Large-scale synthesis and catalytic activity of nanoporous Cu–O system towards CO oxidation

Abstract

Nanoporous Cu–O system catalysts with different oxidation states of Cu have been fabricated through a combination of dealloying as-milled Al_66.7Cu_33.3 alloy powders and subsequent thermal annealing. X-ray diffraction (XRD), field-emission scanning electron microscopy (FESEM), transmission electron microscopy (TEM) and X-ray photoelectron spectroscopy (XPS) have been used to characterize the microstructure and surface chemical states of Cu–O catalysts. The peculiar nanoporous structure can be retained in Cu–O catalysts after thermal treatment. Catalytic experiments reveal that all the Cu–O samples exhibit complete CO conversion below 170 °C. The optimal catalytic performance could be achieved through the combination of annealing in air with hydrogen treatment for the Cu–O catalyst, which shows a near complete conversion temperature (T_90%) of 132 °C and an activation energy of 91.3 KJ mol⁻¹. In addition, the present strategy (ball milling, dealloying and subsequent thermal treatment) could be scaled up to fabricate high-performance Cu–O catalysts towards CO oxidation.

---

1. Introduction

As a highly toxic gas, carbon monoxide (CO) can easily bond to hemoglobin and a concentration of about 650 ppm exposes human beings to the threat of death.¹ A dominant strategy to abate CO emission is the catalytic oxidation of CO to carbon dioxide (CO₂). During the past decades, noble metals including Pt, Pd and Rh have served as important catalysts in the purification of automobile exhaust.² Besides Pt-group metals, Haruta et al.^3,4 also initiated the study of CO catalytic oxidation over metal oxides supported Au nanoparticles in the 1980s. However, the increasing price and limited resources of noble metals have become non-ignorable factors in the wide utilization of these catalysts. In contrast, transition metal oxides have recently captured interest as substitutes for noble metals in CO oxidation due to their excellent catalytic properties.⁵ Among transition metal oxides, the oxides of copper (including Cu₂O and CuO) have aroused particular attention because of their diverse applications in photocatalysis,^6,7 gas sensors,^7,8 Li-ion battery electrode materials,^9,10 etc. There are a few but significant reports pointing out that nanoscale Cu₂O/CuO can exhibit appealing CO catalytic oxidation performance, achieving a near complete conversion at temperatures around or below 200 °C.^11–16 To develop the oxides of copper in the field of CO oxidation, efforts should be made to further enhance catalytic performance, make the catalyst more versatile and reduce the material cost. To this date, various methods have been explored to synthesize the oxides of copper with different morphologies and size distributions. For instance, Zhou et al.¹³ utilized the carbonate (and surfactant)-assisted hydrothermal approach to prepare the nanoporous CuO, with 100% CO conversion achieved at 160 °C. Bao et al.¹⁵ adopted the precipitation method to synthesize Cu₂O nanocrystals from PVP containing precursor solutions and the as-prepared octahedral-Cu₂O became active from 150 °C, obtaining a CO conversion of more than 80% at 200 °C. However, there are still many disadvantages in these methods, such as surfactant usage which could cover or poison active sites, insufficient catalytic activity, the complicated synthesis procedure and low yield. Hence, it is necessary to seek more facile and surfactant-free strategies to design and prepare the oxides of copper catalysts with clean surfaces, decent catalytic properties and the possibility to realize large-scale industrial production.

Recent years have witnessed extensive studies on the synthesis of nanoporous metals/composites and the versatile functionalization of them in catalysis,¹⁷ sensors,¹⁸ actuators,¹⁹ energy conversion/storage systems etc.,^20,21 due to their high specific surface areas and peculiar bicontinuous ligament–channel structure. As an example, Li et al.²² reported the high activity in the selective catalytic oxidation of alcohols over nanoporous silver. Dealloying, as an ancient depletion gilding technology, has been developed again in the synthesis of nanoporous metals/composites because of its effective and facile preparation advantages. It typically refers to the selective etching of one or more less noble element(s) out of suitable precursor alloys.^23,24 Dealloying can be triggered when the concentration of noble element(s) in precursor alloys is below a threshold value and the electrochemical potential is above the critical one.^25,26 Surface diffusion and reorganization of noble element(s) are involved during dealloying, which leads to the unique three-dimensional (3-D) bicontinuous interpenetrating ligament–channel structure.^27–29 It is worth mentioning that the emergence of nanoporous noble metals has greatly contributed to the development of CO oxidation catalysts. For example, Xu et al.³⁰ fabricated np-Au catalyst with ligament size below 6 nm by dealloying Ag–Au alloy in concentrated nitric acid, and the np-Au shows an outstanding catalytic behavior towards CO oxidation at low temperatures. Nevertheless, even though there are several literatures reported on the preparation of nanoporous Cu and its oxides,^6,31,32 the information on their CO catalytic oxidation behaviors is rare.¹³

