Hard-template synthesis of three-dimensional mesoporous Cu–Ce based catalysts with tunable architectures and their application in the CO catalytic oxidation

Abstract

In this paper, three-dimensional (3D) Cu–Ce-Ox catalysts with controllable pore diameters were rationally designed and synthesized by a facile hard-template method. By controlling the calcination temperature, the 3D Cu–Ce-Ox catalysts with different pores were synthesized by nano-replication technology with KIT-6 as hard template. These catalysts expressed 3D pore structure, possessed higher pore volume and large pores, and owned more surface active sites, which were helpful for CO oxidation. When the calcination temperature of Cu–Ce-Ox was 600 °C, this catalyst showed the highest catalytic activity. The reason was due to its relatively stronger redox ability and surface activity species. More Ce³⁺, Cu⁺ and chemisorbed oxygen concentration was very useful in achieving the highest catalytic performance.

---

1. Introduction

CO is a flammable, explosive, harmful air pollutant. With the social development and the improvement of people's living standards, carbon monoxide emissions increased annually. Thus studying the elimination of carbon monoxide is very important to us. So far the most common way to eliminate carbon monoxide is oxidation method. Oxidation technique is widely used in air purifiers, gas sensors, closed-cycle laser, gas mask. Studying catalysts which could oxide carbon monoxide has been one of the hot in the catalytic field.

Since the 1930s, researchers first found that a transition metal oxide catalyst having a catalytic effect on the elimination of CO, people had a strong interest in elimination of CO oxidation reaction. In 1989, Haruta et al.¹ reported highly dispersed nanoparticles supported on a suitable carrier for low temperature CO oxidation reaction showed high activity, the study of the elimination of the catalyst was entering a new stage. Precious metal catalysts^2–6 were shown high activity for carbon monoxide oxidation. Shapovalov et al.⁷ also found that gold doped ceria significantly enhanced catalytic activity, exhibiting better catalytic oxidation, and has good stability and long life. However, due to the cost of expensive precious metal-prone sulfur poisoning and other defects, people have been trying to find a substitute.

Therefore, transition-metal oxides and mixed transition-metal oxides catalysts have become a research hotspot for CO catalytic oxidation at present, such as, TiO₂,⁸ CeO₂ (ref. 9) with different morphologies, copper-containing mesoporous manganese oxides,¹⁰ mesoporous Co₃O₄–CeO₂.¹¹ Because CeO₂ has unique redox property and high oxygen storage capacity,⁹ the various properties of CeO₂ promote different chemical reactions and catalytic activity, CeO₂ and ceria-based catalysts have attracted widely attention in CO catalytic oxidation. In our previous work,¹² it was investigated that the adsorbed oxygen concentration on the surface of meso-CeO₂ was relatively higher than micro-CeO₂ and nano-CeO₂. There was a large surface area and small crystallite size of meso-CeO₂, a higher active oxygen concentration on the surface of catalyst was advantage to improve catalytic activity of catalyst. In addition, in contrasted with non-doped CeO₂ catalyst, metal-doped CeO₂ catalyst seems to have better catalytic performance. The research of metal-doped CeO₂,¹³ showed that Cu-doped CeO₂ catalysts have a high CO catalytic activity and have higher oxygen capacity and thermal stability than Co-doped CeO₂ catalysts and Fe-doped CeO₂ catalysts. These research results showed that the surface area, surface adsorbed oxygen concentration, reducibility and pore structure are the key factors that influence the catalytic performance of a metal oxide catalyst.^14,15

It has been well established that the pore structure could exert a significant influence on the surface textures of transition-metal oxide catalysts, and in turn determined the redox properties and activity of the final catalyst. So far, little attention has been paid to investigate the effect of different pore structure and catalytic activity of Cu-doped CeO₂ catalysts with different pore structure for CO oxidation.

In this work, the Cu-doped CeO₂ catalysts with different calcination temperatures (400, 500, 600 and 700 °C) were successfully prepared using a nano-replication method from three-dimensional mesoporous silica KIT-6 as the hard template. We investigated the effect of different pore structure on the surface textures of Cu-doped CeO₂ catalysts and catalytic activity of Cu-doped CeO₂ catalysts with different pore structures for CO oxidation in detail. Their structures were characterized by powder XRD, BET, TEM, XPS, Raman and FTIR techniques.

2. Experimental

2.1 Synthesis of ordered mesoporous silica KIT-6

The ordered mesoporous silica KIT-6 with three-dimensional mesostructure was synthesized according to the previously reported method.¹⁶ Typically, 6.0 g P123 (EO20PO70EO20, MW = 5800, Aldrich), 220 mL deionized water, and 12.0 g hydrochloric acid (35 wt%) were added into a 500 mL beaker and stirred at 35 °C. After complete dissolution of P123, 6.0 g BuOH was added at once and stirred for 1 h. 12.86 g TEOS was added into the transparent solution and stirred at 35 °C for 24 h. Then, the solution was transferred to a Teflon-lined stainless steel autoclave at 100 °C for 24 h. White precipitates were filtered by a vacuum filter, washed with hydrochloric acid (35 wt%, 20–30 mL)–ethanol (industrial alcohol 300–400 mL) solution and 1 L deionized water twice, and then dried at 100 °C overnight. Finally, the sample with a heating rate of 1 °C min⁻¹ from room temperature to 550 °C and kept at that temperature for 5 h. The white mesoporous KIT-6 powder was obtained.

