44
Acta Chim. Slov. 2005, 52, 44-52
Scientific Paper
Redox Behavior of (CuO)015(CeO2)0 85 Mixed Oxide Catalyst Prepared by Sol-Gel Peroxide Method
Albin Pintar, Jurka Batista, and Stanko Hočevar*
Laboratory for Catalysis and Chemical Reaction Engineering, National Institute of Chemistry, Hajdrihova 19, P.O. Box 660, SI-1001 Ljubljana, Slovenia
Received 17-08-2004
Abstract
Temperature-programmed reduction (TPR), oxidation (TPO) and desorption (TPD) studies were performed on (CuO)015(CeO2)085 copper-ceria mixed oxide sample prepared by sol-gel peroxide route. The obtained results reveal that despite initial drop in specific surface area after consecutive redox cycles the hydrogen consumption remains constant. This is because CuO is highly dispersed over the surface of Ce02 nano-crystallites and remains highly dispersed even after the agglomeration of Ce02 nano-crystallites in more dense secondary structure. The dispersed CuO is reduced to Cu° during the TPR, forming agglomerated metal particles on the surface of partialh/ reduced Ce02. After subsequent TPO step ali the Cu° is oxidized back into CuO and re-dispersed over the Ce02 crystallites.
Key words: catalyst, (CuO)015(CeO2)085, redox properties, TPR, TPO, TPD
Introduction
Cerium oxide and Ce02-containing materials have been extensively studied as a component of heteroge-neous industrial catalysts or as a support for transition metals due to their superior chemical and physical stability, high oxygen mobility, strong interaction with the supported metal and ability to be modified. Cerium oxide has been increasingly used as a thermal stabilizer and oxygen storage medium in the three-way catalysts for automotive emission control.1
Ce02 has face-centered cubic crystal structure, into which various cation dopants can be introduced in order to improve the physicochemical properties of ceria.M The modification of Ce02 with Cu2+ ions leads to creation of oxygen ion vacancy around the Cu2+ ion in the Ce02-based mixed oxide catalysts, to local struc-tural changes and to decrease of the redox potentials of Cu species in the Ce02 matrix.5 The obtained CuO-Ce02 catalyst materials exhibited high activities for the oxidation of carbon monoxide and methane,6-8 the S02 reduction by CO,911 the NO reduction1213, the water gas shift reaction14, and the wet oxidation of phenol.1516 The CuO-Ce02 mixed oxides were also reported to be highly active and selective for oxidation of carbon monoxide in excess of hydrogen.1722 The steam reforming of metha-nol over the Cu-Ce02 catalysts, which were prepared by reduction of CuO-Ce02 mixed oxides, was reported very recently.2325
The catalysts for fundamental studies have been synthesized by various conventional techniques such as co-precipitation,2'6'9 the urea co-precipitation/gela-
tion method,26 the homogeneous co-precipitation us-ing hexamethylenetetramine,27 and by novel chemical routes such as the inert gas condensation technique,2830 and the solution combustion method.5 The preparation conditions and mixed oxide composition influence the prevailing form and distribution of copper species on ceria. The enhanced catalyst activity and stability result from interactions betvveen the copper-cerium oxide phases.
It was revealed that activity and selectivity of two series of (CuO)x(Ce02)1_x catalysts prepared by co-precipitation method and by sol-gel peroxide route increase with the dispersion of copper oxide phase on the cerium oxide.1516 A comparative study of Pt/Y-Al203, Au/a-Fe203 and (CuO)0 05(CeO2)0 95 catalysts for the selective oxidation of carbon monoxide in excess hydrogen showed that the (CuO)0 05(CeO2)0 95 catalyst prepared by the sol-gel peroxide route is superior to the other two catalysts in the low-temperature range, because it has the best compromise betvveen activity, selectivity, and priče of the catalyst.18 The kinetics of selective CO oxi-dation in excess of hydrogen over the (CuO)01(CeO2)09 nanostructured catalyst prepared by sol-gel peroxide route was studied under simulated preferential oxida-tion (PROX) reactor conditions.19
In this work, we report on the examination of a (CuO)015(CeO2)085 mixed oxide catalyst (synthesized by the modified sol-gel technique using hydrogen peroxide16 and referred to as CuCe-2) by means of TPR, TPO and TPD techniques, which will be demonstrated as an efficient tool to obtain information about the re-dox behavior of this solid, understanding of which is of
Pintar et al.    Redox Behavior of Copper-Ceria Mbced Oxide Catalyst
Acta Chim. Slov. 2005, 52, 44-52
45

25
40
55
70
2 Theta,
Figure 1. X-ray powder diffraction patterns of CuCe-2 sample: (a) fresh; (b) after TPR-l/TPO/TPR-2 analysis; (c) after four consecutive TPR/TPO cycles.
practical importance for designing CuO-Ce02 catalysts that can replace the expensive noble metal catalysts in a number of down-stream processes for production of H2-rich gas streams from fossil and renewable fuels used as a fuel for the proton exchange membrane fuel cells (PEMFC).
