Effect of aging atmosphere on catalytic activity for NO-CO-C 3H6-O2 reaction of CeO2-containing oxide supported pd catalysts

Article abstract of DOI:10.1007/s10562-013-1015-5

The catalytic activity for CO, hydrocarbon and NO removal on Al ₂O₃ and CeO₂ based oxides supported Pd catalysts were studied under switching of aging condition between air and N ₂ atmosphere. For CeO₂-containing Pd catalysts, the deterioration of catalytic activities by aging in N₂ was improved by oxidative treatment. Based on results of XPS and FT-IR measurements, it was presumed that the catalytic activity of Pd catalysts was strongly affected by the adsorbed form of CO on Pd, which occurs owing to a change of the chemical state of Pd.

Full text of DOI:10.1007/s10562-013-1015-5

Catal Lett (2013) 143:687–696
DOI 10.1007/s10562-013-1015-5
Effect of Aging Atmosphere on Catalytic Activity for NO–CO–
C3H6–O2 Reaction of CeO2-Containing Oxide Supported Pd
Catalysts
•
Takashi Wakabayashi Sumio Kato
•
•
Yuunosuke Nakahara Ryoichi Oshima
•
•
Masataka Ogasawara Shinichi Nakata
Received: 17 December 2012 / Accepted: 23 April 2013 / Published online: 4 May 2013
Ó Springer Science+Business Media New York 2013
Abstract The catalytic activity for CO, hydrocarbon
and NO removal on Al2O3 and CeO2 based oxides
supported Pd catalysts were studied under switching of
aging condition between air and N2 atmosphere. For
CeO2-containing Pd catalysts, the deterioration of catalytic
activities by aging in N2 was improved by oxidative
treatment. Based on results of XPS and FT-IR measure-
ments, it was presumed that the catalytic activity of Pd
catalysts was strongly affected by the adsorbed form of
CO on Pd, which occurs owing to a change of the
chemical state of Pd.
rhodium [5–7]. TWCs has used under fluctuating atmo-
sphere of the ratio air/fuel at high temperature. Recently,
an improvement of TWCs performance and an efficient use
of the precious metals are demanded in order to accom-
modate the global new legislation and the rising precious
metal prices. Therefore it will be needed saving the con-
sumption of precious metal and preventing the catalyst
from the deterioration of purification performance for
exhaust gas under atmosphere fluctuating O2 concentration
at high temperature. There have been a number of reports
on the deactivation of Pd catalyst [5, 8–10]. It is known that
changes in the chemical states of Pd affect remarkably the
catalytic activity and thermal stability of Pd catalyst [8, 11–
14]. Since the vapor pressure of Pd metal is higher than that
of PdO, PdO decomposition temperature and oxygen par-
tial pressure affect the sintering of Pd [8, 11]. To suppress
sintering of dispersed Pd particles, attempts to increase
PdO decomposition temperature have been made by dop-
ing the electronegative promoters such as alkaline earth
metal in a carrier [5]. Hinokuma et al. [15] has reported
that the Pd/CeO2 catalyst by the treatment of thermal aging
at 900 °C in air was remarkably increased catalytic activ-
ities for the ambient-temperature CO oxidation. However
improved phenomenon of the deteriorated purification
performance for the CeO2 containing Pd catalysts have not
been investigated in detail.
Keywords Palladium ꢀ Cerium oxide ꢀ Oxidation
1 Introduction
Pd has widely used in three-way catalysts (TWCs) of auto-
mobile for active species which oxidize carbon monoxide
(CO) and hydrocarbons (HC) [1–4]. And there have been
several studies on the properties of reduction for nitrogen
oxides (NOx) in Pd catalysts for the alternative material of
In this work, we studied the catalytic activity for NO–
CO–C3H6–O2 reaction and surface states of dispersed Pd
on c-Al2O3, CeO2 or CeO2–ZrO2 mixed oxides under the
switch of aging condition using by air and nitrogen (N2)
atmosphere as simple endurance test of TWCs. Conse-
quently, it was found that the deteriorated purification
performance of the CeO2 containing Pd catalysts by the
aging in N2 atmosphere was recovered by the aging in
air.