2.2 Synthesis of CeCu20 catalysts

A “bi-solvent” method was used in a typical synthesis of a highly ordered meso-CeCu20.¹⁷ The Cu–CeO₂ catalysts with Cu/(Cu + Ce) atomic ratios were 20%, denoted as meso-CeCu20. In a typical synthesis of meso-CeCu20, 2.6 g of Ce (NO₃)₃·6H₂O and 0.29 g of Cu (NO₃)₂·3H₂O were dissolved in 15 mL ethanol with stirring at room temperature. After complete dissolution, 1 g mesoporous silica template (KIT-6) was added. The mixture was stirred at room temperature until a nearly dry powder had been obtained. The sample was then heated slowly to 400 °C, 500 °C, 600 °C and 700 °C calcined at the same temperature for 5 h with a ramp 1 °C min⁻¹ for forming metal oxide. The resulting sample was twice treated with a hot 2 M NaOH solution (200 mL) to remove the silica template, followed by washing with deionized water (about 1 L) several times, and then drying at 60 °C. CeCu20 prepared by calcination at 400 °C, 500 °C, 600 °C, 700 °C was denoted as CeCu20-400, CeCu20-500, CeCu20-600 and CeCu20-700, respectively.

2.3 Catalyst characterization

XRD analysis was performed to verify the crystallographic phase present KIT-6, CeCu20. XRD patterns of the samples were recorded on a Rigaku D/MAX-RB X-ray diffractometer with a target of Cu Kα operated at 60 kV and 55 mA with a scanning speed of 0.5° min⁻¹. The 2θ of wide-angle ranged from 20° to 80° and the 2θ of low-angle ranged from 0.6° to 5°.

TEM experiments were measured on a JEOL JEM-2010 transmission electron microscope equipped with an Oxford energy-dispersive X-ray (EDX) spectrometer attachment operating at 200 kV.

The specific surface area and the mean pore diameter of the catalysts were determined by nitrogen adsorption in accordance with the BET method, with a Micromeritics ASAP 2010 instrument. The BET surface area determinations were based on six measurements at relative pressures of N₂ in the range of 0.05–1.00.

A Fourier transform infrared spectroscope (FTIR, Nexus 870FT-IR) was used for recording the FTIR spectra of the sample ranging from 400 to 4000 cm⁻¹.

H₂-TPR measurements were performed on online GC-7890II gas chromatograph equipped with a thermal conductivity detector (TCD). The reducing gas was 5 vol% H₂ balanced by argon, and a flow rate of 40 mL min⁻¹ was used. The quartz tube reactor was loaded with 20 mg sample in powder form. The test was carried out from room temperature to 800 °C at a heating rate of 10 °C min⁻¹. Before each measurement, the sample was purged with dry air at 400 °C for 1 h.

Chemical states of the atoms in the catalyst surface were investigated by X-ray photoelectron spectroscopy (XPS) on a VG ESCALAB 210 Electron Spectrometer with Mg Kα (1253.6 eV) radiation. XPS data were calibrated using the binding energy of C 1s (284.6 eV) as the standard.

2.4 Measurements of catalytic performance

Catalytic activity tests were performed in a continuous flow fixed-bed microreactor. A glass tube with an inner diameter of 6 mm was chosen as the reactor tube. About 300 mg catalyst with the average diameter of 20–40 mesh was placed into the tube. The reaction gas mixture consisting of 1 vol% CO balanced with air was passed through the catalyst bed at a total flow rate of 50 mL min⁻¹. A typical weigh hourly space velocity (WHSV) was 10 000 mL g⁻¹ h⁻¹. The composition of the influent and effluent gas was detected with an online GC-7890II gas chromatograph equipped with a TCD. In this paper, this change of moles was neglected. Therefore, the CO conversion was calculated based on the outlet CO:

3. Results and discussion

3.1 Results of KIT-6 templates

3.1.1 XRD results. The low-angle XRD for KIT-6 sample is shown in Fig. 1. The appearance of intense diffraction signal at 2θ is near 1° for KIT-6 sample. It was indicated that the structure is an ordered mesoporous structure. The unit cell parameter (a₀) was estimated via KIT-6 (211) line broadening for KIT-6 sample and the result is listed in Table 1.

Fig. 1 Low-angle XRD patterns of KIT-6.

Table 1 Pore structure parameters and particle size of the KIT-6 silica templates

Sample	BET surface area (m^2 g^−1)	Pore volume (cm^3 g^−1)	a0^a (nm)	dBJH^b (nm)	tw^c (nm)
a XRD unit-cell parameter a0 is equal to 6^1/2d211.b Average pore size calculated from the desorption branch of the isotherm using the BJH method.c The wall thickness tw = a0/2 − dBJH.
KIT-6	678.59	0.86	20.88	5.54	4.90

---

3.1.2 BET results. The N₂ adsorption–desorption isotherm and BJH pore size distribution of KIT-6 sample is shown in Fig. 2. The sample exhibit typical IV shape isotherm, and the result showed that the sample is mesoporous silica with cylindrical pore geometry.¹⁵ The result is consistent with the low-angle XRD image. The textural properties of sample obtained from the nitrogen sorption isotherm are summarized in Table 1. The wall thickness of the sample was calculated using the pore size obtained from N₂ desorption data and the unit-cell parameter a₀ from X-ray diffraction, and the result was 4.9 nm in Table 1.

Fig. 2 BJH pore size distributions (A) and N₂ adsorption–desorption isotherms (B) of KIT-6.

