Promotion of low-temperature oxidation of CO over Pd supported on titania-coated ceria

Article abstract of DOI:10.1039/c4ra10167g

The design of a metal oxide support for Pd was investigated to promote the oxidation of CO at lower temperatures. Through a screening study using pure metal oxides, Pd/CeO₂ showed the highest activity above 100 °C, while homemade TiO₂ was more effective below 100°C as a support for Pd. Applying the advantages of CeO₂ and TiO₂, we proposed a TiO₂-CeO₂ support having a monolayer amount of surface TiO₂ supported on CeO₂. The Pd/TiO₂-CeO₂ catalyst showed higher activity than both Pd/TiO₂ and Pd/CeO₂, which was not achieved by using the CeO₂-TiO₂ mixed-oxide. The light-off temperature over Pd/TiO₂-CeO₂ was 100°C lower than Pd/Al₂O₃. The effect of surface TiO₂ was attributed to the promotion of the reduction-oxidation cycle of supported Pd. This journal is

Full text of DOI:10.1039/c4ra10167g

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Promotion of low-temperature oxidation of CO
over Pd supported on titania-coated ceria†
Cite this: RSC Adv., 2014, 4, 54187
ab
a
a
a
a
Atsushi Satsuma,* Masatoshi Yanagihara, Kaoru Osaki, Yurina Saeki, Heng Liu,
c
c
ab
Yuta Yamamoto, Shigeo Arai and Junya Ohyama
The design of a metal oxide support for Pd was investigated to promote the oxidation of CO at lower
temperatures. Through a screening study using pure metal oxides, Pd/CeO showed the highest activity
was more effective below 100 C as a support for Pd. Applying the
, we proposed a TiO –CeO support having a monolayer amount of
supported on CeO . The Pd/TiO –CeO catalyst showed higher activity than both Pd/TiO
2
ꢀ
ꢀ
above 100 C, while homemade TiO
advantages of CeO and TiO
surface TiO
and Pd/CeO
over Pd/TiO
2
2
2
2
2
Received 10th September 2014
Accepted 17th October 2014
2
2
2
2
2
2
, which was not achieved by using the CeO
2
–TiO
2
mixed-oxide. The light-off temperature
was attributed to the
DOI: 10.1039/c4ra10167g
ꢀ
2
–CeO
2
was 100 C lower than Pd/Al
2
O
3
. The effect of surface TiO
2
www.rsc.org/advances
promotion of the reduction–oxidation cycle of supported Pd.
loaded perovskite on which supported Pd retained its high
1
. Introduction
n+
metal dispersion by reversible incorporation of Pd ions under
0
In order to reduce carbon dioxide (CO
2
) emissions from auto- oxidative atmosphere and formation of Pd particles under
2
mobiles, better fuel economy systems, such as hybrid and start– reductive atmosphere. The similar effect on the retention of
stop systems, are becoming popular. The exhaust gas temper- metal dispersion is observed when CeO
2 2
and CeO -based
ature of these systems is lower than ever because of better supports are used as supports by forming solid solutions or the
3,4
energy efficiency. Hence, the enhancement of low-temperature formation of metal–O–Ce bonds.
activity is strongly required for automobile catalysts, especially
The catalytic activity of supported Pd is sometimes more
for combustion of CO and unburned hydrocarbons as the main affected by the oxidation state of Pd. For example, we have
components in low-temperature exhaust fumes.
reported very unique dependence of the oxidation activity of Pd/
Among platinum group metals (Pt, Pd, and Rh) as active CeO on surface area of the CeO support, i.e., the lower surface
elements in a three-way catalyst, a supported Pd catalyst is one area of CeO was more preferable for the propene oxidation
2
2
2
5
of the essential components playing important role in the depending on the reaction and pre-treatment conditions. In
ꢀ
combustion of CO and unburned hydrocarbons. For the this case, the propane oxidation at 200 C proceeds preferably
promotion of low-temperature activity of a supported Pd cata- on metallic Pd species which is more stable in larger Pd parti-
lyst, we can nd various strategies for catalyst design in the cles on CeO
2
having low surface area. The metallic Pd species
literature. Generally, the catalytic activity is correlated to the plays an important role on the oxidation reaction at lower
6,7
dispersion of supported Pd. For example, Baylet et al. reported temperatures, such as CO oxidation. On the other hand,
that both higher surface area of supports and stability of Pd oxidized Pd species shows higher activity for the methane
8
–12
particles are required to achieve higher Pd dispersion on hex- oxidation,
which is rationalized by the role of dispersed Pd
1
13–15
aaluminate. The higher dispersion of Pd can be obtained by oxide species as an oxygen provider.
The oxidation state of
metal–support interaction under cycles between oxidative and supported Pd is strongly affected by the acid–base property of
3,10
reductive atmospheres encountered in an exhaust gas. Nishi- supports.
It is also known that incorporated Pd species in
hata et al. developed so called “intelligent catalyst” based on Pd- metal oxide matrix shows activity for the oxidation reactions.
Hedge and co-workers proposed “noble metal ionic catalysts” in
the formulae of Ce1 Pd O
and Ce
Ti Pd O
(x ¼ 0.01–
ꢁx
x
2ꢁd
1ꢁxꢁy
y
x
2ꢁd
a
Graduate School of Engineering, Nagoya University, Nagoya 464-8603, Japan. E-mail:
0.02, d z x, y ¼ 0.15–0.25) and Ti Pd O
(x ¼ 0.01–0.03),
1ꢁx
x
2ꢁx
satsuma@apchem.nagoya-u.ac.jp; Fax: +81-52-789-3193; Tel: +81-52-789-4608
b
ꢀ
16–21
which showed very high CO oxidation activity below 100 C.
