Composite NiCo2 O4 @CeO2 Microsphere as Cathode Catalyst for High-Performance Lithium-Oxygen Battery.

The large overpotential and poor cycle stability caused by inactive redox reactions are tough challenges for lithium-oxygen batteries (LOBs). Here, a composite microsphere material comprising NiCo2 O4 @CeO2 is synthesized via a hydrothermal approach followed by an annealing processing, which is acted as a high performance electrocatalyst for LOBs. The unique microstructured catalyst can provide enough catalytic surface to facilitate the barrier-free transport of oxygen as well as lithium ions. In addition, the special microsphere and porous nanoneedles structure can effectively accelerate electrolyte penetration and the reversible formation and decomposition process of Li2 O2 , while the introduction of CeO2 can increase oxygen vacancies and optimize the electronic structure of NiCo2 O4 , thereby enhancing the electron transport of the whole electrode. This kind of catalytic cathode material can effectively reduce the overpotential to only 1.07 V with remarkable cycling stability of 400 loops under 500 mA g-1 . Based on the density functional theory calculations, the origin of the enhanced electrochemical performance of NiCo2 O4 @CeO2 is clarified from the perspective of electronic structure and reaction kinetics. This work demonstrates the high efficiency of NiCo2 O4 @CeO2 as an electrocatalyst and confirms the contribution of the current design concept to the development of LOBs cathode materials.

Introduction

The growing requirement for future energy storage systems has promoted the rapid development of battery research.[ , , , ] Among them, nonaqueous rechargeable lithium–oxygen batteries (LOBs) with ultrahigh theoretical energy density (around 3505 Wh kg−1) stand out among numerous rechargeable batteries, showing unprecedented potential.[ , , , ] LOBs have widespread applications in transportation, portable electronic devices, and energy storage applications, which have attracted much attention.[ , , , , ] The typical discharge reaction (oxygen reduction reaction (ORR)) of LOBs is: 2Li + O2 ↔ Li2O2, while the charge reaction (oxygen evolution reaction (OER)) is the decomposition of Li2O2.[ , , ] However, the insoluble and insulating products generated during the discharge process are easy to deposit around the positive electrode material and block the active site, resulting in excessive overpotential,[ , ] poor stability, short cycle life,[ ] and even final death of batteries.[ ] Therefore, the positive electrode material should not only have a high specific surface area to accommodate Li2O2, but also a high catalytic activity to promote the decomposition of lithium peroxide.[ ] The development of bi‐functional electrocatalysts that can effectively enhance OER and ORR is one of the challenges of high‐powered LOBs.[ ]

Currently, bi‐functional catalysts for LOBs mainly include carbon‐based materials, noble metals, transition metal oxides, etc.[ , ] Although noble metals such as Pt/C have been proved to be effective catalysts, the disadvantage of high prices makes people more inclined to use transition metal oxides with low cost, abundant reserves, and relatively high electrochemical activity.[ , ] In particular, the spinel NiCo2O4 has received widespread attention due to its extremely low overpotential and corrosion resistance.[ ] Some studies have shown that NiCo2O4, as a bimetallic oxide, can utilize the synergistic effect between Ni and Co to promote OER and ORR[ , ] and it is also more conductive than single metal oxide. Besides, NiCo2O4 with different morphology would help to reduce discharge/charge overpotential and increase capacity. Several different types of nanostructured NiCo2O4, including nanoparticles,[ ] nanoplates,[ ] nanosheets,[ ] nanowires,[ ] and microspheres,[ , ] have been successfully synthesized and applied in various energy storage devices. However, compared with other reported high‐efficiency cathodes, the cathode structure and catalytic effect of NiCo2O4 still need to be further optimized and strengthened.[ ]

Cerium dioxide (CeO2) possesses high catalytic activity as a rare earth metal oxide. It can be used as an “oxygen buffer” to flexibly change the valence state according to the concentration of oxygen, thus helping to enhance the migration ability of oxygen species and significantly improve the ORR performance of LOBs.[ , , ] This process can be expressed as CeO2↔CeO + (2−x)/2O2 (1.5 < x < 2).[ , ] The oxygen in the lattice of CeO2 has high activity and fluidity, which is conducive to the generation of oxygen vacancies, thereby increasing the active sites of the reaction. Studies have shown that the introduction of CeO2 will increase the specific capacity of LOBs, reduce overpotential, and improve cyclability.[ , , , , ] However, the poor conductivity and minimal area of pure CeO2 nanoparticles limit their development, which needs to be combined with a conductive substrate with a high specific surface area.[ , , ] Therefore, compositing the urchin‐like NiCo2O4 with CeO2 nanoparticles is a feasible solution to significantly promote the performance of LOBs.

