www.elsevier.com/locate/apcatb

Applied Catalysis B: Environmental 69 (2007) 219–225

Effect of ceria–zirconia ratio on the interaction of CO with

PdO/Al2O3–(Cex–Zr1�x)O2 catalysts prepared by sol–gel method

Gabriela Perez-Osorio a, Felipe Castillon b, Andrey Simakov b, Hugo Tiznado a,Francisco Zaera c, Sergio Fuentes b,*

a Posgrado en Fısica de Materiales, Centro de Investigacion Cientıfica y de Educacion Superior de Ensenada, BC (CICESE),

Apdo. Postal 2732, Ensenada, BC, Mexicob Departamento de Catalisis, Centro de Ciencias de la Materia Condensada (CCMC UNAM), Apdo. Postal 2681, Ensenada, BC, Mexico

c Department of Chemistry, University of California at Riverside (UCR), Riverside, CA 92521, USA

Received 15 December 2005; received in revised form 31 May 2006; accepted 6 July 2006

Available online 21 August 2006

Abstract

The current work is devoted to study of CO interaction with PdO/Al2O3–(Cex–Zr1�x)O2 catalysts. Ceria–zirconia–alumina supports with

different Ce/Zr ratio were prepared by sol–gel technique. The FT-IR characterization of CO adsorbed at �120 and 25 8C on oxidized and reduced

samples revealed that Ce/Zr ratio modifies the surface properties of support and oxidation state of palladium. The catalyst with Ce/Zr molar ratio

0.5/0.5 was characterized with the highest ability to stabilize palladium in oxide state and the highest activity to oxidize CO. Redox treatment of

catalysts improves their catalytic activity.

# 2006 Published by Elsevier B.V.

Keywords: Palladium oxide; Al2O3–(Cex–Zr1�x)O2; CO adsorption; In situ FT-IR; Catalytic CO oxidation

1. Introduction

The palladium-only three-way catalysts (Pd-only TWC) are

used for automotive catalytic converters with closed-loop

control systems [1–11], due to their lower cost, their ability to

catalyze the hydrocarbon oxidation and their durability under

high temperature conditions. In fact, the fourth generation of

these catalytic converters has been available in the market since

1995 [12,13]. Pd-only catalysts can be effective as TWC for the

simultaneous removal of NO, CO and hydrocarbons in

automotive exhaust gases [7,14], but they typically show

lower efficiency in NO conversion compared with TWC based

on Rh [15,16]. Nevertheless, the catalytic activity of Pd-only

TWC could be improved by doping with rare earth oxides. For

instance, Trovarelli [17] reported that due to its properties,

structure and capabilities of storing and releasing oxygen, CeO2

is an important component in automotive emission-control

(oxidation) catalysts. The role of oxygen vacancies and their

* Corresponding author. Tel.: +52 646 1744602; fax: +52 646 1744603.

E-mail address: fuentes@ccmc.unam.mx (S. Fuentes).

0926-3373/$ – see front matter # 2006 Published by Elsevier B.V.

doi:10.1016/j.apcatb.2006.07.001

mobility on ceria surfaces, has been recently studied by Esch

et al. [18]. The mechanism of oxygen vacancies appearance and

their interaction upon further ceria reduction have been recently

studied in detail by scanning tunneling microscopy and density

functional calculations [18]. It was shown that on the slightly

reduced surface single vacancies prevail and can be distin-

guished as two types. One type appears as depressions

surrounded by three paired lobes and can be assigned to

surface oxygen vacancies. A second type appears as triple

protrusions centered about of the third layer oxygen sites. Upon

further reduction almost all vacancies interact with each other

with formation so-called linear surface oxygen vacancies with

length of several nm. In addition the authors concluded that

single vacancy formation around single Zr4+ dopants is

facilitated.

The catalytic activity of the palladium catalysts depends as

much on their method of preparation [19,20] as well as the

composition of the support [21–25]. For Pd supported on Al2O3

catalysts modified with ceria and zirconia, the extent of the

effect of this promotion on both the reduction of NO and the

oxidation of CO depends on the interaction between the support

and palladium, which determines the nature of the Pd species

G. Perez-Osorio et al. / Applied Catalysis B: Environmental 69 (2007) 219–225220

and its particle size. Thus, the physical and chemical

characteristics of modifier play an important role in these

reactions [26,27], which requires a strict control of the

composition of the support to obtain the best thermal stability

and optimal catalytic activity [28,29].

