Diagnosis of the Storage Aging of a Commercial Cell with Neutron Powder Diffraction

Abstract

Neutron diffraction is an alternative tool to study the structural changes of a commercial lithium ion cell in operando due to its high penetration, low interaction with matter, and sensitive to low atomic number elements. High resolution neutron diffractometer (Echidna at ANSTO) and high intensity neutron powder diffractometer (Wombat at ANSTO) were used in this study to investigate the structures of a fresh and storage aged commercial 18650 cells (comprised with LiNi0.5Mn0.3Co0.2O2 (NMC)/Li1.1Mn1.9O4 (LMO) composite cathode and graphite anode) and their structural changes upon charge/discharge cycling. It was found that the storage capacity fade increases significantly with increasing storage temperature after 6 months storage. The capacity fades of the cells stored at 0, 25, 50, 75, and 100% DOD are 9.9, 17.5, 7.5, 3.6, and 0.7 %, respectively, after stored at 60oC for 6 months.Figure 1 shows the typical high resolution NPD pattern and its refinement of a fresh cell, while Table 1 lists the results of Rietveld refinement for the NPD patterns of the fresh and aged cells at their the fully discharged and charged states after been stored at 0, 25, and 50 %DOD at 60oC for 6 months. It can be found that the aged cells manifest similar lattice parameters of LMO and NMC at their fully charged state as those of fresh cell, respectively. However, the aged cells show smaller lattice parameter of LMO than that of fresh one and the value decreases with decreasing DOD of storage. That suggests the cell stored at 0% DOD has the most severe structure change in LMO among the samples stored at 60oC for 6 months. From the in-situ 2D ND patterns of the fresh and aged cells collected every 5 minutes with Wombat upon cycling at 215 mA within 2.75 and 4.2 V, shown in Fig. 2, the variation in lattice parameters of LMO and NMC can be determined. The values of the parameters at their charged and discharged states obtained from sequential Rietveld refinement are similar as those shown in Table 1. From the intensity changes of LMO 222 plane after storage for 6 months, it can be found that the sample stored at 0% DOD shows higher LMO loss than other stored samples, which can also be revealed from the abrupt change in lattice parameter of the sample at 4.06 V upon charge/discharge cycling, while linear change in parameter are observed in case of other samples. Aged sample stored at 25% DOD shows lower intensity of LiC6 001 plane manifests the sample has higher lithium loss of inventory (LLI) than others, which can also be revealed from the changes after storage in lattice parameters of NMC at discharged state, shown in Table 1, and intensity of NMC 003 plane at discharged state shown in Fig. 2. These are in consistent with the results obtained from ICA study for the charge/discharge curves and post-mortem studies. The results suggest that the capacity losses of the samples stored at 0 and 25% DOD can be attributed to loss of LMO and LLI, whereas samples stored at depth of discharge higher than 50% DOD can be caused by loss of LLI only. Figure 1