In this paper, we report a fairly convenient approach to directly synthesize a superior composite structure with NiCo2O4 urchin‐like microsphere as the core and CeO2 nanoparticles as an embellishment, working as an effective OER and ORR dual‐function catalyst for the rechargeable nonaqueous LOBs. Such microscale structure can provide sufficient catalytic surface and open space to facilitate the transfer of oxygen or lithium ions, provide more active sites and adequate specific surface area to store the products during the discharge process. In addition, the porous structure of the nanoneedles significantly promotes barrier‐free oxygen transmission and electrolyte penetration, helping to establish multi‐dimensional charge diffusion paths. The introduction of CeO2 can increase oxygen vacancies, which not only generate a great number of nonsaturable sites, but also erect an electron transport bridge, thereby highly promoting the charge transfer kinetics. The excellent catalytic effect of the composite material on the OER and ORR of the LOBs is mainly attributed to these synergistic advantages, which can effectively reduce overpotential to 1.07 V and extend the cycle life that can reach 400 times under the restricted capacity of 500 mA h g−1 with an ampere density of 500 mA g−1. This special urchin‐like composite structure provides a new way to build up the performance of LOBs.

Results and Discussion

Morphology and Structure Characterization

The synthesis method of composite urchin‐like NiCo2O4@CeO2 microspheres is achieved by a two‐step method, as shown in Figure  . First, the NiCo2O4@CeO2 precursor is prepared by a one‐step hydrothermal reaction. After being calcined in air, the precursor changes from smooth nanoneedles to porous nanotubes due to unbalanced diffusion,[ ] which is denoted as NiCo2O4@CeO2 in this work. For comparison, NiCo2O4 microspheres and CeO2 nanoparticles are also prepared by the same procedure, respectively.

Figure 1

a) The schematic diagram of preparation process for NiCo2O4@CeO2. b) XRD patterns of NiCo2O4@CeO2 and NiCo2O4. c) SEM images of NiCo2O4@CeO2 at low magnification (inset: enlarged image of nanoneedles).

The phase structure and composition of the synthesized samples were determined by X‐ray diffraction (XRD) characterization. The XRD pattern of the precursor (Figure S1, Supporting Information) shows that it is composed of Ni2(OH)2CO3·4H2O, Co(CO3)0.5(OH)·0.11H2O, and Ce(CO3)2O·H2O. From Figure 1b, it could be found that in the crystal structure of all composite materials, all typical peaks can be exactly matched with spinel NiCo2O4 (JCPDS no. 73–1702) and CeO2 (JCPDS no. 04–0593). Specifically, the typical diffraction peaks located at 18.9°, 31.1°, 36.7°, 38.3°, 44.6°, 55.4°, 59.1°, and 64.9° can be accurately indexed as (111), (220), (311), (222), (400), (422), (511), and (440) planes of NiCo2O4, and the three characteristic peaks at 28.5°, 33.0°, and 47.4° can be ascribed to (111), (200), and (220) planes of CeO2. The structure of individual CeO2 and its precursor are demonstrated in Figure S2 (Supporting Information).[ , , ] It can be confirmed that the hydroxy carbonate of the precursor is well converted into oxide, and the NiCo2O4@ CeO2 composite structure is successfully formed. It should be noted that compared with a single substance, the diffraction peak intensity of NiCo2O4 decreases with the recombination of CeO2, indicating a slight decrease in crystallinity. In any case, these results confirm that NiCo2O4 and CeO2 coexist in the composite structure, and the prepared samples have high crystallinity without impurities.