Results and discussion
1. X-ray diffraction analysis and UV-VIS spectroscopy
X-Ray powder diffraction patterns of fresh and treated CuCe-2 mixed oxide samples are shown in Figure 1. The X-ray pattern of fresh CuCe-2 sample revealed the characteristic diffraction peaks of Ce02 phase (Cerianite, syn, cubic, PDF 34-0394: 26 = 28.6, 33.1, 47.5, 56.3, 59.1 and 69.4°), assigned to the fluorite structure. The single-phase XRD patterns indicate a high dispersion of copper species (X-ray amorphous) at the surface of Ce02 crystallites. The diffraction peaks of cerianite in CuCe-2 sample prepared by sol-gel peroxide route are strong and sharp, which means that the CuO phase is finely dispersed on the surface of relatively large Ce02 crystallites.16 By peak broadening analysis, the average crystallite size in the direction normal to the (111) plane in cerianite was calculated to be 42 nm.
In order to obtain preliminary information on the copper species present in the CuCe-2 sample, UV-VIS diffuse reflectance spectra of the reference Ce02, CuO and Cu20 materials and prepared mixed oxide were col-lected. The positions of the absorption maxima for CuO (X=875 nm) and Cu20 (X=620 nm) clearly demonstrate
the presence of the d-d transitions of Cu2+ (3d9) in the examined mixed oxide sample.31
2. H2-TPR, 02-TPO, TPD-H2 and TPD-02
Interaction of hydrogen with (CuO)x(Ce02)1_x mixed oxides during TPR involves the adsorption of hydrogen on aH active sites of the cerium oxide surface, storage of hydrogen in the host oxide and reduction of the CuO component.27'3234 The qualitative and quantita-tive characterization of reducibility and reoxidability of different types (Le., well-dispersed, bulk-like) of copper ions present in the prepared CuCe-2 mixed oxide are determined by TPR and TPO measurements carried out in the temperature range from 0 to 400 °C (Figure 2). Figure 2a confirms that highly dispersed copper ions in nanocrystalline (CuO)015(CeO2)0 85 can be readily re-duced and oxidized at temperatures as low as 260 °C.35 The extent of further reoxidation of Cu° to Cu2+, partial consumption of hydrogen sto red in reduced samples and storage of oxygen during the TPO run are illustrated in Figure 2b. The numerical values of the TPR and TPO peak areas of the copper-cerium oxide sample (Figure 2) are evaluated and listed in Table 1. Comparison of TPR profiles of CuCe-2 sample obtained by carrying the TPR experiments at different sample loading (0.10 and 0.25 g, respectively) confirms the reproducibility of TPR profile characteristics: the temperature corresponding to the maximum reduction rate and the shape of the TPR profile. At the same time peak area increases proportionally with increasing sample loading, which implies that the reduction profile is not influenced by
o
Pintar et al.    Redox Behavior of Copper-Ceria Mbced Oxide Catalyst
46
Acta Chim. Slov. 2005, 52, 44-52


		°    exp. -------calc.
		Peak #        T (° C) s,              130
0.06	1 %	b,              149 s2                178 I                  201
	S'I"\-N. J !''¦' 1	b2                212
0.03		
	7' 1   a<--- \\	
0.00		
		
100                 200
Temperature, °C
300
400


		•    exp.
0.016		-------calc. Peak #        T (°C)
0.012	r''¦   V««	s,              49 b,               95
	1           /\	s2                121 b2                170
0.008		l„                219
0.004	i '                  ''• ¦'                  i   \	
	/¦•'        /s-. '¦.        A'bX	
0.000	I^C^--^^:-------~~'i*-'----------^--------l—^^^^gg	
100
200
Temperature, »C
o
300
400
Figure 2. TPR-1 (a) and TPO (b) profiles of CuCe-2 sample measured in the temperature range of 0^100 °C and predicted by means of deconvolution method. Operating conditions: 50 mL/min (STP), H2(5 vol.%)/Ar (a), O2(10 vol.%)/He (b), 5 °C/min. Sample weight: 250 mg. The initial state of (b) is fresh sample following TPR-1 run, cooling in H2(5 vol.%)/Ar to 0 °C and purging at 0 °C with pure Ar. Designation of peaks: sx - partial reduction of Cu2+ -> Cu+ in well-dispersed CuO species; s2 - partial reduction of Cu+ -> Cu°in well-dispersed CuO species; b1 - partial reduction of Cu2+ -> Cu+ in bulk-like CuO phase; b2 - partial reduction of Cu+ -> Cu° in bulk-like CuO phase. Ia - H2 incorporation in the catalyst structure; Ib - consumption of H2 incorporated in the catalyst structure during the TPR-1 analysis.
the operating conditions (Le., initial amount of reducible species, heating rate).36
In order to verify the restoration of redox proper-ties of examined mixed oxide, the second TPR run was performed following TPO after cooling the sample in helium flow to 0 °C. TPR-2 measurement was carried out under the same reaction conditions as the first TPR-1 run. The TPR-l/TPO/TPR-2 cycling of CuCe-2 revealed that the TPR profile characteristics are not re-producible (Figure 3a). We observed in accordance with the findings of Zimmer et al.21 that heat treatment under hydrogen (TPR-1 analysis) reduces the specific surface area of CuCe-2 from 31 to 26 m2/g; no further decrease of specific surface area was observed in subsequent TPR/TPO cvcles. However, a comparison of the XRD patterns for prepared and TPR-l/TPO/TPR-2 treated CuCe-2 samples (Figure 1) provide no evidence that such treatment has substantial influence on the chang-ing of the cerium oxide diffraction peaks. A weak XRD peak associated with the Cu° phase is slightly above the detection limit. Apparently, XRD was not sensitive to the dispersed copper component.
