Paradoxical Role of Structural Degradation in Battery Performance

【Research Background】

The performance decay of lithium-ion batteries (LIBs) is influenced by both cycling and storage. The performance decay caused by cycling is primarily due to the chemical-mechanical degradation of active materials, which has been extensively studied. However, the mechanisms of performance decay due to storage differ from those due to cycling, and research on this is relatively scarce. The state of charge (SoC) of the battery remains nearly constant during storage, and the decay caused by storage largely depends on the SoC during rest. At different SoCs, the structural stability of the positive and negative electrode active materials, side reactions with the electrolyte, and the temperature during battery rest can all impact battery performance. The capacity decay caused by storage exhibits a peculiar relationship with SoC, showing that storage performance at high SoC has a better capacity retention rate compared to intermediate SoC. This may be related to the imbalance of side reactions occurring at the anode and cathode during storage.

【Content Summary】

To fill the research gap in this field, this article studies the performance of two representative batteries at two different SoCs (SoC70 and SoC100) using LiNi_1-x-yCo_xMn_yO₂ (NCM6, 1-x-y > 0.6) and LiNi_1-x-y-zCo_xMn_yAl_zO₂ (NCMA8, 1-x-y-z > 0.8) rich nickel cathodes. The results show that batteries stored at SoC70 have poor capacity retention, while batteries stored at SoC100 produced more gas. Although high Ni cathodes are generally considered more unstable at high SoC, NCMA8/graphite batteries demonstrated comparable or even slightly better capacity retention during storage compared to NCM6/graphite batteries. Through in situ XRD and battery disassembly analysis, the root causes of the reverse capacity decay during storage were revealed. In situ XRD observations indicated severe lithium inventory loss in SoC70 batteries during storage, suggesting that the capacity decay during storage at intermediate SoC is mainly caused by lithium inventory loss. On the other hand, disassembly analysis combined with in situ XRD revealed that the capacity decay during high SoC storage is primarily due to structural degradation of the NCMA8 cathode material, while side reactions at the positive electrode suppressed lithium inventory loss. The research results reveal the structural degradation of cathode active materials (CAM) and their contradictory effect on the overall battery capacity retention. The authors point out that the reductive side reactions at the anode and the oxygen released from CAM are the main causes of storage decay.

【Main Content】

Capacity Retention and Gas Evolution During High-Temperature Storage

Figure 1 Capacity retention and gas evolution during high-temperature storage at SoC70 and SoC100.

Figure 1 shows the capacity retention of pouch batteries and the gas generation characteristics during high-temperature storage. NCM6/graphite (Figure 1a) and NCMA8/graphite (Figure 1b) full batteries stored at SoC70 experienced faster capacity decay compared to those at SoC100. In contrast to capacity retention, gas generation due to side reactions at SoC100 was more severe than at SoC70. The differences in electrode potentials at different SoCs led to remarkably different side reactions during storage. The aliphatic hydrocarbons produced at SoC100 were about three times those at SoC70. The oxygen released from the CAM leads to irreversible oxidation of the electrolyte resulting in the generation of CO and CO₂, with SoC100 being about 12 times higher than SoC70. The higher cathode potential (approximately 300mV, Figure 1c) at SoC100 triggered more severe oxidation reactions, producing more CO₂. This indicates that the oxidation side reactions of CAM at SoC100 are more severe, leading to the decomposition of CAM and generation of CO₂ and lithium oxalate. Lithium oxalate triggers a self-discharge shuttling reaction between the cathode and anode, where CO₂ dissolves in the electrolyte and diffuses to the anode, reacting with lithium ions to reduce to lithium oxalate, which can then diffuse back to the cathode, inserting lithium ions and being oxidized to CO₂. This self-discharge reaction can recover in subsequent cycles, thereby suppressing lithium inventory loss and capacity decay during storage at SoC100. In contrast, the lower CO₂ generated at SoC70 indicates that the more stable SoC70 suppressed irreversible CAM decomposition and associated side reactions. In this case, lithium ions extracted from the anode could not be re-collected into the cathode, thus the aforementioned self-discharge reaction could not occur at SoC70. This led to irreversible lithium inventory loss throughout the battery, resulting in rapid capacity decline. From the results of capacity retention and gas generation, it can be seen that the decay caused by storage at SoC70 and SoC100 is driven by different mechanisms. Additionally, considering that lattice oxygen release typically leads to irreversible degradation of CAM, the superior capacity retention at high SoC100 raises questions about the structural activity of CAM during cycling after experiencing so many side reactions. To determine the impact of structural degradation of CAM after storage at SoC70 and SoC100 on overall battery decay, experimental analysis of the structure of anodes and cathodes during cycling after high-temperature storage was conducted.

