41Pt, (d) Pd0.99Pt, (e) Pd0.43Pt, and (f) Pt.
Fig. 4 shows the N adsorption–desorption isotherms and pore-
trum is composed of the Pd 3d5/2–Pd 3d3/2 spin-orbit doublets, and
the binding energy (BE) of the spin-orbital splitting is 5.26 eV. For
PdO, the BE of Pd 3d5/2 was 336.6–337.7 eV [28]. Fig. 5 illustrates
the Pd 3d and Pt 4f XPS spectra of the samples and their surface
element compositions are summarized in Table 2. Compared to
the bulk element compositions obtained from the ICP–AES tech-
nique, the surface Pd/Pt molar ratios (Table 2) of the PdxPt samples
were higher (except for the Pd0.99Pt sample). By using the curve-
fitting approach, each Pd 3d spectrum could be decomposed into
four components at BE = 335.5, 337.3, 340.7, and 342.6 eV (Fig. 5A):
the components at BE = 335.5 and 340.7 eV were attributed to the
surface Pd0 species, whereas the ones at BE = 337.3 and 342.6 eV
2
size distributions of the mesoporous Pd, Pt, and PdxPt samples, and
their BET surface areas, average pore sizes, and pore volumes are
summarized in Table 1. It is observed from Fig. 4A that each sample
displayed a type II isotherm with a H2 hysteresis loop in the relative
pressure (p/p ) of 0.8–1.0. The H2 hysteresis loop (which showed
0
a large uptake) was related to the distortion of mesopores [27].
The formation of mesopores in these noble metal samples is also
confirmed by their pore-size distributions (Fig. 4B). The pore sizes
were in the range of 5–30 nm, with the average pore sizes being
1
3.7–19.8 nm (Table 1). The BET surface areas and pore volumes of
2
the noble metal samples were in the ranges of 25.5–31.9 m /g and
3
2+
0
.123–0.149 cm /g, respectively.
were assigned to the surface Pd species [28–32]. It can be clearly
observed that with increasing the Pt content in the PdxPt samples,
the surface Pd2+ content decreased but the surface Pd0 content
increased, indicating that the partial amount of surface PdO was
reduced by the Pt. That is to say, the surface Pd2+ concentration
could be modified by the doping of Pt. Similarly, the Pt 4f XPS
3
.2. Surface elemental composition and metal oxidation state
XPS is an effective technique to determine the surface elemental
compositions and metal states of a catalyst. The Pd 3d XPS spec-

X. Zhao et al. / Molecular Catalysis 442 (2017) 191–201
195
Fig. 5. (A) Pd 3d and (B) Pt 4f XPS spectra of (a) Pd, (b) Pd8.52Pt, (c) Pd2.41Pt, (d) Pd0.99Pt, (e) Pd0.43Pt, (f) Pt, and (g) used Pd2.41Pt.
Table 2
◦
Surface element compositions, catalytic activities, TOFs, and specific reaction rates at 280 C and SV = 100,000 mL/(g h) of the samples.
◦
C
Sample
Surface element composition
Catalytic activity
Methane combustion at 280
TOFPdb TOFPtb
Pd2+/Pd0
Pt4+/Pt0
Pd/Pt
molar ratio
T10%
T50%
T90%
( C)
TOFPd+Ptb
(×10
Specific
reaction rate
␮mol/(gcat
◦
◦
◦
−3 −1
−3 −1
−3 −1
s )
molar ratio
molar ratio
( C)
( C)
(×10
s
)
(×10
s
)
(
s))
Pd
12.75
11.10
7.57
1.86
1.03
–
–
–
313
297
272
292
292
282
254
350
338
303
322
328
372
283
368
351
322
338
348
506
306
0.11
0.21
0.85
0.45
0.95
–
–
–
1.06
1.67
4.46
1.49
1.69
2.70
–
Pd8.52Pt
Pd2.41Pt
Pd0.99Pt
Pd0.43Pt
Pt
2.78
2.12
0.90
0.48
0.58
2.93
12.66
2.82
0.90
0.46
–
1.92
1.98
0.45
0.41
0.53
–
0.19
0.59
0.22
0.28
–
Used Pd2.41Pta 40.69
2.78
–
–
a
Used sample after the drop and rise in reaction temperature.
TOFPd, TOFPt, and TOFPd + Pt were calculated according to TOFM = XC0/nM (nM is the molar amount of Pd, Pt or Pd + Pt in the samples, respectively).
b
spectrum of each sample could also be decomposed into four com-
ponents (Fig. 5B): the components at BE = 71.3 and 74.5 eV were
ascribed to the surface metallic Pt (Pt ) species, while the ones at
by the XPS technique in order to investigate the changes in chemi-
cal state of noble metals (Fig. 5g). Obviously, the surface oxidized Pd
and Pt concentrations increased after a recycle of methane combus-
tion when the reaction temperature dropped and rose again. Such
a result suggests that a higher surface PdO–PtO2 species concen-
tration would be favorable for methane combustion.