Kurnatowska and co-workers also reported that nanocrystalline
Ce1ꢁxPd 2ꢁy exhibited a good performance on the low-
temperature CO oxidation. In these catalysts, the low-
support, and the temperature activity is assigned to the formation of Pd–O–Ce
Unit of Elements Strategy Initiative for Catalysts & Batteries, Kyoto University, Kyoto
6
c
15-8530, Japan
x
O
Ecotopia Science Institute, Nagoya University, Nagoya, 464-8603, Japan
Electronic supplementary information (ESI) available: XRX patterns of Pd
supported catalysts, TEM/EDS micrographs of TiO –CeO
estimation of oxygen storage-release rate from TG-DTA. See DOI:
0.1039/c4ra10167g
2
2
†
2
2
bond as a redox site and activation of lattice oxygen. Actually,
1
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 54187–54193 | 54187

View Article Online
RSC Advances
Paper
Colussi et al. clearly demonstrated by means of HRTEM and 3 h. TiO
2
–CeO
2
supports were prepared by a precipitation
DFT calculations that Pd–O sites in Pd–Ce surface enhanced the method as follows. CeO powder and distilled water were put in
2
23
catalytic activity for methane combustion.
As for the role of supports, oxygen storage and release water bath, a prescribed amount of Ti source (TiCl4 or
capacity (abbreviated as OSC, hereaer) strongly controls the Ti(OC ) was added to the solution and stirred for 30 min.
a beaker. Aer the temperature regulation of the beaker in a
3
7 4
H )
catalytic activity especially at lower temperatures. Liu et al. Then an aqueous solution of ammonia was added to adjust pH
demonstrated an effective contribution of iron oxide support to of the solution around pH ¼ 8–9. The precipitates were ltrated
24
low-temperature CO oxidation. They reported that CO oxida- and washed with distilled water, and then dried and calcined at
ꢀ
tion over Pt/FeO
x
and Pd/FeO
x
proceeds over two adjacent but 500 C for 3 h in air. The detailed preparation conditions are
different active sites (Pt, Pd for CO and FeO for oxygen) with described in the Section 3.2.
x
ꢁ1
low apparent activation energies (30–34 kJ mol ). Ikeue et al.,
reported that the use of partially Fe-substituted 10Al O $2B O
as a support of Pd gave an effective catalyst for the catalytic NO with a 10 mg of catalyst at a total ow rate of 100 N cm min
The catalytic activity was evaluated by CO oxidation using a
conventional xed-bed ow reactor at atmospheric pressure
2
3
2 3
3
ꢁ1
ꢁ
1
7
conversion in a fuel rich region because of improved oxygen which corresponds to GHSV ¼ 480 000 h . A catalyst was at
2
5
ꢀ
2 2
-based supports, it is well rst reduced in a ow of 3% H /He at 400 C for 10 min, and the
storage capacity. In the case of CeO
known that partial incorporation of Zr and other elements in catalytic run was carried out under a ow of 0.45% CO/10% O
2
/
26–31
CeO
2
lattice signicantly improves OSC.
Several papers
2
N . The effluent gas was analyzed by nondispersive infrared
demonstrated suitable designs of CeO -based metal oxide (NDIR) CO/CO analyzer (Horiba VIA510).
2
2
3
2–34
supports for promotion of CO oxidation activity using CeO ,
Temperature programmed reduction by CO (CO-TPR) was
2
35–37
38–44
CeO /Co O ,
and CeO –TiO .
There are two major effects carried out using a conventional xed-bed ow reactor as
2
3
4
2
2
of CeO
2
-based supports on the low temperature oxidation: one follows. The sample was oxidized in a ow of 20% O /N (100 N
2
2
3
ꢁ1
ꢀ
is the promotion of redox cycles of Pd particles, and another is cm min ) at 400 C for 30 min. Aer cooling the sample in a
supply of lattice oxygen. As for the latter effect, the strong ow of pure N to room temperature, the CO-TPR measurement
2
adsorption of CO on Pd surface strongly suppresses O
adsorption and activation on Pd particles in Pd/Al
7
ꢀ
ꢁ1
2
was carried out in a ow of 0.4% CO/N
2
at a rate of 5 C min
ꢀ
2
O
3
at lower up to 400 C. The effluent gas was analyzed by nondispersive
temperatures, while CeO -based supports played a role of infrared (NDIR) CO/CO analyzer (Horiba VIA510).
2
2
oxygen supply from lattice even when the Pd surface is fully
covered by strongly adsorbed CO.
On the basis of above mentioned activity-controlling factors ow of O
for supported Pd catalysts, the present study aimed a design of min, and then reduced in a ow of H
The dispersion of Pd was estimated by the CO-pulse
7
46–48
adsorption method.