Structural Changes of NCM6/Graphite Batteries After High-Temperature Storage

Figure 2 Structural evolution of the anode after high-temperature storage.

Figure 3 Structural changes of NCM6 cathode obtained through in situ XRD.

By analyzing the decayed NCM6/graphite batteries, the structural changes of the active materials were analyzed using in situ X-ray diffraction patterns of the anode (Figure 2) and cathode (Figure 3a–c). The “12w-S100” in the figure represents the battery stored for 12 weeks at SoC100. As storage time increased, the lithium content in the anode decreased (Figure 2d), while the lithium released from the cathode failed to fully reinsert into the anode, possibly due to degradation of CAM or loss of cyclable lithium in the battery. The increase in storage time also led to a downward shift in the peak position of the d-spacing in the anode (Figure 2e), indicating a reduction in lithium concentration in GIC. The anode of SoC70 batteries displayed lower lithium content at the charge and discharge endpoints, indicating lithium inventory loss throughout the entire battery after storage. Compared to the initial state, the structural evolution of decayed NCM6 shows: (1) a decrease in the maximum value of the lattice parameter c (i.e., interlayer spacing), (2) a slightly higher SoC at the end of discharge, and (3) a contraction of c at the charge endpoint in SoC100 batteries. Additionally, the lattice parameter a of decayed NCM6 slightly decreased (Figure 3b), and the unit cell volume V also decreased (Figure 3c), which is reflected at the end of discharge (EOD). The corresponding peak shifts in the XRD patterns (Figure 3d) indicate a higher SoC of the cathode. In the 12w-S70 cathode, the increase in SoC is caused by the shift of the anode potential. Since the graphite anode cannot be further delithiated, the NCM6 cathode in batteries stored at SoC70 cannot be sufficiently lithiated at EOD. Under SoC100 conditions, the (003) peak shifts to a higher angle (Figure 3e, f), indicating significant contraction of c at the charge endpoint. As the storage time at SoC100 increased, the contraction of c became more severe. Furthermore, a minor phase known as the fatigue phase appeared in the high SoC region (Figure 3e, f). The fatigue phase has limited capacity. The (003) peak in the figure decomposes into two peaks (Figure 3g), corresponding to the active phase and fatigue phase, respectively. As storage time increased, the fatigue phase in the 12w-S100 battery became more pronounced compared to the 12w-S70. The structural evolution of the fatigue phase is similar to the former.

Structural Degradation of NCM6 Cathode After High-Temperature Storage

Figure 4 Dissection analysis of aged NCM6 cathode.

Through half-cell testing, the structural degradation of CAM was revealed. During the experiment, the NCM6 cathode was taken from the decayed full battery and reassembled with a lithium anode to form a half-cell, which was then measured after two cycles. Despite two cycles in the NCM6 half-cell, the voltage curve changes of SoC100 in the full battery were still reflected in the half-cell (Figure 4a). The corresponding peaks on the dQ/dV curve of the 12w-S100 (Figure 4b) were more pronounced than the initial state (BoL), indicating permanent structural changes in the NCM6 cathode. Even after 12 weeks of storage, the overall cell capacity retention at SoC70 (90.6%) was lower than at SoC100 (94.0%), but the situation for the NCM6 cathode was the opposite (Figure 4a). Therefore, the overall capacity decay at SoC70 is mainly due to lithium inventory loss rather than irreversible CAM degradation. When there is sufficient cyclable lithium in the battery, capacity can be restored. Continuous structural degradation of the NCM6 cathode after high-temperature storage can be observed through in situ synchrotron X-ray diffraction (SXRD) patterns (Figure 4c, d). The presence of a surface reconstruction layer (SRL) was confirmed on the surfaces of the 12w-S70 (Figure 4e) and 12w-S100 (Figure 4f) cathodes. Moreover, analysis of the Ni K-edge XANES indicates the overall average oxidation state of Ni. The results show that the oxidation state of Ni in the active phase of 12w-S100 is higher.

Scheme 1 Schematic of aging caused by storage.

Scheme 1 summarizes the degradation mechanisms revealed by in situ X-ray diffraction and characterization of decayed NCM6/graphite batteries under storage conditions. Although storage at SoC70 does not severely degrade CAM, the loss of lithium inventory during storage shifts the anode potential curve to the right (Scheme 1a). This shift in anode potential prevents the utilization of the cathode in the low SoC region, leading to irreversible capacity decay. This capacity decay primarily stems from the loss of lithium inventory. On the other hand, capacity decay caused by storage at SoC100 is mainly due to the permanent structural degradation of CAM. Although CAM degradation at SoC100 is more severe than at SoC70, the related oxidative side reactions suppressed the shift in anode potential (i.e., loss of lithium inventory) and subsequent capacity decay. Therefore, the capacity loss at SoC100 is due to the degradation of cathode structure rather than loss of lithium inventory during storage.