3.2 Catalyst results

3.2.1 XRD results. Fig. 3 shows the XRD patterns of CeCu20-400, CeCu20-500, CeCu20-600 and CeCu20-700. The wide-angle XRD patterns of CeO₂ are displayed in Fig. 3A. By comparing with the XRD pattern of the standard CeO₂ sample (JCPDS PDF#43-1002), the main reflections at 28.5°, 33.1°, 47.5°, 56.3°, 59.1°, 69.4°, 76.7° and 79.1° (2θ), correspond to the (111), (200), (220), (311), (222), (400), (331) and (420) planes, it judged that CeCu20-400, CeCu20-500, CeCu20-600 and CeCu20-700 have a typical crystalline fluorite structure. The absence of CuO diffraction peaks for CeCu20-400, CeCu20-500, CeCu20-600 and CeCu20-700 catalysts. These results are ascribed to the substitution of copper in the ceria lattice or the formation of extremely small copper oxide clusters, indicating homogeneous dispersion of copper species on the ceria matrices. The finding also supports the results of Raman (Fig. 3C).

Fig. 3 Wide-angle (A), low-angle (B) XRD diffraction and Raman (C) patterns of CeCu20-400 (a), CeCu20-500 (b), CeCu20-600 (c) and CeCu20-700 (d). (B) Low-angle XRD diffraction of fresh catalyst, the inset of (B) is low-angle XRD diffraction of used catalyst (CeCu20-600) after stability test.

Raman spectroscopic technique was used for investigation of the structures of the CeCu20 catalysts. The spectra obtained are compiled in Fig. 3C. Three characteristic peaks could be observed in all the catalysts. The strong band at 465 cm⁻¹ is attributed to the F_2g Raman vibration mode of the cubic fluorite-structure phase and the broad band at 593 cm⁻¹ is attributed to the oxygen vacancies generated by the partial substitution of some copper ions to Ce⁴⁺.¹⁸ In addition, the weak band at 246 cm⁻¹ belongs to the rearrangement of oxygen atoms from their ideal fluorite lattice locations.¹⁹ Usually, the area ratio of peaks 593 and 465 cm⁻¹ can represent the relative concentration of oxygen vacancies.²⁰ According to the passage,²¹ the high concentration of oxygen vacancies is bound to result in a considerably enhanced activity of the catalysts. From Fig. 3C, it also can be seen that with increasing the calcination temperature, the intensity of the band at 465 cm⁻¹ decreases apparently, while the intensity of the band at 593 cm⁻¹ basically unchanged. The area ratio of peaks 593 and 465 cm⁻¹ follows the sequence of CeCu20-600 > CeCu20-500 > CeCu20-700 > CeCu20-400, which indicates that the amount of oxygen vacancies in CeCu20 catalysts is strongly affected by the calcination temperature. The shift of the Raman peak near 465 cm⁻¹ in CeO₂ nanoparticles progressively to higher energy indicates the increase of their particle size and the crystallinity degree,²² which is consistent with the results of XRD. Respectively, the average sizes of crystalline CeO₂ are calculated from the ceria (111) plane using Scherrer's formula, and the result is shown in Table 2. All catalysts exhibit small crystallite size of CeO₂, among which sample CeCu20-600 gave the smallest particle size of only 6.06 nm. The small crystallite size would increase in the amount of reactive edge sites and thus enhance the catalytic activity.

Table 2 Pore structure parameters and particle size of the prepared CeCu20 sample

Sample	BET surface area (m^2 g^−1)	Pore diameter^a (nm)	Pore volume (cm^3 g^−1)	DCeO2^b (nm)	Cu/(Cu + Ce) molar^c (%)
a The pore size distribution is calculated from the desorption isotherm by the BJH method.b Calculated the average size of crystalline CeO2 from line broadening of the ceria (111) plane using Scherrer's formula [D = kλ/(β1 cos θ), k = 0.943, λ = 0.15405 nm, β1 = [(β/180) × 3.14], β is peak FWHM, θ = 14.25°].c Calculated based on ICP results.
CeCu20-400	113.18	5.16	0.13	6.99	23.03
CeCu20-500	132.97	6.32	0.17	6.87	24.23
CeCu20-600	133.57	7.13	0.24	6.06	24.45
CeCu20-700	116.30	5.72	0.15	6.92	23.22

---

The low-angle XRD patterns of the bimetal oxide replicas are shown in Fig. 3B. All catalysts show the same cubic Ia3d symmetry as their mesoporous silica KIT-6 template. The appearance of diffraction signal at 2θ is close to 1° for all catalysts. The low-angle XRD patterns of the used catalyst (CeCu20-600) after 48 hours stability test is shown in the inset of Fig. 3B, strong diffraction peaks well-preserved and there is no significant difference between the Fig. 3B-(c) and the inset of Fig. 3B. It indicates that the catalyst has a good stability, and it could well keep the 3D mesoporous structure after catalytic reaction. Compared with CeCu20-400 catalyst, CeCu20-500, CeCu20-600 and CeCu20-700 catalysts have relatively strong diffraction peaks. It indicates that the structure of CeCu20-500, CeCu20-600 and CeCu20-700 catalysts has relatively higher ordered than the CeCu20-400 catalyst. Four catalysts by using same preparation process and the same template show that the higher the calcination temperature is, the higher the order is (Fig. 3B).

3.2.2 BET results. Fig. 4 shows the N₂ adsorption–desorption isotherms and BJH pore size distributions of mesoporous CeCu20 catalysts. All four isotherms (Fig. 4A) are of typical IV classification with a clear H₁ hysteresis loop, which are characteristics of mesoporous materials. These results are consistent with the low-angle XRD image. Fig. 4B shows a little wide pore-size distribution, the peak centered at 2.94 nm, 2.93 nm, 2.99 nm and 2.42 nm for CeCu20-400, CeCu20-500, CeCu20-600 and CeCu20-700 catalysts, respectively, in good agreement with the wall thickness of KIT-6 (see Table 1). From Table 2, the CeCu20-600 catalyst has largest BET specific pore diameter and pore volume. As we all know, the high pore diameter and pore volume would increase in the amount of reactive edge sites and thus enhance the catalytic activity. The different results show that the catalyst pore structure can be controlled by controlling the calcination temperature. The Cu/(Cu + Ce) molar percentages of all catalysts derived from ICP are listed in Table 2. It can be seen that the results are in good agreement with the ratios in initial mixture.

