Oxygen storage capacity and thermal stability of the CuMnO₂–CeO₂ composite system

Abstract

Fast and reversible oxygen diffusion in solid oxides depending on oxygen partial pressure at low temperatures is a promising strategy for improving the overall performance and service lifetime of many energy-related materials. However, the high energy required for the redox reaction of cations and their high thermodynamic barriers have impeded the realization of fast oxygen diffusion at low temperatures. Herein, we report enhanced oxygen diffusion and storage capacity of monoclinic crednerite CuMnO₂ at a lower temperature by surface modification with CeO₂. The fast and reversible oxygen uptake/release can be attributed to CeO₂ that serves as a fast oxygen diffusion channel between bulk CuMnO₂ and the surrounding atmospheres. Importantly, the amount of CeO₂ in the CuMnO₂–CeO₂ composite system has a great effect on the total oxygen storage capacity and redox behaviour. Our findings could provide useful information for developing effective oxygen storage materials for wide energy-related applications.

---

1. Introduction

Solid-state oxygen storage materials (OSMs) have attracted considerable attention due to their wide applications in numerous oxygen-related energy and environmental fields, such as three-way catalysts for the effective removal of automobile exhaust emissions (e.g., NO_x, CO, and hydrocarbons).^1,2 The development of OSMs is also crucial to the success of new energy technologies, such as oxygen enrichment to improve the efficiency of chemical looping combustion.³ Ideal OSMs for practical applications should satisfy certain requirements and possess some properties, including large oxygen storage capacity (OSC), quick absorption/desorption of oxygen and their dependence on temperature and/or oxygen partial pressure, and good reversibility.

The most widely investigated OSMs are based on ceria (e.g., well-known Ce_1−xZr_xO_2+δ) due to the reversible redox reaction of Ce³⁺/Ce⁴⁺,^2,4–8 however, their OSC is relatively small and usually achieved by the usage of reductive gas. Recently, oxides based on transition metals have attracted remarkable attention due to their excellent properties, such as their flexible oxidation states, various phase structures, possible substitutions, cationic and anionic non-stoichiometry, or lattice oxygen deficient in the framework.^3,9–13 Their oxygen storage/release behaviour is generally based on the oxygen non-stoichiometry, which can be achieved by adjusting the surrounding oxygen partial pressure (i.e., oxidative air, reductive H₂) or temperature.^11,14 Among OSMs based on transition-metals, extensive attention has been paid to delafossite-type oxides with the general formula CuMO₂ (M = trivalent cation) because of the low reductive/oxidative temperature of Cu²⁺/Cu⁺, remarkable oxygen uptake ability, wide potential applications, environmental-friendliness and abundance.^15–17 However, only very limited work has been reported on the effect of trivalent cation species on the oxygen storage properties of the delafossite-type CuMO₂ oxides under oxidative/reductive atmospheres.^15–17 For example, Sumio Kato, et al. reported that CuMnO₂ and CuFeO₂ exhibited larger OSC values at lower temperature than those of CuAlO₂ and CuGaO₂,¹⁵ and OSC values for x = 0.1 and 0.3 in CuFe_1−xAl_xO₂ were larger than that for x = 0 above 500 °C.¹⁷ However, the full details of the phase-transformation process under oxidative/inert atmospheres and the effect of surface-modification by other metal oxides on the oxygen diffusion, oxygen storage capacity and thermal stability of CuMO₂ still remain not well-defined.

In the search for better OSMs, crednerite CuMnO₂ as an alternative phase to delafossite-type CuMO₂ seems to have great potential due to its changeable valence and low-temperature for oxygen uptake/release,^15,18 and its wide applications, such as a hydrogen photo-evolution catalyst,¹⁹ a three-way catalyst for the removal of exhaust gases (e.g., CO, NO_x),¹⁵ and hydrogen storage.²⁰ As reported by several research groups, crednerite CuMnO₂ exhibits a monoclinic structure at room temperature, which consists of edge-shared MnO₆ octahedra and two-coordinated Cu⁺ cations at the interlayer sites, as represented in Fig. 1.^15,19,21 This structure is closely related to the rhombohedral 3R delafossite structure of CuFeO₂ (i.e., the Cu⁺ cations linearly coordinating to two O ions, and the parallel O–Cu–O chains connecting Fe³⁺ cations to form two-dimensional sheets of edge-shared FeO₆ octahedra), but with a distortion of the MnO₆ octahedron due to the Jahn–Teller effect of Mn³⁺.²² However, it has been reported that Cu_1+xMn_1−xO₂ (0.08 < x < 0.12) at high temperature can exhibit a delafossite-like phase with a hexagonal structure because the thermal expansion of the structure would result in a larger lattice distortion than that from the Jahn–Teller effect.^23,24

Fig. 1 Schematic illustration of the crystal structure of CuMnO₂. The illustration was drawn with Diamond 3.1d software for crystal and molecular structure visualization.