2.1. Quantitative analvsis
For quantitative analysis, aH of the TPR, TPO and TPD peaks that can be discerned by computer softvvare (Grams/32, Thermo Galactic, version 4) have been integrated to evaluate their individual amounts. The numerical values obtained from the integrated areas are given in Table 1. Baseline correction and deconvolution of TPR and TPO profiles with five peaks using the PeakFit softvvare package (SPSS, version 4.11) give the area percentage for the copper ion species closely
interacting with the cerium oxide and for the segregated (bulk-like) CuO.
Evidently, the total hydrogen consumption of CuCe-2 solid is larger than the value expected for a com-plete reduction of the CuO component to Cu° (Table 1). The additional hydrogen consumption may be due to surface reduction of Ce02.27'32~34 During the reduction process, storage of hydrogen in the oxide, mainly in the bulk (Le., formation of bronze-like species) and further reaction of these activated hydrogen species with the lattice oxygen ions at T > 230 °C can take plače.34 Furthermore, it was reported that the incorporation of Ni2+ or Cu2+ in ceria affects the hydrogen storage in the host oxide and the degree of reduction of the cations.33'37 In order to determine the amount of hydrogen stored in the host oxide (HSC), the calculated amount of H2, which is needed for complete reduction of CuO to Cu° and the amount of hydrogen that reacts with oxygen stored in copper-cerium mixed oxide (after activation) is subtracted from the total hydrogen uptake (TPR). The obtained H2 consumption is then compared with the amount of hydrogen desorbed from reduced solid in the same temperature range (TPD-H2). Ouantita-tive analysis of the TPO profiles revealed the extent of reoxidation of Cu° on reduced CuCe-2 sample, the part of stored hydrogen that reacts with oxygen during temperature-programmed oxidation and the amount of oxygen stored in the sample during TPO.
Figure 2a shows that the CuCe-2 sample starts to be reduced at temperatures below 100 °C. The reduction steps as well as the simultaneous incorporation of hydrogen into the catalyst structure is completed at temperatures below 260 °C. Data shown in Table 1
o
0
Pintar et al.    Redox Behavior of Copper-Ceria Mhced Oxide Catalyst
Acta Chim. Slov. 2005, 52, 44-52
47
Table 1. Results of TPR-1, TPO, TPR-2, TPD-H2 and			TPD-02 analyses of CuCe-2	sample.
Analyses	CuCe-2		Analyses	CuCe-2
TPR-1			TPD-H2upto650°C	
total H2 uptake, mL/gsolitl	25.2		desorbed H2, mL/gsoM	4.0"
stored H2 (HSCi), mL/gsolitl	4.0 (16%)°		restofH2,mL/gsolitl	0
TPO			irreversibly captured H2,%	0
total 02 uptake, mL/gsolitl	12.6		TPD-H2upto400°C	
partial 02 uptake, mL/gsoM	2.0 (16%)"		desorbed H2, mL/gsolitl	3.2
remained H2, mL/gsolitl	0.0 (0%)c		restofH2,mL/gsolitl	0.8
TPR-2			irreversibly captured H2,%	20
total H2 uptake, mL/gsolitl	25.2		TPD-O2upto400°C	
stored H2 (HSC2), mL/gsolitl	4.0		desorbed 02, mL H2/gsolitl	1.3
HSCC, mL/gsolitl	= 4.0			
" Part of H2 consumed for incorporation into the catalyst structure. b Part of 02 consumed in a reaction with H2 stored in the catalyst struc-ture. c Part of H2 stored in the catalyst structure after the completion of TPO analysis. d The same value was measured during the TPD analysis conducted after four TPR/TPO cycles.
confirm that in the performed TPR analysis complete reduction of CuO phases was obtained. At a first sight, at least three peaks could be distinguished on the TPR profile of CuCe-2: the first peak with a maximum at T = 135 °C, the second one at T = 158 °C and a large reduction peak with a maximum at 212 °C (Figure 2a). The total hydrogen uptake obtained from the integrated area is 25.2 mL H2/gsolid, which is equivalent to the stoi-chiometric total oxygen consumption of 12.6 mL 02/gsolid (Table 1). Assuming ali the Cu content (x = 0.15) in the activated CuCe-2 solid is present as CuO, 21.2 mL H2/gsoM is required to reduce CuO to Cu°. The excess hydrogen uptake of 4.0 mL H2/gsolid is calculated as the difference betvveen the total amount of consumed H2 and the amount of H2 needed for complete reduction of Cu2+ ions in the CuCe-2. After hydrogen uptake up to 400 °C, the CuCe-2 was cooled in H2 and purged with Ar, and then the temperature-programmed desorption of hydrogen was carried out. The obtained amount of H2 desorbed from CuCe-2 in the temperature range up to 400 °C is 3.2 mL H2/gsoM (Table 1). This means that about 20% of stored hydrogen remains irrevers-ibly captured in the CuCe-2 solid after treatment up to 400 °C. Temperature-programmed oxidation of reduced CuCe-2 gives the TPO signal consisting of five overlap-ping peaks with maxima at 49, 95, 121, 170, and 219 °C (Figure 2b). The excess oxygen uptake of 2.0 mL 02/gsolid (Table 1) can be ascribed to the extraction of stored hydrogen from the CuCe-2 previously reduced up to 400 °C (Figure 2a). Comparison of the total hydrogen uptake with the total oxygen consumption shows that 0.0 mL H2/gsoM was remained in the CuCe-2 solid after TPR/TPO runs (Table 1).
