A facile approach for the preparation of biomorphic CuO–ZrO2 catalyst for catalytic combustion of methane

Applied Catalysis A: General 423–424 (2012) 121–129

Contents lists available at SciVerse ScienceDirect

Applied Catalysis A: General journal homepage: www.elsevier.com/locate/apcata

A facile approach for the preparation of biomorphic CuO–ZrO2 catalyst for catalytic combustion of methane Jingjie Luo a , Huiyuan Xu a,b , Yuefeng Liu a,c , Wei Chu a,∗ , Chengfa Jiang a,∗ , Xiusong Zhao d a

Department of Chemical Engineering, Sichuan University, Chengdu 610065, China Department of Chemical Engineering, Yibin University, Yibin, Sichuan 644007, China Laboratoire des Matériaux, Surfaces et Procédés pour la Catalyse, UMR 7515 du CNRS, Université de Strasbourg, 25, rue Becquerel, 67087 Strasbourg Cedex 08, France d Department of Chemical and Biomolecular Engineering, National University of Singapore, Singapore 117576, Singapore b c

a r t i c l e

i n f o

Article history: Received 28 August 2011 Received in revised form 12 February 2012 Accepted 17 February 2012 Available online 24 February 2012 Keywords: Bio-template method Biomorphic CuO–ZrO2 catalyst Methane catalytic combustion

a b s t r a c t A series of novel biomorphic CuO–ZrO2 catalysts were prepared using a cotton bio-template and compared with conventional CuO–ZrO2 catalysts. The physical and chemical properties of the as-obtained catalysts were characterized by techniques including X-ray diffraction (XRD), scanning electron microscope (SEM), transmission electron microscope (TEM), H2 -temperature programmed reduction (H2 -TPR), and O2 -temperature programmed desorption (O2 -TPD). The catalytic combustion of methane was chosen as the probe reaction. The results suggested that the bio-template method prepared porous biomorphic CuO–ZrO2 catalysts consist of hollow microtubes. Comparing with conventional CuO–ZrO2 catalysts, biomorphic CuO–ZrO2 catalysts displayed better reducibility and oxygen mobility, stronger metal-oxides synergistic effect, appropriate particle size distribution, and lower activation energy. The crystalline state of zirconia transformed from a single crystallite phase of t-ZrO2 into a complex of m-ZrO2 and t-ZrO2 after introducing the bio-template. With proper CuO content (20 mol%), the biomorphic CuO–ZrO2 catalyst displayed preponderant properties. The compensation of surface lattice oxygen from bulk lattice oxygen was more available at high reaction temperatures. © 2012 Elsevier B.V. All rights reserved.

1. Introduction Heat production is the main driving force of developing new technologies for fuels combustion. Because of the lowest CO2 emission during the same amount of energy production, methane is viewed as the best fossil fuel for environmental requirements. The catalytic combustion of methane (CCM) becomes a promising approach for decreasing pollutant and also provides wide applications in the industry [1]. Accordingly, approaches have been adopted for preparing catalysts with high activity and stability. These catalysts include precious or promoted transition metals, perovskites and hexaaluminates [1–3]. Comparing with the most efficient precious metal oxide catalysts, copper oxide is considered as a good replacement due to its comparable catalytic performance and low cost [4–6]. Besides, it is well known that ZrO2 possesses high thermal stability and durability at high temperatures. It was reported that zirconia suppressed the sintering of the active components when they were dispersed mutually [7,8]. Zirconia is also the only metal oxide

∗ Corresponding authors. Tel.: +86 28 85403836; fax: +86 28 85461108. E-mail addresses: [email protected], [email protected] (W. Chu), [email protected] (C. Jiang). 0926-860X/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.apcata.2012.02.025

with four important chemical properties, namely acidity, basicity, reducing ability and oxidizing ability, thus it is widely used as support for the activation and stabilization of copper oxide [9]. In the past decades, the investigation and modification of zirconia supported copper oxides have been widely carried out. Jackson and Ekerdt [10,11] have confirmed that the oxygen anion vacancy performed as the active site for CO hydrogenation over zirconia or Y2 O3 doped zirconia. The relationship between the mobility of oxygen and reaction rate was found out. Águila et al. [2] investigated the methane oxidation over copper oxides supported on various solids. They observed that zirconia supported copper oxide was more efficient than the CuO supported on Al2 O3 and SiO2 . Qu et al. [12] anchored the copper oxide on two kinds of zirconia carriers, demonstrating that the textural structure and property of zirconia carrier greatly affected the catalytic activity for CCM. However, the activity of CuO–ZrO2 catalysts for CCM was not as excellent as for the other oxidizing reactions. Recently, researchers have endeavored to find approaches for improving the activity of CuO–ZrO2 catalysts. These investigations are mainly classified as two groups, modifying the chemical and physical properties of the CuO–ZrO2 catalysts by: (i) adding promoters such as ZnO, CeO2 , and other metal oxides [13–16]; (ii) applying novel preparation methods such as glycine-nitrate combustion synthesis, sol–gel method, surfactant-assisted method [17–20],

122

J. Luo et al. / Applied Catalysis A: General 423–424 (2012) 121–129

which take both money and time as the cost. Facile and economic method for preparing CuO–ZrO2 catalyst efficient for the CCM is still required. The activities of catalysts are known to be critically dependent on several factors such as the preparative routes, composition structures and surface properties [21–23]. Varying the preparation routes often leads to changes on the physical and chemical properties of catalysts. Since the preparation and structure of the catalyst act as an important role for the catalytic activity, researchers are trying to improve the applications of catalysts via regulating their microstructures [24–26]. In recent years, attention has been focused on the materials with biomorphic structures. Prion proteins, agaroses, active carbon fibers, and sponges have been selected as templates to produce various advanced biomorphic materials [27–29]. The limitless options for creating materials have opened up new technological perspectives. However, to the best of our knowledge, there is still no literature investigating CCM over CuO–ZrO2 catalyst preparing by using cotton as template. In this study, biomorphic CuO–ZrO2 microtubes using cotton as template were prepared for the first time and compared with conventional catalysts. This method is facile and inexpensive, but efficient and largely reduces the cost from adding promoter to the CuO–ZrO2 catalysts. The physical and chemical properties of biomorphic CuO–ZrO2 catalysts and their performance for CCM were investigated systematically.