Incorporation of Ce–Zr by sol–gel into the alumina matrix

can improve more efficient degree of Ce–Zr dispersion and

thermal stability of alumina. The sol–gel method provides the

capacity to mix several components in a single stage, and

allows controlling the structure and composition of the final

solid mixture at molecular level [30]. In addition, sol–gel

methods give a homogeneous mixture of Al2O3–MOx binary

oxides with enhanced efficiency for NO reduction. In the past,

we have reported that palladium catalysts deposited on Al2O3–

La2O3 supports prepared by the sol–gel method show an

enhancement in the reducibility properties, and that those

changes influence the activity of reduction of NO by hydrogen

[31]. It was suggested in that work that the higher selectivity

towards the production of N2 found on the catalysts of Pd/

Al2O3–La2O3, compared with the Pd/Al2O3 catalysts at

temperatures below 350 8C is due to the interaction of

palladium particles and reduced species of lantana.

The aim of this work was to determine the effect of ceria–

zirconia ratio on the interaction of CO with PdO/Al2O3–(Cex–

Zr1�x)O2 catalysts in order to obtain further information about

the catalyst structure and surface configuration by FT-IR

spectroscopy, that is important in their efficiency in CO

oxidation. The samples of alumina–ceria–zirconia mixed

oxides were prepared by sol–gel method in order to get

homogeneous support for catalysts and to promote the

interaction between PdO and support.

2. Experimental

2.1. Preparation of catalysts

Alumina supports were prepared with different Ce/Zr ratios

(Al2O3–(Cex–Zr1�x)O2 with x = 0, 0.25, 0.33, 0.5, 0.66, 0.75

and 1.0) by the sol–gel method from organic precursors

following a procedure reported previously [32]. A solution of

acetylacetonate of cerium (Aldrich) and/or acetylacetonate of

zirconium (Alfa Aesar) in ethanol with moderate agitation was

added to a mixture of aluminum sec-butoxide (Aesar Alpha) in

Table 1

BET surface area and experimental composition of catalysts

Catalysts Surface area BET (m2 g�1) Exp

Pd

PdO/Al2O3–CeO2 246 0.2

PdO/Al2O3(Ce0.75Zr0.25)O2 253 0.2

PdO/Al2O3(Ce0.67Zr0.33)O2 260 0.2

PdO/Al2O3–(Ce0.5Zr0.5)O2 246 0.1

PdO/Al2O3(Ce0.33Zr0.67)O2 253 0.2

PdO/Al2O3(Ce0.25Zr0.75)O2 257 0.2

PdO/Al2O3–ZrO2 249 0.1

PdO/Al2O3 214 0.2

a Calculation was made by difference based on stoichiometry of Ce, Zr and Al

2-metil-2,4-pentanediol (Aesar Alpha), staying in reflux for

3 h, with moderate agitation at 94 8C. Hydrolysis was made by

adding deionized water. The obtained gel was aged for 10 h.

Samples were dried at vacuum (about 10�2 Torr) at 100 8C for

12 h and then calcined in N2 atmosphere at 450 8C for 12 h,

with a later treatment in air at 650 8C for 4 h.

The support impregnation was carried out using a palladium

chloride solution of 0.12% (w/v), to give a concentration of

0.3 wt.% of palladium. After impregnation, the samples were

calcined in air at 650 8C for 4 h.

Chemical analysis of prepared samples by inductive coupled

plasma (ICP) technique was done using Optical Emission

Spectrometer Optima 4300D. Before analysis samples were

dissolved in sulphuric acid solution (1:1). The experimental

composition of the prepared catalysts determined by ICP as

well as the specific surface area determined by nitrogen

adsorption and BET method are presented in the Table 1.

According to the data presented in a previous paper [33] for

most of the catalysts it could be concluded that (Ce1�xZrx)O2

mixed oxide is present like nano-crystallites which size is

below the XRD detection limits. Only for the PdO/Al2O3–

(Ce0.75Zr0.25)O2 catalyst it was observed formation of a

(Ce1�xZrx)O2 mixed oxide solution. The particle size of PdO

was about 6 nm for PdO/Al2O3–CeO2, it decreased to 3.3 nm

for PdO/Al2O3–(Ce0.25Zr0.75)O2 and finally it increased a little

for PdO/Al2O3–ZrO2 (4.5 nm).

2.2. CO adsorption by FT-IR

The CO adsorption was performed as described in previous

report [34]. Typical experiments were carried out step by step

from low temperatures (LT,�120 8C) to room temperature (RT,

25 8C) and from 5 Torr of CO up to evacuation for 40 min at

10�2 Torr. Infrared spectra of CO adsorbed on the catalysts

were registered with a resolution of 4 cm�1, with a Bruker

Equinox 55 Fourier-transformed infrared (FT-IR) spectrometer

in transmittance mode with DTGS detector [35]. Experiments

were carried out in a quartz laboratory-made cell with windows

of NaCl able to work from �120 to 550 8C and pressures of

10�2 to 760 Torr. The catalyst powders were pressed into self-

supporting discs of 13 mm of diameter and less than 20 mg of

weight. Samples were exposed to 5 Torr of carbon monoxide

(Matheson Research grade) purified using a trap cooled with

erimental composition ICP (wt.%) Calculated valuea (wt.%)

Ce Zr Al2O3

0 4.45 <0.007 95.34

0 3.42 1.45 94.93

4 3.09 2.03 94.64

9 2.27 2.84 94.70

0 1.37 3.68 94.75

0 1.07 4.12 94.61

8 <0.006 5.5. 94.31

9 <0.006 <0.006 99.70

oxides.