Characterization of NCMA8 Cathode Structure Obtained from Disassembled Batteries After High-Temperature Storage

Figure 5 In situ SXRD analysis of aged NCMA8 cathode at charge endpoint.

Figure 6 Dissection analysis of aged NCMA8 cathode.

By further studying NCMA8, the impact of structural degradation of CAM on storage-induced decay was explored. The in situ synchrotron X-ray diffraction results of NCMA8 (Figure 5) show similar c contraction and the presence of fatigue phase as seen in the NCM6 cathode. The proportion of the fatigue phase increases with increasing SoC and storage time (Figure 5a, b). The lattice parameters c, a, unit cell volume V, and c/a ratio of NCMA8 (Figure 6a-d) indicate that the SoC at the discharge endpoint of the decayed cathode is slightly higher (i.e., lithium deficient) than the initial state. In the low SoC region, the c/a ratio increases with increasing SoC, which can be used to assess the SoC of the cathode. Due to the higher proportion of fatigue phase at the charge endpoint (Figure 5b), this indicates a thicker and/or wider surface reconstruction layer (SRL) on the CAM surface, and the presence of SRL hinders lithium insertion. The slow insertion of lithium can be observed by observing the dQ/dV curve during discharge. Subsequently, GITT was used to verify the hindrance of diffusion due to the SRL on the surface of the decayed NCMA8 cathode (Figure 6f, g).

Figure 7 XAS analysis of aged NCMA8 cathode Ni L3-edge.

The changes in the oxidation state of Ni in the discharge states of SoC70 and SoC100 after 12 weeks of storage were studied (Figure 7a, b). The Li defects in the SoC70 cathode lead to a higher oxidation state of Ni (Figure 7a). However, the oxidation state of Ni in the SoC100 cathode significantly decreased. Surface oxidation state changes due to the presence of SRL were studied through sXAS (Figure 7c, d). The formation of SRL leads to a significant decrease in the oxidation state of Ni near the surface region. The NCMA8 stored for 4 weeks exhibited a more pronounced decrease in surface Ni oxidation state compared to the NCM6 stored for 12 weeks, highlighting the surface instability of high-nickel NCM(A).

Scheme 2 Observed CAM structural degradation after high-temperature storage.

Scheme 2 summarizes the structural degradation processes that occurred during storage: (1) During storage, the oxidation state of Ni in CAM decreases, leading to the formation of SRL on the CAM surface. (2) Under high SoC conditions, the appearance of fatigue phase during cycling after storage is one of the main reasons for capacity decay. (3) The mixing of Li/Ni cations slightly increases, but CAM essentially maintains its crystal structure. Storage decay better reflects the unique structural instability during specific SoC storage.

【Conclusion】

The research results comprehensively demonstrate the structural degradation of CAM during storage and its reverse impact on capacity retention. Storage at moderate SoC leads to relatively minor degradation of CAM, but due to severe lithium inventory loss in the full battery, capacity decays rapidly. In contrast, storage at high SoC suppresses capacity decay, but at the cost of severe CAM degradation, leading to the generation of CO₂. Although side reactions may suppress apparent capacity decay in the early stages, the degradation of CAM and related side reactions consume electrolyte, increasing the potential risk of battery failure. Therefore, strategies to mitigate storage-induced decay will focus on suppressing oxygen release from CAM, reducing irreversible lithium inventory loss during intermediate to high SoC storage, and enhancing reductive stability at the anode. The structural analysis provided in this article offers valuable insights for reducing gas generation during high SoC storage and optimizing storage protocols.

Paradoxical role of structural degradation of nickel-rich layered oxides in capacity retention upon storage of lithium-ion batteries.H. Hyun, H. Yoon,S. Choi, J. Kim, S. Y. Kim, T. Regier, Z. Arthur, S. Kim and J. Lim, Energy Environ. Sci., 2023, DOI:10.1039/D3EE02334F.

https://pubs.rsc.org/en/content/articlelanding/2023/ee/d3ee02334f#!divRelatedContent&articles

Research Group of Xin Huo Lin—Polymer Electrolytes: Single Ion Conduction (SIC) or Dual Ion Conduction (DIC)?

2023-08-11