A catalyst was at rst treated in a
ꢀ
2
at 400 C for 15 min followed by purge with He for 15
ꢀ
2
at 400 C for 15 min
novel supported Pd catalysts for low-temperature oxidation of followed by purge with He for 15 min. Then the catalyst was
CO. We focused TiO having a limited promotion effect only at cooled in a ow of He, and a series of CO pulses were injected
lower temperatures. Although mixed oxide of CeO –TiO has with an interval of 2 to 3 min until the amount of slipped CO
this study proposed hierarchical pulses reaches a steady state value. In order to avoid oxidation
conguration of TiO2 and CeO2 for the better CO oxidation of CO to CO followed by adsorption on CeO as carbonate, the
2
2
2
38–44
been already proposed,
2
2
activity at lower temperatures.
sample cell was soaked in a dry ice/ethanol bath and the
ꢀ
adsorption of CO was carried out at ca. ꢁ70 C. The BET
2
specic surface area was measured by N adsorption at liquid
2. Experimental section
N
2
temperature by using a conventional ow-type adsorption
Supported Pd catalysts with Pd content of 1 wt% were prepared apparatus.
by a conventional impregnation method using an aqueous The oxygen storage-release property was evaluated by
palladium nitrate, followed by dryness and calcination in air at weight deviation under the following O –H periodic opera-
tion at 300 C using TG-DTA (Shimadzu DTG-60H). In order to
JRC-ZRO-1) were supplied from the committee of reference suppress the reduction–oxidation of supported Pd, the
2
2
ꢀ
2 2 2
00 C for 3 h. CeO (JRC-CEO-3), TiO (B) (JRC-TIO-8), ZrO
ꢀ
5
(
catalyst, Catalysis society of Japan. SiO
2
(Fuji Silysia, Q10) and oxidation step was carried out under low oxygen concentration
ꢀ
Fe O (Kishida Chemical, 98%) were purchased. Al O was (3%). Aer pretreatment in a ow of 3% O
2
/N
2
at 500 C for 30
2
3
2 3
ꢀ
ꢀ
obtained by calcination of boehmite at 900 C in air for 3 h. min, the oxidized sample was cooled down to 300 C in 3% O
2
/
TiO
2
(A) was prepared by hydrolysis of Ti(OC
3
H
7
)
4
(Kishida
N
2
. Aer purging with pure N
2
for 30 s, a ow gas was switched
until the catalyst weight loss reaches to a steady
, oxidation in 3% O /N
Chemical, 99%) by an aqueous solution of ammonia followed to 3% H
by dryness and calcination in air at 500 C for 3 h. CeO
2
/N
2
ꢀ
–TiO
2
state. Then the purging in pure N
2
2
2
2
(
Ce/Ti ¼ 0.2) and TiO
2
–ZrO
using TiCl
2
(Ti/Zr ¼ 5/5) were prepared by co- until the catalyst weight recovered, the purging in pure N
,
2
37–45
precipitation
Ce(NO
4
(Kishida Chemical, 24%), and the reduction in 3% H
2
/N
2
were repeated for several times
ꢀ
)
4
(Kishida Chemical, 98%) and ZrO(NO
3
)
2
$2H
2
O
at 300 C.
3
(Kishida Chemical, 99%) as precursors. The mixed aqueous
Pd K-edge XANES spectra were measured on BL28B2 of
solutions were neutralized with an aqueous solution of SPring-8 (Hyogo, Japan) operated at 8 GeV. The analyses of X-ray
ammonia (Kishida Chemical, 28%), and the obtained Absorption Near Edge Structure (XANES) were performed using
ꢀ
precipitates were dried and then calcined in air at 500 C for the REX version 2.5 program (RIGAKU).
54188 | RSC Adv., 2014, 4, 54187–54193
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
3. Results and discussion
3.1. Screening study
At rst, a screening study of the effect of support materials on
the oxidation of CO was carried out as summarized in Table 1.
The temperatures at which the CO conversion reached to 20%
and 80% (T20 and T80) were used as indexes of light-off and
nearly complete oxidation, respectively. It should be noted that
the results were reproducible as far as the activity tests were
ꢀ
carried out below 500 C. The values of T20 for Pd catalysts
supported on TiO
below 100 C, while those were higher than 130 C when Al
2
, CeO
2
, and CeO
2
-based mixed oxides were
ꢀ
ꢀ
2
O
3
,
SiO
previous paper, the reason for the high activity of Pd/CeO
Pd/TiO can be assigned to the promotion of reduction–oxida-
2
, ZrO
2
, and Fe
2
O
3
were used as supports. As discussed in the
7
2
and
2
tion cycles of supported Pd particles and the oxygen supply from
the supports. It should be noted that the order of the activity
was not dependent on the surface area or Pd dispersion.
Although Pd dispersion of Pd/CeO
2
is less than a half of that of
Fig. 1 Conversion of CO over 1 wt% Pd supported catalysts.
ꢀ
Pd/Al
2
O
3
, T20 of the former was 71 C lower than that of the
ꢀ
latter. The lowest T20 (44 C) was obtained when homemade
2 2
TiO (A) was used as a support. Interestingly, Pd/TiO (B) avail-
able in a market did not show such high activity at lower
2 3
Pd/Al O was also plotted as a model of practical catalyst.
Although the conversion of CO on Pd/TiO
2
is higher than Pd/
(A)
temperatures. The higher Pd dispersion on Pd/TiO (A) than Pd/
2
ꢀ
CeO
2
below 110 C, the increase in CO conversion on Pd/TiO
2
TiO (B) is one of the possible reasons for the lowest T .