The morphology of NiCo2O4@CeO2, NiCo2O4, and carbon nanotubes (CNTs) was detected by scanning electron microscope (SEM) in Figure 1c and Figure S3 (Supporting Information). Figure S4a (Supporting Information) shows that the precursor of the composited NiCo2O4@CeO2 is about 4 µm urchin‐like microsphere, each of which has sharp and smooth nanoneedles with a length of 2 µm (Figure S4b, Supporting Information). After being calcined in the air atmosphere, the entire urchin‐like sphere swells to about 5 µm (Figure 1c). Meanwhile, the precursor changes from smooth nanoneedles to porous and rough nanotubes due to nonequilibrium diffusion,[ ] and each nanotube is about 250 nm thick, as shown in the inset of Figure 1c. There is no significant difference in the morphology of NiCo2O4 and NiCo2O4@CeO2 because they are both urchin‐like microspheres, as seen in Figure S3a,b (Supporting Information). The only difference between them may be the adsorbed CeO2 nanoparticles on the surface. As can be observed in Figure S3c,d (Supporting Information), the CeO2 particles are about 40 nm in size without the addition of nickel salt and cobalt salts, and the diameter of the CNTs is around 25 nm. The composite of CeO2 particles can also increase the surface area of the composite material, thereby the active sites can be increased, which is consistent with the results of the Brunauer–Emmett–Teller (BET) experiment. This kind of microstructured catalyst can provide enough catalytic surface to enlarge the area of contact between the cathode material and the electrolyte as well as the open space to facilitate the transfer of oxygen or lithium ions, thus providing more active sites for OER and ORR reactions, and sufficient specific surface district for the storage of discharge products. In addition, the porous structure of the nanotubes greatly promotes barrier‐free oxygen transmission and electrolyte penetration, and helps establish multi‐dimensional charge diffusion paths.[ ] The energy‐dispersive X‐ray spectroscopy diagram of Figure S4c–f in the Supporting Information shows the uniform distribution of Ni, Co, and Ce, indicating the successful recombination of Ce, which is also in accord with the results of Figure 1b. The atomic ratio of Ni to Co is about 1:2 (Table S1, Supporting Information), which is close to the original feed ratio, and the Ce content is about 18.7%. The above evidence demonstrates that after CeO2 is deposited, CeO2 nanoparticles are well coated on the surfaces of NiCo2O4 nanotubes, forming a typical hierarchical composite structure. It is noteworthy that although the surface roughness of NiCo2O4@CeO2 increases obviously compared with a single NiCo2O4 microsphere, the open voids are still able to be well preserved without being totally blocked, and the porous structure is retained even after the CeO2 particles are wrapped. The chemical composition and bonding state of NiCo2O4@CeO2 nanocomposites were deeper analyzed by X‐ray photoelectron spectroscopy (XPS), which was shown in Figure S5, Supporting Information.

The more specific microstructure and morphology of NiCo2O4@CeO2 were studied via transmission electron microscope (TEM). Figure  reveals the diameter of the nanotubes that make up NiCo2O4@CeO2 urchin‐like microspheres is about 200 nm, which is particularly consistent with the SEM image situated in the inset of Figure 1c. It is also found that the nanotubes are composed of self‐assembly of nanoparticles and they have a well‐constructed vesicular structure. The high‐resolution TEM (HRTEM) image in Figure 2b clearly shows a distinct heterogenous interface, with a lattice spacing of about 0.24 and 0.27 nm, which are in keeping with the (311) plane of NiCo2O4 and (200) plane of CeO2, respectively. This is in line with the results of the XRD pattern. The relevant selected area electron diffraction pattern (Figure 2c) shows clearly defined rings, indexed as (311) and (422) crystal planes of NiCo2O4 mixed with (200) and (220) crystal plane of CeO2. This indicates the polycrystalline nature of NiCo2O4@CeO2 urchin‐like microspheres, and further proves the successful preparation of NiCo2O4@CeO2 composites. In addition, the homologous energy dispersive X‐ray (EDX) element mapping image of a single nanorod (Figure 2d–g) reveals the uniform distribution of Ni, Co, and Ce elements, indicating that Ce has been successfully incorporated into the composite material. On the other hand, the type IV behavior with the existence of a hysteresis loop was determined by the typical nitrogen adsorption–desorption isotherm characteristics, as displayed in Figure S6 (Supporting Information).

Figure 2

a) TEM image of NiCo2O4@CeO2 nanotubes. b) HRTEM image of NiCo2O4@CeO2 composite nanostructure. c) Electron diffraction pattern of NiCo2O4@CeO2. d–g) Elemental mapping images of mixed, Ni, Co, Ce of the NiCo2O4@CeO2 nanotube.