0
4
+
BE = 72.4 and 75.6 eV were due to the surface oxidized Pt (Pt
)
species [33–35]. The surface Pt4+/Pt molar ratio decreased with
increasing the Pt content in the PdxPt samples, but that on the
Pd0.43Pt surface was slightly lower than that on the monometal-
lic Pd surface. Since Pt is higher than Pd in electronegativity (the
Pauling electronegativity of Pt and Pd is 2.28 and 2.20, respectively)
0
3.3. Reducibility
[
36], a charge transfer would occur from Pd to Pt in the bimetallic
The reducibility of the Pd, Pt, and PdxPt samples was measured
by the CO-TPR technique. Fig. 6 illustrates the CO-TPR profiles of
the samples. The Pd sample showed the highest amount of CO2
0
2+
4+
0
catalysts via Pt + Pd → Pt + Pd . That is to say, Pt addition can
0
enhance the Pd concentration due to the strong electronic inter-
action between Pd and Pt. The XPS results reveal that the surface
◦
formed at 132 C. For the Pt sample, there were three peaks of CO
2
2+
0
Pd concentration decreased but the surface Pd concentration
increased, which are in good agreement with the fact that Pt pos-
sesses higher electron affinity than Pd. Therefore, the doping of Pt
could reduce the ability of Pd to form the PdO phase. The forma-
tion of PtPd/PdO core-shell structures has been proposed by some
researchers [16,37,38]. According to the XPS results, we deduce that
Pd and Pt might be oxidized on the surface after calcination of the
Pd, Pt or PdxPt sample, forming PdOx, PtOx or even Pd–O–Pt bonds
due to their similar electronic properties; whereas Pd–Pt metallic
alloy might be generated in the bulk of PdxPt. It should be pointed
out that slight shifts in BE were seen in the Pd 3d and Pt 4f XPS
spectra of the PdxPt samples, but their BE values were also simi-
lar to those reported in the literature [16,28]. The Pd2.41Pt sample
after the drop and rise in reaction temperature was characterized
◦
formation at 83, 280, and 345 C, respectively. However, the PdxPt
samples displayed different reduction behaviors, depending upon
their Pd/Pt molar ratios. Each of the Pt-containing bimetallic sam-
ples showed one small peak at 60–70 C and one broad peak at
0–300 C, confirming the formation of a Pd–Pt alloy. The adsorbed
◦
◦
9
CO could reduce the oxidized Pd and/or Pt at elevated temperatures.
Haneda et al. [39] studied the reducibility of the PdPt/Al O3 sam-
2
ples by the CO-TPR technique, and observed a new reduction peak
◦
above 200 C (which was ascribed to the reduction of alloyed Pd–Pt
◦
NPs) in addition to the broader reduction peaks at 100–200 C.
However, the catalytic activity cannot be fully correlated with
the result of the CO-TPR study. Castellazzi et al. [13] reported
strong evidence against a direct correlation between activity and
PdO reducibility in the Pt-containing catalysts. Such a lack in cor-

1
96
X. Zhao et al. / Molecular Catalysis 442 (2017) 191–201
Fig. 7. O2–TPD profiles of (a) Pd, (b) Pd8.52Pt, (c) Pd2.41Pt, (d) Pd0.99Pt, (e) Pd0.43Pt,
and (f) Pt.
Fig. 6. CO–TPR profiles of (a) Pd, (b) Pd8.52Pt, (c) Pd2.41Pt, (d) Pd0.99Pt, (e) Pd0.43Pt,
and (f) Pt.
the samples became lower in intensity and shifted to lower tem-
peratures, suggesting that the presence of Pt weakened the Pd–O
bond strength [13,40]. The results of XPS, TPR, and O -TPD studies
2
relation between reducibility and activity implies that the other
factors rather than the surface PdO reduction could partially control
the methane combustion rate over the bimetallic samples. Yash-
nik et al. [28] observed that a rise in the Pt share resulted in a
drop in PdO amount of the PdPt/MnLaAl11O19 catalyst due to for-
mation of the Pd–Pt alloy, and the formed Pd–Pt alloy implicitly
influenced the redox and catalytic performance; when the Pt con-
tent increased, the redox properties of the bimetallic catalysts was
worsen due to the rise in the share of PtO2 particles. Such a mis-
match between reducibility and activity suggests the contributions
of the other factors for methane combustion. Therefore, we con-
ducted the O -TPD and CH -TPD experiments to gain more clear
demonstrate formation of the PdOx and/or PtOx species, which
were mainly distributed on the particle surface of each sample.
We also measured the O2 desorption amounts of the samples, and
they decreased in the sequence of Pd (8.46 mmol/g) > Pd8.52Pt
(6.84 mmol/g) > Pd2.41Pt
(4.30 mmol/g) > Pd0.99Pt
(2.94 mmol/g) > Pt (2.00 mmol/g) > Pd0.43Pt (1.21 mmol/g),
a
result due to the drop in PdO content of the samples. The
decreasing PdO (i.e., increasing Pt) content was in line with
the results of XPS and CO-TPR investigations. However, des-
orbed amounts of the adsorbed oxygen species decreased in
the sequence of Pd (2.32 mmol/g) > Pt (1.42 mmol/g) > Pd8.52Pt
2
4
(1.15 mmol/g) > Pd0.43Pt
(1.02 mmol/g) > Pd0.99Pt
insights to the other factors influencing the combustion of methane
over the PdxPt samples.
(0.90 mmol/g) > Pd2.41Pt (0.64 mmol/g). Although the Pd2.41Pt
sample exhibited a less amount of adsorbed oxygen desorption,
methane combustion activity over it was higher than those over
the other samples (see below). Lapisardi et al. [40] also reported
the same observations, and pointed out that the doping of PdO with
Pt (even in a small Pt content) could induce the Pd–Pt interactions
which led to the improved catalytic activity of the Pt-doped PdO
sample for methane combustion. Persson et al. [16] claimed that
the metallic noble metal phase could be maintained even in an
oxidizing atmosphere due to the formation of a Pd–Pt alloy in
the catalysts, resulting in a better activity as compared with the
monometallic Pd catalyst. Another possible reason for the higher
catalytic activity might be that a more amount of O2 molecules
could be dissociatively adsorbed over the Pd–Pt alloys than over
the PdO phase, so that the alloy provided the PdO with oxygen
after reduction. The metals were much more active than the metal
oxides in dissociating oxygen and the Pt metal was better than the
Pd metal [41]. Hence, the superior property for O2 activation of
the Pd–Pt alloys might be the reason why the sample containing a
lower PdO content could perform better.