2
20
ꢀ
slowed down above 100 C and became the same as Pd/Al
2
O
3
However, T80 of both the Pd/TiO was almost the same. The loss
2
ꢀ
above 180 C. Assuming that the mixing of CeO
2
and TiO
2
may
of the advantage of Pd/TiO
caused by sintering of Pd particles on Pd/TiO
2
(A) at higher temperature is not
(A) because the
result in a better performance, two types of supports were
2
38–44
examined, i.e., CeO –TiO mixed oxide
and TiO –CeO2
2
2
2
activity patterns of all the catalysts were reproducible. The result
supported oxide. The CO oxidation activity of Pd/CeO –TiO
2
2
suggests that the promotion effect is caused by an interaction
ꢀ
(Ce/Ti ¼ 0.2) was comparable to Pd/TiO
2
ꢀ
below 80 C, however,
ꢀ
between Pd and TiO
2
(A), which is limited below 100 C. As
that was lower than Pd/CeO
2
above 110 C. On the other hand,
shown in Fig. 1S,† both the TiO supports showed the diffrac-
2
aer the optimization of chemical composition and preparation
conditions, Pd/TiO –CeO (TiO
¼ 5.4 wt%) showed the best
activity from the light-off to the 100% conversion. The surface
tion lines of anatase phase, while the intensity of the diffraction
lines of TiO (A) were signicantly weaker than those of TiO (B).
2
2
2
2
2
The result suggests major amorphous phase in TiO
2
(A) may
2
ꢁ1
area and Pd dispersion of Pd/TiO –CeO were 64 m g and
2
2
cause the promotive effect of CO oxidation at lower tempera-
tures. In order to obtain more active catalyst, a design of a
support is investigated using TiO and CeO in the next section.
2 2
3
5%, respectively. Since the Pd dispersions of Pd/TiO –CeO2
2
and Pd/CeO –TiO are the same, the better activity of Pd/TiO –
2
2
2
CeO
2
than Pd/CeO
2
–TiO
2
is caused by another factor, as dis-
cussed in the Section 3.3. Interestingly, the CO conversion over
Pd/TiO –CeO –ZrO (TiO
¼ 5.4 wt%, Ce/Zr ¼ 5/5) did not
Fig. 1 shows the activity patterns of Pd catalysts supported on exceed that of Pd/CeO (gure not shown). It should be noted
2 2
3.2. Design of TiO –CeO support
2
2
2
2
2
2 2
2 2
CeO , TiO (A), and their mixed oxides. The activity pattern of that Pd/TiO –CeO showed the CO oxidation activity around
ꢀ
00 C lower than Pd/Al O which is a typical oxidation catalyst.
2 3
1
The optimization of the chemical compositions and prepara-
tion conditions is reported below.
Table 1 BET surface area (SBET), dispersion of Pd (DPd), and temper-
atures at which CO conversions were 20% and 80% (T20 and T80) for 1
wt% Pd supported catalysts
Fig. 2 shows the effect of TiO
activity of Pd/TiO –CeO . The catalytic activity increased with
the increase in the TiO content up to 5.4 wt%, and further
addition of TiO suppressed the catalytic activity. It is inter-
2
content on the CO oxidation
2
2
2
ꢁ1
ꢀ
ꢀ
T80/ C
Catalysts
Pd/TiO (A)
S
BET/m g
D
Pd/%
T
20/ C
2
2
2
56
84
36
69
32
12
23
35
71
16
10
44
59
147
122
168
151
138
176
<200
195
200
esting that the maximum activity was obtained when the theo-
Pd/CeO
Pd/CeO
2
2
–ZrO
2
71
retical amount of monolayer TiO (5.4 wt%) was supported on
2
Pd/TiO
Pd/CeO
2
(B)
–TiO
130
200
242
414
11
84
88
130
149
155
183
CeO2.
2
2
Fig. 3 shows the effect of TiO precursors and precipitation
2
Pd/Al
Pd/SiO
Pd/ZrO
Pd/Fe
2
O
3
temperature on the CO oxidation activity of Pd/TiO
2
–CeO
2
.
2
ꢀ
Using TiCl
4
as a precursor, the precipitation at 4 C (water/ice
2
2
O
3
7.0
0.1
bath) resulted in the better catalytic performance than that
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 54187–54193 | 54189

View Article Online
RSC Advances
Paper
2
on the CO oxidation suggests contacts between TiO and Pd
particles. The promotive role of surface TiO is investigated in
2
the next section.
3
.3. Promotion effect of surface TiO on CeO
2 2
As discussed in the previous paper, the catalytic activity of
supported Pd catalysts for the low temperature oxidation of CO
can be rationalized by the following factors: (1) reducibility of
supported Pd species, (2) oxygen storage-release property of the
7
supports, and (3) dispersion of Pd. The factor (1) is essential for
the activation of CO because coordinatively unsaturated surface
sites of Pd metal particles are necessary for CO adsorption. The
factor (2) is essential for the activation of O
2
. Since the Pd
–TiO were
dispersions of Pd/TiO –CeO , Pd/CeO , and Pd/CeO
2
2
2
2
2
almost the same and lower than Pd/TiO , the factor (3) can be
2
neglected as the reason for the best activity of Pd/TiO –CeO .
2
2
Therefore, the factors (1) and (2) of Pd/TiO –CeO are compared
2
2
2 2
with Pd/CeO and Pd/TiO .
2 2 2
Fig. 2 Conversion of CO over Pd/TiO –CeO having various TiO
contents.