Electrochemical Performances

In order to study the catalytic behavior of this composite material, NiCo2O4@CeO2 was employed as the cathode active substance of nonaqueous LOBs for charge and discharge tests. For comparison, NiCo2O4, CeO2, and pure CNTs‐based cathode materials were also evaluated under similar conditions. Figure  shows the premier discharge/charge curves of NiCo2O4@CeO2, NiCo2O4, CeO2, and pure CNTs electrodes under the voltage window within 2.0–4.5 V at an ampere density of 500 mA g−1. Note that the charge and discharge capacity of pure CNTs‐based cathodes can be negligible, which indicates that the capacity contribution mainly comes from the active material. NiCo2O4@CeO2 catalyst presents a full charge and discharge capacity of 5537/5586 mA h g−1 with a relevant Coulombic efficiency of 99.1%, higher than 4655/4990 mA h g−1 of CeO2 with 93.2% and 3439/3646 mA h g−1 of NiCo2O4 with 94.3%, which confirms that NiCo2O4@CeO2 composite material can improve the Coulombic efficiency. More significantly, the NiCo2O4@CeO2 electrode shows the lowest charge (0.89 V) and discharge (0.18 V) medium capacity potential, which is defined as the potential at half‐capacity, leading to a pretty less polarization of only 1.07 V. As for NiCo2O4@CeO2 cathode, the charging and discharging processes are considerably stable with a fairly smooth voltage plateau in a broad capacity scope, which means that the formation and decomposition of Li2O2 continue to occur at a nearly steady reaction rate.[ ] These results illustrate that the composite material can significantly increase the specific capacity and reduce the overpotential of the LOBs, indicating that NiCo2O4@CeO2 has high electrocatalytic activity and effective reaction kinetics.

Figure 3

a) The initial discharge/charge curves of NiCo2O4@CeO2, NiCo2O4, CeO2, and pure CNTs electrodes at an ampere density of 500 mA g−1. b) The discharge/charge curves of NiCo2O4@CeO2 electrodes with different ampere densities. c) The initial discharge/charge curves of NiCo2O4@CeO2, NiCo2O4, CeO2, and CNTs electrodes with a cut‐off capacity of 500 mA h g−1 at an ampere density of 500 mA g−1. d) Typical discharge/charge curves of NiCo2O4@CeO2 electrode with different cycles at 500 mA g−1 under a limited capacity of 500 mA h g −1. e) Cycling properties of NiCo2O4@CeO2, NiCo2O4, CeO2, and CNTs electrodes with a limited capacity of 500 mA h g−1 at 500 mA g−1.

Figure 3b records the rate performance of the NiCo2O4@CeO2 composite electrode in a 2.0–4.5 V voltage window when the ampere densities increase from 500, 1000, 1500 to 2000 mA g−1. It could be seen that the NiCo2O4@CeO2 electrode exhibits discharge capacities of 5586 mA h g−1 at 500 mA g−1 and 6120 mA h g−1 at 1000 mA g−1. Even at the high current density of 2000 mA g−1, the NiCo2O4@CeO2 electrode still has a discharge capacity of 2979 mA h g−1. Interestingly, with a moderate ampere density of 1000 mA g−1, the capacity retention rate of NiCo2O4@CeO2 electrode is up to 85.7%, showing excellent rate capability. Although the capacity and efficiency, as well as the overpotential of the composite material, deteriorate with the increase of current density due to the growing internal transfer impedance (R ct) of the whole LOBs,[ ] they still show excellent performance among similar materials. This could be imputed to the unique recombination interface of NiCo2O4@CeO2 electrode that enhances electronic interaction, and the existence of vacancies can promote rapid electron transfer, leading to the reversible formation and decomposition of Li2O2.[ ] Figure 3c displays the initial discharge/charge profiles of the four samples with a limited capacity of 500 mA h g−1 with an ampere density of 500 mA g−1. The NiCo2O4@CeO2 electrode shows a smaller polarization than the NiCo2O4 cathode, CeO2 cathode, and pure CNTs cathode, with discharge/charge overpotentials of 0.89/0.18, 1.04/0.23, 1.22/0.50, and 1.36/0.32 V, respectively. NiCo2O4@CeO2 electrode presents better ORR and OER catalytic performance because the overpotential is the lowest (1.07 V), indicating that this active material can effectively reduce the overpotential.