2. Experimental 2.1. Catalyst preparation The bifunctional CuO–ZrO2 samples were prepared by using absorbent cotton as template. A typical procedure to prepare the bio-template CuO–ZrO2 catalyst was as follows: appropriate amounts of Cu(NO3 )2 ·3H2 O and Zr(NO3 )4 ·5H2 O (making the total mass of catalysts as 3 g after calcinations, the final composition of catalyst is assumed to be CuO and ZrO2 ) were applied as the raw materials and were dissolved into 50 ml of distilled water. About 1 g dry cottons were then soaked into the above solution. The cotton submersed with a slightly excessive solution was optimal. Then the mixture solution was aged for 2 h and preliminarily dried in the water bath at 50 ◦ C. After the as-prepared precursor was dried at 80 ◦ C over night, it was calcined at 400 ◦ C for 3 h in a muffle furnace. Five samples with different CuO molar contents were prepared as above. The obtained catalysts were labeled as C0.05 ZB, C0.10 Z-B, C0.15 Z-B, C0.20 Z-B and C0.30 Z-B, which meant that the CuO/(CuO + ZrO2 ) molar contents were 5%, 10%, 15%, 20% and 30%, respectively. For comparison, two other CuO–ZrO2 catalysts, all with 10 mol% CuO were prepared by two different conventional methods, impregnation and co-precipitation methods. The 2 M NaOH aqueous solution was applied as precipitant. In the impregnation method (IMP), the zirconia precursor after precipitation was aged for 2 h and followed by washing and filtering. After drying and calcination, 2.58 g zirconia was impregnated into the copper nitrate aqueous (with 1.27 g Cu(NO3 )2 ·3H2 O solid) and aged for 2 h for preparing the catalyst. In the co-precipitation (CP) method, the procedure was familiar with the IMP method except that the copper and zirconium nitrates were precipitated together and the impregnation process of support was canceled. The as-prepared samples were dried at 80 ◦ C over night and calcined at 400 ◦ C for 3 h in a muffle furnace. The acquired catalysts were labeled as C0.10 Z-I and C0.10 Z-C, respectively. All the samples experienced no further pretreatment and directly tested by the CCM reaction.

2.2. Characterization of catalysts The phase purity of the sample was confirmed by X-ray diffraction (XRD) measurement. It was performed with an MPD type ´˚ X’pert powder diffractometer equipped with Cu-K␣ ( = 1.54056 A) radiation, which was operated at 40 kV and 30 mA for 2 angles ranging from 10◦ to 80◦ . The particle sizes were calculated from the Scherrer equation. To obtain surface textural details of the support, the morphology and structure of the catalysts were studied using a scanning electron microscope (SEM, JEOL/EO, JSM-5900). TEM measurement was performed on the JEOL-JEM-200CX transmission electron microscopy, and the sample was dispersed in ethanol by ultrasonic pretreatment. For the measurement of H2 temperature programmed reduction (H2 -TPR), 50 mg catalyst was placed into a quartz reactor and the sample was saturated in a gas stream of 5.0% (v) H2 in nitrogen at a total flow rate of 30 mL/min. After the stabilization, temperature increased at a rate of 10 ◦ C/min from 50 ◦ C to 750 ◦ C. The amount of H2 consumption was analyzed online via a thermal conductivity detector (TCD). The steps of O2 temperature programmed desorption (O2 -TPD) were similar with our previous work [30–32], 200 mg of fresh catalyst was loaded, and adsorbed in O2 at 300 ◦ C for 60 min. After the powder was cooled to 50 ◦ C, it was blown by N2 for 120 min. The catalyst was then heated to 750 ◦ C at a linear heating rate of 10 ◦ C/min in the N2 flow. The effluent gas was analyzed with a mass spectrometer. 2.3. Measurement of catalytic performances Catalytic activity measurements for methane combustion were carried out in a fixed-bed reactor (di = 6 mm). 200 mg catalyst was mixed with 800 mg quartz sand and packed into the quartz reactor tube. The temperature of the catalyst bed was measured by a thermocouple. The exhaust gas stream was monitored with a gas chromatograph equipped with a TDX-01 column. The reaction flow was a mixture of 5 vol.% CH4 /10 vol.% O2 (balance Ar, GHSV = 45 000 h−1 ). Methane combustion was carried out in the temperature range of 400–700 ◦ C at a heating rate of 5 ◦ C/min. In all measurements, CO2 and H2 O were the only detected products. Methane conversion was expressed as: x% =

CH4 0 − CH4 T CH4 0

× 100

In this function, CH4 0 was the initial CH4 concentration (350 ◦ C) and CH4 T was the concentration at a given temperature T. 3. Results and discussion 3.1. Crystal structure and morphology The XRD profiles of Cx Z-B catalysts with different copper oxide contents have been displayed in Fig. 1. The inset profile is the comparison of C0.10 Z-I and C0.10 Z-C catalysts. No diffraction peak of CuO is observed in the C0.05 Z-B and C0.10 Z-B catalysts. It is shown that the amount of non-crystalline CuO has a threshold limit value at 10 mol%. When the CuO/(CuO + ZrO2 ) molar content increases from 10 mol% to 30 mol%, the diffraction peaks of copper oxide (JCPDS 80-1268) [33] are observed at 2 of 35.5◦ and 38.7◦ and become stronger, indicating a continuous increase in the crystallization of CuO. The CuO crystalline sizes of C0.15 Z-B, C0.20 Z-B and C0.30 Z-B catalysts are calculated by Scherrer equation, which are 15.78, 17.11, 23.45 nm, respectively. In the C0.10 Z-I sample, only the diffraction peaks of tetragonal ZrO2 (2 = 30◦ and 34.8◦ ) [34] can be observed. The C0.10 Z-C sample possesses a weaker and broader peak in the range of 20–50◦ ,

Intensity (a.u.)

