Stability of Pt particles on ZrO2 support during partial oxidation of methane: DRIFT studies of adsorbed CO

Article abstract of DOI:10.1016/j.molcata.2007.11.020

In this work, diffuse reflectance infrared Fourier transformation (DRIFT) spectroscopy of adsorbed CO was studied to investigate the stability of Pt/ZrO₂ catalysts for partial oxidation of methane. The studies confirmed that the preparation method has a significant effect on the catalyst performance and the catalyst treatment using argon glow discharge plasma improves the catalyst stability. TG analysis showed that coke formation is not significant on both samples with and without glow discharge plasma treatment, indicating the coke formation is not responsible for the catalyst deactivation. The IR band intensity of adsorbed CO decreases remarkably with increasing calcination temperature, suggesting the sintering of Pt particles in the presence of O₂. The sintering of plasma treated sample is slower than that of the untreated sample, especially at higher temperature, which indicates that the plasma treated sample is more resistant to sintering under the oxidizing atmosphere.

Full text of DOI:10.1016/j.molcata.2007.11.020

Available online at www.sciencedirect.com
Journal of Molecular Catalysis A: Chemical 282 (2008) 67–73
Stability of Pt particles on ZrO support during partial oxidation
2
of methane: DRIFT studies of adsorbed CO
a
a
a,∗
b
Xinli Zhu , Yongbing Xie , Chang-jun Liu , Yue-ping Zhang
a
Key Laboratory for Green Chemical Technology of Ministry of Education, School of Chemical Engineering, Tianjin University, Tianjin 300072, China
b
Department of Chemistry, Tianjin University, Tianjin 30072, China
Received 13 March 2007; received in revised form 18 November 2007; accepted 23 November 2007
Available online 10 January 2008
Abstract
In this work, diffuse reflectance infrared Fourier transformation (DRIFT) spectroscopy of adsorbed CO was studied to investigate the stability
of Pt/ZrO catalysts for partial oxidation of methane. The studies confirmed that the preparation method has a significant effect on the catalyst
2
performance and the catalyst treatment using argon glow discharge plasma improves the catalyst stability. TG analysis showed that coke formation
is not significant on both samples with and without glow discharge plasma treatment, indicating the coke formation is not responsible for the catalyst
deactivation. The IR band intensity of adsorbed CO decreases remarkably with increasing calcination temperature, suggesting the sintering of Pt
particles in the presence of O
which indicates that the plasma treated sample is more resistant to sintering under the oxidizing atmosphere.
2007 Elsevier B.V. All rights reserved.
2
. The sintering of plasma treated sample is slower than that of the untreated sample, especially at higher temperature,
©
Keywords: Partial oxidation; Methane; Pt/ZrO2; Plasma treatment; Glow discharge; Deactivation; DRIFT; Synthesis gas; Sintering
1
. Introduction
ing and steam reforming of the left CH4 occurs in the second
step [9,10]. Both CO2 and steam reforming of CH4 are highly
endothermic reactions, and thus the potential of coke formation
is high at high reaction temperature. Coke formation has been
suggested as the deactivation reason for POM over Pt/ZrO2 [9].
However, little attention has been paid to the Pt sintering in the
presence of O2 for POM over Pt/ZrO2.
Synthesis gas (CO and H2) is now industrially produced
through steam reforming of methane, which has some obvious
drawbacks, such as intense energy input and high H2/CO ratio
for further synthesis [1]. Catalytic partial oxidation of methane
(POM) to synthesis gas (1) is a promising alternative to steam
reforming. The weak exothermic property of POM leads to some
advantages, such as lower energy input and higher operating
space velocity [2,3]. In addition, the theoretical H2/CO ratio of
this reaction is 2, which is very suitable for further synthesis of
hydrocarbons and/or oxygenated hydrocarbons.
Because of the high price and low availability, reduction in
the amount of Pt usage is highly desirable. However, charac-
terization of supported catalysts with a small amount of metal
is very difficult since the small amount limits characterization
methods, such as X-ray powder diffraction (XRD). Infrared (IR)
spectroscopy using CO as probe molecule is a well established
method for characterizing supported metal catalysts [11–15].