In order to ensure the reversibility of oxygen reduction/evolution, we tested the LOBs with a limited charge/discharge depth, for which the cycle properties of the battery were measured when the ampere density is 500 mA g−1 and the cut‐off capacity for the LOBs is limited to 500 mA h g−1. The typical discharge/charge curves corresponding to different cycles of NiCo2O4@CeO2 electrode (Figure 3d) suggest that the termination charge voltage hardly increased during the first 300 cycles and no decline is observed, which is better than that of NiCo2O4, CeO2, and pure CNTs (Figure S7a–c, Supporting Information) during cycling. Figure S8a in the Supporting Information tests the charge–discharge rate performance of NiCo2O4@CeO2 electrode at different current densities from 200 to 1000 mA g−1 when the constant limited capacity is 500 mA h g−1. As shown in Figure 3e and Figure S8b in the Supporting Information, the charging voltage of all electrodes abruptly increases during the first five cycles, and then gradually stabilizes. The possible reason is that the reaction was not complete at the beginning, and Li2O2 was not decomposed in time. Under the same conditions, the termination charge voltages of pure CNTs, CeO2, and NiCo2O4 cathodes exceed 4.5 V in a short time, and then the terminal discharge voltage begins to drop sharply, showing limited cycle numbers of 70, 100, and 155 cycles, respectively. In sharp contrast to NiCo2O4 and CeO2 electrodes, NiCo2O4@CeO2 electrode can achieve 400 times longer cycle life in 2.0–4.5 V, and shows impressive cycle performance. From the above results, it can be concluded that the NiCo2O4@CeO2 composite material as an effective bi‐functional catalyst shows a huge advantage in nonaqueous LOBs. The synergistic effect of NiCo2O4 and CeO2 enhances its catalytic activity, and has excellent charge and discharge capabilities as well as cycle stability. The catalytic activity and redox mechanism of NiCo2O4@CeO2, NiCo2O4, CeO2, and CNTs were studied by cyclic voltammetry (CV) within a voltage window of 2.0–4.5 V relative to Li/Li + at a scanning speed of 0.1 mV s −1 (Figure S9a, Supporting Information). To demonstrate the excellent kinetic performance of the composite electrode and greatly enhance the catalytic performance of the LOBs, electrochemical impedance spectroscopy (EIS) tests were accomplished on the different states of the electrode in the frequency scope of 100 kHz to 0.01 Hz (Figure S9b,c, Supporting Information). Besides, to understand the final product on the electrode surface, the cell was disassembled in the glove box and the electrode was washed three times with ethylene glycol dimethyl ether. The morphology after charging and discharging is shown in Figure S10 (Supporting Information). Table S2 in the Supporting Information also shows the comparison of the catalyst in this work with other related NiCo2O4 or CeO2 cathode materials for LOBs. These results indicate that the interaction between NiCo2O4 and CeO2 may lead to excellent electronic/ion conductivity and improve catalytic performance. The formed nanotubes are open and porous, which not only facilitates the transmission of lithium ions and oxygen, but also greatly enhances the ion diffusion ability.

Density Functional Theory Calculations

In order to gain deep insight into the effect of CeO2 on the catalytic properties of NiCo2O4 for the OER at the atomic level, we implemented density functional theory (DFT) calculations to investigate the ORR and OER processes from both the thermodynamic and electrochemical perspectives. The computing approach has a detailed description in the Experimental Section. Primarily, the adsorption energy (E ads) of the ORR intermediate was calculated to determine the possible adsorption states at the same surface site, as shown in Figure  . As the adsorption of O2 and Li on the surface of NiCo2O4 and NiCo2O4@CeO2, the value of E ads gradually decreases, especially for the LiO2 molecular, of which the E ads presents the maximal variance. It follows that the adsorption energy of the first lithiation step on these two surfaces is lower than that of O2, indicating the reduction of O2 on a clean surface of NiCo2O4 and NiCo2O4@CeO2 is easier than oxidation. As the adsorption continues, the NiCo2O4 and NiCo2O4@CeO2 surfaces show significantly different adsorption characteristics. Around the interface of NiCo2O4@CeO2, the adsorption energies of the intermediate state Li2O2 and Li3O4 become relatively weak, while a large decrease in variance occurs at the stage of 2(Li2O2). This observation is not notable in NiCo2O4, suggesting that NiCo2O4@CeO2 is more likely to be covered by the Li2O2 aggregates than NiCo2O4. On the other hand, the weak E ads of Li2O and 2(Li2O) suggests that the two surfaces conduct catalytic processes through the favorable 2e− pathway, thus confirming the experimental speculation. It can be inferred that the most significant change for NiCo2O4@CeO2 occurs in the adsorption of LiO2 to 2(Li2O2) molecules, because it largely determines the charge and discharge voltage, thereby affecting the catalytic efficiency.