J. Luo et al. / Applied Catalysis A: General 423–424 (2012) 121–129

20

30

40

50

60

2 Theta (degree) Fig. 1. XRD profiles of five samples with different CuO contents and the comparable C0.10 Z-I and C0.10 Z-C samples.

which is a characteristic of typical amorphous structure of tetragonal ZrO2 . Different from the C0.10 Z-I and C0.10 Z-C samples, the diffraction peaks of both tetragonal and monoclinic ZrO2 (t-ZrO2 and m-ZrO2 ) are observed in all the five biomorphic samples. With the increase of the copper oxide content, the diffraction peaks of ZrO2 become broader and more diffused, suggesting the existence of smaller or higher dispersed zirconia species. However, a small increase in diffraction intensity of both CuO and t-ZrO2 is observed in the C0.30 Z-B sample, which represents a slight increase in crystallite size of CuO and t-ZrO2 . These results confirm that phase transformation of zirconia takes place when the bio-template method is applied. Besides, the CuO molar content is important for the transformation between t-ZrO2 and m-ZrO2 in the biomorphic catalysts. The surface morphologies of samples prepared from three different methods were observed by scanning electron microscopy (SEM) as shown in Fig. 2. The SEM image of C0.10 Z-I sample (Fig. 2(c)) shows that the copper oxides are heterogeneously loaded on the surface of large zirconia bulk. The magnified images of the conventional C0.10 Z-I sample shows clusters with planar faces, whereas the C0.10 Z-C sample displays homogeneous clusters formed by spherical particles. The C0.10 Z-B sample presents mainly microtubes about 50–100 ␮m long. The magnified SEM image (Fig. 2(b)) of the C0.10 ZB sample displays a single microtube with diameter around 7 ␮m. Excess particles formed during the preparation are dispersed on the surface of microtubes. The SEM images of C0.20 Z-B sample are shown in Fig. 2(e) and (f). Although there are massive structures, typical microtubes with length of 50–150 ␮m can be observed. The calcinations temperature used in this work is based on the results of work in our group [35,36]. The XRD analysis without any carbon peaks confirms that those microtubes are composed of zirconia and copper oxides. We also analysis several parts of the fiber-like tubes in the C0.10 Z-B and C0.20 Z-B samples by SEM-EDS, as shown in Fig. 2(g). The results evidence that the fiber-like microtubes is not the residual cotton fiber but consisted of CuO and ZrO2 . The catalyst duplicates the morphology of template and exhibits biomorphic structure. Fig. 3 shows the TEM images of C0.10 Z-B and C0.10 Z-I catalysts and their particle size distribution. It is observed that the profiles of the two catalysts differ from each other very much. The C0.10 Z-B catalyst, which is prepared by using bio-template, displays fibrous morphology. The magnified profiles of C0.10 Z-B sample (Fig. 3(b)) exhibits globular mesoporous structures with the pore size about 2–5 nm, depicting that catalysts are porous and constructed by nanoparticles. As suggested by Boettcher et al., the interconnected pores in the mesoporous materials facilitates the interaction of materials and effective hole transport [24]. Different

123

from the C0.10 Z-B sample, parts of particles in the C0.10 Z-I sample tend to be clustered on the surface of the bulk-like support as shown in Fig. 3(c). The interaction between the CuO particles and the zirconia support seems not very strength since parts of the particles have been separated from the support. The sizes of particles are measured automatically for each catalyst and the particle distribution of two catalysts is plotted. The average particle size is calculated to be 8.64 and 11.42 nm for C0.10 Z-B and C0.10 Z-I samples, respective. It is also observed that the distribution range of the C0.10 Z-I catalyst is broader, mainly centralizes in the range of 2–15 nm. The distribution range of the C0.10 Z-B catalyst is relative narrower and mainly centralizes in the range of 2–8 nm. In other words, the dispersion of active component in the C0.10 Z-B catalyst is largely improved than that in the C0.10 Z-I catalyst. In the biotemplate method prepared catalysts, it is difficult to distinguish the CuO and ZrO2 particles from the TEM images. However, undeniable factor is that both the CuO and ZrO2 particles are better dispersed than those in the conventional C0.10 Z-I catalyst. Thus the synergistic effect can be strengthened from higher component dispersion and smaller crystallite size [37]. In addition, for the active component interacts with smaller carrier particles, the catalytic performance tends to be more distinctive than those with larger carrier particles according to Xu et al. [38]. The higher dispersion of particles in C0.10 Z-B catalyst thus provides much more active centers than the conventional C0.10 Z-I catalyst, which acts as one of the key factor for the catalytic activity. In order to understand the impact of CuO content on the particle size distribution and average particle size, the TEM images and particle size distribution of C0.05 Z-B and C0.20 Z-B catalysts are shown in Fig. 4. Mesoporous structures and fibrous morphology also exist in the C0.05 Z-B and C0.20 Z-B catalysts, which are porous and constructed by uniform nanoparticles. However, several clustered particles exhibit in the C0.20 Z-B samples as shown in the red dotted circles, which is in accordance with the previous results from XRD. In comparison, for the catalyst surface where the catalytic reactions take place, the TEM technique provides more information than the XRD technique [39]. It is common sense that such clusters are futile and even result in a decrease of the active sites amount, which might be harmful for the catalytic activity of CCM [40]. However, such statement seems to be contradictory with the fact that the C0.20 Z-B catalyst never possesses inferior activity during the CCM reaction as shown in Fig. 7. In order to understand this phenomenon, we further profile the particle size distributions from the TEM profiles, as shown in Fig. 4(b). Most particles of the two typical catalysts are 2–6 nm. The amounts of particles smaller than 15 nm in both the two samples are calculated to be 86.7% and 85.2%, respectively. It is a worth note that the ratio of the smaller particles (<15 nm) in the C0.05 Z-B, C0.20 Z-B and above-mentioned C0.10 Z-B samples differ from each other not very much (all about 85–87%). The number of smaller particles is hardly influenced by the CuO content. In this way, those clusters on the surface of C0.20 Z-B catalyst are attributed to the bulk CuO species that are formed from the excessive CuO. Generally speaking, the activity of catalysts depends on the smaller particles of the active component in the copper based catalyst, which provide more active sites [40,41]. In this case, although aggregates and larger average particle size exist, the catalytic activity of C0.20 Z-B sample has been barely blocked. 3.2. Redox property and oxygen mobility According to the Mars-van Krevelen mechanism [42] that is mostly used to describe the oxidation of organic compounds over metal oxides, the reaction takes place through alternative oxidation and reduction on the catalyst surface. Temperature program reduction techniques are performed to get information of sample’s