3.4. Desorption behavior
To gain a more clear insight into the effect of Pt on the redox
behavior of the PdxPt samples, we carried out the O -TPD experi-
2
ments and their profiles are illustrated in Figs. 7 and S3. For all of
the samples, the O desorption peaks centered at low temperatures
2
◦
(
(
5
300–480 C) were due to desorption of adsorbed oxygen species
or O ), whereas the ones at high temperatures
00–700 C) were due to desorption of lattice oxygen in the
−
◦
2−
−
e.g., O2 , O2
oxidized Pd and/or Pt (i.e., PdO and/or PtO ). As shown in Fig. 7,
2
the O -TPD profile was strongly modified by the addition of Pt to
2
the Pd sample, and the Pd-like behavior slowly changed to the
Pt-like behavior. Such observations were in good agreement with
those reported in the literature [5,13,16,40]. Two main phenomena
should be noticed: (i) the concentration of the adsorbed oxygen
species on the samples decreased with the rise in Pt content; and
(
ii) when the Pt content increased, the O2 desorption peaks of

X. Zhao et al. / Molecular Catalysis 442 (2017) 191–201
197
the monometallic Pd sample, respectively. Compared to the Pt-
containing samples, the addition of Pt induced a significant effect on
CO desorption, in which CO formation occurred at lower temper-
2
2
atures. That is to say, the Pt-containing samples exhibited a stronger
methane activation ability than the monometallic Pd sample. This
result demonstrates that Pt was beneficial for methane activation. It
is generally accepted that the Pd-based catalysts perform better for
methane combustion, whereas the Pt-based ones are more suitable
for the oxidation of higher hydrocarbons. Burch and coworkers [43]
proposed that Pt could activate the almost non-polar C–H bonds
of methane via a hemolytic mechanism (dissociative adsorption
of CH4 at noble metal sites). Castellazzi et al. [13] also reported
the dissociative adsorption of CH4 on the Pt surface. According to
the results of CO-TPR, O -TPD, and CH -TPD studies, the Pt sample
2
4
showed lower O2 and CO2 desorption temperatures as compared
with other samples at the low-temperature region, which might be
a result possibly due to the strong dissociation ability of methane
and O2 molecules on the Pt surface. Therefore, the Pt sample was
more active than the bimetallic samples at lower temperatures.
The CO2 desorption amounts of the samples reveal impor-
tant information related to the adsorption of CH . However, the
4
decreased CO quantity with the rise in Pt content of the PdxPt sam-
2
ples were mainly due to the decreased concentration of the oxygen
species. It is noticeable that for the Pd2.41Pt sample, the peak cen-
◦
tered at 381 C was nearly the same intensity as compared with that
◦
for the Pd sample, but the peak centered at 556 C reduced dramat-
Fig. 8. CH4–TPD profiles of (a) Pd, (b) Pd8.52Pt, (c) Pd2.41Pt, (d) Pd0.99Pt, (e) Pd0.43Pt,
and (f) Pt.
ically in intensity. Such results confirm that the methane molecules
adsorbed on the Pd2.41Pt surface could easily be activated and react
with the adsorbed oxygen species to produce CO and H O, a result
2
2
It is commonly known that adsorption of reactants is a key step
in heterogeneous catalysis. For methane combustion, the methane
adsorption ability is possibly the key factor influencing the activ-
possibly due to the strong dissociation ability of methane molecules
on the Pd–Pt alloys. The results of CO-TPR, O -TPD, and CH -TPD
2
4
investigations reveal that the doping of Pt to Pd gave rise to the
bimetallic PdxPt catalysts with decreased amounts of PdO species
and stronger ability to active O or CH , as compared with the pure
ity of a catalyst. CH -TPD experiments were conducted to gain
4
insights into the activation ability of the noble metal samples. How-
ever, it should be mentioned that desorption of methane from the
PdxPt samples could be regarded as the temperature-programmed
oxidation (TPO) of carbon species originated from methane [42].
As detected by the mass spectrometer, there was carbon dioxide
2
4
Pd sample. Such a phenomenon might be due to the interaction
between Pd and Pt, leading to the formation of a Pd–Pt alloy in
PdxPt and PdOx and PtOx species on the surface of the PdxPt sam-
ples. Persson et al. [16] reported that the PdO amount in the sample
did not influence the onset temperature of PdO decomposition, but
only changed its peak intensity. Hence, the shift in temperature of
CO reduction or O2 desorption observed over the PdxPt samples in
(
m/z = 44) signal but almost no methane (m/z = 15) signal from the
surface of the samples. The results indicate that CH4 on the cata-
lyst surface mainly existed in chemical adsorption form and then
reacted with the oxygen species, eventually desorbing as CO . As
2
the CO-TPR and O -TPD profiles was attributed to the promotion
2
shown in the CH -TPD profiles (Fig. 8) of the samples, there were
4
effect of Pt doping to Pd, rather than due to the lower PdO amount.