G. Perez-Osorio et al. / Applied Catalysis B: Environmental 69 (2007) 219–225 221

Fig. 1. FT-IR spectra of catalysts recorded at �50 8C (after introduction of

5 Torr of CO at �120 8C, 40 min evacuation and heating up to �50 8C). The

samples were pretreated at (O550). (A) PdO/Al2O3, (B) PdO/Al2O3–ZrO2, (C)

PdO/Al2O3–(Ce0.25Zr0.75)O2, (D) PdO/Al2O3–(Ce0.33Zr0.67)O2, (E) PdO/

Al2O3–(Ce0.50Zr0.50)O2, (F) PdO/Al2O3–(Ce0.67Zr0.33)O2, (G) PdO/Al2O3–

(Ce0.75Zr0.25)O2 and (H) PdO/Al2O3–CeO2.

liquid nitrogen and then continuously evacuated. Adsorption

experiments were initiated at �120 8C (LT) in order to avoid

changes of catalysts surface, later, temperature was increased

gradually with ramp of 10 8C/min up to room temperature (RT).

Infrared spectra of CO adsorption were obtained after

subtraction of the background recorded before CO adsorption.

Catalysts were pretreated in situ before FT-IR experiments.

Oxidized catalysts refer to samples calcined in oxygen

(100 Torr) at 550 8C for 1 h, while reduced ones correspond

to samples with oxygen treatment followed with reduction in

hydrogen (100 Torr) at 550 8C for 1 h. These treatments were

labeled as O550 and O550/H550, respectively.

2.3. Catalytic activity in CO oxidation

Catalytic activity test using a 1% CO + 1% O2 (helium

balance) at 40 ml/min was carried out in a fixed bed glass U-

tube reactor (i.d. 4 mm). About 75 mg of sample was packed

between two layers of glass wool. Gases of UHP grade were

controlled with mass flow controllers. Catalysts were preox-

idized in situ in O2 flow at 550 8C for 1 h, cooled at the same

atmosphere and purged by helium at room temperature. During

catalytic activity test the temperature of catalyst bed was

increased from room temperature up to 250 8C with ramp of

2 8C/min. Reagents and products were analyzed by gas

chromatography with a SRI instrument 8610C equipped with

a TCD detector and two columns packed with molecular sieve

and silica-gel, operating at 100 8C.

CO oxidation was carried out on two sets of catalysts, where

the first set refers to fresh samples after oxidation treatment in

O2 flow at 550 8C for 1 h and the Second set corresponds to

redox catalysts. Redox treatment includes two cycles of

reduction in H2/Ar (10/90) flow with temperature increase

programmed from 25 up to 550 8C with ramp of 20 8C/min and

re-oxidation of samples in O2 flow at 550 8C for 1 h.

The catalytic activity data were presented in terms of

temperature required for 50 % CO conversion.

3. Results and discussion

3.1. CO adsorption on oxidized (O550) catalysts

In order to minimize changes on the catalysts surface due to

reaction of CO probe with adsorbed oxygen, the temperature of

adsorption was initially lowered and then gradually increased

stepwise at definite rate.

The effect of Ce/Zr ratio on CO chemisorption over PdO

species was studied comparing FT-IR spectra of catalysts

recorded at �50 and 25 8C.

Experiments of CO adsorption were performed introducing

5 Torr of CO for 20 min at �120 8C, followed by 40 min of

evacuation (10�2 Torr) at this temperature then heating from

�50 to 25 8C. The temperature increase from �120 to �50 8Ccaused gradual diminution of bands attributed to CO

chemisorbed on the support. Therefore, at �50 8C it was

possible to observe the CO adsorption corresponding to

palladium oxide species.

Spectra of CO adsorbed on catalysts after O550 treatment

(Fig. 1) presented at �50 8C five main bands, three of them

already observed on supports [34,36], S3 (2185 cm�1) and S4

(2200–2205 cm�1) on alumina, Ce4 (2162 cm�1) on ceria, and

two more related to palladium oxide, L1 (2120 cm�1) and L2

(2156 cm�1)) as in [34,24]. In agreement with reference [34]

the S3 and S4 sites could be assigned to CO on Al3+ tetrahedral

and Al3+ octahedral, respectively. The band at 2162 cm�1 may

be assigned to CO coordinated on Ce4+ cations. This band is

shifted upward from physisorbed CO (2157 cm�1) due to

strong interaction with Lewis acid sites in accordance with

studies of CO adsorption on polycrystalline ceria [26,37]. The

appearance of band at 2185 cm�1 in the absence of Ce ions as in

the case of PdO/Al2O3–ZrO2 sample [34] indicates the role of

alumina sites in stabilization of such type of CO species.