124

J. Luo et al. / Applied Catalysis A: General 423–424 (2012) 121–129

Fig. 2. SEM images of four typical Cx Z-B catalysts: C0.10 Z-B (a) and (b), C0.10 Z-I (c), C0.10 Z-C (d), C0.20 Z-B (e) and (f), and SEM-EDS analysis of C0.10 Z-B and C0.20 Z-B samples (g).

redox properties. The typical H2 -TPR profiles of three CuO–ZrO2 catalysts from different preparation methods are shown in Fig. 5(a). A separate study carried out over pure CuO (not shown) reveals that only a very large peak exists at about 350 ◦ C, which is in accordance with the result reported by Cao et al. [33]. No reduction peak of pure zirconia is observed from 100 ◦ C to 800 ◦ C. According to Fig. 5(a), it is noticeable that the copper oxide species in C0.10 Z-C displays one sharp peak, while the C0.10 Z-B and C0.10 Z-I samples illustrate more than one peak unlike the pure CuO sample. As revealed in the previous studies [31,33,43], the complex of copper with other metal oxides causing the pure CuO reduction peak splitting into several overlapped peaks. This is attributed to the different synergistic effects between metal oxides by various preparation methods. In the C0.10 Z-B and C0.10 Z-I samples, the overlapped peaks are assigned to the reduction of the copper oxide differently interacted with the zirconia. In addition, both the reduction peaks in the three samples shift to lower temperature range comparing to pure CuO. It is suggested that the reduction of copper oxide is promoted due to the synergy between copper oxide and zirconia [44]. For the C0.10 Z-I sample, the temperature of the first reduction peak is about 60 ◦ C lower than that of the C0.10 Z-C sample. But the peak temperature corresponding to the following peak trails to 300 ◦ C, much more hysteretic than the other two samples. This is because of the coexistence of aggregated CuO species for lower surface energy and finer CuO species

impregnated on the surface. The reduction of copper component on the surface is easier while the reduction of the bulk like CuO is much more difficult. In the C0.10 Z-C sample, the copper oxide and zirconia are simultaneously precipitated, the strong combination between CuO and ZrO2 makes the reduction of CuO more difficult at higher temperature. The H2 consumption of the reduction peaks from 100 ◦ C to 500 ◦ C, as shown in Table 1, is calculated by using the reduction peak area of pure CuO as calibration. The H2 uptake over the C0.10 Z-I sample is higher than that over the C0.10 Z-B sample. However, the peak temperature of C0.10 Z-I sample shifted to higher temperature range, and the peak intensity corresponding to the Table 1 H2 -TPR quantitative result and some kinetic data of as-prepared catalysts. Samples

C0.10 Z-I C0.10 Z-C C0.05 Z-B C0.10 Z-B C0.15 Z-B C0.20 Z-B C0.30 Z-B a b

H2 -TPR Ta (◦ C)

H2 uptakeb (mmol/gcat )

201 266 162 160 160 158 184

0.94 0.74 0.43 0.79 1.53 1.96 3.70

Ea (kJ/mol)

ln A

160.32 164.20 150.99 125.01 113.34 98.10 122.30

23.82 23.45 22.32 18.43 16.31 14.41 18.16

The temperature corresponding to the first reduction peak in Fig. 5. The total H2 consumption of the reduction peaks from 100 ◦ C to 500 ◦ C.

J. Luo et al. / Applied Catalysis A: General 423–424 (2012) 121–129

125

Fig. 3. TEM images of C0.10 Z-B (a) and (b), C0.10 Z-I (c) samples, and particle size distributions of two CuO–ZrO2 samples (d).

first reduction peak is similar to that of the C0.10 Z-B sample. It suggested that the excess impregnated copper species transferred from surface to bulk, which contributes little for the catalytic activity [2]. The biomorphic C0.10 Z-B sample possesses the absolute predominance of low peak temperature and large peak area, which evidence better oxygen mobility than the other two catalysts. According to the research of Jackson and Ekerdt [10], the higher the mobility of oxygen was, the higher the reaction rate became. Besides, it is a worth note that the crystalline states of ZrO2 is one of the factors impacting the reducibility of supported CuO as previously reported [45]. Guo et al. [17] also found that the CuO supported on the tZrO2 is reduced at higher temperature than that supported on the m-ZrO2 . From the improved active component reducibility and the oxygen mobility, it is inferred that there exists stronger synergy between copper oxide and zirconia after the addition of biological template. Fig. 5(b) shows the H2 -TPR profiles of five Cx Z-B catalysts. Since there is no obvious peak after 350 ◦ C, only the peaks ranging from 100 ◦ C to 350 ◦ C are illustrated. The profiles of Cx Z-B catalysts again differ from that of pure CuO very much. The overlapped reduction peaks are present in all the samples. The TPR curves of the catalysts vary in the maximum temperature and intensity of peaks, depending upon the CuO content in biomorphic Cx Z-B catalysts. Choudhary et al. [46] reported that the TPR curves essentially reflected the reactivity of the oxygen species. That is, the higher the peak temperature is, the lower reactivity the oxygen species performs. In Fig. 5(b), the ␣ peak with a maximum temperature around 160 ◦ C corresponds to the reduction of the highly dispersed surface copper oxide that strongly interacts with zirconia [47]. The ␤ and ␥ peaks