It is well known that multiple valences of Mn cations can co-exist in many Mn-containing compounds, which could be beneficial to the oxygen-storage properties.^25–27 The valence of Mn in CuMnO₂ has been demonstrated to be +3, however the existence of Mn⁴⁺ in the nonstoichiometric Cu_1+xMn_1−xO₂ oxides is also detected by XPS because the excess Cu atoms would occupy the octahedral site with Cu²⁺ while the electrical neutrality principle would result in the formation of mixed Mn³⁺/Mn⁴⁺.^18,28 In addition, the Cu_xMn_3−xO₄ spinel, one of the oxidation products of CuMnO₂, has been reported to show the coexistence of Mn³⁺/Mn⁴⁺, implying a possible increase in the total oxygen uptake ability under an oxidative atmosphere.^27,29,30 However, pure crednerite CuMnO₂ oxide is usually synthesized using the solid-state reaction at very high temperatures^15,31,32 or the ion-exchange reaction with a long reaction time.³² Such a high reaction temperature or slow chemical conversion rate for reversible redox processes is impractical for many technological applications.¹⁸ Therefore, it is important to improve the oxygen diffusion ability and the overall oxygen storage capacity of crednerite CuMnO₂ at low temperatures.

CeO₂ has been widely investigated as a three-way catalyst or an oxygen promoter due to the fast oxygen ionic mobility between Ce⁴⁺ and Ce³⁺ in a reductive/oxidative atmosphere even though CeO₂ could retain its fluorite structure under oxidizing and mildly reducing atmospheres.^33–37 CeO₂ has also been reported as an interlayer or electrolyte in the La_1−xSr_xCoO_3−δ/CeO₂ composite system to improve the oxygen ionic mobility, in which some chemical reactions can take place at the interfaces between perovskite and CeO₂.^38,39 Therefore, CeO₂ may function as an oxygen ionic mobility channel between oxides and their surrounding atmosphere. Here, we investigated the CuMnO₂–CeO₂ composite system with the purpose of further optimizing the oxygen diffusion ability and overall oxygen storage capacity at low temperatures under alternating oxidative (i.e., air or O₂) and inert (i.e., Ar) atmospheres. Modifying the surfaces of CuMnO₂ with CeO₂ may bring in both benefits of CuMnO₂ and CeO₂, thereafter adjusting their thermochemical properties (e.g., redox properties, non-stoichiometry, oxygen exchange constant, and formation of oxygen vacancies), further enhancing their OSC and thermal stability. Our research results indicate that modifying CuMnO₂ with a small portion of CeO₂ (e.g., the molar ratios of CeO₂ to CuMnO₂ smaller than 20%) can improve the oxygen storage capacity at low temperatures (<600 °C) in a highly reversible manner.

2. Experimental section

2.1 Sample preparation

CuMnO₂–CeO₂ composites were prepared by a conventional Pechini method, followed by a solid state method. In a typical process, 10 mmol of Cu(NO₃)₂·2.5H₂O (98%) and 10 mmol of Mn(CH₃COO)₂·4H₂O (98%) were dissolved in 100 mL of deionized H₂O under continuous stirring. A certain amount (0.5, 1.0, 2.0 or 4.0 mmol) of Ce(NO₃)₃·6H₂O (99.99%) was added into the above solution under stirring, followed by adding 20 mmol of citric acid monohydrate (99.95%) and 10 mL of ethylene glycol. After stirring for 4 h, the solvent was evaporated at 110 °C to obtain a gel. After drying, the powder was ground and pre-fired at 500 °C under static air for 1 h to obtain the precursor, then the precursor was pressed into pellets and fired at 960 °C for 12 h under flowing argon. The obtained products were referred to as CuMnO₂–xCeO₂, in which x is the molar ratio percent of CeO₂ to CuMnO₂ (i.e., x = mol of CeO₂/mol of CuMnO₂ × 100).

