international journal of hydrogen energy 35 (2010) 22472261

Available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/he

Activity and selectivity of CO oxidation in H2 rich stream over the Ag/Co/Ce mixed oxide catalysts
Filiz Balikci Derekaya a,*, Cig dem Guldur b
a b

Department of Advanced Technologies, Institute of Science and Technology, Gazi University, 06500 Ankara, Teknikokullar, Turkey Department of Chemical Engineering, Faculty of Engineering and Architecture, Gazi University, 06570 Ankara, Maltepe, Turkey

article info
Article history: Received 20 October 2009 Received in revised form 26 December 2009 Accepted 29 December 2009 Available online 1 February 2010 Keywords: Silver Cobalt Cerium Carbon monoxide Selective oxidation

abstract
Silver, cobalt and ceria mixed oxide catalysts were prepared at different metal/metal oxide molar ratios by the co-precipitation method, calcined at different temperatures (200  C, 450  C) and tested for the selective CO oxidation reaction in H2 rich gas stream. XRD, XPS, N2 physisorption, SEM and TPR-H2 techniques were used to characterize the catalysts. Catalysts have an average pore diameter in the mesoporous range. Catalysts which were calcined at 200  C had amorphous phase structure. After calcination at 450  C, not only the crystal phase structure but also decrease in BET surface areas of the catalysts and the shift at light off temperatures of the catalysts to the higher temperatures were obtained. The highest activity and selectivity was obtained from the catalysts calcined 200  C which were 50/50 AgCo and 50/50 CoCe mixed oxide catalysts, respectively, which did not loose their activity and selectivity in a reaction period of 5800 min. 2010 Professor T. Nejat Veziroglu. Published by Elsevier Ltd. All rights reserved.

1.

Introduction

Purifying H2 rich fuel by a selective oxidation of a trace amount of CO over different metal/metal oxide catalysts, was investigated by several researchers. In these studies, different metals such as Pt, Ru, Rh, Au, Ag, Cu and different supports such as Al2O3, SiO2, ZrO2, activated carbon and CeO2 were used in the catalyst structure. Among them, supported Pt and Au catalysts were the most extensively studied ones. When the gold based catalysts were used a decrease in the selectivity was often observed at temperature values above 80  C due to the competitive oxidation of hydrogen which was the opposite for the Pt-based catalysts. In addition, the catalytic activities of gold catalysts are dependent on the size of the gold particles, the supports and the preparation conditions. On the other hand, Pt-based catalysts are more robust and less sensitive to preparation procedures [1].

Both low temperature CO oxidation and selective CO oxidation in a H2 rich gas stream were studied with the catalysts that include Ag in their structure [27]. According to these studies, the activity of the catalyst was increased by increasing the Ag molar ratio in the catalyst structure [47]. Since the cobalt oxide had an increasing effect on the activities and surface area of the catalysts, it was chosen as a catalyst support or an active component for both low temperature CO oxidation and selective CO oxidation reactions in a H2 rich gas stream [813]. Different supports were used for the catalysts which were used in CO oxidation reactions. Supports such as TiO2, ZrO2, CeO2 are active in the reaction as a source of oxygen for the reaction, while Al2O3, SiO2, MgO are inert. In recent years, CeO2 was the most studied one. CeO2 is well known to have a high oxygen exchange capacity, which is related to the capacity of cerium to change oxidation states reversibly between Ce (IV)

* Corresponding author. Tel.: 90 3122023701; fax: 90 3122023710. E-mail addresses: lizb@gazi.edu.tr (F. Balikci Derekaya), cguldur@gazi.edu.tr (C. Guldur). 0360-3199/$ see front matter 2010 Professor T. Nejat Veziroglu. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijhydene.2009.12.177

2248

international journal of hydrogen energy 35 (2010) 22472261

and Ce (III) by receiving or giving up oxygen. CO oxidation is thought to proceed at the metal-support interfacial perimeter. The reduction of CeO2 has been demonstrated to proceed via a surface reaction that involves the redundant CO. Hence, the oxygen from the support oxidizes CO adsorbed on the metal particle at the interface, and the oxygen vacancies are then replenished by O2 [14]. After reviewing the studies in the literature, catalysts were prepared at different metal/metal oxide molar ratios; (50/50 (mol.%) AgCo, 50/50 (mol.%) AgCe, 50/50 (mol.%) CoCe and 25/25/50 (mol.%) AgCoCe mixed oxide catalysts) for the CO oxidation and selective CO oxidation reaction in a H2 rich gas stream in this study. To nd out the effects of the pre-treatment conditions on the characteristic and catalytic properties of the catalysts, all samples were calcined at two different temperatures. At the end of the study, the effects of the catalysts composition and pre-treatment condition on the characteristic and catalytic properties of the catalysts were determined.

a voltage of 12.5 kV. Charging effects were corrected by adjusting the C 1s peak to a position of 284.5 eV. The samples were pressed into thin discs and mounted on a sample rod placed in a pre-treatment chamber. The spectra of Ag 3d, Co 2p, O 1s and Ce 3d levels were recorded.

2.2.4.

Scanning electron microscopy (SEM)

Catalyst morphology was examined using a JEOL/JSM-6335F scanning electron microscope. The electron gun was operating at a voltage of 12 kV for the collection of all images.

2.2.5.

Temperature programmed reduction (TPR-H2)

2.
2.1.

Experimental
Catalyst preparation

Temperature programmed methods were carried out by a CHEMBET-3000 system. The catalyst (z0.150 g) was loaded into a U-shape quartz reactor. The catalyst samples were pretreated with the ow of helium at 200  C (or 450  C, depending on the calcination temperature) for 0.5 h. After cooling them to room temperature in the helium ow, the TPR measurement was then performed using an H2N2 mixture with a composition of 595% respectively with a ow rate of 50 ml/min 5% H295% N2 ow while the reactor was heated from room temperature to 650  C with a heating rate of 10  C/ min.

The catalysts were prepared by the co-precipitation method at different metal/metal oxide (molar ratios). AgNO3 (Aldrich, 99.8%), Co(NO3)2.6H2O (Fluka, 99.0%) and Ce(NO3)3.6H2O (Sigma, 99%) at desired molar ratios were dissolved in distilled water. The total concentration of the metals in the combined aqueous solutions was 0.1 M. The metal salt solution was put in to the stirrer. The Na2CO3 (1 M) solution was added to the metal salt solution to adjust the pH of the solution to 8. The precipitates were aged for 3 h at a pH value of 8 and ltered and washed with hot distilled water several times in order to remove excess ions. They were dried overnight at 110  C. The catalysts were calcined at 200  C for 3 h and at 450  C for 3 h under the ow of air.

2.3.

Catalytic activity studies

2.2. 2.2.1.

Characterization X-Ray diffraction (XRD)

The X-Ray diffraction (XRD) technique was used to identify the phases present in the synthesized samples. A rigaku rotating anode X-Ray diffractometer system generating CuKa radiation was used to obtain the XRD patterns.

Catalytic activity and selectivity measurements were carried out in a xed bed quartz tubular reactor using 2535 mg of catalyst. At CO oxidation studies, feed with a composition of 1 vol.% CO, 21 vol.% O2 and the remainder He, was used and the temperature range of the reactor was changed from room temperature to 200  C. In the selectivity measurements, feed consisting of 1 vol.% CO, 2 vol.% O2, 85 vol.% H2 and the remainder He was passed through the catalyst bed. Selectivity measurements were carried out in the temperature of 110 210  C. The ow rate of the feed gas was 25 ml/min. The analysis of the reactor efuent was performed by an on-line Perkin Elmer CLARUS 500 gas chromatograph equipped with a thermal conductivity detector (TCD). The chromatograph column packing was carbosphere and the column temperature was maintained at 50  C. (The catalytic activity system, which was shown by Guldur et al. [7] was used). The CO conversion was calculated based on the carbon monoxide consumption: CO conversion% CO2 100 COin (1)

2.2.2.