T. Wakabayashi (&) ꢀ Y. Nakahara ꢀ R. Oshima
Catalysts Strategic Division, Mitsui Mining & Smelting Co.,
Ltd., 1013-1 Ageoshimo, Ageo, Saitama 362-0025, Japan
e-mail: t_wakabayashi@mitsui-kinzoku.co.jp
S. Kato ꢀ M. Ogasawara ꢀ S. Nakata
Department of Applied Chemistry, Graduate School of
Engineering and Resource Science, Akita University,
1-1 Tegatagakuen-machi, Akita 010-8502, Japan
123

688
T. Wakabayashi et al.
2 Experimental
2.1 Sample Preparation
The catalyst powder loaded 1 mass% Pd supported on c-
Al2O3, CeO2 or CeO2–ZrO2 mixed oxides were prepared
by impregnation of Pd(NO3)2 solution (Tanaka Noble
Metals Co, Ltd), followed by drying at 100 °C and calci-
nations at 600 °C for 3 h in air. CeO2 and CeO2–ZrO2
mixed oxides powder used as supports were prepared using
a precipitation method [16]. La doped c-Al2O3 (Sasol,
Ltd.) was used as a support for Pd/c-Al2O3 catalyst. The
slurry prepared by mixing of each powders and inorganic
binder were coated 100 g/L on a ceramics honeycomb
(NGK INSULATORS, LTD.), followed by drying at 90 °C
for 15 min and calcinations in air at 600 °C for 3 h.
Aging of the catalysts was performed at 900 °C under
air and N2 atmosphere shown in Table 1. Hereinafter,
aging atmospheres of air and N2 are denoted as A and N,
respectively. CeO2–ZrO2 is denoted as CZ-X/Y, where X
and Y indicate the weight percentage of CeO2 and ZrO2,
respectively.
Fig. 1 Schematic drawing of the catalytic reactor
10 °C/min and held for 10 min in flowing air at 100 cm3/
min. Then, the catalysts were purged with He fed at
100 cm3/min for 1 min, and then held for 10 min in
flowing H2 at 100 cm3/min. Subsequently the catalyst
cooled down to 50 °C under He atmosphere, and the cat-
alyst held at 50 °C under switched feed gases. The order of
feed gases was as follows, air (5 min), He (1 min), CO2
(10 min), He (1 min), H2 (5 min) and He 5 min at
100 cm3/min. The CO pulse adsorption was performed at
50 °C. CO concentration in the effluent gas was evaluated
using a Shimadzu GC-8A gas chromatograph with TCD.
Based on an assumed stoichiometry of one adsorbed CO
molecule per exposed Pd atom, the particle size of Pd was
calculated by using Eq. (1) [18].
2.2 Catalytic Activity Test and Characterization
Catalytic activity test was performed raising the tem-
perature from 100 to 500 °C in a simulated stoichiometric
gas containing CO (0.69 vol.%), C3H6 (400 ppm), NO
(500 ppm), O2 (0.5 vol.%), CO2 (14 vol.%), H2O (10 vol.%)
and N2 balance. The total gas flow during the experiments
was 25 L/min, GHSV 20,000 h-1, and the heating rate of
the furnace was 20 °C/min. The experimental set-up for
the activity measurements is presented in Fig. 1. The
concentration of CO, C3H6 and NO in the effluent gas was
continuously monitored by infrared (IR), flame ionization
and chemiluminescence detectors with an on-line gas
analyzer (Horiba, MEXA7100).
The characterization of catalyst which was stripped from
the coated ceramics honeycomb after the aging treatment
was carried out. The dispersion of Pd on Pd/CeO2, Pd/c-
Al2O3 and Pd/CeO2–ZrO2 catalysts was measured by using
CO pulse method [17]. The 0.2 g of catalyst was loaded
into a quartz tube reactor. The catalyst heated to 400 °C at
Dispersion ¼ 1:1=d ðnmÞ
ð1Þ
A specific surface area was estimated by N2 adsorp-
tion measurement using Brunauer–Emmett–Teller (BET)
method (Quantachrome Instrument, Quadrasorb SI). X-ray
diffraction (XRD) patterns of the crystal phases of the aged
Pd catalysts were recorded using a Rigaku RINT-TTRIII
diffract meter with Cu Ka radiation. The data for step-
scanning XRD were recorded at 0.02° intervals in the range
30.0° B 2h B 50.0° with 10 s count accumulation per a
step. Chemical states of Pd on supports were investigated
by X-ray photoelectron spectroscopy (XPS, Shimazu,
KRATOS AXIS-ULTRA DLD). XPS measurement was
performed using monochromatic Al Ka X-rays (15 kV,
10 mA). The peak position was calibrated using the binding
energy of C 1s (284.6 eV). Fourier transform infrared
Table 1 Schematic table of aging condition of Pd catalysts
Symbol
Atmosphere
Aged A
Aged N
Aged for 4 h in air
Aged for 4 h in N2
Aged AA Aged for 8 h in air
Aged AN Switched to aging for 4 h in air after aging for 4 h in N2
Aged NA Switched to aging for 4 h in N2 after aging for 4 h in air
123

Effect of Aging Atmosphere on Catalytic Activity
689
spectroscopy (FT-IR) experiments were carried out using a
JASCO FT/IR-6200 spectrometer equipped with MCT
detector and KBr beam splitter. Spectrum was obtained
with 64-scan data acquisition at resolution of 4 cm-1 in
controlled gas atmosphere and temperature, employing a
diffuse reflectance infrared Fourier transform (DRIFT)
spectroscopy cell. The sample was pretreated in N2 flow at
600 °C for 30 min and then cooled to 50 °C in N2. The
background spectrum of the surface was measured for
spectral correction. Measurement of spectra was carried out
after introducing 25 vol.% CO with the balance of N2 at a
flow rate of 0.1 L/min.