The oxygen storage-release property was evaluated by weight
ꢀ
deviation during the O
2
–H
2
periodic operation at 300 C, and
the results of Pd/TiO , Pd/CeO
2
2
, and Pd/TiO
2
–CeO
2
are shown in
Fig. 4. It is clear that the maximum weight decrease is strongly
dependent on the supports in the following order: Pd/TiO –
2
ꢁ
1
ꢁ1
CeO (0.19 mmol-O g ) $ Pd/CeO (0.14 mmol-O g ) [ Pd/
2
2
2
2
ꢁ
1
TiO2 (0.07 mmol-O2
g
). The weight deviation of Pd/CeO2
to CeO2ꢁd (d ¼
.048), which is close to the maximum oxygen storage capacity
corresponds to the reduction of pure CeO
2
0
27,29
for pure CeO
2
(d ¼ 0.05).
was only a half of that from Pd/CeO
2
On the other hand, the released
. The
modi-
oxygen from Pd/TiO
oxygen storage capacity of CeO
2
2
was retained aer TiO
2
ꢀ

cation. The trend was the same at 200 C: the maximum
weight decreases were Pd/TiO –CeO (0.17 mmol-O g ) $ Pd/
CeO (0.14 mmol-O g ) [ Pd/TiO (undetectable). The rates
for oxygen release and storage at 300 C was estimated from the
ꢁ
1
2
2
2
ꢁ
1
2
2
2
ꢀ
initial slope of the weight deviation, as shown in Fig. 3S.† The
ꢁ
6
ꢁ1 ꢁ1
estimated release rates were 0.54 ꢂ 10 mol-O
2
g
s
for Pd/
2 2 2
Fig. 3 Conversion of CO over Pd/TiO –CeO catalysts. The TiO –
CeO supports were prepared using (a–c) TiCl
2
4
, and (d) Ti(OC
3
H
7 4
) as
ꢀ
precursors and precipitated at (a) room temperature, (b) 4 C, (c)
ꢀ ꢀ
20 C, and (d) 4 C.
ꢁ
precipitated at room temperature. The further decrease in the
ꢀ
temperature to ꢁ20 C (NaCl/water/ice bath) was not effective.
3 7 4
The use of Ti(OC H ) as a precursor was less effective than that
of TiCl . The best catalyst performance was obtained when
4
ꢀ
5
.4 wt% TiO was precipitated to CeO at 4 C using TiCl as a
2
2
4
precursor. The EDX images of the optimized support indicated
uniform dispersion of TiO on CeO support, as shown in
Fig. 2S.† In the XRD pattern, there was no diffraction line
assignable to TiO , indicating TiO is in amorphous phase, very
small particle, or thin layer. The promotion effect of TiO layer
2
2
2
2
Fig. 4 Weight deviation of Pd/TiO
during O –H
2
, Pd/CeO
periodic operation at 300 C.
2
, and Pd/TiO
2
–CeO
2
2
ꢀ
2
2
54190 | RSC Adv., 2014, 4, 54187–54193
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
ꢁ
6
ꢁ1 ꢁ1
TiO
2
(A), and 1.8 ꢂ 10 mol-O
2
g
s
for Pd/CeO
2
and Pd/ respectively) was higher than the stoichiometry of the PdO
TiO –CeO . Pd/CeO and Pd/TiO –CeO showed 3.3 times faster reduction to metallic Pd. The excess formation of CO is caused
rate than Pd/TiO (A). The oxygen storage rates of these three by the oxygen release from the supports. The reduction peak
2
2
2
2
2
2
7
2
ꢁ
6
ꢁ1 ꢁ1
ꢀ
catalysts are the same (0.94 ꢂ 10 mol-O g
s
). The results above 150 C is attributed to the contribution of supports or
2
ꢀ
suggest the oxygen storage-release capacity and rate at 300 C of disproportionation of CO. The maximum reduction tempera-
ꢀ
Pd/TiO
contribution of OSC on higher CO conversion level. The lower Pd/CeO
activity of Pd/TiO
2
–CeO
2
were comparable to Pd/CeO
2
, indicating the ture of the rst peak was in the order of Pd/TiO
2
–CeO
2
(78 C) <
ꢀ
ꢀ
ꢀ
2
(96 C) < Pd/TiO
2
(113 C) ꢃ Pd/Al
2 3
O (165 C). This
2
(A) at higher temperatures can be rationalized order was not well in harmony with that of CO oxidation activity,
by the lower oxygen storage capacity and release rate. Actually, because the reduction temperature of Pd/TiO
2
was higher than
ꢀ
above 180 C, Pd/TiO (A) showed comparable activity to Pd/ Pd/CeO and the positions of the rst peak of Pd/TiO –CeO and
2
2
2
2
Al O having no oxygen storage capacity.
Pd/CeO2 are too close on both temperature and quantity to
2
3
ꢀ
As discussed above, the contribution of OSC on CO oxidation attribute the big difference in the CO conversion at 60 C for
ꢀ
above 100 C can be rationalized, but the contribution of OSC these catalysts (21% for Pd/CeO
2
and 48% for Pd/TiO
2
–CeO
2
,
on the light-off activity cannot be discussed based on TG-DTA respectively). From the results of CO-TPR, we cannot conclude
results because the weight deviation during the H
odic operation was negligible below 100 C. Mukri and co- supported Pd species from PdO
2
–O
2
peri- that the catalytic activity is controlled by the reducibility of
to Pd or OSC. Then, to inves-
ꢀ
x
workers reported that Ti1ꢁxPd
of TiO2 by Pd ion showed high catalytic activity for CO PdO ), in situ XANES spectra are compared.