However, we can not exclude overlapping of this band with

band of CO adsorbed on Ce4+ ions modified by chlorine, which

was described in the report of Binet et al. [37] and Bensalem

and co-workers [38]. Indeed, modification by chlorine could be

due to PdCl2 used as palladium precursor under the catalysts

preparation. Intensity of signal at 2185 cm�1 could be sum of

two bands from different CO adspecies stabilized on alumina

and ceria.

The L1 and L2 bands similar to those of reference [34,24]

were associated to Pd+ and Pd2+ species, respectively [39]. The

O550 treatment caused a different L1/L2 intensity ratio

depending on the Ce–Zr ratio. So, for PdO/Al2O3 catalyst

the L1/L2 ratio was slightly less than one, it later increased

above one for the PdO/ Al2O3–ZrO2 sample and decreased

drastically when Ce concentration was gradually increased.

Finally, for the PdO/ Al2O3–CeO2 catalyst only the L2 band

was detected. Notice that frequency of band at around

G. Perez-Osorio et al. / Applied Catalysis B: Environmental 69 (2007) 219–225222

Fig. 2. FT-IR spectra of catalysts recorded at 25 8C (after exhibition to 5 Torr of

CO and 40 min evacuation at 25 8C). The samples were pretreated at (O550). (A)

PdO/Al2O3, (B) PdO/Al2O3–ZrO2, (C) PdO/Al2O3–(Ce0.25Zr0.75)O2, (D) PdO/

Al2O3–(Ce0.33Zr0.67)O2, (E) PdO/Al2O3–(Ce0.50Zr0.50)O2, (F) PdO/Al2O3–

(Ce0.67Zr0.33)O2, (G) PdO/Al2O3–(Ce0.75Zr0.25)O2 and (H) PdO/Al2O3–CeO2.

Fig. 3. Spectra of catalysts within range 1100–1600 cm�1 (carbonate region)

recorded at 25 8C (after sample exhibition with 5 Torr of CO at �120 8C,

40 min evacuation and heating up to 25 8C, exhibition to 5 Torr of CO and

40 min evacuation at 25 8C). The samples were pretreated at (O550). (A) PdO/

Al2O3, (B) PdO/Al2O3–ZrO2, (C) PdO/Al2O3–(Ce0.25Zr0.75)O2, (D) PdO/

Al2O3–(Ce0.33Zr0.67)O2, (E) PdO/Al2O3–(Ce0.50Zr0.50)O2, (F) PdO/Al2O3–

(Ce0.67Zr0.33)O2, (G) PdO/Al2O3–(Ce0.75Zr0.25)O2 and (H) PdO/Al2O3–CeO2.

The bands assignation is showed in Table 2. Bold curve (E0) corresponds to CO

adsorption over bare Al2O3–(Ce0.50Zr0.50)O2 sample.

Table 2

FT-IR in the 1100–1600 cm�1 region for CO2 adspecies formed at CO

adsorption at 25 8C after O550 treatment of catalyst samples

Band wavenumber

(cm�1)

Assignation Literature

value (cm�1)

Da

(cm�1)

1160 Bridged on alumina 1135b �25

1193 Hydrogen carbonate on ceria 1218b 25

1227 Bridged on ceria 1232c 5

1334 Monodentated on ceria 1351b 17

1437 Polydentated on ceria 1462b 25

1490 Monodentated on ceria 1504b, 1497d 14, 7

a Difference between literature and experimental values.b Ref. [37].c Ref. [40].d Ref. [41].

2120 cm�1 depends on the Ce/Zr ratio reflecting different

nature of Pd–support interface. Changes in position of signal

around 2156 cm�1 manifests different contribution of L2 and

Ce4 bands for samples with different Ce/Zr ratio.

From these results it is concluded that ceria stabilizes

preferentially Pd2+ species. However, some portion of ceria

surface sites is not occupied by palladium species in accordance

with CO adsorption data.

Further increase of temperature up to 25 8C (Fig. 2)

diminished the relative intensity of band assigned to Pd2+

species and affected the relative contribution of Pd+. Note, the

position of L1 band was shifted by about 10 cm�1 for PdO/

Al2O3–ZrO2 and PdO/Al2O3(Ce0.33Zr0.67)O2. This blue shift

could be attributed to formation of some intermediate

palladium species between Pd+ and Pd2+ as was also observed

in Ref. [34].