Figure 4

a) Adsorption energy (E ads) of the ORR intermediates (O2, Li, LiO2, Li2O, Li2O2, Li3O4, 2(Li2O), and 2(Li2O2) on the NiCo2O4 surface and NiCo2O4@CeO2 surface. b) Planar‐averaged CDD of NiCo2O4 surface and NiCo2O4@CeO2 surface. The red solid line represents the value of NiCo2O4@CeO2 surface, and the green dotted line is that of NiCo2O4 surface for comparison. c) PDOS of the clean NiCo2O4 and NiCo2O4@CeO2. d) Potential‐dependent profiles of intermediate discharge adsorption to the surface of NiCo2O4 surface and NiCo2O4@CeO2 surface.

Then, we performed planar‐average charge density difference (CDD) analysis (Figure 4b) to obtain the charge‐transfer properties near the interface of NiCo2O4@CeO2. For comparison, the CDD of NiCo2O4 under the same spatial phase and scale is presented by the green dashed line. It is observed that excess electrons are obtained by Ce and O atoms of CeO2, forming an electron gas near the interface. As a result, the atoms of the NiCo2O4 termination lost excess electrons at the interface. The charge transfer from NiCo2O4 to CeO2 also leads to a drastic charge fluctuation in NiCo2O4, which exists in the range of at least four atomic layers. Figure 4c shows the partial density of states (PDOS) of NiCo2O4 and NiCo2O4@CeO2, respectively, which reveals that these two systems have metallic properties. Except that the PDOS of NiCo2O4@CeO2 is increased at the Fermi level compared with NiCo2O4, the remarkable difference between them is that the electronic orbital overlap for the former is more significant, which can facilitate the charge transfer capability during the electrochemical reaction.

Before analyzing the electrochemical reaction in these two systems, we constructed a phase diagram to quantify the stability of the ORR product, as shown in Figure 4d. By comparison, the Li2O2 nucleation is initiated when the electrode potential drops to 2.95 V, while on the NiCo2O4@CeO2 surface, Li2O2 nucleates below 2.00 V. The next ORR step on NiCo2O4 and NiCo2O4@CeO2, which is the formation of Li3O4 and 2(Li2O2), occurs below 2.95 and 2.67 V for NiCo2O4, while 2.49 and 2.91 V for NiCo2O4@CeO2. The higher ΔG of 2(Li2O) also confirms that NiCo2O4 and NiCo2O4@CeO2 adsorb middle discharge products via the thermodynamically favorable 2e− route, resulting in the formation of 2(Li2O2). In contrast to the discharge course, which involves the middle‐products adsorption of Li O , the charging reaction occurs straightly through the decomposition of Li2O2 at the electrolyte/Li2O2 interface during the OER process. On the base of Figure 4d, beyond 2.27 and 2.91 V, which are the intersections of the 2(Li2O2) and Li2O2 lines for NiCo2O4 and NiCo2O4@CeO2, respectively, 2(Li2O2) dissociates to 2(Li2O2), 2Li, and O2. These threshold potentials correspond to the minimum charge potentials (see discussion below). The lower charge potential of NiCo2O4@CeO2, when compared to NiCo2O4 strongly, suggests that Li2O2 separates more easily.

Figure  shows the illustration of every reaction route consisting of the five foundational reaction steps, as well as the corresponding free energy charts traced with the OER and the ORR routes with the most stable adsorption configurations of the ORR middle‐products on the NiCo2O4 and NiCo2O4@CeO2 surfaces, respectively. To demonstrate the electrocatalytic activity on every surface, discharge potentials (U DC) and the charge potentials (U C) were evaluated as the maximum and minimum voltages, respectively, causing every route to maintain declivity. It can be concluded that the NiCo2O4@CeO2 surface outperforms the NiCo2O4 surface. In this case, the U C of the NiCo2O4@CeO2 surface is noticeably better, with a very low potential of 3.60 V than that of the NiCo2O4 surface (3.87 V). On the other hand, with the higher voltage of 2.20 V, the promotion of the U DC on the NiCo2O4@CeO2 surface appears more prominent.