3 Result and Discussion
3.1 Catalytic Activity and Pd Dispersion
The catalytic activities for NO–CO–C3H6–O2 reaction
using a simulated stoichiometric exhaust gas over Pd/
CeO2, Pd/c-Al2O3 and Pd/CZ were investigated and eval-
uated the temperature at which conversion is attained to
50 % (T50) and a conversion at 500 °C (g500) of CO, C3H6
and NO. Figures 2, 3, 4, 5 and 6 shows values of T50 and
light-off curves of conversions for the fresh and aged (at
900 °C) Pd catalysts, respectively. As shown in Figs. 3, 4,
5 and 6, g500 of CO, C3H6 and NO were reached nearly
100 % in all catalysts. Pd/CeO2 catalyst exhibited the
highest activity for both CO/C3H6 oxidation and NO
reduction, except for the aged-N catalysts. T50 for CO,
C3H6 and NO of fresh Pd/c-Al2O3 catalyst shifted from 221
to 265 °C for the fresh catalyst to over 300 °C after the
aging at 900 °C (Fig. 2). On the other hand, the T50 for CO
of Pd/CeO2 shifted higher temperature from 176 °C for the
fresh catalyst to 348 °C for the aged-N catalyst, and shifted
lower temperature, 185 °C, which is close to that of the
fresh catalyst, after the following oxidative treatment
(aged-NA). This activity recovery behavior also appeared
in the case of C3H6 and NO, and the temperature shift
became larger with increasing CeO2 contents. Additionally,
the T50 for CO of Pd/CeO2 shitted slightly lower from
176 °C for the fresh catalyst to 167 °C after oxidative
treatment (aged-A). However, after the following aging in
N2, it raised to 200 °C (aged-AN). T50 for CO of Pd/CeO2
after oxidative treatment, aged-A and AA, were close to
each other, such as 167 and 169 °C, respectively. Similar
behaviors of T50 for C3H6 and NO were also observed for
the CeO2 containing Pd catalysts. These results suggest that
the catalytic activities over CeO2 containing Pd catalysts
were affected by the aging atmosphere. In addition, it was
found that the catalytic activity of CeO2 containing Pd
catalyst was deteriorated by aging in N2 and recovered by
the following aging in air.
Fig. 2 Light-off temperature of 50 % conversion rate in NO–CO–
C3H6–O2 reaction over Pd catalysts
The BET surface area and Pd dispersions (CO/Pd) on
each sample for fresh and aged at 900 °C are shown in
Table 2. The BET surface area of fresh catalysts for Pd/c-
Al2O3, Pd/CeO2, Pd/CZ-50/50 and Pd/CZ-30/70 were 178,
91, 83 and 90 m2/g, respectively. The surface area of the
catalysts aged for 8 h (aged-AA) is smaller than that for
4 h (aged-A), suggesting that sintering of the support and
Pd proceeded with aging time. The surface areas of Pd/c-
Al2O3, for fresh and after aging under all the aging con-
ditions were higher than those of CeO2-containing Pd
catalysts. On the other hand, surface area of the Pd/c-Al2O3
after aged-AA treatment was the lowest among the aged
Pd/c-Al2O3 catalysts, indicating that aging process in oxi-
dative atmosphere causes to decrease surface area. CO/Pd
of fresh catalysts, Pd/c-Al2O3, Pd/CeO2, Pd/CZ-50/50 and
Pd/CZ-30/70 were 0.130, 0.559, 0.429 and 0.430,
123

690
T. Wakabayashi et al.
Fig. 3 Conversion curves of CO, C3H6 and NO over Pd/CeO2 for NO–C3H6–CO–O2 reaction
Fig. 4 Conversion curves of CO, C3H6 and NO over Pd/Al2O3 for NO–C3H6–CO–O2 reaction
Fig. 5 Conversion curves of CO, C3H6 and NO over Pd/CZ (C/Z = 50/50) for NO–C3H6–CO–O2 reaction
respectively. CO/Pd of each aged Pd catalysts decreased in
the sequence of Pd/CeO2 [ Pd/CZ-50/50 [ Pd/CZ-30/
70 [ Pd/c-Al2O3, except for the catalysts after aged-N
treatment. This sequence is inconsistent with that of the
surface area. In the case of CeO2-containing Pd catalysts
after aging, except for the aged-N catalyst, CO/Pd
decreased with increasing CeO2 contents in the support.