Our previous results have proved that the nanoporous Cu@Cu₂O nanocomposites can be prepared via dealloying and the subsequent oxidation in air.⁶ In this work, we aim to further explore CO catalytic oxidation behaviors over nanoporous Cu–O system, synthesized through a facile and improved strategy combining dealloying and subsequent thermal treatment. Catalytic performance over the samples with different pretreatment conditions has been evaluated and compared. Catalysis tests reveal that the as-prepared nanoporous Cu–O system, particularly the sample after thermal annealing in air at 300 °C and subsequent treatment in hydrogen at 200 °C, exhibits an excellent catalytic activity toward CO oxidation.

2. Experimental section

2.1. Catalyst preparation

The Al_66.7Cu_33.3 (at.%, nominal composition) precursor alloy was prepared from elemental Al and Cu powders (99.9 wt% purity) by ball milling. The detailed milling procedure can be found in the literature.³³ The dealloying of the as-milled Al–Cu powders was performed in a 2 mol L⁻¹ NaOH aqueous solution at ambient temperature. Then the as-dealloyed samples were rinsed by ultra-pure water and dehydrated alcohol (analytical grade) and kept in a drying vacuum vessel. The subsequent treatment of the as-dealloyed samples included thermal annealing in a muffle furnace (air ambience) and/or a controlled atmosphere furnace (hydrogen) with temperature and duration control. The as-dealloyed samples experienced thermal annealing in air at two different temperatures (200 and 300 °C) for 1.5 h and subsequent treatment in hydrogen at 200 °C for 1.5 h.

2.2. Microstructural characterization

The phase constitution of the catalysts was characterized by X-ray diffraction (XRD, Rigaku D/max-rB) with Cu Kα radiation. The field-emission scanning electron microscope (FESEM, Quanta FEG 250) and the transmission electron microscope (TEM, FEI Tecnai G2) were employed to study the morphology, size distribution and nanoporous structure of the catalysts. Surface elemental information was detected by X-ray photoelectron spectroscopy (XPS, ESCALAB 250). The surface area of the catalysts was measured through N₂ adsorption–desorption experiments which were performed on an automatic surface area apparatus (ASAP2000) at 77 K.

2.3. Catalysis test

CO catalytic oxidation tests were carried out in a fixed-bed quartz tube reactor. Quartz sand (0.5 g) was used to disperse the catalysts (300 mg) in the tube reactor. A gas mixture with a volume ratio of CO/O₂/N₂ = 1/10/89 was fed into the reaction chamber at a total flow rate of 40 mL min⁻¹. A temperature control programmer was used to set the initial and final temperatures as well as the heating rate (3 °C min⁻¹). The outlet gas composition was analyzed by an on-line gas analyzer with ppm-precision. The CO conversion was calculated by the following expression: CO conversion = CO_{2 out}/(CO_{2 out} + CO_out) × 100%, where CO_{2 out} and CO_out were the corresponding concentration of CO₂ and CO at the outlet.

3. Results and discussion

3.1. Microstructure of the Cu–O catalysts

The as-milled Al_66.7Cu_33.3 alloy powders of around 30 g can be clearly seen from Fig. 1a. It is worth mentioning that the alloy powders in Fig. 1a were obtained from one of the four milling vials, and near 120 g of the same powders can be produced from one-batch milling in laboratory. This can ensure high yield of as-dealloyed catalysts, far outnumbering the Cu–O catalysts obtained from traditional wet chemical methods such as the hydrothermal strategy. In Fig. 1b, the corresponding XRD results verify that the as-milled sample is comprised of Al₄Cu₉ intermetallic compound, as well indicated from standard diffraction information (PDF #65-7542).

Fig. 1 (a) The as-milled Al_66.7Cu_33.3 (at.%, nominal composition, around 30 g) precursor alloy powders obtained from one of the four milling vials (a). (b) Corresponding XRD pattern of the alloy powders in (a).