Surface areas and pore volumes

The textural properties of samples degassed at 200  C (or 450  C) were studied by N2-sorption measurements at liquid nitrogen temperatures in a Quantochrome Autosorp 1C. BET surface area values were calculated in the relative pressure range P/P0 of 0.050.30. Calculation of pore sizes followed the method of BarretJoynerHalenda (BJH) according to implemented software routines using Halsey thickness equation to relate the thickness of the adsorbed layer to the relative pressure. All calculations are based on the desorption model.

where COin is the inlet carbon monoxide and CO2 is the concentration of carbon dioxide as a product. The O2 conversion was based on the oxygen consumption:

O2 conversion%

O2 in O2 out 100 O2 in

where [O2]in is the inlet oxygen and [O2]out is the unreacted oxygen. The selectivity is dened as the fraction of oxygen that is used up to oxidize carbon monoxide to carbon dioxide: selectivity% 0:5CO2 100 O2 in O2 out

2.2.3.

X-ray photoelectron spectroscopy (XPS)

XPS analyses were performed with an SPECS SAGE-150 spectrometer, using MgKa radiation (1486.6 eV) operating at

international journal of hydrogen energy 35 (2010) 22472261

2249

2.4.

Deactivation studies

The deactivation test was carried out at 150  C for 5800 min. Feed consisting of 1 vol.% CO, 2 vol.% O2, 85 vol.% H2 and the remainder He was used in the deactivation system and the CO oxidation activity and selectivity to CO oxidation of the catalysts were monitored as a function of time.

3.

Results and discussion

Catalysts were characterized by using different techniques. N2 physisorption measurements were made in order to obtain the surface areas, pore volumes and pore diameters of the catalysts. XRD technique was used in order to obtain the phases present in the catalysts. SEM results showed the surface morphology of the catalysts. XPS studies revealed the oxidation states of the catalysts. Temperature programmed reduction gave us the behavior of the catalysts with H2. Finally, the activity measurements of the catalysts to CO oxidation reactions illustrate to us the relationship between the characteristic properties of the catalysts with activity. The results are discussed in the following sections.

3.1. 3.1.1.

Catalyst characterization Nitrogen physisorption measurements

The specic surface area, pore volumes and average pore diameters of the catalysts are shown in Table 1. All values were obtained from the Quantochrome Autosorp 1C instrument. Table 1 shows the sharp decrease in the BET surface areas of the catalysts with an increase in calcination temperature. BET surface areas of the catalysts were decreased sharply after calcination at 450  C. The highest decrease was observed for the 50/50 (mol.%) AgCe mixed oxide catalyst. The decrease in the surface area was due to the decomposition of the silver oxide into the metallic silver phase. The ratio of decrease for the 50/50 (mol.%) AgCe mixed oxide catalyst is greater than the 50/50 (mol.%) CoCe mixed oxide catalyst because of the decomposition of the silver oxide phase. 50/50 (mol.%) CoCe and 25/25/50 (mol.%) AgCoCe mixed oxide catalysts calcined at 450  C gave the same BET

surface area values since silver is in the metallic silver phase in 25/25/50 (mol.%) AgCoCe mixed oxide catalyst. The decrease in the surface area of the catalyst was also due to the decomposition of the porous structure of the catalysts. The small pores are decomposed with an increase in calcination temperature to form the biggest ones. The pore volume values are conrmed by this decomposition of the porous structure. Although the micro and meso pore volume values are small for the catalysts calcined at 200  C, the difference between the total pore volume and micro and meso pore volume values are large for the catalysts calcined at 450  C. The macropores were formed after calcination at 450  C. The highest BET surface area values were obtained from the 50/50 (mol.%) CoCe mixed oxide catalyst which was calcined at 200  C. These biggest surface area values were mostly due to the Co3O4 phase in the catalyst structure. The adsorptiondesorption isotherms of the catalysts t the Type III and Type V isotherms in BDDT classication which were shown at Fig. 1. Type III and V isotherms are characteristic of weak gassolid interactions. The Type III isotherm is given by a nonporous or macroporous solid and the type V isotherm is given by a mesoporous or microporous solid. The 50/50 (mol.%) AgCe mixed oxide catalyst showed the Type III adsorption isotherm behavior. The isotherm behavior of the 50/50 (mol.%) AgCo mixed oxide catalyst, 50/ 50 (mol.%) CoCe mixed oxide catalyst and 25/25/50 (mol.%) AgCoCe mixed oxide catalyst t Type V isotherm in the BDDT classication. The hysterisis loop of the adsorption desorption isotherm of the catalysts calcined at 200  C changed in between P/P0 0.40.8. For the catalysts calcined at 450  C, the hysterisis loop of adsorptiondesorption isotherm started below P/P0 0.1. and the area of hysterisis loop decreased, indicating part of the mesopores had changed, thereby leading to a decrease in the pore volume and surface area. The beginning point of the P/P0 was increased by increasing Co3O4 ratio in the catalyst structure. The value of the P/P0 shows us the starting point of the capillary condensation. Since the pore structure of the catalysts differ from each other, the value of the P/P0 and shape of the hysterisis loop changed from one catalyst to another. Catalysts of which the value of the P/P0 was low showed higher catalytic activity in CO oxidation in comparison to the catalysts where the P/P0

Table 1 The physical properties of the catalysts obtained form N2-sorption measurements. Catalysts Specic Surface Area m2/g Single Point BET 200  C
50/50 AgCo mixed oxide 50/50 AgCe mixed oxide 50/50 CoCe mixed oxide 25/25/50 AgCoCe mixed oxide 149 121 159 126

Multi Point BET 200  C

V(mikromezo) (liquid N2 cm3/g) 200  C

VToplam (liquid N2 cm3/g) 200  C

Average Pore Diameter nm 200  C

450  C
29 19 41 41

450  C
28 20 38 38

450  C
0.20 0.08 0.43 0.20

450  C
0.31 0.17 1.23 0.36

450  C
2.5, 31 3.1 2.7, 34 2.5

2250

international journal of hydrogen energy 35 (2010) 22472261

Fig. 1 The N2 adsorptiondesorption isotherms of the catalysts.

international journal of hydrogen energy 35 (2010) 22472261

2251

value was high. This result relates to the surface area because catalysts with larger a hysterisis loop have a bigger surface area. Fig. 2 shows the pore size distribution patterns of the catalysts. The average pore diameters of the catalysts were changed in the mesoporous scale (2 nm<daverage < 50 nm). Catalysts calcined at 200  C have an average pore diameter value between 2.1 nm daverage 2.5 nm, and 50/50 (mol.%) AgCo mixed oxide catalyst and 50/50 (mol.%) CoCe mixed oxide catalyst have a second average pore diameter, which is 19 nm. Same catalysts calcined at 450  C have an average pore 3.5 nm, and 50/50 diameter between the 2.4 nm daverage (mol.%) AgCo mixed oxide catalyst and 50/50 (mol.%) CoCe mixed oxide catalyst have a second average pore diameter of 31 nm and 34 nm, respectively, in addition to the rst one. The same relationship between the pore diameter and Co3O4 has been reported by Storsaeter et al. [15]. They observed that the average pore diameter of the g-Al2O3, SiO2 and TiO2 supported Co and CoRe catalysts was increased with increasing Co3O4 content. Results showed that average pore diameter of the catalysts slightly increased with an increase in calcination temperature. This result was conrmed by the Jung et al. [16]. They found that with an increase in calcination temperature, the area with the hysterisis loop decreased, indicating that the micropores collapsed suddenly, thereby leading to a decrease in pore volume and BET surface area whereas average pore diameter increased. Results of average pore diameter corresponding to 50/50 (mol.%) CoCe mixed oxide catalyst conrmed the surface area result, since it gave the highest surface area result. The second average pore diameter obtained form these catalysts might be responsible for the high surface area result. The increase in the average pore diameter was due to the decomposition of the small pores which can be seen from the pore volume results shown in Table 1. In addition, a change in the pore size distribution pattern was found with an increase in the calcination temperature of the catalysts. After calcination at 450  C the pore volumes of the catalyst (50/50 (mol.%) AgCe mixed oxide, 25/25/50 (mol.%) AgCoCe mixed oxide) decreased markedly. The small pores in the catalyst structure were decomposed and larger pores were formed at the high temperature calcination.

calcined at 450  C were consistent with the other studies in the literature [1723]. When the calcination temperature increased to 450  C the silver oxide phases in the catalysts were decomposed to metallic silver. The 2q values of the metallic Ag peaks were observed at similar 2q values in the other studies [1820].