The IR spectrum of adsorbed CO is very sensitive to the particle
size, morphology, and electronic structure, and this information
can be obtained in turn.
On the other hand, preparation methods show a significant
influence on the performance of catalysts for partial oxida-
tion of methane [16]. We previously observed a remarkable
improvement in the activity and stability of catalysts for POM
can be achieved by using glow discharge plasma treatment
of the catalyst [17,18]. Glow discharge is one of non-thermal
◦
1
2
CH4 + O2 → CO + 2H2 (ΔH
= −35 kJ mol)
(1)
298 K
Pt/ZrO2 is an active and stable catalyst for CO2 reforming of
methane [4–6]. It has been tested for POM. However, a quick
deactivation was usually observed [7–9]. A two-step reaction
mechanismwasproposedforPOMoverPt/ZrO2;CH4 isdirectly
oxidized to CO2 and H2O in the first step, and then CO2 reform-
∗
Corresponding author. Fax: +86 22 27890078.
E-mail addresses: coronacj@tju.edu.cn, ughg cjl@yahoo.com (C.-j. Liu).
1
381-1169/$ – see front matter © 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2007.11.020

6
8
X. Zhu et al. / Journal of Molecular Catalysis A: Chemical 282 (2008) 67–73
plasma phenomena, characterized by high electron tempera-
ture (10,000–100,000 K) and low gas temperature (as low as
room temperature) [19]. Catalyst preparation using non-thermal
plasma treatment attracts increasing attentions in recent years
ing argon. All gases used are in ultrahigh purity, without further
purification.
2.3. Characterizations
[
17–24]. The energetic species (electrons, ions, and radicals)
in plasmas could modify the metal particle size, morphology,
and metal-support interactions of catalysts [20], leading to some
specific catalytic properties.
The aim of this work is to study the structure and stability of
the Pt/ZrO2 catalyst for POM using IR spectra of adsorbed CO.
The enhanced stability of Pt/ZrO2 catalysts via glow discharge
plasma treatment has also been investigated.
The dispersion of Pt was measured using a volumetric system
(Quantachrome, Autosorb-1). The catalyst sample was firstly
◦
reduced at 300 C for 2 h. Then the CO chemisorption was per-
◦
formed at 40 C. The ratio of CO:Pt = 1:1 was used to calculate
the dispersion of Pt. Pt particle size was estimated by assuming
a spherical particle shape.
The thermogravimetric (TG) analysis of coke formed was
carried out using a TGA-50 system (Shimadzu). The oxidation
◦
2
. Experimental
temperature was increased from room temperature to 800 C at
a rate of 10 C/min under flowing air (25 cm /min).
◦
3
2
.1. Catalyst preparation
Diffuse reflectance infrared Fourier transformation (DRIFT)
spectra of adsorbed CO were recorded by using an IR spectrom-
eter (Bruker, Tensor 27) equipped with a liquid nitrogen cooled
MCT detector, a diffuse reflectance accessory (Praying Man-
tis, Harrick) and a high temperature cell (HVC, Harrick). The
2
ZrO2 (SBET = 8.3 m /g, monoclinic)waspurchasedfromAlfa
Aesar, and used as received. It was impregnated at room tem-
perature for 12 h with aqueous solution of H2PtCl4 6H2O (99%,
Tianjin North China Reagent Co.), followed by drying at 110 C
•
◦
cell was fitted with two CaF windows. The catalyst powder of
2
for another 12 h. The resulting sample was divided into two
parts. One part was directly calcined in a muffle at 600 C for
100 mg was loaded in the sample cup of high temperature cell.
The sample was reduced in situ at 300 C (or 500 C) for 1 h
◦
◦
◦
4
h (denoted by PtZr-C), while the other part was treated with
using flowing H , and then purged by flowing He for 30 min at
2
argon glow discharge plasma for 1 h followed by calcination at
the same temperature. After that, the temperature was decreased
to 25 C, and the background spectrum was recorded. 1.11%
◦
◦
6
00 C for another 4 h (denoted by PtZr-PC). The amount of Pt
loading is 0.5 wt%. Sometimes the calcination was carried out
at 400 C for 4 h or at 800 C for 4 h after calcination at 600 C
for 4 h, which was specially addressed in the text.
