【Background】
Sodium-ion batteries have become one of the most promising supplementary systems for lithium-ion batteries due to their abundant resources and low cost. The development of cathode materials with high specific capacity and high stability is still a research hotspot in the field of sodium-ion batteries. Among many types of sodium-ion batteries, transition metal layered oxides have the advantages of high theoretical specific capacity, low cost, and easy preparation, and are the most promising cathode systems. As a cobalt-free O3-NaxTMO2 cathode, NaNi1/3Fe1/3Mn1/3O2 (NFM) is one of the most ideal cathode materials due to its cost-effective, environmentally friendly and structurally stable advantages. However, although the theoretical specific capacity of NFM is as high as 240 mAh·g-1, its operating voltage is currently limited to less than 4.0 V due to high-voltage instability to ensure a longer cycle life, and its specific capacity is between 130-140 mAh·g-1. The instability of NFM at high pressure (4.0-4.3 V) mainly comes from two aspects: (1) an irreversible phase transition from P3 to OP2 or O'3 occurs, resulting in severe distortion and contraction of the crystal lattice. This transition leads to the formation of microcracks that occur repeatedly throughout the cycle, ultimately leading to structural failure and degradation of electrochemical performance. (2) More serious side reactions will occur at the cathode-electrolyte interface, which will subsequently lead to surface transition metal reduction, lattice heterogeneous reconfiguration, and transition metal ion dissolution accompanied by O loss. The loss of oxygen further stimulates the oxidation of the electrolyte, creating a vicious cycle. The irreversible phase transition in the cathode phase and the degradation at the cathode interface lead to poor cycling stability and Na+ transport kinetics of NFM at high cut-off voltages.
【Job Introduction】
Recently, Wen Zhaoyin's research group at the Shanghai Institute of Ceramics, Chinese Academy of Sciences and others synthesized a CoxB-spinel double-coated NFM cathode material by one-step liquid phase reduction method, and realized the stable cycling of the material at a high cut-off voltage (2-4.3 V). In addition, the stability improvement mechanism of the cathode surface and bulk phase was explored by non-situ EELS, HADDF-STEM and in-situ XRD. The designed cathode system (1%-CoxB@NFM) can obtain a specific capacity of 160 mAh·g-1 at 2-4.3 V, and the capacity retention rate of 300 cycles at 2 C rate is still 70%, which is much higher than the capacity retention rate of 51.2% for the uncoated sample. The article was published in Advanced Energy Materials, a top international journal. Feng Sheng, a doctoral student at the Shanghai Silicon Institute, is the first author of this paper.
【Content Description】
In cathode materials, various modification methods including element doping, surface engineering and morphology design are used for its modification, among which, surface engineering can effectively prevent the cathode from being directly exposed to the electrolyte by applying various coating materials, thereby reducing HF erosion and active species dissolution, and alleviating interfacial side reactions. However, single-layer coating has limited effect in providing interfacial protection and cannot directly improve the stability of the bulk structure of the cathode. The spinel structure (Fd-3m) has excellent stability and an open three-dimensional Na+ transport path, and the in-situ surface reconstruction of a spinel-like structure matching the layered cathode lattice can not only enhance the surface Na+ transport dynamics, but also serve as an anchor point to stabilize the overall layered structure and inhibit its lattice slip. However, the use of spinel-like surfaces alone is not sufficient to inhibit the dissolution of transition metal ions. Therefore, the authors propose a dual-function strategy to simultaneously realize the surface prestructure and cladding layer of the spinel structure, synergistically leveraging the advantages of both surface engineering techniques. The coating material is metallic glass CoxB, which has excellent corrosion resistance and mechanical strength, resists electrolyte attack, and inhibits cracking or fracture of cathode particles. In addition, when CoxB interacts with Na+, it can be transformed into a fast ionic conductor to ensure Na+ transmission. The authors used a one-step room-temperature liquid-phase reduction method to construct a double protective layer composed of spinel layer and amorphous CoxB coating on the surface of NFM in situ, so as to achieve a fine surface structure of layered-spinel-CoxB configuration.