are assigned to the reduction of the larger bulk CuO particles less associating with zirconia. The temperatures corresponding to the first reduction peaks are listed in Table 1. The temperatures of the first peaks in biomorphic Cx Z-B samples are generally located at about 160 ◦ C except the C0.30 Z-B sample. Interestingly, the intensity of the first peak hardly changes after the CuO molar content reaches to 20 mol%. This phenomenon is similar to the previous report [48]. When the CuO content reaches a proper amount, the excess of CuO forms bulk CuO clusters. Combining the XRD results, the introducing of bio-template has changed the crystalline state of zirconia to be a complex of m-ZrO2 and t-ZrO2 . The variations of the TPR peak temperature and intensity not only strongly depend on the CuO molar content, but also have relationships with the crystalline state of ZrO2 (m-ZrO2 and t-ZrO2 ). When the content of CuO is higher than 20 mol%, the t-ZrO2 crystallite phase become intensive and the oxygen mobility in the C0.30 Z-B catalyst greatly decreases. The H2 uptake of the reduction peaks from 100 ◦ C to 500 ◦ C is also shown in Table 1. Obviously, along with the increase of the copper oxide content, the H2 consumption enhances. However, the temperature of ␣ peak in the C0.30 Z-B sample quickly increases to 184 ◦ C, implying the deceased oxygen mobility [44]. The ␤ and ␥ peaks of the C0.30 Z-B sample also shift to much higher temperature and become wider and more intensive, indicating the obvious clustering of bulk CuO species in the C0.30 Z-B sample [49]. The much lower temperature of the first peak in the C0.20 Z-B catalyst demonstrates the better oxygen mobility and reducibility.

126

J. Luo et al. / Applied Catalysis A: General 423–424 (2012) 121–129

Fig. 4. TEM images (a) and particle size distributions (b) of two typical biomorphic Cx Z-B samples.

Temperature programmed desorption of oxygen (O2 -TPD) is also an effective method to determine the mobility of oxygen species. The O2 -TPD profiles of three samples prepared by different methods are performed as displayed in Fig. 6(a)–(c). The dotted lines are the results of fitting (Gauss model). The relevant peak areas and temperatures are given in Table 2. No obvious desorption peak of pure zirconia is observed in the experimental temperature range, which correlates with the previous report [50]. All the curves contain two major peaks, labeled as ␣ and ␤ peaks, respectively. The broad ␣ peak can be fitted into two peaks (␣1 and ␣2 ). The ␣1 desorption peak at about 270 ◦ C is related to molecular O2 , O2 − and O2 2 adsorbed on oxygen vacancies [32,51]. The ␣2 peak is likely related to the release of surface lattice oxygen. The ␤ peak is

Table 2 Quantitative result from O2 -TPD of relative samples. Sample

C0.10 Z-I C0.10 Z-C C0.10 Z-B C0.15 Z-B C0.20 Z-B C0.30 Z-B a

Peak temperature (◦ C)

Relative peak area

␣ peak

␣ peak

␤ peak

0.20 0.58 1.00 1.12 1.50 0.74

1.21 1.58 1.00 1.30 1.41 1.79

␤ peak

␣1

␣2

␤1

␤2

276 265 269 259 250 312

312 367 342 330 327 382

689 628

774 773 592a 589a 581a

569

663

The temperature is corresponding to the total ␤ peak.

(b)

(a)

C0.10Z-C C0.10Z-I

Intencity (a.u.)

Intensity (a.u.)

β γ

C0.20Z-B C0.15Z-B C0.10Z-B C0.05Z-B

C0.10Z-B 100

200

300

Temperature (ºC)

400

500

C0.30Z-B

α

100

200

300

Temperature (ºC)

Fig. 5. H2 -TPR profiles of three samples from three preparation methods (a) and five Cx Z-B catalysts prepared by bio-template method (b).

O2 consumption (mmol/g)

0.30 0.46 0.43 0.52 0.63 0.55

J. Luo et al. / Applied Catalysis A: General 423–424 (2012) 121–129

663 581

327

Intensity (a.u.)

774

689 312

276 300

400

589 330

342

d

259

592

269

Intensity (a.u.)

250

628

367

e

b

a

382

312

265

569

f

773

c

127

500

600

700

800

Temperature (ºC)

300

400

500

600

700

800

Temperature (ºC)

Fig. 6. O2 -TPD profiles of CuO–ZrO2 catalysts: C0.10 Z-B (a), C0.10 Z-I (b), and C0.10 Z-C (c); C0.15 Z-B (d), C0.20 Z-B (e), and C0.30 Z-B (f).

related to the desorption of lattice oxygen bound to metal cations in the bulk of the catalyst, which replenishes and enhances the surface lattice oxygen during high temperature desorption processes [52]. It is remarkable from Table 2 that the total O2 consumptions of the C0.10 Z-B and C0.10 Z-C samples are similar, which are much larger than that of the C0.10 Z-I sample. Besides, the C0.10 Z-B sample displays the ␣ peaks with much larger area and slightly lower temperature than those in the other two samples, suggesting the stronger synergy among metal oxides in the biomorphic C0.10 Z-B sample [53]. Such strong synergy between metal oxides weakens the M O bond and thereby enhances the mobility or reactivity of ␣-oxygen. In addition, the ␤ peak in the C0.10 Z-B sample is narrow and symmetrical. For different preparation methods, the ␤ peaks in C0.10 Z-I and C0.10 Z-C samples split into two overlapping peaks (␤1 and ␤2 peaks), which can be explained by different migration mechanisms of bulk oxygen due to different metal oxides interaction [53]. The ␤ desorption peak of biomorphic C0.10 Z-B sample shifts to much lower temperature along with the reduced area, implying that the mobility or activity of the bulk oxygen can be more easily migrated than those in the other two samples. The smaller ␤ desorption peak in the C0.10 Z-B sample is attributed to the migration of oxygen species from bulk to surface [54]. From Fig. 6(b) and (c), although the ␣1 oxygen of C0.10 Z-I and C0.10 Z-C samples locate at low temperature, the amounts of oxygen species are low and the ␤ oxygen species are too difficult to be desorbed under experimental condition of CCM in this work.