The plasma treatment was carried out in a glow discharge
system as described previously [17]. Briefly, 1 g catalyst sample
loaded in a quartz boat was put in the discharge cell. When
the pressure was adjusted in the range of 200–100 Pa, glow
discharge plasma was ignited by applying voltage of 900 V to
the electrodes. The 100 Hz square wave signal was generated
by a function/arbitrary waveform generator (Hewlett Packard,
CO diluted by He was introduced for adsorption for 30 min.
The sample was then purged by flowing He for another 30 min.
Temperature programmed desorption of CO was performed by
◦
◦
◦
◦
◦
increasing the temperature to 450 C at a rate of 10 C/min under
flowing helium. Spectra were recorded with the background sub-
−
1
tracted at a resolution of 4 cm and 64 scans accumulation. All
spectra were illustrated using Kubelka Munk units, which is
linear with the concentration of surface species.
3. Results and discussion
3
3120A), and amplified to 900 V by a high voltage amplifier
(
Trek, 20/20B). High purity argon was used as plasma form-
3.1. Chemisorption results
ing gas. Each treatment lasted 10 min, and the treatment was
repeated 6 times.
Table 1 gives the results of chemisorption. The dispersion of
the PtZr-C sample is 27.9%, and the estimated Pt particle size is
4.1 nm. For the PtZr-PC sample, the dispersion is slightly lower,
and the particle size is slightly larger.
2
.2. Activity test
The activity test was carried out in a micro tubular quartz
reactor (inner diameter is 4 mm) at atmospheric pressure. The
catalyst sample of 40 mg (40–60 mesh) was packed in the reac-
tor with two layers of quartz wool. Prior to reaction, the catalyst
3.2. Catalytic performance
Fig. 1 shows the catalytic performance of the two samples
3
◦
◦
was reduced by using flowing H2 (30 cm /min) at 500 C for
for POM at 800 C during 25 h time on stream. The conversion
◦
1
h. Then the temperature was increased to 800 C under flow-
of CH is declined with time on stream for both samples. A
4
ing argon. When the temperature was stabilized, CH4 and O2
with a ratio of 2:1 were introduced to the micro reactor by
mass flow controllers. The gases after reaction passed through
an ice trap to remove water, and was then analyzed at every
similar trend is observed for H selectivity. In contrast, both CO
2
Table 1
Chemisorption results of PtZr-C and PtZr-PC
3
0 min by a gas chromatograph (Angilent 6890N), equipped
Samples
Dispersion (%)
Average diameter (nm)
with a thermal conductivity detector and a 2-m TDX-01 packed
column. When the reaction was stopped, the temperature was
slowly decreased to room temperature under protection of flow-
PtZr-C
PtZr-PC
27.9
24.1
4.1
4.7

X. Zhu et al. / Journal of Molecular Catalysis A: Chemical 282 (2008) 67–73
69
◦
4
3
Fig. 1. Stability test of POM over PtZr-C (solid symbols) and PtZr-PC (hollow symbols). Reaction conditions: T = 800 C, P = 1 atm, GHSV = 9 × 10 cm /(g cat h),
CH4/O2 = 2:1.
selectivity and CO2 selectivity only change slightly. A higher
CO selectivity with a lower CO2 selectivity is observed for the
PtZr-PC sample. The deactivation rate based on CH4 conversion
is 0.6%/h for the PtZr-PC sample, and it is 0.9%/h for the PtZr-
C sample. Obviously, the PtZr-PC sample is more stable during
POM reaction.
3.4. IR study of CO adsorption
3.4.1. Effect of reduction temperature
Fig. 3 shows the DRIFT spectra of CO adsorbed on samples
◦
◦
(calcined at 600 C) at 25 C after the catalysts were reduced
◦
at 300 C for 1 h. The dotted line shows the spectra after CO
adsorption for 30 min, and the solid line represents the spectra
after CO desorption for 30 min. The gaseous CO band is located
3
.3. Coke analysis
−
1
at 2173 and 2115 cm . The intensity of gaseous CO bands is
essentiallyidenticalforbothPtZr-CandPtZr-PCsamples, which
make it reasonable to compare both the location and intensity of
adsorbed CO bands. All IR bands of adsorbed CO are located
at higher wavenumber in the presence of gaseous CO than the
bands in the absence of gaseous CO. This is due to the contri-
bution of dipole–dipole interaction between CO molecules at
higher CO coverage [25].