◦
three CO2 desorption peaks centered at 382, 452, and 602 C for
Fig. 9. (A) Methane conversion as a function of temperature over ( ) Pd, ( ) Pd8.52Pt, ( ) Pd2.41Pt, ( ) Pd0.99Pt, ( ) Pd0.43Pt, and (᭿) Pt; (B) methane conversion versus
temperature over the Pd2.41Pt sample when the temperature (a) rose, (b) dropped, and (c) rose again at SV = 100,000 mL/(g h).

1
98
X. Zhao et al. / Molecular Catalysis 442 (2017) 191–201
3.5. Catalytic performance
Fig. 9A shows the methane conversion versus temperature over
the as-prepared samples at a SV of 100,000 mL/(g h). It is con-
venient to compare the catalytic activities of the samples using
the reaction temperatures (T50% and T90%) requiring for achieving
methane conversions of 50 and 90% (Fig. 9A and Table 2), respec-
tively. Among all of the samples, the Pd
best catalytic activity: T50% = 303 C and T
Pt sample showed the
2.41
◦
◦
= 322 C, which were
90%
◦
much lower than those over the monometallic Pd (T
= 350 C
5
0%
◦
◦
◦
and T90% = 368 C) and Pt (T
= 372 C and T
= 506 C) sam-
50%
90%
ples. Therefore, the catalytic activity decreased in the sequence of
Pd2.41Pt > Pd0.99Pt > Pd0.43Pt > Pd8.52Pt > Pd > Pt. It should be noted
that all of the PdxPt alloy samples outperformed the monometal-
lic Pd or Pt sample, which was not in consistency with the results
reported in the literature [15,16]. The excellent catalytic perfor-
mance of the PdxPt alloy samples might be associated with the
co-existence of Pd–Pt alloy and PdO–PtO2.
We measured the methane conversions over the Pd2.41Pt
sample when the temperature rose, dropped, and rose again
◦
(
Fig. 9B), respectively. Methane conversions at 180–360 C during
the temperature-dropping process were higher than those during
the temperature-rising process. When the temperature rose again,
however, methane conversions were rather similar to those during
the temperature-dropping process. Comparing to the T10%, T50%, and
Fig. 10. Methane conversion versus temperature over the (
reduced Pd2.41Pt samples at SV = 100,000 mL/(g and h).
) oxidized and (
)
◦
Fig. 10 shows the catalytic activities over the oxidized and
reduced Pd2.41Pt samples for methane combustion. Obviously,
methane combustion activity significantly decreased after the
T90% (272, 303, and 322 C) over the Pd2.41Pt sample during the first
temperature-rising process, the T10%, T50%, and T90% decreased to
◦
2
54, 283, and 306 C during the second temperature-rising process,
Pd2.41Pt sample was reduced in 10 vol% H –90 vol% N (30 mL/min)
respectively. Similar phenomena for methane combustion during
the temperature-dropping and rising processes have been reported
in the literature [4,32,44]. It is generally believed that Pd is easily
2
2
◦
◦
at 300 C, with the T10%, T50%, and T
increasing by 43, 42,
90%
and 46 C, respectively. The result demonstrates that the oxidized
◦
Pd2.41Pt (PdO–PtO ) sample in air (30 mL/min) at 400 C was more
2
◦
oxidized into the PdO species at a certain temperature (<600 C).
0
0
active than the reduced Pd2.41Pt (metallic Pd –Pt ) sample. Similar
results have been reported by other researchers [2,49]. In addition,
we measured the catalytic activities over the oxidized and reduced
Pd2.41Pt samples after the “temperature rise-drop-rise” cycle, as
shown in Fig. S4. The oxidized sample after the “temperature rise-
drop-rise” cycle exhibited better activity than the fresh sample,
but there was almost no change in activity over the reduced sam-
ple. This phenomenon could be explained by their XPS results. It is
observed that the Pd/Pt molar ratio on the used Pd2.41Pt sample was
close to that on the fresh Pd2.41Pt sample (Table 2), indicating no
significant changes in Pd/Pt molar ratio on the surface of the fresh
Therefore, the enhancement in catalytic activity of the Pd2.41Pt
sample might be due to generation of a more amount of the PdO
species during the reaction process, which was supported by the
2+
0
4+
0
results (much higher surface Pd /Pd and Pt /Pt molar ratios
were detected on the used Pd2.41Pt sample (Table 2)) of the XPS
investigation.
It is more accurate to use the turnover frequencies (TOFs) and
specific reaction rates for evaluating the inherent catalytic activities
of the samples. The TOFM (TOFM = XC /nM, where X is the conversion
0
at a certain temperature, C0 (mol/(g s)) is the initial methane con-
centration per gram per second, and nM (mol) is the molar amount
of Pd, Pt or Pd + Pt) and specific reaction rates can be calculated
according to the activity data and amounts of Pd and Pt in the
Pd, Pt, and PdxPt samples, as summarized in Table 2. It is clearly
and used Pd2.41Pt samples. The concentrations of PdO and PtO2,
however, increased after the “temperature rise-drop-rise” cycle,
resulting in an increase in methane combustion activity.
In the past decades, a number of works on methane com-
bustion have been reported in the literature, and their catalytic
activities (T50% and T90%) over the catalysts reported in the liter-
ature and the Pd2.41Pt catalyst reported in the present work are
summarized in Table S1. Apparently, our Pd2.41Pt catalyst outper-
observed that the Pd
Pt sample exhibited a relatively high TOFPd
2
.41
−
3
−1
−3 −1
), the highest
(
0.85 × 10
s
), the highest TOFPt (1.98 × 10
s
−
3
−1
TOF
(
(0.59 × 10
s
), and the highest specific reaction rate
Pd+Pt
◦
4.46 ␮mol/(gcat s)) in the combustion of methane at 280 C, a result
possibly due to the interaction between Pd and Pt. Usually, it is
more accurate to calculate the TOFs based on the surface atoms.