Fig. 4 N₂ adsorption–desorption isotherms (A) and BJH pore size distributions (B) of CeCu20-400 (a), CeCu20-500 (b), CeCu20-600 (c) and CeCu20-700 (d).

3.3 TEM analysis

The four mesoporous CeCu20 catalysts are examined by TEM (Fig. 5a–h). It can be seen from the low magnification TEM images (Fig. 5a, c, e and g), all four catalysts show 3D mesoporous structure. All four catalysts show mesostructure, suggesting that four samples are well reverse-replica of the KIT-6 template and the mesostructure regularity is well preserved after the nanocasting replication. The result is consistent with BET result (Fig. 4A). By examining many particles, no obvious nonporous particles were observed. The result indicated that almost all the metal nitrate were successfully filled inside the mesopores and then transformed in situ to the fluorite during the calcination. The result is consistent with the low-angle XRD result of catalysts (Fig. 3B). The morphology of the all four catalysts has been also analyzed by HRTEM, which is showed in Fig. 5b, d, f and h. It can be seen that the pore walls are well constituted by metal oxide nanocrystals with a diameter of several nanometers. The lattice spacing d value of the (111) and (200) plane were estimated to be (0.3215, 0.2596 nm), (0.3118, 0.2691 nm), (0.3099, 0.2743 nm), (0.3225, 0.2712 nm) which are close to that (0.3124 nm, 0.2706 nm) of fluorite CeO₂ (JCPDS PDF#43-1002). The selected area electron diffraction (SAED) patterns are given in the inset of (Fig. 5b, d, f and h). There were multiple bright electron diffraction rings, indicating the generation of polycrystalline.

Fig. 5 TEM and HRTEM images of CeCu20-400 (a and b), CeCu20-500 (c and d), CeCu20-600 (e and f), CeCu20-700 (g and h). The inset is the SAED pattern.

3.4 FTIR analysis

The FTIR spectra of KIT-6 and CeCu20-600 are shown in Fig. 6. As for KIT-6, the absorption bands around 1082, 808 and 461 cm⁻¹ may be assigned to Si–O–Si asymmetrical stretching vibration, symmetrical stretching vibration and bending vibration, respectively. The absorption band at around 985 cm⁻¹ is contribution from Si–OH. The peaks of CeCu20-600 centered at around 513 cm⁻¹ can be assigned to vibration of Cu–O band,²³ while the bands centered at 873, 972, 1332 and 1519 cm⁻¹ are related to CeO₂ structure.²⁴ For all samples, the peaks centered at about 1631 and 3436 cm⁻¹ are assigned to the hydroxyl groups of the superficial adsorbed or crystallized water. Compared with KIT-6 and CeCu20-600, there are not absorption peaks belonging to the KIT-6 for CeCu20-600. The result shows that the mesoporous silica template has been completely removed.

Fig. 6 FTIR spectra of KIT-6 (a) and CeCu20-600 (b).

3.5 H₂-TPR results

H₂-TPR is a useful tool to obtain the reducibility of the catalysts, as displayed in Fig. 7. The reduction peak centered at about 400 and 500 °C are attributed to the reduction of surface oxygen for all CeCu20 catalysts.²⁵ A well-defined two-step reduction profile is observed for all CeCu20 catalysts in the range 100–350 °C. It indicated that there are at least two copper species of catalysts. According to the literature,²⁶ the pure CuO reduction temperature is at about 380 °C. Two reduction temperatures are much lower than that of pure CuO for CeCu20 catalysts. It suggested that the reducibility is promoted by the interaction of CuO and CeO₂. It has been reported that CeO₂ promotes the reduction of finely dispersed CuO species, and the smaller the CuO particles, the easier they are to reduce. The reduction peak around at 210 °C is attributed to the reduction of well dispersed Cu species and Cu ions strongly interacting with CeO₂.²⁵ The reduction peak around at 280 °C is attributed to the contribution of CuO crystalline.²⁷ For the there catalysts, the reduction peak temperatures of CuO are as follows: CeCu20-400 (258 and 309 °C), CeCu20-500 (219 and 273 °C), CeCu20-600 (204 and 239 °C). Compared with previously reported catalysts, peaks of CeCu20-400, CeCu20-500, CeCu20-600 catalysts shift to the lower temperature side (CeCu20-400 < CeCu20-500 < CeCu20-600), indicating that they have relatively stronger redox ability, especially CeCu20-600 catalyst. According to the literatures,²⁸ ceria is reduced by H₂ only at temperature higher than 350 °C and two reduction peaks correspond to the reduction of superficial cerium (at about 540 °C) and bulk CeO₂ (at about 840 °C). CeO₂ showed two peaks at 658 and 794 K attributable to the reduction of surface and bulk ceria compositions, respectively. In this paper, the weak peak at about 400 and 520 °C correspond to the reduction of superficial CeO₂. There catalysts are easy to be reduced due to the better dispersion of copper species. This result is also supported by the XRD and Raman results (Fig. 3A and C).

Fig. 7 H₂-TPR profiles of CeCu20-400 (a), CeCu20-500 (b) and CeCu20-600 (c).