The TG analysis of the samples after 25 h reaction on stream
was performed in order to characterize the coke formation. It
can be seen from Fig. 2 that the weight loss of both samples is
very limited. Considering that the TG analysis was not carried
out in situ and the sample may adsorb some water and other
impurities, little coke was formed on both samples. The TG
analyses suggest that the coke formation is notthe main reasonof
catalyst deactivation for the present case. This result is opposite
to the case of POM over Pt/Al2O3 [17]. A large amount of coke
was formed over Pt/Al2O3, which leads to deactivation of the
catalyst [17]. The good anti-coke performance of Pt/ZrO2 is
probably related to the high mobility of lattice oxygen in ZrO2
−
1
For the PtZr-C sample, a band located at 2054 cm
is
−
1
observed, accompanied by a shoulder band at 2081 cm . A
−1
weak and broad band centered at 1813 cm is also present,
which can be assigned to two or three-fold adsorption of CO
[13,14]. The defects sites of Pt particles give a stronger back
donation, and thus the metal carbon bond is strengthened and
n+
and to the Pt-Zr sites that help the elimination of carbon [10].

7
0
X. Zhu et al. / Journal of Molecular Catalysis A: Chemical 282 (2008) 67–73
dance with Greenler et al., who assigned three bands at 2081,
−
1
2
070, and 2063 cm to CO linearly adsorbed on planes, edges,
and corners of Pt crystallites supported on silica [27].
For the PtZr-PC sample, the band of CO linearly adsorbed
−1
on close-packed planes of Pt is also present at 2081 cm
,
whereas the bands of CO linearly adsorbed to defect sites of Pt
−
1
is blue shifted to 2065 cm with respect to that for PtZr-C, and
−1
decreased in intensity to appear as a shoulder of the 2081 cm
band. This suggests that the concentration of defect sites of Pt
particles is significantly decreased. In addition, the full width
at half maximum (FWHM) of linearly adsorbed CO band for
−
1
the PtZr-C sample (68 cm ) is larger than that of the PtZr-PC
−
1
sample (53 cm ), indicating that the Pt particle of the PtZr-C
sample is smaller than that of the PtZr-PC sample [26].
Fig. 4 gives the DRIFT spectra of CO adsorbed on samples
◦ ◦
calcined at 600 C) at 25 C after the catalysts were reduced at
(
5
◦ ◦
00 C for 1 h. Compared to the spectra reduced at 300 C, the
IR intensity of adsorbed CO decreased sharply for both sam-
ples. This decrease in the IR band intensity of adsorbed CO
is not a result of the increase in the Pt particle size [28]. It has
beensuggestedthat, whenthereductiontemperatureincreasedto
Fig. 2. TGA curves of PtZr-C and PtZr-PC samples after 25 h time on stream at
◦
8
00 C of POM.
◦
the carbon oxygen bond is weakened. Consequently, the car-
bonyl vibration band of CO adsorbed on defect sites is present
higher than 300 C, the ZrO near to Pt will be partially reduced
2
to ZrO (x < 2), and migrate onto the surface of Pt particles
x
−
1
at lower wavenumber [26]. We ascribe the band at 2081 cm
to CO bonded to close-packed planes of Pt particle (such as
[28–30]. This state is the so-called strong metal-support inter-
action (SMSI) [31,32]. As a result, the surface of Pt accessible
to CO is decreased. The CO chemisorption capacity is there-
−
1
Pt(1 1 1) plane), and 2054 cm to CO bonded to defect sites of
Pt particles (such as edges and corners). This is in good accor-
◦
fore decreased. Thus, 300 C reduction reflects the true state of
Fig. 3. DRIFT spectra of CO adsorbed on PtZr-C and PtZr-PC catalysts reduced
Fig. 4. DRIFT spectra of CO adsorbed on PtZr-C and PtZr-PC catalysts reduced
◦
◦
at 300 C. Dotted line, exposed to 1.11% CO/He for 30 min; solid line, He purged
at 500 C. Dotted line, exposed to 1.11% CO/He for 30 min; solid line, He purged
◦
◦
for another 30 min. All spectra were recorded at 25 C. PtZr-C and PtZr-PC were
calcined at 600 C.
for another 30 min. All spectra were recorded at 25 C. PtZr-C and PtZr-PC were
calcined at 600 C.