3.1.3.

X-ray photoelectron spectroscopy (XPS)

3.1.2.

X-Ray diffraction (XRD)

Fig. 3 shows the X-Ray spectrum of the catalysts. The 50/50 (mol.%) AgCe mixed oxide, 50/50 (mol.%) CoCe mixed oxide and 25/25/50 (mol.%) AgCoCe mixed oxide catalysts calcined at 200  C exhibited amorphous structure (Fig. 3A). However, 50/50 (mol.%) AgCo mixed oxide catalyst gave peaks which were due to the Co3O4 and AgCoO2 phases which were shown at Guldur et al. [7]. The AgCoO2 peaks were observed at 2q 27.604 , 39.241 , 65.9oand Co3O4 peaks were observed at 2q 36.853 and 65.238 . On the other hand the catalysts calcined at 450  C exhibited the crystal phase structure (Fig. 3B). The Co3O4, CeO2 and metallic silver peaks appeared in the catalysts calcined at 450  C. The Co3O4 peaks and metallic silver peaks were very close and Co3O4 peaks prevented the appearance of the metallic silver peaks. Therefore, the intensities of the metallic silver peaks were poor. The X-Ray results of the catalysts

The XPS measurements were made to determine the oxidation state of the catalysts since the catalysts calcined at 200  C exhibited an amorphous phase structure in the XRD analysis. Table 2 shows the binding energies of the Ag3d, Co2p, O1s and Ce3d photoelectron peaks. The C 1s at 284.5 eV reference peak was used in the analysis. The Ag3d5/2 binding energies of 50/50 (mol.%) AgCo mixed oxide, 50/50 (mol.%) ACe mixed oxide and 25/25/50 (mol.%) AgCoCe mixed oxide catalysts were 368.3 eV, 368.1 eV and 369.1 eV respectively. The Ag3d peaks can be compared with 367.3 eV for Ag2CO3, 367.7 eV for Ag2O, 367.8 for AgO and 368 eV for Ag/Al2O3 [2426]. The Ag3d binding energies obtained in this study, which were shifted to higher temperatures, are slightly greater than the other studies. This difference might be due to the amorphous structure of the catalysts which did not show any peaks in the XRD studies. The Co2p3/2 binding energy is 780 eV, 779.3 eV and 781 eV for the 50/50 (mol.%) AgCo mixed oxide, 50/50 (mol.%) CoCe mixed oxide and 25/25/50 (mol.%) AgCoCe mixed oxide catalysts, respectively. The Co2p1/2 binding energies of 50/50 (mol.%) AgCo mixed oxide, 50/50 (mol.%) CoCe mixed oxide and 25/25/50 (mol.%) AgCoCe mixed oxide catalysts are 795.2 eV, 794.6 eV, 796.5 eV respectively. The Co2p peaks can be compared with 779.4 eV for Co3O4 in CuCo oxide, 779.9 eV for Co3O4 in Co3O4 catalysts [27,28]. The Co2p3/2 binding energy which was obtained from 50/50 (mol.%) CoCe mixed oxide catalyst is very close to the reference studies. According to this result, the Co might be in the Co3O4 phase in the catalysts structure although the catalyst gave amorphous phase structure in the XRD studies. The Co2p3/2 binding energy of 25/ 25/50 (mol.%) AgCoCe mixed oxide catalyst slightly different than 50/50 (mol.%) CoCe mixed oxide catalyst and other catalysts in the reference studies. This difference might be due to the composition of the catalyst. The silver in the catalyst structure might shift the binding energy to the right. The O1s binding energies of 50/50 (mol.%) AgCo mixed oxide, 50/50 (mol.%) AgCe mixed oxide, 50/50 (mol.%) CoCe mixed oxide and 25/25/50 (mol.%) AgCoCe mixed oxide catalysts are 530 eV, 529.3 eV, 529 eV and 531 eV respectively. The O1s peaks can be compared with 532.1 eV and 529.2 eV in Ag2O catalyst; 529.7 eV in Co3O4 catalyst and 529.4 eV in CeO2 catalyst [25,26,28,29]. The O1s binding energy of the 50/50 (mol.%) AgCe mixed oxide catalyst is very close to the reference studies [26,28,29]. On the other hand, the highest binding energy was obtained form the 25/25/50 (mol.%) AgCoCe mixed oxide catalyst which may be due to the composition of the catalysts. Since 25/25/50 (mol.%) AgCoCe mixed oxide catalyst has three metal components in its structure while the 50/50 (mol.%) AgCe mixed oxide and 50/50 (mol.%) CoCe mixed oxide catalysts have two metal components in their structures.

2252

international journal of hydrogen energy 35 (2010) 22472261

Fig. 2 The pore size distribution patterns of the catalysts.

international journal of hydrogen energy 35 (2010) 22472261

2253

Fig. 3 XRD spectrum of the mixed oxide catalysts (A: catalysts calcined at 200 8C, B: catalysts calcined at 450 8C, (C: CeO2, :: Ag2O, >: metallik Ag, -: Co3O4)).

The Ce3d5/2 binding energy of the 50/50 (mol.%) AgCe mixed oxide catalyst is 882.5 eV, 889.4 eV, 901.25 eV and 902.4 eV. The Ce3d5/2 binding energy of the 50/50 (mol.%) CoCe mixed oxide catalyst is 881.97 eV, 887.81 eV and 897.79 eV. The Ce3d5/2 binding energy of the 25/25/50 (mol.%) AgCoCe mixed

oxide catalyst is 883.75 eV, 890.88 eV, 900 eV and 902.31 eV. The Ce3d3/2 peak has a binding energy of 916.67 eV at the 50/50 (mol.%) AgCe mixed oxide catalyst, 916.04 eV at the 50/50 (mol.%) CoCe mixed oxide catalyst and 918.03 eV at the 25/25/ 50 (mol.%) AgCoCe mixed oxide catalyst. The Ce3d peaks can

Table 2 Binding energies obtained from the XPS studies (BE[[]eV). Catalysts
50/50 AgCo mixed oxide 50/50 AgCe mixed oxide 50/50 CoCe mixed oxide 25/25/50 AgCoCe mixed oxide