oxidation because of activated lattice oxygen. In their case, the
formation Pd–O–Ti interface in lattice improved the catalytic catalysts under the CO oxidation conditions below 100 C. The
x
O
2ꢁx prepared by substitution tigate the oxidation step of supported Pd particles (from Pd to
2
+
x
21
Fig. 6 shows in situ Pd K-edge XANES spectra of supported Pd
ꢀ
ꢁ1
activity due to high oxygen storage capacity up to 5.1 mmol g
which was more than one order of magnitude higher than those Under the CO oxidation, the spectra of supported Pd catalysts
for the catalysts in this study. Thus, the promotion effect of the are similar to that aer the reduction in H , indicating the
TiO layer cannot be attributed only to OSC enhancement. The supported Pd species is basically in metallic state. However, the
different type of the effect of TiO layer on Pd species can be shi of the absorption edge to the higher energy suggests
,
feed gas compositions are the same as in the catalytic tests.
2
2
2
expected. Therefore, reduction–oxidation property of the cata- certain contribution of oxidized Pd species. The oxidation state
ꢀ
lysts below 100 C was evaluated by CO-Temperature Pro- of Pd was evaluated in the following manner. Because of the
grammed Reduction and in situ XANES.
interaction between Pd species and supports, the spectra were
Fig. 5 shows CO-Temperature Programmed Reduction (CO- not reproduced by the combination of the spectra of pure PdO
TPR) proles of selected Pd supported catalysts. Before the disk and Pd foil. Therefore, we used the spectra of catalysts aer
TPR measurement, the supported Pd was oxidized to PdO, and calcination as a reference of oxidized state (Ox. in the gures)
2
the TPR measurement was carried out in 0.4% CO/N without
ꢀ
2 2 3
O . Pd/Al O showed a reduction peak around 165 C. The
ꢁ1
amount of evolved CO (0.11 mmol g ) in the rst peak was
2
almost equivalent to the stoichiometry of PdO reduction to Pd
ꢁ
1
(
0.094 mmol g ). The other catalysts showed two reduction
ꢀ
ꢀ
peaks below 150 C and above 150 C. As for the rst peak, the
amount of evolved CO
ꢁ
1
2
(0.13 mmol g for Pd/TiO
2
, 0.37 mmol
2
–CeO ,
ꢁ1
ꢁ1
g
for Pd/CeO
2
, and 0.40 mmol g
for Pd/TiO
2
Fig. 6 Pd K-edge XANES spectra of supported Pd catalysts (Ox.) after
ꢀ
ꢀ
calcination at 500 C in air, (Red.) after reduction in H
2
profile during CO-Temperature Programmed min, and in situ Pd K-edge XANES spectra under the CO oxidation at
2
at 400 C for 30
Fig. 5 Effluent CO
Reduction of supported Pd catalysts.
various temperatures.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 54187–54193 | 54191

View Article Online
RSC Advances
Paper
ꢀ
and aer reduction in H
2
at 400 C as a reference of reduced equilibrium constant (K ¼ k
/k
O R
) under the reaction conditions.
state (Red. in the gures). The measured in situ XANES spectra The contribution of PdO under the reaction conditions should
x
were well reproduced by the combination of these reference indicate the relative difference of K in each sample, and Pd
spectra. Table 2 shows the evaluated contribution of oxidized Pd species on Pd/TiO –CeO is more easily oxidized than that on
2
2
under the CO oxidation. The oxidation state of Pd was strongly Pd/CeO
affected by the supports. The contribution of oxidized Pd was promotive effect of TiO
2
at lower temperatures. The results indicate that the
can be attributed to the promotion of
which was not active for the CO the regeneration step of oxidized PdO species. The promotive
2
less than 10% on Pd/Al
2
O
3
x
ꢀ
oxidation below 100 C. On the other hand, more than 10% of effect was limited at lower temperatures, i.e., contributions of
Pd is oxidized on TiO
2
–CeO
2
and TiO
2
supports even at room oxidized Pd were much higher on Pd/TiO
2
–CeO
2
and Pd/TiO
2
,
ꢀ
temperature. With the increase in the reaction temperature, the while at 100 C that of Pd/CeO became almost the same as Pd/
2
oxidation number of Pd increased for all the catalysts. Although TiO –CeO but Pd/TiO showed lower value. The variation in the
2
2
2
the contribution of oxidized Pd in Pd/CeO was lower than that PdO ratio as a function of reaction temperature is approxi-
2
x
ꢀ
of Pd/TiO
C. As for Pd/TiO
2
–CeO
2
below 50 C, they were almost the same at 100 mately in harmony with the activity pattern for CO oxidation
, the contribution of oxidized Pd was shown in Fig. 1. Therefore, it is reasonable to attribute the
–CeO at room temperature, while promotive effect of TiO modication to the promotion of re-
ꢀ
2
comparable to that of Pd/TiO
2
2
2
that was not sensitive to reaction temperature. The trend in the oxidation step of Pd species under the CO oxidation.
oxidation state of Pd is in harmony with the CO oxidation
activity. The results in the table suggest contribution of the
oxidation step of Pd under the CO oxidation as an activity-
controlling factor.
One may wonder above discussion: although the reducibility
of Pd is one of the activity-controlling factors, the promotion
4
. Conclusions
The effect of metal oxide support was investigated to promote
the CO oxidation over supported Pd catalysts at lower temper-
atures. The use of TiO support enhanced the CO oxidation
below 100 C, and that of CeO2 support was effective above
2
effect of TiO
2
modication was attributed to the rate of oxida-
ꢀ
tion of Pd. The results can be rationalized by considering the
reduction–oxidation cycle of Pd species during the CO oxida-
tion. The oxidation of CO over Pd catalyst can be generally
depicted in the following scheme.