3.6 Catalytic activity

Catalysts with different pore structure will have a great effect on the catalytic activity for CO catalytic oxidation. Four CeCu20 catalysts with different pore structure were investigated in this work. In quantitative terms, low temperature oxidation is usually determined by three parameters: T₁₀, T₅₀ and T₉₀, which denote the temperatures at which a 10, 50 or 90% conversion of the initial reactants is attained. The CO catalytic oxidation activity patterns of the CeCu20-400, CeCu20-500, CeCu20-600 and CeCu20-700 catalysts are shown in Fig. 8. With an increase of reaction temperature, the catalytic activity in the CO oxidation becomes better. The Fig. 9 profiles indicate that the temperature needed for a complete CO conversion over CeCu20-400, CeCu20-500, CeCu20-600 and CeCu20-700 were 75, 57, 55 and 63 °C, respectively. The complete CO conversion values of CeCu20-500 and CeCu20-600 catalysts have not considerable transformation. But the T₅₀ values of four catalysts have a big difference, and these values are 53 °C (CeCu20-400), 45 °C (CeCu20-500), 30 °C (CeCu20-600) and 50 °C (CeCu20-700), respectively. Compared with CeCu20-400 and CeCu20-700, CeCu20-500 and CeCu20-600 catalysts show relatively higher activity. It can be noticed that the catalytic stability is also evaluated over the best catalyst CeCu20-600 (Fig. 10A). The CeCu20-600 catalyst reused three times at 55 °C and the stability diagram was shown in Fig. 10A. As we can see from the Fig. 10A, the CeCu20-600 catalyst can be reused. However, it can be noticed that the catalytic stability can keep 42 h on stream at 55 °C in the first use. Then the stability just can keep 26 h for the CeCu20-600 catalyst on stream at 55 °C in the second time. The catalytic stability in the third used far less than the previous, it can keep 15 h. When water vapor adds to the reaction gas, carbon monoxide conversion rate drop to around eighty percent at 55 °C, but it can keep 12 h stably (Fig. 10B). It indicated that the catalyst has a good stability, and it could well keep the 3D mesoporous structure after catalytic reaction (The low-angle XRD patterns of the catalyst CeCu20-600 with stability test after 48 hours are shown in the inset of Fig. 3B, strong diffraction peaks well-preserved and there was no significant difference between the low-angle XRD patterns Fig. 3B-(c) and the inset of Fig. 3B).

Fig. 8 The catalytic activity of CeCu20-400 (a), CeCu20-500 (b), CeCu20-600 (c) and CeCu20-700 (d) catalysts for CO oxidation.

Fig. 9 The temperature of the CeCu20-400 (sample 1), CeCu20-500 (sample 2), CeCu20-600 (sample 3) and CeCu20-700 (sample 4) catalysts for CO complete oxidation.

Fig. 10 The stability diagram of the CeCu20-600 catalyst recycled three times for CO oxidation at 55 °C (A), the influence of water vapor on the catalytic performance for CeCu20-600 at 55 °C (B).

3.7 Surface composition analysis

In order to speculate the reaction mechanism of CeCu20 catalysts, the surface composition is also investigated over the catalysts of CeCu20-400, CeCu20-500, CeCu20-600 and CeCu20-700 (Fig. 11). Fig. 11A shows the XPS spectra of Ce 3d for CeCu20-400, CeCu20-500, CeCu20-600 and CeCu20-700 catalysts. The fitting process was referred to literature. They (V₀, V₂, V₃, V₄, V₆, V₈) are characteristic of Ce⁴⁺ 3d final states and they (V₁, V₅, V₇) are characteristic of Ce³⁺ 3d final states.²⁹ The phenomena show the coexistence of Ce³⁺ and Ce⁴⁺ species on the surface of CeCu20-400, CeCu20-500, CeCu20-600 and CeCu20-700 catalysts. Fig. 11B shows the XPS spectra of O 1s for CeCu20-400, CeCu20-500, CeCu20-600 and CeCu20-700 catalysts. Three kinds of peaks (O_latt, O_ads and O_wat) can be identified by the deconvolution of the O 1s spectra. The one (O_latt) with lower binding energy (at 529.6 eV) is assigned to lattice oxygen in the metal oxides, the next (O_ads) (at 530.7 eV) is ascribed to chemisorbed oxygen, and the third (O_wat) peak at the highest binding energy (at 532.1 eV) is due to adsorbed water species on the surface.³⁰ The surface chemisorbed O atoms have been reported to be highly active in oxidation reactions because their mobility is higher than that of the lattice O atoms, and a higher concentration of O_ads is helpful for the CO oxidation. The relative percentages of O_ads species are quantified based on the area of their XPS peaks and are shown in Table 3. With higher calcination temperature, the O_ads/(O_latt + O_ads + O_wat) ratio increased to 33.66% for CeCu20-600, meaning the catalysts with higher calcination temperature possessed more chemisorbed oxygen species than those with lower calcination temperature. The results show that CeCu20-600 catalyst possesses the higher catalytic activity than CeCu20-400, CeCu20-500 and CeCu20-700 catalysts. Fig. 11C shows the XPS spectra of Cu 2p for CeCu20-400, CeCu20-500, CeCu20-600 and CeCu20-700 catalysts. There are two sets of peaks for CeCu20-400, CeCu20-500, CeCu20-600 and CeCu20-700 catalysts, corresponding to Cu 2p_3/2, respectively. According to literature,³¹ the peak centered at 934.3 eV is characteristic of Cu²⁺ 2p final state and the peak centered at 932.6 eV is reduced copper species (Cu⁺). The presence of Cu⁺ is most possible in catalyst prepared by calcination in air. The Cu⁺ species is considered to be the active component in the CeCu20 catalysts. The relative percentages of Cu⁺ species are quantified based on the area of their XPS peaks and shown in Table 4. The results show that CeCu20-600 catalyst possesses much more surface active Cu⁺ species than CeCu20-400, CeCu20-500 and CeCu20-700 catalyst, therefore CeCu20-600 catalyst shows the higher catalytic activity.