Ag3d5/2
368.3 368.1 369.1

Ag3d3/2
374.2 374.2 375.2

O1s
530 529 529 531

Ce3d5/2

Ce3d3/2

Co2p3/2
780

Co2p1/2
795.2 794.6 796.5

883, 889,898, 901, 902 882, 888, 898 884, 891, 900, 902

916.7 916 918

779.3 781

2254

international journal of hydrogen energy 35 (2010) 22472261

be compared with 916.8 eV for Ce(IV) and 882.6 eV for Ce(III); 883.0 eV, 889.6 eV, 899 eV, 901.2 eV, 908 eV, 917 eV for Ce(IV) in the CeO2 catalyst [30,31]. The peaks observed at XPS measurement are consistent with the reference studies. According to the XPS measurements results the Ag3d5/2 peaks were obtained from Ag 1; the Co2p3/2 peaks were obtained from Co 3 and Ce3d5/2 and Ce3d3/2 peaks were obtained from Ce (IV) in the catalysts. As a result, the oxidation state of the surface is Ag (1), Co (3) and Ce (IV). Table 3 shows the atomic percentages of the Ag, Co, Ce atoms. The Ag/Co, Ag/Ce, Co/Ce and Ag/Co/Ce ratios were not consistent with the theoretical values. While the theoretical Ag/Co ratio is 2/3, the experimental Ag/Co value was obtained as 1.41. Theoretical Co/Ce is 3/1 in the 50/50 CoCe mixed oxide catalyst; this value was obtained as 2.79 in the XPS measurement. This experimental result was very close to theoretical value. The theoretical Ag/Ce ratio in the 50/50 Ag Ce mixed oxide catalysts is 2/1. This value was obtained as 0.93 in the XPS measurements. This showed that silver was greater than the cerium on the surface. In spite the theoretical Ag/Co/Ce ratio in the 25/25/50 AgCoCe mixed oxide catalyst is 2/3/1, the experimental value is 1.23/2.08/1 in the XPS measurement. This result shows that the ratio of the Ag and Co atoms on the surface is small. The atomic ratio of the Co/Ce in the 50/50 CoCe mixed oxide catalysts is close to the theoretical value.

3.1.4.

Scanning electron microscopy (SEM)

SEM photographs corresponding to AgCo mixed oxides were presented in Fig. 4 which shows the differences of the surface of the catalysts after calcination at different temperatures. The SEM image of the catalyst calcined at 200  C calcination temperature shows that the particles are closer each other, the sized of the gaps between agglomerates are small and the shape of the particles are same. When the catalyst was calcined at 450  C, the major changes which affected the catalyst activity and selectivity were obtained. The sizes of the gaps between agglomerates increased, the shape and the homogeneity of the particle changed after calcination of the catalyst at 450  C. The changes of the surface after calcination at 450  C were mainly due to the sintering of particles.

Fig. 4 SEM photographs of the AgCo mixed oxide catalysts (A: calcined at 200 8C, B: calcined at 450 8C).

3.1.5.

Temperature programmed reduction (TPR-H2)

inuence of the composition of the catalyst and calcination temperature on the reduction properties of the catalysts was also studied by the TPR. Fig. 5 shows the TPR proles of the mixed oxide catalysts. The TPR proles of the catalysts calcined at different temperatures showed the biggest difference. The TPR proles of the 50/50 (mol.%) AgCo mixed oxide

Temperature programmed reduction has been used extensively to characterize the reducibility of the catalysts. The

Table 3 Surface atomic compositions of the catalysts calcined at 200 8C. Catalysts
50/50 AgCo mixed oxide 50/50 AgCe mixed oxide 50/50 CoCe mixed oxide 25/25/50 AgCoCe mixed oxide

Ag
59 48

Co
41

Ce

52 74 26 23

29

48

Fig. 5 Temperature programmed reduction (TPR-H2) prole of the mixed oxide catalysts.

international journal of hydrogen energy 35 (2010) 22472261

2255

catalysts were very similar to that of the pure Co3O4 catalyst. Examples of the TPR-H2 proles of the catalysts that include Co3O4 in their structures were found in the literature. Teodorescu et al. [32] obtained two reduction peaks at 320  C and 390  C from a Co3O4 catalyst. The rst peak was responsible for the decomposition of Co3O4 to CoO and the second peak was responsible for the decomposition of CoO to Co phase, and Co3O4 decomposed to CoO at 500  C. The Co/Al2O3 catalyst that was investigated by the Luo et al. [18] showed different reduction behavior. TPR proles showed that a catalyst composed of four different Co phases had reduction peaks at 400  C, 480  C, 600  C and 690  C. The 50/50 (mol.%) AgCo mixed oxide catalysts both calcined at 200  C and 450  C gave similar TPR proles. While the TPR proles of the 50/50 (mol.%) AgCo mixed oxide catalyst calcined at 200  C has a reduction peak between 108  C and 650  C which has a shoulder at 145  C, the 50/50 (mol.%) AgCo mixed oxide catalyst calcined at 450  C has a reduction peak between 198  C and 552  C which has very small shoulder at 170  C. The 50/50 (mol.%) AgCe mixed oxide catalyst calcined at 200  C gave a very big reduction peak but the 50/50 (mol.%) AgCe mixed oxide catalyst calcined at 450  C gave a small reduction peak. The TPR prole of the 50/50 (mol.%) AgCe mixed oxide catalyst calcined at 200  C showed the broad reduction peak in the range of 83 C261  C with a maxima at 276  C and the 50/50 (mol.%) AgCe mixed oxide catalyst calcined at 450  C showed a small reduction peak in the range of 85 C278  C with a maxima at 196  C. The reduction behavior of the 50/50 (mol.%) AgCe mixed oxide catalyst calcined at 450  C was very similar to the pure CeO2 catalyst because the silver oxide was decomposed to metallic silver at 450  C and the metallic silver was not reduced under H2 atmosphere. Two major reduction peaks observed over the pure CeO2 catalyst calcined at a temperature higher than 450  C in the literature [3335]. Tang et al. [33] observed that pure CeO2 catalysts gave two reduction peaks at 430  C and 900  C. Zhu et al. [34] studied also TPR with pure CeO2 catalyst and they observed two reduction peaks at 500  C and 900  C. The low temperature peak is due to the reduction of CeO2 surface oxygen and the high temperature peak is due to the reduction of bulk oxygen. Xiaoyuan et al. [35] reported that TPR proles of the pure CeO2 catalyst have two major reduction peaks at 450  C and 575  C. The TPR proles of the 50/50 (mol.%) CoCe mixed oxide catalyst calcined at 200  C shows the broad reduction peak in the range of 98 C700  C with a shoulder at 160  C. The TPR prole of the 50/50 (mol.%) CoCe mixed oxide catalyst calcined at 450  C shows a reduction peak in the range of 197 C556  C with two maxima at 366  C and 480  C. The area of the reduction peak which was the summation of the reduction of CeO2 phase and Co3O4 phase decreased when the calcination temperature was increased from 200  C to 450  C. The CeO2 phase present in the 50/50 (mol.%) CoCe mixed oxide catalyst calcined at 450  C had an inuence on the reduction behavior and it caused the decrease in the reduction amount of the catalyst. Kang et al. [36] reported that the CoOx/ CeO2 catalyst calcined at 650  C has two reduction peaks in the range of 227 C427  C with a maxima at 327  C. Liotta et al. [17] observed that the 50/50 CoOx/CeO2 catalyst calcined at 750  C has two reduction peaks between 290  C and 465  C. According to these studies, the reduction of 50/50 (mol.%) CoCe was

taken as a step wise process via Co3 / Co2 / Co0. The TPR proles of the 50/50 (mol.%) CoCe mixed oxide catalyst showed similar results obtained in literature. The TPR prole of the 25/25/50 (mol.%) AgCoCe mixed oxide catalyst calcined at 200  C shows a reduction prole in the range of 125 C650  C with a small shoulder at 171  C. The 25/25/50 (mol.%) AgCoCe mixed oxide catalyst calcined at 450  C showed a reduction peak between 171  C and 474  C with a maxima at 392  C. The reduction peak area of the 25/25/50 (mol.%) AgCoCe mixed oxide catalyst calcined at 450  C is smaller than the 50/50 (mol.%) CoCe mixed oxide catalyst which was calcined at 450  C. Since the cobalt oxide ratio was decreased from 50% to 25% a smaller reduction peak area was obtained. However, the reduction peak area of the 25/25/50 (mol.%) AgCoCe mixed oxide catalysts was smaller than the reduction peak area of the 50/50 (mol.%) CoCe mixed oxide catalysts. This may be due to the interaction between silver and cerium oxide in the catalysts structure of the 25/25/50 (mol.%) AgCoCe mixed oxide catalysts.