Fig. 2 shows the XRD patterns of the Cu–O samples involved in CO catalytic oxidation experiments. The as-dealloyed sample is mainly composed of Cu (PDF # 65-9026) and Cu₂O (PDF # 65-3288, blue curve in Fig. 2). After catalytic experiments, the sample only consists of Cu₂O (cyan curve). With respect to the samples after thermal annealing in air at 200 and 300 °C, the phase compositions are different. Only Cu₂O can be detected by XRD after heat treatment at 200 °C (green curve), whereas a further rise in temperature to 300 °C makes CuO (PDF # 65-2309) the main phase (with weak diffraction peaks indicating the existence of Cu₂O, purple curve). As it indicates, the sample after 300 °C thermal annealing and 200 °C hydrogen treatment (black) is composed of three phases: Cu, Cu₂O and CuO. The 2 theta angles of 43.3, 50.5 and 74.1° correspond to the (111), (200) and (220) reflections of face centered cubic (f.c.c.) Cu, respectively. In addition, the major diffraction peaks at 36.5, 42.4, 61.5 and 73.7° can be indexed to the (111), (200), (220) and (311) reflections of Cu₂O, and the diffraction peaks at 35.7, 38.9, 48.9, 58.5, 66.6 and 68.4° can be attributed to the (−111), (111), (−202), (202), (−311) and (220) reflections of CuO. In comparison to the sample annealed at 300 °C in air, the subsequent hydrogen treatment makes the composites be partially reduced to metallic Cu. Even though there is report on the CO catalytic oxidation performance of Cu,³⁴ the stability of Cu, to some degree, relies on the gas ratio (volume ratio of CO/O₂). In our catalysis system, Cu can be further oxidized after catalytic tests (magenta and red curves). The related phase evolution has been highlighted by dashed lines in Fig. 2. Similar scenario can be noted for the as-dealloyed sample (blue curve) with consequent oxidation of Cu into Cu₂O after catalysis test (cyan curve). As shown in Fig. 2, the following catalytic test (from magenta to red) did not give rise to the further oxidation of Cu(I) into Cu(II), which indicates the relatively stable existence of Cu₂O in the given reaction conditions.

Fig. 2 XRD patterns for all the Cu–O samples involved in CO oxidation. From top to bottom: the as-dealloyed sample after 300 °C thermal annealing in air combined with the hydrogen treatment at 200 °C (black); the same sample after one catalytic cycle (magenta) and subsequent activation energy experiment (red); the samples after thermal annealing in air at 300 and 200 °C respectively (purple and green); the as-dealloyed sample after and before catalytic cycles (cyan and blue).

Fig. 3a and b are the low magnification FESEM images, showing the as-dealloyed sample before and after thermal annealing at 300 °C. Both the as-dealloyed and as-annealed samples are comprised of particles. The average particle size ranges from sub-micron (small particles) to several microns (large particles) for the as-dealloyed sample, Fig. 3a. And the annealing at 300 °C does not significantly change the size distribution, as shown in Fig. 3b. Moreover, the nanoporous structure cannot be discerned by SEM at low magnifications. The nanoporous structure of the as-dealloyed sample (nanoporous Cu@Cu₂O) has been documented in our previous work.⁶

Fig. 3 FESEM images showing the morphology and particle size of (a) the as-dealloyed sample and (b) the sample after thermal annealing at 300 °C in air.

For nanoparticles, the aggregation typically leads to the expected activity loss due to the nanostructure collapse during the annealing process.³⁵ However, this does not imply that the as-dealloyed particles will also lose their fine nanostructures in the annealing process. Fig. 4 provides the TEM images and corresponding selected-area electron diffraction (SAED) patterns of the sample after thermal treatment at 300 °C in air. The bright-field TEM image in Fig. 4a shows the microstructure of one particle. And the nanoporous structure can be clearly observed at higher magnification (Fig. 4b). The nanopores and ligaments are marked by solid and dotted arrows in Fig. 4b respectively. In addition, the as-dealloyed sample has been characterized by the Brunauer–Emmett–Teller (BET) method to obtain the specific surface area. The estimated BET surface area value is 19.5 m² g⁻¹ for the as-dealloyed sample (Fig. 5). In contrast, the sample after thermal treatment at 300 °C has a smaller surface area of 9.3 m² g⁻¹ (inset of Fig. 6), due to the coarsening of ligaments. The obvious hysteresis loop of the adsorption–desorption isotherms indicates the existence of nanopores. And we can also observe the existence of nanopores from the pore size distribution (Fig. 6). Moreover, the as-annealed sample displays a wide pore size distribution (from several nanometers to tens of nanometers). Combined with the TEM observations, it is reasonable to believe that the nanoporous structure has been retained for the sample after annealing at 300 °C. Fig. 4c shows the SAED pattern of the nanoporous area of the as-treated sample (the selected area is ∼200 nm in diameter). As indicated by the polycrystalline diffraction rings of the SAED results, the as-prepared catalysts have ligaments with nanocrystalline nature. Specifically, the (110), (−111), (−112), (202) and (022) reflections of CuO as well as the (111) reflection of Cu₂O can be well indexed from the diffraction rings, as highlighted in Fig. 4c. Fig. 4d shows the high resolution TEM (HRTEM) image of one area in Fig. 4a, which further confirms the nanoporous structure of the as-annealed sample. Some interplanar spacings were determined and marked in Fig. 4d. The values of 2.53 and 2.54 Å are close to 2.52 Å of the (−111) interplanar spacing of CuO. The TEM results are in conformity with the corresponding XRD pattern (purple curve in Fig. 2).