Fig. 11 Ce 3d (A), O 1s (B) and Cu 2p (C) XPS spectra of CeCu20-400 (a), CeCu20-500 (b), CeCu20-600 (c) and CeCu20-700 (d).

Table 3 Relative contents of O 1s in various chemical states

Samples	Chemical states	Binding energy (eV)	Chemisorbed oxygen percent (%)
CeCu20-400	Olatt	529.63	21.00
	Oads	530.7
	Owat	532.1
CeCu20-500	Olatt	529.53	27.83
	Oads	530.7
	Owat	532.0
CeCu20-600	Olatt	529.63	33.66
	Oads	530.7
	Owat	532.2
CeCu20-700	Olatt	529.53	26.45
	Oads	530.7
	Owat	532.1

---

Table 4 BE values (eV) and surface Cu⁺/(Cu⁺ + Cu²⁺) and Ce³⁺/(Ce³⁺ + Ce⁴⁺) atomic ratios

Samples	Cu 2p3/2	Binding energy (eV)	Cu^+/(Cu^+ + Cu^2+) (%)	Ce^3+/(Ce^3+ + Ce^4+) (%)
CeCu20-400	Cu^+	932.7	30.95	15.92
	Cu^2+	934.3
CeCu20-500	Cu^+	932.7	37.38	16.98
	Cu^2+	934.2
CeCu20-600	Cu^+	932.7	42.15	17.76
	Cu^2+	934.2
CeCu20-700	Cu^+	932.7	33.27	16.06
	Cu^2+	934.2

---

3.8 Reaction mechanism speculate

Through the XPS results, there are Ce³⁺ and Cu⁺ on the surface of catalyst. In order to maintain electrostatic equilibrium, it will produce oxygen vacancies. Oxygen vacancy is very important to catalytic activity. The activity center can promote the activation of O₂. Therefore, the CO oxidation reaction can become more easily. There are amount of adsorbed oxygen species on the surface of catalyst. Usually, adsorbed oxygen species play an important role in the oxidation of CO. For reaction mechanism of CO oxidation, maybe adhered to the Langmuir–Hinshelwood + redox.^32,33 The XPS confirms that the percentage of chemisorbed oxygen on the surface of the CeCu20-600 catalyst have an obvious enhancement compared with CeCu20-400, CeCu20-500 and CeCu20-700. Therefore, the CeCu20-600 catalyst exhibited good catalytic activity compared with CeCu20-400, CeCu20-500 and CeCu20-700. Based on the above analysis and discussion, it can be concluded that the good performance of the CeCu20-600 catalyst was associated with the larger pore size, higher adsorbed oxygen and oxygen vacancy concentration, and 3D mesoporous structure. And the most important factor is higher chemisorbed oxygen and oxygen vacancy concentration governing the catalytic activity.

Based on the conventional oxidation mechanism, a carbon monoxide oxidation mechanism over CeCuO_x is proposed (Fig. 12). As shown in Fig. 12, the paths of CO oxidation reaction in surface, CO adsorbed on the copper, react with nearby chemisorbed oxygen species (the path 1), and also can be react with the oxygen species (O₂⁻ and O⁻) formed by transformation lattice oxygen (O²⁻)³⁴ (the path 2). In this mechanism, Cu acts as the sole catalytic center where the reaction takes place, and the oxygen transfer is completed through the support via a redox cycle.³⁵ Chemical reaction equation is shown by the following:

O₂ → O₂⁻ + O⁻ (1)

O₂⁻/O⁻ + CO → CO₂ (2)

O₂⁻ + O⁻ + Ce³⁺ → [O²⁻] + Ce⁴⁺ (3)

[O²⁻] + Cu²⁺ → Cu⁺ + O₂⁻ + O⁻ (4)

O₂⁻/O⁻ + CO → CO₂ (5)

Fig. 12 The patterns of CO reaction mechanism over the interface of CeCuO_x.

The catalytic cycle involves: (1) the adsorbed oxygen on the support surface dissociates into chemisorbed oxygen species O₂⁻ and O⁻; (2) chemisorbed oxygen species O₂⁻ and O⁻ reaction of adsorbed CO and the generation of CO₂; (3) O₂⁻ and O⁻ diffuses into the support lattice vicinity, and subsequently is activated at the oxygen vacancies and converted into lattice oxygen [O²⁻], accompanied by the transformation of Ce³⁺ to Ce⁴⁺; (4) once the surface oxygen is consumed, oxygen vacancies, which lie close to the copper species, can accept oxygen from bulk diffusion in catalyst lattice or gaseous oxygen to form chemisorbed oxygen O₂⁻ and O⁻, accompanied by the transformation of Cu²⁺ to Cu⁺. (5) The chemisorbed oxygen O₂⁻ and O⁻ formed by lattice oxygen reaction of the CO adsorbed on the copper and the generation of CO₂.

Either the chemisorbed oxygen species (O₂⁻ and O⁻) generated on the copper–CeO₂ interface³⁶ (path 1) or the oxygen species (O₂⁻ and O⁻) formed by transformation lattice oxygen (O²⁻) derived from ceria (path 2) can react with the adsorbing CO to form CO₂. Interface activation plays a determining role with Cu⁺ effectively adsorbing CO, and with CeO₂ on the interface serving as dissociation centers for oxygen molecules. Once the surface oxygen is consumed, oxygen vacancies, which lie close to the copper species, can accept oxygen from bulk diffusion in catalyst lattice or gaseous oxygen. It is important to note that the CO oxidation concerns redox couples: Cu²⁺/Cu⁺ and Ce⁴⁺/Ce³⁺. Consequently, the synergetic interactions among Cu and CeO₂ are enhanced due to the mutual promotion of texture and structure, redox properties, and surface composition, leading to improved catalytic activity.