Subsequently, deconvolution of TPR profile il-lustrated in Figure 2a was performed. Among numer-
ous functions tested, the best agreement betvveen the measured and calculated TPR profiles was obtained by using the Pearson IV fit function allowing asymmetric peaks. It can be seen that excellent agreement between the measured and predicted TPR profiles was obtained in the whole temperature range. Five peaks were suf-ficient to accurately describe H2 consumption during the reduction of CuCe-2 sample. This suggests that in the stepwise hydrogenation of CuO the reduction of inter-mediate Cu+ phases to Cu° proceeds at a comparable rate in comparison to the Cu2+ —> Cu+ reduction step. The mathematical analysis of TPR profile illustrated in Figure 2a reveals that the storage of H2 into the catalyst structure occurs in parallel to the reduction of CuO or Cu20 phases. The peak that shows the course of H2 storage as a function of temperature is marked in Figure 2a with letter I (incorporated). Although hydrogen can be stored in the examined catalyst in various forms,3234 no attempt was made during the mathematical analysis of TPR profiles illustrated in Figure 2a to further de-convolute I curve and thus try to differentiate betvveen various processes contributing to hydrogen storage capacity. This is because the measured TPR profiles were rather smooth not exhibiting many obvious peaks, which would be needed to perform deconvolution of I curve. It is interesting to note that the amount of H2 incorporated into the catalyst structure, calculated by means of the performed deconvolution analysis, is in good agreement with the results of TPR analysis (Table 1). For example, the performed simulation shows that 18% of H2 consumed (Ia peak in Figure 2a) was incorporated into the catalyst structure; based on the results of TPR-1 analysis, this value equals to 16% (Table 1). The performed deconvolution analysis can be also used to estimate the amount of well-dispersed CuO species
Pintar et al.    Redox Behavior of Copper-Ceria Mhced Oxide Catalyst
48
Acta Chim. Slov. 2005, 52, 44-52


				------TPR-1
0.42	ty			------TPR-2
		a	\        1*1	.......TPR-3
		f\	\ k	------TPR-4
0.40		1   »     ^5	V	
				
0.33	M			
	JJ			
nse	jj^tf«««^		•	'
u;e
200 Temperature, °C
»:'
'K.O
-0.720
-0J2S


-0.T3O
¦0.735
•0.740
200
Temperature, °C
,i;q
Figure 3. Comparisons of TPR and TPO patterns obtained in the temperature range of (Mi00 °C of CuCe-2 mixed oxide treated in con-secutive TPR/TPO cycles. For operating conditions, see Figure 1. Sample weight: 250 mg.
(sum of s1 and s2 relative peak areas, 15%) and bulk-like CuO species on the surface of Ce02 crystallites (sum of bj and b2 relative peak areas, 85%), which was confirmed recently by N20 selective chemisorption analysis.38
Figure 2b shows a profile obtained during the TPO analysis of pre-reduced CuCe-2 sample. It should be noted that no oxygen was consumed at T < 0 °C. It was also verified by means of pulse chemisorption measure-ments carried out at T = 0 °C that no 02 was consumed during the preceding detector stabilization period (5 min), in which a catalyst sample was exposed to oxy-gen stream. It can be further seen that in comparison to TPR analysis the catalyst reoxidation completes at temperatures similar to the ones required for complete reduction of this solid. As discussed below, the oxidative treatment at temperatures up to 400 °C is sufficient for the examined catalyst sample to completely reoxidize the copper phases. The TPO profile measured during the reoxidation of CuCe-2 sample (Figure 2b) was satis-factorily simulated by assuming the following processes: (i) reoxidation of the well dispersed Cu phase; (ii) re-oxidation of the segregated Cu phase; (iii) consumption of H2 stored in the catalyst structure. An involvement of five peaks was required to satisfactorily simulate the measured TPO profile during the subsequent de-convolution. This is in agreement with results of TPR analysis illustrated in Figure 2a, during which stepwise reduction of CuO was observed. Peaks s1 and s2 in Figure 2b are attributed to the stepwise reoxidation of well-dispersed Cu phase, while peaks bt and b2 belong to the reoxidation of bulk-like Cu phase. The calculated areas of peaks of each pair are very similar. The calculated relative surface areas of peaks belonging to the reoxi-dation of well dispersed (14%) and segregated (86%) Cu phases, are very close to the corresponding TPR-1 values. Furthermore, the calculated relative area of peak Ib, demonstrating the consumption of incorporated H2
during the TPO analysis, was found to be equal to 14%, which is again very close to the measured value of 16% calculated on the basis of data listed in Table 1. Finally, it is reported in Table 1 that in CuCe-2 sample, which exhibits rather low HSC capacitv, hydrogen stored in the catalyst during the preceding TPR step was completeh/ consumed during the TPO analysis.
