Understanding Degradation Mechanism between Garnet Solid Electrolyte Li7La3Zr2O12 and LiNi0.6Mn0.2Co0.2O2 Cathode

Abstract

All-solid-state Li-ion batteries (ASSLBs) using oxide based solid electrolytes have been considered as a promising candidates for the next generation batteries due to their excellent thermal and electrochemical stability. Especially, garnet-type Li7La3Zr2O12 (LLZO) has gained much attention due to its good conductivity and large electrochemical window. Since the electrochemical window of LLZO allows usage of high voltage cathodes such as NMC (LiNixMnyCo1-x-yO2) and Li anode, it is possible to design high voltage, high capacity and safe energy storage medium using the eletrolyte. However, current attempts on exploiting LLZO as a solid electrolyte have been hindered by high resistance at the interface, especially between cathode and the solid electrolyte. Even though the importance of the issue is well known, there has not been a comprehensive understanding yet on the topic. The lack of understanding comes from the difficulty of characterizing buried interface. It is difficult to study thin interfacial region between cathode and solid electrolyte with conventional techniques since the signal from the interface is often overwhelmed by the bulk, and destructive characterization techniques ruin the signal itself. We developed model system by sputtering thin-film cathode on top of pre-prepared solid state electrolyte pellets to overcome this problem. Samples were annealed afterwards to simulate sintering process. Since the interfacial region is near surface, we could use XANES and EXAFS which are surface-sensitive and non-destructive. Those techniques gave us valuable information regarding oxidation state and local chemical environment of each element, which were used to evaluate migration tendency. Moreover, those results can be used to precisely characterize secondary phases by complementing complex XRD results. Results obtained from those characterization techniques were correlated with electrochemical analysis techniques such as EIS and chronovoltammetry to evaluate the effect of the interfacial degradation on cell performance. In this work, we aimed to obtain fundamental understanding on the instability of cathode|electrolyte interface during two different stages: manufacturing stage and operation stage. We systematically varied variables related to manufacture (Sintering temperature, sintering time, gas environment during sintering), and operation (Electrochemical potential, current density). We chose NMC622 (LiNi0.6Mn0.2Co0.2O2) as a cathode, and Al-doped cubic LLZO (Li7La3Zr2O12) as a solid electrolyte. To study degradation of NMC622|LLZO during manufacturing stage, we varied temperature condition (300°C, 500°C, 700°C for 4h respectively) and gas environment (air, O2, humidified O2, N2, CO2) during annealing. Samples annealed in air showed high interfacial resistance and poor performance, which are signs for interfacial degradation. At 700°C, formation of detrimental phase such as La2Zr2O7 and La(Ni,Co)O3 led to complete blockage of Li ions. XANES and EXAFS showed severe change of oxidation state and chemical environment of Ni and Co, which indicated that those species escaped out of the cathode and participate in the secondary phase formation. In contrast, NMC622|LLZO interface remained chemically stable up to 700°C with no formation of detrimental phase when the sample was annealed in O2. On the other hand, when the sample was annealed in humidified O2 at 700°C, detrimental phase(La2Zr2O7) could be found. In addition, we could characterize substantial amount of Li2CO3 from samples annealed in humidified O2, which could have formed by reaction between LiOH and CO2 during sample characterization done in air. Annealing in CO2 condition led to the crystallization of detrimental phases such as Li2CO3, La2O2CO3, NiCO3 at 500°C. In addition, we could see formation of La2Zr2O7 and La2(Ni,Co)O4 from the sample annealed at 700°C. In both air and CO2 environment, we could find La-(Ni,Co)-O secondary phase from annealed samples, which indicates interdiffusion at the interface. However, extent of the degradation was much higher in CO2 environment. This could be seen by the intensity of XRD peaks corresponding to secondary phases, which overwhelmed the ones corresponding to bulk LLZO. Co L-edge XANES and XRD data for samples annealed at 700°C in different gas conditions could be seen in the attached figure. These findings suggest that interfacial degradation is strongly dependent on the sintering temperature and gas environment. Sintering temperature should be kept as low as possible to avoid formation of detrimental phase. In terms of gas environment, O2 environment is the most promising since the sample remained stable up to 700°C. On the other hand, we found that CO2 and H2O(vapor) are detrimental to the sample. This could be because that they could both form phases which could extract Li out of the system, which led to formation of delithiated secondary phases such as La2Zr2O7. Findings from the work can be used to design optimal conditions to manufacture and operate all solid Li-batteries using LLZO with minimal interfacial degradation. Figure 1