Figure 1. Schematic diagram of the synthesis process of the CoxB@NFM (a) and material characterization results: (b) XRD refinement results, (c, d) HRTEM images of the particle surface and the corresponding FFT, (e) EELS results from the cathode surface to the bulk phase, (f) SEM topography of the particles, (g) HADDF-STEM images of the particle surface and associated EDS element maps.
Figure 1a is a schematic diagram of the synthesis process for a series of samples, with XRD refinement of 1wt%-CoxB@NFM showing no significant differences in crystal structure. The HRTEM image shows the surface of 1wt%-CoxB@NFM particles, forming a spinel-like structure. Figure 1c,d shows a three-part structure including amorphous, spinel, and layered structures. The fast Fourier transform (FFT) of Zone 1 indicates that it is a layered structure along the <110 > belt axis, while the near-surface FFT of Zone 2 confirms the coexistence of spinel and layered structure. The EELS analysis was performed at different depths inside the particles, and Co showed a stronger signal on the surface of the particles, with a negative shift in the Mn L3 peak on the surface, indicating a decrease in valence, while no significant peak shift was observed for Ni and Fe. The energy difference (ΔE) between the front and K edges of O is affected by the hybridization of TM 3D and O2p orbitals. A lower ΔE value at the surface indicates a reduced degree of hybridization, which can be attributed to a reduction in transition metals and a loss of oxygen. In the spinel structure, the valence state of Mn is always lower than that of the layered structure, which further confirms the formation of the spinel structure. Figure 1f shows the morphology of the 1wt%-CoxB@NFM particles, where the particle surface and gaps are coated with amorphous CoxB, and using HADDF-STEM, (Figure 1g), we can observe that the particles are coated with a lower contrast film. The corresponding EDS profile showed that the outermost layer was rich in Co, which was consistent with the results of the EELS profile. These results indicate that a CoxB-spinel-layered structure is formed on the surface of the particles at 1wt%-CoxB@NFM.
Figure 2. Electrochemical performance tests of 1wt%-CoxB@NFM and p-NFM. (a) Rate performance of the two cathodes between 2-4 V at 0.2, 0.5, 1, 2, 5 C (1 C = 130 mAh g-1), (b) first charge-discharge curves of 2-4.3 C at 0.1 C, cycling curves between 2-4 V at (c) 1 C, (d) 2 C cycling curves between 2-4.3 V, (e) DQ-dV curves at 2-4.3 V, apparent Na+ calculated by GITT at the first cycle (f) and 100 cycles (g). Diffusion coefficient.
The electrochemical behavior of different samples was investigated in a half-cell with sodium metal as the anode. The specific capacity of CoxB@NFM 99.6 mAh·g-1 (80.8% of 0.2 C) was obtained at 5 C, which was higher than that of 89.97 mAh·g-1 (67.7% of 0.2 C) at 5 C. 1wt%-CoxB@NFM shows significant cycling stability at both low and high cut-off voltages. After 300 cycles at 1 C between 2 and 4 V (Figure 2c), the specific capacity of 1wt%-CoxB@NFM retains 79.6% of its initial specific capacity, significantly better than the 51.4% of p-NFM. Even at a higher cut-off voltage of 4.3 V (Figure 2d), it retains 70% of its initial specific capacity after 300 cycles at 2 C, much higher than p-NFM (51.2%). To further demonstrate the superior electrochemical performance of 1wt%-CoxB@NFM, the authors performed GITT measurements to quantify the apparent diffusion coefficient of Na+ over multiple charging stages at different cycles (Fig. 2f, g). In the first period, the Na+ diffusion coefficient of 1wt%-CoxB@NFM is slightly higher than that of p-NFM. Notably, the 1wt%-CoxB@NFM did not show a significant decrease in the Na+ diffusion coefficient at the end of the charge (Figure 2f), indicating a mitigation of the contraction of the c-axis, which will be confirmed in the next section. After 100 cycles (Figure 2g), the Na+ diffusion coefficient maintained at 1wt%-CoxB@NFM was almost the same as in the first cycle and significantly higher than that of p-NFM. This highlights the excellent stability of the bulk and interfacial phases at 1wt%-CoxB@NFM, thus ensuring rapid migration and cycling stability of Na+.