However, it is well known that the surface element composition of
a sample is not the same as its bulk element composition. Since the
Pd/Pt molar ratios on the surface of the samples were not the same
as those determined by the ICP–AES technique, it is hard to obtain
the exact numbers of surface Pd and Pt atoms, which makes the cal-
culation of the TOF values difficult. Actually, some researchers also
used the total amount of the noble metal to calculate the TOF value.
For example, Li et al. [45] calculated the TOFs of the PtSn/TS-1 cata-
lysts for propane dehydrogenation according to the molar amount
of Pt in the samples; Hutchings and coworkers also estimated the
TOFs on the basis of the total metal loading [46–48]. Therefore, the
TOFs calculated according to the molar amounts of noble metals
could reflect the inherent catalytic performance of the noble metal
samples.
formed the 1.0 wt% Pd/Al O3 [50], 1.92 wt% Pd/Co O4 [51], 1.1 wt%
2
3
Pt/3DOM Ce0.6Zr0.3Y0.1O [52], 3.0 wt% AuPd/3DOM La0.6Sr0.4MnO
2
3
[
[
2
53], 3.1 wt% Pd–2.9 wt% Pt/LaMnAl11O19 [54], 5 wt% PdCo/Al O3
16], and 1.0 wt% Co–1.0 wt% Pd/Al O3 [8] catalysts.
2
3.6. Effect of water vapor, CO , and SO on catalytic activity
2
2
Although the Pd-based catalysts are the most active mate-
rials for methane combustion [1,2,8,55], their activity stability
is not good enough [11,16,55,56]. It has been reported that the
supported monometallic Pt catalysts were less active than the
supported monometallic Pd catalysts for methane combustion
[57]. The partial substitution of Pd with Pt, however, could gen-
erate a catalyst that improve its catalytic stability for methane
combustion [2,58]. Fig. 11 shows the catalytic activities versus

X. Zhao et al. / Molecular Catalysis 442 (2017) 191–201
199
combustion in the absence of H O. In the presence of 2.5 vol% water
2
vapor, however, methane conversion first decreased slightly (ca.
5
% loss in activity) and then recovered to its original value in the
absence of H O. By switching to 5.0 vol% water vapor, the catalytic
2
activity maintained almost unchanged. After cutting off the 5.0 vol%
water vapor, the catalytic activity was totally restored to that in
the absence of water vapor. The results indicate that the Pd2.41Pt
sample was catalytically stable in the presence of water vapor in
a certain concentration below 5.0 vol%. As the main active phase,
PdO is readily to form the Pd(OH)2 species through the reaction
of PdO + H O → Pd(OH) in the presence of water vapor, where the
2
2
Pd(OH)2 was usually regarded as an inactive phase [59]. The deac-
tivation due to water of the Pd-based catalysts was related to the
reaction temperature. Burch and coworkers [49,60] pointed out
that the PdO/Al O3 catalysts were irreversibly deactivated due to
2
hydroxylation of PdO to inactive Pd(OH)2 particles in the presence
◦
of water at a lower temperature (e.g., 300 C), but their deactivation
induced by water was insignificant at a higher temperature (e.g.,
◦
4
50 C). Therefore, the stability of the Pd2.41Pt sample for methane
combustion under the wet conditions might be related to the inter-
action between Pd and Pt, which not only improved the catalytic
activity, but also inhibited the formation of Pd(OH)2 [10–12,40].
In the case of 2.5 or 5.0 vol% CO2 introduction to the reaction sys-
Fig. 11. Methane conversion versus on-stream reaction time over the Pd, Pd2.41Pt,
and Pt samples at 350 C.
◦
◦
tem over the Pd2.41Pt sample at 310 C and SV = 100,000 mL/(g h),
◦
one can clearly see that the catalytic activity was durable after
on-stream methane combustion at 350 C over the Pd and Pt
2
0 h of methane combustion in the presence of 2.5 or 5.0 vol% CO2
samples at SV = 100,000 mL/(g h) and over the Pd2.41Pt catalyst at
SV = 120,000 mL/(g h). Apparently, the Pd2.41Pt sample exhibited
the highest activity and no significant loss in activity during 40 h of
on-stream reaction, indicating that the Pd2.41Pt sample possessed
good catalytic stability. The Pt sample displayed a stability simi-
lar to that of the Pd2.41Pt sample, the catalytic activity was much
lower over the former than that over the latter. Methane conversion
over the Pd sample, however, gradually decreased from 78 to 70%
within 40 h of on-stream methane combustion. We also recorded
the TEM image of the used Pd2.41Pt sample after 40 h of on-stream
methane combustion, as shown in Fig. S5. Obviously, there were no
significant changes in mesostructure of the fresh and used Pd2.41Pt
samples. Therefore, the alloying of Pd with Pt could increase the
catalytic activity and stability. The good stability of the Pd2.41Pt
sample might be associated with the fact that Pt could improve the
anti-sintering ability of Pd NPs [11,15].
(
Fig. 12B). Hence, CO had no significant negative effect on catalytic
2
activity of the Pd2.41Pt sample for methane combustion. Similar
results have been reported in the literature [60].