2.2 Characterization

Structures of all samples were characterized by X-ray powder diffraction (XRD) on a PANalytical Empyrean Reflection Diffractometer with Cu Kα irradiation (λ = 0.15418 nm). HRTEM images and TEM elemental mapping of samples were observed on a JEM-2011 Transmission Electron Microscope (TEM) with an acceleration voltage of 200 kV equipped with an X-ray Energy Dispersive spectrometer (EDS). The morphologies of all samples were observed on a JEOL JSM-6700 Field Scanning Electron Microscope (FESEM). The oxygen uptake behaviours of these samples were measured by thermogravimetric analysis (TGA) on a NETZSCH TG 209 instrument (NETZSCH-Geraetebau GmbH, Selb, Germany) with a TASC 414/3 controller. The measurements were carried out for 50 mg specimens up to 800 °C with a heating rate of 10 °C min⁻¹ under air or O₂ with flowing rate 25 mL min⁻¹. In addition, the reversibility of the oxygen uptake/release was carried out in the following experiments: first increasing the temperature to 600 °C under flowing O₂ gas, then switching the gas from O₂ to Ar with temperature from 600 to 900 °C; after cooling down to 300 °C under flowing Ar, switching the gas from Ar to O₂ with temperature from 300 to 600 °C; this process was repeated between 300 and 900 °C 4 times. The thermal stability (weight and phase structural changes) of the as-prepared samples with temperature from room temperature to 1000 °C under flowing air was checked using a TGA/DTA technique carried out on a Stanton Redcroft STA-780 series thermal analyser.

3. Results and discussion

XRD patterns of these as-prepared CuMnO₂–xCeO₂ composites obtained by post-annealing at 960 °C for 12 h under flowing Ar are shown in Fig. 2. The XRD pattern for CuMnO₂ in Fig. 2a can be indexed to a pure monoclinic structure with a C2/m space group. All the peaks are in good agreement with the JCPDS card no. 50-0860 for the crednerite phase. For the CuMnO₂–xCeO₂ composites, they are obviously composed of crednerite CuMnO₂ and fluorite CeO₂, and no other impurity phases are observed. The XRD Rietveld refinements for CuMnO₂ and CeO₂ were carried out using GSAS software based on the monoclinic structure with the C2/m space group and cubic structure with the Fm3̄m space group, respectively. The typical XRD patterns after final refinement for CuMnO₂ and CuMnO₂–40CeO₂ are shown in Fig. 3, further confirming the phase composition of CuMnO₂ and CeO₂.²¹

Fig. 2 XRD patterns of the as-prepared CuMnO₂–xCeO₂ (i.e., post-annealed at 960 °C under flowing Ar): (a) CuMnO₂, (b) CuMnO₂–5CeO₂, (c) CuMnO₂–10CeO₂, (d) CuMnO₂–20CeO₂, and (e) CuMnO₂–40CeO₂.

Fig. 3 XRD patterns after final Rietveld refinements for the as-prepared products: (a) CuMnO₂ and (b) CuMnO₂–40CeO₂.

Detailed lattice parameters for pure CuMnO₂ and CuMnO₂–xCeO₂ composites after Rietveld refinements are summarized in Table 1. There are no obvious changes in the lattice parameters a, b and c for CuMnO₂ and a for CeO₂ with the increasing CeO₂ amount up to x = 40 for the CuMnO₂–xCeO₂ composite, suggesting that CuMnO₂ and CeO₂ exist mainly as separate phases and there may be only a little contact between the surfaces. The HRTEM images of CuMnO₂ in CuMnO₂–10CeO₂ and CuMnO₂–40CeO₂ (Fig. S1†) show clear lattice fringes and these d spacings of 0.22 and 0.57 nm correspond to the (111) and (001) planes of CuMnO₂, respectively. The TEM elemental mappings of Cu, Mn, O and Ce for CuMnO₂–10CeO₂ (Fig. S2†) and CuMnO₂–40CeO₂ (Fig. S3†) also indicate the individual existence of CuMnO₂ and CeO₂, as well as the inhomogeneous surface contact between CeO₂ and CuMnO₂.