3.2.

Catalytic activity measurements

The CO oxidation activity studies were done in order to compare the results obtained from selective CO oxidation in H2 rich stream studies. The feed used has a composition of 1 vol.% CO, 21 vol.% O2 with the remainder consisting of He. The activity of the catalysts was monitored from room temperature to 200  C. Both of the results obtained from the CO oxidation activity tests and selective CO oxidation activity tests were used to determine the most suitable catalysts that would be used in the deactivation study. Table 4 shows the temperatures at which 50% CO conversion and 100% CO conversion were obtained over the indicated catalysts. Activity of the catalyst increased with an increase in reaction temperature. The highest activities were obtained for the 50/ 50 (mol.%) AgCo and 50/50 (mol.%) AgCe mixed oxide catalysts calcined at 200  C. Activities of these two catalysts are very similar to each other, although the compositions of the Ag2O in these two catalysts were different. Metals such as Ag, Co, Ce are in the oxide phase in these two catalysts. Results of the XRD studies showed that catalysts calcined at 200  C have an amorphous structure. The similarity of the activity results of these two catalysts was mainly due to the same Ag2O molar ratio. The activity curve of the 50/50 (mol.%) AgCe mixed oxide catalyst calcined at 200  C is a little above the activity curve of the 50/50 (mol.%) AgCo mixed oxide catalyst calcined at 200  C which showed the effect of the cerium oxide. Ceria (CeO2) released oxygen by the reductionoxidation process [3740]. However, activity of pure CeO2 and pure Ag2O is reported to be quite low between room temperature and 200  C [41,5]. Activity results showed that interaction between Ag2O and CeO2 lead to increase in activity. Both 50/50 (mol.%) AgCo mixed oxide catalyst calcined at 200  C and 50/50 (mol.%) AgCe mixed oxide catalyst calcined at 200  C showed amorphous structures which were determined from the XRD measurement and both of these catalysts showed similar activities. Activity of the 25/25/50 (mol.%) AgCoCe mixed oxide catalyst calcined at 200  C is lower than the activity of the 50/ 50 (mol.%) AgCo and 50/50 (mol.%) AgCe mixed oxide

2256

international journal of hydrogen energy 35 (2010) 22472261

Table 4 The conversion temperatures of the catalysts for the low temperature CO oxidation (2540 mg catalyst, S.V. 45 000 hL1, 1 vol.% CO, 21 vol.% O2 and the rest being He). Catalysts 50% conversion temperature 200 C
50/50 AgCo mixed oxide 50/50 AgCe mixed oxide 50/50 CoCe mixed oxide 25/25/50 AgCoCe mixed oxide 103  C 101  C 122  C 145  C


100% conversion temperature 200  C

450 C
185  C 173  C T > 200  C T > 200  C

450  C
T > 200  C T > 200  C T > 200  C T > 200  C

catalysts calcined at 200  C. 25/25/50 (mol.%) AgCoCe mixed oxide catalyst calcined at 200  C can be compared with the 50/ 50 (mol.%) AgCe mixed oxide catalyst calcined at 200  C since the molar ratio of the cerium oxide in the two catalysts were the same, but Ag2O molar ratio was decreased by 50%. Decrease in the activity of the 25/25/50 (mol.%) AgCoCe mixed oxide catalyst calcined at 200  C might be due to the decrease in the Ag2O molar ratio in the catalyst structure. Gulari et al. [5] reported that the 50% conversion temperature of the 10% silver 10% cobalt oxide impregnated activated carbon catalyst is around 140  C while the 50% conversion temperature of the 50/50 silvercobalt composite oxide catalyst calcined at 200  C is around 38  C. The activity curves of the catalysts shifted to higher temperatures by increasing calcination temperature of the catalysts. 50/50 (mol.%) AgCe and 50/50 (mol.%) AgCo mixed oxide catalyst calcined at 450  C exhibit better activity than the 50/50 (mol.%) AgCo mixed oxide catalyst calcined at 450  C after reaction temperature of 150  C. The XRD spectrum of the catalyst showed that silver was in the metallic silver phase in the both of the catalyst structures. The main difference was due to CeO2 phase. The ceria (CeO2) acted as a thermal stabilizer and supported oxygen to the 50/50 (mol.%) AgCe mixed oxide catalyst calcined at 450  C. The highest decay was obtained from the 50/50 (mol.%) AgCoCe mixed oxide catalyst calcined at 450  C. Decrease in the silver molar ratio in the catalyst composition and decomposition of the Ag2O into metallic silver phase may lead to the decrease in the activity of the catalyst. Gulari et al. [5] conrmed that metallic silver was an inactive catalyst for the CO oxidation reaction. Also both 50/50 (mol.%) AgCo and 50/50 (mol.%) AgCe mixed oxide catalysts which were calcined at either 200  C or 450  C were reduced between 100  C and 200  C under H2 atmosphere according to the TPR measurements. But, 25/25/50 (mol.%) AgCoCe mixed oxide catalyst calcined at 450  C began to reduce after about 200  C. According to this result surface oxygen was not used until around 200  C. Therefore, 25/25/50 (mol.%) AgCoCe mixed oxide catalyst calcined at 450  C showed the lowest performance for the CO oxidation compared to the other catalysts. The 50/50 (mol.%) CoCe mixed oxide catalysts both calcined at 200  C and 450  C gave lower activity in comparison with the 50/50 (mol.%) AgCo amd 50/50 (mol.%) AgCe mixed oxide catalysts although the surface areas of the 50/50 (mol.%) CoCe mixed oxide catalysts were high. However, several studies showed that the Co3O4/ CeO2 catalyst showed good performance for the CO oxidation

reaction [8,11,42]. Calcination temperature has an important effect on the catalytic activity. Ko et al. reported that an increase in the calcination temperature resulted in a rapid decrease in CO activity and selectivity [13]. Catalytic activity of the catalyst which was calcined at the optimum calcination temperature mostly depend on the stable state of the catalyst solid solution [16]. Since the main goal of this study was to prepare catalysts in order to remove CO which is in the fuel fed into the fuel cell, all the synthesized catalysts were tested for the selective CO oxidation reaction using the H2 rich stream. The selectivities of the catalysts were as important as their CO oxidation activities. CO exists with a high amount of H2 (z85%) in the reformer gas. Therefore, the catalytic activity of the catalyst in oxidizing CO in a H2 rich stream and keeping its activity and selectivity for a long time is very important. A gas mixture containing 1 vol.% CO, 2 vol.% O2, 85 vol.% H2 and remaining being He, was used as a feed for the selectivity measurements. The reactor was heated from 110  C to 210  C and the efuent gas was analyzed by the gas chromatograph at desired temperature intervals. The experiments for the selective oxidation of CO were carried out in the presence of excess hydrogen over the 50/50 (mol.%) AgCo mixed oxide, 50/50 (mol.%) AgCe mixed oxide, 50/50 (mol.%) CoCe mixed oxide and 25/25/50 (mol.%) AgCo Ce mixed oxide catalysts. The CO conversion over mixed oxide catalysts as a function of the reaction temperature in the excess H2 were shown in Figs. 6 and 7. The best CO conversion values were obtained with 50/50 (mol.%) AgCo mixed oxide catalyst and 50/50 (mol.%) CoCe mixed oxide catalysts which were calcined at 200  C. The CO conversion of the 50/50 (mol.%) AgCo mixed oxide catalyst was decreased at around 150  C due to the formation of methane in a minor amount, with further increase in temperature, CO conversion again increased. At 170  C, CO conversion of the catalysts reached 99% and it did not change after that temperature. The CO conversion of the 50/50 (mol.%) CoCe mixed oxide was increased with increasing reaction temperature and it reached 99% at 170  C and it did not change. The CO conversion of the 25/25/50 (mol.%) AgCoCe mixed oxide catalyst calcined at 200  C increased with increasing reaction temperature and it reached 84% at 210  C. The lowest CO conversion among the catalysts calcined at 200  C was obtained for the 50/50 (mol.%) AgCe mixed oxide catalyst of which CO conversion did not increase over 19% in the reaction temperature interval. TPR spectrums of the

international journal of hydrogen energy 35 (2010) 22472261

2257

Fig. 6 Selectivities, CO conversions and O2 conversions of the catalysts calcined at 200 8C as a function of temperature fort he selective CO oxidation reaction (2540 mg catalyst, S.V. 45 000 hL1, 1 vol.% CO, 2 vol.% O2, 85 vol.% H2 and remain He).