To examine the effect of SO2 on catalytic activity, we con-
◦
ducted methane combustion over the Pd2.41Pt sample at 310 C and
SV = 100,000 mL/(g h). As shown in Fig. 13A, methane conversion
decreased from 90.0 to 31.1% after 10 h of 100 ppm SO2 introduc-
tion. When the SO2 was cut off, however, methane conversion
slowly increased to 72.4%. This result indicates that deactivation
of the Pd2.41Pt sample induced SO was partially reversible. To elu-
2
cidate the reason for such a deactivation of the Pd2.41Pt sample,
we carried out the in situ DRIFT experiments, and their spectra
are shown in Fig. 13B. According to the literature [32,61,62], the
2
−
absorption bands of the H O, CO , and SO
3
2−
or SO4
species were
2
2
−
1
−1
−1
at 1640 cm , at 2341 and 2360 cm , and at 1159 and 1045 cm
respectively. It can be clearly seen from Fig. 13B that in addition to
the absorption bands due to H O and CO , there were no absorp-
tion bands assignable to the SO3
within 4 h of methane combustion in the absence of SO , but the
,
H O and CO are the products of methane combustion. It is nec-
2
2
essary to examine the effects of H O and CO on catalytic activity of
2
2
2−
2
2
2−
or SO4
species on the sample
the typical sample. We introduced 2.5 or 5.0 vol% water vapor to the
reaction system and monitored the methane conversions over the
2
−
1
2−
◦
absorption bands at 1159 and 1045 cm (assignable to the SO3
Pd2.41Pt sample at 310 C and SV = 100,000 mL/(g h). As shown in
or SO42 species) were detected and their intensity was much
−
Fig. 12A, methane conversion was not altered within 4 h of methane
◦
Fig. 12. Effects of (A) water vapor and (B) CO2 on methane conversion over the Pd2.41Pt sample at 310 C and SV = 100,000 mL/(g h).

2
00
X. Zhao et al. / Molecular Catalysis 442 (2017) 191–201
◦
Fig. 13. (A) Effect of 100 ppm SO2 on methane conversion over the Pd2.41Pt sample at 310 C and SV = 100,000 mL/(g h), and (B) in situ DRIFT spectra of the Pd2.41Pt sample
◦
◦
◦
C
after (a) 4 h of methane combustion at 310 C, (b) 10 h of methane combustion at 310 C when 100 ppm SO2 was introduced, and (c) 9 h of methane combustion at 310
when 100 ppm SO2 was cut off.
higher after 10 h of SO2 introduction than that after cutting off the
Appendix A. Supplementary data
SO . This result reasonably explains the partially reversible deacti-
2
vation of the Pd2.41Pt sample induced by SO . The negative effect of
Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.mcat.2017.09.
002.
2
SO addition on methane combustion might be due to the attack of
2
SO2 on the PdO component in the oxidized Pd2.41Pt sample, which
made highly active PdO transform into inactive PdSO or PdSO . The
3
4
SO , however, would have little influence on the Pd–Pt alloy, thus
the partially deactivated sample still showed a methane conversion
2
References
of 31.1%. After cutting off the SO , methane conversion was par-
tially recovered (72.4%), a result possibly due to the decomposition
2
[
1] J. Xu, L. Ouyang, W. Mao, X.J. Yang, X.C. Xu, J.J. Su, T.Z. Zhuang, H. Li, Y.F. Han,
ACS Catal. 2 (2012) 261–269.
2] P. Gélin, M. Primet, Appl. Catal. B 39 (2002) 1–37.
3] N.M. Kinnunen, J.T. Hirvi, T. Venäläinen, M. Suvanto, T.A. Pakkanen, Appl.
Catal. A 397 (2011) 54–61.
[
[
of the weakly adsorbed SO species. Furthermore, we activated the
2
1
5
00 ppm SO -treated Pd2.41Pt sample in a N2 flow of 30 mL/min at
2
◦
00 C for 1 h, and measured its catalytic activity for methane com-
[4] M. Cargnello, J.J. Delgado Jaén, J.C. Hernández Garrido, K. Bakhmutsky, T.
Montini, J.J. Calvino Gámez, R.J. Gorte, P. Fornasiero, Science 337 (2012)
◦
bustion. As shown in Fig. S6, methane conversions at 310 C over
the activated Pd2.41Pt sample were almost recovered to the original
values over the fresh Pd2.41Pt sample, and the catalytic activity was
stable within 8 h of methane combustion.
713–717.
[
5] N. Sadokhina, G. Smedler, U. Nylén, M. Olofsson, L. Olsson, Appl. Catal. B 200
(2017) 351–360.
[6] J. Nilsson, P. Carlsson, S. Fouladvand, N.M. Martin, J. Gustafson, M.A. Newton,
E. Lundgren, H. Gronbeck, M. Sköglundh, ACS Catal. 5 (2015) 2481–2489.
[
7] D.N. Gao, C.X. Zhang, S. Wang, Z.S. Yuan, S.D. Wang, Catal. Commun. 9 (2008)
583–2587.
8] K. Persson, A. Ersson, K. Jansson, N. Iverlund, S.G. Järås, J. Catal. 231 (2005)
39–150.
2
4
. Conclusions
[
1
The mesoporous Pd, Pt, and PdxPt alloys could be prepared via
[9] R. Burch, P.K. Loader, Appl. Catal. B 1–2 (1994) 149–164.
10] H. Yamamoto, H. Uchida, Catal. Today 45 (1998) 147–151.