ꢀ
100 C. Applying the advantages of CeO
2
and TiO
2
, Pd/TiO
2
–CeO
2
showed the best catalytic activity from the light-off to the 100%
conversion, which was around 100 C lower than Pd/Al
the optimization of the chemical compositions and preparation
conditions, the best catalyst performance was obtained when
ꢀ
2
O
3
. Aer
theoretical monolayer amount (5.4 wt%) of TiO was precipi-
2
ꢀ
tated to CeO2 at 4 C using TiCl4 as a precursor. By using
TG-DTA, CO-TPR, and in situ XANES, the promotion of the
oxidation step of Pd species during the CO oxidation was
revealed to be an activity-controlling factor.
In lean conditions (oxygen-rich atmospheres), Pd species are
partially oxidized by gaseous oxygen or oxygen from supports.
Acknowledgements
x
to
The oxidation of CO proceeds with the reduction of PdO
metallic Pd. The rates for both Pd oxidation (k ) and PdO
reduction (k ) determine the activity of gas phase CO oxidation
). Comparing Pd/TiO –CeO and Pd/CeO , the rates of PdO
x
O
x
This work was partly supported by JSPS KAKENHI Grant from
the Ministry of Education, Culture, Sports, Science and Tech-
nology, Japan. A part of this work was performed under a
management of “Elements Strategy Initiative for Catalysts &
Batteries (ESICB)” supported by MEXT program “Elements
Strategy Initiative to Form Core Research Centre” (since 2012),
MEXT; Ministry of Education Culture, Sports, Science and
Technology, Japan. The XAFS measurements at SPring-8 were
carried out by the approval (proposal no. 2012B1569) of the
Japan Synchrotron Radiation Research Institute (JASRI).
R
(k
G
2
2
2
reduction to Pd (k ) by CO are not much different because the
R
reduction peaks in CO-TPR were observed at the same
temperature range. It suggests that the rate determining step is
the regeneration of PdO species (k ). Although the rate of PdO
x O x
regeneration cannot be evaluated, the oxidation state under the
reaction conditions estimated by XANES spectra indicates the
Table 2 Contribution of oxidized Pd in Pd particles (% in total Pd)
under the CO oxidation at various temperatures evaluated from in situ
XANES spectra
Notes and references
1
A. Baylet, S. Royer, P. Marecot, J. M. Tatibouet and D. Duprez,
Appl. Catal., B, 2008, 77, 237–247.
ꢀ
ꢀ
Catalyst
Room temp.
50 C
100 C
2
Y. Nishihata, J. Mizuki, T. Akao, H. Tanaka, M. Uenishi,
M. Kimura, T. Okamoto and N. Hamada, Nature, 2002, 418,
Pd/TiO
Pd/CeO
Pd/TiO
Pd/Al
2
–CeO
2
16.2
7.4
10.7
4.0
18.9
13.7
13.1
4.7
23.6
23.7
16.9
6.8
2
164–167.
2
(A)
3
Y. Nagai, T. Hirabayashi, K. Dohmae, N. Takagi, T. Minami,
2
O
3
H. Shinjoh and S. Matsumoto, J. Catal., 2006, 242, 103–109.
54192 | RSC Adv., 2014, 4, 54187–54193
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
4
R. V. Gulyaev, A. I. Stadnichenko, E. M. Slavinskaya, 25 K. Ikeue, K. Watanabe, T. Minekishi, A. Imamura, T. Sato,
A. S. Ivanova, S. V. Koscheev and A. I. Boronin, Appl. Catal.,
A, 2012, 439–440, 41–50.
Y. Nagao, Y. Nakahara and M. Machida, Appl. Catal., B,
2014, 146, 50–56.
5
6
7
8
9
A. Satsuma, R. Sato, K. Osaki and K. Shimizu, Catal. Today, 26 H. C. Yao and F. Y. Yao, J. Catal., 1984, 86, 254–265.
2
012, 185, 61–65.
27 M. Ozawa, M. Kimura and A. Isogai, J. Alloys Compd., 1993,
193, 73–75.
28 M. Sugiura, M. Ozawa, A. Suda, T. Suzuki and T. Kanazawa,
Bull. Chem. Soc. Jpn., 2005, 78, 752–767.
29 G. Dutta, U. Vagmare, T. Baidya, M. S. Hegde, K. R. Priolkar
and P. R. Sarode, Catal. Lett., 2006, 108, 165–172.
30 G. Dutta, U. Vaghmare, T. Baidya, M. S. Hegde, K. R. Priolkar
and P. R. Sarode, Chem. Mater., 2006, 18, 3249–3256.
31 Q. Ding, S. Yin, C. Guo and T. Sato, Chem. Lett., 2012, 41,
1250–1252.
K. B. Kim, M. K. Kim, Y. H. Kim, K. S. Song and E. D. Park,
Res. Chem. Intermed., 2010, 36, 603–611.
A. Satsuma, K. Osaki, M. Yanagihara, J. Ohyama and
K. Shimizu, Appl. Catal., B, 2013, 132–133, 511–518.
R. F. Hicks, H. Qi, M. L. Young and R. G. Lee, J. Catal., 1990,
1
22, 295–306.