◦
◦

X. Zhu et al. / Journal of Molecular Catalysis A: Chemical 282 (2008) 67–73
71
Pt. It should be noted that the peak intensity for the PtZr-C is
significantly stronger than that for PtZr-PC, suggesting a lower
coverage of ZrOx on the Pt surface for PtZr-C. The higher cov-
erage of ZrOx on Pt for PtZr-PC could fix the Pt more strongly
on ZrO2.
PtZr-C sample. Compared with PtZr-C, a significant decrease in
intensity of all adsorbed CO bands is observed for the PtZr-
−1
PC sample. Its main band is centered at 2064 cm also with a
−
1
shoulder at 2087 cm . These results indicate that, when calci-
◦
nation temperature is 400 C, the Pt particles in both samples
are rough with much higher concentration of defect sites. These
results also suggest that the Pt particle of PtZr-C is smaller than
3
.4.2. Effect of calcination temperature
The catalyst sample was sintered by calcination at different
◦
that of PtZr-PC when calcination temperature is 400 C.
◦
temperatures. Fig. 5 shows the CO adsorption on the catalyst
sample calcined at 400 C for 4 h (Fig. 5A) and 600 C for 4 h
followed by calcination at 800 C for another 4 h (Fig. 5B). A
very intense peak is located at 2070 cm along with a shoulder
For the sample calcined at 600 C for 4 h followed by calcina-
◦
◦
◦
tion at 800 C for another 4 h, all bands reduce significantly, as
◦
shown in Fig. 5B. In this case, the IR band of linearly adsorbed
CO for the PtZr-PC sample is higher in intensity than that for the
PtZr-Csample. ThisresultindicatesthatthePtparticlesizeofthe
PtZr-PC sample is smaller when the sample is further calcined
−
1
−
1
−1
at 2087 cm , as well as a very broad band at 1821 cm , for the
◦
at 800 C, which has larger surface area for CO coordination.
Fig. 6 shows the change in the IR band intensity of linearly
adsorbed CO versus calcination temperature. It can be seen that
the IR band intensity decreased sharply with calcination tem-
perature, suggesting that Pt sinters with increasing calcination
temperature. The decrease rate is higher for the PtZr-C sam-
ple. The results suggest that the plasma treated sample is more
resistanttosinteringinthepresenceofO2 athighertemperatures.
3.4.3. Temperature programmed desorption
Temperature programmed desorption of CO adsorbed on
Pt/ZrO2 catalysts was performed by monitoring IR spectra to
obtain the Pt CO bond strength. Fig. 7 gives the normalized
IR intensity of linearly adsorbed CO (COL) [33] during the
temperature programmed desorption of CO. For both sam-
ples, the normalized IR intensity of COL passes a maximum
◦
at temperature of 100–150 C, then decreases with increasing
temperature. This is due to the fact that the linearly adsorbed
CO is more stable than bridge adsorbed CO over Pt cata-
lysts, and a portion of bridge adsorbed CO changes to linearly
Fig. 5. DRIFT spectra of CO adsorbed on PtZr-C and PtZr-PC catalysts reduced
◦
at 300 C. Procedure: recorded after exposed to 1.11% CO/He for 30 min and
◦
◦
He purged for another 30 min. (A) 400 C calcinations for 4 h; (B) 600 C calci-
nations for 4 h followed by 800 C calcinations for another 4 h. All spectra were
recorded at 25 C.
◦
Fig. 6. Changes of integrated IR band intensity of linearly adsorbed CO (CO )
L
◦
◦
at 25 C versus calcination temperature of PtZr-C and PtZr-PC catalysts.