3.9 Compare of different Cu/CeO₂ catalysts on CO oxidation activity

Many studies have shown that CuO/CeO₂ catalysts have high catalytic activities for CO oxidation (as shown in Table 5). Zhang et al.³⁷ synthesized CuO/CeO₂ catalysts by a co-precipitation method and studied their catalytic properties by using a microreactor-GC system. The results indicated that the CuO loadings and calcination temperature had a large influence on the catalytic activity of the catalysts for CO oxidation. The catalyst with 7 wt% CuO loading calcined at 650 °C for 4 h exhibited the highest catalytic activity, and the T₁₀₀ of CuO/CeO₂ catalyst was 120 °C. Tang et al.³⁸ prepared CuO/CeO₂ catalysts by co-precipitation, deposition–precipitation and impregnation methods. CuO/CeO₂ catalysts were extensively investigated for CO oxidation reaction. The catalysts prepared by co-precipitation exhibited the highest catalytic activity in CO oxidation with CO total conversion at 85 °C. Zheng et al.³⁹ synthesized CeO₂ nano-crystals by a sol–gel process and then used as support for CuO/CeO₂ catalysts prepared via impregnation method. The CuO/CeO₂ (sol–gel) catalysts exhibited higher catalytic activity than the CuO/CeO₂ (commercial) catalysts and the T₁₀₀ of CuO/CeO₂ catalyst was 140 °C. Avgouropoulos et al.⁴⁰ synthesized CuO–CeO₂ catalysts via the urea-nitrate combustion method. The catalyst with Cu/(Cu + Ce) ratio equal to 0.15 exhibited the best catalytic performance, and the T₁₀₀ of CuO/CeO₂ catalyst was 160 °C. The sample remained for a period of 75 h under reaction conditions in the presence of 15% CO₂ (with W/F = 0.144 g s cm⁻³), at 160 °C. During that period, the catalyst exhibited a perfectly constant behavior with more than 99.5% CO conversion. Following this first period, 10% H₂O was added in the feed, resulting to a significant decrease of the activity. The conversion of CO dropped from 99.5 to 58%.

Table 5 Comparing of different Cu/CeO₂ catalysts on CO oxidation activity^a

Catalysts	Method	Metal load proportion (%)	Reaction condition	T100 (°C)	Stability (h)	Ref.
a CP: co-precipitation method; DP: deposition–precipitation method; IMP: impregnation method; UNC: urea–nitrate combustion method; HT: hard-template method.
CuO/CeO2	CP	7 wt%	0.2 g catalyst, feed gas 9.9 vol% CO	120	—	37
CuO/CeO2	CP	10 wt%	0.2 g catalyst, feed gas 1.0 vol% CO	85	—	38
	DP	2.5 wt%		140
	IMP	5 wt%		160
CuO/CeO2	IMP	15.12 wt%	0.2 g catalyst, feed gas 1 vol% CO	140	—	39
CuO/CeO2	UNC	Cu/(Cu + Ce) molar ratio 0.15	50–120 mg catalyst, feed gas 1 vol% CO	160	75	40
CeCu20-600	HT	Cu/(Cu + Ce) molar ratio 0.2	0.3 g catalyst, feed gas 1 vol% CO	55	42	This work

---

Compared with CuO/CeO₂ catalysts prepared via other techniques, our catalysts exhibited similar or even better catalytic performance. The temperature needed for a complete CO conversion over CeCu20-600 were 55 °C lower than their best CuO/CeO₂ catalysts (T₁₀₀ = 85 °C). 10% H₂O was added in the feed,⁴⁰ resulting to a significant decrease of the activity and the conversion of CO dropped from 99.5 to 58%. It was similar to our result, H₂O was added in the feed, resulting to a significant decrease of the activity and the conversion of CO dropped from 100 to 80%, it was worth noting that it also could keep 12 h stably. CuO–CeO₂ catalysts synthesized by Avgouropoulos et al.⁴⁰ remained for a period of 75 h under reaction conditions at 160 °C, but in this work the catalytic stability is evaluated over the best catalyst CeCu20-600 (Fig. 10), and no loss of CO conversion is observed before 42 h on stream at 55 °C. Although the stability is better than in this work, the temperature needed for a complete CO conversion over CeCu20-600 were only 55 °C far lower than CuO–CeO₂ catalysts (T₁₀₀ = 160 °C).

4. Conclusions

In summary, porous polycrystalline of CeCu20 mixed metal template by mesoporous silica KIT-6 with different calcination temperature have been fabricated. Compared with CeCu20-400, CeCu20-500 and CeCu20-700, the CeCu20-600 possesses higher pore volume and large pores. Four catalysts were evaluated for the catalytic CO oxidation reaction. The CeCu20-600 showed a highest catalytic activity for CO oxidation, and the T₁₀₀ value is 55 °C. A better catalytic activity for CO oxidation could be achieved over the catalyst with large pore size and higher adsorbed oxygen and oxygen vacancy concentration. Besides, the best catalyst CeCu20-600 exhibited highly stability, and low-angle XRD patterns show it could well keep the 3D mesoporous structure after long-term catalytic reaction.

Acknowledgements

The financial support of The National Natural Science Foundation of China (21407154, 21507137), The National Basic Research Program of China (2013CB933201), Science and Technology Service Network Initiative (STS) of Chinese Academy of Science (KFJ-SW-STS-149) is gratefully acknowledged.