Table 1 Lattice parameters of the as-prepared CuMnO₂–xCeO₂ obtained from Rietveld refinements using GSAS software

As-prepared sample	Composition	a (Å)	b (Å)	c (Å)	Cell volume (Å^3)	β
CuMnO2	CuMnO2	5.592(3)	2.883(1)	5.892(3)	92.1(1)	104.03(3)
CuMnO2–5CeO2	CuMnO2	5.597(3)	2.883(1)	5.892(3)	92.2(1)	104.00(4)
	CeO2		5.412(2)		158.5(2)
CuMnO2–10CeO2	CuMnO2	5.593(1)	2.883(6)	5.892(1)	92.19(3)	104.06(3)
	CeO2		5.412(1)		158.5(2)
CuMnO2–20CeO2	CuMnO2	5.589(2)	2.882(1)	5.891(2)	92.08(8)	104.08(4)
	CeO2		5.411(1)		158.4(1)
CuMnO2–40CeO2	CuMnO2	5.597(3)	2.883(1)	5.891(3)	92.24(9)	104.06(7)
	CeO2		5.412(2)		158.5(3)

---

The oxygen uptake behaviours of the as-prepared CuMnO₂–xCeO₂ composites were investigated by TGA at temperature up to 800 °C under flowing air or O₂. The results shown in Fig. 4 reveal a remarkable oxygen uptake capacity of this CuMnO₂–xCeO₂ composite system under an oxidative atmosphere, which corresponds to the following exothermic oxidation reaction:

CuMnO₂ + 1/(6 − 2y)O₂ → 1/(3 − y)Cu_yMn_3−yO₄ + (3 − 2y)/(3 − y)CuO (1)

Fig. 4 TGA curves of the as-prepared CuMnO₂–xCeO₂ samples (i.e., post-annealed at 960 °C under flowing Ar for 12 h) under flowing air or pure O₂ with 25 mL min⁻¹ from room temperature to 800 °C. The solid lines are the TGA curves under air and the short dashed lines are the TGA curves under pure O₂.

The weight for all these samples starts to increase at about 300 °C, due to the increase in the oxygen content, accompanied by the oxidation of Cu⁺ and/or Mn³⁺. Pure CuMnO₂ had a continuous and smooth weight increase with increasing temperature from 300 to 800 °C, reaching the maximum value of 5.475 wt% at 800 °C under both flowing air and O₂, corresponding to the formation of spinel Cu_1.058Mn_1.942O₄. For pure CuMnO₂ and CuMnO₂–xCeO₂ composites, the oxygen uptake rate in air is a bit slower than that in O₂ and the maximum oxygen uptake amount in air is also a bit smaller than that in O₂, in which CuMnO₂–5CeO₂ exhibits the most weight increase, reaching 6.169 wt% at 591 °C under flowing O₂ and 6.031 wt% at 620 °C under flowing air, respectively, suggesting that higher oxygen partial pressure would enhance the oxygen uptake ability at lower temperatures. With the further increase of CeO₂ amount in the composite, the maximum OSC decreased, which can be attributed to the offset effect of CeO₂ since the weight increase is mainly caused by the oxidation of Cu⁺ and/or Mn³⁺ in CuMnO₂ while CeO₂ could maintain its fluorite-type structure under an inert atmosphere,^40,41 as determined by the XRD results in Fig. 2. After reaching the maximum weight at around 600 °C, further increasing the temperature to 800 °C results in a slight and smooth weight decrease (<0.14 wt%) for all these CeO₂-modified CuMnO₂ samples, which can be attributed to the composition adjustment of spinel Cu_yMn_3−yO₄ with temperature.¹⁸ These results also indicate that CeO₂-modification would favour the oxygen mobility in CuMnO₂ and the formation of spinel Cu_yMn_3−yO₄ with more Mn⁴⁺ at lower temperatures. It should be noted that the oxygen uptake in CuMnO₂ and CuMnO₂–xCeO₂ composites mainly comes from the oxidation of CuMnO₂ to CuO and spinel Cu_yMn_3−yO₄. Assuming that the total weight increase only comes from the oxidation of CuMnO₂, the composition of spinel Cu_yMn_3−yO₄ at the maximum weight under flowing O₂ can be calculated based on eqn (1) to be Cu_1.370Mn_1.630O₄ for the CuMnO₂–5CeO₂ starting mix, Cu_1.317Mn_1.683O₄ for CuMnO₂–10CeO₂, Cu_1.356Mn_1.644O₄ for CuMnO₂–20CeO₂ and Cu_1.316Mn_1.684O₄ for CuMnO₂–40CeO₂, while the composition of spinels at the maximum weight under flowing air can be calculated to be Cu_1.332Mn_1.668O₄ for CuMnO₂–5CeO₂, Cu_1.296Mn_1.704O₄ for CuMnO₂–10CeO₂, Cu_1.310Mn_1.690O₄ for CuMnO₂–20CeO₂ and Cu_1.268Mn_1.732O₄ for CuMnO₂–40CeO₂, respectively. These calculation results suggest that higher oxygen partial pressure favours the formation of spinels with more Mn⁴⁺ from the oxidation of CuMnO₂–xCeO₂ composites.