Adsorption of CO at room temperature obviously should

result in appearance of reduced palladium ions as Pd+ due to

partial reduction of Pd2+, because reduction can proceed even at

�50 8C as was shown by our own experiments presented in

Fig. 1. On the other hand one did not expect any changes in PdO

dispersion at room temperature. So, we rather prefer to assign

observed shift of L1 band to appearance of some Pd species as a

result of an ensemble contribution of Pd+ and Pd2+ ions.

For some oxidized catalysts during heating up carbonate

species were detected (Fig. 3), indicating that oxide palladium

species were reduced by CO. The intensity of bands assigned to

CO adsorption on support sites S3 and Ce4 also changed when

temperature was increased.

The formation of carbonate species in the 1100–1600 cm�1

region was monitored at RT (25 8C) after CO adsorption

(5 Torr, 25 8C). The spectra of all samples shown in Fig. 3

develop carbonate bands at 1160, 1193, 1227, 1334, 1437 and

1490 cm�1. The assignment of these bands is presented in

Table 2. The shift of experimental bands from those reported by

Binet et al. [37] on polycrystalline CeO2 could be attributed to

differences in dispersion of Ce species in our samples. Such an

effect of dispersion is clearly indicated for monodentate CO2

species reported for bulk CeO2 [37,40] and that supported on

alumina [41]. As a result of Zr addition one can expect strong

distortion of Ce oxide structure and it could be the reason for

red shift of experimental bands observed for studied samples in

carbonate region. A similar red shift for bands of methanol

adsorbed over Ce–Zr–O mixed oxide in comparison with pure

ZrO2 was observed in Ref. [42].

The presence of CO2 bands reveals that oxidation of CO

already happened at these temperatures. The formation of

G. Perez-Osorio et al. / Applied Catalysis B: Environmental 69 (2007) 219–225 223

Fig. 5. FT-IR spectra of catalysts recorded at 25 8C (after exhibition to 5 Torr of

CO and 40 min evacuation at 25 8C). The samples were pretreated at (O550/H550).

(A) PdO/Al2O3, (B) PdO/Al2O3–ZrO2, (C) PdO/Al2O3–(Ce0.25Zr0.75)O2, (D)

PdO/Al2O3–(Ce0.33Zr0.67)O2, (E) PdO/Al2O3–(Ce0.50Zr0.50)O2, (F) PdO/Al2O3–

(Ce0.67Zr0.33)O2, (G) PdO/Al2O3–(Ce0.75Zr0.25)O2 and (H) PdO/Al2O3–CeO2.

carbonate bands started during heating from �50 to 25 8C.

These carbonate bands are attributed to adsorption of CO2

formed by reaction between PdO and CO adsorbed on it.

Note, that for samples PdO/Al2O3–(Ce0.5Zr0.5)O2 and PdO/

Al2O3–(Ce0.33Zr0.67)O2 the bands with more pronounced

intensity were detected indicating their high activity in CO

oxidation.

3.2. CO adsorption on reduced (O550/H550) catalysts

The spectra of CO adsorbed on reduced catalysts are

shown in Fig. 4. For the majority of samples the O550/H550

treatment produced an intense band at 2103 cm�1 (L0)

attributed to CO adsorbed on Pd0. Indeed, the presence of

marked shoulders L1 and L2 in many cases indicated the

presence of oxidized palladium species being dependent on

the content of Ce. That was particularly the case for the

Pd/Al–Ce0.50Zr0.50 catalyst, where the L0 band was not

detected, but L1 and L2 were clearly observed. It can be

proposed that Pd2+ and Pd+ species were reduced to Pd0 and

then it was reoxidized by oxygen species provided by

surrounding ceria species. The ability of palladium to be

reoxidized in this catalyst may be determined by the unique

high oxygen mobility characterized for mixed Ce/Zr oxides

and particular for those with Ce/Zr ratio close 1 according to

[41,43].

A shoulder at 2111 cm�1 (Ce3) corresponding to Ce3+–CO

interaction [38] was clearly observed for the reduced catalysts

particular for those with high Ce/Zr ratio.

The spectra in Fig. 5 recorded at 25 8C show the same bands

but with low intensity. The L0 band was predominant for all

Fig. 4. FT-IR spectra of catalysts recorded at �50 8C (after introduction of

5 Torr of CO at �120 8C, 40 min evacuation and heating up to �50 8C). The

samples were treated at (O550/H550). (A) PdO/Al2O3, (B) PdO/Al2O3–ZrO2, (C)

PdO/Al2O3–(Ce0.25Zr0.75)O2, (D) PdO/Al2O3–(Ce0.33Zr0.67)O2, (E) PdO/

Al2O3–(Ce0.50Zr0.50)O2, (F) PdO/Al2O3–(Ce0.67Zr0.33)O2, (G) PdO/Al2O3–

(Ce0.75Zr0.25)O2 and (H) PdO/Al2O3–CeO2.

catalysts except for Pd/Al2O3–(Ce0.50Zr0.50)O2 where L1 and

L2 bands still were detected.