References

1. M. Haruta, N. Yamada and T. Kobayashi, J. Catal., 1989, 115, 301–309.
2. M. Valden, S. Pak, X. Lai and D. W. Goodman, Catal. Lett., 1998, 56, 7–10.
3. N. M. Gupta and A. K. Tripathi, J. Catal., 1999, 187, 343–347.
4. L. Zhu, F. Li, T. Si and Y. He, J. Mol. Catal., 2016, 30, 46–53.
5. A. Holmgren, F. Azamoush and E. Fridell, Appl. Catal., B, 1999, 22, 49–61.
6. A. Martinez-Arias, J. M. Coronado and R. Cataluna, J. Phys. Chem. B, 1998, 102, 4357–4365.
7. V. Shapovalov and H. Metiu, J. Catal., 2007, 245, 205–214.
8. K. Ho and K. Yeung, J. Catal., 2006, 242, 131–141.
9. D. Zhang, C. Pan, L. Shi, L. Huang, J. Fang and H. Fu, Microporous Mesoporous Mater., 2009, 117, 193–200.
10. Y. Hasegawa, K. Fukumoto, T. Ishima, H. Yamamoto, M. Sano and T. Miyake, Appl. Catal., B, 2009, 89, 420–424.
11. C. Ma, D. Wang, W. Xue, B. Dou, H. Wang and Z. Hao, Environ. Sci. Technol., 2011, 45, 3628–3634.
12. Y. Su, Z. Tang, Y. Song, W. Han and P. Zhang, CrystEngComm, 2014, 16, 5189–5197.
13. Y. Su, Z. Tang, Y. Song, W. Han and G. Lu, Catal. Surv. Asia, 2015, 19, 68–77.
14. Z. Qu, F. Yu, X. Zhang, Y. Wang and J. Gao, Chem. Eng. J., 2013, 229, 522–532.
15. Y. Su, Z. Tang, Y. Song, W. Han and G. Lu, Catal. Surv. Asia, 2015, 19, 129–139.
16. F. Kleitz, S. H. Choi and R. Ryoo, Chem. Commun., 2003, 17, 2136–2137.
17. H. Yen, Y. Seo, R. Guillet-Nicolas, S. Kaliaguine and F. Kleitz, Chem. Commun., 2011, 47, 10473–10475.
18. J. Fan, X. Wu, Q. Liang, R. Ran and D. Weng, Appl. Catal., B, 2008, 81, 38–48.
19. H. Vidal, J. Kašpar, M. Pijolat, G. Colon, S. Bernal and A. Cordón, Appl. Catal., B, 2000, 27, 49–63.
20. Q. Liang, X. Wu, D. Weng and H. Xu, Catal. Today, 2008, 139, 113–118.
21. W. Dow and T. Huang, J. Catal., 1996, 160, 171–182.
22. J. Spanier, R. Robinson, F. Zhang, S. Chan and I. Herman, Phys. Rev. B: Condens. Matter Mater. Phys., 2001, 64, 245407.
23. H. Tuysuz, M. Comotti and F. Schuth, Chem. Commun., 2008, 34, 4022–4024.
24. G. Díaz, R. Pérez-Hernández, A. Gómez-Cortés, M. Benaissa, R. Mariscal and J. L. G. Fierro, J. Catal., 1999, 187, 1–14.
25. Z. Zou, M. Meng and Y. Zha, J. Phys. Chem. C, 2010, 114, 468–477.
26. C. He, Y. Yu, L. Yue, N. Qiao, J. Li, Q. Shen, W. Yu, J. Chen and Z. Hao, Appl. Catal., B, 2014, 147, 156–166.
27. A. Gómez-Cortés, Y. Marquez, J. Arenas-Alatorre and G. Díaz, Catal. Today, 2008, 133–135, 743–749.
28. P. Ratnasamy, D. Srinivas, C. Satyanarayana, P. Manikandan, R. Senthil, M. Sachin and V. Shetti, J. Catal., 2004, 221, 455–465.
29. M. A. Henderson, C. L. Perkins, M. H. Engelhard, S. Thevuthasan and C. H. F. Peden, Surf. Sci., 2003, 526, 1–18.
30. J. Zhu, J. Yang, Z. Bian, J. Ren, Y. Liu, Y. Cao, H. Liu, H. He and K. Fan, Appl. Catal., B, 2007, 76, 82–91.
31. G. Avgouropoulos and T. Ioannides, Appl. Catal., A, 2003, 244, 155–167.
32. E. S. Kurkina and E. D. Tolstunova, Appl. Surf. Sci., 2001, 182, 77–90.
33. E. Bekyarova, P. Fornasiero, J. Kaspar and M. Graziani, Catal. Today, 1998, 45, 179–183.
34. X. Tang, B. Zhang, Y. Li, Y. Xu, Q. Xin and W. Shen, Appl. Catal., A, 2005, 288, 116–125.
35. J. Sun, L. Zhang, C. Ge, C. Tang and L. Dong, Chin. J. Catal., 2014, 35, 1347–1358.
36. J. Dubois, M. Bislaux, H. Mimoun and C. Cameron, Chem. Lett., 1990, 19, 967–970.
37. T. Zhang, S. Wang, Y. Yu, Y. Su, X. Guo, S. Wang, S. Zhang and S. Wu, Catal. Commun., 2008, 9, 1259–1264.
38. X. Tang, B. Zhang, Y. Li, Y. Xu, Q. Xin and W. Shen, Catal. Today, 2004, 93, 191–198.
39. X. Zheng, S. Wu, S. Wang, S. Wang, S. Zhang and W. Huang, Appl. Catal., A, 2005, 283, 217–223.
40. G. Avgouropoulos and T. Ioannides, Appl. Catal., A, 2003, 244, 155–167.

---