Figure 3. In situ and ex situ XRD testing of p-NFM and 1wt%-CoxB@NFM. (a) 2D contour plots of in-situ XRD at 2.0-4.0 V, 0.2 C for p-NFM and (d) 1wt%-CoxB@NFM for different states of charge, and the corresponding charge-discharge curves: (b-c)p-NFM and (e-f)1wt%-CoxB@NFM.
In order to further understand the influence of surface engineering on the evolution of the host structure, the authors used in-situ XRD supplemented by quasi-in-situ XRD to study the phase transitions during initial charge-discharge processes. As shown in Figures 3a,d, in-situ XRD measurements show similar phase transitions between 2-4 V. Quasi-in-situ XRD showed that at 4.15 V, the 1wt%-CoxB@NFM had shifted from a P3 structure to an OP2 structure with (002), (010), and (103) peaks. As the charge continues to 4.3 V, the (002) peak continues to move to a higher angle, indicating that lattice parameter c continues to shrink. At 4.3 V, the peak of p-NFM (002) shifted to the high angle more, and the lattice shrinkage was more severe. The alleviation of lattice shrinkage allows the 1wt%-CoxB@NFM to maintain a more intact structure during long cycling, thereby improving the reversibility and cyclic stability of the phase transition. This enhancement can be attributed to the protective effect of surface engineering. On the one hand, the spinel phase acts as an anchor point to inhibit the slippage and shrinkage of the crystal lattice, and on the other hand, CoxB inhibits the formation of interfacial side reactions and cracks. Together, these synergies enhance the structural stability of 1wt%-CoxB@NFM.
Figure 4. Surface composition and structure characterization after 200 cycles between 2-4.3 V. XPS for CEI analysis: (a) C1s for p-NFM, (b) O2p for p-NFM, (c) C1s for 1wt%-CoxB@NFM, (d) O2p for 1wt%-CoxB@NFM, TEM and HRTEM images for (e-f) p-NFM particles, TEM and HRTEM images for (g-h) 1wt%-CoxB@NFM particles.
The authors performed a series of comprehensive characterizations of the electrode after 200 cycles between 2-4.3 V to investigate the effect of surface engineering on the evolution of the interfacial structure. Figures 4a-d are XPS depth profiles for studying the cathode electrolyte intermediate phase (CEI). Both were covered by C═O and C─O species decomposed by PC solvents, and (CFx) belonged to PVDF. As the etching time is extended, undetected transition metal-oxygen species (TM-O) begin to emerge and the CEI peak formed by electrolyte decomposition gradually weakens. In contrast, the 1wt%-CoxB@NFM surface O, Cl2p, and F1s peaks have a faster decrease in intensity and a higher proportion of lattice oxygen, suggesting that a thinner CEI has formed on the surface, thus ensuring better Na+ transport. The thinner CEI showed that the CoxB coating effectively inhibited the side reactions between the cathode surface and the electrolyte. The particle morphology and surface structure of the circulating cathode were observed by TEM and HRTEM. It can be seen that the surface of the p-NFM particles has deep and wide cracks (Figure 4e), while the surface of the 1wt%-CoxB@NFM shows only narrow and shallow cracks (Figure 4g). The HRTEM image (Fig. 4f) shows that there is significant lattice distortion and dislocation in the vicinity of the p-NFM crack, while the lattice near the 1wt%-CoxB@NFM crack is still well structured (Fig. 4h). These results show that the strain on the surface of the 1wt%-CoxB@NFM particles is alleviated and the formation of cracks is effectively suppressed.
Figure 5. After 200 cycles between 2.0-4.3 V, the surface fine microstructure characterization and structural model schematic diagram of the cathode, (a) HRTEM at the surface of p-NFM, (b-c) FFT of the atomic-scale HADDF-STEM and related regions at the surface of p-NFM, (d) 1wt%-CoxB@NFM HRTEM, (e-f)1wt%- Atomic-scale HADDF-STEM and FFT of interest regions on the CoxB@NFM surface, ICP results for (g)TM ion dissolution tests, (h) structural model of the p-NFM particle surface, (i) 1wt%-CoxB@NFM surface structural model.