The spectrum of carbonate bands in Fig. 6 at room

temperature shows that a less amount of CO2 was formed over

reduced samples, compared to oxidized catalysts. Also in this

case, catalyst Pd/Al2O3–(Ce0.50Zr0.50)O2 develops carbonate

bands with the highest intensity.

Fig. 6. Spectra of catalysts within range 1100–1600 cm�1 (carbonate region)

recorded at 25 8C (after sample exhibition with 5 Torr of CO at �120 8C,

40 min evacuation and heating up to 25 8C, exhibition to 5 Torr of CO and

40 min evacuation at 25 8C). The samples were pretreated at (O550/H550). (A)

PdO/Al2O3, (B) PdO/Al2O3–ZrO2, (C) PdO/Al2O3–(Ce0.25Zr0.75)O2, (D) PdO/

Al2O3–(Ce0.33Zr0.67)O2, (E) PdO/Al2O3–(Ce0.50Zr0.50)O2, (F) PdO/Al2O3–

(Ce0.67Zr0.33)O2, (G) PdO/Al2O3–(Ce0.75Zr0.25)O2 and (H) PdO/Al2O3–CeO2.

The bands assignation are shown in Table 2.

G. Perez-Osorio et al. / Applied Catalysis B: Environmental 69 (2007) 219–225224

Fig. 8. Catalytic activity in CO oxidation for redox catalysts. The data for

upwarding curve correspond to light-off curve with temperature increase and

those for downwarding curve to that with temperature decrease.

3.3. CO oxidation

All catalysts are characterized by a gradual increase of CO

conversion from room temperature up to 150 8C. Fig. 7 shows

catalytic activity in CO oxidation of fresh catalysts (O550) with

different Ce/Zr ratio and data for fresh supports. Values for

supports are shifted about 200 8C to high temperature

compared with those for catalysts. The PdO/Al2O3–ZrO2 and

PdO/Al2O3 catalysts presented relatively low activity in CO

oxidation. Addition of ceria–zirconia mixed oxide enhanced

activity of catalysts, compared with those based on individual

oxide supports. The best catalysts were those situated around

the middle of the Ce–Zr composition range.

It has been reported that ceria plays a key role in CO

oxidation by supplying oxygen to the Pd–Ce interface [41,43].

The fact that the lowest temperature for 50% CO conversion

was achieved with catalysts PdO/Al2O3–(Ce0.67Zr0.33)O2, PdO/

Al2O3–(Ce0.5Zr0.5)O2 and PdO/Al2O3–(Ce0.33Zr0.67)O2 corre-

lates with the highest oxygen mobility for bulk Ce–Zr system

with similar composition reported in [42,44].

The redox O550/H550/O550 pretreatment slightly modified the

catalytic properties of catalysts in CO oxidation, as shown in

Fig. 8. As a result of such a treatment the catalytic activity of

PdO/Al2O3–CeO2 and PdO/Al2O3 catalysts was diminished; on

the other hand, PdO/Al2O3–(Ce0.5Zr0.5)O2 and PdO/Al2O3–

(Ce0.33Zr0.67)O2 catalysts increased their activity. Light-off

profiles of redox catalysts are very similar to those of fresh

catalysts and manifest the same tendency with respect to Ce/Zr

ratio, except that they are shifted to lower temperatures.

There are two possible explanations for enhancement of the

activity of Ce–Zr catalysts after redox treatment. The

dispersion of palladium oxide species could be increased by

consequent reduction/oxidation procedures [34]. Otherwise,

redox treatment can result in formation of new vacancies in

Fig. 7. Catalytic activity in CO oxidation for fresh catalysts and for bare

supports. The data for upwarding curve correspond to light-off curve with

temperature increase and those for downwarding curve to that with temperature

decrease. Data for bare supports are presented by dash line.

ceria containing species improving the oxygen mobility to the

palladium–support interface [18].

In addition, for all cases the catalytic activity obtained at

temperature increase (upwarding curve) was higher than that

with temperature decrease (downwarding curve) indicating

some changes in the active sites of the catalysts in the course

of the catalytic reaction. Note, that the difference in

upwarding and downwarding curves decreased slightly after

redox treatment of samples. The later permits to assume that

the main reason for catalyst deactivation during activity test is

partial sample reduction then the blocking of active sites

with carbon deposits formed at CO oxidation. Indeed, for

sample with the highest oxygen mobility (PdO/Al2O3–

(Ce0.5Zr0.5)O2) as could be expected according to data in

[42] the difference in up-warding and down-warding curves is

practically absent.

4. Conclusions

Preparation of Al2O3–(CexZr1�x)O2 oxides by sol–gel

allowed the incorporation of ceria and zirconia additives with

different Ce/Zr ratio to alumina producing homogeneous

supports with high surface area for palladium containing

catalysts.