The second reduction of catalyst sample examined in this study was performed and compared to those measured during the TPR-1 analysis (see Figure 3a). It was found out that the TPR-2 profile is shifted towards higher temperatures. This shift is the consequence of substantial decrease of the sample volume (up to 15 %), which occurred during the TPR-1 analysis. This causes different bulk density as well as different location of the thermocouple in the sample bed. Consequently, the lower heat transfer to thermocouple by the gas convec-tion in comparison with the heat transfer by conduction causes shift to higher temperature. Partialh/, the shift might be attributed also to the fact that the BET surface area drops after TPR-1 analysis (from 31 to 26 m2/g). Interestingly, the same number of peaks (5) was used to satisfactorih/ simulate the TPR-2 profile by means of the deconvolution method as was in the čase of TPR-1 analysis. This analysis, which accurateh/ takes into account the amount of hydrogen incorporated in the catalyst structure during the TPR-2 analysis (Table 1), interestingly shows that the percentage of well-dispersed and bulk-like CuO phases remains constant in the ap-plied temperature range. Although the BET surface area changes during the TPR-1 analysis, these findings suggest that the interface betvveen the CuO and Ce02 phases is not modified significantly. Furthermore, ad-ditional TPR/TPO cycles illustrated in Figure 3 reveal that total hydrogen and oxygen consumptions as well as HSC values remain constant in consecutive TPR/TPO cycles. It is believed that this is due to high dispersion of
Pintar et al.    Redox Behavior of Copper-Ceria Mhced Oxide Catalyst
Acta Chim. Slov. 2005, 52, 44-52
49
-0.328
-0.329


-0.330
-0.331
Measured after:
------TPR-1 aiiElysf5
¦ consscutive TPR/TPO cyclss
100       200       300       400 Temperature, °C
šoO
tsau
Figure 4. Comparison of TPD-H2 patterns of CuCe-2 sample obtained in the temperature range of –20–650 °C after TPR-1 analysis or consecutive TPR/TPO cycles carried out up to 400 °C. Operating conditions: 50 mL/min (STP), pure Ar, 5 °C/min. Sample weight: 250 mg.


ms
- 300
sm
100


m>     a    m   m   1« Time, min
Figure 5. TPD-02 pattern of CuCe-2 sample measured in the temperature range of -20-400 °C. Operating conditions: 50 mL/min (STP), pure He, 5 °C/min. Sample weight: 250 mg.
CuO over the surface of Ce02 nano-crystallites, which is maintained even after the agglomeration of Ce02 nano-crystallites in more dense secondary structure. This is confirmed by the fact that subsequent TPR-2 - TPR-4 profiles (Figure 3a) as well as TPO-2 - TPO-3 patterns (Figure 3b) are not shifted to higher temperatures.
It is evident from Table 1 that rather low quantity of hydrogen was stored in the structure of CuCe-2 sample during the TPR-1 analysis. It is also shown that hydrogen initialh/ incorporated in CuCe-2 sample was completely consumed in the subsequent reoxidation step conducted in the temperature range of 0-400 °C. Since the amounts of hydrogen consumed during the TPR-1 and TPR-2 analyses of CuCe-2 sample were found to be equal, this means that the complete hydro-gen storage capacity (HSCC) of this solid was achieved in the first reduction step. On the basis of performed analysis, it is concluded that the HSCC for CuCe-2 sample is equal to 4.0 mL H2/gsolid.
TPD-H2 profile of pre-reduced CuCe-2 sample (after TPR-1 analysis) was measured at temperatures up to 650 °C, which was sufficiently high so that the signal of TCD detector returned to the initial value. This analysis illustrated in Figure 4 reveals that very small amount of H2 was physisorbed on the catalyst surface (peak at approx. 30 °C) and that the majority of H2 stored in the catalyst during the TPR analysis was chemi-sorbed. H2 desorption occurred in a wide temperature range, and two wide peaks are noted. However, in the applied temperature interval, complete desorption of H2 incorporated in the catalyst structure was achieved. This was confirmed also by TPD-H2 examination of CuCe-2 sample, previously used in four TPR/TPO cy-
cles; a profile similar to the one measured after TPR-1 analysis was obtained (Figure 4). By means of TPD analysis of CuCe-2 sample subjected to consecutive TPR/TPO cycling, the same HSCC value as reported above was obtained. It was estimated on the basis of presented TPD spectra (and confirmed by additional TPD experiments) that about 80% of H2 was desorbed from CuCe-2 sample at temperatures up to 400 °C. Since hydrogen incorporated in the catalyst structure was completeh/ consumed during the TPO analysis, carried out by means of diluted oxygen stream in the temperature range of 0—400 °C, this implies that higher temperatures are obviously required to achieve complete hydrogen desorption when the catalyst sample is purged by inert gas. Finally, it is very interesting to note that the locations of peak maxima in measured TPD -H2 profiles (Figure 4), which appear in the temperature range of 150-250 °C, are in very good agreement with the location of peak in predicted Ib curve, showing the consumption of oxygen in TPO experiment that reacted with stored hydrogen (Figure 2b).