Figure 5

a) Top and side views of the ORR mechanism for the O2 reduction near the NiCo2O4 or NiCo2O4@CeO2 surface. Free energy diagrams for the charge and discharge process on the b) NiCo2O4 surface and c) NiCo2O4@CeO2 surface. The asterisk labels the adsorbed states of the molecules. d) Computed charge overpotentials (η C), discharge overpotentials (η DC), and the aggregated overpotentials (η Total) on the NiCo2O4 and NiCo2O4@CeO2 surfaces. The green and orange bars represent η C and η DC, respectively, and the η Total is displayed above the stacked bars.

To help clarify how much this promotion in the electrocatalytic activity is corresponding to the electro‐catalytic efficiency of NiCo2O4@CeO2 surfaces, the overpotentials are evaluated as shown in Figure 5d. The following formula is used to calculate the overpotential during the first charge and discharge process, 1) η C = U C – U 0, 2) η DC = U 0 − U DC. From the results in Figure 5d, it can be seen that when NiCo2O4 is used as the cathode catalyst of LOBs, the total overpotential for the first charge and discharge is 2.00 V, and when the CeO2 is doped into NiCo2O4, the total overpotential drops to 1.60 V. CeO2 doping of NiCo2O4 reduces the overpotential of the OER and ORR processes during the charging and discharging of the lithium–air battery, and the overpotential reduction (0.29 V) of the OER process is greater than that of the ORR process (0.11 V). Evidently, the NiCo2O4@CeO2 acts as a double‐way catalyst that plays a role in both the charge and discharge process, which is consistent with the experimental detection.

Conclusions

In summary, we synthesized a new type of porous urchin‐like NiCo2O4@CeO2 composite microspheres as a bi‐functional catalytic cathode for OER and ORR in LOBs. A number of CeO2 nanoparticles are tightly adsorbed on NiCo2O4 nanotubes, forming a rich interface. DFT calculations show that there is a strong chemical interaction between NiCo2O4 and CeO2, which stimulates the electronic conductivity of the composite material. When tested in the LOBs, the NiCo2O4@CeO2 electrode shows a relatively low overpotential of 1.07 V, and a highly long life of 400 cycles under a fixed capacity of 500 mA h g−1 at 500 mA g−1. The noteworthy improvement in the electrocatalytic properties of ORR/OER is attributed to the open structure of the composite material and the synergistic effect of the customized catalytic material. The unique porous structure can establish rapid oxygen and electron transport in the cathode and provide enough space for Li2O2 storage. These characteristics enable NiCo2O4@CeO2 to exhibit significantly improved OER/ORR electrocatalytic activity compared with the simplex components of NiCo2O4 and CeO2. Additionally, with the help of DFT calculations, we clarified the 2e− pathway of NiCo2O4@CeO2 during ORR/OER, as well as the enhancement mechanism of electrochemical performance as a cathode catalyst for LOBs from the perspective of electronic structure and reaction kinetics. This work not only demonstrates a highly efficient oxygen electrocatalyst, but also provides a strategy to improve electrocatalytic performance for LOBs through interface/surface engineering.

Experimental Section

Material Preparation

At room temperature, NiCl2·6H2O (0.297 g), CoCl2·6H2O (0.595 g), cerium acetate (0.181 g), and urea (0.27 g) were deliquescent in ultrapure water (25 mL), and the liquor was sonicated for 0.5 h. Then it was removed to a 50 mL stainless steel autoclave lined with polytetrafluoroethylene and heated to 120 °C for 8 h. After cooling to room temperature, the product was obtained by centrifugation, washed three times with deionized water and ethanol, respectively, and then dried overnight under 80 °C vacuum. Finally, the collected product was calcined at 375 °C with a temperature rate of 1 °C min −1 for 2 h to obtain urchin‐like NiCo2O4@CeO2 microspheres in the air. For comparison, NiCo2O4 microspheres and CeO2 nanoparticles were also prepared by the same procedure only by adding the corresponding metal salt.