Fig. 4 (a and b) TEM and (d) HRTEM images of the sample after thermally annealed at 300 °C in air. (c) Corresponding SAED pattern of (a).

Fig. 5 N₂ adsorption–desorption isotherms of the as-dealloyed sample at 77 K.

Fig. 6 Pore size distribution of the sample annealed at 300 °C in air (inset: corresponding N₂ adsorption–desorption isotherms at 77 K).

3.2. Catalytic activity evaluation and catalytic mechanism probe

Fig. 7 presents the curve of CO conversion versus temperature over the as-dealloyed sample. The first cycle exhibits different catalytic progress with that of the last two cycles in Fig. 7, notwithstanding the similar sigmoid profile of these three consecutive activity tests. As summarized in Table 1, the T_5% (temperature of 5% CO conversion) is 96 °C in the first cycle, whereas the second and third cycles show a decrease of about 16 °C (80 and 79 °C obtained from the corresponding curves, respectively). Light-off temperature (T_50%) for the last two cycles also decreases to 125 and 123 °C, compared to the value of 146 °C in the first cycle. Even though the second and third cycles demonstrate better CO conversion within the temperature window than the first cycle, the values of T_90% for the three cycles are relatively close to each other, ranging from 150 to 158 °C (Fig. 7). According to the information provided by the XRD patterns (Fig. 2), phase evolution occurs accompanying with the first catalytic cycle. Exposed to the given reactive atmosphere, Cu in the as-dealloyed sample can easily be oxidized to Cu₂O, which is confirmed by the XRD results in Fig. 2. Phase composition of homogeneous Cu₂O has been maintained during the last two cycles, as indicated from the nearly overlapped curves in Fig. 7. Jernigan and Somorjai³⁴ studied the CO oxidation over Cu species with various oxidation states and they concluded that the metallic Cu, depending on the Langmuir–Hinshelwood mechanism, has better catalytic activity than Cu₂O and CuO. However, the as-dealloyed sample does not exhibit a higher onset catalytic activity in the first cycle than the other two, even though Cu indeed exists at the beginning of the first cycle (Fig. 7). Actually, there are controversies with respect to the intrinsic catalytic activity of metallic Cu. Huang and Tsai³⁶ found that Cu is not active as Cu₂O at low temperature range since Cu cannot provide surface lattice oxygen to react with CO. Their conclusions are on the basis of Mars–van-Krevelen redox mechanism in which two important steps are involved: the reaction between adsorbed CO and surface lattice oxygen of oxides of copper (Cu₂O/CuO) and the diffusion of subsurface oxygen into surface vacancies (or dissociative adsorption of O₂ from gas).^36,37 The former leads to the formation of surface oxygen vacancies (rate-limiting step), whereas the latter replenishes surface lattice oxygen vacancies.^36,37 With respect to the enhanced catalytic behaviors in both second and third cycle (Fig. 7), we cannot simply conclude that Cu₂O has improved activity over Cu. As pointed out by Nagase et al.,³⁸ Cu₂O formed through in situ reoxidation of metallic Cu is different from the initial Cu₂O, and the reoxidized Cu₂O is easily reduced by CO. Surface reconstruction typically happens during heterogeneous catalysis,¹⁵ and during this process the in situ formed Cu₂O may also bring and expose more active sites to reactive mixtures, rendering an enhanced catalytic performance.

Fig. 7 Consecutive three catalytic cycles in CO oxidation for the as-dealloyed sample.