In order to further investigate the mechanism of surface structure degradation of p-NFM particles and the enhanced structural stability of 1wt%-CoxB@NFM particles during cycling, HRTEM and HADDF-STEM were used to characterize their fine structures, as shown in Figure 5. For p-NFM, the surface reconstruction from the layered structure to the spinel structure is very severe. It is worth noting that this surface reconstruction is anisotropic, which is related to the Na+ migration pathway. In the layered structure, Na+ moves along a two-dimensional migration path, especially along the (003) crystal plane. As shown in Figure 5a, on the surface parallel to the (003) crystal plane, the reconstructed spinel structure is very thin, less than 10 nm. In contrast, on the surface perpendicular to the (003) crystal plane, a much thicker spinel structure is formed (>20 nm) (Fig. 5b, c). This suggests that the surface remodeling process may be facilitated by Na+ deintercalation, and is therefore more pronounced in regions experiencing dynamic Na+ transport. EELS analysis was performed from the surface of the particles to the inside of the particles. Both the Mn L3 and Ni L3 peaks show a negative shift, indicating a lower valence state, while the peak of iron has not shifted. The ΔE value of the surface is greatly reduced, indicating a weakened hybridization between TM 3d and O2p, which is attributed to the reduction of transition metal ions and the loss of oxygen. The dissolution of low-valent transition metal ions and the loss of oxygen are involved in the oxidation of the electrolyte, which promotes the acceleration of surface remodeling, resulting in the continuous deterioration of the interfacial structure. For 1wt%-CoxB@NFM, the surface parallel to the (003) crystal plane shows a complete layered structure (Figure 5d). The spinel structure can only be observed on a surface perpendicular to the (003) crystal plane (Fig. 5e, f), which is consistent with the original surface structure. The EELS analysis in Figure 6b shows that only the Mn L3 peak shows a slight negative shift and a slight decrease in ΔE on the surface. The above results show that the reduction of transition metal ions and the loss of O in 1wt%-CoxB@NFM are effectively alleviated, and the surface reconstruction is almost stopped. Figure 5h, i is a schematic representation of the surface structure of p-NFM and 1wt%-CoxB@NFM during cycling. These behaviors further induce the migration, reduction, and dissolution of transition metal ions, triggering the restructuring and formation of spinel structures. However, the lattice mismatch between this spinel structure and the layered structure makes it unstable and leads to further diffusion into the bulk phase. Conversely, thanks to surface engineering, the pre-constructed spinel structure on the surface of the 1wt%-CoxB@NFM (Figure 5i) particles is well matched to the lattice of the layered structure, thus ensuring structural stability. The loss of O, the dissolution of transition metal ions, and the side reactions with the electrolyte are all effectively inhibited, thus ensuring the stability of the interface. Through the synergistic effect of surface engineering on the body and surface, the 1wt%-CoxB@NFM cathode achieves remarkable high-pressure long-cycle stability.
Figure 6. Atomic-scale EELS characterization. HADDF-STEM image with EELS from surface to bulk structure in the red box: (a-b)p-NFM, (c-d)1wt%-CoxB@NFM.
【Conclusion】
In this paper, the authors successfully synthesized spinel and CoxB double cladding on the surface of NFM. The spinel structure matched to the layered lattice provides an open three-dimensional Na+ migration channel, anchors the layered structure of the bulk phase, inhibits lattice distortion and sliding, and improves the reversibility of phase transition. At the same time, the outer layer of CoxB plays a crucial role in reducing the loss of O, the dissolution of transition metal ions, and the side reactions with the electrolyte. With this dual coating, 1wt%-CoxB@NFM ensures continuous reversible phase transitions, interfacial stability, and particle integrity over long cycling periods, resulting in unmatched cycling stability. This dual coating strategy provides an effective solution for the development of stable high-voltage sodium-ion battery cathodes, which is of great significance for improving the energy density of sodium-ion batteries.
S. Feng, Y. Lu, X. Lu, H. Chen, X. Wu, M. Wu, F. Xu, Z. Wen, Surface Engineering through In Situ Construction of CoxB-Spinel Dual Coating Layers for High-Voltage Stable Sodium-Ion Batteries. Adv. Energy Mater. 2024.
https://doi.org/10.1002/aenm.202303773.
Note: Most of the articles reproduced on this site are collected on the Internet, and the copyright of the article belongs to the original author and the original source. The views in the article are only for sharing and exchange, if it involves copyright and other issues, please let us know, and I will deal with it in a timely manner.