11] K. Narui, H. Yata, K. Furuta, A. Nishida, Y. Kohtoku, T. Matsuzaki, Appl. Catal. A
[
[
the KIT-6-templating route. The mesoporous samples displayed
2
a surface area of 25.5–31.9 m /g and an average pore size of
1
79 (1999) 165–173.
[12] N.M. Kinnunen, J.T. Hirvi, M. Suvanto, T.A. Pakkanen, J. Mol. Catal. A 356
2012) 20–28.
1
3.7–19.8 nm. An Pd–Pt alloy was formed in each of the PdxPt
samples. The addition of Pt to Pd exerted a significant effect on
(
[
[
13] P. Castellazzi, G. Groppi, P. Forzatti, Appl. Catal. B 95 (2010) 303–311.
14] G. Corro, C. Cano, J.L.G. Fierro, J. Mol. Catal. A 315 (2010) 35–42.
the redox property of the Pd sample. The PdO–PtO were more
2
0
0
active than the Pd –Pt . Among all of the samples, the Pd2.41Pt
sample showed the highest catalytic performance for methane
[15] R. Strobel, J. Grunwaldt, A. Camenzind, S.E. Pratsinis, A. Baiker, Catal. Lett. 104
2005) 9–16.
[16] K. Persson, A. Ersson, K. Jansson, J.L.G. Fierro, S.G. Järås, J. Catal. 243 (2006)
4–24.
(
◦
◦
combustion (at SV = 100,000 mL/(g h), T
= 272 C, T
= 303 C,
1
0%
50%
1
◦
3
◦
−3 −1
and
T90% = 322 C;
at
280 C,
TOF = 0.85 × 10
s
,
[17] J.T. Zhang, C.M. Li, Chem. Soc. Rev. 41 (2012) 7016–7031.
[18] C.Z. Zhu, D. Du, A. Eychmüller, Y.H. Lin, Chem. Rev. 115 (2015) 8896–8943.
Pd
s
−
−1
−3 −1
, and specific
TOFPt = 1.98 × 10
s
,
TOFPd+Pt = 0.59 × 10
[
19] X.F. Zhang, P.F. Guan, L. Malic, M. Trudeau, F. Rosei, T. Veres, J. Mater. Chem. A
(2015) 2050–2056.
reaction rate = 4.46 ␮mol/(gcat s)). The deactivation of the Pd2.41Pt
3
sample induced by 2.5–5.0 vol% CO or H O addition was reversible,
2
2
[20] Y. Wang, N. Su, L. Ye, Y.H. Ren, X.Y. Chen, Y.J. Du, Z.H. Li, B. Yue, S.C.E. Tsang, Q.
Chen, H.Y. He, J. Catal. 313 (2014) 113–126.
21] H.J. Wang, H.Y. Jeong, M. Imura, L. Wang, L. Radhakrishnan, N. Fujita, T. Castle,
O. Terasaki, Y. Yamauchi, J. Am. Chem. Soc. 133 (2011) 14526–14529.
[22] C.X. Xu, X.H. Xu, J.X. Su, Y. Ding, J. Catal. 252 (2007) 243–248.
[23] A. Saramat, P. Thormählen, M. Skoglundh, G.S. Attard, A.E.C. Palmqvist, J.
Catal. 253 (2008) 253–260.
but its deactivation due to 100 ppm SO2 introduction was irre-
versible. It is concluded that the good mesoporous structure, Pd–Pt
alloy and PdO–PtO2 coexistence, and good methane and oxygen
activation ability were responsible for the excellent catalytic
performance of the Pd2.41Pt sample.
[
[
[
[
[
24] H.M. Duan, C.X. Xu, J. Catal. 332 (2015) 31–37.
25] F. Kleitz, S.H. Choi, R. Ryoo, Chem. Commun. (2003) 2136–2137.
26] C.X. Xu, Q. Hao, H.M. Duan, J. Mater. Chem. A 2 (2014) 8875–8880.
27] C.Z. Zhu, Q.R. Shi, S.F. Fu, J.H. Song, H.B. Xia, D. Du, Y.H. Lin, Adv. Mater. 28
Acknowledgments
(
2016) 8779–8783.
[
[
28] S.A. Yashnik, Y.A. Chesalov, A.V. Ishchenko, V.V. Kaichev, Z.R. Ismagilov, Appl.
Catal. B 204 (2017) 89–106.
29] K.R. Priolkar, P. Bera, P.R. Sarode, M.S. Hegde, S. Emura, R. Kumashiro, N.P.
Lalla, Chem. Mater. 14 (2002) 2120–2128.
This work was supported by the NSF of China (21377008 and
1677004) and National High Technology Research and Develop-
ment Program (“863” Program) of China (2015AA034603).
2

X. Zhao et al. / Molecular Catalysis 442 (2017) 191–201
201
[30] S.H. Xie, J.G. Deng, S.M. Zang, H.G. Yang, G.S. Guo, H. Arandiyan, H.X. Dai, J. [47] M.I.B. Saiman, G.L. Brett, R. Tiruvalam, M.M. Forde, K. Sharples, A. Thetford,
Catal. 322 (2015) 38–48.
R.L. Jenkins, N. Dimitratos, J.A. Lopez-Sanchez, D.M. Murphy, D. Bethell, D.J.
Willock, S.H. Taylor, D.W. Knight, C.J. Kiely, G.J. Hutchings, Angew. Chem. Int.
Ed. 51 (2012) 5981–5985.
[
[
31] M. Brun, A. Berthet, J.C. Bertolini, J. Electron Spectrosc. 104 (1999) 55–60.