Materials Characterization

The powder XRD patterns were measured on Bruker‐Axs D8 X‐ray diffractometer using Cu K radiation at 40 kV and 40 mA. The XPS information was performed by a Thermo Scientific K‐Alpha instrument with an Al K source. The surface morphological characteristics of the products were analyzed by using SEM (Hitachi SU‐70). TECNAI F‐30 TEM operating at 300 kV was acquired to observe the microstructure of the samples. The BET specific surface area and Barrett–Joyner–Halenda pore volume of materials were measured using 3H‐2000PM2 analyzer.

Electrochemical Measurements

2032‐type coin cells with 19 small holes on the cathode side were packed in an Ar‐filled glove‐box (the content of moisture and oxygen <0.1 ppm). Each cell was made up of a fresh Li metal plate as the anode, a glass fiber separator (Whatman, GF/D), 1 m LiTFSI in tetraethylene glycol dimethyl ether as the electrolyte, a piece of carbon paper coated with catalyst as cathode, and a porous nickel foam as O2 diffusion layer. 40 wt% catalyst, 40 wt% mCNT, and 20 wt% polyvinylidene fluoride were evenly mixed with the limited N‐methyl pyrrolidone solution, which was painted on 12 mm carbon paper homogenously using paintbrushes. The samples were dried in a vacuum under 80 °C for 12 h, and the mass loading on the cathode was about 0.3–0.5 mg. After the assembled battery was put in the glove box for 24 h, it was transferred to a bottle full of oxygen, and the NEWARE BTS battery charging system was used at room temperature to perform constant current charge and discharge test in the voltage range of 2.0–4.8 V, and calculate current density and specific capacity by weight according to the cathode catalyst. CV was performed on the CHI660D electrochemical workstation (CH Instrument Co., Ltd., Shanghai, China), with a scan rate of 0.1 mV s−1, within a voltage window of 2.0–4.8 V (vs Li/Li+). EIS was performed on the same electrochemical workstation, and the frequency range was 100 KHz to 0.1 Hz.

Calculation Methods

The DFT calculations were implemented in the Vienna ab initio simulation package (VASP) with a plane‐wave basis set and a projector‐augmented wave method.[ , ] The electron–core interactions were represented by generalized gradient approximation in the parametrization of the Perdew–Burke–Ernzerhof (PBE) pseudopotential,[ ] and the cutoff of the plane‐wave kinetic energy was 400 eV. To solve the underestimation problem of the Ce 4f states by the standard PBE, the DFT+U method was employed and the on‐site Coulomb interaction correction with an effective U value of 5.0 eV was applied in the highly localized Ce 4f states.[ ] The van der Waals interaction was described by using the empirical correction in Grimme's scheme, i.e., DFT + D3.[ ] Spin‐polarized calculations were applied in all the cases.

NiCo2O4 and NiCo2O4@CeO2 were modeled to identify the catalytic activity of the interface formed by NiCo2O4 and CeO2, which was helpful to understand the catalytic mechanism. For convenience, NiCo2O4@CeO2 was constructed containing two nearly symmetric NiCo2O4 (422)||CeO2(220) interfaces, whose rationality can be found in both the literature[ ] and the comparison with the experimental results. The vacuum space was greater than 15 Å, which was sufficient to prevent interaction between periodic images. The adsorption energies relevant to the possible stable configuration of the ORR intermediate products, i.e., (Li)*, (LiO2)*, (Li2O2)*, (Li3O4)*, and 2(Li2O2)* were used to construct free energy diagrams. The Gibbs free energies of these molecules as a result of electrochemical adsorption reactions were calculated using the following expression where ΔE tot is the change in total energy obtained from DFT, ΔE ZPE and S are the changes in zero‐point energy and entropy at standard conditions (T = 298 K and 0 V vs Li/Li+ reference electrode for O2 reduction).

The adsorption energies of Li O intermediates, for a given site, were defined as where E surf is the surface energy, is the energy of molecular Li O species, and is the total system energy.

It should be noted that in this method, lower adsorption energy value, E ads, implied stronger binding between adsorbate molecules and the NiCo2O4 or NiCo2O4@CeO2 surface. The equilibrium potential for bulk Li2O2(s) (Föppl structure) was calculated using the standard free energies of formation as follows where the standard free energy of formation for the bulk Li2O2(s) at 298 K was calculated as

The phonon contributions to entropy were contained in the G based on the harmonic approximation.

Conflict of Interest

The authors declare no conflict of interest.

Supporting Information

Click here for additional data file.