XRD patterns of the oxygenated CuMnO₂ and CuMnO₂–xCeO₂ composites after the TGA test up to 800 °C under flowing air are shown in Fig. 5. While CeO₂ in these oxygenated CuMnO₂–xCeO₂ composites still maintained its fluorite-type cubic structure, CuMnO₂ was oxidized to CuO and spinel Cu_yMn_3−yO₄ oxides, in good agreement with the TGA analysis in Fig. 4 and previous reported results.¹⁸ The lattice parameters of CeO₂ and spinel Cu_yMn_3−yO₄ in the oxygenated CuMnO₂ and CuMnO₂–xCeO₂ composites after the TGA test up to 800 °C in flowing air were refined from the XRD results in Fig. 5 using GSAS software and are summarized in Table 2. The lattice parameters of CeO₂ in the oxygenated CuMnO₂–xCeO₂ composites are a bit smaller than those of the as-prepared CuMnO₂–xCeO₂ composites (as shown in Table 1), indicating the possible doping of Mn and/or Cu in the lattice of CeO₂ after oxidation. In addition, the lattice parameters of spinel Cu_yMn_3−yO₄ in the oxygenated CuMnO₂–xCeO₂ composites are also slightly smaller than those of oxygenated pure CuMnO₂, indicating the higher Cu/Mn ratios in the spinel Cu_yMn_3−yO₄ in the oxygenated CuMnO₂–xCeO₂ composites as the smaller ionic radii of Cu²⁺ than that of Mn²⁺ would result in a decrease of lattice parameters when copper ions replace manganese ions on tetrahedral sites.

Fig. 5 XRD patterns after the TGA test from room temperature to 800 °C under flowing air: (a) CuMnO₂, (b) CuMnO₂–5CeO₂, (c) CuMnO₂–10CeO₂, (d) CuMnO₂–20CeO₂, and (e) CuMnO₂–40CeO₂.

Table 2 Lattice parameters of oxygenated CuMnO₂–xCeO₂ in Fig. 5 from Rietveld refinements using GSAS software

Oxygenated CuMnO2–xCeO2	Composition	a (Å)	Cell volume (Å^3)
CuMnO2	CuyMn3−yO4	8.317(2)	575.30(7)
CuMnO2–5CeO2	CuyMn3−yO4	8.307(2)	573.23(4)
	CeO2	5.407(8)	158.14(7)
CuMnO2–10CeO2	CuyMn3−yO4	8.311(5)	574.06(3)
	CeO2	5.409(1)	158.26(1)
CuMnO2–20CeO2	CuyMn3−yO4	8.309(4)	573.64(9)
	CeO2	5.410(6)	158.39(3)
CuMnO2–40CeO2	CuyMn3−yO4	8.312(2)	574.27(1)
	CeO2	5.410(2)	158.35(8)

---

To further check their thermal stability with temperature under flowing air, we treated the as-prepared samples (i.e., post-annealed at 960 °C for 12 h under flowing Ar) from room temperature up to 1000 °C with heating and cooling rates of ±10 °C min⁻¹ under flowing air, as shown in Fig. 6. As discussed in aforementioned paragraphs, the as-prepared samples can start to take up oxygen near 300 °C and reach the maximum values at near 600 °C for CuMnO₂–xCeO₂ composites and 800 °C for pure CuMnO₂, respectively. On further heating, these samples exhibit a steep weight loss from 960 °C accompanied by an endothermic peak in the DTA curves, as shown in Fig. S4.† Such behaviour can be attributed to the transformation of spinels and CuO back into a crednerite Cu_1+y′Mn_1−y′O₂ phase, accompanied by the release of O₂.^18,32 During the cooling process under flowing air from 1000 °C, there is a sharp continuous weight increase up to around 880 °C, which can be attributed to the re-oxidation of the crednerite phase to CuO and spinel oxides. The slight weight increase from 880 to 600 °C can be attributed to the composition adjustment of spinels with the cooling temperature, as determined by XRD shown in Fig. S5.† Notably, their weights at 1000 °C are higher than their pristine counterparts, suggesting that CuO and spinel Cu_yMn_3−yO₄ at high temperature under air can be converted to Cu_1+y′Mn_1−y′O₂ rather than CuMnO₂ according to eqn (2).