7
2
X. Zhu et al. / Journal of Molecular Catalysis A: Chemical 282 (2008) 67–73
and coalescence (PMC) [34]. It was suggested that sintering of
the supported Pt particle is limited under the H2 atmosphere
but is much more significant under the oxidizing atmosphere.
Datye et al. proposed that Pt sintering is mainly due to the PMC
pathwayundertheH2 atmosphere[34]. However, thePtsintering
pathway changes to OR in the presence of O2. It is suggested
that the Pt sintering is due to the following reaction:
Pt(s) + O2(g) = PtO2(g), ꢀH = 175 kJ/mol
The enthalpy of this reaction is much lower than the sub-
limation energy for Pt (565 kJ/mol) or the energy required for
transferringPtatomsfromaparticletothesubstrate(527 kJ/mol)
[34]. Under oxidizing atmosphere, the Pt particle easily becomes
volatile PtO2, which can be readily decomposed into Pt when the
◦
temperature is above 600 C [35]. Thus, PtO2 helps the migra-
tion of Pt atoms from small Pt particles to large Pt particles
through the OR pathway. In the present case, the Pt sintering
should be easy to occur because of the low specific surface area
of ZrO2 used, on which Pt is difficult to be stabilized.
From the overview point, POM is under reductive atmosphere
because H2 and CO are produced. However, POM over Pt/ZrO2
follows the reaction pathway of methane combustion followed
by CO2 reforming and steam reforming [9,10]. The reaction of
methane combustion is an intensely exothermic reaction under
oxidative atmosphere. Methane combustion produces “hot spot”
Fig. 7. Changes of normalized relative IR band intensity of linearly adsorbed
CO (COL) during temperature programmed desorption of CO.
adsorbed CO with increasing temperature [33]. When tem-
perature is further increased, the linearly adsorbed CO begins
desorbing.
For both samples, with increasing calcination temperature,
the temperature for complete desorption of CO decreases.
In general, the defect site concentration decreases (or close-
packed plane concentration increases) with increasing particle
size, and CO adsorbs linearly on defect sites more strongly
than on close-packed planes because of stronger back donation
in the catalyst bed [1]; i.e., the actual temperature is higher than
◦
800 C for the upper layers of the catalyst bed, which is more
favorable for Pt sintering through the OR pathway. The Pt atoms
couldmigratefromtheupperlayerstothelowerlayersofcatalyst
bed.
AttemptstoobservethesinteringofPt/ZrO2 catalyststhrough
transmission electron microscopy (TEM) failed because of the
low contrast between Pt particles and ZrO2 and the low amount
of Pt loading [28]. However, the intensity of adsorbed CO
significantly decreases with increasing calcination temperature
[
11–13,26]. Therefore, the CO desorption temperaturedecreases
with increasing calcination temperature, suggesting that the Pt
particle size becomes larger.
When the samples are calcined at the same temperature, CO
is completely desorbed at lower temperature for the PtZr-PC
(Fig. 6), which strongly supports the Pt sintering under the oxi-
◦
sample (about 50 C lower), suggesting that Pt particle of PtZr-
dizing atmosphere.
PC sample has higher concentration of the close-packed plane,
on which CO is expected to be desorbed more easily than on
defect sites of the PtZr-C sample [11–13,26]. When the calcina-
The IR band intensity of adsorbed CO for the plasma treated
sample deceases slowly with increasing calcination tempera-
ture compared to that for the untreated sample (Fig. 6). When
◦
tion temperatures are 400 and 600 C, PtZr-PC showed higher
◦
the sample was calcined at 800 C for 4 h, the IR intensity of
concentration of the close-packed plane in line with its larger
adsorbed CO is higher for the plasma treated sample (Fig. 5B),
indicating that the plasma treated sample has a larger surface for
CO coordination. Therefore, the plasma treated sample is more
resistant to sintering in the presence of O2 at higher tempera-
tures. The plasma treated sample has higher concentration of
the close-packed plane of Pt, which is more difficult to be oxi-
dized than the defect sites. Nagai et al. proposed that Pt sintering
could be inhibited through the Pt-support interaction under oxi-
dizing atmosphere [35]. Therefore, the smoothened Pt particles
of PtZr-PC probably spread over the ZrO2 with a larger Pt-ZrO2
interface, leading to a stronger Pt-support interaction. More-
◦
Pt particle size. When the samples calcined at 600 C are fur-
◦
ther calcined at 800 C, PtZr-PC adsorbs a larger amount of CO
(Fig. 5B), but CO is desorbed at lower temperature. Thus, the
lower desorption temperature cannot be related to larger Pt par-
ticle size but only to higher concentration of the close-packed
plane for PtZr-PC, i.e., the smoothened surface of Pt particle.