After aging in the absence oxygen atmosphere (aged-N),
the CO/Pd of CeO2-containing Pd catalysts was significantly
small, but recovered by following oxidative treatment
123

Effect of Aging Atmosphere on Catalytic Activity
691
Fig. 6 Conversion curves of CO, C3H6 and NO over Pd/CZ (C/Z = 30/70) for NO–C3H6–CO–O2 reaction
Table 2 Surface area and Pd dispersion in Pd catalysts after aging at 900 °C
Surface area (m2/g)
Dispersion: CO/Pd
Aged at 900 °C
Aged at 900 °C
Fresh
A
N
AA AN NA Fresh
A
N
AA
AN
NA
Pd/c-Al2O3
178
91
102 131 94 105 104 0.13 (8.6) 0.04 (32.0) 0.03 (36.2)
0.03 (40.0) 0.02 (53.4)
0.04 (27.3)
0.31 (3.7)
0.13 (8.6)
Pd/CeO2
54
46
52
59
73
70
46
41
48
52
44
52
43
50
51
0.56 (2.0) 0.47 (2.4)
0.44 (2.6) 0.18 (6.3)
0.01 (160.2) 0.41 (2.7)
0.03 (38.7) 0.12 (9.4)
0.29 (3.8)
Pd/CZ-50/50 83
Pd/CZ-30/70 90
0.02 (56.1)
0.43 (2.6) 0.08 (13.8) 0.03 (44.9)
0.06 (20.4) 0.01 (112.1) 0.10 (11.4)
() The particle size of Pd species (nm)
(aged-NA). The change of CO/Pd for the aged-N and NA
catalysts exhibited similar behavior with T50 of CO, C3H6
and NO, namely T50 shifted higher temperature with
decreasing CO/Pd. This result suggests that CO/Pd of the
CeO2-containing Pd catalysts was independent on the
surface area and the oxidation of Pd species leads to higher
dispersion of Pd on CeO2. Similar behavior has been
reported for Pd/c-Al2O3 catalyst by Xu et al. [19] that Pd
sintering was not influenced by the sintering of support.
Nagai et al. [20] has reported that the interaction between
Pt and CeO2 at 800 °C under oxidizing condition causes
redispersion of as grown Pt crystallites in Pt/CeO2 catalyst.
In the present study, the redispersion behavior of Pd on
CeO2-containing catalysts was found.
metallic Pd, leading to faster Pd sintering at the higher
temperature in air. T50 of CO was 356 and 226 °C after
aged-N and aged-NA treatment at 950 °C, respectively,
indicating that the catalytic activity at lower temperature
was improved by the oxidative treatment after N2 treat-
ment. A similar phenomenon was also observed for C3H6
oxidation and NO reduction over the Pd/CeO2 and for the
catalysts aged at 1,000 °C. On the other hand, for Pd/c-
Al2O3 catalysts, T50 of CO, C3H6 and NO were shifted high
temperature with increasing the aging time and
temperature.
Figure 8 shows T50 of CO plotted as function of CO/Pd
for Pd/c-Al2O3, Pd/CeO2, Pd/CZ-50/50 and Pd/CZ-30/70
catalysts. It was found that the value of T50 is correlated
with CO/Pd. As the results mentioned above, the value of
CO/Pd for the aged-NA catalyst was higher than that for
the aged-N catalyst in CeO2-containing Pd catalysts. These
results suggest that the recovery behavior of the catalytic
activity after aging in air is attributed to a characteristic of
Pd surface on the support.
Table 3 and Fig. 7 shows BET surface area, CO/Pd and
value of T50 for the Pd catalysts aged at 950 and 1,000 °C,
respectively. The surface area of Pd/CeO2 catalyst aged at
1,000 °C was lower than that of the catalyst aged at 900 °C
in both aged-N and aged-NA conditions. CO/Pd of Pd/
CeO2 catalysts after the aged-N and aged-NA treatment at
950 °C were 0.01 and 0.21, respectively, indicating that
redispersion of Pd occurred by the aged-NA treatment.