1/(3 − y)Cu_yMn_3−yO₄ + (3 − 2y)/(3 − y)CuO → αCu_x′Mn_3−x′O₄ + (2 − α)/2Cu_1+y′Mn_1−y′O₂ + [1/(6 − 2y) − α/2]O₂ (2)

where α = 2y′/(3 + 3y′ − 2x′).

Fig. 6 TGA curves of the as-prepared CuMnO₂–xCeO₂ samples (i.e., post-annealed at 960 °C for 12 h under flowing Ar) from room temperature up to 1000 °C with heating and cooling rates of 10 °C min⁻¹ under flowing air.

The TGA behaviours of the oxygenated CuMnO₂–xCeO₂ composites (i.e., after the TGA test up to 800 °C under flowing O₂) with temperature up to 900 °C under flowing argon are displayed in Fig. 7. The results show that all these samples are stable up to around 600 °C under flowing argon. Further increasing temperature would lead to a continuous weight loss (<1.2 wt%) up to about 790 °C, in which the oxygenated CuMnO₂–5CeO₂ exhibited the most weight loss and the total weight loss decreased with the increase of CeO₂. The small weight loss can be attributed to the removal of oxygen in spinels. At temperatures higher than 790 °C, these sharp weight losses are attributed to the further removal of oxygen to form crednerite Cu_1+y′Mn_1−y′O₂.³¹ To further support this conclusion, we treated the oxygenated CuMnO₂, CuMnO₂–5CeO₂ and CuMnO₂–20CeO₂ under flowing argon at 900 °C for 2 h and their XRD patterns shown in Fig. S6† indicate the reformation of the crednerite CuMnO₂ structure. The total weight losses for the oxygenated CuMnO₂–5CeO₂, CuMnO₂–10CeO₂ and CuMnO₂–20CeO₂ are higher than that for oxygenated CuMnO₂. The enhanced oxygen mobility and storage capacity of the CuMnO₂–xCeO₂ composite system can be attributed to the synergistic effect between CuMnO₂ and CeO₂, in which CeO₂ can act as an oxygen transfer channel or oxygen promoter dependent on temperature and/or oxygen partial pressures. The possible doping or exsolution of Cu/Mn into or from the lattice of CeO₂ on the surface depending on the temperature and oxygen partial pressure may also contribute to the enhanced performance. Therefore, with the surface modification of CeO₂, CuMnO₂ can take up oxygen from CeO₂ rather than directly from the surrounding O₂ atmosphere and vice versa, under an inert Ar atmosphere, CuO and spinel Cu_yMn_3−yO₄ can release oxygen to the CeO₂ lattice and then to the surroundings, rather than directly to the surroundings.

Fig. 7 TGA curves of oxidized CuMnO₂–xCeO₂ samples (i.e., after the TGA test up to 800 °C under flowing O₂) from room temperature to 900 °C under flowing argon with 25 mL min⁻¹.

From the TGA data in Fig. 4, it appears that these CeO₂-modified CuMnO₂ oxides can be oxidized from 300 °C and reach the maximum weight at around 600 °C under flowing O₂, while the oxygenated samples can start to release oxygen from 600 °C under flowing argon based on the TGA results in Fig. 7. These results indicate that CeO₂-modified CuMnO₂ oxides can cause either oxygen uptake or release depending on the temperature and/or oxygen partial pressure. Thus, remarkable oxygen uptake/release behaviours are expected via adjusting the temperature and oxygen partial pressure. As demonstrated in Fig. 8, CuMnO₂ and CeO₂-modified CuMnO₂ oxides can take up/release a large amount of oxygen in which CuMnO₂–5CeO₂ exhibits the highest OSC (ca. 6.0 wt%) and CuMnO₂–20CeO₂ shows the best reversibility under the alternating O₂ and argon between 300 and 900 °C for four cycles, indicating that CeO₂ modification would improve the oxygen storage capacity as well as the reversibility of CuMnO₂. The OSC and reversibility would become worse with the increasing cycle times, as shown in Fig. S7,† but the final weight after each treatment under flowing argon during the cycles almost remains the same, indicating that the gradual weight loss may be due to the formation of spinels with less Mn⁴⁺ during the oxidation process with the increasing cycling times. As shown in Fig. S8,† after a nine cycle test under alternating O₂ and argon, CuMnO₂–5CeO₂ still contains crednerite CuMnO₂ and fluorite CeO₂ phases; however, there are obvious CeO₂ nanoparticles on the CuMnO₂ surface, in comparison to the FESEM images in Fig. S9.† The TEM results for CuMnO₂–10CeO₂ after the TGA test, as shown in Fig. S10,† also indicate the appearance of CeO₂ nanoparticles on the surface of CuMnO₂, which resulted in a worse performance for the oxidation of CuMnO₂. The TEM elemental mapping results for CuMnO₂–10CeO₂ in Fig. S10† suggest that some Cu is doped into the lattice of CeO₂ while Mn is still in CuMnO₂, which can be explained by the bigger ionic radius of Cu²⁺ (0.73 Å) and Cu⁺ (0.77 Å) than that of Mn³⁺ (0.58 Å) and Mn⁴⁺ (0.53 Å).