3
.5. Discussion
The TG analysis results show that the coke formed on both
◦
samples is very limited. Consequently, the coke formation is not
the reason responsible for catalyst deactivation during POM in
the present case. Pt sintering in the presence of O2 is the possible
reason accounting for deactivation. There are two pathways to Pt
particle sintering: Ostwald ripening (OR) and particle migration
over, when the samples were reduced at 500 C, the surface of
Pt was more covered by ZrOx as evidenced by DRIFT spectra
of adsorbed CO (Fig. 4) for PtZr-PC, resulting in a tighter Pt
fixation during the reaction. The plasma treated sample is more
stable than the untreated sample for POM, which is in line with

X. Zhu et al. / Journal of Molecular Catalysis A: Chemical 282 (2008) 67–73
73
the finding that the plasma treated sample is more resistant to
sintering in the presence of O2.
[7] L.V. Mattos, E. Rodino, D.E. Resassco, F.B. Passos, F.B. Noronha, Fuel
Process. Technol. 73 (2003) 147.
[
8] L.V. Mattos, E.R. Oliveria, P.D. Resende, F.B. Noronha, R.B. Passos, Catal.
Today 77 (2002) 245.
4
. Conclusion
[
9] F.B. Passos, E.R. Oliveira, L.V. Mattos, F.B. Noronha, Catal. Lett. 110
(
2006) 161.
10] M.M.V.M. Souza, M. Schmal, Catal. Lett. 91 (2003) 11.
[11] G. Blyholder, J. Phys. Chem. 68 (1964) 2772.
In this work, plasma treated and untreated Pt/ZrO2 samples
[
were tested in the reaction of partial oxidation of methane. TG
analysis showed little coke formation on both samples, indi-
cating the coke formation is not responsible for the catalyst
deactivation. The IR band intensity of adsorbed CO decreases
strongly with increasing calcination temperature, suggesting the
Pt particle sintering in the presence of O2. The sintering of
plasmatreatedsampleisslowerthanthatoftheuntreatedsample,
especially at higher temperature, which indicates that the plasma
treated sample is more resistant to sintering under the oxidizing
atmosphere. Under the partial oxidation atmosphere, the volatile
PtO2 is easily formed, which helps Pt atom to migrate from small
Pt particles to large Pt particles through the Ostwald ripening
pathway. The plasma treated sample is more stable for partial
oxidation of methane, as a result of its resistance to sintering in
the presence of O2.
[
12] S.G. Podkolzin, J.Y. Shen, J.J. de Pablo, J.A. Dumesic, J. Phys. Chem. B
04 (2000) 4169.
13] T. Visser, T.A. Nijhuis, A.M.J. van der Eerden, K. Jenken, Y. Ji, W. Bras,
S. Nikitenko, Y. Ikeda, M. Lepage, B.M. Weckhuysen, J. Phys. Chem. B
1
[
1
09 (2005) 3822.
[14] J.A. Anderson, F.K. Chong, C.H. Rochester, J. Mol. Catal. A 140 (1999)
5.
15] J. Llorca, N. Homs, J. Ara n˜ a, J. Sales, P.R. de la Piscina, Appl. Surf. Sci.
34 (1998) 217.
6
[
1
[
[
16] Y.H. Hu, E. Ruckenstein, Adv. Catal. 48 (2004) 297.
17] X.L. Zhu, P.P. Huo, Y.P. Zhang, C.J. Liu, Ind. Eng. Chem. Res. 45 (2006)
8604.
[
18] J.G. Wang, C.J. Liu, Y.P. Zhang, K.L. Yu, X.L. Zhu, F. He, Catal. Today
89 (2004) 183.