After the aged-NA treatment at 1,000 °C, the CO/Pd was
lower than that after the treatment at 900 °C. This phe-
nomenon would be caused by PdO decomposition to
In order to determine Pd species on the catalysts, XRD
experiments were carried out. The XRD patterns of Pd/c-
Al2O3 and Pd/CeO2 for the aging at 900 °C are shown in
Fig. 9. In Fig. 9a, the most of peaks are consistent with the
characteristic peaks of cubic CeO2, and metallic Pd
123

692
T. Wakabayashi et al.
Table 3 Surface area and Pd dispersion in Pd catalysts after aging at 950 and 1,000 °C
Surface area (m2/g)
Dispersion: CO/Pd
Aged at 950 °C
Aged at 1,000 °C
Aged at 950 °C
Aged at 1,000 °C
N
NA
N
NA
N
NA
N
NA
Pd/c-Al2O3
128
39
104
34
123
20
103
20
0.03
0.01
0.03
0.21
0.02
0.01
0.03
0.02
Pd/CeO2
Fig. 8 Relationships between CO/Pd and T50 of CO
Al2O3. Metallic Pd for the aged-AN and aged-N catalysts
and PdO for the aged-A and NA catalysts were detected.
These crystallite sizes of metallic Pd and PdO were
16–21 nm. This crystallite sizes are not correlated the
particle size of Pd species which estimated from CO/Pd in
Table 2. Xu et al. [21] has reported that the correlation of
Pd metal particle size was obtained between TEM, XRD,
H2 chemisorption and CO oxidation in Pd/h-Al2O3 catalyst.
In their study, the sample was immediately quenched in
liquid N2 after N2 aging to prevent oxidation of metallic Pd
into polycrystalline PdO. The difference between particle
sizes estimated by XRD and CO adsorption in the present
study would be attributed to the existence of polycrystal-
line which formed during the cooling after N2 aging and
the adsorbed form of CO on Pd species such as Pd0–CO,
(Pd0)2–CO and (Pd0)3–CO. These results suggest that a
number of CO absorption sites on Pd affected the catalytic
activity. In order to clarify the recovery behavior of the
catalytic activity of Pd/CeO2 catalysts, the chemical state
of Pd and a form of adsorbed CO were investigated.
Fig. 7 Light-off temperature of 50 % conversion rate in a simulated
exhaust gas over Pd catalysts for aging at 950 and 1,000 °C
(2h = 40.1°) was detected in the aged-N and aged-AN Pd/
CeO2 catalyst. The crystallite sizes of metallic Pd esti-
mated by Scherrer’s equation for the aged-N and aged-AN
catalysts were 46.8 and 55.9 nm, respectively. The peak
observed at 2h = 40.1° was disappeared in the case of the
catalyst after the treatment in inert and following oxidative
atmosphere (aged-NA). The strongest peak of PdO
(2h = 33.8°) could not be observed in all Pd/CeO2, due to
the existence of the peak of CeO2 in the region of
2h = 31–35°. Figure 9b shows the XRD patterns for Pd/c-
3.2 The Chemical State of Pd
The change in the chemical state of Pd on the catalysts
treated each aging condition at 900 °C were studied using
XPS. Pd 3d XPS spectra of Pd/CeO2 and Pd/c-Al2O3 cat-
alysts aged under the each condition are shown in Fig. 10.
123

Effect of Aging Atmosphere on Catalytic Activity
693
Fig. 9 XRD patterns for Pd
catalysts: a Pd/CeO2 and b Pd/
Al2O3
The Pd 3d5/2 peaks of Pd/CeO2 catalysts were observed in
the region of 334–339 eV (Fig. 10a). It is reported that Pd
3d5/2 peak of metallic Pd (Pd0), PdO (Pd2?) and PdO2
(Pd4?) are shown at 335.2, 336.7 and 337.5 eV, respec-
tively [22–27]. The peak observed at 337.7 eV for the fresh
of Pd/CeO2 catalysts was assigned to Pd4?. The peak of
Pd4? was observed for the aged-A and NA catalysts after
oxidative treatment, and it was also observed the aged-AN
catalyst after aging in inert atmosphere. For the aged-N
catalyst, the weak peak observed at 335.4 eV was assigned
to Pd0. The reason for this result of the aged-N catalyst may
be due to a decrease of Pd atoms on the catalyst. These
results indicate that reduction of Pd4? on CeO2 after the
oxidative treatment at 900 °C is difficult and the chemical
state of Pd0 for the aged-N easily changes to that of the
fresh catalyst by oxidative treatment at 900 °C (the aged-
NA). On the other hand, the peaks observed at 334.8 and
336.7 eV for the Pd/c-Al2O3 catalysts were assigned to Pd0
and Pd2? (Fig. 10b). The peaks of Pd0 were observed for
the catalyst aged under inert atmosphere (aged-N and AN).
