Calendar Degradation Mechanism of Lithium Ion Batteries with a LiMn2O4 and LiNi0.5Co0.2Mn0.3O2 Blended Cathode

Abstract

Introduction In recent years, the degradation of lithium-ion batteries (LIBs) has been intensively analyzed to improve the life expectancy of LIBs in electronic vehicles (EVs). Lifetime tests must be conducted under various conditions that simulate the various real environments of EVs. The degradation phenomena of LIBs are classified into cycle and calendar degradation. Most EVs are exposed to longer parking than driving periods; therefore, suppressing the calendar deterioration of LIBs during parking is important. Recently, LIBs with a mixed-phase blended cathode have been employed as energy storage devices. However, the effect of the cathode composition on performance degradation remains unclear. In this study, we conducted systematic calendar life tests at six SOCs and four temperatures using commercial 18650-type lithium-ion batteries with a Li1-x Mn2O4 and Li1-y Ni0.5Co0.2Mn0.3O2 blended cathode. To understand the degradation mechanism, we adopted nondestructive dV/dQ curve analyses[1]. Experimental Calendar life tests were conducted on commercially available lithium-ion cells with a blended cathode (18650-type, 1.4 Ah). The active cathode and anode materials are Li1-x Mn2O4 + Li1-y Ni0.5Co0.2Mn0.3O2 (25:75 wt.%) and graphite, respectively. Calendar life tests were conducted at six SOCs (0%, 40%, 60%, 70%, 80%, 100%) and four temperatures (0°C, 25°C, 45°C, 60°C). The battery performance was periodically measured in low-current (C/20) charge/discharge tests over the full SOC range (0%–100%) at 25°C (measurement interval was approximately 2 months). The dV/dQ curves were calculated from the discharge curve. The cathode and anode dV/dQ curves were also obtained from the discharge curves of the half-cell against lithium metal electrode with 1 mol dm− 3 LiPF6in ethylene carbonate and diethyl carbonate (1:1 in vol.). Results and discussion The capacity retentions during the calendar life tests are shown in Fig. 1. At 25°C, the capacity decreased fast with the increase of SOC. Moreover, at 60°C, the capacity decreased more rapidly at 60% and 70% SOC than at 80% and 100% SOC. The dV/dQ curve changes during the calendar life tests are shown in Fig. 2. Two of the four peaks in the dV/dQ curve are attributable to phase transitions of the cathode (C1 and C2); the others arise from phase transitions of the anode (A1 and A2)[2]. At 25°C, a slip (decrease of the distance between A1(A2) and C1(C2)) occurred in the cathode/anode reaction region and accelerated at higher SOC. Conversely, at 60°C, negative shifts of C1 and C2 occurred. These shifts, indicating cathode degradation, were specific to 60% and 70% SOC and were accompanied by larger cathode/anode reaction region slips than that at 25°C. From the dV/dQ curves, the lithium composition of spinel type Li1-x Mn2O4 at 60% and 70% SOC was estimated as x ~ 0.2. This fraction is the manganese dissolvable composition[3] and manganese dissolution was confirmed in electrolyte analyses. According to these results, the calendar degradation mechanisms of blended cathode LIB are affected by two mechanisms: cathode/anode reaction region slip induced by electrolyte decomposition at high SOCs, and degradation of cathode by manganese dissolution of the spinel-type Li1-x Mn2O4at intermediate SOCs. Acknowledgement This study was supported in part by the New Energy and Industrial Technology Development Organization (NEDO). Reference [1] I. Bloom, et al., J. Power Sources, 139, 295 (2005). [2] K. Ando, et al., 228th ECS Meeting , No 130 (2015). [3] T. Saito, et al., 200th ECS Meeting , No. 180 (2001). Figure 1