catalysts showed that the area under the TPR curve until 210  C of the 50/50 (mol.%) AgCe mixed oxide catalyst calcined at 200  C was the biggest. The oxygen in this catalyst structure was higher than the other catalysts. The oxygen in the gas phase might be used to ll the oxygen vacancies in the catalyst structure. The activity in CO oxidation reaction in selective oxidation of the catalyst was low because the CO oxidation mostly depended on the reaction between the

surface oxygen and adsorbed CO to the surface of the catalyst. CO oxidation activity of 25/25/50 (mol.%) AgCoCe mixed oxide catalyst calcined at 450  C may be low because of the competition between the CO oxidation and catalyst surface at which the oxygen vacancies consist. Among the catalysts calcined at 450  C, the 50/50 (mol.%) AgCo mixed oxide catalyst gave the best result in terms of CO conversion (Fig. 7). While the CO conversion of this catalyst was

2258

international journal of hydrogen energy 35 (2010) 22472261

Fig. 7 Selectivities, CO conversions and O2 conversions of the catalysts calcined at 450 8C as a function of temperature fort he selective CO oxidation reaction (2540 mg catalyst, S.V. 45 000 hL1, 1 vol.% CO, 2 vol.% O2, 85 vol.% H2 and remain He).

low between the temperatures 110  C170  C, it reached a value of 87% at 210  C. The 50/50 (mol.%) AgCe gave maximum 13% CO conversion in the reaction temperature interval and 25/25/ 50 (mol.%) AgCoCe mixed oxide catalyst gave the lowest CO conversion among the synthesized catalysts.

The decrease in the cobalt oxide and silver oxide molar ratio caused a decrease in the CO activity, but addition of the cerium oxide caused an increase in the selective CO oxidation activity. Although the 50/50 (mol.%) AgCo mixed oxide catalyst gave the highest CO oxidation activity, the 50/50 (mol.%)

international journal of hydrogen energy 35 (2010) 22472261

2259

CoCe mixed oxide catalyst gave the highest selectivity since the cerium oxide kept the lattice oxygen of the catalysts against the H2 reduction. Souza et al. [43] reported that the strong interaction between the metal and CeOx, creates new active sites for CO adsorption at the metal-support interface, and this led to an increase in the catalytic activity for CO oxidation. Therefore, the interaction between the cobalt and CeO2 cause gave the highest selectivity result. The selectivity of the catalysts to CO oxidation in the H2 rich gas stream as a function of temperature are shown in Figs. 6 and 7. The best selectivity result was obtained for the 50/50 (mol.%) CoCe mixed oxide catalyst calcined at 200  C. The selectivity of this catalyst increased until 150  C and then it decreased. The selectivity of the 50/50 (mol.%) AgCo mixed oxide catalysts calcined at 200  C was decreased at around 150  C due to the formation of methane in a minor amount and then further increased up to 170  C at which it gave 13% selectivity. After 170  C, selectivity of 50/50 (mol.%) AgCo mixed oxide catalyst calcined at 200  C decreased to 9%. 25/25/ 50 (mol.%) AgCoCe mixed oxide catalysts calcined at 200  C gave maximum 9% selectivity between the reaction temperature intervals. The lowest selectivity was obtained for the 50/ 50 (mol.%) AgCe mixed oxide catalyst calcined at 200  C of which selectivity was not increased. The difference between the selectivities of the catalysts calcined at 200  C and 450  C are shown in Figs. 6 and 5. 50/50 (mol.%) AgCo mixed oxide catalyst calcined at 450  C gave a more stable selectivity curve. Selectivity of the 50/50 (mol.%) AgCo mixed oxide catalyst calcined at 450  C increased up to 38% at 190  C and then reduced to 34% at 210  C. Both the selectivities of the 50/50 (mol.%) AgCo mixed oxide catalyst calcined at 200  C and 50/ 50 (mol.%) AgCo mixed oxide catalyst calcined at 450  C reduced after 170  C and 190  C, respectively. Selectivities of the both 50/50 (mol.%) AgCe mixed oxide and 25/25/50 (mol.%) AgCoCe mixed oxide catalysts calcined at 450  C were very low and did not increase with increasing reaction temperature. Oxygen conversions of the mixed oxide catalysts also were shown in Figs. 6 and 7. Both 50/50 (mol.%) AgCo mixed oxide and 50/50 (mol.%) AgCe mixed oxide catalysts calcined at 200  C gave 100% O2 conversion. The O2 conversion over the 50/50 (mol.%) CoCe mixed oxide catalyst and 25/25/50 (mol.%) AgCoCe mixed oxide catalyst which were calcined at 200  C increased with increasing temperature and these catalysts gave 100% O2 conversion at temperatures higher than 170  C. The highest O2 conversion was obtained from 50/50 (mol.%) AgCe mixed oxide catalyst among the catalysts calcined at 450  C. The O2 conversion of the other three catalysts calcined at 450  C was lower than 15%. According to the TPR measurements, oxygen vacancies were formed on the catalysts structure after H2 reduction. The O2 conversion was greater than the CO conversion because oxygen in the gas phase was used not only to oxidize CO and H2 but also to ll the O2 vacancies after H2 reduction. The difference between the CO oxidation activities and selective CO oxidation activities of the catalysts depended on several parameters. The catalysts were reduced under H2 atmosphere, the CO conversion of the catalysts obtained after selective CO oxidation studies were lower than that obtained from the CO oxidation studies. The decrease in the CO