The TPD-02 profile illustrated in Figure 5 for CuCe-2 sample pre-calcined at 400 °C lies in a very short range of TCD signals, which alludes that in the applied temperature range of -20 400 ° C the examined solid exhibits, in comparison to measured TPD-H2 values, lower ability for the exchange of oxygen. The amount of desorbed oxygen for given temperature range is listed in Table 1. It is seen that TPD-02 value is about 3 times lower compared with TPD-H2 data obtained in the same temperature range. However, the reported TPD-02 value is in good agreement with data found in the literature.39 Finalh/, hydrogen pulse chemisorption
Pintar et al.    Redox Behavior of Copper-Ceria Mhced Oxide Catalyst
50
Acta Chim. Slov. 2005, 52, 44-52
measurements of CuCe-2 sample after TPD-02 treat-ment (not shown) confirmed that no chemisorption of hydrogen occurred at temperature close to 0 °C.
Conclusions
The results of this study reveal that despite a drop in specific surface area of CuCe-2 mixed oxide sample after the first TPR-H2 treatment, which implies that morphological changes take plače, the redox behavior of this solid prepared by the sol-gel peroxide route remains nearly unchanged as total hydrogen consumption, hy-drogen storage capacity and total oxygen consumption remain constant during successive TPR/TPO cycles.
During the TPR-1 analysis the CuO, which is X-ray amorphous and dispersed over the Ce02 crystallites (with dimension of about 42 nm), reduces to Cu°. The Cu° phase can be detected by X-ray diffraction after the reduction process. This means that the reduction process is accompanied by metal particles agglomera-tion process. However, after subsequent TPO the Cu° is again oxidized back into CuO and re-dispersed over the Ce02 crystallites. Consequently, in subsequent reduction/oxidation cycles the TPR and TPO patterns are not shifted towards higher temperatures.
Experimental
1. Preparation of (CuO)015(CeO2)085 mixed oxide sample (CuCe-2)
CuCe-2 sample was prepared by sol-gel peroxide route method15 by reacting CuCl2-2H20 (99 wt.% pu-rity, Aldrich) with H202 water solution (30 wt.%, p.a., Merck) and in separate vessel by reacting H202 water solution with CeCl3-7H20 (99.9 wt.% purity, Aldrich). The concentration of metal chloride in aqueous H202 solution was altered to attain the desired (Cu:Ce) molar ratio in the oxide sample. After the reaction was com-pleted, the solution was vigorously mixed and the excess peroxide was decomposed at about 60 °C and then the ethanol (molar ratio of ethanol to metallic ions = 30) was added slowly during continuous stirring at room temperature. The solvent was removed by evaporation at about 40 °C, and then the remaining viscous prod-uct was slowly dried to obtain the xerogel, which was calcined at 400 °C for 6 h in dry air.
2. Characterization
2.1. X-ray diffraction analysis
The crystalline phases present in the prepared and treated CuCe-2 samples and the average crystallite sizes were examined by X-ray diffraction using a Philips PW-1710 diffractometer with Cu Ka irradiation source (X =
0.15406 nm) operated at 40 kV and 30 mA. The XRD patterns of samples were measured in 0.04° steps from 7 to 70° (in 26) with 1 s per step. The obtained patterns were compared with PDF data files40 to confirm phase identities. The average crystallite sizes were calculated applying the Scherrer equation41 to the line broaden-ing of diffraction peak from (111) plane in cerianite.
2.2. UV-VIS diffuse reflectance spectroscopy
Diffuse reflectance UV-VIS spectra of CuCe-2 sample as well as of reference materials were recorded at room temperature using a Perkin-Elmer Lambda 40P UV-VIS spectrophotometer equipped with the RSA-PE-19M Praying Mantis accessory, which is designed for diffuse reflectance measurements of horizontally positioned powder samples, pastes or rough surface samples. The Spectralon® white reflectance standard was used to perform the instrument background cor-rection in the range of 200-900 nm. The scans were acquired in duplicates with speed of 120 nm/min and slit set to 4 nm.
2.3. BET, TPR, TPO and TPD measurements
Single-point BET surface area, temperature-pro-grammed reduction (TPR) with hydrogen, tempera-ture-programmed oxidation (TPO) with oxygen, tem-perature-programmed desorption (TPD) of hydrogen, temperature-programmed desorption of oxygen, and hydrogen pulse chemisorption measurements of CuCe-2 sample were performed by means of an automated catalyst characterization system (Micromeritics, model AutoChem II 2920), which incorporates a thermal con-ductivity detector (TCD). The sample loadings were 0.10 and 0.25 g. Prior to BET analysis, the sample was degassed at 200 °C in He. Before starting TPR and TPD-02 runs, the sample was activated under flowing O2(10 vol.%)/He at 400 °C for 90 min. TPR, TPO and TPD experiments were carried out at a heating rate of 5 °C/min. The reactive gas compositions were H2(5 vol.%)/Ar for TPR and O2(10 vol.%)/He for TPO. The flow rate was fixed at 50 mL/min (STP). The total reactive gas consumption (TPR and TPO) and desorption (TPD-02 and TPD-H2) were measured. To convert the peak area data to volume data, the analyzer was cali-brated with gas mixtures of known composition.