Pd2? were observed for the catalyst aged under oxidative
atmosphere (aged-A and aged NA), namely the chemical
state of Pd on Al2O3 was easily changed by the aging
atmosphere. From these results of the XPS measurement, it
was found that the chemical state of Pd on CeO2 was
higher oxidation state than that of c-Al2O3 under oxidative
atmosphere.
oxide support and Moroseac et al. [28] have suggested that
the metal-support interactions may stabilize PdO2 for Pd/
SnOx catalyst in hydrogen containing air, respectively. On
the other hand, Colussi et al. [29] reported the formation of
a square planer PdO4 unit for Pd2? state in Pd–O–Ce sur-
face of Pd/CeO2 catalyst as revealed by density-functional
theory (DFT) calculations on the basis of high-resolution
TEM data.
3.3 A Form of Adsorbed CO Over Pd Catalysts
The DRIFT spectra of adsorbed CO at 50 °C on Pd/CeO2
and Pd/c-Al2O3 catalysts aged under each condition at
900 °C are shown in Fig. 11. Figure 11a shows the DRIFT
spectra of CO adsorbed Pd/CeO2 catalysts. It is reported
that the bands for linearly adsorbed CO on Pd0, Pd?, Pd2?
and the bridge bonded CO on Pd0 [(Pd0)2–CO] appear near
2085, 2110, 2156 and 2000, 1960 cm-1 for the Pd/CeO2
catalysts, respectively [12, 30–36]. For the fresh catalyst,
the bands observed at 2,089 and 2,140 cm-1 were assigned
to linear Pd0–CO and Pd2?–CO. Intensity of these band
decreased by aging and the band assigned to linear Pd?–
CO appeared at 2,118 cm-1 after oxidative treatment
(aged-A, NA and AN). For the aged-N catalyst, these linear
Pd–CO bands were absent, but the other band was observed
at 1,997 cm-1, which is assigned to bridge (Pd0)2–CO. It
was found that the Pd in aged-N catalyst is present as
metallic state. The bands of Pd0–CO and Pd2?–CO disap-
peared for the aged-N catalyst, but they were observed by
oxidative treatment (aged-NA) again. These results were
There are a number of reports about the existence of
Pd4?. Otto et al. [25] have reported that PdO2 was stabi-
lized by an interaction of a highly dispersed Pd with an
123

694
T. Wakabayashi et al.
Fig. 10 Pd 3d XPS peaks of Pd
catalysts: a Pd/CeO2 and b Pd/
Al2O3
Fig. 11 DRIFT spectra of CO
adsorbed on the Pd catalysts in a
flow of 25 vol.% CO and
balance N2 at 50 °C for 10 min:
a Pd/CeO2 and b Pd/Al2O3
suggested that the adsorbed form of CO on Pd was changed
by the chemical state of Pd on CeO2, and the lowest CO/Pd
for the aged-N (Table 2) would be caused by absence of
adsorbed linear carbonyl on Pd0. In Pd/CeO2 catalysts, no
bands assigned to adsorb CO on Pd4? were observed for
the fresh, aged-A, aged-AN and aged-NA.
Figure 11b shows the DRIFT spectra of CO adsorbed on
Pd/c-Al2O3 catalysts. It is reported that the bands for
linearly adsorbed of CO on Pd0, Pd?, the bridge bonded
of CO on Pd0 and the three-fold bonded of CO on
Pd0 [(Pd0)3–CO] had shown near 2080, 2158, 1975 and
1911 cm-1 for the Pd/c-Al2O3 catalysts, respectively [6,
123

Effect of Aging Atmosphere on Catalytic Activity
695
34, 37, 38]. The weak bands observed at 1,989 and
2,094 cm-1 were assigned to bridge (Pd0)2–CO and Pd0–
CO for the fresh catalyst. After aging under the each
condition, the band at 2,094 cm-1 disappeared and inten-
sity of the band at 1,989 cm-1 decreased. No bands of
linear carbonyl on Pd2? for the fresh, aged-A and aged-NA
were observed except for the fresh catalyst. These results
for the Pd/Al2O3 catalysts are suggested Pd2? was reduced
to Pd0 in 25 vol.% CO/N2 flow.