oxidation activity was obvious in the case of 50/50 (mol.%) AgCo and 50/50 (mol.%) AgCoCe mixed oxide catalyst calcined at 200  C. The activity of the catalyst mainly depended on the oxygen in the catalyst structure. Because of the reduction of oxygen in the catalyst structure, a sharp decrease in the CO conversion was obtained in the selectivity studies. Changing the structure of the catalyst by reduction of oxygen has affected the catalytic activity results. XRD studies showed that the metallic silver peaks were formed in the catalysts structures which were calcined at 450  C. The 50/50 (mol.%) AgCo and 50/50 (mol.%) AgCe mixed oxide catalysts calcined at 450  C showed mostly the catalytic behavior of the CeO2 and Co3O4. In the literature, the CeO2 catalyst is reported to give 8% CO conversion and 9.1% selectivity to CO oxidation at 95  C [4]. This study supported the low CO conversion and selectivity to CO oxidation results obtained for the 50/50 (mol.%) AgCe mixed oxide catalyst calcined at 450  C. Reaction products, water vapor and CO2, have an effect on the catalytic activity of the catalysts. The production of CO2 and water vapor resulted in a decrease in the catalytic activity, since the active sites of the catalysts are blocked by CO2 and water vapor. Decrease in the catalytic activity because of the CO2 and H2O was also supported by other researchers [44,21]. The stability of the activity and selectivity of the catalyst for a long time is a very important criterion in the catalyst research. Long time activity and stability are preferred from an economical point of view. In this study, catalysts were prepared in order to remove the CO from the feed which was used as a fuel in the fuel cell applications. The feed was composed of approximately 12 vol. CO%, 75 vol.% H2, 25 vol.% CO2. The catalyst was in interaction with H2, which was in the fuel, for a time before fed in to the fuel cell. For this reason the activity and selectivity of the catalyst will be affected after a certain time. The deactivation test was carried out to investigate the effect of gas composition (1 vol.% CO, 2 vol.% O2, 85 vol.% H2 and remaining being He) on the activity and selectivity of the catalyst with time. It was desired that the catalyst used in the industrial plants should not lose its activity and selectivity. So, the activity and selectivity of the catalyst was scanned for 5800 min which was sufcient time. The deactivation study was carried out with 50/50 (mol.%) Ag Co and 50/50 (mol.%) CoCe mixed oxide catalysts which were calcined at 200  C. The 50/50 (mol.%) AgCo mixed oxide catalyst calcined at 200  C was chosen since it gave the highest activity among the catalysts and 50/50 (mol.%) CoCe mixed oxide catalyst calcined at 200  C was chosen since it showed the highest selectivity among the catalysts. The CO conversion and selectivity to CO oxidation as a function of time is shown in Fig. 8. The activity and selectivity of the catalysts were monitored at 150  C throughout the 5800 min. Activity of the 50/50 (mol.%) CoCe mixed oxide catalyst was changed between the 5660%. 50/50 (mol.%) AgCo mixed oxide catalyst gave 30% CO conversion throughout the 5800 min. Both of the catalysts did not lose their selectivity during the deactivation tests and selectivity of the catalysts was increased at the end of the deactivation tests. While the selectivity of the 50/50 (mol.%) CoCe mixed oxide catalyst was 39% at the beginning, it was increased to 50% at the end of the test and the selectivity of the 50/50 (mol.%) AgCo mixed oxide catalyst was increased from 23% to 40%. These experiment results can

2260

international journal of hydrogen energy 35 (2010) 22472261

calcination at 200  C. Catalysts gave crystal phase structure after calcination at 450  C with low surface area, low activity and low selectivity results. In order to show the effects of the cerium oxide on the activity and selectivity for the CO oxidation the results were compared our earlier studies. The preparation the catalysts composed of silver oxide and cobalt oxide gave better activity and selectivity for the selective CO oxidation than the catalyst composed of silver oxide and cerium oxide and the catalyst composed of cobalt oxide and cerium oxide. The catalysts composed of silver oxide, cobalt oxide and cerium oxide gave the lowest activity and selectivity results. The cobalt oxide was found more active component than cerium oxide to the silver oxide for the selective CO oxidation. The 50/50 (mol.%) AgCo mixed oxide catalyst calcined at 200  C was found to be the most active catalyst and 50/50 (mol.%) CoCe mixed oxide catalyst calcined at 200  C was found to be most selective catalyst for the selective CO oxidation. Deactivation tests which were made over the 50/50 (mol.%) AgCo mixed oxide and 50/50 (mol.%) CoCe mixed oxide catalysts calcined at 200  C showed that catalysts did not lose their activity and selectivity at 150  C during the 5800 min. Both catalytic activity and characteristic results showed that the silver oxide and cobalt oxide is good couple for the selective CO oxidation reaction. Fig. 8 Deactivation test results of the 50/50 (mol.%) AgCo mixed oxide and 50/50 (mol.%) CoCe mixed oxide catalysts calcined at 200 8C (2540 mg catalyst, S.V. 45 000 hL1, 1 vol.% CO, 2 vol.% O2, 85 vol.% H2 and remain He, T [ 150 8C).

Acknowledgement
The authors gratefully acknowledge the nancial support of the TUBITAK MISAG-229 and Gazi University BAP 06/2006-17 project.

be supported by the Gulari et al. [5] and Qu et al. [45]. Gulari et al. [5] showed that 50/50 AgCo composite oxide did not loose their activity at 30  C, 50  C and 90  C for 20 000 min (z333 h) for the low temperature CO oxidation activity. In addition, Qu et al. [45] studied the long term stability of the 8% Ag/SiO2 catalyst for selective CO oxidation at 50  C and they observed that catalyst showed good stabilities z490 min. Deactivation results are promising compared with the deactivation results of the catalysts that composed of Pt and Au in the literature. The activities and selectivities of the catalysts composed of Pt decreased sharply at the beginning of the deactivation study which was result of the carbon deposition [43,46,47]. On the other hand PtPd/CeO2 catalyst showed high stability at 90  C for 24 h for the selective CO oxidation [48]. And catalysts composed of Au have good stability for the selective CO oxidation [1,12].

references

4.

Conclusion

In this study the activity and selectivity of CO oxidation in a H2 rich stream were investigated over the Ag/Co/Ce mixed oxide catalysts. It was predicted that at the end of the preparation of the silver oxide and cobalt oxide with cerium oxide as nal catalysts will gave high activity and selectivity for the selective CO oxidation reaction. Results showed that catalysts gave amorphous phase structure besides high surface area after

[1] Huang Y, Wang A, Wang X, Zhang T. Preferential oxidation of CO under excess H2 conditions over iridium catalysts. Int J Hydrogen Energy 2007;32:38806. [2] Manzoli M, Avgouropoulas G, Tabakova T, Papavasiliou T, Ioannides T, Boccuzzi F. Preferential CO oxidation in H2 rich gas mixtures over Au/doped ceria catalysts. Catal Today 2007;138(12):23943. [3] Qu Z, Cheng M, Dong X, Bao X. CO selective oxidation in H2rich gas over Ag nanoparticles-effect of oxygen treatment temperature on the activity of silver particles mechanically mixed with SiO2. Catal Today 2004;9395:24755. [4] Qu Z, Huang W, Zhou S, Zheng H, Liu X, Cheng M, et al. Enhancement of the catalytic performance of supportedmetal catalysts by pretreatment of the support. J Catal 2005; 234:336. [5] Gulari E, Guldur C, Srivannavit S, Osuwan S. CO oxidation by silver cobalt composite oxide. Appl Catal A 1999;182:14763. [6] Guldur C, Balkc F. Catalytic oxidation of CO over AgCo/ Alumina catalysts. Chem Eng Comm 2003;190:98698. [7] Guldur C, Balkc F. Selective carbon monoxide oxidation over Ag-based composite oxides. Int J Hydrogen Energy 2002;27: 21924. [8] Guo Q, Liu Y. MnOx modied Co3O4CeO2 catalysts for the preferential oxidation of CO in H2-rich gases. Appl Catal B 2008;82:1926.