The TPR measurements were carried out follow-ing activation after cooling the sample in helium flow to 0 °C. The sample was then held at 0 °C under flowing helium to remove the remaining adsorbed oxygen so that the TCD signal returned to the baseline. Then the TPR experiments were performed up to a temperature 400 °C at which the sample was maintained for 30 min. The trap was cooled with isopropyl alcohol/liquid ni-trogen slurry (IPA/LN2, T = -80 °C). In order to verify
Pintar et al.    Redox Behavior of Copper-Ceria Mhced Oxide Catalyst
Acta Chim. Slov. 2005, 52, 44–52
51
that mass transfer limitations do not affect the TPR measurements, we have carried out TPR experiments at different sample loadings (Le., 0.10 and 0.25 g).
The TPO experiments were performed follow-ing TPR after cooling the samples in H2(5 vol.%)/Ar flow to 0 °C. The samples were then purged at 0 °C in flowing Ar to remove the residual hydrogen. After that O2(10 vol.%)/He gas mkture was passed over the samples which were heated to 400 °C and then held at 400 °C for 1 h.
To examine the reproducibility of TPR/TPO profiles after the reoxidation, the sample was cooled in helium flow to 0 °C at the end of the TPR-1/TPO cycle and then consecutive reduction/oxidation runs were carried out under the same reaction conditions as described above.
The TPD-H2 was performed following the TPR experiments after cooling the reduced sample in H2(5 vol.%)/Ar gas mkture flow to -20 °C. At that temperature, the sample was maintained under flow of pure argon to remove the residual hvdrogen so that the TCD signal returned to the baseline. After that the TPD-H2 experiment was carried out up to a temperature of 650 °C, at which the sample was held for 30 min.
The TPD-02 was carried out on the activated sample, which was cooled in O2(10 vol.%)/He gas mkture flow to -20 °C and then purged in flowing helium to remove the residual oxygen. The TPD-02 measurement was performed up to a temperature of 400 °C, at which the sample was held for 6 h.
The hvdrogen pulse chemisorption measurements of CuCe-2 sample were applied after TPD-02 measurements to find out whether a chemisorption of hvdrogen occurred at temperature close to 0 °C (Le., during TPR runs). The sample was degassed and cooled under flow-ing argon to -5° C, at which pulses of H2(5 vol.%)/Ar were injected into a stream of Ar flowing through the sample bed. The injection loop (nominal volume 0.5 mL) was calibrated with pulses of N2 in helium flow and compared against a calibration line produced from gas tight svringe injections of N2 under helium flow.
Acknowledgements
The authors gratefully acknowledge the financial support of the Ministry of Education, Science and Šport of the Republic of Slovenia through Grant Pl-0152-0104.
References
1. A. Trovarelli, Catal. Rev. - Sci. Eng.  1996, 38, 439-520.
2.  Y. Zhang, S. Andersson, M. Muhammed, Appl. Catal. B
1995, 6, 325–337.
3.  P. Fornasiero, G. Balducci, R. Di Monte, J. Kaspar, V. Sergo, G. Gubitosa, A. Ferrero, M. Graziani, J. Catal.
1996, 164, 173–183.
4.  A. E. C. Palmqvist, E. M. Johansson, S. G. Järas, M. Muhammed, Catal. Lett. 1998, 56, 69–75.
5.  P. Bera, K. R. Priolkar, P. R. Sarode, M. S. Hegde, S. Emura, R. Kumashiro, N. P. Lalla, Chem. Mater. 2002, 14, 3591–3601.
6.  W. Liu, M. Flytzani-Stephanopoulos, J. Catal. 1995, 153, 304–316.
7.  Lj. Kundakovic, M. Flytzani-Stephanopoulos, J. Catal.
1998, 179, 203–221.
8.  A. Martinez Arias, R. Cataluna, J. C. Conesa, J. Soria, J. Phys. Chem. B 1998, 102, 809–817.
9.  W. Liu, A. F. Sarofim, M. Flytzani-Stephanopoulos, Appl. Catal. B 1994, 4, 167–186.
10.   A. Tschöpe, W. Liu, M. Flytzani-Stephanopoulos, J. Y. Ying, J. Catal. 1995, 157, 42–50.
11.   T. Zhu, Lj. Kundakovic, A. Dreher, M. Flytzani-Stephanopoulos, Catal. Today 1999, 50, 381–397.
12.   M. Fernandez-Garcia, E. Gomez Rebollo, A. Guerrero Ruiz, J. C. Conesa, J. Soria, J. Catal. 1997, 172, 146–159.
13.   P. Bera, S. T. Aruna, K. C. Patil, M. S. Hegde, J. Catal.
1999, 186, 36–44.