Fig. 8 TGA curves of the as-prepared CuMnO₂–xCeO₂ samples (i.e., post-annealed at 960 °C for 12 h under flowing Ar) under alternating O₂ and argon. The samples were first oxidized under O₂ to 600 °C, then the gas was switched to argon and the temperature was increased to 900 °C. After being cooled down to 300 °C under argon, the gas was then switched back to O₂ and oxidized to 600 °C.

4. Conclusions

In conclusion, we found that fast and reversible oxygen mobility, and increased oxygen storage capacity of crednerite CuMnO₂ could be achieved at reduced temperatures (ca. 600 °C) by surface modification with CeO₂ at an amount less than 20 mol% in the CuMnO₂–CeO₂ composite system. The fast and reversible oxygen uptake/release and increased oxygen storage capacity at lower temperatures can be attributed to the synergistic effect of CeO₂ as an oxygen diffusion channel between bulk CuMnO₂ and the surrounding atmospheres, favouring the reversible formation of spinels at lower temperatures. Our findings reported here could provide a pathway for the design and development of effective oxygen storage materials for a range of energy-synergetic related applications at low temperatures.

Acknowledgements

The authors gratefully thank the Engineering and Physical Sciences Research Council (EPSRC) platform grant EP/I022570/1 and EP/I022570/2 for financial support.