[
[
[
[
19] C.J. Liu, G.P. Vissokov, B.W.L. Jang, Catal. Today 72 (2002) 173.
20] J.-J. Zou, C.J. Liu, Y.P. Zhang, Langmuir 22 (2006) 2334.
21] C.K. Shi, B.W.L. Jang, Ind. Eng. Chem. Res. 45 (2006) 5879.
22] Z.H. Li, S.X. Tian, H.T. Wang, H.B. Tian, J. Mol. Catal. A 211 (2004) 149.
Acknowledgments
[23] Z.-J. Wang, Y. Zhao, L. Cui, H. Du, P. Yao, C.-J. Liu, Green Chem. 9 (2007)
54.
5
[
24] X.L. Zhu, K.L. Yu, J. Li, Y.P. Zhang, Q. Xia, C.J. Liu, React. Kinet. Catal.
Lett. 87 (2005) 93.
The supports from National Natural Science Foundation of
China (under contract 20490203) and Tianjin Municipal Sci-
ence and Technology Commission (under contract 043104111)
are greatly appreciated. The instrument supports from ABB
Switzerland and from the Program for Changjiang Scholars and
Innovative Research Team from the Ministry of Education of
China are also appreciated.
[25] P. Bazin, O. Saur, J.C. Lavalley, M. Daturi, G. Blanchard, Phys. Chem.
Chem. Phys. 7 (2005) 187.
[26] H. Song, R.M. Rioux, J.D. Hoefelmeyer, R. Komor, K. Niesz, M. Grass, P.
Yang, G.A. Somorjai, J. Am. Chem. Soc. 128 (2006) 3027.
[
27] R.G. Greenler, K.D. Burch, K. Kretzschar, R. Klauser, A.M. Bradshaw,
Surf. Sci. 152–153 (1985) 336.
28] J.H. Bitter, K. Seshan, J.A. Lercher, Top. Catal. 10 (2000) 295.
29] D. Tibiletti, F.C. Meutier, A. Goguet, D. Reid, R. Burch, M. Boaro, M.
Vicario, A. Trovarelli, J. Catal. 244 (2006) 183.
[
[
References
[
30] S. Ricote, G. Jacobs, M. Milling, Y. Ji, P.M. Patterson, B.H. Davis, Appl.
Catal. A 303 (2006) 35.
[
1] D. Dissanayake, M.P. Rosynek, K.C.C. Kharas, J.H. Lunsford, J. Catal.
32 (1991) 117.
1
[31] S.J. Tauster, S.C. Fung, R.L. Garten, J. Am. Chem. Soc. 100 (1978) 170.
[32] S. Bernal, J.J. Calvino, M.A. Cauqui, J.M. Gatica, C. Larese, J.A. P e´ rez
Omil, J.M. Pintadom, Catal. Today 50 (1999) 175.
[33] S. Zafeiratos, G. Papakonstantinou, M.M. Jacksic, S.G. Neophytides, J.
Catal. 232 (2005) 127.
[
[
2] D.H. Hickman, L.D. Schimdt, Science 259 (1993) 343.
3] S. Alberazzi, P. Arpentinier, F. Basile, P. Del Gallo, G. Fornasari, D. Gary,
A. Vaccari, Appl. Catal. A 247 (2003) 1.
4] J.H. Bitter, K. Seshan, J.A. Lercher, J. Catal. 176 (1998) 93.
5] A.M. O’Conor, Y. Schuurman, J.R.H. Ross, C. Mirodatos, Catal. Today
[
[
[34] A.K. Datye, Q. Xu, K.C. Kharas, J.M. McCarty, Catal. Today 111 (2006)
59.
1
15 (2006) 191.
[
6] F. Pompeo, N.N. Nichio, M.M.V.M. Souza, D.V. Cesar, O.A. Ferretti, M.
Schmal, Appl. Catal. A 316 (2007) 175.
[35] Y. Nagai, T. Hirabayashi, K. Dohmae, N. Takagi, T. Minami, H. Shinjoh,
S. Matsumoto, J. Catal. 242 (2006) 103.