Figure 12 shows the DRIFT spectra during a flow of
25 vol.% CO and balance N2 over the aged-A Pd/CeO2 cat-
alyst at 50 °C. The bands observed at 2,089 and 2,140 cm-1
were assigned to linear Pd0–CO and Pd2?–CO after a flow of
CO/N2 gas. The bands observed at 2,300–2,400 cm-1 were
assigned to CO2 (g). As from 3 min, intensity of band
assigned to linear Pd2?–CO decreased with the exposure
time. On the other hands, intensity of band assigned to linear
Pd0–CO increased with the exposure time, and the band
clearly appeared at 2,118 cm-1 (linear Pd?–CO) as from
10 min. It was suggested that oxidized Pd species were
reduced to Pd? or metallic Pd in the flow of CO/N2 gas.
Therefore, no bands assigned to adsorb CO on Pd4? would be
observed Pd/CeO2 catalysts in this study. From these results
of the FT-IR experiments, it was found that the adsorbed
forms of CO on Pd/CeO2 and Pd/Al2O3 catalysts after the
oxidative treatment were linear and bridge carbonyl types,
respectively. And it was found that the difference between the
crystallite sizes of Pd species estimated by Scherrer’s equa-
tion and the CO adsorption are caused by the adsorbed form
Fig. 12 DRIFT spectra of CO adsorbed on the aged-A Pd/CeO2
catalyst in a flow of 25 vol.% CO and balance N2 at 50 °C. Exposure
time: 0, 1, 3, 5, 10, 15 and 30 min
c-Al2O3. Reduction of Pd4? formed by the oxidative
treatment at 900 °C was difficult, and the chemical state of
Pd0 for the aged-N returned easily to that of the fresh
catalyst by oxidative treatment at 900 °C.
4 Conclusion
´
´
´
of CO on Pd. Fernandez-Garcia [32] and Martınez-Arias et al.
[33] has reported that linear carbonyl type was easily des-
orbed at lower temperature than bridged carbonyl type on Pd,
and it was suggested that the generated metallic adsorption
sites by desorption of CO was became adsorption sites of NO
for CO–NO–O2 reaction. Moreover Craciun et al. [12] has
reported that the higher dispersion and catalytic activity
showed with increasing the ratio of linear carbonyl to bridged
carbonyl peak areas for methane steam reforming. Therefore
it was thought that CO/Pd and the catalytic activity were
affected by the difference of the adsorbed form of CO on Pd.
From the above-mentioned results, it was found that the
catalytic activity of CeO2 containing Pd catalyst was
deteriorated by the aging in N2 but it was recovered by the
oxidative treatment. And the catalytic activity is depended
on CO/Pd. The deterioration of catalytic activity for the
aged-N in CeO2 containing Pd catalysts was caused by the
decrease in CO/Pd, which takes place by the change of
adsorbed form of CO on Pd from linear Pd–CO for fresh
catalyst to bridge (Pd0)2–CO for the aged-N. This adsorbed
form of CO on Pd was affected the chemical state of Pd.
The chemical state of Pd on CeO2 by the oxidative treat-
ment was existed higher oxidation state (Pd4?) than that of
In the present study, we studied the catalytic activity Pd
catalysts under the switch of aging condition using by air
and N2 atmosphere, and found the catalytic activity of
CeO2-containing Pd catalysts improved by oxidative
treatment at 900 °C. It was shown that Pd/CeO2 catalysts
exhibited higher catalytic activities for CO, C3H6 and NO
than Pd/c-Al2O3 and Pd/CZ catalysts, except that for the
aged-N. For CeO2-containing Pd catalysts, the deteriora-
tion of catalytic activities for the aged-N was improved
by oxidative treatment at 900 °C. The catalytic activity
was correlated with CO/Pd rather than BET surface area
and the crystallite size of Pd species. The XPS results
indicated that the chemical state of Pd on CeO2 per-
formed the oxidative treatment was highly ionic state than
that of c-Al2O3. And the FT-IR measurement revealed
that the adsorbed form of CO on Pd/CeO2 catalysts was
linear carboxyl and that of CO on Pd/Al2O3 catalysts was
bridge carboxyl type. Based on these results, it was pre-
sumed that the catalytic activity for Pd catalysts was
strongly affected by the adsorbed form of CO on Pd,
which takes place by a change of the chemical state of
Pd.