As revealed by FT-IR studies of CO adsorbed the oxidation

state of palladium in PdO/Al2O3–(CexZr1�x)O2 catalysts is a

function of the Ce/Zr ratio due to specificity of oxygen mobility

of Ce–Zr mixed oxides. So, in PdO/Al2O3–(Ce0.5Zr0.5)O2

catalyst palladium presents preferably as Pd2+ even after

reductive sample pretreatment due to reoxidation by bulk

support oxygen.

Ceria–zirconia mixed additives enhance the catalytic

activity of palladium oxide catalysts compared with those

based on alumina or alumina with Ce or Zr individually. The

G. Perez-Osorio et al. / Applied Catalysis B: Environmental 69 (2007) 219–225 225

catalysts showing Ce/Zr ratios about 0.4–0.6 presented the

lowest temperature for 50% conversion of CO.

The redox O550/H550/O550 pretreatment increased the

activity of PdO/Al2O3–(Ce0.5Zr0.5)O2 and PdO/Al2O3–

(Ce0.33Zr0.67)O2 catalysts. The highest activity for CO

oxidation was observed for PdO/Al–(Ce0.5Zr0.5)O2 catalyst.

The sol–gel method used for catalysts preparation promotes

the palladium–support interaction, which in turn is very

appropriate for CO oxidation. The observed red shift of the

bands in FT-IR spectra for carbonate region permits to propose

formation of Ce–Zr–O nanocrystallites (islands) on the surface

of alumina prepared by the sol–gel technique where structure of

cerium oxide is hardly distorted by presence of Zr. Similar

conclusion could be done based on the fact that change of

catalytic activity with Ce–Zr ratio for our samples is

comparable with that published for bulk Ce–Zr oxides [39].

Interaction of these nanocrystallites with palladium precursor

produces highly reactive PdO sites for CO oxidation, probably

by increase of dispersion of palladium species and/or by supply

enough oxygen transport in the course of the reaction.

Acknowledgments

The authors acknowledge E. Flores-Aquino, J. Peralta, J.

Palomares and P. Casillas for their technical assistance. The

financial support was obtained by grants from CONACyT

(41331-Y and CO 1-3), UC-MEXUS-CONACYT, U.S.

Material Corridor Initiative and DGAPA-UNAM, IN119602-

3 projects. F.F. Castillon acknowledges support from Depart-

ment of Chemistry of the University of California at Riverside.

G. Perez-Osorio, acknowledges support from CONACyT for

the scholarship 122975.

References

[1] W.B. Williamson, R.G. Silver, J.C. Summers, US Patent 5,672,557 (1997).

[2] T. Sekiba, S. Kimura, H. Yamamoto, A. Okada, in: Proceedings of the

International Forum on Environmental Catalysis, Tokyo, 1993.

[3] J. Noh, O.B. Yang, D.H. Kim, S.I. Woo, Catal. Today 53 (1999) 575.

[4] J.P. Cuif, G. Blanchard, O. Touret, A. Seigneurin, M. Marczi, E. Quemere,

SAE Technical Paper Series, No. 970463 (1997).

[5] J.S. Hepburn, K.S. Patel, M.G. Meneghel and H.S. Gandhi, Soc. Automot.

Eng. Tech. Pap. Ser., Paper No. 941058 (1994).

[6] A. Lemaire, J. Massardier, H. Praliaud, G. Mabilon, M. Prigent, in: A.

Frennet, J.M. Bastin (Eds.), Catalysis and Automotive Pollution Control

III, Elsevier, Amsterdam, Stud. Surf. Sci. Catal. 96 (1995) 97–108.

[7] J.C. Summers, W.B. Williamson, Am. Chem. Soc. Symp. Ser. 552 (1994)

94.

[8] D. Ciuparau, A. Bensalem, L. Pfefferle, Appl. Catal. B 26 (2000) 241.

[9] D.D. Beck, J.W. Sommers, Appl. Catal. B. 6 (1995) 185.

[10] B.H. Engler, D. Lindner, E.S. Lox, A. Shafer-Singlinger, K. Ostgathe, in:

A. Frennet, J.-M. Bastin (Eds.), Catalysis and Automotive Pollution

Control III, Elsevier, Amsterdam, Stud. Surf. Sci. Catal. 96 (1995)

441–460.

[11] R. Lin, M.-F. Luo, Q. Xin, G.-Q. Sun, Catal. Lett. 93 (2004) 139.

[12] Z. Yang, X. Chen, G. Niu, Y. Liu, M. Bian, A. He, Appl. Catal. B 29 (2001)

185.

[13] R.M. Heck, R.J. Farrauto, Catalytic Air Pollution Control: Commercial

Technology, Van Nostrand Reinhold, New York, 1995.