Table 1 Summary of the catalytic performance towards CO oxidation over the different samples

Sample	T5% (°C)	T50% (°C)	T90% (°C)	Activation energy (KJ mol^−1)
As-dealloyed sample, 1st cycle	96	146	158
As-dealloyed sample, 2nd cycle	80	125	153
As-dealloyed sample, 3rd cycle	79	123	150
200 °C annealing, 1st cycle	100	149	174
200 °C annealing, 2nd cycle	96	144	168
300 °C annealing, 1st cycle	98	144	168
300 °C annealing, 2nd cycle	96	142	165
300 °C annealing and 200 °C hydrogen treatment, 1st cycle	84	126	139
300 °C annealing and 200 °C hydrogen treatment, 2nd cycle	80	118	132	91.3
CuO (ref. 12)				88.6–124
CuO (ref. 14)				79–115
CuO/ocatahedral-Cu2O, CuO/cubic-Cu2O (ref. 15)				73.4–110

---

Thermal annealing at moderate temperature is an effective method to produce copper oxides, and different oxidation states of Cu can be obtained through tuning the annealing temperature. The sample after annealing at 200 °C in air is composed of Cu₂O (Fig. 2) and the second cycle exhibits minor enhancement in catalytic conversion within the temperature window (Fig. 8a). The light-off and near complete conversion temperatures (144 and 168 °C, respectively) are just 5 °C lower in the second cycle than those in the first cycle (Fig. 8a and Table 1). Somorjai³⁹ pointed out that adsorbate (O₂ or CO) induced surface reconstruction accompanied by creating active crystal faces could promote the catalytic activity of catalyst. In addition, this minor enhancement detected by our ppm-level gas analyzer could also possibly be attributed to the surface reconstruction accompanied by the in situ oxidation of tiny amounts of Cu into Cu₂O. Compared to the as-dealloyed sample, the T_90% increases noticeably for the sample after annealing at 200 °C in air (Fig. 8a), and is around 15 °C higher in the first and second cycle (174 and 168 °C) than their counterparts in Fig. 7. As mentioned above, the same phase composition (Cu₂O) of the samples in Fig. 7 and 8a experienced different formation conditions, and the low density of active sites in the as-annealed sample (at 200 °C in air) could be an important reason for the catalytic activity decay. In ref. 40, Sadykov and Tikhov suggested that the catalytic activities tend to vary in a range due to the different surface defect nature, even though samples possess the same phase composition. Compared to the sample after annealing at 200 °C, the rise in annealing temperature to 300 °C leads to a slight improvement in catalytic performance, decreasing about 5 °C for light-off (144 °C) and near complete temperature (168 °C) in the first cycle (Fig. 8a and Table 1). The comparison is available and reasonable between the two samples after annealing at 200 and 300 °C, mainly because the formation condition of the two copper oxides (Cu₂O and CuO) is uniform thermal annealing. The minor catalytic enhancement in CuO can therefore be attributed to its relatively high concentration of lattice oxygen, when the same weight of CuO and Cu₂O is compared.³⁶ The high content of lattice oxygen is expected to participate in the redox reaction according to Mars–van-Krevelen mechanism.

Fig. 8 CO conversion versus temperature over (a) the samples after thermal annealing at 200 and 300 °C in air, respectively, and (b) the sample after thermal annealing at 300 °C in air combined with the hydrogen treatment at 200 °C.

With appropriate parameters, the thermal treatment in hydrogen can make the oxidized Cu species be partially reduced (Fig. 2). The sample after annealing at 300 °C in air and subsequent hydrogen treatment at 200 °C provides better catalytic performance (Fig. 8b) than that of the samples discussed in Fig. 7 and 8a, with approximately 20 °C decrease in the values of T_50% (126 °C) and T_90% (139 °C) than the as-dealloyed counterpart in the first cycle (Table 1). Besides Cu(I) species, the partially reductive treatment also gives rise to oxygen vacancies onto which not only the gaseous O₂ but also CO molecules can be adsorbed to further facilitate the catalytic reaction.⁴¹ Also, the mixture of Cu(II) and Cu(I) is regarded as the active structure,^11,37 since catalytic active defects such as grain boundaries exist at the interface between CuO and Cu₂O, serving as an ideal diffusion path for bulk lattice oxygen to compensate for oxygen vacancies.³⁷ During the first cycle, the reduced metallic Cu reoxidizes (Fig. 2), leading to a reconstructed and roughened surface. Such surface possesses more active sites which are responsible for the promoted catalytic activity in the second cycle (Fig. 8b). Corresponding reoxidation information can be obtained from the comparison of the XRD patterns in Fig. 2 (black and magenta). As shown in Fig. 8b, the second cycle exhibits the best catalytic behavior among all the samples, achieving the lowest T_50% of 118 °C and T_90% of 132 °C (Table 1). In comparison to the light-off temperature of reported copper oxides, the T_50% of 118 °C in this experiment is noticeably lower than the value of 210 °C for the commercially prepared CuO after activation under similar reaction conditions.¹² The T_50% of around 175 and 240 °C can be found in ref. 15 for CuO/ocatahedral-Cu₂O and CuO/cubic-Cu₂O nanocomposites. In ref. 14, CuO catalysts after H₂ plasma treatment show a CO conversion of 89% at 140 °C which is still a higher temperature compared to the 90% conversion at 132 °C demonstrated in this work. Considering its relatively high activity, the activation energy experiment and the XPS characterization were performed for the hydrogen treated sample after two catalytic cycles.