14.   X. Qi, M. Flytzani-Stephanopoulos, Ind. Eng. Chem. Res. 2004, 43, 3055–3062.
15.   S. Hočevar, J. Batista, J. Levec, J. Catal. 1999, 184, 39–48.
16.   S. Hočevar, U. Opara Krašovec, B. Orel, A. S. Aricó, H. Kim, Appl. Catal. B 2000, 28, 113–125.
17.   G. Avgouropoulos, T. Ioannides, H. K. Matralis, J. Batista, S. Hočevar, Catal. Lett. 2001, 73, 33–40.
18.   G. Avgouropoulos, T. Ioannides, Ch. Papadopoulou, J. Batista, S. Hočevar, H. K. Matralis, Catal. Today 2002, 75, 157–167.
19.   G. Sedmak, S. Hočevar, J. Levec, J. Catal. 2003, 213, 135–150.
20.   D. H. Kim, J. E. Cha, Catal. Lett. 2003, 86, 107–112.
21.   G. Sedmak, S. Hočevar, J. Levec, J. Catal. 2004, 222, 87–99.
22.   G. Sedmak, S. Hočevar, J. Levec, Topics in Catalysis 2004, 30–31, 445–449.
23.   Y. Liu, T. Hayakawa, K. Suzuki, S.Hamakawa, T. Tsunoda, T. Ishii, M. Kumagai, Appl. Catal. A 2002, 223, 137–145.
24.   Y. Liu, T. Hayakawa, T. Tsunoda, K. Suzuki, S. Hamakawa, K. Murata, R. Shiozaki, T. Ishii, M. Kumagai, Topics in Catalysis 2003, 22, 205–213.
25.   J. Papavasillou, G. Avgouropoulos, T. Ioannides, Catal. Commun. 2004, 5, 231–235.
26.   Lj. Kundakovic, M. Flytzani-Stephanopoulos, Appl. Catal. A 1998, 171, 13–29.
Pintar et al.    Redox Behavior of Copper-Ceria Mhced Oxide Catalyst
52
Acta Chim. Slov. 2005, 52, 44–52
27.   P. Zimmer, A. Tschöpe, R. Birringer, /. Catal. 2002, 205, 339-345.
28.   A. Tschöpe, J. Y. Ying, Nanostruct. Mater. 1994, 4, 617-623.
29.   J. Y. Ying, A. Tschöpe, D. Levin, Nanostruct. Mater. 1995, 6, 237-246.
30.   B. Skarman, T. Nakayama, D. Grandjean, R. E. Benfield, E. Olsson, K. Niihara, L. R. Wallenberg, Chem. Mater. 2002, 14, 3686-3699.
31.   A. B. P. Lever, Inorganic Electronic Spectroscopy; Elsevier: Amsterdam, 1984.
32.   C. Lamonier, A. Bennani, A. D´Huysser, A. Aboukais, G. Wrobel, /. Chem. Soc, Faraday Trans. 1996, 92, 131-136.
33.   G. Wrobel, C. Lamonier, A. Bennani, A. D´Huysser, A. Aboukais, /. Chem. Soc, Faraday Trans. 1996, 92, 2001-2009.
34.   J. L. G. Fierro, J. Soria, J. Sanz, J. M. Rojo,/. Solid State Chem. 1987, 66, 154–162.
35.   A. Tschöpe, M. L. Trudeau, J. Y. Ying, /. Phys. Chem. B 1999, 103, 8858-8863.
36.   G. Fierro, M. Lo Jacono, M. Inversi, P. Porta, R. Lavecchia, F. Cioci, /. Catal. 1994, 148, 709-721.
37.   A. Aboukais, A. Bennani, C. F. Aissi, G. Wrobel, M. Guelton, /. Chem. Soc, Faraday Trans. 1992, 88, 1321-1325.
38.   A. Pintar, J. Batista, S. Hočevar,/. Coll. Interface ScL, in print.
39.   H. C. Yao, Y. F. Yu Yao, /. Catal. 1984, 86, 254-265.
40.   Powder Diffraction File; International Center for Diffraction Data, Newtown, 2000.
41.   H. P. Klug, L. E. Alexander, X-Ray Diffraction Procedures (For Polycrystalline and Amorphous Materials); Wiley & Sons: New York, 1954, p. 491 ff.
Povzetek
Sintetizirali smo nanokristalinični (CuO)0.15(CeO2)0.85 oksid po sol-gel metodi z uporabo vodikovega peroksida. Redoks lastnosti oksidnega vzorca smo preučevali s temperaturno programiranimi tehnikami redukcije (TPR), oksidacije (TPO) in desorpcije (TPD). Rezultati analiz so pokazali, da se pri redukciji CuO, ki je dispergiran na površini CeO2 nanokristalitov, zniža specifična površina (CuO)0.15(CeO2)0.85 oksida, medtem ko celotna poraba vodika v zaporedno izvedenih TPR/TPO ciklih ostane konstantna. To pomeni, da se visoka stopnja disperzije CuO ohrani celo po aglomeraciji CeO2 nanokristalitov v gostejšo sekundarno strukturo. Med TPR procesom poteče popolna redukcija CuO v Cu0 in na površini delno reduciranega CeO2 se tvorijo skupki kovinskih delcev. V TPO stopnji naknadno poteče popolna reoksidacija Cu0 in ponovna disperzija nastalega CuO na površini CeO2 kristalitov.
Pintar et al.    Redox Behavior of Copper-Ceria Mhced Oxide Catalyst