R. Burch, P. K. Loader and F. J. Urbano, Catal. Today, 1996,
7, 243–248.
2
1
1
1
1
0 H. Yoshida, T. Nakajima, Y. Yazawa and T. Hattori, Appl.
Catal., B, 2007, 71, 70–79.
1 P. Castellazzi, G. Groppi, P. Forzatti, E. Finocchio and
G. Busca, J. Catal., 2010, 275, 218–227.
2 S. Colussi, A. Trovarelli, E. Vesselli, A. Baraldi, G. Comelli, 34 H. Cordatos and R. J. Gorte, J. Catal., 1996, 159, 112–118.
G. Groppi and J. Llorca, Appl. Catal., A, 2010, 390, 1–10.
32 T. Jin, T. Okuhara, G. J. Mains and J. M. White, J. Phys.
Chem., 1987, 91, 3310–3315.
33 G. S. Zaris and R. J. Gorte, J. Catal., 1993, 139, 561–567.
35 J.-Y. Luo, M. Meng, X. Li, X.-G. Li, Y.-Q. Zha, T.-D. Hu,
3 T. Schalow, B. Brandt, M. Laurin, S. Schauermann,
Y.-N. Xie and J. Zhang, J. Catal., 2008, 254, 310–324.
S. Guimond, H. Kuhlenbeck, J. Libuda and H.-J. Freund, 36 A. T ¨o mcrona, M. Skoglundh, P. Thorm ¨a hlen, E. Fridell and
Surf. Sci., 2006, 600, 2528–2542. E. Jobson, Appl. Catal., B, 1997, 14, 131–146.
4 S. Specchia, E. Finocchio, G. Busc, P. Palmisano and 37 A. Mart ´ı nez-Arias, M. Fern ´a ndez-Garc ´ı a, A. Iglesias-Juez,
1
1
1
1
V. Specchia, J. Catal., 2009, 263, 134–145.
5 S. Specchia, E. Finocchio, G. Busca, G. Saracco and
V. Specchia, Catal. Today, 2009, 143, 86–93.
A. B. Hungr ´ı a, J. A. Anderson, J. C. Conesa and J. Soria,
Appl. Catal., B, 2001, 31, 39–50.
38 G. Dong, J. Wang, Y. Gao and S. Chen, Catal. Lett., 1999, 58,
37–41.
6 M. S. Hedge, G. Madras and K. C. Patil, Acc. Chem. Res., 2009,
4
2, 704–712.
39 H. Zhu, Z. Qin, W. Shan, W. Shen and J. Wang, J. Catal.,
2004, 225, 267–277.
7 P. Bera, A. Gayen, M. S. Hedge, N. P. Lalla, L. Spadaro,
F. Fursteri and F. Arena, J. Phys. Chem. B, 2003, 107, 6122– 40 H. Zhu, Z. Qin, W. Shan, W. Shen and J. Wang, J. Catal.,
130. 2005, 233, 41–50.
8 T. Baidya, A. Gayen, M. S. Hegde, N. Ravishankar and 41 H. Zhu, Z. Qin, W. Shan, W. Shen and J. Wang, Catal. Today,
L. Dupont, J. Phys. Chem. B, 2006, 110, 5262–5267. 2007, 126, 382–386.
9 T. Baidya, M. S. Hegde, N. Ravishankar, A. Marimuthu and 42 S. Yang, W. Zhu, Z. Jiang, Z. Chen and J. Wang, Appl. Surf.
G. Madras, J. Phys. Chem. C, 2007, 111, 830–839. Sci., 2006, 252, 8499–8505.
0 T. Baidya, G. Dutta, M. S. Hegde and U. V. Waghmare, Dalton 43 F. Liang, H. Zhu, Z. Qin, G. Wang and J. Wang, Catal.
Trans., 2009, 455–464. Commun., 2009, 10, 737–740.
1 B. D. Mukri, G. Dutta, U. V. Waghmare and M. S. Hegde, 44 K. Nakagawa, Y. Murata, M. Kishida, M. Adachim, M. Hiro
Chem. Mater., 2012, 24, 4491–4502. and K. Susa, Mater. Chem. Phys., 2007, 104, 30–39.
2 M. Kurnatowska, L. Kepinski and W. Mista, Appl. Catal., B, 45 H. Zou and Y. S. Lin, Appl. Catal., A, 2004, 265, 35–42.
6
1
1
2
2
2
2
2012, 117–118, 135–147.
46 T. Tanabe, Y. Nagai, T. Hirabayashi, N. Takagi, K. Dohmae,
N. Takahashi, S. Matsumoto, H. Shinjoh, J. N. Kondo,
J. C. Schouten and H. H. Brongersma, Appl. Catal., A, 2009,
370, 108–113.
3 S. Colussi, A. Gayen, M. F. Camellone, M. Boaro, J. Llorca,
S. Fabis and A. Trovarelli, Angew. Chem., Int. Ed., 2009, 48,
8481–8484.
24 L. Liu, F. Zhou, L. Wang, X. Qi, F. Shi and Y. Deng, J. Catal., 47 K. Shimizu, Y. Saito, T. Nobukawa, N. Miyoshi and
2010, 274, 1–10.
A. Satsuma, Catal. Today, 2008, 139, 24–28.
8 K. Shimizu, Y. Saito, T. Nobukawa and A. Satsuma, Top.
Catal., 2010, 53, 584–590.
4
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 54187–54193 | 54193