Fig. 9 shows the XPS results of the sample after the annealing at 300 °C in air and subsequent hydrogen treatment at 200 °C (after two catalytic cycles). The O 1s spectrum owns two peaks in Fig. 9a. Of the two peaks, the one at 530 eV can be attributed to oxygen ions and another at 531.6 eV originates from hydroxyl ions.⁴² As can be seen from Cu LMM auger spectrum in Fig. 9b, the peak at the binding energy of 569 eV well corresponds to Cu(I) instead of Cu(0).⁴³ This information is consistent with the XRD results which indicate that Cu₂O is stably retained after the first cycle and activation energy test (magenta and red curves in Fig. 2). XPS analysis in Fig. 9c confirms the existence of Cu(II) as the typical shake-up satellite peaks appear within the binding energy range from 940 to 945 eV.⁴⁴ Both the Cu LMM auger and XPS spectra verify that the surface of the sample is the mixture of Cu(I) and Cu(II) bi-oxidation states, which have been proposed to form important redox pairs in CO oxidation.³⁸ It should be noted that Nagase et al. found that the redox cycle between Cu(0) and Cu(II) could be present according to the Fourier transform infrared spectroscopy (FT-IR) results.³⁸ Moreover, the absence of Cu(0) in the XPS signal cannot be an exception or counter-example to reflect the irrationality of redox cycle between Cu(0) and Cu(II). Since the refilling of oxygen vacancies is fast during the CO oxidation reaction,³⁶ the Cu(0) could exist as metastable species during reaction and be oxidized at the terminating step of the redox reaction. And the underlying mechanism needs to be further clarified.

Fig. 9 (a) O 1s XPS spectrum, (b) Cu LMM auger spectroscopy and (c) Cu 2p XPS spectrum of the sample after the annealing at 300 °C in air and subsequent hydrogen treatment at 200 °C. Before the XPS measurement, the sample was also subjected to two catalytic cycles for CO oxidation.

To quantify and evaluate the activity, we investigated the activation energy of the sample after the annealing at 300 °C in air and subsequent hydrogen treatment at 200 °C (after two catalytic cycles). The Arrhenius equation is used to describe the mathematic relation between reaction constant and temperature:

where k is the reaction rate constant, A is the Arrhenius factor, E_a is the activation energy, R is the molar gas constant (8.314 J mol⁻¹ K⁻¹) and T is the absolute temperature. The following equation is used to determine reaction orders and rate constant at different temperatures:

R_CO = k[CO]^a[O₂]^b (2)

where a and b are corresponding reaction orders, and [CO] and [O₂] represent the volume concentration of CO and O₂, respectively. The temperature points of 366, 371, 376 and 381 K and different CO or O₂ concentrations (vol%, for [CO] ranging from 0.4% to 1% and [O₂] from 10% to 22%) in a total flow rate 40 mL min⁻¹ have been selected to estimate the corresponding reaction rate constant as well as reaction orders. The plot of ln R_CO versus ln[CO] or ln[O₂] is presented in Fig. 10a, from which the reaction orders for a and b were determined to be 0.54 and 0.17, respectively. Fig. 10b indicates that ln k versus 1/T gives a good linear correlation and the estimated activation energy is 91.3 KJ mol⁻¹. As summarized in Table 1, Pillai and Deevi¹² calculated the activation energy of their CuO catalysts to be within 88.6–124 KJ mol⁻¹, and Feng and Zheng¹⁴ reported from the activation energy of 79–115 KJ mol⁻¹ for their CuO nanowires. The activation energy of as-synthesized CuO/ocatahedral-Cu₂O and CuO/cubic-Cu₂O heterogeneous nanocrystals ranges from 73.4 to 110 KJ mol⁻¹.¹⁵ The comparison data imply that our nanoporous Cu–O catalysts exhibit a decent and comparable catalytic activity to the copper oxides reported in ref. 12, 14 and 15. In addition to the activation energy, synthesis methods are also non-ignorable factors if realistic applications are considered. Ball milling, dealloying and subsequent thermal treatment (oxidation and reduction) have been proven to be facile and promising for large-scale production. Moreover, the surfactant-free dealloying media ensures the catalyst will have a clean surface, resulting in active sites being directly exposed to the reactive gaseous mixture.