[14] M. Shelef, G.W. Graham, Catal. Rev. Sci. Eng. 36 (1994) 433.

[15] D. Lindner, E.S. Lox, R. Van Yperen, K. Ostgathe, T. Kreuzer, Soc.

Automot. Eng. Tech. Pap. Ser., Paper No. 960802 (1996).

[16] D.R. Monroe, M.H. Krueger, D.D. Beck, M.J. D’Aniello, in: A. Crucq

(Ed.), Catalysis and Automotive Pollution Control II, Elsevier, Amster-

dam, Stud. Surf. Sci. Catal. 71 (1991) 593–616.

[17] A. Trovarelli, in: G.J. Hutchings (Ed.), Catalytic Science Series, Imperial

College Press, UK, 2002.

[18] F. Esch, S. Fabris, L. Zhou, T. Montini, C. Africh, P. Fornasiero, G.

Comelli, R. Rosei, Science 309 (2005) 752.

[19] P. Briot, M. Primet, Appl. Catal. 70 (1991) 87.

[20] N. Bogdanchikova, S. Fuentes, M. Avalos-Borja, M.H. Farıas, A. Bor-

oninc, G. Dıas, Appl. Catal. B 17 (1998) 221.

[21] Y. Bi, G. Lu, Appl. Catal. B 41 (2003) 279.

[22] K. Masuda, T. Sano, F. Mizukami, K. Kuno, Appl. Catal. A 133 (1995) 59.

[23] R. Craciun, W. Daniell, H. Knozinger, Appl. Catal. A 230 (2002) 153.

[24] H. Zhu, Z. Qin, W. Shan, W. Shen, J. Wang, Low temperature oxidation of

CO over Pd/Ce–Ti catalysts with different pretreatments, J. Catal. 233

(2005) 41.

[25] A. Iglesias-Juez, A. Martinez-Arias, M. Fernandez-Garcia, J. Catal. 221

(2004) 148.

[26] M. Fernandez-Garcıa, A. Martınez-Arias, A. Iglesias-Juez, A.B. Hungrıa,

J.A. Anderson, J.C. Conesa, J. Soria, J. Appl. Catal. B 31 (1) (2001) 39.

[27] A. Martınez-Arias, M. Fernandez-Garcıa, A. Iglesias-Juez, A.B. Hungrıa,

J.A. Anderson, J.C. Conesa, J. Soria, Appl. Catal. B 31 (1) (2001) 51.

[28] C. Bozo, N. Guilhaume, E. Garbowski, M. Primet, Catal. Today 59 (2000)

33.

[29] C. Bozo, N. Guilhaume, J.-M. Herrmann, J. Catal. 203 (2001) 393.

[30] D.A. Ward, E.I. Ko, Ind. Eng. Chem. Res. 34 (1995) 421.

[31] A. Barrera, M. Viniegra, P. Bosch, V.H. Lara, S. Fuentes, Appl. Catal. B 34

(2) (2001) 97.

[32] S. Fuentes, N.E. Bogdanchikova, G. Dıaz, M. Peraaza, G.C. Sandoval,

Catal. Lett. 47 (1997) 27.

[33] G. Perez Osorio, S. Fuentes Moyado, V. Petranovskii, A. Simakov, Catal.

Lett. (2006), In press.

[34] H. Tiznado-Vazquez, S. Fuentes, F. Zaera, Langmuir 20 (2004) 10490.

[35] F. Zaera, Int. Rev. Phys. Chem. 21 (2002) 433.

[36] A. Zecchina, E. Escalona Platero, C. Otero Arean, J. Catal. 107 (1987)

244.

[37] C. Binet, M. Daturi, J.C. Lavalley, Catal. Today 50 (1999) 207.

[38] A. Bensalem, F. Bozon-Verduraz, M. Delamar, G. Bugli, Appl. Catal. A

121 (1995) 81–93.

[39] A. Davydov, Molecular Spectroscopy of Oxide Catalysts Surfaces, John

Wiley & Sons, England, 2003, p. 236.

[40] T. Jin, Y. Zhon, G.J. Mains, J.M. White, J. Phys. Chem. 91 (1987) 5931.

[41] J.H. Holles, M.A. Switzer, R.J. Davis, J. Catal. 190 (2000) 247.

[42] P. Fornasiero, T. Montini, M. Graziani, J. Kaspar, A.B. Hungria, A.

Martinez-Arias, J.C. Conesa, Phys. Chem. Phys. 4 (2002) 149.

[43] M. Fernandez-Garcıa, A. Martınez-Arias, L.N. Salamanca, J.M. Coro-

nado, J.A. Anderson, J.C. Conesa, J. Soria, J. Catal. 187 (1999) 474.

[44] M. Boaro, C. de Leitenburg, G. Dolcetti, A. Trovarelli, J. Catal. 193 (2000)

338.