Other three biomorphic Cx Z-B catalysts are chosen as typical samples and the corresponding O2 -TPD results are illustrated in Fig. 6(d)–(f). Combining the O2 -TPD result from Fig. 6(a), it is found that the ␣1 and ␣2 peaks locate at lower temperature region with increasing peak areas as the CuO molar content raises until it reaches 30%. This can be attributed to the accelerated synergy between metal oxides. The appropriate increase of CuO content facilitates the release of active oxygen species. Meanwhile, the amount of ␤ oxygen drastically reduced as well as the desorption peak temperature shifts to slightly lower region. However, for the C0.30 Z-B sample, the intensity of ␣ peak lowered obviously and the desorption temperature raises. The ␤ desorption peak is wide and asymmetrical, splitting into two overlapping peaks. The mobility or activity of ␤ oxygen drops suddenly in the C0.30 Z-B sample. It is manifested that, there is an optimum content of copper oxide making the compensation of surface lattice oxygen from bulk lattice oxygen more available. Taking the results from XRD and H2 -TPR into consideration, it can be inferred that the excess CuO species enter into the bulk phase instead of dispersing on the surface when the CuO content is higher. Such clustering of CuO particles weakens the synergy between CuO and ZrO2 , lessens the oxygen mobility, and possibly block the transformation of active oxygen species during the reaction. It is concluded that the C0.20 Z-B catalyst is highly effective at improving the mobility of oxygen species. In a word, the C0.20 Z-B catalyst possesses better characteristics in not only biomorphic structures and appropriate particle size distribution, but also reducibility and oxygen mobility.

128

J. Luo et al. / Applied Catalysis A: General 423–424 (2012) 121–129

100

C0.30Z-B

80

-2.8 -1.6

60

-2.0

y=18.16-14.71x C0.20Z-B

-2.4

40

y=14.42-11.80x

-2.0 20 0 400

lnkCH

CH4 Conversion (%)

-2.4

450

500

550

600

650

y=16.31-13.63x

-2.4 -1.6

700

C0.10Z-B y=18.43-15.04x

-2.0

Temperature (ºC) Fig. 7. Methane catalytic combustion over various CuO–ZrO2 catalysts: () CuO, () C0.10 Z-I, () C0.10 Z-C, () C0.05 Z-B, () C0.10 Z-B, () C0.15 Z-B, () C0.20 Z-B, and (䊉) C0.30 Z-B.

C0.15Z-B

-2.4 C0.05Z-B

-2.0

y=22.32-18.16x

-2.4 -2.8 1.32

3.3. Catalytic activity The catalytic activity for CCM over series of CuO–ZrO2 catalysts are shown in Fig. 7. The CO2 is the only production detected by the chromatograph. Among the three catalysts prepared by different methods, the C0.10 Z-B catalyst exhibits the highest methane conversion throughout the temperature range studied. The lightoff temperature (the temperature of 10% methane conversion) of the C0.10 Z-I catalyst is about 50 ◦ C lower than that of the C0.10 ZC catalyst. When the reaction temperature reaches 575 ◦ C, the catalytic activities of both the impregnation and co-precipitation methods prepared catalysts display a flat. The catalytic activity of the C0.10 Z-I catalyst then drops clearly than the other samples. Such phenomenon might result from the high reaction temperature that limits the thermal stability of copper-based catalyst and forces the formation of CuO species clusters that poison the catalysts. It is remarkable that 90% conversion of methane could be achieved over the bio-template prepared C0.10 Z-B catalyst at a relatively low temperature of 590 ◦ C. The CH4 conversion over biomorphic Cx Z-B samples as a function of reaction temperature is also investigated in Fig. 7. All the curves increase monotonically among the experimental temperature range. The pure CuO sample is not able to achieve 50% CH4 conversion in the temperature range shown. A moderate amount of ZrO2 addition increases the catalytic activity of CuO species. The catalytic activity of biomorphic Cx Z-B catalysts enhanced with the increase of copper oxide molar content from 5 mol% to 20 mol%, but then decreased with the further increase of CuO content. The finding suggests that there is a critical amount of active component content. When the copper oxide content is fewer, along with the increase of CuO content, more catalytically active centers consisting of copper particles are formed and hence the reaction rates are accelerated consequently. Whereas, with high content of copper species, the excess CuO species would not only lead to the bulk CuO, but also cover some of the active sites, which negatively affect the catalytic activity [55].