Notes and references

1. H. C. Yao and Y. F. Y. Yao, J. Catal., 1984, 86, 254–265.
2. C. E. Hori, H. Permana, K. Y. S. Ng, A. Brenner, K. More, K. M. Rahmoeller and D. Belton, Appl. Catal., B, 1998, 16, 105–117.
3. T. Motohashi, Y. Hirano, Y. Masubuchi, K. Oshima, T. Setoyama and S. Kikkawa, Chem. Mater., 2013, 25, 372–377.
4. M. Ozawa, M. Kimura and A. Isogai, J. Alloys Compd., 1993, 193, 73–75.
5. A. Gupta, U. V. Waghmare and M. S. Hegde, Chem. Mater., 2010, 22, 5184–5198.
6. M. P. Yeste, J. C. Hernandez-Garrido, D. C. Arias, G. Blanco, J. M. Rodriguez-Izquierdo, J. M. Pintado, S. Bernal, J. A. Perez-Omil and J. J. Calvino, J. Mater. Chem. A, 2013, 1, 4836–4844.
7. X. Wang, G. Lu, Y. Guo, L. Jiang, Y. Guo and C. Li, J. Mater. Sci., 2009, 44, 1294–1301.
8. Q. Dong, S. Yin, C. Guo, T. Kimura and T. Sato, RSC Adv., 2012, 2, 12770–12774.
9. S. Remsen and B. Dabrowski, Chem. Mater., 2011, 23, 3818–3827.
10. O. Parkkima, H. Yamauchi and M. Karppinen, Chem. Mater., 2013, 25, 599–604.
11. S. Carter, A. Selcuk, R. J. Chater, J. Kajda, J. A. Kilner and B. C. H. Steele, Solid State Ionics, 1992, 53–56, 597–605.
12. M. Cherry, M. S. Islam and C. R. A. Catlow, J. Solid State Chem., 1995, 118, 125–132.
13. J. W. Lekse, S. Natesakhawat, D. Alfonso and C. Matranga, J. Mater. Chem. A, 2014, 2, 2397–2404.
14. H. Jeen, W. S. Choi, M. D. Biegalski, C. M. Folkman, I. C. Tung, D. D. Fong, J. W. Freeland, D. Shin, H. Ohta, M. F. Chisholm and H. N. Lee, Nat. Mater., 2013, 12, 1057–1063.
15. S. Kato, R. Fujimaki, M. Ogasawara, T. Wakabayashi, Y. Nakahara and S. Nakata, Appl. Catal., B, 2009, 89, 183–188.
16. S. Kato, H. Sato, M. Ogasawara, T. Wakabayashi, Y. Nakahara and S. Nakata, Solid State Sci., 2012, 14, 177–181.
17. S. Kato, S. Suzuki, R. Kawashima, M. Ogasawara, T. Wakabayashi, Y. Nakahara and S. Nakata, J. Mater. Sci., 2013, 48, 8077–8083.
18. M. Trari, J. Töpfer, P. Dordor, J. C. Grenier, M. Pouchard and J. P. Doumerc, J. Solid State Chem., 2005, 178, 2751–2758.
19. Y. Bessekhouad, M. Trari and J. P. Doumerc, Int. J. Hydrogen Energy, 2003, 28, 43–48.
20. S. A. M. Abdel-Hameed, F. H. Margha and A. A. El-Meligi, Int. J. Energy Res., 2014, 38, 459–465.
21. Y. Bessekhouad, Y. Gabes, A. Bouguelia and M. Trari, J. Mater. Sci., 2007, 42, 6469–6476.
22. J. Topfer, M. Trari, P. Gravereau, J. P. Chaminade and J. P. Doumerc, Z. Kristallogr., 1995, 210, 184–187.
23. F. C. M. Driessens and G. D. Rieck, Z. Anorg. Allg. Chem., 1967, 351, 48–62.
24. P. Wei, M. Bieringer, L. D. Cranswick and A. Petric, J. Mater. Sci., 2010, 45, 1056–1064.
25. J. Li, S. Xiong, X. Li and Y. Qian, Nanoscale, 2013, 5, 2045–2054.
26. N. K. Radhakrishnan and A. B. Biswas, Phys. Status Solidi A, 1977, 44, 45–49.
27. D. P. Shoemaker, J. Li and R. Seshadri, J. Am. Chem. Soc., 2009, 131, 11450–11457.
28. H. Y. Chen and D. J. Hsu, J. Alloys Compd., 2014, 598, 23–26.
29. I. N. Dubrovina, V. F. Balakirev and A. V. Antonov, Inorg. Mater., 2001, 37, 76–81.
30. E. Ríos, S. Abarca, P. Daccarett, H. Nguyen Cong, D. Martel, J. F. Marco, J. R. Gancedo and J. L. Gautier, Int. J. Hydrogen Energy, 2008, 33, 4945–4954.
31. A. P. Amrute, Z. Łodziana, C. Mondelli, F. Krumeich and J. Pérez-Ramírez, Chem. Mater., 2013, 25, 4423–4435.
32. B. Bellal, B. Hadjarab, N. Benreguia, Y. Bessekhouad and M. Trari, J. Appl. Electrochem., 2011, 41, 867–872.
33. Z. Liu, Z. Wu, X. Peng, A. Binder, S. Chai and S. Dai, J. Phys. Chem. C, 2014, 118, 27870–27877.
34. L. Shi, W. Chu, F. Qu and S. Luo, Catal. Lett., 2007, 113, 59–64.
35. A. Hornés, A. B. Hungría, P. Bera, A. L. Cámara, M. Fernández-García, A. Martínez-Arias, L. Barrio, M. Estrella, G. Zhou, J. J. Fonseca, J. C. Hanson and J. A. Rodriguez, J. Am. Chem. Soc., 2009, 132, 34–35.
36. S. D. Senanayake, D. Stacchiola and J. A. Rodriguez, Acc. Chem. Res., 2013, 46, 1702–1711.
37. I. Moog, C. Feral-Martin, M. Duttine, A. Wattiaux, C. Prestipino, S. Figueroa, J. Majimel and A. Demourgues, J. Mater. Chem. A, 2014, 2, 20402–20414.
38. E. Y. Konysheva, S. M. Francis, J. T. S. Irvine, A. Rolle and R.-N. Vannier, J. Mater. Chem., 2011, 21, 15511–15520.
39. E. Konysheva, R. Blackley and J. T. S. Irvine, Chem. Mater., 2010, 22, 4700–4711.
40. A. S. Ivanova, Kinet. Catal., 2009, 50, 797–815.
41. Y. Zhou and M. N. Rahaman, Acta Mater., 1997, 45, 3635–3639.

---

Footnote

† Electronic supplementary information (ESI) available: Additional XRD, TGA/DTA, HRTEM and FESEM images as well as TEM elemental mappings. See DOI: 10.1039/c5ta01361e

---