123

696
T. Wakabayashi et al.
21. Xu Q, Kharas KC, Croley BJ, Datye AK (2011) ChemCatChem
3:1004
References
22. Bird RJ, Swift P (1980) J Electron Spectrosc Relat Phenom
21:227
1. Yamada T, Kayano K, Funabiki M (1993) SAE Paper 930253
2. Boehman AL, Niksa S (1993) SAE Paper 932761
3. Brisley RJ, Chandler GR, Jones HR, Anderson PJ, Shady PJ
(1995) SAE Paper 950259
23. Johansson G, Hedman J, Berndtsson A, Klasson M, Nilsson R
(1973) J Electron Spectrosc Relat Phenom 2:295
24. Kim DH, Woo SI, Lee JM, Yang O (2000) Catal Lett 70:35
25. Otto K, Haack LP, de Vries JE (1992) Appl Catal B 1:1
26. Bi Y, Lu G (2003) Appl Catal B 41:279
27. Guimara˜es AL, Dieguez LC, Schmal M (2003) J Phys Chem B
107:4311
¨
4. Lassi U, Polvinen R, Suhonen S, Kallinen K, Savimaki A,
¨
¨
Harkonen M, Valden M, Keiski RL (2004) Appl Catal A 263:241
5. Kobayashi T, Nagatomo S, Ishikawa K, Yamada T (2000) SAE
Paper 2000-01-2927
6. Macleod N, Lambert RM (2003) Appl Catal B 46:483
7. Qi G, Yang RT, Rinaldi FC (2006) J Catal 237:381
8. Datye AK, Bravo J, Nelson TR, Atanasova P, Lyubovsky M,
Pfefferle L (2000) Appl Catal A 198:179
9. Farrauto RJ, Hobson MC, Kennelly T, Waterman EM (1992)
Appl Catal A 81:227
10. Xu Q, Kharas KC, Datye AK (2001) Stud Surf Sci Catal 139:157
11. Shinjoh H, Muraki H, Fujitani Y (1991) Stud Surf Sci Catal
71:617
´
´ ´ ´ ´
28. Moroseac M, Skala T, Veltruska K, Matolın V, Matolınova I
(2004) Surf Sci 566–568:1118
29. Colussi S, Gayen A, Camellone MF, Boaro M, Llorca J, Fabris S,
Trovarelli A (2009) Angew Chem Int Ed 48:8481
30. Zhu H, Qin Z, Shan W, Shen W, Wang J (2004) J Catal 225:267
31. Fernandez-Garcia M, Martınez-Arias A, Salamanca LN, Coro-
nado JM, Anderson JA, Conesa JC, Soria J (1999) J Catal
187:474
´
´
´
´
´
´
´
32. Fernandez-Garcia M, Martınez-Arias A, Iglesias-Juez A, Hungrıa
AB, Anderson JA, Conesa JC, Soria J (2001) Appl Catal B 31:39
¨
12. Craciun R, Daniell W, Knozinger H (2002) Appl Catal A 230:153
13. Ciuparu D, Lyubovsky MR, Altman E, Pfefferle L, Datye A
(2002) Catal Rev Sci Eng 44:593
´
´
´
´
33. Martınez-Arias A, Fernandez-Garcia M, Iglesias-Juez A, Hungrıa
AB, Anderson JA, Conesa JC, Soria J (2001) Appl Catal B 31:51
34. Choi KI, Vannice MA (1991) J Catal 127:465
35. Bensalem A, Muller JC, Tessier D, Bozon-Verduraz F (1996) J
Chem Soc Faraday Trans 92:3233
´
14. Gelin P, Primet M (2002) Appl Catal B 30:1
15. Hinokuma S, Fujii H, Okamoto M, Ikeue K, Machida M (2010)
Chem Mater 22:6183
16. Haneda M, Shinoda K, Nagane A, Houshito O, Takagi H,
Nakahara Y, Hiroe K, Fujitani T, Hamada H (2008) J Catal
259:223
17. Takeguchi T, Manabe S, Kikuchi R, Eguchi K, Kanazawa T,
Matsumoto S (2005) Appl Catal A 293:91
36. Haneda M, 32st Meeting on Japan Reference Catalyst Program
I-5 (2011) 9
37. Valden M, Keiski RL, Xiang N, Pere J, Aaltonen J, Pessa M,
¨ ¨ ¨
Maunula T, Savimaki A, Lahti A, Harkonen M (1996) J Catal
161:614
38. Pitchon V, Primet M, Praliaud H (1990) Appl Catal 62:317
18. Costa CN, Christou SY, Georgiou G, Efstathiou AM (2003) J
Catal 219:259
19. Xu Q, Kharas KC, Croley BJ, Datye AK (2012) Top Catal 55:78
20. Nagai Y, Hirabayashi KT, Dohmae K, Takagi N, Shinjoh T,
Matsumoto S (2006) J Catal 242:103
123