1.34

1.36

1.38

1.40

1.42

-1

1000/T (K ) Fig. 8. Arrhenius plots according to Eq. (2) for the Cx Z-B catalysts.

design equation of the plug-flow reactor and the Arrhenius equation, the apparent activation energy (Ea) and preexponential factor (A) can be given by CH4 conversion (x) and the reaction temperature (T) as below ln[−ln(1 − x)] = ln A −

Ea RT

(2)

Arrhenius plots according to Eq. (2) are shown in Fig. 8. Since higher methane conversion might induce external diffusion limitations, the methane conversion involved in this figure is lower than 20%. The pseudo-first-order model of methane makes a good linear relationship of the data. The Ea and ln A of CCM over different catalysts can be calculated via the Arrhenius plots as a function of temperature. The obtained ln A of biomorphic Cx Z-B catalysts changes the same trend as Ea as compensation (Table 1). The apparent activation energies of Cx Z-B catalyst with CuO molar content from 5 mol% to 20 mol% decrease from 150.99 to 98.10 kJ/mol. The reducing trend of the Ea is in accordance with the increase of oxygen mobility of catalysts as well as their catalytic activity for CCM. The ln A and Ea of conventional C0.10 Z-I and C0.10 Z-C samples are also listed in Table 1. The Arrhenius plots over the C0.10 Z-I and C0.10 Z-C catalysts (not shown) are nearly parallel with very close value of Ea, which are much larger than that of the bio-template method prepared catalysts. The Ea values of the samples fall in the range of activation energies reported for methane combustion over various catalysts [1,58]. Besides, the Ea value of biomorphic C0.20 Z-B is very close to that of some precious metal catalysts (70–100 kJ/mol) [59]. Therefore, it can be claimed that the Cx Z-B catalysts in the present study have enough catalytic activity to catalyze the methane combustion for practical purposes.

3.4. Apparent activation energy from Arrhenius equation 4. Conclusion According to the previous researches, the rates are considered to be first-order in the CH4 partial pressure, and the order with respect to oxygen might vary from 0 to 0.5 [3,56,57]. For a simple first-order kinetic model, the intrinsic reaction rate can be expressed as rCH4 = kPCH4

(1)

In the equation, k and PCH4 are the intrinsic reaction rate constant and partial pressure of CH4 , respectively. Combining the

In this work, the CuO–ZrO2 catalysts were prepared by using the bio-template method and regulated with different CuO contents. They have possessed a higher performance in the catalytic combustion of methane (CCM) than those conventional methods prepared CuO–ZrO2 catalysts. The biomorphic CuO–ZrO2 catalyst with a proper 20 mol% CuO displayed the best activity for CCM. The following were the main conclusions:

J. Luo et al. / Applied Catalysis A: General 423–424 (2012) 121–129

• The application of bio-template has formed the biomorphic structure and good particle size distribution in the as-prepared CuO–ZrO2 catalysts. • The crystalline state of zirconia transformed from a single crystallite phase of t-ZrO2 into a complex of m-ZrO2 and t-ZrO2 after the usage of bio-template. The biomorphic C0.20 Z-B catalyst displayed an ideal crystalline structure, which is facilitating for the transition of active lattice oxygen species from the bulk to surface, thus reduced the reaction temperature. • Stronger CuO–ZrO2 synergistic interaction, better redox properties and oxygen mobility of the biomorphic CuO–ZrO2 catalyst were evidenced, especially in the C0.20 Z-B sample with a threshold CuO molar content of 20 mol%.

Acknowledgments This work was financed under National Basic Research Program of China (973 Program) (2011CB201202). The donation of TEM techniques by Analytical & Testing Center of Sichuan University is gratefully acknowledged. References [1] D. Ciuparu, M.R. Lyubovsky, E. Altman, L.D. Pfefferle, A. Datye, Catal. Rev. 44 (2002) 593–649. [2] G. Águila, F. Gracia, J. Cortes, P. Araya, Appl. Catal. B 77 (2008) 325–338. [3] H. Falcon, J.A. Barbero, J.A. Alonso, M.J. Martinez-Lope, J.L.G. Fierro, Chem. Mater. 14 (2002) 2325–2333. [4] L.J. Liu, Z.J. Yao, B. Liu, L. Dong, J. Catal. 275 (2010) 45–60. [5] H. Zhang, W. Chu, H.Y. Xu, J. Zhou, Fuel 89 (2010) 3127–3131. [6] A.C. Furtado, C.G. Alonso, M.P. Cantão, N.R.C. Fernandes-Machado, Int. J. Hydrogen Energy 36 (2011) 9653–9662. [7] C. Perkins, A.W. Weimer, Int. J. Hydrogen Energy 29 (2004) 1587–1599. [8] N. Gokon, S. Takahashi, H. Yamamoto, T. Kodama, Int. J. Hydrogen Energy 33 (2008) 2189–2199. [9] K. Tanabe, Mater. Chem. Phys. 13 (1985) 347–364. [10] N.B. Jackson, J.G. Ekerdt, J. Catal. 126 (1990) 31–45. [11] N.B. Jackson, J.G. Ekerdt, J. Catal. 126 (1990) 46–56. [12] F.F. Qu, W. Chu, L.M. Shi, M.H. Chen, J.Y. Hu, Chin. Chem. Lett. 18 (2007) 993–996. [13] P. Ratnasamy, D. Srinivas, C.V.V. Satyanarayana, P. Manikandan, R.S.S. Kumaran, M. Sachin, V.N. Shetti, J. Catal. 221 (2004) 455–465. [14] X.M. Guo, D.S. Mao, S. Wang, G.S. Wu, G.Z. Lu, Catal. Commun. 10 (2009) 1661–1664. [15] X.F. Dong, H.B. Zou, W.M. Lin, Int. J. Hydrogen Energy 31 (2006) 2337–2344. [16] Q. Yu, X.X. Wu, X.J. Yao, B. Liu, F. Gao, J.M. Wang, L. Dong, Catal. Commun. 12 (2011) 1311–1317. [17] X.M. Guo, D.S. Mao, G.Z. Lu, S. Wang, G.S. Wu, J. Catal. 271 (2010) 178–185. [18] M.F. Luo, J.M. Ma, J.Q. Lu, Y.P. Song, Y.J. Wang, J. Catal. 246 (2007) 52–59.