international journal of hydrogen energy 35 (2010) 22472261

2261

[9] Zhao Z, Yung MM, Ozkan US. Effect of support on the preferential oxidation of CO over cobalt catalysts. Catal Commun 2008;9:146571. [10] Wang H, Zhu H, Qin Z, Wang G, Liang F, Wang J. Preferential oxidation of CO in H2 rich stream over Au/CeO2Co3O4 catalysts. Catal Commun 2008;9:148792. [11] Guo Q, Wu M, Liu Y, Bai X. Mesoporous CeO2-supported Co3O4 catalysts for CO preferential oxidation in H2-rich gases. Chin J Catal 2007;28(11):9537. [12] Yung MM, Zhao Z, Woods MP, Ozkan US. Preferential oxidation of carbon monoxide on CoOx/ZrO2. J Mol Catal A Chem 2008;279:19. [13] Ko EY, Park ED, Lee HC, Lee D, Kim S. Supported PtCo catalysts for selective CO oxidation in a hydrogen rich stream. Angew Chem Int 2007;46:7247. [14] Chen YZ, Liaw BJ, Chang WC, Huang CT. Selective oxidation of CO in excess hydrogen over CuO/CexZr1xO2Al2O3 catalysts. Int J Hydrogen Energy 2007;32:45508. [15] Storster S, Ttdal B, Walmsley JC, Tanem BS, Holmen A. Characterization of alumina-, silica-, and titania-supported cobalt FischerTropsch catalysts. J Catal 2005;236:13952. [16] Jung CR, Han J, Nam SW, Lim TH, Hong SA, Lee HI. Selective oxidation of CO over CuOCeO2 catalyst: effect of calcination temperature. Catal Today 2004;9395:18390. [17] Liotta LF, Gorlo GD, Pantales G, Venezia AM, Deganells G. Co3O4/CeO2 composite oxides for methane emissions abatement. Relationship between Co3O4CeO2 interaction and catalytic activity. Appl Catal B 2006;66:21727. [18] Luo M, Yuan X, Zheng X. Catalyst characterization and activity of AgMn, AgCo and AgCe composite oxide for oxidation of volatile organic compounds. Appl Catal A 1998;175:1219. [19] Taylor PL, Omotoso O, Wiskel JB, Mitlin B, Burel RE. Impact of heat on nanocrystalline silver dressing part I: physical properties. Biomaterials 2005;26:723040. [20] Gao XY, Wang JY, Li J, Zheng YX, Jhang RJ, Zhou P, et al. Study of structure and optical properties of silver oxide lms by ellipsommetry, XRD and XPS methods. Thin Solid Films 2004; 455456:43842. [21] Zou H, Dong X, Lin W. Selective CO oxidation in hydrogen-rich gas over CuO/CeO2 catalysts. Appl Surf Sci 2006;253:28938. [22] Bicford ES, Velu S, Song C. Nano-structured CeO2 supported CuPd bimetallic catalysts for the oxygen assisted water-gas shift reaction. Catal Today 2005;99:34757. [23] Zheng XC, Wu SH, Wang SP, Wang SR, Zhang SM, Huang WP. The preparation and catalytic behavior of coppercerium oxide catalysts for low temperature carbon monoxide oxidation. Appl Catal A 2005;283:21723. [24] Salaita GN, Hazos ZF, Hound GB. Surface characterization study of the thermal decomposition of Ag2CO3 using X-ray photoelectron spectroscopy and electron energy loss spectroscopy. J Electron Spectrosc 2000;107:7381. [25] Boronin AI, Koscheev SV, Murzakhmetov KT, Avdeev VI, Zhidomirov GM. Associative oxygen species on the oxidized silver surface formed under O2 microwave excitation. Appl Surf Sci 2000;165:914. [26] Waterhouse GIN, Bowmaker GA, Metson JB. Oxidation of a polycrystalline silver foil by reaction with ozone. Appl Surf Sci 2000;183:191204. [27] Cesar DV, Perez CA, Schmal M, Salim VMM. Quantitative XPS analysis of silica-supported CuCo oxides. Appl Surf Sci 2000; 157:15966. [28] Hagelin-Weaver HAE, Hound GB, Minahan DM, Salaita GN. Electron energy loss spectroscopic investigation of Co metal, CoO, and Co3O4 before and after Ar bombardment. Appl Surf Sci 2004;235:42088. [29] Larachi F, Pierre J, Adnot A, Bernis A. Ce 3d XPS study of composite CexMn1xO2y wet oxidation catalysts. Appl Surf Sci 2002;195:23650.

[30] Skarman B, Grandjean D, Beneld RE, Hinz A, Andersson A, Wallenberg LR. Carbon monoxide oxidation on nanostructured CuOx/CeO2 composite particles characterization by HREM, XPS, XAS, and high-energy diffraction. J Catal 2002;211:11933. [31] Colina JZ, Nix RM, Weiss H. A novel approach to the epitaxial growth of CeO2 lms on Si(111). Surf Sci 2004;563:2515. [32] Teodorescu M, Georgescu V, Sitaru I, Vass MI, Segal E. Thermoprogrammed reduction and thermoprogrammed desorption studies of Co and Cr bimetallic oxides. Thermochim Acta 1996;274:30713. [33] Tang X, Zhang B, Li Y, Xu Y, Xin Q, Shen W. Carbon monoxide oxidation over CuO/CeO2 catalysts. Catal Today 2004;191:935. [34] Zhu H, Qin Z, Shan W, Shen W, Wang J. Pd/CeO2TiO2 catalyst for CO oxidation at low temperature: a TPR study with H2 and CO as reducing agents. J Catal 2004; 255:26777. [35] Xiaoyuan J, Guanglie L, Renxian Z, Jianxian M, Yu C, Xiaoming Z. Studies of pore structure, temperatureprogrammed reduction performance, and microstructure of CuO/CeO2 catalysts. Appl Surf Sci 2001;173:20820. [36] Kang M, Song MW, Lee CH. Catalytic carbon monoxide oxidation over CoOx/CeO2 composite catalysts. Appl Catal A 2003;251:14356. [37] Lay E, Gauthiera G, Rosini S, Savaniu C, Irvine JTS. Cesubstituted LSCM as new anode material for SOFC operating in dry methane. Solid State Ionics 2008;179:15626. [38] Faria WLS, Dieguez LC, Schmal M. Autothermal reforming of propane for hydrogen production over Pd/CeO2/Al2O3 catalysts. Appl Catal B 2008;85(12):7785. [39] Eguchi Y, Abe D, Yoshitake H. Oxidation state of Ce and ethanol oxygen reaction of mesoporous titania-supported cerium oxide. Microporous Mesoporous Mater 2008;116(13):4450. [40] Monyanon S, Pongstabodee S, Luengnaruemitchai A. Catalytic activity of PtAu/CeO2 catalyst for the preferential oxidation of CO in H2-rich stream. J Power Sources 2006;163: 54754. [41] Chih-Wei Tang, Chao-Chieh Kuo, Ming-Chih Kuo, ChenBin Wang, Shu-Hua Chien. Inuence of pretreatment conditions on low-temperature carbon monoxide oxidation over CeO2/Co3O4 catalysts. Appl Catal A 2006;309:3743. [42] Shao J, Zhang P, Tang X, Zhang B, Song W, Xu Y. Effect of preparation method and calcinations temperature on low temperature CO oxidation over Co3O4/CeO2. Chin J Catal 2007;28(2):1639. [43] Souza MMVM, Ribeiro NFP, Schmal M. Inuence of the support in selective CO oxidation on Pt catalysts for fuel cell applications. Int J Hydrogen Energy 2007;32:4259. [44] Avgouropoulos G, Ioannides T, Matralis H. Inuence of the preparation method on the performance of CuOCeO2 catalysts for the selective oxidation of CO. Appl Catal B 2005; 56(12):8793. [45] Qu Z, Cheng M, Shi C, Bao X. Low-temperature selective oxidation of CO in H2-rich gases over Ag/SiO2 catalysts. J Mol Catal A Chem 2005;239:2231. [46] Luengnaruemitchai A, Nimsuk M, Naknam P, Wongkasemjit S, Osuwan S. A comparative study of synthesized and commercial A-type zeolite-supported Pt catalysts for selective CO oxidation in H2-rich stream. Int J Hydrogen Energy 2008;33:20613. [47] Son IH, Lane AM, Johnson DT. The study of the deactivation of water-pretreated Pt/g-Al2O3 for low-temperature selective CO oxidation in hydrogen. J Power Sources 2003;124:4159. [48] Parinyaswan A, Pongstabodee S, Luengnaruemitchai A. Catalytic performances of PtPd/CeO2 catalysts for selective CO oxidation. Int J Hydrogen Energy 2006;31: 19429.