Fig. 10 (a) The dependence of reaction rate on CO or O₂ concentrations (vol%) for CO oxidation over the sample after thermal annealing at 300 °C in air combined with the hydrogen treatment at 200 °C. (b) The corresponding Arrhenius plot of (a).

4. Conclusions

In summary, we have developed nanoporous Cu–O system catalysts with different combination of Cu oxidation states through a facile dealloying strategy with subsequent thermal annealing. All the Cu–O samples show good catalytic oxidation performance with near complete conversion temperatures below 170 °C. The combination of annealing in air with hydrogen treatment can pursue the optimal catalytic performance, achieving the lowest light-off temperature (T_50%) of 118 °C and near complete conversion temperature (T_90%) of 132 °C. An activation energy of 91.3 KJ mol⁻¹ was achieved for our Cu–O catalysts, indicating their good catalytic performance towards CO oxidation. When comparing the catalytic activity over different oxidation states of Cu, formation pathways need to be considered as the same phase prepared through different ways may bring distinct surface defect nature, leading to different catalytic performance. Although catalytic mechanism has been cautiously and objectively discussed according to the previously and currently reported literatures, a solid in situ study and theoretical foundations from modeling are still necessary as the debates on the activity over different oxidation states of Cu have never been ceased. Even so, current emphasis is put on the synthesis strategy and decent performance. With a promising application prospect, present strategy could be scaled up for fabrication of the nanoporous Cu–O system catalysts, which could play versatile roles in CO elimination to substitute for commonly used Pt-group noble metals.

Acknowledgements

The authors gratefully acknowledge financial support by National Natural Science Foundation of China (51371106), Young Tip-top Talent Support Project (the Organization Department of the Central Committee of the CPC), Open Project of Shanghai Key Laboratory of Modern Metallurgy and Materials Processing (SELF-2011-02), Program for New Century Excellent Talents in University (MOE, NCET-11-0318), Specialized Research Fund for the Doctoral Program of Higher Education of China (20120131110017) and the Program for Professor of Special Appointment (Eastern Scholar) at Shanghai Institutions of Higher Learning (no. TP2014042).