32] S.H. Xie, J.G. Deng, S.M. Zang, Z.H. Zhang, H. Arandiyan, H.X. Dai, Environ. Sci.
Technol. 51 (2017) 2271–2279.
[48] G.T. Whiting, S.A. Kondrat, C. Hammond, N. Dimitratos, Q. He, D.J. Morgan,
N.F. Dummer, J.K. Bartley, C.J. Kiely, S.H. Taylor, G.J. Hutchings, ACS Catal. 5
(2015) 637–644.
[49] R. Burch, F.J. Urbano, Appl. Catal. A 124 (1995) 121–138.
[50] W.R. Schwartz, D. Ciuparu, L.D. Pfefferle, J. Phys. Chem. C 116 (2012)
8587–8593.
[
[
33] P. Bera, K.C. Patil, V. Jayaram, G.N. Subbanna, M.S. Hegde, J. Catal. 196 (2000)
2
93–301.
34] M.Y. Mao, H.Q. Lv, Y.Z. Li, Y. Yang, M. Zeng, N. Li, X.J. Zhao, ACS Catal. 6 (2016)
18–427.
4
[
[
35] H. Arandiyan, H.X. Dai, K.M. Ji, H.Y. Sun, J.H. Li, ACS Catal. 5 (2015) 1781–1793.
36] J. Xu, L. Ouyang, G.J. Da, Q.Q. Song, X.J. Yang, Y.F. Han, J. Catal. 285 (2012)
[51] M.T. Makhlouf, B.M. Abu-Zied, T.H. Mansoure, J. Nanopart. 2013 (2013)
384350.
7
4–82.
[
[
37] A. Morlang, U. Neuhausen, K.V. Klementiev, Appl. Catal. B 60 (2005) 191–199.
38] N.M. Martin, J. Nilsson, M. Skoglundh, E.C. Adams, X. Wang, P. Velin, G.
Smedler, A. Raj, D. Thompsett, H.H. Brongersma, T. Grehl, G. Agostini, O.
Mathon, S. Carlson, K. Nore, F.J. Martinezcasado, Z. Matej, O. Balmes, P.
Carlsson, J. Phys. Chem. C 120 (2016) 28009–28020.
[52] H. Arandiyan, H.X. Dai, K.M. Ji, H.Y. Sun, J.H. Li, ACS Catal. 5 (2015) 1781–1793.
[53] K. Sasaki, O. Tanaike, H. Konno, Can Miner. 36 (1998) 1225–1235.
[54] Y. Wang, H. Arandiyan, J. Scott, M. Akia, H.X. Dai, J.G. Deng, K. Aguey-Zinsou, R.
Amal, ACS Catal. 6 (2016) 6935–6947.
[55] A. Ersson, H. Ku sˇ ar, R. Carroni, T. Griffin, S. Järås, Catal. Today 83 (2003)
[39] M. Haneda, K. Suzuki, M. Sasaki, H. Hamada, M. Ozawa, Appl. Catal. A 475
265–277.
(
2014) 109–115.
[56] K. Persson, K. Jansson, S.G. Järås, J. Catal. 245 (2007) 401–414.
[57] M. Lyubovsky, L.L. Smith, M. Castaldi, H. Karim, B. Nentwick, S. Etemad, R.
LaPierre, W.C. Pfefferle, Catal. Today 83 (2003) 71–84.
[58] Y. Ozawa, Y. Tochihara, A. Watanabe, M. Nagai, S. Omi, Appl. Catal. A 259
(2004) 1–7.
[59] W.R. Schwartz, L.D. Pfefferle, J. Phys. Chem. C 116 (2012) 8571–8578.
[60] R. Burch, F.J. Urbano, P.K. Loader, Appl. Catal. A 123 (1995) 173–184.
[61] W.R. Schwartz, D. Ciuparu, L.D. Pfefferle, J. Phys. Chem. C 116 (2012)
8587–8593.
[
40] G. Lapisardi, L. Urfels, P. Gélin, M. Primet, A. Kaddouri, E. Garbowski, S. Toppi,
E. Tena, Catal. Today 117 (2006) 564–568.
[
[
[
[
41] S. Bedrane, C. Descorme, D. Duprez, Appl. Catal. A 289 (2005) 90–96.
42] J.R. Paredes, E. Díaz, F.V. Díez, S. Ordó n˜ ez, Energy Fuel 23 (2009) 86–93.
43] R. Burch, D.J. Crittle, M.J. Hayes, Catal. Today 47 (1999) 229–234.
44] C. Chen, Y.H. Yeh, M. Cargnello, C.B. Murray, P. Fornasiero, R.J. Gorte, ACS
Catal. 4 (2014) 3902–3909.
45] J.C. Li, J.M. Li, Z. Zhao, X.Q. Fan, J. Liu, Y.C. Wei, A.J. Duan, Z.A. Xie, Q.L. Liu, J.
Catal. 352 (2017) 361–370.
46] L. Kesavan, R. Tiruvalam, M.H.A. Rahim, M.I.B. Saiman, D.I. Enache, R.L.
Jenkins, N. Dimitratos, J.A. Lopez-Sanchez, S.H. Taylor, D.W. Knight, C.J. Kiely,
G.J. Hutchings, Science 331 (2011) 195–199.
[
[
[62] Z.X. Wu, J.G. Deng, Y.X. Liu, S.H. Xie, Y. Jiang, X.T. Zhao, J. Yang, H. Arandiyan,
G.S. Guo, H.X. Dai, J. Catal. 332 (2015) 13–24.