Lithium-ion batteries are widely used as power sources in electric buses, railways, ferries, and other mass transportation systems. To meet the energy needs of these infrastructure systems, it is important to realize battery control that enables high-energy operation for long periods of time. The high-capacity Ni-rich cathode material NCM811 (LiNi0.8Co0.1Mn0.1O2) has attracted research interest as a candidate high-energy-density cathode material. However, NCM811 has been reported to have some degradation problems, especially when operating at potential above 4.2 V versus Li/Li+1).Furthermore, the safety of NCM811 becomes compromised with time. Therefore, understanding the aging mechanism and predicting battery aging behavior are required for effective battery control. We have found that operating NCM811 cathodes at the combination of the potential higher than 4.2 V and lower potential less than 3.6 V that accelerates aging, leading to substantial capacity loss and increased resistance. However, aging was mitigated when NCM811 batteries were operated at a higher or lower potential range. Electrochemical and chemical analyses were performed to clarify the aging mechanism. We prepared 1-Ah LTO/NCM811 pouch cells. Charge/discharge cycling tests were carried out using various state-of-charge (SOC) ranges (SOC0-100%, SOC20-100%, and SOC0-70%) at a 3C rate at 65°C. When the cell was charged at SOC100% and SOC0%, the cathode potential was set to above 4.2 V and to below 3.6 V. The capacity retention and DC resistance (DCR: SOC50%, 0.2 sec) for these cells are shown in Figure1a. The capacity retention of the cell after 1000 cycles at SOC0-100% cycling test was about 75% for the initial capacity. However, a high capacity retention of 93% was obtained at SOC20-100% cycling, despite the cell was charged at SOC100% of 1000 times. The capacity retention of SOC0-70% cycling was nearly 100%. DCR increased 2.5-times after 1000 cycles at SOC0-100% cycling, which was higher increase than that of SOC20-100% cycling and SOC0-70% cycling. To clarify the mechanism, the cathode was removed from the deteriorated pouch cells and its electrochemical properties were investigated using a three-electrode glass beaker cell. The cell was assembled using the cathode and lithium metal as a counter electrode, and a reference electrode was also prepared. The lithium-diffusion coefficient of NCM811 at the time of preparation and after 1000 cycles was measured using the galvanostatic intermittent titration technique method. In the case of a NCM811 cathode operated at SOC0-100% for 1000 cycles, there was little change for the lithium-diffusion coefficient of x < 0.8 in LixNi0.8Co0.1Mn0.1O2, but that of x > 0.8 decreased to less than 1/10. Changes in the crystal structure of the NCM811 particle surface were analyzed by scanning transmission electron microscopy–electron energy loss spectroscopy (STEM-EELS). The thickness of the surface structural change layer was calculated from the EELS spectrum. Also, the thickness and composition of surface film was analyzed by hard X-ray photoelectron spectroscopy (HAX-PES). This measurement was performed at the SPring-8 synchrotron radiation facility. From the STEM-EELS analyses, in the case of SOC0-100% cycling, the surface crystal structure of NCM811 changed from the layered type to the rock-salt or spinel-layer type. The thickness of this transition layer increased by 7 nm compared from beginning. In the case of SOC0-70% cycling, the thickness increased by only 2 nm. It is expected that the cathode operated above 4.2 V at SOC0-100% cycling. From the HAX-PES analysis, the film thickness increased by 20 nm from the initial condition after SOC0-100% cycling. But, after SOC20-100% cycling and SOC0-70% cycling, the film thickness increased was suppressed than that. Based on the results of these analyses, we proposed the following model. In the first step, the structure change of NCM811 was accelerated at high potential. As the rock-salt layer disturbed lithium diffusion, the diffusional resistance increased and the capacity during discharge decreased. In the second step, when NCM811 subjected to a higher potential was discharged to a lower potential, the lithium can’t be intercalated smoothly at the surface x > 0.8 in LixNi0.8Co0.1Mn0.1O2. As a result, the lithium near the surface reacts with electrolyte, forming a film on the NCM811 surface. As the lithium was consumed by the film-growth reaction, the cathode operating potential increased, which accelerated the first step mechanism. The combination of the first and second steps accelerated aging. But this aging can be minimized by controlling the operating voltage. Understanding of the aging mechanism will enable accurate prediction of deterioration.[1]F. Friedrich, B. Strehle, etc, Journal of The Electrochemical Society, 166, A3760-A3774 (2019) Figure 1
Read full abstract