129

[19] M.F. Luo, Y.P. Song, J.Q. Lu, X.Y. Wang, Z.Y. Pu, J. Phys. Chem. C 111 (2007) 12686–12692. [20] J.L. Cao, Y. Wang, X.L. Yu, S.R. Wang, S.H. Wu, Z.Y. Yuan, Appl. Catal. B 79 (2008) 26–34. [21] J.H. Liu, A.Q. Wang, Y.S. Chi, H.P. Lin, C.Y. Mou, J. Phys. Chem. B 109 (2005) 40–43. [22] L.M. Shi, W. Chu, F.F. Qu, S.H. Luo, Catal. Lett. 113 (2007) 59–64. [23] N. Wang, W. Chu, T. Zhang, X.S. Zhao, Chem. Eng. J. 170 (2011) 457–463. [24] S.W. Boettcher, J. Fan, C.K. Tsung, Q.H. Shi, G.D. Stucky, Acc. Chem. Res. 40 (2007) 784–792. [25] B.G. Trewyn, I.I. Slowing, S. Giri, H.T. Chen, V.S.Y. Lin, Acc. Chem. Res. 40 (2007) 846–853. [26] F.Z. Huang, M.F. Zhou, Y.B. Cheng, R.A. Caruso, Chem. Mater. 18 (2006) 5835–5839. [27] T. Scheibel, Proc. Natl. Acad. Sci. U.S.A. 100 (2003) 4527–4532. [28] B. Bai, P. Wang, L. Wu, L. Yang, Z. Chen, Mater. Chem. Phys. 114 (2009) 26–29. [29] R.S. Yuan, X.Z. Fu, X.C. Wang, P. Liu, L. Wu, Y.M. Xu, X.X. Wang, Z.Y. Wang, Chem. Mater. 18 (2006) 4700–4705. [30] H.Y. Xu, W. Chu, J.J. Luo, M. Liu, Catal. Commun. 11 (2010) 812–815. [31] J.J. Luo, W. Chu, H.Y. Xu, C.F. Jiang, T. Zhang, J. Nat. Gas Chem. 19 (2010) 355–361. [32] H.Y. Xu, W. Chu, J.J. Luo, T. Zhang, Chem. Eng. J. 170 (2011) 419–423. [33] J.L. Cao, Y. Wang, T.Y. Zhang, S.H. Wu, Z.Y. Yuan, Appl. Catal. B 78 (2008) 120–128. [34] Z. Liu, M.D. Amiridis, Y. Chen, J. Phys. Chem. B 109 (2005) 1251–1255. [35] L.J. Xie, W. Chu, J.H. Sun, P. Wu, D.G. Tong, J. Mater. Sci. 46 (2011) 2179–2184. [36] L.J. Xie, W. Chu, Y.Y. Huang, D.G. Tong, Mater. Lett. 65 (2011) 153–156. [37] M. Chen, L.P. Fan, L.Y. Qi, X.Y. Luo, R.X. Zhou, X.M. Zheng, Catal. Commun. 10 (2009) 838–841. [38] B.Q. Xu, J.M. Wei, Y.T. Yu, Y. Li, J.L. Li, Q.M. Zhu, J. Phys. Chem. B 107 (2003) 5203–5207. [39] S. Janbroers, P.A. Crozier, H.W. Zandbergen, P.J. Kooyman, Appl. Catal. B 102 (2011) 521–527. [40] H.Y. Pan, Z. Li, Q.B. Xia, H.X. Xi, C. He, Catal. Commun. 10 (2009) 1166–1169. [41] C.Z. Yao, L.C. Wang, Y.M. Liu, G.S. Wu, Y. Cao, W.L. Dai, H.Y. He, K.N. Fan, Appl. Catal. A 297 (2006) 151–158. ˜ [42] P. Hurtado, S. Ordónez, H. Sastre, F.D. Díez, Appl. Catal. B 51 (2004) 229–238. [43] Y. Liu, Q. Fu, M.F. Stephanopoulos, Catal. Today 93–95 (2004) 241–246. [44] Z.Q. Zou, M. Meng, Y.Q. Zha, J. Phys. Chem. C 114 (2010) 468–477. [45] M.D. Rhodes, A.T. Bell, J. Catal. 233 (2005) 198–209. [46] V.R. Choudhary, B.S. Uphade, S.G. Pataskar, Appl. Catal. A 227 (2002) 29–41. [47] G. Avgouropoulos, T. Ioannides, Appl. Catal. A 244 (2003) 155–167. [48] M.F. Luo, Y.J. Zhong, X.X. Yuan, X.M. Zheng, Appl. Catal. A 162 (1997) 121–131. [49] P.H. Matter, D.J. Braden, U.S. Ozkan, J. Catal. 223 (2004) 340–351. [50] K. Otsuka, Y. Wang, M. Nakamura, Appl. Catal. A 183 (1999) 317–324. [51] Q. Liang, X.D. Wu, D. Weng, H.B. Xu, Catal. Today 139 (2008) 113–118. [52] S. Kaliaguine, A.V. Neste, V. Szabo, J.E. Gallot, M. Bassir, R. Muzychuk, Appl. Catal. A 209 (2001) 345–358. [53] L.H. Zhang, F. Li, D.G. Evans, X. Duan, Ind. Eng. Chem. Res. 49 (2010) 5959–5968. [54] S.J. Huang, A.B. Walters, M.A. Vannice, J. Catal. 192 (2000) 29–47. [55] R.X. Zhou, T.M. Yu, X.Y. Jiang, F. Chen, X.M. Zheng, Appl. Surf. Sci. 148 (1999) 263–270. [56] R.F. Hicks, H. Qi, M.L. Young, R.G. Lee, J. Catal. 122 (1990) 280–294. [57] G. Saracco, G. Scibilia, A. Iannibello, G. Baldi, Appl. Catal. B 8 (1996) 229–244. [58] L.A. Isupova, G.M. Alikina, O.I. Snegurenko, V.A. Sadykov, S.V. Tsybulya, Appl. Catal. B 21 (1999) 171–181. [59] J.H. Lee, D.L. Trimm, Fuel Process. Technol. 42 (1995) 339–359.