References

1. R. A. Jones, J. A. Strickland, J. A. Stunkard and J. Siegel, Toxicol. Appl. Pharmacol., 1971, 19, 46.
2. Y. F. Y. Yao, J. Catal., 1984, 87, 152.
3. M. Haruta, N. Yamada, T. Kobayashi and S. Iijima, J. Catal., 1989, 115, 301.
4. M. Haruta, T. Kobayashi, H. Sano and N. Yamada, Chem. Lett., 1987, 16, 405.
5. S. Royer and D. Duprez, ChemCatChem, 2011, 3, 24.
6. T. Kou, C. Jin, C. Zhang, J. Sun and Z. Zhang, RSC Adv., 2012, 2, 12636.
7. Y. Sui, Y. Zeng, L. Fu, W. Zheng, D. Li, B. Liu and B. Zou, RSC Adv., 2013, 3, 18651.
8. K. Yamamoto, T. Kasuga and M. Nogami, Electrochem. Solid-State Lett., 1999, 2, 595.
9. P. Poizot, S. Laruelle, S. Grugeon, L. Dupont and J. M. Tarascon, Nature, 2000, 407, 496.
10. X. Gao, J. Bao, G. Pan, H. Zhu, P. Huang, F. Wu and D. Song, J. Phys. Chem. B, 2004, 108, 5547.
11. B. White, M. Yin, A. Hall, D. Le, S. Stolbov, T. Rahman, N. Turro and S. O’Brien, Nano Lett., 2006, 6, 2095.
12. U. R. Pillai and S. Deevi, Appl. Catal., B, 2006, 64, 146.
13. M. Zhou, Y. Gao, B. Wang, Z. Rozynek and J. O. Fossum, Eur. J. Inorg. Chem., 2010, 2010, 729.
14. Y. Feng and X. Zheng, Nano Lett., 2010, 10, 4762.
15. H. Bao, W. Zhang, Q. Hua, Z. Jiang, J. Yang and W. Huang, Angew. Chem., Int. Ed., 2011, 50, 12294.
16. H. Huang, L. Zhang, K. Wu, Q. Yu, R. Chen, H. Yang, X. Peng and Z. Ye, Nanoscale, 2012, 4, 7832.
17. Z. Qi, H. Geng, X. Wang, C. Zhao, H. Ji, C. Zhang and J. Xu, J. Power Sources, 2011, 196, 5823.
18. Z. Liu, J. Du, C. Qiu, L. Huang, H. Ma, D. Shen and Y. Ding, Electrochem. Commun., 2009, 11, 1365.
19. H. Jin, X. Wang, S. Parida, K. Wang and M. Seo, Nano Lett., 2010, 10, 187.
20. Z. Zhang, C. Zhang, J. Sun, T. Kou, Q. Bai, Y. Wang and Y. Ding, J. Mater. Chem. A, 2013, 1, 3620.
21. X. Lang, A. Hirata, T. Fujita and M. Chen, Nat. Nanotechnol., 2011, 6, 232.
22. Z. Li, J. Xu, X. Gu, K. Wang, W. Wang, X. Zhang, Z. Zhang and Y. Ding, ChemCatChem, 2013, 5, 1705.
23. Z. Zhang, Y. Wang, Z. Qi, C. Somsen, X. Wang and C. Zhao, J. Mater. Chem., 2009, 19, 6042.
24. X. Wang, J. Sun, C. Zhang, T. Kou and Z. Zhang, J. Phys. Chem. C, 2012, 116, 13271.
25. I. Oppenheim, D. J. Trevor, C. E. Chidsey, P. L. Trevor and K. Sieradzki, Science, 1991, 254, 687.
26. R. C. Newman and K. Sieradzki, Science, 1994, 263, 1708.
27. J. Erlebacher, M. Aziz, A. Karma, N. Dimitrov and K. Sieradzki, Nature, 2001, 410, 450.
28. Y. Ding and M. Chen, MRS Bull., 2009, 34, 569.
29. J. Weissmüller, R. C. Newman, H. Jin, A. M. Hodge and J. W. Kysar, MRS Bull., 2009, 34, 577.
30. C. Xu, J. Su, X. Xu, P. Liu, H. Zhao, F. Tian and Y. Ding, J. Am. Chem. Soc., 2007, 129, 42.
31. J. R. Hayes, A. M. Hodge, J. Biener, A. V. Hamza and K. Sieradzki, J. Mater. Res., 2006, 21, 2611.
32. Z. Qi, C. Zhao, X. Wang, J. Lin, W. Shao, Z. Zhang and X. Bian, J. Phys. Chem. C, 2009, 113, 6694.
33. Y. Wang, Y. Wang, C. Zhang, T. Kou and Z. Zhang, CrystEngComm, 2012, 14, 8352.
34. G. Jernigan and G. A. Somorjai, J. Catal., 1994, 147, 567.
35. M. Bagge-Hansen, A. Wichmann, A. Wittstock, J. R. I. Lee, J. Ye, T. M. Willey, J. D. Kuntz, T. Buuren, J. Biener, M. Bäumer and M. M. Biener, J. Phys. Chem. C, 2014, 118, 4078.
36. T. Huang and D. H. Tsai, Catal. Lett., 2003, 87, 173.
37. V. A. Sadykov, S. F. Tikhov, N. N. Bulgakov and A. P. Gerasev, Catal. Today, 2009, 144, 324.
38. K. Nagase, Y. Zheng, Y. Kodama and J. Kakuta, J. Catal., 1999, 187, 123.
39. G. A. Somorjai, Annu. Rev. Phys. Chem., 1994, 45, 721.
40. V. A. Sadykov and S. F. Tikhov, J. Catal., 1997, 165, 279.
41. A. Q. M. Boon, F. V. Looij and J. W. Geus, J. Mol. Catal., 1992, 75, 277.
42. K. Asami, K. Hashimoto and S. Shimodaira, Corros. Sci., 1977, 17, 713.
43. K. Yu, K. S. M. Pillai, P. R. Nalla and O. Chyan, J. Appl. Electrochem., 2010, 40, 143.
44. A. E. Palomares, F. Márquez, S. Valencia and A. Corma, J. Mol. Catal. A: Chem., 2000, 162, 175.

---